Rights / License: Research Collection In Copyright - …41789/... · Peripheral mediatory mechanisms of behaviorally conditioned immunosuppression ... behaviorally conditioned immunosuppression

Research Collection

Doctoral Thesis

Peripheral mediatory mechanisms of behaviorally conditionedimmunosuppression by cyclosporin A

Author(s): Riether, Carsten

Publication Date: 2008

Permanent Link: https://doi.org/10.3929/ethz-a-005804624

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For moreinformation please consult the Terms of use.

ETH Library

https://doi.org/10.3929/ethz-a-005804624

http://rightsstatements.org/page/InC-NC/1.0/

https://www.research-collection.ethz.ch

https://www.research-collection.ethz.ch/terms-of-use

DISS.ETH NO: 18085

Peripheral mediatory mechanisms of behaviorally

conditioned immunosuppression by cyclosporin A

A dissertation submitted to the

SWISS FEDERAL INSTITUTE OF TECHNOLOGY ZURICH

for the degree of

Doctor of Sciences

presented by

CARSTEN RIETHER

Master of Science.; École Superiéure de Biotechnologie de Strasbourg (ESBS)

Born May 4th, 1980

citizen of Germany

accepted on the recommendation of:

Prof. Dr. Joram Feldon, examiner

Prof. Dr. Manfred Schedlowski, co-examiner

Prof. Dr. Rainer H. Straub, co-examiner

2008

Acknowledgments

The research of this dissertation was conducted at the Laboratory of Psychology

and Behavioural Immunobiology, Swiss Federal Institute of Technology (ETH)

Zurich, the Laboratory of Behavioral Neurobiology, ETH Zurich, Switzerland and

the Laboratory of Psychoneuroimmunology, University Medical Center, Utrecht,

the Netherlands. This research was financially supported by the ETH Zurich. The

work also strongly benefited from the excellent research facilities at the

Laboratory of Psychology and Behavioural Immunobiology, ETH Zurich,

Switzerland and the Laboratory of Psychoneuroimmunology, University Medical

Center, Utrecht, the Netherlands.

I would like to take the opportunity to thank the members of the dissertation

committee:

First and foremost, my deepest thanks go to Professor Manfred

Schedlowski for giving me the opportunity to conduct the research for my

dissertation in his laboratory. I am very thankful for the daily supervision,

guidance and support under awkward circumstances, as well as for his trust in me

throughout these years. He provided me with a solid foundation which will

enable me to pursue my scientific goals.

I would also like to express my thanks to Professor R. H. Straub from the

University Hospital of Regensburg, Germany, for taking the time to read my thesis

and for agreeing to be a member of my exam committee.

I am extremely grateful to Professor Joram Feldon for sharing his

neurobiological expertise with me, for supervising the neurobiological projects

included in my thesis and for taking over the responsibility as first examiner of

the present thesis as well as for providing animal husbandry and care in his

facilities.

My thanks also go to all the present and past members of the Laboratory of

Psychology and Behavioural Immunobiology, ETH Zurich:

I am indebted to Dr. Harald Engler, Dr. Gustavo Pacheco-Lopez, Dr. Maj-

Britt Niemi, Dr. Andrea Engler and Raphaël Doenlen. In the last three years they

have probably got to know me better than I know myself. Their contribution with

different kinds of expertise was indispensable for the realization of the present

thesis. I would also like to express my sincere thanks to them for running and

assisting the immunological and neurobiological assays, for the long and for

constantly providing new motivation and a feeling of well-being. The project

would never have achieved such scope without their help.

I would also like to express my sincere thanks to Professor C. Heijnen and

Dr. Annemieke Kavellars for planning and assisting the immunological assays in

their laboratory in Utrecht. Their contribution, especially to the immunological

studies, was indispensable for the realization of the present thesis. I would also

like to acknowledge Hanneke Willemen’s excellent assistance with the Western

Blots and the immuno-precipitation.

In addition, I would like to thank Dr. Urs Meyer, Dr. Irene Knüsel, Severin

Schwendener, and Mary Muhia for joining me in many interesting discussions

about the exciting world of neuro-psycho-endocrino-immuno-biological functions

and for being good buddies throughout the past few years.

A special “thank you” is due to Dr. Dirk Hanebuth for his constant supply of

food, music, energy, motivation and other help.

My thanks and appreciation also go to the Animal Services Department

Schwerzenbach, Jeanne Michel-von Arx, Pascal Guela, Ruedi Blersch, and Dana

Ryser-Stokes for their excellent animal husbandry and care. I also remain indebted

to Rachel Matthey for her editorial assistance in the preparation of this

dissertation.

Finally, a big “thank you” to my parents Rita and Harald Riether and my

brother Sascha Riether, who supported me all along the way and to whom I

dedicate this work. „Ihr ward immer da als ich euch brauchte und habt mich

unglaublich motiviert und mir den Rücken gestärkt. Ich danke euch dafür von

ganzem Herzen“.

“Man kann einem Menschen nichts lehren. Man kann ihm nur helfen, es in sich

selbst zu entdecken.“ (Galileo Galilei)

Carsten Riether

Contents

Contents

List of abbreviations..........................................................................................................................7

Summary............................................................................................................................................ 9

Zusammenfassung...........................................................................................................................11

Chapter 1............................................................................................................................................ 13

General Introduction....................................................................................................................... 13

Bi-directional communication between the central nervous and the immune system.14

Behavorial conditioning of immune functions ................................................................................. 15

Theoretical framework of behavioral conditioning ................................................................... 15

Neural mechanisms underlying behavioural conditioning....................................................18

Peripheral mechanisms of saccharin-cyclosporin A conditioning......................................20

Objectives and outline of the thesis ...................................................................................................... 21

References .........................................................................................................................................................23

Chapter 2.......................................................................................................................................... 26

Cyclosporin A as unconditioned stimulus is detected by the central nervous system at

acquisition time.............................................................................................................................. 26

Materials and Methods ..............................................................................................................................29

Results................................................................................................................................................................. 31

Discussion..........................................................................................................................................................32

References ........................................................................................................................................................ 34

Chapter 3. ......................................................................................................................................... 38

Memorized T lymphocyte response ............................................................................................ 38

Materials and Methods ..............................................................................................................................40

Results................................................................................................................................................................44

Discussion.........................................................................................................................................................50

References .........................................................................................................................................................53

Chapter 4...........................................................................................................................................55

ββββ-adreneroceptor stimulation inhibits calcineurin activity in T lymphocytes via a PKA-

dependent mechanism

............................................................................................................................................................55

Introduction..................................................................................................................................................... 56

Materials and Methods ...............................................................................................................................57

Results................................................................................................................................................................62

Discussion.......................................................................................................................................................... 71

Contents

References .........................................................................................................................................................73

Chapter 5. ..........................................................................................................................................75

General discussion...........................................................................................................................75

Potential mechanisms of conditioned immunosuppression by CsA...................................... 78

Potential use of behavioural conditioning as a supplemental therapy in patients ......... 82

Animal models with clinical relevance ........................................................................................... 82

Behavioural conditioning of immune responses in humans................................................ 83

Concluding remarks ..................................................................................................................................... 87

References ........................................................................................................................................................88

Curriculum Vitae............................................................................................................................. 92

List of publications ......................................................................................................................... 93

List of abbreviations

7


6-OHDA 6-hyroxydopamine

ACTH adrenocorticotropic hormone

AKAP A-kinase anchoring protein

Am amygdala

ANOVA analysis of variance

AR adrenergic receptor

BBB blood-brain barrier

cAMP cyclic adenosine monophosphate

CaN calcineurin

CHS contact hypersensitivity

CIS behaviorally conditioned immunosuppression by

CsA

CNS central nervous system

CR conditioned response

CRH corticotropin-releasing hormone

CS conditioned stimulus

CsA cyclosporin A

CTA conditioned taste avoidance

CY cyclophosphamide

D1/D2R dopamine D1/D2 receptor

DA dopamine

DBcAMP dibuturyl-cyclic adenosine mono phosphate

FACS fluorescence activated cell sorting

GPCR G-protein coupled receptor

HPA hypothalamius-pituitary-adrenal axis

i.p. intraperitoneal

IC insular cortex

IFN interferon

IL interleukin

LHA lateral hypothalamic area

LPS lipopolysaccharide


8

NFAT nuclear factor of activated T-cells

NMDA N-methyl-D-aspartate

PBS phosphate buffered saline

PKA phosphokinase A

PKC phosphokinase C

PLC phospholipase C

RNA ribonucleic acid

SAL saline

SAC saccharin

SNS sympathetic nervous system

TNF tumor necrosis factor

US unconditioned stimulus

VMH ventromedial nucleus of the hypothamulus

Summary

9

Summary

The experiments in this research project are based on data demonstrating the

intense communication between the central nervous system (CNS) and the immune

system when applying the paradigm of conditioned immunosuppression (CIS) by the

immunosuppressive agent cyclosporin A (CsA). Even though progress has been made in

elucidating the central mechanisms of behavioural conditioning during the last two

decades, little is known about the peripheral mechanisms behind this phenomenon. CIS

was shown to be mediated via the splenic nerve and the sympathetic nervous system

(SNS) predominately via β-adrenergic mechanisms. However, the peripheral mediatory

mechanisms of CIS are largely unknown to date.

The research described in the present thesis aimed at identifying and evaluating

distinct fundamental peripheral mechanisms of CIS with particular focus on the cellular

and intracellular pathways, in order to provide a clinical framework for CIS and to further

contribute to a better understanding of the principles of behavioural interaction in

individuals with different neurological and immune histories. To this end, a rat model of

CIS was employed to explore (i) the detection of the unconditioned stimulus (US), i.e. CsA

at acquisition time by the CNS; (ii) the issue of peripheral specificity of CIS; and (iii)

potential cellular and intracellular mechanisms in T lymphocytes mediating CIS.

The first series of experiments revealed that after injection of the

immunosuppressive drug CsA at acquisition time and subsequent reduction in pro-

inflammatory cytokine expression in the spleen 120min after injection, IL-1β mRNA was

concomitantly synthesized de novo in the amygdala (Am; Chapter 2). This suggests that a

more physiological immuno-sensory process is implicated, and that the key brain

structures involved in the acquisition of CIS not only detect peripheral cytokine increases

but also “sense” a reduction in cytokine levels in the periphery, as occurred after the

administration of CsA.

The next series of experiments shows the peripheral specificity and identifies

calcineurin (CaN) as the target molecule of CIS (Chapter 3). Analysis of T lymphocyte

proliferation and Th1-cytokine production after conditioning identified T lymphocytes as

the target cell population of CIS. In addition, CaN activity in splenic homogenates was

reduced in conditioned animals. Importantly, the behavioral conditioned effects on CaN

activity, cytokine production and cell proliferation were comparable in magnitude and

direction to the effects elicited by administration of CsA alone, demonstrating the

peripheral specificity of CIS. Therefore, these data clearly illustrate that the conditioned

Summary

10

response is specifically mediated on a cellular level via the inhibition of CaN and identify

for the first time CaN as the target molecule of CIS.

It has also been demonstrated that CIS suppresses T lymphocyte proliferation and

Th1-cytokine secretion and is mediated via the splenic nerve and β-adrenergic

mechanisms (Exton, Gierse et al. 2002; Xie, Frede et al. 2002; Pacheco-Lopez, Riether et al.

2009). Since CaN activity was shown to be reduced in spleens of conditioned rats

(Pacheco-Lopez, Riether et al. 2009), we modelled the β-adrenergic-T lymphocyte

interaction ex vivo in order to investigate the efferent intracellular mechanism mediating

conditioned CIS (Chapter 4). The current data demonstrate that CaN activity in enriched T

lymphocytes is transiently reduced after specific β-adrenergic agonist stimulation with

terbutaline, with these effects being mediated via protein kinase A (PKA) and most likely

the endogenous CaN inhibitor, A-kinase anchoring protein-150 (AKAP-150; (Liu 2003)).

These suppressive effects were antagonized by the β-adrenergic antagonist nadolol.

These data clearly suggest that CIS is mediated in the periphery exclusively via β-

adrenergic activation of T lymphocytes and subsequent inhibition of CaN.

Together, the experimental investigations presented in this thesis are clear

illustrations of (i) how the US is detected by the CNS at acquisition time and (ii) how CIS is

mediated in the periphery on an intracellular level. The results thus support the

peripheral specificity of CIS and identify for the first time an intracellular mechanism

mediating the conditioned response, thus providing a clinical framework for the better

understanding of neuro-immune interactions. This lends support to the relevance and

feasibility of employing behavioural conditioning as a to supplement to standard

pharmacological regimens in animals as well as in humans, and it highlights the validity

of the model in psychoneuroimmunological research for investigating bi-directional

brain-immune communication.

Zusammenfassung

11

Zusammenfassung

Die in diesem Forschungsprojekt aufgeführten Experimente basieren auf der

intensiven Kommunikation zwischen dem zentralen Nervensystem (ZNS) und dem

peripheren Immunsystem. Afferente und efferente Verbindungswege zwischen dem ZNS

und dem Immunsystem bilden die neuroanatomischen und die biochemischen

Grundlagen für das Phänomen der Klassischen Konditionierung von Immunfunktionen.

Anhand eines Tiermodells zur konditionierten Immunsuppression, bei dem das

immunsuppressive Medikament Cyclosporin A (CsA) als unkonditionierter Stimulus (US)

und saccharinhaltige Trinklösung als konditionierter Stimulus (CS) eingesetzt werden,

konnte in früheren Studien gezeigt werden, dass die konditionierte Immunsuppression

auf dem efferenten Weg über den Milznerv unter Beteiligung des Neurotransmitters

Noradrenalin und ß-adrenerger Rezeptoren an immunkompetente Zellen übermittelt

wird. Bisher war jedoch völlig unklar, wie das ZNS auf dem afferenten Weg die CsA-

induzierte Immunsuppression wahrnimmt. Ebenso war auf dem efferenten Weg bisher

ungeklärt, wie spezifisch der konditionierte suppressive Effekt auf die Immunantwort

ausfällt und welche zellulären Mechanismen daran beteiligt sind.

Die ersten Experimente dieser Dissertation konzentrierten sich daher auf den

afferenten Informationsweg, vom peripheren Immunsystem zum Gehirn. Die Ergebnisse

zeigen, dass es 120 Minuten nach intraperitonealer Injektion von CsA einerseits zu einer

verminderten mRNA-Expression proinflammatorischer Zytokine in der Milz, anderseits

zur de novo Synthese von IL-1β mRNA in der Amygdala (Am) kommt (Kapitel 2). Dies

deutet darauf hin, dass das ZNS nicht nur in der Lage ist einen Anstieg, sondern auch eine

durch CsA-induzierte Verminderung peripherer Zytokine wahrzunehmen. Welche

afferenten Kommunikationswege an der Übermittlung des Signals von der Peripherie

zum Gehirn beteiligt sind, muss in zukünftigen Studien geklärt werden.

In einer zweiten Reihe von Experimenten wurde die Spezifität der konditionierten

Immunsuppression im direkten Vergleich mit der CsA-induzierten Immunsuppression

analysiert (Kapitel 3). Die Messung von T-Zellproliferation und Th1-Zytokinproduktion

nach der Konditionierung ergaben, dass T-Lymphozyten die hauptsächlichen Zielzellen

der konditionierten Immunsuppression sind. Des weiteren zeigte sich, dass die Aktivität

der Protein-Phophatase Calcineurin (CaN) in Milzhomogenisaten nicht nur durch das

Medikament CsA, welches seine immunsuppressiven Eigenschaften durch Inhibition der

Aktivität von CaN induziert, vermindert war, sondern im gleichen Maße auch durch die

Konditionierung inhibiert war. Diese experimentellen Befunde dokumentieren zum

ersten Mal, dass CaN ein wichtiges Zielmolekül bei der konditionierten

Zusammenfassung

12

Immunsuppression ist.

Um die intrazellulären Mechanismen der konditionierten CaN-Inhibition

aufzuklären, wurde im nächsten Schritt ein ex vivo Modell etabliert, mit dem die Rolle β-

adrenerger Rezeptorstimulation und deren mögliche Auswirkungen auf die CaN-

Inhibition in T-Zellen genauer untersucht werden konnten. Die Analysen mit diesem

Modell zeigen, dass die Aktivität von CaN in isolierten T-Lymphozyten nach Stimulation

mit dem β-adrenergen Agonisten Terbutalin vorübergehend vermindert ist. Dieser Effekt

wird durch die Proteinkinase A (PKA) und den endogenen CaN-Inhibitor, A-Kinase-

Anchoring Protein (AKAP)-150, herbeigeführt. Die terbutalininduzierte Suppression der

CaN-Aktivität, der T-Zellproliferation und der Zytokinproduktion konnten durch den

selektiven β-adrenergen Antagonisten Nadolol aufgehoben werden. Diese Ergebnisse

deuten daraufhin, dass die mit dem oben beschrieben Modell induzierte, konditionierte

T-Zellsuppression in der Milz über die Aktivierung β-adrenerger Aktivierung von

Rezeptoren auf T-Lymphozyten und die daraus resultierende Hemmung von CaN

vermittelt wird.

Zusammengefaßt zeigen die in dieser Dissertation beschriebenen

Untersuchungen zum einen, wie der unkonditionierte Stimulus CsA vom ZNS während

der Assoziationsphase wahrgenommen wird und, zum anderen, wie die konditionierte

Immunsuppression intrazellulär in peripheren T-Lymphozyten vermittelt wird. Die

Resultate demonstrieren insbesondere die Spezifität der konditionierten Reaktion im

Immunsystem und liefern damit einen wichtigen Baustein zu einem besseren

Verständnis der funktionellen Interaktion zwischen dem Gehirn und dem peripheren

Immunsystem. Die tierexperimentellen Befunde unterstreichen zudem die potentielle

Relevanz solcher Konditionierungsprotokolle in klinischen Situationen bei Patienten, bei

denen eine Unterdrückung der Immunantwort notwendig ist. Ziel wird es hier sein, durch

den Einsatz solcher Verhaltensprotokolle die Menge an Medikamenten und damit

unerwünschte pharmakologische Nebenwirkungen einzusparen bei einer gleichzeitigen

Maximierung der therapeutischen Effekte.

Chapter 1: General introduction

13

Chapter 1.

General Introduction


14

Bi-directional communication between the central nervous and the immune system

The study of the communication and interaction between the central nervous

system (CNS) and the immune system has developed over the last three decades into an

extensive interdisciplinary field of research termed psychoneuroimmunology (Blalock and

Smith 2007).

An elegant model to investigate the interactions between these systems is the

behavioural conditioning paradigm. In these experiments the administration of an

immunomodulating drug or substance, the unconditioned stimulus (US), is paired with a

neutral stimulus, typically a taste or odour, the conditioned stimulus. After one or several

contingent pairings of the CS with the US during the acquisition phase, re-exposure to

the CS during the evocation phase induces changes in the peripheral immune response,

formerly just elicited by the drug or substance, i.e. the US (Ader 2003). This phenomenon

of neuo-immune associative learning phenomenon is based on the intensive interaction

between the brain and the immune system, which has been documented particularly

during the past two decades (Watkins and Maier 2000; Tracey 2002; Besedovsky and del

Rey 2007; Nance and Sanders 2007; Quan and Banks 2007; Ziemssen and Kern 2007).

Experimental evidence demonstrates that the brain communicates with the immune

system on the efferent arm via direct innervation of primary and secondary lymphoid

organs, such as the thymus and the spleen (Felten, Felten et al. 1985; Elenkov, Wilder et al.

2000; Nance and Sanders 2007; Quan and Banks 2007), and/or via humoral pathways

comprising activation of the hypothalamus-pituitary-adrenal (HPA) axis (Besedovsky and

del Rey 1996). HPA axis activation induces adrenocorticotropic hormone (ACTH) secretion

from the adrenal cortex via corticotropin-releasing hormone (CRH) which results in

elevated cortisol plasma levels. Leukocytes bear intracellular and extracellular receptors

specifically for hormones, neurotransmitters and neuropeptides (Sanders and Straub

2002). Therefore, alterations in plasma cortisol levels can induce tissue-specific changes

in receptor expression of immune cells resulting in impaired cytokine production and

gene expression. In parallel, the CNS impacts immune function via peripheral neural

pathways like the sympathetic nervous system (SNS). The SNS innervates secondary

lymphoid organs, such as the spleen and lymph nodes predominately via noradrenergic

nerve fibres (Felten, Felten et al. 1985; Nance and Sanders 2007; Quan and Banks 2007),

affecting circulation and activity of adrenoceptor-expressing lymphocytes (Elenkov,

Wilder et al. 2000; Nance and Sanders 2007).

In turn, the peripheral immune system gives feedback to the brain on the

interoceptive immune status through the afferent arm via neural and/or humoral


15

afferent pathways. The neural pathway includes, e.g. cytokine stimulation of the vagus

nerve, while the humoral pathway depends on peripheral cytokines crossing the blood-

brain barrier (BBB) via active or passive transport mechanisms (Gaillard 1998; Banks, Farr

et al. 2001; Quan and Banks 2007). Neurons express receptors for pro-inflammatory

cytokines (Diana, Van Dam et al. 1999; Morikawa, Tohya et al. 2000), T cell cytokines

(Neumann, Schmidt et al. 1997; Wang, Lu et al. 2001) and chemokines (Horuk, Martin et al.

1997). For example, the pro-inflammatory cytokine IL-1β, for instance, activates the vagus

nerve via receptors on sensory neurons (Goehler, Gaykema et al. 1998). These alterations

in vagal activity are transmitted via the nucleus of the solitary tract (NTS) to the

hippocampus and the hypothalamic nuclei via the nucleus of the solitary tract, resulting

in up-regulated IL-1β gene expression in microglia cells (Dantzer 2004). Moreover, IL-1β is

capable of directly or indirectly crossing the BBB (Banks, Farr et al. 2001).

The complex bi-directional network illustrated in Figure 1.1 shows that the brain is

capable of detecting signals released by an activated immune system, e.g. during

acquisition time, and vice versa. One of the major issues for future research activities will

be to elucidate the hierarchical, temporal and spatial communication patterns linking the

brain and the peripheral immune system under normal conditions, and to understand in

more detail how, when and where this interaction is disturbed under the different

pathological conditions.

Behavorial conditioning of immune functions

Theoretical framework of behavioral conditioning

Classical or Pavlovian conditioning is often described as the transfer of the

response-eliciting property of a biologically significant stimulus (US) to another stimulus

without that property (Pavlov 1927; Carew and Sahley 1986; Domjan 2005). This transfer

is thought to occur only if the CS serves as a predictor of the US (Rescorla and Wagner

1972; Pearce 1987; Rescorla 1988). The behavioural conditioning of immune functions

basically follows the same rules as classical or Pavlovian conditioning. Two basic steps

compose the conditioning protocol: an acquisition phase in which one or more CS-US

pairings occur inducing an associative learning process, and an evocation phase where

the memory of such an association is retrieved after exposure to the CS. In order to

successfully acquire a typical Pavlovian conditioned response (CR; e.g. salivation, eye

blink) several CS-US pairings occurring in a contingent manner are necessary. A backward

association scheme where the US precedes the CS will not induce an associative learning

process. The features of CS-US timing and predictability regarding classical conditioning

have been reviewed elsewhere (Rescorla 1988). The rest interval between acquisition and


16

evocation is another key factor explaining the magnitude of the CR: typically, the longer

the rest interval, the weaker the CR (passive forgetting), since retention of a given

engram may deteriorate over time. Moreover, extinction of the CR is a classical feature of

behavioural conditioning.

Figure 1.1: Theoretical framework for behavioural conditioning At acquisition time, there are two possible unconditioned stimuli (US) associated with a conditioned stimulus. The US that is directly detected by the central nervous system (CNS) is defined as a “genuine US”, whereas the one that needs one or more intermediary molecules to be released by another system before it can be detected by the CNS is called a “sham US”. For any US, directly or indirectly perceived, there are two possible afferent pathways to the CNS: the neural afferent pathway and the humoral afferent pathway. At evocation time there are two possible pathways by which the CNS can modulate immune functions: the humoral efferent pathway and the neural efferent pathway. The humoral efferent pathway may imply changes in neurohormones that directly or indirectly modify the immune response. The neural efferent pathway is supported by the direct innervation of primary and secondary lymphoid organs (Riether, Doenlen et al. 2008).

This phenomenon occurs at evocation time if the CS is repeatedly presented without the

US (active forgetting), reducing the magnitude of the CR and finally extinguishing it.

Additionally, the CR and the response elicited by the US (unconditioned response, UR), are

neither necessarily of a similar magnitude nor of the same direction, and the kinetics of

each response may also differ (Rescorla 1988). In the terminology of behavioural

conditioning, both the CS (i.e., changes in the external environment) and the US (i.e.


17

changes in the internal environment) must be inputs to the CNS. Thus, in a behavioural

conditioning protocol, only a change in the immune system that is sensed by the CNS at

acquisition time can serve as a US (Pacheco-Lopez, Niemi et al. 2007).

When Ivan P. Pavlov studied the conditioned salivary response in dogs, it was

sufficient to note that before conditioning a dog would not salivate after the sound of a

bell, but that after conditioning the same dog would do so. The emphasis was on

observing conditioning in the individual animal, and it was noted that some animals

were more easily conditioned than others (Pavlov 1927).

The immune system itself is to a certain degree independent of neural activity.

This implies that the experimenter not only has to control for sensitization and

habituation but also for immune memory and tolerance. In addition, behavioural

conditioning protocols involving taste/odour-visceral associations may require more

consideration than standard conditioning protocols. The fact that flavour trace memory

lasts for a number of hours should be taken into account in the experimental design of

behavioural conditioning protocols employing this kind of CS (Garcia, Ervin et al. 1966;

Schafe, Sollars et al. 1995).

Considering all these particularities with regard to the behavioural conditioning of

immune functions, Ader and Cohen provided the following guidelines for the experi-

mental design of such protocols (Ader 2001; Ader 2003):

Conditioned group: The primary experimental group that is conditioned by pairing CS and

US, and at evocation time is exposed to the CS.

Conditioned not evoked group: At association time this group is treated identically to the

conditioned group, but is not exposed to the CS in the evocation phase. This group serves

as a control for any direct or indirect immunomodulating effects of the association

procedure per se, as well as for possible residual effects of the US in the evocation phase

upon re-exposure to the CS.

Unconditioned response group: At association time this group is exposed to the US and CS

at the same times. At evocation time it is exposed to the US only, in the absence of the

CS. This control group defines the magnitude and direction of the UR.

Non-Conditioned but evoked group: At association time this group is exposed to the CS

and US as many times as the conditioned group, but in a non-contingent manner. At

evocation time, the subjects in this group are exposed to the CS. This group is mainly

intended to control for non-associative factors, and also certifies the immunological

neutrality of the CS during the association and evocation phases.

Placebo group: In the association and evocation phases this group is exposed to the CS

which is paired with an immunological neutral stimulus (e.g. sterile saline or phosphate


18

buffered saline) as a placebo. This group controls for residual effects of the CS and

possible artefacts due to the procedure (e.g. handling, injection etc.).

Neural mechanisms underlying behavioural conditioning

Most of the recent studies on behavioural conditioning of immune functions

focussed on immunosuppressive alterations in the immune system (Ader and Cohen

1975). However, after the replication and acceptance of the conditionability of immuno-

pharmacological responses, studies on the interaction between the immune system and

the CNS were broadened resulting in the exploration of both directions in the same

paradigm, i.e. conditioned immunosuppression as well as conditioned immuno-

enhancement (Pacheco-Lopez, Niemi et al. 2007). The following section will provide an

overview of the neural mechanisms involved in behavioural conditioning of immune

functions in both directions.

The phenomenon of associating a flavour (food/drink) with possible immune

consequences has been experimentally appraised in rodents and humans by exploring

the conditioned taste aversion/avoidance (CTA) paradigm (Garcia, Kimeldorf et al. 1955).

Applying this type of associative learning, subjects associate a flavour with a postprandial

malaise (Bermúdez-Rattoni 2004). One discrete neural network involved in taste-visceral

associative learning has already been reported (Sewards and Sewards 2002; Sewards

2004). This neural circuit consists of sensory and hedonic neural pathways, including the

NTS, the parabrachial nucleus, the medial thalamus, the amygdala (Am) and the insular

cortex (IC) (Yamamoto, Shimura et al. 1994). Regarding the central processing of the

gustatory CS and visceral US, cholinergic neurotransmission in the IC and Am seems to be

essential during the initial stages of taste memory formation (Bermudez-Rattoni,

Ramirez-Lugo et al. 2004). Acetylcholine appears to codify the novelty of both the

conditioned and the unconditioned stimulus (Acquas, Wilson et al. 1996; Miranda,

Ferreira et al. 2002). In particular, the IC is essential for the acquisition and retention of

this kind of associative learning process (Bermudez-Rattoni and McGaugh 1991; Cubero,

Thiele et al. 1999). Moreover, it has been postulated that the IC may integrate gustatory

and visceral stimuli (Sewards and Sewards 2001). The preponderant role of the IC in

conditioned antibody production was confirmed when assessing the neuronal activity

marker c-Fos (Chen, Lin et al. 2004). The Am seems to play an important role during the

formation of aversive ingestive associations (Reilly and Bornovalova 2005) and also

appears to be relevant to limbic-autonomic interaction (Swanson and Petrovich 1998).

Based on the central findings on CTA, a series of studies investigated the

involvement of IC and Am, which are reciprocally interconnected, in conditioned


19

immunosuppression of IgM antibody production (Ramirez-Amaya, Alvarez-Borda et al.

1998; Ramirez-Amaya and Bermudez-Rattoni 1999). Both brain areas were identified as

key structures in mediating conditioned immunosuppression after evoking taste-

cyclophosphamide (CY) association. N-methyl-D-aspartic acid (NMDA)-induced lesions

either in the IC or the Am before acquisition and before evocation demonstrated that IC

lesions disrupt both acquisition and evocation of conditioned immunosuppression, while

Am lesions merely affected acquisition. In addition, cortical and amygdaloidal glutamate

release has been related to central visceral processing (Miranda, Ferreira et al. 2002).

Moreover, the IC and the Am were shown to be involved in behavioural interactions that

mediate conditioned enhancement of antibody production (Ramirez-Amaya and

Bermudez-Rattoni 1999). The ventromedial hypothalamic nucleus (VMH), widely

recognized as a satiety centre (Vettor, Fabris et al. 2002), is intimately associated with

sympathetic facilitation in peripheral tissues (Saito, Minokoshi et al. 1989), including

modulation of peripheral immune reactivity (Okamoto, Ibaraki et al. 1996). More recently,

the neural substrates involved in behaviourally conditioned immunosuppression by

cyclosporin A (CsA) in rats were identified (Figure 1.2; Pacheco-Lopez, Niemi et al. 2005).

Excitotoxic brain lesions of the IC, the Am and the VMH were shown to affect the

conditioned suppression of splenocyte proliferation and cytokine production (IL-2, IFN-γ).

More specifically, these results indicate that in the underlying paradigm the IC is

essential for acquisition and evocation of the conditioned response in the underlying

paradigm. In contrast, the Am seems to mediate the input of visceral information

necessary at acquisition time, while the VMH appears to participate in the efferent

output pathway to the immune system to evoke the behaviourally conditioned immune

response.

Investigating the conditioned enhancement of NK cell activity in rodents, it has

been demonstrated that central catecholamines and glutamate are essential factors at

the evocation stage (Hsueh, Kuo et al. 1999; Kuo, Chen et al. 2001). Central and peripheral

catecholamine contents were specifically depleted before the evocation phase by

reserpine and 6-hydroxydopamine (6-OHDA) treatment, respectively. Since reserpine

treatment impaired conditioned NK cell activity and 6-OHDA did not, central

catecholamines seem to be essential for memory retrieval during evocation (Hiramoto,

Solvason et al. 1990). These findings were confirmed by Hsueh and co-workers showing

that pre-treatment with α- and β-adrenoceptor antagonists or dopamine (DA)-1- and DA-

2-receptor antagonists before evocation also blocked the conditioned enhancement of

NK cell activity (Hsueh, Kuo et al. 1999). In addition, glutamate- but not GABA- was

required at evocation time (Kuo, Chen et al. 2001). Furthermore, it has been shown that

both the cholinergic and the serotonergic system are critical for triggering the


20

conditioned NK cell response during association and evocation phases (Hsueh, Chen et al.

2002). At association, acetylcholine seems to act through the nicotinic, M2 and M3-

muscarinic receptors whereas at evocation, neither the latter receptors nor the M1

receptors appear to affect the CR. In both the association and evocation phases, serotonin

acts through the 5-HT1 and 5-HT2 receptors to modulate the CR (Hsueh, Chen et al. 2002).

In addition, it was shown that naltrexone only blocked conditioned enhancement of NK

cell activity when applied before re-exposure to the CS, suggesting that opiate receptors

in the CNS mediate the conditioned response. In contrast, CS-US association does not

seem to involve endogenous opioids, since naltrexone administered prior to aquisition

did not interfere with the conditioning process (Solvason, Hiramoto et al. 1989).

Apart from these reports, there have been no systematic attempts to elucidate

the neural mechanisms involved in the behavioural conditioning of immune functions.

Figure 1.2: Schematic view of brain-immune interaction in the model of saccharin-cyclosporin A conditioning Two steps compose the conditioning protocol: an acquisition phase in which one or more CS-US pairings occur inducing an associative learning process, and an evocation phase where the memory of such an association is retrieved after exposure to the CS. The information on the gustatory CS is processed centrally through brain stem relays (NTS and PBN), reaching the insular cortex (IC). The IC, together with the amygdala (Am), is indispensable in CS-US association processes, and is also necessary for evoking conditioned taste avoidance (CTA). The lateral hypothalamic area (LHA) and the ventromedial nucleus of the hypothalamus (VMH) are also essential for evoking the conditioned immunosuppression in the periphery (modified from Pacheco-Lopez, Niemi et al. 2005).

Peripheral mechanisms of saccharin-cyclosporin A conditioning

Even though progress has been made in elucidating the central mechanisms of

behavioural conditioning during the last two decades, little is known about the


21

peripheral mechanisms behind this phenomenon. This section focusses on a well-

established conditioning model implementing the association of a gustatory stimulus

with the immunosuppressive effects of i.p. administered CsA.

Some paradigms investigating conditioned changes in immune functions were

intended to produce “stress effects” evoked by activation of the HPA axis (Kelley, Dantzer

et al. 1985). However, corticosterone levels remained unchanged after evocation in a

model of CsA conditioning (Exton, von Horsten et al. 1998). Based on the main findings on

neural innervation of secondary lymphoid organs such as the spleen (Nance and Sanders

2007) and the expression of receptors for neurotransmitters on lymphocytes (Elenkov,

Wilder et al. 2000), the splenic nerve was identified as the major efferent mediator

linking the brain and the immune system. Chemical and surgical sympathetic

denervation completely blocked conditioned immunosuppression by CsA (Exton, von

Horsten et al. 1998; Exton, Schult et al. 1999) and noradrenaline was identified as the

predominant neurotransmitter mediating this effect (Exton, Gierse et al. 2002).

Moreover, in vivo administartion of the β-adrenergic receptor anatgonist propranolol and

continuous blockade of β-adrenergic receptors using isoproterenol in an in vitro model

demonstrated that saccharin (SAC)-CsA conditioned immunosuppression seems to be β-

adrenoceptor-dependent. In contrast, splenic denervation did not affect conditioned

suppression of contact hypersensitivity reaction by CsA in the same paradigm, since this

acute immune response occurs predominately in the draining lymph nodes (Exton, Elfers

et al. 2000).

Hence, the fundamental peripheral mechanisms in behavioural conditioning are

still poorly understood and ongoing studies will have to focus particularly on the cellular

and intracellular pathways in order to provide a clinical framework for behavioural

conditioning and to contribute to a better understanding of the principles of behavioural

interaction in individuals with different neurological and immune histories.

Objectives and outline of the thesis

At present the peripheral mechanisms occurring during conditioning as well as

the aspect of specificity of conditioned immunosuppression by CsA are completely

unknown. It is hypothesized that a more physiological immuno-sensory process might

implicate that the CNS not only detects peripheral cytokine increases but also “senses” a

reduction in cytokine levels in the periphery, as occurrs after the administration of an

immunosuppressive drug such as CsA.

In an attempt to advance the understanding of the mechanisms and biological

relevance of conditioned immunomodulation by the immunosuppressive agent CsA, part


22

I of this thesis was designed to analyze the central and peripheral immuno-sensory

processes follwoing intraperitoneal CsA administration. In order to determine the time

point when the CsA (US) signal reaches the CNS, and to analyze whether a decrease in

peripheral cytokines in the spleen is responsible for the mediation of the US during

acquisition time, cytokine mRNA expression in the Am and the spleen were monitored at

various time points.

In part II of the thesis the aspect of the specificity of CIS by CsA in the periphery

was analyzed. CsA exerts its specific actions in the periphery by inhibition of the

phosphate CaN in T lymphocytes (Ho, Clipstone et al. 1996). In order to determine

whether CIS by CsA is mediated on a cellular level also via CaN and to elaborate potential

mechanisms of CIS by CsA, CaN activity in splenic lysates as well as the proliferative

response of T lymphocytes after conditioning were assessed.

The findings of part II identifying CaN as a target molecule and T lymphocytes as the

target cell population of the conditioned response, as well as the findings of Exton and

coworkers showing that the behavioral conditioned effects were mediated in the

periphery via noradrenaline and β-adrenoceptors, led to the establishment of an ex vivo

model to analyze potential mechanisms underlying CIS by CsA (Exton, Gierse et al. 2002;

Pacheco-Lopez, Riether et al. 2009).

In an attempt to advance the understanding of the peripheral mechanisms and

especially the correlation between β2-adrenergic sympathetic stimulation and the

observed reduction in CaN activity in conditioned animals, the β2-adrenergic pathway in

terbutaline-treated T lymphocytes was dissected, identifying AKAP-150 as a potential

mediator of PKA dependent CaN inhibition after terbutaline treatment.

These results allow new insights into the peripheral mediation of conditioned

immunosuppression by CsA and strengthen the aspect of specificity in this phenomenon.

The present thesis mostly consists of research articles originally written for

separate peer-reviewed journal publications. A more detailed description of the

objectives of the separate studies can be found in the introductory sections of the

corresponding chapters.


23

References Acquas, E., C. Wilson, et al. (1996). "Conditioned and unconditioned stimuli increase frontal cortical and

hippocampal acetylcholine release: effects of novelty, habituation, and fear." J Neurosci 16(9): 3089-

96.

Ader, R. (2003). "Conditioned immunomodulation: research needs and directions." Brain Behav Immun 17

Suppl 1: S51-7.

Ader, R. and N. Cohen (1975). "Behaviorally conditioned immunosuppression." Psychosom Med 37(4): 333-40.

Ader, R., Cohen, N. (2001). Conditioning and Immunity. Psychoneuroimmunology. R. Ader, Felten, D., Cohen, N.

New York, Academic Press. 2: 3-34.

Banks, W. A., S. A. Farr, et al. (2001). "Intravenous human interleukin-1alpha impairs memory processing in

mice: dependence on blood-brain barrier transport into posterior division of the septum." J

Pharmacol Exp Ther 299(2): 536-41.

Bermúdez-Rattoni, F. (2004). "Molecular mechanisms of taste-recognition memory." Nat Rev Neurosci 5(3):

209 - 17.

Bermudez-Rattoni, F. and J. L. McGaugh (1991). "Insular cortex and amygdala lesions differentially affect

acquisition on inhibitory avoidance and conditioned taste aversion." Brain Res 549(1): 165 - 70.

Bermudez-Rattoni, F., L. Ramirez-Lugo, et al. (2004). "Molecular signals into the insular cortex and amygdala

during aversive gustatory memory formation." Cell Mol Neurobiol 24(1): 25-36.

Besedovsky, H. O. and A. del Rey (1996). "Immune-neuro-endocrine interactions: facts and hypotheses." Endocr

Rev 17(1): 64-102.

Besedovsky, H. O. and A. D. del Rey (2007). "Physiology of psychoneuroimmunology: a personal view." Brain

Behav Immun 21(1): 34-44.

Blalock, J. E. and E. M. Smith (2007). "Conceptual development of the immune system as a sixth sense." Brain

Behav Immun 21(1): 23-33.

Carew, T. J. and C. L. Sahley (1986). "Invertebrate learning and memory: from behavior to molecules." Annu Rev

Neurosci 9: 435-87.

Chen, J., W. Lin, et al. (2004). "Enhancement of antibody production and expression of c-Fos in the insular

cortex in response to a conditioned stimulus after a single-trial learning paradigm." Behav Brain Res

154(2): 557 - 65.

Cubero, I., T. E. Thiele, et al. (1999). "Insular cortex lesions and taste aversion learning: effects of conditioning

method and timing of lesion." Brain Res 839(2): 323 - 30.

Dantzer, R. (2004). "Cytokine-induced sickness behaviour: a neuroimmune response to activation of innate

immunity." Eur J Pharmacol 500(1-3): 399-411.

Diana, A., A. M. Van Dam, et al. (1999). "Co-localization of interleukin-1 receptor type I and interleukin-1

receptor antagonist with vasopressin in magnocellular neurons of the paraventricular and

supraoptic nuclei of the rat hypothalamus." Neuroscience 89(1): 137-47.

Domjan, M. (2005). "Pavlovian conditioning: a functional perspective." Annu Rev Psychol 56: 179-206.

Elenkov, I. J., R. L. Wilder, et al. (2000). "The sympathetic nerve--an integrative interface between two

supersystems: the brain and the immune system." Pharmacol Rev 52(4): 595-638.

Exton, M. S., A. Elfers, et al. (2000). "Conditioned suppression of contact sensitivity is independent of

sympathetic splenic innervation." Am J Physiol Regul Integr Comp Physiol 279(4): R1310-5.

Exton, M. S., C. Gierse, et al. (2002). "Behaviorally conditioned immunosuppression in the rat is regulated via

noradrenaline and beta-adrenoceptors." J Neuroimmunol 131(1-2): 21-30.

Exton, M. S., M. Schult, et al. (1999). "Conditioned immunosuppression makes subtherapeutic cyclosporin

effective via splenic innervation." Am J Physiol 276(6 Pt 2): R1710-7.


24

Exton, M. S., S. von Horsten, et al. (1998). "Behaviorally conditioned immunosuppression using cyclosporine A:

central nervous system reduces IL-2 production via splenic innervation." J Neuroimmunol 88(1-2):

182-91.

Felten, D. L., S. Y. Felten, et al. (1985). "Noradrenergic and peptidergic innervation of lymphoid tissue." J

Immunol 135(2 Suppl): 755s-765s.

Gaillard, R. C. (1998). "Cytokines in the neuroendocrine system." Int Rev Immunol 17(1-4): 181-216.

Garcia, J., E. Ervin, et al. (1966). "Learning with prolonged delay of reinforcement." Psychon. Sci. 5: 121-122.

Garcia, J., D. J. Kimeldorf, et al. (1955). "Conditioned aversion to saccharin resulting from exposure to gamma

radiation." Science 122(3160): 157 - 8.

Goehler, L. E., R. P. Gaykema, et al. (1998). "Interleukin-1 induces c-Fos immunoreactivity in primary afferent

neurons of the vagus nerve." Brain Res 804(2): 306-10.

Hiramoto, R., B. Solvason, et al. (1990). "Effect of reserpine on retention of the conditioned NK cell response."

Pharmacol Biochem Behav 36(1): 51 - 6.

Ho, S., N. Clipstone, et al. (1996). "The mechanism of action of cyclosporin A and FK506." Clin Immunol

Immunopathol 80(3 Pt 2): S40-5.

Horuk, R., A. W. Martin, et al. (1997). "Expression of chemokine receptors by subsets of neurons in the central

nervous system." J Immunol 158(6): 2882-90.

Hsueh, C., S. Chen, et al. (2002). "Cholinergic and serotonergic activities are required in triggering conditioned

NK cell response." J Neuroimmunol 123(1-2): 102 - 11.

Hsueh, C., J. Kuo, et al. (1999). "Involvement of catecholamines in recall of the conditioned NK cell response." J

Neuroimmunol 94(1-2): 172 - 81.

Kelley, K. W., R. Dantzer, et al. (1985). "Conditioned taste aversion suppresses induction of delayed-type

hypersensitivity immune reactions." Physiol Behav 34(2): 189-93.

Kuo, J., S. Chen, et al. (2001). "The involvement of glutamate in recall of the conditioned NK cell response." J

Neuroimmunol 118(2): 245 - 55.

Miranda, M. I., G. Ferreira, et al. (2002). "Glutamatergic activity in the amygdala signals visceral input during

taste memory formation." Proc Natl Acad Sci U S A 99(17): 11417-22.

Morikawa, Y., K. Tohya, et al. (2000). "Expression of interleukin-6 receptor, leukemia inhibitory factor receptor

and glycoprotein 130 in the murine cerebellum and neuropathological effect of leukemia inhibitory

factor on cerebellar Purkinje cells." Neuroscience 100(4): 841-8.

Nance, D. M. and V. M. Sanders (2007). "Autonomic innervation and regulation of the immune system (1987-

2007)." Brain Behav Immun 21(6): 736-45.

Neumann, H., H. Schmidt, et al. (1997). "Interferon gamma gene expression in sensory neurons: evidence for

autocrine gene regulation." J Exp Med 186(12): 2023-31.

Okamoto, S., K. Ibaraki, et al. (1996). "Ventromedial hypothalamus suppresses splenic lymphocyte activity

through sympathetic innervation." Brain Res 739(1-2): 308 - 13.

Pacheco-Lopez, G., M. B. Niemi, et al. (2007). Behaviorally Conditioned Enhancement of Immune Responses.

Psychoneuroimmunology. R. Ader. New York, Academic Press. 1: 631-660.

Pacheco-Lopez, G., M. B. Niemi, et al. (2005). "Neural substrates for behaviorally conditioned

immunosuppression in the rat." J Neurosci 25(9): 2330-7.

Pacheco-Lopez, G., C. Riether, et al. (2009). "Calcineurin inhibition in splenocytes induced by Pavlovian

conditioning." FASEB J 23 (4): 1161-7.

Pavlov, I. P. (1927). "Conditioned Reflexes: An Investigation of the Physiological Activity of the Cerebral Cortex."

Oxford University Press.

Pearce, J. (1987). "A model for stimulus generalization in Pavlovian conditioning." Psychol Rev 94(1): 61 - 73.


25

Quan, N. and W. A. Banks (2007). "Brain-immune communication pathways." Brain Behav Immun 21(6): 727-35.

Ramirez-Amaya, V., B. Alvarez-Borda, et al. (1998). "Differential effects of NMDA-induced lesions into the

insular cortex and amygdala on the acquisition and evocation of conditioned immunosuppression."

Brain Behav Immun 12(2): 149-60.

Ramirez-Amaya, V. and F. Bermudez-Rattoni (1999). "Conditioned enhancement of antibody production is

disrupted by insular cortex and amygdala but not hippocampal lesions." Brain Behav Immun 13(1):

46-60.

Reilly, S. and M. A. Bornovalova (2005). "Conditioned taste aversion and amygdala lesions in the rat: a critical

review." Neurosci Biobehav Rev 29(7): 1067 - 88.

Rescorla, R. (1988). "Behavioral studies of Pavlovian conditioning." Annu Rev Neurosci 11: 329 - 52.

Rescorla, R. and A. Wagner (1972). A theory of pavlovian conditioning: variations in the efectiveness of

reinforcement and nonreinforcement. Classical conditioning II: current research and theory. A. Black

and W. Prokasy. New York, Appleton-Century-Crofts: 64-99.

Riether, C., R. Doenlen, et al. (2008). "Behavioural conditioning of immune functions: how the central nervous

system controls peripheral immune responses by evoking associativel earning processes." Reviews

in the Neurosciences 19: 1-17.

Saito, M., Y. Minokoshi, et al. (1989). "Accelerated norepinephrine turnover in peripheral tissues after

ventromedial hypothalamic stimulation in rats." Brain Res 481(2): 298 - 303.

Sanders, V. M. and R. H. Straub (2002). "Norepinephrine, the beta-adrenergic receptor, and immunity." Brain

Behav Immun 16(4): 290-332.

Schafe, G., S. Sollars, et al. (1995). "The CS-US interval and taste aversion learning: a brief look." Behav Neurosci

109(4): 799 - 802.

Sewards, T. V. (2004). "Dual separate pathways for sensory and hedonic aspects of taste." Brain Res Bull 62(4):

271 - 83.

Sewards, T. V. and M. Sewards (2002). "Separate, parallel sensory and hedonic pathways in the mammalian

somatosensory system." Brain Res Bull 58(3): 243 - 60.

Sewards, T. V. and M. A. Sewards (2001). "Cortical association areas in the gustatory system." Neurosci

Biobehav Rev 25(5): 395 - 407.

Solvason, H. B., R. N. Hiramoto, et al. (1989). "Naltrexone blocks the expression of the conditioned elevation of

natural killer cell activity in BALB/c mice." Brain Behav Immun 3(3): 247-62.

Swanson, L. W. and G. D. Petrovich (1998). "What is the amygdala?" Trends Neurosci 21(8): 323 - 31.

Tracey, K. J. (2002). "The inflammatory reflex." Nature 420(6917): 853-9.

Vettor, R., R. Fabris, et al. (2002). "Neuroendocrine regulation of eating behavior." J Endocrinol Invest 25(10):

836 - 54.

Wang, G., C. Lu, et al. (2001). "Immunohistochemical localization of interleukin-2 and its receptor subunits

alpha, beta and gamma in the main olfactory bulb of the rat." Brain Res 893(1-2): 244-52.

Watkins, L. R. and S. F. Maier (2000). "The pain of being sick: implications of immune-to-brain communication

for understanding pain." Annu Rev Psychol 51: 29-57.

Yamamoto, T., T. Shimura, et al. (1994). "Neural substrates for conditioned taste aversion in the rat." Behav

Brain Res 65(2): 123-37.

Ziemssen, T. and S. Kern (2007). "Psychoneuroimmunology-cross-talk between the immune system and

nervous system." J Neurol 254 (Suppl 2): II8-II11.

Chapter 2: Cyclosporin A as an unconditioned stimulus is detected by the central nervous system at acquisition time

26

Chapter 2.

Cyclosporin A as an unconditioned stimulus is detected by

the central nervous system at acquisition time


27

Introduction

Psychoneuroimmunological research is demonstrating that there is intense

communication among the central nervous system, the neuroendocrine system and the

immune system (Ader, Cohen et al. 1995; Madden and Felten 1995; Besedovsky and del

Rey 1996; Ader 2001; Exton, Herklotz et al. 2001; Straub and Schedlowski 2002;

Besedovsky and Rey 2007; Tracey 2007; Dantzer, O'Connor et al. 2008)

Employing a conditioned taste aversion (CTA) paradigm, pairing the

immunosuppressive drug cyclosporin A (CsA) as unconditioned stimulus (US) and

saccharin (SAC) taste as conditioned stimulus (CS) behaviorally conditioned suppression

of peripheral immune functions in rats has been exquisitely demonstrated as illustrated

by suppressed splenocyte proliferation, interleukin-2 (IL-2) and interferon-γ (IFN-γ)

production and -cytokine mRNA expression at evocation time (Exton, Schult et al. 1998;

Exton, von Horsten et al. 1998). Behavioral conditioned immunosuppression by CsA in the

spleen is mediated via the efferent sympathetic pathway namely: the splenic nerve,

noradrenaline- and β-adrenoceptors-dependent mechanisms (Exton, Gierse et al. 2002;

Xie, Frede et al. 2002). Moreover, it was recently demonstrated that the insular cortex (IC)

and the amygdala (Am) are essential neuronal structures for acquisition in the behavioral

conditioning of immune functions (Pacheco-Lopez, Niemi et al. 2005).

During behavioral conditioning one essential step is the detection of the CS and

the US by the CNS. Relating to the CS (SAC taste), it has been reported that

neurotransmitters such as acetylcholine in the IC and the Am, and noradrenaline in the

Am seem to be involved in taste recognition memory, which is necessary during the

acquisition phase (Williams, Men et al. 2000; Miranda, LaLumiere et al. 2003).

Acetylcholine appears to codify the novelty of the conditioned (Miranda, Ramirez-Lugo et

al. 2000) as well as the unconditioned stimuli (Acquas, Wilson et al. 1996). Regarding

central visceral processing, cortical and amygdaloidal glutamate release also appear to be

involved. In addition, noradrenergic input into the Am has been related to vagal arousal,

enhancing the associative learning processes like CTA (Berman and Dudai 2001; Miranda,

LaLumiere et al. 2003; Hassert, Miyashita et al. 2004) .

In the periphery, based on anatomical and functional considerations, the vagus

nerve is particularly well situated for an immunosensory function. Therefore, the vagus

nerve, having relays in brain stem nuclei, has been proposed as the main afferent

pathway in mediating the immunosensory process, before the immune information

reaches the forebrain structures. Subsequent to peripheral antigen immunization rapid

and transient increases of peripheral cytokine concentrations are observed, occurring in

parallel with the activation of the central catecholaminergic system (Tracey 2007). In this

context, it has been demonstrated that different immune stimuli (e.g. antigens or


28

cytokines) activate vagal sensory neurons (Niijima 1996; Ek, Kurosawa et al. 1998;

Gaykema, Goehler et al. 1998; Goehler, Gaykema et al. 1998). Furthermore, the CNS

consequences of such immune challenges, e.g. c-Fos expression, fever or anorexia, are

inhibited by sub-diaphragmatic vagotomy (Watkins, Goehler et al. 1995; Sehic and

Blatteis 1996; Romanovsky, Simons et al. 1997; Fleshner, Goehler et al. 1998). Additionally,

the fact that Am activity reflects peripheral antigenic stimulation has been extensively

documented (Tkacs and Li 1999). A more physiological sensory process might implicate

that the CNS not only detects changes in peripheral cytokine increases but also “senses” a

reduction in cytokine production or expression in the periphery, induced by an

immunosuppressive drug such as CsA. However, at present it is completely unknown,

when and how the CNS detects an US with immune consequences such as the

immunosuppressive drug CsA.

Despite its lipophilicity, CsA is not able to reach the CNS with a healthy blood -

brain barrier (Halloran 2001). Therefore, a direct CsA effect on the CNS is not likely.

Originally ascribed to the immune system, cytokines play an active and important role in

the afferent communication between the immune system and the CNS. It has been

reported however, that the CNS is able to detect, or “sense”, changes in cytokine

concentrations in the periphery, in particular changes in pro-inflammatory cytokines such

as interleukin-1 (IL-1), interleukin-6 (IL-6) or tumor-necrosis-factor-alpha (TNF) (Gibertini

1996; Schneider, Pitossi et al. 1998; Banks, Farr et al. 2001; Matsumoto, Yoshida et al. 2001;

Rachal Pugh, Fleshner et al. 2001; Besedovsky and Rey 2007). These cytokines seem to be

partly under central monoaminergic control (Depino, Earl et al. 2003). Apart from these

neuromodulatory properties, pro-inflammatory cytokines seem to play an important role

in the afferent pathway between the immune system and the CNS (Besedovsky and del

Rey 1996; Turnbull and Rivier 1999; Bernik, Friedman et al. 2002; Dantzer, O'Connor et al.

2008). Therefore, it can be hypothesized that central cytokines act as mediators in the

CNS in an “immune-sensing” step during the acquisition time of behavioral conditioned

immunosuppression. This hypothesis is supported by observations that 1) receptors for

these pro-inflammatory cytokines are expressed in the CNS, 2) peripheral immune

changes affect central cytokine production and cytokine receptor expression in the brain,

and 3) cytokines can act as unconditioned stimuli to induce CTA (Tazi, Dantzer et al. 1988;

Hiramoto, Ghanta et al. 1993; Longo, Duffey et al. 1999; Mormede, Palin et al. 2004). In

healthy conditions the blood-brain barrier limits the crossing of peripheral-born cytokines

to the CNS (Dantzer, Konsman et al. 2000), however it has been demonstrated that

peripheral immune challenges as well as cytokines induce the production and release of

cytokines within the CNS (Laye, Parnet et al. 1994; Pitossi, del Rey et al. 1997; Turrin, Gayle

et al. 2001). To date, the role of CNS cytokines within non-pathological settings is still


29

under discussion, however it has been reported that CNS cytokines, such as IL-1, may be

involved in the central processing of immune stimuli (Maier 2003). Apart from these

observations no systematic approach has been undertaken to elucidate the role of pro-

inflammatory cytokines in the central processing of behavioral conditioning of immune

functions. It is hypothesized that a more physiological immunosensory process might

implicate that the CNS not only detects peripheral cytokine increases but also “senses” a

reduction in cytokine levels in the periphery, as occurrs after the administration of an

immunosuppressive drug such as CsA. However, experimental data about the afferent

pathway(s) and central processes at acquisition time are completely lacking. Since we

consider that an “immunosensory” process supports CsA detection by the CNS, we

decided to quantify the expression of selected cytokines within the spleen and Am after

peripheral CsA administration.

Materials and Methods

Animals

Male Dark Agouti (DA) rats, weighing between 220 and 250g, were obtained from

Harlan Netherland (Horst, Netherlands). Animals were individually housed under an

inverted 12:12h light/dark schedule (lights off at 7am) with food and water available ad

libitum. All procedures were approved by the Swiss Cantonal Veterinary Office, and are in

agreement with the Principles of Laboratory Animal Care (NIH publication No. 86-23,

revised 1985).

Experimental protocol

Animals were randomly assined in two groups and injected i.p. with either sterile

saline or cyclosporin A (20mg/kg, Sandimmune, Novartis, Basel, Switzerland). 120min,

240min and 360min after CsA and saline injections, the animals were deeply

anesthetized with isoflurane (Provet AG, Lyssach, Switzerland). Animals were

transcardially perfused with low molarity PBS followed by high molarity PBS containing

4% paraformaldehyde. The brains were removed, postfixed for 24 h and cyroprotected by

immersion in 30% sucrose for 72 h.

Collection of brain samples

After decapitation, rat brains were carefully removed and placed in a sterile

polystyrene tissue culture dish (BD Pharmingen, Allschwil, Switzerland). The tissue

culture dish was tightly sealed to avoid desiccation (Ziploc® plastic bag, SC Johnson AG,

Dietikon, Switzerland) and was frozen in liquid nitrogen. The total brain isolation


30

procedure was restricted to 5min including decapitation. Afterwards brains were stored

at -80°C until dissection. Serial coronal sections (300µm) were collected using a cryostat

set at -5°C (Reichert-Jung Frigocut 2700, Nussloch, Germany) and placed on pre-chilled

glass slides (SuperFrost®, Roth, Karlsruhe, Germany). The tissue was stored in tightly

sealed slide boxes at -80°C until punching. The area of interest was dissected using a

micro-punch technique described elsewhere (Palkovits 1973; Cuello and Carson 1983).

Briefly, a pre-chilled stainless steel sample corer (Fine Science Tools, GmbH, Heidelberg,

Germany) with an internal diameter of 1mm was used to dissect the amygdala from

selected brain sections placed on dry-ice (Figure 2.1). The optical tract and hippocampus

were used as main anatomical landmarks to obtain comparable dissection of the

amygdala across animals. Dissected tissue was immediately processed for mRNA

isolation as described in the method section on mRNA isolation and cDNA synthesis.

Figure 2.1: Micro-dissection of the amygdala applying the micro-punch technique (see Methods section: collection of brain samples)

Cytokine mRNA expression

Total RNA was isolated from the amygdala and the spleen using Trizol reagent

(Invitrogen) according to the manufacturer's instructions, homogenized with a

TissueTearor (BioSpec Products) and stored at –80°C. The extracted RNA was treated

with DNase I and further purified with RNeasy Mini columns (Qiagen, Hombrechtikon,

Switzerland). Single-strand cDNA was generated by reverse transcription (RT) using

Superscript II reverse transcriptase and random hexamer primers (Invitrogen).

The reaction mixture (20µl) contained 5µg total RNA, 4µl of 5x first-strand buffer, 5mM

DTT, 0.5mM dNTPs, 250ng random hexamer primers, and 40U RNaseOut RNase inhibitor

(all from Invitrogen). After incubation for 10 min at 25°C, RT was carried out for 60min at

42°C, followed by RT inactivation for 15min at 70°C and RNase H (Invitrogen) incubation

for 20min at 37°C. Gene expression analysis using Taqman gene expression assays

(Applied Biosystems, Rotkreuz, Switzerland) for IL-1β (RN00580432_m1), IL-6

(RN00561420_m1), IL-2 (Rn00587673_m1), TNF (RN99999017_m1), IFN-γ

(RN00594078_m1) and 18S rRNA (#4319413E) was performed on an Applied Biosystems


31

(Foster City, CA, USA) 7500 Real-Time PCR. Relative gene expression was calculated

according to the 2-∆∆CT method (Livak and Schmittgen 2001). The synthesized cDNA was

stored at -20°C.

Statistical analysis

Cytokine gene expression data were analyzed using independent Student's t test

(two-tailed). Statistical significance was set at P < 0.05. Statistics were calculated using

SPSS for Windows (Version 14.0, SPSS, Chicago, IL, USA).

Results

Cytokine mRNA expression analyses in the spleen revealed that the expression of

the pro-inflammatory cytokines IL-1β, IL-6 and TNF significantly decreases 120min after

CsA injection (p<0.05), with these effects being reversed after 240 and 360min (Figure

2.2B). The relative gene expression profiles for IL-2 and IFN-y at the indicated time points,

however, did not significantly differ from the controls. In addition, plasma cytokine

concentrations were assessed for the corresponding cytokines (Doenlen 2008). However,

statistically insignificant differences were observed after CsA injection at the given time

points.

These data demonstrate that a single CsA injection transiently decreases pro-

inflammatory, but not Th1- cytokine mRNA expression in the spleen, 120 min after

injection without affecting cytokine plasma levels.

To elucidate whether these transient peripheral alterations in pro-inflammatory

cytokine mRNA expression in the spleen are detected by the CNS, Th1- and pro-

inflammatory cytokine mRNA expression was assessed in the Am up to 360min after CsA

injection. The analyses revealed that solely IL-1β mRNA expression was statistically

significantly increased 240 and 360min after CsA injection (p<0.05; Figure 2.2A). The

expression of the pro-inflammatory cytokines IL-6, TNF, IL-2 and IFN-y remained

unaffected.


32

Figure 2.2: Relative pro-inflammatory and Th

1-cytokine mRNA expression in A) amygdala (Am) and B) spleen

(Spl) of vehicle injected control rats (open bars, n=6) and CsA-injected rats 120min (hatched bars, n=8), 240min (cross-hatched bars, n=8) and 360min (solid bars, n=9) after injection. Statistics: Student`s T-Test: *p<0.05 vs vehicle injected controls.

In summary, these data demonstrate that the CNS seems to detect the injection of the

immunosuppressive drug CsA at acquisition and that there is subsequent reduction in

pro-inflammatory cytokine expression in the spleen 120min after injection by de novo

synthesis of IL-1β mRNA expression in the CNS.

Discussion

The immune system and the CNS use a variety of common signalling molecules

participating in local as well intersystemic communication like for example pro-

inflammatory cytokines or neurotransmitters (Besedovsky and del Rey 1996; Tracey 2007).

The pro-inflammatory cytokines IL-1, IL-6, and TNF, for instance have been demonstrated

to play an important role in the afferent communication between the immune system

and the CNS and were shown to be involved in cognitive processes such as learning and

memory (Gibertini 1996; Banks, Farr et al. 2001; Matsumoto, Yoshida et al. 2001; Rachal

Pugh, Fleshner et al. 2001). However, during development a phenomenon where

messengers of two anatomically separated systems communicate with each other

evolved.

Primitive species and non-mammalian species like sponges, insects and molluscs have

been shown to produce pro-inflammatory cytokines in their host defence response to

microbial invaders (for review see Maier 2003). For example, IL-1β was detected in the

hemolymph of molluscs participating in the regulation of local inflammation, injury and


33

energy production (Pipe 1990; Beck and Habicht 1991; Hughes, Smith et al. 1992).

Furthermore, molluscs like Aplysia californica and Mytilus edulis use IL-1β in the

nociceptive process of the withdrawal reflex. Nerve injury results in the accumulation of

hematocytes at the damaged site and leads to the secretion of cytokines (Clatworthy,

Castro et al. 1994; Clatworthy and Walters 1994). In addition, the sensitization of the

withdrawal reflex is triggered by cytokines like TNF and IL-1β by modulation of ion

channels in the neurons (Sawada, Hara et al. 1991). In summary, during early evolution

pro-inflammatory cytokines were implemented in the process of host defence including

energy production. Moreover, IL-1β derived from the immune cells of primitive organisms

was shown to participate in the neuro-immune communication between organisms.

However, none of organisms decribed above have a BBB. In more complex organisms, the

BBB was developed much later. Therefore, the process of immune-to-brain

communication had to be adapted and became independent of direct cytokine actions.

Consequently, a pathway using peripheral-cytokine induced de novo synthesis of

cytokines within the brain by microglia was developed. During evolution this process of

immune-to-brain communication was conserved and implemented in the mediation of

other fundamental processes such as sickness behaviour, stress or pain response (Tracey

2007).

In summary, the results are in accordance with the data exploring the

bidirectional communication between the immune system and the brain. Pro-

inflammatory cytokines expressed in the CNS and the periphery were shown to affect

brain function, alter memory and learning and direct immune-brain communication

(Lynch 2002; Maier 2003; Balschun, Wetzel et al. 2004). In this study, a peripheral

injection of CsA is perceived in the periphery as an immune challenge as illustrated by

reduced pro-inflammatory cytokine mRNA expression in the spleen. These changes elicit

a subsequent increase in IL-1β mRNA expression in the Am which plays a key role during

acquisition of the conditioned immunosuppression by CsA and which was shown to

participate in the detection of peripheral antigenic stimulation by the CNS. Taking into

consideration that cytokines such as IL-α have also been shown to be an appropriate US

during conditioning of NK cell activity (Demissie, Rogers et al. 1996), it can be concluded

that the peripheral reduction in pro-inflammatory cytokine levels, as shown after the

administration of CsA, is sensed by the CNS and triggers the de novo production of

central IL-1β during acquisition of the CIS. Therefore, it could clearly be demonstrated that

CsA as a US is detected by the CNS at acquisition time. However, if peripheral

immunosuppression is also mediated via the activation of vagal neurons as peripheral

immune activation (Niijima 1996; Ek, Kurosawa et al. 1998; Gaykema, Goehler et al. 1998;


34

Goehler, Gaykema et al. 1998), the consequences of IL-1β mRNA up-regulation in the CNS

after CsA administration still have to be elucidated.

References Acquas, E., C. Wilson, et al. (1996). "Conditioned and unconditioned stimuli increase frontal cortical and

hippocampal acetylcholine release: effects of novelty, habituation, and fear." J Neurosci 16(9): 3089-

96.

Ader, R., N. Cohen, et al. (1995). "Psychoneuroimmunology: interactions between the nervous system and the

immune system." Lancet 345(8942): 99-103.

Ader, R., Cohen, N. (2001). Conditioning and Immunity. Psychoneuroimmunology. R. Ader, Felten, D., Cohen, N.

New York, Academic Press. 2: 3-34.

Balschun, D., W. Wetzel, et al. (2004). "Interleukin-6: a cytokine to forget." Faseb J 18(14): 1788-90.

Banks, W. A., S. A. Farr, et al. (2001). "Intravenous human interleukin-1alpha impairs memory processing in

mice: dependence on blood-brain barrier transport into posterior division of the septum." J

Pharmacol Exp Ther 299(2): 536-41.

Beck, G. and G. S. Habicht (1991). "Primitive cytokines: harbingers of vertebrate defense." Immunol Today 12(6):

180-3.

Berman, D. E. and Y. Dudai (2001). "Memory extinction, learning anew, and learning the new: dissociations in

the molecular machinery of learning in cortex." Science 291(5512): 2417-9.

Bernik, T. R., S. G. Friedman, et al. (2002). "Pharmacological stimulation of the cholinergic antiinflammatory

pathway." J Exp Med 195(6): 781-8.

Besedovsky, H. O. and A. del Rey (1996). "Immune-neuro-endocrine interactions: facts and hypotheses." Endocr

Rev 17(1): 64-102.

Besedovsky, H. O. and A. D. Rey (2007). "Physiology of psychoneuroimmunology: a personal view." Brain Behav

Immun 21(1): 34-44.

Clatworthy, A. L., G. A. Castro, et al. (1994). "Induction of a cellular defense reaction is accompanied by an

increase in sensory neuron excitability in Aplysia." J Neurosci 14(5 Pt 2): 3263-70.

Clatworthy, A. L. and E. T. Walters (1994). "Comparative analysis of hyperexcitability and synaptic facilitation

induced by nerve injury in two populations of mechanosensory neurones of Aplysia californica." J

Exp Biol 190: 217-38.

Cuello, C. and S. Carson (1983). Microdissection of Fresh Rat Brain Tissue Slides. Brain Microdisscetion

Techniques. C. Cuello, International Brain Research Organization: 37-125.

Dantzer, R. (2004). "Cytokine-induced sickness behaviour: a neuroimmune response to activation of innate

immunity." Eur J Pharmacol 500(1-3): 399-411.

Dantzer, R., J. P. Konsman, et al. (2000). "Neural and humoral pathways of communication from the immune

system to the brain: parallel or convergent?" Auton Neurosci 85(1-3): 60-5.

Dantzer, R., J. C. O'Connor, et al. (2008). "From inflammation to sickness and depression: when the immune

system subjugates the brain." Nat Rev Neurosci 9(1): 46-56.

Demissie, S., C. F. Rogers, et al. (1995). "Arecoline a muscarinic cholinergic agent conditions central pathways

that modulate natural killer cell activity." J Neuroimmunol 59(1-2): 57-63.

Depino, A. M., C. Earl, et al. (2003). "Microglial activation with atypical proinflammatory cytokine expression in

a rat model of Parkinson's disease." Eur J Neurosci 18(10): 2731-42.


35

Doenlen, R. (2008). Fingerprints of Neural Activity after Peripheral Immune Challenges: An Experimental

Investigation of the Bi-directional Communication between the Central Nervous System and the

Immune System

Institute for Behavioral Sciences. Zurich, ETH Zurich. Dr. sc. .

Ek, M., M. Kurosawa, et al. (1998). "Activation of vagal afferents after intravenous injection of interleukin-

1beta: role of endogenous prostaglandins." J Neurosci 18(22): 9471-9.



Exton, M. S., J. Herklotz, et al. (2001). "Conditioning in the rat: an in vivo model to investigate the molecular

mechanisms and clinical implications of brain-immune communication." Immunol Rev 184: 226-35.

Exton, M. S., M. Schult, et al. (1998). "Behavioral conditioning prolongs heart allograft survival in rats."

Transplant Proc 30(5): 2033.



182-91.

Fleshner, M., L. E. Goehler, et al. (1998). "Thermogenic and corticosterone responses to intravenous cytokines

(IL-1beta and TNF-alpha) are attenuated by subdiaphragmatic vagotomy." J Neuroimmunol 86(2):

134-41.

Gaykema, R. P., L. E. Goehler, et al. (1998). "Bacterial endotoxin induces fos immunoreactivity in primary

afferent neurons of the vagus nerve." Neuroimmunomodulation 5(5): 234-40.

Gibertini, M. (1996). "IL1 beta impairs relational but not procedural rodent learning in a water maze task." Adv

Exp Med Biol 402: 207-17.

Goehler, L. E., R. P. Gaykema, et al. (1998). "Interleukin-1 induces c-Fos immunoreactivity in primary afferent

neurons of the vagus nerve." Brain Res 804(2): 306-10.

Halloran, P. F. (2001). "Mechanism of action of the calcineurin inhibitors." Transplant Proc 33(7-8): 3067-9.

Hassert, D. L., T. Miyashita, et al. (2004). "The effects of peripheral vagal nerve stimulation at a memory-

modulating intensity on norepinephrine output in the basolateral amygdala." Behav Neurosci 118(1):

79-88.

Hiramoto, R., V. Ghanta, et al. (1993). "Identification of specific pathways of communication between the CNS

and NK cell system." Life Sci 53(7): 527-40.

Hughes, T. K., E. M. Smith, et al. (1992). "Immunoreactive cytokines in Mytilus edulis nervous and immune

interactions." Acta Biol Hung 43(1-4): 269-73.

Laye, S., P. Parnet, et al. (1994). "Peripheral administration of lipopolysaccharide induces the expression of

cytokine transcripts in the brain and pituitary of mice." Brain Res Mol Brain Res 27(1): 157-62.

Livak, K. J. and T. D. Schmittgen (2001). "Analysis of relative gene expression data using real-time quantitative

PCR and the 2(-Delta Delta C(T)) Method." Methods 25(4): 402-8.

Longo, D. L., P. L. Duffey, et al. (1999). "Conditioned immune response to interferon-gamma in humans." Clin

Immunol 90(2): 173-81.

Lynch, M. A. (2002). "Interleukin-1 beta exerts a myriad of effects in the brain and in particular in the

hippocampus: analysis of some of these actions." Vitam Horm 64: 185-219.

Madden, K. S. and D. L. Felten (1995). "Experimental basis for neural-immune interactions." Physiol Rev 75(1):

77-106.

Maier, S. F. (2003). "Bi-directional immune-brain communication: Implications for understanding stress, pain,

and cognition." Brain Behav Immun 17(2): 69-85.


36

Matsumoto, Y., M. Yoshida, et al. (2001). "Involvement of cholinergic and glutamatergic functions in working

memory impairment induced by interleukin-1beta in rats." Eur J Pharmacol 430(2-3): 283-8.

Miranda, M. I., R. T. LaLumiere, et al. (2003). "Blockade of noradrenergic receptors in the basolateral amygdala

impairs taste memory." Eur J Neurosci 18(9): 2605-10.

Miranda, M. I., L. Ramirez-Lugo, et al. (2000). "Cortical cholinergic activity is related to the novelty of the

stimulus." Brain Res 882(1-2): 230-5.

Mormede, C., K. Palin, et al. (2004). "Conditioned taste aversion with lipopolysaccharide and peptidoglycan

does not activate cytokine gene expression in the spleen and hypothalamus of mice." Brain Behav

Immun 18(2): 186-200.

Niijima, A. (1996). "The afferent discharges from sensors for interleukin 1 beta in the hepatoportal system in

the anesthetized rat." J Auton Nerv Syst 61(3): 287-91.



Palkovits, M. (1973). "Isolated removal of hypothalamic or other brain nuclei of the rat." Brain Res 59: 449-50.

Pipe, R. K. (1990). "Hydrolytic enzymes associated with the granular haemocytes of the marine mussel Mytilus

edulis." Histochem J 22(11): 595-603.

Pitossi, F., A. del Rey, et al. (1997). "Induction of cytokine transcripts in the central nervous system and pituitary

following peripheral administration of endotoxin to mice." J Neurosci Res 48(4): 287-98.

Rachal Pugh, C., M. Fleshner, et al. (2001). "The immune system and memory consolidation: a role for the

cytokine IL-1beta." Neurosci Biobehav Rev 25(1): 29-41.

Romanovsky, A. A., C. T. Simons, et al. (1997). "The vagus nerve in the thermoregulatory response to systemic

inflammation." Am J Physiol 273(1 Pt 2): R407-13.

Sawada, M., N. Hara, et al. (1991). "Ionic mechanism of the outward current induced by extracellular ejection

of interleukin-1 onto identified neurons of Aplysia." Brain Res 545(1-2): 248-56.

Schneider, H., F. Pitossi, et al. (1998). "A neuromodulatory role of interleukin-1beta in the hippocampus." Proc

Natl Acad Sci U S A 95(13): 7778-83.

Sehic, E. and C. M. Blatteis (1996). "Blockade of lipopolysaccharide-induced fever by subdiaphragmatic

vagotomy in guinea pigs." Brain Res 726(1-2): 160-6.

Straub, R. H. and M. Schedlowski (2002). "Immunology and multimodal system interactions in health and

disease." Trends Immunol 23(3): 118-20.

Tazi, A., R. Dantzer, et al. (1988). "Interleukin-1 induces conditioned taste aversion in rats: a possible

explanation for its pituitary-adrenal stimulating activity." Brain Res 473(2): 369-71.

Tkacs, N. C. and J. Li (1999). "Immune stimulation induces Fos expression in brainstem amygdala afferents."

Brain Res Bull 48(2): 223-31.

Tracey, K. J. (2007). "Physiology and immunology of the cholinergic antiinflammatory pathway." J Clin Invest

117(2): 289-96.

Turnbull, A. V. and C. L. Rivier (1999). "Regulation of the hypothalamic-pituitary-adrenal axis by cytokines:

actions and mechanisms of action." Physiol Rev 79(1): 1-71.

Turrin, N. P., D. Gayle, et al. (2001). "Pro-inflammatory and anti-inflammatory cytokine mRNA induction in the

periphery and brain following intraperitoneal administration of bacterial lipopolysaccharide." Brain

Res Bull 54(4): 443-53.

Watkins, L. R., L. E. Goehler, et al. (1995). "Blockade of interleukin-1 induced hyperthermia by subdiaphragmatic

vagotomy: evidence for vagal mediation of immune-brain communication." Neurosci Lett 183(1-2):

27-31.


37

Williams, C. L., D. Men, et al. (2000). "The effects of noradrenergic activation of the nucleus tractus solitarius

on memory and in potentiating norepinephrine release in the amygdala." Behav Neurosci 114(6):

1131-44.

Xie, Y., S. Frede, et al. (2002). "Beta-adrenoceptor-induced inhibition of rat splenocyte proliferation: cytokine

gene transcription as the target of action." Immunobiology 206(4): 345-53.

Chapter 3: Memorized T lymphocyte response

38

Chapter 3.

Memorized T lymphocyte response

With Gustavo Pacheco-López, Raphael Doenlen, Harald Engler, Maj-Britt Niemi, Andrea

Engler, Annemieke Kavelaars, Cobi J. Heijnen and Manfred Schedlowski

As published in FASEB, (2009)

[Personal contribution to the work: Designing and performing the research, analysis of the

immunological data, and writing the manuscript]


39

Introduction

Experimental and clinical evidence demonstrate that activated peripheral

immune cells secrete soluble messengers such as cytokines, neurotransmitters and

prostaglandins, that can affect the activity of the central nervous system (CNS; Goehler,

Gaykema et al. 2000; Steinman 2004; Raison, Capuron et al. 2006; Dantzer, O'Connor et

al. 2008). Reciprocally, the CNS modulates the activity of immune cells in the periphery

via neural and humoral pathways (Downing and Miyan 2000; Wrona 2006). Conditioned

immunosuppression by CsA (CIS) has been recently documented in both rodents (Exton,

von Horsten et al. 1998; Niemi, Pacheco-Lopez et al. 2006) and humans (Goebel, Trebst et

al. 2002) employing the immunosuppressive drug cyclosporin A (CsA) as unconditioned

stimulus (US) and contingently paired with the distinctive taste of saccharin (SAC) as the

conditioned stimulus. In these studies, we were able to demonstrate that there is a

conditioned reduction in splenocyte proliferation, Th1-cytokine production and cytokine

mRNA expression after ex vivo mitogenic stimulation with concanvalin A. The clinical

relevance of CIS has been demonstrated in vivo showing prolonged heart allograft

survival (Exton, Schult et al. 1998) and attenuation of experimental autoimmune disease

and allergic responses (Klosterhalfen and Klosterhalfen 1990). Advances have also been

made in understanding the central processing and peripheral routes involved in

acquisition and evocation of CIS. The sympathetic innervation of the spleen is essential

for eliciting the conditioned effects on splenocytes (Exton, von Horsten et al. 1998; Xie,

Frede et al. 2002). In addition, the insular cortex (IC) and the ventromedial hypothalamic

(VMH) nucleus are indispensable forebrain structures (Pacheco-Lopez, Niemi et al. 2005)

which are intensively involved in sympathetic transmission of the conditioned response

via the splenic nerve during the evocation phase. However, the specificity of CIS as well as

the cellular mediatory signaling pathways in lymphocytes are unknown.

Calcineurin (CaN) inhibitors, like CsA, play a crucial role in the treatment of graft

rejection (Ho, Clipstone et al. 1996). CsA, produced by the fungus Tolypocladium inflatum

Gams, is effective both in the prevention and in the treatment of acute allograft rejection

crisis and the treatment of autoimmune diseases like psoriasis. The mechanism of action

of CsA shows that in T lymphocytes CsA binds to a specific protein called immunophilin, a

cis-trans isomerase. The CsA-immunophilin complex binds to and blocks the

Ca2+/Calmodulin-dependent phosphatase CaN (Halloran 2001). The latter is required for

the translocation of the nuclear factor of activated T cells (NF-ATc) from the cytosol to the

nucleus, where it would normally bind to and activate enhancers/promoters of early T

lymphocyte activation factors such as IL-2 (Liu 2005). In the presence of CsA, the cytosolic

activation factor NF-ATc is unable to reach the nucleus, and transcription of IL-2 (and


40

other T lymphocyte activation factors) is strongly inhibited. As a result, T lymphocytes do

not proliferate and do not expand clonally (Ho, Clipstone et al. 1996; Halloran 2001).

To investigate the cellular mechanisms of CIS, CaN activity in the spleen, Th1- and

Th2-cytokine production and proliferation of splenocytes after stimulation with the

specific T cell stimulus anti-CD3 were assessed ex vivo.


Animals


Harlan Netherland (Horst, Netherlands) and were individually housed under an inverted

12:12h light/dark schedule (lights off at 7am) with food and water available ad libitum.

The animals were allowed to acclimatize to the new surroundings for two weeks before

initiation of any experimental procedure. All procedures were approved by the Swiss

Cantonal Veterinary Office, and are in agreement with the Principles of Laboratory

Animal Care (NIH publication No. 86-23, revised 1985).

Behavioral protocol

Five days prior to the first acquisition trial, animals were placed on a water

deprivation regimen, allowing them 15 min of drinking at 8 am and 5 pm each day. On

experimental day 6, conditioning started with 72h intervals between acquisition trials,

and 24h intervals between recall trials (Table 3.1). The period between the last acquisition

trial and the first evocation trial for all groups was 72h for all groups. In the morning of

acquisition days, conditioned animals (C) and conditioned not-evoked animals (CNE)

received a 0.2% sodium-saccharin (Sigma, Germany) as conditioned stimulus.

Immediately after the bottles were removed, animals received an i.p. injection of

20mg/kg CsA (Sandimmune ®; Novartis, Basel, Switzerland). In the afternoon session,

water (tap water) was paired with an i.p. saline injection (NaCl 0.9%, B. Braun Medical AG,

Emmenbrücke, Switzerland). Non-contingent conditioned rats (NCC) received the same

stimuli in a non-contingent manner: water paired with a CsA injection in the morning

session and saccharin in combination with saline injection in the afternoon session of the

three acquisition days. During evocation trials conditioned rats were re-exposed to

saccharin in the morning sessions and water in the afternoon sessions, whereas the CNE

group received water in both sessions. The order was reversed for the NCC group, which

received water in the morning and SAC presentation in the afternoon. In addition, a

pharmacological control (CsA) group was included, which was treated similarly to the


41

NCC group, but received an additional CsA injection following water presentation during

morning sessions of evocation days. This group allowed a comparison between the

conditioned immune effect and the pharmacological effect. Finally, a group of untreated

rats (N) was included. These animals remained unmanipulated during the whole

conditioning procedure, except for being subjected to the water deprivation regimen. On

the last recall day, one hour after finishing the morning drinking session the animals

were sacrificed under deep anesthesia (Isoflurane, Attane™, Mirnrad Inc, N.Y., USA), blood

was collected by cardiac puncture and spleens were aseptically removed.

Splenocyte isolation

Spleens were asepctically removed and transferred to tubes containing 10ml

Hanks buffered salt solution (HBSS, Invitrogen, Carlsbad, CA, USA). Splenocyte

suspensions were obtained by disrupting the tissue in 10ml ice-cold medium containing

10% fetal calf serum (Invitrogen Corporation) and 50µg/ml gentamycin (Invitrogen

Corporation) in a petri-dish with a syringe plunger. Splenocytes were washed once with

medium. Red blood cells were removed by a 3-min lysis in diluted PharM Lyse (BD

Pharmingen, Basel, Switzerland). Subsequently, cells were washed twice and filtered

through a 70-µM cell strainer (BD Pharmingen). Cell concentrations were determined

with an automatic cell counter (Vet abc™, Medical solution GmbH, Steinhausen,

Switzerland) and adjusted to 5 x 106 cells/ml.

Proliferation Assay

Splenocytes (2.5 x 105 cells) were stimulated in 96-well flat-bottom plates with

1µg/ml soluble anti-CD3 mAb (clone: G4.18; BD Pharmingen) for 72h in a humified

incubator (37°C, 5% CO2). Cell proliferation was measured using a commercially available

colorimetric assay (CellTiter96 Aqueous One, Promega, Dübendorf, Switzerland). Twenty

microliter of CellTiter96 AQueous One Solution Reagent were added to each well of the

96-well microtiter plates for the last 4h of incubation and absorbance was measured at

490nm. Dehydrogenase enzymes in viable, proliferating cells convert the tetrazolium

Table 3.1: Conditioning protocol consisting of 3 acquisition and 3 evocation trials. Saccharin solution (0.2%, SAC) was employed as conditioned stimulus and CsA (20mg/kg; i.p.) as unconditioned stimulus. The stimuli presented to induce or to control for a conditioned response are illustrated (for details see Methods section). Water: Wat. Acquisition Evocation Experimental groups Morning Afternoon Morning Afternoon Conditioned (C) SAC/CsA Wat/NaCl SAC Wat Conditioned not evoked (CNE) SAC/CsA Wat/NaCl Wat Wat Non-contingent conditioned (NCC) Wat/CsA SAC/NaCl Wat SAC Pharmacological control (CsA) CsA Untreated (N) Wat Wat Wat Wat


42

substrate to the brown, water-soluble product formazan that can be measured by

absorbance at 492nm using an automatic microplate reader. The quantity of formazan as

measured by the amount of absorbance is directly proportional to the number of living

cells in the culture. All samples were assayed in triplicate and data were illustrated as :

the difference between the signal from wells containing stimulated splenocytes and the

signal from wells containing unstimulated splenocytes.

FACS analysis

Total cell counts were obtained with an automatic cell counter. Splenocytes (2.5 x

105 cells) and peripheral blood cells (20µl) were stained with appropriately diluted CD45

(biotinylated, clone OX-1, with streptavidin-PE Texas-Red), CD3 FITC (clone 1F4), CD4 PE

(clone OX-35), CD8a PerCP (clone OX-8), CD161a Alexa647 (clone 10/78) CD45RA PE (clone

OX-33) and CD172a FITC (clone ED9) for 45 min at 4°C. All monoclonal antibodies were

obtained from BD Pharmingen (Allschwil, Switzerland) or AbD Serotec (Düsseldorf,

Germany). For red blood cell lysis, samples were treated with BD FACS Lysing solution (BD

Pharmingen) according to the manufacturer’s instructions. Ten-thousand leukocytes

from each sample were gated based on forward and sideward scatter characteristics and

CD45. Monocytes/macrophages and neutrophils were identified by forward and side

characteristics and differences in CD172a expression. Data were analyzed on a flow

cytometer (LSR II, BD Biosciences, San Jose, CA, USA) using FACS DIVA software. The

percentage of cells for a given cell type is the percentage of positive cells among 10 000

lymphocytes gated for the cell lineage specific markers.

Cytokine production

Concentrations of Th1- and Th2- cytokines in the culture supernatants of anti-CD3-

stimulated and unstimulated splenocytes were measured using commercially available

ELISAs (IL-2 and IFN-γ from Invitrogen; IL-4 from BD Pharmingen) according to the

manufacturer’s instructions. Briefly, 96-well plates (MaxiSorp, Nunc, Rosklide, Denmark)

were coated with purified mouse anti-rat mAb to IL-2, IFN-γ or IL-4 overnight and blocked

for 2h at room temperature (RT). Standards or undiluted test samples were incubated for

2h at RT. After washing biotinylated rat anti-mouse IL-2, IFN- γ and IL-4 were added for 1h

at RT. Following an additional washing step, appropriately diluted streptavidin horse-

radish peroxiadse in assay buffer was added. Thereafter, the chromogene

tetramethylbenzidine (TMB) was added. The reaction was stopped after 30 min by

addition of stopping solution (H2SO4). Optical density was measured at 492nm. Each

standard was expressed as means of duplicates per 96-well plate. Linearity was obtained

by fitting all data points with a log-log curve fit algorithm using Softmax Pro software 5.1


43

(Molecular Devices Corporation, Sunnyvale, CA, USA). Detection limits were 15.6pg/ml for

IL-2, 21.8pg/ml for IFN-γ and 1.5pg/ml for IL-4.

Protein determination

Protein concentration in splenic lysates for CaN activity analysis was measured

using a BCA™ Protein Assay (Pierce, Rockford, IL, USA) according to the manufacturer’s

instructions. (Smith, Krohn et al. 1985).

For Western Blot analysis, protein concentration was determined by the method

of Bradford using the Bio-Rad protein assay reagent (Bio-Rad Laboratories, Hercules, CA,

USA) and bovine serum albumin (BSA) as standard (Bradford 1976).

CaN activity

Cellular CaN phosphate activity in rat splenic lysates was assessed following the

manufacturer’s instructions (EMD chemicals, San Diego, CA, USA). Briefly, spleens were

disrupted with a rotor/stator tissue homogenizer (Tissue Terror, BioSpec Products Inc.,

Bartlesville, OK, USA) in lysing buffer supplemented with a mixture of protease inhibitors.

Lysates were centrifuged at 16000 x g for 1 h at 4°C and free phosphate was eliminated

by passing the supernatant through a chromatography column. Effective removal of

phosphate was qualitatively tested by the addition of Green reagent to the flow-through.

To determine CaN activity, 5 µl of flow through was incubated with RII phosphopeptide, a

known substrate for CaN, and the free phosphate released was detected based on the

Malachite green assay (Martin, Pallen et al. 1985). Since EGTA inhibits calcineurin, total

CaN phosphatase activity was reported as the difference between phosphatase activity

(Total) and phosphatase activity (EGTA buffer).

Western Blot

Splenic cells were sonicated twice for 15s in ice-cold lysis RIPA buffer (20mM

Hepes, pH 7.5, 1% Triton X-100, 150mm NaCl, 10 mM EDTA, 2mM 4-(2-aminoethyl)

benzenesulphonyl fluoride (AEBSF), 20µg/mL leupeptin and 200µg/mL benzamidine) and

lysates were clarified at 15000 x g for 30min at 4°C. The supernatants were collected and

the protein concentration was determined using the Bradford assay. 50µg of lysates were

boiled for 5min in the SDS sample buffer and were analyzed by SDS-PAGE in a 10%

separating gel. After electrophoresis, the proteins were transferred to Hybond-P PVDF

membranes (Amersham, Uppsala, Sweden) by electroblotting. Efficiency of transfer was

verified by Ponceau red staining. Membranes were blocked overnight in TBS-Tween

containing 5% of non-fat milk and then incubated with a specific antibody (1:1000)

against the catalytic alpha subunit of CaN (Cell Signaling Technology, Danvers, MA, USA).


44

Membranes were washed three times with Tris-buffered saline and incubated with

peroxidase-conjugated donkey anti-rabbit IgG (1:5000, Amersham, Buckinghamshire, UK).

Blots were developed with an ECL kit from Amersham (Amersham) and bands were

visualized after exposing blots to X-ray films. Band density was determined using a GS-

700 Imaging Densitometer and analyzed by Molecular Analyst Software v. 1.5 (Bio-Rad

Laboratories, CA, USA). Quantification of CaN was normalized to the amount of β-actin

present.


All data were analyzed using parametric analysis of variance (ANOVA) followed by

Tukey’s HSD post-hoc test whenever a main effect attained statistical significance.

Statistical significance was set at P < 0.05. All statistical analyses were carried out using

the statistical software SPSS for Windows (version 14.0).

Results

Fluid intake.

During the first encounter with the novel taste of SAC, rats initially elicited a

characteristic neophobic behavior. This stereotypic rodent behavior was displayed by all

animals exposed to the SAC (Figure 3.1, 1st association trial (A1)). Stimuli contingency

allowed conditioned animals (C and CNE groups) to associate the SAC with CsA

administration. Clear avoidance behavior was elicited in those groups by later re-

presentations of the SAC. In contrast, no associative learning occurred when the same

stimuli were inverted and the inter-stimuli interval was extended by 24h (NCC group).

Therefore, contingent pairing(s) of the SAC with the CsA during association led to the

establishment of a temporal/causal association of both stimuli, stored within the CNS, as

shown by the strong resistance to extinction (C group) during the evocation phase

(Figure 3.1). This is in agreement with previous reports (Exton, Von Horsten et al. 1998).


45

Figure 3.1: Fluid consumption (water, black symbols; saccharin, grey symbols; A, acquisition; E, evocation) of experimental and control groups during acquisition and evocation phases (Conditioned group: square, n=16; conditioned not-evoked group: triangle downwards, n=14; non-contigent conditioned group: circle, n=14; CsA-treated group: hex, n=14; untreated group: triangle upwards, n=12). Statistics: one-way ANOVA followed by Tukey’s HSD; *p< 0.05 vs. conditioned not-evoked and non-contigent conditioned control groups.

Ex vivo T cell cytokine production and proliferation.

CsA exerts its immunosuppressive properties on T lymphocytes via specific

inhibition of the transcription of early T lymphocyte activation genes such as the Th1-

cytokines IL-2 and IFN-γ. Concomitantly, T lymphocyte activation and clonal expansion is

potently inhibited (Ho, Clipstone et al. 1996). In contrast, it is known that Th2-cytokines

like IL-4 are not affected by CsA treatment (Barten, Rahmel et al. 2006). In order to verify

the CIS, IL-2, IFN-γ and IL-4 production of rat splenocytes 48h after ex vivo stimulation

with anti-CD3 mAb, which the specificically stimulates T cell in the culture, was assessed

(Figure 3.2A-C).

As expected, CsA treated animals (CsA group) that allowed us a comparison between the

conditioned effect and the real drug effect showed the expected reduction in IL-2 and

IFN-γ production (p<0.05: Figure 3.2A-B). More importantly though, conditioned animals

(C group) also displayed a pronounced and significant reduction in IL-2 and IFN-γ

compared to non-contingent conditioned or conditioned not-evoked animals (p<0.05;

NCC and CNE groups, respectively) which received exactly the same pharmacological

treatment differing however in the conditioning assignment. Also important is the fact

that the behavioral conditioned effects on cytokine production and cell proliferation were

comparable in magnitude and direction to the effects elicited by the administration of

CsA alone. Moreover, production of the Th2-cytokine, IL-4, remained unchanged in all

groups (Figure 3.2C). In addition, the changes in cytokine production were not due to


46

alterations in T lymphocyte subpopulations in the spleen as confirmed by 4-color flow

cytometry (Table 3.2).

Tabel 3.2: FACS Phenotyping of splenocytes and peripheral blood cells. Statistics: one-way Anova followed by Tukey`s HSD. *p<0.05 vs NCC group, # p<0.05 vs CsA group. Abbreviations: C, conditioned group; NCC, non-contigent conditioned group; CNE, conditioned not-evoked group; CsA, CsA treated group; N, untreated group) Experimental groups C NCC CNE CsA N Blood Myeloid cells 36.1 ± 1.4# 39.7 ± 1.6 39.6 ± 1.7 43.9 ± 2.1 23.8 ± 1.1*,# Granuolcytes 14.0 ± 0.9 19.5 ± 0.9 19.1 ± 2.1 22.7 ± 0.9 15.1 ± 1.1 Monocytes/Macrophages 21.5 ± 0.9 19.8 ± 1.2 20.1 ± 0.9 20.6 ± 2.5 8.4 ± 0.4*

T lymphocytes 43.6 ± 1.6 44.0 ± 1.7 40.1 ± 1.6 39.2 ± 1.9 60.3 ± 1.3* CD4+ T lymphocytes 33.7 ± 1.4 34.2 ± 1.4 31.1 ± 1.2 30.5 ± 1.5 46.1 ± 1.3* CD8+ T lymphocytes 9.4 ± 0.4 9.5 ± 0.3 9.0 ± 1.7 8.6 ± 0.4 14.1 ± 0.3*

B lymphocytes 9.3 ± 0.4# 9.8 ± 0.4# 10.5 ± 0.3# 7.3 ± 0.6* 11.1 ± 0.3 Natural killer cells 7.5 ± 0.6# 6.3 ± 0.9 5.5 ± 0.5 4.6 ± 0.6 3.4 ± 0.5*

Spleen Myeloid cells 24.6 ± 1.1 24.3 ± 1.2 21.1 ± 1.6 22.2 ± 1.6 17.0 ± 0.5 Granuolcytes 3.6 ± 0.6 4.0 ± 0.6 4.2 ± 1.0 4.2 ± 1.1 3.1 ± 0.3 Monocytes/Macrophages 18.8 ± 0.8* 17.9 ± 0.8 15.2 ± 0.7 14.9 ± 0.6* 11.5 ± 0.7*

T lymphocytes 39.1 ± 1.0 38.1 ± 1.3 39.9 ± 0.9 39.8 ± 0.8 46.7 ± 1.0*,#

CD4+ T lymphocytes 28.3 ± 0.7 27.7 ± 0.9 28.9 ± 5.3 28.3 ± 0.7 34.6 ± 0.7* CD8+ T lymphocytes 9.0 ± 0.5 9.4 ± 0.7 10.6 ± 6.4 11.0 ± 0.5 11.0 ± 0.6

B lymphocytes 22.5 ± 0.8 21.6 ± 1.0 24.8 ± 0.5 23.9 ± 0.7 22.2 ± 0.9 Natural killer cells 5.3 ± 0.3# 4.8 ± 0.3 5.6 ± 0.2# 4.3 ± 0.3* 3.9 ± 0.1*


47

Figure 3.2: (A, B) Th

1- and (C) Th

2- cytokine production 48h after ex vivo anti-CD3 stimulation. Statistics: one-

way ANOVA followed by Tukey’s HSD; *p< 0.05 vs. CNE and NCC control groups. Abbreviations: N, untreated group; CsA, CsA treated group; C, conditioned group; NCC, non-contigent conditioned group; CNE, conditioned not-evoked group)

In addition to cytokine production, ex vivo proliferation of anti-CD3 stimulated T cell was

dtermined. Both conditioned animals and pharmacological control animals showed a


48

pronounced and significant reduction in T cell proliferation (p<0.05; Figure 3.3).

Importantly, the conditioned effects on cell proliferation and cytokine production were

independent of any residual CsA effect as shown by the CNE and NCC groups. Even

though receiving the same pharmacological treatment as C group, CNE and NCC groups

neither showed a reduction in Th1-cytokines nor in proliferation.

Figure 3.3: T lymphocyte proliferation 72h after ex vivo anti-CD3 stimulation. Statistics: one-way ANOVA followed by Tukey’s HSD; *p< 0.05 vs. conditioned not-evoked and non-contigent conditioned control groups; #p<0.05 vs. conditioned, conditioned not-evoked and non-contingent conditioned group.

Figure 3.4: Downstream signalling pathways induced after T cell receptor stimulation with sites of action of cyclosporin A and rapamycin. Reprinted from Pattison 1997.


49

CaN activity and CaNα protein expression.

CsA exerts its immunosuppressive actions specifically on T lymphocytes via

inhibition of the Ca2+/Calmodulin-dependent protein phosphate CaN (Figure 3.4; Ho,

Clipstone et al. 1996; Halloran 2001). To further explore the cellular mechanism and to

elucidate the aspect of specificity of CIS, we assessed both CaN activity and expression of

the catalytic subunit α of CaN (CaNα) in splenic homogenates. As expected, the

pharmacological control displayed a strong reduction in splenic CaN activity (p<0.05;

Figure 3.5). Importantly, conditioned animals showed a pronounced and statistically

significant reduction in CaN activity (p<0.05) compared to behavioral controls (CNE/NCC

groups).

Figure 3.5: CaN phosphatase activity in splenic homogenates 1h after the 3rd evocation trial. N: untreated group; CsA: pharmacological control group; C: conditioned group; NCC: Non-contingent conditioned group¸ CNE: conditioned not-evoked group. Statistics: one-way ANOVA followed by Tukey’s HSD; *p< 0.05 vs. CNE and NCC control groups.

Western blot analysis of the expression of the catalytic subunit alpha (CaNα) confirmed

that the observed changes in CaN activity in both CsA-treated and conditioned animals

were not due to alterations in CaN protein expression (Figure 3.6). Therefore, these data

clearly illustrate that the conditioned response is specifically mediated intracellularly via

the inhibition of CaN. In addition, these data suggest that CIS seems to be mediated by

the same mechanism as pharmological CsA treatment, CaN inhibition.

In summary, both CsA treatment as well as re-exposure to the SAC after

behavioral conditioning suppresses T lymphocyte activation and cytokine production to

the same magnitude and in the same direction upon ex vivo anti-CD3. In addition, CaN

was identified for the first time as intracellular molecule mediating CIS. These results

demonstrate that the immunosuppressive mechanisms of CsA are mimicked during CIS,

highlighting the specificity of the underlying association (i.e. taste-CaN inhibiton).


50

42 kDa - ββββ-actin

61 kDa - CaNαααα

N CsA C NCC CNE

Figure 3.6: Normalized expression of the catalytic subunit CaNα in splenic homogenates 1h after the 3rd evocation trial. N: untreated group; CsA: pharmacological control group; C: conditioned group; NCC: Non-contingent conditioned group¸ CNE: conditioned not-evoked group. Statistics: one-way ANOVA followed by Tukey’s HSD.

Discussion

For more than 30 years, the functionality of CNS-immune system interaction has

been extensively documented by behavioral studies describing how immunosuppression

can be elicited by behavioral conditioning (Ader and Cohen 2001; Exton, Herklotz et al.

2001; Riether, Doenlen et al. 2008). However, little is known about the intracellular

mechanism of CIS. Employing the immunosuppressive drug CsA as the US within a

rodent taste-associative learning paradigm, the present data indicate that CaN is the

common intracellular target for both the pharmacologic effects of CsA and the

conditioned response in splenocytes. These data identify for the first time a fundamental

mediator of CIS.

Upon T cell receptor stimulation, the Ca2+/Calmodulin-dependent phosphatase

CaN becomes activated. The latter is required for the translocation of NF-ATc from the

cytosol to the nucleus, where it binds to enhancers/promoters of early T lymphocyte

activation factors and induces the clonal expansion of T cells (Figure 3.4; Liu 2005). The

data presented here are the first to document splenic CaN modulation by conditioning.


51

After the initial associative learning process, i.e., the repeated pairing of SAC and CsA, re-

exposure to the CS alone resulted in a strong inhibition of the CaN activity in splenocytes.

Consequently, anti-CD3-stimulated splenocytes from conditioned animals showed

impaired Th1-cytokine production and T cell proliferation. Importantly, the magnitude and

the direction of the observed effects were comparable to the pharmacological effects

induced by CsA.

Immunocompetent cells, like many others, can be neurally regulated (Downing

and Miyan 2000). Under hypothalamic control (Saito, Minokoshi et al. 1989), splenic

noradrenergic fibers innervate the periarteriolar sheath, an area densely populated by T

lymphocytes (Felten, Felten et al. 1987; Madden and Felten 1995). The presence of

adrenergic receptors, especially β-adrenoceptors, on lymphoid cells has been well

documented, and the stimulation of these receptors can significantly affect an immune

response (Elenkov, Wilder et al. 2000; Kohm and Sanders 2000). For instance, after brief

VMH stimulation rat splenocyte proliferation was significantly reduced, with this effect

being blocked by both surgical denervation of the spleen as well as β-adrenoceptor

antagonist administration (Okamoto, Ibaraki et al. 1996). The IC and the VMH are brain

structures actively involved in the output pathway to the immune system in CIS, since

their lesion before the evocation phase blocked the expression of the conditioned

immune response (Pacheco-Lopez, Niemi et al. 2005). Surgical denervation of the spleen

prior to association phase, or chemical sympathectomy before evocation phase

completely disrupts CIS (Exton, von Horsten et al. 1998; Exton, Gierse et al. 2002), clearly

indicating that behavioral conditioning of splenocyte function is mainly regulated in the

periphery via the sympathetic nervous system. Continuous administration of the β-

adrenoceptor antagonist propranolol during the evocation phase abrogated CIS (Exton,

Gierse et al. 2002). Furthermore, splenocyte proliferation, cytokine production, and

cytokine mRNA expression were markedly reduced after ex vivo stimulation with the β-

adrenoceptor agonist isoproterenol, ex vivo, with these effects being blocked by addition

of the β-adrenoceptor antagonist propranolol (Exton, Gierse et al. 2002; Xie, Frede et al.

2002).

The data shown here contribute significantly to the understanding of CIS and

document for the first time the conditioned response specifically suppresses splenic T

lymphocyte activity by targeting the common intracellular mediator CaN. Thereby, CIS

uses the same mediatory mechanisms as pharmacologic CsA treatment.

CaN activity was significantly inhibited in splenic homogenates from conditioned

animals after the last evocation trial. Similar to the CsA-treated animals, splenocytes

from conditioned animals showed reduced Th1- but not Th2-cytokine production and low

levels of T cell proliferation. In addition, changes in Th1- cytokine production and T


52

lymphocyte proliferation were not evoked by a stress-related release of glucocorticods,

since plasma corticosterone levels were unaffected at evocation phase (data not shown,

elsewhere reported) (Exton, von Horsten et al. 1998). Furthermore, stress- related changes

in peripheral blood cell composition were not observed during conditioning (see table 3.2)

and were consequently attributable to the mere re-exposure to the SAC as confirmed by

a NCC and a CNE group. These results point out that both conditioned as well as

pharmacological suppression of Th1- like cytokines and T lymphocyte proliferation by CsA

is specifically affecting T lymphocytes on a cellular level. Our findings document for the

first time that the activation of the “neuro-immune axis” (Pacheco-Lopez, Niemi et al.

2005) by behavioral conditioning modulates T lymphocyte function in the spleen by

reducing the phosphatase activity of CaN. However, the exact molecular mechanism by

which CaN activity in T lymphocytes is affected via the sympathetic nervous system has

still to be elucidated. In this regard, endogenous CaN inhibitors (Shibasaki, Hallin et al.

2002; Liu 2003) forming dynamic complexes with β2-adrenoceptors, protein kinases and

phosphatases, including CaN may be potential mediatiors of the observed CaN inhibition

in conditioned animals (Shih, Lin et al. 1999).

The findings of the present study demonstrate that, through contingent pairing, a

formerly neutral taste can be associated with through contingent pairings the specific

pharmacologic properties of CsA can be transferred to an immune-neutral taste. Two

major alternatives, not mutually exclusive, may explain this phenomenon. The first

possibility is fully consistent with behavioral conditioning. The association between

“sensory” (SAC taste) and immunological perturbations (immunosuppression by CaN

inhibition) is encoded and stored in the brain, inducing a specific and independent

engram, which afterwards can be selectively activated by re-exposure to the given taste.

An alternative model proposes that neurally-derived signals triggered by gustatory

stimulation may transmit just a part of the pattern of the engram to selective immune

cells in the lymphoid tissue at acquisition time (Grossman, Herberman et al. 1992). Our

data support associative specificity and at the same time control for residual effects of

the CsA. However, conclusive evidence can only be provided by experiments specifically

designed to assess passive forgetting and from discriminative conditioning paradigms

(Ader 2003).

In summary, the results demonstrate that through behavioral conditioning the

particular pharmacologic properties of CsA can be transferred to an immune-neutral

taste. In this study CaN activity was inhibited in splenocytes from conditioned rats by a

specific gustatory stimulation. Importantly, the behaviorally conditioned effects were

comparable in magnitude and direction to the pharmacologic actions of CsA alone,

resulting in analog immunosuppression as illustrated by reduced T lymphocyte


53

proliferation and cytokine production. For the first time, these data identify CaN as the

intracellular target molecule of CIS, highlighting the specificity of the underlying

association (i.e. taste-CaN inhibition), as well as contributing to our understanding of the

mechanisms behind “learned placebo effects”.

References Ader, R. (2003). "Conditioned immunomodulation: research needs and directions." Brain Behav Immun 17

Suppl 1: S51-7.

Ader, R. and N. Cohen (2001). Psychoneuroimmunology. R. Ader, D. Felten and N. Cohen. New York, Academic

Press. 2: 3-34.

Barten, M. J., A. Rahmel, et al. (2006). "Cytokine analysis to predict immunosuppression." Cytometry A 69(3):

155-7.

Bradford, M. M. (1976). "A rapid and sensitive method for the quantitation of microgram quantities of protein

utilizing the principle of protein-dye binding." Anal Biochem 72: 248-54.

Dantzer, R., J. C. O'Connor, et al. (2008). "From inflammation to sickness and depression: when the immune

system subjugates the brain." Nat Rev Neurosci 9(1): 46-56.

Downing, J. E. and J. A. Miyan (2000). "Neural immunoregulation: emerging roles for nerves in immune

homeostasis and disease." Immunol Today 21(6): 281-9.

Elenkov, I. J., R. L. Wilder, et al. (2000). "The sympathetic nerve--an integrative interface between two

supersystems: the brain and the immune system." Pharmacol Rev 52(4): 595-638.





Exton, M. S., M. Schult, et al. (1998). "Behavioral conditioning prolongs heart allograft survival in rats."

Transplant Proc 30(5): 2033.



182-91.

Exton, M. S., S. Von Horsten, et al. (1998). "Conditioned taste aversion produced by cyclosporine A:

concomitant reduction in lymphoid organ weight and splenocyte proliferation." Physiol Behav 63(2):

241-7.

Felten, D. L., S. Y. Felten, et al. (1987). "Noradrenergic sympathetic neural interactions with the immune

system: structure and function." Immunol Rev 100: 225-60.

Goebel, M. U., A. E. Trebst, et al. (2002). "Behavioral conditioning of immunosuppression is possible in

humans." Faseb J 16(14): 1869-73.

Goehler, L. E., R. P. Gaykema, et al. (2000). "Vagal immune-to-brain communication: a visceral chemosensory

pathway." Auton Neurosci 85(1-3): 49-59.

Grossman, Z., R. B. Herberman, et al. (1992). "Neural modulation of immunity: conditioning phenomena and

the adaptability of lymphoid cells." Int J Neurosci 64(1-4): 275-90.



54

Ho, S., N. Clipstone, et al. (1996). "The mechanism of action of cyclosporin A and FK506." Clin Immunol

Immunopathol 80(3 Pt 2): S40-5.

Klosterhalfen, S. and W. Klosterhalfen (1990). "Conditioned cyclosporine effects but not conditioned taste

aversion in immunized rats." Behav Neurosci 104(5): 716-24.

Kohm, A. P. and V. M. Sanders (2000). "Norepinephrine: a messenger from the brain to the immune system."

Immunol Today 21(11): 539 - 42.

Liu, J. O. (2003). "Endogenous protein inhibitors of calcineurin." Biochem Biophys Res Commun 311(4): 1103-9.

Liu, J. O. (2005). "The yins of T cell activation." Sci STKE 2005(265): re1.

Madden, K. S. and D. L. Felten (1995). "Experimental basis for neural-immune interactions." Physiol Rev 75(1):

77-106.

Martin, B., C. J. Pallen, et al. (1985). "Use of fluorinated tyrosine phosphates to probe the substrate specificity

of the low molecular weight phosphatase activity of calcineurin." J Biol Chem 260(28): 14932-7.

Niemi, M. B., G. Pacheco-Lopez, et al. (2006). "Murine taste-immune associative learning." Brain Behav Immun

20(6): 527-31.

Okamoto, S., K. Ibaraki, et al. (1996). "Ventromedial hypothalamus suppresses splenic lymphocyte activity

through sympathetic innervation." Brain Res 739(1-2): 308-13.

Pacheco-Lopez, G., M. Niemi, et al. (2007). Psychoneuroimmunology. R. Ader. New York, Acacdemic Press. 1:

631-660.



Raison, C. L., L. Capuron, et al. (2006). "Cytokines sing the blues: inflammation and the pathogenesis of

depression." Trends Immunol 27(1): 24-31.

Riether, C., R. Doenlen, et al. (2008). "Behavioural conditioning of immune functions: how the central nervous

system controls peripheral immune responses by evoking associativel earning processes." Reviews

in the Neurosciences 19: 1-17.

Saito, M., Y. Minokoshi, et al. (1989). "Accelerated norepinephrine turnover in peripheral tissues after

ventromedial hypothalamic stimulation in rats." Brain Res 481(2): 298-303.

Shibasaki, F., U. Hallin, et al. (2002). "Calcineurin as a multifunctional regulator." J Biochem (Tokyo) 131(1): 1-15.

Shih, M., F. Lin, et al. (1999). "Dynamic complexes of beta2-adrenergic receptors with protein kinases and

phosphatases and the role of gravin." J Biol Chem 274(3): 1588-95.

Smith, P. K., R. I. Krohn, et al. (1985). "Measurement of protein using bicinchoninic acid." Anal Biochem 150(1):

76-85.

Steinman, L. (2004). "Elaborate interactions between the immune and nervous systems." Nat Immunol 5(6):

575-81.

Wrona, D. (2006). "Neural-immune interactions: an integrative view of the bidirectional relationship between

the brain and immune systems." J Neuroimmunol 172(1-2): 38-58.



Chapter 4: β-adrenoceptor stimulation inhibits calcineurin activity in T lymphocytes via a PKA dependent mechanism

55

Chapter 4.

ββββ-adreneroceptor stimulation inhibits calcineurin activity in T

lymphocytes via a PKA-dependent mechanism


56

Introduction

Behaviorally conditioned immunosuppression by CsA was shown to be mediated

via neural innervation of the spleen (Exton, Gierse et al. 2002; Xie, Frede et al. 2002).

Surgical denervation of the spleen prior conditioning reduced the noradrenaline content

in the spleen and abrogated the conditioned effect indicating that behavioural

conditioning by CsA. In addition, it has been shown that conditioning by CsA is regulated

by the SNS and β-adrenergic mechanisms (Exton, Gierse et al. 2002; Xie, Frede et al.

2002).

These data suggest that the efferent communication between the CNS and the

immune system during conditioning primarily affects immunocompetent cells located in

the spleen. In addition, T lymphocytes were identified to be the mediator of conditioning

by CsA in the spleen and CaN inhibition in splenocytes to be responsible for the observed

effect (Pacheco-Lopez, Riether et al. 2009). However, the intracellular mediatory

mechanisms of CIS are still poorly understood.

Neural innervation of the spleen is predominantly sympathetic, with zones of T

lymphocytes and macrophages especially, richly innervated by noradrenergic fibers

(Mignini, Streccioni et al. 2003). In addition, it has been shown that naïve T cells

predominantly express β2-adrenergic receptors (Sanders, Baker et al. 1997; Sanders and

Straub 2002) and mediate cAMP-dependant inhibition of T lymphocyte proliferation and

IL-2 production after mitogen-induced stimulation (Kammer 1988; Bartik, Bauman et al.

1994; Tasken and Stokka 2006). Elevated intracellular levels of cAMP have an inhibitory

effect on T lymphocyte activity (Torgersen, Vang et al. 2002), keeping T lymphocytes in an

inactive state via the A-kinase anchoring protein (AKAP), predominantly via AKAP-150

(Williams 2002). AKAP-150 was shown to anchor and regulate PKA activity (Scott 2006). In

addition, AKAP-150 binds calcineurin, PLC and PKC and clusters the enzymes to control

signal transduction as specific intracellular compartments (Coghlan, Perrino et al. 1995;

Klauck, Faux et al. 1996).

Therefore, β-adrenergic mechanisms may be responsible along the efferent

pathway for the observed conditioned reduction in T lymphocyte proliferation and CaN

inhibition. Thus, the aim of this study was to investigate the effect of β-adrenergic

stimulation on CaN activity in T lymphocytes and to elucidate whether CaN-AKAP-150

interaction mediate the observed CaN inhibition in CIS in an ex vivo model.


57


Experimental animals


Harlan Netherland (Horst, Netherlands) and were individually housed under an inverted

12:12h light/dark schedule (lights off at 7am) with food and water available ad libitum.

The animals were allowed to acclimate to the new surroundings for 2 weeks before

initiation of any experimental procedure. All procedures were approved by the Swiss

Cantonal Veterinary Office, and are in agreement with the Principles of Laboratory

Animal Care (NIH publication No. 86-23, revised 1985).

Reagents

Terbutaline, Nadolol, H-89 and DBcAMP were purchased from Sigma-Aldrich

(Sigma, Buchs, Switzerland) and were dissolved and diluted in culture medium

immediately before addition to cultures. InCELLect® AKAP St-Ht31 inhibitor and St-Ht31P

control peptide were purchased from Promega (Promega, Dübendorf, Switzerland). CsA

was purchased from Novartis Pharma AG (Sandimmune™, Novartis, Basel, Switzerland).

Isolation of splenocytes

Spleens were aspectically removed and transferred to tubes containing 10ml

Hank’s buffered salt solution (HBSS, Invitrogen, Carlsbad, CA, USA). Splenocyte

suspensions were obtained by disrupting the tissue in 10ml ice-cold medium containing

10% fetal calf serum (Invitrogen Corporation) and 50µg/ml gentamycin (Invitrogen

Corporation) in a petri-dish with a syringe plunger. Splenocytes were washed once with

medium. Red blood cells were removed by a 3-min lysis in diluted PharM Lyse (BD

Pharmingen, Allschwil, Switzerland). Subsequently, cells were washed twice and filtered

through a 70-µm cell strainer (BD Pharmingen) before CD4+ T lymphocyte isolation.

Isolation of CD4+ T lymphocytes

Purified CD4+ T lymphocytes were obtained using a two-step magnetic cell

separation (MACS) protocol. First, CD4+ monocytes/macrophages and B cells were

eliminated by positive selection using indirect magnetic labelling with purified mouse

anti-rat CD172a (clone ED9, isotype IgG1, AbD Serotec), anti-rat CD45RA (clone OX-33,

istoype IgG1, BD Pharmingen) and rat anti-mouse IgG1 MicroBeads (Miltenyi Biotec,

Bergisch Gladbach, Germany). Second, T helper cells were positively selected from the

CD172a/CD45RA-depleted splenocytes using rat CD4 MicroBeads (Miltenyi Biotec). The


58

purity of the isolated CD4+ T lympocytes is typically between 92 and 95%, as verified by

flow cytometry (Figure 4.1, LSR II, BD Immunocytometry Systems, San Jose, CA, USA).

Figure 4.1: FACS staining of CD4+T cells before and after purification To analyze the degree of T cell purification, splenocytes (A, 2.5 x 105) and isolated T cells (B, 2.5 x 105) were stained with APC conjugated anti-rat CD3, PE conjugated CD45, FITC conjugated CD4 and PerCP conjugated CD8a before and after a two step T lymphocyte selection, respectively. The purity of the isolated CD4+ T lymphocytes is typically between 92-95% (see Table 4.1)

Anti-CD3/CD28 stimulation

Isolated CD4+ T lymphocytes (5 x 105) were stimulated for 72h with plate-bound

mouse anti-rat CD3 (clone G4.18, 10mg/ml, BD Pharmingen) and soluble mouse anti-rat

CD28 (clone JJ319, 1mg/ml, BD Pharmingen) in T cell activation plates (BD Pharmingen) in

the presence and absence of indicated concentrations of terbutaline, nadolol, CsA, H-89,

DBcAMP, St-Ht31 peptide or St-Ht31P control peptide, in which two isoleucine residues

have been replaced by proline residues, thereby blocking its ability to disrupt the AKAP-

PKA complex. Supernatants were collected for cytokine measurements and cell

proliferation was measured by flow cytometry using the Click-iT™ EdU cell proliferation

assay (Invitrogen) according to the manufacturer’s instructions.

Cytokine measurements

Cytokine measurements for IL-2, IFN-y and IL-4 in supernatants were performed

using a multiplexed bead-based assay (Bio-Plex Cytokine Assays, Biorad, Hercules, CA,

USA) according to the manufacturer’s instructions. Briefly, beads covalently coupled to

capture antibodies for IL-2, IFN-y and IL-4, respectively, were incubated with 20µl of

diluted cell culture supernatant at 4°C overnight. After washing with PBS containing

0.05% Tween-20, samples were incubated with the respective biotinylated detection

A B


59

antibodies under agitation on an orbital shaker at 450rpm for 2h RT. Samples were

washed, incubated with diluted streptavidin-PE (1:2000, BD Pharmingen) and washed

again. Re-suspended samples were analyzed on a flow cytometer (LSR II, BD Pharmingen)

using FACS DIVA software version 6.0. Results are expressed as the mean of triplicate

cultures ± SEM of at least two independent experiments. The calibration standards

ranged from 0.1 to 3200pg/ml. Each standard was expressed as the mean of duplicates

per 96-well plate. Linearity was obtained by fitting all data points with a log-log

regression using Softmax Pro 4.7.1 software (Molecular Devices, Sunnyvale, CA, USA).

Phospho-CREB/CREB measurements

Phospho-CREB (pCREB) and total CREB (CREB) in CD4+ T lymphocyte lysates were

analyzed by flow cytometry using the Bio-Plex™ Phosphoprotein Detection Kit (Biorad)

according to the manufacturer’s instructions. Briefly, treated and untreated CD4+ T cells (1

x 106) were harvested and washed with 400µl of ice-cold wash buffer. Supernatants were

removed and cells were lysed in a total volume of 200µl of lysing solution under agitation

on a microplate shaker at 300rpm for 20min at 4°C. Samples were centrifuged for 20min

at 4500 x g at 4°C and 20µl of lysates were incubated with beads covalently coupled to

antibodies directed against pCREB or CREB on an orbital shaker at 300rpm overnight. The

next day, samples were washed three times and incubated with the corresponding

biotinylated detection antibodies for 30min under agitation. After another three washes,

samples were incubated with streptavidin-PE for 10min and washed three times. Samples

were re-suspended and 200 events per sample were analyzed on a flow cytometer (LSR II,

BD Pharmingen) using FACS DIVA software version 6.0. Data are represented as the

percentage of median fluorescence of control samples ± SEM of at least two independent

experiments.

Flow cytometry

Phenotyping

To assess the degree of T cell purification before and after the two-step magnetic

cell separation, total splenocytes, CD172a/CD45RA-positive and CD172a/CD45RA-depleted

splenocytes as well as purified T helper cells were washed with PBS (Invitrogen)

containing 2% FCS (Invitrogen) and 0.1% NaN3 (Sigma, FASC buffer) and stained in a total

volume of 100µl with predetermined, appropriately diluted APC-conjugated mouse anti-

rat CD3 (clone: 1F4, BD Pharmingen), PE conjugated mouse anti-rat CD45 (clone OX-1, BD

Pharmingen), PerCP-conjugated CD8a (clone OX-8, BD Pharmingen) and FITC-conjugated

CD4 (clone OX-35, BD Pharmingen) in Panel 1 and appropriately diluted APC conjugated


60

mouse anti-rat CD45RA (clone: OX-33, BD Pharmingen), PE-conjugated mouse anti-rat

CD45 (clone OX-1, BD Pharmingen), Alexa647-conjugated CD161 (clone OX-78, Serotec)

and FITC-conjugated CD172a (clone ED9, Serotec) in Panel 2 for 45 min at 4°C. For RBC

lysis, samples were treated for 10min with 1:10 diluted BD FACS Lysing solution (BD

Pharmingen). Ten-thousand lymphocytes were gated on forward and sideward scatter

characteristics and the expression of CD45. Data were analyzed on a flow cytometer (LSR

II, BD Pharmingen) using FACS DIVA software version 6.0.

EdU proliferation

Proliferation of isolated T helper cells was measured by flow cytometry using the

Click-iT™ EdU cell proliferation assay according to the manufacturer’s instructions. Briefly,

isolated T helper cells were stimulated with plate-bound anti-CD3 and soluble anti-rat

CD28 in the presence and absence of different concentrations of terbuatline and nadolol

for 72h at 37°C and 5% CO2. The thymidine analog, 5-ethynyl-2´-deoxyuridine (EdU) was

added to the cells at a concentration of 10µM for the last 18h of culture. After harvesting,

cells were washed with FACS buffer. Supernatants were removed and cells were stained

in total volume of 100µl with predetermined, appropriately diluted APC conjugated

mouse anti-rat CD3 (clone: 1F4, BD Pharmingen) for 30min at 4°C. After 1 washing FACS

buffer cells were fixed in 100µl of Click-iT™ fixative for 15min at RT and washed again.

Cells were permeabilized in a volume of 100µl of Triton X-100-based permeabilization

reagent for 30min at RT, washed and incubated for 30min with 500µl of Click-iT™

reaction cocktail. After a final washing step, cells were re-suspended in 500µl FACS

buffer. Samples were analyzed using a LSR II flow cytometer and FACSDiVA Software

Version 6.0 (BD Pharmingen). Results are expressed as the percentage of proliferation

from anti-CD3/anti-CD28 stimulated cells (control) ± SEM of triplicate culture of at least

two independent experiments.


61

Figure 4.2:T cell proliferation results obtained using the Click-it EdU reagents showing a dual-parameter plot of Click-it EdU Alexa Fluor 488 and the CD3 T cell surface marker staining. CaN activity

Cellular CaN phosphate activity in CD4+ T cell lysates was assessed following the

manufacturer’s instructions (EMD Chemicals, San Diego, CA, USA). Briefly, CD4+ T cells

were lysed in a buffer supplemented with a mixture of protease inhibitors for 30min at

4°C under agitation. Lysates were centrifuged at 4500xg for 20h at 4°C and free

phosphate was eliminated by passing the supernatant through a chromatography

column. Effective removal of phosphate was qualitatively tested by the addition of Green

reagent to the flow-through. To determine CaN activity, 5µl of flow through was

incubated with RII phosphopeptide, a known substrate for CaN, and the free phosphate

released was detected based on the Malachite green assay (Martin, Pallen et al. 1985).

Since EGTA inhibits CaN, total CaN phosphatase activity was reported as the difference

between: phosphatase activity (Total) and phosphatase activity (EGTA buffer).

Protein determination

Protein levels in T cell lysates for CaN activity analysis was determined using the

BCA protein assay according to the manufacturer’s instructions (Pierce, Rockford, IL)

(Smith, Krohn et al. 1985).


62


Data were analyzed using parametric one-way or two way analysis of variance

(ANOVA), where indicated, followed by Tukeys’ HSD post-hoc test whenever a main

effect attained statistical significance. Statistical significance was set at P<0.05. All

statistical analyses were carried out using the statistical software SPSS for Windows

(version 14.0).

Results

Cytokine production and proliferation of CD4+ T lymphocytes after APC-independent

activation

Exton and coworkers have shown that CIS is mediated in the periphery via the

splenic nerve and β-adrenergic mechanisms (Exton, Gierse et al. 2002). Chemical

sympathectomy via 6-OHDA completely blocked the conditioned suppression of

splenocyte proliferation in response to mitogens and Th1-cytokine production.

Furthermore, administration of the beta-adrenoceptor (βAR) antagonist propranolol

abrogated the conditioned effect on splenocyte proliferation, indicating that the

behaviorally conditioned T cell suppression is a βAR-dependent process. Most recent

studies have shown that CaN activity is significantly reduced in rat splenocytes after

conditioning with SAC-CsA (Pacheco-Lopez, Riether et al. 2009).

Earlier studies revealed that CD4+ T lymphocytes most prominently express the

β2AR causing a significant reduction in IL-2 production by these cells after stimulation

with the β2AR- agonist terbutaline (Ramer-Quinn, Swanson et al. 2000; Sanders and

Straub 2002). To determine whether stimulation of the βAR subtype alone was required

to alter the level of Th1-cytokine production by CD4+ T lymphocytes and to validate our ex

vivo model, enriched T lymphocytes (see Table 4.1) were activated with plate-bound anti-

CD3 and soluble anti-CD28 mAb in an APC-independent manner in the presence and

absence of various concentrations of terbutaline.

Table 4.1: FACS staining of naïve T cells before and after purification To analyze the degree of T cell purification, isolated T cells (2.5 x 105) were stained with APC conjugated anti-rat CD3, PE conjugated CD45, FITC conjugated CD4 and PerCP conjugated CD8a before and after a two step T lymphocyte selection, respectively. The ranges indicate results for all experiments implemented in the study. Percentages indicate the amount of CD45+ leukocytes for spleen cell suspensions (pre-purified) and T lymphocyte suspensions (post-purified) that express the indicated marker. T lymphocytes CD3+ CD3+ CD4+ CD3+ CD8+ Pre-purified 49.3 – 50.8 % 76.4 – 77.3 % 22.7 – 23.6 % Post-purified 94.0 – 97.4 % 96.7 – 97.4 % 2.6 – 3.3 %


63

As shown in Figure 4.3, terbutaline dose-dependently reduced the amount of the Th1-

cytokines, IL-2 and IFN-y, produced by CD4+ T lymphocytes (p<0.05) with these effects

being reversed by the concomitant addition of the βAR antagonist nadolol. However,

production of the Th2-cytokine IL-4 was not affected by βAR stimulation.

Figure 4.3: (A) IL-2 production and (B) IFN-γ production by freshly isolated CD4+ T lymphocytes exposed to the βAR agonist terbutaline. Isolated CD4+ T cells were activated with plate-immobilized anti-CD3 mAb (10µg/ml) and soluble anti-CD28 (1µg/ml). Increasing concentrations of terbutaline were added at the time of cell activation or terbutaline (10-5

M) was added in the presence or absence of indicated concentrations of nadolol. All culture supernatants were collected 72h after cell activation and analyzed for IL-2 and IFN-γ production. Data are represented as the mean and SEM of at least three independent experiments. An asterisk indicates


64

significant differences from naïve T lymphocytes activated by anti-CD3 and anti-CD28 only (p<0.05). Statistics: one-way ANOVA followed by Tukey’s HSD.

To confirm cytokine data, CD4+ T lymphocytes were stimulated with anti-

CD3/CD28 and various concentrations of terbutaline after T lymphocyte stimulation, and

T cell proliferation was assessed using the EDU cell proliferation assay. T lymphocyte

proliferation was dose-dependently decreased by terbutaline with the effects being

reversed by the non-selective βAR antagonist nadolol (p<0.05; Figure 4.4).

In summary, our data illustrate and confirm previous studies showing that

stimulation of the βAR on CD4+ T lymphocytes at the time of T cell activation with anti-

CD3 and anti-CD28 decreases the level of T lymphocyte proliferation and Th1-cytokines

produced without affecting Th2-cytokine production.

Figure 4.4: Proliferation of freshly isolated CD4+ T cells exposed to the βAR agonist terbutaline. Isolated CD4+ T lymphocytes were activated with plate-immobilized anti-CD3 mAb (10µg/ml) and soluble anti-CD28 (1µg/ml). Increasing concentrations of terbutaline were added at the time of cell activation or terbutaline (10-

5M) was added in the presence and absence of indicated concentrations of nadolol. T cell proliferation was assessed 72h after cell activation by measuring EDU incorporation using flow cytometry. Data are represented as the mean (% of control) EDU proliferation and SEM of at least three independent experiments. An asterisk indicates significant differences from T lymphocytes activated by anti-CD3 and anti-CD28 only (p<0.05). Statistics: one-way ANOVA followed by Tukey’s HSD.

Effects of terbutaline on CaN activity in T lymphocytes

To elaborate a potential link between βAR stimulation and inhibition in CaN

activity, T lymphoctes were exposed to terbutaline (10-5M) for up to 480min upon anti-

CD3/CD28 stimulation and CaN activity in T lymphocyte lysates was assessed using a

commercial cellular CaN activity kit. To verify successful β-adrenergic stimulation pCREB

and total CREB expression was analyzed in parallel over a period of 90min. Exposure of T


65

lymphocytes to a β-agonist results in intracellular accumulation of the second messenger

cAMP and concomitant PKA-dependent phosphorylation of the transcription factor CREB

peaking at 15min and returning to baseline after approximately 90min after stimulation

(Mayr and Montminy 2001).

Figure 4.5: (A) pCREB and (B) total CREB measurements in lysates of freshly isolated CD4+ T lymphocytes exposed to terbutaline. Isolated CD4+ T lymphocytes were activated with plate-immobilized anti-CD3 mAb (10µg/ml, open bars) and soluble anti-CD28 (1µg/ml) in the presence and absence of terbutaline (10-5

M, grey bars), nadolol (10-5

M, hatched bars) or CsA (100µg/l, closed bars). Cells were harvested and lysed after indicated times of activation and pCREB and CREB were assessed using a beads-based pCREB/CREB assay on a flow cytometer. Data are represented as the mean (% of control) pCREB and CREB and SEM of at least three independent experiments. An asterisk indicates significant differences from T lymphocytes activated by anti-CD3 and anti-CD28 only (p< 0.05). Statistics: two-way ANOVA followed by Tukey’s HSD.


66

As expected, pCREB and total CREB levels in anti-CD3/CD28-stimulated control

samples as well as in CsA treated samples remained unchanged upon agonist

stimulation. Terbutaline-treated samples however displayed the characteristic kinetic of

pCREB protein expression after βAR stimulation. pCREB levels peaked at 15min (p<0.05)

and returned to baseline after 90min of stimulation (Figure 4.5). These changes in pCREB

were completely independent of changes in intracellular total CREB. These data clearly

demonstrate successful βAR stimulation being reversed by addition of the βAR nadolol.

CsA exerts its immunosuppressive properties in vivo and ex vivo by inhibition of

the Ca2+/Calmodulin-dependent phosphatase CaN (Halloran 2001). CsA-treated CD4+ T

lymphocytes showed a pronounced and stable reduction in CaN activity over time

(p<0.05; Figure 4.6). Surprisingly, however, also terbutaline-treated cells displayed a

transient but strong reduction in CaN activity up to 90min after β-adrenergic stimulation

(p<0.05) returning back to baseline after 480min. Concomitant exposure of the cells to

the non-selective βAR antagonist nadolol abrogated the observed effect on CaN activity,

supporting the position that βAR- stimulation transiently affects CaN activity, with these

effects being blocked by the addition of nadolol. Importantly, these changes in CaN

activity were not due to changes in CaN expression in T lymphocytes.

In summary, our data illustrate for the first time that stimulation of the βAR on

CD4+ T lymphocytes at the time of T lymphocytes activation with anti-CD3/CD28

transiently decreases CaN activity.

Figure 4.6: CaN activity measurement in lysates of freshly isolated CD4+ T lymphocytes exposed to terbutaline. Isolated CD4+ T lymphocytes were activated with plate-immobilized anti-CD3 mAb (10µg/ml, open bars) and soluble anti-CD28 (1µg/ml) in the presence and absence of terbutaline (10-5M, grey bars),


67

nadolol (10-5M, hatched bars) or CsA (100µg/l, closed bars). Cells were harvested and lysed after indicated times of activation and CaN activity was assessed. Data are represented as the mean (% of control) CaN activity and SEM of at least three independent experiments. An asterisk indicates significant differences from T lymphocytes activated by anti-CD3 and anti-CD28 only (p<0.05). Statistics: two-way ANOVA followed by Tukey’s HSD. Inhibition of CaN activity in T lymphocytes after terbutaline is PKA-dependent

Previous studied have shown that terbutaline-associated suppression of mitogen-

induced proliferation and cytokine production in T lymphocytes is a PKA-dependent

mechanism (Torgersen, Vang et al. 2002; Hughes-Fulford, Sugano et al. 2005; Tasken and

Stokka 2006). However, it has recently been shown that elevated concentrations of cAMP

per se are not responsible for this inhibition but rather the level of PKA activity generated

(Bauman, Bartik et al. 1994). The isoquinoline H-89 inhibits PKA (Hughes-Fulford, Sugano

et al. 2005). To see, whether the reduction in CaN activity in T lymphocytes after βAR

stimulation is PKA dependent, CD4+ T lymphocytes were exposed to terbutaline in the

presence and absence of H-89. Terbutaline-induced CREB phosphorylation was blocked

by H-89 (Figure 4.7). More importantly, however, H-89 dose-dependently inhibited

terbutaline-induced inhibition in CaN activity demonstrating that βAR stimulation-

induced inhibition of the phosphatase CaN is a PKA-dependent process (Figure 4.8).

Figure 4.7: pCREB measurement in lysates of freshly isolated CD4+ T lymphocytes exposed to terbutaline (10-

5M, grey bar) in the presence and absence of the PKA-inhibitor H-89 (10µM, hatched bar). Statistics: two-way ANOVA followed by Tukey’s HSD; *p<0.05 vs. anti-CD3/CD28 stimulated controls (solid bar).


68

Figure 4.8: CaN activity measurement in lysates of freshly isolated CD4+ T lymphocytes exposed to terbutaline (10-5M, grey bar) in the presence and absence of the PKA-inhibitor H-89 (10µM, hatched bar). Statistics: two-way ANOVA followed by Tukey’s HSD; *p<0.05 vs. anti-CD3/CD28 stimulated controls (solid bar).

Inhibition of CaN activity in T lymphocytes after terbutaline is ββββAR-receptor dependent

To dissect the mechanism by which the βAR agonist terbutaline induced

reduction in CaN activity and to investigate whether a βAR independent rise in the

second messenger cAMP is essential for the terbutaline-induced reduction in CaN

activity, T lymphocytes were treated with the cell membrane-permeable cAMP analog,

DBcAMP, and CaN activity in T lymphocyte lysates was examined. Treatment of CD4+ T

lymphocytes with DBcAMP as well as terbutaline resulted in the characteristical pCREB

expression profile (p<0.05; Figure 4.9). CaN activity, however, was not affected at any

time after treatment with DBcAMP compared to terbutaline-treated samples (p<0.05;

Figure 4.10). These results indicate that terbutaline-induced inhibition in CaN activity is a

βAR dependent process.


69

Figure 4.9: pCREB measurement in lysates of freshly isolated CD4+ T lymphocytes exposed to terbutaline (10-

5M, grey bar) or the cell-permeable cAMP-analog DBcAMP (10-4M, hatched bar). Statistics: two-way ANOVA followed by Tukey’s HSD; *p<0.05 vs. anti-CD3/CD28 stimulated controls (solid bar).

Figure 4.10: CaN activity measurement in lysates of freshly isolated CD4+ T lymphocytes exposed to terbutaline (10-5M, grey bar) or the cell-permeable cAMP-analog DBcAMP (10-4M, hatched bar). Statistics: two-way ANOVA followed by Tukey’s HSD; *p<0.05 vs. anti-CD3/CD28 stimulated controls (solid bar). Inhibition of CaN activity in T lymphocytes after Terbutaline is AKAP-150/PKA complex-

dependent

During the last years a series of endogenous CaN inhibitors have been identified

amongst others AKAPs (Liu 2003). AKAPs target PKA to different subcelluar locations and


70

are thought to play important roles in the cAMP signalling pathway through their

association with PKA in T lymphocytes (Baillie, Scott et al. 2005; Scott 2006). At least 8

AKAPS were identified in CD4+ T lymphocytes and Jurkat cells forming multivalent

signaling complexes with kinases and phosphatases involved in GPCR mediated

signalling inter alia AKAP-150 (Scott 1997; Schillace, Andrews et al. 2002; Malbon, Tao et

al. 2004). AKAP-150 contains binding sites for the major signaling molecules in T

lymphocyte activation such as PKC and CaN suggesting that both are targeted to

subcelluar sites to regulate substrate phosphorylation (Coghlan, Perrino et al. 1995; Faux,

Rollins et al. 1999).

Disruption of the AKAP-PKA interaction by the AKAP inhibitor peptide St-Ht31 rendered T

lymphocytes insensitive to cAMP-elevating agents (Williams 2002). To determine

whether disruption of the AKAP-PKA complex abolishes terbutaline-induced inhibition of

CaN activity, CD4+ T lymphocytes were exposed to terbutaline in the presence and

absence of St-Ht31 (Figure 4.11) and CaN activity was assessed in T lymphocyte lysates.

Incubation of cells with terbutaline showed a statistically significant reduction in CaN

activity after 60min of incubation (p<0.05) with this effect being reversed after 360min

of incubation. However, St-Ht31 disruption of the AKAP-PKA complex completely

abolished CaN inhibition in T lymphocytes (Figure 4.11). As a control we used the peptide

St-Ht31P in which two isoleucine residues have been replaced by proline residues, thereby

blocking its ability to disrupt the AKAP-PKA complex. CaN activity in these control

samples remained unaffected (data not shown). These results clearly indicate that the

AKAP-PKA complex is indispensable to the terbutaline-induced inhibition in CaN activity

in T lymphocytes and that disruption of the AKAP-PKA interaction rendered T

lymphocytes insensitive to βAR stimulation. In addition, these data suggest that the

temporale terbutaline-induced inhibition of CaN activity in CD4+ T lymphocytes is

mediated via binding to AKAP-15O on an intracellular level.


71

Figure 4.11: CaN activity measurement in lysates of freshly isolated CD4+ T lymphocytes exposed to terbutaline (10-5M, grey bar) in the presence and absence of AKAP inhibitor peptide St-Ht31 (50µM, hatched bar). Statistics: two-way ANOVA followed by Tukey’s HSD; *p<0.05 vs. anti-CD3/CD28 stimulated controls (solid bar).

Discussion

It has been previously demonstrated that CIS suppresses T lymphocyte

proliferation and Th1-cytokine secretion and is mediated via the splenic nerve and β-

adrenergic mechanisms (Exton, Gierse et al. 2002; Xie, Frede et al. 2002; Pacheco-Lopez,

Riether et al. 2009). Since CaN activity was shown to be reduced in spleens of conditioned

rats (Pacheco-Lopez, Riether et al. 2009), the β-adrenergic-T lymphocyte interaction was

modelled ex vivo to investigate the cellular efferent mechanism of CIS. The current data

indicate that CaN activity in enriched T lymphocytes is transiently reduced after specific

β-adrenergic agonist stimulation with terbutaline with these effects being mediated via

PKA and most likely the endogenous CaN inhibitor, AKAP-150 (Liu 2003). The suppressive

effects were antagonized by the non-selective β-adrenergic antagonist nadolol.

Distinct lymphocyte populations like NK, B and T cells were shown to express β2-

adrenoceptors in different densities (Maisel, Fowler et al. 1989; Sanders, Kasprowicz et al.

2001). In addition, it has been shown that β2-adrenoceptors are predominantly expressed

on mouse Th1 cells but not Th2 cells (Ramer-Quinn, Baker et al. 1997; Sanders, Baker et al.

1997; Ramer-Quinn, Swanson et al. 2000; Sanders and Straub 2002) and mediate cAMP-

dependant inhibition of T lymphocyte proliferation and IL-2 production after mitogen-

induced stimulation (Kammer 1988; Bartik, Bauman et al. 1994; Tasken and Stokka 2006).


72

Recently, it was shown that disruption of the AKAP-PKA complex abolishes β-agonist

induced inhibition of T cell function by blocking the interaction of AKAP with β2-

adrenergic receptors (Williams 2002). AKAPs represent dynamic multivalent signalling

platforms that regulate PKA activity by translocating the enzyme to various subcelluar

compartments (Malbon, Tao et al. 2004; Scott 2006). Moreover, CaN was demonstrated

to associate with AKAP-150 and with the β-adrenergic receptor, while CaN phosphatase

activity is silenced by association with AKAP-150 and the β2-adrenergic receptor in

interaction studies (Liu, Farmer et al. 1991; Coghlan, Perrino et al. 1995; Scott 1997;

Kashishian, Howard et al. 1998; Liu 2003) . AKAP-150 binds the catalytic subunit α of CaN

distinct from the CsA binding domain (Kashishian, Howard et al. 1998). However, the

putative CaN binding site of AKAP-150 is almost identical to the binding site of the

immunophilin FKBP-12, the only binding protein mediating T lymphocyte inhibition by the

CaN inhibitor FK506 (Liu, Farmer et al. 1991; Coghlan, Perrino et al. 1995). The FKBP12-

FK506 complex was shown to inhibit CaN in a competitive manner. Taken these data

together, one may argue that binding of CaN to the AKAP complex leads to a competitive

inhibition of CaN activity. One the other hand, it was demonstrated that CaN activity can

be regulated by phosphorylation of CaN by the Ca2+/Calmodulin-dependent kinase II and

PKC resulting in a subsequent modification of the CaM-binding domain (Hashimoto and

Soderling 1989; Martensen, Martin et al. 1989). The data demonstrate that the regulation

of T lymphocyte function after specific β-adrenergic stimulation results in a pronounced

but transient inhibition in CaN activity. In contrast, it was previously shown that α-

adrenergic mechanisms have no impact on mediation of CIS by CsA (Xie, Frede et al.

2002). However, whether the observed inhibition in CaN activity is directly mediated by

AKAP-150 or indirectly via kinases like CAMKII or PKA still has to be elucidated (Figure 5.1).

In summary, the current data demonstrate that β-adrenoceptor stimulation

specifically inhibits CD4+ T lymphocyte responsiveness on a transcriptional level by

temporal inhibition of the phosphatase CaN via a PKA dependent mechanism and the

involvement of the endogenous CaN inhibitor AKAP-150. For the first time, we identified

ex vivo the mechanism mediating conditioned immunosuppression by CsA involving

sympathetic activation of β-adrenoceptors on CD4+ T lymphocytes.


73

References Baillie, G. S., J. D. Scott, et al. (2005). "Compartmentalisation of phosphodiesterases and protein kinase A:

opposites attract." FEBS Lett 579(15): 3264-70.

Bartik, M. M., G. P. Bauman, et al. (1994). "Costimulatory signals modulate the antiproliferative effects of

agents that elevate cAMP in T cells." Cell Immunol 158(1): 116-30.

Bauman, G. P., M. M. Bartik, et al. (1994). "Induction of cAMP-dependent protein kinase (PKA) activity in T cells

after stimulation of the prostaglandin E2 or the beta-adrenergic receptors: relationship between

PKA activity and inhibition of anti-CD3 monoclonal antibody-induced T cell proliferation." Cell

Immunol 158(1): 182-94.

Coghlan, V. M., B. A. Perrino, et al. (1995). "Association of protein kinase A and protein phosphatase 2B with a

common anchoring protein." Science 267(5194): 108-11.



Faux, M. C., E. N. Rollins, et al. (1999). "Mechanism of A-kinase-anchoring protein 79 (AKAP79) and protein

kinase C interaction." Biochem J 343 Pt 2: 443-52.


Hashimoto, Y. and T. R. Soderling (1989). "Regulation of calcineurin by phosphorylation. Identification of the

regulatory site phosphorylated by Ca2+/calmodulin-dependent protein kinase II and protein kinase

C." J Biol Chem 264(28): 16524-9.

Hughes-Fulford, M., E. Sugano, et al. (2005). "Early immune response and regulation of IL-2 receptor subunits."

Cell Signal 17(9): 1111-24.

Kammer, G. M. (1988). "The adenylate cyclase-cAMP-protein kinase A pathway and regulation of the immune

response." Immunol Today 9(7-8): 222-9.

Kashishian, A., M. Howard, et al. (1998). "AKAP79 inhibits calcineurin through a site distinct from the

immunophilin-binding region." J Biol Chem 273(42): 27412-9.

Klauck, T. M., M. C. Faux, et al. (1996). "Coordination of three signaling enzymes by AKAP79, a mammalian

scaffold protein." Science 271(5255): 1589-92.

Liu, J., J. D. Farmer, Jr., et al. (1991). "Calcineurin is a common target of cyclophilin-cyclosporin A and FKBP-

FK506 complexes." Cell 66(4): 807-15.

Liu, J. O. (2003). "Endogenous protein inhibitors of calcineurin." Biochem Biophys Res Commun 311(4): 1103-9.

Maisel, A. S., P. Fowler, et al. (1989). "A new method for isolation of human lymphocyte subsets reveals

differential regulation of beta-adrenergic receptors by terbutaline treatment." Clin Pharmacol Ther

46(4): 429-39.

Malbon, C. C., J. Tao, et al. (2004). "AKAPs (A-kinase anchoring proteins) and molecules that compose their G-

protein-coupled receptor signalling complexes." Biochem J 379(Pt 1): 1-9.

Martensen, T. M., B. M. Martin, et al. (1989). "Identification of the site on calcineurin phosphorylated by

Ca2+/CaM-dependent kinase II: modification of the CaM-binding domain." Biochemistry 28(24):

9243-7.

Martin, B., C. J. Pallen, et al. (1985). "Use of fluorinated tyrosine phosphates to probe the substrate specificity

of the low molecular weight phosphatase activity of calcineurin." J Biol Chem 260(28): 14932-7.

Mayr, B. and M. Montminy (2001). "Transcriptional regulation by the phosphorylation-dependent factor

CREB." Nat Rev Mol Cell Biol 2(8): 599-609.

Mignini, F., V. Streccioni, et al. (2003). "Autonomic innervation of immune organs and neuroimmune

modulation." Auton Autacoid Pharmacol 23(1): 1-25.


74

Pacheco-Lopez, G., C. Riether, et al. (2009). "Calcineurin inhibition in splenocytes induced by Pavlovian

conditioning." FASEB J 23 (4): 1161-7.

Ramer-Quinn, D. S., R. A. Baker, et al. (1997). "Activated T helper 1 and T helper 2 cells differentially express the

beta-2-adrenergic receptor: a mechanism for selective modulation of T helper 1 cell cytokine

production." J Immunol 159(10): 4857-67.

Ramer-Quinn, D. S., M. A. Swanson, et al. (2000). "Cytokine production by naive and primary effector CD4+ T

cells exposed to norepinephrine." Brain Behav Immun 14(4): 239-55.

Sanders, V. M., R. A. Baker, et al. (1997). "Differential expression of the beta2-adrenergic receptor by Th1 and

Th2 clones: implications for cytokine production and B cell help." J Immunol 158(9): 4200-10.

Sanders, V. M., D. J. Kasprowicz, et al. (2001). Neurotransmitter receptors on lymphocytes and other lymphoid

cells. Psychoneuroimmunology. R. Ader, D. L. Felten and N. Cohen. San Diego, Academic Press: 161-

196.

Sanders, V. M. and R. H. Straub (2002). "Norepinephrine, the beta-adrenergic receptor, and immunity." Brain

Behav Immun 16(4): 290-332.

Schillace, R. V., S. F. Andrews, et al. (2002). "Identification and characterization of myeloid translocation gene

16b as a novel a kinase anchoring protein in T lymphocytes." J Immunol 168(4): 1590-9.

Scott, J. D. (1997). "Dissection of protein kinase and phosphatase targeting interactions." Soc Gen Physiol Ser

52: 227-39.

Scott, J. D. (2006). "Compartmentalized cAMP signalling: a personal perspective." Biochem Soc Trans 34(Pt 4):

465-7.

Smith, P. K., R. I. Krohn, et al. (1985). "Measurement of protein using bicinchoninic acid." Anal Biochem 150(1):

76-85.

Tasken, K. and A. J. Stokka (2006). "The molecular machinery for cAMP-dependent immunomodulation in T-

cells." Biochem Soc Trans 34(Pt 4): 476-9.

Torgersen, K. M., T. Vang, et al. (2002). "Molecular mechanisms for protein kinase A-mediated modulation of

immune function." Cell Signal 14(1): 1-9.

Williams, R. O. (2002). "Cutting edge: A-kinase anchor proteins are involved in maintaining resting T cells in

an inactive state." J Immunol 168(11): 5392-6.



Chapter 5: General discussion

75

Chapter 5.

General discussion


76

Overview

The experiments presented in this thesis are based on data demonstrating intense

communication between the CNS and the immune system and applying the paradigm of

conditioned immunosuppression by CsA (CIS). The study provides for the first time

experimental data on the afferent mechanism mediating the detection of the US and the

intracellular mechanism evoking the conditioned response in the periphery. The purpose of

these experiments was to investigate and to advance the understanding of the mechanisms

and biological relevance of CIS. The items are:

(i) the central and peripheral immuno-sensory process induced by the i.p. CsA (US)

administration and

responsible for the mediation of “learning” during acquisition time (Chapter 2)

(ii) the aspect of specificity of CIS by CsA (Chapter 3)

(iii) potential intracellular mechanisms of CIS (Chapter 4).

The results from the experiment described in chapter 2 demonstrated that central

cytokines, such as IL-1β, are involved within the central processing of immune stimuli during

acquisition time of CIS. These data illustrate that the CNS seems to detect both the injection of

the immunosuppressive drug (US) and subsequent reduction in pro-inflammatory cytokine

expression in the periphery and in response induces the up-regulation of IL-1β mRNA

expression in the CNS. This confirms that an “immunosensory” process supports the CsA

detection by the CNS during acquisition time.

The experiments included in Chapter 3 (Pacheco-Lopez, Riether et al. 2009) emphasize

the specificity of CIS and demonstrate that the pharmacologic immunosuppressive properties

of CsA can be transferred to an immune neutral stimulus (e.g. the taste of SAC). Analysis of CaN

activity in splenic lysates after the evocation phase and analysis of T lymphocyte proliferation

and cytokine production after specific stimulation with the mAb anti-CD3 revealed that CIS is

mediated on a cellular level by the inhibition CaN activity in T lymphocytes. This confirms that

the conditioned response is triggered by recruiting the same intracellular signalling cascade

that is required for immunosuppression by CsA and highlights the specificity of the underlying

association (i.e. taste-CaN inhibition).

The experimental findings described in Chapter 4 demonstrate that β2-adrenergic

stimulation specifically inhibits T lymphocyte responsiveness on a transcriptional level by

temporal inhibition of the phosphatase CaN via a PKA dependent mechanisms. These data

suggest the temporal association of CaN with the endogenous CaN inhibitor AKAP-150 to be

crucial for the transient inhibition of CaN activity. For the first time, we were able to identify in


77

an ex vivo model the mechanism mediating CIS and involving sympathetic activation of βAR on

T lymphocytes.

Together, the results presented in this thesis are a clear illustration of how the CNS

detects the immunosuppressive US at acquisition time and they demonstrate for the first time

how the conditioned response is mediated in the periphery at intracellular level at evocation

time. The sensory capabilties of the CNS to detect changes in cytokine concentrations in the

periphery, in particular changes in pro-inflammatory cytokines such as IL-1β, IL-6 or TNF, has

already been documented (Gibertini 1996; Schneider, Pitossi et al. 1998; Banks, Farr et al. 2001;

Matsumoto, Yoshida et al. 2001; Rachal Pugh, Fleshner et al. 2001; Besedovsky and del Rey

2007). Data demonstrate that an “immunosensory” process mediates the perception of CsA at

acquisition time. The injection of the immunosuppressive drug CsA and the subsequent

reduction in pro-inflammatory cytokine expression in the spleen centrally induces the up-

regulation of IL-1β mRNA expression.

For the first time we identified an intracellular mechanism mediating CIS. The

conditioned response is mediated via the sympathetic nervous systems, predominantly via β-

adrenergic receptors on T lymphocytes. β-adrenergic activation of T lymphocytes leads to the

temporal inhibition of CaN activity and a subsequent reduction in T lymphcyte responsiveness.

Thus, both CIS and immunosuppression by CsA is mediated via inhibition of a common

intracellular target molecule, CaN, with the involvement of two different intracellular

inhibitors, namely CsA and most likely AKAP-150 respectively. An integrative diagram of the

identified intracellular mechanisms ruling CIS is shown in Figure 5.1.


78

Figure 5.1: Schematic representation of the intracellular events in CD4+ T lymphocytes after β-adrenergic stimulation mediated by the temporal association of A-kinase anchoring protein (AKAP)-150 with calcineurin (CaN). The example shown is that of AKAP-150 which associates reversibly with the β-adrenergic receptor (βAR) in response to agonist-induced GPCR activation. AKAP is an intriguing scaffold protein representing a dynamic and multivalent signalling complex. AKAP-150 binds CaN, PLC and PKC and clusters the enzymes to control signal transduction as specific intracellular compartments. Upon β-agonist CD4+ T lymphocyte activation, cAMP is generated and AKAP-150 complex associates with β AR to foster protein kinase A (PKA) in the proximity of local cAMP accumulation. PKA gets activated and phoshorylates both the receptor as well as AKAP to enhance the docking and for the further resensitization process. Subsequently, CD4+ T lymphocyte responsiveness is inhibited by temporal binding of CaN to the AKAP-complex. The Ca2+/Calmodulin-dependant phosphatase CaN that is a fundamental intracellular mediator of early T cell activation is shown in the current study to be reversibly associated with the endogenous CaN inhibitor, AKAP-150, upon β-AR stimulation and PKA activation. (abbreviations: AC, adenylate cyclase; PKC, protein kinase C; PLC, phospholipase C; PKA, protein kinase A; CREB, cAMP responsive element-binding; CBP, CREB binding protein; GPCR, G-protein coupled receptor; cAMP, cyclic adenosine monophosphate; β-Arr, β-Arrestin)

Potential mechanisms of conditioned immunosuppression by CsA

Behavioural conditioning is a unique model for examining in parallel both the efferent

as well as the afferent communication between the CNS and the immune system (for review

see Riether, Doenlen et al. 2008). Employing a CTA paradigm by pairing the

immunosuppressive drug CsA as US and SAC as CS behaviourally conditioned suppression of

peripheral immune functions in rats has been exquisitely demonstrated (Ader and Cohen 2001;

Exton, Herklotz et al. 2001; Riether, Doenlen et al. 2008). The relevance and feasibility of

employing behavioural conditioning as a supplement to standard pharmacological regimens in

animals as well as in humans has been extensively demonstrated (Giang, Goodman et al. 1996;

Goebel, Trebst et al. 2002). The following section will provide an overview of current knowledge

on the neural and peripheral mechanisms involved in both afferent and efferent mediation of

CIS.


79

Based on the central findings on CTA (see general introduction), a series of studies

investigated the involvement of the insular cortex (IC) and the amygdala (Am), which are

reciprocally interconnected, in conditioned immunosuppression of IgM antibody production

(Ramirez-Amaya, Alvarez-Borda et al. 1998; Ramirez-Amaya and Bermudez-Rattoni 1999). The IC

and Am were identified as key structures in mediating conditioned immunosuppression after

evoking taste-CY association. Excitotoxic lesions either in the IC or the Am before acquisition

demonstrated that both IC and Am lesions disrupt acquisition of conditioned

immunosuppression. These data provided evidence for the involvement of IC and Am in taste-

immune memory formation. In addition, cortical and amygdaloidal glutamate release has been

related to central visceral processing (Miranda, Ferreira et al. 2002). Moreover, the IC and the

Am were shown to be involved in behavioural interactions that mediate conditioned

enhancement of antibody production (Ramirez-Amaya and Bermudez-Rattoni 1999).

Furthermore, it has been shown that both the cholinergic as well as the serotonergic system

are critical for triggering the conditioned NK cell response during the association phases

(Hsueh, Chen et al. 2002). Acetylcholine seems to act through the nicotinic, M2 and M3-

muscarinic receptors. In contrast, CS-US association does not seem to involve endogenous

opioids, since naltrexone, an opioid antagonist, administered prior to acquisition did not

interfere with the conditioning process (Solvason, Hiramoto et al. 1989).

More recently, the neural substrates involved in behaviourally conditioned

immunosuppression by CsA in rats were identified (Pacheco-Lopez, Niemi et al. 2005), showing

that excitotoxic brain lesions of the IC and the Am before acquisition modulate CIS on the

immune system as illustrated by the analysis of splenocyte proliferation and cytokine

production. More specifically, these results indicate that the IC and the Am are essential for

mediating the input of visceral information necessary at acquisition time (Figure 1.2).

As already discussed in detail in Chapter 2, the induction of peripheral pro-

inflammatory cytokines in the spleen after administration of the immunosuppressive US at

acquisition time may trigger the detection of the US. Subsequent analysis of cytokine contents

in the Am have confirmed that US-induced up-regulation of peripheral pro-inflammatory

cytokines is sensed by the CNS and triggers the de-novo production of central IL-1β in a kind of

“ïmmunosensory” process during acquisition of CIS .

Since recent studies have show that peripheral cytokine elevation in response to

infection induces a set of behavioural and physiological changes mediated by the vagus nerve

and referred to as sickness behaviour (Tracey 2007; Dantzer, O'Connor et al. 2008), and one

could argue that the vagus may transmit the perception of the US (Doenlen 2008). However,

several alternative mechanisms exist through which the elevation in peripheral cytokines after

administration of immunomodulating agents ia able to bring about changes in central cytokine

mRNA expression and up-regulation of neurotransmitters, but the exact mechanism of the US

detection still has to be elucidated.


80

CIS has been shown to be mediated centrally on the efferent arm by key brain

structures such as the IC and the VMH. Furthermore it has been shown that the sympathetic

nervous system mediated the CR via the splenic nerve and βAR mechanisms in the periphery

(Exton, Gierse et al. 2002; Xie, Frede et al. 2002; Pacheco-Lopez, Niemi et al. 2005; Riether,

Doenlen et al. 2008). Investigating the conditioned enhancement of NK cell activity in rodents,

it was demonstrated that central catecholamines and glutamate are essential factors at the

evocation stage (Hiramoto, Solvason et al. 1990; Hsueh, Kuo et al. 1999; Kuo, Chen et al. 2001).

Furthermore, it has been shown that both the cholinergic as well as the serotonergic system

are critical for triggering the conditioned NK cell response during the evocation phase (Hsueh,

Chen et al. 2002). In addition, it was demonstrated that naltrexone only blocked conditioned

enhancement of NK cell activity when applied before re-exposure to the CS, suggesting that

opiate receptors in the CNS mediate the CR.

The VMH, widely recognized as a satiety centre (Vettor, Fabris et al. 2002), is intimately

associated with sympathetic facilitation in peripheral tissues (Saito, Minokoshi et al. 1989),

including modulation of peripheral immune reactivity (Okamoto, Ibaraki et al. 1996). Excitotoxic

brain lesions of the IC, the AM and the VMH identified the neural substrates of CIS. While the

VMH appears to participate in the efferent output pathway to the immune system to evoke CIS,

the IC was identified as an essential brain structure in both acquisition and evocation of the CR

in the underlying paradigm (Figure 2.1).

Based on the main findings on neural innervation of secondary lymphoid organs such

as the spleen (Nance and Sanders 2007) and the expression of receptors for neurotransmitters

on lymphocytes (Elenkov, Wilder et al. 2000), the splenic nerve was identified as the major

efferent mediator linking the brain and the immune system, because chemical and surgical

sympathetic denervation completely blocked conditioned immunosuppression by CsA (Exton,

von Horsten et al. 1998; Exton, Schult et al. 1999). In parallel, noradrenaline was identified as

the predominant neurotransmitter mediating the conditioned CsA effects via β-adrenergic

mechanisms (Exton, Gierse et al. 2002). In contrast, splenic denervation did not affect

conditioned suppression of contact hypersensitivity (CHS) reaction using CsA in the same

paradigm, since this acute immune response occurs predominantly in the draining lymph

nodes (Exton, Elfers et al. 2000).

The present data (Chapter 3) contribute significantly to the understanding of this

peripheral mediatory mechanism of CIS. The documented findings show for the first time that

(Pacheco-Lopez, Niemi et al. 2005) by behavioral conditioning T lymphocyte function in the

spleen is modulated by reducing the phosphatase activity of CaN. Subsequent analysis of CaN

activity in T lymphocytes after treatment with the specific βAR agonist terbutaline indicated

that βAR activation of T lymphocytes leads to a temporal inhibition of CaN activity (Chapter 4).


81

The A-kinase anchoring protein (AKAP)-150, was shown to keep T lymphocytes in an inactive

state (Williams 2002). In addition, AKAP-150 anchors and regulates PKA activity (Scott 2006)

and binds to CaN, PLC and PKC, clustering the enzymes to control signal transduction as

specific intracellular compartments (Coghlan, Perrino et al. 1995; Klauck, Faux et al. 1996). The

current data demonstrate that β-adrenergic stimulation specifically inhibits CD4+ T lymphocyte

responsiveness on a transcriptional level by temporal inhibition of the phosphatase CaN

mediated via a PKA dependent mechanism and the involvement of the endogenous CaN

inhibitor AKAP-150. For the first time, the mechanism mediating CIS that is characterized by the

sympathetic activation of β-adrenergic receptors on CD4+ T lymphocytes was identified ex vivo.

Several alternative mechanisms exist by which AKAP-150 may inhibit CaN activity in T

lymphocytes.

- Firstly, via direct mechanisms and the binding of AKAP-150 to the FKBP-12 binding

domain of CaN, a domain that is involved in mediation of the immunosuppressive

actions of CaN inhibitor FK506 (Ho, Clipstone et al. 1996). AKAP-150 was shown to bind

to CaN at a domain similar to the FKBP-12 binding domain. Therefore, AKAP-150 may act

as a competitive inhibitor CaN.

- Second, indirectly, via the recruitment of kinases like PKA and CaMK II to the multivalent

AKAP complex. These kinases were shown to inhibit CaN activity by phosphorylation of

the CaM binding pocket and subsequent conformational and physiological alterations

in CaN (Hashimoto and Soderling 1989).

In summary, the present data demonstrate that through contingent pairings the

specific pharmacological properties of CsA are mimicked by a specific taste. Two major

alternatives, not mutually exclusive, may explain this phenomenon. First, the association

between a specific taste (sensory input) and a specific drug effect (suppression of T cell

function by CaN inhibition) is encoded and stored in the brain, inducing a specific and

independent engram, which afterwards can be selectively activated by re-exposure to the given

taste. Second, neurally-derived signals triggered by gustatory stimulation may transmit just a

part of the pattern of the engram to selective immune cells in the lymphoid tissue at

acquisition time (Grossman, Herberman et al. 1992). The data support associative specificity

and control at the same time for residual effects of the CsA. However, conclusive evidence can

only be provided by experiments specifically designed to assess passive forgetting and by

discriminative conditioning paradigms (Ader 2003).

The present study contributes to a better understanding of the afferent and efferent

mechanisms underlying CIS and provides for the first time experimental data how the

conditioned suppression of T cell function is mediated at the cellular level. However, the aspect

of specificity and the question of whether the regulatory pathway discovered is a common

process mediating all different kinds of conditioned immunosuppressions, or whether the


82

pathway discovered is a specific mode of action for CIS still has to be elucidated. Therefore,

extending the conditioning model byimplementating other immunosuppressive agents, which

are independent of CaN would help to further assess the aspect of specificity of the model. For

example, rapamycin, an immunosuppressive drug that suppresses T cell proliferation via

inhibition of the mammalian target of rapamycin (mTOR; Figure 3.4; Abraham and

Wiederrecht, 1996). Furthermore, the issue whether CIS specifically conditions the effects of the

CsA or whether it is just general sympathetic activation of the IS has not yet been answered,

but this knowledge would be helpful in assessing the full potential of CIS and other

immunosuppressive agents, in order to transfer this model as a supplement therapy from

bench to bedside.

Potential use of behavioural conditioning as a supplemental therapy in patients

Conditioning experiments performed in humans have indicated the feasibility of

modulating immune reactivity through such behavioural approaches (Giang, Goodman et al.

1996; Goebel, Hubell et al. 2005). It may be advisable to treat some illnesses or syndromes with

several therapeutical approaches in order to optimize the treatment outcome. In this context,

behavioural conditioning regimens should be employed as complementary therapy supporting

the main pharmacological treatment. In the following section the clinical relevance of selected

paradigms for behavioural conditioning of immune functions in animal and human subjects

will be illustrated.

Animal models with clinical relevance

In 1975, Ader and Cohen performed pioneer experiments on conditioned

immunosuppression by pairing the flavour of SAC solution (CS) with an i.p. injection of

cyclophosphamide (CY) (US) in a single acquisition trial (Ader and Cohen 1975). CY is a drug

primarily used in chemotherapy and is also referred to as an immunosuppressive agent in

autoimmune diseases. Following this protocol, it was demonstrated that mice that had been

given several pairings of CY in association with SAC displayed prolonged skin allograft survival

after subsequent exposure to the gustatory CS (Gorczynski 1990). When investigating the

impact of conditioned immunosuppression by CY on the development of systemic lupus

erythematosus, it was shown that recalling such an association was sufficient to delay the

onset of this autoimmune disease and increase the survival rate of conditioned animals (Ader

and Cohen 1982). Moreover, Grochowicz et al. demonstrated that conditioned

immunosuppression by a CsA paradigm prolonged graft rejection processes in an established

heart transplantation model (Grochowicz, Schedlowski et al. 1991). Recipient rats were

transplanted with heterotopic heart allograft of donor strains. Animals received two

acquisition trials, namely, SAC taste was paired with an i.p. injection of CsA. One day and three

days prior to heart transplantation, animals were re-exposed to the CS alone. Conditioned


83

animals displayed prolongation of heart allograft survival as well as exhibiting CTA. These

findings were consistently reproduced and extended by Schedlowski’s group, showing that

conditioned immunosuppression enhanced the effectiveness of CsA treatment in postponed

heart allograft survival (Exton, Schult et al. 1998). In addition, it was demonstrated that

subtherapeutic treatment with this immunosuppressive drug and behavioural conditioning

synergized, indicating the usefulness of behavioural models as a supplement to standard

immunomodulatory drug regimens.

In a model of asthma, it has been reported that treatment with diazepam before

evocation prevented the conditioned histamine release (Irie, Nagata et al. 2004). This indicates

that such sedative treatment prevents evocation under unconscious conditions. This finding

may have important clinical implications for the therapy of allergies, particularly for acute

episodes of bronchial asthma, by helping to prevent stress- or anxiety-related airway responses.

In a study on conditioned NK cell responses, polyinisinic: polycytidylic acid (poly I:C), a

synthetic TLR3 agonist that mimics double-stranded viral RNA was used as the US and camphor

odour as the CS. Re-exposure to the CS alone led to an increase in natural killer cell (NK) activity

and, after in vivo challenged with myeloma cells, conditioned mice showed resistance to

myeloma growth after repeatedly being exposed to the CS only (Ghanta, Miura et al. 1988), and

this gives a potential indication of the potential use of treating NK cell mediated responses.

In sum, these examples clearly demonstrate the clinical relevance of merging

pharmacological regimens with behavioural approaches in the treatment of immune-related

diseases.

Behavioural conditioning of immune responses in humans

Clinical observation and experimental data indicate that behavioural conditioning of

immune functions can also be induced in humans. In 1957, Decker et al. described two

asthmatic patients suffering from skin sensitivities to house dust and grass pollen (Dekker,

Pelser et al. 1957). These patients were exposed to these allergens by inhalation. After a series of

conditioning trials, they experienced allergic attacks after inhalation of the neutral solvent used

to deliver the allergens. The experiment showed fast conditioning of the asthmatic attack with

a lack of extinction. This observation, together with data from animal experiments (Noelpp and

Noelpp-Eschenhagen 1952; Noelpp and Noelpp-Eschenhagen 1952; Noelpp and Noelpp-

Eschenhagen 1952), resulted in the early hypothesis that asthma could be conceived of as a

“learned response” (Turnbull 1962). Much progress has been made in understanding the

principles underlying allergic reactions and several pharmacological treatments have been

developed (Rubin and Fink 2007). Moreover, allergic reactions were demonstrated to be

affected by behavioural conditioning and emotional status. For instance, after the acquisition

phase, elevated mast cell tryptase was observed when an intranasal saline application was


84

given simultaneously with the CS in patients with allergic rhinitis (Gauci, Husband et al. 1994).

In another study with asthmatic children, a bronchodilator inhalation (salbutamol) was paired

with a vanilla odour twice a day for 15 days. Re-exposure to the odour alone increased

pulmonary function for the conditioned children (Castes 1998). In a more recent experiment, a

novel-tasting drink was paired with the H1-receptor antagonist (desloratadine) for patients

suffering from allergic house-dust-mite rhinitis. After the acquisition phase, the results showed

that re-exposure to the novel-tasting drink decreased basophile activation. Moreover, the skin

prick test and the subjective symptom score showed results similar to those of the group that

received the drug only (desloratadine) (Goebel 2008), indicating that behavioural conditioning

is able to mimic the anti-histamine effects. It has also been shown that placebo bronchodilator

administration significantly reduced bronchial hyperreactivity (Kemeny, Rosenwasser et al.

2007). Recently, both fundamental and clinical research has focused on elucidating the

neurobiology behind the placebo effect (de la Fuente-Fernandez, Schulzer et al. 2002;

Benedetti, Rainero et al. 2003). Experimental data indicate that conscious expectation and

unconscious behavioural conditioning processes appear to be the major neurobiological

mechanisms capable of releasing endogenous neurotransmitters and/or neurohormones that

mimic the expected or conditioned pharmacological effects (Figure 5.2; for a review see

Pacheco-López, Engler et al. 2006).


85

Figure 5.2: Neurobiology of immunomodulatory placebo effects. Conscious expectation and unconscious behavioural conditioning processes appear to be the major neurobiological mechanisms of placebo effects. These processes are capable of releasing endogenous neurotransmitters that mimic the expected or conditioned pharmacological effects, i.e., the placebo effect. A discrete neural network has been identified that modulates peripheral immune functions and which can possibly be elicited by the psychosocial context of placebo treatments. Major neural efferent pathways, through which expectations and/or memories could affect peripheral immune functions include limbic and hypothalamic relays the HPA axis, and the brain stem–vagus–cholinergic pathway. Thus, placebo effects can affect end organ functioning and the overall health of an individual through the healing power of belief, positive expectations and conditioning processes. Abbreviations: ACTH, adrenocorticotropic hormone; Ach, acetylcholine; NE, noradrenaline (Pacheco-Lopez, Engler et al. 2006).


86

Associative learning has been consistently reported in the context of cancer treatment,

particularly chemotherapy (Bovbjerg 2003). Chemotherapeutic agents generally have

immunosuppressive effects. These agents are typically administered in cycles, with each

outpatient treatment infusion followed by a period of recovery prior to the next infusion. From

a conditioning perspective, clinic treatment visits can be viewed as "acquisition trials" in which

the distinctive salient features of the clinic environment (CS) are contingently paired with the

infusion of agents (e.g., CY; US) that affects immune functions. This view is supported by data

analyzing blood samples from 20 cancer patients prior to chemotherapy in the hospital and a

second time in the home environment. Proliferative responses of isolated blood immune cells

were lower when blood was taken in the hospital (i.e. after evocation) were lower than for

home samples (Bovbjerg, Redd et al. 1990). These results were replicated in 22 ovarian patients

(Lekander, Fürst et al. 1995) and 19 paediatric patients also receiving chemotherapy (Stockhorst,

Spennes-Saleh et al. 2000). In addition, chemotherapy patients often develop anticipatory or

conditioned nausea (Andrykowski 1988; Bovbjerg, Redd et al. 1990; Morrow, Lindke et al. 1991;

Matteson, Roscoe et al. 2002), anxiety (Jacobsen, Bovbjerg et al. 1993; DiLorenzo, Jacobsen et al.

1995) and fatigue responses (Bovbjerg, Montgomery et al. 2005) as reminders of chemotherapy.

The conditioned nausea and anxiety responses can also be elicited by thoughts and images of

chemotherapy (Redd, Dadds et al. 1993; Dadds, Bovbjerg et al. 1997), raising the possibility that

conditioned effects may affect patients during the course of normal life for years after

treatment.

Based on the knowledge that adrenaline administration leads to an increase in

numbers abd functions of NK cells in the periphery (Benschop, Rodriguez-Feuerhahn et al. 1996;

Schedlowski, Hosch et al. 1996), one research group assessed the conditionability of NK cell

numbers and their lytic activity in healthy volunteers following behavioural conditioning.

Although positive results were reported after evoking a taste–adrenaline association (Buske-

Kirschbaum, Kirschbaum et al. 1992), these effects could not be replicated by other research

groups (Buske-Kirschbaum, Kirschbaum et al. 1994).

The efficacy of behavioural conditioning was also tested in multiple sclerosis patients.

Four monthly CY infusions (US) were contingently paired with the taste of aniseed-flavoured

syrup (CS) (Giang, Goodman et al. 1996). Long-term treatment with CY decreases blood

leukocyte numbers leading to leukopenia. One month after the fifth acquisition trial, half of the

subjects were exposed to the gustatory CS paired with a sham infusion whereas the other half

was treated with CY. The results show that after the placebo infusion paired with the drink,

eight out of ten patients showed a conditioned reduction in peripheral leukocytes numbers. In

connection with the treatment of multiple sclerosis, it was possible to induce an elevation of

neopterin and quinolinic acid serum levels by pairing interferon-γ injections (US) with a

distinctively flavoured drink (CS) (Longo, Duffey et al. 1999). However, it has been hypothesized


87

that more than a single associative learning trial, pairing a distinctive taste (CS) with

interferon-γ injections (US), is necessary to produce immune conditioned effects (Goebel,

Hübell et al. 2005).

More closely related to the present data, healthy male volunteers received four times

the immunosuppressive drug CsA (US) paired with a distinctively flavoured/coloured solution

(CS) (Goebel, Trebst et al. 2002). Re-exposure to the drink alone (CS) resulted in conditioned

inhibition of cytokine (IL-2 and IFN-γ) mRNA expression and cytokine release, as well as reduced

proliferation of human peripheral blood lymphocytes.

In summary, the above mentioned studies demonstrate the relevance and feasibility of

employing behavioural conditioning to supplemental therapy to pharmacological standard

regimens in animals as well as in humans with the aim of maximizing the therapeutic benefit,

reducing the side effects and lowering the costs.

Concluding remarks

The behavioural conditioning model using CsA as a US was developed based on a taste-

aversion paradigm to study the intensive and clinically relevant anatomical and functional

interactions between the immune system and the central nervous system. This taste-immune

associative learning phenomenon proves to be a valid model to investigate in parallel the

afferent as well as the efferent arm of this bi-directional communication. Unique to this model

is its holistic appreciation of intricate bi-directional interactions under pathological and non-

pathological conditions. It is highly suitable for the investigation and evaluation of the

hierarchical, temporal and spatial communication patterns linking the brain and the peripheral

immune system under normal conditions, and to understand in more detail how, when and

where this interaction is disturbed under different pathological conditions. Its continual

evaluation and examination are expected to reveal multiple neuroimmune mechanisms with

further potential as subordinate therapy in the long-term treatment oftransplant medicine and

autoimmune disease. The constellation that behavioural conditioning by CsA was shown to be

possible in rodents and humans under different pathological and non-pathological conditions

is indicative of the relevance of this model in “placebo responses”.

The current findings reveal for the first time an intracellular mechanism that represents

the efferent mediation of brain-to-immune communication in the underlying taste-CsA

conditioning paradigm. However, the current findings do not rule out that CIS specifically

conditions the effects of the CsA nor that it is just a sympathetic activation of the IS. Therfore,

extending the conditioning model to include the implementation of other immunosuppressive

agents would help to further assess the aspect of specificity of the model. It would also enable

us to assess the full potential of CIS and other immunosuppressive agents, in order to transfer

this model from bench to bedside as a supplemental therapy.


88

References Abraham RT, Wiederrecht GJ (1996) Immunopharmacology of rapamycin. Annu Rev Immunol 14:483-510.

Ader, R. (2003). "Conditioned immunomodulation: research needs and directions." Brain Behav Immun 17 Suppl 1: S51-

7.

Ader, R. and N. Cohen (1975). "Behaviorally conditioned immunosuppression." Psychosom Med 37(4): 333-40.

Ader, R. and N. Cohen (1982). "Behaviorally conditioned immunosuppression and murine systemic lupus

erythematosus." Science 215(4539): 1534-6.

Ader, R. and N. Cohen (2001). Psychoneuroimmunology. R. Ader, D. Felten and N. Cohen. New York, Academic Press. 2:

3-34.

Andrykowski, M. A. (1988). "Defining anticipatory nausea and vomiting: differences among cancer chemotherapy

patients who report pretreatment nausea." J Behav Med 11(1): 59 - 69.

Banks, W. A., S. A. Farr, et al. (2001). "Intravenous human interleukin-1alpha impairs memory processing in mice:

dependence on blood-brain barrier transport into posterior division of the septum." J Pharmacol Exp Ther

299(2): 536-41.

Benedetti, F., I. Rainero, et al. (2003). "New insights into placebo analgesia." Curr Opin Anaesthesiol 16(5): 515-9.

Benschop, R. J., M. Rodriguez-Feuerhahn, et al. (1996). "Catecholamine-induced leukocytosis: early observations,

current research, and future directions." Brain Behav Immun 10(2): 77 - 91.

Besedovsky, H. O. and A. D. del Rey (2007). "Physiology of psychoneuroimmunology: a personal view." Brain Behav

Immun 21(1): 34-44.

Bovbjerg, D. (2003). "Conditioning, cancer, and immune regulation." Brain Behav Immun 17 Suppl 1: S58 - 61.

Bovbjerg, D. H., G. H. Montgomery, et al. (2005). "Evidence for classically conditioned fatigue responses in patients

receiving chemotherapy treatment for breast cancer." J Behav Med 28(3): 231 - 7.

Bovbjerg, D. H., W. H. Redd, et al. (1990). "Anticipatory immune suppression and nausea in women receiving cyclic

chemotherapy for ovarian cancer." J Consult Clin Psychol 58(2): 153 - 7.

Buske-Kirschbaum, A., C. Kirschbaum, et al. (1994). "Conditioned manipulation of natural killer (NK) cells in humans

using a discriminative learning protocol." Biol Psychol 38(2-3): 143-55.

Buske-Kirschbaum, A., C. Kirschbaum, et al. (1992). "Conditioned increase of natural killer cell activity (NKCA) in

humans." Psychosom Med 54(2): 123-32.

Castes, M., Palenque, M., Canelones, P., Hagel, I., Lynch, N. (1998). Classic conditioning and placebo effects in the

bronchiodilator response of asthmatic children. Neuroimmunomodulation. 5: 70.

Coghlan, V. M., B. A. Perrino, et al. (1995). "Association of protein kinase A and protein phosphatase 2B with a

common anchoring protein." Science 267(5194): 108-11.

Dadds, M. R., D. H. Bovbjerg, et al. (1997). "Imagery in human classical conditioning." Psychol Bull 122(1): 89 - 103.

Dantzer, R., J. C. O'Connor, et al. (2008). "From inflammation to sickness and depression: when the immune system

subjugates the brain." Nat Rev Neurosci 9(1): 46-56.

de la Fuente-Fernandez, R., M. Schulzer, et al. (2002). "The placebo effect in neurological disorders." Lancet Neurol

1(2): 85-91.

Dekker, E., H. E. Pelser, et al. (1957). "Conditioning as a cause of asthmatic attacks; a laboratory study." J Psychosom

Res 2(2): 97-108.

DiLorenzo, T. A., P. B. Jacobsen, et al. (1995). "Sources of anticipatory emotional distress in women receiving

chemotherapy for breast cancer." Ann Oncol 6(7): 705 - 11.

Doenlen, R. (2008). Fingerprints of Neural Activity after Peripheral Immune Challenges: An Experimental

Investigation of the Bi-directional Communication between the Central Nervous System and the Immune

System

Institute for Behavioral Sciences. Zurich, ETH Zurich. Dr. sc..


89

Elenkov, I. J., R. L. Wilder, et al. (2000). "The sympathetic nerve--an integrative interface between two supersystems:

the brain and the immune system." Pharmacol Rev 52(4): 595-638.

Exton, M. S., M. Schult, et al. (1998). "Behavioral conditioning prolongs heart allograft survival in rats." Transplant Proc

30(5): 2033.



Exton, M. S., M. Schult, et al. (1999). "Conditioned immunosuppression makes subtherapeutic cyclosporin effective

via splenic innervation." Am J Physiol 276(6 Pt 2): R1710-7.



Exton, M. S., S. von Horsten, et al. (1998). "Behaviorally conditioned immunosuppression using cyclosporine A: central

nervous system reduces IL-2 production via splenic innervation." J Neuroimmunol 88(1-2): 182-91.

Exton, M. S., A. Elfers, et al. (2000). "Conditioned suppression of contact sensitivity is independent of sympathetic

splenic innervation." Am J Physiol Regul Integr Comp Physiol 279(4): R1310-5.

Gauci, M., A. J. Husband, et al. (1994). "Pavlovian conditioning of nasal tryptase release in human subjects with

allergic rhinitis." Physiol Behav 55(5): 823-5.

Ghanta, V. K., T. Miura, et al. (1988). "Augmentation of natural immunity and regulation of tumor growth by

conditioning." Ann N Y Acad Sci 521: 29-42.

Giang, D. W., A. D. Goodman, et al. (1996). "Conditioning of cyclophosphamide-induced leukopenia in humans." J

Neuropsychiatry Clin Neurosci 8(2): 194-201.

Gibertini, M. (1996). "IL1 beta impairs relational but not procedural rodent learning in a water maze task." Adv Exp

Med Biol 402: 207-17.

Goebel, M., D. Hübell, et al. (2005). "Behavioral conditioning with interferon beta-1a in humans." Physiol Behav 84(5):

807 - 14.

Goebel, M. U., Meykadeh, N., Kou, W., Schedlowski, M., Hengge, U.R. (2008). "Behavioral conditioning of anti-

hisatmine effects in patients with allergic rhintis." Psychoth Psychosom In press.

Goebel, M. U., A. E. Trebst, et al. (2002). "Behavioral conditioning of immunosuppression is possible in humans."

FASEB J 16(14): 1869 - 73.

Gorczynski, R. M. (1990). "Conditioned enhancement of skin allografts in mice." Brain Behav Immun 4(2): 85-92.

Grochowicz, P. M., M. Schedlowski, et al. (1991). "Behavioral conditioning prolongs heart allograft survival in rats."

Brain Behav Immun 5(4): 349-56.

Grossman, Z., R. B. Herberman, et al. (1992). "Neural modulation of immunity: conditioning phenomena and the

adaptability of lymphoid cells." Int J Neurosci 64(1-4): 275-90.

Hashimoto, Y. and T. R. Soderling (1989). "Regulation of calcineurin by phosphorylation. Identification of the

regulatory site phosphorylated by Ca2+/calmodulin-dependent protein kinase II and protein kinase C." J Biol

Chem 264(28): 16524-9.

Hiramoto, R., B. Solvason, et al. (1990). "Effect of reserpine on retention of the conditioned NK cell response."

Pharmacol Biochem Behav 36(1): 51-6.

Ho, S., N. Clipstone, et al. (1996). "The mechanism of action of cyclosporin A and FK506." Clin Immunol Immunopathol

80(3 Pt 2): S40-5.

Hsueh, C. M., J. S. Kuo, et al. (1999). "Involvement of catecholamines in recall of the conditioned NK cell response." J

Neuroimmunol 94(1-2): 172-81.

Irie, M., S. Nagata, et al. (2004). "Diazepam attenuates conditioned histamine release in guinea pigs." International

Journal of Psychophysiology 51(3): 231-238.

Jacobsen, P. B., D. H. Bovbjerg, et al. (1993). "Anticipatory anxiety in women receiving chemotherapy for breast

cancer." Health Psychol 12(6): 469 - 75.


90

Kemeny, M. E., L. J. Rosenwasser, et al. (2007). "Placebo response in asthma: A robust and objective phenomenon." J

Allergy Clin Immunol 119(6): 1375-81.

Klauck, T. M., M. C. Faux, et al. (1996). "Coordination of three signaling enzymes by AKAP79, a mammalian scaffold

protein." Science 271(5255): 1589-92.

Kuo, J. S., S. F. Chen, et al. (2001). "The involvement of glutamate in recall of the conditioned NK cell response." J

Neuroimmunol 118(2): 245-55.

Lekander, M., C. J. Fürst, et al. (1995). "Anticipatory immune changes in women treated with chemotherapy for

ovarian cancer." Int J Behav Med 2(1): 1 - 12.

Longo, D. L., P. L. Duffey, et al. (1999). "Conditioned Immune Response to Interferon-[gamma] in Humans." Clinical

Immunology 90(2): 173-181.

Matsumoto, Y., M. Yoshida, et al. (2001). "Involvement of cholinergic and glutamatergic functions in working memory

impairment induced by interleukin-1beta in rats." Eur J Pharmacol 430(2-3): 283-8.

Matteson, S., J. Roscoe, et al. (2002). "The role of behavioral conditioning in the development of nausea." Am J Obstet

Gynecol 186(5 Suppl Understanding): S239 - 43.

Miranda, M. I., G. Ferreira, et al. (2002). "Glutamatergic activity in the amygdala signals visceral input during taste

memory formation." Proc Natl Acad Sci U S A 99(17): 11417-22.

Morrow, G. R., J. Lindke, et al. (1991). "Anticipatory nausea development in cancer patients: replication and extension

of a learning model." Br J Psychol 82 ( Pt 1): 61 - 72.

Nance, D. M. and V. M. Sanders (2007). "Autonomic innervation and regulation of the immune system (1987-2007)."

Brain Behav Immun.

Noelpp, B. and I. Noelpp-Eschenhagen (1952). "[Experimental bronchial asthma in the guinea pig. IV. Experimental

asthma in the guinea pig as an experimental model.]." Int Arch Allergy Appl Immunol 3(3): 207-17.

Noelpp, B. and I. Noelpp-Eschenhagen (1952). "[Experimental bronchial asthma of guinea-pigs. III. Studies on the

significance of conditioned reflexes; ability to develop conditioned reflexes and their duration under

stress.]." Int Arch Allergy Appl Immunol 3(2): 108-36.

Noelpp, B. and I. Noelpp-Eschenhagen (1952). "[The experimental bronchial asthma of the guinea pig. V. Experimental

pathophysiological studies.]." Int Arch Allergy Appl Immunol 3(4): 302-23.

Okamoto, S., K. Ibaraki, et al. (1996). "Ventromedial hypothalamus suppresses splenic lymphocyte activity through

sympathetic innervation." Brain Res 739(1-2): 308 - 13.

Pacheco-Lopez, G., M. B. Niemi, et al. (2005). "Neural substrates for behaviorally conditioned immunosuppression in

the rat." J Neurosci 25(9): 2330-7.

Pacheco-Lopez, G., C. Riether, et al. (2009). "Calcineurin inhibition in splenocytes induced by Pavlovian conditioning."

FASEB J 23 (4): 1161-7.

Pacheco-Lopez, G., H. Engler, et al. (2006). ”Expectations and associations that heal: Immunomodulatory placebo

effects and its neurobiology.” Brain Behav Immun 20(5): 430-36.

Rachal Pugh, C., M. Fleshner, et al. (2001). "The immune system and memory consolidation: a role for the cytokine IL-

1beta." Neurosci Biobehav Rev 25(1): 29-41.

Ramirez-Amaya, V., B. Alvarez-Borda, et al. (1998). "Differential effects of NMDA-induced lesions into the insular

cortex and amygdala on the acquisition and evocation of conditioned immunosuppression." Brain Behav

Immun 12(2): 149-60

Ramirez-Amaya, V. and F. Bermudez-Rattoni (1999). "Conditioned enhancement of antibody production is disrupted

by insular cortex and amygdala but not hippocampal lesions." Brain Behav Immun 13(1): 46-60.

Redd, W. H., M. R. Dadds, et al. (1993). "Nausea induced by mental images of chemotherapy." Cancer 72(2): 629 - 36.

Riether, C., R. Doenlen, et al. (2008). "Behavioural conditioning of immune functions: how the central nervous system

controls peripheral immune responses by evoking associativel earning processes." Reviews in the

Neurosciences 19: 1-17.


91

Rubin, B. K. and J. B. Fink (2007). "Novel medications for asthma: a look at the future." Expert Opin Investig Drugs

16(6): 889-97.

Saito, M., Y. Minokoshi, et al. (1989). "Accelerated norepinephrine turnover in peripheral tissues after ventromedial

hypothalamic stimulation in rats." Brain Res 481(2): 298 - 303.

Schedlowski, M., W. Hosch, et al. (1996). "Catecholamines modulate human NK cell circulation and function via

spleen-independent beta 2-adrenergic mechanisms." J Immunol 156(1): 93 - 9.

Schneider, H., F. Pitossi, et al. (1998). "A neuromodulatory role of interleukin-1beta in the hippocampus." Proc Natl

Acad Sci U S A 95(13): 7778-83.

Scott, J. D. (2006). "Compartmentalized cAMP signalling: a personal perspective." Biochem Soc Trans 34(Pt 4): 465-7.

Solvason, H. B., R. N. Hiramoto, et al. (1989). "Naltrexone blocks the expression of the conditioned elevation of natural

killer cell activity in BALB/c mice." Brain Behav Immun 3(3): 247-62.

Stockhorst, U., S. Spennes-Saleh, et al. (2000). "Anticipatory symptoms and anticipatory immune responses in

pediatric cancer patients receiving chemotherapy: features of a classically conditioned response?" Brain

Behav Immun 14(3): 198 - 218.

Tracey, K. J. (2007). "Physiology and immunology of the cholinergic antiinflammatory pathway." J Clin Invest 117(2):

289-96.

Turnbull, J. W. (1962). "Asthma conceived as a learned response." J Psychosom Res 6: 59-70.

Vettor, R., R. Fabris, et al. (2002). "Neuroendocrine regulation of eating behavior." J Endocrinol Invest 25(10): 836 - 54.

Williams, R. O. (2002). "Cutting edge: A-kinase anchor proteins are involved in maintaining resting T cells in an

inactive state." J Immunol 168(11): 5392-6.

Xie, Y., S. Frede, et al. (2002). "Beta-adrenoceptor-induced inhibition of rat splenocyte proliferation: cytokine gene

transcription as the target of action." Immunobiology 206(4): 345-53.

Curriculum Vitae

92

Curriculum Vitae Personal data

First name Carsten

Last name Riether

Date of birth May 04, 1980

Nationality German

Home address Schwerzenbachstrasse 18, 8117 Fällanden, Switzerland

Office Address Swiss Federal Institute of Technology (ETH) Zurich

Laboratory of Psychology and Behavioral Immunobiology

Turnerstrasse 1, 8096 Zurich, Switzerland

[email protected]

Education

2005-2008 Doctoral studies (Ph.D.)

Laboratory of Psychology and Behavioral Immunobiology

Swiss Federal Institute of Technology (ETH) Zurich

2005 Master Thesis

Title: Modulation of primary and secondary T cell activation by

different immunosuppressive mechanisms

Supervisor: Barbara Metzler, Ph.D.

Autoimmunity and Transplantation unit

Novartis Institue for Biomedical Research, Basel, Switzerland

2002-2005 Master studies

Biotechnology

École Superieure de Biotechnologie de Strasbourg (ESBS), France

2000-2002 Undergraduate studies in Biology

Albert-Ludwig University Freiburg i. Brsg., Germany

1991-1999 Max-Planck Gymnasium Lahr, Germany, Abitur (University

Entrance Examination)

List of publications

93

List of publications PEER-REVIEWED PUBLICATIONS

Niemi MB, Pacheco-Lopez G, Engler H, Riether C, Doenlen R, Schedlowski M. (2008). “Neuro-

Immune Associative Learning” in: Handbook of Neurochemistry and Molecular

Neurobiology: Neuroimmunology. Ed A. Lajtha, A Galoyan and HO Besedovsky, Springer

Press, pp. 123-150.

Pacheco-Lopez G, Niemi MB, Engler H, Engler A, Riether C, Doenlen R, Espinosa E, Oberbeck R ,

Schedlowski M. (2008). Weaken taste-LPS association during endotoxin tolerance.

Physiol Behav, 93 (1-2), 261-266.

Riether C*, Doenlen R*, Pacheco-López G, Niemi MB, Engler A, Engler H, Schedlowski M (2008).

Behavioural conditioning of immune functions: How the CNS controls peripheral

immune responses by evoking associative learning processes. Rev. Neurosciences. 19, 1-

17. (*equal contribution)

Eckart C, Engler H, Riether C, Kolassa S, Elbert T, Kolassa IT. Cortisol diurnal profiles and quality

of sleep in survivors of organized violence living under stressful conditions.

Psychoneuroendocrinology. In press.

Riether C*, Pacheco-López G*, Doenlen R, Engler H, Niemi MB, Engler A, Kavelaars A, Heijnen

CJ, Schedlowski M (2009). Calcineurin inhibition in splenocytes induced by Pavlovian

conditioning. FASEB J 23 (4): 1161-7. (*equal contribution)

Engler H, Doenlen R, Riether C, Engler A, Niemi MB, Besedevosky H, Del Rey A, Pacheco-Lopez

G, Feldon J, Schedlwoski M (2009). Time-dependent alterations of peripheral immune

parameters after nigro-striatal dopamine depletion in a rat model of Parkinson’s

disease. BBI. In Press.

SUBMITTED MANUSCRIPTS

Sommerhof A, Basler M, Riether C, Engler H, Gröttrup M. Attenuation of the cytotoxic T

lymphocyte response to lymphocyticchoriomeningitis virus in mice subjected to chronic

social stress. FASEB J

Weber K, Engler H, Awiszus B, Riether C, Schedlowski M, Rockstroh B. Early life stress affects

diurnal and reactive salivary cortisol in psychatric patients. Psychoneuroendocrinology

List of publications

94

PUBLISHED ABSTRACTS Riether C, Doenlen R, Engler H, Niemi MB, Pacheco-López G, Schedlowski M (2007).

Behaviorally conditioned suppression of peripheral T lymphocyte function and

signalling. Exp Dermatol, 16 (4), 347-383.

Riether C, Doenlen R, Engler H, Niemi MB, Pacheco-López G, Schedlowski M (2007). Specificity

of immunopharmacological conditioned effects. Brain Behav Immun, 21 (4), e1–e42.

Documents

Rights / License: Research Collection In Copyright - …41789/... · Peripheral mediatory mechanisms of behaviorally conditioned immunosuppression ... behaviorally conditioned immunosuppression