Autoreactive Anti-DNA Transgenic B Cells in Lupus-Prone New Zealand Black/New Zealand White Mice Show Near Perfect L Chain Allelic Exclusion

Abstract

A nti-DNA autoantibodies (1, 2) are the hallmark of human systemic lupus erythematosus (3) and of mouse models of this prototypic autoimmune disease (4). The extensively studied anti-DNA Abs resemble secondary response Igs in experimental animals mostly comprising somatically mutated IgG with high affinity for DNA (5). In contrast, mostly lowaffinity, germline-encoded IgM "natural" anti-DNA Abs can be found in the sera of healthy individuals (6). This difference has been attributed to mechanisms of self-tolerance, leading to negative selection of autoreactive B and T cells (7). Several recent studies have suggested that intrinsic B cell defects may bear the prime responsibility for the loss of self-tolerance in systemic lupus erythematosus (8). Therefore, the mechanisms of B cell tolerance, particularly those involving clonal deletion (9), clonal anergy (10), and receptor editing (11) have become the focus of intensive investigation. Receptor editing provides the cell with a new receptor through replacement of the autoreactive L chain with a new one by continued rearrangement and is the primary mechanism for removal of a self-reactive BCR in the bone marrow (12).Recent studies have suggested that the process of secondary rearrangement is not necessarily accompanied by inactivation of the autoreactive L chain and may lead to a state of allelic or isotype inclusion (13)(14)(15)(16)(17)(18)(19). This results in the expression of more than one functional BCR on the same B cell and challenges the dogma of "one lymphocyte, one receptor" (20). Estimates of L chain-included B cell frequency in nonautoimmune, wild-type mice, using different methodologies, range from 3 to 10% of all B cells (17, 18). However, in transgenic (Tg) 3 animals expressing an autoreactive BCR, this frequency could reach 40% (15). It has been argued that coexpression of a nonself-reactive or L chain can rescue the B cell from negative selection by diluting out the selfreactive receptor (13)(14)(15)(16)18). Additionally, it was suggested that under certain circumstances, such as in aging mice, these B cells are potentially functional and can differentiate into autoantibodysecreting plasma cells (15).All of the studies so far have been conducted in wild-type or Tg normal mice. However, this newly described mechanism of L chain allelic inclusion could have particular implications for specific autoimmune responses (e.g., anti-DNA) in susceptible animals, such as the lupus-prone NZB/NZW F1 mouse (4). On one hand, these animals could use allelic or isotype inclusion to avoid B cell tolerance and thus significantly increase the frequency and selection of included B cells, especially in the fraction of autoantibody producing cells. Alternatively, an opposite scenario could also be envisaged, where the lack of effective tolerance mechanisms would reduce the need for L chain inclusion and their frequency would therefore be diminished.To test these different possibilities, we used our well-defined, anti-DNA, knock-in mouse lines (21,22) backcrossed to the NZ...

L Chain Allelic Inclusion Does Not Increase Autoreactivity in Lupus-Prone New Zealand Black/New Zealand White Mice

“…This finding is also supported by the low percentage of L chain-included hybridomas among both spontaneous and LPSactivated anti-DNA hybridomas from hCk/mCk NZB/NZW mice (Table II). In anti-DNA transgenic hCk/mCk D42H NZB/ NZW mice, the percentage of anti-DNA hybridomas from total hybridomas was three to four times higher than in wild-type NZB/ NZW mice, but the fraction of L chain-included hybridomas in this population was only 1-2% (Table II), in agreement with our previous study (18).…”

“…Although our previous hybridoma study (18) was limited to anti-DNA hybridomas and to a small number of L chains, it was still puzzling that the level of allelic inclusion was much lower than that observed in this study, using single sorted B cells. We therefore compared the present analysis to hybridomas obtained from activated B cells (i.e., spontaneous hybridomas) and to LPSinduced hybridomas, presumably derived from nonactivated B cells (20,24,25).…”

“…Our results are in agreement with earlier studies (11,13,15) with regard to the close association between L chain allelic inclusion and receptor editing. The earlier discrepancy concerning the fraction of included hybridomas (18) was probably due to the difference between spontaneous and LPS-induced hybridomas. This difference suggests that included B cells are generally not activated and are not recruited to the GC for class switching and somatic mutation, possibly because of their low effective BCR density (9,11).…”

“…Hybridoma production, flow cytometric analysis, reverse transcription, sequence determinations, and DNA-binding tests were performed as described previously (18).…”

“…We have followed this work with lupus-prone NZB/NZW mice by studying anti-DNA-producing hybridomas from our knockin mice, which extensively edit their BCRs (18). We found, surprisingly, that the extent of k/k and k/l allelic inclusion in spontaneous (non-LPS-induced) anti-DNA hybridomas was negligibly small (∼0.5%).…”