.. 7 25...:
Z: . .1
A . .

in:
{5 Opt: . I:
I¢ v5

. .

if:
. 1 .
flu.» u.

{0|} '9'
. ‘ at. lJo-Inéftfli
1; a; ‘ a ”find “cur-v.“
t. 4-} . I il'
a ‘0 if).
.. 5”: v :2... r 3.
, .1. .. v
19. Hulfi’czbna
‘ 111).. u ””09.

 

in
I It; a
A {.79 it. 1 afi-

l’. 1! . ll , 1. fl. '
141,... war . aizwfifzn! 13...; r
. . . .5 .r...»
.h.dfiw.n....rlfz. a... .
i” ll;

,ll.‘a 11.1.9". (‘1':

.1 ‘io‘é .f’tzl-fctlr!£l-5OI.
0.
fine: if“??? $9.5...

-

 

vl .f- '1. .lmL-u .
I.:d.hh..n.l.(3v [xv-Ch”; (If:
huv‘wdtpu...lr‘ct . .
1|, .(rl.

l

130?, 4 .

‘Ig'f'flcl‘
ID .

I u 1"!ch M

..
nails-1.1.;
lit 7! I II
(firth...
:27:

Asl

..
.. .1. I...

It! Ila-It: .
I... {In}: ‘

..-y-.—v-—I‘

I: a..l.

 

v 31".”
I)! illitll¢
If. 11...". {a'
04..) 0‘09}:

7. "v va‘ '
w‘v‘xw‘r‘“ ‘
7" .

O
. Vivit
plilrvxll-CIIOVA
.0!.4|I.1.I' .’Ov:..-f
(.a.tl;\‘vrvcn .v:
.I£D0..Ol..).f’ 0"
‘1’! ‘v’i‘b’: ‘
,al

I .51..
4. a .11.: . ‘
{v0 :7, )3 {.3 J. .

'_'.yuvv~"'1. "'

 

.I\-l. ‘
‘ z ‘ Vonnhl.

1.253!

i

LL7|II

Illllllllllllllllllllllllllllllllllllllllllllllllllllllllllll

3 1293 00793 7604

This is to certify that the
dissertation entitled
Effects of Monocrotaline Pyrrole on Cultured
Endothelial Cell Function

presented by

Cindy Marie Hoorn, D.V.M., Ph.D.

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in Pharmacology/
Environmental Toxicology

[Km/til???

Major professor /

Date October 20, 1992

MS U is an Affirmative Action/Equal Opportunity Institution 0—12771

 

 

 

;___

w—v—w—

LIBRARY
Michigan State
University

L _,

7

re

 

 

—_.—.

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES teturn on or before date due.

DATE DUE DATE DUE DATE DUE

1 l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

l"? :l

MSU Is An Affirmative ActiorVEquel Opportunity lnetitution
cmmma-m

 

 

 

‘ fl hit-L-“

“AA .w‘.“ --

EFFECTS OF MONOCROTALINE PYRROLE ON CULTURED
ENDOTI-IELIAL CELL FUNCTION .

By

Cindy Marie Hoorn

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Pharmacology and Toxicology
Institute for Environmental Toxicology

1992

ABSTRACT

EFFECTS OF MONOCROTALINE PYRROLE ON CULTURED
ENDOTHELIAL CELL F UNCI'ION

By

Cindy Marie Hoom

Monocrotaline pyrrole (MCI?) is a putative, toxic metabolite of the
pyrrolizidine alkaloid, monocrotaline. The pneumotoxicity produced by
administration of MCTP in vivo is delayed in onset and progressive in nature and is
characterized by chronic pulmonary vascular alterations. Although the mechanism
by which MCI'P produces lung vascular injury is unknown, there is evidence that
damage to the pulmonary endothelium is involved.

A primary goal of this work was to identify alterations in cultured endothelial
cell (EC) function in response to MCI'P treatment in vitro consistent with the ability
of this compound to cause lung vascular injury in vivo. Changes in activity of
angiotensin—converting enzyme (ACE) have been associated with a number of
conditions that result in lung vascular injury. A decline in the activity of this enzyme
was observed that corresponded to the degree of overt MCI'P-induced cytotoxicity,
suggesting that altered EC ACE activity is unlikely to be a direct effect of MCI? on
the enzyme but may occur in response to the delayed cell injury caused by this
compound.

ECs treated with 150 uM MCTP were unable to proliferate but continued to
synthesize DNA, RNA and protein, ultimately at higher levels than controls. Thus,

basic cellular activity was maintained but was altered substantially by exposure to

Cindy Marie Hoorn
MCI'P. Altered EC production and / or release of mesenchymal cell growth factors
in response to MCI'P treatment could contribute to the vascular remodeling seen in
viva. However, untreated ECs in culture constitutively released high levels of growth-
stimulatory activity, and MCTP treatment did not augment this release.

BCs from the rat, a species sensitive to the pneumotoxic effects of MCTP,
exhibited cytolytic and cytostatic effects in response to MCTP treatment in vitra that
might contribute to the pulmonary vascular changes seen in viva. Persistent DNA
crosslinking, which occurred in BCs treated with MCTP in vitra, may be responsible
for the cytostasis and limited repair capacity evident in cultured ECs treated with
MCI'P. Treatment of cultured ECs with mitomycin C (MMC), another bifunctional
alkylating agent that causes pulmonary vascular injury in viva, resulted in
morphologic, cytolytic and cytostatic changes similar to those caused by treatment
with MCI'P. As with MCTP, DNA crosslinking was observed at concentrations of
MMC that were cytostatic but whichcaused limited cytolysis, suggesting that the
abilities to crosslink DNA and inhibit proliferation are associated and may be
important in the delayed and progressive injury caused by these compounds.

In summary, cultured ECs are inhibited in their ability to proliferate and
demonstrate altered synthetic activity in response to treatment with MCI'P. The
cytostatic effects of MCTP, which impede the endothelial repair process, appear to
be related to the ability of this compound to crosslink DNA. The EC morphological
and functional alterations caused by MC'I'P are determined in part by the extent of
the cytolytic changes that occur in the monolayer which create the need for repair.
DNA crosslinking may be responsible for the delayed and progressive endothelial

response to MCTP treatment.

Copyright by
CINDY MARIE HOORN
1992

"Two roads diverged in a wood, and I--
I took the one less traveled by,
And that has made all the difference."

-Robert Frost
"The Road Not Taken"

To my family,
for believing in me and supporting me
in my seemingly endless quest for knowledge.

ACKNOWLEDGMENTS

First and foremost, I would like to express my gratitude to my research advisor
and friend, Dr. Robert Roth, for his guidance and support. Many years ago, his
enthusiasm and encouragement inspired a skeptical, young veterinary student to
consider a career in research; I thank him for pointing out that road to me.

Iwould like to thank the members of my guidance committee, Dr. Margarita
Contreras, Dr. Frederick Derksen, Dr. Jay Goodman and Dr. Steven Kunkel, for
their advice, direction and patience. I am also grateful to Dr. Kenneth Moore, Dr.
Lawrence Fischer, Dr. Robert Leader and Dr. Robert Poppenga for their interest in
my progress and their helpful career advice and encouragement.

Iwould like to thank Dr. James Varani of the Department of Pathology at the
University of Michigan, and Dr. B.V. Madhukar of the Department of Pediatrics and
Dr. John Wang of the Department of Biochemistry at Michigan State University for
the donation of cells, technical assistance and helpful advice. I would also like to
thank Diane Hummel, Nelda Carpenter, Mickie Vandeer and Marty Burns for
helping me through the administrative "red tape" inherent to University life, and
Terri Schmidt for helping me to gather and organize my voluminous literature files.

I have been fortunate to work with an innovative group of people on the
monocrotaline project. I am especially indebted to Dr. James Reindel for teaching
me the fine art of cell culture and for many stimulating discussions. I also thank Drs.

Eric Schultze, Susan White and Bruce Bowdy, and some dedicated research

assistants: James Wagner, Kathy Vorick, Judy Angelini and Maria Colligan, for their
help, patience, and general bonhomie. In addition to my research cronies, I would
like to thank my other coworkers, who, despite their misdirected interest in an
inferior organ system, have made the past few years memorable and (dare I say it?)
fun. To Drs. Lonnie Dahm, Jim Hewett, Marc Bailie and Patti Ganey, and to Eric
Shobc and Robert Stachlewitz: many thanks, and I wish you all the best. And to Drs.
Sandy Hewett and Nan Kanagy: thank you for your friendship and emotional support.

I would also like to acknowledge the sources of my support. My research was
funded by NIH Grant ESOZSSI. The majority of my stipend was provided by NRSA
BSOS477; I also received support from the Department of Pharmacology &
Toxicology and the College of Veterinary Medicine, and I received a travel award
from Proctor and Gamble.

On a more personal note, I want to thank the Williamson family and my
parents, Julie and George Cromer and Francis and Louise Hoorn, for their loving
we of my son, Christopher. Without their help there would be no dissertation. I
thank all of my family, especially my sister Chris, for their love and understanding;
and I thank my wonderful friends, Bob Byelich, Ellen Kaufman, Dan and Barb
Palazuk, Mike and Sue Rollert, and Linda Van Horn, for the emotional support, the
laughs, and the diversions they provided.

Finally, I gratefully acknowledge my husband, Jay, for putting his life on hold
for a few years so I could attain this goal...thank you for your patience and love and

for always being there to remind me what life is really all about.

vii

TABLE OF CONTENTS

LIST OF TABLES
LIST OF FIGURES
LIST OF ABBREVIATIONS
CHAPTER I - INTRODUCTION
1. Pyrrolizidine Alkaloids
II. Monocrotaline

A. General

B. Metabolic Activation
C. Biological Effects (General)

1 Hepatotoxicity
2 N ephrotoxicity
3. Carcinogenicity
4 Other Effects

D. Cardiopulmonary Effects

1. Pathology
a. Macroscopic changes
b. Microscopic changes
2. Pulmonary Mechanics
3. Vascular Alterations
a. Hemodynamics
b. Vascular reactivity

4. Biochemical Changes

viii

ax-AANHEEEEE

ll
11
12
13
14
15
16

16
16

22
24

24

27

Table of Contents (cont)

E.

F.
III.

A.

B
IV.

FWPOF’?

Monocrotaline Pyrrole Pneumotoxicity
Progression of Cardiopulmonary Alterations

MCI' Pneumotoxicity as a Model of Pulmonary Hypertension

in People

Primary Pulmonary Hypertension
Pulmonary Hypertension in Response to Injury

1. Adult Respiratory Distress Syndrome
2. Other Forms of Pulmonary Hypertension

Mechanisms of MCI‘ (P) Pneumotoxicity

The Role of Vasoconstriction

The Role of Thrombosis

The Role of the Platelet

The Role of Inflammation

The Role of Cell Growth and Proliferation
The Role of Alkylation

V. The Vascular Endothelium

Structure
Homeostatic Functions of Endothelium

Regulation of Vascular Permeability
Regulation of Vasoactivity
Regulation of Hemostasis
Regulation of Vascular Cell Growth
Other Roles

.U'P'P’N!‘

Functional Response of Endothelium to Injury or Disease

1. Alterations in Permeability
2. Alterations in Vasoactivity

ix

30
33

36

36
39

40
41

43

50
51
56
63
68

72

72
76

76
79
82
85
86

87
88

Table of Contents (cont)

3. Alterations in Hemostasis
4. Enhancement of Vascular Cell Growth
5. The "Activated" Endothelial Cell

VI. Endothelial Cell Alterations in MCI‘ (P) Pneumotoxicity

MCT(P)-Induced Endothelial Cell Alterations In Vzva

The Endothelial Cell in Culture

MCT(P)-Induced Endothelial Cell Alterations In Vztra

Potential Role of Endothelium in the Development
of Pulmonary Hypertension

$7.05”?

VII. Research Goals

CHAPTER H - MONOCROTALINE PYRROLE-INDUCED
CHANGES IN ANGIOTENSIN-CONVERTING
ENZYME ACTIVITY OF CULTURED PULMONARY
ARTERY ENDOTHELIAL CELLS

Summary

Introduction

Materials and Methods
Results

Discussion

CHAPTER III - MONOCROTALINE PYRROLE ALTERS DNA, RNA
AND PROTEIN SYNTHESIS IN PULMONARY
ARTERY ENDOTHELIAL CELLS

Summary

Introduction

Materials and Methods
Results

Discussion

91
97
102

102
105
109

118

122

125

126
128
130
134
139

145

146
147
148
152
163

Table of Contents (cont)

Bags:
CHAPTER IV - MONOCROTALINE PYRROLE TREATMENT
DOES NOT ENHANCE THE RELEASE OF
MESENCHYMAL CELL GROWTH FACTORS
FROM CULTURED ENDOTHELIAL CELLS 169
Summary 170
Introduction 171
Materials and Methods 173
Results 176
Discussion 182
CHAPTER V - EFFECTS OF MONOCROTALINE PYRROLE ON
CULTURED RAT PULMONARY ENDOTHELIUM 188
Summary 189
Introduction 190
Materials and Methods 191
Results 196
Discussion 206
CHAPTER VI - THE EFFECTS OF MITOMYCIN C ON
CULTURED ENDOTHELIUM:
COMPARISON WITH MONOCROTALINE
PYRROLE-INDUCED CYTOTOXICITY 210
Summary 211
Introduction 213
Materials and Methods 215
Results 217
Discussion 229
SUMMARY AND CONCLUSIONS 234

BIBLIOGRAPHY 246

LIST OF TABLES

TABLE 1: DNA synthesis by PECs as evaluated by autoradiography. 159

TABLE 2: Lactate dehydrogenase (LDH) isozyme profile of
conditioned medium from rat endothelial cells. 203

xii

Figure 1:
Figure 2:
Figure 3:

Figure 4:

Figure 5:

Figure 6:
Figure 7:
Figure 8:

Figure 9:
Figure 10:
Figure 11:

Figure 12:

Figure 13:

LIST OF FIGURES

Structures of monocrotaline and monocrotaline pyrrole.
Homeostatic functions of endothelium.

Endothelial modulation of vasoactivity in health
and disease.

Endothelial modulation of the hemostatic system
in health and disease.

Endothelial modulation of vascular cell growth
in health and disease.

The ”activated" endothelial cell.
Effects of MCTP on monolayer cellularity.

Release of lactate dehydrogenase from MCTP-treated
monolayers.

Effects of MCTP on colony-forming efficiency.

Effects of MCTP on PEC morphology.

Role of MCT(P)-induced endothelial cell injury in
the development of pulmonary vascular alterations:

An hypothesis.

Potential endothelial cell alterations after exposure
to MCTP.

Effects of MCTP on cellularity of PEC (A) and BBC (B)
monolayers.

xiii

77

89

92

95

98

112

113

114

115

119

121

135

List of Figures (cont)

Figure 14:

Figure 15:

Figure 16:
Figure 17:

Figure 18:

Figure 19:

Figure 20:
Figure 21:

Figure 22:

Figure 23:

Figure 24:

Figure 25:

Figure 26:

Figure 27:

Figure 28:

Figure 29:

Angiotensin-converting enzyme activity of MCTP-
treated monolayer.

Angiotensin-converting enzyme activity expressed as
a function of cell number.

Effects of MCTP on endothelial cell number.
Effects of MCTP on cellular protein content.
Cellular DNA content of MCTP-treated PECs.
Effects of MCTP on incorporation of [3H]thymidine.
Effects of MCTP on incorporation of [3H]leucine.
Effects of MCTP on incorporation of [3H]uridine.

Effects of MCTP on cellularity of monolayers
grown on inserts.

Effects of MCTP on cellularity of monolayers grown
in flasks.

Effects of co-culture with MCTP-treated PECs on Swiss
3T3 cell number.

Effects of medium conditioned by MCTP-treated PECs
on Swiss 3T3 cell number.

Photomicrographs of REC monolayers 4 days after a single
administration of vehicle (A) or 150 uM MCTP (B).

Effects of MCTP on REC monolayer cellularity.

Release of lactate dehydrogenase activity from REC
monolayers.

Representative electrophoretic patterns of LDH isozymes
in RECs.

xiv

137

138
153
155
156
158
161

162

177

178

179

181

198

199

200

201

List of Figures (cont)

Em
Figure 30: Effects of MCTP on colony-forming efficiency of RECs. 204
Figure 31: DNA crosslink factor for RECs treated once with
MCTP at time 0. 205
Figure 32: Structures of mitomycin C, the reactive metabolite,
7-amino-1,2-aziridinomitosene, and dehydro-
monocrotaline (MCTP) 212

Figure 33: Photomicrographs of PEC monolayers 5 days after a single
administration of 0 (A), 0.1 (B), 1 (C) or 10 (D) uM MMC. 218

Figure 34: Photomicrographs of REC monolayers 5 days after a single
administration of 0 (A), 0.1 (B), 1 (C) or 10 (D) uM MMC. 219

Figure 35: Effects of MMC on PEC monolayer cellularity. 220
Figure 36: Effects of MMC on REC monolayer cellularity. 221
Figure 37: Release of LDH activity from PEC monolayers. 223
Figure 38: Release of LDH activity from REC monolayers. 224
Figure 39: Effects of MMC on colony-forming efficiency of PECs. 226
Figure 40: Effects of MMC on colony-forming efficiency of RECs. 227
Figure 41: DNA.cros(s)link factor for PECs treated once with MMC 228
at time .

Ab/Am
ACE
ACH
ADP
Ang I
Ang II
AN OVA
ANTU
ARDS
BALF
BEC

BK

BSA

CI

CM
CMF-HBSS

CO
DFMO
DHR
Di-I-Ac-LDL
DMEM
EC
ECGF
EDCF
EDGF
EDRF
EGF
EGIP
ELAM
ET

F CS
FGF
FH
G-CSF

LIST OF ABBREVIATIONS

Antibiotic/Antimycotic Solution

Angiotensin-Converting Enzyme

Acetyl Choline

Adenosine Diphosphate

Angiotensin I

Angiotensin II

Analysis of Variance

a-Napthylthiourea

Adult Respiratory Distress Syndrome

Bronchoalveolar Lavage Fluid

Bovine Pulmonary Artery Endothelial Cell

Bradykinin

Bovine Serum Albumin

Cardiac Index

Conditioned Medium

Calcium- and Magnesium-Free Hanks’ Balanced
Salt Solution

Cardiac Output

a-Difluoromethylornithine

Dehydroretronecine

Acetylated Low Density Lipoprotein

Dulbecco’s Modified Eagle’s Medium

Endothelial Cell

Endothelial Cell Growth Factor

Endothelium-Derived Contracting Factor

Endothelium-Derived Growth Factor

Endothelium-Derived Relaxing Factor

Epidermal Growth Factor

Endothelial Growth Inhibitory Protein

Endothelial Leukocyte Adhesion Molecule

Endothelin

Fetal Calf Serum

Fibroblast Growth Factor

Fawn-Hooded (Rats)

Granulocyte Colony Stimulating Factor

xvi

list of Abbreviations (cont.)

GM-CSF
GMP
GSH
GSH-DHP
GSH-DHR
HBSS
HDL
HETE
HODE
SHT
ICAM
IGF

Granulocyte-Macrophage Colony Stimulating Factor
Granule-Membrane Protein
Glutathione
Glutathionyldihydropyrrolizine
7-Glutathionyldehydroretronecine
Hanks’ Balanced Salt Solution
High-density Lipoprotein
Hydroxyeicosatetraenoic Acid
Hydroxyoctadecadienoic Acid
S-Hydroxytryptamine (Serotonin)
Intercellular Adhesion Molecule
Insulin-like Growth Factor
Interleukin

Antiserum Directed Against Lymphocytes
Lactate Dehydrogenase

LOW-dCl‘lSity Lipoprotein

Leukotriene

Monoamine Oxidase

Mixed Function Oxidase
Monocrotaline

Monocrotaline Pyrrole

Medium 199

Magnesium Aspartate

Mitomycin C

Norepinephrine

Nitric Oxide

Ornithine Decarboxylase

Platelet Activating Factor
Plasminogen Activator Inhibitor
Pulmonary Arterial Pressure
Antiserum Directed Against Platelets
Phosphate Buffered Saline
chhlorophenylalanine
Platelet-Derived Growth Factor
Porcine Pulmonary Artery Endothelial Cell
Platelet-Endothelial Cell Adhesion Molecule
Prostaglandin

Prostacyclin

Polymorphonuclear Cell (Neutrophil)
Primary Pulmonary Hypertension
Pulmonary Vascular Resistance

xvii

List of Abbreviations (cont.)

PZA
RBC
REC
RVH
SAM-DC
SAT
SDS
SFM
SMC
TCA

TGF

tPA

uPA
V-CAD
VCAM
VLDL

vWF

Pyrrolizidine Alkaloid

Red Blood Cell

Rat Pulmonary Endothelial Cell

Right Ventricular Hypertrophy
S-Adenosyl Methionine Decarboxylase
Spermine/Spermidine Acetyltransferase
Sodium Dodecyl Sulfate

Serum-Free Medium

Smooth Muscle Cell

Trichloroacetic Acid

Tissue Factor

Transforming Growth Factor
Thrombomodulin

Tumor Necrosis Factor

Tissue-type Plasminogen Activator
Thromboxane

Urokinase-type Plasminogen Activator
Vascular Cadherin

Vascular Cell Adhesion Molecule
Very Low-density LipOprotein

Oxygen Consumption

von Willebrand Factor

xviii

Chapter I

INTRODUCTION

I. E 1.1. ”11.!
The pyrrolizidine alkaloids (PZAs) are a group of structurally related

compounds produced by a wide variety of plant species. The PZAs share a common
nucleus, consisting of two fused, five-membered rings joined with a nitrogen atom at
the bridgehead (McLean, 1970). To date, PZA-containing species have been
identified in over 60 plant genera belonging to 13 unrelated families. More than 150
alkaloids of this class have been isolated and characterized (Bull et aL, 1968;
Huxtable, 1979; Mattocks et aL, 1986; Huxtable, 1989; Smith and Culvenor, 1981;
Mattocks, 1986). Many occur as free alkaloids and others as alkaloidal N-oxides;
cycling between these states may provide a redox system for the plant (Huxtable,
1980).

PZA-containing plant species are found world-wide (Bull et al., 1968; Cheeke,
1989; Smith and Culvenor, 1981), and poisoning of livestock (Bull et aL, 1968;
McLean, 1970) as well as of humans (Hill et aL, 1951; Huxtable, 1989) occurs as a
result of accidental or intentional ingestion of these plants. In the United States,
PZA-containing plants such as Senecia jacabaea, which contaminates millions of acres
of cropland in the Pacific Northwest, and several Cratalaria spp., prevalent in the
southeast, are responsible for significant loss of livestock (Heath, 1969; Snyder, 1972;
Huxtable, 1980). The toxic effects of PZAs, as well as target organ specificity,
appear to vary with animal species, age and sex, plant species consumed, duration of
exposure and total dose of alkaloid consumed (McLean, 1970; Culvenor et 01., 1976;

Huxtable, 1979; Shull et aL, 1976).

3

Toxic PZAs may affect a number of organ systems, but hepatic changes are
most commonly identified. These include veno-occlusive disease, hepatomegaly and
megalocytosis (Hill et aL, 1951; Schoental and Head, 1955; Bull et al., 1968; Bras et
aL, 1957; McLean, 1970; Petry et aL, 1984). Several of the PZAs are mutagenic and
genotoxic in a variety of systems (Williams and Mori, 1980; Yamanaka et aL, 1979;
Griffin and Segall, 1986; Petry et at, 1986; Takanashi et at, 1980; Styles et al., 1980),
and liver cell carcinomas (Cook et aL, 1950; Schoental et aL, 1954; Schoental, 1968)
and rhabdomyosarcomas (Allen et aL, 1975; Huxtable et aL, 1978), and
cardiopulmonary effects have been noted with some frequency under these conditions
(Turner and Lalich, 1965; Kay and Heath, 1969; Chesney and Allen, 1973d; Chesney
and Allen, 1973b; Chesney et aL, 1974a). In many cases, the subtlety of the lesions
and delayed develOpment of toxicity may complicate the diagnosis of PZA toxicoses
(Huxtable, 1979; Huxtable, 1989), however, neurological (Hooper et aL, 1974;
Mattocks et at, 1986), renal (Schoental and Head, 1955; Masugi et aI., 1965; Hayashi
and Lalich, 1967), pancreatic (Putzke and Persaud, 1976) and gastrointestinal (Hsu
et aL, 1974; Hooper, 1975) effects have been documented. The PZAs are potent
antimitotic agents, affecting liver cells and tissues with rapid cell turnover (Hsu et aL,
1973). Teratogenicity and embryotoxicity have also been reported (Peterson and
J ago, 1980)

Monocrotaline (MCT) is a PZA found in the seeds, leaves and stems
of plants of the genus Cratalan'a in the family Leguminasae (Neal et aL, 1935);
Cratalan'a spectabilis is the species most commonly encountered in the United States

(Heath, 1969; Kay and Heath, 1969). Of the many PZAs currently under

4

investigation, MCT is perhaps the most intensively studied, and the remainder of this

dissertation will be confined to a discussion of MCT.

11- Monmtaline
A General
Cratalan'a spectabilis (common name: rattlebox) and other MCT-
containing plant species are found throughout the tropical and subtropical regions
of the world, including the southeastern United States. Human exposure to MCT
occurs when these plants are eaten as vegetables, or when decoctions or extracts of
the leaves and stems are taken as herbal teas or remedies. Inadvertent exposure also
occurs when food grains contaminated with seeds of MCT»containing plants are
consumed. livestock may ingest Cratalan'a spp. while grazing or encounter them as
contaminants in grain and forage (Huxtable, 1989; Kay and Heath, 1969).
Epidemiologically, it is often difficult to demonstrate a direct
connection between consumption of MCT-containing plants and toxicity; however,
liver, lung and cardiotoxicity have been associated with ingestion of these plants by
a number of animal species, including horses (Stalker, 1884; Rose et aL, 1957; Cox
et aL, 1958), cattle (Sippel, 1964; Sanders et at, 1936), pigs (Emmel et al., 1935;
Peckham et aL, 1974), goats (Dickinson, 1980), poultry (Thomas, 1934; Allen et at,
1960; Allen et at, 1963; Sippel, 1964) and people (Bierer et aL, 1960; Huxtable, 1989;
Heath, 1969). Administration of MCT to experimental animals such as rats (Heath,
1969; Allen and Carstens, 1970; Kay et aL, 1967a; Lalich and Merkow, 1961; lalich,

1964; Masugi et aL, 1965; Roth et aL, 1981a), rabbits (Gardiner et aL, 1965), dogs

S
(Miller et al., 1978) and non-human primates (Allen et al., 1965; Raczniak et al.,

1979) also results in toxicity. Mice are less sensitive (Miranda er al., 1981b; Molteni
et al., 1989a),and guinea pigs, gerbils and hamsters appear to be resistant, to MCT
toxicity (Chesney and Allen, 1973a; Cheeke and Pierson-Goeger, 1983). Young
animals are generally more sensitive than older animals (Chesney and Allen, 1973b;
Stuart and Bras, 1957; Todd er al., 1985).

Chemically, MCT consists of a branched-chain, dicarboxylic acid,
monocrotalic acid, esterified to a retronecine nucleus [Figure 1] (Adams and
Rodgers, 1939). The alkaloid has a molecular weight of 325.3, and is a colorless
crystal with a melting point of 202-203°C (IARC Monograph). The structure of MCT
has been confirmed by infrared spectroscopy, nuclear magnetic resonance and mass

spectrometry (Culvenor and DalBon, 1964; Bull er al., 1968).

      

5 3

MONOCROTALINE MONOCROTALINE
P 7128 O L E

Figure 1: Structures of monocrotaline and monocrotaline pyrrole.
(* potential reactive centers)

B. Wu
As is the case with other PZAs, the parent alkaloid, MCT, is not toxic
but must undergo metabolic bioactivation (Mattocks, 1968). Mattocks and White
(1971) identified the mixed-function oxidase (MFO) system of the liver as the site of
metabolism of PZAs to hepato— and pneumotoxic species. Bioactivation is dependent
on the presence of NADPH and oxygen, is inhibited by carbon monordde and SKF-

525A, and is enhanced by phenobarbital pretreatment of animals (Mattocks and

White, 1971; White and Mattocks, 1971; Chesney er al., 1974b). The 1:2 double bond
of the pyrrolizidine nucleus, a structural requirement for toxicity, is necessary for
dehydrogenation of the parent alkaloid to reactive, pyrrolic derivatives (Mattocks,
1969; Mattocks, 1968; Culvenor et al., 1969; Huxtable, 1990a). In the case of MCT,
monocrotaline pyrrole (MCTP; dehydromonocrotaline) is the primary pyrrolic
metabolite [Figure 1]. Recently, cytochrome P450 isozyme IIIA-2 (PCN-E) has been
identified as the form of P450 responsible for conversion of PZAs to reactive pyrroles
(Williams et al., 1989). The hepatic MFO system (cytochrome P450 isozyme IIC11
[UT-A1) also produces N-oxide derivatives of the PZAs; however, these metabolites
do not appear to be responsible for the toxic effects of PZAs (Mattocks and White,
1971; Culvenor et al., 1970; Culvenor et al., 1976; Chesney et al., 1974b; Williams et
al., 1989). Liver microsomes of the guinea pig are high in esterase activity and ester
hydrolysis accounts for > 90% of MCT metabolism in this species;

dehydroretronecine (DHR), rather than MCTP, is the primary pyrrolic metabolite

7
generated by this species, and may explain the guinea pig’s resistance to PZA
intoxication (Dueker et al., 1992).

Although the liver is not the only organ to possess MFO activity, it is
thought to be the primary organ involved in conversion of MCT and other PZAs to
pyrrolic metabolites. Microsomes prepared from lung are unable to bioactivate
PZAs (Mattocks and White, 1971). Isolated, perfused livers and liver slices produce
toxic pyrroles when exposed to PZAs in vitra, but isolated, perfused lungs and lung
slices do not (Mattocks, 1968; LaFranconi er al., 1984; Gillis et al., 1978; Hilliker et
al., 1983c; LaFranconi and Huxtable, 1984). This has recently been confirmed in
tandem liver-lung preparations (Pan et al., 1991). Finally, pretreatment of animals
with phenobarbital preferentially induces hepatic, but not pulmonary, MFOs;
phenobarbital induction increases the lung and liver toxicities of MCT and other
PZAs, supporting the concept that the liver is the important locus for bioactivation
(Mattocks, 1968).

Until recently, little was known about the toxicokinetics of MCT. In
an early study, Hayashi (1966) followed the metabolism and excretion of [3H]MCT
after administration of a single, subcutaneous injection to rats and found rapid
accumulation of radioactivity in urine and bile. The parent alkaloid was the
predominant species identified in urine, whereas the biliary form appeared to be a
metabolite (Hayashi, 1966). More recently, Segall et a1. (1990) administered a single,
i.v. injection of [“C]MCT to male rats and found rapid elimination of radioactivity,
With 90% of the injected dose appearing in urine and bile by 7 hr. Most of the

urinary material was parent alkaloid; little N-oxide was identified. Most of the

8

biliary radioactivity was in the form of pyrrolic metabolites. Plasma levels of
radioactivity dropped rapidly by 7 hr, but red blood cell (RBC) radioactivity
remained relatively high. Enterohepatic recirculation was not a major feature of
MCT metabolism (Estep et al., 1991).

Hepatic metabolism of [“C]MCI‘ was examined in the isolated,
perfused rat liver in situ (Lame et al., 1991). N o retronecine and little N-oxide were
identified in liver perfusate, supporting the contention that ester hydrolysis and N -
oxidation are not major metabolic pathways for MCT. Dehydrogenation to MCTP
was identified as the major metabolic pathway. Secondary reaction of MCTP with
nucleophiles such as glutathione (GSH) resulted in the production of water-soluble
conjugates (e.g. 7-glutathionyldehydroretronecine [GSH-DHR] or
glutathionyldihydropyrrolizine [GSH-DHP]) and free monocrotalic acid. A number
of investigators have identified pyrrolic glutathione conjugates in the bile of rats
treated with MCT (LaFranconi et al., 1985; Buhler et al., 1990; Lame et al., 1990;
Huxtable et al., 1990b) and an N -acetylcysteine conjugate has been identified in rat
urine (Estep et al., 1990).

Segall and coworkers also examined tissue distribution and covalent
binding of a dose of ["C]MCT injected subcutaneously (Estep et al., 1991). F our
hours after injection, binding of radioactivity was evident in liver, RBCs, lung and
kidney, with maximum binding in liver. By 24 hr, bound radioactivity was decreased
in all organs, but substantial levels remained in RBCs, liver, lung and kidney.

Most evidence suggests that pyrrolic derivatives are the proximate toxic

metabolites of MCT and other PZAs. Pyrrolic metabolites of the PZAs are highly

9

reactive, bifunctional electrophiles, capable of binding to cellular macromolecules
(Mattocks, 1969; Mattocks and Bird, 1983b; Mattocks, 1972; White and Mattocks,
1972; Robertson et al., 1977). Covalently bound pyrroles have been identified in
tissues of animals treated with MCT and other PZAs (Mattocks and White, 1970;
Allen et al., 1972a; Mattocks, 1972; Lame et al., 1991; Mattocks and Jukes, 1990), and
the quantity of pyrrole found in tissues correlates with the degree of tissue injury
(Mattocks, 1972). Induction of the hepatic MFO system by pretreatment of animals
with phenobarbital or other inducers results in enhanced production of pyrrolic
metabolites and more severe hepato- and pneumotoxicity following MCT
administration, while inhibition of hepatic MFO does the opposite (Allen et al.,
1972a; Tuchweber et al., 1974; Chesney et al., 1974b). Finally, administration of
chemically synthesized pyrroles (e.g., MCTP) to animals results in toxicity which is
similar to that caused by the parent alkaloid (Mattocks, 1968; Butler, 1970a; Butler
et al., 1970b; Bruner et al., 1983).

Although MCTP formed by the liver is a convincing proximate toxicant
for this organ, a number of investigators are less convinced of its role in MCT
pneumotoxicity. MCTP is highly reactive, and has a half-fife of 3.5-5 seconds in
aqueous solution (Mattocks and Bird, 1983a; Bruner er al., 1986). Mattocks et a1.
(1990) demonstrated that a small amount of reactive MCTP does escape from the
liver into the circulation; whether MCTP can then survive the trip from the liver to
the lung in the aqueous milieu of the blood with its alkylating functions intact is
unknown. It has been proposed that some of the toxic effects of PZAs are due to

secondary metabolites or breakdown products of primary pyrroles. However, MCTP

10

rapidly becomes non-toxic in aqueous media, and its pneumotoxic effects are not
influenced by the presence of hepatic MFO inducers or inhibitors (Bruner et al.,
1986). Also, although the secondary pyrrolic metabolite, DHR, has been identified
in urine of MCT-treated animals, it lacks the potency required to be the pneumotoxic
metabolite (Hsu et al., 1973).

Conjugation with GSH or mercapturic acid could allow reactive
metabolites of MCT to survive in the aqueous environment of the bloodstream
(Huxtable et al., 1990b; Estep et al., 1990). GSH-DHR appears to be pneumotoxic
to rats when administered at doses comparable to MCT (75 mg/kg) (LaFranconi et
al., 1985; Huxtable et al., 1990b). However, these doses are much higher than
pneumotoxic doses of MCTP (1-5 mg/kg) (Bruner et al., 1986; Pan et al., 1992).
Doses of 12-24 mg of the GSH conjugate of MCT / kg or 12 mg of the cysteine
conjugate/kg are not pneumotoxic in the rat (Pan et al., 1992). It has also been
suggested that RBCs serve as transporters for reactive MCT metabolites. RBCs
appear to sequester MCT-derived [“C], which is then capable of covalent interaction
with lung tissue. In addition, RBCs stabilize and augment the transport of reactive
MCT metabolites from liver to lung (Estep et al., 1991; Pan et al., 1991).

In summary, MCT is metabolized by the hepatic MFO system to a
reactive, pyrrolic metabolite(s). At least one primary metabolite is MCTP. It has
been proposed that MCTP enters the bloodstream and proceeds to the lung, perhaps
bound to RBCs, where it then binds and causes injury. Indeed, MCTP injected
intravenously can reproduce the pneumotoxic effects of MCT itself. Some of the

MCTP formed from MCT in the liver is also converted to GSH-conjugated forms.

11

These are secreted into bile; they also enter the bloodstream, since mercapturic acid
conjugates have been identified in urine. GSH-conjugated metabolites of MCT are

also pneumotoxic but only at relatively high doses.

C. mm
The pathophysiology of MCT has been studied extensively. MCT has
been administered in a variety of forms, ranging from the addition of crushed C.
spectabilir seeds to the diet, to injection of chemically synthesized MCTP. MCT has
been given in food and drinking water, applied to the skin, and injected
subcutaneously and intravenously. It has been given as a single administration, or
chronically over the course of several weeks. In many cases, the total dose of
chemical administered can only be roughly estimated. Such diversity in experimental
design makes detailed comparison of the findings difficult. Although the form of
MCT and mode of administration influence the time course and extent of the
toxicity, the pathologic changes seen are qualitatively quite similar. Therefore, to
facilitate discussion of the biological effects of MCT, 1.) specific experimental
parameters will not be described unless they provide clarification, and 2.) the
designation "MCT' will refer to C. spectabilis as well as the purified alkaloid, unless
otherwise noted. Studies utilizing the metabolite, MCTP, will be incorporated as
deemed appropriate, but will be discussed more specifically in Section 11E.
1. Hepatotum'u
The acute response of the liver to MCT treatment is

centrilobular hemorrhagic necrosis with dilation, congestion and thrombosis of portal

12

hepatic veins. The injury progresses to a veno-occlusive lesion (Bras and Hill, 1956),
with consequent development of portal hypertension (Harris et al., 1942; Schoental
and Head, 1955; Turner and Lalich, 1965; Merkow and Kleinerman, 1966b; Allen
and Carstens, 1968). Hepatic fibrosis, cirrhosis and ascites have also been reported
(Peckham et al., 1974; Allen et al., 1963; Emmel et al., 1935). liver function is
impaired, as indicated by decreased total serum proteins, a shift in albumin:globulin
ratio (Allen et al., 1963; Allen and Chesney, 1972b), increased prothrombin time
(Rose et al., 1945; Schoental and Head, 1955), and depressed excretion of
indocyanine green (Roth et al., 1981a); however, hepatic bile production is not
altered (Roth et al., 1981a). Plasma glutamic pyruvic transaminase activity is
increased (Roth et al., 1981a). With severe, chronic liver disease, ammonia levels in
blood may increase, resulting in hepatic encephalopathy (Rose et al., 1957).
Megalocytic hepatic parenchymal cells, a hallmark of chronic PZA intoxication (J ago,
1969), are also evident in MCT-treated animals (Turner and lalich, 1965 ; Allen and
Chesney, 1972b).
2. W911!

MCT treatment results in injury to the renal vasculature of rats,
primarily in the glomerular region of the kidney, and is characterized by glomerular
swelling and necrosis. A single injection of MCT produces hyaline thrombosis of
glomerular capillaries and afferent arterioles, and medial hypertrophy is evident in
interlobular arteries (Hayashi and Lalich, 1967). Similarly, chronic administration
of C. spectabilis seeds results in marked vascular and glomerular changes but also

causes degenerative changes in the proximal tubules of the kidney (Masugi er al.,

13

1965). Renal hemosiderosis occurs, perhaps as a result of intravascular hemolysis
associated with microvascular damage (Hayashi and Lalich, 1967; Schoental and
Head, 1955). Glomerular and tubular lesions have also been reported in swine
(Peckham et al., 1974; Emmel et al., 1935).

In addition, several indices of renal function are altered with
MCT treatment. Blood urea nitrogen is increased in animals receiving MCT in the
drinking water, suggesting impaired glomerular filtration. Kidney slices from these
animals also demonstrate altered ability to accumulate organic anions and cations,
a change consistent with proximal tubular damage (Roth et al., 1981a).

3. 91mm

MCT is mutagenic in several strains of Drasaphila (Cook and
Holt, 1966), and is a hepatic carcinogen in rats (Newberne and Rogers, 1973;
Shumaker et al., 1976). Administration of MCT’s secondary pyrrolic metabolite,
DHR, also results in neoplastic transformation (Shumaker et al., 1976), and causes
rhabdomyosarcomas in rats (Allen et al., 1975). Both MCTP and DHR serve as
initiators in mouse skin tumor formation when croton oil is used as a promotor
(Mattocks and Cabral, 1982). MCTP, when administered at high doses, causes a
marked inflammatory response and may serve as a complete carcinogen (Mattocks
and Cabral, 1982; Hooson and Grasso, 1976). Interestingly, the metabolites of MCT
are potent antiproliferative agents in cultured hepatocytes (Mattocks and Legg, 1980)
and other cell lines (Reindel and Roth, 1991; Reindel et al., 1991), and this

characteristic of MCT toxicity may tend to inhibit tumor formation in viva.

14
Cancer metastasis is enhanced in lungs of MCT-treated animals

(Vincic et al., 1989). This has been attributed to alterations in the pulmonary
vasculature compatible with increased tumor cell retention.
4- W

Although the major effects of MCT intoxication involve the liver
and lung, a number of other pathologic features have been reported. Whereas some
of these changes may be due to direct effects of MCTP or other metabolites, others
are secondary responses to changes in hepatic or pulmonary function.

Animals treated with MCT fail to thrive and do not gain weight
like paired control animals (Schoental and Head, 1955; Merkow and Kleinerman,
1966a; Peckham er al., 1974; Roth et al., 1981a; Meyrick and Reid, 1979). Necrosis
of the thymic cortex has been reported in mice treated with a single, high dose of
MCT (Harris et aL, 1942). Generalized atrophy of lymphoid tissue and gastric
ulceration are seen following oral administration of C. spectabilis seeds to pigs
(Peckham et al., 1974); gastric mucosa] erosions are also reported in rats (Turner and
Lalich, 1965). Mice receiving MCT by oral gavage for 14 days show suppressed
immune function in the absence of marked liver or lung toxicity (Deyo and Kerkvliet,
1990). Vascular lesions of the mesentery and pancreas occur infrequently in animals
receiving MCT for an extended period of time (McLean, 1970; Schoental and Head,
1955).

15
D. W

Although a broad spectrum of lesions are associated with MCT
intoxication, the cardiopulmonary effects of this compound are perhaps the most
notable (McLean, 1970). Animals treated with high doses of MCT generally
experience massive liver injury. However, when certain species (including the rat)
are given lower doses of this alkaloid, liver damage is minimal but the lungs exhibit
progressive injury (Valdivia et al., 1967a; Roth et al., 1981a). As is the case in both
the liver and kidney, a prominent feature of MCT pneumotoxicity is vascular injury.
C. spectabilis has been dubbed the "pulmonary hypertension plant" in
acknowledgment of the frequent development of chronic pulmonary vascular disease
following exposure to either the plant itself or its purified alkaloid, MCT (Kay and

Heath, 1969).
The pulmonary toxicity of C. spectabilis was first reported by Schoental and
Head in 1955. However, a systematic investigation of the pneumotoxic effects of this
plant by lalich and Merkow in 1961 really set the stage for future work in this arena.
Chronic cardiopulmonary changes in response to C. spectabilis or MCT have been
reported in swine (Emmel et al., 1935; Peckham er al., 1974), dairy calves (Pringle
et al., 1991) and non-human primates (Chesney and Allen, 1973b), but the vast
majority of reports on MCT pneumotoxicity describe studies performed in rats. As
the pulmonary changes follow a similar pattern in all affected species, the following

discussion will be confined to findings in the rat model.

l5?

16
1. Karma
a. i h e

Rats treated with pneumotoxic doses of MCT experience
tachypnea and dyspnea at rest. They become weak and listless, have poor appetites,
and fail to gain weight. Respiratory distress and cyanosis are evident in the terminal
stages of intoxication. The lungs of these animals are heavy and edematous, and
subpleural and parenchymal hemorrhage and regional atelectasis are evident.
Pulmonary vessels are dilated and congested, and occasional thrombi are noted.
Pulmonary lymphatic vessels are also dilated. (Schoental and Head, 1955; Masugi et
al., 1965; Merkow and Kleinerman, 1966a; Lalich and Merkow, 1961; Turner and
lalich, 1965; Merkow and Kleinerman, 1966b)

Cardiac enlargement occurs later in the course of toxicity.
This cardiomegaly involves only the right ventricular free wall, and does not affect
the septum or left ventricle (Kay and Heath, 1966; Kay et al., 1967a; Hislop and
Reid, 1974; Meyrick et al., 1980; Roth, 1981b; Ghodsi and Will, 1981; Hilliker et al.,
1982; Molteni et al., 1985; Gillis et al., 1978; Roth et al., 1981a).

b. MW

As stated previously, MCT causes injury to the
vasculature of the lung. Alterations in pulmonary endothelial cells (ECs) are among
the earliest changes noted in vessels of MCT-treated animals, occurring as early as
24 hours (Valdivia et al., 1967a). ECs are prominent, often protruding into the
Vascular lumen, and vesicular "blebbing" is reported (Butler, 1970a; Valdivia et al.,

1967a; Merkow and Kleinerman, 1966b; Valdivia et al., 1967b; Turner and Lalich,

17
1965; Rosenberg and Rabinovitch, 1988; Kay et al., 1969). ECs from treated animals

have marked nuclear enlargement, an increased number of cytoplasmic organelles
and prominent pinocytotic vesicles (Meyrick and Reid, 1982; Kay et al., 1969;
Rosenberg and Rabinovitch, 1988). Capillary ECs show alternate areas of thickening
and thinning, and swollen, hypertrophic endothelial cells occasionally appear to
occlude capillaries (Valdivia et al., 1967b; Rosenberg and Rabinovitch, 1988; Lalich
and Merkow, 1961; Meyrick and Reid, 1982; Turner and Lalich, 1965 ; Hislop and
Reid, 1974; Vincic et al., 1989; Plestina and Stoner, 1972; Chesney and Allen, 1973c).
ECs in alveolar wall vessels incorporate increased amounts of [3H]thymidine at 7 and
21 days in rats receiving C. spectabilis seeds in their diet (Meyrick and Reid, 1982);
vessels were not examined for this phenomenon earlier in the course of treatment.

Extension of smooth muscle into small (<25 u)
pulmonary arterioles which are normally nonmuscular is another early change, noted
as soon as 3 days after treatment with MCT (Hislop and Reid, 1974; Turner and
Lalich, 1965; Meyrick and Reid, 1979; Langleben and Reid, 1985). Medial
hypertrophy of the small pulmonary arteries is consistently observed in lungs of
MCT-treated animals (Heath and Kay, 1967; Turner and Lalich, 1965; Masugi et al.,
1965; Hislop and Reid, 1974; Hayashi et al., 1967; Merkow and Kleinerman, 1966b;
Kay and Heath, 1966; Kay and Heath, 1969; Lalich et aL, 1977). Increased thickness
of the media is primarily due to increased smooth muscle and an increased number
of elastic laminae; however, vasoconstriction has occasionally been suggested as a
contributor as well. Extension of smooth muscle into small arterioles and thickening

of the media in small pulmonary arteries are changes commonly associated with

18
pulmonary arterial hypertension (Harris and Heath, 1962). Medial smooth muscle

cells (SMCs) contain increased numbers of mitochondria, increased rough
endoplasmic reticulum and prominent golgi complexes, suggesting enhanced
metabolic activity consistent with cellular hypertrophy (Merkow and Kleinerman,
1966a). Medial SMCs also show a small increase in [3H]thymidine uptake indicative
of increased DNA synthesis (Meyrick and Reid, 1982). This may represent SMC
proliferation, but could also represent DNA repair or increased DNA synthesis
without subsequent cell division.

Diffuse, necrotizing arteritis of small muscular pulmonary
arteries and arterioles is described by Merkow and Kleinerman (1966a). They
differentiate this lesion from that reported by Turner and Lalich (1965) as being
fibrinoid in nature, without the presence of leukocytes. A number of investigators
describe arteritis and periarteritis of small, medium or larger pulmonary arteries
(Lalich and Merkow, 1961; Turner and Lalich, 1965; Merkow and Kleinerman,
1966a; Masugi er al., 1965; Lalich, 1964; Lalich and Ehrhart, 1962; Merkow and
Kleinerman, 1966b) . These inflammatory lesions exhibit edema of the vascular wall,
with lymphocytes, plasma cells, eosinophils and polymorphonuclear leukocytes
(PMNs) present in some cases. Proliferation of fibroblasts frequently accompanies
the inflammation. While this lesion is often associated with pulmonary hypertension,
the occasional absence of this inflammatory change suggests it may be related instead
to a sudden increase in pulmonary vascular pressure (Kay and Heath, 1966; Heath,
1969). Bronchial arteries are not typically affected by MCT treatment (Merkow and r

Kleinerman, 1966a).

l9

Reported changes in the main pulmonary artery include
thickening of the media, widening of the subendothelial space (with fluid, cell debris
and collagen) and focal muscle necrosis and elastolysis. Alterations in polarity, shape
and organelle content of smooth muscle cells, suggestive of a change from a
contractile to a secretory phenotype, occur as well. The adventitia is also thickened
with increased collagen and variable amounts of cardiac muscle (Guzowski and
Salgado, 1987).

Pruning of the pulmonary vascular tree follows treatment
with C. spectabilzls seeds (Hislop and Reid, 1974; Meyrick and Reid, 1979). Blockage
of small, peripheral vessels by swollen ECs or thrombi is thought to contribute to a
decrease in cross-sectional area for pulmonary blood flow, thus resulting in increased
pulmonary pressure. Kay et al. (1982c) disagree with these findings, and suggest that
the techniques used to enumerate vessels in the preceding studies led to misleading
results. Kay and coworkers utilized MCT in their studies, rather than C. spectabilir
seeds, and this difference might account for the difference in their results. However,
a recent cast corrosion study performed by Schraufnagel and Schmid (1989) reported
a reduced pulmonary capillary density in MCT-treated animals.

MCT-induced angiogenesis has also been reported in the
lung (Schraufnagel, 1990). New vessels are found primarily at the pleural surface
and in the perivascular connective tissues of the bronchial vessels, areas reported to
accumulate mast cells following MCT treatment (Takeoka et al., 1962).

Neovascularization does not occur in the alveolar capillary beds.

20

Prominent fibromuscular pads are evident in the intima
of small pulmonary veins of animals receiving C. spectabilis seeds in the diet. These
may be large enough to impede pulmonary venous outflow (Kay and Heath, 1966).
The SMCs in these vessels swell, evaginate into the intima, and press into the ECs
(Smith and Heath, 1978). Such changes have not been reported in MCT-treated rats
and could reflect differences in the rate or character of the development of toxicity
or may be due to other contaminating substances present in C. spectabilis seeds.

Blockages of vessels by intraluminal thrombi and fibrin
are associated with MCT pneumotoxicity, and pulmonary capillaries may be occluded
with platelets (Turner and Lalich, 1965; Lalich and Ehrhart, 1962; Merkow and
Kleinerman, 1966b; Masugi et al., 1965; Merkow and Kleinerman, 1966a; Valdivia
et al., 1965; Valdivia et al., 1967a; Kay and Heath, 1966). Vascular thrombosis has
been reported both early (Valdivia et al., 1967a; Valdivia et al., 1967b) and late
(Turner and Lalich, 1965; Merkow and Kleinerman, 1966b; Hayashi et al., 1967) in
the course of MCT pneumotoxicity, and its role in the development of hemodynamic
changes is controversial.

In addition to vascular alterations, changes occur in the
parenchymal epithelium of the lungs with MCT treatment. Valdivia et al. (1967a)
have shown that MCT treatment affects all cells and the elastic membranes of the
alveolar wall, and these changes precede the appearance of many of the vascular
changes. Hyperplasia of the epithelial cells of the alveoli and terminal bronchioles
has been reported (Schoental and Head, 1955; Valdivia et al., 1967a; Kay et al., 1969;

Sugita et al., 1983a), and extension of bronchiolar epithelium to the alveolar ducts

21
and alveoli occurs as well (Kay and Heath, 1966). Wilson and Segall (1990) found

cyto- and karyomegaly of alveolar type II pneumocytes in the absence of obvious
cytolytic changes. Type 11 cells were increased in volume and nuclear diameter, but
decreased in number. They also had an increased thymidine labeling index. This is
reminiscent of changes seen in hepatic parenchymal cells following PZA
administration (McLean, 1970) and may be related to an inability of these cells to
divide in the face of a stimulus for proliferation.

There are accumulations of mast cells (Kay et al., 1967b)
and enlarged macrophages with foamy cytoplasm (Kay and Heath, 1966; Sugita et al.,
1983a) in the interstitium of the lungs of treated animals. Edema fluid is evident in
the interstitium and in the alveolar spaces as early as 4 hours after treatment with
MCT and hemorrhage into the alveoli is occasionally seen (Masugi et al., 1965;
Schoental and Head, 1955 ; Valdivia et al., 1967a). Interstitial fibrosis is a later
development (Kay et al., 1969; Butler, 1970a; Meyrick and Reid, 1979; Meyrick et al.,
1980). These are collectively referred to as exudative lesions, and their presence
suggests a loss of vascular integrity. Exudative lesions typically occur when capillary
permeability and vascular pressure are increased (Wilson and Segall, 1990; Heath,
1969; Valdivia er al., 1967b; Molteni et al., 1984; Sugita et al., 1983b).

The left ventricle of the heart is morphologically normal
in MCT-treated animals. In the right ventricle, changes consistent with hypertrophy
and increased energy production are evident. The number and size of mitochondria
increases by 14 days posttreatment (Kajihara, 1970). As right ventricular

enlargement progresses, most cardiac myocytes become hypertrophic with large

22

nuclei. Increased RNA synthesis is evidenced by changes in the golgi apparatus,
dilation of the endoplasmic reticulum, and an increased number of free ribosomes
in the cytoplasm. An increase in RNA is also indicated by a decrease in the
DNA/RNA ratio in the right ventricle (LaFranconi et al., 1984). Protein synthesis,
and protein and collagen content, of the right ventricle increase with the ventricular
enlargement (Huxtable et al., 1977; LaFranconi et al., 1984).

Small accumulations of lymphocytes and monocytes
between myocardial fibers and interstitial edema have been reported (Lalich and
Merkow, 1961; Hayashi and Lalich, 1967; Werchan et al., 1989). However, right
ventricular myocytes show no histologic evidence of failure; the arrangement and
mean diameter of myofilaments are not different from control animals (Loskutoff et
al., 1983).

A report by Blaustein et al. (1965) identified coronary
vascular changes. Intimal atheromatous plaques were evident in Wistar rats treated
with MCT. Coronary vessel effects as well as subendocardial fibrosis of the right

ventricle have also been reported in monkeys (Chesney and Allen, 1973b and c).

2. W
The effects of MCT on pulmonary function were examined by
Gillespie et al. (1985). 20 days after receiving a single injection of MCT, rats showed
evidence of mechanical, ventilatory and gas exchange deficits. Total lung capacity
and residual volume were decreased, as were respiratory frequency, tidal volume, and

dynamic and quasistatic compliance. There was a marked increase in pulmonary

”CC";
15.5. N

X‘
5;)

u

:'§-.
\

rm:

re.‘
0"”

“‘C

I

.‘A. r

“1’!

”ALI"
‘H

23

resistance, and the coefficient of diffusion for carbon monoxide was reduced. These
changes in function were associated with morphologic changes in the alveolar wall:
alveolar septae were thickened, with edema fluid, cellular debris and many mast cells
present in the interstitial space, and alveolar integrity was compromised. Functional
and morphological findings suggested a severe, combined restrictive and obstructive
airway disorder.

Further work in this area demonstrated that changes in the lung
parenchyma and the resultant ventilatory dysfunction occurred early in the course of
MCT intoxication, preceding the development of pulmonary vascular changes (Lai
et al., 1991). Although these findings suggest a potential causative role for
hypoventilation and hypoxia in the deve10pment of subsequent vascular alterations,
the observation that arterial oxygen tension is reduced only very late in the course
of toxicity would argue that this is not the case (Meyrick et aL, 1980). It may be that
the changes in the pulmonary parenchyma and vasculature are coincident, both
contributing to the more terminal cardiopulmonary changes seen following MCT
administration.

Werchan et al. (1989) evaluated right ventricular function in
animals with MCT-induced pulmonary hypertension, which produces a chronic
pressure overload on the right heart. The hypertrophic right ventricle demonstrated
enhanced performance with no evidence of failure at 2-5 weeks posttreatment, which
is in agreement with previous morphologic findings. In addition, Bruuer et a1. (1983)

found that although MCTP causes changes in the electrocardiogram consistent with

24

right ventricular enlargement, there is no evidence of dysfunction at 14 days
posttreatment.
3. V 1 r '
The angiotoxic effects of MCT have been emphasized previously.
The pulmonary lesions of MCT characteristically center on the vasculature, and
include a vascular injury phase and a subsequent vascular remodeling phase. The
structural alterations which follow MCT treatment have been discussed above
(Section II.D.1.b.). This section will deal with the hemodynamic changes which are
associated with MCT pneumotoxicity, including alterations in vascular reactivity.
2» W
The characteristic pathology of the pulmonary vasculature
observed after MCT treatment, together with the consistent development of right
ventricular enlargement late in the course of intoxication, were accepted as evidence
of elevated pulmonary arterial pressure long before hemodynamic studies were
performed in this model. Further indirect evidence of pulmonary hypertension was
added with reports of increased right ventricular pressure in MCT- and MCTP-
treated animals (Kay et al., 1967a; Chesney et al., 1974a). Eventually, direct
measurement of pulmonary arterial pressure (PAP) in C. spectabilir-treated rats
confirmed the delayed development of pulmonary hypertension and its relationship
to right ventricular hypertrophy (RVH) in the MCT model (Huxtable et al., 1977).
A thorough study of the hemodynamic changes in
conscious rats following ingestion of C. spectabilis was performed by Meyrick et al.

(1980). These authors found significant increases in oxygen consumption (V0,) and

25

cardiac index (CI) after 7 days of treatment; only the latter change was persistent.
Changes in V02 and CI were correlated with the muscularization of pulmonary
arterioles, and the increase in CI was attributed to dilatation and recruitment of
vessels in response to vascular inflammation and damage.

PAP was increased by 14 days, coincident with the
increase in medial thickness of small pulmonary arteries. Pulmonary vascular
resistance was markedly increased by 33 days. Changes in the larger vessels of the
lungs occurred late in the course of toxicity, coincident with a loss of peripheral
vessels and evidence of RVH. No evidence of arterial hypoxemia was evident until
long after the development of pulmonary hypertension. From this work, the authors
suggested that precapillary vascular remodeling was responsible for the increased
PAP seen after treatment with MCT.

b. mm

It has been hypothesized that altered pulmonary vascular
responsiveness could contribute to the increases in PAP seen in MCT
pneumotoxicity. Vasoconstriction is thought to be the primary vascular change in
some models of pulmonary hypertension (Voelkel and Reeves, 1979). Although this
may not be the case with MCT, several investigators have examined changes in
vascular reactivity in response to vasoactive mediators. Gillespie et al. (1986) found
an early, transient increase in pressor response of isolated, perfused lungs from MCT-
treated animals to both angiotensin II (Ang II) and hypoxia, but concluded that this
was unlikely to contribute to sustained increases in arterial pressure. This course is

somewhat different than that found by Hilliker and Roth (1985) using perfused lungs

26
from rats treated with MCTP (see below). Altiere et al. (1986) found vessels from

MCT-treated rats demonstrated an early, transient increase in responsiveness to Ang
II, norepinephrine (NE) and KC] and reduced responsiveness by 14 days after MCT
treatment; relaxation in response to acetylcholine (ACH) and isoprotereuol were also
decreased late in the course of injury. Shubat and cowbrkers (1987) found a
progressive decrease in contractility of isolated pulmonary vessels in response to 5-
hydroxytryptamine (SHT), NE and K+ which became significant with the
establishment of pulmonary hypertension. A generalized reduction in responsiveness
may represent direct damage to the vascular wall, or could reflect compensatory
changes in response to increased PAP. Decreased vascular responsiveness coincides
with thickening of the small pulmonary arteries, and with greater vascular compliance
and reduced ability to generate force (Altiere et al., 1986; Langleben er al., 1988).

Recent work by Ito and coworkers (1988a) confirmed the
decreased responsiveness to ACH, and reported enhanced contraction in response
to FGF”. However, these investigators found no difference in response to Aug I or
Ang 11, nor did they detect a change in responsiveness to NE or K+.

Differences in the pathology of large and small
pulmonary vessels after MCT treatment have led to investigation of differences in the
SMCs of these vessels during pulmonary hypertension. SMCs of the main pulmonary
artery are depolarized as compared to non-hypertensive controls, due to increased
chloride conductance of the cell membrane. SMCs of smaller pulmonary arteries
(100-200 11) are hyperpolarized, and this is attributed to an increase in Na+ /K+

pump activity. SMCs of small pulmonary arteries are those which undergo

27
hyperplasia and / or hypertrophy during MCT toxicity, and the hyperpolarization may

be linked to co-transport of N a+ and amino acids in the process of cell growth
(Gillespie et al., 1986).

Subsequent studies by Shubat et a1. (1990) on larger (500-
800 u) arteries isolated from MCT-treated animals demonstrated a diminished
contractile response to NE and SHT and evidence for dysfunction of the N a+ /K+
pump by 4-5 days posttreatment. This pump may be a relatizely early target of MCT

toxicity in SMCs, and alterations in activity may be a function of vessel diameter.

4. ' ' l h n
Evidence exists for altered biochemical function in the lung after
MCT treatment. Although studies emphasizing pulmonary pathology reported
changes consistent with altered function, it was not until the late 19705 that the
nonrespiratory function of the lung was quantitated in MCT-treated animals.

A number of investigators have demonstrated that
clearance of SHT by the lungs is reduced, and this reduction is accompanied by
decreased release of metabolite. This may reflect a specific decrease in SHT uptake
by pulmonary ECs rather than a defect in metabolism, as monoamiue oxidase
(MAO) activity is not impaired; however, changes in the distribution of pulmonary
blood flow can also result in decreased SHT uptake (Huxtable et al., 1978; Gillis et
al., 1978; Roth et al., 1981a; Hilliker et al., 1983c). Changes in SHT clearance occur
coincident with markers of pulmonary vascular leak, such as increased lung weight-to-

body weight ratio and increased release of lactate dehydrogenase (LDH) into cell-

28
free, bronchoalveolar lavage fluid (BALF) (Roth et al., 1981a; Roth, 1981b; Roth et

al., 1981a).

Removal of NE by the pulmonary vasculature of MCT-treated
animals is more controversial. Uptake of NE is also a carrier-mediated uptake
process at the endothelial surface, and NE is metabolized by the same MAO system
as is SHT (Roth, 1985). Gillis et al. (1978) and Hilliker et al. (1984) found decreased
removal of NE by lungs isolated from MCT-treated animals; however, Huxtable and
coworkers reported that N E transport was unaffected by MCT treatment (Huxtable
et al., 1978). Discrepancies in these results may be due to different lung perfusion
techniques.

There is reduced degradation of prostaglandins in the pulmonary
circulation after MCT treatment (Ito et al., 1988b; Ito et al., 1988a). This is true for
exogenously administered PGE, and FGF,“ and, presumably, for endogenous
prostaglandins as well. Prostaglandin clearance is also a function of the pulmonary
endothelium and requires uptake into the cell. Whether the uptake mechanism is
impaired or metabolism is inhibited is not known.

The loss of relaxation of pulmonary vessels from MCT-treated
animals in response to ACH is suggestive of a defect in generation of endothelium-
derived relaxing factor (EDRF) (Ito et al., 1988a). The reactivity of vascular smooth
muscle is unimpaired in these vessels, suggesting that an endothelial defect may be
responsible for the lack of response to ACH. Generation of EDRF has not been

investigated by direct means in the MCT model of lung injury.

29
The lung vasculature, specifically the endothelial surface, is

important in generation of a number of factors involved in hemostasis. Molteni et
al. (1984,1989b) found increased production of prostacyclin (PGIZ) and decreased
fibrinolytic activity in lung homogenates from MCT-treated rats. Although the
decrease in fibrinolytic activity occurred relatively late in the course of lung injury,
PGI2 levels rose progressively throughout the treatment period. The authors
concluded that progressive EC damage in the lungs of MCT-treated animals resulted
in these alterations, but they were uncertain whether these were primary effects of
the toxicaut or were secondary changes occurring in response to increased PAP. The
production of thromboxane A2 (TXAZ) is also increased in MCT-treated animals
(Molteni et al., 1988a; Molteni et aL, 1989b).

Synthesis of RNA and protein is increased in the lungs and right
ventricle of MCT-treated animals, and the DNA:RN A ratio is decreased (LaFranconi
et al., 1984). There is an increase in absolute protein content of these organs
(LaFranconi et al., 1984; LaFranconi and Huxtable, 1983). In the lung, the largest
increase in mass is in the cytoplasmic proteins; whereas in the heart, the largest
increase is in matrix proteins, including collagen. Biochemical changes of this nature
support the observation that these organs respond to MCT treatment with
hypertrophy, but whereas the lungs respond to direct chemical damage, the right
ventricle of the heart responds to an increased work load caused by pulmonary
hypertension (LaFranconi et al., 1984).

In contrast to the functions mentioned above, 5’-nucleotidase activity

is unaffected by MCT lung injury (Huxtable et al., 1978), and the effects on

30

angiotensin-converting enzyme (ACE; kininase II) are controversial. [See Chapter

II for a discussion of this subject]

The metabolic requirement for MCT toxicity complicates the
investigation of its effects. Species differences in hepatic metabolism (Cheeke and
Pierson-Goeger, 1983; Shull et al., 1976) and individual differences in level of
activating enzyme (Allen et al., 1972a; Tuchweber et al., 1974; Chesney et al., 1974b)
can influence the type and amount of metabolites produced; therefore, administration
of a carefully calculated dose of purified MCT could theoretically produce variable
effects in viva. In an effort to circumvent these difficulties, some investigators have
used chemically synthesized MCTP in their studies.

MCTP, a primary pyrrolic metabolite of MCT, is chemically reactive
and, when injected into animals, causes injury in the first vascular bed it encounters.
Subcutaneous injection of MCTP results in local transient necrosis and subsequent
endothelial injury (Hooson and Grasso, 1976). Administration via the mesenteric
vein results in liver injury (Butler et al., 1970b). Whereas a single, large dose of
MCTP (> 10 mg/kg) injected into the tail vein of rats results in acute, fulmiuant
pulmonary edema and death (Hurley and J ago, 1975; Plestina and Stoner, 1972),
similar administration of a low dose of MCTP (2-5 mg/ kg) results in pulmonary
vascular injury which is similar to that seen after administration of MCT (Butler et
al., 1970b; Bruner et al., 1986; Bruner et al., 1983). Administration of a low dose of

MCTP via the tail vein does not cause detectable injury to the liver or other organs.

31

The time course of MCTP pneumotoxicity is somewhat more rapid than
that seen after MCT but follows a similar pattern (Bruner et al., 1986; Schultze et al.,
1990). Early injury is characterized by progressive vascular leak, pulmonary edema
and congestion (Butler et al., 1970b; Plestina and Stoner, 1972; Reindel et al., 1990;
Schultze et al., 1990). Changes in endothelium, alveoli, the pulmonary interstitium
and vessel walls have been reported which are analogous to MCT-induced changes
(Butler, 1970a; Chesney et al., 1974a; Reindel et al., 1990; Butler et al., 1970b;
Raczniak et al., 1979; lalich et al., 1977). Similarly, fibrin and platelet thrombi have
been noted in pulmonary arteries, arterioles, capillaries and veins (Chesney et al.,
1974a; Lalich er al., 1977).

Increased LDH activity and protein are evident in BALF from MCTP-
treated rats (Bruner et al., 1983; Hilliker er al., 1984; Reindel et al., 1990). Removal
of biogenic amines is impaired in isolated lungs and lung slices from MCTP-treated
animals, suggesting similar changes in biochemical functions of the lung (Hilliker et
al., 1983a; Hilliker et al., 1984). Development of pulmonary hypertension ensues as
a result of earlier vascular changes and is followed within a few days by the
development of RVH (Chesney et al., 1974a; Lalich et al., 1977; Bruner et al., 1983;
. Hilliker et al., 1984; Reindel et al., 1990; Schultze et al., 1990).

Hilliker and Roth (1985b) were able to produce injury in isolated,
perfused lungs with MCTP administered in vitro. The injury was characterized by
increased vascular leak, increased perfusion pressure, and decreased removal of SHT
after injection of a moderate dose of MCTP into the pulmonary artery. Plestina and

Stoner (1972) showed that more than half of the MCTP administered to the isolated

32

lung in this way is retained within the lung in the initial 30 seconds of perfusion, most
likely on its first pass through the pulmonary vasculature. Ehrlich-positive material
(pyrroles) was distributed throughout the lung following perfusion, but was most
concentrated in the larger pulmonary vessels.

Minor differences in the response to MCTP vs. MCT exist. Platelet
numbers do not decrease after MCTP treatment (Bruner et al., 1983; White and
Roth, 1988) as they do after MCT (Hilliker et al., 1982). The transient, enhanced
response to vasoconstrictors that is evident soon after MCT treatment is more
persistent after MCTP treatment (Gillespie et al., 1986; Hilliker and Roth, 1985a).
Type I pneumocytes are affected with MCTP but not with MCT (Butler, 1970a).
These differences may be due to metabolites of MCT (other than MCTP) which are
produced by the liver, differences in the amount of active material which reaches the
lung, or modification of some of the pyrrolic effects of MCTP by the vehicle used for
its administration. However, the general character of the pneumotoxicity caused by
MCTP is similar to that caused by the parent alkaloid, and the advantages its use
provides are great. Therefore, for the purpose of inducing chronic pulmonary
vascular injury in the rat and carrying out investigations on the mechanisms of this
injury, MCTP serves as a useful alternative to MCT. Chemically synthesized MCTP
has an added benefit, as its use permits studies in vitra with tissues unable to

generate the necessary reactive metabolite.

33
F. Wanna:

One of the more interesting features of MCT (P) pneumotoxicity is the
delayed and progressive nature of the injury. Whether MCT is administered
chronically at low levels in the food or water, is given as a single, bolus injection, or
is administered as the pyrrolic metabolite, overt damage to the lung is not evident
immediately after treatment but requires several days to manifest itself (Valdivia et
al., 1967a; Valdivia et al., 1967b; Raczniak et al., 1979; Wilson and Segall, 1990;
Meyrick er al., 1980; Shubat et al., 1990; Bruner et al., 1983; Hilliker et al., 19833).

BC changes in small pulmonary vessels are among the first noted and
are evident as early as 1-3 days after treatment with MCT (P) (Valdivia et al., 1967a;
Valdivia et al., 1967b; Rosenberg and Rabinovitch, 1988; Reindel et al., 1990).
Evidence of vascular leak, consequent pulmonary edema, and inflammatory changes
also are reported relatively early in the course of intoxication (Valdivia et al., 1967a;
Valdivia et al., 1967b; Miller et al., 1978; Roth et al., 1981a; Sugita et al., 1983b;
Reindel et al., 1990), as are alveolar changes and alterations in lung respiratory
function (Valdivia et al., 1967a; Gillespie et al., 1985a; Lai et al., 1991). This may be
considered the "lung injury phase" of MCT toxicity, and it typically precedes vascular
remodeling and the "pulmonary hypertensive phase" (Roth et al., 1989).

Vascular remodeling involves thickening of precapillary resistance
vessels of the lung (Lalich et al., 1977; Merkow and Kleinerman, 1966b; Hislop and
Reid, 1974) and, in combination with increased fluid and fibrous tissue in the lung
parenchyma (Valdivia et al., 1967a), leads to increased PAP (Huxtable et al., 1977;

Kay et al., 1967a; Chesney et al., 1974a). RVH follows, in response to the increased

34

work of pumping against a more resistant pulmonary vascular bed, and death of the
animal usually results (Werchan et al., 1989; Meyrick et al., 1980).

This delayed and progressive development of injury suggests that lung
cells are sublethally but irreversibly damaged by even a single exposure to the
reactive metabolite(s) of MCT. The reactivity and lability of MCTP (Bruner et al.,
1986) suggest that the initial event in the toxicity occurs rapidly (ie. covalent binding
of MCTP to the cell); however, the lag period implies cellular alterations which are
slow to develop but which are progressive and trigger a cascade of molecular changes
and pathophysiological events that result ultimately in chronic pulmonary vascular
disease.

Hayashi et al. (1979) examined the effects of diet restriction on MCT
toxicity in rats. They found that rats receiving a single injection of MCT and fed ad
libitum (Group 1) gained weight initially, developed typical lung pathology, and died
between 22 and 40 days posttreatment. MCT-treated rats that were diet-restricted
(Group 2) did not gain weight, had minimal pulmonary changes, and survived the
entire experimental period. A third group of rats was treated with MCT and diet
restricted for 30 days, during which time they behaved like the animals in Group 2.
After 30 days post-MCT, they were fed an ad libitum diet, began to gain weight,
rapidly developed pulmonary lesions, and proceeded to die between 45 and 85 days
post-MCT. This interesting study supports the idea of early, irreversible injury which
is delayed in its expression; additionally, expression of the pulmonary injury appears
to be dependent on either adequate nutrition or actual growth of the animal. Similar

effects of diet restriction were later reported for MCTP by Gauey et al. (1985),

35

indicating that the effect is not a consequence of reduced food intake on MCT
bioactivation.

Once begun, the vascular and parenchymal changes are progressive,
and rarely is the injury caused by this compound reversible. Hislop and Reid (1974)
reported a regression of the arterial medial thickening and RVH when feeding of C.
spectabilis seeds was discontinued after 34 days, but they saw no recovery of lost
peripheral vessels. Adult rats were used in these studies and their age may account
for their recovery and survival, as young animals are typically more sensitive to PZAs
than are older animals (McLean, 1970; Schraufnagel, 1990). Recovery has not been
reported previously, and animals are typically moribund and / or dead by this point
in similar investigations (Kay and Heath, 1966; Kay et al., 1967a).

However, it does appear that there is a minimum cumulative dose of
MCT necessary for the development of pulmonary hypertension and RVH (Shubat
er al., 1989), and discontinuation of a chronic dosing regimen before this critical dose
is reached may result in regression of some lesions. Shale et al. (1986) gave MCT
in the drinking water at 20 mg/ liter for 3, 7 or 15 days, then examined a number of
parameters 21 days after initiation of treatment. Three or more days of MCT
resulted in an increased wet lung-to-body weight ratio, 7 or more days of MCT
resulted in increased lung protein content and dry weight, but RVH occurred only
after 15 or more days of MCT feeding, thus separating the development of
pulmonary hypertension from increases in lung mass. The investigators estimated
that rats in the 7 day group received 14.1 mg MCT/ kg, suggesting that the critical

dose of MCT is slightly higher (Shale et al., 1986; Shubat et al., 1989).

36

III. r101. hr ._!!0. .54.; .1 n-l . l . I , 1|‘l'!_r'!.-"l -
Little is known about the mechanisms underlying certain forms of chronic
pulmonary hypertension in people, and few good animal models exist for the study
of this condition. MCT can be characterized as an angiotoxin which causes lung
vascular injury that is delayed in onset and progressive in nature, and results in the
development of pulmonary hypertension. Chronic pulmonary vascular disease in
people and the pulmonary vascular toxicity produced in rats given low doses of MCT
have many similar features. Study of MCT pneumotoxicity in the rat may therefore
provide valuable insights with regard to pathogenesis of pulmonary vascular disease
in people. .
A W

Primary pulmonary hypertension (PPH) is a heterogeneous disease of
the pulmonary vasculature, and more than one causative factor is probably involved
in the pathogenesis. Few cases of PPH are diagnosed early in the course of the
disease. This makes it difficult to determine the etiology of the condition and
seriously limits the benefits of therapeutic intervention. One of the biggest stumbling
blocks investigators have faced is a lack of good animal models with which to study
this problem. Although not identical, many aspects of the pathology seen in the rat
treated with pneumotoxic doses of MCT are similar to those seen in patients with
PPH.

PPH is characterized by increased precapillary vascular resistance in
the lungs which ultimately results in RVH and failure. The clinical definition of PPH

requires an elevated PAP in the absence of a discernible cause. Also by definition,

37
the left atrial (or pulmonary wedge) pressure is normal (Dresdale et aL, 1951;

Hatano and Strasser, 1975). In its purest form, the histopathological pattern which
emerges is a plexogenic pulmonary arteriopathy, characterized by concentric intimal
fibrosis, necrotizing arteritis and dilatation lesions.

The most severe pathological changes occur in the muscular arteries
of 50-500 u. diameter. There is marked medial hypertrophy, with hyperplasia and
hypertrophy of smooth muscle, and duplication of the elastic laminae. Concentric
intimal proliferation is prominent in 100-200 M. diam. vessels. It has been suggested
that these vascular changes are a response of the pulmonary vasculature to
endothelial damage as a result of shear stress (Brenner, 1935 ; Clowes et al., 1983).

lesions in the pulmonary arterioles are less common but resemble
those of the muscular arteries. In capillaries, swelling of ECs and thickening of the
basement membrane have been noted; occasionally, capillaries may be obliterated
(Pietra and Schloo, 1986). Embolic or in situ thrombosis is seen with plexogenic
arteriopathy and is thought to occur in response to the disease process. Hyaliue
thrombi are occasionally found occluding small pulmonary arteries (Pietra et al.,
1987), accounting for further pruning of the vascular bed.

Plexiform lesions occur in 70% of PPH cases but are also seen in a
variety of other chronic pulmonary vascular diseases in humans (Pietra and Schloo,
1986). These lesions occur rarely in rats treated with MCT (Watauabe and Ogata,
1976). Plexiform lesions are thought to be a remodeling response of the vasculature

to very chronic elevations in vascular pressure, and they may occur less often in

38

laboratory animals due to their short lifespan and limited survival time after the
onset of pulmonary hypertension.

Necrotizing arteritis is seen in cases with very high PAP. This is
characterized by trausmural vasculitis with fibrin deposition and necrosis, and occurs
in response to rapid increases in pressure (Pietra et al., 1987).

The general pattern of hemodynamic changes noted with PPH includes
high PAP and pulmonary vascular resistance, normal pulmonary wedge pressure, and
low cardiac output (CO) and CI. Systemic pressure is typically normal or low
(Voelkel and Reeves, 1979). These findings are not unique to PPH, but are seen
with many forms of PH, including that caused by administration of MCT to animals.

There is limited respiratory impairment with PPH, but alterations in
pulmonary mechanics have been reported (Fernandez-Bonett, 1983). Biochemical
function of the lung is altered as well, and the inability to clear circulating biogenic
amines is suggestive of endothelial dysfunction (Heath er al., 1987; Peach et al.,
1987). A number of EC alterations may underlie the altered reactivity of the
pulmonary vasculature in PPH, including a decreased ability to generate endogenous
relaxants or metabolize vasoconstrictors, increased interaction with circulating blood
components, altered response to vasoactive substances, increased production of
mitogens and chemotaxins, or liberation of inflammatory mediators, to name a few
(Geggel et al., 1987; Hogg, 1988; Barst and Stalcup, 1985).

A common feature of many of the proposed etiologic factors for PPH
is their ability to cause pulmonary vasoconstriction (Hogg, 1988). Vasoconstriction

of the precapillary resistance vessels is thought by some to play a role early in the

39

disease process, and the occasional success of vasodilator therapy supports this
hypothesis (Hogg, 1988; Voelkel and Reeves, 1979). With the advance of disease,
there is focal obstruction of pulmonary arterioles as a result of severe intimal fibrosis.
These vascular segments are occluded and ultimately lost, leaving a small fraction of
functional pulmonary arterioles. The burden of vessel obliteration adds to the
vasoconstriction, and together these are thought to account for the increased vascular
resistance typically seen with PPH (Wagenvoort and Wagenvoort, 1970).

PPH is not a disease that affects millions of people each year; in fact,
not more than a thousand cases have been reported to date. Nevertheless, it
continues to hold the interest of physicians and biomedical scientists, because, despite
years of scrutiny, PPH remains a disease that is difficult to diagnose, untreatable,

progressive, and typically fatal.

B. W

Although rare cases of PPH occur, pulmonary hypertension is seen
more frequently as a sequel to a number of pathological or toxicological conditions.
At first glance, the primary conditions have little in common aside from the eventual
obstruction of pulmonary blood flow which results in increased PAP. However, some
common themes have been emerging with study of these conditions, including the
concept of vascular cell injury as an underlying factor in the development of
increased pulmonary vascular resistance. Again, the similarity in the development
and progression of pathological changes in the MCT model to those seen in cases of

secondary pulmonary hypertension is marked. Study of the rat treated with MCT (P)

40

may provide clues as to the common features underlying this diverse group of
conditions.
1. MW

Adult respiratory distress syndrome (ARDS) occurs as a
complication of a number of seemingly unrelated clinical conditions, including sepsis,
chemical injury, shock, thermal injury or trauma (Hyers and Fowler, 1986). ARDS
is a progressive condition, which is characterized initially by disruption of the
pulmonary rnicrovasculature and increased capillary permeability. Endothelial injury
is evident on histopathology, and EC function is impaired, as evidenced by decreased
removal of SHT (Morel et al., 1985 ). Thromboembolism and inflammation are
evident, and PAP is increased. The vascular alterations are accompanied by
pulmonary parenchymal cell injury and altered pulmonary mechanics (Snow et al.,
1982; Tomashefski et al., 1983; Cochrane et al., 1983). These acute pulmonary
changes resemble the alterations seen in rats treated with high doses of MCTP
(Plestina and Stoner, 1972; Hurley and J ago, 1975).

The earliest manifestations of ARDS may resolve, but in some
cases the condition progresses to a more chronic form, characterized by pulmonary
interstitial fibrosis, vascular remodeling and pulmonary hypertension (Snow et al.,
1982). The vascular lesions seen in this chronic stage are similar to those seen in
rats treated with low doses of MCT (P). There may be a common denominator
involved in the initiation of vascular lesions in ARDS, such as injury to cells of the

microvasculature by mediators released during sepsis or traumatic injury. Study of

41

the MCT model may provide insights as to how these vascular cell alterations lead
to the development of pulmonary hypertension.
2. WWW

Pulmonary hypertension also occurs as a result of exposure to
chronic hypoxia (Abraham et al., 1971; Meyrick and Reid, 197 8; Rabinovitch et al.,
1979; Barer et al., 1982), radiation (Phillips, 1966; Gross, 1981; Gross, 1977; Perkett
et al., 1986) or chemotherapeutic agents (see Chapter VI). The mechanisms involved
in the development of chronic vascular alterations under these conditions are still
undefined, but a direct, injurious effect(s) on vascular cells has been proposed as the
initiating factor in each case.

Brief exposures to hypoxia result in immediate and reversible
increases in PAP and PVR (Fishman, 1980). Sustained hypoxia prolongs these
changes and leads to structural remodeling of the pulmonary vasculature and
eventual RVH (Abraham et al., 1971; Meyrick and Reid, 1978). EC alterations are
evident within hours of exposure to hypoxia, and platelet adherence is enhanced
(Meyrick and Reid, 1978).

Enhanced vasoconstriction is thought to be responsible for both
the increased PAP and vascular remodeling in hypoxic pulmonary hypertension
(Stanbrook et al., 1984); however, it may be that early changes in vascular cells are
responsible for both vasoconstriction and the initiation of remodeling. Hypoxia has
numerous effects on the endothelium, including perturbation of its barrier and

anticoagulant functions (Ogawa et al., 1990b; Ogawa et al., 1990a), enhanced release

42
of mitogens (Vender et al., 1987) and decreased ACE activity (Schweigerer et al.,

1987; Shanahan and Korn, 1984).

Exposure to radiation rapidly results in increased capillary
permeability, pulmonary inflammation and altered pulmonary blood flow. These
changes are persistent, and pulmonary lesions progress to a chronic, fibrotic state
with pronounced vascular remodeling and increased PAP (Perkett et al., 1986).
Changes in type II pneumocytes and capillary endothelium are evident early in the
course of injury, and ECs irradiated in vitra respond with an inhibition of
proliferation, hypertrophy, enhanced adherence for PMNs and platelets, increased
release of PGIZ, increased permeability, and altered ACE activity (Rubin et al., 1984;
Dunn et al., 1986; Rosen et al., 1989).

Repeated exposure to hyperoxia, (Jones et al., 1983; Crapo et
al., 1980), endotoxiu (Meyrick and Brigham, 1983; Meyrick and Brigham, 1986), and
chemicals such as a-naphthylthiourea (ANTU) (Hill et al., 1984) can also result in
pulmonary vascular remodeling and hypertension. The endothelium has been
implicated as a primary site of lung injury by these agents as well (Crapo et al., 1980;
Meyrick et al., 1989; Rutili et al., 1982). Although the early changes which follow
acute endothelial cell injury by these agents differ somewhat from those seen in
hypoxic or MCT-induced lung injury, the chronic alterations they produce after

prolonged or intermittent exposure are very similar.

Initial EC damage may be the shared event linking a diverse group of

etiologic factors to a common pathophysiological endpoint. Chronic exposure to

43

some agents (eg. hypoxia, ANTU) is necessary to induce persistent vascular
alterations, whereas, a single administration of MCT, or a single dose of radiation,
is sufficient to produce similar injury. If endothelial injury is central to the
development of subsequent alterations, then MCT(P) treatment might be expected
to produce more persistent EC damage than does hypoxia. It has been proposed that
persistent endothelial injury is the crucial event in the development of chronic
pulmonary hypertension (Barst and Stalcup, 1985 ; Meyrick et al., 1987).

These models of pulmonary hypertension share other common features
which have been proposed to play a role in the development of chronic pulmonary
vascular disease, including persistent pulmonary vascular leak (Sugita et al., 1983b),
chronic inflammation (Meyrick et aL, 1987) and altered pulmonary mechanics (Lai
et al., 1991). Investigation into the potential mechanisms which underlie MCT(P)-
induced lung injury could increase our understanding of how vascular injury can lead

to chronic pulmonary alterations.

IV. WWW

Despite intensive study, the mechanisms by which MCT (P) produces delayed
and progressive cardiopulmonary toxicity are not yet understood. On the basis of the
progression of lesions, the sequence in which pathophysiological alterations occur,
and the response to a number of pharmacological interventions, several potential
mechanisms of action have been proposed for MCT (P). It is possible to group these
into six broad categories: 1.) Vasoconstriction, 2.) Thrombosis, 3.) Platelets, 4.)

Inflammation, 5.) Cell Growth and Proliferation, and 6.) Alkylation. Although these

44

categories overlap to a certain extent, they represent recurrent themes in the
MCT (P) literature and provide a structure to facilitate discussion of this complex
subject.
A. Wen

Pulmonary vasoconstriction, with a resultant increase in pulmonary
arterial pressure, is a primary defect in a number of chronic lung diseases, including
hypoxic pulmonary hypertension (Abraham et al., 1971; Staubrook et al., 1984) and
some cases of primary pulmonary hypertension (PPH) (Wagenvoort and Wagenvoort,
1977; Hogg, 1988; Voelkel and Reeves, 1979). In these conditions, vascular
remodeling occurs secondarily in response to increased pulmonary vascular pressure.
Direct vasoconstriction of pulmonary vessels by MCT (P) was proposed as a
mechanism for increased PAP by Smith and Heath (1978) based on morphologic
changes they saw in pulmonary veins. The reactive nature of the pyrrolic metabolites
of MCT and the delayed development of pulmonary hypertension in the rat argues
against this theory, although intravenous administration of MCT does result in rapid
increases in PVR (Miller et al., 1978; Czer et al., 1986) and PAP (Czer et al., 1986)
in the dog. This response may be due to a direct vasoconstrictive effect of MCT
metabolites in this species, but further study will be required to confirm this theory.

Secondary vasoconstriction in response to hypoxemia has been
suggested in the development of MCT-induced pulmonary hypertension (Turner and
Lalich, 1965; Chesney er al., 1974a). Similarities in the end-stage pathology of MCT-
and hypoxia-induced hypertension led to this theory before the delayed and

progressive nature of MCT pneumotoxicity was well understood. Subsequent

45

investigations have clearly established that vascular remodeling is evident prior to the
development of pulmonary hypertension in this model (Meyrick et al., 1980; Ghodsi
and Will, 1981), and marked hypoxemia does not occur until pulmonary hypertension
and RVH are well established after ingestion of C. spectabilis seeds (Meyrick et al.,
1980). However, hypoxic vasoconstriction continues to receive interest in the MCT
model.

Gillespie and coworkers have reported that ventilatory dysfunction
precedes the development of vascular changes in MCT-injected rats, suggesting that
hypoventilation, impaired gas exchange, and consequent hypoxemia may be important
in the evolution of pulmonary hypertension (Gillespie et al., 1985a; Lai et al., 1991).
Hill et al. (1989) detected a mild hypoxemia 21 days (but not 10 days) after injection
of MCT. In addition, they found that exposure to 35% oxygen on days 10 - 21 after
MCT administration reversed the hypoxemia and significantly reduced vascular
remodeling, right ventricular systolic pressure and right ventricular hypertrophy, but
supplemental oxygen during the first 10 days following MCT was not protective.
Thus, it appears that hypoxemia may be involved in the development or maintenance
of MCT-induced pulmonary hypertension, perhaps by potentiating smooth muscle cell
proliferation (Vender er al., 1987) or vasoconstriction (Meyrick and Reid, 1978);
however, it is not a factor in the early injury caused by MCT.

After MCT treatment, changes in pulmonary vascular cells which result
in altered vascular responsiveness may also contribute to increased vasoconstriction.
[Please refer to sections D.3.b. and D.4.] Treatment with MCT results in some of

the same alterations in vascular reactivity as does endothelial denudation, including

46

decreased response to the vasodilator ACH and increased response to PGFZ, (Ito
et al., 1988a). Decreased clearance of biogenic amines (Gillis er al., 1978; Hilliker
et al., 1983c; Hilliker et al., 1984) impaired degradation of prostaglandins (Ito et al.,
1988b; Ito et al., 1988a), decreased generation of EDRF (Ito et al., 1988a) and other
dysfunctions of pulmonary endothelium which occur following MCT treatment may
contribute to abnormal vascular tone and responsiveness.

Vascular smooth muscle changes in response to MCT treatment which
could support the vasoconstriction hypothesis are less consistent. Vessels
demonstrate enhanced responsiveness to vasoconstrictors relatively early in the
course of injury, but it is uncertain whether this effect is sustained (Gillespie et al.,
1986; Altiere et al., 1986; Hilliker and Roth, 1985a). The pulmonary vasculature
exhibits a heightened response to hypoxia, but this is seen either very early (Gillespie
et al., 1986) or quite late, after vascular remodeling is evident (Rosenberg and
Rabinovitch, 1988). Most investigators believe that these changes are incompatible
with a major role for vasoconstriction in the development or maintenance of
pulmonary hypertension (Gillespie er al., 1986; Petry et al., 1986) , Meyrick et al.,
1980).

The response to several pharmacological interventions in MCT toxicity
suggests that vasoconstriction may play a modulatory role, at least in the later
cardiopulmonary alterations, and perhaps in some of the earlier changes as well. The
vasodilator, hydralazine, reduces the degree of RVH and the increased BALF protein

concentration in MCTP-treated rats (Hilliker and Roth, 1984b). This drug has other

47
effects, including inhibition of platelet thromboxane synthesis, but part of its

protective effect may be in reducing pulmonary vascular resistance.

The calcium channel blocking agent, verapamil, attenuates MCT-
iuduced pulmonary hypertension by limiting the extent of vascular remodeling of the
small and large pulmonary vessels, but has no effect on the earlier lung injury phase
(Mathew et al., 1990). Interference with a number of calcium-dependent processes
in endothelium and smooth muscle may account for these findings. ,B-blockade with
propranalol reduces PAP and RVH (Huxtable er al., 1977), and chemical
sympathectomy with 6-hydroxyd0pamine also reduces the degree of RVH after MCT
administration. Sympathetic input to the heart and vasculature may be enhanced in
MCT-treated animals (Tucker et al., 1983) and may thus contribute to both the
increased vascular resistance and the cardiac response to increased pressure. It may
be that generally reducing vascular tone early in the course of injury can modulate
the extent of vascular remodeling and limit the development of cor pulmonale.

Concurrent administration of magnesium aspartate (Mg-Asp) also
reduces arterial pathology, attenuates the increases in PAP and reduces RVH in
MCT-treated rats (Mathew et al., 1988). Mg+ + has a number of functions in
vascular cells, acting in part by regulating calcium movement; it also serves as a
cofactor in many enzyme systems. Mg+ + blocks vasoconstriction and enhances
vasodilation in response to many agonists, modulates prostaglandin production, and
inhibits coagulation and platelet adherence. Although it is not clear how Mg-Asp
limits the toxic effects of MCT, many of its actions would be consistent with a

relaxant effect on constricted vessels.

48
An important function of the puhnonary vascular endothelium is

regulation of a number of circulating vasoactive mediators. ACE, or kininase II,
found on the surface of ECs, converts Ang I to the potent vasoconstrictor, Aug II,
and cleaves bradykinin to an inactive form (Yang et al., 1970; Erdos, 1975). It has
been proposed that agents which result in pulmonary endothelial injury (e.g. MCT)
may alter the distribution or activity of ACE (Dobuler et al., 1982; Catravas et al.,
1988b), thus influencing local and systemic levels of the vasoactive substances it
regulates. Response to Ang II is variable in vessels from MCT-treated animals
(Gillespie et al., 1986; Ito et al., 1988a); nevertheless, altered amounts of this potent
pressor agent in the pulmonary vasculature may influence the degree of
vasoconstriction. In addition to its vasoconstrictive effects, Ang II has a number of
other effects on vascular cells that could affect the development of MCT-induced
injury (Bell and Madri, 1990) (see Chapter II).

A number of investigators have evaluated ACE activity in lungs of rats
treated with MCT in an effort to correlate changes in enzyme activity to some aspect
of the development or progression of MCT-induced cardiopulmonary changes. Some
investigators report a decrease in lung ACE activity which corresponds to an increase
in pulmonary pressure (Kay et al., 1982b; Keane et al., 1982), whereas others contend
that apparent changes in lung ACE activity are due to a dilution of its activity by the
increased lung mass seen in this model (Huxtable et al., 1978; LaFranconi and
Huxtable, 1983). Yet another group reports an increase in lung ACE activity shortly
after MCT treatment, with a subsequent, sustained decline in activity later on

(Molteni er al., 1984). A fourth group reports a sustained decline in ACE activity

49

which precedes the onset of pulmonary hypertension and which occurs at doses that
do not produce RVH (Shale et al., 1986). From these reports, it is difficult to draw
definitive conclusions about ACE activity in lung vasculature injured by MCT.

Captopril and a number of other thiol and non-thiol ACE inhibitors
reduce MCT-induced pulmonary vascular remodeling, interstitial fibrosis and RVH
(Molteni et al., 1985; Molteni et al., 1986; Molteni et aL, 1987; Molteni et al., 1988a).
Interestingly, their protective effects are apparently not attributable to ACE
inhibition, thiol effects or sparing of the endothelium in the early lung injury phase
(Molteni et al., 1987). It is possible that the anti-inflammatory or systemic
hypotensive actions of these compounds are responsible for their salutary effects, or
they may alter the pharmacokinetics of MCT itself. Captopril and certain other ACE
inhibitors also inhibit weight gain (Molteni et al., 1986), which has been shown to be
similarly protective (Hayashi et al., 1979; Ganey er al., 1985). ACE inhibition in
other forms of vascular injury and hypertension is currently receiving a lot of
attention in the literature. ACE inhibitors strongly suppress neointimal proliferation
following balloon catheter injury (Powell et al., 1990). They are autifibrotic (Ward
et al., 1989), and enhance endothelium-dependent relaxation by increasing release of
EDRF (Goldschmidt and Tallarida, 1991; Clozel, 1991). This is an area which merits
further study in the MCT (P) model. [See Chapter II for a more complete discussion
of this subject]

In summary, it appears unlikely that vasoconstriction plays a major
role in the early vascular alterations caused by MCT (P). There is evidence for

increased tone in the vasculature, and relaxation of vascular smooth muscle and relief

50

of sympathetic input seems to attenuate the cardiopulmonary response to the early
injury. However, extensive vascular remodeling, which is seen in this and other
models of chronic pulmonary hypertension, results in serious reductions in arterial
diameter and elasticity that must be severely limiting to vascular responsiveness. It
seems likely that these fixed changes ultimately eclipse those due to altered vascular

reactivity in this model of lung injury.

B. le f r m o i

Although there is no question that MCT-induced pulmonary vascular
injury is associated with platelet and fibrin thrombi [see section D.1.b.], it remains
controversial whether thrombosis plays an active role in the development of lung
injury or merely occurs in response to the injury. Vascular thrombosis has been
reported early in the course of MCT pneumotoxicity (Valdivia et al., 1967a; Lalich
et al., 1977), but this is not a consistent finding. Pruning of the peripheral
vasculature has been attributed, in part, to occlusion of small vessels and capillaries
with thrombi (Schraufnagel and Schmid, 1989; Hislop and Reid, 1974), and this
contributes to the decrease in cross-sectional area available for blood flow in the
lungs of these animals.

The presence of vascular thrombi is suggestive of either excessive
procoagulant activity or defective fibrinolysis (Bauer and Rosenberg, 1987). The
endothelium is an important component of the hemostatic system, generating both
anti- and procoagulent factors. [Please refer to Section V.] Injury to the

endothelium as a result of MCT (P) treatment ,appears to result in alterations

5.1

consistent with a hypercoagulable state, including increased platelet adherence in the
pulmonary vasculature (White and Roth, 1988) and decreased fibrinolytic activity
(Molteni et al., 1988a; Molteni et al., 1986).

Unfortunately, few studies have been directed at this aspect of MCT
pneumotoxicity. Fasules et al. (1987) found that anticoagulant heparin inhibited
clotting in rats treated with MCT but did not protect against early lung injury,
vascular remodeling, increased PAP or the development of RVH. Heparin
administered at similar doses did attenuate hypoxia-induced pulmonary hypertension
in rats (Geggel et al., 1986). It is possible that localized thrombus formation in the
lung could still occur in heparin and MCT-treated animals, as the lungs were not
scrutinized closely for the presence of thrombi. However, it is difficult to reconcile
these results with a major role for vascular thrombosis in MCT lung injury.

Direct examination of the coagulation system in MCT-treated rats is
currently under investigation. It appears that vascular thrombosis is not mediated by
an excess of procoagulent activity as evaluated in the peripheral blood (Schultze et
al., 1991); however, preliminary results suggest that there is decreased fibrinolytic

activity in the lungs of rats treated with MCTP (Schultze and Roth, 1992). ,

C. Weight
Platelets were originally noted as components of the thrombi seen in
the pulmonary rnicrovasculature after administration of MCT (Turner and Lalich,
1965; Valdivia et al., 1967a; Chesney et al., 1974a). Subsequent to these observations,

it was noted that numbers of circulating platelets are moderately decreased in MCT-

52

treated rats at 2, 5, and 10 days posttreatment (Hilliker et al., 1982); this is a
relatively early change following MCT administration. Thrombocytopenia is not
evident in rats treated with MCTP (Bruner et al., 1983), however platelet
sequestration in the lungs is evident by 8 days posttreatment (White and Roth, 1988).
These findings sparked interest in the platelet as a potential effector in this model,
and its role in mediating MCT-induced cardiopulmonary injury has been investigated
in some detail.

Hilliker er al. (1984) used antiserum directed against rat platelets (PAS)
to create 48-hour windows of moderate thrombocytopenia during the course of
development of MCT-induced lung injury. PAS administered 12 hours prior to
MCTP was not protective; however, thrombocytopenia during days 3-5 or days 68
posttreatment reduced the development of RVH by 19% or 41%, respectively, but
did not alter any of the markers of lung injury examined. The three intervals of
thrombocytopenia were designed to bracket the initial lung injury (02 d), the
development of major lung injury (3-5 d), and the increase in PAP (6.8 d) seen after
MCTP administration. Thus, it appears that platelets do not mediate the lung injury
but do contribute to the development of pulmonary hypertension (Hilliker et al.,
1984a; Ganey et al., 1988).

Injury to vascular endothelium can result in enhanced platelet-
endothelial cell interactions (Ratliff et al., 1979) and may account for platelet
sequestration and activation. Activated platelets can, in turn, release a number of
potent vasoactive substances which result in further platelet aggregation and

stimulation (adenosine diphosphate [ADP], TXAz). altered vascular permeability

53
(SHT), vasoconstriction (SHT and TXAz) and proliferation of mesenchymal cells

(platelet-derived growth factor; PDGF) (Holmsen et al., 1969; Longenecker, 1985;
Niewiarowski and Holt, 1985).

Hilliker et al. (1983b) investigated the effects of MCT on the response
of platelets to aggregating agents and found that platelets isolated from MCTP-
treated rats 7 or 14 days posttreatment were less sensitive to stimulation than
platelets from control animals. The authors proposed that this effect might be due
to endothelial alterations which could influence the platelet milieu, such as increased
production of PGIZ. Nevertheless, direct hyperresponsiveness of the platelet to
aggregatory stimuli was not observed. In addition, Ganey and Roth (1987a) found
that MCTP treatment of rats did not affect platelet SHT content, nor did it alter
basal or stimulated release of TXA2 in vitra. Hence, platelet function and content
appear to be essentially unaltered by MCTP treatment, with the exception of
depressed aggregation quite late in the course of injury, suggesting that external
stimuli present in MCT(P)-treated animals may trigger platelet responses in this
model.

Which of the platelet’s many activities are important in the
development of MCTP-induced pulmonary hypertension? In an efiort to begin to
address this question, Hilliker et al. (1983) compared the responses of Sprague-
Dawley and F awn-Hooded (FH) rats to MCTP. FH rats have a congenital platelet
defect which consists of a reduced ability to take up SHT and to store SHT and
ADP, decreased release of these substances when stimulated, and reduced

aggregatory responses to collagen and arachidonate (Raymond and Dodds, 1975).

54

MCTP caused similar pneumotoxic injury in both strains of rats, suggesting that SHT
and ADP may not be essential to the pathogenesis. However, FH rat platelets do
generate T'XA2 and PDGF normally, and may have a small amount of residual dense
granule function; perhaps these may be adequate for them to fulfill their role in
MCTP pneumotoxicity (Hilliker er al., 1983a).

SHT has received quite a lot of attention as a potential mediator of the
pneumotoxic effects of MCT, and not only as a component of platelets; mast cells,
which also release SHT, have been identified in the lungs of rats fed C. spectabilis
seeds. SHT causes increased vascular permeability in vessels and is a vasoconstrictor
as well; both of these effects could contribute to MCT pneumotoxicity (Sugita et al.,
1983b; Hilliker and Roth, 1985a). Damage to the pulmonary endothelium by
MCT (P) results in a decreased ability to take up and metabolize this vasoactive
mediator (Gillis et al., 1978;(Hilliker et al., 1982; Hilliker er al., 1983c), which might
lead to increased levels of SHT in the pulmonary vasculature; the pressor response
to SHT is also increased in the lungs of MCTP-treated rats (Hilliker and Roth,
1985a). Finally, p—chlorophenylalanine (PCPA), a SHT synthesis inhibitor, attenuated
the increased PAP and RVH seen following MCT treatment (Carillo and Aviado,
1969; Tucker et al., 1983; Kay et al., 1985); however, this may be due to effects on
metabolism or body weight gain rather than a result of decreased SHT synthesis.
Ganey et al. (1986) gave the SHT receptor antagonists, metergoline and ketanserin,
to MCTP-treated rats. These agents had no impact on either the lung injury or the

pulmonary hypertensive effects of MCTP (Ganey et al., 1986), suggesting that SHT

55

does not play an important role in development of the cardiovascular changes in this
model.

The platelet-derived, vasoactive mediator, TXAZ, has also received its
share of attention in the MCT literature. Initial studies utilized a group of agents
which shared the ability to inhibit platelet prostaglandin synthesis. Hydralazine,
dexamethasone and sulphiupyrazone all reduced the degree of RVH which resulted
from MCTP treatment, and hydralazine and dexamethasone also reduced BALF
protein content, suggesting that platelet-derived vasoconstrictors could play a role in
the development of MCTP-induced pulmonary hypertension (Hilliker and Roth,
1984b). However, these compounds could have a number of other potential effects
on the lungs as well, including vasodilation and generalized anti-inflammatory effects,
and the results may have been explained in a number of ways.

Arachidonic acid metabolites, particularly TXAZ, are thought to be
inmortant in endotoxiu-induced pulmonary hypertension (Hales et al., 1981). Interest
focused on this mediator in the MCT model when it was discovered that lungs from
rats treated with MCT (Stenmark et al., 1985) or MCTP (Ganey and Roth, 1987b)
release increased amounts of TXA2 late in the course of injury. The source of the
TXA2 has not been identified and could be either platelets or lung tissue. However,
subsequent studies with the thromboxane synthesis inhibitor, dazmegrel (Langleben
et al., 1986; Ganey and Roth, 1986), the cyclooxygenase inhibitor, ibuprofen, and the
T'XA2 receptor antagonist, L-640,035 (Ganey and Roth, 1986) have shown that
interference with TXA2 synthesis or action does not attenuate the cardiopulmonary

effects of MCT (P). Thus, another platelet mediator was eliminated from the running.

56
The role of the mesenchymal cell mitogen, PDGF, has also been

investigated. This growth factor, released from platelets or ECs (Ross et al., 1986),
could play a role in the vascular remodeling phase of MCT (P) pneumotoxicity.
However, PDGF neutralizing antibody was not protective when administered daily
for 14 days following MCTP treatment (Ganey et al., 1988). It is possible that
systemically-administered anti-PDGF would not block the paracrine efiects of PDGF
in the microenvironment of the vascular wall, therefore it is difficult from this result
to eliminate PDGF (or other growth factors) as important players in the vascular
remodeling seen with MCT (P) lung injury.

In summary, platelets appear to be - important after MCT(P)-induced
lung injury is established but prior to the development of pulmonary hypertension
and RVH. Although a few of the potential mechanisms for platelet involvement
have been eliminated, there are still several to explore, including a number of other
platelet-associated growth factors such as transforming growth factors a and B and

epidermal growth factor (Assoian and Sporn, 1986; Oka and Orth, 1983).

D. When
A majority of investigators report a distinct inflammatory component
to MCT (P) pneumotoxicity which is progressive throughout the course of the lung
injury. Inflammatory infiltrates are most evident in the alveolar walls and
perivascular interstitium, and consist primarily of mononuclear cells, including mast
cells, macrophages, lymphocytes and plasma cells (Takeoka et al., 1962; Kay et al.,

1967b; Valdivia et al., 1967a; Sugita et al., 1983b; Kay and Heath, 1966; Sugita et al.,

57
1983a; Stenmark et al., 1985; Dahm er al., 1986). There is also a slight, transient

increase in polymorphonuclear cells (PMNs) in the BALF at 24 hr, and progressively
increasing numbers are recoverable from the lungs after 5 days posttreatment with
MCTP or 21 days after MCT treatment (Stenmark et al., 1985; Ilkiw et al., 1989; Czer
er al., 1986; Schultze et al., 1990).

Although these inflammatory cells are generally present, their role in
the deve10pment of pulmonary lesions in this model have not been established.
These cells respond to and release a number of mediators which could be important
in the vasoactive and structural changes which develop in the lungs of MCT(P)-
treated animals, and anti-inflammatory drugs such as dexamethasone and
methylprednisolone protect against the development of RVH (Hilliker and Roth,
1984b); however, these drugs also decreased weight gain and the results could be
explained on this basis. Exudation of fluid and protein from the pulmonary
vasculature occurs relatively early in the course of MCT(P)-induced lung injury
(Wilson and Segall, 1990; Heath, 1969; Valdivia et al., 1967b; Molteni er al., 1984;
Sugita et al., 1983b; Schultze et al., 1990), and may contribute to or arise as a
consequence of pulmonary inflammation. Increased collagen in the adventitia of
pulmonary arteries (Merkow and Kleinerman, 1966a; Guzowski and Salgado, 1987;
Meyrick and Reid, 1982) and pulmonary interstitial fibrosis (Kay et al., 1967b; Kay
et al., 1969; Butler, 1970a; Meyrick and Reid, 1979; Meyrick et al., 1980) have been
reported as later developments, and are likely sequelae to chronic inflammation.

Mast cells accumulate in the lungs of C. spectabilis-treated rats

(Takeoka et al., 1962; Kay et al., 1967b; Valdivia et al., 1967a). These cells release

58

a number of vasoactive mediators, including SHT (discussed above with platelets),
heparin, histamine and leukotrienes D4 and E4. It appears that the accumulation of
mast cells correlates with the severity of pulmonary exudative changes rather than
with the extent of pulmonary hypertension (Kay et al., 1967b), suggesting that they
do not play a critical role in the development of the hypertensive lesions. However,
when mast cells are present, the mediators they release could contribute to the
disease process. In the discussion of the platelet above, SHT was eliminated as a
major player in MCT (P) pneumotoxicity. However, mast cells and other
inflammatory cells also generate leukotrienes (LTs), and a case has been made for
their involvement in MCT (P) pneumotoxic injury. These arachidonic acid
metabolites are potent vasoconstrictors and cause increased vascular permeability as
well (Samuelsson, 1983). They also interact with the cyclooxygenase pathway,
inducing the release of TXA2 (Omini et al., 1982). LT synthesis inhibitors block
hypoxic pulmonary vasoconstriction, and are proposed to play an important role in
hypoxia-induced pulmonary hypertension (Morganroth et al., 1984).

Stenmark et al. (1985) found increased levels of LTs B4, D4 and E4 in
the BALF of MCT-treated rats, and treatment with the 5-lipoxygenase inhibitor,
diethylcarbamazine (DEC), was protective against both the early lung injury and the
later development of pulmonary hypertension in MCT-treated animals. However,
Bruner and Roth (1984) found that whereas DEC protected against early lung injury
in MCTP-treated animals, it did not influence the development of later lung injury
and pulmonary hypertension. DEC also retarded weight in these studies, and this

effect may have been responsible for any protective effects. Nevertheless, this area

59

merits further investigation. More specific LT synthesis inhibitors are currently
available, which should make it possible to define better the role of LTs in MCT(P)-
induccd cardiopulmonary toxicity.

The lipid autacoid, platelet-activating factor (PAF), can stimulate LT
synthesis and can induce pulmonary hypertension when infused chronically into
rabbits (Ohar er al., 1991). This inflammatory mediator was evaluated in MCT-
treated rats by Ono and Voelkel (1991). PAF levels were increased in the lungs of
treated animals at 1-3 weeks posttreatment, and administration of specific PAF
antagonists reduced pulmonary vascular leak at 1 week posttreatment and inhibited
the development of pulmonary hypertension and RVH. The authors propose that
injury to endothelial cells or activation of macrophages may result in release of this
mediator during the early phases of MCT-induced lung injury, which could in turn
stimulate LT synthesis and activate platelets. This may be an important area to
pursue.

The cytokines have received little attention as possible mediators in this
model. Interleukin 1 (IL-1) is transiently increased in the BALF of MCT-treated rats
at 4 days posttreatment, and is increased again on days 14 and 21 (Gillespie et al.,
1988). It is not certain what role, if any, these mediators play in MCT-induced lung
injury, but macrophage and PMN products might be involved in the development of
the toxicity.

Myeloperoxidase activity is increased in BALF at 21 days posttreatment
(Gillespie et al., 1988), suggesting that PMNs are present in the lungs after MCT(P)

treatment. Granulocytes and phagocytic cells are capable of generating reactive

60
oxygen species (Fantone and Ward, 1982; Repine et al., 1979), and it has been

postulated that these could play a role in MCT pneumotoxicity. Dahm et al. (1986)
examined the ability of cells recovered from the BALF of MCTP-treated rats to
generate superoxide anion. Cells collected from MCTP-treated rats on days 7, 10
and 14 posttreatment generated less superoxide than cells from control rats when
stimulated in vitra. These cells may have already been stimulated to generate
superoxide in viva so that their ability to generate more in vitra was exhausted, or
they may have been down-regulated in this capacity. Bruner et al. (1987b) pursued
the role of reactive oxygen species in this model by attempting to intervene with
drugs which prevent formation of toxic species or degrade them. Cotreatment with
deferoxamine, dimethyl sulfoxide or catalase was not protective in MCT-induced lung
injury, suggesting that toxic oxygen metabolites do not play an important role in this
model. It could be argued that these drugs were unable to block the effects of
oxygen metabolites generated at the inflammatory cell / lung tissue interface; however,
this regimen of treatment is effective in other models of neutrophil-mediated injury
(Ward et al., 1983; Till et al., 1985).

Another agent released by activated PMNs is elastase. Neutrophil
elastase degrades elastin and types 111 and IV collagen and causes damage to ECs
in certain models of vascular injury (Mainardi et al., 1980; Smedley et al., 1986).
Increased fragmentation of the internal elastic lamina of larger muscular arteries has
been observed 4 days after MCT administration, and elastases have been proposed
to play a role in MCT pneumotoxicity (Todorovich-Hunter et al., 1988). C0-

administration of the neutrophil serine elastase inhibitor, SC39026, attenuated the

61
vascular remodeling and reduced the PAP and RVH in MCT-treated animals;

however, it did not protect against early lung injury, including EC changes, or reduce
the inflammation (Ilkiw et al., 1989; Molteni et al., 1989b). When administration of
the inhibitor is discontinued at 8 days posttreatment, pulmonary changes are not
present at 2 weeks but are evident at 3 weeks posttreatrneut (Ilkiw et al., 1989).
Thus, neutrophil elastase appears to be important in the initial remodeling of
pulmonary vessels and in the maintenance of vascular changes that result in
pulmonary hypertension. Platelets (James et al., 1986) and macrophages (Banda and
Werb, 1981) also produce elastases, and these may be important in MCT-induced
vascular remodeling as well.

A number of agents prevent pulmonary fibrosis in MCT-treated
animals. Penicillamine, which inhibits collagen crosslinking, and the ACE inhibitors,
captoprfl and C1242817, all decrease lung hydroxyproline content and attenuate the
development of pulmonary hypertension and RVH in this model (Molteni et aL,
1985; Molteni et al., 1986). None of these compounds affords protection against the
early lung injury caused by MCT, however penicillamine and captopril do ameliorate
the decrease in fibrinolytic activity. It has been proposed that sparing of this EC
function may account for the antifibrotic effect of these compounds (Molteni er al.,
1985; Law, 1981).

CL242817 prevents weight gain, and could influence MCT toxicity via
this mechanism (Hayashi er al., 1979; Ganey er al., 1985); however, the other 2 agents
do not affect body weight. Penicillamine does not inhibit ACE activity but does

chelate capper, which is elevated in serum in MCT(P)-treated rats (Molteni et al.,

62
1985; Ganey and Roth, 1987; Molteni et al., 1988b). Although it is possible that this

ability to decrease serum copper concentration is a mechanism for the protective
effects of penicillamine, it seems unlikely. In general, ACE inhibitors do not chelate
copper, but serum copper is also lower when Cl242817 is co-administered with MCT.
Serum copper concentration appears to vary with the severity of the cardiopulmonary
damage induced by MCT, and agents which reduce the injury may indirectly reduce
the hypercupremia (Molteni et al., 1988b). Pulmonary fibrosis is a common sequel
to many types of lung injury and frequently occurs as a result of chronic
inflammation. By limiting the development of fibrosis, many of the later events in
MCT toxicity appear to be prevented; the mechanisms of this protection remain
uncertain.

The possibility that an immune-mediated response may account for
some of the effects of MCTS(P) has also been explored (Bruner, 1986; Bruner et al.,
1987a). Immunosuppression has been achieved by either administration of an
antiserum directed against rat lymphocytes (LAS) or the administration of
cyclosporin A. Administration of LAS has no effect on MCT pneumotoxicity.
Cyclosporin A is protective against both the early lung injury and the development
of RVH, however, this compound also restricts weight gain and may protect in this
way (Hayashi et al., 1979; Ganey et al., 1985). MCTP does not activate complement
in viva or in vitra, and complement depletion is not protective (Bruner et al., 1988).
From these results, the participation of either cell-mediated or humoral immunity in

MCT pneumotoxicity seems unlikely.

63

Shubat et al. (1987) reported the development of vascular alterations
and RVH after treatment with MCT in the absence of inflammation, suggesting this
feature of MCT toxicity is not essential to the development of vascular injury.
However, MCT(P)-induced inflammation, when present, may influence the
progression of the injury. It is still unclear which of the specific cells and mediators
are essential to the progression of lung injury and which are secondary contributors.
Leukotrienes and PAP are still viable candidates, and elastases seem to be quite
important as well; these will presumably receive further attention as more selective
inhibitors are identified. The antifibrotic agents are also quite interesting, as they
may represent a relatively late intervention which affords some protection against

chronic pulmonary vascular injury.

E. W

Several lines of evidence suggest that alterations in cell growth and
proliferation are fundamental to the expression of MCT (P) cardiopulmonary toxicity:
young, growing animals are more susceptible to the effects of MCT than older
animals (Allen and Chesney, 1972b; Todd et al., 1985), proliferative vascular
alterations precede the development of MCT-induced pulmonary hypertension
(Hislop and Reid, 1974; Meyrick and Reid, 1982; Masugi et al., 1965; Kay and Heath,
1966; Heath and Kay, 1967), and hypertrophy is evident in a number of cell types in
the lung and other organs (Turner and Lalich, 1965; Masugi et al., 1965; Valdivia et

al., 1967a; Wilson and Segall, 1990). Perhaps most convincing are the protective

64

effects of diet restriction and the involvement of polyamines in the expression and
progression of the MCT-induced pulmonary vascular changes.

Diet restriction of MCT-treated animals is almost completely protective
against the lung injury and lethality of the toxicaut. However, without further
exposure to MCT, return to a normal diet (and consequent resumption of growth)
results in the rapid development of puhnonary injury and death (Hayashi et al., 1979).
Diet restriction of MCTP-treated rats is also somewhat protective, although less
markedly so, suggesting that protection is not due solely to decreased bioactivation
of MCT (Ganey et al., 1985). It is not clear whether a particular aspect of diet
restriction or general suppression of weight gain is most critical to the protective
effects; however, altered sodium intake has been ruled out as a potential factor
(Ganey et al., 1985). In any event, it appears that suppression of the growth of the
animal, or alteration of its nutritional status, results in suppression of whatever
mechanism it is that causes initial, irreversible changes in target cells to progress to
overt pulmonary vascular lesions.

Diet restriction increases mortality from hyperoxia, and this has been
attributed to compromised repair (i.e. inhibition of proliferation) in the lungs
(Elsayed et al., 1988). Among its myriad effects, diet restriction decreases polyamine
content and protein, lipid and DNA synthesis in the lung (Gail et al., 1977; Faridy,
1970; Elsayed et al., 1988). Refeeding after food restriction accelerates polyamine
and DNA synthesis above normal levels in the liver (Domschke and Soling, 1973),
and increases ODC activity in the lung as well (Elsayed et al., 1988). Polyamines are

necessary for optimal cell growth and proliferation, under normal circumstances and

65

following injury, and agents which alter PA metabolism have been associated with
reduced cell proliferation (Heby, 1981; Pegg and McCaun, 1982). It may be that
these small, organic cations are the link between diet restriction and protection
against MCT pneumotoxicity.

The polyamines have been implicated as important mediators in the
development of MCT-induced pulmonary hypertension. There is a sustained increase
in the activity of lung ornithine decarboxylase (0CD) in rats treated with MCT
(Olson et al., 1984b). This precedes the development of pulmonary hypertension and
right ventricular hypertrophy (Olson et al., 1984b). Lung S-adenosylmethionine
decarboxylase (SAM-DC) activity is also increased by MCT treatment. CDC and
SAM-DC are the rate-limiting enzymes in the polyamine synthetic pathway; they
regulate production of these organic cations, which are associated with cell growth,
proliferation and differentiation (Pegg and McCaun, 1982). levels of the diamine,
putrescine, and the polyamine, spermidine, are elevated in whole lung and pulmonary
arterial wall by day 4 and day 7, respectively, after MCT treatment; spermine content
is increased in lung and vessels at 21 days posttreatment (Olson et al., 1984a;
Orlinska et al., 1988). In addition, lung spermidine/spermine acetyltransferase (SAT)
activity and levels of the secondary polyamine, Nl-acetylspermidine, are increased by
MCT treatment (Orlinska et al., 1989).

Continuous treatment with a-difluoromethylornithine (DFMO), a
specific, irreversible inhibitor of ODC activity, decreases the levels of putresciue and
spermidine in the lungs and attenuates the medial wall thickening, increased PAP

and RVH associated with the pulmonary hypertensive phase of MCT pneumotoxicity

66
(Olson et al., 1984a; Olson er al., 1985). DFMO treatment also blunts the

development of pulmonary edema and attenuates the early, transient increase in
vascular reactivity which follows MCT treatment (Gillespie et al., 1985b; Olson et al.,
1985).

DFMO does not alter the bioactivation of MCT, thus does not afford
protection in this manner (Olson et al., 1985). When animals receiving MCT and
DFMO are supplemented with ornithine, the immediate precursor to the polyamines,
their lung polyamine levels again increase to those seen in MCT-treated animals and
the protective effects of DFMO are reversed (Olson et al., 1989), suggesting that
polyamines are important in the evolution of MCT-induced pulmonary hypertension.
More specifically, these data support the concept that cell growth and/or
proliferation is important in the expression of MCT-induced pulmonary effects:
increased polyamine levels are essential for cell growth in response to any stimulus
(Heby, 1981). The search for the specific stimulus (or, more likely, stimuli) for cell
proliferation in the MCT model continues.

Increased epidermal growth factor (EGF)—like immunoreactivity is
detectable in the perivascular regions of the lung in rats treated 4 days earlier with
MCT (Gillespie et al., 1989). This coincides with a time when platelet depletion
protects against the pulmonary hypertensive effects of MCT, and platelets may serve
as a source for this EGF. Chronic administration of exogenous EGF for 7 days
results in increased medial thickening in arteries 100-200 H in diameter, increased
PAP and increased lung polyamine content. Smaller muscular arteries are not

affected, but in general, the effects of EGF on the lung are quite similar to those

67

caused by administration of MCT. It seems likely that this smooth muscle cell
mitogen may serve as one stimulus for cell proliferation after MCT treatment.
Studies using EGF-neutralizing antibody may help to confirm the importance of this
mediator in the MCT model.

Much of the cell proliferation which occurs after administration of
MCT involves the smooth muscle cells of the pulmonary vasculature. In addition to
its anticoagulant effects, heparin has several properties that can interfere with
vascular remodeling. It inhibits platelet adherence and degranulation and thereby
eliminates this source of PDGF in the vessel wall (Heiden et al., 1977). Heparin also
inhibits smooth muscle cell proliferation and migration, both in viva and in vitra
(Clowes and Karnovsky, 1977; Clowes and Clowes, 1986; Hoover et al., 1980). It has
been shown to reduce significantly vascular remodeling, PAP and RVH in mice
exposed to chronic hypoxia (Hales et al., 1983), and is useful in preventing intimal
hyperplasia after balloon angioplasty as well (Clowes and Karnovsky, 1977).

Fasules et al. (1987) found that neither anticoagulant nor
nonanticoagulant heparin attenuated MCT-induced lung injury. They attributed this
lack of protection to a number of factors: increased inflammation, more widespread
injury, different mediators involved in stimulation of proliferation, different smooth
muscle cell responsiveness to heparin, or altered pharmacokinetics for heparin in the
MCT model. These results suggest that despite certain similarities in the chronic
pulmonary vascular alterations caused by hypoxia and MCT, there may be important

differences in the mechanisms underlying cell proliferation in the two models.

68
The evidence supporting a critical role for cell growth and altered

proliferation in the expression of MCT (P) lung injury is compelling. Numerous cell
types and mediators may be involved, and regulation of these phenomena is likely

to be quite complex. Nevertheless, this is an area that merits further investigation.

F. Widget;

The importance of the proliferative response in the development of
cardiopulmonary toxicity after MCT treatment is interesting, particularly in light of
the fact that the reactive metabolite, MCTP, is itself a potent antiproliferative agent
(Mattocks and Legg, 1980; Reindel and Roth, 1991; Reindel et al., 1991). Many of
the PZAs are potent antimitotic agents in the liver, inhibiting cell division of
hepatocytes but not interfering with cell growth or DNA synthesis and resulting in
the development of very large cells with large nuclei (Bull et al., 1968). Similar
effects have also been noted in type II pneumocytes after treatment with MCT
(Wilson and Segall, 1990). The antiproliferative effects of the PZAs have generally
been attributed to the ability of reactive pyrrolic metabolites to act as a bifunctional
alkylating agents and to crosslink DNA and other cellular macromolecules (Culvenor
er al., 1969).

Alkylation involves the interaction of an electrophilic center in one
molecule with a nucleophilic center in another, resulting in the formation of a
covalent bond. Electron-rich sulfur, nitrogen or oxygen atoms of cellular
macromolecules are frequent targets for alkylating agents in viva (Robertson et al.,

1977; Robertson, 1982). Monofunctional alkylating agents can be toxic, perhaps

69

through binding to active sites in proteins and inhibiting their activity. If some of the
proteins affected are those essential to cell division, monofunctional alkylation could
result in an inhibition of cell proliferation. Carbon 7 (C-7) of DHR, a pyrrolic
metabolite of MCT, covalently binds to thiol groups on cysteine and GSH in vitra,
suggesting that metabolites of MCT may alkylate thiol groups on proteins within cells
(Robertson et al., 1977).

Bifunctioual alkylating agents (eg. MCTP) have two reactive centers
and can function as crosslinking agents. Mattocks and Leg (1980) compared the
effects of pyrroles with mono- or bifunctional alkylating ability on a rat liver cell line.
The antimitotic effect of these compounds was dependent on the chemical reactivity
of the pyrroles and their mono- or bifunctional activity. Some of the highly reactive
monofunctional pyrroles demonstrated weak antimitotic ability, but bifunctional
pyrroles with similar reactivity were much more effective antimitotic agents.
Furthermore, a number of potent agents used in cancer chemotherapy are
bifunctional alkylating agents, and their cytotoxic potential has been associated with
their ability to crosslink DNA (Tokuda and Bodell, 1987; Bodell, 1990). For the
most part, investigators in this field equate decreased colony-forming ability with cell
death; thus, the cytotoxic effect most commonly associated with DNA crosslinking is
inhibition of cell proliferation.

Pyrrolic derivatives of MCT crosslink DNA in vitra (Robertson, 1982;
White and Mattocks, 1972; Hincks et al., 1991), and DNAzDNA interstrand crosslinks
and DNAzprotein crosslinks are detectable in livers of rats treated with MCT in viva

(Petry et al., 1984). The initial interaction of DHR involves C—7 of the pyrrole and

70
N2 of deoxyguanosine (Robertson, 1982). Recent work with MCTP demonstrates

that both the C-7 and 09 positions of MCTP can react with a number of uucleosides
at their nitrogen atoms, which could indicate a much less specific interaction with
DNA than was previously supposed (Niwa et al., 1991).

Crosslinking in rat liver is maximal 12 hours after MCT treatment, and
slowly decreases so that it is indistinguishable from controls by 96 hours.
Interestingly, a single administration of PZA can result in prolonged antimitotic
effects in the liver (Culvenor et al., 1969; Jago, 1969). Accordingly, it is difficult to
correlate inhibition of proliferation with DNA crosslinking in viva, as the former
effect is persistent while the latter is transient. Perhaps the antimitotic effects of
MCT and other PZAs are not due to the actual presence of crosslinks in the DNA,
but are due to the consequences of repair of these lesions. Alternatively, the assay
may not be sensitive enough to detect a small but persistent fraction of crosslinks.
Further investigation will be required to resolve this issue.

It is possible to protect against the liver toxicity caused by MCT and
the PZA-containing plants, Senecia jacabea and S. vulgan's, by adding
mercaptoethylamine or cysteine to the diet (Hayashi and Lalich, 1968; Buckmaster
et al., 1976). These agents have sulfhydryl groups which may bind the reactive PZA
moieties, limiting their ability to alkylate cellular macromolecules. Cysteine may also
protect by increasing cellular GSH levels. Administration of the antioxidant,
ethoxyquin, is protective against the lethal effects of MCT in mice. Other

antioxidants are not protective, suggesting that ethoxyquin’s ability to increase

71

glutathione S-transferase activity may be responsible for its beneficial effects
(Miranda et al., 1981a).

It is possible that the interaction of a reactive metabolite of MCT with
DNA or other nucleophiles in cells of the pulmonary vasculature is the most
immediate consequence of exposure to this compound. The delayed consequences
of such an interaction may initiate a cascade of events that results in the
development of the characteristic pathophysiology we associate with MCT
pneumotoxicity. The lungs of MCT(P)-treated animals have not been examined for
DNA crosslinking, nor has anyone looked at the ability of cysteine and other
sulfhydryl-containing compounds to protect the lung from MCT(P)-induced injury.
Studies of this type would provide important information with regard to the
importance of crosslinking to the development of pulmonary vascular injury. In
addition, there is some question as to the nature of the MCT metabolite that reaches
the lung vasculature. The identification of DNAzDNA or DNAzprotein crosslinks in
the lung would support the concept that MCTP or another bifunctional metabolite

is directly involved in the injury.

From this discussion, it is apparent that the mechanisms involved in the
induction of lung injury and the development of pulmonary hypertension in this
model are neither simple nor fully understood. The complexity of the response to
MCT suggests that a single mechanism is unlikely to account for all aspects of the
toxicity. More realistically, the chronic cardiopulmonary changes which result may

be the cumulative response to effects on a number of vascular and parenchymal cells.

72

A dissection of the model to its component parts may provide information about the
direct effects of MCT (P) on specific cell types and the possible consequences of these

effects on the development of lung vascular injury.

V. V n h 1'

Damage to the endothelium is a common feature of several models of lung
injury and pulmonary hypertension. Situated at the interface of the blood and the
vascular wall, the endothelium is first among lung cells to encounter blood-borne
mediators and toxicants. Once thought to serve as a passive barrier, the endothelium
is now recognized for the important role it plays in the maintenance of vascular
homeostasis. This tissue is capable of generating, modifying and responding to a
variety of bioactive substances, and alterations in endothelial cell function may be
central to many forms of vascular and systemic disease.

There is a massive body of literature which focuses on the endothelial cell in
both healthy and diseased states. Although it is not possible to review this work in
detail, I will attempt to give a brief sketch of normal endothelial cell structure and
function, as well as an overview of the response to injury, in order to set the stage
for a discussion of the involvement of the endothelium in MCT(P)-induced

pulmonary vascular disease.

A Stimuli;
The vascular endothelium covers the intimal surface of the blood

vessels and consists of a single layer of flattened, polygonal cells. In profile, ECs

73
exhibit an attenuated cytoplasm with a bulging nucleus. They range from 10-50 is in

diameter and from 0.1-0.5 11 in thickness. In viva, ECs tend to orient themselves
longitudinally in the direction of blood flow (Wehrmacher, 1988). The morphologic
character of the endothelial layer, like many other features of ECs, varies with the
vascular source and may be continuous as in the brain or lung vessels, fenestrated as
in glomerular capillaries, or discontinuous as in the spleen,liver or bone marrow
(Renkin, 1988).

Electron microscopy reveals that ECs are covered with many finger-like
projections and caveolae, greatly increasing their surface area (Ryan, 1982c). The
luminal surface of ECs is covered with a carbohydrate-rich coating called the
glycocalyx. This is a specialization of the plasma membrane consisting of
glycosaminoglycans and the polysaccharide side chains of plasma membrane
glycoproteins and glycolipids. The glycocalyx is typically 15-40 nm thick and gives the
surface of the EC a fuzzy appearance on electron micrographs (Luft, 1966; Parson
and Subjeck, 1972). A number of plasma proteins are found adsorbed to this
carbohydrate surface, including albumin, az-macroglobulin, lipoprotein lipase,
fibrinogen, antithrombin III, heparin cofactor II, protein C and protein S
(Wehrmacher, 1988). The glycocalyx is proposed to play a role in regulation of
endothelial permeability by influencing the distribution of anionic sites on the cell
surface, and it may serve as a size, shape and charge barrier to the movement of
small molecules (Simionescu and Simionescu, 1986). It also may act to mask surface
receptors, thus playing a role in recognition between ECs and blood elements (Ryan,

1986).

74

ECs are able to synthesize their own subcellular matrix, which consists
of collagen (Types 1, III, IV or V), elastin, and the glyc0proteins laminin and
fibronectin. The composition of the subendothelial matrix varies with the state of
growth, cellular environment, vessel type and species and can influence EC
proliferation, migration, spreading and multicellular organization (Sage, 1984; Jafi'e
et al., 1976; Carnes et al., 1979; Madri and Williams, 1983; Jaffe and Masher, 1978;
Gospodarowia and Ill, 1980; Form et al., 1986). The ablumiual surface of the EC
is anchored to the subendothelial matrix by specialized matrix binding proteins
known as integrins. These proteins bind to laminin, fibronectin, collagen and
vitronectin and are involved in securing ECs not only to the matrix but also to
adjoining ECs (Macarak and Howard, 1983; lampugnani et al., 1991).

Fibronectin is not restricted to the subendothelium but covers all
surfaces of the EC and is released to the intercellular stroma and blood or culture
medium, as are collagen and elastin. A layer of fibronectin is essential for EC
adhesion, spreading and locomotion. In addition, it interacts with the
glycosaminoglycans, dermatan sulfate, heparin sulfate and chondroitin sulfate, which
are synthesized by ECs and deposited on the cell surface or released to the
subendothelium and blood or medium. The glycosaminoglycans bind lipoprotein
lipase and several coagulation factors to the endothelial surface, trap lipid in the
vessel wall, and play a role in EC regulation of SMC proliferation (Barnes et al.,
1978; Buonassisi, 1973; Castellot et al., 1982; Camejo et al., 1985; Colburn and

Buonassisi, 1982).

75

ECs are interconnected a three types of junctions: occluding (tight),
adherens, and connecting (gap) junctions. Desmosomes are generally not present,
however there are occasional desmosome-like regions evident at the ablumiual
plasma membrane. The presence or absence of a particular type of intercellular
junction depends a great deal on the vessel being considered. For example, capillary
ECs do not have gap junctions; postcapillary venular ECs are joined by modified
tight junctions (Larson and Sheridan, 1982; Simionescu et al., 1975; Franke et al.,
1988). In addition, ECs in confluent monolayers are connected by endothelial growth
inhibitory protein (EGIP) and calcium-dependent vascular cadherin (V-CAD).
Disruption of these intercellular attachments may result in stimulation of endothelial
proliferation (Heimark and Schwartz, 1988). ECs also send processes through the
subendothelial matrix and internal elastic lamina to form myoendothelial junctions
with the underlying SMCs, and these two cell types communicate via gap junctions
as well (Wehrmacher, 1988; Clowes et al., 1989).

In addition to the integrins and cadherins mentioned previously, two
other forms of EC-related cell adhesion molecules are associated with the plasma
membrane. The selectins and addressins are responsible for interaction of white
blood cells with the vessel wall (Lelkes, 1991). These molecules are important in the
EC response to inflammation, and they will be discussed in this context later.

EC contraction, spreading, migration, proliferation, and attachment to
other cells and the subendothelial matrix are all activities mediated by the
cytoskeleton. Vascular ECs contain microtubules and actin microfilaments which are

highly dynamic and contribute to a number of cell functions. Desmin and cytokeratin

76

intermediate filaments are evident in the ECs of some vascular beds in select species,
but vimentin is the only intermediate filament present in all mammalian ECs. These
intermediate filaments are less dynamic and may be involved in some way with the
state of cell differentiation. In general, cytoskeleton-mediated activity is involved in
the maintenance of endothelial integrity and control of vascular permeability; a
functional cytoskeleton is also critical to the monolayer repair process (Gottlieb er
al., 1991; Alexander et al., 1991).
B. W [Figure 2l
l. Regulatign Qf Vgculgr Peggeability

One of the earliest recognized functions of the vascular
endothelium was its role as a barrier to movement of water, solutes and
macromolecules between the vascular space and the interstitium. An intact layer of
ECs was thought to serve as a predominantly physical barrier to fluid movement, as
a loss of endothelial monolayer integrity resulted in vascular leak and its
consequences. Although the basic structure of the endothelial layer is important to
its ability to regulate vascular permeability, it is now recognized that this regulatory
process is quite active and complex.

Transport pathways for fluids and solutes through unfenestrated
microvascular endothelium (e.g., pulmonary vascular endothelium) include passive
movement through the cell membrane, vesicular, small pore and large pore pathways
of the cell, and passage through the intercellular junctions. Water and small,
lipophilic molecules generally move through the lipid bilayer directly. Shuttling of

transcytotic vesicles from all membrane surfaces results in no net fluid movement,

77

 

 

 

 

 

 

 

 

 

 

 

 

 

Hemostasis
Lipid metabolism (lasaactjujz‘y
_ x
\ r —- L —————— — —————=:-“-_—:_*‘ ;g§f
Kraut/7 . .
. Perl/7930111 t
fiegzdatz on y

 

fingjagenesjs

Figure 2: Homeostatic functions of endothelium.

78

but does contribute to the equilibration of luminal and ablumiual fluids. Small and
large pore channels provide a continuous route for movement of water and solutes,
including plasma proteins, across the endothelial layer. Larger, lipid-soluble
molecules may move through the intercellular junctions by diffusion in the membrane
(Renkin, 1977; Renkin, 1988).

Fenestrated endothelial layers possess all of the pathways
mentioned above, but also allow transport of much larger molecules. Open
fenestrae, such as those found in the glomerular capillaries, are 700-800 A in
diameter and would thus restrict only the cellular components of the blood.
However, in this organ, the basement membrane and epithelium help to limit
perfusion of larger plasma proteins. In other vascular beds, the fenestrae are covered
by a thin diaphragm which restricts the movement of plasma proteins but allows freer
flow of water, ions and small molecules than is possible through continuous
endothelium. Discontinuous endothelial layers lack a continuous basement
membrane, and very large proteins and even cells are able to enter the bloodstream
at these locations (Renkin, 1988; Deen et al., 1983).

In addition to the passive means of transport, there are
endocytotic and transcytotic receptors which bind and transport larger molecules,
including low density lipoprotein (LDL), insulin, growth factors, and plasma transport
proteins such as albumin and transferrin. As was mentioned previously, the
endothelial surface is covered by a glycocalyx, and its proteoglycans, sialoconjugates
and adsorbed proteins provide a negative surface charge. Specialized coated pits

bind the more cationic ligands, but it has been proposed that charge-differentiated

79

microdomains exist along the surface of the cells which may serve as sites for
transport of anionic molecules that would otherwise be repelled (Simionescu and
Simionescu, 1986).

. 2. l i n V a ivi

Another important function of the endothelium involves its
ability to take up and metabolize circulating vasoactive molecules, generating both
active and inactive mediators. This function is especially critical in the pulmonary
endothelium, as the lung receives 100% of the cardiac output and can therefore
adjust the level of vasoactive substances present in the blood delivered to the
systemic circulation.

The numerous caveolae present on the surface of ECs serve to
increase the vascular surface area exposed to the blood and provide a specialized
microenvironment for the activity of metabolic enzymes. Several endothelial
ectoenzymes have been localized to these structures, including ACE, ATPase,
ADPase, 5’-nucleotidase, carboxypeptidase N and carbonic anhydrase (Ryan and
Ryan, 1985; Ryan, 1986). ACE is the carboxy terminal dipeptidase responsible for
conversion of Ang I to the active vasoconstrictor, Ang II, and degradation of
bradykinin (BK). Thus, activity of this enzyme tends to promote vasoconstriction
(Yang et al., 1970; Erdos, 1975; Ryan, 1982c; Ryan and Catravas, 1990). ACE
activity is much greater in ECs than in other cells of the vascular wall, and has been
used as a marker to identify this cell type in culture (Auerbach et al., 1982).

The endothelial ectonucleotidases (ATPase, ADPase and 5’-

nucleotidase) are the major regulators of circulating adenine nucleotide levels in viva.

80

AMP is relatively inactive in the blood vessels, but ADP and adenosine have marked
efiects on platelets and are vasoactive as well (Smith and Ryan, 1972; Pearson et al.,
1980; Slakey et al., 1988). Carboxypeptidase N is important in the pulmonary
clearance of anaphylatoxins (Ryan and Ryan, 1985; Ryan et al., 1982b), and carbonic
anhydrase presumably plays a role in the exchange of CO2 and bicarbonate in the
vessels (Ryan et al., 1982a).

Endothelial cells also possess monoamine oxidase activity which
is responsible for degrading the biogenic amines, SHT and NE. MAO is
intracellular, thus removal of these substances necessitates facilitated transport into
the cell (Hughes et al., 1969; Iwasawa et al., 1973; Cross at al., 1974). Enzymes that
metabolize histamine are also present in ECs, although not in all vascular beds or
all species (Robinson-White and Beaven, 1982). In addition, the lung vasculature
serves to clear Ang II, adenosine, numerous prostaglandins, and the vasoconstrictor
peptide endothelin (ET) from the blood. Angiotensinases A and C have been
identified in pulmonary ECs, and it is likely that the clearance of prostaglandins and
ET are endothelial-mediated as well (Gillis and Roth, 1976; Piper er al., 1970;
Johnson and Erdos, 1977; Pearson et al., 1978; Sirvio er al., 1990). Although
endothelial uptake and subsequent phosphorylation or deamination of adenosine
occurs, it is not clear how this compound enters ECs, and the importance of
endothelial clearance of adenosine in viva is unknown (Catravas et al., 1988a).

A number of receptors for vasoactive mediators have been
identified on the surface of endothelial cells, including a and B adrenergic receptors,

M1 and M2 muscarinic receptors, H1 and H2 histamine receptors, and receptors for

81
Aug II, SHT and bradykinin (Buonassisi and Venter, 1976; Gryglewski et al., 1988;

Borsum, 1991). Mechanosensors, which respond to shear stress by opening and
closing K+ channels, are also found on the endothelial cell surface (Lansman, 1988;
Olesen et al., 1988), and it is thought that ECs can sense changes in oxygen tension
as well (Pohl, 1990). These receptors and sensors allow the endothelial cells to sense
and react to changes in their environment and permit them to respond to stimulation
by generating their own set of vasoactive mediators.

Endothelium-derived relaxing factors (EDRFs) are released in
response to a number of stimuli, including acetyl choline (ACH; M2), NE (a2),
thrombin, BK, histamine (H1), SHT (SHT 1), adenine nucleotides, stretch, hypoxia,
shear stress, and others. As the name implies, the factors generated in response to
these stimuli cause relaxation of vascular smooth muscle; they also have a
disaggregatory effect on platelets (Furchgott and Zawadzki, 1980; Furchgott, 1984;
Gryglewski et al., 1988; Auerbach, 1981; Azuma et al., 1986). Nitric oxide (NO)
appears to be one of the relaxing factors released by ECs. In addition to its
vasoactive properties, it may also serve as a protective chemical barrier against
cytotoxic injury caused by superoxide (Palmer et al., 1987; Gryglewski et al., 1988).
Some of the agonists (e.g., ACH an M1 receptors) also cause ECs to release a
substance which acts on K+ channels of SMCs to cause hyperpolarization. This is
referred to as endothelium-derived hyperpolarizing factor (EDI-IF), and also results
in relaxation of smooth muscle (Taylor and Weston, 1988). In addition, several of
the mediators which cause ECs to release EDRFs also stimulate generation of PGI2

by ECs, which contributes to vasodilation and inhibits platelet activation. Of the

82

agonists mentioned, ACH and BK may be the most physiologically relevant in
maintaining the vasculature in an unconstricted and nonthrombogenic state under
unstimulated conditions (Radomski et al., 1987; Gryglewski et al., 1988; Furchgott and
Zawadzki, 1980).

In addition to relaxant substances, ECs can generate potent
vasoconstrictive substances in response to stimuli. Endotheliumderived contracting
factors (EDCFs) fall into three categories: 1.) vasoconstrictive arachidonic acid
metabolites (eg. PGH2 or TXAZ) are generated in response to Ang I or II, SHT,
substance P, adenine nucleotides, superoxide or stretch; 2.) an unnamed EDCF
produced in response to severe hypoxia; and 3.) the potent vasoconstrictor peptide,
ET, produced in response to Ang II, vasopressin, transforming growth factor B
(T GF,) or thrombin (Katusic and Shepherd, 1991; Yanagisawa et al., 1988;
Yanagisawa and Masaki, 1989). Several of these stimuli also release EDRFs, and
it is thought that local regulation of blood flow may be modulated by a balanced
generation of these endothelial-derived mediators.

3- Remuuuuutflemususu

The role of the endothelium in regulation of the hemostatic
system has been investigated in detail. [For a review of this subject, please see
Nawroth et al. (1985)] Any regulatory adjustment of this system involves a complex
interplay of anticoagulant and procoagulant activities. Under normal conditions, the
vascular endothelium must provide a nonthrombogenic surface for smooth,
unimpeded blood flow; therefore, the balance is shifted in favor of the ROS

anticoagulant functions.

83
In the previous section, the roles of EDRF and PGI2 in

vasorelaxation were discussed. These substances also decrease platelet binding and
aggregation (Radomski et al., 1987; Moncada et al., 1977) and therefore decrease the
thrombogenicity of the vessel wall. Vascular ECs have a variable capacity to produce
a number of other arachidonic acid metabolites, including the cyclooxygenase
products TXAZ, PGFq, PGDZ, and FGF”, and the lipid hydroperoxides, 5-, 12- and
15-hydroxyeicosatetraenoic acid (5-HETE, lZ-HETE and 15-HETE) and 9- and 13-
hydroxyoctadecadienoic acid (9-HODE and 13-HODE) (Nawroth et al., 1985;
Buchanan et al., 1985; Buchanan and Bastida, 1988). Some of these products are
Pro— and some are anticoagulant in nature, and again a balance must be maintained
in the production of these mediators.
In addition to EDRF and PGIZ, the lS-lipoxygenase product, 13-
HODE, plays an imortant role in maintaining a nonthrombogenic vascular surface.
This mediator is found in the cytosol of ECs but is able to inhibit platelet adherence
and activity at the EC surface. It also inhibits PGI2 synthase; thus, levels of PGI2
vary inversely with concentrations of 13-HODE (Buchanan et al., 1985). lS-HETE
inhi1>its synthesis of PGI2 and decreases platelet activity as well but tends to enhance
VeSSel thrombogenicity. It appears that the relative concentrations of these
00111I><>uuds influence the expression of adhesive moieties on a number of cell types,
including ECs and platelets. 13-HODE production is predominant under
unstinuulared conditions and inhibits cell adhesion; 15-HETE predominates after
Still'llllation or injury, and enhances cell adhesivity (Hopkins et aL, 1984; Buchanan

and Basuda, 1988).

84
The glycosaminoglycan, heparan sulfate, is produced by

endothelial cells and then binds onto the cell surface or in the subendothelial matrix.

Heparan sulfate binds antithrombin III, which inactivates thrombin and may also

inactivate coagulation factors IXa, Xa and XIIa (Castellot et al., 1982; Stern et al.,
1985).

Thrombomodulin (TM) and protein S are two more EC products

with anticoagulant function. TM is located on the surface of ECs where it acts as a
receptor for thrombin. Once bound to TM, thrombin loses its procoagulant ability
and instead activates the anticoagulant protein C; activated protein C in turn
inactivates factors Va and VIIIa. Protein S, also generated by ECs, facilitates the
activation of protein C by thrombin (Esmon and Owen, 1981; Stern et al., 1986).

Tissue-type plasminogen activator (tPA) and urokiuase-type
Plasminogen activator (uPA) are serine proteases which are generated by ECs in
resli>0nse to stimuli similar to those that release EDRF, including thrombin. tPA and
“PA give endothelial cells their fibrinolytic capacity, and the EC surface provides a
Site for assembly of the fibrinolytic components (Levin and Loskutoff, 1982; loskutoff
and Edgington, 1977).

The ability of ECs to metabolize adenine nucleotides, some of
which have potent effects on platelets, was discussed in the previous section.
ADPase is quite important in this regard, as ADP is a powerful aggregatory stimulus
for Platelets (Smith and Ryan, 1972; Ryan and Ryan, 1985). The maintenance of an
intaq endothelial surface is also important in preventing platelet adherence to the

suberldothelial stroma (Fajardo, 1989)-

85

Vascular endothelium has procoagulant activity as well, but this
is either negligible in unstimulated ECs or is overbalanced by the anticoagulant
activity under normal circumstances. For this reason, the procoagulant features of
ECs will be covered in the discussion of the response to EC injury.
4. ' V l r
ECs modulate the growth of SMCs (and other mesenchymal
cells) in the vascular wall by a number of mechanisms. The composition of the
extracellular matrix, which is influenced by a number of factors (see Section V.A.),
may in turn influence SMC phenotype, movement and proliferation (Carey, 1991).
It seems that under quiescent conditions, proliferation of vascular smooth muscle is
kept in check by inhibitory substances produced by ECs. Heparin inhibits smooth
muscle cell proliferation in viva and in vitra (Clowes and Karnovsky, 1977 ; Clowes
er 01—, 1989; Hoover et al., 1980). ECs produce heparan sulfate that is deposited into
the subendothelial matrix and stroma adjacent to the SMCs. Factors present in the
b1°°d release heparin-like activity from this bound heparan sulfate which could
accollnt for the growth-inhibitory effects of confluent ECs on SMC proliferation
(CaStellor et al., 1981; Castellot et al., 1982; Herman and Castellot, 1987). In
addition, extracellular heparan sulfate binds proteins of the fibroblast growth factor
fun-i138 and thus regulates the growth of vascular cells by modulating the availability
and Stability of these mitogens (D’Amore, 1990).
Growth may also be modulated to some extent by gap junctional
mumlunicatiou between ECs and SMCs (larson and Sheridan, 1932; Davies et al.,

1985 ; Davies, 1986). This type of intercellular communication does influence cell

86
proliferation in other cell types (Loewenstein, 1979). However, although these

junctions exist between ECs and SMCs, their role in this process has yet to be
determined.

ECs synthesize and release a number of peptide growth factors
which are capable of stimulating the proliferation of SMCs and other mesenchymal
cells- Again, the production of these mediators is negligible in the undisturbed vessel

wall, therefore endothelium-derived growth factors will be discussed as a part of the
response to injury.
5. We:

The endothelium plays an active role in the uptake and handling
0f lipids by the vessel wall. ECs accumulate acetate and mevalonate and incorporate
them into sterol (Borsum, 1991). They also have surface receptors that bind low
density lipoprotein (LDL), high density lipoprotein (HDL), B-very low density
1iPoprotein (fi-VLDL) and chylomicrons. Both transcytotic receptors and endocytotic
rec<=ptors are present, thus lipids may be transported across the endothelium or may
be degraded or oxidatively modified within endothelium (Reckless et al., 1978; Morel
et ‘11—, 1984). ECs, like macrophages, also possess receptors to scavenge modified, or
acctylated, LDL. This property is used in characterization of cultured ECs, but it
may also contribute to EC damage in the presence of large amounts of modified lipid
(VOYta et al., 1984). The glycosaminoglycans present on the EC surface bind

llpo131‘otein lipase, therefore ECs are also able to hydrolyze triglycerides (Camejo er
“I" 1985).

87

Angiogenesis, both during growth and in response to injury or

neoplasia, appears to be an EC-mediated process. It involves a fibrinolytic-type
response of ECs (primarily venular) to a number of chemotactic and mitogenic
substances, including acidic and basic fibroblast growth factors (aFGF and bFGF),
EGF, PGEI, PGEz, tumor necrosis factor a (TNF,) and fibrin. The angiogenic

process has been reviewed in detail by Folkman (1982) and Auerbach (1981).

 

The most obvious result of endothelial injury is the development
0f interstitial edema as a result of increased vascular leak. A loss of the structural
and functional integrity of the endothelial monolayer occurs during inflammation
Primarily as a result of changes in the postcapillary venules in response to
inflammatory mediators (Pober, 1988; Pober and Cotran, 1990; Majno et al., 1967).
with discontinuous tight junctions, venular endothelium is somewhat leaky under
nol‘Inal circumstances, and this leakiness is enhanced during inflammation.
EEldothelial barrier function can also be impaired in capillaries or precapillary vessels
by oxidant injury (Shasby et al., 1982) or chemical damage (Meyrick et al., 1989;
Plestina and Stoner, 1972).

Alterations in certain cellular macromolecules represents one important
Inechanism leading to increased vascular permeability. For example, cytoskeletal
elements are important in determining EC shape and in maintaining their

conIlections with other cells and the subcellular matrix. Disruption of the

88

cytoskeleton occurs with some forms of EC injury, resulting in retraction of individual
cells and a loss of intercellular contacts. These structural changes in the endothelial
monolayer can contribute to a loss of vascular integrity and consequent increases in
permeability (Phillips et al., 1988; Hinshaw et al., 1989; Molony and Armstrong,
1991). In addition, changes in the glycocalyx that covers the endothelial surface may
influence permeability to a number of substances in addition to water by masking or
unmasking transport receptors, or by altering the surface charge distribution
(Simionescu and Simionescu, 1986).
2. ra i in V o ivi

Endothelial injury may result in disruption of the balanced
PrOduction of relaxing and constricting factors, leading to pathologically altered
Vascular responsiveness and tone. Many of the protective attributes of EC function
appear to be lost or depressed with mechanical or toxic injury, resulting in a vascular
bed that has little ability to protect itself from the direct effects of circulating
vElsoactive mediators. [Figure 3]

EC production of PGI2 is often increased as a result of EC
inqu or stimulation, perhaps in an effort to modulate the vasospastic and
pr Occagulant responses to vascular damage (see below). However, injured vessels
may also lose the ability to generate EDRFs and the vasodilatory prostaglandins.
This may be due to frank denuding of the luminal surface of the vessel, but it is more
fictiuently due to a decreased endothelial capacity to generate these substances.
Se"eral of the stimuli that cause the release of EDRF also stimulate generation of

EJDCZFS, and this ability is often maintained or increased in injured vessels. These

 

 

 

 

 

 

 

 

 

 

 

 

  

89

 

 

 

 

 

 

Figure 3: Endothelial modulation of vasoactivity in health and disease.

 

 

 

Health Disease
vasoactivity
angiotensin I
“CE _angiotensin 11
%
7i ”Iifié’fii't‘il')
, Bradykinin \
0 \ (active) 0 7
a R
I s 0 sur 3
0
I; 5 3
L N
7 a S
l r '
r 3
° 3:
“ P812 fl: ;
j EDRF g
- lrcrz. %
f song
7 K Leaner ‘.

 

 

90

changes in EC responsiveness shift the balance to favor vasoconstriction (Rubanyi,
1991; Shepherd and Katusic, 1991; Katusic and Shepherd, 1991; Chandler and Giri,
1983; Cartier et al., 1991; Adnot et al., 1991; Cartier et al., 1991). Also contributing
to the pressor response are agents like SHT and ACH, which stimulate the release
of EDRF in healthy vessels but are themselves direct vasoconstrictors. With
increased permeability or actual loss of the intervening endothelium, these mediators
gain direct access to receptors on the surface of SMCs and stimulate contraction
(Rubanyi, 1991; Becker, 1991; Ludmer et al., 1986).

EC injury can also interfere with the metabolism of circulating
vasc>active mediators, resulting in altered blood concentrations of these compounds
and their metabolites. Although ACE activity and levels of Ang II are increased in
some models of lung vascular injury (Brizio-Molteni et al., 1984; Zakheim et al.,
19‘76), decreases in enzyme activity are more frequently reported and may serve as
an indicator of pulmonary endothelial or microvascular dysfunction (Ryan and
Catravas, 1990; Catravas et al., 1988b). Loss of pulmonary ACE activity with a
concurrent increase in plasma ACE activity has been reported after administration
of agents which cause acute EC injury and may reflect sloughing of the enzyme from
the Surface of the EC (Hollinger et al., 1980b; Hdllinger et al., 1980a). Decreased
ACE activity may also be a compensatory response of the endothelium to chronic
eleVations in pulmonary blood pressure, or may occur in response to direct EC injury
(Keane et al., 1982; Chandler and Giri, 1983; Rounds et al., 1985; Stalcup et al., 1979;
Bell and Madri, 1990).

91

Some models of lung vascular injury result in a decreased ability
to remove circulating biogenic amines from the circulation, and this is considered to
be an indication of EC injury in viva (Block et al., 1985; Cross et al., 1974; Gillis,
1973; Iwasawa et al., 1973; Gillis et al., 1978). Inability to clear SHT or NE from the
blood could result in elevated concentrations of these mediators in the circulation,
thus exacerbating the situation described above. This may be of particular
imPortance when high concentrations of amines are released locally from platelets
or as a result of inflammation.
3. flieratigns in Hemostasis
The endothelium plays a critical role in modulating the
helrxlostatic process in response to injury or disease. The balance between
coa-gulation and fibrinolysis that results in a nonthrombogenic vascular surface under
“0111131 conditions shifts to favor a procoagulant state during injury, inflammation or
I1<=<>I)lasia. When functioning properly, EC act not only to initiate thrombosis but
“50 to limit the extent of this response. However, with some types of vascular injury,
the hemostatic response may become excessive; under these circumstances, the
hYIDer-coagulative state can actually contribute to the disease process. [Figure 4]
Tissue factor (TF; thromboplastin, factor III) is a cellular
cof‘I'iCtor that binds factor VII and initiates the coagulation process. The subintimal
Vaseular wall contains high levels of TF, but quiescent ECs express very little of this
proVain. Thus, the EC layer provides a barrier between the blood and this potent
Stilllulus for coagulation. Disruption of the intimal layer results in exposure of TF,

and a thrombotic response is initiated rapidly (Maynard et al., 1977; Nawroth et al.,

Health
1

P

/// RD
\

nDPase

Y no

 

 

  

92

Disease
Hemostasis

Factor 0

Factor XIIa

I

    

 

 

 

 

\

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 4:

   

 

P812
EDRF

protein c \,
(inactive) U tissue
Factor Ullaex ”“9" IF f
Protein C \\ ‘\‘iTF -
(active) Factor 011 r' TF
j?) TF
Factors Ua & 01113
i
.j
KS-) Thromhin f
Factors Ixa, X3, 1
XIIa
90 c
0 t3
(-) Platelets ( )Pl t 1 t
0 O \ ‘9‘ a e e S
13““00“: 15 rlien: "W l
i

u-Plasminog

 

 

 

 

new ‘1

t-Plasminogen activator

en activator (_)

Plasminogen
activator
Inhibitor 1

 

    
  

 

 

 

 

 

 

 

 

 

 

Endothelial modulation of the hemostatic system in health and disease.

93

1985). Perturbation of cultured ECs can result in increased expression of TF by
these cells (Galdal, 1984; Jaffe, 1987); it is not yet known if this is a response to
injury in viva as well. If so, this would permit an intact but injured endothelium to
initiate the coagulation process.

Just as the EC surface can serve as a site for assembly of the
fibrinolytic components (see above), so can it serve as a base for propagation of the
coagulation process. Factor IX is bound on the surface of ECs, where it can be
activated by factor VIIa. Factor IXa then promotes activation of factor X, which can
also bind to the EC surface (Stern er al., 1985; Nawroth et al., 1985). Bound factor
Xa can then bind and activate prothrombin. ECs also synthesize and bind
coagulation factor V and promote activation of factor XII (Cerveny et al., 1984;
Fajardo, 1989; Wiggins et al., 1980).

ECs synthesize, store and secrete von Willebrand factor (vWF;
factor VIII-associated antigen). This is a carrier protein for factor VIII, but vWF is
also important in mediating platelet adherence to the subendothelium during
denuding injury (Hoyer et al., 1973; Jaffe et al., 1974). In primates, cattle and rats,
vWF is concentrated and stored in Weibel—Palade bodies within the EC cytoplasm.
Identification of these structures, or immunolocalization of vWF in the cytoplasm,
helps to confirm the identity of EC in culture (Reinders et al., 1984; Wagner et al.,
1982). Mechanical or toxic injury to endothelium leads to increased intracellular
vWF, primarily associated with the endoplasmic reticulum, and increased vWF in the
subendothelial matrix (Reidy et al., 1989), and circulating levels of vWF increase

after irradiation injury or during renal failure. Thus, vWF may contribute to the

94

hypercoagulable state which exists with some forms of vascular disease, and increased
plasma levels of vWF may serve as an indicator of endothelial injury (Sporn et al.,
1984; Warrell et al., 1979).

ECs may promote hemostasis by modulation of the fibrinolytic
system. ECs produce tissue plasminogen activator inhibitor 1 (PAI-l), which limits
the activity of tPA and thus decreases fibrinolysis (Loskutoff and Edgington, 1977).
Injured ECs may also release more of the platelet activating and vasoconstrictive
lipid mediators, such as lS-HETE and TXAZ, and less 13-HODE. Stimulated ECs
may continue to release PGIz, but perhaps at insufficient levels to overcome
procoagulant influences (Chandler and Giri, 1983; Buchanan and Bastida, 1988;
Ingerman—Wojenski et al., 1981; Cotran, 1987; Jaffe, 1987). As noted above,
denuding and nondenuding injury to blood vessels can result in a decreased ability
to generate EDRFs, and this would also contribute to enhanced platelet adherence
and thrombosis.

4. V l r l r

Quiescent ECs in viva do not generate or release mitogens to
an appreciable extent; in fact, the vasodilatory and antithrombotic mediators that are
constitutively released by these cells (eg. EDRF or P612) actually inhibit SMC
proliferation (Shepherd and Katusic, 1991). However, after toxic or physical injury
or when stimulated by inflammatory mediators, ECs synthesize and release a variety
of substances that are potent SMC chemoattractants and / or mitogens. These may
be important in initiating or exacerbating the structural remodeling which is a part

of many chronic vascular diseases. [Figure 5]

He;

 

95

th . D'
Heal Growth Regulat1on lsease

V§s

\\

rcr,//

 

\\\~

Via

§§

     
 

c-csr
(‘l cn-csr
IL-1

 

 

 

’\

sh

 

\\‘\‘\‘

'HA
3'39
'V'Iv

\\\\\

_«e

/I\\
V§§a

us
‘II
FGF

\\\\\\\\

A
4
v

\\

 

PDGF

a, \\

 

Man-0H
'H'l'l
\\

 

 

Figure 5: Endothelial modulation of vascular cell growth in health and disease.

96

Many of the vasoconstrictive mediators stimulate SMC
proliferation. This is true for ET, Ang II, and the catecholamines and may be the
use for all vasoconstrictors. Similarly, many of the peptide growth factors have
vasoconstrictive properties (N emecek et al., 1986; Campbell-Boswell and Robertson,
1981; Yanagisawa and Masaki, 1989; Blaes and Boissel, 1983; Kourembanas et al.,
1990; Gibbons and Dzau, 1990; Shepherd and Katusic, 1991; Dzau and Gibbons,
1991). Their common ability to increase intracellular calcium may account in part
for this correspondence (Ryan et al., 1988; Dzau and Gibbons, 1991; Berk et al.,
1987; Shepherd and Katusic, 1991).

Mitogens produced by ECs include a platelet-derived growth
factor-er substance (PDGF) (DiCorleto and Bowen-Pope, 1983; Fox and DiCorleto,
1984; Vlodavsky er al., 1987), acidic and basic fibroblast growth factors (aFGF and
bFGF) (Schweigerer et al., 1987; Vlodavsky er al., 1987; Burgess and Maciag, 1989),
insulin-er growth factor 1 (IGF—l) (Dzau and Gibbons, 1991), endothelial cell-
derived growth factor (ECGF) (Gajdusek et al., 1980), granulocyte and granulocyte—
macrophage colony stimulating factors (G-CSF and GM-CSF) (J affe, 1987; Zsebo et
al., 1988) and the cytokine IL-l (Shanahan and Korn, 1984; Giri et al., 1985 ; Mizel,
1989). All of these are mitogenic for SMCs, and some are chemotactic as well
(Grotendorst et al., 1982; Clowes et al., 1989). A few (e.g., aFGF, bFGF) are also
auto-stimulatory and cause EC proliferation (Schweigerer et al., 1987; Sato and
Riflcin, 1988); IL-l inhibits BC proliferation (Cozzolino et al., 1990). Stimulated ECs
also release a connective tissue growth factor which is related to but distinct from

PDGF, a hepatocyte growth factor which is hypothesized to be a ubiquitous factor

97
for tissue repair (Bradham et al., 1991; Noji et al., 1990), and TGF“, which promotes

angiogenesis, inhibits EC proliferation, and modulates SMC proliferation
(Antonelli-Orlidge et al., 1989; Assoian and Sporn, 1986) .

The release of growth factors from ECs in combination with
other possible alterations as a result of injury, such as loss of cell attachments, loss
of gap junctional communication, or alteration of the composition of the extracellular
matrix, may result in an environment favorable to SMC migration and proliferation
(Simionescu and Simionescu, 1986; Ryan, 1986; Carey, 1991). In addition, the
increased platelet adherence which may occur as a result of EC injury could result
in increased release by platelets of growth factors, most notably PDGF, TGF, TGF,3
and EGF (258,422,423}, which could also contribute to the vascular remodeling
process. Platelets also release thrombin, which is a potent stimulus for EC
production of growth factors and other mediators (Schini et al., 1989). Increased
production and release of mesenchymal cell growth factors by ECs has been
documented under a variety of adverse conditions (Hsieh et al., 1991; Madri et aL,
1991; Vender et al., 1987; Kourembanas et al., 1990; Albelda et aL, 1989; Fox and
DiCorleto, 1984) and may represent enhancement of the repair process involved in
wound closure.

5. " 'v " n e i ll
ECs, particularly those of the postcapillary venules, respond to
cytokines and other inflammatory mediators in a manner quite similar to
macrophages; ECs in this stimulated state are said to be "activated" (Cotran, 1987).

[Figure 6] Agents to which endothelial cells respond include endotoxin, thrombin,

98

    

 

I I
Interleukin 1 . “Ctluatlon
Ill-or Necrosis Factor
Thronbin
Others
lnterleukins 1,6,8 vHF
Tumor Necrosis Factor t4,“
Others PHI-1
Complement

activation

 

Figure 6: The "activated" endothelial cell.

99
IL-l, tumor necrosis factor (TNF), phorbol esters, lymphotoxin, and others. During

the activation process, ECs undergo a number of structural and functional alterations.
They become plumper and more prominent than quiescent ECs, and biosynthetic
organelles increase. There is a reorganization of the actin and microtubule networks
of the cells and a redistribution of some of the focal contact proteins, but interaction
with the matrix is not disrupted (Molony and Armstrong, 1991). Thus, activated cells
retract and change shape, but generally do not detach. Gaps appear between the
retracted cells and vascular permeability increases. Activated cells become more
mobile and they divide more rapidly than quiescent endothelial cells (Cotran, 1987).
They are also able to act as phagocytes and engulf viruses and other small particles
0.1-20 u in diameter (Ryan, 1988).

Addressins are members of the immunoglobulin superfamily and
include intercellular adhesion molecules 1 and 2 (ICAM-1 and ICAM-2), vascular
cell adhesion molecule 1 (VCAM-1) and platelet-endothelial cell adhesion molecule
1 (PECAM-l). Although ICAM-1 may be expressed by unstimulated ECs, ICAM-2
and VCAM-1 are only expressed on the surface of activated ECs, predominantly
those of postcapillary venules; all three of these addressins are targets for lymphocyte
homing molecules. PECAM-l is more evident at the intercellular junctions of
activated ECs and is a site for platelet adherence on the EC surface (Marlin and
Springer, 1987; Osborn et al., 1989; Newman et al., 1990; Albelda and Buck, 1990).
Another set of adhesion molecules expressed on the surface of activated ECs are the
selectins. These include endothelial leukocyte adhesion molecule-1 (ELAM-1) and

granule-membrane protein 140 (GMP-140). ELAM-1 is expressed primarily on

100
venular ECs and binds PMNs. GMP-140 is present on the surface of Weibel-Palade

bodies within ECs. With activation, these adhesion molecules are inserted into the
cell membrane where they bind PMNs (Bevilacqua et al., 1987; Johnston et al., 1989).

BC activation results in the unmasking of PC and C3b receptors
on the endothelial surface, perhaps through alteration of the glycocalyx. Activated
ECs express major histocompatibility Class I and II antigens and the ABO blood
group antigens. The protein factor, Clq, is found on their surface as well, which
binds circulating immune complexes and activates complement (Ryan et al., 1981;
Wedgewood et al., 1988; Jaffe et al., 1973).

Activated ECs not only respond to inflammatory mediators, but
they release a tremendous array their own soluble mediators, including the cytokines
IL-1, IL—6, IL-8, TNF and lymphotoxin, which promote chemotaxis and activation of
PMNs and other leukocytes. They release factors that inhibit (tPA and PGIZ) and
promote (PAl-l and vWF) platelet activation and hemostasis. Activated ECs
generate complement components, including C5a, a potent neutrophil chemotactic
factor; they also produce PAF (see Section IV .D. above), a variety of growth factors
(see Section IV .E. above), and collagenases that can degrade collagens I-V, thus
influencing cell attachment and motility. They can also metabolize LTA4 to LTB4,
LTC4, LTD4 and LTE4, substances that attract PMNs and participate in anaphylaxis
(Miossec et al., 1986; Whatley et al., 1988; Warren et al., 1987; Cotran, 1987;
Bussolino et al., 1988; Ibe and Campbell, 1988; Modat et al., 1990). At our present

level of understanding, it is difficult to appreciate the potential consequences of this

drainatic, multifaceted response to stimulation. Many of these studies were

101

performed in vitro, and in a variety of cell types; a localized response to inflammatory
mediators in viva may result in a more limited spectrum of effects appropriate to the

stimulus.

In summary, the vascular endothelium performs a number of functions in
addition to its role as a barrier to fluid movement. In fact, ECs are a dynamic,
interactive component of the blood vessel wall, and their functional status must be
considered as well as their structural integrity when evaluating toxic injury to the
vasculature.

Study of endothelial function has revealed that the activity observed at a
specific time is actually the sum of many activities which are constantly being
modulated to achieve an appropriate balance for a specific set of conditions. This
balance of stimulation and inhibition, coagulation and fibrinolysis, quiescence and
activation, appears to allow the endothelium a broad spectrum of responses to
sublethal injury. ECs in different vascular beds and in different species demonstrate
heterogeneity. It has been proposed that many EC surface and metabolic properties
are inducible, not constitutive, and represent an adaptation of otherwise identical
cells to different vascular environments (Gimbrone, 1987). If this is true, then the
response of ECs to different types of injury may also be heterogeneous. In addition,
the importance of EC function to vascular homeostasis and the careful balance of
activities required for homeostasis suggests that EC dysfunction might contribute to

certain disease processes.

102
V1. W

A. MQIKEHnduged Endothelial Cell Alterations 14 Viva

Endothelial damage occurs in the pulmonary vasculature with MCT(P)
treatment in viva, and the morphological and functional evidence of this damage has
been discussed in previous sections. However, it may be useful to synthesize these
observations and review them in light of the preceding discussion of EC function.

Morphologic alterations in pulmonary ECs are evident as early as 24
hours after administration of MCT (P), and these include swelling, nuclear
enlargement, vesicular blebbing and an increase in the number of cytoplasmic
organelles (Merkow and Kleinerman, 1966b; Valdivia et al., 1967a,b; Rosenberg and
Rabinovitch, 1988; Meyrick and Reid, 1982; Reindel et al., 1990 and others). ECs
in alveolar wall vessels incorporate increased amounts of [3H]thymidine (Meyrick and
Reid, 1982), which is indicative of an increased level of DNA synthesis associated
with proliferation or DNA repair in these cells. EC hypertrophy, a rather unusual
response of these cells to toxic injury, has been noted in later stages of MCI‘ (P) lung
injury. Occlusion of capillaries and small arterioles with swollen, hypertrophic ECs
is thought to contribute to pruning of the peripheral vasculature (Butler, 1970a;
Hislop and Reid, 1974; Raczniak et al., 1979; Valdivia et al., 1967b; Meyrick and
Reid, 1979).

A number of other morphologic changes in the lung are suggestive of
altered EC function. The presence of interstitial edema and other exudative changes
suggest a loss of vascular integrity and increased capillary permeability in the lungs

of MCT(P)-treated animals (Wilson and Segall, 1990; Heath, 1969; Valdivia et al.,

103
1967b; Molteni et al., 1984; Sugita et al., 1983b). Decreased ability of the

endothelium to serve as a barrier is also reflected by an increased lung weight-to-
body weight ratio, increased release of protein into BALF, and increased
accumulation of circulating [1251]albumin in the lungs after treatment with MCT (P)
(Roth et al., 1981a; Roth, 1981b; Bruner et al., 1983,1986; Reindel et al., 1990). In
addition, the increased numbers of inflammatory cells noted in the lungs (Kay et al.,
1967b; Sugita et al., 1983b; Stenmark et al., 1985; Ilkiw et al., 1989; Schultze et al.,
1990) may be a result of EC activation with consequent increased production of
chemoattractants and altered adhesivity, as well as a loss of vascular integrity.
Intraluminal thrombi and fibrin blockages are seen both early (Valdivia
et al., 1967a) and late (Turner and lalich, 1965; Merkow and Kleinerman, 1966b;
Hayashi er al., 1967) in the course of MCT-induced lung injury and might be
indicative of an imbalance in endothelial production of pro- and anti-coagulant
substances. The observation that lungs from MCT(P)-treated rats demonstrate
decreased fibrinolytic ability supports this hypothesis (Molteni et al., 1988a, 1989b;
Schultze and Roth, 1992). Platelet thrombi are reported in the pulmonary
rnicrovasculature of MCT(P)-treated animals (Turner and lalich, 1965; Valdivia et
al., 1967a; Chesney et al., 1974a), and platelet sequestration occurs by 8 days after
treatment with MCTP (White and Roth, 1988). Changes in platelet activity or
adherence may reflect exposure of the subendothelial matrix or may occur as a result
of altered EC production/ release of coagulation factors or mediators such as EDRF

or PGIZ.

104

MCT-induced angiogenesis occurs at the pleural surface and in the

penbronchial region (Schraufnagel, 1990). ECs of postcapillary venules are sensitive
to angiogenic stimuli and generally initiate this response. Atheromatous plaques
have been reported in the coronary vasculature of MCT-treated rats, and might be
a result of altered lipid metabolism by ECs in these vessels (Chesney and Allen,
1973c). Finally, the muscularization of normally nonmuscular arterioles (Hislop and
Reid, 1974; Turner and lalich, 1965 ; Meyrick and Reid, 1979; Langleben and Reid,
1985; Reindel et al., 1990) and the increased medial smooth muscle seen in the small
pulmonary arteries (Turner and Lalich, 1965; Merkow and Kleinerman, 1966b;
I'Iislop and Reid, 1974; Heath and Kay, 1967; Kay and Heath, 1966; Reindel et al.,
1990) which are characteristic of MCT(P)-induced vascular remodeling may also
reflect EC dysfunction, since the endothelium plays an important role in the
regulation of SMC proliferation.

More direct evidence of functional change in ECs comes from
biochemical studies in the lungs of animals treated with MCT (P). Uptake and
Clearance of biogenic amines (e.g., SHT and NE) is reduced in isolated perfused
lungs and in lung slices from MCT(P)-treated rats (Huxtable er al., 1978; Gillis et al.,
1978; Roth et al., 1989; Hilliker et al., 1983c; Hilliker et al., 1983a; Hilliker et al.,
1984). This function has been localized to the endothelium, and altered clearance
0f amines has been used as a marker of pulmonary endothelial dysfunction.
Pulmonary clearance of prostaglandins is also reduced (Ito et al., 1988b; Ito et al.,

1988a). Changes in ACE activity have been reported but are controversial. Changes

 

3C1

105

in biochemical function of endothelium such as these may contribute to alterated
vasoactivity in MCT(P)-treated animals.

There is a loss of pulmonary vasodilation in response to ACH after
MCT treatment which suggests a decreased ability to generate EDRF (Ito et al.,
1988a; Altiere et al., 1986). PGI2 and TXA2 production are increased in lungs of
MCT(P)-treated animals, perhaps as a result of altered endothelial synthesis (Molteni
et al., 1984; Ganey and Roth, 1987b).

The important structural and functional properties of the endothelium
and its ability to respond heterogeneously to stimulation or injury have made this
tissue a focus in a number of models of systemic vascular injury, including
atherosclerosis and essential hypertension. The anatomic situation of the pulmonary
endothelium, just downstream from the liver and at the interface of the blood and
vessel wall, makes it a likely target for blood-borne toxic metabolites. Considering
the evidence demonstrating that changes in EC structure and function occur as a
result of treatment with MCT (P) in viva, endothelium would seem to be a logical
tissue on which to focus studies on mechanisms of MCT (P) pneumotoxicity.

B. h Ii 11 in r

The value of direct examination of an individual cell type has been
appreciated for many years; it permits identification of characteristic cell functions
and investigation into the direct effects of exogenous agents on those functions.
Isolated cell systems are also useful in the investigation of mechanisms of toxicity,
since they can facilitate the detection of very early and / or subtle alterations in cell

activity after exposure to toxicants (Ramos and Cox, 1987).

106

Although the cumulative mass of the endothelium in the body is nearly
a kilogram (Wolinsky, 1980), the fact that these cells exist in monolayers closely
apposed to other vascular cells makes it difficult to study their function in situ. For
this reason, much of what we know about EC function has been learned from work
done in cultured ECs (Borsum, 1991). Studies in viva are needed to confirm that
similar functions are carried out by ECs in situ; nevertheless, the isolated EC systems
currently available provide a means to study both EC biology and response to injury.

Isolated, unfixed ECs in blood vessels have been studied in "en face”
Hautchen preparations or attached to cellulose acetate paper (Pastan et al., 1977;
Smith and Ryan, 1973a; Tomlinson et al., 1991). These preparations are short-lived
and fragile and therefore of limited usefulness, but they are representative of the
endothelium in situ. A technique for the culture of ECs from the human umbilical
vein was first reported by Maruyama (1963). Since that time, a number of
investigators have developed methods for the isolation and culture of endothelial
cells from a variety of vessels and species, ranging from the aorta of the cow to the
pulmonary rnicrovasculature of the rat (Lewis et al., 1973; Habliston et al., 1979; Giri
et al., 1985; Booyse et al., 1975; Ryan and White, 1986). A variety of methods, both
mechanical and enzymic, have been utilized to collect EC cells; dozens of media
formulations have been advocated to sustain them; and a variety of propagation
methods have been used to perpetuate them, all with a common goal: a system in
vitra that is as representative of the situation in viva as possible.

In many respects, this goal has been achieved. ECs replicate well in

culture and form a contact-inhibited monolayer. Olltured ECs retain many of the

107

morphologic features that are characteristic of related ECs in situ, including the
presence of Weibel-Palade bodies, pinocytotic vesicles, surface projections and
caveolae, a glycocalyx and fenestrae (Ryan et al., 1978; Smith and Ryan, 1972; Smith
and Ryan, 1973b). ECs in culture carry out some of the biochemical and metabolic
functions which have been reported in viva. Among these are the conversion of Ang
Ito Ang II and degradation of BK by ACE (Ryan et al., 1976), uptake of acetylated
LDL (Voyta et al., 1984), and synthesis and storage of factor VIII-related antigen
(Wagner et al., 1982; Ryan et al., 1978), all of which are used in the identification and
characterization of cultured ECs.

Limitations exist in any isolated cell system and the EC in culture is no
exception. ECs in the body do not exist as an isolated mass of cells; they are in
constant contact and in communication with the surrounding matrix, smooth muscle
cells and blood. In addition, they are subject to physical forces such as shear stress
and mechanical stretch. The importance of these contacts to the maintenance of
normal cell function is becoming more evident as techniques to study the
endothelium in situ are improved.

The proliferative rate of normal vascular EC in viva is low, with
average doubling times that range from months to years. In culture, these cells have _
a population doubling time of 18-65 hours. This rate, influenced by the culture
medium and the concentration of serum and other biological additives used, is more
representative of ECs after wounding or under ne0plastic conditions than of
quiescent cells in viva (Cotran, 1965; Greenburg and Hunt, 1978; Duthu and Smith,

1980; Hobson and Denekamp, 1984). ECs in viva and in vitra have a limited

108

doubling potential and, unless the cells are transformed, a finite lifespan (Duthu and
Smith, 1980). Thus, a more rapid aging process occurs in cultured cells, and this
should be considered when evaluating structural and functional alterations in
response to treatment.

Although ECs in culture synthesize extracellular matrix components
which are similar to those formed in viva, it is difficult to achieve conditions that
allow them to synthesize a functionally intact basement membrane (Kramer et al.,
1984; Huber and Weiss, 1989). This has hampered permeability studies, as well as
investigation of the interactions between ECs and inflammatory cells. EC matrix
characteristics also affect cell shape, spreading, motility and proliferation, and these
may in turn affect cell function (Madri and Stenn, 1982; Gospodarowicz and Lui,
1981; Hennig et al., 1989; Madri er al., 1991).

Cultured ECs constitutively synthesize and release growth factors, most
notably PDGF (DiCorleto and Bowen-Pope, 1983). They can be stimulated to
produce large amounts of tissue factor, a capacity which has not been demonstrated
in viva (Jaffe, 1987). Cultured microvascular ECs tend to form tubes readily, thus
providing a model for study of angiogenesis (F olkman, 1982; Madri and Pratt, 1986).
These and other differences between ECs in viva and in vitra may be attributed to
an absence of the regulation provided by other cell types and mediators normally
present in the vascular wall. Despite such differences, the benefit of being able to
examine the direct effects of exogenous mediators and toxicants on a specific target
cell type is great, and EC culture remains a valuable tool. Certain nuances of

MCT (P) toxicity render this model a difficult one to study in viva with regard to

109

elucidating mechanisms of action on the lung vasculature. There is evidence in viva
to suggest that the pulmonary endothelium is a likely target for reactive MCTP
arriving via the bloodstream, and EC cultures are a simplified but useful model of
certain aspects of MCTP-induced toxicity.
C. - n 1i 1 Al r ' V

Reindel and Roth (1981) examined the time- and dose-dependent
effects of MCTP on cultured bovine pulmonary artery endothelial cells (BECs). BEC
injury in response to a single exposure to MCTP was delayed in onset and
progressive in nature, much like the pulmonary toxicity in viva. In the 24 hour period
after exposure to MCTP, monolayers of BECs showed no evidence of cytotoxicity as
evidenced by morphologic change or release of LDH. By 48 hours posttreatment,
there was a subtle increase in cell detachment followed by a rise in LDH release and
increased generation of PGIZ. These responses became progressively more
pronounced over the duration of the study. As cells detached, a monolayer free of
obvious gaps was maintained by enlargement and spreading of the remaining cells,
which in some cases enlarged as much as 20-fold. These cells were bizarre in
appearance, having enlarged, occasionally multiple nuclei with multiple, prominent
nucleoli, dilated end0plasmic reticulae and pronounced cytoplasmic stress fibers and
vacuoles; however, they remained viable as they were able to exclude trypan blue.
In addition, colony forming efficiency assays revealed that MCTP-treated BECs were
unable to divide; these antiproliferative effects were evident at lower concentrations

of MCTP than those needed to produce overt cytotoxicity (Reindel and Roth, 1991).

110

The repair processes which occur in endothelial monolayers after
wounding follow a predictable sequence which involves cell spreading, migration and
proliferation. Whereas the former two processes may be sufficient to repair small,
discrete defects in a monolayer, cell proliferation is required to restore monolayer
integrity when larger or more widespread defects are present (Wong and Gotlieb,
1985; Coomber and Gotlieb, 1990). In BEC monolayers treated with MCTP, defects
in the monolayer which result from the cytotoxic effects of MCTP are compensated
for initially by the spreading of adjacent cells. As cell detachment continues, the
remaining cells, which are unable to divide, become dramatically enlarged. In time,
when this response becomes inadequate, gaps appear in the monolayer. A similar
response by the endothelium in viva could result in the delayed and progressive
development of vascular leak which is characteristic of MCT(P) pneumotoxicity
(Reindel and Roth, 1991).

Porcine pulmonary artery endothelial cells (PECs) studied under
similar conditions respond somewhat differently from BECs (Reindel et al., 1991).
PECs are much less sensitive to the cytolytic effects of MCTP than are BECs,
exhibiting less cell detachment [Figure 7] and LDH release [Figure 8]. Although
they become more spindle-like in shape, they do not enlarge markedly. However,
treated PECs demonstrate a delayed and progressive increase in generation of PGIZ,
and they are equally sensitive to the antiproliferative effects of MCTP [Figure 9]
(Reindel et al., 1991).

The more subtle morphologic response of PEC monolayers to MCTP

treatment does not appear to be due to a qualitative difference in the response of

111
this cell type to injury. When MCTP-treated PEC monolayers are artificially

wounded to mimic the loss of monolayer integrity seen in treated BECs, the adjacent
PECs migrate into the defect and enlarge markedly [Figure 10] (Hoom et al., 1990).
These results suggest that, whereas BECs are much more sensitive to the cytolytic
effects of MCTP than are PECs, surviving cells of each species respond very similarly
to a loss of monolayer integrity. Both cell types are inhibited in their ability to divide
when treated with MCTP yet are viable and able to respond to a loss of cells from
the monolayer by cell migration and enlargement.

Cattle and swine differ in their target organ susceptibility to ingested
C. spectabilis; cattle typically develop veno-occlusive disease of the liver (Sanders et
al., 1936; Sippel, 1964), whereas swine may experience pneumotoxic changes in
response to ingestion of this plant (Emmel et al., 1935; Peckham et al., 1974). It has
been suggested that species differences in response to PZAs may be due to
differences in metabolism (Shull et al., 1976; Cheeke and Pierson-Goeger, 1983), but

the results of studies in vitra suggest that there are interspecies differences in ECs,

112

 

 

 

 

 

 

800--
a 0/0
"’0 / A
._ 600 <~ O A/
x /
v %‘ \O O ug/ml
A
g g 0.5 [lg/ml
8 400 0 k“
g o\. ‘l
53 200 .. \ \o .
2 \g ‘
O i r i §
1 2 3 4 5
AYS AFTER MCTP TREATMENT
800 T
n
c:
" 600 4-
X
at
.J
._J
:3 400‘»
u.
{so-K 0 Aug/ml
ICE) .ég .* t\g 0'5 I'l'g/r"l
33‘ mm \f*~_°t\ti 5 "9/m'
g "‘4 50 pg/ml
Z
O t i t t i
l 2 .3 4 5

DAYS AFTER MCTP TREATMENT

Figure 7: Effects of MCTP on monolayer cellularity.

BEC (A) and PEC (B) monolayers were exposed to a single administration of MCTP
on day 0. Values represent mean + /- SE of 5 (BECs) or 4 (PECs) independent
experiments. " Significantly different from control (p<0.05)

113

 

 

70’ 0—0 0 lug/ml l———-—-—-§
-—0 50 l a
60‘ . #g/m I/
LL] 0
‘32 l
U 50.
d .
a: 40 .
S 30« I
7: 1
g: 20. l
T: r/O/l ‘l
a o 6/9 1
o l 2 3 4 5

DAYS AFTER MCTP TREATMENT

 

 

704
u 60‘ o—o Orig/ml
m .__.. so ng/ml
:5 50~
d
0:. 40‘
S 304
E zol
>3 .——-—-—"'—""'_"'2
- lD‘ ‘ t
.e . .
a? O A£3:349\§
O l 2 .3 4 5

DAYS AFTER MCTP TREATMENT

Figure 8: Release of lactate dehydrogenase from MCTP-treated monolayers.

BECs (A) and PECs (B) were exposed to 0 or 50 ug MCTP/ml medium on day 0.
Each value represents mean + /- SE of 6 independent experiments. ’ Significantly
different from controls at same time (p<0.05)

114

1004—4 fi'°\.
80‘
601 .
40«

20‘ 9 e

COLONY FORMING EFFICIENCY
(25 OF CONTROL)

 

 

O [I [I r r r . .—
0 0.005 0.05 0.5 5 50
MCTP CONCENTRATION (pg/ml)

 

120- r
.

100+fl//L\o
80-
604

40-

COLONY FORMING EFFICIENCY
(95 OF CONTROL)

20‘ e o

 

 

0 71,1 U I U . .
0 0.005 0.05 0.5 5 50
MCTP CONCENTRATION (fig/ml)

Figure 9: Effects of MCT P on colony-forming efficiency.

BECs (A) and PECs (B) were plated at low density and treated with a single
adminsitration of MCTP or vehicle on day 0. Values represent means + /- SE of 4
independent experiments. " Significantly different from control (p<0.05)

115

  
 

3n; -
,n‘z' Nil, 3“

.', ‘
l l/ t

m.- xxx-t:
. '7" * 1 his!"
.2., 1‘ \. .

 

Figure 10: Effects of MCTP on PEC morphology.

PEC monolayers were treated with MCTP or vehicle and
scraped 4 hours posttreatment. Fixed and stained cells
were photographed at 5 days posttreatment.

116

at least in culture, which might also account in part for differences in sensitivity to
these compounds (Reindel er al., 1991). This area of investigation merits further
study.

In high concentrations, the parent alkaloid, MCT, is also directly toxic
to PECs and BECs in culture. Administration of millimolar quantities of MCT to
ECs results in pronounced vacuolation, decreased monolayer cellularity and increased
permeability to albumin within hours of treatment. The increased monolayer
permeability is attenuated by pretreatment of ECs with SKF-SZSA, an inhibitor of
the cytochrome P450 MFO system. Injured cells are also inhibited in their ability to
proliferate; they become hypertrophic and exhibit increased quantities of DNA and
increased polyamine content (Hacker et al., 1991; Salone et al., 1991). These results
suggest that pulmonary vascular endothelium is able to metabolize MCI‘ to a
bioactive form(s). MCT-induced cytolytic changes are evident much earlier than
those seen after administration of MCI'P, but this may be a phenomenon related to
the presence of high levels of MCT. The identity of the putative cytotoxic
metabolite(s) has not been established in this system; however, the effects on cell
proliferation are similar to those seen with MCTP, and preliminary studies revealed
that cytotoxic concentrations of MCT also cause DNA crosslinking in PECs (Wagner
et al., unpublished observations), suggesting that a bifunctional pyrrole may be
generated. The relevance of these Observations to MCT (P) pneumotoxicity in viva
has not been established.

MCTP treatment inhibits the proliferation of a variety of primary and

transformed cells in culture in addition to ECs, including Madin-Darby canine kidney

117

cells, Crandel-Rous feline kidney cells, undifferentiated human keratinocytes and
bovine and porcine pulmonary artery SMCs. Although these cell types vary in their
susceptibility to the cytolytic effects of MCTP, colony forming ability is inhibited to
a similar extent in all cell types after exposure to identical concentrations of this
compound (Reindel et al., 1988; Reindel and Roth, 1991) Reindel, Hoom and Roth,
unpublished observations).

Many of the PZAs exert antiproliferative effects on the liver in viva,
and this has been attributed to their ability to act as a bifunctional alkylating agents
which form DNA crosslinks (Bullet al., 1968; Mattocks and Legg, 1980; Culvenor et
al., 1969; Jago, 1969; Petry er al., 1986). DNA crosslinking occurs in the livers of rats
treated with MCT (Petry et al., 1984) and in cultured bovine kidney epithelial cells
treated with MCT in the presence of a metabolic activating system (Hincks et al.,
1991). In this cultured kidney cell system, the ability of various PZAs to cause DNA
crosslinking is associated with their ability to inhibit cell proliferation, but the exact
nature of this relationship remains conjectural.

It is not known if MCT(P) treatment in viva results in DNA
crosslinking in the pulmonary vasculature. However, recent studies in cultured ECs
have begun to address the relationship between the ability of MCTP to act as a
biflmctional alkylating agent and its cytotoxic effects. Cultured PECs treated with
MCTP in vitra manifest bath DNAzDNA and DNAzprotein crosslinking at
concentrations that inhibit cell proliferation (Wagner et al., 1991; Wagner et al.,
1992). Further investigation is required: 1.) to determine if there is a cause/effect

relationship between DNA crosslinking and the inhibition of EC proliferation, and

118

2.) to ascertain if and how these alterations in vascular ECs contribute to the

development of the delayed pneumotoxicity caused by MCT (P) in viva.

 

It is possible to fit together what we currently know about MCT (P)
pneumotoxicity and EC injury to form a tentative hypothesis to explain how early
endothelial changes lead to vessel leakage and pulmonary edema [Figure 11]. ECs
line the pulmonary vasculature in a monolayer. In the normal lung vasculature,
occasional EC death results in a small defect in the monolayer. This stimulates cell
migration and division which covers the defect quickly, before significant vascular
leak occurs.

When MCTP is given intravenously in viva, the endothelium of the
pulmonary vasculature is probably a major site of covalent binding of this
bifunctional electrophile. The effects of this binding could be lethal (e.g., cytolysis,
perhaps as a result of membrane disruption or altered calcium homeostasis) or
nonlethal but with profound biological consequences (e.g., DNA crosslinking). As
ECs die due to normal cell turnover or as a result of MCTP treatment, a small defect
forms in the monolayer. This stimulates cells to divide, but the crosslinked DNA is
unable to replicate correctly; thus, cells migrate and enlarge but are unable to divide.

In time, these damaged ECs may also die, resulting in a larger defect
in the monolayer. Surrounding cells are again stimulated to divide but are frustrated

in the attempt; these cells enlarge and die, too. Eventually, the defect in the

119

endothelial cell
monolayer

normal turnover
MCTP-induced?

 

y

cell death

1

small detect ln monolayer

stimulus for cell dlvlslon

””\ ... \a.......

X-llnltlng
stimulus for ——> frustrated attempt ‘1'
cell dlvislon to dlvlde repair or detect

cell enlargement
Change in gene products

l \

larger detect 6— 09“ 993‘" ' camera/unease or
In monolayer encore noses: can.

i 3120ch
~~—_ ~ “*“3 menu manuals

Figure 11: Role of MCT(P)-induced endothelial injury in the development of
pulmonary vascular alterations: An hypothesis.

120

Surviving ECs, if still functional, may express an altered profile of gene
products or demonstrate altered enzyme activity which could in turn influence
surrounding vascular cells. For example, if there were a change in the balance of EC
signals regulating SMC growth, vascular remodeling could ensue. Altered ACE
activity on the surface of the endothelium could result in abnormal levels of Ang I,
Ang II and BK in the pulmonary circulation, leading to altered vasoactivity. Figure
12 illustrates a number of EC changes which could contribute to the pulmonary
vascular changes seen with MCT (P) pneumotoxicity.

Cultured ECs have thus far been used to examine the time- and dose-
dependent effects of chemically synthesized MCTP on EC morphology and on limited
aspects of cellular function. These early studies demonstrated that there are direct
effects on ECs treated with MCTP and that expression of these effects appears to be
delayed and progressive. Further identification and clarification of functional
changes in ECs in response to treatment with MCTP may help to identify early,
subtle alterations that lead to the delayed and progressive changes characteristic of
MCT(P)-induced lung injury. In addition, manipulation of this model in vitra may

help to define mechanisms which underlie the critical EC alterations.

With. MCTP:

 

 
  
 

  
    
 

 

altered.
altered Surface Coagulation
Proteins -
Changes in
BBC adherence *Chaggsfiggtffi
. laltered Cene
Matrix Expression
Changes Enzy-e
Leakage anlt. Protein
altered Synthesis
Cell-Cell
49.
w. Contact up
.A . , {(4)
mauve»? ., 5:: ti
e======b
__-‘

 

accumulation of Edema Fluid
and Proteins

Figure 12: Potential-endothelial cell alterations after exposure to MCT P.

122

VIL mm

It is proposed that the endothelium is a primary target of MCI? and that the
functional consequences of MCTP-induced EC injury contribute to the development
of pulmonary vascular lesions in this model of pneumotoxicity. A major objective of
my dissertation research was to test the hypothesis that MCTP causes changes in
endothelium in vitra that are consistent with its ability to cause lung vascular injury
and pulmonary hypertension in viva. To do this, I utilized cultured ECs treated with
chemically synthesized MCTP. This model was developed in our laboratory to
facilitate examination of the direct effects of MCTP on vascular endothelium
(Reindel and Roth, 1991; Reindel et al., 1991). Cultured ECs have proven useful in
preliminary studies identifying changes in EC morphology and proliferative ability
after MCTP treatment. I have extended these studies to examine the effects of
MCTP treatment on: 1.) a surface enzyme function (ACE activity); 2.) basic
intracellular function (ability to synthesize DNA, RNA and protein); and 3.) a
specific aspect of EC function (regulation of mesenchymal cell proliferation).

Interspecific differences are evident in the response of cultured ECs to MCTP
treatment. These differences may reflect inherent differences which are expressed
when cells of various species are maintained in culture. However, if these differences
exist in viva as well, they may in part explain differences in target organ toxicity in
response to PZAs. A second Objective Of my work has been to continue to
investigate species differences in the response Of cultured ECs to MCTP treatment
in an effort to correlate these alterations with the responses seen in viva. To do this,

I have characterized the response of cultured rat pulmonary endothelium to MCTP

123

treatment and compared this response to that described in BECs and PECs treated
with MCI'P. This is an extremely relevant species comparison, as the rat is typically
used to study MCI‘ (P) pneumotoxicity in viva. In addition, I have extended the
species comparison of porcine and bovine endothelium to a functional parameter,
ACE activity. Studies with ECs of various species have contributed to a better
understanding of the endothelial response to injury as well, and this subject will be
discussed in my concluding remarks.

My third objective was to begin to investigate DNA crosslinking as a
mechanism in MCTP-induced EC injury. Damage to monolayers as a consequence
of treatment with certain other EC toxicants (e.g. endotoxiu, cyclosporin A), is
typically much more rapid in onset (Meyrick, 1986; Paulsen et al., 1989; Zoja et al.,
1986). Cell detachment and LDH release are evident within a few hours, and
obvious monolayer disruption is evident by 24 hours. These effects are generally not
progressive, and with removal of the toxicant the cells may recover and repair the
damage to the monolayer. The hypertrophic response which gradually develops after
a single administration of MCTP does not occur with these other toxic compounds.
Thus, the delayed and progressive nature of MCTP toxicity in ECs may represent a
unique response of this cell type to toxic injury, and may reflect its inability to effect
repair. In PECs, MCTP causes DNA crosslinking which is quite persistent (Wagner
et al., 1992). These findings led to the development of a second hypothesis:
crosslinking is responsible for the inhibition of proliferation seen in cultured ECs
treated with MCTP, and this cytostasis results in the characteristic delayed,

progressive and hypertrophic EC response to injury. In partial test of this hypothesis,

124

I examined the effects of another crosslinking agent, Mitomycin C (MMC), on

cultured porcine and rat endothelium.

Chapter II

MONOCROTALINE PYRROLE-INDUCED CHANGES IN ANGIOTENSIN-

CONVERTING ENZYME ACTIVITY OF CULTURED PULMONARY ARTERY

ENDOTHELIAL CELLS

125

126
Summary

Changes in structural and functional integrity of endothelium have been recognized
as relatively early features of the delayed and progressive pulmonary vascular injury
caused by the pyrrolizidine alkaloid, monocrotaline (MCT). Although a number of
investigators have evaluated angiotensin-converting enzyme (ACE) activity in the
lungs of rats treated with MCT, the exact nature of changes in activity of this enzyme
and the role they may play in MCT pneumotoxicity remain controversial.
Monocrotaline pyrrole (MCTP) is a putative toxic metabolite of MCT. Treatment
of cultured porcine and bovine pulmonary artery endothelial cells (PECs and BECs,
respectively) with chemically-synthesized MCTP results in toxicity which is also
delayed and progressive in nature. In this study, we examined the direct effects of
MCI? on cultured endothelial cell ACE activity. Post-confluent monolayers of PECs
or BECs were treated with a single administration of MCTP at time 0, then they
were examined for their ability to degrade the synthetic peptide, [3H]benzoyl-phe-ala-
pro. In PECs, which are relatively insensitive to the direct cytolytic effects of MCTP,
monolayer ACE activity was unchanged initially, but gradually decreased by 4 days
after treatment with a high dose of MCTP. This decrease was transient, and PEC
monolayer ACE activity returned to the control value by 10 days posttreatment.
BEC monolayer ACE activity was also unchanged initially, but rapidly declined 4
days after MCI'P treatment and remained depressed throughout the posttreatment
period. BECs are quite sensitive to the cytolytic effects of MCTP, and the decline

in ACE activity occurred coincident to the decrease in monolayer cellularity and

127

appearance of marked cytotoxicity previously reported in these cells. When ACE
activity was expressed as a function of cell number, cellular ACE activity was still
significantly decreased 4 days after a high concentration of MCTP in both PECs and
BECs. In summary, high concentrations of MCTP appear to decrease endothelial
ACE activity in a model which excludes complicating hemodynamic factors present
in viva. The decline in ACE activity is delayed, and the magnitude and duration of
the decrease corresponds to the degree of overt MCTP-induced cytotoxicity. This
suggests that altered endothelial ACE activity is unlikely to be a direct effect of
MCTP on the enzyme but may be a response to the delayed cell injury which results

from exposure to this compound.

TC

0]

lit

128

Changes in endothelial structural and functional integrity have been noted
relatively early in the course of MCI‘ pneumotoxicity in viva (Reindel er al., 1990;
Bruner et al., 1983; Hilliker et al., 1982), and it has been suggested that sublethal but
persistent changes in endothelial cell function may play a role in the development
and/ or maintenance of MCT-induced pulmonary hypertension (Reindel et al., 1990;
Reindel and Roth, 1991; Roth and Reindel, 1990). An important function of the
pulmonary vascular endothelium is cleavage of circulating, inactive angiotensin I
(Ang I) to the potent vasoconstrictor, angiotensin II (Ang II), by the action of the
exopeptidase known as angiotensin-converting enzyme (ACE). This membrane-
bound enzyme also serves to inactivate the vasodilator peptide, bradykinin (Yang et
al., 1970; Erdos, 1975). ACE has been localized to the luminal surface of endothelial
cells in structures called caveolae intracellulares (Ryan et al., 1975). Pulmonary
conversion of these peptides appears to be a function of pulmonary vascular surface
area and the time it takes for blood to transit the lung vasculature (Fanburg and
Glazier, 1973), thus it has been suggested that pulmonary vascular disease might be
associated with changes in pulmonary ACE activity (Gillis and Catravas, 1982;
Dobuler er al., 1982; Catravas er al., 1988b). It is also possible that agents which
result in pulmonary endothelial injury could directly alter the distribution or activity
of this surface enzyme.

A number of investigators have evaluated ACE activity in lungs of rats treated

with MCT in an effort to correlate changes in enzyme activity to some aspect of the

129

development or progression of MCT-induced cardiopulmonary changes. The results
of these studies have been disparate and controversial. Some investigators report
that there is a decrease in lung ACE activity which corresponds to an increase in
pulmonary pressure (Kay et al., 1982a; Keane et al., 1982), whereas others contend
that apparent changes in lung ACE activity are due to a dilution of this activity by
the increased lung mass seen in this model (Huxtable et al., 1978; LaFranconi and
Huxtable, 1983). Yet another group reports an increase in lung ACE activity shortly
after MCT treatment, with a subsequent sustained decline in activity later on
(Molteni et al., 1984). A fourth group finds a sustained decline in ACE activity which
precedes the onset of pulmonary hypertension and which occurs at doses that do not
produce right ventricular hypertrophy (Shale et al., 1986). From these reports, it has
been difficult to draw definitive conclusions about ACE activity in lung vasculature
injured by MCT.

MCTP, a putative toxic metabolite of MCT, is toxic to cultured bovine and
porcine pulmonary artery endothelial cells (BECs and PECs, respectively) (Reindel
and Roth, 1991; Reindel et al., 1991). This cultured endothelial cell model makes
it possible to investigate the direct effects of MCTP on endothelial ACE activity.
The purpose of this study was to determine whether MCTP affects the activity of this
ectoenzyme in cultured endothelium and to characterize the development of any

changes that occur.

130

W: Lines of PECs and BECs were derived from
segments of pulmonary artery by modifications of the methods of Jaffe et al. (1987)

and Booyse er al. (1975), as described by Reindel et al. (1991). Sections of
pulmonary artery were collected aseptically from freshly killed young market animals,
rinsed in sterile Hank’s balanced salt solution (HBSS; Sigma Chemical Company, St.
Louis, MO), and placed into sterile Puck’s saline containing 3%
antibiotic/antimycotic solution (300 units / ml penicillin, 300 ug/ ml streptomycin, 0.75
rig/ml Fungizone; Ab/Am; GIBCO, Grand Island, NY) on ice for transport. In a
laminar flow hood, the adventitia, external elastic laminae and outer medial smooth
muscle layers were removed by careful dissection. Trimmed squares of artery were
placed into 60 mm tissue culture plates with luminal surfaces submerged in a thin
film of collagenase (Type IA, 0.1%; Sigma). After incubation for 3 to 5 minutes at
37°C, the luminal surface of the vessel was gently brushed with a single stroke of a
rubber policeman, and the brushings were resuspended in fresh, calcium- and
magnesium-free HBSS (CMF-HBSS; Sigma). Under an inverted microscope (TMS;
Nikon, Tokyo, Japan), small clusters of endothelial cells were collected with a
micropipette and transferred to wells of 12-well tissue culture clusters (Corning, Park
Ridge, IL) containing 0.5 ml of Opti-MEM (GIBCO) with 5% fetal calf serum (FCS;
Biocell, Rancho Dominguez, CA) and 1% Ab/Arn (PECs) or Medium 199 (M199;
GIBCO) containing 10% FCS and 1% Ab/Am (BECs). Plates were incubated for

5 to 7 days under standard incubation conditions, 37°C with 7.5% CO,/92.5% air.

131

At this time, wells with colonies exhibiting typical cobblestone morphology and that
were free of spindle cells were passed into larger wells using 0.25% trypsin/0.01M
EDTA (GIBCO). These cells were grown to confluence, and wells were selected for

passage on the basis of characteristic endothelial cell morphology.

At the time of their third passage, cells were plated onto glass coverslips for further
characterization, including uptake of Di-l—Ac-LDL (acetylated low density lipoprotein
labeled with l,1’-dioctadecyl-1-3,3,3’,3’-tetramethyl-indo-carbocyanine perchlorate;
Biomedical Technologies, Inc., Stoughton, MA), angiotensin-converting enzyme
activity (Ventrex, Portland, ME), positive staining for factor VIII-related antigen and
electron microscopic examination, and were tested for contamination with
mycoplasma (Hoechst Stain Kit; Flow Laboratories, McLean, VA). Cultured
endothelial cell lines were maintained in Opti-MEM with 3% FCS and 1% Ab/Am
(PECs) or M199 with 10% FCS and 1% Ab/Am (BECs) and were split at a ratio of
1:3 or 1:4 at weekly intervals by either enzymic (trypsin/EDTA) or mechanical

dissociation. All studies were performed with cells between passages 4 and 9.

W: MCTP was prepared from MCT (Transworld Chemical,
Washington, DC) via an N-oxide intermediate by the method of Mattocks (1968).
MCTP isolated from this synthesis procedure has Ehrlich activity (Mattocks and
White, 1971) and a structure compatible with MCI‘P as determined by mass
spectrometry and nuclear magnetic resonance (Bruner et al., 1986). MCTP was

dissolved in N,N-dimethylformamide (DMF; Sigma Chemical Co., St. Louis, M0) at

132
a concentration of 20 mg/ml, and all dilutions of MCTP were made with DMF. A

25 or volume of MCI? solutions or DMF vehicle (0 uM MCI‘P) per milliliter of
medium was used to achieve the nominal concentrations of MCTP (0, 1.5, 15 or 150

uM) used in the study.

WWW: PECs and BECs were plated into 12-

well tissue culture clusters (25 mm diameter) and allowed to form confluent
monolayers. Monolayers were not used for assay of ACE activity until at least 10
days after plating as ACE activity of cultured endothelial cells is not restored after
dissociation and replating until cells have been maintained in a confluent state for
this length of time (DelVecchio and Smith, 1981). Mature monolayers were treated
with a single administration of MCTP (0, 1.5, 15 or 150 llM) on day 0. ACE activity
of monolayers was analyzed at 8 hours and 2, 4 and 7 days (also at 10 days, for PEC
monolayers) posttreatment using the standard radioassay protocol provided by
Ventrex Laboratories, Inc. (Portland, ME). The assay involves cleavage of the
synthetic peptide [3H]benzoyl-Phe-Ala-Pro to form [3H]benzoyl-phenylalanine and the
dipeptide alanyl-proline by endothelial ACE. The tritiated portion of the peptide
becomes soluble in organic scintillant when cleaved, and can be extracted, separated
and counted in a liquid scintillation counter. Activity in the presence of the specific
ACE inhibitor, Captopril (SQ 14,255; IO‘M), was subtracted from total activity to
ensure that conversion of substrate was due to the action of ACE. Calculation of

ACE activity was carried out as described in the Ventrex protocol, using a formula

133

representing a simplified form of the integrated first order Michaelis-Menton

 

 

equation:
4000(T-I)
fp x 80
E = -..
t
where, E = enzyme activity as % substrate utilized / minute / ml
t = time in minutes of reaction
T = c.p.m. of TEST monolayer
I = c.p.m. of INHIBIT OR-treated monolayer
fp = fractional partition of product into counting phase = 0.5
So = c.p.m. of substrate (total c.p.m.)

A unit of ACE activity is defined as that required to hydrolyze substrate at an initial
rate of one percent per minute at 37°C.

WW: After determination of ACE activity as
described above, monolayers were washed three times with CMF-HBSS (Sigma) to

remove non-adherent cells. Adherent cells were enzymically removed from the plate
surface with 0.025% trypsin-0.27 mM EDTA solution (GIBCO) and counted using

a Coulter counter (Model ZM; Coulter Electronics, Luton, England).

Wm: ACE activity data are presented as means of 5 (PECs) or 3
(BECs) separate experiments, each consisting of two wells per concentration of
MCTP. The standard error of difference (Steel and Torrie, 1980) for each data set

is presented. Data were analyzed with a blocked analysis of variance (ANOVA).

134

Individual comparisons between treatment groups (MCTP and DMF vehicle) were

made using Tukey’s omega as the past hac test (p<0.05).

Cell numbers are presented as means + /- SEM. These data were analyzed using a
completely random AN OVA, and individual comparisons were made using the

Tukey’s omega test (p < 0.05).

The effects of MCTP an cellularity of PEC and BBC monolayers are shown in
Figures 13A and 13B, respectively. Cell monolayers were treated once with MCTP
at zero time. Whereas PEC and BBC cell numbers continued to increase slightly
with time in wells treated with vehicle or the lowest concentration of MCTP, they
declined in monolayers treated with the higher doses of MCTP, beginning at day 4
posttreatment. When examined using phase contrast microscopy, all monolayers
appeared to remain intact and free of obvious gaps throughout the posttreatment
period. PECs and BECs treated with higher concentrations of MCI? (15 or 150
uM) demonstrated distinct morphological changes as monolayer cellularity decreased.
These changes have been reported previously (Reindel and Roth, 1991; Reindel et
al., 1991) and include cell enlargement, vacuolization and nuclear changes in BECs

and a milder enlargement with a shift toward a more spindle-er shape in PECs.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2000 L
A
8
o 1500
3:
v
3 1000
3 0 Vehicle
\ o 1.5 yum MCTP
l) V 15 pM MCTP
3 500 - v 150 pm MCTP
U
0 1 1 l 1 .
0 2 4 6 8 10
Time (Days)
2000.. B. BBC Number
8 0 Vehicle
8 150° ” o 1.5 nM MCTP
.. v 15 yum MCTP
\x/ V 150 pM MCTP
g 1000 -
3
\
.‘i’
7.3 500 -
o
O J l l l l l i
0 l 2 3 4 5 5 7

Time (Days)
Figure 13: Effects of MCTP on cellularity of PEC (A) and BBC (B) monolayers.
Post-confluent monolayers were exposed to a single administration of MCTP on day

0. Values represent the means of replicate studies (n=5 for PECs; n=3 for BECs).
" Significantly different from vehicle control (p < 0.05)

136
Changes inACE activity of mature monolayers of PECs and BECs after a single

exposure to MCTP are shown in Figure 14. ACE activity of PEC monolayers (Figure
14A) treated with 150 llM MCTP decreased gradually, reaching a nadir at 4 days
posttreatment then returning to control levels by 10 days posttreatment. BEC
monolayers also showed a concentration-dependent decline in ACE activity with
treatment, but the decrease was much more marked and persistent (Figure 14B). In
monolayers treated with the highest concentration of MCTP, ACE activity was
significantly decreased by 2 days posttreatment, and after 4 days posttreatment these
monolayers demonstrated virtually no ACE activity. In BECs, the intermediate

MCTP concentration also led to reduced ACE activity after day 2.

Cellular ACE activity after MCTP treatment is expressed in Figure 15 as a
function of PEC and BBC number (units of activity per million cells). While the
total activity of vehicle-treated PEC and BBC monolayers were similar (see Figure
14), activities normalized to cell number were quite different. BECs (Figure 15B)
had about twice the ACE activity per million cells than PECs (Figure 15A). The
delayed, transient decrease in ACE activity of PECs treated with 150 llM MCTP was
still evident, but cellular ACE activity returned to the control value by 7 days and
was increased by 10 days posttreatment. Cells treated with 15 llM MCTP showed
no change in activity until day 10, at which time cellular ACE activity was elevated.
In BECs, cellular ACE activity was significantly depressed by 4 days posttreatment
in cells treated with the highest concentration of MCTP. However, decreases in

cellular ACE activity were not evident in BECs treated with 15 11M MCTP.

137
A. PEC ACE Activity

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

100 T
90 -
2"?
9 O
__ ........... o ......................
g ._ ,9 """"""""" c m;
v /
v/
E): \\ I./"
2 .0 \'
S 0 Vehicle
<2 30 - 0 1.5 pM MCTP
8 .20 _ BED- 3-0 v 15 pM MCTP
< 10- v 150 pM MCTP
0 L ‘ ' ‘ '
o 2 4 6 8 10
Time (Days)
100r B. BEC ACE Activity
90 -
£7? 80 -
E
z ..
3 . 7O
60 -
E 50 -
>
5: 4O -
3 30 . 0 Vehicle \ _
8 20 _ o 1.5 nM MCTP [5'30" 7'5]
< v 15 MM MCTP .
10 '- v 150 MM MCTP ' v
0 J 1 m l 1 1 I
o 1 2 3 4 5 6 7

Time (Days)
Figure 14: Angiotensin-converting enzyme activity of MCTP-treated monolayers.

Monolayers of PECs (A) and BECs (B) were exposed to a single administration of
MCTP at time 0. One unit of ACE activity is defined as that required to hydrolyze
substrate at an initial rate of one percent per minute at 37°C. Values represent the
means of replicate studies (n=5 for PECs; n=3 for BECs). Standard error of
difference = 3.89 PECs; 7.55 BECs). ‘ Significantly different from vehicle control

(p < 0.05 )

601-

20-

.10 -

ACTIVITY (U/10E6 CELLS)

 

120,
110a
100-
90-
.80~
70-
50-
50-
40-
30-
20»
10-

ACTIVITY (U/10E6 CELLS)

 

138

A. PEC ACE Activity

 

 

 

0 Vehicle 7

 

o 1.5 ,u.M MCTP
v 15 [.LM MCTP
V 150 pM MCTP

 

 

4. 6 8 10

Time (Days)

B. BEC ACE Activity

 

 

 

 

O

V
V

 

Vehicle

1.5 ,uM MCTP
15 pM MCTP "

4

150 nM MCTP '

 

 

4L

l

To
j.-
l—

l I

 

1

2

3 4 5 6 7
Time (Days)

Figure 15: Angiotensin-converting enzyme activity
expressed as a function of cell number.

Mature monolayers of PECs (A) and BECs (B) were exposed to a single
administration of MCTP at time 0. Values represent the means of replicate studies
(n=5 PECs; n=3 BECs). Standard error of difference = 2.44 PECs; 9.58 BECs.

“ Significantly different from vehicle controls (p<0.05)

139

A number of conditions which result in pulmonary injury or alterations in
pulmonary blood flow have been associated with changes in angiotensin-converting
enzyme activity in lung or serum. In some cases, these changes may reflect direct
damage to the vascular endothelium. For example, administration to rats of alpha-
naphthylthiourea (ANTU) (Hollinger et al., 1980a) or paraquat (Hollinger et al.,
1980b) at doses which cause acute lung injury results in a rapid, transient decrease
in lung ACE and a corresponding increase in serum ACE. These changes are
thought to reflect a loss of endothelial integrity, which is supported by the rapid onset
of pulmonary edema.

In other cases, changes in ACE activity appear to be compensatory and occur
coincident with or subsequent to the development of puhnonary hypertension. This
appears to be the case in animals exposed to conditions of chronic hypobaric hypoxia
(J ederlinic et al., 1988; Keane et al., 1982) or in the lungs of hypoxia-adapted rats
(Oparil et al., 1988). Fetal rats and rabbits, in which pulmonary arterial pressure is
normally greater than in adults, also have low puhnonary ACE activity which
increases soon after birth (Wallace et al., 1979; Stalcup et al., 1978). Under these
conditions of increased pulmonary arterial pressure, decreased ACE activity might
result in decreased conversion of Ang I to Ang II, a potent vasoconstrictor, as well
as less degradation of the vasodilator, bradykinin.

A similar scenario has been described for MCT pneumotoxicity in the rat.

Kay er al. (1982a) report that a single, subcutaneous injection of 60 mg MCT/kg

140

causes delayed and progressive pulmonary vascular injury which results in vascular
remodeling and consequent pulmonary hypertension by 10-14 days posttreatment.
These changes are accompanied by a decrease in lung ACE activity per mg protein
beginning by day 10, which is supported by decreased conversion of Ang I to Ang 11
(Keane and Kay, 1984). Molteni et al. (1984) report a transient increase in
pulmonary ACE activity after 1 week of MCI‘ followed by a persistent decline in
activity during weeks 2 through 6. In these studies, rats received MCT in their
drinking water (20 mg/ liter) throughout the experimental period. No changes in
serum ACE were reported under either set of experimental conditions. Kay and
Keane maintain that the alterations in ACE activity seen with MCT are a result of
the pulmonary hypertensive conditions, and may be a protective mechanism to limit
this physiological change (Kay et al., 1982a; Keane et al., 1982). This view is held by
Molteni et al. (1984) as well, but they also suggest that the transient increase in ACE
which they find in their model precedes vascular remodeling and reflects endothelial
injury and dysfunction.

In a separate study, Shale et al. (1986) found that lung ACE (reported as
nmol/min/lung and nmol/min/mg protein) decreased when MCT was given in an
oral dosing regimen which did not produce pulmonary hypertension or right
ventricular hypertrOphy. Although this supports the idea that ACE does not play a
causative role in the development of these sequelae, it also suggests that the decrease
in ACE activity seen with MCT treatment is not solely a response to increased
pulmonary vascular pressure. Rather, it suggests an early, direct effect of MCT on

the lung. The increased pulmonary tissue mass due to inflammation and hyperplasia

14 1
after MCT treatment may in fact dilute what is really an unchanged endothelial ACE

content at this stage in the injury.

The concept that a decrease in lung ACE activity may be due to increased
total lung protein is one that has been maintained by Huxtable and his co-workers
for some time (Huxtable et al., 1978; LaFranconi and Huxtable, 1983). Although
they found that MCT administered in the drinking water for 3 weeks did decrease
lung ACE per gram of dried lung, total lung ACE activity was unchanged. Any
decrease in lung ACE activity was attributed to dilution by an increased total lung
mass. However, Kay and Keane maintain that ACE decreases in their model
whether it is calculated on a per tissue weight or total lung basis, and they suggest
that the difference in response may be due to differences in experimental design
(Keane and Kay, 1984).

My objective in performing this study was to determine if MCI'P would
directly alter ACE activity in an isolated endothelial cell system. Although we found
that MCTP does decrease endothelial ACE activity in vitro, our results raise a
number of interesting points with respect to changes in ACE activity after cell injury.
In both PECs and BECs, decreases in ACE activity did not occur immediately after
administration of MCTP; rather, changes were delayed for several days. This
suggests that the decrease in ACE activity was not due to a direct interaction of
MCTP with the enzyme or to other changes which may occur at the cell surface at
the time of treatment. MCTP is quite reactive and much of it probably binds to cells

rapidly after administration in vitro. However, the inactivation of ACE does not

142

appear to be an early or an immediate change effected by MCTP, even at relatively
high concentrations.

A more rapid decrease in ACE activity has been reported in cultured
endothelial cells when they are subjected to hypoxia (Stalcup et al., 1979). This
decrease is rapidly reversible with restoration of normoxia, and it occurs without
marked structural changes in the cells. The authors of that report suggest that the
decrease in enzyme activity results from a response of the endothelial cells to the
hypoxia rather than as a direct effect of hypoxia on the enzyme. Our results suggest
that this may be the case with MCTP treatment of endothelial cells as well. Ryan
and Catravas (1990) suggest that, in addition to enzyme dysfunction, changes in ACE
activity may occur as a result of changes in the microenvironment of the enzyme.
Some of the delayed cytotoxic changes which occur as a consequence of exposure to
hypoxia or MCTP treatment could result in subtle alterations of the endothelial cell
surface which are incompatible with ACE function.

There are clear species differences in the response of pulmonary artery
endothelial cells to treatment with MCTP. Species differences have been reported
with respect to cell detachment, release of lactate dehydrogenase and changes in cell
morphology (Reindel et al., 1991). While BECs are quite sensitive to the cytolytic
effects of MCTP, PECs appear to be relatively resistent. However, both cell types
are inhibited in their abilities to proliferate at the concentrations used in this study,
and both show distinct changes in morphology after MCTP treatment. The
differences noted with respect to ACE activity of BBC and PEC monolayers after

MCTP treatment are also pronounced, and these seem to parallel the morphologic

143

and cytotoxic changes. For example, the transient nature of the decrease in PECs
corresponds with the decline in cell numbers, and the more dramatic and persistent
fall in ACE activity in BECs reflects the more marked cytotoxic response of this cell
type.

ACE inhibitors, such as captopril, cilazapril or CL242817, and angiotensin
receptor antagonists prevent or attenuate the vascular remodeling seen after
denuding or toxic injury to the endothelium (Grotendorst et al., 1982; Schweigerer
er al., 1987; Molteni et al., 1985; Powell et al., 1990; Clozel et al., 1991; Zakheim et
al., 1975), suggesting that Ang II may be important in the pathophysiology of the
vascular response to injury. It is possible that endothelial modulation of production
of this potent mediator may occur at a local level under conditions which might
predispose the vessel to myointimal hyperplasia. Bell and Madri (1990) have shown
that ACE inhibition or angiotensin receptor blockade in cultured bovine aortic
endothelial cells ((BAECs) and smooth muscle cells (SMCs) results in increased
BAEC migration and plasminogen activator levels, while limiting SMC migration.
They suggest that blockade of this autocrine pathway might reduce vascular wall
injury by enhancing wound closure, increasing the antithrombotic tendency of the
area and decreasing the rate of SMC infiltration. ACE activity declined substantially
in our cells after treatment with MCTP only after there was a decrease in monolayer
cellularity and evidence of cytotoxicity. This suggests that the change in enzyme
activity may have occurred in response to these alterations and may function to limit
the local damage. In PECs, in which the cellular damage was less pronounced, the

changes in ACE activity were also quite subtle and transient. These cells eventually

144

had increased ACE activity on a cellular basis by 10 days posttreatment, which may
reflect more persistent alterations in the cells’ response to injury.

It is important to point out that the most dramatic changes in ACE activity
in both PECs and BECs occurred only at the highest concentrations of MCTP. The
pulmonary vascular endothelium may not experience such a high exposure after a
pulmonary hypertension-producing dose of MCTP in viva (Reindel et al., 1990),
except, perhaps, in certain regions. The pattern of MCI‘- or MCTP-induced lung
injury in viva is multifocal in nature, suggesting nonuniform distribution of toxicant
to the various regions of the lung vasculature. Although endothelial damage does
result after treatment with MCT or MCTP in viva, the damage is unlikely to be
evidenced by a noticeable decrease in cellular ACE activity unless rat pulmonary
endothelium is much more sensitive to these compounds than the species we have
tested. Considering this, it seems likely that changes in ACE activity seen after MCI‘
treatment in viva occur either as a result of altered lung mass, pulmonary vascular
surface area, or blood flow, or in response to changes in pulmonary vascular
pressure, but probably do not reflect altered endothelial ACE activity as a direct
result of treatment.

In summary, ACE activity in pulmonary artery endothelium is altered after
MCTP treatment in vitro, and this alteration is clearly not secondary to hypoxia,
increased pressure, or other complicating hemodynamic factors that may be present
in viva. These changes appparently are not due to a direct action of MCTP on the

enzyme, but they more likely occur as a response to endothelial cell injury.

Chapter HI

MONOCROTALINE PYRROLE ALTERS DNA, RNA AND PROTEIN SYNTHESIS

IN PULMONARY ARTERY ENDOTHELIAL CELLS

145

146
Summer

Olltured porcine pulmonary artery endothelial cells (PECs) treated with
monocrotaline pyrrole (MCTP) remain viable but are unable to divide and exhibit
an altered morphology. Such responses raise a question about the extent to which
affected cells carry out normal functions such as RNA and protein synthesis.
Accordingly, the cellular activity of MCTP-treated PECs was examined in this study.
PECs were treated with a single administration of MCTP or vehicle, and
determinations of cell number, protein and DNA content were made at times up to
7 days posttreatment. DNA, RNA and protein synthesis were quantified by
incorporation of [3H]—labeled thymidine, uridine and leucine, respectively. Increases
in cell number that occurred with time in the control cells were reduced in MCTP-
treated cells. At 7 days post-treatment, both protein and DNA' content increased
above control levels. Synthesis of DNA, RNA and protein continued in all treatment
groups throughout the post-treatment period, but cells treated with high
concentrations of MCI'P showed less synthetic activity than controls during the initial
48 hours posttreatment. By 7 days, MCTP-treated cells were producing significantly
more DNA, RNA and protein. These results indicate that cells treated with MCTP
continue to synthesize DNA, resulting in an increased DNA content. In addition,
treated cells continue to synthesize RNA and translate RNA into protein. Thus,

cellular activity is maintained but altered substantially by MCTP exposure.

147

An interesting feature of MCTP pneumotoxicity is its delayed and progressive
nature. There is a delay of several days after a single administration of MCT or
MCTP before major lung injury is evident. This suggests that initially lung cells may
be sublethally but irreversibly damaged by MCTP. Although the reactivity of MCTP
suggests that the initial event in the toxicity occurs rapidly (ie, covalent binding of
MCTP to the cell), the lag period implies cellular alterations which are slow to
develop but which are progressive and trigger a cascade of molecular changes and
pathophysiological events that result ultimately in chronic pulmonary vascular disease.

By virtue of its anatomic location, the pulmonary vascular endothelium is a
likely target for binding of MCTP that is delivered to the lung via the circulation, and
there are indications that endothelial cells are altered with MCTP treatment in viva.
In vitro, administration of a sublethal concentration of MCTP causes injury to
cultured endothelial cells. A single treatment of endothelial cells from porcine
pulmonary arteries with MCTP at noncytotoxic concentrations inhibits cell
proliferation (Reindel and Roth, 1991; Reindel et al., 1991). PECs treated with
MCTP become spindle-shaped and show a delayed and progressive increase in
release of prostacyclin (Reindel et al., 1991). If PEC monolayer integrity is
compromised by mechanical wounding after MCTP treatment, the remaining cells
migrate into the defect, enlarge dramatically, and remain viable but do not divide as
normal cells would do (Hoorn et al., 1990). Although PECs treated with MCTP are

unable to proliferate normally, they do remain viable in culture for long periods of

148
time. MCTP-treated cells may be functionally different from normal ECs, and these

changes in function might contribute to the delayed and progressive development of
MCTP pneumotoxicity in viva. Many types of cellular stresses, such as heat shock
and hypoxia, alter the amounts and types of RNA and protein which cells produce
(Williams et al., 1989; Thomas et al., 1982; King et al., 1989; Kourembanas et al.,
1990; Ogawa er al., 1990b). This study was designed to investigate the effects of
MCTP treatment on synthesis of RNA and protein by PECs to determine if MCTP
treatment influences basic cellular functions.

The arrest of cell proliferation observed after MCTP treatment and the ability
of this compound to act as a bifunctional alkylating agent (Robertson, 1982; Petry et
al., 1984) suggest that changes may occur in treated ECs at the level of DNA. This
suggestion is supported by observations that DNA crosslinking occurs in the livers of
rats treated with MCT (Petry et al., 1984) and in cultured PECs treated with MCTP
(Wagner et al., 1991; Wagner et al., 1992). Although MCTP-treated cells do not
divide, it is possible that they continue to undergo some DNA synthesis either as part
of the DNA repair process or in a frustrated attempt to proliferate. Accordingly, a
second objective of this study was to determine the effect of MCTP treatment on

DNA synthesis and DNA content of cultured PECs.

(2W Primary isolation and subculture of PECs was performed as described

previously in chapter II. These studies were performed with subconfluent as well as

149

confluent cultures of endothelial cells (as indicated) to determine if a stimulus for
cell proliferation (ie., subconfluency) further alters cell function after MCI‘P
treatment. PECs were plated at low (40,000 cells /well) or high (80,000 cells/well)
density into 12-well tissue culture clusters containing 1.5 ml Opti-MEM with 5% F CS
and 1% Ab/Am per well. Plates were incubated under standard conditions of 37°C
and 7.5% CO,/ 92.5% air until the high-density cultures reached confluence (typically,
48-72 hours). At this time, the medium in each well was replaced with Medium 199
(GIBCO) containing 0.5% FCS for 48-72 hours to growth-arrest the PECs. At the
end of this period, the low-serum medium was removed and replaced with 2 ml fresh

Opti-MBM (5% ms, 1% Ab/Am).

W: MCTP was synthesized from MCI‘ as described previously
in Chapter H. MCI‘P was dissolved DMF and diluted to the indicated
concentrations. Equal volumes of DMF served as vehicle controls in all experiments.
At time 0, growth-arrested PECs were placed in fresh Opti-MEM and received a
single administration of MCTP at a volume of 5 ul/Z ml medium to achieve the
nominal concentrations of 1.5 or 150 uM MCTP. These concentrations, selected on
the basis of previous work with MCTP in cultured PECs, typically result in minimal
(low concentration) or marked (high concentration) toxic responses in cultured PECs,
respectively (Reindel et al., 1991). Treated cells were incubated under standard
conditions for the indicated time periods with one change of medium on day 4

posttreatment. No additional MCTP was added with the fresh culture medium.

150
W. At the designated times of 0, 3, 6, 12, 24, 36, 48 and 168

hours post-treatment, three wells of PECs per treatment group were analyzed for
each of the following: cell number, protein content, DNA content, and radiolabeled
thymidine, uridine and leucine incorporation.

Number of cells per well was determined as described previously in Chapter II.

Protein content of the monolayer was determined by the method of Bradford
(1976). After washing with HBSS, monolayers were dissolved in 1 ml of 0.1 M ~
NaOH. Two 100 pl aliquots of the cell lysate per well were analyzed, and the results
reported as mean protein content per cell.

Total DNA content per well was determined by the spectrofluorometric method
of LaBarca and Paigen (1980). Monolayers were washed and cells removed with
trypsin/EDTA. Cells were transferred to a 3-ml polypropylene tube, spun in a
centrifuge at 1400 x g, and resuspended in 0.65 ml DNA assay buffer [2 M NaCl, 0.05
M Nazi-1P04 and 2 mM EDTA-ZNa in water, pH 7.4]. The suspension was sonicated
and refrigerated until analysis. Each cell sonicate was analyzed in duplicate, and the
results reported as mean DNA content per cell.

Thymidine, uridine and leucine incorporation were determined by a
modification of the method of Madhukar et al. (1989). At the post-treatment times
indicated, cells were pulse-labeled with 2 uCi/ ml [5-methyl-3H] thymidine (40-60
Ci/mmol; Amersham, Arlington Heights, IL), 0.2 uCi/ ml [5,6-3H] uridine (35-50
Ci/mmol; NEN/Dupont, Hoffman Estates, IL), or 2 uCi/ml L-[4,5-3H(N)] leucine
(40-60 Ci / mmol; NEN/Dupont) under standard incubation conditions for 2 hours.

The medium used for all labeling procedures contained 105M cold thymidine, as well

151

as 120 mg cold leucine and 0.3 mg cold uracil / liter, ensuring that the radiolabeled
compounds would serve as tracers only and would not significantly impact on the
precursor pools. After incubation, the culture plates were placed on ice for 30
minutes, then they were washed three times with cold, phosphate-buffered saline
(PBS) to remove all unincorporated tracer. The cells were treated with cold 10%
trichloroacetic acid (TCA) on ice for 10 minutes, washed first with 70% ethanol, then
with 95% ethanol and air-dried. 0.5 ml 1 N NaOH was added to solubilize the
precipitate overnight, then 0.5 ml 1 N HCl was added to neutralize the digest; 200
pl aliquots were mixed with 15 ml Safety-Solve” liquid scintillant (Research Products
International Corporation, Mount Prospect, IL), and radioactivity was determined in
a scintillation counter. Disintegrations per minute (d.p.m.) in each well were divided
by cell number to determine average incorporation per cell.

[3H]Thymidine incorporation was also evaluated with autoradiography by a
modification of the method of Wang et al. (1989). Cells were grown at high or low
density in Lab Tek’ two-chamber glass slides (Nunc, Inc., N aperville, IL), then
treated with MCTP as described above. At 24, 48 or 168 hours post-treatment, [3H]
thymidine was added to each well at a final activity of 2 uCi/ml for a 2 hour
incubation, after which the slides were washed with cold PBS (4 x 15 minutes) and
fixed with cold 5% TCA for 20 minutes. Slides were dipped briefly in absolute
ethanol and air-dried. They were dipped in NTB-2 emulsion (Kodak, Rochester,
NY) and air-dried, then sealed in a light-tight can with desiccant at 4°C for 48 hours.
Autoradiographs were developed in Kodak D-19 developer, then they were fixed and

stained with Giemsa stain. Four chambers were evaluated per treatment group.

152

Results are reported as the fraction of labeled cells per total cells counted and the

number of silver grains per labeled cell.

mm. Data are presented as means of 5 (high density) or 3 (low

density) separate experiments, each of which consists of 3 wells per experimental
parameter. The standard error of difference (Steel and Torrie, 1980) for each data
set is presented. Data were analyzed by a blocked AN OVA. Individual comparisons
between treatments (MCTP and DMF vehicle) were made using Tukey’s omega as

the post hoc test (p< 0.05).

The effects of MCTP on PEC number are presented in Figure 16. Although
PECs plated at high density had formed a confluent monolayer by the time of
treatment, cells treated with DMF vehicle or the low concentration of MCTP
continued to divide and increase in number throughout the experimental period.
This has been a consistent finding with PECs. The cells continue to divide in the
confluent monolayer after release from arrest, so that the cell density increases with
time. PECs treated with the higher concentration of MCTP were inhibited in their
ability to proliferate and did not increase in number with time (Figure 16A). Similar
results were obtained when PECs were treated at subconfluent densities (Figure

16B); however, cell numbers in control wells doubled in the high-density cultures but

153

600000 ? A. CELL NUMBER: HIGH DENSITY

 

 

 

 

 

 

 

 

 

 

 

 

 

 

V
500000 '-
__ 400000
'5'
p
\ 300000
2
3
200000 -
O VEHICLE
v 1.5 p.“ MCTP
100000 __ V ‘50 fl “CT?
0 . : ///’ o
0 24 48 168
Time (Hours)
600000
F 8. CELL NUMBER: Low DENSITY
500000 '-
0 VEHICLE
= ‘00000 ’ v 1.5 nu MCTP
; v 150 nu MCTP -—-—e
\ 300000 - ---v
3 .
3
200000
100000
0 . : 1 ///I ‘
0 24 48 168

Time (Hours)

Figure 16: Effects of MCTP on endothelial cell number.

Monolayers of PECs at high (A) or low (B) density received a single administration
of MCTP at time 0. Values represent means of replicate studies. (n=5, High
Density; n=3 Low Density) [Standard error of difference = 15,900 (HD); 11,190
(LD)] ‘ Significantly different from vehicle control (p<0.05)

154

increased between three- and four-fold in the low density cultures. Because of these
changes in cell numbers, remaining data are presented on a “per cell" basis.

Changes in cellular protein reflect changes in cell size. In the high density
cultures (Figure 17A), protein content per cell increased very modestly between 12
and 24 hours in cells treated with DMF or with 1.5 uM MCTP and then returned to
a basal value of 0.3 ng/ cell after 48 hours, presumably once cell division had taken
place. Cells treated with the higher concentration of MCTP increased in protein
content very slightly over time to 0.4 ng / cell at 7 days post-treatment. The response
was similar but exaggerated in the cells plated at low density (Figure 17B): controls
increased in number and reached confluence by one week post-treatment, whereas
cells treated with the high concentration of MCTP failed to increase in number but
instead enlarged and spread out to cover the surface of the culture dish and had
twice the protein content of vehicle-treated cells.

DNA content of PECs treated with MCTP is shown is Figure 18. DNA content
of cells treated at high density with vehicle or with the low concentration of MCTP
remained nearly constant throughout the treatment period (Figure 18A). Cells
treated with 150 uM MCTP had an increased DNA content by 7 days post-treatment.
PECs treated at low density initially had a greater DNA content compared to cells
treated at high density. In cells treated with vehicle or with the low concentration
of MCTP, DNA content decreased initially then slowly increased over 24 hours,
before ultimately declining to a basal level. PECs treated at low density with the

high concentration of MCTP increased dramatically in DNA content late in the post-

 

155

 

 

 

 

 

 

 

 

 

 

 

 

 

0-7 i A. PROTEIN CONTENT: HIGH DENSITY
0.6 -
e VEHICLE
V 1.5 [2M MCTP
3 0.5 - v 150 an MCTP
’6
0
\
U
5
i
.1
O
‘-
c- 0.2 ~
0.1 -
0.0 . . . , L .
O 24 4a 168
Time (Hours)
07 - B. PROTEIN CONTENT: Low DENSITY
0°6V """" v
72:,
E? 0'5 "w
E v
u 0.4 " . .
= ---v
.5 0.3 ~ "“
E e VEHICLE
g. 02 ,_ V 1.5 [2M MCTP
v 150 n MCTP
0.1 -
0.0 . . . ..
0 24 43 168

Time (Hours)

Figure 17: Effects of MCTP on cellular protein content.

PECs at high (A) and low (B) density were exposed to a single administration of
MCTP at time 0. Values represent means of replicate studies. (n=5, High Density;
n=3, Low Density) [Standard error of difference = 0.028 (HD); 0.053 (LD)]

‘ Significantly different from vehicle control (p<0.05)

P \mu.

1
{\ clan

 

156

25 r A. DNA CONTENT: IIICII DENSITY

 

20 .. 0 VEHICLE
v 1.5 pH MCTP
v 150 pM MCTP

 

 

 

Cellular DNA (pg/cell)

 

 

O 24 48 ' 168

Time (Hours)

25

B. DNA CONTENT: LOW DENSITY

 

 

 

 

 

 

E :-.. - ----- v
3 ." ..... t ‘7‘
\ ’1'. 0"..3.
g e" ‘ x" ...........
I \ '....
v J \
‘ M
z
:3
l- ---
.2
2 e VEHICLE
8 10 - V 1.5 uM MCTP
v 150 1.4M MCTP
5 I /

 

o 2.4 48 108
Time (Hours)

Figure 18: Cellular DNA content of MCTP-treated PECs.

PECs at high (A) or low (B) density were exposed to a single administration of
MCTP at time 0. Values represent means of replicate studies.

(n=5, High Density; n=3, Low Density) [Standard error of difference = 0.75 (HD);
1.70 (LD)] " Significantly different from vehicle control (p<0.05)

157

treatment period, so that by 7 days these had a final DNA content nearly double that
of controls.

DNA synthesis as determined by incorporation of [3H]thymidine is presented in
Figure 19. Incorporation of radiolabel into vehicle-treated cells peaked at 24 hours
in both high (Figure 19A) and low density (Figure 19B) cultures, but cells treated at
low density incorporated more than twice as much labeled thymidine as cells treated
at high density. A similar response was seen in cells treated with the low
concentration of MCTP. In contrast, incorporation was consistently less in cells
treated with 150 pM MCTP than in controls through 48 hours. However, thymidine
incorporation increased gradually in the high concentration group with time, and by
7 days post-treatment incorporation was equal to (high density) or much greater than
(low density) incorporation by vehicle-treated cells. In the low-density group, cells
treated with the low concentration of MCTP also incorporated significantly more
thymidine than controls by day 7.

Autoradiography was performed to analyze DNA synthesis at the level of
individual cells and to determine whether the changes seen in thymidine
incorporation were due to changes in thymidine pool size, dilution or altered
transport of labeled thymidine rather than changes in DNA synthesis (Wang et al.,
1989). Table 1 shows the results from these studies. In all treatment groups, cells
treated at low density had a higher percentage of labeled nuclei than cells treated at
high density. At 24 hours post-treatment, there was little difference between treated
and control groups in the percent of cells labeled, but cells treated with 150 uM

MCTP were labeled less intensely. At 48 hours post-treatment, the percent of

158

125 A. [3H] THYMIDINE INCORPORATION:

 

 

 

 

 

 

 

F HIGH DENSITY
IOO -
3 e VEHICLE
3 v I.5 nu MCTP
° 75 - v 150 nu MCTP
O
O
O
\ 50 .
E"
E
25 -
O ' . J ' ' I I/ -
0 24 48 168
Time (Hours)
125 8. [3H] THYMIDINE INCORPORATION:
( . Low DENSITY

a

‘ e VEHICLE i
l v 1.5 nu MCTP I
I v I50 nu MCTP;

 

DI’M / 1000 Cells
1 <1}

.............
0"
.....

 

 

O 24 es 163
Time (Hours)

Figure 19: Effects of MCTP on incorporation of [3H]thymidine.

PECs grown at high (A) or low (B) density were exposed to a single administration
of MCTP at time 0. Values represent means of replicate studies. (n=5, High
Density; n=3, Low Density) [Standard error of difference = 4.86 (HD); 7.00 (LD)]

" Significantly different from vehicle control (p<0.05)

159

«on

$0

«an

..mch... .233 22:2. ES. 32...... 2:38:53 e
duo... E2533 .2. 3:238 N .o :03 :e 23.58 32.. N .0 2mm .. ... 2.2:: Eons—.2 253) .o.

 

 

 

 

 

 

 

 

 

 

 

 

 

«mam eNnd .\+ mmd— thus. 5.: On—
mma fimv 1.. pm .8 FN enmd .\+ mod thus. 5.: mé
25— 3 K ....mu $0 8.0 .\+ 00.0 0.0.30) 30..
0mm *0 $N— mama coed ... + No. 3 ...—.05. 5.: amp
moo— «one 8am $6 «.36 A... mod ...—.05. 5.: m;
050— ..me $c— fin n—d .\ + m3.— o.u...o> :22 .25 b
acN ....o in: fimm LE... .1. 095— ...—.05. .2: amp
53 8mm .... .n 80— NNd .\+ mmfi .302 5.: m;
was ....No flaw fin th .\ + 5.x. 0.9.30) 26..
one 86 Raw 8 K echd .\+ mmé ....OS. 5.: amp
onop 80v ion fimn and .\+ hmfi ...—.05. 5.: mé
86— $60 $00 iv and .\ + —v.N o.u...o> so... .50... av
Nam .... 3 «one ....ON emmd .s + «QNN thus. 5.: On—
hno *2. flow $0 3.0 .s + 5.0. ...—.05. E: m;
5N5 «one ism $O— cmd -\ + gum— o.o...e> 30..
009— *0— $00 $8 0.6 4+ Na; .....02 5.: amp
moo— $ 5 ion *m— mmd .\+ «ed ...—.02 2: mé
pno— 3N5 imp flop cod .\ + NON o.u...o> £3... 8:0: «N
Dub—.500 ...“..de On A an . Op a— v .3 .505 22>: h2w2b<w¢h >....m2mo mi;
“.0 $3232 ....mu 03me can. wz.<c@ .3ng “.0 .8 430
“.0 zogamztma

>Ia<m005<¢0h3< >m Omh<34<>m m< aOwn. >0 m.mm......2>m <20

u p m..m<...

160
labeled cells increased in PECs treated with the high concentration of MCTP and

most of these cells were labeled much less densely. Although the percentage of
labeled cells remained similar to controls in the group treated with 1.5 uM MCTP,
these cells were labeled less intensely. By day 7, there was no difference in labeling
between low and high density controls. However, cells treated with either
concentration of MCTP showed a higher percentage of labeled cells, with a shift
fiom intense incorporation of thymidine (> 50 grains/ cell) in fewer cells to a lesser
incorporation (<10 grains/ cell) in a larger population of cells.

Incorporation of radiolabeled leucine and uridine were measured as indicators
of protein and RNA synthesis, respectively. In control cells, incorporation of each
of these was higher in the low density groups (Figures 20 and 21). Although leucine
incorporation was depressed shortly after treatment with 150 uM MCTP as compared
to controls, measurable leucine incorporation was maintained. By 7 days post-
treatment, leucine incorporation into cells treated with MCTP was significantly
greater than that in vehicle-treated controls. The same was true for uridine
incorporation: whereas incorporation per cell was significantly lower during the first
36 hours after treatment with high concentrations of MCTP, treated cells continued
to synthesize RNA throughout the post-treatment period and by 7 days were
incorporating more labeled uridine than controls. Incorporation was generally much
greater in cells at low density, but, as control PECs reached confluence (ie., day 7),
incorporation decreased to the levels seen in high-density cultures (Figures 20A &

21A).

161

10° r A. [an] LEUCINE INCORPORATION:
IIIGII DENSITY

 

\l
U‘
I

e VEHICLE
v I.5 rm MCTP
v I50 nu MCTI’

 

 

 

DPM / 1000 Cells
in
o

 

 

 

‘~ 0
i '- -.._“~.':~ . ---v
'.- ‘-. -7- ................. —.
25 .' ......... g\§ ______
o 7] i /;i—
0 24 ‘8 168

Time (Hours)

 

 

 

 

 

 

 

too B. [an] LEUCINE INCORPORATION: ,
[ LOW DENSITY ...... v
.2 , ---;
3 . ,v
U .‘. ’4 \‘
Q '_ r’ \
O ‘.. \
o “
.2 . e\"z
a. ‘0
C: II
25 ~ 0 VEHICLE fl
v 1.5 [1.” MCTP
v 150 uM MCTP
0 f j X ’//_I_
o 24 43 168

Time (Hours)

Figure 20: Effects of MCTP on incorporation of [3H]leucine.

PECs grown at high (A) or low (B) density were exposed to a single administration
of MCTP at time 0. Values represent means of replicate studies.

(n=5, High Density; n=3, Low Density) [Standard error of difference = 4.7 (HD);
10.2 (LD)] ‘ Significantly different from vehicle control (p<0.05)

 

I62

 

 

 

 

 

 

 

 

 

 

 

 

 

450 ' A. [an] URIDINE INCORPORATION:
IIIGII DENSITY
400 ~
350 e
D
:3 300 - O VEHICLE
o v 1.5 nu MCTP
o 250 _ v I50 ,III MCTP
8
" 200 ~
\ .
E 150 . """ '
G W8 ..... 'g ...... :3
'00 t1..-“ ...... .-
. e . """""""" V """" e
50 - ‘
on---n._-n-: :1.
O 24 48 168
Time (llours)
450 ~ 8. [3H] URIDINE INCORPORATION:
v Low DENSITY
400 ‘ /”.‘s\
350 ”I --“~.’ .
3 ’I ./ \
7, 300 °
U ' ----------- v" .
g 250 '_ ' ' ..... v
S -'
V. '0 e e
\ 200 i“: v, ---v
2 150 5'"
D.
a ‘ e VEIIICLE '
100 ,_ V 1.5 u” MCTP
v I50 nu MCTP
50 ~
on--....-.- t
0 24 48 168

Time (Hours)

Figure 21: Effects of MCTP on incorporation of [3H]uridine.

PECs grown at high (A) or low (B) density were exposed to a single administration
of MCTP at time 0. Values represent means of replicate studies.

(n=5, High Density; n=

3, Low Density) [Standard error of difference = 13.1 (HD);

41.1 (LD)] ‘ Significantly different from vehicle control (p<0.05)

163
Recognizing that larger cells may synthesize more protein as their cytoplasmic

volume increases, protein synthesis was also normalized to protein content (data not
shown). Although leucine incorporation into high density, MCTP-treated cells was
no longer significantly increased at 7 days, in all other respects the pattern of leucine

incorporation was similar whether normalized to protein content or cell number.

This study was undertaken to understand how MCTP treatment of cultured
porcine pulmonary artery endothelial cells affects DNA synthesis and normal cellular
functions as represented by synthesis of RNA and protein. PECs treated with MCTP
in vitm are unable to divide yet remain attached to the culture dish and continue to
exclude trypan blue, suggesting that they remain viable in culture for long periods of
time (Reindel et al., 1991). Treated cells also increase in size and maintain an
apparently intact monolayer throughout a 7 day treatment period. It has yet to be
determined what functional changes occur in cultured endothelial cells after
treatment with MCTP and how similar changes in vivo might contribute to the
progression of lung vascular injury and the development of pulmonary hypertension
which are characteristic of MCTP pneumotoxicity (Roth and Ganey, 1988; Roth et
al., 1989). Our data indicate that although cells treated with high concentrations of
MCTP do not divide, they are not functionally quiescent but continue to synthesize

DNA, RNA and protein.

CIO!

164
MCTP is a bifunctional alkylating agent, capable of crosslinking DNA. DNA

crosslinks have been demonstrated in livers of MCT-treated rats (Petry et al., 1984)
as well as in cultured PECs treated with MCTP (Wagner et al., 1991). In PECs, the
DNA crosslinking appears to be long-lasting and may be responsible for the arrest
of cell proliferation. It is not yet clear whether or not cultured PECs retain the
ability to repair these crosslinks, but our results suggest that DNA synthesis and/or
repair continues at low levels in these cells after MCTP treatment (Figure 19; Table
1). Although PECs treated with 150 uM/ml MCTP incorporated significantly less
labeled thymidine than controls for the first 48 hours after treatment, they
nevertheless continued to incorporate thymidine gradually, and by 7 days post-
treatment they were synthesizing DNA at levels equal to or greater than controls.
This efiect was greater in PECs treated at low density, suggesting a larger population
of active cells. Results using incorporation of labeled thymidine into TCA-
precipitable material were supported by autoradiographic studies: in all cases,
altered incorporation of [3H] thymidine reflected a subpopulation of cells actively
synthesizing DNA.

From previous work with synchronized PECs we know that, following release
from arrest, DNA synthesis peaks between 18 and 24 hours, preceding the increase
in cellular DNA content by a few hours. Cell division occurs between 36 and 48
hours. After this time, synchronization is lost to some extent and the pattern is much
less distinct. As stated above, PECs in an apparently confluent monolayer undergo
some increase in cell number after release from arrest. In this study, whereas a

majority of cells at low density divided (note 100% increase in vehicle-treated cell

165
number fi'om 24 to 48 hours, Figure 16B), a relatively low percentage of those at

high density did so (vehicle-treated cell numbers increased only 25% during the same
time interval, Figure 16A). This was also evidenced by the autoradiographic results
in Table 1. While 2% of high density, vehicle-treated cells were labeled with
thymidine during a 2 hour period, 16% of low density, vehicle treated cells were
labeled. Incorporation of [3H]thymidine per cell was also much lower in the high
density controls than in low density controls (Figures 19A and 19B, 24 hours),
supporting the suggestion of a smaller population of active cells.

Taken together, the results suggest that changes in thymidine incorporation
seen in Figure 4 and Table 1 are consistent with changes in DNA synthesis. While
changes in DNA content at 24 hours were subtle in the high density cells, it must be
considered that relatively few cells were increasing their DNA content at this time,
and these cells were not acting in perfect synchrony. Some cells may have increased
their DNA content earlier and gone on to divide sooner (some increase in cell
numbers was evident by 36 hours), whereas others may have done so somewhat later.

In cells treated with MCI'P, incorporation of thymidine occurred at a much
reduced rate per cell. Some of the incorporated thymidine may have been used in
repair of damaged DNA. little is known about the DNA repair mechanisms in this
cell type in culture. There may be a basal level of repair which is comparable to that
in the endothelium in vivo, or there may be an altered capacity for repair, perhaps
as a result of increased DNA damage under culture conditions. The extent to which
DNA repair was responsible for the increased thymidine incorporation in these cells

is unknown, but since DNA crosslinking occurs with exposure to MCTP, repair of

166

damaged DNA is likely to represent at least part of the incorporation. However,
some of the thymidine seems to have been utilized in the synthesis of new DNA
despite the fact that cells treated with a high concentration of MCTP did not divide.
This is supported by the gradually increasing DNA content of MCTP-treated PECs
throughout the 7 day post-treatment period. By 7 days, PECs treated at low density
with 150 IIM MCTP had nearly twice the DNA content of controls.

Changes in DNA content were also less evident in cells at high density since
the percentage of cells actively incorporating [3H]thymidine was apparently very small
initially (Table 1). At low density, a higher percentage of treated cells were
incorporating labeled thymidine relative to controls, and this may explain why the
DNA content increased in parallel in treated and control cells despite a lower rate
of incorporation in those treated with 150 uM MCTP. But while control cells
divided and reestablished their basal DNA content, the treated cells maintained or
increased their DNA content over time. Low-density culture conditions provide a
strong stimulus for cell proliferation and may have triggered cells to enter the DNA
‘synthesis phase of the cell cycle, but since they were unable to divide some of the
MCTP-treated cells may have remained in a tetraploid configuration.

We examined the synthesis of RNA and protein in MCTP-treated PECs to
determine whether these basic cellular functions were maintained after treatment.
Whereas synthesis of both RNA and protein was somewhat lower than controls for
the first 48 hours after treatment with 150 IIM MCTP, their formation did continue
above a basal level. Indeed, by 7 days post treatment, MCTP-treated cells were

synthesizing significantly more RNA and protein than controls. These responses

167

were much more pronounced in cells treated at low density. This may reflect a
greater percentage of synthetically active cells and / or an increase in the synthetic
activity of the general population of cells. Undoubtedly, cells treated at low density
with 150 IIM MCTP had an increased cellular protein content that reflects increased
cell volume; but whereas some of the increased protein synthesis may be attributed
to increased cytoplasmic volume, there is a disproportionate increase in the synthetic
activity of these cells, suggesting that the cells are quite active. Of primary interest,
however, is the fact that both RNA and protein synthesis are ongoing under all
conditions, suggesting that MCTP-treated cells are not functionally quiescent.
Endothelial cells are metabolically complex and play a role in the
maintenance of vascular integrity and homeostasis. Disruption of normal endothelial
cell activity could result in a loss of the delicate balance these cells help to maintain.
It has been reported recently that exposure to a variety of cellular stresses, including
exposure to endotoxin (Chesney et al., 1974a; Fox and DiCorleto, 1984; Albelda et
al., 1989), heat-shock (Kelley and Schlesinger, 1978; Thomas et al., 1982; Williams
et al., 1989), hypoxia (King et al., 1989; Kourembanas et al., 1990; Ogawa et al.,
1990b) and acetylated low density lipoprotein (Fox and DiCorleto, 1986), may result
in altered gene expression in target tissues such as the vascular endothelium.
Changes in the amount or type of particular proteins produced by cells may be
involved in the expression of injury seen under these conditions. It is possible that
MCTP treatment causes a change in the production of proteins by endothelial cells
in a similar manner. Many of the changes seen in the pulmonary vasculature after

MCTP treatment in viva, such as vascular remodeling (Bruner et al., 1983; Reindel

168
et al., 1990), thrombosis of the pulmonary vessels (Bruner et al., 1983; Hilliker et al.,

1982), accumulation of platelets (Hilliker et aL, 1982; White and Roth, 1988) and
altered vascular reactivity (Hilliker and Roth, 1985a), could be initiated or
exacerbated by changes in the profile of endothelial cell-derived proteins, including
growth factors, chemotaxins, coagulation and fibrinolytic factors and vasoactive
agents.

These results suggest that marked changes in PEC function occur with MCTP
treatment. The increased RNA and protein synthesis observed may reflect
overexpression of proteins normally synthesized by cultured PECs, but it is also
possible that a somewhat different profile of proteins is expressed in MCTP-treated
cells. Changes of this nature might play a role in the delayed and progressive

development of MCTP pneumotoxicity.

Chapter IV

MONOCROTALINE PYRROLE TREATMENT DOES NOT ENHANCE THE

RELEASE OF MESENCHYMAL CELL GROWTH FACTORS

FROM CULTURED ENDOTHELIAL CELLS

169

170
Summary

Cultured porcine pulmonary artery endothelial cells (PECs) treated with MCI'P are
unable to divide but continue to synthesize RNA and protein, ultimately in greater
amounts than controls. This increase in protein synthesis may reflect overexpression
of normally synthesized proteins, but cell injury could also result in expression of a
new profile of endothelial cell gene products. In certain models of vascular injury,
endothelial cells demonstrate increased production of mesenchymal cell mitogens
which may contribute to the vascular remodeling process. The purpose of this study
was to determine whether PECs treated with MCTP also release factors that possess
growth-stimulatory activity. MCTP-treated PEC monolayers were co-cultured with
growth-arrested Swiss 3T3 fibroblasts in serum-free medium. In separate studies,
conditioned, serum-flee medium was collected from treated PECs and added to
growth-arrested Swiss 3T3 cells. At various times posttreatment, proliferation of
SWiss 3T3 cells was determined. Under coculture conditions, grth stimulation of
3T3 cells by untreated PECs was maximal. Medium conditioned by untreated PECs
also stimulated 3T3 cell proliferation but to a lesser extent; MCTP treatment did not
afigment the release of mitogenic activity in this system. Cultured endothelial cells
from other species constitutively produce and release high levels of platelet-derived
STOWth factor (PDGF) and other mitogens, and PECs may do this as well. MCTP
treatment does not appear to increase the release of growth factors above this basal
level in cultured PECs; however, the response of the quiescent endothelium in situ

t0 MCTP may be quite different.

171

Pulmonary vascular alterations which occur after MCT (P) treatment in viva
include both endothelial cell changes and remodeling of the vessel walls. The
vascular remodeling is Characterized by extension of smooth muscle into normally
nonmuscular arterioles and increased medial thickness in the small pulmonary
arteries (Hislop and Reid, 1974; Kay and Heath, 1969; Lalich et al., 1977; Merkow
and Kleinerman, 1966b; Meyrick and Reid, 1979; Meyrick et al., 1980; Reindel et al.,
1990). Similar lesions are evident in other forms of chronic pulmonary vascular
injury, including hypoxia- and irradiation-induced pulmonary hypertension (Abraham
et al., 1971; Meyrick and Reid, 1978; Perkett et al., 1986; Reid, 1979). Vascular
remodeling may precede the development of pulmonary hypertension or can occur
secondarily in response to increased pulmonary vascular pressure; regardless, once
established, it contributes to the maintenance of the hypertensive state by decreasing
the compliance and reducing the cross-sectional area of the pulmonary vasculature
(Reid, 1979; Meyrick and Reid, 1979).

The endothelium exerts a regulatory influence on the growth and proliferation
of other vascular cells, and changes in EC function with respect to this capacity may
contribute to the vascular remodeling process (Davies, 1986; Reid, 1979; Schwartz
and Ross, 1984). ECs modulate smooth muscle cell (SMC) proliferation through
composition of the extracellular matrix, direct communication via intercellular
junctions, and the release of humoral growth stimulatory and inhibitory factors

(Davies, 1986; Carey, 1991; Castellot et aL, 1981). Increased synthesis and/or release

172

of endothelium-derived growth factors (EDGFs) have been demonstrated in vitra
under conditions which cause EC injury, including exposure to chronic hypoxia,
irradiation, endotoxin and mechanical wounding (Vender et al., 1987; Witte et al.,
1989; Albelda et al., 1989; Libby et al., 1986; McNeil et al., 1989), and which are also
associated with EC alterations and vascular remodeling in viva (Meyrick and Reid,
1978; Perkett et al., 1986; Meyrick and Brigham, 1986; Reidy, 1990; Powell et al.,
1990). Similar release of EDGFs has not been established in situ but, if present,
could help to define a role for the endothelium in the development of Chronic
pulmonary vascular disease.

Qfltured PECs treated with MCTP, a putative pneumotoxic metabolite of
MCI, demonstrate limited cytolysis but are unable to divide (Reindel et al., 1991).
Treated cells become spindle-shaped in appearance or enlarge if monolayer integrity
is disturbed, but they remain viable in culture for long periods of time (Reindel et
al., 1991; Hoom et al., 1990). Basic cellular function, as illustrated by the ability to
synthesize RNA and protein, is maintained in treated PECs; indeed, by 7 days after
MCTP treatment, cells demonstrate higher synthetic activity than vehicle-treated
controls (Hoom and Roth, 1992). The profile of proteins synthesized by MCTP-
treated cells is not known. They may continue to synthesize the normal spectrum of
EC proteins but in larger amounts; alternatively, they might express a new set of
gene products in response to injury, including EDGFs. These studies were conducted
to determine if MCTP-treated ECs release increased amounts of mesenchymal cell

growth factors.

173

W: Primary isolation and subculture of PECs was performed as described
previously in Chapter II. PECs were maintained in Opti-MEM supplemented with
5% FCS and 1% Ab/Am under standard conditions of 37°C and 7.5% CO2 and
92.5% air. Swiss 3T3 fibroblasts (CCL 92; American Type Culture Collection,
Rockville, MD) were maintained in Dulbecco’s modified Eagle’s medium (DMEM;
GIBCO) supplemented with 10% FCS and 1% Ab /Am under similar conditions.

Medium was changed three times per week, unless otherwise indicated.

WM: MCTP was prepared from MCT as described previously in
Chapter II. MCI'P was dissolved in DMF at a concentration of 20 mg/ ml. All
dilutions were performed in DMF, and equal volumes of DMF served as the vehicle
control in all studies. Untreated cells and cells stimulated with 2-5 units/ml bovine

thrombin were also utilized as indicated.

E . I 2 l' . :

a.) Caculture Experiments: PECs were plated on 22 mm polycarbonate or
collagen culture plate inserts (Costar). When the PECs reached confluence (48-72
hr), 1.5 ml of fresh medium was added to each insert and the monolayers received
a single administration of MCTP or DMF vehicle (5 III vol). Meanwhile, 3T3 cells
were plated at 25,000 cells per 35 mm well. 24 hours later, the medium on the 3T3

cells was replaced with DMEM containing 0.25 % bovine serum albumin (BSA) and

174
1% Ab/Am [serum-free medium; SFM] for 72 hours to growth-arrest the cells. At

the end of this time, the PEC-containing inserts (72 hours posttreatment with MCTP)
were rinsed twice with CMF-HBSS and added to the growth-arrested 3T3 cells.
Fresh SFM was added to both chambers, and the cocultures were maintained for 72
hours. A second set of growth-arrested 3T3 cells was prepared for coculture with
PECs 8 days posttreatment with MCTP, and these were also maintained for 72 hours.

In addition, several wells of growth-arrested 3T3 cells were cocultured with
untreated PEC monolayers or monolayers that had been stimulated with thrombin
24 hours previously. These were handled in the same way as those cocultured with
MCTP or vehicle. Additional wells of 3T3 cells were incubated with cell-free inserts
in the presence either of SF M or of DMEM containing 10% F C8 to define minimum
and maximum proliferation for the system, respectively.

b.) WW. PECs were plated in T-75 tissue culture
flasks (Costar) and were incubated under standard conditions for 7-10 days prior to
treatment to establish quiescent monolayers. At this time, monolayers were rinsed
twice with CMF-HBSS and 15 ml fresh Opti-MEM was added. PECs received a
single administration of MCTP or DMF vehicle (37.5 III) to achieve the nominal
concentrations utilized in these studies. At 48 hours posttreatment, the monolayers
were rinsed twice with CMF-HBSS, and 15 ml of SFM was added to each flask. The
PEC-conditioned medium (CM) was collected after 72 hours (5 days posttreatment
with MCTP), and 15 ml of fresh SFM was again added to each flask; this was

collected 48 hours later (7 days posttreatment with MCTP).

175

The CM was spun in a centrifuge at 800 x g for 10 minutes, and a portion was
diluted 1:1 with fresh SFM. 3-ml aliquots of 100% or diluted CM were added to
wells containing growth-arrested 3T3 cells (prepared as above), and the plates were
incubated under standard conditions for 72 hours.

As described above, CM from untreated PECs or PECs stimulated with
thrombin for 24 hr was also used on separate wells of growth-arrested 3T3 cells.
Additional wells maintained with unconditioned SFM or with DMEM containing

10% FCS were used to determine minimum and maximum proliferation for the

system, respectively.

W: At the end of the coculture period, or subsequent to the
collection of CM from the PEC monolayers, PECs were removed from the inserts or

flasks and counted as described previously in Chapter II.

a.) Enumeration of 3T3 cells: At the end of the 72 hour period of coculture
or incubation with CM, adherent 3T3 cells were removed fiom the wells and counted
as described previously in Chapter II.

b.) Incorporation af[3H]thymidine: In some studies, cell proliferation was also
quantified by the incorporation of radiolabeled thymidine over the 824 hr and 24-48
hr intervals of the coculture period or during the 18-36 hr interval after the addition
of CM. [5-methyl-3H] thymidine (40-60 Ci / mmol; Amersham) was added to the

culture medium at a concentration of 1 uCi / ml for the period indicated. After

176

incubation, incorporation of thymidine into TCA-precipitable material was quantified

as described in Chapter III.

mm. Data are presented as means + /- SEM of 3 separate
experiments, each consisting of 3 wells per experimental parameter (CM studies: 2
T-75 flasks). Percentage data were transformed by the arcsin‘l/2 transformation
before statistical analysis. Data were analyzed using a random AN OVA, and
individual comparisons were made using Tukey’s omega as the past hac test. The

criterion for significance was P< 0.05.

Beams

Cellularity of M CT P-treated PE C monolayers on inserts or in flasks followed
a pattern typical for this cell type. Vehicle-treated monolayers and monolayers
treated with a low concentration of MCTP contained similar numbers of cells to
untreated PEC monolayers, whereas those treated with higher concentrations of
MCTP contained fewer cells. This effect was no more pronounced at 10 days
posttreatment than at 5 days. [Figures 22 & 23]

The effects of coculture with MCTP-treated PECs on the proliferation of
growth-arrested 3T3 cells is shown in Figure 24. 313 cell number is presented as a

percent of the response achieved by medium containing 10% FCS (i.e., maximum

177

PEC Number

 

-5d Posiireotment
D 10d Posiireotmenl

 

 

 

1200 l- . _ --...- W..-

IOOO -

BOO -

6OO —

 

400 L

Cells/Well (X 1000)

ZOO -

 

 

 

 

 

 

 

 

 

 

NT VEH M 1.5 M 150 THR
Treatment

Figure 22: Effects of MCTP on cellularity of monolayers grown on inserts.

Confluent monolayers of PECs grown on culture plate inserts were treated with
MCTP at time 0. At the times indicated posttreatment, cells were removed from the
inserts and counted. Values represent means + /- SEM of 3 replicate experiments,
each treatment group consisting of 3 individual wells. " Significantly different from

vehicle control (p < 0.05).

7000

6000
A
8

o 5000
x

V 4000
x
(I)
2

L. 3000
\
2
'6

0 2000

1000

0.

Figure 23:

178

PEC Number

 

 

NT VEH M 1.5 M15 M150 THR
Treatment

Effects of MCTP on cellularity of monolayers grown in flasks.

Post-confluent monolayers of PECs received a single administration of MCI? at time
0. After collection of conditioned medium at 7 days posttreatment, cells were
removed from the flasks and counted. Values represent means +/- SEM of 4
replicate studies, each treatment group consisting of 2 flasks. ' Significantly
different from vehicle control (p<0.05).

179

A. Coculture (3-5 days)

 

IOO— I
Tr?
.. T I l
“- 80-
N
o
C’ so *
E I
3
E
'; 40-
U
2
N 20..
10% 07. NT van—ms MISO THR

PCS PCS PEC Treatment

8. Coculiure (8-10 days)

100 r-

i
i l
i
80—
60- [
40~
20—

107: 09: NT VEH—MI.5—MISO THR
FCS FCS PEC Treatment

7.; Maximum (10% FCS)

 

 

Figure 24: Effects of co-culture with MCTP-treated PECs on
Swiss 313 cell number.

Growth-arrested 3T3 cells were co-cultured with MCTP-treated PECs during the
indicated intervals posttreatment. Values represent means + /- SEM of 3 replicate
experiments, each treatment group consisting of 3 wells. Results are expressed as a
percent of the proliferation stimulated by exposure to 10% FCS. ‘ Significantly
different from vehicle control (p<0.05)

180

response). Untreated PECs stimulated Swiss 3T3 cell proliferation at a level 80-
100% of that stimulated by 10% FCS, and treatment with MCTP or thrombin did not
augment this response. Cells treated with a high concentration of MCTP stimulated
less proliferation than vehicle controls; however, PEC monolayers in this treatment
group were also less cellular. Similar results were obtained with PECs 3-5 days
posttreatment [Figure 24A] and 8-10 days posttreatment [Figure 24B]. Incorporation
of [3H]thymidine paralleled these increases in 3T3 cell number, but results were less
uniform [data not shown].

The efi'ects of CM from MCTP-treated monolayers on 3T3 cell number are
shown in Figure 25. CM from untreated PECs stimulated proliferation of growth-
arrested 3T3 cells at approximately 80% of maximum, and CM from MCTP-treated
cells at 2-5 days [Figure 25A] or 5-7 days [Figure 258] posttreatment did not
significantly increase this response. Dilution to 50% strength reduced the growth
stimulation capacity of the CM slightly in all treatment groups [Figure 25], and
dilution to 25 % strength reduced 3T3 cell proliferation to a level not significantly
different from that in SFM [data not shown]. CM from PECs treated with higher
concentrations of MCI? did not demonstrate a decreased ability to stimulate
proliferation under these conditions, despite significantly lower PEC numbers in
these treatment groups. Again, results of [3H]thymidine incorporation reflected
comparable effects on cell proliferation' [data not shown].

CM from thrombin-stimulated PECs did increase 313 cell proliferation under

these conditions, but this response was quite variable among experiments. 50%

181

A. Conditioned Medium (2-5 days)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

100
6? ] i I
o l l
h.
N 30 l
0
V 60
E
3
E
'§ 40
O
2
N 20
o ._ -
10% 07. NT VEH MLS m5 M150 THR
rcs FCS PEC Treatment
-CM
[:]1:1 on
B. Conditioned Medium (5-7 days)
100»-
8 I T
“- 80-
.. i
o _
V 60-
E .
3
E
g 40-
o .
2
N 20-
0 _
10% 0% NT VEH MI.5 ms r4150
FCS FCS PEC Treatment

Figure 25: Effects of medium conditioned by MCTP-treated PECs on
Swiss 3T3 cell number.

Growth-arrested 3T3 cells were incubated for 72 hr with conditioned medium
collected during the indicated intervals posttreatment. Medium was applied at full
strength or diluted 1:1 with fresh, serum-free medium as indicated. Values represent
means + /- SEM of four replicate experiments, each treatment group consisting of
3 individual wells. No significant difference from controls.

182

dilution of this CM decreased its ability to stimulate 3T3 cell proliferation to a

greater extent than dilution of CM from untreated or MCTP-treated cells.

The endothelium plays an important role in the regulation of SMC
growth and proliferation in the vessel wall, and several models of lung injury which
are characterized by structural remodeling of the vasculature also demonstrate
evidence of relatively early EC alterations (Meyrick and Reid, 1978; Perkett et al.,
1986; Crapo et al., 1980). Cultured ECs from a variety of species synthesize and
release factors which are mitogenic and chemotactic for SMCs and other
mesenchymal cells in response to stimulation with thrombin or after toxic or physiml
injury (Harlan et al., 1986; Schini et al., 1989; Hsieh et al., 1991; Vender er al., 1987;
Witte et al., 1989). MCTP-treated PECs remain active synthetically, and it is not
unreasonable to suggest that they may produce altered amounts of normally
expressed proteins or, alternatively, that they may express a new complement of
proteins in response to exposure to MCTP. Therefore, it is plausible that similar
alterations in EDGF production and / or release could occur in MCTP-treated PECs,
and the studies presented here were designed to determine the effects of MCTP
treatment on the release of soluble mediators from PECs that influence mesenchymal

cell proliferation.

183

A number of preliminary experiments were performed in order to optimize
the experimental conditions and allow appropriate controls to be incorporated into
the design. The use of a transformed cell line of 3T3 fibroblasts eliminated the
difficulties of maintaining two primary cell lines (i.e., PECs and SMCs) at optimal
levels simultaneously. 3T3 cells are frequently used to bioassay for grth factor
production and respond with cell division to FCS and SMC mitogens like PDGF,
bFGF and ET (Vogel et al., 1978; Wang et al., 1989; Csonka and Jellinek, 1986;
Wang et al., 1981). Preliminary work with the growth-arrested 3T3 cells used in
these studies demonstrated that they were able to respond appropriately to these
factors (at concentrations of 1-10 ng/ml) in our hands as well [data not shown]. In
addition, serum-flee conditions were established that could sustain PECs and permit
a "normal” response to MCTP yet would not mask the production of growth factors
by these cells. Despite these and other precautions, the results of the studies did not
provide evidence that MCTP altered release of growth factor(s) by PECs.

The 3T3 cell bioassay system used in these studies is a very sensitive one and
is frequently used to detect the production or release of low levels of SMC mitogens;
concentrations of growth factor which are undetectable by Western blotting or
radioimmunoassay can stimulate cell proliferation under these conditions (Vogel et
al., 1978; Wang et al., 1989). However, bioassay is of limited usefulness in the
presence of high levels of growth factor because once 100% proliferation has been
achieved there is no way to detect further increases in growth factor concentration.

Under coculture conditions, stimulation of 313 cell proliferation by untreated

PECs was maximal: virtually 100% of 3T3 cells divided. Thus, MCTP treatment and

184

thrombin stimulation could not augment this response. PECs treated with 150 uM
MCTP stimulated less proliferation of 3T3 cells, but there were also 40-50% fewer
cells on the inserts in this treatment group. Apparently, cellular production of
mitogenic activity is maintained after MCTP treatment and is not increased by
treatment with high concentrations of MCTP; however, enhanced release of
mitogenic activity in other treatment groups cannot be detected in this system.

The effects of basement membrane composition can influence EC shape and
behavior and may affect the release of mitogens (Huber and Weiss, 1989; Madri and
Pratt, 1986; Gospodarowicz and Lui, 1981; Gospodarowicz and Ill, 1980). In an
effort to decrease basal EDGF production, both collagen and polycarbonate culture
plate inserts were utilized, but this variable had minimal impact on the outcome of
the studies. PEC monolayers were also maintained for longer periods of time in
culture to establish a more quiescent monolayer prior to treatment, but this interval
was limited by the tendency for older monolayers to detach from the insert during
the repeated manipulations required for feeding, treatment, etc.

Because of the limitations of the coculture system, studies with PEC CM were
initiated. Monolayers grown in flasks could be maintained in culture for longer
periods of time, and the conditioned medium generated in this manner could be
diluted as necessary to modify the 3T3 cell proliferative response. CM from
untreated cells varied in its ability to stimulate 3T3 cell proliferation but generally
elicited a response 70-80% of maximum. Dilution of the CM resulted in a further
decrease in this proliferative response. Maintenance of the monolayers for 21 days

in collagen-coated flasks prior to treatment reduced the mitogenic potential of the

185

CM to 50-60% of maximum, but still did not result in completely quiescent
monolayers. However, CM from thrombin-stimulated PECs was able to increase
proliferation to a variable extent, suggesting that the bioassay system was able to
detect an augmentation in the release of mitogenic activity under these conditions.

CM from MCTP-treated PECs did not increase or decrease the proliferation
of 313 cells as compared to CM from untreated or vehicle-treated controls, and
dilution decreased the growth-promoting activity of the CM to a similar extent in all
treatment groups. These results suggest that MCTP does not alter the production
of mesenchymal cell growth factors by cultured PECs. Decreases in 3T3 cell
proliferation which were seen in the coculture system when PECs were treated with
high concentrations of MCI? are not evident in this system despite similar decreases
in monolayer cellularity. It is possible that the cells in the flasks treated with 150 uM
MCTP generate more mitogenic activity per cell than those in the other treatment
groups, but this would be of questionable significance in a vessel as the release per
unit area is not apparently different from unstimulated cells.

It is possible that the time intervals examined in these studies were
inappropriate. Using information from previous studies, intervals for coculture and
collection of CM were selected to bracket the onset of increased RNA and protein
synthesis in MCTP-treated PECs (Chapter 111). It is possible that a spike of mitogen
release would be missed using this experimental protocol; however, a transient
release of growth factor would be of questionable significance in the development of

vascular remodeling in this model.

186

There are certain difficulties inherent in a cultured cell system, particularly
in dealing with the regulation of proliferation. Although the endothelium in situ is
relatively quiescent in terms of growth factor production (Shepherd and Katusic,
1991), the act of taking ECs out of their normal environment and growing them in
culture changes them in such a way that their unstimulated production of mitogens
(predominantly PDGF) is increased greatly (Vlodavsky et al., 1987; DiCorleto and
Bowen-Pope, 1983; Barrett et al., 1984). Growth under low—serum conditions appears
to enhance this effect in human ECs. Despite this high level of basal activity, EC
injury in vitro has been associated with increased release of EDGFs (Madri et al.,
1991; Vender et al., 1987; Albelda et al., 1989). Thus, our results would appear to
indicate that increased production and/or release of mitogens is not a feature of
MCTP-induced EC injury.

Vascular remodeling is a characteristic feature of MCT(P)-induced
pneumotoxicity. Whether this is due to hyperplasia, hypertrophy and/or increased
chemotaxis of SMCs has not been definitively established in this model, but a change
in the regulation of SMC growth and behavior is evident (Meyrick and Reid, 1982;
Orlinska et aL, 1988; Olson et al., 1985). The elaboration of EDGFs is only one way
in which ECs may influence activity of SMCs. Changes in intercellular
communication between ECs and SMCs or in the composition of the extracellular
matrix can affect the growth and phenotype of SMCs (Sheridan and Atkinson, 1985;
Herman, 1990; Carey, 1991; Davies, 1986). SMC chemotactic factors and mitogens
may be derived from platelets, which accumulate in the lung during MCTP-induced

pneumotoxicity, perhaps as a result of EC alterations (Ross et al., 1986; Assoian and

187
Sporn, 1986; Oka and Orth, 1983; White and Roth, 1988). Alternatively, there may

be decreased production of growth-inhibitory substances (i.e., heparan sulfate) by
ECs (Castellot et al., 1981). These alterations would not be evident in the systems
utilized to evaluate cell proliferation in this study.

Despite our findings, the contribution to vascular remodeling of EDGFs
produced by pulmonary endothelium exposed to MCT (P) in vivo should not be ruled
out. Although a valuable tool in many ways, the cultured cell system is a limited one
for studies of this nature. Cultured ECs exist in a perpetually activated state with
respect to growth factor production. Quiescent endothelium in viva may respond to
MCT(P)-induced injury (and other forms of injury) quite differently in this respect.
Little work has been done in viva on the role of endothelium-derived growth factors
in this and other models of vascular injury, perhaps because the technology has not
been available until recently. Preliminary studies with PDGF-neutralizing antibody
suggest that this mediator is not involved in MCTP pneumotoxicity (Ganey et al.,
1988); however, the potential for paracrine modulation within the microenvironment
of the vessel wall cannot be ruled out by those studies. In situ hybridization or
immunohistochemical techniques with cDNA probes and/ or antibodies for the
mediators of interest (as they become available) may be able to determine more
conclusively whether changes in growth factor production by the endothelium play

a part in this model of lung vascular injury.

Chapter V

EFFECTS OF MONOCROTALINE PYRROLE 0N CULTURED

RAT PULMONARY EN DOTHELIUM

188

189
Summary

Monocrotaline pyrrole (MCTP) causes direct toxicity to cultured bovine and porcine
pulmonary artery endothelial cells (BECs and PECs, respectively), but there exist
species differences both in whole-animal response to the parent alkaloid and in
cellular response to direct application of MCTP. In this study, the changes in
cultured rat pulmonary vascular endothelial cells (RECs) after a single administration
of MCTP were characterized in order to compare these with changes previously
identified in this species in viva. MCTP caused a delayed and progressive release of
lactate dehydrogenase from REC monolayers which consisted primarily of isozymes
4 and 5. Progressive cell detachment was evident and remaining cells became
enlarged, with morphologic changes comparable to those reported previously in
BECs, including cytoplasmic vacuolization and nuclear enlargement. MCTP also
caused an inhibition of cell proliferation at concentrations of 0.15 uM MCTP or
greater, and DNA crosslinking was evident at 24 and 48 hours posttreatment. These
results suggest that MCTP is directly toxic to cultured RBCs, and the development
of changes is reminiscent of that seen in the rat in viva. The cytostatic nature of the
compound, in combination with its cytolytic effect on RECs, could contribute to the
development of pulmonary edema and other lung vascular changes seen in rats

treated with MCT or MCTP.

190

MCTP, a putative toxic metabolite of MCT, is unstable in aqueous solution
(Bruner et al., 1986), and it is likely that rapid binding of this electrophile to cellular
constituents of the pulmonary vasculature is the initiating event in the pneumotoxicity
seen with this compound. DNA crosslinking has been reported as a relatively early
phenomenon following MCT treatment in vivo (Petty et al., 1984), as well as in cells
treated with MCTP in vitro (Wagner et al., 1991; Wagner et al., 1992). Whereas the
pattern and progression of the pulmonary injury caused by MCTP in viva have been
well characterized in the rat, the link between binding of this short-lived electrophile
and the delayed vascular changes has not been clearly identified.

The endothelium has been suggested as a target for binding of MCTP in the
lung, and changes in pulmonary vascular endothelial cell structure and function have
been reported as relatively early events in MCT and MCI'P pneumotoxicity (Reindel
et al., 1990; Bruner et al., 1983; Hilliker et al., 1982; Molteni et al., 1984; Rosenberg
and Rabinovitch, 1988). We have shown that a single administration of MCTP to
cultured bovine and porcine pulmonary artery endothelial cells (BECs and PECs,
respectively) causes delayed and progressive injury which is reminiscent of that which
occurs in viva (Reindel and Roth, 1991; Reindel et al., 1991). However, the response
of endothelial cells of these two species to MCTP is quite different. Both BECs and
PECs are sensitive to the antiproliferative effects of MCTP, and both respond with
a release of prostacyclin (Reindel et al., 1991). However, whereas BECs are sensitive

to the cytolytic and hypertrophic effects of MCTP, PECs are relatively insensitive,

191
showing minimal cell detachment and release of LDH and more subtle morphologic

changes.

The manifestation of toxicity of the parent alkaloid, MCT, in cattle and pigs
is also quite different in viva. In cattle, MCT causes a hepatic venoocclusive disease
and pneumotoxicity is not evident (Sanders et al., 1936; Sippel, 1964). Clinical
reports indicate that ingestion of MCI‘ by pigs results in delayed and progressive lung
injury which is similar to that described in the rat. These differences may be due in
part to differences in metabolism or in distribution of the toxic metabolite(s) (Shull
et al., 1976; Cheeke and Pierson-Goeger, 1983). However, differences in sensitivity
of endothelial cells to reactive pyrroles might also influence the organ pathology
observed in animals intoxicated with monocrotaline or other pyrrolizidine alkaloids
(Reindel et al., 1991).

Because most attention has centered on the rat in the MCT model of
pulmonary vascular injury and in light of the species differences in response to MCT
and MCTP both in viva and in vitro, this study was conducted to determine how

cultured rat endothelial cells respond to direct administration of MCTP.

WW5

WM. Rat pulmonary vascular endothelial cells were
generously provided by Dr. James Varani (Dept. of Pathology, University of

Michigan, Ann Arbor, Michigan). These cells were isolated by the method of Ryan

192

and White (1986) and demonstrate characteristic cobblestone morphology. RBCs

were further characterized as described in Chapter II.

RBCs were maintained in DMEM (GIBCO) containing 10% FCS and 1% Ab/Am
and were incubated under standard conditions of 37°C and 6% C02/ 94% air. Cells
were passed at a ratio of 1:10 to 1:20 following enzymic (0.025% trypsin—0.27 mM
EDTA) or mechanical dissociation. Cells used for these studies were between
passages 6 and 10. Unless otherwise noted, culture medium was replaced with fresh

DMEM containing 10% FCS and 1% Ab/Am every other day.

W: MCTP was prepared from MCT as described previously in
Chapter II. MCTP was dissolved in DMF at a concentration of 20 pg MCTP/ml,
and all dilutions of MCTP were made with DMF. A 2.5 ul volume of MCI?
solutions or DMF vehicle (0 uM MCTP) per milliliter of medium was used in all
studies to achieve the nominal concentrations of MCTP (0.015-150 pM) used in these

studies.

WWW: After exposure of

confluent REC monolayers in 12-well tissue culture clusters to a single administration
of MCTP on day 0, cells remaining in the monolayers were photographed and

counted (as described previously in Chapter II) on days 1-5 posttreatment.

 

' : RECs were plated into 12-well tissue
culture clusters and allowed to proliferate until monolayers reached confluence. At
this time (day 0), fresh medium was added and monolayers were exposed to a single
administration of 0, 1.5, 15 or 150 pM MCTP in the culture medium. On each of the
subsequent 5 days, one 12-well plate of cells was used for determination of LDH
activity by the method of Bergmeyer and Bemt (1974). Medium from above
monolayers was removed for LDH analysis, and monolayers were washed twice with
CMF-HBSS. Two milliliters of fresh CMF-HBSS were added, and cells were lysed

with 15 pl of a 10% solution of Triton X-100 (Sigma).

Percent LDH release was determined by the following formula:

LDH in medium above monolayer
% release = x 100
LDH from lysed cells +
LDH in medium above monolayer

 

W. Aliquots of culture medium and cell lysate collected

each day for analysis of total LDH release as described above were also submitted
for LDH isozyme analysis. Prior to electrophoresis, samples were diluted with
phosphate-buffered saline to a total activity of 400-600 IU/L. Isozymes were
separated in agarose gels (Paragon Lactate Dehydrogenase Isoenzyme
Electrophoresis Kit; Beckman Instruments, Inc., Fullerton, CA). A commercially
prepared control serum (ID Zone Abnormal LDH Control; Beckman Instruments,

Inc.) containing all five isozymes was run with each gel. Gels were dried and

194

scanned on a densitometer (Appraise1M Densitometer, Beckman Instruments, Inc.).
The LDH isozyme profile of a blank (DMEM containing 10% FCS, which was not
exposed to cells) was also evaluated, and these values were subtracted from the
values for cell-conditioned medium to arrive at the LDH isozyme profile of LDH
released from RECs.

Wm. RECs were plated at low density (500 cells per plate) in

100-mm tissue culture plates. Cells were incubated for 3-4 hours to allow for cell
attachment and then were exposed to a single administration of 0, 0.015, 0.15, 1.5,
15 or 150 pM MCTP. Fresh medium (without MCTP) was added on day 7
posttreatment; on day 14, cell colonies on plates were fixed with 10% neutral
buffered formalin and stained with crystal violet (0.5%). Colonies of >50 cells were

counted. Results are expressed as a percentage of the number of colonies in plates

exposed to vehicle.

mm. Two days prior to MCTP treatment, confluent monolayers of
RBCs were harvested by trypsin digestion and plated (approx. 100,000 cells/T-75
flask) in medium containing 0.1 pCi [3H]thymidine (40-60 Ci/mmol; Amersham,
Arlington Heights, IL) / ml to allow for incorporation into cellular DNA. labeling
medium was replaced with fresh medium containing no radioactivity after 24 hours.
Cells were treated one day later (day 0) with 0, 1.5, 15 or 150 pM MCTP and

incubated for 24 or 48 hours.

195
At this time, cells were harvested by scraping, resuspended, and assayed for DNA

crosslinking by the alkaline elution technique of Kohn et al. (1981). Briefly,
suspensions of cells in phosphate buffered saline (pH 7.2) were X-irradiated (100
rads) using a Variable Flux Gamma Irradiator (Model E-0117-M-1; US. Nuclear
Corp., Burbank, CA) to induce DNA single strand breaks. Aliquots of cells were
then loaded onto polyvinyl chloride filters (2.0 pm pore size; Omega Specialty
Instruments, Chelmsford, MA) which were mounted in Swinnex filter assemblies
(Millipore, Bedford MA) affixed with smokestack funnels (Millipore). Cells were
lysed with 10 ml of lysing buffer (2% sodium dodecyl sulfate [SDS] and 25 mM
Na,EDTA; pH 9.7) and rinsed with 10 ml of a solution containing 0.1% SDS and 20
mM ILEDTA (pH 10). Outflow needles were affixed via polyethylene tubing to a
peristaltic pump, and 30 ml elution buffer (0.1% SDS and 20 mM H4EDTA; pH
12.3) was added to the funnels. This buffer was drawn through the filters at a rate
of approximately 0.04 ml/min. The alkaline eluting buffer denatures DNA into
single strands, leaving covalently bonded crosslinks intact. Crosslinked DNA
complexes are excluded or retarded from passing through the filter as compared with
noncrosslinked, single-stranded DNA. Fractions were collected at 90 min. intervals,
and incorporated [3H]thymidine was quantified using a liquid scintillation counter
(Beckman, Model 1.8-3 lSOP). A crosslinking factor was calculated as described by

Kohn and coworkers (1981):

x. = (1~R,/ 1-R,)‘” - 1

196

where X. is the crosslinking factor induced by treatment t, R‘ is the fraction of DNA
from control cells remaining on the filter at 18 ml of elution, and R. is the fraction
of DNA from MCTP-treated cells remaining on the filter at the same elution volume.
Hence, the crosslink factor normalizes the elution rate of DNA from treated cells to
that of a corresponding vehicle-treated control. An increase in crosslinking factor

indicates an increase in total DNA crosslinking.

Wis: Data are presented as means + /- SEM or as means with
standard error of differences (SED) presented for the data set (Steel and Torrie,

'1/2 transformation before

1980). Percentage data were transformed by the arcsin
statistical analysis. The data from the time-course assessment of monolayer
cellularity were analyzed by a completely blocked analysis of variance (AN OVA), and
individual comparisons were made using Tukey’s omega as the past hac test. DNA
crosslinking data were analyzed using Student’s t-test for single comparisons between
treated cells and their vehicle control. Data for the remaining studies were analde

using a random ANOVA, and individual comparisons were made using Tukey’s

omega as the past hac test. The criterion for significance was P<0.05.

Morphology of REC monolayers. RECs showed dose-dependent injury first

evident as increased cell detachment which was noticeable between 1 and 2 days

197

posttreatment. Cell detachment became more pronounced with time, and cells which
remained attached to the dish progressively enlarged. These enlarged cells became
increasingly abnormal with time, developing large cytoplasmic vacuoles and stress
fibers and demonstrating prominent nuclear changes [Figure 26]. However, they
remained viable and were able to exclude trypan blue. By 4 days posttreatment,
small gaps were evident in the monolayers treated with the highest concentration of
MCTP (150 pM). Monolayer integrity appeared to be maintained by cellular
hypertrophy throughout the posttreatment period in the other groups treated with
MCTP.

Monolayer cell density was quantified by counting the cells remaining attached
to the wells after MCTP treatment [Figure 27]. Whereas cell numbers remained
relatively constant or increased somewhat in monolayers treated with 0 or 1.5 pM
MCTP, they progressively decreased in monolayers treated with the higher
concentrations of MCTP, resulting in a large disparity in monolayer cellularity
between treatment groups by day 5.

Cellular release of LDH activity expressed as a percentage of total releasable
activity is shown in Figure 28. RBCs treated with 150 pM MCTP released
significantly more LDH activity than vehicle controls by 2 days posttreatment, and
this persisted throughout the 5 day period. In none of the other treatment groups
was released LDH activity significantly elevated.

The LDH isozyme profile was examined in cells treated with 0 or 150 pM
MCTP. Cultured RBCs primarily contained isozymes 4 and 5 (39% and 58%,

respectiveIY). with small amounts of isozyme 3 present as well (Figure 29A).

3;. _
. 4.1 . o
. 1 s. '00,.V:

 

Photomicrographs of REC monolayers 4 days after a single

Figure 26:

180X

[Phase contrast.

administration of vehicle (A) or 150 pM MCTP (B).

Magnification]

3000
2500
2000

1500

Calls / Well (x1000)

1000

500

I

I

I

I

 

199

REC Number

 

 

 

0 Vehicle

 

o 1.5 an MCTP \ I;

V 15 pM MCTP
V 150 pM MCTP

 

 

 

Time (Days)

Figure 27: Effects of MCTP on REC monolayer cellularity.

Confluent monolayers of RBCs received a single administration of MCTP on
day 0. Values represent means of four replicate studies. [Standard error of
‘ Significantly different from vehicle control (p<0.05)

difference = 158]

 

 

 

 

 

 

 

4° ' LDH Release by RECs
0 Vehicle

3° " a 1.5 pm MCTP
C; V 15 pH MCTP
"5 v 150 pm MCTP
*- at:
N y a:
v
g 20 ~ * 1
3°: all
. I
Q:
I
o }/J
_, .

o

 

 

Time (Days)

Figure 28: Release of lactate dehydrogenase activity from REC monolayers.

MCI'P was administered once at time 0 at the concentrations indicated. Data are
presented as means + /- SEM of four replicate studies. ‘ Significantly different
from vehicle control (p<0.05)

201

 

 

 

 

 

A
Lactate Dehydrogenase
15mm am IDA
3 3 33
4 37 408
5 60 658

 

 

 

 

Tour lU/L . 248

Figure 29: Representative electrophoretic patterns of LDH isozymes in RECs.

Cells were treated with vehicle (A) or 150 pM MCTP at time 0 and were lysed 5
days later with 15 pl of 10% Triton X-100 for analysis.

202

Isozymes 1 and 2were not detected. This pattern did not change qualitatively with
MCTP treatment. Figure 29B shows the electrophoresis pattern of LDH isozymes
in RECs 5 days after treatment with MCTP. Isozymes 4 and 5 still represent 43%
and 55%, respectively, of the total LDH activity. Table 2 shows the pattern of LDH
isozymes found in culture medium over the 5 days posttreatment. Although the
percent of total LDH released from RECs increased with time for the first 3 days
after MCI'P treatment, the profile of LDH isozymes released from the treated cells
into the culture medium was essentially the same as that of the cell lysates. The
conditioned medium contained isozymes 3, 4 and 5, in similar proportions to the cell
lysates.

Colony-fanning efficiency of RECs treated with MCTP is shown in Figure 30.
Concentrations of 0.15 or 1.5 pM MCTP partially inhibited REC proliferation,
whereas higher concentrations resulted in a complete inhibition of cell division. In
plates exposed to concentrations of 15 pM MCI'P or greater, individual cells were
attached to the plate surface, but there was no colony formation. These cells were
greatly enlarged and vacuolated, but they remained viable as determined by their
ability to exclude trypan blue.

The ability of M CT? to cause DNA crasslinkfarmatian in RECs is demonstrated
in Figure 31. There was a concentration-dependent increase in DNA crosslinking at
24 hours posttreatment in RECs exposed to MCTP as compared to vehicle controls.
After 48 hours of continuous exposure to the MCTP-containing medium, there was

a further increase in DNA crosslinking in all treatment groups.

203

Table 2: Lactate Dehydrogenase (LDH) isozyme Profile at
Conditioned Medium from Rat Endothelial Cells

isozyme Fraction (96 of total)‘

 

Imminent: 2e! .3. A. .5. W

Vehicle 1 o o o 1
2 11 56 o 1
3 1o 42 4 1
4 28 66 1 1
5 2 36 62 9

MCTP 1 9 43 14 6
2 5 44 49 23
3 4 41 55 31
4 6 43 42 13
5 6 39 47 20

'lsozymes were separated by electrophoresis in agarose gels and were quantified by densitomitry.
Values were calculated from a single analysis of a representative sample submitted on the day
indicated. The profile of isozymes 3, 4 and 5 found in the supernatant is presented as the percent of

total LDH units Gsozymes 1—5) each represents.

' Cells were treated once at time 0 with 150 pM monocrotaline pyrrole (MCTP) or vehicle.

' This value represents the quantity of LDH released into supernatant expressed as a percent of the

sum of supernatant and cell lysate LDH activity.

204

Rat Pulmonary Endothelium

 

 

moi—- ’§\
6‘ 15
5°, 80 -
...;
“-4 I
I11 0
DD 50 l’
.5
E
L.
L2 40 -
>x
£1
.9.
o 20 ~
0

0, // l- t 1 .—~ ‘ .___
o ’ 0.015 0.15 1.5 15 150

MCTP Concentration QzM)

Figure 30: Effects of MCT P on colony-forming efficiency of RECs.

RECs plated at very low density received a single administration of MCT P on day
0. Colonies were fixed, stained and counted 14 days later. Values represent means
+/- SEM of four replicate studies. “ Significantly different from vehicle control
(p<0.05)

l\)
O
U:

Rat Pulmonary Endothelium

 

 

 

 

 

 

 

3.0-- *
T
I
m 2'5“ o 24 HOUR
2 O 48 HOUR
3 2.0——
Er...
E
7‘ 1.5——
3
O *
‘3‘
:2 1.0—» %
Z *
Q §
0.5--
* >1:
...—0 0
ES J
(10%/1 ] T ]
15 15 150

MCTP CONCENTRATION (,u.M)

Figure 31: Total DNA crosslinking in RECs.

RBCs were treated once with MCTP at time 0. Values represent means + /- SEM
of replicate studies (n=3-4 per group). "' Significantly different from vehicle control
(p<0.05)

206

The pulmonary vascular endothelium, by virtue of its proximity to the flowing
blood and its large surface area, has been proposed as a likely site for binding of
much of the reactive metabolite(s) of MCI“ (Roth and Reindel, 1990). The
consequences of this binding may be very important, both in terms of overt
endothelial cell structural injury and more subtle changes in endothelial function.
Changes in endothelial structural and functional integrity have been noted as
relatively early features of the delayed and progressive pulmonary vascular injury
caused by the administration of MCT or MCTP to rats in viva. Although many of
the reported endothelial cell changes are subtle initially, their onset precedes the
onset of vascular remodeling and increases in pulmonary pressure, suggesting a role
for the altered endothelium in the development and progression of the
pneumotoxicity caused by these compounds (Plestina and Stoner, 1972; Rosenberg
and Rabinovitch, 1988; Allen and Carstens, 1970).

The present studies in cultured rat endothelium demonstrated that MCTP was
directly cytotoxic to these cells. The injury we observed in RECs was delayed and
progressive, reminiscent of that seen in the rat in viva. Injury to monolayers was not
apparent for at least 24 hours after a single administration of MCTP. From this time
forward, there were concentration-dependent cell detachment and release of LDH
which were persistent. The cells which remained in the culture plates became
markedly enlarged, with cytoplasmic vacuolization, prominent stress fibers, and

enlarged nuclei. These cytolytic and hypertrophic changes closely resemble those

207
previously reported in cultured BECs treated with MCTP (Reindel and Roth, 1991;

Reindel et al., 1991) but are more pronounced than those which occur in PECs.

The electrophoretic pattern of LDH isozymes in these cells was investigated
in an effort to identify endothelium as a potential source of LDH detected in the
bronchoalveolar lavage fluid (BALF) and serum of rats treated with MCIP in viva.
The profile of LDH isozymes in RECs did not change with MCTP treatment. The
predominant isozymes detected in lysates of either MCTP- or vehicle-treated cells
were 4 and 5, and these isozymes were released into the medium in the same relative
proportions from injured RECs. N at surprisingly, it appears that inasmuch as LDH
isozymes 4 and 5 make up a substantial part of the LDH released into
bronchoalveolar lavage fluid in viva, other isozymes are detected as well (Schultze
et al., unpublished observations). Thus, the endothelium may be a source of some,
but not all, of the activity noted in BALF and serum of MCTP-treated rats.

In addition to these cytolytic consequences of MCTP treatment, RECs were
also found to be very susceptible to the antiproliferative effects of MCI'P, perhaps
more so than are either BECs or PECs (Reindel et al., 1991). Concentrations of as
little as 0.15 pM MCTP caused a significant decrease in REC colony formation, and
concentrations >1.5 pM MCTP completely inhibited cell division. We have found
MCTP to be a potent antiproliferative agent in all of the cultured cell 'types we have
tested, including BECs, PECs, bovine and porcine smooth muscle cells, Madin-Darby
canine kidney cells and Crandel-Rous feline kidney cells (Reindel er al., 1988;

Reindel and Roth, 1991; Reindel et al., 1991) ;unpublished observations).

208
DNA:DNA and DNAzprotein crosslinking has been demonstrated in the livers

of rats treated with MCT, suggesting that the reactive metabolite generated by the
liver is a potent, bifunctional alkylating agent (Petry et al., 1984). Studies in our
laboratory revealed that MCTP treatment results in persistent DNA crosslinking in
cultured PECs (Wagner et al., 1991; Wagner er al., 1992), and this has also been
demonstrated in RECs in the present study. Alkylation leading to DNA crosslinking
may represent one of the earliest events in the toxicity. It is likely that the DNA
crosslinking which results shortly after MCTP treatment interferes with cell division.
However, the DNA crosslinking does not appear to inhibit DNA synthesis
completely, since MCTP-treated PECs continue to synthesize DNA and increase their
DNA content although they are unable to divide (Hoorn and Roth, 1992).

The cytostatic effects of MCTP appear to be long-lived and may interfere with
the ability of the endothelial monolayer to effect repair in a normal fashion. In the
face of endothelial cell loss in viva, due either to intimal injury or normal cell
turnover, a persistent defect in proliferative repair could result in a gradual loss of
vascular integrity. The progressive vascular leak seen in the lungs of MCTP-treated
rats reflects such a deterioration of the endothelial barrier (Sugita et al., 1983b;
Bruner er al., 1983; Reindel et al., 1990).

In summary, MCTP causes direct injury to rat endothelium in vitna which is
delayed and progressive, supporting the contention that rat pulmonary endothelium
is a likely target for MCTP binding and injury in viva. MCTP-induced injury in
cultured RECs is both cytostatic and cytolytic and suggests several ways in which

endothelial impairment or altered function could contribute to the pulmonary

209

vascular injury in this species when treated in viva. Impaired cell proliferation,
perhaps as a consequence of DNA crosslinking, would leave the pulmonary
endothelium vulnerable to any action resulting in injury or stimuli for increased cell
turnover. This aspect of MCTP cytotoxicity, combined with the cytolytic nature of
the injury, could contribute to the vascular alterations which occur in the rat treated

with MCT in viva.

Chapter VI

THE EFFECTS OF MITOMYCIN C ON CULTURED ENDOTHELIUM:

COMPARISON WITH MONOCROTALINE PYRROLE-INDUCED

CYTOTOXICITY

210

211
Summary

EC alterations are proposed to underlie the delayed and progressive development of
pneumotoxicity which is seen after exposure to MCT (P), but the mechanism by which
this compound initiates injury remains unknown. MCTP injures ECs treated in vitro,
resulting in an inhibition of proliferation at low concentrations and marked cytolytic
and morphologic alterations at higher concentrations. MCTP is a bifunctional
alkylating agent, and DNA crosslinking has been observed in cultured ECs treated
in vitro with concentrations of MCTP that are cytostatic but which cause little
cytolysis. DNA crosslinking, by inhibiting EC proliferation and limiting monolayer
repair capability, may engender the delayed and progressive cytotoxic response of
ECs monolayers to MCTP administration. Mitomycin C (MMC) is a bifunctional
alkylating agent used in cancer chemotherapy. Like MCTP, reactive metabolites of
MMC can also crosslink DNA, and the cytotoxicity of this compound has been
attributed to this capacity as well as to other capacities such as redox cycling.
Treatment of cultured EC monolayers with MMC resulted in a pattern of injury
reminiscent of MCTP-induced EC injury, including both cytolytic and cytostatic
effects. MMC treatment also resulted in DNA crosslinking at concentrations which
inhibited proliferation but caused limited overt cytotoxicity, supporting an association
between DNA crosslinking and cytostasis. MMC was much more potent than MCTP
in producing EC effects. Comparable morphologic alterations were evident at 1 pM
MMC and 150 pM MCTP, and 0.01 pM MMC and 1.5 pM MCTP inhibited

proliferation to a similar extent. Nevertheless, the similarity of the EC reaction to

212

bifunctional alkylating agents of different classes suggests that DNA crosslinking in
ECs results in a characteristic response of this cell type to a loss of monolayer

integrity.

0 .
HzN. HN O
N?) if,

     

\ N

Dehydromonocrotaiine

Figure 32: Structures of mitomycin C, the reactive metabolite, 7-amino-1,2-
aziridinomitosene, and dehydromonocrotaline (MCTP).

213

Many of the pyrrolic metabolites of PZAs, as well as a number of potent
cancer chemotherapeutic agents, are bifunctional electrophiles. The antiproliferative
and cytotoxic efiects of these compounds have generally been attributed to their
common ability to crosslink DNA and other cellular macromolecules, as related
monofunctional alkylating agents, which do not form crosslinks, are generally much
less cytotoxic (Bull et al., 1968; Culvenor et al., 1969; Zwelling et al., 1979; Wheeler,
1975; Bodell er al., 1985; Mattocks and Legg, 1980).

DNA crosslinking occurs as a relatively early event in the course of MCT(P)
toxicity, both in the liver after MCT treatment in viva and in cultured endothelial
cells exposed to MCTP in vitro (Petry et al., 1984; Wagner et al., 1992). Cell
proliferation is inhibited in both systems, and surviving cells become markedly
enlarged, with pronounced nuclear alterations (Culvenor et al., 1969; J ago, 1969;
Reindel er al., 1991). It is hypothesized that similar binding to DNA in the
endothelium of the lung vasculature may initiate a cascade of cellular alterations that
contribute to the development of MCT(P) pneumotoxicity in viva (Roth et al., 1989).

Mitomycin C (MMC) is an antitumor antibiotic used in the treatment of
neoplasia of the gastrointestinal tract or breast (Crooke and Bradner, 1976). Like
MCT, MMC is bioactivated to a potent, bifunctional alkylating species that is capable
of crosslinking DNA, and much of the cytotoxic potential of this agent has been
attributed to this capacity (Dulhanty et al., 1989; Iyer and Szybalski, 1963).

Reductively activated MMC bears a structural similarity to the PZAs, but has an

214
added quinone function (Weidner et al., 1990) [Figure 32]. Interestingly, intravenous

administration of MMC occasionally has been reported to cause delayed-onset,
progressive pulmonary vascular injury which is characterized by EC enlargement,
vascular leak, and the development of pulmonary hypertension (Jolivet et al., 1983;
Waldhom et al., 1984; Cantrell er al., 1985 ; Chang et al., 1986; McCarthy and Staats,
1986). Structural analogues of MMC have also been shown to cause a delayed
pneumotoxicity in rats, with vascular alterations which are reminiscent of those
caused by MCI‘ (P). These alterations include necrotizing arteritis and medial
hypertrophy, with the development of pulmonary edema and right ventricular
dilatiation and hypertrophy. Endothelial changes were not mentioned in this study,
and the pulmonary effects were considered, perhaps erroneously, to be a
consequence of cardiotoxicity (Bregman et al., 1989).

To begin to understand the importance of DNA crosslinking to the inhibition
of proliferation and the progressive development of cytotoxic alterations which occur
in ECs treated with MCTP, the effects of MMC on cultured EC structure and
function were examined. Although MMC has been used as an experimental tool to
inhibit EC proliferation, the antimitotic effects of this compound have not been
directly correlated with DNA crosslinking in ECs, and the overall response of the
endothelium to MMC treatment has not been well characterized. The identification
of similarities in the response of ECs to bifunctional alkylating agents of different
classes may help to define a role for DNA crosslinking in the early cellular

alterations that lead to the delayed development of chronic pulmonary vascular

injury.

215

W. Porcine pulmonary artery endothelial cells (PECs)
were isolated and characterized as described previously in Chapter II. Rat
pulmonary endothelial cells (RECs) were generously provided by Dr. James Varani
(Dept. of Pathology, University of Michigan, Ann Arbor, Michigan). PECs and
RECs were maintained and subcultured as described previously in Chapters 11 and
IV, respectively. PECs used in these studies were between passages 4 and 10; RBCs
were at passages 27 and 28. Unless otherwise noted, culture medium was replaced
with fresh Opti-MEM (5% FCS, 1% Ab/Am) or DMEM (10% FCS, 1% Ab/Am)

every other day.

WM. MMC (Sigma) was dissolved in sterile water at a concentration
of 100 pM, and all dilutions of MMC were made with sterile water. A 5 pl volume
of MMC solution or vehicle (0 pM MMC) per milliliter of medium was used in all
studies to achieve the nominal concentrations of MMC (0.001-10 pM) used in these

studies.

WWW: PECs 0r RECS were Plated

into 12-well tissue culture clusters and allowed to proliferate until monolayers
reached confluence. At this time (day 0), fresh medium was added and monolayers

were exposed to a single administration of 0, 0.1, 1 or 10 pM MMC. At 4 hrs and

216
on days 1, 2, 3, 5, 7 and 10 (PECs only) posttreatment, cells remaining in the

monolayers were examined and counted as described previously in Chapter 11.

WW: PECs and RECs were plated and treated as

described above. At the indicated times posttreatment, one lZ-well plate of each cell

type was used for determination of LDH activity as described in Chapter V.

W: PECs and RECs were plated at low density (1000

cells/plate) in 100 mm tissue culture plates. Cells were incubated for 3-4 hr to allow
for cell attachment and then were exposed to a single administration of 0, 0.001, 0.01,
0.1, 1 or 10 pM MMC. Fresh medium (without MMC) was added on day 7
posttreatment. On day 14, cell colonies were fixed, stained and counted as described

in Chapter V.

W. These studies were carried out (on PECs only) as described
previously in Chapter V. PECs were treated with 0, 0.01, 0.1 or 1 pM MMC, and the

alkaline elution assay was performed at 24 and 48 hr posttreatment.

mm: Data are presented as means + /- SEM. Percentage data were
transformed by the arcsin‘l/2 transformation before statistical analysis. DNA
crosslinking data were analyzed using Student’s t-test for single comparisons between

treated cells and their vehicle controls. Data for the remaining studies were analyzed

217
using a random AN OVA, and individual comparisons were made using Tukey’s

omega as the past hac test. The criterion for significance was P<0.05.

Morphology of cell monolayers. Both PECs and RECs showed concentration-
dependent injury first evident as increased cell detachment by 24 hr posttreatment.
Cell detachment was more pronounced in RECs but progressed with time in both
cell types. As monolayers became less cellular, the remaining cells progressively
enlarged and became irregular in shape. The concentration-dependent effects of
MMC on PEC and REC monolayers at 5 days posttreatment are shown in Figures
33 and 34. By this time, few cells remained in the monolayers treated with 10 pM
MMC, and gaps were evident in the monolayers treated with 1 pM MMC.
Remaining PECs became spindle-shaped, whereas RECs were markedly enlarged and
abnormal in appearance with cytoplasmic vacuoles and enlarged nuclei. All adherent
cells in these dose groups remained viable and were able to exclude trypan blue.
PEC and REC monolayers treated with 0.1 pM MMC showed only minor
morphologic alterations, but they appeared less dense than controls at this time;
there was little cell detachment evident in this treatment group.

Monolayer cell density was quantified by counting the cells remaining attached
to the wells after MMC treatment [PEC: Figure 35; RECs: Figure 36]. PEC number

increased slightly over the 10 day period of observation in the control group. Cell

218

 

Figure 33: Photomicrographs of PEC monolayers 5 days after a single
administration of 0 (A), 0.1 (B), l (C) or 10 (D) pM MMC. [Phase contrast].

219

 

Figure 34: Photomicrographs of REC monolayers 5 days after a single
administration of 0 (A), 0.1 (B), l (C) or 10 (D) pM MMC. [Phase contrast].

220

PEC Number

1000 r-

 

 

 

 

750
A
O
O
O
1‘,
E 500
3
\
2
‘6
0 “xi
1a
250 _ 0 Vehicle
0 0.1 pM MMC
v 1 pM MMC

v 10 [4M MMC

 

 

 

 

 

I“\\* 3 t
o r t "“J “r-—-V . L
O 2 4 6 8 10
Time (Days)

Figure 35: Effects of MMC on PEC monolayer cellularity.

PEC monolayers received a single administration of MMC at time 0. Values
represent means +/- SEM of 3 replicate studies. " Significantly different from
vehicle control (p<0.05) '

IQ
IQ
e—l

 

 

 

  

 

 

 

 

 

 

 

REC N umber
4000
3500
A 3000
o
o
o
3', 2500
v Vehicle
; 2000 0.1 H” We
1 M MMC
\ #-
m . 10 #M We
= 1500 - \
8 a...”
a:
\\ at a“.
1000 _ , i
0.... *
...”...anoouu..uou u *
500 r- \.“ ‘17 w w?
!\ 1' ,, ...
o l J \Y I y L— L
o 1 2 3 4 5 6 7

Time (Days)

Figure 36: Effects of MMC on REC monolayer cellularity.

REC monolayers received a single administration of MMC at time 0. Values
represent means +/- SEM of 3 replicate studies. " Significantly different from
vehicle control (p < 0.05)

222

number remained relatively constant in PECs treated with 0.1 pM MMC and
declined gradually in monolayers treated with 1 pM MMC. PEC monolayers treated
with 10 pM MMC rapidly decreased in cellularity, with the number of adherent cells
less than 5% of controls by 5 to 7 days posttreatment.

REC monolayer cellularity increased dramatically in both the control group
and the lowest MMC treatment group (0.1 pM) during the first two days of the
experiment but remained constant thereafter. The rate at which cell number
increased was less in the cells treated with 0.1 pM MMC. By 7 days posttreatment,
these dense REC monolayers began to detach spontaneously from the culture plates
in sheets irrespective of treatment, making evaluation at 10 days posttreatment
impossible. In monolayers treated with higher concentrations of MMC, REC number
decreased progressively with time. REC monolayers treated with 10 pM MMC had
few adherent cells remaining after 3 days posttreatment.

Release of LDH activity expressed as a percentage of total releasable activity
is illustrated in Figures 37 and 38 (PECs and RECs, respectively). PEC LDH release
was minimal for the first 24 hours posttreatment. After this time, LDH activity in
the medium from PEC monolayers treated with 10 pM MMC increased rapidly until
3 days posttreatment, then began to decline by day 5 when few cells remained in the
monolayers. LDH release was minimal in other treatment groups; monolayers
treated with 1 pM MMC showed a slight increase in release of LDH activity, but
only at 7 days posttreatment.

IDH release from REC monolayers was minimal at 4 hr posttreatment in all

groups but increased dramatically during the next three days in cells treated with 10

223

 

  
  

‘00 ' LDH Release by PECs
80 b 0 Vehicle
' e 0.1 p.14 MMC
V 1 pM MMC
v 10 pM MMC

 

 

 

LDH Release (% Toial)

 

 

 

 

 

 

Time (Days)

Figure 37: Release of LDH activity from PEC monolayers.

MMC was administered once at time 0 at the concentrations indicated. Data are

presented as means + /- SEM of three replicate studies. ’ Significantly different
from vehicle controls p<0.05)

224

 
  
 
 
  

 

 

 

 

100 -
LDH Release by RBCs
T
80 h I\ f
O
'- 60 -
5°, 0 Vehicle
“‘0’ O 0.1 “M MMC
E v 1 pM MMC
(55” 40 l. v 10 1.4M MMC
I
Q
_J .

 

 

 

Time (Days)

Figure 38: Release of LDH activity from REC monolayers.

MMC was administered once at time 0 at the concentrations indicated. Data are
presented as means + /- SEM of three replicate studies. “ Significantly different
from vehicle controls p<0.05)

225

pM MMC. The increase in LDH release was more gradual and progressive from
cells treated with 1 pM MMC, peaking at about 25 % by 3 days posttreatment. LDH
release was minimal from REC monolayers treated with 0.1 pM MMC until 7 days
posttreatment; by this time, LDH release was increased from control monolayers as
well.

Colony-farming efiiciency of PECs treated with MMC is shown in Figure 39.
MMC at a concentration of 0.01 pM inhibited PEC proliferation by 50%, whereas
higher concentrations resulted in a complete inhibition of cell proliferation. At 0.1
pM MMC, enlarged, abnormal cells could be seen as individual cells or small groups
attached to the plate surface. Very few cells remained on the plates treated with 1
pM MMC, and no cells were evident in the highest treatment group.

MMC treatment inhibited REC proliferation in a similar manner [Figure 40].
Concentrations of 0.01 pM decreased colony formation by 33%, and higher
concentrations completely inhibited cell proliferation.

DNA crosslinking in PECs in response to MMC treatment is shown in Figure
41. A concentration-dependent increase in total DNA crosslinking was evident at 24
hr posttreatment as compared to controls, and this crosslinking persisted through 48

hours.

226

Porcine Pulmonary Artery Endothelium

1001—

 

 

 

8

3x; 80 -
ӣ5

E 60 -

E

L.

O

“- 40 -

>~

C

O

'6

° 20 -

o, // r . 1 . .—
0 ’0001 0.01 0.1 1 10

MMC Concentration (p M)

Figure 39: Effects of MMC on colony-forming efficiency of PECs.

PECs plated at very low density received a single administration of MMC on day 0.
Colonies were fixed, stained and counted 14 days later. Values represent means + /-

SEM of three replicate studies. ‘ Significantly different from vehicle control
(p<0.05)

227

Ra’t Pulmonary Endoihelium

tooi— —-§

 

 

 

8
5 80 i-
:3
c» 60 -
.5
E
L.
O
“- 40 -
g .
C
2
0
° 20 -
t t
O // l . l ‘ ._____
0 ”0.001 0.01 0.1 10

1
MMC Concentration (p. M)

Figure 40: Effects of MMC on colony-forming efficiency of RBCs.

RBCs plated at very low density received a single administration of MMC on day 0.
Colonies were fixed, stained and counted 14 days later. Values represent means + /-

SPEN‘I) :5) three replicate studies. ‘ Significantly different from vehicle control
< .

Porcine Pulmonary Artery Endothelium

 

 

 

 

 

 

1.001-
O
m a 24 HOUR
g 075-- o 48 HOUR
O
<
It.
:4
Z
I? 0.504—
VJ
(I)
O
6 4:
g 4
c: 0.25» .
*/§
4: —/
3|:
2|:
O-OOT‘?” I I l I
‘0.001 0.01 0.1 t

MMC CONCENTRATION (pM)

Figure 41: Total DNA crosslinking in PECs.

PECs were treated once with MMC at time 0. Values represent means + /- SEM
of replicate studies (11 = 4-5 per group). " Significantly different from vehicle
control (p<0.05)

229

Treatment with MCTP causes injury to cultured ECs derived from a number
of species (Reindel et al., 1991) , Chapter V). PECs demonstrate subtle morphologic
alterations in response to treatment with high concentrations of MCTP; they become
spindle-shaped in appearance and demonstrate some nuclear enlargement.
However,they are relatively insensitive to the cytolytic effects of this compound, and
demonstrate only limited cell detachment and LDH release (Reindel er al., 1991).
RECs are much more sensitive to the cytolytic effects of MCTP and respond with
marked increases in cell detachment, increased LDH release and dramatic
morphologic changes in the surviving cells (Chapter V). However, MCI'P causes a
similar inhibition of proliferation in both PECs and RECs at concentrations which
cause little or no other evidence of cytotoxicity.

MCTP treatment also results in persistent DNA crosslinking in cultured PECs
and RECs. Significant crosslinking is evident after treatment with low concentrations
of MCTP that cause limited cytolytic injury but significant cytostasis (Wagner et al.,
1992; Chapter V). These responses suggest that DNAzDNA or DNAzprotein
crosslinking may be responsible for the antimitotic effects of MCTP but is probably
not directly responsible for its cytolytic effects.

As with MCTP, treatment of cultured PECs and RECs with the bifunctional
alkylating agent, MMC, resulted in delayed and progressive injury that was
concentration-dependent. MMC caused cytostasis at lower concentrations, and

increased cell detachment and release of LDH were evident at higher concentrations.

230

DNA crosslinking occurred in PECs at concentrations that caused marked inhibition
of proliferation but only modest cytolysis. Accordingly, MMC causes an EC response
that is quite similar to that caused by MCTP.

The cytolytic alterations in ECs developed somewhat more rapidly in response
to MMC than to MCTP, and both PECs and RECs were sensitive to these effects.
This may be a consequence of the concentrations of MMC used in this study.
Nevertheless, the endothelial response (i.e., cell spreading, enlargement and shape
change) to a loss of monolayer integrity was very consistent within each species to
both MCTP and MMC.

Although the effects of MMC on cultured endothelium are similar to those
reported for MCTP-treated endothelium, the concentrations required to cause these
effects are quite different. Concentrations of 1 pM MMC caused changes in
monolayer cellularity and morphology comparable to concentrations of 150 pM
MCTP or greater, and the DNA crosslink factor associated with MMC-induced
cytostasis was much lower than that associated with an inhibition of proliferation by
MCTP {Reindel et al., 1991; Chapter V). The amount of bioactive MMC metabolite
may, in fact, be much lower than the nominal concentrations indicated, thus MMC
appears to be a much more potent EC toxicant than MCTP.

MMC, like MCT, is metabolically activated to one or more bifunctional
alkylating species; covalent binding to DNA and other cellular macromolecules
occurs only after reduction. Under aerobic conditions, NADPHzcytochrome P450
reductase or xanthine oxidase are thought to be responsible for bioactivation (Keyes

et al., 1984; Pan et al., 1986). However, under aerobic conditions, reduction by DT-

231
diaphorase is thought to predominate, and inhibition of this enzyme by dicumerol

increases resistance to the toxicity of MMC (Ernster, 1967; Keyes et al., 1985;
Dulhanty et al., 1989). Endothelial DT-diaphorase activity has not been identified
per se in the literature; however, the results of these studies and those of others
(Duperray et al., 1988) in cultured ECs would suggest that these cells are able to
bioactivate MMC to a cytotoxic metabolite that crosslinks DNA.

MMC is thought to exert its cytotoxic effects by covalent binding of a
reductively activated metabolite to DNA or other cellular macromolecules and/or
by the formation of reactive oxygen species as a result of redox cycling (Dulhanty et
al., 1989; Monks et al., 1992; Iyer and Szybalski, 1963). DNAzDNA or DNAzprotein
crosslink formation by the quinone methide or aziridinornitosene metabolites of
MMC, or by reactive MCTP, may be responsible for their common antiproliferative
effects, resulting in an endothelial monolayer that is defective in its capacity to effect
repair. MMC and MCTP demonstrate a similar sequence preference for DNA
binding, resulting in crosslinking between deoxyguanosine residues on complementary
strands at 5’-CG duplex sequences (Niwa et al., 1991; Weidner et al., 1990). MC
and MCTP also have a similar distance between electrophilic centers, and these
features may result in crosslinking at similar sites in EC DNA and consequently
similar effects.

Cell death may ensue as a delayed response to the DNA damage, but there
may be other, dissimilar toxic consequences of exposure to MMC or MCI'P which
are responsible for the cytolytic effects of these compounds, such as alteration of

calcium homeostasis, alkylation of cellular proteins or redox cycling and resultant

232

oxidative stress. For example, the quinone function of MMC, which MCTP lacks,
may contribute to the direct cytotoxic effects of this compound (Monks et al., 1992).

Cell detachment or lysis results in the formation of gaps in the EC monolayer,
providing a stimulus for cell migration, spreading and proliferation. Although MMC
and MCTP may cause a loss of monolayer integrity by similar or dissimilar
mechanisms, they similarly impair the ability of ECs to respond to a stimulus for
proliferation. Thus, a critical consequence of persistent DNA crosslinking in the
endothelium may be an altered capacity for monolayer repair, which leaves the
vascular wall vulnerable to endothelial disruption of any sort, including cell death in
response to disease, toxic insult or aging.

Lung vascular injury is not a common sequel to chemotherapy with MMC in
viva, but a number of cases have been reported after intravenous administration of
this compound. EC alterations, vascular leak, arterial medial hypertrophy, and the
development of pulmonary hypertension and right ventricular enlargement are
characteristic of the pulmonary vascular damage which can result from treatment
with MMC. Similar changes have been reported in the lungs of animals exposed to
MCT (P). The pneumotoxic effects of both MMC and MCTP typically are delayed
in onset and progressive in nature (McCarthy and Staats, 1986; Chang et al., 1986;
Jolivet et al., 1983; White and Roth, 1989). The formation of fibrin thrombi in
capillaries, increased vascular permeability and other evidence of endothelial
dysfunction have been reported after administration of both compounds, and the
development of hemolytic anemia after MMC treatment has been attributed to

changes in vascular EC function (McCarthy and Stats, 1986; Jolivet et al., 1990;

233
Valdivia et al., 1967a,b; Merkow and Kleinerman, 1966b). Thus, not only are the

effects of MMC and MCTP on cultured endothelium very similar, but the
consequences of exposure of the pulmonary vasculature to these agents in viva are
similar as well.

In conclusion, the nature of the endothelial response to MMC demonstrated
by these studies in vitro and the spectrum of reported pulmonary vascular alterations
as a result of treatment with this agent in viva bear a marked similarity to those
reported previously for MCTP. These compounds are derived from different sources,
are activated by different mechanisms, and possess some distinct chemical properties;
nevertheless, they share a common pyrrolizidine nucleus and both are bioactivated
to bifunctional alkylating species. Their shared ability to crosslink DNA (and other
macromolecules) may underlie the inhibition of proliferation seen after
administration of both of these agents and may contribute to the unusual, delayed
and progressive endothelial response to MMC and MCTP treatment in vitro. Similar
effects on the capacity for proliferation and repair in the endothelium of pulmonary
vessels in viva may contribute to the evolution of chronic lung vascular disease caused

by these, and perhaps other, bifunctional alkylating agents.

SUMMARY AND CONCLUSIONS

234

235

I began these studies hoping to determine whether there existed changes in
EC function in response to MCTP treatment that were consistent with an endothelial
contribution to the development of MCTP-induced pulmonary vascular alterations.
MCTP treatment does affect basic cellular function of ECs, increasing their synthesis
of DNA, RNA and protein in a delayed and progressive manner. This is an
important point in understanding the endothelial response to injury. In the study of
cancer chemotherapeutic agents, inhibition of proliferation is typically considered to
be synonymous with cell death (Rosenblum et al., 1978). Our work with MCTP
suggested that these conditions should not be viewed as equivalent as they apply to
chronic lung injury; cells may be inhibited in their ability to divide yet remain viable
and continue to function, though perhaps in an altered manner (Hoom and Roth,
1992). Although the specific profile of proteins produced by MCTP-treated cells
remains unknown, the fact that these cells continue to function and, in fact,
synthesize increased amounts of RNA and protein after treatment suggests that ECs
could contribute to the vascular alterations seen after MCT (P) treatment.

My efforts to define some of the specific EC functional alterations in response
to MCTP met with limited success. ACE activity decreased in response to MCTP
treatment, but only at high concentrations. In addition, the delayed nature of the
decrease and its correlation with the extent of overt cytotoxicity suggested that
changes in ACE activity occur as part of the EC response to injury, not as a direct
efiect of MCTP on the enzyme itself. Although this tells us a little more about the
EC response to injury, surface ACE activity of ECs does not appear to be directly

affected by low concentrations of MCTP. These findings suggest that changes in

236

ACE activity in viva, if present, likely occur as secondary changes in response to an
altered physical or chemical environment for the enzyme.

The cultured EC system did not lend itself well to investigation of the role of
MCTP-induced EC alterations in regulation of SMC growth and proliferation. I was
unable to demonstrate increased release of mesenchymal cell growth factors from
MCTP-treated ECs. It is possible that the system was incapable of upregulating the
production and release of mitogens. This was difficult to determine definitively, as
the basal release of EDGFs was quite high and the response to thrombin stimulation
was inconsistent. However, after many modifications of the system and exhaustive
repetition of these studies, it seems evident that MCTP treatment does not increase
markedly the release of growth factors from constitutively active cells.

There is still a case to be made for a contribution of EC-derived mitogens in
the vascular remodeling process associated with MCT (P) treatment in viva.
Quiescent endothelium in viva could respond to injury quite differently with regard
to growth factor synthesis than the model in vitro used in these studies. The relevant
question, "Does MCTP treatment increase the expression, synthesis or release of
growth factors from quiescent endothelium in viva?" may be quite different from the
question we actually asked, "Does MCTP augment the release of growth factors from
cells in vitro that constitutively express these factors at high levels?". In retrospect,
the cultured EC system at its present state of development is an inadequate model
to address the former question. Examination of gene expression in the lung

vasculature in situ may be a more appropriate way to investigate this hypothesis.

237

From these investigations with cultured endothelium treated with MCTP, we
have learned a great deal about the endothelial response to injury in vitro. Damage
to endothelial monolayers as a consequence of treatment with certain other EC
toxicants (e.g., endotoxin, cyclosporin A) is typically more rapid in onset and is less
persistent than the damage that results from exposure to MCTP. The limited
proliferative repair capacity and the dramatic hypertrophic response seen with MCTP
treatment does not occur with these more acutely toxic compounds, suggesting that
delayed and progressive EC injury may represent a response by endothelium to toxic
damage that is unique to certain toxicants.

The EC alterations observed after the administration of MCTP in vitro appear
to represent a summation of responses to a number of distinct events which occur as
a result of this treatment. A relatively early event is the binding of MCTP to cellular
nucleophiles such as DNA. The highly reactive nature of MCTP and its instability
in aqueous solution suggest that initial binding to ECs occurs shortly after
administration of the compound. Since DNA crosslinking increases with time during
the 48 hr period after a single administration of MCTP, this initial binding may be
transient, with MCTP shifting from nucleophile to nucleophile within the cell.
Bifunctioual alkylating agents generally bind at one reactive center initially; if
conditions are favorable and the initial site of alkylation is not repaired, the other
nucleophilic center also binds, forming a crosslink. Thus, stable DNA crosslinks may
take some time to form in MCTP-treated ECs. By 48 hr posttreatment with MCTP,

crosslinking appears to be maximal; this suggests that either 1) final formation of

238

stable crosslinks has occurred by this time, or 2) the rate of crosslink formation is in
equilibrium with the rate of crosslink removal.

Although DNA crosslinking is frequently associated with an inhibition of cell
proliferation, it is difficult to determine conclusively whether a cause/effect
relationship exists. linking of complementary strands of DNA should hinder strand
separation and progression of the replication fork, thus physically impeding the
mitotic process. In support of this theory, most crosslinking agents are cycle active
and arrest cells late in M phase, during G1 phase, or early in S phase; cells are least
sensitive during late S phase (Meyn and Murray, 1984; Murray and Meyn, 1986).
Bifunctioual alkylating agents are typically more cytostatic than their monofunctional
counterparts (Mattocks and Driver, 1983; Mattocks and Legg, 1980; Tokuda and
Bodell, 1987; Bodell, 1990), suggesting that a bifunctional event is critical; however,
crosslinking of other vital cellular proteins may occur coincident with DNA
crosslinking, and these may have antiproliferative effects, too. Nevertheless, DNA
crosslinking remains a plausible mechanism underlying the antimitotic effects of such
agents.

The ability of ECs in a blood vessel to respond to a loss of monolayer
integrity by spreading, migrating into the defect and proliferating is probably critical
to the repair capacity of this tissue. Unrepaired disruption of the intimal surface of
the blood vessel or capillary wall can result in increased vascular permeability,
increased thrombogenicity of the intimal surface, and altered vasoactivity, perhaps
with severe consequences to the surrounding tissues. Migration and spreading of ECs

may be sufficient to maintain vascular integrity temporarily in the face of limited,

239

discrete monolayer disruption; however, proliferation is vital in the repair of larger
defects or more diffuse intimal damage.

MCTP-treated ECs are able to migrate and spread but are persistently
impaired in their ability to divide. DNA crosslinking has frequently been associated
with cytostasis. This primary defect in ECs treated with MCTP (e.g., crosslinking of
DNA or other critical macromolecules), with the consequent inhibition of
proliferation, may account for the characteristic pattern of injury seen in vitro. PECs
treated with a single administration of MCTP exhibit little cell detachment or lysis
but are inhibited in their ability to proliferate. Monolayers of treated PECs
demonstrate only subtle morphologic alterations, and synthesize DNA, RNA and
protein at slightly elevated levels. However, if the monolayer is artificially disrupted,
or if the cells are treated before they reach confluence, they enlarge dramatically,
become vacuolated, have marked nuclear changes, and synthesize substantially larger
amounts of DNA, RNA and protein. Thus, a stimulus for proliferation appears to
unmask the effects of persistent crosslinking in these cells.

There are differences in the response of ECs derived from different species
and sources to MCTP treatment. In the case of BECs or RECs, treatment with
MCTP causes significant cell detachment and cytolysis. This may occur as a delayed
response to the DNA damage or may be due to other alterations which occur as a
result of MCTP treatment. In any case, the cytolytic response results in a loss of
monolayer integrity and provides a stimulus for cell spreading and proliferation.

Surviving cells are unable to divide, but they do spread and enlarge greatly. Thus,

240
although MCTP causes cytolytic damage in these cells which is not evident in PECs,

the basic response to injury is the same in all three cell types.

Our studies with MMC suggest that DNA crosslinking and the resultant
inhibition of proliferation may be the common denominator in the atypical EC
response to injury. MMC and MCTP may cause overt cytolytic changes by very
difl'erent mechanisms, but they share a common ability to crosslink DNA and the
response of MMC- or MCTP-treated endothelial monolayers to a loss of monolayer
integrity is very similar. Further investigation with this comparative model should
provide many answers with respect to the role of DNA crosslinking in the inhibition
of proliferation and in the development of other functional alterations in ECs, such
as changes in DNA, RNA and protein synthesis.

In addition to their ability to inhibit cell proliferation, DNA alkylating agents
have the potential to influence gene expression. Gene amplification may occur as
a result of endoreduplication of the genome when there is an arrest of proliferation
but continued DNA synthesis. MCTP-treated PECs continue to synthesize DNA and
demonstrate an increased cellular DNA content (Chapter III). Some of the
increased RNA and protein synthesis which is evident in MCTP-treated ECs may
reflect the expression of genes amplified in this manner, and this possibility requires
further investigation. In addition, the act of binding to DNA necessitates repair, and
increased rates of repair may lead to defective repair and alterations in DNA
sequence, folding and transcription. Furthermore, the very nature of interstrand
DNA crosslinks may increase the likelihood of inaccurate repair. Future studies

which identify specific alterations in EC function in response to MCTP and MMC

241

may point to areas of the genome on which to focus a search for critical DNA
alterations.

In order to demonstrate a cause-and-effect relationship for DNA crosslinking
and various EC responses, it will be necessary to identify crosslink formation after
administration of relevant concentrations of MCTP or MMC. Crosslinking should
occur at critical posttreatment intervals which precede or coincide with the effects
of interest, and it should persist . Conditions which enhance crosslinking should
enhance the cytotoxic effects. Ideally, we should also be able to inhibit crosslink
formation and prevent occurrence of the EC effects. Although several of these
criteria have been met by MCTP and MMC, there is still much work to be done
before DNA crosslinking is can be established as a mechanism for the chronic
pulmonary vascular injury caused by these compounds. This work may be
confounded by the fact that we still know very little about the extent of crosslinking
required for cytotoxic injury; and we have yet to confirm that the relevant crosslink
formation is that which occurs in DNA. However, comparison of the EC response
to MCTP treatment with the response to MMC and other, unrelated, bifunctional
agents will allow us to begin to investigate these relationships.

Eventually, the results of studies performed in cultured cells will need to be
confirmed in studies performed in viva. DNA crosslinking in the pulmonary
endothelium in situ has not been demonstrated; it may be difficult to detect small but
relevant levels of crosslink formation in this tissue in viva. Parallel studies in viva

utilizing MCTP and MMC (and perhaps other bifunctional agents of different

242
classes) may provide information about the relevance of DNA crosslinldng to

pulmonary vascular alterations in viva.

A final area addressed by the work presented in this dissertation involves the
importance of species differences in the endothelial response to MCTP. In a
previous study, the responses of BECs and PECs to MCTP treatment were compared
(Reindel et aL, 1991). We found that BECs, like RECs, were quite sensitive to the
cytolytic effects of MCTP. PECs appeared to be more resistent to these effects,
evidencing less cell detachment and release of LDH and only subtle changes in cell
size and morphology. However, PECs and BECs were equally sensitive to the
cytostatic effects of MCTP. From these studies, we concluded that species
differences in sensitivity to reactive pyrrolic metabolites exist and might in part
explain the species differences observed in organ pathology after exposure to MCI‘
in viva.

Similar studies utilizing cultured endothelium from a highly relevant species
such as the rat have not been performed previously. The results presented here raise
new questions with respect to species differences in endothelial sensitivity to MCTP
and how these differences may relate to the patterns of toxicity observed in viva. For
example, cattle typically develop venoocclusive lesions of the liver shortly after
exposure to MCT (Sanders er al., 1936) rather than the chronic, progressive
pulmonary vascular lesions which are more typically reported in the rat (Chesney and
Allen, 1973d; Meyrick and Reid, 1979; Reindel et al., 1990) or the pig (Peckham et

al., 1974). The former lesions might be attributable to a rapid deterioration of the

243

hepatic vasculature after interaction of the reactive metabolite(s) of MCT with
sensitive bovine endothelial cells. On the other hand, in a species with endothelium
that is less susceptible to cytolytic injury (e.g., porcine), changes in the hepatic
vasculature might be more subtle, allowing time for the consequences of extensive
but delayed, progressive pulmonary vascular cell injury to be expressed (Reindel et
al., 1991). Assuming this scenario, one might predict that the endothelium of the rat
would resemble that of the pig in its sensitivity to MCT. However, this does not
appear to be the case; in fact, RECs more closely resemble BECs in their response
to injury by MCTP in vitro.

Presumably, species differences in ability to generate the necessary toxic
metabolite(s) in sufficient quantity from the parent compound and differences in
distribution of the reactive metabolites once they are formed may influence the
pattern of organ toxicity seen after exposure to MCT (Chesney et al., 1974a; Cheeke
and Pierson-Goeger, 1983; Mattocks, 1968). In large animals, susceptibility of the
endothelium to injury by MCTP may be quite important in determining the ultimate
effects of a broadly distributed dose of toxicant. Bovine endothelium is very sensitive
to the cytolytic effects of MCTP, and the deterioration of the liver in this species
after MCT ingestion may reflect endothelial injury at the site of metabolite
generation. Delayed pneumotoxic injury is more commonly seen in the pig, perhaps
because endothelial injury in both the liver and the lung is more subtle, allowing time
for pulmonary consequences of inhibited EC proliferation to develop. In addition,
the relatively longer time required for an unstable hepatic metabolite to reach the

lungs via the circulation in larger animals may reduce pulmonary susceptibility.

244

Given the short distance from the liver to the lung and the extremely rapid blood
circulation time in a small animal like the rat, a larger proportion of metabolite may
reach the lung vasculature. In this case, the sensitivity of the endothelium may be
of secondary importance compared to the relative concentration of metabolite
reaching the pulmonary vascular bed.

Finally, it is important to consider that the species differences observed in
response to this compound in vitro may be due to inherent differences which are
expressed when endothelial cells of different species are maintained in culture.
These may include differences in matrix production, glycocalyx composition, protein
production or a number of other factors. Although these species differences have not
been investigated in detail, a number of differences have been reported between
endothelial cells of large and small vessel origin (Kumar et al., 1987; Gumkowski et
al., 1987; Stolz and Jacobson, 1991). Some of these may be important in interpreting
the cellular response to injury, as the BECs and PECs used in many of these studies
were of pulmonary artery origin, whereas the RECs used in these investigations were

isolated from smaller, intrapulmonary vessels.

Clearly, much work remains to be done, both in identifying specific, relevant
functional alterations in ECs in response to MCTP treatment, and in determining the
role of DNA crosslinking in the development of these alterations. These studies in
cultured endothelium have established that changes occur in vitro which appear to
be consistent with the development of chronic vascular alterations in viva, but

eventually these observations must be confirmed in the pulmonary vasculature of

245

treated animals in situ. Although one must bear in mind the limitations of the
cultured cell system, this model of EC injury should continue to provide important
information on mechanisms of MCTP pneumotoxicity, as well as on the role of this

pluripotent tissue in the development of vascular disease.

BIBLIOGRAPHY

246

ABRAHAM, A.S., KAY, J.M., COLE, RB. and PINCOCK, AC. (1971).
Haemodynamic and pathological study of the effect of chronic hypoxia and
subsequent recovery of the heart and pulmonary vasculature of the rat.
Cardiovascular Res. 5, 95-102.

ADAMS, R. and RODGERS, ER (1939). The structure of monocrotaline, the

alkaloid in Crotaiag'a spectabilis and Crgtalgrig remsa I. JAmenChemSac. 61,
2815-2819.

ADNOT, S., RAFFESTIN, B., EDDAHIBI, S., BRAQUET, P. and CHABRIER,
RE. (1991). Loss of endothelium-dependent relaxant activity in the pulmonary
circulation of rats exposed to chronic hypoxia. .LClinInvest. 87, 155-162.

ALBELDA, S.M., ELIAS, IA, LEVINE, EM. and KERN, LA. (1989). Endotoxin
stimulates platelet-derived growth factor production from cultured human pulmonary
endothelial cells. AmlPhysiaL (Lung CelLMaLPhysiaLI) 257, L65-L70.

ALBELDA, SM. and BUCK, CA (1990). Integrins and other cell adhesion
molecules. FASEB J. 4, 2868-2880.

ALEXANDER, J.S., PATTON, W.F., YOON, M.U. and SHEPRO, D. (1991).
Cytokeratin filament modulation in pulmonary microvessel endothelial cells by
vasoactive agents and culture confluency. Tissue Cell 23, 141-150.

ALLEN, LR. and CARSTENS, LA. (1968). Veno-occlusive disease in Rhesis
monkeys. AmerJ.Vet.Res. 29, 1681-1694.

ALLEN, J. R. and CARSTENS, LA. (1970). Clinical signs and pathologic changes
in Wis intoxicated rats. Amen]. Vet. Res. 31, 1059-1070.

ALLEN, J.R., CHILDS, GR. and CRAVEN S, W.W. (1960). Crotalaria spectabilis
toxicity in chickens. PracSac.Exp.BiaLMed. 104, 434-436.

ALLEN, J.R., LALICH, JJ. and SCHMITTLE, S.C. (1963). Crotalaria
spectabilis-induced cirrhosis in turkeys. Laanvest. 12, 512-517.

‘ ALLEN, J .R., CARSTENS, LA. and KNEZEVIC, AL. (1965). W
intoxication in rhesus monkey. AmerJ. Vet.Res. 26, 753-755.

ALLEN, J .R. and CARSTEN S, LA (1970). Pulmonary vascular occlusions initiated
by endothelial lysis in monocrotaline-intoxicated rats. Exp.Mal.Pathal. 13, 159-171.

ALLEN, J .R., CHESNEY, CF. and FRAZEE, W.J. (1972a). Modifications of
pyrrolizidine alkaloid intoxication resulting from altered hepatic rnicrosomal enzymes.
Taxical. Appl. Pharmacol. 23, 470-479.

247

ALLEN, JR. and CHESNEY, C.F. (1972b). Effect of age on development of cor
pulmonale in non-human primates following pyrrolizidine alkaloid intoxication.
ExpenMaLPathal. 17, 220-232.

ALLEN, J .R., HSU, LC. and CARSTEN S, LA. (1975). Dehydroretronecine-induced
rhabdomyosarcomas in rats. Cancer Res. 35, 997-1002.

ALT'IERE, RJ., OLSON, J.W. and GILLESPIE, MN. (1986). Altered pulmonary
vascular smooth muscle responsiveness in monocrotaline-induced pulmonary
hypertension. lP/tannacaLEprIter. 236, 390-395.

ANTONELJJ-ORLIDGE, A, SAUNDERS, K.B., SMITH, S. and D’AMORE, PA
(1989). An activated from of transforming growth factor-beta is produced by
cocultures of endothelial cells and pericytes. Prac.NatLAcad.Sci. USA 516, 4544-4568.

ASSOIAN, R.K. and SPORN, MB. (1986). Type beta transforming growth factor in
human platelets: Release during platelet degranulation and action on vascular
smooth muscle cells. .LCeILBiaL 102, 1217-1223.

AUERBACH, R. (1981). Angiogenesis-inducing factors: a review. In Lymphakines.
Vol.4 (E. Pick, Ed.), pp.69-88. Academic Press, New York.

AUERBACH, R., ALBY, L, GRIEVES, J., JOSEPH, J., LINGREN, C.,
MORRISSEY, LW., SIDKY, Y.A, TU, M. and WATT, $1.. (1982). Monoclonal
antibody against angiotensin-converting enzyme: its use as a marker for murine,
bovine, and human endothelial cells. Prac.Natl.Acad.Sci. USA 79, 7891-7895.

AZUMA, H., ISHIKAWA, M. and SEKIZAKI, S. (1986). Endothelium-dependent
inhibition of platelet aggregation. BrJ.PharmacaL 88, 411-415.

BANDA, MJ. and WERB, A (1981). Mouse macrophage elastase. Purification and
characterization as a metalloproteinase. BiachemJ. 193, 589-605.

BARER, G.F., FINLAY, M., BEE, D. and WACH, RA (1982). Pulmonary
hypertension of chronic hypoxia in a rat model: anatomical and rheological factors.
BulLEunPhysiapathaLRespir. 18, 69-73.

BARNES, MJ., MORTON, LF. and LEVENE, CJ. (1978). Synthesis of interstitial
collagens by pig aortic endothelial cells in culture. Biachem.Biaphys.Res.Cammun. 84,
646-653.

BARRETT, T.B., GAJDUSEK, C.M., SCHWARTZ, S.M., MCDOUGALL, J .K. and
BENDITT, ER (1984). Expression of the sis gene by endothelial cells in culture and
in vivo. Prac.NatLAcad.Sci. USA 81, 6772-6774.

248

BARST, RJ. and STALCUP, SA. (1985). Endothelial function in clinical pulmonary
hypertension. Chest 88, 2168-2198.

BAUER, KA and ROSENBERG, RD. (1987). The pathophysiology of
prethrombotic state in humans: Insights gained from studies using markers of
hemostatic system activation. Blood 70, 343-350.

BECKER, R.C. (1991). Seminars in thrombosis, thrombolysis and vascular biology.
I. The vascular endothelium. Cardiology 78, 13-22.

BELL, L and MADRI, J.A (1990). Influence of the angiotensin system on
endothelial and smooth muscle cell migration. AmerJPathal. 137, 7-12.

BERGMEYER, H.U., and BERNT, E. (1974). Lactate dehydrogenase UV assay with
pyruvate and NADPH. In Methods of Enzymatic Analysis, Vol. 2 (H.U. Bergmeyer,
Ed.), pp.574-579. Academic Press, New York.

BERK, B.C., BROCK, T.A., GIMBRONE, MA. and ALEXANDER, R.W. (1987).
Early agonist-mediated ionic events in cultured vascular smooth muscle cells: calcium
mobilization is associated with intracellular acidification. J.Bial. Chem. 262, 5065-5072.

BEVILACQUA, M.P., STENGELIN, SJR., GIMBRONE, MA. and SEED, B.
(1987). Endothelial leukocyte adhesion molecule 1: an inducible receptor for
neutrophils related to complement regulatory proteins and lectins.
PracNatLAcadSci. USA 84, 9238-9242.

BIERER, B.W., VICKERS, C.L, RHODES, W.H. and THOMAS, J.B. (1960).
Comparison of the toxic effects of Crotalaria spectabilis and Crotalaria giant striata
as complete feed contaminants. J.Amer. Vet.Med.Assac. 318-322.

BLAES, NA and BOISSEL, J .P. (1983). Growth-stimulating effect of catecholamines
on rat aortic smooth muscle cells in culture. .LCelLPhysial. 116, 167-172.

BLAUSTEIN, R.L, BLAUSTEIN, A. and ROSENBLAT, M. (1965).
Monocrotaline-induced atheromatous lesions in coronary artery of the rat. ArchPath.
79, 335-339.

BLOCK, E.R., PATEL, J .M. and SHERIDAN, N.P. (1985). The effect of oxygen and
endotoxin on lactate dehydrogenase release, S-hydroxytryptamine uptake and
antioxidant enzyme activities in endothelial cells. lCelLPhysial. 122, 240-248.

BODELL, W.J., AIDA, T., BERGER, MS. and ROSENBLUM, ML. (1985). Repair
of O6-(2-chloroethyl)guanine mediates the biological effects of
chloroethylnitrosoureas. Env. Health Persp. 62, 119-126.

249

BODELL, W.J. (1990). Molecular dosimetry for sister-chromatid exchange induction
and cytotoxicity by monofunctional and bifunctional alkylating agents. Mutation Res.
233, 203-210.

BOOYSE, F.M., SEDLAK, BJ. and RAFELSON, ME. (1975). Culture of arterial
endothelial cells. Characterization and growth of bovine aortic cells.
Thramb. DiatILHaemarrh. 35, 825-839.

BORSUM, T. (1991). Biochemical properties of vascular endothelial cells. Vrrchaws
Archiv. B CelLPathaL 60, 279-286.

BRADFORD, M.M. (1976). A rapid and sensitive method for the quantitation of
microgram quantities of protein utilizing the principle of protein-dye binding.
AnaLBiachem. 72, 248-254.

BRADHAM, D.M., IGARASHI, A, POTTER, R.L. and GROTENDORST, GR.
(1991). Connective tissue growth factor: a cysteine-rich mitogen secreted by human
vascular endothelial cells is related to the SRC-induced immediate early gene
product CEF-IO. .LCelLBial. 114, 1285-1294.

BRAS, G. and HILL, KR. (1956). Veno-occlusive disease of the liver, essential
pathology. Lancet 2, 161-163.

BRAS, G., BERRY, D.M. and GYORGY, P. (1957). Plants as aetiological factor in
veno-occlusive disease of the liver. Lancet i, 960-962.

BREGMAN, C.L, BUROKER, R.A., BRADNER, W.T., HIRTH, RS. and
MADISSOO, H. (1989) Cardiac, renal, and pulmonary toxicity of several mitomycin
derivatives in rats. FundAppl. Taxical. 13, 46-64.

BRENNER, O. (1935). Pathology of the vessels of the pulmonary circulation.
ArcthLMed. 56, 211.

BRIZIO-MOLTENI, L, PIANO, G., RICE, PL, WARPEHA, R., FRESCO, R.,
SOLLIDAY, NH. and MOLTENI, A. (1984). Effect of wood combustion smoke
inhalation on angiotensin-l-converting enzyme in the dog. AnnClinLab.Sci. 14,
381-389.

BRUNER, LH., HILLIKER, KS. and ROTH, RA (1983). Pulmonary hypertension
and ECG changes from monocrotaline pyrrole in the rat. AmLPhysial. 245,
H300-H306.

BRUNER, LH., CARPENTER, L.J., HAMLOW, P. and ROTH, RA. (1986). Effect
of a mixed function oxidase inducer and inhibitor on monocrotaline pyrrole
pneumotoxicity. Taxical. Appl. Pharmacol 85, 416-427.

250

BRUNER, LH. and ROTH, RA (1984). Diethylcarbamazine protects rats from the
cardiopulmonary effects of monocrotaline pyrrole. Fed.Prac. 43, 923-923.(Abstract)

BRUNER, LH. (1986). The Role of Immune Efi‘ectars in Manacnataline
Pyrrole-Induced Pulmonary Injury: A Dissertation. , .

BRUNER, LH., BULL, R.W. and ROTH, R.A (1987a). The effect of
immunosuppressants and adoptive transfer in monocrotaline pyrrole pneumotoxicity.
Toxicol. Appl. Pharmacol. 91, 1-12.

BRUNER, LH., JOHNSON, K., CARPENTER, LJ. and ROTH, R.A. (1987b). Lack
of effect of deferoxamine, dimethyl sulfoxide, and catalase on monocrotaline pyrrole
pulmonary injury. J.Taxical.Enviran.Health 21, 205-217.

BRUNER, LH., JOHNSON, KJ., TILL GO. and ROTH, RA. (1988). Complement
is not involved in monocrotaline pyrrole-induced pulmonary injury.Am.J.PhysiaL 254,
H258-H264.

BUCHANAN, MR. and BASTIDA, E. (1988). Endothelium and underlying
membrane reactivity with platelets, leukocytes and tumor cells: regulation by the
lipoxygenase-derived fatty acid metabolites, 13-HODE and HETES. Medical
Hypotheses 27, 317-325.

BUCHANAN, M.R., HAAS, T.A., LAGARDE, M. and GUICHARDANT, M.
(1985). 13-Hydroxyoctadecadienoic acid is the vessel wall chemorepellant factor,
LOX. J.Bial. Chem. 260, 16056-16059.

BUCKMASTER, G.W., CHEEKE, RR. and SHULL, LR. (1976). Pyrrolizidine
alkaloid poisoning in rats. J.AnimSci. 43, 464-473.

BUHLER, D.R., MIRANDA, C.L., KEDZIERSKI, B. and REED, R.L. (1990).
Mechanisms for pyrrolizidine alkaloid activation and detoxification.Adv.Exp.Med.Bial.
283, 597-603.

BULL, LB., CULVENOR, C.C..l., and DICK, AT. (1968). The Pyrrolizidine
Alkaloids: Their chemistry, pathogenicity and other bialagicalpraperties. pp. 1-293.North
Holland Publishing Company, Amsterdam.

BUONASSISI, V. (1973). Sulfated mucopolysaccharide synthesis and secretion in
endothelial cell cultures. Exp. Cell Res. 76, 363-368.

BUONASSISI, V. and VENTER, J .C. (1976). Hormone and neurotransmitter
receptors in an established vascular endothelial cell line. Prac.Natl.Acad.Sci. USA 73,
1612-1616.

251

BURGESS, W.H. and MACIAG, T. (1989). The heparin-binding (fibroblast) growth
factor family of proteins. AnnRev.Biachem. 58, 575-606.

BUSSOLINO, F ., CAMUSSI, G. and BAGLIONI, C. (1988). Synthesis and release
of platelet-activating factor by human vascular endothelial cells treated with tumor
necrosis factor or interleukin lalpha. J.Biol Chem. 263, 11856-11861.

BUTLER, W.H. (1970a). An ultrastructural study of the pulmonary lesion induced
by pyrrole derivatives of the pyrrolizidine alkaloids. J.Path. 102, 15-19.

BUTLER, W.H., MATTOCKS, AR. and BARNES, J.M. (1970b). Lesions in the
liver and lungs of rats given pyrrole derivatives of pyrrolizidine alkaloids. J.Path. 100,
169-175.

CAMEJO, G., LOPEZ, A., LOPEZ, F. and OUINONES, J. (1985). Interaction of
low density lipoproteins with arterial proteoglycans. The role of charge and sialic acid
content. Atherosclerosis 55, 93-105.

CAMPBELL-BOSWELL, M. and ROBERTSON, AL. (1981). Effects of angiotensin
II and vasopressin on human smooth muscle cells in vitro. Exp.Mal.Pathal. 35,
265-276.

CANTRELL, J.E.JR., PHILLIPS, T.M. and SCI-IEIL, RS. (1985).
Carcinoma-associated hemolytic-uremic syndrome: a complication of mitomycin C
chemotherapy. J.Clin. Oneal. 3, 723-734.

CAREY, DJ. (1991). Control of growth and differentiation of vascular cells by
extracellular matrix proteins. AnnRevJ’hysioL 53, 161-177.

CARILLO, L and AVIADO, D.M. (1969). Monocrotaline-induced pulmonary
hypertension and p-chlorophenylalanine (PCPA). Laanvest. 20, 243-248.

CARNES, W.H., ABRAHAM, RA. and BUONASSISI, V. (1979). Biosynthesis of
elastin by an endothelial cell culture. BiachemBiaphys.Res.Cammun. 90, 1393- 1399.

CARTIER, R., PEARSON, PJ., LIN, PJ. and SCHAFF, H.V. (1991). Time course
and extent of recovery of endothelium-dependent contractions and relaxations after
direct arterial injury. J.Tharac.Cardiovasc.Surg. 102, 371-377.

CASTELLOT, JJJR., ADDONIZIO, M.L, ROSENBERG, R. and KARNOVSKY,
M. (1981). Chrltured end0thelial cells produce a heparinlike inhibitor of smooth
muscle cell growth. .LCelLBial. 90, 372-379.

252

CAST'ELLOT, JJJR., FAVREAU, LV., KARNOVSKY, MJ. and ROSENBERG,
RD. (1982). Inhibition of vascular smooth muscle cell growth by endothelial
cell-derived heparin. Possible role of a platelet endoglycosidase. J.BiaLChem. 257,
11256-11259.

CATRAVAS, J.D., BASSINGTHWAIGHTE, J.B., and SPARKS, H.VJR. (1988a).
Adenosine transport and uptake by cardiac and pulmonary endothelial cells. In
Endothelial Cells. Vol.1 (U.S. Ryan, Ed.), pp.65-84. CRC Press, Boca Raton.

CATRAVAS, J.D., BURCH, S.E., SPURLOCK, B. and MILLS, LR. (1988b). Early
effects of ionizing radiation on pulmonary endothelial angiotensin converting enzyme
and S’-nucleotidase, in viva. TaxicaL Appl. Pharmacol. 94, 342-355.

CERVENY, T‘J., FASS, D.N. and MANN, KG. (1984). Synthesis of coagulation
factor V by cultured aortic endothelium. Blood 63, 1467-1474.

CHANDLER, DB. and GIRI, SN. (1983). Changes in plasma concentrations of
prostaglandins and plasma angiotensin-converting enzyme during bleomycin-induced
lung fibrosis in hamsters. AmRev.Respir.Dis. 128, 71-76.

CHANG, AY., KUEBLER, J.P., PANDYA, K.J., ISRAEL, R.H., MARSHALL, BC.
and TORMEY, DC. (1986). Pulmonary toxicity induced by mitomycin C is highly
responsive to glucocorticoids. Cancer 57, 2285-2290.

CHEEKE, RR. (1989). Pyrrolizidine alkaloid toxicity and metabolism in laboratory
animals and livestock. In Taxicants of Plant Origin: Val! Alkaloids (P.R. Cheeke, Ed.),
pp.1-22. CRC Press, Boca Raton, FL.

CHEEKE, RR. and PIERSON-GOEGER, ML (1983). Toxicity of Seneg’ojagohaea
and pyrrolizidine alkaloids in various laboratory animals and avian species.
TaxicaLLett. 18, 343-349.

CHESNEY, CF. and ALLEN, J.R. (1973a). Resistance of the guinea pig to
pyrrolizidine alkaloid intoxication. Toxicol. Appl. Pharmacol. 28, 385-392.

CHESNEY, CF. and ALLEN, J .R. (1973b). Endocardial fibrosis associated with
monocrotaline-induced pulmonary hypertension in nonhuman primates (Mm
mags).AmerJ. Vet.Res. 34, 1577-1581.

CHESNEY, CF. and ALLEN, J .R. (1973c). Monocrotaline induced pulmonary
vascular lesions in non-human primates. Cardiovascular Res. 7, 508-518.

CHESNEY, CF. and ALLEN, J.R. (1973d). Cardiopulmonary disease.Amer.J.PathaL
70, 489-492.

253

CHESNEY, C.F., ALLEN, JR. and HSU, I.C. (1974a). Right ventricular hypertrophy
in monocrotaline treated rats. Exper.MoLPathaL 20, 257-268.

CHESNEY, C.F., HSU, LC. and ALLEN, J.R. (1974b). Modifications of the myjtm
metabolism of the hepatotoxic pyrrolizidine alkaloid, monocrotaline.
ResCammunChemPathaLphannacal. 8, 567-570.

CLOWES, AW. and KARNOVSKY, MJ. (1977). Suppression by heparin of smooth
muscle cell proliferation in injured arteries. Nature (Land. ) 265, 625-626.

CLOWES, AW., REIDY, M.A and CLOWES, M.M. (1983). Mechanism of stenosis
after arterial injury. Laanvest. 49, 208.

CLOWES, AW. and CLOWES, M.M. (1986). Kinetics of cellular proliferation after
arterial injury: IV. Heparin inhibits rat smooth muscle rnitogenesis and migration.
CircRes. 58, 839-845.

CLOWES, AW., CLOWES, M.M., FINGERLE, J. and REIDY, M.A (1989).
Regulation of smooth muscle cell growth in injured artery. .LCardiavascPharmacaL
14(Suppl. 6), $12-$15.

CLOZEL, J.P., HESS, P., MICHAEL, C., SCI-IIETINGER, K. and
BAUMGARTNER, HR. (1991). Inhibition of converting enzyme and neointima

formation after vascular injury in rabbits and guinea pigs. Hypertension 18(suppl II),
II-55-II-59.

CLDZEL M. (1991). Mechanism of action of angiotensin converting enzyme
inhibitors on endothelial function in hypertension. Hypertension 18[suppl II],
II-37-II-42.

COCHRANE, C.G., SPRAGG, R. and REVAK, SD. (1983). Pathogenesis of the
adult respiratory distress syndrome. Evidence of oxidant activity in bronchoalveolar
lavage fluid. J.Clin.Invest. 71, 754-759.

COLBURN, P. and BUONASSISI, V. (1982). Anti-clotting activity of endothelial cell
cultures and heparan sulfate proteoglycans. BiachemBiaphys.Res.Cammun. 104,
220-227.

COOK, J .W., DUFFY, E. and SCHOENTAL, R. (1950). Primary liver tumors in rats
following feeding with alkaloids of mm. BrJ. Cancer 4, 405-410.

COOK, LM. and HOLT, AGE. (1966). Mutagenic activity of two pyrrolizidine
alkaloids. J.Genet. 59, 273-274.

254

COOMBER, BL and GOTLIEB, A1. (1990). In vitro endothelial wound repair.
Interaction of cell migration and proliferation. Arteriasclerasrls 10, 215-222.

COTRAN, RS. (1965). The delayed and prolonged vascular leakage in inflammation.
11. An elecron microscope study of the vascular response after thermal injury.
AmerJJ’athal. 46, 589-620.

COTRAN, RS. (1987). New roles for the endothelium in inflammation and
immunity. AmerJJ’athal. 129, 407-413.

COX, D.H., HARRIS, D.L and RICHARD, T.A (1958). Chemical identification of
Crotalaria poisoning in horses. J.Amer.Vet.Med.Assac. 133, 425-426.

COZZOLINO, F ., TORCIA, M., ALDINUCCI, D., ZICHE, M., ALMERIGOGNA,
F., BANI, D. and STERN, D.M. (1990). Interleukin 1 is an autocrine regulator of
human endothelial cell growth. Prac.NatlAcadSci. USA 87, 6487-6491.

CRAPO, J.D., BARRY, F.E., FOSCUE, HA. and SHELBOURNE, J. (1980).
Structural and biochemical changes in rat lungs occurring during exposures to lethal
and adaptive doses of oxygen. AmRev.Respir.Dis. 122, 123-143.

CROOKE, S.T. and BRADNER, W.T. (1976). Mitomycin C. A review. Cancer
Treat.Rev. 3, 121-139.

CROSS, S.A.M., ALABASTER, V.A., BAKHLE, Y.S. and VANE, J .R. (1974). Sites
of uptake of 3H-5-hydroxytryptamine in rat isolated lung. Histachemistry 39, 83-91.

CSONKA, E. and JELLINEK, H. (1986). Aortic endothelial and smooth muscle cell
interactions in culture. Acta Morphologica Hungarica 34, 157-161.

CULVENOR, C.C.J. and DALBON, R. (1964). Carbonyl stretching frequencies of
pyrrolizidine alkaloids. AustrJ.Chem. 17, 1296-1300.

CULVENOR, C.C.J., DOWNING, D.T., EDGAR, J.A. and JAGO, M.V. (1969).
Pyrrolizidine alkaloids as alkylating and antimitotic agents. Ann.N. Y.AcadSci. 163,
837-847.

CULVENOR, C.C.J., EDGAR, J .A., SMITH, LW. and TWEEDDALE, HJ. (1970).
Dihydropyrrolizines. III. Preparation and reactions of derivatives related to
pyrrolizidine alkaloids. AustrJ.Chem. 23, 1853-1867.

CULVENOR, C.C.J., EDGAR, J.A, JAGO, M.V., OUTTERIDGE, A.,
PETERSON, J.E. and SMITH, LW. (1976). Hepato- and pneumotoxicity of
pyrrolizidine alkaloids and derivatives in relation to molecular structure.
ChemBiaLInteract. 12, 299-324.

255

CZER, G.T., MARSH, J., KONOPKA, R. and MOSER, KM. (1986). Low-dose
PGIZ prevents monocrotaline-induced thromboxane production and lung injury.
J.AppLPhysiaL 60, 464-471.

D’AMORE, RA (1990). Heparin-endothelial cell interactions. Haemastasis 20(suppl
1), 159-165.

DAHM, LJ., BRUNER, LH. and ROTH, RA. (1986). Superoxide (02-) production
by bronchoalveolar cells harvested from monocrotaline pyrrole (MCI'P)-treated rats.
FedJ’rac. 45, 220-220.(Abstract)

DAVIES, P.F., GANZ, D. and DIEHL, PS. (1985). Reversible
microcarrier-mediated junctional communication between endothelial and smooth

muscle cell monolayers: an in vitro model for vascular interactions. Lab.Invest. 85,
710-715.

DAVIES, RF. (1986). Vascular cell interactions with special reference to the
pathogenesis of atherosclerosis. Laanvest. 55, 5-24.

DEEN, W.M., BRIDGES, CR. and BRENNER, BM. (1983). Biophysical basis of
glomerular permeability. J.MemaniaL 71, 1-10.

DELVECCHIO, PJ. and SMITH, JR. (1981). Expression of angiotensin-converting
enzyme activity in cultured pulmonary artery endothelial cells. ICelLPhysiaL 108,
337-345.

DEYO, J.A and KERKVLIET, N .I. (1990). lmmunotoxicity of the pyrrolizidine
alkaloid monocrotaline following subchronic administration to C57Bl/6 mice.
FundamAppl. Toxicol. 14, 842-849.

DICKINSON, JD. (1980). Release of pyrrolizidine alkaloids into milk.
PraaWestJ’harmacaLSoc. 23, 377-379.

DICORLETO, RE. and BOWEN-POPE, DP. (1983). Cultured endothelial cells
produce a platelet-derived growth factor-like protein. Prac.Natl.Acad.Sci. USA 80,
1919-1923.

DOBULER, K.J., CATRAVAS, JD. and GILLIS, ON. (1982). Early detection of
oxygen-induced lung injury in conscious rabbits: reduced activity of angiotensin
converting enzyme and removal of 5-hydroxytryptamine. Am.Rev.Respir.Dis. 126,
534-539.

DOMSCHKE, S. and SOLING, H.D. (1973). Polyamine metabolism in rat liver:
effect of starvation and refeeding. Harm.Metab.Res. 5, 97-101.

256

DRESDALE, D.T., SCHULTZ, M. and MICHTOM, RJ. (1951). Primary pulmonary
hypertension. 1. Clinical and haemodynarnic study. AmerlMed. 11, 686-705.

DUEKER, S.R., LAME, M.W., MORIN, D., WILSON, D.W. and SEGALL, HJ.
(1992). Guinea pig and rat hepatic microsomal metabolism of monocrotaline. The
Taxicalogr'st 12, 386-386.(Abstract)

DULFIANTY, AM., LI, M. and WHIT'MORE, GP. (1989). Isolation of Chinese
hamster ovary cell mutants deficient in excision repair and mitomycin C
bioactivation. Cancer Res. 49, 117-122.

DUNN, M.M., DRAB, EA. and RUBIN, DB. (1986). The effect of irradiation on
endothelial cell-polymorphonuclear leukocyte interaction. J.AppLPhysial. 60,
1932-1937.

DUPERRAY, A, T'RANQUI, L, ALIX, J .L and CORDONNIER, D. (1988) Effect
of mitomycin C on prostacyclin synthesis by human endothelial cells.
BiochemPhannacal. 24, 4753-4757.

DUTHU, GS. and SMITH, J .R. (1980). In vitro proliferation and lifespan of bovine
aorta endothelial cells: effect of culture conditions and fibroblast growth factor.
.LCelLPhysial. 103, 385-392.

DZAU, V.]. and GIBBONS, G.H. (1991). Endothelium and growth factors in
vascular remodeling of hypertension. Hypertension 18[suppl III], III-115-III-121.

ELSAYED, N.M., TIERNEY, DP. and HACKER, AD. (1988). Effects of food
restriction and hyperoxia on rat survival and lung polyamine metabolism.
AmlPhysial. 64, 2010-2016.

EMMEL M.W., SANDERS, DA. and HEALY, W.W. (1935). W15
seed poisoning in swine. J.Amer. Vet.Med.Assac. 86, 43-54.

ERDOS, E.G. (1975). Angiotensin converting enzyme. CircRes. 36, 247-255.

ERNSTER, L (1967) DT-diaphorase. Methods Enzymal. 10, 309-317.
ESMON, CT. and OWEN, W.G. (1981). Identification of an endothelial cell cofactor
for thrombin-catalyzed activation of protein C. Prac.Natl.Acad. Sci. USA 78, 2249-2252.

ESTEP, J.B., LAME, M.W., JONES, AD. and SEGALL, HJ. (1990).
N-Acetylcysteine-conjugated pyrrole identified in rat urine following administration
of two pyrrolizidine alkaloids, monocrotaline and senecionine. TaxicaLLett. 54, 61-69.

257

ESTEP, J.B., LAME, M.W., MORIN, D., JONES, AD., WILSON, D.W. and
SEGALL, HJ. (1991). [14C]Monocrotaline kinetics and metabolism in the rat. Drug
Metab.Di.spas. 19, 135-139.

FAJARDO, LF. (1989). The complexity of endothelial cells: A review.
AmJ.Clin.Pathal. 92, 241-250.

FANBURG, BL and GLAZIER, J .B. (1973). Conversion of angiotensin I to
angiotensin H in the isolated perfused dog lung. J.AppLPhysial. 35, 325-331.

FANT ONE, J .C. and WARD, RA. (1982). Role of oxygen-derived free radicals and
metabolites in leukocyte-dependent inflammatory reactions. AmerJJ’athal. 107,
397-418.

FARIDY, E.E. (1970). Effect of food and water deprivation on surface activity of
lungs of rats. J.AppLPhysiol. 29, 493-498.

FASULES, J.W., STENMARK, K.R., HENSON, P.M., VOELKEL NP. and
REEVES, J.T. (1987). Neither anticoagulant nor nonanticoagulant heparin affects
monocrotaline lung injury. AmJ.PhysiaL 62, 816-820.

FERNANDEZ-BONET‘TI (1983). .AmJ.Physial

FISHMAN, AP. (1980). Vasomotor regulation of the pulmonary circulation.
AnnRevJ’hysial. 42, 211-220.

FOLKMAN, J. (1982). Angiogenesis: initiation and control. J.CelLBial. 104, 212-227.

FORM, D.M., PRATI‘, B.M. and MADRI, J .A. (1986). Endothelial cell proliferation
during angiogenesis. In vitro modulation by basement membrane components.
Lab.Invest. 55, 521-530.

FOX, PL and DICORLETO, RE. (1984). Regulation of production of a
platelet-derived growth factor-like protein by cultured bovine aortic endothelial cells.
J.CelLPhysiol. 121, 298-308.

FOX, PL and DICORLETO, RE. (1986). Modified low density lipoprotein suppress
production of a platelet-derived growth factor from cultured endothelial cells.
PraaNatLAcadSci. USA 83, 4774-4800.

FRANKE, W.W., COWIN, P., GRUND, C., KUHN, C., and KAPPRELL, HP.
(1988). The endothelial junction. The plaque and its components. In Endothelial Cell
Biology in Health and Disease (11. Simionescu, and m. Simionescu, Eds), pp.147-166.
Plenum Press, New York.

258

FURCHGOTT, RF. (1984). The role of the endothelium in the responses of
vascular smooth muscle to drugs. AnnRev.PharmacaL Toxicol. 24, 175-197.

FURCHGOTT, RF. and ZAWADZKI, J .V. (1980). The obligatory role of
endothelial cells in the relaxation of arterial smooth muscle by acetylcholine. Nature
(Land) 288, 373.

GAIL D.B., MASSARO, G.D. and MASSARO, D. (1977). Influence of fasting on
the lung. J.AppLPhysial. 42, 88-92.

GAJDUSEK, C., DICORLETO, P.E., ROSS, R. and SCHWARTZ, SM. (1980). An
endothelial cell-derived growth factor. J.Cell Biol. 85, 467-472.

GALDAL KS. (1984). Thromboplastin synthesis in endothelial cells. Haemastasis
14, 378-385.

GANEY, P.E., FINK, G.D. and ROTH, RA (1985). The effect of dietary restriction
and altered sodium intake on the cardiopulmonary toxicity of monocrotaline pyrrole.
Toxicol. Appl. Pharmacal. 78, 55-62.

GANEY, RE. and ROTH, RA. (1986). Thromboxane does not mediate pulmonary
vascular response to monocrotaline pyrrole. AmJ.PhysiaL 252, H743-H748.

GANEY, RE. and ROTH, R.A. (1987a). S-I-Iydroxytryptarnine and thromboxane in
platelets from rats treated with monocrotaline pyrrole. Toxicol. Appl. Pharmacol. 88,
157-164.

GANEY, RE. and ROTH, R.A. (1987b). 6-Keto prostaglandin FI-alpha and
thromboxane B2 in isolated, buffer-perfused lungs from monocrotaline
pyrrole-treated rats. ExpLung Res. 12, 195-206.

GANEY, P.E., SPRUGEL, K.H., HADLEY, KB. and ROTH, RA (1986).
Monocrotaline pyrrole-induced cardiopulmonary toxicity is not altered by metergoline
or ketanserin. J.PharmacaLEprher: 237, 226-231.

GANEY, P.E., HILLIKER-SPRUGEL, K, WHITE, S.M., WAGNER, J.G. and
ROTH, RA (1988). Pulmonary hypertension due to monocrotaline pyrrole is
reduced by moderate thrombocytopenia. AmJ.Physial. 255, H1165-H1172.

GANEY, RE. and ROTH, RA. (1987). Elevated serum copper concentration in
monocrotaline pyrrole treated rats with pulmonary hypertension. BiachemPharmacol.
36, 3535-3537.

GARDINER, M.R., ROYCE, R. and BOKOR, A (1965). Studies on W
grim a newly recognized cause of Kimberly horse disease.J.Path.Bact. 89, 43-55.

259

GEGGEL R.L, HU, LM. and REID, LM. (1986). Effect of heparin during chronic
hypoxia: a hemodynamic study including acute hypoxia. FASEB J. 45,
943-943.(Abstract)

GEGGEL R.L, CARVALHO, AC.A., HOYER, LW. and REID, LM. (1987). Von
Willegrand factor abnormalities in primary pulmonary hypertension.
AmRev.Respir.Dis. 135, 294-299.

GHODSI, R. and WILL J .A (1981). Changes in pulmonary structure and function
induced by monocrotaline intoxication. AmJ.Physial. 240, H149-H155.

GIBBONS, G.H. and DZAU, VJ. (1990). Angiotensin converting enzyme, inhibition
and vascular hypertrophy in hypertension. Cardiovasc.Drugs Ther. 4, 237-242.

GILLESPIE, M.N., FREDERICK, W.B., ALTIERE, R.J., OLSON, J.W. and
KIMMEL E.C. (1985a). Pulmonary mechanical, ventilatory, and gas exchange

abnormalities in rats with monocrotaline-induced pulmonary hypertension. Fncp.Lung
Res. 8, 189-197.

GILLESPIE, M.N., DYER, KK, OLSON, J .W., O’CONNOR, W.N. and ALTIERE,
R.J. (1985b). Alpha-difluoromethylornithine, an inhibitor of polyamine synthesis,
attenuates monocrotaline-induced pulmonary vascular hyperresponsiveness in isolated
perfused rat lungs. Res. Commun. Chem.PathaI.phannacal. 50, 365-378.

GILLESPIE, M.N., OLSON, J.W., REINSEL C.N., O’CONNOR, W.N. and
ALTIERE, RJ. (1986). Vascular hyperresponsiveness in perfused lungs from
monocrotaline-treated rats. AmJ.PhysiaL 251, H109-H114.

GILLESPIE, M.N., GOLDBLUM, D., COHEN, C. and MACCLAIN, J. (1988).
Interleukin 1 bioactivity in the lungs of rats with monocrotaline-induced pulmonary
hypertension. Proc.Sac.Exp.BioLMed. 187, 26-26.(Abstract)

GILLESPIE, M.N., RIPPETOE, P.E., HAVEN, C.A, SHIAO, R.T., ORLIN SKA, U.,
MALEY, B. and OLSON, J .W. (1989). Polyamines and epidermal growth factor in
monocrotaline-induced pulmonary hypertension. Am.Rev.Respir.Dis. 140, 1463-1466.

GILLIS, C.N. (1973). Metabolism of vasoactive hormones by lung. Anesthesiology 39,
626-632.

GILLIS, C.N., HUXT ABLE, RJ. and ROTH, RA (1978). Effects of monocrotaline
pretreatment of rats on removal of 5-hydroxytryptamine and noradrenaline by
perfused lung. Br.J.Phamracal 63, 435-443.

260

GILLIS, C.N. and CATRAVAS, JD. (1982). Altered removal of vasoactive
substances in the injured lung: detection of lung microvascular injury.
Ann.N.YAcadSci. 384, 458-474.

GILLIS, C.N. and ROTH, JA. (1976). Pulmonary disposition of circulating
vasoactive hormones. BiachemPharmacal. 25, 2547-2553.

GIMBRONE, M.A. (1987). Vascular endothelium: nature’s blood-compatible
container. AnnMYAcadSci. 516, 5-11.

GIRI, J.G., LOMEDICO, RT. and MIZEL SB. (1985). Studies on the synthesis and
secretion of interleukin 1. I. A 33,000 molecular weight precursor for interleukin 1.
I.Immunal. 134, 343-349.

GOLDSCHMIDT, IE. and TALLARIDA, RJ. (1991). Pharmacological evidence
that captopril possesses an end0thelium-mediated component of vasodilation: effect
of sulfliydryl groups on endothelium-derived relaxing factor. .LPhannacaLEprher.
257, 1136-1145.

GOSPODAROWICZ, D. and ILL D. (1980). Extracellular matrix and control of
proliferation of vascular endothelial cells. J.Clianvest. 65, 1351-1364.

GOSPODAROWICZ, D. and LUI, GM. (1981). Effect of substrata and fibroblast
growth factor on the proliferation in vitro of bovine aortic endothelial cells.
J.CelLPhysial. 109, 69-81.

GOTTLIEB, AL, LANGILLE, B.L, WONG, M.KK and KIM, D.W. (1991).
Biology of disease: Structure and function of the endothelial cytoskeleton. Lab.Invest.
65, 123- 137.

GREENBURG, G.H. and HUNT, T.K (1978). The proliferative response in vitro
of vascular endothelial and smooth muscle cells exposed to wound fluid, and
macrophages. J.CelLPhysiaL 97, 353-360.

GRIFFIN, D.S. and SEGALL HJ. (1986). Genotoxicity and cytotoxicity of selected
pyrrolizidine alkaloids, a possible alkenal metabolite of the alkaloids, and related
alkenals. Toxicol. Appl. Phannacal. 86, 227-234.

GROSS, NJ. (1977). Pulmonary effects of radiation therapy. AnantemMed. 86,
81-92.

GROSS, NJ. (1981). The pathogenesis of radiation-induced lung damage. Lung 159,
115-125.

261

GROTENDORST, G.R., CHANG, T., SEPPA, H.EJ., KLEINMAN, H.K and
MARTIN, GR. (1982). Platelet-derived growth factor is a chemoattractant for
vascular smooth muscle cells. J.CelLPhysiaL 113, 261-266.

GRYGLEWSKI, RJ., BOTTIN G, RM. and VAN E, J .R. (1988). Mediators produced
by the endothelial cell. Hypertension 12, 530-548.

GUMKOWSKI, F., KAMIN SKA, M., KAMIN SKI, M., MORRISSEY, LW. and
AUERBACH, R. (1987). Heterogeneity of mouse vascular endothelium. Blood
Vessels 24, 11-23.

GUZOWSKI, DE. and SALGADO, ED. (1987). Changes in main pulmonary artery
of rats with monocrotaline-induced pulmonary hypertension. ArchPathoLLab.Med.
111, 741-745.

HABLIST ON, D.L, WHITAKER, C., HART, M.A, RYAN, US. and RYAN, J.W.
(1979). Isolation and culture of end0thelial cells from the lungs of small animals.
AmRev.Respir.Dis. 199, 853-868.

HACKER, A, CHRISTENSEN, C., EFFROS, RM. and REIN, D. (1991).
Monocrotaline, endothelial cells and polyamines. FASEB J. 5, A534.(Abstract)

HALES, C.A, SONNE, L and PETERSON, M. (1981). Role of thromboxane and
prostacyclin in pulmonary vasomotor changes after endotoxin in dogs. J.Clianvest.
68, 497-505.

HALES, C.A, KRADIN, R.L, BRANDSTETTER, RD. and ZHU, Y. (1983).
Impairment of hypoxic pulmonary artery remodeling by heparin in mice.
AmRev.Respir.Dis. 128, 747-751.

HARLAN, J.M., THOMPSON, PJ., ROSS, RR. and BOWEN-POPE, DP. (1986).
Alpha-thrombin induces release of platelet-derived growth factor-like molecule(s) by
cultured human endothelial cells. J.CelLBial 103, 1129-1133.

HARRIS, P., and HEATH, D. (1962). The Human Pulmonary Circulation. pp.
194-194.Livingstone, Edinburgh.

HARRIS, P.N., ANDERSON, R.C. and CHEN, KK (1942). The action of
monocrotaline and retronecine. .LPIzannacaLExp.Ther: 75, 78-82.

HATANO, S., and STRASSER, T. (1975). Primary Pulmonary Hypertension: Report
an a WHO Meeting. pp. 7-21.World Health Organization, Geneva.

HAYASHI, Y. (1966). Excretion and alteration of monocrotaline in rats after a
subcutaneous injection. FASEB J. 25, 688.(Abstract)

262

HAYASHI, Y. and LALICH, J .J . (1967). Renal and pulmonary alterations induced
in rats by a single injection of monocrotaline. PracSac.Exp.Biol.Med. 124, 392-396.

HAYASHI, Y., HUSSA, J.F. and LALICH, JJ. (1967). Cor pulmonale in rats.
Laanvest. 16, 875-881.

HAYASHI, Y., KATO, M. and orsuxA, H. (1979). inhibitory effects of
diet-reduction on monocrotaline intoxication in rats. TaxicaLLett. 3, 151-155.

HAYASHI, Y. and LALICH, J J. (1968). Protective effect of mercaptoethylamine
and cysteine against monocrotaline intoxication in rats. Toxicol. Appl. Phannacal. 12,
36-43.

HEATH, D. (1969). Pulmonary hypertension due to W. In The
Pulmonary Circulation and Interstitial Space (AP. Fishman, and H.H. Hecht, Eds.),

pp.305-319. Univ. of Chicago Press, Chicago.

HEATH, D. and KAY, J .M. (1967). The medial thickness of the pulmonary trunk in
rats with car puhnonale induced by ingestion of Crotalaria spectabilis seeds.
Cardiovascular Res. 1, 74-79.

HEATH, D., SMITH, P., GOSNEY, J., MULCAHY, D., FOX, K, YACOUB, M.
and HARRIS, P. (1987). The pathology of the early and late stages of primary
pulmonary hypertension. Br.I-Ieart J. 58, 204-213.

HEBY, O. (1981). Role of polyamines in the control of cell proliferation and
differentiation. Difi‘erentiatian 19, 1-20.

HEIDEN, D., MIELKE, CH. and RODVIEN, R. (1977). Impairment by heparin of
primary haemostasis and platelet [14C]5-hydroxytryptamine release. BrlHaematal.
36, 427-436.

HEIMARK, R.L, and SCHWARTZ, SM. (1988). Endothelial morphogenesis. In
Endothelial Cell Biology in Health and Disease (n. Simionescu, and m. Simionescu,
Eds.), pp.123-146. Plenum Press, New York.

HENNIG, B., BOISSONNEAULT, GA. and GLAUERT, HP. (1989). Effects of
serum type on growth and permeability properties of cultured endothelial cells.
Exp.Cell Res. 181, 589-596.

HERMAN, LM. (1990). Endothelial cell matrices modulate smooth muscle cell
growth, contractile phenotype and sensitivity to heparin. Haemastasis 20(suppl 1),
166-177.

263

HERMAN, LM. and CASTELLOT, JJ. (1987). Regulation of vascular smooth
muscle cell grth by endothelial-synthesized extracellular matrices. Arteriasclerasis
7, 463-469.

HILL KR., RHODES, K, STAFFORD, J .L and AUB, R. (1951). Liver disease in
Jamaican children (serous hepatosis). West Indian Med. 1, 49-63.

HILL N.S., O’BRIEN, R.F. and ROUNDS, S. (1984). Repeated lung injury due to
alpha-napthylthiourea causes right ventricular hypertrophy in rats. J.AppLPhysial. 56,
388-396.

HILL N.S., JEDERLINIC, P. and GAGNON, J. (1989). Supplemental oxygen
reduces right ventricular hypertrophy in monocrotaline-injected rats.Am.J.Physial. 66,
1642-1648.

HILLIKER, KS., BELL T.G. and ROTH, R.A (1982). Pneumotoxicity and
thrombocytopenia after a single injection of monocrotaline. AmlPhysial. 242,
H573-H579.

HILLIKER, KS., BELL T.G. and ROTH, R.A (1983a). Monocrotaline
pyrrole-induced pulmonary hypertension in fawn-hooded rats with platelet storage
pool deficiency: S-hydroxytryptamine uptake by isolated, perfused lungs. Thramb.
Haemastas. (Stuttgart) 50, 844—847.

HILLIKER, KS., DEYO, J .A, BELL T.G. and ROTH, R.A (1983b). Aggregation
of platelets from monocrotaline pyrrole-treated rats. Thromb.Res. 32, 325-333.

HILLIKER, KS., GARCIA, CM. and ROTH, R.A (1983c). Effects of
monocrotaline and monocrotaline pyrrole on 5-hydroxytryptamine and paraquat
uptake by lung slices. Res. Commun. ChemPathaLpharmacol. 40, 179-197.

HILLIKER, KS., BELL T.G., LORIMER, D. and ROTH, R.A (1984a). Effects of
thrombocytopenia on monocrotaline pyrrole-induced pulmonary hypertension.
AmJPhysiaL 246, H747-H753.

HILLIKER, KS. and ROTH, R.A. (1984b). Alteration of monocrotaline
pyrrole-induced cardiOpulmonary effects in rats by hydrallazine, dexamethasone or
sulphinpyrazone. BrJPharmacaL 82, 375-380.

HILLIKER, KS., IMLAY, M. and ROTH, R.A (1984). Effects of monocrotaline
treatment on norepinephrine removal by isolated, perfused rat lungs.
BiachemPharrnacaL 33, 2692-2695.

264

HILLIKER, KS. and ROTH, R.A. (1985a). Increased vascular responsiveness in
lungs of rats with pulmonary hypertension induced by monocrotaline pyrrole.
AmRev.Respir.Dis. 131, 46-50.

HILLIKER, KS. and ROTH, R.A. (1985b). Injury to the isolated, perfused lung by
exposure in vitro to monocrotaline pyrrole. Exp.Lung Res. 8, 201-212.

HINCKS, J.R., KIM, H.Y., SEGALL HJ., MOLYNEUX, RJ., STERMITZ, RR.
and COULDMBE, R.AJR. (1991). DNA cross-linking in mammalian cells by
pyrrolizidine alkaloids: Structure-activity relationships. Toxicol Appl. Pharmacol. 111,
90-98.

HINSHAW, D.B., BURGER, J .M., ARMSTRONG, BC. and HYSLOP, PA (1989).
Mechanism of endothelial cell shape change in oxidant injury.J.Surg.Res. 46, 339-349.

HISLOP, A and REID, L (1974). Arterial changes in Crotalaria spectabilis-induced
pulmonary hypertension in rats. Br.J.Exp.Path. 55, 153-162.

HOBSON, B. and DENEKAMP, J. (1984). Endothelial proliferation in tumours and
normal tissues: continuous labelling studies. Br.J.Cancer 49, 405-413.

HOGG, J.C. (1988). Primary pulmonary hypertension. Chest 93, 1728-1758.

HOLLINGER, MA, GIRI, S.N., PATWELL 8., ZUCKERMAN, J.B., GORIN, A
and PARSONS, G. (1980a). Effect of acute lung injury on angiotensin converting
enzyme in serum, lung lavage, and effusate. Am.Rev.Respir.Di.s. 121, 373-376.

HOLLINGER, M.A, PATWELL 8., ZUCKERMAN, J.B., GORIN, AB.,
PARSONS, G. and GIRI, 8.N. (1980b). Effect of paraquat on serum angiotensin
converting enzyme. AmRev.Respir.Dis. 121, 795-798.

HOLMSEN, H., DAY, HJ. and STORMOKEN, H. (1969). The blood platelet
release reaction. ScandIHaematal. Suppl. 8, 3-26.

HOOPER, P.T., BEST, S.M. and MURRAY, DR. (1974). Hyperammonaemia and
spongy degeneration of the brain in sheep affected with hepatic necrosis. Res. Vet.Sci.
16, 216-222.

HOOPER, RT. (1975). Experimental acute gastrointestinal disease caused by
pyrrolizidine alkaloid lasiocarpine. J.Camp.PathaL 85, 341-349.

HOORN, C.M., REINDEL J.F., WAGNER, J.G. and ROTH, R.A (1990).
Differences in response to MCTP of bovine and porcine vascular endothelium in
culture. The Taxicalagist 10, 33.(Abstract)

265

HOORN, CM. and ROTH, R.A. (1992). DNA, RNA and protein synthesis by
porcine pulmonary artery endothelial cells after treatment with monocrotaline
pyrrole. AmJPhysiol. (Lung CelLMaLPhysial. I) (In Press)

HOOSON, J. and GRASSO, P. (1976). Cytotoxic and carcinogenic response to
monocrotaline pyrrole. J.Path. 118, 121-129.

HOOVER, R.L, ROSENBERG, R.D., HAERING, W. and KARNOVSKY, MJ.
(1980). Inhibition of rat arterial smooth muscle cell proliferation by heparin. CircRes.
47, 578-583.

HOPKINS, N.K, OGLESBY, T.D., BUNDY, G.W. and GORMAN, RR. (1984).
Biosynthesis and metabolism of 15-hydroxyperoxy-5,8,l1,13-eicosatetraenoic acid by
human umbilical vein endothelial cells. J.Bial. Chem. 259, 14048-14058.

HOYER, LW., DE LOS SANTOS, RP and HOYER, JR. (1973). Antihemophilic
factor antigen. Localization in endothelial cells by immunofluorescent microscopy.
J.Clianvest. 52, 2737-2744.

HSIEH, HJ., Ll, NO. and FRANGOS, J.A (1991). Shear stress increases
endothelial platelet-derived growth factor mRNA levels. AmJ.Physial. 260,
H642-H646.

HSU, I.C., ALLEN, J .R. and CHESNEY, CF. (1973). Identification and toxicological
effects of dehydroretronecine, a metabolite of monocrotaline. Prac.Sac.Exp.BiaLMed.
144, 834-838.

HSU, I.C., SHUMAKER, R.C. and ALLEN, LR. (1974). Tissue distribution of
tritium-labeled dehydroretronecine. Chem.Bial.Interact. 8, 163-170.

HUBER, AR. and WEISS, SJ. (1989). Disruption of the subendothelial basement
membrane during neutrophil diapedesis in an in vitro construct of a blood vessel
wall. J.Clianvest. 83, 1122-1136.

HUGHES, J., GILLIS, C.N. and BLOOM, EB. (1969). The uptake and disposition
of d,1-norepinephrine in perfused rat lung. J.PharmacaLEprher. 169, 237-248.

HURLEY, J.V. and JAGO, M.V. (1975). Pulmonary oedema in rats given
dehydromonocrotaline: A topographic and electron-microscope study. J.Path. 117,
23-41.

HUX'T ABLE, R., PAPLANUS, S. and LAUGHARN, J. (1977). The prevention of
monocrotaline-induced right ventricular hypertrophy. Chest 718, 308-310.

266

HUXT ABLE, R., CIARAMITARO, D. and EISENSTEIN, D. (1978). The effect of
a pyrrolizidine alkaloid, monocrotaline, and a pyrrole, dehydroretronecine, on the
biochemical functions of the pulmonary endothelium. MoLPharmacal. 14, 1189-1203.

HUXTABLE, RJ., PAPLANUS, S. and LAUGHARN, J. (1977). The prevention of
monocrotaline-induced right ventricular hypertrophy. Chest 71S, 308-310.

HUXTABLE, RJ. (1979). New aspects of the toxicology and pharmacology of
pyrrolizidine alkaloids. GenPharmac. 10, 159-167.

HUXTABLE, RJ. (1980). Problems with pyrrolizidines. TIPS 1, 299-303.

HUXT ABLE, RJ. (1989). Human health implications of pyrrolizidine alkaloids and
herbs containing them. In Taxicants of Plant Origin: Vol.1 Alkaloids (P.R. Cheeke,
Ed.), pp.41-86. CRC Press, Boca Raton, FL.

HUXTABLE, RJ. (1990a). Activation and pulmonary toxicity of pyrrolizidine
alkaloids. Pharmac. Ther. 47, 371-389.

HUXTABLE, RJ., BOWERS, R., MATTOCKS, AR. and MICHNICKA, M.
(1990b). Sulfur conjugates as putative pneumotoxic metabolites of the pyrrolizidine
alkaloid, monocrotaline. Adv.Exp.MedBioL 283, 605-612.

HYERS, T.M. and FOWLER, AA. (1986). Adult respiratory distress syndrome:
causes, morbidity, and mortality. FedPrac. 45, 25-29.

IBE, 8.0. and CAMPBELL W.B. (1988). Synthesis and metabolism of leukotrienes
by human endothelial cells: influence on prostacyclin release. Biachim. Biophys. Acta
960, 309-321.

ILKIW, R., TODOROVICH-HUNTER, L, MARUYAMA K, SHIN, J. and
RABINOVITCH, M. (1989). SC-39026, a serine elastase inhibitor, prevents
muscularization of peripheral arteries, suggesting a mechanism of
monocrotaline-induced pulmonary hypertension in rats. Circ.Res. 64, 814-825.

INGERMAN-WOJENSKI, C., SILVER, MJ., SMITH, J.B. and MACARAK, E.
(1981). Bovine endothelial cells in culture produce thromboxane as well as
prostacyclin. J.Clianvest. 67, 1292-1296.

ITO, K, NAKASHIMA, T., MURAKAMI, K. and MURAKAMI, T. (1988a). Altered
function of pulmonary endothelium following monocrotaline-induced lung vascular
injury in rats. Br.J.Pharmacal. 94, 1175-1183.

267

ITO, K, HAYASHI, K, NAKASHIMA T. and MURAKAMI, T. (1988b). Alteration
of fate of vasoactive autocoids in pulmonary circulation following
monocrotaline-induced lung vascular injury in rats. Br.J.Pharrnacal. 94, 1167-1174.

IWASAWA, Y., GILLIS, C.N. and AGHAJANIAN, G. (1973). Hypothermic
inhibition of 5-hydroxytryptamine and norepinephrine uptake by lung: cellular
localization of arrrines after uptake. .LPhannacaLEprher. 186, 498-507.

IYER, V.N. and SZYBALSKI, W. (1963). A molecular mechanism of mitomycin
action: Linking of complementary DNA strands. Microbiology 50, 355-362.

JAFFE, E.A., HOYER, LW. and NACHMAN, R.L (1974). Synthesis of von
Willebrand factor by cultured human endothelial cells. PraaNatLAcadSci. USA 71,
1906-1909.

JAFFE, E.A, MINICK, R., ADELMAN, B., BECKER, CG. and NACHMAN, R.
(1976). Synthesis of basement membrane collagen by cultured human endothelial
cells. J.Bxp.Med. 144, 209-225.

JAFFE, EA and MOSHER, DE (1978). Synthesis of fibronectin by cultured human
endothelial cells. J.Exp.Med 147, 1779-1791.

JAFFE, EA, NACHMAN, R.L, BECKER, CG. and MINICK, CR. (1973). Culture
from human endothelial cells derived from umbilical veins. Identification by
morphologic and immunologic criteria. J.Clianvest. 52, 2745-2756.

JAFFE, EA (1987). Cell biology of endothelial cells. HumPathal. 18, 234-239.

JAGO, M.V. (1969). The development of hepatic megalocytosis of chronic
pyrrolizidine alkaloid poisoning. Amen/.Pathal. 56, 405-422.

JAMES, H.L, JAMES, P.L, PAINTER, R.G., ZAHLER-BENT‘Z, K and COHEN,
AB. (1986). Subcellular localization of platelet elastase and its retention during the
release reaction. Semin. Thramb.Hemast. 12, 250-252.

JEDERLINIC, P., HILL N.S., OU, LC. and FANBURG, BL (1988). Lung
angiotensin converting enzyme activity in rats with differing susceptibilities to chronic
hypoxia. Thorax 43, 703-707.

JOHNSON, AR. and ERDOS, E.G. (1977). Metabolism of vasoactive peptides by
human endothelial cells in culture. Angiotensin I converting enzyme (kininase II) and
angiotensinase. J. Clin.1nvest. 59, 684-695.

268

JOHNSTON, G.I., COOK, R.G. and MCEVER, R.F. (1989). Cloning of GMP-140,
a granule membrane protein of platelets and endothelium: sequence similarity to
proteins involved in cell adhesion and inflammation. Cell 56, 1033-1044.

JOLIVET, J., GIROUX, L, LAURIN, 8., GRUBER, J., BETTEZ, P. and BAND,
RR. (1983). Microangiopathic hemolytic edema, renal failure and noncardiogenic
pulmonary edema: a chemotherapy-induced syndrome. Cancer Treat.Rep. 67, 429-434.

JONES, R., ZAPOL W.M. and REID, L (1983). Pulmonary arterial wall injury and
remodelling by hyperoxia. Chest 83, 408-428.

KAJIHARA, H. (1970). Electron microscopic observation of hypertrophied
myocardium of rat produced by injection of monocrotaline. ActaPathJap. 20,
183-206.

KATUSIC, 2.8. and SHEPHERD, J .T. (1991). Endothelium-derived vasoactive
factors: II. Endothelium-dependent contraction. Hypertension l8[suppl III],
III-86-III-92.

KAY, J.M. and HEATH, D. (1966). Observations on the pulmonary arteries and
heart weight of rats fed on Crotalaria spectabilis seeds.J.PathaLBacterial. 92, 385-394.

KAY, J .M., HARRIS, P. and HEATH, D. (1967a). Pulmonary hypertension produced
in rats by ingestion of Crotalaria spegtabilis seeds. Thorax 22, 176-179.

KAY, J .M., GILLUN D, TD. and HEATH, D. (1967b). Mast cells in the lungs of rats
fed on Crotalaria spectabilis seeds. Amen/.Pathal. 51, 1031-1042.

KAY, J .M., and HEATH, D. (1969). Crotalaria spectabilis, the Pulmonary
Hypertension Plant. Charles C. Thomas, Springfield, IL

KAY, J.M., SMITH, P. and HEATH, D. (1969). Electron microscopy of Crotalaria
pulmonary hypertension. Thorax 24, 511-526.

KAY, J.M., KEANE, P.M., SUYAMA KL and GAUTHIER, D. (19823).
Angiotensin converting enzyme activity and evolution of pulmonary vascular disease
in rats with monocrotaline pulmonary hypertension. Thorax 37, 88-96.

KAY, J.M., KEANE, P.M., SUYAMA KL. and GAUTHIER, D. (1982b).
Angiotensin converting enzyme activity and evaluation of pulmonary vascular disease
in rats with monocrotaline pulmonary hypertension. Thorax 37, 88-96.

KAY, J.M., SUYAMA, KL. and KEANE, P.M. (1982c). Failure to show decrease
in small pulmonary blood vessels in rats with experimental pulmonary hypertension.
Thorax 37, 927-930.

269

KAY, J.M., KEANE, RM. and SUYAMA KL (1985). Pulmonary hypertension
induced in rats by monocrotaline and chronic hypoxia is reduced by
p-chlorophenylalanine. Respiration 47, 48-56.

KEANE, P.M., KAY, J.M., SUYAMA KL, GAUTHIER, D. and ANDREW, K
(1982). Lung angiotensin converting enzyme activity in rats with pulmonary
hypertension. Thorax 37, 198-204.

KEANE, RM. and KAY, J.M. (1984). Lung angiotensin converting enzyme activity
in monocrotaline pulmonary hypertension (letter). Thorax 39, 159-160.

KELLEY, PM. and SCHLESINGER, MJ. (1978). The effect of amino acid
analogues and heat shock on gene expression in chicken embryo fibroblasts. Cell 15,
1277-1286.

KEYES, S.R., F RACASSO, P.M., HEIMBROOK, D.C., ROCKWELL 8., SLIGAR,
8.G. and SARTORELLI, AC. (1984) Role of NADPHzcytochrome c reductase and
DT-diaphorase in the biotransforrnation of mitomycin C. Cancer Res. 44, 5638-5643.

KEYES, S.R., ROCKWELL S. and SARTORELLI, AC. (1985) Enhancement of
mitomycin C cytotoxicity to hypoxic cells by dicumerol in viva and in vitro. Cancer
Res. 45, 213-216.

KING, 8J., BOOYSE, P.M., LIN, P., TRAYLOR, M., NARKATES, AJ. and
OPARIL, S. (1989). Hypoxia stimulates endothelial cell angiotensin-converting
enzyme antigen synthesis. Am.J.Physial. 256(Cell Physiol.25), C1231-C1238.

KOHN, KW., EWIG, R.A, ERICKSON, LC., and ZWELLING, LA (1981).
Measurement of strand breaks and cross-links in DNA by alkaline elution. In DNA
Repair: A Laboratory Manual of Research Procedures, Vol I, Part B (E. F riedberg, and
P. Hanawalt, Eds.), pp.379-402. Marcel Dekker, Inc., New York.

KOUREMBANAS, 8., HAN NAN, R.L. and FALLER, D.V. (1990). Oxygen tension
regulates the expression of the platelet-derived growth factor-B chain gene in human
endothelial cells. J.Clianvest. 86, 670-674.

KRAMER, R.H., BENSCH, KG., DAVISON, RM. and KARASEK, M.A (1984).
Basal lamina formation by cultured microvascular endothelial cell monolayers.
J.CelLBiaI. 99, 692-698.

KUMAR, 8., WEST, DC. and AGER, A. (1987). Heterogeneity in endothelial cells
from large vessels and rnicrovessels. Differentiation 36, 57-70.

LABARCA C. and PAIGEN, K. (1980). A simple, rapid, and sensitive DNA assay
procedure. AnalBiachem. 102, 344-352.

270

LAFRANCONI, W.M., DUHAMEL R.C., BRENDEL K. and HUXTABLE, RJ.
(1984). Differentiation of the cardiac and pulmonary toxicity of monocr0taline, a
pyrrolizidine alkaloid. Biachem.PhannacaL 33, 191-197.

LAFRANCONI, W.M. and HUXT ABLE, RJ. (1983). Changes in
angiotensin-converting enzyme activity in lungs damaged by the pyrrolizidine alkaloid
monocrotaline. Tharwr 38, 307-309.

LAFRANCONI, W.M. and HUXTABLE, RJ. (1984). Hepatic metabolism and
pulmonary toxicity of monocrotaline using isolated perfused liver and lung.
BiochemPharmacal. 33, 2479-2484.

LAFRANCONI, W.M., OHKUMA 8. and HUXTABLE, RJ. (1985). Biliary
excretion of novel pneumotoxic metabolites of the pyrrolizidine alkaloid,
monocrotaline. Taxican 23, 983-992.

LAI, Y.L, OLSON, J.W. and GILLESPIE, MN. (1991). Ventilatory dysfunction
precedes pulmonary vascular changes in monocrotaline-treated rats. JAppLPhysial.
70, 561-566.

LALICH, JJ. (1964). Influence of rat strain on pulmonary vascular responses to
monocrotaline feeding. PathaLMicrabial 27, 965-73.

LALICH, JJ. and EHRHART, LA (1962). Monocrotaline induced pulmonary
arteritis in rats. J.Atherasclerasis Res. 2, 482-492.

LALICH, J J ., JOHNSON, W.D., RACZNIAK TJ. and SHUMAKER, R.C. (1977).
Fibrin thrombosis in monocrotaline pyrrole-induced cor pulmonale in rats.
ArchPathoLLab.Med. 101, 69-73.

LALICH, JJ. and MERKOW, L. (1961). Pulmonary arteritis produced in rats by
feeding Crotalaria spectabilis. Laanvest. 10, 744-750.

LAME, M.W., MORIN, D., JONES, AD., SEGALL HJ. and WILSON, D.W.
(1990). Isolation and identification of a pyrrolic glutathione conjugate metabolite of
the pyrrolizidine alkaloid monocrotaline. TaxicaLLett. 51, 321-329.

LAME, M.W., JONES, AD., MORIN, D. and SEGALL HJ. (1991). Metabolism
of [14C]monocrotaline by isolated perfused rat liver. Drug Metab.Dispas. 19, 516-524.

LAMPUGNANI, M.G., RESNATI, M., DEHANA, E. and MARCHISIO, RC.
(1991). The role of integrins in the maintenance of endothelial monolayer integrity.
J.CelLBiaL 112, 479-490.

271

LANGLEBEN, D. and REID, LM. (1985). Effect of methylprednisolone on
monocrotaline-induced pulmonary vascular disease and right ventricular hypertrophy.
Laanvest. 52, 298-303.

LANGLEBEN, D., CARVALHO, AC.A and REID, LM. (1986). The platelet
thromboxane inhibitor, dazmegrel, does not reduce monocrotaline-induced pulmonary
hypertension. AmRev.Respir.Dis. 133, 789-791.

LAN GLEBEN, D., SZAREK, J .L, COFLESKY, J .T., J ON ES, R.C., REID, LM. and
EVANS, J .N. (1988). Altered artery mechanics and structure in monocrotaline
pulmonary hypertension. J.AppLPhysial. 65, 2326-2331.

LANSMAN, J.B. (1988). Endothelial mechanosensors. Going with the flow. Nature
(Land) 331, 481-482.

LARSON, D.M. and SHERIDAN, JD. (1982). Intercellular junctions and transfer
of small molecules in primary vascular endothelial cells. J.CelLBial. 92, 183-191.

LAW, MP. (1981). Radiation-induced vascular injury and its relation to late effects
in normal tissues. Adv.Radiat. 9, 37-73.

LELKES, RI; (1991). New aspects of endothelial cell biology. J.CelLBiochem. 45,
242-244.

LEVIN, E.G. and LOSKUTOFF, DJ. (1982). Cultured bovine endothelial cells
produce both urokinase and tissue-type plasminogen activators. J.CelLBial. 94,
631-636.

LEWIS, LJ., HOAK, J.C., MACA RD. and FRY, G.L (1973). Replication of
human endothelial cells in culture. Science 181, 453-454.

LIBBY, P., ORDOVAS, J .M., AUGER, KR., ROBBINS, AH., BIRINYI, LK and
DINNARELLO, CA (1986). Endotoxin and tumor necrosis factor induce

interleukin-1 gene expression in adult human vascular endothelial cells.
AmerlPathol. 124, 179-185.

LOEWENSTEIN, W.R. (1979). Junctional intercellular communication and the
control of growth. Biachim. Biophys. Acta 560, 1-65.

LONGENECKER, G.L. (1985). Platelet arachidonic acid metabolism. In The
Platelets: Physiology and Pharmacology (G.L. Longenecker, Ed.), pp.49-83. Academic
Press, New York.

LOSKUTOFF, DJ. and EDGINGTON, TS. (1977). Synthesis of a fibrinolytic
activator and inhibitor by endothelial cells. Prac.NatLAcadSci. USA 74, 3903-3907.

272

LOSKUTOFF, DJ ., VAN MOURIK J.A, ERICKSON, LA. and LAWRENCE, D.
(1983). Detection of an unusually stable fibrinolytic inhibitor produced by bovine
endothelial cells. Prac.NatLAcad.Sci. USA 80, 2956-2960.

LUDMER, P.L, SELWYN, AP., SHOOK, TL and ET AL (1986). Paradoxical
vasoconstriction induced by acetylcholine in atherosclerotic coronary arteries.
N.Eng.J.Med 315, 1046-1051.

LUFT, J .H. (1966). Fine structure of capillary and endocapillary layer as revealed by
ruthenium red. FedProc. 25, 1173-1183.

MACARAK, EJ. and HOWARD, RS. (1983). Adhesion of endothelial cells to
extracellular matrix proteins. J.CellPhysial. 116, 76-86.

MADHUKAR, B.V., OH, 8.Y., CHANG, C.C., WADE, M. and TROSKO, J.B.
(1989). Altered regulation of intercellular communication by epidermal growth factor,
transforming growth factor-beta and peptide hormones in normal human
keratinocytes. Carcinogenesis 10, 13-20.

MADRI, J.A and STENN, KS. (1982). Aortic endothelial cell migration. 1. Matrix
requirements and composition. AmerlPathal 106, 180-186.

MADRI, J.A and WILLIAMS, SK (1983). Capillary endothelial cell cultures:
Phenotypic modulation by matrix components. J.CelLBiaL 97, 153-165.

MADRI, J.A, BELL L, MARX, M., MERWIN, J.R., BASSON, C. and PRINZ, C.
(1991). Effects of soluble factors and extracellular matrix components on vascular cell
behavior in vitro and in vivo: models of de-endothelialization and repair.
J.CelLBiachem. 45, 123-130.

MADRI, J.A and PRATT, BM. (1986). Endothelial cell-matrix interactions: in vitro
models of angiogenesis. J.Histachem.Cytachem. 34, 85-91.

MAINARDI, C.L, DIXIT, SN. and KANG, AH. (1980). Degradation of type IV '
(basement membrane) collagen by a proteinase isolated from human
polymorphonuclear leukocyte granules. J.Bial Chem. 255, 5435-5441.

MAJNO, G., GILMORE, V. and LEVENTHAL M. (1967). On the mechanism of
vascular leakage caused by histamine-type mediators. Circ.Res. 21, 933-837.

MARLIN, 8D. and SPRINGER, T.A. (1987). Purified intercellular adhesion
molecule-1 (ICAM-1) is a ligand for lymphocyte function-associated antigen 1
(LFA-l). Cell 51, 813-819.

273

MARUYAMA Y. (1963). The human endothelial cell tissue culture.
ZZellfarschMikraskAnat. 60, 69.

MASUGI, Y., OAMI, H., AIHARA, K, HASHIMOTO, K and JHAKOZAKI, T.
(1965). Renal and puhnonary vascular changes induced by W in
rats. Acta PathJap. 15, 407-415.

MATHEW, R., GLOSTER, E.S., ALTURA ET. and ALTURA, BM. (1988).
Magnesium aspartate hydrochloride attenuates monocrotaline-induced pulmonary
artery hypertension in rats. Clinical Science 75, 661-667.

MATHEW, R., GUZOWSKI, DE and GLOSTER, ES. (1990). Effect of verapamil
on monocrotaline-induced pulmonary artery hypertension and endothelial cell
dysfunction in rats. Exp.Lung Res. 16, 627-644.

MATTOCKS, AR. (1968). Toxicity of pyrrolizidine alkaloids. Nature (Land) 217,
723-728. -

MATTOCKS, AR. (1969). Dihydropyrrolizine derivatives from unsaturated
pyrrolizidine alkaloids. J.Chem.Sac. C, 1155-1162.

MATTOCKS, AR. (1972). Acute hepatotoxicity of pyrrolic metabolites in rats dosed
with pyrrolizidine alkaloids. Chem.Bial.Interact. 5, 227-242.

MATTOCKS, AR. and BIRD, I. (1983a). Pyrrolic and N-oxide metabolites formed
from pyrrolizidine alkaloids by hepatic rnicrosomes in vitro: Relevance to in vivo
hepatotoxicity. Chem.BiaL Interact. 43, 209-222.

MATTOCKS, AR. and BIRD, I. (1983b). Alkylation by dehydroretronecine, a
cytotoxic metabolite of some pyrrolizidine alkaloids: An in vitro test. TaxicaLLett. 16,
1-8.

MATTOCKS, AR. and CABRAL J.R.P. (1982). Carcinogenicity of some pyrrolic
pyrrolizidine alkaloid metabolites and analogues. Cancer Letters 17, 61-66.

MATTOCKS, AR., DRIVER, E., BARBOUR, RH. and ROBINS, DJ. (1986).
Metabolism and toxicity of synthetic analogues of macrocyclic diester pyrrolizidine
alkaloids. Chem.Bial.Interact. 58, 95-108.

MATTOCKS, AR. and LEGG, R.F. (1980). Antimitotic activity of
dehydroretronecine, a pyrrolizidine alkaloid metabolite, and some analogous
compounds, in a rat liver parenchymal cell line. Chem.Bial.Interact. 30, 325-336.

274

MATTOCKS, AR. and WHITE, I.N.H. (1971). The conversion of pyrrolizidine
alkaloids to N-oxides and to dihydropyrrolizidine derivatives by rat liver microsomes
in vitro. Chem.BiaLInteract. 3, 383-396.

MATTOCKS, AR. (1986). Chemistry and Toxicology of Pyrrolizidine Alkaloids.
Academic Press, London.

MATTOCKS, AR. and DRIVER, HE. (1983). A comparison of the pneumotoxicity
of some pyrrolic esters and similar compounds analogous to pyrrolizidine alkaloid
metabolites, given intravenously to rats. Toxicology 27, 159-177.

MATTOCKS, AR. and JUKES, R. (1990). Recovery of the pyrrolic nucleus of
pyrrolizidine alkaloid metabolites from sulphur conjugates in tissues and body fluids.
Chem.Bial.Interact. 75, 225-239.

MATTOCKS, AR. and WHITE, I.N.H. (1970). Estimation of metabolites of
pyrrolizidine alkaloids in animal tissues. AnalBiachem. 38, 529-535.

MATTOCKS, AR., LEGG, R.F. and JUKES, R. (1990). Trapping of short-lived
electr0philic metabolites of pyrrolizidine alkaloids escaping from perfused rat liver.
TaxicaLLett. 54, 93-99.

MAYNARD, J .R., DREYER, E.E., STEMERMAN, MB. and PITIICK, RA (1977).
Tissue factor coagulant activity of cultured human endothelial and smooth muscle
cells and fibroblasts. Blood 50, 387-396.

MCCARTHY, J.T. and STAATS, BA (1986). Pulmonary hypertension, hemolytic
anemia, and renal failure. A mitomycin-associated syndrome. Chest 89, 608-611.

MCLEAN, E.K (1970). The toxic actions of pyrrolizidine alkaloids. PharmacolRev.
22, 429-483.

MCNEIL, P.L, MUTHUDRISHNAN, L, WARDER, E. and D’AMORE, PA
(1989). Growth factors are released by mechanically wounded endothelial cells.
J.CelLBiaL 109, 811-822.

MERKOW, L and KLEINERMAN, J. (1966a). An electron microscopic study of
pulmonary vasculitis induced by monocrotaline. Laanvest. 15, 547-564.

MERKOW, L and KLEINERMAN, J. (1966b). An electron microscopic study of
pulmonary vasculitis induced by monocrotaline. Laanvest. 12, 547-564.

MEYN, RE. and MURRAY, D. (1984). Cell cycle effects of alkylating agents.
Pharmacol. Ther. 24, 147-163.

275

MEYRICK, B. and BRIGHAM, KL (1983). Acute effects of Escherichia coli
endotoxin on the pulmonary microcirculation of anesthetized sheep. Lab.Inve.st. 48,
458-470.

MEYRICK, B. and BRIGHAM, KL. (1986). Repeated Escherichia coli
endotoxin-induced pulmonary inflammation causes chronic pulmonary hypertension
in sheep. Lab.Invest. 55, 164-176.

MEYRICK, B., PERKETT, EA and BRIGHAM, KL (1987). Inflammation and
models of chronic pulmonary hypertension. Am.Rev.Respir.Dis. 136, 767-769.

MEYRICK B., HOOVER, R., JONES, M.R., BERRY, LCJR. and BRIGHAM,
KL (1989). In vitro effects of endotoxin on bovine and sheep lung microvascular and
pulmonary artery endothelial cells. J.CelLPhysiaL 138, 165-174.

MEYRICK, B. and REID, L (1978). The effect of continued hypoxia on rat
pulmonary arterial circulation. Lab.Invest. 38, 188-200.

MEYRICK, B. and REID, L (1979). Development of pulmonary arterial changes in
rats fed Crotalaria spectabilis. AmerJPathal. 94, 37-50.

MEYRICK, B., GAMBLE, W. and REID, L (1980). Development of Crotalaria,
pulmonary hypertension: hemodynamic and structural study. Am.J.Physial. 239,
H692-H702.

MEYRICK, B0. (1986). Endotoxin-mediated pulmonary endothelial cell injury.
FedPrac. 45, 19-24. .

MEYRICK, B.O. and REID, LM. (1982). Crotalaria-induced pulmonary
hypertension. Uptake of 3H-thymidine by the cells of the pulmonary circulation and
alveolar walls. AmerJPathal. 106, 84-94.

MILLER, W.C., RICE, D.L, KREUSEL R.G. and BEDROSSIAN, C.W.M. (1978).
Monocrotaline model of noncardiogenic pulmonary edema in dogs. J.AppLPhysial. 45,
962-965.

MIOSSEC, P., CAVENDER, D. and ZIFF, M. (1986). Production of interleukin 1
by human endothelial cells. J.Immunal. 136, 2486-2491.

MIRANDA, C.L, CARPENTER, H.M., CHEEKE, RR. and BUHLER, D.R.
(1981a). Effect of ethoxyquin on the toxicity of the pyrrolizidine alkaloid

monocrotaline and on hepatic drug metabolism in mice. Chem.Bial.Interact. 37,
95-107.

276

MIRANDA, C.L, REED, R.L., CHEEKE, RR. and BUHLER, D.R. (1981b).
Protective effects of butylated hydroxyanisole against the acute toxicity of
monocrotaline in mice. Taxical Appl. Phamtacal. 59, 424-430.

MIZEL S.B. (1989). The interleukins. FASEB J. 3, 2379-2388.

MODAT, G., DORNAND, J., BERNAD, N., JUNQUERO, D., MARY, A,
MULLER, A and BONNE, C. (1990). LPS-stimulated bovine aortic endothelial cells
produce IL-1 and IL-6 like activities. Agents Actions 30, 403-411.

MOLONY, L and ARMSTRONG, L (1991). Cytoskeletal reorganization in human
umbilical vein endothelial cells as a result of cytokine exposure. Etp.Cell Res. 196,
40-48.

MOLTENI, A, WARD, W.F., TS’AO, C.H., PORT, CD. and SOLLIDAY, NH.
(1984). Monocrotaline-induced pulmonary endothelial dysfunction in rats.
PraaSacExpBiaLMed 176, 88-94.

MOLTENI, A, WARD, W.F., TS’AO, C.H., SOLLIDAY, NH. and DUNNE, M.
(1985). Monocrotaline-induced pulmonary fibrosis in rats: amelioration by captopril
and penicillarrrine (42151). PracSac.Exp.Bial.Med. 180, 112-120.

MOLTENI, A, WARD, W.F., TS’AO, CH. and SOLLIDAY, NH. (1986).
Monocrotaline-induced cardiopulmonary damage in rats: amelioration by the
angiotensin-converting enzyme inhibitor CL242817 (42370). Proc.Sac.Exp.Bial.Med
182, 483-493.

MOLTENI, A, WARD, W., TS’AO, C. and SOLLIDAY, N. (1987).
Monocrotaline-induced cardiopulmonary injury in rats: Modification by thiol and
nonthiol ACE inhibitors. ClirLExper. Theory and Practice A9, 381-385.

MOLTENI, A, WARD, W.F., TS’AO, CH. and SOLLIDAY, N.H. (1988a).
Monocrotaline-induced cardiopulmonary injury in rats: modification by the nonthiol
ACE inhibitors CGSIB945 and CG816617. Arch.lnt.Pharmaca.Ther. 291, 21-40.

MOLTENI, A, WARD, W.F., TS’AO, CH. and FITZSIMONS, EJ. (1988b). Serum
copper concentration as an index of cardiopulmonary injury in monocrotaline-treated
rats. AnnClinLabSci. 18, 476-483.

MOLTENI, A, WARD, W.F., TS’AO, CH. and SOLLIDAY, N.H. (1989a).
Monocrotaline pneumotoxicity in mice. Vrrchaws Archiv. B Cell.Pathal. 57, 149-155.

MOLTENI, A, WARD, W.F., TS’AO, CH. and HINZ, J.M. (1989b).
Monocrotaline-induced cardiopulmonary injury in rats: Modification by the
neutrophil elastase inhibitor SC39026. BiochemPharmacaL 38, 2411-2419.

277

MONCADA 8., HERMAN, AG., HIGGS, EA and VANE, JR. (1977).
Differential formation of prostacyclin (PGX or PG12) by layers of the arterial wall.
An explanation for the anti-thrombotic properties of vascular endothelium.
Thramb.Res. 11, 323-344.

MONKS, TJ., HANZLIK, R.P., COHEN, G.M., ROSS, D. and GRAHAM, D.G.
(1992). Quinone chemistry and toxicity. TaxicaL Appl. Pharmacol. 112, 2-16.

MOREL D.R., DARGENT, K, BACHMANN, M., SUTER, RM. and JUNROD,
AF. (1985). Pulmonary extraction of serotonin and propranalol in patients with adult
respiratory distress syndrome. Am.Rev.Respir.Dis. 132, 479-484.

MOREL D.W., DICORLETO, RE. and CHISOLM, GM. (1984). Endothelial and
smooth muscle cells alter low density lipoprotein in vitro by free radical oxidation.
Arteriasclerasis 4, 357-364.

MORGANROTH, M.L, REEVES, J.T. and MURPHY, R.C. (1984). Leukotriene
synthesis and receptor blockers block hypoxic pulmonary vasoconstriction.
J.AppLPhysial. 56, 1340-1346.

MURRAY, D. and MEYN, RE. (1986). Cell cycle-dependent cytotoxicity of
alkylating agents: determination of nitrogen mustard-induced DNA cross-links and
their repair in Chinese hamster ovary cells synchronized by centrifugal elutriation.
Cancer Res. 46, 2324-2329.

NAWROTH, P., KISIEL W. and STERN, D. (1985). The role of endothelium in the
homeostatic balance of haemostasis. Clinics in Haematology 14, 531-546.

NEAL W.M., RUSOFF, LL and AHMANN, CF. (1935). The isolation and some
properties of an alkaloid from Crotalaria spectabilis Roth. J.Arn. ChemSac. 57,
2560-2561.

NEMECEK, G.M., COUGHLIN, S.R., HAN DLEY, DA and MOSKOWITZ, M.A
(1986). Stimulation of aortic smooth muscle cell mitogenesis by seratonin.
PracNatLAcadSci. USA 83, 674-678.

NEWBERNE, P.N., and ROGERS, AE. (1973). Nutrition, monocrotaline and
aflatoxicn B1 in liver carcinogenesis. In Plant Foods for Man (P.N. Newman, Ed.),
pp.23-31. , .

NEWMAN, PJ., BERNDT, M.C., GORSKI, J., WHITE, G.C.lI., LYMAN, 8.,
PADDOCK C. and MULLER, WA. (1990). PECAM-l (CD31) cloning and relation
to adhesion molecules of the immunoglobulin gene superfamily. Science 247,
1219-1222.

278

NIEWIAROWSKI, 8., and HOLT, J.C. (1985). Platelet alpha-granule proteins:
Biochemical and pathological aspects. In The Platelets: Physiology and Pharmacology
(G.L Longenecker, Ed.), pp.49-83. Academic Press, New York.

NIWA, H., OGAWA, T., OKAMOTO, O. and YAMADA, K. (1991). Alkylation of
uucleosides by dehydromonocrotaline, the putative toxic metabolite of the
carcinogenic pyrrolizidine alkaloid monocrotaline. Tetrahedron Let. 32, 927-930.

NOJI, 8., TASHIRO, K, KOYAMA E., N OHNO, T., OHYAMA, K, TAN IGUCI-II,
8. and NAKAMURA, T. (1990). Expression of hepatocyte growth factor gene in
endothelial and kupffer cells of damaged rat livers, as revealed by m
hybridization. BiachemBiaphys.Res.Cammun. 173, 42-47.

OGAWA 8., SHREENIWAS, R., BUTURA, C., BRETT, J. and STERN, D.M.
(1990a). Modulation of endothelial function by hypoxia: Perturbation of barrier and

anticoagulant function, and induction of a novel factor X activator. Adv.£xp.MedBial.
281, 303-312.

OGAWA, 8., GERLACH, H., ESPOSITO, C., PASAGIAN-MACAULAY, A,
BRETT, J. and STERN, D. (1990b). Hypoxia modulates the barrier and coagulant
function of cultured bovine endothelium: Increased monolayer permeability and
induction of procoagulent properties. J.Clianvest. 85, 1090-1098.

OHAR, J.A, WALLER, KS., DEMELLO, D. and LAGUNOFF, D. (1991).
Administration of chronic intravenous platelet-activating factor induces pulmonary
arterial atr0phy and hypertension in rabbits. Laanvest. 65, 451-458.

OKA, Y. and ORTH, D.N. (1983). Human plasma epidermal growth
factor/beta-orogastrone is associated with blood platelets. J. Clianvest. 72, 249-259.

OLESEN, S.P., CHAPMAN, DE. and DAVIES, PF. (1988). Hemodynamic shear
stress activates a K+ current in vascular endothelial cells. Nature (Land) 331,
168-170.

OLSON, J.W., HACKER, AD., ALTIERE, RJ. and GILLESPIE, M.N. (1984a).
Polyamines and the development of monocrotaline-induced pulmonary hypertension.
AmJ.PhysiaL 247, H682-H685.

OLSON, J.W., ALTIERE, RJ. and GILLESPIE, M.N. (1984b). Prolonged activation
of rat lung ornithine decarboxylase in monocrotaline-induced puhnonary
hypertension. BiochemPharmacol. 33, 3633-3637.

279

OLSON, J .W., ATKINSON, J .E., HACKER, AD., ALTIERE, RJ. and GILLESPIE,
MN. (1985). Suppression of polyamine biosynthesis prevents monocrotaline-induced
pulmonary edema and arterial medial thickening. Toxicol. Appl. Pharmacol. 81, 91-99.

OLSON, J.W., ORLINSKA U. and GILLESPIE, MN. (1989). Polyamine synthesis
blockade in monocrotaline-induced pneumotoxicity. BiochemPhannacal. 38,
2903-2910.

OMINI, C., FOLEO, CC. and VIGANO, T. (1982). Leukotriene-C4 induces
generation of PGIZ and TXAZ in guinea pig in vivo. Pharm.Res.Cammun. 13,
633-640.

ONO, S. and VOELKEL, NP. (1991). PAF antagonists inhibit
monocrotaline-induced lung injury and pulmonary hypertension. J.AppLPhysial. 71,
2483-2492.

OPARIL 8., NARKATES, AJ., JACKSON, R. and ANN, HS. (1988). Altered
angiotensin-converting enzyme in lung and extrapulmonary tissues of hypoxia-adapted
rats. J.AppLPhysial. 65, 218-227.

ORLINSKA, U., OLSON, J .W. and GILLESPIE, MN. (1988). Polyamine content in
pulmonary arteries from rats with monocrotaline-induced pulmonary hypertension.
Res. Commun. ChemPathaLpharmacal. 62, 187-194.

ORLINSKA U., OLSON, J.W., GEBB, SA. and GILLESPIE, MN. (1989).
Acetylated polyamines in lungs from rats with monocrotaline-induced pneumotoxicity.
FundamAppl. Taxical. 13, 277-284.

OSBORN, L, HESSION, C., TIZARD, R., VASSALIO, C., LUHOWSKYL 8.,
CHI-ROSSO, G. and LOBB, R. (1989). Direct expression cloning of vascular cell
adhesion molecule 1, a cytokine-induced endothelial protein that binds to
lymphocytes. Cell 59, 1203-1211.

PALMER, R.MJ., FERRIGE, AG. and MON CADA 8. (1987). Nitric oxide release
accounts for the biological activity of endothelium-derived relaxing factor. Nature
(Land) 327, 524-526.

PAN, LC., LAME, M.W., MORIN, D., WILSON, D.W. and SEGALL HJ. (1991).
Red blood cells augment transport of reactive metabolites of monocrotaline from
liver to lung in isolated and tandem liver and lung preparations. Toxicol. Appl.
Pharmacol. 110, 336-346.

280

PAN, LC., LAME, M.W., WILSON, D.W. and SEGALL HJ. (1992). Cor
pulmonale is caused by putative monocrotaline metabolite MCTP, but not
glutathione or cysteine conjugates. The Toxicologist 12, 34-34.(Abstract)

PARSON, DP. and SUBJECK, JR. (1972). The morphology of the polysaccharide
coat of mammalian cells. Biachim. Bi0phys. Acta 265, 85-113.

PASTAN, 1., WILLINGHAM, M., ANDERSON, W. and GALLOW, M. (1977).
Localization of serum-derived alpha2-macroglobulin in cultured cells and decrease
after Moloney sarcoma virus transformation. Cell 12, 609-617.

PAULSEN, D.B., MOSIER, D.A., CLINKENBEARD, KD. and CONFER, AW.
(1989). Direct effects of Pasteurella haemolytiga lipopolysaccharide on bovine
pulmonary endothelial cells in vitro. AmerJ.Vet.Res. 50, 1633-1637.

PEACH, MJ., JOHNS, R.A. and 1220, NJ. (1987). Endothelium-mediated
modulation of tone in pulmonary vessels..From: International Conference on Primary
Pulmonary Hypertension. (In Prep.)

PEARSON, J .D., CARLETON, J .S., HUTCHINGS, A and GORDON, J .L (1978).
Uptake and metabolism of adenosine by pig aortic endothelial and smooth muscle
cells in culture. BiachemJ. 170, 265-271.

PEARSON, J.D., CARLETON, SJ. and GORDON, J.L (1980). Metabolism of
adenine nucleotides by ectoenzymes of vascular endothelial and smooth muscle cells
in culture. BiachemJ. 190, 421-429.

PECKHAM, J.C., SANGSTER, LT. and JONES, O.H.JR. (1974). Maria
spectabilis poisoning in swine. J.Amer.Vet.Med.Assac. 165, 633-638.

PEGG, AE. and MCCANN, PP. (1982). Polyamine metabolism and function.
Am.J.Physial. 243(Cell Physiol.12), C212-C221.

PERKETT, E.A, BRIGHAM, KL and MEYRICK, B. (1986). Increased
vasoreactivity and chronic pulmonary hypertension following thoracic irradiation in
sheep. J.AppLPhysial. 61, 1875-1871.

PETERSON, J.E. and JAGO, M.V. (1980). Comparison of the toxic effects of
dehydroheliotridine and heliotrine in prignant rats and their embryos. J.Path. 131,
339-355.

PETRY, T.W., BOWDEN, G.T., HUXT ABLE, RJ. and SIPES, LG. (1984).
Characterization of hepatic DNA damage induced in rats by the pyrrolizidine
alkaloid monocrotaline. Cancer Res. 44, 1505-1509.

281

PETRY, T.W., BOWDEN, G.T., BUHLER, DR. and SIPES, LG. (1986).
Genotoxicity of the pyrrolizidine alkaloid jacobine in rats. TaxicaLLett. 32, 275-281.

PHILLIPS, P.G., HIGGINS, PJ., MALIK, AB. and TSAN, MP. (1988). Effect of
hyperoxia on the cytoarchitecture of cultured endothelial cells. AmerlPathal. 132,
59-72.

PHILLIPS, TL (1966). An ultrastructural study of the development of radiation
injury in the lung. Radiology 87, 49-54.

PIETRA G.G., and SCHLOO, BL (1986). Pathology of primary pulmonary
hypertension. In Abnormal Pulmonary Circulation (E.H. Bergofsky, Ed.), Churchill
Livingstone, New York.

PIETRA, G.G., EDWARDS, W.D. and KAY, J.M. (1987). The spectrum of
pulmonary arterial lesions in patients with primary pulmonary hypertension. From:
International Conference on Primary Pulmonary Hypertension. (In Prep.)

PIPER, P.S., VANE, J .R. and WYLLIE, J .H. (1970). Inactivation of prostaglandins
by the lungs. Nature (Land) 225, 600-604.

PLEST'INA, R. and STONER, H.B. (1972). Pulmonary oedema in rats given
monocrotaline pyrrole. J.Path. 106, 235-249.

POBER, J.S. (1988). Cytokine-mediated activation of vascular endothelium:
physiology and pathology. AmerJ.Pathal. 133, 426-433.

POBER, 1.8. and COTRAN, RS. (1990). Cytokines and endothelial cell biology.
PhysioLRev. 70, 427-450.

POHL, U. (1990). Endothelial cells as part of a vascular oxygen-sensing system:
hypoxia-induced release of autacoids. Experientia 46, 1175-1179.

POWELL J.S., MULLER, R.KM., ROUGE, M., KUHN, H., HEFTI, F. and
BAUMGARTNER, HR. (1990). The proliferative response to vascular injury is
suppressed by angiotensin-converting enzyme inhibition. J.CardiavasaPharmacal.
16(Suppl.4), $42-$49.

PRINGLE, J.K, BRIGHT, J.M., DUNCAN, R.B., KERR, L, LINNABARY, RD.
and TARRIER, MR (1991). Pulmonary hypertension in a group of dairy calves.
J.Amer. Vet.Med.Assac. 198, 857-861.

PUTZKE, HF. and PERSAUD, T.V.N. (1976). Cytotoxicity of the pyrrolizidine
alkaloid fulvine to pancreatic acinar cells in the rat. Light microscopic histochemical
and ultrastructural studies. ExpLPath. 12, 329-335.

282

RABINOVITCH, M., GAMBLE, W., NADAS, AS., MIETTINEN, O. and REID,
L (1979). Rat pulmonary circulation after chronic hypoxia: hemodynamic and
structural features. Am.J.Physial. 236, H817-H827.

RACZNIAK, TJ., SHU MAKER, R.C., ALLEN, J .R., WILL J .A and LALICH, JJ.
(1979). Pathophysiology of dehydromonocrotalinc-induced pulmonary fibrosis in the
beagle. Respiration 37, 252-260.

RADOMSKI, M.W., PALMER, R.MJ. and MONCADA S. (1987). Endogenous
nitric oxide inhibits human platelet adhesion to vascular endothelium. Lancet 2,
1057-1058.

RAMOS, K and COX, LR. (1987). Primary cultures of rat aortic endothelial and
smooth muscle cells: an in vitro model to study xenobiotic-induced vascular toxicity.
In Vitra CelLDev.BiaL 23, 288-296.

RAT'LIFF, N.B., GERRARD, J.M. and WHITE, J.G. (1979). Platelet-leukocyte
interaction following arterial endothelial injury. AmerJ.Pathal. 96, 567-580.

RAYMOND, S.L and DODDS, WJ. (1975). Characterization of the fawn-hooded
rat as a model for hemostatic studies. Thramb.DiathHaemorrh. 33, 361-369.

RECKLESS, J.P.D., WEINSTEIN, DB. and STEINBERG, D. (1978). Lipoprotein
and cholesterol metabolism in rabbit arterial endothelial cells in culture. Biachim.
Biophys. Acta 529, 475-487.

REID, LM. (1979). The pulmonary circulation: Remodeling in growth and disease.
Am.Rev.Respir.Dis. 119, 531-546.

REIDY, M.A, CHOPEK, M., CHAO, 8., MCDONALD, T. and SCHWARTZ, S.M.
(1989). Injury induces increase of von Willebrand factor in rat endothelial cells.
AmerlPathal. 134, 857-864.

REIDY, M.A. (1990). Proliferation of smooth muscle cells at sites distant from
vascular injury. Arteriasclerasis 10, 298-305.

REINDEL J.F., CHANG, C.C., MADHUKAR, B.V. and ROTH, RA (1988).
Monocrotaline pyrrole induces antiproliferative and delayed cytotoxic responses in
cultured cells. FASEB J. 2, A1181.(Abstract)

REINDEL J.F., GANEY, P.E., WAGNER, J.G., SLOCOMBE, R.F. and ROTH,
R.A. (1990). Development of morphologic, hemodynamic, and biochemical changes
in lungs of rats given monocrotaline pyrrole. TaxicaL Appl. Pharmacol. 106, 179-200.

283

REINDEL J.F. and ROTH, R.A. (1991). The effects of monocrotaline pyrrole on
cultured pulmonary artery endothelial and smooth muscle cells. AmerJPathal. 138,
707-719.

REINDEL J.F., HOORN, C.M., WAGNER, J.G. and ROTH, R.A. (1991).
Comparison of the responses of bovine and porcine pulmonary artery endothelial
cells to monocrotaline pyrrole. Am.J.Physial. (Lung CelLMoLPhysiaL5) 261,
LAO6-L414.

REINDERS, J.H., DEGROOT, P.G., GONSALVES, M.D., ZANDBERGEN, J.,
LOESBERG, C. and VAN MOURIK J .A. (1984). Isolation of a storage and secretory
organelle containing von Willebrand protein from cultured human endothelial cells.
Biachim. Biophys. Acta 804, 361-369.

RENKIN, EM. (1977). Multiple pathways of capillary permeability. Cancer Res. 41,
735-743.

RENKIN, EM. (1988). Transport pathways and processes. In Endothelial Cell Biology
in Health and Disease (11. Simionescu, and m. Simionescu, Eds.), pp.51-68. Plenum
Press, New York.

REFINE J .E., EATON, J.W., ANDERS, M.W., HOIDAL J .R. and FOX, RB.
(1979). Generation of hydroxyl radical by enzymes, chemicals and human phagocytes
in vitro. Detection with the anti-inflammatory agent, dimethylsulfoxide. J.Clianvest.
64, 1642-1651.

ROBERTSON, KA, SEYMOUR, J.L, HSIA MT. and ALLEN, LR. (1977).
Covalent interaction of dehydroretronecine, a carcinogenic metabolite of the
pyrrolizidine alkaloid monocrotaline, with cysteine and glutathione. Cancer Res. 37,
3141-3144.

ROBERTSON, KA (1982). Alkylation of N2 in deoxyguanosine by
dehydroretronecine, a carcinogenic metabolite of the pyrrolizidine alkaloid
monocrotaline. Cancer Res. 42, 8-14.

ROBINSON-WHITE, A and BEAVEN, M.A. (1982). Presence of histamine and
histamine-metabolizing enzyme in rat and guinea-pig microvascular endothelial cells.
J.PharmacaLExpTher. 223, 440-445.

ROSE, AL, GARDINER, C.A, MCCONNEL J .D. and BULL LB. (1957). Field
and experimental investigation of "walk-about" disease of horses (Kimberly horse

disease) in Northern Australia. Crotalaria poisoning in horses. Australian VetJ. 33,
25-62.

284

ROSE, C.L, FINK, R.D., HARRIS, RN. and CHEN, KK. (1945). The effect of
hepatotoxic alkaloids on the prothrombin time of rats. IPhannacalEprher. 83,
265-269.

ROSEN, E.M., VINTER, D.W. and GOLDBERG, ID. (1989). Hypertrophy of
cultured bovine aortic endothelium following irradiation. Radiat.Res. 117, 395-408.

ROSENBERG, HO and RABINOVITCH, M. (1988). Endothelial injury and
vascular reactivity in monocrotaline pulmonary hypertension. Am.J.Physial. 255,
H1484-H 1491.

ROSENBLUM, M.L, VASOUEZ, D.A, HOSHINO, T. and WILSON, CB. (1978).
Development of a clonogenic cell assay for human brain tumors. Cancer 41,
2305-2314.

R088, R., RAINES, E.W. and BOWEN-POPE, DE (1986). The biology of
platelet-derived growth factor. Cell 46, 155-169.

ROTH, R.A, DOT‘ZLAF, B., BARANYI, B., KUO, CH. and HOOK, J.B. (1981a).
Effect of monocrotaline ingestion on liver, kidney, and lung of rats. Toxicol. Appl.
Pharmacol. 60, 193-203.

ROTH, R.A (1981b). Effect of pneumotoxicants on lactate dehydrogenase activity
in airways of rats. Toxicol. Appl. Pharmacol. 57, 69-78.

ROTH, R.A (1985). Biochemistry, physiology and drug metabolism -- Implications
regarding the role of the lungs in drug disposition. ClinPhysr'aLBiachem 3, 66-79.

ROTH, R.A and GANEY, RE. (1988). Platelets and the puzzles of pulmonary
pyrrolizidine poisoning. Toxicol. Appl. Pharmacal 93, 463-471.

ROTH, R.A, and REINDEL J.F. (1990). Lung vascular injury from monocrotaline
pyrrole, a putative hepatic metabolite. In Biological Reactive Intermediates IV (C.M.
Witrner, and et al., Eds.), pp.477-487. Plenum Press, New York, NY.

ROTH, R.A, WHITE, S.M. and REINDEL J .F . (1989). Mechanisms of lung injury
and pulmonary hypertension from monocrotaline: An hypothesis. Comm. Taxical. 3,
131-144.

ROUNDS, 8., FARBER, H.W., HILL NS. and O’BRIEN, R.F. (1985). Effects of
endothelial cell injury on pulmonary vascular reactivity. Chest 88, 2138-2158.

RUBANYI, GM. (1991). Endothelium-derived relaxing and contracting factors.
J.CelLBiachem. 46, 27-36.

285

RUBIN, D.B., DRAB, E.A, WARD, W.F., SMITH, LJ. and F OWELL S.M. (1984).
Enzymatic responses to radiation in cultured vascular endothelial and smooth muscle
cells. Radiat.Res. 99, 420-432.

RUTILI, G., KVIETYS, P., MARTIN, D., PARKER, J.C. and TAYLOR, AE.
(1982). Increased pulmonary microvascular permeability induced by
alpha-napthylthiourea. J.AppLPhysial. 52, 1316-1323.

RYAN, J.W., and CATRAVAS, JD. (1990). Angiotensin converting enzyme as an
indicator of pulmonary microvascular function. In Focus on Pulmonary Pharmacology
and Toxicology (M.A Hollinger, Ed.), pp.183-210. CRC Press, Inc., Boca Raton, FL

RYAN, J.W., RYAN, U.S., SHULTZ, D.R., WHITAKER, C. and CHUNG, A
(1975). Subcellular localization of pulmonary angiotensin-converting enzyme (kininase
II). BiachemJ. 146, 497-499.

RYAN, U.S., RYAN, J .W. and CHIU, A. (1976). Kininase II (angiotensin converting
enzyme) and endothelial cells in culture. Adv.Exp.MedBiol. 70, 217-227.

RYAN, U.S., CLEMENTS, E., HABLISTON, K and RYAN, J .W. (1978). Isolation
and culture of pulmonary artery endothelial cells. Tissue Cell 10, 535-554.

RYAN, U.S., SCHULTZ, DR. and RYAN, J.W. (1981). F0 and C3b receptors on
pulmonary endothelial cells: induction by injury. Science 214, 557-559.

RYAN, U.S., WHITNEY, PL. and RYAN, J.W. (1982a). Localization of carbonic
anhydrase on pulmonary endothelial cells in culture. J.AppLPhysial. 53, 914-919.

RYAN, U.S., RYAN, J.W. and PLUMMER, TH. (1982b). Carboxypeptidase N is
a surface enzyme of pulmonary endothelial cells. Cardiology 66, II-242.(Abstract)

RYAN, U.S. (1982c). Structural bases for metabolic activity. Ann.Rev.Physial. 44,
223-239.

RYAN, US. (1986). Metabolic activity of pulmonary endothelium: modulations of
structure and function. AnnRev.Physial. 48, 263-277.

RYAN, U.S., AVDONIN, P.V., POSIN, E.Y., POPOV, E.G., DANILOV, S.M. and
TKACHUK V.A (1988). Influence of vasoactive agents on cytoplasmic free calcium
in vascular endothelial cells. J.Appl.PhySi0L 65, 2221-2227.

RYAN, US. and RYAN, J .W. (1985). Relevance of endothelial surface structure to
the activity of vasoactive substances. Chest 88, 2038-2068.

286

RYAN, US. (1988). Phagocytic properties of endothelial cells. In Endothelial Cells.
VaLIII (U.S. Ryan, Ed.), pp33-50. CRC Press, Boca Raton.

RYAN, US. and WHITE, L. (1986). Microvascular endothelium, isolation with
microcarriers: arterial, venous. J. Tissue Cult.Methads 10, 9-14.

SAGE, H. (1984). Collagen synthesis by endothelial cells in culture. In Biology of
endothelial cells (E.A Jaffe, Ed.), pp.161-177. Martinus-Nijhoff, Boston.

SALONE, B., HENNIG, B., GILLESPIE, MN. and OLSON, J.W. (1991).
Monocrotaline (MCT) is toxic to cultured porcine pulmonary artery endothelial cells
(ECs). FASEB J. 5, A535.(Abstract)

SAMUELSSON, B. (1983). Leukotrienes: Mediators of immediate hypersensitivity
reactions and inflammation. Science 220, 568-575.

SANDERS, D.A., SHEALY, AL and EMMEL M. W. (1936). The pathology of
Winks poisoning in cattle. J.Amer. Vet. MedAssac. 89, 150-162.

SATO, Y. and RIFKIN, DB. (1988). Autocrine activities of basic fibroblast growth
factor: regulation of endothelial cell movement, plasminogen activator synthesis, and
DNA synthesis. J.CellBiaL 107, 1199-1205.

SCHINI, V.B., HENDRICKSON, H., HEUBLEIN, D.M., BURNETT, J .C., JR. and
VANHOUTTE, RM. (1989). Thrombin enhances the release of endothelin from
cultured porcine aortic endothelial cells. Eur.J.Pharmacal. 165, 333-334.

SCHOENTAL R., HEAD, M.A and PEACOCK, RF. (1954). Senecio alkaloids:
Primary liver tumors in rats, as a result of treatment with (1) a mixture of alkaloids
from S. jacobaea lin. (2) retrorsine (3) isatidine. BrJ.Cancer 8, 458-465.

SCHOENTAL R. and HEAD, M.A (1955). Pathological changes in rats as a result
of treatment with monocrotaline. BritJ. Cancer 9, 229-237.

SCHOENT AL R. (1968). Toxicology and carcinogenic action of pyrrolizidine
alkaloids. Cancer Res. 28, 2237-2246.

SCHRAUFNAGEL DE. and SCHMID, A (1989). Pulmonary capillary density in
rats given monocrotaline. Am.Rev.Respir.Dis. 140, 1405-1409.

SCHRAUFNAGEL DE. (1990). Monocrotaline-induced angiogenesis: Differences
in the bronchial and pulmonary vasculature. AmerlPathal. 137, 1083-1090.

287

SCHULTZE, AE., WAGNER, J .G., WHITE, S.M. and ROTH, R.A (1990). Early
indications of monocrotaline pyrrole-induced lung injury in rats. Toxicol. Appl.
Pharmacol. (In Press)

SCHULTZE, AE., WAGNER, J.G. and ROTH, R.A. (1991). An evaluation of
procoagulant activity in the peripheral blood of rats treated with monocrotaline
pyrrole. Taxical. AppL Pharmacol. 109, 421-431.

SCHULTZE, R.A and ROTH, R.A. (1992). Fibrinolytic activity in blood and lungs
of rats treated with monocrotaline pyrrole. (In Prep.)

SCHWARTZ, S.M. and R088, R. (1984). Cellular proliferation in atherosclerosis
and hypertension. Prag.Cardiavasc.Di.s. XXVI, 355-371.

SCHWEIGERER, L, NEUFELD, G., FRIEDMAN, J., ABRAHAM, J .A, FIDDES,
J.C. and GOSPODAROWICZ, D. (1987). Capillary endothelial cells express basic
fibroblast growth factor, a mitogen that promotes their own growth. Nature (Land)
325, 257-259.

SHALE, DJ., WISEMAN, MS. and COOKSON, W.O.C.M. (1986). Effect of
monocrotaline ingestion on the distribution of protein and angiotensin converting
enzyme activity in the rat lung. Thorax 41, 914-918.

SHANAHAN, W. and KORN, J. (1984). Endothelial cells express IL-l-like activity
as assessed by enhancement of fibroblast PGE synthesis. Clin.Res. 32, 666A

SHASBY, D.M., SHASBY, 8.8., SULLIVAN, J.M. and PEACH, MJ. (1982). Role
of endothelial cell cytoskeleton in control of endothelial permeability. Circ.Res. 51,
657-665.

SHEPHERD, J.T. and KATUSIC, 2.8. (1991). Endothelium-derived vasoactive
factors: I. End0thelium-dependent relaxation. Hypertension 18[suppl III],
III-76-III-85.

SHERIDAN, JD. and ATKINSON, M.M. (1985). Physiological roles of permeable
junctions: some possibilities. Ann.Rev.Physial. 47, 337-353.

SHUBAT, PJ., BANNER, W. and HUXTABLE, RJ. (1987). Pulmonary vascular
responses induced by the pyrrolizidine alkaloid, monocrotaline, in rats. T oxican 25,
995-1002. -

SHUBAT, PJ., HUBBARD, AK. and HUXTABLE, RJ. (1989). Dose-response
relationships in intoxication by the pyrrolizidine alkaloid, monocrotaline.
J. TaxicoLEnviroaHealth 28, 445-460.

288

SHUBAT, PJ., BOWERS, RJ. and HUXT ABLE, RJ. (1990). N a+ /K+ -Adenosine
triphosphatase activity of pulmonary arteries after intoxication with the pyrrolizidine
alkaloid, monocrotaline. J.Phannacal.Exp.Ther. 252, 70-76.

SHULL LR., BUCKMASTER, G.W. and CHEEKE, RR. (1976). Factors
influencing pyrrolizidine (Senecio) alkaloid metabolism: species, liver sulfliydryls and
rumen fermentation. J.AnimSci. 43, 1247-1257.

SHUMAKER, R.C., ROBERTSON, KA, HSU, LC. and ALLEN, JR. (1976).
Neoplastic transformation in tissues of rats exposed to monocrotaline or
dehydroretronecine. J.Natl Cancer Inst. 56, 787-789.

SIMIONESCU, M., SIMIONESCU, N. and PALADE, GE. (1975). Segmental
differentiations of cell junctions in the vascular endothelium. J.CellBial 67, 863-868.

SIMIONESCU, M. and SIMIONESCU, N. (1986). Functions of the endothelial cell
surface. AmRev.Physial 48, 279-293.

SIPPEL W.L (1964). Crotalaria poisoning in livestock and poultry.
Ann.N.YAcadSci. 111, 562-570.

SIRVIO, M.L, METSARIN NE, K, SAIJONMAA, O. and FYHRQUIST, F. (1990).
Tissue distribution and half-life of 1251-endothelin in the rat: Importance of
pulmonary clearance. BiachemBiaphys.Res.Cammun. 167, 1191-1195.

SLAKEY, LL, GOLDMAN, SJ., and GORDON, EL (1988). Kinetics of
endothelial ectonucleotidases. In Endothelial Cells. ValI (U.8. Ryan, Ed.), pp.55-63.
CRC Press, Boca Raton.

SMEDLEY, LA, TONNESEN, M.C., SANDHAUS, R.A, HASSLET, C.,
GUTHRIE, LA, JOHNSTON, R.BJR., HENSON, RM. and WORTHEN, GS.
(1986). Neutrophil-mediated injury to endothelial cells. Enhancement by endotoxin
and essential role of neutrophil elastase. J.Clin.Invest. 77, 1233-1243.

SMITH, LW. and CULVENOR, C.CJ. (1981). Plant sources of hepatotoxic
pyrrolizidine alkaloids. J.Nat.Prods. 44, 129-152.

SMITH, P. and HEATH, D. (1978). Evagination of vascular smooth muscle cells
during the early stages of Crotalaria pulmonary hypertension. J.Path. 124, 177-183.

SMITH, U. and RYAN, J .W. (1972). Pulmonary endothelial cells and metabolism of
adenine nucleotides, kinins and angiotensin I. Adv.Exp.Med.Bial 21, 267-276.

SMITH, U. and RYAN, J.W. (1973a). Electron microscopy of endothelial cells
collected on cellulose acetate paper. Tissue Cell 5, 333-336.

289

SMITH, U. and RYAN, J.W. (1973b). Electron microscopy of endothelial and
epithelial components of the lungs: correlations of structure and function. FedProc.
32, 1957-1966.

SNOW, R.L, DAVIES, P., PONTOPPIDAN, H., ZAPOL W.M. and REID, LM.
(1982). Pulmonary vascular remodeling in Adult Respiratory Distress Syndrome.
Am.Rev.Respir.Dis. 126, 887-892.

SNYDER, SP. (1972). Livestock losses due to tansy ragwort poisoning. Oregon
Agrirecard 255, 3-4.

SPORN, LA, RUBIN, P., MARDER, U.J. and WAGNER, DD. (1984). Irradiation
induces release of von Willebrand protein from endothelial cells in culture. Blood 64,
567-570.

STALCUP, S.A, PANG, LM., LIPSET, J.S., ODYA C.E., GOODFRIEND, TL
and MELLINS, RB. (1978). Gestational changes in pulmonary converting enzyme
activity in the fetal rabbit. Circ.Res. 43, 705-711.

ST ALCUP, S.A, LIPSET, J .S., WOAN, J .M., LEUENBERGER, P. and MELLINS,
RB. (1979). Inhibition of angiotensin converting enzyme activity in cultured
endothelial cells by hypoxia. J.Clin.Invest. 63, 966-976.

STALKER, M. (1884). Crotalism--a new disease among horses. BullIawa State
AgnCall 114.

ST ANBROOK, H.S., MORRIS, KG. and MCMURTRY, LF. (1984). Prevention and
reversal of hypoxic pulmonary hypertension by calcium antagonists 1-3.
Am.Rev.Respir.Dis. 130, 81-85.

STEEL R.G.D., and TORRIE, J .H. (1980). Principles and procedures of statistics: A
biametrical approach. pp. 381.McGraw-Hill,lnc., New York.

STENMARK, KR., MORGANROTH, M.L, REMIGIO, LK, VOELKEL N.F.,
MURPHY, R.C., HENSON, P.M., MATHIAS, M.M. and REEVES, J.T. (1985).
Alveolar inflammation and arachidonate metabolism in monocrotaline-induced
pulmonary hypertension. Am.J.Physial. 248, H859-H866.

STERN, D., NAWROTH, P., MARCUM, J., HANDLEY, D., KISIEL W.,
ROSENBERG, R. and STERN, K (1985). Interaction of antithrombin III with
bovine aortic segments. Role of heparin in binding and enhanced anticoagulant
activity. J.Clin.Invest. 75, 272-279.

290

STERN, D., BRETT, J., HARRIS, K and NAWROTH, P. (1986). Participation of
endothelial cells in the protein C-protein S anticoagulant pathway: The synthesis and
release of protein S. J.CellBial 102, 1971-1978.

STERN, D.M., NAWROTH, P.P., KISIEL W., VEHAR, G. and ESMON, CT.
(1985). The binding of factor IXa to cultured bovine aortic endothelial cells.
Induction of a specific site in the presence of factor VIII and X. J.Bial Chem. 260,
6717-6722.

ST OLZ, DB. and JACOBSON, BS. (1991). Macro- and microvascular endothelial
cells in vitro: maintenance of biochemical heterogeneity despite loss of ultrastructural
characteristics. In Vitra Cell.Dev.Bial. 27A, 169-182.

STUART, KL and BRAS, G. (1957). Veno-occlusive disease of the liver. QlMed
26, 291-315.

STYLES, J., ASHBY, J. and MATTOCKS, AR. (1980). Evaluation m of
several pyrrolizidine alkaloid carcinogens: observations on the essential pyrrolic
nucleus. Carcinogenesis 1, 161-164.

SUGITA T., STENMARK, KR., WAGNER, W.W., HEN SON, P.M., HY ERS, T.M.
and REEVES, J.T. (1983a). Abnormal alveolar cells in monocrotaline induced
pulmonary hypertension. ExLung Res. 5, 201-215.

SUGIT A T., HYERS, T.M., DAUBER, LM., WAGNER, W.W., MCMURTRY, LF.
and REEVES, J.T. (1983b). Lung vessel leak precedes right ventricular hypertrophy
in monocrotaline-treated rats. JAppLPhysial 54, 371.

TAKANASHI, H., UMEDA M. and HIRONO, I. (1980). Chromosomal aberrations
and mutation in cultured mammalian cells induced by pyrrolizidine alkaloids.
Mutation Res. 78, 67-77.

TAKEOKA, O., ANGEVINE, D.M. and LALICH, JJ. (1962). Stimulation of mast
cells in rats fed various chemicals. Amen/.Pathal 40, 545-554.

TAYLOR, 8.G. and WESTON, AH. (1988). Endothelium-derived hyperpolarizing
factor: A new endogenous inhibitor from the vascular endothelium. TIPS 9, 272-274.

THOMAS, E.E. (1934). The toxicity of certain species of Crotalaria seed for the
chicken, quail, turkey and dove. J.Amer. Vet.Med.Assac. 85, 617-622.

THOMAS, G.P., WELCH, W.J., MATTHEWS, M. and FERAMISCO, JR. (1982).
Molecular and cellular effects of heat shock and related treatments of mammalian
tissue culture cells. Cold Spring Harbor Symp. Quant. Biol 46, 985-996.

291

TILL G.O., HATHERILL J.R., TOURTELLOTTE, W.W., LUTZ, MJ. and
WARD, RA (1985). Lipid peroxidation and acute lung injury following thermal
trauma to the skin: Evidence for the role of hydroxyl radical. Amer.J.Pathal. 119,
376-384.

TODD, L, MULLEN, M., OLLEY, PM. and RABINOVITCH, M. (1985).
Pulmonary toxicity of monocrotaline differs at critical periods of lung development.
PediatnRes. 19, 731-737.

TODOROVICH-HUNTER, L, JOHNSON, D., RANGER, P., KEELEY, F. and
RABINOVITCH, M. (1988). Elastin and collagen synthesis associated with
progressive pulmonary hypertension induced by monocrotaline: A biochemical and
ultrastructural study. Laanvest. 58, 184-195.

TOKUDA K and BODELL WJ. (1987). Cytotoxicity and sister chromatid
exchanges in 9L cells treated with monofunctional and bifunctional nitrogen
mustards. Carcinogenesis 8, 1697-1701.

TOMASHEFSKI, J .F., DAVIES, P., BOGGIS, C., GREENE, R., ZAPOL W.M. and
REID, LM. (1983). The pulmonary vascular lesions of the adult respiratory distress
syndrome. AmerJ.Pathal 112, 112-126.

TOMLINSON, A, VANVLIJMEN, H., LOESCH, A. and BURNSTOCK, G. (1991).
An immunohistochemical study of endothelial cell heterogeneity in the rat:
observations in "en face" Hautchen preparations. Cell Tissue Res. 263, 173-181.

TUCHWEBER, B., KOVACS, K, JAGO, M.V. and BEAULIEU, T. (1974). Effect
of steroidal and non-steroidal microsomal enzyme inducers in the hepatotoxicity of
pyrrolizidine alkaloids in rats. Res. Commun. ChemPathalphannacal 7, 459-480.

TUCKER, A, BRYANT, S.E., FROST, H.H. and MIGALLY, N. (1983). Chemical
sympathectomy and serotonin inhibition reduce monocrotaline-induced right
ventricular hypertrophy in rats. CanJ.PhysialPhannacal 61, 356-362.

TURNER, J .H. and LALICH, J J. (1965). Experimental cor pulmonale in the rat.
ArchPathal 79, 409-419.

VALDIVIA, E., HAYASHI, Y., LALICH, JJ. and SONNAD, J. (1965). Capillary
obstruction in experimental cor pulmonale. Circ. 32(supp.ll), 211-211.

VALDIVIA E., LALICH, J J ., HAYASHI, H. and SONNAD, J. (1967a). Alterations
in pulmonary alveoli after a single injection of monocrotaline. Arch.Pathal 84, 64-76.

VALDIVIA, E., SONNAD, J., HAYASHI, Y. and LALICH, JJ. (1967b).
Experimental interstitial pulmonary edema. Ang. 18, 378-383.

292

VENDER, R.L, CLEMMONS, D.R., KWOCK, L and FRIEDMAN, M. (1987).
Reduced oxygen tension induces pulmonary endothelium to release a pulmonary
smooth muscle cell mitogen(s). Am.Rev.Respir.Dis. 135, 622-627.

VINCIC, L, ORR, F .W., WARNER, DJ .A, SUYAMA KL and KAY, J.M. (1989).
Enhanced cancer metastasis after monocrotaline-induced lung injury. Taxical Appl
Pharmacol 100, 259-270.

VLODAVSKY, I., FRIDMAN, R., SULLIVAN, R., SASSE, J. and KLAGSBRUN,
M. (1987). Aortic endothelial cells synthesize basic fibroblast growth factor which
remains cell associated and platelet-derived growth factor-like protein which is
secreted. J.CellPhysial 131, 402-408.

VOELKEL N., and REEVES, J .T. (1979). Primary pulmonary hypertension. In
Pulmonary Vascular Diseases (KM. Maser, Ed.), pp.573-628. Marcel Dekker, New
York.

VOELKEL N.F., and REEVES, J .T. (1979). Primary pulmonary hypertension. In
Pulmonary Vascular Disease (pp.573-628. Dekker, New York.

VOGEL A, RAINES, B., KARIYA B., RIVEST, MJ. and R088, R. (1978).
Coordinate control of 3T3 cell proliferation by platelet-derived growth factor and
plasma components. PracNatLAcadSci. USA 75, 2810-2814.

VOYTA J.C., VIA D.P., BUTTERFIELD, CE. and ZETTER, BR. (1984).
Identification and isolation of endothelial cells based on their increased uptake of
acetylated-low density lipoprotein. J. CelLBial. 99, 2034-2040.

WAGENVOORT, CA and WAGENVOORT, N. (1970). Primary pulmonary
hypertension: a pathologic study of the lung vessels in 156 clinically diagnosed cases.
Circ. 42, 1163-1182.

WAGENVOORT, CA, and WAGENVOORT, N. (1977). Pathology of Pulmonary
Hwertensian. Wiley, New York.

WAGNER, D.D., OLMST'ED, J .B. and MARDER, VJ. (1982). Immunolocalization
of von Willebrand protein in Weibel-Palade bodies of human endothelial cells.
J.CellBial. 95, 355-360.

WAGNER, J.G., ROTH, R.A, WHITE, S.M. and PETRY, T.W. (1991).
Monocrotaline pyrrole-induced DNA crosslinking in cultures of endothelial cells from
pulmonary artery. FASEB 1.5, A534.(Abstract)

293

WAGNER, J.G., PETRY, T.W. and ROTH, R.A (1992). Characterization of
monocrotaline pyrrole-induced DNA cross-linking in pulmonary artery endothelium.
(Submitted)

WALDHORN, R.B., TSOU, B., SMITH, RP. and KERWIN, D.M. (1984).
Pulmonary veno-occlusive disease associated with rnicroangiopathic hemolytic anemia
and chemotherapy of gastric adenocarcinoma. Medical and Pediatric Oncology 12,

394-396.

WALLACE, KB., BAILIE, MD. and HOOK J.B. (1979). Development of
angiotensin-converting enzyme in fetal rat lungs. Am.J.Physial. 236, R57-R60.

WANG, C.H., LARGIS, E.E. and SCHAFFER, SA (1981). The effects of
endothelial cell-conditioned media on the proliferation of aortic smooth muscle cells
and 3T3 cells in culture. Artery 9, 358-371.

WANG, J.L, SUN, 0., and VOSS, PG. (1989). Growth control in cultured 3T3
fibroblasts: Molecular properties of a growth regulatory factor isolated from
conditioned medium. In Growth control during cell aging (E. Wang, and HR. Warner,
Eds.), pp.149-162. CRC Press, Boca Raton, Florida.

WARD, P.A, TILL, G.O., KUNKEL S. and BEAUCHAMP, C. (1983). Evidence
for a role of hydroxyl radical in complement and neutrophil-dependent tissue injury.
J.Clin.Invest. 72, 789-801.

WARD, W.F., MOLTENI, A and TS’AO, CH. (1989). Radiation-induced
endothelial dysfunction and fibrosis in rat lung: modification by the angiotensin
converting enzyme inhibitor CL242817. Radiat.Res. 117, 342-350.

WARRELL R., HULTIN, MB. and COLLER, BS. (1979). Increased factor
VIII/van Willebrand factor antigen and von Willebrand factor activity in renal
failure. AmerJMed. 66, 226-228.

WARREN, H.B., PANTAZIS, P. and DAVIES, RF. (1987). The third component
of complement is transcribed and secreted by cultured human endothelial cells.
AmerlPathal 129, 9-13. A

WATANABE, S. and OGATA T. (1976). Clinical and experimental study upon
primary puhnonary hypertension. Jpn. Cich. 40, 603-610.

WEDGEWOOD, J.F., HATAM, L and BONAGURA, V.R. (1988). Effect of
interferon-gamma and tumor necrosis factor on the expression of class I and class 11
major histocompatibility molecules by cultured human umbilical vein endothelial
cells. CellImmunal. 111, 1-9.

294

WEHRMACHER, W.H. (1988). Endothelium: From whence to whither.
Semin. Thramb.Hemast. 14, 1-11.

WEIDNER, M.F., SIGURDSSON, S.T. and HOPKINS, RB. (1990). Sequence
preferences of DNA interstrand cross-linking agents: dG-to-dG cross-linking at 5’-CG
by structurally simplified analogues of mitomycin C. Biochemistry 29, 9225-9233.

WERCHAN, P.M., SUMMER, W.R., GERDES, AM. and MCDONOUGH, KH.
(1989). Right ventricular performance after monocrotaline-induced pulmonary
hypertension. Am.J.Physial. 256, H1328-H1336.

WHATLEY, R.B., ZIMMERMAN, G.A., MCINTYRE, T.M. and PRESCOTT, S.M.
(1988). Endothelium from diverse vascular sources synthesizes platelet-activating
factor. Arteriasclerasis 8, 321-331.

WHEELER, GP. (1975). A review of studies on the mechanism of action of
nitrosoureas. Cancer Chemother. 30, 87--119.

WHITE, I.N.H. and MATTOCKS, AH. (1971). Some factors affecting the conversion
of pyrrolizidine alkaloids to N-oxides and to pyrroloic derivatives m.Xenabiatica
1, 503-505.

WHITE, I.N.H. and MATTOCKS, AR. (1972). Reaction of dihydropyrrolizines with
deoxyribonucleic acids in vitro. BiachemJ. 128, 291-297.

WHITE, S.M. and ROTH, R.A. (1988). Pulmonary platelet sequestration is increased
following monocrotaline pyrrole treatment of rats. Toxicol Appl Phannacal 96,
465-475.

WIGGINS, R.C., LOSKUTOFF, DJ., COCHRANE, C.G., GRIFFIN, J.H. and
EDGINGTON, TS. (1980). Activation of rabbit Hageman factor by homogenates of
cultured rabbit endothelial cells. J.Clin.Invest. 65, 197-206.

WILLIAMS, D.B., REED, R.L., KEDZIERSKI, B., GUENGERICH, RP. and
BUHLER, DR. (1989). Bioactivation and detoxication of the pyrrolizidine alkaloid

senecionine by cytochrome P450 enzymes in rat liver. Drug Metab.Dispas. 17,
387-392.

WILLIAMS, GM. and MORI, H. (1980). Genotoxicity of pyrrolizidine alkaloids in
the hepatocyte primary culture / DNA-repair test. Mutation Res. 79, 1-5.

WILLIAMS, KJ., LANDGRAF, B.E., WHITING, N.L and ZURLO, J. (1989).
Correlation between the induction of heat shock protein 70 and enhanced viral
reactivation in mammalian cells treated with ultraviolet light and heat shock. Cancer
Res. 49, 2735-2742.

295

WILSON, D.W. and SEGALL HJ. (1990). Changes in type II cell populations in
monocrotaline pneumotoxicity. AmerJPathal 136, 1293-1299.

WITTE, L, FUKS, Z., HAIMOVITZ-FRIEDMAN, A, VLODAVSKY, L,
GOODMAN, D.S. and ELDOR, A. (1989). Effects of irradiation on the release of
growth factors from cultured bovine, porcine, and human endothelial cells. Cancer
Res. 49, 5066-5072.

WOLINSKY, H. (1980). A proposal linking clearing of circulating lipoproteins to
tissue metabolic activity as a basis for understanding atherogenesis. Cancer Res. 47,
301-311.

WONG, M.KK. and GOTLIEB, A1. (1985). Control of reendothelialization: The
importance of endothelial microfilaments, microtubules and centrosomes in
endothelial locomotion. Surv.SynthPathalRes. 4, 341-356.

YAMANAKA, H., NAGAO, M. and SUGIMURA, T. (1979). Mutagenicity of
pyrrolizidine alkaloids in the salmonella/mammalian-microsome test. Mutation Res.
68, 211-216.

YANAGISAWA M., KURIHARA H., KIMURA S. and ET AL (1988). A novel
potent vasoconstrictor peptide produced by vascular end0thelial cells. Nature (Land)
332, 411-415.

YANAGISAWA M. and MASAKI, T. (1989). Endothelin, a novel
endothelium-derived peptide. BiachemPhamracal 38(12), 1877-1883.

YANG, H.Y.T., ERDOS, E.G. and LEVIN, Y. (1970). A dipeptidyl carboxypeptidase
that converts angiotensin I and inactivates bradykinin. Biachem.Biaphys.Acta 214,
374-376.

ZAKHEIM, R.M., MATTIOLI, L, MOLTENI, A, MULLIS, KB. and BARTLEY,
J. (1975). Prevention of pulmonary vascular changes of chronic alveolar hypoxia by
inhibition of angiotensin-l-converting enzyme in the rat. Lab.Invest. 33, 57-61.

ZAKHEIM, R.M., MOLTENI, A, MATTIOLI, L. and MULLIS, KB. (1976).
Angiotensin 2 levels during acute and chronic hypoxia in the unanesthetized rabbit.
J.ApplPhysial 41, 462-465.

ZOJA, C., GHILARDI, F., ZILIO, P., BENIGNI, A. and REMUZZI, G. (1986).
Cyclosporin-induced endothelial injury. Lab.Invest. 55, 455-462.

296

ZSEBO, KM., YUSCHENKOFF, V.N., SCHIFFER, 8., CHANG, D., MCCALL E.,
DINARELLO, C.A, BROWN, M.A, ALTROCK B. and BAGBY, GO (1988).
Vascular endothelial cells and granulopoiesis: Interleukin 1 stimulates release of
G-CSF and GM-CSF. Blood 71, 99-103.

ZWELLING, LA, ANDERSON, T. and KOHN, KW. (1979). DNA-protein and
DNA interstrand crosslinking by cis- and trans-platinum(ll)diamminedichloride in
L1210 mouse leukemic cells and relation to cytotoxicity. Cancer Res. 39, 365-369.