Pivotal studies in mice models have pointed out that Treg are indispensable for the maintenance of peripheral immune tolerance. Also in humans a similar role of Tregs is likely, prompting discussions about their clinical applicability. Though comparable in many aspects, several differences between mouse and human Treg phenotype, function and mechanisms of suppression have been identified in the past few years. For instance, the expression of FOXP3 seems to be a more consistent marker for functional Treg in mice, than it is in humans 
. As for mechanisms of suppression, IL-35 production by Treg is important for suppression in mice 
, while IL-35 is not even expressed by human Treg 
. Since Treg are currently tested for therapeutic applications in humans, it is especially important to determine to what extend results obtained in mice can be translated to human Treg.
Recently, Pandiyan et al. exemplified a new mechanism of action of Treg in mice, namely their capacity to induce apoptosis in Teff, based on specific cytokine consumption as Treg can consume IL-2 produced by the Teff. Also, addition of IL-2 to co-cultures of Teff and Treg prevented the apoptosis of Teff. Though they did not directly show that addition of IL-2Rγ-chain binding cytokines, which diminished apoptosis, also prevented suppression in vitro
, in vivo
they did find that induction of Teff apoptosis is indeed important for Treg function. Furthermore, previous reports show that suppression in vitro
by murine Treg is prevented by addition of IL-2Rγ-chain binding cytokines 
. Our current data show some similarities between the mouse and human system, but also reveal an essential difference between mouse and human Tregs; human Treg function is not mediated by apoptosis of Teff. Obviously, human experiments such as these are restricted to in vitro
assays, and only limited numbers of cells are available. However, in vitro
Treg assays, similar to those used for mice, can be performed with human cells as well and compared to data obtained in experimental models.
Similar to mice, we show that naturally occurring human Treg very efficiently suppress both proliferation and cytokine production by effector T cells, which can be reversed by addition of IL-2Rγ-chain binding cytokines. These results are consistent with earlier reports on human and murine Treg which show both inhibition of Teff IL-2 mRNA production, as well as Teff proliferation by Treg, and a decrease of suppression of Teff proliferation by addition of high levels of exogenous IL-2 
. Also, Treg derived from a highly inflammatory environment, synovial fluid from the joints of JIA patients, suppress Teff proliferation and cytokine production. Obviously, mouse splenocytes differ in many aspects from human PBMC 
. Here we show that human Teff seem to be less prone to apoptosis than mouse Teff. When comparing cell death in cultures with only Teff, human Teff show hardly any apoptosis (2%), whereas mouse Teff show a higher level of apoptotic cells (20%) 
. And, importantly, we show that suppression by human Treg does not involve induction of apoptosis in Teff: the absolute numbers of apoptotic cells decrease in the presence of Treg.
IL-2 is an important cytokine for Treg function, both in mice and humans. However, we do not find a decrease of apoptosis in Teff upon addition of IL-2. This may again be due to the low level of apoptosis in Teff in general. However, it could also be explained by the fact that Teff do not necessarily require IL-2 to survive or become activated. This is confirmed by recent data obtained by in vitro
tests on peripheral blood cells from a specific group of IPEX patients. In these patients Teff produce only low levels of IL-2 and, remarkably, the deficit in Treg function can be overcome by addition of IL-2 to cell cultures. Thus, the in vivo
lack of Treg function could be explained by the decreased production of IL-2 by Teff in these IPEX patients 
.Altogether, this suggests that in humans IL-2 is very important for Treg function, but is not required for Teff survival and function, as these Teff, despite low IL-2 production, are still highly activated and causing disease.
We show here, in line with earlier publications, that addition of IL-2 and IL-7 abrogates suppression of both Teff proliferation and cytokine production. This could be due to a higher activation of the Teff, as the Teff cultured alone proliferate more and produce more cytokines in the presence of IL-2 and IL-7, or, in case of IL-2, to abrogation of Treg anergy. In addition, we do not find a decrease of the added IL-2 in these cultures with Treg present. This suggests that IL-2 is not consumed by the Treg, although we can not exclude that the level of exogenous IL-2 is simply too high to detect consumption by Treg.
In conclusion, we here point out an important difference between human and murine Treg function: human Treg do not induce apoptosis in Teff to achieve suppression. With these data we emphasize that experimental data from mouse models should be carefully validated in human cells to identify discrepancies, and to ensure that further therapeutic applications are efficient and safe. This does not mean that Treg are less valuable targets for intervention. It could even be argued that if human Treg, instead of eliminating Teff by inducing apoptosis, render Teff either anergic, or even turn them into suppressor cells themselves 
, may be able to exert a stronger bystander suppression in an ongoing inflammatory response.
Our functional data here support that Treg are excellent clinical targets to counteract autoimmune diseases. For optimal functional outcome in human clinical trials, future work should focus on the ability of Treg to suppress proliferation and cytokine production of Teff, rather than induction of Teff apoptosis.