Stress and immunity: A unifying concept

Veterinary Immunology and Immunopathology, 20 (1989) 263-312 Elsevier Science Publishers B.V., Amsterdam - - Printed in The Netherlands

Stress and I m m u n i t y : a U n i f y i n g Concept

J. FRANK T. GRIFFIN

Department of Microbiology, University of Otago, Dunedin (New Zealand) TABLE OF CONTENTS 1.0 INTRODUCTION 2.0 PHYSIOLOGY OF STRESS 2.1 Definition of stress 2.2 Host factors which influence stress 2.2.1 Genetics 2.2.2 Perception 2.3 Types of stressors 2.3.1 Environmental 2.3.2 Behavioural 2.3.3 Psychological 3.0 NEUROENDOCRINOLOGY OF STRESS 3.1 Sympathetic adrenomedullary stimulation 3.1.1 The 'flight-fight' syndrome 3.1.2 CNS activation and immunocompetence 3.2 The hypothalamic-pituitary-adrenal (HPA) response to stress 3.2.1 The 'conservation-withdrawal' syndrome 3.2.2 Glucocorticoids and immunocompetence 3.3 Neuropeptides-neurohormones 3.3.1 'Learning, behaviour and pain' 3.3.2 Neurohormones and immunocompetence 3.3.3 Opioids and immunocompetence 4.0 N E U R O P E P T I D E - L Y M P H O I D CELL INTERACTIONS 4.1 Innervation of the LRS 4.2 Receptors on immunocytes for hormones and neuropeptides 4.3 Immune factors which influence CNS function 5.0 STRESS AND IMMUNOCOMPETENCE 5.1 Infectious disease susceptibility 5.2 Stress and cancer 6.0 E X P E R I M E N T A L MODELS FOR STRESS IN IMMUNOLOGY 6.1 Stress and humoral immunity 6.2 Stress and cell-mediated immunity 6.2.1 Stress and CMI in vivo 6.2.2 Stress and CMI in vitro 6.3 A new large animal model for stress studies 6.4 Conditioning of immune reactivity 7.0 STRESS: DIAGNOSIS AND MANAGEMENT 8.0 CONCLUSIONS 9.0 REFERENCES

0165-2427/89/$03.50

© 1989 Elsevier Science Publishers B.V.

263

264 1.0 INTRODUCTION "The constancy of the'milieu interieur' is the condition of free and independent existence" (Claude Bernard).

The domestication of animals by man throughout the past 8000 years represents the single most significant challenge to their free and independent existence. This process has dramatically changed the behaviour of animals in the small number of species which have been domesticated. It has been successful because of the peculiar traits of individual species, which include; tolerance of humans, a non-migratory nature, catholic food preference, and adaptability to a wide range of environments (Hale, 1969). However, inappropriate human interactions with animals may challenge their degree of adaptability and result in an abuse of their wellbeing, which may limit their productive potential or, in extreme cases, threaten their lives. In recent years there has been considerable public debate concerning the wellbeing of animals, farmed under intensified management practices, and laboratory animals used in biological research, so it has become necessary for scientists to establish criteria by which animal wellbeing may be defined objectively (Ewbank, 1985 ). Depending on the nature of management and the animals, genetic background, age and experience, the host response to environmental stress will vary both qualitatively and quantitatively. The net effect is that different forms of stress may evoke neurophysiological responses which may produce host reactions ranging from minor stimulation to total exhaustion. The endocrine and neuropeptide factors elicited in response to varying stimuli may ultimately impact on the immune system to produce immunoenhancement, immunosuppression or have no effect. A proper understanding of the patterns of the physiological and neuroendocrine responses in animals exposed to different conditions of management or stress is fundamental, if we are to discover the true impact of environmental stimuli on disease susceptibility. Such information would also be of great value in optimizing aspects of preventive medicine for treatment of disease in animals and man. 2.0 PHYSIOLOGYOF STRESS In 1878, Claude Bernard first recognised the physiological response of animals to environmental stimuli, when he described the "milieu interieur" of the animal, characterized by its constancy, and the external environment, characterized by its variability. Cannon (1914) further refined this concept as 'homeostasis', where a steady state is obtained by the optimum interaction of

265

counteracting processes within the host. He also defined the 'flight-fight' reflex of animals exposed to undue stress and identified the potentially damaging effects of stress on the individual (Cannon, 1929). The physiological basis of the stress response was attributed to activation of endocrinological factors involving the sympathetic adrenomedullary axis (Cannon, 1935). In the next decade, Selye (1946) described the response of laboratory animals exposed experimentally to noxious stimuli, which he classified as the 'general adaptation syndrome' (GAS). The syndrome was classified as having three distinct phases: first, an alarm reaction; second, a resistance phase; and finally, if stress was not dissipated, exhaustion of the biological system. Because different noxious stimuli produced an apparently similar physiological response, involving activation of the pituitary-adrenal-cortical axis, he concluded that the response to stress was 'non-specific'. This led to the widely held view that the stress largely involved non-specific production of corticosteroids, and it has subsequently led to the measurement of glucocorticoids as the primary indicator of stress and animal wellbeing. As the field of neuroendocrinology evolved, Selye's concept of the non-specific stress response has been refuted. Mason (1968) showed that different types of stress, such as fasting and heat shock, produce dramatically different corticosteroid levels in experimental monkeys. He demonstrated characteristically different responses unique for each stressor, and showed that activation of the autonomic adrenomedullary response as well as the pituitary adrenocortical response occurred, and produced characteristic responses to the different stressors. It is now widely accepted that there are unique but different endocrine responses to different physical, chemical and psychological stressors, which vary within and between different species of animals (Dantzer and Mormede, 1985). Experimental studies carried out during the past 20 years have seen the marriage of psychology, neuroendocrinology and immunology into a new subspeciality called psychoneuroimmunology. Research in this multifaceted system has identified and integrated the physiological factors produced by the neurological and immunological systems. Soluble factors produced by the neurological and immunological tissues include neurotransmitters, neuropeptides, adrenocorticotrophin (ACTH), catecholamines, corticosteroids, opioids and cytokines; and lymphokines and monokines (Besedovsky et al., 1985). 2.1 Definition of stress

In the veterinary context Fraser et al. (1975) have defined stress as "an abnormal or extreme adjustment in the physiology of an animal to cope with adverse effects of its environment and management". In this paper the adverse effect will be designated the stressor. The term stress has been further qualified by other workers (Selye, 1973; Ewbank, 1985) to include the term distress.

266 This term is used to identify the extreme response to adverse stimuli which cause a damaging pathophysiological reaction in the host, producing associative changes in behaviour, physiology and disease susceptibility. It has also been recognised that other forms of stimuli potentiate the physiological responses of the host without producing any adverse effect, and such responses are said to involve eustress (Selye, 1973).

2.2 Host factors which influence stress 2.2.1 Genetics Evolutionary changes have led to marked differences in the adaptative physiological response of different species exposed to stress. This is reflected in variations in the endocrine response of species to individual stressors and in the variable effects of specific endocrine factors on the target cells of different animal species. Variations of responses are also found within species and may be accentuated by selective breeding, age, sex and conditioning. 2.2.2 Perception The most important factor which influences stress is the individual's perception of the challenge posed by the specific stimulus. Factors which influence the individual's perception of stress include bonding during rearing, the novelty or predictability of the stimulus, and training which frustrates the avoidance response and evokes a state of 'learned helplessness' (Moberg, 1985). Conditioning to stimuli may also occur where the association between an aversive chemical stimulus and an immunomodulating agent, given in combination, can evoke a conditioned state of immunosuppression (Ader and Cohen, 1975) or immunoenhancement (Gorczynski et al., 1982) when the aversive stimulus is applied in the absence of the immunomodulating agent. 2.3 Types of stressors Stressors include a wide variety of environmental stimuli which evoke significant homeostatic alterations in the host. They may result from natural changes in the environment, or artificial changes, imposed by management constraints, or be caused by a range of experimental stimuli which have been used to study stress physiology. The effect produced by exposure to stress may be influenced by the severity of the stressor, the time during which it is applied (acute vs. chronic), or whether the animal can escape the stressor if it is applied repeatedly (escapable vs. inescapable).

2.3.1 Environmental Natural exposure to extremes of heat or cold under climatic conditions, or artificially in animal housing or holding facilities, are amongst the commonest

267

CNS Recognition

Endocrine Response

Autonomic Nervous Stimulation

Systemic Reactivity

"Flight/Fight"

Physiological

"ConservationWithdrawal"

Reactivity

Pathological Reactivity

I Disease I

"Exhaustion"

Fig. 1. Organisedmulti-systemresponse to stress. stressors. Physical restraint associated with separation, t r e a t m e n t or transport of animals is another important stressor. Environmental changes which produce strange or novel sounds, sights, odours or tastes can also evoke stress. Drugs or chemicals used in m a n a g e m e n t or t r e a t m e n t of animals can act as stressors, while toxic products released by infectious agents, environmental pollutants or inadequate ventilation can have a similar effect. Shearing, docking and castration, carried out during routine management, also cause stress.

2.3.2 Behavioural Overcrowding, hierarchical challenge, weaning, exposure to unfamiliar surroundings or isolation can have a major behavioural impact on animals and evoke stress. Changes or restrictions in diet can also act as stressors alone or in combination with stressors such as physical restraint.

268

2.3.3 Psychological Capture of wild animals, or the exposure of domesticated animals to restraint, transport or management extremes may evoke adaptative stress in the host. In the context of Selyes general adaptation syndrome, the response is recognizable initially as anxiety, which may progress to fright or terror. Failure of the animal to express the 'flight-fight' response results in the expression of anger or rage, and if the stimulus persists, frustration or helplessness. Because stress results largely from the individual's perception of the threat posed by the stimulus, rather than its nature per se, a cognitive or psychological component is central to all stress. The biological response within the host which results from exposure to stress is outlined in Fig. 1, and produces an impact on many of the organs within the affected host. 3.0 NEUROENDOCRINOLOGY OF STRESS

The cognitive recognition of adverse stimuli evokes a series of neuroendocrine responses outlined schematically in Fig. 2. A graded response occurs involving four discrete systems of neuroendocrine response. These include: I, autonomic nervous stimulation; II, hypothalamic-pituitary-adrenal (HPA) axis responses; III, neuropeptide and neurotransmitter production; IV, neuroimmunological peptides and receptors, which facilitate bidirectional signals to be transmitted between the central nervous system (CNS) and lymphoreticular system (LRS).

3.1 Sympathetic adrenomedullary stimulation 3.1.1 The 'flight-fight' syndrome Cognitive stimuli received by the cerebral cortex of the brain produce neurological impulses which cause hypothalamic stimulation of the autonomic nervous system with an associated production of sympathetic neurotransmitters. Activation of the autonomic nervous system produces an immediate response which results in the production of catecholamines by two discrete pathways: (a) directly, by release of norepinephrine from sympathetic nerve endings; (b) indirectly, by release of epinephrine and small amounts of norepinephrine from the innervated adrenal medulla. The somatic response to catecholamines causes dramatic increases in cardiovascular function and metabolism to effect the increased physical potential of the host which is manifest by the 'flight-fight' response. More specific responses are manifest within the haematological compartment where epineph~ rine may cause vasoconstriction and splenic contraction with an associated release of splenic erythrocytes and leukocytes into the blood vasculature (Cross et al., 1989). Because different types of psychological and physical stimuli produce vary-

269

# III.

IV.

Beta-era orphin

ACTH

Cortic Corticosteroids

Growth Hormone

Catecholamines

I.

Other Hormones

Enkephallns

|

~V

Fig. 2. Stress and immunodulation.

ing levels of norepinephrine (NE) and epinephrine (E) (Birkenboch, 1983), it is necessary to understand the interaction between these factors and their impact on host function. Increased plasma levels of NE are considered (Axelrod and Reisine, 1984) to result from activation of sympathetic nerves and to be a feature of the acute stress response. Elevation of plasma E levels results from activation of the adrenomedullary tissues and is closely related to corticosteroid production within the adrenal cortex following stimulation of the HPA system. There also appears to be a relationship between catecholamine and production of adrenocorticotrophin. High levels of brain NE are known to inhibit ACTH production by the pituitary, whereas plasma E activates ACTH production (Axelrod and Reisine, 1984). There appears to be an association between stress, learning and adaptation, and catecholamine levels in animals and humans (Levine, 1985). Whereas footshock stress in rats causes a simul-

270

taneous increase in plasma catecholamine and corticosteroid levels, their respective levels change dramatically following termination of stress. Catecholamine levels return to normal immediately following termination of stress, whereas corticosteroid levels remain high for 30 min. Adaptation to chronic stress results in persistent elevation of plasma catecholamines whereas plasma corticosteroids return to basal levels much earlier.

3.1.2 CNS activation and immunocompetence The influence of the integrity of the cortical areas of the brain has been demonstrated by electrolytic ablation of neocortical areas of the brain in mice (Biziere et al., 1985). A dramatic difference was seen between the immunocompetence of animals which had lesions produced in the left (LCL) or right (RCL) neocortical areas of the brain. Lesions in the left cortical area of mouse brain produced dramatic changes in the immunocompetence of the affected animals. There was a significant depression in the alloreactivity and the lymphoproliferative activity of splenocytes from these animals when stimulated with phytohaemagglutinin (PHA) or concanavalin A (ConA) but the response to pokeweed mitogen (PWM) remained unaltered. Natural killer (NK) activity and specific splenic IgG plaque-forming cell (PFC) reactivity were significantly reduced in animals with LCL. By contrast these lesions had no effect on IgM PFC or monocyte activity of the affected animals. In striking contrast, RCL in animals caused no alterations in their lymphoproliferation, NK, B cell or monocytic activity. These data show that the integrity of the brain cortical region of mouse, especially in the left neocortex, has a significant impact on immunocompetence. Because hypothalamic activation is considered to have an important effect on neuroendocrine function, a number of studies (Hall et al., 1985; Roszman et al., 1985; Belluardo et al., 1987) have examined the effect of lesions on the hypothalamus on immunocompetence. Roszman et al. (1985) have demonstrated that electrolytic lesions in the anterior hypothalamus (AHT) cause a significant reduction in the NK cell, PFC and ConA lymphoproliferative response of splenic cells from treated rats. The effect of AHT lesions decreased with increasing age of the rats and was independent of corticosteroid levels. Lesions produced in the mouse brain nucleus locus ceruleus (Hall et al., 1980), which is the major site for catecholamine production in the mouse brain, cause a significant reduction in norepinephrine and impair the production of factors which cause granular monocyte (GM) activation in lesion-affected mice. Lesions produced in the median hypothalamus consistently produced the suppressive effect but no such change was evident when lesions were produced in the anterior or posterior hypothalamic regions (Hall and Goldstein, 1985). Belluardo et al. (1987) have found that electrothermocoagulation (ETC) lesions in the median hypothalamic (MH) region of mouse brain also produced a significant systemic effect on immunocompetence of the affected animals.

271 There was a significant reduction in the NK and large granular lymphocyte (LGL) activity in these animals. The hypophyseal function of these animals was intact and there was no suggested role of corticosteroids in the observed suppressive effect. ETC lesions in the anterior or posterior region of the hypothalamus had no effect on the NK activity of the test animals. Altered activity through prostaglandin (PGE2) production was proffered as a likely explanation for the immunosuppressive effect produced by MH lesions. Whereas the integrity of the median hypothalamus appears vital in the immunocompetent response of the mouse, the intact function of the anterior region of the hypothalamus appears more vital in the rat (Cross et al., 1984). The central role of sympathetic nerves in immunocompetence (Miles et al., 1981 ) is seen from studies involving chemical sympathectomy (Besedovsky et al., 1979). Such treatment results in reduced immunological activity in the spleen and lymph nodes of immunized animals. The NE content of rat spleen following immunization is inversely related to antibody production, where low levels of NE are found in spleen at the peak of the splenic PFC response. Specific pathogen-free (SPF) rats have significantly lower splenic NE levels when exposed to environmental pathogens, as compared with conventionally raised rats which mount a much less vigorous response to infectious organisms. Activation of fl receptors on macrophage membranes (Hadden, 1983) augments adenylate cyclase activity. It is also thought that activation of fl adrenergic receptors on lymphocytes may assert an immunosuppressive effect (Feldman et al., 1987) by activation of c-AMP activity (Bourne et al., 1974) and that fl adrenergic activation of mast cells inhibits histamine release and has an ameliorating effect on the anaphylactic response. Acute stress causes can increase in the levels of catecholamines in both blood and urine. It may also be manifest in altered levels of catecholamines within the brain, especially the hypothalamus (Kvetnansky et al., 1978). Because catecholamine levels in the brain influence the production of corticotropinreleasing factor (CRF) from the hypothalamus, it has been proposed (Axelrod and Reisine, 1984) that the catecholamines produced in the acute phase of stress may act indirectly to stimulate adrenocortical activity through the hypothalamic-pituitary-adrenal axis (HPA). 3.2 The hypothalamic-pituitary-adrenal (HPA) response to stress 3.2.1 The 'conservation-withdrawal' syndrome Clinical observations by Engel (1967) identified two distinct behavioural responses to stress in which the acute 'flight-fight' reaction was followed by a 'conservation-withdrawal' reaction. It has subsequently been demonstrated that the 'flight-fight' reaction is a result of the physiological changes due to the production of catecholomines by the sympathetic-adrenal medullary system (Axelrod and Reisine, 1984). Failure of this system to resolve stress leads

272

to activation of the hypothalamic-pituitary-adrenocortical response which is manifest behaviourally by the 'conservation-withdrawal' reaction. Persistence of HPA activation causes biological changes which may produce a pre-pathological state where the animal becomes susceptible to disease (Fig. 1). If this latter state persists, pathological changes will occur in the host (Kagan and Levi, 1974). Harris (1948) first demonstrated that stimulation of the anterior hypothalamus produced corticotropin-releasing factor (CRF), which stimulates the anterior pituitary gland to secrete adrenocorticotropins (ACTH). The production of ACTH in response to stress may also be influenced by a number of other hormones and neuropeptides. Neuronal norepinephrine (Mains and Eipper, 1981) and epinephrine (Axelrod and Reisine, 1984) from the adrenal medulla, vasoactive intestinal peptide (VIP) (O'Dorisio, 1987) and vasopressin (Gibbs, 1986) all cause an increase in pituitary ACTH production. The target organ affected by ACTH is the adrenal cortex which is stimulated to produce significant levels of circulating glucocorticoids (cortisol/corticosterone). Glucocorticoids (GC) have an effect on most of the homeostatic systems in the body, but are thought to have their dominant effect on gluconeogenesis, and on the inflammatory and immunological systems. The name glucocorticoid implicates the primary effect produced by activation of the HPA system which is the production of a hyperglycaemia through enhanced hepatic gluconeogenesis. This metabolic response facilitates the 'conservation-withdrawal' response. Depending on the species, different responses are produced by the action of cortisol on inflammation and blood leukocyte dynamics. Animals with relatively high lymphocyte numbers, such as mice, rabbits, chickens and cattle, respond with a lymphopaenia and neutrophilia but produce a net decrease in leukocytes. Those with relatively low lymphocyte numbers (dogs, cats, horses, cattle, pigs and man) respond with a leukocytosis because of the neutrophilia (Burton and Beljan, 1970; Gwazdauskas et al., 1980; Blecha et al., 1982 ). There is generally an eosinopaenia in most animals which produce high levels of cortisol. GC exerts a generalized anti-inflammatory action by causing a reduction in capillary permeability and endothelial swelling, reducing vascular flow. Activation of the HPA system and the associated increase in the levels of the plasma cortisol has been used widely to monitor stress levels in domestic animals (Franzmann et al., 1975; Johnston and Buckland, 1976; Leach, 1982). In cattle, elevated plasma cortisol levels have been found following handling (Crookshank et al., 1979), weaning (Gwazdauskas et al., 1978) castration and dehorning (Johnston and Buckland, 1976), acute pain (Stephens, 1980), forced exercise (Arave et al., 1978), and transportation (Shaw and Nichols, 1964; Crookshank et al., 1979). However, it must be recognised that corticosteroids are only one hormone among many which are produced in the stress syndrome in animals (Roth, 1985). Thus, single blood samples analysed for corticoste-

273

roids are unlikely to give a meaningful measure of the relevance of such factors in the overall response to stress.

3.2.2 Glucocorticoids and immunocompetence Selye's original studies on stress and endocrine function (Selye, 1946) suggested that activation of the adrenal gland to produce plasma glucocorticoids (GC) was central to all alterations in host physiology during stress. Until Mason ( 1968 ) identified the effect of adrenomedullary activation and the production of catecholamines in stress, corticosteroids reigned supreme as the hormones of stress. Whereas GC production obviously exerts a major impact on metabolism, inflammation, the lymphatic organs and peripheral blood leukocytes (Marsh and Rasmussen, 1960), it is important to reevaluate their role in light of the variety of hormones and peptides that are now known to be produced during stress. The central pathway by which the adrenal cortex is stimulated during stress has now been thoroughly described and is known to involve the hypothalamic (CRF) - pituitary (ACTH) - adrenocortical (GC) pathway (Kreiger, 1983). From the earliest studies it was recognised that GC production in many species of animals and man was associated with adrenal hypertrophy, thymic involution, lymphocytopaenia, eosinopaenia and neutrophilia (Selye, 1946; Jensen, 1969; Riley, 1981; Cupps and Fauci, 1982). The administration of GC to animals of many species has been shown to markedly increase their susceptibility to infectious disease and cause activation of latent infection. The specific examples evident from their use in cattle show that GC injection decreases resistance to infectious bovine rhinotracheitis virus (IBRV) (Davies and Duncan, 1974), coccidiosis (Niilo, 1970; Stockdale and Niilo, 1976), herpesvirus (Sheffey and Davies, 1972), fatal bovine viral diarrhoea (BVD) (Shope et al., 1976) and parasites (Callow and Parker, 1969). GC also appears to influence immunocompetence in cattle which show reduced levels of humoral immunity (Gwadzdauskas et al., 1978; May et al., 1979), and in vitro lymphocyte blastogenesis (Roth et al., 1982). By contrast, the administration of GC simultaneously with BVD vaccine causes increased levels of resistance to subsequent challenge (Roth and Kaeberle, 1983). The activation of adrenocortical production of GC by injection of cattle with ACTH does not always produce a lymphopaenia (Roth et al., 1982) and may have variable effects on immunocompetence (Roth and Kaeberle, 1983). Riley (1981) has found that the timing of administration of GC relative to immune challenge, also appears to be of considerable importance. In tumour rejection studies, he showed that administration of GC to rats 3-7 days prior to challenge with tumour increases the rate of tumour rejection while administration of GC after (4-6 days) implantation of tumour decreases rejection rates. The level of GC hormone may also be important in that a moderate increase in GC levels may cause immune enhancement rather than immunosuppression (Croiset et al., 1987). The effects of GC may be mediated by their influence on inflammatory cells,

274 antigen-presenting cells or lymphocytes (Kelso and Munk, 1984). GC acts through lipocortin (Blackwell et al., 1980) to limit the activation of phospholipase and cause a reduction in prostaglandins, thromboxane and leukotrienes. It influences the circulatory cell levels by its effect on high endothelial vessels, whereby it limits reentry of leukocytes from tissues, as well as causing margination of cells within the vasculature (Dougherty, 1952; Ellis, 1981). Steroids may also have a direct effect on the lymphoreticular system through their interaction with receptors on mononuclear cells (Comsa et al., 1982 ). GC has been shown to suppress the expression of class II MHC molecules on monocytes (Snyder and Unanue, 1982) and decrease production of interleukin-1 (IL-1) (MacDermott and Stacey, 1981). It inhibits transcription of ILl genes and post-transcriptional expression of mRNA (Knudsen et al., 1987 ). Production of tumour necrosis factor (TNF) by murine (Beutler et al., 1986) and human (Waage and Bakke, 1988) leukocytes is also inhibited by steroids. During characterization of interleukin-2 (IL-2) production by T-helper cells, Gillis et al. (1979) showed that steroids acted as potent inhibitors of IL-2. Naive T-cells appear much more sensitive to the effects of steroids than antigen-primed memory cells (Cupps and Fauci, 1982). Although there is no doubt that steroids can have a major influence on LRS function when present at significantly elevated levels, their effects may be much more varied when present at lower doses, or in species which are relatively resistant to the effects of steroids (Tornello et al., 1982). It may be that the role of steroids in stress has been overestimated because most experimental studies have been carried out on laboratory rodents (Parillo and Fauci, 1979), which are classified as corticosteroid-sensitive animals. By contrast primates (Parillo and Fauci, 1979) and domestic animals such as cattle (Roth and Kaeberle, 1982) and pigs (Westly and Kelley, 1984) belong to the corticosteroidresistant species.

3.3 Neuropeptides-neurohormones 3.3.1 'Learning, behaviour and pain' Because stress is affected mainly by the individual's perception of the external stimulus, it is important to consider the production and function of the brain peptides and transmitters which could influence the adaptive response following stimulation of the CNS. In recent years many peptides have been characterized and reclassified (Kreiger, 1983) as neurotransmitters and neuropeptides (Table 1 ). Whereas a few were discovered in association with CNS tissue and called the 'hypothalamic releasing hormones', others were first discovered in neurohypophyseal tissue or as hormonal factors associated with the pituitary or adrenal glands, or intestinal tissue. The fundamental role of these molecules in cell communication is suggested by their conservation throughout the microbial, plant and animal kingdoms (Roth et al., 1985). They have been isolated from unicellular organisms and

275 TABLE 1 Neuropeptides and neurohormones Hypothalamic -releasing hormones Tyrotropin-releasing hormone (TRH) Gonadotropin-releasing hormone (GRH) Growth hormone-releasing hormone (GHRH) Corticotropin-releasing factor (CRF) Somatostatin (So)

Neurohypophyseal hormones Vasopressin (VP) Oxytocin (O)

Pituitary peptides Adrenocorticotropic hormone {ACTH) fl-Endorphin (fl-End) ~-Melanocyte-stimulating hormone (~-MSH) Prolactin (PRL) Luteinizing hormone (LH) Growth hormone (GH) Thyrotropin {TH)

Gastrointestinal peptides Vasoactive intestinal peptide (VIP) Substance P (SP) Methionine-enkephalin {Met-ENK) Leucine-enkephalin {Leu-ENK) Insulin (I) Somatostatin (So)

plants and in the lower orders of animals such as coelentera and annelids. Their presence in animals lacking endocrine tissue points to their role in evolution as neurotransmitters. The peptides may be found in secretory or neuronal cells or only in neurons. The common amino acids shared by many of these peptides suggest that they may have arisen from a common primordial gene (Kreiger, 1983). A prototype molecule which has been widely studied is proopiomelanocortin (POMC). This precursor molecule can, following cleavage, produce a series of important neuropeptides, which play an important role in the sensory function of the CNS (Liolta et al., 1980). Products generated by POMC include melanocyte-stimulating hormone (MSH), ACTH and fl-endorphin (Kreiger, 1983). RNA hybridisation studies have shown that POMC genes are expressed within CNS tissues of higher animals, and their production has been associated primarily with hypothalamic activation. Recent work (Kapcala, 1988) has shown that ACTH and fl-endorphin can also be produced by cortical brain tissues. These factors have also been found in reproductive tissues of the ewe (Lim et al., 1983 ). Whereas the precise role of these peptides is not fully understood, a number of independent studies has implicated them as factors which are produced in increased amounts in animals subjected to stress (Takayama et al., 1986). The opioid peptides (Lewis et al., 1985), fl-endorphin (Haynes and Timms, 1987), vasopressin and oxytocin (Wideman and Murphy, 1985; Williams et al., 1985) and aminobutyric acid {Yonada et al., 1983) have all been found at increased levels in stressed animals. The production of neuropeptides causes activation of physiological and behavioural responses which are central to our understanding of stress (Gersh-

276

bein et al., 1941). The generalized impact of neuroendocrine factors, neuropeptides and neurotransmitters on host physiology has been reviewed recently by Breazile (1987). Their more selective impact on the host response is seen where altered CNS levels of neurohormones and neuropeptides have been associated with learning, feeding, immobilisation, temperature regulation and pain control. CRF (Brown et al., 1982), ACTH, MSH, enkephalins, fl-endorphins (Bohus, 1979; McGaugh, 1983), oxytocin (Bohus et al., 1978) and vasopressin are associated with learning and memory. Feeding behaviour is influenced by fl-endorphins and opioids (Grandison and Guidolta, 1977). Temperature regulation is influenced by fl-endorphins, opioids (Holaday et al., 1978) and vasopressin (Veale et al., 1981 ). Displacement activities in response to aversive stimuli, such as nibbling and fighting, have been associated with altered levels of ACTH, cortisol, endorphins and opioids (Ulrich and Azrin, 1962; Conner et al., 1971; Panksepp et al., 1978). It is likely that neuropeptides may have an important role in many of the behavioural and physiological changes found in stressed animals, especially through their effect of increasing tolerance to the pain associated with exposure to noxious stimuli. 3.3.2 Neurohormones and immunocompetence Serotonin interacts with receptors on lymphocytes and is known to influence immune function, because destruction of serotoninergic neurons, or injection of serotonin antagonists or agonists, all affect immunocompetence (Hall and Goldstein, 1985). Injection with serotonin precursors increases serotonin levels in guinea pigs and rats and causes an associated reduction in delayed type hypersensitivity (DTH) and antibody production (Boranic et al., 1987). Serotonin antagonists increase splenic IgM and IgG PFC levels in mice, and destruction of serotoninergic neurons increases humoral immunity. Hypothalamic serotonin levels are increased in the brain following immunisation. The major effect of serotonin may be through its activation of hypothalamic CRF. In this way serotoninergic activity seems to have the directly opposite effect of fi-adrenergic activity within the hypothalamus. Substance P interacts with lymphocytes and inflammatory cells and causes an increase in PHA transformation of lymphocytes (Payan et al., 1984; McGilles et al., 1987), enhanced release of inflammatory mediators and neutrophil chemotaxis. The enhanced effect on PHA reactivity is inhibited by SP analogues. Because of its enhancement of the inflammatory system, it may be important in the urticarial and allergic reactions which occur in certain stress conditions (Breazile, 1987). Vasoactive intestinal peptide (VIP) also has an effect on inflammatory and immune cell function (O'Dorisio, 1987). It has an enhancement on lymphocyte circulation and migration, mast cell mediator release and NK activity. Detailed studies by Ottaway (1987) have shown that these are changes in the expression of receptor density for VIP on activated immunocytes. It appears to bind selectively to helper/inducer ( L 3 T 4 + ) T-cells but does not bind to

277 suppressor/cytotoxic ( L y t 2 - ) T-cells. VIP has an inhibitory effect on splenic ConA responsiveness and impairs the production of IL-2. It has no effect on ConA activation of T suppressor cells. The special activity of VIP within mucosal tissues may impact on mucosal immunity. Vasopressin and oxytocin, insulin and growth hormone are also known to produce enhanced immunomodulation. Johnson and Torres (1985) showed that vasopressin or oxytocin could replace the need for IL-2 in the production of 7-interferon (~-IFN) by T helper cells. Insulin and growth hormone both cause enhanced lymphocyte activation, and growth hormone has been shown to counteract the inhibitory effects of corticosteroids on lymphocyte function (Snow, 1985). ACTH synthesis can be promoted in vitro by IL-1 (Lumpkin, 1987). A major role of enkephalins and endorphins may be as analgesic molecules which allow for increased tolerance to pain following exposure of animals to noxious stimuli. They may be secreted by brain or peripheral tissues. Molecules other than opiates may also be involved in stress-associated analgesia as some kinds of stress evoke analgesia which is not blocked by opiate antagonists (Lewis et al., 1980). The opioid, but not the nonopioid form of stress analgesia, is attenuated by hypophysectomy, adrenalectomy or adrenal demedullation or deneuration. Because the adrenal medulla is known to contain enkephalin-like peptides, which are secreted in response to stress, Lewis et al. (1982) have concluded that opioid stress analgesia depends on the release of adrenal reedullary enkephalins. Maier et al. (1983) have shown that unavoidable stress in laboratory animals, which evokes a state of 'learned helplessness', when the animals are exposed to inescapable shock, is associated with secretion of opioid analgesics. Opioid analgesia can be mimicked by administration of morphine (Terman et al., 1984). The similarity between POMC-derived ACTH and immunoreactive interferon, produced by activation of lymphocytes, is seen from studies which have required gene cloning to distinguish the molecular differences between ACTH and c~-IFN (Smith and Blalock, 1981). The functional similarity between c~-IFN and fl-endorphin is evident from experiments (Blalock and Smith, 1981) which have shown that ~-IFN has an analgesic effect, but even greater than that of fl-endorphin or morphine.

3.3.30pioids and immunocompetence Endorphins (Gilman et al., 1982; Johnson et al., 1982) and enkephalins (Johnson et al., 1982) have both been shown to alter humoral or cellular immune reactivity, when the respective molecules were added to splenic cells cultures. Antibody production as measured by in vitro plaque-forming cell (PFC) activation was dramatically suppressed by c~-endorphin but only marginally suppressed by y-endorphin, and was unaffected by fl-endorphin (Johnson et al., 1982). T-cell mitogenesis in rat cell cultures was unaffected by aendorphin and was enhanced by fl-endorphin (Gilman et al., 1982). Met- and Leu-enkephalin do not appear to have any effect on T-cell mitogenesis of hu-

278 man peripheral blood leukocytes. Endorphine and enkephalins caused enhanced NK cell function as demonstrated in vitro by target cell cytotoxicity (Mathews et al., 1983) or by facilitating tumour rejection in vivo (Plotnikoff et al., 1985). 4.0 NEUROPEPTIDE-LYMPHOIDCELL INTERACTIONS The central nervous system (CNS) and the lymphoreticular system (LRS) share the common trait that each has evolved as a specialised tissue which can generate appropriate but highly heterogeneous responses to extrinsic stimuli. Until recently, it was considered that the LRS functioned autonomously and was not influenced by the neuroendocrine system (McCain et al., 1982). Data are now accumulating which provide evidence that the LRS may be influenced directly by the afferent response of the CNS to cognitive environmental stimuli (Besedovsky and Del Rey, 1986). The cycle of reactivity between the CNS and LRS has now been completed, and there is increasing evidence that reciprocal efferent reactions also occur, where factors produced by activated cells within the LRS during immune reactivity to infectious agents or malignant cells have an important non-cognitive impact on CNS function (Besedovsky and Del Rey, 1986; Carr and Blalock, 1986). Evidence for afferent and efferent channels of communication between the CNS and the LRS is seen from three distinct lines of research: (1) compartmentalised innervation, with noradrenergic sympathetic nervous fibres occurs within the immunologically active areas of the primary and secondary lymphoid organs (Ackerman et al., 1986); (2) immune cells express receptors for many of the hormones, neuropeptides and neurotransmitters produced by the brain, in response to cognitive stimuli (Besedovsky and Del Rey, 1986); (3) the products of activated immune cells (cytokines) have neuroendocrine-like activity which can influence brain function (Carr and Blalock, 1986). 4. I Innervation of the L R S The presence of nerves has been demonstrated in thymus, bone marrow (Calvo, 1968), spleen and lymph nodes (Anderson and Anderson, 1975; Reilly et al., 1979). More recent work (Bullock and Moore, 1981; Williams and Felp ten, 1981; Livnat et al., 1985; Walcott and McLean, 1985) indicates that these nerves are distributed in specific locations within these organs. Noradrenergic (NE) nerve fibres are found dominantly in the subcapsular region and the corticomedullary junction of the thymus. In the spleen NE nerve fibres are found particularly within the periarteriolar lymphoid sheath (PALS) and the marginal zone of white pulp (Williams et al., 1981). In lymph nodes NE fibres occur mainly in the medullary cords and in the paracortical and cortical regions surrounding the follicles and in the subcapsular zone (Giron et al., 1980 ).

279 NE fibres are found mainly in the T-dependent areas of the gut-associated lymphoid tissues (GALT). More detailed histochemical studies (Ackerman et al., 1986), using fluorescent staining for catecholamines, have demonstrated that the major sites of NE innervation in spleen and lymph nodes occur within sites of entry, antigenic capture and exit of lymphocytes. These findings suggest that all the immunologically active areas within primary and secondary lymphoreticular tissues are potentially responsive to the neurological stimulation which results from cognitive stimuli which cause activation of adrenergic fibres in the CNS.

4.2 Receptors on immunocytes for hormones and neuropeptides Evidence for the link between the CNS, endocrine and immunological systems is seen from the extensive repertoire of receptors on immune cells for hormones and neuropeptides. Receptors for corticosteroids (Cake and Litwack, 1975; Werb et al., 1978), insulin (Hollenburg and Cuatrecasas, 1974; Helderman and Strom, 1978), growth hormone (Arrenbrecht, 1974 ), VIP (Ottaway and Greenberg, 1984; Ottaway, 1987) estradiol (Gillette and Gillette, 1979), testosterone (Abraham and Bug, 1976), arginine, vasopressin and oxytocin (Johnson and Torres, 1985), substance P (Payan et al., 1984), fl-adrenergic agents (Hadden et al., 1970; Bourne et al., 1974; Singh et al., 1979), acetylcholine (Strom et al., 1974; Richman and Arnason, 1979), endorphins (Hazum et al., 1979) and enkephalins (Wybran et al., 1979), have been demonstrated on the surface of immunologically active mononuclear cells. Receptor density and affinity for certain of these hormones are influenced by the state of activation of the immune cell, where increasing receptor density is found with increasing levels of cell activation. Altered density for glucocorticoids (Crabtree et al., 1980; Comsa et al., 1982), fl-endorphin (Gilman et al., 1982 ), enkephalins ( Plotnikoff et al., 1985 ) and insulin (Helderman and Strom, 1978), has been found on activated lymphocytes. Taken together the presence of hormone receptors and the modulation of immune reactivity by altered levels of hormones, analogous to that found in stress, provide compelling evidence that stress may influence immunocompetence through activation of the neuroendocrine and neuropeptide systems.

4.3 Immune factors which influence CNS function The bidirectional interaction between the CNS and the lymphoreticular system (LRS) is seen by the production of factors by activated immune cells which influence CNS activity. The possible interactions are outlined schematically in Fig. 3. Changes in CNS activity following immunisation have been shown to increase blood glucocorticoid and thyroxine levels (Shek and Sabiston, 1983;

280

I[~

Immune Activation ...."1

~

'~

h node

"

Thymus

I lnterleukins and Inter rerons

'",

I Corticosteroids

Catecholamines

Endorphin~

I

A C T H

III.

,[

I.

--

_

|

Fig. 3. I m m u n e activation a n d CNS function.

Besedovsky et al., 1985 ) and cause decreased sympathetic activity within spleen (Besedovsky et al., 1979) and in noradrenergic neurons in the hypothalamus (Besedovsky et al., 1983; Livnat et al., 1985). There is also an increased rate of firing of neurons in the medial hypothalamus following immunisation (Besedovsky et al., 1977). There is also a significant difference in the neuronal activity within lymphoid tissues of high and low responder animals following immunisation (Del Rey et al., 1982 ). Smith et al. ( 1985 ) have shown that virus infection of lymphocytes produces an immunoreactive hormone (ir-ACTH) analogous to ACTH. Besedovsky et al. ( 1985 ) have isolated a factor from lymphocytes which stimulates glucocorticoid production, and they have designated it glucocorticoid increasing factor (GIF). Smith et al. ( 1985 ), using bacterial lipopolysaccharide (LPS) to activate lymphocytes, have also observed that immunologically activated cells produce an endorphin (Jr-END). This factor has identical biochemical properties to pituitary fl-endorphin. Immunoreactive thyrotropin has also been produced following activation of lymphoid cells (Smith et al., 1983).

281

IL-1 produced following activation of monocytes causes release of ACTH from pituitary cells (Woloski et al., 1985; Lumpkin, 1987). IL-1 and IL-2 can replace CRF as functional activators for m RNA production of POMC molecules in pituitary tissues (Brown et al., 1987). Cytokines can also cause an increase in blood insulin (Besedovsky and Del Rey, 1986) and a decrease in the activity of brain noradrenergic neurons (Besedovsky et al., 1983). Cytokines also activate glial cells and they may play an important role in defense against brain infection (Fontana et al., 1980; Billingsley et al., 1982). Arginine vasopressin and oxytocin are capable of replacing the IL-2 requirement of T cells for the production of 7-interferon (Johnson and Torres, 1985). The relationship between the thymus and the neuroendocrine system has been suspected for 50 years, since patients with reproductive disorders were treated with thymic extracts (Andersen, 1932). The embryonic origin of thymus is in the pharyngeal pouch ectoderm and cephalic neural crest (Bockman and Kirby, 1985). The similarity between the origins of thymus, thyroid and parathyroid tissues suggests the possibility of some central role of endocrine factors in the function of thymic tissue. The thymic hormone, thymosin fraction V (TF5), has been shown to have a direct effect on pituitary function and the HPA axis, causing elevated levels of ACTH, fl-endorphin and cortisol, (Healy et al., 1983). The ability of TF5 to stimulate a glucocorticoid-induced hyperglycemia (Sivas et al., 1982) may significantly affect host defense and tissue repair following injury. Primordial Ig-like molecules, belonging to the immunoglobulin supergene family, have been found to be produced uniquely by glial cells of the CNS (Williams, 1987), providing further evidence for a functional link between the LRS and CNS. 5.0 STRESS AND IMMUNOCOMPETENCE

Whereas the earlier sections considered the generalised multisystem impact of the neurohormonal factors produced in response to stress, the remainder of this review will attempt to elucidate the hypothesis "That the immune system may be modified by stress to produce both inappropriate (suppressive) and appropriate (enhanced) changes in immunocompetence" (Golub and Gershwin, 1985). The nature of stresses likely to produce either effect will be considered in detail by examining the neurohormonal responses involved in different forms of stress and the concomitant changes in immune function in the respective hosts. A recent workshop involving a group of immunologists, physiologists and psychologists (Cohen, 1987) has looked critically at the methodological requirements necessary to unravel the complex interplay between environmental stimuli, neuroendocrine factors and host immunological reactivity. The findings were that, whereas there is compelling evidence that a competent immune response is fundamental in sustaining health and resistance to infection

282 and malignancy, there are few objective immune parameters that can be measured in vivo or in vitro, which provide an incontrovertible link between immune function and disease. Recognising that the experimental methodologies applied always involve some compromise, the greatest care should be taken nonetheless, to choose the most appropriate immune parameter for measurement, rather than the most accessible. It is important therefore that the researcher first establishes a meaningful link between the immune function being measured and the disease being studied in a model system. Care must also be exercised in extrapolating from data obtained between, or even within species, because of the extremely variable response of individuals as a result of their genotype or phenotypic expression, which may be influenced by their training, memory, learning or conditioning experiences. It has also been suggested (Cohen, 1987) that, where possible, outbred animals should be used in such studies, as the only biologically relevant responses are those expressed in a heterozygous animal, subject to varied phenotypic expression in response to complex environmental stimuli. The major body of research has tested the hypothesis that stress produces changes which result in defective immunocompetence with harmful consequences for the host. To accommodate such as approach one must first reconcile how such a potentially damaging change in immunity could have survived in evolution. Explanations advanced by Cohen (1987) involve three propositions. ( 1 ) The immunological changes found during stress are an unfortunate but expedient concomitant of the more fundamental 'flight-fight' response, which is essential for survival. (2) Acute stress does not exert a significant effect on immunocompetence, and it is only chronic stress which causes an impairment of immunity compatible with the development of disease. Chronic stress may have emerged in recent times as the product of h u m a n development and lifestyle, and following domestication of animals. The redundancy of the 'flightfight' or the 'conservation-withdrawal' responses during chronic stress, may have been associated with the deficit produced in the current 'expectation/ fulfillment' equation for humans, or the cost inherent in animal domestication, to satisfy the 'profit/loss' equation. (3) There is no objective evidence that immunity is altered at a functional level, and the components identified as different during stress are not biologically relevant, and may represent the most accessible measurement of immunity rather than the most relevant parameters. Data discussed throughout the remainder of this review provide evidence to support the above propositions, showing the highly complex nature of the interactions between stress and immunity. Is is now 110 years since Claude Bernard made his fundamental observation on homeostasis, and it may be worthwhile once again to take the wholist view that we are dealing with an array of reactions which require a complex but appropriate range of host responses. No

283 single parameter will adequately explain the mechanisms which could accurately implicate immunity in stress.

5. I Infectious disease susceptibility It is a striking coincidence that the year 1878 provided not only the foundation statement in physiology by Claude Bernard, which has become the basis for our understanding of homeostasis and stress physiology, but in that year Pasteur (cited by Nicol, 1974) also made the empirical observation that hypothermia increases the susceptibility of fowl to anthrax infection, during the experimental demonstration of his 'germ theory' for infectious disease. This work identified the link between the host, the microorganism and the environment in the aetiology of infectious disease. Since then many workers have identified links between the stress associated with physical and psychological stimuli, and the increased susceptibility of animals to infectious, organic and neoplastic disease (reviewed by Kelley, 1980 and Riley, 1981). Because of the complex interplay between factors, such as risk exposure to infectious agents and host response to infection, there are remarkably few documented reports using domesticated species which provide objective scientific information which implicate management stress as the causal factor in increased disease susceptibility (Henry and Stephens-Larson, 1985 ). To compound the complexity of the system, sophisticated management practices involving vaccination and therapy often obscure the natural susceptibility of domesticated animals to clinically obvious infectious diseases, under intensive farming conditions. Because pigs are one of the most intensively farmed domesticated species of animals, considerable work has been carried out on the effect of management on susceptibility to disease in pig herds. Perinatal stress has been shown to cause an increase of incidence of gastroenteritis and diarrhoea in piglets (Curtis, 1974). Exposure of piglets to cold in the first day of life is of special importance in influencing disease resistance and uptake of maternal colostrum by neonatal piglets (Blecha and Kelley, 1981). Shimizu et al. (1978) have demonstrated a direct link between the temperature at which pigs are housed and their clinical response to transmissable gastroenteritis (TGE). When 3-monthold pigs, housed under warm conditions (30 °C), were exposed to TGE, they showed no clinical signs of diarrhoea. Pigs housed at low temperatures for 2 to 4 weeks prior to TGE exposure showed a dramatic increase in the incidence of clinical disease. The temperature at which the pigs were housed after exposure to TGE was also important, as was adaptation of the animals to a given temperature prior to exposure. 'Shipping fever' is a well documented disease (Hambdy et al., 1963; Hoerlein, 1980) associated with transportation of cattle. The onset and severity of' response of individual animals to transport may be influenced by a number of variables, including the quality of transport, the ambient temperature, the stock

284

density, the novelty of transport and the time involved in transportation. Hoerlein and Marsh ( 1957 ) first suggested that the infectious complex causing 'shipping fever' was due to reduced resistance of animals, precipitated by transport stress, which caused establishment of secondary viral and bacterial infections. The age at which animals are exposed to stress appears to be extremely important in the manifestation of increased susceptibility of cattle. Young calves are much more susceptible to stress-associated episodes of infection, found during transport (Staples and Haugse, 1974), or following meteorological changes (Martin et al., 1975) or wetting (Jennings and Glover, 1952). Calves transported during winter have a higher mortality than those transported under less severe weather conditions (Staples and Haugse, 1974). Lactation in cattle involves an increased metabolic demand which may predispose cows to secondary stress. Increased susceptibility to infection is seen early in lactation in high-milk-producing cows (Newbould, 1974). Temperature extremes (McDowell and McDaniel, 1965) and nutritional stress (Morrow, 1976) cause further increase in the incidence of infection in lactating COWS.

Many years ago, Graham et al. (1919) produced their classic study of salmonellosis in horses and mules when they isolated salmonella from a high percentage of horses and mules suffering from diarrhoea following transport. Diarrhoea in horses induced by stress associated with transportation or surgery has become well recognised in recent years (Owen, 1975). Reactivation of salmonella excretion and diarrhoea has been identified in ponies subjected to surgery and transportation (Owen et al., 1983). Clinical manifestation of disease and shedding of organisms was not found until at least 3 days after stress. A neutropaenia developed within 24 h after stress and it lasted for 5 days. Pasteur's original observation of stress-induced susceptibility of fowl to anthrax has been confirmed in other studies on fowl exposed to management stress, analogous to that found associated with intensive management systems. Social stress involving isolation or high stocking densities is known to impact on and produce variable effects on the resistance of fowl to infectious agents (Gross, 1976). Gross and Colmano (1969) found that excessive social interactions reduced the susceptibility of chickens to S. aureus and E. coli, but increased their susceptibility to Newcastle disease virus (NDV) and Mycoplasma gaUisepticum. High-level social stress in chickens has also been shown to increase their resistance to mites (Ornithonyssus sylviarium) (Hall et al., 1979), but made them more susceptible to Marek's disease (Gross, 1972). Varying levels of social stress have been shown to produce opposite effects on the resistance of fowl to E. coli; low-grade social stress enhances disease resistance while high-grade social stress reduces disease resistance (Gross, 1984). Infectious disease resistance has been shown to increase with exposure to high temperatures in fowl (Thaxton, 1978). The highly variable effect of stress on

285 individual infections and the response to different types of infections highlight variations not only in the modes of protection involved in different diseases but also the variable physiological response of fowl to different types of stress. Laboratory animal models of stress (rodents) have been used extensively and provide incontrovertible evidence for the link between physical and psychosocial stress and altered susceptibility to infection. The first comprehensive series of experiments (Friedman and Glasgow, 1966) on psychosocial stimulation of mice demonstrated altered resistance of mice to Herpes Simplex, Coxsackie B, poliomyelitis and vesicular stomatitis viruses, Trichinella spiralis and tuberculosis. Social stress, imposed by high population density, decreased the resistance of mice to the specific rodent malarial parasite Plasmodium berghei (Plaut et al., 1969). Reduced resistance to encephalomyocarditis (EMC) virus was found in mice housed in isolation by Friedman et al. ( 1969 ). High population density also significantly increased mortality in mice exposed to S. typhirnurium. The virulence of the organism used for exposure of stressed animals appears to be of prime importance in the manifestation of altered disease resistance patterns. The effects of stress are more likely to be manifest when opportunistic organisms or agents causing chronic or insidious infection are used to challenge stressed animals. Exposure of mice to temperature stress (5 ° C) prior to inoculation with avirulent strains of S. aureus or S. typhimurium significantly increased their mortality when compared with animals housed under warm conditions (25 °C). Cold temperatures increased the mortality of mice exposed to avirulent strains of S. aureus from 21% to 91%, and from 61% to 100% for avirulent S. typhirnurium (Previte and Berry, 1962). By contrast, housing of mice in cold temperatures did not affect their mortality following exposure to virulent strains of S. aureus (70% vs. 70%) or S. typhimurium (94% vs. 100% ). 5.2 Stress and cancer

There is a vast body of literature which has considered the association between stress and cancer (reviewed by Riley, 1981; Sklar and Anisman, 1981; Justice, 1985 ). The findings from such studies are as varied as the approaches used to study the system. This is largely because of the lack of the fundamental understanding of protective mechanisms involved in different types of cancer and the extremely heterogeneous response of subjects exposed to stress. It is possible to clarify the position somewhat by excluding all studies which have not used objective and quantifiable parameters, and only compare data from work which has used a specific type of cancer model and defined stress conditions. This means that whereas there is circumstantial evidence for a link between cancer and stress in humans, such data are not included for comparison because the link is largely untestable under rigorous experimental conditions.

286

Early work (1950-1980) in this field has mainly involved the use of mice and rats in experimental stress regimes, which were often severe, and would be largely unacceptable in the light of the current ethical constraints being properly applied in laboratory animal experimentation. Four important criteria (Justice, 1985) must be included in experimental studies which attempt to associate stress with the initiation or outcome of cancer. These include, the nature and time span in which the stress is applied, the type of tumour under study, and the parameters used to study the development of the cancer. The most predictable regimes used to produce defined levels of stress are restraint (1 h per day), electric shock (1 h per day), forced activity (swimming or rotation for many hours each day), or cold (usually close to freezing). The type of cancer under study is of equal importance as there are significant differences in the kinetics of tumour development with tumours such as viral tumours, passaged spontaneous tumours, ascites or chemically induced tumours. Even viral tumours of different origins produce highly variable disease kinetics; murine sarcoma virus (MSV) (Fefer et al., 1968) causes tumours which develop to a maximum size in 2-4 weeks and then regress, whereas polyoma virus tumour is palpable within 1 month but is usually lethal within 3 months (Burchfield et al., 1978). By contrast mammary tumour virus (MTV) does not produce serious pathology or cause a potentially lethal effect for 1-2 years (Murphy, 1966). This highlights the desirability of applying rigorous criteria in the selection of tumour models appropriate for the study of the impact of stress, and suggests that important chronological considerations must be used in assessing the course of disease. Of equal importance are the criteria used to assess tumour development and growth. Because of the complex effects of stress on different physiological systems, it is probably inappropriate to use death as an objective measurement of tumour growth. The composite effects of stress and cancer may interact to produce the situation where a small tumour may be lethal for stressed animals, whereas a significantly larger tumour may not cause death in a non-stressed animal (Baker, 1977). More objective measurements of tumour development are likely to be latency of appearance, tumour size, number of tumours, rejection or metastasis (Justice, 1985). If one applies rigorous criteria in the selection of data from different laboratories, more cohesive and reproducibility models can be constructed. The summary findings using viral tumours suggest that when stress is experienced while tumours are present it usually facilitates tumour growth (Matthes, 1963; Amkraut and Solomon, 1972; Riley, 1981 ). Stress applied before implantation of a tumour usually inhibits its growth (Amkraut and Solomon 1972; Burchfield et al., 1978; Riley, 1981; Smith, 1982). This is interpretated as a biphasic response where immune facilitation of tumour growth due to immunosuppression occurs during exposure to stress, while an immune rebound occurs, due to

287 immunoenhancement and increased surveillance, following termination of stress. The levels at which these factors operate are influenced by the chronicity of exposure to stress and the degree of adaptation of the subject to the stress (Sklar and Anisman, 1981). Other factors which can influence the development of viral tumours include the strain of animal used, intercurrent infection and treatment with steroids. Riley (1981) has shown that even substrains of mice (C3H) have dramatically different responses to implantation with lymphosarcoma; strain C3H/He is relatively resistant while C3H/Bi is highly susceptible to tumour growth. Exposure of the resistant mouse strain (C3H/He) to infection by an unrelated non-oncogenic lactate dehydrogenase-elevating (LDH) virus causes acute but significant increase in plasma corticosteroids and causes a dramatic increase in growth of implanted lymphosarcoma. Injection of synthetic corticoids (dexamethasone, DMS ) also has a dramatic effect on the growth of implanted lymphosarcoma (6C3HED) in the resistant mouse strain. Mice given DMS 7 days before tumour implantation show a significantly increased rejection of tumour whereas injection of DMS 7 days after implantation of tumour causes increased growth of tumour. Riley (1981) has also studied the effect of low-stress housing conditions on the development of the slow-growing MTV mammary tumour in mice (C3H/HeJ) exposed to MTV during suckling, and found that animals exposed to low-stress housing conditions show a 10% incidence of tumours at 13 months of age, while animals housed in higher stress (conventional) conditions had a 68-92% incidence of tumours at the same age. Another extremely important factor which influences the impact of stress on tumours is whether the stress is escapable or inescapable. Independent studies (Sklar and Anisman, 1981; Visintainer et al., 1982) have shown that when mice are inoculated with tumours and exposed to inescapable shock there was enhancement of tumour growth. Exposure of mice to escapable shock had no impact on tumour development (Shavit et al., 1985). Chemically induced tumours or ascites have different kinetics of growth and spread, and the period of exposure to stress appears to be of fundamental importance in determining the impact of stress on tumour development. Severe acute stress, such as electrical stimulation (Kaliss and Fuller, 1968), or chronic low-grade cold stress for 150 days (Baker, 1977) is required before its impact on the growth of a slow-growing methylcholanthrene (MC) tumour is manifest. Faster-growing 7,12-dimethylbenz- ( a ) -anthracene (DMBA) tumours require stress to be applied for a minimum of 4 weeks. The growth of the naturally occurring Walker carcinoma, derived from mammary tissue in the rat, which is a fast-growing transplantable tumour, can be modified by stress applied for a short time course: 10 days forced exercise (Gershbein et al., 1974). Similarily the development of fast-growing ascites can be retarded by application of stress over a short time course (Matthes, 1963; Sklar and Anisman, 1981 ), prior to implantation of ascites cells. Whereas it is widely accepted that tumours of viral origin are consistently

288 responsive to immunosurveillance (Allison and Taylor, 1967 and Riley, 1981 } there is no such evidence that chemically induced tumours are controlled by immunological mechanisms (Good, 1972). It is possible therefore that stress has an impact on chemically induced tumours which does not involve immunological mechanisms. Justice ( 1985 ) has postulated that stress-related mechanisms which influence chemically induced tumour development may involve a 'stress-inhibition' response when stress is applied during the time of tumour growth. This may be more closely related to the catabolic-anabolic stress response proposed by Mason (1968). It is suggested that the compensatory metabolic mechanisms which produce a 'stress-rebound' response following recovery from stress may also have a non-immunological effect causing faster growth of tumours of non-viral origin. 6.0 EXPERIMENTALMODELSFOR STRESS IN IMMUNOLOGY There is a large body of circumstantial evidence that certain kinds of stress evoke physiological changes which influence susceptibility to infection and malignancy. Such changes may be due to alterations in the activation or function of the inflammatory or immunological pathways. Whereas it has been relatively easy to establish such an association, it is far more difficult to define the cellular mechanisms which may be involved or the predictable impact of different types of stress on immune function. Before considering this question more fully it is appropriate to consider an extreme state of altered immunity, AIDS, to highlight possible mechanisms which are vital in immune protection. It is now firmly established that the human disease AIDS is produced by a lymphotrophic virus (HIV) which specifically affects T cell subpopulations (T helper) to produce a state of generalised immunodeficiency. In this extreme state of immunological compromise the host becomes more susceptible to infectious disease and malignancy. AIDS sufferers are generally affected by chronic low-grade infections caused by parasites, protozoa, mycobacteria or latent virus infection. Surprisingly, AIDS patients do not appear to be peculiarly susceptible to acute bacterial or viral infections. These findings highlight the fact that even in the presence of severe immunodeficiency there are apparently only insidious changes in disease susceptibility, usually involving an increased incidence of chronic infections with opportunistic organisms. These findings suggest that care must be taken in selecting model systems which attempt to assess the impact of stress on immunity and protection. Studies quoted earlier (Previte and Berry, 1962 ), which have used virulent and avirulent strains of S. aureus and S. typhimurium in mice, confirm the likelihood that altered immunity associated with stress is evident only in selected infectious processes.

289 6.1 Stress and humoral immunity Early studies in mice (Rasmussen et al., 1959) demonstrated that avoidance learning stress decreased the susceptibility of passive anaphylaxis, indicating that stress could affect the expression of effector cell functions involved in anaphylaxis (Goetzl et al., 1985). Subsequently it has been demonstrated that aversive stimuli such as noise, light, movement or housing conditions could suppress antibody production following immunisation of stressed mice (Vessey, 1964). Solomon (1969) showed a decreased primary and secondary antibody response in acutely stressed rats. By contrast, exposure to chronic lowgrade stress, caused by low-voltage electrical shock, caused enhanced antibody production in rats (Hirata-Hibi, 1967; Solomon, 1969). A clearer picture of the effects of stress on antibody production is now emerging which suggests that while acute stress tends to cause a reduction in antibody production, repeated exposure results in an adaptation response which can cause enhancement of the immune response. Gisler and Schenkel-Hulliger (1971) showed that stress associated with restraint or crowding, applied in a single session, suppressed the antibody responses in mice, but the responses returned to pre-stress levels after 3 days of exposure. Monjan and Collector (1977) found that in vitro activation of B cells was suppressed in mice exposed to sound stress for up to 20 days, but animals exposed over extended periods had enhanced B cell reactivity. The complexity of the impact of stress on humoral immunity is highlighted by the study of Joasoo and McKenzie (1976) who found differential effects of different stressors on antibody production in animals of different sex. Acceleration or ether anaesthesia induced increased plasma corticosteroid levels in mice and there was an associated diminution of antibody production by their splenic B lymphocytes, exposed to antigen in vitro. The depressed response of the splenic B cells of stressed response could be reversed by treating the animals with somatotrophin which appeared to abrogate the effect of corticosteroids as inhibitors of in vitro antibody production by B cells. Recent studies have attempted to characterise more critically qualitative aspects of stress and its impact on humoral immunity. Croiset et al. (1987) found that mild emotional stress induced by exposure of rats to light/dark preference conditions, in a one-trial-learning passive avoidance test, caused enhanced primary antibody-forming cell activity. This enhanced response was ablated by exposure of the stressed rats to conflict through the avoidance response evoked after a single learning trial. By contrast, mild isolation stress causes reduced antibody production in mice (Edwards et al., 1980). Esterling and Rabin (1987) have attempted to assess the impact of an acute but more severe form of rotational stress on immunity, as measured by the ability of animals to produce antibody and splenic plaque-forming cells (PFC), following immunisation of stressed mice with sheep erythrocytes (SRBC). Adrenalectomy was carried out to measure the impact of stress on the adrenocortical

290 hormone response. The results showed that stress produced a significant decrease in splenic Lyl (T helper lymphocytes) cells in intact animals, and that adrenalectomy and stress caused a significant increase in Ly2 (T suppressor/ cytotoxic ) cells. Whereas adrenalectomy influenced the T cell subpopulations in stressed animals it had no impact on B cell activity as measured by splenic PFCs in stressed animals. These data imply that whereas adrenal function and increased corticosteroid levels may influence T cell subpopulations in the spleen of stressed animals, the influence of stress on B cell function is independent of the adrenal corticosteroid system (Blalock et al., 1984). Boranic et al. (1987) have shown that chronic restraint produces a reduced PFC response in immunised rats and that the impairment of immunity can be correlated with, or produced in vitro by serotonin. Chronic restraint has also been shown (Pericic et al., 1987) to cause a significant decrease in PFC production in mice immunised with SRBCs. The response was associated with decrease levels of brain noradrenalin and increased plasma corticosteroids in stressed animals. A similar level of immunosuppression was seen in control mice treated with diazepam. A comprehensive series of experiments carried out in Japan recently (Okimura and Nigo, 1986; Okimura et al., 1986a) has attempted to evaluate the effect of the time of stress application on the cells involved in antibody production in mice exposed to restraint stress; 12 h per day for two consecutive days. Okimura and Nigo (1986) have demonstrated a significant reduction in PFC levels and antibody production to SRBCs in mice restrained before antigen injection. By contrast, there was no change in antibody production when stress was applied after immunisation. Application of the inhibitory stress conditions to mice immunised with T-independent antigens (TNP-ficoll or TNP-lipopolysaccharide) did not cause any reduction in PFC activity. These data suggest that stress impacts on T cell regulatory activity and does not affect B cells. Okimura et al. (1986a) have also shown that restraint stress caused a significant reduction in the activity of T-helper and T-suppressor cells, while it had no impact on the function of antigen-presenting monocytic cells. Whereas T-cell-dependent antibody production to SRBC in stressed mice was 50% lower than in the control mice, the antibody response to T-cell-independent antigen (TNP-LPS) was 40% higher in stressed animals than in the controls. Proliferative responses of splenic lymphocytes to T cell and B cell polyclonal activators, ConA and lipopolysaccharide (LPS), were similarly affected by stress. Splenic cell responses to ConA were suppressed in stressed animals, but unaffected by stress when stimulated with LPS. Blecha and Kelley (1981) showed that exposure of pigs to cold increased serum antibody levels by 60% and caused a 360% increase in their capacity to synthesise antibody. Exposure of calves to cold temperature stress caused an increase in antibody production (Kelley et al., 1982a). Stress associated with early weaning ( < 5 weeks) significantly depressed B cell function in young pigs (Blecha and Kelly, 1981 ). Tethering of sows caused a significant reduction in antibody

291 production and colostral transfer of antibody to the offspring, when the sow was immunised with SRBC during restraint. Machado-Neto et al. (1987) have found that by comparison with control piglets, there were reduced IgG levels in piglets from sows exposed to heat stress (32 °C) from day 100 of pregnancy to 8 hours before farrowing. These piglets had significantly lower levels of IgG for the first 20 days and had higher levels of serum cortisol at birth. Lower IgG levels in piglets from heat-stressed sows were associated with higher levels of preweaning mortality. A genetic basis for stress susceptibility in pigs was found by Edfors-Lilja et al. (1987) when they showed that the genes associated with hyperthermia (Hal/n) also provided a marker for susceptibility of these animals to transportation stress and high plasma corticosteroid levels. Animals which were homozygous ( n / n ) showed the greatest response to stress and had a significantly reduced antibody response when immunised with E. coli (0149) antigen, after exposure to transport stress. The impact of temperature extremes on antibody production in fowl has been evaluated. Acute intermittent heat stress (43°C) has been reported (Thaxton, 1978) to suppress the antibody production in fowl immunised with SRBC. Heller et al. (1979) observed that exposure of fowl to 42°C for short periods (2 h) caused enhanced antibody production, whereas Regnier et al. (1980) found no alteration in antibody production in heat-or cold-stressed fowl. Regnier and Kelley ( 1981 ) have shown that heat stress (36 ° C for 5 days) in fowl causes elevated antibody production in certain breeds (New Hampshire), though they could not find any change in antibody levels in immunised chickens kept in the cold (1°C for 5 days), as compared with birds kept at 26°C.

6.2 Stress and cell-mediated immunity Cell-mediated immunity (CMI) is produced by the activation of mononuclear cells following exposure to antigen. The main cell types involved in this reaction are T cells and mononuclear phagocytes. Different subpopulations of T cells are involved in the regulatory (T helper, Th, and T suppressor, Ts) and the effector phases of CMI (T delayed hypersensitivity, Td, or T cytotoxic, Tc ). CMI reactivity is difficult to measure accurately and is usually monitored in vivo by the delayed type hypersensitivity (DTH) skin test or contact sensitisation (CS) reaction. The reaction has been studied much more extensively in vitro where function can be assessed by lymphocyte blastogenesis (Soder and Hellstrom, 1987) or lymphokine production using polyclonal T cell activators [phytohaemagglutinin (PHA), concanavalin A (ConA) or pokeweed mitogen (PWM) ], or antigens in animals displaying specific CMI reactivity. Special emphasis has been placed on the role of CMI in stress because T cells are uniquely sensitive to soluble modulating factors (Kelley, 1980), so it is likely that the neuroendocrine response in stress elaborates hormones and

292

peptides which may have a major impact on CMI function. It has been established that activation of the HPA system and the release of corticosteroids has a major effect on the T cell system (Claman, 1972; Cupps and Fauci, 1982) of humans, domestic animals and laboratory rodents, which is manifest most dramatically by thymic involution (Riley, 1981) and lymphopaenia within hours of the exposure to severe stress. However, other neuroendocrine factors released as a consequence of stress may also influence T cell function because of the presence of receptors on T cells for most of the hormones and peptides (Besedovsky and Del Rey, 1986; Carr and Blalock, 1986) produced in the brain and periphery during stress (Kelley, 1985). 6.2.1 Stress and CMI in vivo The first report (Pitkin, 1965) which considered the impact of heat stress on CMI in mice showed a reduction in the in vivo D T H of heat-stressed anireals. The impact of restraints stress on D T H in mice has been studied (Blecha et al., 1982 ) by measuring D T H reactions 4 days following immunisation with low doses of SRBC or following contact sensitisation (CS) with dinitrofluorobenzene (DNFB). Stress caused a 40% reduction in D T H to SRBC when stress was imposed at the time of vaccination, whereas stress applied at rechallenge, prior to measurement of DTH, caused a 30% reduction in D T H by comparison with unrestrained control mice. The CS reaction to DNFB showed completely different patterns of responsiveness when the stress was applied at the time of sensitisation or rechallenge. Stress at sensitisation caused a 78% greater CS response and stress at rechallenge caused a 140% increase in CS reactivity in stressed animals by comparison with unrestrained controls. Adrenalectomy of animals prior to the application of stress and immunisation or sensitisation, reversed the inhibitory effect of stress on D T H to SRBC, but had no effect on the enhanced response seen with CS to DNFB. The results suggest that the T cells involved in D T H are different from those involved in CS, and they are sensitive to the immunosuppressive effects of corticosteroids released by acute stress. It is possible that the immunoenhanced CS response to stress results from the potentiation of the reaction of immunoregulatory cells by neuropeptides such a fl-endorphin, which are known to react with (Wybran et al., 1979), and enhance T cell activity (Gillman et al., 1982). Cold (5°C), heat (35°C), or restraint stress applied to mice tested in the above D T H and CS model systems, showed even more complex effects (Blecha et al., 1982). The CS response to DNFB was consistently enhanced by restraint, heat or cold stress, irrespective of whether it was applied at the inductive or effector phase of the challenge. The induction of the D T H to SRBC was enhanced by heat stress and either enhanced or suppressed by cold stress depending on the timing of stress relative to the induction or expression of the CMI reaction. These results show that a single type of stress can either enhance or suppress CMI activity in mice, and that the effect produced may depend on the type of regulatory cell that controls the T-cell response under

293 test. Okimura et al. (1986b) have confirmed that restraint applied before or after the induction of DTH caused a 50% reduction in the DTH reaction, irrespective of when the stress was applied. Studies have also been carried out to measure the impact of heat and cold stress on the DTH or CS response in calves (Kelley et al., 1982b), pigs (Westly and Kelley, 1984) and chickens (Regnier and Kelley, 1981). Kelley et al. (1982b) exposed 3-week-old calves to a 2-week period of heat (35 ° C) or cold ( - 5 ° C ) prior to testing for DTH reactivity to M. tuberculosis (PPD) or CS following sensitisation with DNFB, or intradermal injection with phytohaemagglutinin (PHA). Heat exposure reduced the DTH reaction by 42 % and the CS reaction by 38%. The PHA reaction was unaffected by a 1-week exposure to heat stress but was reduced to 80% of the control values after 2 weeks of exposure to heat stress. The response to cold stress was more varied in that cold stress reduced CS reactions by 39% after 1 or 2 weeks. The specific DTH reaction to PPD was increased by 42% after 1 week but reduced by 14% after 2 weeks exposure to cold as compared with control calves, providing evidence in favour of immune adaptation. The in vivo PHA reaction was reduced by 39% but only after 2 weeks exposure to cold. Field studies have also provided evidence that exposure of adult cattle to subtropical conditions reduces the sensitivity of TB-infected cattle to intradermal DTH reactivity when tested with PPD. Recent work in our laboratory suggests that the sensitivity of tuberculin skin testing in recently captured deer is lower than that found in animals adapted to farming conditions (Griffin, 1989). A significantly higher incidence of TB infection in breeding hinds implies that the endocrinological factors produced during pregnancy may also decrease CMI resistance to TB. Weaning also causes in vivo suppression of PHA reactivity in young piglets (Blecha et al., 1983). A similar finding has been noted for humans where pregnant women appear more susceptible to TB (Rich, 1951 ), and exacerbation of the symptoms of leprosy (Duncan et al., 1982), both of which could best be explained by a reduced level of CMI protection during pregnancy. Regnier and Kelley ( 1981 ) have also used the CS reaction to DNFB and the in vivo PHA reaction as a monitor for DTH, by measuring wattle swelling in chickens exposed to cold ( 1 ° C ) or heat (36 ° C ) stresses, compared with a thermoneutral (26 °C) housing temperature. DNFB sensitisation and PHA-induced wattle swelling were significantly reduced in fowl exposed to 36°C or 1 °C for 5 days. 6.2.2 Stress and CMI in vitro A relatively large number of studies has been carried out using mononuclear cells from laboratory animals or domesticated animals exposed to stress (Roberts, 1979). As in most areas of experimental immunology, the majority of the work has been carried out in laboratory rodents whose responses are well characterised. Folch and Waksman (1974) serendipiditously observed that splenic cells from mice exposed to loud noise and water deprivation, during the course

294

of new building construction, had reduced responsiveness to PHA which persisted for 1-3 weeks after termination of the stimuli. Monjan and Collector ( 1977 ) found that mice exposed to short-term noise stress gave a reduced splenic cell response to ConA. Exposure to similar noise levels over extended periods evoked an enhanced response of the splenic lymphocytes in culture with ConA. Okimura et al. (1986a) showed that mice exposed to restraint stress had reduced ConA reactivity in splenic cell cultures and that the induction of Ts cells by ConA was impaired in stressed mice. Teshima et al. (1987) demonstrated a significant change in T cell subpopulations in the thymus of mice exposed to immobilisation stress. Lyt-2 (Ts) cells in the thymus were significantly depleted 3 days after cessation of stress. There was an initial decrease in thymic Lyt-1 (Th) cells for 1 day following cessation of stress, after which the cells returned to normal levels. There was no significant impact on Lyt-l:Lyt-2 cell ratios in the spleen or bloodstream of the stressed mice. Thymic atrophy associated with stress reduced the weight of the thymus to 58% of the control values and this atrophy increased to produce a thymic weight 2 days after cessation of stress which was 35% of the weight of control thymus. The nature of stress and its relative impact on lymphocyte function using in vitro culture assays has been studied comprehensively in the rat, highlighting some of the subtle changes produced in different stress systems. Rats exposed to graded electric tail shocks (Keller et al., 1981 ) showed a decreasing level of peripheral blood lymphocyte responsiveness to PHA. The degree of suppression was directly related to the severity of the shock applied to the rats. Apparatus control rats which did not receive any shock also had a small but significant reduction in their PHA responses by comparison with home cage controls. Shock also caused a significant reduction in the lymphocyte numbers present in the peripheral blood. Further studies by this group (Keller et al., 1983) showed that whereas high-grade electric shock had a significant effect on plasma corticosteroid levels in the stressed animals, there was still a significant reduction in the PHA responses of peripheral blood lymphocytes from shocked animals which had been previously adrenalectomised, and produced no plasma corticosteroids. The major difference in the shocked adrenalectomised animals was that they showed a less dramatic drop in blood lymphocytes by comparison with the intact shocked animals. Treatment of unshocked adrenalectomised animals with corticosterone produced a lymphocytopaenia equivalent to that found in intact stressed animals. These studies suggest that whereas electric shock causes activation of the HPA system and a corticosteroid-related reduction in the absolute numbers of blood lymphocytes, the reduced PHA responsiveness of lymphocytes from stressed rats was in some measure independent of adrenal function. The influence of conditioning and adaptation on immunosuppression in rats has been illustrated in a study (Laudenslager et al., 1983 ) which showed that PHA responsiveness of peripheral blood lymphocytes from rats exposed to a

295 series of electric shocks was suppressed only if the shock was inescapable. Exposure of rats to equivalent levels of escapable shock had no influence on the PHA responses. These data suggest that is was the psychological response rather than the physical severity of the stress which produced the altered lymphoproliferative response. Whereas the type of shock applied in Keller's experiments evokes an immediate analgesia (Maier et al., 1983), irrespective of whether it is escapable or inescapable, different neuroendocrine factors are responsible for the respective analgesic effects. Shavit et al. (1984) found that escapable foot shock was mediated by a non-opiate analgesic which had no effect on immune function, whereas inescapable shock produced an opioid form of analgesia which was immunosuppressive. It is likely therefore that opioid neuropeptides produced following electric shock are implicated in the immunosuppressive effects associated with this form of stress. Croiset et al. (1987) have provided another example of conditioning in studies which have shown that the mild emotional stress evoked in the one-triallearning passive avoidance test to light and darkness caused an enhanced ConA peripheral blood cell response in rats. Animals showing enhanced ConA responses had significantly higher corticosteroid levels than those found in the home cage controls, demonstrating that enhanced ConA reactivity was produced when elevated corticosteroids were present, providing further evidence that elevation of plasma corticosteroids was not necessarily the sole explanation for stress-related immunosuppression. Studies with monkeys (Reite et al., 1981) have demonstrated that social separation of paired pigtailed monkeys, which had been raised together from infancy, caused a significant reduction in T-cell mitogenesis. This effect was evident for 2 weeks following separation, after which the T-cell response returned to normal. Moberg (1985) has elegantly demonstrated that rearing patterns in monkeys have a marked impact on the response of animals exposed to stresses, such as novel housing or restraint. Animals raised with inanimate surrogate parents show dramatic differences in heart rate, behavioural activity, vocalisation, plasma corticosteroids and growth hormone, following exposure to the novel environment. Although the effect of rearing patterns on immunocompetence has not been measured, it is likely that this would also be affected differently by exposure of such animals to stress. In domesticated animals transport stress has been identified as a causal factor in the reduced lymphoproliferative T-cell responses of cattle, when their peripheral blood leukocytes are cultured in vitro (Kelley et al., 1981; Kent and Ewbank, 1983; Blecha et al., 1984). By contrast another group (Murata et al., 1987) found enhanced in vitro lymphoproliferation in the peripheral blood of calves after a 1-h transportation experiment. This group demonstrated a small but significant reduction in spontaneous lymphoproliferation in leukocytes from calves transported for 4 h prior to testing, but failed to identify any transport-associated reduction in the response of calf leukocytes to PHA or PWM.

296

Capture of Bighorn sheep (Hudson, 1973) has been shown to cause a significant decrease in their in vitro lymphocyte blastogenesis to PHA. The response recovers to normal levels within 3 weeks post-capture. Studies in our laboratory (Griffin, unpublished) have also identified a dramatic reduction in leukocyte blastogenesis of peripheral blood cells from deer in the 4 weeks following capture using PHA, ConA, or PWM. The mitogenic reactivity of lymphocytes was 90% suppressed in the first week post-capture, but it had recovered, so that only 30% suppression was evident 1 month post-capture. 6.3 A new large animal model for stress studies In the past 20 years New Zealand has been involved in one of the largest experiments in animal behaviour seen in recent centuries, involving the domestication of wild deer for farming under intensive husbandry conditions. This represents a uniquely accessible system to study the stress associated with domestication of a wild animal. Not only has it proved itself to be a viable alternative form of agricultural enterprise, but it has highlighted many unique problems imposed by husbandry, which invariably challenges the 'free and independent existence' of a wild animal. The success of this enterprise, which has resulted in the emergence within N.Z. of a national farmed herd of 500 000 deer, has been dependent primarily on the development of management and handling systems which allow farmed deer to be managed under a system of minimal stress. The primary response of these animals to the stress associated with human intervention and management is manifest majestically by their physical expression of 'flight' reflex as a component of their 'flight-fight' response to stress. The stresses imposed on this recently domesticated species are now well recognised, and vary in severity from the severe form of acute stress in the wild capture of feral animals, to the chronic and insidious farming stress during physical restraint, transport, handling, adverse climate, inadequate nutrition, weaning, social hierarchy and breeding (Griffin, 1987a,b). Many infectious diseases, found in increased incidence in farmed deer, are exacerbated by management stress. They range in severity from acutely fulminating bacterial disease involved in foot abscesses, which produce a rapidly fatal toxaemia in animals post-capture (Griffin, 1987a), to acutely lethal infection caused by malignant catarrhal fever (MCF) virus in animals following transport or after exposure to adverse climatic changes (McAllum, 1980). Episodic outbreaks of yersiniosis in weaned deer (Mackintosh and Henderson, 1984) and exacerbation of tuberculosis in breeding females (Griffin, 1989) have also been found in farmed deer. 6.4 Conditioning of immune reactivity Ader and Cohen (1975) made the startling discovery that the influence of an immunosuppressant drug could in part be conditioned by association be-

297 tween the specific drug and an independent aversive chemical. When cyclophosphamide or methotrexate was given to animals simultaneously with saccharin in the drinking water, rats subsequently developed a state of conditioned immunosuppression when given saccharin alone. This was reflected in reduced antibody responses in conditioned rats given saccharin alone. The data have been viewed with great skepticism and it has been suggested that the conditioned immunosuppressive effect may be no different from HPA stimulation of animals, to produce corticosteroids, when they are exposed to aversive chemicals. The response has, however, been shown to be independent of corticosteroid production through adrenal activation. Jenkins et al. (1983) have also shown conditioned immunosuppression of antibody production in animals treated with lithium chloride. However, a better argument in favour of conditioning of immune reactivity is conditioned immunoenhancement which is evident when animals are given immunopotentiating agents combined with aversive chemicals. This has now been confirmed by Gorczynski et al. (1982) who demonstrated a conditioned immunoenhancement of CMI during skin graft rejection. Evidence for enhancement of NK function has also been found recently (Ghanta et al., 1985) where increased resistance of rats to implanted osteosarcoma was produced in a conditioned enhancement of NK activity, in animals treated with poly-IC in association with camphor odours. Exposure of rats to camphor alone was shown to have no immunomodulating effect but when conditioned rats were exposed there was enhanced resistance to tumour cells. The implications of such findings are of major significance if such strategies could be applied to animals to evoke a state of immunosuppression to control autoimmunity or graft rejection or conversely to facilitate reactivity to vaccines. The ability of the subconscious responses to influence host immune reactivity was identified many years ago by Black et al. (1963) who demonstrated that delayed type hypersensitivity in humans could be ablated in tuberculinreactive subjects under the influence of hypnotic suggestion. Smith and McDaniel (1983) have confirmed this finding, showing that behavioural conditioning reduced the oedema, erythema and induration of tuberculin-positive patients. Histological studies showed that there was an accumulation of mononuclear cells equivalent to control levels, within the skin test sites of the conditioned subjects, but that the inflammatory changes associated with a conventional DTH reaction did not occur. These data imply that CNS responses, far less dramatic than those likely to occur in stress, have a significant impact on the expression of immune reactivity. The underlying contributions by neurohormones and neuropeptides have not been characterised in the above experiments. It is conceivable that such reactivity may also have a role in acupuncture, which has been used so widely in the Orient to treat an array of psychological and organic diseases. Similarly, the placebo effect seen in the treatment of human conditions may involve cognitive responses in the CNS,

298 which produces an appropriate physiological response and improves the health of the individual. The prospect of using conditioning or suggestion to facilitate therapeutic treatment has exciting prospects for treatment of disease and in vaccination protocols. 7.0 STRESS: DIAGNOSISAND MANAGEMENT In recent years significant public debate has highlighted areas of animal welfare which have demanded the re-evaluation of many of the practices used in intensified farming and biological research. Curators of zoological gardens have recognised the benefits of expanding the scope of housing and in providing free-range living conditions for their animals. Some go so far as to use social enrichment environments to create the greatest sense of wellbeing in their captive guests. By contrast, the highly intensified husbandry practices used for the management of fowl, pigs, cattle and laboratory animals often involve housing and management systems which at best pay little attention to social enrichment. Using anthropomorphic systems to derive models for management, it is likely that for animals an environment which makes them most contented may also be the most productive. Central to such a debate is the consideration of how different animal treatments may be perceived as stressful and how they could be monitored or eliminated. In civilised human society it should be realistic to expect that the animals which we use for food production or as companions should live their lives free from abuse or adverse exploitation. At worst, pragmatism should demand the recognition that the profit we derive from animals is directly related to their state of wellbeing. Because the immune system is so sensitive to extrinsic stimuli it is possible that the monitoring of immune function may provide a sensitive bioassay of animal wellbeing. By first critically defining reference parameters of inflammatory cell and immune function for animals belonging to different species it should be feasible to use immunological markers as indicators of stress or disease susceptibility in individual animals. Because of the highly heterogeneous neuroendocrine response to environmental stimuli, single point hormone assay are unlikely to be of value in determining animal wellbeing, and functional immunoassays could provide more definitive markers for pain or distress. The recognition that handling, restraint and long-term management conditions may cause acute or chronic changes in animal physiology demands that animals used in experimentation should be housed and managed under conditions which do not cause undue stress. The serendipiditous discovery that disruption of laboratory animals' housing conditions (Folch and Waksman, 1974) altered cellular immune function in mice, highlights the care required to maintain animals in a natural state appropriate for analysis of the effects of experimental intervention. Riley ( 1981 ) has also emphasized that acute and chronic changes in haematological or immunological parameters can be pro-

299 duced within minutes of manual restraint or through systematic bleeding of a group of animals from a single cage. The impact of chronic stress was highlighted by their use of animals raised in stress-free environments for 1 to 2 years, which produced different patterns of resistance to tumours than similar animals raised in the more stressful conditions found in conventional housing. Significant advances have been made in legislation recently to focus the attention of experimental workers on the minimal conditions necessary for the breeding, housing and care of laboratory animals. The recent Animals (Scientific Procedures) Act passed through the British Parliament (Hollands, 1986) should advance the cause of quality animal experimentation by placing stricter responsibilities on the individual research worker for the selection, design and execution of experimental procedures. The 'cost-benefit' involved in individual experimentation must now be justified by the experimental worker to ensure that worthwhile scientific endeavour occurs without undue suffering to the animal. It is interesting that laboratory animal procedures once considered as routine, such as blood collection and antibody production, now fall within the definition of experimental intervention. The emergence of a sophisticated standard of guidelines for animal experimentation should ensure not only that laboratory animal wellbeing is given paramount consideration but also that many artefactual responses, due to inappropriate host reactions, contingent to poor animal management, are obviated. Accepting that the basic mechanisms involved in biology can only be elucidated by studying natural responses, it is timely that due attention is paid to ensure that animals under experiment are responding naturally. In addition to housing, handling and experimental intervention, it is also important that due care be given to the appropriate selection and use of anaesthetics and analgesics, where pain or distress is likely to occur, during or consequent to an experimental protocol (Morton, 1987). Physical restraint should not be used as a substitute for the appropriate implementation of analgesia, sedation or anaesthesia. Whereas it recognised that the absence of stress occurs only in death (Selye, 1946), it is obvious that a great deal of work will be required to properly characterise the environmental stimuli which potentiate and enhance the host response during stress, as compared with those stimuli which produce such severe changes in host physiology as to be manifest in a prepathological state. The ethical constraints now placed by legislative bodies mean that more justification will be needed if cost-effective experiments on stress are to be carried out using laboratory animals. A closer examination of naturally occurring stress states associated with routine management of farm animals may provide more relevant and accessible model systems for the study of stress in the future. Such an approach demands that a complete database of normal endocrine parameters and immune function tests be available for the individual species under test. Recent studies in our laboratory using deer have shown us that valuable findings involving stress may result from the use of

300

highly controlled experimental studies without draconian intervention by the scientist. By placing greater emphasis on the parameters of immune function (e.g., active immunisation with novel primary immunogens } to be studied, and by comparison between experimental findings and a complete reference database, it may be feasible to clarify the mechanisms whereby stress may affect immune function in this species, using a spectrum of different natural stressors ranging through weaning, hierarchial interactions, transport, handling and breeding. 8.0 C O N C L U S I O N S

The physiological changes which occur under the influence ~f CNS activation during stress produce widespread changes within the host. The initial 'flight-fight' response is due to an increase in metabolic and cardiovascular activity, under the control of the sympathetic adrenomedullary system. This is followed by the 'conservation-withdrawal' response, under the influence of adrenocortical stimulation. The cycle is completed by the 'learning-adaptation' response conditioned by the production of a wide array of neuroendocrine factors. Does this increased capacity to run, hide, tolerate pain and adapt, place undue demands on other biological systems within the host? The evidence forthcoming from the extensive studies which have examined the direct influence of stress on immune physiology would suggest that immunity is impaired transiently during the acute phase of the stress response. However, the generation of a protective immune response de novo which requires i to 2 weeks to become fully activated, may be of little relevance during the acute phase of stress, which demands an immediate effector response. The integrity of the inflammatory system, and its activation by pre-existing immune cofactors, is likely to be of far greater significance in the protection of the host during acute challenge. The increase in circulating blood neutrophils may contribute to enhanced nonspecific protection during acute stress. There is also a significant body of data which suggests that resolution of acute stress has contingent side effects which can potentiate subsequent immune activation. The bidirectional interactions between the CNS and LRS, due to communications mediated by neuroendocrine factors and cytokines, explain how cognitive responses in the brain can impact on immune function. However, when stress is of extreme severity or persists chronically, it may place demands on the host which have more telling effects on immune function. Should chronic stress compromise immune function, it is of considerable importance, as the persistence of stress may threaten the integrity of immunosurveillance. Data available from the study of chronic stress suggest that the inability of the host to cope with or counteract chronic aversive stimuli produces severe immune compromise through physiological imbalances which re-

301 sult from chronic activation of the neuroendocrine pathways. By contrast, the ability of the host to cope with chronic stress evokes the production of a different array of neuropeptides and neuroendocrine factors, which are produced during learning and adaptation. Coping or adaptation may be expressed behaviourally by avoidance or escape from the aversive stimulus or the development of an assertive or aggressive response to challenge. The physiological factors produced during adaptation appear to cause potentiation of immunocompetence and may be of significant benefit for the host. Future research into the underlying mechanisms which link stress physiology and immune function demand t h a t we focus on new questions and identify different pathways for study. Key topics in this new phase of research include the following: (1) the need to define the kinetics for the activation and production of the array of different but interrelated neuroendocrine factors produced during defined states of acute or chronic stress. It also appears vital t h a t CNS and peripheral levels of these hormones should be measured simultaneously. (2) the necessity of studying the impact of stress on immune reactivity using de novo experiments of activation of the immune system with specific antigenic probes in vivo. (3) the urgency to establish a reference database for the accurate assessment of both neuroendocrine and immunological factors for individual species of animal under different states of activation. Above all, the field now demands t h a t we a t t e m p t to design experiments which apply more predictable stressors, known to evoke consistent neurophysiological responses, under defined conditions of immune challenge, all of which can be measured objectively.

9.0 REFERENCES Abraham, A. and Bug, G., 1976. H-testosterone distribution and binding in rat thymus cells in vivo. Mol. Cell. Biochem., 13: 157-163. Ackerman, K.D., Felten, S.Y., Bellinger, D.L., Livnat, S. and Felten, D.L., 1986. Noradrenergic sympathetic innervation of spleen and lymph nodes in relation to cellular compartments. In: B. Cinader and R.G. Miller (Editors), Progress in ImmunologyVI. AcademicPress, Toronto, Ont., pp. 588-600. Ader, R. and Cohen, N., 1975. Behaviourallyconditionedimmunosuppression.Psychosom.Med., 37: 333-340. Allison, A.C. and Taylor, R.B., 1967. Observations on thymectomy and carcinogenesis. Cancer Res., 27: 703-707. Amkraut, A. and Solomon,G.F., 1972. Stress and murine sarcoma virus (Maloney) induced tumours. Cancer Res., 32: 1428-1433. Andersen, D.H., 1932. The relationship between the thymus and reproduction. Physiol. Rev., 12: 1-22. Anderson, A.O. and Anderson, N.D., 1975. Studies on the structure and permeabilityof the microvasculature in normal rat and lymph nodes. Am. J. Pathol., 80: 387-412. Arave, C.W.,Walters, J.L. and Lamb, R.C., 1978. Effects of exercise on glucocorticoidsand other cellular components of blood. J. Dairy Sci., 61: 1567-1572.

302

Arrenbrecht, S., 1974. Specific binding of growth hormone to thymocytes. Nature, (London), 252: 255-257. Axelrod, J. and Reisine, T.D., 1984. Stress hormones: their interaction and regulation. Science, 224: 452-459. Baker, D.G., 1977. Influence of chronic environmental stress on the incidence of methylcholanthrene-induced tumors. Cancer Res., 37: 3939-3944. Belluardo, N., Mudo, G., Cella, S., Santoni, A., Forni, G. and Bindoni, M., 1987. Hypothalamic control of certain aspects of natural immunity in the mouse. Immunology, 62: 321-327. Bernard, C., 1878. Les Phenomenes de la Vie. Librairie J.B. Bailliere et Fils, Paris, Vol. 1, pp. 879 893. Besedovsky, H.O, and Del Rey, A., 1986. Immune-neuroendocrine network. In: B. Cinader and R.G. Miller {Editors), Progress in Immunology VI. Academic Press, Toronto, Ont., pp. 578587. Besedovsky, H., Sorkin, E., Felix, D. and Haas, H., 1977. Hypothalamic changes during the immune response. Eur. J. Immunol., 7: 323-325. Besedovsky, H., Del Rey, A., Sorkin, E., Da Prada, M. and Keller, H.H., 1979. Immuno-regulation mediated by the sympathetic nervous system. Cell. Immunol., 18: 346-355. Besedovsky, H., Del Rey, A., Sorkin, E., Da Prada, M., Burri, R. and Honegger, C., 1983. The immune response evokes changes in brain noradrenergic neurons. Science, 221: 564-568. Besedovsky, H.O., Del Rey, A.E. and Sorkin, E., 1985. Immune-neuroendocrine interactions. J. Immunol., 135: 750s-754s. Beutler, B., Krochin, N., Misark, I.W., Luedke, C. and Cerami, A., 1986. Control of cachectin (tumour necrosis factor) synthesis: mechanism of endotoxin resistance. Science, 232: 977979. Billingsley, M.L., Hall, N.R. and Mandel, H.G., 1982. Trauma induced glial proliferation: possible involvement of the immune system. Immunopharmacology, 5: 95-101. Birkenboch, F., 1983. The role of catecholamines in the control of the secretion of pro-opiomelanocortin derived peptides from the rat's pituitary gland and its implications in the response to stress. Ph.D. Thesis, Univ. Amsterdam, pp. 78-101. Biziere, K., Guillamin, J.M., Degenne, D., Bardos, P., Renoux, M. and Renoux, G., 1985. Lateralized neocortical modulation of the T-cell lineage. In: I.M. Guillemin, M. Cohn and T. Melnechuck (Editors), Neuralmodulation of Immunity. Raven Press, New York, NY, pp. 81-94. Black, S., Humphrey, J.H. and Niven, J., 1963. Inhibition of the Mantoux reaction by direct suggestion under hypnosis. Br. Med. J,, 346: 1649-1652. Blackwell, G.J., Cornuccio, M. and DiRosa, R.J., 1980. Macrocortin: a polypeptide causing the antiphospholypase effect of glucocorticoids. Nature (London), 187: 147-149. Blalock, J.E. and Smith, E.M., 1981. Human leukocyte interferon (HuIFN-o~): potent endorphinlike opioid activity. Biochem. Biophys. Res. Commun., 101: 472-478. Blalock, J.E, Johnson, H.M., Smith, E.M. and Torres, B.A., 1984. Enhancement of in vitro antibody response by thyrotropin. Biochem. Biophys. Res. Commun., 125: 30-34. Blecha, F. and Kelley, K.W., 1981. Effects of cold and weaning stressors on the antibody-mediated immune response of pigs. J. Anim. Sci., 53: 439-447. Blecha, F., Barry, R.A. and Kelley, K.W., 1982. Stress induced alterations in delayed-type hypersensitivity to SRBC and contact sensitivity to DNFB in mice. Proc. Soc. Exp. Biol. Med., 169: 239-246. Blecha, F., Pollman, D.S. and Nichols, D.A., 1983. Weaning pigs at an early age decreases cellular immunity. J. Anim. Sci., 56: 396-400. Blecha, F., Boyles, S.L. and Riley, J.G., 1984. Shipping suppresses lymphocyte blastogenic responses in Angus and Brahman X Angus feeder calves. J. Anim. Sci., 59: 576-583. Bockman, D.E. and Kirby, M.L., 1985. Neural crest interactions in the development of the immune system. J. Immunol., 135: 7665-7685.

303 Bohus, B., 1979. Effects of ACTH-like neuropeptides on animal behaviour and man. Pharmacology, 18: 113-122. Bohus, B., Urban, T., Wimersma, G. and De Werd, J., 1978. Opposite effects of oxytocin and vasopressin on avoidance behaviour and hippocampal theta rhythm in the rat. Neuropharmacology, 17: 239-247. Botanic, M., Pericic, D., Poljak-Blazi, M., Sverko, V. and Marolte, T., 1987. Suppression of immune response in rats by stress and drugs interfering with metabolism of serotorium. Ann. N.Y. Acad. Sci., 496: 485-491. Bourne, H., Lichtenstein, L., Melmon, K., Henney, C., Weinstein, Y. and Shearer, G., 1974. Modulation of inflammation and imunity by cyclic AMP. Science, 184: 19-28. Breazile, J.E., 1987. Physiologic basis and consequences of distress in animals. J. Am, Vet. Med. Assoc., 191: 1212-1215. Brown, M.R., Fisher, L.A., Speiss, C. Rivier, C., Rivier, J. and Vale, W., 1982. Corticotropin releasing factor: actions on the sympathetic nervous system and metabolism. Endocrinology, 111: 928-931. Brown, S.L., Smith, L.R. and Blalock, J.E., 1987. Interleukin 1 and interleukin 2 enhance propiomelonocortin gene expression in pituitary cells. J. Immunol., 139: 3181-3184. Bullock, K. and Moore, R.Y., 1981. Innervation of the thymus gland by brain stem and spinal cord in mouse and rat. Am. J. Anat., 162: 157-166. Burchfield, S.R., Woods, S.C. and Elich, M.S., 1978. Effects of cold stress on tumour growth. Physiol. Behav., 21: 537-540. Burton, R.R. and Beljan, J.R., 1970. Animal restraint applications in space (weightless) environment. Aerosp. Med., 41: 1060-1065. Cake, M. and Litwack, G., 1975. The glucocorticoid receptor. In: G. Litwack (Editor), Biochemical Actions of Hormones. Vol. 3. Academic Press, New York, NY, pp. 317-390. Callow, L.L. and Parker, R.J., 1969. Cortisone induced relapses in Babesia argentina infections of cattle. Aust. Vet. J., 45: 103-104. Calvo, W., 1968. The innervation of the bone marrow in laboratory animals. Am. J. Anat., 123: 315-328. Cannon, W.B., 1914. The emergency function of the adrenal medulla in pain and major emotions. Am. J. Physiol., 33: 356-372. Cannon, W.B., 1929. Bodily Changes in Pain, Hunger, Fear and Rage: an Account into the Function of Emotional Excitement. Appleton, New York, NY, 203 pp. Cannon, W.B., 1935. Stresses and strains of homeostosis. Am. J. Med. Sci., 189: 1-12. Carr, D.J.J. and Blalock, J.E., 1986. A molecular basis for bidirectional communication between the immune and neuroendocrine systems In: B. Cinader and R.G. Miller (Editors), Progress in Immunology VI. Academic Press, Toronto, Ont., pp. 619-629. Claman, H.N., 1972. Corticosteroids and lymphoid cells. N. Engl. J. Med., 287: 388-397. Cohen, J.J., 1987. Methodological issues in behavioural immunology. Immunol. Today 8: 33-34. Comsa, J., Leonhardt, H. and Wekerle, H., 1982. Hormonal coordination of the immune response. Rev. Physiol. Biochem. Pharmacol., 92:116-191. Conner, R.L., Vernikos-Danellis, J. and Levine, S., 1971. Stress fighting and neuroendocrine function. Nature (London), 234: 564-566. Crabtree, G.R., Munck, A. and Smith, K.A., 1980. Glucocorticoids and lymphocytes. 1. Increased glucocorticoids receptor levels in antigen stimulated lymphocytes. J. Immunol., 124: 24302435. Croiset, G., Veldhuis, H.D., Ballieux, R.E., De Wied, D. and Heijnen, C.J., 1987. The impact of mild emotional stress induced by passive avoidance procedure on immune reactivity. Ann. N.Y. Acad. Sci., 496: 477-484. Crookshank, H.R., Elissdale, M.H., White, R.G., Clanton, D.C. and Smalley, H.E., 1979. Effect

304 of transportation and handling calves upon blood serum composition. J. Anim. Sci., 48: 430435. Cross, J.P., Mackintosh, C.G. and Griffin, J.F.T., 1989. The effect of physical restraint and xylazine sedation on haematological values in red deer (Cervus elaphus ). Res. Vet. Sci. (in press ). Cross, R.H., Markesbery, W.R., Brooks, W.H. and Roszman, T,L., 1984. Hypothalamic immune interactions: neuromodulation of natural killer activity by lesioning of the anterior hypothalamus. Immunology, 51: 399-405. Cupps, T.R. and Fauci, A.S., 1982. Corticosteroid mediated immunoregulation in man. Immunol. Rev., 65: 133-155. Curtis, S.E., 1974. Response of piglets to perinatal stressors. J. Anim. Sci., 38: 1031-1037. Dantzer, R. and Mormede, P., 1985. Stress in domestic animals: a psychoneuroendocrine approach. In: G.P. Moberg (Editor), Animal Stress. Am. Physiol. Soc., Waverley Press, Bethesda, MD, pp. 81-95. Davies, D.H. and Duncan, J.R., 1974. The pathogenesis of recurrent infections with infectious bovine rhinotracheitis virus induced in calves by treatment with corticosteroids. Cornell Vet., 64: 340-366. Del Rey, A., Besedovsky, H.O., Sorkin, E., Da Prada, M. and Bondiolotte, G.P., 1982. Sympathetic immunoregulation: differences between high-and low-responder animals. Am. J. Physiol., 242: R30-R32. Dougherty, T.F., 1952. Effect of hormones on lymphatic tissues. Physiol. Rev., 32: 379-401. Duncan, M.E., Pearson, J.M.H., Ridley, D.H., Melson, R. and Bjune, G., 1982. Pregnancy and leprosy: the consequences of alterations of cell-mediated and humoral immunity during pregnancy and lactation. Int. J. Lepr., 50: 425-435. Edfors-Lilja, I., Lundstrom, K., Nyberg, L. and Rundgren, M., 1987. Influence of the Hal locus and standarised stress on antibody response and in vitro reactivity of peripheral blood lymphocytes, in pigs. Vet. Immunol. Immunopathol., 14: 157-171. Edwards, E.A., Rahe, R.H., Stephens, P.M. and Henry, J.S., 1980. Antibody response to bovine serum albumin in mice: the effects of psychosocial environmental change. Proc. Soc. Exp. Biol. Med., 164: 478-481. Ellis, E.F., 1981. Adverse effects of corticosteroid therapy. J. Allergy Clin. Immunol., 80: 515-518. Engel, G.L., 1967. A psychological setting of somatic disease: the "giving up-given up" complex. Proc. R. Soc. Med., 60: 553-557. Esterling, B. and Rabin, B.S., 1987. Stress-induced alterations of T-lymphocyte subsets and humoral immunity in mice. Behav. Neurosci., 101: 115-119. Ewbank, R., 1985. Behavioural responses to stress in farm animals. In: G.P. Moberg (Editor), Animal Stress, Am. Physiol. Soc., Waverley Press. Bethesda, MD, pp. 71-80. Fefer, A., McCoy, J.L., Perk, K. and Glynn, J.P., 1968. Immunologic, virologic and pathologic studies of regression of autochthonous Maloney sarcoma virus induced tumors in mice. Cancer Res., 28: 1577-1585. Feldman, R.D., Hunninghake, G.W. and McArdle, W.L., 1987. fl-andrenergic receptor-mediated suppression of interleukin 2 receptors in human lymphocytes. J. Immunol., 139: 3355-3360. Folch, H. and Waksman, B.H., 1974. The splenic suppressor cell. J. Immunol., 113: 127-139. Fontana, A., Greider, S., Arrenbrecht, S. and Grob, P., 1980. In vitro stimulation of glial cells by a lymphocyte produced factor, J. Neurol. Sci., 46: 55-62. Franzmann, A.W., Flynn, A. and Arneson, P.D., 1975. Serum corticoid levels relative to handling stress in Alaskan moose. Can. J. Zool., 53: 1424-1426. Fraser, D., Richie, J.S.D. and Frase, A.F., 1975. The term 'stress' in the veterinary context. Br. Vet. J., 131: 653-662. Friedman, S.B. and Glasgow, L.A., 1966. Psychologic factors and resistance to infectious disease. Pediatr. Clin. North Am., 13: 315-355.

305 Friedman, S.B., Glasgow, L.A. and Ader, R., 1969. Psychological factors modifying host resistance to experimental infections. Ann. N.Y. Acad. Sci., 164: 381-392. Gershbein, L.L., Cortell, R. and Feldman, J., 1941. The effect of emotion, sham rage and hypothalamic stimulation on the vago-insulin system. Am. J. Physiol., 132: 532-541. Gershbein, L.L., Benuck, I. and Shurrager, P.S., 1974. Influence of stress on lesion growth and on survival of animals bearing parenteral and intracerebral leukemia L120 and Walker tumors. Oncology, 30: 429-435. Ghanta, V.K., Hiramoto, R.N., Salvason, H.B. and Spector, N.H., 1985. Neural and environmental influences on neoplasia and conditioning NK activity. J. Immunol., 135: 848s-852s. Gibbs, D.M., 1986. Vasopressin and oxytocin: hypothalamic modulators of the stress response. Psychoneuroendocrinology, 11: 131-139. Gillette, S. and Gillette, R., 1979. Changes in thymic estrogen receptor expression following orchidectomy. Cell. Immunol., 42: 194-196. Gillis, S., Crabtree, G.R. and Smith, K.A., 1979. Glucocorticoid-induced inhibition of T cell growth factor. I. The effect on mitogen-induced lymphocyte proliferation. J. Immunol., 123: 16241631. Gilman, S.C., Schwartz, J.M., Milner, R.J., Bloom, F.E. and Feldman, J.D., 1982. fl-endorphin enhances lymphocyte proliferative responses. Proc. Natl. Acad. Sci., U.S.A., 79: 4226-4230. Giron, L.T., Crutcher, K.A. and Davis, J., 1980. Lymph nodes. A possible site for sympathetic neuronal regulation of immune responses. Ann. Neurol., 8: 520-525. Gisler, R.H. and Schenkel-Hulliger, L., 1971. Hormonal regulation of the immune response. Influence of pituitary and adrenal activity on immune responsiveness in vitro. Cell. Immunol., 2: 646-657. Goetzl, E.J., Chernov, T., Renold, F. and Payan, D.G., 1985. Neuropeptide regulation of the expression of immediate hypersensitivity. J. Immunol., 135: 802s-805s. Golub, M.S. and Gershwin, M.E., 1985. Stress induced immunomodulation: what is it, if it is. In: G.P. Moberg {Editor), Animal Stress. Am. Physiol. Soc., Waverley Press, Bethesda, MD, pp. 177-192. Good, R.A., 1972. Relations between immunity and malignancy. Proc. Natl. Acad. Sci., U.S.A., 69: 1026-1032. Gorczynski, R.M., Macrae, S. and Kennedy, M., 1982. Conditioned immune response associated with allogenic skin grafts in mice. J. Immunol., 129: 704-709. Graham, R., Reynolds, F.H.K. and Hill, J.F., 1919. Bacteriologic studies of a peracute disease of horses and mules. J. Am. Vet. Med. Assoc., 56: 378-393. Grandison, L. and Guidolta, A., 1977. Stimulation of food intake by muscimol and beta endorphin. Neuropharmacology, 169: 533-536. Griffin, J.F.T., 1987a. Acute bacterial infections in farmed deer. Ir. Vet. J., 41: 328-331. Griffin, J.F.T., 1987b. Stress and disease in farmed deer. Publ. Vet. Deer Soc., 2: 3-9. Griffin, J.F.T., 1989. The aetiology and diagnosis of tuberculosis in farmed deer. Ir. Vet. J. (in press). Gross, W.B., 1972. Effect of social stress on occurrence of Marek's disease in chickens. Am. J. Vet. Res., 33: 2275-2279. Gross, W.B., 1976. Plasma steroid tendency, social environment and Erineria necatrix infection. Poult. Sci., 55: 1508-1572. Gross, W.B., 1984. Effect of a range of social stress severity on Escherichia coli challenge infection. Am. J. Vet. Res., 45: 2074-2076. Gross, W.B. and Colmano, G., 1969. Effect of social isolation on resistance to some infectious agents. Poult. Sci., 48: 514-520. Gwazdauskas, F.C., Gross, W.B., Bibb, T.L. and McGillard, M.L., 1978. Antibody titres and plasma glucocorticoid concentrations near weaning in steers and heifer calves. Can. Vet. J., 19: 150154.

306 Gwazdauskas, F.C., Paape, M.J., Peery, D.A. and McGillard, M.L., 1980. Plasma glucocorticoid and circulating blood leukocyte responses in cattle after sequential intramuscular injections of ACTH. Am. J. Vet. Res., 41: 1052-1056. Hadden, J.W., 1983. Cyclic nucleotides and related mechanisms in immune regulation: a minireview. In: N. Fabris, E. Caraci, J. Hadden and N.A. Mitchinson (Editors), Immunoregulation. Plenum, New York, NY, pp. 201-239. Hadden, J., Hadden, E. and Middleton, E., 1970. Lymphocyte blast transformation. Demonstration of adrenergic receptors in human peripheral lymphocytes. Cell. Immunol., 1: 583-595. Hale, E.B., 1969. Domestication and evolution in behaviour. In: E.S.E. Hafez (Editor), The Behaviour of Domestic Animals, 2nd edn. Williams and Wilkins, Baltimore, MD, pp. 22-42. Hall, N.R. and Goldstein, A.L., 1985. Neurotransmitters and host defense. In: R. Guillemin, M. Cohn and T. Melnichuk (Editors), Neuromodulation of Immunity. Raven Press, New York, NY, pp. 143-156. Hall, N.R., McLure, J.E., Hsu, S.K., Tick, N.T., Seales, C.M. and Goldstein, A.L., 1980. Effect of chemical sympatectomy upon thymus dependent immune responses. Soc. Neurosci. Abstr., 26: 4. Hall, N.R., McGillis, J.P., Spangelo, B.L. and Goldstein, A.L., 1985. Evidence that thymosins and other biologic response modifiers can function as neuroactive immunotransmitters. J. Immunol., 135: 806s-811s. Hall, R.D., Gross, W.B. and Turner, E.C.Jr., 1979. Population developments of Ornithonyssus sylviarum on Leghorn roosters inoculated with steroids and subjected to extremes of social interactions. Vet. Parasitol., 5: 287-297. Hambdy, A.H., Trapp, A.I., Gale, C., 1963. Experimental transmission of shipping fever in calves. Am. J. Vet. Res., 24: 287-294. Harris, G.W., 1948. Neural control of the pituitary gland. Physiol. Rev., 28: 139-179. Haynes, L.W. and Timms, R.J., 1987. Stress-induced release of pituitary fl-endorphin may be mediated by activation of the brain stem defence areas. Int. J. Tissue React., 9:55 59. Hazum, E., Chang, K. and Cuatrecasas, P., 1979. Specific nonopiate receptors for fl-endorphin. Science, 205: 1033-1035. Healy, D.L., Hodgen, G.D., Schulte, H.M., Chrouros, G.P., Loriaux, D.L., Hall, N.R. and Goldstein, A.L., 1983. The thymus-adrenal connection: thymosin has corticotropin-releasing activity in primates. Science, 222: 1353-1355. Helderman, J. and Strom, T., 1978. Specific insulin binding site on T and B lymphocytes as a marker of cell activation. Nature (London), 274: 62-63. Heller, E.D., Nathan, D.B. and Perek, M., 1979. Short heat stress as an immunostimulant in chicks. Avian Pathol., 8: 195-203. Henry, J.P. and Stephens-Larson, P., 1985. Specific effects of stress on disease processes. In: G.P. Moberg (Editor), Animal Stress. Am. Physiol. Soc., Waverley Press, Baltimore, MD, pp. 161175. Hirata-Hibi, M., 1967. Plasma cell reaction and thymic germinal centres after a chronic form of electrical stress. J. Reticuloendothel. Soc., 4: 370-389. Hoerlein, A.B., 1980. Shipping fever. In: H.E. Amstutz (Editor), Bovine Medicine and Surgery. Am. Vet. Publ., Santa Barbara, CA, pp. 99-106. Hoerlein, A.B. and Marsh, G.L., 1957. Studies on the epizootiology of shipping fever in calves. J. Am. Vet. Med. Assoc., 131: 123-127. Holaday, J.W., Loh, H. and Li, C.H., 1978. Unique behavioural effects of fl-endorphin and their relationship to thermoregulation and hypothalamic function. Life Sci., 22: 1525-1536. Hollands, C., 1986. The Animals (Scientific Procedures) Act. Lancet, July 5: 32-33. Hollenburg, M. and Cuatrecasas, P., 1974. In: B. Clarkson and R. Baserga (Editors), Control of Proliferation of Animal Cells. Cold Spring Harbour, NY, pp. 77-85.

307 Hudson, R.J., 1973. Stress and in vitro lymphocyte stimulation by phytohaemagglutinin in Rocky Mountain Bighorn sheep. Can. J. Zool., 51: 479-482. Jenkins, P.E., Chadwick, R.A. and Nevin, J.A., 1983. Classically conditioned enhancement of antibody production. Bull. Psychosom. Soc., 21: 485-491. Jennings, A.R. and Glover, R.E., 1952. Enzootic pneunomia in calves. II. The experimental disease. J. Comp. Pathol., 62: 6-22. Jensen, M.M., 1969. Changes in leukocyte counts associated with various stressors. J. Reticuloendothel. Soc., 8: 457-465. Joasoo, A. and McKenzie, J.M., 1976. Stress and the immune response in rats. Int. Arch. Allergy Appl. Immunol., 50: 659-663. Johnson, H.M. and Torres, B.A., 1985. Regulation of lymphokine production by arginine vasopressin and oxytocin: modulation of lymphocyte function by neurohypophyseal hormones. J. Immunol., 135: 773s-775s. Johnson, H.M., Smith, E.M., Torres, B.A. and Blaloc, J.E., 1982. Neuroendocrine hormone regulation of in vitro antibody production. Proc. Natl. Acad. Sci. U.S.A., 79: 4171-4174. Johnston, N.E. and Buckland, R.B., 1976. Response of male Holstein calves from seven sires to four management stresses as measured by plasma corticoid levels. Can. J. Anim. Sci., 56: 727732. Justice, A., 1985. Review of the effects of stress on cancer in laboratory animals: importance of time of stress application and type of tumour. Psychol. Bull., 98: 108-138. Kagan, A.R. and Levi, L., 1974. Health and environment-psychosocial stimuli: a review. Soc. Sci. Med., 8: 225-241. Kaliss, N. and Fuller, J.L., 1968. Incidence of lymphatic leukemia and methylcholanthrene-induced cancer in laboratory mice subjected to stress. J. Natl. Cancer Inst., 41: 967-981. Kapcala, L.P., 1988. Stimulated release of immunoreactive adrenocorticotropin and b-endorphin from extra hypothalamic brain. Neuroendocrinology, 47: 50-54. Keller, S.E., Weiss, J.M., Scheifler, S.J., Miller, N.E. and Stein, M., 1981. Suppression of immunity by stress: effect of a graded series of stressors on lymphocyte stimulation in the rat. Science, 213: 1397-1400. Keller, S.E., Weiss, J.M., Miller, N.E. and Stein, M., 1983. Stress-induced suppression of immunity in adrenalectomised rats. Science, 221: 1301-1304. Kelley, K.W., 1980. Stress and immune function: a bibliographic review. Ann. Rech. Vet., 11: 445478. Kelley, K.W., 1985. Immunological consequences of changing environmental stimuli. In: G.P. Moberg (Editor), Animal Stress. Am. Physiol. Soc., Waverley Press, Bethesda, MD, pp. 193223. Kelley, K.W., Osborne, C.A., Evermann, J.F., Parish, S.M. and Hinrichs, D.G., 1981. Whole blood leukocytes vs separated mononuclear cell blastogenesis in calves. Time-dependent changes after shipping. Can. J. Comp. Med., 45: 249-258. Kelley, K.W., Osborne, C.A., Evermann, J.F., Parish, S.M. and Gaskins, C.T., 1982a. Effects of chronic heat and cold stressors on plasma immunoglobulin and mitogen induced blastogenesis in calves. J. Dairy Sci., 65: 1514-1528. Kelley, K.W., Randall, E., Greenfield, B.S., Evermann, J.F., Parish, S.M. and Perryman, L.E., 1982b. Delayed type hypersensitivity, contact sensitivity, and phytohaemagglutinin skin-test responses of heat and cold stressed calves. Am. J. Vet. Res., 43: 775-779. Kelso, A. and Munk, A., 1984. Glucocorticoid inhibition of lymphokine secretion by alloreactive T lymphocyte clones. J. Immunol., 133: 784-791. Kent, T.E. and Ewbank, R., 1983. The effect of road transportation on blood constituents and behaviour of calves. I. Six months old. Br. Vet. J., 139: 228-235. Knudsen, P.J., Dinorello, C.A. and Strom, T.B., 1987. Glucocorticoids inhibit transcriptional and post transcriptional expression of interleukin 1 in U937 cells. J. Immunol., 139: 4129-4134.

308 Kreiger, D.T., 1983. Brain peptides: what, where, why? Science, 222: 975-985. Kvetnansky, R., Kopi, I,J. and Saavedra, J.M., 1978. Changes in epinephrine in individual hypothalmic nuclei after immobilization stress. Brain Res., 155: 387-390. Laudenslager, M.L., Ryan, S.M., Drugan, R.C., Hyson, R.L. and Maier, S.F,, 1983. Coping and immunosuppression: inescapable but not escapable shock suppresses lymphocyte proliferation. Science, 221: 568-570. Leach, T.M., 1982. Physiology of the transport of cattle. In: B. Moss (Editor), Transport of Animals Intended for Breeding, Production and Slaughter. Martinus Nijhoff, The Hague, pp. 5774. Levine, S., 1985. A definition of stress? In: G.P. Moberg (Editor), Animal Stress. Am. Physiol. Soc., Waverley Press, Bethesda, MD, pp. 51-69. Lewis, J.W., Cannon, J.T. and Liebeskind, J.C., 1980. Opioid and non-opioid mechanisms of stress analgesia. Science, 208: 623-625. Lewis, J.W., Tordoff, M.G., Sherman, J.E. and Liebeskind, J.C., 1982. Adrenal medullary enkephalin-like peptides may mediate opioid stress analgesia. Science, 217: 557-559. Lewis, J.W., Shavit, Y., Terman, G.W., Nelson, L.R., Martin, F.C., Gale, R.P. and Liebeskind, J.C., 1985. Involvement of opioid peptides in the analgesic, immunosuppressive and tumorenhancing effects of stress. Psychopharmacol. Bull., 21: 479-484. Lim, A.T., Lolait, S. and Barlow, J.W., 1983. Immunoreactive fl-endorphin in sheep ovary. Nature (London), 303: 709-711. Liolta, A.S., Loudes, C., McKelry, J.F. and Kreiger, D.T., 1980. Biosynthesis of precursor corticotropin/endorphin, corticotropin, and melanotropin, fl lipotropin, and fl-endorphin-like material by cultured neonatal rat hypothalmic neurons. Proc. Natl. Acad. Sci. U.S.A., 77: 18801884. Livnat, S., Felten, S.Y., Carlson, S.L. and Felten, D.L., 1985. Involvement of peripheral and central catecholamine systems in neural-immune interractions. J. Neuroimmunol., 10: 5-30. Lumpkin, M.D., 1987. The regulation of ACTH synthesis by IL-1. Science, 238: 452-455. MacDermott, R.P. and Stacey, M.C., 1981. Further characterisation of the human autologous mixed leukocyte reaction (MLR). J. Immunol., 126: 729-734. Machado-Neto, R., Graves, C.N. and Curtis, S.E., 1987. Immunoglobulins in piglets from sows heat stressed pre partum. J. Anita. Sci., 65: 445-455. Mackintosh, C.G. and Henderson, T.G., 1984. The epidemiology of yersiniosis in deer. Proc. Deer Course for Veterinarians, N.Z.V.A., 1: 34-42. Maier, S.E., Sherman, J.E., Lewis, J.G., Termon, G.W. and Liebeskind, J.C., 1983. The opioidnonopioid nature of stress induced analgesia and learned helplessness. J. Exp. Psychol. (Anim. Behav.), 9: 80-86. Mains, R. and Eipper, B., 1981. Coordinate equimolar secretion of smaller peptide products derived from pro-ACTH/endorphin by mouse pituitary tumor cells. J. Cell Biol., 89: 21-28. Marsh, J.T. and Rasmussen, A.F.Jr., 1960. Response of adrenals, thymus, spleen and leukocytes to shuttle box and confinement stress. Proc. Soc. Exp. Biol. Med., 104: 180-183. Martin, S.W., Schwabe, C.W. and Franti, C.E., 1975. Dairy calf mortality rate. Association of daily meteorological factors. Can. J. Med., 39: 377-388. Mason, J.W., 1968. "Over-all" hormonal balance as a key to endocrine function. Psychosom. Med., 30: 791-808. Mathews, P.M., Froelich, C.J., Sibbitt, W.L.Jr. and Bankhurst, A.D., 1983. Enhancement of natural cytfltoxicity by fl-endorphin. J. Immunol., 130: 1658-1661. Matthes, T.H., 1963. Experimental contribution to the question of emotional stress reactions on the growth of tumours in animals. Proc. VIIIth Anti-Cancer Congr., 3: 1608-1610. May, I., Manoiu, I., Donta, C., Tetu, M., Vior, C. and St Maldovan, C., 1979. Stress und Immunit~it beim Rind. Arch. Exp. Veterinaermed., 33: 87-98.

309 McAllum, H.J.F., 1980. Common diseases in farmed red deer. Ag link FPP 254. Min. Agric. Fish., Wellington, New Zealand, 3 pp. McCain, H.W., Lamster, I.B., Bozzone, J.M. and Grbic, J.T., 1982./? endorphin modulates human immune activity via non-opiate receptor mechanisms. Life Sci., 31: 1619-1624. McDowell, R.E. and McDaniel, B.T., 1965. Interbreed matings in dairy cattle. 11. Herd health and viability. J. Dairy Sci., 57: 1275-1283. McGaugh, J.L., 1983. Hormonal influences on memory. Annu. Rev. Psychol., 34: 297-323. McGillis, J.P., Organist, M.L. and Payan, D.G., 1987. Substance P and immunoregulation. Fed. Proc., 46: 196-199. Miles, K., Quintons, J., Chelmicko-Schorr, E. and Arneson, B.G., 1981. The sympathetic nervous system modulates antibody response to thymus dependent antigens. J. Neuroimmunol., 1: 101108. Moberg, G.P., 1985. Biological response to stress: key to assessment of animal well being. In: G.P. Moberg (Editor), Animal Stress. Am. Physiol. Soc., Waverley Press, Bethesda, MD, pp. 2749. Monjan, A.A. and Collector, M.I., 1977. Stress-induced modulation of the immune response. Science, 196: 307-308. Morrow, D.A., 1976. Fat cow syndrome. J. Dairy Sci., 59: 1625-1629. Morton, D.B., 1987. Epilogue: Colloquium on recognition and alleviation of animal pain and distress. J. Am. Vet. Med. Assoc., 191: 1292-1296. Murata, H., Takahashi, H. and Matsumoto, H., 1987. The effect of road transportation on peripheral blood lymphocyte subpopulations, lymphocyte blastogenesis and neutrophil function in calves. Br. Vet. J., 143: 166-174. Murphy, E.D., 1966. Characteristic tumors. In: E.G. Green (Editor), Biology of the Laboratory Mouse. McGraw-Hill, New York, NY, pp. 521-562. Newbould, F.H.S., 1974. Microbial diseases of the mammary gland. In: Lactation. Academic Press, New York, NY, pp. 269-316. Nicol, L., 1974. L'Epop~e Pastorienne et al Medicine Veterinaire. L. Nicol (Publ.), Garches, France, pp. 197-211. Niilo, L., 1970. The effect of dexamethasone on bovine coccidiosis. Can. J. Comp. Med., 34: 325328. O'Dorisio, M.S., 1987. Biochemical characteristics of receptors for vasoactive intestinal poly-peptide in nervous, endocrine and immune systems. Fed. Proc., 46: 192-195. Okimura, T. and Nigo, Y., 1986. Stress and immune responses. I. Suppression of T cell function in restraint stressed mice. Jpn. J. Pharmacol., 40:505-511. Okimura, T., Satomi-Sasaki, Y. and Ohkuma, S., 1986a. Stress and immune responses. II. Identification of stress-sensitive cells in murine spleen cells. Jpn. J. Pharmacol., 40: 513-525. Okimura, T., Ogawa, M. and Yamouchi, T., 1986b. Stress and immune responses. III. Effect of restraint stress on delayed type hypersensitivity (DTH) responses, natural killer (NK) activity and phagocytosis in mice. Jpn. J. Pharmacol., 41: 229-235. Ottaway, C.A., 1987. Selective effects of vasoactive intestinal peptide on the mitogenic response of murine T cells. Immunology, 62: 291-297. Ottaway, C.A. and Greenberg, G., 1984. Interaction of vasoactive intestinal peptide with mouse lymphocytes: specific binding and the modulation of mitogen response. J. Immunol., 132: 417422. Owen, Rh.ap Rh., 1975. Poststress diarrhoea in the horse. Vet. Rec., 96: 267-270. Owen, Rh.ap Rh., Fullerton, J. and Barnum, D.A., 1983. Effect of transportation, surgery, and antibiotic therapy in ponies infected with salmonella. Am. J. Vet. Res., 44: 46-50. Panksepp, J., Herman, B.H., Vilberg, T., Bishop, P. and De Eskinazi, F.G., 1978. Endogenous opioids and social behaviour. Neurosci. Biobehav. Rev., 4: 473-487.

310 Parillo, J.E. and Fauci, A.S., 1979. Mechanisms of glucocorticoid action on immune processes. Ann. Rev. Pharmacol. Toxicol., 19: 179-201. Payan, F.G., Levine, J.D. and Goetzl, E.J., 1984. Modulation of immunity and hypersensitivity by sensory neuropeptides. J. Immunol., 132: 1601-1604. Pericic, D., Manev, H., Boranic, M., Poljak-Blazi, M. and Lekic, N., 1987. Effect of diazepam on brain neurotransmitters, plasma corticosterone and the immune system of stressed rats. Ann. N.Y. Acad. Sci., 496: 450-458. Pitkin, D.H., 1965. Effects of physiological stress on the delayed hypersensitivity reaction. Proc. Soc. Exp. Biol. Med., 120: 350-351. Plaut, S.M., Ader, R., Friedman, S.B. and Ritterson, A.L., 1969. Social factors and resistance to malaria in the mouse: effects of group versus individual housing on resistance to Plasmodium berghei infection. Psychosom, Med., 31: 536-552. Plotnikoff, N.P., Murgo, A.J., Miller, C.G., Corder, C.N. and Faith, R.E., 1985. Enkephalins: immunomodulators. Fed. Proc. Am. Soc. Exp. Biol., 44: 118-122. Previte, J.J. and Berry, J.L., 1962. The effect of environmentaltemperatures on the host-parasite relationship in mice. J. Infect. Dis., 110: 201-209. Rasmussen, A.F.Jr., Spenser, E.T. and Marsh, J.R., 1959. Decrease in susceptibility in mice to passive anaphylaxis following avoidance-learningstress. Proc. Soc. Exp. Biol. Med., 100: 878879. Regnier, J.A. and Kelley, K.W., 1981. Heat- and cold-stress suppresses in vivo and in vitro cellular immune responses of chickens. Am. J. Vet. Res., 41: 294-299. Regnier, J.A., Kelley, K.W. and Gaskins, C.T., 1980. Acute thermal stressors and synthesis of antibodies in chickens. Poult. Sci. 59: 985-990. Reilley, F.D., McCuskey, R.S. and Meineke, H.A., 1976. Studies of the hemopoietic microenvironment. VIII. Andrenergic and cholinergic innervation of the murine spleen. Anat. Rec., 185: 109-119. Reilly, F., McCuskey, P., Miller, M., McCuskey, R. and Meineke, H., 1979. Innervation of the periarteriolar lymphatic sheath of the spleen. Tissue Cell, 11: 121-126. Reite, M., Harbeck, R. and Hoffman, A., 1981. Altered cellular immune response following peer separation. Life Sci., 29: 1133-1136. Rich, A.R., 1951. The Pathogenesis of Tuberculosis. Charles C. Thomas, Springfield, IL, pp. 181195. Richman, D. and Arnason, B., 1979. Nicotinic actylcholine receptor on human lymphocytes. Proc. Natl. Acad. Sci. U.S.A., 76: 4632-4635. Riley, V., 1981. Psychoneuroendocrine influences on immunocompetence and neoplasia. Science, 212: 1100-1109. Roberts, N.J.Jr., 1979. Temperature and host defence. Microbiol. Rev., 43: 241-259. Roszman, T.L., Cross, R.J., Brooks, W.H. and Markesbery, W.R., 1985. Neuroimmodulation: effects of neural lesions on cellular immunity, In: R. Guillemin, M. Cohn and T. Melnechuk (Editors), Neural Modulation of Immunity. Raven Press, New York, NY, pp. 95-109. Roth, J.A., 1985. Cortisol as a mediator of stress associated immunosuppression in cattle. In: G.P. Moberg (Editor), Animal Stress. Am. Physiol. Soc., Waverley Press, Bethesda, MD, pp. 225243. Roth, J.A. and Kaeberle, M.L., 1982. Effect of the glucocorticoids on the bovine immune system. J. Am. Vet. Med. Assoc., 180: 894-901. Roth, J.A. and Kaeberle, M.L., 1983. Suppression of neutrophil and lymphocyte function induced by a vaccinal strain of bovine viral diarrhoea virus with and without the administration of ACTH. Am. J. Vet. Res., 44: 2366-2372. Roth, J.A., Kaeberle, M.L. and Hsu, W.H., 1982. Effect of adrenocorticotropin administration on bovine polymorphonuclear leukocyte function and lymphocyte blastogenesis. Am. J. Vet. Res., 43: 412-416.

311 Roth, J.O., LeRoth, D., Collier, E.S., Weaver, N.R., Watkinson, A., Cleland, C.F. and Glick, S.M., 1985. Evolutionary origins of neuropeptides, hormones, and receptors: possible applications to immunology. J. Immunol., 135: 816s-818s. Selye, H., 1946. The general adaptive syndrome and diseases of adaption. J. Clin. Endocrinol., 6: 117-230. Selye, H., 1973. The evolution of the stress concept. Am. Sci., 61: 692-699. Shavit, Y., Lewis, J.W., Terman, G.W., Gale, R.P. and Liebeskind, J.C., 1984. Opioid peptides mediate the suppressive effect of stress on natural killer cell cytotoxicity. Science, 223: 188190. Shavit, Y., Terman, G.W., Martin, F.C., Lewis, J.W., Liebeskind, J.C. and Gale, R.P., 1985. Stress, opioid peptides, the immune system and cancer. J. Immunol., 135: 834s-837s. Shaw, K.E. and Nichols, R.E., 1964. Plasma 17fl-hydroxy-corticosteroids in calves: the effects of shipping. Am. J. Vet. Res., 25: 252-253. Sheffey, B. and Davies, D.H., 1972. Reactivation of bovine herpesvirus after corticosteroid treatment. Proc. Soc. Exp. Biol. Med., 140: 974-976. Shek, P. and Sabiston, B., 1983. Neuroendocrine regulation of immune processes: change in circulating corticosterone levels induced by the primary antibody response in mice. Int. J. Immunopharmacol., 5: 23-33. Shimizu, M., Shimizu, Y. and Kodama, Y., 1978. Effect of ambient temperatures on induction of transmissible gastroenteritis in feeder pigs. Infect. Immun., 21: 747-752. Shope, R.E., Muscoplat, C.C., Chen, A.W. and Johnson, D.W., 1976. Mechanism of protection from primary bovine viral diarrhea virus infection. 1. The effect of dexamethasone. Can. J. Comp. Med., 40: 355-359. Singh, U., Millson, D., Smith, P. and Owen, J., 1979. Identification of fl-adrenoreceptors during thymocyte ontogeny in mice. Eur. J. Immunol., 9: 31-35. Sivas, A., Nysal, M. and Oz, H., 1982. The hyperglycemic effect ofthymosin F5, a thymic hormone. Horm. Metab. Res., 14: 330-331. Sklar, L.S. and Anisman, H., 1981. Stress and cancer. Psychol. Bull., 89: 369-406. Smith, E.M. and Blalock, J.E., 1981. Human lymphocyte production of ACTH and endorphin like substances: association with leukocyte interferon. Proc. Natl. Acad. Sci. U.S.A., 78: 75307534. Smith, E.M., Harbour-McMenamin, D. and Blalock, J.E., 1985. Lymphocyte production of endorphins and endorphin mediated immunoregulatory activity. J. Immunol., 135: 779s-782s. Smith, E.M., Phan, M., Coopenhaver, D., Kruger, T.E. and Blalock, J.E., 1983. Human leukocyte production of immunoreactive thyrotropin. Proc. Natl. Acad. Sci. U.S.A., 80: 6010-6013. Smith, K.A., 1982. T-cell growth factor and glucocorticoids: opposing regulatory hormones in neoplastic cell growth. Immunobiology, 161: 157-173. Smith, R.G. and McDaniel, S.M., 1983. Psychologically mediated effects on the delayed hypersensitivity reaction in humans. Psychosom. Med., 45: 65-69. Snow, E.C., 1985. Insulin and growth hormone function as minor growth factors that potentiate lymphocyte activation. J. Immunol., 135: 776s-778s. Snyder, D. and Unanue, E., 1982. Corticosteroids inhibit murine macrophage Ia expression and interleukin I production. J. hnmunol., 129: 1803-1805. Soder, O. and Hellstrom, P.M., 1987. Neuropeptide regulation of human thymocytes, guinea pig lymphocyte-T and rat lymphocyte-B mitogenesis. Int. Arch. Allergy Appl. Immunol., 84: 205212. Solomon, G.F., 1969. Stress and antibody response in rats. Int. Arch. Allergy Appl. Immunol., 35: 97-101. Staples, G.E. and Haugse, C.N., 1974. Losses in young calves after transportation. Br. Vet. J., 130: 374-378.

312 Stephens, D.B., 1980. Stress and its measurement in domestic animals. Adv. Vet. Sci. Comp. Med., 24: 179-210, Stockdale, P.H.G. and Niilo, L., 1976. Production of bovine coccidiosis with Eimeria zuervii. Can. Vet. J., 17: 35-37. Strom, T., Sytowski, A., Carpenter, C. and Merrill, J., 1974. Cholinergic augmentation of lymphocyte-mediated cytotoxicity. A study of the cholinergic receptor of cytotoxic T lymphocytes. Proc. Natl. Acad. Sci. U.S.A., 71: 1330-1333. Takayama, H., Ota, Z. and Ogawa, N., 1986. Effect of immobilisation stress on neuropeptides and their receptors. Regul. Pept., 15: 239-248. Terman, G.W., Shaint, Y., Lewis, J.W., Cannon, J.J. and Liebeskind, J.C., 1984. Intrinsic mechanisms of pain inhibition: activation by stress. Science, 226: 1270-1277. Teshima, H., Sogawa, H., Kihara, S., Ago, Y. and Nagawa, T., 1987. Changes in populations of Tcell subsets due to stress. Ann. N.Y. Acad. Sci., 496: 459-466. Thaxton, P., 1978. Influence of temperature on the immune response of birds. Poult. Sci., 57: 1430-1440. Tornello, S., Orti, E., De Nicola, A., Rainbow, T. and McEwen, B., 1982. Regulation of glucocorticoid receptors in the brain by corticosterone treatment of adrenalectomized rats. Neuroendocrinology, 35: 411-417. Ulrich, R.E. and Azrin, N.H., 1962. Reflexive fighting in response to aversive stimulation. J. Exp. Anim. Behav., 5: 511-520. Veale, W.I., Kasting, K.I. and Cooper, K.E., 1981. Arginine vasopressin and endogenous antipyreses: evidence and significance. Fed. Proc. Fed. Am. Soc. Exp. Biol., 40: 2750-2753. Vessey, S.H., 1964. Effects of grouping on levels of circulating antibodies in mice. Proc. Soc. Exp. Biol., 115: 252-255. Visintainer, M.A., Volpicelli, J.R. and Seligman, M.E.P., 1982. Tumor rejection in rats after exposure to inescapable and escapable shock. Science, 216: 437-439. Waage, A. and Bakke, O., 1988. Glucocorticoids suppress the production of tumour necrosis factor by lipopolysaccharide stimulated human monocytes. Immunology, 69: 299-302. Walcott, B. and McLean, J.R., 1985. Catecholamine-containing neurons and lymphoid cells in a lacrimal gland of the pigeon. Brain Res., 328: 129-137. Werb, Z., Foley, R. and Munck, A., 1978. Interaction of glucocorticoids with macrophages. J. Exp. Med., 147: 1684-1694. Westly, H.J. and Kelley, K.W., 1984. Physiological concentrations of cortisol suppress cell-mediated immune events in the domestic pig. Proc. Soc. Exp. Biol. Med., 177: 156-164. Wideman, C.H. and Murphy, H.M., 1985. Effects of vasopressin deficiency, age and stress on stomach ulcer induction in rats. Peptides (Suppl.), 6: 63-67. Williams, A.F., 1987. A year in the life of the immunoglobulin family. Immunol. Today, 8:298 303. Williams, A.R., Carey, R.J. and Miller, M., 1985. Altered emotionality of the vasopressin deficient Brattleboro rat. Peptides (Suppl)., 6: 69-76. Williams, J.M. and Felten, D., 1981. Sympathetic innervation of murine thymus and spleen. A comparative histofluorescence study. Anat. Rec., 199:531-542. Williams, J.M., Peterson, R.G., Shea, R.G., Bauer, D.C. and Felten, D.L., 1981. Sympathetic innervation of the murine thymus and spleen: evidence of a functional link between the nervous and immune systems. Brain Res. Bull., 6: 83-94. Woloski, B., Smith, E.M., Meyer, W.J., Fuller, G.M. and Blalock, J.E., 1985. Corticotropin-releasing activity of monokines. Science, 230: 1035-1037. Wybran, J., Appleboom, T., Famaey, J. and Govaerts, A., 1979. Suggestive evidence for receptors for morphine and methionine-enkephalin on normal human blood T-lymphocytes. J. Immunol., 123: 1068-1070. Yoneda, Y., Kanmori, K., Ida, S. and Kuriyama, K., 1983. Stress induced alterations in metabolism of 7-aminobutyric acid in rat brain. J. Neurochem., 40: 350-356.