- Email: [email protected]
Immunosuppression in mice induced by cold water stress
BRAIN,
BEHAVIOR,
AND
4, 278-291 (1990)
IMMUNITY
Immunosuppression
in Mice Induced by Cold Water Stress
CHENG GAN JIANG, JULIE L. MORROW-TESCH, DAVID I. BELLER, ELINOR M. LEVY AND PAUL H. BLACK Department of Microbiology,
Boston University School of Medicine, 80 East Concord Street, Boston, Massachusetts 02118
A number of studies indicate that stress can result in suppression of the immune system in animals and man. Most of the studies have focused on alterations of lymphocyte function while only a few have investigated alterations of macrophage function or macrophage cytokine production. Macrophages play an essential role in homeostasis of the immune response. Indeed, the earliest events of the immune response occur in cells of the monocytic lineage, and their secretion of various cytokines may have both immunological and nonimmunological effects. The present studies were undertaken to determine whether alterations in macrophage physiology occur in mice subjected to a stress stimulus. Our studies in mice exposed to cold water stress for 4 days indicated reduced numbers of thymocytes and splenocytes, decreased T-cell blastogenesis, and reduced NK activity. Examination of elicited peritoneal macrophages from stressed mice revealed increased prostaglandin E, (PGE,) secretion and decreased immune region associated antigen (la) expression in response to interferon-y. Despite elevated PGE, levels, indomethacin was generally unable to restore depressed immune function. Of special interest was the finding that cell-associated and secreted interleukin 1 were significantly higher from unstimulated elicited macrophages from stressed mice. These results suggest that early in the response to stress, functions of a variety of cells within the immune system, especially macrophages, are altered and that dysregulated macrophage function may well contribute to the generalized suppression of the immune response in cold water stressed mice. 0 1990 Academic Press, Inc.
INTRODUCTION
Numerous investigations in both humans and animals have indicated that the immune system is affected by the response to stress (Dantzer & Kelley, 1989). It is known that psychological stress may influence the host’s response to infectious agents (Laudenslager, 1987). A number of investigations have documented physiological, hormonal, and immunological changes as a result of stress. The results of recent research indicate that stressful stimuli may depress immune functioning in humans and in animal models (Jesberger & Richardson, 1985; Levy & Black, 1988; Stein, Keller & Schleifer, 1985; Veldhuis, Croiset, Ballieux, Heijnen, & DeWied, 1987). Decreased T-cell mitogen response to concanavalin A (Con A)’ has been found ’ Abbreviations used: ACTH, adrenocorticotropic hormone; CNS, central nervous system; Con A, concanavalin A; DMEM, Dulbecco’s modified Eagle minimal essential medium; FITC, fluorescein isothiocyanate; FBS, fetal bovine serum; Ia, immune region associated antigen; IFN-y, interferon-y; IL-l, interleukin 1; IL-6, interleukin 6; INDO, indomethacin; LPS, Escherichia coli lipopolysaccharide; PEC, peritoneal exudate cells; PBS, sodium phosphate (10 mM) buffered saline; PGE,, prostaglandin E,; TNF, tumor necrosis factor; NK cell, natural killer cell; NP-40, nonidet P-40; SP, substance P. 278 0889-1591/90 $3.00 Copyright 6 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
IMMUNOSUPPRESSION
AND
COLD
WATER
STRESS
279
in recently bereaved widowers and hospitalized, severely depressed patients (Irwin, 1988; Kronfol et al., 1983; Schleifer, Keller, Camerino, Thornton, & Stein, 1983; Schleifer, Keller, Siris, Davis, & Stein, 1985; Stein, 1989). Significantly low levels of natural killer (NK) cell activity, decreased mitogenesis, low levels of IFN-y, and elevated levels of antibody to certain herpesviruses have been demonstrated in blood samples collected from medical students during examination periods (Kiecolt-Glaser, Garner, Speicher, Penn, Holliday, & Glaser, 1984; Workman & Via, 1987). In animal experiments, investigators have confirmed that lymphocyte proliferation (Keller, Weiss, Schleifer, Miller, & Stein, 1981; Weiss, Sundar, Becker, & Cierpial, 1989), antibody synthesis (Okimura, Satomi-Sasaki, & Ohkuma, 1986a), NK activity (Aarstad, Gardernack, & Seljelid, 1983; Okimura, Ogawa & Yamauchi, 1986b), and macrophage phagocytosis and tumoricidal function (Pavlidis & Chirigos, 1980) are impaired subsequent to various stressful stimuli. Accessory cells such as macrophages are essential for the induction of most immunological responses (Unanue, 1981; Unanue & Allen, 1987). The expression of class II MHC gene products (the Ia antigens in the mouse) and the processing of nominal antigen are essential for antigen presentation to and activation of helper T cells (Uhing 8z Adams, 1989). Macrophages further contribute to the regulation of immunologic responses by releasing various cytokines such as interleukin 1 (IL-l) and prostaglandins (Bonta, Elliott, Tak, & Ben-Efraim, 1989; Dore-Duffy, Guha, Rothman, & Zurier, 1988; Goodwin et al., 1981; Goodwin & Webb, 1980; Taffet & Russell, 1981). The aim of the studies reported here was to examine certain important immune functions subsequent to a stressful stimulus focusing, in particular, on the role of macrophages. MATERIALS
AND METHODS
Animals Male C57BL/6J mice were obtained from Jackson Laboratory (Bar Harbor, ME). Animals were housed, four or five per cage, under pathogen-free conditions on a 12-h light/dark cycle (lights off at 6 PM) and were provided commercial rodent chow and water ad libitum. These mice were used between 6 and 8 weeks of age as the source of splenocytes, thymocytes, and peritoneal exudate cells. Animals were allowed a minimum of 1 week acclimatization after arrival at our animal facility before experiments were conducted. No differences in the immunological parameters we have utilized, in response to stress, have been detected between animals acclimatized for 1 week or longer. Stress Procedure Cold water stress was induced as described with some modifications (Aarstad et al., 1983). The stress paradigm was administered twice daily (9 AM and 3 PM), for a duration of 1 min each time. Control and stressed treatment groups contained two to five and four to five mice, respectively. Control animals were transported from the animal room to the laboratory along with stressed mice. Test mice were
280
CHENG
ET
AL.
treated for 4 consecutive days. Briefly, mice were placed, one at a time, in a 4-liter glass container filled with 1 liter of cold water (4”C), and watched continuously for the entire time during which they were in the cold water; the water was deep enough to cover the back. At the end of each test, wet mice were dried with paper towels. This stress paradigm represents a complex interaction of physical and psychological stressors including anxiety, hypothermia, and exercise and was taken as the cummulative effects of all of these. Mice were injected with 1.5 ml of 10% proteose peptone (endotoxin level ~2.4 rig/ml as determined by the Limulus amebocyte assay) intraperitoneally 3 days prior to sacrifice to elicit peritoneal macrophages (after the first day of stress). Animals were sacrificed by cervical dislocation 1 day after the last stress session. All experiments were conducted three times unless otherwise stated. Preparation
of Immune
Cells
All media used to prepare immune cells were low in endotoxin as indicated by the manufacturer (FCS < 1 ng endotoxin/ml and medium co.05 &ml as tested in the Limulus Amoebocyte Lysate assay). Peritoneal exudate cells were obtained by lavage (Cowing, Schwartz, & Dickler, 1978). Briefly, the ventral surface was soaked with 70% ethanol and using two sterile forceps, the abdominal skin was pulled simultaneously in cephalad and caudad directions to expose the ventral surface of the parietal peritoneum. Ten milliliters of sterile, low endotoxin PBS (co.1 ng endotoximml, MA Bioproducts, Walkersville, MD) was injected into the peritoneal cavity using a syringe with a 22-gauge needle and the peritoneal cells were removed and placed immediately into centrifuge tubes on ice. The peritoneal cells were washed and centrifuged at 5OOg for 10 min at 4°C. Mononuclear cells were counted and the viability was determined by trypan blue exclusion. Cells harvested from individual animals were adjusted to lo6 mononuclear cells/ml in Dulbecco’s modified Eagle minimal essential medium (DMEM, Sigma Chemical Co., St. Louis, MO). After a 2-h incubation at 37°C in an humidified atmosphere of 5% CO,, nonadherent cells were removed by washing three times with DMEM. Spleens were removed aseptically into iced DMEM and cell suspensions were made by mincing the spleens using a sharp scissors and then thoroughly disrupting the fragments by aspiration several times through a 21-gauge needle. Debris was removed by passing through a stainless-steel wire mesh. The cells were washed twice and resuspended to 5 x 10’ cells/ml in red blood cell lysing buffer at 4°C for 5 min, in order to lyse red blood cells. The cells were washed twice in DMEM and resuspended to 10’ cells/ml. Thymocytes were prepared in a similar manner. Mitogen-Induced
Lymphocyte
Proliferation
The response of splenic lymphocytes to Con A stimulation was assessed as previously described (Evans, 1984). Briefly, lo6 cells were added to each of three wells on a 96-well, flat-bottomed, microtiter plate (Costar, Data Packaging Corp., Cambridge, MA) in a final volume of 0.2 ml with or without 2.5 kg/ml of Con A. This mitogen concentration and cell number have previously been shown in our laboratory to be optimal. INDO was dissolved in 95% ethanol to a concentration of lop2 M and then diluted in DMEM. The concentration of indomethacin (INDO)
IMMUNOSUPPRESSION
AND
COLD
WATER
STRESS
281
utilized was 5 x lo-’ M unless otherwise indicated. Cell cultures were incubated at 37°C for a total of 72 h. The splenic cells were pulsed with 1 PCi [3H]thymidine during the last 16 h of the incubation. Plates were then harvested with a multiwell cell harvester onto glass fiber filters and the counts per minute of radiolabeled DNA precursor incorporated by the cells was determined with a Beckman liquid scintillation counter. la Antigen Induction Macrophages harvested from individual animals by peritoneal lavage were seeded into chambers of a multichambered slide (Nunc, Inc., Naperville, IL) at a concentration of 2 x lo5 cells/chamber. To each chamber, either recombinant IFN-y (Amgen, Thousand Oaks, CA) or medium alone was added and the slides were incubated for 5 days at 37°C in an humidified 5% CO, atmosphere. The final concentration of IFN-y was 10 IU/ml unless otherwise stated. We have determined these conditions to be optimal for Ia induction by IFN-y in mouse peritoneal exudative cells in our lab (Poutsiaka, Schroder, Taylor, Levy & Black, 1985a; Poutsiaka, Taylor, Levy, & Black, 1985b). At the termination of the culture period, the percentage of macrophages expressing surface Ia antigen was determined (Steeg, Moore, Johnson, & Oppenheim, 1982). Approximately equal numbers of cells from stressed vs unstressed mice plated onto the slides as verified after 4 h by protein assay (Pierce, Rockford, IL) and at the end of the culture period. Macrophages were washed four times with PBS and fixed with 0.25% glutaraldehyde in PBS for 30 min at room temperature, and the slides were washed three times with PBS. Ia antigen was determined visually by immunoflourescence, using a monoclonal anti-I-Ab antibody (IgM) produced from the mouse hybridoma 25-g-38 (obtained from NIAID). The monoclonal antibody was added to each chamber, incubated for 1 h at 37°C and then washed three times with PBS. Fluorescein isothiocyanate (FITC)-conjugated rabbit anti-mouse immunoglobulin (DAK0 Corp., Santa Barbara, CA) was added to each chamber and the slides were incubated for an additional hour at 37°C and then washed three times with PBS. Coverslips were mounted on the chamber-slides and the slides were stored in the dark at 4°C. Each chamber was examined using a Nikon fluorescence microscope and 200 total cells were examined, with each treatment condition being assayed in duplicate. Ia positive cells were defined as those cells showing bright fluorescence. IL-1 Activity Assay The production of IL-1 by peritoneal macrophages was determined by proliferation using the DlO.G4.1 cell assay (Kaye et al., 1984; Struhar, Hat-beck, Gegen, Kawada, & Mason, 1989). For this assay, cells were pooled from all animals within a treatment group. Proteose peptone-elicited peritoneal macrophages (2 x lo4 to 1 x 105) were added to each well of a 96-well plate. Nonadherent cells were removed by gentle washing twice after a 2-h incubation at 37°C. Different concentrations of LPS (4 to 500 &ml; Sigma Chemical Co.) were added into cultures prepared with low endotoxin medium (co.05 ng endotoxin/ml). After 20 h of incubation, supernatants were collected and stored at - 70°C for determination of
282
CHENG
ET
AL.
secreted IL-l. Cells were also fixed with 1% paraformaldehyde for 10 min and then washed twice with PBS for assay of cell-associated IL-l. DlO.G4.1 cells, a clone of T cells which responds to IL-l in the presence of Con A and conalbumin, were utilized to assay IL-l. These cells were maintained using the supematant from rat splenocytes stimulated with Con A. For the secreted IL-I assay, each well of a 96-well tissue culture plate was seeded with 2 x lo4 cells in the presence or absence of spent media. The final concentration of Con A added was 2.5 l&ml. To determine cell-associated IL-l, DlO.G4.1 cells and Con A were added directly to wells containing fixed macrophages. These cells were incubated for 3 days with [3H]thymidine being added during the last 16 h. Proliferation was assayed by incorporation of [3H]thymidine into DNA. NK-cell
Cytotoxicity
Assay
YAC-1 cells, a Moloney virus-induced T-cell lymphoma, were used as target cells in the NK assay (Aarstad et al., 1983); they were maintained in RPM1 1640 medium supplemented with antibiotics (penicillin and streptomycin), L-glutamine, and 10% fetal bovine serum (FBS). Target cells were labeled with sodium chromate (“0; New England Nuclear, Boston, MA). Four effector-to-target cell ratios ranging from 100: 1 to 12.5: 1 were tested on microtiter plates with lo4 target cells per well. Medium was added to target cells for spontaneous lysis determinations, and 1% Nonidet P-40 (NP-40) was added for determination of 100% lysis. Plates were incubated for 4 h at 37°C in a humidified 5% CO, atmosphere. The supernatant fluid was harvested using a multiwell cell harvester for radioactive counting in a gamma counter. The percentage of specific 51Cr release is equal to cpm test sample - cpm medium control x 100. cpm NP-40 control - cpm medium control PGE, Determination PGE, concentrations in the culture supematants of peritoneal macrophages derived from normal and stressed mice were determined using a radioimmunoassay kit (Advanced Magnetics, Inc., Cambridge, MA.). Cells were incubated (37”C, 5% COJ with or without IFN-y at the final concentration of 10 IU/ml, unless otherwise stated, for 5 days. Supernatants were collected and stored at -70°C until assays were performed (Leizer et al., 1987). Statistical
Analysis
Immunological assays were performed on peritoneal macrophages and splenocytes from individual animals except for the IL-l assay when cells from mice within a treatment group were pooled. Significant differences between stressed and control mice were determined within each experiment by Student’s t test. Differences were considered to be significant when probability @) values <.05 were obtained. Within all experiments, statistically significant differences were consistent (I, < .05).
IMMUNOSUPPRESSION
AND
COLD
WATER
STRESS
283
RESULTS Enumeration
of Splenocytes
and Thymocytes
in Stressed
Mice
The impact of exposure to a stressful stimulus on spleen and thymus cell numbers was examined. In one experiment, the number of thymocytes was decreased in the stressed mice group (14.78 + 2.47 and 6.16 + 0.82 x IO’ cells for control and stressed mice, respectively; mean + SE; p < .Ol). The average decrease across three experiments was 59.6%, ranging from 50.1 to 67.1%. There was a concomitant decrease of splenic lymphocyte numbers in stressed mice (11.18 + 0.09 and 0.87 k 0.07 x lo* cells for control and stressed mice, respectively; p < .Ol). The average decrease was 58.6%, ranging from 19.9 to 92.9% across three experiments. Blastogenic
Responses
in Stressed
Mice
Mitogenesis in splenic lymphocytes induced by Con A was used to assess T-lymphocyte function. In one experiment, splenic lymphocytes from stressed mice exhibited decreased blastogenesis of approximately 44% (60.5 + 9.46 and 33.9 + 9.94 x lop3 cpm for control and stressed mice, respectively, mean 2 SE; p < .05). The average decrease across four experiments was 29.3%, ranging from 18.8 to 44%. Because PGs, produced largely by activated macrophages and monocytes, are inhibitors of mitogenesis, blastogenesis experiments were carried out in the presence of INDO, an inhibitor of PG production. INDO alone did not activate lymphocytes either in normal mice or in stressed mice. In normal splenocytes, the mitogenic response induced by Con A was not changed by the addition of INDO. With a stress exposure of 4 days, Con A induced blastogenesis decreased to 56% of control values and was not increased by the presence of INDO (33.9 + 9.94 and 37.98 k 2.07 x lop3 cpm for Con A alone or with INDO, respectively; p > .lO). Natural
Killer Activity
in Stressed
Mice
NK activity in spleen cells from cold water stressed mice was compared with those from control mice. NK activity of the test mice enduring cold water swim stress was reduced (p s .05) by at least 60% of controls at each effector:target ratio (percentage “Cr release, mean + SE; 31.2 + 2.0 vs 7.8 + 2.4 at lOO:l, 15.2 2 4.8 vs 5.9 2 2.7 at 50:1, 13.8 + 5.8 vs 1.9 + 0.8 at 25:l and 6.9 + 4.3 vs 1.7 + 0.6 at 12.5:1; control vs. stress, respectively). Although PGs can inhibit NK activity (Parhar & Lala, 1988), in the present experiments, NK activity of splenic lymphocytes derived from either stressed or normal mice was not increased significantly by the addition of INDO (data not shown). Increased
PGE, Synthesis
in Stressed
Mice
To explore some possible regulatory factors which may influence immune function, the production and release of PGE, were investigated. The baseline levels of PGE, produced by unactivated macrophages from stressed mice were higher than normal, control values, but were low compared to activated macrophages (Fig. 1). Macrophages from stressed mice activated with 10 IU/ml of IFN-y synthesized more PGE, than those from normal mice; the levels of PGE, in culture media
284
CHENG ET AL.
!
i+c hl
400 300
k
5 F B 5 s z
200
100 0
Normal
2 Days
4 Days
FIG. 1. PGE, concentration in culture supematants of peritoneal macrophages derived from normal or stressed mice. Open bars represent macrophages without the addition of INF-?I, and cross-hatched bars represent macrophages with the addition of INF-?I. Results are means 2 standard deviation. *Significantly different from the control group (p < .Ol).
increased an average of 253% (range between 163 and 351% in four separate experiments) when compared to levels obtained from macrophages from nonstressed mice. The synthesis of PGE, in peritoneal macrophages was dramatically increased in stressed mice in response to IFN-y, compared to macrophages obtained from nonstressed mice at all concentrations of IFN-y tested (Fig. 2). In macrophages from stressed mice, a dose-response relationship is evident. la Antigen Induction
in Stressed Mice
Although many investigations have examined immunological functions requiring T-cell help in stressed populations, the capacity of macrophages from stressed animals to synthesize and express Ia antigen, which is required for induction of T-cell help, has not been investigated. Elicited peritoneal macrophages from
1000 c x 55
LOO--
0 m
Normal Stressed
800 -700 --
1
600 --
E -6 400 -t-2 300 -8N -8 loo-a 500
200
0
r-l
0
10 IFN-gamma
25
I.7
50
100
(lU/ml)
FIG. 2. Dose responses of PGE, secretion by peritoneal macrophages in stressed mice. There were four mice in each group. Open bars represent control mice, and solid bars represent stressed mice.
IMMUNOSUPPRESSION
AND COLD WATER
STRESS
285
stressed and control mice were utilized to investigate Ia antigen expression. Peritoneal macrophages from stressed mice exhibited a significant inhibition of Ia antigen induction in response to IFN-y of approximately 58% in one experiment, Table 1 (range from four experiments of 49.8 to 69.7%). When the amount of IFN-y added to the culture was increased, there was a positive dose-dependent relationship between the amount of IFN-y and Ia antigen expression in normal macrophages (Table 2). There was also a dose-dependent relationship in macrophages from stressed mice; however, the percentage of Ia positive macrophages in stressed mice was much lower than that in nonstressed mice at all concentrations of IFN-y added. Therefore, the response of macrophages to IFN-y was decreased in stressed macrophages and could not be restored by increasing the IFN-y concentrations. The presence of INDO (5 x lo-’ A4) with IFN-y (10 IU/ml) did not increase Ia expression in either normal or stressed mice suggesting that PG were not involved in the suppression of IFN-v induced Ia expression (Table 2). This was also true when INDO was used at higher doses (10V6 and lo-’ M; data not shown). Production
of IL-l
by Macrophages
from Stressed Mice
IL-1 production by macrophages was determined using the IL-l-dependent DlO.G4.1 cell line. Both cell-associated IL-1 and secreted IL-l from unstimulated peritoneal macrophages were greatly increased in stressed mice (Fig. 3). In three experiments cell-associated IL-1 from unstimulated peritoneal macrophages was increased an average of 368.3% (range 305 to 409%) and secreted IL-1 was increased an average of 644.5% (range 480 to 1424%) compared to controls. There were no significant differences in IL-1 levels between normal and stressed adherent macrophages activated with LPS over a wide dose-response range (data not shown). DISCUSSION
The cold water stress paradigm used in these studies showed effects on lymphocytes similar to those of other stress models. Numbers of thymocytes and splenocytes in stressed mice dropped dramatically; these results are consistent with the findings of other laboratories (Keller et al., 1988). Lymphocyte blastogenesis and cytotoxicity of NK cells were also impaired in stressed mice compared to controls. NK activity and lymphocyte blastogenesis in response to miTABLE I Peritoneal Macrophage Ia Antigen Expression Induced by Recombinant INF-y in Stressed Mice Ia+ macrophages (%) Group
n
Control Stress
4 4
6.4 + 1.2” 10.1 + 2.2
D Values presented are means 2 standard deviation. * Significantly different from normal mice (p < .Ol).
IFN 69.1 + 2.9 29.1 + 1.3’
286
CHENG ET AL.
TABLE 2 Dose-Response Relationship Between INF-y Concentration and Ia Induction in Stressed Mice Iaf macrophages (%) Treatments
Control
Stressed
Control Indo Indo + IFN 10 W/ml IFN 5 IU/ml 10 25 50 100
5.3 2 0.8” 4.7 5 1.5 74.5 ‘- 5.1
5.2 t 0.5 4.4 2 0.3 37.9 2 2.9b
31.3 70.2 75.2 76.7 79.7
14.7 33.2 31.7 43.0 42.5
2 1.7 -+ 4.3 2 5.7 2 4.9 2 2.0
2 k + 5 e
l.Ob 4.0b 1.36 1.3b 2.1b
u Values presented are means f standard deviation. b Significantly different from control group @ < .Ol).
togens have also been shown to be decreased in rodents stressed by restraint and electric shock (Cunnick, Lysle, Armfield, & Rabin, 1988; Laudenslager, Ryan, Drugan, Hyson, & Maier, 1983; Mormede, Dantzer, Michaud, Kelly, & Le Moal, 1988; Okimura et al., 1986b). Thus, this cold water stress paradigm produced immunological changes consistent with other models of stress. The major new findings of the present study are that stress activated macrophages in an apparently dysregulated manner in that some functions such as IL-l and PGE, production were enhanced while other functions such as the ability to induce Ia antigen expression by INF-y were suppressed. The effects of stress on macrophage function, particularly the increased IL-l production and decreased Ia expression, are novel. Although the precise localization of IL-l in the fixed macrophage is unresolved (Kurt-Jones, Beller, Mizel, & Unanue, 1985; Suttles, Carruth, & Mizel, 1989) the finding of comparably elevated cell-associated and se-
1
FIG. 3. Synthesis of cell-associated and secreted IL-l by peritoneal macrophages in stressed mice. There were four mice in each group. Open bars represent macrophages from control mice, and solid bars represent macrophages from mice stressed for 4 days.
IMMUNOSUPPRESSION
AND
COLD
WATER
STRESS
287
creted IL-l in fixed macrophages suggests that the enhanced levels of IL-1 are attributed to an increase in IL-l production, rather than enhanced or accelerated secretion/release from the viable stressed macrophage. Although we did not carry out measurements of various neurohormones and/or neuromediators, one may speculate about the mechanisms that may be involved in the immunological changes we have found. The extent to which glucocorticoids are elevated in mice exposed to cold water stress is not known. One would suppose that they are elevated since corticosteroid elevation generally accompanies stress. Corticosteroids have been shown to suppress mitogen-induced lymphocyte blastogenesis in a number of species (Dantzer & Kelley, 1989) and may be a factor in the suppression of blastogenesis seen in our studies. In vitro, corticosteroids have been shown to have a synergistic inhibitory action along with PGE, on lymphocyte proliferation. One possible mechanism suggested is that cortisol sensitizes lymphocytes to the suppressive effects of PGE, (Berenbaum, Cope, & Bundick, 1976). In our experiments, macrophages from stressed mice produced increased levels of PGE, (similar to the rise seen in activated macrophages; Snider, Fettle, & Zwilling, 1982) and IL-l. Since increases in corticosteroids generally accompany stress, it is of interest that increases in IL-l and PGE, occurred since glucocorticoids can block the production of PGs and IL-l by macrophages (Besedovsky, Del Rey, Sorkin, & Dinarello, 1986; Okimura & Nigo, 1986; Snyder & Unanue, 1982). It should be noted, however, that some evidence exists indicating that the rise in glucocorticoids after stress may not be the first event and other peptides may actually activate initial immune events (Morley, Kay, & Solomon, 1989). Our studies have revealed that macrophages from stressed mice express less Ia antigen than control macrophages after exposure to INF-y. This may well be responsible for certain immunosuppressive characteristics of stressed mice. The diminished Ia antigen expression in macrophages from stressed mice may compromise macrophage functions such as antigen presentation and other accessory cell functions. Studies to determine this are currently in progress. The reason for the diminished Ia presence on the cell surface and whether this is due to decreased production and/or processing of the Ia glycoprotein are not known at present. Our preliminary observations indicate that it occurs in the presence of INDO suggesting that PGs are not responsible. In vitro, PGE, at the level produced by our stressed macrophages has been shown to inhibit IL-l secretion (Knudsen, Dinarello, & Strom, 1986). Because this was not observed in our experiments, the possibility also exists that macrophages from stressed mice are less sensitive to the suppressive effects of PGE,. Evidence for resistance to PGE, or receptor down regulation is also suggested by the fact that INF-y stimulated macrophages from stressed mice were not affected by the addition of INDO. Whether the lack of sensitivity to PGE, in macrophages from stressed animals may represent a change in the cell subpopulations of our elicited peritoneal macrophages from stressed vs control mice remains to be tested. In our present experiments, it is interesting that unstimulated macrophages derived from stressed mice secrete high levels of IL-l. This is a novel finding and raises the question whether macrophages in vivo are activated directly by CNS
288
CHENG
ET AL.
stimulation. It has been well documented that primary and secondary lymphoid organs such as the spleen, lymph nodes, thymus, and bone marrow are innervated by sympathetic noradrenergic fibers and fibers containing substance P (SP) (Felten, Ackerman, Wiegand, & Felten, 1987; Felten & Felten, 1988). Both the catecholamines and SP can affect macrophages directly as receptors for both these substances are found on their surface. Substance P is a chemoattractant for macrophages and can stimulate the production of IL-l, IL-6, and TNF (Hartung, 1988). Substance P is released during activation of the autonomic nervous system and has been demonstrated to be present in primary and secondary lymphoid tissues (Felten, Lot-ton, & Felten, 1989; Lorton, Bellinger, Felten, & Felten, 1989). Thus, during stress, macrophages could be exposed to SP which may be responsible for some of the alterations we have seen in macrophage activity. Alternatively, CRF has recently been found to be increased after stress, and this neuropeptide has been shown to stimulate IL-l production (Kavelaars, Ballieux, & Heijnen, 1989). It is of interest that the changes in macrophages we have described are similar, in many respects, to the changes induced by endotoxin. Thus, endotoxin may induce or enhance IL-l and PGE, production and, at certain doses, may decrease Ia production in response to INF-y. We have not found elevated endotoxin levels in the peripheral blood of 4-day stressed mice at the time of sacrifice (although this does not rule out elevations at earlier times). While a low level of endotoxin was detectable in the proteose peptone used to elicit peritoneal macrophages (~2.4 rig/ml), the absence of such changes in control cells, elicited by the same preparation, argues against endotoxin causing the observed changes in cells from stressed mice. In preliminary experiments moreover, similar findings for Ia expression and IL-l were obtained in macrophages isolated from the endotoxinresistant mouse strain C,H/HeJ and subjected to the same stress paradigm, suggesting that endotoxin is not responsible for the stress-associated changes (a 45.5% decrease, p < .05, in INF-y-induced Ia production and a 23.8% increase, p < .05, in unstimulated secreted IL-l for stress vs unstressed C,H/HeJ mice; our unpublished data). Although we find no evidence that endotoxin itself is involved, activation of macrophages with the elaboration of the endogenous pyrogen, IL-l, does share some characteristics with the response to certain infectious agents. We do not yet know whether other endogenous pyrogens such as TNF and IL-6 are found at elevated levels in macrophages from stressed animals. It will be of interest to determine this and to determine whether any acute phase reactants are also induced following stress since certain acute phase reactants are induced by the endogenous pyrogens mentioned above (Prowse & Baumann, 1989). If stress can induce factors which cause inflammatory reactions, it raises the question whether stress might be a promoter or contributing factor in certain inflammatory diseases of unknown etiology such as rheumatoid arthritis. Our results indicate that a complex picture of macrophage dysregulation follows stress, particularly the increase in production of IL-l, PGE,, and the diminished ability to synthesize Ia antigen following stimulation with INF-y. These studies may reveal the mechanism(s) whereby other immune functions, particularly those
IMMUNOSUPPRESSION
AND COLD WATER
STRESS
289
of lymphocytes, are altered, and may further elucidate the local regulatory mechanisms that affect an animal’s ability to respond to an immunological challenge. ACKNOWLEDGMENTS This research was supported by a grant from the NC1 (CA8882) and an American Cancer Society Institutional Research Grant (ACS 97-M) awarded to Boston University Medical School. We are indebted to Barbara Pugh and Cynthia Peirce for excellent technical help.
REFERENCES Aarstad, H. J., Gardernack, G., & Seljelid, R. (1983). Stress causes reduced natural killer activity in mice. &and. J. Immunol. 18,461-464. Berenbaum, M. C., Cope, W. A., 8r Bundick, R. V. (1976). Synergistic effect of cortisol and prostaglandin E, on the PHA response. Clin. Exp. Immunol. 26, 534-537. Besedovsky, H., Del Rey, A., Sorkin, E., & Dinarello, C. A. (1986). Immunoregulatory feedback between interleukin-1 and glucocorticoid hormones. Science 233, 652-654. Bonta, I. L., Elliott, G. R., Tak, C. & Ben-Efiaim, S. (1989). Indomethacin stimulation of macrophage cytostasis against MOPC-315 tumor cells is enhanced by endogenous metabolites of lipoxygenase and counteracted by prostaglandin E,. Agents Actions 26, 167-169. Cowing, C., Schwartz, B. D. & Dickler, H. B. (1978). Macrophage Ia antigens. 1. Macrophage populations differ in their expression of Ia antigens. J. Immunol. 120, 378-384. Cunnick, J. E., Lysle, D. T., Armfield, A. & Rabin, B. S. (1988). Shock-induced modulation of lymphocyte responsiveness and natural killer activity, differential mechanisms of induction. Brain Behav. Immun. 2, 102-113. Dantzer, R., & Kelley, K. W. (1989). Stress and immunity: An integrated view of relationships between the brain and the immune system. Life Sci. 44, 1995-2008. Dore-Duffy, P., Guha, A., Rothman, B. L., & Zurier, R. B. (1988). Synthesis of prostaglandin E by peritoneal macrophages from NZB/W mice. Life Sci. 42, 2669-2676. Evans, R. (1984). Regulation of T and B lymphocyte responses to mitogens by tumor-associated macrophages: The dependency of the stage of tumor growth. J. Leuk. Biol. 35, 549-559. Felten, D. L., Ackerman, K. D., Wiegand, S. J., & Felten, S. Y. (1987). Noradrenergic sympathetic innervation of the spleen. I. Nerve fibers associated with lymphocytes and macrophages in specific compartments of the splenic white pulp. J. Neurosci. Res. 18, 28-36. Felten, D. L., & Felten, S. Y. (1988). Sympathetic noradrenergic innervation of immune organs. Brain Behav.
Immun.
2, 293-300.
Felten, S. Y., Lorton, D. L., & Felten, D. L. (1989). Substance P (SP) innervation of the rat thymus. Sot. Neurosci. Abstr. 15, 714. Goodwin, J. S., & Webb, D. R. (1980). Regulation of the immune response by prostaglandins. Clin. Immunol.
Immunopharmacol.
15, 106-122.
Goodwin, J. S., Bromberg, S., Staszak, C., Kaszubowski, P. A., Messner, R. P., & Neal, J. F. (1981). Effect of physical stress on sensitivity of lymphocytes to inhibition by prostaglandin E,. J. Immunol.
127, 518-522.
Hartung, H. P. (1988). Activation of macrophages by neuropeptides. Brain Behav. Immun. 2,275-281. Irwin, M. (1988). Depression and immune function. Stress Med. 4, 95-103. Jesberger, J. A., & Richardson, J. S. (1985). Animal models of depression: Parallels and correlates to severe depression in humans. Biol. Psychiatry 20, 764-784. Kavelaars, A., Ballieux, R. E., & Heijnen, C. J. (1989). The role of IL-1 in the corticotropin-releasing factor and arginine vasopressin-induced secretion of immunoreactive B-endorphin by human peripheral blood mononuclear cells. J. Immunol. 142, 2338-2342. Kaye, J., Gillis, S., Mizel, S. B., Schevach, E. M., Malek, T. R., Dinarello, C. A., Lachman, L. B., & Janeway, C. A. (1984). Growth of a cloned helper T cell line induced by a mono-clonal antibody
290
CHENG ET AL.
specific for the antigen receptor: Interleukin 1 is required for the expression of receptors for interleukin 2. J. Immunol. 133, 1339-1345. Keller, S. E., Weiss, J. M., Schleifer, S. J., Miller, N. F., & 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., Schleifer, S. J., Liotta, A. S., Bond, R. N., Farhoody, N., & Stein, M. (1988). Stressinduced alterations of immunity in hypophysectomized rats. Proc. Natl. Acad. Sci. USA 85, 9297-930 1. Kiecolt-Glaser, J. K., Gamer, W., Speicher, C., Penn, G. M., Holliday, J. E., & Glaser, R. (1984). Psychosocial modifiers of immunocompetence in medical students. Psychosom. Med. 46, 7-14. Knudsen, P. J., Dinarello, C. A., & Strom, T. B. (1986). Prostaglandins post-transcriptionally inhibit monocyte expression of interleukin 1 activity by increasing intracellular cyclic adenosine monophosphate. J. Immunol. 137, 3189-3194. Kronfol, Z., Silva, J. Jr., &eden, J., Dembinski, S., Gardner, R., h Carroll, B. (1983). Impaired lymphocyte function in depressive illness. Life Sci. 33, 241-247. Kurt-Jones, E. A., Beller, D. I., Mizel, S. B., & Unanue, E. R. (1985). identification of a membraneassociated interleukin 1 in macrophages. Proc. Natl. Acad. Sci. USA 82, 1204-1208. Laudenslager, M. L. (1987). Psychosocial stress and susceptibility to infectious illness. In E. Kurstak, Z. J. Liposki, & P. V. Morozov (Eds.), Viruses, immunity, and mental disorders. Plenum: New York. Laudenslager, M. L., Ryan, S. M., Drugan, R. C., Hyson, R. L., & Maier, S. F. (1983). Coping and immunosuppression: Inescapable but not escapable shock suppresses lymphocyte proliferation. Science 221, 568-570. Leizer, T., Clarris, B. .I., Ash, P. E., van Damme, J., Saklatvala, J., & Hamilton, J. A. (1987). Interleukin 1B and interleukin-la stimulate the plasminogen activator activity and prostaglandin E, levels of human synovial cells. Arth. Rheum. 30, 562-566. Levy, E. M., & Black, P. H. (1988). Depression and the immune system. Clin. Immunol. Newsl. 9, 4-6. Lorton, D., Bellinger, D. L., Felten, S. Y., & Felten, D. L. (1989). Substance P (SP) and calcitonin gene-related peptide (CGRP) innervation of the rat spleen. Sot. Neurosci. Abstr. 15, 714. Morley, J. E., Kay, N., & Solomon, G. F. (1989). Opioid peptides, stress, and immune function. In Y. Tache, J. E. Morley, & M. R. Brown (Eds.), Neuropeptides and stress, p 222-234. SpringerVerlag: New York. Mormede, P., Dantzer, R., Michaud, B., Kelley, K. W. & Le Moal, M. (1988). Influence of stressor predictability and behavioral control on lymphocyte reactivity, antibody responses and neuroendocrine activation in rats. Physiol. Behav. 43, 577-583. Okimura, T., & Nigo, Y. (1986). Stress and immune responses. I. Suppression of T cell function in restraint-stressed mice. Japan. J. Pharmacol. 40, 505-511. Okimura, T., Satomi-Sasaki, Y., & Ohkuma, S. (1986a). Stress and immune responses. II. Identitication of stress-sensitive cells in murine spleen cells. Japan. J. Pharmacol. 40, 513-525. Okimura, T., Ogawa, M., & Yamauchi, T. (1986b). Stress and immune responses. III. Effect of restraint stress on delayed type hypersensitivity (DTH) response, natural killer (NK) activity and phagocytosis in mice. Japan. J. Pharmacol. 41, 229-235. Parhar, R. S., & Lala, P. K. (1988). Prostaglandin EZ-mediated inactivation of various killer lineage cells by tumor-bearing host macrophages. J. Leuk. Biol. 44, 474-484. Pavlidis, N., & Chirigos, M. (1980). Stress-induced impairment of macrophage tumoricidal function. Psychosom. Med. 42, 47-54. Poutsiaka, D. D., Schroder, E. W., Taylor, D. D., Levy, E. M., & Black, P. H. (1985a). Membrane vesicles shed by murine melanoma cells selectively inhibit the expression of Ia antigen by macrophages. J. Immunol. 134, 131144. Poutsiaka, D. D., Taylor, D. D., Levy, E. M., & Black, P. H. (1985b). Inhibition of recombinant interferon-y-induced Ia antigen expression by shed B16 FlO melanoma cell membrane vesicles. J. Immunol. 134, 145-150. Prowse, K. R., & Baumann, H. (1989). Interleukin-1 and interleukind stimulate acute-phase protein production in primary mouse hepatocytes. J. Leuk. Biol. 45, 55-61.
IMMUNOSUPPRESSION
AND COLD WATER
STRESS
291
Schleifer, S. J., Keller, S. E., Came&o, M., Thornton, J. C., & Stein, M. (1983). Suppression of lymphocyte stimulation following bereavement. JAMA 250, 374-377. Schleifer, S. J., Keller, S. E., Siris, S. G., Davis, K. L., & Stein, M. (1985). Depression and immunity. Lymphocyte function in ambulatory depressed patients, hospitalized schizophrenic patients, and patients hospitalized for hemiorrhaphy. Arch. Gen. Psychiatry 42, 129-133. Snider, M. E., Fertel, R. H., & Zwilling, B. S. (1982). Prostaglandin regulation of macrophage function: Effect of endogenous and exogenous prostaglandins. Ceil. Immunol. 74, 234-242. Snyder, D. S., & Unanue, E. R. (1982). Corticosteroids inhibit murine macrophage Ia expression and interleukin 1 production. J. Zmmunol. 129, 1803-1805. Steeg, P. S., Moore, R. N., Johnson, H. M., & Oppenheim, J. J. (1982). Regulation of murine macrophage Ia antigen expression by a lymphokine with immune interferon activity. J. Exp. Med. 156, 1780-1793. Stein, M. (1989). Stress, depression, and the immune system. J. Clin. Psychiatry 50, 3-2. Stein, M., Keller, S. E., & Schleifer, S. J. (1985). Stress and immunomodulation: The role of depression and neuro-endocrine function. J. Immunol. 135, 827s-833~. Struhar, D. J., Harbeck, R. J., Gegen, N., Kawada, H., & Mason, R. J. (1989). Increased expression of class II antigens on the major histocompatibility complex on alveolar macrophages and alveolar type II cells and interleukin-1 (IL-l) secretion from alveolar macrophages in an animal model of silicosis. Clin. Exp. Immunol. 77, 281-284. Suttles, J., Carruth, L. M., & Mizel, S. B. (1989). Detection of IL-la and IL-1E in the supematants of paraformaldehyde-treated human monocytes: Evidence against a membrane form of IL-l. J. Immunol. 144, 170-174. Tatfet, S. T., & Russell, S. W. (1981). Macrophage-mediated tumor cell killing: Regulation of expression of cytolytic activity by prostaglandin E. J. Immunol. 126, 424-427. Uhing, R. J., & Adams, D. 0. (1989). Molecular events in the activation of murine macrophages. Agents Actions 26, 9-14. Unanue, E. R. (1981). The regulatory role of macrophages in antigenic stimulation. II. Symbiotic relationship between lymphocytes and macrophages. Adv. Immunol. 31, I-136. Unanue, E. R., & Allen, P. M. (1987). The basis for the immunoregulatory role of macrophages and other accessory cells. Science 236, 551-557. Veldhuis, H. D., Croiset, G., Ballieux, R. E., He&en, C. J., & DeWied, D. (1987). Emotional stress and modulation of immunological reactivity. In D. Nerozzi, F. K. Goodwin, & E. Costa (Eds.), Hypothalamic dysfunction in neuro-psychiatric disorders, pp. 237-245. Raven Press: New York. Weis, J. M., Sundar, S. K., Becker, K. J., & Cierpial, M. A. (1989). Behavioral and neural irdluences on cellular immune responses: Effects of stress and interleukin-1. J. Clin. Psychiatry 50, 43-55. Workman, E. A., & Via, M. F. (1987). T-lymphocyte polyclonal proliferation: Effects of stress response style on medical students taking national board examinations. Clin. Zmmunol. Zmmunopathol. 43, 308-313. Received February 20, 1990
BEHAVIOR,
AND
4, 278-291 (1990)
IMMUNITY
Immunosuppression
in Mice Induced by Cold Water Stress
CHENG GAN JIANG, JULIE L. MORROW-TESCH, DAVID I. BELLER, ELINOR M. LEVY AND PAUL H. BLACK Department of Microbiology,
Boston University School of Medicine, 80 East Concord Street, Boston, Massachusetts 02118
A number of studies indicate that stress can result in suppression of the immune system in animals and man. Most of the studies have focused on alterations of lymphocyte function while only a few have investigated alterations of macrophage function or macrophage cytokine production. Macrophages play an essential role in homeostasis of the immune response. Indeed, the earliest events of the immune response occur in cells of the monocytic lineage, and their secretion of various cytokines may have both immunological and nonimmunological effects. The present studies were undertaken to determine whether alterations in macrophage physiology occur in mice subjected to a stress stimulus. Our studies in mice exposed to cold water stress for 4 days indicated reduced numbers of thymocytes and splenocytes, decreased T-cell blastogenesis, and reduced NK activity. Examination of elicited peritoneal macrophages from stressed mice revealed increased prostaglandin E, (PGE,) secretion and decreased immune region associated antigen (la) expression in response to interferon-y. Despite elevated PGE, levels, indomethacin was generally unable to restore depressed immune function. Of special interest was the finding that cell-associated and secreted interleukin 1 were significantly higher from unstimulated elicited macrophages from stressed mice. These results suggest that early in the response to stress, functions of a variety of cells within the immune system, especially macrophages, are altered and that dysregulated macrophage function may well contribute to the generalized suppression of the immune response in cold water stressed mice. 0 1990 Academic Press, Inc.
INTRODUCTION
Numerous investigations in both humans and animals have indicated that the immune system is affected by the response to stress (Dantzer & Kelley, 1989). It is known that psychological stress may influence the host’s response to infectious agents (Laudenslager, 1987). A number of investigations have documented physiological, hormonal, and immunological changes as a result of stress. The results of recent research indicate that stressful stimuli may depress immune functioning in humans and in animal models (Jesberger & Richardson, 1985; Levy & Black, 1988; Stein, Keller & Schleifer, 1985; Veldhuis, Croiset, Ballieux, Heijnen, & DeWied, 1987). Decreased T-cell mitogen response to concanavalin A (Con A)’ has been found ’ Abbreviations used: ACTH, adrenocorticotropic hormone; CNS, central nervous system; Con A, concanavalin A; DMEM, Dulbecco’s modified Eagle minimal essential medium; FITC, fluorescein isothiocyanate; FBS, fetal bovine serum; Ia, immune region associated antigen; IFN-y, interferon-y; IL-l, interleukin 1; IL-6, interleukin 6; INDO, indomethacin; LPS, Escherichia coli lipopolysaccharide; PEC, peritoneal exudate cells; PBS, sodium phosphate (10 mM) buffered saline; PGE,, prostaglandin E,; TNF, tumor necrosis factor; NK cell, natural killer cell; NP-40, nonidet P-40; SP, substance P. 278 0889-1591/90 $3.00 Copyright 6 1990 by Academic Press, Inc. All rights of reproduction in any form reserved.
IMMUNOSUPPRESSION
AND
COLD
WATER
STRESS
279
in recently bereaved widowers and hospitalized, severely depressed patients (Irwin, 1988; Kronfol et al., 1983; Schleifer, Keller, Camerino, Thornton, & Stein, 1983; Schleifer, Keller, Siris, Davis, & Stein, 1985; Stein, 1989). Significantly low levels of natural killer (NK) cell activity, decreased mitogenesis, low levels of IFN-y, and elevated levels of antibody to certain herpesviruses have been demonstrated in blood samples collected from medical students during examination periods (Kiecolt-Glaser, Garner, Speicher, Penn, Holliday, & Glaser, 1984; Workman & Via, 1987). In animal experiments, investigators have confirmed that lymphocyte proliferation (Keller, Weiss, Schleifer, Miller, & Stein, 1981; Weiss, Sundar, Becker, & Cierpial, 1989), antibody synthesis (Okimura, Satomi-Sasaki, & Ohkuma, 1986a), NK activity (Aarstad, Gardernack, & Seljelid, 1983; Okimura, Ogawa & Yamauchi, 1986b), and macrophage phagocytosis and tumoricidal function (Pavlidis & Chirigos, 1980) are impaired subsequent to various stressful stimuli. Accessory cells such as macrophages are essential for the induction of most immunological responses (Unanue, 1981; Unanue & Allen, 1987). The expression of class II MHC gene products (the Ia antigens in the mouse) and the processing of nominal antigen are essential for antigen presentation to and activation of helper T cells (Uhing 8z Adams, 1989). Macrophages further contribute to the regulation of immunologic responses by releasing various cytokines such as interleukin 1 (IL-l) and prostaglandins (Bonta, Elliott, Tak, & Ben-Efraim, 1989; Dore-Duffy, Guha, Rothman, & Zurier, 1988; Goodwin et al., 1981; Goodwin & Webb, 1980; Taffet & Russell, 1981). The aim of the studies reported here was to examine certain important immune functions subsequent to a stressful stimulus focusing, in particular, on the role of macrophages. MATERIALS
AND METHODS
Animals Male C57BL/6J mice were obtained from Jackson Laboratory (Bar Harbor, ME). Animals were housed, four or five per cage, under pathogen-free conditions on a 12-h light/dark cycle (lights off at 6 PM) and were provided commercial rodent chow and water ad libitum. These mice were used between 6 and 8 weeks of age as the source of splenocytes, thymocytes, and peritoneal exudate cells. Animals were allowed a minimum of 1 week acclimatization after arrival at our animal facility before experiments were conducted. No differences in the immunological parameters we have utilized, in response to stress, have been detected between animals acclimatized for 1 week or longer. Stress Procedure Cold water stress was induced as described with some modifications (Aarstad et al., 1983). The stress paradigm was administered twice daily (9 AM and 3 PM), for a duration of 1 min each time. Control and stressed treatment groups contained two to five and four to five mice, respectively. Control animals were transported from the animal room to the laboratory along with stressed mice. Test mice were
280
CHENG
ET
AL.
treated for 4 consecutive days. Briefly, mice were placed, one at a time, in a 4-liter glass container filled with 1 liter of cold water (4”C), and watched continuously for the entire time during which they were in the cold water; the water was deep enough to cover the back. At the end of each test, wet mice were dried with paper towels. This stress paradigm represents a complex interaction of physical and psychological stressors including anxiety, hypothermia, and exercise and was taken as the cummulative effects of all of these. Mice were injected with 1.5 ml of 10% proteose peptone (endotoxin level ~2.4 rig/ml as determined by the Limulus amebocyte assay) intraperitoneally 3 days prior to sacrifice to elicit peritoneal macrophages (after the first day of stress). Animals were sacrificed by cervical dislocation 1 day after the last stress session. All experiments were conducted three times unless otherwise stated. Preparation
of Immune
Cells
All media used to prepare immune cells were low in endotoxin as indicated by the manufacturer (FCS < 1 ng endotoxin/ml and medium co.05 &ml as tested in the Limulus Amoebocyte Lysate assay). Peritoneal exudate cells were obtained by lavage (Cowing, Schwartz, & Dickler, 1978). Briefly, the ventral surface was soaked with 70% ethanol and using two sterile forceps, the abdominal skin was pulled simultaneously in cephalad and caudad directions to expose the ventral surface of the parietal peritoneum. Ten milliliters of sterile, low endotoxin PBS (co.1 ng endotoximml, MA Bioproducts, Walkersville, MD) was injected into the peritoneal cavity using a syringe with a 22-gauge needle and the peritoneal cells were removed and placed immediately into centrifuge tubes on ice. The peritoneal cells were washed and centrifuged at 5OOg for 10 min at 4°C. Mononuclear cells were counted and the viability was determined by trypan blue exclusion. Cells harvested from individual animals were adjusted to lo6 mononuclear cells/ml in Dulbecco’s modified Eagle minimal essential medium (DMEM, Sigma Chemical Co., St. Louis, MO). After a 2-h incubation at 37°C in an humidified atmosphere of 5% CO,, nonadherent cells were removed by washing three times with DMEM. Spleens were removed aseptically into iced DMEM and cell suspensions were made by mincing the spleens using a sharp scissors and then thoroughly disrupting the fragments by aspiration several times through a 21-gauge needle. Debris was removed by passing through a stainless-steel wire mesh. The cells were washed twice and resuspended to 5 x 10’ cells/ml in red blood cell lysing buffer at 4°C for 5 min, in order to lyse red blood cells. The cells were washed twice in DMEM and resuspended to 10’ cells/ml. Thymocytes were prepared in a similar manner. Mitogen-Induced
Lymphocyte
Proliferation
The response of splenic lymphocytes to Con A stimulation was assessed as previously described (Evans, 1984). Briefly, lo6 cells were added to each of three wells on a 96-well, flat-bottomed, microtiter plate (Costar, Data Packaging Corp., Cambridge, MA) in a final volume of 0.2 ml with or without 2.5 kg/ml of Con A. This mitogen concentration and cell number have previously been shown in our laboratory to be optimal. INDO was dissolved in 95% ethanol to a concentration of lop2 M and then diluted in DMEM. The concentration of indomethacin (INDO)
IMMUNOSUPPRESSION
AND
COLD
WATER
STRESS
281
utilized was 5 x lo-’ M unless otherwise indicated. Cell cultures were incubated at 37°C for a total of 72 h. The splenic cells were pulsed with 1 PCi [3H]thymidine during the last 16 h of the incubation. Plates were then harvested with a multiwell cell harvester onto glass fiber filters and the counts per minute of radiolabeled DNA precursor incorporated by the cells was determined with a Beckman liquid scintillation counter. la Antigen Induction Macrophages harvested from individual animals by peritoneal lavage were seeded into chambers of a multichambered slide (Nunc, Inc., Naperville, IL) at a concentration of 2 x lo5 cells/chamber. To each chamber, either recombinant IFN-y (Amgen, Thousand Oaks, CA) or medium alone was added and the slides were incubated for 5 days at 37°C in an humidified 5% CO, atmosphere. The final concentration of IFN-y was 10 IU/ml unless otherwise stated. We have determined these conditions to be optimal for Ia induction by IFN-y in mouse peritoneal exudative cells in our lab (Poutsiaka, Schroder, Taylor, Levy & Black, 1985a; Poutsiaka, Taylor, Levy, & Black, 1985b). At the termination of the culture period, the percentage of macrophages expressing surface Ia antigen was determined (Steeg, Moore, Johnson, & Oppenheim, 1982). Approximately equal numbers of cells from stressed vs unstressed mice plated onto the slides as verified after 4 h by protein assay (Pierce, Rockford, IL) and at the end of the culture period. Macrophages were washed four times with PBS and fixed with 0.25% glutaraldehyde in PBS for 30 min at room temperature, and the slides were washed three times with PBS. Ia antigen was determined visually by immunoflourescence, using a monoclonal anti-I-Ab antibody (IgM) produced from the mouse hybridoma 25-g-38 (obtained from NIAID). The monoclonal antibody was added to each chamber, incubated for 1 h at 37°C and then washed three times with PBS. Fluorescein isothiocyanate (FITC)-conjugated rabbit anti-mouse immunoglobulin (DAK0 Corp., Santa Barbara, CA) was added to each chamber and the slides were incubated for an additional hour at 37°C and then washed three times with PBS. Coverslips were mounted on the chamber-slides and the slides were stored in the dark at 4°C. Each chamber was examined using a Nikon fluorescence microscope and 200 total cells were examined, with each treatment condition being assayed in duplicate. Ia positive cells were defined as those cells showing bright fluorescence. IL-1 Activity Assay The production of IL-1 by peritoneal macrophages was determined by proliferation using the DlO.G4.1 cell assay (Kaye et al., 1984; Struhar, Hat-beck, Gegen, Kawada, & Mason, 1989). For this assay, cells were pooled from all animals within a treatment group. Proteose peptone-elicited peritoneal macrophages (2 x lo4 to 1 x 105) were added to each well of a 96-well plate. Nonadherent cells were removed by gentle washing twice after a 2-h incubation at 37°C. Different concentrations of LPS (4 to 500 &ml; Sigma Chemical Co.) were added into cultures prepared with low endotoxin medium (co.05 ng endotoxin/ml). After 20 h of incubation, supernatants were collected and stored at - 70°C for determination of
282
CHENG
ET
AL.
secreted IL-l. Cells were also fixed with 1% paraformaldehyde for 10 min and then washed twice with PBS for assay of cell-associated IL-l. DlO.G4.1 cells, a clone of T cells which responds to IL-l in the presence of Con A and conalbumin, were utilized to assay IL-l. These cells were maintained using the supematant from rat splenocytes stimulated with Con A. For the secreted IL-I assay, each well of a 96-well tissue culture plate was seeded with 2 x lo4 cells in the presence or absence of spent media. The final concentration of Con A added was 2.5 l&ml. To determine cell-associated IL-l, DlO.G4.1 cells and Con A were added directly to wells containing fixed macrophages. These cells were incubated for 3 days with [3H]thymidine being added during the last 16 h. Proliferation was assayed by incorporation of [3H]thymidine into DNA. NK-cell
Cytotoxicity
Assay
YAC-1 cells, a Moloney virus-induced T-cell lymphoma, were used as target cells in the NK assay (Aarstad et al., 1983); they were maintained in RPM1 1640 medium supplemented with antibiotics (penicillin and streptomycin), L-glutamine, and 10% fetal bovine serum (FBS). Target cells were labeled with sodium chromate (“0; New England Nuclear, Boston, MA). Four effector-to-target cell ratios ranging from 100: 1 to 12.5: 1 were tested on microtiter plates with lo4 target cells per well. Medium was added to target cells for spontaneous lysis determinations, and 1% Nonidet P-40 (NP-40) was added for determination of 100% lysis. Plates were incubated for 4 h at 37°C in a humidified 5% CO, atmosphere. The supernatant fluid was harvested using a multiwell cell harvester for radioactive counting in a gamma counter. The percentage of specific 51Cr release is equal to cpm test sample - cpm medium control x 100. cpm NP-40 control - cpm medium control PGE, Determination PGE, concentrations in the culture supematants of peritoneal macrophages derived from normal and stressed mice were determined using a radioimmunoassay kit (Advanced Magnetics, Inc., Cambridge, MA.). Cells were incubated (37”C, 5% COJ with or without IFN-y at the final concentration of 10 IU/ml, unless otherwise stated, for 5 days. Supernatants were collected and stored at -70°C until assays were performed (Leizer et al., 1987). Statistical
Analysis
Immunological assays were performed on peritoneal macrophages and splenocytes from individual animals except for the IL-l assay when cells from mice within a treatment group were pooled. Significant differences between stressed and control mice were determined within each experiment by Student’s t test. Differences were considered to be significant when probability @) values <.05 were obtained. Within all experiments, statistically significant differences were consistent (I, < .05).
IMMUNOSUPPRESSION
AND
COLD
WATER
STRESS
283
RESULTS Enumeration
of Splenocytes
and Thymocytes
in Stressed
Mice
The impact of exposure to a stressful stimulus on spleen and thymus cell numbers was examined. In one experiment, the number of thymocytes was decreased in the stressed mice group (14.78 + 2.47 and 6.16 + 0.82 x IO’ cells for control and stressed mice, respectively; mean + SE; p < .Ol). The average decrease across three experiments was 59.6%, ranging from 50.1 to 67.1%. There was a concomitant decrease of splenic lymphocyte numbers in stressed mice (11.18 + 0.09 and 0.87 k 0.07 x lo* cells for control and stressed mice, respectively; p < .Ol). The average decrease was 58.6%, ranging from 19.9 to 92.9% across three experiments. Blastogenic
Responses
in Stressed
Mice
Mitogenesis in splenic lymphocytes induced by Con A was used to assess T-lymphocyte function. In one experiment, splenic lymphocytes from stressed mice exhibited decreased blastogenesis of approximately 44% (60.5 + 9.46 and 33.9 + 9.94 x lop3 cpm for control and stressed mice, respectively, mean 2 SE; p < .05). The average decrease across four experiments was 29.3%, ranging from 18.8 to 44%. Because PGs, produced largely by activated macrophages and monocytes, are inhibitors of mitogenesis, blastogenesis experiments were carried out in the presence of INDO, an inhibitor of PG production. INDO alone did not activate lymphocytes either in normal mice or in stressed mice. In normal splenocytes, the mitogenic response induced by Con A was not changed by the addition of INDO. With a stress exposure of 4 days, Con A induced blastogenesis decreased to 56% of control values and was not increased by the presence of INDO (33.9 + 9.94 and 37.98 k 2.07 x lop3 cpm for Con A alone or with INDO, respectively; p > .lO). Natural
Killer Activity
in Stressed
Mice
NK activity in spleen cells from cold water stressed mice was compared with those from control mice. NK activity of the test mice enduring cold water swim stress was reduced (p s .05) by at least 60% of controls at each effector:target ratio (percentage “Cr release, mean + SE; 31.2 + 2.0 vs 7.8 + 2.4 at lOO:l, 15.2 2 4.8 vs 5.9 2 2.7 at 50:1, 13.8 + 5.8 vs 1.9 + 0.8 at 25:l and 6.9 + 4.3 vs 1.7 + 0.6 at 12.5:1; control vs. stress, respectively). Although PGs can inhibit NK activity (Parhar & Lala, 1988), in the present experiments, NK activity of splenic lymphocytes derived from either stressed or normal mice was not increased significantly by the addition of INDO (data not shown). Increased
PGE, Synthesis
in Stressed
Mice
To explore some possible regulatory factors which may influence immune function, the production and release of PGE, were investigated. The baseline levels of PGE, produced by unactivated macrophages from stressed mice were higher than normal, control values, but were low compared to activated macrophages (Fig. 1). Macrophages from stressed mice activated with 10 IU/ml of IFN-y synthesized more PGE, than those from normal mice; the levels of PGE, in culture media
284
CHENG ET AL.
!
i+c hl
400 300
k
5 F B 5 s z
200
100 0
Normal
2 Days
4 Days
FIG. 1. PGE, concentration in culture supematants of peritoneal macrophages derived from normal or stressed mice. Open bars represent macrophages without the addition of INF-?I, and cross-hatched bars represent macrophages with the addition of INF-?I. Results are means 2 standard deviation. *Significantly different from the control group (p < .Ol).
increased an average of 253% (range between 163 and 351% in four separate experiments) when compared to levels obtained from macrophages from nonstressed mice. The synthesis of PGE, in peritoneal macrophages was dramatically increased in stressed mice in response to IFN-y, compared to macrophages obtained from nonstressed mice at all concentrations of IFN-y tested (Fig. 2). In macrophages from stressed mice, a dose-response relationship is evident. la Antigen Induction
in Stressed Mice
Although many investigations have examined immunological functions requiring T-cell help in stressed populations, the capacity of macrophages from stressed animals to synthesize and express Ia antigen, which is required for induction of T-cell help, has not been investigated. Elicited peritoneal macrophages from
1000 c x 55
LOO--
0 m
Normal Stressed
800 -700 --
1
600 --
E -6 400 -t-2 300 -8N -8 loo-a 500
200
0
r-l
0
10 IFN-gamma
25
I.7
50
100
(lU/ml)
FIG. 2. Dose responses of PGE, secretion by peritoneal macrophages in stressed mice. There were four mice in each group. Open bars represent control mice, and solid bars represent stressed mice.
IMMUNOSUPPRESSION
AND COLD WATER
STRESS
285
stressed and control mice were utilized to investigate Ia antigen expression. Peritoneal macrophages from stressed mice exhibited a significant inhibition of Ia antigen induction in response to IFN-y of approximately 58% in one experiment, Table 1 (range from four experiments of 49.8 to 69.7%). When the amount of IFN-y added to the culture was increased, there was a positive dose-dependent relationship between the amount of IFN-y and Ia antigen expression in normal macrophages (Table 2). There was also a dose-dependent relationship in macrophages from stressed mice; however, the percentage of Ia positive macrophages in stressed mice was much lower than that in nonstressed mice at all concentrations of IFN-y added. Therefore, the response of macrophages to IFN-y was decreased in stressed macrophages and could not be restored by increasing the IFN-y concentrations. The presence of INDO (5 x lo-’ A4) with IFN-y (10 IU/ml) did not increase Ia expression in either normal or stressed mice suggesting that PG were not involved in the suppression of IFN-v induced Ia expression (Table 2). This was also true when INDO was used at higher doses (10V6 and lo-’ M; data not shown). Production
of IL-l
by Macrophages
from Stressed Mice
IL-1 production by macrophages was determined using the IL-l-dependent DlO.G4.1 cell line. Both cell-associated IL-1 and secreted IL-l from unstimulated peritoneal macrophages were greatly increased in stressed mice (Fig. 3). In three experiments cell-associated IL-1 from unstimulated peritoneal macrophages was increased an average of 368.3% (range 305 to 409%) and secreted IL-1 was increased an average of 644.5% (range 480 to 1424%) compared to controls. There were no significant differences in IL-1 levels between normal and stressed adherent macrophages activated with LPS over a wide dose-response range (data not shown). DISCUSSION
The cold water stress paradigm used in these studies showed effects on lymphocytes similar to those of other stress models. Numbers of thymocytes and splenocytes in stressed mice dropped dramatically; these results are consistent with the findings of other laboratories (Keller et al., 1988). Lymphocyte blastogenesis and cytotoxicity of NK cells were also impaired in stressed mice compared to controls. NK activity and lymphocyte blastogenesis in response to miTABLE I Peritoneal Macrophage Ia Antigen Expression Induced by Recombinant INF-y in Stressed Mice Ia+ macrophages (%) Group
n
Control Stress
4 4
6.4 + 1.2” 10.1 + 2.2
D Values presented are means 2 standard deviation. * Significantly different from normal mice (p < .Ol).
IFN 69.1 + 2.9 29.1 + 1.3’
286
CHENG ET AL.
TABLE 2 Dose-Response Relationship Between INF-y Concentration and Ia Induction in Stressed Mice Iaf macrophages (%) Treatments
Control
Stressed
Control Indo Indo + IFN 10 W/ml IFN 5 IU/ml 10 25 50 100
5.3 2 0.8” 4.7 5 1.5 74.5 ‘- 5.1
5.2 t 0.5 4.4 2 0.3 37.9 2 2.9b
31.3 70.2 75.2 76.7 79.7
14.7 33.2 31.7 43.0 42.5
2 1.7 -+ 4.3 2 5.7 2 4.9 2 2.0
2 k + 5 e
l.Ob 4.0b 1.36 1.3b 2.1b
u Values presented are means f standard deviation. b Significantly different from control group @ < .Ol).
togens have also been shown to be decreased in rodents stressed by restraint and electric shock (Cunnick, Lysle, Armfield, & Rabin, 1988; Laudenslager, Ryan, Drugan, Hyson, & Maier, 1983; Mormede, Dantzer, Michaud, Kelly, & Le Moal, 1988; Okimura et al., 1986b). Thus, this cold water stress paradigm produced immunological changes consistent with other models of stress. The major new findings of the present study are that stress activated macrophages in an apparently dysregulated manner in that some functions such as IL-l and PGE, production were enhanced while other functions such as the ability to induce Ia antigen expression by INF-y were suppressed. The effects of stress on macrophage function, particularly the increased IL-l production and decreased Ia expression, are novel. Although the precise localization of IL-l in the fixed macrophage is unresolved (Kurt-Jones, Beller, Mizel, & Unanue, 1985; Suttles, Carruth, & Mizel, 1989) the finding of comparably elevated cell-associated and se-
1
FIG. 3. Synthesis of cell-associated and secreted IL-l by peritoneal macrophages in stressed mice. There were four mice in each group. Open bars represent macrophages from control mice, and solid bars represent macrophages from mice stressed for 4 days.
IMMUNOSUPPRESSION
AND
COLD
WATER
STRESS
287
creted IL-l in fixed macrophages suggests that the enhanced levels of IL-1 are attributed to an increase in IL-l production, rather than enhanced or accelerated secretion/release from the viable stressed macrophage. Although we did not carry out measurements of various neurohormones and/or neuromediators, one may speculate about the mechanisms that may be involved in the immunological changes we have found. The extent to which glucocorticoids are elevated in mice exposed to cold water stress is not known. One would suppose that they are elevated since corticosteroid elevation generally accompanies stress. Corticosteroids have been shown to suppress mitogen-induced lymphocyte blastogenesis in a number of species (Dantzer & Kelley, 1989) and may be a factor in the suppression of blastogenesis seen in our studies. In vitro, corticosteroids have been shown to have a synergistic inhibitory action along with PGE, on lymphocyte proliferation. One possible mechanism suggested is that cortisol sensitizes lymphocytes to the suppressive effects of PGE, (Berenbaum, Cope, & Bundick, 1976). In our experiments, macrophages from stressed mice produced increased levels of PGE, (similar to the rise seen in activated macrophages; Snider, Fettle, & Zwilling, 1982) and IL-l. Since increases in corticosteroids generally accompany stress, it is of interest that increases in IL-l and PGE, occurred since glucocorticoids can block the production of PGs and IL-l by macrophages (Besedovsky, Del Rey, Sorkin, & Dinarello, 1986; Okimura & Nigo, 1986; Snyder & Unanue, 1982). It should be noted, however, that some evidence exists indicating that the rise in glucocorticoids after stress may not be the first event and other peptides may actually activate initial immune events (Morley, Kay, & Solomon, 1989). Our studies have revealed that macrophages from stressed mice express less Ia antigen than control macrophages after exposure to INF-y. This may well be responsible for certain immunosuppressive characteristics of stressed mice. The diminished Ia antigen expression in macrophages from stressed mice may compromise macrophage functions such as antigen presentation and other accessory cell functions. Studies to determine this are currently in progress. The reason for the diminished Ia presence on the cell surface and whether this is due to decreased production and/or processing of the Ia glycoprotein are not known at present. Our preliminary observations indicate that it occurs in the presence of INDO suggesting that PGs are not responsible. In vitro, PGE, at the level produced by our stressed macrophages has been shown to inhibit IL-l secretion (Knudsen, Dinarello, & Strom, 1986). Because this was not observed in our experiments, the possibility also exists that macrophages from stressed mice are less sensitive to the suppressive effects of PGE,. Evidence for resistance to PGE, or receptor down regulation is also suggested by the fact that INF-y stimulated macrophages from stressed mice were not affected by the addition of INDO. Whether the lack of sensitivity to PGE, in macrophages from stressed animals may represent a change in the cell subpopulations of our elicited peritoneal macrophages from stressed vs control mice remains to be tested. In our present experiments, it is interesting that unstimulated macrophages derived from stressed mice secrete high levels of IL-l. This is a novel finding and raises the question whether macrophages in vivo are activated directly by CNS
288
CHENG
ET AL.
stimulation. It has been well documented that primary and secondary lymphoid organs such as the spleen, lymph nodes, thymus, and bone marrow are innervated by sympathetic noradrenergic fibers and fibers containing substance P (SP) (Felten, Ackerman, Wiegand, & Felten, 1987; Felten & Felten, 1988). Both the catecholamines and SP can affect macrophages directly as receptors for both these substances are found on their surface. Substance P is a chemoattractant for macrophages and can stimulate the production of IL-l, IL-6, and TNF (Hartung, 1988). Substance P is released during activation of the autonomic nervous system and has been demonstrated to be present in primary and secondary lymphoid tissues (Felten, Lot-ton, & Felten, 1989; Lorton, Bellinger, Felten, & Felten, 1989). Thus, during stress, macrophages could be exposed to SP which may be responsible for some of the alterations we have seen in macrophage activity. Alternatively, CRF has recently been found to be increased after stress, and this neuropeptide has been shown to stimulate IL-l production (Kavelaars, Ballieux, & Heijnen, 1989). It is of interest that the changes in macrophages we have described are similar, in many respects, to the changes induced by endotoxin. Thus, endotoxin may induce or enhance IL-l and PGE, production and, at certain doses, may decrease Ia production in response to INF-y. We have not found elevated endotoxin levels in the peripheral blood of 4-day stressed mice at the time of sacrifice (although this does not rule out elevations at earlier times). While a low level of endotoxin was detectable in the proteose peptone used to elicit peritoneal macrophages (~2.4 rig/ml), the absence of such changes in control cells, elicited by the same preparation, argues against endotoxin causing the observed changes in cells from stressed mice. In preliminary experiments moreover, similar findings for Ia expression and IL-l were obtained in macrophages isolated from the endotoxinresistant mouse strain C,H/HeJ and subjected to the same stress paradigm, suggesting that endotoxin is not responsible for the stress-associated changes (a 45.5% decrease, p < .05, in INF-y-induced Ia production and a 23.8% increase, p < .05, in unstimulated secreted IL-l for stress vs unstressed C,H/HeJ mice; our unpublished data). Although we find no evidence that endotoxin itself is involved, activation of macrophages with the elaboration of the endogenous pyrogen, IL-l, does share some characteristics with the response to certain infectious agents. We do not yet know whether other endogenous pyrogens such as TNF and IL-6 are found at elevated levels in macrophages from stressed animals. It will be of interest to determine this and to determine whether any acute phase reactants are also induced following stress since certain acute phase reactants are induced by the endogenous pyrogens mentioned above (Prowse & Baumann, 1989). If stress can induce factors which cause inflammatory reactions, it raises the question whether stress might be a promoter or contributing factor in certain inflammatory diseases of unknown etiology such as rheumatoid arthritis. Our results indicate that a complex picture of macrophage dysregulation follows stress, particularly the increase in production of IL-l, PGE,, and the diminished ability to synthesize Ia antigen following stimulation with INF-y. These studies may reveal the mechanism(s) whereby other immune functions, particularly those
IMMUNOSUPPRESSION
AND COLD WATER
STRESS
289
of lymphocytes, are altered, and may further elucidate the local regulatory mechanisms that affect an animal’s ability to respond to an immunological challenge. ACKNOWLEDGMENTS This research was supported by a grant from the NC1 (CA8882) and an American Cancer Society Institutional Research Grant (ACS 97-M) awarded to Boston University Medical School. We are indebted to Barbara Pugh and Cynthia Peirce for excellent technical help.
REFERENCES Aarstad, H. J., Gardernack, G., & Seljelid, R. (1983). Stress causes reduced natural killer activity in mice. &and. J. Immunol. 18,461-464. Berenbaum, M. C., Cope, W. A., 8r Bundick, R. V. (1976). Synergistic effect of cortisol and prostaglandin E, on the PHA response. Clin. Exp. Immunol. 26, 534-537. Besedovsky, H., Del Rey, A., Sorkin, E., & Dinarello, C. A. (1986). Immunoregulatory feedback between interleukin-1 and glucocorticoid hormones. Science 233, 652-654. Bonta, I. L., Elliott, G. R., Tak, C. & Ben-Efiaim, S. (1989). Indomethacin stimulation of macrophage cytostasis against MOPC-315 tumor cells is enhanced by endogenous metabolites of lipoxygenase and counteracted by prostaglandin E,. Agents Actions 26, 167-169. Cowing, C., Schwartz, B. D. & Dickler, H. B. (1978). Macrophage Ia antigens. 1. Macrophage populations differ in their expression of Ia antigens. J. Immunol. 120, 378-384. Cunnick, J. E., Lysle, D. T., Armfield, A. & Rabin, B. S. (1988). Shock-induced modulation of lymphocyte responsiveness and natural killer activity, differential mechanisms of induction. Brain Behav. Immun. 2, 102-113. Dantzer, R., & Kelley, K. W. (1989). Stress and immunity: An integrated view of relationships between the brain and the immune system. Life Sci. 44, 1995-2008. Dore-Duffy, P., Guha, A., Rothman, B. L., & Zurier, R. B. (1988). Synthesis of prostaglandin E by peritoneal macrophages from NZB/W mice. Life Sci. 42, 2669-2676. Evans, R. (1984). Regulation of T and B lymphocyte responses to mitogens by tumor-associated macrophages: The dependency of the stage of tumor growth. J. Leuk. Biol. 35, 549-559. Felten, D. L., Ackerman, K. D., Wiegand, S. J., & Felten, S. Y. (1987). Noradrenergic sympathetic innervation of the spleen. I. Nerve fibers associated with lymphocytes and macrophages in specific compartments of the splenic white pulp. J. Neurosci. Res. 18, 28-36. Felten, D. L., & Felten, S. Y. (1988). Sympathetic noradrenergic innervation of immune organs. Brain Behav.
Immun.
2, 293-300.
Felten, S. Y., Lorton, D. L., & Felten, D. L. (1989). Substance P (SP) innervation of the rat thymus. Sot. Neurosci. Abstr. 15, 714. Goodwin, J. S., & Webb, D. R. (1980). Regulation of the immune response by prostaglandins. Clin. Immunol.
Immunopharmacol.
15, 106-122.
Goodwin, J. S., Bromberg, S., Staszak, C., Kaszubowski, P. A., Messner, R. P., & Neal, J. F. (1981). Effect of physical stress on sensitivity of lymphocytes to inhibition by prostaglandin E,. J. Immunol.
127, 518-522.
Hartung, H. P. (1988). Activation of macrophages by neuropeptides. Brain Behav. Immun. 2,275-281. Irwin, M. (1988). Depression and immune function. Stress Med. 4, 95-103. Jesberger, J. A., & Richardson, J. S. (1985). Animal models of depression: Parallels and correlates to severe depression in humans. Biol. Psychiatry 20, 764-784. Kavelaars, A., Ballieux, R. E., & Heijnen, C. J. (1989). The role of IL-1 in the corticotropin-releasing factor and arginine vasopressin-induced secretion of immunoreactive B-endorphin by human peripheral blood mononuclear cells. J. Immunol. 142, 2338-2342. Kaye, J., Gillis, S., Mizel, S. B., Schevach, E. M., Malek, T. R., Dinarello, C. A., Lachman, L. B., & Janeway, C. A. (1984). Growth of a cloned helper T cell line induced by a mono-clonal antibody
290
CHENG ET AL.
specific for the antigen receptor: Interleukin 1 is required for the expression of receptors for interleukin 2. J. Immunol. 133, 1339-1345. Keller, S. E., Weiss, J. M., Schleifer, S. J., Miller, N. F., & 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., Schleifer, S. J., Liotta, A. S., Bond, R. N., Farhoody, N., & Stein, M. (1988). Stressinduced alterations of immunity in hypophysectomized rats. Proc. Natl. Acad. Sci. USA 85, 9297-930 1. Kiecolt-Glaser, J. K., Gamer, W., Speicher, C., Penn, G. M., Holliday, J. E., & Glaser, R. (1984). Psychosocial modifiers of immunocompetence in medical students. Psychosom. Med. 46, 7-14. Knudsen, P. J., Dinarello, C. A., & Strom, T. B. (1986). Prostaglandins post-transcriptionally inhibit monocyte expression of interleukin 1 activity by increasing intracellular cyclic adenosine monophosphate. J. Immunol. 137, 3189-3194. Kronfol, Z., Silva, J. Jr., &eden, J., Dembinski, S., Gardner, R., h Carroll, B. (1983). Impaired lymphocyte function in depressive illness. Life Sci. 33, 241-247. Kurt-Jones, E. A., Beller, D. I., Mizel, S. B., & Unanue, E. R. (1985). identification of a membraneassociated interleukin 1 in macrophages. Proc. Natl. Acad. Sci. USA 82, 1204-1208. Laudenslager, M. L. (1987). Psychosocial stress and susceptibility to infectious illness. In E. Kurstak, Z. J. Liposki, & P. V. Morozov (Eds.), Viruses, immunity, and mental disorders. Plenum: New York. Laudenslager, M. L., Ryan, S. M., Drugan, R. C., Hyson, R. L., & Maier, S. F. (1983). Coping and immunosuppression: Inescapable but not escapable shock suppresses lymphocyte proliferation. Science 221, 568-570. Leizer, T., Clarris, B. .I., Ash, P. E., van Damme, J., Saklatvala, J., & Hamilton, J. A. (1987). Interleukin 1B and interleukin-la stimulate the plasminogen activator activity and prostaglandin E, levels of human synovial cells. Arth. Rheum. 30, 562-566. Levy, E. M., & Black, P. H. (1988). Depression and the immune system. Clin. Immunol. Newsl. 9, 4-6. Lorton, D., Bellinger, D. L., Felten, S. Y., & Felten, D. L. (1989). Substance P (SP) and calcitonin gene-related peptide (CGRP) innervation of the rat spleen. Sot. Neurosci. Abstr. 15, 714. Morley, J. E., Kay, N., & Solomon, G. F. (1989). Opioid peptides, stress, and immune function. In Y. Tache, J. E. Morley, & M. R. Brown (Eds.), Neuropeptides and stress, p 222-234. SpringerVerlag: New York. Mormede, P., Dantzer, R., Michaud, B., Kelley, K. W. & Le Moal, M. (1988). Influence of stressor predictability and behavioral control on lymphocyte reactivity, antibody responses and neuroendocrine activation in rats. Physiol. Behav. 43, 577-583. Okimura, T., & Nigo, Y. (1986). Stress and immune responses. I. Suppression of T cell function in restraint-stressed mice. Japan. J. Pharmacol. 40, 505-511. Okimura, T., Satomi-Sasaki, Y., & Ohkuma, S. (1986a). Stress and immune responses. II. Identitication of stress-sensitive cells in murine spleen cells. Japan. J. Pharmacol. 40, 513-525. Okimura, T., Ogawa, M., & Yamauchi, T. (1986b). Stress and immune responses. III. Effect of restraint stress on delayed type hypersensitivity (DTH) response, natural killer (NK) activity and phagocytosis in mice. Japan. J. Pharmacol. 41, 229-235. Parhar, R. S., & Lala, P. K. (1988). Prostaglandin EZ-mediated inactivation of various killer lineage cells by tumor-bearing host macrophages. J. Leuk. Biol. 44, 474-484. Pavlidis, N., & Chirigos, M. (1980). Stress-induced impairment of macrophage tumoricidal function. Psychosom. Med. 42, 47-54. Poutsiaka, D. D., Schroder, E. W., Taylor, D. D., Levy, E. M., & Black, P. H. (1985a). Membrane vesicles shed by murine melanoma cells selectively inhibit the expression of Ia antigen by macrophages. J. Immunol. 134, 131144. Poutsiaka, D. D., Taylor, D. D., Levy, E. M., & Black, P. H. (1985b). Inhibition of recombinant interferon-y-induced Ia antigen expression by shed B16 FlO melanoma cell membrane vesicles. J. Immunol. 134, 145-150. Prowse, K. R., & Baumann, H. (1989). Interleukin-1 and interleukind stimulate acute-phase protein production in primary mouse hepatocytes. J. Leuk. Biol. 45, 55-61.
IMMUNOSUPPRESSION
AND COLD WATER
STRESS
291
Schleifer, S. J., Keller, S. E., Came&o, M., Thornton, J. C., & Stein, M. (1983). Suppression of lymphocyte stimulation following bereavement. JAMA 250, 374-377. Schleifer, S. J., Keller, S. E., Siris, S. G., Davis, K. L., & Stein, M. (1985). Depression and immunity. Lymphocyte function in ambulatory depressed patients, hospitalized schizophrenic patients, and patients hospitalized for hemiorrhaphy. Arch. Gen. Psychiatry 42, 129-133. Snider, M. E., Fertel, R. H., & Zwilling, B. S. (1982). Prostaglandin regulation of macrophage function: Effect of endogenous and exogenous prostaglandins. Ceil. Immunol. 74, 234-242. Snyder, D. S., & Unanue, E. R. (1982). Corticosteroids inhibit murine macrophage Ia expression and interleukin 1 production. J. Zmmunol. 129, 1803-1805. Steeg, P. S., Moore, R. N., Johnson, H. M., & Oppenheim, J. J. (1982). Regulation of murine macrophage Ia antigen expression by a lymphokine with immune interferon activity. J. Exp. Med. 156, 1780-1793. Stein, M. (1989). Stress, depression, and the immune system. J. Clin. Psychiatry 50, 3-2. Stein, M., Keller, S. E., & Schleifer, S. J. (1985). Stress and immunomodulation: The role of depression and neuro-endocrine function. J. Immunol. 135, 827s-833~. Struhar, D. J., Harbeck, R. J., Gegen, N., Kawada, H., & Mason, R. J. (1989). Increased expression of class II antigens on the major histocompatibility complex on alveolar macrophages and alveolar type II cells and interleukin-1 (IL-l) secretion from alveolar macrophages in an animal model of silicosis. Clin. Exp. Immunol. 77, 281-284. Suttles, J., Carruth, L. M., & Mizel, S. B. (1989). Detection of IL-la and IL-1E in the supematants of paraformaldehyde-treated human monocytes: Evidence against a membrane form of IL-l. J. Immunol. 144, 170-174. Tatfet, S. T., & Russell, S. W. (1981). Macrophage-mediated tumor cell killing: Regulation of expression of cytolytic activity by prostaglandin E. J. Immunol. 126, 424-427. Uhing, R. J., & Adams, D. 0. (1989). Molecular events in the activation of murine macrophages. Agents Actions 26, 9-14. Unanue, E. R. (1981). The regulatory role of macrophages in antigenic stimulation. II. Symbiotic relationship between lymphocytes and macrophages. Adv. Immunol. 31, I-136. Unanue, E. R., & Allen, P. M. (1987). The basis for the immunoregulatory role of macrophages and other accessory cells. Science 236, 551-557. Veldhuis, H. D., Croiset, G., Ballieux, R. E., He&en, C. J., & DeWied, D. (1987). Emotional stress and modulation of immunological reactivity. In D. Nerozzi, F. K. Goodwin, & E. Costa (Eds.), Hypothalamic dysfunction in neuro-psychiatric disorders, pp. 237-245. Raven Press: New York. Weis, J. M., Sundar, S. K., Becker, K. J., & Cierpial, M. A. (1989). Behavioral and neural irdluences on cellular immune responses: Effects of stress and interleukin-1. J. Clin. Psychiatry 50, 43-55. Workman, E. A., & Via, M. F. (1987). T-lymphocyte polyclonal proliferation: Effects of stress response style on medical students taking national board examinations. Clin. Zmmunol. Zmmunopathol. 43, 308-313. Received February 20, 1990