- Email: [email protected]
Corticosterone-independent alteration of lymphocyte mitogenic function by amphetamine
BRAIN,
BEHAVIOR,
AND
IMMUNITY
6, 293-299
(1992)
Corticosterone-Independent Mitogenic Function MICHAEL
Alteration of Lymphocyte by Amphetamine
A. PEZZONE, KELLY A. RUSH, ALEXANDER W. KUSNECOV, PAUL G. WOOD, AND BRUCE S. RABIN
The Brain, Behavior and Immunity Center and The Division of Clinical Immunopathology, Department of Pathology, Universiry of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15213 Amphetamine, a neural stimulatory agent with acute effects mimicking those of stress, is shown here to elevate plasma corticosterone levels and suppress spleen and peripheral blood lymphocyte (PBL) mitogenic responses to concanavalin A (Con A) and phytohemagglutinin (PHA) when administered to rats. Pretreatment of the rats with propranolol, a nonselective P-adrenergic receptor antagonist, totally prevented the amphetamineinduced suppression of lymphocyte mitogenic reactivity to Con A and PHA in the spleen and to PHA in the peripheral blood; however, the PBL mitogenic response to Con A was only partially restored. Although the amphetamine-induced alterations in immune function were prevented by propranolol pretreatment, the elevated plasma corticosterone response was not. This suggests that corticosterone is not modulating the mitogenic activity of splenic lymphocytes or PHA-reactive PBLs. On the other hand, Con A-reactive PBLs may be affected by corticosterone and/or other mechanisms, which may D 1992 Academic press. IIIC. include the catecholamines.
Adrenal glucocorticoids become elevated in plasma following exposure of animal and human subjects to a variety of stressors (Munck & Guyre, 1991). Glucocorticoids, which are known to exert a suppressive effect on lymphocyte function (Cupps & Fauci, 1982), are utilized therapeutically to help prevent allograft rejection (Turk, 1975). Thus, it seems reasonable to hypothesize that glucocorticoids may mediate stressor-induced suppression of immune function by affecting the viability of lymphoid cells and/or the production of mononuclear cell-derived cytokines (Munck & Guyre, 1991). Recently, the role of glucocorticoids in stressor-induced suppression of immune function has been questioned. In both rats and humans, stress has been found to elevate glucocorticoids in the absence of a concomitant alteration of lymphocyte mitogenic activity (Flores, Hernandez, Hargreaves, & Bayer, 1990; Jessop & Bayer, 1989; Manuck, Cohen, Rabin, Muldoon, & Bachen, 1991). Further, a shock-induced reduction of peripheral blood and spleen lymphocyte responses to nonspecific mitogens has been demonstrated in adrenalectomized rats (Keller, Weiss, Schleifer, Miller, & Stein, 1983). The injection of rats with morphine causes lymphocyte functional alterations along with an elevation of corticosteronc (Bayer, Daussin, Hernandez, & Irvin, 1990); however, the lymphocyte functional changes are completely reversed by naltrexone, while the corticosterone elevation persists. Environmental situations, such as the number of mice housed in a cage, also influence the responsiveness of lymphocytes to mitogenic stimulation by a corticosterone-independent mechanism (Rabin, Lyte, & Hamill, 1987). Finally, the role of corticosterone in classically conditioned immune modulation has been questioned (Ader & Cohen, 1985) with recent evidence indicating a disso293 0889-1591192
$5.00
Copyright Q I%? by Academic Press, Inc. All rights of reproduction in any form reserved.
294
PEZZONE
ET AL.
ciation between corticosterone elevation and immune suppression in mice conditioned with cyclophosphamide (Roudebush & Bryant, 1991). The hypothalamic-pituitary-adrenal (HPA) axis may participate in stressorinduced immune alterations without involvement of its adrenocorticoid component. An adrenal-independent role for corticotropin-releasing factor (CRF) in stress-induced immune alterations has recently been suggested (Jain, Zwickler, Hollander, Brand, Saperstein, Hutchinson, Brown, & Audhya, 1991). By treating rats with CRF antisera or a CRF antagonist, the stress-induced suppression of lymphocyte mitogenic activity was similarly ameliorated in both adrenalectomized and intact rats. CRF, which is present throughout the rat brain (Chappell Smith, Kilts Bissette, Ritchie, Anderson, & Nemeroff, 1986), may function as a mediator of sympathetic nervous system activation during stress (Munck & Guyre, 1991). This role of CRF is supported by a study in which interleukin-I (IL-l) infusion into rat brains (Sundar, Cierpial, Kilts, Ritchie, & Weiss, 1990) decreased immune function by a mechanism involving CRF activation of the sympathetic nervous system. Alternatively, CRF may induce production of IL-l by monocytes, which may subsequently induce B-endorphin production from B lymphocytes (Kavelaars, Ballieux, & Heijnen, 1990). B-Endorphin may then alter the function of T lymphocytes. There is evidence which suggests that the HPA axis may counteract the suppressive influence of stress on lymphocyte function. Following a stressor, hypophysectomized rats have a more pronounced depression of lymphocyte mitogenie function than do intact rats (Keller, Schleifer, Liotta, Bond, Farhoody, & Stein, 1988). Consistent with this, our unpublished data show that lesions of the hypothalamic paraventricular nucleus in rats result in a more profound stressorinduced suppression of lymphocyte mitogenic function when compared to shamlesioned animals (Pezzone, manuscript in preparation). The immunomodulating potential of acutely administered amphetamine was investigated in this study, as many of its neurochemical, physiological, and behavioral effects are identical to those elicited by various stressors (Antelman & Chiodo, 1983). Furthermore, because the P-adrenergic receptor antagonist propranolol prevents the suppression of lymphocyte mitogenic activity induced by a footshock stressor (Cunnick, Lysle, Kucinski, & Rabin, 1990), its ability to mod.ify the potential immunomodulating effects of amphetamine was also investigated. We have also sought to provide additional information regarding the role of glucocorticoids in the functional alteration of lymphocyte activity in amphetaminetreated rats, since acutely administered amphetamine elevates corticosterone levels independent of propranolol administration (Knych & Eisenberg, 1979). METHODS
Experimental animals. Male rats of the Lewis strain, 65 days old and 250-300 g in weight, were purchased from Harlan Sprague Dawley (Indianapolis, IN). The animals were individually housed in hanging wire cages under a 12-h day-night cycle (lights on at 0700) of artificial illumination and ad lib access to food and water. Upon arrival in the animal housing facility, rats were given a 2-week acclimatization to the new surroundings. Furthermore, the animals were gently handled approximately 2 mm/day for 5 consecutive days during their first week in the new colony room. On the day prior to experimental manipulation all rats were weighed.
CORTICOSTERONE-INDEPENDENT
IMMUNE
MODULATION
295
Treatment groups. (D)-Amphetamine sulfate (Sigma Chemical Co., St. Louis, MO) and (s)-(-)-propranolol hydrochloride (Sigma) were each dissolved in normal saline to achieve concentrations of 2.5 and 2.0 mg/ml, respectively. Intraperitoneal (IP) injections consisting of normal saline, the amphetamine solution, or the propranolol solution were given at a volume of 1 ml/kg body wt. The final doses of amphetamine and propranolol, 2.5 mg/kg and 2 mg/kg, respectively, were chosen on the basis of their ability to stimulate corticosterone release (Kynch & Eisenberg, 1979) and block the shock-induced suppression of lymphocyte mitogenie activity (Cunnick et al., 1990), respectively. Subjects were momentarily removed from their home cages during each of two IP injections which were spaced exactly one-half hour apart. Saline/saline-treated animals (n = 6) received two saline injections; saline/amphetamine-treated animals (n = 6) received a saline injection followed 30 min later by an amphetamine injection; and propranoloYamphetamine-treated animals (n = 7) received a propranolol injection followed by an amphetamine injection. All animals (N = 19) were sacrificed exactly 1 h after their second IP injection. Because previous studies have shown that propranolol alone has no effect on mitogenic responsiveness (Cunnick et al., 1990) and corticosterone levels in home cage control rats (Kynch & Eisenberg, 1979), this group was not utilized in this study. Specimen collection. Each rat was sacrificed by cervical dislocation with a clamp, and within 1 min blood was collected from the abdominal aorta into a 5-ml heparinized syringe fitted with a 21-gauge needle. Approximately 1.5 ml of this blood was placed into an Eppendorf microcentrifuge tube and kept on ice for later corticosterone measurement. From the same animal, the spleen was then extracted and placed into a 15-ml polypropylene centrifuge tube, containing RPMI1640 media supplemented with 10 mM Hepes, 2 mM glutamine, and 50 p,g gentamicimml . Lymphocyte mitogenic assay. Spleens were dissociated into a single cell suspension using sterile frosted glass slides. Splenic leukocytes were counted using a Coulter counter (Model ZBI) and diluted to 5 x lo6 leukocytes/ml in supplemented RPMI-1640 with 10% fetal calf serum. Leukocytes per milliliter of whole blood were determined using a Unopette and hemocytometer. Mitogen stimulation assays were performed as previously described (Lysle, Lyte, Fowler, & Rabin, 1987) on whole spleen suspensions and whole blood (diluted 1:lO with supplemented media containing heparin, 5 units/ml). Phytohemagglutinin (PHA) and concanavalin A (Con A) were used at concentrations of 1 and 10 pg!ml. The spleen cultures were pulsed with 1 &i [3H]thymidine (sp act 6.7 CilmM; DupontNew England Nuclear) in 50 ~1 of supplemented RPMI-1640 during the last 5 h of a 48-h incubation. The cultures were harvested onto glass filter paper using a Skatron cell harvester (Skatron, Inc.), and the incorporation of [3H]thymidine was determined with a liquid scintillation counter (Packard Model 1500) and was expressed as counts per minute (cpm). The blood leukocyte cultures were incubated for a total of 96 h and were pulsed with [3H]thymidine (1 &i/well) during the last 18 h of the incubation. The blood cultures were harvested and counted in the same manner as the spleen cultures. The average cpm/well for the blood cultures were normalized to lo5 leukocytes/well. Plasma corticosterone assay. The samples of heparinized blood collected and maintained on ice were centrifuged at 12,000 rpm for 3 min in a microcentrifuge, and 0.5 ml of plasma was removed and frozen at -70°C. Plasma corticosterone
296
PEZZONE
ET AL.
concentrations were determined by a competitive protein-binding radioassay which is sensitive to 0.2 &dl (Murphy, 1967). Data analysis. The data presented in this study were obtained from two identical experiments, each with three to four rats per group. As the results of the two studies were the same, they were pooled for statistical analysis. Raw data were analyzed by a one-way analysis of variance (ANOVA) using a statistical package (CLR ANOVA). In the event of a significant main effect, the group means were subjected to multiple post hoc comparisons using the Neuman-Keuls test. RESULTS Plusma corticosterone. The injection
of amphetamine, with or without a preceding injection of propranolol, produced a marked elevation of plasma corticosterone that did not differ between the saline/amphetamine and the propranolol/ amphetamine groups (Table 1). ANOVA of the corticosterone data revealed a significant main effect [F(2,18) = 22.556, p < O.OOl]. Post hoc multiple comparisons of the group means revealed that the saline/amphetamine and the propranolol/amphetamine groups were signilicantly higher than the saline/saline group (p < 0.01); however, the saline/amphetamine and the propranolol/amphetamine groups did not differ from each other. Thus, the elevation of corticosterone by amphetamine is not blocked by the p-adrenergic receptor antagonist. Splenic lymphocytes. Table 2 shows the mean cpm and statistical analysis of spleen lymphocyte responses to optimum concentrations of Con A and PHA. It should be noted that all the data presented in Tables 1 and 2 were collected from the same rats. The saline/amphetamine-injected animals showed a profound suppression of mitogenic activity in comparison to the saline/saline-injected animals. When the animals were treated with propranolol 30 min before the amphetamine injection, the spleen mitogenic activity was maintained at the levels found in animals that were saline/saline injected (Table 2). Thus, there is no decrease or functional alteration of mitogenic function in the spleen lymphocytes of the propranolol/amphetamine injected rats, even though there is a markedly elevated plasma corticosterone concentration (Table 1). Peripheral blood lymphocytes. The decreased mitogenic response that occurred in the saline/amphetamine-injected animals in comparison to the saline/salineinjected animals indicates that amphetamine is capable of inducing suppression of peripheral blood mitogenic activity (Table 2); however, when the animals were treated with propranolol prior to the amphetamine injections there was only a partial restoration of peripheral blood mitogenic activity for Con A, despite a complete restoration for PHA. Comparing the cpms of the propranolol/ amphetamine-injected animals to those of the saline/saline-injected animals indicates that approximately 50% of the Con A suppression is due to a mechanism involving the p-adrenergic receptor. There may be an additive effect of both TABLE 1 Plasma Corticosterone Concentrations Saline/saline Saline/amphetamine PropranololIamphetamine
(*g/d1 )(n = 6/group) 1.2 2 2.1 (p&dl) 53.4 k 15.6 (p,g/dl) 38.4 k 20.2 (pg/dl) .-
CORTICOSTERONE-INDEPENDENT Effect of Amphetamine
IMMUNE
297
MODULATION
TABLE 2 and B-Adrenergic Blocking on Lymphocyte Mitogenic Function
Saline/saline (n = 6) Saline/amphetamine (n = 6) PropranoloVamphetamine (n = 7)
Saline/saline (n = 6) Saline/amphetamine (n = 6) Propranolollamphetamine (n = 7)
Spleen (10 p&ml Con A)”
Spleen (10 &ml PHA)O
165,058 * 35.140 25,241 2 12,016 170,627 t 45,802
303,020 ? 30,269 40,606 2 3 1,537 274,913 ? 54,838
Blood (10 &ml Con A)’
Blood (10 pg/ml PHA)*
186,793 2 23,077 61,421 f 26,261 114,121 2 33,917
43,720 + 7.953 10,672 i 9,770 37,730 i_ 16.871
n ANOVA of the spleen Con A and PHA data reveals a significant main effect [F(2.16) = 55.11, p < 0.001, and F(2,17) = 81.85, p < 0.0011, respectively. Post hoc tests for each mitogen show that the saline/saline and propranolol/amphetamine groups are significantly different than the saline/ amphetamine group @ < 0.01) and that the saline/saline and propranolol/amphetamine groups do not differ from each other (p > 0.05). ’ ANOVA of the blood Con A and PHA data reveals a significant main effect [F(2,14) = 27.54, p < 0.001, and F(2,14) = 11.34, p < O.OOl], respectively. Post hoc tests for PHA show that the saline/saline and propranolol/amphetamine groups are significantly different from the saline/ amphetamine group (p < 0.01) but not different from each other @ > 0.05). Post hoc tests for Con A reveal that all three groups are significantly different from each other (p < 0.01).
catechoiamines and corticosterone in suppressing the responsiveness eral blood lymphocytes to Con A.
of periph-
DISCUSSION
This study investigated the effects of amphetamine on immune function, since many of the neurochemical, physiological, and behavioral responses to acute amphetamine administration mimic those elicited by stress (Antelman & Chiodo, 1983). Interestingly, amphetamine was found to produce a marked reduction of lymphocyte mitogenic activity in both the spleen and peripheral blood and a marked elevation of plasma corticosterone, responses identical to those elicited by footshock and other immune-modulating stimuli (Cunnick et al., 1990; Lysle et al., 1987; Rabin et al., 1987). The suppressive effect of amphetamine on immune function further enhances its classification as a stressor as previously described by Antelman and Chiodo (1983). Following pretreatment with propranolol, the amphetamine-induced suppression of splenic lymphocyte mitogenic activity was completely prevented, a finding comparable to that of Cunnick et al. (1990) who showed that footshock-induced suppression of splenic lymphocyte mitogenic activity was also ameliorated by propranolol. Furthermore, they concluded that this effect was mediated via peripheral P-adrenergic receptors, as the response was also attenuated by peripherally administered nadolol, a l3-adrenergic receptor antagonist that does not cross the blood-brain barrier. Because amphetamine and propranolol possess multiple pharmacological properties, determining exactly how they modulate immune function is complex; however, by primarily considering their sympathomimetic and sympatholytic activities, respectively, one can hypothesize that the sympathetic nervous system is mediating this response via innervation of lymphoid tissue and catecholamine release (Felten & Felten, 1991).
298
PEZZONE
ET
AL.
The amphetamine-induced suppression of spleen cell mitogenic reactivity to both Con A and PHA was completely abrogated by propranolol pretreatment, as was the peripheral blood lymphocyte (PBL) response to PHA; however, the amphetamine-induced suppression of the PBL mitogenic response to Con A was only partially ameliorated, suggesting that PHA-responsive and Con A-responsive lymphocytes in the peripheral blood may be affected by different mechanisms (Haynes & Fauci, 1978). Numerous hormones and neuropeptides have been implicated as possible mediators of stressor-induced immune alteration (Carr & Blalock, 1991). Because the amphetamine-induced elevation of plasma corticosterone was not attenuated by propranolol pretreatment, the presence of normal mitogenic activity in these animals indicates that this response is unlikely modulated by corticosterone. These results are at variance with published studies which suggest that the glucocorticoid receptor modulates lymphocyte function in response to nonspecific mitogenic stimulation (Miller, Spencer, Trestman, Kim, McEwen, & Stein, 1991). Overnight exposure to dexamethasone, the highly potent, synthetic glucocorticoid, may account for their observation, as it is possible that the acute release of the less potent, endogenous glucocorticoid, corticosterone, would not have provided similar results. These data, in conjunction with those previously cited in the introduction, support the view that corticosterone does not participate in the functional alteration of spleen lymphocytes and PHA-reactive peripheral blood lymphocytes, but it may participate in the altered function of Con A reactive PBLs. Alternatively, corticosterone may indeed be responsible for the observed alterations in immune function. Because glucocorticoids can up-regulate B-adrenergic receptors on lung tumor cell lines (Nakane, Szentendrei, Stern, Virmani, Seely, & Kunos, 1990), the same phenomenon may occur on lymphocytes which are known to possess B-adrenergic receptors (Loveland, Jarrot, & McKenzie, 1981). Consequently, the corticosterone released by amphetamine administration may enhance the effect of norepinephrine on lymphocyte mitogenic activity, which was previously shown to be suppressive (Goodwin, Messner, & Williams, 1979). Correspondingly, propranolol could then negate this effect. ACKNOWLEDGMENTS The authors thank Ms. Ada Armlield and Ms. Ellen Hamill for technical assistance. This work was supported in part by National Research Service Award MHl0157 to M.A.P. from the National Institute of Mental Health (NIMH) and by NIMH Research Grant MH43411 to B.S.R. Preliminary results from these studies were presented at the Research Perspectives in Psychoneuroimmunology III Conference in Columbus, Ohio, 1991.
REFERENCES Ader, R., & Cohen, N. (1985). CNS-immune
interactions: Conditioning phenomena. BehavioralBrain
Sci. 8, 379-394.
Antelman, S. M.. & Chiodo, L. A. (1983). Amphetamine as a stressor. In I. Creese (Ed.), Stimulants: Neurochemical, Behavioral, and Clinical Perspectives, pp. 269-299. Raven Press: New York. Bayer, B. M., Daussin, S., Hemandez, M., & Irvin, L. (1990). Morphine inhibition of lymphocyte activity is mediated by an opioid dependent mechanism. Neurophnrmacology 29, 369-374. Carr, D. J. J., & Blalock, J. E. (1991). Neuropeptide hormones and receptors common to the immune and neuroendocrine systems: Bidirectional pathway of intersystem communication. In R. Ader. D. Felten, & N. Cohen (Eds.), Psychoneuroimmunology, 2nd ed., pp. 573-588. Academic Press: New York.
CORTICOSTERONE-INDEPENDENT
IMMUNE
MODULATION
299
Chappell, P. B., Smith, M. A., Kilts, C. D., Bissette, G., Ritchie, J., Anderson, C., & Nemeroff, C. B. (1986). Alterations in corticotropin-releasing factor-like immunoreactivity in discrete rat brain regions after acute and chronic stress. J. Neurosci. 6, 29082914. Cunnick, J. E., Lysle, D. T., Kucinski, B. J., & Rabin, B. S. (1990). Evidence that shock-induced immune suppression is mediated by adrenal hormones and peripheral 8-adrenergic receptors. Pharmacol. Biochem. Behav. 36, 645-651. Cupps, T. R., & Fauci, A. S. (1982). Corticosteroid-mediated immunoregulation in man. Immunol. Rev. 65, 133-155. Felten, S. Y., & Felten, D. L. (1991). Innervation of lymphoid tissue. In R. Ader, D. Felten. & N. Cohen (Eds.), Psychoneuroimmunology, 2nd ed., pp. 2769. Academic Press: New York. Flores, C. M., Hemandez, M. C., Hargreaves, K. M., & Bayer, B. M. (1990). Restraint stressinduced elevations in plasma corticosterone and 8-endorphin are not accompanied by alterations in immune function. J. Neuroimmunol. 28, 21%225. Goodwin, J. S., Messner, R. P., & Williams, R. C. (1979). Inhibitors of T-cell mitogenesis: Effect of mitogen dose. Cell. Immunol. 45, 303-308. Haynes. B. F., & Fauci, A. S. (1978). The differential effect of in viva hydrocortisone on the kinetics of subpopulations of human peripheral blood thymus-derived lymphocytes. .I. C/in. Invest. 61,703-707. Jain, R.. Zwickler, D.. Hollander, C. S., Brand, H., Saperstein, A.. Hutchinson, B., Brown, C.. & Audhya, T. (1991). Corticotropin-releasing factor modulates the immune response to stress in the rat. Endocrinology 128, 1329-1336. Jessop, J. J., & Bayer, B. M. (1989). Time-dependent effects of isolation on lymphocyte and adrenocortical activity. .I. Neuroimmunol. 23, 143-147. Kavelaars, A., Ballieux, R. E., & Heijnen. C. J. (1990). B-Endorphin secretion by human peripheral blood mononuclear cells: Regulation by glucocortoids. Life Sci. 46, 1233-1240. Keller, S. E., Schliefer, S. J., Liotta, A. S., Bond, R. N., Farhoody, N., & Stein, M. (1988). Stressinduced alterations of immunity in hypophysectomized rats. Proc. Nat. Acad. Sci. USA 85, 9297-9301. Keller, S. E., Weiss, J. M., Schleifer, S. J., Miller, N. E., & Stein M. (1983). Stress-induced suppression of immunity in adrenalectomized rats. Science 221, 1301-1304. Knych, E. T., & Eisenberg, R. M. (1979). Effect of amphetamine on plasma corticosterone in the conscious rat. Neuroendacrinology 29, 110-I 18. Loveland, B. E., Jarrott, B., and McKenzie, I. F. (1981). The detection of beta adrenoceptors on murine lymphocytes. Int. .I. Zmmunopharmacol. 3, 45-55. Lysle, D. T., Lyte, M., Fowler, H., & Rabin, B. S. (1987). Shock-induced modulation of lymphocyte reactivity: Suppression, habituation. and recovery. Life Sci. 41, 1805-1814. Manuck, S. B., Cohen, S., Rabin, B. S., Muldoon, M. F., & Bachen, E. A. (1991). Individual differences in cellular immune response to stress. Psycho/. Sci. 2, 11 l-l 16. Miller, A. H., Spencer, R. L., Trestman, R. L., Kim, C., McEwen, B. S.. & Stein, M. (1991). Adrenal steroid receptor activation in vivo and immune function. Am. J. Physiol. 261, El26E131. Munck, A., & Guyre, P. M. (1991). Glucocorticoids and immune function. In R. Ader, D. Felten, & N. Cohen (Eds.), Psychoneuroimmunology, 2nd ed., pp. 447-474. Academic Press: New York. Murphy, B.E.P. (1967). Some studies of the protein-binding of steroids and their application to the routine micro and ultramicro measurement of various steroids in body fluids by competitive protein-binding radioassay. J. C&n. Endocrinoi. Metab. 27, 973-990. Nakane, T., Szentendrei, T., Stem, L.. Virmani, M., Seely, J., & Kunos, G. (1990). Effects of IL-! and cortisol on 8-adrenergic receptors, cell proliferation, and differentiation in cultured human A549 lung tumor cells. J. Zmmunol. 145, 260-266. Rabin, B. S., Lyte. M., & Hamill. E. (1987). The influence of mouse strain and housing on the immune response. J. Neuroimmunol. 17, 11-16. Roudebush, R. E., & Bryant, H. U. (1991). Conditioned immunosuppression of a murine delayed type hypersensitivity response: Dissociation from corticosterone elevation. Brain Behav. Immun. 5, 308317. Sundar, S. K., Cierpial, M. A., Kilts, C., Ritchie, J. C., & Weiss, J. M. (1990). Brain IL-l-induced immunosuppression occurs through activation of both pituitary-adrenal axis and sympathetic nervous system by corticotropin releasing factor. J. Neurosci. 10, 3701-3706. Turk, J. L. (1975). Immunological unresponsiveness. In A. Neuberger & E. L. Tatum (Eds.), Delayed Hypersensitivity, 2nd ed.. pp. 129-180. North-Holland: Amsterdam. Received December 11, 1991
BEHAVIOR,
AND
IMMUNITY
6, 293-299
(1992)
Corticosterone-Independent Mitogenic Function MICHAEL
Alteration of Lymphocyte by Amphetamine
A. PEZZONE, KELLY A. RUSH, ALEXANDER W. KUSNECOV, PAUL G. WOOD, AND BRUCE S. RABIN
The Brain, Behavior and Immunity Center and The Division of Clinical Immunopathology, Department of Pathology, Universiry of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania 15213 Amphetamine, a neural stimulatory agent with acute effects mimicking those of stress, is shown here to elevate plasma corticosterone levels and suppress spleen and peripheral blood lymphocyte (PBL) mitogenic responses to concanavalin A (Con A) and phytohemagglutinin (PHA) when administered to rats. Pretreatment of the rats with propranolol, a nonselective P-adrenergic receptor antagonist, totally prevented the amphetamineinduced suppression of lymphocyte mitogenic reactivity to Con A and PHA in the spleen and to PHA in the peripheral blood; however, the PBL mitogenic response to Con A was only partially restored. Although the amphetamine-induced alterations in immune function were prevented by propranolol pretreatment, the elevated plasma corticosterone response was not. This suggests that corticosterone is not modulating the mitogenic activity of splenic lymphocytes or PHA-reactive PBLs. On the other hand, Con A-reactive PBLs may be affected by corticosterone and/or other mechanisms, which may D 1992 Academic press. IIIC. include the catecholamines.
Adrenal glucocorticoids become elevated in plasma following exposure of animal and human subjects to a variety of stressors (Munck & Guyre, 1991). Glucocorticoids, which are known to exert a suppressive effect on lymphocyte function (Cupps & Fauci, 1982), are utilized therapeutically to help prevent allograft rejection (Turk, 1975). Thus, it seems reasonable to hypothesize that glucocorticoids may mediate stressor-induced suppression of immune function by affecting the viability of lymphoid cells and/or the production of mononuclear cell-derived cytokines (Munck & Guyre, 1991). Recently, the role of glucocorticoids in stressor-induced suppression of immune function has been questioned. In both rats and humans, stress has been found to elevate glucocorticoids in the absence of a concomitant alteration of lymphocyte mitogenic activity (Flores, Hernandez, Hargreaves, & Bayer, 1990; Jessop & Bayer, 1989; Manuck, Cohen, Rabin, Muldoon, & Bachen, 1991). Further, a shock-induced reduction of peripheral blood and spleen lymphocyte responses to nonspecific mitogens has been demonstrated in adrenalectomized rats (Keller, Weiss, Schleifer, Miller, & Stein, 1983). The injection of rats with morphine causes lymphocyte functional alterations along with an elevation of corticosteronc (Bayer, Daussin, Hernandez, & Irvin, 1990); however, the lymphocyte functional changes are completely reversed by naltrexone, while the corticosterone elevation persists. Environmental situations, such as the number of mice housed in a cage, also influence the responsiveness of lymphocytes to mitogenic stimulation by a corticosterone-independent mechanism (Rabin, Lyte, & Hamill, 1987). Finally, the role of corticosterone in classically conditioned immune modulation has been questioned (Ader & Cohen, 1985) with recent evidence indicating a disso293 0889-1591192
$5.00
Copyright Q I%? by Academic Press, Inc. All rights of reproduction in any form reserved.
294
PEZZONE
ET AL.
ciation between corticosterone elevation and immune suppression in mice conditioned with cyclophosphamide (Roudebush & Bryant, 1991). The hypothalamic-pituitary-adrenal (HPA) axis may participate in stressorinduced immune alterations without involvement of its adrenocorticoid component. An adrenal-independent role for corticotropin-releasing factor (CRF) in stress-induced immune alterations has recently been suggested (Jain, Zwickler, Hollander, Brand, Saperstein, Hutchinson, Brown, & Audhya, 1991). By treating rats with CRF antisera or a CRF antagonist, the stress-induced suppression of lymphocyte mitogenic activity was similarly ameliorated in both adrenalectomized and intact rats. CRF, which is present throughout the rat brain (Chappell Smith, Kilts Bissette, Ritchie, Anderson, & Nemeroff, 1986), may function as a mediator of sympathetic nervous system activation during stress (Munck & Guyre, 1991). This role of CRF is supported by a study in which interleukin-I (IL-l) infusion into rat brains (Sundar, Cierpial, Kilts, Ritchie, & Weiss, 1990) decreased immune function by a mechanism involving CRF activation of the sympathetic nervous system. Alternatively, CRF may induce production of IL-l by monocytes, which may subsequently induce B-endorphin production from B lymphocytes (Kavelaars, Ballieux, & Heijnen, 1990). B-Endorphin may then alter the function of T lymphocytes. There is evidence which suggests that the HPA axis may counteract the suppressive influence of stress on lymphocyte function. Following a stressor, hypophysectomized rats have a more pronounced depression of lymphocyte mitogenie function than do intact rats (Keller, Schleifer, Liotta, Bond, Farhoody, & Stein, 1988). Consistent with this, our unpublished data show that lesions of the hypothalamic paraventricular nucleus in rats result in a more profound stressorinduced suppression of lymphocyte mitogenic function when compared to shamlesioned animals (Pezzone, manuscript in preparation). The immunomodulating potential of acutely administered amphetamine was investigated in this study, as many of its neurochemical, physiological, and behavioral effects are identical to those elicited by various stressors (Antelman & Chiodo, 1983). Furthermore, because the P-adrenergic receptor antagonist propranolol prevents the suppression of lymphocyte mitogenic activity induced by a footshock stressor (Cunnick, Lysle, Kucinski, & Rabin, 1990), its ability to mod.ify the potential immunomodulating effects of amphetamine was also investigated. We have also sought to provide additional information regarding the role of glucocorticoids in the functional alteration of lymphocyte activity in amphetaminetreated rats, since acutely administered amphetamine elevates corticosterone levels independent of propranolol administration (Knych & Eisenberg, 1979). METHODS
Experimental animals. Male rats of the Lewis strain, 65 days old and 250-300 g in weight, were purchased from Harlan Sprague Dawley (Indianapolis, IN). The animals were individually housed in hanging wire cages under a 12-h day-night cycle (lights on at 0700) of artificial illumination and ad lib access to food and water. Upon arrival in the animal housing facility, rats were given a 2-week acclimatization to the new surroundings. Furthermore, the animals were gently handled approximately 2 mm/day for 5 consecutive days during their first week in the new colony room. On the day prior to experimental manipulation all rats were weighed.
CORTICOSTERONE-INDEPENDENT
IMMUNE
MODULATION
295
Treatment groups. (D)-Amphetamine sulfate (Sigma Chemical Co., St. Louis, MO) and (s)-(-)-propranolol hydrochloride (Sigma) were each dissolved in normal saline to achieve concentrations of 2.5 and 2.0 mg/ml, respectively. Intraperitoneal (IP) injections consisting of normal saline, the amphetamine solution, or the propranolol solution were given at a volume of 1 ml/kg body wt. The final doses of amphetamine and propranolol, 2.5 mg/kg and 2 mg/kg, respectively, were chosen on the basis of their ability to stimulate corticosterone release (Kynch & Eisenberg, 1979) and block the shock-induced suppression of lymphocyte mitogenie activity (Cunnick et al., 1990), respectively. Subjects were momentarily removed from their home cages during each of two IP injections which were spaced exactly one-half hour apart. Saline/saline-treated animals (n = 6) received two saline injections; saline/amphetamine-treated animals (n = 6) received a saline injection followed 30 min later by an amphetamine injection; and propranoloYamphetamine-treated animals (n = 7) received a propranolol injection followed by an amphetamine injection. All animals (N = 19) were sacrificed exactly 1 h after their second IP injection. Because previous studies have shown that propranolol alone has no effect on mitogenic responsiveness (Cunnick et al., 1990) and corticosterone levels in home cage control rats (Kynch & Eisenberg, 1979), this group was not utilized in this study. Specimen collection. Each rat was sacrificed by cervical dislocation with a clamp, and within 1 min blood was collected from the abdominal aorta into a 5-ml heparinized syringe fitted with a 21-gauge needle. Approximately 1.5 ml of this blood was placed into an Eppendorf microcentrifuge tube and kept on ice for later corticosterone measurement. From the same animal, the spleen was then extracted and placed into a 15-ml polypropylene centrifuge tube, containing RPMI1640 media supplemented with 10 mM Hepes, 2 mM glutamine, and 50 p,g gentamicimml . Lymphocyte mitogenic assay. Spleens were dissociated into a single cell suspension using sterile frosted glass slides. Splenic leukocytes were counted using a Coulter counter (Model ZBI) and diluted to 5 x lo6 leukocytes/ml in supplemented RPMI-1640 with 10% fetal calf serum. Leukocytes per milliliter of whole blood were determined using a Unopette and hemocytometer. Mitogen stimulation assays were performed as previously described (Lysle, Lyte, Fowler, & Rabin, 1987) on whole spleen suspensions and whole blood (diluted 1:lO with supplemented media containing heparin, 5 units/ml). Phytohemagglutinin (PHA) and concanavalin A (Con A) were used at concentrations of 1 and 10 pg!ml. The spleen cultures were pulsed with 1 &i [3H]thymidine (sp act 6.7 CilmM; DupontNew England Nuclear) in 50 ~1 of supplemented RPMI-1640 during the last 5 h of a 48-h incubation. The cultures were harvested onto glass filter paper using a Skatron cell harvester (Skatron, Inc.), and the incorporation of [3H]thymidine was determined with a liquid scintillation counter (Packard Model 1500) and was expressed as counts per minute (cpm). The blood leukocyte cultures were incubated for a total of 96 h and were pulsed with [3H]thymidine (1 &i/well) during the last 18 h of the incubation. The blood cultures were harvested and counted in the same manner as the spleen cultures. The average cpm/well for the blood cultures were normalized to lo5 leukocytes/well. Plasma corticosterone assay. The samples of heparinized blood collected and maintained on ice were centrifuged at 12,000 rpm for 3 min in a microcentrifuge, and 0.5 ml of plasma was removed and frozen at -70°C. Plasma corticosterone
296
PEZZONE
ET AL.
concentrations were determined by a competitive protein-binding radioassay which is sensitive to 0.2 &dl (Murphy, 1967). Data analysis. The data presented in this study were obtained from two identical experiments, each with three to four rats per group. As the results of the two studies were the same, they were pooled for statistical analysis. Raw data were analyzed by a one-way analysis of variance (ANOVA) using a statistical package (CLR ANOVA). In the event of a significant main effect, the group means were subjected to multiple post hoc comparisons using the Neuman-Keuls test. RESULTS Plusma corticosterone. The injection
of amphetamine, with or without a preceding injection of propranolol, produced a marked elevation of plasma corticosterone that did not differ between the saline/amphetamine and the propranolol/ amphetamine groups (Table 1). ANOVA of the corticosterone data revealed a significant main effect [F(2,18) = 22.556, p < O.OOl]. Post hoc multiple comparisons of the group means revealed that the saline/amphetamine and the propranolol/amphetamine groups were signilicantly higher than the saline/saline group (p < 0.01); however, the saline/amphetamine and the propranolol/amphetamine groups did not differ from each other. Thus, the elevation of corticosterone by amphetamine is not blocked by the p-adrenergic receptor antagonist. Splenic lymphocytes. Table 2 shows the mean cpm and statistical analysis of spleen lymphocyte responses to optimum concentrations of Con A and PHA. It should be noted that all the data presented in Tables 1 and 2 were collected from the same rats. The saline/amphetamine-injected animals showed a profound suppression of mitogenic activity in comparison to the saline/saline-injected animals. When the animals were treated with propranolol 30 min before the amphetamine injection, the spleen mitogenic activity was maintained at the levels found in animals that were saline/saline injected (Table 2). Thus, there is no decrease or functional alteration of mitogenic function in the spleen lymphocytes of the propranolol/amphetamine injected rats, even though there is a markedly elevated plasma corticosterone concentration (Table 1). Peripheral blood lymphocytes. The decreased mitogenic response that occurred in the saline/amphetamine-injected animals in comparison to the saline/salineinjected animals indicates that amphetamine is capable of inducing suppression of peripheral blood mitogenic activity (Table 2); however, when the animals were treated with propranolol prior to the amphetamine injections there was only a partial restoration of peripheral blood mitogenic activity for Con A, despite a complete restoration for PHA. Comparing the cpms of the propranolol/ amphetamine-injected animals to those of the saline/saline-injected animals indicates that approximately 50% of the Con A suppression is due to a mechanism involving the p-adrenergic receptor. There may be an additive effect of both TABLE 1 Plasma Corticosterone Concentrations Saline/saline Saline/amphetamine PropranololIamphetamine
(*g/d1 )(n = 6/group) 1.2 2 2.1 (p&dl) 53.4 k 15.6 (p,g/dl) 38.4 k 20.2 (pg/dl) .-
CORTICOSTERONE-INDEPENDENT Effect of Amphetamine
IMMUNE
297
MODULATION
TABLE 2 and B-Adrenergic Blocking on Lymphocyte Mitogenic Function
Saline/saline (n = 6) Saline/amphetamine (n = 6) PropranoloVamphetamine (n = 7)
Saline/saline (n = 6) Saline/amphetamine (n = 6) Propranolollamphetamine (n = 7)
Spleen (10 p&ml Con A)”
Spleen (10 &ml PHA)O
165,058 * 35.140 25,241 2 12,016 170,627 t 45,802
303,020 ? 30,269 40,606 2 3 1,537 274,913 ? 54,838
Blood (10 &ml Con A)’
Blood (10 pg/ml PHA)*
186,793 2 23,077 61,421 f 26,261 114,121 2 33,917
43,720 + 7.953 10,672 i 9,770 37,730 i_ 16.871
n ANOVA of the spleen Con A and PHA data reveals a significant main effect [F(2.16) = 55.11, p < 0.001, and F(2,17) = 81.85, p < 0.0011, respectively. Post hoc tests for each mitogen show that the saline/saline and propranolol/amphetamine groups are significantly different than the saline/ amphetamine group @ < 0.01) and that the saline/saline and propranolol/amphetamine groups do not differ from each other (p > 0.05). ’ ANOVA of the blood Con A and PHA data reveals a significant main effect [F(2,14) = 27.54, p < 0.001, and F(2,14) = 11.34, p < O.OOl], respectively. Post hoc tests for PHA show that the saline/saline and propranolol/amphetamine groups are significantly different from the saline/ amphetamine group (p < 0.01) but not different from each other @ > 0.05). Post hoc tests for Con A reveal that all three groups are significantly different from each other (p < 0.01).
catechoiamines and corticosterone in suppressing the responsiveness eral blood lymphocytes to Con A.
of periph-
DISCUSSION
This study investigated the effects of amphetamine on immune function, since many of the neurochemical, physiological, and behavioral responses to acute amphetamine administration mimic those elicited by stress (Antelman & Chiodo, 1983). Interestingly, amphetamine was found to produce a marked reduction of lymphocyte mitogenic activity in both the spleen and peripheral blood and a marked elevation of plasma corticosterone, responses identical to those elicited by footshock and other immune-modulating stimuli (Cunnick et al., 1990; Lysle et al., 1987; Rabin et al., 1987). The suppressive effect of amphetamine on immune function further enhances its classification as a stressor as previously described by Antelman and Chiodo (1983). Following pretreatment with propranolol, the amphetamine-induced suppression of splenic lymphocyte mitogenic activity was completely prevented, a finding comparable to that of Cunnick et al. (1990) who showed that footshock-induced suppression of splenic lymphocyte mitogenic activity was also ameliorated by propranolol. Furthermore, they concluded that this effect was mediated via peripheral P-adrenergic receptors, as the response was also attenuated by peripherally administered nadolol, a l3-adrenergic receptor antagonist that does not cross the blood-brain barrier. Because amphetamine and propranolol possess multiple pharmacological properties, determining exactly how they modulate immune function is complex; however, by primarily considering their sympathomimetic and sympatholytic activities, respectively, one can hypothesize that the sympathetic nervous system is mediating this response via innervation of lymphoid tissue and catecholamine release (Felten & Felten, 1991).
298
PEZZONE
ET
AL.
The amphetamine-induced suppression of spleen cell mitogenic reactivity to both Con A and PHA was completely abrogated by propranolol pretreatment, as was the peripheral blood lymphocyte (PBL) response to PHA; however, the amphetamine-induced suppression of the PBL mitogenic response to Con A was only partially ameliorated, suggesting that PHA-responsive and Con A-responsive lymphocytes in the peripheral blood may be affected by different mechanisms (Haynes & Fauci, 1978). Numerous hormones and neuropeptides have been implicated as possible mediators of stressor-induced immune alteration (Carr & Blalock, 1991). Because the amphetamine-induced elevation of plasma corticosterone was not attenuated by propranolol pretreatment, the presence of normal mitogenic activity in these animals indicates that this response is unlikely modulated by corticosterone. These results are at variance with published studies which suggest that the glucocorticoid receptor modulates lymphocyte function in response to nonspecific mitogenic stimulation (Miller, Spencer, Trestman, Kim, McEwen, & Stein, 1991). Overnight exposure to dexamethasone, the highly potent, synthetic glucocorticoid, may account for their observation, as it is possible that the acute release of the less potent, endogenous glucocorticoid, corticosterone, would not have provided similar results. These data, in conjunction with those previously cited in the introduction, support the view that corticosterone does not participate in the functional alteration of spleen lymphocytes and PHA-reactive peripheral blood lymphocytes, but it may participate in the altered function of Con A reactive PBLs. Alternatively, corticosterone may indeed be responsible for the observed alterations in immune function. Because glucocorticoids can up-regulate B-adrenergic receptors on lung tumor cell lines (Nakane, Szentendrei, Stern, Virmani, Seely, & Kunos, 1990), the same phenomenon may occur on lymphocytes which are known to possess B-adrenergic receptors (Loveland, Jarrot, & McKenzie, 1981). Consequently, the corticosterone released by amphetamine administration may enhance the effect of norepinephrine on lymphocyte mitogenic activity, which was previously shown to be suppressive (Goodwin, Messner, & Williams, 1979). Correspondingly, propranolol could then negate this effect. ACKNOWLEDGMENTS The authors thank Ms. Ada Armlield and Ms. Ellen Hamill for technical assistance. This work was supported in part by National Research Service Award MHl0157 to M.A.P. from the National Institute of Mental Health (NIMH) and by NIMH Research Grant MH43411 to B.S.R. Preliminary results from these studies were presented at the Research Perspectives in Psychoneuroimmunology III Conference in Columbus, Ohio, 1991.
REFERENCES Ader, R., & Cohen, N. (1985). CNS-immune
interactions: Conditioning phenomena. BehavioralBrain
Sci. 8, 379-394.
Antelman, S. M.. & Chiodo, L. A. (1983). Amphetamine as a stressor. In I. Creese (Ed.), Stimulants: Neurochemical, Behavioral, and Clinical Perspectives, pp. 269-299. Raven Press: New York. Bayer, B. M., Daussin, S., Hemandez, M., & Irvin, L. (1990). Morphine inhibition of lymphocyte activity is mediated by an opioid dependent mechanism. Neurophnrmacology 29, 369-374. Carr, D. J. J., & Blalock, J. E. (1991). Neuropeptide hormones and receptors common to the immune and neuroendocrine systems: Bidirectional pathway of intersystem communication. In R. Ader. D. Felten, & N. Cohen (Eds.), Psychoneuroimmunology, 2nd ed., pp. 573-588. Academic Press: New York.
CORTICOSTERONE-INDEPENDENT
IMMUNE
MODULATION
299
Chappell, P. B., Smith, M. A., Kilts, C. D., Bissette, G., Ritchie, J., Anderson, C., & Nemeroff, C. B. (1986). Alterations in corticotropin-releasing factor-like immunoreactivity in discrete rat brain regions after acute and chronic stress. J. Neurosci. 6, 29082914. Cunnick, J. E., Lysle, D. T., Kucinski, B. J., & Rabin, B. S. (1990). Evidence that shock-induced immune suppression is mediated by adrenal hormones and peripheral 8-adrenergic receptors. Pharmacol. Biochem. Behav. 36, 645-651. Cupps, T. R., & Fauci, A. S. (1982). Corticosteroid-mediated immunoregulation in man. Immunol. Rev. 65, 133-155. Felten, S. Y., & Felten, D. L. (1991). Innervation of lymphoid tissue. In R. Ader, D. Felten. & N. Cohen (Eds.), Psychoneuroimmunology, 2nd ed., pp. 2769. Academic Press: New York. Flores, C. M., Hemandez, M. C., Hargreaves, K. M., & Bayer, B. M. (1990). Restraint stressinduced elevations in plasma corticosterone and 8-endorphin are not accompanied by alterations in immune function. J. Neuroimmunol. 28, 21%225. Goodwin, J. S., Messner, R. P., & Williams, R. C. (1979). Inhibitors of T-cell mitogenesis: Effect of mitogen dose. Cell. Immunol. 45, 303-308. Haynes. B. F., & Fauci, A. S. (1978). The differential effect of in viva hydrocortisone on the kinetics of subpopulations of human peripheral blood thymus-derived lymphocytes. .I. C/in. Invest. 61,703-707. Jain, R.. Zwickler, D.. Hollander, C. S., Brand, H., Saperstein, A.. Hutchinson, B., Brown, C.. & Audhya, T. (1991). Corticotropin-releasing factor modulates the immune response to stress in the rat. Endocrinology 128, 1329-1336. Jessop, J. J., & Bayer, B. M. (1989). Time-dependent effects of isolation on lymphocyte and adrenocortical activity. .I. Neuroimmunol. 23, 143-147. Kavelaars, A., Ballieux, R. E., & Heijnen. C. J. (1990). B-Endorphin secretion by human peripheral blood mononuclear cells: Regulation by glucocortoids. Life Sci. 46, 1233-1240. Keller, S. E., Schliefer, S. J., Liotta, A. S., Bond, R. N., Farhoody, N., & Stein, M. (1988). Stressinduced alterations of immunity in hypophysectomized rats. Proc. Nat. Acad. Sci. USA 85, 9297-9301. Keller, S. E., Weiss, J. M., Schleifer, S. J., Miller, N. E., & Stein M. (1983). Stress-induced suppression of immunity in adrenalectomized rats. Science 221, 1301-1304. Knych, E. T., & Eisenberg, R. M. (1979). Effect of amphetamine on plasma corticosterone in the conscious rat. Neuroendacrinology 29, 110-I 18. Loveland, B. E., Jarrott, B., and McKenzie, I. F. (1981). The detection of beta adrenoceptors on murine lymphocytes. Int. .I. Zmmunopharmacol. 3, 45-55. Lysle, D. T., Lyte, M., Fowler, H., & Rabin, B. S. (1987). Shock-induced modulation of lymphocyte reactivity: Suppression, habituation. and recovery. Life Sci. 41, 1805-1814. Manuck, S. B., Cohen, S., Rabin, B. S., Muldoon, M. F., & Bachen, E. A. (1991). Individual differences in cellular immune response to stress. Psycho/. Sci. 2, 11 l-l 16. Miller, A. H., Spencer, R. L., Trestman, R. L., Kim, C., McEwen, B. S.. & Stein, M. (1991). Adrenal steroid receptor activation in vivo and immune function. Am. J. Physiol. 261, El26E131. Munck, A., & Guyre, P. M. (1991). Glucocorticoids and immune function. In R. Ader, D. Felten, & N. Cohen (Eds.), Psychoneuroimmunology, 2nd ed., pp. 447-474. Academic Press: New York. Murphy, B.E.P. (1967). Some studies of the protein-binding of steroids and their application to the routine micro and ultramicro measurement of various steroids in body fluids by competitive protein-binding radioassay. J. C&n. Endocrinoi. Metab. 27, 973-990. Nakane, T., Szentendrei, T., Stem, L.. Virmani, M., Seely, J., & Kunos, G. (1990). Effects of IL-! and cortisol on 8-adrenergic receptors, cell proliferation, and differentiation in cultured human A549 lung tumor cells. J. Zmmunol. 145, 260-266. Rabin, B. S., Lyte. M., & Hamill. E. (1987). The influence of mouse strain and housing on the immune response. J. Neuroimmunol. 17, 11-16. Roudebush, R. E., & Bryant, H. U. (1991). Conditioned immunosuppression of a murine delayed type hypersensitivity response: Dissociation from corticosterone elevation. Brain Behav. Immun. 5, 308317. Sundar, S. K., Cierpial, M. A., Kilts, C., Ritchie, J. C., & Weiss, J. M. (1990). Brain IL-l-induced immunosuppression occurs through activation of both pituitary-adrenal axis and sympathetic nervous system by corticotropin releasing factor. J. Neurosci. 10, 3701-3706. Turk, J. L. (1975). Immunological unresponsiveness. In A. Neuberger & E. L. Tatum (Eds.), Delayed Hypersensitivity, 2nd ed.. pp. 129-180. North-Holland: Amsterdam. Received December 11, 1991