Targeting Syk-activated B cells in murine and human chronic graft-versus-host disease

Abstract

Novel therapies for chronic graft-versus-host disease (cGVHD) are needed. Aberrant B-cell activation has been demonstrated in mice and humans with cGVHD. Having previously found that human cGVHD B cells are activated and primed for survival, we sought to further evaluate the role of the spleen tyrosine kinase (Syk) in cGVHD in multiple murine models and human peripheral blood cells. In a murine model of multiorgan system, nonsclerodermatous disease with bronchiolitis obliterans where cGVHD is dependent on antibody and germinal center (GC) B cells, we found that activation of Syk was necessary in donor B cells, but not T cells, for disease progression. Bone marrow-specific Syk deletion in vivo was effective in treating established cGVHD, as was a small-molecule inhibitor of Syk, fostamatinib, which normalized GC formation and decreased activated CD80/86(+) dendritic cells. In multiple distinct models of sclerodermatous cGVHD, clinical and pathological disease manifestations were not eliminated when mice were therapeutically treated with fostamatinib, though both clinical and immunologic effects could be observed in one of these scleroderma models. We further demonstrated that Syk inhibition was effective at inducing apoptosis of human cGVHD B cells. Together, these data demonstrate a therapeutic potential of targeting B-cell Syk signaling in cGVHD.

Figure 1

Syk activation during cGVHD. (A) B cells were purified from the spleens of healthy control (BM only) or cGVHD (BM + T) mice and stimulated with anti-IgM at 5 μg/mL. Splenocytes from B10.BR mice transplanted with B6 BM and low numbers of T cells were analyzed for pSyk at Y348 after ex vivo stimulation by 5 μg/mL of anti-IgM. (B) Cumulative expression of phosphorylated Syk. *P < .5. Error bars represent standard error of the mean (SEM); n = 8 representative data from 2 experiments.

Figure 2

Presence of Syk in BM-derived B cells, but not donor T cells, is necessary for development of cGVHD pathology in a BO model. (A) Day-60 pulmonary function tests of mice transplanted with tamoxifen-induced Syk KO BM and WT T cells. (B) Frequency of B cells in transplanted mice on day 60 after transplant. (C) Pulmonary function tests from mice transplanted with WT BM and either WT T cells or Syk KO T cells. (D) Pathology scores from lung and liver of mice transplanted with Syk KO T cells. *P < .5; **P < .01. Error bars represent SEM; representative data from 3 independent experiments with n = 8 per group.

Figure 3

Genetic deletion of Syk during active disease prevents pathogenic pulmonary function. Mice were transplanted with either WT or Syk fl/fl × ERT2-Cre BM. On day 8, mice were treated with tamoxifen for 5 days to delete Syk in donor-BM–derived cells. Pulmonary function tests of mice on day 60. *P < .5; ** P < .01. Error bars represent SEM; n = 10 per group.

Figure 4

Inhibition of Syk by R788 decreases cGVHD in a murine BO model. (A) Day-56 pulmonary function tests on mice treated with 30 mg/kg R788 twice daily from days 28 to 56. (B) GVHD pathology scores from the lungs of mice on day 60. (C) Number of GCs present in situ in spleens of mice. (D) Frequency of GC B cells (gated on CD19⁺GL7⁺CD95^hi) present in the spleens of transplanted mice on day 60. (E-G) Frequency of CD11c cells expressing CD80, CD86, and MHC class II in the spleens of transplanted mice on day 60. *P < .5; **P < .01; ***P < .001. Error bars represent SEM; representative data from 3 independent experiments with n = 8 per group.

Figure 5

Fostamatinib does not alter skin inflammation in a B6 into B6D2F1 model of sclerodermatous GVHD but decreases peripheral blood Ly6C^hi monocytes. (A) Frequency of CD19⁺ B cells in spleens of mice on day 21 in vehicle-treated (gray bars), R788-treated (black bars), and non-GVHD controls (white bars). (B) Total inflammation score. (C) Absolute number of CD11b⁺ F4/80⁺ peripheral blood monocytes. Peripheral blood monocytes were examines for (D) Ly6C high or (E) Ly6C low expression. *P < .5; **P < .01. Error bars represent SEM; n = 6 per group. PB mono, peripheral blood monocyte.

Figure 6

Fostamatinib has little clinical therapeutic benefit in a B10.D2 into BALB/c sclerodermatous cGVHD model but decreases expression of CXCR4 on CD11b⁺ and CD4⁺ cells. (A) Clinical GVHD scores or (B) skin GVHD scores in vehicle-treated (gray squares) or R788-treated (black circles) mice in the B10.D2 into BALB/c model. (C) Frequency of viable CD11b⁺ cells. (D) Expression of CXCR4 on CD11b⁺ cells. (E) Absolute number CD19⁺ B cells. (F) Expression of CXCR4 on B cells. (G) Expression of CXCR4 on CD4⁺ cells. *P < .5. Error bars represent SEM; n = 5 per group. CLN, cervical lymph node; MLN, mesenteric lymph node.

Figure 7

Increased apoptosis in human B cells when treated with R788. (A) Representative flow plots of human PBMCs with or without cGVHD. Consistent with previous work, unmanipulated, untreated cGVHD B cells had superior survival compared with B cells from patients without disease. PBMCs from patients without cGVHD (n = 6, open circles) and with cGVHD (n = 6, filled squares) treated with R406 (0, 0.01, and 0.1 μM) as indicated for 48 hours. Apoptotic B cells were defined as CD19⁺ Annexin V⁺ 7AAD⁻ cells (B) or as CD19⁻ non–B cells (C). Fold increase in apoptosis by R406 divided by phosphate-buffered saline is depicted. Data are median ± range pooled from 2 independent experiments. *P < .5; **P < .01; ***P < .001. n.s., not significant.