Brain, Behavior, and Immunity 36 (2014) 183–192

Contents lists available at ScienceDirect

Brain, Behavior, and Immunity journal homepage: www.elsevier.com/locate/ybrbi

Adrenergic and glucocorticoid modulation of the sterile inflammatory response Stewart S. Cox a, Kristin J. Speaker a, Lida A. Beninson a, Wendy C. Craig a, Madeline M. Paton a, Monika Fleshner a,b,⇑ a b

Department of Integrative Physiology, University of Colorado, Boulder, United States Center for Neuroscience, University of Colorado, Boulder, United States

a r t i c l e

i n f o

Article history: Received 12 August 2013 Received in revised form 21 November 2013 Accepted 26 November 2013 Available online 7 December 2013 Keywords: Acute stress Cytokine Monocyte chemotactic protein-1 Heat shock protein Catecholamines Glucocorticoids Sterile inflammation

a b s t r a c t Exposure to an intense, acute stressor, in the absence of a pathogen, alters immune function. Exposure to a single bout of inescapable tail shock increases plasma and tissue concentrations of cytokines, chemokines, and the danger associated molecular pattern (DAMP) Hsp72. Although previous studies have demonstrated that adrenergic receptor (ADR) and glucocorticoid receptor (GCR)-mediated pathways alter pathogen or microbial associated molecular pattern (MAMP)-evoked levels of cytokines, chemokines, and Hsp72, far fewer studies have tested the role of these receptors across multiple inflammatory proteins or tissues to elucidate the differences in magnitude of stress-evoked sterile inflammatory responses. The goals of the current study were to (1) compare the sterile inflammatory response in the circulation, liver, spleen, and subcutaneous (SQ) adipose tissue by measuring cytokine, chemokine, and DAMP (Hsp72) responses; and (2) to test the role of alpha-1 (a1), beta-1 (b1), beta-2 (b2), and beta-3 (b3) ADRs, as well as GCRs in signaling the sterile inflammatory response. The data presented indicate plasma and SQ adipose are significantly more stress responsive than the liver and spleen. Further, administration of ADR and GCR-specific antagonists revealed both similarities and differences in the signaling mechanisms of the sterile inflammatory response in the tissues studied. Finally, given the selective increase in the chemokine monocyte chemotactic protein-1 (MCP-1) in SQ tissue, it may be that SQ adipose is an important site of leukocyte migration, possibly in preparation for infection as a consequence of wounding. The current study helps further our understanding of the tissue-specific differences of the stress-induced sterile inflammatory response. Ó 2013 Elsevier Inc. All rights reserved.

1. Introduction The inflammatory response is an integral part of successful host immunity and is capable of being triggered by both pathogenic and non-pathogenic challenges. While pathogenic inflammation occurs as a result of exposure to bacteria and other microorganisms, inflammation in response to a non-pathogenic insult like stress, known as sterile inflammation, is stimulated from several signals, including danger associated molecular patterns (DAMPs) (Maslanik et al., 2013; Campisi and Fleshner, 2003). DAMPs are endogenous signals that originate from cells following damage or in response to danger or stress. The sterile inflammatory response could promote survival if triggered acutely in young, healthy organisms by activation of innate immunity that can prevent infection and facilitate healing consequent to potential wounding ⇑ Corresponding author. Address: University of Colorado, Department of Integrative Physiology, 1725 Pleasant Street, Clare Small Room 114, UCB 354, Boulder, CO 80309-0354, United States. Tel.: +1 303 492 1483; fax: +1 303 492 6778. E-mail address: fl[email protected] (M. Fleshner). 0889-1591/$ - see front matter Ó 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.bbi.2013.11.018

(Dhabhar, 2009; Fleshner, 2013). Furthermore, inflammatory proteins are crucial for the benefits of this stress-evoked sterile inflammatory response (Fleshner et al., 2006; Johnson et al., 2005b). Pro-inflammatory cytokines, such as interleukin-1 beta (IL-1b) (O’Connor et al., 2003) and interleukin-6 (IL-6) (Ando et al., 1998) can increase in the blood and in tissues in response to acute stressor exposure. Additionally, our lab, as well as others, have demonstrated that the anti-inflammatory cytokine, interleukin10 (IL-10) (Maslanik et al., 2012a), as well as the chemokine monocyte chemotactic protein-1 (MCP-1) (Maslanik et al., 2012b) and DAMPs (heat shock protein 72 (Hsp72), Campisi and Fleshner, 2003; Nickerson et al., 2006; uric acid, Maslanik et al., 2013) also increase after tail shock stress. Few studies, however, take into consideration the network of inflammatory proteins and DAMPs produced in vivo, as well as the interaction and communication between them during stressor exposure (Engler et al., 2005; Maslanik et al., 2012a). The sterile inflammatory response encompasses changes in proand anti-inflammatory cytokines, chemokines, DAMPs, and can

184

S.S. Cox et al. / Brain, Behavior, and Immunity 36 (2014) 183–192

occur in the circulation and numerous tissues throughout the body. For example, stressor exposure increases inflammatory proteins and/or Hsp72 in plasma (Fleshner et al., 2002, 2007; Zhou et al., 1993), liver (Kubes and Mehal, 2012), spleen (Maslanik et al., 2012a; You et al., 2011), and subcutaneous (SQ) adipose tissue (Coppack, 2001; Mohamed-Ali et al., 1998; Speaker et al., 2013). However, most of the previous work assessing stress-evoked sterile inflammation only measured responses in one or two tissues and a few simultaneously measured inflammatory proteins and DAMPs (Persoons et al., 1995; Pertsov et al., 2009; Zhou et al., 1993). Additionally, the neuroendocrine signals responsible for stressevoked sterile inflammation are poorly understood. It has been well established that microbe-associated molecular patterns (MAMPs), such as lipopolysaccharide (LPS), a cell wall component of gram-negative bacteria, potently activate a pathogen-evoked inflammatory response. Depending on the cytokine (i.e., inflammasome independent vs. dependent), MAMPs stimulate cytokines either alone or in concert with DAMPs (Griffiths et al., 1995; Netea et al., 2010). This pathogen-evoked response characterized by an increase in immunoregulatory factors such as cytokines, can be regulated by both arms of the stress response, the sympathetic nervous system (SNS) and the hypothalamic pituitary adrenal (HPA) axis. Although both of these neuroendocrine signaling pathways demonstrate complex regulation of the sterile inflammatory response depending on the timing and presence of MAMPs/DAMPs, numerous studies have illustrated catecholamines generally immunoenhance, while glucocorticoids generally immunosuppress. For example, catecholamines binding to adrenoceptors (ADR) potentiate LPS-induced levels of several cytokines in the plasma (Haskó et al., 1995; Pastores et al., 1996). There are also reports that glucocorticoids can reduce the pathogen-evoked inflammatory response due to their potent anti-inflammatory properties. For example, it has been observed that administration of dexamethasone, a synthetic glucocorticoid, reduced serum IL-1b and tumor necrosis factor-alpha (TNF-a) following an LPS challenge (Brattsand and Linden, 1996; Zuckerman et al., 1989; Steer et al., 2000). Far fewer studies have focused on the impacts of these neuroendocrine signals on levels of immunoregulatory factors during a pathogen-free, stress-evoked inflammatory response, although there is a growing body of research. We have reported, for example, that ADRs mediate stress-induced increases in IL-1b and IL-6 in the circulation and brain (Johnson et al., 2005b). Studies have also implicated alpha-1 (a1, located primarily on vascular endothelium; Elliott, 1997), beta-2 (b2, expressed on cardiac and muscle tissue; Rang, 2003) and beta-3 (b3, expressed on coronary and peripheral vasculature, with the highest density expressed on adipocytes; Dessy et al., 2005; Krief et al., 1993; Michel et al., 2011; Rath et al., 2012) ADRs in modulating sterile inflammatory increases in cytokines and chemokines. For example, the addition of a b2 ADR agonist to macrophages in vitro increased IL-1b and IL-6 production in the absence of MAMP signaling, like LPS (Tan et al., 2007). Further, sterile activation of b3 ADRs increased IL-6 and MCP-1 levels in adipose tissue (Mottillo et al., 2010). In contrast, the binding of glucocorticoids to the stress-responsive GCR, found on nearly every cell in the body, has potent immunosuppressive effects on monocytes and other mononuclear cells. These immunosuppressive effects can reduce cytokine levels after an administration of synthetic glucocorticoids in vivo (Kunicka et al., 1993). Only a small number of studies have explored the ADR and GCR-mediated regulation of DAMP levels during a stress-evoke sterile inflammatory response. Johnson et al. (2005a) reported that Hsp72 is released into the circulation in the face of acute stress via an a1 ADR-dependent mechanism. Although some of the

mechanisms have been parsed out, many components of sterile inflammation, such as the impact of b3 ADR and GCR-specific antagonists on inflammatory proteins and Hsp72, are understudied. In addition, a single study has not explored the neuroendocrine signaling responsible for changes in levels of numerous inflammatory proteins and Hsp72 in different peripheral tissues to observe the potential differences in the regulation of sterile inflammation. Based on previous work that an intense stressor can evoke a systemic sterile inflammatory response and evidence that suggests the response is modulated in whole or in part by catecholamines binding to ADR and glucocorticoids binding to GCR, the aim of the current study is to (1) compare the sterile inflammatory response in the circulation, liver, spleen, and SQ adipose tissue by measuring cytokine, chemokine, and DAMP (Hsp72) responses; and (2) to test the role of a1, b1, b2, and b3 ADRs, as well as GCR in signaling the sterile inflammatory response within each tissue. 2. Materials and methods 2.1. Subjects Adult, male, Fischer 344 rats (200–250 g, 8–9 weeks old) were maintained on a 12:12-h light–dark schedule (lights on from 0700 to 1900). Rats were allowed to acclimate for one week prior to experimental manipulation, during which time rats were handled briefly the four days prior to experimentation. Animals were housed individually in Nalgene plexiglass cages (45  25.2  14.7 cm) in a temperature (23 °C) and humidity-controlled, pathogen-free animal facility. Rats were also allowed food (standard rat chow, Harlan Laboratories, Denver, CO) and water ad libitum. The care and treatment of the animals were in accordance with protocols approved by the University of Colorado Institutional Animal Care and Use Committee. 2.2. Stressor protocol On the day of the experiment, animals either remained in their home cages as controls (Control) or were exposed to 100, 1.5 mA, 5-s, intermittent, inescapable tail shocks lasting a total of 1 h 48 min (Stress) as previously described (Campisi and Fleshner, 2003; Kennedy et al., 2005). This stressor protocol was used because of its well-established activation of the stress response and modulation of the immune system (Campisi and Fleshner, 2003; Johnson et al., 2005b; Kennedy et al., 2005). During the stress procedure, rats were placed in a plexiglass tube (23.4  7 cm) and electrodes were placed across the tail protruding from the shock tube. Shocks were administered by an automated shock system (Precision Calculated Animal Shocker, Colbourne Instruments). All animals were shocked between 0700 and 1100. Immediately following stressor termination, animals were sacrificed via rapid decapitation. 2.3. Adrenergic receptor antagonist pretreatment Animals were injected i.p. with sterile saline vehicle, 2.0 mg/kg prazosin (Sigma Aldrich, Milwaukee, WI; cat. #P-7791), 10.0 mg/kg propranolol (Sigma, cat. #P-8688) or 3.0 mg/kg SR59230A (Tocris Biosciences, Bristol, UK; cat. #1511) (n = 8 per group) 30 min prior to stressor exposure. Prazosin is a selective a1-adrenoceptor antagonist demonstrated to be able to cross the blood–brain barrier (Cavero and Roach, 1980; Menkes et al., 1981) and is dissolved in endotoxin-free water plus heat. Propranolol and SR59230A, b1/b2 and b3-adrenoceptor antagonists, respectively, readily cross the blood–brain barrier (Berg et al., 2010; Pardridge et al., 1983) and are dissolved in sterile, endotoxin-free saline. These doses have

S.S. Cox et al. / Brain, Behavior, and Immunity 36 (2014) 183–192

been previously reported to effectively antagonize adrenoceptormediated stress responses (Johnson et al., 2005b; Kong et al., 2010; Thompson et al., 2012). 2.4. Glucocorticoid receptor antagonist pretreatment Rats were injected subcutaneously with either propylene glycol vehicle or 50 mg/kg RU486 (Caymen Chemical, Ann Arbor, MI; cat. #10006317) (n = 8 per group) 1 hr prior to stressor exposure. RU486 is a specific (type II) glucocorticoid receptor antagonist and can cross the blood–brain barrier, but is observed in the brain only at a fraction of the levels measured in serum (Heikinheimo and Kekkonen, 1993). RU486 is readily dissolved in propylene glycol and the dose was chosen based on its previously reported efficacy to antagonize glucocorticoid-mediated stress responses (Fleshner et al., 1996; Nadeau and Rivest, 2003). 2.5. Blood and tissue collection Immediately following stressor termination, animals were rapidly decapitated and trunk blood, spleen, liver, and subcutaneous (SQ) adipose tissue were collected. Trunk blood was collected in 10 ml EDTA tubes and spun at 3000 g for 15 min at 4 °C and divided into 220 ll aliquots to obtain plasma samples. Spleens, left medial lobe of the livers nearest to the portal vasculature, and the right inguinal SQ pad were aseptically dissected, collected in propylene tubes, and snap frozen in liquid nitrogen. All samples were frozen at 80 °C until needed. 2.6. Spleen, liver, and SQ adipose homogenizations Approximately 150 mg of spleens and livers, and 300 mg of SQ adipose samples were added to ice-cold Radioimmunoprecipitation (RIPA) lysis buffer (0.5 M Tris–HCl, pH 7.4, 1.5 M NaCl, 2.5% deoxycholic acid, 10% NP-40, 10 mM EDTA, 1 mM NaF, 1 mM sodium orthovanadate, 1 mM phenylmethylsulfonylfluoride) containing protease and phosphatase inhibitors and 0.01% phosphatase inhibitor cocktail as described previously (Pond, 2000). Tissue samples were homogenized via rapid shaking with ceramic beads (2  50 s at 5000 rpm) using a Precellys 24 high-throughput tissue homogenizer (Bertin Corp, Rockville, MD). Each sample was weighed using a digital scale (Sartorius R200D) and the volume of RIPA lysis buffer was adjusted to a ratio of 3 ml of buffer per gram of tissue in order to correct for total protein concentrations. Following homogenization, sample lysates were divided into aliquots of 120–150 ll and stored at 80 °C for later analysis. 2.7. Measurement of Hsp72 Hsp72 levels were measured in plasma and tissue samples via commercially available sandwich ELISA according to manufacturer’s instructions. Plasma Hsp72 was measured in a high sensitivity ELISA from Enzo Biosciences (prod. #ADI EKS-715) and tissue Hsp72 was measured using a regular sensitivity ELISA (Enzo, prod. #ADI EKS-700B). All homogenates were diluted 1:2 and optical densities were measured using a SpectraMax Plus 354 plate reader (Molecular Devices, Sunnyvale, CA). 2.8. Measurement of inflammatory proteins Concentrations of inflammatory proteins (IL-1b, IL-6, IL-10, and MCP-1) were measured in plasma, spleen, liver, and SQ adipose tissue lysates by ELISA. MCP-1 was measured using ELISAs from Invitrogen (Carlsbad, CA, prod. #KRC1011). All homogenates were analyzed at a 1:2 dilution for MCP-1 and the optical densities were measured using a SpectraMax Plus 354 plate reader. IL-1b, IL-6 and

185

IL-10 were measured using custom multiplex ELISAs from Aushon (Billerica, MA) and individual sandwich ELISAs from R&D Systems (Minneapolis, MN; prod. #870358). Both assays were run according to manufacturer’s instructions, with minor alterations to the wash steps of the multiplex; an extra wash was added to reduce background and ensure accurate readings. Spleen, liver, and SQ adipose tissue homogenates were run using multiplex ELISAs at a 1:3 dilution and plasma was run neat. Optical densities of the multiplex were measured using an Aushon Biosystems Signature PLUS imager. 2.9. Assessment of total protein from tissue samples Total protein was measured from liver, spleen, and SQ adipose tissue homogenates via bicinchonic acid (BCA) assay (Thermo Scientific, Waltham, MA) according to the manufacturer’s instructions. Spleen and liver homogenates were diluted 1:100 and SQ adipose homogenates were diluted 1:10 for protein quantification. Total protein was used to normalize each of the tissue homogenates for comparison of levels of inflammatory proteins and Hsp72. 2.10. Statistical analysis A one-way analysis of variance (ANOVA) was used to compare the stress responsivity of each cytokine, chemokine, and DAMP across plasma, liver, spleen, and SQ tissue. A two-way ANOVA was used to test the effects of adrenergic or glucocorticoid receptor antagonists and stress on cytokine, chemokine, and Hsp72 concentrations within each tissue. Consistent with previously recommended statistical practice (Ott, 1984; Petersen, 1985), post hoc analyses were then performed if an interaction between treatment effects or two significant treatment effects (i.e., effects of both drug treatment and stress) were found. Post hoc analyses were conducted using a Fisher protected least significant difference test (F-PLSD) to determine stress-induced changes in measured cytokines, chemokines, and DAMPs, as well as significant differences between the vehicle + stress group and groups of the different drug treatments + stress. Outliers were removed from the data set according to the Grubb’s test for outliers (Grubbs, 1969). Data are presented as means ± standard error of the mean. Results were significant if p < 0.05. 2.11. Combining vehicle groups In order to combine the two distinct vehicle groups, a one-way ANOVA was performed to examine the effects of the different vehicle injections, sterile saline and propylene glycol, on each protein in each tissue. This was done to determine if the vehicle + no stress groups for each vehicle were statistically different from one another. The analyses revealed that none of the measured levels of IL-b, IL-6, IL-10, MCP-1, or Hsp72 were statistically different between the two vehicle groups in the circulation, SQ adipose, liver, or spleen (p > 0.05). Therefore, the two separate vehicle groups were combined into a single vehicle group (n = 16), to simplify the analyses, reduce the variability and allow for a comparison of the separate adrenergic and glucocorticoid receptor antagonist experiments in the analyses of cytokine, chemokine, and Hsp72 levels. 3. Results 3.1. Comparison of the stress response across plasma and tissues In order to compare the stress-induced changes of inflammatory proteins and Hsp72 between the circulation, spleen, liver,

186

S.S. Cox et al. / Brain, Behavior, and Immunity 36 (2014) 183–192

and SQ adipose tissue, the percent change compared to each tissue’s baseline was calculated. The change from baseline for each inflammatory protein and Hsp72 are illustrated in Fig. 1. There was a robust increase in the percent change of IL-1b (Fig. 1A) and IL-6 (Fig. 1B) in both plasma and SQ adipose tissue. The percent changes of both plasma and SQ adipose were significantly higher than changes observed in the liver (p < 0.0001) and spleen (p < 0.0001). However, the IL-1b response measured in plasma was significantly larger compared to SQ adipose (p < 0.001), while the opposite trend was observed for IL-6; SQ adipose was significantly more responsive than the circulation (p = 0.02). SQ adipose tissue also expressed over a fourfold increase in the percent change of MCP-1 levels (Fig. 1D). This stress-induced change in MCP-1 measured in SQ adipose was significantly higher than plasma, liver, and spleen (p < 0.001). In comparison, the slight changes from

baseline observed in the circulation, liver, and spleen were statistically insignificant from one another. Similar to the other inflammatory proteins, the liver and spleen showed a small percent change in IL-10 levels statistically insignificant from each other (p = 0.9029) (Fig. 1C). However, unlike the other inflammatory proteins, the percent change of IL-10 levels observed in SQ adipose were not statistically significant from that seen in the spleen (p = 0.5534). Plasma again showed a robust response, significantly larger than the other tissues (p < 0.001). As an interesting complement to the percent changes of the inflammatory proteins measured, Hsp72 changes were more ubiquitous across the tissues (Fig. 1E). The percent change observed in the plasma was significantly larger than all three tissues (p < 0.0001). The percent change of splenic Hsp72 was significantly higher than the one observed in the SQ adipose (p = 0.0095) but not in the liver (p = 0.0884). In

Fig. 1. The percent change from non-stressed home cage controls (Control depicted with the dashed line) in (A) IL-1b, (B) IL-6, (C) IL-10, (D) MCP-1, and (E) Hsp72 levels across plasma, liver, spleen, and SQ adipose tissue. Rats either remained in their home cages (Control) or were subjected to inescapable tail shocks (Stress) (n = 16 per group). Animals were sacrificed immediately following termination of stress and plasma, liver, spleen, and SQ were harvested. (a) Plasma levels significantly differed from spleen, liver, and SQ adipose levels (p < 0.05). (b) SQ adipose levels significantly differed from plasma, liver, and spleen levels (p < 0.05).

S.S. Cox et al. / Brain, Behavior, and Immunity 36 (2014) 183–192

addition, the change in the liver and SQ adipose were not reliably different from one another (p = 0.3507). 3.2. The Effect of adrenoceptor and glucocorticoid receptor antagonists on inflammatory protein and Hsp72 levels In order to examine the role of both adrenoceptor and glucocorticoid receptors in elevating inflammatory protein and Hsp72 levels in the circulation, liver, spleen, and SQ adipose tissues, animals were injected with vehicle, prazosin, propranolol, SR59230A, or RU486 prior to tail shock exposure. Pretreatment with any of the three adrenoceptor antagonists, prazosin (p < 0.0001), propranolol (p = 0.0002), or SR59230A (p < 0.0001), significantly reduced the levels of circulating IL-1b, resulting in a main effect of drug treatment (F4,4 = 9.900, p < 0.0001). A main effect of stress (F4,118 = 257.569, p < 0.0001), as well as a significant interaction of treatment  stress (F4,118 = 7.708, p < 0.0001; Fig. 2A) were also observed. SQ adipose IL-1b was also significantly altered due to drug treatment (F4,86 = 3.235, p = 0.01) and stressor exposure (F1,86 = 164.127, p < 0.0001; Fig. 2B). Post hoc analyses revealed that rats exposed to stress and administered RU486 had elevated levels of SQ adipose IL-1b compared to the vehicle + stress group (p = 0.0123). There was also a trend for a reduction in stress-induced IL-1b levels in the propranolol + stress group, although it was not significant (p = 0.07). Analysis of splenic IL-1b revealed main effects of drug treatment (F4,82 = 17.307, p < 0.0001) and stress (F1,82 = 114.436, p < 0.0001), as well as an interaction of treatment  stress (F4,82 = 9.934, p < 0.0001; Fig. 2C). The stress-induced increase of IL-1b levels was blunted by propranolol (p < 0.0001) indicated by the lack of statistical difference between the propranolol + no

187

stress and the propranolol + stress groups (p = 0.9701). In addition, both prazosin (p < 0.0001) and SR59230A (p = 0.01) significantly attenuated the splenic IL-1b stress response. The analysis of liver IL-1b levels revealed main effects of drug treatment (F4,85 = 3.031, p = 0.02) and stress (F1,85 = 89.074, p < 0.0001), as well as a significant interaction (F4,85 = 4.591, p = 0.002; Fig. 2D). Post hoc analyses demonstrated IL-1b levels in the liver were only altered by propranolol. The IL-1b levels in the propranolol + stress group were significantly lowered compared to the vehicle + stress group (p < 0.0001) and no significant difference was seen between the propranolol + no stress and the propranolol + stress groups (p = 0.72), indicating that propranolol blocked the increase in stress-induced IL-1b in the liver. Analysis of circulating IL-6 levels revealed a main effect of stress (F1,85 = 77.148, p < 0.0001; Fig. 3A). IL-6 concentrations in SQ adipose (Fig. 3B) and spleen (Fig. 3C) are both increased after stress. Analyses revealed reliable main effects of stress of IL-6 in SQ adipose tissue (F1,85 = 55.207, p < 0.001) and spleen (F1,83 = 9.323, p = 0.0030; Fig. 3C). There were no main effect of drug and no drug by stress interactions in plasma, SQ or spleen. Compared to the circulation and the other tissues, the levels of IL-6 within the liver were unchanged in the face of stress (data not shown). Fig. 4 depicts the effect of stress and drug treatment on IL-10. Statistical analyses revealed main effects of stress on IL-10 concentrations in plasma (F1,133 = 271.239, p < 0.0001) and SQ adipose (F1,86 = 105.089, p < 0.0001). There were also significant main effects of drug treatment on IL-10 levels in the plasma (F4,133 = 4.678, p = 0.0015; Fig. 4A) and SQ adipose (F4,86 = 3.616, p = 0.009; Fig. 4B), and a significant interaction between drug treatment and stress (F4,118 = 4.754, p = 0.001) in the the plasma. Given the detection of statistically significant main effects of drug and

Fig. 2. Effects of adrenergic and glucocorticoid receptor antagonists on stress-induced increases in concentrations of IL-1b compared to non-stressed home cage control rats (Control) in (A) plasma, (B) SQ adipose, (C) spleen, and (D) liver. Rats were injected i.p. or subcutaneously with vehicle (n = 16), prazosin (2 mg/kg), propranolol (10 mg/kg), SR59230A (3 mg/kg), or RU486 (50 mg/kg) (n = 8 per group) 30 min–1 h prior to tail shock (stress). Animals were sacrificed immediately following termination of stress and plasma and tissues were harvested. IL-1b levels were measured by ELISA. Values represent mean ± standard error of measurement. #Significant difference between control and stress within a drug condition (p < 0.05). ⁄Significant difference between vehicle + stress and drug treatment + stress groups (p < 0.05).

188

S.S. Cox et al. / Brain, Behavior, and Immunity 36 (2014) 183–192

Fig. 4. Effects of adrenergic and glucocorticoid receptor antagonists on stressinduced increases in concentrations of IL-10 compared to non-stressed home cage control rats (Control) in (A) plasma and (B) SQ adipose. Rats were injected i.p. or subcutaneously with vehicle, prazosin (2 mg/kg), propranolol (10 mg/kg), SR59230A (3 mg/kg), or RU486 (50 mg/kg) 30 min–1 h prior to tail shock (stress). Animals were sacrificed immediately following termination of stress and plasma and tissues were harvested. Tissue was homogenized and IL-10 levels were measured by ELISA. Values represent mean ± standard error of measurement. # Significant difference between control and stress within a drug condition (p < 0.05). ⁄Significant difference between vehicle + stress and drug treatment + stress groups (p < 0.05).

Fig. 3. Effects of adrenergic and glucocorticoid receptor antagonists on stressinduced increases in concentrations of IL-6 compared to non-stressed home cage control rats (Control) in (A) plasma, (B) SQ adipose, and (C) spleen. Rats were injected i.p or subcutaneously with vehicle, prazosin (2 mg/kg), propranolol (10 mg/ kg), SR59230A (3 mg/kg), or RU486 (50 mg/kg) 30 min–1 h prior to tail shock (stress). Animals were sacrificed immediately following termination of stress and plasma and tissues were harvested. Tissue was homogenized and IL-6 levels were measured by ELISA. Values represent mean ± standard error of measurement. ### Main effect of stress (p < 0.0001). #Significant difference between control and stress within a drug condition (p < 0.05). ⁄Significant difference between vehicle + stress and drug treatment + stress groups (p < 0.05).

stress in plasma and SQ adipose as well as the significant interaction, additional post hoc analyses were performed. They revealed that prazosin (p = 0.0022) and RU486 (p < 0.0001) augmented the stress-induced increases in circulating IL-10 compared to the vehicle + stress group. An error occurred with the spleen homogenates of the adrenoceptor antagonists during the analysis of IL-10. However, the homogenates of the glucocorticoid receptor antagonists showed a reliable stress response (F1,27 = 34.080, p < 0.0001) and were therefore considered to be accurate based on previous work that demonstrates the stress responsivity of IL-10 in the spleen (Maslanik et al., 2012a). Despite their reliability, the pretreatment

with RU486 had no effect on the stress-induced increase of IL-10 in the spleen (F1,27 = 0.005, p = 0.94; data not shown). A similar issue emerged with the IL-10 ELISA during the analysis of the liver homogenates. These data are not shown. Fig. 5 depicts the effect of stress and drug treatment on MCP-1. Statistical analyses revealed main effects of stress in the plasma (F1,86 = 30.436, p < 0.0001), SQ adipose (F1,85 = 101.769, p < 0.001), and spleen (F1,83 = 12.576, p = 0.006; Fig. 5C). There were also significant main effects of drug treatment on MCP-1 in the circulation (F4,86 = 3.397, p = 0.0125; Fig. 5A) and SQ adipose tissue (F4,85 = 3.965, p = 0.0053; Fig. 5B). Given the detection of statistically significant main effects of drug and stress in plasma and SQ MCP-1, additional post hoc analyses were performed and revealed that prazosin attenuated the stress-induced increase of plasma MCP-1 (p = 0.0223) compared to vehicle treated stressed rats. A significant attenuation of SQ adipose MCP-1 levels in the prazosin + stress group (p = 0.0005), as well as the SR59230A + stress group (p = 0.0161) compared to the vehicle + stress group were also observed. MCP-1 levels within the liver were unchanged after stressor exposure (data not shown). Fig. 6 depicts the effects of stress and drug treatment on Hsp72. Main effects of drug treatment (F4,73 = 3.332, p = 0.0146; Fig. 6A) and stressor exposure (F1,73 = 40.296, p < 0.0001) were observed in the analysis of circulating Hsp72 levels. Further post hoc analyses revealed that both prazosin (p = 0.001) and RU486 (p = 0.02) attenuated stress-induced increases in circulating Hsp72 compared to vehicle treated stressed rats. Importantly, neither prazosin nor RU486 reliably changed non-stressed control

S.S. Cox et al. / Brain, Behavior, and Immunity 36 (2014) 183–192

189

well as an interaction of drug treatment  stress were observed (F4,82 = 6.709, p = 0.0001; Fig. 6C) in the analysis of splenic Hsp72. Further post hoc analyses indicated that the prazosin blocked the stress-induced increase of splenic Hsp72 with the stress group levels unchanged compared to the home cage control animals (p = 0.8). Propranolol also significantly reduced the stress-induced increase of Hsp72 (p = 0.0021). Conversely, there was a significant increase of splenic Hsp72 in the RU486 + stress group compared to the vehicle + stress group (p = 0.0183). Levels of Hsp72 in the liver were also modulated by drug treatment (F4,85 = 4.347, p = 0.003) and increased after stressor exposure (F1,85 = 9.880, p = 0.002; Fig. 6D). A stress-induced increase of Hsp72 was not found in the prazosin + stress (p = 0.7178), or the propranolol + stress (p = 0.9) groups. There was no significant difference between the unstressed and stressed groups given the SR59230A (p = 0.1368). Additionally, a significant increase of Hsp72 was observed in the RU486 + stress group compared to vehicle + stress (p = 0.04).

4. Discussion

Fig. 5. Effects of adrenergic and glucocorticoid receptor antagonists on stressinduced increases in concentrations of MCP-1 compared to non-stressed home cage control rats (Control) in (A) plasma, (B) SQ adipose, and (C) spleen. Rats were injected i.p. or subcutaneously with vehicle, prazosin (2 mg/kg), propranolol (10 mg/kg), SR59230A (3 mg/kg), or RU486 (50 mg/kg) 30 min–1 h prior to tail shock (stress). Animals were sacrificed immediately following termination of stress and plasma and tissues were harvested. Tissue was homogenized and MCP-1 levels were measured by ELISA. Values represent mean ± standard error of measurement. ## Main effect of stress (p < 0.01). #Significant difference between control and stress within a drug condition (p < 0.05). ⁄Significant difference between vehicle + stress and drug treatment + stress groups (p < 0.05).

concentrations compared to vehicle treated non-stressed controls. A significant interaction between drug treatment  stress was observed in the analysis of SQ adipose Hsp72 levels (F4,86 = 8.716, p < 0.0001; Fig. 6B). Post hoc analyses revealed rats injected with prazosin and subjected to stress had attenuated levels of Hsp72 compared to levels observed in vehicle + stress animals (p = 0.02). In contrast, SQ adipose Hsp72 levels in the RU486 + stress group were potentiated compared to the vehicle + stress group (p = 0.0133). Main effects of drug treatment (F4,82 = 10.960, p < 0.0001) and stressor exposure (F1,82 = 55.961, p < 0.0001), as

The impact of stress on the immune system is well established. Stress-evoked changes in immunoregulatory factors in peripheral tissues can assist in clearance and resolution of infections as a consequence of potential wounding (Fleshner, 2013; Dhabhar, 2009). The present study confirmed that exposure to an intense, acute stressor modulates numerous inflammatory proteins and Hsp72 in the circulation, liver, spleen, and SQ adipose. Furthermore, this study illustrated for the first time the distinct differences in the stress responsiveness of different peripheral tissues. Plasma concentrations of IL-1b, IL-6, IL-10, and Hsp72 were shown to be strongly stress responsive as demonstrated previously (Campisi and Fleshner, 2003; Maslanik et al., 2012b, 2013; Nguyen et al., 2000). SQ adipose tissue levels of IL-6 were also highly stress responsive and, interestingly, SQ adipose was the only tissue that expressed a large stress-evoked increase in MCP-1 levels immediately after stressor exposure. Administration of selective a1, b1/2, b3 ADR, as well as (type II) GCR receptor antagonists (prazosin, propranolol, SR59230A, and RU486, respectively) revealed both similarities and differences between the mechanisms of catecholamine and glucocorticoidmediated changes in inflammatory protein and DAMP levels in the periphery due to stress. IL-1b was attenuated, at least in part, by all three ADR antagonists in the plasma, spleen, and SQ adipose (strong trends were present in SQ adipose, though not significant); conversely, levels in the liver were reduced by propranolol alone, suggesting that liver IL-1b is induced by stress via a b1/2 ADR-specific pathway. Additionally, RU486 caused a potentiation of IL-1b levels only in SQ adipose. This observation suggests glucocorticoids may exert stronger regulation of the sterile inflammatory response in SQ adipose compared to other tissues. Levels of IL-6 were not affected by treatment of ADR or GCR antagonists in the circulation, adipose, spleen, or liver. Interestingly, stress-evoked increases in plasma levels of IL-10 were potentiated with the administration of prazosin and RU486, indicating the a1 ADR and glucocorticoid pathway may amplify the pro-inflammatory state of the circulation in response to a stressor by reducing IL-10, a potent anti-inflammatory cytokine. In contrast, treatment with prazosin reduced stress-evoked increases in MCP-1 levels in the circulation and SQ adipose, and SR59230A attenuated stress-evoked MCP-1 only in SQ adipose tissue. Finally, stress-induced levels of Hsp72 were reduced with the administration of prazosin, as previously demonstrated (Johnson et al., 2005a). This effect was observed in all three tissues as well. RU486 had contrasting effects on Hsp72 levels in plasma compared to the liver and SQ adipose. The GCR antagonist caused an increase

190

S.S. Cox et al. / Brain, Behavior, and Immunity 36 (2014) 183–192

Fig. 6. Effects of adrenergic and glucocorticoid receptor antagonists on stress-induced increases in concentrations of Hsp72 compared to non-stressed home cage control rats (Control) in (A) plasma, (B) SQ adipose, (C) spleen, and (D) liver. Rats were injected i.p. or subcutaneously with vehicle, prazosin (2 mg/kg), propranolol (10 mg/kg), SR59230A (3 mg/kg), or RU486 (50 mg/kg) 30 min–1 h prior to tail shock (stress). Animals were sacrificed immediately following termination of stress and plasma and tissues were harvested. Tissue was homogenized and Hsp72 levels were measured by ELISA. Values represent mean ± standard error of measurement. #Significant difference between control and stress within a drug condition (p < 0.05). ⁄Significant difference between vehicle + stress and drug treatment + stress groups (p < 0.05).

in stress-induced Hsp72 levels in the liver and SQ adipose, while attenuating it in the circulation. The RU486-induced decrease of circulating Hsp72 contradicts previous results that indicated plasma levels were unaffected by adrenalectomy in rats (Johnson et al., 2005a). However, due to the selectivity of the treatment of the GCR-selective antagonist RU486 compared to adrenalectomy, these data may more accurately depict the role of glucocorticoids alone on plasma Hsp72 levels. The function of Hsp72 is dependent on its location in relation to the cell. In all likelihood, Hsp72 in the tissues is found both intraand extracellularly and therefore would perform numerous functions within the tissue. Intracellularly, Hsp72 facilitates the folding of naïve proteins, the refolding of denatured proteins, and can act as a chaperone to aid in protein trafficking (Hartl, 1996; Fleshner et al., 2006; Lancaster and Febbraio, 2005). Extracellularly, Hsp72 is capable of inducing pro-inflammatory cytokine production by antigen-presenting cells, cause the maturation of dendritic cells, and can act as a chemokine for leukocytes like neutrophils (Asea, 2007; Pittman and Kubes, 2013). Due to the preparation of tissues being analyzed, we are unable to determine its specific location relative to the cell, and therefore cannot definitively conclude on its function within the tissue. Interestingly, tissue concentrations of MCP-1, a strong chemokine for monocytes and other immune cells, was highly elevated immediately after tail shock stress in SQ adipose tissue compared to other tissues. The SQ adipose-specific upregulation of MCP-1 could indicate immunological priming is initiated with innate immune cells migrating into the tissue, and is regulated to occur in areas that are at high risk for injury or infection. It has been previously demonstrated that leukocytes can migrate from the circulation to the skin after exposure to stress (Dhabhar, 1998; Dhabhar and McEwen, 1997). These studies hypothesized that leukocyte migration to the skin would benefit the organism to prevent or

clear infection immediately following a stressful experience due to its anatomical location and critical function as a physical barrier from the environment. The current study suggests that a similar phenomenon may occur in SQ adipose tissue. Thus the 4-fold increase in MCP-1 concentrations in the SQ adipose depot after tail shock could set the stage for increased migration or retention of leukocytes in this tissue. Due to its location directly below the skin, priming the SQ adipose would also benefit the organism if the cutaneous layer were compromised during an attack from a predator. And although we did not measure macrophages in the SQ adipose, it has previously been demonstrated that a single bout of restraint stress can cause an increase in mononuclear cell content in adipose tissue (Uchida et al., 2012), further supporting the possibility of leukocytes migrating into SQ adipose to initiate or enhance the sterile inflammatory response. The current study is also novel in its elucidation of the role of b3 ADR in stress-induced SQ adipose MCP-1 regulation. Additionally, prazosin an a1 ADR antagonist, also modulated SQ adipose MCP1 levels and previous studies have reported MCP-1 can be secreted by endothelial cells (Rollins et al., 1990), a tissue with a high proportion of a1 ADR. Further, while b3 ADR have been identified on other tissues like the coronary system and peripheral vasculature (Dessy et al., 2005; Michel et al., 2011; Rath et al., 2012), they are densely expressed on adipocytes, where they may alter MCP-1 levels in SQ adipose during stress. This is the first study to implicate adipocytes specifically in the recruitment of leukocytes via MCP-1 during an acute stressor. Therefore, both the vascular endothelium, as well adipocytes, may participate in enhancing the sterile inflammatory response within SQ adipose tissue. To conclude, the current study illustrates the differences in stress responsivity across tissues; the circulation and SQ adipose tissue were highly stress responsive compared to liver and spleen.

S.S. Cox et al. / Brain, Behavior, and Immunity 36 (2014) 183–192

In addition, the neuroendocrine mechanisms responsible for stress-induced increases in IL-b, IL-6, IL-10, MCP-1, and Hsp72 were analyzed in plasma and across several peripheral tissues to elucidate similarities and differences underlying their changes due to catecholamines and glucocorticoids. MCP-1, a potent leukocyte chemokine, was selectively increased 4-fold in SQ adipose after stressor exposure, suggesting a mechanism for increased leukocyte migration into this tissue and possible immunological priming. One limitation of the current study is that the drugs tested have some ability to cross the blood–brain barrier (Berg et al., 2010; Cavero and Roach, 1980; Heikinheimo and Kekkonen, 1993; Pardridge et al., 1983), thus the effects observed could be mediated by central or peripheral catecholamines or glucocorticoids. Although more research is needed to explicate the cellular source of these inflammatory proteins and Hsp72 within the tissues, these results add to our understanding of the complexity and tissuespecificity of the sterile inflammatory response and its modulation during an acute stressor. 5. Conflicts of Interest The authors declare no conflicts of interest. The authors alone are responsible for the content and writing of this manuscript. Acknowledgments This research was supported by Grants awarded to MF by the National Science Foundation (#IOS 1022451). The authors would also like to thank Tyler Woodworth for assisting in homogenizations. References Ando, T., Rivier, J., Yanaihara, H., Arimura, A., 1998. Peripheral corticotropinreleasing factor mediates the elevation of plasma IL-6 by immobilization stress in rats. Am. J. Physiol. 275, R1461–R1467. Asea, A., 2007. Mechanisms of HSP72 release. J. Biosci. 32, 579–584. Berg, T., Piercey, B.W., Jensen, J., 2010. Role of beta1-3-adrenoceptors in blood pressure control at rest and during tyramine-induced norepinephrine release in spontaneously hypertensive rats. Hypertension 55, 1224–1230. Brattsand, R., Linden, M., 1996. Cytokine modulation by glucocorticoids: mechanisms and actions in cellular studies. Aliment. Pharmacol. Ther., 81–92. Campisi, J., Fleshner, M., 2003. Role of extracellular HSP72 in acute stress-induced potentiation of innate immunity in active rats. J. Appl. Physiol. 94, 43–52. Cavero, I., Roach, A.G., 1980. The pharmacology of prazosin, a novel antihypertensive agent. Life Sci. 27, 1525–1540. Coppack, S.W., 2001. Pro-inflammatory cytokines and adipose tissue. Proc. Nutr. Soc. 60, 349–356. Dessy, C., Saliez, J., Ghisdal, P., Daneau, G., Lobysheva, I.I., Frérart, F., Belge, C., Jnaoui, K., Noirhomme, P., Feron, O., Balligand, J.L., 2005. Endothelial beta3adrenoreceptors mediate nitric oxide-dependent vasorelaxation of coronary microvessels in response to the third-generation beta-blocker nebivolol. Circulation 112, 1198–1205. Dhabhar, F.S., 1998. Stress-induced enhancement of cell-mediated immunity. Ann. N.Y. Acad. Sci. 840, 359–372. Dhabhar, F.S., 2009. Enhancing versus suppressive effects of stress on immune function: implications for immunoprotection and immunopathology. NeuroImmunoModulation 16, 300–317. Dhabhar, F.S., McEwen, B.S., 1997. Acute stress enhances while chronic stress suppresses cell-mediated immunity in vivo: a potential role for leukocyte trafficking. Brain Behav. Immun. 11, 286–306. Elliott, J., 1997. Alpha-adrenoceptors in equine digital veins: evidence for the presence of both alpha1 and alpha2-receptors mediating vasoconstriction. J. Vet. Pharmacol. Ther. 20, 308–317. Engler, A., Roy, S., Sen, C.K., Padgett, D.A., Sheridan, J.F., 2005. Restraint stress alters lung gene expression in an experimental influenza A viral infection. J. Neuroimmunol. 162, 103–111. Fleshner, M., 2013. Stress-evoked sterile inflammation, danger associated molecular patterns (DAMPs), microbial associated molecular patterns (MAMPs) and the inflammasome. Brain Behav. Immun. 27, 1–7. Fleshner, M., Brennan, F.X., Nguyen, K., Watkins, L.R., Maier, S.F., 1996. RU-486 blocks differentially suppressive effect of stress on in vivo anti-KLH immunoglobulin response. Am. J. Physiol. 271, R1344–R1352. Fleshner, M., Campisi, J., Deak, T., Greenwood, B.N., Kintzel, J.A., Leem, T.H., Smith, T.P., Sorensen, B., 2002. Acute stressor exposure facilitates innate immunity

191

more in physically active than in sedentary rats. Am. J. Physiol. Regul. Integr. Comp. Physiol. 282, R1680–R1686. Fleshner, M., Sharkey, C.M., Nickerson, M., Johnson, J.D., 2006. Endogenous extracellular Hsp72 release is an adaptive feature of the acute stress response. Psychoneuroimmunology. Elsevier/Academic Press, San Diego, pp. 1013–1034. Fleshner, M., Johnson, J.D., Friedman, J., 2007. Extracellular Hsp72: a double-edged sword for host defense. Heat Shock Proteins: Potent Mediators of Inflammation and Immunity. Springer, Dordrecht, The Netherlands, pp. 235–263. Griffiths, R.J., Stam, E.J., Downs, J.T., Otterness, I.G., 1995. ATP induces the release of IL-1 from LPS-primed cells in vivo. J. Immunol. 154, 2821–2828. Grubbs, F.E., 1969. Procedures for detecting outlying observations in samples. Technometrics 11, 1–21. Hartl, F.U., 1996. Molecular chaperones in cellular protein folding. Nature 381, 571– 579. Haskó, G., Elenkov, I.J., Kvetan, V., Vizi, E.S., 1995. Differential effect of selective block of alpha 2-adrenoreceptors on plasma levels of tumour necrosis factoralpha, interleukin-6 and corticosterone induced by bacterial lipopolysaccharide in mice. J. Endocrinol. 144, 457–462. Heikinheimo, O., Kekkonen, R., 1993. Dose–response relationships of RU 486. Ann. Med. 25, 71–76. Johnson, J.D., Campisi, J., Sharkey, C.M., Kennedy, S.L., Nickerson, M., Fleshner, M., 2005a. Adrenergic receptors mediate stress-induced elevations in extracellular Hsp72. J. Appl. Physiol. (Bethesda, Md.: 1985) 99, 1789–1795. Johnson, J.D., Campisi, J., Sharkey, C.M., Kennedy, S.L., Nickerson, M., Greenwood, B.N., Fleshner, M., 2005b. Catecholamines mediate stress-induced increases in peripheral and central inflammatory cytokines. Neuroscience 135, 1295–1307. Kennedy, S.L., Nickerson, M., Campisi, J., Johnson, J.D., Smith, T.P., Sharkey, C., Fleshner, M., 2005. Splenic norepinephrine depletion following acute stress suppresses in vivo antibody response. J. Neuroimmunol. 165, 150–160. Kong, Y., Zhang, Y., Li, N., Zhang, L., Gao, Y., Xue, H., Li, Y., Li, W.M., 2010. Association between beta3-adrenergic receptor and oxidative stress in chronic heart failure rats. Zhonghua Xinxueguanbing Zazhi 38, 435–439. Krief, S., Lönnqvist, F., Raimbault, S., Baude, B., Van Spronsen, A., Arner, P., Strosberg, A.D., Ricquier, D., Emorine, L.J., 1993. Tissue distribution of beta 3-adrenergic receptor mRNA in man. J. Clin. Invest. 91, 344–349. Kubes, P., Mehal, W.Z., 2012. Sterile inflammation in the liver. Gastroenterology 143, 1158–1172. Kunicka, J.E., Talle, M.A., Denhardt, G.H., Brown, M., Prince, L.A., Goldstein, G., 1993. Immunosuppression by glucocorticoids: inhibition of production of multiple lymphokines by in vivo administration of dexamethasone. Cell. Immunol. 149, 39–49. Lancaster, G.I., Febbraio, M.A., 2005. Mechanisms of stress-induced cellular HSP72 release: implications for exercise-induced increases in extracellular HSP72. Exerc. Immunol. Rev. 11, 46–52. Maslanik, T., Bernstein-Hanley, I., Helwig, B., Fleshner, M., 2012a. The impact of acute-stressor exposure on splenic innate immunity: a gene expression analysis. Brain Behav. Immun. 26, 142–149. Maslanik, T., Tannura, K., Mahaffey, L., Loughridge, A.B., Benninson, L., Ursell, L., Greenwood, B.N., Fleshner, M., 2012b. Commensal bacteria and MAMPs are necessary for stress-induced increases in IL-1b and IL-18 but not IL-6, IL-10 or MCP-1. PLoS ONE 7, e50636. Maslanik, T., Mahaffey, L., Tannura, K., Beninson, L., Greenwood, B.N., Fleshner, M., 2013. The inflammasome and danger associated molecular patterns (DAMPs) are implicated in cytokine and chemokine responses following stressor exposure. Brain Behav. Immun. 28, 54–62. Menkes, D.B., Baraban, J.M., Aghajanian, G.K., 1981. Prazosin selectively antagonizes neuronal responses mediated by alpha1-adrenoceptors in brain. Naunyn Schmiedebergs. Arch. Pharmacol. 317, 273–275. Michel, M.C., Harding, S.E., Bond, R.A., 2011. Are there functional b3-adrenoceptors in the human heart? Br. J. Pharmacol. 162, 817–822. Mohamed-Ali, V., Pinkney, J.H., Coppack, S.W., 1998. Adipose tissue as an endocrine and paracrine organ. Int. J. Obes. Relat. Metab. Disord. 22, 1145–1158. Mottillo, E.P., Shen, X.J., Granneman, J.G., 2010. Beta3-adrenergic receptor induction of adipocyte inflammation requires lipolytic activation of stress kinases p38 and JNK. Biochim. Biophys. Acta 1801, 1048–1055. Nadeau, S., Rivest, S., 2003. Glucocorticoids play a fundamental role in protecting the brain during innate immune response. J. Neurosci. 23, 5536–5544. Netea, M.G., Simon, A., Van de Veerdonk, F., Kullberg, B.J., Van der Meer, J.W.M., Joosten, L.A.B., 2010. IL-1beta processing in host defense: beyond the inflammasomes. PLoS Pathog. 6, e1000661. Nguyen, K.T., Deak, T., Will, M.J., Hansen, M.K., Hunsaker, B.N., Fleshner, M., Watkins, L.R., Maier, S.F., 2000. Timecourse and corticosterone sensitivity of the brain, pituitary, and serum interleukin-1beta protein response to acute stress. Brain Res. 859, 193–201. Nickerson, M., Kennedy, S.L., Johnson, J.D., Fleshner, M., 2006. Sexual dimorphism of the intracellular heat shock protein 72 response. J. Appl. Physiol. (Bethesda, Md.: 1985) 101, 566–575. O’Connor, K.A., Johnson, J.D., Hansen, M.K., Wieseler Frank, J.L., Maksimova, E., Watkins, L.R., Maier, S.F., 2003. Peripheral and central proinflammatory cytokine response to a severe acute stressor. Brain Res. 991, 123–132. Ott, L., 1984. An Introduction to Statistical Methods and Data Analysis. Duxbury Press, Boston (pp. 365–370). Pardridge, W.M., Sakiyama, R., Fierer, G., 1983. Transport of propranolol and lidocaine through the rat blood–brain barrier: primary role of globulin-bound drug. J. Clin. Invest. 71, 900–908.

192

S.S. Cox et al. / Brain, Behavior, and Immunity 36 (2014) 183–192

Pastores, S.M., Hasko, G., Vizi, E.S., Kvetan, V., 1996. Cytokine production and its manipulation by vasoactive drugs. New Horiz. 4, 252–264. Persoons, J.H., Schornagel, K., Brevé, J., Berkenbosch, F., Kraal, G., 1995. Acute stress affects cytokines and nitric oxide production by alveolar macrophages differently. Am. J. Respir. Crit. Care Med. 152, 619–624. Pertsov, S.S., Koplik, E.V., Stepanyuk, V.L., Simbirtsev, A.S., 2009. Blood cytokines in rats with various behavioral characteristics during emotional stress and treatment with interleukin-1beta. Bull. Exp. Biol. Med. 148, 196–199. Petersen, R.G., 1985. Design and Analysis of Experiments. Marcel Dekker, Inc., New York and Basel (pp. 77–84). Pittman, K., Kubes, P., 2013. Damage-associated molecular patterns control neutrophil recruitment. J Innate Immun. 5, 315–323. Pond, C.M., 2000. Adipose tissue: quartermaster to the lymph node garrisons. Biologist 47, 147–150 (London, UK). Rang, H.P., 2003. Pharmacology. Churchill Livingstone, Edinburgh (pp. 163–270). Rath, G., Balligand, J.L., Dessy, C., Chantal, D., 2012. Vasodilatory mechanisms of beta receptor blockade. Curr. Hypertens. Rep. 14, 310–317. Rollins, B.J., Yoshimura, T., Leonard, E.J., Pober, J.S., 1990. Cytokine-activated human endothelial cells synthesize and secrete a monocyte chemoattractant, MCP-1/JE. Am. J. Pathol. 136, 1229–1233. Speaker, K.J., Cox, S.S., Paton, M.M., Serebrakian, A., Maslanik, T., Greenwood, B.N., Fleshner, M., 2013. Habitual exercise exerts a depot specific, anti-inflammatory effect on stress-evoked cytokines in the white adipose tissue of lean rats. Brain Behav. Immun., 2013. http://dx.doi.org/10.1016/j.bbi.2013.10.028. Steer, J.H., Kroeger, K.M., Abraham, L.J., Joyce, D.A., 2000. Glucocorticoids suppress tumor necrosis factor-alpha expression by human monocytic THP-1 cells by

suppressing transactivation through adjacent NF-kappa B and c-Jun-activating transcription factor-2 binding sites in the promoter. J. Biol. Chem. 275, 18432– 18440. Tan, K.S., Nackley, A.G., Satterfield, K., Maixner, W., Diatchenko, L., Flood, P.M., 2007. Beta2 adrenergic receptor activation stimulates pro-inflammatory cytokine production in macrophages via PKA- and NF-kappaB-independent mechanisms. Cell. Signal. 19, 251–260. Thompson, R.S., Strong, P.V., Fleshner, M., 2012. Physiological consequences of repeated exposures to conditioned fear. Behav. Sci. 2, 57–78. Uchida, Y., Takeshita, K., Yamamoto, K., Kikuchi, R., Nakayama, T., Nomura, M., Cheng, X.W., Egashira, K., Matsushita, T., Nakamura, H., Murohara, T., 2012. Stress augments insulin resistance and prothrombotic state: role of visceral adipose-derived monocyte chemoattractant protein-1. Diabetes 61, 1552– 1561. You, Z., Luo, C., Zhang, W., Chen, Y., He, J., Zhao, Q., Zuo, R., Wu, Y., 2011. Pro- and anti-inflammatory cytokines expression in rat’s brain and spleen exposed to chronic mild stress: involvement in depression. Behav. Brain Res. 225, 135– 141. Zhou, D., Kusnecov, A.W., Shurin, M.R., DePaoli, M., Rabin, B.S., 1993. Exposure to physical and psychological stressors elevates plasma interleukin 6: relationship to the activation of hypothalamic–pituitary–adrenal axis. Endocrinology 133, 2523–2530. Zuckerman, S.H., Shellhaas, J., Butler, L.D., 1989. Differential regulation of lipopolysaccharide-induced interleukin 1 and tumor necrosis factor synthesis: effects of endogenous and exogenous glucocorticoids and the role of the pituitary–adrenal axis. Eur. J. Immunol. 19, 301–305.