5.... .r3.,Mfu.....3§ Erna. .
fififiewwuu

T 3%

, Br .3 .55... mg

_ . .r! 1.... I. . .

«Fax any: my}?
Lax. .r; n..." W .1: .

.:( :5» .. .xfil

.. .u‘ 25?.

. Mum“... .§

#1

<
31:? z
p???

.

"Rh.
i4}.
T‘s... ..

.1.

Ezmmaméz

. .5.in
.. .. Vanna
A ‘

initqmufili ‘

a? .2. .3
$5 ,1}...
:i! .1.
A: g: 3..

.. 1.!
. If.

.3. in 4&3“;
a“... x as

. w a?

. ”El.
It 19!);

 

 

r 23.4.,

:31?
. ’QO‘L." ’0‘-
XIII 1-! n v v
i ‘0'; vl .A
.

{v.16 s
«.1
t 5‘

I .2 .13}.
«no-i, .lrlufiu
a ”a a

Hana?
L31. 1529 ,3
L w !, .x‘frr-

. ‘ .(..I41
»v In? ..u..\

\.. L! «I ‘ .u...
L;:... .1; .‘
I f 2...;
. Pie}

. zin§ .r. .

 

 

 

um. 31.». .

TERMS

 

IIRDARV’

Michlbi. atate
Unwmmmy

 

 

 

PLACE IN REIURN Box to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDuepw-pJS

MECHANISMS OF ENHANCED ALLYL ALCOHOL
HEPATOTOXICITY BY ENDOTOXIN

By

Shawn Kinser

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

2001

ABSTRACT
MECHANISMS OF ENHANCED ALLYL ALCOHOL
HEPATOTOXICITY BY ENDOTOXIN
By

Shawn Kinser

Lipopolysaccharide (LPS) causes enhancement of toxicity from a
number of xenobiotics including allyl alcohol. The mechanism by which
LPS enhances the hepatotoxicity of allyl alcohol was investigated. Kupffer
cells are critical for the enhancement of allyl alcohol hepatotoxicity by
LPS. Stimulation of Kupffer cells with LPS in either Kupffer cell-
hepatocyte cocultures or isolated, perfused livers from nai‘ve rats did not
induce the enhancement of allyl alcohol hepatotoxicity. These data
indicate that Kupffer cells require the participation of other factors not
present in the medium-perfused liver. Extrahepatic factors critical to
large dose LPS-induced liver injury include neutrophils and thrombin. A
role for each of these factors in the potentiation of allyl alcohol-induced
liver injury by LPS was investigated. Neutrophils were previously known
to play a critical role in this potentiation model. Allyl alcohol is well
known to rapidly deplete glutathione stores in hepatocytes. Therefore, it
was hypothesized that neutrophils enhance the toxicity of allyl alcohol

through the release of reactive oxygen species onto hepatocytes depleted

of their glutathione. While neutrophil-derived reactive oxygen species
caused a slight enhancement of allyl alcohol-induced cytotoxicity in vitro,
in vivo administration Of the antioxidants superoxide dismutase and
catalase or apocynin did not afford protection from the potentiation of
allyl alcohol toxicity by LPS. These data indicate that neutrophil-derived
reactive oxygen species do not play a critical role in this potentiation
model. To begin to investigate the role of thrombin rats were pretreated
with the anticoagulants heparin or warfarin. Both of these anticoagulants
protected against LPS enhancement of allyl alcohol-induced liver injury.
These results suggested that thrombin might play a critical role in this
potentiation model. To determine whether thrombin was acting through a
receptor mediated mechanism to enhance the toxicity of allyl alcohol,
isolated, perfused livers from LPS-treated rats were perfused with
medium containing allyl alcohol in the absence or presence of thrombin.
In these experiments thrombin did not enhance the toxicity of allyl alcohol
indicating that the role of thrombin in this potentiation model may be
dependent on other coagulation factors or other components of the blood.
In conclusion, LPS enhances the hepatotoxicity of allyl alcohol through a
complex mechanism in which the roles of Kupffer cells, neutrophils and

thrombin are interdependent.

Copyright by
Shawn Kinser
2001

Dedications

I dedicate this accomplishment to those I love the most. My
parents Jan and Martin Kinser, my brother Eric, and of course my girl

Mary.

Acknowledgments

There are a great number of people that deserve many thanks for
their help in my progress toward graduation. All of the faculty and staff
in the department of Pharmacology and Toxicology were very supportive
and encouraging through the long hard years of my study.

A great deal of thanks goes to all the peOple of the Roth/Ganey
lab. Dwayne Hill for his wonderful mentorship during my first year
rotation. Fredo for his part in keeping the warm, friendly, and humorous
air about the lab. Also for handing down to me the isolated perfused
liver technique and its many intricacies. Terry Schmitt for her undying
will to help and make happy any and all around her. Jim Wagner for his
help in understanding just what it takes to be a good scientist. Rosie
Sneed for her preliminary work on my project, her friendly guidance, and
her love of animals. Steve Yee for his constant concern, willingness to
help in all things short of straining his back, and for sharing adventures
with electric circuitry. John Buchweitz for his calm demeanor with my
constant complaints and desires for this that and everything else. If it
wasn’t there or it didn’t work, John was the man that could fix it. Bryan
Copple for his help with everything scientific. From hypothesis to
conclusion he had an idea, which at times I was sorry to hear. John

Phipps for his kind support of all my hair brained ideas, and his ability

vi

drop the most important moments Of his day to help with my most trivial
problems. Jane Maddox for her wonderful advice on everything thing
from glue to conclusions, and of course for taking my wonderful Office,
with which I may have never wanted to graduate. And everyone else in
the lab for their caring and kind willingness to help.

I thank my family who has supported my work without question.
My mother for her persistence and love. She always wanted to help with
whatever she could even with work in the lab. My dad for his never-
ending desire to learn and think up solutions to problems. My brother,
Eric, who was always there for support when every I needed him. And for
helping me have a good time away from work. My girl, Mary, for doing
much of my lab work, keeping me going, putting up with my frustrations,
and for her love.

I thank all the members of my committee who guided my way
through graduation. Jon Hotchkiss, who was always supportive and full
of enthusiasm. Jay Goodman for his constant ability to see those
Obvious points for needed controls and tests of hypothesizes. Bob Roth
for his thoughtful points and constant ideas, which could come from the
most obscure directions. And most of all Patricia Ganey, my advisor, for
her patient understanding of a student that did things slightly different
than the rest. Also for her tireless efforts attempting to teach me the

proper way to write.

vii

TABLE OF CONTENTS

LIST OF TABLES ................................................................................ xi

LIST OF FIGURES ............................................................................. xii

LIST OF ABBREVIATIONS .................................................................. xv
CHAPTER 1

General Introduction .......................................................................... 1

1.1 Endotoxin ........................................................................... 2

1.1.1 Structure ............................................................... 2

1.1.2 Cell Signaling ....................................................... 10

1.1.3 Critical Factors for LPS-Induced Responses in Liver 24

1.1.3.1 Hepatic Response .................................... 24

1.1.3.2 Kupffer Cells ............................................ 26

1.1.3.3 Neutrophils .............................................. 31

1.1.3.4 Thrombin ................................................ 34

1.1.3.5 Summary ................................................ 41

1.2 Endotoxin Enhancement of Xenobiotic Toxicity ................... 42

1.3 Allyl Alcohol ..................................................................... 44

1.4 The Potentiation Of Allyl Alcohol-Induced Liver Injury .......... 53
CHAPTER 2

Involvement of Kupffer Cells in the Potentiation of Allyl Alcohol

Hepatotoxicity by LPS ...................................................................... 60

2.1 Introduction ..................................................................... 61

2.2 Methods ........................................................................... 62

2.2.1 Materials ............................................................. 62

2.2.2 Animals ............................................................... 63

2.2.3 KC Isolation ......................................................... 63

2.2.4 Hepatocyte Isolation ............................................. 64

2.2.5 In Vitro Coculture Experiments .............................. 65

2.2.6 Slide Coculture Experiment ................................... 66

2.2.7 Liver Homogenate Experiment .............................. 66

2.2.8 Isolation and Perfusion of Rat Livers ..................... 67

2.2.9 Statistical Analysis ............................................... 68

2.3 Results ............................................................................. 69

2.4 Discussion ....................................................................... 87

viii

CHAPTER 3
Role of Reactive Oxygen Species in the Potentiation of Allyl Alcohol-

Induced Liver Injury by LPS .............................................................. 90
3.1 Introduction ..................................................................... 91
3.2 Methods ........................................................................... 92
3.2.1 Animals ............................................................... 92
3.2.2 Hepatocyte Isolation ............................................. 93
3.2.3 Neutrophil Isolation .............................................. 94
3.2.4 Detection of GSH Concentrations .......................... 94
3.2.5 Detection of Superoxide Anion .............................. 96
3.2.6 Effect Of PMN-Derived Superoxide Anion on
Hepatocellular Viability In Vitro ...................................... 96
3.2.7 Effect of PMN-Derived Proteases on Hepatocellular
Viability ........................................................................ 97
3.2.8 Treatment In Vivo with Antioxidants ...................... 98
3.2.9 Statistical Analysis ............................................... 99
3.3 Results ........................................................................... 100
3.3.1 Effect of Superoxide on Allyl Alcohol-Treated
Hepatocytes ................................................................ 100
3.3.2 Effect of NeutrophiI-Derived Proteases on Allyl
Alcohol-Treated Hepatocytes ....................................... 112
3.3.3 Effect Of Antioxidants In Vivo ............................... 112
3.4 Discussion ..................................................................... 119
CHAPTER 4
The Role of Thrombin in the Potentiation of Allyl Alcohol-Induced Liver
Injury by LPS ................................................................................. 125
4.1 Introduction ................................................................... 126
4.2 Methods ......................................................................... 127
4.2.1 Materials ........................................................... 127
4.2.2 Animals ............................................................. 127
4.2.3 In Vivo Experiments with Heparin .......... 128
4.2.4 Determination of CINC-l and COX-2 mRNA Expression
Levels ......................................................................... 129
4.2.5 In Vivo Experiments with Large Dose Allyl Alcohol 130
4.2.6 In Vivo Experiments with Warfarin ....................... 130
4.2.7 Isolation and Perfusion of Rat Livers ................... 131
4.2.8 Statistical Analysis ............................................. 132
4.3 Results ........................................................................... 133
4.4 Discussion ..................................................................... 167

CHAPTER 5

Summary and Conclusions ............................................................. 175
5.1 Summary ....................................................................... 176
5.2 Revised Proposed Mechanism ......................................... 182
5.3 Importance of Work ........................................................ 187
BIBLIOGRAPHY ............................................................................. 191

LIST OF TABLES

Title Page

Table 4.1 Plasma ALT, AST, and fibrinogen values in rats 3 hours
after LPS administration ............................................................... 135

xi

LIST OF FIGURES

Title Page
Figure 1.1 Structure of LPS molecule ................................................ 9
Figure 1.2 Signaling cascade from LPS-stimulated TLR4 receptor ..... 23
Figure 1.3 The coagulation cascade ................................................. 36
Figure 2.1 LPS stimulates release of TNF-a from KCs ....................... 71

Figure 2.2 LPS stimulation does not enhance allyl alcohol toxicity in
cocultures of K05 and HCs ............................................................... 74

Figure 2.3 Altering the ratio Of KCs to HCs does not elicit an LPS-
induced enhancement of allyl alcohol toxicity .................................... 76

Figure 2.4 LPS-induced enhancement of allyl alcohol toxicity is not
Observed in simulated hepatic sinusoids ........................................... 78

Figure 2.5 Altering the time of incubation with allyl alcohol does
not elicit an LPS-induced enhancement of allyl alcohol toxicity ........... 81

Figure 2.6 Altering the time of incubation with LPS does not elicit
an LPS-induced enhancement of allyl alcohol toxicity ........................ 83

Figure 2.7 Perfusion with LPS does not enhance allyl alcohol toxicity
in the isolated liver .......................................................................... 86

Figure 3.1 Glutathione depletion occurs within 20 minutes after
allyl alcohol treatment of hepatocytes ............................................. 102

Figure 3.2 PMA stimulates maximal superoxide generation from
neutrophils at 6 minutes ................................................................ 104

Figure 3.3 Superoxide generation by PMA-stimulated neutrophils is
not attenuated by allyl alcohol ........................................................ 107

Figure 3.4 AIIyI alcohol-induced glutathione depletion is not
enhanced by PMA-stimulated neutrophils ........................................ 109

xii

Figure 3.5 PMA-stimulated release of reactive oxygen species
enhances allyl alcohol cytotoxicity in hepatocyte-neutrophil
cocultures ..................................................................................... 111

Figure 3.6 Conditioned medium does not enhance the cytotoxicity

of allyl alcohol ............................................................................... 114
Figure 3.7 Superoxide dismutase and catalase do not attenuate
LPS-enhanced allyl alcohol hepatotoxicity ....................................... 116
Figure 3.8 Apocynin does not attenuate LPS-enhanced allyl alcohol
hepatotoxicity ................................................................................ 118
Figure 4.1 Heparin pretreatment attenuates fibrinogen depletion in

rats cotreated with LPS and allyl alcohol ......................................... 138
Figure 4.2 Heparin pretreatment attenuates LPS-enhanced allyl

alcohol hepatotoxicity in vivo .......................................................... 140
Figure 4.3 Neutrophil infiltration is not attenuated by heparin ........ 143

Figure 4.4 ClNC-l mRNA expression is not attenuated by heparin 145
Figure 4.5 COX-2 mRNA expression is not attenuated by heparin 147
Figure 4.6 Heparin does not inhibit the toxicity of allyl alcohol alone 149

Figure 4.7 Allyl alcohol does not decrease Circulating fibrinogen
concentration ................................................................................ 151

Figure 4.8 LPS-enhanced, allyl alcohol-induced decrease in plasma
fibrinogen is blocked by warfarin pretreatment ................................ 154

Figure 4.9 Warfarin attenuates LPS-enhanced allyl alcohol
hepatotoxicity ................................................................................ 156

Figure 4.10 Thrombin does not enhance the hepatotoxicity Of allyl
alcohol in isolated, perfused livers from LPS-treated rats .............. 158

Figure 4.11 Livers isolated from LPS-treated rats are more sensitive
to allyl alcohol-induced injury than control livers ............................. 161

Figure 4.12 Thrombin is activated by 2 hours in LPS-treated rats 163

xiii

Figure 4.13 Thrombin does not enhance the hepatotoxicity of allyl
alcohol in isolated, perfused livers from rats treated with heparin
and LPS ........................................................................................ 166

Figure 5.1 Proposed mechanism for LPS-enhancement of allyl
alcohol hepatotoxicity .................................................................... 186

xiv

a
AA
AALAS
ADP
ALT
ANOVA
AP-l
AST
ATP
BPI

C

Ca++
CETP
CINC-l
C02
COX-2
DAG
DD
ECSIT
ERK
Fe3+
fMLP
8

Gal
GBS
GdCI3
GIc
GICNAc
GSH
GSSG
HBSS
HCs
Hep
IKK
lL-lR
ILs

ip
IRAK
iv

IX

JNK

LIST OF ABBREVIATIONS

activated factor

allyl alcohol

American Association of Laboratory Animal Sciences
adenosine diphosphate

alanine aminotransferase

analysis of variance

adaptor protein-1

asparate aminotransferase

adenosine triphosphate
bactericidal/permeability-increasing protein
Celsius

calcium

Cholesterol ester transfer protein
cytokine-induced neutrophil chemoattractant
carbon dioxide '
cyclooxygenase-2

diacylglycerol

death domain

Evolutionarily-Conserved Signaling Intermediate in Toll
extracellular regulated kinase

iron
n-formyl-methionyl-Ieucyl~phenylaIanine
gravity

D-galactose

Gey's balanced salt solution

gadolinium Chloride

D-glucose

N-acetyI-D-glucosamine

glutathione (reduced form)

glutathine (oxidized form)

Hanks' balanced salt solution

hepatocytes

Heptose

IK-B kinase

interleukin-1 receptor

interleukins

intraperitoneal

lL-1 receptor associating kinase

intravenous

plasma thromboplastin component (Christmas factor)
C-Jun NH2-Terminal kinase

XV

Kupffer cells

2-ketO-3-deoxyoctonic acid

kilogram

liter

LPS binding protein

Iipopolysaccharide

molar

D-mannose

mitogen-activated protein

mitogen-activated protein kinase kinase kinase
membrane-bound C014

mitogen-activated protein kinase/ERK kinase kinase
microgram '
milligram

micromolar

millimolar

minute

mitogen-activated protein kinase kinase
number

nicotine adenine dinucleotide, reduced form
nicotine adenine dinucleotide phosphate, reduced form
nuclear transcription factor kappa B
nanogram

NF-KB-inducing kinase

molecular oxygen

superoxide anion

platelet activating factor

protease-activated receptor
phosphatidylcholine-specific phospholipase C
prostaglandin E2

prostacyclin

phosphafidyhnosnOLB

phospholipid surface

phospholipase A2

phospholipid transfer protein

phorbol myristate acetate

neutrophil

rhamnose

reactive oxygen species

reverse transcriptase-polymerase chain replication
soluble CD14

standard error of the mean

superoxide dismutase

xvi

TAKl
TGFa
11R
TLR4
TLRs
TNsz
TRAF6
VH

XI
XII

transforming growth factor-B activated kinase
transforming growth factor a
TON/His

Toll-like receptor 4

Toll-like receptors

tumor necrosis factor or

TNF receptor-associated factor
proconverhn

Stuart-Prower factor

plasma thromboplastin
Hageman factor

xvii

CHAPTER 1

General Introduction

INTRODUCTION

1.1 Endotoxin

1.1.1 Structure

Endotoxin is a component Of gram-negative bacteria that elicits a
strong inflammatory response in mammals. The term endotoxin was
first coined by Richard Pfeiffer in 1893 to describe the toxic component
found in the Iysates of heat-inactivated Vibrio Cholerae. This toxic
component was different from other known exotoxins in that it was heat
stable and its toxicity was only found in Iysed cells, indicating that it may
be attached to the cell body. Because the first endotoxin components
identified were polysaccharide and lipid, the endotoxins were also called
Iipopolysaccharides (LPS). This molecule is found in the outer portion of
cell wall Of gram-negative bacteria. One bacterial cell contains
approximately 3.5 X 105 LPS molecules that occupy an area of about 4.9
pmz. Because the surface Of the cell is about 6.7 pmz, LPS comprises
approximately three-quarters of the bacterial surface (Nikaido et al,
1987).

A mutation in the early steps Of LPS synthesis is fatal to the
bacterium, indicating that LPS is an essential component Of the outer

membrane. LPS represents the main surface antigen (O-antigen) of

gram-negative bacteria. It is essential for the specific recognition and
elimination of gram-negative bacteria from the host. It may be that,
because gram-negative bacteria display this essential component on
their surface, the host defense mechanisms were able to evolve an ability
to recognize this component at very low concentrations in the plasma.
LPS isolated from different species of gram-negative bacteria elicit
different pathological and physiological activities. Depending on the
exact structure, LPS may also help prevent activation of complement
and phagocytosis by macrophages, thereby protecting the pathogen
from cellular host defenses.

Over the past several decades scientists have investigated the
structural components of LPS to understand better the ability Of host
cells to recognize gram-negative bacteria. LPS structure can be broken
down into 4 basic segments: the O-speCifiC Chain, outer core, inner core,
and lipid A. Each segment is different in structure and biosynthesis and
elicits different functions and biological activities. A brief understanding
of each of these segments will be described. For additional information
the reader is encouraged to read any of the following reviews: Morrison
et al, 1992; Decker, 1997; Haeffner-Cavaillon, 1999; Darveau, 1998;
Raetz, 1988; Rietschel, 1994; Wang, 1996; Fenton, 1998; Tobias, 1999;
Ulevitch, 1999; Cunningham, 2000; Antal-Salma's, 2000; Ulevitch,

2000; Doun, 2000; Medzhitov, 2000; Beutler, 2000; Means, 2000.

The O-speCific region is a carbohydrate polymer consisting of up
to 50 Oligosaccharides (Figure 1.1). This oligosaccharide chain is made
up of many repeated units, each unit consisting of up to seven
monosaccharides. The structure of the O-specifiC region is highly
variable among species and even within a single bacterium. Variation
can occur in the ring form, linkage type, monosaccharide,
monosaccharide sequence within a repeating unit, number of repeating
units, and extent Of branching. The diversity of sugar constituents and
the variability of linkage sites allows for a vast number of possible 0-
specific Chain structures (Rietschel et al, 1982).

O-specific Chains are either synthesized on the surface of the
bacteria or synthesized within the bacterium and transported to the
surface (Mulford ef al, 1983). S. typhimurium O-specific chain
biosynthesis occurs by first making oligosaccharide repeating units and
then connecting the units using a polymerase found on the periplasmic
surface Of the inner membrane. Elongation Of the chain occurs at the
reducing end. In other bacteria (i.e. Samone/Ia) elongation occurs at the
nonreducing end (Makela et al, 1984).

A Cluster of genes termed rfb encodes proteins critical for the
synthesis of the O-specific Chain. Mutants with a defect in the rfb gene
synthesize LPS that lacks the O-speCifiC region. LPS lacking the 0-

specific region is called R-form LPS because it produces bacteria that

look rough on their surface. These mutant bacteria are able to grow in
vitro, indicating that the O-specific Chain is not required for viability.
However, when these mutants are put into tissue or body fluids they do
not survive, suggesting a protective role for the O-specific chain (Jann et
al, 1984).

The O-specific Chain plays a role in many of the biological
properties of LPS. For example, these chains determine the serological
O-specificity of LPS and can act as receptors for bacteriophages
(Luderity et al, 1982; Lindberg et al, 1983). Some bacteria utilize their
O-specific chain to help bind to their leguminous host (Wolpert et al,
1976) or inhibit attachment of the C5b-9 membrane attack complex
(Makela et al, 1990), which makes it important for bacterial virulence.
The O-specific chain of other bacteria is able to activate the alternative
complement pathway (i.e. phagocytosis via C3b) leading to removal from
the host (Grossman et al, 1984). Due to the large variability in the
structure of this region of LPS, hosts may not have been able to evolve to
recognize this region with specificity. Thus, the O-specific region is not
critical for host recognition and often serves to protect the bacteria from
recognition and can even down-regulate the endotoxic activity of LPS
(Morrison et al, 1985; Luderitz ef al, 1989).

The core region of LPS is a branched heteropolysaccharide

without repeating glycosyl structures. Its biosynthesis starts at the lipid

A-bound 2-ketO-3-deoxyoctonic acid (Kdo). The chain then elongates by
the addition of activated glycosyl residues to the nonreducing end of the
Chain (Jann et al, 1984). This biosynthesis occurs independently of and
with separate pathways from O-specific chain synthesis. The core
oligosaccharide can be broken into two regions based on the type of
sugars incorporated; the outer core and the inner core.

Most species variation in core structure is found in the outer core.
The outer core is also called the hexose region due to its content of
various hexoses such as D—glucose, D-galactose, and N-acetyl-D-
glucosamine (Figure 1.1). Not much is known about the biological
activities that are elicited by this region. However, it can recognize an
Ra-specific serum factor (Ihara et al, 1982) and may be involved in
binding certain bacterial mutants to lymphocytes (Lehmann et al, 1980;
Jirillo et al, 1990).

The inner core connects the outer core and O-specific region to the
lipid A region (Figure 1.1). It is made up of heptose sugars such as L-
glycero-D-manno-heptose. The structure of the inner core varies across
species but all gram-negative bacterial LPS contain at least one a-bound
pyranosidic or furanosidic Kdo residue. This Kdo group is always found
in a lipid A-proximal location and serves to anchor the polysaccharide to

the lipid portion of LPS. The inner core structure has a slightly negative

Charge which is responsible for the negative charge of LPS. This
charged region of LPS aids in membrane stability.

The inner core region is a rather conserved region of LPS. This
may have enabled the host to evolve the mechanisms that recognize this
region of LPS. In fact this region is required for some Of the biological
activities of lipid A and for bacterial viability (Haeffner-Cavaillar et al,
1989; Vaara et al, 1984; Osborn MJ, 1979).

The lipid A region of LPS represents the covalently bound lipid
component (Figure 1.1). Though its structure varies among species and
within a given LPS preparation, much of the structural format remains
similar. Nearly all lipid A contains gluco- configured and pyranosidic- D-
hexosamine residues which are usually present as a B(1-6)-Iinked
disaccharide. This disaccharide makes up the backbone of the lipid.
Hydroxy fatty acids are linked to the disaccharide via amine or ester
bonds. Typically the hydroxy fatty acids are 10 to 28 carbons long and
are often acylated at their 3-hydroxyl group. The amount and position of
fatty acid acylation and phosphorylation play key roles in host-cell
responses to LPS (Takada et al, 1992). Some structural prerequisites
for bioactivity have been elucidated by analyzing the biological activities
of synthetic lipid A molecules. In general, a biologically active lipid A will
contain two hexosamine residues, two phosphoryl groups, and six fatty

acids in a defined location as it is presented in E. coli.

Figure 1.1. Structure Of LPS molecule.

Structure depicting the four major segments of LPS the 0-
specific chain, outer core, inner core, and lipid A. Abbreviations:
Glcz D-glucose; Gal: D-galactose; GICNAcz N-acetyl-D-glucosamine;
Hepzheptose; Kdo: 2-keto-3-deoxyoctonic acid; Man: D-mannose;
Rha: rhamnose; n: the number of repeating units may vary from 0
to approximately 50. (adapted from Haeffner-Cavaillon et al,

1999)

Inner Core

f—‘——~_'\

Outer Core P' ”WW“

f—~'——\

COOH

Em
Lipid A

WEN . ,__..__.
Wfiéf : .n. .
3., 3&1:

W
[(1.3) GIcNAc (1.2) wan (1.3) Gal (1.4) m] Chain
0

Cu°H

Cu
cu CuOH C..OH

Cu

Lipid A represents the least variable region of LPS. This may be the
reason that Lipid A contains the endotoxic principle of LPS. Lipid A has
been shown to induce the various Classical effects Of endotoxin such as
fever, local Schwartzman reactivity (a standard in vivo test of LPS tissue
toxicity), and lethal toxicity. In addition to all Of the physiological effects
elicited by LPS, the lipid A region is essential for bacterial viability

(Osborn MJ, 1979).

1.1.2 Cell Signaling

The ability of cells to recognize LPS has been an area of intense
study for many years. The cellular receptor for LPS has only recently
been discovered and is thought to be Toll-like receptor 4 (TLR4). The
ability of cells to detect levels of LPS in the picogram per milliliter range
is due to the LPS carrier proteins LPS binding protein (LBP) and CD14
which effectively deliver LPS to its receptor. A brief review of the
mechanism by which these proteins interact to elicit LPS detection by
cells follows as reviewed by: Tobias, 1999;Ulevitch, 1999; Cunningham,
2000; AntaI-Szalma’s, 2000; Ulevitch, 2000; Daun, 2000; Medzhitov,
2000; Beutler, 2000; Means, 2000.

LBP is a 61-65 kDa glycoprotein, acute-phase reactant that is
found in the blood at concentrations of about 220 pg/ml (Gallay et al,

1993, Myc et al, 1997, Calvano ef al, 1994). The levels Of LBP can

10

increase 50-100 fold during inflammation by its formation mainly in the
liver and to a lesser extent in the lungs and other tissues (Martin et al,
1997). LBP is structurally very similar to three other proteins:
bactericidal/permeability-increasing protein (BPI), Cholesterol ester
transfer protein (CETP), and phospholipid transfer protein (PLTP) (Day
at al, 1994; Tobias et al, 1988; Gray et al, 1989). Because no x-ray
crystallography has been done on LBP, much Of the knowledge of its
structure is based on the x-ray crystallography of the very similar BPI
protein. Like LBP, BPI will also bind LPS molecules, however it does not
transfer them to CD14. The BPI protein will deposit LPS molecules back
onto LPS aggregates, effectively neutralizing LPS (Tobias et al, 1997).
According to its crystallographic structure, BPI contains two apolar
pockets that interact with the acyl Chains of phosphatidylcholine
(Beamer et al, 1997). These apolar pockets could be the sites of
interaction with LPS for both BPI and LBP. These proteins appear to
associate with the fatty acid portion of LPS. Although LBP binds to LPS
from different bacterial strains with differing affinities (Cunningham et al,
1996), this is not thought to be the biological mechanism by which the
host can differentiate among LPS molecules.

In general, LBP binds LPS entering the blood in very small
concentrations and helps present it to cells expressing C014. LPS binds

to 0014 as a monomer, however LPS is found as aggregates in aqueous

ll

environments such as tissue culture medium or blood. These
aggregates spontaneously release LPS monomers at a very slow rate.
The addition of LBP dramatically accelerates the binding of LPS
monomers from LPS aggregates to C014 (Hailman et al, 1994), and in
this capacity LBP increases the sensitivity of cells to LPS (Wright et al,
1991). Though LBP can transfer LPS to membrane-bound C014 and
form an LBP-LPS-C014 complex, LBP does not form a complex with the
soluble C014 (sC014). LBP can also transfer LPS to lipoproteins, which
functionally neutralizes LPS (Wurfel et al, 1994; Munford et al, 1982),
and in these instances LBP decreases the sensitivity of cells to LPS.

The importance Of LBP is demonstrated by its ability to increase
the sensitivity Of cells to LPS. For example, the threshold concentration
of LPS required to induce tissue factor release from a monocyte cell line
was reduced 20-fold by purified LBP (Steinemann et al, 1994). In
addition, antibodies to LBP prevented LPS-induced activation of cells
and lethality from LPS administration (Gallay et al, 1994). Blood from
LBP knock-out mice was less sensitive to LPS-stimulated release of TNF-
a, but LBP knock-out mice responded similarly to wild type mice during
LPS exposure in vivo (Wurfel et al, 1997). These contrasting results may
be explained by local production in vivo Of the sC014 or some alternate
protein that can act as an LPS carrier. However, galactosamine-

sensitized LBP knock-out mice were less sensitive to LPS than

12

heterozygous controls, indicating the importance of LBP. In addition,
LBP knock out mice were more sensitive to Salmonella-induced death,
indicating that LBP plays a critical role in host defense (Fenton et al,
1998). Despite the inconsistency of these results, it is clear that LBP
plays an important role in the recognition of LPS.

0014 is a membrane glycoprotein of 356 amino acids with a 19
amino acid N-terminal leader peptide. Its gene is found on Chromosome
5 in the region q23-21. It has two exons and encodes a single 1.4 kb
mRNA transcript (Ferrero et al, 1988). After translation, 28-30 amino
acids of the C-terminal end are replaced with a glycosyl-phosphatidyl
inositol group which serves to anchor C014 to the membrane but does
not facilitate a transmembrane signal (Haziot et al, 1988). Although only
one mRNA transcript has been reported for C014, different subtypes
have been proposed based on differences in antigenicity, LPS binding
affinity, and sensitivity to enzymatic cleavage (Pedron et al, 1995). Cells
known to express 0014 on their surface include monocytes,
macrophages, and neutrophils (Goyert et al, 1986).

The number of C014 molecules on a human neutrophil has been
estimated to be about 3000 (AntaI-Szalmas et al, 1997) and on
monocytes between 6000 and 190,000 (Marchant ef al, 1992, Couturier
et al, 1991). The number of membrane bound C014 receptors varies

between monocytes of different tissues. Peritoneal, pleural, and

13

perivascular brain macrophages express large numbers, whereas
microgliae, alveolar macrophages, Kupffer cells, and macrophages of the
intestinal mucosa express smaller copy numbers of C014 (Antal-
Szalmas et al, 2000).

The expression of C014 can be altered by various stimuli. Tumor
necrosis factor-alpha, granulocyte-colony stimulating factor, n-formyl-
met-leu-phe (fMLP), and LPS can upregulate 0014 about two fold in
neutrophils. In monocytes, interleukin-4 and interleukin-13 decrease
C014 mRNA, whereas interferon-or, interferon-y, interleukin-2, and
transforming growth factor beta induce a rapid increase in C014. LPS
also alters the monocytic-expression of C014; however, the response
appears to be dependent on dose and time. At early time points (30180
min.) LPS causes a rapid increase in C014 expression (SO-100%)
followed by a decrease (SO-75%) at 3-6 hours. Finally, C014 expression
increases dramatically (ZOO-300%) after 1-6 days.

sC014 exists in two different forms and can be found in the
plasma at concentrations Of 3-5 pig/ml (Kruger et al, 1991). One form is
produced from the shedding of membrane bound C014 (mC014) as a
result of various stimuli including PMA, IFN-y, Ca2+-ionophore A23187,
or LPS. This stimulus-initiated shedding is thought to involve a
membrane-associated serine protease and results in a protein with a

molecular mass of 48-49 kDa. The second form consists of C014

14

molecules that escape glycosyl-phosphatidyl inositol anchor attachment,
retaining their C-terminal leader which results in a protein with a
molecular mass of 55-56 kDa (Haziot et al, 1988). This form may be
stored intracellularly. In neutrophils sC014 molecules appear to be
stored in secretory vesicles and azurophilic granules. Their localization
in monocytes is more diffuse and in the perinuclear region (Detmers et
al, 1995). There as yet does not appear to be any biologically functional
difference between the two forms Of 50014.

C014 binds and shuttles a wide variety of microbial ligands. The
mechanism by which C014 binds to LPS is not entirely known. There
are no crystallographic data available for C014; based on what is known
about the sequence for C014, it is thought to bind the polar portion of
LPS. There are many regions within the amino terminal portion of 0014
that are required for LPS binding (Shapiro et al, 1997). Only the first
152 amino acids Of the 356 amino acids in 0014 are important for
functionality (Viriyakosol ef al, 1996; Juan et al, 1995). Charge reversal
mutations are the most effective at reducing LPS binding affinity
(Shapiro et al, 1997). It is thought that many of the critical regions are
on the same side of the molecule forming a charged pocket that
effectively works as a selective molecular flypaper, similar to

macrophage scavenger protein (Cunningham et al, 2000).

15

There is a vast body Of data supporting a central role for C014 in
LPS responses. C014 has a high affinity for and binds stoichiometrically
to LPS (Hailman et al, 1994). Specific monoclonal antibodies against
0014 inhibit LPS-enhanced phagocytosis (Wright et al, 1990; Lynn et al,
1991; Wright et al, 1991; Kitchens et al, 1992). 50014 is found in the
blood and it allows for the activation of cells that do not express 0014
(Bazil et al, 1989; Frey ef al, 1992; Arditi et al, 1993; Haziot et al, 1993).
Transfection Of C014 into cell lines enhances their responsiveness to
LPS more than 1000-fold (Lee et al, 1992; Golenbock et al, 1993;
Viryakosol et al, 1995). C014-deficient knock-out mice are 10,000-fold
less sensitive to LPS than wild-type mice (Haziot et al, 1996; Haziot et al,
1995). Whereas small concentrations of sC014 increase the sensitivity
of cells to LPS, high concentrations of sC014 can act to neutralize LPS
activity (Troelstra et al, 1997).

0014 was originally thought to be the receptor for LPS, however,
it was proven to assist merely in the recognition of LPS by the actual
LPS receptor (Kitchens et al, 1995). Until recently, the LPS receptor
remained unknown. It is now known to be TLR4.

The discovery of the LPS receptor occurred rather serendipitously.
Initial studies in Drosophilla melanogaster showed that a protein family
known as Toll played a critical role in innate immune responses to

microbial pathogens. This finding prOmpted the search for a human

16

homologue of the Toll receptor family. Human homologs were
discovered and named Toll-like receptors (TLRs) (Medzhitov et al, 1997;
Chaudhary ef al, 1998). The members of the TLR family, TLR2 and
TLR4, have been found to respond to microbial pathogens (Kirschning ef
al, 1998; Yang et al, 1998; Chow ef al, 1999).

Structurally the TLR group contains transmembrane regions with
leucine-rich repeats in the extracellular region, and a hydrophobic
transmembrane domain (Rock et al, 1998). The cytoplasmic tail is
similar to the interleukin-1 receptor (IL-1R) and is termed the TOII/lL-lR
(TIR) domain. This region shows 45% sequence identity with IL-1R (Gay
et al, 1991). It is estimated that there are 1000 or fewer molecules of
TLR4 on a cell. Currently the amount Of TLR4 expressed on the surface
Of cells is thought to regulate the limit of LPS sensitivity. Little is known
about the regulation Of the TLR4 gene.

Shortly after the discovery of human homologs to the Toll
receptor, their potential contributions toward the immune system were
investigated. Two groups independently showed that TLR2 supports a
COM-dependent LPS signal that activates nuclear factor kappa B (NF-
KB) using a pathway similar to lL-1R (Yang et al, 1998; Kirschning et al,
1998). Another group found that hamster cells carrying a null allele for
TLR2 are capable of responding to LPS (Heine et al, 1999). These

findings suggested that while TLR2 is sufficient for an LPS response, it is

17

not required. It was later found that activation of TLR2, in the presence
of LPS, may have actually been stimulated by bioactive contaminants
found in commercial preparations of LPS. Deletion of TLR2 in mice
resulted in loss of peptidoglycan recognition and responsiveness with no
inhibition Of LPS response. TLR2 recognized a wide variety Of other
microbial products in vitro, including components of gram-positive
bacterial cell walls, zymosan, bacterial lipoproteins, as well as
mycobacteria and glycolipids derived from mycobacterial cell walls
(Underhill et al 1999; Schwandner et al, 1999; Aliprants et al, 1999;
Brightbill et al, 1999; Flo ef al, 2000). While it appears that TLR2 can
recognize many different molecules, it must be remembered that neither
in vitro transfection nor targeted deletion Of individual TLRs can tell the
specificity of a single TLR type. It has been hypothesized that some
TLRs may form heterodimers which would influence ligand specificity. In
general, TLR2 is thought to recognize predominately gram-positive
bacterial cell wall components. Thus there had to be some other LPS-
responsive receptor.

Beutler and coworkers suggested the role for TLR4 in LPS
signaling (Poltorak et al, 1998). There is a large amount of data
supporting a role for TLR4 in LPS signaling; the most convincing data
come from LPS-resistant mouse strains. The defect in the Lps gene that

caused LPS resistance in the mouse strains C3H/HeJ and

18

C57BL/IOScCr was found to encode the TLR4 protein. In the C3H/HeJ
strain there is a proline-histidine change at position 712 and in the
C57BI/IOSCCr strain there is a null mutation in the TLR4 gene (Poltorak
et al, 1998). In summary, mutation of TLR4 in mice resulted in a
complete loss of LPS responsiveness.

TLR4 intracellular signaling has been the most extensively studied
LPS signaling pathway and reviewed by Ulevitch, 2000; Medzhitov,
2000; Means, 2000 (Figure 1.2). LPS is well known to stimulate the
phosphorylation of p38, extracellular regulated kinase (ERK) 1 and 2, C-
Jun NHz-Terminal kinase (JNK), adaptor protein-1 (AP-1), and NFx-B.
Most of what is known about this signaling pathway involves the proteins
associated with the activation of NF'KB. A protein called MD-2
associates with TLR4 to aid in LPS stimulation. MD-2 lacks a
transmembrane domain but can form a complex on the extracellular
domain Of TLR4 (Akashi et al, 2000; Shimazu et al, 1999). Upon
stimulation the protein My088 associates with TLR4 via an interaction at
the TIR region (Muzio et al, 1998; Medzhitov et al, 1998). My088,
originally isolated as a myeloid differentiation primary response gene,
contains both an N-terminal death domain and a C-terminal TIR domain
(Daun JM, 2000). My088-deficent mice are resistant to the lethal effect
of LPS, and macrophages from these mice do not produce lL-6, TNF-or,

or NO in response to LPS (Kawai et al, 1999). However, LPS is capable

l9

of signaling through both My088-dependent and -independent pathways
(Daun JM, 2000). My088 recruits lL-l receptor associating kinase
(IRAK) to the TLR4 receptor. IRAK is a serine/threonine protein kinase
that contains an N-terminal death domain. It is the homophilic
interaction between My088 and IRAK death domains that leads to
association and autophosphorylation of IRAK (Wesche et al, 1997). This
stimulates the activation of the downstream kinase TNF receptor-
associated factor (TRAF6), which leads to the stimulation of the
transforming growth factor-B-activated kinase (TAKl) (Cao et al, 1996;
Ninomiya-Tsuji et al, 1996). TAKl activation leads to induction of NF-
KB-inducing kinase (NIK), which leads to the activation of the Ix-B
kinases IKK1 and IKK2 (Ninomiya-Tsuji ef al, 1996). Activated lKKs
phosphorylate IK-B and stimulate its degradation through the
proteasome pathway leading to the release of NFK-B (NinomiyaTsuji et
al, 1996). Once NFx-B is activated and translocates to the nucleus many
Of the genes involved in the inflammatory response are activated.

Other portions of LPS-stimulated signal transduction have been
discovered, however, much more work must be done to get a clear
picture Of the intracellular pathways activated by LPS. Some of the
other portions of LPS-stimulated pathways include TRAF6 association
with Evolutionarin-Conserved Signaling Intermediate in Tollipathways

(ECSIT). ECSIT appears to bridge TRAF6 to the mitogen activated

20

 

 

protein kinase kinase kinase (MAP3K), MEKK-l via a direct interaction
(Kopp et al, 1999). MEKK-l also can activate JNK (Means TK, 2000).
Alternatively JNK may be activated by TAK.l/TABl, which are thought to
activate the mitogen activated protein kinase kinases MKK3/6 and MKK4
that in turn activate p38 and JNK respectively (Ninomiya-Tsuji et al,
1999; Muzio etal, 1998). MAPKs can be activated by LPS independently
of My088 (Means TK, 2000). Along other pathways LPS stimulates the
activation Of phosphatidylinositol-3 (Pl3) kinase. Pl3 kinase may
associate with the NlK/IKK complex leading to lK-B phosphorylation and
degradation (Figure 1.2). LPS stimulation also results in the breakdown
of phosphatidylcholine to the second messengers diacylglycerol and
ceramide, which leads to mitogen activated protein (MAP) kinase
activation (Daun JM, 2000). A phosphatidylcholine-specific
phospholipase C (PC-PLC) is important for ERK activation with LPS-
stimulation (Daun JM, 2000). It is easy to see that LPS stimulates
many different intracellular pathways. This causes the stimulated cells
to produce many different acute phase proteins and other mediators that

can alter the physiology of the cells and the organs surrounding them.

21

Figure 1.2. Signaling cascade from LPS-stimulated TLR4

receptor.

Schematic diagram representing the signaling cascade of
the TLR4 receptor initiated by LPS as described in the text.
Abbreviations: LPS, lipopolysaccaride; LBP, lipopolysaccaride
binding protein; TLR4, toll-like receptor-4; TIR, toll/interleukins-l
receptor; 00, death domain; IRAK, interleukins-1 receptor
associating kinase; TRAF6, tumor necrosis factor receptor-
associated factor-6; TAKl, transformaing growth factor B-activated
kinase; NIK, NFK'B inducing kinase; IKK, I-KB kinases; ECSIT,
evolutionarily-conserved signaling intermediate in toll pathways;
MKK, mitogen activated protein kinase kinase; MEKK-l, mitogen
activated protein kinase/ERK kinase kinase; ERK, extracellular
regulated kinase; JNK, C-Jun NHz-terminal kinase; DAG,
diacylglycerol; PC-PLC, phosphatidylcholine-specific

phospholipase C.

22

Extracellular

 

 

 

W

 

Intracellular
TRAF6
I
\

 

 

A gene expression

23

 

1.1.3 Critical Factors for LPS-Induced Responses in Liver

LPS stimulates cells through a receptor-mediated process to
release a large array of cellular mediators. The physiological effects
induced by LPS are many. Only the effects of LPS on the liver will be
discussed here. In the liver LPS can induce injury at sufficient doses.
Liver injury induced by LPS is critically dependent on the activity Of
several factors in the liver including Kupffer cells, neutrophils, thrombin,
and hepatocytes (Fujita ef al, 1995; limuro et al, 1994; Hewett et al,
1992; Sato et al, 1993; Pearson et al, 1996; Moulin et al, 1996). Each of

these factors will be discussed further below.

1.1.3.1 Hepatic Response

Although the exact mechanisms are unknown, LPS stimulates the
release of many cellular mediators including complement factors,
clotting factors, oxygen radicals, lysosomal enzymes, nitric oxide,
arachidonic acid metabolites, platelet activating factor (PAF), tumor
necrosis factor-or (TNF-a), interleukins (lLs), and endothelin. The
number of physiological effects elicited by such mediators is enormous.

After LPS administration, various receptors are down-regulated
including hepatic vasopressin, al-adrenergic, prostaglandin F2a, and

glucocorticoid receptors. Other receptors become sensitized such as

24

receptors for glucagons, insulin, and TNF-a. Cells exposed to LPS
exhibit increased sensitivity to cyclic AMP-dependent hormones, which
may be due to LPS-induced increase in membrane-bound adenylate
cyclase activity.

Upon LPS administration in vivo, inflammatory cells release
mediators that stimulate the production of acute phase proteins.
Medium from LPS treated monocytes is capable of inducing hepatocytes
to produce acute phase proteins, supporting the contention that
stimulation of hepatocytes in vivo is mediated through the release of
cytokines from monocytes. Three main cytokines released from Kupffer
and other cells that are responsible for induction of acute phase proteins
are IL-1, IL-6, and TNF-a. Hepatocytes are stimulated to increase the
synthesis of many acute phase proteins and decrease the synthesis of
negative acute phase proteins, such as albumin. Synthesizing results
from many labs it is apparent that the net effect on total protein
synthesis is dependent on the dose of LPS administered and the time
proteins are measured. At early time points there is an increase in total
protein synthesis.

Glycogen depletion is an early event after LPS administration that
causes hyperglycemia. This is stimulated by catecholamines and
prostaglandins at an early phase and by glucagons at a later phase of

endotoxemia. Additionally, gluconeogenesis is increased in LPS-treated

25

animals. Cyclooxygenase products are thought to play a role in the LPS-
stimulated glycogenolytic burst.

It also affects lipid metabolism in hepatocytes. LPS causes a
switch in fatty acid synthesis from oxidation to esterification. It also
causes a marked increase in triacylglycerolemia, which is partly due to a
decrease in lipoprotein lipase activity and an increase in fatty acid

synthetase activity.

1.1.3.2 Kupffer Cells

Kupffer cells originate from promyloblast stem cells in the bone
marrow. Once the cells differentiate into monocytes they are released
into the Circulation. The monocytes may circulate for several days
before settling in various organs where they differentiate into tissue
macrophages. The macrophages that settle in the liver are called
Kupffer cells. These represent 80% of all resident macrophages, and
they represent about 10 to 12 percent of the liver cell population.

In addition to LPS, Kupffer cells play an important role in the
development of liver injury from many different xenobiotics. For
example, activating Kupffer cells with retinol enhances carbon
tetrachloride and vinylidene chloride hepatotoxicity (Badger ef al, 1996;
Wueweera et al, 1996). Furthermore, inhibition of Kupffer cells with

gadolinium chloride, dextran sulfate, or glycinecontaining diet protects

26

against hepatotoxicity induced by carbon tetrachloride, vinylidene
Chloride, acetaminophen, D-galactosamine, 1,2-dichlorobenzene, LPS,
ethanol, diethyldithiocarbamate, and ischemia reperfusion (Badger et al,
1996; Wueweera et al, 1996; Michael et al, 1999; Stachlewitz ef al, 1999;
Hoglen et al, 1998; Sarphie et al, 1996; Fujita et al, 1995; limuro ef al,
1994; Bautista et al, 1999; Hisama et al, 1996; lshiyama et al, 1995;
Adachi et al, 1994).

The mechanism by which Kupffer cells contribute to liver injury is
not fully understood. l have found that the most common mechanisms
reported are the generation of oxidative stress and the release of TNF-a.
Kupffer cell-derived oxidative stress is important for the toxicity of many
compounds including acetaminophen, 1,2-dichlorobenzene, LPS, and
diethyldithiocarbamate (Michael et al, 1999; Hoglen et al, 1998; limuro
et al, 1994; lshiyama et al, 1995). TNF-a release is important for the
toxicity of other compounds such as D-galactosamine (Stachlewitz et al,
1999).

Though Kupffer cells can increase the toxicity of many xenobiotics,
they can also protect against liver injury from xenobiotics. Based on a
review of the literature it appears that the ability of the Kupffer cell to
protect'the liver as opposed to contributing to injury is dependent upon
the time at which the Kupffer cells are activated relative to chemical

exposure. Activation Of Kupffer cells with Corynebacterium parvum five

27

days before treatment protected against carbon tetrachloride and
acetaminophen toxicity. This protective effect may be due to inhibition
of microsomal oxidizing enzymes and an increased production of
glutathione (Pereira et al, 1997).

Kupffer cells are the major cell type responsible for LPS clearance.
After an i.v. injection Of LPS, 100 times more LPS is found in Kupffer
cells than in endothelial cells and 1000 times more than in hepatocytes
(Fox, 1990). Most Of the effects of LPS administration have been
attributed at least in part to signal molecules generated by resident
macrophages and by blood monocytes (Decker, 1997). Upon
stimulation with LPS, Kupffer cells change from a quiescent state to an
inflammatory or active state. While in the quiescent state, Kupffer cells
release only low levels of apolipoprotein E and prostanoids. They also
respond very weakly to external stimuli. Upon stimulation with LPS they
become swollen and contain an increased number of cytoplasmic
lysosomal granules and phagocytic vacuoles. The endoplasmic
reticulum appears dilated and damage to the plasma and nuclear
membranes is also apparent at large doses of LPS. In the active state
Kupffer cells increase their secretion of mediators and are much more
sensitive to external stimuli. Kupffer cells are the major suppliers of
LPS-stimulated TNF-a in the liver. Upon LPS-stimulation they also

release IL-lp, IL-6, lL-8, lL-lO, Interferon-OMB, TGF-a, TGF-p, PGEZ,

28

PGDz, PGFZa, PGIZ, thromboxane A2, platelet activating factor,
superoxide, and nitric oxide (Decker, 1997).

Though no one single cytokine can be responsible for all of the
pathological effects Of LPS, TNF'G is widely considered the most relevant
and typical cytokine mediating the effects of LPS. Plasma TNF-or levels
reach their peak at about 90 minutes after LPS administration and
administration of TNF-or elicits many effects similar to those of LPS.
Furthermore, the effects Of LPS administration can be blocked by
inhibition of TNF-a (Decker, 1997; Stachlewitz et al, 1999; Hewett et al,
1993).

The effects of lL-lp are very similar to those of TNF-a. lL-lfl plays
a role in antigen presentation by increasing adhesion molecules on
antigen presenting cells. Glucose homeostasis can also be modulated
by lL-1. Blockage of lL-IB inhibits the effects of LPS (Decker, 1997). IL-
1 from Kupffer cells stimulates hepatocytes to release the Chemokine
cytokine-induced neutrophil chemoattractant-1 ClNC-l (Mawet et al,
1996)

lL-6 plays a role in enhancing the synthesis of acute phase
proteins. The amount of lL-6 produced by Kupffer cells is only a fraction
of what blood monocytes produce; however, Kupffer cells are spatially

much Closer to hepatocytes and are thus thought to play a critical role in

29

the regulation of acute-phase protein synthesis in hepatocytes (Decker,
1997).

Platelet activating factor released from LPS-stimulated Kupffer
cells stimulates the Kupffer cells to form oxygen free radicals and
peripheral blood monocytes to produce TNF-or (Lo et al, 1997). Platelet
activating factor plays a role in the early stages of neutrophil infiltration
Of the liver (Coughlan ef al, 1994; Zimmerman et al, 1994; Lorant et al,
1991). Platelet activating factor primes neutrophils to respond to
secondary immunological stimuli more robustly (Kulikov et al, 1998). It
is thought to play a role in the early stages Of endotoxemia (Balsa et al,
1997; Terashita et al, 1992, Qi et al; 1990).

Prostaglandins are released from Kupffer cells upon LPS-
stimulation of phospholipase A2, which provides arachidonic acid by
Cleavage of membrane-attached phospholipids. The major
prostaglandins released from LPS-stimulated Kupffer cells are PGE2 and
PGDZ. These prostaglandins are involved in vasodilation and are also
thought to play a role in the regulation of inflammation (Decker, 1997).

Kupffer cells can release nitric oxide upon stimulation by LPS.
Nitric oxide leads to the relaxation Of contracted hepatic stellate cells. It
can also be converted to the highly toxic molecule peroxynitrite in the
presence of superoxide. Peroxynitrite plays a role in the development Of

tissue injury in many models of toxicity (Decker, 1997).

30

LPS alone does not stimulate superoxide production by Kupffer
cells. LPS enhances phorbol ester- or phagocytotic stimuli-activated
production of superoxide catalyzed by membrane-bound nicotinamide
adenine dinucleotide phosphate (NADPH) oxidase. This is likely due to
the ability of LPS to enhance the formation of glucose-6-phosphate
dehydrogenase, which is an enzyme that limits the rate of NADPH

regeneration needed for oxygen reduction (Decker, 1997).

1.1.3.3 Neutrophils

Neutrophils originate in the bone marrow from promyelocytic stem
cells. The majority of their maturity occurs in the bone marrow before
release into the systemic Circulation. Neutrophils play a key role in the
host’s defense against bacterial infection. This is illustrated by the fact
that animals depleted Of their circulating neutrophils have a greater
susceptibility to bacterial infection (Matsumoto et al, 1991).

Accumulation at the site of infection is the first step in the
neutrophil response to a bacterial infection. This involves a complex
series of events that can be subdivided into chemotaxis, adherence, and
diapedesis. Chemotaxis refers to the process by which neutrophils
follow a concentration gradient of soluble Chemotactic factors that
directs them to the site of infection. Adherence is used to describe the

ability of the neutrophil to adhere to the vascular endothelium.

31

Diapedesis is the process by which neutrophils migrate from the
vascular lumen into the interstitium.

Once the neutrophils arrive at the site of infection they can kill the
bacteria by either phagocytosis or the release Of cytotoxic factors into
the surrounding area. Bacteria that are phagocytosed are subsequently
killed and degraded by cytotoxic factors released into the phagocytic
vacuole. The same Cytotoxic factors are also released into the
extracellular milieu. These include reactive oxygen metabolites and
proteases.

The production of reactive oxygen metabolites is dependent upon
the activation of the enzyme NADPH oxidase. This enzyme is assembled
on the plasma membrane upon LPS-stimulation and catalyzes the
transfer of one electron from NADPH to molecular oxygen to produce
superoxide. Though superoxide is itself cytotoxic it is also converted into
more toxic oxidants such as hydrogen peroxide, hydroxyl radical, and
peroxynitrites. With the additional release of myeloperoxidase the major
oxidant hypochlorous acid can be formed. Oxygen metabolites can
damage cells by initiation of lipid peroxidation and alterations in
proteins and DNA.

LPS stimulates the expression of neutrophil adhesion molecules

on the surface of vascular endothelial cells and the production of

Chemokines, which stimulate the migration of neutrophils into the liver.

32

Molecules that act as chemokines include IL-8 and C5a as well as other
complement products.

Neutrophils contribute to LPS-induced liver injury. This is
demonstrated by the fact that depletion of Circulating neutrophils
protects against LPS-induced liver injury (Hewett et al, 1992; Sato et al,
1993). The mechanism by which neutrophils impart LPS-induced
toxicity has been debated. There are many reports that describe an
increased release of reactive oxygen during endotoxemia (Bautista et al,
1990; Spitzer et al, 1993). Administration of antioxidants improve
survival of endotoxemic animals (Rahman et al, 1999; Villa at al, 1995).
Superoxide dismutase (SOD) protects against liver injury from
Corynebacterium parvum and endotoxin treatment (Aurthur et al, 1985).
In addition, phorbol myristate acetate (PMA)-stimulated neutrophils
injure the isolated, perfused liver (Dahm et al, 1991). However, in
cocultures of hepatocytes and neutrophils, superoxide does not
contribute to hepatotoxicity (Guigui et al, 1988; Ganey et al, 1994).
Rather, proteases released from stimulated neutrophils are responsible
for hepatocellular death (Mavier et al, 1988; Harbrecht et al, 1993;
Ganey ef al, 1994). Cathepsin G and elastase are the cytotoxic proteases
released from stimulated neutrophils (H0 at al, 1996). In vivo, reactive
oxygen species may increase the hepatotoxicity of proteases by

inhibiting the activity of antiproteases found in the plasma (Jaeschke et

33

al, 1997). Additionally, reactive oxygen species are known to activate
NF-KB pathways that also may contribute to hepatocellular injury by
enhancing the inflammatory process (Schreck et al, 1991; Canty et al,
1999; Feng et al, 1995). Thus, considering all of the results, one can
conclude that neutrophils may contribute to LPS-induced liver injury
through a protease-dependent mechanism that is enhanced by reactive

oxygen species.

1.1.3.4 Thrombin

Initiation of coagulation can occur either intrinsically by surface-
mediated reactions or extrinsically through a tissue-derived factor
pathway. In hemostasis, first platelets adhere to macromolecules in
subendothelial regions of the injured blood vessel. This aggregate forms
the primary hemostatic plug. The platelets then stimulate local
activation of plasma coagulation factors. At each step of the coagulation
cascade a previously activated enzyme activates the next enzyme by
Cleavage of one or more peptide bonds in the proenzyme. This activity
requires a cofactor, usually calcium, and an organizing surface, usually
platelets. At the end of the cascade prothrombin is converted into its
active form, thrombin, by factor Xa, which Cleaves two bonds on
prothrombin. Thrombin can then convert fibrinogen into the insoluble

fibrous protein fibrin (Figure 1.3).

34

Figure 1.3. The coagulation cascade.

Schematic diagram of the coagulation cascade representing
both the intrinsic and extrinsic pathways. Abbreviations: Ca”,
calcium; PL, phospholipid surface; a, activated factor; XII,
Hageman factor; Xl, plasma thromboplastin antecedent; IX,
plasma thromboplastin component (Christmas factor); VII,

proconvertin; X, Stuart-Prower factor.

35

Intrinsic

Surface contact

(collagen, platelets)
Prekallikrein

I

XII —->XIIa
XI —' XIa

IX"""IXa

Extrinsic

Tissue injury

I

Tissue
factor

VIIa ‘— VII

VIIa
R X /
Ca

IIa

V—’ Va

 

I

I

PL

Xa

Prothrombin —> Thrombin

(11)

(11a)

Fibrinogen —’ Fibrin

36

The proenzymes involved in the coagulation cascade include
factors ll (prothrombin), VII, IX, X, XI, XII, and prekallikrein. Each of
these proenzyme proteases contain a trypsin-like domain containing
about 200 amino acids Of the carboxyl end. In addition, 9 to 12
glutamate residues near the amino-terminal ends of factors ll, VII, IX,
and X are converted to y-carboxyglutamate residues during biosynthesis
in the liver. These y-carboxyglutamate residues are important for
calcium binding and are required for coagulation.

Factors V and VII act as cofactors once they are activated by
thrombin Cleavage into Va and Vlla. Va and Vlla have no enzymatic
activity themselves but will increase the proteolytic efficiency of the
factors Xa and lXa respectively. Tissue factor is also a nonenzymatic
cofactor that greatly increases the proteolytic efficiency of Vlla. It is
found on the surface of cells not normally exposed to blood. In addition,
monocytes and endothelial cells also may express tissue factor when
exposed to a variety of stimuli, including LPS, TNF-a, and lL-l.

LPS-stimulated enhancement of tissue factor gene activity plays
an important role in the initiation of LPS-induced coagulation (Franco et
al, 2000). This is supported by the ability of tissue factor inhibition to
block LPS-activated coagulation (de Jonge et al, 2000; Levi et al, 1994).

In particular, neutrophils release a large amount of tissue factor in

37

response to LPS, and it is their contribution Of tissue factor that seems
to be important (Todoroki et al, 2000).

Thrombin has been found to elicit many effects beyond its
“Classical" role in the coagulation cascade. For instance, human or-
thrombin stimulates migration Of human peripheral blood monocytes in
a dose-dependent manner. Thrombin stimulates endothelial cells to
produce and release many cellular mediators including PAF, prostacyclin
(PGIZ), nitric oxide, adenine nucleotides, plasminogen activator,
plasminogen activator inhibitor, von Willebrand factor, fibronectin,
superoxide, and platelet-derived growth factor (Carney et al, 1992; Bar-
Shavit ef al, 1992; Ritchie et al, 1995; De Meyer et al, 1995; Holland et
al, 1998). Thrombin enhances the release of TNF-a and lL-l from
monocytes (Hoffman et al, 1995) and stimulates platelet activating factor
release from stellate cells (Pinzani et al, 1994). In isolated, perfused
livers thrombin stimulates glycogenolysis (Yamanaka et al, 1992). In
platelets, thrombin stimulates the release Of neutrophil-activating
peptide 2 variants and adherence to endothelial cells (Piccardoni et al,
1996; Carney et al, 1992). In neutrophils, thrombin may enhance
generation of superoxide, adherence to endothelial cells and release of
thromboxane (Chan et al, 1988; Carney et al, 1992; Bizios et al, 1987).
Thrombin also acts as a growth factor for smooth muscle cells (Bar-

Shavit et al, 1992).

38

Some Of the effects of thrombin are dependent on its proteolytic
activity whereas others are dependent on different sites on thrombin
called exosites or subsites. These sites provide recognition centers for
receptors, of which thrombin has several. One Class of receptors that
thrombin has been found to activate is called the protease-activated
receptor (PAR). Our knowledge of this family of G-protein-coupled
receptors is growing. There are currently four known receptors in this
family (PARs 1-4). These receptors are activated by proteolytic cleavage
of the N-terminal sequence, which then acts as a tethered receptor
ligand. It is well known that PAR-1 plays a key role in hemostasis.
Activation of PAR-1 contributes to proinflammatory effects by
stimulating vasodilatation, increased vascular permeability to plasma
proteins, and Chemotaxis. Thrombin, via activation Of PAR-1, stimulates
various inflammatory cells to release mediators such as histamine,
eicosanoids, and cytokines, and to induce leukocyte rolling and
adherence on post-capillary mesenteric venules (reviewed in Vergnolle et
al, 2001). Thrombin induced lL-6 production by endothelial cells is
potentiated by LPS and TNFa, indicating that thrombin and
inflammatory mediators may cooperate to amplify vascular inflammation
(Chi ef al, 2001).

With the many cellular and physiological effects stimulated by

thrombin, it should be no surprise that thrombin plays a role in LPS-

39

induced liver injury. Within three hours of a lethal dose of LPS to rats,
circulating fibrinogen concentrations decrease by more than 90%
(Margaretten et al, 1967; Prager et al, 1979). At the same time fibrin
clots appear in the microcirculation of the liver. Infusion of thrombin
into the portal vein leads to morphological Changes in the liver that are
similar to portal venous infusion of LPS. These changes include fibrin
deposition, neutrophil accumulation, and hepatic injury (Shibayama,
1987; Hewett et al, 1993). The anticoagulants, heparin and warfarin,
and the thrombin specific inhibitor, hirudin, protect against LPS-induced
liver injury (Pernerstorfer et al, 1999; Moulin et al, 1996; Pearson ef al,
1996), underlining the critical role played by thrombin. However,
ancrod, which depletes Circulating fibrinogen and thereby prevents
formation of fibrin Clots, does not protect against LPS-induced liver
injury (Hewett et al, 1995; Moulin et al, 1996). This result suggests that
thrombin contributes to injury by a mechanism independent of Clot
formation. This interpretation was supported by the Observation that
perfusion of livers isolated from LPS-pretreated rats with medium
containing thrombin produced injury whereas no injury was Observed in
the absence Of thrombin (Moulin et al, 1996).

Though thrombin induces neutrophil Chemotaxis and aggregation,
this does not appear to be the mechanism by which thrombin

contributes to LPS-induced liver injury (Bizios et al, 1986). Many

40

investigators have found that thrombin is not critical for LPS-induced
neutrophil accumulation into the liver (Hoffmann et al, 2000; Woodman
et al, 2000; Pearson et al, 1996). However, the role Of thrombin in
neutrophil extravasation has not yet been fully investigated. The ability
of the neutrophil to extravasate may be a critical step for neutrophils to

damage hepatocytes (Jaeschke, 1997).

1.1.3.5 Summary

From the text above it is clear that hepatocyte injury induced by
LPS is critically dependent on the roles of Kupffer cells, neutrophils, and
thrombin. The role of each of these factors is intertwined with that of
other factors. Though the exact mechanism Of LPS-induCed liver injury
has yet to elucidated, it is Clear that endotoxin stimulates the
inflammatory system, and when the inflammatory system is stimulated
to some threshold, cellular and/or soluble mediators cause hepatocytes
to die. The point at which inflammatory cells reach this threshold, lose
their anti-microbial specificity for bacteria and begin to injure host cells
is not known. Equally unknown is the mechanism that allows an
underlying inflammatory state to enhance the toxicity of a xenobiotic

agent

41

1.2 Endotoxin Enhancement of Xenobiotic Toxicity

LPS administration can enhance the toxicity of many compounds
including ozone, aflatoxin B1, ethanol, benzyl alcohol, verotoxin 2,
halothane, carbon tetrachloride, and galactosamine (Peavy et al, 1987;
Barton ef al, 2000; Thurman, 1998; Cebula et al, 1984; Sugatani et al,
2000; Lind et al, 1984; Formal ef al, 1960; Galanos et al, 1979). The
biological relevance of this enhancement was demonstrated through the
reduction of xenobiotic toxicity by eliminating LPS from animals.
Animals pretreated with LPS binding compounds such as polymyxin B,
lactulose, or LPS-neutralizing antibody were protected against the
toxicity ofcarbon tetrachloride, halothane, and galactosamine (Nolan et
al, 1978; Lind et al, 1984; Czaja et al, 1994). D-galactosamine-induced
acute liver injury was attenuated by removal of normal gut flora (Camara
et al, 1983). Induction of endotoxin tolerance by pretreatment with LPS
protected against the toxicity of ozone, carbon tetrachloride, and
acetaminophen (Peavy et al, 1987; Liu et al, 2000). A dose equivalent to
1% Of the lethal dose of LPS negated the protective effects of COIectomy
on murine hepatitis induced by frog virus 3 (Kim et al, 1982).

The culmination of these data and others not described, Clearly is
that LPS is capable of altering the toxic threshold Of a wide variety of
chemicals. Synthesizing the literature it becomes evident that the

timing of exposure to LPS relative to exposure to the chemical is

42

important in determining the nature of the response. In general,
pretreatment with LPS 16 or more hours before a toxic insult decreases
an animal’s toxic threshold. The mechanism for this is thought to be
dependent on either decreased cytochrome P450 levels (Liu et al, 2000)
or an increased amount Of antioxidants (Peavy et al, 1987). Cotreatment
with LPS at or near the time of administration of the toxicant generally
results in a decreased toxic threshold; this may be due to injury from the
inflammatory response such as reactive oxygen species and initiation of
coagulation. This thought is based on the data indicating that inhibition
of cyclooxygenase-Z, Kupffer cells, neutrophils, and the coagulation
cascade can attenuated LPS enhancement of xenobiotic hepatotoxicity
(Ganey et al, 2001; Sneed et al, 1997; Barton et al, 2000). Therefore,
whereas underlying inflammation induced by LPS exposure may enhance
human health risk of xenobiotics, the timing of these two episodes is
critical.

The potential for LPS to decrease the toxic threshold Of
xenobiotics found in the environment or work setting is a concern for
human health. People are often exposed to small doses Of LPS from
many sources. Perhaps the most common source of LPS exposure is
translocation through the gastrointestinal tract. This normally occurs in
small quantities but is greatly increased during stresses such as

abdominal surgery, drinking alcohol, and even running marathons

43

(Palmer etal. 1980; Deitch, 1994; Bode et al, 1987; Camus efaI, 1997).
Though large doses of LPS cause toxic injury, over the past several years
small, nontoxic doses of LPS have been shown to potentiate the toxicity
of many xenobiotics as described above. By seeking to understand the
mechanism by which LPS potentiates the toxicity of xenobiotics, we may
be able to predict which xenobiotic-induced toxicities LPS will enhance
and perhaps inhibit the possible potentiation of drugs administered to
patients susceptible to endotoxemia or underlying inflammation. To
help elucidate this mechanism we Chose to examine the mechanism Of
allyl alcohol-induced liver injury enhancement by LPS. Allyl alcohol is a
well-studied model periportal hepatotoxicant with a great deal of
literature describing the mechanism of its toxicity. A review of this

literature is presented here.

1.3 Allyl Alcohol

Allyl alcOhol (CH2=CH-CH20H) is commonly used as an
intermediate in the synthesis of many industrial compounds. It is also a
constituent of some perfumes, food colorings, and pharmaceuticals. In
addition, allyl alcohol can be formed from the hydrolysis of the allyl ester
Of weak acids (Kaye, 1973). As of 1977 there were no specific
regulations for materials used in food flavorings for the United Kingdom.

The Council of Europe classified allyl alcohol as “an artificial flavoring

substance that may be added temporarily to foodstuffs without hazard to
public health", and it recommended it be further studied in acute and
short-term experiments in animals of various species. It is a
lachrymator causing eye, skin, and upper respiratory irritation. Humans
can readily detect allyl alcohol by smell at 0.78 ppm, and it causes nose
and eye irritation intolerably at 25 ppm. The symptoms usually go away
after about 24-48 hours (Dunlap et al, 1958). “Typical signs of acute allyl
alcohol toxicity in animals, regardless Of the route of administration were
pulmonary effusion, congestion of internal organs, marked hypotension,
Cloudy swelling of the kidney cells, and necrosis of the periportal areas
of the liver" (Carpanini et al, 1978). The toxicity Of allyl alcohol has been
thoroughly studied to elucidate the mechanism.

TO gain understanding of the mechanism for allyl alcohol toxicity
an understanding of its metabolic pathways are necessary. The rate of
metabolism for allyl alcohol was calculated to be about 23 mg/kg/h
during constant i.v. infusion (Kodama et al, 1958). When rats were given
an i.p. injection of 1"C-Iabeled allyl alcohol most of the label was found in
the periportal region Of the liver; however, when an inhibitor of alcohol
dehydrogenase was added the amount of 1‘IC-labeled allyl alcohol in the
liver was reduced by 80 percent. This large difference was not found in
any other organs (Reid, 1972). This indicated that alcohol

dehydrogenase is involved in the metabolism of allyl alcohol and causes

45

allyl alcohol to accumulate in the liver. This enzyme produces the major
oxidative metabolite, acrolein, via the cofactor NAD+ (Patel ef al, 1980,
Serafini-Cessi, 1972). It is generally accepted that acrolein is the
metabolite that is responsible for cellular damage. This was
demonstrated by showing that inhibition of alcohol dehydrogenase
activity blocks hepatic injury (Patel et al, 1980, Serafini-Cessi, 1972) and
that addition of acrolein leads to cytotoxicity (Silva ef al, 1989).

Acrolein is a very reactive aldehyde that is found in cigarette
smoke and automobile exhaust and is a by-product of many industrial
processes including burning Oils. In fact, the word “acrolein" derives
from the acrid smell of overheated frying oil (Rikans, 1987). It is also a
metabolite of the common anticancer and immunosuppressant drug,
cyCIophosphamide. Acrolein has four main metabolic fates within the
cell. First, about 15% of acrolein is converted to the non-reactive acrylic
acid via the enzyme aldehyde dehydrogenase in the presence of NAD+
(Patel et al, 1980). Second, glutathione has a high affinity for binding
the reactive acrolein molecules forming S-3-hydroxypropylglutathione,
which is further broken down to 3-hydroxypropylmercapturic acid.
Third, acrolein can be converted into an epoxide, glycidaldehyde, via
liver microsomes in the presence of NADPH. Glycidaldehyde is then
hydrolyzed to glyceraldehyde by epoxide hydrolase. Both glycidaldehyde

and glycidol are substrates for glutathione-s-epoxide transferase (Patel

46

et al, 1980). Finally, acrolein will bind to protein sulfhydryl groups.
These metabolic fates of acrolein act in concert to inhibit protein
function through thiol binding, reduce cellular glutathione levels, and
provoke the production of free radicals, which in culmination can lead to
cellular injury.

Allyl alcohol is considered a classical, periportal hepatotoxicant.
Necrotic doses of allyl alcohol cause a periportal-specific lesion which
can then spread through the liver acinus toward the pericentral region as
the dose of allyl alcohol is increased. The mechanism for the periportal
specificity has yet to be determined. Originally it was thought that
periportal hepatocytes had a higher alcohol dehydrogenase activity
(Serafini-Cessi, 1972). This has since been shown to not be the case
through several experiments. Penttila isolated pericentral and periportal
hepatocytes and found not only that alcohol dehydrogenase activities
were similar in these two populations, but also that periportal
hepatocytes actually tended to be more resistant to allyl alcohol toxicity.
This might have been due to their 26% higher GSH content after
preincubation with L-methionine, because glutathione has been shown to
be protective against allyl alcohol hepatotoxicity (Penttila, 1988). Liver
sinusoids were also microdissected, and both alcohol dehydrogenase
and aldehyde dehydrogenase activities were similar in periportal and

pericentral regions (Sasse et al, 1991). In the isolated, perfused liver the

47

maximal increase in NADH fluorescence due to allyl alcohol metabolism
was greater in the pericentral region (Belinsky et al, 1984).
Cumulatively, these data clearly prove that allyl alcohol is not periportal-
specific because of a higher alcohol dehydrogenase activity in this
region.

The concentrations of allyl alcohol in the V. portae and V. cava
were equal 30 minutes after dosing with allyl alcohol (Belinsky et al,
1984), which would suggest that both areas of the liver received the
same amount Of allyl alcohol. On repetition of this experiment, a
different group of investigators found that the concentration of allyl
alcohol at one minute after administration was 2800 pM in the V. portae
and only 120 [M in the V. hepatica. By five minutes the V. portae
concentration had decreased to 236 LIM (Sasse et al, 1991). In addition,
it was found that perfusing allyl alcohol in the retrograde direction
caused pericentral-specific damage (Badr et al, 1986). From these
experiments it is Clear that there must be some concentration gradient
within the liver causing the periportal hepatocytes to receive a larger
dose of allyl alcohol.

Further experiments in the isolated, perfused liver demonstrated
the importance of oxygen to allyl alcohol-induced hepatic injury.
Because oxygen uptake was decreased only in the damaged region and

the amount of injury was dependent on the oxygen content of the

48

perfused medium (Badr et al, 1986), it was suggested that oxygen
concentration may play an important role in the zone-specificity of allyl
alcohol. The concentration of oxygen did not affect the toxicity of allyl
alcohol in isolated hepatocytes indicating that some other cell type is
involved in the oxygen-dependence. The role of Kupffer cells, which are
predominately found in the periportal region, in the oxygen-dependence
of allyl alcohol injury was investigated by blocking Kupffer cell activity
with gadolinium Chloride (GdCI3). Rats treated with GdCI3 developed
less hepatic injury from allyl alcohol suggesting that Kupffer cells may be
involved. Interestingly, blocking Kupffer cell activity with GdCI3 in the
isolated, medium-perfused liver had no effect on the hepatotoxicity of
allyl alcohol, suggesting that Circulating blood elements may also
contribute to the injury (Przybocki et al, 1992). This experiment was
repeated by our laboratory and resulted in contradictory findings.
Neither inhibition Of Kupffer cells with GdCl3 nor depletion of neutrophils
with anti-neutrophil antibodies was protective against allyl alcohol
hepatotoxicity (Ganey et al, 1995). Therefore, the importance of Kupffer
cells in allyl alcohol hepatotoxicity may be dependent on other factors
and is likely not the cause of periportal specificity.

It was hypothesized that glutathione concentrationscould explain
the. periportal specificity of allyl alcohol hepatotoxicity. Glutathione was

depleted throughout the liver within 20 minutes after allyl alcohol

49

injection, but liver injury was not markedly increased until 40 minutes
and was observed only in the periportal region. Because glutathione
levels were depleted in periportal and pericentral areas, glutathione
depletion alone could not explain the specificity of allyl alcohol for the .
periportal region. Interestingly, after 60 minutes of allyl alcohol
perfusion of isolated livers, the ATP/ADP ratio was only decreased in the
periportal region, indicating the possibility that periportal cells had less
energy to protect and repair themselves. However, it was argued that
because decreasing the ATP/ADP ratio with potassium cyanide did not
induce liver injury, the decrease in energy status did not explain the
zone-specificity of allyl alcohol (Badr et al, 1986). Thus, the reason for
the periportal localization remains unknown.

Equally mysterious is the actual cause of cell death from allyl
alcohol toxicity. Some groups think that hepatocytes die from the direct
effects of acrolein binding to and disrupting critical proteins. Others
argue that acrolein indirectly causes toxicity by depleting glutathione
stores, leaving the cell unprotected from oxidative injury. One of the
cell’s mechanisms for protection from protein thiol binding and oxidative
injury is glutathione. Allyl alcohol caused a rapid decrease in reduced
glutathione, without the formation of oxidized glutathione, ' and a
decrease in protein thiols (Dogterom et al, 1988; Jaeschke et al, 1986;

Silva et al, 1989; Comporti et al, 1991). Depletion of glutathione by 65%

50

with ethacrynic acid increased acrolein-induced hepatocyte toxicity (Silva
et al, 1989). Hepatocytes incubated with acrolein and washed were
protected from injury by the addition of dithiothreitol, glutathione, or N-
acetylcysteine indicating that initial events in this cytotoxicity are
reversible (Silva et al, 1989; Rikans et al, 1994).

The loss of protein thiol groups can lead to the inactivation of
proteins necessary for maintaining cellular energy levels (i.e. ATP). This
decrease in energy levels was thought to be a cause Of cell death from
allyl alcohol. In isolated, rat hepatocytes allyl alcohol (50 pM) caused a
15 percent decrease in ATP. Because a similar decrease in ATP
concentration using sodium azide did not induce the same injury, it was
concluded that alteration of energy homeostasis alone did not explain
cell death from allyl alcohol (Rikans et al, 1995).

Another hypothesis is that lipid peroxidation is more important to
cell death from allyl alcohol than protein lOss. Vitamin E, an antioxidant,
protected hepatocytes from lipid peroxidation and death from allyl
alcohol (Dogterom et al, 1989, Comporti et al, 1991). The importance of
lipid peroxidation was called into question with the observation that
disulfiram, an inhibitor of aldehyde dehydrogenase, and its metabolite,
diethyldithiocarbamate, protected from lipid peroxidation but not cell
death in hepatocytes (Dogterom et al, 1988). This interpretation may

have been flawed, however, because disulfiram alone increased cell

51

death. On the other hand, the prevention of lipid peroxidation with
desferrioxamine, an iron Chelator, was accompanied by decreased cell
death and protein thiol loss of membranes with no decrease in thiol loss
invthe cytosol (Pompella et al, 1991; Dogterom et al, 1988; Jaeschke et
al, 1987; Miccadei et al, 1988). This indicates the possibility that
protein thiol loss in the membrane is in part from lipid-derived radicals.
Because of the protective effect of desferrioxamine, it was thought
that the lipid peroxidation occurs through an iron-dependent
mechanism. In support of this, ferrous sulfate increased allyl alcohol-
induced damage. In addition, NADH and Fe3+ or NADH and ferritin
induced lipid peroxidation in hepatocytes (Jaeschke et al, 1992).
Furthermore, lipid peroxidation was Observed only when glutathione
stores were depleted by allyl alcohol to 15-30% Of normal levels (Silva et
al, 1989). This is less glutathione depletion than is thought necessary to
allow endogenous lipid peroxidation to occur. There are other proposed
sources of free radicals for the initiation of lipid peroxidation. Acrolein
and its glutathione adduct, glutathionylpropionaldehyde, induce oxygen
radical formation: the glutathionylpropionaldehyde metabolite of acrolein
interacts with xanthine oxidase and aldehyde dehydrogenase to produce
superoxide and hydroxyl radicals, and acrolein forms acroleinyl radical

and superoxide (Adams et al, 1993).

52

The evidence presented suggests that allyl alcohol depletes
glutathione stores which leaves the cell more susceptible to lipid
peroxidation. Lipid peroxidation may be initiated through several sources
including Kupffer cell- or neutrophil-derived reactive oxygen
intermediates, released iron, xanthine oxidase, or aldehyde
dehydrogenase activity. If these events that occur with larger, toxic
doses of allyl alcohol also occur with threshold doses, then the reactive
oxygen intermediates released from LPS-stimulated inflammatory cells

may significantly increase injury.

1.4 The Potentiation of Allyl Alcohol-Induced Liver Injury

To understand better the mechanism by which LPS enhances the
toxicity of many xenobiotics our lab has focused on the ability of LPS to
potentiate the toxicity of allyl alcohol. Rats were used to produce a dose-
response curve for LPS and allyl alcohol. Doses Of each compound were
found that did not cause toxicity alone. The rats were given a
nonhepatotoxic dose of LPS two hours before a nonhepatotoxic dose of
allyl alcohol. This combination resulted in a significant hepatotoxicity
(Sneed et al, 1997). Furthermore, treatment of LPS-treated rats with
gadolinium chloride abolished the LPS enhancement of allyl alcohol
hepatotoxicity (Sneed et al, 1997). These studies demonstrate the

potentiation of allyl alcohol-induced liver injury by LPS and the

53

importance of the activation state of the inflammatory system in this
potentiation.

The LPS potentiation Of allyl alcohol-induced liver injury was
studied further to determine its mechanism. The conclusion that LPS
potentiates the toxicity of allyl alcohol is based on histology. Injury from
large doses of LPS is focused around the midzonal region, and the
potentiated injury is focused around the periportal region similar to what
occurs after a large dose of allyl alcohol (Sneed et al, 1997). For the
potentiation, maximal liver injury occurs six hours after allyl alcohol
administration as measured by the release of alanine aminotransferase
(ALT), indicating that the injury is rapid in onset. This injury is
dependent on the metabolism of allyl alcohol into the reactive metabolite
acrolein through the enzyme alcohol dehydrogenase. An inhibitor of
alcohol dehydrogenase, 4-methylpyrazole, blocked the potentiation Of
allyl alcohol hepatotoxicity by LPS (Sneed et al, 1997). In addition, LPS
administration does not appear to affect the rate of allyl alcohol
metabolism or the amount of glutathione found in the liver (Sneed et al,
1997). Thus, the basic mechanism of allyl alcohol toxicity appears to be
unaltered by LPS, and LPS does not make the liver more susceptible to
allyl alcohol by depleting protective glutathione.

The importance of other cell types was demonstrated through

several experiments. In isolated hepatocytes, LPS did not shift the dose-

54

response curve for allyl alcohol, indicating that the potentiation observed
in vivo could not be replicated with hepatocytes alone. TNF-a is involved
in many mechanisms of injury; however, it is not thought to be involved
in this potentiation due to the lack of protection from anti-TNF
antibodies (Sneed etal, 2000). It remains possible that TNF-a acts in an
autocrine fashion and that inhibition with anti-TNF-a antibodies could
not scavenge the TNF-a before acting on a TNF-a receptor. However,
when TNF was added to isolated hepatocyte cultures no alteration in allyl
alcohol toxicity was observed, indicating that TNF is not involved in the
potentiation through direct stimulation of hepatocytes (Sneed et al,
2000).

The ability of gadolinium chloride to attenuate the potentiation
response demonstrates the importance of Kupffer cells to this injury
(Sneed et al, 1997). Because the ability of gadolinium Chloride to inhibit
Kupffer cells specifically is the main evidence that indicates the
importance of Kupffer cells in this model of potentiation, it is important
that we have little question about the mechanism by which gadolinium
chloride attenuates the potentiation of allyl alcohol toxicity by endotoxin.
Gadolinium Chloride is a rare earth metal that selectively inhibits Kupffer
cell activity. The mechanism by which gadolinium Chloride inhibits
Kupffer cell activity is not completely understood. Gadolinium Chloride

is soluble in an acidic solution. When this acidic solution is injected into

55

the systemic Circulation, the pH increases and the gadolinium Chloride
forms particulate clumps in the blood that are then phagocytosed by
macrophages. Once inside the acidic lysosomal compartment the
particulates may release free gadolinium into the macrophage that may
then be toxic to the cell (reviewed in Adachi et al, 1994). The specificity
of gadolinium chloride is likely due to the enormous phagocytic capacity
of the Kupffer cell. Kupffer cells represent about 85 to 95 percent of the
total intravascular phagocytic capacity (Ruttinger et al, 1996).
Gadolinium chloride does have other effects including inhibition of
stretch-activated ion Channels and physiological responses of tissues to
mechanical stimulation (Adding et al, 2001). However, in the liver the
predominant effect of gadolinium chloride is to decrease phagocytic
activity of Kupffer cells. Though gadolinium Chloride treatment alone
can cause a decrease in P450 activity and slight increases in serum
markers of liver injury (ALT and AST), cytokine release (TNF-or and lL-6),
bile flow, and the number of nonperfused sinusoids, there was no change
in observable liver injury, mean arterial blood pressure, or amount of
circulating endotoxin in gadolinium Chloride treated animals (Ruttinger
et al, 1996; Badger et al, 1997). At smaller doses, gadolinium Chloride
can cause Kupffer cell death and at larger doses Kupffer cells are absent
from the liver, presumably because they have died and been washed out

by circulating blood (Adachi et al, 1994). Gadolinium Chloride

56

pretreatment Of LPS-treated animals increased survival and decreased
liver injury as well as cytokine release (Vollmar et al, 1996; Sarphie et al,
1996; limuro et al, 1994; Fujita et al, 1995).

Most investigators demonstrate inhibition Of Kupffer cell activity
by gadolinium chloride treatment by measuring either phagocytic
capacity of the Kupffer cells or by counting the number of Kupffer cells
in the liver. These are usually done in a separate experiment and do not
demonstrate that Kupffer cells were inhibited during the experiment
from which they drew conclusions. Furthermore, whereas they
demonstrate the inhibition Of Kupffer cells in a separate experiment they
do not monitor many, if any, other effects that gadolinium chloride could
have on the liver. Thus conclusions drawn from experiments using
gadolinium Chloride are based on the drug's main effect, inhibition of
Kupffer cell activity. While there are other inhibitors of Kupffer cell
' activity, each has the same basic problems of secondary drug effects.
To prove that Kupffer cells play a critical role would require further
examination, such as in vitro experiments.

Experiments demonstrating the ability Of gadolinium Chloride to
attenuate LPS enhancement of allyl alcohol toxicity did not demonstrate
inhibition Of Kupffer cells (Sneed ef al, 1997). In these experiments the
dose of gadolinium Chloride used was able to inhibit the activity of

Kupffer cells as reported by others (Adachi ef al, 1994; Ganey et al,

57

1995; Brown et al, 1997). Furthermore, gadolinium Chloride does not
inhibit the activity of alcohol dehydrogenase (Przybocki 'et al, 1992),
indicating that attenuation of LPS potentiation of allyl alcohol toxicity
with gadolinium chloride is not through decreased metabolism of allyl
alcohol into the reactive metabolite, acrolein. The inability of gadolinium
chloride to inhibit the toxicity of allyl alcohol given alone indicates that
gadolinium Chloride does not have any effects on the toxicity of allyl
alcohol. Thus, the inhibition of LPS-enhanced allyl alcohol hepatotoxicity
by gadolinium chloride is likely due to its ability to inhibit Kupffer cell
activity or some other LPS-mediated portion Of the potentiation. To
further examine the role Of Kupffer cells in the potentiation of allyl
alcohol toxicity by LPS, hepatocyte-Kupffer cell cocultures were used in
experiments described below.

Neutrophils were also found to contribute to the potentiation of
allyl alcohol toxicity by LPS. Depletion Of neutrophils with an anti-
neutrophil antibody afforded protection from liver injury (Sneed, 2000).
The specificity of the anti-neutrophil antibody for depleting circulating
neutrophils was determined by counting white blood cells. Whereas the
number of neutrophils and monocytes was significantly decreased the
number of lymphocytes was not significantly decreased. While the
possibility remains that monocytes may play a critical role in the

potentiation of allyl alcohol hepatotoxicity by LPS, injury develops very

58

rapidly in this model Of toxicity, and neutrophils are known to elicit their
actions before monocytic infiltration by several hours. Furthermore,
neutrophils play a critical role in LPS-induced liver injury (Hewett et al,
1992). The possible contributions that neutrophils provide to the
potentiation of allyl alcohol by LPS needs to be investigated.

The potentiation of allyl alcohol hepatotoxicity by endotoxin
appears to involve a complicated mechanism that includes the activities
of several cell types in the liver. The focus of this thesis is to elucidate
further the mechanism by whichendotoxin enhances the hepatotoxicity
of allyl alcohol. This includes investigating the roles Of Kupffer cells,

neutrophils, and other factors involved in inflammation.

59

CHAPTER 2

Involvement of Kupffer Cells in the Potentiation of

Allyl Alcohol Hepatotoxicity by LPS

60

2.1 Introduction

The mechanism underlying the enhancement of allyl alcohol-
induced injury by LPS is unknown. Although LPS can alter the hepatic
metabolism of many compounds through changes in cytochrome P450
levels (reviewed in Morgan, 2001), LPS did not affect the rate Of allyl
alcohol metabolism (Sneed RA et al, 1997). There is evidence to
suggest that the LPS enhancement of allyl alcohol toxicity involves
components of the inflammatory system. For example, hepatocytes
(HCs) exposed to LPS were not more sensitive to allyl alcohol than naive
HCs, indicating that LPS does not increase toxicity by directly altering
HCs (Sneed et al, 1997). In addition, inhibition of activity of Kupffer cells
(KCs), the resident macrophage of the liver, significantly decreased the
potentiation of allyl alcohol hepatotoxicity by LPS, indicating that KCs
play a critical role in the potentiation (Sneed et al, 1997). Inhibition of
Kupffer cells can attenuate the toxicity of many xenobiotics including:
carbon tetrachloride (Pereira et al, 1997; Badger et al, 1996; Wueweera
et al, 1996), vinylidene chloride (Wueweera et al, 1996), acetaminophen
(Michael et al, 1999), D-galactosamine (Stachlewitz et al, 1999), 1,2-
dichlorobenzene (Hoglen et al, 1998), LPS (Sarphie ef al, 1996; Fujita et
al, 1995; limuro et al, 1994), ethanol (Bautista et al, 1999; Adachi et al,

1994), and diethyldithiocarbamate (lshiyama et al, 1995). Because

61

Kupffer cells are thought to play a critical role in the development of liver
injury of many compounds we sought to understand better the
involvement of Kupffer cells in the potentiation of allyl aICOhOI
hepatotoxicity by LPS.

The ability of LPS to stimulate KCs to release many mediators of
inflammation that can act directly or indirectly on HCs to alter their
physiology led to the hypothesis that LPS-stimulated KCs sensitize HCs
to allyl alcohol-induced toxicity through the release of inflammatory
mediators. Results presented here to test this hypothesis demonstrated
that the model is more complex than a simple interaction among KCs,

LPS, and H05.

2.2 Methods

2.2.1 Materials

Williams' Medium E, gentamicin, RPMI, and NCTC-109 medium
were purchased from GibcoBRL, ROCkville, MD. Fetal Calf Serum was
purchased from lntergen, Purchase, NY. TNF-or ELISA kits were
purchased from Biosource, Camarillo, CA. ProbeOn Plus microscope
slides were purchased from Fisher Scientific, Pittsburgh, PA. All other

materials were purchased from Sigma Chemical CO., St. Louis, MO.

62

2.2.2 Animals

Male, Sprague-Dawley rats (CD-CrlzCD-(SD)BR VAF/Plus; Charles
River, Portage, MI) weighing 175.225 g were allowed food (Wayne Lab-
Blox, Allied Mills, Chicago, IL) and water ad Iibitum. They were housed
under conditions Of controlled temperature, humidity and 12-hr light and
dark cycle. All procedures on animals were carried out according to the
humane guidelines of the AALAS and the University Laboratory Animal

Research Unit at MSU.

2.2.3 Kupffer Cell Isolation

KCs were isolated using a method based on Knook, 1977. Rats
were anesthetized with sodium pentobarbital (50 mg/kg). The portal
vein was cannulated and perfused with Ca2+/Mg2*-free Hanks’ balanced
salt solution to Clear the blood from the liver. The liver was then
perfused with 200 ml of collagenase type II (0.4 mg/ml) in Gey’s
balanced salt solution (GBS). Pronase (0.2% in CBS with 2 pg/ml
DNAse) was then perfused in a recirculating manner for 10-15 minutes.
The liver was removed and gently combed to loosen the cells. The cells
were filtered through gauze and centrifuged (50 x g, 2 min) to remove
HC5, and the supernatant was collected from centrifugation (50 X g, 2
min) of cells twice. The cells found in the supernatants were then pooled

via centrifugation (600 x g, 5 min). These cells were resuspended in

63

GBS containing 2 ug/ml DNAse and loaded onto a centrifugal elutriator
(Beckman J6-Ml centrifuge with JE-GB elutriator rotor). GBS was
pumped through the rotor at a rate of 12 ml/min for 250 ml. Flow rate
was then increased to 24 ml/min for 150 ml to remove endothelial cells.
Flow was increased again to 42 ml/min for 150 ml and collected. This
fraction was spun in a centrifuge (600 x g, 5 min), and the pellet was
resuspended in RPMI culture medium supplemented with 1% Medium
NCTC-109 and 15% bovine calf serum. The KCs were then plated in 12-
well Falcon Primaria culture plates at 1 X 106 cells/well (37°C, 7.5%
C02, 92.5% air). KC purity was found to be greater than 90% via
peroxidase staining and latex bead phagocytosis. KC responsiveness
was assessed by measuring TNF-or levels in the medium 90 min after

stimulation with LPS (105 EU/ml) using a rat TNF-or ELISA kit.

2.2.4 Hepatocyte Isolation

Hepatic parenchymal cells were isolated using methods
established by Seglen, 1973. Briefly, rats were anesthetized with
sodium pentobarbitol (50 mg/kg, i.p.). The liver portal vein was
cannulated and perfused with Ca2+/Mg2*-free Hanks' balanced salt
solution to remove red blood cells and then digested by perfusion with a
solution containing 60 mg collagenase type II in 250 ml Williams’

Medium-E for 30 min. The digested liver was gently combed, filtered

through sterile gauze, and centrifuged (50 X g, 2 min) to pellet the H05.
The cells were washed twice with Williams’ Medium-E and resuspended
in Williams’ Medium-E containing 1% gentamycin and 10% fetal calf
serum. Viability was assessed via trypan blue exclusion. If cells were
greater than 80% viable, they were plated in 12-well Falcon Primaria
culture plates at a density of 2.5 X 105 cells/well. The cells were then
allowed to incubate (37°C, 7.5% C02, 92.5% air) for four hours before

the cells were washed and used in an experiment.

2.2.5 In I/itra Coculture Experiments

Kupffer cellswere isolated and plated as described above. After
20 hours of incubation medium was removed, and hepatocytes isolated
as described above were plated with the Kupffer cells. After an
additional four hours of incubation, cocultures were treated with LPS
and allyl alcohol. Medium was collected 1.5 hours after addition of allyl
alcohol, and cells were lysed with 1 ml Of 1% Triton X. Cytotoxicity was
assessed from release of alanine aminotransferase (ALT) into the
medium. Activity of ALT was determined in cell-free supernatant fluids
using Sigma Diagnostics Kit NO. 52. Percent total ALT was calculated by
dividing ALT activity in medium by the total ALT activity found in
medium and cell lysates. Total ALT activity in the medium and cell

Iysates did not Change with any culture treatment. Results were

65

analyzed using analysis of variance (ANOVA) blocked for each day Of the

experiment.

2.2.6 Slide Coculture Experiment

Hepatocytes and Kupffer cells were isolated as described above
and plated on ProbeOn Plus microscope slides. At the time of the
experiment two slides were Clipped together face to face so that the cells
were within the capillary space between the slides provided by the thick
frosted corners of the ProbeOn Plus microscope slides. The slides were
set vertically in a small trough with a section of filter paper set on top,
which served to oxygenate and distribute the medium evenly over the
length of the slide. Medium was then pumped in a Circulating manner
from the trough to the filter were it flowed between the slides and was
collected in the trough. Cells were treated with LPS and allyl alcohol and

assessed for injury as described for coculture experiments above.

2.2.7 Liver Homogenate Experiment

Hepatocytes were isolated and plated as described above. Non-
parenchymal cells, consisting of the cells remaining in the supernatant
after the liver homogenate was centrifuged (50 X g, 2 min), were plated
in 12-well Falcon Primaria culture plates at 1 X 106 cells/well containing

hepatocytes. After incubation the cells were treated with LPS and allyl

66

alcohol, and cytotoxicity was assessed from release of ALT as described

above.

2.2.8 Isolation and Perfusion of Rat Livers

Rats were weighed, then anesthetized with sodium pentobarbital
(50 mg/kg, i.p.), and the portal vein was exposed. Livers were perfused
via portal vein cannulation with polyethylene tubing (PE 190, Clay
Adams, Parsippany, NJ). The perfusion medium was Krebs-Henseleit
bicarbonate buffer supplemented with 2% bovine serum albumin and
saturated with 95% 02 and 5% 002 gas. Flow was constant at rate of
0.14 ml/min/g body weight. The thoracic portion of the inferior vena
cava was cannulated (PE 240 Clay Adams, Parsippany, NJ) for outflow.
The liver was placed in a temperature-controlled cabinet maintained at
37°C as previously described (Moulin et al, 1996). The livers were
allowed to stabilize for 10 minutes with single pass perfusion. A sample
of the perfusion medium was then taken (time = 0), and the system was
switched to recirculating perfusion with medium containing LPS (1.2 X
105 EU/kg body weight). Samples (500uL) of the perfusion medium
were taken every 30 minutes for two hours. At two hours allyl alcohol
(30, 25, or 20 mg/kg body weight) was added to the perfusion reservoir,
and medium samples were taken every 15 minutes for another two

hours. ALT activity in the samples was determined as described above.

67

Previous experience demonstrated that perfusion of naive livers with this
dose of LPS does not produce injury. Accordingly, livers for which ALT
activity in the perfused medium at two hours was greater than 200 U/L
were considered damaged during the isolation procedure and were

removed from the experiment.

2.2.9 Statistical Analysis

Data are expressed as means :I: S.E.M. For all results presented,
N represents the number of individual experiments. Data with non-
homogeneous variances were square root- or log-transformed prior to
analysis. Data represented as percents were angular-transformed prior
to analysis. Data were analyzed using repeated measures ANOVA
provided in the statistics program NCSS 2000. Comparisons among
groups were performed using Fisher’s Least Squares Difference test.

The criterion for statistical significance was p s 0.05.

68

2.3 Results

KCs are critical to LPS enhancement of AA toxicity in vivo (Sneed
et al, 1997), accordingly initial studies were designed to examine
whether LPS-stimulated KCs or their mediators increased hepatocellular
sensitivity to allyl alcohol. It has been reported that LPS stimulates the
release of TNF-a from KCs (LiChtman et al, 1996). To ensure that
isolated KCs to be used in these studies were functional and responsive
to LPS, TNF-a concentration in medium of LPStreated KCs was
assessed. The concentration of TNF-a in the medium of untreated KCs
was low (Figure 2.1). Treatment with LPS (105 EU/ml) increased TNF-a
release significantly. The presence of allyl alcohol (25, 50, 200 pM)
decreased the amount Of LPS-stimulated TNF-0t release in a
concentration-dependent manner.

ALT release in KC/HC cocultures was lower than release in
cultures of H05 alone by about 5 to 10% ALT release (data not shown).
Exposure of cells to allyl alcohol induced a concentration-dependent
increase in ALT release (Figure 2.2). Treatment of KC/HC cocultures
with LPS (1X105 EU/ml) significantly decreased allyl alcohol induced
ALT release compared to vehicle. A larger concentration Of LPS (4X105
EU/ml), also did not induce a greater sensitivity to allyl alcohol toxicity

in vitro (data not shown).

69

Figure 2.1. LPS stimulates release of TNF-or from K05.

K05 were isolated via centrifugal elutriation and cultured for
24 hours before stimulation with LPS (105 EU/ml). TNF-a
concentration was measured using an ELISA kit. Data are
expressed as means i S.E.M. N=3-7. *, significantly different

from control.

70

 

1 200

     
    
    
      
  
  
 
  
   
    
    
     
    
    
    
   
   
  
  
    

   

 
  
  
   

              

 
  
 
   
 
       
      
      

          
  
 
 

   

   
    
 
     
 
  
     

 

  
 
  
 

     
   
   
   
   
  
    
 
  
  

    
 

  
  
 
    

    
 

     
   
  

  
  
 
 

 
 

 

   
  

  
   
  
        

 

  
 
 
 

    
    
   

  
      
   
  

              
 

   
 
 
    
 
 

      
       
    
   
  

  
     
 
  
       

      
 

  
    
   
   

  
  

 
 

       
    
    
        
        
    
      
   

 

  
 

   

 
 

  
   
 
  
  
   
  
  
  
  

  

  
 

 
   
 
   
   

      
  
 

   
    
   
   
   

      

      

      

    

   

 

- Control
1000 - m LPS "
— .O.Q *
N * *
soo Wot
- ‘ 9999
5') I93}: 9 9
9 9 9 9 ’9'9 9 4
CL ifififiv ’VVVZ r999
\a’ 600. 9994 9999 -9994
9999 -9994 9999
.004 O... ’9094 vvv
5 9999 »9994 9999 i-9994
9994 9999 -9994 9999
| 9999 .,,,. 9999 .9994
9994 9999 -9994 9999
9999 -9994 9999 »9994
lL 400. 9994 9999 ~9994 WOO.
9.9.9.9, ‘9.9.9.4 9.9.9.” ~ 9 9 9 4
_ - 9999
:5 9999 9994 9999 -9994
9994 9999 -9994 9999
h. 9999 -9994 9999 .9994
. - 9999 -9994
200 9994 9999 -9994 9999
9999 -9994 9999 .9994
9994 9999 -9994 9999
9999 9994 9999 .9994
in”! ’9??? v??? 8330:

   

o o 25 50 200
Allyl alcohol (pM)

71

 

In vivo, the ratio of KCs to HCs is about 1:10 (Altin et al, 1988);
however, toxic responses observed in KC/HC cocultures require a higher
ratio of KCs to HCs (Lysz et al, 1990; Kausalya et al, 1993). Cocultures
with a higher ratio of KCszHCs produce less PGEZ upon LPS stimulation
than cocultures with lower ratios (Billiar et al, 1990). Because previous
experiments have shown that prostaglandins can increase the toxicity of
allyl alcohol in hepatocytes (Ganey ef al, 2000), we did not want a
decreased production of prostaglandins. To examine whether the lack of
effect of LPS was related to the high ratio of KCs to HCs, cocultures were
plated using different ratios. Allyl alcohol caused a concentration-
dependent cytotoxicity, and LPS did not increase the sensitivity of H05
to allyl alcohol at any of the KCzHC ratios tested (Figure 2.3).

The lack of LPS effect on allyl alcohol hepatotoxicity could be
dependent on reaching a threshold concentration of mediators or some
labile mediator. To address this possibility a simulated sinusoid
environment was used in a limited number of experiments. In this
experiment hepatocytes and Kupffer cells were cocultured on glass
slides placed a capillary distance apart. Although allyl alcohol caused a
dose-dependent increase in ALT release, LPS treatment did not enhance

the cytotoxicity of allyl alcohol (Figure 2.4).

72

Figure 2.2. LPS stimulation does not enhance allyl alcohol toxicity
in cocultures of KCs and H05.

Cocultures of KCs and HCs (4:1) were stimulated with LPS
(105 EU/ml) and exposed to allyl alcohol at the concentrations
indicated. ALT release was assessed 1.5 hours after allyl alcohol
exposure as described in Methods. Data are expressed as means
i S.E.M. N=4. *, significantly different from control at the

indicated concentration.

73

 

/

 

 

—0— Control

-v— LPS
80 l

 

 

 

 

 

m m m o

:82 § .8823. H2

0
o
1

7'5 200

Allyl Alcohol (pM)

50

74

Figure 2.3. Alterimhe ratio Of KCs to HCs does not elicit an LPS-
induced enhancement of allyl alcohol toxicity.

KCs were cocultured with HCs at a ratio of (A) 4 KCs to 1 HC
(8) 1 KC to 1 H0 or (C) 1 KC to 10 H05 and were exposed to
vehicle (black bars) or LPS (105 EU/ml) (white bars) with
increasing concentrations of allyl alcohol. ALT release was
assessed 1.5 hours after allyl alcohol exposure as described in
Methods. Data are expressed as means i S.E.M. N=4. No
significant differences between LPS-treated and control values

were observed.

75

% ALT Release W

O

80

 

60-

40«

20-

 

80

 

f — Control

E: LPS

4:1

 

 

o..ll:_lCI_i __

 

 

 

 

 

60-
40-

zol

 

04

80

 

 

 

 

60«
404

20-

 

 

 

 

25 50 200
Allyl Alcohol (M)

76

 

Figure 2.4. LPS-induced enhancement of allyl alcohol toxicity is
not Observed in simulated hepatic sinusoids.

KCs were cocultured with HCs on microscope slides placed
a capillary distance apart with medium flowing between the slides
in a Circulating manner. Cells were exposed to vehicle (black bars)
or LPS (105 EU/ml) (white bars) with increasing concentrations of
allyl alcohol. ALT release was assessed 1.5 hours after allyl
alcohol exposure as described in Methods. Data are expressed as
means, N=2. No differences between LPS-treated and control

values were observed.

77

Percent ALT Release

100

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

— Control
80 LPs
60
4O
20 d_. : Ting}
° 25
Allyl alcohol (pM)

78

To determine whether the lack of LPS effect may be due to other
cell types in the liver that are not found in the coculture, a liver
nonparenchymal cell fraction was used in place of Kupffer cells. This
system was also used to determine the effect of altering incubation
times. To examine a time-course of cytotoxicity after exposure to allyl
alcohol, ALT release was assessed 30, 60, and 90 minutes after addition
of allyl alcohol. Allyl alcohol caused a dose dependent increase in
hepatocellular injury. This injury was observed after 60 minutes of allyl
alcohol incubation and increased slightly by 90 minutes. The response
was unaffected by addition of LPS (Figure 2.5). In addition, the effect of
the duration of exposure to LPS was examined by incubating
homogenates with LPS for 0, 0.5, 2, or 8 hours. Though allyl alcohol
caused a dose-dependent increase in hepatocellular injury, LPS did not
cause a significant enhancement of allyl alcohol toxicity during any of the
LPS-incubation times (Figure 2.6).

Though KCs were found to be a critical component in the
potentiation of allyl alcohol toxicity by LPS in vivo, there may be other
critical factors not present in the KC-HC coculture such as cell surface
receptors that are cleaved during isolation or three-dimensional
structural interactions. To assess these possibilities isolated, buffer-
perfused livers from na'l've rats were utilized. Allyl alcohol caused a time-

and dose-dependent increase in ALT release in the perfused livers

79

Figure 2.5. Alterirg the time of incubation with allyl alcohol does
not elicit an LPS-induced enhancement of allyl alcohol toxicity.
Liver homogenates were exposed to vehicle (gray bars) or
LPS (105 EU/ml) (black bars) for 0 hours with increasing
concentrations of allyl alcohol for 30, 60, or 90 minutes. ALT
release was assessed at various times after allyl alcohol exposure
as described in Methods. Data are expressed as means, N=2. No
differences between LPS-treated and control values were

observed.

80

\nl

 

 

 

 

87

.E:

0.0
1

:32 i .8323. .72

20-/

.w
M

O

9
0%
6W
“
om
mm
w
m

0
2&9

A

200

44:0,.

Figure 2.6. Altering the time of incubation with LPS does not elicit
an LPS-induced enhancement of allyl alcohol toxicity

Liver homogenates were exposed to vehicle (gray bars) or
LPS (105 EU/ml) (black bars) for various amounts of time before
treatment with increasing concentrations of allyl alcohol. ALT
release was assessed 1.5 hours after allyl alcohol exposure as
described in Methods. Data are expressed as means, N=2. No
differences between LPS-treated and control values were

observed.

82

 

 

 

 

 

 

00000000
87654321

:32 § 882% 54.

83

(Figure 2.7). LPS alone did not induce ALT release from isolated livers.
LPS also did not significantly affect allyl alcohol-induced ALT release.
These results suggest that an extra-hepatic factor is critical for the

ability of LPS to increase allyl alcohol-induced liver injury.

84

Figure 2.7. Perfusion with LPS does not enhance allyl alcohol
toxicity in the isolated liver.

Medium-perfused, rat livers were treated with LPS (1.2 X
105 EU/kg body weight) (gray lines) or vehicle (black lines) for two
hours and subsequently exposedto allyl alcohol at the indicated
doses (in mg/kg) for an additional two hours. ALT released into
the medium was assessed over time as a marker of liver injury.
Data are expressed as means :SEM. N=6-14. NO significant LPS

effect was observed.

85

 

.3
0|
0
0

ALT (Units/L)
2 §
0 O

 

 

+
-O
+—
“a.

+

... 0...,
_._

Control

LPS

20 AA

20 AA 8. LPS
25 AA

25 AA 8. LPS
30 AA

30 AA & LPS

 

 

 

O

.4.

I

,. ‘ “5*“
.——¢_-_.___——-JQ .

LPS Allyl Alcohol
IA ..

:N9

J
4.
9
.I‘
.
.7 A
v ,
v
If
a; ‘ "“ °

H

,’ 4
fl .‘ ‘.
I1 0

9"“

. II

 

o i 2
Hours

86

 

2.4 Discussion

Previous results indicate that KCs play a critical role in LPS
enhancement of allyl alcohol-induced liver injury (Sneed et al, 1997).
Studies presented here were designed to determine if KCs are sufficient
for the LPS-induced enhancement of hepatocellular injury by allyl
alcohol. Isolated KCs were activated by LPS as indicated by production
of TNF upon exposure to LPS. However, under the conditions used for
HC/KC coculture experiments, LPS-stimulated KCs did not increase the
sensitivity of HCs to allyl alcohol.

Several alterations were made to the KC/HC cocultures in an
attempt to reproduce the LPS-induced enhancement of allyl alcohol
hepatotoxicity observed in vivo. For example, increasing the
concentration of LPS did not increase KC/HC coculture sensitivity to allyl
alcohol. In addition, the ratio of KCs to HCs was altered. The ratio of
KCs to HCs can influence the ability of KCs to produce inflammatory
mediators (Billiar et al, 1990), an effect that may be due to factors
produced by HCs that control KC function. To account for this
possibility we altered the KC/HC ratios and found that LPS did not affect
the cytotoxicity of allyl alcohol to HCs at any of the KC/HC ratios

examined.

87

The hypothesis that Kupffer cell mediators were not reaching a
critical concentration for cellular activation, via either metabolism or
diffusion in the culture plate medium, was tested using a simulated
sinusoid. This system in effect increases cell number and decreases the
diffusion space for metabolites. LPS did not enhance allyl alcohol
cytotoxicity under these conditions. We simultaneously tested the
hypotheses that other cells types in the liver may be involved, that LPS
enhances the rate of allyl alcohol cytotoxicity, and that cells need to be
exposed to LPS for some duration before addition of allyl alcohol using
liver homogenates in place of isolated Kupffer cells. None of these
alterations produced an LPS enhancement of allyl alcohol cytotoxicity.
In summary, the lack in vitro of the LPS enhancement of allyl alcohol
observed in vivo did not appear to be due to the concentration of LPS,
the ratio of KCs to HCs, a decrease in mediator concentration, loss of
other liver cell types, time of incubation with LPS, or time of incubation
with allyl alcohol.

There are many other factors that may explain why KC/HC
cocultures fail to mimic the LPS-induced enhancement of allyl alcohol
toxicity found in vivo. An obvious possibility is that the coculture system
lacks an element critical to the response. These factors may include the
loss of Kupffer cell-derived reactive oxygen species. Kupffer cells are not

stimulated to release reactive oxygen species upon LPS exposure in vitro,

88

although they are in vivo. This may be due to the loss of C014 from
Kupffer cells during isolation with collagenase-pronase perfusion (lkejima
et al, 1999). Additionally, the isolation process may remove critical
receptors from the K05 or HCs, or factors released from the K05 could
be highly labile and depend on the Close association of KCs and HCs that
is found in vivo. To account for these possibilities, the isolated, perfused
liver was used as a model to reproduce the response observed in vivo.
Exposure of livers to LPS did not enhance allyl alcohol-induced liver
injury, suggesting that even in this system a critical factor was missing.
This missing factor could be any of a number of factors including
neutrophils, platelet activating factor, thrombin, and other components
found in the Circulatory system.

In summary, we found that LPS stimulation of Kupffer cells in vitro
does not enhance the toxicity of allyl alcohol. This work demonstrates
that although Kupffer cells play a critical role in the potentiation of allyl
alcohol hepatotoxicity in vivo, they are not sufficient to enhance the
sensitivity of hepatocytes to allyl alcohol. Other critical factors required
for the potentiation of allyl alcohol toxicity by LPS may be present in the

blood.

89

Chapter 3

Role of Reactive Oxygen Species in the Potentiation of

Allyl Alcohol-Induced Liver Injury by LPS

90

3.1 Introduction

The sensitivity of rats to allyl alcohol-induced liver injury is
dramatically increased by exposure to small doses of LPS (Sneed et al,
1997). This toxicity occurs rapidly and is maximal at 6 hours after allyl
alcohol administration. Kupffer cells play a critical role in this
potentiation, demonstrated by the protective effect of inhibiting Kupffer
cell activity with gadolinium Chloride (Sneed et al, 1997). However, the
presence of Kupffer cells is not sufficient to cause LPS enhancement of
allyl alcohol-induced liver injury as demonstrated by experiments in
isolated perfused livers (Chapter 2).

An additional factor that may be required in this model of LPS-
potentiated liver injury is the neutrophil (PMN). The toxicity caused by
large doses of LPS is dependent on the presence of circulating
neutrophils (Hewett ef al, 1992; Sato et al, 1993) and smaller doses of
LPS can cause neutrophil infiltration into the liver (Barton et al, 2000).
Depletion of Circulating neutrophils with anti-neutrophil antibodies
protects against the potentiation of allyl alcohol-induced liver injury by
LPS indicating that neutrophils are required for this model of
potentiation (Sneed et al, 2000). In addition, LPS stimulates production
of reactive oxygen species in vivo (Rahman et al, 1999; Robinson et al,

1997; Liu et al, 1995). Though, in addition to reactive oxygen species,

91

the development of hepatocellular injury may be induced by neutrophil
derived proteases (Ganey et al, 1994), depletion of hepatocellular
antioxidants may allow the development of injury from reactive oxygen
species. Allyl alcohol is well known to cause reductive stress in
hepatocytes through a rapid depletion of hepatocellular reduced
glutathione stores without the formation of oxidized glutathione
(Dogterom ef al, 1988; Jaeschke et al, 1986; Silva et al, 1989; Comporti
et al, 1991).

Since glutathione is used by cells to protect against oxidative
injury, the ability of allyl alcohol to deplete hepatocellular glutathione
stores may cause the hepatocytes to become susceptible to oxidative
injury. The reactive oxygen species provided by neutrophils exposed to
LPS may increase the toxicity of allyl alcohol. Accordingly, we tested the
hypothesis that LPS-induced, neutrophil-derived reactive oxygen species
increase the sensitivity of hepatocytes to the reductive stress of allyl

alcohol.

3.2 Methods

3.2.1 Animals
Male, Sprague-Dawley rats [CD-Crl:CD-(SD)BR VAF/Plus]; Charles
River, Portage, MI) weighing 200-300 g were used in all studies. The

animals were allowed food (Rodent Chow, Teklad, Madison, WI) and

92

water ad Iibitum. They were maintained on a 12-hour light and dark cycle

under conditions of controlled temperature and humidity.

3.2.2 Hepatocyte Isolation

Hepatic parenchymal cells were isolated using methods
established by Seglen, 1973. Briefly, rats were anesthetized with
sodium pentobarbital (50 mg/kg, i.p.). The liver was perfused via the
portal vein with Cagt/Mggt-free Hanks’ balanced salt solution to remove
red blood cells and then digested by perfusion with a collagenase type I I
(Sigma Chemical Co., St. Louis, MO) solution (240 pg/ml collagenase in
Williams’ Medium E) for 30 min. The digested liver was gently combed,
filtered through sterile gauze, and centrifuged (50 X g, 2 min) to pellet
the hepatocytes. The cells were washed twice with Williams' Medium E,
resuspended in Williams' Medium E containing 1% gentamycin (GIBCO)
and 10% fetal calf serum (lntergen, Purchase, NY). Viability was
assessed via trypan blue exclusion. If cells were greater than 80%
viable, they were plated in 12-well Falcon Primaria culture plates at a
density of 2.5 X 105 cells/well. After a 4-hr incubation period (37°C,

7.5% C02, 92.5% air) the cells were washed and used in experiments.

93

3.2.3 Neutrophil Isolation

Glycogen-elicited neutrophils were isolated from the peritoneum of
male, SpragueDawley (Charles River Laboratories, Portage, MI), retired
breeder rats as described previously (Hewett et al. 1988). Briefly, rats
were anesthetized with diethyl ether and given 1% glycogen (30 ml i.p.).
After four hours, rats were anesthetized again, decapitated, and 0.01M.
PBS containing heparin (1 unit/ml) (30 ml i.p.) was injected into the
peritoneum. The fluids of the peritoneal cavity were collected, filtered
through gauze, and centrifuged at 500 X g for 7 min. Contaminating red
blood cells were Iysed with 15 ml of 0.15 M NH4CI for two minutes. The
remaining neutrophils were washed one time with 0.1 M PBS and again
with Williams’ Medium E. The percentage of neutrophils and cell
viability were routinely greater than 95%. Cells were diluted to 2.5 X 106

cells/ml for use in experiments.

3.2.4 Detection of GSH Concentration

The concentrations of glutathione (GSH) in cell preparations were
determined using HPLC (Fariss and Reed, 1987; Jean et al, 1995). Cells
were plated (5 X 106 cells/ml) in 6-well plates and treated as described
below in the superoxide anion experiment. At the end of the experiment
cells were lysed with 0.5 mL of 10% perchloric acid (v/v in milIi-Q water)

containing 1 mM bathophenanthroline-disulfonic acid and 0.5 mM

94

glutamyl glutamate (internal standard). The plates were scraped, and
the contents of duplicate wells were combined. The Iysates were
centrifuged at 15,000 X g for 1 min, and 0.5 mL of the supernatant was
collected. The lysate supernatants were derivatized by adding 50 LIL of
100 mM iodoacetic acid containing 0.2 M m-cresol purple. This was
then brought to pH 8-9 by the addition of 0.48 ml of KOH (2 M)-KHCO3
(2.4 M) and allowed to incubate for 10 min. in the dark. Next 100 LIL of
5% fluorodinitrobenzene (v/v in ethanol) was added and mixed with a
vortex mixer. This was set in the dark in the refrigerator overnight.

The HPLC system consisted of two Waters HPLC 510 pumps, a
Waters 717 WISP autosampler, a Waters 486 variable wavelength
detector, and a 3-aminopropyl column (5p Custom LC, Inc., Houston,
TX). The Millenium 2010 Chromatography Manager software (Millipore
Corp., Milford, MA) was used to control the HPLC and integrate peaks.
Samples were diluted 50:50 with methanol, and 100 pL of sample was
injected onto the column. Samples were eluted using a linear gradient of
80% methanol and 0.5 M sodium acetate in 64% methanol. The amount

of GSH was calculated from the area under the UV absorbance curves.

95

3.2.5 Detection of Superoxide Anion

Superoxide anion generation by phorbol myristate acetate (PMA)-
stimulated neutrophils was detected by the reduction of cytochrome c in
the presence and absence of superoxide dismutase (SOD). Neutrophils
in 96 well culture plates (5 X 105 cells/well in calcium- and magnesium-
containing HBSS), containing cytochrome C, allyl alcohol, and :tSOD
were stimulated with PMA (20 ng/ml). PMA is an effective stimulator of
superoxide production in rat neutrophils. The absorbance of each well
was measured at 550 nm every two minutes for up to 90 minutes.
Superoxide anion production was calculated using an extinction
coefficient of 18.5 cm'lmM'1 and the difference in absorbance of wells
with and without SOD. Rate of superoxide anion production was
estimated by Change in superoxide anion production every two minutes.

Area under the curve was calculated using the trapezoidal rule.

3.2.6 Effect of PMN-Derived Superoxide Anion on Hepatocellular
Viability In Vitra

After 4 hours of incubation, hepatocyte cultures were washed with
Williams' Medium E, and Williams’ Medium E with or without freshly
isolated neutrophils was added. The neutrophils were allowed to settle
for 20 min, and the cultures were treated with allyl alcohol (0, 25, 50,

75, 100, 125, 150 and 200 pM). PMA (20 ng/ml) was added to the

96

cultures 20 minutes after the addition of allyl alcohol. Following 90
minutes of allyl alcohol treatment cell viability was assessed via ALT
release from hepatocytes. Medium was collected, and cells were lysed
with 1 ml of 1% Triton X100. Percent total ALT released was then
calculated by dividing ALT activity released into medium by the total ALT
activity found in medium and cell lysates. Results were analyzed using

ANOVA blocked for each day of the experiment.

3.2.7 Effect of PMN-Derived Proteases on Hepatocellular Viability
Freshly isolated neutrophils in Williams’ Medium E were exposed
to cytochalasin B (10.4 nM), a primer for neutrophil degranulation, for 5
min before addition of formyl-methionyl-leucyl-phenylalanine (fMLP)
(100 nM). The cell suspension was placed in a shaking water bath at
37°C for 30 min. The cells were then centrifuged at 500 X g for 10 min
and the resulting conditioned medium was added to hepatocyte cultures
that had been washed after 4 hours of incubation. Medium containing
cytochalasin B and fMLP was used as a control. The cells were then
treated with allyl alcohol (0, 25, 50, 75, 100, 125, 150 and 200 nM).
Following 90 minutes of allyl alcohol treatment cell viability was
assessed via alanine aminotransferase (ALT) release from hepatocytes as

described above.

97

3.2.8 Treatment In Viva with Antioxidants

Rats were given LPS (1.2 X 105 EU/kg body weight, i.v.)
(Escherichia coli, serotype 0127:88) or an equivalent volume of saline
vehicle (Abbott Labs, Abbott Park, IL) one hour before treatment with
polyethylene glycol-coupled superoxide dismutase (10,000 IU/kg i.v.),
polyethylene glycol-coupled catalase (40,000 lU/kg i.v.) or an equivalent
volume of saline. One hour later allyl alcohol (30 mg/kg ip) was
administered. Six hours after treatment with allyl alcohol or saline
control, rats were anesthetized with sodium pentobarbital (50 mg/kg ip),
and blood was collected from the abdominal aorta into syringes
containing 3.8% sodium Citrate (final concentration 0.38%). The activity
of ALT was determined in plasma. SOD activity was confirmed in plasma
samples taken one hour after allyl alcohol administration by measuring
superoxide production by neutrophils as described above in the presence
of rat plasma diluted 1:3.

In a separate experiment rats were given apocynin (5 mg/kg, p.o.)
or an equivalent volume of 5% ethanol vehicle 2 hours before treatment
with of LPS (1.2 X 105 EU/kg body weight, i.v.) (Escherichia coli, serotype
0127:88) or an equivalent volume of saline vehicle (Abbott Labs, Abbott
Park, IL). Allyl alcohol (30 mg/kg i.p.) was administered two hours after
LPS. Six hours after treatment with allyl alcohol or saline control, rats

were anesthetized with sodium pentobarbital (50 mg/kg i.p.), and blood

98

was collected from the abdominal aorta into syringes containing 3.8%
sodium Citrate (final concentration 0.38%). Hepatotoxicity was

determined via the activity of ALT in plasma.

3.2.9 Statistical Analysis

Data are expressed as means :I: S.E.M. For all results presented,
N represents the number of individual experiments. Data with non-
homogeneous variances were square root- or log-transformed prior to
analysis. Data were analyzed using repeated measures ANOVA provided
in the statistics program NCSS 2000. Comparisons among groups were
performed using Fisher’s Least Squares Difference test. The criterion for

statistical significance was p s 0.05.

99

3.3 Results

3.3.1 Effect of Superoxide on Allyl Alcohol-Treated Hepatocytes
To test the hypothesis that LPS-induced, neutrophil-derived
reactive oxygen species increase the sensitivity of hepatocytes to the
reductive stress of allyl alcohol leading to an enhanced hepatotoxicity,
neutrophil-hepatocyte cocultures were used. Since the hypothesis was
that glutathione depletion by allyl alcohol renders hepatocytes more
sensitive to neutrophil-derived reactive oxygen species, it was important
to stimulate neutrophils to produce superoxide anion at a time when
glutathione was low. Accordingly, the time courses for glutathione
depletion by allyl alcohol in hepatocytes and superoxide generation by
PMA-stimulated neutrophils were determined. Allyl alcohol caused a
concentration-dependent decrease in hepatocellular GSH concentrations
within the first 5 minutes of exposure (Figure 3.1). GSH continued to
decrease for 20 minutes and then began to increase slowly.
' In freshly isolated neutrophils, PMA stimulated the production of
superoxide anion, and the rate of production was maximal by 6 min
(Figure 3.2). Based on these results, neutrophil-hepatocyte cocultures

were treated with allyl alcohol 20 minutes before addition of PMA.

100

Figure 3.1. Glutathione depletion occurs within 20 minutes after
allyl alcohol treatment of hepatocytes.

Isolated hepatocytes were treated with allyl alcohol at the
indicated concentrations and assessed for glutathione content at
the indicated times after allyl alcohol treatment. Percent initial
glutathione concentration is based on the concentration of
glutathione 5 minutes after treatment with 0 pM allyl alcohol.

Data are represented as means. N=4.

101

°/o Initial GSH

 

 

 

+0AA
v 25M
—I—-5OAA
0 75AA
—+-1OOAA
—9— 125M
—0— 150M
v 200M

 

 

 

Time (min)

102

 

Figure 3.2. PMA stimulates maximal superoxidggeneration from
neutrophils at 6 minutes.

Isolated neutrophils were plated in 96 well plates containing
cytochrome c for determination of superoxide generation.
Neutrophils were treated with either phorbol myristate acetate
(PMA) at 20 ng/ml (open Circle) or vehicle control (black Circle).

Data are represented as mean i S.E.M. N=4.

103

Rate of 02' generation

(AAbsorbance per 2 min)

0.04

 

9 .o
O O
N (A

0.01 -

0.00 .

 

-0.01

—9— Control

-—0— PMA 20 nglmj

 

.w.&.‘.A9§.o.4 ‘ .

 

20 40 6'0 8.0
Time (min)

104

 

During each individual experiment, superoxide anion production
from neutrophils was monitored in the presence of allyl alcohol. PMA
caused a consistent production of superoxide anion from the isolated
neutrophils. Allyl alcohol had a minimal effect on superoxide anion
production by PMA-stimulated neutrophils (Figure 3.3). Only the highest
concentration of allyl alcohol caused a small but statistically significant
decrease in superoxide anion production.

The glutathione concentration of hepatocytes cultured alone or
with PMNs was not altered by PMA (Figure 3.4). Allyl alcohol caused a
concentration dependent decrease in cellular GSH. Neither the presence
of PMNs nor PMA treatment had a significant effect on the concentration
ofGSH.

Allyl alcohol caused a concentration dependent increase in ALT
release from both neutrophil-hepatocyte cocultures and hepatocyte
cultures (Figure 3.5). PMA stimulation increased slightly but
significantly the sensitivity of hepatocytes to allyl alcohol in the

neutrophil cocultures but did not affect the hepatocytes cultured alone.

105

Figure 3.3. Superoxide generation by PMA-stimulated neutrophils
is not attenuated by allyl alcohol.

Isolated neutrophils were plated in 96 well plates containing
cytochrome C for determination Of superoxide anion production.
Total superoxide generation after 70 minutes of PMA (20 ng/ml)
stimulation was calculated. Data are represented as mean i
S.E.M. N=4. *, significatly different from control (0 pM allyl

alcohol).

106

 

 

100

 

 

5
2

.. .
AEEcoocaeomnm.
02:0 .89.: no.3

75

50

25

Allyl Alcohol (pM)

107

Figure 3.4. Allyl alcohol-inducedglutathione depletion is not
enhanced by PMA-stimulated neutrophils.

Isolated hepatocytes were cocultured with or without
isolated neutrophils (PMNs). The cells were treated with allyl
alcohol at concentrations indicated and 20 minutes later treated
with PMA (20 ng/ml) or vehicle. control. After a 90 minute
incubation with allyl alcohol, glutathione (GSH) concentration was
determined. Data are represented as means, N = 3. NO statistical

differences among groups.

108

 

 

 

 

 

 

 

 

 

01 . + PMN-I-PMA
' v —o— PMN

A 50 g + PMA

=_”..’ v \o -v— Control

8 \

c 40 'l V\~ .

9- v 9

:9 so '

E ‘ .3 '

0) v

(D 20 .

1O . . . . .

O 25 5O 75 100

Allyl Alcohol pM

109

 

Figure 3.5. PMA-stimulated release of reactive oxygen species
enhances allyl elcohol cytotoxicity in hepatocfie-neutrophil
co It r

Isolated hepatocytes were cultured alone (A) or cocultured
with neutrophils (B). Cells were treated allyl alcohol at the
indicated concentrations and 20. minutes later with phorbol
myristate acetate (PMA) (20 ng/ml) or vehicle. After 90 minutes
of allyl alcohol incubation cell viability was assessed by measuring
release of ALT activity. Data are represented as mean i S.E.M.
N=4. *, significantly different from control at same allyl alcohol

concentration.

110

tn % ALT Release

% ALT Release

 

80
70-
60-
50-
40-
304
20-
104

—<>-IMWA
+ Control

 

 

 

Hepatocyte culture

 

 

 

 

—o— PMA
+ Control

 

Hepatocyte-neutrophil
coculture

 

 

50 75 100

Allyl Alcohol (11M)

111

 

 

3.3.2 Effect of Neutrophil-Derived Proteases on Allyl Alcohol-Treated
Hepatocytes

Allyl alcohol caused a concentration dependent increase in ALT
release from hepatocyte cultures (Figure 3.6). Conditioned medium
from fMLP-stimulated neutrophils did not affect the amount of ALT

released into the medium.

3.3.3 Effect of Antioxidants In Viva

Animals treated with both LPS and allyl alcohol developed
significant liver injury as assessed by plasma ALT activity. Pretreatment
with superoxide dismutase and catalase did not offer any protection
from the hepatotoxicity induced by LPS and allyl alcohol cotreatment
(Figure 3.7). Plasma samples taken from rats one hour after treatment
with allyl alcohol were analyzed for superoxide dismutase activity.
Neutrophil-generated superoxide anion reduction of cytochrome C was
decreased 60% by plasma from superoxide dismutase treated animals
compared to vehicle treated animals (data not shown). Pretreatment
with the NADPH oxidase inhibitor apocynin did not induce liver injury
when administered alone (41 i 7 Units/L plasma ALT activity) and did
not afford protection from the cotreatment of LPS and allyl alcohol

(Figure 3.8).

112

Figure 3.6. Conditioned medium does not enhance the
cytotoxicity of allyl alcohol.

Isolated hepatocytes were cultured in naive medium
containing fMLP and cytochalasin B (black) or conditioned
medium containing supernatant from cytochalasin B- and fMLP-
treated neutrophils. Cells were treated with allyl alcohol at the
indicated concentrations. After 90 minutes, cell viability was
assessed by measuring ALT activity. Data are represented as

mean i S.E.M. N=4.

113

ALT Release in Medium (Units/L)

a
h

.3
N

.3
O

 

. + Naive medium

 

Clo-FOG

 

F0— Conditioned medium

 

 

 

 

6 2'5 5'0 7'5 160
Allyl Alcohol (pM)

114

 

Figure 3.7. Superoxide dismutase and catalase do not attenuate

LPS-enhanced allyl alcohol hepatotoxicity.

Rats were treated with LPS (1.2 X 105 EU/kg body weight,
i.v.). One hour later rats were treated with polyethylene-coupled
superoxide dismutase (10,000 units/kg, i.v.) and polyethylene-
coupled catalase (40,000 units/kg, i.v.) (SOD/CAT) or vehicle.
One hour later rats were treated with allyl alcohol (30 mg/kg, i.p.).
After an additional six hours liver injury was assessed via plasma

ALT activity. Data are represented as mean i S.E.M. N = 4.

115

 

Plasma ALT (Units/L)

 

4000

3500

3000 -
2500 -
2000 -
1500 -
1000 -l

Cl
C
O

 

O
L

 

Vehicle

116

 

SOD/CAT

 

Figure 3.8. Apocynin does not attenuate LPS-enhanced allyl
alcohol hepatotoxicity.

Rats were given apocynin (5 mg/kg, p.o.) or an equivalent
volume of 5% ethanol vehicle 2 hours before treatment with LPS
(100 pg/kg, i.v.).l Allyl alcohol (30 mg/kg i.p.) was administered
two hours after LPS. Six hours after treatment with allyl alcohol
liver injury was assessed via plasma ALT activity. Data are

represented as mean i S.E.M. N = 3.

117

 

 

Plasma ALT (Units/L)

 

1400 -
1200 .
1000 -
800 -
600 <
400 .
200 .

 

o .

Vehicle

118

 

Apocynin

 

3.4 Discussion

Neutrophils are a critical component of liver injury in this
potentiation model. Large numbers of neutrophils accumulate in the
liver sinusoids. Removal of neutrophils with an anti-neutrophil antibody
(Sneed, 2000) protects animals from LPS enhancement of allyl alcohol
hepatotoxicity, confirming the important role neutrophils play in organ
damage during LPS enhancement of allyl alcohol-induced injury.

It is widely held that neutrophils must become activated to
contribute to tissue injury. One function of activated neutrophils is
production of reactive oxygen species. Allyl alcohol decreases the
cellular levels of GSH (Figure 3.1), which leaves the cells more
susceptible to attack from reactive oxygen species. For example,
depletion of hepatic GSH with maleiC acid diethyl ester or phorone
increases oxidative hepatic injury in rats (Carbonell et al, 2000; Moriya et
al, 2000). Thus the hypothesis was posed that neutrophil-derived
superoxide anion contributes to LPS enhancement of allyl alcohol
toxicity.

To test this possibility in vitro experiments were performed with
neutrophil-hepatocyte cocultures. PMA was used to stimulate the
production of reactive oxygen species from PMNs at a time that

corresponded with maximal GSH loss in allyl alcohol-treated

119

hepatocytes. A small but significant increase in ALT release was
observed in cocultures of hepatocytes and PMA-stimulated neutrophils
(Figure 3.5). The small magnitude of this increase in allyl alcohol
toxicity raises questions about the pathophysiological relevance of this
effect. The ability of PMA-stimulated neutrophils to enhance the
cytotoxicity of allyl alcohol was greater on some days than others. This
was not correlated with the amount of superoxide produced by PMNs or
with the degree of glutathione depletion. The cytotoxicity-enhancing
effect of PMA-stimulated neutrophils in cocultures might depend on
other factors that were not monitored, such as vitamin E content,
glutathione peroxidase and/or reductase activity, or release of proteases
from PMA-stimulated neutrophils.

To examine the possibility that the increase in toxicity observed
with PMA-stimulated neutrophils was related to preoteases released into
the medium, experiments were performed with conditioned medium
from fMLP-activated neutrophils. fMLP stimulates PMNs to degranulate
and to produce reactive oxygen species. Because superoxide anion has
a very short half-life and the PMNs are spun out of the conditioned
medium, the conditioned medium was likely free of reactive oxygen
species and contained degranulation products including proteases. The
lack of effect of conditioned medium to increase the toxicity of allyl

alcohol in this paper indicates that the observed enhancement of allyl

120

alcohol cytotoxicity by PMA-stimulated neutrophils may be attributed to
superoxide anion production rather than PMA-stimulated protease
release. Nonetheless, the inability of conditioned medium to enhance
the cytotoxicity of allyl alcohol in vitro may not lead us to conclude a lack
of importance for protease release in vivo. The lack of enhanced allyl
alcohol cytotoxicity may be due to the short duration of the experiment.
Our lab has previously found that the proteases cathepsin G and elastase
are major contributors to hepatocellular injury from exposure in vitro to
conditioned medium from fMLP-activated neutrophils; however this
injury takes 16 hours to develop (Ganey et al, 1994; H0 at al, 1996).
Thus, the role of neutrophil-derived proteases cannot be ruled out by
experiments in this study.

Pretreatment of rats with two different antioxidants did not afford
protection from the potentiation of allyl alcohol hepatotoxicity by LPS.
The antioxidants superoxide dismutase and catalase act in concert to
scavenge superoxide anion. Polyethylene glycol-coupled enzymes were
used to increase the half-life of superoxide dismutase and catalase
(Tamura et al, 1988). The lack of protection with this antioxidant
treatment could be due to an insufficient concentration of antioxidants
for effective oxidant scavenging. However, plasma from rats treated with
polyethylene glycol-coupled superoxide dismutase and catalase

demonstrated a significant superoxide anion scavenging activity.

121

Furthermore, the doses used in our experiments were similar to those
used by other investigators for studies in which protective effects against
oxidative injury without toxic effects of antioxidant treatment alone were
observed (elSisi et al, 1993; Tamura et al, 1988; Thomson et al, 1990;
Suzuki et al, 1992; Lehman et al, 1992; Liu et al, 1989). For example,
vitamin A potentiation of carbon tetrachloride hepatotoxicity was
attenuated by treating animals with the same dose of superoxide
dismutase and catalase used in experiments presented here. The level
of lipid peroxidation was also reduced as measured by ethane
concentration in expired air (elSisi et al, 1993).

Lack of antioxidant-induced protection against LPS enhancement
of allyl alcohol toxicity could be due to an insufficient concentration of
antioxidants in the small space between the neutrophils and the
hepatocytes to which they adhere. However, in a model of reperfusion
injury in isolated rabbit hearts, injury was not as severe in hearts
isolated from rabbits pretreated with polyethylene glycol-coupled
sUperoxide dismutase. Because treatment with the antioxidant occurred
in vivo and the perfusion medium did not contain the superoxide
dismutase, the protective effect was most likely due to intramyocardial
or membrane-bound superoxide dismutase. This indicates that
polyethylene glycol linkage may facilitate intracellular access (Lehman et

al, 1992). Thus, polyethylene glycol-coupled superoxide dismutase and

122

catalase may reside on or in the membranes within the space between
the neutrophils and hepatocytes, and the inability of this antioxidant
treatment to protect against the potentiation of allyl alcohol
hepatotoxicity by LPS is not likely due to insufficient antioxidant
capacny.

To further examine the involvement of neutrophil-derived reactive
oxygen species in the potentiation of allyl alcohol hepatotoxicity by LPS
another antioxidant was used. Apocynin is metabolically activated by
reactive oxygen species and myeloperoxidase enzyme activity from
activated phagocytes into a compound that potently inhibits NADPH
oxidase assembly on the membrane (Stolk et al, 1994; Wang et al,
1994). Apocynin is thought to reduce the production of reactive oxygen
species by activated neutrophils without affecting phagocytosis,
exocytosis, or intracellular killing of bacteria (Wang et al, 1994). Though
apocynin pretreatment did not attenuate the potentiation of allyl alcohol
hepatotoxicity by LPS, 3 similar dose of apocynin was shown to
attenuate oxidative injury in other models of tissue injury without causing
toxicity on its own (Supinski et al, 1999; Salmon et al, 1998). For
example, disruption of diaphragmatic function after endotoxin
administration in rats is dependent on oxygen-derived free radicals.
Administration of apocynin at doses similar to those used in our

experiments was protective against endotoxin-induced decreases in force

123

production of diaphragm strips. Furthermore, this treatment regimen
resulted in decreased formation of 4-hydroxynonenal (a marker of lipid
peroxidation) and nitrotyrosine (a marker of free radical-mediated
protein modification) (Supinski et al, 1999). Thus the inability of
apocynin to attenuate the enhancement of allyl alcohol hepatotoxicity by
LPS was not likely due to insufficient inhibition of NADPH oxidase.
However, apocynin elicits other effects such as inhibition of thromboxane
A2 release and the stimulation of release of prostaglandins E2 and F2 as
well as increases in glutathione synthesis (Engels et al, 1992, Lapperre et
al, 1999). The ability of apocynin to stimulate release of prostaglandins
may have a toxic effect in this potentiation model, because a
cyclooxygenase-2 inhibitor attenuates the potentiation of allyl alcohol
hepatotoxicity by LPS (Ganey et al, 2001). This might explain the
somewhat greater toxicity observed in allyl alcohol/LPScotreated rats
given apocynin.

Although neutrophils are required for hepatotoxicity in this model,
reactive oxygen species do not appear to play a critical role in LPS
enhancement of allyl alcohol-induced hepatocellular injury. The small
effect of reactive oxygen species observed in vitro may indicate that the
presence of both reactive oxygen species and proteases is necessary to

elicit the injury observed in vivo.

124

CHAPTER 4

The Role of Thrombin in the Potentiation of Allyl Alcohol-

lnduced Liver Injury by LPS

125

4.1 Introduction

The potentiation of allyl alcohol-induced liver injury by LPS is
dependent on Kupffer cell activity. Inhibition of Kupffer cell activity with
gadolinium Chloride protects against the LPS enhancement of allyl
alcohol toxicity. However, we have previously shown in Kupffer
cell/hepatocyte cocultures and the isolated, perfused liver that LPS-
stimulated Kupffer cells (KCs) do not sensitize hepatocytes (HCs) to
injury from allyl alcohol (Chapter 2). These data indicate that KCs are
not sufficient for LPS to enhance the toxicity of allyl alcohol and that
other extrahepatic factors must play a critical role in the potentiation of
allyl alcohol-induced liver injury by LPS.

Though liver injury from large doses of allyl alcohol alone is not
dependent on the presence of extrahepatic factors, liver injury from large
doses of LPS is critically dependent on the extrahepatic factor thrombin
(Moulin et al, 1996). Thrombin induces liver injury in livers isolated from
LPS pretreated rats. Thrombin elicits its damaging effects on liver
through a thrombin receptor-mediated mechanism (Moulin, 1999).
Studies presented here were designed to test whether thrombin plays a
role in LPS enhancement of allyl alcohol injury that is similar to its role
in liver damage from large doses of LPS. The hypothesis was that

thrombin participates in the potentiation of allyl alcoholoinduced liver

126

injury by LPS through a mechanism involving a thrombin receptor in the

liver.

4.2 Methods

4.2.1 Materials

Primers for cytokine-induced neutrophil chemoattractant-1 (CINC-
1) and cyclooxygenase-2 (COX-2) were purchased from Biosource
International, Camarillo, CA and Ambion, Austin, TX respectively. Tri
Reagent was from Molecular Research Center, Cincinnati, OH.
RETROscript protocol and Quantum RNA kit were from Ambion, Austin,
TX. All other materials were purchased from Sigma Chemical Co., St.

Louis, MO.

4.2.2 Animals

Male, Sprague-Dawley rats (CD-Crl:CD-(SD)BR VAF/Plus; Charles
River, Portage, MI) weighing 175-225 g were allowed food (Wayne Lab-
Blox, Allied Mills, Chicago, IL) and water ad Iibitum. They were housed
under conditions of controlled temperature, humidity and 12-hr light and

dark cycle.

127

4.2.3 In Vim Experiments with Heparin

Rats were treated with heparin (2000 Units/kg body weight, i.v.)
or an equivalent volume of saline as control one hour before
administration of LPS. Rats were treated with LPS (1.2X105 EU/kg body
weight, i.v.), and two hours later they were treated with allyl alcohol (30
mg/kg body weight, i.p.). At 2, 3, and 8 hours after LPS administration
rats were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) for
removal of blood (in 0.38% sodium Citrate) and liver samples. Plasma
samples were analyzed for ALT, aspartate aminotransferase (AST) and
alkaline phosphatase activities using Sigma Diagnostic Kits No. 52, 51,
and 245 respectively, and for fibrinogen concentration using a BBL '
fibrometer (Becton, Dickinson and Company, Hunt Valley, MD). Livers
were weighed, and a portion of the liver was frozen immediately in liquid
nitrogen and stored at —80°C for preparation of RNA (see below). The
remaining liver sample was fixed in formalin and processed for
histological evaluation. Liver sections were stained for PMNs as
described previously (Pearson et al, 1995) using an anti-PMN antibody
prepared as described previously (Hewett et al, 1992). Stained cells
were counted under light microscopy in 20 random fields at 400x

magnification.

128

4.2.4 Determination of CINC-l and COX-2 mRNA Expression Levels
Total RNA was isolated from frozen liver tissue using Tri Reagent
(Molecular Research Center, Cincinnati, OH) as described by
Chomczynski (Chomczynski, 1993; Chomczynski and Mackey, 1995).
The concentration and purity of RNA was determined from absorbance
at 260 and 280nm. RNA was adjusted to 50pg/ml with RNAse-free
water. First strand CDNA was synthesized using the RETROscript
protocol (Ambion, Austin, TX). Semiquantitative polymerase chain
reaction (PCR) was performed using the Quantum RNA kit (Ambion).
188 rRNA was used as the internal control. Primers for ClNC-l and
COX-2 were obtained from Biosource International and Ambion
respectively. Samples were denatured for 90 seconds at 94°C, followed
by 35 cycles of 30 seconds at 94°C, 45 seconds at 60°C, and 45
seconds at 72°C with a final 7 minute extension step at 72°C. PCR
products were separated via electrophoresis on a 1.5 % agarose gel
containing ethidium bromide. Band intensity was quantified using
Quantity One quantitation software (version 4, BIO-RAD). Band
intensities for COX-2 and ClNC-l were normalized to the band intensity

of the 18S rRNA internal standard.

129

4.2.5 In Viva Experiments with a Hepatotoxic Dose of Allyl Alcohol

Rats were treated with heparin (2000 Units/kg body weight, i.v.)
or an equivalent volume of saline as control. Three hours after heparin,
rats were treated with allyl alcohol (50 mg/kg body weight, i.p.). At 9
hours after heparin administration, rats were anesthetized with sodium
pentobarbital (50 mg/kg, i.p.) for removal of blood (in 0.38% sodium
citrate). Blood plasma samples were analyzed for ALT and fibrinogen as

described above.

4.2.6 In Viva Experiments with Warfarin

Rats were treated with warfarin (20 mg/kg body weight, p.o.) or
vehicle twice, 24 hours apart. 18 hours after the last treatment, rats
were given LPS (1.2X105 EU/kg body weight, i.v.) or saline vehicle, and
two hours after LPS they were treated with allyl alcohol (30 mg/kg body
weight, i.p.) or vehicle. 8 hours after LPS administration, rats were
anesthetized with sodium pentobarbital (50 mg/kg, i.p.) for removal of
blood (in 0.38% sodium citrate) and liver samples. Plasma samples

were analyzed for ALT and fibrinogen concentration as described above.

130

4.2.7 Isolation and Perfusion of Rat Livers

Rats were weighed and then treated with LPS (1.2X105 EU/kg
body weight, i.v.) or an equivalent volume of saline as control. After two
hours, the rats were anesthetized with sodium pentobarbital (50 mg/kg,
i.p.), and the portal vein was exposed. Livers were perfused via portal
vein cannulation with polyethylene tubing (PE 190, Clay Adams,
Parsippany, NJ). The perfusion medium was Krebs-Henseleit
bicarbonate buffer supplemented with 2% bovine serum albumin and
saturated with 95% 02 and 5% 002 gas. Flow was constant at a rate of
0.14 ml/min/g body weight. The thoracic portion of the inferior vena
cava was cannulated (PE 240 Clay Adams, Parsippany, NJ) for outflow.
The liver was placed in a temperature-controlled cabinet maintained at
37°C as previously described (Moulin et al, 1996). The livers were
allowed to stabilize for 10 minutes with single pass perfusion. A sample
of the perfusion medium was then taken (time = SPP), and the system
was switched to recirculating perfusion. Thrombin (70 units/rat) or
vehicle control was added to the perfusion medium before the addition
of allyl alcohol (32.5 mg/kg body weight) or vehicle control. Medium
samples (350pL) were taken every 15 minutes for two hours. ALT

activity in the samples was determined as described above.

131

4.2.8 Statistical Analysis

Data are expressed as means :I: S.E.M. For all results presented,
N represents the number of individual experiments. Data with non-
homogeneous variances were square root- or log-transformed prior to
analysis. Data were analyzed using repeated measures ANOVA provided
in the statistics program NCSS 2000. Comparisons among groups were
performed using Fisher’s Least Squares Difference test. The criterion for

statistical significance was p s 0.05.

132

4.3 Results

One extrahepatic factor that is critical for liver injury induced by
large doses of LPS alone is thrombin (Hewett et al, 1995; Moulin et al,
1996). To begin to assess the importance of thrombin in the LPS
enhancement of allyl alcohol hepatotoxicity, rats were treated with
heparin (2000 Units/kg) one hour prior to LPS. Heparin is an
anticoagulant that enhances the binding of antithrombin III to thrombin,
inhibiting the formation of active thrombin. Plasma fibrinogen
concentration was monitored as a marker of thrombin activity and drug
effectiveness. Thrombin proteolytically cleaves fibrinogen to fibrin; thus
activation of thrombin is associated with a decrease in Circulating
fibnnogen.

The plasma fibrinogen concentration of animals treated only with
LPS did not decrease, whereas that of animals treated with allyl alcohol
alone decreased a small but insignificant amount. Plasma fibrinogen
concentrations in animals co-treated with LPS and allyl alcohol were
significantly decreased 3 hours after LPS administration (Table 4.1).
Pretreatment with heparin significantly blocked the decreases in plasma
fibrinogen concentration. Plasma ALT activity was not elevated above
those of animals receiving saline alone in any of the treatment groups
indicating that significant liver injury had not developed by 3 hours after

LPS administration. Plasma ’AST activity was slightly but significantly

133

Table 4.1. Plasma ALT. AST. and fibrinogen values in rats 3 hours
after LPS administration.

Rats were treated with heparin (Hep; 2000 Units/kg body
weight, i.v.) or saline one hour before LPS (1.2 X 105 EU/kg body
weight, i.v.) or saline. Two hours after LPS, rats received allyl
alcohol (AA; 30 mg/kg body weight, i.p.) or saline. Plasma ALT
and AST activities, and fibrinogen concentration were measured 3
hours after LPS administration. Data are expressed as means i

S.E.M. N = 3-5. *, significantly different from saline only.

134

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

HeplLPSI AAIFibrinogen (mgldL!T ALT (Units/L) | AST (Units/L) !
- - - 200 :I: 18 120 1:25 76 21:23
+ - - 225 121 121 :I: 15 103 :I: 12
- + - 233 1:26 138 1:29 107 125
+ + - 203 1:27 144 :I:31 85 :I: 18
- - 4' 152 141 107 1:27 142 136
'1' - 4' 226 :I:8 128 :27 87 :I: 15
. + + *39 114 139 :24 *226 :I:66
4' + + 214 1:35 173 1:34 149 :I:41

 

 

 

 

 

 

 

135

elevated in animals receiving LPS and allyl alcohol co-treatment
compared to animals receiving saline only.

The effectiveness of heparin at inhibiting thrombin was also
demonstrated by measuring plasma fibrinogen concentrations at 8 hours
post LPS administration (Figure 4.1). Fibrinogen was depleted only in
the LPS-allyl alcohol cotreatment group. Heparin pretreatment
significantly attenuated this depletion of fibrinogen, indicating that an
effective dose of heparin was used. Liver injury was assessed via the
release of AST and ALT into the plasma 8 hours after LPS
administration. Neither LPS nor allyl alcohol alone induced significant
ALT release; however, LPS-allyl alcohol cotreatment induced significant
ALT release at 8 hours. Similar results were observed with AST activity.
There were no significant Changes in alkaline phosphatase activity or
liver weight/body weight index at any time after treatment (data not
shown). Treatment with heparin significantly attenuated'the LPS-
enhanced toxicity of allyl alcohol as assessed by plasma ALT and AST
aCtivities at 8 hours post LPS administration (Figure 4.2).

Heparin has been shown to inhibit various processes of
inflammation (Downing ef al, 1998; Darien et al, 1998; Shin et al, 1997).
To assess its effect on inflammation in our study, the number of

neutrophils in the liver, expression of the Chemokine, CINC-l and

136

Figure 4.1. Heparin pretreatment attenuates fibrinogen eepletien
in rats cotreated with LPS and allyl alcohol.

Rats were treated with heparin (2000 Units/kg body weight,
i.v.) or vehicle one hour before LPS (1.2 X 105 EU/kg body weight,
i.v.) or vehicle (saline). Two hours after LPS, rats received allyl
alcohol (30 mg/kg body weight, i.p.) or vehicle. Plasma fibrinogen
was measured 8 hours after LPS administration. Data are
expressed as means i S.E.M. N = 3-5. *, significantly different
from saline only; T, significantly different from LPS/AA in the

absence of heparin.

137

 

 

 

350

 

A - Saline
—l 300 - r=l Heparin
E
g, 250 -
V200 . . *.1'
C
8,150 . l
O l
5 100. *
h l
:2 50 .
LL l l
o I I

 

 

 

 

 

Saline LPS Saline LPS
Saline Allyl Alcohol

Figure 4.2. Heparin pretreatment attenuates LPS-enhanced allyl
alcohol hepatotoxicity in vivo.

Rats were treated with heparin (2000 Units/kg body weight,
i.v.) or vehicle (control) one hour before LPS (1.2 X 105 EU/kg
body weight, i.v.) or vehicle (saline). Two hours after LPS, rats
received allyl alcohol (30 mg/kg body weight, i.p.) or vehicle
(saline). (A) ALT and (B) AST activities were measured in plasma
samples as markers of liver toxicity 8 hours after LPS
administration. Data are expressed as means i S.E.M. N = 6. *,
significantly different from saline only (control). 1, Significantly

different from respective value in the absence of heparin.

139

 

 

 

>

3000

 

N
0|
O
O

2000 -

1500 -

1000 .

Plasma ALT (Units/L)

500 -

 

 

— Saline
l:::l Heparin

   
  

   
  

  
  

LP

Saline

Saline

LPS

 

 

Saline

Allyl Alcohol

 

Plasma AST (Units/L)

 

— Saline
l:=:.l Heparin

  

LPS

Saline

*

 

 

LPS

Saline

 

 

Saline

140

Allyl Alcohol

 

 

expression of COX-2 were measured in LPS-treated groups. In livers
from control rats, few neutrophils (PMNs) were present (Figure 4.3).
Treatment with LPS in the absence or presence of allyl alcohol resulted
in accumulation of PMNs in liver tissue (Figure 4.3). Heparin
pretreatment did not attenuate the LPS-dependent increase in the
number of PMNs in the liver. ClNC-l mRNA expression was increased
by treatment with LPS in the absence or presence of allyl alcohol, and
heparin pretreatment significantly increased its expression level (Figure
4.4). Allyl alcohol treatment alone did not enhance the expression of
ClNC-1 mRNA. COX-2 mRNA expression was increased by treatment
with LPS in the absence or presence of allyl alcohol, and heparin
pretreatment did not significantly alter its expression level (Figure 4.5).
Allyl alcohol treatment alone did not significantly enhance the expression
of COX-2 mRNA.

To determine whether heparin would affect the toxicity of allyl
alcohol in the absence of LPS, a large dose of allyl alcohol (50 mg/kg)
was administered to rats that had been treated with heparin or vehicle 3
hours earlier. Whereas allyl alcohol caused significant liver injury,
pretreatment with heparin did not significantly attenuate the injury (ALT
activity in plasma was 631 j: 189 and 393 i 102, respectively) (Figure
4.6). In addition, the concentration of plasma fibrinogen from rats

treated with a large dose of allyl alcohol did not decrease from normal

141

Figure 4.3. Neutrophil infiltration is not attenuated by heparin.
Rats were treated with heparin (2000 Units/kg body weight,
i.v.) or vehicle one hour before LPS (1.2 X 105 EU/kg body weight,
i.v.) or vehicle. Two hours after LPS, rats received allyl alcohol (30
mg/kg body weight, i.p.) or vehicle. Liver sections were taken 3
and 8 hours after LPS administration and stained for neutrophils
(PMNs). The number of stained cells were counted under light
microscopy in 20 fields at 4OOX magnification. Data are
expressed as means i S.E.M. N = 3-6. *, significantly different

from saline only (control).

142

 

 

:2 Control

 

             

M + M
.m
L. m + *
s 8 ps
P P e P
L L H L 1. weeveexex
m m _ 33949494943949?”

.T7///////n
m

 

 

 

 

00 0000
65 321

"E: Quiz—amt

8 hour

 

143

3 hour

 

Figure 4.4. ClNC-l mRNA expression is not attenuated by
hepann.

Rats were treated with heparin (2000 Units/kg body weight,
i.v.) or vehicle one hour before LPS (1.2 X 105 EU/kg body weight,
i.v.) or vehicle. Two hOurs after LPS, rats received allyl alcohol (30
mg/kg body weight, i.p.) or vehicle. Liver sections were taken 3
hours after LPS administration and assessed for ClNC-l mRNA by
RT-PCR. ClNC-l band intensity was normalized to 18S rRNA band
intensity. Data are expressed as means i S.E.M. N = 4. *,

significantly different from saline only (control); 1‘, significantly

different from LPS/AA in the absence of heparin.

144

 

Relative Intensi

.3
O

 

l l l

O-éNw-FGO’NQCD

 

ClNC-1 mRNA

 

 

 

gags:

 

, ' \ . If
. . _..’_.
t _ ‘9;
~ '.

.4 ,_ .‘.

 

 

 

 

 

 

Control

LPS

AA LPS Heparin
AA LPS
AA

145

 

Figure 4.5. COX-2 mRNA expression is not attenuated by heparin.

Rats were treated with heparin (2000 Units/kg body weight,
i.v.) or vehicle one hour before LPS (1.2 X 105 EU/kg body weight,
i.v.) or vehicle. Two hours after LPS, rats received allyl alcohol (30
mg/kg body weight, i.p.) or vehicle. Liver sections were taken 3
hours after LPS administration, and assessed for COX-2 mRNA by
RT-PCR. COX-2 band intensity was normalized to 188 rRNA band

intensity. Data are expressed as means i S.E.M. N = 4. ,

significantly different from saline only (control).

146

 

2.0

.3
0|

Relative Intensity
3

P
a:

 

0.0

COX-2 mRNA

     

 

 

 

 

 

 

 

, . L
~ . I.-- -.“
. a 1
,.. ..
l .. C"
. ' ,li'u

. . . 1|!"
u .. '\ .';
-’ L"_ 3&5
.7 ‘ - ‘ l
L

 

Control Li’s KA Li’s Hefiarin
AA LPS
AA

147

 

Figure 4.6.‘ Heparin does not inhibit the toxicity of allyl alcohol

alone.

 

Rats were treated with heparin (2000 Units/kg body weight,
i.v.) or vehicle three hours before allyl alcohol (50 mg/kg body
weight, i.p.). Plasma samples were taken 6 hours after allyl
alcohol administration for determination of ALT activity. Data are

represented as mean i S.E.M. N = 14-15.

148

 

 

 

 

 

 

 

 

 

 

1000

H .m

r

.l .m

. e

H

e

.m

.l. .4.

S
0 0 0 0
0 0 0 0
8 6 4 2
35:5. 5< «Emma

149

Figure 4.7. Allyl alcohol does not decrease circulating fibringgg
concentration.

Rats were treated with heparin (2000 Units/kg body weight,
i.v.) or vehicle three hours before allyl alcohol (50 mg/kg body
weight, i.p.). Plasma samples. were taken 6 hours after allyl
alcohol administration for determination of fibrinogen

concentration. Data are represented as mean i S.E.M. N = 7-8.

150

 

 

 

 

 

 

Hepafin

 

 

 

Sahne

 

 

0
5
2

u I I d

o
2

Same. Sneeze“.

 

o
0 5 o 5
1

151

levels and was not altered by pretreatment with heparin (plasma
fibrinogen concentration was 209 .t 7 and 210 :t 15, respectively) (Figure
4.7).

To confirm that the attenuation of LPS-enhanced allyl alcohol
hepatotoxicity was due to inhibition of coagulation by heparin another
anticoagulant was used, Warfarin. Warfarin inhibits the coagulation
cascade through inhibition of the synthesis of functional coagulation
factors VII, IX, X, and II (prothrombin). Plasma fibrinogen concentration
was significantly decreased in rats treated with both LPS and allyl
alcohol, and pretreatment with warfarin blocked the LPS and allyl
alcohol-induced decrease in plasma fibrinogen (Figure 4.8). Liver injury
from LPS and allyl alcohol cotreatment was attenuated in rats pretreated
with warfarin (Figure 4.9). Warfarin treatment alone did not cause liver
injury.

To determine whether thrombin plays a role in the potentiation of
allyl alcohol toxicity by LPS through a mechanism mediated by a
thrombin-receptor in the liver, as it does in LPS-induced liver injury, the
isolated, perfused liver model was used. Rats were treated in vivo with
LPS, to allow neutrophil accumulation in the liver, and two hours later
their livers were isolated and perfused with buffer in a recirculating
manner. Two hours corresponds to the time that rats would receive allyl

alcohol during in vivo experiments. Medium-perfused livers were treated

152

Figure 4.8. LPS-enhanced. allyl alcohol-induced decrease in
plasma fibrinogen is blocked by warfarin pretreatment.

Rats were treated with warfarin (20 mg/kg body weight,
p.o.) or saline vehicle twice 24 hours apart. Then 18 hours after
the last warfarin treatment rats were treated with LPS (1.2X105
EU/kg body weight, i.v.), and .two hours after LPS they were
treated with allyl alcohol (30 mg/kg body weight, i.p.). Eight
hours after LPS administration plasma samples were analyzed for
fibrinogen concentration. Data are represented as mean i S.E.M.

N = 4-6. *, significantly different from saline/saline control.

153

Fibrinogen (mgldL)

 

h
D
O

c»
O
O

N
O
O

.3
O
O

 

— Saline
'l' c: Warfarin

 

 

 

 

T

 

  

 

 

 

 

 

 

 

Saline LPS
Allyl Alcohol

154

Figure 4.9. Warfarin attenuates LPS-enhanced allyl alcohol
hepatotoxicity.

Rats were treated with warfarin (20 mg/kg body weight,
p.o.) or saline vehicle twice 24 hours apart. Then 18 hours after
the last warfarin treatment rats were treated with LPS (1.2X105
EU/kg body weight, i.v.) and two hours later with allyl alcohol (30
mg/kg body weight, i.p.). Eight hours after LPS administration
plasma samples were analyzed for ALT activity. Data are
represented as mean 1r S.E.M. N = 4-15. *, significantly different

from control at p<0.05.

155

 

 

AAAAA

 

 

 

 

 

 

   

 

E

.22 3000

'E — Saline

3 2500 . i::::l Warfarin

E‘ 2000

..>.

‘6 1500 .

<

E 1000 -

u 500 ‘

G Saline . LPS
5: Allyl Alcohol

156

 

Figure 4.10. Thrombin does not enhance the hepatotoxicity of
allyl alcohol in isolated. buffer-perfused livers from LPS-treated
m5...

Livers were isolated from rats 2 hours after treatment with
LPS (1.2 X 105 EU/kg body weight) and subsequently perfused
with allyl alcohol at the indicated doses (numbers in legend are
mg/kg) and with either 70 units/liver thrombin (open symbols) or
vehicle (filled symbols) for an additional two hours. ALT released
into the medium was assessed over time as a marker of liver
injury. Data are. expressed as means iS.E.M. N=2-8. No

significant thrombin effect was observed.

157

 

ALT Activity (Units/L)

 

1200-

1000-

800-

600-

400

—1»-Vbhfi
—o— Throm,0
+ Veh,20
—v— Throm,20
+ Veh,25
—<>— Throm,25
-Ih-Vbh30
—Cl— Throm,30
+ Veh,32.5
—A— Throm,32.5
+ Veh,35
—o— Throm,35

 

 

200-

 

 

 

 

0 SPP 0.25 0.5 0.75 1 1.25 1.5 1.75 2

Time (hours)

158

 

with thrombin and the indicated concentrations of allyl alcohol (Figure
4.10). Liver injury was assessed from the ALT activity of the perfusion
medium. Allyl alcohol-induced liver injury was found to be both time-
and concentration-related. However, addition of thrombin did not
enhance the hepatotoxicity of allyl alcohol at any concentration of allyl
alcohol used.

In a separate isolated, perfused liver experiment the effect of LPS
on allyl alcohol-induced hepatotoxicity was determined. Rats were
treated in vivo with LPS or saline, and two hours later their livers were
isolated and perfused in a recirculating manner as described above.
Allyl alcohol (32.5 mg/kg body weight) was then added to the perfusion
medium, and samples were taken for an additional two hours. Liver
injury as assessed by ALT activity increased with time during perfusion
with allyl alcohol. Livers from rats pretreated with LPS developed
significantly greater injury than livers from rats pretreated with saline
(Figure 4.11).

Thrombin did not increase injury in livers isolated from LPS-
treated rats. It is possible that activation of thrombin in vivo prior to
iSolation of the livers could have caused thrombin receptors to become
refractory to further thrombin stimulus. To determine whether this was
a possibility, plasma was collected from rats two hours after

administration of LPS or saline (i .e., at the time livers were isolated for

159

Figure 4.11. Livers isolated from LPS-treated rats are more
sensitive to allyl alcohol-induced injury than control livers.

Livers were isolated from rats 2 hours after treatment with
LPS (1.2 X 105 EU/kg body weight) (closed squares) or saline
(open circles). Subsequently, livers were perfused with medium
containing allyl alcohol (32.5 mg/kg) for a period of two hours.
ALT released into the medium was assessed over time as a marker
of liver injury. Data are expressed as means i S.E.M. N = 5-6. *,

significantly different from saline at corresponding time.

160

800

700 -

#0105
O
0

ALT Activity (Units/L)
O
O

 

7 -o— Saline

 

 

O
O

300 -
200 .
100 «

 

 

 

 

 

o SPP 0.25 0.5 0.75 1 1.25 1.5 1.75 2
Time (hours)

161

 

Figure 4.12. Thrombin is activated by 2 hours in LPS-treated rats.

Blood plasma was isolated from rats 2 hours after
treatment with LPS (1.2 X 105 EU/kg body weight) or saline
control. Plasma fibrinogen concentration was determined as
described in Methods. Data are expressed as means 1r S.E.M. N =

15. *, significantly different from saline control.

162

 

 

 

 

 

 

 

 

 

 

*T
T
0 0 0 0 0
0 5 0. 5
2 1 1

3.2922 59.55:. new“...

Les

Sahne

163

perfusion), and plasma fibrinogen concentration was determined. A
significant decrease in plasma fibrinogen concentration was observed
(saline, 152 i 8 mg/dL; LPS, 113 :t 9 mg/dL) (Figure 4.12). To test
whether this activation in vivo prevented the effect of thrombin in isolated
livers, rats were pretreated with heparin (2000 Units/kg body weight)
one hour before LPS treatment in vivo. Two hours later the livers were
isolated and perfused with medium containing allyl alcohol in the
absence or presence of thrombin. Though allyl alcohol caused an
increase in ALT release over time, addition of thrombin did not enhance
the hepatotoxicity of allyl alcohol (Figure 4.13).

The ability of thrombin to induce liver injury in the isolated liver is
dependent on the presence of neutrophils (Moulin, 1999). We
considered the possibility that at the small dose of LPS used in these
studies, not enough neutrophils were infiltrating the liver to support this
effect of thrombin in the isolated, perfused liver. To examine this
possibility the number of neutrophils was counted in liver sections from
animals two hour after treatment with LPS. Hepatic neutrophil numbers
were also determined just after single-pass perfusion of isolated livers
from LPS-treated rats and at the end of isolated, perfused liver
experiments. In livers from LPS-treated rats there were about 33
neutrophils per 20X field at two hours (i.e. the time livers are isolated).

After isolation of the livers and single-pass perfusion for 15 minutes the

164

Figure 4.13. Thrombin does not enhance the h ot xi i
all I alcohol in isol ted erfused livers from rats tr wi h
heparin and LPS.

Livers were isolated from rats 2 and 3 hours after treatment
with LPS (1.2 X 105 EU/kg body weight) and heparin (2000
Units/kg body weight), respectively. Subsequently, livers were
perfused with allyl alcohol (32.5 mg/kg) and either thrombin (7O
Units/rat) (closed circle) or vehicle control (open circle) for a
period of two hours. ALT released into the medium was assessed
over time as a marker of liver injury. Data are expressed as

means. N = 2.

165

 

 

 

ALT (Units/L)

 

 

1400 - —0— Control
+ Thrombin

 

 

 

 

 

6 $1390.55 0.5 0.215 1 1.125 1:51.215 23
Time (hours)

166

number of neutrophils decreased to 13 neutrophils per 20X field. By the
end of the two~hour recirculating perfusion experiment about 7
neutrophils per 20X field remained in the liver. This is much greater

than the 0.2 neutrophils per 20X field found in livers from naive rats.

4.4 Discussion

In previous studies isolated, perfused livers from naive rats were
perfused with buffer containing LPS or saline for two hours and allyl
alcohol for an additional two hours (Chapter 2). The inability of LPS to
enhance the hepatotoxicity of allyl alcohol as observed in vivo led to the
possibility that extrahepatic factors play a critical role in the potentiation
of allyl alcohol-induced liver injury by LPS. Injury from larger doses of
allyl alcohol alone is not dependent on extrahepatic factors as evidenced
by the lack of protection offered by heparin or neutrophil depletion
(Ganey et al, 1995) and the ability of allyl alcohol to injure isolated
hepatocytes (Sneed et al, 1997). However, LPS-induced liver injury from
large doses of LPS is critically dependent on extrahepatic factors (Moulin
et al, 1996; Hewett et al, 1992). One extrahepatic factor that plays a
critical role in LPS-induced liver injury is thrombin. In a rat model of
liver injury induced by large doses of LPS, inhibitors of the coagulation
cascade such as heparin, warfarin, or a specific inhibitor of thrombin,

hirudin, each attenuated LPS-induced liver injury (Hewett et al, 1995;

167

Pearson et al, 1996). Thus it was of interest to examine a possible role
for thrombin in the LPS enhancement of allyl alcohol hepatotoxicity.

To begin to assess whether thrombin or the coagulation cascade
plays a role in LPS-enhanced allyl alcohol hepatotoxicity, heparin was
used to inhibit thrombin activity. Heparin enhances the ability of anti-
thrombin III to bind to thrombin, thereby inhibiting the activity of
thrombin. Pretreatment with heparin reduced liver injury from the
cotreatment of allyl alcohol and LPS in vivo. These data suggest that
thrombin participates in the response. However, although heparin
inhibits thrombin it also has other actions. Heparin has been shown to
inhibit the infiltration of PMNs into tissues and the production of
superoxide anion and nitric oxide from stimulated neutrophils (Downing
etal, 1998; Darien et al, 1998; Shin et al, 1997; Riesenberg et al, 1995;
Beltran et al, 1999). To determine if parts of the inflammatory response
were decreased in our experiment several aspects of inflammation were
assessed. ClNC-l is a Chemokine that is released during inflammation
and helps to bring neutrophils into tissues (Zhang et al, 1995). Neither
the expression of ClNC-l mRNA nor LPS~induced neutrophil
accumulation in liver was attenuated by pretreatment with heparin. ln -
addition, LPS and other inflammatory stimuli upregulate the inducible
cyclooxygenase enzyme, COX-2 (Ruetten et al, 1997). We know from

studies in our laboratory that expression of COX-2 plays a critical role in

168

the potentiation of allyl alcohol hepatotoxicity by LPS (Ganey et al,
2001). Expression of COX-2 mRNA was increased in livers from rats
cotreated with LPS and allyl alcohol, but this increase was unaffected by
pretreatment with heparin. These findings demonstrate that heparin
pretreatment did not have an inhibitory effect on several aspects of the
inflammatory response in our experiment. Therefore the protective
effect of heparin against liver injury was likely not due to an inhibition of
the inflammatory response but rather to its ability to block the
coagulation cascade.

Though the results of heparin pretreatment suggest that the
mechanism of protection was through its inhibition of coagulation, we
confirmed this by using another anticoagulant that works through a '
completely different mechanism to inhibit the coagulation cascade.
Warfarin is an anticoagulant that inhibits the regeneration of reduced
vitamin K. Reduced vitamin K is critical for the carboxylation of
glutamate to y-carboxyglutamate residues on precursor protein factors
Vll, IX, X, and II (prothrombin). Thus warfarin inhibits the coagulation
cascade by inhibiting the formation of functional coagulation factors.
Using this alternate method of inhibiting the coagulation cascade, the
potentiation of allyl alcohol toxicity by LPS also was attenuated.
Collectively these data suggest that the coagulation cascade plays a

critical role in the potentiation of allyl alcohol hepatotoxicity by LPS.

169

To confirm that attenuation of LPS-enhanced allyl alcohol-induced
liver injury was specific to the potentiation and not simply attenuation of
the toxicity of allyl alcohol, heparin pretreated rats were treated with a
large dose of allyl alcohol. Not only did heparin fail to decrease allyl
alcohol-induced liver injury at this large dose, but this dose of allyl
alcohol did not activate the coagulation cascade as assessed by plasma
fibrinogen concentration. This indicates that the observed protective
effect of heparin against LPS enhancement of allyl alcoholinduced liver
injury is unique to the potentiation model and provides evidence that the
mechanism of injury in the potentiation model differs from allyl alcohol-
induced liver injury. Further, it provides evidence that heparin did not
decrease the bioactivation of allyl alcohol.

As mentioned above thrombin plays a critical role in liver injury
from large doses of LPS. The mechanism by which thrombin contributes
to LPS-induced injury involves a specific receptor for thrombin, protease-
activated receptor-1 (PAR-1) found in the liver (Moulin et al, 1996;
Moulin, 1999). To examine the possibility of thrombin acting through a
receptor-mediated mechanism in the LPS-induced potentiation of allyl
alcohol, isolated livers from LPS-treated rats were perfused with allyl
alcohol and either thrombin or vehicle. Addition of thrombin to'the
medium did not enhance the hepatotoxicity of allyl alcohol. However,

livers from LPS-treated rats were more sensitive to allyl alcohol-induced

170

injury than livers from saline-treated rats. Analysis of plasma fibrinogen
concentration from rats two hours after saline or LPS injection revealed
that LPS caused a slight but significant activation of the coagulation
cascade in this time frame. This indicates that thrombin had been
activated in vivo by LPS before isolation of the livers. Thus, one
explanation for the lack of effect of thrombin in the isolated, perfused
liver is that prior thrombin activation caused a downregulation or
internalization of thrombin receptors in the liver preventing their
subsequent activation in the isolated, perfused liver. To test this
hypothesis, 3 limited number of rats was pretreated with heparin one
hour before LPS, and livers were isolated and perfused as in the previous
experiments. Thrombin did not enhance allyl alcohol-induced injury in
these livers. This suggests that prior activation of thrombin and possible
internalization of thrombin receptors does not explain the inability of
thrombin to enhance the hepatotoxicity of allyl alcohol. Taken together
the data suggest that enhancement of allyl alcohol-induced liver injury is
dependent on thrombin activation of blood components or some other
components of the coagulation cascade. One potential component is the
neutrophil.

Though there were a greater number of neutrophils in livers from
LPS treated rats than in livers from naive rats, it is possible that there

were not enough neutrophils present to induce injury upon thrombin

171

stimulus. In experiments demonstrating the ability of thrombin to
induce injury in livers from LPS-treated rats, maximal injury occurred in
the presence of neutrophils (Moulin, 1999). At the end of a two hour
recirculating experiment, these livers were found to contain about 41
neutrophils per 20X field (Moulin and Copple, unpublished), which is
four times as many neutrophils in the liver at the end of the experiment
than there were at the beginning of experiments presented here. Though
the larger number of neutrophils may be due to a three-fold larger dose
of LPS than was used in the results presented here, it raises the
possibility that a minimum number of neutrophils was needed to observe
a thrombin-mediated injury and that this minimum number of
neutrophils was not present in the livers for studies presented here.
Furthermore, the potentiation of allyl alcohol hepatotoxicity by LPS in
vivo is dependent on the presence of neutrophils (Sneed, 2000). Thus, it
is possible that the lack of thrombin effect was due to a decrease in the
number of neutrophils. The number of liver-infiltrating neutrophils found
in vivo may be further increased after the administration of allyl alcohol
through added stress and increased Chemokine release. Therefore, it
may be that thrombin-induced liver injury requires a critical number of
infiltrating neutrophils.

Livers from LPS-treated rats were more sensitive to allyl alcohol

hepatotoxicity than livers from saline-treated rats. If not enough

172

neutrophils were present in the livers, then by what mechanism did LPS
enhance the hepatotoxicity of allyl alcohol? Kupffer cells play a critical
role in the potentiation of allyl alcohol by LPS (Sneed ef al, 1997).
Therefore it is possible that LPS-stimulated Kupffer cells released
mediators into the liver that altered the hepatocytes to become more
sensitive to allyl alcohol hepatotoxicity. However, enhanced allyl alcohol
hepatotoxicity was not observed during perfusion of isolated livers from
naive rats with LPS. Accordingly, the enhancement of allyl alcohol
hepatotoxicity in livers from LPS-treated rats is not likely due solely to
the release of Kupffer cell-derived mediators into the perfusion medium.
Therefore, it is possible that there are enough neutrophils present in the
isolated livers from LPS treated rats to develop injury and that LPS
treatment in vivo activates components of the coagulation cascade which
alter the liver to make it more susceptible to allyl alcohol. In this
scenario pretreatment with heparin would inhibit the activation of
coagulation. Therefore, the inability of thrombin to enhance the toxicity
of allyl alcohol is likely due to the absence of other critical factors of the
coagulation cascade or other thrombin-activated factors present in the
blood.

Results presented in this paper indicate that thrombin does not
appear to play a role in the potentiation of allyl alcohol-induced liver

injury by LPS through a liver receptor-mediated mechanism. However,

173

the ability of anticoagulants to attenuate the potentiation indicate that
some component of the, coagulation cascade does play a critical role in
the ability of LPS to enhance allyl alcohol-induced liver injury. The role
of thrombin in the potentiation of allyl alcohol hepatotoxicity by LPS may

be dependent on other coagulation factors or components of the blood.

174

CHAPTER 5

Summary and Conclusions

175

5.1 Summary

Humans are exposed to many different drugs and chemicals from
day to day. Investigators have shown that endotoxin exposure can
enhance the toxicity of varioUs xenobiotics. Because endotoxin is well
known to be a potent stimulator of the inflammatory system, the ability
of endotoxin to enhance the toxicity of xenobiotics may be due to the
stimulated inflammatory system. Humans exposed to a xenobiotic
during a simultaneous inflammatory bout may be at greater risk for
toxicity.

To elucidate the mechanism by which inflammation can enhance
the toxicity of xenobiotics we chose allyl alcohol as a model xenobiotic
and endotoxin to stimulate an inflammatory state. The ability of LPS to
enhance the hepatotoxicity of allyl alcohol has been investigated
previously. Early findings demonstrated that endotoxin acts to enhance
the sensitivity of the liver to the toxicity of allyl alcohol based on the
observation that the lesion remains similar to that observed with a large
dose of allyl alcohol. Furthermore, exposure of isolated hepatocytes to
LPS did not alter the hepatotoxicity of allyl alcohol. This indicated that
some other cell type or cellular mediator is required for LPS to enhance
the toxicity of allyl alcohol. Inhibition of Kupffer cell activity attenuated
LPS enhancement of the toxicity of allyl alcohol. This implicated a

critical role for Kupffer cells in LPS-enhanced allyl alcohol hepatotoxicity.

176

The purpose of the present body of work was to investigate further the
mechanism by which underlying inflammation induced by a small dose
of LPS enhances the hepatotoxicity of allyl alcohol.

In Chapter 2 the critical role of Kupffer cells in the potentiation of
allyl alcohol hepatotoxicity by LPS was investigated. Because the role of
Kupffer cells would most easily be studied in vitro, a Kupffer cell-
hepatocyte coculture system was utiliZed. However, many alterations to
the coculture system failed to replicate the potentiation observed in vivo.
We also discovered that LPS could not potentiate the hepatotoxicity of
allyl alcohol in isolated, buffer-perfused livers from naive rats. This
observation indicated that a Kupffer cell-hepatocyte coculture system
would not be sufficient to study the role of Kupffer cells. Furthermore,
the critical role of Kupffer cells in the potentiation was dependent on
other extrahepatic factors. Thus, work was conducted to determine
which extrahepatic factors play a critical role in the potentiation. .

Because allyl alcohol-induced hepatotoxicity is not known to be
dependent on extrahepatic factors, the mechanism of LPS-induced
hepatotoxicity was reviewed. LPS-induced hepatotoxicity is critically
dependent on neutrophils and thrombin, both of which are extrahepatic
factors (Hewett et al, 1992; Moulin et al, 1996). The work presented in
Chapters 3 and 4 investigated the roles of neutrophils and thrombin,

respectively, in the potentiation of allyl alcohol hepatotoxicity by LPS.

177

The critical role of neutrophils in the potentiation of allyl alcohol
hepatotoxicity by LPS was discovered by Sneed (Sneed, 2000). This led
to the hypothesis that allyl alcohol-induced depletion of glutathione
renders hepatocytes more susceptible to LPS-induced, neutrophil-
derived reactive oxygen species. Experiments presented in Chapter 3
tested this hypothesis. Neutrophil-hepatocyte cocultures were used to
test in vitro whether neutrophil-derived reactive oxygen species could
enhance the hepatotoxicity of allyl alcohol. The results from this set of
experiments demonstrated that neutrophil-derived reactive oxygen
species were capable of enhancing the toxicity of allyl alcohol; however,
the magnitude of effect was very small and cast doubt on the role of
reactive oxygen in this response in the whole animal.

The role of reactive oxygen species in the potentiation of allyl
alcohol-induced liver injury by LPS was assessed in vivo using two
different inhibitors of reactive oxygen species. In one experiment rats
were pretreated with polyethylene-coupled superoxide dismutase and
catalase. This inhibitor would neutralize the reactive oxygen species
formed by activated neutrophils. Though the dose used in the
experiments presented in Chapter 3 caused a 66 percent decrease in
superoxide detection and was similar to the dose used in a previous
study demonstrating protection from vitamin A potentiation of carbon

tetrachloride hepatotoxicity (elSisis et al, 1993), we did not observe any

178

protection from injury in the LPS/allyl alcohol model. These data
suggested that reactive oxygen species do not play a critical role in the
potentiation of allyl alcohol toxicity by LPS. It could be argued, however,
that the dose of polyethylene-coupled superoxide dismutase and
catalase was not large enough to neutralize the generated superoxide
sufficiently. In particular, the small space between the sequestered
neutrophil and the hepatocyte could contain a very high concentration of
superoxide in comparison to the concentration of superoxide dismutase.
Therefore another inhibitor of reactive oxygen species was used.

Apocynin is a drug that inhibits the assembly of the NADPH
oxidase enzyme complex on the surface of cells. It is this enzyme that is
responsible for the production of superoxide anion by activated cells.
Thus, by using apocynin we could prevent reactive oxygen species from
being formed. Using a dose that was reported to protect against
neutrophil-induced respiratory muscle dysfunction during sepsis
(Supinski et al, 1999), we did not find a significant attenuation of liver
injury. These data further demonstrated that superoxide does not play a
critical role in the potentiation of allyl alcohol-induced liver injury by
LPS.

This result has been found in other models of liver injury that
involve neutrophils. It was surprising that allyl alcohol, being an agent

that causes reductive stress in cells, did not become very toxic in the

179

presence neutrophil-derived superoxide anion. However, our laboratory
has found that hepatocellular injury is also dependent on neutrophil.
derived proteases, specifically cathepsin G and elastase, via their ability
to induce hepatocellular injury from activated neutrophils (Ho et al,
1996). This may also be the mechanism by which neutrophils play a
critical role in the potentiation of allyl alcohol hepatotoxicity by LPS.

In Chapter 4 we investigated the potential role of thrombin in the
potentiation of allyl alcohol toxicity by LPS. Thrombin acts through a
receptor-mediated mechanism in a model of large-dose, LPS-induced
liver injury (Moulin, 1999). This led to the hypothesis that thrombin
plays a critical role in the potentiation of allyl alcohol hepatotoxicity by
LPS through a receptor-mediated mechanism. The ability of heparin and
warfarin to inhibit the potentiation of allyl alcohol damage by LPS
indicated that thrombin was playing a role in the potentiation of allyl
alcohol. However, neither heparin nor warfarin is a specific inhibitor of
thrombin. These drugs also inhibit other factors within the coagulation
cascade. To demonstrate the specific effect of thrombin on the liver the
isolated, perfused liver model was used. Addition of thrombin to
isolated, perfused livers from LPS-treated rats did not enhance the
hepatotoxicity of allyl alcohol, indicating that thrombin was not acting
through a liver receptor-mediated mechanism. Thus the role of the

coagulation cascade in the potentiation of allyl alcohol heptotoxicity by

180

LPS could be dependent on other factors of the coagulation cascade or
other factors within the blood that thrombin can activate. Alternatively,
it is possible that too many neutrophils were flushed out of the livers
from LPS-treated rats and that the role of thrombin is dependent on
some critical number of neutrophils in the liver.

The mechanism by which underlying inflammation induced by low
doses of LPS enhances the hepatotoxicity of allyl alcohol remains
unknown. It is clear from this body of work that this is a complex
mechanism which is critically dependent on several different aspects of
inflammation including Kupffer cells, neutrophils, and the coagulation
cascade. It is interesting that while none of these factors is critical for
the hepatotoxicity of allyl alcohol alone, all of these factors play a critical
role in the hepatotoxicity of large doses of LPS. From this point of view
it would appear that allyl alcohol enhances the hepatotoxicity of LPS.
The liver lesion from the potentiation of allyl alcohol and LPS is located
in an area representative of allyl alcohol hepatotoxicity. This could be
due to the toxicity of LPS that is being enhanced in the region of allyl
alcohol-induced stress. In the simplest terms the mechanism of the
potentiation with LPS and allyl alcohol may be due to the combination of
a primed inflammatory system with a hepatocellular stress. The stress
of allyl alcohol toxicity, though not enough to induce cell death, may

induce release of cellular mediators that can trigger the already primed

181

immune system to become active whereas under normal circumstances

the immune system would not be activated.

5.2 Revised Proposed Mechanism

Using information from this thesis and the literature a simplified
mechanism by which LPS enhances the toxicity of allyl alcohol is
proposed. Upon LPS exposure, Kupffer cells are activated to release
many different mediators including prostaglandins, interleukins, TNF,
and reactive oxygen species. The effects of activated Kupffer cells in the
liver include decreased protein synthesis, synthesis of acute phase
proteins which include chemokines, recruitment of neutrophils, and
activation of coagulation through tissue factor expression. The initiation
of coagulation acts as a proinflammatory stimulus that can further
activate and prime Kupffer cells, endothelial cells, and neutrophils. The
proinflammatory actions of coagulation may be due to thrombin.
Thrombin activates endothelial cells to release platelet activating factor,
adhesion molecules, and reactive oxygen species. Thrombin also
enhances the release of mediators from both Kupffer cells and
neutrophils. Though inhibition of thrombin activity does not block
neutrophil infiltration, it does protect against neutrophil-dependent liver
injury (Pearson et al, 1996). Thus, thrombin may play a major role in

the propagation of liver injury.

182

Endothelial cell activation by either LPS, Kupffer cell-derived
mediators, or thrombin can led to expression of adhesion molecules
such as intercellular adhesion molecule-1, vascular cell adhesion
molecule-1, and E-selectin. Neutrophils also begin to express adhesion
molecules and infiltrate into the liver. The liver is now primed to protect
itself from invading bacteria. However, as reviewed by Jaeschke
(Jaeschke, 1997) the presence of neutrophils in the liver does not mean
that they are contributing to the injury. There are many examples in
which neutrophils must accumulate in the liver for several hours before
injury results and still others in which liver injury does not occur despite
a large number of neutrophils in the liver. Only neutrophils that adhere
to hepatocytes appear to be involved in injury. For neutrophils to
extravasate there must be some signal from the hepatocytes to call them
in such as ClNC-l. This may be the point at which allyl alcohol toxicity
may act as the trigger for neutrophil damage. The slight injury or insult
induced by small doses of allyl alcohol may cause the hepatocyte to
release chemokines such as ClNC-l which would call neutrophils into the
extravascular space, allowing direct contact between the hepatocyte and
the neutrophil. This contact with a partially damaged cell may trigger
the neutrophil to release its antimicrobial milieu resulting in damage to
the hepatocyte. In addition the activated neutrophil is capable of

releasing tissue factor both from itself and from the injured hepatocyte.

183

The increased release of tissue factor would serve to activate further the
coagulation cascade. In addition, activation of the coagulation cascade
would prime further the inflammatory response leading to propagation of

the injury (Figure 5.1).

184

 

Figure 5.1 Proposed mechanism for LPS-enhancement of 1111'
alcohol hepatotoxicity.

Cartoon representation of a liver sinusoid exposed to LPS
and allyl alcohol. Allyl alcohol acts directly on hepatocytes and
causes stress that may enhance the production of chemokines.
LPS stimulates Kupffer cells which then release mediators such as
prostaglandins, cytokines, reactive oxygen species, and tissue
factor that stimulate neutrophil infiltration and release of
chemokines from hepatocytes. Chemokine release from
hepatocytes induces neutrophils to extravasate and subsequently
injure the hepatocytes. Release of tissue factor from stimulated
cells activates the coagulation cascade which then exacerbates the
inflammatory response and consequent injury. Arrow thickness
represents an approximate order of events with thin arrows
representing early events and thick arrows representing later

events.

185

Prostaglandins
lnterleukins Tissue Factor I :f‘
TNF 0
01'

pa
8
E.
3
(D
a

 

186

5.3 Importance of Work

Each chapter of this thesis has helped to further our
understanding of the mechanism by which an underlying inflammatory
state can enhance the toxicity of a xenobiotic. In Chapter 2 the
involvement of Kupffer cells in the potentiation of allyl alcohol
hepatotoxicity by LPS was found to be dependent on extrahepatic
factors. Kupffer cells play a critical role in many types of toxic
responses. Most investigators find that inhibition of Kupffer cells in vivo
attenuates injury and do not investigate the role of Kupffer cells in vitro.
Thus, the critical dependence of Kupffer cells on other factors is often
overlooked in the literature. Prevention of Kupffer cell-dependent toxicity
via inhibition of Kupffer cells may not be beneficial or a practical
approach to therapy due to the critical role these cells play in many
other normal physiological functions. Inhibition of the other factors on
which Kupffer cells depend, such as thrombin, may be a more beneficial
target. For this reason, understanding the nature of the factors and the
interdependence of Kupffer cells with them is important.

In Chapter 3 we found that the enhancement of allyl alcohol
toxicity by LPS was not due to reactive oxygen species. For many years
the role of reactive oxygen intermediates in the development of allyl

alcohol-induced liver injury has been debated in the literature. While

187

 

some have shown that protein adduct formation and consequent loss of
protein function is the cause of allyl alcohol-induced liver injury, others
have presented evidence that cell damage is dependent on oxidative
injury. If allyl alcohol-induced cell damage were dependent on oxidative
injury then the addition of reactive oxygen species would be expected to
increase cell damage. Our findings indicating that enhancement of allyl
alcohol hepatotoxicity by LPS is not dependent on the release of reactive
oxygen species from activated inflammatory cells suggest that the
toxicity of allyl alcohol alone is also not dependent on oxidative injury.
Therefore, the data presented here lend support toward the mechanism
of allyl alcohol-induced liver injury being dependent on protein adduct
formation and loss of protein function.

In Chapter 4 we found that the coagulation cascade plays a critical
role in the potentiation of allyl alcohol-induced liver injury by LPS while it
does not play a role in the toxicity of allyl alcohol alone. The ability of
anticoagulants to afford protection could potentially be used to identify
idiosyncratic drug interactions that can be explained by an underlying
inflammatory state. The literature could be searched to compare the
number of idiosyncratic drug interactions that occur with anticoagulants
compared to the number of idiosyncratic drug interactions that occur

with several other drug types. A smaller number of idiosyncratic

188

interactions with anticoagulants may indicate that anticoagulants protect
against underlying inflammation as a cause for adverse interactions.
Work presented here has advanced our current knowledge of the
mechanism by which underlying inflammation can enhance the toxicity
of a xenobiotic agent. Each of the mechanisms found to play a role in
enhanced toxicity has increased our current understanding of liver
toxicology. Although studies in which inhibition of Kupffer cells in vivo
alters a response can implicate these cells in a biological effect, they do
not reveal the complexity of the contribution of these cells. Their
complexity is only partially uncovered when attempting to recreate the
response in an in vitro system. In this model of toxicity Kupffer cells
appear to play a critical role in the development of injury through their
ability to induce neutrophil accumulation and activate the coagulation
cascade. Though Kupffer cells may be involved in the potentiation of
allyl alcohol hepatotoxicity by endotoxin through other mechanisms, this
body of work demonstrates the complex nature of toxic responses.
Furthermore, results presented here have laid a foundation upon which
further investigations can be based to further elucidate the mechanism
by which LPS exposure can enhance the hepatotoxicity of allyl alcohol.
By understanding factors involved in the enhancement of Xenobiotic
toxicity we may be better prepared to predict conditions that could

cause possible enhancement of xenobiotic toxicity and take measures to

189

protect against such enhancements through inhibition of critical steps in

the mechanism.

190

 

BIBLIOGRAPHY

191

 

BIBLIOGRAPHY

Adachi, Y., Bradford, B.U., Gao, W., Bojes, H.K., & Thurman, R.G.
(1994). Inactivation of Kupffer cells prevents early alcohol-induced liver
injury. Hepatology, 20, 453-460.

Adams, J.D.J. & Klaidman, L.K. (1993). Acrolein-induced oxygen radical
formation. Free Radic.8io/ Med, 15, 187-193.

Adding, L.C., Bannenberg, G.L., & Gustafsson, LE. (2001). Basic
experimental studies and clinical aspects of gadolinium salts and
chelates. Cardiovasc Drug Rev, 19, 41-56.

Akashi, S., Ogata, H., Kirikae, F., Kirikae, T., Kawasaki, K., Nishijima, M.,
Shimazu, R., Nagai, Y., Fukudome, K., Kimoto, M., & Miyake, K. (2000).
Regulatory roles for C014 and phosphatidylinositol in the signaling via
toll-like receptor 4-MD-2. Biochem Biophys Res Commun, 268, 172-177.

Altin, J.G. & Bygrave, F.L. (1988). Prostaglandin F28 and the
thromboxane A2 analogue ONO-11113 stimulate Ca2+ fluxes and other
physiological responses in rat liver. Biochem.J., 249, 677-685.

Antal-Szalmas, P. (2000). Evaluation of CD14 in host defence. EurJ Clin
Invest, 30, 167-179.

AntaI-Szalmas, P., Strijp, J.A., Weersink, A.J., Verhoef, J., & Van Kessel,
K.P. (1997). Quantitation of surface CD14 on human monocytes and
neutrophils. J Leukoc Biol, 61, 721-728.

Arditi, M., Zhou, J., Dorio, R., Rong, G.W., Goyert, S.M., & Kim, KS.
(1993). Endotoxin-mediated endothelial cell injury and activation: role of
soluble CDl4. Infect Immun, 61, 3149-3156.

Arthur, M.J., Bentley, I.S., Tanner, A.R., Saunders, P.K., Millward-Sadler,
G.H., & Wright, R. (1985). Oxygen-derived free radicals promote hepatic
injury in the rat. Gastroenterology, 89, 1114-1122.

192

 

Badger, D.A., Kuester, R.K., Sauer, J.M., & Sipes, LG. (1997).
Gadolinium chloride reduces cytochrome P450: relevance to chemical-
induced hepatotoxicity. Toxicology, 121, 143-153.

Badger, D.A., Sauer, J.M., Hoglen, N.C., Jolley, C.S., & Sipes, LG.
(1996). The role of inflammatory cells and cytochrome P450 in the
potentiation of CCl4-induced liver injury by a single dose of retinol.
Toxicol Appl Pharmacol, 141, 507-519.

Badr, M.Z., Belinsky, S.A., Kauffman, F.C., & Thurman, R.G. (1986).
Mechanism of hepatotoxicity to periportal regions of the liver Iobule due
to allyl alcohol: Role of oxygen and lipid peroxidation. Journal of
Pharmacology and Experimental Therapeutics, 238, 1138-1142.

Balsa, D., Merlos, M., Giral, M., Ferrando, R., Garcia-Rafanell, J., & Forn,
J. (1997). Effect of endotoxin and platelet-activating factor on rat
vascular permeability: role of vasoactive mediators. J Lipid Mediat Cell
Signal, 17, 31-45.

 

Bar-Shavit, R., Benezra, M., Sabbah, V., DeJana, E., Vlodavsky, l., &
Wilner, GD. (1992). functional domains in thrombin outside the catalytic
site. In L.J. Berliner (Ed.), thrombin: structure and function (pp. 315New
York: Plenum Press.

Barton, 00., Barton, E.X., Ganey, P.E., Kunkel, S.L., & Roth, RA.
(2000). Bacterial lipopolysacchardie enhances aflatoxin Bl
hepatotoxicity in rats by a mechanism that depends on tumor necrosis
factor-alpha. Hepatology,

Barton, 00, Ganey, P.E., & Roth, RA (2000). Lipopolysaccharide
Augments Aflatoxin B(1)-lnduced Liver Injury through Neutrophil-
Dependent and -lndependent Mechanisms. Toxicol Sci, 58, 208-215.

Barton, 00, Hill, D.A., Yee, S.B., Barton, E.X., Ganey, P.E., & Roth, RA
(2000). Bacterial Iipopolysaccharide exposure augments aflatoxin B(1)-
induced liver injury. Toxicol Sci, 55, 444-452.

 

193

Bautista, A.P., Meszaros, K., Bojta, J., & Spitzer, J.J. (1990). Superoxide
anion generation in the liver during the early stage of endotoxemia in
rats. J Leukoc Biol, 48, 123-128.

Bautista, A.P. & Spitzer, J.J. (1999). Role of Kupffer cells in the ethanol-
induced oxidative stress in the liver. Front Biosci, 4, D589-D595

Bazil, V., Baudys, M., Hilgert, l., Stefanova, l., Low, M.G., Zbrozek, J., &
Horejsi, V. (1989). Structural relationship between the soluble and

membrane-bound forms of human monocyte surface glycoprotein CD14. E
Mol Immunol, 26, 657-662.

Beamer, L.J., Carroll, S.F., & Eisenberg, D. (1997). Crystal structure of
human BPI and two bound phospholipids at 2.4 angstrom resolution.
Science, 276, 1861-1864.

 

Belinsky, S.A., Matsumura, T., Kauffman, F.C., & Thurman, R.G. (1983).
Rates of allyl alcohol metabolism in periportal and pericentral regions of
the liver Iobule. Molecular Pharmacology, 25, 158-164.

Beltran, A.E., Concepcion, F., Manzanares, D., Garrido, G., Glaria, L.A., &
Rojas, A. (1999). Heparin and low molecular weight heparin decrease
nitric oxide production by human polymorphonuclear cells. Arch Med
Res, 30, 116-119.

Beutler, B. (2000). Endotoxin, toll-like receptor 4, and the afferent limb
of innate immunity. Curr Opin Microbiol, 3, 23-28.

Billiar, T.R., Lysz, T.W., Curran, R.D., Bentz, B.G., Machiedo, G.W., &
Simmons, R.L. (1990). Hepatocyte modulation of Kupffer cell
prostaglandin E2 production in vitro. Journal of Leukocyte Biology, 47,
304-31 1.

Bizios, R., Lai, L., Fenton, J.W., ll, & Malik, AB. (1986). Thrombin-
induced chemotaxis and aggregation of neutrophils. Journal of Cellular
Physiology, 128, 485-490.

 

194

 

Bode, C., Kugler, V., & Bode, J.C. (1987). Endotoxemia in patients with
alcoholic and non-alcoholic cirrhosis and in subjects with no evidence of
chronic liver disease following acute alcohol excess. J Hepatol, 4, 8-14.

Brightbill, H.D., Libraty, D.H., Krutzik, S.R., Yang, R.B., Belisle, J.T.,
Bleharski, J.R., Maitland, M., Norgard, M.V., Plevy, S.E., Smale, S.T.,
Brennan, P.J., Bloom, B.R., Godowski, P.J., & Modlin, R.L. (1999). Host
defense mechanisms triggered by microbial lipoproteins through toll-like
receptors. Science, 285, 732-736.

Brown, A.P., Harkema, J.R., Schultze, A.E., Roth, R.A., & Ganey, PE.
(1997). Gadolinium chloride pretreatment protects against hepatic injury
but predisposes the lungs to alveolitis after Iipopolysaccharide
administration. Shock, 7, 186-192.

Calvano, S.E., Thompson, W.A., Marra, M.N., Coyle, S.M., de Riesthal,
H.F., Trousdale, R.K., Barie, P.S., Scott, R.W., Moldawer, L.L., & Lowry,
SF. (1994). Changes in polymorphonuclear leukocyte surface and
plasma bactericidal/permeability-increasing protein and plasma
Iipopolysaccharide binding protein during endotoxemia or sepsis. Arch
Surg, 129, 220-226.

Camara, D.S., Caruana, J.A.J., Schwartz, K.A., Montes, M., & Nolan, JP.
(1983). D-Galactosamine liver injury: absorption of endotoxin and

protective effect of small bowel and colon resection in rabbits. Proc Soc
Exp Biol Med, 172, 255-259.

Camara, D.S., Caruana, J.A., Jr., Schwartz, K.A., Montes, M., & NOlan,
JP. (1983). D-Galactosamine liver injury: absorption of endotoxin and
protective effect of small bowel and colon resection in rabbits.
Proceedings of the Society for Experimental Biology and Medicine, 172,
255-259.

Camus, P., Joshi, U.M., Lockard, V.G., Petrini, M.F., Lentz, D.L., &
Mehendale, HM. (1989). Effects of drug-induced pulmonary
phospholipidosis on lung mechanics in rats. J Appl Physiol, 66, 2437-
2445.

195

 

Canty, T.G.J., Boyle, E.M.J., Farr, A., Morgan, E.N., Verrier, E.D., &
Pohlman, T.H. (1999). Oxidative stress induces NF-kappaB nuclear
translocation without degradation of lkappaBalpha. Circulation, 100,
ll361-ll364

Cao, Z., Xiong, J., Takeuchi, M., Kurama, T., & Goeddel, D.V. (1996).
TRAF6 is a signal transducer for interleukin-1. Nature, 383, 443-446.

Carbonell, L.F., Nadal, J.A., Llanos, M.C., Hernandez, l., Nava, E., &
Diaz, J. (2000). Depletion of liver glutathione potentiates the oxidative
stress and decreases nitric oxide synthesis in a rat endotoxin shock
model. Crit Care Med, 28, 2002-2006.

Carney, DH. (1992). postclotting cellular effects of thrombin mediated
by interaction with high-affinity thrombin receptors. In L.J. Berliner (Ed.),
Thrombin: structure and function (pp. 351New York: plenum Press.

Carpanini, F.M.B., Gaunt, I.F., Hardy, J., Gangolli, S.D., Butterworth,
K.R., & Lloyd, AG. (1978). Short-term toxicity of allyl alcohol in rats.
Toxicology, 9, 29-45.

Cebula, T.A., El-Hage, A.N., & Ferrans, V.J. (1984). Toxic interactions of
benzyl alcohol with bacterial endotoxin. Interaction and Immunity, 44(1),
91-96.

Chan, K.L., Bizios, R., & Malik, AB. (1988). Thrombin enhances
opsonized zymosan-induced chemiluminescence of neutrophils. Tissue
Cell, 20, 13-17.

Chaudhary, P.M., Ferguson, 0., Nguyen, V., Nguyen, 0., Massa, H.F.,
Eby, M., Jasmin, A., Trask, B.J., Hood, L., & Nelson, PS. (1998). Cloning
and characterization of two Toll/ Interleukin-1 receptor-like genes TIL3
and TIL4: evidence for a multi-gene receptor family in humans. Blood,
91, 4020-4027.

Chi, L., Li, Y., Stehno-Bittel, L., Gao, J., Morrison, D.C., Stechschulte,
D.J., & Dileepan, K.N. (2001). Interleukin-6 production by endothelial
cells via stimulation of protease-activated receptors is amplified by

196

endotoxin and tumor necrosis factor-alpha. J Interferon Cytokine Res, 21,
231-240.

Chow, J.C., Young, D.W., Golenbock, D.T., Christ, W.J., & Gusovsky, F.
(1999). Toll-like receptor-4 mediates Iipopolysaccharide-induced signal
transduction. JBiol Chem, 274, 10689-10692.

Comporti, M., Maellaro, E., Del Bello, B., & Casini, AF. (1991).
Glutathione depletion: Its effects on other antioxidant systems and
hepatocellular damage. Xenobiotica, 21, 1067-1076.

Coughlan, A.F., Hau, H., Dunlop, L.C., Berndt, M.C., & Hancock, WW.
(1994). P-selectin and platelet-activating factor mediate initial
endotoxin-induced neutropenia. J. Exp. Med, 179, 329-334.

Couturier, C., Haeffner-Cavaillon, N., Caroff, M., & Kazatchkine, MD.
(1991). Binding sites for endotoxins (Iipopolysaccharides) on human
monocytes. Jlmmunol, 147, 1899-1904.

Cunningham, M.D., Seachord, C., Ratcliffe, K., Bainbridge, B., Aruffo, A.,
& Darveau, RP. (1996). Helicobacter pylori and Porphyromonas
gingivalis Iipopolysaccharides are poorly transferred to recombinant
soluble CD14. Infect Immun, 64, 3601-3608.

Cunningham, M.D., Shapiro, R.A., Seachord, C., Ratcliffe, K., Cassiano,
L., & Darveau, RP. (2000). CD14 employs hydrophilic regions to
"capture" Iipopolysaccharides. J Immunol, 164, 3255-3263.

Czaja, M.J., Xu, J., Ju, Y., Alt, E., & Schmiedeberg, P. (1994).
Lipopolysaccharide-neutralizing antibody reduces hepatocyte injury from
acute hepatotoxin administration. Hepatology, 19, 1282-1289.

Dahm, L.J., Schultze, A.E., & Roth, RA. (1991). Activated neutrophils
injure the isolated, perfused rat liver by an oxygen radical-dependent
mechanism. American Journal of Pathology, 139(5), 1009-1020.

197

 

 

 

Darien, B.J., Fareed, J., Centgraf, K.S., Hart, A.P., MacWilliams, P.S.,
Clayton, M.K., Wolf, H., & Kruse-Elliott, K.T. (1998). Low molecular
weight heparin prevents the pulmonary hemodynamic and
pathomorphologic effects of endotoxin in a porcine acute lung injury
model. Shock, 9, 274-281.

Darveau, RP. (1998). Lipid A diversity and the innate host response to
bacterial infection. Curr Opin Microbiol, 1, 36-42.

Daun, J.M. & Fenton, M.J. (2000). lnterleukin-l/Toll receptor family
members: receptor structure and signal transduction pathways. J
Interferon Cytokine Res, 20, 843-855.

Day, J.R., Albers, J.J., Lofton-Day, C.E., Gilbert, T.L., Ching, A.F., Grant,
F.J., O'Hara, P.J., Marcovina, S.M., & Adolphson, J.L. (1994). Complete
cDNA encoding human phospholipid transfer protein from human
endothelial cells. J Biol Chem, 269, 9388-9391.

de Jonge, E., Dekkers, P.E., Creasey, A.A., Hack, C.E., Paulson, S.K.,
Karim, A., Kesecioglu, J., Levi, M., van Deventer, S.J., & van Der, P.
(2000). Tissue factor pathway inhibitor dose-dependently inhibits
coagulation activation without influencing the fibrinolytic and cytokine
response during human endotoxemia. Blood, 95, 1124-1129.

Decker, K. (1997). Bacterial Endotoxin. In R.S. McCuskey & D.L. Earnest
(Eds.), Hepatic and Gastrointestinal Toxicology (pp. 433New York: Elsevier
Science Inc.

Deitch, EA. (1994). Bacterial translocation: the influence of dietary
variables. Gut, 35, S23-S27

Detmers, P.A., Zhou, 0., Powell, D., Lichenstein, H., Kelley, M., &
Pironkova, R. (1995). Endotoxin receptors (CD14) are found with CD16
(Fc gamma Rlll) in an intracellular compartment of neutrophils that
contains alkaline phosphatase. J Immunol, 155, 2085-2095.

Dogterom, P., Mulder, G.J., & Nagelkerke, J.F. (1989). Lipid
peroxidation-dependent and -independent protein thiol modifications in

198

 

isolated rat hepatocytes: differential effects of vitamin E and disulfiram.
Chemico-Biological Interactions, 71, 291-306.

Dogterom, P., Nagelkerke, J.F., Van Steveninck, J., & Mulder, G.J.
(1988). Inhibition of lipid peroxidation by disulfiram and
diethyldithiocarbamate does not pervent hepatotoxin-induced cell death
in isolated rat hepatocytes. a study with allyl alcohol, tert-butyl
hydroperoxide, diethyl maleate, bromoisovalerylurea and carbon
tetrachloride. Chemico-Biological Interactions, 66, 251-265.

Downing, L.J., Strieter, R.M., Kadell, A.M., Wilke, C.A., Greenfield, L.J., &
Wakefield, T.W. (1998). Low-dose low-molecular-weight heparin is anti-
inflammatory during venous thrombosis. J Vasc Surg, 28, 848-854.

ElSisi, A.E.D., Earnest, D.L., 8. Sipes, l.G. (1993a). Vitamin A
potentiation of carbon tetrachloride hepatotoxicity: enhanced lipid
peroxidation without enhanced biotransformation. Toxicology and Applied
Pharmacology, 119, 289-294.

ElSisi, A.E.D., Earnest, D.L., & Sipes, I.G. (1993b). Vitamin A
potentiation of carbon tetrachloride hepatotoxicity: role of liver
macrophages and active oxygen species. Toxicology and Applied
Pharmacology, 119, 295-301.

Engels, F., Renirie, B.F., Hart, B.A., Labadie, R.P., & Nijkamp, F.P.
(1992). Effects of apocynin, a drug isolated from the roots of Picrorhiza
kurroa, on arachidonic acid metabolism. FEBS Lett, 305, 254-256.

Fariss, M.W. & Reed, DJ. (1987). High-performance liquid
chromatography of thiols and disulfides: dinitrophenol derivatives;
Methods Enzymol., 143:101-9, 101-109.

Feng, L., Xia, Y., Garcia, G.E., Hwang, D., & Wilson, 08. (1995).
Involvement of reactive oxygen intermediates in cyclooxygenase-2
expression induced by interleukin-1, tumor necrosis factor-alpha, and
Iipopolysaccharide. JCIin Invest, 95, 1669-1675.

199

 

!,_-.

 

 

Fenton, M.J. & Golenbock, D.T. (1998). LPS-binding proteins and
receptors. J Leukoc Biol, 64, 25-32.

Ferrero, E. & Goyert, SM. (1988). Nucleotide sequence of the gene
encoding the monocyte differentiation antigen, CD14. Nucleic Acids Res, ’
16, 4173

Flo, T.H., Halaas, 0., Lien, E., Ryan, L., Teti, G., Golenbock, D.T.,
Sundan, A., & Espevik, T. (2000). Human toll-like receptor 2 mediates
monocyte activation by Listeria monocytogenes, but not by group B
streptococci or Iipopolysaccharide. J Immunol, 164, 2064-2069.

Formal, S.B., Noyes, H.E., & Schneider, H. (1960). Experimental Shigella
infections. lll. Sensitivity of normal, starved and carbon tetrachloride
treated guinea pigs to endotoxin. Proceedings of the Society for
Experimental Biology and Medicine, 103, 415-418.

Fox, E.S., Broitman, S.A., & Thomas, P. (1990). Bacterial endotoxins and
the liver. Lab. Invest, 63(6) , 733-739.

Franco, R.F., de Jonge, E., Dekkers, P.E., Timmerman, J.J., Spek, C.A.,
van Deventer, S.J., van Deursen, P., van Kerkhoff, L., van Gemen, B., ten
Cate, H., van der Poll, T., & Reitsma, PH. (2000). The in vivo kinetics of
tissue factor messenger RNA expression during human endotoxemia:
relationship with activation of coagulation. Blood, 96, 554-559.

Frey, E.A., Miller, D.S., Jahr, T.G., Sundan, A., Bazil, V., Espevik, T.,
Finlay, B.B., & Wright, SD. (1992). Soluble CD14 participates in the
response of cells to Iipopolysaccharide. J Exp Med, 176, 1665-1671.

Fujita, S., Aril, S., Monden, K., Adachi, Y., Funaki, N., Higashitsuji, H.,
Furutani, M., Mise, M., lshiguro, S., & Kitao, T. (1995). Participation of
hepatic macrophages and plasma factors in endotoxin-induced liver
injury. J.Surg.Res., 59, 263-270.

Galanos, C., Freudenberg, M.A., & Reutter, W. (1979). Galactosamine-
induced sensitization to the lethal effects of endotoxin. Proc Natl Acad
Sci U SA, 76, 5939-5943.

200

 

Gallay, P., Heumann, D., Le Roy, D., Barras, C., & Glauser, MP. (1993).
Lipopolysaccharide-binding protein as a major plasma protein
responsible for endotoxemic shock. Proc Natl Acad Sci U SA, 90, 9935-
9938.

Gallay, P., Heumann, D., Le Roy, D., Barras, C., & Glauser, MP. (1994).
Mode of action of anti-lipopolysaccharide-binding protein antibodies for
prevention of endotoxemic shock in mice. Proc Natl Acad Sci U SA, 91,
7922-7926.

Ganey, P.E., Bailie, M.B., & Roth, RA. (1993). Modulation of Kupffer cell
function with gadolinium chloride does not alter methylene dianiline or
allyl alcohol hepatotoxicity. The Toxicologist, 13, 429(Abstract)

Ganey, P.E., Bailie, M.B., VanCise, S., Colligan, M.E., Madhukar, B.V.,
Robinson, J.P., & Roth, RA. (1994). Activated neutrophils from rat
injured isolated hepatocytes. Lab. Invest, 70(1), 53-60.

Ganey, P.E., Barton, Y.W., Kinser, S., Sneed, R.A., Barton, 00., & Roth,
RA. (2001). Involvement of cyclooxygenase-2 in the potentiation of allyl
alcohol-induced liver injury by bacterial Iipopolysaccharide. ToxicolAppI
Pharmacol, 174, 113-121.

Ganey, P.E. & Schultze, A.E. (1995). Depletion of neutrophils and
modulation of Kupffer cell function in allyl alcohol-induced
hepatotoxicity. Toxicology, 99, 99-106.

Gay, N.J. & Keith, F.J. (1991). Drosophila Toll and lL-1 receptor. Nature,
351. 355-356.

Golenbock, D.T., Liu, Y., Millham, F.H., Freeman, M.W., & Zoeller, RA.
(1993). Surface expression of human CD14 in Chinese hamster ovary
fibroblasts imparts macrophage-like responsiveness to bacterial
endotoxin. J Biol Chem, 268, 22055-22059.

Goyert, S.M., Ferrero, E.M., Seremetis, S.V., Winchester, R.J., Silver, J.,
& Mattison, AC. (1986). Biochemistry and expression of
myelomonocytic antigens. J Immunol, 137, 3909-3914.

201

 

 

Gray, P.W., Flaggs, G., Leong, S.R., Gumina, R.J., Weiss, J., Ooi, C.E., &
Elsbach, P. (1989). Cloning of the cDNA of a human neutrophil
bactericidal protein. Structural and functional correlations. J Biol Chem,
264, 9505-9509.

Grossman, N. & Leive, L. (1984). Complement activation via the
alternative pathway by purified Salmonella Iipopolysaccharide is affected
by its structure but not its O-antigen length. J Immunol, 132, 376-385.

Guigui, B., Rosenbaum, J., Preaux, A.-M., Martin, N., Zafrani, E.S.,
Dhumeaux, D., & Mavier, P. (1988). Toxicity of phorbol myristate
acetate-stimulated polymorphonuclear neutrophils against rat
hepatocytes. Demonstration and mechanism. Lab. Invest, 59(6), 831-
837.

Gumerlock, P.H., Edwards, B.F., Deitch, A.D., & Meyers, F.J. (1988).
Human renal clear cell carcinoma: establishment and characterization of
a new cell line (G-2101). In Vitro Cellular and Developmental Biology,
24(5), 429-434.

Haeffner-Cavaillon, N., Caroff, M., & Cavaillon, J.M. (1989). Interleukin-1
induction by Iipopolysaccharides: structural requirements of the 3-deoxy-
D-manno-2-octulosonic acid (KDO). Mol Immunol, 26, 485-494.

Haeffner-Cavaillon, N., Carreno, M.P., Aussel, L., & Caroff, M. (1999).
Molecular aspects of endotoxins relevant to their biological functions.
Nephrol Dial Transplant, 14, 853-860.

Hailman, E., Lichenstein, H.S., Wurfel, M.M., Miller, D.S., Johnson, D.A.,
Kelley, M., Busse, L.A., Zukowski, M.M., & Wright, S.D. (1994a).
Lipopolysaccharide (LPS)-binding protein accelerates the binding of LPS
to CD14. JExp Med, 179, 269-277.

Hailman, E., Lichenstein, H.S., Wurfel, M.M., Miller, D.S., Johnson, D.A.,
Kelley, M., Busse, L.A., Zukowski, M.M., & Wright, S.D. (1994b).
Lipopolysaccharide (LPS)-binding protein accelerates the binding of LPS
to CD14. JExp Med, 179, 269-277.

202

Harbrecht, B.G., Billiar, T.R., Curran, R.D., Stadler, J., & Simmons, R.L.
(1993). Hepatocyte injury by activated neutrophils in vitro is mediated by
proteases. Annals of Surgery, 218(2), 120-128.

Haziot, A., Chen, S., Ferrero, E., Low, M.G.,-Silber, R., & Goyert, SM.
(1988). The monocyte differentiation antigen, CD14, is anchored to the
cell membrane by a phosphatidylinositol linkage. J Immunol, 141, 547-
552.

Haziot, A., Ferrero, E., Kontgen, F., Hijiya, N., Yamamoto, S., Silver, J.,
Stewart, C.L., & Goyert, SM. (1996). Resistance to endotoxin shock and
reduced dissemination of gram-negative bacteria in CD14-deficient mice.
Immunity, 4, 407-414.

Haziot, A., Ferrero, E., Lin, X.Y., Stewart, C.L., & Goyert, SM. (1995).
CD14-deficient mice are exquisitely insensitive to the effects of LPS.
Prog Clin Biol Res, 392, 349-351.

Haziot, A., Rong, G.W., Silver, J., & Goyert, SM. (1993). Recombinant
soluble CD14 mediates the activation of endothelial cells by
Iipopolysaccharide. Jlmmunol, 151, 1500-1507.

Heemann, U., Szabo, A., Hamar, P., Muller, V., Witzke, 0., Lutz, J., &
Philipp, T. (2000). Lipopolysaccharide pretreatment protects from renal
ischemia/reperfusion injury : possible connection to an interleukin-6-
dependent pathway. Am J Pathol, 156, 287-293.

Heine, H., Kirschning, C.J., Lien, E., Monks, B.G., Rothe, M., &
Golenbock, D.T. (1999). Cutting edge: cells that carry A null allele for
toll-like receptor 2 are capable of responding to endotoxin. J Immunol,
162, 6971-6975.

Hewett, J.A. & Roth, RA (1988). Dieldrin activates rat neutrophils in
vitro. Toxicology and Applied Pharmacology, 96, 269-278. *

Hewett, J.A. & Roth, RA. (1993). Hepatic and extrahepatic pathobiology
of bacterial Iipopolysaccharides. Pharmacol Rev, 45, 382-411.

203

Hewett, J.A. & Roth, RA. (1995). The coagulation system, but not
circulating fibrinogen, contributes to liver injury in rats exposed to
Iipopolysaccharide from gram-negative bacteria. J Pharmacol Exp Ther,
272, 53-62.

Hewett, J.A., Schultze, A.E., VanCise, S., & Roth, RA. (1992). Neutrophil
depletion protects against liver injury from bacterial endotoxin. Lab
Invest, 66, 347-361.

Hisama, N., Yamaguchi, Y., lshiko, T., Miyanari, N., lchiguchi, 0., Goto,
M., Mari, K., Watanabe, K., Kawamura, K., Tsurufuji, S., & Ogawa, M.
(1996). Kupffer cell production of cytokine-induced neutrophil
chemoattractant following ischemia/reperfusion injury in rats.
Hepatology, 24, 1 193-1198.

Hitchcock, P.J., Leive, L., Makela, P.H., Rietschel, E.T., Strittmatter, W.,
& Morrison, DC. (1986). Lipopolysaccharide nomenclature-past,
present, and future. J Bacteriol, 166, 699-705.

Ho, J.S., Buchweitz, J.P., Roth, R.A., & Ganey, PE. (1996). Identification
of factors from rat neutrophils responsible for cytotoxicity to isolated
hepatocytes. J Leukoc Biol, 59, 716-724.

Hoffman, M. & Cooper, ST. (1995). Thrombin enhances monocyte .
secretion of tumor necrosis factor and interleukin-1 beta by two distinct
mechanisms. Blood Cells Mal Dis, 21, 156-167.

Hoffmann, J.N., Vollmar, B., Inthorn, D., Schildberg, F.W., & Menger,
MD. (2000). The thrombin antagonist hirudin fails to inhibit endotoxin-
induced leukocyte/endothelial cell interaction and microvascular
perfusion failure. Shock, 14, 528-534.

Hoglen, N.C., Younis, H.S., Hartley, D.P., Gunawardhana, L., Lantz, R.C.,
& Sipes, LG. (1998). 1,2-Dichlorobenzene-induced lipid peroxidation in
male Fischer 344 rats is Kupffer cell dependent. Toxicol Sci , 46, 376-
385.

204

Holland, J.A., Meyer, J.W., Chang, M.M., O'Donnell, R.W., Johnson, D.K.,
& Ziegler, L.M. (1998). Thrombin stimulated reactive oxygen species
production in cultured human endothelial cells. Endathelium, 6, 113-
121.

Ihara, l., Harada, Y., Ihara, S., & Kawakami, M. (1982). A new
complement-dependent bactericidal factor found in nonimmune mouse
sera: specific binding to polysaccharide of Ra chemotype Salmonella. J
Immunol, 128, 1256-1260.

limuro, Y., Yamamoto, M., Kahno, H., ltakura, J., Fujii, H., & Matsumoto,
Y. (1994). Blockade of liver macrophages by gadolinium chloride
reduces lethality in endotoxemic rats-analysis of mechanisms of lethality
in endotoxemia. Journal of Leukocyte Biology, 55, 723-728.

lkejima, K., Enamoto, N., Seabra, V., lkejima, A., Brenner, D.A., &
Thurman, R.G. (1999). Pronase destroys the Iipopolysaccharide receptor
CD14 on Kupffer cells. Am J Physiol, 276, G591-G598

lshiyama, H., Ogino, K., & Hobara, T. (1995). Role of Kupffer cells in rat
liver injury induced by diethyldithiocarbamate. Eur.J Pharmacol, 292,
135-141.

Jaeschke, H. (1997). Cellular adhesion molecules: regulation and
functional significance in the pathogenesis of liver diseases. Am J
Physiol, 273, G602-G61 1

Jaeschke, H., Bautista, A.P., Spolarics, Z., & Spitzer, J.J. (1992).
Superoxide generation by neutrophils and Kupffer cells during in vivo
reperfusion after hepatic ischemia in rats. Journal of Leukocyte Biology,
52, 377-382.

Jaeschke, H., Kleinwaechter, C., & Wendel, A. (1987). The role of
acrolein in allyl alcohol-induced lipid peroxidation and liver cell damage
in mice. Biachem.PharmacaI., 36(1), 51-57.

Jann, K. & Jann, B. (1984). Structure and biosynthesis. of O-antigens.
Chemistry of Endotoxin, 1, 138.

205

Jean, P.A., Bailie, M.B., & Roth, RA. (1995). 1-Naphthylisothiocyanate-
induced elevation of biliary glutathione. Biachem.PharmacaI., 49, 197-
202.

Jirillo, E., De Simone, 0., Covelli, V., Kiyono, H., McGhee, J.R., &
Antonaci, S. (1990). LPS-mediated triggering of T lymphocytes in

immune response against gram-negative bacteria. Adv Exp Med Biol,
256, 417-425.

Juan, T.S., Kelley, M.J., Johnson, D.A., Busse, L.A., Hailman, E., Wright,
S.D., & Lichenstein, HS (1995). Soluble CD14 truncated at amino acid
152 binds Iipopolysaccharide (LPS) and enables cellular response to
LPS. JBioI Chem, 270, 1382-1387. '

Kausalya, S., Malla, N., Ganguly, N.K., & Mahajan, RC. (1993).
Leishmania donovani: in vitro evidence of hepatocyte damage by Kupffer
cells and immigrant macrophages in a murine model. Exp Parasitol, 77,
326-333.

Kawai, T., Adachi, 0., Ogawa, T., Takeda, K., & Akira, S. (1999).
Unresponsiveness of MyD88-deficient mice to endotoxin. Immunity, 11,
115-122.

Kaye, OM. (1973). Biosynthesis of mercapturic acids from allyl alcohol,
allyl esters and acrolein. Biochem.J., 134, 1093-1101.

Kim, A., Gut, J.P., & Gendrault, J.L. (1982). Interaction of viruses with
sinusoidal cells. Prog Liver Dis, 7, 377-392.

Kirschning, C.J., Wesche, H., Merrill, A.T., & Rothe, M. (1998). Human
toll-like receptor 2 confers responsiveness to bacterial
Iipopolysaccharide. J Exp Med, 188, 2091-2097.

Kitchens, R.L. & Munford, RS. (1995). Enzymatically deacylated
lipopolysaccharide (LPS) can antagonize LPS at multiple sites in the LPS
recognition pathway. J Biol Chem, 270, 9904-9910.

206

Kitchens, R.L., Ulevitch, R.J., & Munford, RS. (1992).
Lipopolysaccharide (LPS) partial structures inhibit responses to LPS in a
human macrophage cell line without inhibiting LPS uptake by a CD14-
mediated pathway. J Exp Med, 176, 485-494.

Knook, D.L. & Sleyster, EC. (1977). Preparation and characterization of
kupffer cells from rat and mouse liver. pp. 273-88. In: Wisse.E.,

Knook. DL., ed. Kupffer. cells and other liver sinusoidal.cells.Amsterdam.,
Elsevier. / North-Holland. Biomedical, :

Kopp, E., Medzhitov, R., Carothers, J., Xiao, 0., Douglas, |., Janeway,
C.A., & Ghosh, S. (1999). ECSIT is an evolutionarily conserved
intermediate in the Toll/lL-1 signal transduction pathway. Genes Dev,
13, 2059-2071.

Kruger, C., Schutt, C., Obertacke, U., Joka, T., Muller, F.E., Knoller, J.,
Koller, M., Konig, W., & Schonfeld, W. (1991). Serum CD14 levels in
polytraumatized and severely burned patients. Clin Exp Immunol , 85,
297-301. '

Kulikov, V.l. & Muzya, G.l. (1998). The bioregulatory role of platelet-
activating factor in intracellular processes and cell-cell interactions.
Biochemistry (Mosc), 63, 47-54.

Lapperre, T.S., Jimenez, L.A., Antonicelli, F., Drost, E.M., Hiemstra, P.S.,
Stalk, J., MacNee, W., & Rahman, l. (1999). Apocynin increases
glutathione synthesis and activates AP-l in alveolar epithelial cells. FEBS
Lett, 443, 235-239.

Lee, J.D., Kato, K., Tobias, P.S., Kirkland, T.N., & Ulevitch, R.J. (1992).,
Transfection of CD14 into 702/3 cells dramatically enhances the
sensitivity to complexes of lipopolysaccharide (LPS) and LPS binding
protein. J Exp Med, 175, 1697-1705.

Lehman, J.D., Dyke, C., Abd-Elfattah, A., Yeh, T.J., Ding, M., Ezrin, A., &
Wechsler, AS. (1992). Polyethylene glycol-conjugated superoxide
dismutase attenuates reperfusion injury when administered twenty-four
hours before ischemia. J Thorac Cardiovasc Surg, 104, 1597-1601.

207

Lehmann, V., Streck, H., Minner, l., Krammer, P.H., & Ruschmann, E.
(1980). Selection of bacterial mutants from Salmonella specifically
recognizing determinants on the cell surface of activated T lymphocytes.
A novel system to define cell surface structures. EurJ Immunol, 10, 685-
693.

Lichtman, S.N., Wang, J., Zhang, C., & Lemasters, J.J. (1996).
Endocytosis and Ca2+ are required for endotoxin-stimulated TNF-alpha
release by rat Kupffer cells. Am J Physiol, 271, G920-G928

Lind, R.C., Gandolfi, A.J., Sipes, l.G., & Brown, BR. (1984). The
involvement of endotoxin in halothane-associated liver injury.
Anesthesiology, 61, 544-550.

Lindberg, A. A., Wollin, R., Bruse, G., Ekwall, E., Svenson, SB. (1983).
Immunology and immunochemistry of synthetic and semisynthetic
Salmonella O-antigen specific glycoconjugates, ACS Symp. Ser., 231, 83.

Liu, J., Sendelbach, L.E., Parkinson, A., & Klaassen, CD. (2000).
Endotoxin pretreatment protects against the hepatotoxicity of
acetaminophen and carbon tetrachloride: role of cytochrome P450
suppression. Toxicology, 147, 167-176.

Liu, P., McGuire, G.M., Fisher, M.A., Farhood, A., Smith, C.W., &
Jaeschke, H. (1995). Activation of Kupffer cells and neutrophils for
reactive oxygen formation is responsible for endotoxin-enhanced liver
injury after hepatic ischemia. Shock, 3, 56-62.

Liu, T.H., Beckman, J.S., Freeman, B.A., Hogan, E.L., & Hsu, CY.
(1989). Polyethylene glycol-conjugated superoxide dismutase and
catalase reduce ischemic brain injury. Am J Physiol, 256, H589-H593

Lo, C.J., Cryer, H.G., Fu, M., & Kim, B. (1997). Endotoxin-induced
macrophage gene expression depends on platelet-activating factor. Arch
Surg, 132, 1342-1347.

Lorant, D.E., Patel, K.D., McIntyre, T.M., McEver, R.P., Prescott, S.M., &
Zimmerman, GA (1991). Coexpression of GMP-l40 and PAF by

208

endothelium stimulated by histamine or thrombin: a juxtacrine system
for adhesion and activation of neutrophils. J Cell Biol, 115, 223-234.

Luderitz, 0., Freudenberg, M. A., Galanose, C., Lehman, V., Rietschel, E.
Th., & Shaw, D. H. (1982). Lipopolysaccarides of Gram-negative
bacteria. Microbial Membrane Lipids, 17, 79.

Luderitz, Th., Brandenburg, K., Seydel, U., Roth, A., Galanos, C., &
Rietschel, E. Th. (1989). Sturctural and physicochemical requirements of
endotoxins for the activation of arachidonic acid metabolism in mouse
peritoneal macrophages in vitro. EurJ Biochem, 179, 11.

Lynn, W.A., Raetz, C.R., Qureshi, N., & Golenbock, D.T. (1991).
Lipopolysaccharideinduced stimulation of CD11b/CD18 expression on
neutrophils. Evidence of specific receptor-based response and inhibition
by lipid A-based antagonists. J Immunol, 147, 3072-3079.

Lysz, T.W., Billiar, T.R., Curran, R.D., Simmons, R.L., & Machiedo, G.W.
(1990). Kupffer cell-hepatocyte interactions and the changes in 1-14C-
arachidonate incorporation in response to endotoxin in yitro.
Prostaglandins, 39(5), 497-514.

Makela, P.H., Hovi, M., Sazen, H., Muotiala, A., & Rhen, M. (1990). Role
of LPS in the pathogenesis of salmonellosis. Cellular and Molecular
Aspects of Endotoxin Reactions, 1, 537.

Marchant, A., Duchow, J., Delville, J.P., & Goldman, M. (1992).
Lipopolysaccharide induces up-regulation of CD14 molecule on
monocytes in human whole blood. EurJ Immunol, 22, 1663-1665.

Margaretten, W., McKay, D.G., 8. Phillips, LL. (1967). The effect of
heparin on endotoxin shock in the rat. American Journal of Pathology, 51,
61-68.

Martin, T.R., Rubenfeld, G.D., Ruzinski, J.T., Goodman, R.B., Steinberg,
K.P., Leturcq, D.J., Moriarty, A.M., Raghu, G., Baughman, R.P., &
Hudson, L.D. (1997). Relationship between soluble CD14,
Iipopolysaccharide binding protein, and the alveolar inflammatory

209

response in patients with acute respiratory distress syndrome. Am J
Respir Crit Care Med, 155, 937-944.

Matsumoto, M., Tamura, M., Matsubara, S., Matsuno, T., Ono, M., &
Yokota, T. (1991). Mechanism of protective effect of recombinant human
granulocyte colony-stimulating factor (rG-CSF) on Pseudomonas
infection. Microbiol Immunol, 35, 461-474.

Mavier, P., Preaux, A.-M., Guigui, B., Lescs, M.-C., Zafrani, E.S., &
Dhumeaux, D. (1988). In vitro toxicity of polymorphonuclear neutrophils
to rat hepatocytes: evidence for a proteinase-mediated mechanism.
Hepatology, 8(2), 254-258.

Mawet, E., Shiratori, Y., Hikiba, Y., Takada, H., Yoshida, H., Okano, K.,
Komatsu, Y., Mastumura, M., Niwa, Y., & Omata, M. (1996). Cytokine-
induced neutrophil chemoattractant release from hepatocytes is
modulated by Kupffer cells. Hepatology, 23, 353-358.

Means, T.K., Golenbock, D.T., & Fenton, M.J. (2000). Structure and
function of Toll-like receptor proteins. Life Sci, 68, 241-258.

Means, T.K., Golenbock, D.T., & Fenton, M.J. (2000). The biology of Tall-
like receptors. Cytokine Growth Factor Rev, 11, 219-232.

Medzhitov, R. & Janeway, C.J. (2000). The Toll receptor family and
microbial recognition. Trends Microbiol, 8, 452-456.

Medzhitov, R., Preston-Hurlburt, P., & Janeway, C.A.J. (1997). A human
homologue of the Drosophila Toll protein signals activation of adaptive
immunity. Nature, 388, 394-397.

Medzhitov, R., Preston-Hurlburt, P., Kopp, E., Stadlen, A., Chen, C.,
Ghosh, S., & Janeway, C.A.J. (1998). MyD88 is an adaptor protein in the
hToll/lL-l receptor family signaling pathways. Mol Cell, 2, 253-258.

210

Miccadei, S., Nakae, D., Kyle, M.E., Gilfor, D., & Farber, J.L. (1988).
Oxidative cell injury in the killing of cultured hepatocytes by allyl alcohol.
Archives of Biochemistry and Biophysics, 265(2), 302-310.

Michael, S.L., Pumford, N.R., Mayeux, P.R., Niesman, M.R., & Hinson,
J.A. (1999). Pretreatment of mice with macrophage inactivators
decreases acetaminophen hepatotoxicity and the formation of reactive
oxygen and nitrogen species. Hepatology, 30, 186-195.

Morgan, ET. (2001). Regulation of cytochrome p450 by inflammatory
mediators: why and how? Drug Metab Dispos, 29, 207-212.

Moriya, S., Yokoyama, H., Fukuda, M., Okamura, Y., Kamegaya, Y.,
Mizukami, T., Ohgo, H., & lshii, H. (2000). Glutathione depletion
enhances the formation of superoxide anion released into hepatic
sinusoids after Iipopolysaccharide challenge. Alcohol Clin Exp Res, 24,
598-638.

Morrison, D.C., Vukajlovich, S.W., Goodman, S.A., & Wollenweber, H.W.
(1985). Regulation of Iipopolysaccharide biologic activity by
polysaccaride. The Pathogenesis of Bacterial Infections (Bayer Symp. VIII),
68.

Morrison, D.C., Lei, M.G., Chen, T.Y., Flebbe, L.M., Hailing, J., & Field, S.
(1992). Identification and characterization of mammalian cell membrane
receptors for LPS-endotoxin. Adv Exp Med Biol, 319, 23-29.

Moulin, F., Pearson, J.M., Schultze, A.E., Scott, M.A., Schwartz, K.A.,
Davis, J.M., Ganey, P.E., & Roth, RA. (1996). Thrombin is a distal
mediator of Iipopolysaccharide-induced liver injury in the rat. J Surg Res,
65, 149-158.

Mulford, C.A. & Osborn, M.J. (1983). An intermediate step in
translocation of lipopolysaccharide to the outer membrane of Salmonella
typhimurium. Proc Natl Acad Sci, 80, 1159.

Munford, R.S., Hall, C.L., Lipton, J.M., & Dietschy, J.M. (1982).
Biological activity, lipoprotein-binding behavior, and in vivo disposition of

211

extracted and native forms of Salmonella typhimurium
Iipopolysaccharides. J Clin Invest, 70, 877-888.

Muzio, M., Natoli, G., Saccani, S., Levrero, M., & Mantovani, A. (1998).
The human toll signaling pathway: divergence of nuclear factor kappaB
and JNK/SAPK activation upstream of tumor necrosis factor receptor-
associated factor 6 (TRAF6). J Exp Med, 187, 2097-2101.

Myc, A., Buck, J., Gonin, J., Reynolds, B., Hammerling, U., & Emanuel,
D. (1997). The level of Iipopolysaccharide-binding protein is significantly
increased in plasma in patients with the systemic inflammatory response
syndrome. Clin Diagn Lab Immunol, 4, 113-116.

Nikaido, H. & Vaara, M. (1987). Outer membrane. Escherichia coli and
salmonella typhimurium Cellular and Molecular Biology, 7-22.

Ninomiya-Tsuji, J., Kishimoto, K., Hiyama, A., lnoue, J., Cao, Z., &
Matsumoto, K. (1999). The kinase TAK1 can activate the NlK-l kappaB
as well as the MAP kinase cascade in the lL-1 signalling pathway.
Nature, 398, 252-256.

Nolan, J.P. & Leibowitz, AL (1978). Endotoxin and the liver. lll.
Modification of acute carbon tetrachloride injury by polymyxin b--an
antiendotoxin. Gastroenterology, 75, 445-449.

Osborn, M.J. (1979). Biosynthesis and assembly of Iipopolysaccharides
of the outer membrane. Bacterial Outer Membranes, Biogenesis and
Functions, 15.

Palmer, K.R., Duerden, B.l., & Holdsworth, CD. (1980). Bacteriological
and endotoxin studies in cases of ulcerative colitis submitted to surgery.
Gut, 21, 851-854.

Patel, J.M., Wood, J.C., & Leibman, KC. (1980). The biotransformation
of allyl alcohol and acrolein in rat liver and lung perparations. Drug
Metabolism and Disposition, 8(5), 305-308.

212

 

Paul, A., Wilson, 8., Belham, C.M., Robinson, C.J., Scott, P.H., Gould,
G.W., & Plevin, R. (1997). Stress-activated protein kinases: activation,
regulation and function. Cell Signal, 9, 403-410.

Pearson, J.M., Schultze, A.E., Jean, P.A., & Roth, RA. (1995). Platelet
participation in liver injury from gram-negative bacterial
Iipopolysaccharide in the rat. Shock, 4, 178-186.

Pearson, J.M., Schultze, A.E., Schwartz, K.A., Scott, M.A., Davis, J.M., &
Roth, RA. (1996). The thrombin inhibitor, hirudin, attenuates
lipopolysaccharide-induced liver injury in the rat. J Pharmacol Exp Ther,
278, 378-383.

Peavy, D.L. & Fairchild, E.J. (1987). Toxicologic interactions between
ozone and bacterial endotoxin. Environ Res, 42, 63-71.

Pedron, T., Girard, R., & Chaby, R. (1995). Variation of LPS-binding
capacity, epitope expression, and shedding of membrane-bound CD14
during differentiation of human monocytes. J Immunol, 155, 1460-1471.

Penttila, K.E. (1988). Allyl alcohol cytotoxicity and glutathione depletion
in isolated periportal and perivenous rat hepatocytes. Chemica-
Biological Interactions, 65, 107-121.

Pereira, F.E., Motta, L., & Cardoso, AA. (1997). Kupffer cell activation
with BCG. Corynebacterium parvum or zymosan protects against acute
liver injury induced by carbon tetrachloride in rats. Arq Gastroenteral, 34,
157-162.

Pernerstorfer, T., Hollenstein, U., Hansen, J., Knechtelsdorfer, M.,
Stohlawetz, P., Graninger, W., Eichler, H.G., Speiser, W., & Jilma, B.
(1999). Heparin blunts endotoxin-induced coagulation activation.
Circulation, 100, 2485-2490.

Piccardonl, P., Evangelista, V., Piccoli, A., De Gaetano, G., Walz, A., &
Cerletti, C. (1996). Thrombin-activated human platelets release two
NAP-2 variants that stimulate polymorphonuclear leukocytes. Thromb
Haemost, 76, 780-785.

213

Pinzani, M., Carloni, V., Marra, F., Riccardi, D., Laffi, G., & Gentilini, P.
(1994). Biosynthesis of platelet-activating factor and its 10-acyl
analogue by liver fat-storing cells. Gastroenterology, 106, 1301-1311.

Poltorak, A., He, X., Smirnova, l., Liu, M.Y., Huffel, C.V., Du, X., Birdwell,
D., Alejos, E., Silva, M., Galanos, C., Freudenberg, M., Ricciardi-
Castagnoli, P., Layton, B., & Beutler, B. (1998). Defective LPS signaling
in C3H/HeJ and CS7BL/108cCr mice: mutations in Tlr4 gene. Science,
282, 2085-2088.

Pompella, A., Romani, A., Benedetti, A., & Comporti, M. (1991). Loss of
membrane protein thiols and lipid peroxidation in allyl alcohol
hepatotoxicity. Biachem.PharmacaI., 41, 1255-1259.

Prager, R.L., Dunn, E.L., Kirsh, M.M., & Penner, J.A. (1979). Endotoxin-
induced intravascular coagulation (DIC) and its therapy. Advances in
Shock Research, 2, 277-287.

Przybocki, J.M., Reuhl, K.R., Thurman, R.G., & Kauffman, PC (1992).
Involvement of nonparenchymal cells in oxygen-dependent hepatic injury
by allyl alcohol. Toxicology and Applied Pharmacology, 115, 57-63.

Qi, M. & Jones, SB. (1990). Contribution of platelet activating factor to
hemodynamic and sympathetic responses to bacterial endotoxin in
conscious rats. Circulatory Shock, 32, 153-163.

Raetz, C.R.H., Brozek, K.A., Clementz, T., Coleman, J.D., Galloway, S.M.,
Golenbock, D.T., & Hampton, R.Y. (1988). Gram-negative endotoxin: A
biologically active lipid. Cold Spring Harbor Symposia on Quantitative Biol.,
Llll, 973-982. _

Rahman, M.M., Suzuki, M., Unno, M., Endo, K., Takeuchi, H., Kakita, T.,
& Matsuno, S. (1999). Hepatic phosphatidylcholine hydroperoxide
content in noncirrhotic, cirrhotic, and antioxidant-treated rats with
endotoxemia. Surg Today, 29, 1047-1052.

Reid, W.D. (1972). Mechanism of allyl alcohol-induced hepatic necrosis.
Experientia, 28, 1058-1061.

214

Riesenberg, K., Schlaeffer, F., Katz, A., & Levy, R. (1995). Inhibition of
superoxide production in human neutrophils by combinations of heparin
and thrombolytic agents. Br HeartJ, 73, 14-19.

Rietschel, E.T., Kirikae, T., Schade, F.U., Mamat, U., Schmidt, G.,
Loppnow, H., Ulmar, A.J., Zahringer, U., Seydel, U., di Padova, F.,
Schreier, M., & Brade, H. (1994). Bacterial endotoxin: molecular

relationships of structure to activity and function. FASEB Journal, 8, 217-
225.

Rikans, LE. (1987). The oxidation of acrolein by rat liver aldehyde
dehydrogenases. Relation to allyl alcohol hepatotoxicity. Drug
Metabolism and Disposition, 15, 356-362.

Rikans, L.E. & Cal, Y. (1994). Dithiothreitol reversal of allyl alcohol
cytotoxicity in isolated rat hepatocytes. Toxicology, 86, 147-161.

Rikans, L.E., Cai, Y., & Hornbrook, KR. (1995). Allyl alcohol cytotoxicity
in isolated rat hepatocytes: effects of azide, fasting and fructose.
J. Tax. Env. Health, 44, 1-11.

Ritchie, H., Jamieson, A., & Booth, NA. (1995). Thrombin modulates
synthesis of plasminogen activator inhibitor type 2 by human peripheral
blood monocytes. Blood, 86, 3428-3435.

Rock, F.L., Hardiman, G., Timans, J.C., Kastelein, R.A., & Bazan, J.F.
(1998). A family of human receptors structurally related to Drosophila
Toll. Proc Natl Acad Sci U SA, 95, 588-593.

Roth, R.A., Harkema, J.R., Pestka, J.P., & Ganey, PE. (1997). Is
exposure to bacterial endotoxin a determinant of susceptibility to
intoxication from xenobiotic agents? ToxicolAppl Pharmacol, 147, 300-
311'.

Ruetten, H. & Thiemermann, C. (1997). Effects of tyrphostins and
genistein on the circulatory failure and organ dysfunction caused by
endotoxin in the rat: a possible role for protein tyrosine kinase. BrJ
Pharmacol, 122, 59-70.

215

Ruttinger, D., Vollmar, B., Wanner, G.A., & Messmer, K. (1996). In vivo
assessment of hepatic alterations following gadolinium chloride-induced
Kupffer cell blockade. J Hepatol, 25, 960-967.

Rylander, R. & Nordstrand, A. (1974). Pulmonary cell reactions after
exposure to cotton dust extract. British Journal of Industrial Medicine, 31,
220-223.

Salmon, M., Koto, H., Lynch, D.T., Haddad, E.B., Lamb, N.J., Quinlan,
G.J., Barnes, P.J., & Chung, K.F. (1998). Proliferation of airway
epithelium after ozone exposure: effect of apocynin and dexamethasone.
Am J Respir Crit Care Med, 157, 970-977.

Sarphie, T.G., D'Souza, N.B., & Deaciuc, IV (1996). Kupffer cell
inactivation prevents lippolysaccharide-induced structural changes in the
rat liver sinusoid: and electron-microscopic study. Hepatology, 23, 788-
796.

Sarphie, T.G., D'Souza, N.B., Spitzer, J.J., & Deaciuc, IV (1996).
Chronic alcohol feeding in liquid diet or in drinking water has similar
effects an electron microscopic appearance of the hepatic sinusoid in the
rat. Alcohol Clin Exp Res, 20, 973-979.

Sasse, D. & Maly, LP. (1991). Studies on the periportal hepatotoxicity of
allyl alcohol. Progress in Histo- and Cytochemistry , 23, 146-149.

Sato, T., Shinzawa, H., Abe, Y., Takahashi, T., & Arai, S. (1993).
lnhbition of corynebacterium parvum-primed and Iipopolysaccharide-
induced hepatic necrosis in rats by selective depletion of neutrophils
using a monoclonal antibody. Journal of Leukocyte Biology, 53, 144-150.

Schreck, R., Rieber, P., & Baeuerle, PA. (1991). Reactive oxygen
intermediates as apparently widely used messengers in the activation of
the NF-kappa B transcription factor and HIV-1. EMBO J, 10, 2247-2258.

Schwandner, R., Dziarski, R., Wesche, H., Rothe, M., & Kirschning, C.J.
(1999). Peptidoglycan- and lipoteichoic acid-induced cell activation is
mediated by toll-like receptor 2. J Biol Chem, 274, 17406-17409.

216

 

Seglen, PO. (1973). Preparation of rat liver cells. 3. Enzymatic
requirements for tissue dispersion. Exp Cell Res, 82, 391-398.

Serafini-Cessi, F. (1972). Conversion of allyl alcohol into acrolein by rat
liver. Biochem.J., 128, 1103-1107.

Shapiro, R.A., Cunningham, M.D., Ratcliffe, K., Seachord, C., Blake, J.,
Bajorath, J., Aruffo, A., & Darveau, RP. (1997). Identification of CD14
residues involved in specific Iipopolysaccharide recognition. Infect
Immun, 65, 293-297.

Shibayama, Y. (1987). Sinusoidal circulatory disturbance by
microthrombosis as a cause of endotoxin-induced hepatic injury. Journal
of Pathology, 151, 315-321.

Shimazu, R., Akashi, S., Ogata, H., Nagai, Y., Fukudome, K., Miyake, K.,
& Kimoto, M. (1999). MD-2, a molecule that confers Iipopolysaccharide
responsiveness on Toll-like receptor 4. J Exp Med, 189, 1777-1782.

Shin, C.S., Han, J.U., Kim, J.L., Schenarts, P.J., Traber, L.D., Hawkins,
H.,'& Traber, D.L. (1997). Heparin attenuated neutrophil infiltration but
did not affect renal injury induced by ischemia reperfusion. Yonsei Med
J, 38, 133-141.

Silva, J.M. & O'Brien, P.J. (1989). Allyl alcohol-and acrolein-induced
toxicity in isolated rat hepatocytes. Archives of Biochemistry and
Biophysics, 275(2), 551-558.

Sneed, R.A., Buchweitz, J.P., Jean, P., & Ganey, PE. (2000).
Pentoxifylline attenuates bacterial lipopolysaccharide-induced
enhancement of allyl alcohol hepatotoxicity. Toxicol Sci, 56, 203-210.

Sneed, R.A., Grimes, S.D., & Ganey, PE. (1995). Lipopolysaccharide
enhances the hepatotoxicity of allyl alcohol in sprague-dawley rats.
Fundamental and Applied Toxicology, 15, 219(Abstract)

217

 

Sneed, R.A., Grimes, S.D., Schultze, A.E., Brown, A.P., & Ganey, P.E.
(1997). Bacterial endotoxin enhances the hepatotoxicty of allyl alcohol.
Toxicology and Applied Pharmacology, 144, 77-87.

Spitzer, J.A. & Mayer, MS. (1993). Hepatic neutrophil influx: eicosanoid
and superoxide formation in endotoxemia. J.Surg.Res. , 55, 60-67.

Stachlewitz, R.F., Seabra, V., Bradford, B., Bradham, C.A., Rusyn, l.,
Germolec, D., & Thurman, R.G. (1999). Glycine and uridine prevent D-
galactosamine hepatotoxicity in the rat: role of Kupffer cells. Hepatology,
29, 737-745.

Steinemann, S., Ulevitch, R.J., & Mackman, N. (1994). Role of the
Iipopolysaccharide (LPS)-binding protein/CD14 pathway in LPS
induction of tissue factor expression in monocytic cells. Arteriascler
Thromb, 14, 1202-1209.

Stalk, J., Hiltermann, T.J., Dijkman, J.H., & Verhoeven, A.J. (1994).
Characteristics of the inhibition of NADPH oxidase activation in
neutrophils by apocynin, a methoxy- substituted catechol. Am J Respir
Cell Mol Biol, 11., 95-102.

Sugatani, J., lgarashi, T., Munakata, M., Komiyama, Y., Takahashi, H.,

Komiyama, N., Maeda, T., Takeda, T., & Miwa, M. (2000). Activation of
coagulation in 057BL/6 mice given verotoxin 2 (VT2) and the effect of

co-administration of LPS with VT2. Thromb Res, 100, 61-72.

Supinski, G., Stofan, D., Nethery, D., Szweda, L., & DiMarco, A. (1999).
Apocynin improves diaphragmatic function after endotoxin
administration. J Appl Physiol, 87, 776-782.

Suzuki, Y., Tanigaki, T., Helmer, D., Wang, W.Z., Ross, W.G., Sussman,
H.H., & Raffin, TA (1992). Polyethylene glycol-conjugated superoxide

dismutase attenuates septic lung injury in guinea pigs. Am Rev Respir

Dis, 145, 388-393.

218

 

Takada, H. &Kotani, S. (1992). Structure function relations of lipid A.
Bacterial Endotoxic Lipopolysaccharides, Molecular Biochemistry and Cellular
Biology, 107-130.

Tamura, Y., Chi, L.G., Driscoll, E.M.J., Hoff, P.T., Freeman, B.A.,
Gallagher, K.P., & Lucchesi, BR. (1988). Superoxide dismutase
conjugated to polyethylene glycol provides sustained protection against
myocardial ischemia/reperfusion injury in canine heart. Circ Res, 63,
944-959.

Terashita, Z.-l., Kawamura, M., Takatani, M., Tsushima, 8., lmura, Y., &
Nishikawa, K. (1992). Beneficial effects of TCV-309, a novel potent and
selective platelet activating factor antagonist in endotoxin and
anaphylactic shock in rodents. J. of Pharmacology and Experimental
Theraputics, 748-755.

Thomson, P.D., Till, G.O., Woolliscroft, J.O., Smith, D.J., & Prasad, J.K.
(1990). Superoxide dismutase prevents lipid peroxidation in burned
patients. Burns, 16, 406-408.

Thurman, R.G. (1998). ll. Alcoholic liver injury involves activation of
Kupffer cells by endotoxin. Am J Physiol, 275, G605-G611

Tobias, P.S., Mathison, J.C., & Ulevitch, R.J. (1988). A family of
lipopolysaccharide binding proteins involved in responses to gram-
negative sepsis. J Biol Chem, 263, 13479-13481.

Tobias, P.S., Soldau, K., lovine, N.M., Elsbach, P., & Weiss, J. (1997).
Lipopolysaccharide (LPS)-binding proteins BPI and LBP form different
types of complexes with LPS. J Biol Chem, 272, 18682-18685.

Tobias, P.S., Tapping, R.l., & Gegner, J.A. (1999). Endotoxin interactions
with lipopolysaccharide-responsive cells. Clin Infect Dis, 28, 476-481.

Todoroki, H., Nakamura, S., Higure, A., Okamoto, K., Takeda, S.,
Nagata, N., ltoh, H., & Ohsato, K. (2000). Neutrophils express tissue
factor in a monkey model of sepsis. Surgery, 127, 209-216.

219

 

Troelstra, A., Giepmans, B.N., Van Kessel, K.P., Lichenstein, H.S.,
' Verhoef, J., & Van Strijp, J.A. (1997). Dual effects of soluble CD14 on
LPS priming of neutrophils. J Leukoc Biol, 61, 173-178.

Ulevitch, R.J. (2000). Molecular mechanisms of innate immunity.
ImmunoI Res, 21, 49-54.

Ulevitch, R.J. & Tobias, PS (1999). Recognition of gram-negative
bacteria and endotoxin by the innate immune system. Curr Opin l'
Immunol, 11, 19-22. ‘

Underhill, D.M., Ozinsky, A., Hajjar, A.M., Stevens, A., Wilson, 08.,
Bassetti, M., & Aderem, A. (1999). The Toll-like receptor 2 is recruited to

macrophage phagosomes and discriminates between pathogens. Nature, 3
401, 811-815.

 

Vaara, M. & Nikaido, H. (1984). Molecular organization of bacterial outer
membrane. Chemistry of Endotoxin, 1.

Vergnolle, N., Wallace, J.L., Bunnett, N.W., & Hollenberg, MD. (2001).
Protease-activated receptors in inflammation, neuronal signaling and
pain. Trends Pharmacol Sci, 22, 146-152.

Villa, P., Sartor, G., Angelini, M., Sironi, M., Conni, M., Gnocchi, P.,
lsetta, A.M., Grau, G., Buurman, W., & van Tits, L.J. (1995). Pattern of
cytokines and pharmacomodulation in sepsis induced by. cecal ligation
and puncture compared with that induced by endotoxin. Clin Diagn.Lab
Immunol, 2, 549-553.

Viriyakosol, S. & Kirkland, TN. (1996). The N-terminal half of membrane
CD14 is a functional cellular Iipopolysaccharide receptor. Infect Immun,
64, 653-656.

Vollmar, B., Ruttinger, D., Wanner, G.A., Leiderer, R., & Menger, MD.
(1996). Modulation of kupffer cell activity by gadolinium chloride in
endotoxemic rats. Shock, 6, 434-441.

220

 

Wang, W., Suzuki, Y., Tanigaki, T., Rank, D.R., & Raffin, TA (1994).
Effect of the NADPH oxidase inhibitor apocynin on septic lung injury in
guinea pigs. Am J Respir Crit Care Med, 150, 1449-1452.

Wang, Y. & Hollingsworth, RI. (1996). An NMR spectroscopy and
molecular mechanics study of the molecular basis for the
supramolecular structure of Iipopolysaccharides. Biochemistry, 35,
5647-5654.

Wesche, H., Henzel, W.J., Shillinglaw, W., Li, S., & Cao, Z. (1997).
MyD88: an adapter that recruits IRAK to the lL-1 receptor complex.
Immunity, 7, 837-847.

Wolpert, J.S. & Albersheim, P. (1976). Host-symbiont interactions. l. The
lectins of legumes interact with the o-antigen-containing
Iipopolysaccharides of their symbiont Rhizobia. Biochem Biophys Res
Commun , 70, 729-737.

Woodman, R.C., Teoh, D., Payne, D., & Kubes, P. (2000). Thrombin and
leukocyte recruitment in endotoxemia. Am J Physiol Heart Circ Physiol,
279, H1338-H1345

Wright, S.D., Ramos, R.A., Hermanowski-Vosatka, A., Rockwell, P., &
Detmers, P.A. (1991a). Activation of the adhesive capacity of CR3 on
neutrophils by endotoxin: dependence on Iipopolysaccharide binding
protein and CD14. JExp Med, 173, 1281-1286.

Wright, S.D., Ramos, R.A., Hermanowski-Vosatka, A., Rockwell, P., &
Detmers, P.A. (1991b). Activation of the adhesive capacity of CR3 on
neutrophils by endotoxin: dependence on Iipopolysaccharide binding
protein and CD14. JExp Med, 173, 1281-1286.

Wright, s.o., Ramos, R.A., Tobias, P.S., Ulevitch, R.J., & Mathison, J.C.
(1990). C014, a receptor for complexes of Iipopolysaccharide (LPS) and
LPS binding protein. Science, 249, 1431-1433.

221

Wueweera, J.B., Gandolfi, A.J., Badger, D.A., Sipes, I.G., & Brendel, K.
(1996). Vitamin A potentiation of vinylidene chloride hepatotoxicity in
rats and precision-cut rat liver slices. Fundam Appl Toxicol, 34, 73-83.

Wurfel, M.M., Kunitake, S.T., Lichenstein, H., Kane, J.P., & Wright, SD.
(1994). Lipopolysaccharide (LPS)-binding protein is carried on
lipoproteins and acts as a cofactor in the neutralization of LPS. J Exp
Med, 180, 1025-1035.

Wurfel, M.M., Monks, B.G., Ingalls, R.R., Dedrick, R.L., Delude, R., Zhou,
D., Lamping, N., Schumann, R.R., Thieringer, R., Fenton, M.J., Wright,
S.D., & Golenbock, D. (1997). Targeted deletion of the
Iipopolysaccharide (LPS)-binding protein gene leads to profound
suppression of LPS responses ex vivo, whereas in vivo responses remain
intact. J Exp Med, 186, 2051-2056.

Yamanaka, H., Nukina, 8., Handler, J.A., Currin, R.T., Lemasters, J.J., &
Thurman, R.G. (1992). Transient activation of hepatic glycogenolysis by
thrombin in perfused rat livers. Eur.J Biochem, 208, 753-759.

Yang, R.B., Mark, M.R., Gray, A., Huang, A., Xie, M.H., Zhang, M.,
Goddard, A., Wood, W.l., Gurney, A.L., & Godowski, P.J. (1998). Toll-like
receptor-2 mediates lipopolysaccharide-induced cellular signalling.
Nature, 395, 284-288.

Yee, S.B., Kinser, S., Hill, D.A., Barton, 00, Hotchkiss, J.A., Harkema,
J.R., Ganey, P.E., & Roth, RA (2000). Synergistic Hepatotoxicity from
Coexposure to Bacterial Endotoxin and the Pyrrolizidine Alkaloid
Monocrotaline. Toxicology and Applied Pharmacology, 166, 173-185.

Zhang, P., Xie, M., Zagorski, J., & Spitzer, J.A. (1995). Attenuation of
hepatic neutrophil sequestration by anti-CINC antibody in endotoxic rats.
Shock, 4, 262-268.

Zimmerman, B.J., Paulson, J.C., Arrhenius, T.S., Gaeta, F.C.A., &
Granger, D.N. (1994). Thrombin receptor peptide-mediated leukocyte
rolling in rat mesenteric venules: roles of P-selectin and sialyl Lewis X.
Am.J.Physiol., 267, H1049-H1053

222