- Email: [email protected]
VH replacement in mice and humans
Review
TRENDS in Immunology
Vol.28 No.3
VH replacement in mice and humans Zhixin Zhang Division of Developmental and Clinical Immunology, Departments of Medicine and Microbiology, University of Alabama at Birmingham, Birmingham, AL 35294, USA
Heavy chain variable segment (VH) replacement refers to recombination activating gene (RAG) product-mediated secondary recombination between a previously rearranged VH gene and an upstream unrearranged VH gene. VH replacement was first observed in mouse pre-B cell lines and later demonstrated in knock-in mouse models carrying immunoglobulin heavy chain (IgH) genes encoding self-reactive or mono-specific antibodies or non-functional IgH rearrangements on both IgH alleles. Despite these findings, it is still difficult to find VH replacement intermediates during normal murine B cell development. In humans, ongoing VH replacement was found in a clonal B lineage EU12 cell line and in human bone marrow immature B cells. The identification of potential VH replacement products also suggested a potential contribution of VH replacement to the antibody repertoire. Here, I review the evidence for whether VH replacement genuinely offers an in vivo RAG-mediated recombinatorial mechanism to alter preformed IgH genes in mice and humans. Generating a diversified antibody repertoire through V(D)J recombination To combat infectious agents, a diversified antibody repertoire is generated through somatic recombination of previously separated variable (V), diversity (D) [for the heavy (H) chain only] and joining (J) gene segments to form the variable domain exons of immunoglobulin (Ig) genes in developing B lineage cells [1–4]. V(D)J recombination is catalyzed by a pair of recombination activating gene products, RAG1 and RAG2 [5,6]. Specific recombination of the V, D and J gene segments is directed by the recombination signal sequences (RSS), which contain a highly conserved heptamer (50 -CACTGTG-30 ) and a nonamer (50 -ACAAAAACC-30 ) separated by a non-conserved spacer region either 12 bp or 23 bp in length [7] (Figure 1a). The heptamer and nonamer motifs are required for RAG1 and RAG2 binding and subsequent DNA cleavage [8,9]. The RSS spacer regions are important for directing the specific formation of the coding joints. As dictated by the 12/23 rule, coding joints preferentially form between a pair of 12-bp and 23-bp RSS-tagged gene segments for Ig and T cell receptor (TCR) genes [7]. The random nature of V(D)J recombination is essential for generating a diversified antibody repertoire; however, it can also produce unwanted Ig genes. First, owing to the imprecise joining of the coding ends, two-thirds of the VHDJH and VL-JL joints will be out of reading frame and Corresponding author: Zhang, Z. ([email protected]). Available online 26 January 2007. www.sciencedirect.com
unable to produce Ig peptides. Second, functionally rearranged VHDJH genes might produce m heavy chains that fail to pair with surrogate or conventional light chains to form the pre-B cell receptor (pre-BCR) or BCR, respectively. Third, the random nature of V(D)J recombination can also generate Ig genes encoding self-reactive BCRs. Such unwanted rearrangements must be changed to enable continuous B cell development and maturation. Editing the Ig light chain genes Receptor editing refers to RAG-mediated secondary recombination of Ig genes to alter the previously rearranged Ig genes. Most previous studies of receptor editing focused mainly on the Ig light chain (IgL) genes, because Vk and Jk gene segments are tagged by complimentary RSS sites and, thus, the organization of the mouse and human Igk loci enables continuous secondary rearrangements by simply joining an upstream Vk with a downstream Jk gene segment to delete the preformed VkJk joints. As an additional choice, B cells can delete the Igk locus and initiate de novo rearrangement of the Igl locus. For human Igl loci, there are three functional Jl and Cl clusters for additional recombinations with upstream Vl genes. The mouse Igl loci have limited numbers of Vl genes and are organized differently from human Igl loci. Secondary recombination of mouse Igl genes might be difficult. Secondary IgL gene rearrangement occurs frequently during early B cell development, as indicated by the usage of downstream Jk genes or the deletion of the k locus to express Igl genes. In heterozygous mice carrying a human Ck constant region on one of the k alleles and rearranged Igk genes on the other allele, 25% of B cells expressed the newly rearranged Igk gene with the human Ck region, suggesting that at least 25% of B cells changed their Igk genes [10]. The caveat of this estimate is that it was still based on mouse models expressing rearranged mouse Igk genes on one of the alleles. The true frequency of secondary Igk recombination might be even higher. When immature B cells are stimulated through crosslinking their BCRs, 70% of them change their IgL genes [11,12]. It has also been proposed that the ability to edit light chain genes is confined to early immature B cells in the bone marrow [13]. The later immature B cells become more sensitive to apoptosis, presumably to enable clonal deletion. The concept of VH replacement Secondary rearrangement of an upstream VH to a preformed VHDJH gene is conceptually difficult, because all the intervening DH segments are deleted during the primary VH to
1471-4906/$ – see front matter ß 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.it.2007.01.003
Review
TRENDS in Immunology
Figure 1. VH replacement and different RSS sites. (a) The 12-bp RSS (black triangle), 23-bp RSS (open triangle), and cRSS (gray triangle). (b) Proposed model for serial VH replacement. During each round of VH replacement recombination, the previously rearranged VH gene is replaced by the upstream VH gene and the intervening DNA is released as an excision circle from the chromosome. Different VH germline genes are color coded in green, red, blue and purple, such that the VH replacement ‘footprint’ left over from each round of VH replacement recombination can be traced in the elongated IgH CDR3 regions (highlighted in yellow circles). Open triangles and gray triangles are specified as in (a).
DJH rearrangement, leaving no 12-bp RSS to recombine with 23-bp RSS-tagged VH or JH genes. Previous analysis of transformed murine pre-B cell lines found that functional VHDJH genes can be generated from non-functional VHDJH rearrangements [14–16]. Comparing the functional and non-functional Ig heavy chain (IgH) sequences suggested a novel VH to VHDJH recombination mechanism mediated by a cryptic RSS (cRSS) embedded near the 30 end of the rearranged VH genes, known as VH replacement [14–16] (Figure 1b). This proposed VH replacement mechanism is particularly intriguing, because the cRSS motif can be found in 40 out of 44 human functional VH genes and in almost two-thirds of mouse VH genes [17]. Therefore, VH replacement could offer a unique RAG-mediated recombination mechanism to change preformed IgH genes. VH replacement in knock-in mouse models Currently, there is no distinction between VH replacement for editing an active receptor and correcting a non-functional gene. The occurrence and biological potential of VH replacement were first tested in knock-in mice carrying a rearranged IgH V(D)J joint at the JH locus encoding antiDNA antibodies. In these mice, the artificially inserted VHDJH genes can be efficiently deleted by three types of recombination events: VH to VHDJH, DH to VHDJH, and VH to DH to VHDJH [18,19]. Only the VH to VHDJH recombination mediated by the cRSS can occur in normal mice. The other two types of recombination are the artifacts of the www.sciencedirect.com
Vol.28 No.3
133
targeted IgH locus, in which the intact DH region is located upstream of the rearranged VHDJH gene. In these mice, the abnormally early expression of the functional IgH transgenes drives pro-B cells or even early progenitor cells directly into the pre-B cell stage. Thus, the developmental stage(s) during which VH replacement occurs could not be determined clearly. Evidence for VH replacement has also been obtained in the quasimonoclonal (QM) mouse [10]. The QM mouse carries a knocked-in functional VHDJH gene segment at one of the IgH alleles, and the other IgH allele is nonfunctional owing to targeted deletion of all the JH segments. The QM mouse cannot generate functional Igk genes and must rely on Igl gene rearrangement to produce B cells. The knocked-in IgH chain, when paired with any of the endogenous Igl chains, will generate antibodies against the hapten (4-hydroxy-3-nitrophenyl) acetyl (NP). Approximately 20% of peripheral B cells and up to 50% of the peritoneal B cells in the QM mouse have lost the knocked-in VH gene and NP reactivity, presumably through VH replacement [10]. A potential contribution of VH replacement to antibody repertoire diversification has also been suggested by the capability of the QM mouse to mount a protective antiviral antibody response and to produce monoclonal antibodies against various antigens using IgH genes generated by VH replacement [20,21]. However, these suggestions were based mainly on sequence analysis. A comprehensive study of the VH replacement process in the QM mouse is required to understand this mouse model clearly. Other types of secondary IgH rearrangements had also been observed in knock-in mice carrying different rearranged IgH genes. For instance, in knock-in mice with a rearranged VH11D42 gene, secondary VH to VHDJH recombinations rarely produced functional IgH genes, and most of the secondary recombinations on the knockin allele targeted the two cryptic heptameric sites in the VH gene leader sequence region [22]. These secondary recombinations cannot produce functional IgH genes. In another knock-in mouse carrying a rearranged VHT15 gene, secondary recombinations also frequently disrupted the knock-in IgH allele. In the heterozygous mice, 60% of B cells express the endogenous IgH genes and have lost the expression of the VHT15 transgene [23]. It is unclear why secondary IgH recombination is targeted to different cryptic heptameric sites in these models rather than to the expected cRSS embedded at the end of most VH genes. Recently, two knock-in mouse models carrying non-functional IgH rearrangements on both IgH alleles were generated independently by two research groups [24,25]. In both models, VH replacement is employed to rescue pro-B cells with non-functional IgH rearrangements. An almost normal number of B cells and a diversified B cell repertoire were generated through VH replacement or other types of recombinations, such as VH to JH recombination [24] or VH to DH to VHDJH recombination [25]. In heterozygous mice, the cRSS-mediated VH replacement process is not efficient enough to compete with de novo VH to DJH recombination on the unrearranged IgH allele [24,25]. Nevertheless, VH replacement mediated by the cRSS is more efficient than direct VH to JH recombination [25]. VH replacement did
134
Review
TRENDS in Immunology Vol.28 No.3
not occur in the RAG-deficient background [25], directly confirming that VH replacement is a RAG-mediated event. These results provide the long awaited in vivo evidence for the originally observed VH replacement event in murine preB cell lines more than two decades ago, and also suggest that VH replacement can serve as a powerful repertoire diversification mechanism. VH replacement during murine B cell development The occurrence of VH replacement has been well documented in knock-in mice carrying different IgH genes within the IgH locus. VH replacement might offer a unique mechanism to rescue pro-B cells with non-functional IgH rearrangements and to edit B cells expressing self-reactive or mono-specific antibodies. Theoretically, a substantial number of B cells should carry non-functional IgH rearrangements on both IgH alleles or IgH genes encoding self-reactive BCRs. RAG-mediated double-stranded DNA breaks at the VH–cRSS border are the direct indicators for ongoing VH replacement, which can be detected by ligationmediated PCR (LM-PCR). However, it is apparently difficult to detect double-stranded DNA breaks at the cRSS border by LM-PCR in mouse bone marrow pro-B, pre-B or immature B cells [26]. From a total of 48 LM-PCR reactions, only three of them identified specific double-stranded DNA breaks at the expected cRSS sites in bone marrow immature B cells. In most of the reactions, the artificial linkers were ligated to other sites and were not indicative of authentic VH replacement [26]. The low success rate of specific amplification of double-stranded DNA breaks at the cRSS borders argues that VH replacement might be a rare event during B cell development in non-engineered mice. In addition, the analysis of >500 mouse IgH sequences identified only one IgH gene with a potential VH replacement ‘footprint’ at the V-D junction [26], suggesting that the sequence-analysis-based method to identify potential VH replacement products does not work on mouse IgH sequences. These results and conclusions are remarkably different to those obtained using knock-in mouse models, in which VH replacement efficiently rescues B cells carrying non-functional IgH rearrangements and generates a diversified antibody repertoire. In the knock-in mice, VH replacement occurs in homozygous mice carrying non-functional IgH genes on both IgH alleles. VH replacement is not efficient enough to compete with the de novo recombination of the wild-type IgH allele. Pro-B cells carrying non-functional rearrangements on both IgH alleles are blocked for future development and accumulate in the bone marrow of knock-in mice. In normal mice, these pro-B cells might be eliminated instead of being given additional chances for secondary recombination. Additional studies are required to address the disparity between the efficient VH replacement in the knock-in mouse models and the difficulty in finding VH replacement intermediates in normal mice. Ongoing VH replacement in human EU12 cells Owing to the lack of suitable experimental model systems and analytic approaches, the molecular basis and natural occurrence of VH replacement in human B lineage cells were not fully studied until recently. The human EU12 cell www.sciencedirect.com
line was established from a childhood acute lymphoblastic leukemia patient. EU12 cells in vitro undergo differentiation from pro-B cells to pre-B cells and immature B cells under normal tissue culture conditions [27]. Analysis of the IgH sequences derived from the EU12 cells indicated that almost half of them were generated through serial VH replacement recombinations mediated by the cRSS site at the end of VH gene [28] (Figure 1b). This prediction was confirmed by the detection of double-stranded DNA breaks at the VH2 cRSS border by LM-PCR and the amplification of VH replacement excision circles that were generated during each step of VH replacement recombination [28]. Moreover, purified RAG1 and RAG2 proteins were shown to bind to the cRSS probes derived from representative VH germline genes and to cleave the cRSS substrates in a fashion similar to that of the conventional 12-bp or 23-bp RSS, despite a relatively lower efficiency [28]. Ongoing VH replacement in human bone marrow immature B cells Using the LM-PCR approach, double-stranded DNA breaks at the VH3 cRSS border have been detected in human bone marrow immature B cells but not in pro-B, pre-B, naive B or memory B cells [28]. This result provided the first definitive evidence that VH replacement occurs during human B cell development [28]. Theoretically, VH replacement should occur in pro-B cells with nonproductive VHDJH rearrangements or in pre-B cells expressing IgH chains that do not pair with the surrogate or conventional light chains. However, LM-PCR assays failed to detect double-stranded DNA breaks at the cRSS border in pro-B or pre-B cells. Despite the different success rates of LM-PCR amplification of VH-cRSS double-stranded DNA breaks in human and mouse bone marrow immature B cells, the results indicated the ongoing VH replacement in the bone marrow immature B cells. It should be emphasized that LM-PCR detects double-stranded DNA breaks that occur at the time of the study. Thus, the number of immature B cells and their differentiation stage are crucial for the successful detection of double-stranded DNA breaks at the cRSS borders. In studies using human bone marrow samples, immature B cells were identified as CD24hiIgM+IgD– cells, which only account for 10–15% of CD19+IgM+ B cells in the bone marrow, based on fluorescence-activated cell sorting (FACS) analysis. The number of purified immature B cells used per LM-PCR reaction is 300. However, the LM-PCR products specific for the VH3 cRSS borders might be amplified from only a few cells of the 300 immature B cells. Reducing the number of cells or changing the FACS sorting protocol might have undesired consequences for the success rate of LM-PCR amplification. Searching for potential VH replacement products in human IgH sequences VH replacement renews almost the entire VH coding region to delete the unwanted IgH genes. However, owing to the location of the cRSS, a short stretch of nucleotides following the cRSS might remain at the V-D junction after VH replacement recombination, which can be used as a VH replacement ‘footprint’ to trace the occurrence of VH replacement.
Review
TRENDS in Immunology
Such VH replacement ‘footprints’, containing at least 5 nucleotides at their V-D junctions that match with those of another VH gene, can be found in 5% of the human IgH genes derived from peripheral blood B cells [28], suggesting that VH replacement products contribute significantly to the diversification of the human B cell repertoire. However, a recent survey of human 6329 IgH sequences using the VH3-23 gene reached a totally opposite conclusion: there is no evidence of VH replacement in the human IgH repertoire [29]. In fact, potential VH replacement ‘footprints’ within the V-D junctions were found in both studies and the different conclusions are mainly owing to the negative controls used to determine the VH replacement ‘footprints’ contributed by random events. In the first study, 23 out of 414 IgH sequences contain pentameric VH replacement ‘footprints’ at the V-D junctions. This frequency is significantly higher than that at the D-J junctions (6 out of 414 sequences) or that using a randomly generated database of sequences 6 nucleotides in length (6 out of 200 sequences). Thus, it was concluded that the occurrence of pentameric VH replacement ‘footprints’ at the V-D junctions were contributed by VH replacement rather than by random events. In the second study, 692 pentameric VH replacement ‘footprints’ were identified from 3126 V-D junctions (or 22%) [29]; this number was not significantly higher than that using a permutated sequence database derived from the V-D junction sequences (21%). In another test, the frequency of potential ‘footprints’ contributed by the VH genes upstream of the VH3-23 gene is not significantly higher than that contributed by the VH genes downstream of the VH3-23 gene. Based on the reasoning that the VH3-23 gene can replace only downstream VH genes, it was concluded that none of the ‘footprints’ are indicative of VH replacement. Currently, it is under discussion whether the permutated sequence database represents truly random sequences, because the frequency of VH replacement ‘footprints’ in the permutated sequence database is significantly higher than that in the randomly generated sequence database. For the second test, although the assumption that only an upstream VH gene can replace a downstream VH gene is correct, the ‘footprints’ contributed by VH genes upstream of the VH3-23 gene are similar to those contributed by VH genes downstream of the VH3-23 gene, which makes this test practically infeasible (as agreed by T. Barington, personal communication). Clearly, the protocol for assigning potential VH replacement products based on the identification of potential VH replacement ‘footprints’ at the V-D junction needs to be refined further. Currently, this is the only way of estimating the potential contribution of VH replacement products to the antibody repertoire. Currently, a joint effort is in progress to test additional sequence sets to refine the protocol. It should be pointed out that although potential VH replacement products might be assigned based on the identification of VH replacement ‘footprints’ within the V-D junctions, the actual frequency of VH replacement products in the IgH repertoire must be higher. In the knock-in mouse carrying non-functional IgH rearrangements, almost all the functional IgH genes were generated through VH replacement; however, only a fraction (20%) of them contained www.sciencedirect.com
Vol.28 No.3
135
identifiable VH replacement ‘footprints’ [24]. There are several possible explanations for the loss of the VH replacement ‘footprints’. First, the VH replacement ‘footprints’ can be nibbled away from either end during recombination by exonuclease activities. Second, it has been suggested that VH replacement often proceeds through microhomology-mediated recombination, which might delete the VH replacement ‘footprints’. Third, VH replacement ‘footprints’ preferentially encode charged amino acids in the CDR3 regions, and VH replacement products containing charged CDR3s might be removed by negative selection during B cell development. Thus, it is more likely that the sequenceanalysis-based assignment of VH replacement products will underestimate the true contribution of VH replacement products to the antibody repertoire. Unsolved questions concerning VH gene replacement Studies in mouse pre-B cell lines, knock-in mice carrying functional or non-functional IgH genes, mouse bone marrow immature B cells, the human EU12 cell line and human bone marrow immature B cells all demonstrated the occurrence of VH replacement. VH replacement can rescue pro-B cells with non-functional IgH rearrangements on both alleles and can also delete IgH genes encoding self-reactive or monoclonal BCRs (Figure 2). The stage and frequency of VH replacement during normal B cell development, and the potential contribution of VH replacement products to the diversification of antibody repertoires in humans and mice, are the remaining questions. One of the intrinsic and provocative features of VH replacement is that the VH replacement ‘footprints’ frequently encode charged amino acids in both humans and mice [17,24,28,30]. This feature is conserved in both humans and mice – the reason is unknown. In the knock-in mouse model of VH replacement, charged residues contributed by VH replacement ‘footprints’ are negatively selected during B cell maturation [24]. Charged amino acids within the IgH CDR3 region are not well tolerated in the normal antibody repertoire but are frequently found in self-reactive or antiviral antibodies [31,32]. It has yet to be determined if the charged amino acids contributed by the VH replacement ‘footprints’ have any biological significance. Despite the different success rates for LM-PCR amplifications in humans and mice, double-stranded DNA breaks at the cRSS border were detected in the bone marrow immature B cells in both humans and mice, suggesting the natural occurrence of VH replacement during normal B cell development. The occurrence of VH replacement in immature B cells is consistent with the abilities of these cells to express RAG genes and to alter their IgL genes. Thus, the immature B cell stage seems to be the dedicated stage for editing both IgH and IgL genes. The ongoing VH replacement in bone marrow immature B cells also raises additional important questions of whether VH replacement is a regulated process and how it is regulated. Immature B cells are generated throughout life from the bone marrow, and they migrate into the circulation as they mature into naive B cells [33]. Immature B cells
136
Review
TRENDS in Immunology Vol.28 No.3
Figure 2. VH replacement during early B lineage cell development. The potential stages at which VH replacement occurs to rescue pro-B cells containing non-functional IgH genes, and to edit preformed IgH genes in immature B cells or in the newly emigrated immature B cells in the periphery are indicated. Terminal deoxynucleotidyltransferase (TdT) indicates TdT expression.
can be detected at low frequency in the peripheral blood as the newly emigrated immature B cells [34]. During inflammatory responses, elevated numbers of these cells are released from the bone marrow [35]. The newly emigrated immature B cells have similar features to their bone marrow counterparts, as they can express RAG genes and edit their IgL genes [36]. It has yet to be determined if VH replacement also occurs in the newly emigrated immature B cells in the periphery to diversify the antibody repertoire further. Collectively, VH replacement offers a unique RAG-mediated recombinatorial mechanism to change preformed IgH genes, which can rescue pro-B cells with nonfunctional IgH rearrangements and delete IgH genes encoding self-reactive BCRs. Understanding the potential contribution of VH replacement products to the antibody repertoire and the regulatory mechanism for VH replacement should be the focuses of future studies. Acknowledgements I am supported in part by a K01 grant AR048592 from NIH/NIAMS. I thank Max D. Cooper and Peter D. Burrows for critically reviewing this manuscript. I also thank Michael Zemlin and Torben Barington for many helpful discussions regarding the identification of potential V H replacement products.
References 1 Rajewsky, K. (1996) Clonal selection and learning in the antibody system. Nature 381, 751–758 www.sciencedirect.com
2 Tonegawa, S. (1983) Somatic generation of antibody diversity. Nature 302, 575–581 3 Jung, D. and Alt, F.W. (2004) Unraveling V(D)J recombination: insights into gene regulation. Cell 116, 299–311 4 Jung, D. et al. (2006) Mechanism and control of V(D)J recombination at the immunoglobulin heavy chain locus. Annu. Rev. Immunol. 24, 541– 570 5 Schatz, D.G. and Baltimore, D. (1988) Stable expression of immunoglobulin gene V(D)J recombinase activity by gene transfer into 3T3 fibroblasts. Cell 53, 107–115 6 Schatz, D.G. et al. (1989) The V(D)J recombination activating gene. RAG-1. Cell 59, 1035–1048 7 Lewis, S.M. (1994) The mechanism of V(D)J joining: lessons from molecular, immunological, and comparative analyses. Adv. Immunol. 56, 27–150 8 Difilippantonio, M.J. et al. (1996) RAG1 mediates signal sequence recognition and recruitment of RAG2 in V(D)J recombination. Cell 87, 253–262 9 Ramsden, D.A. et al. (1996) Distinct DNA sequence and structure requirements for the two steps of V(D)J recombination signal cleavage. EMBO J. 15, 3197–3206 10 Casellas, R. et al. (2001) Contribution of receptor editing to the antibody repertoire. Science 291, 1541–1544 11 Melamed, D. and Nemazee, D. (1997) Self-antigen does not accelerate immature B cell apoptosis, but stimulates receptor editing as a consequence of developmental arrest. Proc. Natl. Acad. Sci. U. S. A. 94, 9267–9272 12 Melamed, D. et al. (1998) Developmental regulation of B lymphocyte immune tolerance compartmentalizes clonal selection from receptor selection. Cell 92, 173–182 13 Sandel, P.C. and Monroe, J.G. (1999) Negative selection of immature B cells by receptor editing or deletion is determined by site of antigen encounter. Immunity 10, 289–299
Review
TRENDS in Immunology
14 Kleinfield, R. et al. (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 15 Reth, M. et al. (1986) A novel VH to VHDJH joining mechanism in heavychain-negative (null) pre-B cells results in heavy-chain production. Nature 322, 840–842 16 Covey, L.R. et al. (1990) VH to VHDJH rearrangement is mediated by the internal VH heptamer. Int. Immunol. 2, 579–583 17 Zhang, Z. et al. (2004) The molecular basis and biological significance of VH replacement. Immunol. Rev. 197, 231–242 18 Chen, C. et al. (1995) Immunoglobulin heavy chain gene replacement: a mechanism of receptor editing. Immunity 3, 747–755 19 Chen, C. et al. (1995) The site and stage of anti-DNA B-cell deletion. Nature 373, 252–255 20 Lopez-Macias, C. et al. (1999) Secondary rearrangements and hypermutation generate sufficient B cell diversity to mount protective antiviral immunoglobulin responses. J. Exp. Med. 189, 1791–1798 21 Madan, M.K. et al. (2000) Specific antibody production by VH-gene replacement. Eur. J. Immunol. 30, 2404–2411 22 Pewzner-Jung, Y. et al. (1998) B cell deletion, anergy, and receptor editing in ‘knock in’ mice targeted with a germline-encoded or somatically mutated anti-DNA heavy chain. J. Immunol. 161, 4634–4645 23 Taki, S. et al. (1993) Targeted insertion of a variable region gene into the immunoglobulin heavy chain locus. Science 262, 1268–1271 24 Koralov, S.B. et al. (2006) Antibody repertoires generated by VH replacement and direct VH to JH joining. Immunity 25, 43–53 25 Lutz, J. et al. (2006) VH replacement rescues progenitor B cells with two nonproductive VDJ alleles. J. Immunol. 177, 7007–7014
Vol.28 No.3
26 Watson, L.C. et al. (2006) Paucity of V-D-D-J rearrangements and VH replacement events in lupus prone and nonautoimmune TdT/ and TdT+/+ mice. J. Immunol. 177, 1120–1128 27 Wang, Y.H. et al. (2003) V(D)J recombinatorial repertoire diversification during intraclonal pro-B to B-cell differentiation. Blood 101, 1030–1037 28 Zhang, Z. et al. (2003) Contribution of VH gene replacement to the primary B cell repertoire. Immunity 19, 21–31 29 Ohm-Laursen, L. et al. (2006) No evidence for the use of DIR, D-D fusions, chromosome 15 open reading frames or VH replacement in the peripheral repertoire was found on application of an improved algorithm, JointML, to 6329 human immunoglobulin H rearrangements. Immunology 119, 265–277 30 Liu, Y. et al. (2005) Potential contribution of VH gene replacement in immunity and disease. Ann. N. Y. Acad. Sci. 1062, 175–181 31 Radic, M.Z. et al. (1993) Residues that mediate DNA binding of autoimmune antibodies. J. Immunol. 150, 4966–4977 32 Radic, M.Z. and Zouali, M. (1996) Receptor editing, immune diversification and self-tolerance. Immunity 5, 505–511 33 Nunez, C. et al. (1996) B cells are generated throughout life in humans. J. Immunol. 156, 866–872 34 Wardemann, H. et al. (2003) Predominant autoantibody production by early human B cell precursors. Science 301, 1374–1377 35 Ueda, Y. et al. (2004) Inflammation controls B lymphopoiesis by regulating chemokine CXCL12 expression. J. Exp. Med. 199, 47– 58 36 Sims, G.P. et al. (2005) Identification and characterization of circulating human transitional B cells. Blood 105, 4390–4398
Elsevier.com – linking scientists to new research and thinking Designed for scientists’ information needs, Elsevier.com is powered by the latest technology with customer-focused navigation and an intuitive architecture for an improved user experience and greater productivity. The easy-to-use navigational tools and structure connect scientists with vital information – all from one entry point. Users can perform rapid and precise searches with our advanced search functionality, using the FAST technology of Scirus.com, the free science search engine. Users can define their searches by any number of criteria to pinpoint information and resources. Search by a specific author or editor, book publication date, subject area – life sciences, health sciences, physical sciences and social sciences – or by product type. Elsevier’s portfolio includes more than 1800 Elsevier journals, 2200 new books every year and a range of innovative electronic products. In addition, tailored content for authors, editors and librarians provides timely news and updates on new products and services. Elsevier is proud to be a partner with the scientific and medical community. Find out more about our mission and values at Elsevier.com. Discover how we support the scientific, technical and medical communities worldwide through partnerships with libraries and other publishers, and grant awards from The Elsevier Foundation. As a world-leading publisher of scientific, technical and health information, Elsevier is dedicated to linking researchers and professionals to the best thinking in their fields. We offer the widest and deepest coverage in a range of media types to enhance cross-pollination of information, breakthroughs in research and discovery, and the sharing and preservation of knowledge.
Elsevier. Building insights. Breaking boundaries. www.elsevier.com www.sciencedirect.com
137
TRENDS in Immunology
Vol.28 No.3
VH replacement in mice and humans Zhixin Zhang Division of Developmental and Clinical Immunology, Departments of Medicine and Microbiology, University of Alabama at Birmingham, Birmingham, AL 35294, USA
Heavy chain variable segment (VH) replacement refers to recombination activating gene (RAG) product-mediated secondary recombination between a previously rearranged VH gene and an upstream unrearranged VH gene. VH replacement was first observed in mouse pre-B cell lines and later demonstrated in knock-in mouse models carrying immunoglobulin heavy chain (IgH) genes encoding self-reactive or mono-specific antibodies or non-functional IgH rearrangements on both IgH alleles. Despite these findings, it is still difficult to find VH replacement intermediates during normal murine B cell development. In humans, ongoing VH replacement was found in a clonal B lineage EU12 cell line and in human bone marrow immature B cells. The identification of potential VH replacement products also suggested a potential contribution of VH replacement to the antibody repertoire. Here, I review the evidence for whether VH replacement genuinely offers an in vivo RAG-mediated recombinatorial mechanism to alter preformed IgH genes in mice and humans. Generating a diversified antibody repertoire through V(D)J recombination To combat infectious agents, a diversified antibody repertoire is generated through somatic recombination of previously separated variable (V), diversity (D) [for the heavy (H) chain only] and joining (J) gene segments to form the variable domain exons of immunoglobulin (Ig) genes in developing B lineage cells [1–4]. V(D)J recombination is catalyzed by a pair of recombination activating gene products, RAG1 and RAG2 [5,6]. Specific recombination of the V, D and J gene segments is directed by the recombination signal sequences (RSS), which contain a highly conserved heptamer (50 -CACTGTG-30 ) and a nonamer (50 -ACAAAAACC-30 ) separated by a non-conserved spacer region either 12 bp or 23 bp in length [7] (Figure 1a). The heptamer and nonamer motifs are required for RAG1 and RAG2 binding and subsequent DNA cleavage [8,9]. The RSS spacer regions are important for directing the specific formation of the coding joints. As dictated by the 12/23 rule, coding joints preferentially form between a pair of 12-bp and 23-bp RSS-tagged gene segments for Ig and T cell receptor (TCR) genes [7]. The random nature of V(D)J recombination is essential for generating a diversified antibody repertoire; however, it can also produce unwanted Ig genes. First, owing to the imprecise joining of the coding ends, two-thirds of the VHDJH and VL-JL joints will be out of reading frame and Corresponding author: Zhang, Z. ([email protected]). Available online 26 January 2007. www.sciencedirect.com
unable to produce Ig peptides. Second, functionally rearranged VHDJH genes might produce m heavy chains that fail to pair with surrogate or conventional light chains to form the pre-B cell receptor (pre-BCR) or BCR, respectively. Third, the random nature of V(D)J recombination can also generate Ig genes encoding self-reactive BCRs. Such unwanted rearrangements must be changed to enable continuous B cell development and maturation. Editing the Ig light chain genes Receptor editing refers to RAG-mediated secondary recombination of Ig genes to alter the previously rearranged Ig genes. Most previous studies of receptor editing focused mainly on the Ig light chain (IgL) genes, because Vk and Jk gene segments are tagged by complimentary RSS sites and, thus, the organization of the mouse and human Igk loci enables continuous secondary rearrangements by simply joining an upstream Vk with a downstream Jk gene segment to delete the preformed VkJk joints. As an additional choice, B cells can delete the Igk locus and initiate de novo rearrangement of the Igl locus. For human Igl loci, there are three functional Jl and Cl clusters for additional recombinations with upstream Vl genes. The mouse Igl loci have limited numbers of Vl genes and are organized differently from human Igl loci. Secondary recombination of mouse Igl genes might be difficult. Secondary IgL gene rearrangement occurs frequently during early B cell development, as indicated by the usage of downstream Jk genes or the deletion of the k locus to express Igl genes. In heterozygous mice carrying a human Ck constant region on one of the k alleles and rearranged Igk genes on the other allele, 25% of B cells expressed the newly rearranged Igk gene with the human Ck region, suggesting that at least 25% of B cells changed their Igk genes [10]. The caveat of this estimate is that it was still based on mouse models expressing rearranged mouse Igk genes on one of the alleles. The true frequency of secondary Igk recombination might be even higher. When immature B cells are stimulated through crosslinking their BCRs, 70% of them change their IgL genes [11,12]. It has also been proposed that the ability to edit light chain genes is confined to early immature B cells in the bone marrow [13]. The later immature B cells become more sensitive to apoptosis, presumably to enable clonal deletion. The concept of VH replacement Secondary rearrangement of an upstream VH to a preformed VHDJH gene is conceptually difficult, because all the intervening DH segments are deleted during the primary VH to
1471-4906/$ – see front matter ß 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.it.2007.01.003
Review
TRENDS in Immunology
Figure 1. VH replacement and different RSS sites. (a) The 12-bp RSS (black triangle), 23-bp RSS (open triangle), and cRSS (gray triangle). (b) Proposed model for serial VH replacement. During each round of VH replacement recombination, the previously rearranged VH gene is replaced by the upstream VH gene and the intervening DNA is released as an excision circle from the chromosome. Different VH germline genes are color coded in green, red, blue and purple, such that the VH replacement ‘footprint’ left over from each round of VH replacement recombination can be traced in the elongated IgH CDR3 regions (highlighted in yellow circles). Open triangles and gray triangles are specified as in (a).
DJH rearrangement, leaving no 12-bp RSS to recombine with 23-bp RSS-tagged VH or JH genes. Previous analysis of transformed murine pre-B cell lines found that functional VHDJH genes can be generated from non-functional VHDJH rearrangements [14–16]. Comparing the functional and non-functional Ig heavy chain (IgH) sequences suggested a novel VH to VHDJH recombination mechanism mediated by a cryptic RSS (cRSS) embedded near the 30 end of the rearranged VH genes, known as VH replacement [14–16] (Figure 1b). This proposed VH replacement mechanism is particularly intriguing, because the cRSS motif can be found in 40 out of 44 human functional VH genes and in almost two-thirds of mouse VH genes [17]. Therefore, VH replacement could offer a unique RAG-mediated recombination mechanism to change preformed IgH genes. VH replacement in knock-in mouse models Currently, there is no distinction between VH replacement for editing an active receptor and correcting a non-functional gene. The occurrence and biological potential of VH replacement were first tested in knock-in mice carrying a rearranged IgH V(D)J joint at the JH locus encoding antiDNA antibodies. In these mice, the artificially inserted VHDJH genes can be efficiently deleted by three types of recombination events: VH to VHDJH, DH to VHDJH, and VH to DH to VHDJH [18,19]. Only the VH to VHDJH recombination mediated by the cRSS can occur in normal mice. The other two types of recombination are the artifacts of the www.sciencedirect.com
Vol.28 No.3
133
targeted IgH locus, in which the intact DH region is located upstream of the rearranged VHDJH gene. In these mice, the abnormally early expression of the functional IgH transgenes drives pro-B cells or even early progenitor cells directly into the pre-B cell stage. Thus, the developmental stage(s) during which VH replacement occurs could not be determined clearly. Evidence for VH replacement has also been obtained in the quasimonoclonal (QM) mouse [10]. The QM mouse carries a knocked-in functional VHDJH gene segment at one of the IgH alleles, and the other IgH allele is nonfunctional owing to targeted deletion of all the JH segments. The QM mouse cannot generate functional Igk genes and must rely on Igl gene rearrangement to produce B cells. The knocked-in IgH chain, when paired with any of the endogenous Igl chains, will generate antibodies against the hapten (4-hydroxy-3-nitrophenyl) acetyl (NP). Approximately 20% of peripheral B cells and up to 50% of the peritoneal B cells in the QM mouse have lost the knocked-in VH gene and NP reactivity, presumably through VH replacement [10]. A potential contribution of VH replacement to antibody repertoire diversification has also been suggested by the capability of the QM mouse to mount a protective antiviral antibody response and to produce monoclonal antibodies against various antigens using IgH genes generated by VH replacement [20,21]. However, these suggestions were based mainly on sequence analysis. A comprehensive study of the VH replacement process in the QM mouse is required to understand this mouse model clearly. Other types of secondary IgH rearrangements had also been observed in knock-in mice carrying different rearranged IgH genes. For instance, in knock-in mice with a rearranged VH11D42 gene, secondary VH to VHDJH recombinations rarely produced functional IgH genes, and most of the secondary recombinations on the knockin allele targeted the two cryptic heptameric sites in the VH gene leader sequence region [22]. These secondary recombinations cannot produce functional IgH genes. In another knock-in mouse carrying a rearranged VHT15 gene, secondary recombinations also frequently disrupted the knock-in IgH allele. In the heterozygous mice, 60% of B cells express the endogenous IgH genes and have lost the expression of the VHT15 transgene [23]. It is unclear why secondary IgH recombination is targeted to different cryptic heptameric sites in these models rather than to the expected cRSS embedded at the end of most VH genes. Recently, two knock-in mouse models carrying non-functional IgH rearrangements on both IgH alleles were generated independently by two research groups [24,25]. In both models, VH replacement is employed to rescue pro-B cells with non-functional IgH rearrangements. An almost normal number of B cells and a diversified B cell repertoire were generated through VH replacement or other types of recombinations, such as VH to JH recombination [24] or VH to DH to VHDJH recombination [25]. In heterozygous mice, the cRSS-mediated VH replacement process is not efficient enough to compete with de novo VH to DJH recombination on the unrearranged IgH allele [24,25]. Nevertheless, VH replacement mediated by the cRSS is more efficient than direct VH to JH recombination [25]. VH replacement did
134
Review
TRENDS in Immunology Vol.28 No.3
not occur in the RAG-deficient background [25], directly confirming that VH replacement is a RAG-mediated event. These results provide the long awaited in vivo evidence for the originally observed VH replacement event in murine preB cell lines more than two decades ago, and also suggest that VH replacement can serve as a powerful repertoire diversification mechanism. VH replacement during murine B cell development The occurrence of VH replacement has been well documented in knock-in mice carrying different IgH genes within the IgH locus. VH replacement might offer a unique mechanism to rescue pro-B cells with non-functional IgH rearrangements and to edit B cells expressing self-reactive or mono-specific antibodies. Theoretically, a substantial number of B cells should carry non-functional IgH rearrangements on both IgH alleles or IgH genes encoding self-reactive BCRs. RAG-mediated double-stranded DNA breaks at the VH–cRSS border are the direct indicators for ongoing VH replacement, which can be detected by ligationmediated PCR (LM-PCR). However, it is apparently difficult to detect double-stranded DNA breaks at the cRSS border by LM-PCR in mouse bone marrow pro-B, pre-B or immature B cells [26]. From a total of 48 LM-PCR reactions, only three of them identified specific double-stranded DNA breaks at the expected cRSS sites in bone marrow immature B cells. In most of the reactions, the artificial linkers were ligated to other sites and were not indicative of authentic VH replacement [26]. The low success rate of specific amplification of double-stranded DNA breaks at the cRSS borders argues that VH replacement might be a rare event during B cell development in non-engineered mice. In addition, the analysis of >500 mouse IgH sequences identified only one IgH gene with a potential VH replacement ‘footprint’ at the V-D junction [26], suggesting that the sequence-analysis-based method to identify potential VH replacement products does not work on mouse IgH sequences. These results and conclusions are remarkably different to those obtained using knock-in mouse models, in which VH replacement efficiently rescues B cells carrying non-functional IgH rearrangements and generates a diversified antibody repertoire. In the knock-in mice, VH replacement occurs in homozygous mice carrying non-functional IgH genes on both IgH alleles. VH replacement is not efficient enough to compete with the de novo recombination of the wild-type IgH allele. Pro-B cells carrying non-functional rearrangements on both IgH alleles are blocked for future development and accumulate in the bone marrow of knock-in mice. In normal mice, these pro-B cells might be eliminated instead of being given additional chances for secondary recombination. Additional studies are required to address the disparity between the efficient VH replacement in the knock-in mouse models and the difficulty in finding VH replacement intermediates in normal mice. Ongoing VH replacement in human EU12 cells Owing to the lack of suitable experimental model systems and analytic approaches, the molecular basis and natural occurrence of VH replacement in human B lineage cells were not fully studied until recently. The human EU12 cell www.sciencedirect.com
line was established from a childhood acute lymphoblastic leukemia patient. EU12 cells in vitro undergo differentiation from pro-B cells to pre-B cells and immature B cells under normal tissue culture conditions [27]. Analysis of the IgH sequences derived from the EU12 cells indicated that almost half of them were generated through serial VH replacement recombinations mediated by the cRSS site at the end of VH gene [28] (Figure 1b). This prediction was confirmed by the detection of double-stranded DNA breaks at the VH2 cRSS border by LM-PCR and the amplification of VH replacement excision circles that were generated during each step of VH replacement recombination [28]. Moreover, purified RAG1 and RAG2 proteins were shown to bind to the cRSS probes derived from representative VH germline genes and to cleave the cRSS substrates in a fashion similar to that of the conventional 12-bp or 23-bp RSS, despite a relatively lower efficiency [28]. Ongoing VH replacement in human bone marrow immature B cells Using the LM-PCR approach, double-stranded DNA breaks at the VH3 cRSS border have been detected in human bone marrow immature B cells but not in pro-B, pre-B, naive B or memory B cells [28]. This result provided the first definitive evidence that VH replacement occurs during human B cell development [28]. Theoretically, VH replacement should occur in pro-B cells with nonproductive VHDJH rearrangements or in pre-B cells expressing IgH chains that do not pair with the surrogate or conventional light chains. However, LM-PCR assays failed to detect double-stranded DNA breaks at the cRSS border in pro-B or pre-B cells. Despite the different success rates of LM-PCR amplification of VH-cRSS double-stranded DNA breaks in human and mouse bone marrow immature B cells, the results indicated the ongoing VH replacement in the bone marrow immature B cells. It should be emphasized that LM-PCR detects double-stranded DNA breaks that occur at the time of the study. Thus, the number of immature B cells and their differentiation stage are crucial for the successful detection of double-stranded DNA breaks at the cRSS borders. In studies using human bone marrow samples, immature B cells were identified as CD24hiIgM+IgD– cells, which only account for 10–15% of CD19+IgM+ B cells in the bone marrow, based on fluorescence-activated cell sorting (FACS) analysis. The number of purified immature B cells used per LM-PCR reaction is 300. However, the LM-PCR products specific for the VH3 cRSS borders might be amplified from only a few cells of the 300 immature B cells. Reducing the number of cells or changing the FACS sorting protocol might have undesired consequences for the success rate of LM-PCR amplification. Searching for potential VH replacement products in human IgH sequences VH replacement renews almost the entire VH coding region to delete the unwanted IgH genes. However, owing to the location of the cRSS, a short stretch of nucleotides following the cRSS might remain at the V-D junction after VH replacement recombination, which can be used as a VH replacement ‘footprint’ to trace the occurrence of VH replacement.
Review
TRENDS in Immunology
Such VH replacement ‘footprints’, containing at least 5 nucleotides at their V-D junctions that match with those of another VH gene, can be found in 5% of the human IgH genes derived from peripheral blood B cells [28], suggesting that VH replacement products contribute significantly to the diversification of the human B cell repertoire. However, a recent survey of human 6329 IgH sequences using the VH3-23 gene reached a totally opposite conclusion: there is no evidence of VH replacement in the human IgH repertoire [29]. In fact, potential VH replacement ‘footprints’ within the V-D junctions were found in both studies and the different conclusions are mainly owing to the negative controls used to determine the VH replacement ‘footprints’ contributed by random events. In the first study, 23 out of 414 IgH sequences contain pentameric VH replacement ‘footprints’ at the V-D junctions. This frequency is significantly higher than that at the D-J junctions (6 out of 414 sequences) or that using a randomly generated database of sequences 6 nucleotides in length (6 out of 200 sequences). Thus, it was concluded that the occurrence of pentameric VH replacement ‘footprints’ at the V-D junctions were contributed by VH replacement rather than by random events. In the second study, 692 pentameric VH replacement ‘footprints’ were identified from 3126 V-D junctions (or 22%) [29]; this number was not significantly higher than that using a permutated sequence database derived from the V-D junction sequences (21%). In another test, the frequency of potential ‘footprints’ contributed by the VH genes upstream of the VH3-23 gene is not significantly higher than that contributed by the VH genes downstream of the VH3-23 gene. Based on the reasoning that the VH3-23 gene can replace only downstream VH genes, it was concluded that none of the ‘footprints’ are indicative of VH replacement. Currently, it is under discussion whether the permutated sequence database represents truly random sequences, because the frequency of VH replacement ‘footprints’ in the permutated sequence database is significantly higher than that in the randomly generated sequence database. For the second test, although the assumption that only an upstream VH gene can replace a downstream VH gene is correct, the ‘footprints’ contributed by VH genes upstream of the VH3-23 gene are similar to those contributed by VH genes downstream of the VH3-23 gene, which makes this test practically infeasible (as agreed by T. Barington, personal communication). Clearly, the protocol for assigning potential VH replacement products based on the identification of potential VH replacement ‘footprints’ at the V-D junction needs to be refined further. Currently, this is the only way of estimating the potential contribution of VH replacement products to the antibody repertoire. Currently, a joint effort is in progress to test additional sequence sets to refine the protocol. It should be pointed out that although potential VH replacement products might be assigned based on the identification of VH replacement ‘footprints’ within the V-D junctions, the actual frequency of VH replacement products in the IgH repertoire must be higher. In the knock-in mouse carrying non-functional IgH rearrangements, almost all the functional IgH genes were generated through VH replacement; however, only a fraction (20%) of them contained www.sciencedirect.com
Vol.28 No.3
135
identifiable VH replacement ‘footprints’ [24]. There are several possible explanations for the loss of the VH replacement ‘footprints’. First, the VH replacement ‘footprints’ can be nibbled away from either end during recombination by exonuclease activities. Second, it has been suggested that VH replacement often proceeds through microhomology-mediated recombination, which might delete the VH replacement ‘footprints’. Third, VH replacement ‘footprints’ preferentially encode charged amino acids in the CDR3 regions, and VH replacement products containing charged CDR3s might be removed by negative selection during B cell development. Thus, it is more likely that the sequenceanalysis-based assignment of VH replacement products will underestimate the true contribution of VH replacement products to the antibody repertoire. Unsolved questions concerning VH gene replacement Studies in mouse pre-B cell lines, knock-in mice carrying functional or non-functional IgH genes, mouse bone marrow immature B cells, the human EU12 cell line and human bone marrow immature B cells all demonstrated the occurrence of VH replacement. VH replacement can rescue pro-B cells with non-functional IgH rearrangements on both alleles and can also delete IgH genes encoding self-reactive or monoclonal BCRs (Figure 2). The stage and frequency of VH replacement during normal B cell development, and the potential contribution of VH replacement products to the diversification of antibody repertoires in humans and mice, are the remaining questions. One of the intrinsic and provocative features of VH replacement is that the VH replacement ‘footprints’ frequently encode charged amino acids in both humans and mice [17,24,28,30]. This feature is conserved in both humans and mice – the reason is unknown. In the knock-in mouse model of VH replacement, charged residues contributed by VH replacement ‘footprints’ are negatively selected during B cell maturation [24]. Charged amino acids within the IgH CDR3 region are not well tolerated in the normal antibody repertoire but are frequently found in self-reactive or antiviral antibodies [31,32]. It has yet to be determined if the charged amino acids contributed by the VH replacement ‘footprints’ have any biological significance. Despite the different success rates for LM-PCR amplifications in humans and mice, double-stranded DNA breaks at the cRSS border were detected in the bone marrow immature B cells in both humans and mice, suggesting the natural occurrence of VH replacement during normal B cell development. The occurrence of VH replacement in immature B cells is consistent with the abilities of these cells to express RAG genes and to alter their IgL genes. Thus, the immature B cell stage seems to be the dedicated stage for editing both IgH and IgL genes. The ongoing VH replacement in bone marrow immature B cells also raises additional important questions of whether VH replacement is a regulated process and how it is regulated. Immature B cells are generated throughout life from the bone marrow, and they migrate into the circulation as they mature into naive B cells [33]. Immature B cells
136
Review
TRENDS in Immunology Vol.28 No.3
Figure 2. VH replacement during early B lineage cell development. The potential stages at which VH replacement occurs to rescue pro-B cells containing non-functional IgH genes, and to edit preformed IgH genes in immature B cells or in the newly emigrated immature B cells in the periphery are indicated. Terminal deoxynucleotidyltransferase (TdT) indicates TdT expression.
can be detected at low frequency in the peripheral blood as the newly emigrated immature B cells [34]. During inflammatory responses, elevated numbers of these cells are released from the bone marrow [35]. The newly emigrated immature B cells have similar features to their bone marrow counterparts, as they can express RAG genes and edit their IgL genes [36]. It has yet to be determined if VH replacement also occurs in the newly emigrated immature B cells in the periphery to diversify the antibody repertoire further. Collectively, VH replacement offers a unique RAG-mediated recombinatorial mechanism to change preformed IgH genes, which can rescue pro-B cells with nonfunctional IgH rearrangements and delete IgH genes encoding self-reactive BCRs. Understanding the potential contribution of VH replacement products to the antibody repertoire and the regulatory mechanism for VH replacement should be the focuses of future studies. Acknowledgements I am supported in part by a K01 grant AR048592 from NIH/NIAMS. I thank Max D. Cooper and Peter D. Burrows for critically reviewing this manuscript. I also thank Michael Zemlin and Torben Barington for many helpful discussions regarding the identification of potential V H replacement products.
References 1 Rajewsky, K. (1996) Clonal selection and learning in the antibody system. Nature 381, 751–758 www.sciencedirect.com
2 Tonegawa, S. (1983) Somatic generation of antibody diversity. Nature 302, 575–581 3 Jung, D. and Alt, F.W. (2004) Unraveling V(D)J recombination: insights into gene regulation. Cell 116, 299–311 4 Jung, D. et al. (2006) Mechanism and control of V(D)J recombination at the immunoglobulin heavy chain locus. Annu. Rev. Immunol. 24, 541– 570 5 Schatz, D.G. and Baltimore, D. (1988) Stable expression of immunoglobulin gene V(D)J recombinase activity by gene transfer into 3T3 fibroblasts. Cell 53, 107–115 6 Schatz, D.G. et al. (1989) The V(D)J recombination activating gene. RAG-1. Cell 59, 1035–1048 7 Lewis, S.M. (1994) The mechanism of V(D)J joining: lessons from molecular, immunological, and comparative analyses. Adv. Immunol. 56, 27–150 8 Difilippantonio, M.J. et al. (1996) RAG1 mediates signal sequence recognition and recruitment of RAG2 in V(D)J recombination. Cell 87, 253–262 9 Ramsden, D.A. et al. (1996) Distinct DNA sequence and structure requirements for the two steps of V(D)J recombination signal cleavage. EMBO J. 15, 3197–3206 10 Casellas, R. et al. (2001) Contribution of receptor editing to the antibody repertoire. Science 291, 1541–1544 11 Melamed, D. and Nemazee, D. (1997) Self-antigen does not accelerate immature B cell apoptosis, but stimulates receptor editing as a consequence of developmental arrest. Proc. Natl. Acad. Sci. U. S. A. 94, 9267–9272 12 Melamed, D. et al. (1998) Developmental regulation of B lymphocyte immune tolerance compartmentalizes clonal selection from receptor selection. Cell 92, 173–182 13 Sandel, P.C. and Monroe, J.G. (1999) Negative selection of immature B cells by receptor editing or deletion is determined by site of antigen encounter. Immunity 10, 289–299
Review
TRENDS in Immunology
14 Kleinfield, R. et al. (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 15 Reth, M. et al. (1986) A novel VH to VHDJH joining mechanism in heavychain-negative (null) pre-B cells results in heavy-chain production. Nature 322, 840–842 16 Covey, L.R. et al. (1990) VH to VHDJH rearrangement is mediated by the internal VH heptamer. Int. Immunol. 2, 579–583 17 Zhang, Z. et al. (2004) The molecular basis and biological significance of VH replacement. Immunol. Rev. 197, 231–242 18 Chen, C. et al. (1995) Immunoglobulin heavy chain gene replacement: a mechanism of receptor editing. Immunity 3, 747–755 19 Chen, C. et al. (1995) The site and stage of anti-DNA B-cell deletion. Nature 373, 252–255 20 Lopez-Macias, C. et al. (1999) Secondary rearrangements and hypermutation generate sufficient B cell diversity to mount protective antiviral immunoglobulin responses. J. Exp. Med. 189, 1791–1798 21 Madan, M.K. et al. (2000) Specific antibody production by VH-gene replacement. Eur. J. Immunol. 30, 2404–2411 22 Pewzner-Jung, Y. et al. (1998) B cell deletion, anergy, and receptor editing in ‘knock in’ mice targeted with a germline-encoded or somatically mutated anti-DNA heavy chain. J. Immunol. 161, 4634–4645 23 Taki, S. et al. (1993) Targeted insertion of a variable region gene into the immunoglobulin heavy chain locus. Science 262, 1268–1271 24 Koralov, S.B. et al. (2006) Antibody repertoires generated by VH replacement and direct VH to JH joining. Immunity 25, 43–53 25 Lutz, J. et al. (2006) VH replacement rescues progenitor B cells with two nonproductive VDJ alleles. J. Immunol. 177, 7007–7014
Vol.28 No.3
26 Watson, L.C. et al. (2006) Paucity of V-D-D-J rearrangements and VH replacement events in lupus prone and nonautoimmune TdT/ and TdT+/+ mice. J. Immunol. 177, 1120–1128 27 Wang, Y.H. et al. (2003) V(D)J recombinatorial repertoire diversification during intraclonal pro-B to B-cell differentiation. Blood 101, 1030–1037 28 Zhang, Z. et al. (2003) Contribution of VH gene replacement to the primary B cell repertoire. Immunity 19, 21–31 29 Ohm-Laursen, L. et al. (2006) No evidence for the use of DIR, D-D fusions, chromosome 15 open reading frames or VH replacement in the peripheral repertoire was found on application of an improved algorithm, JointML, to 6329 human immunoglobulin H rearrangements. Immunology 119, 265–277 30 Liu, Y. et al. (2005) Potential contribution of VH gene replacement in immunity and disease. Ann. N. Y. Acad. Sci. 1062, 175–181 31 Radic, M.Z. et al. (1993) Residues that mediate DNA binding of autoimmune antibodies. J. Immunol. 150, 4966–4977 32 Radic, M.Z. and Zouali, M. (1996) Receptor editing, immune diversification and self-tolerance. Immunity 5, 505–511 33 Nunez, C. et al. (1996) B cells are generated throughout life in humans. J. Immunol. 156, 866–872 34 Wardemann, H. et al. (2003) Predominant autoantibody production by early human B cell precursors. Science 301, 1374–1377 35 Ueda, Y. et al. (2004) Inflammation controls B lymphopoiesis by regulating chemokine CXCL12 expression. J. Exp. Med. 199, 47– 58 36 Sims, G.P. et al. (2005) Identification and characterization of circulating human transitional B cells. Blood 105, 4390–4398
Elsevier.com – linking scientists to new research and thinking Designed for scientists’ information needs, Elsevier.com is powered by the latest technology with customer-focused navigation and an intuitive architecture for an improved user experience and greater productivity. The easy-to-use navigational tools and structure connect scientists with vital information – all from one entry point. Users can perform rapid and precise searches with our advanced search functionality, using the FAST technology of Scirus.com, the free science search engine. Users can define their searches by any number of criteria to pinpoint information and resources. Search by a specific author or editor, book publication date, subject area – life sciences, health sciences, physical sciences and social sciences – or by product type. Elsevier’s portfolio includes more than 1800 Elsevier journals, 2200 new books every year and a range of innovative electronic products. In addition, tailored content for authors, editors and librarians provides timely news and updates on new products and services. Elsevier is proud to be a partner with the scientific and medical community. Find out more about our mission and values at Elsevier.com. Discover how we support the scientific, technical and medical communities worldwide through partnerships with libraries and other publishers, and grant awards from The Elsevier Foundation. As a world-leading publisher of scientific, technical and health information, Elsevier is dedicated to linking researchers and professionals to the best thinking in their fields. We offer the widest and deepest coverage in a range of media types to enhance cross-pollination of information, breakthroughs in research and discovery, and the sharing and preservation of knowledge.
Elsevier. Building insights. Breaking boundaries. www.elsevier.com www.sciencedirect.com
137