- Email: [email protected]
restraint stress inhibition of host resistance to Listeria monocytogenes
Journal of Neuroimmunology 125 (2002) 94 – 102 www.elsevier.com/locate/jneuroim
Sympathetic nervous system plays a major role in acute cold/restraint stress inhibition of host resistance to Listeria monocytogenes Ling Cao, Nikolay M. Filipov 1, David A. Lawrence* Wadsworth Center, New York State Department of Health, Empire State Plaza, P.O. Box 509, Albany, NY 12201-0509, USA Received 26 October 2001; received in revised form 28 January 2002; accepted 30 January 2002
Abstract BALB/c mice exposed to acute cold/restraint stress (ACRS) had significantly lower host resistance to Listeria monocytogenes (LM) than controls. The stress hormones corticosterone (CORT) and norepinephrine (NE), which are known to modulate immune responses, were evaluated as the cause of the decline in immune defense. The involvement of CORT and NE was investigated by pretreating mice with the CORT synthesis inhibitor metyrapone and the chemical sympathectomy drug 6-hydroxydopamine (6-OHDA), respectively. LM burdens in spleen and liver were determined three days post-infection. 6-OHDA significantly decreased the LM burden in both control and stressed animals. 6-OHDA also completely blocked the stress effects observed in spleens while only partially affecting the liver. The 6-OHDA-uptake inhibitor desipramine aided confirmation that peripheral sympathetic adrenergic nerves and NE depletion, rather than the direct action of 6OHDA, were responsible for the decreased susceptibility to LM. The results suggest that the peripheral sympathetic nervous system (SNS) postganglionic neurotransmitter NE plays a major role in LM host resistance and has significant tissue-dependent effects after ACRS. In contrast, metyrapone-treated animals had further decreased host resistance to LM, suggesting a potential protective effect of CORT after ACRS. Altogether, the results suggest that stress hormones play an important role in stress-modulated host resistance and that NE is the major hormone involved in ACRS-induced suppression of host resistance. D 2002 Published by Elsevier Science B.V. Keywords: Cold/restraint stress; Listeria monocytogenes; Host resistance; Corticosterone; Norepinephrine; Desipramine
1. Introduction Stress, as a state of disharmony or threatened homeostasis resulting from external or internal challenge (Fricchione and Stefano, 1994), has been associated with a variety of diseases including infectious diseases, neurodegenerative diseases, and cancers (Ader and Cohen, 1993; Biondi and Zannino, 1997; Wu et al., 2000). Although psychological stress also has been indicated in exacerbation of autoimmune diseases (Walker et al., 1999; Herrmann et al., 2000), inadequate stress hormone responses have been associated with enhancement of autoimmune diseases (Sternberg, 2001). In terms of infectious disease, stress has been shown to be involved in the onset, progression, and outcome in both human and animal studies. The stress-induced effect is dependent on
*
Corresponding author. Tel.: +1-518-402-5684; fax: +1-518-474-1412. E-mail address: [email protected] (D.A. Lawrence). 1 Current address: Center for Environmental Health Sciences, Department of Basic Sciences, College of Veterinary Medicine, Mississippi State University, Box 6100, Mississippi State, MS 39762, USA. 0165-5728/02/$ - see front matter D 2002 Published by Elsevier Science B.V. PII: S 0 1 6 5 - 5 7 2 8 ( 0 2 ) 0 0 0 3 9 - 5
the pathogen and the particular immune responses involved (Cohen et al., 1991; Biondi and Zannino, 1997), but the mechanisms by which stress modulates host resistance to infectious diseases have not been well delineated. Stress involves activation of both the hypothalamic-pituitary-adrenal (HPA) axis and the sympathetic nervous system (SNS) (Fricchione and Stefano, 1994). Repeated studies have demonstrated that both glucocorticoids and norepinephrine (NE) can modulate immune responses at different levels. In vitro and ex vivo, glucocorticoids suppress T cell proliferation in response to mitogens or antigens, natural killer (NK) cell activity, macrophage phagocytosis, and cytokine secretion from T cells and macrophages (Brown and Zwilling, 1994; Joyce et al., 1997; McEwen et al., 1997; Wilckens and De Rijk, 1997). In vivo, glucocorticoids modulate leukocyte trafficking and delayed-type hypersensitivity (DTH) responses (Dhabhar et al., 1996; Dhabhar and McEwen, 1998). With regard to NE, it is well known that SNS innervates both primary and secondary lymphoid organs. There are direct contacts between SNS nerve terminals (detected by anti-tyrosine hydroxylase antibody) and lymphocytes and macrophages within rodent spleens (Stevens-Felten et al.,
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
1998). Peripheral SNS denervation by the neurotoxin 6hydroxydopamine (6-OHDA) enhanced mitogen-induced T cell proliferation and macrophage phagocytosis (Lyte et al., 1991). In vitro, NE inhibits type 1 helper T cells (RamerQuinn et al., 1997; Sanders et al., 1997), stimulates antigenspecific B cell precursors (Sanders and Powell-Oliver, 1992), and alters cytokine, chemokine and nitric oxide production by macrophages (Hasko et al., 1998; Boomershine et al., 1999; Szelenyi et al., 2000). NE also can be a powerful chemoattractant for monocytes in vitro (Straub et al., 2000). Thus, NE has many immunomodulatory effects. Few studies have assessed NE involvement in modulation of host resistance, but it was recently demonstrated that NE significantly suppresses host resistance to Listeria monocytogenes (LM) in vivo (Rice et al., 2001). Both glucocorticoids and NE have been proposed to be important components in a two-stage model explaining stress modulation of immune responses, i.e. stress activates certain circuits and/or areas in the central nervous system (CNS), such as the limbic system and the hypothalamus, leading to the activation of the HPA axis and peripheral SNS. These activations result in the production of stress hormones, such as glucocorticoids and NE, which in turn modulate certain immune responses in the periphery, such as host susceptibility to infectious agents. In agreement with this model, glucocorticoids and NE have been shown to be involved in many stress-induced effects on the immune system. For example, adrenalectomy, which eliminates glucocorticoids, prevented acute restraint stress-induced changes in blood leukocyte distribution (Dhabhar et al., 1996). Modulation of allergic dermatitis by acute restraint stress in B6.129 mice was, in part, CORT-dependent (Flint et al., 2000). The CORT receptor antagonist RU486 prevented chronic restraint stressinduced mononuclear cell trafficking and alterations in cytokine responses to influenza infection in B6 mice (Hermann et al., 1995; Dobbs et al., 1996). NE was involved in restraint stress-induced alteration of antibody responses (Okimura et al., 1986). Peripheral SNS denervation markedly attenuated the cytokine response to immobilization stress in CD rats (Takaki et al., 1994). Important bi-directional regulation between corticoids and NE exists such that endogenous glucocorticoids restrain catecholamine turnover, synthesis, and release in the presence of immobilization stress (Kvetnansky et al., 1993). The interactions between CORT and NE can further affect stressinduced effects mediated by these hormones. In terms of modulation of host resistance, CORT and NE can mediate stress-induced effects in either an antagonistic or agonistic fashion. For example, in DBA/2 mice stress-induced glucocorticoids were suggested to provide protective effects, whereas SNS has been shown to play an important role in limiting the activation of virus-specific effector cells, thus suppressing host defense (Hermann et al., 1994). Glucocorticoids and NE together were responsible for restraint stressinduced enhancement of the skin DTH response (Dhabhar and McEwen, 1998).
95
Herein, we used a combination of acute cold/restraint stress (ACRS) and LM infection to investigate whether CORT and/or NE are involved in ACRS inhibition of host resistance. ACRS is a stress model widely used in physiology, pharmacology, immunology, and behavioral neurobiology (Glavin et al., 1994), and it likely elicits physical and psychological stress. LM is an intracellular pathogen used for studying the cell-mediated host resistance in mice. The CORT synthesis inhibitor metyrapone and the chemical sympathectomy drug 6-OHDA aided the elucidation of CORT vs. NE involvement in the observed ACRS suppression.
2. Materials and methods 2.1. Animals and acute/cold restraint stress ACRS Male BALB/cByJ mice (Wadsworth Center) were housed at four per cage in a specified pathogen-free environment with food and water ad libitum. All animals were maintained on a 12-h light/dark cycle with lights on from 7 AM to 7 PM. Animals were allowed at least 1 week of habituation in the animal room before use in experiments. For ACRS, 10-weekold mice were individually restrained in well-ventilated plastic 60-ml syringes (Sherwood Medical, St. Louis, MO) at 4 C for 1 h. Mice can move forward and backward in the syringe, but cannot turn head to tail. The ACRS was always performed between 10 AM and 12 AM. Control mice were left in their original cages undisturbed during the same time period. The day at which mice were given ACRS was considered day 0 for all experiments. All animal procedures were approved by the IACUC of Wadsworth Center (protocol # 00-278). 2.2. Drug administration All drugs were injected intraperitoneally (IP) and the respective vehicles were given to the control animals at the same time. 6-Hydroxydopamine HBr (6-OHDA, Sigma, St. Louis, MO) was dissolved in sterile 0.01% ascorbic acid/ saline and injected (200 mg/kg body weight) on day-3. Desipramine HCl (Sigma) was prepared in sterile saline and injected (25 mg/kg body weight) 30 min before 6-OHDA or vehicle treatment. Metyrapone (Sigma) was dissolved in sterile saline and injected (200 mg/kg body weight) 2 h before ACRS on day 0. 6-OHDA selectively enters and destroys sympathetic noradrenergic terminals, thus depleting tissue catecholamines. It does not cross a mature blood –brain barrier, thereby not having an effect on the central catecholamines. The 6-OHDA dose used here can significantly reduce the basal tissue NE levels and decrease splenic NE about 80% for up to 3 weeks (Kostrzewa and Jacobowitz, 1974; Kruszewska et al., 1995). Since 6-OHDA has to enter NE terminals in order to exert its neurotoxicity, the use of a NE uptake blocker, such as desipramine, is a very effective way of minimizing the noradrenergic toxicity of 6-OHDA
96
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
(Kostrzewa and Jacobowitz, 1974). Metyrapone, an 11bhydroxylase inhibitor, blocks the conversion of 11b-deoxycorticosterone to CORT thus blocking CORT synthesis (Orth and Kovacs, 1998). Metyrapone has been used together with aminoglutethimide for complete chemical adrenalectomy (Plotsky and Sawchenko, 1987); however, by itself, metyrapone attenuates stress-induced CORT increase without affecting the basal CORT level (Orth and Kovacs, 1998). The metyrapone dose that we employed can suppress stressinduced CORT for about 8 h (Plotsky and Sawchenko, 1987).
Boston, MA) was added and tubes were incubated at 37 C for 60 min. A cold dextran-coated charcoal (200 ml, Sigma) suspension was added into cooled tubes followed by a 10-min incubation (on ice) and centrifugation (2000 g, for 15 min at 4 C). The supernatant (250 ml) was mixed with 3 ml scintillation fluid (Ecoscint H, National Diagnostics, Atlanta, GA) and counted in a Beckman LS 6500 scintillation counter (Fullerton, CA). CORT concentrations were determined from a standard curve and normalized with a normal mouse serum control.
2.3. Listeria monocytogens (LM) infection, sample collection and analysis
2.5. HPLC-ECD analysis of tissue NE
LM was originally isolated from a LM meningitis patient and passaged in mice for multiple (> 9) generations before LM stocks were prepared. A LM stock was maintained by passaging in mice and stored at 80 C at 108 –109 colony forming units (CFU)/ml. Mice were intravenously (IV) injected with 5 –6 103 CFU LM/mouse on day 0 immediately after ACRS. This dose of LM allows all mice (control and ACRS) to survive and to eventually clear LM. Mice were sacrificed on day 3 post-infection. Upon sacrifice, bloods were collected after CO2 anesthesia, and sera were obtained by centrifugation for a serum CORT radioimmunoassay (RIA). Spleens and livers were removed aseptically. A small piece of each spleen (for all experiments) and liver (only for desipramine/6-OHDA experiments) was frozen immediately for later NE analysis by high-performance liquid chromatography with electrochemical detection (HPLC-ECD). The remaining part of each organ was homogenized in lysis buffer (stock buffer 50 mM Tris, 120 mM NaCl and 4 mM Na2EDTA, pH 8.0, with freshly added 30 mM phenylmethylsulfonyl fluoride (PMSF), 42 mM leupeptin, 0.002% aprotinin and 3 mM NaN3). Serial dilutions of organ homogenates were plated on blood-agar plates for enumeration of viable LM. Because in some experiments, a piece of either spleen or liver was used for HPLC, viable LM numbers are presented on the tissue weight basis (i.e. viable LM number in organ = viable LM/weight (g) of organ fragment used for enumerating viable LM). The LM was assumed to be evenly distributed. 2.4. CORT RIA A rabbit antiserum against corticosterone-21-thyroglobulin (Sigma) was used for the CORT RIA as described previously (Kim et al., 1999). Briefly, sera were deproteinized by precipitation with absolute ethanol (serum: ethanol = 1:4) followed by centrifugation (500 g, 10 min). Supernatants (10 ml) were mixed with 90 ml of assay buffer (0.05 M Tris – HCl containing 0.1 M NaCl, 0.1%NaN3 and 0.1% bovine serum albumin, pH 8.0). CORT (ICN, Aurora, OH) standard was prepared and diluted in 100 ml assay buffer. Diluted antiserum (500 ml) was then added into each sample and standard tube, and all tubes were incubated at room temperature for 30 min. [3H]-CORT (15 pg in 100 ml, NET,
Following respective treatment, spleens (all experiments) and livers (subset of experiments) were rapidly removed on ice. A piece of spleen (approximately one quarter of the spleen, always from the head portion of the spleen) and a piece of liver (approximately 60 mg, always from the end portion of the same lobe) tissue were collected from each mouse, weighed, sonicated in 500 ml saline, and rapidly frozen at 80 C. At the day of NE analysis, samples were thawed, and 100 ml of each sample homogenate was mixed with an equal volume of 0.4 N perchloric acid, containing EGTA (100 mg/l). Samples were then centrifuged at 12,000 rpm in a microcentrifuge for 2 min. Supernatants (50 ml) of each sample were analyzed for NE by HPLC-ECD with chromatographic conditions and equipment as previously described (Seegal et al., 1986), with the pH of the mobile phase being 4.07. As an additional between-runs quality control standard, pooled spleen samples from normal mice (N) and from 6-OHDA-treated mice (depleted, D) were generated, and at least one N and one D sample vial were included in each run. The overall between-runs CV was < 5%. Data for both spleen and liver NE is presented as ng NE per mg organ. 2.6. Statistical analysis In cases where three or more groups were compared, appropriate analysis of variance (ANOVA) was used initially. If a significant main effect or interaction was identified by the ANOVA analysis (p < 0.05), the respective group means were compared using the Student –Newman –Keuls (SNK) post-hoc test. In experiments involving two groups only, Student’s t-test was employed.
3. Results 3.1. ACRS and host resistance to LM infection after ACRS Mice restrained in the 60-ml syringes at 4 C for 1 h had significantly ( p < 0.001) increased serum CORT levels until 1 h after ending ACRS (Fig. 1A). By 3 h after ACRS, the serum CORT was back to the normal level. The change in CORT level suggests a temporary ACRS-induced activation of the
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
97
Fig. 1. Serum CORT (A) and host resistance to LM (B) after ACRS. Male BALB/c mice (10 weeks old) were individually restrained in 60-ml syringes at 4 C for 1 h. Control non-stressed mice remained in their original cages without disturbance for the same time period. Bloods were collected by cardiac puncture after CO2 anesthesia from control and stressed animals at 0, 1 and 3 h after ACRS. Serum CORT was measured by RIA. Data were collected from three to four mice/group and are presented as mean F SEM. One-way ANOVA and SNK post-hoc analysis were used and a vs. b p < 0.001. At 0 h after ACRS, mice were infected with LM (5 – 6 103 CFU/mouse, IV). As an index of host resistance, viable LM numbers were determined at day 3 post-infection by plating serial dilutions of spleen (white) and liver (gray) homogenates from control (open) and ACRS-treated (hatched) mice onto blood agar plates. Data were collected from six to seven mice/ group and are presented as mean F SEM. Student’s t-test was used for comparison within spleen and liver data and * indicates control vs. ACRS p < 0.01.
HPA axis. To determine the host resistance of the stressed mice, LM (5 –6 103 CFU / mouse, IV) was given immediately after ACRS, and the mice were sacrificed at day 3 postinfection. Preliminary experiments indicated that the LM burden in spleen and liver peaked on day 3. Both spleens and livers were removed for enumerating viable LM numbers by plating serial diluted homogenates onto blood-agar plates. Stressed mice had a higher LM burden in both spleen and liver ( p < 0.002) than the control mice (Fig. 1B). The greater LM burden after ACRS indicates that stress significantly suppressed host resistance to LM infection. 3.2. CORT and NE levels after metyrapone and 6-OHDA pretreatments Since stress hormones are important immune response modulators and have been implicated in many stress-induced effects, the roles of CORT and NE in our ACRS/infection model were investigated. Metyrapone and 6-OHDA were used to inhibit the ACRS-induced increase of serum CORT and to eliminate peripheral SNS innervation, respectively. The basal levels of serum CORT and splenic NE in stressed and non-stressed animals after the drug administrations were determined. Mice were randomly assigned into eight groups: vehicle control, metyrapone-pretreatment, 6-OHDA-pretreatment and both metyrapone and 6-OHDA-pretreatment, and each of these four groups included both a non-stress control and ACRS-treatment. Serum CORT (Fig. 2A) and splenic NE (Fig. 2B) were measured immediately after ACRS. Metyrapone or 6-OHDA did not affect serum CORT levels in non-stressed mice, while metyrapone significantly blocked (p < 0.001) the stress-induced increase of serum CORT up to 90%. Metyrapone did not affect splenic NE levels either with or without ACRS. 6-OHDA decreased
( p < 0.001) the basal splenic NE levels up to 80%, and this decrease was preserved after ACRS. There was no significant interaction between metyrapone and 6-OHDA for either serum CORT or splenic NE levels ( p>0.1). There was a slight, decrease of splenic NE after ACRS, which has been reported previously (Sudo, 1985). This decrease has been suggested to be due to organ release into the circulation. 3.3. LM burden in metyrapone and/or 6-OHDA pretreated mice We further investigated whether the inhibition of CORT or NE could prevent ACRS-impaired host resistance to LM infection. Mice were randomly divided into eight groups and pretreated with metyrapone and 6-OHDA as described earlier. All mice were infected with LM (5 –6 103 CFU/ mouse, IV) immediately after ACRS and sacrificed at day 3 after infection. Spleens and livers were collected for enumerating viable LM burden (Fig. 3A and B). As we initially observed, the LM burdens were significantly greater in stressed mice than in control mice, in both spleen ( p < 0.05) and liver ( p < 0.02). In the absence of stress, no difference in LM burden in either organ was found between metyraponetreated and vehicle-treated mice, while after ACRS, metyrapone further decreased the host resistance to LM in stressed mice in both spleen ( p < 0.001) and liver ( p < 0.05). This suggests a protective effect of ACRS-induced CORT in defense against the LM infection, as suggested in host resistance to influenza (Hermann et al., 1994). On the other hand, 6-OHDA administration caused a dramatic organdependent effect. In spleen, 6-OHDA significantly ( p < 0.05) enhanced the host resistance of both non-stressed and stressed animals (Fig. 3A). The enhancement of spleen host resistance by 6-OHDA in the absence of stress was consistent
98
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
Fig. 2. Serum CORT (A) and NE (B) after metyrapone and 6-OHDA pretreatments. Male BALB/c mice (10 weeks old) were pretreated with metyrapone (200 mg/kg body weight IP, 2 h before ACRS) and/or 6-OHDA (200 mg/kg body weight IP, 3 days before ACRS). Respective vehicles were given at the same time. At day 0, mice were given ACRS (gray bars) or control (white bars) treatments and then sacrificed immediately after ACRS period. Serum CORT (A) and splenic NE (B) were measured by RIA and HPLC-ECD, respectively. Data were collected from five mice/group and are presented as mean F SEM. Three-way ANOVA (ACRS, metyrapone and 6-OHDA) was followed by SNK post-hoc analysis to compare different treatment groups. Letters a and b indicate group differences within ACRS or control groups ( p < 0.05) and * indicates stress effects within each drug treatment ( p < 0.05). Note that ‘‘met + OHDA’’ denotes metyrapone and 6-OHDA.
with a previous study (Rice et al., 2001). However, in liver, 6OHDA alone significantly ( p < 0.05) enhanced the host resistance of non-stressed control mice (Fig. 3B), but only moderately affected ACRS-treated mice (i.e. a stress-induced increase in LM numbers in livers was still seen in the presence of 6-OHDA), suggesting an incomplete blockage of the ACRS effects on host resistance (Fig. 3B). Treatment with both metyrapone and 6-OHDA produced no significant interaction between metyrapone and 6-OHDA in both spleen and liver in either non-stressed control or ACRS-treated mice (Fig. 3A and B). However, in liver, metyrapone administration had a tendency to increase LM burden, which was
decreased by 6-OHDA (Fig. 3B). Serum CORT levels and splenic NE levels at day 3 post-infection (data not shown) were measured to assess whether early or later changes in concentration could be responsible for the host defense changes. In all animals, serum CORT levels were at nearbasal levels, which is consistent with the idea that the low dose of LM infection is a minimal stressor. Splenic NE levels showed that 6-OHDA-treated animals had a similar degree of denervation as we measured at day 0. Vehicle- and metyrapone-treated animals also showed a significant ( p < 0.001) decrease in splenic NE levels from the level before infection, i.e., from about 0.6 to 0.1 ng/mg tissue in all non-6-OHDA
Fig. 3. Viable LM number in organs after LM infection of metyrapone and/or 6-OHDA pretreated mice. Male BALB/c mice were pretreated as described in Fig. 2. Immediately after ACRS or control period (day 0), the mice were infected with LM (5 – 6 103 CFU/mouse, IV), and 3 days later, all mice were sacrificed for enumerating viable LM in spleens (A) and livers (B), as described in Materials and methods. Data were obtained from 9 to 17 mice/group and are presented as mean F SEM. Three-way ANOVA (ACRS, metyrapone and 6-OHDA) was followed by SNK post-hoc analysis to compare different treatment groups. Letters a, b and c indicate group differences within ACRS or control groups ( p < 0.05) and * indicates stress effects within each drug treatment ( p < 0.05). Note that ‘‘met + OHDA’’ denotes metyrapone and 6-OHDA.
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
99
Fig. 4. Splenic NE levels after desipramine and/ or 6-OHDA pretreatments. Male BALB/c mice (10 weeks old) were pretreated with desipramine (25 mg/kg body weight IP, 30 min before 6-OHDA administration) and/or 6-OHDA (200 mg/kg body weight IP, 3 days before ACRS). Respective vehicles were given at the same time by IP. At day 0, mice were given ACRS (gray bars) or control (white bars) treatments and then sacrificed immediately after ACRS. Spleen (A) and liver (B) NE were measured by HPLC-ECD. Data were collected from five mice/group and are presented as mean F SEM. Three-way ANOVA (ACRS, 6OHDA and desipramine) was followed by SNK post-hoc analysis to compare different treatment groups. Letters a and b indicate group differences within ACRS or control groups ( p < 0.05). Note that ‘‘des + OHDA’’ denotes desipramine and 6-OHDA.
treated groups, which also had been reported in earlier study (Kim et al., 1999). This decrease suggests an interesting phenomenon, i.e., animals tend to down-regulate tissue NE levels after infection to aid host resistance. 3.4. LM burden in desipramine/6-OHDA pretreated mice Since NE appeared to play more of a major role in the ACRS-induced reduction of host resistance to LM infection than CORT, we needed to confirm the basis for the 6-OHDA effects. By pretreating mice with the 6-OHDA uptake-inhib-
itor desipramine, support is provided that the inhibition of the ACRS effect by 6-OHDA was due to the selective uptake of 6-OHDA into sympathetic nerve terminals and the subsequent depletion of tissue NE rather than due to direct toxic effects of 6-OHDA. Due to the observed organ-dependent differences, both spleen and liver NE basal levels were measured at day 0 (the day of ACRS) after 6-OHDA and desipramine treatments (Fig. 4). In both spleen and liver, 6OHDA significantly eliminated sympathetic nerve innervation (Fig. 4A and B) ( p < 0.001 and p < 0.02 in spleen and liver, respectively). Desipramine by itself did not affect
Fig. 5. Viable LM number in organs after LM infection in desipramine and 6-OHDA pretreated mice. Male BALB/c mice were pretreated as described in Fig. 4. Subsequently, mice were infected with LM (5 – 6 103 CFU/mouse, IV), and 3 days later, all mice were sacrificed and viable LM numbers in spleen (A) and liver (B) were determined as described above. Data were collected from 5 to 17 mice/group and are presented as mean F SEM. Three-way ANOVA (ACRS, 6-OHDA and desipramine) was followed by SNK post-hoc analysis to compare different treatment groups. Letters a and b indicate group differences within ACRS or control groups ( p < 0.05) and * indicates stress effects within each drug treatment ( p < 0.05). Note that ‘‘des + OHDA’’ denotes desipramine and 6-OHDA.
100
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
spleen or liver NE levels in both control and stressed mice, while it almost completely prevented NE depletion by 6OHDA in both spleen and liver ( p < 0.001) (Fig. 4A and B). Interestingly, in the liver, there also was a significant ( p < 0.001) decrease in NE after stress (Fig. 4B). In addition, the basal liver NE level was much lower than that in spleen (about 10-fold), suggesting that sympathetic nerve innervation was much less in liver (on ng/mg basis) compared to that in spleen or there is either less release of NE or more rapid catabolism of NE in the liver. These differences may be contributory to the differences in the organ-dependent stress effects in the presence of 6-OHDA. Desipramine, on its own, did not significantly affect LM burdens in either spleen (Fig. 5A) or liver (Fig. 5B) under both basal and ACRS conditions. However, desipramine pretreatment significantly prevented the ability of 6-OHDA to improve host resistance in non-stressed mice in both spleen and liver LM burdens ( p < 0.05). Under ACRS conditions, however, desipramine significantly reversed the 6-OHDA effects on LM burdens in the spleen ( p < 0.05) and completely failed to do so in the liver.
4. Discussion This study showed that ACRS substantially impaired host resistance to LM in BALB/c mice, and chemical sympathectomy by 6-OHDA blocked the stress effect completely in spleen and partially in liver, indicating that the stress hormone NE from SNS plays a major role in ACRS-mediated suppression of overall host resistance. The CORT synthesis inhibitor metyrapone actually lowered the host resistance of ACRS-treated mice, suggesting that CORT from the stressinduced activation of HPA axis may have protective effects in defense against LM infection. It is well known that noradrenergic nerve fibers are present in many peripheral tissues, including both primary and secondary lymphoid organs and that lymphocytes, macrophages, and neutrophils express certain types of adrenergic receptors (Stevens-Felten and Bellinger, 1997). Multiple studies have shown that NE modulates immune responses both in vitro and in vivo (Josefsson et al., 1994; Madden et al., 1994, 2000; Delrue-Perollet et al., 1995; Kruszewska et al., 1995; Leo et al., 1998; Sanders, 1998; Kohm and Sanders, 1999). Moreover, NE is suggested to be responsible for various stress-induced changes including stress-induced peripheral cytokine elevation, activation of virus-specific effector cells, and DTH responses (Hermann et al., 1994; Takaki et al., 1994; Dhabhar et al., 1996). In the current study, NE appeared to mediate the ACRS-impairment of host resistance, which could involve several possible mechanisms. First, neutrophil recruitment may be modified. LM is a facultative intracellular bacterium. Neutrophils are cells of the first-line of defense. They migrate into infected areas, lysing infected cells, and thus controlling and confining the infection before phagocytosis by macrophages and activation of
adaptive immunity. It has been reported that host resistance to LM was enhanced by 6-OHDA in the absence of stress, as a result of an increased number of neutrophils infiltrating the infected organs during the innate stage of immunity (Rice et al., 2001). We postulate that, after ACRS, NE from the activated SNS inhibits neutrophil migration into LM targeted organs, leading to increased LM burdens in these organs. NE depletion blocks this effect by allowing an increase in neutrophils in these infected sites, thus significantly enhancing host resistance in both control and stressed animals. The possible underlying mechanisms include a direct inhibitory action of NE on neutrophils through their adrenergic receptors and/or down-regulation of endothelial adhesion molecules such as ICAM-1 or chemokine levels within the infected organs (Gabriel and Kindermann, 1998; Goebel and Mills, 2000). In addition, defects in macrophages may also be responsible. The number of LM peaks on day 3, the critical transition time from innate immunity to adaptive immunity. Macrophages play an important role in the phagocytosis of LM and LM-infected cells, and in the presentation of bacterial antigens to helper T cells. NE can affect the macrophage functions through their adrenergic receptors by inhibiting bactericidal activity or responsiveness to IFNg (the macrophage-activating cytokine secreted mainly by NK cells and type 1 helper T (Th1) cells) (Koff and Dunegan, 1986; Hasko et al., 1998; Boomershine et al., 1999; Szelenyi et al., 2000), thereby suppressing host resistance. Alternatively, NE action on neutrophils and macrophages may be receptor-independent, particularly under conditions of stress. One of the consequences of stress is a temporary state of ischemia driven by the stress-induced increase of circulating NE. In this regard, Koch et al. (1996) have demonstrated that infusion of NE in rabbits before bacterial infection impairs bacterial clearance with a net result of increased tissue burdens of bacteria. These authors suggest that NE-driven ischemia causes impairment in phagocytosis of bacteria by macrophages and neutrophils. Finally, inhibition of Th1 cell function may be the cause of lowered host resistance. NE has been shown to selectively inhibit Th1 cells (Ramer-Quinn et al., 1997; Sanders et al., 1997; Sanders, 1998). Th1 cells are the major stimulators of adaptive immune responses dealing with intracellular pathogens, including LM. Although at day 3 post-infection, innate immunity is still the dominant immune response in LM infection, dysfunction of Th1 cells may contribute to some of the stress effects. Chemical sympathectomy blocked ACRS effects completely in spleen but only partially in liver. These organdependent effects may be due to differential sympathetic nerve innervation and/or local regulatory factors. Spleen and liver NE levels (Fig. 4) clearly showed that spleen has about 10-fold higher levels of NE (based on ng/mg tissue weight) than liver, therefore, spleen denervation would have much greater consequences to the animal than a denervation of the liver. For instance, stress-induced NE can up-regulate IL-10 but down-regulate IFNg levels (Elenkov et al., 2000), and IL-10 has been reported to significantly enhanced host
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
defense in the spleen but not in the liver (Samsom et al., 2000). NE depletion was almost complete in both spleen (75 – 80%) and liver (> 90%). Hence, incomplete depletion (e.g. due to different vascular structure or anatomical diffusion barrier causing a differential distribution of 6-OHDA between organs; Kostrzewa and Jacobowitz, 1974) is not a likely explanation for this local effect. With regard to local factors, it has been found that stress slightly reduced tissue NE levels in some organs (Sudo, 1985), which is shown in greater degree in the liver than in the spleen (Fig. 4). This reduction in NE may lessen the influence of denervation in the liver after ACRS. It also provides evidence that liver responds to ACRS differently from spleen. Various factors could exist in liver to directly or indirectly influence NEmodulated changes, such factors may include local endothelia expressing adhesion molecules and resident macrophage products, such as reactive oxygen species. The Kupffer cells of the liver may respond differently from the macrophages of the spleen. The results with desipramine support the contention that 6OHDA exerted its effects through uptake by nerve terminals. Desipramine is a tricyclic antidepressant, which blocks 6OHDA effects by preventing its uptake into adrenergic nerve terminals. Although the restoration of tissue NE was complete in both spleen and liver (Fig. 4A and B), desipramine completely reversed the 6-OHDA effects in non-stressed animals, but not in the stressed animals (Fig. 5A and B). These results may be due to the effects of 6-OHDA on postsynaptic adrenergic receptor sites (Kostrzewa and Jacobowitz, 1974). 6-OHDA can either form a covalent bond in the region of adrenergic receptor or cause conformational changes on receptor sites to block NE action. These effects can last up to 5 days for a-adrenergic receptors. Thus, due to the lack of available receptors, ACRS-induced NE release cannot exert the same significant effects in desipramine/6OHDA pretreated animals as in non-treated animals after ACRS. Metyrapone-treated mice had less host resistance to LM infection compared with vehicle-treated mice after ACRS. This result suggests that the short-term increase of CORT by the ACRS may protect animals from the further stress of infection, possibly by dampening the stress-induced release of NE (Kvetnansky et al., 1993). The use of CORT or dexamethasone supplementation and glucocorticoid receptor antagonist administration will help to confirm this potential effect. However, the possible suppressive effects by the accumulated deoxycorticosterone after metyrapone treatment cannot be excluded at this point. In summary, we suggest that NE plays a major role in ACRS inhibition of host resistance to LM infection in BALB/c mice, while a short-term elevation of the CORT concentration has a potential preventive effect. Stress can substantially modulate host defense, and stress hormones are important mediators. Moreover, the balance between different stress hormones and their kinetics, and the interaction between stress hormones and organ-dependent local factors
101
are critical in determining the net effects of stress on the host resistance.
Acknowledgements We thank Dr. Richard F. Seegal for assistance with the norepinephrine assessment by HPLC-EDC.
References Ader, R., Cohen, N., 1993. Psychoneuroimmunology: conditioning and stress. Ann. Rev. Psychol. 44, 53 – 85. Biondi, M., Zannino, L.-G., 1997. Psychological stress, neuroimmunomodulation, and susceptibility to infectious diseases in animals and man: a review. Psychother. Psychosom. 66, 3 – 26. Boomershine, C.S., Lafuse, W.P., Zwilling, B.S., 1999. Beta 2-adrenergic receptor stimulation inhibits nitric oxide generation by mycobacterium avium infected macrophages. J. Neuroimmunol. 101, 68 – 75. Brown, D.H., Zwilling, B.S., 1994. Activation of the hypothalamic-pituitary-adrenal axis differentially affects the anti-mycobacterial activity of macrophages from BCG-resistant and susceptible mice. J. Neuroimmunol. 53, 181 – 187. Cohen, S., Tyrrell, D.A.J., Smith, A.P., 1991. Psychological stress and susceptibility to the common cold. N. Engl. J. Med. 325, 606 – 612. Delrue-Perollet, C., Li, K.-S., Vitiello, S., Neveu, P.J., 1995. Peripheral catecholamine are involved in the neuroendocrine and immune effects of LPS. Brain Behav. Immun. 9, 149 – 162. Dhabhar, F.S., McEwen, B.S., 1998. Enhancing versus suppressive effects of stress hormones on skin immune function. Proc. Natl. Acad. Sci. U. S. A. 96, 1059 – 1064. Dhabhar, F.S., Miller, A.H., McEwen, B.S., Spencer, R.L., 1996. Stressinduced changes in blood leukocyte distribution role of adrenal steroid hormones. J. Immunol. 157, 1638 – 1644. Dobbs, C.M., Feng, N., Beck, F.M., Sheridan, J.F., 1996. Neuroendocrine regulation of cytokine production during experimental influenza viral infection effects of restraint stress-induced elevation in endogenous corticosterone. J. Immunol. 157, 1870 – 1877. Elenkov, I.J., Wilder, R.L., Chrousos, G.P., Vizi, E.S., 2000. The sympathetic nerve—an integrative interface between two supersystems: the brain and the immune system. Pharmacol. Rev. 52, 595 – 638. Flint, M.S., Miller, D.B., Tinkle, S.S., 2000. Restraint-induced modulation of allergic and irritant contact dermatitis in male and female B6.129 mice. Brain Behav. Immun. 14, 256 – 269. Fricchione, G.L., Stefano, G.B., 1994. The stress response and autoimmunoregulation. Adv. Neuroimmunol. 4, 13 – 27. Gabriel, H., Kindermann, W., 1998. Adhesion molecules during immune response to exercise. Can. J. Physiol. Pharmacol. 76, 512 – 523. Glavin, G.B., Pare, W.P., Sandbak, T., Bakke, H.-K., Murison, R., 1994. Restraint stress in biomedical research: an update. Neurosci. Biobehav. Rev. 18, 223 – 249. Goebel, M.U., Mills, P.J., 2000. Acute psychological stress and exercise and changes in peripheral leukocyte adhesion molecule expression and density. Psychosom. Med. 62, 664 – 670. Hasko, G., Shanley, T.P., Egnaczyk, G., Nemeth, Z.H., Salzman, A.L., Vizi, E.S., Szabo, C., 1998. Exogenous and endogenous catecholamines inhibit the production of macrophage inflammatory protein (MIP) 1-alpha via a beta adrenoceptor mediated mechanism. Br. J. Pharmacol. 125, 1297 – 1303. Hermann, G., Beck, F.M., Tovar, C.A., Malarkey, W.B., Allen, C., Sheridan, J.F., 1994. Stress-induced changes attributable to the sympathetic nervous system during experimental influenza viral infection in DAB/2 inbred mouse strain. J. Neuroimmunol. 53, 173 – 180. Hermann, G., Beck, F.M., Sheridan, J.F., 1995. Stress-induced glucocorti-
102
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
coid response modulates mononuclear cell trafficking during an experimental influenza viral infection. J. Neuroimmunol. 56, 179 – 186. Herrmann, M., Scholmerich, J., Straub, R.H., 2000. Stress and rheumatic diseases. Rheum. Dis. Clin. North Am. 26, 737 – 763 viii. Josefsson, E., Mansson, J.-E., Blennow, K., Tarkowski, A., 1994. Immunomodulating and anti-inflammatory properties of the sympatholytic compound 6-hydroxydopamine. J. Neuroimmunol. 55, 161 – 169. Joyce, D.A., Steer, J.H., Abraham, L.J., 1997. Glucocorticoid modulation of human monocyte/macrophage function: control of TNF-alpha secretion. Inflammation Res. 46, 447 – 451. Kim, D., Carlson, J.N., Seegal, R.F., Lawrence, D.A., 1999. Differential immune responses in mice with left- and right-turning preference. J. Neuroimmunol. 99, 164 – 171. Koch, T., Heller, S., van Ackern, K., Schiefer, H.G., Neuhof, H., 1996. Impairment of bacterial clearance induced by norepinephrine infusion in rabbits. Intensive Care Med. 22, 637 – 643. Koff, W.C., Dunegan, M.A., 1986. Neuroendocrine hormones suppress macrophage-mediated lysis of herpes simplex virus-infected cells. J. Immunol. 136, 705 – 709. Kohm, A.P., Sanders, V.M., 1999. Suppression of antigen-specific Th2 celldependent IgM and IgG1 production following norepinephrine depletion in vivo. J. Immunol. 162, 5299 – 5308. Kostrzewa, R.M., Jacobowitz, D.M., 1974. Pharmacological actions of 6hydroxydopamine. Pharmacol. Rev. 26, 199 – 288. Kruszewska, B., Felten, S.Y., Moynihan, J.A., 1995. Alterations in cytokine and antibody production following chemical sympathectomy in two strains of mice. J. Immunol. 155, 4613 – 4620. Kvetnansky, R., Fukuhara, K., Pacak, K., Cizza, G., Goldstein, D.S., Kopin, I.J., 1993. Endogenous glucocorticoids restrain catecholamine synthesis and release at rest and during immobilization stress in rats. Endocrinology 133, 1411 – 1419. Leo, N.A., Callahan, T.A., Bonneau, R.H., 1998. Peripheral sympathetic denervation alters both the primary and memory cellular immune responses to herpes simplex virus infection. NeuroImmunoModulation 5, 22 – 35. Lyte, M., Ernst, S., Driemeyer, J., Aissa, B., 1991. Strain-specific enhancement of splenic T cell mitogenesis and macrophage phagocytosis following peripheral axotomy. J. Neuroimmunol. 31, 1 – 8. Madden, K.S., Felten, S.Y., Felten, D.L., Hardy, C.A., Livnat, S., 1994. Sympathetic nervous system modulation of the immune system: II. Induction of lymphocyte proliferation and migration in vivo by chemical sympathectomy. J. Neuroimmunol. 49, 67 – 75. Madden, K.S., Stevens, S.Y., Felten, D.L., Bellinger, D.L., 2000. Alteration in T lymphocyte activity following chemical sympathectomy in young and old Fischer 344 rats. J. Neuroimmunol. 103, 131 – 145. McEwen, B.S., Biron, C.A., Brunson, K.W., Bulloch, K., Chambers, W.H., Dhabhar, F.S., Goldfarb, R.H., Kitson, R.P., Miller, A.H., Spencer, R.L., Weiss, J.M., 1997. The role of adrenocorticoids as modulators of immune function in health and disease: neural, endocrine and immune interactions. Brain Res. Rev. 23, 79 – 133. Okimura, T., Ogawa, M., Yamauchi, T., Sasaki, Y., 1986. Stress and immune responses IV. Adrenal involvement in the alteration of antibody responses in restraint-stressed mice. Jpn. J. Pharmacol. 41, 237 – 245. Orth, D.N., Kovacs, W.J., 1998. The adrenal cortex. In: Wilson, J.D., Foster, D.W., Kronenberg, H.M., Larsen, P.R. (Eds.), Section 4-12 Williams Textbook of Endocrinology, 9th edn. W.B. Saunders, Philadelphia, p. 620. Plotsky, P.K., Sawchenko, P.E., 1987. Hypophysial-portal plasma levels, median eminence content, and immunohistochemical staining of corticotrophin-releasing factor, arginine vasopressin, and oxytocin after pharmacological adrenalectomy. Endocrinology 120, 1361 – 1369.
Ramer-Quinn, D.S., Baker, R.A., Sanders, V.M., 1997. Activated T helper 1 and T helper 2 cells differentially express the beta2-adrenergic receptor a mechanism for selective modulation of T helper1 cell cytokine production. J. Immunol. 159, 4857 – 4867. Rice, P.A., Boehm, G.W., Moynihan, J.A., Bellinger, D.L., Stevens, S.Y., 2001. Chemical sympathectomy increases the innate immune response and decreases the specific immune response in the spleen to infection with Listeria monocytogenes. J. Neuroimmunol. 114, 19 – 27. Samsom, J.N., Annema, A., Greertsma, M.F., Langermans, J.A.M., Groeneveld, P.H.P., De Heer, E., Van Furth, R., 2000. Interleukin-10 has different effects on proliferation of Listeria monocytogenes in livers and spleens of mice. Infect. Immun. 68, 4666 – 4672. Sanders, V.M., 1998. The role of norepinephrine and beta-2-adrenergic receptor stimulation in the modulation of Th1, Th2, and B lymphocyte function. Adv. Exp. Med. Biol. 437, 269 – 278. Sanders, V.M., Powell-Oliver, F.E., 1992. Beta2-adrenoceptor stimulation increases the number of antigen-specific precursor B lymphocytes that differentiate into IgM-secreting cells without affecting burst size. J. Immunol. 148, 1822 – 1828. Sanders, V.M., Baker, R.A., Ramer-Quinn, D.S., Kasprowicz, D.J., Fuchs, B.A., Street, N.E., 1997. Differential expression of the beta2-adrenergic receptor by Th1 and Th2 clones implications for cytokine production and B cell help. J. Immunol. 158, 4200 – 4210. Seegal, R.F., Brosch, K.O., Bush, B., 1986. High-performance liquid chromatography of biogenic amines and metabolites in brain, cerebrospinal fluid, urine and plasma. J. Chromatogr. 377, 131 – 144. Sternberg, E.M., 2001. Neuroendocrine regulation of autoimmune/inflammatory disease. J. Endocrinol. 169, 429 – 435. Stevens-Felten, S.Y., Bellinger, D.L., 1997. Noradrenergic and peptidergic innervation of lymphoid organs. In: Blalock, J.E. (Ed.), Neuroimmunoendocrinology, 3rd rev. edn. Chem. Immunol. vol. 69, Karger, Basel, pp. 99 – 131. Stevens-Felten, S.Y., Madden, K.S., Bellinger, D.L., Kruszewska, B., Moynihan, J.A., Felten, D.L., 1998. The role of the sympathetic nervous system in the modulation of immune responses. Adv. Pharmacol. 42, 583 – 587. Straub, R.H., Mayer, M., Kreutz, M., Leeb, S., Scholmerich, J., Falk, W., 2000. Neurotransmitters of the sympathetic nerve terminal are powerful chemoattractants for monocytes. J. Leukocyte Biol. 67, 553 – 558. Sudo, A., 1985. Accumulation of adrenaline in sympathetic nerve endings in various organs of the rat exposed to swimming stress. Jpn. J. Pharmacol. 1985, 367 – 374. Szelenyi, J., Kiss, J.P., Vizi, E.S., 2000. Differential involvement of sympathetic nervous system and immune system in the modulation of TNFalpha production by alpha (2)- and beta-adrenoceptors in mice. J. Neuroimmunol. 103, 34 – 40. Takaki, A., Huang, Q.-H., Somogyvari-Vigh, A., Arimura, A., 1994. Immobilization stress may increase plasma interleukin-6 via central and peripheral catecholamines. NeuroImmunoModulation 1, 335 – 342. Walker, J.G., Littlejohn, G.O., McMurray, N.E., Cutolo, M., 1999. Stress system response and rheumatoid arthritis: a multilevel approach. Rheumatology 38, 1050 – 1057. Wilckens, T., De Rijk, R., 1997. Glucocorticoids and immune function: unknown dimensions and new frontiers. Immunol. Today 18, 418 – 424. Wu, W., Yamaura, T., Murakami, K., Murata, J., Matsumoto, K., Watanabe, H., Saiki, I., 2000. Social isolation stress enhanced liver metastasis of murine colon 26-l5 carcinoma cells by suppressing immune responses in mice. Life Sci. 66, 1827 – 1838.
Sympathetic nervous system plays a major role in acute cold/restraint stress inhibition of host resistance to Listeria monocytogenes Ling Cao, Nikolay M. Filipov 1, David A. Lawrence* Wadsworth Center, New York State Department of Health, Empire State Plaza, P.O. Box 509, Albany, NY 12201-0509, USA Received 26 October 2001; received in revised form 28 January 2002; accepted 30 January 2002
Abstract BALB/c mice exposed to acute cold/restraint stress (ACRS) had significantly lower host resistance to Listeria monocytogenes (LM) than controls. The stress hormones corticosterone (CORT) and norepinephrine (NE), which are known to modulate immune responses, were evaluated as the cause of the decline in immune defense. The involvement of CORT and NE was investigated by pretreating mice with the CORT synthesis inhibitor metyrapone and the chemical sympathectomy drug 6-hydroxydopamine (6-OHDA), respectively. LM burdens in spleen and liver were determined three days post-infection. 6-OHDA significantly decreased the LM burden in both control and stressed animals. 6-OHDA also completely blocked the stress effects observed in spleens while only partially affecting the liver. The 6-OHDA-uptake inhibitor desipramine aided confirmation that peripheral sympathetic adrenergic nerves and NE depletion, rather than the direct action of 6OHDA, were responsible for the decreased susceptibility to LM. The results suggest that the peripheral sympathetic nervous system (SNS) postganglionic neurotransmitter NE plays a major role in LM host resistance and has significant tissue-dependent effects after ACRS. In contrast, metyrapone-treated animals had further decreased host resistance to LM, suggesting a potential protective effect of CORT after ACRS. Altogether, the results suggest that stress hormones play an important role in stress-modulated host resistance and that NE is the major hormone involved in ACRS-induced suppression of host resistance. D 2002 Published by Elsevier Science B.V. Keywords: Cold/restraint stress; Listeria monocytogenes; Host resistance; Corticosterone; Norepinephrine; Desipramine
1. Introduction Stress, as a state of disharmony or threatened homeostasis resulting from external or internal challenge (Fricchione and Stefano, 1994), has been associated with a variety of diseases including infectious diseases, neurodegenerative diseases, and cancers (Ader and Cohen, 1993; Biondi and Zannino, 1997; Wu et al., 2000). Although psychological stress also has been indicated in exacerbation of autoimmune diseases (Walker et al., 1999; Herrmann et al., 2000), inadequate stress hormone responses have been associated with enhancement of autoimmune diseases (Sternberg, 2001). In terms of infectious disease, stress has been shown to be involved in the onset, progression, and outcome in both human and animal studies. The stress-induced effect is dependent on
*
Corresponding author. Tel.: +1-518-402-5684; fax: +1-518-474-1412. E-mail address: [email protected] (D.A. Lawrence). 1 Current address: Center for Environmental Health Sciences, Department of Basic Sciences, College of Veterinary Medicine, Mississippi State University, Box 6100, Mississippi State, MS 39762, USA. 0165-5728/02/$ - see front matter D 2002 Published by Elsevier Science B.V. PII: S 0 1 6 5 - 5 7 2 8 ( 0 2 ) 0 0 0 3 9 - 5
the pathogen and the particular immune responses involved (Cohen et al., 1991; Biondi and Zannino, 1997), but the mechanisms by which stress modulates host resistance to infectious diseases have not been well delineated. Stress involves activation of both the hypothalamic-pituitary-adrenal (HPA) axis and the sympathetic nervous system (SNS) (Fricchione and Stefano, 1994). Repeated studies have demonstrated that both glucocorticoids and norepinephrine (NE) can modulate immune responses at different levels. In vitro and ex vivo, glucocorticoids suppress T cell proliferation in response to mitogens or antigens, natural killer (NK) cell activity, macrophage phagocytosis, and cytokine secretion from T cells and macrophages (Brown and Zwilling, 1994; Joyce et al., 1997; McEwen et al., 1997; Wilckens and De Rijk, 1997). In vivo, glucocorticoids modulate leukocyte trafficking and delayed-type hypersensitivity (DTH) responses (Dhabhar et al., 1996; Dhabhar and McEwen, 1998). With regard to NE, it is well known that SNS innervates both primary and secondary lymphoid organs. There are direct contacts between SNS nerve terminals (detected by anti-tyrosine hydroxylase antibody) and lymphocytes and macrophages within rodent spleens (Stevens-Felten et al.,
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
1998). Peripheral SNS denervation by the neurotoxin 6hydroxydopamine (6-OHDA) enhanced mitogen-induced T cell proliferation and macrophage phagocytosis (Lyte et al., 1991). In vitro, NE inhibits type 1 helper T cells (RamerQuinn et al., 1997; Sanders et al., 1997), stimulates antigenspecific B cell precursors (Sanders and Powell-Oliver, 1992), and alters cytokine, chemokine and nitric oxide production by macrophages (Hasko et al., 1998; Boomershine et al., 1999; Szelenyi et al., 2000). NE also can be a powerful chemoattractant for monocytes in vitro (Straub et al., 2000). Thus, NE has many immunomodulatory effects. Few studies have assessed NE involvement in modulation of host resistance, but it was recently demonstrated that NE significantly suppresses host resistance to Listeria monocytogenes (LM) in vivo (Rice et al., 2001). Both glucocorticoids and NE have been proposed to be important components in a two-stage model explaining stress modulation of immune responses, i.e. stress activates certain circuits and/or areas in the central nervous system (CNS), such as the limbic system and the hypothalamus, leading to the activation of the HPA axis and peripheral SNS. These activations result in the production of stress hormones, such as glucocorticoids and NE, which in turn modulate certain immune responses in the periphery, such as host susceptibility to infectious agents. In agreement with this model, glucocorticoids and NE have been shown to be involved in many stress-induced effects on the immune system. For example, adrenalectomy, which eliminates glucocorticoids, prevented acute restraint stress-induced changes in blood leukocyte distribution (Dhabhar et al., 1996). Modulation of allergic dermatitis by acute restraint stress in B6.129 mice was, in part, CORT-dependent (Flint et al., 2000). The CORT receptor antagonist RU486 prevented chronic restraint stressinduced mononuclear cell trafficking and alterations in cytokine responses to influenza infection in B6 mice (Hermann et al., 1995; Dobbs et al., 1996). NE was involved in restraint stress-induced alteration of antibody responses (Okimura et al., 1986). Peripheral SNS denervation markedly attenuated the cytokine response to immobilization stress in CD rats (Takaki et al., 1994). Important bi-directional regulation between corticoids and NE exists such that endogenous glucocorticoids restrain catecholamine turnover, synthesis, and release in the presence of immobilization stress (Kvetnansky et al., 1993). The interactions between CORT and NE can further affect stressinduced effects mediated by these hormones. In terms of modulation of host resistance, CORT and NE can mediate stress-induced effects in either an antagonistic or agonistic fashion. For example, in DBA/2 mice stress-induced glucocorticoids were suggested to provide protective effects, whereas SNS has been shown to play an important role in limiting the activation of virus-specific effector cells, thus suppressing host defense (Hermann et al., 1994). Glucocorticoids and NE together were responsible for restraint stressinduced enhancement of the skin DTH response (Dhabhar and McEwen, 1998).
95
Herein, we used a combination of acute cold/restraint stress (ACRS) and LM infection to investigate whether CORT and/or NE are involved in ACRS inhibition of host resistance. ACRS is a stress model widely used in physiology, pharmacology, immunology, and behavioral neurobiology (Glavin et al., 1994), and it likely elicits physical and psychological stress. LM is an intracellular pathogen used for studying the cell-mediated host resistance in mice. The CORT synthesis inhibitor metyrapone and the chemical sympathectomy drug 6-OHDA aided the elucidation of CORT vs. NE involvement in the observed ACRS suppression.
2. Materials and methods 2.1. Animals and acute/cold restraint stress ACRS Male BALB/cByJ mice (Wadsworth Center) were housed at four per cage in a specified pathogen-free environment with food and water ad libitum. All animals were maintained on a 12-h light/dark cycle with lights on from 7 AM to 7 PM. Animals were allowed at least 1 week of habituation in the animal room before use in experiments. For ACRS, 10-weekold mice were individually restrained in well-ventilated plastic 60-ml syringes (Sherwood Medical, St. Louis, MO) at 4 C for 1 h. Mice can move forward and backward in the syringe, but cannot turn head to tail. The ACRS was always performed between 10 AM and 12 AM. Control mice were left in their original cages undisturbed during the same time period. The day at which mice were given ACRS was considered day 0 for all experiments. All animal procedures were approved by the IACUC of Wadsworth Center (protocol # 00-278). 2.2. Drug administration All drugs were injected intraperitoneally (IP) and the respective vehicles were given to the control animals at the same time. 6-Hydroxydopamine HBr (6-OHDA, Sigma, St. Louis, MO) was dissolved in sterile 0.01% ascorbic acid/ saline and injected (200 mg/kg body weight) on day-3. Desipramine HCl (Sigma) was prepared in sterile saline and injected (25 mg/kg body weight) 30 min before 6-OHDA or vehicle treatment. Metyrapone (Sigma) was dissolved in sterile saline and injected (200 mg/kg body weight) 2 h before ACRS on day 0. 6-OHDA selectively enters and destroys sympathetic noradrenergic terminals, thus depleting tissue catecholamines. It does not cross a mature blood –brain barrier, thereby not having an effect on the central catecholamines. The 6-OHDA dose used here can significantly reduce the basal tissue NE levels and decrease splenic NE about 80% for up to 3 weeks (Kostrzewa and Jacobowitz, 1974; Kruszewska et al., 1995). Since 6-OHDA has to enter NE terminals in order to exert its neurotoxicity, the use of a NE uptake blocker, such as desipramine, is a very effective way of minimizing the noradrenergic toxicity of 6-OHDA
96
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
(Kostrzewa and Jacobowitz, 1974). Metyrapone, an 11bhydroxylase inhibitor, blocks the conversion of 11b-deoxycorticosterone to CORT thus blocking CORT synthesis (Orth and Kovacs, 1998). Metyrapone has been used together with aminoglutethimide for complete chemical adrenalectomy (Plotsky and Sawchenko, 1987); however, by itself, metyrapone attenuates stress-induced CORT increase without affecting the basal CORT level (Orth and Kovacs, 1998). The metyrapone dose that we employed can suppress stressinduced CORT for about 8 h (Plotsky and Sawchenko, 1987).
Boston, MA) was added and tubes were incubated at 37 C for 60 min. A cold dextran-coated charcoal (200 ml, Sigma) suspension was added into cooled tubes followed by a 10-min incubation (on ice) and centrifugation (2000 g, for 15 min at 4 C). The supernatant (250 ml) was mixed with 3 ml scintillation fluid (Ecoscint H, National Diagnostics, Atlanta, GA) and counted in a Beckman LS 6500 scintillation counter (Fullerton, CA). CORT concentrations were determined from a standard curve and normalized with a normal mouse serum control.
2.3. Listeria monocytogens (LM) infection, sample collection and analysis
2.5. HPLC-ECD analysis of tissue NE
LM was originally isolated from a LM meningitis patient and passaged in mice for multiple (> 9) generations before LM stocks were prepared. A LM stock was maintained by passaging in mice and stored at 80 C at 108 –109 colony forming units (CFU)/ml. Mice were intravenously (IV) injected with 5 –6 103 CFU LM/mouse on day 0 immediately after ACRS. This dose of LM allows all mice (control and ACRS) to survive and to eventually clear LM. Mice were sacrificed on day 3 post-infection. Upon sacrifice, bloods were collected after CO2 anesthesia, and sera were obtained by centrifugation for a serum CORT radioimmunoassay (RIA). Spleens and livers were removed aseptically. A small piece of each spleen (for all experiments) and liver (only for desipramine/6-OHDA experiments) was frozen immediately for later NE analysis by high-performance liquid chromatography with electrochemical detection (HPLC-ECD). The remaining part of each organ was homogenized in lysis buffer (stock buffer 50 mM Tris, 120 mM NaCl and 4 mM Na2EDTA, pH 8.0, with freshly added 30 mM phenylmethylsulfonyl fluoride (PMSF), 42 mM leupeptin, 0.002% aprotinin and 3 mM NaN3). Serial dilutions of organ homogenates were plated on blood-agar plates for enumeration of viable LM. Because in some experiments, a piece of either spleen or liver was used for HPLC, viable LM numbers are presented on the tissue weight basis (i.e. viable LM number in organ = viable LM/weight (g) of organ fragment used for enumerating viable LM). The LM was assumed to be evenly distributed. 2.4. CORT RIA A rabbit antiserum against corticosterone-21-thyroglobulin (Sigma) was used for the CORT RIA as described previously (Kim et al., 1999). Briefly, sera were deproteinized by precipitation with absolute ethanol (serum: ethanol = 1:4) followed by centrifugation (500 g, 10 min). Supernatants (10 ml) were mixed with 90 ml of assay buffer (0.05 M Tris – HCl containing 0.1 M NaCl, 0.1%NaN3 and 0.1% bovine serum albumin, pH 8.0). CORT (ICN, Aurora, OH) standard was prepared and diluted in 100 ml assay buffer. Diluted antiserum (500 ml) was then added into each sample and standard tube, and all tubes were incubated at room temperature for 30 min. [3H]-CORT (15 pg in 100 ml, NET,
Following respective treatment, spleens (all experiments) and livers (subset of experiments) were rapidly removed on ice. A piece of spleen (approximately one quarter of the spleen, always from the head portion of the spleen) and a piece of liver (approximately 60 mg, always from the end portion of the same lobe) tissue were collected from each mouse, weighed, sonicated in 500 ml saline, and rapidly frozen at 80 C. At the day of NE analysis, samples were thawed, and 100 ml of each sample homogenate was mixed with an equal volume of 0.4 N perchloric acid, containing EGTA (100 mg/l). Samples were then centrifuged at 12,000 rpm in a microcentrifuge for 2 min. Supernatants (50 ml) of each sample were analyzed for NE by HPLC-ECD with chromatographic conditions and equipment as previously described (Seegal et al., 1986), with the pH of the mobile phase being 4.07. As an additional between-runs quality control standard, pooled spleen samples from normal mice (N) and from 6-OHDA-treated mice (depleted, D) were generated, and at least one N and one D sample vial were included in each run. The overall between-runs CV was < 5%. Data for both spleen and liver NE is presented as ng NE per mg organ. 2.6. Statistical analysis In cases where three or more groups were compared, appropriate analysis of variance (ANOVA) was used initially. If a significant main effect or interaction was identified by the ANOVA analysis (p < 0.05), the respective group means were compared using the Student –Newman –Keuls (SNK) post-hoc test. In experiments involving two groups only, Student’s t-test was employed.
3. Results 3.1. ACRS and host resistance to LM infection after ACRS Mice restrained in the 60-ml syringes at 4 C for 1 h had significantly ( p < 0.001) increased serum CORT levels until 1 h after ending ACRS (Fig. 1A). By 3 h after ACRS, the serum CORT was back to the normal level. The change in CORT level suggests a temporary ACRS-induced activation of the
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
97
Fig. 1. Serum CORT (A) and host resistance to LM (B) after ACRS. Male BALB/c mice (10 weeks old) were individually restrained in 60-ml syringes at 4 C for 1 h. Control non-stressed mice remained in their original cages without disturbance for the same time period. Bloods were collected by cardiac puncture after CO2 anesthesia from control and stressed animals at 0, 1 and 3 h after ACRS. Serum CORT was measured by RIA. Data were collected from three to four mice/group and are presented as mean F SEM. One-way ANOVA and SNK post-hoc analysis were used and a vs. b p < 0.001. At 0 h after ACRS, mice were infected with LM (5 – 6 103 CFU/mouse, IV). As an index of host resistance, viable LM numbers were determined at day 3 post-infection by plating serial dilutions of spleen (white) and liver (gray) homogenates from control (open) and ACRS-treated (hatched) mice onto blood agar plates. Data were collected from six to seven mice/ group and are presented as mean F SEM. Student’s t-test was used for comparison within spleen and liver data and * indicates control vs. ACRS p < 0.01.
HPA axis. To determine the host resistance of the stressed mice, LM (5 –6 103 CFU / mouse, IV) was given immediately after ACRS, and the mice were sacrificed at day 3 postinfection. Preliminary experiments indicated that the LM burden in spleen and liver peaked on day 3. Both spleens and livers were removed for enumerating viable LM numbers by plating serial diluted homogenates onto blood-agar plates. Stressed mice had a higher LM burden in both spleen and liver ( p < 0.002) than the control mice (Fig. 1B). The greater LM burden after ACRS indicates that stress significantly suppressed host resistance to LM infection. 3.2. CORT and NE levels after metyrapone and 6-OHDA pretreatments Since stress hormones are important immune response modulators and have been implicated in many stress-induced effects, the roles of CORT and NE in our ACRS/infection model were investigated. Metyrapone and 6-OHDA were used to inhibit the ACRS-induced increase of serum CORT and to eliminate peripheral SNS innervation, respectively. The basal levels of serum CORT and splenic NE in stressed and non-stressed animals after the drug administrations were determined. Mice were randomly assigned into eight groups: vehicle control, metyrapone-pretreatment, 6-OHDA-pretreatment and both metyrapone and 6-OHDA-pretreatment, and each of these four groups included both a non-stress control and ACRS-treatment. Serum CORT (Fig. 2A) and splenic NE (Fig. 2B) were measured immediately after ACRS. Metyrapone or 6-OHDA did not affect serum CORT levels in non-stressed mice, while metyrapone significantly blocked (p < 0.001) the stress-induced increase of serum CORT up to 90%. Metyrapone did not affect splenic NE levels either with or without ACRS. 6-OHDA decreased
( p < 0.001) the basal splenic NE levels up to 80%, and this decrease was preserved after ACRS. There was no significant interaction between metyrapone and 6-OHDA for either serum CORT or splenic NE levels ( p>0.1). There was a slight, decrease of splenic NE after ACRS, which has been reported previously (Sudo, 1985). This decrease has been suggested to be due to organ release into the circulation. 3.3. LM burden in metyrapone and/or 6-OHDA pretreated mice We further investigated whether the inhibition of CORT or NE could prevent ACRS-impaired host resistance to LM infection. Mice were randomly divided into eight groups and pretreated with metyrapone and 6-OHDA as described earlier. All mice were infected with LM (5 –6 103 CFU/ mouse, IV) immediately after ACRS and sacrificed at day 3 after infection. Spleens and livers were collected for enumerating viable LM burden (Fig. 3A and B). As we initially observed, the LM burdens were significantly greater in stressed mice than in control mice, in both spleen ( p < 0.05) and liver ( p < 0.02). In the absence of stress, no difference in LM burden in either organ was found between metyraponetreated and vehicle-treated mice, while after ACRS, metyrapone further decreased the host resistance to LM in stressed mice in both spleen ( p < 0.001) and liver ( p < 0.05). This suggests a protective effect of ACRS-induced CORT in defense against the LM infection, as suggested in host resistance to influenza (Hermann et al., 1994). On the other hand, 6-OHDA administration caused a dramatic organdependent effect. In spleen, 6-OHDA significantly ( p < 0.05) enhanced the host resistance of both non-stressed and stressed animals (Fig. 3A). The enhancement of spleen host resistance by 6-OHDA in the absence of stress was consistent
98
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
Fig. 2. Serum CORT (A) and NE (B) after metyrapone and 6-OHDA pretreatments. Male BALB/c mice (10 weeks old) were pretreated with metyrapone (200 mg/kg body weight IP, 2 h before ACRS) and/or 6-OHDA (200 mg/kg body weight IP, 3 days before ACRS). Respective vehicles were given at the same time. At day 0, mice were given ACRS (gray bars) or control (white bars) treatments and then sacrificed immediately after ACRS period. Serum CORT (A) and splenic NE (B) were measured by RIA and HPLC-ECD, respectively. Data were collected from five mice/group and are presented as mean F SEM. Three-way ANOVA (ACRS, metyrapone and 6-OHDA) was followed by SNK post-hoc analysis to compare different treatment groups. Letters a and b indicate group differences within ACRS or control groups ( p < 0.05) and * indicates stress effects within each drug treatment ( p < 0.05). Note that ‘‘met + OHDA’’ denotes metyrapone and 6-OHDA.
with a previous study (Rice et al., 2001). However, in liver, 6OHDA alone significantly ( p < 0.05) enhanced the host resistance of non-stressed control mice (Fig. 3B), but only moderately affected ACRS-treated mice (i.e. a stress-induced increase in LM numbers in livers was still seen in the presence of 6-OHDA), suggesting an incomplete blockage of the ACRS effects on host resistance (Fig. 3B). Treatment with both metyrapone and 6-OHDA produced no significant interaction between metyrapone and 6-OHDA in both spleen and liver in either non-stressed control or ACRS-treated mice (Fig. 3A and B). However, in liver, metyrapone administration had a tendency to increase LM burden, which was
decreased by 6-OHDA (Fig. 3B). Serum CORT levels and splenic NE levels at day 3 post-infection (data not shown) were measured to assess whether early or later changes in concentration could be responsible for the host defense changes. In all animals, serum CORT levels were at nearbasal levels, which is consistent with the idea that the low dose of LM infection is a minimal stressor. Splenic NE levels showed that 6-OHDA-treated animals had a similar degree of denervation as we measured at day 0. Vehicle- and metyrapone-treated animals also showed a significant ( p < 0.001) decrease in splenic NE levels from the level before infection, i.e., from about 0.6 to 0.1 ng/mg tissue in all non-6-OHDA
Fig. 3. Viable LM number in organs after LM infection of metyrapone and/or 6-OHDA pretreated mice. Male BALB/c mice were pretreated as described in Fig. 2. Immediately after ACRS or control period (day 0), the mice were infected with LM (5 – 6 103 CFU/mouse, IV), and 3 days later, all mice were sacrificed for enumerating viable LM in spleens (A) and livers (B), as described in Materials and methods. Data were obtained from 9 to 17 mice/group and are presented as mean F SEM. Three-way ANOVA (ACRS, metyrapone and 6-OHDA) was followed by SNK post-hoc analysis to compare different treatment groups. Letters a, b and c indicate group differences within ACRS or control groups ( p < 0.05) and * indicates stress effects within each drug treatment ( p < 0.05). Note that ‘‘met + OHDA’’ denotes metyrapone and 6-OHDA.
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
99
Fig. 4. Splenic NE levels after desipramine and/ or 6-OHDA pretreatments. Male BALB/c mice (10 weeks old) were pretreated with desipramine (25 mg/kg body weight IP, 30 min before 6-OHDA administration) and/or 6-OHDA (200 mg/kg body weight IP, 3 days before ACRS). Respective vehicles were given at the same time by IP. At day 0, mice were given ACRS (gray bars) or control (white bars) treatments and then sacrificed immediately after ACRS. Spleen (A) and liver (B) NE were measured by HPLC-ECD. Data were collected from five mice/group and are presented as mean F SEM. Three-way ANOVA (ACRS, 6OHDA and desipramine) was followed by SNK post-hoc analysis to compare different treatment groups. Letters a and b indicate group differences within ACRS or control groups ( p < 0.05). Note that ‘‘des + OHDA’’ denotes desipramine and 6-OHDA.
treated groups, which also had been reported in earlier study (Kim et al., 1999). This decrease suggests an interesting phenomenon, i.e., animals tend to down-regulate tissue NE levels after infection to aid host resistance. 3.4. LM burden in desipramine/6-OHDA pretreated mice Since NE appeared to play more of a major role in the ACRS-induced reduction of host resistance to LM infection than CORT, we needed to confirm the basis for the 6-OHDA effects. By pretreating mice with the 6-OHDA uptake-inhib-
itor desipramine, support is provided that the inhibition of the ACRS effect by 6-OHDA was due to the selective uptake of 6-OHDA into sympathetic nerve terminals and the subsequent depletion of tissue NE rather than due to direct toxic effects of 6-OHDA. Due to the observed organ-dependent differences, both spleen and liver NE basal levels were measured at day 0 (the day of ACRS) after 6-OHDA and desipramine treatments (Fig. 4). In both spleen and liver, 6OHDA significantly eliminated sympathetic nerve innervation (Fig. 4A and B) ( p < 0.001 and p < 0.02 in spleen and liver, respectively). Desipramine by itself did not affect
Fig. 5. Viable LM number in organs after LM infection in desipramine and 6-OHDA pretreated mice. Male BALB/c mice were pretreated as described in Fig. 4. Subsequently, mice were infected with LM (5 – 6 103 CFU/mouse, IV), and 3 days later, all mice were sacrificed and viable LM numbers in spleen (A) and liver (B) were determined as described above. Data were collected from 5 to 17 mice/group and are presented as mean F SEM. Three-way ANOVA (ACRS, 6-OHDA and desipramine) was followed by SNK post-hoc analysis to compare different treatment groups. Letters a and b indicate group differences within ACRS or control groups ( p < 0.05) and * indicates stress effects within each drug treatment ( p < 0.05). Note that ‘‘des + OHDA’’ denotes desipramine and 6-OHDA.
100
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
spleen or liver NE levels in both control and stressed mice, while it almost completely prevented NE depletion by 6OHDA in both spleen and liver ( p < 0.001) (Fig. 4A and B). Interestingly, in the liver, there also was a significant ( p < 0.001) decrease in NE after stress (Fig. 4B). In addition, the basal liver NE level was much lower than that in spleen (about 10-fold), suggesting that sympathetic nerve innervation was much less in liver (on ng/mg basis) compared to that in spleen or there is either less release of NE or more rapid catabolism of NE in the liver. These differences may be contributory to the differences in the organ-dependent stress effects in the presence of 6-OHDA. Desipramine, on its own, did not significantly affect LM burdens in either spleen (Fig. 5A) or liver (Fig. 5B) under both basal and ACRS conditions. However, desipramine pretreatment significantly prevented the ability of 6-OHDA to improve host resistance in non-stressed mice in both spleen and liver LM burdens ( p < 0.05). Under ACRS conditions, however, desipramine significantly reversed the 6-OHDA effects on LM burdens in the spleen ( p < 0.05) and completely failed to do so in the liver.
4. Discussion This study showed that ACRS substantially impaired host resistance to LM in BALB/c mice, and chemical sympathectomy by 6-OHDA blocked the stress effect completely in spleen and partially in liver, indicating that the stress hormone NE from SNS plays a major role in ACRS-mediated suppression of overall host resistance. The CORT synthesis inhibitor metyrapone actually lowered the host resistance of ACRS-treated mice, suggesting that CORT from the stressinduced activation of HPA axis may have protective effects in defense against LM infection. It is well known that noradrenergic nerve fibers are present in many peripheral tissues, including both primary and secondary lymphoid organs and that lymphocytes, macrophages, and neutrophils express certain types of adrenergic receptors (Stevens-Felten and Bellinger, 1997). Multiple studies have shown that NE modulates immune responses both in vitro and in vivo (Josefsson et al., 1994; Madden et al., 1994, 2000; Delrue-Perollet et al., 1995; Kruszewska et al., 1995; Leo et al., 1998; Sanders, 1998; Kohm and Sanders, 1999). Moreover, NE is suggested to be responsible for various stress-induced changes including stress-induced peripheral cytokine elevation, activation of virus-specific effector cells, and DTH responses (Hermann et al., 1994; Takaki et al., 1994; Dhabhar et al., 1996). In the current study, NE appeared to mediate the ACRS-impairment of host resistance, which could involve several possible mechanisms. First, neutrophil recruitment may be modified. LM is a facultative intracellular bacterium. Neutrophils are cells of the first-line of defense. They migrate into infected areas, lysing infected cells, and thus controlling and confining the infection before phagocytosis by macrophages and activation of
adaptive immunity. It has been reported that host resistance to LM was enhanced by 6-OHDA in the absence of stress, as a result of an increased number of neutrophils infiltrating the infected organs during the innate stage of immunity (Rice et al., 2001). We postulate that, after ACRS, NE from the activated SNS inhibits neutrophil migration into LM targeted organs, leading to increased LM burdens in these organs. NE depletion blocks this effect by allowing an increase in neutrophils in these infected sites, thus significantly enhancing host resistance in both control and stressed animals. The possible underlying mechanisms include a direct inhibitory action of NE on neutrophils through their adrenergic receptors and/or down-regulation of endothelial adhesion molecules such as ICAM-1 or chemokine levels within the infected organs (Gabriel and Kindermann, 1998; Goebel and Mills, 2000). In addition, defects in macrophages may also be responsible. The number of LM peaks on day 3, the critical transition time from innate immunity to adaptive immunity. Macrophages play an important role in the phagocytosis of LM and LM-infected cells, and in the presentation of bacterial antigens to helper T cells. NE can affect the macrophage functions through their adrenergic receptors by inhibiting bactericidal activity or responsiveness to IFNg (the macrophage-activating cytokine secreted mainly by NK cells and type 1 helper T (Th1) cells) (Koff and Dunegan, 1986; Hasko et al., 1998; Boomershine et al., 1999; Szelenyi et al., 2000), thereby suppressing host resistance. Alternatively, NE action on neutrophils and macrophages may be receptor-independent, particularly under conditions of stress. One of the consequences of stress is a temporary state of ischemia driven by the stress-induced increase of circulating NE. In this regard, Koch et al. (1996) have demonstrated that infusion of NE in rabbits before bacterial infection impairs bacterial clearance with a net result of increased tissue burdens of bacteria. These authors suggest that NE-driven ischemia causes impairment in phagocytosis of bacteria by macrophages and neutrophils. Finally, inhibition of Th1 cell function may be the cause of lowered host resistance. NE has been shown to selectively inhibit Th1 cells (Ramer-Quinn et al., 1997; Sanders et al., 1997; Sanders, 1998). Th1 cells are the major stimulators of adaptive immune responses dealing with intracellular pathogens, including LM. Although at day 3 post-infection, innate immunity is still the dominant immune response in LM infection, dysfunction of Th1 cells may contribute to some of the stress effects. Chemical sympathectomy blocked ACRS effects completely in spleen but only partially in liver. These organdependent effects may be due to differential sympathetic nerve innervation and/or local regulatory factors. Spleen and liver NE levels (Fig. 4) clearly showed that spleen has about 10-fold higher levels of NE (based on ng/mg tissue weight) than liver, therefore, spleen denervation would have much greater consequences to the animal than a denervation of the liver. For instance, stress-induced NE can up-regulate IL-10 but down-regulate IFNg levels (Elenkov et al., 2000), and IL-10 has been reported to significantly enhanced host
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
defense in the spleen but not in the liver (Samsom et al., 2000). NE depletion was almost complete in both spleen (75 – 80%) and liver (> 90%). Hence, incomplete depletion (e.g. due to different vascular structure or anatomical diffusion barrier causing a differential distribution of 6-OHDA between organs; Kostrzewa and Jacobowitz, 1974) is not a likely explanation for this local effect. With regard to local factors, it has been found that stress slightly reduced tissue NE levels in some organs (Sudo, 1985), which is shown in greater degree in the liver than in the spleen (Fig. 4). This reduction in NE may lessen the influence of denervation in the liver after ACRS. It also provides evidence that liver responds to ACRS differently from spleen. Various factors could exist in liver to directly or indirectly influence NEmodulated changes, such factors may include local endothelia expressing adhesion molecules and resident macrophage products, such as reactive oxygen species. The Kupffer cells of the liver may respond differently from the macrophages of the spleen. The results with desipramine support the contention that 6OHDA exerted its effects through uptake by nerve terminals. Desipramine is a tricyclic antidepressant, which blocks 6OHDA effects by preventing its uptake into adrenergic nerve terminals. Although the restoration of tissue NE was complete in both spleen and liver (Fig. 4A and B), desipramine completely reversed the 6-OHDA effects in non-stressed animals, but not in the stressed animals (Fig. 5A and B). These results may be due to the effects of 6-OHDA on postsynaptic adrenergic receptor sites (Kostrzewa and Jacobowitz, 1974). 6-OHDA can either form a covalent bond in the region of adrenergic receptor or cause conformational changes on receptor sites to block NE action. These effects can last up to 5 days for a-adrenergic receptors. Thus, due to the lack of available receptors, ACRS-induced NE release cannot exert the same significant effects in desipramine/6OHDA pretreated animals as in non-treated animals after ACRS. Metyrapone-treated mice had less host resistance to LM infection compared with vehicle-treated mice after ACRS. This result suggests that the short-term increase of CORT by the ACRS may protect animals from the further stress of infection, possibly by dampening the stress-induced release of NE (Kvetnansky et al., 1993). The use of CORT or dexamethasone supplementation and glucocorticoid receptor antagonist administration will help to confirm this potential effect. However, the possible suppressive effects by the accumulated deoxycorticosterone after metyrapone treatment cannot be excluded at this point. In summary, we suggest that NE plays a major role in ACRS inhibition of host resistance to LM infection in BALB/c mice, while a short-term elevation of the CORT concentration has a potential preventive effect. Stress can substantially modulate host defense, and stress hormones are important mediators. Moreover, the balance between different stress hormones and their kinetics, and the interaction between stress hormones and organ-dependent local factors
101
are critical in determining the net effects of stress on the host resistance.
Acknowledgements We thank Dr. Richard F. Seegal for assistance with the norepinephrine assessment by HPLC-EDC.
References Ader, R., Cohen, N., 1993. Psychoneuroimmunology: conditioning and stress. Ann. Rev. Psychol. 44, 53 – 85. Biondi, M., Zannino, L.-G., 1997. Psychological stress, neuroimmunomodulation, and susceptibility to infectious diseases in animals and man: a review. Psychother. Psychosom. 66, 3 – 26. Boomershine, C.S., Lafuse, W.P., Zwilling, B.S., 1999. Beta 2-adrenergic receptor stimulation inhibits nitric oxide generation by mycobacterium avium infected macrophages. J. Neuroimmunol. 101, 68 – 75. Brown, D.H., Zwilling, B.S., 1994. Activation of the hypothalamic-pituitary-adrenal axis differentially affects the anti-mycobacterial activity of macrophages from BCG-resistant and susceptible mice. J. Neuroimmunol. 53, 181 – 187. Cohen, S., Tyrrell, D.A.J., Smith, A.P., 1991. Psychological stress and susceptibility to the common cold. N. Engl. J. Med. 325, 606 – 612. Delrue-Perollet, C., Li, K.-S., Vitiello, S., Neveu, P.J., 1995. Peripheral catecholamine are involved in the neuroendocrine and immune effects of LPS. Brain Behav. Immun. 9, 149 – 162. Dhabhar, F.S., McEwen, B.S., 1998. Enhancing versus suppressive effects of stress hormones on skin immune function. Proc. Natl. Acad. Sci. U. S. A. 96, 1059 – 1064. Dhabhar, F.S., Miller, A.H., McEwen, B.S., Spencer, R.L., 1996. Stressinduced changes in blood leukocyte distribution role of adrenal steroid hormones. J. Immunol. 157, 1638 – 1644. Dobbs, C.M., Feng, N., Beck, F.M., Sheridan, J.F., 1996. Neuroendocrine regulation of cytokine production during experimental influenza viral infection effects of restraint stress-induced elevation in endogenous corticosterone. J. Immunol. 157, 1870 – 1877. Elenkov, I.J., Wilder, R.L., Chrousos, G.P., Vizi, E.S., 2000. The sympathetic nerve—an integrative interface between two supersystems: the brain and the immune system. Pharmacol. Rev. 52, 595 – 638. Flint, M.S., Miller, D.B., Tinkle, S.S., 2000. Restraint-induced modulation of allergic and irritant contact dermatitis in male and female B6.129 mice. Brain Behav. Immun. 14, 256 – 269. Fricchione, G.L., Stefano, G.B., 1994. The stress response and autoimmunoregulation. Adv. Neuroimmunol. 4, 13 – 27. Gabriel, H., Kindermann, W., 1998. Adhesion molecules during immune response to exercise. Can. J. Physiol. Pharmacol. 76, 512 – 523. Glavin, G.B., Pare, W.P., Sandbak, T., Bakke, H.-K., Murison, R., 1994. Restraint stress in biomedical research: an update. Neurosci. Biobehav. Rev. 18, 223 – 249. Goebel, M.U., Mills, P.J., 2000. Acute psychological stress and exercise and changes in peripheral leukocyte adhesion molecule expression and density. Psychosom. Med. 62, 664 – 670. Hasko, G., Shanley, T.P., Egnaczyk, G., Nemeth, Z.H., Salzman, A.L., Vizi, E.S., Szabo, C., 1998. Exogenous and endogenous catecholamines inhibit the production of macrophage inflammatory protein (MIP) 1-alpha via a beta adrenoceptor mediated mechanism. Br. J. Pharmacol. 125, 1297 – 1303. Hermann, G., Beck, F.M., Tovar, C.A., Malarkey, W.B., Allen, C., Sheridan, J.F., 1994. Stress-induced changes attributable to the sympathetic nervous system during experimental influenza viral infection in DAB/2 inbred mouse strain. J. Neuroimmunol. 53, 173 – 180. Hermann, G., Beck, F.M., Sheridan, J.F., 1995. Stress-induced glucocorti-
102
L. Cao et al. / Journal of Neuroimmunology 125 (2002) 94–102
coid response modulates mononuclear cell trafficking during an experimental influenza viral infection. J. Neuroimmunol. 56, 179 – 186. Herrmann, M., Scholmerich, J., Straub, R.H., 2000. Stress and rheumatic diseases. Rheum. Dis. Clin. North Am. 26, 737 – 763 viii. Josefsson, E., Mansson, J.-E., Blennow, K., Tarkowski, A., 1994. Immunomodulating and anti-inflammatory properties of the sympatholytic compound 6-hydroxydopamine. J. Neuroimmunol. 55, 161 – 169. Joyce, D.A., Steer, J.H., Abraham, L.J., 1997. Glucocorticoid modulation of human monocyte/macrophage function: control of TNF-alpha secretion. Inflammation Res. 46, 447 – 451. Kim, D., Carlson, J.N., Seegal, R.F., Lawrence, D.A., 1999. Differential immune responses in mice with left- and right-turning preference. J. Neuroimmunol. 99, 164 – 171. Koch, T., Heller, S., van Ackern, K., Schiefer, H.G., Neuhof, H., 1996. Impairment of bacterial clearance induced by norepinephrine infusion in rabbits. Intensive Care Med. 22, 637 – 643. Koff, W.C., Dunegan, M.A., 1986. Neuroendocrine hormones suppress macrophage-mediated lysis of herpes simplex virus-infected cells. J. Immunol. 136, 705 – 709. Kohm, A.P., Sanders, V.M., 1999. Suppression of antigen-specific Th2 celldependent IgM and IgG1 production following norepinephrine depletion in vivo. J. Immunol. 162, 5299 – 5308. Kostrzewa, R.M., Jacobowitz, D.M., 1974. Pharmacological actions of 6hydroxydopamine. Pharmacol. Rev. 26, 199 – 288. Kruszewska, B., Felten, S.Y., Moynihan, J.A., 1995. Alterations in cytokine and antibody production following chemical sympathectomy in two strains of mice. J. Immunol. 155, 4613 – 4620. Kvetnansky, R., Fukuhara, K., Pacak, K., Cizza, G., Goldstein, D.S., Kopin, I.J., 1993. Endogenous glucocorticoids restrain catecholamine synthesis and release at rest and during immobilization stress in rats. Endocrinology 133, 1411 – 1419. Leo, N.A., Callahan, T.A., Bonneau, R.H., 1998. Peripheral sympathetic denervation alters both the primary and memory cellular immune responses to herpes simplex virus infection. NeuroImmunoModulation 5, 22 – 35. Lyte, M., Ernst, S., Driemeyer, J., Aissa, B., 1991. Strain-specific enhancement of splenic T cell mitogenesis and macrophage phagocytosis following peripheral axotomy. J. Neuroimmunol. 31, 1 – 8. Madden, K.S., Felten, S.Y., Felten, D.L., Hardy, C.A., Livnat, S., 1994. Sympathetic nervous system modulation of the immune system: II. Induction of lymphocyte proliferation and migration in vivo by chemical sympathectomy. J. Neuroimmunol. 49, 67 – 75. Madden, K.S., Stevens, S.Y., Felten, D.L., Bellinger, D.L., 2000. Alteration in T lymphocyte activity following chemical sympathectomy in young and old Fischer 344 rats. J. Neuroimmunol. 103, 131 – 145. McEwen, B.S., Biron, C.A., Brunson, K.W., Bulloch, K., Chambers, W.H., Dhabhar, F.S., Goldfarb, R.H., Kitson, R.P., Miller, A.H., Spencer, R.L., Weiss, J.M., 1997. The role of adrenocorticoids as modulators of immune function in health and disease: neural, endocrine and immune interactions. Brain Res. Rev. 23, 79 – 133. Okimura, T., Ogawa, M., Yamauchi, T., Sasaki, Y., 1986. Stress and immune responses IV. Adrenal involvement in the alteration of antibody responses in restraint-stressed mice. Jpn. J. Pharmacol. 41, 237 – 245. Orth, D.N., Kovacs, W.J., 1998. The adrenal cortex. In: Wilson, J.D., Foster, D.W., Kronenberg, H.M., Larsen, P.R. (Eds.), Section 4-12 Williams Textbook of Endocrinology, 9th edn. W.B. Saunders, Philadelphia, p. 620. Plotsky, P.K., Sawchenko, P.E., 1987. Hypophysial-portal plasma levels, median eminence content, and immunohistochemical staining of corticotrophin-releasing factor, arginine vasopressin, and oxytocin after pharmacological adrenalectomy. Endocrinology 120, 1361 – 1369.
Ramer-Quinn, D.S., Baker, R.A., Sanders, V.M., 1997. Activated T helper 1 and T helper 2 cells differentially express the beta2-adrenergic receptor a mechanism for selective modulation of T helper1 cell cytokine production. J. Immunol. 159, 4857 – 4867. Rice, P.A., Boehm, G.W., Moynihan, J.A., Bellinger, D.L., Stevens, S.Y., 2001. Chemical sympathectomy increases the innate immune response and decreases the specific immune response in the spleen to infection with Listeria monocytogenes. J. Neuroimmunol. 114, 19 – 27. Samsom, J.N., Annema, A., Greertsma, M.F., Langermans, J.A.M., Groeneveld, P.H.P., De Heer, E., Van Furth, R., 2000. Interleukin-10 has different effects on proliferation of Listeria monocytogenes in livers and spleens of mice. Infect. Immun. 68, 4666 – 4672. Sanders, V.M., 1998. The role of norepinephrine and beta-2-adrenergic receptor stimulation in the modulation of Th1, Th2, and B lymphocyte function. Adv. Exp. Med. Biol. 437, 269 – 278. Sanders, V.M., Powell-Oliver, F.E., 1992. Beta2-adrenoceptor stimulation increases the number of antigen-specific precursor B lymphocytes that differentiate into IgM-secreting cells without affecting burst size. J. Immunol. 148, 1822 – 1828. Sanders, V.M., Baker, R.A., Ramer-Quinn, D.S., Kasprowicz, D.J., Fuchs, B.A., Street, N.E., 1997. Differential expression of the beta2-adrenergic receptor by Th1 and Th2 clones implications for cytokine production and B cell help. J. Immunol. 158, 4200 – 4210. Seegal, R.F., Brosch, K.O., Bush, B., 1986. High-performance liquid chromatography of biogenic amines and metabolites in brain, cerebrospinal fluid, urine and plasma. J. Chromatogr. 377, 131 – 144. Sternberg, E.M., 2001. Neuroendocrine regulation of autoimmune/inflammatory disease. J. Endocrinol. 169, 429 – 435. Stevens-Felten, S.Y., Bellinger, D.L., 1997. Noradrenergic and peptidergic innervation of lymphoid organs. In: Blalock, J.E. (Ed.), Neuroimmunoendocrinology, 3rd rev. edn. Chem. Immunol. vol. 69, Karger, Basel, pp. 99 – 131. Stevens-Felten, S.Y., Madden, K.S., Bellinger, D.L., Kruszewska, B., Moynihan, J.A., Felten, D.L., 1998. The role of the sympathetic nervous system in the modulation of immune responses. Adv. Pharmacol. 42, 583 – 587. Straub, R.H., Mayer, M., Kreutz, M., Leeb, S., Scholmerich, J., Falk, W., 2000. Neurotransmitters of the sympathetic nerve terminal are powerful chemoattractants for monocytes. J. Leukocyte Biol. 67, 553 – 558. Sudo, A., 1985. Accumulation of adrenaline in sympathetic nerve endings in various organs of the rat exposed to swimming stress. Jpn. J. Pharmacol. 1985, 367 – 374. Szelenyi, J., Kiss, J.P., Vizi, E.S., 2000. Differential involvement of sympathetic nervous system and immune system in the modulation of TNFalpha production by alpha (2)- and beta-adrenoceptors in mice. J. Neuroimmunol. 103, 34 – 40. Takaki, A., Huang, Q.-H., Somogyvari-Vigh, A., Arimura, A., 1994. Immobilization stress may increase plasma interleukin-6 via central and peripheral catecholamines. NeuroImmunoModulation 1, 335 – 342. Walker, J.G., Littlejohn, G.O., McMurray, N.E., Cutolo, M., 1999. Stress system response and rheumatoid arthritis: a multilevel approach. Rheumatology 38, 1050 – 1057. Wilckens, T., De Rijk, R., 1997. Glucocorticoids and immune function: unknown dimensions and new frontiers. Immunol. Today 18, 418 – 424. Wu, W., Yamaura, T., Murakami, K., Murata, J., Matsumoto, K., Watanabe, H., Saiki, I., 2000. Social isolation stress enhanced liver metastasis of murine colon 26-l5 carcinoma cells by suppressing immune responses in mice. Life Sci. 66, 1827 – 1838.