Mens sana in corpore sano--and vice versa. The role of the autonomic nervous system in the immune-neuroendocrine dialogue.

Adrenergic and cholinergic transmitters of the autonomic nervous system have important roles in the mutual interrelationships between the brain and the immune system. Besides expressing functional adrenergic and cholinergic receptors, lymphocytes and other immune cells were found to synthesize and release catecholamines and acetylcholine pointing to a possible role of these mediators in the intrinsic regulation of the immune system. In this review we will summarize concepts of Psychoneuroimmunology on the basis of data as obtained in vitro and in experimental studies in animal models, and discuss their relevance to human clinical medicine.

Prolonged alpha-adrenergic stimulation causes changes in leukocyte distribution and lymphocyte apoptosis in the rat.

We have previously shown in the rat model that acutely or chronically increased peripheral catecholamines lead to suppression of lymphocyte responsiveness via alpha(2)-adrenoceptor activation. Here we investigated the effects of alpha-adrenergic treatment on total leukocyte numbers and proportions of leukocyte subsets in peripheral blood and lymphoid tissues. It was found that a 12-h treatment with subcutaneously implanted tablets, one containing norepinephrine (NE) and one propranolol, leads to an increase in total blood leukocyte counts, due to a pronounced increase in granulocytes. In contrast, the numbers of all classes of lymphocytes other than NK cells were decreased. This decrease in blood lymphocytes is apparently not due to redistribution, since in the thymus, spleen, mesenteric and peripheral lymph nodes, the total numbers of lymphocytes were decreased as well, without any changes in subpopulations. Analogous results were obtained with rats adrenalectomized before the catecholamine treatment. Animals that received the alpha-adrenergic treatment displayed significantly more apoptotic cells in the lymphoid organs, as determined by the TUNEL technique. In the spleen, the enhanced rate of apoptosis was confined to the white pulp; red pulp areas exhibited significantly fewer apoptotic cells. Thus, an increased alpha-adrenergic tone in rats led to a general loss of lymphocytes due to lymphocyte directed apoptosis that was independent of glucocorticoids.

The tumor-associated shift in immunoglobulin G1/G2 is expressed at the messenger RNA level of peripheral blood B lymphocytes in patients with gynecologic malignancies.

BACKGROUND: Previously, it could be demonstrated that human patients with malignant diseases of various tissues exhibited characteristic and highly significant changes in the serum patterns of immunoglobulin (Ig)G subclasses, consisting of a decrease in IgG1 and an increase in IgG2 relative to total IgG. The aim of the current study was to determine whether this phenomenon was detectable at the level of IgG-producing B lymphocytes. METHODS: Using a competitive reverse transcriptase polymerase chain reaction specific to IgG1 and IgG2, the gene expression of these 2 IgG subclasses in peripheral B cells from 10 patients with carcinomas of various sites within the female reproductive tract and 10 healthy controls was quantitatively determined, in parallel with the concentrations of the respective serum proteins. RESULTS: Absolute levels of IgG subclass messenger ribonucleic acid (mRNA) showed a slight but not significant decrease in IgG1 and an increase in IgG2 in patients with gynecologic malignancies. However, the ratio of IgG1 to IgG2 expression showed a highly significant (P < 0.001) decrease in tumor patients compared with healthy controls, and corresponded to the change in the ratio of IgG1 to IgG2 serum proteins. CONCLUSIONS: These data suggest that the shifts in the serum patterns of IgG1 and IgG2 observed in patients with gynecologic malignancies are due to irregular biosynthesis of these IgG subclasses at the B-cell level.

The dialogue between the brain and immune system involves not only the HPA-axis.

Starting out from our previous observations that defects in the immune system-brain feedback predispose to pathogenic immune responses, our interest focuses at the roles of adrenergic/cholinergic neurotransmitters in brain-immune interactions. We have shown in rodent models that 1) both catecholamines and acetylcholine are potent modulators of peripheral immune functions, 2) cholinergic signals are involved in the afferent signalling of the immune system, and 3) lymphocytes not only express functional adrenergic and cholinergic receptors, but synthesize and release neurotransmitters, such as acetylcholine, in quantitative dependence of differentiation and activation. Studies are presently being initiated to investigate the role(s) of these non-neuronal neurotransmitters within immune tissues, and to explore the relevance of excitatory amino acids as important central neurotransmitters in the brain-immune system dialogue.

In vivo immunomodulation by peripheral adrenergic and cholinergic agonists/antagonists in rat and mouse models.

Our work is devoted to defining relationships between the immune system and the adrenergic and cholinergic systems in vivo. In the rat model, we have shown that the cells of different immune compartments express the genes of a defined set of adrenergic/cholinergic receptors, and it was shown that lymphocytes are a site of non-neuronal production of norepinephrine and acetylcholine. Furthermore, using implantable slow-release tablets containing adrenergic or cholinergic agonists/antagonists, distinct and partly opposite effects were observed on peripheral immune functions. Concerning sympathetic immunoregulation, our data--in contrast to those of other studies--suggest that an enhanced adrenergic tonus leads to immunosuppression primarily via alpha 2-receptor-mediated mechanisms. Beta-blockade strongly enhances this effect, most likely by inhibition of pineal melatonin synthesis. In recent experiments on the kinetics it was found that the continuous alpha-adrenergic treatment entails a strong suppression of cellular responsiveness during the first few hours, which is increasingly followed by a general loss of lymphocytes in blood and lymphoid organs most likely due to enhanced apoptosis. More recently, we have extended our studies to the mouse model. First data obtained with RNAse protection assays suggest a biphasic effect on the gene expression of several cytokines in spleen cells due to adrenergic in vivo treatment.

Early rheumatoid arthritis is associated with diminished numbers of TH1 cells in stimulated peripheral blood.

Rheumatoid arthritis (RA) has been associated with an altered TH1/TH2 balance. Here we present first results with a technique to quantitate stimulated TH1 and TH2 subsets in whole blood by means of cytofluorimetric detection of intracytoplasmic TH1 and TH2 cytokines. A group of 10 patients suffering from initial stages of RA exhibited a significantly reduced percentage of TH1 cells (11.0 +/- 2.0%) in the peripheral blood as compared to 13 healthy, age matched controls (28.8 +/- 2.0%). The TH2 response, as determined by intracellular expression of IL-4, remained unchanged. The data may indicate a defect in the TH0-TH1 differentiation or/and a selective trapping of TH1 cells into the affected joints.

The IgG1/G2 subclass shift--a sensitive, tissue non-specific marker for malignancy. Diagnostic performance with squamous cell carcinoma of the head and neck.

A significant decrease in %IgG1 accompanied by an increase in %IgG2 in total serum IgG has been previously reported as a highly sensitive marker for detecting early stages of carcinomas of various localizations. Here we investigated the question as to whether this phenomenon is also observed in sera of patients with squamous cell carcinoma of the head-neck region (SCC-HN), and to evaluate its diagnostic performance in the post-operative monitoring. Using quantitative affinity chromatography, serum concentrations of IgG1, IgG2 and total IgG were determined in 81 patients with different stages of primary and untreated SCC-HN, in 51 SCC-HN patients in post-therapeutical follow up, and in 33 patients with organ matched benign diseases. The data were compared with a total of 174 healthy controls. It was found that (i) 105 SCC-HN patients exhibited a mean value of 56.0 +/- 0.7% IgG1, which likewise differed from healthy controls (63.2 +/- 0.5) and benign diseases (61.5 +/- 1.0) with P < 0.0005, (ii) sensitivities and specificities for discriminating primary malignancies from healthy controls were 70 and 74% respectively, and from benign diseases 65 and 76%, (iii) highest sensitivities and specificities were observed with post-therapeutic cases suffering from tumour recurrence (88% and 75%) or patients with distant metastases (87% and 86%), (iv) apparently tumour-free post-therapeutic patients showed a mean %IgG1 not different from the normal value. The decrease in %IgG1 accompanied by increased %IgG2 is an efficient, sensitive and early marker of SCC-HN, which appears particularly useful for the post-therapeutic monitoring.

Adrenergic/cholinergic immunomodulation in the rat model--in vivo veritas?

For several years, our group has been studying the in vivo role of adrenergic and cholinergic mechanisms in the immune-neuroendocrine dialogue in the rat model. The main results of these studies can be summarized as follows: (1) exogenous or endogenous catecholamines suppress PBL functions through alpha-2-receptor-mediated mechanisms, lymphocytes of the spleen are resistant to adrenergic in vivo stimulation, (2) direct or indirect cholinergic treatment leads to enhanced ex vivo functions of splenic and thymic lymphocytes leaving PBL unaffected, (3) cholinergic pathways play a critical role in the "talking back" of the immune system to the brain, (4) acetylcholine inhibits apoptosis of thymocytes possibly via direct effects on thymic epithelial cells, and may thereby influence T-cell maturation, (5) lymphocytes of the various immunological compartments were found to be equipped with the key enzymes for the synthesis of both acetylcholine and norepinephrine, and to secrete these neurotransmitters in culture supernatants.

Melatonin and the immune system.

In recent years, melatonin, i.e. the major endocrine product of the pineal gland, was investigated as to its possible regulatory role in the communication between the neuroendocrine and the immune systems. First indications that melatonin may be an endocrine immunomodulator came from early reports about antitumor effects in animals and humans. Since then evidence has accumulated suggesting that melatonin-as a well-preserved molecule during evolution-is indeed involved in the feedback between neuroendocrine and immune functions. At present we begin to discover molecular mechanisms, by which melatonin affects cellular functions in general, and from the variety of possible direct and indirect interactions it appears that melatonin may play a complex physiological role in neuroimmunomodulation. In this article we want to give a critical review of the numerous reports on melatonin influencing immune functions, and to discuss the possible different mechanisms of action, which were suggested recently.

Characterization of the spontaneous apoptosis of rat thymocytes in vitro.

Unlike splenic or blood lymphocytes, rat thymocytes spontaneously undergo continuously increasing apoptosis during culture. In this study we characterized apoptotic thymus cells of rats according to cell size, nuclear dye binding and surface marker expression. Furthermore, the effects of cell density in culture, the age of the donor animals, glucocorticoids, and inhibition of protein synthesis were studied. It was found that: (1) apoptotic rat thymocytes are recognized in flow cytometry as small, acridine orange low, CD4low, CD8high cells; (2) the rate of apoptosis is dependent on the cell density in the culture in a biphasic manner; (3) thymic apoptosis increases with age of the donor animal in fresh, as well as in 24-hour cultivated cell suspensions; (4) neither adrenalectomy nor in vivo or in vitro treatment with the glucocorticoid antagonist RU 38486 influenced spontaneous apoptosis of thymocytes, and (5) inhibition of protein synthesis, which decreases apoptosis induced by corticosterone, had no effect on spontaneous apoptosis of thymocytes.

Reactive disulfide bonds in immunoglobulin G. A unique feature in serum proteins of different species.

A reactive disulfide bond (SS)* was detected and characterized in IgG of humans, rats and mice by virtue of disulfide interchange with dithionitrobenzoate. (SS)* was found exclusively in human IgG1 and rat IgG2b. In human IgG1 (SS)* was identified as the upper one of the two interheavy bridges in the hinge, where it appears to take part in complement activation. The biological significance of (SS)* in IgG was underlined by the fact that no other serum proteins were found to exhibit a similar reactivity.

Beta-blockade enhances adrenergic immunosuppression in rats via inhibition of melatonin release.

We have recently shown in rats that an in vivo treatment with catecholamines via alpha 2-receptors leads to a pronounced suppression of T- and B-cell mitogen responses of peripheral blood lymphocytes (PBL), provided that a beta-blocker is administered concomitantly. Since melatonin (MEL) reportedly has stress-protective effects on several immune functions, and since the release of MEL from the pineal gland is inhibited by beta-blockade, we tested the effect of MEL substitution on T- and B-cell mitogen responses of PBL in rats treated with two s.c. implanted retard tablets containing noradrenaline (NA) and propranolol. It was found that an oral treatment with MEL (about 40 micrograms/animal) abolished the adrenergic immunosuppression. Furthermore, functional pinealectomy induced by constant light had a similar enhancing effect on the alpha 2-adrenergic immunosuppression as observed with beta-blockers, whereas PBL from animals kept at the regular light/dark interval were resistant to the treatment with the selective alpha 2-agonist clonidine. It is concluded that endogenous MEL effectively protects rat PBL from adrenergic immunosuppression, and that beta-blockers enhance the immunosuppressive property of alpha 2-adrenergic agents via blocking the night-time release of MEL.

Adrenergic suppression of peripheral blood T cell reactivity in the rat is due to activation of peripheral alpha 2-receptors.

A 20-h treatment of rats with catecholamines using s.c.implantable retard tablets markedly suppresses the in vitro reactivity of peripheral blood (PBL) T lymphocytes, provided that beta-receptors are blocked with propranolol (Felsner et al., 1992). The results can be summarized as follows: (i) the suppressive effect of noradrenaline+propranolol to the concanavalin A (ConA) response of PBL was abolished by the simultaneous application of the alpha-blocker phentolamine. Using selective agonists, the relevant receptor was identified to belong to the alpha 2-subtype. (ii) The alpha-adrenergic suppression of the PBL T cell response was likewise observed in adrenalectomized animals, which rules out the participation of secondarily induced glucocorticoids. Furthermore, the combination of noradrenaline with the watersoluble beta-blocker nadolol was equally effective to suppress the ConA response of PBL. (iii) An analogous alpha-mediated suppression of T cell function of PBL, but not spleen cells, was observed 1 h after i.p. treatment with tyramine, which leads to the release of endogenous noradrenaline. From these results it is concluded that the adrenergic suppression of PBL T cell functions is primarily due to the activation of peripheral alpha 2-receptors and that it is likewise observed under acute indirect sympathomimetic treatment.

Cholinergic signals to and from the immune system.

This article reviews recent data from our laboratory towards the impact of the autonomous nervous system on mutual interactions between the immune system and the central nervous system. Using a pharmacological approach in rats it is shown that shifts in the adrenergic/cholinergic balance in vivo affect in vitro functions of the non-specific and specific immune system, whereby adrenergic and cholinergic stimulation in general have opposite effects. A high degree of integration appears to exist between cells of the immune system with the cholinergic system. Lymphocytes were found to react to acetylcholine, but are also able to produce and to degradate this neurotransmitter. In addition, changes in the cholinergic tonus were found to affect immune signaling to the brain and to protect thymocytes from apoptosis, possibly via a direct effect on thymic epithelial cells.

Cholinergic stimulation modulates apoptosis and differentiation of murine thymocytes via a nicotinic effect on thymic epithelium.

Effects of cholinergic agonists/antagonists on apoptosis and differentiation of murine thymus cells were investigated in two in vitro models. Treatment with 10(-7) M carbachol was found to counteract the effects of a cortical thymus epithelial cell line (TEC 1.4), on apoptosis and the ratio of CD4 CD8 DP/DN cells in cocultured fetal thymus lobes. A medullary line (TEC 2.3) did not influence apoptosis in fetal thymus lobes. Both TEC lines had the same strong apoptotic effect on thymus cells in suspension, but only the effect of TEC 1.4 was counteracted by carbachol. This cholinergic influence on TEC 1.4 cells is mediated via nicotinic cholinergic receptors, since d-tubocurarine, but not atropine, effectively blocked the effect of carbachol. The results suggest that cholinergic signals to thymic epithelial cells may have regulatory influence on thymic differentiation and selection processes.

Continuous in vivo treatment with catecholamines suppresses in vitro reactivity of rat peripheral blood T-lymphocytes via alpha-mediated mechanisms.

A 20 h continuous treatment of rats with catecholamines, using subcutaneously implantable retard tablets, had either no (adrenaline, isoproterenol, midodrine) or a slight (noradrenaline) suppressive effect on the in vitro responsiveness of peripheral blood T-lymphocytes. A marked suppression of the mitogen response ensued when adrenaline, noradrenaline or midodrine, but not isoproterenol, was applied together with the beta-receptor blocker propranolol, whereas the combination with the alpha-receptor blocker phentolamine had no effect. The mitogen response of splenic lymphocytes was not affected by any of these treatments. This alpha-mediated adrenergic suppression of peripheral blood T-cells was not correlated with general metabolic alterations, shifts in white blood cell counts or CD4+/CD8+ subsets, or with elevated glucocorticoid levels. The data suggest that to consistently influence the reactivity of rat peripheral blood lymphocytes by chronic adrenergic stimuli in vivo requires both high catecholamine levels and a bias towards alpha-adrenergic receptivity.

Bidirectional interaction between immune and neuroendocrine systems. An experimental approach.

The purpose of this article is to review recent experimental data from our laboratory on immune-neuroendocrine communications, whereby the following findings will be discussed: (i) Experiments using an experimental stress model in rats revealed different doses of a defined stressor to exert opposite effects on the in vitro reactivity of peripheral blood lymphocytes (PBL). Stress effects are also dependent on the tissue origin of lymphocytes, and they seem to be exclusively mediated by adrenal hormones. (ii) The balance between the sympathetic and parasympathetic nervous system may be critical in extrinsic immunoregulation: Experimentally induced hypercatecholaminemia in rats seem to protect lymphocytes from immunosuppressing effects of other, endogenous stress hormones, but causes suppression of PBL activation, if beta-receptors are blocked at the same time. Chronic cholinergic stimuli exert enhancing effects on the cells of the thymus. (iii) A defect in the immune-neuroendocrine crosstalk may contribute to the occurrence of forbidden immune reactions, as has been shown in spontaneous and experimentally induced autoimmune diseases in animals models. Recent data indicate that the parasympathetic nervous system is involved in the transmission of immune messages to the central nervous system.

Norepinephrine triggers medullar epinephrine depletion during normoglycemia.

Our experiments did show that chronic 12 hours administration of norepinephrine (NE) to rats by means of subcutaneously implantable retard tablets, led to a highly significant epinephrine (E) depletion of the adrenal medulla during normoglycemia. The expected rise of free plasma NE at 6 and 12 hours was accompanied by increased free plasma E values at 12 hours. At this very time point the liver contents of glycogen and free intracellular glucose showed their most pronounced decrease. Since at 12 hours both liver glycogen and medullar E values were at their lowest, a second experiment was performed to examine a possible causal relationship. In order to curb the breakdown of liver glycogen, rats were force fed with 50% glucose solution 9 hours after NE tablet implantation. Glucose feeding not only caused a much less pronounced liver glycogen fall at 12 hours, but, at the same time also prevented E depletion of the adrenal medulla. These observations suggested that rapid fall of liver glycogen and/or liver intracellular free glucose might be the trigger for medullar E depletion, even before hypoglycemia.

Preventive action of phentolamine on adrenaline induced blood glucose elevation in humans.

A comparison of the action of adrenaline infusion and a combined adrenaline + alpha blocker (phentolamine, Regitine) infusion on blood glucose (BG), plasma immunoreactive insulin (IRI), BG/IRI ratio, C-peptide, and plasma cortisol levels was made in healthy young human subjects. The purpose of the experiment was to check, whether alpha block could abolish adrenaline-induced enhancement of blood glucose levels. The results show that during enhanced adrenaline levels, the use of regitine could indeed normalize blood glucose levels, not so much by increasing the IRI secretion, but by diminishing adrenaline-induced liver glycogenolysis via alpha receptors. This could be a model to prevent stress (adrenaline) induced metabolic deviations in diabetics, especially before and during predictable stress situations, e.g. examinations or surgery.