Rapid growth of the elderly population and the rising medical costs associated with their care make it important to gain a better understanding of immune function and regulation in the aged. Among the changes in immune function with aging, dysregulation in T lymphocyte function appears to be the most dramatic (14,19). Age-related thymic involution and a relative increase in the number of memory (CD45RO+) versus naive (CD45RA+) lymphocytes are thought to contribute to the changes in T cell function (19). Recent evidence also indicates that aging leads to macrophage and natural killer (NK) cell hyporesponsiveness (13,18,29,31). The dysregulation of immune function seen with the aging likely contributes to the increased risk of infection and cancer and poorer prognoses in the afflicted elderly (14,32).
The immune response to exercise has received increased attention since the documentation of elevated incidence rates of upper respiratory tract infection (URTI) after stressful exercise (23) and lower cancer incidence rates in physically active people (30). In young subjects, NK cell activity, various measures of T and B lymphocyte function, upper respiratory neutrophil (PMN) function, and salivary immunoglobulin A (IgA) levels have all been reported to be suppressed following intense or prolonged exercise (23). In contrast, moderately intense exercise may lead to immunopotentiation (3). Unfortunately, despite the fact that older people are being encouraged to adopt a physically active lifestyle and that the elderly exhibit increased incidence rates for URTI and other diseases, the effects of exercise on the senescent immune system has received little attention. Therefore, the purpose of this study was to determine the effects of an acute bout of maximal exercise on blood leukocyte mobilization and lymphocyte responsiveness to mitogenic stimulation in young and old sedentary subjects. We chose to study acute maximal exercise because of the high degree of cell trafficking that occurs in response to this type of exercise and the fact that stressful exercise is suspected to lead to a transient immunosuppression.
MATERIALS AND METHODS
The subjects consisted of sedentary elderly (N = 33, 65.3 ± 0.8 yr) and young (N = 14, 22.4 ± 0.7) volunteers recruited from the community. They were considered sedentary if they had not performed exercise of 15 min duration or longer more than two times per week for the previous 6 months. Subjects were excluded if they smoked, if they were taking any medications (i.e., aspirin, antiinflammatory drugs, antidepressants) known to affect immune function, or if they had any previous history of cancer, arthritis, or immune disorders. Body fat was measured by total body electrical conductivity (Model HA-2, EM-Scan, Inc., Springfield, IL). This study was approved by the Human Subjects Committee of the University of Illinois and was in accordance with the requirements of the American College of Sports Medicine for human experimentation.
Treadmill testing. The subjects performed a physician-supervised modified Balke treadmill test to volitional fatigue. The initial treadmill speeds were set at 2.5-3.0 and 3.5-4.0 mph for the old and young, respectively. This was done to ensure similar times to fatigue in both groups. The test incremented in 2 min stages with a 2% increase in grade at each stage. Measurements of oxygen uptake (V̇O2), heart rate (electrocardiogram), and blood pressure were continuously monitored. V̇O2 was measured from expired air samples taken at 30 s intervals until a peak V̇O2 (i.e., plateau in V̇O2 despite increasing workload) was attained, or until symptom limitation and/or volitional fatigue. An Applied Electrochemistry (Ametek, Pittsburgh, PA) metabolic measurement system was used for measurement of metabolic variables and V̇O2, V̇CO2, V̇E, and respiratory exchange ratio (RQ) was determined by computer software. Antecubital venous blood samples were collected pre-, post-, and 20 min postexercise.
Complete blood cell counts. Blood was collected in 3-mL Vacutainers containing sodium EDTA and analyzed on a Cell-Dyn 3500 automated cell analysis system. Measurements included quantification of total leukocytes, PMN, lymphocytes, monocytes, eosinophils, and basophils. In addition, hematocrit (Hct) and hemoglobin (Hb) values were assessed and used for correcting data for exercise-induced plasma volume shifts according to the method of Greenleaf (11).
Leukocyte subsets. Cells (100 μL of blood) were incubated with 1 μg of the following direct and indirect (2 and 3 color) combinations of monoclonal antibodies according to the manufacturer (Coulter Corp., Miami, FL): fluoroscein isothiocyanate (FITC)- anti CD3, phycoerythrin (PE)-anti CD14, FITC-anti CD45RA (naive), PE-anti CD45RO (memory), and biotin-anti CD4 and CD8. Streptavidin Cy-chrome was added as the second antibody for the detection of CD4 and CD8+ T cells. Isotypic control antibodies and autofluorescence was run with each sample. CD56+ NK cells and NK cytolytic activity were also analyzed in isolated peripheral blood mononuclear cells (PBMC) and results can be found in a previously published report (31). The samples were incubated on ice for 30-45 min and were washed twice with phosphate-buffered saline (PBS) with azide. Contaminating red blood cells were lysed with ammonium chloride and fixed in 4% paraformaldehyde for analysis. A minimum of 5000 cells were analyzed on a Coulter XL-MCL four-color analyzer (Coulter Corp., Miami, FL) and data analysis was performed using XL System II software. Lymphocyte analysis was performed by gating on lymphocytes based on forward and side light scatter. The CD3+ gate was set to minimize (0-2%) contaminating monocytes (CD14+ cells). Memory and naive lymphocyte analysis was performed on the CD4+ and CD8+ cell subsets.
Lymphocyte proliferation. We used a whole blood lymphocyte mitogenesis assay modified from Bloemena et al. (2). Blood was collected before, immediately after, and 20 min after exercise in 10-mL Vacutainer tubes containing preservative-free sodium heparin. The heparinized venous blood was diluted 1:10 with RPMI-1640 (Sigma, St. Louis, MO) supplemented with 5% heat inactivated fetal bovine serum (FBS, Sigma), penicillin (100 U·mL−1), streptomycin (100 μg·mL−1), glutamine (200 mM), and 2-mercaptoethanol (5 × 10−5 M) at room temperature. One hundred fifty μl of diluted blood was plated in triplicate wells of a 96-well round bottomed microtiter plate. The polyclonal T cell mitogens concanavalin A (Con A) and phytohemagglutinin (PHA) were diluted to final concentrations of 50 (Con A only), 20, 10, 5, 2.5, 1, and 0 μg·mL−1. The plates were incubated for 48 h at 37°C, 5% CO2 and then pulsed with 25 μL (1 μCi·well−1) of [3H]thymidine (6.7 Ci·mmol−1, ICN Biomedicals, Costa Mesa, CA). The plates underwent a second incubation for 16 h at which point they were harvested with a cell harvester (Cambridge Tech. Inc., Watertown, MA). Three milliliters of scintillation fluid was added to the filters and they were counted in a Packard 1600 TR scintillation counter. The values are expressed as mean ± SE for the triplicate wells and are presented as uncorrected (raw counts per minute) and corrected (i.e., cpm/number of CD3+ cells per well) values. The number of CD3+ T cells in the assay wells was determined by multiplying the percentage of CD3+ cells by the lymphocyte count.
Statistical analyses. Differences in descriptive data between the young and old were analyzed using a Student's t-test. Leukocyte counts, subsets, and mitogenic responses were analyzed using a 2 (age) × 3 (time) repeated measures ANOVA with SigmaStat (Jandel Scientific, San Rafael, CA) software. In instances where assumptions of normality or equal variance were violated, a Geisser-Greenhouse F test was used to determine significance. This analysis is a very conservative approach for testing differences in a repeated measures design. Student Newman-Keuls multiple comparisons procedures were used to compare differences between means. Statistical significance was set at P < 0.05. Upon separate analysis of the data, we found no gender related differences among any of the dependent variables measured in this study. Therefore, to simplify interpretation we grouped both males and females together in the final data analysis representing the tables and figures.
Descriptive data. Body weight and percentage body fat were higher in the old when compared with the young subjects (Table 1). V̇O2max, expressed in L·min−1 or in mL·Kg−1·min−1, was significantly lower in the old subjects compared with the young (Table 1). Despite this, both groups attained age-predicted maximal heart rates and achieved RQ values >1.0 (Table 1) signifying that both groups were performing at high exercise intensities. In addition, there were no apparent differences in percentage increases in V̇O2 in the latter stages of the exercise tests between the groups (data not shown). Time to fatigue was not significantly different between the groups (Table 1); therefore any effects seen were most likely a result of the relative exercise intensity and not the duration of the exercise test.
Complete blood cell counts.Table 2 contains leukocyte counts corrected for exercise-induced changes in plasma volume. At rest, elderly subjects demonstrated increased absolute numbers of blood monocytes and basophils. In response to maximal exercise, plasma volume decreased about 7% in both groups; however, by 20 min postexercise the aged still manifested a significant plasma volume shift. Total leukocytes increased significantly in both the elderly and the young subjects, the magnitude of which was greater (44% vs 30%) in the young (Table 2). The leukocytosis was transient in the young but remained elevated in the elderly at 20 min postexercise. The exercise-induced leukocytosis included increases in PMN (23% and 16%, in the young and old, respectively), lymphocytes (73% and 56%), monocytes (65% and 22%), eosinophils (22% and 6%), and basophils (67% and 40%). PMN and lymphocytes remained elevated in the old, but not the young, 20 min postexercise.
Leukocyte subsets. The absolute numbers of leukocyte subsets corrected for postexercise-induced changes in plasma volume can be found in Table 3. Comparison of the pre-exercise values revealed that the elderly subjects had significantly higher numbers of CD3+, CD8+, and CD4+, and CD8+ memory (i.e., CD45RO+) cells. In contrast, the number of naive (i.e., CD45RA+) CD4+ and CD8+ cells and double positive (CD45RA+/CD45RO+) CD4+ and CD8+ cells were lower in the old when compared with those in the young.
In the immediate postexercise samples, CD3+ cell number increased to a similar extent (∼50%) in the young and old. This increase was reflective of an increase in the absolute numbers of both CD4+ (57% and 22% increase, in the young and old, respectively) and CD8+ (153% and 112% increase, in the young and old, respectively) cell subsets. CD8+ cell number remained elevated at 20 min postexercise in the old but not the young. The relatively larger increase in CD8+ versus CD4+ cells resulted in a lower CD4/CD8 ratios immediately postexercise, which persisted at 20 min of recovery in the old subjects.
Exercise resulted in a significant increase in CD4+ and CD8+ naive and memory cells in both the young and old. Regardless of age, the increase in CD8+ naive and memory cells was greater than that of the CD4+ subset. In the young, CD4+/CD45RA+ and CD8+/CD45RA+ cells rose 53% and 188%, respectively, while in the old the increase was not as large (22% and 157%, for CD4+/CD45RA+ and CD8+/CD45RA+, respectively). CD4+/CD45RO+ and CD8+/CD45RO+ cells increased by 47% and 155% in the young, respectively, and 28% and 85% in the elderly when measured immediately postexercise. Therefore, when expressed as a percent of the pre-exercise value, the exercise-induced increase in T cells bearing CD45RA or CD45RO was greater in the young when compared with that in the old. At 20 min of recovery, the elderly still had a significantly higher number of CD8+/CD45RO+ cells in the blood when compared with those in pre-exercise. Transitional cells bearing both CD45RA and CD45RO also demonstrated this same response, except that the increase in CD8+ double positive cells was statistically significant when compared with CD4+ double positive cells.
Figure 1 compares the composition of the exercise-induced CD4+ and CD8+ cellular influx into blood of young and old subjects immediately postexercise. The old recruited fewer CD4+ and more CD8+ cells to the blood in response to exercise. The composition of the recruited cells differs in that approximately equal portions of memory and naive CD8+ cells are recruited regardless of age, whereas the old recruited significantly more CD4+ memory cells and fewer naive CD4+ cells when compared with young subjects.
Lymphocyte proliferation. Whole blood proliferative responses to a wide range of Con A and PHA doses, expressed in counts per minute uncorrected for changes in CD3+ T cells, can be found in Tables 4 and 5. The common age-related decrease in mitogenic response was evident in unstimulated and stimulated (i.e., Con A doses of 0 and 50 μg·mL−1 and PHA doses of 0, 10, and 20 μg·mL−1) cultures. The lack of an age-related decrease in responsiveness across all doses is likely because of the large interindividual variation in this measure and the fact that the elderly had significantly elevated numbers of CD3+ cells when compared with the young subjects (Table 3). In response to maximal exercise, there was a tendency toward increased proliferation immediately postexercise and a reduced proliferation at 20 min postexercise (Tables 4 and 5). This was more apparent in the elderly.
Because the elderly had significantly higher numbers of T cells and exercise altered T cell numbers, we also expressed proliferation on a per cell basis by dividing radioactive counts by the number of CD3+ cells per culture well. These adjusted data can be found in Figure 2. PHA gave similar responses (data not shown). The elderly had significantly reduced proliferative responses when compared with the young. Immediately after maximal exercise young subjects had a significant reduction in proliferation on a per CD3+ cell basis at 5, 10, 20, and 50 μg·mL−1 doses of Con A, whereas the elderly subjects failed to demonstrate this effect.
In this report we compared the effects of maximal exercise on leukocyte counts and mitogenic responses in young and old subjects. We found, in general, that while the elderly responded similarly to the young with an exercise-induced leukocytosis, the magnitude of the increase was lower. In addition, the elderly were less resilient, in that many leukocyte subsets remained elevated above pre-exercise levels when measured 20 min postexercise, whereas leukocyte counts quickly returned to baseline levels by 20 min postexercise in the young. Another important finding was that maximal exercise reduced proliferation on a per CD3+ cell basis when measured immediately postexercise in the young but not in the old.
While there have been many acute exercise studies performed in young subjects, there have been few studies regarding the effects of a single bout of exercise on immune function in the elderly. Several studies have documented that acute exercise increases NK cell number and activity in the elderly to a similar extent when compared with that in young subjects (5,8,31); however, none of these studies examined T lymphocyte function which is known to be highly age sensitive. Mazzeo et al. (16) compared leukocyte counts and T lymphocyte subsets before and immediately after a moderately intense (∼50% peak work capacity) 20 min exercise bout in the young and old (16). Old subjects manifested a 33% increase in total leukocytes (vs 15% for young) and a 44% increase in total lymphocytes (vs 24% in young). This exercise resulted in similar increases in CD3+, CD56+, CD4+, and CD8+ lymphocytes in the young and old. Unfortunately, it is difficult to directly compare our data with that of Mazzeo because of the different intensities and durations of the exercise bouts employed. We found that in response to maximal exercise the elderly had an attenuated exercise-induced leukocytosis when compared with the young. Total leukocytes increased 44% versus 30%, PMN increased 23% versus 16%, and lymphocytes increased 73% versus 56% in the young when compared with the old immediately postexercise. The percentage increase (when expressed relative to pre-exercise values) in CD3+, CD4+, CD8+, and CD4+ and CD8+ memory and naive cells was also higher in the young versus the old. Our data are similar to that provided by Cannon et al. (4) who found that the elderly had an attenuated increase in PMN in response to 45 min of eccentric exercise at 78% maximal heart rate (4). However, this attenuation occurred at 4 and 8 h postexercise and not immediately postexercise. Based on these limited number of studies, it appears, not surprisingly, that the immune response to exercise may differ in young and old subjects depending on exercise intensity and duration.
We also demonstrate, as others have (7,19), that aging results in differential expression of T lymphocyte subsets, such that the elderly express significantly more CD45RO+ memory T lymphocytes and fewer CD45RA+ naive and transitional (i.e., CD45RA+/CD45RO+) T lymphocytes (Table 3). Some believe this may be a factor in the reduction in T cell mediated immunity in the elderly (19). In response to maximal exercise both young and old recruited naive and memory CD4+ and CD8+ cells to the blood (Fig. 1). We found that the young and old recruited approximately equal numbers and percentages of CD8+ naive and memory cells to the blood. In contrast, the aged recruited significantly fewer numbers of CD4+ naive and transitional (CD45RA+RO+) cells and a higher percentage of CD4+/CD45RO+ cells. In young subjects, Gabriel et al. (10) demonstrated a significant increase in CD8+/CD45RO+ cells and a nonsignificant increase in CD4+/CD45RO+ cells in response to 1 min of maximal exercise. In their study they did not stain for CD45RA. Therefore, it appears that exercise results in a differential mobilization of memory and naive T lymphocytes dependent on CD4 or CD8 expression and age.
The mechanisms responsible for the different exercise-induced responses between young and old were not directly examined in this study. However, the lack of resilience in the leukocyte response 20 min postexercise in the elderly is analogous to the documented lack of resilience in blood stress hormone concentrations in the elderly in response to a wide variety of stressors (26). Indeed, the levels of glucocorticoids and catecholamines stay elevated for a prolonged period after hypothalamic-pituitary-adrenal (HPA) axis activation in the old, perhaps because of a loss of negative feedback regulation (24). Moreover, some evidence suggests that the elderly demonstrate a greater stress hormone response to a given stressor such as an exercise bout (16,27). Elevated stress hormone levels or sustained increases could lead to differential responses in the elderly when compared with those in the young. Indeed, epinephrine has been postulated to contribute to the exercise-induced leukocytosis by binding to marginating leukocytes and reducing their interactions with vascular endothelial cells (1,20) and glucocorticoids are thought to contribute to exercise-induced leukocytosis by promoting PMN release from bone marrow (17).
There is a large literature on the effects of acute (single bout) exercise on in vitro T lymphocyte mitogenesis in response to mitogens in young subjects. Human isolated PBMC or whole blood mitogenic responsiveness transiently decreases 30-60% following (up to 4 h post) acute moderate or intense exercise (9,15,22,25). This suppressant effect is greater following intense exercise when compared with moderate exercise (15), and some have shown no suppression resulting from moderate exercise (21). Our data are in agreement with the majority of studies in young subjects that have demonstrated that intense exercise decreases lymphocyte proliferation. However, in our data, this was only apparent after we corrected for the number of CD3+ cells in the culture wells and not if we expressed data as raw counts per minute. Aged subjects in this study did not demonstrate reduced per cell proliferation in response to acute maximal exercise. In the only other human study to examine lymphocyte proliferation in the elderly after an acute exercise bout, Mazzeo et al. (16) found that young subjects had a significantly higher PHA proliferative response immediately postexercise, whereas the old subjects demonstrated a small nonsignificant increase (16). These data differ from ours, but it should be remembered that exercise intensity was different between the two studies and they did not express their data relative to the number of responsive cells in the assay wells. In partial support of our findings, De la Fuente et al. (6) found that the PHA response of lymphocytes obtained from young and old mice was reduced after a single bout of exhaustive swimming (6). However, the suppression in the elderly mice (∼26%) was lower than that observed for the young mice (∼62%).
The exercise-induced changes in lymphocyte proliferation are thought to result largely from exercise-induced changes in the proportions or numbers of responsive cells (i.e., CD3+, CD4+, CD8+, naive, memory) in peripheral blood samples (12,25). Exercise increases NK cell number and reduces the CD4/CD8 ratio, both of which may contribute to the decline in mitogenic responsiveness by altering the composition of responsive cells. However, intense exercise may also suppress T lymphocyte mitogenesis independently of cell number and, while the mechanisms are unknown, it is believed that exercise-induced increases in immunosuppressive hormones like glucocorticoids and epinephrine may be responsible (22). Others argue that postexercise monocyte production of PGE2 depresses T cell responsiveness to mitogens (28). In this study we found that proliferation was reduced on a per CD3+ cell basis immediately after exercise in the young; arguing against the hypothesis that changes in CD3+ cell numbers account for the exercise-induced decrease in mitogenic responsiveness. Upon analysis, we failed to correlate any cell subset change with the change in proliferation in the young postexercise.
In summary, we have found differential leukocyte subset and mitogenic responses to acute maximal exercise between young and old subjects. The mechanisms responsible for these differences are unknown, but stress hormones are likely to be involved. The clinical significance of the exercise-induced changes in T lymphocyte responsiveness in this study are also unknown. However, the transient nature of the response make it unlikely that changes such as this would lead to clinical illness. Clearly, more studies need to be performed to describe better and understand mechanistically how exercise affects immune function in the elderly.
1. Benschop, R. J., F. G. Oosterveen, C. J. Heijnen, and R. E. Ballieux. β2-adrenergic stimulation causes detachment of natural killer cells from cultured endothelium. Eur. J. Immunol.
2. Bloemena, E., M. T. L. Roos, J. L. A. M. Van Heijst, J. M. J. J. Vossen, and P. T. A. Schellekens. Whole-blood lymphocyte cultures. J. Immunol. Methods
3. Brines, R., L. Hoffman-Goetz, and B. K. Pedersen. Can you exercise to make your immune system fitter? Immunol. Today
4. Cannon, J. G., M. A. Fiatarone, R. A. Fielding, and W. J. Evans. Aging and stress-induced changes in complement activation and neutrophil mobilization. J. Appl. Physiol.
5. Crist, D. M., L. T. Mackinnon, R. F. Thompson, H. A. Atterbom, and P. A. Egan. Physical exercise increases natural killer cellular-mediated tumor cytotoxicity in elderly women. Gerontol.
6. De La Fuente, M., M. D. Ferrandez, J. Miquel, and A. Hernanz. Changes with aging and physical exercise in ascorbic acid content and proliferative response of murine lymphocytes. Mech. Ageing Dev.
7. Ernst, D. N., O. Weigle, and M. V. Hobbs. Aging and lymphokine gene expression by T cell subsets. Nutr. Rev.
8. Fiatarone, M. A., J. E. Morley, E. T. Bloom, D. Benton, G. F. Solomon, and T. Mackinodan. The effects of exercise on natural killer cell activity in young and old subjects. J. Gerontol.
9. Field, C. J., R. Gougeon, and E. B. Marliss. Circulating mononuclear cell numbers and function during intense exercise and recovery. J. Appl. Physiol.
10. Gabriel, H., A. Urhausen, and W. Kindermann. Mobilization of circulating leucocyte and lymphocyte subpopulations during and after short, anaerobic exercise. Eur. J. Appl. Physiol.
11. Greenleaf, J. Plasma volume shifts during stress in man: osmolality and red cell volume. J. Appl. Physiol.
12. Hinton, J. R., D. G. Rowbottom, D. Keast, and A. R. Morton. Acute intensive interval training and in vitro
T-lymphocyte function. Int. J. Sport Med.
13. Kutza, J. and D. M. Murasko. Effects of aging on natural killer cell activity and activation by interleukin-2 and IFN-α. Cell. Immunol.
14. Mackinodan, T. and K. Hirokawa. Normal aging of the immune system. In: Relations Between Normal Aging and Disease.
H. A. Johnson (Ed.). New York: Raven Press, 1985, pp. 117-132.
15. MacNeil, B., L. Hoffman-Goetz, A. Kendall, et al. Lymphocyte proliferation responses after exercise in men: fitness, intensity, and duration. J. Appl. Physiol.
16. Mazzeo, R. S., C. Rajkumar, J. Rolland, B. Blaher, G. Jennings, and M. Esler. Immune response to a single bout of exercise in young and elderly subjects. Mech. Ageing Dev.
17. McCarthy, D. A. and M. M. Dale. The leukocytosis of exercise: a review and model. Sports Med.
18. McLachlan, J. A., C. D. Serkin, K. M. Morrey, and O. Bakouche. Antitumoral properties of aged human monocytes. J. Immunol.
19. Miller, R. A. Aging and immune function: Cellular and biochemical analyses. Expt. Gerontol.
20. Murray, D. R., M. Irwin, C. A. Rearden, M. Ziegler, H. Motulsky, and A. S. Maisel. Sympathetic and immune interactions during dynamic exercise: mediation via β2-adrenergic dependent mechanism. Circulation
21. Nehlsen-Cannerella, S. L., D. C. Nieman, J. Jessen, et al. The effects of acute moderate exercise on lymphocyte function and serum immunoglobulin levels. Int. J. Sports Med.
22. Nieman, D. C., A. R. Miller, D. A. Henson, et al. Effect of high-versus moderate-intensity exercise on lymphocyte subpopulations and proliferative response. Int. J. Sports Med.
23. Nieman, D. C. Immune response to heavy exertion. J. Appl. Physiol.
24. Sapolsky, R. M., L. C. Krey, and B. S. McEwen. The neuroendocrinology of stress and aging: the glucocorticoid cascade hypothesis. Endocr. Rev.
25. Shinkai, S., S. Shore, P. N. Shek, and R. J. Shephard. Acute exercise and immune function: relationship between lymphocyte activity and changes in subset counts. Int. J. Sports Med.
26. Seeman, T. E. and R. J. Robbins. Aging and hypothalamic-pituitary-adrenal response to challenge in humans. Endocr. Rev.
27. Silverman, H. G. and R. S. Mazzeo. Hormonal responses tomaximal and submaximal exercise in trained and untrained men of various ages. J. Gerontol. Biol. Sci.
28. Smith, J. D., S. Chi, G. Salazar, et al. Effect of moderate exercise on proliferative responses of peripheral blood mononuclear cells. J. Sport Med. Phys. Fitness
29. Wallace, P. K., T. K. Eisenstein, J. J. Meissler, Jr., and P. S. Morahan. Decreases in macrophage mediated antitumor activity with aging. Mech. Ageing Dev.
30. Woods, J. A. Exercise and resistance to neoplasia. Can. J. Physiol. Pharmacol.
31. Woods, J. A., J. K. Evans, B. W. Wolters, M. A. Ceddia, and E. McAuley. Effects of maximal exercise on natural killer cell function and responsiveness to interferon-α in the young and old. J. Gerontol. Biol Sci.
32. Yoshikawa, T. T. Geriatric infectious diseases: an emerging problem. J. Am. Geriat. Soc.