Fig. 1: Metabolic remodelling of immunized LNs and B cell-dependent GABA production.
From: B cell-derived GABA elicits IL-10+ macrophages to limit anti-tumour immunity
a–e, Mice were injected in the foot pad with OVA + CFA, and iLNs and cLNs were collected for metabolite MS and histology at day 7: WT (n = 11), Cd3–/– (n = 5), muMt–/– (n = 4) and Rag1–/– (n = 3). a, Immunohistochemistry of B cells (B220), T cells (CD3) and myeloid cells (CD11c). Scale bars, 200 µm. b, Principal-component analysis of metabolites in iLNs and cLNs. PC, principal component. c, Pathway analysis of metabolites with significantly different abundance between iLNs and cLNs in WT mice (two-tailed unpaired t-test, P < 0.05). d, Relative Cd3–/– (B cell signature) and muMt–/– (T cell signature) iLN metabolites, showing only metabolites that also differed between iLNs and cLNs in WT mice. e, GABA levels in the indicated cLNs or iLNs, relative to WT cLNs (data are shown as mean ± s.e.m.; two-tailed unpaired t-test: ***P < 0.001, ****P < 0.0001; NS, not significant). f, Correlation of plasma GABA levels with disease activity scores (Simplified Disease Activity Index (SDAI) or Activity Score 28 using C-reactive protein (DAS28-CRP)) and plasma anti-cyclic citrullinated peptide (CCP) antibody levels in patients with rheumatoid arthritis (n = 138). Pearson’s and Spearman’s r and P values (two tailed) are shown. n indicates the number of biological replicates. Data are representative of two experiments (a–e). Exact P values are provided in the Source Data.
