Changes in natural killer cell subpopulations over a winter training season in elite swimmers

Luís Rama, Ana Maria Teixeira, Alice Matos, Grasiely Borges, Ana Henriques, Michael Gleeson, Susana Pedreiro, Edith Filaire, Francisco Alves, Artur Paiva

Immune changes and increased susceptibility to infection are often reported in elite athletes. Infectious episodes can often impair training and performance with consequences for health and sporting success. This study monitored the occurrence of episodes of upper respiratory symptoms (URS) and the variation in circulating NK cells, CD56(bright) and CD56(dim) NK cells subpopulations, over a winter swimming season. Nineteen national elite swimmers and 11 non-athlete controls participated in this study. URS episodes were monitored using daily log books. Blood samples were taken at rest at four time points during the season: before the start of the season (t1--middle September), after 7 weeks of an initial period of gradually increasing training load (t2--early November), after 6 weeks of an intense training cycle (t3--late February) and 48 h after the main competition (t4--early April) and from the controls at three similar time points (t1--early November; t2--late February; t3--early April). In the swimmers, the occurrence of URS clustered around the periods of elevated training load (67 %). No URS were reported at equivalent time points in the non-athletes. Athletes showed a decrease in the percentage (t2 = 21 %; t3 = 27 %; t4 = 17 %) and absolute counts of circulating NK cells (t2 = 35 %; t3 = 22 %; t4 = 22 %), coinciding with the periods of increased training load, never recovering to the initial values observed at the start of the season. The reduction in the CD56(dim) and an increase in the CD56(bright) NK cell subpopulations were significant at t2 and t3 (p < 0.05). Concomitant with the fall in values of NK cells, in athletes that shown more than three URS episodes, a moderate correlation (r = 0.493; p = 0.036) was found between CD56(bright)/CD56(dim) ratio and the number of URS episodes after the more demanding training phase (t3). At t3, a lower value of CD56 cell counts was found in the group who reported three or more URS episodes (t = 2.239; p = 0.032). A progressive significant decrease in the expression of CD119, the receptor for IFN-γ, on the CD56(dim) cells was found over the season and an elevation in Granzyme B expression was coincident with the more demanding training phases. Periods of highly demanding training seem to have a negative impact on innate immunity mediated by NK cell subsets, which could partially explain the higher frequency of URS observed during these training phases.