Regular, Moderate Exercise is Associated with Improved Overall Health
Regular moderate physical activity positively influences many aspects of health. For example, a physically active lifestyle is associated with decreased risks of coronary heart disease and high blood pressure. A physically active lifestyle may also decrease the risk of bacterial or viral illness. We have proposed that the reported reduction in infectious disease associated with habitual physical activity is caused by an indirect health benefit of exercise, that is, stress reduction. Exposure to physical and/or psychological stressors modulates the immune response. Stressor exposure is neither globally immunosuppressive nor immunopotentiating, but rather immunomodulatory. Factors that modulate the impact that the stress response has on the immune function include the following: the duration of stressor exposure; the perceived controllability of the stressor; the measure of the immune response; and the physiological state of the organism (e.g., young vs old, anxious vs calm, healthy vs ill, and physically active vs sedentary). Voluntary physical activity may prevent stress-induced suppression of the immune system, thereby reducing the increased susceptibility and severity of infectious disease caused by stress.
Exercise, Stress, Disease, and Immune Function
There is early evidence to support the stress-modulatory effect of regular moderate exercise. Brown and Siegal (1) assessed teenage girls (364 subjects) for their levels of self-reported exercise schedules, stress levels, and disease incidence. Although there are limitations to the conclusions that can be drawn from this study because of the failings of self-report and disease verification, the primary findings of this study are clear. As expected, girls who were sedentary and under high levels of stress had elevated disease incidence. In contrast, girls who were moderately physically active and under high stress were protected against the stress-induced increases in disease incidence. Thus, the hypothesis that physical activity may improve health by preventing the immunologically deleterious consequences of stress has support in the human literature.
Similar support can be found in the animal literature using immunological measures as an endpoint. Moraska and Fleshner (13) have recently reported a stress-buffering effect of freewheel running on stress-induced suppression in antibody or immunoglobulin generated against keyhole limpet hemocyanin (αKLH Ig). In this study (depicted in part in Fig. 1), adult male rats (10 per group) were allowed to live with running wheels or remain sedentary in their home cages. After 4 wk of voluntary running, rats were immunized with keyhole limpet hemocyanin (KLH) immediately before exposure to a single session of inescapable tail shock stress (100, 1.6 mA, 5-s uncontrollable and unpredictable tail shock). The concentration of antibody generated against KLH was measured in blood using enzyme-linked immunosorbent assay. Assessment of αKLH Ig is an excellent in vivo measure of antigen-specific or acquired immunity because the immune cells remain in the organism, the response requires interactions between several cells of the immune system, the Ig response generated is specific to the antigen used for challenge, and αKLH Ig can be quantitatively measured repeatedly across time. In addition, reductions in an organism’s ability to generate αKLH Ig response may be clinically meaningful because it has been reported in humans, as well as in animals, to decline with age and psychological stress. Using this measure we reported that exposure to tail shock stress reduced the αKLH Ig response in sedentary, but not in physically active, rats. Freewheel running alone had no effect on the antibody response.
The data presented in Figure 1 suggest that regular, moderate, physical activity can prevent the negative consequences of stress on immune function. The potential immunophysiological mechanism(s) for the stress-buffering effect of exercise are currently undergoing investigation and are the topic of this review. The approach my laboratory has taken to investigate these mechanisms is to conduct research using an accepted model of regular, moderate, habitual exercise (voluntary freewheel running) and a well-characterized model of stress-induced immunosuppression (tail shock–induced suppression of αKLH Ig). Before presenting the proposed immunophysiological mechanisms for the stress-buffering effect of exercise, I first briefly describe what is known about the cellular interactions necessary to generate antibody. Second, the potential cellular and neuroendocrine mechanisms for stress-induced suppression in the αKLH Ig response in sedentary animals are elucidated. And finally, our current hypothesis concerning the potential physiological adaptations in the physical active animal that may contribute to the stress-buffering effect of exercise on stress-induced suppression of αKLH Ig are described. Because of strict citation restrictions, this review focuses on work from my laboratory using a single model of exercise (voluntary freewheel running) and a single model of stress-induced immunosuppression (tail shock–induced suppression of αKLH Ig).
CELLULAR INTERACTIONS NECESSARY TO GENERATE αKLH IMMUNOGLOBULIN
As depicted in Figures 2 and 3, the generation of an antibody response to a T-cell-dependent soluble protein, such as KLH, involves the interaction of antigen-presenting cells (APC) (B cells and/or dendritic cells), T-helper cells, and B cells (9). Within hours after intraperitoneal injection of KLH, antigen is transported to the draining lymph nodes and spleen. APC, likely B cells in this case, appropriately present antigen to T-helper cells. Presentation of antigen by B cells that are functioning as APCs involves the binding of KLH to the B cell receptor (BCR) and the upregulation of costimulator molecules on the B cell surface (i.e., B7 or CD80/86). KLH is then presented in association with self-molecules (major histocompatibility complex class II) to a Th cell that has a receptor that recognizes KLH plus major histocompatibility complex class II. If adequate costimulation is provided via CD28 (on the Th cell) binding to B7 on the APC, then the αKLH Th cell proliferates and differentiates into a functional helper cell. As shown in Figure 3, B cells that express B cell receptor that can bind KLH must receive T cell help from the KLH-specific T-helper cells in the form of cytokines, such as IL2, INFγ, IL4, IL5, or IL6, and costimulation. The Th “help” facilitates B cell proliferation, B cell differentiation into antibody secreting cells, and Ig isotype switching (IgM to IgG). The proliferation of KLH-specific Th and B cells is greatest in the draining lymph nodes and spleen 4–7 d after KLH (4–6). Although the “APC” function and the “antibody secreting” function are depicted schematically by two separate cells in Figures 2 and 3, it is possible that the same B cell performs both tasks at different stages of the Ig response.
MECHANISMS OF STRESS-INDUCED SUPPRESSION OF ANTI-KLH Ig: CELLULAR
Rats that are immunized with KLH and exposed to a single session of inescapable tail shock have a long-term reduction in serum levels of αKLH IgM and αKLH IgG (3,6). Importantly the suppressive effect is specific to the generation of antibody to the antigen, because total serum IgM and IgG is not reduced by stress. Using flow cytometric analysis and antigen-specific proliferative assays, we have determined that the suppression in αKLH IgM and IgG is caused by a failure of the stressed rats to increase KLH-specific T-helper cell numbers (4–6). With fewer αKLH T-helper cells, there is less T cell help, and fewer KLH-specific B cells in the spleen. Fewer KLH-specific B cells leads to a reduction in serum αKLH Ig (Fig. 4).
Evidence suggests that the final site of stress-induced immunomodulation is the spleen, because if we remove the spleen from adult male rats (N = 10; splenectomy; Fig. 5) before immunization with KLH and stressor exposure, in addition to reducing the total amount of αKLH Ig 30-fold, we eliminate the stress-induced reduction in αKLH IgG found in sham-operated rats exposed to tail shock stress (Fig. 5). The tail shock–induced suppression in αKLH IgG is isotype specific. Tail shock exposure results in reduced αKLH IgG2a but not αKLH IgG1 (4–6,13). We hypothesize that the isotype specificity is caused by a selective suppression in αKLH Th1-like, interferon-γ producing T cells (4–6). A reduction in the Th1 cell subset leads to less αKLH IgG2a because interferon-γ induces B cells to make IgG2a, whereas IL4 induces B cells make IgG1. In addition, a reduction in Thl-like IL2 secreting T cells could also lead to a reduction in Th2 cells. Th2 cells are classically associated with making cytokines necessary for B cell growth and differentiation (e.g., IL4 and IL5). Finally, we have recently narrowed the window of time (0–5 h after KLH) when exposure to an intense acute stressor (i.e., tail shock) will result in suppression in αKLH Ig. Based on the recent work by Kennedy et al., evidence suggests that exposure to stress is disrupting the very earliest stages in the generation of Ig against KLH, i.e., antigen presentation and costimulation by antigen specific B cells. As a consequence of suboptimal antigen presentation and costimulation, fewer KLH-specific Th cells are generated, resulting in fewer antibody secreting KLH B cells and less antibody released into the blood.
MECHANISIM(S) OF STRESS-INDUCED SUPPRESSION OF ANTI-KLH Ig: THE ROLE OF THE SYMPATHETIC NERVOUS SYSTEM
Although the signal(s) responsible for stress-induced suppression of αKLH Ig remain unclear, changes in the sympathetic nervous system (SNS) are likely playing a major role. Stimulation of the SNS is a hallmark of the acute stress response. SNS activation has many physiological consequences that work in concert to promote the “fight/flight” response. In addition, most primary and secondary lymphoid tissues (including the spleen) receive dense SNS innervation and Th1 cells, but not Th2 cells, B cells, and monocytes/macrophages/dendritic cells express adrenergic receptors (β2ADR). If we focus on the role of the SNS in stress-induced immunomodulation, there is evidence that the SNS contributes to specifically stress-induced suppression of the αKLH Ig response (14). Central nervous system stimulation of the sympathetic nervous system, for example, suppresses the antibody response to KLH. Although earlier work suggested that high concentrations of norepinephrine can suppress various aspects of immunity, more recent data support the hypothesis that splenic norepinephrine depletion, not circulating or splenic norepinephrine elevation, is more likely responsible for stress-induced suppression of in vivo αKLH Ig responses.
There are several lines of evidence to support this shift in dogma from “too much norepinephrine to too little norepinephrine.” First, if one examines the earlier literature demonstrating that high levels of norepinephrine are immunosuppressive, most studies were performed in vitro, examined mitogen-stimulated proliferative or cytokine responses, and tested pharmacological concentrations of norepinephrine. Under these circumstances, clearly norepinephrine can suppress immune function and in fact can be fatal to cells of the immune system. Second, rarely was activation status of the immune cells considered in these earlier studies. For example, β2AR are differentially expressed on naïve versus stimulated B cells (14). Thus, previous research supporting a simple view that too much norepinephrine is responsible for stress-induced suppression of in vivo immune responses has limitations.
Recent evidence is more consistent with the dogmatic shift that too little norepinephrine may be responsible for stress-induced suppression of in vivo antibody responses and that dynamic interactions between SNS and immune cells occur to produce optimal Ig responses. For example, during the generation of an in vivo antibody response to KLH, norepinephrine is released from peripheral nerves innervating the spleen (14). As shown in Figure 6, norepinephrine binding to the B cell β2ADR stimulates the expression of costimulatory molecules (B7.2 or CD86), Ig production, and splenic germinal center formation. And norepinephrine activation of the Th1 β2ADR stimulates interferon-γ cytokine production and Th1 differentiation. Importantly, depletion of splenic norepinephrine content by pharmacological lesion (6-OHDA) before in vivo KLH immunization reduces αKLH Ig.
In addition to these studies, recent work from my laboratory suggests that stress-induced suppression of αKLH Ig requires splenic norepinephrine depletion and not circulating norepinephrine elevation, also lending support to the idea that “too little and not too much norepinephrine” may be important. Exposure to tail shock stress increases circulating norepinephrine (via tissue spillover) and drives the SNS to the extent that splenic norepinephrine concentration is depleted or reduced below control levels for several hours after stressor termination (8). In a recent study by Kennedy et al., rats that were immunized 4 h after stressor termination when splenic norepinephrine concentrations have returned to normal, no longer have suppression in the αKLH Ig. In contrast, if rats are immunized with KLH 2 h after stressor termination when blood concentrations of norepinephrine are normal but splenic norepinephrine remains depleted, stress-induced suppression of αKLH Ig persists. In addition, rats treated with tyrosine (norepinephrine precursor molecule) before tail shock are protected from stress-induced splenic norepinephrine depletion and αKLH Ig suppression. Importantly, blood concentrations of norepinephrine in the tyrosine-treated stressed rats were equal to saline injected stressed rats, yet tyrosine completely prevented the suppression in αKLH Ig. Finally, central activation of the SNS in the absence of stressor exposure with mirtazapine, a α2ADR antagonist that acts in the brain to release the SNS from α2ADR-mediated inhibition, increases blood concentrations of norepinephrine higher than that produced by tail shock and for an equal or longer duration. Mirtazapine does not produce splenic norepinephrine depletion and does not suppress αKLH Ig. Mirtazapine likely increases norepinephrine in the blood but does not produce splenic norepinephrine depletion because it globally activates peripheral SNS output across many tissues leading to greater tissue norepinephrine spillover and hence higher norepinephrine concentration in the blood. In contrast, exposure to tail shock may produce a more selective activation of SNS, such that specific peripheral tissues are driven to norepinephrine depletion with less total spillover and less norepinephrine increases in the blood. There is precedent in the literature for a selective activation of SNS to specific peripheral tissues in response to a variety of stressors (10). These data support the hypothesis that stress-induced suppression of αKLH is caused by too little, and not too much, splenic norepinephrine.
Central Nervous Systems Pathways of Splenic Autonomic Output after Stress
The central control of the autonomic nervous system involves highly complex reciprocal interactions between areas in the cerebral cortex, limbic system, basal ganglia, hypothalamus, thalamus, and medulla. Specifically, if we focus on the innervation pathway of the spleen, a more detailed description of the central sympathetic circuits responsible for splenic innervation can be described using retrograde pseudorabies virus staining and c-Fos immunohistochemistry. Pseudorabies virus can be used to track neural pathways because it is retrogradely transported from nerve terminals to cell bodies. This virus has the advantage over other retrograde tracer in that it will cross synapses, thus allowing the tracing of multisynaptic pathways (2). C-Fos is produced after activation of the immediate early gene cfos. Increased expression of c-Fos protein is indicative of neural activation or increased neural metabolism. Using these techniques, Cano et al. (2) have elegantly described several specific areas in the brain that are active during stress (c-Fos+) and that innervate the spleen (pseudorabies virus+).
Depicted in Figure 7 are the regions in the brain that are consistently among the first to contain virus after splenic peripheral injection (2). They include the medial parvicellular subdivisions of the paraventricular hypothalamic nucleus (medial parvicellular subdivisions of the paraventricular hypothalamic nucleus autonomic subnuclei), A5 cell group, ventromedial medulla, rostral ventrolateral medulla, and caudal raphe magnus. Additionally Barrington’s nucleus and the locus ceruleus, regions not traditionally thought to be directly involved in modulation of the peripheral sympathetic nervous system, have been shown to contain pseudorabies virus rapidly after splenic injection (2). We refer to these brain nuclei with close synaptic connections to peripheral sympathetically innervated targets as the “central sympathetic circuit” (Fig. 7). Importantly, modulation of nuclei within this circuit is known to affect splenic nerve activity and splenic immune function, supporting a functional and regulatory role for the central sympathetic circuit in splenic sympathetic and immune modulation. Areas depicted in Figure 7 but not discussed include dorsal parvicellular cap of the paraventricular nucleus, ventral pontine reticular nucleus, alpha region of the gigantocellular reticular nucleus, perifacial zone, raphe pallidus, and lateral paragigantocellular nucleus.
WORKING HYPOTHESIS OF STRESS-INDUCED SUPPRESSION OF αKLH Ig IN SEDENTARY ORGANISMS
Exposure to stress in a sedentary animal results in the activation of the central sympathetic circuit (Fig. 7), which leads to activation of the splenic sympathetic neurons and the release of norepinephrine from splenic sympathetic nerve terminals. With continued stressor exposure (10–50, 5-s, 1.6-mA tail shocks), splenic norepinephrine nerve terminals, and hence splenic norepinephrine content become depleted (8). A depletion of splenic norepinephrine content reduces the stimulation of the B cell β2ADR and consequently results in a reduction in expression of the costimulatory molecule (B7.2/CD86). Suboptimal B7 expression on the antigen presenting B cell reduces αKLH Th1 proliferation, differentiation, and cytokine production. This, in turn, results in fewer αKLH Th1 cells, less T cell help for αKLH B cells, and less αKLH IgM, IgG, and IgG2a, but not IgG1 (4–6).
THE EFFECT OF STRESS IN PHYSICALLY ACTIVE ORGANISMS: SYMPATHETIC NERVOUS SYSTEM ADAPTATIONS
Can physical activity prevent the negative effect of stress on αKLH Ig by reducing stress-induced sympathetic output and preventing splenic norepinephrine depletion? Historically, it is clear that endurance training reduces plasma catecholamine levels in response to submaximal exercise and differentially impacts tissue sympathetic content. The majority of the animal literature on this topic is not directly relevant to our model because rats were trained using forced treadmill running and tested after an additional treadmill exercise challenge. This is problematic because treadmill running produces immunological and physiological changes indicative of chronic stress (12). Recent data from our laboratory demonstrate that freewheel running can blunt sympathetic nervous system output and prevent tissue norepinephrine depletion after exposure to tail shock stress (8). We have previously reported that 6 wk of voluntary wheel running prevents stress-induced depletion of splenic norepinephrine. Although the precise mechanism for this effect remains a current topic of investigation in our laboratory, we have evidence that freewheel running produces adaptations in peripheral sympathetic nerve norepinephrine synthesis/release and central sympathetic circuit activation that function to prevent/delay tissue norepinephrine depletion. Interestingly, the adaptations produced by physical activity are tissue-specific. For example, 4–6 wk of voluntary wheel running prevents stress-induced norepinephrine depletion in the liver and spleen (8), but not the adrenals (8). Wheel running produces an increase rate of synthesis/release (k) of norepinephrine in the liver (P = 0.004) and not in the spleen or adrenals. Thus, in case of liver, physical activity may prevent norepinephrine depletion via peripheral adaptations in synthesis, whereas for spleen a different adaptation, possibly in the central sympathetic circuit, may be responsible. In contrast, freewheel running produces neither peripheral nor central adaptations in the adrenals and hence does not prevent stress-induced norepinephrine depletion. It is clear, therefore, that exercise does impact stress-induced sympathetic nervous system output. The effect of exercise is dependent on the tissue examined and may be caused by adaptations in both peripheral sympathetic nerve synthesis/release rates and central sympathetic circuit activation.
THE EFFECT OF STRESS IN PHYSICALLY ACTIVE ORGANISMS: CENTRAL SYMPATHETIC CIRCUIT ADAPTATIONS
The majority of the literature examining the impact of physical activity on the central nervous system deals with central regulation of peripheral physiological systems responding directly to the demands of exercise. In addition, the majority of these studies, once again, use forced treadmill exercise regimens. There is some work investigating the changes in brain activity associated with voluntary freewheel running; however, there are few studies that test the effect of voluntary freewheel running on the brain’s response to a stressor or environmental challenge. Lambert and Jonsdottier (11) demonstrated that 5–6 wk of voluntary freewheel running reduced hypothalamic concentration of norepinephrine in spontaneously hypertensive (physiologically stressed) rats. The reduction in hypothalamic norepinephrine was responsible for reducing sympathetic nervous system activity and improving cardiovascular function and immune function also reported in these rats. Dishman (3) has also conducted research supporting the hypothesis that voluntary freewheel running can modulate the brain’s response to stress. He has reported that voluntary freewheel running can prevent norepinephrine depletion in locus ceruleus and reduce prefrontal cortex norepinephrine elevations (in vivo microdialysis) after foot shock and that freewheel running decreases in pons-medullary norepinephrine concentration, which is indicative of decreased of reduced peripheral noradrenergic activity.
We have recently reported that voluntary freewheel running produces changes in stress-induced activation of the central sympathetic circuit (8). If we focus on those areas that based on the pseudorabies tracing studies are sites responsible for splenic sympathetic innervation, we find that rats that voluntarily run on a running wheel for 6 wk before exposure to 100 5-s, 1.6-mA tail shocks have a reduction in neural activation (cFos+) of cells in the medial parvicellular subdivisions of the paraventricular hypothalamic nucleus (medial parvicellular subdivisions of the paraventricular hypothalamic nucleus autonomic subnuclei), Barrington’s nucleus, locus ceruleus, A5 cell group, rostral ventrolateral medulla, and caudal raphe magnus nuclei. Importantly, not all regions involved with the central autonomic pathway were affected. For example, tail shock stress activated A7 equally in freewheel run and in sedentary rats (8). What specific adaptations occur in the brain that result in a more constrained SNS response to stressor exposure or environmental challenge are currently unknown and may involve increases in inhibitory neurotransmitters or neurocircuitry, as well as upregulation of inhibitory autoreceptors. We are currently investigating these ideas.
WORKING HYPOTHESIS OF PROTECTION AGAINST STRESS-INDUCED SUPPRESSION OF αKLH IG IN PHYSICALLY ACTIVE ORGANISMS
To date, our results support the hypothesis that 4–6 wk of daily physical activity before exposure to an acute stressor blunts the negative effect of stress on the immune response (Fig. 1). The mechanism(s) for this protective remain a topic of inquiry; however, our current hypothesis is that regular moderate physical activity produces central nervous system adaptations such that exposure to stress results in a constraint in the activation of brain autonomic circuits, including the central sympathetic circuit responsible for innervation of the spleen. A decrease in sympathetic drive during stress prevents splenic norepinephrine depletion. With normal concentrations of norepinephrine present in the spleen, antigen-presenting B cells that have taken up KLH would receive adequate β2AR stimulation and hence express optimal levels of B7.2 (CD86). Optimal expression of costimulatory molecules present on KLH-specific APC would result in normal (i.e., not reduced) numbers of KLH-specific Th cells and Th1 cytokine production. With normal numbers of αKLH Th cells and Th1 cytokines, αKLH Ig levels are also normal in the physically active stressed rat. Importantly, we believe that the protective effect of stress in physically active rats is constrained SNS drive and not caused by a change in ADR receptor function. This is supported by the observation that splenocytes from both sedentary and physically active non-stressed rats respond equally to norepinephrine added in vitro.
CONCLUSION: IMPLICATIONS OF STRESS BUFFERING EFFECTS OF DAILY, MODERATE PHYSICAL ACTIVITY
The goal of this review was to provide evidence using a well-characterized animal model of stress-induced immunosuppression that physical active organisms are more resistant to the negative effects of stress and to offer a potential immunophysiological mechanism for this effect. These data add to a growing literature that physically active organisms are more stress-resistant than sedentary counterparts (7). One immunophysiological mechanism for the stress-buffering effect is that daily moderate physical activity constrains SNS output in response to stressor exposure. Adaptations in stress-induced SNS output produced by physical activity are likely caused by central nervous system and peripheral nervous system changes. Although speculative at this time, the potential positive health consequences of better constraint on SNS activation to stressors are numerous and could include reduction in hypertension, heart failure, oxidative stress, and immunosuppression (15). One intriguing alternative interpretation of the current work is our results also suggest that the sedentary organisms may be more stress-sensitive or more negatively impacted by activation of the stress response. Regardless of one’s interpretation, it is clear that regular moderate physical activity in the form of daily wheel running changes stress physiology, allowing animals to better-buffer the negative impact of stress on acquired immunity.
1. Brown, J.D., and J.M. Siegal. Exercise as a buffer of life stress
: A prospective study of adolescent health. Health Psychol
. 7:341–353, 1988.
2. Cano, G., A.F. Sved, L. Rinaman, B.S. Rabin, and J.P. Card. Characterization of the central nervous system innervation of the rat spleen using viral transneuronal tracing. J. Comp. Neurol
. 439:1–18, 2001.
3. Dishman, R.K. Brain monamines, exercise, and behavioral stress
: Animal models. Med. Sci. Sports Exercise
. 29:63–74, 1997.
4. Fleshner, M., J. Hermann, L.L. Lockwood, M.L. Laudenslager, L.R. Watkins, and S.F. Maier. Stressed rats fail to expand the CD45RC+CD4+ (Th1-like) T cell subset in response to KLH: Possible involvement of IFN-gamma. Brain Behav. Immun
. 9:101–112, 1995.
5. Fleshner, M., K.T. Nguyen, C.S. Cotter, L.R. Watkins, and S.F. Maier. Acute stressor exposure both suppresses acquired immunity and potentiates innate immunity. Am. J. Physiol
. 275:R870–878, 1998.
6. Gazda, L.S., T. Smith, L.R. Watkins, S.F. Maier, and M. Fleshner. Stressor exposure produces long-term reductions in antigen-specific T and B cell responses. Stress
7. Greenwood, B.N., T.E. Foley, H.E. Day, J. Campisi, S.H. Hammack, S. Campeau, S.F. Maier, and M. Fleshner. Freewheel running prevents learned helplessness/behavioral depression: Role of dorsal raphe serotonergic neurons. J. Neurosci
. 23:2889–2898, 2003.
8. Greenwood, B.N., S. Kennedy, T.P. Smith, S. Campeau, H.E. Day, and M. Fleshner. Voluntary freewheel running selectively modulates catecholamine content in peripheral tissue and c-Fos expression in the central sympathetic circuit following exposure to uncontrollable stress
in rats. Neuroscience
9. Janeway, C.A., T. Travers, M. Walport, and M.J., Shlochik. ImmunoBiology: The Immune System in Health and Disease
. New York and London: Current Biology/Garland Science, 2005.
10. Kregel, K.C., D.G. Johnson, and D.R. Seals. Tissue-specific noradrenergic activity during acute heat stress
in rats. J. Appl. Physiol
. 74:1988–1993, 1993.
11. Lambert, G.W., and I.H. Jonsdottir. Influence of voluntary exercise on hypothalamic norepinephrine. J. Appl. Physiology
. 85:962–966, 1998.
12. Moraska, A., T. Deak, R.L. Spencer, D. Roth, and M. Fleshner. Treadmill running produces both positive and negative physiological adaptations in Sprague-Dawley rats. Am. J. Physiol. Regul. Integr. Comp. Physiol
. 279:R1321–R1329, 2000.
13. Moraska, A., and M. Fleshner. Voluntary physical activity prevents stress
-induced behavioral depression and anti-KLH antibody suppression. Am. J. Physiol. Regul. Integr. Comp. Physiol
. 281:R484–489, 2001.
14. Sanders, V.M., and A.P. Kohm. Sympathetic nervous system interaction with the immune system. Int. Rev. Neurobiol
. 52:17–41, 2002.
15. Seals, D., and F. Dinenno. Collateral damage: Cardiovascular consequences of chronic sympathetic activation with human aging. Am. J. Physiol. Heart Circ. Physiol
. 287:H1895–905, 2004.