Exercise and blood lymphocyte subset responses: intensity, duration, and subject fitness effects ADAM KENDALL, LAURIE HOFFMAN-GOETZ, MICHAEL HOUSTON, BRIAN MAcNEIL, AND YOGA ARUMUGAM Departments of Kinesiology and Health Studies, University of Waterloo, Waterloo, Ontario NZL 3Gl, Canada
KENDALL, ADAM, LAURIE HOFFMAN-GOETZ, MICHAEL HOUSTON, BRIAN MACNEIL, AND YOGA ARUMUGAM. Exercise and blood lymphocyte subset responses: intensity, duration, and subject fitness effects. J. Appl. Physiol. 69(l): 2X-260, 1990.This study examined the effect of exercise intensity and duration on the percent blood lymphocytes in men of low [LF; maximal 0, uptake (vo2 ,,,) < 50 ml. kg-‘. min-l and sedentary], moderate (MF; vo2 max = 50-60 ml. kg-l’min-’ and recreationally active), and high (HF; vo2 max > 60 ml kg-‘. min-’ and recent training history) fitness. Thirty healthy adult men (aged 20-31 yr) participated in four randomly ordered cycle ergometer rides: ride 1 (65% VO, m8X, 30 min), ride 2 (30% VO 2 max9 60 min), ride 3 (75% VOgmax, 60 min), and ride 4 (65% VO 2 max9 120 min). Blood samples were drawn at various times before and after the exercise sessions. Lymphocyte subsets were determined by flow cytometry using monoclonal antibodies for total T (CD3+), T-helper (CD4+), and T-suppressor (CDS’) lymphocytes and for a subset of cells expressing a natural killer (NK) cell antigen (Leu7+). Plasma catecholamines were assayed to determine exercise stress. There were sharp reductions (P < 0.01) in the percentage of pan-T and T-helper lymphocytes immediately after exercise across all fitness levels; the magnitude of this reduction was greatest after the highest intensity (ride 3) or longest duration (ride 4) work. In contrast, the absolute number of T and T-helper cells tended to increase after exercise and significantly so in the HF subjects (P < 0.005). There was no significant effect of exercise or subject fitness category on the percentage of T-suppressor lymphocytes, although the absolute numbers of this subset increased significantly after exercise in LF subjects. Marked increases (P c 0.01) in the percentage of NK cells occurred immediately after exercise at all intensities and durations tested; numerical increases in total NK cells were significant in all fitness groups after the highest intensity work (ride 3; P c 0.005). Irrespective of whether the changes were expressed as percentage or total numbers, recovery to base line occurred at 30 min after exercise. The results suggest that the exercise effect on blood lymphocyte subset percentages in men is transient and occurs across all fitness levels. Concomitant changes in plasma catecholamine concentrations are only weakly associated with these lymphocyte subset percentage responses to exercise. Furthermore, this study shows that the exercise-induced changes in lymphocyte percentages do not consistently reflect changes in the absolute numbers of cells. l
lymphocyte cholamines
phenotypes;
submaximal
work; conditioning;
cate-
MANY INDIVIDUALS believe that regularly performed exercise helps improve resistance to infection, but only limited epidemiological data support this claim (20). 0161-7567/90
$1.50
Copyright
0 1990
Exercise is a form of physical stress because it releases a number of hormones in common with stereotypical stress responses (21). During exercise the hormones released are dependent on the relative intensity of exercise (8). Therefore, by manipulating the intensity and duration of an exercise bout, it is theoretically possible to manipulate the concentration of a specific hormone in the blood. Many of the hormones released during exercise have immunoregulatory properties. For example, elevations in blood catecholamines during exercise may mediate the immunological changes during exercise (1, 13). Immunosuppression has been observed after in vivo administration of epinephrine (7). In vivo administration of epinephrine has also been reported to upregulate natural killer (NK) cell activity and number in humans (22). Recirculation or lymphocyte traffic patterns may alter subset proportions in the lymphoid tissue, and this recirculation may be differentially affected by exercise and the attendant hormonal changes (6). Studies in exercise immunology are often limited by their experimental design. Exercise protocols have ranged from the 5-min stair climb (9) to the 24-h military march (10). Intensity levels have ranged from recreational tennis (18) to intense endurance bicycling (16). Keast et al. (11) emphasized that exercise protocols need to be better controlled with respect to exercise duration and intensity to explore the role of exercise stress on the immune system. The purpose of this study was to evaluate the effect of exercise duration, intensity, and fitness level on the percentage and numerical responses of human lymphocytes after single bouts of exercise. The underlying hypothesis was that the shorter the duration and/or intensity of the work, the smaller the impact on the percentage change in lymphocyte subsets. Conversely, the longer and more intense the work, the greater the impact on lymphocyte subsets. Accordingly, we determined the effects of four different bicycle ergometer rides on changes in the percentage of peripheral blood T-lymphocytes, Tlymphocyte subsets (helper/inducer and cytotoxic/suppressor), and large granular lymphocytes, including NK cells, by using monoclonal antibodies and flow cytometry. Additionally, we wanted to determine whether the changes in lymphocyte percentage were related to concurrent changes in plasma epinephrine and norepinephrine concentrations. the American
Physiological
Society
251
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
252
EXERCISE
INTENSITY
AND
DURATION
MATERIALSANDMETHODS
LS’ubJects
Thirty young adult males were selected from the university population. All subjects were administered a selection criterion, activity inventory questionnaire, and an informed consent approved by the Office of Human Research. The subjects were recruited in four groups. Group 1 (n = 8) were control subjects and did not participate in any exercise testing. Group 2 (n = 8) were individuals recruited for their lack of training history and involvement in sport; they were referred to as a sedentary low-fit sample. Low-fit subjects were also classified as having maximal aerobic capacities (VO, max) ~50 ml OZ. min-’ kg body wt-‘. Group 3 (n = 6) was classified as moderately active, with Tj02max in the range of 50-60 ml OZ. min-’ . kg-‘. These subjects had an activity profile of recreational athletes with no formal exercise training history. Group 4 (n = 8) was defined as very active WO2max> 60 ml 02. min-’ . kg-‘). This group comprised high-fit individuals who had a recent history of active involvement in endurance exercise training. To ensure compliance, all subjects received an honorarium on completion of the study. To aid in fitness classification, the subjects completed an activity profile questionnaire. This questionnaire requested the type and frequency of activities performed the week before the study. A composite score was calculated based on the relative energy expenditure above basal metabolic rate; this was termed metabolic equivalents (METS) per week. The composite score equaled the time spent per week on a particular activity multiplied by the average energy expenditure for that activity. This method of determining recent physical activity was adopted and modified from Shouten et al. (19). l
Experimental Design
Each subject performed two iTOpmax tests before the exercise tests. All exercise sessions and VOW maxtests were performed on an electronically braked bicycle ergometer (Quinton model 870); subjects were required to maintain a pedaling frequency of 60 pedal revolutions per minute. Each V02maxtest began with 4 min of loadless pedaling to allow the subjects to warm up and adjust to the equipment. Power output was subsequently increased by 30 W/min until exhaustion. The bicycle ergometer was modified to include an analog-integrating circuit to produce a linear increase in power output or a ramp signal. Maximal test or exhaustion was defined by one or more of the following: a plateau or decline of oxygen uptake (VO,), a respiratory exchange ratio in excess of 1.15, a maximal heart rate consistent with the subject’s agepredicted maximal heart rate, and the subject’s inability to maintain the desired pedaling frequency. Bicycle ergometry allows for precise control of power output over a wide range and is an almost completely concentric rhythmic activity that does not produce undue muscle or joint trauma. Heart rate was monitored by electrocardiography immediately before and every 2 min during the exercise and VOW maxtests. Four exercise sessions were randomly administered 1 wk apart: two bouts of cycling at a fixed duration of 60
EFFECTS
ON
LYMPHOCYTES
min at intensities of 30% (ride 2) and 75% (ride 3) of VO, maxand two bouts at a fixed intensity of 65% VOW max for durations of 30 (ride I) and 120 (ride 4) min. For each exercise session, the power output was increased progressively from 0 W to the desired level over 5 min, at which point the timer was started. VOW maxwas monitored at 5- to lo-min intervals to ensure that the appropriate work load was maintained. Blood samples were drawn from a forearm vein by venipuncture at the following times: 24 h before each exercise bout, 3 min before and 3 min after the termination of the ride, and 30 min, 2 h, and 24 h after exercise. The 3-min before and after samples will interchangeably be referred to as immediately preexercise and immediately postexercise in the text. Blood samples at rest were drawn after the subjects had relaxed in a chair for 15 min. The exercise bouts were initiated at 9 A.M. every week to minimize the effects of hormonal diurnal variation. Experimental Procedures
Mononuclear leukocytes were isolated from heparinized blood by a modified method of Boyum (4). Briefly, blood was diluted twofold with phosphate-buffered saline (PBS; 0.01 M, pH 7.2), layered over Histopaque (Sigma Chemical, St. Louis, MO), and centrifuged at 400 g for 30 min. The mononuclear cell interface was collected, washed three times, suspended in 1 ml of PBS, counted, and concentration adjusted to 2 x lo7 cells/ml. Fiftymicroliter aliquots of this cell suspension were incubated 45 min at 4°C together with 20 ~1 of each of the selected monoclonal antibodies. Labeled cell suspensions were fixed in 2% paraformaldehyde for analysis within 1 wk of sample collection. All mononuclear cell counts were performed on a Nikon Optiphot phase-contrast microscope, and cell viability was assessed by the trypan blue exclusion technique and was always >95%. Lymphocyte subpopulations were quantitated by the technique of direct immunofluorescence using monoclonal antibodies specific for peripheral total T-cells (CD3’), helper/inducer T-cells (CD4+), suppressor/cytotoxic T-cells (CD8+), and large granular lymphocytes, a subset of which expresses NK functional activity (Leu7+). Monoclonal antibodies were obtained from Becton Dickinson and used with a fluorescence-activated cell sorter (Coulter EPICS IV FACS) with an argon laser (488-nm wavelength) to excite the fluorescein-isothiocyanate-conjugated monoclonal antibodies. Background autofluorescence was determined for each sample, and an average of 10,000 cells per sample was counted. Hematocrit and Hemoglobin
Venous whole blood was collected in EDTA and was used for the determinations of hematocrit and hemoglobin content. For hematocrit determination, blood was centrifuged in heparinized microcapillary tubes at 11,500 revolutions/min for 6 min. Samples were determined in triplicate. Hemoglobin was determined by adding 6.0 ml cyanomethemoglobin reagent to 20 ~1 of the whole blood mixed in EDTA. This was then vortexed and left to stand for 20 min. Absorbance was read at 540 nm against
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
EXERCISE
Group
n
C LF MF HF
8 8 6 8
Age, yr 24.3k0.3 26.1kl.O 25.71k1.3 23.1k0.5
INTENSITY
Height, cm
Weight, kg
VO 2 maxp ml. min-’ kg-’
176.6k3.1 176.5k3.3 18O.lk2.6 176.9t1.7
78.1t2.1 71.2k4.4 77.1t2.9 71.423.4
52.4t2.3* 44.9k1.5t 55.2tl.6” 63.3+1.8$
AND
.
DURATION
MET Score, per wk
2,475+561 1,103&296 1,457+308 3,138+676
EFFECTS
ON
LYMPHOCYTES
253
difference between groups for MET score; however, the low-fit group had the lowest MET score (1,103 t 296 METs/wk), and the high-fit group had the highest (3,138 t 676 METs/wk) MET score. The control group was in the middle with a score of 2,475 t 561 METs/wk. Lymphocyte
Subset Percentages
Total T lymphocytes (CD3+). Total T-lymphocyte percentage responses after exercise are shown by group in Fig. 1, A-D. Across all fitness groups, the 30% Vozrnax ride elicited the smallest decrease in CD3+ cells after exercise and was significantly (F3,78 = 5.02, P < 0.003) different from all other rides. The immediately postexercise samples had significantly (F5,130 = 41.95, P < 0.0001) lower percentages of CD3+ cells compared with a reagent blank, and hemoglobin concentration values other sample time points when the results were pooled were determined from a standard curve. for all levels of fitness. The main time effects were significant in the moderate- (Fig. lC, F5 15 = 46.21, P < Catecholamine Determination 0.0001) and high- (Fig. lD, F5 15 = 25.95,‘P c 0.0001) fit Aliquots (1.5 ml) were extracted from ethylene gly- populations, whereas there was no statistically significol-bis(P-aminoethyl ether)-N,N,N’,N’-tetraacetic acid- cant decrease in CD3+ cells in the control (Fig. 1A) and chelated plasma for catecholamine determination and low-fit populations (Fig. IB). The decrease in CD3+ stored in Eppendorf tubes at -90°C until assayed. The phenotype occurred immediately postexercise and in all concentrations of norepinephrine (NE) and epinephrine cases had returned to base line by 30 min postexercise. (E) were analyzed by high-performance liquid chromaT-suppressor lymphocytes (CD8+). The T-suppressor tography using a modification of the method of Weicker lymphocyte population was highly resistant to changes and co-workers (25). Catecholamine concentrations were in percentage after all exercise treatments. There was no adjusted to reflect exercise-induced changes in plasma overall ride main effect, nor were there any differences volume using the van Beaumont equation (24). between fitness groups in their response to exercise. The t tests contrasting time within each ride for each group revealed no differences in T-suppressor percentage when Statistical Analysis compared with the preexercise sample. Data were analyzed by three-way analysis of variance T-helper lymphocytes (CD4+). Figure 2 shows the T(ANOVA) for significant between-group main effects, by helper lymphocyte percentage response to exercise in a series of two-way ANOVA for within-group main ef- men of defined fitness levels. Across all fitness groups, fects, and by Pearson correlation analyses. Specific dif- the least intense ride (ride 2 at 30% VOWmax) produced ferences between ride and time points were analyzed post the smallest reduction in the percentage of CD4+ lymhoc by Student-Newman-Keuls (SNK) test using an cy phocytes (F~,T~s= 6.04, P < 0.0009) compared with all = 0.01. Student’s t tests with CY= 0.005 were employed other rides. The T-helper lymphocyte percentage was to determine time differences within each ride; these significantly (F5,130 = 49.84, P c 0.0001) depressed only conservative cylevels were chosen to reduce the possibilimmediately postexercise, irrespective of fitness group. ity of type I error. The T-helper lymphocyte response of the low-fit subjects to exercise is shown in Fig. 2B. The mean T-helper RESULTS lymphocyte percentage fell by 23%, when averaged across all rides. In this group, the ride responses did not differ, Subject Characteristics as there was a significant reduction in CD4+ cells even The subject characteristics for each fitness group are after ride 2. Fig. 2C illustrates the T-helper lymphocyte presented in Table 1. There was no difference between percentage changes as a result of exercise for the modthe mean age, height, or weight of individuals in the erate-fit group. CD4+ cell percentage dropped significantly (P < 0.005) immediately postexercise in rides 1, control, low-fit, moderate-fit, or high-fit categories. There was, however, a significant (8& = 17.19, P < 3, and 4. Contrasts comparing immediately preexercise 0.0001) difference between the VOW,,, of the three exer- to 2 h postexercise revealed a significant difference from cising groups. The low-fit group had the lowest Vozrnax base line for ride 3. The CD4’ cell percentage response of the high-fit individuals to exercise is seen in Fig. 20. with a mean of 44.9 t 1.5 ml OZomin-’ kg-‘. The modA significant (F5,15 = 27.50, P < 0.0001) reduction in Terate-fit group was next highest with a mean of 55.2 t 1.6 ml OZ. min-’ . kg-’ for six subjects. The high-fit pophelper cell phenotype occurred when averaged across all ulation had the highest VO 2 maxwith a mean of 63.3 t 1.8 rides immediately postexercise; this reduction was -26%) and values returned to base line by 30 min postexercise. ml OZ. min-’ . kg? The control group’s vOzmax (52.4 t 2.3 ml 020min-‘o kg-‘) was not significantly different Ride 2 had a significantly (Fs,zl = 5.55, P < 0.0057) smaller effect on T-helper cell percentage than did ride from that of the moderate-fit group but had a different 3. mean from that of the low- and high-fit groups. Pairwise Large granular l.ymphoc-ytes with NK activity (Leu7+). comparisons using SNK at CY= 0.01 did not reveal anv Values are means t SE; n, no. of subjects. MET score was calculated based on activity questionnaires for the week before the study. This measure classifies physical activities according to energy expenditure independent of body weight as described in Ref. 19. C, control; LF, low fitness; MF, moderate fitness; HF, high fitness. *t"f Different at P < 0.01 by SNK test except where symbols are the same.
l
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
254
EXERCISE
INTENSITY
AND
DURATION
EFFECTS
A 90
ON
LYMPHOCYTES
B 0 Ride1
Ride2
l
0 Ride3
90
B Ride4
0 Ride1
l
Ride2
•I Ride3
n
Ride4
40 0 -24h
-3min
0 Ride1
+3min TIME
l
Ride2
+30min
0 Ride3
+2h
+24h
-24h
u Ride4
-3min
0 Rtdel
0
l
+3min TIME
Ride2
+30min
0 Ride3
+2h
+24h
m Ride4
0 -24h
-3min
+3min TIME
+30min
+2h
+24h
-24h
-3min
+3min TIME
FIG. 1. Percent pan-T lymphocytes (CD3’) of total mononuclear cells in control (A), low- (B), moderate(C), high-fit (D) subjects after ride treatments. Lymphocytes isolated by flow cytometry with argon laser set at 488 Ride 1 = 30 min, 65% Vozmax; ride 2 = 60 min, 30% Vo2 max; ride 3 = 60 min, 75% VOW max; ride 4 = 120 min, Abscissa is not time scaled. Percentage based on 10,000 cells. * Significant difference compared VO 2 max. immediately preexercise (P < 0.005).
Figure 3, A-D, presents the NK cell percentages as a function of exercise and fitness group. There was a significant (&26 = 6.92, P < 0.0014) difference between groups in the response of NK cells to exercise; the lowfit population had a greater mean percentage of Leu7+ cells compared with the other groups. This was possibly because of the different absolute numbers of these cells between fitness populations at rest. The low-fit subjects had -78% greater NK cells at rest, compared with the results for the other fitness categories. There was no difference between rides in terms of their effect on circulating Leu7+ cell percentage, even though ride 2 elicited the smallest increase in the NK phenotype. There was, however, a significant (F5,130= 75.25, P < 0.0001) increase in NK cell percentage immediately postexercise. All rides in the low-fit group were associated with the same magnitude level of NK cell percentage increase. A
+30min
+2h
+24h
and nm. 65% with
significant (F5,15 = 48.28, P < 0.0001) increase in NK cell percentage occurred immediately postexercise; a 68% increase in circulating NK phenotype was found when averaged across all rides. This relative increase was transient and returned to base-line levels by 30 min postexercise. Analysis by individual ride reinforced the significant (P c 0.005) increase immediately pre- to immediately postexercise for all rides. Fig. 3C represents the NK response to exercise in the moderate-fit subjects. Similar to the low-fit subjects, there was no significant difference between rides for the pattern of NK percentage increase in moderate-fit subjects. Likewise, there was a significant (F5 15= 57.02, P < 0.0001) increase of 50% in the cells bearing the NK phenotype immediately postexercise. Evaluation of individual ride effects showed that rides 1, 2, and 4 were significantly different from base line at 3 min postexercise. Ride 3 did not reach statistical signif-
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
EXERCISE
AND
DURATION
EFFECTS
A
60
Ride2
l
0 Ride3
n
ON
0 Ride1
Ride4
T 9 n -0
50
255
LYMPHOCYTES
B
60 0 Ride1
T d n -0 a g
INTENSITY
Ride2
l
0 Ride3
l
Ride4
50
a 40
iii I t%
a" cd I I%
30
20
40
30
20
un
0
-24h
-3min
+3min
+30min
+2h
-24h
+24h
-3min
+3min
TIME 60
T d n -0
l
Ride2
o Ride3
n
60
Ride4
ii I t%
+24h
D o
50
a e
+2h
TIME
C 0 Ride1
+30min
40
+^ d n 0
50
a g
40
iii I I%
30
20
Ridel‘
l
Ride2
0 Ride3
I) Ride4
30
20
0
0 -24h
-3min
+3min
+30min
+2h
+24h
-24h
-3min
TIME
icance at cy = 0.005, possibly because of large standard deviations in the response. NK cell responses to the exercise treatments in the high-fit individuals are given in Fig. 30. Again, there was no significant ride difference, but there was a highly significant (8’S,15 = 47.99, P c 0.0001) increase in NK cell percentage immediately after exercise. With post hoc pairwise comparison, rides 1, 2, and 3 were significantly different from base line (P < 0.005 in each case). Lymphocyte Subset Total Numbers
The total numbers fraction are shown (immediately before were no significant
of lymphocytes in the mononuclear by subset, fitness group, and ride and after exercise) in Table 2. There effects of time on the lymphocyte
+30min
+2h
+24h
TIME
2. Percent T-helper lymphocytes (CD4’) of total mononuclear cells in control (A), low- (B), and high-fit (D) subjects after ride treatments. Lymphocytes isolated by flow cytometry with argon nm. Ride 1 = 30 min, 65% V02max; ride 2 = 60 min, 30% V02max; ride 3 = 60 min, 75% VO* max; ride 4 VO 2 mnx- Abscissa is not time scaled. Percentage based on 10,000 cells. * Significant difference immediately preexercise (P < 0.005). FIG.
+3min
moderate(C), laser set at 488 = 120 min, 65% compared with
numerical responses in the controls. The total number of CD3+ (pan-T) lymphocytes increased after exercise at intensities of 65% VO 2MBXor higher in all fitness groups. However, these increases were significant (P c 0.005) only after ride 1 (all groups) and ride 4 (low- and highfit subjects). The same pattern of increase in the total number of CD4+ (T-helper) lymphocytes after exercise occurred; this increase was significant after rides 3 and 4 in the high-fit subjects (P < 0.005 vs. preexercise). In contrast to the generally flat percentage response of CD8+ (T-suppressor) lymphocytes with exercise, the absolute numbers increased significantly (P < 0.005) after work at intensities of 65% 7j02,,, or greater in the low-fit subjects; the total number of T-suppressor lymphocytes was also significantly elevated (P c 0.005) in the high-fit subjects after the highest-intensity work
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
256
EXERCISE
INTENSITY
AND
DURATION
A
EFFECTS
30 o Ride1
Ride2
l
0 Ride3
n
ON
LYMPHOCYTES
YB 0 Ride1
Ride4
l
Ride2
q Ride3 / 4P \\
Ride4
l
it
20
10
tu Z M 0
0 -24h
-3min
+3min
+30min
+2h
-24h
+24h
-3min
+3min
+30min
+2h
+24h
+2h
+24h
TIME
TIME
D
C 0 Ride1
l
Ride2
-24h
-3min
q
Ride3
+3min
n
+30min
0 Ride1
Ride4
+2h
-24h
+24h
l
-3min
Ride2
0 Ride3
+3min
l
+30min
Ride4
TIME
TIME
FIG. 3. Percent total mononuclear cells bearing natural killer (Leu7+) phenotype in control (A), low- (B), moderate(C), and high-fit (D) subjects after ride treatments. Lymphocytes isolated by flow cytometry with argon laser set at 488 nm. Ride I = 30 min, 65% VOW,,,; ride 2 = 60 min, 30% $70~ max; ride 3 = 60 min, 75% voz max; ride 4 = 120 min, difference compared with 65% ire, max. Abscissa is not time scaled. Percentage based on 10,000 cells. * Significant immediately preexercise (P < 0.005).
(ride 3). There was no differential effect of fitness level on the total number of NK (Leu7+) cells after work at 75% vo, max(ride 3): all groups had a significant increase in the total number of Leu7+ cells after exercise (P < 0.005). Exercise at intensities of 65% VO, max or less resulted in a more variable response across fitness groups. Low- and high-fit subjects had significant increases (P < 0.005) in the number of NK cells; moderatefit subjects followed the same pattern of increase after exercise, albeit not significantly. Total Mononuclear
other sampling intervals. The ride effect was due to the greater increases in leukocyte number after rides 3 and 4. These increases were on the order of two- to threefold above basal levels. For example, in ride 3 the increases were from 5.6 t 1.3 to 14.4 t 3.0 cells X 107/ml in highfit, from 4.1 t 0.2 to 9.4 t 1.4 cells x 107/ml in moderatefit, and from 2.7 t 0.2 to 5.4 t 0.5 cells X 107/ml in lowfit subjects, and in ride 4 the increases were from 5.3 t 1.1 to 12.8 t 2.9 cells X 107/ml in high-fit, from 3.7 t 0.5 to 8.8 t 1.3 cells X 107/ml in moderate-fit, and from 2.8 -+ 0.3 to 6.9 t 0.7 cells X 107/ml in low-fit subjects.
Cell Number
The total number of mononuclear cells recovered after density gradient centrifugation of whole blood was significantly increased as a function of time (& 130= 37.59, P < 0.0001) and ride (F3,78 = 3.97, P c O.Oi) (data not shown). The main effect of time was due to a marked leukocytosis immediately after exercise relative to all
Plasma
Catecholamines
Because catecholamines are released during exercise and have immunomodulatory as well as circulatory effects, plasma concentrations of E and NE were determined at the serial time points corresponding to immune measurements. The NE responses of subjects to the
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
EXERCISE
INTENSITY
AND
DURATION
EFFECTS
Lymphocyte Group
Ride
No.
CD3’ -3 min
C
1
2 3 LF
1
2 3 MF
1
2 3 HF
1
2 3
Subset,
ON
x 10’ cells/ml
CD4’ +3 min
-3
min
257
LYMPHOCYTES
CD8’ +3 min
-3
min
Leu7’ +3 min
-3 min
+3 min
3.6t0.3 4.6-e0.5
4.0t0.5 5.3tl.l
2.0t0.2 2.6k0.3
2.1t0.3 3.0t0.7
0.9kO.2
1.3t0.1
1.020.2 1.5kO.4
0.4kO.l 0.6kO.l
0.6kO.l 0.8t0.3
4.1t0.4 2.0k0.3 1.9-co.2 1.8k0.2 1.9t0.2 2.6t0.4 3.4k0.6 3.OkO.l 2.8k0.4 3.6t0.6 4.4t0.8 3.7t0.9 3.8t0.9
4.6-1-0.4 3.2t0.5* 2.3k0.3 3.3kO.4 4.0t0.6* 4.4&0.4* 3.1t0.4 5.1kO.7 5.321.0 5.3&0.8* 4.5k0.8 7.2tl.3* 6.8k1.4*
2.3t0.3 1.2t0.2 1.1kO.l l.ltO.l 1.1t0.1 1.6kO.2 2.1kO.4 1.9t0.6 1.7t0.3 2.3t0.4 2.9kO.7 2.4-e0.7 2.520.7
2.5t0.2 1.61!10.2* 1.2zko.2 1.4t0.1 2.0k0.3 2.6t0.3 1.9kO.3 2.8k0.4 2.8t0.4 3.1t0.5 2.8k0.6 4.0t0.8* 4.1*0.7*
1.lt0.2 0.7kO.l 0.6tO.l 0.6kO.l 0.7kO.l 0.7tO.l 0.9kO.l 0.7kO.l 0.7kO.l 0.9t0.2 1.3t0.2 l.OkO.2 1.0t0.2
1.3k0.2 1.2+0.2* 0.7gO.2 1.4t0.2* 1.5t0.3* 1.3t0.2 0.9kO.l 1.9kO.4 1.8t0.3 2.1kO.5 1.4t0.3 3.2&0.6* 2.7k0.7
0.5t0.1 0.4t0.2 0.3t0.1 0.4kO.O 0.4t0.1 0.2kO.O 0.3kO.l 0.2to.o 0.2tO.l 0.4kO.l 0.3t0.1 0.4kO.8 0.3t0.1
0.7t0.1 1.lt0.2* 0.6tO.l 1.5t0.2* 1.5t0.2* 1.420.4 0.6kO.l 2.0t0.6* 1.620.4 1.7t0.5 0.8&0.2* 2.7&0.6* 1.9t0.5"
Values are means t SE; SE values < been rounded up to single decimal place exercise (-3 min).
0.05 have been rounded down to single decimal (0.1). Lymphocyte subset is from mononuclear
exercise tests are shown in Fig. 4. There was a significant main effect of ride (F3 78 = 11.61, P < 0.0001) for plasma NE concentrations, with the longest duration ride (ride 4, 65% \jozmax for 2 h) or highest intensity ride (ride 3, 75% Topmax for 1 h) producing the largest increase in plasma NE. Plasma NE levels peaked immediately postexercise and returned to basal levels 2 h after completion of physical work (F15,130= 10.36, P < 0.0001 for main effect of time). There was virtually no change in plasma NE in control subjects (Fig. 4A) as a function of time. In contrast, low- (Fig. 4B), moderate- (Fig. 4C), and high-fit (Fig. 40) subjects demonstrated a sharp increase in plasma NE levels immediately postexercise and especially in ride 3 (P c 0.0005 in all cases); this highintensity ride effect on plasma NE occurred independently of fitness group. In addition, high-fit subjects had a significant NE peak immediately after ride 4. Figure 5 shows the within-group effect for changes in plasma E as a function of ride and time. Not surprisingly, there were overall significant main effects of group (Fs,zG = 4.00, P < O.Ol), ride (F378 = 10.77, P < O.OOOl), and the V&30 = 31.51, P < O.Obol) for plasma E. All groups, except controls, had an increase in plasma E immediately postexercise after work at high intensity (ride 3) and long duration (ride 4). However, because of large withingroup variation, the peak increase was only significant in the high-fit subjects (Fig. 5D). Pearson correlation analyses were done to determine the relationship between the change in a specific hormone (NE, E) pre- to postexercise and the change in the percentage of a given lymphocyte subset. There was no significant relationship between either E or NE and changes in CD3+ or CD8+ cells. The postexercise decrease in the percentage of CD4+ cells was correlated with the increase in NE levels postexercise (r = 0.38, P C 0.0003). Postexercise increases in NE were also correlated with the increase in the percentage of Leu7+ cells (r = 0.40, P c 0.0002). DISCUSSION
This study evaluated the effect of individual fitness level on the percentage and total number of T-lympho-
place (0.0) and SE values between 0.05 and 0.099 have cell-rich fraction. * P < 0.005 vs. immediately before
cyte subsets in blood after single bouts of exercise at controlled durations (60 and 120 min) and intensities (30, 65, and 75% VO Zmax). Thirty healthy adult males (aged 20-31 yr) were assigned to fitness categories based on functional aerobic capacity and physical activity profiles. All three fitness groups had significantly different Vo2 m8Xvalues. However, VO, maxrepresents continuous rather than categorical data; therefore, the cutoff points used for the minimum and maximum values for the fitness categories were, to some degree, arbitrary. Activity profiles (MET scores) were generated from questionnaires to aid in subject classification. The MET values for the four different groups were not significantly different. Subject recall, over and underestimation of activity, and subjectivity of the measures are clearly potential confounds when utilizing questionnaire methodology. Nevertheless, despite wide variations in MET scores across subjects, individuals in the low-fit group had the lowest average MET and those in the high-fit group had the highest MET (1,103 t 296 and 3,138 t 676, respectively). It was hypothesized that exercise of shorter duration and lower intensity would produce a smaller change in the percentage of T-lymphocyte subpopulations in peripheral blood, regardless of fitness level. This was hypothesized because the release of immunomodulatory hormones in the blood generally increases in proportion with relative exercise intensity and duration (5). Exercise that elicits less stress or immunomodulatory hormone release should be associated with smaller lymphocyte and leukocyte numerical and percentage shifts. Analysis of the percentage of lymphocytes in peripheral blood as a function of fitness level and exercise treatment revealed several consistent patterns. First, the T-suppressor (CD8+) cell percentage response was extremely resilient; there was minimal change after exercise of either high intensity (75% VOW max)or long duration (120 min). This insensitivity to exercise occurred irrespective of whether the subjects were of low, moderate, or high fitness. Some investigators have reported
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
258
EXERCISE
0 Ride1
l
INTENSITY
Ride2
0 Ride3
l
AND
DURATION
EFFECTS
ON
1600
-
Ride4
E
1400
-
1400
-
s
1200
-
1200
-
1000
-
1000
-
800
-
800
-
iii!
600
-
600
-
0 Z
400
400
-
200
-
-
E 2 w Z [L
a I m
.a*- . .-. . . . . .- --
-.--
200
P
a
5
I
*.
. . . . . . .--. . . -
-H-+-4=-3
0 ’
-l h U
-24h
-3min
+3min
t30min
+2h
+24h
a I to a 2
LYMPHOCYTES
0 Ride1
0 Ride2
-24h
-3min
+3min
Ride2
0 Ride3
l
a 0 Z a r v)
-
200
-
1400
a E
1000
2
a zl"
0 A -24h
-3min
+3min
+30min
+24h
+2h
+24h
TIME
Ride2
l
0 Ride3
n
Ride4
800 I
CL
a 2
n”
si -
600 400
0 Ride1
Ride4
z E loo0 g 800 a. w
+2h
D
1600
F
l
+30min
TIME
C
Ride1
m Ride4
0
TIME
0
0 Ride3
T
600
1
-I
0 -24h
-3min
+3min
+30min
+2h
+24h
TIME
FIG. 4. Norepinephrine concentration in plasma of control (A), low- (B), moderate(C), and high-fit (D) subjects after treatments. Norepinephrine determination was by high-performance liquid chromatography. Ride I = 30 min, 65% vo2 ,,,ax; ride 2 = 60 min, 30% VOW,,,; ride 3 = 60 min, 75% vo2 max;ride 4 = 120 min, 65% vo2 max. Abscissa is not time scaled. * Significant difference compared with immediately preexercise (P < 0.005).
significant changes in the percentage of CD8+ cells immediately after acute physical work at maximal intensities in humans (2), but many reports of the effect of exercise on the T-suppressor subset have been inconclusive. There have been relative increases (14, 15) or no net change (23). Our data indicate that at submaximal exercise intensities (~75% Vo2 max), the percentage of T suppressor cells in blood will not change. It is not known from these data whether higher intensity work (but still below maximal) or repetitive bouts of submaximal exercise will affect the relative percentage of this lymphocyte subset. Second, irrespective of subject fitness level, the percentages of CD3+ and CD4+ lymphocytes were reduced immediately after exercise. Furthermore, the magnitude of these reductions was influenced by exercise intensity and duration: in most cases, the higher intensity work (75% vo2,,, for 60 min) produced the largest drop in the percentage of these cells. Longer duration bouts of aerobic work (65% VO 2 1118X for 120 min) resulted in almost
as great reductions in the percentage of these cells. Exercise of the lowest intensity (30% I702 m8x for 60 min) resulted in some reductions in the percentage of these cells albeit not as dramatically or consistently as the other work sessions. Thus it appears that the higher the relative intensity of aerobic work, the greater the impact on the relative percentage of CD3+ and CD4’ lymphocytes. Subject fitness level does little to modify this basic effect. These results agree with Oshida et al. (17), who found similar decrements in T-helper lymphocyte percentage immediately after exercise. Their subjects were both trained and untrained and rode a bicycle ergometer for 2 h at 60% \jo2 max. These results also agree with Tvede et al. (23), who reported reductions in the CD3+ and CD4+ lymphocyte percentage after exercise, and they attributed the decrease in the CD3+ percentage to the decrease in the CD4+ subset. This conclusion is logical because the pan-T marker, CD3, is present on all T-lymphocytes. If the CD8+ subset remains constant during exercise, and there is a marked reduction in the
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
EXERCISE
INTENSITY
AND
DURATION
EFFECTS
w z
0 Ride2
0 Ride3
n
a w z L w < I
I
Ride4
0 Ride1
l
Ride2
T
0 Ride3
Ride4
l
w z
250
a I
259
LYMPHOCYTES
1 B
350 0 Ride1
ON
200
a I 2
-j
150 100
m
a
50
-a
2 0
i
’
-24h
se.........
*.........* --
f**
4.
&=---d -3min
+3min
+30min
+2h
+24h
-24h
-3min
+3min
TIME
z > -n w z “r tl w z a. w < I CA u
-
-
300
-
250
-
250
-
200
-
200
-
150
-
100
-
-
300
c 0 Ride1
l
Ride2
0 Ride3
l
D
Ride4
0 Ride1
.. *. .‘. ;: .. ...
150-
+24h
TIME 350
350
+2h
+30min
l
Ride2
:
loo50 -
T
•I Ride3
n
Ride4
‘..
50 -
2
-
0 ’
0 -24h
-3min
+3min
+30min
+2h
+24h
TIME
-24h
-3min
+3min
+30min
+2h
+24h
TIME
5. Epinephrine concentration in plasma of control (A), low- (B), moderate(C), and high-fit (D) subjects after treatments. Epinephrine determination by high-performance liquid chromatography. Ride I = 30 min, 65% ride 4 = 120 min, 65% vO2 max. Abscissa is not VO 2 max9 - ride 2 = 60 min, 30% vOzmax; ride 3 = 60 min, 75% vO2,,,; time scaled. * Significant difference immediately pre- to immediately postexercise (P < 0.005). FIG.
CD4’ population, then the CD3+ decrease should be due to the reduction in CD4 phenotype. There were similar percentage decrements in the CD4+ and CD3+ cell in this present study as well. Third, exercise at intensities ~65% vozrnax produced marked increases in the percentage of Leu7+ (a subset of NK) cells immediately after cessation of exercise. As with the T-cells, the magnitude of the percentage change was greatest after ride 3 (75% Vop maxfor 60 min). Surprisingly, even at exercise of low intensity (30% VOg max for 60 min in ride Z), NK cell percentage increased significantly. Although the clinical significance of the increase in the relative percentage of NK cells cannot be determined from this study, it is tempting to speculate that an increase may be potentially beneficial; NK cells have the capacity to lyse tumor targets. Because the significant increase in NK cells occurred after rides even at relatively low intensity (ride Z), the potential immunological benefits of changes in this parameter may be achieved at relatively low metabolic costs.
The results of this study show that at all fitness levels, the higher the intensity of work the greater the impact on the percentages of T lymphocyte subsets. In contrast, the impact of exercise and fitness level on the total number of lymphocytes in each subset was more variable. Although high-intensity or long-duration work produced dramatic increases in the number of virtually all lymphocytes in the high-fit subjects, moderate-fit subjects had increases only in the total number of NK cells and only after the high-intensity ride (ride 3); low-fit subjects had significant increases in the number of T, T suppressor, and NK cells (but not T-helper lymphocytes) after rides 3 and 4. Given that all fitness groups experienced a leukocytosis after exercise and especially with the higher intensity work (ride 3), it was expected that a uniform increase in the total number of lymphocytes by subset would occur independent of fitness group. The reasons for this differential numerical response by fitness groups to exercise are not known. Factors such as differences in the absolute work loads (but not relative work
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
260
EXERCISE
INTENSITY
AND
DURATION
loads) affecting lymphocyte trafficking from tissue reservoirs and differences in perceived stress within fitness groups may be involved. The contribution of endocrine factors in driving the observed changes in the percentage of lymphoid cells was also addressed in this study. Lymphocytes have cy- and P-adrenergic receptors (12). E is a classic stress hormone with a blood concentration that increases with exercise at >60% vozrnax (3). Our data confirm this, with levels of E and NE increasing markedly immediately after exercise of high intensity (75% VOWm8X,60 min) or moderate intensity for long durations (65% Tj02 max, 120 min) relative to preexercise values. Although it is tempting to suggest that the changes in catecholamine levels resulted in the lymphocyte percentage changes, plasma concentrations of E and NE do not provide information about tissue effects. Furthermore, catecholamine samples taken after exercise do not always accurately reflect catecholamine values during exercise. Plasma E returns to basal levels quickly after short-term maximal exercise; 1 min postexercise the concentration of arterialized plasma E decreases by 35% and the half time is calculated to be -2-3 min (12). Therefore, it is not surprising that the temporal relationships between catecholamines and lymphocyte subset percent changes were weak. In summary, the findings of this study suggest that the lymphocyte phenotype percent shifts with aerobic work are strongly affected by the intensity of exercise but not by fitness level. The total number of lymphocytes in each subset in response to exercise is, however, differentially affected by fitness level. The numerical increases in lymphocyte subsets also tended to be larger with increasing intensity of work. Differences between the total number and relative percent for a given subset serve to emphasize the importance of utilizing more than one measure of immune status in interpreting exercise-immunology studies. These findings also raise questions about the biological significance of transient shifts in blood lymphocyte subsets after single episodes of work. The direction, magnitude, and meaning of numerical changes in blood lymphocytes with exercise need to be assessed in the broader context of other concurrent physiological stresses (e.g., work under environmentally challenging conditions), with reference to lymphocyte recirculation kinetics, and in relationship to functional measures of immune status. We thank Drs. J. Randall Simpson and P. Pauls for assistance with this project. This research was supported by the Department of National Defence (Canada) and National Sciences and Engineering Research Council of Canada Grant 7645. Address for reprint requests: L. Hoffman-Goetz, Dept. of Health Studies, University of Waterloo, Waterloo, Ontario NZL 3G1, Canada. Received
6 October
1989; accepted
in final
form
8 February
EFFECTS
3.
8.
9.
10.
11. 12.
13.
14.
15.
16.
17.
18. 19.
20. 21.
22.
23.
1990.
REFERENCES 1. AHLBORG, B., AND G. AHLBORG. Exercise leukocytosis with and without beta-adrenergic blockade. Acta Med. Stand. 187: 241-249, 1970. 2. BERK, I,. S., D. NEIMAN, A. TAN, S. NEHLSEN-CANNERELLA, J. KRAMER, W. C. EBY, AND M. OWENS. Lymphocyte subset changes
24. 25.
ON
LYMPHOCYTES
during acute maximal exercise (Abstract). Med. Sci. Sports Exercise 18: 706, 1986. BLOOM, S. R., R. H. JOHNSON, D. M. PARK, M. J. RENNIE, AND W. R. SULAIMAN. Differences in the metabolic and hormonal response to exercise between racing cyclists and untrained individuals. J. Physiol. Lond. 258: 1-18, 1976. BOYUM, A. Isolation of mononuclear cells and granulocytes from blood. Stand. J. Clin. Lab. Invest. Suppl. 97: 77-89, 1968. BUNT, J. C. Hormonal alterations due to exercise. Sports Med. 3: 331-345,1986. BUTCHER, E. C. The regulation of lymphocyte traffic. Curr. Top. Microbial. Immunol. 128: 85-92, 1986. CRARY, B., M. BORYSENKO, D. C. SUTHERLAND, I. KUTZ, J. Z. BORYSENKO, AND H. BENSON. Decrease in mitogen responsiveness of mononuclear cells from peripheral blood after epinephrine administration in humans. J. Immunol. 130: 694-697, 1983. DEUSTER, P. A., G. P. CHROUSOS, A. LUGER, J. E. DEBOLT, L. L. BERNIER, U. H. TROSTMANN, S. B. KYLE, L. C. MOTGOMERY, AND D. L. LORIAUX. Hormonal and metabolic responses of untrained, moderately trained, and highly trained men to three exercise intensities. Metabolism 38: 141-148, 1989. EDWARDS, A. J., T. H. BACON, C. A. ELMS, R. VERARDI, M. FELDER, AND S. C. KNIGHT. Changes in the populations of lymphoid cells in human peripheral blood following exercise. Clin. Exp. Immunol. 58: 420-428,1984. GALUN, E., R. BURSTEIN, E. ASSIA, I. TUR-KASPA, J. ROSENBLUM, AND Y. EPSTEIN. Changes in white blood cell count during prolonged exercise. Int. J. Sports Med. 8: 253-256, 1987. KEAST, D., K. CAMERON, AND A. R. MORTON. Exercise and the immune response. Sports Med. 5: 248-267, 1988. KJAER, M., P. A. FARRELL, N. J. CHRISTENSEN, AND H. GALBO. Increased epinephrine response and inaccurate glucoregulation in exercising athletes. J. Appl. Physiol. 61: 1693-1700, 1986. LANDMANN, R. M. A., F. B. MULLER, C. H. PERINE, M. WESP, P. ERNE, AND F. R. BUHLER. Changes in immunoregulatory cells induced by psychological and physical stress: relationship to plasma catecholamines. Clin. Exp. Immunol. 58: 127-135, 1984. LEWICKI, R., H. TCHORZEWSKI, E. MAJEWSKA, Z. NOWAK, AND Z. BAJ. Effect of maximal physical exercise on T-lymphocyte subpopulations and on interleukin-1 (IL-l) and interleukin-2 (IL2) production in vitro. Int. J. Sports Med. 9: 114-117, 1988. MASUHARA, M., K. KAMI, K. UMEBAYASHI, AND N. TATSUMI. Influence of exercise on leukocyte count and size. J. Sports Med. 27: 285-290, 1987. MACKINNON, L. T., T. W. CHICK, A. VAN As, AND T. B. TOMASI. Effect of exercise on secretory and natural immunity. Adv. Exp. Med. Biol. 216: 869-876, 1986. OSHIDA, Y., K. YAMANOUCHI, S. HAYAMIZU, AND Y. SATO. Effect of acute physical exercise on lymphocyte subpopulations in trained and untrained subjects. Int. J. Sports Med. 9: 137-140, 1988. OYSTER, N. Changes in plasma eosinophils and cortisol of women in competition. Med. Sci. Sports Exercise 12: 148-159, 1980. SHOUTEN, W. V., R. VERSHUUR, AND H. C. G. KEMPER. Physical activity and upper respiratory tract infections in a normal population of young men and women: the Amsterdam growth and health study. Int. J. Sports Med. 9: 451-455, 1988. SIMON, H. B. Exercise and infection. Physician Sportsmed. 15: 135-141,1987. TERJUNG, R. L. Endocrine responses to exercise. In: Exercise and Sport Science Reviews, edited by R. Hutton and D. Miller. Philadelphia, PA: Franklin Institute, 1979, vol. 7, p. 153-180. TBNNESON, E., N. J. CHRISTENSEN, AND M. M. BRINKLOV. Natural killer activity during cortisol and adrenaline infusion in healthy volunteers. Eur. J. Clin. Invest. 17: 497-503, 1987. TVEDE, N., B. K. PEDERSEN, F. R. HANSEN, T. BENDIX, L. D. CHRISTENSEN, H. GALBO, AND J. HALKJAER-KRISTENSEN. Effect of physical exercise on blood mononuclear cell subpopulations and in vitro proliferative responses. Stand. J. Immunol. 29: 383-389, 1989. VAN BEAUMONT, W. Evaluation of hemoconcentration from hematocrit measurements. J. Appl. Physiol. 31: 712-713, 1972. WEICKER, H., M. FERANDI, H. HAGELE, AND R. PLUTO. Electrochemical detection of catecholamine in urine and plasma after separation with HPLC. Clin. Chim. Acta 141: 17-25, 1984.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
KENDALL, ADAM, LAURIE HOFFMAN-GOETZ, MICHAEL HOUSTON, BRIAN MACNEIL, AND YOGA ARUMUGAM. Exercise and blood lymphocyte subset responses: intensity, duration, and subject fitness effects. J. Appl. Physiol. 69(l): 2X-260, 1990.This study examined the effect of exercise intensity and duration on the percent blood lymphocytes in men of low [LF; maximal 0, uptake (vo2 ,,,) < 50 ml. kg-‘. min-l and sedentary], moderate (MF; vo2 max = 50-60 ml. kg-l’min-’ and recreationally active), and high (HF; vo2 max > 60 ml kg-‘. min-’ and recent training history) fitness. Thirty healthy adult men (aged 20-31 yr) participated in four randomly ordered cycle ergometer rides: ride 1 (65% VO, m8X, 30 min), ride 2 (30% VO 2 max9 60 min), ride 3 (75% VOgmax, 60 min), and ride 4 (65% VO 2 max9 120 min). Blood samples were drawn at various times before and after the exercise sessions. Lymphocyte subsets were determined by flow cytometry using monoclonal antibodies for total T (CD3+), T-helper (CD4+), and T-suppressor (CDS’) lymphocytes and for a subset of cells expressing a natural killer (NK) cell antigen (Leu7+). Plasma catecholamines were assayed to determine exercise stress. There were sharp reductions (P < 0.01) in the percentage of pan-T and T-helper lymphocytes immediately after exercise across all fitness levels; the magnitude of this reduction was greatest after the highest intensity (ride 3) or longest duration (ride 4) work. In contrast, the absolute number of T and T-helper cells tended to increase after exercise and significantly so in the HF subjects (P < 0.005). There was no significant effect of exercise or subject fitness category on the percentage of T-suppressor lymphocytes, although the absolute numbers of this subset increased significantly after exercise in LF subjects. Marked increases (P c 0.01) in the percentage of NK cells occurred immediately after exercise at all intensities and durations tested; numerical increases in total NK cells were significant in all fitness groups after the highest intensity work (ride 3; P c 0.005). Irrespective of whether the changes were expressed as percentage or total numbers, recovery to base line occurred at 30 min after exercise. The results suggest that the exercise effect on blood lymphocyte subset percentages in men is transient and occurs across all fitness levels. Concomitant changes in plasma catecholamine concentrations are only weakly associated with these lymphocyte subset percentage responses to exercise. Furthermore, this study shows that the exercise-induced changes in lymphocyte percentages do not consistently reflect changes in the absolute numbers of cells. l
lymphocyte cholamines
phenotypes;
submaximal
work; conditioning;
cate-
MANY INDIVIDUALS believe that regularly performed exercise helps improve resistance to infection, but only limited epidemiological data support this claim (20). 0161-7567/90
$1.50
Copyright
0 1990
Exercise is a form of physical stress because it releases a number of hormones in common with stereotypical stress responses (21). During exercise the hormones released are dependent on the relative intensity of exercise (8). Therefore, by manipulating the intensity and duration of an exercise bout, it is theoretically possible to manipulate the concentration of a specific hormone in the blood. Many of the hormones released during exercise have immunoregulatory properties. For example, elevations in blood catecholamines during exercise may mediate the immunological changes during exercise (1, 13). Immunosuppression has been observed after in vivo administration of epinephrine (7). In vivo administration of epinephrine has also been reported to upregulate natural killer (NK) cell activity and number in humans (22). Recirculation or lymphocyte traffic patterns may alter subset proportions in the lymphoid tissue, and this recirculation may be differentially affected by exercise and the attendant hormonal changes (6). Studies in exercise immunology are often limited by their experimental design. Exercise protocols have ranged from the 5-min stair climb (9) to the 24-h military march (10). Intensity levels have ranged from recreational tennis (18) to intense endurance bicycling (16). Keast et al. (11) emphasized that exercise protocols need to be better controlled with respect to exercise duration and intensity to explore the role of exercise stress on the immune system. The purpose of this study was to evaluate the effect of exercise duration, intensity, and fitness level on the percentage and numerical responses of human lymphocytes after single bouts of exercise. The underlying hypothesis was that the shorter the duration and/or intensity of the work, the smaller the impact on the percentage change in lymphocyte subsets. Conversely, the longer and more intense the work, the greater the impact on lymphocyte subsets. Accordingly, we determined the effects of four different bicycle ergometer rides on changes in the percentage of peripheral blood T-lymphocytes, Tlymphocyte subsets (helper/inducer and cytotoxic/suppressor), and large granular lymphocytes, including NK cells, by using monoclonal antibodies and flow cytometry. Additionally, we wanted to determine whether the changes in lymphocyte percentage were related to concurrent changes in plasma epinephrine and norepinephrine concentrations. the American
Physiological
Society
251
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
252
EXERCISE
INTENSITY
AND
DURATION
MATERIALSANDMETHODS
LS’ubJects
Thirty young adult males were selected from the university population. All subjects were administered a selection criterion, activity inventory questionnaire, and an informed consent approved by the Office of Human Research. The subjects were recruited in four groups. Group 1 (n = 8) were control subjects and did not participate in any exercise testing. Group 2 (n = 8) were individuals recruited for their lack of training history and involvement in sport; they were referred to as a sedentary low-fit sample. Low-fit subjects were also classified as having maximal aerobic capacities (VO, max) ~50 ml OZ. min-’ kg body wt-‘. Group 3 (n = 6) was classified as moderately active, with Tj02max in the range of 50-60 ml OZ. min-’ . kg-‘. These subjects had an activity profile of recreational athletes with no formal exercise training history. Group 4 (n = 8) was defined as very active WO2max> 60 ml 02. min-’ . kg-‘). This group comprised high-fit individuals who had a recent history of active involvement in endurance exercise training. To ensure compliance, all subjects received an honorarium on completion of the study. To aid in fitness classification, the subjects completed an activity profile questionnaire. This questionnaire requested the type and frequency of activities performed the week before the study. A composite score was calculated based on the relative energy expenditure above basal metabolic rate; this was termed metabolic equivalents (METS) per week. The composite score equaled the time spent per week on a particular activity multiplied by the average energy expenditure for that activity. This method of determining recent physical activity was adopted and modified from Shouten et al. (19). l
Experimental Design
Each subject performed two iTOpmax tests before the exercise tests. All exercise sessions and VOW maxtests were performed on an electronically braked bicycle ergometer (Quinton model 870); subjects were required to maintain a pedaling frequency of 60 pedal revolutions per minute. Each V02maxtest began with 4 min of loadless pedaling to allow the subjects to warm up and adjust to the equipment. Power output was subsequently increased by 30 W/min until exhaustion. The bicycle ergometer was modified to include an analog-integrating circuit to produce a linear increase in power output or a ramp signal. Maximal test or exhaustion was defined by one or more of the following: a plateau or decline of oxygen uptake (VO,), a respiratory exchange ratio in excess of 1.15, a maximal heart rate consistent with the subject’s agepredicted maximal heart rate, and the subject’s inability to maintain the desired pedaling frequency. Bicycle ergometry allows for precise control of power output over a wide range and is an almost completely concentric rhythmic activity that does not produce undue muscle or joint trauma. Heart rate was monitored by electrocardiography immediately before and every 2 min during the exercise and VOW maxtests. Four exercise sessions were randomly administered 1 wk apart: two bouts of cycling at a fixed duration of 60
EFFECTS
ON
LYMPHOCYTES
min at intensities of 30% (ride 2) and 75% (ride 3) of VO, maxand two bouts at a fixed intensity of 65% VOW max for durations of 30 (ride I) and 120 (ride 4) min. For each exercise session, the power output was increased progressively from 0 W to the desired level over 5 min, at which point the timer was started. VOW maxwas monitored at 5- to lo-min intervals to ensure that the appropriate work load was maintained. Blood samples were drawn from a forearm vein by venipuncture at the following times: 24 h before each exercise bout, 3 min before and 3 min after the termination of the ride, and 30 min, 2 h, and 24 h after exercise. The 3-min before and after samples will interchangeably be referred to as immediately preexercise and immediately postexercise in the text. Blood samples at rest were drawn after the subjects had relaxed in a chair for 15 min. The exercise bouts were initiated at 9 A.M. every week to minimize the effects of hormonal diurnal variation. Experimental Procedures
Mononuclear leukocytes were isolated from heparinized blood by a modified method of Boyum (4). Briefly, blood was diluted twofold with phosphate-buffered saline (PBS; 0.01 M, pH 7.2), layered over Histopaque (Sigma Chemical, St. Louis, MO), and centrifuged at 400 g for 30 min. The mononuclear cell interface was collected, washed three times, suspended in 1 ml of PBS, counted, and concentration adjusted to 2 x lo7 cells/ml. Fiftymicroliter aliquots of this cell suspension were incubated 45 min at 4°C together with 20 ~1 of each of the selected monoclonal antibodies. Labeled cell suspensions were fixed in 2% paraformaldehyde for analysis within 1 wk of sample collection. All mononuclear cell counts were performed on a Nikon Optiphot phase-contrast microscope, and cell viability was assessed by the trypan blue exclusion technique and was always >95%. Lymphocyte subpopulations were quantitated by the technique of direct immunofluorescence using monoclonal antibodies specific for peripheral total T-cells (CD3’), helper/inducer T-cells (CD4+), suppressor/cytotoxic T-cells (CD8+), and large granular lymphocytes, a subset of which expresses NK functional activity (Leu7+). Monoclonal antibodies were obtained from Becton Dickinson and used with a fluorescence-activated cell sorter (Coulter EPICS IV FACS) with an argon laser (488-nm wavelength) to excite the fluorescein-isothiocyanate-conjugated monoclonal antibodies. Background autofluorescence was determined for each sample, and an average of 10,000 cells per sample was counted. Hematocrit and Hemoglobin
Venous whole blood was collected in EDTA and was used for the determinations of hematocrit and hemoglobin content. For hematocrit determination, blood was centrifuged in heparinized microcapillary tubes at 11,500 revolutions/min for 6 min. Samples were determined in triplicate. Hemoglobin was determined by adding 6.0 ml cyanomethemoglobin reagent to 20 ~1 of the whole blood mixed in EDTA. This was then vortexed and left to stand for 20 min. Absorbance was read at 540 nm against
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
EXERCISE
Group
n
C LF MF HF
8 8 6 8
Age, yr 24.3k0.3 26.1kl.O 25.71k1.3 23.1k0.5
INTENSITY
Height, cm
Weight, kg
VO 2 maxp ml. min-’ kg-’
176.6k3.1 176.5k3.3 18O.lk2.6 176.9t1.7
78.1t2.1 71.2k4.4 77.1t2.9 71.423.4
52.4t2.3* 44.9k1.5t 55.2tl.6” 63.3+1.8$
AND
.
DURATION
MET Score, per wk
2,475+561 1,103&296 1,457+308 3,138+676
EFFECTS
ON
LYMPHOCYTES
253
difference between groups for MET score; however, the low-fit group had the lowest MET score (1,103 t 296 METs/wk), and the high-fit group had the highest (3,138 t 676 METs/wk) MET score. The control group was in the middle with a score of 2,475 t 561 METs/wk. Lymphocyte
Subset Percentages
Total T lymphocytes (CD3+). Total T-lymphocyte percentage responses after exercise are shown by group in Fig. 1, A-D. Across all fitness groups, the 30% Vozrnax ride elicited the smallest decrease in CD3+ cells after exercise and was significantly (F3,78 = 5.02, P < 0.003) different from all other rides. The immediately postexercise samples had significantly (F5,130 = 41.95, P < 0.0001) lower percentages of CD3+ cells compared with a reagent blank, and hemoglobin concentration values other sample time points when the results were pooled were determined from a standard curve. for all levels of fitness. The main time effects were significant in the moderate- (Fig. lC, F5 15 = 46.21, P < Catecholamine Determination 0.0001) and high- (Fig. lD, F5 15 = 25.95,‘P c 0.0001) fit Aliquots (1.5 ml) were extracted from ethylene gly- populations, whereas there was no statistically significol-bis(P-aminoethyl ether)-N,N,N’,N’-tetraacetic acid- cant decrease in CD3+ cells in the control (Fig. 1A) and chelated plasma for catecholamine determination and low-fit populations (Fig. IB). The decrease in CD3+ stored in Eppendorf tubes at -90°C until assayed. The phenotype occurred immediately postexercise and in all concentrations of norepinephrine (NE) and epinephrine cases had returned to base line by 30 min postexercise. (E) were analyzed by high-performance liquid chromaT-suppressor lymphocytes (CD8+). The T-suppressor tography using a modification of the method of Weicker lymphocyte population was highly resistant to changes and co-workers (25). Catecholamine concentrations were in percentage after all exercise treatments. There was no adjusted to reflect exercise-induced changes in plasma overall ride main effect, nor were there any differences volume using the van Beaumont equation (24). between fitness groups in their response to exercise. The t tests contrasting time within each ride for each group revealed no differences in T-suppressor percentage when Statistical Analysis compared with the preexercise sample. Data were analyzed by three-way analysis of variance T-helper lymphocytes (CD4+). Figure 2 shows the T(ANOVA) for significant between-group main effects, by helper lymphocyte percentage response to exercise in a series of two-way ANOVA for within-group main ef- men of defined fitness levels. Across all fitness groups, fects, and by Pearson correlation analyses. Specific dif- the least intense ride (ride 2 at 30% VOWmax) produced ferences between ride and time points were analyzed post the smallest reduction in the percentage of CD4+ lymhoc by Student-Newman-Keuls (SNK) test using an cy phocytes (F~,T~s= 6.04, P < 0.0009) compared with all = 0.01. Student’s t tests with CY= 0.005 were employed other rides. The T-helper lymphocyte percentage was to determine time differences within each ride; these significantly (F5,130 = 49.84, P c 0.0001) depressed only conservative cylevels were chosen to reduce the possibilimmediately postexercise, irrespective of fitness group. ity of type I error. The T-helper lymphocyte response of the low-fit subjects to exercise is shown in Fig. 2B. The mean T-helper RESULTS lymphocyte percentage fell by 23%, when averaged across all rides. In this group, the ride responses did not differ, Subject Characteristics as there was a significant reduction in CD4+ cells even The subject characteristics for each fitness group are after ride 2. Fig. 2C illustrates the T-helper lymphocyte presented in Table 1. There was no difference between percentage changes as a result of exercise for the modthe mean age, height, or weight of individuals in the erate-fit group. CD4+ cell percentage dropped significantly (P < 0.005) immediately postexercise in rides 1, control, low-fit, moderate-fit, or high-fit categories. There was, however, a significant (8& = 17.19, P < 3, and 4. Contrasts comparing immediately preexercise 0.0001) difference between the VOW,,, of the three exer- to 2 h postexercise revealed a significant difference from cising groups. The low-fit group had the lowest Vozrnax base line for ride 3. The CD4’ cell percentage response of the high-fit individuals to exercise is seen in Fig. 20. with a mean of 44.9 t 1.5 ml OZomin-’ kg-‘. The modA significant (F5,15 = 27.50, P < 0.0001) reduction in Terate-fit group was next highest with a mean of 55.2 t 1.6 ml OZ. min-’ . kg-’ for six subjects. The high-fit pophelper cell phenotype occurred when averaged across all ulation had the highest VO 2 maxwith a mean of 63.3 t 1.8 rides immediately postexercise; this reduction was -26%) and values returned to base line by 30 min postexercise. ml OZ. min-’ . kg? The control group’s vOzmax (52.4 t 2.3 ml 020min-‘o kg-‘) was not significantly different Ride 2 had a significantly (Fs,zl = 5.55, P < 0.0057) smaller effect on T-helper cell percentage than did ride from that of the moderate-fit group but had a different 3. mean from that of the low- and high-fit groups. Pairwise Large granular l.ymphoc-ytes with NK activity (Leu7+). comparisons using SNK at CY= 0.01 did not reveal anv Values are means t SE; n, no. of subjects. MET score was calculated based on activity questionnaires for the week before the study. This measure classifies physical activities according to energy expenditure independent of body weight as described in Ref. 19. C, control; LF, low fitness; MF, moderate fitness; HF, high fitness. *t"f Different at P < 0.01 by SNK test except where symbols are the same.
l
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
254
EXERCISE
INTENSITY
AND
DURATION
EFFECTS
A 90
ON
LYMPHOCYTES
B 0 Ride1
Ride2
l
0 Ride3
90
B Ride4
0 Ride1
l
Ride2
•I Ride3
n
Ride4
40 0 -24h
-3min
0 Ride1
+3min TIME
l
Ride2
+30min
0 Ride3
+2h
+24h
-24h
u Ride4
-3min
0 Rtdel
0
l
+3min TIME
Ride2
+30min
0 Ride3
+2h
+24h
m Ride4
0 -24h
-3min
+3min TIME
+30min
+2h
+24h
-24h
-3min
+3min TIME
FIG. 1. Percent pan-T lymphocytes (CD3’) of total mononuclear cells in control (A), low- (B), moderate(C), high-fit (D) subjects after ride treatments. Lymphocytes isolated by flow cytometry with argon laser set at 488 Ride 1 = 30 min, 65% Vozmax; ride 2 = 60 min, 30% Vo2 max; ride 3 = 60 min, 75% VOW max; ride 4 = 120 min, Abscissa is not time scaled. Percentage based on 10,000 cells. * Significant difference compared VO 2 max. immediately preexercise (P < 0.005).
Figure 3, A-D, presents the NK cell percentages as a function of exercise and fitness group. There was a significant (&26 = 6.92, P < 0.0014) difference between groups in the response of NK cells to exercise; the lowfit population had a greater mean percentage of Leu7+ cells compared with the other groups. This was possibly because of the different absolute numbers of these cells between fitness populations at rest. The low-fit subjects had -78% greater NK cells at rest, compared with the results for the other fitness categories. There was no difference between rides in terms of their effect on circulating Leu7+ cell percentage, even though ride 2 elicited the smallest increase in the NK phenotype. There was, however, a significant (F5,130= 75.25, P < 0.0001) increase in NK cell percentage immediately postexercise. All rides in the low-fit group were associated with the same magnitude level of NK cell percentage increase. A
+30min
+2h
+24h
and nm. 65% with
significant (F5,15 = 48.28, P < 0.0001) increase in NK cell percentage occurred immediately postexercise; a 68% increase in circulating NK phenotype was found when averaged across all rides. This relative increase was transient and returned to base-line levels by 30 min postexercise. Analysis by individual ride reinforced the significant (P c 0.005) increase immediately pre- to immediately postexercise for all rides. Fig. 3C represents the NK response to exercise in the moderate-fit subjects. Similar to the low-fit subjects, there was no significant difference between rides for the pattern of NK percentage increase in moderate-fit subjects. Likewise, there was a significant (F5 15= 57.02, P < 0.0001) increase of 50% in the cells bearing the NK phenotype immediately postexercise. Evaluation of individual ride effects showed that rides 1, 2, and 4 were significantly different from base line at 3 min postexercise. Ride 3 did not reach statistical signif-
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
EXERCISE
AND
DURATION
EFFECTS
A
60
Ride2
l
0 Ride3
n
ON
0 Ride1
Ride4
T 9 n -0
50
255
LYMPHOCYTES
B
60 0 Ride1
T d n -0 a g
INTENSITY
Ride2
l
0 Ride3
l
Ride4
50
a 40
iii I t%
a" cd I I%
30
20
40
30
20
un
0
-24h
-3min
+3min
+30min
+2h
-24h
+24h
-3min
+3min
TIME 60
T d n -0
l
Ride2
o Ride3
n
60
Ride4
ii I t%
+24h
D o
50
a e
+2h
TIME
C 0 Ride1
+30min
40
+^ d n 0
50
a g
40
iii I I%
30
20
Ridel‘
l
Ride2
0 Ride3
I) Ride4
30
20
0
0 -24h
-3min
+3min
+30min
+2h
+24h
-24h
-3min
TIME
icance at cy = 0.005, possibly because of large standard deviations in the response. NK cell responses to the exercise treatments in the high-fit individuals are given in Fig. 30. Again, there was no significant ride difference, but there was a highly significant (8’S,15 = 47.99, P c 0.0001) increase in NK cell percentage immediately after exercise. With post hoc pairwise comparison, rides 1, 2, and 3 were significantly different from base line (P < 0.005 in each case). Lymphocyte Subset Total Numbers
The total numbers fraction are shown (immediately before were no significant
of lymphocytes in the mononuclear by subset, fitness group, and ride and after exercise) in Table 2. There effects of time on the lymphocyte
+30min
+2h
+24h
TIME
2. Percent T-helper lymphocytes (CD4’) of total mononuclear cells in control (A), low- (B), and high-fit (D) subjects after ride treatments. Lymphocytes isolated by flow cytometry with argon nm. Ride 1 = 30 min, 65% V02max; ride 2 = 60 min, 30% V02max; ride 3 = 60 min, 75% VO* max; ride 4 VO 2 mnx- Abscissa is not time scaled. Percentage based on 10,000 cells. * Significant difference immediately preexercise (P < 0.005). FIG.
+3min
moderate(C), laser set at 488 = 120 min, 65% compared with
numerical responses in the controls. The total number of CD3+ (pan-T) lymphocytes increased after exercise at intensities of 65% VO 2MBXor higher in all fitness groups. However, these increases were significant (P c 0.005) only after ride 1 (all groups) and ride 4 (low- and highfit subjects). The same pattern of increase in the total number of CD4+ (T-helper) lymphocytes after exercise occurred; this increase was significant after rides 3 and 4 in the high-fit subjects (P < 0.005 vs. preexercise). In contrast to the generally flat percentage response of CD8+ (T-suppressor) lymphocytes with exercise, the absolute numbers increased significantly (P < 0.005) after work at intensities of 65% 7j02,,, or greater in the low-fit subjects; the total number of T-suppressor lymphocytes was also significantly elevated (P c 0.005) in the high-fit subjects after the highest-intensity work
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
256
EXERCISE
INTENSITY
AND
DURATION
A
EFFECTS
30 o Ride1
Ride2
l
0 Ride3
n
ON
LYMPHOCYTES
YB 0 Ride1
Ride4
l
Ride2
q Ride3 / 4P \\
Ride4
l
it
20
10
tu Z M 0
0 -24h
-3min
+3min
+30min
+2h
-24h
+24h
-3min
+3min
+30min
+2h
+24h
+2h
+24h
TIME
TIME
D
C 0 Ride1
l
Ride2
-24h
-3min
q
Ride3
+3min
n
+30min
0 Ride1
Ride4
+2h
-24h
+24h
l
-3min
Ride2
0 Ride3
+3min
l
+30min
Ride4
TIME
TIME
FIG. 3. Percent total mononuclear cells bearing natural killer (Leu7+) phenotype in control (A), low- (B), moderate(C), and high-fit (D) subjects after ride treatments. Lymphocytes isolated by flow cytometry with argon laser set at 488 nm. Ride I = 30 min, 65% VOW,,,; ride 2 = 60 min, 30% $70~ max; ride 3 = 60 min, 75% voz max; ride 4 = 120 min, difference compared with 65% ire, max. Abscissa is not time scaled. Percentage based on 10,000 cells. * Significant immediately preexercise (P < 0.005).
(ride 3). There was no differential effect of fitness level on the total number of NK (Leu7+) cells after work at 75% vo, max(ride 3): all groups had a significant increase in the total number of Leu7+ cells after exercise (P < 0.005). Exercise at intensities of 65% VO, max or less resulted in a more variable response across fitness groups. Low- and high-fit subjects had significant increases (P < 0.005) in the number of NK cells; moderatefit subjects followed the same pattern of increase after exercise, albeit not significantly. Total Mononuclear
other sampling intervals. The ride effect was due to the greater increases in leukocyte number after rides 3 and 4. These increases were on the order of two- to threefold above basal levels. For example, in ride 3 the increases were from 5.6 t 1.3 to 14.4 t 3.0 cells X 107/ml in highfit, from 4.1 t 0.2 to 9.4 t 1.4 cells x 107/ml in moderatefit, and from 2.7 t 0.2 to 5.4 t 0.5 cells X 107/ml in lowfit subjects, and in ride 4 the increases were from 5.3 t 1.1 to 12.8 t 2.9 cells X 107/ml in high-fit, from 3.7 t 0.5 to 8.8 t 1.3 cells X 107/ml in moderate-fit, and from 2.8 -+ 0.3 to 6.9 t 0.7 cells X 107/ml in low-fit subjects.
Cell Number
The total number of mononuclear cells recovered after density gradient centrifugation of whole blood was significantly increased as a function of time (& 130= 37.59, P < 0.0001) and ride (F3,78 = 3.97, P c O.Oi) (data not shown). The main effect of time was due to a marked leukocytosis immediately after exercise relative to all
Plasma
Catecholamines
Because catecholamines are released during exercise and have immunomodulatory as well as circulatory effects, plasma concentrations of E and NE were determined at the serial time points corresponding to immune measurements. The NE responses of subjects to the
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
EXERCISE
INTENSITY
AND
DURATION
EFFECTS
Lymphocyte Group
Ride
No.
CD3’ -3 min
C
1
2 3 LF
1
2 3 MF
1
2 3 HF
1
2 3
Subset,
ON
x 10’ cells/ml
CD4’ +3 min
-3
min
257
LYMPHOCYTES
CD8’ +3 min
-3
min
Leu7’ +3 min
-3 min
+3 min
3.6t0.3 4.6-e0.5
4.0t0.5 5.3tl.l
2.0t0.2 2.6k0.3
2.1t0.3 3.0t0.7
0.9kO.2
1.3t0.1
1.020.2 1.5kO.4
0.4kO.l 0.6kO.l
0.6kO.l 0.8t0.3
4.1t0.4 2.0k0.3 1.9-co.2 1.8k0.2 1.9t0.2 2.6t0.4 3.4k0.6 3.OkO.l 2.8k0.4 3.6t0.6 4.4t0.8 3.7t0.9 3.8t0.9
4.6-1-0.4 3.2t0.5* 2.3k0.3 3.3kO.4 4.0t0.6* 4.4&0.4* 3.1t0.4 5.1kO.7 5.321.0 5.3&0.8* 4.5k0.8 7.2tl.3* 6.8k1.4*
2.3t0.3 1.2t0.2 1.1kO.l l.ltO.l 1.1t0.1 1.6kO.2 2.1kO.4 1.9t0.6 1.7t0.3 2.3t0.4 2.9kO.7 2.4-e0.7 2.520.7
2.5t0.2 1.61!10.2* 1.2zko.2 1.4t0.1 2.0k0.3 2.6t0.3 1.9kO.3 2.8k0.4 2.8t0.4 3.1t0.5 2.8k0.6 4.0t0.8* 4.1*0.7*
1.lt0.2 0.7kO.l 0.6tO.l 0.6kO.l 0.7kO.l 0.7tO.l 0.9kO.l 0.7kO.l 0.7kO.l 0.9t0.2 1.3t0.2 l.OkO.2 1.0t0.2
1.3k0.2 1.2+0.2* 0.7gO.2 1.4t0.2* 1.5t0.3* 1.3t0.2 0.9kO.l 1.9kO.4 1.8t0.3 2.1kO.5 1.4t0.3 3.2&0.6* 2.7k0.7
0.5t0.1 0.4t0.2 0.3t0.1 0.4kO.O 0.4t0.1 0.2kO.O 0.3kO.l 0.2to.o 0.2tO.l 0.4kO.l 0.3t0.1 0.4kO.8 0.3t0.1
0.7t0.1 1.lt0.2* 0.6tO.l 1.5t0.2* 1.5t0.2* 1.420.4 0.6kO.l 2.0t0.6* 1.620.4 1.7t0.5 0.8&0.2* 2.7&0.6* 1.9t0.5"
Values are means t SE; SE values < been rounded up to single decimal place exercise (-3 min).
0.05 have been rounded down to single decimal (0.1). Lymphocyte subset is from mononuclear
exercise tests are shown in Fig. 4. There was a significant main effect of ride (F3 78 = 11.61, P < 0.0001) for plasma NE concentrations, with the longest duration ride (ride 4, 65% \jozmax for 2 h) or highest intensity ride (ride 3, 75% Topmax for 1 h) producing the largest increase in plasma NE. Plasma NE levels peaked immediately postexercise and returned to basal levels 2 h after completion of physical work (F15,130= 10.36, P < 0.0001 for main effect of time). There was virtually no change in plasma NE in control subjects (Fig. 4A) as a function of time. In contrast, low- (Fig. 4B), moderate- (Fig. 4C), and high-fit (Fig. 40) subjects demonstrated a sharp increase in plasma NE levels immediately postexercise and especially in ride 3 (P c 0.0005 in all cases); this highintensity ride effect on plasma NE occurred independently of fitness group. In addition, high-fit subjects had a significant NE peak immediately after ride 4. Figure 5 shows the within-group effect for changes in plasma E as a function of ride and time. Not surprisingly, there were overall significant main effects of group (Fs,zG = 4.00, P < O.Ol), ride (F378 = 10.77, P < O.OOOl), and the V&30 = 31.51, P < O.Obol) for plasma E. All groups, except controls, had an increase in plasma E immediately postexercise after work at high intensity (ride 3) and long duration (ride 4). However, because of large withingroup variation, the peak increase was only significant in the high-fit subjects (Fig. 5D). Pearson correlation analyses were done to determine the relationship between the change in a specific hormone (NE, E) pre- to postexercise and the change in the percentage of a given lymphocyte subset. There was no significant relationship between either E or NE and changes in CD3+ or CD8+ cells. The postexercise decrease in the percentage of CD4+ cells was correlated with the increase in NE levels postexercise (r = 0.38, P C 0.0003). Postexercise increases in NE were also correlated with the increase in the percentage of Leu7+ cells (r = 0.40, P c 0.0002). DISCUSSION
This study evaluated the effect of individual fitness level on the percentage and total number of T-lympho-
place (0.0) and SE values between 0.05 and 0.099 have cell-rich fraction. * P < 0.005 vs. immediately before
cyte subsets in blood after single bouts of exercise at controlled durations (60 and 120 min) and intensities (30, 65, and 75% VO Zmax). Thirty healthy adult males (aged 20-31 yr) were assigned to fitness categories based on functional aerobic capacity and physical activity profiles. All three fitness groups had significantly different Vo2 m8Xvalues. However, VO, maxrepresents continuous rather than categorical data; therefore, the cutoff points used for the minimum and maximum values for the fitness categories were, to some degree, arbitrary. Activity profiles (MET scores) were generated from questionnaires to aid in subject classification. The MET values for the four different groups were not significantly different. Subject recall, over and underestimation of activity, and subjectivity of the measures are clearly potential confounds when utilizing questionnaire methodology. Nevertheless, despite wide variations in MET scores across subjects, individuals in the low-fit group had the lowest average MET and those in the high-fit group had the highest MET (1,103 t 296 and 3,138 t 676, respectively). It was hypothesized that exercise of shorter duration and lower intensity would produce a smaller change in the percentage of T-lymphocyte subpopulations in peripheral blood, regardless of fitness level. This was hypothesized because the release of immunomodulatory hormones in the blood generally increases in proportion with relative exercise intensity and duration (5). Exercise that elicits less stress or immunomodulatory hormone release should be associated with smaller lymphocyte and leukocyte numerical and percentage shifts. Analysis of the percentage of lymphocytes in peripheral blood as a function of fitness level and exercise treatment revealed several consistent patterns. First, the T-suppressor (CD8+) cell percentage response was extremely resilient; there was minimal change after exercise of either high intensity (75% VOW max)or long duration (120 min). This insensitivity to exercise occurred irrespective of whether the subjects were of low, moderate, or high fitness. Some investigators have reported
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
258
EXERCISE
0 Ride1
l
INTENSITY
Ride2
0 Ride3
l
AND
DURATION
EFFECTS
ON
1600
-
Ride4
E
1400
-
1400
-
s
1200
-
1200
-
1000
-
1000
-
800
-
800
-
iii!
600
-
600
-
0 Z
400
400
-
200
-
-
E 2 w Z [L
a I m
.a*- . .-. . . . . .- --
-.--
200
P
a
5
I
*.
. . . . . . .--. . . -
-H-+-4=-3
0 ’
-l h U
-24h
-3min
+3min
t30min
+2h
+24h
a I to a 2
LYMPHOCYTES
0 Ride1
0 Ride2
-24h
-3min
+3min
Ride2
0 Ride3
l
a 0 Z a r v)
-
200
-
1400
a E
1000
2
a zl"
0 A -24h
-3min
+3min
+30min
+24h
+2h
+24h
TIME
Ride2
l
0 Ride3
n
Ride4
800 I
CL
a 2
n”
si -
600 400
0 Ride1
Ride4
z E loo0 g 800 a. w
+2h
D
1600
F
l
+30min
TIME
C
Ride1
m Ride4
0
TIME
0
0 Ride3
T
600
1
-I
0 -24h
-3min
+3min
+30min
+2h
+24h
TIME
FIG. 4. Norepinephrine concentration in plasma of control (A), low- (B), moderate(C), and high-fit (D) subjects after treatments. Norepinephrine determination was by high-performance liquid chromatography. Ride I = 30 min, 65% vo2 ,,,ax; ride 2 = 60 min, 30% VOW,,,; ride 3 = 60 min, 75% vo2 max;ride 4 = 120 min, 65% vo2 max. Abscissa is not time scaled. * Significant difference compared with immediately preexercise (P < 0.005).
significant changes in the percentage of CD8+ cells immediately after acute physical work at maximal intensities in humans (2), but many reports of the effect of exercise on the T-suppressor subset have been inconclusive. There have been relative increases (14, 15) or no net change (23). Our data indicate that at submaximal exercise intensities (~75% Vo2 max), the percentage of T suppressor cells in blood will not change. It is not known from these data whether higher intensity work (but still below maximal) or repetitive bouts of submaximal exercise will affect the relative percentage of this lymphocyte subset. Second, irrespective of subject fitness level, the percentages of CD3+ and CD4+ lymphocytes were reduced immediately after exercise. Furthermore, the magnitude of these reductions was influenced by exercise intensity and duration: in most cases, the higher intensity work (75% vo2,,, for 60 min) produced the largest drop in the percentage of these cells. Longer duration bouts of aerobic work (65% VO 2 1118X for 120 min) resulted in almost
as great reductions in the percentage of these cells. Exercise of the lowest intensity (30% I702 m8x for 60 min) resulted in some reductions in the percentage of these cells albeit not as dramatically or consistently as the other work sessions. Thus it appears that the higher the relative intensity of aerobic work, the greater the impact on the relative percentage of CD3+ and CD4’ lymphocytes. Subject fitness level does little to modify this basic effect. These results agree with Oshida et al. (17), who found similar decrements in T-helper lymphocyte percentage immediately after exercise. Their subjects were both trained and untrained and rode a bicycle ergometer for 2 h at 60% \jo2 max. These results also agree with Tvede et al. (23), who reported reductions in the CD3+ and CD4+ lymphocyte percentage after exercise, and they attributed the decrease in the CD3+ percentage to the decrease in the CD4+ subset. This conclusion is logical because the pan-T marker, CD3, is present on all T-lymphocytes. If the CD8+ subset remains constant during exercise, and there is a marked reduction in the
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
EXERCISE
INTENSITY
AND
DURATION
EFFECTS
w z
0 Ride2
0 Ride3
n
a w z L w < I
I
Ride4
0 Ride1
l
Ride2
T
0 Ride3
Ride4
l
w z
250
a I
259
LYMPHOCYTES
1 B
350 0 Ride1
ON
200
a I 2
-j
150 100
m
a
50
-a
2 0
i
’
-24h
se.........
*.........* --
f**
4.
&=---d -3min
+3min
+30min
+2h
+24h
-24h
-3min
+3min
TIME
z > -n w z “r tl w z a. w < I CA u
-
-
300
-
250
-
250
-
200
-
200
-
150
-
100
-
-
300
c 0 Ride1
l
Ride2
0 Ride3
l
D
Ride4
0 Ride1
.. *. .‘. ;: .. ...
150-
+24h
TIME 350
350
+2h
+30min
l
Ride2
:
loo50 -
T
•I Ride3
n
Ride4
‘..
50 -
2
-
0 ’
0 -24h
-3min
+3min
+30min
+2h
+24h
TIME
-24h
-3min
+3min
+30min
+2h
+24h
TIME
5. Epinephrine concentration in plasma of control (A), low- (B), moderate(C), and high-fit (D) subjects after treatments. Epinephrine determination by high-performance liquid chromatography. Ride I = 30 min, 65% ride 4 = 120 min, 65% vO2 max. Abscissa is not VO 2 max9 - ride 2 = 60 min, 30% vOzmax; ride 3 = 60 min, 75% vO2,,,; time scaled. * Significant difference immediately pre- to immediately postexercise (P < 0.005). FIG.
CD4’ population, then the CD3+ decrease should be due to the reduction in CD4 phenotype. There were similar percentage decrements in the CD4+ and CD3+ cell in this present study as well. Third, exercise at intensities ~65% vozrnax produced marked increases in the percentage of Leu7+ (a subset of NK) cells immediately after cessation of exercise. As with the T-cells, the magnitude of the percentage change was greatest after ride 3 (75% Vop maxfor 60 min). Surprisingly, even at exercise of low intensity (30% VOg max for 60 min in ride Z), NK cell percentage increased significantly. Although the clinical significance of the increase in the relative percentage of NK cells cannot be determined from this study, it is tempting to speculate that an increase may be potentially beneficial; NK cells have the capacity to lyse tumor targets. Because the significant increase in NK cells occurred after rides even at relatively low intensity (ride Z), the potential immunological benefits of changes in this parameter may be achieved at relatively low metabolic costs.
The results of this study show that at all fitness levels, the higher the intensity of work the greater the impact on the percentages of T lymphocyte subsets. In contrast, the impact of exercise and fitness level on the total number of lymphocytes in each subset was more variable. Although high-intensity or long-duration work produced dramatic increases in the number of virtually all lymphocytes in the high-fit subjects, moderate-fit subjects had increases only in the total number of NK cells and only after the high-intensity ride (ride 3); low-fit subjects had significant increases in the number of T, T suppressor, and NK cells (but not T-helper lymphocytes) after rides 3 and 4. Given that all fitness groups experienced a leukocytosis after exercise and especially with the higher intensity work (ride 3), it was expected that a uniform increase in the total number of lymphocytes by subset would occur independent of fitness group. The reasons for this differential numerical response by fitness groups to exercise are not known. Factors such as differences in the absolute work loads (but not relative work
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
260
EXERCISE
INTENSITY
AND
DURATION
loads) affecting lymphocyte trafficking from tissue reservoirs and differences in perceived stress within fitness groups may be involved. The contribution of endocrine factors in driving the observed changes in the percentage of lymphoid cells was also addressed in this study. Lymphocytes have cy- and P-adrenergic receptors (12). E is a classic stress hormone with a blood concentration that increases with exercise at >60% vozrnax (3). Our data confirm this, with levels of E and NE increasing markedly immediately after exercise of high intensity (75% VOWm8X,60 min) or moderate intensity for long durations (65% Tj02 max, 120 min) relative to preexercise values. Although it is tempting to suggest that the changes in catecholamine levels resulted in the lymphocyte percentage changes, plasma concentrations of E and NE do not provide information about tissue effects. Furthermore, catecholamine samples taken after exercise do not always accurately reflect catecholamine values during exercise. Plasma E returns to basal levels quickly after short-term maximal exercise; 1 min postexercise the concentration of arterialized plasma E decreases by 35% and the half time is calculated to be -2-3 min (12). Therefore, it is not surprising that the temporal relationships between catecholamines and lymphocyte subset percent changes were weak. In summary, the findings of this study suggest that the lymphocyte phenotype percent shifts with aerobic work are strongly affected by the intensity of exercise but not by fitness level. The total number of lymphocytes in each subset in response to exercise is, however, differentially affected by fitness level. The numerical increases in lymphocyte subsets also tended to be larger with increasing intensity of work. Differences between the total number and relative percent for a given subset serve to emphasize the importance of utilizing more than one measure of immune status in interpreting exercise-immunology studies. These findings also raise questions about the biological significance of transient shifts in blood lymphocyte subsets after single episodes of work. The direction, magnitude, and meaning of numerical changes in blood lymphocytes with exercise need to be assessed in the broader context of other concurrent physiological stresses (e.g., work under environmentally challenging conditions), with reference to lymphocyte recirculation kinetics, and in relationship to functional measures of immune status. We thank Drs. J. Randall Simpson and P. Pauls for assistance with this project. This research was supported by the Department of National Defence (Canada) and National Sciences and Engineering Research Council of Canada Grant 7645. Address for reprint requests: L. Hoffman-Goetz, Dept. of Health Studies, University of Waterloo, Waterloo, Ontario NZL 3G1, Canada. Received
6 October
1989; accepted
in final
form
8 February
EFFECTS
3.
8.
9.
10.
11. 12.
13.
14.
15.
16.
17.
18. 19.
20. 21.
22.
23.
1990.
REFERENCES 1. AHLBORG, B., AND G. AHLBORG. Exercise leukocytosis with and without beta-adrenergic blockade. Acta Med. Stand. 187: 241-249, 1970. 2. BERK, I,. S., D. NEIMAN, A. TAN, S. NEHLSEN-CANNERELLA, J. KRAMER, W. C. EBY, AND M. OWENS. Lymphocyte subset changes
24. 25.
ON
LYMPHOCYTES
during acute maximal exercise (Abstract). Med. Sci. Sports Exercise 18: 706, 1986. BLOOM, S. R., R. H. JOHNSON, D. M. PARK, M. J. RENNIE, AND W. R. SULAIMAN. Differences in the metabolic and hormonal response to exercise between racing cyclists and untrained individuals. J. Physiol. Lond. 258: 1-18, 1976. BOYUM, A. Isolation of mononuclear cells and granulocytes from blood. Stand. J. Clin. Lab. Invest. Suppl. 97: 77-89, 1968. BUNT, J. C. Hormonal alterations due to exercise. Sports Med. 3: 331-345,1986. BUTCHER, E. C. The regulation of lymphocyte traffic. Curr. Top. Microbial. Immunol. 128: 85-92, 1986. CRARY, B., M. BORYSENKO, D. C. SUTHERLAND, I. KUTZ, J. Z. BORYSENKO, AND H. BENSON. Decrease in mitogen responsiveness of mononuclear cells from peripheral blood after epinephrine administration in humans. J. Immunol. 130: 694-697, 1983. DEUSTER, P. A., G. P. CHROUSOS, A. LUGER, J. E. DEBOLT, L. L. BERNIER, U. H. TROSTMANN, S. B. KYLE, L. C. MOTGOMERY, AND D. L. LORIAUX. Hormonal and metabolic responses of untrained, moderately trained, and highly trained men to three exercise intensities. Metabolism 38: 141-148, 1989. EDWARDS, A. J., T. H. BACON, C. A. ELMS, R. VERARDI, M. FELDER, AND S. C. KNIGHT. Changes in the populations of lymphoid cells in human peripheral blood following exercise. Clin. Exp. Immunol. 58: 420-428,1984. GALUN, E., R. BURSTEIN, E. ASSIA, I. TUR-KASPA, J. ROSENBLUM, AND Y. EPSTEIN. Changes in white blood cell count during prolonged exercise. Int. J. Sports Med. 8: 253-256, 1987. KEAST, D., K. CAMERON, AND A. R. MORTON. Exercise and the immune response. Sports Med. 5: 248-267, 1988. KJAER, M., P. A. FARRELL, N. J. CHRISTENSEN, AND H. GALBO. Increased epinephrine response and inaccurate glucoregulation in exercising athletes. J. Appl. Physiol. 61: 1693-1700, 1986. LANDMANN, R. M. A., F. B. MULLER, C. H. PERINE, M. WESP, P. ERNE, AND F. R. BUHLER. Changes in immunoregulatory cells induced by psychological and physical stress: relationship to plasma catecholamines. Clin. Exp. Immunol. 58: 127-135, 1984. LEWICKI, R., H. TCHORZEWSKI, E. MAJEWSKA, Z. NOWAK, AND Z. BAJ. Effect of maximal physical exercise on T-lymphocyte subpopulations and on interleukin-1 (IL-l) and interleukin-2 (IL2) production in vitro. Int. J. Sports Med. 9: 114-117, 1988. MASUHARA, M., K. KAMI, K. UMEBAYASHI, AND N. TATSUMI. Influence of exercise on leukocyte count and size. J. Sports Med. 27: 285-290, 1987. MACKINNON, L. T., T. W. CHICK, A. VAN As, AND T. B. TOMASI. Effect of exercise on secretory and natural immunity. Adv. Exp. Med. Biol. 216: 869-876, 1986. OSHIDA, Y., K. YAMANOUCHI, S. HAYAMIZU, AND Y. SATO. Effect of acute physical exercise on lymphocyte subpopulations in trained and untrained subjects. Int. J. Sports Med. 9: 137-140, 1988. OYSTER, N. Changes in plasma eosinophils and cortisol of women in competition. Med. Sci. Sports Exercise 12: 148-159, 1980. SHOUTEN, W. V., R. VERSHUUR, AND H. C. G. KEMPER. Physical activity and upper respiratory tract infections in a normal population of young men and women: the Amsterdam growth and health study. Int. J. Sports Med. 9: 451-455, 1988. SIMON, H. B. Exercise and infection. Physician Sportsmed. 15: 135-141,1987. TERJUNG, R. L. Endocrine responses to exercise. In: Exercise and Sport Science Reviews, edited by R. Hutton and D. Miller. Philadelphia, PA: Franklin Institute, 1979, vol. 7, p. 153-180. TBNNESON, E., N. J. CHRISTENSEN, AND M. M. BRINKLOV. Natural killer activity during cortisol and adrenaline infusion in healthy volunteers. Eur. J. Clin. Invest. 17: 497-503, 1987. TVEDE, N., B. K. PEDERSEN, F. R. HANSEN, T. BENDIX, L. D. CHRISTENSEN, H. GALBO, AND J. HALKJAER-KRISTENSEN. Effect of physical exercise on blood mononuclear cell subpopulations and in vitro proliferative responses. Stand. J. Immunol. 29: 383-389, 1989. VAN BEAUMONT, W. Evaluation of hemoconcentration from hematocrit measurements. J. Appl. Physiol. 31: 712-713, 1972. WEICKER, H., M. FERANDI, H. HAGELE, AND R. PLUTO. Electrochemical detection of catecholamine in urine and plasma after separation with HPLC. Clin. Chim. Acta 141: 17-25, 1984.
Downloaded from www.physiology.org/journal/jappl by ${individualUser.givenNames} ${individualUser.surname} (129.186.138.035) on January 17, 2019.
