HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

J. Biol. Chem., Vol. 276, Issue 31, 29126-29133, August 3, 2001

This Article Abstract Full Text (PDF) All Versions of this Article: 276/31/29126    most recent M103797200v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Raghavan, S. C. Articles by Lieber, M. R. Search for Related Content PubMed PubMed Citation Articles by Raghavan, S. C. Articles by Lieber, M. R.

Analysis of the V(D)J Recombination Efficiency at Lymphoid Chromosomal Translocation Breakpoints^*

Sathees C. Raghavan, Ilan R. Kirsch^§, and Michael R. Lieber^¶

From the  Departments of Pathology, Biochemistry & Molecular Biology, Molecular Microbiology & Immunology, and Biological Sciences, Norris Comprehensive Cancer Center, University of Southern California Keck School of Medicine, Los Angeles, California 90089-9176 and the ^§ Medicine Branch and Department of Genetics, NCI, National Institutes of Health, Bethesda, Maryland 20889-5105

Received for publication, April 27, 2001, and in revised form, June 3, 2001

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Chromosomal translocations and deletions are among the major events that initiate neoplasia. For lymphoid chromosomal translocations, misrecognition by the RAG (recombination activating gene) complex of V(D)J recombination is one contributing factor that has long been proposed. The chromosomal translocations involving LMO2 (t(11;14)(p13;q11)), Ttg-1 (t(11;14)(p15;q11)), and Hox11 (t(10;14)(q24;q11)) are among the clearest examples in which it appears that a D or J segment has synapsed with an adventitious heptamer/nonamer at a gene outside of one of the antigen receptor loci. The interstitial deletion at 1p32 involving SIL (SCL-interrupting locus)/SCL (stem cell leukemia) is a case involving two non-V(D)J sites that have been suggested to be V(D)J recombination mistakes. Here we have used our human extrachromosomal substrate assay to formally test the hypothesis that these regions are V(D)J recombination misrecognition sites and, more importantly, to quantify their efficiency as V(D)J recombination targets within the cell. We find that the LMO2 fragile site functions as a 12-signal at an efficiency that is only 27-fold lower than that of a consensus 12-signal. The Ttg-1 site functions as a 23-signal at an efficiency 530-fold lower than that of a consensus 23-signal. Hox11 failed to undergo recombination as a 12- or 23-signal and was at least 20,000-fold less efficient than consensus signals. SIL has been predicted to function as a 12-signal and SCL as a 23-signal. However, we find that SIL actually functions as a 23-signal. These results provide a formal demonstration that certain chromosomal fragile sites can serve as RAG complex targets, and they determine whether these sites function as 12- versus 23-signals. These results quantify one of the three major factors that determine the frequency of these translocations in T-cell acute lymphocytic leukemia.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Alterations in chromosome structure are among the major genetic changes that initiate neoplasia. The list of causes of chromosomal translocations and deletions is not yet fully established, but the following factors are thought to be included: ionizing radiation, free radicals, and enyzmes of DNA metabolism (polymerases, nucleases, and topoisomerases) (1-4). In lymphoid cells, an additional cause is misrecognition by the RAG complex (5-7). The normal function of the RAG¹ complex is to initiate the breaks of V(D)J recombination at pairs of consensus heptamer/nonamer sequences in which one recombination signal has a 12-base pair (bp) spacer between the heptamer and nonamer (12-signal) and the other signal has a 23-bp spacer (23-signal). Misrecognition by the RAG complex has been speculated for numerous lymphoid chromosomal translocations; however, there have been minimal or no quantitative tests of this issue.

All chromosomal deletions and reciprocal translocations require at least two double strand DNA breaks at separate chromosomal regions. In some of these cases, one of the breaks is at a known V, D, or J segment, making the cause of the break at this location clearly attributable to the RAG complex. However, the cause of the other break is often less certain. If the non-V(D)J site has a convincing heptamer/nonamer sequence and consistently occurs in many patients, then it is reasonable to conjecture that the RAG complex is responsible. However, even for V, D, and J elements, the heptamer and nonamer portions of the signal sequence vary from the consensus, which has the following configuration: [coding end]/CACAGTG- - -12 or 23 spacer- - -ACAAAAACC. The number of theoretically possible heptamer/nonamer signals is well over 10⁹ variations, and fewer than 100 have been tested for their in vivo efficiency (8-12). None have been tested in human cells. The consensus heptamer/nonamer is thus far the most efficient sequence for directing V(D)J recombination. Given that such a small fraction of the total number of variants has been tested, the vast majority of variations from the consensus are uncharacterized as to how efficiently they can direct V(D)J recombination if at all. This makes predicting the V(D)J recombination potential of a chromosomal breakpoint site very uncertain, and many such breakpoints may turn out to be cleaved by a mechanism unrelated to V(D)J recombination. Hence, one can only be truly certain of the cause of the breakage if one can recapitulate it under experimental conditions. An additional advantage of this is that one can then quantify the frequency of the breakage, which provides very important information for evaluating these sites as breakpoints.

Some events involve two chromosomal locations, neither of which is a V, D, or J segment. In these cases where the translocation or rearrangement involves two non-V(D)J sites, both are subject to causal uncertainty. It is conceivable that one or both of these breakage sites are caused by a mechanism other than V(D)J recombination. Again, consistent location adjacent to a near-consensus heptamer/nonamer sequence has been the only basis for suggesting the RAG complex. In cases in which the position varies and/or the adjacent sequences do not contain convincing heptamer/nonamer sequences, the cause of the break is even more ambiguous. Only experimental testing of such breakpoints can provide a reliable answer.

The complexity of evaluating non-V(D)J breakage sites is increased because mechanistic studies of V(D)J recombination have been unable to define the frequency with which a non-RAG breakage site can recombine with a RAG breakage site. RAG biochemical studies have shown that double strand breaks caused by the RAG complex do so at pairs of signal sites, one with a 12-signal and one with a 23-signal. Double strand breaks because of the RAGs at a pair of 12-signals (or a pair of 23-signals) occur but at frequencies that are 50- to more than 100-fold lower than for a 12/23 pair (8, 9, 11). Double strand breaks by the RAG complex at a single signal have not been possible to reliably distinguish from background levels of pairing and cleavage of two 12- or two 23-signals (13). It remains a matter of speculation as to the efficiency with which the RAG complex would cleave one signal and recombine it with a break caused, for example, by another nuclease. The efficiency of V(D)J recombination when one signal only has a heptamer but no nonamer is reduced markedly. When the nonamer of the 12-signal is altered at every position, but the 23-signal is a consensus one, then the efficiency is about 20-fold reduced for one particular nonamer ablation substrate (8). When the 23-signal lacks a nonamer but the 12-signal is a consensus one, then the efficiency is reduced by at least 100-fold.

The chromosomal translocations involving LMO2 (t(11;14)(p13;q11)) (14-17), Ttg-1 (t(11;14)(p15;q11)) (18-20), and Hox11 (t(10;14)(q24;q11)) (21-23) are among the clearest examples in which it appears that a D or J segment has partnered with an adventitious heptamer at a gene outside one of the antigen receptor loci (Figs. 1 and 2). The interstitial deletion at 1p32, SIL/SCL, is a case involving two non-V(D)J sites that have been proposed as being cleaved by RAGs (24, 25).

View larger version (33K): [in this window] [in a new window]   Fig. 1.   Diagrams of the LMO2, Ttg-1, Hox11, and SIL/SCL events studied. The dark arrows in each panel indicates the proposed break region on the respective chromosomes. The triangles represent the proposed signals at the translocation junctions. LMO2 Translocation, the breakpoint region of the LMO2 gene at chromosome 11p13 is depicted based on previous sequencing of translocations from four patients (14, 17, 27). The breakpoint junction at chromosome 14 is located at the 23-signal of D1. In two patients, the breakpoint was reported at the 12-signal of T-cell receptor 1 of chromosome 7q35 (not included in the figure). Ttg1 Translocation, the translocation breakpoint of the Ttg-1 gene at chromosome 11p15 is depicted (18-20). At chromosome 14, the break occurs at the 12-signal of D1. Hox11 Translocation, the translocation breakpoint of the Hox11 gene is at chromosome 10q24; on chromosome 14, it is at the 12-signal of the D2 (21, 22). SIL/SCL Deletion, depicts the SIL/SCL interstitial deletion. Adjacent to the breakpoints, heptamer/nonamer-like sequences (indicated as triangles) are present (25, 34, 35). ~100 kb indicates the approximate length of the interstitial deletion; the triangles are intermediately shaded between the 12- and 23-signals because of the uncertainty as to which signal they are mimicking.

View larger version (26K): [in this window] [in a new window]   Fig. 2.   Depiction of breakpoint spanning sequences used in this study. The potential heptamers and nonamers are indicated in bold letters with underlining. In the cases where the identity of the signal (12 versus 23) is not clear, nonamers are marked for the use of the signals as both a 12- and a 23-signal. In the SIL/SCL panel, the SCL sequences are presented at the top, and the SIL sequences are at the bottom. The downward pointing arrows indicate the proposed breakpoint on the respective chromosomes, and the upward pointing arrows indicate the exact nucleotide position at which the recombination is predicted to occur.

In the t(11;14)(p13;q11) translocation, recombination occurs between the LMO2 (Ttg-2, rhombotin-2) gene on chromosome 11 and T-cell receptor D elements on chromosome 14 (14, 26). This translocation results in the aberrant expression of LMO2 and occurs in 10-20% of T-cell acute lymphoblastic lymphoma (T-ALL). Most of the breakpoints are clustered within a 2-kilobase pair region. Of five patients in which the breakpoints have been sequenced, four breakpoints have apparent adjacent heptamer/nonamer-like sequences, and three of these are positioned identically (14, 15, 17, 27-29).

The t(11;14)(p15;q11) is one of the least studied translocations and is between the Ttg-1 (rhombotin-1) gene on chromosome 11 (a gene normally involved in brain development) and the T-cell receptor locus (D1) on chromosome 14. It is reported to be associated with <5% of T-ALL (18-20).

The t(10;14)(q24;q11) translocation involves the Hox11 (TCL3) gene on chromosome 10 (30), which is normally not expressed in T cells. Hox11 is deregulated upon translocation to the T-cell locus (31). This translocation is seen in 5-10% of T-ALL cells (22, 23, 32).

The SIL/SCL interstitial deletion occurs in about 25% of T-ALL. This leads to the deletion of about 100-kilobase pairs at chromosome 1p32 (24). This rearrangement juxtaposes SCL within the SIL-transcribed region, resulting in the activation of SCL (also known as Tal-1, TCL5), which is normally not expressed in T-cells (33). The breakpoint regions at both SIL and SCL possess heptamer-like sequences, although there is no obvious nonamer (25, 34, 35).

Here we have used an extrachromosomal substrate assay for human cells to pair each of these fragile sites with an optimal 12- or 23-signal (Fig. 3). We find that the LMO2 fragile site functions as a 12-signal at an efficiency that is only 27-fold lower than that of a consensus 12-signal. Ttg-1 serves as a 23-signal at an efficiency that is 530-fold lower than that of a consensus 23-signal. Hox11 failed to undergo recombination as a 12- or 23-signal and was at least 20,000-fold less efficient than consensus signals. SCL has been predicted to serve as a 23-signal and SIL as a 12-signal. However, we find that SIL functions as a 23-signal at an efficiency that is 750-fold lower than a consensus 23-signal. Consistent with this finding, SIL did not appear to function as a 12-signal when screened to levels 25,000-fold lower than consensus 12-signals. SCL failed to recombine as a 12- or as a 23-signal and was at least 20,000-fold lower than consensus signals. These results provide a formal demonstration that certain chromosomal fragile sites can serve as RAG complex targets, determining whether they function as 12- versus 23-signals. These results also quantify one of the major factors that determine the lymphoid translocation efficiency.

View larger version (25K): [in this window] [in a new window]   Fig. 3.   Outline of the human V(D)J recombinase assay. The diagram depicts the introduction of circular plasmids into a human cell line, Reh, active for V(D)J recombination. After 48 h within the cells, the minichromosomes are harvested by rapid alkaline lysis and transformed into E. coli for genetic detection of recombinants on ampicillin plus chloramphenicol (Amp + Cam) LB agar plates. The recombination is depicted between a consensus 12 (open triangle)- and 23 (dark triangle)-signal of pGG49, leading to signal joint formation. When analyzing the recombination efficiency of the translocation breakpoint regions, we ligated the breakpoint regions into the plasmid in place of either the 12- or 23-signal. The pSCR19 substrate, for example, has the LMO2 breakpoint region cloned in place of the 12-signal. Hence, on pSCR19, the LMO2 site is paired with a consensus 23-signal. cat denotes the chloramphenicol acetyl transferase gene, and Stop denotes the prokaryotic transcription terminator (49). The E. coli lac promoter is denoted as Plac. The resistance gene for ampicillin is denoted as Ampr. Resistance for ampicillin and chloramphenicol is denoted as Ampr Camr.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Reagents-- Chemical reagents are purchased from Sigma Chemical Co (St Louis, MO). Restriction endonucleases and DNA modifying enzymes are from New England Biolabs (Beverly, MA). T4 DNA ligase is obtained from from Roche Molecular Biochemicals. Components for cell culture are from Irvine Scientific (Santa Ana, CA).

Oligomers-- Oligomers were synthesized from the Microchemical Core Facility, Norris Cancer Center, University of Southern California with SalI or BamHI complementary ends to facilitate cloning. The following oligomers were used in this study: SCR1, 5'-TCGACTCCCCCTTTTCCTTACGCAATATACAGAAATGCGCGAGGCTGTGGTTGGTTTTCTCTAGAG-3'; SCR2, 5'-TCGACTCTAGAGAAAACCAACCACAGCCTCGCGCATTTCTGTATATTGCGTAAGGAAAAGGGGGAG-3'; SCR3, 5'TCGACTCTGGCTCACACTCTGCTACGTAGTAAGGGATCAGTTAATGTTTGAAGTTCATCTAGAG-3'; SCR4, 5'-TCGACTCTAGATGAACTTCAAACATTAACTGATCCCTTACTACGTAGCAGAGTGTGAGCCAGAG-3'; SCR7, 5'-GATCCTTGTCTTGAGCTCACACAGTGGCTCACCACCCCACACAGCCCTCACTCTGGCATGCGGATCTAGAG-3'; SCR8, 5'-GATCCTCTAGATCCGCATGCCAGAGTGAGGGCTGTGTGGGGTGGTGAGCCACTGTGTGAGCTCAAGACAAG-3'; SCR9, 5'-GATCCTCCCCCTTTTCCTTACGCAATATACAGAAATGCGCGAGGCTGTGGTTGGTTTTCTCTAGAG-3'; SCR10, 5'-GATCCTCTAGAGAAAACCAACCACAGCCTCGCGCATTTCTGTATATTGCGTAAGGAAAAGGGGGAG-3'; SCR11, 5'-GATCCTCTGGCTCACACTCTGCTACGTAGTAAGGGATCAGTTAATGTTTGAAGTTCATCTAGAG-3'; SCR12, 5'-GATCCTCTAGATGAACTTCAAACATTAACTGATCCCTTACTACGTAGCAGAGTGTGAGCCAGAG-3'; SCR13, 5'-GATCCCGCGCACAGCCAATGGAGAGACCCAGTCGAAACCGCGAACGTCTCTTTCTAGAG-3'; SCR14, 5'-GATCCTCTAGAAAGAGACGTTCGCGGTTTCGACTGGGTCTCTCCATTGGCTGTGCGCGG-3'; SCR15, 5'-TCGACCCGAAATTATTGCTGGGTAAGACAATACTGTGTGTGTTCTAGAG-3'; SCR16, 5'-TCGACTCTAGAACACACACAGTATTGTCTTACCCAGCAATAATTTCGGG-3'; SCR25, 5'-TCGACCGCGCACAGCCAATGGAGAGACCCAGTCGAAACCGCGAACGTCTCTTTCTAGAG-3'; SCR26, 5'-TCGACTCTAGAAAGAGACGTTCGCGGTTTCGACTGGGTCTCTCCATTCGCTGTGCGCGG-3'.

The oligomers were purified using 8-10% denaturing polyacrylamide gel electrophoresis. After phosphorylation, the complementary oligomers were annealed in 0.1 M NaCl and 1 mM EDTA by heating in a beaker of boiling water for 10 min, followed by slow cooling (leaving 5' overhangs to facilitate cloning).

The oligomers used for PCR amplification are as follows. SCR21, 5'-AGTGCCACCTGACGTCTAAG-3'; SCR22, 5'-CCCGAGGGTTTTTGTACAGC-3'; SCR23, 5'-CTCTAGAAAGAGACGTTCGC-3'; SCR24, 5'-TCCTCTAGATCCGCATGCCA-3'. These oligomers were obtained from Operon Technologies (Richmond, CA).

Plasmid Construction-- The plasmid constructs were made by modifying the SV40-based plasmid pGG49 (36). The annealed (double-stranded) oligomers were cloned into the pGG49 vector after removing the 12-signal (by SalI digestion), the 23-signal (by BamHI digestion), or both. The SCR15/16 oligomers carrying the LMO2 breakpoint region was cloned into the SalI site of pGG49 (pSCR19). The oligomers with the Ttg-1 breakpoint (SCR7/8) were cloned into the BamHI site of the pGG49 after removing the 23-signal (pSCR8). Hox11 oligomers were cloned as either a 23-signal, pSCR17 (SCR13/14), or a 12-signal, pSCR29 (SCR25/26). The oligomers covering the SIL breakpoint region were cloned at the SalI site of pGG49 creating pSCR12 (oligomers SCR3/4) or at the BamHI site (oligomers SCR11/12) creating pSCR15. Similarly, SCL breakpoint junctions containing oligomers SCR1/2 and SCR9/10 were cloned into the SalI site (pSCR10) or the BamHI site (pSCR15) of pGG49, respectively. pSCR27 was constructed after deleting both 12- and 23-signals from pGG49 and replacing them with SCL (SCR1/2) and SIL (SCR11/12) containing oligomers.

Cell Line and Tissue Culture-- The human pre B-cell line, Reh, was obtained from ATCC (Manassas, VA). This cell line was originally derived from a patient with acute lymphophoblastic leukemia (37, 38). Reh cells were maintained in RPMI 1640 supplemented with 10% fetal bovine serum, 100 units of penicillin G/ml, 100 µg of streptomycin sulfate/ml, 0.292 mg of L-glutamine/ml, and 50 µM mercaptoethanol.

V(D)J Recombination Assay and Analysis of Recombinants-- The human lymphoid cell line, Reh, was grown logarithmically, transfected with the appropriate plasmid substrates using the electroporation/DEAE-dextran method as described previously (39), and cultured for 48 h at 37 °C. The various steps involved in the human V(D)J recombinase assay are summarized in Fig. 3, which we have described previously (40, 41). In brief, recombinant substrates confer both ampicillin and chloramphenicol resistance (designated as AC). The ratio of AC colonies to A colonies reflects the fraction of recovered substrates that underwent V(D)J recombination. A more meaningful ratio can be obtained by focusing on those ampicillin-resistant molecules that have replicated in the eukaryotic cells (42). The ratio of AC to A among replicated molecules is designated R ((DAC/DA) × 100), where DAC is the number of transformants that arose because of transformation with replicated recombinant molecules and DA is the number of transformants that arose because of replicated plasmids. R represents the recombination frequency of the substrate. We control for replication by digesting the recovered substrates with DpnI before bacterial transformation. DpnI cleaves the plasmid that did not lose its prokaryotic dam methylation pattern by replicating in eukaryotic cells. Only molecules replicated in eukaryotic cells (DA) remain undigested by DpnI and will transform the bacteria. The replication frequency of the given substrate is calculated by the equation (DA/A) × 100. Each eukaryotic transfection is typically analyzed with 10-20 Escherichia coli transformations to determine R ((DAC/DA) × 100). The averaging of multiple different eukaryotic transfections of the same substrate in Reh is done by summing the total number of DAC counts (after restriction analysis and sequencing verification of representative transformants) divided by the total number of transformants from replicated plasmids (DA). The resulting number is expressed as a percentage and is the weighted average, R' (= (DAC/DA)100). Direct transformation of the substrate into E. coli does not yield any DAC transformants at a frequency of 10⁶. At lower frequencies, rare DAC transformants after direct E. coli transformation do not score as V(D)J recombination events based on ApaLI digestion and sequencing.

With the V(D)J recombination positive cell line used in this study, more than 99% of the DAC colonies represented bona fide V(D)J events, based on high resolution restriction analysis of all DAC colonies picked and sequencing of representative samples. The very rare DAC colonies (<1%) that did not have the expected restriction pattern based on V(D)J rearrangement were excluded from subsequent analysis.

PCR Amplification of Recombinants-- Polymerase chain reaction (PCR) amplifications were carried out in which one primer was positioned between the BamHI sites (23-signal) of the plasmids pSCR19, pSCR17, pSCR8, and pGG49 (primers SCR22, SCR23, and SCR24 of which SCR24 is common for pSCR19 and pGG49). The second primer, which is ~500 bp away from the first one, was common for all different constructs (SCR21). The PCR reactions were performed using a thermocycler (Stratagene) under the following conditions: 1× reaction mix (10 mM Tris, pH 8.3, 50 mM KCl, 0.01% gelatin) with 1.25 mM MgCl_2, 200 µM dNTPs, 0.2 µl of [-³²P]dCTP, and 2 units of Taq polymerase in a volume of 25 µl. Amplification was carried out for 35 cycles: 94 °C for 45 s, 58 °C for 60 s, and 72 °C for 30 s. Reamplification for an additional 35 cycles was performed with 1 µl of sample from the parental reaction in a 25-µl volume. The products were resolved on a 1.3% agarose gel and dried, and signals were detected using a Molecular Dynamics PhosphorImager.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Use of the Human V(D)J Recombination Assay to Quantitatively Assess Lymphoid Translocation Regions-- We were interested in determining the efficiency with which certain lymphoid translocation breakpoints serve as targets for the RAG complex. To accomplish this, we annealed synthesized oligonucleotides that span each of these translocation breakpoint sites and then ligated them into a human V(D)J recombination substrate plasmid (Fig. 3). For a single RAG-mediated recombination event, both a 12- and a 23-signal are needed. Thus, the breakpoint region must serve as a 12- or 23-like signal when the region is synapsed with an ideal 23- or 12-signal, respectively. For those translocations that normally involve a T-cell receptor D segment, at its 23-signal side the chromosomal breakpoint site must be serving as a 12-signal (Fig. 1). For example, in the cases of LMO2, the naturally occurring chromosomal translocations are indicative of a breakpoint site serving as a 12-like signal. The test substrate that we generated here has an LMO2 breakpoint site paired with an optimal 23-signal. For those breakpoints for which it is not clear whether it is serving as a 12- or 23-like signal, we tested for both cases. That is, we paired it with both an optimal 12-signal in one substrate configuration and we with an optimal 23-signal in a second substrate configuration. We used these substrates in the cellular assay depicted in Fig. 3. Using this method, the substrates were transfected into the human pre-B cell line, Reh. After 48 h of incubation, the plasmid DNA (substrate and any recombined product) was harvested from the Reh cells and analyzed as described (see "Experimental Procedures").

Efficiency with Which the LMO2 Locus Serves as a Target for V(D)J Recombination-- Previous studies from patients with T-cell ALL due to a t(11;14)(p13;q11) or a t(7;11)(q35;p13) translocation suggested a V(D)J recombinase-mediated double strand break at chromosome 11. It was noted from the sequence analysis of three of five patients that the translocation breakpoints fell at exactly the same nucleotide position within the LMO2 gene at chromosome 11, with a heptamer/nonamer-like sequence positioned adjacently. Here, we have used oligomers SCR15/16, which cover the LMO2 breakpoint region (Fig. 2). The oligomers were cloned into the SalI site of pGG49, the vector used for testing V(D)J recombination (see "Experimental Procedures") thereby replacing the 12-signal normally at the SalI site. The consensus 23-signal remained in its normal location on the substrate.

The substrate was transfected into Reh cells and then recovered for analysis. We found that the LMO2 region examined here underwent V(D)J recombination such that 0.34% of the substrates were converted to recombinant product (Fig. 4). This efficiency is 27-fold lower than the recombination efficiency of a standard 12-signal. The signal joint sequences of the recombinant junctions were indistinguishable from those of a standard 12/23 substrate such as pGG49. All of the signal ends of the signal joints were preserved precisely. Nucleotide additions were observed at 37% of the signal joints as determined by sequencing of a randomly chosen subset (data not shown). This feature of signal joint nucleotide addition has been previously documented as a normal feature of V(D)J recombination in murine and human cells when the cells express terminal deoxynucleotidyltransferase (36, 43). Hence, this LMO2 translocation region serves as a target of normal V(D)J recombination, although its efficiency is 27-fold lower than a consensus 12-signal.

View larger version (29K): [in this window] [in a new window]   Fig. 4.   The recombination frequency of the LMO2 breakpoint region of t(11;14)(p13;q11). The LMO2 breakpoint oligomers (SCR15/16) were cloned into pGG49 in place of the 12-signal (boxed triangle) as diagrammed at the top. The dark triangle represents the consensus 23-signal. The resulting pSCR19 was transfected into the Reh cell line, recovered 48 h after transfection, digested with DpnI, and transformed into bacteria. R, recombination frequency (DAC/DA × 100) of the substrate is measured as the ratio of the number of recombined substrate molecules (designated DAC) divided by the number of substrate molecules replicated (designated DA, for DpnI-resistant; a subset of the total number of ampicillin-resistant substrates (designated A). DA/A × 100 represents the percentage of plasmid replication. The recombination frequencies of each transfection into Reh cells obtained after multiple transformations of E. coli are presented as separate rows. The standard deviation is not calculated because the weighted average recombination frequency, R' (see "Experimental Procedures") was calculated as the total DAC of all transfections divided by the total DA of all transfections.

Ttg-1 Serves as a 12-signal but at Low Efficiency-- In this study we have cloned the oligomers (SCR7/8) that span the breakpoint region of the Ttg-1 gene of chromosome 11, which is known to be associated with T-ALL due to the lymphoid translocation t(11;14)(p15;q11) (Fig. 2). This breakpoint region possesses a consensus heptamer, CACAGTG; however, there is no obvious nonamer. This breakpoint has been predicted to function as a 23-signal because it appears to synapse at a 12-signal on chromosome 14 in the patient translocations (Fig. 1). Transfection with the Ttg-1 construct pSCR8 yielded recombinant products in most, but not all, transfections (Fig. 5). The ApaLI digestion results showed that all examined recombinants were sensitive to digestion, indicating precise joining (which was further confirmed by sequencing; data not shown). In brief, Ttg-1 can function as a 23-signal. The recombination frequency is 20-fold lower than for LMO2 and about 530-fold lower than for a normal V(D)J 23-signal.

View larger version (21K): [in this window] [in a new window]   Fig. 5.   The recombination frequency of Ttg-1 breakpoint region of t(11;14)(p15;q11). The Ttg-1 breakpoint containing oligomers (SCR7/8) were cloned into pGG49 after replacing the 23-signal (boxed). The open triangle represents the consensus 12-signal to which Ttg-1 is paired. The resulting pSCR8 is transfected into lymphoid cells (see "Experimental Procedures"). The recombination frequency, R, is calculated as DAC/DA × 100 (see Fig. 4 legend for further details).

The Hox11 Fragile Site of the t(10;14) Translocation Recombines at Least 20,000-Fold Less Efficiently than Consensus Signals-- Studies from T-cell acute lymphoblastic leukemia patients carrying the t(10;14)(q24;q11) chromosomal translocation showed that the breakpoints at chromosome 10q24 are located at the Hox11 gene. We used a set of oligomers, SCR13/14, spanning the Hox11 breakpoint (Fig. 2) corresponding to three of the six patients for whom the breakpoint has been sequenced (21, 22).

Based on the naturally occurring patient DNA sequences, this region of the Hox11 gene appears to act as a 23-signal when it recombines with the 12-signal on chromosome 14. The Hox11 oligomers were cloned into pGG49, and the construct, pSCR17, was transfected into Reh cells. However, multiple transfections and repeated transformations showed no evidence of recombination despite analysis of 246,000 replicated plasmid molecules (data not shown). Hence, the recombination frequency is <0.00041%, which is 830-fold lower than for LMO2 and 22,000-fold lower than for a V(D)J recombination 23-signal.

Because no recombinants were detected, we tested the possibility that Hox11 acts as a 12-signal. However, after multiple transfections of the corresponding substrate, pSCR29, no recombinant molecules could be detected (data not shown). Based on this finding, the recombination frequency was <0.00047% for the Hox11 as a 12-signal. These studies indicate that Hox11 may be recombining with a much lower efficiency than is detectable.

PCR Assay to Determine the Recombination Efficiencies of LMO2, Ttg-1, and Hox11-- Using a PCR strategy, we independently assayed the recombination of LMO2, Ttg-1, and Hox11. The PCR strategy involved the amplification of the recovered DNA. One of the primers was placed at the 23-signal sequence, and the other one was positioned outside of the 12-signal. Amplification of recombined products would yield a band of ~250 bp, whereas unrecombined molecules would yield a PCR fragment of ~500 bp. The detectability of the ~250 bp band was improved by reamplification of PCR products in a second round of PCR in the presence of [-³²P]dCTP.

The results showed that in the case of LMO2 (three of three independent harvests of transfection products) and Ttg-1 (one of two harvests of transfection products) could yield the expected 250-bp band, confirming our earlier results (data not shown). However, for Hox11, we still could not detect any band at the 250-bp size even after reamplification in presence of [-³²P]dCTP (data not shown). Therefore, PCR did not provide any greater sensitivity than the genetic assay.

Analysis of the Mechanism of the Interstitial Deletion at Chromosome 1p32 between SIL and SCL-- Earlier studies by Aplan et al. (24) described an illegitimate V(D)J recombination event in the interstitial deletion at chromosome 1p32. Both SIL and SCL possess heptamer-like sequences. However, there is no obvious nonamer present at the SIL site, and the possible nonamers at the SCL site could be either 12 or 23 bp from the proposed heptamer. Because in these cases we were not sure which breakpoint would act as the 12- versus 23-signal, we tested the oligomers as both 12- and 23-signals.

Our results showed that pSCR15 (SIL as a 23-signal) does yield recombinants (Fig. 6A). The recombination frequency of 0.012% was 750 times lower than that of a consensus pair of 12- and 23-signals. A total of 52 recombinant molecules was obtained from independent transfections after multiple transformations. Of that number, 33 colonies were analyzed by restriction digestion analysis. The ApaLI digestion showed that 26 (79%) colonies possessed precisely joined signal ends resulting in ApaLI sensitivity, whereas 7 (21%) were joined with modification at the junction making them resistant to ApaLI cleavage. Sequencing showed that joining occurred with insertion of 4-6 nucleotides, a normal feature of V(D)J recombination, as mentioned above. When SIL was cloned as a 12-signal (pSCR12), it did not give any positive recombinants after transfection (Fig. 6B). Here, the recombination frequency was less than 0.00036%, which is 25,000-fold lower than that of a consensus V(D)J recombination 23-signal. These results show that during SIL/SCL interstitial deletion, SIL acts as a 23-signal and does so with a moderate recombination efficiency.

View larger version (27K): [in this window] [in a new window]   Fig. 6.   The recombination frequency of the SIL breakpoint region of the SIL/SCL interstitial deletion. A, the SIL breakpoints containing oligomers SCR11/12 were cloned into pGG49 in place of the 23-signal to generate a pSCR15. B, the SCR3/4 was cloned as a 12-signal to generate pSCR12. The recombination frequencies of each transfection (obtained after multiple transformations) are presented as separate rows (see Fig. 4 legend for more details).

We tested the possibility of SCL acting as a 12-signal by transfecting the plasmid pSCR10 into Reh cell lines. The results showed that the recombination efficiency was much lower than that of SIL (<0.00031%) (Fig. 7). We could not detect any positive recombinants after screening 326,000 replicated DNA molecules. In this case, the recombination frequency is at least 29,000-fold lower than a consensus signal. We also tested SCL as a 23-signal (pSCR13), and again could not detect any positive recombinants. Here the recombination frequency was less than 0.00056% (Fig. 7).

View larger version (21K): [in this window] [in a new window]   Fig. 7.   Comparison of recombination efficiency of different translocation breakpoint regions with consensus V(D)J recombination signals. R' is the fraction of the replicated molecules that underwent V(D)J recombination (DAC/DA × 100). The recombination frequency in each lane was calculated from multiple transfections (after repeated transformations). 12 + 23 indicates the optimal 12- and 23-signals of pGG49 in which recombination leads to signal joint formation. In all other lanes, the breakpoint spanning oligomers of the gene concerned (represented by boxes) was paired with one of the consensus signals, except in SCL + SIL where both of the signals were replaced. To more readily convey the relative magnitudes of the recombination frequencies, the recombination efficiency of pGG49 is normalized to 20,000, and other values are given in proportion.

From these studies, we infer that the SIL/SCL interstitial deletion occurs through a V(D)J mechanism where SIL acts as 23-signal and SCL may act as a 12-signal. Because we could not detect positive recombinants of SCL as a 12-signal, we constructed a plasmid containing both SIL and SCL (pSCR27) to test for any special advantage that these two regions might have for recombination with one another. However, in this configuration we still could not detect any recombinant clones (data not shown). Here the recombination frequency was less than 0.00044% (Fig. 7), which is at least 20,000-fold lower than consensus 12- and 23-signals.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Our study shows that mammalian episomes carrying the breakpoint regions of LMO2 and Ttg-1 translocations or the SIL interstitial deletion site are capable of undergoing V(D)J recombination. The recombination frequency is different for each breakpoint site (Fig. 7). The LMO2 site recombines with the highest efficiency, 0.34%, which is only 27-fold less than a consensus V(D)J recombination 12-signal efficiency. The Ttg-1 breakpoint recombines 530-fold lower than consensus signals, and the Hox11 site recombines at least 20,000-fold less efficiently than consensus signals. Our study also shows that the SIL fragile site functions as a 23-signal with a recombination efficiency of 0.012%. SCL presumably acts as a 12-signal, but the efficiency is at least 29,000-fold lower than consensus 12-signals.

The reason that the recombination frequencies of Hox11 and SCL are so low is unclear. The nonamer sequences of the Hox11 site deviate strongly from the consensus for V(D)J recombination, and the nonamer of the SCL site examined here also deviates from the consensus. In addition, the heptamer of both Hox11 and SCL is [coding end]/CACAGCC (22, 24), which deviates at the distal two heptamer positions from the [coding end]/CACAGTG consensus (8). The absence of any definitive proof that Hox11 or SCL can be targets of V(D)J recombination means that if these sites are cleaved by the RAG complex, it occurs at extremely low efficiencies. Alternatively, there remains the possibility that these sites may be cleaved by a non-RAG mechanism.

SIL Acts as a 23-Signal Rather than as a 12-Signal-- Our data show that SIL acts as a 23-signal when paired with a consensus 12-signal. Consistent with this observation, it did not recombine when paired with a 23-signal. These results suggest that during the SIL/SCL interstitial deletion, SIL acts as a 23-signal and SCL presumably acts as a 12-signal.

Our results provide some clarification regarding the secondary breakpoints within the SCL locus (34, 35). Of 63 SIL/SCL deletions that were sequenced previously (34, 35), 55 occurred at the site assayed here. An additional 8 occurred at a secondary site located 1.7 kilobase pairs more proximal to SIL. This secondary site has clearer sequence features being a 12-site. Our experimental results showing that the SIL site functions as a 23-signal would be consistent with it synapsing with 12-signals. These results raise the question of why the secondary SCL site (not tested here) might have a sequence closer to consensus and yet appear 7-fold less frequently in T-ALL. Clearly, two additional factors, chromatin accessibility and growth advantage, will affect the frequency of a site being seen (see below). Presumably these two additional factors must favor the predominant SCL breakpoint site (the one tested in this study).

One previous study tested one SIL-bearing substrate in murine cells in combination with a consensus 23-signal (12). However, the focus of that interesting study was on the identification of cryptic sites in a random stretch of DNA (the plasmid backbone). Only a limited effort was directed at analyzing the SIL site; in fact, only one recombinant colony was identified in a test of SIL as a 12-signal. Furthermore, the SIL region studied was too short to include the length of SIL DNA that might serve as a nonamer for 23-signal. Our study provides substantial data showing that SIL functions instead as a 23-signal.

Although SIL acts as a 23-signal when paired with a consensus 12-signal, we could not detect any recombinants when pSCR27, a plasmid containing SCL and SIL in place of both the 12- and the 23-signals, was tested. A likely reason for this is that the efficiency of SIL and SCL individually is very low when each is paired with an optimal signal. Hence, the efficiency of SIL and SCL together is likely to be even lower.

Efficiency of Recombination Relative to the Frequencies of the Corresponding Leukemias-- At least two other factors affect the frequency of these chromosomal rearrangements in addition to misrecognition by the RAG complex. One additional factor is the local chromatin structure. CpG methylation can dictate a chromatin structure that is markedly resistant to V(D)J recombination (44, 45). Methylated regions have deacetylated nucleosomes and are transcriptionally inactive (46, 47). Because the transcriptional activity correlates with active chromatin, it is useful to note which of the chromosomal breakpoints are within transcriptionally active regions in the genome. LMO2 and SIL are both actively transcribed in the T-cells in which the translocations occur, whereas Ttg-1, Hox11, and SCL are inactive (see the Introduction).

A second additional factor is the growth advantage conferred by the translocation. Transgenic mice that overexpress the gene affected by the translocation manifest varying efficiencies of leukemogenesis. In one study, the CD2 promoter was used to drive either LMO2 or Ttg-1 in transgenic mouse models (48). The percentage of mice succumbing to leukemia was higher for the LMO2 transgenic mice than for the Ttg-1 mice. This type of study gives some sense of the probability of neoplasia conferred by corresponding levels of expression. Of course, in the actual translocations, the levels of expression may differ.

The three factors discussed here may balance one another to give leukemogenesis rates that appear to be within one to two orders of magnitude of one another. The contribution of our study is to provide quantitation to the V(D)J recombination component, which has previously only been subject to conjecture.

	ACKNOWLEDGEMENTS

We thank Dr. Chih-Lin Hsieh and Dr. Frederic Chedin for comments on the manuscript, and we thank members of the Lieber laboratory for helpful comments during the course of the study.

	FOOTNOTES

^* This research was supported by National Institutes of Health grants (to M. R. L.).The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

^¶ The Rita and Edward Polusky Basic Cancer Research Professor. To whom correspondence should be addressed: University of Southern California Keck School of Medicine, Rm. 5428, 1441 Eastlake Ave., Los Angeles, CA 90089-9176. Tel.: 323-865-0568; Fax: 323-865-3019; E-mail: lieber@usc.edu.

Published, JBC Papers in Press, June 4, 2001, DOI 10.1074/jbc.M103797200

	ABBREVIATIONS

The abbreviations used are: RAG, recombination activating gene; bp, base pair(s); T-ALL, T-cell acute lymphoblastic lymphoma; PCR, polymerase chain reaction; 12-signal, recombination signal having a 12-bp spacer between the heptamer and nonamer; 23-signal, recombination signal having a 23-bp spacer between the heptamer and nonamer; SCL, stem cell leukemia; SIL, SCL-interrupting locus.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

This article has been cited by other articles:

				M. Zhang and P. C. Swanson V(D)J Recombinase Binding and Cleavage of Cryptic Recombination Signal Sequences Identified from Lymphoid Malignancies J. Biol. Chem., March 14, 2008; 283(11): 6717 - 6727. [Abstract] [Full Text] [PDF]

				J. D. Curry, D. Schulz, C. J. Guidos, J. S. Danska, L. Nutter, A. Nussenzweig, and M. S. Schlissel Chromosomal reinsertion of broken RSS ends during T cell development J. Exp. Med., October 1, 2007; 204(10): 2293 - 2303. [Abstract] [Full Text] [PDF]

				W. A. Dik, B. Nadel, G. K. Przybylski, V. Asnafi, P. Grabarczyk, J. M. Navarro, B. Verhaaf, C. A. Schmidt, E. A. Macintyre, J. J. M. van Dongen, et al. Different chromosomal breakpoints impact the level of LMO2 expression in T-ALL Blood, July 1, 2007; 110(1): 388 - 392. [Abstract] [Full Text] [PDF]

				J. M. Murray, J. P. O'Neill, T. Messier, J. Rivers, V. E. Walker, B. McGonagle, L. Trombley, L. G. Cowell, G. Kelsoe, F. McBlane, et al. V(D)J Recombinase-Mediated Processing of Coding Junctions at Cryptic Recombination Signal Sequences in Peripheral T Cells during Human Development J. Immunol., October 15, 2006; 177(8): 5393 - 5404. [Abstract] [Full Text] [PDF]

				M. S. Schlissel, C. R. Kaffer, and J. D. Curry Leukemia and lymphoma: a cost of doing business for adaptive immunity. Genes & Dev., June 15, 2006; 20(12): 1539 - 1544. [Full Text] [PDF]

				A. Allam and D. Kabelitz TCR trans-Rearrangements: Biological Significance in Antigen Recognition vs the Role as Lymphoma Biomarker J. Immunol., May 15, 2006; 176(10): 5707 - 5712. [Abstract] [Full Text] [PDF]

				S. C. Raghavan, P. C. Swanson, Y. Ma, and M. R. Lieber Double-Strand Break Formation by the RAG Complex at the Bcl-2 Major Breakpoint Region and at Other Non-B DNA Structures In Vitro Mol. Cell. Biol., July 15, 2005; 25(14): 5904 - 5919. [Abstract] [Full Text] [PDF]

				S. C. Raghavan, P. Chastain, J. S. Lee, B. G. Hegde, S. Houston, R. Langen, C.-L. Hsieh, I. S. Haworth, and M. R. Lieber Evidence for a Triplex DNA Conformation at the bcl-2 Major Breakpoint Region of the t(14;18) Translocation J. Biol. Chem., June 17, 2005; 280(24): 22749 - 22760. [Abstract] [Full Text] [PDF]

				H. Tsuji, H. Ishii-Ohba, T. Katsube, H. Ukai, S. Aizawa, M. Doi, K. Hioki, and T. Ogiu Involvement of Illegitimate V(D)J Recombination or Microhomology-Mediated Nonhomologous End-Joining in the Formation of Intragenic Deletions of the Notch1 Gene in Mouse Thymic Lymphomas Cancer Res., December 15, 2004; 64(24): 8882 - 8890. [Abstract] [Full Text] [PDF]

				S. C. Raghavan, S. Houston, B. G. Hegde, R. Langen, I. S. Haworth, and M. R. Lieber Stability and Strand Asymmetry in the Non-B DNA Structure at the bcl-2 Major Breakpoint Region J. Biol. Chem., October 29, 2004; 279(44): 46213 - 46225. [Abstract] [Full Text] [PDF]

				G. S. Boehden, A. Restle, R. Marschalek, C. Stocking, and L. Wiesmuller Recombination at chromosomal sequences involved in leukaemogenic rearrangements is differentially regulated by p53 Carcinogenesis, August 1, 2004; 25(8): 1305 - 1313. [Abstract] [Full Text] [PDF]

				Y. Kitagawa, K. Inoue, S. Sasaki, Y. Hayashi, Y. Matsuo, M. R. Lieber, H. Mizoguchi, J. Yokota, and T. Kohno Prevalent Involvement of Illegitimate V(D)J Recombination in Chromosome 9p21 Deletions in Lymphoid Leukemia J. Biol. Chem., November 22, 2002; 277(48): 46289 - 46297. [Abstract] [Full Text] [PDF]

				J. L. Wiemels, B. C. Leonard, Y. Wang, M. R. Segal, S. P. Hunger, M. T. Smith, V. Crouse, X. Ma, P. A. Buffler, and S. R. Pine Site-specific translocation and evidence of postnatal origin of the t(1;19) E2A-PBX1 fusion in childhood acute lymphoblastic leukemia PNAS, November 12, 2002; 99(23): 15101 - 15106. [Abstract] [Full Text] [PDF]

				R. Marculescu, T. Le, P. Simon, U. Jaeger, and B. Nadel V(D)J-mediated Translocations in Lymphoid Neoplasms: A Functional Assessment of Genomic Instability by Cryptic Sites J. Exp. Med., January 7, 2002; 195(1): 85 - 98. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: 276/31/29126    most recent M103797200v1