n“

05:?

 

; 'iffi‘éflzfii

      

o
o

-
>—

 

~ m’rrfi

mi“

»

33$

7.— 49*

M
4

h...‘

17
"‘59

 

F.

g; .
“.1“

“m.

M“... a"
"u“ 0...

an

 

65’“

3’3.

‘3’

' -
'w.....

, J .3.
..
AMP-:fiiifi
. r ‘5, ‘
c- ‘. -

Y
5' 7‘9“
7. .3

‘1
'%

 

‘. ““4

2'

..
.v

....
~.vh9u‘
. 4-.. .
“-1,
~77.
Er» vhf-3's"

.

7". V

5a
1: 3'7"
. .
1‘3"

"N film
.23?»-

an.

4‘12?

. ’11:»:
(V g

:3.

‘3'

7
u
2

on I
«am- ’-
‘ Ja '

‘5
ude—I

I’

.4“ ~11.-
Er:

..

- I an

‘ .4-“
I

‘f'
.. bum.
4. .
mo
..
u
-v
‘.

..~...

.
‘
.ttx' ,
9521
~ :33
3

 

.. )4-J-'
a
...~ 7-
. u
o
w
.u
wu
“a
.n
..

A .- .
717.7.

7'-

l .1
uh."
.

'l
,,.,<..
_..
‘-
.,_
-~/'
(.1.N-
7,»

7

 

.v “r

'0‘:

0..-

w 2~
,.

Elm
tri-

. L- n» o
«nu-A4,
c w»
mm a»
"1'

u...

 

.I 7“

:3:

«.23
. .;

'7.
‘ .3... .vur-rvnr

,V. 1 n” -
hey-r
«.1

M
air“? ,

~77. .. _..7

.4" 3‘. w

. .
, , .. 4
An
....= I
..‘.-
"in.

b.‘

 

¢‘

+7-

sat-w-

.\’
l
I

I
0‘:
. 74‘};
0‘ ‘1

a.

C3.

..u

{-7 1— ~
'u' .~ an"
M )3751’23:
" 3!; d‘ 3“
Eu

 

A

 

.‘t,
l“’ in?!
2" oh >

 

'
t 21

.. ..—‘f..,

I
C IR C1 (A

MICHIGAN sun

I III IIIIIIIII IIIII’I III IIIIIIII

01405 1878

II

 

 

 

 

 

 

 

 

This is to certify that the

thesis entitled

.DRREAGE, THE STRESS RESPONSE,
AND THE HEAT SHOCK RESPONSE.

presented by

David Michael Kan

has been accepted towards fulfillment
of the requirements for

H. 3- degree in SURGERY

 

 

Major professor /\

Date €////?é

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

 

 

 

 

LIBRARY
Michigan State
Untversity

 

 

 

PLACE ll RETURN BOX to roman this chockout from your rooord.

TO AVOID FINES Mum on or Moro Moduo.

  

DATE DUE DATE DUE DATE DUE

  
 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

  

 

 

 

l l I
-F —I-
{ME

 

 

 

 

 

 

 

Hemorrhage, the stress response,
and the heat shock response

by

David Michael Kam, MD

A THESIS

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

MASTER OF SCIENCE

Department of Surgery

1 995

ABSTRACT

HEMORRHAGE, THE STRESS RESPONSE,
AND THE HEAT SHOCK RESPONSE

By
David Michael Kam

The 70kD family of heat shock proteins (HSP70 / HSP72) are
purportedly cytoprotective during stressfinjury. Although it is known that Kupffer
cell (KC) cytokine release and cytotoxicity is increased following hemorrhage
but depressed in peritoneal macrophages (pMD), it is unknown what role HSP's
play in these changes. ‘

Mice were bled to and maintained at a mean arterial pressure of
35mmHg for one hour and then adequately resuscitated, followed by KC and
pMD isolation at 2 and 24 hours. Western blot analysis was performed and
HSP72 expression quantified densitometrically. In addition, TNF—a and lL—6
productive capacities of the two macrophage populations were measured.
Results were compared to those from sham—operated mice.

Hemorrhaged mice exhibited significantly increased expression of HSP
72 and increased productive capacities of both lL—6 and TNF—a at both 2 and
24 hours when compared to shams. The ability to mount a heat shock
response may preserve the capacity of KC to release inflammatory cytokines

and maintain cytotoxicity.

Table of Contents

Page Number
List of Figures v
List of Tables vl
Introduction 1-16
A. General 1
B. Heat shock proteins 3

1. Historical aspects
2. lnducers of the heat shock response,
thermotolerance, and protection against stress
3. The HSP 70 family of proteins
4. The stress response
C. Physiological aspects of hemorrhage 12
1. Hemorrhage and the stress response
2. Cytokines and immune deficiency
D. Heat shock protein 72 expression in a murine hemorrhage model 15
1. Specific aims
2. Hypothesis

IV. Materials and methods 17-22

A Hemorrhage Model 17
B. Macrophage cell preparations 18
Preparation of Peritoneal macrophages

Preparation of Kupffer Cells

Cytokine bioassays

Protein extraction and quantitation

Tricine PAGE electrophoresis

Western Blot and quantitation of HSP 70 synthesis

Statistics

NP’P'PSDN?

V. Results 23-43

A Macrophage Cytokine Productive Capacities 23
in Response to Hemorrhage / Resuscitation
1. Cell yields and viability of macrophage preparation and
isolation
2. Kupffer cells
3. Peritoneal macrophages

B. HSP72 Expression in Response to Hemorrhage/
Resuscitation
1. Kupffer cells
2. Peritoneal macrophages

VI. Discussion: Implications and clinical relevance

A Hemorrhage and its Immunologic Consequences

B. Differential Expression of HSP72 from Kupffer Cells and
Peritoneal Macrophages in Response to Hemorrhage]
Resuscitation

VII. Summary and Conclusions

A Summary
B. Conclusions
C. Limitations and future studies

VIII. Appendices
IX. Bibliography

iv

LIST OF FIGURES
Page number

Figure 1: TNF—or productive capacities of Kupffer cells 2 hours after 27

hemorrhage and resuscitation.

Figure 2: TNF—a productive capacities of Kupffer cells 24 hours after 28

hemorrhage and resuscitation.

Figure 3: IL-6 productive capacities of Kupffer cells 2 hours after 29

hemorrhage and resuscitation.

Figure 4: lL—6 productive capacities of Kupffer cells 24 hours after 30

hemorrhage and resuscitation.

Figure 5: TNF—a productive capacities of peritoneal macrophages 2 31

hours after hemorrhage and resuscitation.

Figure 6: TNF—a productive capacities of peritoneal macrophages 24 32

hours after hemorrhage and resuscitation.

Figure 7: IL—6 productive capacities of peritoneal macrophages 2 33

hours after hemorrhage and resuscitation.

Figure 8: lL—6 productive capacities of peritoneal macrophages 24 34

hours after hemorrhage and resuscitation.

Figure 9: hsp 72 expression of Kupffer cells 2 hours after hemorrhage 38

Figure 10:

Figure 11:

Figure 12:

Figure 13:

Figure 14:

and resuscitation.

hsp 72 expression of Kupffer cells 24 hours after hemorrhage 39
and resuscitation.

hsp 72 expression of peritoneal macrophages 2 hours after 40
hemorrhage and resuscitation.

hsp 72 expression of peritoneal macrophages 24 hours after 41
hemorrhage and resuscitation.

Representative sample of chemiluminescent radiographs from 42
hsp72 Western blot analysis: Kupffer cells

Representative sample of chemiluminescent radiographs from 43
hsp72 Western blot analysis: peritoneal macrophages.

LIST OF TABLES
Page Number

Table 1: ISummary of results representing the productive capacities 26
for TNF—a and IL-6 from PMa (Per) and KC (Kup) in
control (Sham) and hemorrhage groups (Hem) at 2 and 24
hours.

Table 2: TSummary of results comparing the expression of hsp72 at 37
2 and 24 hours for Kupffer cells (KC) and peritoneal
macrophges (PMD).

vi

III. INTRODUCTION

GENERAL

An organism's response to a potentially injurious external stimulus is
usually one of self preservation. This is a core principle that is not only
germane to a single being but one that can be extended vertically to the group
or social structure in which that being participates or intrinsically to the organ
systems that makes up the organism. For example, much in the same way that
human societies contain within them and make use of military forces to defend
the society from both internal and external threats, homeostatic mechanisms
protect an organism from that which may harm the organism. This analogy,
albeit simplistic, does have relevance to the way one approaches the study of
basic responses to injury. Defensive mechanisms are inherent and pervasive
at both the macroscopic and microscopic level and function as part of a greater
homeostatic mechanism that involves the interaction of individual organ
systems. If the result of marshaling defense forces allows an organism (or
society) to maintain or allow a return to normal homeostasis (or normal social
order), then that organism (or society) is able to survive.

However, if the insult is so great that normal homeostatic mechanisms
are not allowed to return or are sacrificed in order to maintain a defensive
posture, it is often the case that the organism will succumb as a result of the
response to the injury. In my experience as a surgeon I have seen patients die
not only as a result of the initial injury or physiological derangement but

secondarily as a result of progressive multi—organ dysfunction syndrome

1

2
(MODS) in response to sepsis, trauma, hemorrhage, or a systemic

inflammatory response syndrome (SIRS).

The goal of this thesis is to provide an overview of mammalian
responses at a cellular and systemic level to an insult such as hemorrhage.
Specifically, I will discuss the interaction of the inflammatory response, in the
form of cytokines, with a phylogenetically ancient stress response, the heat
shock response. In so doing, I will introduce the results of experiments that
measure both cytokine release and expression of heat shock proteins in two
distinct populations of macrophages in a fixed—pressure murine model of
severe hemorrhage followed by resuscitation. I will then discuss the
implications of the findings in furthering the understanding of cellular

homeostasis and the preservation of the organism.

HEAT SHOCK PROTEINS

II' | . l | ,

In 1962, FM. Ritossa described a phenomenon of chromosomal
'puffing' in the polytene chromosomes of salivary glands of the fruit fly
Drosophila melanogaster which was induced by exposing isolated salivary
glands to temperatures warmer than what was believed most optimal for the
fruit fly's normal growth and development.‘ These 'puffs' appeared at specific
sections of the chromosomes as identified by Giemsa staining. This response
was consistently seen immediately after exposing the salivary glands to the
higher temperatures and the chromosomal changes persisted and increased in
size in a reproducible manner 1: 2. The investigators termed this phenomenon
the 'heat shock response." Tissieres and Mitchell in 1974 identified the
elaboration of specific, unique protein sequences associated with these
chromosomal puffs and coined the term, 'heat shock proteins' 3. Eventually it
was found that the heat shock response is an elemental function in all
organisms and that in many cases the response is triggered by exposure to
temperatures only slightly higher than the usual ambient temperature in which
that organism exists, such as that seen in Ritossa's fruit flies or in the
mammalian response to fever 4. As work progressed in the years following
Tissieres' work, it became evident that this general response of all cells was not
limited only to thermal stimuli. Evidence accumulated that a number of different
stimuli induced a heat shock response through the common pathway of
disruption of the protein tertiary sequence. These various stimuli include

arsenite poisoning, ethanol, heavy metal poisoning, oxygen toxicity, and oxygen

4
free radicals 2- 5- ‘- 7. lschemia - reperfusion injury, anoxia. and hemorrhage

have also been found to be strong inducers of the heat shock response.
Eventually an entire group of related proteins have been identified that
are highly conserved among species and even kingdoms; are constitutively
produwd as well as inducible; and are important to many cellular homeostatic
functions including protein folding and unfolding, protein assembly, transport
across intracellular membranes, and rescue/transport of improperly folded,
partially denatured, and abnormal proteins!» ' In their constitutive form they
are considered 'chaperone proteins' or 'housekeeping proteins' because they
accompany proteins through major phases of production, assembly, and
transport and as such they have a major role in baseline cellular homeostasis.
It is now clear that the stress proteins fall within certain 'families' based on their
size. The best —known are the 28 kD, 60 kD, 70 kD, 90 RD, and 110 kD
families of heat shock proteins (hsp28, hsp60, hsp70, hsp90, and hsp110,

respectively).

0 O
.0 : o : lzz Jo‘s}, : 0,01,: 1:1“00: .101: :_ o o 03°19! :02.

Recent evidence indicates a possible role of inducible stress proteins,
especially members of the hsp7o family, in protecting the cell against external
insults. Blockimg the activity of hsp70 of cells in vitro, via antibody binding or
competitive inhibition, prior to heat shock renders previously normal cells
thermosensitivefir 1° Therrnotolerance is the appropriate term for the protective
effect the heat shock response confers in reaction to a non-lethal thermal
insult. Interestingly, a number of studies have shown that thermotolerance is
also conferred when the heat shock response (especially increases in hsp72) is

initiated chemically such as in sodium arsenite poisoning. Naturally, the focus

5
then switches to whether induction of the heat shock response will also protect

cells and organisms from other insults. Investigative work in such matters have
found promising results and extends the role of the heat shock response in
maintenance of cellular homeostasis beyond the bounds of simple
thermotolerance.

Yellon and colleagues found specific improvements in the physiological
function of myocardium as a result of the heat shock response."- ‘1 In their
series of experiments, rabbit ex—vivo hearts, previously exposed to heat-shock
in—vivo, had improved cardiac function, decreased creatine phosphokinase
release, and significantly conserved ATP levels and mitochondrial function after
ischemia—reperfusion than did nonheat—shocked controls. It was noted that
hsp72, the primary inducible form of the hsp70 family, was the principle product
of the induced heat shock response. However, it is unlikely that heat shock
proteins are involved in classical ischemic myocardial preconditioning since
heat stress failed to limit infarct size in the rabbit myocardial reperfusion injury
model. “-13

Villar, Ribeiro, and Slutsky found an analogous result when thermal pre-
treatment, associated with a rise in a 72 kD member of the hsp70 family,
attenuated tissue damage and mortality in a rat model of experimental lung
injury.“ The same group, in more recent experiments, found that both thermal
pro-treatment and injection of sodium arsenite induces a heat shock response
with measurable and significant increases in hsp72. Interestingly, thermal pre—
treatment resulted in decreased mortality and reduced end-organ damage of
liver and lung in sepsis-induced lung injury.15 Sodium arsenite induced a
significant increase in hsp72 in lungs of treated rats and correlated with
decreased mortality during the time periods hsp72 was expressed above basal

values."3

6
It is interesting to note also that the mechanism of protection may be

related to protection from cytotoxic forces within the host. This has important
implications in disease processes such as sepsis and trauma where the
elaboration of inflammatory cytokines, especially interleukin-1 (IL-1),
interleukin-6 (IL-6), and tumor necrosis factor alpha (T NF—a), secondary to a
systemic inflammatory response syndrome (SIRS) may be cytotoxic to the
host's organ systems and progress to multiorgan dysfunction syndrome
(MODS) or frank failure. It has been found, for example. that heat—treated .
mouse cal-IA cells were resistant to the cytotoxic effects of TNF-a despite
retaining the ability to bind TNF as well as nonheat-treated cells."
Transfecting TNF-sensitive murine fibrosarcoma cells with human hsp-70 and
hsp-27 genes protected these cells from TNF cytotoxicity, TNF-mediated
monocyte cytotoxicity, and oxidative stress."

Therefore, there seems to be strong in vivo evidence of a relative
protective effect induced by the heat shock response. Teleologically, this
affirms that inherent mechanisms exist to preserve cell function and possibly
protect from further insult. Furthermore, because heat shock protein structure
identity homology is preserved among eucaryotes (often exwedlng 50%), it is
clear that this ancient, evolutionarily conserved response was meant to confer a
survival advantagefiv 1’ The common inducer of the heat shock response
appears to be the presence of abnormal or partially denatured proteins in
response to a number of stimuli, as previously stated. It would be overly
simplistic to ascribe the responsibility of maintaining what Claude Bernard
referred to as the 'milieu interieur' (internal environment) and William Cannon
later termed 'homeostasis' 2° to the heat shock response. Conversely, it is
proven that the heat shock response confers more than just thermotolerance. It

is most likely that a multimodal stress response works in concert with other

7
adaptive mechanisms to result in the survival of the organism as a whole. The

role that the heat shock response plays in this stress response is discussed in
the next two sections.

II IISEZDI 'I I I'

The human hsp70 family is comprised of at least four distinct members 4.
They are, according to the nomenclature of Watowich and Morimoto 2‘ hsp70,
hsp72, p72, and grp78. The major heat inducible 70kD hsp70 is also basally
expressed. It combines with hsp90 to form the glucocorticoid receptor 22, is
thought to stabilize the cellular oncogene p53 (along with p72), and localizes to
the nucleoli in response to stress where its mechanism of action may be to
stabilize RNA splicing and structure, especially in response to hyperthermiafi’a as
Hsp72 is a protein that is stress inducible only and is not basally produced. p72
is slightly heat inducible but to afar lesser degree than hsp70; it is elemental in
clathrin uncoating, nuclear matrix/ nucleoli binding, and also interacts with
p53.4 In humans, it can be summarized that hsp 70 is the primary inducible
form of the gene while p72 is the primary constitutive form but can also be
dramatically increased with the appropriate stimulus. grp78 (a.k.a. BiP) is
primarily located within the endoplasmic reticulum and functions primarily to fold
and unfold proteins as they make their transmembrane crossing to and from the
cytosolic sector by binding to them transiently as they assume their correct
tertiary structure. It also appears to bind permanently to proteins that are
misfolded or unfolded thereby acting as an 'editor' of misfit proteins.“ It
should be noted at this time that rodents have three known hsp70 variants very
similar in size and function to the human complement, but do not have an

equivalent of p72 4.

8
The genes for the hsp family of proteins are multiply duplicated and

located on human chromosome 6, in association with the major
histocompatibility complex, 14, and 21 4- 25. Specific and highly restricted heat
shock transcription factors (HSFs) mediate the inducible expression of heat
shock proteins via binding to the heat shock response element. Since they are
restricted, they respond to specific stimuli. For example, HSF1 is the principle
mediator of stress induced hsp expression from heat and restraint to ischemia—
reperfusion injury 22. 33. However, the activation of transcription factors and the
expression of their resulting gene products is a complex interplay of cellular
systems sometimes acting in concert and sometimes in exclusion of others, as

will be discussed in the next section.

W

Some investigators such as William Welch have advocated replacing the
terms “heat shock response' and 'heat shock proteins' with stress response
and proteins, respectively.“ This Is too narrowly defined a term. Historically
speaking, the stress response is more correctly defined as a whole body
phenomenon Involving adaptive responses including those of the autonomic
nervous system, the acute phase response, the hypothalamio-pituitary—adrenal
axis (HPA) and endocrine responses, and the heat shock response.22 In fact,
within the cell there are at least four genetically discrete responses to stress:
the heat shock response, the acute phase response, the oxidative stress
response, and the ultra violet stress response.’- 27- 2°: 29. 30 It appears that
although some of these responses are triggered by similar stimuli, as in the
case of the oxidative stress response and the heat shock response, the acute

phase and heat shock responses are discrete and exclusive of each other with

9
the heat shock response taking precedence as the more primitive and

preemptive of the two.

Local tissue injury, such as that seen in surgery and trauma, initiates an
inflammatory response with the release of circulating cytokines such as lL—1,
IL-6, and TNF-ct. These cytokines mediate the systemic effects of the acute
phase response such as fever, hypothalamic—pituitary—adrenal axis (HPA)
activation, leukocytosis, immune / reticuloendothelial system activation, and
production of acute phase proteins‘“: 32 Cytokines, including lL-1, IL-2, TNF,
and interferon-«y (INF-7), have been shown to either induce or decrease the
threshold for induction of heat shock proteins.“ 34 This phenomenon may be a
response to the cytotoxic effects of various cytokines, especially TNF, and may
also prove to be a protective regulatory effect for host tissues to remain viable
in the face of a potentially injurious inflammatory response‘”- 3‘ One has to be
cautious when interpreting the data from experimental conditions where the
heat shock response induction is brief, episodic, and artificial. In the clinical
setting (a.k.a. 'the real world'), the potential stimulus for a heat shock response
may be protracted, ongoing, and difficult to reverse. This is true in cases of
prolonged hypotension secondary to hypovolemia l hemorrhage, cardiac pump
failure, massive tissue injury and lschemia (as in crush injuries and burns), and
septic shock. However, there is evidence that critically ill human patients
express the inducible hsp72 in granulocytes; ostensibly as a result of circulating
factors during a prolonged inflammatory state.

Experimental evidence exists in vivo that severe surgical stress as well
as simple restraint stress in rats resulted in specific increases in the expression
of inducible forms of hsp70 in the adrenal gland and aorta.37- 3' In a series of
inspired and elegant experiments using chronic hormone therapy

(glucocorticoids and antiglucocorticoids), hypophysectomy, and catecholamine

1O
administration in the same model, Udelsman and his associates were able to

show that the inducible hsp70 expression in the adrenal cortex and aorta were
due to an interaction of the hypothalamic pituitary adrenal axis (i.e.,
glucocorticoid production) and catecholamines, and that the control of these
responses in these two different tissues were the result of a specific
transcription factor, heat shock factor 1 (HSF 1), which has also been shown to
be the mediator of hsp induction in response to heat and other classic
stresses”. 39 Other organs did not show an appreciable increase in the
expression of hsp70 in their models of surgical or restraint stress but their work
clearly shows that there is differential organ-specific regulation of hsp
expression based, in part, on the effect of the HPA axis in association with
catecholamine regulation.

Buchman and colleagues“ ‘0: 4‘ showed in in vivo models of
cardiogenic shock and hemorrhage resuscitation that the expression of hsp72
in the liver as well as two acute phase genes (on-anti chemotrypsin and inter-
a trypsin inhibitor). Additionally, these same investigators were able to show
involvement of oxidative stress in the form of superoxide anion generation as
an inducer of hsp70.‘2 It is unclear in these groups of experiments which cell
type in the liver (hepatocyte, Kupffer cell, endothelial cell, etc.) is primarily
responsible for the generation of hsp72. Nor is it clear whether the
phenomenon is due to ischemia alone or reperfusion through the splanchnic
bed. However, what is remarkable is that while hsp72 transcription is initiated
by the low flow state, it continues long after blood flow is restored.

In summary, the stress response involves at least four distinct genetic
responses that are regulated, in part, by the host's stressed internal
environment. Mediators of these responses include the hypothalamic—

pituitary—adrenal axis, catecholamines, and inflammatory cytokines. The

11
control of these responses is most likely located at a transcriptional level and,

as in the case of the acute phase and heat shock response, may be exclusive

of each other.

PHYSIOLOGICAL ASPECTS OF HEMORRHAGE

WWW

Sepsis, and the antecedent phenomena of multiorgan dysfunction
syndrome (MODS) and failure, is of continued interest and frustration in surgical
intensive care units world—wide. It remains the leading cause of mortality and
morbidity in surgical intensive care units.“- “- 45. 4‘ These derangements are
significantly represented in the surgical, trauma, and cardiac critical care patient
populations. Of interest in these patient populations is their susceptibility to
overwhelming infections and sepsis which may be due to a breakdown in host
immune response and intrinsic defenses. Although modern surgical critical
care and trauma care has significantly improved our ability to initially treat these
patients, it is increasingly evident that a significant number die of these
sequelae. Therefore, a significant amount of investigative effort has been spent
in dissecting out the various causes and consequences of the breakdown of
host defenses resulting from injurious insults.

Modern medicine is limited by its inability to act effectively in a proactive
manner. Although it is possible to support a critically ill patient (volume and
inotropic support in the face of shock, ventilators and positive pressure
ventilation in the face of acute respiratory distress syndrome (ARDS), antibiotics
in the face of infection, total parenteral nutrition, and surgical debridement),
overall mortality in the surgical intensive care unit has changed only
incrementally. Part of this may be selection: as more patients survive their first
insult they gain the opportunity to succumb to the more insidious process of the

inability to overcome anti—homeostatic forces set off by the initial insult.43- 44. 45

12

13
In hemorrhagic shock, the initial result of hemodynamically significant

hypovolemia results in the host's response to facilitate maximal support of vital
structures (i.e., the central nervous system) at the expense of others (e.g., skin
and integument, visceral organs, kidneys). Restoration of volume often returns
a semblance of homeostasis to the previously deranged system but this is not
universal. In trauma, for example, there are recognized patient subsets
stratified on the basis of age (pediatric; elderly), injury severity scores, and with
concomitant disease processes (immunodeficient; severe pre-existing
circulatory or cardiac problems) that tend to do poorly."7 Patients within these
groups are more likely to succumb to not only the initial insult, but the sequelae
of that insult as well: from simple opportunistic and nosocomial infections to

multiorgan failure.

CII' I' II"

It is known that burn and soft tissue injury results in a profound
suppression of cell-mediated as well as humoral immunity“: 47. 4'
Hemorrhage, with and without associated tissue injury, has also been proven
(in the same non-lethal murine hemorrhage model described here) to
significantly depress macrophage immune function, increases susceptibility to
sepsis, decreases splenocyte proliferation and lymphokine release“ 50, and
yet is associated with a concurrent increase in Kupffer cell TNF and lL—6
release as well as increased circulating levels of TNF at 2 hours post-
hemorrhage‘o- 51- 52' 53.

It appears that inflammatory cytokines, especially lL—6 and TNF-a, may
have a direct effect in causing depressed Kupffer cell antigen presenting ability
as a result of increased levels of these cytokines in both the local environment

and in blood.“- 53 Prostaglandins (especially PGEg) may play a significant role

14
since blockade by chloroquine (at phospholipase A2) and ibuprofen (at

cyclooxygenase) at least partially restores Kupffer cell antigen presentation and
prevents depression of macrophage functions“ 55- 5‘. Moreover, it has been
shown that ibuprofen also decreases the incidence of sepsis in the hemorrhagic
rodent model." However, the applicability to humans is limited due the fact
that pretreatment of humans is sometimes not even possible and overall this
modality has not made a significant decrease in the mortality of the sequlae of
hemorrhage.

There is an interesting anomaly in the results of these previous studies.
Although macrophages from liver, peritoneum, and spleen are all shown to be
relatively immunosuppressed as a consequence of hemorrhage and
resuscitation, only Kupffer cells have a significantly elevated capacity to
express the inflammatory cytokines IL—6 and TNF-ct. This disparity may be
due a difference in the relative stress that these cell populations are subjected

to. Questions about this form the basis of my experiment as described next.

HEAT SHOCK PROTEIN 72 EXPRESSION IN A MURINE HEMORRHAGE
MODEL

W5:

The differential expression of inflammatory cytokines between PMO and
KC in a murine fixed pressure model of hemorrhage may be due to a difference
in the amount of relative stress seen between these two macrophage
populations. This stress is reflected by a difference in the expression of the
inducible 72 kD heat shock protein (hsp72). Therefore, increased hsp72
expression may preserve the capacity of KC to mount a cytotoxic response and

release inflammatory cytokines in response to hemorrhage.

S '[i . ,
The previous discussion leads to some interesting general questions
regarding heat shock protein expression and function during as a result of
hemorrhage:
- Is there a measurable difference in the inducible 72 kD heat shock
protein (hsp72) expression during hemorrhage from baseline?
- What is the relationship between cytokines synthesis and heat shock
protein expression?
- What is the role, protective or otherwise, of heat shock proteins in
hemorrhage?
- Is there co—induction of gene expression of some cytokines and the
stress proteins during hemorrhage and/or sepsis and what is the

role of stress proteins in this response?

15

16
In this experiment we hope to:

1.) Compare differences in inducible murine hsp72 expression of peritoneal
macrophages (PMO) and Kupffer Cells (KC) in a fixed pressure model of
hemorrhage and the relationship to known differences in cytokine expression
0.9., TNF—alpha and IL-6) in the same model

2.) Discuss the teleological and possible functional significance of this
difference in response and how it impacts on the global immunosuppression

that results from pure hemorrhagic shock.

IV. MATERIALS AND METHODS

HEMORRHAGE MODEL:

A fixed—pressure murine model of hemorrhagic shock was used as
previously described!“- 50 All protocols were carried out in accordance with the
guidelines set forth in the Animal Welfare Act and as outlined in the Guide for
Care and Use of Laboratory animals by the National Institutes of Health
Publications. This project was approved by the Institutional Review Board and
Research Committee of Butterworth Hospital and the Institutional Animal Care
and Use Committee of Michigan State University.

Briefly, 8-10 week old inbred male C3H/HeN mice (Charles River
Laboratories, Portage, MI) were randomly divided into hemorrhage (HEM) and
control sham operation groups (SHAM). Prior to randomization and
experimentation, the animals were fasted overnight but allowed water ad
Iibitum. The HEM group was anesthetized with methoxyflurane inhalation,
shaved, prepped sterilely, and underwent bifemoral artery cannulation with P-
10 tubing. They were then bled to and maintained at a mean arterial pressure
of 35 mm Hg for one hour as measured by calibrated pressure tracings. After
one full hour at stable hypovolemia the HEM group was adequately resuscitated
with the shed blood plus twice that volume balanced electrolyte solution
(Ringer's lactate). SHAM groups underwent cannulation of both femoral
arteries in a similar manner without hemorrhage / resuscitation. HEM and
SHAM groups were then randomly divided into groups for procurement of KC

and PMO at 2 and 24 hours.

17

MACROPHAGE CELL PREPARATIONS:

KC and PMO are harvested at the time of sacrifice for each of the groups

of animals using aseptic techniques and isolated as follows:

E l' I 'I l I ,

A peritoneal lavage technique was utilized as previously described by
Ayala et al.50 Briefly, this consisted of harvesting the PMO by peritoneal lavage
using ice-cold Dulbecco's minimal essential media with 10% fetal calf serum
(DMEM, Gibco, Grand Island NY, see Appendix A). Cells were washed once
(2809, 15 min., 4°C) with Hank's balanced salt solution (HBSS, see Appendix
B). 1 x 107 PMO from animals within each group were plated into tissue culture
bottles with 10 ml RPMl-1640 (Gibco, see Appendix C) with 10% fetal calf
serum for a final concentration of 1 x 106 cells / ml in preparation for cytokine
and Western blot analysis. After 3 h at 37°C in a 5% 002 atmosphere, non—
viable and non-adherent cells were removed by 4 sequential gentle washings.
Cell yields and viability as determined by trypan blue exclusion, were compared
for SHAM and HEM groups.

EmmnmcLKunfleLcells:

An in situ collagenase digestion method was utilized as previously
described by Ayala et al. (23). Briefly, after infusion of the liver with ice-cold
HBSS, followed by warm DMEM with collagenase, the liver was resected en
bloc and incubated 10 minutes at 37°C in a humidified incubator. After mincing
and resuspension with, sequentially, a standard 25 ml and 10 ml pipet, the cells
were pelleted, collected, and washed (4°C, in RPMl-1640 with 10% fetal calf

18

19
serum, at 280 g). Parenchymal and non-parenchymal cells were separated

from each other by centrifugation over Metrizamide (Accurate Chem. Corp.,
Westbury, NY) and the cells at the interface pooled and washed twice by
centrifugation (2809, 20 min., 4°C) with cold media with 10% fetal calf serum
(FCS). Viability via trypan blue exclusion and yields were analyzed after the
cell pellet is dispersed in and resuspended in RPMl-1640 media. Then 1 x 107
KC from animals within the individual groups were plated into tissue culture
bottles (1 x 106 cells / ml) in preparation for cytokine , Northern blot , and
Western blot analysis as described for PMD above. After 5 hrs at 37°C in a
humidified, 5% 002 atmosphere, non-viable and non-adherent cells were
removed by 4 sequential gentle washings. Cell yields and viability were
compared for SHAM and HEM groups. The cells from each group were
washed and then stimulated for 24 hrs with LPS to induce cytokine production.
After the supematants were collected for TNF-a and lL—6 bioassays, the cells
were lysed for protein extraction (see below). The cells were then prepared for
protein electrophoresis and Western blot analysis using a chemiluminescent

HSP 70 probe

[2 | l . | . .

Interleukin-6 activity in the culture supematants were determined by the
amount of proliferation of the murine B-cell hybridoma cell-line 7TDI, which
grows only in the presence of lL-6.5° 4 X 103 7TDI cells/ml were grown for 72
hours at 37°C in 5% C02 after stimulation with serial dilutions of serum or
macrophage supematants. For the last 4 hours of incubation, 20 ul of a 3-(4,5-
dimethylthiazol-2-L)-2,5-diphenyltetrazolium-bromide solution (MTT; 5 mg/ml in
RPMl-1640, Gibco) were added to each well. Only viable cells incorporate
MTT. The assay was stopped by aspiration of 150 pl supernatant from each

20
well, with subsequent replacement by 150 pl of a 10% solution of sodium

dodecyl sulfate in PBS (Lauryl Sulfate, Sigma Chemical Co., St. Louis, M0) to
dissolve the dark-blue forrnazan crystals. Using an automated microplate
reader (EL-311, Bio-Tek Instruments Inc., Winooski, VT) the light-absorbence
for each well was measured at 595 nm. The amount of lL—6 in supernatant is
directly proportional to the proliferation of 7TD1 cells.

TNF in culture supematants was determined by the inherent cytotoxic
effect on the fibrosarcoma cell line WEHI 164 subclone 13.58- 59 The
proliferation of unlysed WEHI cells was measured, as described above, using a
MTT assay. The amount of TNF in supernatant is indirectly proportional to the
proliferation of WEHI cells.

In the bioassays described, the units of cytokine activity per milliliter of
supernatant were determined by comparison to standard curves produced from
dilutions of a purified recombinant lL—6 standard (200 U/ml, Amgen Corp.,
Thousand Oaks, CA) or a murine TNF standard (200 U/ml, Amgen Corp.) after
the methods described by Mizel.‘0

EI' "I. I I'll"
After heat shock or hemorrhage, cell samples are prepared for protein

electrophoresis. After washing three times with PBS at the end of any

individual treatments, the cell collections are lysed by the addition of ice cold

lysis buffer:
50 mM Tris (pH 8.0) 2 mM phenylmethylsulfonylurea
150 mM NaCl 0.5 % Triton X—100
5 mM EDTA

After addition of lysis buffer, the Iysates are incubated on ice for 30 min. The

Iysates were clarified by microcentrifugation and the supernatants stored at

21
-70°C until ready for analysis. Protein concentrations were determined using a

modified Lowery protein essay.

I" EEEEII I .,

Protein samples were prepared for tricine gel electrophoresis by first
heating similar quantities of protein for each sample at 85°C in 1X sample
buffer and then running in a precast gel (10% acrylamide l2.6%Bis—acrylam ide,
Novex®, San Diego, CA) for approximately 1 hr. at 100V constant voltage.
After electrophoresis, the gel contents are electroblotted to nitrocellulose
membranes for Western blot analysis. The gel is then fixed and stained with
Coomassie blue.

The prepared protein blots are blocked with a 5% milk solution, washed,
and then incubated at 4°C with primary antibody for hsp72 (murine anti hsp72
MAb, lgG1,SPA 810, StressGen Biotechnologies Corp., Victoria, BC,
Canada). After this, the membranes are again washed and then incubated with
a secondary antibody conjugated to peroxidase for a luminol—based
chemiluminescence reaction (ECL systems, Amersham, lnc., Manchester,
68.). At the time of chemiluminescence, the membranes were exposed to
autoradiography film (X—Omat AR, Eastman Kodak, Rochester, N.Y.). The
exposures were assessed for band intensity by densitometric analysis with a
JAVA image analyzer (Jandel Scientific, San Rafael, CA). The band intensity
was directly proportional to quantify.

22
Statistics:

Values are presented as means i SEM for continuous variables or as
percentages / ratios for discrete variables. Differences between groups were
compared using analysis of variance (ANOVA) and Fisher's protected least
significant difference (PLSD). Statistical significance was recognized at p <
0.05.

V. Results

MACROPHAGE CYTOKINE PRODUCTIVE CAPACITIES IN RESPONSE TO
HEMORRHAGE

 

Trypan blue exclusion for cell viability discriminates viable cells (no dye
uptake) from non-viable cells (positive dye uptake). Our analysis revealed that
no significant differences existed in the percentage of viable cells of the total
number isolated whether the animal was subjected to sham operation or
hemorrhage / resuscitation.

WM

Kupffer cells isolated from hemorrhaged mice showed an increased
capacity to release TNF-ct (see Figures 1 and 2) and IL—6 (see Figures 3 and
4) when compared to controls. At 2 hours, Kupffer cell TNF-a productive
capacity from hemorrhaged animals (6 animals, mean = .048 :I: .0098
units/mI/105 cells) was increased 20-fold when compared to sham—operated
animals (7 animals, mean = .002 :l: .0004). This was statistically significant with
p < 0.05 (ANOVA, Fisher's PLSD). This was true also at 24 hours for Kupffer
cells from hemorrhaged (7 animals, mean = .110 :l: .0130) versus sham-
operated animals (10 animals, mean = .001 i .0002) and was also statistically
significant with p < 0.05 (ANOVA, Fisher's PLSD).

IL—6 productive capacities for Kupffer cells were similarly affected. At 2
hours Kupffer cells from the hemorrhage group (6 animals, mean = 23.5 :I: 7.08)

had a significantly increased lL—6 productive capacity when compared to shams

23

24
shams (7 animals, mean = 4.9 i: 1.03). At 24 hours this difference remained

significant as the hemorrhage group (7 animals, mean = 12.6 :I: 1.56) had a six-
fold increase in lL—6 productive capacity as compared to shams (10 animals,
mam=2Ji0An.

As previously described in 'Methods and Materials,‘ productive
capacities are initiated by stimulation with LPS. Therefore, this does not
represent circulating inflammatory cytokine levels but is an in vitro measure of
where the circulating cytokines are produced. Although the absolute values of
TNF are much smaller than lL—6 values, the measured values are consistent,

as represented by the standard errors of the mean (SEM).

W

PMO at 2 and 24 hours showed no significant differences in the capacity
to release TNF-ct (see figure 5 and 6) and IL-6 (see figure 7 and 8) when
comparing hemorrhaged and sham groups.

At 2 hours, peritoneal macrophage TNF-ct productive capacity from
hemorrhaged animals (6 animals, mean = 17.6 :I: 3.71 units/mll105 cells) was
not significantly different compared to sham-operated animals (7 animals,
mean = 12.4 :I: 1.68). At 24 hours, hemorrhaged (7 animals, mean = 3.5 :l: 0.31)
versus sham-operated animals (10 animals, mean = 3.3 :l: 0.64) showed no
significant differences in peritoneal macrophage TNF-a productive capacity.

lL-6 productive capacities for peritoneal macrophages were similarly
affected. At 2 hours the hemorrhage group (6 animals, mean = 1206.5 :l: 100.8)
had no significant differences in lL—6 productive capacity when compared to
shams (7 animals, mean = 1131.7 i 222.7). At 24 hours, there were also no
significant differences between the hemorrhage group (7 animals, mean =

580.1 :I: 72.9) and shams (10 animals, mean = 446.7 1 122.8).

25
These results mirror those previously described in our lab. It establishes

the validity of this model for the experiments involving the expression of hsp 72,
which is described in the next section.

26

 

 

 

 

Cell Type in) TNF-a * IL—6 *
HemPer 2 (6) 17.6 i 3.71 1206.5 i 100.77
ShamPer 2 (7) 12.4 :I: 1.68 1131.7 1 222.67
HemPer 24 (7) 3.5 :I: 0.31 580.1 :I: 72.95
ShamPer 24 (10) 3.3 :I: 0.64 446.7 :I: 122.77
HemKup 2 (6) 0.0480 at 0.0098 23.5 :I: 7.08
ShamKup2 (7) 0.0022 1 0.0004 4.9 :t 1.03
HemKup 24 (7) 0.1100 :I: 0.0130 12.6 i 1.56
ShamKup24 (10) 0.0010 i 0.0002 2.1 :I: 0.47

 

Table 1: Summary of results representing the productlve capacltles for
TNF-or and IL-6 from PMD (Per) and KC (Kup) In control (Sham) and

hemorrhage groups (Hem) at 2 and 24 hours. Number per group In
parentheses. * All values means 1: SEM In unltslmllIOs cells.

 

27

 

TNF (units/ml)

 

 

 

HemKup 2 ShamKup 2
n=6 n=7

Figure 1: TNF-ct productlve capacities of Kupffer cells 2 hours after
hemorrhage and resuscitation. Mean 1: SEM, n = # of animals In each
group.

' p < 0.05 (ANOVA, Fisher‘s PLSD).

28

 

 

 

 

HemKup 24 ShamKup 24
n=7 n=10

Flgure 2: TNF-a productlve capacities of Kupffer cells 24 hours after
hemorrhage and resuscltatlon. Mean :12 SEM, n a a of animals In each
group.

* p < 0.05 (ANOVA, Fisher's PLSD).

29

 

lL—6 (units/ml)

 

 

 

 

5 - l___L___|
o - .
HemKup 2 ShamKup 2
n=6 n=7

Flgure 3: IL-6 productlve capacities of Kupffer cells 2 hours after
hemorrhage and resuscitation. Mean i SEM, n a t of anlmals In each
group.

' p < 0.05 (ANOVA, Flsher's PLSD).

 

  

IL-6 (units/ml)

 

 

 

HemKup 24 ShamKup 24
n=7 n=10

Flgure 4: IL-6 productlve capacities of Kupffer cells 24 hours after
hemorrhage and resuscitation. Mean :t SEM, n = i of anlmals In each
group.

* p < 0.05 (ANOVA, Flsher's PLSD).

 

20‘

 

17.5 ‘

.s
0|
1

 

12.5 "‘

 

 

TNF (units/ml)
\V‘

 

 

 

 

 

O

HemPer 2 Sham Per 2
n=6 n=7

Flgure 5: TNF-a productlve capacltles of perltoneal macrophages 2
hours after hemorrhage and resuscltatlon. Mean :I: SEM, n = s of anlmals
In each group. No dlfference, by ANOVA, Flsher's PLSD.

 

 

 

 

 

 

 

 

 

 

q ‘ ‘ q ‘ q 1 - 1 - ‘ 1‘ q ‘
555555555

Flgure 6: TNF-a productlve capacltles of peritoneal macrophages 24
hours after hemorrhage and resuscitatlon. Mean :I: SEM, n =

t of animals

Flsher's PLSD.

Ifference, by ANOVA,

In each group. No d

 

'5

_e
N
8

l

 

'é’

lL—6 (units/ml)
s s s S
a
83

 

 

 

 

 

O

HemPer 2 ShamPer 2
n=6 n=7

Figure 7: IL—6 productlve capacltles of peritoneal macrophages 2 hours
after hemorrhage and resuscitation. Mean :t SEM, n a s of animals In each
group.

No dlfference, by ANOVA, Flsher's PLSD.

 

 

 

 

 

 

 

 

 

 

 

:2: l
“:1 4 .

Figure 8: lL—6 productlve capacltles of peritoneal macrophages 2 4 hours
after hemorrhage and resuscitation. Mean 3: SEM, n s s of animals In each
group. No difference, by ANOVA, Flsher's PLSD.

HSP72 EXPRESSION IN RESPONSE TO HEMORRHAGE

KunfleLQells;

Hsp72 levels in Kupffer cells were significantly elevated in response to
hemorrhage at both 2 hours (see Figure 9) and 24 hours (see Figure 10) when
compared to controls(see Table 2). At 2 hours, hsp72 expression in Kupffer
cells were significantly increased in the hemorrhage group (8 animals, mean =
161.6 1 18.7 intensity units x 1000) versus shams (7 animals, mean = 50.4 :I:
5.8). This was statistically significant at p < 0.05 (ANOVA, Fisher's PLSD). At
24 hours post-hemorrhage, this difference remained statistically significant
between hemorrhage (7 animals, mean = 76.1 :I: 15.4) and sham groups (7

animals, mean = 39.9 3: 2.7) for Kupffer cell hsp 72 expression.

W

No significant differences were seen in PMO between HEM and SHAM
groups at either 2 (Figure 11) or 24 hours (Figure 12). It also appears that
although there is some diminution of hsp levels at 24 hours, differences still
remain significant. At 2 hours, peritoneal macrophages from hemorrhaged
animals (6 animals, mean = 21.9 :I: 1.3) exhibited equivalent hsp72 expression
when compared to shams (3 animals, mean = 22.9 i: 0.8). At 24 hours, there
were still no significant differences noted between hemorrhaged animals (6
animals, mean = 29.9 :49) and shams (6 animals, mean = 28.3 i: 4.8).

Sample Western blots exhibiting the differences in expression of hsp72
in the two cell populations are shown in Figures 13 (Kupffer cells) and 14
(peritoneal macrophages). Reference values from a known molecular weight

standard on the original tricine PAGE gel were used to confirm the size of the

35

36
band of interest. By setting the background signal to a standard intensity

(zero), using consistent amounts of protein per gel lane (5 ng), and
standardized exposure times during the zenith of chemiluminescence (5

seconds), we were able to obtain semi-quantitative data which could be used

for comparison in our experiments.

37

 

 

 

Cell Type Time (hrs) Hemorrhage' (n) Sham' in)

KC 2 161.6 :I: 18.7 (a) 50.4 :I: 5.8 (7)

PMQ 2 21.9 :I: 1.3 (6) 22.9 :I: 0.8 (3) NS
KC 24 76.1 :I: 15.4 (7) 39.9 1 2.7 (7) s
PMQ 24 29.9 :1: 4.9 (6) 28.3 :I: 4.8 (6) NS

 

Table 2: fSummary of results comparing the expression of hsp72 at 2

and 24 hours for Kupffer cells (KC) and peritoneal macrophges (PMD).

Number per group In parentheses.’ All values means 2 SEM In density

units x 1000.

P: Significance set at p < 0.05 (ANOVA, Flsher's PLSD); S = significant, NS
= not significant.

 

 

 

  

Intensity (x 1000)

08888

 

 

 

 

 

 

HemKup2 ShamKup2
n=8 n=7

Figure 9: hsp 72 expression of Kupffer cells 2 hours after hemorrhage
and resuscitation. Mean :I: SEM, n a s of animals In each group.
1 p < 0.05 (ANOVA, Flsher's PLSD).

 

 

 

 

 

 

 

 

100
8 9°: .
O ..
:3 3°,
F
x 70‘
V I
60 -l
2? 5°:
Q . L
: 1°.
3 30'
c J
— 20 «II
10 -
o s
HemKup24 ShamKup24
n=7 n=7

Figure 10: hsp 72 expression of Kupffer cells 24 hours after hemorrhage
and resuscitation. Mean :t SEM, n = if of animals In each group.
* p < 0.05 (ANOVA, Flsher's PLSD).

 

8

'0
0|
1

 

(x1000)

.e
O
l

Intensity

 

 

 

 

 

 

W

HemPer2 ShamPer2
n=6 n=3

Figure 11: hsp 72 expression of peritoneal macrophages 2 hours
hemorrhage and resuscitation. Mean :I: SEM, n a s of animals In eac
group.

No difference by ANOVA, Flsher's PLSD.

41

 

 

(x1000)

Intensity
a

 

 

 

 

 

 

HemPer24 snameer24
n=6 n=6

Figure 12: hsp 72 expression of peritoneal macrophages 24 hours after
hemorrhage and resuscitation. Mean :I: SEM, n = if of animals In each
group.

No difference by ANOVA, Flsher's PLSD.

42

 

 

 

 

Figure 13: Representative sample of chemiluminescent radiographs from
hsp72 Western blot analysis: Kupffer cells. Lanes 1 and 3 — extracts from
Kupffer cells 2 hours post - hemorrhage; lane 2 - extract from Kupffer
cells 2 hours after sham operation; lane 4 - blank; lane 5 - extract from
Kupffer cells 24 hours after sham operation; lane 6 — extract from Kupffer
cells 24 hours post — hemorrhage. Arrow — hsp72.

-..__——

43

 

Figure 14: Representative sample of chemiluminescent radiographs from
hsp72 Western blot analysis: peritoneal macrophages. Lanes 1 and 3 —
extracts from peritoneal macrophages 2 hours after sham operation;
lanes 2 and 4 - extracts from peritoneal macrophages 2 hours post -
hemorrhage; lane 5 - extract from peritoneal macrophages 24 hours after
sham operation; lane 6 - extract from peritoneal macrophages 24 hours
post - hemorrhage.

Arrow - hsp72.

VII. DISCUSSION

HEMORRHAGE AND ITS IMMUNOLOGIC CONSEQUENCES

Hemorrhage, and injury in general, results in a cascade of events that
are intended to act as a measure of defense for the host. These include, but
are not limited to, a metabolic response reflected by an increase in
counterregulatory hormones, corticosteroids, and catecholamines;
implementation of nonspecific inflammatory responses mediated by kinins, the
complement sytem, coagulation, and the reticuloendothelial system; and an
immune response mediated by a complex network of macrophages, T—cells, B-
cells and their inflammatory mediators (e.g., cytokines, arachidonic acid
metabolites, and colony-stimulating factors).47- ‘1 However, the result of this
activation is often injurious given the correct environment. As such, the final
common pathway is often multiorgan failure, with mortality directly related to the
number of dysfunctional systems of the affected host.“2

It appears that many of the functional aspects of the immune sytem that
act in a protective manner for the host also act to damage many of the
structural elements that support homeostasis. It is clear that when this situation
is combined with any element of immune dysfunction, such as depressed
antigen presentation by macrophages and depressed phagocytic function, the
host is at risk for sepsis caused by secondary infection from endogenous as
well as exogenous sources. Chaudry and Ayala have clearly outlined the
mechanisms believed to play a role in the breakdown of normal immune

function that leads to depressed lymphocyte and macrophage functions and

44

45
immunosuppression“- ‘4 Hemorrhage is believed to cause regional hypoxia

mediated by decreased oxygen delivery secondary to hypovolemia and
preferential shunting away from non-critical organs. This situation is further
exacerbated by early systemic mediators (e.g., arachidonic acid metabolites,
catecholamines, the acute—phase response) that initially act to counter the
initial insult. Regional hypoxia, in turn, initiates a cycle of cellular derangements
in energy stores (i.e.: ATP) and calcium homeostasis that may stimulate an
increase in inflammatory cytokines which propagate the host's mediator
response and amplifies it. It appears that this amplification leads to further
suppression of macrophage and lymphocyte function. This is reflected in the
decreased ability of macrophages to present antigen, decreased lymphokine
activation important for the coordination of a competent cell— and humoral—
mediated immune response, and an increase in the elaboration of potent
immunosuppresive agents such as transforming growth factor beta (T GF-B)
and platelet activating factor (PAF).55- 5‘

It is apparent that the inflammatory cytokines playing large roles in this
cascade of events are lL—1, lL—6, and TNF—0t. As was indicated earlier, these
cytokines are upregulated in response to hemorrhage / resuscitation in a
differential manner among two different cell populations: increased productive
capacities in Kupffer cells and decreased in peritoneal macrophages. Despite
this difference, there is a global effect on many mononuclear cells for
decreased antigen presentation.“- 53- 33 Reasons for these differences may be
due to intrinsic as well as environmental factors.

Kupffer cells lie in a matrix rich in immunodulating factors. They are
exposed to bacterial by—products from the splanchnic circulation, receive
signals from surrounding hepatocytes, clear toxins, and act as accessory cells

in the induction of antigen-specific immune responses”, 53 As such, they exist

46
in a much more dynamic immunomodulatory milieu than do peritoneal

macrophages. It may be surmised that signals received by the Kupffer cell
acting to increase its cytotoxic abilities and ability to elaborate cytokines in an
amplified manner are inherently different as a result of this difference in
environment. However, it is not clear what mechanisms are affecting the ability
of the Kupffer cell to maintain its cytotoxic nature despite a supression of
antigen presentation. One mechanism may be its increased expression of

hsp72.

DIFFERENTIAL EXPRESSION OF HSP72 FROM KUPFFER CELLS AND
PERITONEAL MACROPHAGES IN RESPONSE TO HEMORRHAGE]
RESUSCITATION

These experients show that there is a differential expression of the heat
shock response, as measured by hsp72, between Kupffer cells and peritoneal
macrohages. In response to hemorrhage, shock, and resuscitation, Kupffer
cells exhibit an increased expression of hsp72 when compared to shams.
Peritoneal macrophages do not exhibit any increase in hsp expression in
response to the same insult. When coupled with the findings of differential
inflammatory cytokine productive capacities as previously outlined, one can
infer that the ability to mount a heat shock response may be a protective
mechanism which preserves the cytotoxic abilities of Kupffer cells but not
peritoneal macrophages. The reasons for this may again be due to the
environment of the Kupffer cell during hemorrhage / resuscitation.

During the resuscitative phase of our model, the liver is exposed to many
of the stimuli of the heat shock response including bacterial by products and
endotoxin, oxygen radicals, and reperfusion mediators“. 42» ‘7 This is in
addition to the potent stimulus of regional hypoxia induced in the liver
secondary to the low flow state of acute hemorrhage. Given its metabolic load,
it is easy to surmise that significant reductions in blood flow and oxygen
delivery to the liver will have more profound effects than a similar reduction to
peritoneal cells. Therefore, the effective stress response will be significantly
more magnified in the liver parenchyma than in the peritoneal lining.

Teleologically, given the inherent cytoprotective properties of the heat

shock response, it is likely that certain cell populations able to mount such a

47

48
response will preserve more of their natural cell function than those unable to

mount a heat shock response. Therefore, the preservation of cytotoxic
responses in Kupffer cells may be related to its ability to express hsp72. What
remains to be elucidated are the specific genetic mechanisms that are involved
in this phenomenon.

Previous work by Udelsman and associates 22- 37: 3" 3' as well as
by Buchman and associates 31, ‘0: 41 had shown measurable increases in the
mRNA of the hsp70 family of proteins in response to surgical stress, restraint
stress, cardiogenic shock, and hemorrhage (see 'The stress response', pp. 8—
10). These investigators demonstrated single organ elevations of the message
and postulated genetic mechanisms mediated by specific transcription factors
such as HSF—I that are specific to certain stress stimuli. In our study, we have
demonstrated that a specific cell population within the reticuloendothelial
system, Kupffer cells, exhibit increased expression of hsp72 by Western blot
analysis. Our attempts at Northern blot analysis of hsp72 mRNA expression
showed that, in some prliminary data, there was an increased expression of this
message in response to hemorrhage within liver as a whole and from some of
the isolated Kupffer cell samples. Unfortunately, our results were too
inconsistent to make any meaningful interpretation possible and are not
included in this thesis.

Further work in this area should be directed to genetic mechanisms
involved in the elucidation of the heat shock response. It is quite possible that
potential mechanism involved in the differences between Kupffer cells and
peritoneal macrophages in our specific model may be due to an upregulation in
the transcription factors responsible for initiating and amplifying the production
of hsp72 in Kupffer cells. The effect of this would be to augment the differences

in environmental factors affecting the two cell populations resulting in an

49
increased heat shock response in Kupffer cells. It is hoped that future

investigators with an interest in this area may be able to demonstrate these

differences at the molecular level.

VIII. SUMMARY AND CONCLUSIONS
SUMMARY

This project confirmed that hemorrhage and resuscitation in mice
resulted in a significantly increased capacity to release TNF—0t and IL—6 from
Kupffer cells when compared to sham-operated controls. This was not
reflected by a similarly, significant upregulation of inflammatory cytokine release
in peritoneal macrophages from the same animals (see Figures 1 to 8 and
Table 1). Similar results were previously described by Ayala et al. in the same
hemorrhage and resuscitation model“- 53» 54 In these studies, it was shown
that hemorrhage and resuscitation resulted in a significantly and similarly
decreased capacity for antigen presentation among both PMO and KC and
susceptibility of the subject to early death from sepsis when compared to
sham-operated controls.

This study, in addition, showed that differences in cytokine productive
capacities are mirrored by differences in hsp72 expression between PMO and
KC in response to hemorrhage. Here the response of KC is an increased
expression of hsp72 as a result of hemorrhage and resuscitation in 03H I HeN
mice when compared to sham-operated controls (Figures 9 to 12 and Table 2).
Although an association between increased cytokine productive capacities and
increased hsp72 expression exists temporally within Kupffer cells of mice
undergoing hemorrhage / resuscitation, it is uncertain mechanistically whether

or not these two phenomena are related.

50

CONCLUSIONS

These studies demonstrated differential effects of hemorrhage /
resuscitation on two specific cell populations. Kupffer cells and peritoneal
macrophages experience marked cellular immune dysfunction. However,
inflammatory cytokine (IL-6 and TNF—0r) production is markedly increased in
Kupffer cells compared to peritoneal macrophages. This is mirrored by an
increased ability to mount a heat shock response in the former. It is
conceivable that the depression of cellular immunity in the face of differential
inflammatory cytokine productive capacities occur through separate but
temporally related mechanisms. However, in light of the findings of this project,
it is also likely that preservation of cytokine productive capacities in Kupffer
cells in this model is related to the ability to mount a heat shock response.

Peritoneal macrophages are unable to mount a heat shock response
equal to that of Kupffer cells. The reasons for these differences and their
consequences may be as follows:

1. There are relative differences in the stress experienced by the two
different cell populations based on differences in the environment
of each of the cell types and the relative effects of hemorrhage on
each environment. '

2. Reperfusion injury may have a bigger impact on the liver and its
constituents, especially Kupffer cells, which is reflected by the
more intense heat shock response in Kupffer cells.

3. The heat shock response elicited in the Kupffer cell population may
be a protective mechanism preserving basic cellular functions in

the face of an injurious insult.

51

52
4. The preservation of function in Kupffer cells is not 'complete',

resulting in a depression of antigen presentation despite an
increased inflammatory cytokine productive capacity and cell
cytotoxicity.

5. The critical mechanism for the expression of the heat shock response
may be that specific signals required to upregulate transcription
factors needs to complete the initiation and amplification of hsp72
mRNA production in our model exist in the splanchnic
environment of the liver during hemorrhage I resuscitation but are

inadequate within the circulation of the peritoneum.

UMITATIONS AND FUTURE STUDIES

The findings of this study are unique in that previous work had not
demonstrated increased heat shock response in one cell population from an
organ as a response to severe stress such as hemorrhage I resuscitation.
Furthermore, our findings clearly show a difference in mononuclear cell activity
and function depending on the local environment that cell was in during the
inception of stress. However, although there is an association of increased
cytokine productive capacity to hsp72 expression in response to hemorrhage,
the mechanisms and causal relationships need to be worked out.

It can be argued that because LPS stimulation is used in the cytokine
bioassays the expression of hsp72 from the two cell populations of interest in
this study also need to be treated with LPS. This is a valid point. It would be of
vital interest to know if LPS pretreatment has any effect on the expression of
hsp72 in our model. This would then show whether or not hsp72 production is
also stimulated by the same mechanisms in play during the stimulation phase
of the cytokine bioassays. If hsp72 is upregulated equally along all cells and
treatments (i.e. sham or hemorrhage), then this upregulation is a background
increase that does not necessarily invalidate our conclusions. If there are
differential alterations in hsp72 expression due to LPS stimulation that does not
parallel our findings, we must rethink the mechanisms involved in the cytokine
bioassay and hold off on the notion that cytokine expression and the heat shock
response are necessarily linked. However, as a point of reference, our data do
show the intrinsic levels of heat shock proteins as a result of the stress of
hypovolemlc shock due to severe hemorrhage followed by acute resuscitation

in peritoneal macrophages and Kupffer cells are significantly different. This

53

54
difference may be due to differential responses to a similar stress in two

different cell populations insofar as their respective heat shock responses are
concerned.

These limitations in our study form the basis of possible future studies in

this model:

1. Elucidate the genetic expression of hsp72 production, whether by
Northern blot analysis or amplification via polymerase chain
reaction (PCR).

2. Gene knockout experiments, whether by antibody binding to hsp72 or
antisense RNA prior to stimulation for cytokine release, may be
able to delineate whether or not hsp72 is required for cytokine
production or if they are even linked mechanistically.

3. Perform the experiments where LPS is used prior to harvesting
protein for Western blot analysis and compare to assays where
there is no pretreatment with LPS.

These points, in conjunction with studies of the mechanisms of hsp72
upregulation at the transcriptional level, may further our understanding of the
reasons why differential alteration of macrophage function in response to
hemorrhage / resuscitation leads to a relative global depression of immune

function in the affected organism.

APPENDIX A

APPENDIX A

 

 

 

Arginine HCL 84.0
Cystine 48.0
Glutamine 584.0
Glycine 30.0
Histidine.HCl.HzO 42.0
Isoleucine 104.8
Leucine 104.8
Lysine.HCl 146.2
Methionine 30.0
Phenylalanine 66.0
Serine 42.0
Threonine 95.2
Tryptophan 16.0
Tyrosine 72.0
Valine 93.6
Vitamins melml
Choline Chloride 4.0
Folic acid 4.0
Nicotinamide 4.0
Calcium d—pantothenate 4.0
Pyridoxal.HCl 4.0
Riboflavin 0.4

55

56

 

Thiamine HCI 4.0
Inositol 7.0
WILL mall
NaCl 6400.0
KCI 400.0
NaH2P04.2H20 125.0
MgSO4.7H20 200.0
CaCl2 200.0
NaHCOa 3700.0
Fe(N03)3.9H20 0.1
Other Components
n—Butyl—p—hydroxybenzoate 0.2 mg/l
Glucose 4500.0 mg/l
Penicillin G potassium 5000 units
Phenol red 15.0 mg/l
Sodium pyruvate 110.0 mg/l
Streptomycin sulfate, equivalent base 100.0 mg/l

APPENDIX B

APPENDIX B

 

 

W
1.280110% infill
Arginine 200.0
Asparagine 50.0
Aspartate 20.0
Cystine 50.0
Glutamate 20.0
Glutamine 300.0
Glycine 10.0
Histidine 10.0
Hydroxyproline 20.0
Isoleucine 50.0
Leucine 50.0
Lysine.HCl 40.0
Methionine 15.0
Phenylalanine 15.0
Proline 20.0
Serine 30.0
Threonine 20.0
Tryptophan 5.0
Tyrosine 20.0
Valine 20.0
MitammL mall
Biotin 0.2

57

58

 

812 0.005
Calcium pantothenate 0.25
Choline Chloride 3.0
Folic acid 1.0
Nicotinamide 1.0
p—aminobenzoic acid 1.0
Pyridoxine.HCl 1.0
Riboflavin 0.2
Thiamine HCI 1.0
102mm mall
NaCl 6000.0
KCI 400.0
NagHPO4.7H20 1512.0
MgSO4.7H20 100.0
NaHCOa 2000.0
Ca(N03)2.4H20 100.0
Other Components
Glucose 2000.0
Glutathione (reduced) 1.0

Phenol red 5.0

APPENDIX C

APPENDIX C

E | l' I. ,
Kam DM, Ayala A, Chaudry IH. Insights into the mechanism of
increased Kupffer cell cytokine release in response to hemorrhage: increased

expression of HSP 72. Surgical Forum, 1995; 46: 152-154.

mm

Kam DM, Ayala A, Chaudry IH. Insights into the mechanism of
increased Kupffer cell cytokine release in response to hemorrhage: increased
expression of HSP 72. Presented at:

1.) American College of Surgeons Surgical Forum, New Orleans, LA
(1995)

2.) Society of University Surgeons, Resident Section, Denver, CO (1995)

3.) Michigan Chapter American College of Surgeons, Troy, MI (1995)

4.) GRAMEC Research Day, Grand Rapids, MI (1995). Winner, Best
Trauma Study

59

BIBUOGRAPHY

IX. BIBLIOGRAPHY

1.) Ritossa F. Chromosomal puffing after heat-shock in Drosophila.
Experientia. 1962; 18: 571-573

2.) Welch WJ. Mammalian stress response: cell physiology, structure function
of stress proteins, and implications for medicine and disease. Physiologic Rev.
1992; 72: 1063—1081

3.) Tissieres A, Mitchell HK. Tracy UM. Protein synthesis in salivary glands of
Drosophila melanogaster: Relation to chromosomal puffs. J Mol Biol. 1979;
84: 389-398.

4.) Lindquist S, Craig EA The Heat Shock Proteins. Annu Rev Genet. 1988;
222: 631 - 77.

5.) Ananthan J, Goldberg AL, Voellmy R. Abnormal proteins serve as
eucaryotic stress signals and trigge rthe activation of heat shock genes.
Science. 1986; 232: 522 - 24.

6.) Donati YRA, Slosman 00, Polls BS. Oxidative injury and the heat shock
response. Biochem Pharrnacol. 1990; 40: 2571 - 77.

7.) Mizzen LA, Welch WJ. Characterization of the therrnotolerant cell. I: Effects
on protein synthesis activity and the regulation of heat-shock protein 70
expression. J Cell Biol. 1988; 106: 1105 - 16.

8.) Morimoto RI. Heat shock: the role of transient inducible responses in cell
damage, transformation, and differentiation. Cancer Cells. 1991; 3: 295—301.

9.) Riabowol KT, Mizzen LA, Welch WJ. Heat shock is lethal to fibroblasts
microinjected with antibodies against hsp70. Science. 1988; 242: 433 — 6.

10.) Johnston RN, Kucey BL. Competetive inhibition of hsp70 gene causes
then'rlosensitivity. Science. 1988; 242: 1551 -4.

61

11.) Yellon DM, Pasini E, Cargnoni A, Marber MS, Latchman 08, Ferrari R.
The protective role of heat stress in the ischaemic and reperfused rabbit
myocardium. J Mol Cell Cardiol. 1992; 24: 895 — 907.

12.) Yellon DM, Marber MS. HSP70 in myocardial ischemia. Experientia.
1994; 50: 1075 - 84.

13.) Yellon DM, lliodrimitis E, Latchman DS, VanWinkle DM, Downey JM.
Whole body heat stress failed to limit infarct size in the reperfused rabbit heart.
Cardiovascfies. 1992; 26: 342 - 6.

’ _V.L"

14.) Villar J, Edelson JD, Post M, Mullin JB, Shutsky AS. Induction of heat
stress proteins is associated with decreased mortality in an animal model of
acute lung injury. Am Rev Resp Dis. 1993; 147: 177 - 81

“ND,

‘um‘fl-‘a A-'

15.) Villar J, Ribeiro SP, Mullen JBM, Kuliszewski M, Post M, Slutsky AS.
Induction of the heat shock response reduces mortality rate and organ damage
in a sepsis—induced acute lung injury model. Crit Care Med. 1994; 22: 914 -
21.

16.) Ribeiro SP, Villar J, Downey GP, Edelson JD, Slutsky AS Sodium
arsenite induces heat shock protein-72 kilodalton expression in the lungs and
protects rats against sepsis. Crit Care Med. 1994; 22: 922 - 29.

17.) Kusher DI, Ware CF, Gooding LR. Induction of the heat shock response
protects cells from lysis by tumor necrosis factor. J Immunol. 1990; 145: 2925
- 31.

18.) Jaatela M, Wissing D. Heat shock proteins protect cells from monocyte
cytotoxicity. possible mechanism of self-protection. J Exp Med. 1993; 177: 231
- 6.

19.) Welch WJ. How cells respond to stress. Scientific American. May 1993,
pp. 56 - 64.

20.) Cannon, Physiol rev, 1929

21 .) Watowich SS, Morimoto RJ. Complex regulation of heat shock and
glucose responsive genes in human cells. Mol Cell Biol. 1988; 8: 393 - 405.

62

22.) Udelsman R, Holbrook NJ. Endocrine and moolecular responses to
surgical stress. Curr Probl Surg. 1994; 8: 655 - 721.

23.) Yost HJ, Lindquist S. RNA splicing is interrupted by heat shock and is
rescued by heat shock protein synthesis. Cell. 1986; 45: 185 - 93

24.) Munro S, Pelham H. An hsp70—like protein in the ER: identity with the
78kD glucose regulated protein and immunoglobulin heavy chain binding
protein. Cell. 1986; 46: 291 - 300, 1986

25.) Sargent CA, Durham l, Trowsdale J, Cambell RD. Human major
histocompatibility complex contains genes for the major heat shock protein
HSP70. Proc Natl Acad Sci. 1989: 86: 1968 - 72.

26.) Sorger PK. Heat shock factor and the heat shock response. Cell. 1991;
65: 363 - 6.

27.) Buchman TG. Manipulation of stress gene expresion: a novel therapy for
the treatment of sepsis? Crit Care Med. 1994; 22: 901 - 3.

28.) Applegate LA. Luscher P, Tyrrell RM. Induction of heme oxygenase: a
general response to oxidant stress in cultured mammalian cells. Cancer Res.
1991; 51: 974-8.

29.) Luethy JD, Fargnoli J, Park JS, Fomace AJ, Holbrook NJ Isolation and
characterization of the hamster gadd 153 gene. Activation of promoter activity
by agents tat damage DNA. J Biol Chem. 1990; 265: 16521 - 6.

30.) Buchman TG, Cabin DE. Molecular biology of circulatory shock Ill. Hesz
cells demonstrate two patterns of shock—induced gene expression which are
independent, exclusive, and prioritized. Surgery. 1990; 108: 902 - 11.

31 .) Kusher l. The phenomenon of the acute phase response. Ann NY Acad
Sci. 1982; 389: 39 — 48.

32.) Fey GH, Gauldie J. The acute phase response of the liver in
inflammation. Prog Liver Dis. 1990; 9: 89 - 116.

33.) Polla BS, Kantengwa S. Heat shock proteins and inflammation. Curr
Topics Microbiol Immunol. 1991; 167: 93—105

63

34.) Low—Friedrich I, Weisensee D, Mitrou P, Schoeppe W. Cytokines induce
stress protein termination In cultured cardiac myocytes. Basic Res Cardiol.
1992; 87:12-18

35.) Fincato G, Polentarutti N, Sica A, Mantovani A, Colotta F. Expression of a
heat-inducible gene of the hsp70 family in human myelomonocytic cells:
regulation by bacterial products and cytokines. Blood. 1991; 77: 579 — 86.

36.) Konig W, Koller M, Brom J. Priming mechanisms and induction of heat
shock proteins in human PMN granulocytes induced by eicosanoids and
cytokines. Eicosanoids. 1992; 5 (suppl): S 39 - 41.

37.) Blake MJ, Udelsman R, Feulner GJ, Norton DD, Holbrook NJ. Stress
induced HSP70 expression in adrenal cortex: a glucocorticoid-sensitive, age
dependent response. Proc Natl Acad Sci USA 1991; 88: 9873 - 7.

38.) Udelsman R, Blake MJ, Holbrook NJ. Molecular response to surgical
stress: specific and simultaneous heat shock protein induction in the adrenal
cortex. aorta, and vena cava. Surgery 1991; 110: 1125 - 31 .

39.) Sarge KD, Murphy SP, Morimoto RI. Activation of heat shock gene
transcription by HSF1 involves oligomerization, acquisition of DNA binding
activity, and nuclear localization and can occur in the absence of stress. Mol
Cell Biol. 1993; 13: 1392 - 1407.

40.) Buchman TG, Cabin DE, Porter JM, Bulkley GB. Change in hepatic gene
expression after shock / resuscitation. Surgery. 1989; 106: 283-291

41.) Schoeniger L0, Reilly P, Bulkley G, Buchman TG. Heat shock gene
expression excludes hepatic acute phase gene expression after resuscitation
from shock Surgery; 1992

42.) Schoeniger L0, Andreoni KA, Ott GR, Risby TH, Bulkley GB, Udelsman R,
Buchman TG. Induction of heat shock gene expression in the post ischemic pig
liver is dependent on upon superoxide anion generation at reperfusion.
Gastroenterology. 1994; 106: 177 - 84.

43.) Cerra FB. Multiple organ failure syndrome. In Bihari DJ, Cerra FB (eds).
Multiple Organ Failure. Fullerton, CA: Society of Critical Care Medicine, 1989:
pp. 1 - 24.

54

44.) Faist E, Ertel W, Mewes A, Alkan S, Walz A, Strasser T. Trauma-induced
alterations iof the lymphokine cascade. lin Faist E, Ninneman J, Green D (eds.).
Immune consequences of Trauma, Shock, and Sepsis. New York, NY:Springer
Verlag, 1989: pp 79 - 84.

45.) Field BE, Devich LE, Carlson RW. Impact of a commehensive supportive
care team in the management of hopelessly ill patients with multiple organ
failure. Chest 1989; 96: 353-362.

46.) Deitch EA Multiple organ failure: summary and overview. In Deitch EA
(ed.): Multiple Organ Failure. New York, NY: Thieme Medical publishers. 1990:
pp. 285—299.

47.) Baker 0, Faist E. Immunologic response. In Moore EE, Mattox KL,
Feliciano DV (eds): Trauma, 2d ed. East Norwalk, CT: Appleton 8 Lange,1991:
pp. 881 - 90.

48.) Miller CA. Secondary immunodeficiency in burns and after surgical
trauma. Clin Immunol Allergy. 1981; 1: 641.

49.) Stephan RN, Kupper TS, Geha AS, Chaudry IH. Hemorrhage without
tissue trauma produces immunosuppression and enhances susceptibility to
sepsis. Arch Surg. 1987; 122: 62-68

50.) Ayala A, Perrin MM, Wang P, Ertel W, Chaudry IH. Hemorrhage induces
enhanced Kupffer cell cytotoxicity while decreasing peritoneal or splenic
macrophage capacity: involvement of cell—associated TNF and reactive
nitrogen. J Immunol. 1991; 147: 4147 - 54.

51.) Ayala A, Perrin MM, Meldrum DR, Ertel W, Chaudry IH. Hemorrhage
induces an increase in serum TNF which is not asociated with an increase in
endotoxin. Cytokine. 1990; 2: 170 - 4.

52.) Ertel W, Morrison MH, Wang P, Ba ZF, Ayala A, Chaudry IH. The
complex pattern of cytokines in sepsis: association between prostaglandins,
cachectin, and interleukins. Ann Surg. 1991; 214: 141-148.

53.) Ayala A, Wang P, Ba ZF, Chaudry IH. Differential effects of hemorrhage
on Kupffer cells: Decreased antigen presentation despite increased
inflammatory cytokine (IL—1, lL—6 and TNF) release. Cytokine. 1992; 4: 66 — 75.

65

54.) Ayala A. Perrin MM, Chaudry IH. Defective macrophage antigen
presentation following hemorrhage is associated with the loss of MHC Class II
(Ia) antigens. Immunology. 1990; 70:33-39

55.) Ertel W, Morrison MH, Ayala A, Chaudry IH. Chloroquine attenuates
hemorrhagic shock induced suppression of Kupffer Cell antigen presentation
and Major Histocompatibility Complex Class II antigen expression through
blockade of tumor necrosis factor and prostaglandin release. Blood. 1991; 78:
1781 - 8.

prostaglandin production increases cachectin synthesis and prevents
depression of macrophage functions following hemorrhagic shock. Ann Surg.

56.) Ertel W, Morrison MH, Ayala A, Perrin MM, Chaudry IH. Blockade of ET
1991; 213: 265. 5

57.) Ertel W, Morrison MH, Meldrum DR, Ayaa A, Chaudry IH. Ibuprofen
restores cellular immunity and decreases susceptibility to sepsis following i
hemorrhage. J Surg Res. 1992; 53: 55 -61.

 

58.) Espevik T, Nissan—Meyer J. A highly sensitive cell line, WEHI 164
subclone 13, for measuring cytotoxic factor / tumor necrosis factor from human
monocytes. J Immunol Methods. 1986; 95: 99.

59.) Eskandari MK, Nguyen DT, Kunkel SL, Remick DG. WEHI 164 subclone
13 assay for TNF: Sensitivity, specificity, and reliability. lmmunol Invest. 1990;
19: 69.

60.) Mizel SB. Production and quantitation of lymphocyte activating factor
(interleukin 1). In Herscowitz HB, Holden HT, Bellanti JA, Ghaffar A (eds):
Manual of Macrophage Methodology. New York, NY: Marcel Dekker, Inc., 1981:
p. 407.

61.) Chaudry IH, Ayala A, Ertel W, Stephan RN. Hemorrhage and
resuscitation: immunological aspects. Am J Physiol. 1990; 259: R663—R678

62.) Knaus WA, Draper EA, Wagner DP, Zimmerman JE. Prognosis in acute
organ-system failure. Ann Surg. 1985; 202: 685

63.) Chaudry IH, Ayala A Immunological aspects of hemorrhage. Austin, TX:
Medical Inteligence Unit, R.G. Landes Company, 1992: 1 - 32.

66

64.) Meldrum DR, Ayala A, Perrin MM, Ertel W, Chaudry IH. Diltiazem
restores lL—2, lL-3, lL—6, and lFN—gamma synthesis and decreases
susceptibility to sepsis following hemorrhage. J Surg Res. 1991; 51: 158 — 64.

65.) Chaudry IH, Ayala A Mechanism of increased susceptibility to infection
following hemorrhage. Am J Surg. 1993; 165 (suppl): 598 - 67S.

66.) Zellwger R, Ayala A, Schmand JF, Morrison MH, Chaudry IH. PAF—
antagonist administration after hemorrhage—resuscitation prevents splenocyte
immunodepresslon. J Surg Res. 1995; 59: 366 — 70.

67.) Bemelli—Zazzera A, Cairo G, Schiaffonati L, Tacchini L. Stress proteins
and reperfusion stress in the liver. Ann NY Acad Sci. 1992; 663: 120 - 4