In vivo studies of T-cell homeostasis in humans have, until now, been hindered by the lack of available tools that are safe for human use. But now, reporting in The Journal of Clinical Investigation, Hellerstein and colleagues use a highly innovative technique to define the dynamics of T-cell proliferation in vivo during HIV-1 infection.

The T-cell pool can be divided into populations that are functionally and kinetically distinct: memory cells have a long life span and high proliferative capacity; by contrast, effector cells typically die quickly by activation-induced cell death. The size of the T-cell pool is mainly regulated and maintained by proliferation of long-lived progenitor cells. In patients infected with HIV, a shortened average life span of T cells has been previously described. However, it remains controversial whether this results from direct cell killing by HIV or indirect effects of chronic activation. Without the benefit of cell-type specific markers that distinguish long- and short-lived cells, Hellerstein et al. developed an alternative approach to characterize the variation in life span in the memory/effector T-cell pool. By administration of deuterium — a safe, stable, non-radioactive isotope of hydrogen — in the form of deuterated water and deuterated glucose, which is incorporated into the DNA of dividing cells, the authors were able to measure T-cell kinetics in vivo over long periods of time. Long-term incorporation of deuterated water into DNA indicated that effector/memory T-cell subpopulations, but not naive T cells, had biphasic kinetics, consistent with the presence of T-cell subpopulations in humans with different life spans. This technique also enabled the detection in normal individuals of long-lived quiescent T cells that did not divide over the 9 week deuterium administration, probably representing the reservoir of progenitor cells.

They then went on to show that individuals with advanced HIV-1 infection had higher proportions of T cells that were short lived, in both the CD4+ and CD8+ memory/effector T-cell subpopulations, compared with healthy controls. These results were confirmed by analysis of die-away kinetics after short-term labelling with deuterated glucose, and together with long-term kinetic analysis, pool sizes and turnover rates of kinetically distinct subpopulations of T cells were calculated. In advanced HIV-1 infection, total pool sizes of short-lived cells were only moderately affected, whereas the size of the long-lived pool was reduced by a factor of more than four. This indicates that patients with HIV-1 have a reduced ability to generate long-lived progenitor cells for maintaining the T-cell pool. Effective treatment with antiretroviral therapy was able to restore the capacity to generate these progenitor cells, by reducing the levels of T-cell proliferation and T-cell death in infected individuals.

These data provide evidence against a model in which the characteristic depletion of CD4+ T cells that occurs in HIV-1 infection results from direct HIV-mediated cell killing, as long-lived progenitors of both CD4+ and CD8+ T-cell subpopulations were reduced in patients infected with HIV-1. Moreover, they favour a model in which increased levels of proliferation of effector/memory T cells in HIV infection reflect chronic activation.

Understanding how infection with HIV-1 affects T-cell homeostasis has implications for vaccine design as well as potential new therapeutic strategies.