Graft-Versus-Host Disease Impairs Early B Lymphopoiesis in the Bone Marrow

Abstract 2976

Graft-versus-host disease (GVHD) and infectious complications are main causes of non-relapse mortality after allogeneic stem cell transplantation (SCT). Impaired immune function after SCT is usually attributed to the immunosuppressive medication applied for GVHD prophylaxis or therapy. Using a major histocompatibility complex (MHC)–mismatched murine model of GVHD (C57BL/6→BALB/c), we now examined the influence of GVHD on B cell immunity after SCT in the absence of pharmacologic immunosuppression. Lethally irradiated BALB/c (H-2^d) recipients were transplanted with T cell-depleted bone marrow (TCD BM; 2.5×10⁶) from C57BL/6 (H-2^b) donors and parallel groups received CD4⁺CD25⁻ conventional donor T cells (Tconv; 0.25 × 10⁶) 2d later. Mice that received TCD BM alone (n =10) did not develop GVHD and showed a rapid and complete reconstitution of B cells in peripheral blood (PB) (25 ± 7% CD19⁺ B cells at d21; 55 ± 5% at d100). Mice that received additional donor Tconv cells (n =12) developed severe GVHD and completely lacked donor and host B cells in PB until their early death or throughout the observation period of 100d (p<0.001). Animals that were protected from severe GVHD by the co-infusion of donor CD4⁺CD25⁺Foxp3⁺ regulatory T cells (Treg; 0.25×10⁶ Treg transplanted together with TCD BM; 2d later transfusion of 0.25 × 10⁶ Tconv) showed a delayed, but finally full reconstitution of their B cell compartment in PB (9 ± 12% CD19⁺ B cells at d21; 42 ± 17% at d100). Similarly, animals without GVHD after TCD BMT and animals protected from GVHD by co-transplanted Treg cells showed a complete reconstitution of their B cell compartment in spleen and BM at d100 (spleen: 26±4, 7×10⁶ and 31 ± 9.3×10⁶ CD19⁺ B cells, respectively; BM: 2, 2 ± 0, 3×10⁶ and 2.9 ± 0.9×10⁶ B cells, respectively). In contrast, B cells were not only undetectable in peripheral lymphoid organs in animals with severe GVHD but also in the BM, suggesting that B cell precursors were affected. To examine whether GVHD solely impedes B cell regeneration or actively contributes to B cell eradication, GVHD was induced after B cell reconstitution at d21 after TCD BMT by donor lymphocyte infusions (DLI). Within 1wk after the transfer of 8×10⁶ or 12×10⁶ donor CD4⁺ lymphocytes, a significant reduction of B cells in PB was detected (from 30.3 ± 5.2% to 10 ± 6.9% and 36.3 ± 9.2% to 5.9 ± 1.3%, respectively; n =4). Thus, GVHD not only affected B cell reconstitution, but even eradicated stem cell-derived B cells that were syngeneic to the GVHD-inducing T cells, suggesting that GVHD-induced inflammation contributed to B cell depletion. To examine the influence of GVHD on precursor cells, serial transplants were performed. Yet, TCD BM from both, animals with and without GVHD, reconstituted their B cell compartment upon secondary transplantation (n =18; 33.1 ± 14.8% vs. 32.4 ± 17% at d100), thereby proving that the stem cell compartment was not affected. Next, we examined the effect of GVHD on precursor cells. Multipotent BM precursors (lin⁻, Sca-1⁺, c-kit⁺ [LSK]) were not significantly different in GVHD animals (TCD BM plus Tconv; n =12) as compared to controls (TCD BM only; n =10; 3.5×10³ ± 2.8×10³ vs. 5.8×10³ ± 2.5×10³, respectively). However, common lymphoid precursors (CLP; Lin⁻, FLt3⁺, CD127⁺) in the BM were significantly reduced in animals with GVHD (0.3×10³ ± 0.17×10³) as compared to transplant recipients without GVHD (4.4×10³ ± 2.2×10³, p<0.001). These results suggest that the dysregulated production of pro-inflammatory cytokines during GVHD is toxic for early B cell precursors and/or that the alloresponse destroys the BM niche for developing B cells. As IFN- γ and TNF are known to be elevated in GVHD and to impair B lymphopoiesis even in a non-transplant setting, we generated mixed chimeras using BM from wt and cytokine receptor deficient animals. Yet, a selective B cell reconstitution from receptor deficient BM was not observed in GVHD, suggesting that neither of these cytokines is exclusively responsible for its toxic effects on B cell precursors. Taken together, our results show that GVHD not solely affects immune reconstitution by the well known destruction of secondary lymphoid organs, but it disturbs early lymphoid progenitors in the BM through inflammatory, but not necessarily allo-specific immune responses.