Cis-acting regulatory sequences promote high-frequency gene conversion
http://www.100md.com
《核酸研究医学期刊》
Department of Molecular Biology and Genetics, College of Biological Science, University of Guelph, Guelph, Ontario, Canada N1G 2W1
* To whom correspondence should be addressed. Tel: +1 519 824 4120; Ext. 54788; Fax: +1 519 837 2075; Email: mdbaker@uoguelph.ca
ABSTRACT
In mammalian cells, little is known about the nature of recombination-prone regions of the genome. Previously, we reported that the immunoglobulin heavy chain (IgH) μ locus behaved as a hotspot for mitotic, intrachromosomal gene conversion (GC) between repeated μ constant (Cμ) regions in mouse hybridoma cells. To investigate whether elements within the μ gene regulatory region were required for hotspot activity, gene targeting was used to delete a 9.1 kb segment encompassing the μ gene promoter (Pμ), enhancer (Eμ) and switch region (Sμ) from the locus. In these cell lines, GC between the Cμ repeats was significantly reduced, indicating that this ‘recombination-enhancing sequence’ (RES) is necessary for GC hotspot activity at the IgH locus. Importantly, the RES fragment stimulated GC when appended to the same Cμ repeats integrated at ectopic genomic sites. We also show that deletion of Eμ and flanking matrix attachment regions (MARs) from the RES abolishes GC hotspot activity at the IgH locus. However, no stimulation of ectopic GC was observed with the Eμ/MARs fragment alone. Finally, we provide evidence that no correlation exists between the level of transcription and GC promoted by the RES. We suggest a model whereby Eμ/MARS enhances mitotic GC at the endogenous IgH μ locus by effecting chromatin modifications in adjacent DNA.
INTRODUCTION
Homologous recombination (HR) is the process by which genetic information is exchanged between two similar or identical DNA duplexes. In eukaryotes, recombination plays an important role in generating genetic diversity in meiosis, and in mitotic cells, represents a pathway for the repair of DNA damage. Nevertheless, these benefits are associated with the risk of aberrant rearrangements due to recombination between the myriad of repeated sequences in the eukaryotic genome. Germane to this risk is the knowledge that the frequency of recombination is not uniform, in that some loci exhibit relatively high frequencies of recombination (hotspots), while others exhibit relatively low frequencies of recombination (coldspots). This fact has important implications for the maintenance of genomic stability.
Relatively little is known about the nature of mitotic recombination hotspots in mammalian cells. However, since the primary role of HR in somatic cells is as a DNA repair mechanism, it stands to reason that regions of the genome that are prone to DNA damage would be active in HR. Current evidence suggests that errors that occur in the course of normal DNA metabolism, such as transcription (1), and replication (2), are a major source of recombinogenic lesions. It follows then, that DNA regulatory elements could act as recombination hotspots as an indirect by-product of their normal biological function. For example, the yeast recombination hotspot, HOT1, contains an initiation site and enhancer of transcription by RNA polymerase I (3). The stimulatory activity of HOT1 correlates with its capacity to promote transcription through the recombining sequences (4).
Other DNA motifs are thought to stimulate genomic rearrangement through their effects on the secondary structure of DNA. Regions of alternating purines and pyrimidines that can adopt the Z-DNA conformation (5–7), trinucleotide repeats (8) and inverted (palindromic) repeats that can extrude and form hairpin or cruciform structures (9,10) are all prone to rearrangement in eukaryotic cells.
Indirect evidence suggests higher order chromatin structure might also influence HR rates. It is plausible that a change in chromatin structure facilitates the access of recombination proteins, or possibly, leads to hypersensitivity to DNA-damaging agents. Hyper-recombination phenotypes were reported for certain yeast mutants defective in proteins involved in chromatin-mediated repression of transcription (11). A correlation between Sir2-mediated DNA silencing and a more ‘closed’ chromatin structure was shown by Fritze et al. (12). Since DNA silencing also correlates with reduced recombination (13), it was suggested that ‘closed’ chromatin has a double effect in repressing both transcription and recombination.
Our laboratory previously reported that the mouse immunoglobulin heavy-chain (IgH) μ locus acts as a hotspot for spontaneous mitotic gene conversion (GC) (14). The assay system monitors intrachromosomal GC events between closely linked direct repeats of the IgH μ gene constant (Cμ) region. In this paper, we report the identification of a 9.1 kb segment of DNA encompassing the IgH μ gene regulatory region, which stimulates GC between adjacent Cμ repeats both at the endogenous IgH locus, and when appended to the same Cμ repeats stably integrated at ectopic genomic sites. We also show that deletion of the IgH major intronic enhancer (Eμ) and flanking matrix attachment regions (MARs) from the regulatory region abolishes GC hotspot activity at the chromosomal IgH μ locus. However, the Eμ/MARs fragment by itself does not stimulate recombination at ectopic genomic sites. Finally, we provide evidence that there is no correlation between the level of transcription and the level of GC promoted by the IgH μ gene regulatory region. We suggest a model whereby the endogenous Eμ/MARs effects chromatin modifications in adjacent chromosomal regions that enhance spontaneous mitotic GC.
MATERIALS AND METHODS
Hybridoma cell lines
The hybridoma cell line, Sp6/HL, bears a single copy of the trinitrophenyl (TNP)-specific chromosomal IgH μ chain gene and makes normal, cytolytic TNP-specific IgM (-chain) (15,16). The following Sp6/HL-derived hybridoma cell lines were used in this study. Hybridoma igm10 is a mutant that has lost the TNP-specific chromosomal μ gene (16). Hybridoma Im/RCμ2A2 was constructed by gene targeting in the Cμ region of the haploid, TNP-specific chromosomal μ gene, as described previously (17). The conditions for hybridoma cell growth are described elsewhere (15,16).
DNA transfer
Linear plasmid vector DNA (8.7 pmol) was introduced into 2 x 107 hybridoma cells by electroporation as described previously (17). Individual transformants were recovered by limited dilution cloning procedures as described in (18,19).
Quantitative PCR (qPCR) assay
The conditions used for qPCR and details of the construction of the vector control have been reported previously (14).
DNA analysis
Preparation of genomic DNA was performed by the method of Gross-Bellard et al. (20) with the exception of DNA samples used in the qPCR assay, which were prepared with the QIAamp DNA mini kit (QIAGEN). For Southern analysis, restriction enzymes were purchased from New England Biolabs Inc. (Mississauga, ON, Canada) and used in accordance with the manufacturer's specification. Gel electrophoresis, transfer of DNA onto nitrocellulose membrane, 32P-labeled probe preparation and hybridization were all performed according to standard procedures (21). Primers used in this study were synthesized at Sigma (Oakville, ON, Canada), and their sequences have been reported previously (14). New primers used in this study are as follows: 23819, 5'-CAT CCT CCT CCT CAT CAT CGT CAT-3'; CμXbaRI, 5'-GAA TCT GTC TTC TTG CCT CCT GTC-3'; 24226, 5'-GCT GTG TAG AAG TAC TCG CCG ATA-3'.
RNA analysis
Hybridoma cells were grown to a density of 3–5 x 105 cells/ml. Total RNA was isolated using Trizol (Gibco) in accordance with the manufacturer's specification, and assayed by dot blotting for hybridization to a Cμ fragment (Probe F) and a ?-actin probe. The procedures of Baumann et al. (22) were adopted for RNA dot blotting, while hybridizations were performed according to standard methods (21).
RESULTS
Role of the IgH μ gene regulatory region in spontaneous HR hotspot activity
In this study, we investigated whether cis-acting DNA elements within the IgH μ gene regulatory region promote intrachromosomal GC. The haploid IgH μ locus in the hybridoma cell line, Im/RCμ2A2, bears a pair of Cμ region heteroalleles that exhibit high-frequency intrachromosomal GC (14). The upstream (5') Cμ region is derived from the wild-type Sp6/HL hybridoma and resides in its normal position to be expressed from the TNP-specific VH region. Vector pSV2neo sequences separate the 5' wild-type Cμ region from the downstream (3') Cμ region, which bears a 2 bp deletion in exon Cμ3 (referred to as the mutant igm482 Cμ3 region) (15,16).
The 9.6 kb omega ()-form vector, pVHCμKO, was used in the targeted modification of the Im/RCμ2A2 μ gene (Figure 1A). The pVHCμKO backbone consists of an enhancer-trap derivative of pSV2hyg in which the 372 bp NsiI/NdeI fragment encompassing the simian virus 40 (SV40) early region enhancer residing upstream of the hygromycin phosphotransferase (hyg) gene was deleted. Although not relevant to this study, the Herpes Simplex Virus-1 (HSV-1) thymidine kinase (tk) gene is also present. In pVHCμKO, the flanking arms consist of a 2.3 kb EcoRI/HpaI segment with homology 5' of the μ gene promoter (Pμ) and a 1.6 kb XbaI/MfeI Cμ region fragment, positioned to the left and right of pSV2hyg, respectively. Flanking arm alignment with isogenic regions of the Im/RCμ2A2 chromosomal μ gene is indicated by the dashed lines in Figure 1A.
Figure 1. Construction of 2A2VH cell lines. (A) The structure of the haploid, chromosomal μ locus in the recipient murine hybridoma cell line, Im/RCμ2A2, is presented along with the 9.7 kb replacement vector, pVHCμKO. Regions of alignment between the homologous arms of the vector and Im/RCμ2A2 chromosomal μ locus are indicated by the dashed lines. (B) The structure of the recombinant μ locus following targeted gene replacement by the vector pVHCμKO. In both panels, the mutant igm482 Cμ3 exon is indicated by an open box, while wild-type Cμ exons are indicated by filled boxes. Although not labeled, matrix attachment regions (MARs) are indicated by filled diamonds. The diagnostic fragment sizes generated following digestion with the indicated restriction enzymes are presented. DNA probe fragment F is an 870 bp XbaI/BamHI fragment while probe tk is a 1.3 kb fragment from the HSV-1 thymidine kinase (tk) gene. VH, TNP-specific μ-heavy-chain variable region; Pμ, μ-gene promoter; Eμ, major intronic enhancer; Sμ, μ-gene switch region; Cμ, μ-gene constant region; neo, neomycin phosphotransferase gene; tk, HSV-1 thymidine kinase gene; hyg, hygromycin resistance gene; B, BamHI; E, EcoRI; H, HpaI; M, MfeI; Sc, ScaI; X, XbaI. The figure is not drawn to scale.
Targeted gene replacement by pVHCμKO deletes all coding and regulatory DNA sequences required for expression of the TNP-specific μ heavy chain gene in Im/RCμ2A2 (Figure 1B) (23). Thus, following electroporation, culture supernatant from a total of 270 hygR transformants was tested for the presence of TNP-specific IgM, by complement-dependent lysis of TNP-coupled sheep erythrocytes in spot tests (15). The structure of the IgH μ locus in six IgM–, hygR transformants was examined by Southern analysis to identify those in which targeted gene replacement had occurred. The fragment sizes of the recipient μ locus in Im/RCμ2A2 (Figure 1A), and of correctly targeted transformants (Figure 1B), following digestion with BamHI and hybridization with Cμ-specific probe F are shown. Two independent hybridoma cell lines, 2A2VH-82 and 2A2VH-236 were identified as having the correctly targeted structure (data not shown).
For determination of GC frequencies, a quantitative PCR (qPCR) assay was used, as described previously (14). In brief, the assay makes use of four primers. Primers 1 and 2 bind upstream of the 5' and 3' Cμ regions, respectively. Primers 3 and 4 are specific for the wild-type and mutant igm482 Cμ3 regions, respectively. Two related hybridomas were used to standardize the assay, Im/RCμ3/1-7 (abbreviated 3/1-7) and Im/RCμD7-7 (abbreviated D7-7) (Figure 2A). In hybridoma cell line, 3/l-7, both the 5' and 3' Cμ regions are wild-type (designated, double wild-type), whereas, in hybridoma cell line, D7-7 they are both mutant igm482 Cμ3 regions (designated double mutant). Figure 2A indicates the primer binding sites and product sizes generated by PCR amplification of 3/1-7 and D7-7. To circumvent the problem of variability in efficiency of amplification, 10–9 pmol of a heterologous vector control, sharing only the primer binding sites with the target sequence, was included in each PCR reaction. Amplification of the vector generates a 1.3 kb product (Figure 2A). Genomic mixtures of 3/1-7 and D7-7 were prepared in ratios of 1:50, 1:200, 1:500 and 1:1000, and amplified using primer pair 2/3 for generation of the wild-type, 1.9 kb 3' Cμ region product (Figure 2B, lanes 2–5). Following gel electrophoresis, quantitation of the EtBr staining in the 1.9 kb target band divided by that in the 1.3 kb vector control band yields a ratio that, when plotted against the wild-type Cμ region copy number, generates a standard curve (Figure 2B, inset). In a similar manner, the qPCR assay was standardized with the primer pairs 1/3, 1/4 and 2/4 to detect the wild-type 5' Cμ, mutant 5' Cμ and mutant 3' Cμ regions, respectively.
Figure 2. qPCR assay to measure gene conversion between the Cμ region repeats. (A) Hybridoma 3/1-7 has the wild-type sequences in both the 5' and 3' Cμ regions (double wild-type), while hybridoma D7-7 bears the mutant igm482 Cμ3 exon in both the 5' and 3' Cμ regions (double mutant). The PCR primers and the sizes of the amplified products they produce are indicated. Primer 1 binds upstream of the 5' Cμ region and outside the region of homology in the 3' Cμ region. Primer 2 binds within pSV2neo vector sequences upstream of the 3' Cμ region. Primers 3 and 4 bind specifically to the complementary strand of the wild-type and mutant igm482 Cμ3 exons, respectively. The vector control contains the primer 1 and 2 binding sites and primer 3 and 4 binding sites on opposite strands flanking a segment of the thymidine kinase (tk) gene. (B) Lanes 2–5, standard solutions of genomic DNA consisting of 3/1-7 and D7-7 prepared in ratios of 1:50, 1:200, 1:500, 1:1000 along with a constant amount of vector control (10–9 pmol) were amplified using primer pair 2/3 for generation of the 1.9-kb wild-type 3' Cμ region product and 1.3-kb vector control product. DNA from the double mutant, D7-7 cell line, was included as a negative control (lane 6). Quantitation of the EtBr staining in the target band divided by that in the vector control band yielded the standard curve (inset). Lanes 7–11, representative samples from 2A2VH-82 subclone 3 subculture wells 1–5, amplified with primer pair 2/3 to detect gene conversion in the 3' Cμ region to the wild-type sequence. Positive signals are present in subcultures 2 and 4. Abbreviations are the same as in Figure 1. The figure is not drawn to scale.
To quantify the GC frequency in the cell lines, replicate genomic DNA samples, along with a constant amount of vector control, were amplified using primer pair 2/3. Following gel electrophoresis, and quantitation of the EtBr staining, the resulting target:vector ratio was used to determine the GC frequency from the standard curve. In cases where the GC frequency was below the sensitivity of the assay (<0.001), the culture was distributed at a density of 500 cells/well in 24-well plates. Following cell growth, genomic DNA was prepared from each culture well and qPCR analysis was performed on the separate DNA preparations using primer pair 2/3. As an example, Figure 2B (lanes 7–11) presents five representative genomic DNA samples for hybridoma 2A2VH-82 subclone 3. From the fraction of negative wells in the qPCR assay (as examples, lanes numbered 1, 3 and 5 in Figure 2B) and the Poisson distribution, the mean GC frequency between the Cμ repeats was calculated.
Similar GC frequencies were detected previously for the 5' and 3' Cμ regions in Im/RCμ2A2 (14), and therefore, in cell lines, 2A2VH-82 and 2A2VH-236, the frequency of GC of the recipient, 3' mutant Cμ region by the donor, 5' wild-type Cμ region was determined. Three subclone cultures of 2A2VH-82 and two subclone cultures of 2A2VH-236 were each started from a single cell. Southern blot and PCR analysis revealed that the μ gene structure in each subclone was identical to that in the parent cell line (data not shown). Each subclone was grown for 25 generations in medium supplemented with G418 (0.6 mg/ml) and hygromycin (0.6 mg/ml), and then assayed by qPCR using primer pair 2/3. This analysis revealed that the mean GC frequency between the Cμ repeats in 2A2VH-82 and 2A2VH-236 was 0.2 x 10–3 recombinants/cell (Table 1). In comparison, GC between the same repeats integrated at the wild-type μ locus of hybridoma Im/RCμ2A2 was previously shown to occur at a frequency of 6.1 x 10–3 recombinants/cell (14). The procedures and conditions used for cell growth, and determination of GC frequencies were identical for both studies (14). The 31-fold difference in the GC frequencies is highly significant (t-test of log-transformed data; P = 0.0002), suggesting that the IgH μ gene regulatory region is required for GC hotspot activity.
Table 1. Frequency of gene conversion between Cμ region repeats
The IgH μ gene regulatory region stimulates intrachromosomal gene conversion between Cμ region repeats at ectopic sites in the hybridoma genome
Since deletion of the regulatory region from the endogenous IgH locus reduced the frequency of GC between adjacent Cμ repeats, it was of interest to determine whether the region stimulated GC when appended to the same Cμ repeats integrated outside the IgH locus. For these studies, we used a derivative of the vector pCμ-repeat described previously (Figure 3A) (14). It bears 4.3 kb segments encoding the mutant igm482 and wild-type Cμ regions flanking a pSV2neo vector backbone. The 10.8 kb EcoRI/MfeI fragment encompassing the VH region and the VH–Cμ intron (Figure 1A) was obtained from the cloned wild-type Sp6 μ gene (24), and inserted into pCμ-repeat in its correct position and orientation, upstream of the mutant igm482 5' Cμ region to create the vector, pVHCμRep (Figure 3B). The μ gene structure in pVHCμRep is isogenic to that in Im/RCμ2A2, with the exception of an 2.8 kb deletion in the μ gene switch (Sμ) region that occurred during cloning of the genomic DNA in Escherichia coli, leaving an 0.4 kb residual Sμ segment (24). The vector pVHCμRep was linearized at the unique NotI site and transferred by electroporation into the Sp6-derived hybridoma cell line igm10, which lacks the endogenous chromosomal μ gene (16). A total of 96 individual G418R transformants were isolated and screened by PCR analysis in conjunction with Southern blotting (Figure 3B) to determine the structure and copy number of the integrated vector. Five cell lines, designated VHCμRep-10, VHCμRep-12, VHCμRep-15, VHCμRep-19 and VHCμRep-21, were identified as having a single copy of the integrated vector with the structure shown in Figure 3B (data not shown). Each transformant was expanded from a single cell and grown for 25 generations under G418 selection. Genomic DNA was prepared from each culture, and qPCR utilizing primer pair 1/3 was performed to measure the frequency of GC in the 5' Cμ region. It should be noted that randomly integrated constructs are susceptible to position effects, which could influence GC rates. As an unusually high or low frequency of GC in one of the cell lines could have a misleading effect on the mean GC frequency (due to the low sample size), we felt that comparison of median values was more appropriate for the ectopic data. When compared to transformants containing just the Cμ repeats integrated at ectopic genomic sites (R/CμRepeat cell lines in Table 1) (14), the frequency of GC was stimulated up to as much as 30-fold in the VHCμRep transformants (Table 1) (Mann–Whitney test, P = 0.036).
Figure 3. Constructs integrated at ectopic sites in the R/CμRepeat and VHCμRep transformants. The structures of the linearized vectors transfected into hybridoma igm10, which lacks the endogenous chromosomal μ gene (16), are shown. (A) The vector pCμ-repeat bears the 4.3-kb Cμ region repeats inserted into a pSV2neo vector backbone in the tandem orientation (14). The upstream (5') Cμ region bears the 2-bp igm482 Cμ3 deletion (open box), but is otherwise isogenic with the downstream (3') Cμ region. (B) The vector pVHCμRep contains the EcoRI/MfeI VH-region fragment from the hybridoma Sp6/HL inserted into the MfeI site of pCμ-repeat in its correct position and orientation with respect to the Cμ regions. As described in the text, a residual 0.4 segment remains of the Sμ region (denoted Sμ*) (24). The diagnostic fragment sizes generated following digestion with EcoRI and SpeI, and hybridization with probe F (Figure 1) and with probe Eμ, a 992-bp XbaI fragment, are presented. Southern blotting using both single cutters and noncutters was used to determine the copy number of the integrated vector (data not shown). N, NotI; Sp, SpeI; other abbreviations are the same as in Figure 1. The figure is not drawn to scale.
The μ gene switch region, Sμ, is not required for high-frequency gene conversion at the IgH locus
The frequency of GC promoted by the IgH μ gene regulatory region at ectopic sites was only 20–30% of the mean value measured in Im/RCμ2A2 (Table 1). This might suggest a role for Sμ in stimulating GC, since as indicated above, it was largely deleted from the construct used in the ectopic studies. To investigate this, the frequency of GC was determined in two related hybridomas, Emut1-Sμ+ and Emut1-Sμ–, that were isolated in a previous study (14). As shown in Figure 4, both cell lines contain Cμ repeats positioned at the IgH μ locus: the 3' Cμ region is wild type, while the 5' Cμ region is mutant as a consequence of a 4 bp insertion in Cμ exon 1. Cell line Emut1-Sμ+ bears the chromosomal Sμ region upstream of both the 5' and 3' Cμ regions, whereas Emut1-Sμ– contains the 2.8 kb Sμ deletion upstream of both Cμ regions (Figure 4). The Sμ deletion does not affect μ gene expression (24).
Figure 4. Structure of the haploid, chromosomal IgH μ gene in the Emut1-Sμ+ and Emut1-Sμ– hybridoma cell lines. In each cell line, the mutant, recipient 5' Cμ region bears a BspHI marker in Cμ exon 1 and is separated from the wild-type, donor 3' Cμ region by the integrated pSV2neo vector sequences. In hybridoma Emut1-Sμ+, the endogenous μ gene switch region (denoted Sμ) is present upstream of both the 5' and 3' Cμ regions. Whereas, hybridoma Emut1-Sμ– bears an 2.8 kb Sμ deletion (denoted Sμ*) upstream of both Cμ regions. The figure is not drawn to scale.
The frameshift mutation in the expressed 5' Cμ region causes Emut1-Sμ+ and Emut1-Sμ– to produce a truncated μ heavy chain that cannot form pentameric IgM and activate complement-dependent lysis of TNP-coupled sheep erythrocytes (14). During growth under G418-selective conditions, conversion of the recipient, 5' mutant Cμ region by the donor, 3' wild-type Cμ region can restore normal, TNP-specific IgM production in the hybridoma cells, allowing them to be detected as plaque-forming cells (PFC) in a TNP-specific plaque assay (17). Previously, we showed that there was no difference in the frequency of GC of the 4 bp mutation when inserted at different sites spanning the Cμ region (14). Furthermore, the frequency of GC of the 4 bp insertion was similar to that measured for the 2 bp deletion in Im/RCμ2A2 (14). Three subclone cultures of the Emut1-Sμ+ and Emut1-Sμ– mutant hybridoma cell lines were started from a single cell. Southern analysis of each subclone revealed the same μ gene structure in the subclones and parental cultures (data not shown). Each subclone culture was grown for 24 generations and then subjected to the TNP-specific plaque assay. As shown in Table 1, the mean frequency of generating TNP-specific PFC in each cell line was not significantly different (t-test, P = 0.30). This suggests that the Sμ region is not required for high-frequency intrachromosomal GC.
The IgH major intronic enhancer, Eμ, is required for stimulation of gene conversion by the IgHμ gene regulatory region
To determine whether the major intronic enhancer (Eμ) and/or the flanking matrix attachment regions (MARs) play a role in HR hotspot activity, a ‘hit-and-run’ gene targeting technique (25,26) was used to delete these elements from the endogenous IgH μ locus in the hybridoma, Im/RCμ2A2. The ‘hit’ step takes advantage of the 10.8 kb enhancer-trap, insertion vector pVHEμ (Figure 5A). pVHEμ contains the 8.9 kb XbaI fragment encompassing the VH region exon and VH–Cμ intron, inserted into the enhancerless pSV2hyg backbone bearing the HSV-1 tk gene. The region of homology was modified by deleting the 992 bp XbaI Eμ/MARs-containing fragment. The vector was linearized at the unique AfeI site, which provides 3.0 and 2.2 kb of homology on the 5' and 3' sides of the cut site, respectively, and transferred by electroporation into Im/RCμ2A2 (17). A total of 446 independent hygR transformants were isolated, 106 of which were screened by PCR using primers 23819 and CμXbRI for the 9.5 kb product that identifies the endogenous, chromosomal IgH region (Figure 5A). Southern analysis was performed on five cell lines from which the 9.5 kb PCR product was not amplified to identify those in which a single copy of the vector had correctly integrated at the IgH locus. The fragment sizes of the recipient μ locus of cell line Im/RCμ2A2 (Figure 5A), and of correctly targeted transformants (Figure 5B), following digestion with BamHI and hybridization with Probe N, are shown. Two hybridoma cell lines, Im/2VHRCμ2A2-17 and Im/2VHRCμ2A2-26 were identified as having the correctly targeted structure (data not shown).
Figure 5. Construction of 2A2Eμ+ and 2A2Eμ– cell lines by ‘Hit’ and ‘Run’ gene targeting at the chromosomal μ locus. (A) The ‘hit’ step involves targeted integration of the enhancer-trap insertion vector pVHEμ into the chromosomal μ gene of the hybridoma Im/RCμ2A2. pVHEμ contains a cloned segment of theSp6/HL μ gene regulatory region from which the 992 bp Eμ/MARs-containing XbaI fragment (indicated by an inverted ‘V’) was deleted. Regions of alignment between the vector and the Im/RCμ2A2 chromosomal μ locus are indicated by the dashed lines. (B) The structure of the Im/RCμ2A2 chromosomal μ gene following targeted integration of a single copy of the pVHEμ vector. (C) The ‘run’ step involves excision of the integrated vector and one copy of homologous DNA by intrachromosomal HR, an event that exploits the ability to select against the HSV-1 tk gene with gancyclovir (Ganc). Intrachromosomal reciprocal crossover that occurs in the region of homology 5' or 3' of the Eμ/MARs deletion generates the μ gene structure depicted in (D) and (E), respectively. The diagnostic fragment sizes generated following digestion with BamHI are presented. DNA probe N consists of adjacent 475 and 495 bp NheI fragments. The positions of primers 23819 and CμXbRI are indicated. Af, AfeI; other abbreviations are the same as in Figure 1. The figure is not drawn to scale.
The ‘run’ step involves excision of the integrated vector by intrachromosomal HR between the duplicated region of homology that removes the integrated vector, hyg and tk genes along with one copy of homologous DNA (Figure 5C). Duplicate subcultures of one cell line (Im/2VHRCμ2A2-17) were plated in media lacking hygromycin to allow for growth of hygS hybridomas in which vector excision had occurred. Following growth, each subculture was distributed at low cell density in 480 individual culture wells in media supplemented with gancyclovir (Ganc) to select against the tk gene (25). A total of 71 and 141 GancR colonies were recovered from the duplicate subcultures, respectively. As shown in Figure 5D and E, and outlined in the figure legend, multiple IgH μ gene structures are possible following vector excision depending on the location of the crossover. Genomic DNA was prepared from the GancR colonies and screened by PCR using primer pair 23819–CμXbRI (Figure 5A) (data not shown). From each of the duplicate subcultures, two GancR cell lines that contained the 8.5 kb PCR product diagnostic of the μ gene excision product that is deleted for Eμ/MARs (Figure 5D) (designated, 2A2Eμ–#1–5, 2A2Eμ–#1–6, 2A2Eμ–#2–4 and 2A2Eμ–#2–10), and two GancR cell lines that contained the 9.5 kb PCR product diagnostic of the normal μ gene locus (Figure 5E) (designated, 2A2Eμ+#1–4, 2A2Eμ+#1–8, 2A2Eμ+#2–16 and 2A2Eμ+#2–20) were isolated. Southern analysis using BamHI and Probe N (Figure 5D and E) confirmed the IgH μ gene structures in the eight cell lines (data not shown).
For determination of GC frequencies, each cell line was expanded from a single cell for 25 generations under G418 selection. Genomic DNA was prepared from each culture, and qPCR utilizing primer pair 1/4 was performed to measure the frequency of GC in the 5' Cμ region. This analysis revealed a mean GC frequency of <0.11 x 10–3 recombinants/cell in the Eμ/MARs deleted cell lines (Table 1). This value is 30-fold lower than the mean GC frequency measured in the cell lines with an intact μ gene regulatory region (3.4 x 10–3 recombinants/cell) (Table 1) (t-test of log-transformed data; P = 4 x 10–6). This suggests a role for the Eμ/MARs fragment in promoting high-frequency GC between the Cμ regions at the IgH μ locus.
The Eμ/MARs segment alone is not sufficient to stimulate gene conversion between Cμ region repeats at ectopic sites in the hybridoma genome
Since the DNA fragment containing the entire IgH μ gene regulatory region was able to stimulate GC between Cμ region repeats at ectopic sites in the genome, it was of interest to determine whether the Eμ/MARs fragment alone also stimulated GC outside the IgH μ locus. To examine this, the 992 bp XbaI fragment containing Eμ/MARs was inserted into pCμ-repeat, in its correct orientation, upstream of the mutant 5' Cμ region (Figure 6). The vector was linearized at the unique NotI site and transferred by electroporation into the hybridoma, igm10 (16). A total of 88 individual G418R transformants were isolated and screened by PCR analysis in conjunction with Southern blotting using Cμ probe F (Figure 6) to determine the structure and copy number of the integrated vector (data not shown). From this screening, five cell lines were recovered that contained a single, intact vector integrated in the genome, designated EμCμRep-17, EμCμRep-46, EμCμRep-61, EμCμRep-75 and EμCμRep-81. Each of the transformants was expanded from a single cell and grown for 25 generations under G418 selection. Genomic DNA was prepared from each culture, and qPCR analysis utilizing primer pair 1/3 was performed to measure the frequency of GC in the 5' Cμ region. The median GC frequency was 0.10 x 10–3 recombinants/cell, which is similar to the median GC frequency of 0.07 x 10–3 recombinants/cell measured in the R/CμRepeat cell lines (Table 1). Thus, there was no significant difference in the frequency of GC between Cμ repeats at ectopic genomic sites, with or without the Eμ/MARs fragment (Mann–Whitney test, P = 0.24).
Figure 6. Influence of Eμ on gene conversion between ectopic Cμ repeats. The structure of hybridoma cell lines bearing the linearized vector, pEμCμRep, as determined by Southern blot and PCR analysis is shown. The vector was derived from pCμ-repeat (Figure 3A), by inserting the 995-bp Eμ/MARs-containing XbaI fragment into the MfeI site, in its correct orientation 5' of the Cμ regions. The diagnostic fragment sizes generated following digestion with XbaI and SpeI, and hybridization with probe F (Figure 1), are presented. PCR amplification with primer pair 24226–CμXbRI generates a 1.9 kb PCR product. Southern blotting using both single cutters and noncutters was used to determine the copy number of the integrated vector (data not shown). Abbreviations are the same as in Figure 1. The figure is not drawn to scale.
Evidence that transcription initiated at the μ gene promoter is not necessary for the stimulation of gene conversion by the IgH μ gene regulatory region
It remained a possibility that the reduced GC frequency in the absence of the endogenous Eμ/MARs fragment was an indirect consequence of a reduced level of μ gene transcription. To determine what effect the Eμ/MARs deletion had on μ gene transcript level, the amount of μ-specific RNA in the cell lines was examined by RNA dot blot analysis (Figure 7) (16,22,27). Previously, deletion of the majority of the VH–Cμ intron, including the Eμ/MARs elements, was shown to cause equivalent changes in both the levels of μ mRNA and nuclear run-on activity, suggesting that no elements are present within this region that affect RNA stability or processing efficiency (28). Therefore, we suggest that the μ-specific RNA levels measured in our cell lines is probably a good indicator of the level of transcription through the Cμ regions. Densitometric analysis revealed that the μ-specific RNA levels in the Eμ/MARs-deleted cell lines, 2A2Eμ–#1–5 and 2A2Eμ–#2–10, were 25% and 10% of that in Im/RCμ2A2, respectively (Table 2). In comparison, μ RNA levels in the transformants bearing the intact Eμ/MARs fragment (2A2Eμ+ #1–4 and 2A2Eμ+ #2–20), were 144% and 92% of that in Im/RCμ2A2, respectively.
Figure 7. RNA dot blot analysis of μ RNA. Serial 1:4 dilutions of total RNA were applied to a nitrocellulose filter starting with 10 μg as described in (22). Relative levels of μ RNA were determined by hybridization with 32P-labeled Cμ-specific Probe F (Figure 1), and separately with a 32P-labeled 2.0-kb PstI fragment containing the chicken ?-actin cDNA (27). The cell line igm10 is an Sp6-derived mutant that lacks the endogenous chromosomal μ gene (16).
Table 2. Cμ-specific RNA levels
To investigate whether μ-specific RNA levels correlated with the frequency of GC between the Cμ repeats, a 154 bp XbaI/NcoI fragment encompassing the Pμ TATA box, and octamer motif, 5'-ATTTGCAT-3', was deleted from the vector pVHCμRep to generate pVHCμRepPμ (Figure 8). Protein binding to the octamer motif of Pμ is required for high-level expression of the IgH gene in B-cells (29,30). The vector pVHCμRepPμ was linearized at the unique NotI site and transferred by electroporation into the igm10 hybridoma (16). Southern blot and PCR analysis of genomic DNA from 84 individual G418R transformants identified five cell lines that contained a single copy of the integrated vector (Figure 8), designated VHCμRepPμ-1, VHCμRepPμ-5, VHCμRep Pμ-7, VHCμRepPμ-15 and VHCμRepPμ-26 (data not shown). Next, the amount of μ-specific RNA in the Pμ-deleted cell lines was compared to those with an intact promoter region (the VHCμRep cell lines reported in Table 1) (Figure 9). As shown in Table 2, the μ RNA levels in the VHCμRep cell lines ranged between 21 and 183% of that in Im/RCμ2A2. In comparison, the μ RNA in the VHCμRepPμ cell lines was between 6 and 20% of that in Im/RCμ2A2 (Table 2).
Figure 8. Construction of μ gene promoter (Pμ)-deleted cell lines. The structure of the linearized vector, pVHCμRep, is shown. The vector is identical to pVHCμRep (Figure 3B) with the exception of a 154 bp XbaI/NcoI deletion encompassing the μ gene promoter, Pμ, as explained in the text. The diagnostic fragment sizes generated following digestion with EcoRI and SpeI, and hybridization with probe F (Figure 1) and probe Eμ (Figure 3), are presented. Southern blotting using both single cutters and noncutters was used to determine the copy number of the integrated vector (data not shown). N, NotI; Sp, SpeI; other abbreviations are the same as in Figure 1. This figure is not drawn to scale.
Figure 9. RNA dot blot analysis of μ RNA. Serial 1:4 dilutions of total RNA were applied to a nitrocellulose filter starting with 10 μg as described (22). Relative levels of μ RNA were determined by hybridization with 32P-labeled Cμ-specific Probe F (Figure 1). To verify loading, the blot was reprobed with a 32P-labeled 2.0 kb PstI fragment containing the chicken ?-actin cDNA (27). The cell line igm10 is an Sp6-derived mutant that lacks the endogenous chromosomal μ gene (16).
Each of the VHCμRepPμ transformants was expanded from a single cell and grown for 25 generations under G418 selection. Genomic DNA was prepared from each culture, and qPCR utilizing primer pair 1/3 was performed to measure the frequency of GC in the 5' Cμ region. This analysis revealed a median GC frequency of 1.7 x 10–3 recombinants/cell (Table 3). This value is not significantly different from the median value measured in the VHCμRep transformants (1.5 x 10–3 recombinants/cell) (Table 1) (Mann–Whitney test, P = 0.46). Importantly, no significant correlation exists between μ RNA levels and GC frequencies in the VHCμRep and VHCμRepPμ transformants (Pearson's correlation, P = 0.36).
Table 3. Frequency of gene conversion at ectopic genomic sites
Interestingly, the GC frequency in transformant VHCμRepPμ-5 was 5.3% (Table 3), a value that is 761-fold higher than the median frequency measured in the R/CμRepeat transformants (Table 1). To determine whether this exceptionally high frequency resulted from an early recombination event during expansion of the culture (i.e. a ‘jackpot’), three subclone cultures of VHCμRepPμ-5, each started from a single cell, were grown for 25 generations and the frequency of GC in the 5' Cμ region was determined by qPCR. As shown in Table 3, the frequency of GC in the subclone cultures ranged from 1.9 to 10%, suggesting that the vector may have fortuitously integrated into an unusually recombinogenic region of the genome.
DISCUSSION
Previously, we showed that the IgH μ locus behaved as a hotspot for spontaneous, mitotic intrachromosomal GC between direct repeats of the Cμ region in murine hybridoma cells (14). Here, we report the identification of a 9.1 kb segment of the μ locus regulatory region that confers GC hotspot activity in this system: deleting RES from the endogenous IgH locus abolished GC between the Cμ repeats, while a stimulation of GC was observed when RES was appended to Cμ repeats in ectopic positions in the hybridoma genome. The RES encompasses DNA sequences beginning 5' of the VH region and extending to immediately 5' of the first Cμ region exon. Several regulatory elements that function in the assembly, expression and replication of the IgH μ gene reside within this fragment. This includes the μ gene promoter (Pμ), the major intronic μ gene enhancer (Eμ) (23) and the μ gene switch (Sμ) region (31). Also, associated with Eμ are two matrix attachment regions (MARs) (32), a promoter of sterile transcripts (Iμ) (23,33) and a putative origin of replication (34).
It is noteworthy that RES-stimulated GC between Cμ repeats in ectopic genomic positions is only 20–30% of the mean value measured at the endogenous μ locus. As the RES fragment used in the ectopic studies contained an 2.8 kb deletion encompassing most of the Sμ region, this might suggest a role for Sμ sequences in RES activity. The role of Sμ in class switch recombination (CSR) is well documented (31). In addition, Sμ is frequently involved in translocations observed in B-cell lymphomas (35), and is a preferred site for insertion of transfected DNA (36). However, the failure to observe a reduction in the frequency of GC between Cμ repeats positioned at the IgH locus, which contained the same 2.8 kb Sμ deletion, argues against a role for Sμ in RES activity.
Rather, the reduced RES activity at ectopic sites might suggest that DNA sequences or elements at or near the IgH locus, which are not included within the RES fragment, are necessary to generate the optimal environment. A possible candidate is the 3' enhancer (E) complex that resides 200 kb 3' of the Cμ region (37). E has previously been shown to be essential for CSR (38), and is believed to function with Eμ to maintain high levels of expression of the IgH locus in fully differentiated B-cells (39). A second possibility is that elements within the RES itself might generate a domain that is conducive to HR at the endogenous μ locus, but at ectopic sites, cannot fully recreate this environment due to the influence of neighboring DNA sequences or chromatin structure at the site of chromosomal integration. That chromatin structure can influence spontaneous HR has been suggested previously (11,13).
In the IgH μ locus, Eμ and the flanking MARs generate a domain of chromatin accessibility, as suggested by increased sensitivity to DNA-damaging agents (40,41). Analysis of Eμ/MARs deleted cell lines revealed a 30-fold decrease in the frequency of GC at the IgH μ locus, indicating that the Eμ/MARs segment is required for RES activity. However, by itself, the Eμ/MARs fragment did not have a stimulatory effect on adjacent Cμ repeats when integrated at ectopic genomic sites. This finding might suggest that other elements present in the larger RES segment, but not included within the smaller Eμ/MARs fragment, are necessary for GC-stimulating activity; or that the context of the Eμ/MARs elements, with respect to the adjacent nucleotide sequences, is important for its function.
Transcription has previously been shown to stimulate mitotic HR in mammalian cells (42–44). In comparison to the hybridomas with an intact regulatory region, a substantial reduction in μ-specific RNA levels was observed in the Eμ-deleted cell lines. This raised the possibility that the Eμ/MARs complex is indirectly related to RES activity through its role as a transcriptional activator. In contrast to our results, several B-cell lines that lack both Eμ and the MARs at the endogenous IgH locus, have been shown to produce nearly normal levels of Ig mRNA or protein (45,46). This is generally ascribed to the presence of functionally redundant elements, possibly the 3' enhancers (39). The apparent discrepancy, however, can be resolved by the finding that introduction of a gpt cassette 3' of the endogenous IgH μ gene renders expression dependent on the Eμ/MARs elements (28). In that study, deletion of Eμ and the MARs depressed μ expression to 2% the normal level. Our findings suggest that the IgH locus is similarly insulated from the effects of any redundant elements, quite possibly by the integrated pSV2neo vector sequences located downstream of the 5' Cμ region.
To directly determine whether GC between the Cμ repeats correlated with the high level of IgH μ gene transcription, Pμ was crippled in the otherwise intact RES fragment by deleting two highly conserved elements, the TATA box and an octamer motif. The mean μ RNA level in the Pμ-deleted transformants was reduced 80% compared to transformants bearing the intact RES fragment, a result which agrees with previous studies showing that mutation of either motif drastically reduces μ transcription (29,30). More importantly however, the reduction in μ RNA levels was not associated with a corresponding decrease in the frequency of GC promoted by the RES. The lack of any correlation between the GC frequency and μ RNA levels suggests that transcriptional activity is unlikely to be responsible for RES-stimulated recombination in this system. However, since a residual level of μ RNA persisted in the stable transformants, possibly as a result of sterile transcripts initiating at Iμ (33), we cannot discount the possibility that a low level of transcription may still account for the RES activity.
Assuming the Eμ/MARs are the important elements, what role do they play in GC hotspot activity at the IgH μ locus? The simplest hypothesis is that enhancer-mediated chromatin changes within the Cμ repeats increases their susceptibility to nicks or breaks, which invoke HR for repair. This notion is consistent with previous results that suggest GC events initiate at random sites across the entire Cμ region (14). Also, a high frequency of mutation in the chromosomal Ig μ and genes has been reported for hybridoma and myeloma cell lines (15,47). It would be interesting to know whether the high rate of mutation is also dependent on the enhancer elements.
Alternatively, chromatin remodeling mediated by the Eμ/MARs elements might facilitate the adoption of a non-B-DNA conformation within the Cμ repeats. Regions of alternating purines and pyrimidines readily form left-handed Z-DNA when contained in negatively supercoiled DNA (48). Interestingly, a potential Z-DNA forming run of 28 repeats of the dinucleotide, GT, is present within the Cμ region of homology (49). Smaller regions of alternating purines and pyrimidines are also present throughout the repeated Cμ regions (49). Sequences capable of forming Z-DNA have previously been shown to stimulate mitotic HR (5–7).
In summary, we have engineered various mouse hybridoma cell lines to investigate the importance of IgH μ gene regulatory elements in promoting GC in a recombination reporter consisting of a pair of homologous, repeated Cμ region segments. Our results suggest that a region we refer to as the RES, probably, through a main contribution from the internal Eμ/MARs fragment, promotes GC hotspot activity between the Cμ repeats at the endogenous IgH μ locus. While the exact mechanism is unknown, it is possible that Eμ/MARs exerts its effect through changes in chromatin accessibility or chromatin remodeling rendering the adjacent Cμ repeats more susceptible to DNA breaks, which invoke GC for repair. One could imagine that in the absence of proper repair by HR, a similar mechanism acting at the endogenous IgH μ locus in normal B cells might promote genomic instability, e.g. by stimulating chromosomal translocations into the IgH μ locus (35). While our results provide support for cis-acting regulatory elements promoting GC at the IgH μ locus, the analysis of recombination in cell line, VHCμRepPμ-5, suggests that GC hotspot activity can also be a feature of other loci in the mammalian genome. Finally, it is possible that the high frequency, Eμ/MAR-stimulated GC that we observe in this model system bears some mechanistic relationship to DNA transactions that are required to generate antibody diversity in normal B cells (50). In this regard, it is worth noting that the high frequency of GC that we observe in our system is known to be an important mechanism of antibody diversification in chickens and rabbits (51,52), and there is evidence that GC might play a role in some murine antibody responses as well (53–55).
ACKNOWLEDGEMENTS
This research was supported by a PhD studentship from the Canadian Institutes of Health Research (CIHR) to S.J.R., and a CIHR operating grant to M.D.B.
REFERENCES
Aguilera,A. ( (2002) ) The connection between transcription and genomic instability. EMBO J., , 21, , 195–201.
Kuzminov,A. ( (2001) ) DNA replication meets genetic exchange: chromosomal damage and its repair by homologous recombination. Proc. Natl Acad. Sci. USA, , 98, , 8461–8468.
Voelkel-Meiman,K., Keil,R.L. and Roeder,G.S. ( (1987) ) Recombination-stimulating sequences in yeast ribosomal DNA correspond to sequences regulating transcription by RNA polymerase I. Cell, , 48, , 1071–1079.
Stewart,S.E. and Roeder,G.S. ( (1989) ) Transcription by RNA polymerase I stimulates mitotic recombination in Saccharomyces cerevisiae. Mol. Cell. Biol., , 9, , 3464–3472.
Stringer,J.R. ( (1985) ) Recombination between poly sequences in simian virus 40-infected cultured cells. Mol. Cell. Biol., , 5, , 1247–1259.
Bullock,P., Miller,J. and Botchan,M. ( (1986) ) Effects of poly and poly repeats on homologous recombination in somatic cells. Mol. Cell. Biol., , 6, , 3948–3953.
Wahls,W.P., Wallace,L.J. and Moore,P.D. ( (1990) ) The Z-DNA motif d(TG)30 promotes reception of information during gene conversion events while stimulating homologous recombination in human cells in culture. Mol. Cell. Biol., , 10, , 785–793.
Lenzmeier,B.A. and Freudenreich,C.H. ( (2003) ) Trinucleotide repeat instability: a hairpin curve at the crossroads of replication, recombination, and repair. Cytogenet. Genome Res., , 100, , 7–24.
Gordenin,D.A., Lobachev,K.S., Degtyareva,N.P., Malkova,A.L., Perkins,E. and Resnick,M.A. ( (1993) ) Inverted DNA repeats: a source of eukaryotic genomic instability. Mol. Cell. Biol., , 13, , 5315–5322.
Farah,J.A., Hartsuiker,E., Mizuno,K., Ohta,K. and Smith,G.R. ( (2002) ) A 160-bp palindrome is a Rad50·Rad32-dependent mitotic recombination hotspot in Schizosaccharomyces pombe. Genetics, , 161, , 461–468.
Malagón,F. and Aguilera,A. ( (1996) ) Differential intrachromosomal hyper-recombination phenotype of spt4 and spt6 mutants of S. cerevisiae. Curr. Genet., , 30, , 101–106.
Fritze,C.E., Verschueren,K., Strich,R. and Easton Esposito,R. ( (1997) ) Direct evidence for SIR2 modulation of chromatin structure in yeast rDNA. EMBO J., , 16, , 6495–6509.
Gottlieb,S. and Esposito,R.E. ( (1989) ) A new role for a yeast transcriptional silencer gene, SIR2, in regulation of recombination in ribosomal DNA. Cell, , 56, , 771–776.
Raynard,S.J., Read,L.R. and Baker,M.D. ( (2002) ) Evidence for the murine IgH μ locus acting as a hot spot for intrachromosomal homologous recombination. J. Immunol., , 168, , 2332–2339.
K?hler,G. and Shulman,M.J. ( (1980) ) Immunoglobulin M mutants. Eur. J. Immunol., , 10, , 467–476.
K?hler,G., Potash,M.J., Lehrach,H. and Shulman,M.J. ( (1982) ) Deletions in immunoglobulin μ chains. EMBO J., , 1, , 555–563.
Baker,M.D., Pennell,N., Bosnoyan,L. and Shulman,M.J. ( (1988) ) Homologous recombination can restore normal immunoglobulin M production in a mutant hybridoma cell line. Proc. Natl Acad. Sci. USA, , 85, , 6432–6436.
Baker,M.D. and Read,L.R. ( (1992) ) Ectopic recombination within homologous immunoglobulin μ gene constant regions in a mouse hybridoma cell line. Mol. Cell. Biol., , 12, , 4422–4432.
Ng,P. and Baker,M.D. ( (1999) ) Mechanisms of double-strand-break repair during gene targeting in mammalian cells. Genetics, , 151, , 1127–1141.
Gross-Bellard,M., Qudet,P. and Chambon,P. ( (1973) ) Isolation of high-molecular weight DNA from mammalian cells. Eur. J. Biochem., , 36, , 32–38.
Sambrook,J., Fritsch,E.F. and Maniatis,T. ( (1989) ) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Baumann,B., Potash,M.J. and K?hler,G. ( (1985) ) Consequences of frameshift mutations at the immunoglobulin heavy chain locus of the mouse. EMBO J., , 4, , 351–359.
Staudt,L.M. and Lenardo,M.J. ( (1991) ) Immunoglobulin gene transcription. Annu. Rev. Immunol., , 9, , 373–398.
Ochi,A., Hawley,R.G., Hawley,T., Shulman,M.J., Traunecker,A., K?hler,G. and Hozumi,N. ( (1983) ) Functional immunoglobulin M production after transfection of cloned immunoglobulin heavy and light chain genes into lymphoid cells. Proc. Natl Acad. Sci. USA, , 80, , 6351–6355.
Ng,P. and Baker,M.D. ( (1998) ) High efficiency site-specific modification of the chromosomal immunoglobulin locus by gene targeting. J. Immunol. Methods, , 214, , 81–96.
Serwe,M. and Sablitzky,F. ( (1993) ) V(D)J recombination in B cells is impaired but not blocked by targeted deletion of the immunoglobulin heavy chain intron enhancer. EMBO J., , 12, , 2321–2327.
Cleveland,D.W., Lopata,M.A., MacDonald,R.J., Cowan,N.J., Rutter,W.J. and Kirschner,M.W. ( (1980) ) Number and evolutionary conservation of - and ?-tubulin and cytoplasmic ?- and -actin genes using specific cloned cDNA probes. Cell, , 20, , 95–105.
Oancea,A.E., Berru,M. and Shulman,M.J. ( (1997) ) Expression of the (recombinant) endogenous immunoglobulin heavy-chain locus requires the intronic matrix attachment regions. Mol. Cell. Biol., , 17, , 2658–2668.
Mason,J.O., Williams,G.T. and Neuberger,M.S. ( (1985) ) Transcription cell type specificity is conferred by an immunoglobulin VH gene promoter that includes a functional consensus sequence. Cell, , 41, , 479–487.
Eaton,S. and Calame,K. ( (1987) ) Multiple DNA sequence elements are necessary for the function of an immunoglobulin heavy chain promoter. Proc. Natl Acad. Sci. USA, , 84, , 7634–7638.
Manis,J.P., Tian,M. and Alt,F.W. ( (2002) ) Mechanism and control of class-switch recombination. Trends Immunol., , 23, , 31–39.
Cockerill,P.N. ( (1990) ) Nuclear matrix attachment occurs in several regions of the IgH locus. Nucleic Acids Res., , 18, , 2643–2648.
Su,L.K. and Kadesch,T. ( (1990) ) The immunoglobulin heavy-chain enhancer functions as the promoter for Iμ sterile transcription. Mol. Cell. Biol., , 10, , 2619–2624.
Ariizumi,K., Wang,Z. and Tucker,P.W. ( (1993) ) Immunoglobulin heavy chain enhancer is located near or in an initiation zone of chromosomal DNA replication. Proc. Natl Acad. Sci. USA, , 90, , 3695–3699.
Bergsagel,P.L., Chesi,M., Nardini,E., Brents,L.A., Kirby,S.L. and Kuehl,W.M. ( (1996) ) Promiscuous translocations into immunoglobulin heavy chain switch regions in multiple myeloma. Proc. Natl Acad. Sci. USA, , 93, , 13931–13936.
Baar,J. and Shulman,M.J. ( (1995) ) The Ig heavy chain switch region is a hotspot for insertion of transfected DNA. J. Immunol., , 155, , 1911–1920.
Pettersson,S., Cook,G.P., Bruggemann,M., Williams,G.T. and Neuberger,M.S. ( (1990) ) A second B cell-specific enhancer 3' of the immunoglobulin heavy-chain locus. Nature, , 344, , 165–168.
Cogné,M., Lansford,R., Bottaro,A., Zhang,J., Gorman,J., Young,F., Cheng,H.L. and Alt,F.W. ( (1994) ) A class switch control region at the 3' end of the immunoglobulin heavy chain locus. Cell, , 77, , 737–747.
Lieberson,R., Ong,J., Shi,X. and Eckhardt,L.A. ( (1995) ) Immunoglobulin gene transcription ceases upon deletion of a distant enhancer. EMBO J., , 14, , 6229–6238.
Jenuwein,T., Forrester,W.C., Fernandez-Herrero,L.A., Laible,G., Dull,M. and Grosschedl,R. ( (1997) ) Extension of chromatin accessibility of nuclear matrix attachment regions. Nature, , 385, , 269–272.
Nikolajczyk,B.S., Sanchez,J.A. and Sen,R. ( (1999) ) ETS protein-dependent accessibility changes at the immunoglobulin μ heavy chain enhancer. Immunity, , 11, , 11–20.
Nickoloff,J.A. and Reynolds,R.J. ( (1990) ) Transcription stimulates homologous recombination in mammalian cells. Mol. Cell. Biol., , 10, , 4837–4845.
Nickoloff,J.A. ( (1992) ) Transcription enhances intrachromosomal homologous recombination in mammalian cells. Mol. Cell. Biol., , 12, , 5311–5318.
Thyagarajan,B., Johnson,B.L. and Campbell,C. ( (1995) ) The effect of target site transcription on gene targeting in human cells in vitro. Nucleic Acids Res., , 23, , 2784–2790.
Wabl,S. and Burrows,P.D. ( (1984) ) Expression of immunoglobulin heavy chain at a high level in the absence of a proposed immunoglobulin enhancer element in cis. Proc. Natl Acad. Sci. USA, , 81, , 2452–2455.
Zaller,D.M. and Eckhardt,L.A. ( (1985) ) Deletion of a B-cell-specific enhancer affects transfected, but not endogenous, immunoglobulin heavy-chain gene expression. Proc. Natl Acad. Sci. USA, , 82, , 5088–5092.
Yu,H. and Eckhardt,L.A. ( (1986) ) DNA rearrangement causes a high rate of spontaneous mutation at the immunoglobulin heavy-chain locus of a mouse myeloma cell line. Mol. Cell. Biol., , 6, , 4228–4235.
Nordheim,A., Peck,L.J., Lafer,E.M., Stollar,B.D., Wang,J.C. and Rich,A. ( (1983) ) Supercoiling and left-handed Z-DNA. Cold Spring Harb. Symp. Quant. Biol., , 47, , 93–100.
Richards,J.E., Gilliam,A.C., Shen,A., Tucker,P.W. and Blattner,F.R. ( (1983) ) Unusual sequences in the murine immunoglobulin μ- heavy-chain region. Nature, , 306, , 483–487.
Alt,F.W., Blackwell,T.K. and Yancopoulos,G.D. ( (1987) ) Development of the primary antibody repertoire. Science, , 238, , 1079–1087.
Reynaud,C.A., Dahan,A., Anquez,V. and Weill,J.C. ( (1989) ) Somatic hyperconversion diversifies the single VH gene of the chicken with a high incidence in the D region. Cell, , 59, , 171–183.
Becker,R.S. and Knight,K.L. ( (1990) ) Somatic diversification of immunoglobulin heavy chain VDJ genes: evidence for somatic gene conversion in rabbits. Cell, , 63, , 987–997.
David,V., Folk,N.L. and Maizels,N. ( (1992) ) Germ line variable regions that match hypermutated sequences in genes encoding murine anti-hapten antibodies. Genetics, , 132, , 799–811.
Xu,B. and Selsing,E. ( (1994) ) Analysis of sequence transfers resembling gene conversion in a mouse antibody transgene. Science, , 265, , 1590–1593.
Selsing,E., Xu,B. and Sigurdardottir,D. ( (1996) ) Gene conversion and homologous recombination in murine B cells. Semin. Immunol., , 8, , 51–158.(Steven J. Raynard and Mark D. Baker*)
* To whom correspondence should be addressed. Tel: +1 519 824 4120; Ext. 54788; Fax: +1 519 837 2075; Email: mdbaker@uoguelph.ca
ABSTRACT
In mammalian cells, little is known about the nature of recombination-prone regions of the genome. Previously, we reported that the immunoglobulin heavy chain (IgH) μ locus behaved as a hotspot for mitotic, intrachromosomal gene conversion (GC) between repeated μ constant (Cμ) regions in mouse hybridoma cells. To investigate whether elements within the μ gene regulatory region were required for hotspot activity, gene targeting was used to delete a 9.1 kb segment encompassing the μ gene promoter (Pμ), enhancer (Eμ) and switch region (Sμ) from the locus. In these cell lines, GC between the Cμ repeats was significantly reduced, indicating that this ‘recombination-enhancing sequence’ (RES) is necessary for GC hotspot activity at the IgH locus. Importantly, the RES fragment stimulated GC when appended to the same Cμ repeats integrated at ectopic genomic sites. We also show that deletion of Eμ and flanking matrix attachment regions (MARs) from the RES abolishes GC hotspot activity at the IgH locus. However, no stimulation of ectopic GC was observed with the Eμ/MARs fragment alone. Finally, we provide evidence that no correlation exists between the level of transcription and GC promoted by the RES. We suggest a model whereby Eμ/MARS enhances mitotic GC at the endogenous IgH μ locus by effecting chromatin modifications in adjacent DNA.
INTRODUCTION
Homologous recombination (HR) is the process by which genetic information is exchanged between two similar or identical DNA duplexes. In eukaryotes, recombination plays an important role in generating genetic diversity in meiosis, and in mitotic cells, represents a pathway for the repair of DNA damage. Nevertheless, these benefits are associated with the risk of aberrant rearrangements due to recombination between the myriad of repeated sequences in the eukaryotic genome. Germane to this risk is the knowledge that the frequency of recombination is not uniform, in that some loci exhibit relatively high frequencies of recombination (hotspots), while others exhibit relatively low frequencies of recombination (coldspots). This fact has important implications for the maintenance of genomic stability.
Relatively little is known about the nature of mitotic recombination hotspots in mammalian cells. However, since the primary role of HR in somatic cells is as a DNA repair mechanism, it stands to reason that regions of the genome that are prone to DNA damage would be active in HR. Current evidence suggests that errors that occur in the course of normal DNA metabolism, such as transcription (1), and replication (2), are a major source of recombinogenic lesions. It follows then, that DNA regulatory elements could act as recombination hotspots as an indirect by-product of their normal biological function. For example, the yeast recombination hotspot, HOT1, contains an initiation site and enhancer of transcription by RNA polymerase I (3). The stimulatory activity of HOT1 correlates with its capacity to promote transcription through the recombining sequences (4).
Other DNA motifs are thought to stimulate genomic rearrangement through their effects on the secondary structure of DNA. Regions of alternating purines and pyrimidines that can adopt the Z-DNA conformation (5–7), trinucleotide repeats (8) and inverted (palindromic) repeats that can extrude and form hairpin or cruciform structures (9,10) are all prone to rearrangement in eukaryotic cells.
Indirect evidence suggests higher order chromatin structure might also influence HR rates. It is plausible that a change in chromatin structure facilitates the access of recombination proteins, or possibly, leads to hypersensitivity to DNA-damaging agents. Hyper-recombination phenotypes were reported for certain yeast mutants defective in proteins involved in chromatin-mediated repression of transcription (11). A correlation between Sir2-mediated DNA silencing and a more ‘closed’ chromatin structure was shown by Fritze et al. (12). Since DNA silencing also correlates with reduced recombination (13), it was suggested that ‘closed’ chromatin has a double effect in repressing both transcription and recombination.
Our laboratory previously reported that the mouse immunoglobulin heavy-chain (IgH) μ locus acts as a hotspot for spontaneous mitotic gene conversion (GC) (14). The assay system monitors intrachromosomal GC events between closely linked direct repeats of the IgH μ gene constant (Cμ) region. In this paper, we report the identification of a 9.1 kb segment of DNA encompassing the IgH μ gene regulatory region, which stimulates GC between adjacent Cμ repeats both at the endogenous IgH locus, and when appended to the same Cμ repeats stably integrated at ectopic genomic sites. We also show that deletion of the IgH major intronic enhancer (Eμ) and flanking matrix attachment regions (MARs) from the regulatory region abolishes GC hotspot activity at the chromosomal IgH μ locus. However, the Eμ/MARs fragment by itself does not stimulate recombination at ectopic genomic sites. Finally, we provide evidence that there is no correlation between the level of transcription and the level of GC promoted by the IgH μ gene regulatory region. We suggest a model whereby the endogenous Eμ/MARs effects chromatin modifications in adjacent chromosomal regions that enhance spontaneous mitotic GC.
MATERIALS AND METHODS
Hybridoma cell lines
The hybridoma cell line, Sp6/HL, bears a single copy of the trinitrophenyl (TNP)-specific chromosomal IgH μ chain gene and makes normal, cytolytic TNP-specific IgM (-chain) (15,16). The following Sp6/HL-derived hybridoma cell lines were used in this study. Hybridoma igm10 is a mutant that has lost the TNP-specific chromosomal μ gene (16). Hybridoma Im/RCμ2A2 was constructed by gene targeting in the Cμ region of the haploid, TNP-specific chromosomal μ gene, as described previously (17). The conditions for hybridoma cell growth are described elsewhere (15,16).
DNA transfer
Linear plasmid vector DNA (8.7 pmol) was introduced into 2 x 107 hybridoma cells by electroporation as described previously (17). Individual transformants were recovered by limited dilution cloning procedures as described in (18,19).
Quantitative PCR (qPCR) assay
The conditions used for qPCR and details of the construction of the vector control have been reported previously (14).
DNA analysis
Preparation of genomic DNA was performed by the method of Gross-Bellard et al. (20) with the exception of DNA samples used in the qPCR assay, which were prepared with the QIAamp DNA mini kit (QIAGEN). For Southern analysis, restriction enzymes were purchased from New England Biolabs Inc. (Mississauga, ON, Canada) and used in accordance with the manufacturer's specification. Gel electrophoresis, transfer of DNA onto nitrocellulose membrane, 32P-labeled probe preparation and hybridization were all performed according to standard procedures (21). Primers used in this study were synthesized at Sigma (Oakville, ON, Canada), and their sequences have been reported previously (14). New primers used in this study are as follows: 23819, 5'-CAT CCT CCT CCT CAT CAT CGT CAT-3'; CμXbaRI, 5'-GAA TCT GTC TTC TTG CCT CCT GTC-3'; 24226, 5'-GCT GTG TAG AAG TAC TCG CCG ATA-3'.
RNA analysis
Hybridoma cells were grown to a density of 3–5 x 105 cells/ml. Total RNA was isolated using Trizol (Gibco) in accordance with the manufacturer's specification, and assayed by dot blotting for hybridization to a Cμ fragment (Probe F) and a ?-actin probe. The procedures of Baumann et al. (22) were adopted for RNA dot blotting, while hybridizations were performed according to standard methods (21).
RESULTS
Role of the IgH μ gene regulatory region in spontaneous HR hotspot activity
In this study, we investigated whether cis-acting DNA elements within the IgH μ gene regulatory region promote intrachromosomal GC. The haploid IgH μ locus in the hybridoma cell line, Im/RCμ2A2, bears a pair of Cμ region heteroalleles that exhibit high-frequency intrachromosomal GC (14). The upstream (5') Cμ region is derived from the wild-type Sp6/HL hybridoma and resides in its normal position to be expressed from the TNP-specific VH region. Vector pSV2neo sequences separate the 5' wild-type Cμ region from the downstream (3') Cμ region, which bears a 2 bp deletion in exon Cμ3 (referred to as the mutant igm482 Cμ3 region) (15,16).
The 9.6 kb omega ()-form vector, pVHCμKO, was used in the targeted modification of the Im/RCμ2A2 μ gene (Figure 1A). The pVHCμKO backbone consists of an enhancer-trap derivative of pSV2hyg in which the 372 bp NsiI/NdeI fragment encompassing the simian virus 40 (SV40) early region enhancer residing upstream of the hygromycin phosphotransferase (hyg) gene was deleted. Although not relevant to this study, the Herpes Simplex Virus-1 (HSV-1) thymidine kinase (tk) gene is also present. In pVHCμKO, the flanking arms consist of a 2.3 kb EcoRI/HpaI segment with homology 5' of the μ gene promoter (Pμ) and a 1.6 kb XbaI/MfeI Cμ region fragment, positioned to the left and right of pSV2hyg, respectively. Flanking arm alignment with isogenic regions of the Im/RCμ2A2 chromosomal μ gene is indicated by the dashed lines in Figure 1A.
Figure 1. Construction of 2A2VH cell lines. (A) The structure of the haploid, chromosomal μ locus in the recipient murine hybridoma cell line, Im/RCμ2A2, is presented along with the 9.7 kb replacement vector, pVHCμKO. Regions of alignment between the homologous arms of the vector and Im/RCμ2A2 chromosomal μ locus are indicated by the dashed lines. (B) The structure of the recombinant μ locus following targeted gene replacement by the vector pVHCμKO. In both panels, the mutant igm482 Cμ3 exon is indicated by an open box, while wild-type Cμ exons are indicated by filled boxes. Although not labeled, matrix attachment regions (MARs) are indicated by filled diamonds. The diagnostic fragment sizes generated following digestion with the indicated restriction enzymes are presented. DNA probe fragment F is an 870 bp XbaI/BamHI fragment while probe tk is a 1.3 kb fragment from the HSV-1 thymidine kinase (tk) gene. VH, TNP-specific μ-heavy-chain variable region; Pμ, μ-gene promoter; Eμ, major intronic enhancer; Sμ, μ-gene switch region; Cμ, μ-gene constant region; neo, neomycin phosphotransferase gene; tk, HSV-1 thymidine kinase gene; hyg, hygromycin resistance gene; B, BamHI; E, EcoRI; H, HpaI; M, MfeI; Sc, ScaI; X, XbaI. The figure is not drawn to scale.
Targeted gene replacement by pVHCμKO deletes all coding and regulatory DNA sequences required for expression of the TNP-specific μ heavy chain gene in Im/RCμ2A2 (Figure 1B) (23). Thus, following electroporation, culture supernatant from a total of 270 hygR transformants was tested for the presence of TNP-specific IgM, by complement-dependent lysis of TNP-coupled sheep erythrocytes in spot tests (15). The structure of the IgH μ locus in six IgM–, hygR transformants was examined by Southern analysis to identify those in which targeted gene replacement had occurred. The fragment sizes of the recipient μ locus in Im/RCμ2A2 (Figure 1A), and of correctly targeted transformants (Figure 1B), following digestion with BamHI and hybridization with Cμ-specific probe F are shown. Two independent hybridoma cell lines, 2A2VH-82 and 2A2VH-236 were identified as having the correctly targeted structure (data not shown).
For determination of GC frequencies, a quantitative PCR (qPCR) assay was used, as described previously (14). In brief, the assay makes use of four primers. Primers 1 and 2 bind upstream of the 5' and 3' Cμ regions, respectively. Primers 3 and 4 are specific for the wild-type and mutant igm482 Cμ3 regions, respectively. Two related hybridomas were used to standardize the assay, Im/RCμ3/1-7 (abbreviated 3/1-7) and Im/RCμD7-7 (abbreviated D7-7) (Figure 2A). In hybridoma cell line, 3/l-7, both the 5' and 3' Cμ regions are wild-type (designated, double wild-type), whereas, in hybridoma cell line, D7-7 they are both mutant igm482 Cμ3 regions (designated double mutant). Figure 2A indicates the primer binding sites and product sizes generated by PCR amplification of 3/1-7 and D7-7. To circumvent the problem of variability in efficiency of amplification, 10–9 pmol of a heterologous vector control, sharing only the primer binding sites with the target sequence, was included in each PCR reaction. Amplification of the vector generates a 1.3 kb product (Figure 2A). Genomic mixtures of 3/1-7 and D7-7 were prepared in ratios of 1:50, 1:200, 1:500 and 1:1000, and amplified using primer pair 2/3 for generation of the wild-type, 1.9 kb 3' Cμ region product (Figure 2B, lanes 2–5). Following gel electrophoresis, quantitation of the EtBr staining in the 1.9 kb target band divided by that in the 1.3 kb vector control band yields a ratio that, when plotted against the wild-type Cμ region copy number, generates a standard curve (Figure 2B, inset). In a similar manner, the qPCR assay was standardized with the primer pairs 1/3, 1/4 and 2/4 to detect the wild-type 5' Cμ, mutant 5' Cμ and mutant 3' Cμ regions, respectively.
Figure 2. qPCR assay to measure gene conversion between the Cμ region repeats. (A) Hybridoma 3/1-7 has the wild-type sequences in both the 5' and 3' Cμ regions (double wild-type), while hybridoma D7-7 bears the mutant igm482 Cμ3 exon in both the 5' and 3' Cμ regions (double mutant). The PCR primers and the sizes of the amplified products they produce are indicated. Primer 1 binds upstream of the 5' Cμ region and outside the region of homology in the 3' Cμ region. Primer 2 binds within pSV2neo vector sequences upstream of the 3' Cμ region. Primers 3 and 4 bind specifically to the complementary strand of the wild-type and mutant igm482 Cμ3 exons, respectively. The vector control contains the primer 1 and 2 binding sites and primer 3 and 4 binding sites on opposite strands flanking a segment of the thymidine kinase (tk) gene. (B) Lanes 2–5, standard solutions of genomic DNA consisting of 3/1-7 and D7-7 prepared in ratios of 1:50, 1:200, 1:500, 1:1000 along with a constant amount of vector control (10–9 pmol) were amplified using primer pair 2/3 for generation of the 1.9-kb wild-type 3' Cμ region product and 1.3-kb vector control product. DNA from the double mutant, D7-7 cell line, was included as a negative control (lane 6). Quantitation of the EtBr staining in the target band divided by that in the vector control band yielded the standard curve (inset). Lanes 7–11, representative samples from 2A2VH-82 subclone 3 subculture wells 1–5, amplified with primer pair 2/3 to detect gene conversion in the 3' Cμ region to the wild-type sequence. Positive signals are present in subcultures 2 and 4. Abbreviations are the same as in Figure 1. The figure is not drawn to scale.
To quantify the GC frequency in the cell lines, replicate genomic DNA samples, along with a constant amount of vector control, were amplified using primer pair 2/3. Following gel electrophoresis, and quantitation of the EtBr staining, the resulting target:vector ratio was used to determine the GC frequency from the standard curve. In cases where the GC frequency was below the sensitivity of the assay (<0.001), the culture was distributed at a density of 500 cells/well in 24-well plates. Following cell growth, genomic DNA was prepared from each culture well and qPCR analysis was performed on the separate DNA preparations using primer pair 2/3. As an example, Figure 2B (lanes 7–11) presents five representative genomic DNA samples for hybridoma 2A2VH-82 subclone 3. From the fraction of negative wells in the qPCR assay (as examples, lanes numbered 1, 3 and 5 in Figure 2B) and the Poisson distribution, the mean GC frequency between the Cμ repeats was calculated.
Similar GC frequencies were detected previously for the 5' and 3' Cμ regions in Im/RCμ2A2 (14), and therefore, in cell lines, 2A2VH-82 and 2A2VH-236, the frequency of GC of the recipient, 3' mutant Cμ region by the donor, 5' wild-type Cμ region was determined. Three subclone cultures of 2A2VH-82 and two subclone cultures of 2A2VH-236 were each started from a single cell. Southern blot and PCR analysis revealed that the μ gene structure in each subclone was identical to that in the parent cell line (data not shown). Each subclone was grown for 25 generations in medium supplemented with G418 (0.6 mg/ml) and hygromycin (0.6 mg/ml), and then assayed by qPCR using primer pair 2/3. This analysis revealed that the mean GC frequency between the Cμ repeats in 2A2VH-82 and 2A2VH-236 was 0.2 x 10–3 recombinants/cell (Table 1). In comparison, GC between the same repeats integrated at the wild-type μ locus of hybridoma Im/RCμ2A2 was previously shown to occur at a frequency of 6.1 x 10–3 recombinants/cell (14). The procedures and conditions used for cell growth, and determination of GC frequencies were identical for both studies (14). The 31-fold difference in the GC frequencies is highly significant (t-test of log-transformed data; P = 0.0002), suggesting that the IgH μ gene regulatory region is required for GC hotspot activity.
Table 1. Frequency of gene conversion between Cμ region repeats
The IgH μ gene regulatory region stimulates intrachromosomal gene conversion between Cμ region repeats at ectopic sites in the hybridoma genome
Since deletion of the regulatory region from the endogenous IgH locus reduced the frequency of GC between adjacent Cμ repeats, it was of interest to determine whether the region stimulated GC when appended to the same Cμ repeats integrated outside the IgH locus. For these studies, we used a derivative of the vector pCμ-repeat described previously (Figure 3A) (14). It bears 4.3 kb segments encoding the mutant igm482 and wild-type Cμ regions flanking a pSV2neo vector backbone. The 10.8 kb EcoRI/MfeI fragment encompassing the VH region and the VH–Cμ intron (Figure 1A) was obtained from the cloned wild-type Sp6 μ gene (24), and inserted into pCμ-repeat in its correct position and orientation, upstream of the mutant igm482 5' Cμ region to create the vector, pVHCμRep (Figure 3B). The μ gene structure in pVHCμRep is isogenic to that in Im/RCμ2A2, with the exception of an 2.8 kb deletion in the μ gene switch (Sμ) region that occurred during cloning of the genomic DNA in Escherichia coli, leaving an 0.4 kb residual Sμ segment (24). The vector pVHCμRep was linearized at the unique NotI site and transferred by electroporation into the Sp6-derived hybridoma cell line igm10, which lacks the endogenous chromosomal μ gene (16). A total of 96 individual G418R transformants were isolated and screened by PCR analysis in conjunction with Southern blotting (Figure 3B) to determine the structure and copy number of the integrated vector. Five cell lines, designated VHCμRep-10, VHCμRep-12, VHCμRep-15, VHCμRep-19 and VHCμRep-21, were identified as having a single copy of the integrated vector with the structure shown in Figure 3B (data not shown). Each transformant was expanded from a single cell and grown for 25 generations under G418 selection. Genomic DNA was prepared from each culture, and qPCR utilizing primer pair 1/3 was performed to measure the frequency of GC in the 5' Cμ region. It should be noted that randomly integrated constructs are susceptible to position effects, which could influence GC rates. As an unusually high or low frequency of GC in one of the cell lines could have a misleading effect on the mean GC frequency (due to the low sample size), we felt that comparison of median values was more appropriate for the ectopic data. When compared to transformants containing just the Cμ repeats integrated at ectopic genomic sites (R/CμRepeat cell lines in Table 1) (14), the frequency of GC was stimulated up to as much as 30-fold in the VHCμRep transformants (Table 1) (Mann–Whitney test, P = 0.036).
Figure 3. Constructs integrated at ectopic sites in the R/CμRepeat and VHCμRep transformants. The structures of the linearized vectors transfected into hybridoma igm10, which lacks the endogenous chromosomal μ gene (16), are shown. (A) The vector pCμ-repeat bears the 4.3-kb Cμ region repeats inserted into a pSV2neo vector backbone in the tandem orientation (14). The upstream (5') Cμ region bears the 2-bp igm482 Cμ3 deletion (open box), but is otherwise isogenic with the downstream (3') Cμ region. (B) The vector pVHCμRep contains the EcoRI/MfeI VH-region fragment from the hybridoma Sp6/HL inserted into the MfeI site of pCμ-repeat in its correct position and orientation with respect to the Cμ regions. As described in the text, a residual 0.4 segment remains of the Sμ region (denoted Sμ*) (24). The diagnostic fragment sizes generated following digestion with EcoRI and SpeI, and hybridization with probe F (Figure 1) and with probe Eμ, a 992-bp XbaI fragment, are presented. Southern blotting using both single cutters and noncutters was used to determine the copy number of the integrated vector (data not shown). N, NotI; Sp, SpeI; other abbreviations are the same as in Figure 1. The figure is not drawn to scale.
The μ gene switch region, Sμ, is not required for high-frequency gene conversion at the IgH locus
The frequency of GC promoted by the IgH μ gene regulatory region at ectopic sites was only 20–30% of the mean value measured in Im/RCμ2A2 (Table 1). This might suggest a role for Sμ in stimulating GC, since as indicated above, it was largely deleted from the construct used in the ectopic studies. To investigate this, the frequency of GC was determined in two related hybridomas, Emut1-Sμ+ and Emut1-Sμ–, that were isolated in a previous study (14). As shown in Figure 4, both cell lines contain Cμ repeats positioned at the IgH μ locus: the 3' Cμ region is wild type, while the 5' Cμ region is mutant as a consequence of a 4 bp insertion in Cμ exon 1. Cell line Emut1-Sμ+ bears the chromosomal Sμ region upstream of both the 5' and 3' Cμ regions, whereas Emut1-Sμ– contains the 2.8 kb Sμ deletion upstream of both Cμ regions (Figure 4). The Sμ deletion does not affect μ gene expression (24).
Figure 4. Structure of the haploid, chromosomal IgH μ gene in the Emut1-Sμ+ and Emut1-Sμ– hybridoma cell lines. In each cell line, the mutant, recipient 5' Cμ region bears a BspHI marker in Cμ exon 1 and is separated from the wild-type, donor 3' Cμ region by the integrated pSV2neo vector sequences. In hybridoma Emut1-Sμ+, the endogenous μ gene switch region (denoted Sμ) is present upstream of both the 5' and 3' Cμ regions. Whereas, hybridoma Emut1-Sμ– bears an 2.8 kb Sμ deletion (denoted Sμ*) upstream of both Cμ regions. The figure is not drawn to scale.
The frameshift mutation in the expressed 5' Cμ region causes Emut1-Sμ+ and Emut1-Sμ– to produce a truncated μ heavy chain that cannot form pentameric IgM and activate complement-dependent lysis of TNP-coupled sheep erythrocytes (14). During growth under G418-selective conditions, conversion of the recipient, 5' mutant Cμ region by the donor, 3' wild-type Cμ region can restore normal, TNP-specific IgM production in the hybridoma cells, allowing them to be detected as plaque-forming cells (PFC) in a TNP-specific plaque assay (17). Previously, we showed that there was no difference in the frequency of GC of the 4 bp mutation when inserted at different sites spanning the Cμ region (14). Furthermore, the frequency of GC of the 4 bp insertion was similar to that measured for the 2 bp deletion in Im/RCμ2A2 (14). Three subclone cultures of the Emut1-Sμ+ and Emut1-Sμ– mutant hybridoma cell lines were started from a single cell. Southern analysis of each subclone revealed the same μ gene structure in the subclones and parental cultures (data not shown). Each subclone culture was grown for 24 generations and then subjected to the TNP-specific plaque assay. As shown in Table 1, the mean frequency of generating TNP-specific PFC in each cell line was not significantly different (t-test, P = 0.30). This suggests that the Sμ region is not required for high-frequency intrachromosomal GC.
The IgH major intronic enhancer, Eμ, is required for stimulation of gene conversion by the IgHμ gene regulatory region
To determine whether the major intronic enhancer (Eμ) and/or the flanking matrix attachment regions (MARs) play a role in HR hotspot activity, a ‘hit-and-run’ gene targeting technique (25,26) was used to delete these elements from the endogenous IgH μ locus in the hybridoma, Im/RCμ2A2. The ‘hit’ step takes advantage of the 10.8 kb enhancer-trap, insertion vector pVHEμ (Figure 5A). pVHEμ contains the 8.9 kb XbaI fragment encompassing the VH region exon and VH–Cμ intron, inserted into the enhancerless pSV2hyg backbone bearing the HSV-1 tk gene. The region of homology was modified by deleting the 992 bp XbaI Eμ/MARs-containing fragment. The vector was linearized at the unique AfeI site, which provides 3.0 and 2.2 kb of homology on the 5' and 3' sides of the cut site, respectively, and transferred by electroporation into Im/RCμ2A2 (17). A total of 446 independent hygR transformants were isolated, 106 of which were screened by PCR using primers 23819 and CμXbRI for the 9.5 kb product that identifies the endogenous, chromosomal IgH region (Figure 5A). Southern analysis was performed on five cell lines from which the 9.5 kb PCR product was not amplified to identify those in which a single copy of the vector had correctly integrated at the IgH locus. The fragment sizes of the recipient μ locus of cell line Im/RCμ2A2 (Figure 5A), and of correctly targeted transformants (Figure 5B), following digestion with BamHI and hybridization with Probe N, are shown. Two hybridoma cell lines, Im/2VHRCμ2A2-17 and Im/2VHRCμ2A2-26 were identified as having the correctly targeted structure (data not shown).
Figure 5. Construction of 2A2Eμ+ and 2A2Eμ– cell lines by ‘Hit’ and ‘Run’ gene targeting at the chromosomal μ locus. (A) The ‘hit’ step involves targeted integration of the enhancer-trap insertion vector pVHEμ into the chromosomal μ gene of the hybridoma Im/RCμ2A2. pVHEμ contains a cloned segment of theSp6/HL μ gene regulatory region from which the 992 bp Eμ/MARs-containing XbaI fragment (indicated by an inverted ‘V’) was deleted. Regions of alignment between the vector and the Im/RCμ2A2 chromosomal μ locus are indicated by the dashed lines. (B) The structure of the Im/RCμ2A2 chromosomal μ gene following targeted integration of a single copy of the pVHEμ vector. (C) The ‘run’ step involves excision of the integrated vector and one copy of homologous DNA by intrachromosomal HR, an event that exploits the ability to select against the HSV-1 tk gene with gancyclovir (Ganc). Intrachromosomal reciprocal crossover that occurs in the region of homology 5' or 3' of the Eμ/MARs deletion generates the μ gene structure depicted in (D) and (E), respectively. The diagnostic fragment sizes generated following digestion with BamHI are presented. DNA probe N consists of adjacent 475 and 495 bp NheI fragments. The positions of primers 23819 and CμXbRI are indicated. Af, AfeI; other abbreviations are the same as in Figure 1. The figure is not drawn to scale.
The ‘run’ step involves excision of the integrated vector by intrachromosomal HR between the duplicated region of homology that removes the integrated vector, hyg and tk genes along with one copy of homologous DNA (Figure 5C). Duplicate subcultures of one cell line (Im/2VHRCμ2A2-17) were plated in media lacking hygromycin to allow for growth of hygS hybridomas in which vector excision had occurred. Following growth, each subculture was distributed at low cell density in 480 individual culture wells in media supplemented with gancyclovir (Ganc) to select against the tk gene (25). A total of 71 and 141 GancR colonies were recovered from the duplicate subcultures, respectively. As shown in Figure 5D and E, and outlined in the figure legend, multiple IgH μ gene structures are possible following vector excision depending on the location of the crossover. Genomic DNA was prepared from the GancR colonies and screened by PCR using primer pair 23819–CμXbRI (Figure 5A) (data not shown). From each of the duplicate subcultures, two GancR cell lines that contained the 8.5 kb PCR product diagnostic of the μ gene excision product that is deleted for Eμ/MARs (Figure 5D) (designated, 2A2Eμ–#1–5, 2A2Eμ–#1–6, 2A2Eμ–#2–4 and 2A2Eμ–#2–10), and two GancR cell lines that contained the 9.5 kb PCR product diagnostic of the normal μ gene locus (Figure 5E) (designated, 2A2Eμ+#1–4, 2A2Eμ+#1–8, 2A2Eμ+#2–16 and 2A2Eμ+#2–20) were isolated. Southern analysis using BamHI and Probe N (Figure 5D and E) confirmed the IgH μ gene structures in the eight cell lines (data not shown).
For determination of GC frequencies, each cell line was expanded from a single cell for 25 generations under G418 selection. Genomic DNA was prepared from each culture, and qPCR utilizing primer pair 1/4 was performed to measure the frequency of GC in the 5' Cμ region. This analysis revealed a mean GC frequency of <0.11 x 10–3 recombinants/cell in the Eμ/MARs deleted cell lines (Table 1). This value is 30-fold lower than the mean GC frequency measured in the cell lines with an intact μ gene regulatory region (3.4 x 10–3 recombinants/cell) (Table 1) (t-test of log-transformed data; P = 4 x 10–6). This suggests a role for the Eμ/MARs fragment in promoting high-frequency GC between the Cμ regions at the IgH μ locus.
The Eμ/MARs segment alone is not sufficient to stimulate gene conversion between Cμ region repeats at ectopic sites in the hybridoma genome
Since the DNA fragment containing the entire IgH μ gene regulatory region was able to stimulate GC between Cμ region repeats at ectopic sites in the genome, it was of interest to determine whether the Eμ/MARs fragment alone also stimulated GC outside the IgH μ locus. To examine this, the 992 bp XbaI fragment containing Eμ/MARs was inserted into pCμ-repeat, in its correct orientation, upstream of the mutant 5' Cμ region (Figure 6). The vector was linearized at the unique NotI site and transferred by electroporation into the hybridoma, igm10 (16). A total of 88 individual G418R transformants were isolated and screened by PCR analysis in conjunction with Southern blotting using Cμ probe F (Figure 6) to determine the structure and copy number of the integrated vector (data not shown). From this screening, five cell lines were recovered that contained a single, intact vector integrated in the genome, designated EμCμRep-17, EμCμRep-46, EμCμRep-61, EμCμRep-75 and EμCμRep-81. Each of the transformants was expanded from a single cell and grown for 25 generations under G418 selection. Genomic DNA was prepared from each culture, and qPCR analysis utilizing primer pair 1/3 was performed to measure the frequency of GC in the 5' Cμ region. The median GC frequency was 0.10 x 10–3 recombinants/cell, which is similar to the median GC frequency of 0.07 x 10–3 recombinants/cell measured in the R/CμRepeat cell lines (Table 1). Thus, there was no significant difference in the frequency of GC between Cμ repeats at ectopic genomic sites, with or without the Eμ/MARs fragment (Mann–Whitney test, P = 0.24).
Figure 6. Influence of Eμ on gene conversion between ectopic Cμ repeats. The structure of hybridoma cell lines bearing the linearized vector, pEμCμRep, as determined by Southern blot and PCR analysis is shown. The vector was derived from pCμ-repeat (Figure 3A), by inserting the 995-bp Eμ/MARs-containing XbaI fragment into the MfeI site, in its correct orientation 5' of the Cμ regions. The diagnostic fragment sizes generated following digestion with XbaI and SpeI, and hybridization with probe F (Figure 1), are presented. PCR amplification with primer pair 24226–CμXbRI generates a 1.9 kb PCR product. Southern blotting using both single cutters and noncutters was used to determine the copy number of the integrated vector (data not shown). Abbreviations are the same as in Figure 1. The figure is not drawn to scale.
Evidence that transcription initiated at the μ gene promoter is not necessary for the stimulation of gene conversion by the IgH μ gene regulatory region
It remained a possibility that the reduced GC frequency in the absence of the endogenous Eμ/MARs fragment was an indirect consequence of a reduced level of μ gene transcription. To determine what effect the Eμ/MARs deletion had on μ gene transcript level, the amount of μ-specific RNA in the cell lines was examined by RNA dot blot analysis (Figure 7) (16,22,27). Previously, deletion of the majority of the VH–Cμ intron, including the Eμ/MARs elements, was shown to cause equivalent changes in both the levels of μ mRNA and nuclear run-on activity, suggesting that no elements are present within this region that affect RNA stability or processing efficiency (28). Therefore, we suggest that the μ-specific RNA levels measured in our cell lines is probably a good indicator of the level of transcription through the Cμ regions. Densitometric analysis revealed that the μ-specific RNA levels in the Eμ/MARs-deleted cell lines, 2A2Eμ–#1–5 and 2A2Eμ–#2–10, were 25% and 10% of that in Im/RCμ2A2, respectively (Table 2). In comparison, μ RNA levels in the transformants bearing the intact Eμ/MARs fragment (2A2Eμ+ #1–4 and 2A2Eμ+ #2–20), were 144% and 92% of that in Im/RCμ2A2, respectively.
Figure 7. RNA dot blot analysis of μ RNA. Serial 1:4 dilutions of total RNA were applied to a nitrocellulose filter starting with 10 μg as described in (22). Relative levels of μ RNA were determined by hybridization with 32P-labeled Cμ-specific Probe F (Figure 1), and separately with a 32P-labeled 2.0-kb PstI fragment containing the chicken ?-actin cDNA (27). The cell line igm10 is an Sp6-derived mutant that lacks the endogenous chromosomal μ gene (16).
Table 2. Cμ-specific RNA levels
To investigate whether μ-specific RNA levels correlated with the frequency of GC between the Cμ repeats, a 154 bp XbaI/NcoI fragment encompassing the Pμ TATA box, and octamer motif, 5'-ATTTGCAT-3', was deleted from the vector pVHCμRep to generate pVHCμRepPμ (Figure 8). Protein binding to the octamer motif of Pμ is required for high-level expression of the IgH gene in B-cells (29,30). The vector pVHCμRepPμ was linearized at the unique NotI site and transferred by electroporation into the igm10 hybridoma (16). Southern blot and PCR analysis of genomic DNA from 84 individual G418R transformants identified five cell lines that contained a single copy of the integrated vector (Figure 8), designated VHCμRepPμ-1, VHCμRepPμ-5, VHCμRep Pμ-7, VHCμRepPμ-15 and VHCμRepPμ-26 (data not shown). Next, the amount of μ-specific RNA in the Pμ-deleted cell lines was compared to those with an intact promoter region (the VHCμRep cell lines reported in Table 1) (Figure 9). As shown in Table 2, the μ RNA levels in the VHCμRep cell lines ranged between 21 and 183% of that in Im/RCμ2A2. In comparison, the μ RNA in the VHCμRepPμ cell lines was between 6 and 20% of that in Im/RCμ2A2 (Table 2).
Figure 8. Construction of μ gene promoter (Pμ)-deleted cell lines. The structure of the linearized vector, pVHCμRep, is shown. The vector is identical to pVHCμRep (Figure 3B) with the exception of a 154 bp XbaI/NcoI deletion encompassing the μ gene promoter, Pμ, as explained in the text. The diagnostic fragment sizes generated following digestion with EcoRI and SpeI, and hybridization with probe F (Figure 1) and probe Eμ (Figure 3), are presented. Southern blotting using both single cutters and noncutters was used to determine the copy number of the integrated vector (data not shown). N, NotI; Sp, SpeI; other abbreviations are the same as in Figure 1. This figure is not drawn to scale.
Figure 9. RNA dot blot analysis of μ RNA. Serial 1:4 dilutions of total RNA were applied to a nitrocellulose filter starting with 10 μg as described (22). Relative levels of μ RNA were determined by hybridization with 32P-labeled Cμ-specific Probe F (Figure 1). To verify loading, the blot was reprobed with a 32P-labeled 2.0 kb PstI fragment containing the chicken ?-actin cDNA (27). The cell line igm10 is an Sp6-derived mutant that lacks the endogenous chromosomal μ gene (16).
Each of the VHCμRepPμ transformants was expanded from a single cell and grown for 25 generations under G418 selection. Genomic DNA was prepared from each culture, and qPCR utilizing primer pair 1/3 was performed to measure the frequency of GC in the 5' Cμ region. This analysis revealed a median GC frequency of 1.7 x 10–3 recombinants/cell (Table 3). This value is not significantly different from the median value measured in the VHCμRep transformants (1.5 x 10–3 recombinants/cell) (Table 1) (Mann–Whitney test, P = 0.46). Importantly, no significant correlation exists between μ RNA levels and GC frequencies in the VHCμRep and VHCμRepPμ transformants (Pearson's correlation, P = 0.36).
Table 3. Frequency of gene conversion at ectopic genomic sites
Interestingly, the GC frequency in transformant VHCμRepPμ-5 was 5.3% (Table 3), a value that is 761-fold higher than the median frequency measured in the R/CμRepeat transformants (Table 1). To determine whether this exceptionally high frequency resulted from an early recombination event during expansion of the culture (i.e. a ‘jackpot’), three subclone cultures of VHCμRepPμ-5, each started from a single cell, were grown for 25 generations and the frequency of GC in the 5' Cμ region was determined by qPCR. As shown in Table 3, the frequency of GC in the subclone cultures ranged from 1.9 to 10%, suggesting that the vector may have fortuitously integrated into an unusually recombinogenic region of the genome.
DISCUSSION
Previously, we showed that the IgH μ locus behaved as a hotspot for spontaneous, mitotic intrachromosomal GC between direct repeats of the Cμ region in murine hybridoma cells (14). Here, we report the identification of a 9.1 kb segment of the μ locus regulatory region that confers GC hotspot activity in this system: deleting RES from the endogenous IgH locus abolished GC between the Cμ repeats, while a stimulation of GC was observed when RES was appended to Cμ repeats in ectopic positions in the hybridoma genome. The RES encompasses DNA sequences beginning 5' of the VH region and extending to immediately 5' of the first Cμ region exon. Several regulatory elements that function in the assembly, expression and replication of the IgH μ gene reside within this fragment. This includes the μ gene promoter (Pμ), the major intronic μ gene enhancer (Eμ) (23) and the μ gene switch (Sμ) region (31). Also, associated with Eμ are two matrix attachment regions (MARs) (32), a promoter of sterile transcripts (Iμ) (23,33) and a putative origin of replication (34).
It is noteworthy that RES-stimulated GC between Cμ repeats in ectopic genomic positions is only 20–30% of the mean value measured at the endogenous μ locus. As the RES fragment used in the ectopic studies contained an 2.8 kb deletion encompassing most of the Sμ region, this might suggest a role for Sμ sequences in RES activity. The role of Sμ in class switch recombination (CSR) is well documented (31). In addition, Sμ is frequently involved in translocations observed in B-cell lymphomas (35), and is a preferred site for insertion of transfected DNA (36). However, the failure to observe a reduction in the frequency of GC between Cμ repeats positioned at the IgH locus, which contained the same 2.8 kb Sμ deletion, argues against a role for Sμ in RES activity.
Rather, the reduced RES activity at ectopic sites might suggest that DNA sequences or elements at or near the IgH locus, which are not included within the RES fragment, are necessary to generate the optimal environment. A possible candidate is the 3' enhancer (E) complex that resides 200 kb 3' of the Cμ region (37). E has previously been shown to be essential for CSR (38), and is believed to function with Eμ to maintain high levels of expression of the IgH locus in fully differentiated B-cells (39). A second possibility is that elements within the RES itself might generate a domain that is conducive to HR at the endogenous μ locus, but at ectopic sites, cannot fully recreate this environment due to the influence of neighboring DNA sequences or chromatin structure at the site of chromosomal integration. That chromatin structure can influence spontaneous HR has been suggested previously (11,13).
In the IgH μ locus, Eμ and the flanking MARs generate a domain of chromatin accessibility, as suggested by increased sensitivity to DNA-damaging agents (40,41). Analysis of Eμ/MARs deleted cell lines revealed a 30-fold decrease in the frequency of GC at the IgH μ locus, indicating that the Eμ/MARs segment is required for RES activity. However, by itself, the Eμ/MARs fragment did not have a stimulatory effect on adjacent Cμ repeats when integrated at ectopic genomic sites. This finding might suggest that other elements present in the larger RES segment, but not included within the smaller Eμ/MARs fragment, are necessary for GC-stimulating activity; or that the context of the Eμ/MARs elements, with respect to the adjacent nucleotide sequences, is important for its function.
Transcription has previously been shown to stimulate mitotic HR in mammalian cells (42–44). In comparison to the hybridomas with an intact regulatory region, a substantial reduction in μ-specific RNA levels was observed in the Eμ-deleted cell lines. This raised the possibility that the Eμ/MARs complex is indirectly related to RES activity through its role as a transcriptional activator. In contrast to our results, several B-cell lines that lack both Eμ and the MARs at the endogenous IgH locus, have been shown to produce nearly normal levels of Ig mRNA or protein (45,46). This is generally ascribed to the presence of functionally redundant elements, possibly the 3' enhancers (39). The apparent discrepancy, however, can be resolved by the finding that introduction of a gpt cassette 3' of the endogenous IgH μ gene renders expression dependent on the Eμ/MARs elements (28). In that study, deletion of Eμ and the MARs depressed μ expression to 2% the normal level. Our findings suggest that the IgH locus is similarly insulated from the effects of any redundant elements, quite possibly by the integrated pSV2neo vector sequences located downstream of the 5' Cμ region.
To directly determine whether GC between the Cμ repeats correlated with the high level of IgH μ gene transcription, Pμ was crippled in the otherwise intact RES fragment by deleting two highly conserved elements, the TATA box and an octamer motif. The mean μ RNA level in the Pμ-deleted transformants was reduced 80% compared to transformants bearing the intact RES fragment, a result which agrees with previous studies showing that mutation of either motif drastically reduces μ transcription (29,30). More importantly however, the reduction in μ RNA levels was not associated with a corresponding decrease in the frequency of GC promoted by the RES. The lack of any correlation between the GC frequency and μ RNA levels suggests that transcriptional activity is unlikely to be responsible for RES-stimulated recombination in this system. However, since a residual level of μ RNA persisted in the stable transformants, possibly as a result of sterile transcripts initiating at Iμ (33), we cannot discount the possibility that a low level of transcription may still account for the RES activity.
Assuming the Eμ/MARs are the important elements, what role do they play in GC hotspot activity at the IgH μ locus? The simplest hypothesis is that enhancer-mediated chromatin changes within the Cμ repeats increases their susceptibility to nicks or breaks, which invoke HR for repair. This notion is consistent with previous results that suggest GC events initiate at random sites across the entire Cμ region (14). Also, a high frequency of mutation in the chromosomal Ig μ and genes has been reported for hybridoma and myeloma cell lines (15,47). It would be interesting to know whether the high rate of mutation is also dependent on the enhancer elements.
Alternatively, chromatin remodeling mediated by the Eμ/MARs elements might facilitate the adoption of a non-B-DNA conformation within the Cμ repeats. Regions of alternating purines and pyrimidines readily form left-handed Z-DNA when contained in negatively supercoiled DNA (48). Interestingly, a potential Z-DNA forming run of 28 repeats of the dinucleotide, GT, is present within the Cμ region of homology (49). Smaller regions of alternating purines and pyrimidines are also present throughout the repeated Cμ regions (49). Sequences capable of forming Z-DNA have previously been shown to stimulate mitotic HR (5–7).
In summary, we have engineered various mouse hybridoma cell lines to investigate the importance of IgH μ gene regulatory elements in promoting GC in a recombination reporter consisting of a pair of homologous, repeated Cμ region segments. Our results suggest that a region we refer to as the RES, probably, through a main contribution from the internal Eμ/MARs fragment, promotes GC hotspot activity between the Cμ repeats at the endogenous IgH μ locus. While the exact mechanism is unknown, it is possible that Eμ/MARs exerts its effect through changes in chromatin accessibility or chromatin remodeling rendering the adjacent Cμ repeats more susceptible to DNA breaks, which invoke GC for repair. One could imagine that in the absence of proper repair by HR, a similar mechanism acting at the endogenous IgH μ locus in normal B cells might promote genomic instability, e.g. by stimulating chromosomal translocations into the IgH μ locus (35). While our results provide support for cis-acting regulatory elements promoting GC at the IgH μ locus, the analysis of recombination in cell line, VHCμRepPμ-5, suggests that GC hotspot activity can also be a feature of other loci in the mammalian genome. Finally, it is possible that the high frequency, Eμ/MAR-stimulated GC that we observe in this model system bears some mechanistic relationship to DNA transactions that are required to generate antibody diversity in normal B cells (50). In this regard, it is worth noting that the high frequency of GC that we observe in our system is known to be an important mechanism of antibody diversification in chickens and rabbits (51,52), and there is evidence that GC might play a role in some murine antibody responses as well (53–55).
ACKNOWLEDGEMENTS
This research was supported by a PhD studentship from the Canadian Institutes of Health Research (CIHR) to S.J.R., and a CIHR operating grant to M.D.B.
REFERENCES
Aguilera,A. ( (2002) ) The connection between transcription and genomic instability. EMBO J., , 21, , 195–201.
Kuzminov,A. ( (2001) ) DNA replication meets genetic exchange: chromosomal damage and its repair by homologous recombination. Proc. Natl Acad. Sci. USA, , 98, , 8461–8468.
Voelkel-Meiman,K., Keil,R.L. and Roeder,G.S. ( (1987) ) Recombination-stimulating sequences in yeast ribosomal DNA correspond to sequences regulating transcription by RNA polymerase I. Cell, , 48, , 1071–1079.
Stewart,S.E. and Roeder,G.S. ( (1989) ) Transcription by RNA polymerase I stimulates mitotic recombination in Saccharomyces cerevisiae. Mol. Cell. Biol., , 9, , 3464–3472.
Stringer,J.R. ( (1985) ) Recombination between poly sequences in simian virus 40-infected cultured cells. Mol. Cell. Biol., , 5, , 1247–1259.
Bullock,P., Miller,J. and Botchan,M. ( (1986) ) Effects of poly and poly repeats on homologous recombination in somatic cells. Mol. Cell. Biol., , 6, , 3948–3953.
Wahls,W.P., Wallace,L.J. and Moore,P.D. ( (1990) ) The Z-DNA motif d(TG)30 promotes reception of information during gene conversion events while stimulating homologous recombination in human cells in culture. Mol. Cell. Biol., , 10, , 785–793.
Lenzmeier,B.A. and Freudenreich,C.H. ( (2003) ) Trinucleotide repeat instability: a hairpin curve at the crossroads of replication, recombination, and repair. Cytogenet. Genome Res., , 100, , 7–24.
Gordenin,D.A., Lobachev,K.S., Degtyareva,N.P., Malkova,A.L., Perkins,E. and Resnick,M.A. ( (1993) ) Inverted DNA repeats: a source of eukaryotic genomic instability. Mol. Cell. Biol., , 13, , 5315–5322.
Farah,J.A., Hartsuiker,E., Mizuno,K., Ohta,K. and Smith,G.R. ( (2002) ) A 160-bp palindrome is a Rad50·Rad32-dependent mitotic recombination hotspot in Schizosaccharomyces pombe. Genetics, , 161, , 461–468.
Malagón,F. and Aguilera,A. ( (1996) ) Differential intrachromosomal hyper-recombination phenotype of spt4 and spt6 mutants of S. cerevisiae. Curr. Genet., , 30, , 101–106.
Fritze,C.E., Verschueren,K., Strich,R. and Easton Esposito,R. ( (1997) ) Direct evidence for SIR2 modulation of chromatin structure in yeast rDNA. EMBO J., , 16, , 6495–6509.
Gottlieb,S. and Esposito,R.E. ( (1989) ) A new role for a yeast transcriptional silencer gene, SIR2, in regulation of recombination in ribosomal DNA. Cell, , 56, , 771–776.
Raynard,S.J., Read,L.R. and Baker,M.D. ( (2002) ) Evidence for the murine IgH μ locus acting as a hot spot for intrachromosomal homologous recombination. J. Immunol., , 168, , 2332–2339.
K?hler,G. and Shulman,M.J. ( (1980) ) Immunoglobulin M mutants. Eur. J. Immunol., , 10, , 467–476.
K?hler,G., Potash,M.J., Lehrach,H. and Shulman,M.J. ( (1982) ) Deletions in immunoglobulin μ chains. EMBO J., , 1, , 555–563.
Baker,M.D., Pennell,N., Bosnoyan,L. and Shulman,M.J. ( (1988) ) Homologous recombination can restore normal immunoglobulin M production in a mutant hybridoma cell line. Proc. Natl Acad. Sci. USA, , 85, , 6432–6436.
Baker,M.D. and Read,L.R. ( (1992) ) Ectopic recombination within homologous immunoglobulin μ gene constant regions in a mouse hybridoma cell line. Mol. Cell. Biol., , 12, , 4422–4432.
Ng,P. and Baker,M.D. ( (1999) ) Mechanisms of double-strand-break repair during gene targeting in mammalian cells. Genetics, , 151, , 1127–1141.
Gross-Bellard,M., Qudet,P. and Chambon,P. ( (1973) ) Isolation of high-molecular weight DNA from mammalian cells. Eur. J. Biochem., , 36, , 32–38.
Sambrook,J., Fritsch,E.F. and Maniatis,T. ( (1989) ) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Baumann,B., Potash,M.J. and K?hler,G. ( (1985) ) Consequences of frameshift mutations at the immunoglobulin heavy chain locus of the mouse. EMBO J., , 4, , 351–359.
Staudt,L.M. and Lenardo,M.J. ( (1991) ) Immunoglobulin gene transcription. Annu. Rev. Immunol., , 9, , 373–398.
Ochi,A., Hawley,R.G., Hawley,T., Shulman,M.J., Traunecker,A., K?hler,G. and Hozumi,N. ( (1983) ) Functional immunoglobulin M production after transfection of cloned immunoglobulin heavy and light chain genes into lymphoid cells. Proc. Natl Acad. Sci. USA, , 80, , 6351–6355.
Ng,P. and Baker,M.D. ( (1998) ) High efficiency site-specific modification of the chromosomal immunoglobulin locus by gene targeting. J. Immunol. Methods, , 214, , 81–96.
Serwe,M. and Sablitzky,F. ( (1993) ) V(D)J recombination in B cells is impaired but not blocked by targeted deletion of the immunoglobulin heavy chain intron enhancer. EMBO J., , 12, , 2321–2327.
Cleveland,D.W., Lopata,M.A., MacDonald,R.J., Cowan,N.J., Rutter,W.J. and Kirschner,M.W. ( (1980) ) Number and evolutionary conservation of - and ?-tubulin and cytoplasmic ?- and -actin genes using specific cloned cDNA probes. Cell, , 20, , 95–105.
Oancea,A.E., Berru,M. and Shulman,M.J. ( (1997) ) Expression of the (recombinant) endogenous immunoglobulin heavy-chain locus requires the intronic matrix attachment regions. Mol. Cell. Biol., , 17, , 2658–2668.
Mason,J.O., Williams,G.T. and Neuberger,M.S. ( (1985) ) Transcription cell type specificity is conferred by an immunoglobulin VH gene promoter that includes a functional consensus sequence. Cell, , 41, , 479–487.
Eaton,S. and Calame,K. ( (1987) ) Multiple DNA sequence elements are necessary for the function of an immunoglobulin heavy chain promoter. Proc. Natl Acad. Sci. USA, , 84, , 7634–7638.
Manis,J.P., Tian,M. and Alt,F.W. ( (2002) ) Mechanism and control of class-switch recombination. Trends Immunol., , 23, , 31–39.
Cockerill,P.N. ( (1990) ) Nuclear matrix attachment occurs in several regions of the IgH locus. Nucleic Acids Res., , 18, , 2643–2648.
Su,L.K. and Kadesch,T. ( (1990) ) The immunoglobulin heavy-chain enhancer functions as the promoter for Iμ sterile transcription. Mol. Cell. Biol., , 10, , 2619–2624.
Ariizumi,K., Wang,Z. and Tucker,P.W. ( (1993) ) Immunoglobulin heavy chain enhancer is located near or in an initiation zone of chromosomal DNA replication. Proc. Natl Acad. Sci. USA, , 90, , 3695–3699.
Bergsagel,P.L., Chesi,M., Nardini,E., Brents,L.A., Kirby,S.L. and Kuehl,W.M. ( (1996) ) Promiscuous translocations into immunoglobulin heavy chain switch regions in multiple myeloma. Proc. Natl Acad. Sci. USA, , 93, , 13931–13936.
Baar,J. and Shulman,M.J. ( (1995) ) The Ig heavy chain switch region is a hotspot for insertion of transfected DNA. J. Immunol., , 155, , 1911–1920.
Pettersson,S., Cook,G.P., Bruggemann,M., Williams,G.T. and Neuberger,M.S. ( (1990) ) A second B cell-specific enhancer 3' of the immunoglobulin heavy-chain locus. Nature, , 344, , 165–168.
Cogné,M., Lansford,R., Bottaro,A., Zhang,J., Gorman,J., Young,F., Cheng,H.L. and Alt,F.W. ( (1994) ) A class switch control region at the 3' end of the immunoglobulin heavy chain locus. Cell, , 77, , 737–747.
Lieberson,R., Ong,J., Shi,X. and Eckhardt,L.A. ( (1995) ) Immunoglobulin gene transcription ceases upon deletion of a distant enhancer. EMBO J., , 14, , 6229–6238.
Jenuwein,T., Forrester,W.C., Fernandez-Herrero,L.A., Laible,G., Dull,M. and Grosschedl,R. ( (1997) ) Extension of chromatin accessibility of nuclear matrix attachment regions. Nature, , 385, , 269–272.
Nikolajczyk,B.S., Sanchez,J.A. and Sen,R. ( (1999) ) ETS protein-dependent accessibility changes at the immunoglobulin μ heavy chain enhancer. Immunity, , 11, , 11–20.
Nickoloff,J.A. and Reynolds,R.J. ( (1990) ) Transcription stimulates homologous recombination in mammalian cells. Mol. Cell. Biol., , 10, , 4837–4845.
Nickoloff,J.A. ( (1992) ) Transcription enhances intrachromosomal homologous recombination in mammalian cells. Mol. Cell. Biol., , 12, , 5311–5318.
Thyagarajan,B., Johnson,B.L. and Campbell,C. ( (1995) ) The effect of target site transcription on gene targeting in human cells in vitro. Nucleic Acids Res., , 23, , 2784–2790.
Wabl,S. and Burrows,P.D. ( (1984) ) Expression of immunoglobulin heavy chain at a high level in the absence of a proposed immunoglobulin enhancer element in cis. Proc. Natl Acad. Sci. USA, , 81, , 2452–2455.
Zaller,D.M. and Eckhardt,L.A. ( (1985) ) Deletion of a B-cell-specific enhancer affects transfected, but not endogenous, immunoglobulin heavy-chain gene expression. Proc. Natl Acad. Sci. USA, , 82, , 5088–5092.
Yu,H. and Eckhardt,L.A. ( (1986) ) DNA rearrangement causes a high rate of spontaneous mutation at the immunoglobulin heavy-chain locus of a mouse myeloma cell line. Mol. Cell. Biol., , 6, , 4228–4235.
Nordheim,A., Peck,L.J., Lafer,E.M., Stollar,B.D., Wang,J.C. and Rich,A. ( (1983) ) Supercoiling and left-handed Z-DNA. Cold Spring Harb. Symp. Quant. Biol., , 47, , 93–100.
Richards,J.E., Gilliam,A.C., Shen,A., Tucker,P.W. and Blattner,F.R. ( (1983) ) Unusual sequences in the murine immunoglobulin μ- heavy-chain region. Nature, , 306, , 483–487.
Alt,F.W., Blackwell,T.K. and Yancopoulos,G.D. ( (1987) ) Development of the primary antibody repertoire. Science, , 238, , 1079–1087.
Reynaud,C.A., Dahan,A., Anquez,V. and Weill,J.C. ( (1989) ) Somatic hyperconversion diversifies the single VH gene of the chicken with a high incidence in the D region. Cell, , 59, , 171–183.
Becker,R.S. and Knight,K.L. ( (1990) ) Somatic diversification of immunoglobulin heavy chain VDJ genes: evidence for somatic gene conversion in rabbits. Cell, , 63, , 987–997.
David,V., Folk,N.L. and Maizels,N. ( (1992) ) Germ line variable regions that match hypermutated sequences in genes encoding murine anti-hapten antibodies. Genetics, , 132, , 799–811.
Xu,B. and Selsing,E. ( (1994) ) Analysis of sequence transfers resembling gene conversion in a mouse antibody transgene. Science, , 265, , 1590–1593.
Selsing,E., Xu,B. and Sigurdardottir,D. ( (1996) ) Gene conversion and homologous recombination in murine B cells. Semin. Immunol., , 8, , 51–158.(Steven J. Raynard and Mark D. Baker*)