Neurochemical, endocrine and immunological responses to stress in young and old Fischer 344 male rats

Neurobiologyof Aging, Vol. 11, pp. 139-150. ©Pergamon Press plc, 1990. Printed in the U.S.A.

0197-4580/90 $3.00 + .00

Neurochemical, Endocrine and Immunological Responses to Stress in Young and Old Fischer 344 Male Rats S T A N L E Y A . L O R E N S , 1 N O R I O H A T A , R O B E R T J. H A N D A , L O U I S D. V A N DE K A R , M A R I A N N E G U S C H W A N , J O A N N A G O R A L , J O H N M . L E E , M A R G A R E T E. H A M I L T O N , C Y N T H I A L. B E T H E A * A N D J O H N C L A N C Y , JR.

Departments of Pharmacology and Anatomy, Stritch School of Medicine, Loyola University of Chicago 2160 South First Avenue, Maywood, IL 60153 and *Oregon Regional Primate Research Center, Beaverton, OR R e c e i v e d 27 F e b r u a r y 1989; A c c e p t e d 6 D e c e m b e r 1989

LORENS, S. A., N. HATA, R. J. HANDA, L. D. VAN DE KAR, M. GUSCHWAN, J. GORAL, J. M. LEE, M. E. HAMILTON, C. L. BETHEA AND J. CLANCY, JR. Neurochemical, endocrine and immunological responses to stress in young and old Fischer 344 male rats. NEUROBIOL AGING 11(2) 139-150, 1990.--Two experiments were performed. In the first, a 20 min conditioned emotional response (CER) paradigm was used to compare the neurochemical, endocrine and immunological responses to stress of 7and 22-month-old Fischer 344 (F344) male rats. In the second, corticosterone levels 20 min following ether stress, and regional brain type I and II corticosterone receptor densities were examined using 7- and 17.5-month-old F344 male rats. Dopamine (DA) metabolism in old nonstressed rats was significantly reduced in the medial frontal cortex, neostriatum, nucleus accumbens and hypothalamus, but not in the amygdala. The CER procedure, nevertheless, increased medial frontal cortical, nucleus accumbens and amygdaloid DA turnover in both the young and old rats. The young and old nonstressed rats did not evidence differences in norepinephrine (NE) and serotonin (5-HT) concentrations. However, stress resulted in a decrease in medial frontal cortex NE content in the young but not in the old rats. In contrast, the CER procedure resulted in increased medial frontal cortical 5-hydroxyindoleacetic acid (5-HIAA) and hypothalamic 5-HT levels in old but not in young animals. These observations suggest age-related differences in the response of central NE and 5-HT systems to stress. Ether and the CER procedure led to exaggerated corticosterone responses in the old rats (17.5 and 22 month, respectively). Hippocampal type I but not type II corticosterone receptors were decreased by 47% in the 17.5-month-old rats. Thus, age-related changes in hippocampal corticosterone receptor types do not occur in unison, and the exacerbated corticosterone response to stress precedes the reported down-regulation of hippocampal type II corticosterone receptors in aged rats. Age-related changes were not observed in the concentrations of corticosterone receptors in other brain regions, or in the prolactin response to stress. The old rats, however, evidenced a reduction in the availability of the renin substrate, angiotensinogen, and in stress-induced renin secretion. Immune function was impaired in the old nonstressed rats, and further compromised by exposure to the CER procedure. In comparison to the young control rats, the old nonstressed rats showed an increased percentage of splenic large granular lymphocytes, reduced splenic natural killer cytotoxicity, and impaired Con-A-stimulated splenic T lymphocyte proliferation. Reductions in T splenic cell proliferation and natural killer cytotoxicity were observed in the young rats subjected to the CER paradigm, but not to the same extent as in the old rats. These observations indicate that aging male F344 rats evidence major alterations in basal central monoamine, endocrine and immune functions, and an increased sensitivity of these systems to stress. Aging Amygdala Angiotensinogen Conditioned emotional response paradigm Corticosterone Corticosterone receptors Dopamine Fischer 344 male rats Hippocampus Hypothalamus Immune function Large granular lymphocytes Natural killer cells Medial frontal cortex Neostriatum Norepinephrine Nucleus accumbens Prolactin Renin Serotonin Stress T lymphocytes

THE aging rat has been reported to show significant decreases in the basal levels of circulating prolactin and renin, increased plasma corticosterone concentrations, and a slower recovery in the corticosterone response to stress (9, 14-16, 52, 60--62). Aging animals, furthermore, evidence progressive neuronal cell loss, especially in the hippocampal formation (7, 14, 21, 31). These neuropathological changes have been ascribed to the toxic effects

of age-related increases in circulating corticosterone levels (32, 33, 63). Age-related reductions in central nervous system (CNS) dopamine (DA) and norepinephrine (NE) levels, but not in serotonin (5-HT) concentrations, have been reported (50, 51, 69). Moreover, age-related decreases in DA, NE and 5-HT metabolism and/or receptor sensitivity have been observed (50, 51, 69). Concomitantly, aging rats exhibit a decrease in a number of

IRequests for reprints should be addressed to Dr. S. A. Lorens, Laboratory of Behavioral Pharmacology, Loyola University Medical Center, 2160 South First Avenue (Building 135), Maywood, IL 60153.

139

LORENS El' AL.

14(1

immune functions (2, 3, 38, 66). Such decreases can affect not only major histocompatibility complex (MHC) restricted T-cell activities (38), but non-MHC restricted natural killer (NK) cell activities (3). Various types of stressors, including immobilization, electric footshock and conditioned fear, have been reported to enhance 5-HT (17), NE (29) and DA (59) metabolism, as well as to stimulate corticosterone, prolactin and renin secretion (47) and to suppress immune function (10, 19, 30, 36, 64). Few studies, however, have compared the effects of stress on young and old rats, and none have simultaneously compared the neurochemical, endocrine and immunological responses to stress in young and old rats. It is now well established that there is a dynamic interaction between the immune and neuroendocrine systems (4), and that neuroendocrine systems play an important role in the aging process (41). The objective of the present study was to compare basal and stress-induced changes in neurochemical, endocrine and immune parameters in young and old Fischer 344 (F344) male rats. In addition, the corticosterone response of young and old F344 rats to ether stress was examined, and potential differences in regional CNS type I and type II corticosterone receptors investigated. The results from these experiments suggest important age-related differences in basal neurochemical, endocrine and immune functions, and their responses to stress. METHOD

Animals Barrier-reared young (5.5 month) and old (16 and 20.5 month) Fischer 344 (F344) male rats (Harlan Sprague-Dawley Inc., Indianapolis), weighing 375-450 g at the time of arrival, were used. The animals were housed individually in conventional "shoe box" plastic cages (44 cm long x 20 cm high × 23 cm wide), and maintained on a 12-hr light/dark cycle (lights on at 07:00 hr) in a temperature- (20-22°C) and humidity- (52-55%) controlled AAALAC approved facility. Food and water were available ad lib.

Experimental Protocols Experiment 1. Conditioned emotional response (CER) paradigm. The animals were handled 4-5 mirdday and weighed twice/week for six weeks. The rats then were subjected to a CER paradigm, and sacrificed as described below. Four groups (n = 8/group) of rats were analyzed: Young Control, Young Stress, Old Control, and Old Stress. At the time of sacrifice the young and old animals were 7 and 22 months of age, respectively. Conditioning procedure. The CER was elaborated in a clear Plexiglas chamber (10 cm 3) with a grid floor composed of steel rods (2.0 mm diameter) spaced 1.3 cm center-to-center. A LVE Model No. 1407 shocker was used to deliver scrambled constant current shock through the grid floor. The chamber was located in a sound attenuated cubicle (50 cm wide x 55 cm high x 35 cm deep) which contained a one-way mirror (27 x 32 cm) mounted in the front wall. Illumination was provided by a 7.5 W bulb in the ceiling of the cubicle. A ventilation fan mounted in the upper rear wall of the cubicle provided a background white noise. The testing apparatus was located in a room immediately adjacent to the animals' living quarters. The conditioning and tissue collection procedures were conducted between 09:00-13:30 hr. Each animal was transported to the experimental room in its home cage, then placed individually into the experimental chamber 20 min/day for 4 consecutive days. The stressed rats received 10 sec (1 mA) inescapable footshock at the end of the 20 min session. Control rats were never shocked. On Days 1-3 each animal was removed from the chamber after the 20

min session, placed in its home cage, and returned to its living quarters. On Day 4, no shock was administered and the animals were sacrificed immediately after the 20 rain session in a room located 2.0 m from the testing room. This procedure was utilized in order to avoid exposing the stressed animals to nociceptive stimuli on the day of sacrifice. The objective was to train the animals to expect an aversive footshock after being placed in the experimental chamber. That this was accomplished was evident by the third day of conditioning. On both the third and fourth days, the stressed rats micturated, defecated and remained in a crouched posture virtually throughout the 20 rain session. In contrast, the control rats slowly moved about the chamber, exploring the walls and floor. These animals, furthermore, rarely urinated or defecated. Animal sacrifice and tissue collection. At the end of the session on Day 4, the rats were euthanized instantaneously using a guillotine. This method of sacrifice was necessary since sedative and anesthetic drugs are known to affect the neurochemical and endocrine measures which were obtained postmortem. Following decapitation, the brain was removed and dissected over ice using a modification of the method detailed by Heffner et al. (22). Accordingly, serial 2.0 mm coronal sections were obtained, and the medial frontal cortex, neostriatum, nucleus accumbens, amygdaloid complex, hippocampal formation, and hypothalamus dissected as described by Herman et al. (26). These brain regions as well as the spinal cord (2.5 cm of the cervicothoracic region) were frozen on dry ice, then stored at - 80°C in polyethylene tubes prior to assay (within 3 months). Trunk blood was collected in centrifuge tubes containing a 0.5 ml solution of 0.3 M ethylenediaminetetraacetic acid (EDTA; pH 7.4). The blood was centrifuged at 1000 x g for 20 min at 4°C and the plasma saved in individually marked tubes. The plasma from each rat was divided into five aliquots: 1) 1.0 ml for the determination of plasma renin activity (PRA); 2) 0.2 ml for the determination of plasma renin concentration (PRC); 3) 0. l ml for the assay of angiotensinogen; 4) 0.04 ml for the determination of corticosterone levels; and, 5) 0.5 ml for prolactin analysis. The spleen was harvested using a sterile field, and spleen cell suspensions were prepared immediately as previously described (8). Experiment 2. Ether stress and corticosterone receptors. The results of the first experiment indicated that old (22 month) F344 male rats exhibit an exaggerated corticosterone response to the CER procedure. We thereby decided to investigate whether old rats would show an enhanced corticosterone response to a different type stressor, and whether the increased corticosterone secretion occurred in slightly younger animals (17.5 month). We also wanted to determine if the enhanced corticosterone response of the old rats was associated with a significant change in regional CNS type I and/or II corticosterone receptors. Thus, we first measured the corticosterone response of young and old animals to ether stress. Subsequently, the animals were adrenalectomized and type I and/or type II corticosterone receptors were measured in several brain regions. Adrenalectomy was required in order to measure total receptor concentrations. The stress and tissue collection procedures were conducted between 09:30-12:30. The animals were handled 4-5 min/day and weighed twice/ week for six weeks. The young (7 months of age) and old (17.5 month) rats then were stressed by placing them individually into a covered desiccant jar (20 cm high x 20 cm diameter) containing ether soaked cotton balls below the wire mesh floor. Following the induction of anesthesia (approximately 1.5 min), the animals were removed and their tails clipped with a razor blade. Approximately 1.0 ml of blood was collected from the tail vein into 1.5 ml conical centrifuge tubes containing 50 ILl of 0.3 M EDTA. Blood

RESPONSE TO STRESS IN AGED RATS

sampling was completed within 3.0 min from the time of In'st exposure to ether. A second sample of blood was collected 20 min later using an identical procedure. Pre- and poststress plasma corticosterone levels were determined by radioimmunoassay (RIA) as described below. Either immediately or 2-3 days following the above procedure, the rats were anesthetized with ether and their adrenal glands removed bilaterally using a dorsal approach. The rats were maintained with 0.9% saline in their drinking water and sacrificed by decapitation 24 hr postadrenalectomy. Trunk blood was collected in tubes containing 500 ixl of 0.3 M EDTA for determination of plasma corticosterone levels. The brains were dissected over ice as described above. The medial frontal cortex, septal area, hippocampus, hypothalamus and midbrain raphe (dorsal and median raphe nuclei) were obtained and immediately processed for type I and type II corticosterone receptor analysis as detailed below. Biochemical Analyses Determination of CNS monoamines and their metabolites. HPLC (electrochemical detection) analyses were performed using a Bioanalytic Systems (BAS; West Lafayette, IN) LC-4 amperometer and TL-5 glassy carbon electrode. Samples were injected automatically every 15-20 min using a Beckman Model 504 Autosampler (20-p,1 injection loop) and a Beckman Model 110 B Solvent Delivery System. The analytical column was a reverse phase 5.0 p,m Ultrasphere ODS 4.6 mm × 15 cm column (Rainin Instruments Co. Woburn, MA), and was protected from tissue contaminants by a 4.0 cm × 4.6 mm Ultrasphere ODS BAS precolumn. The concentrations of the endogenous amines and their acid metabolites were analyzed by a Beckman 427 Integrator by measuring the areas under their curves relative to that of the internal standards and the external standard line. Levels of norepinephrine (NE), dopamine (DA), serotonin (5-HT), 3,4-dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), and 5-hydroxyindoleacetic acid (5-HIAA) were determined using a procedure based on the methods described by Mefford (39). The frozen samples were weighed to the nearest mg and transferred to polyethylene microcentrifuge tubes at 4°C. Each sample then was sonicated at 4°C for 30 sec in 100 ~1 0.1 N perchloric acid (PCA) containing 1.0 mM EDTA, 2.0 mM sodium metabisulfite (an antioxidant), 50 ng of the internal standard N-methyl-5-hydroxytryptamine (NMe5-HT), and 10 ng of the internal standard 3,4-hydroxybenzylamine (DHBA). Following sonication, 50 p~l of chloroform was added to each tube and vortexed. The samples then were centrifuged in an Eppendorf 542 Microfuge at 10,000 × g at 4°C for 15 min. For CNS areas that did not require concentration of amines and metabolites (caudate-putamen, nucleus accumbens, hippocampus, hypothalamus, and spinal cord), 20 p,1 of the supernatant containing 50 ng of the internal standard was injected into the system. The mobile phase contained 0.15 M monochloroacetic acid (pH 3.2), 1.0 mM EDTA, 16% methanol and 100 mg/1 octanylsulfonic acid (OSA). The flow rate was 0.9 ml/min and +0.74 V was applied across the LC-5 electrode. The mobile phase was passed through 0.2 mm filters, and degassed under vacuum. Retention times were: DA (5.2 min), DOPAC (6.5 min), 5-HIAA (9.9 min), 5-HT (11.2 min), NMe5-HT (13.3 min), and HVA (15.2 min). For brain regions that required concentration of the monoamines and their metabolites (medial frontal cortex and amygdala), an alumina extraction was performed. The supernatant (80 ml) from the 10,000 × g centrifugation (4°C for 15 min) was suspended with 5.0 mg of activated alumina and 35 p,l 3.0 M Tris-HCl (pH 8.4) buffer. The suspension was shaken vigorously for 10 min then centrifuged at 10,000 × g for 2.0 min at 4°C. The supernatant, containing 40 Ixl of 0.1 N PCA, was put in 0.8 ml

141 minivials for injection (20 Ixl). This supernatant portion contained 5-HT, NMe5-HT, and 5-HIAA. The mobile phase consisted of 0.15 M monochloroacetic acid (pH 3.2) containing 1.0 mM EDTA, 20% methanol and 10 mg/l OSA. The samples were analyzed as detailed above. The retention times were: 5-HT (5.7 min), NMe5-HT (6.9 rain) and 5-HIAA (9.9 rain). In a separate extraction, the resulting alumina pellet was washed twice by the addition of 150 p,1 double distilled water followed by swift vortexing and 5.0 sec centrifugation. After the second wash, 50 Ixl of 0.5 N HCI, containing 2.0 mM sodium metabisulfite, was added to the alumina. The samples were shaken vigorously for 1.0 min and centrifuged at 5,000 >¢g for 5.0 sec. The supernatant was aspirated and placed in minivials for injection (20 ixl). This supernatant contained NE, DA and DOPAC with the internal standard DHBA. The mobile phase contained 0.15 M monochloroacetic acid (pH 3.2), 1.0 mM EDTA, 15% methanol and 750 mg/1 OSA. The retention times were: NE (5.1 min), DOPAC (6.4 min), DHBA (8.6 min), and DA (12.4 min). The limit of sensitivity was 10-15 ng substance/g tissue, and the intra- and interassay variability has been estimated to be 3.0% and 6.0%, respectively. Endocrine Assays Plasma corticosterone RIA. The corticosterone RIA was performed on unextracted plasma (5.0 and 10.0 p,1) in which binding proteins had been heat denatured using procedures and antisera from Radioassay Systems Laboratories (Carson, CA), as described previously (70). The intra- and interassay variabilities have been calculated to be 4.5% and 11.9%, respectively. CNS corticosterone receptor. Type I and/or type II corticosterone receptor concentrations were analyzed in the regional CNS samples immediately following their dissection. The tissues were homogenized in ice-cold TEGMD buffer (10 mM TRIS, 12.5 mM EDTA, 10% glycerol, 25 mM molybdate, 1.0 mM dithiothreitol; pH = 7.4) in Dounce homogenizers (Wheaton Inc., Millville, NJ). A cytosolic preparation was made by centrifugation (4°C) of the homogenate at 106,000x g for 15 min in a Sorval OTD 55-B ultracentrifuge using a TFT 80.4 rotor. Aliquots of cytosol were incubated with increasing amounts of 3H-corticosterone (0.1-25 nM; 1,2,6,7-3H-corticosterone, 087.1 Ci/mM; NEN, Wilmington, DE). Incubations were carded out at 0°C for 18-24 hr. Parallel incubation tubes containing radioinert RU28362 (4.0 p,M), or RU28362 and dexamethasone (4.0 IxM), in addition to 3Hcorticosterone were run to estimate binding to type II and type I corticosterone receptors, respectively. Since RU28362 only binds to the type II receptor, the binding of 3H-corticosterone in the presence of RU28362 subtracted from total binding was used to estimate binding to type II receptor. Similarly, because dexamethasone binds to both type I and II receptors, but not to corticosterone binding globulin, the concentration of type I receptors could be calculated by subtracting 3H-corticosterone binding in the presence of radioinert dexamethasone and RU28362 from 3H-corticosterone binding in the presence of RU28362 alone. Following incubation, bound and free ligand were separated on miniature Sephadex LH-20 columns as previously described (23). Bound 3H-corticosterone was determined by the addition of 3.0 ml of Atomlight (NEN). Samples were counted in a Beckman LS 7000 liquid scintillationcounter. All receptor data are expressed as femtomoles bound per mg protein. Total cytosol protein was determined by the method of Lowry et al. (35). Plasma prolactin RIA. The prolactin RIA was performed with reagents provided by the NIADDK. The anti-rat prolactin serum S-8 was used at a dilution of 1:5,000, as described previously (71). We have calculated that the intraassay variability is 6.8%. Plasma renin activity (PRA). PRA was measured by RIA

LORENS ET AL.

142

for generated angiotensin II (AII) as detailed previously (20,67). The intra- and interassay variabilities have been estimated to be 4.4% and 7.4%, respectively (57,71). Plasma renin concentration (PRC). The measurement of PRA depends on the concentration of renin and renin substrate (angiotensinogen) in plasma. The normal angiotensinogen concentration is near or less than the Kin, and thus is much less than the concentration required to generate AI at maximum velocity (Vmax). We measured PRC by adding a saturating concentration of exogenous renin substrate to each plasma sample. The renin substrate was plasma obtained from nephrectomized rats which received dexamethasone (0.2 mg/rat, IP) 24 hr before sacrifice. To each plasma sample (0.2 ml), 0.1 ml of 0.5 M phosphate buffer (pH 6.0), and 5.0 txl of PMSF and of 8hydroxyquinoline (to a final concentration of 2.5 mM and 3.4 mM, respectively), was added. The mixture was incubated at 37°C for 1.0 hr, and the reaction stopped by immersing the tubes in boiling water for 3.0 min (34). The RIA of AI was conducted as described above. Plasma angiotensinogen. A saturating concentration of renin (from kidney homogenates) was added to the plasma samples to generate angiotensin I (AI) from endogenous renin substrate. The plasma samples (0.1 ml) received 0.1 ml of 0.5 M phosphate buffer pH 6.0, 0.1 ml renin (a 1:1000 dilution of a 1.0 g kidney homogenized in 1.0 ml water), and 5.0 txl each of PMSF and 8-hydroxyquinoline. The mixture was incubated at 37°C for 1.0 hr. The reaction was stopped by immersion of the tubes in a boiling water bath for 3.0 min. The RIA of AI was performed as described above.

Immunological Analyses Solutions. Hank's balanced salt solution (HBSS) was obtained from KC Biological (Lenexa, KS). Culture media consisted of RPMI 1640 supplemented with 10% FCS, 100 units/ml penicillin, 100 units/ml streptomycin, 2.0 mM glutamine (MA Bioproducts, Walkersville, MD), as described by Oehler et al. (45). This will be referred to as "complete medium." Target cells and chromium labeling. Mouse YAC- 1 lymphoma target cells were maintained as cultures in RPMI t640 medium supplemented with 10% fetal calf serum (KC Biological), according to Oehler et al. (46). The cells (1 x 106) were suspended in 0.5 ml of complete medium. NaSICrO4 (100 ~Ci in 0.9% saline; New England Nuclear) was added to the target cells and the suspension incubated for 1.0 hr at 37°C with frequent gentle agitation. The cells then were washed five times and suspended to a concentration of 5 x 104 cells/ml in complete medium. Evaluation of cell morphology. Cytocentrifuged Giemsa stained spleen cell preparations were analyzed microscopically. Percent large granular lymphocytes (LGL) were evaluated by counting 200--400 mononuclear cells. Cytotoxicity assay. The 51Cr release assay was performed as detailed previously (8). Suspensions of cell preparations at concentrations to provide effector:target (E:T) ratios of 6:1, 12:1, 25:1 and 50:1 were added to YAC-1 target cells in 96 well round bottom sterile microtiter plates. For an E:T ratio of 50:1, for example, 0.1 ml of a suspension of 2.5 x 10 6 effectors cells/ml was added to 0.1 ml of target cells at 5 x 104 cells/ml. Controls for the spontaneous release of 51Cr were YAC-1 target cells plus 0.1 ml of complete medium. Total release was determined with YAC-1 cells and 0.5% Triton X-100 (Sigma, St. Louis, MO). Spontaneous release always was less than 15% of total release. All determinations were performed in triplicate. All plates were incubated for 4.0 hr at 37°C in a humidified incubator with 5.0% CO 2. Supernatants were harvested after centrifugation.at 200 x g for 5.0 min. The supernatant (100 txl) from each well was removed

and analyzed for S~Cr release using a gamma counter. Calculation of cytotoxiciO, and lytic units. Percent cytotoxicity was determined by the formula: (cpm experimental - cpm spontaneous release) % Cytotoxicity = (cpm total release - cpm spontaneous release) x 100 The "experimental" was the dilution of test or control harvested cells being studied. All control and test values were the means of triplicate determinations. All standard errors typically were less than 5.0%. The % cytotoxicity Ibr each of the E:T ratios was plotted in order to establish a lytic unit (LU). One LU was defined as the number of leukocytes in 10 6 effectors necessary to lyse 20% of 5 X 103 targets [LU(20%)/107]. Interleukin-2 (IL-2) activated natural killer (NK) cells. Human recombinant interleukin-2 (rlL-2), containing t.0 x 106 units/ml, was generously provided by Cetus Corporation (Emoryville, CA). Spleen cells were cultured with 10 6 units/ml of rlL-2 in complete medium. The lymphoid cells were incubated in T-75 flasks (Coming) lying flat at a density of 2 x 10 6 viable cells/ml complete medium in 5.0% COz/95% air at 37°C for 24 hr. Cytotoxicity was determined using YAC-1 targets as described above, and according to Vujanovic et al. (73). Mitogen stimulation of T-lymphocytes. Spleen cell suspensions were added to 96 well microtiter plates in 200 ~tl aliquots at concentrations ranging between 1-2 x 105 in complete medium in the presence of 4.0 ixg/ml Con-A (Pharmacia, Uppsala, Sweden) for 72 hr. The cells were pulsed for 4.0 hr with 1.0 txCi of 3H-thymidine (New England Nuclear), harvested, and thymidine incorporation determined as detailed previously (8).

Data Analysis Data reduction and analyses were performed by an IBM PC-XT utilizing the following statistical packages: SYSTAT, version 2.1 (Systat Inc., Evanston, IL), and PC ANOVA, version 1.0 (Human Systems Dynamics, Northridge, CA). The data from Experiment 1 first were subjected to an analysis of variance (ANOVA) using a two-factor (age x treatment) factorial design, followed, if warranted, by a Newman-Keuls' Multiple Range Test for post hoc comparisons (75). The corticosterone response to ether stress (Experiment 2) was analyzed using a treatments-by-treatmentsby-subjects (repeated measures:two factors) design. The corticosterone receptor data were analyzed using a t-test for independent means (two-tail). RESULTS

Experiment 1. Conditioned Emotional Response Animals. All rats completed the study. During the course of their stay in the laboratory the animals maintained their body weights and appeared healthy. At the time of sacrifice the rats weighed between 361-443 g. Necropsy did not reveal any gross pathological conditions in either the young or old rats. Neurochemistry. Medial frontal cortex (MFC). The CER procedure resulted in a significant age effect, F(1,25)=6.6, p<0.02, on MFC NE content. The age × stress interaction, although suggestive, was not statistically significant, F(1,25)=3.5, p<0.07. The post hoc analysis showed that stress reduced MFC NE levels in the young but not in the old rats (Table 1). This observation suggests that stress affects MFC NE metabolism in young but not in old rats. No significant group differences in MFC DA concentrations

RESPONSE TO STRESS IN AGED RATS

143

TABLE 1 MEDIAL FRONTAL CORTICAL, AMYGDALOIDAND HYPOTHALAMIC CATECHOALMINE AND METABOLITE CONCENTRATIONS IN STRESSED AND NONSTRESSED YOUNG (7 MONTH) AND OLD (22 MONTH) F344 MALE RATS Group

Brain Region Medial Frontal Cortex

Age

Procedure

NE

Young

Control

389 ±20

Young

Stress

340:~ -+17

Old

Control

Old

Stress

DA

Amygdala

Hypothalamus

DOPAC

NE

DA

DOPAC

DA

DOPAC

114 ±7

55 ±2

461 ±16

313 ±22

53 -+4

357 -+27

96 +8

111 ±10

73* ±5

422 ±24

339 ±19

69* ±5

452:~ ±28

90 ---8

401 ±22

100 ±7

38;t ±3

474 ±30

288 ±27

53 ±5

295 ±17

67t ±7

417 ±9

105 -+6

64 ±5

438 -+21

388* ±23

66* -+3

360 -+30

78 ±5

Mean ± S.E.M. (n = 6--8/group) ng/g wet weight. *p<0.01 (frontal cortex) or p<0.05 (amygdala) compared to age-matched Control group. tp<0.05 compared to both Young groups. $p<0.05 compared to the other three groups. Abbreviations: NE = Norepinephrine. DA = Dopamine. DOPAC = 3,4-Dihydroxyphenylacetic acid.

were found (Table 1). However, significant age, F ( 1 , 2 5 ) = 9 . 0 , p < 0 . 0 1 , and stress, F ( 1 , 2 5 ) = 2 8 . 0 , p < 0 . 0 0 0 1 , effects were obtained for MFC DOPAC levels (Table 1). DOPAC concentrations in the Old Control group were significantly less than in the other three groups (31% lower than in the Young Control group). The CER procedure, however, significantly increased DOPAC levels in both the young and old rats, relative to their age-matched controls (Young, 38%; Old, 68%). No significant group differences were found in MFC 5-HT concentrations. A significant, F ( 1 , 2 7 ) = 7 . 7 , p < 0 . 0 1 , effect of age on 5-HIAA levels, however, was observed. As seen in Table 2, this effect was due to the elevated MFC 5-HIAA concentration in the Old Stress group (20% higher than in the Young Control group). In sum, DA metabolism was significantly reduced in the MFC of the old nonstressed rats. The CER procedure led to a significant increase in MFC DA turnover in both young and old animals. Stress also induced a significant decrease in MFC NE levels in the young but not in the old rats. In contrast, stress provoked a significant increase in MFC 5-HIAA levels in the old but not in the young rats. Amygdala. The CER procedure produced a significant, F ( 1 , 2 3 ) = 7.4, p < 0 . 0 1 , increase in amygdaloid DA concentration (Table 1). The post hoc analysis showed that this effect was due to the stress-induced increase in DA concentration in the old rats (35% higher than in their age-matched controls). Stress also increased (25-30%) amygdaloid DOPAC concentrations in both the young and old rats, F(1,23) = 10.8, p < 0 . 0 0 3 . No group differences were found in amygdaloid NE (Table 1), 5-HT or 5-HIAA (Table 3) levels. Thus, stress increases amygdaloid DA turnover in young rats, and apparently similar changes in old rats, without concomitant alterations in NE or 5-HT metabolism. Hypothalamus. The CER procedure produced significant age, F ( 1 , 2 7 ) = 9 . 0 , p < 0 . 0 0 6 , and stress, F ( 1 , 2 7 ) = 9 . 7 , p < 0 . 0 0 4 , effects on hypothalamic DA concentrations (Table 1). Post hoc analysis revealed that these effects were due to the increased (27% higher than the Young Control group) DA content in the Young

Stress group. A significant, F(1,27)--8.6, p < 0 . 0 0 7 , age effect also was observed on DOPAC concentrations. This effect was due to the reduced (26-30% versus the young groups) DOPAC levels in the Old Control group (Table 1). A significant, F ( 1 , 2 7 ) = 5.2, p < 0 . 0 3 , age effect also was found on hypothalamic 5-HT levels (Table 2). Overall, these observations suggest that stress may lead to increased DA and 5-HT synthesis in the hypothalamus of young and old animals, respectively, but no change in the release of these monoamines. Old nonstressed rats, furthermore, evidence a decrease in hypothalamic DA turnover.

TABLE 2 MEDIAL FRONTAL CORTICAL AND HYPOTHALAMIC 5-HT AND 5-HIAA CONCENTRATIONS IN STRESSED AND NONSTRESSED YOUNG (7 MONTH) AND OLD (22 MONTH) F344 MALE RATS Group

Brain Region Medial Frontal Cortex

Age

Hypothalamus

Procedure

5-HT

Young

Control

835 -+55

321 -+19

867 --_48

604 ---19

Young

Stress

791 ± 36

314 - 19

977 ± 59

504 -+ 32

Old

Control

820 + 57

364 - 16

1017 ± 50

511 -+29

Old

Stress

905 - 23

386* ± 26

1058" -+42

553 --+37

5-HIAA

5-HT

5-HIAA

Mean _+ S.E.M. (n = 7-8/group) ng/g wet tissue weight. *p<0.05 compared to young rats. Abbreviations: 5-HT = 5-Hydroxytryptamine (serotonin). 5-HIAA = 5Hydroxyindoleacetic acid.

].C)RENS /:T Ai

144

TABLE 3

70-

REGIONAL CNS SEROTONIN AND 5-HIAA LEVELS 1N YOUNG CONTROL RATS

CNS Region

5-HT

Neostriatum Nucleus accumbens Hippocampus Amygdala Spinal cord

894 1050 577 1013 347

=

5-HIAA

± 16 _+ 41 ± 24 _+ 41 ± 35

288 374 358 438 126

-+ _+ ± _+ ±

I CO.r~OL

T

~"~STRESS

500

10 12 12 17 13

0

o C-IT o0

Mean _ S.E.M. (n=6-8) ng/g wet weight. Only the Young Control group data are presented since no significant group differences were observed in the 5-HT (5-hydroxytryptamine; serotonin) and 5-HIAA (5-hydroxyindoleacetic acid) concentrations in these areas.

g

40 30" 20" '10"

/ .... F/Ill, IlIA 11111

0 '160'='-

g

Neostriatum. Although no group differences were observed in caudate-putamen DA levels, a significant effect of age was found in both DOPAC, F(1,28)=4.3, p < 0 . 0 5 , and HVA, F(1,28)= 34.0, p<0.0001, concentrations (Table 4). Post hoc comparisons showed that the HVA levels in both of the old groups were significantly lower than in the young groups. The Newman-Keuls' analysis did not reveal individual group differences in DOPAC levels. However, when the neostriatal DOPAC concentrations in the young rats (regardless of treatment) were compared to those in the old rats a significant (p<0.05) difference was observed. No significant group differences in caudate-putamen 5-HT and 5HIAA levels were obtained (Table 3). Overall, these data confirm previous studies that DA turnover is reduced in the neostriatum of old animals, and that stress does not significantly affect DA metabolism in the caudate-putamen. Nucleus accumbens. No group differences were found in accumbens DA content. Significant age effects, however, were obtained on DOPAC, F(1,28)=22.7, p<0.0001, and HVA, F(1,28)= 38.2, p<0.0001, levels. In addition, a significant, F ( l , 2 8 ) = 12.1, p < 0 . 0 0 2 , effect of stress on accumbens HVA concentrations also was obtained (Table 4). Post hoc comparisons showed that the DOPAC contents in the old groups were significantly lower than in the two young groups. The Newman-Keuls' analysis indicated that

t

60-

140"

120" z
==-

lO0-

8o60 40 PO 0

OLD

YOUNG

FIG. 1. Effects of CER procedure on corticosterone and prolactin levels (mean +--S.E.M.) in young (7 month) and old (22 month) F344 male rats (n=8/group). op<0.05 compared to Young Control group. *p<0.01 compared to both Control groups. **p<0.05 compared to the other three groups.

accumbens HVA content in the Old Control group was significantly lower than in the other three groups, and that the HVA concentrations in the Young Stress group was significantly higher than in the other three groups (Table 4). Accumbens 5-HT and

TABLE 4 NEOSTRIATAL AND NUCLEUS ACCUMBENS DOPAMINE AND METABOLITE CONCENTRATIONS IN STRESSED AND NONSTRESSED YOUNG (7 MONTH) AND OLD (22 MONTH) F344 MALE RATS

Group

Brain Region Neostriatum

Age

Nucleus Accumbens

Procedure

DA

DOPAC

HVA

DA

DOPAC

HVA

Young

Control

6384 ± 334

893 _+39

386 ± 24

5717 ± 349

988 ± 56

380 ± 24

Young

Stress

6107 ±243

905 --+25

407 ±15

6060 +315

1079 ±39

467+ --+15

Old

Control

6211 +_171

803 ±59

283* ± 16

5442 ___239

701 * ±62

263t +--24

Old

Stress

6112 ±310

815 ±44

295* ± 17

5394 ±231

821 * ±68

323 ±20

Mean ___ S.E.M. (n = 8/group) ng/g wet weight. *p<0.05 compared to both of the Young groups. tp<0.05 compared to the other three groups. Abbreviations: DA = Dopamine. DOPAC = 3,4-Dihydroxyphenylacetic acid. HVA = Homovanillic acid.

RESPONSE TO STRESS IN AGED RATS

145

TABLE 5 ANGIOTENSINOGENAND RENIN RESPONSESTO 20 MIN CER PROCEDUREIN YOUNG(7 MONTH) AND OLD (22 MONTH) F344 MALE RATS Group

30-

Plasma Hormone Levels

Age

Procedure

ANG

PRA

PRC

20-

Young

Control

61.2 ___7.4

6.7 +__1.7

19.2 -2_3.0

I0-

Young

Stress

59.7 +-7.0

16.15 +-2.3

36.8t +-2.9

Old

Control

37.3 ---8.9

3.6 __.1.1

12.2 ___1.7

Old

Stress

42.2 ---6.5

5.8 --- 1.8

24.3" ---3.1

Mean - S.E.M. (N= 8/group). *p<0.01 compared to the Old Control group. tp<0.05 compared to all other groups. :~p<0.01 compared to all other groups. Abbreviations: ANG=Angiotensinogen (ng AI/ml/hr). PRA=Plasma renin activity (ng AI/3 hr/ml). PRC=Plasma renin concentration (ng AI/ml/hr). 5-HIAA levels were not affected (Table 3). Thus, DA metabolism is reduced in the nucleus accumbens of old rats since both DOPAC and HVA levels were reduced without a concomitant change in DA concentration. Further, stress induces a significant increase in accumbens HVA content in both young and old rats, in comparison to their age-matched controls. Hippocampus and spinal cord. No group differences were observed in the 5-HT or 5-HIAA concentrations in these CNS regions (Table 3).

Hormone levels. Corticosterone. A significant, F(1,28) = 11.2, p<0.002, effect of the CER procedure on plasma corticosterone levels was observed (Fig. 1). The age effect did not reach statistical significance, F(1,28)=3.6, p<0.07. Post hoc comparisons showed that the stress-induced increase in corticosterone levels were of a greater magnitude in the old rats. The levels in the Young Stress group were 88% higher than in the Young Control group, whereas the corticosterone levels in the Old Stress group were 155% higher than in the Old Control group (Fig. 1). Thus, the old rats exhibited an exacerbated corticosterone response to stress. Prolactin. Stress significantly, F(1,28)=35.8, p<0.0001, increased prolactin levels (Fig. 1). Although the increase was greater in the old rats (191% compared to their age-matched controls) than in the young rats (119%), this difference did not reach statistical significance. Angiotensinogen and renin. Plasma angiotensinogen levels (Table 5) were significantly lower in the old rats, F(1,28) = 7.6, p<0.01. No effect of stress was observed. In contrast, significant age, F(1,28)= 13.7, p<0.001, stress, F(1,28)= 10.2, p<0.004, and age × stress, F(1,28)=4.0, p<0.05, effects were found on plasma renin activity (PRA). The post hoc analysis revealed that these effects were due to the elevated PRA in the Young Stress group relative to the other three groups (Table 5). A significant effect of age, F(1,28)= 12.7, p<0.002, and stress, F(1,28)= 0.0001, on plasma renin concentration (PRC) also was observed. The Newman-Keuls' analysis showed that both stress groups exhibited an increase in PRC, but that the PRC in the Young Stress group was significantly higher than in the Old Stress group (Table

*I*

40-

I yc

ys

Oc

os

FIG. 2. Mean ( -+S.E.M.) percent of large granular lymphocytes (%LGL) in the spleens (n = 3/group) of young (7 month) and old (22 month) male F344 rats exposed to the CER procedure, yc = Young Control group. ys = Young Stress group, oc = Old Control group, os = Old Stress group. **Significantly (p<0.01) greater than both the yc and ys groups.

5). Overall, these data suggest that increasing age is associated with a reduction in the availability of the renin substrate, angiotensinogen, and with a reduction in renin secretion during stress. Immunological measures. The percent of large granular lymphocytes (LGLs) was significantly, F(1,8)=66.3, p<0.0001, greater (>300%) in the old rats (Fig. 2). Significant effects of age, F(1,8) = 130.7, p<0.00001, stress, F(1,8)= 22.1, p<0.002, and age × stress, F(1,8)=20.0, p<0.02, on the cytotoxicity (expressed as lytic units) of freshly isolated splenic natural killer (NK) cells were obtained (Fig. 3). The post hoc analysis showed that splenic NK cytotoxicity was significantly lower in the Young Stress group than in the Young Control group, and that splenic NK cytotoxicity was significantly higher in both of the young groups than in the two old groups. Analysis of IL-2 activated splenic NK cytotoxicity gave identical results (Fig. 3). Significant effects of age, F ( 1 , 8 ) = l l l . 0 , p<0.0001, and stress, F(1,8)=76.0, p<0.0001, also were observed on Con-A-stimulated 3H-thymidine uptake (Fig. 4). The post hoc analysis showed that the

25 20

g¢

I

15. o I0-

5

i

yc

I

ys

OI2

OS

FIG. 3. Effects of CER procedure on splenic natural killer cell cytotoxicity expressed as lytic units (LU) in young (7 month) and old (22 month) male F344 rats (n = 3/group). Hatched bars represent the LUs (mean _+S.E.M.) of cells incubated with HBSS for 18 hr, and the open bars represent the level of LUs generated by these cells following activation with rlL-2 for 18 hr. Abbreviations as in Fig. 2. *Significantly (p<0.01) less than yc group. **Significantly (p<0.05) less than both the yc and ys groups. ***Significantly (p<0.05) less than the other three groups.

146

LORENS ET AL.

TABLE 6 80

REGIONAL CNS TYPE II CORTICOSTERONE RECEPTORS IN YOUNG (7 MONTH) AND OLD (I7.5 MONTH) F344 MALE RATS

70 Brain Area/Age

60

co 0

50 ~L

40 30 20 I0

yc

ys

oc

i os

FIG. 4. Effect of the CER procedure on Con-A-stimulated 3H-thymidine uptake (mean --+S.E.M.) by splenic T lymphocytes in young (7 month) and old (22 month) male F344 rats (n = 3/group). CPMs without Con-A were 1-2 x 103 for all groups. Abbreviations as in Fig. 2. *Significantly (p<0.01) less than the yc group. **Significantly (p<0.01) less than the other three groups.

mitogen-stimulated 3H-thymidine uptake in the Young Stress and Old Control groups was significantly lower (25%) than in the Young Control group. The Con-A-stimulated 3H-thymidine uptake in the Old Stress group, furthermore, was significantly less than in the other three groups. Overall, these data indicate that although there is an increase in the number of large granular lymphocytes in old rats, splenic NK cytotoxicity is reduced. The CER procedure reduced splenic NK cytotoxicity and T-cell proliferation in the young rats, suggesting that stress can impair

1--I 50-

BASEL|NE POST-STRESS

C 0 40R T 30-

/////////~ U g 20-

/

Midbrain raphe Young Old

Type II Receptors (fmole/mg protein)

6 5

161 _+ 10 147 ± 3

6 5

138 _+ 5 140 -- 8

6 5

159 --+ 4 153 -+ 5

6 5

107 -+ 6 94 -+ 10

immune function. The stress-induced inhibition of T-cell proliferation, moreover, is exaggerated in old animals. Experiment 2. Ether Stress and Corticosterone Receptors All of the rats maintained their weights and appeared healthy at the time of the ether stress. Unfortunately, adrenalectomy resulted in the death of 50% of the 12 old rats. One additional adrenaiectomized old animal was dropped from the study because a pituitary tumor was noted at sacrifice. Three young animals were dropped from the corticosterone receptor analysis because plasma corticosterone levels at the time of sacrifice were above detectable levels, indicating an incomplete adrenalectomy. Also, one of the young hippocampal samples was lost due to technical error. Plasma corticosterone levels. The analysis of variance revealed significant age, F(1,19) = 12.2, p<0.003, stress, F(1,19) = 376.0, p<0.0001, and age x stress, F(1,19)=9.9, p<0.005, effects (Fig. 5). Post hoc analysis showed that the corticosterone response to stress was significantly higher (37%) in the old than in the young rats. These results are similar to those obtained in Experiment 1 which employed older rats and a different stressor. CNS corticosterone receptors. Type I corticosterone receptors in the hippocampus, but not in the septal area, were significantly, t(8) = 3.43, p<0.01, lower (47%) in the old than in the young rats (Fig. 6). No significant age differences in Type II corticosterone receptors were found in any of the brain regions analyzed (Table 6). It should be noted that the values (fmole/mg protein) we obtained for type I and II binding agree well with those reported by De Kloet et al. (12). DISCUSSION

~'////////

/ 1 1 / I / I / /

//'//////.6 ////////// " / / / / / / / / ~

D L

Medial frontal cortex Young Old Amygdala Young Old Hypothalamus Young Old

N

iO-

I i 0 YOUNG

OLD

FIG. 5. Corticosterone (CORT) levels (mean ---S.E.M.) before and 20 min after exposure to ether stress in young (7.0 month; n=9) and old (17.5 month; n= 12) male F344 rats. *Significantly (p<0.0001) higher than baseline. **Significantly greater than baseline (p<0.0001), and the poststress values in the young animals (p<0.01).

Major changes in the basal functionality and reactivity to stress of central monoamine, endocrine and immune systems were observed in aging animals. In agreement with several previous reports (16, 50, 51), the old nonstressed rats evidenced a significant decrease in DA metabolism in the medial frontal cortex (MFC), neostriatum, nucleus accumbens and hypothalamus, but not in the amygdala. In contrast, the young and old control rats did not exhibit significant differences in NE, 5-HT or 5-HIAA levels in any of the CNS regions examined. For the most part these results are in agreement with previous observations (50, 51, 69). Stress consistently has been observed to increase CNS DA metabolism (18, 26, 59). In the present study, the CER paradigm

RESPONSE TO STRESS IN AGED RATS

147

TYPE I RECEFI'OR I 200

I YOUNG OLD

150 I00 °,u

ot...

50 0

TYPE II RECEPTOR O

200 150 I00 50 0 HIPPOCAMPUS

SEPTUM

BRAIN REGION FIG. 6. Hippocampaland septal area type I and II corticosteronereceptors (mean---S.E.M.) in young (7.0 month) and old (17.5 month) male F344 rats (n = 5-6/group). *Significantly(p<0.01) less than in young animals.

was found to increase DA metabolism in the MFC, amygdala and nucleus accumbens of young rats, as reported earlier (26,59), but in old rats as well. To our knowledge, this is the first demonstration of stress-induced increases in regional CNS DA turnover in old animals. Although basal DA metabolism is reduced in the MFC and nucleus accumbens of old nonstressed rats, it is clear that stress can activate the mesolimbocortical system in aged animals. Interestingly, the CER procedure led to a reduction in MFC NE content in the young but not in the old rats. No group differences in amygdaloid NE concentration were observed. In contrast, stress increased MFC 5-HIAA levels in the old but not in the young rats. These observations suggest that the MFC NE and 5-HT systems in young and old animals respond differently to stress. Stress has been reported to increase NE synthesis in young animals (29). However, since we did not measure levels of the NE metabolite, MHPG-SO4, we cannot conclude that stress increases MFC NE metabolism in young but not in old rats. The basal plasma corticosterone levels in the young and old nonstressed rats did not differ when measured between 09:00-13:30. This observation is in agreement with DeKosky et al. (14) who reported that basal plasma corticosterone levels were higher in old than in young rats late in the day (18:00 and 23:30) but not in the morning (08:00). On the other hand, stress led to an exaggerated corticosterone response in the old male rats. This response does not appear to depend upon the type of stressor since enhanced corticosterone secretion was observed in old rats following ether stress as well as the CER procedure. We recently have found (unpublished data) that old (22 month) F344 male rats also show an enhanced corticosterone response to novelty stress (20

min exposure to a novel open field). These data are consistent with those of Sapolsky et al. (61) who reported an increased secretion of corticosterone in old male rats following one hour of immobilization stress. Whether this exaggerated secretion of corticosterone is due to augmented neural stimulation or to a decrease in negative feedback has not been determined. In order to explore this issue, we measured corticosterone receptors in the brains of young and old rats. At present, two types of receptors for corticosterone can be distinguished in brain tissue (37, 53, 54). These receptors are separable based on biochemistry and primary structure (1, 37, 39, 53, 54). The hippocampal type I receptor is occupied even with low circulating titers of corticosterone (55), and thus appears to be involved in the regulation of basal ACTH (11) and corticosterone secretion, and the synchronizationof activities such as feeding and sleep (11, 13, 42, 43, 73). The hippocampal type II receptor is occupied during times of elevated corticosterone secretion and appears to regulate glucocorticoid feedback following stress, and glucocorticoid effects on the extinction of fear motivated behavior (5, 6, 13, 55). Since we did not detect changes in hippocampal type II receptors in 17.5-month-old animals, the present data suggest that the observed age-related stress-induced increases in corticosterone secretion are the result of enhanced hypothalamic stimulation rather than a deficit in the negative feedback response. In addition, the decreases in hippocampal type I receptor were not accompanied by changes in basal corticosterone secretion. It is possible that the decreases in hippocampal type I receptor could cause changes in ACTH which are not reflected by changes in corticosterone levels since previous studies have reported reduced adrenal sensitivity to ACTH in old animals (11). Reul et al. (56) reported reductions in both type I and II corticosterone receptors in the hippocampus of 28-month-old rats. We found a decrease in hippocampal type I but no change in CNS type II corticosterone receptors in our 17.5-month-old rats. Our data suggest that deficits in hippocampal sensitivity to corticosterone occur with age. However, receptor subtype specific changes do not occur in unison. The exaggerated corticosterone response to stress in old rats thus precedes the down-regulation of hippocampal type II receptors. In contrast to earlier reports (15, 16, 52), we did not observe any age-related differences in basal prolactin levels or in the prolactin response to stress. This finding is rather surprising since hypothalamic DA turnover was reduced in the old nonstressed rats, and 5-HT levels were increased. 5-HT is thought to play a stimulatory role in prolactin secretion (70), whereas DA has been demonstrated to inhibit prolactin release. This discrepancy may be due to a difference in rat strain as Long-Evans rats were used in the previous studies (15, 16, 52). Plasma angiotensinogen levels were significantly lower in the old animals. The old rats also showed a reduction in stress-induced renin secretion. These findings are in agreement with those of Corman and Michel (9) who suggested that the activity of the renin-angiotensin system is decreased with age. These data also suggest that the age-associated changes in the hypothalamic regulation of corticosterone secretion are unique in that they are not found in other hormonal systems. NK cells have been implicated in the in vivo resistance to syngeneic tumors (25,68). Stress has been documented to decrease the efficiency of the immune system in general (58,65), to decrease NK cell activity (10, 30, 49), and thus to increase the incidence, size, and rate of primary tumor metastases (20,48). While the mechanism of stress-induced suppression has not been clearly elucidated, the role of opiate receptors (10,64) and suppressor cells (49) and molecules (27, 28, 76) have been implicated. Two potential suppressor molecules previously documented to decrease NK activity are prostaglandin E2 from macrophages

148

IORENS ET AL.

(2,76) and corticosteroids (27, 28, 44), both of which are released during stress. Although the present report did not evaluate the mechanism, it did demonstrate a decrease in Con-A-induced T-cell proliferation and in NK activity after four days of exposure to the CER procedure. Additional studies are required in order to determine whether the existence of suppressor cells (49), directly suppressor cells (27, 28, 76), or corticosteroid-induced suppressor cells (44) are responsible. In addition, the role of opioid peptides in the age-related changes in the immune response to stress requires further study. Previous studies by Ghoneum et al. (19) have demonstrated that isolation-induced stress in 3- and 12-month-old SpragueDawley (SD) male rats induced a decrease in splenic and peripheral blood NK activity only in the older group. Since the data in their report were presented as percentage of control, it is not possible to directly compare our data with theirs. Nevertheless, in agreement with previous studies (10, 36, 64), the present report clearly shows a significant effect of stress on young (7 month) splenic NK activity in the F344 rat. Stress, however, did not further compromise the depressed NK activity of the old rats. On the other hand, the CER procedure significantly decreased both

young and old Con-A proliferative activity. The increased LGL levels but decreased functional splenic NK activity found in the old F344 rats in the present study, to our knowledge, has not been observed previously. The cause and significance of this finding is unclear, except that older F344 rats have a high incidence of LGL tumors (74). Thus many of the LGL cells observed may be abnormal and not functionally active. The present studies have demonstrated age-related alterations in neural, endocrine and immune functions. Importantly, the findings were obtained using the same animals, permitting the direct correlation of changes in one system with changes in another system. This interdisciplinary approach will be especially useful in future studies designed to determine if age-related deficits in one homeostatic system precede and/or are responsible for dysfunctions in another. ACKNOWLEDGEMENTS This research was supported in part by grants from Crinos Farmacobiological S.p.A (Como, Italy). the Heart Research Foundation, the Potts Foundation, the National Institute on Drug Abuse (RFP No. 271-87-81171. and the National Institute of Health (No. AI-23718).

REFERENCES 1. Arriza, J. L.; Weinberger, C.; Cerelli, G.; Glaser, T. M.; Handelin, B. L.; Housman, D. E.; Evans, R. M. Cloning of human mineralocorticoid receptor complementary DNA: structural and functional kinship with the glucocorticoid receptor. Science 237:268-275; 1987. 2. Bash, J. A.; Vogel, D. Cellular immunosenescence in F344 rats: decreased natural killer (NK) cell activity involves changes in regulatory interactions between NK cells, interferon, prostaglandin and macrophages. Mech. Ageing Dev. 24:49-65; 1984. 3. Blair, P. B.; Staskawicz, M. O.; Sam, J. S. Suppression of natural killer cell activity in young and old mice. Mech. Ageing Dev. 40:57-70; 1987. 4. Blalock, J. E. A molecular basis for bidirectional communication between the immune and neuroendocrine systems. Physiol. Rev. 69:1-32; 1989. 5. Bohus, B.; de Kloet, E. R. Adrenal steroids and extinction behavior: antagonism by progesterone, d'eoxycorticosterone and dexamethasone of a specific effect of corticosterone. Life Sci. 28:433-440; 1981. 6. Bohus, B.; de Kloet, E. R.; Veldhuis, H. D. Adrenal steroids and behavioural adaptation: relationship to brain corticosterone receptors. In: Ganten, D.; Pfaff, D., eds. Current topics in neuroendocrinology. vol. 2. Berlin: Springer-Verlag; 1982:107-148. 7. Brizzee, K. R.; Ordy, J. M. Age pigments, cell loss and hippocampal function. Mech. Ageing Dev. 9:143-162; 1979. 8. Clancy, J., Jr.; Mauser, L.; Chapman, A. Level and temporal pattern of naturally cytolytic cells during acute graft-vs-host disease in the rat. Cell. Immunol. 79:1-10; 1983. 9. Corman, B.; Michel, J-B. Renin angiotensin system, converting enzyme inhibition and kidney function in aging female rats. Am. J. Physiol. 251 :R450-R455; 1986. 10. Cunnick, J. E.; Lysle, D. T.; Armfield, A.; Rabin, B. S. Shockinduced modulation of lymphocyte responsiveness and natural killer cell activity: differential mechanisms of induction. Brain Behav. Immun. 2:102-113; 1988. 11. Dallman, M. F.; Akana, S. F.; Casciao, C. S.; Darlington, D. N.; Jacobson, L. ; Levin, N. Regulation of ACTH secretion: variations on a theme of B. Recent Prog. Horm. Res. 43:113-173; 1987. 12. De Kloet, E. R.; Ratka, A.; Reul, J. M. H. M.; Sutanto, W.; Van Eekelen, J. A. M. Corticosteroid receptor types in brain: Regulation and putative function. Ann. NY Acad. Sci. 512:351-361; 1987. 13. De Kloet, E. R.; Reul, J. M. H. M. Feedback action and tonic influence of corticosteroids on brain function: a concept arising from the heterogeneity of brain receptor systems. Psychoneuroendocrinology 12:83-105; 1987. 14. DeKosky, S. T.; Scheff, S. W.; Cotman, C. W. Elevated corticosterone levels: possible causes of reduced axon sprouting in aged animals. Neuroendocrinology 38:33-38; 1984.

15. Demarest, K. T.; Moore, K. E.; Riegle, G. D. Adenohypophyseal dopamine content and prolactin secretion in the aged male and female rat. Endocrinology 116:1316-1323; 1985. 16. Demarest, K. T.; Riegle, G. D.; Moore, K. E. Characteristics of dopaminergic neurons in the aged male rat. Neuroendocrinology 31:222-227; 1980. 17. De Souza, E. B.; Van Loon, G. R. Brain serotonin and catecholamine responses to repeated stress in rats. Brain Res. 367:77-86; 1986. 18. Deutch, A. Y.; Tam, S-Y.; Roth, R. H. Footshock and conditioned stress increase 3,4-dihydroxyphenylacetic acid (DOPAC) in the ventral tegmental area but not substantia nigra. Brain Res. 333:143-146: 1985. 19. Ghoneum, M.; Gill, G.; Assanah, P.; Stevens, W. Susceptibility of natural killer cell activity of old rats to stress. Immunology 60: 461-465; 1987. 20. Gottfried, B.; Malomut, N.; Skaredoff, L. Effect of repeated surgical wounding on the growth rate of tumor graft. Ann. Surg. 153:138-142; 1961. 21. Greene, E.; Naranjo, J. N. Degeneration of hippocampal fibers and spatial memory deficit in the aged rat. Neurobiol. Aging. 8:35-43: 1987. 22. Haber, C.; Loerner, D.; Page, L. B.; Kliman, B.; Purnode, A. Application of a radioimmunoassay for angiotensin I to the physiologic measurements of plasma renin activity in normal human subjects. J. Clin. Endocrinol. 29:1329-1355; 1969. 23. Handa, R. J.; Reid, D. L.; Resko, J. A. Androgen receptors in brain and pituitary of female rats: cyclic changes and comparisons with the male. Biol. Reprod. 34:293-303; 1987. 24. Heffner, T. G.; Hartman, J. A.; Seiden, L. S. A rapid method for the regional dissection of the rat brain. Pharmacol. Biochem. Behav. 13:453-456; 1980. 25. Herberman, R. B.; Nunn, M. E.; Holden, H. T.; Lavrin, D. H. Natural cytotoxic reactivity of mouse lymphoid cells against syngeneic and allogeneic tumors II. Characterization of effector cells. Int. J. Cancer 16:230-239; 1975. 26. Herman, J. P.; Guillonneau, D.; Dantzer, R.; Scatton, B.; Semerdjian-Rouquier, L.; Le Moal, M. Differential effects of inescapable footshocks and of stimuli previously paired with inescapable footshocks on dopamine turnover in cortical and limbic areas of the rat. Life Sci. 30:2207-2214; 1982. 27. Hochman, P. S.; Cudkowicz, G. Suppression of natural cytotoxicity by spleen cells of hydrocortisone-treated mice. J. Immunol. 123: 968-976; 1979. 28. Holbrook, N. J.; Cox, W. I.; Homer, H. C. Direct suppression of natural killer activity in human peripheral blood leukocyte cultures by glucocorticoids and its modulation by interferon. Cancer Res. 43:

RESPONSE TO STRESS IN AGED RATS

4019--4025; 1983. 29. Iimori, K.; Tanaka, M.; Kohno, Y.; Ida, Y.; Nakagawa, R.; Hoaki, Y.; Tsuda, A.; Nagasaki, N. Psychological stress enhances noradrenaline turnover in specific brain regions in rats. Pharmacol. Biochem. Behav. 16:637-640; 1982. 30. Irwin, M. R.; Hauger, R. L. Adaptation to chronic stress: temporal pattern of immune and neuroendocrine correlates. Neuropsychopharmacology 1:239-242; 1988. 31. Landfield, P. W.; Rose, G.; Sandles, L.; Wohlstadter, T. C.; Lynch, G. Patterns of astroglial hypertrophy and neuronal degeneration in the hippocampus of aged, memory-deficient rats. J. Gerontol. 32:3-12; 1977. 32. Landfield, P. W.; Waymire, J. C.; Lynch, G. Hippocampal aging and adrenocorficoids: quantitative correlations. Science 202:1098-1102; 1978. 33. Landfield, P. W.; Baskin, R. K.; Pitier, T. A. Brain aging correlates: retardation by hormonal-pharmacological treatments. Science 214: 581-584; 1981. 34. Lorens, S. A.; Van de Kar, L. D. Differential effects of serotonin (5-HTIA and 5-HT2) agonists and antagonists on renin and corticosterone secretion. Neuroendocrinology 45:305-310; 1987. 35. Lowry, O. H.; Rosebrough, N. J.; Farr, A. L.; Randall, R. J. Protein determination with the folin phenol reagent. J. Biol. Chem. 193: 265-275; 1951. 36. Lysle, D. T.; Lyte, M.; Fowler, H.; Rabin, B. S. Shock-induced modulation of lymphocyte reactivity: suppression, habituation, and recovery. Life Sci. 41:1805-1814; 1987. 37. McEwen, B. S.; de Kloet, E. R.; Rostene, W. Adrenal steroid receptors and actions in the central nervous system. Physiol. Rev. 66:1121-1188; 1986. 38. Mankinodan, T.; Kay, M. M. B. Age influence on the immune system. Adv. Immunol. 29:287-330; 1980. 39. Mefford, I. N. Application of high performance liquid chromatography with electrochemical detection to neurochemical analysis: measurement of catecholamines, serotonin and metabolites in rat brain. J. Neurosci. Methods 3:207-224; 1981. 40. Meisfeld, R.; Rusconi, S.; Godowski, P. J.; Maler, B. A.; Okret, S.; Wikstrom, A-C.; Gustafsson, J-A.; Yamamoto, K. R. Genetic complementation of a glucocorticoid receptor deficiency by expression of cloned receptor cDNA. Cell 46:389-399; 1986. 41. Meites, J.; Goya, R.; Takahashi, S. Why the neuroendocrine system is important in aging processes. Exp. Gerontol. 22:1-15; 1987. 42. Micco, D. J.; McEwen, B. S. Glucocorticoids, the hippocampus and behavior: interactive relation between task activation and steroid hormone binding specificity. J. Comp. Physiol. Psychol. 94:624-633; 1980. 43. Micco, D. J.; Meyer, J. S.; McEwen, B. S. Effects of corticosterone replacement on the temporal patterning of activity and sleep in adrenalectomized rats. Brain Res. 200:206-212; 1980. 44. Nair, M. P. N.; Schwartz, S. A. Immunoregulation of human natural killer cells (NK) by corticosteroids: inhibitory effect of culture supernatants. J. Allergy Clin. Immunol. 82:1089-1097; 1988. 45. Oehler, J. R.; Herberman, R. B.; Campbell, D. A., Jr.; Djeu, J. Y. Inhibition of rat mixed lymphocyte cultures by suppressor macrophages. Cell Immunol. 29:238-250; 1977. 46. Oehler, J. R.; Lindsay, L. R.; Nunn, M. E.; Herberman, R. B. Natural cell-mediated cytotoxicity in rats. I. Tissue and strain distribution, and demonstration of a membrane receptor for the Fc portion of IgG. Int. J. Cancer 21:204-209; 1978. 47. Paris, J. M.; Lorens, S. A.; Van de Kar, L. D.; Urban, J. H.; Richardson-Morton, K. D.; Bethea, C. L. A comparison of acute stress paradigms: hormonal responses and hypothalamic serotonin. Physiol. Behav. 39:33-43; 1987. 48. Pollack, R. E.; Babcock, G. F.; Romsdahl, M. M.; Nishivka, K. Surgical stress-mediated suppression of murine natural killer cell cytotoxicity. Cancer Res. 44:3888-3892; 1984. 49. Pollack, R. E.; Lutzova, E.; Stanford, S. D.; Romsdahl, M. M. Effect of surgical stress on murine natural kill cell cytotoxicity. J. Immunol. 138:171-178; 1987. 50. Ponzio, F.; Calderini, G.; Lomuscio, G.; Vantini, G.; Toffano, G.; Algeri, S. Changes in monoamines and their metabolite levels in some brain regions of aged rats. Neurobiol. Aging 3:23-29; 1982. 51. Pradhan, S. N. Central neurotransmitters and aging. Life Sci. 26:

149

1643-1656; 1980. 52. Reigle, G. D.; Meites, J. Effects of aging on LH and prolactin after LHRH L-dopa, methyl-dopa, and stress in male rat. Proc. Soc. Exp. Biol. Med. 151:507-511; 1979. 53. Reul, J. M. H.; de Kloet, E. R. Two receptor systems for corticosterone in rat brain: microdistribution and differential occupation. Endocrinology 117:2505-2511; 1985. 54. Reul, J. M. H.; de Kloet, E. R. Anatomical resolution of two types of corticosterone receptor sites in rat brain with in vitro autoradiography and computerized image analysis. J. Steroid Biochem. 24:269-272; 1986. 55. Reul, J. M. H.; van den Bosch, F. R.; de Kloet, E. R. Relative occupation of type-I and type-II corticosteroid receptors in rat brain following stress and dexamethasone treatment: functional implications. J. Endocrinol. 115:459-467; 1987. 56. Reul, J. M. H.; Tonnaer, J. A. D. M.; de Kloet, E. R. Neurotrophic ACTH analogue promotes plasticity of Type I corticosteroid receptor in brain of senescent male rats. Neurobiol. Aging 9:253-260; 1988. 57. Richardson Morton, K. D.; Van de Kar, L. D.; Brownfield, M. S.; Bethea, C. L. Neuronal cell bodies in the hypothalamic paraventricular nucleus mediate stress-induced renin and corticosterone secretion. Neuroendocrinology 50:73-80; 1989. 58. Riddle, P. R.; Berenbaum, M. C. Postoperative depression of the lymphocyte response to phytohaemagglutinin. Lancet 1:746-748; 1967. 59. Roth, R. H.; Wolf, M. E.; Deutch, A. Y. Neurochemistry of midbrain dopamine systems. In: Meltzer, H. Y., ed. Psychopharmacology: the third generation of progress. New York: Raven Press; 1987:81-94. 60. Sapolsky, R. M.; Krey, L. C.; McEwen, B. S. Corticosterone receptors decline in a site-specific manner in the aged rat brain. Brain Res. 289:235-240; 1983. 61. Sapolsky, R. M.; Krey, L. C.; McEwen, B. S. Stress down-regulates corticosterone receptors in a site-specific manner in the brain. Endocrinology 114:287-292; 1984. 62. Sapolsky, R. M.; Krey, L. C.; McEwen, B. S. Glucocorticoidsensitive hippocampal neurons are involved in terminating the adrenocortical stress response. Proc. Natl. Acad. Sci. USA 81:6174--6177; 1984. 63. Sapolsky, R. M.; McEwen, B. S. Stress, glucocorticoids, and their role in degenerative changes in the aging hippocampus. In: Crook, T., et al., eds. Treatment development strategies for Alzheimer's disease. Madison, CT: Mark Powley Assoc.; 1986:151-171. 64. Shavit, Y.; Lewis, J. W.; Terman, G. W.; Gale, R. P.; Liebeskind, J. C. Opioid peptides mediate the suppressive effect of stress on natural killer cell cytotoxicity. Science 223:188-190; 1984. 65. Slade, M. S.; Simmons, R. L.; Yunis, E.; Greenberg, L. J. Immunodepression after major surgery in normal patients. Surgery 78: 363-372; 1975. 66. Staiano-Coico, L.; Darzynkiewicz, Z.; Melamed, M. R.; Weksler, M. E. Immunological studies of aging. IX. Impaired proliferation of T lymphocytes detected in elderly humans by flow cytometry. J. Immunol. 132:1788-1792; 1984. 67. Stockigt, J. R.; Collins, R. D.; Biglierio, E. G. Determination of plasma renin concentration by angiotensin I radioimmunoassay. Circ. Res. 28/29:175-180; 1971. 68. Talmadge, J. E.; Meyers, K. M.; Prieur, D. J.; Starkey, J. R. Role of NK cells in tumor growth and metastasis in beige mice. Nature 284:622-624; 1980. 69. Timiras, P. S.; Hudson, D. B.; Segall, P. E. Lifetime brain serotonin: Regional effects of age and precursor availability. Neurobiol. Aging 5:235-242; 1984. 70. Van de Kar, L. D.; Karteszi, M.; Bethea, C. L.; Ganong, W. F. Serotonergic stimulation of prolactin and corticosterone secretion is mediated by different pathways from the mediobasal hypothalamus. Neuroendocrinology 41:380-384; 1985. 71. Van de Kar, L. D.; Lorens, S. A.; McWilliams, C.; Kunimoto, K.; Urban, J. H.; Bethea, C. L. Role of midbrain raphe in stress-induced renin and prolactin secretion. Brain Res. 311:333-341; 1984. 72. Veldhuis, H. D.; van Koppen, C.; van Ittersum, M.; de Kloet, E. R. Adrenalectomy reduces exploratory behavior activity in rat: a specific role of corticosterone. Horm. Behav. 16:191-198; 1982. 73. Vujanovic, N. L.; Herberman, R. B.; Hiserodt, J. L. Lymphokineactiviated killer cells in rats: analysis of tissue and strain distribution, ontogeny, and target specificity. Cancer Res. 48:878-883; 1988.

150

74. Ward, J. M.; Reynolds, C. W. Large granular lymphocyte leukemia: a heterogeneous lymphocytic leukemia in F344 rats. Am. J. Pathol. 111:1-10; 1983. 75. Winer, B. J. Statistical principles in experimental design. New York:

LORENS ET AL,

McGraw-Hill; 1971. 76. Zielinski, C. C.; Gisinger, L.; Binder, C.; Mannhalter, J. W.; Eibl, M. M. Regulation of NK cell activity by prostaglandin E2: the role of T cells. Cell. Immunol. 87:65-72; 1984.