Fig 2: Recombinant low affinity anti-insulin IgM neutralizes insulin in vivo. (A) Schematic representation of the recombinant in-house purified anti-insulin IGHV highlighting the two mutations in the CDR2 which were reverted to the germline version of the IGHV3-74*01 allele. Orange: α-insulin IgMhigh (WT-IGHV); blue: α-insulin IgMlow (gl-IGHV). (B) Coomassie-blue stained SDS-PAGE showing purified α-insulin IgMhigh and α-insulin IgMlow under reducing conditions (with β-mercaptoethanol). The image is representative of three independent experiments. (C) Insulin-binding affinity of α-insulin IgMhigh and α-insulin IgMlow measured by bio-layer interferometry. KD (dissociation constant) was calculated by the software. The experiment shown is representative of 3 independent experiments. (D) Blood glucose concentrations of WT mice intravenously (i.v.) injected with 100 µg α-insulin IgG (black bar, n = 4), α-insulin IgMlow (blue bar, n = 4), α-insulin IgMhigh (orange bar, n = 4) or α-insulin IgG + α-insulin IgMhigh (n = 4, green bar) measured at indicated time points. Mean ± SD, statistical significance was calculated using two-way ANOVA with Sidak’s multiple comparisons test. *p < 0,05; **p < 0,01.

Fig 3: Recombinant low-affinity RF is polyreactive and binds DNA. (A) Schematic representation of the immunoglobulin heavy and light variable genes (IGHV and IGLV, respectively) of the recombinant purified low-affinity RF as compared to the closest germline respective alleles. Mutations are highlighted in purple. IGHM, immunoglobulin heavy constant mu; IGVK, immunoglobulin variable kappa. (B) Coomassie blue-stained SDS-PAGE showing recombinant monoclonal (in-house purified) low affinity RF (RFlow), commercial RF from Rheumatoid Arthritis patients (RFhigh) and monoclonal control IgM (mIgM) under reducing conditions (with β-mercaptoethanol). The image is representative of three independent experiments. (C) IgG-binding affinity of RFlow (purple line), RFhigh (green line) and monoclonal IgM (blue line) measured by bio-layer interferometry. KD (dissociation constant) was calculated by the software. The experiment shown is representative of 3 independent experiments. (D) Anti-IgG IgM concentrations detected in purified RFlow (purple bar, n = 3), RFhigh (green bar, n = 3) and mIgM control (blue bar, n=3) measured by ELISA (coating: human IgG). Mean ± SD, statistical significance was calculated using ordinary one-way ANOVA with Tukey’s multiple comparisons test. **p < 0,01. (E) Anti-dsDNA-IgM concentrations of RFlow (purple bar, n = 3), RFhigh (green bar, n = 3) and IgM control (blue bar, n = 3) measured by ELISA (coating: calf-thymus dsDNA). Mean ± SD. Results are representative of three independent measurements. (F) HEp-2 slides showing anti-nuclear structure-reactive IgM (ANA). Purple square: RFlow; green: RFhigh; blue: mIgM control, respectively. Scale bar 65 µm. Green fluorescence indicates IgM binding to HEp-2 cells. Images are representative of three independent experiments. (G) Schematic summary of the characteristics of RFhigh and RFlow.

Fig 4: High affinity RF enhances the effect of autoreactive IgG. (A) Blood glucose concentrations of WT mice intravenously injected with 100 µg α-insulin IgG alone (black bar, n = 4) or in combination with 20 µg RF concentrate from rheumatoid arthritis patients (RFhigh,green bar, n = 4) or monoclonal IgM control (mIgM, blue bar, n = 4) measured at indicated time points. Mean ± SD, statistical significance was calculated using two-way ANOVA with Tukey’s multiple comparisons test. **p < 0,01. (B) Scheme depicting the procedure for isolation of total IgM from healthy donors (HD) sera. (C) Coomassie blue-stained SDS-PAGE showing total IgM isolation from n=2 healthy donors (IgMHD) under reducing conditions (with β-mercaptoethanol). The image is representative of three independent experiments. (D) IgG-binding affinity of IgM isolated from healthy donors (IgMHD, dark red line), RFhigh (green line) and monoclonal IgM control (mIgM, blue line) measured by bio-layer interferometry. KD (dissociation constant) was calculated by the software. The experiment shown is representative of 3 independent experiments. (E) HEp-2 slides showing anti-nuclear structure-reactive IgM (ANA) for total IgM isolation (IgMHD,dark red square), RFhigh (green square) and monoclonal IgM control (mIgM, blue square). Scale bar 65 µm. Green fluorescence indicates IgM binding to HEp-2 cells. Images are representative of three independent experiments. (F) Blood glucose concentrations of WT mice intravenously (i.v.) injected with 100 µg α-insulin IgG combination with 20 µg total IgM purified from healthy donors (dark red line, n = 4) or with monoclonal IgM control (blue line, n = 4) measured at indicated time points. Mean ± SD, statistical significance was calculated using two-way ANOVA with Sidak’s multiple comparisons test. *p < 0,01.

Fig 5: SCFAs reduce class-switched and hypermutated T-dependent and T-independent antibody responses.a–d C57BL/6 mice fed fiber diet on SCFAs water (SCFAs) or plain water (Nil) starting at the age of 5 weeks, and i.p. injected with NP16-CGG (in Alum) or NP-LPS (in PBS) at the age of 8 weeks. The mice were boost-injected 3 weeks later with NP16-CGG or NP-LPS, respectively, and killed a week after the boost injection. a, b Titers of total and NP3-binding serum IgM, IgG1, and IgA in NP16-CGG injected C57BL/6 mice (a), as well as total and NP3-binding serum IgM, IgG3, and IgG2b titers in NP-LPS injected C57BL/6 mice (b), as measured by ELISA (n = 4 per group, pooled from two experiments). Bars represent the mean ± SD. c, d Somatic point mutations in the V186.2 regions from V186.2DJH-C?1 transcripts of NP16-CGG injected C57BL/6 mice (c), or from V186.2DJH-C?3 transcripts of the NP-LPS injected C57BL/6 mice (d). Sequence data were pooled from three mice in each group. Pie charts depict the proportions of sequences that carry different numbers of point mutations; listed below the pie charts is the overall mutation frequency (change/base). Donut charts depict the spectrum of point mutations. e–h Tcrß–/–Tcrd–/– mice on SCFAs water or plain water were injected with NP-LPS. Total and NP3-specific serum IgM, IgG3, and IgG2b titers were measured by ELISA (n = 5 per group) (e). Somatic point mutations in the V186.2 region of V186.2DJH-C?3 and V186.2DJH-C?2b transcripts. Sequence data were pooled from three mice in each group (f). AID and Blimp1 expression in spleen B cells of these mice, AID expression in spleen B cells of Aicda-/- mice, and Blimp1 expression in naive B cells of C57BL/6 mice, as analyzed by intracellular staining followed by fluorescence microscopy (g), and flow cytometry (h). Data in g and h are representative of three independent experiments. i Quantification of AID+ and Blimp1+ cells in h and g. n = 3 per group. *p < 0.05, **p < 0.01, ***p < 0.001 (unpaired t test). Scale bar = 10 µm. The source data are provided in Source Data file.