Natural CD4 CD25 FOXP3 regulatory T cells in graftversus-host disease - Immune Biology of Allogeneic Hematopoietic Stem Cell Transplantation - page 245

the maturing T cells in the thymus. The fact that only very few Foxp3 + T cells
in the thymus belong to the immature double-positive thymocyte subset,
however, suggests that cTEC play only a minor role in Treg cell differentia-
tion [23] . In contrast, there is ample evidence that both mTEC and DC are
equally potent in driving thymic Treg cell development [10] . Yet, it seems
plausible that antigens presented by DC of intrathymic origin, which repre-
sent approximately 2/3 of thymic DC, are different from those sampled by
migratory DC that enter the thymus on their way through the body, and that
both of these differ yet again from those presented by mTEC [24] . mTEC
(and cTEC) are unique among nonhematopoietic cells in that they show
a constitutive high expression of MHC class II molecules. This, together
with their specific capacity to ectopically express a whole array of tissue-
restricted antigens (partially controlled by the “autoimmune regulator”
( AIRE ) gene), makes them the central players for antigen presentation and
tolerance induction in the thymus [25] .

In summary, a cascade of interdependent steps is necessary to ensure Treg
differentiation and the development of a functional Treg compartment in
the thymus. It is thus easily conceivable that disruption of this multistep
process, e.g. by destruction of thymic epithelial cells through alloreactive
donor T cells after allogeneic SCT [26,27] , leads to erroneous positive and
negative selection processes and, consequently, to a severe imbalance in
both central and peripheral tolerance mechanisms.

248

One of the predominant characteristics of Treg cells in vitro is their hypo-
proliferative response to standard T cell stimulation, which led to the
assumption that these cells are generally anergic. In vivo, however, Treg cells
proliferate substantially, not only after transfer into lymphopenic hosts but
also in response to their cognate antigen in lymphocyte-replete animals
and to a certain extent even under steady-state conditions [28] . In fact, they
require much lower antigen concentrations for activation than their non-
regulatory counterparts [29] . For their survival (and function), Treg cells also
crucially depend on exogenous IL-2. They themselves are unable to pro-
duce this cytokine because Foxp3 directly binds to the IL-2 promoter [30]
thereby inhibiting adequate chromatin remodeling of that region after TCR
stimulation [31] . Consequently, mice lacking either IL-2 or the IL-2 receptor
(IL-2R) have profoundly diminished Treg cell numbers in the periphery and
spontaneously develop autoimmune diseases [32,33] . Thus, CD25 as part of
the high-affinity IL-2R is not only a useful surface marker of Treg cells, but
indispensable for their peripheral maintenance.

Treg cells in the periphery account for approximately 5-10% of the CD4 +
T cell pool. However, there is now increasing evidence that Treg cells not
only develop as a separate lineage in the thymus, but also can differenti-
ate from peripheral naïve CD4 + CD25 − T cells under certain stimulatory
conditions, such as TCR activation in the presence of TGF-β [34] or after
targeted delivery of low-dose antigen to steady-state DC via anti-DEC205
antibodies [35] . To what extent these “induced” or “converted” Treg cells
contribute to the peripheral Treg cell population under physiological and
pathological conditions and to what degree they share functional proper-
ties with their thymus-derived counterparts is currently under intensive
investigation. One significant challenge of such studies is, again, the lack of