Recipient FRCs and donor CD4⁺ T cells are potential sources of increased production of soluble BAFF in mice that develop cGVHD. (A) Spleen weights in BM + Sp group mice vs BM only control group mice on day 8 after allo-BMT; n = 10 in each group. (B) Representative flow cytometry of podoplanin/gp38⁺ (PDPN⁺) CD31⁻ fibroblastic reticular cells (FRCs) after pregating on Zombie⁻CD45⁻H-2K^d+ (recipient) cells. FRCs were isolated from spleen and lymph nodes (LNs) at day 8 post-BMT using an established enzyme digestion strategy as previously described.⁴⁶  Numbers shown on the flow cytometric plots are cell percentages within each gate. (C) The proportion (%) of FRCs in BM + Sp group vs BM only group at day 8 post-BMT (n = 5 each). Data shown are representative of 3 repeats. (D) Total numbers of PDPN⁺CD31⁻MadCaM⁻ FRCs acquired from pooled spleen and identical LNs harvested from within each group. BM + Sp cGVHD group was compared with BM only control group at day 8 post-BMT, as assessed by flow cytometry. Pooled data from 2 independent experiments are shown; n = 10 (BM only) and n = 9 (BM + Sp). (E) BAFF transcript levels in FRCs on day 8 after transplantation isolated from BM + Sp mice or BM only controls. Quantitative polymerase chain reaction (qPCR) data were analyzed using a ΔΔCT method and normalized to BM only group data (set to mean of 1). Fold change, 2^−ΔΔCT; n = 6 (BM only) and n = 11 (BM + Sp). (F) BAFF transcript levels as measured by qPCR in CD4⁺ T cells taken from mice without cGVHD (BM only) vs mice with cGVHD (BM + Sp) or syngeneic BMT recipients (Syn). CD4⁺ T cells were sorted by flow cytometry from recipient splenocytes at day 50 post-BMT (>95% purity). BAFF transcript was measured by qPCR, analyzed using the ΔΔCT method, and normalized to BM only group data (set to mean of 1). Fold change, 2^−ΔΔCT; n = 5 (BM only), n = 4 (BM + Sp), and n = 5 (Syn). (G) BAFF transcripts measured in flow cytometric purified (>98%) CD4⁺CXCR5⁺ double-positive T cells, classical T_FH cells, taken from peripheral blood mononuclear cells of patients without vs with clinically active cGVHD. BAFF transcripts were assessed by qPCR and analyzed by ΔΔCT method. Fold change, 2^−ΔΔCT. Results were normalized to no-cGVHD patient data; n = 3 (no-cGVHD) and n = 6 (active-cGVHD). Statistical analysis was performed by unpaired Student t test with Welch’s correction (A-E,G) or ordinary 1-way analysis of variance with Tukey’s multiple comparisons test (F) using GraphPad Prism 8 software. *P < .05, **P < .01, ***P < .001, ****P < .0001. NS, not significant (P > .05).