The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Watson, L. C.
Right arrow Articles by Feeney, A. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Watson, L. C.
Right arrow Articles by Feeney, A. J.
The Journal of Immunology, 2006, 177: 1120-1128.
Copyright © 2006 by The American Association of Immunologists

Paucity of V-D-D-J Rearrangements and VH Replacement Events in Lupus Prone and Nonautoimmune TdT–/– and TdT+/+ Mice1

Lisa C. Watson, Chantelle S. Moffatt-Blue, R. Zachary McDonald, Elizabeth Kompfner, Djemel Ait-Azzouzene, David Nemazee, Argyrios N. Theofilopoulos, Dwight H. Kono and Ann J. Feeney2

The Scripps Research Institute, Department of Immunology, La Jolla, CA 92037


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
CDR3 regions containing two D segments, or containing the footprints of VH replacement events, have been reported in both mice and humans. However, the 12–23 bp rule for V(D)J recombination predicts that D-D rearrangements, which would occur between 2 recombination signal sequences (RSSs) with 12-bp spacers, should be extremely disfavored, and the cryptic RSS used for VH replacement is very inefficient. We have previously shown that newborn mice, which lack TdT due to the late onset of its expression, do not contain any CDR3 with D-D rearrangements. In the present study, we test our hypothesis that most D-D rearrangements are due to fortuitous matching of the second apparent D segment by TdT-introduced N nucleotides. We analyzed 518 sequences from adult MRL/lpr- and C57BL/6 TdT-deficient B cell precursors and found only two examples of CDR3 with D-D rearrangements and one example of a potential VH replacement event. We examined rearrangements from pre-B cells, marginal zone B cells, and follicular B cells from mice congenic for the Lbw5 (Sle3/5) lupus susceptibility loci and from other strains of mice and found very few examples of CDR3 with D-D rearrangements. We assayed B progenitor cells, and cells enriched for receptor editing, for DNA breaks at the "cryptic heptamer" but such breaks were rare. We conclude that many examples of apparent D-D rearrangements in the mouse are likely due to N additions that fortuitously match short stretches of D genes and that D-D rearrangements and VH replacement are rare occurrences in the mouse.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
In each B cell precursor, one of the several DH genes rearranges to one of the four JH genes, and this DJ rearrangement process occurs on both chromosomes. Subsequently, one of the DJ rearrangements joins to one of the ~100 VH genes to make a complete VDJ H chain rearrangement. VH genes and JH genes are flanked by recombination signal sequence (RSS)3 with 23-bp spacers, and the small D gene segments are flanked on both sides by RSS with 12-bp spacers. Direct VH to JH rearrangement is precluded by the 12–23 bp rule, which states that an RSS with a 12-bp spacer can only efficiently rearrange to an RSS with a 23-bp spacer. Extensive studies with recombination substrates have provided the basis for this rule, showing that two RSSs with 12-bp spacers or two RSSs with 23-bp spacers join with each other at <0.1–2% of the frequency of a 12-bp RSS and a 23-bp RSS (1, 2, 3). Recently, an elegant in vivo experiment with a mouse lacking all D genes clearly demonstrated that direct VH to JH rearrangement, i.e., rearrangement between two RSSs both containing 23-bp spacers, is indeed extremely inefficient (4).

This 12–23 bp rule would also not allow rearrangements of one DH segment to another DH segment because both sides of DH genes are flanked by 12-bp spacers. Nonetheless, analyses of CDR3 regions in both mice and humans have occasionally shown two stretches of nucleotides that show identity to different DH genes, and a few reports have claimed to observe very high frequencies of CDR3 with two DH genes in lupus prone mice (5, 6, 7). Hence, the issue of whether D-D rearrangements actually occur in CDR3 of Ig H chains in vivo, and their frequency, is quite controversial (7, 8, 9, 10, 11, 12, 13, 14). Rearrangements with two apparent D segments in CDR3 have been observed in ~5–15% of human CDR3 peripheral blood sequences and in pre-B cells from human bone marrow (15, 16, 17). More recently, a small population of circulating human B cells coexpressing both VpreB and conventional L chains has been described that has unusually long CDR3s, shows increased usage of the most 3' JH, and possesses a high frequency of D-D joinings in CDR3 (9, 10).

Longer CDR3s, which would be created by potential D-D rearrangements, have been associated with self-reactivity and polyreactivity in mAbs (18, 19, 20, 21). D-D rearrangements have been proposed to be important in generating the arginines that are important in anti-DNA Abs, and several anti DNA hybridomas from the lupus prone MRL/lpr mice do, in fact, have apparent D-D rearrangements (21, 22, 23, 24). Furthermore, there have been reports of unusually high frequency of D-D rearrangements in two different strains of lupus-prone mice (5, 6). We previously reported that we could only identify ~2–4% of rearranged sequences as having potential D-D rearrangements in CDR3 from adult bone marrow pre-B cells in both lupus-prone and nonautoimmune strains (25). We further showed that newborn sequences from these same strains displayed no D-D rearrangements (25). Hence, the probability that the potential second D segment in the CDR3 sequences was fortuitous identity of N region addition with a D gene seemed a reasonable concern, which we address in this current study.

Another type of unconventional V(D)J rearrangement is that of VH replacement. This process involves the recombination of a VH gene upstream of a VDJ rearrangement with a "cryptic heptamer" located near the 3' end of the V gene that is in a VDJ rearrangement, as diagrammed in Fig. 2A (26, 27, 28). This cryptic heptamer is encoded within the conserved Tyr-Cys-Ala (TAC TGT GCN) that is present in almost all VH genes close to the end of framework 3. Thus, the VH replacement event replaces almost all of the original VH gene with the new VH gene. The original CDR3 is maintained, and additional nucleotides can be contributed to the 5' side of the CDR3 by the 3' end of the new invading VH gene. Like D-D rearrangements, VH replacement does not follow the canonical rules for V(D)J rearrangement. The cryptic heptamer that would be used in VH replacement has no nonamer upstream of it. In fact, we previously made recombination substrates with this region and demonstrated that it was indeed very ineffectual as an RSS (29).


Figure 2
View larger version (32K):
[in this window]
[in a new window]
  		FIGURE 2. Analysis of VH replacement. A, Diagram of VH replacement. B, Location of the primers and the linker for the LM-PCR assay for DSBs at the cryptic RSS. C, This table shows the number of individual PCR done on each indicated cell type. Tg+ (transgene) refers to the {kappa}-macroself Ag transgenic bone marrow. Tg– are littermate controls. The final two columns are the immature B cells from the receptor editing cultures. Multiple independent cell preparations were used for each cell type. The majority of PCRs was performed with 30 cycles each in the primary and secondary PCR to maximize the detection of breaks at the cryptic RSS, and the number of PCRs with products that were approximately the anticipated size (~275 bp), and the number of those that were cloned, are shown. The last line shows the number of sequences showing breaks at the FR3 cryptic RSS. D, The four most common locations of DSBs are shown. The location of the break ligated to the linker is indicated with an arrow. Cryptic heptamer sequences are underlined, as is the sequence in the location of the putative nonamer (i.e., 12-bp 5' of the heptamer), although none of the examples shown except the last one bears any resemblance to a consensus nonamer.

 
The issue of whether VH replacement is a mechanism that contributes much to repertoire formation is unclear, and based on the lack of a good RSS, one might assume that it would be rare under normal circumstances. Since the initial observations of VH replacement in cell lines, only a small number of examples of VH replacement in vivo have been documented (30, 31), with the notable exception of site-directed knock-in mice, in which a rearranged VDJ is inserted by gene-targeting into the JH locus (32, 33, 34, 35, 36, 37, 38, 39, 40). This type of evidence for VH replacement must be evaluated with the caveat that the knocked-in VDJ is in an unusual location, with the entire DH as well as VH loci upstream of the rearranged VDJ. Nonetheless, several BCR knock-in models support the hypothesis that there are levels of VH replacement events sufficient to find the V-V-D-JH rearrangements in the periphery, which is usually where these events are assayed in the form of hybridomas (32). However, peripheral selection and preferential expansion, particularly in BCR transgenic tolerance models where editing of the original receptor is required to escape self-reactivity, preclude frequency estimates of the initial VH replacement events, which is our aim here. Without having a situation such as a knock-in VDJ, it is very difficult to demonstrate that a VDJ sequence is the result of a VH replacement event because the remnants of the original VDJ (only the CDR3 and JH) will look like a primary VDJ, and the only "footprint" will be a few extra nucleotides at the VH-VDJ junction. These are extremely difficult to distinguish from N nucleotides that may fortuitously match the 3' end of a VH gene.

Thus, the issue of whether VH replacement actually happens in vivo to any physiologically relevant extent has remained enigmatic. A recent study, however, suggested that VH replacement is a frequent event in human immature B cells (41). Using ligation-mediated PCRs (LM-PCRs), bands were observed that were consistent with RAG-mediated double strand breaks (DSBs) at the cryptic RSS in immature B cells from bone marrow but not from pro-B cells or pre-B cells. This suggests that the VH replacement breaks were occurring predominantly as a result of receptor editing in human bone marrow but not as a rescue mechanism for pro-B cells with two nonproductive rearrangements. Analysis of the CDR3 sequences of many rearranged human H chain genes has led the authors to estimate that VH replacement is a common occurrence in the human B cell repertoire (41). Finding that VH replacement is common at the receptor editing stage is somewhat surprising because receptor editing at the H chain locus is much more difficult to perform than at the L chain loci. First, the L chain loci are well organized for continuing rearrangement, whereas the H chain locus is not (42). Second, the VH region of the H chain locus is likely to be less accessible to the recombinase during the pre-B/immature B cell stage than it was during pro-B cell stage. It has been shown that the histones associated with VH genes are less acetylated in pre-B cells than in pro-B cells, correlating with the lowered accessibility after H chain rearrangement is completed (43). Hence, a priori, one would predict that VH replacement events would be rare at all times due to the poor quality of the cryptic RSS and even less likely after the pro-B cell stage when the chromatin of the VH region is less accessible, unless receptor editing signals result in increased accessibility. Nonetheless, it appears that VH replacement is a frequent outcome of receptor editing in human bone marrow immature B cells (41). Whether the same VH replacement events occur at such high frequencies in the mouse as a result of receptor editing has not been investigated.

In this study, we analyzed the frequency of creation of CDR3 with either D-D rearrangements or VH replacement events in the mouse. We used TdT-deficient mice on both lupus-prone and nonautoimmune backgrounds so that we could analyze CDR3 sequences without the ambiguity of N nucleotides. In these mice, all nucleotides in the CDR3 must be of germline origin. Even if a VH replacement event only retained a few base pairs from the original VH gene, we would be able to unambiguously identify them as such in the absence of TdT, whereas distinguishing a 2- to 4-bp footprint from N nucleotides is impossible. Likewise, any CDR3 sequences containing two D segments would be obvious. Our analysis of the CDR3 sequences of 518 cloned H chain rearrangements from bone marrow cells pre-B cells and immature B cells of MRL/lpr TdT–/– and C57BL/6 (B6) TdT–/– mice showed only one potential example of VH replacement and two examples of D-D rearrangements. Because VH replacement in the human takes place predominantly in cells undergoing receptor editing, we performed LM-PCR analysis of pre-B cells, immature B cells, and B cells from receptor editing cultures and from a mouse undergoing extensive polyclonal receptor editing in vivo (44). Our data indicate that DSBs at the cryptic RSS were very infrequent, leading us to conclude that VH replacement as a result of receptor editing signaling in immature bone marrow B cells is rare in the mouse. In addition, we show that D-D rearrangements are rare in the spleens of several strains of nonautoimmune and lupus-prone mice. In this latter survey, we included mice congenic for the Sle3/5(Lbw5) susceptibility loci because this interval has been reported to induce a high frequency of D-D rearrangements in CDR3 regions; however, we found none (6). Hence, we conclude that many apparent D-D rearrangements or VH replacements in the mouse are more likely to be fortuitous N region additions that match a portion of a VH footprint or of a germline D gene and that both events are rarely generated in murine B cells.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Mice

B6 TdT–/– mice were from the 10th backcross generation, and MRL/lpr TdT–/– mice were from the 15th backcross generation (45, 46). Mice expressing the 3-83 BCR transgene on the B10.D2 background and mice expressing the macroself Ag (anti-C{kappa}) transgene (pUlik line no. 2) have been described previously (44, 47, 48). MRL/lpr.H-2d/d 3-83 transgenic mice were made by backcrossing the 3-83 transgene from the B10.D2 background onto the MRL/lpr background, screening for the presence of the transgene at each backcross generation. Because the 3-83 transgene is reactive with H-2Kk and because MRL/lpr mice are H-2k, mice were also screened for heterozygosity at H-2 and were intercrossed at the seventh generation to generate the MRL/lpr.H-2d/d 3-83 transgenic mice used in these experiments. NZB.NZW-Lbw5 mice were generated by marker selected backcrossing of the New Zealand White (NZW) Lbw5 interval on chromosome 7 onto the New Zealand Black (NZB) background (N7 generations). The minimal NZW interval is defined by D7Mit294 (8 cM) and D7Mit31 (44 cM), which overlaps with Sle3/5. The NZW Lbw5 interval on the NZB background results in enhanced autoantibody production, glomerulonephritis, and hemolytic anemia consistent with the presence of Sle3/5 (D. H. Kono and A. N. Theofilopoulos, manuscript in preparation). All mice were maintained in our breeding colony at The Scripps Research Institute, and these animal studies were approved by The Scripps Institutional Animal Care and Use Committee.

Isolation of B cell subsets

Bone marrow was obtained from 6- to 8-wk-old mice. Bone marrow cells were sorted as pro/pre-B cells (CD19+IgM–), pre-B cells (B220+, CD43–IgM–), or immature B cells (B220+IgM+IgD– or B220lowIgM+) in the Flow Cytometry Facility at Scripps. Due to the presence of B220+CD4–CD8–TCR+ cells in the MRL/lpr mice, MRL/lpr bone marrow cells were always pre-enriched for CD19+ cells with anti-CD19-conjugated magnetic beads (Miltenyi Biotec) before cell sorting, and B6 B cell progenitors were often pre-enriched in this way also. The B cell progenitors from {kappa} macroself transgenic mice were sorted as B220+ cells. Marginal zone and follicular B cells were obtained from 9- to 11-wk-old mice by sorting for AA4–CD23low/negCD21bright cells and for AA4–CD21+CD23+ cells, respectively. All Abs were obtained from eBioscience, Biolegend, or BD Pharmingen.

Receptor editing cultures

These cultures were performed as described by Melamed et al. (49). Briefly, unfractionated bone marrow cells were cultured for 5 days in IMDM supplemented with 10% FCS and IL-7, which was obtained as a supernatant from IL-7-transfected cells given to us by A. Rolink (University of Basel, Basel, Switzerland). After 5 days, the cells were washed and cultured for 2 additional days in the absence of IL-7 and in the presence of anti-Id Ab or control Ab. DNA was made with a Qiagen kit (Qiagen).

LM-PCR

LM-PCR was performed as described previously (50). A total of 100–300 ng of genomic DNA prepared with a Qiagen kit was ligated to the BW linker (50) overnight. Nested PCRs were performed with AmpliTaq for one cycle of 95°C for 10 min, followed by 12–30 cycles of amplification at 94°C for 30 s, 59°C for 30 s, and 72°C for 1 min, then 10 min at 72°C. VH7183 primers, AF303 and AF215 (51), were located in FR1 and were used with the BWH primer, which is identical to the BW linker plus 3 bp of the heptamer (50). Five microliters of the primary PCR was amplified in a secondary PCR with nested primers for 18–30 cycles. The VH replacement PCR was run on an agarose gel, and bands that were ~275–300 bp were cloned. Control LM-PCRs were done with J{kappa} primers as previously described (50), using 12 cycles of primary PCR and 15–20 cycles of secondary PCR to ensure that proper linker ligation had occurred.

PCR, cloning, and sequencing

DNA and RNA were made with kits from Qiagen. cDNA was made with SuperScript reverse transcriptase (Invitrogen Life Technologies) using oligo(dT) primers. PCR conditions and primers for amplification of VH7183, VHJ558, and VHS107 rearrangements were performed as described previously (25). PCR products were gel purified, and the eluted DNA was ligated into pBluescript or Topo (Invitrogen Life Technologies) vectors. Minipreps were made and either sequenced by dideoxy sequencing (51) or sequenced by Seqwright (Fisher Scientific). In the analysis of the sequences, N nucleotides were defined as any nucleotide that could not be accounted for by VH, DH, or JH genes or their associated P nucleotides. D gene segments could all be accounted for by D genes in the noninverted orientation.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Absence of D-D rearrangements in CDR3 from pre-B cells in TdT-deficient mice

We previously showed that 2–4% of CDR3 sequences from adult pre-B cells contained sequences in which a second potential D segment could be observed (25). We arbitrarily defined a second D segment as 5 consecutive bp that were not accounted for by V or J genes or P nucleotides and that completely matched a germline D gene. In contrast, we previously demonstrated that newborn mice showed no CDR3 sequences with D-D rearrangements. Since the absence of N nucleotides makes the assignment of nucleotides in CDR3 unambiguous, we cloned and sequenced 424 rearrangements from 6- to 8-wk-old adult pro/pre-B cells (surface Ig–) from TdT-deficient MRL/lpr mice and TdT-deficient C57L/6 mice using family-specific VH primers from VHJ558, VHS107, and VH7183 genes. The VHJ558 and VH7183 gene families are very large and are located in the distal and proximal halves of the VH locus, respectively. Thus, this analysis covers the majority of the Ig repertoire. Pro/pre-B cells (surface Ig–) were used for the majority of the analysis, since we wished to determine the frequency with which CDR3 with two D segments were initially created. In addition, 94 sequences were analyzed from immature B cells from these two strains. Only two D-D rearrangements were observed in the 518 rearrangements from TdT-deficient B cell precursors (Table I, Fig. 1).


View this table:
[in this window]
[in a new window]
  		Table I. D-D rearrangements and VH replacement events in CDR3 of TdT–/– mice

 

Figure 1
View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 1. Sequences of CDR3 from TdT-deficient B progenitors containing potential D-D rearrangements and potential VH replacements. A, The sequences containing two D segments in CDR3. Nucleotides, which could be encoded by either of two adjacent coding regions and that are likely microhomologies used for recombination in the absence of N nucleotides, are listed in parentheses. P nucleotides are underlined. B, The one sequence which represents a potential VH replacement event.

 
Absence of footprints of receptor editing in CDR3 of TdT-deficient mice

In the absence of TdT-generated N nucleotides, any extra nucleotides present between the VH gene and DH gene are very evident. Thus, even if only 2–3 bp remain from the original CDR3 in a VH replacement event, they can be unambiguously identified. Since immature B cells become phenotypically pre-B cell (sIg–) after receptor editing is initiated and would remain in this compartment until a successful replacement event has occurred, it seemed that the pre-B compartment was a likely place to look for footprints of VH replacement. Any nonproductive or nonselectable VH replacement events would still be captured by PCR from the pre-B compartment. Therefore, we analyzed the 518 VDJ rearrangements from pre-B cells and immature B cells of TdT-deficient B6 and TdT-deficient MRL/lpr mice described above for footprints of VH replacement. VH71883 genes, as depicted in Fig. 2A, have potentially 7 bp (plus P nucleotides) in the footprint, VHS107 genes V1 and V11 have 8 and 6 bp, respectively, in their footprint, and VHJ558 genes have only 5 bp. This number can be increased by P nucleotides and can be decreased by deletion of coding ends. However, only one potential example of VH replacement was observed, and it only retained 2 bp of the footprint (Table I, Fig. 1). In addition, 1 of the 518 sequences had 2 bp (CG) at the V-D junction that could not be accounted for as either a D-D rearrangement or a VH replacement footprint, and 6 of the 518 sequences had one extra bp at the VH-DH junction, which is common for TdT-deficient mice. VDJ rearrangements from TdT-deficient mice have been shown to contain 1–3% of sequences with 1–2 bp in CDR3 that cannot be accounted for by V, D, or J genes or their associated P nucleotides (52, 53). A similar frequency (one example of 2 bp and three examples of 1 bp) of unaccounted nucleotides was observed at the D-J junctions. Of the six sequences with an extra single base at the V-D junction, four had a single T at the VH-DH junction and two had a G. Only the latter two could possibly have derived from a VH replacement event because there are no Ts in the footprint sequence. However, it is more likely that all are the few junctional sequences with a single unaccounted base pair seen in TdT-deficient mice (52, 53). Thus, we have minimal evidence for footprints of VH replacement in a situation in which the ambiguity of TdT-added nucleotides is absent.

LM-PCR analysis of ds breaks

To determine whether breaks at the cryptic heptamer in FR3 occur during receptor editing, we performed LM-PCR for breaks in VH7183 genes. Since the study of Zhang et al. (41) indicated that most DSBs at the cryptic heptamer occurred in human bone marrow predominantly at the immature B cell stage, the stage at which receptor editing is initiated, we initially performed PCR with DNA from these cells. To maximize our chance of detecting the VH replacement-related breaks at the cryptic RSS, the large VH7183 family was chosen because these VH genes are more likely than any other VH family to be accessible at the immature B cell stage (43, 54, 55). We sorted bone marrow cells into pre-B and immature B cell subpopulations and analyzed them by LM-PCR. Since the "signal end" for VH replacement is the former VH portion of the initial VDJ reaction, the primers used for this LM-PCR are located in FR1, as diagrammed in Fig. 2B. Thus, traditional VDJ recombination intermediates of a VH to DH recombination event at VH7183 genes are not amplified in these PCRs. Initial PCRs were done with 12 cycles in the primary PCR and 15–25 cycles in the secondary nested PCR. Strong bands were obtained in control PCRs on the linker-ligated DNA to detect RAG-mediated DSBs at J{kappa}, demonstrating that the DNA was properly linker ligated, but amplified products were not observed for breaks at the VH cryptic RSS. Therefore, we increased the number of cycles for both primary and secondary PCRs to detect potential ds breaks at the cryptic VH RSS, mainly using 30 cycles in each for the PCRs listed in Fig. 2C. Under these conditions, two-thirds of the PCRs still gave no bands, but we did observe some LM-PCR products of variable intensity that were consistent with breaks at the cryptic heptamer, although they often appeared slightly larger than expected. Other bands were also observed in some PCRs, most notably in the 100–125 bp range. A few of these were cloned and they did arise from breaks in the coding regions of VH7183 genes (data not shown). Such breaks in the coding region have been observed previously (34).

We cloned many of the darker bands that were approximately the right size to be VH replacement events. For each PCR that was cloned, we primarily obtained only one to two unique sequences, not surprisingly given the large number of cycles required to detect a band. Of the 48 PCRs done on immature B cells, 18 gave bands of approximately the correct size, 11 PCR products were cloned, but only 2 were at the cryptic RSS in FR3 (Fig. 2C). No breaks at the cryptic RSS were obtained from the cloned PCRs from pre-B cells. Hence, we conclude that RAG-mediated breaks at the cryptic RSS were rare in immature B cells.

To increase the proportion of cells undergoing receptor editing, we analyzed DNA from immature B cells derived from cultures undergoing receptor editing (49). In these cultures, bone marrow cells from BCR transgenic mice are cultured for 5 days in IL-7, by which time all cells are of the immature B cell phenotype. The cells are then washed and cultured for 2 more days in the absence of IL-7 and the presence of anti-Id Abs. This BCR ligation by the anti-Id Ab gives an effective receptor editing signal to all cells in the culture. Since these mice are BCR transgenic mice, they cannot undergo VH replacement. However, receptor editing signals result in up-regulation of RAG genes and rearrangement of endogenous {kappa} genes, demonstrating that the endogenous loci react appropriately to receptor editing signals (49). In our assay, we are merely looking for RAG-mediated breaks at endogenous VH genes, not for completed VH replacement events. As a second approach to enriching the number of cells undergoing receptor editing, we sorted B220+ cells from the bone marrow of {kappa}-macroself transgenic mice, which express membrane-bound synthetic self-Ag (anti-C{kappa}) in all cells (44). As a result, essentially all immature B cells in this mouse undergo receptor editing as soon as each cell expresses surface IgM using a {kappa} L chain. Importantly, the macroself Ag transgenic mice have an unmanipulated endogenous Ig locus, and thus normal repertoire formation and physiological receptor editing, albeit at a very high frequency, should be occurring in all bone marrow immature B cells. A somewhat higher percentage of PCR showed bands of approximately the correct size from the receptor editing cultures and macroself transgenic bone marrow cells, but, upon cloning, only two were to the cryptic RSS (Fig. 2C). Thus, even under these conditions of enrichment for cells undergoing receptor editing, only a few breaks were observed at the FR3 cryptic heptamer.

Fig. 2D shows the predominant location of the breaks that we cloned. Many occurred at a cryptic heptamer located in the middle of the spacer that is identical in sequence to the cryptic heptamer in FR3. However, the RSS spacer sequence is only present in unrearranged VH genes, and thus, these DSBs could not give rise to VH replacement. Thus, we observed only a relatively low incidence of DSBs at the cryptic RSS in populations of cells enriched for cells undergo receptor editing and none in the pre-B populations.

Germline VH gene created by VH replacement

We previously reported a novel VH7183 gene in MRL/lpr and C3H mice, both of which are Ighj haplotype (25). In this study, we have found this VH gene in another lupus-prone strain, NZB, which is also Ighj for most of the VH locus (56). It is of interest that this novel gene was created by a process similar to VH replacement, i.e., invasion of a germline VH7183 gene by an upstream VH7183 germline gene. (Fig. 3). No such gene with a germline VH replacement-generated VH gene is present in the VH7183 family in the nonautoimmune B6 (Ighb) or 129 strains (Igha). Curiously, we previously reported that MRL/lpr mice also have a novel DH gene that appears to be a germline D-D fusion (25). Given this propensity for aberrant gymnastics at the IgH locus in these lupus-prone strains, it is even more striking that VH replacement events or D-D rearrangement events in B cell precursors from the MRL/lpr mice are so rare.


Figure 3
View larger version (8K):
[in this window]
[in a new window]
  		FIGURE 3. Germline VH gene likely created by VH replacement. The possible origin of a germline VH7183 gene in Ighj haplotype is shown. In this scenario, an upstream VH gene joined to a downstream VH gene at the cryptic RSS in FR3.

 
Rare D-D rearrangements in CDR3 of Sle3/5 congenic mice and other lupus-prone and nonautoimmune mice in the bone marrow and spleen DNA

It has been reported that B6.Sle3/5 congenic mice have a very high frequency of CDR3 with D-D rearrangements in IgG cDNA, suggesting to the authors that this susceptibility interval results in the creation of abnormal CDR3 regions (6). Therefore, we analyzed IgH sequences from Sle3/5 congenic mice for these rearrangements using the NZB.NZW-Lbw5 congenic line, in which the Sle3/5 interval from NZW has been crossed onto the NZB background. To analyze these CDR3s, we first had to determine the germline sequences of the D genes in these NZB-derived mice. NZB mice had been reported to have a D locus of the Igha haplotype by Southern blotting pattern (56), and we confirmed this by PCR amplification, cloning, and sequencing of germline D genes (data not shown).

We first analyzed pre-B sequences from these mice to determine whether CDR3 sequences with D-D rearrangements were initially made at any higher frequency than in other strains we have studied. Twenty-nine sequences were analyzed, and only one potential D-D containing CDR3 sequence was observed (Table II, Fig. 4), a frequency consistent with the low frequency we observed in other strains (25). Next, we analyzed CDR3 sequences obtained from genomic DNA derived from cell sorter-purified splenic marginal zone B cells and follicular B cells from NZB.NZW-Lbw5 and from other lupus-prone strains, as well as from the nonautoimmune B6 strain. None of the splenic sequences from the lupus-prone mice had any potential D-D rearrangements, and only three sequences from B6 mice had potential V-D-D-J rearrangements (Table II, Fig. 4). Two of these have the D-D in the incorrect genomic order and thus are unlikely to be real D-D rearrangements. The third CDR3 contains a second potential D segment with a 5 bp match (GGGCT) with DST4.2. This D gene has a nonconsensus RSS and rearranges very infrequently (57). Since N nucleotides are G/C rich and often contain stretches of Gs, this GGGCT sequence is more likely to be an N nucleotide addition. Hence, it is unlikely that any of these three B6 sequences is a real V-D-D-J rearrangement.


View this table:
[in this window]
[in a new window]
  		Table II. V-D-D-J rearrangements in B cell subpopulations

 

Figure 4
View larger version (21K):
[in this window]
[in a new window]
  		FIGURE 4. Sequences of all potential V-D-D-J rearrangements from the splenic subpopulations described in Table II. P nucleotides are underlined. Since the sequence of D locus is known in Igha and Ighb haplotypes, we can determine whether the DH#1 segment is located 5' of the DH#2 segment, i.e., the correct genomic order for deletional rearrangement, as shown in the last column.

 
Lack of RAG mRNA in peripheral B cells and of D-D rearrangements in IgM and IgG mRNA

CDR3 sequences are made during V(D)J rearrangement, normally during the pro-B cell stage in the bone marrow. However, if receptor revision/editing takes place in the periphery, presumably during the course of an immune response, these unusual CDR3 sequences could be generated at this late stage. It was reported that mature B cells (AA4–) from B6.Sle3/5 congenic mice express RAG mRNA (6), so we prepared RNA from the sorted subpopulations of splenic B cells from NZB.NZW-Lbw5 mice and analyzed them for expression of RAG mRNA by real-time PCR. However, we did not detect any RAG transcripts in follicular B cells or marginal zone B cells from NZB.NZW-Lbw5 mice, nor from any other strains of mice (data not shown). RAG transcripts were also absent from peritoneal cells of NZB mice (data not shown). Due to the sensitivity of real-time PCR, we conclude that there is essentially no RAG mRNA in the periphery of these mice. This does not, however, preclude the potential existence of RAG expression in a very small proportion of cells such as those in the midst of an immune response.

Since the D-D rearrangements in the B6.Sle3/5 mice were all obtained from IgG cDNA after NP-immunization (6), we also analyzed IgG mRNA from marginal zone cells of unimmunized NZB.NZW-Lbw5 mice. These IgG CDR3 sequences presumably all derived from B cells that had been stimulated by any of a variety of environmental Ags at some time in their history. It can be difficult to definitively analyze CDR3 regions from mutated Abs, but we were able to analyze most of these specific CDR3 sequences quite accurately. A few changes from the germline sequence of VH and JH coding ends were observed, consistent with the likely presence of somatic hypermutation. Nonetheless, we did not observe any CDR3 sequences with any potential D-D rearrangements from these IgG cDNA sequences (Table II). We also analyzed IgM cDNA sequences from MRL/lpr mice, and found one potential V-D-D-J rearrangement out of 37 sequences (Table II, Fig. 4). Thus, in all splenic populations analyzed, D-D rearrangements in CDR3 are rare.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
The existence of D-D rearrangements in CDR3, particularly in lupus-prone mice or in autoantibodies, has often been cited, despite the fact that the creation of such a rearrangement violates the 12–23 bp rule for effective V(D)J recombination. Our hypothesis is that the second D segment is more likely to be a fortuitous match of a stretch of nucleotides added by TdT with a short region of a germline D segment. This hypothesis was strengthened by our previous observation of the absence of any potential D-D rearrangements in neonatal pre-B cells that naturally lack N regions due to the absence of expression of TdT early in ontogeny, although we did observe 2–4.6% of sequences from adult pre-B cells that contained potential D-D rearrangements in that previous study (25). Therefore, to test our hypothesis, we cloned and sequenced 518 rearrangements from adult pre-B cells and immature B cells from TdT-deficient mice, in which the origin of any nucleotides in the CDR3 must be germline. Only two of these sequences had any potential D-D rearrangements (0.039%). Although we cannot exclude the unlikely hypothesis that TdT is required to facilitate the creation of D-D rearrangements, we conclude that it is more likely that many potential D-D rearrangements in murine CDR3 are due to N region additions instead.

We also analyzed these same sequences for the presence of the footprint of VH replacement. VH replacement has been shown in the human to be uniquely prevalent in immature B cells, the stage at which receptor editing occurs. Since the absence of N nucleotides in the CDR3 of TdT-deficient mice makes identification of even a short footprint of VH replacement unambiguous, we also analyzed these CDR3 sequences for VH replacement events. Studies using 3H9 anti-dsDNA transgenic or knock-in mice suggested that there may be more H chain editing in MRL/lpr mice than in nonautoimmune mice bearing the same receptor (39, 58, 59), so the CDR3 sequences from TdT-deficient mice on the lupus-prone MRL/lpr background were of particular interest. However, we found only one potential example of VH replacement from the 518 sequences. It is, however, likely that we are underestimating the number of potential VH replacements because, in the absence of TdT, junctions are often made at the site of short 1- to 3-bp sequence homologies (60, 61). When VH genes are joined to a footprint coding end during VH replacement, the most likely homology to be used would be the CA at the terminus of both coding ends, which would result in a duplication of the CAAGA sequence and would produce an out-of-frame rearrangement. This CA at the end of all VH7183 genes is retained in the vast majority of these sequences and thus would be available for homology-directed recombination. Such nonproductive VH replacement events would be absent at later stages of B cell development but would be captured in PCR from the pre-B cell DNA, which is the reason why we predominantly analyzed pre-B cells. However, if the CAAGA sequence further 5' was used for homology-directed recombination event, any potential VH replacement events would not be evident, becoming masked by the recombination event. Such homology-directed recombination has been observed by K. Rajewsky and colleagues (unpublished observation) during VH replacement. This homology-directed recombination mechanism is unlikely to significantly affect our analysis of D-D rearrangements, however, since the D gene coding ends are more diverse, and the DH genes are much longer.

We also examined many populations of cells enriched in cells undergoing receptor editing, as well as normal pre-B and immature B cell populations, for RAG-mediated breaks at the cryptic heptamer, the approach used to detect frequent VH replacement in human bone marrow. Although DSBs were always readily observed at the J{kappa} RSS, only a small number of breaks were found at the cryptic heptamer of the H chain despite many more rounds of amplification. We did not observe any breaks in our samples of pre-B cells suggesting that, although VH replacement is rare in the mouse as a result of receptor editing, it is even less common in pre-B cells. The number of ds breaks events is so low that relative values are hard to evaluate with any precision. Nonetheless, it is clear that such events are relatively rare.

Any VH replacement events created in the bone marrow are unlikely to make it into the peripheral repertoire of normal mice for several reasons. First, VH replacement events have the chance of occurring at sites other than the cryptic RSS in FR3, and such rearrangements would all be nonfunctional (34). Second, even with replacement into the FR3 cryptic RSS in the correct reading frame, the new CDR3 would be longer, perhaps long enough that it would be selected against (62). Third, the 7 bp that can potentially be retained from the primary VDJ, i.e., the footprint, consists of the sequence encoding the highly conserved alanine and arginine at the end of FR3. Thus, VH replacement events would create VDDJ sequences that are more likely to have two arginines in FR3/CDR3: the arginine from the VH replacement footprint and the one from the upstream invading VH gene. Anti-DNA autoantibodies are enriched in arginines in the CDRs (63), and thus CDR3s with more than one arginine are more likely to be selected against as being potentially autoreactive before leaving the bone marrow. If, however, tolerance is compromised, such as might happen in an autoimmune-prone mouse, then B cells with these rare VH replacement events with an additional arginine in CDR3 might escape into the periphery and give rise to an expanded autoimmune clone. Under normal circumstances, however, if VH replacement is stimulated by receptor editing at the immature B cell stage, any rearrangement that is nonfunctional, or is destined to be eliminated by subsequent receptor editing or clonal deletion, would only be found in the phenotypic pre-B compartment and not in the spleen. For this reason, we predominantly analyzed sequences from sIg– pre-B cells for the footprint of VH replacement. However, we found only one sequence with a very short potential VH replacement event in the 444 analyzed pre-B cells.

It is not clear why human immature B cells appear to have such a relatively high frequency of VH replacement in the immature B cell stage compared with murine immature B cells. In both humans and mice, examples of receptor editing have been well documented, but in the majority of cases, it is unclear if the VH replacement occurred in the bone marrow or the periphery because the sequences are often obtained from peripheral B cells (30, 32). It has been shown that culture of spleen cells with IL-7 or IL-4 and anti-CD40 results in increased accessibility of the largeVHJ558 gene family, and thus it is possible that during the course of an immune response these signals can induce transient accessibility of VH genes in peripheral B cells, allowing VH replacement (43). It has also been shown that anti-CD40 and IL-7 leads to RAG up-regulation in splenic B cells (64), although this initial report has not been confirmed. Thus, it is conceivable that some higher level of VH replacement could occur in the periphery during the short window of time when B cells are actively engaged in an immune response, although the extent of such VH replacement would still be limited by the poor efficiency of the cryptic RSS.

We also found little evidence for D-D rearrangements in the peripheral cells in either genomic DNA or cDNA. It should be noted that our definition of a D-D rearrangement has been arbitrarily set with the second D segment containing a minimum of 5 bp of the D gene. The average length of a D segment in murine CDR3 is 9–10 bp. The second potential D segments seen here and also in our previous study is almost always 5 or 6 bp. Again, this supports our argument that the identity with the second D segment is often fortuitous. However, D-to-D rearrangements have been detected at the genomic level (12) and hence can theoretically contribute at a low level to the murine B cell repertoire.

The apparent frequency of V-D-D-J rearrangements varies widely among studies. One of the reasons is that the CDR3 analysis of any sequences derived from somatically mutated B cells is fraught with uncertainty due to the presence of somatic mutation. Investigators who analyze mutated sequences sometimes gave allowances for a mismatch in identifying D segments (21, 65, 66). Thus, these analyses are likely to inadvertently identify a significant number of false positive D-D rearrangements (8, 14). One way to avoid this uncertainty is to analyze unmutated CDR3 sequences. We and others have analyzed surface Ig– pro/pre-B cells as a means of analyzing sequences that have not been subject to any selection (other than binding to surrogate light chains). Even here, however, there can be some ambiguity. For example, much of the claim for high frequencies of D-D rearrangements stems from absence of information concerning the germline D gene sequences for the particular strain of mice being analyzed. In a large repertoire study of MRL/lpr pre-B cells, it was reported that a very high frequency (~33%) of IgH rearrangements contained D-D rearrangements (5, 7). However, the germline D genes of the MRL/lpr mouse had not been cloned at that time. We subsequently cloned these D genes, revealing the existence in the Ighj haplotype of MRL/lpr mice of two new germline DH genes, one of which resembles a D-D rearrangement (25). With the knowledge of the germline genes in hand, we reanalyzed the CDR3 sequences previously described as containing D-D rearrangements, as well as new sequences that we cloned, but we found only 2–4% of the sequences contained potential V-D-D-J rearrangements. Likewise, many of the D genes invoked as being in D-D rearrangements in B6.Sle3/5 mice are not present in the B6 genome, which is now fully sequenced (6, 67), so those rearrangements are unlikely to be D-D rearrangements (6). This latter study also included many expanded clones of cells in the B6.Sle3/5 mice that also contributed to the very high frequency of apparent D-D rearrangements. From our analysis of NZB.NZW-Lbw5 congenic mice encompassing most of the Sle3/5 locus, we conclude that the Sle3/5 lupus susceptibility interval does not lead to an increased frequency of CDR3 containing two D segments. However, NZB.NZW-Lbw5 mice and the B6.Sle3/5 mice do differ in the background genes, which could affect the results. Also, it should be noted that the Lbw5 interval analyzed here is slightly shorter than Sle3/5 on the centromeric end where the Sle5 interval is located. Lbw5 extends from between D7Mit340 (1.2 cM) and D7Mit294 (8 cM), whereas Sle3/5 extends to D7Mit178 (0.5 cM). Thus, it is possible that part of the Sle5 subinterval is not present in the Lbw5 interval. The Lbw5 interval, however, does include the CD22a polymorphism (9 cM) associated with reduced levels of CD22 in NZW compared with B6 mice (68). Due to the presence of increased disease susceptibility in the NZB.NZW-Lbw5 mice, it is likely that the two congenic lines are similar.

If D-D rearrangements or VH replacement events are primarily created during peripheral receptor revision, we would not expect to see them in the bone marrow pre-B cells. However, the existence of peripheral receptor editing is uncertain (69). There have been reports of RAG expression in peripheral B cells in lupus-prone mice (6, 70), which could be very important due to the potential for continuing rearrangement in these mice (7). Furthermore, continuing V(D)J rearrangement in the periphery, particularly in lupus-prone strains, may be under less stringent tolerance checkpoint control and could thus contribute to the induction of autoimmune disease (71, 72). Since autoantibodies and polyreactive Abs often have longer CDR3s and since D-D rearrangements would by necessity create longer CDR3s, this is a relevant issue (18, 19, 20, 21, 22, 23, 24). It should be noted that D-D rearrangements could only be made during receptor revision if the original productive rearrangement underwent a somatic mutation that rendered it nonproductive and if the second allele containing a D-J rearrangement underwent a V to D and D to DJ rearrangement. Using real-time PCR, we have been unable to detect any RAG expression in purified peripheral subpopulations of lupus-prone mice, although this does not preclude expression of RAG in a small subpopulation of peripheral cells. It is possible that variability in the environment of different animal colonies, and the pathogen load, might contribute to this difference in peripheral RAG expression reported from different labs. Nonetheless, we find little evidence for much peripheral RAG expression.

DH genes in the human are much longer than murine D segments. There have been reports of relatively high frequency of D-D rearrangements in the very small subpopulation of circulating B cells that simultaneously express the surrogate L chain Vpre-B as well as traditional light chains, and the sequences published from these rearrangements contain a second potential D segment that is much longer than any of the second D segments in any murine D-D rearrangements (9). Hence, whether there are differences in the frequency of D-D rearrangements in human vs mice is not clear. However, here we clearly show that, in the absence of TdT, there are only 2 sequences with V-D-D-J rearrangements from the 518 sequences (0.039%) from murine adult pre-B and immature B cells. Furthermore, we find very few V-D-D-J rearrangements in a mouse congenic for a lupus susceptibility interval Sle3/5 on the NZB background. We conclude that many examples of use of two D segments in CDR3 are likely to be due to fortuitous matching of N nucleotides to a germline gene.


   Acknowledgments
 
We thank Drs. Klaus Rajewsky and Martin Weigert for helpful discussions and also thank Dr. Klaus Rajewsky for sharing unpublished data. We also thank Sara Salerno for excellent technical assistance.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 This work was supported by National Institutes of Health Grants R01 AI61167, R01 AI29672, and R01 AR42242. R.Z.M. was supported by National Institutes of Health Grant T34 GM08303, and D.A.-A. was supported by National Institutes of Health Grant T32 HL07195. Back

2 Address correspondence and reprint requests to Dr. Ann J. Feeney, The Scripps Research Institute, Department of Immunology, IMM-22, 10550 North Torrey Pines Road, La Jolla, CA 92037. E-mail address: feeney{at}scripps.edu Back

3 Abbreviations used in this paper: RSS, recombination signal sequence; LM-PCR, ligation-mediated PCR; DSB, double strand break; NZW, New Zealand White; NZB, New Zealand Black. Back

Received for publication March 8, 2006. Accepted for publication April 21, 2006.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Akamatsu, Y., N. Tsurushita, F. Nagawa, M. Matsuoka, K. Okazaki, M. Imai, H. Sakano. 1994. Essential residues in V(D)J recombination signals. J. Immunol. 153: 4520-4529. [Abstract]
  2. Hesse, J. E., M. R. Lieber, K. Mizuuchi, M. Gellert. 1989. V(D)J recombination: a functional definition of the joining signals. Genes Dev. 3: 1053-1061. [Abstract/Free Full Text]
  3. Akira, S., K. Okazaki, H. Sakano. 1987. Two pairs of recombination signals are sufficient to cause immunoglobulin V-(D)-J joining. Science 238: 1134-1138. [Abstract/Free Full Text]
  4. Koralov, S. B., T. I. Novobrantseva, K. Hochedlinger, R. Jaenisch, K. Rajewsky. 2005. Direct in vivo VH to JH rearrangement violating the 12/23 rule. J. Exp. Med. 201: 341-348. [Abstract/Free Full Text]
  5. Klonowski, K. D., L. L. Primiano, M. Monestier. 1999. Atypical VH-D-JH rearrangements in newborn autoimmune MRL mice. J. Immunol. 162: 1566-1572. [Abstract/Free Full Text]
  6. Wakui, M., L. Morel, E. J. Butfiloski, C. Kim, E. S. Sobel. 2005. Genetic dissection of systemic lupus erythematosus pathogenesis: partial functional complementation between Sle1 and Sle3/5 demonstrates requirement for intracellular coexpression for full phenotypic expression of lupus. J. Immunol. 175: 1337-1345. [Abstract/Free Full Text]
  7. Klonowski, K. D., M. Monestier. 2000. Heavy chain revision in MRL mice: a potential mechanism for the development of autoreactive B cell precursors. J. Immunol. 165: 4487-4493. [Abstract/Free Full Text]
  8. Corbett, S. J., I. M. Tomlinson, E. L. L. Sonnhammer, D. Buck, G. Winter. 1997. Sequence of the human immunoglobulin diversity (D) segment locus: a systematic analysis provides no evidence for the use of DIR segments, inverted D segments, "minor" D segments or D-D recombination. J. Mol. Biol. 270: 587-597. [Medline]
  9. Meffre, E., E. Davis, C. Schiff, C. Cunningham-Rundles, L. B. Ivashkiv, L. M. Staudt, J. W. Young, M. C. Nussenzweig. 2000. Circulating human B cells that express surrogate light chains and edited receptors. Nat. Immunol. 1: 207-213. [Medline]
  10. Meffre, E., M. Chiorazzi, M. C. Nussenzweig. 2001. Circulating human B cells that express surrogate light chains display a unique antibody repertoire. J. Immunol. 167: 2151-2156. [Abstract/Free Full Text]
  11. Collins, A. M., M. Ikutani, D. Puiu, G. A. Buck, A. Nadkarni, B. Gaeta. 2004. Partitioning of rearranged Ig genes by mutation analysis demonstrates D-D fusion and V gene replacement in the expressed human repertoire. J. Immunol. 172: 340-348. [Abstract/Free Full Text]
  12. Meek, K. D., C. A. Hasemann, J. D. Capra. 1989. Novel rearrangements at the immunoglobulin D locus: inversions and fusions add to IgH somatic diversity. J. Exp. Med. 170: 39-57. [Abstract/Free Full Text]
  13. Feeney, A. J.. 1990. Lack of N regions in fetal and neonatal mouse immunoglobulin V-D-J junctional sequences. J. Exp. Med. 172: 1377-1390. [Abstract/Free Full Text]
  14. Souto-Carneiro, M. M., N. S. Longo, D. E. Russ, H. W. Sun, P. E. Lipsky. 2004. Characterization of the human Ig heavy chain antigen binding complementarity determining region 3 using a newly developed software algorithm, JOINSOLVER. J. Immunol. 172: 6790-6802. [Abstract/Free Full Text]
  15. Sanz, I.. 1991. Multiple mechanisms participate in the generation of diversity of human H chain CDR3 regions. J. Immunol. 147: 1720-1729. [Abstract]
  16. Yamada, M., R. Wasserman, B. A. Reichard, S. Shane, A. J. Caton, G. Rovera. 1991. Preferential utilization of specific immunoglobulin heavy chain diversity and joining segments in adult human peripheral blood B lymphocytes. J. Exp. Med. 173: 395-407. [Abstract/Free Full Text]
  17. Moore, B. B., K. Meek. 1995. Recombination potential of the human DIR elements. J. Immunol. 154: 2175-2187. [Abstract]
  18. Wardemann, H., S. Yurasov, A. Schaefer, J. W. Young, E. Meffre, M. C. Nussenzweig. 2003. Predominant autoantibody production by early human B cell precursors. Science 301: 1374-1377. [Abstract/Free Full Text]
  19. Ichiyoshi, Y., P. Casali. 1994. Analysis of the structural correlates for antibody polyreactivity by multiple reassortments of chimeric human immunoglobulin heavy and light chain V segments. J. Exp. Med. 180: 885-895. [Abstract/Free Full Text]
  20. Crouzier, R., T. Martin, J. L. Pasquali. 1995. Heavy chain variable region, light chain variable region, and heavy chain CDR3 influences on the mono- and polyreactivity and on the affinity of human monoclonal rheumatoid factors. J. Immunol. 154: 4526-4535. [Abstract]
  21. Shlomchik, M., M. Mascelli, H. Shan, M. Z. Radic, D. Pisetsky, A. Marshak-Rothstein, M. Weigert. 1990. Anti-DNA antibodies from autoimmune mice arise by clonal expansion and somatic mutation. J. Exp. Med. 171: 265-297. [Abstract/Free Full Text]
  22. Eilat, D., D. M. Webster, A. R. Rees. 1988. V region sequences of anti-DNA and anti-RNA autoantibodies from NZB/NZW F1 mice. J. Immunol. 141: 1745-1753. [Abstract]
  23. Eilat, D., R. Fischel. 1991. Recurrent utilization of genetic elements in V regions of antinucleic acid antibodies from autoimmune mice. J. Immunol. 147: 361-368. [Abstract]
  24. Gangemi, R. M., A. K. Singh, K. J. Barrett. 1993. Independently derived IgG anti-DNA autoantibodies from two lupus-prone mouse strains express a VH gene that is not present in most murine strains. J. Immunol. 151: 4660-4671. [Abstract]
  25. Kompfner, E., P. Oliveira, A. Montalbano, A. J. Feeney. 2001. Unusual germline DSP2 gene accounts for all apparent V-D-D-J rearrangements in newborn, but not adult, MRL mice. J. Immunol. 167: 6933-6938. [Abstract/Free Full Text]
  26. Reth, M., P. Gehrmann, E. Petrac, P. Wiese. 1986. A novel VH to VHDJH joining mechanism in heavy-chain-negative (null) pre-B cells results in heavy-chain production. Nature 322: 840-842. [Medline]
  27. Kleinfield, R., R. R. Hardy, D. Tarlinton, J. Dangl, L. A. Herzenberg, M. Weigert. 1986. Recombination between an expressed immunoglobulin heavy-chain gene and a germline variable gene segment in a Ly 1+ B cell lymphoma. Nature 322: 843-846. [Medline]
  28. Covey, L. R., P. Ferrier, F. W. Alt. 1990. VH to VHDJH rearrangement is mediated by the internal VH heptamer. Int. Immunol. 2: 580-582.
  29. Nadel, B., A. Tang, A. J. Feeney. 1998. VH replacement is unlikely to contribute significantly to receptor editing due to an ineffectual embedded recombination signal sequence. Mol. Immunol. 35: 227-232. [Medline]
  30. Itoh, K., E. Meffre, E. Albesiano, A. Farber, D. Dines, P. Stein, S. E. Asnis, R. A. Furie, R. I. Jain, N. Chiorazzi. 2000. Immunoglobulin heavy chain variable region gene replacement as a mechanism for receptor revision in rheumatoid arthritis synovial tissue B lymphocytes. J. Exp. Med. 192: 1151-1164. [Abstract/Free Full Text]
  31. Wilson, P. C., K. Wilson, Y. J. Liu, J. Banchereau, V. Pascual, J. D. Capra. 2000. Receptor revision of immunoglobulin heavy chain variable region genes in normal human B lymphocytes. J. Exp. Med. 191: 1881-1894. [Abstract/Free Full Text]
  32. Chen, C., Z. Nagy, E. Luning Prak, M. Weigert. 1995. Immunoglobulin heavy chain gene replacement: a mechanism of receptor editing. Immunity 3: 747-755. [Medline]
  33. Cascalho, M., J. Wong, M. Wabl. 1997. VH gene replacement in hyperselected B cells of quasimonoclonal mice. J. Immunol. 159: 5795-5801. [Abstract]
  34. Taki, S., F. Schwenk, K. Rajewsky. 1995. Rearrangement of upstream DH and VH genes to a rearranged immunoglobulin variable region gene inserted into the DQ52-JH region of the immunoglobulin heavy chain locus. Eur. J. Immunol. 25: 1888-1896. [Medline]
  35. Bertrand, F. E., R. Golub, G. E. Wu. 1998. VH gene replacement occurs in the spleen and bone marrow of non-autoimmune quasimonoclonal mice. Eur. J. Immunol. 28: 3362-3370. [Medline]
  36. Madan, M. K., R. Golub, M. Wabl, G. E. Wu. 2000. Specific antibody production by VH gene replacement. Eur. J. Immunol. 30: 2404-2411. [Medline]
  37. Li, J., E. D. Geissal, W. Li, B. D. Stollar. 2005. Repertoire diversification in mice with an IgH-locus-targeted transgene for the rearranged VH domain of a physiologically selected anti-ssDNA antibody. Mol. Immunol. 42: 1475-1484. [Medline]
  38. Sekiguchi, D. R., R. A. Eisenberg, M. Weigert. 2003. Secondary heavy chain rearrangement: a mechanism for generating anti-double-stranded DNA B cells. J. Exp. Med. 197: 27-39. [Abstract/Free Full Text]
  39. Li, Y., H. Li, D. Ni, M. Weigert. 2002. Anti-DNA B cells in MRL/lpr mice show altered differentiation and editing pattern. J. Exp. Med. 196: 1543-1552. [Abstract/Free Full Text]
  40. Phan, T. G., M. Amesbury, S. Gardam, J. Crosbie, J. Hasbold, P. D. Hodgkin, A. Basten, R. Brink. 2003. B cell receptor-independent stimuli trigger immunoglobulin (Ig) class switch recombination and production of IgG autoantibodies by anergic self-reactive B cells. J. Exp. Med. 197: 845-860. [Abstract/Free Full Text]
  41. Zhang, Z., M. Zemlin, Y. H. Wang, D. Munfus, L. E. Huye, H. W. Findley, S. L. Bridges, D. B. Roth, P. D. Burrows, M. D. Cooper. 2003. Contribution of Vh gene replacement to the primary B cell repertoire. Immunity 19: 21-31. [Medline]
  42. Nemazee, D.. 2000. Receptor editing in B cells. Adv. Immunol. 74: 89-126. [Medline]
  43. Chowdhury, D., R. Sen. 2003. Transient IL-7/IL-7R signaling provides a mechanism for feedback inhibition of immunoglobulin heavy chain gene rearrangements. Immunity 18: 229-241. [Medline]
  44. Ait-Azzouzene, D., L. Verkoczy, J. Peters, A. Gavin, P. Skog, J. L. Vela, D. Nemazee. 2005. An immunoglobulin C {kappa}-reactive single chain antibody fusion protein induces tolerance through receptor editing in a normal polyclonal immune system. J. Exp. Med. 201: 817-828. [Abstract/Free Full Text]
  45. Feeney, A. J., B. R. Lawson, D. H. Kono, A. N. Theofilopoulos. 2001. Terminal deoxynucleotidyl transferase deficiency decreases autoimmune disease in MRL-Faslpr mice. J. Immunol. 167: 3486-3493. [Abstract/Free Full Text]
  46. Robey, I. F., M. Peterson, M. S. Horwitz, D. H. Kono, T. Stratmann, A. N. Theofilopoulos, N. Sarvetnick, L. Teyton, A. J. Feeney. 2004. Terminal deoxynucleotidyltransferase deficiency decreases autoimmune disease in diabetes-prone nonobese diabetic mice and lupus-prone MRL-Faslpr mice. J. Immunol. 172: 4624-4629. [Abstract/Free Full Text]
  47. Russell, D. M., Z. Dembic, G. Morahan, J. F. Miller, K. Burki, D. Nemazee. 1991. Peripheral deletion of self-reactive B cells. Nature 354: 308-311. [Medline]
  48. Tiegs, S. L., D. M. Russell, D. Nemazee. 1993. Receptor editing in self-reactive bone marrow B cells. J. Exp. Med. 177: 1009-1020. [Abstract/Free Full Text]
  49. Melamed, D., D. Nemazee. 1997. Self-antigen does not accelerate immature B cell apoptosis, but stimulates receptor editing as a consequence of developmental arrest. Proc. Natl. Acad. Sci. USA 94: 9267-9272. [Abstract/Free Full Text]
  50. Schlissel, M., A. Constantinescu, T. Morrow, M. Baxter, A. Peng. 1993. Double-strand signal sequence breaks in V(D)J recombination are blunt, 5'-phosphorylated, RAG-dependent, and cell cycle regulated. Genes Dev. 7: 2520-2532. [Abstract/Free Full Text]
  51. Williams, G. W., A. Martinez, A. Montalbano, A. Tang, A. Mauhar, K. M. Ogwaro, D. Merz, C. Chevillard, R. Riblet, A. J. Feeney. 2001. Unequal VH gene rearrangement frequency within the large VH7183 gene family is not due to RSS variation, and mapping of the genes shows a bias of rearrangement based on chromosomal location. J. Immunol. 167: 257-263. [Abstract/Free Full Text]
  52. Komori, T., A. Okada, V. Stewart, F. W. Alt. 1993. Lack of N regions in antigen receptor variable region genes of TdT-deficient lymphocytes. Science 261: 1171-1175. [Abstract/Free Full Text]
  53. Gilfillan, S., A. Dierich, M. Lemeur, C. Benoist, D. Mathis. 1993. Mice lacking TdT: mature animals with an immature lymphocyte repertoire. Science 261: 1175-1178. [Abstract/Free Full Text]
  54. Costa, T. E. F., H. Suh, M. C. Nussenzweig. 1992. Chromosomal position of rearranging gene segments influences allelic exclusion in transgenic mice. Proc. Natl. Acad. Sci. USA 89: 2205-2208. [Abstract/Free Full Text]
  55. Roldan, E., M. Fuxa, W. Chong, D. Martinez, M. Novatchkova, M. Busslinger, J. A. Skok. 2005. Locus "decontraction" and centromeric recruitment contribute to allelic exclusion of the immunoglobulin heavy-chain gene. Nat. Immunol. 6: 31-41. [Medline]
  56. Brodeur, P., R. Riblet. 1984. The immunoglobulin heavy chain variable region (Igh-V) locus in the mouse. I. One hundred Igh-V genes comprise seven families of homologous genes. Eur. J. Immunol. 14: 922-930. [Medline]
  57. Feeney, A. J., R. Riblet. 1993. DST4: a new, and probably the last, functional DH gene in the BALB/c mouse. Immunogenetics 37: 217-221. [Medline]
  58. Brard, F., M. Shannon, E. L. Prak, S. Litwin, M. Weigert. 1999. Somatic mutation and light chain rearrangement generate autoimmunity in anti-single-stranded DNA transgenic MRL/lpr mice. J. Exp. Med. 190: 691-704. [Abstract/Free Full Text]
  59. Roark, J. H., C. L. Kuntz, K. A. Nguyen, L. Mandik, M. Cattermole, J. Erikson. 1995. B cell selection and allelic exclusion of an anti-DNA Ig transgene in MRL-lpr/lpr mice. J. Immunol. 154: 4444-4455. [Abstract]
  60. Gu, H., I. Förster, K. Rajewsky. 1990. Sequence homologies, N sequence insertion and JH gene utilization in VHDJH joining: implications for the joining mechanism and the ontogenetic timing of Ly1 B cell and B-CLL progenitor generation. EMBO J. 9: 2133-2140. [Medline]
  61. Feeney, A. J.. 1992. Predominance of VH-D-JH junctions occurring at sites of short sequence homology results in limited junctional diversity in neonatal antibodies. J. Immunol. 149: 222-229. [Abstract]
  62. Ivanov, I. I., R. L. Schelonka, Y. Zhuang, G. L. Gartland, M. Zemlin, H. W. Schroeder, Jr. 2005. Development of the expressed Ig CDR-H3 repertoire is marked by focusing of constraints in length, amino acid use, and charge that are first established in early B cell progenitors. J. Immunol. 174: 7773-7780. [Abstract/Free Full Text]
  63. Radic, M. Z., J. Mackle, J. Erikson, C. Mol, W. F. Anderson, M. Weigert. 1993. Residues that mediate DNA binding of autoimmune antibodies. J. Immunol. 150: 4966-4977. [Abstract]
  64. Hikida, M., Y. Nakayama, Y. Yamashita, Y. Kumazawa, S. I. Nishikawa, H. Ohmori. 1998. Expression of recombination activating genes in germinal center B cells: involvement of interleukin 7 (IL-7) and the IL-7 receptor. J. Exp. Med. 188: 365-372. [Abstract/Free Full Text]
  65. Mortari, F., J. A. Newton, J. Y. Wang, H. W. Schroeder, Jr. 1992. The human cord blood antibody repertoire: frequent usage of the VH7 gene family. Eur. J. Immunol. 22: 241-245. [Medline]
  66. Brezinschek