• Mediator-s in -typhoid fever · - Radboud Repository

PDF hosted at the Radboud Repository of the Radboud University

Nijmegen

The following full text is a publisher's version.

For additional information about this publication click this link.

http://hdl.handle.net/2066/146563

Please be advised that this information was generated on 2022-07-10 and may be subject to

change.

http://hdl.handle.net/2066/146563

• Mediator-s in -typhoid fever · •сΙίη¡Όari and experimental'studies

nique Ceuter ·

Mediators in typhoid fever

clinical and experimental studies

ISBN 90-9011526-9

©1998 Monique Keuter. No part of this publication may be reproduced, stored in a

retrieval system, or transmitted in any form or by any means without the prior written

permission of the author, or, where appropriate, of the publishers of the publications.

Mediators in typhoid fever clinical and experimental studies

Een wetenschappelijke proeve

op het gebied van de Medische Wetenschappen

Proefschrift

ter verkrijging van de graad van doctor aan de

Katholieke Universiteit Nijmegen,

volgens besluit van het College van Decanen in het

openbaar te verdedigen op dinsdag 26 mei 1998

des namiddags om 1.30 uur precies

door

Monique Keuter

geboren op 13 augustus 1956 te Utrecht

Promotores Prof. Dr. J.W.M, van der Meer Prof. Dr. R. Djokomoeljanto (Diponegoro University)

Co-promotores Dr. B.J. Kullberg

Dr. W.M.V. Dolmans

Manuscriptcommissie Prof. Dr. D. Ruiter

Prof. Dr. J.A.A. Hoogkamp-Korstanje

Dr. W.P.M. Eling

The studies presented in this thesis were performed at the Division of General Internal

Medicine, Department of Medicine, University Hospital Nijmegen, The Netherlands,

and at the Dr. Kariadi Hospital, Diponegoro University, Semarang, Indonesia

To my mother

In memory of my father

Mediators in typhoid fever Clinical and experimental studies

Chapter 1 Introduction

Chapter 2 Experimental studies on the pathogenesis of Salmonella infections

Chapter 3 Patterns of pro-inflammatory cytokines and inhibitors during

typhoid fever

Chapter 4 Phospholipase A2 is a circulating mediator in typhoid fever

Chapter 5 A semi-quantitative reverse transcriptase polymerase chain reaction

method for measurement of mRNA for TNFa and IL-Iß in whole

blood cultures: Its application in typhoid fever and eccentric

exercise

Chapter 6 Pro-inflammatory cytokine profile in Salmonella typhimurium

infection in mice and the effect of recombinant IL-la

Chapter 7 G-CSF enhances survival of CBA/J mice with Salmonella

typhimurium infection

Chapter 8 Hyperlipoproteinemia protects low-density-lipoprotein receptor

deficient mice against lethal Salmonella typhimurium infection

Chapter 9 General discussion

Chapter 10 Summary

Chapter 11 Ringkasan

Chapter 12 Samenvatting

Dankwoord en Curriculum Vitae

Chapter 1

Introduction

10 Chapter 1

Introduction 11

Typhoid fever Typhoid fever is a severe systemic infection usually caused by Salmonella typhi

and occasionally by Salmonella paratyphi. The disease is confined to humans and

spreading is by the fecal-oral route. Great improvements in Western countries have

been achieved with controlled sewage systems and better living conditions. Still,

typhoid fever is a cause of morbidity in many countries of the world, where water is not

properly purified [1]. The World Health Organization (WHO) have estimated that annu

ally, 16.6 million cases of typhoid fever occur, with nearly 600 000 deaths. In South

East Asia, the incidence is highest in Indonesia, with more than 1000 cases per 100 000

inhabitants [2]. Most afflicted are children and young adults, with resulting economic

loss.

Pathogenesis of typhoid fever and other Salmonella infections involves several

processes. In a natural infection, Salmonellae enter via the intestinal tract. First, the

bacteria have to survive the acidic pH of the stomach. The outcome of the infection

depends upon a number of factors including size of the inoculum, the virulence of the

strain, and the ability to mobilize defense mechanisms [1]. The virulence of the Salmo

nella will influence bacterial invasion, survival and replication mechanisms [3, 4]. On

the side of the host, cellular immunity has always been considered the most important

defense mechanism in Salmonella infection [5]. These host factors will be discussed in

Chapter 2.

Clinical manifestations of infection with S. typhi start after an incubation time from

5-30 days. Symptoms range from asymptomatic infection to death, but most often

patients suffer a severe febrile disease with abdominal pain and constipation. Cough

and diarrhea may be prominent in the second week. Complications may occur in the

third week, with gastrointestinal perforation or bleeding and cerebral abnormalities or

even coma. When untreated, typhoid fever begins to wane in the fourth week, with

relapses occurring in 10-30% of the patients.

The definite diagnosis is made by isolating 5. typhi or S. paratyphi from blood or

bone marrow, the latter being frequently positive even after the use of antibiotics [6].

The diagnosis of typhoid fever with the Widal test alone is considered prone to error

[7]. Other serological tests are currently being developed [8].

The incidence of complications has decreased after the introduction of antimicro

bial therapy [9]. In Indonesia antibiotics reduced the mortality from 18% to 4%. Resis

tance to antibiotics is becoming a serious problem globally, and has spread from the

Indian subcontinent to other countries in Asia, northern Africa and Latin America [2].

Still, chloramphenicol can be used as first line drag in many places; third generation

12 Chapter 1

cephalosporines and quinolones are important, although expensive alternatives. In Indo

nesia, resistance to chloramphenicol is not wide-spread [10].

Vaccination with either parenteral Vi polysaccharide or live oral Ty21a Salmonella

vaccines is protective in 65-70%. However, a study conducted in Indonesia with the

latter Salmonella vaccine reported only 53% protection [11]. Newer strains of attenua

ted Salmonella typhi seem promising as oral vaccines, but these are still under develop

ment [12].

Mediators of fever With few exceptions, patients with full-blown typhoid fever have a continuous

fever. The final pathway in the pathogenesis of fever is the prostaglandin E2 (PGE2)

production in the organum vasculosum of the lamina terminalis (OVLT) in the central

nervous system (CNS) [13, 14]. The manner in which this is achieved is not completely

clear. It is generally believed that the release of PGE2 is induced by circulating cyto

kines (also referred to as endogenous pyrogens) which are produced at the site of infec

tion or inflammation and reach the CNS by fenestrated capillaries. Cytokines involved

in the pathogenesis of fever are interleukin-lß (IL-Iß), IL-6, tumor necrosis factor-α

(TNFct) and interferons, which act independently as endogenous pyrogens. Experimen

tal studies have indeed shown, that potent febrile responses could be induced by injec

tion of cytokines directly in the hypothalamus or in the circulation [13]. However, it

may be questioned whether this classical concept of the origin of fever is valid in all

diseases [15].

Regarding the symptoms in typhoid fever, pyrogenic cytokines or other pyrogenic

mediators would be expected to be present in the circulation during weeks, as these

patients have a continuous fever. Since this is not the case, it is suggested that other

mechanisms are involved. This may include other, yet unidentified circulating endoge

nous pyrogens or different mechanisms, including the local release of cytokines.

We asked the question whether secretory phospholipase A2 (sPLA2) may play a

role as a circulating pyrogen. Indeed, intraventricular injection of inhibitors of sPLA2

were shown to suppress fever in rabbits [16]. Phospholipases are lipolytic enzymes

which catalyse the degradation of phospholipids. To date, three varieties of PLA2 have

been characterized: group I (pancreatic) and group II (non-pancreatic) 14kD sPLA2 and

a cytosolic (85kD) PLA2. The group Π sPLA2 is present in and is secreted by a variety

of cells, and this form has been implicated in the generalized inflammatory responses

found in several experimental models and clinical syndromes such as sepsis and adult

respiratory distress syndrome (ARDS) [17, 18]. Its release is induced by IL-1 and

Introduction 13

TNFa and the enzyme is able to mediate the production of arachidonic acid (AA) [19,

20]. If the metabolic effects of these cytokines during infection are caused by sPLA2,

this would imply a central role for sPLA2 in inflammation as an intermediate between

IL-1 and TNFa and AA [18]. Support for a role of sPLA2 as a mediator of the systemic

inflammatory response came from studies in rabbits in which sPLA2 induced a fall in

blood pressure similar to that found with endotoxin infusion, and administration of

sPLA2 inhibitor p-bromophenacylbromide protected against the hypotensive effect of

purified rabbit sPLA2 [16,21].

Collaboration with Indonesia Joint research between the Catholic University Nijmegen, the Netherlands and the

Diponegoro University, Semarang, Java Indonesia was initiated in 1989. The purpose

of this collaboration was to investigate clinical and immunological features of patients

with typhoid fever, as well as to study the pathogenesis of Salmonella infection in an

experimental animal model.

A prospective study was conducted in adult patients with culture proven typhoid

fever. The aim of this project was to determine the role of mediators i.e., cytokines

present in the circulation and of importance otherwise in both complicated and uncom

plicated typhoid fever. Clinical data were collected and blood samples for cytokines

were processed locally according to a slightly adapted whole blood method, which can

be applied in countries where sophisticated laboratory facilities are not readily available

[22]. Since the immune response to many infections is also determined by HLA type

[23], the genetic characteristics of patients were also subject of study.

Furthermore, a scoring system was developed in order to predict the occurrence of

complications in patients with typhoid fever on admission. This typhoid fever admis

sion severity score (TYFASS) was derived from the clinical data collected in a pilot

study of 112 patients and will be validated by a running study. To determine risk factors

for contracting typhoid fever, social factors and micronutrients of patients and age and

sex-matched controls were determined. Salmonella typhi strains from the patients with

typhoid fever were investigated on resistance [10] and phage typed.

Finally, a therapeutic trial was performed to optimize antimicrobial therapy in

patients with typhoid fever. From the pilot study on clinical aspects, it had become clear

that bone marrow cultures had a reasonably good yield. Especially after the use of anti

biotics, bone marrow cultures remained positive as many as seven days after the start of

chloramphenicol [6]. As this permanent presence of bacteria in the bone marrow may

be the reason for relapse and carrierstate after treatment with chloramphenicol, a rando-

14 Chapter 1

mized clinical trial was done to study the sterilization-of bone marrow after treatment

with chloramphenicol compared to quinolones.

To gain insight in the pathogenesis and host defense of Salmonella infections, we

investigated whether experimental Salmonella infections in mice could be modulated

by cytokines.

Outline of this thesis In Chapter 2, the pathogenesis of Salmonella infections is described and the role of

cytokines involved in these infections are reviewed.

Chapter 3 addresses the question which concentrations of proinflammatory cyto

kines are detectable in plasma of patients with typhoid fever. We measured circulating

cytokines (IL-lß, TNFa, TNFß (lymphotoxin-a) and IL-6), and the concentrations of

interleukin-8 and the cytokine inhibitors interleukin-1 receptor antagonist (IL-IRA) and

the soluble TNF receptors p55 and p75 (sTNF-R). Moreover, the capacity of circulating

blood cells to produce cytokines after stimulation by lipopolysaccharide (LPS) ex vivo

was determined. For this purpose, we have used the whole blood cytokine test as descri

bed by us and Nerad [24, 25].

In Chapter 4, we report investigations to answer the question whether sPLA2 is a

circulating mediator in typhoid fever and whether its presence could be related to cyto

kine responses and severity of disease. Therefore, we sequentially measured circulating

pyrogenic cytokines and their inhibitors and sPLA2. In addition, we examined the

capacity of blood cells to produce IL-lß, TNFa, IL-IRA and sPLA2, upon LPS stimu

lation in samples obtained from patients with typhoid fever at various stages of the dis

ease.

In Chapter 5, the role of messenger RNA of IL-lß and TNFa in the whole blood

procedure is described, to contribute further to the understanding of how the production

of cytokines during typhoid fever is regulated.

In order to better understand the host defense against Salmonella, we investigated

the cytokine responses in mice (Chapter 6) and compared these with those found in

patients with typhoid fever (Chapters 3, 4 and 5) In line with other studies from our

laboratory, we asked the question whether administration of recombinant IL-1 would

exert a protective effect in a murine model of 5. typhimurium infection (Chapter 6).

Neutrophils are able to ingest and kill salmonellae, and they do play a role in the

outcome of Salmonella infection in vivo (see Chapter 2). We asked the question whe

ther stimulating neutrophil production and function by administering rG-CSF before

and during a lethal S.t yphimurium infection would affect survival, outgrowth of micro-

Introduction 15

organisms, histology and cytokines (Chapter 7)

Although Salmonella is a Gram-negative rod, the role of endotoxin in the pathoge

nesis of infection caused by these organism is controversial. We approached this pro

blem by using low density lipoprotein receptor knock-out (LDLR-/-) mice, of which the

circulating lipoproteins are able to bind and neutralize bacterial LPS, thereby reducing

the induction of proinflammatory cytokines (Chapter 8).

References start on page 122

16 Chapter 1

Chapter 2

Experimental studies on the pathogenesis of Salmonella infections

18 Chapter 2

Experimental Salmonella infecüon 19

Whereas Salmonella typhi is not pathogenic to mice, experimental Salmonella

typhimurium infection in mice leads to replication of bacteria in the spleen and the liver

and induction of a disease similar to human typhoid fever. Therefore the mouse model

with S. typhimurium is used to study typhoid fever.

Genetic differences in host defense Inbred mouse strains differ in their susceptibility to S. typhimurium infection. The

most important gene involved is Ity (Immunity to typhimurium), which encodes for the

recently described natural resistance associated membrane protein 1 (Nrampl) on pro

fessional phagocytes [26-28]. The gene is located on chromosome 1. Macrophages of

Ity resistant (R) mice are better able to restrict the growth of Salmonella compared to

those of Ity susceptible (S) mice, that lack the Nramp function. Genetic control of the

innate resistance of mice to S. typhimurium (Ity) is noticeable 24 hours after the start of

infection [29]. Infections in ItyR and Ity S mice can be influenced by recombinant cyto

kines [30-33]. It is not clear whether Ity itself has a regulatory role on cytokine produc

tion [34, 35].

The role of endotoxin Salmonella is a facultative intracellular Gram-negative rod; it contains lipopoly-

saccharide (LPS) but Salmonella spp. do not induce an illness similar to Gram-negative

extracellular bacteremia such as Klebsiella or Escherichia coli do. In animal models of

salmonellosis, TNFoc was detectable in the circulation only several days after S. typhi

murium infection, whereas TNFa did rise 1 hour after LPS administration in these ani

mals [36, 37]. Free endotoxin has never been found in the circulation in S. typhi

infection and endotoxin tolerance in humans did not prevent the development of

typhoid fever [38, 39]. However, LPS in Salmonella is important, as shown by

СЗН/HeJ mice in which cytokines can not be induced. СЗН/HeJ mice are Lps-d and

lack the Lps-gene. The Lps-gene is located on chromosome 4. These mice are not able

to respond to LPS and therefore they have no cytokine production upon infection with

Salmonella. They die after a low dose of Salmonella, although they are ItyR i.e. they

possess the Nramp function. This Lps-d strain benefits from TNFa and IL-1 post

Salmonella infection [40]. LPS of the Salmonella has also been reported to be a viru

lence factor in colonization of the gut and in vaccine immunity [41].

20 Chapter 2

Phases of experimental Salmonella infection The host defense against S. typhimurium infection requires several effector cells,

such as neutrophils, monocytes/macrophages and Τ lymphocytes. These cells are acti

vated and communicate via cytokines. A systemic S. typhimurium infection in mice is

thought to proceed in phases [42,43].

The first phase of an intraperitoneal (i.p.) or intravenous (i.v.) infection takes only

a few hours, in which initial distribution of the microorganisms take place. Whether

macrophages are already activated before this infection seems important, as Salmonella

injected into mice with activated macrophages by a previous infection with S. typhimu

rium or bacillus Calmette-Guerin (BCG), led to a significant lower numbers of micro

organisms in the liver and spleen after 3 and 8 hours respectively, compared to those in

organs of normal previously uninfected mice [44,45]. The transition phase of the bacte

rium, lipids and circulating antibodies are probably also able to influence this stage.

The second phase, during the first days of infection is called the exponential

growth stage and takes place before the acquired immunity has developed. Genetic

factors of the host (such as Ity) play a role [46,47] as well as virulence and inoculum of

the microorganisms. Silica treatment to destroy macrophages leads to a deterioration of

this phase of the infection [48].

Neutrophils are crucial in this stage. Salmonellae invade hepatocytes, which are

destroyed by neutrophils [49]. Natural killer (NK) cells and Τ lymphocytes are probably

not involved in the lysis of hepatocytes. Earlier studies have shown that the liver and

spleen of mice infected with S. typhimurium contained many neutrophils and micro-

abcesses after 24 hours of infection. Hardly any monocytes were found at that point of

time [49-51]. Blocking the migration of neutrophils through administration of monoclo

nal antibodies to the leukocyte adhesion molecule ß-integrin, lethally exacerbated the

infection. 100 times more salmonellae were recovered from the livers of these mice

compared to control mice [49]. In addition, neutropenic mice developed a lethal syste

mic salmonellosis after infection with low numbers of microorganisms. In these mice,

especially the hepatic microvasculature was invaded with salmonellae [50, 52]. There

fore it appears, that neutrophils are crucial for effective host defense against Salmonella

infection. Neutrophils act early in the infection to restrict Salmonella growth to a level

that can be adequately dealt with by other host defenses later.

After 3-7 days, the outgrowth in liver and spleen during Salmonella infection

Experimental Salmonella infection 21

reaches a plateau under the influence of activated macrophages producing proinflam

matory cytokines In this third phase, Τ lymphocytes do not have an important func

tion, as it was shown that athymic mice could reach the plateau phase without difficul

ties [53,54].

There is vast evidence that macrophages play an important role in the defense

against facultative intracellular infections such as Salmonella infection [5, 55-58]

Treatment of mice with silica, leading to destruction of macrophages in experimental

Salmonella infections, led to deterioration of the infection [48] Before the discovery of

the Ity locus on chromosome 1, genetic differences (ItyS and ItyR) in susceptibility in

mice were already found to be based on the difference ш ability of macrophages and

granulocytes to kill S typhimunum [46,47].

Whether the activation of macrophages is crucial for the host defense (killing)

against salmonellae is matter of much debate Although during an infection with S.

typhimunum, macrophages became activated, as was shown by their enhanced capacity

to inhibit proliferation of Toxoplasma gondii and production of nitric oxide, these

activated macrophages were unable to kill Salmonella better in vitro [44, 59] Also

macrophages activated by BCG infection did not have enhanced intracellular killing

abilities for Salmonella [45], although BCG pretreatment in vivo increased survival

from 5. typhimunum infection [60] Macrophages might be activated not to better kill,

but to enhance killing by other cells by producing cytokines. These other cells could be

NK cells, as these cells were activated during vaccination with a Salmonella strain [61]

or granulocytes [49, 51, 52]. The latter would imply that granulocytes do not only play

a role in the early phases of Salmonella infection

Some investigators query that Salmonella is an intracellular pathogen in the tradi

tional definition of being able to multiply inside macrophages Studies with murine and

human macrophages showed that survival of S typhi or S. typhimunum within macro

phages was dependent on host specificity and virulence of the strain [62] There is also

evidence that Salmonella are readily killed by neutrophils and peritoneal macrophages

[63] Intracellular proliferation of Salmonella proved possible within hepatocytes and

other parenchymal cells [50, 51]. Other bacterial and protozoal pathogens currently

believed to be intramacrophage pathogens in fact may parasitize non-phagocytic cells in

vivo. Integrity of these cells might also be influenced by cytokines and this could be an

important defense mechanism with a role for macrophages It has been shown for

cultured HEp-2 cells, that TNFa inhibited invasion of Salmonella [64] In L mono

cytogenes infection a similar phenomenon has also been shown for interferon-gamma

(IFN-γ) [65, 66]

22 Chapter 2

Following the plateau phase, there is the clearance phase during the third week of

infection, which involves the activity of Τ lymphocytes. Athymic mice or T-cell deple

ted animals show an increase in the number of microorganisms in the liver and spleen

after the plateau phase. In this phase, CD4+ T-lymphocytes mediate bacterial clearance

and IFN-γ mRNA expression was seen in CD4+, but also in CD4- (NK) cells. CD4+

cells are required in this phase as infection with S. typhimurium dependent on aromatic

precursors (aroA strain) could not be cleared by mice that lacked these cells or the IFN-

γ-receptor [67].

Cytokines In animal models of extracellular Gram-negative septicemia, high circulating

concentrations of proinflammatory cytokines such as IL-1 and TNFa have been consi

dered the main cause of organ damage and death. In contrast, against infections with

intracellular pathogens like Leishmania spp., Listeria monocytogenes or Mycobacte

rium spp., TNFa is supposed to be needed locally (in small amounts) and adminis

tration of TNFa inhibited the outgrowth of the microorganisms, whereas administration

of antibodies to this cytokine proved to be detrimental [68-73]. Moreover, mice without

TNFa receptor I (TNFRI-knockout mice) were resistant against endotoxin-induced

shock, but succumbed to infection with L monocytogenes [74, 75].

It is generally accepted that the main role of T-cells in resistance to intracellular

pathogens, is to produce IFN-γ to increase bactericidal abilities of macrophages [58]. In

the past 10 years, a concept has been developed in dividing Τ helper cells according to

the cytokines they produce (Thl versus Th2 response) [76]. Thl cells produce IFN-γ

and TNF-ß (lymphotoxin-α) inducing cellular immunity, whereas Th2 cells produce IL-

4, IL-5, IL-10 and IL-13, which are responsible for strong antibody responses. The

differentiation between Thl/Th2 is also under control of cytokines as IL-12 induces

Thl, and IL-4 induces Th2 development. In general, for the host defense of intracellular

infections a type 1 response is required.

Indeed, Thl-cell-produced cytokines such as IL-1, TNFa and IFN-γ have been

shown to be beneficial to animals with Salmonella infections and blocking these sub

stances deteriorated the infection. In contrast, antibodies to most Th2 produced cyto

kines, such as IL-4 and IL-10 increased resistance to S. typhimurium infection. How

ever, this division in Thl and Th2 should not become dogma, as there are conflicting

reports on both sides [77]. The role of the cytokines, regarded the most important in

Salmonella infections, will be described.


Tumor necrosis factor-α

The family of tumor necrosis factors comprises three members: TNFa, TNF-ß

(also known as lymphotoxin-a, LT-a) and LT-ß. TNFa has a key role in inflammatory

diseases. Mononuclear phagocytes are the major source of TNFa. The 26 kDa propro-

tein is cleaved by TNFa converting enzyme (TACE) and yields the 17kDa protein [78].

The effects of TNFa are transmitted via crosslinking of the membrane bound receptor I

(TNFRI, p55) and TNFRII (p75). The extracellular parts of these receptors can be shed

and these soluble receptors can bind TNFa and limit some of its effects [79]. TNFa is

produced after a stimulus such as LPS. The sepsis syndrome that develops after

injection of LPS can be mimicked by intravenous administration of IL-1 or TNFa. The

two proinflammatory cytokines, TNFa and IL-1 can induce each other's release.

Although TNFa in large amounts can be deleterious in extracellular Gram-negative

infections, TNFa is required for host defense in intracellular infections. TNFa induces

IFN-γ production by NK-cells in the early phases of the infection [80]. In infections

with intracellular pathogens like Leishmania spp, Listeria monocytogenes and Toxo

plasma gondii or Mycobacterium spp., administration of TNFa inhibited the outgrowth

of the microorganisms, whereas injection of antibodies to this cytokine proved to be

detrimental [70-72, 81, 82]

In experimental S. typhimurium infection in mice, the role of TNFa is similar to

that in other intracellular infections [83-85]. Inhibition of TNFa during Salmonella

infection has worsened outcome [84, 86], whereas administration of TNFa resulted in

better survival [83, 87].

The exact function of TNFa in Salmonella infections however, is not clear. In-vivo

studies of S. typhimurium infection in mice showed that TNFa was involved after the

early exponential growth phase. TNFa was especially required in the plateau and clear

ance phases. On histopathology, the effect of anti-TNFa antibodies was also seen not

earlier than after 4 days. In these anti-TNFa treated mice, most microorganisms were

seen in Kupffer cells and it was shown that the recruitment of monocytes was impaired.

There was less granuloma formation with fewer mononuclear cells. Administration of

antibodies on day 7 resulted in relapse of the infection and regression of established

granulomas [86]. No apparent effect on neutrophils was shown in the latter study, while

in other experiments with anti-TNFa antibodies fewer neutrophils were seen in the liver

[88]. TNFa is also important for immunity. Anti-TNFa antibodies prior to oral chal

lenge completely abolished protection conferred by the vaccine [85]. From these studies

one can conclude that TNFa acts primarily on monocytes (granuloma formation) and

24 Chapter 2

these cells have their function in the plateau and clearance phase in Salmonella infec

tions. TNFa is essential for the activation of monocytes in an autocrine fashion [81].

Furthermore, cultured hepatocytes (HEp-2 cells) were resistant against invasion of Sal

monella after incubation with TNFa in vitro, which suggests a role for TNFa on hepa

tocytes too [64].

Interferon-γ

IFN-γ is produced by CD4+ Τ lymphocytes, CD8+ T-cells, natural killer (NK)

cells and γδ Τ lymphocytes. IFN-γ has profound effects on host resistance and affects

the function of polymorphonuclear leukocytes and macrophages. Macrophage activa

tion is necessary to kill intracellular pathogens [58]. IFN-γ was shown to be the key

lymphokine in macrophage activation in in vitro animal experiments, as well as in trials

with patients with leprosy or AIDS [89]. IFN-γ increases the respiratory burst and pro

duction of nitric oxide, thereby enhancing the capacity to kill intracellular microorga

nisms. It also induces expression of Fc receptors and surface markers such as the cyto

kine IL-2 and major histocompatibility complex (MHC) class-Π antigens [90].

IFN-γ has been shown to have a beneficial effect on infections with intracellular

pathogens such as Toxoplasma gondii, Leishmania spp. and Listeria [91]. In infections

with Listeria monocytogenes, the mRNA for IFN-γ correlated with antibacterial resis

tance and was lower in susceptible mice [35,92, 93]. Mice that lack the IFN^-receptor

displayed increased susceptibility to Listeria infections, but had normal lymphocyte

subset development and MHC antigen presentation [91].

IFN-γ is also an important mediator in Salmonella infections. Mouse strain differ

ences in cytokine production have been reported previously [94] and in mice with S.

typhimurium infection, lower IFN-γ production by isolated spleen cells stimulated with

live S. typhimurium from ItyS mice has been shown compared to ItyR mice [95 ]. Viru

lence of the bacteria could play a role next to genotype in this difference [96]. A diffe

rence in the ability of production of IFN-γ is denied by other studies, in which IFN-γ

mRNA appeared at the same time in susceptible and resistant mice during a S.typhimu-

rium infection, and kinetics of this cytokine were not different between the strains [35].

Cytokine production in vivo after S. dublin infection was not different either between

ItyS and ItyR mice [34]. In contrast with Listeria infections, mRNA of IFN-γ correlated

with numbers of Salmonella cfu in the spleen and not with enhanced killing abilities

[35,92].

Anti-IFN-γ antibodies decreased host defense to Salmonella infection in mice.

These antibodies exerted a maximal effect during the plateau-phase of the infection [54,


84]. During this phase, IFN-γ is presumably produced by NK cells and not by T-cells

[53, 54, 83]. In orally challenged mice, anti-IFN-γ antibodies abrogated the resistance

and caused more mortality and outgrowth of Salmonella in the tissues [97]. There are

conflicting results concerning the role of IFN-γ in the clearance phase. Anti IFN-γ

antibodies administered in the late stage of salmonellosis did not alter the elimination of

certain strains of Salmonella [98], whereas infection with S. typhimurium dependent on

aromatic precursors (aroA strain) could not be cleared by mice that lacked the IFN-γ-

receptor [67].

Pretreatment of mice with exogenous rIFN-γ before injection of LPS resulted in

higher circulating concentrations of TNFa [99]. Some of the functions of activated

macrophage by rlFN-yare also mediated by the production of endogenous TNFa [100].

This might have been the case in studies where pretreatment with exogenous rIFN-γ

had an inhibiting effect on multiplication on the first day of the in vivo infection, but

not on the growth rate of Salmonella during days 1 to 4 [54]. In vitro macrophage acti

vation by IFN-γ with decreased multiplication of Salmonella has been shown by some

[30], but denied by other investigators, albeit in shorter observation periods [101, 102].

Orally administered recombinant murine IFN-γ had a suppressive effect on the

development of a natural S. typhimurium infection [103].

Interleukin-1

IL-1 is the name for two polypeptides (IL-la and IL-Iß) that possess a wide

spectrum of inflammatory, metabolic, physiologic, hematopoietic and immunologic

properties. These forms are distinct gene products, but they bind to the same receptor

and share various biologic activities [104]. The most potent stimulus for IL-1 is LPS.

When LPS is injected in mice or human volunteers, IL-1 concentrations are found in

the circulation after 3 hours. A similar time course was found after injection of Salmo

nella LPS [36]. However, in experimental S. typhimurium infection, circulating IL-la

was not found [37], showing that Salmonella infection and LPS injection is pathophy

siological^ distinct.

Intervention studies with antibodies against IL-1 in animals with Salmonella infec

tions have not been done. Pretreatment with recombinant IL-1 ameliorates a variety of

infections. It prolongs survival in Gram-negative and Gram-positive infections in

normal and neutropenic mice and increases resistance against candidal and plasmodial

infections in mice [105-108].

Intraperitoneal recombinant IL-1, given 24 hours before a lethal S. typhimurium

infection, increased survival in ItyR mice and reduced outgrowth in the organs [32].

26 Chapter 2

Addition of rTNFa to rIL-1 had a synergistic effect in ityR mice.

The mechanisms of protection by rIL-1 are not completely known yet. In Gram-

negative Pseudomonas and Klebsiella infections, reduction of proinflammatory cyto

kine production by peritoneal macrophages of Swiss mice has been found after admi

nistration of rIL-1 and this ameliorated the lethal cytokinemia [109]. In addition, cyto

kine receptors were modulated.

Whether IL-1 is involved in the microbicidal activity of murine macrophages is a

matter of debate. Incubation with rIL-1 did not enhance the capacity of peritoneal

macrophages to kill Toxoplasma gondii and to produce nitric oxide [100]. However, in

another study LPS-stimulated peritoneal macrophages were found to produce more

nitric oxide after pretreatment of mice in vivo with rIL-1 [110]. Induction of nitric

oxide synthesis is considered to be important for the killing of S. typhimurium in mice

and granuloma formation in the livers of these mice was impaired when nitric oxide

production was inhibited [111].

InterIeukin-4

Interleukin-4 (a 18-20 kDa glycoprotein) is a multifunctional cytokine, which dis

plays both inhibitory and stimulatory properties. IL-4 is considered to be produced by

the Th2 subset of Τ cells. IL-4 is also able to promote this Th2 type of response and

synthesis of IgE and it inhibits the production of proinflammatory cytokine [112].

In systemic Salmonella infection, IL-4 is detrimental. In a recent study with S.

typhimurium infection in IL-4 knock-out mice, these mice had longer survival and less

abcesses in the liver than their IL-4 +/+ littermates [113]. Expression of the murine IL-4

gene in an attenuated strain of S. typhimurium caused more outgrowth of salmonellae

when used as a live vaccine. The killing of Salmonella by macrophages of these

vaccinated mice was impaired [114].

Interleukin-6

IL-6 is a pleiotropic cytokine, involved in the regulation of immune responses, the

acute phase response and haematopoiesis. Infusion of rIL-6 results in fever, but not in

shock. IL-6 is produced by mononuclear phagocytes and other cell types. IL-1 induces

production of IL-6, and some properties of IL-1 such as synthesis of acute phase pro

teins are mediated by IL-6. In various diseases with cytokine activation, IL-6 can be

detected in the circulation, as it has a long halflife. IL-6 induced IL-IRA and sTNFRs

and inhibited production of IL-1 and TNFct and [115] is therefore considered to be an

antiinflammatory cytokine [116].


Interleukin-6-knock-out mice proved highly susceptible to Listeria monocytogenes

infection. In this study, the absence of IL-6 impaired neutrophilia in the peripheral

blood [117]. Treatment with recombinant IL-6 could not elevate the resistance against

L monocytogenes infection in normal mice in contrast to rIL-1 and rTNFa [118].

In S. typhimurium infection in mice, a gradual increase of circulating concentra

tions of IL-6 has been found during the first week of infection [37]. The effect of anti

bodies to IL-6 or recombinant IL-6 has not been investigated in Salmonella infection.

Interleukin-10

IL-10 is a product of T-cells, В cells and macrophages. Like IL-4 it is produced by

the Th2 subset of CD4+ Τ cells and it suppresses Thl cells [119]. IL-10 is a potent anti

inflammatory cytokine. It was reported to be able to downregulate IFN-γ production by

NK cells and CD4+ Τ lymphocytes, presumably by its inhibition of TNFa and IL-1

production by macrophages [120]. IL-4 together with IL-10 can inhibit the delayed type

hypersensitivity responses, which are Thl properties [121].

IL-10 gene expression in spleen cells in ItyS mice was reported to be higher than in

ItyR mice during the early phase of S. typhimurium infection. This difference correlated

with outgrowth and was a consequence rather than a cause of Ity susceptibility [35]. In

S. cholerasuis infection, the administration of anti-IL-10 antibodies was benificial for

the mice and resulted in increased proinflammatory cytokine production of infected

peritoneal cells. Bacterial growth in the liver and in the peritoneal cavity was reduced in

these mice [122]. However, anti-IL-10 antibodies did not influence the outcome of 5.

typhimurium infection in ItyS mice in another study [35].

Interleukin-12

Interleukin-12 is a heterodimeric cytokine, composed of 40- and 35-kDa subunits.

IL-12 is produced by mononuclear cells, macrophages, neutrophils, B-cells and dendri

tic cells, in response to a variety of stimuli [123]. IL-12 activates NK cells and Τ lym

phocytes to produce IFN-γ. This combination of IL-12 and IFN-γ shifts the immune

response towards Thl type immunity. Down-regulation of IL-12 can be achieved by IL-

4, IL-10, IL-13, TGFß and PGE2 and possibly IL-12 p40 homodimers [124, 125]

Recombinant IL-12 is being investigated as a potential stimulant of the immune

responses to malaria, tuberculosis, leishmaniasis and possibly AIDS (and as a vaccine

additive). For instance in experimental leishmaniasis, IL-12 exerted therapeutic activity

and induced IFN^and IL-10 and suppressed IL-4 independently of IFN-γ [126, 127].

However, high doses of IL-12 induced a toxic shock like syndrome.

28 Chapter 2

Endogenous IL-12 is required for the host defense against Salmonella. After immu

nization with attenuated Salmonella spp. or killed organisms, protection correlated with

IL-12 p40 mRNA in Peyer's patches and in peritoneal macrophages [128]. In mice chal

lenged orally with S. dublin, endogenous IL-12 was found in the mesenteric lymph

nodes. The administration of antibodies to IL-12 increased the outgrowth of microorga

nisms and reduced survival times [129]. In S. typhimurium infection in ItyR mice, anti-

IL-12 antibodies were shown to impair the establishment of the plateau phase leading to

higher number of Salmonella cfu in spleen and liver at day 7 of infection [130].

The neuropeptide substance Ρ has been shown to increase secretion of IL-12 p40

and expression of IL-12 p40 and p35 mRNA in cultured murine macrophages [131].

Spantide Π, a substance Ρ antagonist reduced survival of mice infected orally with

Salmonella spp., reducing IFN-yand IL-12 p40 expression at mucosal sites [132].

After incubation with S. dublin LPS in vitro, murine macrophages released IL-

12p40, but not IL-12p70 [133]. The IL-12p40 homodimers competed with the IL-12

(p35 and p40) heterodimers for binding to the high affinity IL-12 receptor and therefore

inhibited bioactivity. These Salmonella LPS-induced homodimers of IL-12p40 might

attenuate the cytokine response in acute endotoxemia [134].

Granulocyte colony-stimulating factor

Granulocyte colony-stimulating factor (G-CSF) is a hematopoietic growth factor,

which stimulates activation and proliferation of neutrophils [135]. The effects on matu

re neutrophils include enhanced phagocytosis and respiratory burst as well as reduced

apoptosis. Administration of rG-CSF has been studied in many neutropenic, as well as

non-neutropenic animal models [136]. In a peritonitis model in non-neutropenic mice,

rG-CSF was highly effective in improving survival [137, 138].

In models of intracellular infection, experience with rG-CSF is scarce. In mycobac

terial infections there are divergent results [139]. rG-CSF was beneficial in Listeria

monocytogenes infection in mice, five days of pretreatment with high doses of human

rG-CSF reducing lethality and outgrowth of microorganisms [140,141].

Natural resistance in murine salmonellosis is necessary in the period before acqui

red immunity. Neutrophils are important in the early SalmonelL· infection [49] and

therefore it may be envisaged that administration of rG-CSF is beneficial.

Apart from its capacity to increase neutrophil function, G-CSF has antiinflamma

tory properties. rG-CSF has been shown to be able to impair the LPS-induced IFN-γ

excretion, and increase the production of the anti-inflammatory cytokine IL-10. Fur

thermore, rG-CSF has been shown to be able to influence TNFa production [142,143].


With intracellular pathogens, proinflammatory cytokines are required to overcome the

infection Therefore with a resulting decreased TNFa production by treatment with rG-

CSF, administration of rG-CSF could harm animals with Salmonella infection Since

we do not know, how neutrophil-activating and antiinflammatory functions of rG-CSF

are balanced, it is interesting to study the administration of rG-CSF in a mouse model

of S typhtmunum infection

References start on page 122

30 Chapter 2

Chapter 3

Patterns of pro-inflammatory cytokines and inhibitors during typhoid fever

Monique Keuter, Edi Dharmana, M. Hussein Gasem, Johanna van der Ven-Jongekrijg,

Robert Djokomoeljanto, Wil M.V. Dolmans, Piene N.M. Demacker, Robert Sauerwein,

Harald Gallati and Jos W.M. van der Meer.

J Infect Dis 1994;169:1306-1311

32 Chapter 3

Abstract Cytokines and inhibitors in plasma were measured in 44 patients with typhoid

fever. Ex-vivo production of the cytokines was analyzed in a whole blood culture sys

tem with or without lipopolysaccharide (LPS). Acute phase circulating concentrations

of cytokines (± SD) were: interleukin-lß (IL-lß) < 140 pg/mL, tumor necrosis factor α

(TNFa) 130 ± 50 pg/mL, interleukin-6 (IL-6) 96 ± 131 pg/mL and interleukin-8 (IL-8)

278 ± 293 pg/mL. Circulating inhibitors were elevated in the acute phase: Interleukin-1

receptor antagonist (IL-IRA) 2304 ± 1427 pg/mL and soluble TNF receptors 55 and 75

4973 ±2644 and 22865 ±15143 pg/mL, respectively. LPS-stimulated production of

cytokines was lower during the acute phase than during convalescence (mean values:

IL-lß: 2547 vs. 6576 pg/mL; TNFa: 2609 vs. 6338 pg/ml; IL-6: 2416 vs. 7713 pg/mL).

LPS-stimulated production of IL-IRA was higher in the acute than during the convales

cent phase (5608 vs 3977 pg/mL). Inhibited production of cytokines in the acute phase

may be due to a switch from a proinflammatory to an antiinflammatory mode.

Cytokines in typhoid fever 33

Introduction

Typhoid fever is caused by the facultative intracellular Gram-negative bacillus

Salmonella typhi and occasionally by Salmonella paratyphi. Although salmonellae con

tain lipopolysaccharide (bacterial endotoxin, LPS), the clinical picture of typhoid fever

differs from Gram-negative sepsis, and the role of endotoxin in the pathophysiology of

typhoid fever is controversial [1].

The proinflammatory cytokines interleukin-lß (IL-lß), tumor necrosis factor alpha

(cachectin, TNFa), interleukin-6 (IL-6), and interleukin-8 (IL-8) have been implicated

in the pathogenesis of sepsis caused by Gram-negative microorganisms [2-4]. When

LPS is injected intravenously into animals or human volunteers, elevated concentrations

of these cytokines can be detected and the symptoms and signs of sepsis are mimicked

[5-7]. Elevated circulating levels of TNFa have been shown to correlate with poor

prognosis in sepsis, meningococcemia and also in cerebral malaria [7-10]. In contrast,

in infections with intracellular pathogens like Leishmania spp, Listeria monocytogenes

or mycobacteria administration of TNFa inhibited the outgrowth of the microorga

nisms, whereas administration of antibodies to this cytokine proved to be detrimental

[11-16]. In experimental Salmonella typhimurium infection in mice, the role of TNFa

seems to be similar to that in other intracellular infections [17-19]. In calves with

Salmonella typhimurium sepsis, the cytokine pattern appears to differ from that seen

after intravenously administered LPS: where TNFa did rise 1 hour after LPS admini

stration, Salmonella sepsis caused a hardly detectable increase of TNFa [20].

In contrast to these animal studies, circulating cytokines (TNFa, IL-6 and IL-lß)

have been found elevated in children with typhoid fever in Chile [21]. Butler et al.

studied outcome of typhoid fever in adult patient in Nepal and found higher values of

IL-6 and soluble TNF receptor p55 related to poorer outcome [22].

From 1989 joint research on several aspects of typhoid fever was started between

Nijmegen University, the Netherlands and Semarang University, Indonesia. To obtain

more insight in the pathophysiology of typhoid fever, not only did we measure

circulating pyrogenic cytokines (IL-lß, TNFa, TNFß (lymphotoxin) and IL-6), but also

the concentrations of interleukin-8 and the cytokine inhibitors interleukin-1 receptor

antagonist (IL-1RA) and the soluble TNF receptors p55 and p75 (sTNF-R). In addition,

we have investigated the capacity of blood cells to produce IL-lß, TNFa, IL-6, and

IL-IRA ex vivo in the acute and convalescent phase of hospitalized patients with

typhoid fever.

We used the whole blood cytokine test as described by van Deuren [23] and Nerad

34 Chapter 3

et al.[24]. This assay is simple, reproducible and especially suitable for use in laborato

ries that are not particularly well-equipped for work with cytokines. In addition, the

method may be less artificial than isolating mononuclear cells over a gradient, and pro

bably is a more natural mirror of what is happening in vivo, because plasma factors and

other cells are left in situ.

Patients and methods

The study was done in Dr. Kariadi Hospital, Diponegoro University, Semarang,

Indonesia beginning in December 1990. Blood and bone-marrow cultures were done for

all adult patients (> 14 years old) hospitalized with suspected typhoid fever (defined as

patients with fever > 38.5° С and at least one of the following signs: relative brady

cardia, abdominal complaints, mental changes, signs of complicated typhoid fever, an

enlarged liver or spleen and no apparent other disease). If blood or bone marrow cul

tures were positive for Salmonella typhi or paratyphi A or patients were found to have

perforated ilea at surgical exploration, typhoid fever was considered proven. A total of

44 patients were studied. Patient characteristics are shown in Table 1. Complications of

typhoid fever were defined as: gastrointestinal bleeding; intestinal perforation; shock;

delirium, stupor or coma; pneumonia; diffuse intravascular coagulation.

Table 1. Characteristics of 44 patients with culture-proven typhoid fever

Characteristic complicated disease uncomplicated disease

Number of patients

Mean age (range), years

Males/females

Median (range) of days with

fever before admission

Leukocyte count at admission

(range)

Complications

pneumonia

delirium

perforation

bleeding

16

20(14-34)

7/9

10 (4-20)

7.3xl0 9/L(2.6-37)

16

10

2

5

1

28

24 (14-60)

12/16

8.5 (4-30)

4.4xl09/L(1.6-7.4)

0


Treatment consisted of chloramphenicol (40 mg/kg/day orally) if leukocyte counts

were > 2 χ 109/L. If fever did not subside within 6 days, treatment was changed to sul

famethoxazole (800 mg) and trimethoprim (160 mg) twice daily or ampicillin (4 χ lg).

Surgical patients received ampicillin, metronidazole and gentamicin during and after

surgery. No cyclooxygenase inhibitors were given. Only 2 patients received a single

dose of 120 mg dexamethasone, but not before blood was obtained for cytokine mea

surement. Most patients were discharged 7-10 days after defervescence, which was the

definition of convalescence. No patients died.

Cytokine measurements

On admission and at convalescence blood was drawn for cytokine measurements.

Venous blood samples were aseptically collected into sterile 4 mL tubes containing

EDTA (Vacutainer, Becton Dickinson, Rutherford, NJ). Unless stated otherwise, 3

tubes were taken from each patient [23]. To each tube 250 \ÍL aprotinin (Trasylol,

Bayer, Leverkusen, Germany; final concentration 625 kallikreine inactivating units/mL)

was added through the stopper by a tuberculin needle and syringe. One tube was

centriraged directly (1250 g for 10 minutes) the platelets from the supernatant plasma

were removed by second centrifugation (15000 g, 1 min) and the plasma was collected

and stored at -20°C until assayed for cytokines. To one of the two remaining tubes 50

μ ι LPS (E. coli serotype 055:B5; Sigma, St Louis, USA; final concentration 10 μ^πτί)

were added to stimulate cytokine production. Unstimulated samples contained only

aprotinin, but no LPS. Both tubes were incubated at 37°C for 24 hours.

For 17 (random) patients 1 tube was added containing indomethacin (0.5 μg/mL

final concentration) in the acute phase. Furthermore, for 26 (random) patients we added

2 tubes and removed the plasma and replaced this with a same amount of phosphate

buffered saline.

TNFa was determined by a radioimmunoassay (RIA) as described earlier (detec

tion level 100 pg/mL) [25]. Normal values for our laboratory: circulating concentrations

and ex vivo production without LPS below detection limit, ex vivo production after 24

hours stimulation with LPS 3780 ± 950 pg/mL. IL-lß was measured by RIA according

to Lisi et al., but without chloroform extraction (detection level 140 pg/mL) [26]. Nor

mal values for our laboratory: circulating concentrations and ex vivo production without

LPS below detection limit, ex vivo production after 24 hours stimulation with LPS 6930

±3160 pg/mL. IL-6 was measured by an ELISA as described (detection level 20

pg/mL) [27]. Normal values for our laboratory: circulating concentrations and ex vivo

production without LPS below detection limit. IL-8 was measured by ELISA (Quanti-

36 Chapter 3

kine, R&D Systems, Abingdon, UK), detection limit 45 pg/mL, normal values below

detection limit. TNFß (lymphotoxin-α) was measured by ELISA (Quantikine). In our

laboratory, we did not ever measure any detectable TNFß. IL-IRA was determined by a

RIA according to Poutsiaka et al. [28] (detection level 300 pg/mL). Normal values for

our laboratory: circulating concentrations and ex vivo production without LPS below

detection limit, ex vivo production after 24 hours stimulation with LPS 5757 ± 1060

pg/mL. sTNF-R were measured by an enzyme linked inumino binding assay ELIBA

(Hoffmann-La Roche, Basel, Switzerland) (detection level 80 pg/mL for p55 and 300

pg/ml for p75). Normal values: circulating concentrations 1.50 ng/mL (p55) and 2.51

ng/mL (p75). All samples from the same patient were analyzed in the same run in dupli

cate to minimize analytical errors.

Statistics

When frequency distribution was paramedical, paired and unpaired Student's t-test

were used. If not, Wilcoxon signed-rank test or Mann-Whitney U test were used. Ρ <

.05 was considered significant.

Results

Circulating cytokines and inhibitors during the acute phase and convalescence

Concentrations of pyrogenic cytokines in the acute phase (IL-Iß, IL-6, TNFa, lym-

photoxin) are shown in figure 1. IL-Iß concentrations were below the detection limit in

both acute and convalescent phases. IL-6 concentrations ranged from undetectable (<

20 pg/mL) to 600 pg/mL (median 73). TNFa concentrations ranged from below detec

tion limit to 310 pg/mL (median 110) in the acute phase and to 300 pg/mL (median was

below detection limit) during convalescence. All lymphotoxin concentrations were

below detection limit as well during the acute as during the convalescent phase. IL-8

concentrations were detectable in the acute phase (median 145, range 47 - 998 pg/mL),

but lower during convalescence (median 46, range 46 - 180).

However, nanograms per mL of inhibitors like IL-IRA and sTNF-R (p55 and p75)

were measured in the acute phase, concentrations being significantly higher than during

convalescence: IL-IRA 2304 ± 1427 pg/mL vs 469 ± 324 and sTNF-R55 4973 ± 2644

vs 1671 ± 532 and sTNF-R75 22865 ± 15143 vs 5971 ± 2750 pg/mL (figure 2).

No differences were found in circulating cytokines or inhibitors between the 16

patients with complicated and the 28 with uncomplicated disease course.


1000-,

800.

600.

400.

200.

0 .

о

о

О о

» ^ J I +

о

о

о

о

*» — 8 — θ 8

IL-6 TNF IL-1 ß IL-8

Figure 1 Circulating concentrations of pyrogenic cytokines IL-IB, TNFct, IL-6, and IL-8 in patients

during acute phase of typhoid fever Patients had been ill >1 week Horizontal continuous circles =

detection limit, horizontal bars = median values In comparison with normal values IL-6, IL-8 and TNFct

are slightly elevated

Ex vivo production of cytokines and inhibitors during the acute phase and conva

lescence.

Unstimulated whole blood cultures did not show any detectable IL-Iß, TNFa or

IL-6 (not shown). After incubation with LPS for 24 hours, the supematants contained

detectable amounts of these cytokines, which were significantly lower in the acute

phase than in convalescence (IL-lß 2547 ± 3319 pg/mL vs 6576 ± 6275, ρ < .001;

TNFa 2609 ± 2443 vs 6338 ± 4366, ρ < 001; IL-6 2416 ± 1531 vs 7713 ± 3809, ρ =

.01; figure 3).

In the acute phase a correlation was found between the LPS-stimulated production

of IL-lß and TNFa (r = 0.664), IL-lß and IL-6 (r = 0 531) and between TNFa and IL-6

(r = 0.434). Such correlations were not found during convalescence.

When indomethacin was added to the stimulated samples, the concentrations of

produced TNFa and IL-lß in the acute phase were not different from those without

indomethacin (2859 ± 2630 vs 2609 ± 2443 and 2782 ± 2821 vs 2547 ±3319 pg/mL for

TNFa and IL-lß, with and without indomethacin respectively) Also, the removal of

plasma and replacement with phosphate buffered saline did not change the stimulated

production of TNFa and IL-lß in the acute phase (TNFa 3307 ± 3920 and IL-lß 2244

±2512)

38 Chapter 3

acute convalescent acute convalescent acute convalescent

Figure 2 Individual concentrations of circulating IL-1RA, sTNF-R 55 and sTNF-R 75 in patients during

the acute and convalescent phases of typhoid fever. Inhibitors like IL-IRA and sTNF-R (p55 and p75)

were significantly higher in acute phase than during convalescence.

The IL-IRA concentrations found in unstimulated cultures (not shown) were

similar as the circulating TL-IRA during the acute phase. However, the LPS-stimulated

production of IL-IRA was high and reached significantly higher concentrations in the

acute phase of the disease (5608 ± 1832 pg/mL) than during convalescence (3977 ±

1974 pg/mL; ρ < .05). sTNF-R, IL-8, lymphotoxin, were not generated in the cultures.

Ex vivo production of cytokines and inhibitors in the complicated and uncomplica

ted course of disease

In the acute phase, patients with complicated disease had significantly lower IL-Iß

production after ex vivo stimulation with LPS (1341 ± 1373 pg/mL vs 6563 ± 1342; ρ <

.005) and a trend towards lower TNFa production (1650 ± 1407 pg/mL vs 3064 ± 2770,

ρ = .06). Such differences were not found for the production of IL-IRA or IL-6. At

convalescence, the ex vivo stimulated production of IL-Iß and TNFa between compli

cated and uncomplicated cases did not differ.

Discussion

In this study we found several signs of cytokine activation during typhoid fever.

The concentrations of circulating inhibitors like IL-IRA and sTNF-R were high in the

acute phase of the disease. IL-IRA is known not to be detectable in the circulation of

normal subjects and normal values for sTNF-R are 1.50 ng/mL (p55) and 2.51 ng/mL


convalescence

Figure 3. Individual production capacity of IL-IB, TNFa and IL-6 in acute and during convalescent

phases of typhoid fever. After incubation with LPS for 24 h, supernatants contained detectable amounts

of these cytokines, which were significantly lower in acute phase than during convalescence.

(p75) [29, 30]. We also found that the production capacity of pyrogenic cytokines in

whole blood is depressed in the acute phase of typhoid fever. This capacity restores

during convalescence. Although the patients in the present study generally had severe

typhoid fever, we found that patients with a complicated course had significantly lower

proinflammatory cytokine production capacity than the uncomplicated cases.

A low production capacity of cytokines has been found in other serious conditions

like severe post-operative infection [31], sepsis [32-34], and attacks of familial Mediter

ranean fever [35, 36]. In these reports, the investigators all used isolated peripheral

blood mononuclear cells or tissue macrophages. Like in the present paper, we have also

found depressed cytokine production capacity in the whole blood culture system during

the acute phase of meningococcal disease and of Pneumocystis carinii infection [37,

38]. From these studies and the work presented here, we may conclude that the depres

sed cytokine production capacity is not a consequence of a lower number of white

blood cells during the acute phase of the infection, as we did not find a correlation

between leucocyte count and cytokine production (Table 1).

Most investigators [32-36] have interpreted the finding of low cytokine production

capacity as exhaustion of cytokine-producing cells, which could be a consequence of

exposure in vivo to stimuli such as endotoxin. Our finding that IL-1 receptor antagonist

is being produced in high concentrations argues against such a hypothesis and excludes

also that the decreased production of pro-inflammatory cytokines is due to an enhanced

lysis of producing cells or to increased inactivation of LPS by lipids in the acute phase.

40 Chapter 3

Although we have not been able to demonstrate that the proinflammatory cytokines and

IL-IRA are being produced by the same kind of cells, it is our hypothesis that after the

initial phase of infection, cytokine-producing cells switch from a balanced proinflam

matory to an anti-inflammatory repertoire. Our findings that patients in the acute phase

of typhoid fever have high concentration of soluble TNF receptors in their blood is in

agreement with this notion.

Since the cultures were performed with whole blood, we investigated whether

some circulating common factor could be responsible for the correlated low production

capacity of the cytokines IL-Iß, TNFa and IL-6 in the acute phase. Cyclooxygenäse

products, such as PGE2, which are known to inhibit production of IL-1 and TNFa [39]

were not responsible, since addition of indomethacin to the whole blood cultures did not

lead to significant changes in cytokine production. Likewise, removal of plasma and

addition of saline before incubation did not overcome the suppression in the acute phase

of the disease.

It is possible that exposure in vivo to other inhibitory factors explains the low cyto

kine production capacity. Schindler et al [40] demonstrated that exposure of isolated

mononuclear cells to IL-6 inhibits the production of IL-1 and TNFa. In the present

study we could not find any correlation between IL-6 concentrations in plasma and the

magnitude of the production of IL-lß and TNFa (r = 0.041, г = 0.035 respectively).

Exposure to other cytokines such as IL-4, IL-10 and TGFß could, however, play a role

here. Recently, Vannier et al provided evidence that exposure of cells to IL-4 suppres

ses the IL-1 production but upregulates the synthesis of IL-Ira [41].

With few exceptions, patients with typhoid fever have a continuous fever. Hence,

pyrogenic cytokines would be expected to be present in the circulation in the acute

phase of the disease. In our series of febrile patients with typhoid fever, we were not

able to detect appreciable concentrations of the pyrogenic cytokines IL-lß, TNFa and

lymphotoxin. The concentrations of IL-6, generally considered a relatively weak pyro

gen [42], were low as compared to other febrile conditions [3, 9]. We could however

detect elevated concentrations of IL-8, but this cytokine is considered non-pyrogenic

[43]. Thus, the question which pyrogens are responsible for the continuous fever in

typhoid fever remains unanswered at the present time.

Acknowledgments

We would like to thank Dr. James Vannice (Synergen) for his supply of reagents

for IL-1RA measurements.


References

1. Greisman SE. The role of endotoxin during typhoid fever and tularemia in man. J

Clin Invest 1969; 48: 613-629.

2. Cannon JG, Thompkins RG, Gelfand JA, Michie HR, Stanford GG, Van der Meer

JWM, Endres S, Lonnemann G, Corsetti J, Chemow B, Wilmore DW, Wolff SM,

Burke JF, Dinarello CA. Circulating interleukin-1 and tumor necrosis factor in

septic shock and experimental endotoxin fever. J Infect Dis 1990; 161: 79-84.

3. Hack CE, De Groot ER, Felt-Bersma RJF. Increased plasma levels of interleukin-6

in sepsis. Blood 1989; 74: 1704-1710.

4. Mitchie HR, Spriggs DR. Manogue KR. Tumor necrosis factor and endotoxin indu

ce similar metabolis responses in human beings. Surgery 1988; 104: 280-285.

5. Fischer E, Marano MA, Barber AE. Comparison between the effects of interleukin-

1 alpha administration and sublethal endotoxemia in primates. Am J Physiol 1991;

261: R442-452.

6. Okusawa S, Gelfand JA, Ikejima T, Connolly RJ, Dinarello CA. Interleukin-1 indu

ces a shock-like state in rabbits. J Clin Invest 1988; 81:1162-1172.

7. Calandra Τ, Baumgarten JD, Grau GE. Prognostic values of tumor necrosis factor/

cachectin, interleukin-1 interferon-alpha and interferon-gamma in the serum of pa

tients with septic shock. J Infect Dis 1990; 161: 982-987.

8. Damas Ρ, Reuter A, Gysen Ρ, Demonty J, Lamy M, Franchimont P. Tumor necrosis

factor and interleukin-1 serum levels during severe sepsis in humans. Crit Care

Med 1989; 17: 975-978.

9. Waage A, Brandtzaeg P, Halstensen A, Kierulf P, Espevik T. The complex pattern

of cytokines in serum from patients with meningococcal septic shock. Association

between interleukin-6, interleukin-1 and fatal outcome. J Exp Med 1989; 169: 333-

338.

10. Grau GE, Fajardo LF, Piguet PF, Allet В, Lambert PH, Vassalli P. Tumor necrosis

factor (cachectin) as an essential mediator in murine cerebral malaria. Science

1987;237: 1210-1212.

11. Barnes PF, Chatterjee D, Brennan PJ, Rea TH, Modlin RL. Tumor necrosis factor

production in patients with leprosy. Infect Immun 1992; 60: 1441-1446.

12. Havell EA. Production of tumor necrosis factor during murine listeriosis. J Immu

nol 1987; 139: 4225-4231.

13. Liew FJ, Parkinson C, Millot S, Severn A, Carrier M. Tumor necrosis factor (TNF)

in leishmaniasis 1 .TNF mediates host protection against cutaneous leishmaniasis.

42 Chapter 3

Immunology 1990; 69: 570-573.

14. Nakane A, Minagawa T, Kato K. Endogenous tumor necrosis factor (cachectin) is

essential to host resistance against Listeria monocytogenes infection. Infect Immun

1988; 56: 2563-2569.

15. Silva CL, Foss NT. Tumor necrosis factor in leprosy patients. J Infect Dis 1989;

159: 787-790.

16. Titus RG, Sherry B, Cerami A. Tumor necrosis factor plays a protective role in ex

perimental murine cutaneous leishmaniasis. J Exp Med 1989; 170: 2097-2104.

17. Nakano Y, Onozuka К, Terada Y, Shinomiya H, Nakano M. Protective effect of re

combinant tumor necrosis factor-α in murine salmonellosis. J Immunol 1990; 144:

1935-1941.

18. Nauciel C, Espinasse-Maes F. Role of gamma-interferon and tumor necrosis factor

alpha in resistance to Salmonella typhimurium infection. Infect Immun 1992; 60:

450-454.

19. Tite JP, Dougan G, Chatfield SN. The involvement of tumor necrosis factor in im

munity to Salmonella infection. J Immunol 1991; 147: 3161-3164.

20. Peel JE. Induction of circulating tumor necrosis factor cannot be demonstrated

during septicemic salmonellosis in calves. Infect Immun 1990; 58: 439-42.

21. Roine I, Herrera Ρ, Ledermann W, Peltola Η. Tumor necrosis factor-α, interleukin-

lß and interleukin-6 levels in typhoid fever. 30th Interscience Conference on Anti

microbial Agents and Chemotherapy. Atlanta, GA, 1990: Abstract 299.

22. Butler T, Ho M, Acharya G, Tiwari M, Gallati H. Interleukin-6, gamma interferon,

and tumor necrosis factor receptors in typhoid fever related to outcome of anti

microbial therapy. Antimicrob Agents Chemother 1993; 37: 2418-2421.

23. Van Deuren M, Van der Ven-Jongekrijg J, Keuter M, Demacker PNM, Van der

Meer JWM. Cytokine production in whole blood cultures. J Int Fed Clin Chem

1993;5:216-221.

24. Nerad JL, Griffiths JK, Van der Meer JWM, Endres S, Poutsiaka DD, Keusch GT,

Bennish M, Salam MA, Dinarello CA, Cannon JG. Interleukin-lß (IL-lß), IL-1 re

ceptor antagonist, and TNF-α production in whole blood. J Leukocyte Biol 1992;

52: 687-692.

25. Van der Meer JWM, Endres S, Lonnemann G, Cannon JG, Dcejima T, Okusawa S,

Gelfand JA, Dinarello CA. Concentrations of immunoreactive human tumor necro

sis factor alpha produced by human mononuclear cells in vitro. J Leucocyte Biol

1988; 43: 216-223.

26. Lisi PJ, Chu CW, Koch GA, Endres S, Lonnemann G, Dinarello CA. Development


and use of radioimmunoassay for human interleukin-lß. Lymphokine Res 1987; 6:

229-244.

27. Barrera Ρ, Boerbooms AMT, Janssen EM. Circulating soluble TNF receptors and

interleukin-2 receptors, tumor necrosis factor-alpha and interleukin-6 in rheumatoid

arthritis. Longitudinal evaluation during methotrexate and azathioprine therapy.

Arthritis Rheuma 1993; 36: 1072-1079.

28. Poutsiaka DD, Clark BD, Vannier E, Dinarello CA. Production of interleukin-1 re

ceptor antagonist and interleukin-lß by peripheral blood mononuclear cells is diffe

rentially regulated. Blood 1991; 78: 1275-1281.

29. Dinarello CA. Interleukin-1 and interleukin-1 antagonism. Blood 1991; 77: 1627-

1652.

30. Shapiro L, Clark BD, Orencole SF, Poutsiaka DD, Granowitz EV, Dinarello CA.

Detection of tumor necrosis factor soluble receptor p55 in blood samples from

healthy and endotoxemic humans. J Infect Dis 1993; 167: 1344-1350.

31. Luger A, Graf H, Schwarz HP, Stummvoll HK, Luger TA. Decreased serum inter

leukin-1 activity and monocyte interleukin-1 production in patients with fatal

sepsis. Crit Care Med 1986; 14:458-461.

32. Simpson SQ, Modi H, Balk RA, Bone RC, Casey LC. Reduced alveolar macropha

ge production of tumor necrosis factor during sepsis in mice and man. Crit Care

Med 1991; 19: 1060-1066.

33. Srugo I, Berger A, Lapidot Z, Katz R, Pollak S. Interleukin-1 secretion by blood

monocytes of septic premature infants. Infection 1991; 3: 150-154.

34. Helminen M. Interleukin-1 production from peripheral blood monocytes in septic

infections in children. Scand J Infect Dis 1991; 23: 607-611.

35. Rozenbaum M, Katz R, Rozner I, Pollack S. Decreased interleukin-1 activity relea

sed from circulating monocytes of patients with familial Mediterranean fever

during in vitro stimulation by lipopolysaccharide. J Rheumatol 1992; 19: 416-418.

36. Schattner A, Lachmi M, Livneh A, Pras M, Hahn T. Tumor necrosis factor in fami

lial Mediterranean fever. Am J Med 1991; 90: 434-438.

37. Van Deuren M, Van der Ven-Jongekrijg J, Demacker PNM, Bartelink AKM, Van

Dalen R, Sauerwein RW, Gallati H, Vannice JL, Van der Meer JWM. Differential

expression of proinflammatory cytokines and their inhibitors during the course of

meningococcal infections. J Infect Dis 1994; 169: 157-161.

38. Perenboom RM, Van Schijndel ACHW, Beckers P, Sauerwein R, Van Hamersvelt

HW, Festen J, Gallati H, Van der Meer JWM. Cytokine profiles in broncho-

alveolar lavage fluid and blood in HIV-seronegative patients with Pneumocystis

44 Chapter 3

carinii pneumonia. Eur J Clin Invest 1994; 26: 159-166.

39. Endres S, Cannon JG, Dempsey RA, Sissson S, Ghorbani R, Lonnemann G,

Dinarello CA. In vitro production of IL-Iß, IL-la, TNF and IL-2 in healthy

subjects: distribution, effect of oral cyclooxygenase inhibitors and evidence of

independent gene regulation. Eur J Immunol 1989; 19: 2327-2333.

40. Schindler R, Mancilla J, Endres S. Correlations and interactions in the production

of interleukin-6 (IL-6), interleukin-1 (IL-1) and tumor necrosis factor (TNF) in

human blood mononuclear cells: IL-6 suppresses IL-1 and TNF. Blood 1990; 75:

40-47.

41. Vannier E, Miller LC, Dinarello CA. Coordinated antiinflammatory effects of inter-

leukin 4: interleukin 4 suppresses interleukin 1 production but up-regulates gene

expression and synthesis of interleukin 1 receptor antagonist. Proc Natl Acad Sci

USA 1992; 89: 4076-4080.

42. Dinarello CA, Cannon JG, Mancilla J, Bishai I, Lees J, Coceani F. Interleukin-6 as

an endogenous pyrogen: induction of prostaglandin E2 in brain but not in periphe

ral blood mononuclear cells. Brain Res 1991; 562: 199-206.

43. Van Damme J. Interleukin-8 and related molecules. In: Thomson AW, ed. The

cytokine handbook. London: Academic Press, 1991.

Chapter 4

Phospholipase A2 is a circulating mediator in typhoid fever

Monique Keuter, Edi Dhaimana, Bart-Jan Kullberg, Casper Schalkwijk, M. Hussein

Gasem, Liesbeth Seuren, Robert Djokomoeljanto, Wil M. V. Dolmans, Henk Van den

Bosch and Jos W.M. van der Meer

J Infect Dis 1995;172:305-308

46 Chapter 4

Abstract

Circulating proinflammatory mediators have not been found in studies on typhoid

fever, although the patients suffer from a systemic disease with characteristic protracted

fever. The 14-kDa group Π extracellular phospholipase A2 (PLA2) is induced by inter-

leukin-1 (IL-1) and tumor necrosis factor (TNF), and may mediate some of the effects

of these cytokines. Circulating PLA2 concentrations were measured in 12 typhoid fever

patients on various days after admission and after recovery. On admission, mean

concentrations of PLA2 were elevated (1444 ± 1560 ng/ml) and decreased gradually

and significantly to day 14 (55 ± 48 ng/ml). Patients with complicated disease had

significantly higher PLA2 levels on admission. PLA2 was not produced in a lipopoly-

saccharide-stimulated whole blood culture, indicating that PLA2 originates from other

types of cells. These data indicate that PLA2 may be a mediator of disease in protracted

inflammatory diseases such as typhoid fever.

PLA2 in typhoid fever 47

Introduction

Despite the presence of protracted fever and other generalized signs of illness in

typhoid fever, several studies have failed to demonstrate circulating proinflammatory

mediators in these patients. In a previous report, we presented evidence for cytokine

activation in terms of raised interleukin-1 receptor antagonist (IL-IRA) and soluble

tumor necrosis factor receptors (sTNF-r) in patients with typhoid fever, but no apparent

rise in circulating concentrations of the proinflammatory cytokines interleukin-1 ß (IL-

lß), tumor necrosis factor alpha (TNFa) and interleukin-6 (IL-6) [1]. This is in agree

ment with other observations in patients with typhoid fever in Nepal and Indonesia [2,

3], although one study revealed elevated cytokine levels in a minority of patients [4].

Phospholipases are lipolytic enzymes which catalyze the degradation of phospho

lipids. To date, three varieties of phospholipase A2 (PLA2) have been characterized:

group I (pancreatic) and group Π (non-pancreatic) 14kD PLA2 and a cytosolic (85kD)

PLA2. The group Π PLA2 occurs in and is secreted by a variety of cells, and has been

implicated in the generalized inflammatory responses in several experimental models

and clinical syndromes such as sepsis and adult respiratory distress syndrome (ARDS)

[5]. Its release is induced by IL-1 and TNFa and the enzyme mediates the production of

arachidonic acid [6, 7]. Some of the metabolic effects of these cytokines during infec

tion may therefore be mediated by PLA2, potentially giving this enzyme a central role

in inflammation [8].

The absence of sizable proinflammatory cytokinemia in typhoid fever made us

consider extracellular PLA2 a candidate circulating mediator of the systemic response.

Besides this we wanted to relate its presence to cytokine responses and severity of

disease. Therefore we measured circulating PLA2 and pyrogenic cytokines and their

inhibitors sequentially. In addition, we examined the capacity of peripheral blood cells

to produce PLA2, IL-Iß, TNFa, and IL-Ira ex vivo of patients with typhoid fever ad

mitted to the hospital.

Patients and Methods

The study was conducted in Dr. Kariadi Hospital, Diponegoro University, Sema-

rang, Indonesia, from December 1990 onward. Patient selection and treatment has been

described elsewhere [1]. In brief, the diagnosis of typhoid fever was confirmed by posi

tive blood or bone marrow culture in all patients, and treatment was with chlorampheni

col. Most patients were discharged 7 to 10 days after defervescence, which was the defi-

48 Chapter 4

nition of convalescence. None of the patients died.

Measurements of cytokines and other mediators

On admission and on day 2, 4 and 6 and after recovery (day 14), blood was drawn

for cytokine and PLA2 measurements. Circulating concentrations and ex vivo produc

tion of cytokines were determined as described earlier[l, 9]. Endotoxin was measured in

platelet-rich plasma by a chromogenic limulus amoebocyte lysate assay (Kabi Vitrum,

Stockholm, Sweden). TNFa was measured by RIA as described [10] (detection level,

30 pg/ml). This RIA detects both free TNFa and TNFa bound to its soluble receptors.

IL-Iß was measured by RIA according to Lisi et al but without chloroform extraction

(detection level 30 pg/ml)[ll]. IL-1RA was determined by a RIA as described by

Poutsiaka et al. [12] (detection level 300 pg/ml). sTNF-R were measured by an enzyme

linked immuno binding assay ELIBA (Hoffmann-La Roche, Basel, Switzerland)

(detection level 80 pg/ ml for p55 and 300 pg/ml for p75).

In a group of healthy controls normal values for our laboratory were as follows.

TNFa: circulating concentrations and ex vivo production without LPS 106 ± 25 pg/ml,

ex vivo production after 24 hours stimulation with LPS 3780 ± 950 pg/ml. IL-lß:

circulating concentrations and ex vivo production without LPS below detection limit, ex

vivo production after 24 hours stimulation with LPS 6930 ±3160 pg/ml. IL-IRA:

circulating concentrations and ex vivo production without LPS below detection limit, ex

vivo production after 24 hours stimulation with LPS 5757 ± 1060 pg/ml. For sTNF-R:

Normal values: circulating concentrations 1.50 ng/ml (p55) and 2.51 ng/ml (p75).

PLA2 measurements

PLA2 activity was assayed with 0.2 mM l-acyl-2[l-C14]linoleoylphosphatidyl-

ethanolamine (specific activity 1000 dpm/nmol in 0.1M Tris-HCl (pH 8.5) containing

10 mM CaC12 and 0.05% Triton X-100. After incubation for 30 min at 37°C the [C14]

linoleate released was extracted by a modified Dole extraction procedure[13]. The

radioactive substrate was prepared biosynthetically as described previously [13].

Immunoreactive group II PLA2 was determined with an ELISA modified from

Smith et al. [14]. Two different monoclonal antibodies against human group Π PLA2

(kindly provided by Dr FB Taylor jr., Oklahoma Medical Research Foundation, Okla

homa City, OK) were used as coating and catching antibodies. Results were compared

with those obtained with culture medium from HepG2 cells stimulated with human IL-

6. The amount of group II PLA2 in this culture medium was assessed by comparison

with purified recombinant human group II PLA2. The lower limit of detection was 1


ng/ml and normal plasma values in 19 healthy volunteers amounted to 20 ± 7 ng/ml

(range 9-30 ng/ml).

Statistics

All samples from the same patient were analyzed in the same run in duplicate to

minimize analytical errors. Values are expressed as mean ± SD unless otherwise indi

cated. When frequency distribution was parametrical, paired and unpaired Student's t-

test were used. If not, Wilcoxon signed-rank test or Mann-Whitney U test were used.

Analysis of variance was done when considered suitable. Ρ < .05 was considered signi

ficant.

Results

Patients

Of 12 patients studied, 4 had a complicated course of disease (3 pneumonia and 1

delirium). The 9 female and 3 male (3:1) patients had a mean age of 16 years (range 14-

53). There was no difference in white blood cell count, female/male ratio or hemoglobin

between patients with a complicated or uncomplicated disease course. No circulating

endotoxin was found in any of the samples.

Circulating PLA2

On admission, mean concentrations of immunoreactive PLA2 were elevated (1444

± 1560 ng/ml) and decreased gradually and significantly to day 14 (55 ± 48 ng/ml);

figure 1). Patients with complicated disease had significantly higher mean PLA2 values

on admission (2520 ± 1284 ng/ml) than patients with uncomplicated disease (235 ± 206

ng/ml; p=.001). Immunoreactive PLA2 concentrations correlated well with PLA2 found

in the bioassay (r = 0.894; figure 1).

Ex vivo production capacity of PLA2

Unstimulated whole blood cultures (data not shown) showed PLA2 concentrations

virtually equal to circulating concentrations. After incubation with LPS for 24 hours,

the concentrations of enzyme in the supernatants did not change, indicating that no

additional PLA2 had been produced by peripheral blood cells. Cells incubated with LPS

after removal of plasma did not produce PLA2 either (data not shown).

50 Chapter 4

Days from admission

Figure 1. Circulating concentrations of immunoreactive PLA2 in 12 patients with typhoid fever.

Inset: correlation between bioactive and immunoreactive PLA2 in patients with typhoid fever.

Circulating cytokines and inhibitors On admission the mean concentrations of circulating proinflammatory cytokines

were: IL-Iß 95 ± 24 pg/ml and TNFa 183 ± 56 pg/ml. IL-Iß concentrations remained

low; TNFa concentrations decreased gradually but significantly to 120 ± 53 on day 14

(p < .05). There was no difference in circulating concentrations of TNFa on admission

between patients with complicated disease and those with uncomplicated disease.

Relatively high concentrations of circulating inhibitors were found on admission,

which also decreased gradually but significantly during the hospital stay. The mean

concentration of IL-IRA decreased from 1329 ± 498 pg/ml on day 0 to 318 ± 129 pg/ml

on day 14 (p < .001); sTNF-R (p55) from 7875 ± 2733 pg/ml on day 0 to 2848 ± 578

pg/ml on day 14 (p < .001); sTNF-R (p75) from 27395 ± 9796 pg/ml on day 0 to 7938 ±

2138 pg/ml on day 14 (p < .001). Patients with complicated disease had significantly

higher concentrations of inhibitors on admission than patients with uncomplicated dis

ease course (p < .05; table 1).


Ex vivo production of cytokines and inhibitors Similar to what we reported earlier [1], the production capacity of proinflammatory

cytokines was significantly suppressed during the acute phase of typhoid fever In these

sequential samples, the production returned to normal after day 4 for TNFa and after

day 6 IL-lß (data not shown). Throughout the penod of illness, the LPS-stimulated pro

duction capacity of IL-1RA was preserved (mean 9483 pg/ml) and did not change

Table 1 Circulating concentrations of anti-inflammatory mediators in the acute phase of typhoid fever in

patients with complicated or uncomplicated disease

inhibitor [pg/ml]

IL-1RA

sTNF-r55

sTNF-r75

complicated

[n = 4]

1825 ±484

10, 380 ±2345

34, 813 ±10108

uncomplicated

[n = 8]

1081 ±280

6621 ±1712

23,656 ±6775

P-value

< 01

< 01

< 05

Data are pg/ml, means ± SD IL-IRA, interleukin-1 receptor antagonist, sTNF-R, soluble tumor necrosis

factor receptor

Discussion

In the 12 patients with proven typhoid fever in the present study, extracellular

group II PLA2 circulated m high concentrations during the febrile phase of the infec

tion PLA2 concentrations were about 10-fold higher in patients with complicated dis

ease. Dunng the course of disease, the PLA2 concentrations fell significantly. PLA2

detected by immunoassay was bioactive The relatively low or undetectable concen

trations of proinflammatory cytokines in typhoid fever [1-3] suggest that this bioactive

PLA2 is responsible for at least some of the systemic signs in typhoid fever.

Although circulating concentrations of TNFa in the present senes were found to be

slightly albeit significantly elevated m the acute phase as compared to convalescence, it

should be stressed that these concentrations probably reflect mainly the TNFa bound to

soluble receptors, as extremely high concentrations of s-TNFreceptors were found on

admission.

Support for a role of PLA2 as a mediator of the systemic inflammatory response

comes from recent studies m rabbits in which PLA2 induced a fall in blood pressure

similar to that found with endotoxin infusion, and PLA2 inhibitor p-bromophenacyl-

bromide protected against this hypotensive effect [5]

The results of our ex-vivo production assay suggest that PLA2 is not being produ-

52 Chapter 4

ced by circulating cells, but rather outside the bloodstream (e.g. the liver or in endo

thelial cells). As IL-lß and TNFa are proximal signals for PLA2 [6,7], it is conceivable

that in typhoid fever, proinflammatory cytokines are first produced.

Production of these cytokines may be confined to the site of the infection in the tis

sues, rather than in the circulation. PLA2 locally induced by these cytokines may reach

the circulation and mediate the systemic signs of the disease [6, 15]. This concept of

PLA2 as a circulating mediator could explain why circulating cytokines are not found in

clinical syndromes that are thought to be cytokine-mediated and that can easily be

mimicked by injection of cytokines.

Important issues remain to be investigated. It is for instance not clear how intravas

cular release of PLA2 is counter regulated, nor has the role of PLA2 in the production

of arachidonate derivatives been satisfactorily defined, especially in view of the lack of

specificity of group II PLA2 for arachnoidate. Moreover, the role of other secreted

(group Γ) PLA2 as a mediator of inflammatory processes has to be elucidated. Our study

suggests that PLA2 is a crucial mediator in a protracted inflammatory disease such as

typhoid fever. Inhibition of PLA2 may provide a new therapeutic intervention for

febrile diseases mediated by this enzyme.

References 1. Keuter M, Dharmana E, Gasem MH, Van der Ven-Jongekrijg J, Djokomoeljanto R,

Dolmans WMV, Demacker PMN, Sauerwein R, Gallati H, Van der Meer JWM.

Patterns of proinflammatory cytokines and inhibitors during typhoid fever. J Infect

Dis 1994; 169: 1306-1311.

2. Butler T, Ho M, Acharya G, Tiwari M, Gallati H. Interleukin-6, gamma interferon,

and tumor necrosis factor receptors in typhoid fever related to outcome of anti

microbial therapy. Antimicrob Agents Chemother 1993; 37: 2418-2421.

3. McGladderly S, Larasati R, Silitonga N, Punjabi N, Lesmana M, Pulungsih S,

O'Hanley P. Acute inflammatory cytokine responses in typhoid fever. Clin Infect

Dis 1993; 17: 578.

4. Roine I, Herrera Ρ, Ledermann W, Peltola Η. Tumor necrosis factor-α, interleukin-

lß and interleukin-6 levels in typhoid fever. 30th Interscience Conference on Anti

microbial Agents and Chemotherapy. Atlanta, GA, 1990: Abstract 299.

5. Vadas P, Hay JB. Involvement of circulating phospholipase A2 in the pathogenesis

of the hemodynamic changes in endotoxin shock. Can J Physiol Pharmacol 1983;

61: 561-566.

6. Crawl RM, Stoller TJ, Conroy RR, Stoner CR. Induction of phospholipase A2 gene


expression in human hepatoma cells by mediators of the acute phase response. J

Biol Chem 1991; 266: 2647-2651.

7. Schalkwijk С, Pfeilschifter J, Marki F, Van den Bosch H. Interleukin-lß, tumor ne

crosis factor and forskolin stimulate the synthesis and secretion of group II phos-

pholipase A2 in rat mesangial cells. Biochem Biophys Res Commun 1991; 174:

268-275.

8. Pruzanski W, Vadas P. Phospholipase A2. A mediator between proximal and distal

effectors of inflammation. Immunol Today 1991; 12: 143-146.



1993;5:216-221.

10. Van der Meer JWM, Endres S, Lonnemann G, Cannon JG, Dcejima T, Okusawa S,

Gelfand JA, Dinarello CA. Concentrations of immunoreactive human tumor necro

sis factor alpha produced by human mononuclear cells in vitro. J Leucocyte Biol

1988; 43: 216-223.

11. Lisi PJ, Chu CW, Koch GA, Endres S, Lonnemann G, Dinarello CA. Development

and use of radioimmunoassay for human interleukin-lß. Lymphokine Res 1987; 6:

229-244.

12. Poutsiaka DD, Clark BD, Vannier E, Dinarello CA. Production of interleukin-1 re

ceptor antagonist and interleukin-lß by peripheral blood mononuclear cells is diffe

rentially regulated. Blood 1991; 78: 1275-1281.

13. Van den Bosch H, Aarsman AJ. A review on methods of phospholipase A determi

nation. Agents Actions 1979; 9: 382-389.

14. Smith GM, Ward RL, McGuigan L, Rajkovic IA, Scott KF. Measurement of phos

pholipase A2 in arthritis plasma using a newly developed sandwich ELISA. Br J

Rheum 1992; 31: 175-178.

15. Inada M, Tojo H, Kawata S, Tarui S, Okamoto M. Induction of group Il-like phos

pholipase A2 by lipopolysaccharide in the liver of BCG-primed rats. Biochem

Biophys Res Commun 1991; 174: 1077-1083.

54 Chapter 4

Chapter 5

A semi-quantitative reverse transcriptase polymerase chain reaction method for measurement of mRNA for TNFa and IL-Iß in whole blood cultures: Its application in typhoid fever and eccentric exercise

Mihai G. Netea, Joost P. H. Drenth, Natasja de Bont, Anneke Hijmans, Monique

Keuter, Edi Dharmana, Pierre N.M. Demacker and Jos W.M. van der Meer

Cytokine 1996; 8:739-744

56 Chapter 5

Abstract

Whole blood cultures are used to study cytokine stimulation and release ex-vivo. In

the present study we compared this method with a more direct approach and a quan

titative reverse transcriptase polymerase chain reaction (RT-PCR) was used to assess

mRNA expression for interleukin (IL)-Iß and tumor necrosis factor (TNF)ct and mRNA

in whole blood.

Stimulation of whole blood from normal donors with lipopolysaccharide (LPS) at

various time intervals showed a parallel rise of immunogenic IL-Iß and TNFct as well

as a rise of mRNA expression for IL-Iß and TNFct with peak levels for IL-Iß after 4-6h

stimulation and for mRNA TNFct expression after 2h stimulation. These methods were

used to explore cytokine production during the course of typhoid fever and after a 5 km

run.

In both conditions circulating cytokine concentrations were not influenced, but the

TNFa and IL-Iß mRNA gene expression in circulating whole blood cells was increased

in patients with typhoid fever. The LPS stimulated production of TNFa and IL-Iß was

decreased in both but there was no change for the mRNA content in whole blood for

these cytokines. These findings demonstrate that RT-PCR is an attractive method to

study the gene expression of cytokines in whole blood, an increased TNFa and IL-Iß

gene expression is present in typhoid fever, and that the LPS stimulated down regula

tion of cytokines in exercise and typhoid fever may be mediated by post transcriptional

processes.

RT-PCR for cytokine mRNA 57

Introduction

Whole blood culture (WBC) is a practical and reproducible method for assessment

of cytokine production from human cells [1, 2]. Although a variety of stimulants have

been used, we prefer the addition of bacterial lipopolysaccharide (LPS) which is added

as an immunostimulus to venous blood. After a stationary incubation for 24h at 37°C,

the release of cytokines is measured and compared to the concentration in a control

tube. Minimal amounts of blood are required and insight is provided in the capacity to

produce cytokines [3]. In healthy volunteers addition of LPS to whole blood will lead to

measurable cytokine concentrations within few hours after incubation, certainly as a

result of an increased production and or cellular release into plasma [3]. The time

course of ex-vivo production of cytokines in WBC is comparable to that in healthy vo

lunteers after a single LPS injection [4]. We have used this method to assess the cyto

kine response in various clinical situations. Application of WBC revealed that the LPS

stimulated ex-vivo production of the inflammatory cytokines tumor necrosis factor-α

(TNFoc) and interleukin-lß (EL-Iß) becomes down regulated not only in the acute phase

of infections such as typhoid fever [5], meningococcal sepsis [6], and Pneumocystis

carinii infections [7], during attacks of familial Mediterranean fever [8] and after major

surgery [9] but also after strenuous physical exercise [10]. In all these clinical situa

tions, the production capacity of these cytokines restores during convalescence.

The mechanism of the depressed cytokine production is incompletely understood.

In LPS-stimulated whole blood from patients with sepsis, Northern blot analysis

showed a decreased expression of mRNA for TNFa and IL-6 and led to the conclusion

that transcription is a decisive mechanism for the modulation of cytokine production

[И].

Apart from this study in sepsis, the molecular mechanism of down-modulation of

cytokines in the above mentioned conditions is not known. In addition, the kinetics of

cytokine mRNA expression in WBC is not fully elucidated. The present study describes

a simple and reproducible method developed to assess the IL-Iß and TNFa mRNA in

whole blood with a semi-quantitative reverse transcriptase polymerase chain reaction

(RT-PCR). This method appears to be suitable to determine the kinetics of cytokine

mRNA expression in circulating cells after stimulation with LPS. Furthermore, with this

method it is possible to evaluate the transcription of mRNA in the down-modulation of

cytokine production capacity in various clinical conditions such as physical exercise

and typhoid fever.

58 Chapter 5

Materials and methods

Controls

Four healthy volunteers participated in the study in order to obtain data for study of

the kinetics of TNFa and IL-Iß protein production and gene expression in LPS stimu

lated whole blood.

Typhoid fever

Four patients with culture proven typhoid fever were recruited in Dr. Kariadi Hos

pital, Diponegoro University, Semarang, Indonesia as part of a project on cytokines in

the pathophysiology of typhoid fever. Samples were obtained on admission (acute) and

after 7-10 days after defervescence (recovery).

Exercise

Seven well-trained males were recruited in Wijchen, The Netherlands to perform a

5 km exercise run. The distance was covered between 18 min 40 sec and 22 min 16 sec.

Samples were drawn before (recovery) and immediately after the exercise (acute).

Blood samples

Blood for cytokine measurements was collected into 2-mL or 4-mL (ex-vivo pro

duction) endotoxin-free tubes containing EDTA (Vacutainer Systems, Becton and Dick

inson, Rutherford, NJ). Cytokine production was measured using a whole blood culture

system as described elsewhere [3]. Briefly, two 2-mL tubes containing 24μL EDTA-K3

(Bayer, Leverkusen, Germany) (4 ml; 48μί EDTA-K3) were drawn. One tube was

incubated immediately, the other tube was incubated after addition of 25μί LPS

(Escherichia coli serotype 055 :B5; Sigma, St Louis, MO; final concentration 10 μg per

mL blood).

After 24 hours of incubation at 37°C the tube was centrifuged at 2250 χ g for 10

minutes and secondly at 15 000 χ g for 5 minutes to obtain platelet-poor plasma. After

4h (exercise volunteers) or 2h (typhoid fever patients) after LPS-stimulation, aliquots of

500 [¡L blood were taken, and after addition of an equivalent amount of guanidinium-

isothiocyanate (GITC) stored at -70°C until RNA isolation. For studies on kinetics of

mRNA in healthy volunteers the 4-mL EDTA tube was incubated for 24 hours and after

0,0.5, 1, 2,4, 6, 8, and 24h 500μί whole blood was taken and subsequently centrifuged

at 2250 χ g for 10 minutes; plasma was used for cytokine protein determination and the

cell pellet was dissolved in 500 μί, (4M) GITC enriched with 7 μ ι ß-mercaptoethanol


for isolation of total RNA. Aliquots were stored at -70°C until assay.

Materials

M-MLV Reverse Transcriptase, DTT, RT buffer, Agarose and Taq DNA polyme

rase were all purchased from Life Technologies (Breda, The Netherlands) RNAsin was

purchased from Promega (Leiden, The Netherlands). dNTPs and pd(N)6 were purcha

sed from Pharmacia (Woerden, The Netherlands). DNAse I was purchased from Boeh-

ringer Mannheim. (Almere, The Netherlands) Ethidium-bromide was purchased from

Sigma (St. Louis, MO). PCR primers for human TNFot, TL-Iß and ß2M were obtained

from Dr. E. Mensink, (Division of Hematology, University Hospital Nijmegen, the

Netherlands) Quantitative densitometry of the gels was performed on a Molecular Ana-

lyst™/PC densitometer (Model GS-670, Biorad, Veenendaal, The Netherlands).

RNA isolation

Total RNA was isolated by the method of Chomczynski and Sacchi with minor

modifications [12]. Briefly, 500 μ ι of whole blood was resuspended in 1 mL GITC,

and sonicated for 10 min, followed by the addition of 2M sodium acetate, phenol and

chloroform / isoamylalcohol (49:1). After centrifugation, the RNA was precipitated

twice from aqueous phase with acidified ethanol. The RNA was dissolved in RNAase-

free sterile water. The amount and quality of RNA were determined by spectrophoto

metry and analysis by agarose gel electrophoresis.

RT-PCR Analysis of cytokine mRNA and ß2-microglobuIin (ß2M) mRNA

For each sample 0.5 μg of total RNA was reverse transcribed in a volume of 20 μL·

reverse transcriptase buffer (50mM Tris-HCL pH 8.3, 75 mM KCl, 3 mM MgCl2) con

taining lOmM dithiothreitol, 5μΜ random hexamers, 250μΜ dNTPs, 20U RNAsin,

200U M-MLV reverse transcriptase. Reaction mixtures were overlaid with mineral oil.

RT reaction was performed for 10 minutes at 20°C, followed by 45 minutes at 42°C,

and finishing for 10 minutes at 95°C using a Mastercycler 5330 (Eppendorf, Hamburg,

Germany), and the samples were stored at -20°C until PCR analysis was performed.

Sequences of the PCR primers for human TNFa, IL-Iß and ВгМ are listed in Table

1. Each primer pair was tested to determine the annealing temperature and the linear

range of the reaction (Table 1). PCR reactions consisted of 3 μΐ cDNA in 50 μΐ^ PCR

buffer (20mM Tris-HCL, pH 8.4, 50 mM KCl, 1.5 mM MgCl2 , 0.001% gelatin) con

taining ΙΟΟμΜ dNTP's, 0.3μΜ of each primer and 1.25U Taq polymerase. Reaction

mixtures were overlaid with mineral oil. PCR cycles were performed (30 sec denatu-

60 Chapter 5

ration at 92°C, 30 sec annealing at 55°C and 90 sec extension at 72°C) on the same

Mastercycler 5330. The linearity of the PCR reactions was checked at different number

of cycles. The plateau phase of the ВгМ reaction became apparent after 30 cycles and of

TNFa and IL-Iß after 32 cycles. Therefore, for serial determinations of mRNA we

selected 26 cycles for ВгМ and 29 cycles for IL-Iß and TNFa. PCR products were

analyzed by electrophoresis on 2% agarose gels stained with ethidium bromide in order

to quantify the intensity of the banding pattern. Gels were scanned on a densitometer

(GS-670, Biorad) and analyzed using Molecular Analyst™ software. Linearity of the

densitometer and software was controlled in a separate experiment using consecutive

dilutions of a sample with a known DNA content. To correct for fluctuations in leuko

cyte count, we corrected for the amount of TNFa and IL-Iß mRNA in a sample by

expressing it as a ratio versus the amount of the house-keeping gene ВгМ mRNA.


Polyclonal antibodies for IL-Iß were kindly provided by Sciavo (Siena, Italy) and

antibodies for TNFa were a gift of Dr. C.A. Dinarello (Denver, CO). EL-Iß, and TNFa

in plasma were measured by non-equilibrium radioimmunoassays (RIA) as described

extensively elsewhere. (7) The sensitivity of the assay with 100 μΐ sample was 40 pg/ml

(IL-Iß) and 20 pg/ml (TNFa). To minimize analytical errors, all samples from the same

patients were analyzed in the same run in duplicate. The inter-assay variation of our

RIA is estimated at less than 15%, while the intra-assay variation is less than 10%,

which are typical figures for these assays.

Figure 1. Kinetics of TNFa (O) and IL-lfl ( · ) production (left panel, mRNA; right panel, protein) in LPS

stimulated human whole blood. Blood was obtained from normal healthy volunteers and stimulated with

LPS and incubated at 37°C for the indicated intervals. The plasma cytokine concentrations were then

measured using specific RIAs. The values shown represent mean for four individual donors.

0 1 2 3 4 5 6 7 8 23 24


Table 1 List of pnmer pairs used for mRNA amplification, with data regarding sequence, annealing tem

perature (Tm°C), size of the PCR product (bp) and number of PCR cycles within the exponential phase of

the reaction

pnmer sequence bp Tm(°C) cycles

ß2M sense S'-CCAGCAGAGAATGGAAAGTC-З· 268 55 26

antisense 5'-GATGCTGCTTACATGTCTCG-3'

TNFa sense 5'-ACAAGCCTGTAGCCCATGTT-3' 427 55 29

antisense 5'-AAAGTAGACCTGCCCAGACT-3'

IL-IB sense 5'-GGATATGGAGCAACAAGTGG-3' 263 55 29

antisense 5'-ATGTACCAGTTGGGGAACTG-3'

Statistical analysis

The non-paired non-parametric Mann-Whitney test was used for statistical compa

rison of results. Probability (p) values were calculated on the basis of two-tailed tests. A

ρ value of < 0.05 considered to be the lowest level of significance.

Results

Time course of IL-lß and TNFa gene expression

The induction of TNFa mRNA in healthy volunteers was very rapid, and high

levels were attained after 30 min. Maximum expression was found after 2h of incu

bation with LPS. After this time point, the TNFa mRNA levels declined and after 24h

incubation values were still higher compared to those at baseline. IL-lß mRNA expres

sion became increased 2h after LPS stimulation, but maximum values were reached

now until only after 4-6h. Compared to TNFa, IL-lß mRNA decreased slower (Fig la).

The kinetics of TNFa and IL-lß protein secretion production and release follow that of

the mRNA transcription (Fig lb).

Table 2 Circulating concentrations (circ) and ex-vivo production of IL-IB and TNFa in supematants of

whole blood stimulated for 24 h at 37°C with LPS in patients suffering from typhoid fever and healthy

volunteers subjected to a 5 km run

Clinical situation

Typhoid fever

Physical exercise

circ

ex-vivo

circ

ex-vivo

TNFa

acute

0 1 6 ± 0 0 2

4 4 ± 1 1

0 0 9 ± 0 0 2

3 0 ± 1

(ng/ml)

recovery

0 I 9 ± 0 1 2

7 1 ± 3 4

0 0 9 ± 0 0 3

6 9 ± 1 8

IL-lß (ng/ml)

acute

0 17±003

3 5 ± 2 8

0 0 5 ± 0 0 1

1 7 2 ± 9 9

recovery

0 14 ± 0 02

5 2 ± 4 8

0 0 6 ± 0 1 8

22±8 5

62 Chapter 5

Ladder Acute circulating B2M Acute circulating TNFo Acute circulating IL-lfl

: Recovery circulating B2M Recovery circulating TNFa Recovery circulating DL-lß

Figure 2. RT-PCR analysis of whole blood from a patient during and after the acute stage of typhoid

fever. Whole blood was stimulated with LPS and at the 2h time point mRNA was extracted and subjected

to RT-PCR. The lanes show mean values for the ratio of TNFa and IL-Iß to ВгМ expression for four

volunteers.

Circulating and ex-vivo production of cytokines

Typhoidfever

In patients with typhoid fever, circulating concentrations of IL-Iß and TNFa

remained unchanged regardless of the phase of the disease (Table 2). However, mRNA

analysis from the circulating uncultured cells of typhoid fever patients in the acute

phase of the disease revealed an increase of both IL-lß/ß2M and TNFa/ß2M ratios

when compared to the recovery phase (Fig 2).

LPS stimulated WBC of patients in the acute phase of typhoid fever showed a decreased

production of IL-1 ß and TNFa, but the production of these proteins restored during

convalescence. In contrast to the decreased ex-vivo LPS-stimulated production of the

DL-lß and TNFa proteins, the ПЛВ/ВгМ and TNFct/BîM mRNA ratios did not differ

significantly between the acute and recovery phase of typhoid fever (Table 3).

Exercise

The 5 km run did not influence the circulating IL-Iß or TNFa concentrations, nor

was there any effect on mRNA content for these cytokines in uncultured whole blood

(Table 2, 3). Similar to typhoid fever, we observed an inhibition of LPS-stimulated ex-

vivo production of both IL-Iß and TNFa in blood samples from athletes after a 5 km

run compared to pre-exercise values but the difference was larger for TNFa. Remarka

bly, the IL-Iß production capacity obtained before exercise was substantially higher in

the athletes compared to the recovered typhoid fever patients and to the values obtained

from healthy sedentary controls for our laboratory. The mRNA values for IL-Iß and

TNFa were increased by the 5 km run, although these differences were not statistically

significant (Table 3).


Table 3 mRNA for IL-Iß and TNFa from circulating whole blood cells (arc) and from whole blood cul

tures stimulated for 4 h at 37°C with LPS in patients with typhoid fever and healthy volunteers running 5

km Results are expressed as a ratio of specific cytokine to the presence of the housekeeping gene B2 M

TNFa IL-lß

Clinical situation acute recovery acute recovery

Typhoid fever arc 0 28 ± 0 06 0 19±0 19 0 58 ± 0 2 0 31 ± 0 09

ex-vivo 0 6 9 ± 0 2 0 5 9 ± 0 2 1 0 5 ± 0 1 0 9 3 ± 0 1

Physical exercise arc 0 0 0 0

ex-vivo 1 3 5 ± 0 9 0 6 2 ± 0 4 1 83 ± 1 5 13 + 0 2

Discussion

Much of the knowledge regarding the gene expression of cytokines in humans has

been obtained in isolated peripheral blood mononuclear cells A whole blood culture

system avoids possible confounding factors on gene expression that may be associated

with the isolation of mononuclear cells such as the adherence-induced increase of

TNFa mRNA [13].We used EDTA-coated sampling tubes because coagulation of

whole blood stimulates IL-lß gene expression [14] Moreover, whole blood represents a

more physiological environment for examining cytokine production to LPS because the

cellular interactions are preserved and the presence of plasma factors such as the LPS-

binding protein is maintained [15] We have developed a simple, sensitive, and semi

quantitative RT-PCR for the detection of mRNA for cytokines in whole blood, which

also could be applied in a country like Indonesia. The amount of mRNA of cytokines

was compared to the amount of the housekeeping gene ВгМ mRNA. This allows IL-lß

and TNFa gene expression to be assessed in whole blood. Because we normalized for

the presence of ВгМ, our data express the mRNA content per individual cell which cor

rects for the increase of the number of leukocytes occurring after exercise. The kinetic

data from this study confirm that incubation of LPS in whole blood leads to a very rapid

induction of TNFa [16]. The TNFa mRNA expression peaked at 2h which is in agree

ment with other studies and corroborates the notion that TNFa is a proximal mediator

of the response to LPS [17]. In contrast, the expression of mRNA for IL-lß followed a

different course with maximum mRNA levels following those of TNFa at 4-6h of incu

bation. In another study, the TNFa mRNA expression m whole blood peaked at lh post

stimulation while IL-lß reached its maximum already after 2h

Compared to the present study, relatively low LPS concentrations were used to

stimulate whole blood (10 μg/ml vs 1 ng/ml) [18] The cytokine mRNA expression in

64 Chapter S

circulating whole blood cells in typhoid fever patients could be assessed. Despite the

absence of differences of cytokine concentrations, there is an increase of mRNA expres

sion for IL-Iß and TNFa in patients with acute typhoid fever. This presence of cytokine

mRNA in circulating cells of these patients suggests that the cytokine network is activa

ted. The pathophysiological significance of these findings for typhoid fever are present

ly unclear. The absence of increased cytokine proteins in acute typhoid fever could be

due to an increased turnover of protein for these cytokines. However, it remains possi

ble that the increased expressed mRNA is not translated and thus does not lead to the

appearance of these proteins in the circulation. The absence of any change in the circu

lating TNFa and IL-Iß concentrations in athletes is associated with absence of mRNA

for these cytokines. Apparently, the stress related to a 5 km run is not intense enough to

elicit detectable cytokines in the circulation of these athletes.

Study of the mRNA expression of cytokines in LPS-stimulated whole blood in

patients with typhoid fever and athletes allowed us to make an important observation.

Despite a decreased cytokine production in the acute phase of typhoid fever and post-

exercise, both IL-Iß and TNFa mRNA expression was similar in the acute and recovery

phase. These findings suggest that under the conditions studied, the inhibition of the

LPS-stimulated ex-vivo cytokine production is a post-transcriptional event. In this

respect our results differ from those in a recent study in sepsis patients, where a decrea

sed TNFa and IL-6 mRNA expression in LPS stimulated whole blood was found [11].

This suggests that the molecular mechanism of down-modulation of cytokine produc

tion may be influenced by type of infection.

The rate of mRNA degradation is the most important mechanism of post transcrip

tional deactivation [19]. Several additional post-transcriptional processes are potentially

able to influence the presence of cytoplasmic mRNA, including capping, splicing, poly-

adenylation, nuclear export and compartmentalization [20-24]. Furthermore, kinetic

influences may play a role and the rate of degradation of IL-Iß and TNFa proteins may

precede the degradation of mRNA. Nevertheless, our results are in line with ex-vivo

experiments with LPS-tolerant mice, showing that peritoneal macrophages did not pro

duce TNFa when restimulated, while its mRNA was still induced. This strongly suggest

a posttranscriptional regulation of TNF down-regulation [25].

Acknowledgments J.P.H. Drenth is a recipient of a Dutch Organization for Scientific Research fellow

ship for Clinical Investigators. (KWO 900-716-065) The authors wish to thank the ath

letes from "Athletiekvereniging Wijchen" (Wijchen, The Netherlands) for their coope-


ration with this study. R. Krebbers and Dr. S.H.M. van Uum are thanked for their help

in performing the study.

References

1. Nerad JL, Griffiths JK, Van der Meer JWM, Endres S, Poutsiaka DD, Keusch GT,

Bennish M, Salam MA, Dinarello CA, Cannon JG. Interleukin-lß (IL-lß), IL-1

receptor antagonist, and TNF-a production in whole blood. J Leukocyte Biol 1992;

52: 687-692.

2. Finch-Arietta MB, Cochran FR. Cytokine production in whole blood ex vivo.

Agents Actions 1991; 34:49-52.



1993;5:216-221.

4. Van Zee KJ, Kohno T, Fischer E, Rock CS, Moldawer LL, Lowry SF. Tumor

necrosis factor soluble receptors circulate during experimental and clinical inflam

mation and can protect against excessive tumor necrosis factor-alpha in vitro and in

vivo. Proc Natl Acad Sci 1992; 89: 4845-4849.

5. Keuter M, Dharmana E, Gasem MH, Van der Ven-Jongekrijg J, Djokomoeljanto R,



Dis 1994; 169: 1306-1311.





7. Perenboom RM, Van Schijndel ACHW, Beckers P, Sauerwein R, Van Hamersvelt

HW, Festen J, Gallati H, Van der Meer JWM. Cytokine profiles in broncho-alveo

lar lavage fluid and blood in Ш -seronegative patients with Pneumocystis carinii

pneumonia. Eur J Clin Invest 1994; 26: 159-166.

8. Rozenbaum M, Katz R, Rozner I, Pollack S. Decreased interleukin-1 activity relea

sed from circulating monocytes of patients with familial Mediterranean fever

during in vitro stimulation by lipopolysaccharide. J Rheumatol 1992; 19: 416-418.

9. Kruimel JW, Naber AHJ, Van der Vliet AJ, Buskens FGM, Pesman G, Sweep F,

Liem T, Kho HG, Van der Meer JWM, Jansen JBMJ. Effects of abdominal surgery

on cytokines in a clinical trial designed to study the effects of parenteral admini-

66 Chapter S

stration of lipid emulsions. Neth J Med 1994; 45: A17-A18.

10. Drenth JPH, Van Uum SHM, Van Deuren M, Pesman GJ, Van der Ven-Jongekrijg

J, Van der Meer JWM. Endurance run increases circulating IL-6 and IL-Ira but

downregulates ex vivo TNF-a and IL-lß production. J Appi Physiol 1995; 79:

1497-1503.

11. Ertel W, Kremer JP, Kenney J, Steckholzer U, Jarrar D, Trentz O, Schildberg FW.

Downregulation of proinflammatory cytokine release in whole blood from septic

patients. Blood 1995; 85: 1341-1347.

12. Chomczynski P, Sacchi N. Single step method of RNA isolation by guanidinum

thiocyanate phenol chloroform extraction. Anal Biochem 1987; 162: 156-159.

13. Haskill S, Johnson C, Eierman D, Becker S, Warren K. Adherence induces selec

tive mRNA expression of monocyte mediators and proto-oncogenes. J Immunol

1988; 140: 1690-1695.

14. Mileno MD, Margolis NH, Clark BD, Dinarello CA, Burke JF, Gelfand JA. Coagu

lation of whole blood stimulates interleukin-lß gene expression. J Infect Dis 1995;

172:308-311.

15. Wright SD, Ramos RA, Tobias PS, Ulevitch RJ, Mathison JC. CD14, a receptor for

complexes of lipopolysaccharide (LPS) and LPS binding protein. Science 1990;

249: 1431-1433.

16. Oliver JC, Bland LA, Oettinger CW, Arduino MJ, McAllister SK, Agüero SM,

Favero MS. Cytokine kinetics in an in vitro whole blood model following an endo

toxin challenge. Lymphokine and Cytokine Research 1993; 12: 115-120.

17. DeForge LE, Kenney JS, Jones ML, Warren JS, Remick DG. Biphasic production

of IL-8 in lipopolysaccharide (LPS)-stimulated human whole blood. J Immunol

1992; 148:2133-2141.

18. Dedrick RL, Conlon PJ. Prolonged expression of lipopolysaccharide (LPS)-induced

inflammatory genes in whole blood cultures requires continual exposure to LPS.

Infect Immun 1995; 63: 1362-1368.

19. Sachs AB. Messenger RNA degradation in eukaryocytes. Cell 1993; 74:

20. Konarska MM, Padgett RA, Sharp PA. Recognition of cap structure in splicing in

vitro mRNA precursors. Cell 1984; 38: 731-736.

21. Amrein H, Hedley ML, Maniatis R. The role of specific protein-RNA and protein-

protein interactions in positive and negative control of pre-mRNA splicing by

transformer 2. Cell 1994; 76: 735-746.

22. Sachs A, Wähle E. Poly(A) tail metabolism and function in eukaryocytes. J Biol

Chem 1993; 268: 22955-22958.


23. Jarmolowski A, Boelens WCM, Izauralde E, Mattaj IW. Nuclear export of different

classes of RNA is mediated by specific factors. J Cell Biol 1994; 124: 627-635.

24. Hill MA, Schedlich L, Gunning P. Serum-induced signal transduction determines

the peripheral location of ß-actin mRNA within the cell. J Cell Biol 1994; 126:

1221-1230.

25. Zuckerman SH, Evans GF, Snyder YM, Roeder WD. Endotoxin-macrophage inter

action: posttranslational regulation of tumor necrosis factor expression. J Immunol

1989; 143: 1223-1237.

68 Chapter 5

Chapter 6

Pro-inflammatory cytokine profile in Salmonella

typhimurium infection in mice and the effect of recombinant IL-la

Monique Keuter, Edi Dharmana, В art-Jan Kullberg, Ineke Verschueren and

Jos W.M. van der Meer

Submitted

70 Chapter 6

Abstract

The endogenous cytokine response and the effect of recombinant human interleu-

kin-loc (rhIL-la) pretreatment in S. typhimurium infection was investigated in CBA/J

mice. Administration of 80 ng rhIL-la i.p. 24 h before infection significantly decreased

bacterial outgrowth in organs on days 3 and 7.

In plasma, circulating concentrations of tumor necrosis factor-α (TNFcc), IL-la

and IL-Iß before and after infection were low. On day 7, TNFa was 145 ± 7 pg/mL

(mean ± SD). Ex vivo LPS-induced cytokine production of blood without infection was

negligible; after Salmonella infection, production increased to day 7. Bone marrow

cells produced low cytokine concentrations without infection; in cells obtained from

infected mice, production increased. Peritoneal cells produced cytokines without infec

tion: IL-la 4081 ± 1554 pg/mL, IL-lß 68 ± 25 pg/mL and TNFa 414 ± 185 pg/mL. In

infected cells, production increased, except for IL-la.

Pretreatment with rhIL-la did not affect circulating cytokines but increased LPS-

induced IL-lß production of blood ex vivo at day 7 (1234 ± 765 vs 2137 ± 581 pg/mL

(p < .005)). RhIL-la pretreatment without infection significantly increased production

of IL-la by peritoneal (to 6190 ± 2190 pg/mL: ρ < .01) and bone marrow macrophages

(from 57 ± 14 pg/mL to 161 ± 69 pg/mL: ρ < .0001), respectively.

In conclusion, production of IL-lß and TNFa increased during S. typhimurium in

fection. Pretreatment with rhIL-la decreased outgrowth of microorganisms and increa

sed endogenous IL-la production of uninfected bone marrow and peritoneal cells.

rIL-1 in S. typhimunum infection 71

Introduction

Antibiotic therapy has reduced complications and mortality in typhoid fever, the

most important Salmonella infection in man, but the mortality of the disease is still 5%

[1]. Therefore, methods are sought to increase resistance of the host against S. typhi. To

this end, more insight in the pathophysiology of Salmonella infections is required. In

studies in patients with typhoid fever, we have measured production of cytokines after a

stimulus such as lipopolysaccharide (LPS) in a whole blood culture system [2]. Howe

ver, sequential cytokine responses in the blood and information on cytokines at the

tissue level cannot be obtained in humans.

Therefore, the mouse model of typhoid fever, induced with S. typhimurium, can

provide useful information on host defense in typhoid fever. Low concentrations of

circulating proinflammatory cytokines have been reported in this experimental infection

model in C57B1/6J mice [3]. However, no data exist on the cytokine production of mu

rine macrophages infected with Salmonella in the experimental model of typhoid fever.

Pretreatment with recombinant interleukin-1 (rIL-1) has been shown to ameliorate

the outcome of Gram-negative and Gram-positive infections in normal and neutropenic

mice. It also increases resistance against candidal and plasmodial infections in mice [4-

7]. The mechanism of this protection is multifactorial, the main mechanism in Gram-

negative infections being a down-regulation of TNFcc production and a modulation of

cytokine receptors [8]. In addition IL-1 pretreatment augments antifungal activity by a

mechanism that has not been fully elucidated [4].

Recent studies on experimental S. typhimurium infection in mice have shown that

rIL-1 pretreatment increased survival of CBA/J mice [9]. This finding is remarkable,

since rIL-1-induced inhibition of TNFa production might be deleterious for the defense

against facultative intracellular Gram-negative bacteria such as the salmonellae. TNFa

has been shown to contribute to the intracellular killing of Salmonella by macrophages

[10,11].

In the present study, the cytokine profiles and bacterial outgrowth during S. typhi

murium infection were investigated in CBA/J mice, with and without pretreatment with

recombinant human (rh)IL-la.

Material and methods

Animals

Specific pathogen-free female CBA/J mice, weighing 20-25 grams, were housed

72 Chapter 6

under standard laboratory conditions and fed sterilized laboratory chow (Hope Farms,

Woerden, The Netherlands) and water ad libitum. The experiments were approved by

the ethics committee for animal experiments at the Catholic University Nijmegen.

Materials

rhIL-la, a gift from Hoffmann-La Roche, was diluted in pyrogen-free saline to a

final concentration of 0.8 \ig/n\L. The endotoxin content was < 20 pg/mg.

Salmonella typhimurium infection

A serum-resistant strain of S. typhimurium (phage type 510) was grown to statio

nary phase culture by overnight incubation at 37CC in nutrient broth (BHI Oxoid). The

LD50 in the ItyR CBA/J mice was 5 χ IO3 bacteria. СВАЯ mice were injected i.p. with

5 χ 103 cfu of S. typhimurium.

rhIL-la was given as an i.p. injection of 80 ng in 100 цЬ of pyrogen free saline at

24 hours before the infection (day -1). Control mice received pyrogen free saline.

Just before the infection with S. typhimurium and on days 1, 3 or 7 after the injec

tion, mice were killed by cervical dislocation and outgrowth of the microorganisms

from the liver and spleen was quantified. For this purpose, the liver and spleen were

removed aseptically, weighed, and homogenized in sterile saline in a tissue grinder. The

number of viable salmonellae was determined by plating several dilutions on Brilliant

Green agar (BGA) plates and cfu were counted after overnight incubation at 37°C. The

results were expressed as the log cfu per gram of tissue.

Cytokine studies

Circulating cytokines

Before infection and on days 1, 3 and 7 of infection, separate groups of mice were

anesthetized with ether to collect blood from the intraorbital plexus for the measu

rement of circulating cytokine concentrations. TNFa, IL-la and IL-Iß concentrations

were determined using specific radioimmunoassays (RIA) developed in our laboratory,

as previously described [12].

Ex-vivo cytokine production

Whole blood cells, resident peritoneal macrophages and bone marrow macro

phages were stimulated ex vivo with LPS before infection and on day 1, 3 and 7 of in

fection with S. typhimurium. Production of cytokines before and during infection of

control mice was compared with that of rhIL-la -pretreated mice.

rIL-1 in S. typhimunum infection 73

For that purpose, EDTA blood was diluted thrice in RPMI 1640 (Dutch modi

fication; Flow Laboratories, Irvine, UK) and stimulated with lipopolysaccharide (LPS)

(E. coli serotype 055:B5; Sigma, St Louis, USA; final concentration l(^g/mL). After

incubation for 24 hours at 37°C, the tubes were centrifuged and the supernatant was

stored at -70°C until assayed.

Before infection, and on days 1, 3 and 7 after the injection of S. typhimurium, peri

toneal macrophages were harvested by rinsing the peritoneal cavity aseptically with 2

ml cold PBS containing 0.38% (w/v) sodium citrate. Bone marrow cells were flushed

from the femora of the mice with 1 ml PBS. After centrifugation, cells were resuspen-

ded in RPMI 1640, containing 1 mM pyruvate, 2 mM L-glutamine and 100 μg gentami-

cin per mL, and incubated for 1 h at 37°C (lO^well) in 96-wells microtiter plates (Co-

star Corporation, Cambridge, MA). The non-adherent cells were discarded and the re

maining cell population consisted of more than 90% macrophages, as assessed by light

microscopy.

Cells (105/ well) were incubated with LPS (1 ng in 100μί of RPMI) at 37°C in 5%

CO2. After 24h, the supernatants were collected and stored at -70°C until assayed. To

the remaining macrophages, 200 \iL of RPMI was added and the cells were disrupted

by three freeze-thaw cycles to determine the cell-associated cytokines. The samples

were stored at -70°C until assayed for cytokines as described above.


Differences in concentrations of cytokines and in numbers of the microorganisms

were analyzed by the Mann-Whitney U test. Differences were considered significant at

ρ < .05. All the experiments were at least performed in duplicate.

Results

Outgrowth of Salmonella in the organs

After infection with 5 χ 103 cfu S. typhimurium, the numbers of microorganisms in

the livers and spleens rose rapidly in control mice (Figure 1). Administration of 80 ng

rhfL-la i.p. 24 hours before injection of S. typhimurium significantly decreased

bacterial outgrowth in liver and spleen on day 3 and 7 of infection (Figure 1).

Effect of rhIL-la pretreatment on peripheral blood and peritoneal cell counts

There was no difference in numbers of peripheral blood neutrophils between rhlL-

loc pretreated animals and control mice before infection and on day 3 of infection (data

74 Chapter 6

Is f о Ç4

Spleen Liver

0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 day

Figure 1 Outgrowth of S typhimunum in the liver and spleen of control СВАЛ mice on days 1, 3 and 7

after ι ρ infection with 5 χ ICH cfu of S typhimunum, and of mice that have received rhIL-la 24 hours

earlier Each point represents the mean ± SD for at least 10 animals Significant differences between

rhIL-la treated and control mice are indicated (*,Ρ < 0l,**,P< 001, Mann-Whitney U test)

not shown). No significant differences were seen in the number of peritoneal macro

phages either: 3.17 ± 1.12 χ lOfymL in control mice and 3.12 ± 1.34 χ lOfymL in rhlL-

lot pretreated mice, 24 h after pretreatment (η.s.). After l.p. infection with 5 χ 103 S.

typhimunum, the number of peritoneal macrophages equally increased in control mice

(13.22 ± 6 26 χ lOfymL) and in rhIL-la pretreated mice (13.91 ± 5.84 χ lOfymL), over

a 7 days time course (n.s.).


Circulating cytokines

Circulating concentrations of IL-la and IL-Iß were < 20 pg/mL in mice without

infection and with Salmonella infection (5 χ 103 cfu) on day 1, 3 and 7. Pretreatment

with rhIL-la had no effect on the circulating cytokine concentrations. Plasma TNFot

was below detection limit (< 40 pg/mL) on days 1 and 3 after infection, but on day 7,

TNFcc was 145 ± 7 pg/mL (p < .01). In mice pretreated with rhIL-la, TNFa concen

trations (60 ± 53 pg/mL) were not significantly different from control mice on day 7

after infection.

rIL-1 in S. typhimuríum infection 75

Ex vivo cytokine production in whole blood

Before infection, LPS-induced cytokine production of whole blood cells ex vivo

was negligible. During Salmonella infection, production of IL-la and IL-Iß significant

ly increased from day 3 to day 7 (p < .001; Figure 2). Production of TNFa significantly

rose from day 0 to day 3 (p < .01) and from day 3 to day 7 (p < .001).

The same pattern was seen after rbJL-Ια pretreatment. Pretreatment with rhIL-Ια

led to a significant increase of IL-lß production at day 7 (p < .01), but did not affect

production of IL-la or TNFa (Figure 2).

Ex vivo cytokine production by bone marrow macrophages

LPS induced IL-lß production by bone marrow cells in mice without infection was

below the detection limit. IL-la production was 57 ± 14 pg/mL and TNFa production

was 206 ± 99 pg/mL. During 5. typhimuríum infection, IL-la production significantly

increased from day 0 (without infection) to day 3 (p < .0001), whereas IL-lß and TNFa

rose non-significantly during infection. Pretreatment with rhIL-Ια significantly increa

sed the production of cell-associated IL-la before infection (p = .0001; Figure 2).

Ex vivo cytokine production by peritoneal macrophages

Production of IL-lß and TNFa was significantly increased 3 days after Salmonella

infection (p < .001). In contrast, IL-la production was significantly decreased at day 3

(p < .0001). Pretreatment with rhIL-Ια only increased the IL-la production before in

fection (p = 0.01 ; Figure 2).

Discussion

In the present study, we demonstrate that treatment with rhIL-Ια strongly reduces

outgrowth of S. typhimuríum in the organs of mice, and this is probably an explanation

for the improved survival of rhIL-Ια-treated animals. The microbiological effect of a

single dose of rhIL-Ια was established early in the infection, and lasted throughout the

study.

A possible mechanism of the reduction in outgrowth of S. typhimuríum in the

organs includes recruitment and activation of neutrophils with enhanced killing of

microorganisms. Since granulocyte numbers were not found to be increased in our ex

periments (data not shown) and in those of others [13], the effect is probably not due to

induction of G-CSF by IL-1. This is underscored by our recent finding that recombinant

murine G-CSF increases survival of mice with 5. typhimuríum infection, but does not

76 Chapter 6

pgM

J i_

H

ч ¡1

I i

Ш

H

li

ss

pgM

i I * t

si 4

4

Ш

I i

sa

го**

S §

ш

^

ч · ?

ш Η

Figure 2. Cytokine production upon LPS stimulation of whole blood, bone marrow and peritoneal macro

phages of control СВАЛ mice, obtained on days 0, 3 and 7 after l.p. infection with 5 χ 1(P cfu of S.

typhimunum, and of mice that have received rhIL-la 24 hours earlier. Significant differences between

control and rhEL-la pretreated mice (*, Ρ < .01; **, Ρ < .001) and significant differences in production

between subsequent days are indicated (#, Ρ < 01, Mann-Whitney U test).

rIL-1 in S. typhimurium infection 77

reduce outgrowth of Salmonella in the organs [14].

Thus, we addressed the question whether the increased microbicidal effect was due

to an influence on cytokines, especially as it is known that 1L-1 is able to stimulate its

own production and that of TNFa [15]. TNFa is capable to inhibit the outgrowth of

intracellular salmonellae, probably by augmenting macrophage function [10, 11, 16].

The increased nitrite production by peritoneal macrophages after pretreatment with rlL-

1 in vivo may also play a role [17].

The effect of rhIL-la pretreatment on cytokine production was found to occur at

24 hours after administration, leading to significantly increased cell-associated IL-la

production by peritoneal and bone marrow macrophages at that time point. This sug

gests that the rhIL-Ια-induced modulation of cytokine production may have caused the

reduction of microorganisms, since the microbiological effect was also established early

in the infection.

Peritoneal macrophages of rhIL-la treated mice might have reduced the numbers

of 5. typhimurium by killing the microorganisms, before the distribution of Salmonella

to the circulation and the subsequent invasion of the organs. This would mean confi

nement of the infection to the peritoneal space and a direct effect of rhIL-la on distri

bution of Salmonella in the early phase of the infection. The outgrowth pattern does

suggest such a scenario, as the rate of outgrowth in the organs of control mice and rhlL-

l a pretreated mice is similar.

RhIL-la had no effect on outgrowth of microorganisms during infection with

Pseudomonas aeruginosa in neutropenic mice [8]. In those experiments, it has been

shown that administration of either rIL-1 or rTNFa increased survival of mice, which

was partially mediated by reduction of circulating concentrations of proinflammatory

cytokines and reduction of LPS-stimulated cytokine production by peritoneal macro

phages [8]. In contrast to extracellular Pseudomonas infections, the cause of death of

mice infected with the facultative intracellular pathogen Salmonella is most probably

organ dysfunction rather than cytokinemia. In the present animal study, concentrations

of TNFa remained low throughout the infection.

Although it has been reported previously, that TNFa is required in Salmonella

infection [10, 11, 16], we did not find an effect of rhIL-la on LPS-stimulated produc

tion of TNFa by peritoneal cells or bone marrow macrophages either before or during

S. typhimurium infection. Moreover, the beneficial effect of rhIL-la on numbers of

microorganisms slightly increased during 7 days of infection, whereas an inhibitory

effect on macrophage activation through reducing TNFa production would have led to

increased outgrowth.

78 Chapter 6

Whereas LPS-stimulated cytokine production during S. typhimurium infection was

generally increased, the IL-la production by infected peritoneal cells was decreased.

This is an important observation, since peritoneal macrophages are easy to obtain and

frequently used as representatives for the resident or activated macrophage population

of the mouse. The present study shows that the cytokine response of peritoneal and

other macrophages are divergent. The attention to peritoneal cells alone, might not yield

information on cytokine production at other sites of infection.

There are important differences in cytokine patterns between these mouse

experiments and the studies in patients with typhoid fever. During S. typhimurium in

fection in mice, the circulating TNFa concentration was increased on day 7, as well as

the stimulated production of IL-la, IL-Iß and TNFa in whole blood. Production of IL-

lß and TNFa by peritoneal and bone marrow cells was also increased during infection.

This general upregulation of proinflammatory cytokine production in this animal model

is in contrast with what we have found in humans with typhoid fever [2]. In these

patients, downregulation of LPS-stimulated production of cytokines in whole blood was

found. These differences seem not to be caused by the time point of measuring cytokine

production, as the patients with typhoid fever had also been ill for a week. More early

down-regulation of cytokine production was shown in patients with Gram-negative

infection such as meningococcemia [18]. It is obviously impossible to draw conclusions

from these animal experiments and extrapolate the results to the human situation. In

conclusion, rhIL-la pretreatment decreased outgrowth of S. typhimurium, partly by a

modest influence on peritoneal cell cytokine production.

References

1. Van den Bergh ET AM, Hussein Gasem M, Keuter M, Dolmans WMV. Outcome in

three groups of patients with typhoid fever treated in Indonesia between 1948 and

1990. 1998; submitted.




Dis 1994; 169: 1306-1311.

3. Jotwani R, Tanaka Y, Watanabe K, Tanaka K, Kato N, Ueno К. Cytokine stimula

tion during Salmonella typhimurium sepsis in ItyS mice. J Med Microbiol 1995;

42: 348-52.

4. Kullberg BJ, Van 't Wout JW, Van Fürth R. Role of granulocytes in enhanced host

rIL-1 in S. typhimurium infection 79

resistance to Candida albicans induced by recombinant interleukin-1. Infect Im

mun 1990;58:3319-3324.

5. Czuprynski CJ, Brown JF. Recombinant murine interleukin-la enhancement of

nonspecific antibacterial resistance. Infect Immun 1987; 55: 2061-2065.

6. Van der Meer JWM, Barza M, Wolff SM, Dinarello CA. A low dose of recom

binant interleukin 1 protects granulocytopenic mice from lethal Gram-negative in

fection. Proc Natl Acad Sci USA 1988; 85: 1620-1623.

7. Curfs JHAJ, van der Meer JWM, Sauerwein R, Eling WMC. Low dosages of inter

leukin-1 protect mice against lethal cerebral malaria. In: Dinarello CA, Kluger M,

Oppenheim J, Powanda M, ed. Interleukin-1 and related cytokines. New York:

AlanLiss, 1990.

8. Vogels MTE, Mensink EJBM, Ye K, Boerman OC, Verschuren CMM, Dinarello

CA, Van der Meer JWM. Differential gene expression for IL-1 receptor antagonist,

IL-1, and TNF receptors and IL-1 and TNF synthesis may explain IL-1-induced

resistance to infection. J Immunol 1994; 153: 5772-5780.

9. Morrissey PJ, Charrier К. Treatment of mice with IL-1 before infection increases

resistance to a lethal challenge with Salmonella typhimurium - the effect correlates

with the resistance allele at the ity locus. J Immunol 1994; 153: 212-219.

10. Mastroeni P, Skepper JN, Hormaeche CE. Effect of anti-tumor necrosis factor

alpha antibodies on histopathology of primary Salmonella infections. Infect Immun

1995; 63: 3674-3682.



450-454.



downregulates ex vivo TNF-oc and IL-lß production. J Appi Physiol 1995; 79:

1497-1503.

13. Kullberg BJ, Van 't Wout JW, Van Fürth R. No effect of recombinant human inter

leukin-1 on the numbers of peripheral blood and peritoneal leukocytes during an

acute inflammation. Inflammation 1991; 15:457-470.

14. Keuter M, Dharmana E, Kullberg BJ, Curfs J, Djokomoeljanto R, Van der Meer

JWM. G-CSF enhances survival of Salmonella typhimurium infection in CBA/J

mice, submitted 1998;

15. Dinarello CA, Ikejima T, Warner SJC, Orencole SF, Lonnemann G, Cannon JG,

Libby P. Interleukin 1 induces interleukin 1.1. Induction of interleukin 1 in rabbits

80 Chapter 6

in vivo and in human mononuclear cells in vitro. J Immunol 1987; 139:1902-1910.

16. Nakano Y, Onozuka К, Terada Y, Shinomiya H, Nakano M. Protective effect of re

combinant tumor necrosis factor-α in murine salmonellosis. J Immunol 1990; 144:

1935-1941.

17. Vogels MTE, Van Rooyen N, Bemelmans MHA, Jansen M, Van der Meer JWM.

Interleukin-1 (IL-l)-induced resistance to bacterial infection: role of the macropha

ge and soluble TNF receptors. In: Vogels MTE, ed. Mechanisms of enhanced natu

ral resistance to infection induced by interleukin-1. Thesis: Catholic University

Nijmegen, 1994.

18. Van Deuren M, Van der Ven-Jongekrijg J, Bartelink AKM, Van Dalen R,

Sauerwein RW, Van der Meer JWM. Correlation between proinflammatory cyto

kines and antiinflammatory mediators and the severity of disease in meningococcal

infections. J Infect Dis 1995; 172: 433-439.

Chapter 7

G-CSF enhances survival of CBA/J mice with Salmonella typhimurium infection

Monique Keuter, Edi Dhaimana, Bart-Jan Kullberg, Jo H.A.J. Curfs

and Jos W.M. van der Meer

Submitted

82 Chapter 7

Abstract

This study shows that recombinant murine granulocyte colony-stimulating factor

(rmG-CSF) improves survival of CBA/J mice with a potentially lethal Salmonella

typhimurium infection. Of the control mice, 8% survived, compared with 33% survival

in mice treated with a single dose of rmG-CSF one day before infection (p < .0001).

The outgrowth of S. typhimurium in liver, spleen, bone marrow, lung, brain, kidney and

heart on day 1, 3 and 5 was not different between the groups. Circulating cytokine con

centrations did not differ between the groups either. One day after of rmG-CSF treat

ment, the lipopolysaccharide (LPS) stimulated tumor necrosis factor-α (TNFa) produc

tion of resident peritoneal macrophages was significantly higher in treated mice: 1.125

± 0.476 ng/mL vs 0.621 ± 0.173 ng/mL in control mice (p < .01). Histological

examination of the liver on day 1 and 5 showed more localized inflammatory foci in the

rmG-CSF treated animals. These results show that activation of neutrophils by rmG-

CSF is beneficial in the facultative intracellular S. typhimurium infection in mice.

rG-CSF in S. typhimurium infection 83

Introduction

Salmonella infections, especially typhoid fever, remain significant causes of mor

bidity and mortality in tropical countries. More insight in the host response to Sal

monella is necessary to develop more effective therapeutic strategies, e.g. immunother

apy, which might improve the outcome of infection. Patients with typhoid fever usually

present with a relative leukopenia. In these patients, we found a correlation between

leukopenia and the development of complications (unpublished results). These observa

tions may suggest that granulocytes play a role in the host defense against Salmonella

infection in man.

In experimental S. typhimurium infection in mice, the bacteria induce a disease

similar to human typhoid fever. The survival of mice during the first days of infection

with the facultative intracellular S. typhimurium , before the development of specific

acquired immunity, depends on natural immunity. Neutrophils are thought to play an

important part in this phase of S. typhimurium infection in mice. Earlier studies have

shown that liver and spleen of mice infected with Salmonella typhimurium contained

many polymorphonuclear phagocytes (PMN's) and microabcesses after 24 hours of in

fection. Hardly any macrophages were found at that time point [1-3]. Blocking the mi

gration of PMN's through administration of mAb to leukocyte adhesion molecules,

lethally exacerbated the infection [1]. In addition, neutropenic mice developed a lethal

systemic salmonellosis after infection with low numbers of microorganisms [2].

Granulocyte colony-stimulating factor (G-CSF) is a hematopoietic growth factor,

which stimulates proliferation and activation of neutrophils and their precursors [4].

The effects on mature neutrophils include enhanced phagocytosis and respiratory burst

as well as reduced apoptosis. rG-CSF treatment has been shown to be beneficial in

many neutropenic animal models, but also in infections with non-neutropenic animals

[5]. In a peritonitis model in non-neutropenic mice, rG-CSF was highly effective in

improving survival [6, 7].

In models of intracellular infection, experience with rG-CSF is scarce, and most

studies have been performed with human rG-CSF, of which the activity is considered

sub optimal for mice. The present study reports survival, outgrowth of microorganisms,

and cytokines in CBA/J mice with a potentially lethal S. typhimurium infection after

treatment with murine rG-CSF (rmG-CSF), which has recently become available.

84 Chapter 7

Materials and methods

Animals

Specific pathogen-free female Ity resistant (ItyR) CBA/J mice, weighing 20-25 g

were used. The course of Salmonella infections in mice is regulated by the Ity locus on

chromosome 1, which most likely encodes for the so-called natural resistance associa

ted membrane protein 1 (Nrampl) on professional phagocytes [8-10]. The CBA/J mice

belong to the ItyR, or Salmonella resistant mice.

Animals were housed under standard laboratory conditions and fed sterilized labo

ratory chow (Hope Farms, Woerden, The Netherlands) and water ad libitum. The expe

riments were approved by the ethical committee for animals experiments at the Catholic

University Nijmegen.

Materials

Recombinant murine G-CSF, a gift from Amgen (Thousand Oaks, CA)was diluted

in pyrogen-free saline to a final concentration of 2.5 μ^πύ.


A serum-resistant strain of S. typhimurium (phage type 510) was grown to statio

nary phase culture by overnight incubation at 37°C in nutrient broth (BHI Oxoid). The

LD50 in the ItyR CBA/J mice was 0.5 χ IO4 bacteria. Mice were injected i.p. with 1 χ

IO4 cfu of S. typhimurium. rmG-CSF was given as a s.c. injection of 500 ng (25 |ig/kg)

in 100 mL of pyrogen free saline. One group of mice received 500 ng s.c. 24 hours

before the infection (Pretreatment group). The multiple treatment group received 500

ng s.c. once daily from 24 hours before infection until day 7 during infection. Control

mice received pyrogen free saline s.c. daily. Survival was assessed twice daily for 21

days in groups of at least 20 animals.

On day 1, 3 or 5 after the injection of S. typhimurium, mice were killed by cervical

dislocation and outgrowth of the microorganisms from the liver, spleen and bone

marrow was quantified in groups of at least 8 mice. For this purpose, the liver and

spleen were removed aseptically, weighed, and homogenized in sterile saline in a tissue

grinder. Bone marrow was flushed from the femur aseptically with 1 mL of sterile

saline. In 4 mice per group, outgrowth in lung, heart, kidney and brain was also quanti

fied. The number of viable salmonellae was determined by plating serial dilutions on

Brilliant Green agar (BGA) plates and cfu were counted after overnight incubation at

37°C. The results were expressed as the log cfu per gram of tissue.

iG-CSF in S. typhimuríum infection 85

Cytokine studies


On days 0,1,3 or 5, separate groups of mice (5 mice per group) were anesthetized

with ether to collect blood from the intraorbital plexus for the measurement of circu

lating cytokine concentrations. TNFa, IL-la and IL-lß concentrations were determined

using specific radioimmunoassays (RIA) developed in our laboratory, as previously

described [11].

Ex-vivo cytokine production

On day 0 (24h after rmG-CSF injection) and on day 3 after infection, resident peri

toneal macrophages of mice were harvested by rinsing the peritoneal cavity aseptically

with 2 ml cold PBS containing 0.38% (w/v) sodium citrate. After centrifugation, cells

were resuspended in RPMI1640 (Dutch modification; Flow Laboratories, Irvine, UK),

containing 1 mM pyruvate, 2 mM L-glutamine and 100 μg of gentamicin per mL, and

105 cells /well were incubated for 1 h at 37°C in 96-wells microtiter plates (Costar

Corporation, Cambridge, MA). The non-adherent cells were discarded and the remai

ning cell population consisted of more than 90% macrophages, as assessed by light

microscopy. LPS (E. coli serotype 055:B5; Sigma, St Louis, USA; final concentration

10μg/mL) was added to the peritoneal macrophages to a final volume of 100 μ ι and

incubated at 37°C in 5% C02. After 24h, the supematants were collected and stored at

-70°C until assayed. To the remaining macrophages, 200 μ ι of RPMI were added and

the cells were disrupted by three freeze-thaw cycles to determine the cell-associated

cytokine contents. The samples were stored at -70°C until assayed for TNFa, IL-la and

IL-lß concentrations as described above.

Blood samples

Blood samples (20 \iL) were collected from the retroorbital plexus and the leuko

cytes were counted in a Coulter Counter (Coulter Electronics, Luton, England). The

total numbers of granulocytes, lymphocytes and monocytes per mm3 were calculated

from the total numbers of leukocytes per mm3 and the differential counts of 200 leuko

cytes in 2 Giemsa-stained blood smears.

Histology

From other groups of animals, the liver was removed, fixed in buffered formalin

(4%), and serial sections were examined microscopically after staining with hematoxy

lin and eosin.

86 Chapter 7

100

_ 80 л .> i 60 (Я

40 •

20 '

rmG-CSF day -1 (—)

rmG-CSF days -1 - +7 (—)

Controls

12 16 20 24 28 days

Figure 1. Survival of СВАЛ mice (at least 20 animals per group) after i.p. injection of 104 S. typhi-murium. Mice were pretreated with rmG-CSF 500 ng s.c. 24h before the infection. Other mice received 500 ng s.c. once daily from 24h before infection until day 7 of infection. Control mice received pyrogen free saline s.c. daily.


Differences in concentrations of cytokines and numbers of microorganisms in the

organs of mice were analyzed by the Mann-Whitney U test. Differences were consi

dered significant at ρ < .05. All the experiments were performed at least in duplicate.

Results

Effect of rmG-CSF on the numbers of granulocytes during infection

In untreated mice, the mean number of peripheral blood granulocytes was 5.5 ± 2.2

χ 107/L. Twenty four hours after injection of rmG-CSF, the mean number was 5.7 ±1.5

χ 107/L (n.s.). On day 3 of infection, peripheral blood granulocytes had increased in all

groups, but were higher in the multiple treatment group (20.4 ± 6.8 χ 107/L) than in the

pretreatment group (14.5 ± 6.7 χ 107/L; ρ = .06). Also, peripheral blood granulocytes

were higher in the multiple treatment group (20.4 ± 6.8 χ 107/L) compared with the

control group (11.9 ± 6.2 χ 107/L; ρ < .01); the difference between the control and pre

treated mice was not significant (p = .4).

After 5 days of infection, the differences between the pretreatment (12.7 ± 3.9 χ

107/L) and control group (13.8 ± 5.3 χ 107/L), with respect to granulocyte numbers had

disappeared. Peripheral blood granulocytes were still significantly higher in the multi

ple treatment group (20.5 ± 7.5 χ 107/L; ρ < .05).

rG-CSF in S. typhimuríum infection 87

!> 6 Ь Q 5

Spleen 8-1

7-

3

t5 σι _

3 5

4

3

Liver τ

^À , ^

^ l Г ^

Γ" • 1 » Π 1 — — I — · — I

4 5 days

Figure 2. Outgrowth of S. typhimurium in the liver and spleen of CBA/J mice after i.p. injection of 104

cfu. Mice were pretreated ( · ) with rmG-CSF 500 ng s.c. 24h before the infection. The multiple treatment group (•) received 500 ng s.c. once daily from 24h before infection until day 7 of infection. Control mice (O) received pyrogen free saline s.c. Each point represents the mean ± SD for at least 10 animals. There were no significant differences.

Effect of rmG-CSF on survival during Salmonella typhimuríum infection in non-

neutropenic mice

After i.p. injection of IO4 cfu S. typhimurium, 92% of the control mice died

between day 4 and day 15 of infection, after this day none died. In the pretreatment

group, a single s.c. injection of 500 ng rmG-CSF 24h before infection resulted in a sig

nificantly better survival of 33% (p < .0001). In the multiple treatment group a 27%

survival was seen, which was also significantly different from control mice (p < .01;

Figure 1)

Effect of rmG-CSF on the outgrowth of Salmonella in the organs

On day 1, 3 and 5 after i.p. infection with 104 cfu of S. typhimurium, the outgrowth

of Salmonella was not significantly different in liver, spleen and bone marrow of

treated mice compared with that of control mice, despite the difference in survival

(figure 2). On day 5, there was a trend for an increased number of cfu in liver and

spleen in the multiple treatment group, compared to the control group and the pretreat

ment group.

Since there was no effect of rmG-CSF on the outgrowth of Salmonella in the liver,

spleen or bone marrow, it was investigated whether Salmonella had accumulated at dif

ferent locations (in other organs). However, outgrowth of S. typhimurium in the lung,

88 Chapter 7

heart, kidney and brain was similar in control and rmG-CSF treated animals (data not

shown).

Circulating cytokines Circulating concentrations of TNFct, IL-la and IL-Iß were determined on day 0

before infection (24 hours after rmG-CSF) and day 1, 3 and 5 after infection of 104 cfu

S. typhimurium. Before infection and on day 1 of infection, cytokine concentrations in

all samples were under the detection limits. No detectable concentrations of IL-Iß (<

0.020 ng/mL) were found at any time point during the infection.

On day 3, TNFa was 0.109 + 0.061 ng/mL (mean ± SD) in rmG-CSF pretreated

mice, 0.115 ± 0.075 ng/mL in the multiple treatment group and 0.127 ± 0.105 ng/mL in

control mice (n.s.). The IL-la concentrations were not significantly different in the

three groups either: 0.067 ± 0.067 ng/mL in pretreated mice, 0.049 ± 0.025 ng/mL in

the multiple treatment group and 0.062 ± 0.039 ng/mL in control mice.

On day 5, circulating concentrations of TNFa and IL-la were higher, but not

significantly different in rmG-CSF multiple dose treated groups than in control mice:

for TNFa 0.323 ± 0.232 ng/mL vs 0.264 ± 0.160 ng/mL (n.s.) and for IL-la 0.093 ±

0.085 ng/mL in the multiple treatment group vs 0.070 ± 0.042 ng/mL in the control

mice (n.s.).

Ex-vivo cytokine production To investigate whether a difference in cytokine production of macrophages after

rmG-CSF could explain the difference in outcome, resident peritoneal macrophages of

the groups of mice were harvested before infection (24h after rmG-CSF administration)

and on day 3 of infection, and cells were stimulated with LPS in vitro.

Before infection, LPS stimulated TNFa concentrations in supematants of macro

phages were significantly higher in the rmG-CSF pretreated group than in controls (p <

.01). On the same day, cell-associated IL-la production tended to be lower in the rmG-

CSF treated group than in controls (n.s.). IL-Iß concentrations in supematants of

macrophages were not different in the 2 groups (Figure 3).

On day 3 during infection, TNFa concentrations in supematants of macrophages

were not significantly higher in the rmG-CSF pretreated group (2.856 ± 2.252 ng/mL)

and multiple treatment group (3.529 ± 2.964 ng/mL) than in controls (2.112 ± 0.897

ng/mL).

Cell-associated IL-la production tended to be higher (but did not reach signifi

cance) in the rmG-CSF pretreated (2.593 ± 1.241 ng/mL) and multiple treatment group


51

Ш ém A l TNFa IL-1a IL-1B

Figure 3. Cytokine production of resident peritoneal macrophages before infection (24h after rmG-CSF administration). Macrophages were stimulated with LPS in vitro. TNFa concentrations in supematants of macrophages were significantly higher in the rmG-CSF treated group (И) than in controls (•; ρ < .01). Cell-associated IL-la production and IL-IB concentrations in supematants of macrophages were not different in the 2 groups.

(3.542 ± 3.470 ng/mL) than in controls (2.291 ± 1.069 ng/mL). IL-Iß concentrations in

supematants of macrophages were not different in the 3 groups: 0.318 ±0.110 ng/mL in

the pretreatment group, 0.294 ± 0.113 ng/mL in rmG-CSF multiple treatment and 0.248

± 0.082 ng/mL in controls (n.s.).

Histology

In the livers of mice pretreated with rmG-CSF, foci of infection with granulocytes

and mononuclear cells were smaller and more compact than in livers of control mice. In

this latter group granulocytes were diffusely scattered throughout the liver with occa

sional necrosis (Figure 4).

Discussion In the present study we found that rmG-CSF reduced mortality of CBA/J mice

during S. typhimurium infection. Innate immunity against extracellular microorganisms

is mostly based on phagocytosis and opsonization. To overcome infection with intracel

lular microorganisms, cell-mediated immunity is required involving cytotoxic T-cells

and T-cell derived cytokines, e.g. interferon-γ (IFN-γ). Natural resistance in murine

salmonellosis however, is necessary in the period before the acquired immunity is esta

blished. In the early phase of infection, Salmonella invades Kupffer-cells and adjacent

90 Chapter 7

Suf' І * I ^* .* "I"*-'¡ г

Figure 4. Histology of the liver of mice pretreated with rmG-CSF (A magnification χ 10; В magnifica

tion χ 40) and control mice (C magnification χ 10; D magnification χ 40). In rmG-CSF treated mice,

more granulocytes can be seen in compact foci in the liver.

rG-CSF in S. typhimurìum infection 91

hepatocytcs [1]. Neutrophils are important for the killing of Salmonella in that stage.

Blocking the migration of these cells through administration of mAb to leukocyte adhe

sion molecules such as CD lib/CD 18 in a sublethal S. typhimurìum infection in mice,

resulted in extracellular outgrowth of salmonellae in the hepatic sinusoids and in a fail

ure to destroy Salmonella-lozded hepatocytes, leading to a lethal infection [12]. Natural

killer (NK) cells and Τ lymphocytes are not involved in the lysis of hepatocytes. There

fore it appears, that neutrophils play an important role in the early host defense against

Salmonella infection. Enhancing the recruitment or activation of neutrophils by rG-CSF

may be helpful at this stage of the infection.

In the present study, administration of rmG-CSF had a beneficial effect on survi

val, but not on the outgrowth of Salmonella in the organs. In general the numbers of

Salmonella cfu in the liver and the spleen correlate with mortality in experimental S.

typhimurìum infection in mice [13-15], however the number of bacteria may not reflect

the extent of organ destruction. In the rmG-CSF treated animals we have seen that mi

croorganisms were contained in a granuloma, whereas there was more diffuse granulo

cytosis and necrosis in the control group.

Although rmG-CSF had increased the granulocytes, this was not the only reason

for the difference in survival on day 5, as the number of PMN's were highest in the mul

tiple treatment group and similar in the control and pretreated mice at that time point.

Apart from its effect on neutrophil function, G-CSF has antiinflammatory properties

and has been shown to be able to influence TNFa production [16, 17]. In this respect, it

is important to note that in intracellular infection, the proinflammatory cytokines play a

beneficial role, partly through activation of macrophages, resulting in better survival

[18]. Inhibition of TNFa during Salmonella infection in mice has been shown to wor

sen outcome [19, 20]. We found that the LPS-stimulated TNFa production by peri

toneal macrophages was increased 24 hours after s.c. administration of rmG-CSF.

Granulocyte-macrophage colony-stimulating factor (GM-CSF) has been known to indu

ce fever as a side-effect, which possibly is associated with its ability to induce produc

tion of TNFa by macrophages in vitro [21]. In contrast, G-CSF, which is supposed not

to induce TNFa is not associated with fever. Moreover, G-CSF protected against a

lethal injection of LPS in rodents [22]. In human volunteers pretreated with rG-CSF,

endotoxin infusion increased the peak of circulating TNFa [23]. However, it was found

that ex vivo production of TNFa in human blood was lower after G-CSF treatment [16,

17]. Apparently, differences in species (man-mouse-rat), strains, time points and way of

delivery of G-CSF can play a role in these discrepancies.

In previous studies we have shown that the clinical picture and pattern of circula-

92 Chapter 7

ting cytokines of typhoid fever in humans differs from that of Gram-negative sepsis or

endotoxinemia [24]. Also in animal models, TNFoc was detectable in the circulation

only several days after S. typhimurium infection, whereas TNFa did rise 1 hour after

LPS administration in these animals [25, 26] .The circulating TNFa concentrations

found in our previous studies on murine salmonellosis almost always seemed to be a re

sult of the outgrowth of microorganisms in the tissues [27].

It could have been envisaged that the increased local TNFa production, as we have

found in our experiments, may have led to increased activation of macrophages and

subsequently to a reduction of the outgrowth in the organs. However, we only found

significantly increased TNFa production 24 hours after administration of rmG-CSF and

not at later time points during the infection. In contrast, TNFa plays a role especially

after 4-5 days of Salmonella infection [19]. Besides, cytokine production of peritoneal

macrophages might not be representative for production by macrophages in the liver.

It may be questioned whether rG-CSF has a beneficial effect on intracellular infec

tions through enhanced (early) neutrophil recruitment, whereas the (late) antiinflamma

tory properties of rG-CSF may be deleterious, through inhibition of TNFa mediated

macrophage activation. The trend for an increased number of cfu in liver and spleen in

the multiple treatment group could be an illustration of this suggestion and may be due

to such an adverse late effect. Pretreatment alone seems beneficial as has been shown in

Listeria monocytogenes infection in mice, in which pretreatment with high dose human

rG-CSF during 5 days, reduced mortality as well as outgrowth of microorganisms [28,

29].

In conclusion, rmG-CSF protects mice against a lethal 5. typhimurium infection in

mice, although no effect was found on the outgrowth of microorganisms in the organs.

Further studies are needed to elucidate the mechanism, afore rG-CSF can be used as im

munotherapy in Salmonella infection.

References 1. Conlan JW, North RJ. Early pathogenesis of infection in the liver with the faculta

tive intracellular bacteria Listeria monocytogenes, Francisella tularensis, and

Salmonella typhimurium involves lysis of infected hepatocytes by leukocytes.

Infect Immun 1992; 60: 5164-5171.

2. Conlan JW, North RJ. Neutrophil-mediated lysis of infected hepatocytes. ASM

News 1993; 59: 563-567.

3. Hsu HS. Pathogenesis and immunity in murine salmonellosis. Microbiol Rev 1989;

53: 390-409.


4. Dale DC, Liles WC, Summer WR, Nelson S. Review: Granulocyte colony-stimula

ting factor. Role and relationships in infectious diseases. J Infect Dis 1995; 172:

1061-1075.

5. Kullberg BJ, Netea MG, Curfs JHAJ, Keuter M, Meis JFGM, Van der Meer JWM.

Recombinant murine granulocyte colony-stimulating factor protects against acute

disseminated Candida albicans infection in non-neutropenic mice. J Infect Dis

1998; 177: 175-181.

6. Nelson S. Role of granulocyte colony-stimulating factor in the immune response to

acute bacterial infection in the nonneutropenic host: an overview. Clin Infect Dis

1994;18(suppl2):S197-204.

7. Barsig J, Bundschuh DS, Härtung Τ, Bauhofer Α, Sauer A, Wendel A. Control of

fecal peritoneal infection in mice by colony-stimulating factors. J Infect Dis 1996;

174: 790-799.

8. Vidal S, Tremblay ML, Govoni G, Gauthier S, Sebastiani G, Malo D, Skamene E,

Olivier M, Jothy S, Gros P. The Ity/Lsh/Bcg locus: Natural resistance to infection

with intracellular parasites is abrogated by disruption of the Nrampl gene. J Exp

Med 1995; 182: 655-666.

9. Vidal S, Lepage Ρ, Gauthier S, Gros P. Natural resistance to intracellular infec

tions. J of Imm 1996; 157: 3559-3568.

10. Govoni G, Vidal S, Gauthier S, Skamene E, Malo D, Gros P. The BcgЯty/Lsh

locus: Genetic transfer of resistance to infections in C57B1/6J mice transgenic for

the N r a m p l G l y 1 6 9 allele. Infect Immun 1996; 64: 2923-2929.


J, Van der Meer JWM. Endurance run increases circulating IL-6 and IL-lra but

downregulates ex vivo TNF-a and IL-Iß production. J Appi Physiol 1995; 79:

1497-1503.

12. Conlan JW. Neutrophils prevent extracellular colonization of the liver microvascu-

lature by Salmonella typhimurium. Infect Immun 1996; 64: 1043-1047.

13. Morrissey PJ, Charrier К. Interleukin-1 administration to СЗН/HeJ mice after but

not prior to infection increases resistance to Salmonella typhimurium. Infect Immun

1991;59:4729-4731.


resistance to a lethal challenge with Salmonella typhimurium - the effect correlates

with the resistance allele at the ity locus. J Immunol 1994; 153: 212-219.

15. Van Dissel JT, Stikkelbroeck JJM, Sluiter W, Leijh PCJ, Van Fürth R. Differences

in initial rate of intracellular killing of Salmonella typhimurium by granulocytes of

94 Chapter 7

salmonella susceptible C57B1/10 mice and salmonella resistant CBA mice. J

Immunol 1986; 136: 1074-1080.

16. Kitabayashi A, Hirokawa M, Hatano Y, Lee M, Kuroki J, Niitsu H, Miura AB.

Granulocyte colony-stimulating factor dowmegulates allogeneic immune responses

by posttranscriptional inhibition of tumor necrosis factor-alpha production. Blood

1995; 86: 2220-2227.

17. Härtung Τ, Docke WD, Gantner F, Krieger G, Sauer A, Stevens Ρ, Volk HD,

Wendel Α. Effect of granulocyte colony-stimulating factor treatment on ex vivo

blood cytokine response in human volunteers. Blood 1995; 85: 2482-2489.

18. Nakano Y, Onozuka К, Terada Y, Shinomiya H, Nakano M. Protective effect of

recombinant tumor necrosis factor-α in murine salmonellosis. J Immunol 1990;

144: 1935-1941.



1995; 63: 3674-3682.



450-454.

21. Sisson SD, Dinarello CA. Production of interleukin-1 alpha, interleukin-1 beta and

tumor necrosis factor by human mononuclear cells stimulated with granulocyte-

macrophage colony-stimulating factor. Blood 1988; 72: 1368-1374.

22. Gorgen I, Härtung Τ, Leist M, Niehorster M, Tiegs G, Uhlig S, Weitzel F, Wendel

A. Granulocyte colony-stimulating factor treatment protects rodents against lipo-

polysaccharide-induced toxicity via suppression of systemic tumor necrosis factor.

J Immunol 1992; 149: 918-924.

23. Pajkrt D, Manten A, Van der Poll T, Van Buul TMM, Jansen J, Ten Cate WJ, Van

Deventer SJ. Modulation of cytokine release and neutrophil function by granulo

cyte colony-stimulating factor during endotoxemia in humans. Blood 1997; 90:

1415-1424.


Dolmans WMV, Demacker PMN, Sauerwein R, Gallati Η, Van der Meer JWM.


Dis 1994; 169: 1306-1311.


during septicemic salmonellosis in calves. Infect Immun 1990; 58:439-42.

26. Jotwani R, Tanaka Y, Watanabe K, Tanaka K, Kato N, Ueno К. Cytokine stimula-



42: 348-52.

27. Keuter M, Netea MG, Kullberg BJ, Van Dissel JT, Demacker PNM, Stalenhoef

AFH, Van der Meer JWM. Hyperlipoproteinemia protects low-density-lipoprotein

receptor deficient mice against lethal Salmonella typhimurium infection. 1998;

submitted.

28. Serushago BA, Yoshikai Y, Handa T, Mitsuyama M, Muramori K, Nomoto K.

Effect of recombinant human granulocyte colony-stimulating factor (rh G-CSF) on

murine resistance to Listeria monocytogenes. Immunology 1992; 75: 475-480.

29. Shinomiya N, Tsuru S, Katsura Y, Kayashima S, Nomoto K. Enhanced resistance

against Listeria monocytogenes achieved by pretreatment with granulocyte colony-

stimulating factor. Infect Immun 1991; 59: 4740-4743.

96 Chapter 7

Chapter 8

Hyperlipoproteinemia protects low-density-lipoprotein receptor deficient mice against lethal Salmonella typhimurium infection

Monique Keuter, Minai G. Netea, Bart-Jan Kullberg, Jaap T. van Dissel, Pierre N.M.

Demacker, Anton F.H. Stalenhoef and Jos W.M. van der Meer

Submitted

98 Chapter 8

Abstract

Lipoproteins improve the outcome of severe Gram-negative infections through

neutralizing lipopolysaccharides (LPS) and thus reducing the induction of proinflam

matory cytokines. During a lethal Salmonella typhimurium infection in C57B1/6J mice,

comparing low density lipoprotein receptor deficient (LDLR-/-) and LDLR+/+ animals,

survival in LDLR-/- mice was greater when compared with LDLR+/+ (100% vs 5%, ρ

< .001). The outgrowth of S. typhimurium in the liver, spleen and bone marrow of

LDLR-/- mice was 100-1000 times lower than in controls on day 3 and 7 after infec

tion, with lower circulating cytokine concentrations. Upon intravenous high dose Sal

monella infection of LDLR+/+ mice, the bacteria disappeared from the blood within 30

minutes with rapid invasion of liver and spleen. In contrast, in LDLR-/- mice, Salmo

nella remained in the circulation for up to 2h, with a significantly decreased (70-80%)

organ count. Distribution differences appear to be a reason for improved survival to

salmonellosis of mice with hyperlipoproteinemia.

S. typhimurium in LDLR-/- mice 99

Introduction

Salmonella infections, especially typhoid fever, pose important public health pro

blems in tropical countries. Despite preventive measures such as hygiene and vaccina

tion, and the availability of antibiotics, typhoid fever remains a significant cause of

morbidity and mortality. To develop more effective therapeutic strategies, e.g. immuno

therapy, more insight in the host response to Salmonella is necessary.

The role of endotoxin in typhoid fever has been debated [1]. Although salmonellae

contain LPS, the clinical picture and pattern of circulating cytokines of typhoid fever in

humans differs from that in other Gram-negative sepsis [2]. Also in animal models of

typhoid fever, TNFcc was undetectable in the circulation until several days after 5.

typhimurium infection, whereas TNFa did rise 1 hour after LPS administration or extra

cellular Gram-negative infection in these animals [3,4]. salmonellae behave as faculta

tive intracellular parasites. It has been shown that in models of intracellular infection,

the proinflammatory cytokines play a beneficial role, resulting in better survival [5] and

inhibition of TNFa during Salmonella infection has worsened outcome [6, 7].

Circulating lipoproteins are able to bind and neutralize bacterial lipopolysaccha-

rides (LPS), thereby reducing its toxic properties by decreasing the induction of proin

flammatory cytokines such as IL-1 and TNFa [8,9]. In low density lipoprotein receptor

deficient (LDLR-/-) mice, the loss of the receptor results in а 7 to 9 times higher LDL-

cholesterol level than in their wild-type LDLR+/+ littermates. Our group has shown

that LDLR-/- mice survive longer and have lower proinflammatory cytokine concen

trations compared to control mice after bacterial endotoxin injection and after infection

with Klebsiella pneumoniae [10].

We realized that, although hyperlipoproteinemic LDLR-/- mice are able to with

stand extracellular Gram-negative infections like with Klebsiella pneumoniae, the sus

ceptibility to Salmonella infection would be more complex to predict in view of the

protective role of TNFa and the limited role of Salmonella endotoxin in these infec

tions. In the present study, we have investigated the susceptibility of LDLR knock out

mice to S. typhimurium.

Methods

Animals

Homozygous C57B1/6J mice lacking low density lipoprotein receptors (LDLR-/-)

and their wild-type littermates (LDLR+/+) were obtained from Jackson Laboratory (Bar

100 Chapter 8

Harbour, ME) as mating pairs and bred in our local facility. For the experiments, 6-8

weeks old mice, weighing 20-25 grams were used. The animals were fed standard labo

ratory chow (Hope Farms, Woerden, The Netherlands) and housed under specific

pathogen free conditions. The experiments were approved by the ethical committee for

animal experiments at the Catholic University Nijmegen.


A serum-resistant strain of 5. typhimurium (phage type 510) was grown to statio

nary phase culture by overnight incubation at 37°C in nutrient broth (BHI Oxoid). The

LD50 after either intraperitoneal (i.p.) or intravenous (i.v.) injection into C57B1/6J

(LDLR+/+) mice was < 102 bacteria. Mice were injected i.v. with 1 χ IO2 cfu of 5.

typhimurium. Survival was assessed daily for 21 days in groups of at least 20 animals.

On days 1, 3 and 7, separate groups of mice were anesthetized with ether to collect

blood from the intraorbital plexus for the measurement of circulating cytokine concen

trations. On day 1, 3 and 7 after the injection of S. typhimurium, mice were killed by

cervical dislocation and outgrowth of the microorganisms from the liver, spleen and

bone marrow was quantified. For this purpose, the liver and spleen were removed asep-

tically, weighed, and homogenized in sterile saline in a tissue grinder. Bone marrow

was flushed from the femur aseptically with 1 ml of sterile saline. The number of viable

salmonellae was determined by plating several dilutions on Brilliant Green agar (BGA)

plates and cfu were counted after overnight incubation at 37°C. The results were ex

pressed as the log cfu per gram of tissue. Previous experiments had shown that i.p. and

i.v. injection of the microorganisms led to an identical outcome.

Distribution of Salmonella cfu

In a separate experiment, Salmonella cfu (105 /mouse) were injected i.v. Distri

bution was determined in blood, liver and spleen after 30, 60, 120 and 360 minutes by

plating serial dilutions of blood and homogenized tissue samples on BGA. After over

night incubation at 37°C the cfu were counted. Groups of 5 mice were used for each

time point.

In vitro cytokine production

Resident peritoneal macrophages of LDLR-/- and LDLR+/+ mice were harvested

by rinsing the peritoneal cavity aseptically with 2 ml cold PBS containing 0.38% (w/v)

sodium citrate. After centrifugation, cells were resuspended in RPMI 1640 (Dutch


modification; Flow Laboratories, Irvine, UK), containing 1 гпМ pyruvate, 2 mM L-

glutamine and 100 μg gentamicin per ml, and incubated for 1 h at 37°C (1 Orwell) in

96-wells microtiter plates (Costar Corporation, Cambridge, MA). The non-adherent

cells were discarded and the remaining cell population consisted of more than 90%

macrophages, as assessed by light microscopy.

Heat-killed (30 min, 100°C) S. typhimurium were washed three times in RPMI.

Fifty uL of suspension containing 106 microorganisms in RPMI were added to perito

neal macrophages (105/ well) to a final volume of 100 \iL and incubated at 37°C in 5%

C02. After 24h, the supematants were collected and stored at -70°C until assayed. To

the remaining macrophages, 200 uJ_ of RPMI was added and the cells were disrupted

by three freeze-thaw cycles to determine the cell-associated cytokine contents. The

samples were stored at -70°C until assayed.

In order to assess the effect of the various lipoprotein subfractions on the produc

tion of cytokines, macrophages of LDLR+/+ mice were incubated with LPS-free lipo

protein subfractions of the plasma from LDLR-/- mice: LDL, VLDL and HDL respec

tively. Isolation was performed as described before [11]. Final concentration of isolated

cholesterol was 0.15 mmol/L. Heat-killed Salmonella were used to stimulate these

macrophages in the same procedure as described above.


TNFa, IL-la and IL-Iß concentrations were determined using specific radioim

munoassays (RIA) developed in our laboratory, as previously described [12].

Cytokine mRNA expression

Total RNA from spleen cells 24 hours after infection was isolated by the method of

Chomczynski and Sacchi [13]. For each sample 0.5 μg of total RNA was reverse trans

cribed and PCR was performed as described [14]. The following primers were used for

the PCR reactions (from a 0.5 μg sample RNA): GAPDH, sense, 5'-AACTCCCTCAA

GATTGTCAGCA-3', and antisense, 5'-TCCACCACCCTGTTGCTGTA-3'; TNFa,

sense, 5'-TCTCATCAGTTCTATGGCCC-3', and antisense, 5-GGGAGTAGACAAG

GTACAAC- 3'; IL-la, sense, 5'-CAGTTCTGCCATTGACCATC-3', and antisense, 5'-

TCTCACTGAAACTCAGCCGT-3·, IL-lß, sense, 5-TTGACGGACCCCAAAAGAT

G-3', and antisense, 5'-AGAAGGTGCTCATGTCCTCA-3' (Eurogentec, Seraing, Bel

gium). After checking the reactions to be in the log phase, thirty PCR cycles were per

formed with sets at 92°C for 30 sec, 55°C for 30 sec, and 72°C for 90 sec, using a

Mastercycler 5330 (Eppendorf, Hamburg, Germany). PCR products were run on 2%

102 Chapter 8

agars gels stained with ethidium bromide. The gels were scanned on a densitometer

(GS-670, Biorad, Veenendaal, The Netherlands) and analyzed using Molecular Ana

lyst software (Biorad). The relative amount of TNFa, IL-la and IL—1 β mRNA in a

sample was expressed as a ratio versus the amount of mRNA for the housekeeping gene

GAPDH.

Growth of Salmonella in vitro

To investigate the effect of lipoproteins on the in vitro growth, 0.5 χ IO3 cfu S.

typhimurium in 0.5 mL BHI were incubated with 0.5 mL of plasma obtained from

LDLR+/+ mice (cholesterol concentration 0.2 mM) or plasma obtained from LDLR-/-

mice (cholesterol concentration 1.5 mM). After 2, 7, 12 and 24 hours, aliquots of 0.1

mL were removed, serial dilutions were plated on BGA agar, and cfu were counted

after overnight incubation at 37°C.

Intracellular killing of S. typhimurium by peritoneal macrophages

Intracellular killing of S. typhimurium by macrophages in vitro [described in 15]

was determined as a decrease in the number of viable intracellular bacteria during incu

bation of macrophages containing ingested bacteria after phagocytosis in vivo.

For the in vivo phagocytosis, a suspension (1 mL) containing 2xl08 Salmonella cfu

in broth suppleted with 10% newborn calf serum was injected i.p. into anesthetized

mice (LDLR-/- and LDLR+/+). After 15 minutes, the animals were killed by cervical

dislocation; the peritoneal macrophages were collected after injection of 2 ml cold PBS

containing 50 U/ml heparin. After differential centrifugation and three washes to re

move extracellular bacteria, macrophages containing bacteria were reincubated with

RPMI containing 10% newborn calf serum and 10% autologous (LDLR-/- or LDLR

+/+) serum at 37°C and 5% CO2 and rotated (4 rpm). After 30, 60, and 90 minutes, 50

|lL aliquots of the suspension were administered to 450 uL distilled water with 0.01%

bovine serum in order to lyse the macrophages. The number of viable bacteria was de

termined by plating dilutions of this solution on BGA plates and counting cfu after

overnight incubation at 37°C.


Survival curves in control and LDLR-/- mice were compared by the Kaplan-Meyer

log-rank test. Differences in concentrations of cytokines and in organ counts of the

microorganisms were analyzed by the Mann-Whitney U test. Differences were consi

dered significant at ρ < .05. All the experiments were at least performed in duplicate.

S. typhimuríum in LDLR7- mice 103

100

(0 > ¡> (Λ

80

60

40

20

-η η η

LDLR-/-

LDLR+/+

12 16 20 days

Figure 1. Survival of LDLR-/- and LDLR+/+ C57B1/6J mice (at least 20 animals per group) after i.v. in

jection of 102 S. typhimuríum.

Results

Survival of mice and outgrowth oí Salmonella in the organs

After i.v. infection with 102 cfu of S. typhimuríum, only 5% of the LDLR-/- mice

died (p < .001), whereas all LDLR+/+ died within 12 days of infection (fig. 1). Out

growth of Salmonella 3 days after infection with 102 microorganisms was at least 10-

fold lower in liver, spleen and bone marrow of LDLR-/- mice compared with control

LDLR+/+ mice (fig. 2). These differences were significant at ρ <.001 (liver) and ρ < .01

(spleen and bone marrow). On day 7 the differences were even greater (p < .001; fig. 2).

The outgrowth after i.p. infection did not significantly differ from that after i.v. infec

tion.

Distribution of Salmonella

Since the differences in outgrowth could be due to a difference in distribution, we

determined disappearance of microorganisms from the circulation after i.v. injection of

105 S. typhimuríum cfu. Blood from LDLR+/+ mice contained significantly less Salmo-

nella cfu than that of LDLR-/- mice 30 minutes after injection (fig. 3, ρ < .01). The

numbers of cfu in the liver and spleen were 70-80% lower in LDLR-/- mice than that in

LDLR+/+ mice at that point in time, and remained lower throughout the experiment,

being significant at 30, 60, 120 and 360 minutes for the liver (p < .05) and at 30 and 60

minutes for the spleen (p <.05 ; fig. 3).

104 Chapter 8

υ α> 3

Bone marrow

.E э υ

Figure 2. Outgrowth of S. typhimurium in the liver, spleen and bone marrow of LDLR-/- and LDLR+/+

C57B1/6J mice after i.v. injection of 102 cfu. Each point represents the mean ± SD for at least 10 animals.

Significant differences between LDLR-/- and LDLR+/+ mice are indicated (*, Ρ < .01; **, Ρ < .001;

Mann-Whitney U test).

Circulating cytokine measurements

Circulating concentrations of IL-la, IL-Iß and TNFa were determined on day 1, 3

and 7 of infection with 102 Salmonella cfu. On day 1, cytokine concentrations in all

samples were under the detection limits. No detectable concentrations of IL-Iß (< 20

pg/ml) were found at any time point during the infection. On day 3, IL-la was under

the detection limit in both groups. TNFa was under the detection limit (< 40 pg/ml) in

LDLR-/- mice, while TNFa concentration was low but detectable (45 ± 10 pg/ml) in

LDLR+/+ mice (n.s.).

Figure 3. Distribution of S. typhimurium to the blood (cfu/ml), the liver (cfu/organ), and the spleen (cfu/

organ) at various time points after i.v. injection of 10s Salmonella cfu. Each point represents the mean ±

SD for at least 5 animals. Significant differences between LDLR-/- and LDLR+/+ mice are indicated (*,

Ρ < .05; **, Ρ < .01; Mann-Whitney U test).

S. typhimunum in LDLR-/- mice 105

On day 7, circulating concentrations of IL-Ια and TNFa were significantly higher

in LDLR+/+ than in LDLR-/- mice: 95 ± 63 pg/ml vs 30 ± 10 pg/ml for IL-la (p < .02)

and 1140 ± 290 pg/ml vs 43 ± 6 pg/ml for TNFa (p < .02) (Figure 4).

Cytokines in the early stages of the infection with 105 Salmonella cfu were asses

sed as well. After 60 minutes, IL-la concentrations in the plasma were higher in the

LDLR-/- mice compared to those in the LDLR+/+ mice: 910 ± 370 pg/ml vs 310 ± 140

pg/ml (p < .05). This difference had disappeared after 2 hours. There were no differen

ces in the TNFa concentrations between the 2 strains (Figure 4).

In vitro cytokine production

To investigate whether a difference in cytokine production capacity in response to

Salmonella could explain the difference in outcome, resident (uninfected) peritoneal

macrophages of LDLR-/- and LDLR+/+ mice were stimulated with heat-killed Salmo

nella in vitro. TNFa concentrations in supernatants of macrophages were similar in

LDLR-/ mice and LDLR+/+ mice: In unstimulated cultures 194 ± 91 pg/ml and 186 ±

55 pg/ml and in macrophage cultures incubated with heat-killed Salmonella 550 ± 166

pg/ml and 520 ±135 pg/ml for LDLR-/- and LDLR+/+ mice, respectively.

Figure 4. Circulating concentrations of TNFa (Δ,Α) and IL-la (•,•) on day 1, 3, and 7 of infection

with of 1 χ 102 CFU S. typhimurium in LDLR-/- (open symbols) and LDLR+/+ (closed symbols) mice.

Significant differences are indicated (*, Ρ < 05; Mann-Whitney U test)

Inset: Circulating concentrations of TNFa (Д,А) and IL-la (C,B) at the early stage (60 and 120

minutes) of infection with 1 χ 105 CFU of S typhimunum in LDLR-/- (open symbols) and LDLR+/+

(closed symbols) mice. Significant differences are indicated (*, Ρ < .05; Mann-Whitney U test).

106 Chapter 8

100

fl ш 80

• ^ υ to

Л 60 со '> (D

S) 40 СО

• ^ с φ и Ё 20 о.

0 0 15 30 45 60 75 90

minutes

Figure S. Intracellular killing of S. typhimunum by resident peritoneal macrophages in vitro, after l.p

injection of 2xl0 8 Salmonella cfu and in vivo phagocytosis during IS min. Data are expressed as percen

tage of the initial number of microorgansims phagocitized by macrophages. No significant differences

between LDLR-/- and LDLR+/+ were found.

Since lipoproteins are able to reduce the LPS-induced cytokine production, we

added various lipoprotein subfractions, i.e. LDL, VLDL and HDL isolated from LDLR

-/- mouse plasma, to normal (LDLR+/+) macrophages, in order to investigate the effect

of these lipoproteins on cytokine production after stimulation. Stimulated with heat-

killed Salmonella in the presence of either of these subfractions, macrophages from

LDLR+/+ mice produced 520 ±135 pg/ml TNFa in the absence of lipoproteins, versus

406 ±191 pg/ml in the presence of LDL (n.s.), 345 ± 48 pg/ml in the presence of

VLDL (p < .05) and 322 ± 48 pg/ml with HDL (p < .02), suggesting that local TNFa

production is reduced in the presence of elevated VLDL or HDL concentrations.

Cytokine mRNA measurements in spleen

Another possible mechanism for the difference between the 2 strains could have

been higher local production of cytokines in LDLR-/- mice in the early stages of infec

tion. In spleens of both strains, mRNA for IL-lß and TNFa was measured 24 hours

after i.v. injection of 102 Salmonella cfu and was not significantly different in the two

mouse strains (expressed as ratio cytokine versus GAPDH mRNA). TNFa was 0.677 ±

0.040 in LDLR-/- mice vs 1.142 ± 0.465 in LDLR+/+ mice ; IL-lß was 0.584 ± 128 in

LDLR-/- mice vs 0.511 ± 0.097 in LDLR+/+ mice.


Growth of Salmonella in vitro We investigated whether there was an effect of lipoproteins on in vitro growth of S.

typhimurium. The growth curves of Salmonella was similar after 2, 7, 12 and 24 hours

in broth with plasma obtained from LDLR-/- mice and broth with plasma obtained from

LDLR+/+ mice (data not shown).

Intracellular killing of Salmonella in vitro A difference in killing capacity of the respective macrophages could account for a

difference in outgrowth of Salmonella cfu in the organs. After in vivo phagocytosis, the

numbers of cfu phagocytized by. macrophages of LDLR-/- and LDLR+/+ mice were

similar (data not shown), which shows that uptake of salmonellae was not enhanced by

peritoneal macrophages from one of the strains examined. The results of the intracel

lular killing experiments showed that resident peritoneal cells of LDLR-/- mice and

LDLR+/+ mice do not differ in their ability to kill S. typhimurium. The killing rate did

not differ when lipoprotein-rich serum of LDLR-/- mice was added to LDLR+/+ control

macrophages, and likewise, serum from LDLR+/+ mice did not affect the killing of

Salmonella by LDLR-/- macrophages (Figure 5).

Discussion

In the present study we have shown that hyperlipoproteinemic LDLR-/- mice are

protected against S. typhimurium infection. In LDLR-/- mice, survival was significantly

greater than that of control mice, outgrowth of microorganisms in the organs was redu

ced, and circulating proinflammatory cytokine concentrations at late stages of infection

were lower.

In previous studies, it has been shown that lipoproteins are able to bind and neutra

lize LPS, and the subsequently reduced induction of circulating proinflammatory cyto

kines is the main mechanism for the beneficial effect of lipoproteins in Gram-negative

infection [10].

The situation in the intracellular S. typhimurium infection appears to be more com

plicated. First of all, in contrast to an extracellular Gram-negative infection such as that

caused by K. pneumoniae, endotoxin release as judged by the cytokine response is not

prominent. More important is the finding that the outgrowth of Salmonella is reduced in

LDLR-/- mice and the lower bacterial load may be the reason for the lesser degree of

cytokinemia. In lethal K. pneumoniae infection, there was no difference in outgrowth of

microorganisms in the organs between the two strains of mice [10]. In view of the bene-

108 Chapter 8

ficial effect of TNFa in host defense against intracellular infections [5-7, 16], we consi

dered that the reduced outgrowth of bacteria in the tissues could be due to increased

cytokine production at the tissue level not reflected in the circulation. Therefore we

investigated whether the decreased outgrowth of microorganisms in the organs could be

a result of higher local TNFa production. In peritoneal macrophages of LDLR-/- mice,

a slight increase in TNFa and especially IL-Iß production after stimulation with LPS

has been observed previously [10]. In the present study, when peritoneal macrophages

of LDLR-/- mice were stimulated in vitro with heat-killed Salmonella, no increase of

cytokine production by macrophages was detected. These in vitro experiments are

further sustained by our observations that no increased local production of messenger

RNA in the spleens of LDLR-/- mice could be found during infection with S. typhi-

murium. This implies that the decreased outgrowth of Salmonella in the liver and spleen

is unlikely to be explained by a local increase of proinflammatory cytokines. Another

explanation could be a difference in microbicidal capacity of phagocytic cells.

High concentrations of lipoproteins may induce changes at the cell-membrane level

[17, 18], thereby changing the susceptibility to Salmonella infection. Although diver

gent rates of S. typhimurium outgrowth in the organs of infected mice may suggest that

differences either in killing or in virulence of Salmonella might play a role, we found

that both phagocytosis and intracellular killing of S. typhimurium by macrophages from

LDLR-/- mice were similar to that of control mice.

A direct effect of lipoproteins on the bacteria may also have influenced the decrea

sed outgrowth of Salmonella in LDLR-/- mice. In this respect, it is of interest that HDL

has been found to be cidal against Trypanosoma cruzi [14]. However, the growth of

Salmonella was similar in LDL-rich medium and in control medium, excluding direct

involvement of lipoproteins as an inhibiting factor for Salmonella growth in LDLR-/-

mice.

An important finding in our studies is the difference in distribution of Salmonella

during the first hours after infection. It is conceivable that these short term distribution

differences determine the short and long term outgrowth and subsequent better survival

of LDLR-/- mice to a considerable extent. To explain this phenomenon there are two

possibilities: either the salmonellae are kept in the circulation or the organs have im

paired capabilities for uptake of the microorganisms. It could be that lipoproteins

directly interact with Salmonella uptake, for example by coating the bacterial surface

structures with lipids. Alternatively, the absence of the LDL receptor could be the

reason for an altered uptake of bacteria. In this respect it is interesting that malaria

parasites compete for the same receptor with certain lipoproteins [19], and that Strepto-

S. typhimurìum in LDLR-Λ mice 109

coccus pneumoniae has been shown to bind to a lipid receptor [20]. However, the ex

periments in which the in vivo phagocytosis of salmonellae by peritoneal macrophages

was investigated argue against such a mechanism, since equal uptake was found by

LDLR-A and LDLR+/+ macrophages. Still, further studies are necessary before this

hypothesis can be rejected.

Acknowledgments The authors thank Ineke Verschueren for performing the cytokine determinations

and Margo van de Brink, Monique Bakker, Yvette Brom, Theo van der Ing, and Maria

van den Barselaar for assistance with the animal experiments.

References


Clin Invest 1969; 48: 613-629.




Dis 1994; 169: 1306-1311.


during septicemic salmonellosis in calves. Infect Immun 1990; 58: 439-42.

4. Jotwani R, Tanaka Y, Watanabe K, Tanaka K, Kato N, Ueno К. Cytokine stimula


42: 348-52.



144: 1935-1941.



1995; 63: 3674-3682.



450-454.

8. Harris HW, Grunfeld C, Feingold KR, et al. Chylomicrons alter the fate of endo

toxin, decreasing tumor necrosis factor release and preventing death. J Clin Invest

1993; 91: 1028-1034.

по Chapter S

9. Flegel WA, Wolpl A, Mannel DN, Northoff H. Inhibition of endotoxin-induced

activation of human monocytes by human lipoproteins. Infect Immun 1989; 57:

2237-2245.

10. Netea MG, Demacker PNM, Kullberg BJ, Boerman OC, Verschueren I, Stalenhoef

AFH, Van der Meer JWM. Low-density-lipoprotein receptor deficient mice are

protected against lethal endotoxinemia and severe Gram-negative infections. J Clin

Invest 1996; 97: 1366-1372.

11. Demacker PNM, Vos-Jansen HE, Jansen AP, Van 't Laar A. Evaluation of the dual

precipitation method by comparison with the ultracentrifugation methods for the

measurement of lipoproteins in serum. Clin Chem 1977; 23: 1238-1244.



downregulates ex vivo TNF-ct and IL-Iß production. J Appi Physiol 1995; 79:

1497-1503.

13. Chomczynski P, Sacchi N. Single step method of RNA isolation by guanidinum

thiocyanate phenol chloroform extraction. Anal Biochem 1987; 162: 156-159.

14. Smith AB, Esko JD, Hajduk SL. Killing of trypanosomes by the human haptoglo-

bin-related protein. Science 1995; 268: 284-286.

15. Van Dissel JT, Stikkelbroeck JJM, Sluiter W, Leijh PCJ, Van Fürth R. Differences

in initial rate of intracellular killing of Salmonella typhimurium by granulocytes of

salmonella susceptible C57B1/10 mice and salmonella resistant CBA mice. J

Immunol 1986; 136: 1074-1080.

16. Tite JP, Dougan G, Chatfield SN. The involvement of tumor necrosis factor in im

munity to Salmonella infection. J Immunol 1991; 147: 3161-3164.

17. Esfahani M, Bigler RD, Alfieri JL, Lund-Katz S, Baum JD, Scerbo L. Cholesterol

regulates the cell surface expression of glycophospholipid-anchored CD 14 antigen

on human monocytes. Biochim Biophys Acta 1993; 1149: 217-223.

18. Rothberg KG, Ying YS, Kamen BA, Anderson RGW. Cholesterol controls the

clustering of the glycophospholipid-anchored membrane receptor for 5-methyl-

tetrahydrofolate. J Cell Biol 1990; 111: 2931-2938.

19. Sinnis P, Willnow TE, Briones MRS, Herz J, Nussenzweig V. Remnant lipopro

teins inhibit malaria sporozoite invasion of hepatocytes. J Exp Med 1996; 184:

945-954.

20. Cundell DR, Gerard NP, Gerard C, Idanpaan-Heikkila I, Tuomanen EI. Strepto

coccus pneumoniae anchor to activated human cells by the receptor for platelet-

activating factor. Nature 1995; 377: 435^38.

Chapter 9

General Discussion

112 Chapter 9

General Discussion ИЗ

Mediators of fever in patients with typhoid fever In the classical ideas about the induction of fever, pyrogenic cytokines circulate

and stimulate the synthesis of PGE2 in the lamina terminalis of the central nervous

system (CNS), which results in behavioral, endocrine and autocrine responses [13]. In

the study described in Chapter 3, we were not able to detect appreciable concentrations

of the pyrogenic cytokines IL-Iß, TNFa and LTa in patients with typhoid fever. The

concentrations of IL-6, generally considered a relatively weak pyrogen [144], were low

as compared to other febrile conditions [145, 146]. The suggestion that synthesis of

prostaglandins such as PGE2 in the CNS might be induced by other substances or in

other ways than by circulating endogenous pyrogenic cytokines, should be considered

[15].

Many years ago, Cranston's group showed that intraventricular injection of inhi

bitors of sPLA2 suppressed fever in rabbits [16]. Since sPLA2 release is induced by IL-

1 and TNFa and the enzyme could mediate the production of arachidonic acid (AA)

[20], sPLA2 might be this intermediate.

In Chapter 4, we found elevated concentrations of extracellular group II 14-kDa

sPLA2 during the acute phase of typhoid fever in 12 patients. In patients with compli

cated disease the bioactive sPLA2 concentrations were 10-fold higher than in those

with uncomplicated disease. sPLA2 concentrations fell significantly towards recovery.

We considered that this bioactive sPLA2 might be responsible for the symptoms of

typhoid fever, as sPLA2 is locally induced by proinflammatory cytokines at the site of

the infection and reaches the circulation. In this manner sPLA2 could mediate the sys

temic signs of the disease [19, 147].

However, this concept of sPLA2 as a circulating mediator of fever has become

more doubtful lately. The role of sPLA2 in the generation of AA and other eicosanoids

which takes place intracellularly, is not clear. It has been questioned why cells should

secrete an enzyme involved in intracellular functions, when there is already an intracel

lular enzyme for AA formation. This enzyme is the cytosolic 85 kDa PLA2. Further

more, we were not able to induce fever in rabbits with purified rat sPLA2 (unpublished

results). Recombinant human sPLA2 did not express inflammatory activity in normal

rat tissues either [148].

Recent publications suggest a role for sPLA2 together with C-reactive protein

(CRP) in the removal of injured cells. For instance in myocardial infarction, it is hypo

thesized that CRP binds to muscle cells via lyso-phosphatidylcholine, generated in the

outer leaflet of the cell membrane by sPLA2. Inner leaflet phosphatidylserine and etha-

nolamine move to the outer leaflet in "flip-flopped" membranes, making phosphatidyl-

114 Chapter 9

choline more susceptible to phospholipolysis by sPLA2. This damage or flip-flop

occurs, when the cells are ischemic. Complement is bound and neutrophils activated

when CRP binds to these lysophospholipids [149]. In this hypothesis sPLA2 has a func

tion in waste disposal probably to allow quick repair. sPLA2 can only be damaging in

situations, where membranes of cells are already injured. This might explain the ab

sence of biological effects when recombinant sPLA2 is injected into healthy animals

[148].

sPLA2 is relevant to the pathogenesis of various clinical conditions, including

rheumatoid arthritis (RA) [150]. As females tended to have higher group IIPLA2 acti

vity in plasma and cells, the authors of this study suggested that sPLA2 activity might

be a risk factor for RA in view of the increased prevalence of the disease in women

[151]. So, although the function of sPLA2 in myocardial infarction [18], RA and

septicemia could be the removal of injured cells, which might be increased by an excess

of sPLA2, the pathobiology of excess sPLA2 production is not completely understood.

In the concept of fever, it is important to note that although concentrations of circu

lating pyrogenic cytokines were low in the acute febrile phase of typhoid fever, mRNA

for TNFa and IL-Iß was found to be present in circulating whole blood cells (Chapter

J). Thus these circulating cells could themselves produce or induce pyrogenic cytokines

in the endothelial cells of the OVLT [15]. Moreover, the recent finding of membrane-

bound TNFa, might explain the absence of free circulating pyrogens in the circulation.

One could hypothesize that this membrane bound TNFa could be released at the site

(OVLT) [152].

Cytokine profile in patients with typhoid fever In Chapter 3, we report low concentrations of circulating proinflammatory cytoki

nes, but high concentrations of the antiinflammatory IL-IRA and sTNFRI and sTNFRII

during the acute phase of typhoid fever. At that time point, the capacity of LPS-induced

production of proinflammatory cytokines in whole blood of patients with typhoid fever

was down-regulated. This depressed cytokine production was not a consequence of a

lower number of white blood cells during the acute phase of the infection.

Down-regulation of cytokine production In Chapter 3, we describe down-regulation of the production of the proinflam

matory cytokines IL-1, IL-6 and TNFa, but not of the anti-inflammatory products

sTNF-receptors and IL-IRA. This is in accordance with later studies on endotoxemia in

volunteers, in which production of proinflammatory cytokines (but also interestingly of

General Discussion 115

the antiinflammatory cytokine IL-10) was found to be suppressed, whereas production

of receptors of TNFa and IL-IRA was increased [153].

It has been described that exposure to endotoxin leads to tolerance [154]. Tolerance

is a general biological phenomenon observed for many ligand-receptor interactions. A

low production of cytokines has not only been found in serious conditions like severe

post-operative infection [155] and sepsis [156-158], but also in patients with familial

Mediterranean fever, during attacks which are not endotoxin-mediated [159,160]. Also

after major surgery or blood loss and strenuous exercise [161, 162], down-regulation of

cytokines was observed and it has been suggested that any major stress can give an im

paired inflammatory response.

This down-regulation of cytokine production occurs only a short time after the in

jury. At 2 hours after endotoxin infusion in volunteers, cytokine production upon LPS

stimulation of whole blood was down-regulated [153]. Also in isolated peripheral blood

mononuclear cells (PBMCs), down-regulation of proinflammatory cytokine production

was apparent few hours after the first injection of endotoxin in volunteers [163].

Decreased cytokine production capacity as a consequence of exposure in vivo to

substances such as for instance endotoxin, has often been interpreted as exhaustion of

cytokine-producing cells [155-160]. Our finding that IL-1 receptor antagonist and

sTNFRs are being produced in high concentrations in whole blood argues against

exhaustion of cells, although these antiinflammatory cytokines have different kinetics

and might be produced by different cell types. PBMCs of patients with typhoid fever

produced fewer proinflammatory cytokines, but were still able to produce EL-IRA in

the acute phase of typhoid fever (unpublished results), indicating that the IL-IRA found

in the whole blood method was at least partly produced by monocytes and not only by

granulocytes.

To investigate the mechanism of down-regulated cytokine production in typhoid

fever further, messenger RNA for IL-Iß and TNFa in whole blood was measured and

this study was described in Chapter 5. The finding that the amounts of mRNA for IL-Iß

and TNFa in the acute phase of typhoid fever were similar to those in the recovery

phase, also denies an exhausted production and is in favor of post-transcriptional inhibi

tion of cytokine production.

LPS tolerance may in part be reproduced by post-endotoxin infusion plasma, sug

gesting that soluble factors produced in response to the LPS challenge are involved

[153]. These soluble factors might include glucocorticosteroids, norepinephrin, lipo

proteins and antiinflammatory cytokines [112, 164, 165]. For instance antiinflammatory

IL-4, IL-10 and TGF-ß are able to down-regulate the production of IL-1, TNFa, IL-6,

116 Chapter 9

IL-8, and to induce the production of IL-IRA [120]. The mechanisms of endotoxin

tolerance are not clear so far. In acute Gram-negative sepsis, down-regulation of cyto

kine production was shown to be caused partly by norepinephrin, which is able to

attenuate cytokine production [165]. In addition, proinflammatory cytokines themselves

may be responsible for the down-regulation, as was shown in mice treated with a single

low dose of rIL-1 [109]. Apparently, after the initial phase of infection there is a post-

transcriptional switch-off of the production of proinflammatory cytokines and a switch-

on to an anti-inflammatory repertoire in mononuclear cells. Theoretically one could

hypothesize that within the mononuclear cells there are two subtypes: those that are sus

ceptible to Thl stimulation and those that are more Th2-sensitive.

Down-regulation and the concept of a switch have been assented by several inves

tigators and have also been shown in vitro in isolated peripheral blood mononuclear

cells. In such experiments, preincubation of PBMCs with a low dose of LPS rendered

these cells tolerant for subsequent stimulation with LPS, resulting in lower TNFa as

well as lower IL-10 responses, but not lower IL-RA production [166]. IL-10 mRNA up-

regulation has been reported in these LPS-tolerant monocytes, whereas TNFa mRNA

was found to be reduced [167]. The CD14 receptor was still expressed in these cells as

well as the transcription factor NF-κΒ, which suggests that LPS could reach its receptor

and cytokine production was not inhibited by a transcription factor [168]. In vitro, anti

inflammatory cytokines such as IL-10 and TGF-ß were able to induce the process of

down-regulation, and neutralizing these cytokines could prevent it [166]. These expe

riments show that IL-10 and TGF-ß are important in the switch to antiinflammatory

cytokine production. However, in vitro down-regulation of cultured cell populations,

which are repeatedly stimulated with LPS is not similar to ex vivo stimulation of whole

blood from patients with for instance endotoxemia.

It was also found that endotoxin tolerance could be antagonized by IFN-γ in vitro

[167]. Recent clinical studies have shown that this LPS tolerance or "immunoparalysis",

in which monocytes of septic patients were neither able to express HLA-DR nor to

produce TNFa, could be reversed in these patients by treatment with IFN-γ [169]. Re

covery of this immune function was reported to lead to survival in 8 of 9 patients.

In conclusion, in-vivo down-regulation of cytokine production with preserved

mRNA of proinflammatory cytokines, which we found during the acute phase of

typhoid fever, is a common phenomenon in many circumstances. Apparently, it is

found in endotoxin-mediated diseases [153, 164, 170, 171] as well as in the intracellular

Salmonella infection. Down-regulation is most likely a protection against overwhel-


ming cytokinemia, but can impair immune functions.

The benefit of down-regulation in patients with typhoid fever is difficult to under

stand in view of experimental animal studies with Salmonella infection, in which the

proinflammatory cytokines are required to survive the infection [82-84, 86, 87] (see

Chapter 2). Possibly the deeper downregulated proinflammatory cytokines in compli

cated cases could reflect a less effective host defense and immunotherapy might be

needed to improve host defense.

Cytokines during experimental S. typhimurium infection In Chapter 6, the cytokine profile during experimental S. typhimurium infection in

mice is described. Without infection, circulating cytokine concentrations but also LPS-

stimulated ex-vivo production in whole blood were undetectable. The latter inability to

produce cytokines is the reverse of the human situation, since human peripheral blood

cells can be stimulated to produce large amounts of proinflammatory cytokines without

infection. In contrast, uninfected bone marrow and peritoneal macrophages were able to

produce proinflammatory cytokines after LPS stimulation.

During S. typhimurium infection, the circulating TNFa concentration was increa

sed on day 7, as was the LPS-stimulated production of IL-la, IL-lß and TNFa in

whole blood. Production of IL-lß and TNFa by peritoneal and bone marrow cells was

also increased during infection. Whereas LPS-stimulated cytokine production during S.

typhimurium infection was generally increased, the IL-la production by infected peri

toneal cells was decreased. This is an important observation, since peritoneal macro

phages are easy to obtain and frequently used as representatives for the resident or acti

vated macrophage population of the mouse. This shows that the cytokine response of

peritoneal and other macrophages are divergent. The attention to peritoneal cells alone

might not yield information on cytokine production at other sites of infection.

The general upregulation of proinflammatory cytokine production in this animal

model is also in contrast with what we have found in humans with typhoid fever, des

cribed in Chapter 3. In these patients, down-regulation of LPS-stimulated production of

cytokines in whole blood was found. This phenomenon was even more profound in

complicated cases. These differences seem not to be caused by the time point of

measuring cytokine production: The patients with typhoid fever had been ill for a week

before admission, and down-regulated production was found at that time point. In the

experimental study, mice were followed for 7 days and upregulation of cytokine

production was predominantly found at day 7. It is obvious that one should be careful to

draw conclusions from animal experiments and extrapolate the results to the human

118 Chapter 9

situation.

The effect of IL-1 pretreatment In Chapter 6, the effect of pretreatment of mice with rhIL-la on S. typhimurium

infection has been described. Administration of rhIL-la led to a significantly decreased

outgrowth of microorganisms in the liver and spleen of mice. This effect of rhIL-la

pretreatment was seen as early as 24 hours after the onset of the infection. The micro

biological effect of a single dose of rhIL-la lasted throughout the study.

Pretreatment with rhIL-la had no effect on circulating cytokine concentrations.

The ex-vivo cell-associated IL-la production by peritoneal and bone marrow macro

phages however, was significantly increased 24 hours after pretreatment.

RhIL-la had no effect on outgrowth of microorganisms during infection with

Pseudomonas aeruginosa in neutropenic mice [172]. In those experiments, it has been

shown that administration of either rIL-1 or rTNFa increased survival of mice. This

effect was partially due to reduction of circulating concentrations of proinflammatory

cytokines and reduction of LPS-stimulated cytokine production by peritoneal macro

phages [109]. It is very unlikely that the mechanism of reducing TNFa production

applies to both Salmonella and Pseudomonas infections. It is known that lacking the

TNFa receptor-1 is deleterious for mice with intracellular Listeria infection whereas it

is life-saving for the same mice with endotoxemia [74, 75]. Besides, we did not find

downregulation of TNFa production after rhIL-la treatment in our experiments with

Salmonella-infected mice.

Possible mechanisms of the reduction in outgrowth of S. typhimurium in the organs

include neutrophil recruitment and activation. Intraperitoneally administered IL-1 in

creased the numbers of neutrophils [106]. IL-1 is able to induce granulocyte colony-

stimulating factor (G-CSF) in vitro, although concentrations of G-CSF have never been

measured in vivo after rIL-1 pretreatment. However, granulocyte numbers were not

found to be increased in our experiment and in those of others at the time point that the

effect on microbial outgrowth was seen [173]. We have shown in Chapter 7 that recom

binant murine G-CSF increased survival, but did not reduce outgrowth of S. typhim

urium in the organs of mice with a sublethal infection. This suggests that induction of

G-CSF is not likely to be the mechanism behind the protection by rIL-1 pretreatment.

There is some debate whether administration of rIL-1 leads to activation of macro

phages [100, 110], and if so, whether these activated macrophages are important in the

earliest stage of the infection, including the distribution phase, or as more efficient

killers later during the infection. Studies with macrophages activated through a prece-


ding S. typhìmurìum or BCG infection showed that this led to lower numbers of salmo-

nellae in the organs during the first hours of a subsequent infection [44,45]. However,

these previously activated macrophages, though able to inhibit T. gondii did not display

a significant effect on growth rate of microorganisms in the organs during this Salmo

nella infection [44, 45]. The outgrowth pattern in our experiment also suggests a distri

bution effect of rhIL-la pretreatment, as the rate of outgrowth in the organs of control

mice and rhIL-la pretreated mice is similar between day 1 and 7. This early effect may

be established by several mechanisms. First, antibodies (specific or non-specific) could

be induced by previous infections and these could influence the distribution of bacteria.

Second, the macrophages themselves may be involved, although the unaffected killing

rate argues against this hypothesis. Third, the activated macrophages may produce

cytokines that exert their effect on other cells, such as granulocytes or NK cells, to

enhance killing. Alternatively, these cytokines may influence invasion or intracellular

killing of microorganisms in hepatocytes or other cells [64,174].

The fact that IL-1 pretreatment protects against different types of infection is in

favor of a general phenomenon such as macrophage activation. The hypothesis that this

effect of rIL-1 is exerted through an effect on distribution is supported by the obser

vation that a similar rate of outgrowth in the organs of control mice and rIL-1 pretreated

mice is also found in Candida infection [173]. Also the beneficial effect of BCG-activa-

ted macrophages on Candida infection was seen in the first hours of infection [175].

The effect of G-CSF treatment In Chapter 7, we described that recombinant murine granulocyte colony-stimula

ting factor (rmG-CSF) improved survival of mice with a potentially lethal Salmonella

typhimurium infection. However, the outgrowth of S. typhimurium in the liver, spleen,

bone marrow, lungs, brain, kidneys and heart of rmG-CSF mice was not different from

that of control mice. Histological examination of liver and spleen on day 1 and 5

showed more localized inflammatory foci in the rmG-CSF treated animals. These

results show that pretreatment with rmG-CSF is beneficial in the facultative intracel

lular 5. typhimurium infection in mice.

The effect of rG-CSF on proinflammatory cytokines has been studied by several

investigators. In contrast to granulocyte-macrophage colony-stimulating factor (GM-

CSF), rG-CSF did not induce fever, suggesting that the induced TNFa production in

GM-CSF treated macrophages [176] was absent with rG-CSF. In human volunteers,

rG-CSF treatment increased plasma concentrations of IL-IRA, TNFa and sTNFRs

without any side effects such as fever [177]. It was found that ex vivo production of

120 Chapter 9

TNFcc in human blood was lower after rG-CSF treatment [142, 143], although Pajkrt et

al. found a biphasic response [178]. In human volunteers pretreated with rG-CSF 2

hours or 12 hours before endotoxin infusion, the peak of circulating TNFa was increa

sed [177, 178]. In rodents, pretreatment with rG-CSF 1 hour before endotoxin infusion

protected against LPS-induced death, but rG-CSF did not prevent TNFa-induced death,

suggesting an effect on TNFa production [179].

In our experiments in mice, basal circulating cytokine concentrations did not differ

between the rmG-CSF treated and control group. One day after treatment, LPS-stimu-

lated TNFa production of resident peritoneal macrophages was significantly higher in

rmG-CSF treated mice. Apparently, differences in species, time points and way of ad

ministration of G-CSF (in vivo or in vitro) play a role in these discrepancies.

It may be questioned, whether rG-CSF has a beneficial effect on intracellular infec

tions through enhanced (early) neutrophil recruitment, and whether the (late) antiin

flammatory properties of rG-CSF are deleterious, e.g., through inhibition of TNFa

mediated macrophage activation. The trend for an increased number of cfu in liver and

spleen in the group, that received continuous rmG-CSF treatment, could be an illus

tration of the latter suggestion and may be due to such an adverse late effect. Late treat

ment with rmG-CSF was also deleterious in experimental Candida infection in mice

(Kullberg, unpublished results).

Pretreatment alone is beneficial in several infections such as with Candida [136].

In L. monocytogenes infection in mice, pretreatment with high dose human rG-CSF

during 5 days, reduced mortality as well as outgrowth of microorganisms [140,141].

In conclusion, rmG-CSF pretreatment protects mice against a lethal S. typhimurium

infection in mice, although no effect was found on the outgrowth of microorganisms in

the organs. Further studies are needed to elucidate the mechanism, before rG-CSF can

be used as immunotherapy in Salmonella infection.

In recent studies, rG-CSF increased the in vitro antimicrobial activity of neutro

phils with ciprofloxacin [180]. Neutrophils treated with rG-CSF for 1 hour were incuba

ted with ciprofloxacin. rG-CSF increased the intracellular concentration of cipro

floxacin in these neutrophils. Inhibition of microorganisms was greatest when neutro

phils, rG-CSF and quinolones were combined. As ciprofloxacin is an important antimi

crobial against Salmonella infection, the combination of these substances is worth

investigating in Salmonella infection.

Lipids, endotoxin and Salmonella infection In patients with Gram-negative bacteremia, lipid infusions are considered for


treatment, since lipoproteins can bind and neutralize LPS and thereby reduce proinflam

matory cytokine production. Therefore we thought it interesting to investigate the effect

of lipoproteins on infection with the Gram-negative intracellular Salmonella typhimu-

rium.

In Chapter 8, the presence of high concentrations of lipoproteins were shown to

improve the outcome of S. typhimurium infection. We show, that neutralization of

endotoxin, which in Klebsiella infection is the beneficial effect of lipids, is not the

mechanism behind the better survival of S. typhimurium infection.

During a lethal S. typhimurium infection in C57B1/6J mice, survival in LDLR-/-

mice was greater when compared with LDLR+/+. The outgrowth of S. typhimurium in

the liver, spleen and bone marrow of LDLR-/- mice was 100-1000 times lower than in

controls on day 3 and 7 after infection. Remarkably, circulating cytokine concentrations

were lower in LDLR-/- mice.

A most important finding in this study is the difference in distribution of micro

organisms during the first hours of infection. After intravenous injection of Salmonella

into LDLR+/+ mice, the bacteria disappeared from the blood within 30 minutes with

rapid invasion of liver and spleen. In contrast, in LDLR-/- mice, Salmonella remained

in the circulation for up to 2h, with a significantly decreased (70-80%) organ count.

How these short term distribution differences relate to the long term outgrowth and

subsequent better survival of LDLR-/- mice, has still to be elucidated. Apparently, two

phenomena play a role: on the one hand the difference in early distribution and on the

other hand the difference in growth rate of salmonellae in the organs of mice.

Distribution effects may be explained by lipids in the circulation, as these are able

to influence the uptake of for instance endotoxin by various cells. Infusion of chylo

microns into rats has been shown to reduce the endotoxin uptake in Kupffer cells and to

increase uptake by hepatocytes [181]. Although LPS is not the same as Salmonella bac

teria, it could well be that lipoproteins directly interfere with Salmonella uptake, for

example by coating Salmonella surface structures with lipids. Also, it may be hypothe

sized that the LDL-receptor influences the uptake of bacteria. In this respect it is inter

esting that malaria parasites compete for the same receptor with certain lipoproteins

[182], and that Streptococcus pneumoniae has been shown to bind to a lipid receptor

[183].

Still, the difference in growth rate is not explained by any of these phenomena. We

found that phagocytosis of Salmonella by peritoneal macrophages, as well as the subse

quent intracellular killing did not differ between LDLR+/+ and LDLR-/- mice. This

suggests that the observed difference in growth rate is to be explained by an effect

122 Chapter 9

through other phagocytic cells, e.g. neutrophils, by induction of humoral antibacterial

factors or by divergent effects of LDLR-/- on macrophage types other than the perito

neal cells, which we studied.

References

1. Jones B, Falkow S. Salmonellosis: Host immune responses and bacterial virulen

ce. Annu Rev Immunol 1996; 14: 533-561.

2. Pang T, Bhutta ZA, Finlay BB, Altwegg M. Typhoid fever and other salmonello

sis: a continuing challenge. Trends Microbiol 1995; 3: 253-255.

3. Rathman M, Barker LP, Falkow S. The unique trafficking pattern of Salmonella

typhimurium-conlaining phagosomes in murine macrophages is independent of

the mechanism of bacterial entry. Infect Immun 1997; 65: 1475-1485.

4. Tang P, Foubister V, Pucciarelli MG, Finlay BB. Methods to study bacterial inva

sion. J Microbiol Meth 1993; 18: 227-240.

5. Mackaness GB, Blanden RV, Collins FM. Host-parasite relations in mouse

typhoid. J Exp Med 1966; 124: 573-583.

6. Hussein Gasem M, Dolmans WMV, Bambang Isbandrio B, Hendro Wahyono,

Keuter M, Djokomoeljanto R. Culture of Salmonella typhi and Salmonella para

typhi from blood and bone marrow in suspected typhoid fever. Trop Geograph

Med 1995; 47: 164-167.

7. Onuigbo MAC. Diagnosis of typhoid fever in Nigeria: misuse of the Widal test.

Trans R Soc Trop Med Hyg 1990; 84: 129-131.

8. Butler T, Rolfe RD, James GM, Hentges DJ. Soluble plasma antigen in experi

mental Salmonella typhimurium infection in mice. FEMS Immunol Med

. Microbiol 1994; 9: 325-332.

9. Van den Bergh ET AM, Hussein Gasem M, Keuter M, Dolmans WMV. Outcome

in three groups of patients with typhoid fever treated in Indonesia between 1948

and 1990. 1998; submitted.

10. Bambang Isbandrio B, Hussein Gasem M, Dolmans WMV, Hoogkamp-Korstanje

JAA. Comparative activities of three quinolones and seven comparison standard

drugs against Salmonella typhi from Indonesia. J Antimicrob Chemother 1994;

33: 1055-1056.

11. Simanjuntak CH, Paleólogo FP, Punjabi NH, Darmowigoto R, Harjining Toto-

sudirjo S, Haryanto P, Suprijanto E, Witham ND, Hoffman SL. Oral immuniza

tion against typhoid fever in Indonesia with Ty21a vaccine. Lancet 1991; 338:


1055-1059.

12. Sztein MB, Wasserman SS, Tacket CO, Edelman R, Hone D, Lindberg AA,

Levine MM. Cytokine production patterns and lymphoproliferative responses in

volunteers orally immunized with attenuated vaccine strains of Salmonella typhi.

J Infect Dis 1994; 170: 1508-1517.

13. Dinarello CA, Cannon JG, Wolff SM. New concepts on the pathogenesis of fever.

Rev Infect Dis 1988; 10: 168-189.

14. Dinarello CA, Wolff SM. Pathogenesis of fever in man. N Engl J Med 1979; 298:

607-612.

15. Netea MG, Kullberg BJ, Van der Meer JWM. Fever: Do endogenous pyrogens

circulate in the bloodstream? Clin Infect Dis 1998; in press:

16. Cranston WI, Hellon RF, Mitchell D, Townsend Y. Intraventricular injections of

drugs which inhibit phospholipase A2 suppress fever in rabbits. J Physiol 1983;

339: 97-105.

17. Vadas P, Pruzanski W, Stefanski E, Sternby В, Mustard R, Bohnen J, Fraser I,

Farewell V, Bombardier C. Pathogenesis of hypotension in septic shock: Correla

tion of circulating phospholipase A2 levels with circulatory collaps. Crit Care

Med 1988; 16: 1-7.

18. Pruzanski W, Vadas P. Phospholipase A2. A mediator between proximal and

distal effectors of inflammation. Immunol Today 1991; 12: 143-146.

19. Crowl RM, Stoller TJ, Conroy RR, Stoner CR. Induction of phospholipase A2

gene expression in human hepatoma cells by mediators of the acute phase respon

se. J Biol Chem 1991; 266: 2647-2651.

20. Schalkwijk С, Pfeilschifter J, Marki F, Van den Bosch H. Interleukin-lß, tumor

necrosis factor and forskolin stimulate the synthesis and secretion of group II

phospholipase A2 in rat mesangial cells. Biochem Biophys Res Commun 1991;

174: 268-275.

21. Vadas P, Hay JB. Involvement of circulating phospholipase A2 in the pathogene

sis of the hemodynamic changes in endotoxin shock. Can J Physiol Pharmacol

1983;61:561-566.

22. Nerad JL, Griffiths JK, Van der Meer JWM, Endres S, Keusch GT, Bennish M,

Dinarello CA, Cannon JG. A simple rapid method for studying cytokine produc

tion by endotoxin-stimulated leukocytes. Cytokine 1989; 1: 139.

23. Bendtzen K, Morling N, Fomsgaard A, Svenson M, Jakobson В, Odum Ν,

Svejgaard A. Association between HLA-DR2 and production of tumor necrosis

factor alpha and interleukin-1 by mononuclear cells activated by lipopoly-

124 Chapter 9

saccharide. Scand J Immunol 1988; 28: 599-606.

24. Nerad JL, Griffiths JK, Van der Meer JWM, Endres S, Poutsiaka DD, Keusch

GT, Bennish M, Salam MA, Dinarello CA, Cannon JG. Interleukin-lß (IL-lß),

IL-1 receptor antagonist, and TNF-a production in whole blood. J Leukocyte Biol

1992; 52: 687-692.



1993;5:216-221.

26. Vidal S, Tremblay ML, Govoni G, Gauthier S, Sebastiani G, Malo D, Skamene E,

Olivier M, Jothy S, Gros P. The Ity/Lsh/Bcg locus: Natural resistance to infection

with intracellular parasites is abrogated by disruption of the Nrampl gene. J Exp

Med 1995; 182: 655-666.

27. Vidal S, Lepage Ρ, Gauthier S, Gros P. Natural resistance to intracellular infec

tions. J of Imm 1996; 157: 3559-3568.

28. Govoni G, Vidal S, Gauthier S, Skamene E, Malo D, Gros P. The Bcg/Ity/Lsh

locus: Genetic transfer of resistance to infections in C57B1/6J mice transgenic for

the N r a m p l G l y 1 6 9 allele. Infect Immun 1996; 64: 2923-2929.

29. Swanson RN, O'Brien AD. Genetic control of the innate resistance of mice to

Salmonella typhimurium: Ity gene is expressed in vivo by 24 hours after infection.

J of Immunol 1983; 131: 3014-3020.

30. Kagaya K, Watanabe K, Fukazawa Y. Capacity of recombinant gamma interferon

to activate macrophages for Salmonella-killing activity. Infect Immun 1989; 57:

609-615.

31. Mastroeni P, Arena A, Costa GB, Liberto MC, Bonina L, Hormaeche CE. Serum

TNFalpha in mouse typhoid and enhancement of a Salmonella infection by anti-

TNFalpha antibodies. Microb Pathogen 1991; 11: 33-38.


resistance to a lethal challenge with Salmonella typhimurium - the effect correla

tes with the resistance allele at the ity locus. J Immunol 1994; 153: 212-219.

33. Morrissey PJ, Charrier К. GM-CSF administration augments the survival of ΓΓΥ-

resistant A/J mice, but not ΓΤΥ-susceptible C57B1/6 mice to a lethal challenge

with Salmonella typhimurium. J Immunol 1990; 14: 557-561.

34. Eckmann L, Fierer J, Kagnoff MF. Genetically resistant (ItyR) and susceptible

(ItyS) congenie mouse strains show similar cytokine responses following infec

tion with Salmonella Dublin. J Immunol 1996; 156: 2894-2900.

35. Pie S, Matsiota-Bernard P, Truffa-Bachi Ρ, Nauciel С. Gamma-interferon and


interleukin-10 gene expression in innately susceptible and resistant mice during

the early phase of Salmonella typhimurium infection. Infect Immun 1996; 64:

849-854.


during septicemic salmonellosis in calves. Infect Immun 1990; 58: 439^42.

37. Jotwani R, Tanaka Y, Watanabe K, Tanaka K, Kato N, Ueno К. Cytokine stimu

lation during Salmonella typhimurium sepsis in ItyS mice. J Med Microbiol 1995;

42: 348-52.

38. Greisman SE, Hornick RB, Wagner HN, Woodward TE. The integrity of the

endotoxin tolerance mechanisms during typhoid fever in man. Trans Assoc Am

Physicians 1967; 80: 250-258.


Clin Invest 1969; 48: 613-629.

40. Morrissey PJ, Charrier К, Vogel SN. Exogenous tumor necrosis factor alpha and

interleukin-1 alpha increase resistance to Salmonella typhimurium: efficacy is

influenced by the Ity and LPS loci. Infect Immun 1995; 63: 3158-3165.

41. Licht TR, Krogfelt KA, Cohen PS, Poulsen LK, Urbance J, Molin S. Role of lipo-

polysaccharide in colonization of the mouse intestine by Salmonella typhimurium

studied by in situ hybridization. Infect Immun 1996; 64: 3811-3817.

42. Hormaeche CE, Maskell DJ, Harrington К, Joysey H, Brock J. Mechanisms of

natural resistance to mouse typhoid. Bull Europ Physiopath Resp 1983; 19: 137-

142.

43. Hormaeche CE, Villareal В, Mastroeni Ρ, Dougan G, Chatfield SN. Immunity

mechanisms in experimental salmonellosis. In: Cabello F, Hormaeche С,

Mastroeni Ρ, Bonina L, ed. Biology of Salmonella. New York: Plenum Press,

1993:223-235.

44. Langermans JAM, Van der Hulst MB, Nibbering PH, Van Fürth R. Activation of

mouse peritoneal macrophages during infection with Salmonella typhimurium

does not result in enhanced intracellular killing. J Immunol 1990; 144: 4340-

4346.

45. Van Dissel JT, Stikkelbroeck JJM, Van den Barselaar MT, Sluiter W, Leijh PCJ,

Van Fürth R. Divergent changes in antimicrobial activity after immunologic acti

vation of mouse peritoneal macrophages. J Immunol 1987; 139: 1665-1672.

46. Van Dissel JT, Leijh PCJ, Van Fürth R. Differences in initial rate of intracellular

killing of Salmonella typhimurium by resident peritoneal macrophages from

various mouse strains. J Immunol 1985; 134: 3404-3410.

126 Chapter 9

47. Van Dissel JT, Stikkelbroeck JJM, Sluiter W, Leijh PCJ, Van Fürth R. Differen

ces in initial rate of intracellular killing of Salmonella typhimurium by granulo

cytes of salmonella susceptible C57B1/10 mice and salmonella resistant CBA

mice. J Immunol 1986; 136: 1074-1080.

48. O'Brien AD, Scher I, Formal SB. Effect of silica on the innate resistance of inbred

mice to Salmonella typhimurium infection. Infect Immun 1979; 25: 513-520.

49. Conlan JW, North RJ. Early pathogenesis of infection in the liver with the facul

tative intracellular bacteria Listeria monocytogenes, Francisella tularensis, and

Salmonella typhimurium involves lysis of infected hepatocytes by leukocytes.

Infect Immun 1992; 60: 5164-5171.

50. Conlan JW, North RJ. Neutrophil-mediated lysis of infected hepatocytes. ASM

News 1993; 59: 563-567.

51. Hsu HS. Pathogenesis and immunity in murine salmonellosis. Microbiol Rev

1989; 53: 390-409.

52. Conlan JW. Neutrophils prevent extracellular colonization of the liver micro-

vasculature by Salmonella typhimurium. Infect Immun 1996; 64: 1043-1047.

53. Hormaeche CE, Mastroeni P, Arena A, Uddin J, Joysey HS. T-cells do not media

te the initial suppression of a salmonella infection in the RES. Immunology 1990;

70: 247-250.

54. Muotiala A, Makela PH. The role of IFN-γ in murine Salmonella typhimurium

infection. Microb Pathogen 1990; 8: 135-141.

55. Mackaness GB. Cellular resistance to infection. J Exp Med 1962; 116: 381-417.

56. Mackaness GB. The immunological basis of acquired cellular resistance. J Exp

Med 1964; 120: 105-120.

57. Mackaness GB, Blanden RV. Cellular immunity. Progr Allergy 1967; 11: 89-140.

58. Mackaness GB. Resistance to intracellular infection. J Infect Dis 1971; 123: 439-

445.

59. Langermans JAM, Nibbering PH, Van Fürth R. Activation of murine peritoneal

macrophages by Salmonella typhimurium. Agents Actions 1989; 26: 208-210.

60. Senterfitt VC, Shands JW. Salmonellosis in mice infected with Mycobacterium

tuberculosis BCG. I. Role of endotoxin in infection. J Bacterid 1968; 96: 287-

292.

61. Schafer R, Eisenstein TK. Induction of natural killer cell activity by a Salmonella

vaccine. FASEB J 1988; 2: A677.

62. Vladoianu IR, Chang HR, Pechere JC. Expression of host resistance to

Salmonella typhi and Salmonella typhimurium: bacterial survival within macro-


phages of murine and human origin. Microb Pathogen 1990; 8: 83-90.

63. Lin FR, Hsu HS, Mumaw VR, Moncure CW. Confirmation of destruction of

salmonellae within murine peritoneal exudate cells by immunocytochemical

technique. Immunology 1989; 67: 394^00.

64. Degre M, Bukholm G, Czamlecki CW. In vitro treatment of HEp-2 cells with

human tumor necrosis factor-alpha and human interferons reduces invasiveness of

Salmonella typhimurium. J Biol Regul Homeost Agents 1989; 3: 1-7.

65. Portnoy DA, Schreiber RD, Connelly P, Tilney LG. γ-Interferon limits access of

Listeria monocytogenes to the macrophage cytoplasm. J Exp Med 1989; 170:

2141-2146.

66. Gregory SH, Wing EJ. IFN-gamma inhibits the replication of Listeria

monocytogenes in hepatocytes. J Immunol 1993; 151: 1401-1409.

67. Hess J, Ladel C, Miko D, Kaufmann SH. Salmonella typhimurium aroA- infection

in gene targeted immunodeficient mice: major role of CD4+ TCR-alpha beta cells

and IFN-gamma in bacterial clearance independent of intracellular location. J

Immunol 1996; 156: 3321-3326.

68. Barnes PF, Chatterjee D, Brennan PJ, Rea TH, Modlin RL. Tumor necrosis factor

production in patients with leprosy. Infect Immun 1992; 60: 1441-1446.

69. Havell EA. Production of tumor necrosis factor during murine listeriosis. J

Immunol 1987; 139: 4225-4231.

70. Liew FJ, Parkinson C, Millot S, Severn A, Carrier M. Tumor necrosis factor

(TNF) in leishmaniasis l.TNF mediates host protection against cutaneous leish

maniasis. Immunology 1990; 69: 570-573.

71. Nakane A, Minagawa T, Kato K. Endogenous tumor necrosis factor (cachectin) is

essential to host resistance against Listeria monocytogenes infection. Infect

Immun 1988; 56: 2563-2569.

72. Silva CL, Foss NT. Tumor necrosis factor in leprosy patients. J Infect Dis 1989;

159: 787-790.

73. Titus RG, Sherry B, Cerami A. Tumor necrosis factor plays a protective role in

experimental murine cutaneous leishmaniasis. J Exp Med 1989; 170: 2097-2104.

74. Pfeffer К, Matsuyama Τ, Kundig TM, Wakeham A, Kishihara К, Shahinian A,

Wiegmann К, Ohashi PS, Kronke M, Мак TW. Mice deficient for the 55 kDa

tumor necrosis factor receptor are resistant to endotoxic shock, yet succumb to L.

monocytogenes infection. Cell 1993; 73: 457-467.

75. Rothe J, Lesslauer W, Lotscher Η, Lang Y, Koebel Ρ, Kontgen F, Althage A,

Zinkernagel R, Steinmetz M, Bluethmann H. Mice lacking the tumour necrosis

128 Chapter 9

factor receptor-1 are resistant to TNF-mediated toxicity but highly susceptible to

infection by Listeria monocytogenes. Nature 1993; 364: 798-802.

76. Mosmann TR, Cherwinski H, Bond MW, Giedlin MA, Coffman RL. Two types

of murine helper T-cell clone. I. Definition according to profiles of lymphokine

activities and secreted proteins. J Immunol 1986; 136: 2348-2357.

77. Allen JE, Maizels RM. Thl-Th2: reliable paradigm or dangerous dogma. Immu

nol Today 1997; 18: 387-392.

78. Black RA, Rauch CT, Kozlosky CJ, Peschon JJ, Slack JL, Wolfson MF, Castner

BJ, Stocking KL, Reddy P, Srinivasan S, Nelson N, Boiani N, Schooley KA,

Gerhart M, Davis R, Fitzner JN, Johnson RS, Paxton RJ, March CJ, Cerretti DP.

A metalloproteinase disintegrin that releases tumour necrosis factor-alpha from

cells. Nature 1997; 385: 729-733.

79. Eigler A, Sinha B, Hartmann G, Endres S. Taming TNF: strategies to restrain this

proinflammatory cytokine. Immunology Today 1997; 18: 487-492.

80. Bancroft GJ, Sheehan KCF, Schreiber RD, Unanue ER. Tumor necrosis factor is

involved in the Τ cell-independent pathway of macrophage activation in SCID

mice. J Immunol 1989; 143: 127-130.

81. Langermans JAM, Van der Hulst MEB, Nibbering PH, Van Fürth R. Endogenous

tumor necrosis factor alpha is required for enhanced antimicrobial activity against

Toxoplasma gondii and Listeria monocytogenes in recombinant gamma

interferon-treated mice. Infect Immu 1992; 60: 5107-5112.

82. Mastroeni P, Villareal-Ramos B, Hormaeche CE. Effect of late administration of

anti-TNFalpha antibodies on a Salmonella infection in the mouse model. Microb

Pathogen 1993; 14: 473^80.



144: 1935-1941.

84. Nauciel C, Espinasse-Maes F. Role of gamma-interferon and tumor necrosis

factor alpha in resistance to Salmonella typhimurium infection. Infect Immun

1992; 60:450-454.

85. Tite JP, Dougan G, Chatfield SN. The involvement of tumor necrosis factor in

immunity to Salmonella infection. J Immunol 1991; 147: 3161-3164.


alpha antibodies on histopathology of primary Salmonella infections. Infect

Immun 1995; 63: 3674-3682.

87. Degré M, Bukholm G. Effect of tumor necrosis factor-alpha on infection with


Salmonella typhimurium in a mouse model. J Biol Regul Homeost Agents 1990;

4: 157-161.

88. Langermans JAM. The cytokine network, macrophage activation, and enhanced

antimicrobial activity. Leiden: thesis, 1992:

89. Murray HW. Interferon-gamma, the activated macrophage, and host defense

against microbial challenge. Ann Int Med 1988; 108: 595-608.

90. Cavaillon JM. Cytokines and macrophages. Biomed & Pharmacother 1994; 48:

445-453.

91. Huang S, Hendriks W, Althage A, Hemm S, Bluetmann H, Kamijo R, Vilcek J,

Zinkernagell RM, Aguet M. Immune response in mice that lack the interferon-γ

receptor. Science 1993; 259: 1742-1745.

92. Pie S, Truffa-Bachi Ρ, Pia M, Nauciel С. Thl response in Salmonella

typhimurium-inkctcd mice with a high or low rate of bacterial clearance. Infect

Immun 1997; 65:4509-4514.

93. Iizawa Y, Wagner RD, Czuprynski CJ. Analysis of cytokine mRNA expression in

Listeria-resistant C57BL/6 and Listeria-susceptible A/J mice during Listeria

monocytogenes infection. Infect Immun 1993; 61: 3739-44.

94. Neta R, Salvin SB. Resistance and susceptibility to infection in inbred murine

strains. Cell Immunol 1983; 75: 173-180.

95. Ramarathinam L, Niesei DW, Klimpel GR. Ity influences the production of IFN-y

by murine plenocytes stimulated in vitro with Salmonella typhimurium. J of

Immunol 1993; 150: 3965-3972.

96. Benbernou N, Nauciel С Influence of mouse genotype and bacterial virulence in

the generation of interferon-gamma-producing cells during the early phase of

Salmonella typhimurium infection. Immunol 1994; 83: 245-249.

97. Ramarathinam L, Shaban RA, Niesei DW, Klimpel GR. Interferon gamma (IFN-

y) production in the gut-associated lymphoid tissue and spleen following oral

Salmonella typhimurium challenge. Microb Pathogen 1991; 11: 347-356.

98. Muotiala A, Makela PH. Role of gamma interferon in late stages of murine sal

monellosis. Infect Immun 1993; 61: 4248^253.

99. Heinzel FP. The role of interferon-gamma in the pathology of experimental endo-

toxemia. J Immunol 1990; 145: 2920-2924.

100. Langermans JAM, Van der Hulst MEB, Nibbering PH, Hiemstra PS, Fransen L,

Van Fürth R. IFN-y-induced L-arginine-dependent toxoplasmastatic activity in

murine peritoneal macrophages is mediated by endogenous tumor necrosis factor-

a. J Immunol 1992; 148: 568-574.

130 Chapter 9

101. Langermans JAM, Nibbering PH, Van der Hulst MEB, Van Fürth R. Microbici

dal activities of Salmonella typhimurium- and interferon gamma activated mouse

peritoneal macrophages. Pathobiology 1991; 59: 189-193.

102. Osanto S, Van den Barselaar MT, Van Dissel JT. Increased intracellular killing of

bacteria in vitro by monocytes of patients with metastatic melanoma before and

during treatment with interferon-gamma and interferon-alpha. Eur j Cancer 1991;

27: 478-482.

103. Degre M, Bukholm G. Orally administered interferon-gamma but not tumor

necrosis factor-alpha suppress infection with Salmonella typhimurium in a mouse

model. J Biol Reg Homeost Agents 1995; 9: 15-20.

104. Dinarello CA. Interleukin-1 and interleukin-1 antagonism. Blood 1991; 77: 1627-

1652.

105. Kullberg BJ, Van 't Wout JW, Van Fürth R. Role of granulocytes in enhanced

host resistance to Candida albicans induced by recombinant interleukin-1. Infect

Immun 1990; 58: 3319-3324.

106. Czuprynski CJ, Brown JF. Recombinant murine interleukin-la enhancement of

nonspecific antibacterial resistance. Infect Immun 1987; 55: 2061-2065.

107. Van der Meer JWM, Barza M, Wolff SM, Dinarello CA. A low dose of recombi

nant interleukin 1 protects granulocytopenic mice from lethal Gram-negative

infection. Proc Natl Acad Sci USA 1988; 85: 1620-1623.

108. Curfs JHAJ, van der Meer JWM, Sauerwein R, Eling WMC. Low dosages of

interleukin-1 protect mice against lethal cerebral malaria. In: Dinarello CA, Klu

ger M, Oppenheim J, Powanda M, ed. Interleukin-1 and related cytokines. New

York: Alan Liss, 1990.

109. Vogels MTE, Mensink EJBM, Ye K, Boerman OC, Verschuren CMM, Dinarello

CA, Van der Meer JWM. Differential gene expression for IL-1 receptor anta

gonist, IL-1, and TNF receptors and IL-1 and TNF synthesis may explain IL-1-

induced resistance to infection. J Immunol 1994; 153: 5772-5780.

110. Vogels MTE, Van Rooyen N, Bemelmans MHA, Jansen M, Van der Meer JWM.

Interleukin-1 (IL-l)-induced resistance to bacterial infection: role of the macro

phage and soluble TNF receptors. In: Vogels MTE, ed. Mechanisms of enhanced

natural resistance to infection induced by interleukin-1. Thesis: Catholic Univer

sity Nijmegen, 1994.

111. Umezawa K, Akaike T, Fujii S, Suga M, Setoguchi К, Ozawa A, Maeda H.

Induction of nitric oxide synthesis and xanthine oxidase and their roles in the anti

microbial mechanism against Salmonella typhimurium infection in mice. Infect


Immun 1997; 65: 2932-2940.

112. Vannier E, Miller LC, Dinarello CA. Coordinated antiinflammatory effects of

interleukin 4: interleukin 4 suppresses interleukin 1 production but up-regulates

gene expression and synthesis of interleukin 1 receptor antagonist. Proc Natl

Acad Sci USA 1992; 89: 4076-W80.

113. Everest P, Allen J, Papakonstantinopoulou A, Mastroeni P, Roberts M, Dougan

G. Salmonella typhimurium infections in mice deficient in Interleukin-4 produc

tion: role of IL-4 in infection associated pathology. J Immunol 1997; 159: 1820-

1827.

114. Denich K, Borlin P, O'Hanley PD, Howard M, Heath AW. Expression of the

murine interleukin-4 gene in an attenuated aroA strain of Salmonella typhimu

rium: persistence and immune response in BALB/c mice and susceptibility to

macrophage killing. Infect Immun 1993; 61:4818-4827.

115. Schindler R, Mancilla J, Endres S. Correlations and interactions in the production

of interleukin-6 (IL-6), interleukin-1 (IL-1) and tumor necrosis factor (TNF) in

human blood mononuclear cells: IL-6 suppresses IL-1 and TNF. Blood 1990; 75:

40-47.

116. Tilg H, Dinarello CA, Mier JW. IL-6 and APPs: anti-inflammatory and immuno-

suppresive mediators. Immunol Today 1997; 18: 429-432.

117. Dalrymple SA, Lucían LA, Slattery R, McNeil Τ, Aud DM, Fuchino S, Lee F,

Murray R. Interleukin-6-deficient mice are highly susceptible to Listeria mono

cytogenes infection: correlation with inefficient neutrophilia. Infect Immun 1995;

63: 2262-2268.

118. Czuprynski CJ, Haak-Frendscho M, Maroushek N, Brown JF. Effects of recom

binant human interleukin-6 alone and in combination with recombinant inter

leukin-la and tumor necrosis factor-alpha on antibacterial resistance in mice.

Antimicrob Agents Chemother 1992; 36: 68-70.

119. Howard M, O'Garra A, Ishida H, de Waal Malefyt R, de Vries J. Biological

properties of interleukin-10. J Clin Immunol 1992; 12: 239-247.

120. Goldman M, Stordeur P. Interleukin-10 as an anti-stress cytokine. Eur Cytokine

Netw 1997; 8: 301-302.

121. Powrie F, Menon S, Coffman RL. Interleukin-4 and interleukin-10 synergize to

inhibit cell-mediated immunity in vivo. Eur J Immunol 1993; 23: 3043-3049.

122. Arai Τ, Hiromatsu К, Nishimura Η, Kimura Y, Kobayashi Ν, Ishida Η, Nimura

Y, Yoshikai Y. Effects of in vivo administration of anti-EL-10 monoclonal anti

body on the host defence mechanism against murine Salmonella infection. Immu-

132 Chapter 9

nology 1995; 85: 381-388.

123. Hall SS. IL-12 at the crossroads. Science 1995; 268: 1432-1434.

124. Van der Pouw Kraan ТСГМ, Boeije LCM, Smeenk RJT, Wijdenes J, Aarden LA.

Prostaglandin-E2 is a potent inhibitor of human interleukin-12 production. J Exp

Med 1995; 181:775-779.

125. Skeen MJ, Miller MA, Shinnick TM, Ziegler HK. Regulation of murine macro

phage IL-12 production. Activation of macrophages in vivo, restimulation in vitro

and modulation by other cytokines. J Immunol 1996; 156: 1196-1206.

126. Murray HW, Hariprashad J. Interleukin-12 is effective treatment for an esta

blished systemic intracellular infection: experimental visceral leishmaniasis. J

Exp Med 1995; 181: 387-391.

127. Wang Z, Zheng S, Cony DB, Dalton DK, Seder RA, Reiner SL, Locksley RM.

Interferon-gamma independent effects of interleukin-12 administered during acute

or established infection due to Leishmania major. Proc Natl Acad Sci USA 1994;

91: 12932-12936.

128. Chong C, Bost KL, Clements JD. Differential production of interleukin-12

mRNA by murine macrophages in response to viable or killed Salmonella spp.

Infect Immun 1996; 64: 1154-1160.

129. Kincy-Cain T, Clements JD, Bost KL. Endogenous and exogenous interleukin-12

augment the protective immune response in mice orally challenged with Salmo

nella dublin. Infect Immun 1996; 64: 1437-1440.

130. Mastroeni P, Harrison JA, Chabalgoity JA, Hormaeche CE. Effect of interleukin-

12 neutralization on host resistance and gamma-interferon production in mouse

typhoid. Infect Immun 1996; 64: 189-196.

131. Kincy-Cain T, Bost KL. Substance P-induced IL-12 production by murine macro

phages. J Immunol 1997; 158: 2334-2339.

132. Kincy-Cain T, Bost KL. Increased susceptibility of mice to Salmonella infection

following in vivo treatment with the substance Ρ antagonist, spantide Π. J Immu

nol 1996; 157: 255-264.

133. Bost KL, Clements JD. Intracellular Salmonella dublin induces sustantial secre

tion of the 40-kilodalton subunit of interleukin-12 (IL-12) but minimal secretion

of IL-12 as a 70-kilodalton protein in murine macrophages. Infect Immun 1997;

65:3186-3192.

134. Heinzel FP, Hujer AM, Ahmed FN, Rerko RM. In vivo production and function

of IL-12 p40 homodimers. J Immunol 1997; 158: 4381-4388.

135. Dale DC, Liles WC, Summer WR, Nelson S. Review: Granulocyte colony-stimu-


lating factor. Role and relationships in infectious diseases. J Infect Dis 1995; 172:

1061-1075.

136. Kullberg В J, Netea MG, Curfs JHAJ, Keuter M, Meis JFGM, Van der Meer

JWM. Recombinant murine granulocyte colony-stimulating factor protects

against acute disseminated Candida albicans infection in non-neutropenic mice. J

Infect Dis 1998; 177: 175-181.

137. Nelson S. Role of granulocyte colony-stimulating factor in the immune response

to acute bacterial infection in the nonneutropenic host: an overview. Clin Infect

Dis 1994; 18(suppl2): S197-204.

138. Barsig J, Bundschuh DS, Härtung Τ, Bauhofer Α, Sauer A, Wendel A. Control of

fecal peritoneal infection in mice by colony-stimulating factors. J Infect Dis 1996;

174: 790-799.

139. Appelberg R. G-CSF in a murine model of Mycobacterium avium infection. In:

Andresen J, ed. Abstracts G-CSF Research Meeting. Santa Monica, CA: Amgen,

1996;36-38.

140. Serushago BA, Yoshikai Y, Handa T, Mitsuyama M, Muramori K, Nomoto K.

Effect of recombinant human granulocyte colony-stimulating factor (rh G-CSF)

on murine resistance to Listeria monocytogenes. Immunology 1992; 75: 475-480.

141. Shinomiya N, Tsuru S, Katsura Y, Kayashima S, Nomoto K. Enhanced resistance

against Listeria monocytogenes achieved by pretreatment with granulocyte colo

ny-stimulating factor. Infect Immun 1991; 59:4740-4743.

142. Kitabayashi A, Hirokawa M, Hatano Y, Lee M, Kuroki J, Niitsu H, Miura AB.

Granulocyte colony-stimulating factor downregulates allogeneic immune respon

ses by posttranscriptional inhibition of tumor necrosis factor-alpha production.

Blood 1995; 86: 2220-2227.

143. Härtung Τ, Docke WD, Gantner F, Krieger G, Sauer A, Stevens Ρ, Volk HD,

Wendel Α. Effect of granulocyte colony-stimulating factor treatment on ex vivo

blood cytokine response in human volunteers. Blood 1995; 85: 2482-2489.

144. Dinarello CA, Cannon JG, Mancilla J, Bishai I, Lees J, Coceani F. Interleukin-6

as an endogenous pyrogen: induction of prostaglandin E2 in brain but not in peri

pheral blood mononuclear cells. Brain Res 1991; 562: 199-206.

145. Hack CE, De Groot ER, Felt-Bersma RJF. Increased plasma levels of interleukin-

6 in sepsis. Blood 1989; 74: 1704-1710.

146. Waage A, Brandtzaeg P, Halstensen A, Kierulf P, Espevik T. The complex

pattern of cytokines in serum from patients with meningococcal septic shock.

Association between interleukin-6, interleukin-1 and fatal outcome. J Exp Med

134 Chapter 9

1989; 169: 333-338.

147. biada M, Tojo H, Kawata S, Tarui S, Okamoto M. Induction of group Il-like

phospholipase A2 by lipopolysaccharide in the liver of BCG-primed rats.

Biochem Biophys Res Commun 1991; 174: 1077-1083.

148. Morgan DW, Anderson CM, Meyers KP. Recombinant human type Π phospholi

pase A2 lacks edema producing activity in rat. Eur J Pharmacol 1993; 235: 45-50.

149. Hack CE, Wolbink GJ, Schalkwijk С, Speijer H, Hermens WT, Van den Bosch

H. A role for secretory phospholipase A2 and C-reactive protein in the removal of

injured cells. Immunology Today 1997; 18: 111-115.

150. Pruzanski W, Vadas P. Secretory synovial fluid phospholipase A2 and its role in

the pathogenesis of inflammation in arthritis. J Rheumatol 1988; 15: 1601-1603.

151. Kuslys T, Vishwanath BS, Frey FJ, Frey BM. Differences in phospholipase A2

activity between males and females and Asian Indians and Caucasians. Eur J Clin

Invest 1996;26:310-315.

152. Lucas R, Juillard P, Decoster E, Redard M, Burger D, Donaty Y, Giroud C,

Monso-Hinard C, De Kesel T, Buurman WA, Moore MW, Dayer JM, Fiers W,

Bluethmann H, Grau GE. Crucial role of tumor necrosis factor (TNF) receptor 2

and membrane-bound TNF in experimental cerebral malaria. Eur J Immunol

1997; 27: 1719-1725.

153. Van der Poll T, Coyle SM, Moldawer LL, Lowry SF. Changes in endotoxin-indu-

ced cytokine production by whole blood after in vivo exposure of normal humans

to endotoxin. J Infect Dis 1996; 174: 1356-1360.

154. Mengozzi M, Ghezzi P. Cytokine down-regulation in endotoxin tolerance. Eur

Cytokine Netw 1993; 4: 89-98.

155. Luger A, Graf H, Schwarz HP, Stummvoll HK, Luger TA. Decreased serum

interleukin-1 activity and monocyte interleukin-1 production in patients with fatal

sepsis. Crit Care Med 1986; 14:458-461.

156. Simpson SQ, Modi H, Balk RA, Bone RC, Casey LC. Reduced alveolar macro

phage production of tumor necrosis factor during sepsis in mice and man. Crit

Care Med 1991; 19: 1060-1066.

157. Srugo I, Berger A, Lapidot Z, Katz R, Pollak S. Interleukin-1 secretion by blood

monocytes of septic premature infants. Infection 1991; 3: 150-154.

158. Helminen M. Interleukin-1 production from peripheral blood monocytes in septic

infections in children. Scand J Infect Dis 1991; 23: 607-611.

159. Rozenbaum M, Katz R, Rozner I, Pollack S. Decreased interleukin-1 activity

released from circulating monocytes of patients with familial Mediterranean fever


during in vitro stimulation by lipopolysaccharide. J Rheumatol 1992; 19: 416-

418.

160. Schattner A, Lachmi M, Livneh A, Pras M, Hahn T. Tumor necrosis factor in

familial Mediterranean fever. Am J Med 1991; 90:434-438.

161. Drenth JPH, Van Uum SHM, Van Deuren M, Pesman GJ, Van der Ven-

Jongekrijg J, Van der Meer JWM. Endurance run increases circulating IL-6 and

IL-lra but downregulates ex vivo TNF-a and IL-lß production. J Appi Physiol

1995; 79: 1497-1503.

162. Kruimel JW, Naber AHJ, Van der Vliet AJ, Buskens FGM, Pesman G, Sweep F,

Liem T, Kho HG, Van der Meer JWM, Jansen JBMJ. Effects of abdominal surge

ry on cytokines in a clinical trial designed to study the effects of parenteral admi

nistration of lipid emulsions. Neth J Med 1994; 45: A17-A18.

163. Granowitz EV, Porat R, Mier JW, Orencole SF, Kaplanski G. Lynch EA, Ye K,

Vannier E, Wolff SM, Dinarello CA. Intravenous endotoxin suppresses the cyto

kine response of peripheral blood mononuclear cells of healthy humans. J Immu

nol 1993; 151: 1637-1645.

164. Van der Poll T, de Waal Malefijt R, Coyle SM, Lowry SF. Antiinflammatory

cytokine responses during clinical sepsis and experimental endotoxemia: sequen

tial measurements of plasma soluble interleukin (IL)-l receptor type Π, IL-10 and

IL-13. J Infect Dis 1997; 175: 118-122.

165. Van der Poll T, Jansen J, Endert E, Sauerwein HP, Van Deventer SJH.

Noradrenaline inhibits lipopolysaccharide-induced tumor necrosis factor and

interleukin-6 production in human whole blood. Infect Immun 1994; 62: 2046-

2050.

166. Randow F, Syrbe U, Meisel C, Krausch D, Zuckermann H, Platzer С, Volk HD.

Mechanism of endotoxin desensitization: involvement of interleukin-10 and trans

forming growth factor-ß. J Exp Med 1995; 181: 1887-1892.

167. Frankenberger M, Pechumer H, Loms Ziegler-Heitbrock HW. Interleukin-10 is

upregulated in LPS tolerance. J Inflamm 1995; 45: 56-63.

168. Loms Ziegler-Heitbrock HW, Frankenberger M, Wedel A. Tolerance to lipopoly

saccharide in human blood monocytes. Immunobiol 1995; 193: 217-223.

169. Docke WD, Randow F, Krausch D, Khusru A, Reinke P, Volk HD, Кох W.

Monocyte deactivation in septic patients: restoration by IFN-gamma treatment.

Nature Med 1997; 3: 678-681.

170. Van Deuren M. Kinetics of tumour necrosis factor alpha, soluble tumour necrosis

factor receptors, interleukin-1-beta and its receptor antagonist during serious

136 Chapter 9

infections. Eur J Clin Microbiol Infect Dis 1994; 13 Suppl 1: 12-16.





172. Vogels MTE, Van der Meer JWM. Use of immune modulators in nonspecific

therapy of bacterial infections. Antimicrob Agents Chemother 1992; 36: 1-5.

173. Kullberg BJ, Van 't Wout JW, Van Fürth R. No effect of recombinant human

interleukin-1 on the numbers of peripheral blood and peritoneal leukocytes during

an acute inflammation. Inflammation 1991; 15: 457-470.

174. Lajarin F, Rubio G, Galvez J, Garcia-Penarrubia P. Adhesion, invasion and intra

cellular replication of Salmonella typhimurium in a murine hepatocyte cell line.

Effect of cytokines and LPS on antibacterial activity of hepatocytes. Microb

Pathog 1996; 21: 319-329.

175. Van 't Wout JW, Poell R, Van Fürth R. The role of BCG-PPD-activated macro

phages in resistance against systemic candidiasis in mice. Scand J Immunol 1992;

36:713-719.

176. Sisson SD, Dinarello CA. Production of interleukin-1 alpha, interleukin-1 beta

and tumor necrosis factor by human mononuclear cells stimulated with granulo

cyte-macrophage colony-stimulating factor. Blood 1988; 72: 1368-1374.

177. Pollmacher T, Korth C, Mullington J, Schreiber W, Sauer J, Vedder H, Galanos

С, Holsboer F. Effects of granulocyte colony-stimulating factor on plasma cytoki

ne and cytokine receptor levels and on the in vivo host response to endotoxin in

healthy men. Blood 1996; 87: 900-905.

178. Pajkrt D, Manten A, Van der Poll T, Van Buul TMM, Jansen J, Ten Cate WJ,

Van Deventer SJ. Modulation of cytokine release and neutrophil function by gra

nulocyte colony-stimulating factor during endotoxemia in humans. Blood 1997;

90: 1415-1424.

179. Gorgen I, Härtung Τ, Leist M, Niehorster M, Tiegs G, Uhlig S, Weitzel F,

Wendel A. Granulocyte colony-stimulating factor treatment protects rodents

against lipopolysaccharide-induced toxicity via suppression of systemic tumor

necrosis factor. J Immunol 1992; 149: 918-924.

180. Andresen J. Enhanced antimicrobial activity of polymorphonuclear leukocytes

and antibiotics in the presence of filgrastim. In: Andresen J, ed. Abstracts G-CSF

Research Meeting. Santa Monica, CA: Amgen, 1996;36-38.

181. Harris HW, Grunfeld C, Feingold KR, Read TE, Kane JP, Jones AL, Eichbaum


EB, Bland GF, Rapp JH. Chylomicrons alter the fate of endotoxin, decreasing

tumor necrosis factor release and preventing death. J Clin Invest 1993; 91: 1028-

1034.

182. Sinnis P, Willnow TE, Briones MRS, Herz J, Nussenzweig V. Remnant lipopro

teins inhibit malaria sporozoite invasion of hepatocytes. J Exp Med 1996; 184:

945-954.

183. Cundell DR, Gerard NP, Gerard C, Idanpaan-Heikkila I, Tuomanen EI. Strepto

coccus pneumoniae anchor to activated human cells by the receptor for platelet-

activating factor. Nature 1995; 377: 435-438.

138 Chapter 9

Chapter 10

Summary

140 Chapter 10

Summary 141

Typhoid fever is a severe systemic infection, confined to humans and is usually

caused by Salmonella typhi. Mostly children and young adults may contract the disease

in places where water is not properly purified, as the spreading is by the fecal-oral

route. The incidence of typhoid fever is high in Indonesia, with more than 1000 cases

per 100 000 inhabitants. In Indonesia antibiotics reduced the mortality from 18% to 4%.

Joint research between the Catholic University Nijmegen, the Netherlands, and the

Diponegoro University, Semarang, Java, Indonesia was initiated in 1989. The purpose

of this collaboration was to investigate clinical and immunological features of patients

with typhoid fever, as well as to study the pathogenesis of Salmonella infection in an

experimental animal model.

In Chapter 2, the pathogenesis of Salmonella infections is described and the role of

cytokines involved in these infections is reviewed. Whereas Salmonella typhi is not

pathogenic to mice, experimental Salmonella typhimurium infection in mice leading to

a disease similar to human typhoid fever, is used to investigate pathogenesis. Therefore

studies with this mouse model are described in Chapter 2. In contrast to extra cellular

Gram-negative septicemia, where high circulating concentrations of proinflammatory

cytokines such as interleukin-1 (IL-1) and tumor necrosis factor (TNFa) have been con

sidered the main cause of organ damage and death and inhibition of these proinflamma

tory cytokines is beneficial, in intracellular infections this very treatment is deleterious.

TNFa is supposed to be needed locally (in small amounts) and administration of TNFa

inhibited the outgrowth of the microorganisms in Salmonella infections.

Pathogenesis of typhoid fever and other Salmonella infections involves several

processes. Regarding the symptoms especially fever in typhoid fever, pyrogenic cyto

kines or other pyrogenic mediators would be expected to be present in the circulation. It

is generally believed that the release of prostaglandin (PGE2) in the brain is the final

pathway in the pathogenesis of fever and this is thought to be induced by circulating

proinflammatory cytokines. These so-called endogenous pyrogens are produced at the

site of infection. In Chapter 3 we report low concentrations of circulating proinflam

matory cytokines, but high concentrations of the antiinflammatory and non-pyrogenic

interleukin-1 receptor antagonist (IL-IRA) and soluble tumor necrosis factor receptors

(sTNFRI and sTNFRII) during the acute phase of typhoid fever. Since pyrogenic cyto

kines were not found in febrile patients with typhoid fever, it was suggested that other

mechanisms were involved. These included other mediators such as secretory phospho-

lipase A2 (sPLA2). Phospholipases are lipolytic enzymes which catalyse the degrada-

142 Chapter 10

tion of phospholipids. The release of sPLA2 is induced by interleukin-1 (IL-1) and

TNFct and the enzyme is able to mediate the production of arachidonic acid and PGE2.

As intraventricular injection of inhibitors of sPLA2 had been shown to suppress fever

in rabbits, sPLA2 could play a role as a circulating pyrogen. In Chapter 4K we indeed

found elevated concentrations of extracellular group II 14-kDa sPLA2 during the acute

phase of typhoid fever. Concentrations of sPLA2 fell to recovery. We considered that

this bioactive sPLA2 might be responsible for the symptoms of typhoid fever.

In the concept of fever, it is important to note that although concentrations of circu

lating pyrogenic cytokines were low in the acute febrile phase of typhoid fever, mRNA

for TNFa and EL-18 was found to be present in circulating whole blood cells (Chapter

5). Another hypothesis is that these circulating cells themselves produce or induce

pyrogenic cytokines in the central nervous system.

Also in Chapter 3, we report down-regulation of the capacity of circulating blood

cells to produce proinflammatory cytokines after stimulation by lipopolysaccharide

(LPS) in the acute phase of typhoid fever. For this purpose, we have used a whole blood

cytokine test. This depressed cytokine production was not a consequence of a lower

number of white blood cells or exhaustion of these cells during the acute phase of the

infection. After the initial phase of infection there is a posttrancriptional switch-off of

the production of proinflammatory cytokines and a switch-on to an anti-inflammatory

repertoire in mononuclear cells, as antiinflammatory factors such as IL-IRA were not

suppressed.

To investigate the mechanism of down-regulated cytokine production in typhoid

fever further, messenger RNA for DL-lß and TNFa in whole blood was measured. In

Chapter 5, the finding of preserved mRNA for IL-Iß and TNFa in the acute phase of

typhoid fever, also denies an exhausted production and is in favor of post-transcriptio-

nal inhibition of cytokine production.

In-vivo down-regulation of cytokine production is a common phenomenon in many

circumstances. Most likely, this is meant to protect against overwhelming cytokinemia,

but down-regulation can impair immune functions. The benefit of down-regulation in

patients with typhoid fever is difficult to understand in view of experimental animal

studies with Salmonella infection, in which the proinflammatory cytokines are required

to survive the infection (Chapter 2). Possibly the deeper downregulated proinflam

matory cytokines in complicated cases could reflect a less effective host defense and

immunotherapy might be needed to improve host defense.

Summary 143

In order to better understand the host defense against Salmonella, we investigated

the cytokine responses in mice (Chapter 6) and compared these with those found in

patients with typhoid fever (Chapters 3, 4 and 5). Without infection, circulating cyto

kine concentrations but also LPS-stimulated ex-vivo production in whole blood were

undetectable. The latter inability to produce cytokines is the reverse of the human situa

tion, since human peripheral blood cells can be stimulated to produce large amounts of

proinflammatory cytokines without infection. We found a general upregulation of pro

inflammatory cytokine production in this animal model during infection, which is also

in contrast with what we have found in humans with typhoid fever, described in

Chapter 3.

In line with other studies from our laboratory, we asked the question whether admi

nistration of recombinant IL-1 would exert a protective effect in a murine model of

Salmonella typhimurium infection (Chapter 6). Administration of rhIL-la led to а

significantly decreased outgrowth of microorganisms in the liver and spleen of mice.

This effect of rhIL-la pretreatment was seen as early as 24 hours after the onset of the

infection and the microbiological effect of a single dose of rhIL-lot lasted throughout

the study. Peritoneal cells produced significantly more cell-associated IL-la 24 hours

after pretreatment with rhIL-la. The early beneficial effect of this pretreatment on

Salmonella infection might have been established by several mechanisms. The fact that

IL-1 pretreatment protects against different types of infection is in favor of a general

phenomenon. We considered this to be macrophage activation.

Neutrophils are able to ingest and kill Salmonella, and they are important in the

outcome of this infection in vivo (Chapter 2). In Chapter 7, we describe that recom

binant murine granulocyte colony-stimulating factor (rmG-CSF) improved survival of

mice with a potentially lethal S. typhimurium infection. However, the outgrowth of S.

typhimurium in the organs of rmG-CSF pretreated mice was not different from that of

control mice. In our experiments, LPS-stimulated TNFoc production of resident perito

neal macrophages was significantly higher in rmG-CSF treated mice, 24 hours after

pretreatment. Histological examination of the liver after infection showed more loca

lized inflammatory foci in the rmG-CSF treated animals. These results show that pre

treatment with rmG-CSF is beneficial in the facultative intracellular S. typhimurium

infection in mice.

As we found a trend for more outgrowth in continuously treated animals, it was

questioned, whether rG-CSF has a beneficial effect on intracellular infections through

enhanced (early) neutrophil recruitment, and whether the (late) antiinflammatory

144 Chapter 10

properties of rG-CSF are deleterious, e.g., through inhibition of TNFa mediated

macrophage activation.

Although Salmonella is a Gram-negative rod, the role of endotoxin in the patho

genesis of infection caused by these organism is controversial. We approached this

problem (Chapter 8) by using low density lipoprotein receptor knock-out (LDLR-/-)

mice, of which the circulating lipoproteins are able to bind and neutralize bacterial LPS,

thereby reducing the induction of proinflammatory cytokines. The presence of high

concentrations of lipoproteins were shown to improve the outcome of S. typhimurium

infection. We show, that neutralization of endotoxin, which in extracellular Klebsiella

infection is the beneficial effect of lipids, is not the mechanism behind the better survi

val of S. typhimurium infection.

We found effects on distribution and later outgrowth. The former may be explained

by lipoproteins directly interfering with Salmonella uptake, for example by coating

Salmonella surface structures with lipids. Also, it may be hypothesized that the LDL-

receptor influences the uptake of bacteria.

For the effect on outgrowth, we found that phagocytosis of Salmonella by perito

neal macrophages, as well as the subsequent intracellular killing did not differ between

mice with or without the LDL-receptor. This suggests that the observed difference in

growth rate is to be explained by an effect through other phagocytic cells than the

peritoneal cells, which we studied.

Conclusions In patients with typhoid fever, we found down-regulation of proinflammatory cyto

kine production, which has been shown to be a common phenomenon in infections and

other situations. This observation supports the notion that immunotherapy may be bene

ficial for Salmonella typhi infection and that such interventions should be aimed at sup-

pletion or upregulation of immunomodulators that enhance host defense to the infec

tion.

We studied immunomodulation in experimental Salmonella typhimurium infection

in mice. It was found that pretreatment with recombinant IL-1 had beneficial effects,

with less outgrowth of microorganisms in the organs of mice. It is hypothesized that

this is due to a distribution effect probably mediated by activated macrophages.

Pretreatment with rmG-CSF had also a favorable effect in mice in terms of survival

to S. typhimurium infection. Outgrowth of microorganisms in the organs however was

not affected. The early effect of rmG-CSF on neutrophil recruitment and activation was

Summary 145

considered the most important mechanism for the survival. In addition, rmG-CSF led to

an increase of the LPS-induced TNFa production by peritoneal macrophages at the

time of infection, which may have contributed to the outcome.

The beneficial effect of lipoproteins on S. typhimurium infection was observed on

both early distribution and subsequent outgrowth of the microorganisms. This indicates

that lipoproteins and their receptors are crucial players in the host defense against

Salmonella infections, in the early distribution as well as in the intracellular killing by

phagocytes.

These observations in experimental S. typhimurium infection in mice give direction

to development of immunotherapy for Salmonella typhi infection in patients.

146 Chapter 10

Chapter 11

Ringkasan

148 Chapter 11

Ringkasan 149

Demam tifoid merupakan penyakit infeksi sistemile yang menyerang manusia dan

pada umumnya disebabkan oleh Salmonella typhi. Penyakit ini terutama menyerang

anak-anak dan dewasa muda dengan cara penyebaran melalui jalur fekal-oral dan oleh

karenanya banyak ditemukan pada daerah yang purifikasi airaya tidak sempuma.

Insidensi demam tifoid di Indonesia tinggi mencapai 1000 kasus per 100.000 penduduk.

Dengan penggunaan antibiotika mortalitas dapat kurangi dari 18% menjadi 4%.

Kerjasama dalam bidang penelitian antara Universitas Katolik Nijmegen dengan

Universitas Diponegoro Semarang telah diawali sejak tahun 1989. Tujuan dari kerja

sama ini adalah untuk mempelajari secara lebih lengkap gambaran klinik dan imuno-

logik penderita demam tifoid serta penelitian patogénesis infeksi Salmonella pada bina-

tang coba.

Pada Bah 2 dibicarakan patogénesis dan peran sitokin pada infeksi Salmonella.

Berdasarkan pada kenyataan bahwa Salmonella typhi bersifat apatogen pada mencit

maka untuk mempelajari patogénesis demamtifoid pada binatang (mencit) dipergu-

nakan Salmonella typhimurium yang akan memberikan penyakit serupa dengan demam

tifoid pada manusia. Studi dengan mempergunakan mencit sebagai model dijelaskan

pada Bah 2. Berlawanan dengan septikemia kuman Gram-negatif ekstraseluler dimana

konsentrasi tinggi sitokin proinflamasi seperti interleukin-1 (IL-1) dan tumor nekrosis

faktor (TNFa) yang dianggap sebagai penyebab utama dari kerusakan organ dan kema-

tian serta penghambatan sitokin proinflamasi akan bermanfaat, pada infeksi kuman

intraseluler perlakuan semacam ini justru akan membahayakan. TNFa diperkirakan

diperlukan secara lokal (dalam jumlah kecil) dan pemberian TNFa dapat mencegah

pertumbuhan microorganisme pada infeksi Salmonella.

Patogénesis demam tifoid dan infeksi Salmonella lainnya melibatkan beberapa

proses. Dengan terdapatnya gejala febris pada demam tifoid diharapkan terdapatnya

sitokin pirogenik dan mediator pirogenik lainnya pada sirkulasi. Secara umum dianggap

bahwa pelepasan prostaglandin (PGE2) di otak yang dirangsang produksinya oleh

sitokin proinflamasi adalah merupakan jalur akhir dari patogénesis febris. Sitokin yang

disebut juga sebagai pirogen endogenik ini diproduksi secara lokal pada tempat terjadi-

nya infeksi.

Pada Bab 3 kami laporkan terdapatnya sitokin proinflamasi dengan kadar rendali di

dalam sirkulasi disertai dengan terdapatnya sitokin antiinflamasi (non-pirogenik) inter

leukin-1 reseptor antagonis (IL-1RA) serta "soluble tumor necrosis factor receptors"

(sTNFRI dan sTNFRII) dengan kadar tinggi pada fase akut demam tifoid. Dengan tidak

150 Chapter 11

diternukannya sitokin pirogenik pada fase febris pada demam tifoid maka diperkirakan

bahwa terdapat mekanisme lain berperan.

Termasuk diantaranya adalah "secretory phospholipase A2" (PLA2). Fosfolipase

adalah ensim lipolitik yang dapat memecah fosfolipid. Pelepasan sPLA2 dipacu oleh

IL-1 dan TNFcc. Ensim ini dapat berperan sebagai mediator untuk terbentuknya asam

arakbidonat dan PGE2. Pemberian injeksi inhibitor sPLA2 pada kelinci dapat mence-

gah timbulnya panas. Oleh karena itu sPLA2 diperkirakan merupakan zat pirogen pada

sirkulasi. Pada Bab 4 kami dapatkan peningkatan kadar Π 14-kDa sPLA2 pada fase

akut demam tifoid. Sedangkan kadar sPLA2 menurun pada fase menyembuhan. Berda-

sarkan hal tersebut di atas kami perkirakan bahwa sPLA2 merupakan penyebab terjadi-

nya panas pada demam tifoid.

Penting diketahui bahwa walaupun konsentrasi pirogen di dalam sirkulasi rendah

pada fase akut demam tifoid, namun mRNA dari TNFa dan IL-Iß dapat ditemukan

pada sei daran pada sirkulasi {Bab 5). Hipotesa lain yang dapat dikemukakan adalah

bahwa sel-sel ini menghasilkan atau menginduksi sitokin pirogenik di susunan saraf

pusat.

Pada Bab 3 kami juga melaporkan terdapatnya regulasi yang bersifat menekan

kapasitas produksi sitokin proinflamasi pada fase akut demam tifoid. Hal ini dapat

dilihat pada hasil pengukuran kadar sitokin pada sel darah yang diberi stimulasi lipo-

polisakharida (LPS). Depresi produksi sitokin tersebut tidak disebabkan karena jumlah

sel darah putih yang rendah ataupun sebagai akibat kelelahan sei pada fase akut infeksi.

Untuk penelitian ini kami pergunakan darah total.

Setelah fase permulaan infeksi, terjadi "switch-off pada fase pascatranskripsi dari

produksi sitokin proinflamasi dan "switch-on" sitokin antiinflamasi pada sel-sel mono-

nuklear yang dapat dibuktikan dengan tidak tersupresinya sintesa IL-IRA.

Untuk mengatui lebih lanjut mekanisme penekanan produksi sitokin pada demam

tifoid dilakukan pengukuran mRNA dari IL-lß dan TNFa dalam darah. Pada Bab 5

dilaporkan bahwa berdasarkan pemeriksaan mRNA dari IL-lß dan TNFa penekanan

produksi sitokin pada demam tifoid fase akut terjadi fase pascatranskripsi dan bukan

disebabkan karena akibat "kelelahan" sel yang sitokin tersebut.

Regulasi yang bersifat menekan produksi sitokin in vivo merupakan fenomena

yang umum terjadi pada beberapa keadaan. Proces penekanan ini kemungkinan meru

pakan suatu upaya tubuh untuk menghambat terjadinya sitokinemia yang berlebihan

waluapun hai ini dapat pula melumpuhkan fungsi imunitas. Keuntungan dari proses

penekanan pada demam tifoid sulit dimengerti sebab pada infeksi Salmonella binatang

Ringkasan 151

coba sitokin proinflamasi diperlukan hospes untuk bertahan hidup {Bab 2). Penekanan

lebih kuat terhadap sitokin proinflamasi pada penderita demam tifoid dengan komplika-

si dapat merupakan gambaran akan adanya sistem kekebalan tubuh yang kurang efektif

sehingga dengan demikian pemberian khemoterapi diperlukan untuk memperkuat perta-

hanan tubuh.

Dalam upaya untuk mengetahui lebih lanjut pertahanan tubuh terhadap Salmonella

kami melakukan studi respon sitokin pada mencit {Bab 6) untuk dibandingkan dengan

respon pada manusia yang menderita demam tifoid (Bab 3, 4 dan 5). Tanpa adanya in-

feksi, sitokin pada sirkulasi dan juga pada kultur darah (ex-vivo) tidak terdeteksi walau-

pun dengan pemberian stimulasi LPS. Ketidak mampuan untuk memproduksi sitokin

pada mencit tersebut diatas berlawanan dengan manusia dimana pemberian stimulasi

LPS pada sei darah tepi menghasilkan sitokin proinflamasi dalam jumlah yang besar

walaupun tanpa infeksi. Kami juga menemukan bahwa pada mensit yang terinfeksi ter-

jadi stimulasi sitokin proinflamasi. Hai ini juga berlawanan dengan yang ditemukan

pada demam tifoid pada manusia {Bab 3).

Searah dengan penelitian lain pada laboratorium kami, timbul suatu pertanyaan

tentang kemungkinan terdapatnya pengaruh yang bersifat protektif dengan tindakan

pemberian rekombinan IL-1 pada mencit yang terinfeksi Salmonella typhimurium {Bab

6). Pemberian rhIL-la secara signifikan dapat menurunkan pertumbuhan mikroorga-

nisme pada hepar dan lien. Efek pemberian rhIL-la ргарегіакиап terlihat pada 24 jam

pertama setelah infeksi dan efek mikrobiologik dari dosis tunggal rhIL-la bertahan

selama eksperimen berlangsung. Sel-sel peritoneum secara signifikan memproduksi

lebih banyak "cell associated" IL-Ια pada 24 jam setelah pemberian rhIL-la.

Efek menguntungkan pada stadium dini dari pemberian rhIL-la ргарегіакиап

mungkin melalui beberapa mekanisme. Kenyataan bahwa pemberian IL-1 praperlakuan

dapat memberikan perlindungan terhadap berbagai macam infeksi merupakan fenome-

na umum. Kami perkirakan hai ini disebabkan oleh aktivasi makrofag.

Netrofil mempunya kemampuan untuk memakan dan membunuh Salmonella dan

berpengaruh terhadap perjalanan penyakit in vivo {Bab 2). Pada Bab 7 dijelaskan

tentang efek rekombinan "murine granulocyte colony-stimulating factor (rmG-CSF)"

yang dapat meningkatkan daya hidup mencit yang mengalami infeksi letal kuman

Salmonella. Temyata, tidak terdapat perbedaan pertumbuan kuman S. typhimurium

mencit kontrol dibandingkan mencit yang diberi praperlakuan rmG-CSF. Pada

eksperimen kami, 24 jam setelah pemberian rmG-CSF praperlakuan, produksi TNFa

152 Chapter 11

oleh makrofag peritoneum (dengan stimulasi LPS) lebih tinggi secara signifikan.

Pada mencit yang diberi rmG-CSF dengan pemeriksaan histologik pada hepar di-

dapatkan fokus-fokus inflamasi yang lebih terlokalisir. Hasil ini menunjukkan bahwa

pemberian rmG-CSF praperlakuan memberikan hasil yang menguntungkan pada mencit

yang terinfeksi kuman fakultatif intaseluler S. typhimurium.. Yang menjadi pertanyaan

adalah, rG-CSF mempunyai efek yang menguntungkan pada infeksi intraseluler melalui

rekrutmen netrofll pada fase dim ataukah sebaliknya justru merugikan karena pada fase

lanjut rG-CSF mempunyai kemampuan antiinflamasi sehingga dapat mengambat

aktivasi makrofag.

Meskipun Salmonella merupakan kuman berbentuk batang Gram-negatif, akan

tetapi masih terdapat kontroversi mengenai peran endotoksin pada patogénesis infeksi.

Kami berusaha memecahkan masalah ini (Bab 8) dengan melakukan percobaan pada

mencit "low density lipoprotein receptor knock-out (LDLR-/-)". Pada mencit jenis ini

lipoprotein yang terdapat pada sirkulasi dapat mengikat dan menetralisasi LPS kuman

sehingga dapat mengurangi induksi sitokin proinflamasi. Konsentrasi tinggi lipoprotein

dapat memberikan pengaruh yang baik pada mencit yang terinfeksi S. typhimurium.

Efek menguntungkan dari lipid yang mempunyai kemampuan untuk menetralisasi

endotoksin seperti halnya yang terjadi pada infeksi kuman ekstraseluler Klebsiella ter-

nyata bukan merupakan mekanisme yang mendasari terjadinya peningkatan daya hidup

pada infeksi S. typhimurium.. Pada penelitian ini terlihat bahwa pengaruh lipoprotein

adalah terhadap distribusi dan pertumbuhan fase lanjut dari kuman. Efek pada distribusi

kuman dihubungkan dengan kemampuan lipoprotein yang secara langsung dapat meng-

ganggu "uptake" dari Salmonella misalkan dengan сага menyelubungi struktur permuka

kuman dengan lipid. Hypotesa lain adalah bahwa reseptor LDL mempengarahi

"uptake" bakteri.

Terhadap pertumbuhan kuman, tidak kami dapatkan perbedaan bermakna pada

makrofag peritoneum dal am kemampuan fagositosis dan membunuh kuman antara

mencit dengan atau tanpa reseptor LDL. Hal ini menunjukkan bahwa perbedaan dalam

kecepatan pertumbuhan kuman disebabkan karena peran dari fagosit lain sel perito

neum.

Kesimpulan

Pada penderita demam tifoid kami dapatkan adanya regulasi yang menekan pro-

duksi sitokin proinflamasi yang juga merupakan fenomena umum pada suatu infeksi

Ringkasan 153

dan keadaan lainnya. Observasi ini mendukung pendapat yang menyatakan bahwa

imunoterapi merupakan hal yang menguntungkan pada infeksi Salmonella dan inter-

vensi yang dilakukan harus ditujukan untuk peningkatan sistim kekebalan tubuh.

Kami mempelajari imunomodulasi pada suatu eksperimen yang dilakukan pada

mencit yang terinfeksi S. typhimurium.. Pada penelitian ini didapatkan adanya pengaruh

yang menguntungkan dari pemberian rekombinan IL-1 praperlakuan terbukti dengan

berkurangnya pertumbuhan kuman pada organ-organ tubuh mencit. Dapat dikemukakan

suatu hipotesa bahwa hai ini terjadi sebagai akibat terdapatnya efek distribusi yang

diperankan oleh makrofag yang teraktivasi.

Pemberian nnG-CSF praperlakuanjuga memberikan pengaruh yang menguntung

kan pada daya hidup mencit terhadap S. typhimurium walaupun pertumbuhan kuman

pada organ tidak terpengaruh. Efek rmG-CSF pada rekrutmen dan aktivasi netrofil yang

terjadi pada saat awal merupakan mekanisme yang paling utama untuk daya hidup.

Dapat juga dilihat bahwa rmG-CSF dapat memacu peningkatan produksi TNFa oleh

makrofag peritoneum yang dipacu oleh LPS. Keadaan ini ikut mempengaruhi perjala-

nan penyakit.

Efek menguntungkan dari lipoprotein pada infeksi S. typhimurium ditemukan baik

pada distribusi kuman pada saat awal maupun pertumbuhan selanjutnya dari mikroorga-

nisme. Hai ini menunjukkan bahwa lipoprotein beserta reseptomya berperan penting

dalam pertahanan tubuh terhadap infeksi Salmonella baik pada distribusi awal maupun

dalam kemampuan "intracellular killing" fagosit.

Observasi infeksi Salmonella typhimurium yang dilakukan secara eksperimental

dapat memberikan aran pada perkembangan imunoterapi pada infeksi S. typhi pada ma-

nusia.

154 Chapter 11

Chapter 12

Samenvatting

156 Chapter 12

Samenvatting 157

Buiktyfus is een ernstige gegeneraliseerde infektie veroorzaakt door Salmonella

typhi. De ziekte wordt van mens op mens orofecaal overgedragen. Meestal zijn het

kinderen en jonge volwassenen, die buiktyfus krijgen. In Indonesie is de incidentie

hoog, met meer dan 1000 gevallen per 100 000 inwoners. De mortaliteit van buiktyfus

is in Indonesie door antibiotica gedaald van 18% tot 4%.

In 1989 werd gezamelijk onderzoek opgezet tussen de Katholieke Universiteit Nij

megen en de Diponegoro Universiteit in Semarang, Java, Indonesië. Het doel van deze

samenwerking was om klinische en immunologische factoren te onderzoeken bij buik-

tyfuspatienten. Daarbij was het ook de bedoeling de Pathogenese van Salmonella infek-

ties te bestuderen in een experimenteel diermodel.

In Hoofdstuk 2 wordt de Pathogenese van Salmonella infekties beschreven alsmede

de rol van cytokinen die bij deze infekties betrokken zijn. Omdat S. typhi niet patho-

geen is voor muizen, wordt de experimentele S. typhimurium infektie gebruikt om de

Pathogenese te bestuderen, aangezien deze infektie eenzelfde ziektebeeld als buiktyfus

geeft in muizen. Studies die gedaan zijn met dit muizenmodel worden in Hoofdstuk 2

gerefereerd. Bij extracellulaire Gram-negatieve infekties, waar hoge circulerende

concentraties van proinflammatoire cytokinen zoals interleukiene-l (IL-1) en tumor

necrosis factor (TNFot) worden beschouwd als de oorzaak van orgaanfalen en dood, is

remming van deze substanties gunstig. Daarentegen is bij intracellulaire infekties

remming van proinflammatoire cytokinen meestal nadelig. TNFa lijkt bij deze infekties

lokaal nodig (in kleine hoeveelheden) en toediening van recombinant TNFa voorkomt

uitgroei van microorganismen bij Salmonella infekties.

Wanneer men de symptomen, met name de koorts overziet bij buiktyfus, is het te

verwachten dat endogene pyrogenen in het bloed circuleren. Aangenomen wordt dat het

vrijkomen van Prostaglandine (PGE2) in de hersenen de uiteindelijke stap is bij het

onstaan van koorts en dat die stap gebeurt onder invloed van circulerende proinflam

matoire cytokinen. Deze zogenaamde endogene pyrogenen worden gemaakt op de

plaats van de infektie. In Hoofdstuk 3 rapporteren we dat er in de acute fase van buik

tyfus lage concentraties circulerende proinflammatoire cytokinen gevonden worden en

daarbij hoge concentraties antiinflammatoire niet-pyrogene stoffen zoals interleukin-1

receptor antagonist (IL-IRA) and oplosbare receptoren van tumor necrosis factor

(sTNFRI and sTNFRII). Omdat bij buiktyfuspatienten met koorts pyrogene cytokinen

niet gevonden werden, leek het dat andere mechanismen een rol zouden kunnen spelen.

Deze andere mechanismen zouden andere mediatoren kunnen zijn zoals het secretoire

158 Chapter 12

phospholipase A2 (sPLA2). Phospholipases zijn lipolytische enzymen die de afbraak

van phospholipiden katalyseren. Het vrijkomen van sPLA2 wordt geïnduceerd door IL-

1 en TNFa en het enzym кап de produktie van arachidonzuur en PGE2 aanzetten. Aan

getoond werd dat intraventriculaire injektie van sPLA2-remmers in hersenen van konij

nen, koorts kon tegengaan. sPLA2 zou dus mogelijk een rol spelen als circulerend pyro-

geen. In Hoofdstuk 4 vonden we inderdaad verhoogde concentraties sPLA2 in de acute

fase van buiktyfus. Deze waarden daalden naarmate de patiënten verbeterden. We be

schouwen het als mogelijk dat dit bioactieve sPLA2 verantwoordelijk zou zijn voor een

gedeelte van de symptomen van buiktyfus.

Hoewel circulerende pyrogene cytokinen niet gevonden werden in de acute fase

van buiktyfus, is het belangrijk te vermelden in het "koortsconcept" dat het transport

RNA voor TNFa en IL-Iß aanwezig bleef in circulerende witte bloedcellen. Dit werd

beschreven in Hoofdstuk 5. Een andere hypothese (dan die aangaande PLA2) voor

koorts is dat zulke circulerende cellen zelf de pyrogene cytokines maken, wanneer ze in

het centraal zenuwstelsel komen.

In Hoofdstuk 3 rapporteren we in de acute fase van buiktyfus ook down-regulatie

van de produktie van proinflammatoire cytokines door circulerende cellen na stimulatie

door lipopolysaccharide (LPS). Om dit te bestuderen gebruikten we een vol-bloed cyto

kine test. De gevonden verminderde cytokine productie was niet het gevolg van leuko-

penie of uitputting van deze cellen gedurende de acute fase van de infektie. Er blijkt een

posttrancriptionele uitschakeling van de produktie van proinflammatoire cytokines te

bestaan in de beginfase van de infektie, terwijl tegelijkertijd de anti-inflammatoire

respons in de mononucléaire cellen behouden blijft, want de produktie van IL-IRA was

niet verminderd op dat moment.

Om dit mechanisme van verminderde cytokine produktie of down-regulatie verder

te bestuderen in buiktyfus werd transportRNA voor IL-Iß en TNFa bepaald in vol

bloed. De bevinding dat transport RNA voor IL-Iß en TNFa niet verlaagd was in de

acute fase van buiktyfus, beschreven in Hoofdstuk 5, verwerpt ook uitgeputte produktie

als verklaring en onderstreept posttranscriptionele remming van cytokine produktie.

Deze in-vivo down-regulatie van cytokine produktie is sind enige tijd een bekend

fenomeen ook in andere omstandigheden. Het is zeer waarschijnlijk bedoeld als be

scherming tegen overweldigende cytokinenstromen, maar de verminderde produktie

kan ook afweerfunkties verslechteren. Het voordeel van down-regulatie in patiënten

met buiktyfus is niet goed te rijmen met de experimentele dierproeven met Salmonella

infekties waar proinflammatoire cytokinen nodig zijn om te overleven {Hoofdstuk 2).

Samenvatting 159

Het zou kunnen dat de nog lagere proinflammatoire cytokines bij gecompliceerde ge

vallen een afspiegeling zijn van een minder effektieve afweer, en dat immunothérapie

de gastheerweerstand hier zou kunnen verbeteren.

Om de afweer tegen Salmonella, infektie beter te begrijpen, bestudeerden we de

cytokine produktie in muizen (Hoofdstuk 6) en vergeleken deze met de gegevens uit de

voorgaande hoofdstukken van patiënten met buiktyfus (Hoofdstukken 3, 4 en 5). Circu

lerende cytokinen concentraties maar ook LPS-gestimuleerde ex-vivo produktie in vol

bloed waren niet detecteerbaar zonder infektie. Het onvermogen om cytokinen te

maken is het omgekeerde van de humane situatie, aangezien gestimuleerde humane

perifere bloedcellen grote hoeveelheden proinflammatoire cytokinen produceren zonder

infektie. Tijdens infektie vonden we meer proinflammatoire cytokinen produktie in dit

diermodel, wat ook in tegenstelling is tot wat we vonden bij patiënten met buiktyfus,

beschreven in Hoofdstuk 3.

Overeenkomstig andere studies van ons laboratorium bekeken we of toediening

van recombinant IL-1 (rhIL-la) een beschermend effekt had in een muizenmodel van

Salmonella typhimurium infekties (Hoofdstuk 6). RhJL-la toediening leidde tot een sig

nificant verminderde uitgroei van microorganismen in de organen van de muizen. Dit

effekt van rhIL-la voorbehandeling werd al 24 uur na het begin van de infektie zicht

baar, en het microbiologische effekt van deze eenmalige dosis van гшЪ-Іос bleef gedu

rende de gehele studie bestaan. Peritoneaalcellen van muizen, voorbehandeld met rhlL-

1 α produceerden significant meer cel-geassocieerd IL-la na 24 uur. Verschillende

mechanismes zouden ten grondslag kunnen liggen aan dit gunstige effekt van voorbe

handeling op een Salmonella infektie. Het feit dat IL-1 voorbehandeling beschermt

tegen zeer uiteenlopende infekties doet vermoeden dat het om een algemeen mechanis

me gaat. Wij postuleren dat dit macrofaagactivatie zou kunnen zijn.

Neutrofiele granulocyten zijn in staat Salmonella te fagocyteren en te doden. Van

deze cellen is gebleken dat ze belangrijk zijn voor het verloop van de infektie in vivo

(Hoofdstuk 2). In Hoofdstuk 7 beschrijven we dat recombinant murine granulocyte

colony-stimulating factor (rmG-CSF) de overleving verbeterd van muizen met een

potentieel lethale S. typhimurium infektie. De uitgroei van 5. typhimurium in de orga

nen van rmG-CSF voorbehandelde muizen was echter niet verschillend van die van

controle muizen. De gestimuleerde TNFct produktie van peritoneaal macrofagen van

met rmG-CSF voorbehandelde muizen was hoger dan die van controle muizen na 24

uur. Histologische beoordeling van de lever na infektie liet meer gelokaliseerde (tegen-

160 Chapter 12

over diffuse) infektiehaarden zien in de met rmG-CSF behandelde dieren. Voorbehan

deling met rmG-CSF is dus van voordeel bij muizen met een infektie met de facultatief

intracellulaire S. typhimurium.

Omdat we bij doorbehandelde dieren een trend vonden naar meer uitgroei van

Salmonella vroegen we ons af, of het gunstige effekt van rG-CSF op intracellulaire

infekties tot stand komt via (vroege) verhoogde neutrofielen activiteit en aantallen en of

later de antiinflammatoire eigenschappen die rG-CSF heeft, mogelijk door remming

van TNFa gemedieerde macrofaag activatie, nadelig zouden zijn.

Hoewel Salmonella een Gram-negatieve staaf is, is de rol van endotoxine in de

Pathogenese van door dit microorganisme veroorzaakte infekties controversieel. We

benaderden dit probleem {Hoofdstuk 8) door studie van low density lipoprotéine recep

tor knock-out (LDLR-/-) muizen, waarvan de circulerende lipoproteinen in staat zijn

om bacterieel LPS te binden en te neutraliseren, waarbij de aanmaak van proinflam

matoire cytokinen geremd wordt. De hoge concentraties lipoproteinen bleken het ver

loop van een S. typhimurium infektie gunstig te beïnvloeden. We laten zien dat het

gunstige effekt op de overleving na een S. typhimurium infektie niet tot stand komt via

neutralisatie van endotoxine, wat bij infektie met de extracellulaire Klebsiella het me

chanisme is.

We vonden effekten op distributie en uitgroei. Het eerste zou verklaard kunnen

worden door lipoproteinen die direkt de Salmonella opname verhinderen doordat ze

zich bijvoorbeeld hechten op oppervlaktestrakturen van de bacterie. Ook zou het moge

lijk zijn dat de receptor voor LDL de opname van Salmonella direkt beïnvloed.

Voor het latere effekt op uitgroei vonden we niet meteen een verklaring. Zowel de

fagocytose van Salmonella als ook de intracellulaire bacteriocidie door peritoneaal-

macrofagen was niet verschillend tussen muizen met of zonder LDL-receptor. Deze

resultaten suggereren dat het gevonden verschil in groeisnelheid verklaard moet worden

door een effekt op andere fagocyterende cellen dan de peritoneaalcellen, die wij

onderzochten.

Conclusies Wij vonden down-regulatie van de produktie van proinflammatoire cytokinen bij

patiënten met buiktyfus. Deze bevinding onderstreept het idee dat immunothérapie

gunstig kan zijn voor infekties met Salmonella typhi en dat zulke interventies gericht

moeten zijn op het suppleren of het doen toenemen van die immuunmodulatoren, die de

afweer tegen deze infektie verhogen.

Samenvatting 161

We bestudeerden immuunmodulatie door middel van de experimentele Salmonella

typhimurium infektie bij muizen. Voorbehandeling met recombinant IL-1 bleek gunstig

te werken op deze infektie met minder uitgroei van microorganismen in de organen. Als

hypothese werd verdedigd, dat dit door een distributie effekt komt, gemedieerd door

geactiveerde macrofagen.

Voorbehandeling met rmG-CSF vergrootte ook de overleving van muizen geduren

de een S. typhimurium infektie. De uitgroei van microorganismen daarentegen veran

derde niet. Het vroege effekt op neutrofiele granulocyten aantallen en activatie werd het

meest belangrijke mechanisme geacht voor de overleving. Voorbehandeling met rmG-

CSF verhoogde de LPS-gestimuleerde TNFa produktie door peritoneaal macrofagen op

het tijdstip van infektie. Dit kan bijgedragen hebben tot de betere overleving.

Lipoproteinen verminderden de distributie en de latere uitgroei tijdens een S. typhimu

rium infektie. De resultaten geven aan dat lipoproteinen en hun receptoren cruciale

factoren zijn in de afweer tegen Salmonella infekties zowel in de vroege distributiefase

als bij de intracellulaire fagocytaire bacteriocidie. Deze bevindingen in experimentele S.

typhimurium infekties bij muizen geven richting aan de ontwikkeling van immunothé

rapie voor Salmonella typhi infekties in patiënten.

162

Dankwoord

Pertama-tama saya berterima kasih untuk pasien-pasien, yang beri daran untuk

penelitian ini di Semarang. Kemudian Edi Dharmana dan M. Hussein Gasem teman-

teman saya dan semua perawat, Jujuk dan dari bagian mikrobiologi Bambang Isban-

drio, Hendro Wahyono dan Pak Untung: terimah kasih. Di rumah (dan restorán) Ninik

Jamari dan begitu di pak Supardio keramahan besar.

Mijn promotor Prof. Dr. J. W. M van der Meer. Beste Jos, toen je jaren geleden mij

aannam op het buiktyfusprojekt had je niet kunnen vermoeden dat het zo lang en 2 kin

deren verder, zou duren tot de afsluiting met dit proefschrift. Met je kenmerkende opti

misme bleef je ons drieën (Hussein, Edi en mij) stimuleren. We zullen zorgen dat de

andere 2 proefschriften snel volgen.

Prof. Dr. R. Djokomoeljanto. Matumuun Prof. untuk undangan dari Semarang

kerjasama di penelitian tentang demam tifus. Ken a tidak selalu gampang, tetapi teman-

teman di Semarang dan makanan Java bisa mengumpan saya berkali-kali.

Mijn co-promotor Dr. B.J. Kullberg. Beste Bart-Jan, jij hebt zeer veel werk verricht

aan dit proefschrift vanaf 1993. Van je zorgvuldigheid heb ik veel geleerd. Heel veel

dank voor alles.

Mijn co-promotor Dr. W. M.V. Dolmans. Beste Wil, jij bent de grote facilitator

van deze samenwerking met Indonesie geweest en nog. Dank je zeer.

Mihai Netea, dank voor je hulp en stimulerende gesprekken over de muizenstudies.

Van het Laboratorium Algemene Interne Geneeskunde (Hoofd Dr. P. N. M. De-

macker) wil ik met name danken Ineke Verschueren, Johanna van der Ven-Jongekrijg

en Liesbeth Seuren en natuurlijk Pierre. Ook Gerard Pesman en Marielle Spruijtenburg

van het Laboratorium Endocrinologie en Voortplanting dank ik. Jullie weten beter dan

ik dat zonder jullie werk dit proefschrift er niet zou zijn.

Van het Centraal Dierenlaboratorium (Hoofd Dr. J. Koopman) dank ik Jos

Hendriks, Yvette Brom, Theo van der Ing, Gerrie Grutters, Margo van de Brink en

Monique Bakker voor hun hulp bij de dierproeven.

Joep van Oosterhout en Arthur de Meyer dank ik voor de hulp bij de voorberei

dingen van het dierproevenonderzoek.

Tenslotte zou alles niet zo zijn gelopen als de kinderen niet vertroeteld werden

door mijn schoonouders, mijn moeder en door Angela Scholten-Peeters en Miep Don

kers- Awater (in Kekerdom) en door Lucia Tenten, Ine van Hest en Hermine Dekkers

(in Heeswijk). Voor de prettige sfeer in een schoon huis ben ik jullie zeer dankbaar.

163

Curriculum Vitae De schrijfster van dit proefschrift werd geboren op 13 augustus 1956 te Utrecht. In

1974 behaalde zij het einddiploma gymnasium-ß aan het Christelijk Gymnasium te

Utrecht. Hetzelfde jaar werd gestart met de studie geneeskunde aan de Rijksuniversiteit

Utrecht. Het doctoraal werd behaald in 1980 en het artsexamen op 26 juni 1981. In

augustus 1981 werd begonnen met de specialisatie tot internist in het Diaconessenhuis

in Arnhem (opleider Dr. C. van Gastel). Van 1983 tot 1984 deed zij onderzoek naar de

relatie tussen neuropeptiden en circulatie aan het Rudolf Magnus Instituut voor farma

cologie te Utrecht (Prof. Dr. W. de Jong, Prof. Dr. D. de Wied). Vanaf 1984 werd de

opleiding tot internist vervolgd in het Onze Lieve Vrouwe Gasthuis te Amsterdam (op

leider Dr. K.J .Roosendaal, Dr. J. Silberbusch). Na de afsluiting met tropische genees

kunde in het Academisch Medisch Centrum (AMC), Amsterdam (Prof. Dr. A. de

Geus), werd zij op 1 januari 1987 geregistreerd als internist.

Van 1987 tot 1989 werkte zij als internist in Mumias, Kakamega Diocese, Western

Province Kenya en deed malaria-onderzoek in samenwerking met het AMC, het Kenya

Medical Research Institute (Dr. A.D. Brandling-Bennett en J. Were) en Wellcome Trust

(Dr. W. Watkins).

Na terugkeer in Nederland werkte zij 1 jaar op de afdeling Intensieve Zorg,

Radboud Ziekenhuis, Nijmegen (Hoofd: Dr. J. Gimbrere). Vanaf 1990 is zij gedeeltelijk

verbonden als universitair docent aan de facultaire werkgroep Gezondheidszorg

Ontwikkelingslanden (Thans: Nijmegen Institute of International Health; Hoofd: Dr.

W.M.V. Dolmans) en aan de afdeling Algemeen Interne Geneeskunde (Hoofd: Prof.

Dr. J.W.M, van der Meer) van de Katholieke Universiteit Nijmegen. In 1998 werd zij

geregistreerd als intemist-infectioloog (Opleider Dr. B.J. Kullberg).

Zij is gehuwd met Albert van Linge. Zij hebben twee dochters, Christine en Vita.

Stellingen

1. Down-regulatie van proinflammatoire cytokinenproduktie wordt gezien in de acute fase van

buiktyfus, maar dit fenomeen is een gebruikelijke reaktie bij infekties of acute stress zoals

operaties of excessieve inspanning.

2. Recombinant interleukine-1 verbetert de uitkomst van Salmonella infektie bij muizen

waarschijnlijk door in een vroege fase de distributie van bacteriën te beïnvloeden.

3. Het gunstige effekt van lipoproteinen op een infektie met Salmonella typhimurium bij

muizen berust op beïnvloeding van distributie en uitgroei van micro-organismen.

4. Recombinant granulocyte colony-stimulating factor (rG-CSF) heeft een gunstig effekt op de

overleving van muizen met een Salmonella infektie en bevordert de opname van chinolonen in

neutrofiele granulocytes Daarom zou de combinatiebehandeling van G-CSF en deze antibiotica

patiënten met buiktyfus van nut kunnen zijn.

Dit proefschrift en Andresen J, ed. Abstracts G-CSF Research Meeting. Santa Monica, CA: Amgen,

1996;36-38.

5. Het grote struikelblok in de ontwikkeling van IL-12 als immunothérapie is dat het mogelijk

werkt op ziekten die slechts in arme zuidelijke landen veelvuldig voorkomen.

Hail SS. IL-12 at the crossroads. Science 1995; 268:1432-1434.

6. Elk ziekenhuis zou willen ruilen met de accomodatie van ziektekostenverzekeraars.

7. Het verlies van stemmen van D66 illustreert het verschil in respect dat men heeft voor oudere

vrouwen tegenover dat voor oudere mannen.

8. Dat Indonesië het land met de hoogste incidentie van buiktyfus is, is pas afgelopen jaar voor

ieder aanschouwelijk gemaakt. (TU<j>oç=beneveling / тл)фоі=гоок maken)

9. Een сорго-motor is een milieuvriendelijke aandrijver van de (meestal juist papierverslin-

dende) promovendus.

10. De meeste vrouwen willen meer sexegenoten op hogere posities, mits zij het niet zelf

hoeven te zijn.

Nijmegen 26 mei 1998 Monique Keuter

с

G,

• Mediator-s in -typhoid fever · - Radboud Repository

Documents

Transcript of • Mediator-s in -typhoid fever · - Radboud Repository