DISCUSSION
The main findings of this study were that prolonged, intensive exercise on consecutive mornings was associated with reduced antigen-stimulated proliferative responses before exercise on day 2 compared with day 1. These responses occurred despite no apparent effect of exercise on antigen-stimulated T-cell cytokine mRNA expression. In addition, CHO beverage ingestion attenuated the decrease in antigen-stimulated proliferative responses before exercise on day 2 but had no effect on T-cell cytokine mRNA expression. Furthermore, CHO ingestion was associated with greater T-cell proliferative responses after exercise on day 2 following mitogen stimulation only.
To the authors' knowledge, this is the first study to have assessed the acute effect of prolonged, intensive exercise on in vitro antigen-stimulated T-cell cytokine mRNA expression and proliferative responses in humans. It is also the first study to have assessed the role of CHO beverage ingestion on these antigen-specific responses. Several previous studies of acute, prolonged, strenuous exercise report a decrease in T-cell proliferative responses to a variety of mitogens, including the plant lectins PHA and concanavalin A (9, 12). It has been speculated that this may account for the observed higher incidence of respiratory infection in athletes involved in heavy schedules of training and competition. However, mitogens activate other cell types, including B cells and NK cells, in addition to a high percentage of circulating T cells (19). Therefore, mitogen stimulation may result in unrealistically large alterations in proliferative responses that may not accurately reflect specific responses to antigens of relevance to this population, such as influenza. The findings of the present study appear to support this; the pattern of change for influenza- and tetanus toxoid-stimulated proliferative responses was similar, yet, following exercise on day 2, CHO ingestion was found to influence proliferative responses to PHA only. It could be argued that differences in the magnitude of the proliferation response to PHA and antigen stimulation could be related to the dose of mitogen and antigen used. However, in the present study, this is unlikely to be the major influencing factor in the responses observed, because the concentration of PHA and each antigen used was designed to give an equated amount of proliferation, as determined from the results of an initial pilot study of resting young, healthy subjects. Caution should always be taken when extrapolating the findings of any in vitro measure, be it antigen or mitogen stimulated, to the in vivo situation. Nevertheless, these findings may highlight a need for some caution when extrapolating mitogen-stimulated T-cell responses to those of common pathogens and to a subsequent risk of infection.
In an attempt to gain an indication of the effect of exercise (and CHO) on the dominant T-cell response elicited, we also investigated T-cell expression of Type 1 and Type 2 cytokines using antigens that would predominantly stimulate a Type 1 (influenza) or a Type 2 (tetanus toxoid) T-cell response. Assessment of T-cell subset activation is challenging. In our study, secreted cytokine protein levels in culture supernatants were insufficient for measurement, and, therefore, we evaluated the relative Type 1 and Type 2 T-cell responses by real-time PCR for cytokine mRNA expression. One possible reason for the observed higher incidence of infections in elite athletes may be a change in the bias of the T-cell cytokine response for a particular antigen, in addition to a reduction in T-cell proliferation. Prolonged, strenuous exercise is associated with a decrease in the percentage of Type 1 T cells in the circulation (21), and shorter intensive exercise is associated with decreases in mitogen-stimulated T-cell release of IL-2 and IFN-γ (20). Therefore, we hypothesized that any reductions in antigen-stimulated proliferative responses would be associated with a decrease in Type 1 T-cell cytokine mRNA expression and a dominant Type 2 T-cell response. However, our findings do not support this: neither exercise (nor CHO ingestion) affected Type 1 or Type 2 T-cell cytokine expression, despite exercise-induced reductions in antigen-stimulated cellular proliferation, although it must be borne in mind that mRNA expression does not necessarily relate directly to synthesis and secretion of functional cytokine protein. Previous studies that have assessed T-cell cytokine release in response to exercise have used phorbol myristate acetate as a stimulating agent (20, 21); therefore, it could be speculated that our findings are reflecting a specific response to specific antigens. Unfortunately, we cannot confirm this since mitogen-stimulated T-cell cytokine mRNA expression was not assessed in this study.
On the other hand, there may be an alternative explanation for these observations: Green and Rowbottom (7) report that the observed reduction in mitogen-stimulated T-cell proliferation following intensive exercise is not due to a decrease in mitosis but rather an increase in the rate of cell death in culture; i.e., there is a reduction in the number of cells that are able to respond. In the present study, cellular proliferation and cytokine mRNA expression were assessed from cultures that contained a T-cell concentration of 106/ml for all samples collected. Therefore, any exercise-induced increase in the rate of cell death would reduce the number of responsive cells and could be expected to account for the observed decrease in antigen-stimulated proliferation before and after exercise on day 2 in the PLA trial. However, any alteration in the number of responsive cells in response to exercise would not be reflected in the observations for T-cell cytokine mRNA expression. This is reported as a ratio relative to the expression of mRNA for the T-cell-specific gene Cβ, which corrects cytokine mRNA expression for the relative presence of total viable T-cell mRNA. In this way, if there is a decrease in the expression of mRNA for a particular T-cell cytokine, the ratio will decrease, and vice versa. If increased rates of cell death are responsible for the alterations in proliferative responses observed here, they would simply not be reflected in the observations for T-cell cytokine mRNA expression, since dead cells will not express patent mRNA for either Cβ or the T-cell cytokines; i.e., the ratio would be unaffected. Nevertheless, an increase in the rate of cell death could still account for a decreased percentage of circulating Type 1 T cells following prolonged, intensive exercise (21).
Altered rates of cell death may also account for the finding of higher antigen- and mitogen-stimulated proliferative responses in CHO than in PLA before exercise on day 2, since Green et al. (8) found that regular CHO ingestion during exercise was associated with reduced rates of cell death within mitogen-stimulated cell cultures. Previously, the effect of CHO ingestion in preventing the reduction of PHA-stimulated T-cell proliferative responses to prolonged, intensive exercise has been associated with a concurrent attenuation of the cortisol response (9). However, the findings of our study and others (8) suggest that the role of cortisol is perhaps not as crucial as was previously thought, since the effects of CHO on antigen- (and mitogen-) stimulated T-cell proliferation occurred without any influence of CHO on plasma cortisol concentration. Nevertheless, the higher proliferative responses following antigen stimulation reported here may reflect a protective effect of CHO for subsequent infection risk, since a recent study in horses reported decreased T-cell proliferative responses to equine influenza virus following 5 days of strenuous exercise (5). These horses also demonstrated increased susceptibility to influenza following a later challenge exposure to the same strain of the virus.
The effect of exercise performed on consecutive mornings was assessed, since many athletes, particularly games players, typically train only once a day. Recently, playing a game of soccer on consecutive days reduced the expression of lymphocyte adhesion and signaling molecules 6 h after the second game (11). However, this study did not take any measurements after the first game or immediately before the second game, making it difficult to assess whether there was any carryover effect from the first game. In PLA in the present study, influenza- and tetanus toxoid-stimulated proliferative responses were lower before exercise on day 2 compared with the same time on day 1. Furthermore, our findings suggest that CHO ingestion may help to diminish this effect. These responses cannot be accounted for by alterations in the numbers of circulating CD3+ cells, since proliferation was assessed in a constant concentration of T cells for all samples.
In an attempt to overcome previous concerns (7) regarding the assessment of exercise-induced changes in T-cell proliferation using cultures containing a constant number of PBMCs, we assessed antigenic and mitogenic proliferative responses in cultures of PBMCs containing a constant number of T cells. This aimed to overcome potential concerns arising from the interpretation of post hoc mathematical adjustments to proliferation data from mixed-cell cultures as changes in T-cell function (7). However, this methodology itself gives rise to a number of limitations that must be acknowledged when interpreting these findings. Although the cultures contained constant numbers of T cells, there would have been alterations in other cell populations (e.g., monocytes and B cells) present in the cultures that could contribute to the overall proliferative response observed. Whereas PHA is a potent stimulant of T cells because it binds mainly to CD3 (19), we were not able to determine which cells are actively proliferating in response to the specific a