Pavlovian conditioning of immune function: animal investigation and the challenge of human application

Pavlovian Conditioning of Immunological and Neuroendocrine Functions

The phenomenon of behaviorally conditioned immunological and neuroendocrine functions has been investigated for the past 100 yr. The observation that associative learning processes can modify peripheral immune functions was first reported and investigated by Ivan Petrovic Pavlov and his co-workers. Their work later fell into oblivion, also because so little was known about the immune system’s function and even less about the underlying mechanisms of how learning, a central nervous system activity, could affect peripheral immune responses. With the employment of a taste-avoidance paradigm in rats, this phenomenon was rediscovered 45 yr ago as one of the most fascinating examples of the reciprocal functional interaction between behavior, the brain, and peripheral immune functions, and it established psychoneuroimmunology as a new research field. Relying on growing knowledge about efferent and afferent communication pathways between the brain, neuroendocrine system, primary and secondary immune organs, and immunocompetent cells, experimental animal studies demonstrate that cellular and humoral immune and neuroendocrine functions can be modulated via associative learning protocols. These (from the classical perspective) learned immune responses are clinically relevant, since they affect the development and progression of immune-related diseases and, more importantly, are also inducible in humans. The increased knowledge about the neuropsychological machinery steering learning and memory processes together with recent insight into the mechanisms mediating placebo responses provide fascinating perspectives to exploit these learned immune and neuroendocrine responses as supportive therapies, the aim being to reduce the amount of medication required, diminishing unwanted drug side effects while maximizing the therapeutic effect for the patient’s benefit.

Learned placebo responses in neuroendocrine and immune functions

The phenomenon of learned placebo responses in neuroendocrine and immune functions is a fascinating example of communication between the brain and both the endocrine and peripheral immune systems. In this chapter, we will give a short overview of afferent and efferent communication pathways, as well as the central mechanisms, which steer the behavioral conditioned immune response. Subsequently, we will focus on data that provides evidence for learned immune responses in experimental animals and learned neuroendocrine and immune placebo responses in humans. Finally, we will take a critical look at these learning protocols, to determine whether or not they can be considered a viable additional treatment option to pharmacological regimens in clinical routine. This is fundamental, since there are still a number of issues, which need to be solved, such as the potential reproducibility, predictability, and extinction of the learned neuroendocrine and immune responses. Together, these findings not only provide an excellent basis to increase our understanding of human biology but may also have far reaching clinical implications. They pave the way for the ultimate aim of employing associative learning protocols as supportive treatment strategies in pharmacological regimens. As a result, medication levels may be reduced, as well as their unwanted side effects, providing a maximized therapeutic outcome to the benefit of the patient.

Neurobehavioural activation during peripheral immunosuppression

Like other physiological responses, immune functions are the subject of behavioural conditioning. Conditioned immunosuppression can be induced by contingently pairing a novel taste with an injection of the immunosuppressant cyclosporine A (CsA) in an associative learning paradigm. This learned immunosuppression is centrally mediated by the insular cortex and the amygdala. However, the afferent mechanisms by which the brain detects CsA are not understood. In this study we analysed whether CsA is sensed via the chemosensitive vagus nerve or whether CsA directly acts on the brain. Our experiments revealed that a single peripheral administration of CsA increases neuronal activity in the insular cortex and the amygdala as evident from increased electric activity, c-Fos expression and amygdaloid noradrenaline release. However, this increased neuronal activity was not affected by prior vagal deafferentation but rather seems to partially be induced by direct action of CsA on cortico-amygdaloid structures and the chemosensitive brainstem regions area postrema and nucleus of the solitary tract. Together, these data indicate that CsA as an unconditioned stimulus may directly act on the brain by a still unknown transduction mechanism.

Repeated recall of learned immunosuppression: evidence from rats and men

Akin to other physiological responses, the immune system can be modified, via Pavlovian or behavioral conditioning. It is unknown, however, whether and to what extent learned immune responses can be repeatedly recalled over time. Here we demonstrate in both rats and humans that repeated contingent pairing of a novel taste (conditioned stimulus, CS) together with the immunosuppressive drug cyclosporine A as unconditioned stimulus (US) leads to the acquisition of a learned immunosuppression. Sole presentation of the CS caused a significant inhibition of interleukin (IL)-2 and interferon (IFN)-γ production by rat splenic T cells and human peripheral T lymphocytes, closely mimicking the effect of the drug. More importantly, a comparable suppression of cytokine production was also observed after a second, unreinforced exposure to the CS that was separated from the first evocation by an interval of 6 (rats) or 11 (humans)days, respectively. Together, our findings demonstrate that a learned immunosuppression can be repeatedly recalled in both animals and humans, which is an important prerequisite for the implementation of conditioning paradigms as supportive therapy.

The learned immune response: Pavlov and beyond

The ability to associate physiological changes with a specific flavor was most likely acquired during evolution as an adaptive strategy aimed at protecting the organism while preparing it for danger. The behaviorally conditioned or learned immune response is an exquisite example of the bidirectional communication between the central nervous system (CNS) and the peripheral immune system. How is it possible that specific immuno-modulating properties of a drug or substance (unconditioned stimulus) can be re-enlisted just by the mere re-exposure to a particular taste, odor or environment (conditioned stimulus)? To answer this key question, we review the neurobiological mechanism mediating this type of associative learning, as well as the pathways and mechanisms employed by the brain to harness the immune system during the execution of the conditioned immune response. Finally, we focus on the potential therapeutic relevance of such learned immune responses, and their re-conceptualization within the framework of "learned placebo effects".

Taste-immunosuppression engram: Reinforcement and extinction

Several Pavlovian conditioning paradigms have documented the brain's abilities to sense immune-derived signals or immune status, associate them with concurrently relevant extereoceptive stimuli, and reinstate such immune responses on demand. Specifically, the naturalistic relation of food ingestion with its possible immune consequences facilitates taste-immune associations. Here we demonstrate that the saccharin taste can be associated with the immunosuppressive agent cyclosporine A, and that such taste-immune associative learning is subject to reinforcement. Furthermore, once consolidated, this saccharin-immunosuppression engram is resistant to extinction when avoidance behavior is assessed. More importantly, the more this engram is activated, either at association or extinction phases, the more pronounced is the conditioned immunosuppression.

Neural substrates for behaviorally conditioned immunosuppression in the rat

We have previously demonstrated behaviorally conditioned immunosuppression using cyclosporin A as an unconditioned stimulus and saccharin as a conditioned stimulus. In the current study, we examined the central processing of this phenomenon generating excitotoxic lesions before and after acquisition to discriminate between learning and memory processes. Three different brain areas were analyzed: insular cortex (IC), amygdala (Am), and ventromedial nucleus of the hypothalamus (VMH). The results demonstrate that IC lesions performed before and after acquisition disrupted the behavioral component of the conditioned response (taste aversion). In contrast, Am and VMH lesions did not affect conditioned taste aversion. The behaviorally conditioned suppression of splenocyte proliferation and cytokine production (interleukin-2 and interferon-gamma) was differentially affected by the excitotoxic lesions, showing that the IC is essential to acquire and evoke this conditioned response of the immune system. In contrast, the Am seems to mediate the input of visceral information necessary at the acquisition time, whereas the VMH appears to participate within the output pathway to the immune system necessary to evoke the behavioral conditioned immune response. The present data reveal relevant neural mechanisms underlying the learning and memory processes of behaviorally conditioned immunosuppression.

Mechanisms of the placebo effect and of conditioning

A placebo is a sham treatment, such as a pill, liquid, or injection without biological activity, used in pharmacology to control for the activity of a drug. However, in many cases this placebo induces biological or psychological effects in the human. Two theories have been proposed to explain the placebo effect: the conditioning theory, which states that the placebo effect is a conditioned response, and the mentalistic theory, which sees the patient's expectation as the primary cause of the placebo effect. The mechanisms involved in these processes are beginning to be understood through new techniques of investigation in neuroscience. Dopamine and the endorphins have been clearly shown to be mediators of placebo effects. Brain imaging has demonstrated that placebos can mimic the effect of the active drugs and activate the same brain areas. This is the case for placebo-dopamine in Parkinson's disease, for placebo-analgesics or antidepressants, and for placebo-caffeine in the healthy subject. It remains to be understood how conditioning and expectation are able to activate memory loops in the brain that reproduce the expected biological responses.

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

Immune responses can be modulated by Pavlovian conditioning techniques. In this study, to evaluate the conditionability of antibody response via a single-trial conditioning paradigm, we used a protein antigen ovalbumin as an unconditioned stimulus (UCS) that was paired with a novel taste of saccharin in a single-trial learning protocol. A significant enhancement of anti-ovalbumin antibody production was observed in the conditioned rats at Days 15, 20 and 25 after re-exposure to the conditioned stimulus. The pattern of conditioned antibody response is similar to that of antigen-induced antibody response. Furthermore, to identify the involvement of a limbic brain structure in the expression of conditioned antibody response, immediate-early gene c-fos expression was used as a marker of neuronal activation to detect the functional activation in the insular cortex (IC) in response to the conditioned stimulus. The re-exposure of conditioned rats to the conditioned stimulus resulted in a significant increase of c-Fos immunoreactivity in all three areas of the IC including the agranular, dysgranular, and granular areas, suggesting that IC is involved in the neural mechanism of expression of conditioned immune response.

Conditioned taste aversion with lipopolysaccharide and peptidoglycan does not activate cytokine gene expression in the spleen and hypothalamus of mice

Several reports show that behavioural and physiological components of the acute phase reaction can be conditioned. However, the mechanisms responsible for these effects remain obscure. The underlying assumption that the changes observed in conditioned animals are dependent on a conditioned production of cytokines has never been demonstrated. In the present study, the possibility of conditioning the production of cytokines or molecules implicated in their signalling pathways was tested by submitting mice to conditioned taste aversion with a new saccharin taste paired with intraperitoneal (i.p.) injections of lipopolysaccharide (LPS, 0.83 microg/g) or peptidoglycan (PGN, 20 microg/g). After two conditioning sessions, conditioned mice developed a clear aversion to saccharine that was not associated with activation of genes of the cytokine network either at the periphery, or in the hypothalamus, as demonstrated by a macroarray approach and confirmed by real time RT-PCR. In contrast, there was an activation of the genes coding for nuclear factor kappa B (NFkappaB) and mitogen activated protein kinase (MAPK) signalling pathways in the spleen and to a lesser extent in the hypothalamus. This modulation of the NFkappaB and MAPK signalling pathways is interpreted in terms of a possible conditioned sensitisation of the immune system.

Behavioural endocrine immune-conditioned response is induced by taste and superantigen pairing

Administration of bacterial superantigen, such as staphylococcal enterotoxin B (SEB), induces in vivo stimulation of T cell proliferation and cytokine production such as interleukin-2 (IL-2). It has been previously reported that SEB administration induces fever, c-Fos expression in the brain, and hypothalamus-pituitary-adrenal axis activation, demonstrating that the brain is able to sense and respond to SEB. Previously it had been shown that immune functions can be behaviourally conditioned pairing a novel gustatory stimulus together with an immunomodulatory drug or an antigen. We designed an experimental protocol using Dark Agouti rats in which saccharin taste, as conditioned stimulus, was paired with an i.p. injection of SEB (2 mg/kg), as unconditioned stimulus. Six days later, when conditioned animals were re-exposed to the conditioned stimulus they displayed strong conditioned taste avoidance to the saccharin. More importantly, re-exposure to the conditioned stimulus significantly increased IL-2, interferon-gamma and corticosterone plasma levels, in comparison with conditioned animals which had not been re-exposed to saccharin taste. These results demonstrate a behavioural-immune-endocrine conditioned response using a superantigen as unconditioned stimulus. In addition, they illustrate the brain abilities to mimic the unconditioned effects of a superantigen by yet unknown mechanisms.

Conditioned place aversion with interleukin-1beta in mice is not associated with activation of the cytokine network

Several distinct findings argue in favor of conditioning of some components of the acute phase reaction. However, the possibility of a conditioned cytokine response has not been assessed. In the present study, this possibility was tested by submitting mice to place aversion conditioning with interleukin-1beta (2 microgram/mouse, ip) as the unconditioned stimulus and an odorous compartment of a two-compartment cage as the conditioned stimulus. After two pairings, conditioned mice developed place aversion towards the odorous compartment. However, this behavioral conditioning was not accompanied by any alteration in peripheral (spleen) and brain (hypothalamus) cytokine levels (interleukin-1, interleukin-6, and interleukin-10). These data do not support the possibility of conditioned alterations in the cytokine network.

Pavlovian conditioning of endotoxin-tolerance in rats

The most fascinating example of the bi-directional interaction between the central nervous system (CNS) and immune system is the behavioral conditioning of immune functions. We therefore investigated the behavioral conditioning of lipopolysaccharide (LPS)-induced endotoxin tolerance using the taste aversion paradigm. The conditioned stimulus (CS) saccharin was paired with the unconditioned stimulus (UCS) LPS over a five (CONDl) or four (COND2) days learning trial. Controls received drinking water with (SHAM) or without (UNT) LPS. Endotoxin tolerance was tested by determination of LPS-induced tumor necrosis factor (TNF)-alpha release. After the avoidance of the induced endotoxin-tolerance the CS saccharin was re-presented in all experimental groups. A the end of the re-exposure period a complete endotoxin tolerance was noticed in the CONDl- and COND2-group. In contrast, no effect of saccharin administration was observed in the SHAM- or UNT-group. Our data demonstrate for the first time the behavioral conditioning of endotoxin tolerance. Furthermore, these results contribute new aspects to the mechanisms underlying the development and modulation of endotoxin tolerance.

The inflammatory reflex

Inflammation is a local, protective response to microbial invasion or injury. It must be fine-tuned and regulated precisely, because deficiencies or excesses of the inflammatory response cause morbidity and shorten lifespan. The discovery that cholinergic neurons inhibit acute inflammation has qualitatively expanded our understanding of how the nervous system modulates immune responses. The nervous system reflexively regulates the inflammatory response in real time, just as it controls heart rate and other vital functions. The opportunity now exists to apply this insight to the treatment of inflammation through selective and reversible 'hard-wired' neural systems.

Behavioral conditioning of immunosuppression is possible in humans

Behavioral conditioned immunosuppression has been described in rodents as the most impressive demonstration of brain-to-immune system interaction. To analyze whether behavioral conditioned immunosuppression is possible in humans, healthy subjects in this double-blind, placebo-controlled study were conditioned in four sessions over 3 consecutive days, receiving the immunosuppressive drug cyclosporin A as an unconditioned stimulus paired with a distinctively flavored drink (conditioned stimulus) each 12 h. In the next week, re-exposure to the conditioned stimulus (drink), but now paired with placebo capsules, induced a suppression of immune functions as analyzed by the IL-2 and IFN-gamma mRNA expression, intracellular production, and in vitro release of IL-2 and IFN-gamma, as well as lymphocyte proliferation. These data demonstrate for the first time that immunosuppression can be behaviorally conditioned in humans.

Behaviorally conditioned immunosuppression in the rat is regulated via noradrenaline and beta-adrenoceptors

Using Cyclosporin A (CsA) as an unconditioned stimulus has previously demonstrated that behaviorally conditioned inhibition of splenocyte proliferation and cytokine production is mediated via the splenic nerve. Therefore, we currently examined the adrenergic modulation of conditioned suppression of splenocyte function. Chemical sympathectomy via 6-OHDA completely blocked the conditioned suppression of splenocyte proliferation to mitogens and cytokine (IL-2, IFN-gamma) production. Furthermore, administration of beta-adrenoceptor antagonist propranolol abrogated the conditioned effect on splenocyte proliferation. Supporting the position that conditioning is beta-adrenergic-dependent, addition of beta-adrenoceptor agonist, but not alpha-adrenoceptor agonists, to splenocytes in vitro mimicked the conditioned suppression of splenocyte functions, with these effects blocked by propranolol. Therefore, these data indicate that behavioral conditioning of splenocyte function in the rat is regulated by the sympathetic nervous system, predominantly via beta-adrenergic mechanisms.