Autoreactivity in human IgG+ memory B cells

Abstract

More than half of the nascent B cells in humans initially express autoreactive antibodies. However, most of these autoantibodies are removed from the repertoire at two checkpoints before maturation into naive B cells. A third checkpoint excludes remaining autoantibodies from the antigen-experienced IgM(+) memory B cell pool. Nevertheless, low-affinity self-reactive antibodies are frequently found in the serum of normal humans. To determine the source of these antibodies, we cloned and expressed antibodies from circulating human IgG(+) memory B cells. Surprisingly, we found that self-reactive antibodies including anti-nuclear antibodies were frequently expressed by IgG(+) memory B cells in healthy donors. Most of these antibodies were created de novo by somatic hypermutation during the transition between mature naive and IgG(+) memory B cells. We conclude that deregulation of self-reactive IgG(+) memory B cells may be associated with autoimmunity.

Figure 1. IgH and IgL chain gene features from IgG memory B cell antibodies

(A) IgG subclass distribution in single IgG+CD27+ memory B cells from three healthy donors. (B) The number of mutations in VH, Vκ, and Vλ genes in antibodies from IgG+ memory B cells. (C) The frequency of mutations in VH, Vκ, and Vλ genes in antibodies from IgG+ memory B cells calculated from the number of replacement (R; black bar) and silent (S; white bar) nucleotide exchanges per base pair in FWRs and CDRs. The R/S ratio for each region is indicated. (D) IgH V and J gene repertoire and IgH CDR3 positive charges and length in amino acids (aa) of antibodies from IgG+ memory B cells compared to mature naïve B cells. Pie charts depict VH and JH family usage and the proportion of IgH CDR3s with 0, 1, 2, or ≥3 positive charges. Bar graphs show frequencies of IgH CDR3s with 9 aa (white bars), 10–14 aa (light gray bars), 15–19 aa (dark gray bars), and ≥20 aa (black bars). The absolute number of sequences analyzed is indicated in the center of each pie chart. Values for mature naïve B cells in this and other figures were published previously and are shown here for comparison (Tsuiji et al., 2006; Wardemann et al., 2003).

Figure 2. Self-reactive antibodies are enriched in the IgG+ memory B cell pool

(A) IgG+ memory B cell antibodies from healthy donors were tested for self-reactivity by HEp-2 cell lysate ELISA and IFA. Horizontal line shows cut-off OD₄₀₅ for positive reactivity determined by comparison to low positive control serum (red line). Typical HEp-2 cell IFA staining patterns of antibodies cloned from IgG+ memory B cells are shown. (B) Pie charts summarizing the frequency of HEp-2 self-reactive IgG+ memory B cell antibodies with nuclear (black), nuclear plus cytoplasmic (dark gray), and cytoplasmic (light gray) IFA staining patterns, and the frequency of non-reactive antibodies (white) in comparison to mature naïve and IgM+ memory B cell clones (Tsuiji et al., 2006; Wardemann et al., 2003). The number of tested antibodies is indicated in each pie chart center. P-values are in comparison to mature naïve B cells and IgM+ memory B cells (Tsuiji et al., 2006; Wardemann et al., 2003). (C) Serum IgG ANA levels of three healthy donors and one patient with Systemic lupus erythematosus (SLE101; Yurasov et al., 2005) were determined by HEp-2-ANA-ELISA. The manufacturer’s instructions were followed to calculate relative units based on internal positive and negative control sera. Horizontal lines show positive, low positive or negative cut-off titres as indicated

Figure 3. Polyreactive antibodies contribute to the IgG+ memory B cell compartment

IgG+ memory B cell antibodies from healthy donors were tested for polyreactivity by ELISA with ds/ssDNA, LPS, and insulin. Dotted lines represent the high positive control antibody ED38 (Meffre et al., 2004). Horizontal lines show cut-off OD₄₀₅ for positive reactivity as determined by comparison to the negative control antibody mGO53 (green line) and low positive control eiJB40 (red line; Wardemann et al., 2003). Pie charts show the frequency of polyreactive clones from IgG+ memory B cells from all three donors compared to mature naïve B cell and IgM+ memory B cell antibodies (Tsuiji et al., 2006; Wardemann et al., 2003). The number of tested antibodies is indicated in the center. P-values are in comparison to mature naïve B cells and IgM+ memory B cells (Tsuiji et al., 2006; Wardemann et al., 2003).

Figure 4. Somatic hypermutation contributes to self-reactivity in IgG memory B cell antibodies

(A) IgH and IgL chains from IgG+ memory B cell antibodies were reverted into their germline counterparts by PCR. Recombinant polyreactive (upper left) and non-polyreactive (lower left) IgG+ memory B cell antibodies and their germline counterparts (right) were tested for polyreactivity by ELISA with ds/ssDNA, insulin and LPS. Representative graphs with dsDNA as antigen are shown. Dotted lines represent the high positive control antibody ED38 (Meffre et al., 2004). Horizontal lines show cut-off OD₄₀₅ for positive reactivity as determined by comparison to the previously published control antibodies mGO53 (negative control: green line; Wardemann et al., 2003) and eiJB40 (low positive control: red line; Wardemann et al., 2003). (B) Typical HEp-2 cell IFA staining patterns of mutated IgG+ memory B cell antibodies (top) and their germline counterparts (bottom).

Figure 5. Monomeric IgM from human serum is less self-reactive than serum IgG

Polymeric IgM from pooled human plasma of healthy donors (open circles) and from serum of two healthy donors (PN, HW) was reduced under mild conditions to its monomeric IgM subunits (filled circles). Serum IgG antibodies from pooled human plasma of healthy donors and from three healthy donors (PN, HW, VB) were purified with Protein-G beads (squares). Equal molar quantities of the antibodies were tested by ELISA for reactivity against dsDNA. Control antibodies were the highly polyreactive ED38 (dotted line; Meffre et al., 2004) and negative mGO53 (triangles; Wardemann et al., 2003).