European Journal of Applied Physiology

, Volume 103, Issue 1, pp 79–87

Aerobic training increases the stimulated percentage of CD4+CD25+ in older men but not older women


    • Institute of Food, Nutrition and Human HealthMassey University
  • Gregory Gass
    • Faculty of Health Science and MedicineBond University
Original Article

DOI: 10.1007/s00421-007-0664-8

Cite this article as:
Broadbent, S. & Gass, G. Eur J Appl Physiol (2008) 103: 79. doi:10.1007/s00421-007-0664-8


The purpose of the present study was to determine whether 12 months of moderate intensity cycling would increase the expression of IL-2 (CD25+) receptors in T helper (CD4+) lymphocytes in men and women aged 65–75 years. Fourteen men and 10 women completed 52 weeks of moderate intensity cycling (60% VO2peak). Subjects trained (TR) three times per week for 45 min per session. Eight age-matched untrained (UT) male and eight UT female subjects acted as controls. Resting blood samples were taken from TR and UT subjects every 4 weeks. Leukocyte concentration was measured using a full blood count. PHA-stimulated CD4+ lymphocytes were analysed for changes in the expression of CD25+, by flow cytometry. Training significantly increased VO2peak (l min−1, ml kg−1 min−1) in male (+14.3, +16%) and female (+16.7, +27.8%) groups. The TR male group showed a significantly lower percentage of CD4+CD25+ than the male UT in January but the TR male percentage was significantly higher than the UT male group during February, March, April, May, June, September B and December. The female TR group showed a significantly higher percentage CD4+CD25+ than the female UT only during July. There were also significant sequential monthly changes in the percentage of CD4+CD25+ for male and female UT and TR groups. Significant increases in the percentage of CD4+CD25+ in the male TR group suggest training-enhanced lymphocyte mitogenic responsiveness. Moderate intensity long-term training may increase the recruitment of active memory CD4+CD25+ in men rather than women.


IL-2IL-2RLymphocyte responseModerate intensity training


It is well established that many immune responses decline with the ageing process, with a corresponding increase in the incidence of infections and cancers in older individuals (Bruunsgaard and Pedersen 2000; Song et al. 1993). The decline in functional adaptive immune cells such as CD4+ T lymphocytes has been attributed to thymic involution and an increase in memory CD4+ lymphocytes (CD45RO+) at the expense of naïve lymphocytes (CD45RA+) (Ceddia et al. 1999; Shinkai et al. 1995; Song et al. 1993). The CD4+ T lymphocyte response may be further compromised by decreased production of the cytokine Interleukin-2 (IL-2), which stimulates lymphocyte proliferation, and reduced expression of IL-2 receptors (IL-2R) (Neckers and Cossman 1983; Whisler et al. 1996). While there is some evidence to suggest that regular, moderate intensity exercise may increase the number and/or mitogenic responsiveness of CD4+ lymphocytes (Crist et al. 1989; Shinkai et al. 1995; Smith et al. 2004) the effects of long-term (52 weeks) moderate intensity training on lymphocyte activation, IL-2R expression and CD4+ proliferation are largely unreported.

CD4+ lymphocyte activation is characterised by the binding of a mitogen or antigen to the T cell receptor (TCR/CD3+), followed by the release of IL-2 and the expression of the IL-2R (high affinity α sub-unit CD25, β sub-unit CD122 and γ sub-unit) via intracellular signalling pathways (Neckers and Cossman 1983; Rhind et al. 1994). The IL-2R is expressed once antigen/mitogen has bound to the CD3+ receptor and IL-2 has been released from the CD4+ lymphocyte. Therefore the expression of the IL-2R, especially the high-affinity CD25, indicates the ability of the lymphocyte to respond the antigenic/mitogenic challenge. IL-2 activates natural killer (NK) cells, cytotoxic CD8+ lymphocytes and antibody-producing B lymphocytes, and decreased CD4+ lymphocyte activation and CD25 expression may impair the response of CD8+, NK and B lymphocytes to infectious agents (Whisler et al. 1996). Some research has suggested that moderate-intensity exercise may increase CD25 expression and CD4+ proliferation, thus enhancing the CD4+ lymphocyte antigenic response and decreasing the risk of infection (Ahluwalia et al. 2001; Bruunsgaard and Pedersen 2000). Our aim was to test the experimental hypothesis that 12 months of moderate intensity training in older men and women would result in a significantly increased percentage of CD4+CD25+ and increased CD25 density in both genders.



The study was 52 weeks in duration (5 January 2001 to 1 January 2002) and involved 2 groups of men (TR, N = 14; UT, N = 8) and women (TR, N = 10; UT, N = 8), aged 65–75 years (mean age ± SD, 68.5 ± 4.4 years). Volunteers were required to have medical clearance to exercise from their medical practitioner, and were randomly allocated to either the TR or UT groups. Volunteers were not undertaking regular exercise, and those diagnosed with cardiovascular disease, glucose intolerance, diabetes, asthma, orthopaedic conditions, currently taking anti-inflammatory medications, currently smoking, or with any other medical condition which contraindicated vigorous exercise, were excluded from participation. The UT subjects undertook no aerobic or resistance training, or recreational sports for the duration of the study. All the subjects were volunteers recruited from the local area. After completing a PAR-Q and full medical history questionnaire, all subjects underwent a resting 12-lead ECG and detailed medical examination. Body mass was determined using electronic scales to the nearest 0.05 kg. Standing height (Holtain stadiometer) was determined to the nearest 10 mm. After a period of quiet rest and with the subject in the supine position, blood pressure was measured by standard auscultation. Griffith University Human Ethics Committee approved the study and written consent was obtained from all subjects before participation.

Incremental graded exercise test to exhaustion

The subjects were familiarised with the Lode cycle ergometer (Lode Excalibur, Groningen, Holland) and other equipment to be used in the incremental and steady state exercise tests. One week later, the subjects visited the laboratory for the incremental exercise test to exhaustion on the Lode cycle ergometer under direct medical supervision. Incremental exercise tests to exhaustion were undertaken at the beginning (Pre-Test, 0 month) and at the end of the study (Post-Test, 12 months), between the hours of 0600 and 0900. Each male and female subject commenced cycling at 15 W, and at zero W, respectively. After a 3 min warm up period, the workload was increased by 15 and 10 W per min for the males and the females respectively. The workload increased until exhaustion or clinical signs and symptoms prevented further exercise. Peak VO2 (VO2peak), VCO2, ventilation (VE BTPS l min−1), ventilatory equivalents (VE/VO2, VE/VCO2) and respiratory exchange ratio (RER) were measured using open circuit spirometry (Exerstress, Sydney, Australia), and recorded every 30 s, with final values determined from the average of the two highest values attained over two collection periods during the exercise test. Throughout each incremental exercise test, heart rate and rhythm were monitored from bipolar leads in the CM5 position (Lohmeier ECG, Munich, Germany). Blood pressure was monitored using a standard cuff size and mercury manometer (Baum & Co., New York, USA). Power (W) was recorded every 30 s (Lode Excalibur, Groningen, Holland). Peak VO2 was determined to have been reached when (1) VO2 increased by ≤0.15 l min−1 despite an increase in power, (2) heart rate was within ten beats of age-predicted maximum, and (3) when RER exceeded 1.15.

Steady state cycle tests

Two days after the incremental exercise test, the participants in the TR groups performed two steady state cycle exercise tests (Monark, Vansbro, Sweden) of 20 min duration to determine the power output and heart rates corresponding to exercise intensities of 50 and 60% of VO2peak. Oxygen uptake and heart rate were measured for at rest, and for 1 min during 5 min intervals when each subject was cycling at 50% VO2peak. Heart rate was monitored continuously with bipolar leads placed in the CM5 configuration (Lohmeier, Munich, Germany), and heart rate and rhythm were recorded during the last 10 s of every 5 min period. The subjects underwent a 15 min rest period, and then the steady state exercise test was repeated with the cycle ergometer resistance set at a load corresponding to 60% VO2peak. The power corresponding to 50% VO2peak provided the initial power and heart rate for the training programme. The power was gradually increased to 60% VO2peak after the first month of training. As steady state heart rates decreased during the study, increments in workload (W) were made to maintain the subjects’ training intensity at a heart rate corresponding to 60% VO2peak.

Blood samples

Fasting venous blood samples were taken every 4 weeks from January 2001–2002. Two blood tests occurred in September, 4 weeks apart, referred to as September A and B. After 10 min of rest in the supine position, a venous blood sample (15 ml) was taken from an antecubital vein using sterile venipuncture procedures (Becton Dickinson, Sydney, Australia). Subjects (TR, UT) were requested to avoid strenuous exercise 24 h prior to the venipuncture. Venous blood was analysed for leukocyte concentration (WBC) with a full cell count, using a blood analyser (Coulter T66, Miami, USA). Unstimulated CD4+ lymphocytes were separated using a density gradient, stimulated with phytohemagglutinin (PHA) and incubated for 48 h. After incubation CD25 was measured with flow cytometry.

Cell Separation

Venous blood samples for each subject were layered into two sterile 5 ml centrifuge tubes (Sarstedt, Adelaide, Australia). Mononuclear cells were separated using Ficoll Histopaque 1077 (Sigma Aldrich, Sydney, Australia), according to the manufacturer’s instructions, and leukocytes were suspended in 1 ml of phosphate buffered saline (PBS) (Sigma Aldrich, Sydney, Australia). CD4+ lymphocytes were then separated from leukocytes using a CD4 positive isolation kit (Dynal, Melbourne, Australia). A measure of 10 μl of CD4 positive isolation beads were added to a test tube containing 1 ml of lymphocyte suspension and rotated gently for 20 min at 2–5°C. The test tube was placed on a magnetic stand for 3 min, allowing the CD4 positive isolation beads to adhere to the test tube walls. The supernatant was removed and 1 ml of PBS was added to the tube. After washing, the PBS was removed and 1 ml of RPMI-1640 containing l-glutamine (R8758, Sigma Aldrich, Sydney, Australia) was added, followed by 5 μl of CD4 detach beads. The test tubes were incubated for 60 min at room temperature, with gentle rotation, and then each tube was placed in the magnetic stand for 3 min. The supernatant now carried the released CD4 lymphocytes, and was pipetted into a fresh tube. CD4+ lymphocytes were washed in PBS and centrifuged at 300×g at room temperature. Lymphocyte purity was >96%, measured by flow cytometry in the forward and side scatter modes, as demonstrated by Mooren et al. (2002). Lymphocyte viability was 98 ± 2% as demonstrated by the Calcein cell viability test (Molecular Probes, USA). The lymphocyte pellet was resuspended in 1 ml of PBS with 5 mM glucose, and 10% fetal calf serum (FCS) (C8056, Sigma Aldrich, Sydney, Australia).

Mitogen stimulation

The lymphocyte concentration (0.5 × 106 ml) was placed in a 5 ml sterile culture tube (Quantum Scientific, Brisbane, Australia) with 10 μl of 10% PHA (Sigma Aldrich, Sydney, Australia) and PBS, 10 μl of 10% Gentamicin sulphate solution (Sigma Aldrich, Sydney, Australia) and sufficient RPMI-1640 with l-glutamine (Sigma Aldrich, Sydney, Australia) supplemented with 20% FCS to make a final volume of 1 ml. The procedure was carried out in duplicate for each subject. Culture tubes were incubated for 48 h at 37°C with 5% CO2. The lymphocytes were then washed twice with ice-cold PBS with 5 mM glucose (0.5 ml) and resuspended in RPMI-1640 with l-glutamine (0.5 ml).

Flow cytometric calibration

The flow cytometer (Facscalibur, Becton Dickinson, Brisbane, Australia) was calibrated using “fluorospheres”. The data analysis region was gated around lymphocytes, with controls analysed first for every sample in order to set positive and negative regions (quadrants) for fluorescence. The positive region was set so that less than 2% of unstimulated cells registered as positive. Receptor density was measured with mean channel log fluorescence, with 5,000 cells analysed per sample. Colour compensation settings had been previously established using a preparation of mAb-stained whole blood. B cells and T cells registered as green and red fluorescent regions, respectively. Any overlap of the emission spectra of the dyes was corrected. Only quadrants 1 and 4 showed positive for red and green fluorescence.

Flow cytometric analysis

The lymphocyte samples were divided into two aliquots in separate 5 ml culture tubes (Quantum Scientific, Brisbane, Australia). Half the samples had 5 μl of fluorescent labelled monoclonal antibodies (mAb) added to the cell suspension. CD25 expression was measured using Phycoerythrin (PE)-conjugated monoclonal mouse anti-human interleukin-2 receptor (CD25+) mAb (Dako, Brisbane, Australia). The remaining 50% of samples had PE-conjugated isotype negative control mouse IgG1 mAb (5 μl) (Dako, Brisbane, Australia) added to each tube. Samples were incubated on ice and in the dark for 30 min. The lymphocytes were then washed in PBS and 5 mM glucose (0.5 ml) and centrifuged at 300×g at 4°C for 5 min. The supernatant was aspirated and the lymphocytes fixed in 1% paraformaldehyde (0.5 ml). Samples were stored on ice in the dark at 4°C prior to flow cytometric analysis, which was within 30 min of fixing for all samples. For purposes of quality control, the reagents and monoclonal antibodies used during the study were from the same batch.

Statistical analysis

Data was analysed with the SPSS Version 12-Windows 98 package (SPSS Inc, Chicago, USA). Values are reported as mean ± SD. Data was distributed normally and therefore analysed using a two-factor analysis of variance (ANOVA) with repeated measures and Bonferroni adjustment to detect significant differences between TR and UT male and female groups at each sampling point (between-group comparison), and between sequential monthly sampling points for TR and UT male and female groups for each variable (time effect). Variables measured were leukocyte concentration, percentage of CD4+ positive for CD25 and CD25 density. Significant differences between groups in pre- and post-study measurements (incremental exercise test variables) were detected using an independent T-test. The P value for significance was set at <0.05.


Subject characteristics and incremental exercise test

Fourteen TR men and ten TR women, and eight UT men and women, started and completed the study. Pre- and post-study characteristics of TR and UT subjects are presented in Table 1. The female TR group had significantly higher pre-training body mass and BMI than the female UT (P = 0.032, 0.031, respectively). The pre-training systolic blood pressure was significantly higher for TR females compared to TR males (P = 0.03). There were no significant post-training differences in either systolic or diastolic blood pressure for the four groups.
Table 1

Pre- and post-study characteristics of male and female untrained (UT) and trained (TR) subjects


UT male N = 8

TR male N = 14

UT female N = 8

TR female N = 10

Age (years)

68.8 ± 4.0

69.2 ± 5.2

68.8 ± 4.4

67.0 ± 4.0

Pre-study body mass (kg)

81.8 ± 11.7

85.8 ± 12.3

58.7 ± 10.2c

66.9 ± 10.7a,c

Post-study body mass (kg)

83.2 ± 12.3

83.5 ± 12.4

58.6 ± 10.2c

64.9 ± 10.4c

Pre-study BMI (kg m2)

25.8 ± 2.9

26.2 ± 2.9

22.3 ± 2.3c

23.3 ± 3.1a

Post-study BMI (kg m2)

26.8 ± 3.8

25.9 ± 3.0

22.4 ± 2.3c

22.4 ± 2.6c

Height (m)

1.78 ± 0.10

1.79 ± 0.12

1.67 ± 0.10

1.70 ± 0.12

Pre-study systolic blood pressure (mmHg)

126 ± 7

124 ± 6

130 ± 6

133 ± 7b

Pre-study diastolic blood pressure (mmHg)

88 ± 4

86 ± 5

88 ± 5

86 ± 4

Post-study systolic blood pressure (mmHg)

128 ± 7

121 ± 6

132 ± 6

129 ± 6

Post-study diastolic blood pressure (mmHg)

88 ± 5

82 ± 5

88 ± 5

82 ± 4

Data presented as mean ± SD, P < 0.05

aTR female group was significantly different to UT female group

bTR female group was significantly different to TR male group

cUT and TR female group significantly different to TR and UT male groups

Both TR groups significantly increased their VO2peak (P < 0.001), VEpeak (P < 0.05) and peak power (P < 0.001) during the study. The male and female TR group increased VO2peak by 14.3 and 16.7%, respectively. The results for the pre- and post-training incremental exercise tests are presented in Table 2. Peak power increased significantly for both male and female TR groups during the study. Peak heart rate was significantly higher post-training for the TR male group compared to their pre-training mean peak heart rate. Peak VE increased significantly with training for male and female groups, but the UT female group had a significantly lower VEpeak post-training compared to pre-training.
Table 2

Pre- and post-study incremental exercise test results for untrained (UT) and trained (TR) groups


UT male

TR male

UT female

TR female









VO2peak (l min−1)

2.0 ± 0.2

1.9 ± 0.2

2.1 ± 0.4

2.4 ± 0.3a,b

1.2 ± 0.1

1.04 ± 0.2a

1.2 ± 0.2

1.4 ± 0.2a,b

VO2peak (ml kg−1 min−1)

24.8 ± 4.6

24.4 ± 4.3

24.6 ± 4.2

29.1 ± 4.3a,b

20.1 ± 2.4

18.4 ± 3.1

18.1 ± 2.5

23 ± 3.1a,b

Powerpeak (W)

156 ± 14

154 ± 12

160 ± 12

186 ± 12a,b

86 ± 7

83 ± 6a

89 ± 9

107 ± 9a,b

Peak heart rate (b min−1)

155 ± 9

154 ± 10

149 ± 10

152 ± 9a

153 ± 7

149 ± 12

147 ± 10

157 ± 10

VEpeak BTPS (l min−1)

85 ± 11

84 ± 10

84 ± 10

99 ± 9a

51 ± 8

46 ± 8a

52 ± 9

58 ± 10a,b


1.19 ± 0.04

1.19 ± 0.03

1.17 ± 0.04

1.17 ± 0.03

1.18 ± 0.04

1.16 ± 0.03

1.19 ± 0.04

1.17 ± 0.03

Data presented as mean ± SD

UT Untrained, TR trained

aPost-study value significantly different from pre-study value, P < 0.05

bTR group significantly different from UT group, P < 0.05

Leukocyte concentration

The concentration of leukocytes remained within the normal range (4–11 × 109 l−1, Coulter, Miami, USA) for all groups during the study. The leukocyte concentrations for male and female groups during 52 weeks are presented in Fig. 1. The male UT group had a significantly higher leukocyte concentration than the male TR and female TR groups during May, June and November, and was significantly higher compared to the female UT group in April, May, September B, October and November.
Fig. 1

Monthly changes in leukocyte concentration (109 l−1) in untrained and trained male and female subjects. * Leukocyte concentration male UT group was significantly different to male TR group. # Leukocyte concentration male UT group was significantly different to female UT group. § Leukocyte concentration male UT group was significantly different to female TR group. Data presented as mean ± SD (P < 0.05)

Significant decreases in leukocyte concentration between sequential months occurred in the male UT group (between November and December), the male TR group (between April and May), and the female TR group (between July and August).

Interleukin-2 receptor (CD25) expression

The monthly changes in the percentage of CD4+ lymphocytes positive for CD25, for male and female groups, are presented in Fig. 2a, men and b women. The male UT group initially had a significantly higher percentage of CD4+CD25+ than the male TR (29.4 ± 4% positive compared to 18.4 ± 3% positive, respectively), but the male UT February percentage positive decreased by 55.5 ± 5% from the baseline value and remained lower than the TR group for the duration of the study. The male TR group showed a significantly higher percentage positive than the male UT in February with an increase of 92.4 ± 4% from the January baseline value, and values remained significantly higher in March, April, May, June, September B and December compared to the UT group. The male TR group was also significantly higher than the female UT group during June and July. The female TR group had a significantly higher percentage of CD4+CD25+ than the female UT group during July.
Fig. 2

a Monthly changes in percentage of CD4+CD25+ in untrained and trained male subjects. * TR group was significantly different to UT group. # Male TR group was significantly higher than the female UT group. Data presented as Mean ± SD (P < 0.05). b Monthly changes in percentage of CD4+CD25+ in untrained and trained female subjects. * TR group was significantly different to UT group. Data presented as mean ± SD (P < 0.05)

Significant sequential monthly increases in CD4+CD25+ occurred between January and February (male TR), April and May (female TR), August and September A (male UT, male TR, female UT) and October and November (male TR).

Changes in CD25 density were measured by mean channel log fluorescence and are presented in Fig. 3 a, men and b, women. The male TR group showed significantly greater CD4+CD25+ density than the female TR group in February, but there were no other significant differences in CD25+ cell surface expression between UT and TR male and female groups.
Fig. 3

a Monthly changes in CD4+CD25+ density in untrained and trained male subjects. * UT group value was significantly different to the previous month. # TR group value was significantly different to the previous month. Data presented as mean ± SD (P < 0.05). b Monthly changes in CD4+CD25+ density in untrained and trained female subjects. * UT group value was significantly different to the previous month. # TR group value was significantly different to the previous month. Data presented as Mean ± SD (P < 0.05)

Significant sequential monthly increases in CD25 density occurred between June and July (female TR), August and September A (male UT, male TR, female TR), and November and December (male TR).


We hypothesised that 12 months of training would increase the percentage of CD4+ positive for CD25, and CD25 density, in men and women. The principal finding of our study is that the male TR group showed a significantly higher percentage of CD4+CD25+ lymphocytes during 12 months of moderate intensity training compared to female TR and male/female UT groups. Notably, the female TR group showed a significantly higher percentage of CD4+CD25+ than the female UT during July only. Moderate intensity training did not increase CD25 density in either the male or female group post-study, but we did find significant sequential monthly changes in both percentage of CD4+CD25 and CD25 density during the study for male and female UT and TR groups, which suggest that there may be seasonal variations in CD25+ expression. Seasonal decreases in mitogenic responsiveness and CD25 expression may be associated with increased risk of infection during these seasons (Brock 1987).

Twelve months of training significantly increased VO2peak in men and women (14.3 and 16.7%, respectively, P < 0.001) but did not increase resting leukocyte concentration in male and female TR groups. Others have also found that total resting leukocyte concentration remains essentially unchanged with training (Nieman et al. 1993; Woods et al. 1999), although there may be training-induced changes in leukocyte subset cell concentrations (Crist et al. 1989; Yan et al. 2001). There is also considerable evidence that total resting leukocyte concentration does not change with increasing age (Ahluwalia et al. 2001; Woods et al. 1999), even though there is a decline in the function of some leukocyte subsets (e.g. NK cell cytotoxic activity) (Bruunsgaard and Pedersen 2000; Ceddia et al. 1999; Song et al. 1993). Thymic involution is primarily responsible for the decrease in naïve CD4+ lymphocytes with increasing age (Whisler et al. 1996; Schindowski et al. 2001) but older individuals have an increased percentage of memory lymphocytes (CD4+ and CD8+ lymphocytes that have already undergone division in response to an antigen) (Ceddia et al. 1999; Shinkai et al. 1995; Smith et al. 2004). The ability of memory lymphocytes to respond to antigens/mitogens depends upon the rounds of clonal expansion undergone (Bruunsgaard and Pedersen 2000; Whisler et al. 1996), and may be compromised due to impaired signal transduction and IL-2 gene transcription (Wang et al. 2000), diminished production of IL-2, and reduced expression of CD25 and the co-receptor CD28 (Neckers and Cossman 1983). We found some significant between-group differences in leukocyte concentration during the study, but as we were unable to perform a leukocyte differential, we could not determine changes in leukocyte subsets. However, the between-group differences in leukocyte concentration seemed to be random differences not attributable to moderate intensity training or to gender. Since leukocyte concentration remained in the normal ranges for all groups, the significant between-group differences during 4 months of the year may not be clinically significant or indicative of impaired leukocyte function.

We found that long-term, moderate intensity training significantly increased the percentage of CD4+CD25+ lymphocytes in older men, but not older women. Few studies have investigated the long-term effects of exercise on lymphocyte function. Crist et al. (1989) found that moderate intensity training significantly increased the percentage of CD4+ cells expressing CD25, while others found that training increased the concentration of CD4+ lymphocytes in older men (Woods et al. 1999). Shinkai et al. (1995) and Woods et al. (1999) found significantly greater responsiveness of CD4+ lymphocytes to PHA, and increased IL-2 production, after 3 and 6 months of training, respectively. Smith et al. (2004) found that trained older men had a greater T lymphocyte antibody response than sedentary older men, while Wang et al. (2000) reported that aerobically-trained older men showed enhanced protein kinase C (a second messenger in lymphocyte gene activation) activity in response to PHA. The male UT group began the study with a significantly higher baseline percentage (55.5 ± 5%) of CD4+CD25+ than the male TR group, which is not unusual given that there is normally some variation in the mitogen-stimulated percentage of CD4+CD25+ in a normal population (Neckers and Cossman 1983). However, TR group increased their percentage of CD4+CD25+ by 92.4 ± 4% after 4 weeks of training, and their values remained significantly higher during 6 months of the study. Our results suggest that long-term moderate intensity training increases the recruitment and concentration of functional memory CD4+CD25+ lymphocytes in response to PHA, possibly as a compensatory immune response to a diminished proliferative ability and an age-related decrease in naïve CD4+ lymphocytes in males. Training may increase the recruitment of active lymphocytes from the endothelial linings of blood vessels, spleen and lymph nodes through increased blood flow, catecholamine concentration, signal transduction and IL-2 production (Ahluwalia et al. 2001; Nieman et al. 1993; Wang et al. 2000). Since older individuals have a reduced number of naïve CD4+ lymphocytes, it seems logical to suggest that the blood concentration of memory CD4+ lymphocytes could increase with training, in an attempt to maintain the lymphocyte response. The significant increase in the percentage of active CD4+CD25+ rather than an increase in CD25 density in our male TR subjects suggests that lymphocyte mitogenic responsiveness is enhanced by a greater number of active memory lymphocytes, rather than an increased number of CD25 receptors expressed on cell membranes. Shinkai et al. (1995) reported that older trained male runners had a higher percentage of active memory CD4+ lymphocytes, and greater CD4+ responsiveness to PHA, compared to older sedentary men, yet there was no difference in CD25+ density between groups, and the authors suggested that training increased the recruitment of active memory CD4+ from peripheral reservoirs. Thus the risk of infections in older men may be reduced by long-term, moderate-intensity exercise-induced increases in memory CD4+CD25+. One of the limitations of our study was that we were unable to measure CD4+ memory and naïve phenotypes, and further studies examining the concentrations of CD4+CD45RA+ and CD4+CD45RO+ during long-term training would be of benefit.

The reasons for the lack of a training-induced increase in CD4+CD25+ in older women remain unclear. There are few studies investigating the long-term effects of moderate intensity training on immune function in older women and our study is the first to examine the effects of 12 months of training on lymphocyte responsiveness. Gueldner et al. (1997) found that older trained women had a significantly higher percentage of CD4+CD25+ lymphocytes after 4 months of aerobic exercise. Nieman et al. (1993) found that older well-trained female endurance athletes (65–84 year) showed improved T cell responsiveness to mitogens but 12 weeks of moderate intensity training had no effect on T cell function in previously sedentary older women. It is possible that older women might show a greater CD4+ mitogenic response to more training sessions per week, or to a higher training intensity than was used in the present study (60% VO2peak) (Fahlman et al. 2000). It is also possible that more than 12 months of moderate intensity training is needed for older women to show improved CD4+CD25+ expression, and that there is a gender difference in the time needed to increase CD4+ lymphocyte responsiveness to mitogens. Several studies have found that older women have an increased concentration of NK cells rather than T lymphocytes after training (Crist et al. 1989; Fahlman et al. 2000; Nieman et al. 1993), and it is possible that there is a gender difference in the recruitment of different lymphocyte subsets with long-term training. As the cell-mediated immune response is an important defence against cancer and infections, further studies are needed to ascertain how training might improve the CD4+CD25+ response in women.

We found significant sequential monthly changes in both leukocyte concentration and CD25 expression, consistent with the findings of others (Maes et al. 1994, 1997; Mann et al. 2000). Monthly and seasonal changes in leukocyte concentration have been attributed to seasonal changes in cortisol, prolactin and melatonin concentration (Brock 1987; Maes et al. 1994, 1997), with increased cortisol production inversely related to leukocyte concentration. Though our subjects showed leukocyte concentration in the normal range during the year, reported seasonal variations in leukocyte concentration suggest that there may be a greater risk of individuals succumbing to infections during certain months of the year when leukocyte concentration is lower. We were not able to measure cortisol concentration during the study, but we found that CD4+ lymphocyte concentration decreased for all groups during April and again during late September, in agreement with others (Brock 1987; Maes et al. 1994; Mann et al. 2000), suggesting that individuals might be at more risk of infection during these months.

We found significant sequential monthly changes in the percentage of CD4+CD25+, and in CD25 density (Figs. 2, 3), in agreement with others (Brock 1987; Pati et al. 1987). As with seasonal changes in leukocyte concentration, the changes in CD25 expression may be due to seasonal variations in cortisol, melatonin and prolactin production (Brock 1987; Maes et al. 1997), and suggest that March–April is a period of reduced lymphocyte responsiveness and proliferation, while August–September and January are periods of increased lymphocyte responsiveness. Others have also reported an increase in PHA lymphocyte proliferative responses from September to November (Pati et al. 1987), and definite seasonal variations in both T and NK cell function (Brock 1987; Maes et al. 1994). In our study, the seasonal changes occurred regardless of training status, and may be of clinical interest in that there may be certain months of the year where older individuals have reduced lymphocyte antigenic/mitogenic responsiveness and proliferation (Maes et al. 1994; Mann et al. 2000; Neckers and Cossman 1983). Our findings support the results of others in that autumn (southern hemisphere) and spring (northern hemisphere), rather than winter, may be seasons of reduced lymphocyte proliferation where older individuals are more at risk of infection.

Our study concluded that long-term moderate intensity training enhanced lymphocyte function in older men by significantly increasing the percentage of active CD4+CD25+. Long-term moderate intensity training may increase the responsiveness of CD4+ lymphocytes to mitogenic stimulation, improving CD4+ function and reducing the risk of infection in older men. The increased percentage of CD4+CD25+ may be due an increase in the recruitment and redistribution of functional memory CD4+CD25+ from other lymphoid compartments. The same training response was not found in older women, and further investigations are needed to determine whether more frequent doses of exercise or higher training intensities enhance the CD4+ lymphocyte response in women.


The authors would like to acknowledge the contributions made to the study by Dr Norman Morris and Dr Bon Gray of Griffith University, Gold Coast, QLD, Australia.

Copyright information

© Springer-Verlag 2008