Abstract
T cells that undergo homeostatic proliferation under lymphopenic conditions are characterized by greater effector and anti-tumor function than naïve T cells. But the ability of these T cells in causing graft-versus-host disease (GVHD) is not known. We tested the hypothesis that T cells that have undergone homeostatic proliferation would cause more severe acute GVHD than naive T cells by utilizing a well-characterized experimental model of GVHD, B6 (H2b) à BALB/c (H2d). We first injected 2 x 106 CD45.1+ spleen T cells from B6Ly5.2 (H2b) animals along with 5 x 106 CD45.1− T cell depleted bone marrow (TCD BM) from B6Ly5.1 (H2b) animals into lethally irradiated (11Gy) B6Ly5.1 recipients [B6Ly5.2→B6Ly5.1] to induce lymphopenia driven homeostatic proliferation of CD45.1+ T cells for two weeks. Irradiated (8Gy) BALB/c animals were then transplanted with TCD BM from naïve B6 Ly5.2 animals along with 1 x 106 splenic T cells from either naive B6 Ly5.2 or from the [B6Ly5.2→B6Ly5.1] animals. Contrary to our hypothesis, BALB/c animals that received homeostatically proliferated, [B6Ly5.2→B6Ly5.1], donor T cells showed significantly greater survival (70% vs. 10%, P<0.005) and less clinical GVHD (week 3 GVHD severity score, 2.8 vs. 5.5, P<0.01) than those that received B6 Ly5.2 naïve T cells. Consistent with improved survival, allogeneic recipients of homeostatically proliferated T cells showed significantly lower levels of serum TNF-α and LPS (P<0.05) compared to the recipients of naïve T cells. Homeostatically proliferated donors T cells that were generated under unirradiated lymphopenic conditions [B6→B6.SCID] also significantly improved GVHD survival, thus ruling any confounding effects of the inflammatory milieu at the time of their generation on the function of these T cells. To explore the mechanisms of the reduction in GVHD we performed phenotypic analyses of these cells and found that T cells that have undergone homeostasis driven proliferation demonstrated greater expression of ‘memory’ markers such as CD44+ and CD122+. We then sorted the homeostatically proliferated [B6Ly5.2→B6Ly5.1] donor T cells for CD44 low CD3+ and CD44 hi CD3+ and found that 100% of the allogeneic BALB/c animals that received CD44 low CD3+ [B6Ly5.2→B6Ly5.1] T cells died by day 32 with signs of GVHD while all of the syngeneic B6 animals were alive ruling out the non-specific toxicity of these T cells. By contrast, only 30% of the allogeneic animals that received homeostatically proliferated T cells enriched for the CD44 hi CD3+ ‘memory’ fraction died by the end of the observation period (P<0.01). We found no difference in the expansion of CD4+CD25+ regulatory T cells after homeostatic proliferation of donor T cells. Furthermore depletion of CD25+ cells from the homeostatically expanded donor T cells did not enhance the mortality of the allogeneic recipients. Together these results demonstrate: 1) homeostatic proliferation of donor T cells reduces their ability to induce GVHD (2) the reduced ability is due to an increase in the expansion of CD44 hi ‘memory’ type T cells (3) and CD4+CD25+ regulatory T cells are not critical for the reduction in GVHD. Because homeostatic proliferation of T cells has been reported to increase their anti-tumor capability, but reduces GVHD induction, our data thus provide for a new perspective on the requirement of homeostatic driven proliferation in GVHD and GVL responses.
Control of homeostatic proliferation by regulatory T cells