- Email: [email protected]
Effects of voluntary exercise on immune function in rats
Physiology&Behavior,Vol. 54, pp. 771-774, 1993
0031-9384/93 $6.00 + .00 Copyright © 1993 PergamonPress Ltd.
Printed in the USA.
BRIEF COMMUNICATION
Effects of Voluntary Exercise on Immune Function in Rats KAREN
J. C O L E M A N 1 A N D D A W N
R. R A G E R
Department of Psychology, University of Georgia, Athens, GA 30602 R e c e i v e d 18 A u g u s t 1992 COLEMAN, K. J. AND D. R. RAGER. Effects of voluntary exercise on immunefunction in rats. PHYSIOL BEHAV 54(4) 771774, 1993.--We investigated effects of voluntary wheel running on specific antibody responses to Keyhole Limpet Hemocyanin (KLH) and on mitogen-stimulated lymphocyte proliferation in adult male rats. Each subject was placed in a running wheel for 12 h daily during the dark portion of the light cycle for a total of 8 weeks. For experimental animals the wheels rotated freely, enabling subjects to exercise at will, while for control animals the wheels were prevented from rotating. Subjects were immunized with KLH (25 #g) 5 weeks into the study, and blood samples were collected intermittently from the tail for 3 weeks and later assayed for anti-KLH antibody levels. At the end of the study, subjects were sacrificed and spleens were dissected and assayed for lymphocyte proliferative responses to the mitogen Concanavalin A. Exercised rats gained less weight and had higher splenic proliferative responses than control rats; however, there were no significant group differences in anti-KLH antibody levels. Voluntary exercise
Proliferation
Antibody
Rat
Stress
EXERCISE is reported to be correlated with improved mood, self-esteem, sexual, and work performance (3,16). Exercise has also been shown to prevent and/or reduce the severity of heart disease, colon cancer, and breast cancer (14). These effects of exercise on disease prevention and outcome may be due, at least in part, to its reported ability to enhance i m m u n e function (4,15). Much of the research on exercise and immunity, however, has yielded conflicting results (1,2,6,10-12,14). For example, in animals, exercise has been shown to enhance (10), suppress (6), or have no effect on i m m u n i t y (1). In humans, while exercise is generally perceived of as a stress-reducing, health-promoting behavior (4), when exercise is forced or excessive it may lead to increased disease susceptibility (4). However, h u m a n s are rarely forced to exercise. Rather, most people choose to exercise, and, thus, exert control over the type and a m o u n t of exercise used. Nevertheless, animal studies of exercise effects on i m m u n e function generally use forced exercise such as swimming (2) or treadmill r u n n i n g induced by electric foot shock and physical prodding ( 1,6). Therefore, it is not surprising that m a n y animal studies refer to exercise as a stressor with potentially detrimental effects on health (6,11). Forced exercise may involve psychological stress, such as lack of control (9), in addition to the physical stress of the exercise activity itself. Indeed, forced exercise and psychological stress seem to exert similar effects. For example, both forced exercise
Immunity
and psychological stress suppress i m m u n e responses (6,7,8,1 1). This exercise-induced i m m u n e suppression could result from the lack of control introduced in a forced exercise paradigm. This experiment sought to determine whether voluntary exercise would affect humoral i m m u n e responses to in vivo challenge with a novel antigen (Keyhole Limpet Hemocyanin or KLH) or in vitro splenic lymphocyte proliferation in response to the mitogen Concanavalin A (ConA) in adult male rats. A voluntary exercise protocol was used to eliminate the possible confounding effects of psychological stress on i m m u n e function. Psychological stress has been found to suppress i m m u n e measures such as antibody responses to antigenic challenge (7,8) and mitogen-stimulated lymphocyte proliferation (11). It was predicted that voluntary exercise would enhance both antibody responses and mitogen stimulated lymphocyte proliferation. METHOD
Subjects Twenty Long - E v a n s male rats were obtained from a commercial breeder (Charles River Laboratories, Wilmington, DE) at approximately 60 days of age. Animals were maintained on a 12:12 light:dark cycle with light onset at 0800 h, and were housed individually in hanging galvanized steel wire-mesh cages. Standard rat chow and water were available ad lib.
Requests for reprints should be addressed to Karen J. Coleman.
771
772
Procedure Upon arrival at the laboratory, animals were left undisturbed and allowed to acclimate to laboratory conditions for 1 week. Subjects were then randomly divided into two groups of 10 animals each: an experimental group (E) that was permitted to exercise freely during the dark (active) phase of the cycle, and a control group (C) that was not permitted the opportunity to exercise. During the experiment, each animal was transferred daily from the home cage to a running wheel (area of wheel = 1017.88 cm2: area of housing box = 364 cm 2) at dark onset, and left undisturbed in the wheel for 12 h. Food and water were available ad lib in the wheels. For subjects in group E, the wheels rotated freely so that the animals had an opportunity to run throughout the night. For subjects in group C, running wheels were locked, preventing running during the dark phase. Immediately following light onset, all subjects were removed from the wheels and returned to their home cages where they were left undisturbed for the remainder of the light portion of the cycle. The distance run in cm/h by each subject in group E was monitored throughout the experiment by recording the total number of wheel revolutions per 12-h running period for each subject and entering this number into the formula: (2pi × radius) × revolutions/h. Body weight measures in g were taken weekly for the first 2 weeks oftbe study, and biweekly thereafter for all subjects. To habituate subjects to the procedure that would be used during the last 3 weeks of the study for blood sampling, subjects were briefly restrained three times a week. Specifically, following removal from the wheels at light onset, each animal was lightly restrained by wrapping it in a towel and fastening a velcro strap around the animal to prevent escape. Each subject remained restrained for approximately 1 min and then was transferred to its home cage. The experiment continued for 8 weeks, because running activity has been reported to peak between 5-7 weeks and then begin to decline by week 8 (13). At the end of the fifth week, each subject received an intraperitoneal injection of 25 ug of KLH suspended in 0.1 ml sterile 0.9% saline. Peripheral blood samples were collected 3, 5, 7, 9, 14, and 21 days after immunization. Blood was obtained by restraining animals as previously described, nicking the tail vein with a clean scalpel, and gently milking approximately 0.5 ml of blood from the tail of each subject. This procedure caused the animals minimal discomfort/ distress and was usually accomplished in less than 5 min. Immediately after collection, samples were allowed to clot and then were centrifuged at 1600 × g for 15 min. The serum was then removed and stored at - 2 0 ° C for later analysis of KLH-specific antibody levels as described in the following paragraphs. Following the final exercise period of the 8-week experiment, each subject was euthanized by CO2 asphyxiation, and the spleen was immediately dissected and assayed for ConA-stimulated lymphocyte proliferation as described in the following paragraphs.
Assay of KLH-Specific Antibody Levels Levels of anti-KLH IgG and lgM antibodies in peripheral sera were determined using an enzyme-linked immunosorbent assay (ELISA). The wells of 96-well flat-bottomed microtiter plates were coated with antigen by adding 200 ttl of coating buffer (0.5 mg/ml KLH in sodium bicarbonate buffer) to each well and incubating at 5°C overnight. The following day the wells of the plates were washed three times with phosphate-buffered saline containing 0.05% Tween 20 (PBS-T). Serum samples were thawed, diluted 1:100 with PBS-T, and 150/~1 of each serum dilution was added in duplicate to the wells of an antigen-coated plate. Also included on each plate was a positive control sample
COLEMAN AND RAGER (pooled sera from rats that had previously responded with antiKLH lgG and IgM antibodies, diluted 1:100 with PBS-T), and a negative control sample (pooled sera from rats that had not been immunized with KLH, diluted 1:100 with PBS-T). Plates containing sera samples were then sealed with plastic tape, incubated at 37°C for 3 h and washed three times with PBS-T. Next, 150 #1 of alkaline phosphatase-conjugated antiantibody (Zymed rabbit antirat IgM or rabbit antirat IgG) diluted 1:1000 with PBS-T was added to the wells, and plates were sealed and incubated (1.5 h for IgM, 1.0 h for IgG) at 37°C. At the end of the secondary antibody incubation plates were washed three times with PBS-T, 150 #1 of enzyme substrate (1 mg Sigma Reagent 104 per 1 ml diethanolamine substrate buffer) was added to each well of the plate, and plates were protected from light during the enzyme-substrate reaction. Reactions were terminated after approximately 30 min for lgG and 60 min for IgM by adding 100 #1 of 1.5 M NaOH to each well. Plates were read 5 min following addition of NaOH at 410 nm on a Dynatech Minireader II, and optical density was recorded in absorbance units. The optical density of each well is proportional to the amount of enzyme, which is, in turn, related to the concentration of anti-KLH antibody in the serum sample. To control for interplate variability within an assay, data are expressed as percentage of plate positive absorbance (18), and average plate positive absorbance values for each set of duplicate wells were used in statistical analyses.
Assay of Mitogen-Stimulated Lymphocyte Proliferation Splenic lymphocyte proliferative responses to ConA were determined using the CellTiter 96 NonRadioactive Cell Proliferation/Cytotoxicity Assay (Promega, Madison, WI). This assay involves addition of a dye solution containing a tetrazolium salt to cell cultures. Live cells convert the salt to a blue formazan by-product that can then be solubilized and detected using a photometric plate reader. The optical density (expressed in absorbance units) is directly proportional to the number of viable cells in culture. This assay method has been reported to yield similar results as compared to the traditional 3H-thymidine incorporation method (5,17). Unpublished data from our own laboratory are in agreement with these reports, with correlation coefficients for ConA-stimulated proliferation ranging between 0.57 and 0.89 (n = 30, p < .001) for the two methods. For the assay, splenic lymphocytes were separated from tissue by gently pressing the whole spleen between sterile frosted glass slides, and the separated cells were suspended in 7 ml of culture medium containing RPMI-1640/Hepes supplemented with 1% penicillin (5000 U/ml)/streptomycin (5000 t~g/ml), 1% L-glutamine (2 mM/ml), 0.1% 2-mercaptoethanol (5 × 10-2 M/ml), and 10% heat-inactivated fetal calf serum. Splenic white blood cell counts and viability were then determined using a hemacytometer and Trypan blue exclusion (red blood cells were excluded from cell counts). The number of viable white blood cells per ml (which always exceeded 95%) was then adjusted to 2 × 10 6 cells/ml by dilution with culture medium. Fifty microliters of each cell suspension ( 100,000 cells) was then added in triplicate to the wells of sterile fiat-bottomed 96 well tissue culture plates. Concanavalin A was diluted with culture medium to concentrations of 40, 20, 10, and 5 t~g/ml, and 50 tsl of each mitogen concentration was added in triplicate to the wells of the plate containing the spleen cell suspensions, yielding final mitogen concentrations of 20, 10, 5, and 2,5 tsg/ml. Unstimulated control wells containing 50 tal of spleen cell suspension plus 50 ~tl of culture medium were also included in triplicate for each subject. Plates were then incubated at 37 °C in a humidified atmosphere
EFFECTS OF EXERCISE ON IMMUNE FUNCTION
TABLE 1 RUNNING ACTIVITY OF EXERCISED RATS AND BODY WEIGHT OF CONTROL AND EXERCISED RATS AS MEASURED FOR EACH WEEK OF THE STUDY
Wl W2 W3 W4 W5 W6 W7 W8
Running (cm/h)
Weight (g) Controls)
Weight (g) Exercise
3835 _+ 473.62 11027 + 1027.0 16579 +_2089.8 17940 + 2011.9 17842 _+2161.5 16611 + 2095.6 13381 _+ 1688.6 11156_ 1014.6
279 _+ 3.16 328 _+ 8,22 -357 _+ 10,4 -383 _+ 11.7 -426 ___14.6
270 + 3.79 288 _+ 6.64 -327 + 3.16 -354 + 8.54 -401 _+ 10.4
Mean _+SEM. Weights for all rats were taken weekly for the first 2 weeks of the experiment and then biweeklythereafter.
with 5% CO2 for 44 h. After 44 h 15 ~tl of dye solution was added to each well of the plate, and plates were incubated for an additional 4 h at 37°C in a humidified 5% CO2 incubator. Following this incubation, 100 ~1 of solubilization solution was added to each well of the plate, and plates were covered, and incubated in a humidified 37°C chamber overnight. Plates were then read at 570 nm on a Dynatech Minireader II, and median absorbance values for each set of triplicate wells were used in statistical analyses. RESULTS
Running Activity
773
mean negative serum control samples by two standard deviations were excluded from statistical analyses. Mean percent plate-positive values (+SEM) for anti-KLH IgM on days 3, 5, 7, 9, 14, and 21 were 23.18 _+ 2.85, 41.97 _+ 5.36, 41.29 _+ 5.35, 58.27 + 8.42, 57.29 + 8.94, and 55.03 _+ 10.28 for exercised subjects and 34.50 _+4.96, 56.40 __+9.01, 58.36 _+ 10.69, 79.98 _+ 10.18, 79.15 _+ 10.16, and 52.84 _+ 8.59 for control rats. For anti-KLH IgG, mean percent plate-positive values for days 5, 7, 9, 14, and 21 were 30.36 _+ 3.7, 37.02 _+ 3.86, 46.79 _+ 6.09, 68.10 _+ 7.93, and 76.31 p 9.39 for exercised subjects and 40.67 _+ 7.23, 47.81 _+7.94, 59.81 _+8.07, 73.33 _+6.77, and 77.62 + 6.63 for control rats. These data were analyzed using separate two-way ANOVAs with one repeated factor (day of response). The analyses revealed no significant differences between exercised and control rats for either anti-KLH IgM or anti-KLH IgG responses.
Mitogen-Stimulated Lymphocyte Proliferation Splenic lymphocyte proliferative responses to ConA are illustrated in Fig. I. One control subject was excluded from the data analysis due to an infection incurred during the last week of the study. A two-way ANOVA with one repeated factor (mitogen concentration) was performed on median absorbance values for each set of triplicate wells. The analysis yielded a significant ConA concentration × condition interaction, F(4, 68) = 3.24, p = 0.017, as well as significant main effects of condition, F(1, 17) = 5.6 l, p = 0.03, and ConA concentration, F(4, 68) = 259. l 0, p < 0.001. Thus, subjects in both conditions exhibited typical inverted U-shaped dose response curves, with exercised rats generally having higher splenic proliferative responses than control rats. Tukey's Multiple Comparisons revealed that exercised rats exhibited higher proliferative responses (p < 0.05) at lower concentrations of ConA (2.5, 5, and l0 #g/ml), however, for both ConA 20/~g/ml and unstimulated control cultures there were no significant differences between conditions. In addition, there
The average weekly running activity (expressed in cm/h) for subjects in group E is illustrated in Table 1. A one-way repeated measures analysis of variance (ANOVA) performed on these data showed that running changed significantly as a function of time, F(7, 63) = 19.906, p < 0.001. Tukey's Multiple Comparisons revealed a significant increase in running activity during the first 3 weeks of the study, followed by a plateau beginning on week 4 and continuing through week 6, and a significant decline in activity by week 8 (p < 0.05). The pattern of wheel running activity observed in the present experiment is similar to what has previously been reported (13).
~
Body Weight
. o.s
Mean body weights for exercised and control groups throughout the experiment are also illustrated in Table 1. Changes in body weight for exercised and control subjects were analyzed using a two-way ANOVA with one repeated factor (time). A significant condition × time interaction was found, F(4, 72) = 3.059, p -- 0.022, along with significant main effects of both condition, F(1, 18) = 5.18 l, p = 0.035, and time, F(4, 72) = 282.448, p < 0.001. Thus, in general, exercised rats weighed less than control rats, and all subjects gained weight throughout the study. Tukey's Multiple Comparisons showed that although body weights did not differ between exercised and control rats at the start of the experiment, control rats gained significantly more weight than exercised rats within the first 2 weeks of the study (p < 0.05).
Antibody Responses to KLH Subjects whose peak anti-KLH IgG and IgM antibody responses failed to exceed the percent plate positive values of the
1.0
0
Exercise (N-=IO)
)
0.9 0.8 0.7
~0.5
~ 0.4 o.a
~ 0.2 0.1 0.0
I
I
I
i
I
I
I
0.0 2.5 5.0 10.0 Conoonovolin A ( u g / m l )
I
i
20.0
FIG. 1. Spleniclymphocyteproliferation(expressedas median absorbance _+ SEM) in response to various concentrations of Concanavalin A for exercised and control subjects.
774
COLEMAN AND RAGER
were no significant differences between unstimulated control cultures and proliferative responses to ConA 20 ~g/ml regardless of condition. DISCUSSION The results of the present experiment showed that when adult male rats are allowed to exercise freely in running wheels, running activity peaks and plateaus after about 5 weeks and then begins to decline by the eighth week of exercise. Because animals were immunized with KLH during week 5, it could be speculated that the subsequent change in running activity was a result of immunization. However, this interpretation seems unlikely given that similar patterns of running activity have been reported to occur in n o n i m m u n i z e d animals (13). It was also observed that exercised rats gained significantly less weight than control rats, especially within the first 2 weeks of the experiment. These differences in body weight are also consistent with the findings of Roebuck et at. (13) that exercised rats weighed less and had decreased fat deposits as compared to control rats during an 8week voluntary wheel running paradigm. With regard to the effects of voluntary wheel running on i m m u n e function, the present experiment found that although there were no significant differences between exercised and control rats in specific antibody responses to KLH, exercised rats exhibited higher splenic lymphocyte proliferative responses to the mitogen ConA as compared to nonexercised controls. It has been reported that antibody responses in mice were enhanced by wheel running (10), although in these studies mice were immunized with a large dose of antigen and then reexposed to the same antigen 10 days after the initial immunization, during the first 4 weeks of exercise. This reexposure paradigm may have involved both primary and secondary (memory) responses to
the antigen. The present study, however, suggests that the primary antibody response is unaffected by exercise, and is consistent with other reports that exercise has no effect on this response in rodents (1,2). In humans, although enhanced immunoglobulin levels following exercise have occasionally been reported, most of the evidence suggests that exercise has no effect on humoral i m m u n i t y (12,15). Splenic lymphocyte proliferation has been reported to be suppressed in rats following an acute bout of exhaustive swimming (11) and a 4-week conditioned treadmill running paradigm (6). In contrast to these findings, the present study found enhanced lymphocyte proliferation in voluntarily exercised rats. An explanation for these apparently contradictory findings may involve differential effects of forced vs. voluntary exercise paradigms. Forced exercise paradigms may involve aspects of psychological stress (9), not appearing in a voluntary exercise protocol, that have been reported to suppress i m m u n e responses (6-8,11). The enhanced lymphocyte proliferation observed in exercised rats could alternatively be interpreted as suppressed proliferative responses in controls. Both exercise and control rats were exposed to the acute stress of handling, brief exposure to a novel environment (i.e., the CO2 chamber), and 02 deprivation immediately prior to sacrifice at the end of the present study. It is possible that exercise attenuated the response to these potential stressors that was evidenced in control animals by suppression of lymphocyte proliferation. Although Dishman (3) has summarized evidence that exercise is associated with reduced forms of depression and anxiety in the h u m a n population, studies of the attenuating effects of voluntary exercise on stress in rodents are lacking. This is clearly an important question that deserves further investigation.
REFERENCES 1. Barnes, C. A.; Forster, M. J.; Fleshner, M.; et al. Exercise does not modify spatial memory, brain autoimmunity or antibody response in aged F-44 rats. Neurobiol. Aging 12:47-53; 1991. 2. Carmack, M. A. Exercise induced modifications in immune responsiveness in rats. Unpublished doctoral dissertation, University of Oregon. Eugene, OR. 1984. 3. Dishman, R. K. Psychological eft~cts of exercise for disease resistance and health promotion. In: Eisinger, M.; Watson, R., eds. Exercise: Disease and cancer resistance. Boca Raton, FL.: CRC Press; 1992:
10. 11. 12.
1-48.
4. Fitzgerald, L. Exercise and the immune system, lmmun. Today 9: 337-339; 1988. 5. Hansen, M. B.; Nielsen, S. E.; Berg, K. Reexamination and further development of a precise and rapid dye method for measuring cell growth/cell kill. J. lmmunol. Methods 119:203-210; 1989. 6. Hoffman-Goetz, L.; Keir, R.; Thorne, R.; Houston, M. E.; Young, C. Chronic exercise stress in mice depresses splenic T-lymphocyte mitogenesis in vitro. Clin. Exp. lmmunol. 6:551-557; 1986. 7. Kiecolt-Glaser, J. K.; Glaser, R. Stress and immune function in humans. In: Ader, R., ed. Psychoneuroimmunology, 2nd ed. New York: Academic Press, Inc,; 1991:849-867. 8. Laudenslager, M. L.; Fleshner, M.; Hofstadter, P.; Held, P. E.; Sb mons, L.; Maier, S. F. Suppression specific antibody production by inescapable shock: Stability under varying conditions. Brain Behav. lmmunol. 2:92-101; 1988. 9. Levine, S.; Coe, C.; Wiener, S. G. Psychoneuroendocrinology of
13. 14. 15. 16. 17. 18.
stress: A psychobiological perspective. In: Psychoendocrinology. New York: Academic Press, Inc.; 1989:341-377. Liu, Y.; Wang, S. Y. The enhancing effect of exercise on the production of antibody to Salmonella typhi in mice. Immunol. Lett. 14:117-120; 1986/1987. Mahan, M. P.: Young, M. R. Immune parameters of untrained or exercise-trained rats after exhaustive exercise. J. Appl. Physiol. 66: 282-287; 1989. Niemann, D. C.; Nehlsen-Cannarella, S. L. The effects of acute and chronic exercise on immunoglobulins. Sports Med. 11:183-201; 1991. Roebuck, B. D.; McCaffrey, J.; Baumgartner, K. J. Protective effects of voluntary exercise during the postinitiation phase of pancreatic carcinogenesis in the rat. Cancer Res. 50:6811-6816; 1990. Simon, H. B. Exercise and human immune function. In: Ader, R., ed. Psychoneuroimmunology, 2nd ed. New York: Academic Press, Inc.; 1991:869-895. Simon, H. B. The immunology of exercise. JAMA 252:2735-2738; 1984. Steptoe, A.; Cox, S. Acute effects of aerobic exercise on mood. Health Psychol. 7:329-340; 1988. Tada, H.; Shiho, O.: Kuroshima, K.; Koyama, M.; Tsukamoto, K. An improved colorimetric assay for interleukin 2. J. Immunol. Methods 93:157-165; 1986. Voller, A.; Bidwell, D. E.: Bartlett, A. Microplate ELISA and its applications. In: Malvano, R., ed. Immunoenzymatic assay techniques. Boston: Martinus Nijhoff Publishers; 1980:105-132.
0031-9384/93 $6.00 + .00 Copyright © 1993 PergamonPress Ltd.
Printed in the USA.
BRIEF COMMUNICATION
Effects of Voluntary Exercise on Immune Function in Rats KAREN
J. C O L E M A N 1 A N D D A W N
R. R A G E R
Department of Psychology, University of Georgia, Athens, GA 30602 R e c e i v e d 18 A u g u s t 1992 COLEMAN, K. J. AND D. R. RAGER. Effects of voluntary exercise on immunefunction in rats. PHYSIOL BEHAV 54(4) 771774, 1993.--We investigated effects of voluntary wheel running on specific antibody responses to Keyhole Limpet Hemocyanin (KLH) and on mitogen-stimulated lymphocyte proliferation in adult male rats. Each subject was placed in a running wheel for 12 h daily during the dark portion of the light cycle for a total of 8 weeks. For experimental animals the wheels rotated freely, enabling subjects to exercise at will, while for control animals the wheels were prevented from rotating. Subjects were immunized with KLH (25 #g) 5 weeks into the study, and blood samples were collected intermittently from the tail for 3 weeks and later assayed for anti-KLH antibody levels. At the end of the study, subjects were sacrificed and spleens were dissected and assayed for lymphocyte proliferative responses to the mitogen Concanavalin A. Exercised rats gained less weight and had higher splenic proliferative responses than control rats; however, there were no significant group differences in anti-KLH antibody levels. Voluntary exercise
Proliferation
Antibody
Rat
Stress
EXERCISE is reported to be correlated with improved mood, self-esteem, sexual, and work performance (3,16). Exercise has also been shown to prevent and/or reduce the severity of heart disease, colon cancer, and breast cancer (14). These effects of exercise on disease prevention and outcome may be due, at least in part, to its reported ability to enhance i m m u n e function (4,15). Much of the research on exercise and immunity, however, has yielded conflicting results (1,2,6,10-12,14). For example, in animals, exercise has been shown to enhance (10), suppress (6), or have no effect on i m m u n i t y (1). In humans, while exercise is generally perceived of as a stress-reducing, health-promoting behavior (4), when exercise is forced or excessive it may lead to increased disease susceptibility (4). However, h u m a n s are rarely forced to exercise. Rather, most people choose to exercise, and, thus, exert control over the type and a m o u n t of exercise used. Nevertheless, animal studies of exercise effects on i m m u n e function generally use forced exercise such as swimming (2) or treadmill r u n n i n g induced by electric foot shock and physical prodding ( 1,6). Therefore, it is not surprising that m a n y animal studies refer to exercise as a stressor with potentially detrimental effects on health (6,11). Forced exercise may involve psychological stress, such as lack of control (9), in addition to the physical stress of the exercise activity itself. Indeed, forced exercise and psychological stress seem to exert similar effects. For example, both forced exercise
Immunity
and psychological stress suppress i m m u n e responses (6,7,8,1 1). This exercise-induced i m m u n e suppression could result from the lack of control introduced in a forced exercise paradigm. This experiment sought to determine whether voluntary exercise would affect humoral i m m u n e responses to in vivo challenge with a novel antigen (Keyhole Limpet Hemocyanin or KLH) or in vitro splenic lymphocyte proliferation in response to the mitogen Concanavalin A (ConA) in adult male rats. A voluntary exercise protocol was used to eliminate the possible confounding effects of psychological stress on i m m u n e function. Psychological stress has been found to suppress i m m u n e measures such as antibody responses to antigenic challenge (7,8) and mitogen-stimulated lymphocyte proliferation (11). It was predicted that voluntary exercise would enhance both antibody responses and mitogen stimulated lymphocyte proliferation. METHOD
Subjects Twenty Long - E v a n s male rats were obtained from a commercial breeder (Charles River Laboratories, Wilmington, DE) at approximately 60 days of age. Animals were maintained on a 12:12 light:dark cycle with light onset at 0800 h, and were housed individually in hanging galvanized steel wire-mesh cages. Standard rat chow and water were available ad lib.
Requests for reprints should be addressed to Karen J. Coleman.
771
772
Procedure Upon arrival at the laboratory, animals were left undisturbed and allowed to acclimate to laboratory conditions for 1 week. Subjects were then randomly divided into two groups of 10 animals each: an experimental group (E) that was permitted to exercise freely during the dark (active) phase of the cycle, and a control group (C) that was not permitted the opportunity to exercise. During the experiment, each animal was transferred daily from the home cage to a running wheel (area of wheel = 1017.88 cm2: area of housing box = 364 cm 2) at dark onset, and left undisturbed in the wheel for 12 h. Food and water were available ad lib in the wheels. For subjects in group E, the wheels rotated freely so that the animals had an opportunity to run throughout the night. For subjects in group C, running wheels were locked, preventing running during the dark phase. Immediately following light onset, all subjects were removed from the wheels and returned to their home cages where they were left undisturbed for the remainder of the light portion of the cycle. The distance run in cm/h by each subject in group E was monitored throughout the experiment by recording the total number of wheel revolutions per 12-h running period for each subject and entering this number into the formula: (2pi × radius) × revolutions/h. Body weight measures in g were taken weekly for the first 2 weeks oftbe study, and biweekly thereafter for all subjects. To habituate subjects to the procedure that would be used during the last 3 weeks of the study for blood sampling, subjects were briefly restrained three times a week. Specifically, following removal from the wheels at light onset, each animal was lightly restrained by wrapping it in a towel and fastening a velcro strap around the animal to prevent escape. Each subject remained restrained for approximately 1 min and then was transferred to its home cage. The experiment continued for 8 weeks, because running activity has been reported to peak between 5-7 weeks and then begin to decline by week 8 (13). At the end of the fifth week, each subject received an intraperitoneal injection of 25 ug of KLH suspended in 0.1 ml sterile 0.9% saline. Peripheral blood samples were collected 3, 5, 7, 9, 14, and 21 days after immunization. Blood was obtained by restraining animals as previously described, nicking the tail vein with a clean scalpel, and gently milking approximately 0.5 ml of blood from the tail of each subject. This procedure caused the animals minimal discomfort/ distress and was usually accomplished in less than 5 min. Immediately after collection, samples were allowed to clot and then were centrifuged at 1600 × g for 15 min. The serum was then removed and stored at - 2 0 ° C for later analysis of KLH-specific antibody levels as described in the following paragraphs. Following the final exercise period of the 8-week experiment, each subject was euthanized by CO2 asphyxiation, and the spleen was immediately dissected and assayed for ConA-stimulated lymphocyte proliferation as described in the following paragraphs.
Assay of KLH-Specific Antibody Levels Levels of anti-KLH IgG and lgM antibodies in peripheral sera were determined using an enzyme-linked immunosorbent assay (ELISA). The wells of 96-well flat-bottomed microtiter plates were coated with antigen by adding 200 ttl of coating buffer (0.5 mg/ml KLH in sodium bicarbonate buffer) to each well and incubating at 5°C overnight. The following day the wells of the plates were washed three times with phosphate-buffered saline containing 0.05% Tween 20 (PBS-T). Serum samples were thawed, diluted 1:100 with PBS-T, and 150/~1 of each serum dilution was added in duplicate to the wells of an antigen-coated plate. Also included on each plate was a positive control sample
COLEMAN AND RAGER (pooled sera from rats that had previously responded with antiKLH lgG and IgM antibodies, diluted 1:100 with PBS-T), and a negative control sample (pooled sera from rats that had not been immunized with KLH, diluted 1:100 with PBS-T). Plates containing sera samples were then sealed with plastic tape, incubated at 37°C for 3 h and washed three times with PBS-T. Next, 150 #1 of alkaline phosphatase-conjugated antiantibody (Zymed rabbit antirat IgM or rabbit antirat IgG) diluted 1:1000 with PBS-T was added to the wells, and plates were sealed and incubated (1.5 h for IgM, 1.0 h for IgG) at 37°C. At the end of the secondary antibody incubation plates were washed three times with PBS-T, 150 #1 of enzyme substrate (1 mg Sigma Reagent 104 per 1 ml diethanolamine substrate buffer) was added to each well of the plate, and plates were protected from light during the enzyme-substrate reaction. Reactions were terminated after approximately 30 min for lgG and 60 min for IgM by adding 100 #1 of 1.5 M NaOH to each well. Plates were read 5 min following addition of NaOH at 410 nm on a Dynatech Minireader II, and optical density was recorded in absorbance units. The optical density of each well is proportional to the amount of enzyme, which is, in turn, related to the concentration of anti-KLH antibody in the serum sample. To control for interplate variability within an assay, data are expressed as percentage of plate positive absorbance (18), and average plate positive absorbance values for each set of duplicate wells were used in statistical analyses.
Assay of Mitogen-Stimulated Lymphocyte Proliferation Splenic lymphocyte proliferative responses to ConA were determined using the CellTiter 96 NonRadioactive Cell Proliferation/Cytotoxicity Assay (Promega, Madison, WI). This assay involves addition of a dye solution containing a tetrazolium salt to cell cultures. Live cells convert the salt to a blue formazan by-product that can then be solubilized and detected using a photometric plate reader. The optical density (expressed in absorbance units) is directly proportional to the number of viable cells in culture. This assay method has been reported to yield similar results as compared to the traditional 3H-thymidine incorporation method (5,17). Unpublished data from our own laboratory are in agreement with these reports, with correlation coefficients for ConA-stimulated proliferation ranging between 0.57 and 0.89 (n = 30, p < .001) for the two methods. For the assay, splenic lymphocytes were separated from tissue by gently pressing the whole spleen between sterile frosted glass slides, and the separated cells were suspended in 7 ml of culture medium containing RPMI-1640/Hepes supplemented with 1% penicillin (5000 U/ml)/streptomycin (5000 t~g/ml), 1% L-glutamine (2 mM/ml), 0.1% 2-mercaptoethanol (5 × 10-2 M/ml), and 10% heat-inactivated fetal calf serum. Splenic white blood cell counts and viability were then determined using a hemacytometer and Trypan blue exclusion (red blood cells were excluded from cell counts). The number of viable white blood cells per ml (which always exceeded 95%) was then adjusted to 2 × 10 6 cells/ml by dilution with culture medium. Fifty microliters of each cell suspension ( 100,000 cells) was then added in triplicate to the wells of sterile fiat-bottomed 96 well tissue culture plates. Concanavalin A was diluted with culture medium to concentrations of 40, 20, 10, and 5 t~g/ml, and 50 tsl of each mitogen concentration was added in triplicate to the wells of the plate containing the spleen cell suspensions, yielding final mitogen concentrations of 20, 10, 5, and 2,5 tsg/ml. Unstimulated control wells containing 50 tal of spleen cell suspension plus 50 ~tl of culture medium were also included in triplicate for each subject. Plates were then incubated at 37 °C in a humidified atmosphere
EFFECTS OF EXERCISE ON IMMUNE FUNCTION
TABLE 1 RUNNING ACTIVITY OF EXERCISED RATS AND BODY WEIGHT OF CONTROL AND EXERCISED RATS AS MEASURED FOR EACH WEEK OF THE STUDY
Wl W2 W3 W4 W5 W6 W7 W8
Running (cm/h)
Weight (g) Controls)
Weight (g) Exercise
3835 _+ 473.62 11027 + 1027.0 16579 +_2089.8 17940 + 2011.9 17842 _+2161.5 16611 + 2095.6 13381 _+ 1688.6 11156_ 1014.6
279 _+ 3.16 328 _+ 8,22 -357 _+ 10,4 -383 _+ 11.7 -426 ___14.6
270 + 3.79 288 _+ 6.64 -327 + 3.16 -354 + 8.54 -401 _+ 10.4
Mean _+SEM. Weights for all rats were taken weekly for the first 2 weeks of the experiment and then biweeklythereafter.
with 5% CO2 for 44 h. After 44 h 15 ~tl of dye solution was added to each well of the plate, and plates were incubated for an additional 4 h at 37°C in a humidified 5% CO2 incubator. Following this incubation, 100 ~1 of solubilization solution was added to each well of the plate, and plates were covered, and incubated in a humidified 37°C chamber overnight. Plates were then read at 570 nm on a Dynatech Minireader II, and median absorbance values for each set of triplicate wells were used in statistical analyses. RESULTS
Running Activity
773
mean negative serum control samples by two standard deviations were excluded from statistical analyses. Mean percent plate-positive values (+SEM) for anti-KLH IgM on days 3, 5, 7, 9, 14, and 21 were 23.18 _+ 2.85, 41.97 _+ 5.36, 41.29 _+ 5.35, 58.27 + 8.42, 57.29 + 8.94, and 55.03 _+ 10.28 for exercised subjects and 34.50 _+4.96, 56.40 __+9.01, 58.36 _+ 10.69, 79.98 _+ 10.18, 79.15 _+ 10.16, and 52.84 _+ 8.59 for control rats. For anti-KLH IgG, mean percent plate-positive values for days 5, 7, 9, 14, and 21 were 30.36 _+ 3.7, 37.02 _+ 3.86, 46.79 _+ 6.09, 68.10 _+ 7.93, and 76.31 p 9.39 for exercised subjects and 40.67 _+ 7.23, 47.81 _+7.94, 59.81 _+8.07, 73.33 _+6.77, and 77.62 + 6.63 for control rats. These data were analyzed using separate two-way ANOVAs with one repeated factor (day of response). The analyses revealed no significant differences between exercised and control rats for either anti-KLH IgM or anti-KLH IgG responses.
Mitogen-Stimulated Lymphocyte Proliferation Splenic lymphocyte proliferative responses to ConA are illustrated in Fig. I. One control subject was excluded from the data analysis due to an infection incurred during the last week of the study. A two-way ANOVA with one repeated factor (mitogen concentration) was performed on median absorbance values for each set of triplicate wells. The analysis yielded a significant ConA concentration × condition interaction, F(4, 68) = 3.24, p = 0.017, as well as significant main effects of condition, F(1, 17) = 5.6 l, p = 0.03, and ConA concentration, F(4, 68) = 259. l 0, p < 0.001. Thus, subjects in both conditions exhibited typical inverted U-shaped dose response curves, with exercised rats generally having higher splenic proliferative responses than control rats. Tukey's Multiple Comparisons revealed that exercised rats exhibited higher proliferative responses (p < 0.05) at lower concentrations of ConA (2.5, 5, and l0 #g/ml), however, for both ConA 20/~g/ml and unstimulated control cultures there were no significant differences between conditions. In addition, there
The average weekly running activity (expressed in cm/h) for subjects in group E is illustrated in Table 1. A one-way repeated measures analysis of variance (ANOVA) performed on these data showed that running changed significantly as a function of time, F(7, 63) = 19.906, p < 0.001. Tukey's Multiple Comparisons revealed a significant increase in running activity during the first 3 weeks of the study, followed by a plateau beginning on week 4 and continuing through week 6, and a significant decline in activity by week 8 (p < 0.05). The pattern of wheel running activity observed in the present experiment is similar to what has previously been reported (13).
~
Body Weight
. o.s
Mean body weights for exercised and control groups throughout the experiment are also illustrated in Table 1. Changes in body weight for exercised and control subjects were analyzed using a two-way ANOVA with one repeated factor (time). A significant condition × time interaction was found, F(4, 72) = 3.059, p -- 0.022, along with significant main effects of both condition, F(1, 18) = 5.18 l, p = 0.035, and time, F(4, 72) = 282.448, p < 0.001. Thus, in general, exercised rats weighed less than control rats, and all subjects gained weight throughout the study. Tukey's Multiple Comparisons showed that although body weights did not differ between exercised and control rats at the start of the experiment, control rats gained significantly more weight than exercised rats within the first 2 weeks of the study (p < 0.05).
Antibody Responses to KLH Subjects whose peak anti-KLH IgG and IgM antibody responses failed to exceed the percent plate positive values of the
1.0
0
Exercise (N-=IO)
)
0.9 0.8 0.7
~0.5
~ 0.4 o.a
~ 0.2 0.1 0.0
I
I
I
i
I
I
I
0.0 2.5 5.0 10.0 Conoonovolin A ( u g / m l )
I
i
20.0
FIG. 1. Spleniclymphocyteproliferation(expressedas median absorbance _+ SEM) in response to various concentrations of Concanavalin A for exercised and control subjects.
774
COLEMAN AND RAGER
were no significant differences between unstimulated control cultures and proliferative responses to ConA 20 ~g/ml regardless of condition. DISCUSSION The results of the present experiment showed that when adult male rats are allowed to exercise freely in running wheels, running activity peaks and plateaus after about 5 weeks and then begins to decline by the eighth week of exercise. Because animals were immunized with KLH during week 5, it could be speculated that the subsequent change in running activity was a result of immunization. However, this interpretation seems unlikely given that similar patterns of running activity have been reported to occur in n o n i m m u n i z e d animals (13). It was also observed that exercised rats gained significantly less weight than control rats, especially within the first 2 weeks of the experiment. These differences in body weight are also consistent with the findings of Roebuck et at. (13) that exercised rats weighed less and had decreased fat deposits as compared to control rats during an 8week voluntary wheel running paradigm. With regard to the effects of voluntary wheel running on i m m u n e function, the present experiment found that although there were no significant differences between exercised and control rats in specific antibody responses to KLH, exercised rats exhibited higher splenic lymphocyte proliferative responses to the mitogen ConA as compared to nonexercised controls. It has been reported that antibody responses in mice were enhanced by wheel running (10), although in these studies mice were immunized with a large dose of antigen and then reexposed to the same antigen 10 days after the initial immunization, during the first 4 weeks of exercise. This reexposure paradigm may have involved both primary and secondary (memory) responses to
the antigen. The present study, however, suggests that the primary antibody response is unaffected by exercise, and is consistent with other reports that exercise has no effect on this response in rodents (1,2). In humans, although enhanced immunoglobulin levels following exercise have occasionally been reported, most of the evidence suggests that exercise has no effect on humoral i m m u n i t y (12,15). Splenic lymphocyte proliferation has been reported to be suppressed in rats following an acute bout of exhaustive swimming (11) and a 4-week conditioned treadmill running paradigm (6). In contrast to these findings, the present study found enhanced lymphocyte proliferation in voluntarily exercised rats. An explanation for these apparently contradictory findings may involve differential effects of forced vs. voluntary exercise paradigms. Forced exercise paradigms may involve aspects of psychological stress (9), not appearing in a voluntary exercise protocol, that have been reported to suppress i m m u n e responses (6-8,11). The enhanced lymphocyte proliferation observed in exercised rats could alternatively be interpreted as suppressed proliferative responses in controls. Both exercise and control rats were exposed to the acute stress of handling, brief exposure to a novel environment (i.e., the CO2 chamber), and 02 deprivation immediately prior to sacrifice at the end of the present study. It is possible that exercise attenuated the response to these potential stressors that was evidenced in control animals by suppression of lymphocyte proliferation. Although Dishman (3) has summarized evidence that exercise is associated with reduced forms of depression and anxiety in the h u m a n population, studies of the attenuating effects of voluntary exercise on stress in rodents are lacking. This is clearly an important question that deserves further investigation.
REFERENCES 1. Barnes, C. A.; Forster, M. J.; Fleshner, M.; et al. Exercise does not modify spatial memory, brain autoimmunity or antibody response in aged F-44 rats. Neurobiol. Aging 12:47-53; 1991. 2. Carmack, M. A. Exercise induced modifications in immune responsiveness in rats. Unpublished doctoral dissertation, University of Oregon. Eugene, OR. 1984. 3. Dishman, R. K. Psychological eft~cts of exercise for disease resistance and health promotion. In: Eisinger, M.; Watson, R., eds. Exercise: Disease and cancer resistance. Boca Raton, FL.: CRC Press; 1992:
10. 11. 12.
1-48.
4. Fitzgerald, L. Exercise and the immune system, lmmun. Today 9: 337-339; 1988. 5. Hansen, M. B.; Nielsen, S. E.; Berg, K. Reexamination and further development of a precise and rapid dye method for measuring cell growth/cell kill. J. lmmunol. Methods 119:203-210; 1989. 6. Hoffman-Goetz, L.; Keir, R.; Thorne, R.; Houston, M. E.; Young, C. Chronic exercise stress in mice depresses splenic T-lymphocyte mitogenesis in vitro. Clin. Exp. lmmunol. 6:551-557; 1986. 7. Kiecolt-Glaser, J. K.; Glaser, R. Stress and immune function in humans. In: Ader, R., ed. Psychoneuroimmunology, 2nd ed. New York: Academic Press, Inc,; 1991:849-867. 8. Laudenslager, M. L.; Fleshner, M.; Hofstadter, P.; Held, P. E.; Sb mons, L.; Maier, S. F. Suppression specific antibody production by inescapable shock: Stability under varying conditions. Brain Behav. lmmunol. 2:92-101; 1988. 9. Levine, S.; Coe, C.; Wiener, S. G. Psychoneuroendocrinology of
13. 14. 15. 16. 17. 18.
stress: A psychobiological perspective. In: Psychoendocrinology. New York: Academic Press, Inc.; 1989:341-377. Liu, Y.; Wang, S. Y. The enhancing effect of exercise on the production of antibody to Salmonella typhi in mice. Immunol. Lett. 14:117-120; 1986/1987. Mahan, M. P.: Young, M. R. Immune parameters of untrained or exercise-trained rats after exhaustive exercise. J. Appl. Physiol. 66: 282-287; 1989. Niemann, D. C.; Nehlsen-Cannarella, S. L. The effects of acute and chronic exercise on immunoglobulins. Sports Med. 11:183-201; 1991. Roebuck, B. D.; McCaffrey, J.; Baumgartner, K. J. Protective effects of voluntary exercise during the postinitiation phase of pancreatic carcinogenesis in the rat. Cancer Res. 50:6811-6816; 1990. Simon, H. B. Exercise and human immune function. In: Ader, R., ed. Psychoneuroimmunology, 2nd ed. New York: Academic Press, Inc.; 1991:869-895. Simon, H. B. The immunology of exercise. JAMA 252:2735-2738; 1984. Steptoe, A.; Cox, S. Acute effects of aerobic exercise on mood. Health Psychol. 7:329-340; 1988. Tada, H.; Shiho, O.: Kuroshima, K.; Koyama, M.; Tsukamoto, K. An improved colorimetric assay for interleukin 2. J. Immunol. Methods 93:157-165; 1986. Voller, A.; Bidwell, D. E.: Bartlett, A. Microplate ELISA and its applications. In: Malvano, R., ed. Immunoenzymatic assay techniques. Boston: Martinus Nijhoff Publishers; 1980:105-132.