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


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nakajima, P. B.
Right arrow Articles by Bosma, M. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nakajima, P. B.
Right arrow Articles by Bosma, M. J.
The Journal of Immunology, 2002, 169: 3094-3104.
Copyright © 2002 by The American Association of Immunologists

Variable Diversity Joining Recombination: Nonhairpin Coding Ends in Thymocytes of SCID and Wild-Type Mice1

Pamela B. Nakajima2 and Melvin J. Bosma

Institute for Cancer Research, Fox Chase Cancer Center, Philadelphia, PA 19111


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Initiation of V(D)J recombination results in broken DNA molecules with blunt recombination signal ends and covalently sealed (hairpin) coding ends. In SCID mice, coding joint formation is severely impaired and hairpin coding ends accumulate as a result of a deficiency in the catalytic subunit of DNA-dependent protein kinase, an enzyme involved in the repair of DNA double-strand breaks. In this study, we report that not all SCID coding ends are hairpinned. We have detected open J{delta}1 and D{delta}2 coding ends at the TCR{delta} locus in SCID thymocytes. Approximately 25% of 5'D{delta}2 coding ends were found to be open. Large deletions and abnormally long P nucleotide additions typical of SCID D{delta}2-J{delta}1 coding joints were not observed. Most J{delta}1 and D{delta}2 coding ends exhibited 3' overhangs, but at least 20% had unique 5' overhangs not previously detected in vivo. We suggest that the SCID DNA-dependent protein kinase deficiency not only reduces the efficiency of hairpin opening, but also may affect the specificity of hairpin nicking, as well as the efficiency of joining open coding ends.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Assembly of Ag receptor genes is accomplished by a site-specific recombination process in which one of many V gene segments is joined to any one of an array of J, and in some cases D, gene segments encoded at TCR and Ig loci (reviewed in Refs. 1 and 2). V(D)J recombination is targeted byrecombination signal sequences (RSSs)3 that lie adjacent to V, D, and J coding segments. RSS consist of a conserved heptamer and nonamer motif separated by a spacer of 12 or 23 bp. A pair of RSS with unlike spacers serves to direct a single recombination reaction and to ensure assembly of a biologically useful product.

The first phase of V(D)J recombination consists of recognition of a pair of RSS, synapsis of the 12- and 23-RSS, and cleavage at the signal-coding borders. Both recognition and cleavage steps are mediated by the lymphoid-specific recombination activating gene (RAG) products, RAG1 and RAG2 (3, 4, 5). The cleavage step produces two types of broken molecules corresponding to V(D)J recombination intermediates: those with coding ends and those with signal ends (6, 7). As a consequence of the cleavage mechanism, coding ends are covalently sealed to form a DNA hairpin (4, 5). In contrast, signal ends are almost always blunt with no sequence loss or addition (8, 9).

The second phase involves the processing and joining of DNA ends, and includes hairpin opening, insertions and deletions at the open coding ends, and formation of signal and coding joints. One feature unique to the processing of coding ends, and observed at junctional sequences of joined coding segments, is the addition of palindromic (P) nucleotides (10, 11). Such additions correspond to the gain of several nucleotides complementary to the last few nucleotides of a full-length coding segment. Addition of P nucleotides is thought to reflect the mechanism whereby hairpin coding ends are opened (reviewed in Ref. 12). Nicking by a single-strand specific endonuclease at a site either 3' or 5' of the hairpin tip would produce P nucleotide overhangs. The second phase of V(D)J recombination also depends upon the function of ubiquitous factors that participate in DNA nonhomologous end joining (13). These include DNA-dependent protein kinase (DNA-PK), composed of a catalytic subunit (DNA-PKcs) associated with a DNA end-binding Ku70/Ku86 heterodimer (reviewed in Ref. 14), and the XRCC4 gene product and DNA-ligase IV, which are involved in DNA end-joining (15, 16, 17, 18).

The critical role of DNA-PK in V(D)J recombination is clear from the phenotype of SCID mice (reviewed in Ref. 19). These mice are extremely impaired in the joining of coding ends and lack functional B and T cells. Hairpin coding ends accumulate in developing SCID B and T cells (7), and rare coding joints often exhibit abnormal nucleotide loss or long P nucleotide additions (20, 21, 22, 23). The scid defect is caused by a nonsense mutation in DNA-PKcs that results in a truncated protein with little or no detectable kinase activity (24, 25, 26, 27). The exact role of DNA-PKcs in the resolution of hairpin coding ends is unclear. However, one model suggests that the hairpins are normally sequestered in unique DNA-protein complexes, and in SCID, these remain unavailable for opening due to the DNA-PK deficiency (28). Efficient coding end processing may require DNA-PK to disassemble or remodel the postcleavage complex. Another proposal is that DNA-PK recruits and/or activates the nuclease that opens hairpin coding ends (25).

DNA hairpin opening activities have been observed in vitro for the Mre11/Rad50/Nbs1 complex (29), which functions in the cellular response to DNA damage (30, 31) for the RAG1/RAG2 proteins (32, 33) and most recently, for the Artemis protein (34). Because the SCID DNA-PK deficiency is not known to affect the expression of any of these proteins, it was of interest to re-examine whether a portion of coding ends are open in SCID mice. Using a sensitive technique to detect broken molecules at the TCR{delta} locus in unmanipulated SCID and wild-type (wt) thymocytes, we found that a fraction of SCID coding ends are open. Most open SCID coding ends were structurally similar to those with 3' overhangs in wt thymocytes; however, a subpopulation of open coding ends unique to SCID had 5' overhangs. This subpopulation, which we also detected in thymocytes with a targeted disruption of the DNA-PKcs gene, has not previously been detected in vivo. We suggest that: 1) coding ends with 5' overhangs may reflect altered hairpin opening activity as a result of the DNA-PK deficiency; and 2) in the absence of normal DNA-PK activity, open coding ends are not efficiently joined.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Animals

BALB/cAnICR and C.B-17 scid/scid (designated as SCID) mice were bred at the Fox Chase Cancer Center. DNA-PKcs-null mice, generously provided by G. Taccioli (Boston University School of Medicine, Boston, MA), were previously described (35). Fetal mice were obtained from females mated overnight. Day 0 of embryonic development was designated the day the male was removed. Newborn mice were used 1–3 days after birth. Adult mice were 4–6 wk of age.

Preparation of DNA samples and Southern analysis

Genomic DNA samples were prepared from cells embedded in agarose as described (36). For Southern analysis, DNA in half of an agarose block was digested with EcoRI (Life Technologies, Rockville, MD) and subjected to agarose gel electrophoresis as described (36). After gel processing, DNA fragments were transferred to Nytran Plus nylon membrane (Schleicher and Schuell, Keene, NH) by a downward alkaline transfer method (37) and then hybridized to random primed 32P-labeled probes (38) prepared from gel-purified DNA fragments. The DJ and 3'J{delta}1 probes were as described previously (6, 39). The J{delta}1 probe was a 1.4-kb PstI fragment containing J{delta}1 and sequences ~800 bp 5' and 600 bp 3' of J{delta}1. The D{delta}2 probe was a 331 fragment containing D{delta}2 with 5' and 3' flanking sequences amplified from liver DNA with oligonucleotides MB61 and MB125. Hybridization and washing of blots were conducted as described (6). The amount of radioactivity hybridized to DNA fragments was measured with a BAS1000 Mac Bio-imaging Analyzer (Fuji Photo Film, Tokyo, Japan). Quantitative analysis of molecules broken at D{delta}2 or J{delta}1 in genomic DNA was performed as described previously (6, 36).

Exonuclease, endonuclease, and T4 DNA polymerase (T4 pol) treatments

Exonuclease treatment of DNA was performed with 15 U Micrococcus luteus ATP-dependent exonuclease (Amersham Pharmacia Biotech, Piscataway, NJ) as outlined (40). For endonuclease treatment, DNA in one-quarter of an agarose block was equilibrated with endonuclease buffer (10 mM Tris (pH 7.9), 50 mM NaCl, 10 mM MgCl2, 1 mM ZnSO4, 1 mM DTT, and 100 µg acetylated BSA per ml) and incubated with 30 U mung bean nuclease (MBN; NEB, Beverly, MA) for 2 h at 37°C. To convert single-strand extentions to blunt ends, untreated DNA or samples pretreated with MBN were equilibrated on ice with T4 pol buffer containing 50 mM NaCl, 10 mM Tris (pH 7.9), 10 mM MgCl2, 1 mM DTT, and 0.2 mM each of dATP, dTTP, dCTP, and dGTP, then incubated at 37°C for 1 h with 3 U of T4 pol (NEB). After enzyme treatment, plugs were extensively washed with 10 mM Tris (pH 7.5) and 1 mM EDTA.

Oligonucleotide primers and linkers

Specificity and sequences of the TCR{delta} primers used are as follows: MB59 (3'J{delta}1), 5'-AAAAAGCTTACTCAACACGACTGGA; MB61 (5'D{delta}2) 5'-AAAAGATCTGGCCTGAACTAACTGCCA; MB125 (3'D{delta}2), 5'-AGGGCAGGCTGCGGGCTGTGTTTAC; MB145 (3'J{delta}1), 5'-TTAAGCTTTCCCAGGCAATCTTACTCA; and MB218 (5'D{delta}2), 5'-TGGCTTGACATGCAGAAAACACCTG. Sequences for {alpha}-actin primers and conditions to amplify the {alpha}-actin PCR product were as described (41). The oligonucleotide pairs used to make annealed linkers for ligation are as follows: MB216, 5'-CACGAATTCCC and MB217, 5'-GCTATGTACTACCCGGGAATTCGTG (blunt-end linker); MB217 and MB385-388, 5'-(N)2–5CACGAATTCCC (5' overhanging linker); MB485, 5'-CGGGAATTCGTG and MB477-480, 5'-(N)2–5CACGAATTCCCGGGTAGTAC (5'-phosphorylated, overhanging linker); MB301, 5'-CACGAATTCCCGGGTAGTACATAGC and MB381-384, 5'-GGGAATTCGTG(N)2–5 (3' overhanging linker). MB301 and MB477-480 were 5'-phosphorylated with ATP using T4 polynucleotide kinase (NEB). Oligonucleotides for linker ligation were annealed as described (42). The oligonucleotide MB481, 5'-GTACTACCCGGGAATTCGTG, was used with 3'J{delta}1 primers to amplify coding ends ligated to MB477-480.

Ligation-mediated PCR (LMPCR)

Linker-ligation. This procedure was performed as described (8, 9, 43) with minor modifications. Briefly, one-quarter of an agarose block containing untreated or pretreated DNA was equilibrated with ligation buffer containing 5 mM MgCl2, 50 mM Tris (pH 7.6), 5 mM DTT, 1 mM hexamine cobalt, 1 mM ATP, and 0.05 mg acetylated BSA per ml. Ligation was done with 50 µl ligation buffer containing 4 µM annealed ligation linkers and 3 U T4 DNA ligase (Life Technologies). For linker ligation to untreated DNA with overhanging ends, linkers with degenerate 5' or 3' overhangs two to five nucleotides in length were used as described (43). Following overnight ligation at 16°C, blocks were equilibrated with PCR buffer (50 mM Tris (pH 8.3), 10 mM KCl, 2 mM MgCl2, 0.1% Triton X-100, and 0.001% gelatin) and then heated 10 min at 95°C in 50 µl PCR buffer.

LMPCR amplification. Linker-ligated DNA was amplified in a 50 µl PCR volume containing one-tenth of the ligation reaction, 0.2 mM each of dATP, dTTP, dCTP, and dGTP, 1.25 U of AmpliTaq DNA polymerase (PerkinElmer, Wellesley, MA), and 25 pmol each of the appropriate primers. J{delta}1-associated broken ends were amplified with primers MB217 and MB145 using conditions of 20 s at 94°C, 30 s at 58°C, and 30 s at 72°C for 10 cycles, then 20 s at 94°C, 30 s at 68°C, and 30 s at 72°C for 16 cycles. The same cycle conditions were used to amplify D{delta}2-associated broken ends with primer MB217 and either the 3'D{delta}2-specific primer, MB125, or the 5'D{delta}2-specific primer, MB61. To characterize the 1.1-kb product, LMPCR was first performed with the 3'J{delta}1-specific primer (MB145) for 15 cycles, then gel-purified products ranging in size from 1.0–1.2 kb were subjected to 25 cycles of LMPCR using primer MB125.

Assays for 5' and 3' overhanging coding ends shown in Fig. 5GoC used overhanging linkers and seminested PCR. Primers MB145 and MB481 were used with conditions of 20 s at 94°C, 30 s at 58°C, and 30 s at 72°C for 15 cycles, then 1 µl of the first PCR was amplified for 23 cycles using primers MB59 and MB481 with conditions of 20 s at 94°C, 30 s at 54°C, and 30 s at 72°C for 9 cycles, then 20 s at 94°C, 30 s at 65°C, and 30 s at 72°C for 14 cycles. LMPCR products were subjected to electrophoresis through 1% SeaKem agarose or through 6% polyacrylamide gels, then transferred to membranes by Southern blotting or electroblotting, respectively.



View larger version (17K):
[in this window]
[in a new window]
  		FIGURE 5. Nucleotide sequences of broken D{delta}2 coding ends amplified by LMPCR from wt (+/+) fetal thymus DNA pretreated with T4 pol. Top line shows the germline nucleotide sequence of the D{delta}2 gene segment (uppercase) and its flanking 5' and 3' heptamers (lowercase). The frequency specifies the number of clones with the indicated sequence. Number of nucleotides deleted from each coding end is indicated in parentheses. A, Nucleotide sequences of wt 5'D{delta}2 coding ends. Boldface shows nongermline encoded sequence additions. Deletions of 5'D{delta}2 coding ends ranged from 2–6 bp; a pattern similar to that shown in Fig. 4GoC for D{delta}2 coding segments rearranged to J{delta}1. B, Nucleotide sequences of wt 3'D{delta}2 coding ends. Cloned 3'D{delta}2 coding end products showed deletions ranging from 2–11 bp, with the loss of 4 bp exhibited by the majority.

 
Semiquantitative analysis of LMPCR products

Relative levels of J{delta}1-associated breaks in fetal and newborn thymocytes were determined by measuring the amount of 32P-labeled J{delta}1 probe hybridized to the 200-bp LMPCR product, and DNA sample levels were evaluated by measuring products amplified with {alpha}-actin primers (41). Levels of open and hairpin 5'D{delta}2 coding ends in SCID thymocytes were determined by measuring the amount of 32P-labeled D{delta}2 probe hybridized to the 155-bp coding end product amplified from DNA pretreated with T4 pol only or with both MBN and T4 pol. The amount of radioactivity associated with the 139-bp 3'D{delta}2 signal end product was used to adjust for differences in sample loads.

Cloning and sequencing of coding and signal end products

5'D{delta}2 coding, 3'D{delta}2 signal, and 5'J{delta}1 coding end products for cloning were obtained as follows. One-tenth of a linker ligation reaction was amplified for 15 cycles with primers MB145 and MB217. The PCR products were electrophoresed through 3% NuSieve agarose (BioWhittaker, Walkersville, MD). Gel sections containing DNA of ~1.1 kb and 200 bp were excised and the DNA recovered using {beta}-agarase digestion as recommended by the manufacturer (NEB). Gel-purified 200-bp DNA was amplified by a second PCR using primers MB217 and MB59 with conditions of 20 s at 94°C, 30 s at 55°C, and 30 s at 72°C for 10 cycles, then 20 s at 94°C, 30 s at 65°C, and 30 s at 72°C for 15 cycles. To recover 5' coding and 3' signal ends of unrearranged D{delta}2 segments, gel-purified 1.1-kb DNA was reamplified with primers MB217 and MB125 using conditions described above. To amplify 3'D{delta}2 coding, and 5'D{delta}2 signal end products for cloning, LMPCR was performed with linker-ligated DNA and the primer pair MB217 and MB61. Gel-purified products of 0.1–0.3 kb were amplified in a secondary PCR using primers MB218 and MB217 for 25 cycles of 15 s at 94°C, 30 s at 68°C, and 30 s at 72°C. Clones for DNA sequence analysis were produced using the TA or TOPO cloning kits (Invitrogen, San Diego, CA). DNA sequences were obtained from cycle sequencing reactions of plasmid DNA that were analyzed with a Model 377 genetic analyzer (Applied Biosystems, Foster City, CA).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
In previous analyses of EcoRI-digested DNA from SCID thymocytes, we identified several nongermline species corresponding to double-strand breaks at the TCR{delta} locus (7). As illustrated in Fig. 1GoA, these species correspond to molecules with signal ends and hairpin coding ends. Similar analyses of thymus DNA from newborn wt mice revealed broken molecules with signal, but not coding ends (6, 7). In this study, more sensitive techniques were used to detect molecules with open coding ends, first in wt and then in SCID thymocytes. The structure of these ends was then analyzed and compared.



View larger version (28K):
[in this window]
[in a new window]
  		FIGURE 1. Broken molecules at the TCR{delta} locus identified in thymocyte DNA. A, Map shows relevant EcoRI (E) restriction sites in BALB/c genomic DNA with filled bars above representing sequences used as hybridization probes. Coding elements are depicted by rectangles and RSS elements containing 12 or 23 nucleotide spacers are denoted by open and filled triangles, respectively. Cleaved species with either coding or signal ends are depicted as products of EcoRI digestion; the expected sizes of the restriction fragments are indicated. B, Genomic DNA preparations were either digested with EcoRI (adult liver, lane 1; fetal thymus, lane 2) or treated with exonuclease before EcoRI digestion (fetal thymus, lane 3). Fragments were separated by electrophoresis, blotted, and hybridized to a 3'J{delta}1 probe. Lengths of relevant fragments are indicated on the left and a schematic of the 4-kb species produced by EcoRI digestion is shown below. C, EcoRI-digested genomic DNA from adult liver (lane 1), newborn wt (wt, lane 2), fetal wt (wt, lane 3), and newborn SCID (lane 4) thymus were subjected to electrophoresis, blotted, and hybridized to a DJ probe. Lengths of relevant fragments are indicated on the left and a schematic of the 4.9-kb species produced by EcoRI digestion is shown below.

 
Nonhairpin (open) coding ends in fetal wt thymocytes

To test for molecules broken near J{delta}1, a Southern blot of EcoRI-digested DNA from fetal wt thymocytes was hybridized to a 3'J{delta}1 probe. We used day 16 fetal thymocytes because cleavage events at the TCR{delta} locus are more abundant in fetal than newborn wt mice (our unpublished observations). As shown in Fig. 1GoB (lane 2), a 7.4-kb germline fragment and multiple nongermline fragments, including one of ~4 kb, were detected. This 4-kb fragment, detected in similar assays of independently prepared fetal wt thymocyte DNA (data not shown), corresponded to ~0.5% of the J{delta}1 hybridizing signal (see Materials and Methods for details of quantitative analysis). Treatment of fetal thymocyte DNA with ATP-dependent exonuclease before EcoRI digestion (Fig. 1GoB, lane 3) specifically destroyed the 4-kb fragment but not other nongermline fragments which represent completed D-J and V-D-J rearrangements (36, 44). This result suggested that the 4-kb species is primarily derived from broken molecules with nonhairpinned, unblocked termini.

From Southern analysis (Fig. 1GoB), identification of a 4.9-kb fragment (depicted in Fig. 1GoA) was not possible because of the complex pattern of 3'J{delta}1-hybridizing nongermline fragments. However, hybridization to a DJ probe, specific for intervening sequences between D{delta}2 and J{delta}1, revealed a fragment in both EcoRI-digested fetal (Fig. 1GoC, lane 3) and newborn (Fig. 1GoC, lane 2) thymocyte DNA that comigrated with the 4.9-kb fragment observed in SCID thymus DNA (Fig. 1GoC, lane 4). Whether the 4.9-kb fragment from wt thymocytes included both D{delta}2 signal and coding end species (Fig. 1GoA) could not be directly determined in this analysis since these two species differ in size by only 16 nucleotides, the size of the D{delta}2 coding element. Identification and structural characterization of D{delta}2 and J{delta}1 wt coding ends are presented later.

Open coding ends from wt thymocytes show blunt and nonblunt termini

To characterize the broken molecules identified in fetal wt thymocytes further, we used LMPCR (8, 9). In this assay, blunt, 5'-phosphorylated broken ends are ligated with a partially double-stranded oligonucleotide, then amplified by PCR using the ligation primer and a locus-specific primer. To assay for double-strand breaks near J{delta}1, we used a primer specific for a sequence ~200-bp 3' of J{delta}1. As shown in Fig. 2GoA (lane 3), a 200-bp LMPCR product that hybridized to a J{delta}1 probe was obtained from fetal thymocytes. No such product was recovered from assays of liver DNA (Fig. 4GoA, lane 1). Although not visible in Fig. 2GoA, the 200-bp product was also detected in newborn thymus (lane 2) after prolonged blot exposure and was at least 20-fold less abundant than that recovered from fetal thymus. Sequence analysis of the 200-bp products (see Fig. 4Go) revealed two general populations: those corresponding to J{delta}1 coding ends and those corresponding to 5'D{delta} coding ends derived from a completed D{delta}-J{delta}1 rearrangement. A prominent 1.1-kb product was amplified from both fetal and newborn thymocyte samples. When the membrane was stripped and hybridized to the DJ probe, only the 1.1-kb product was observed (Fig. 2GoB), suggesting that it corresponded to molecules unrearranged at J{delta}1 and broken at D{delta}2 (Fig. 1GoA).



View larger version (75K):
[in this window]
[in a new window]
  		FIGURE 2. LMPCR analysis of broken molecules in wt thymocytes. A, Genomic DNA from liver (lane 1), newborn (nb; lane 2) or fetal (ft; lane 3) wt thymocytes was directly linker-ligated and assessed for molecules broken near J{delta}1 by LMPCR (26 cycles) with primers MB217 and MB145. Products were separated by electrophoresis, blotted, and hybridized to the J{delta}1 probe. Broken molecules corresponding to LMPCR products are depicted on the left with filled rectangles, indicating the ligation linker and arrows indicating PCR primers. Sizes of the products of interest are on the right. Sequence analysis showed that the 300-bp product from newborn thymus corresponded to a completed DV10S7-D{delta}2-J{delta}1 rearrangement cleaved at a heptamer sequence (5'-CACAGTC) embedded within the V gene segment (data not shown). B, The membrane shown in A was stripped and rehybridized to the DJ probe. C, DNA from ft or nb wt thymocytes were directly linker-ligated (lanes 4, 6, and 8) or pretreated with T4 pol (lanes 5, 7, and 9) before linker ligation. LMPCR was performed and the products analyzed as for A. Quantitative analysis of several experiments showed that the abundance of blunt coding ends recovered from fetal thymocytes was somewhat variable.

 


View larger version (34K):
[in this window]
[in a new window]
  		FIGURE 4. Nucleotide sequences of open coding end products amplified by LMPCR from SCID and wt thymocyte DNA pretreated with T4 pol. Top line shows the germline nucleotide sequence of the D{delta}1 gene segment (uppercase) and its 3' flanking heptamer (lowercase). Second line shows germline sequence of the D{delta}2 gene segment (uppercase) and its 5' and 3' flanking heptamers (lowercase), and the third line show the germline nucleotide sequence of the J{delta}1 gene segment (uppercase) and its 5' flanking heptamer (lowercase). Below are nucleotide sequences of cloned LMPCR products amplified from fetal SCID (A), adult SCID (B), and fetal wt (+/+) (C) thymus DNA pretreated with T4 pol before linker ligation. The frequency specifies the number of clones with the indicated sequence. Underlined nucleotides denote 5' P overhangs. Number of nucleotides deleted from each coding end is indicated in parentheses.

 
To obtain more structural information about the broken ends, LMPCR assays were repeated with or without pretreating genomic DNA preparations with T4 pol before primer ligation. This polymerase can both fill in 5' overhangs and remove 3' overhangs to produce blunt ligatable ends. Although T4 pol pretreatment enhanced recovery of the 200-bp product from fetal thymus DNA ~6-fold (Fig. 2GoC, compare lanes 4 and 5), the 200-bp product was not readily detectable in assays of two independent newborn thymus DNA samples with or without T4 pol pretreatment (Fig. 2GoC, lanes 6–9). In contrast, the 1.1-kb LMPCR product was easily detected in all LMPCR assays of fetal and newborn thymocytes. To analyze the 1.1-kb product further, secondary LMPCR products were amplified using a 3'D{delta}2-specific primer (MB125, ~115-bp 3' of D{delta}2) and separated by high-resolution PAGE. Recovery of a minor 5'D{delta}2 coding end product was slightly increased by T4 pol pretreatment (our unpublished observations). Thus, the 1.1-kb product from wt primarily corresponds to a 3'D{delta}2 signal end. The basis for the 200-bp coding end product being significantly more abundant in day 16 fetal than newborn thymocytes remains to be explored.

Open coding ends in SCID thymocytes

To determine whether SCID thymocytes contain open TCR{delta} coding ends, the LMPCR assay was repeated. As shown in Fig. 3GoA, pretreatment with both MBN to open hairpins (28), and T4 pol to create blunt ends, allowed J{delta}1 coding ends to be detected (lane 2). Without any DNA pretreatment, only the 1.1-kb signal end product was obtained (Fig. 3GoA, lane 5). Strikingly, pretreatment of two independent fetal SCID thymocyte DNA samples with T4 pol alone (Fig. 3GoA, lanes 3 and 4) resulted in detection of J{delta}1 coding end products that did not appear grossly different in size from wt products (Fig. 3GoA, lane 6, 0.2-kb product). Excessive deletions often exhibited at SCID coding joints (20, 21, 23, 45) were not apparent in the open SCID coding ends amplified with a primer specific for a sequence 200-bp 3' of J{delta}1, although ends with very large deletions would not be amplified by this primer.



View larger version (55K):
[in this window]
[in a new window]
  		FIGURE 3. Presence of open coding ends with overhangs in SCID thymocytes. A, Liver (lane 1) and SCID (lanes 2–5) or wt (lane 6) fetal thymocyte DNA was pretreated with T4 pol only (lanes 1, 3, 4, and 6), pretreated sequentially with MBN and T4 pol (lane 2), or directly linker-ligated (lane 5). Lanes 3 and 4, Independent SCID samples. LMPCR products were analyzed as in Fig. 2GoA. B, High resolution gel analysis of LMPCR coding end products. Adult SCID (ad scid; lanes 7-9) or fetal wt (wt; lane 10) thymocyte DNA samples were directly linker-ligated (lane 7), or pretreated with MBN (lane 8) or T4 pol (lanes 9 and 10) before linker ligation. LMPCR was performed as in Fig. 2GoA and products were separated by PAGE, electroblotted, and hybridized to a J{delta}1 probe. C, LMPCR analysis of 5'D{delta}2 coding ends from SCID thymocyte DNA pretreated with T4 pol (lane 11) or pretreated sequentially with MBN and T4 pol (lane 12). LMPCR was performed with MB125, a 3'D{delta}2-specific primer and products were separated through a 6% polyacylamide gel and hybridized to a D{delta}2 probe. A schematic of the coding (155 bp) and signal (139 bp) end products is given on the right.

 
To assay for small differences in size between SCID and wt open J{delta}1 coding ends, LMPCR products were resolved by PAGE. Coding ends recovered after T4 pol pretreatment of adult SCID thymocyte DNA (Fig. 3GoB, lane 9) appeared slightly smaller than both wt (Fig. 3GoB, lane 10) and full-length SCID coding ends recovered after MBN pretreatment (Fig. 3GoB, lane 8). Similar bulk assays of independently prepared SCID thymocyte DNA showed that most J{delta}1 coding end products were only somewhat smaller than full-length, although minor bands with deletions of ~20–30 bp were obtained from some samples (data not shown).

The relative abundance of open and hairpin coding ends in SCID thymocytes was estimated as follows. Thymocyte DNA was pretreated with MBN and T4 pol or with T4 pol alone and LMPCR products corresponding to 5'D{delta}2 coding and 3'D{delta}2 signal ends were amplified. Products were separated by PAGE, electroblotted, and hybridized to a D{delta}2 probe (Fig. 3GoC). Quantitative analysis (see Materials and Methods for details) showed that recovery of 5'D{delta}2 coding ends after sequential MBN and T4 pol pretreatment (Fig. 3GoC, lane 12) was 4-fold higher than after T4 pol pretreatment alone (Fig. 3GoC, lane 11). This result indicates that ~25% of the 5'D{delta}2 coding ends in SCID thymocytes are open, assuming that all hairpins in D{delta}2 coding ends are opened by MBN.

Presence of novel 5' overhangs in SCID open coding ends

The 200-bp LMPCR products obtained from T4 pol pretreated SCID and wt thymus DNA were cloned and sequenced. The nucleotide sequences of cloned products from SCID and wt thymocyte DNA revealed D{delta}1, D{delta}2, and J{delta}1 coding ends (Fig. 4Go). Products with D{delta}1 or D{delta}2 coding ends represent a completed D{delta}1-D{delta}2-J{delta}1 or D{delta}2-J{delta}1 rearrangement, respectively, cleaved at the 5'D{delta} RSS. Fig. 4GoA shows that the majority of J{delta}1 coding ends (27 of 34 clones) obtained from fetal SCID thymocytes had deletions that ranged from 3–12 nucleotides, only slightly more than observed for wt (Fig. 4GoC). Note that P nucleotides were present in 6 of 34 J{delta}1 and one of two 5' D{delta}2 SCID coding end sequences, for a total P nucleotide frequency of ~20% (7 of 37 total clones or 2 of 9 unique clones). A slightly higher percentage (30–40%) of coding ends with P nucleotides (10 of 31 total clones or 4 of 9 unique clones) was observed with adult SCID clones (Fig. 4GoB). The recovery of clones with P nucleotides after T4 pol treatment indicated polymerase fill in of coding ends with 5' overhangs. No full-length coding ends with 5' P overhangs were observed in wt thymocytes. As shown in Fig. 4GoC, both 5'D{delta}2 and J{delta}1 coding ends had 5' end deletions (ranging from 2–7 bp), consistent with removal of 3' overhangs by T4 pol treatment. Junctional sequences of D{delta}2-J{delta}1 coding joints were heterogeneous (11 of 17 were unique). Analysis of additional wt coding end products obtained after T4 pol pretreatment revealed loss of nucleotides only, consistent with 3' overhangs (Fig. 5Go). In contrast, 5' and 3'D{delta}2 signal end sequences retained the complete signal (our unpublished observations), as expected from previous signal end analyses (8). The lack of full-length wt coding ends with 5' P overhangs indicates that detection of such overhangs in SCID thymocytes is due to the DNA-PK deficiency.

Absence of 5' overhangs in wt open coding ends

Characterization of nonblunt coding ends in both wt and SCID thymocytes was extended by repeating the LMPCR with linkers designed to recover staggered ends directly. As previously described (43), the use of duplex oligonucleotide linkers with degenerate 5' or 3' overhanging ends (2–5 nt in length) allows linker ligation to genomic DNA with 5' or 3' overhanging ends, respectively. In the present study, a phosphorylated 5' overhanging linker was also prepared with the long-strand kinased to enable ligation to DNA with nonphosphorylated, 5' overhanging ends. The ligation specificity of each linker was tested with restriction enzyme-digested plasmid DNA representing blunt, 3' overhanging, 5' overhanging, or 5' overhanging dephosphorylated ends, followed by LMPCR analysis. The specificity assay showed that blunt end linkers only detected blunt ends, and the 3' and 5' overhanging linkers specifically detected target DNA with the appropriate overhanging ends (Fig. 6Go). However, the 5' overhanging linkers also detected DNA with blunt ends, as was observed previously (43).



View larger version (23K):
[in this window]
[in a new window]
  		FIGURE 6. Control LMPCR assay of linker ligation specificity. Linkers with blunt (lanes 1–4), 3' overhanging (lanes 5–8), 5' overhanging (lanes 9–13), and 5'-phosphorylated overhanging (lanes 14–18) ends were ligated to p5'D{delta}2 plasmid DNA pretreated as follows: SmaI-digested (S); BamHI-digested (B); BamHI-digested, calf intestinal phosphatase-treated (B/C); or PstI-digested (P). Linker ligations were performed with aqueous mixtures of 1.5 µg liver DNA and 0.1 ng of plasmid DNA overnight at 16°C. LMPCR assays of ligated DNA used primers MB61 and MB481 for 26 cycles, each consisting of 20 s at 94°C, 30 s at 61°C, and 30 s at 72°C. Shown is an ethidium-stained gel of LMPCR products. The LMPCR product detected with the nonphosphorylated 5' overhanging linkers (lane 12) was probably due to incomplete removal of 5' phosphates on the target DNA. As shown in lanes 10 and 15, the 5' overhanging linkers also detected DNA with blunt ends. This was observed previously (43 ) and attributed to linker ligation after "transient breathing" of target DNA ends and strand displacement by the linker.

 
Fig. 7GoA shows a Southern blot of LMPCR products amplified from thymus DNA. The 200-bp coding end product was efficiently recovered from wt thymocytes using 3' overhanging linkers (Fig. 7GoA, lane 4), but not with 5' overhanging linkers (lane 2) or with blunt end linkers (lane 1). The amount of 200-bp product recovered with 3' overhanging linker was almost as much as that obtained with blunt end linkers and T4 pol pretreated thymus DNA (Fig. 7GoA, compare lanes 4 and 6). A similar analysis performed with fetal SCID thymocytes indicated that the majority of open SCID coding ends also have 3' rather than 5' overhangs (Fig. 7GoB, lanes 9 and 8, respectively). When the LMPCR analysis was repeated by using two rounds of PCR with seminested primers (total of 38 cycles), then SCID coding end products were detected using both the 5' and 5'-phosphorylated overhanging linkers (Fig. 7GoC, lanes 2 and 6). Coding ends with 5' overhangs were similarly amplified from thymus DNA of DNA-PKcs null mice (Fig. 7GoC, lanes 3 and 7), but were not detected in wt thymus DNA with either of the two 5' overhanging linkers (Fig. 7GoC, lanes 1 and 5).



View larger version (48K):
[in this window]
[in a new window]
  		FIGURE 7. LMPCR analysis of DNA double-strand breaks using ligation linkers with 5' or 3' overhanging ends. A and B, DNA samples from liver (lanes 3 and 5), fetal wt (ft wt) thymocytes (lanes 1, 2, and 4) and fetal SCID (ft scid) thymocytes (lanes 7–9) were directly ligated to either the blunt linker (lanes 1 and 7), or to linkers with 5' overhanging ends (lanes 2, 3, and 8) or 3' overhanging ends (lanes 4, 5, and 9). After PCR amplification (26 cycles), products were analyzed as in Fig. 2GoA. LMPCR assays of fetal wt (lane 6) and SCID (lane 10) thymocyte DNA pretreated with T4 pol and ligated to the blunt end linker were controls for detection of open coding ends. A and B correspond to Southern blots of two different agarose gels. C, DNA samples from liver (lanes 4, 8, and 12), fetal wt (lanes 1, 5, and 9), fetal SCID (lanes 2, 6, and 10) and adult DNA-PKcs-/- (lanes 3, 7, and 11) thymocytes were directly ligated to linkers with 5' overhanging ends (lanes 1–4), 5' P overhanging ends (lanes 5–8), or 3' overhanging ends (lanes 9–12). Ligated samples were analyzed by a seminested PCR (38 cycles total) described in Materials and Methods. Only the relevant area of the blot is shown. All blots were hybridized to the J{delta}1 probe.

 
Comparison of wt and SCID coding ends with 3' overhangs

The 200-bp LMPCR products recovered from wt and SCID mice with 3' overhanging linkers were cloned and sequenced. All wt sequences corresponded to D{delta}2 or J{delta}1 coding ends (Fig. 8GoA) with the exception of a 5'D{delta}1 coding end product corresponding to a D{delta}1-D{delta}2-J{delta}1 rearrangement cleaved at the 5'D{delta}1 signal sequence. Nucleotide sequence analysis revealed that 51% of the wt coding ends (24 of 47 clones) had P nucleotides (Fig. 8GoA). Of clones with 5'D{delta}2 coding ends, half (10 of 20 clones) were full-length with one to two P nucleotides and the remaining had 5' deletions from 2–7 bp. The vast majority of J{delta}1 coding ends (25 of 27 clones or 92%) were full-length and of these 56% (14 of 25 clones) had from one to four P nucleotides.



View larger version (35K):
[in this window]
[in a new window]
  		FIGURE 8. Nucleotide sequences of open coding end products amplified from wt and SCID thymocytes by LMPCR using 3' overhanging ligation linkers. Top three lines show the germline encoded nucleotide sequences of D{delta}1, D{delta}2, and J{delta}1 gene segments and the flanking heptamers as in Fig. 4Go. Below are nucleotide sequences of cloned LMPCR products recovered with 3' overhanging linkers from fetal wt (A) and SCID (B) thymocytes. Sequences of all experimental groups are represented as in Fig. 4Go with nucleotides underlined to denote 3' P overhangs.

 
The sequence of fetal SCID clones (Fig. 8GoB) showed that 40% (11 of 28 clones) had P nucleotides. All but four SCID clones corresponded to J{delta}1 coding ends and 79% of these were full-length (19 of 24 clones). Of the four remaining clones, three corresponded to completed D{delta}2-J{delta}1 rearrangements in which the 5'D{delta}2 coding end of two clones contained two P nucleotides and one clone had four P nucleotides. The fourth example corresponded to a completed D{delta}1-D{delta}2-J{delta}1 rearrangement cleaved 5' of D{delta}1. Together, these results indicate that open hairpin coding ends with 3' P overhangs in SCID are comparable to those found in wt thymocytes.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
This report is the first to describe open coding ends in SCID thymocytes. Although temperature-dependent opening of hairpin coding ends was recently reported in a bcl-2 transgenic SCID pre-B cell line transformed by a temperature-sensitive Abelson murine leukemia virus (ts-Abl-MLV) (46), the present work is first to demonstrate open coding ends in unmanipulated SCID lymphocytes. In this study, the portion of SCID 5'D{delta}2 coding ends that were open was estimated to be ~25%. In both SCID and wt thymocytes, the majority of open coding ends were nonblunt with 3' single-strand overhangs. An additional unique population of nonblunt coding ends with 5' single-strand overhangs was detected in SCID and DNA-PK null thymocytes. The importance of these findings for the role of DNA-PK in V(D)J recombination is discussed below.

Detection of open TCR{delta} coding ends

There are several explanations for our detection of open TCR{delta} coding ends and the inability to detect such species in earlier studies (6, 7, 28). First, for studies of wt mice, we primarily used embryos since we found that TCR{delta} coding ends were at least 20-fold more abundant in thymocytes from fetal than newborn wt mice. In scid mice, age-dependent differences were not observed in the abundance of open coding ends. Second, detection of low abundance species by Southern blotting was improved by using downward alkaline transfer (37). Third, sensitivity of both our genomic Southern and LMPCR assays was enhanced by using high m.w. DNA prepared from cells embedded in agarose (36). This DNA preparation method has been shown to be crucial for detecting coding ends by LMPCR, probably because there is less randomly sheared DNA competing for linker-ligation than in DNA prepared in solution (43). Previous LMPCR assays for D{delta}2 coding ends in wt and SCID mice employed DNA prepared in solution (28).

Open coding ends in thymocytes from fetal wt mice

Broken molecules with TCR{delta} coding ends were surprisingly abundant in day 16 fetal wt thymocytes. Together, broken molecules with D{delta}2 or J{delta}1 coding ends (4-kb species in Fig. 1GoB) corresponded to ~0.5% of genomic DNA hybridizing to the J{delta}1 probe. Because 5'D{delta}2 and J{delta}1 coding ends appeared equally abundant by Southern and nucleotide sequence analysis (Figs. 4GoC and 8A), each can be inferred to correspond to ~0.25% of J{delta}1-hybridizing DNA. This level of coding ends is ~10-fold less abundant than that of signal ends, which represented 2.5–3% of fetal thymocyte DNA hybridizing to the 5'D{delta}2 probe (P. B. Nakajima and M. J. Bosma, unpublished observations).

We identified blunt and nonblunt open TCR{delta} coding ends in wt fetal thymocytes. Blunt coding ends were at least 6-fold less abundant than nonblunt coding ends (Fig. 2GoC). Blunt coding ends were reported to be the major J{kappa}1-coding end species in a wt pre-B cell line transformed by a ts-Abl-MLV (47), but were not observed in two other studies (43, 48). Rather than being full-length, as would result from hairpin opening at the tip, nucleotide sequence analysis of the blunt coding ends identified in this study revealed nonrandom deletions particular to each coding segment (5'D{delta}2 deletions of seven nucleotides and J{delta}1 deletions of three nucleotides, sequence data not shown). Assuming that these blunt ends were generated from a hairpin precursor, coding sequence could affect processing by influencing the site of hairpin opening and the extent of subsequent nucleotide loss. This explanation was proposed in studies of coding joints at endogenous loci (49) and extra-chromosomal substrates (50, 51) in which nucleotide deletion distinct for different coding segments was noted. Whether the blunt coding ends reported in this study are used for coding joints or represent dead-end intermediates is unclear. However, we suggest that they are not dead-end products because TCR{delta} coding joints recovered from fetal thymocytes include those with nucleotide deletions (10, 52) similar to those reported in this study for D{delta}2 and J{delta}1 blunt coding ends.

Most open TCR{delta} coding ends detected in wt fetal thymocytes were nonblunt. Complementary evidence obtained from two experimental approaches showed that these ends have 3' overhangs. When T4 pol was used to fill-in 5' overhangs and degrade 3' overhangs before LMPCR, all recovered coding end products had deletions, consistent with exonucleolytic removal of 3' single-strand overhangs. In a second approach, we used specially designed ligation linkers to recover overhanging ends directly by LMPCR. Linkers with 3' but not 5' overhangs resulted in amplification of coding end products and the abundance of these products was comparable to that obtained after T4 pol treatment. Most coding ends recovered with 3' overhanging linkers were either full-length or full-length with P nucleotides. In agreement with our findings, J{alpha}, J{kappa}, and JH coding ends were also found to have 3' overhangs (43, 48). Although it remains to be determined whether hairpin coding ends are predominantly opened in vivo at sites 3' of the tip, the resulting termini with 3' overhangs should be joined no less efficiently than 5' overhangs by Xrcc4 and DNA ligase IV (16). With regard to coding joint formation, there could be special features related to using coding ends with 3' overhangs. As noted in this context by Schlissel (43), substrates with 3' overhangs are preferred by TdT for addition of nontemplated nucleotides (53). In addition, although 3' overhangs are potential targets for exonuclease trimming, they would not be filled in and may be more available for homology-mediated V(D)J joining.

Open TCR{delta} coding ends in thymocytes from SCID and DNA-PKcs null mice

The most novel finding of our study is the detection of open D{delta}2 and J{delta}1 coding ends in SCID thymocytes. Similar to wt coding ends, the majority of open TCR{delta} coding ends in SCID had 3' overhangs, many of which (40%) were full-length with P nucleotides. Surprisingly, full-length coding ends with P nucleotides were also detected after T4 pol pretreatment, indicating the presence of coding ends with 5' overhangs. Such 5' P overhangs were found among clones of fetal (20%) and adult (30%) SCID coding end products, in striking contrast to T4 pol-treated wt coding ends where P nucleotides were not observed. Using 5' overhanging linkers, coding ends with 5' overhangs were also detected by bulk LMPCR analyses of DNA from SCID and DNA-PK null thymocytes (Fig. 7GoC). Open coding ends obtained from DNA-PKcs null mice appeared qualitatively and quantitatively similar to those from SCID, negating the possibility that partial function of the truncated DNA-PKcs in SCID allows some hairpin nicking. It should be emphasized that 5' overhanging coding ends were not identified after temperature-dependent opening of hairpin coding ends in ts-Abl-MLV-transformed SCID pre-B cells (46). Depending on cell culture conditions, these transformants predominantly generated either blunt coding ends with minor deletions or staggered ends with 3' overhangs that were progressively deleted with extended culture at the nonpermissive temperature. The resolution and degradation of coding ends in the transformed SCID cell line may be influenced by expression of the bcl-2 transgene and/or the v-abl oncoprotein.

One question raised by our results is whether appreciable levels of hairpin opening occur at other receptor loci in SCID and DNA-PKcs null mice. If only D{delta}2 and J{delta}1 hairpins are subject to DNA-PKcs-independent opening, this might explain the relatively high levels of D{delta}2-J{delta}1 coding joints observed in these mice compared with levels of coding joints at other loci (39, 54). With respect to these observations, it was previously proposed that D{delta}2 and J{delta}1 gene segments are "privileged sites" that can undergo a DNA-PKcs-independent pathway of rearrangement (54). An alternative idea to the privileged site notion is that low level, DNA-PKcs-independent hairpin opening can occur at any locus, with the relative abundance of a particular open coding end depending on the frequency at which the corresponding gene segment is targeted for recombination. As the D{delta}2 and J{delta}1 gene segments appear to be the primary targets of V(D)J recombination activity in early thymocyte development (39, 55), D{delta}2 and J{delta}1 recombination intermediates are readily detectable in SCID thymocytes (7, 39). This could in large part account for our ability to detect open D{delta}2 and J{delta}1 coding ends.

Our data are consistent with the generation of open coding ends in SCID and DNA-PKcs null thymocytes by normal V(D)J cleavage. Most open coding ends detected in SCID mice were strikingly similar to wt coding ends. Less-than-full-length D{delta}2 and J{delta}1 coding ends exhibited only slightly larger deletions than those from wt thymocytes. Gross deletions and abnormally long P nucleotide additions typical of many SCID D{delta}2 and J{delta}1 coding joints (23) were not observed, suggesting that such features result from selection or additional processing of SCID coding ends during end joining. The simplest explanation for coding ends with 5' overhangs is that they arise from asymmetric nicking of the hairpin at sites 5' of the tip as a result of the DNA-PK deficiency. Another explanation is that hairpin opening generates 5' and 3' overhanging coding ends in both SCID and wt mice and the 5' ends are preferentially resolved into coding joints in wt, but not in SCID. Thus the inefficient resolution allows detection of coding ends with 5' overhangs in SCID but not wt thymocytes.

Role of DNA-PK in hairpin opening

Different candidate activities for opening V(D)J hairpin intermediates have been suggested. Endonuclease cleavage of fully base-paired hairpins was observed in vitro with a protein complex comprised of three components: Mre11, RAD50, and Nsb1 (29). In this case, hairpins were cleaved one or two nucleotides 3' of the tip to produce a 3' overhang. However, the Mre11, RAD50, and Nsb1 complex is unlikely to play a major role in processing hairpin coding ends in vivo. No differences were discernable in the quality or frequency of coding joints in wt cells and those with a mutation or the Nbs1 gene (56), and patients possessing Mre11 mutations do not exhibit immunodeficiency (57).

The RAG1 and RAG2 proteins have also been suggested to play a role in hairpin opening. Artificial hairpins are opened in vitro by truncated core RAG proteins and, similar to RAG-mediated RSS cleavage, the endonuclease activity requires paired 12- and 23-RSS substrates in the presence of Mg2+, and is enhanced by the addition of the DNA bending proteins HMG1 or HMG2 (32, 33). Hairpin opening under these conditions occurs exactly at the tip and one to two nucleotides 5' of the tip (32, 33). Additional support for RAG-mediated hairpin opening comes from studies of mutated RAG core proteins. One of the mutant RAG1 proteins was shown to perform coupled cleavage, but was extremely defective for hairpin opening in vitro and for formation of coding and signal joints in vivo (58). Similarly, two mutants of RAG2 were identified that were cleavage-competent, but impaired for hairpin opening in vitro (59). Although both RAG1 and RAG2 proteins have consensus motifs for DNA-PKcs-mediated phosphorylation, mutation of these sites in the core RAG proteins did not detectably alter V(D)J recombination (60). Thus, there is no evidence that directly links the function of the RAG complex to DNA-PKcs activity.

The primary candidate for physiologic opening of coding-end hairpins appears to be the recently identified Artemis protein. Mutations of the Artemis gene result in impaired coding joint formation and increased sensitivity to ionizing radiation (61). In vitro studies show that DNA-PKcs stably associates with and phosphorylates Artemis, enabling Artemis to display hairpin endonuclease activity and generate 3' overhangs (34). Efficient hairpin opening activity in vitro depends upon Artemis remaining physically associated with DNA-PKcs (34).

Given the above candidates for hairpin opening, the generation of open coding ends observed in this study in SCID and DNA-PKcs null mice could have several explanations. One possible explanation is that in vivo, Artemis alone has low level endonuclease activity for hairpins. Alternatively, a protein kinase related to DNA-PKcs may have limited ability to phosphorylate and recruit Artemis to the postcleavage complex in the absence of DNA-PKcs. Relevant to this possibility, the protein kinase ataxi-telangiectasia mutated has been shown to localize to broken DNA intermediates of V(D)J recombination (62). Another possible explanation is that RAG proteins and/or the Mre11, RAD50, and Nsb1 complex may be responsible for some hairpin opening, when DNA-PKcs activity is low or absent. The coding ends with 5' overhangs detected in this study are consistent with results of a previous study of hairpin opening by the core RAG proteins in vitro (33).

Further studies are needed both to characterize coding ends in Artemis-deficient cells and to evaluate hairpin cleavage by Artemis and full-length RAG proteins in the presence of other factors involved in V(D)J recombination. These factors, especially DNA-binding proteins such as Ku70 and Ku86, may act in the fully assembled postcleavage complex to restrict hairpin opening whether mediated by Artemis or another endonuclease activity. Efficient coding end processing may require DNA-PK activity to disassemble or remodel the complex, in addition to activating fully the hairpin endonuclease.

Processing and joining of open SCID coding ends

Coding joint formation is two to three orders of magnitude less in SCID than in wt cells (63, 64). Thus, if hairpin nicking is reduced no more than 4-fold in SCID cells (Fig. 3GoC), subsequent steps of V(D)J recombination must also be impaired to account for the known severity of the SCID defect. In the absence of DNA-PKcs activity, resolution of open coding ends may be affected at the level of overhang processing and/or end joining. Artemis requires phosphorylation by DNA-PKcs to cleave 5' and 3' overhangs in vitro (34). This observation suggests that overhang trimming and removal of flap structures that arise from microhomology annealing (reviewed in Ref. 65) may be reduced in SCID cells. DNA-PKcs is also likely to play an important role in nonhomologous end joining mediated by XRCC4 and DNA ligase 4. Intermolecular ligation catalyzed by DNA ligase-XRCC4 was recently shown to be enhanced in vitro by DNA-PKcs (66). Previous studies have shown that XRCC4 interacts with and is phosphorylated by DNA-PK in vitro (67, 68). XRCC4 can bind to DNA with broken ends (69) and may serve to align DNA termini before joining (15). In view of these findings and those showing a reduction in DNA double-strand break repair in SCID and DNA-PK-null cells in response to DNA damaged by ionizing radiation or bleomycin (35, 70, 71, 72), it is understandable that joining of open SCID coding ends would be greatly impaired in the absence of DNA-PK activity.


   Acknowledgments
 
We are grateful to A. Farley and K. Wiley for valuable assistance with plasmid sequencing, to G. Bosma and C. Saunders for timed matings, and to G. Taccioli for generously providing DNA-PKcs null mice. The assistance of the following CORE facilities of the Fox Chase Cancer Center is gratefully acknowledged: DNA Sequencing, DNA Synthesis, Laboratory Animal, and Special Services. We also thank R. Daniel, D. Kappes, R. Katz, R. Perry, and N. Reutsch for critically reading the manuscript. The secretarial help of R. Diehl is also gratefully acknowledged.


   Footnotes
 
1 This work was supported by the National Institutes of Health (Grants CA06929 and CA04946), Center of Research Excellence Grant CA06927, and an appropriation from the Commonwealth of Pennsylvania. Back

2 Address correspondence and reprint requests to Dr. Pamela B. Nakajima, Institute for Cancer Research, Fox Chase Cancer Center, Philadelphia, PA 19111. E-mail address: PB_Nakajima{at}fccc.edu Back

3 Abbreviations used in this paper: RSS, recombination signal sequence; RAG, recombination-activating gene; P, palindromic; DNA-PK, DNA-dependent protein kinase; DNA-PKcs, catalytic subunit of DNA-PK; wt, wild type; MBN, mung bean nuclease; T4 pol, T4 DNA polymerase; LMPCR, ligation-mediated PCR; ts-Abl-MLV, temperature-sensitive Abelson murine leukemia virus. Back

Received for publication March 18, 2002. Accepted for publication July 10, 2002.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Gellert, M.. 1997. Recent advances in understanding V(D)J recombination. Adv. Immunol. 64:39.[Medline]
  2. Lewis, S. M.. 1994. The mechanism of V(D)J joining: lessons from molecular, immunological, and comparative analyses. Adv. Immunol. 56:27.[Medline]
  3. Oettinger, M. A., D. G. Schatz, C. Gorka, D. Baltimore. 1990. RAG-1 and RAG-2, adjacent genes that synergistically acitvate V(D)J recombination. Science 248:1517.[Abstract/Free Full Text]
  4. McBlane, J. F., D. C. van Gent, D. A. Ramsden, C. Romeo, C. A. Cuomo, M. Gellert, M. A. Oettinger. 1995. Cleavage at a V(D)J recombination signal requires only RAG1 and RAG2 proteins and occurs in two steps. Cell 83:387.[Medline]
  5. van Gent, D. C., J. F. McBlane, D. A. Ramsden, M. J. Sadofsky, J. E. Hesse, M. Gellert. 1995. Initiation of V(D)J recombination in a cell-free system. Cell 81:925.[Medline]
  6. Roth, D. B., P. B. Nakajima, J. P. Menetski, M. J. Bosma, M. Gellert. 1992. V(D)J recombination in mouse thymocytes: double-strand breaks near T cell receptor {delta} rearrangement signals. Cell 69:41.[Medline]
  7. Roth, D. B., J. P. Menetski, P. B. Nakajima, M. J. Bosma, M. Gellert. 1992. V(D)J recombination: broken DNA molecules with covalently sealed (hairpin) coding ends in scid mouse thymocytes. Cell 70:983.[Medline]
  8. Roth, D. B., C. Zhu, M. Gellert. 1993. Characterization of broken DNA molecules associated with V(D)J recombination. Proc. Natl. Acad. Sci. USA 90:10788.[Abstract/Free Full Text]
  9. Schlissel, M., A. Constantinescu, T. Morrow, M. Baxter, A. Peng. 1993. Double-strand signal sequence breaks in V(D)J recombination are blunt, 5'-phosphorylated, RAG-dependent, and cell cycle regulated. Genes Dev. 7:2520.[Abstract/Free Full Text]
  10. Lafaille, J. J., A. DeCloux, M. Bonneville, Y. Takagaki, S. Tonegawa. 1989. Junctional sequences of T cell receptor {gamma}{delta} genes: implications for {gamma}{delta} T cell lineages and for a novel intermediate of V-(D)-J joining. Cell 59:859.[Medline]
  11. McCormack, W. T., L. W. Tjoelker, L. M. Carison, B. Petryniak, C. F. Barth, E. H. Humphries, C. B. Thompson. 1989. Chicken IgL gene rearrangement involves deletion of a circular episome and addition of single nonrandom nucleotides to both coding segments. Cell 56:785.[Medline]
  12. Lewis, S. M.. 1994. P nucleotides, hairpin DNA and V(D)J joining: making the connection. Semin. Immunol. 6:131.[Medline]
  13. Taccioli, G. E., G. Rathbun, E. Oltz, T. Stamato, P. A. Jeggo, F. W. Alt. 1993. Impairment of V(D)J recombination in double-strand break repair mutants. Science 260:207.[Abstract/Free Full Text]
  14. Smith, G. C. M., S. P. Jackson. 1999. The DNA-dependent protein kinase. Genes Dev. 13:916.[Free Full Text]
  15. Li, Z., T. Otevrel, Y. Gao, H.-L. Cheng, B. Seed, T. Stamato, G. E. Taccioli, F. W. Alt. 1995. The XRCC4 gene encodes a novel protein involved in DNA double-strand break repair and V(D)J recombination. Cell 83:1079.[Medline]
  16. Grawunder, U., M. Wilm, X. Wu, P. Kulesza, T. E. Wilson, M. Mann, M. R. Lieber. 1997. Activity of DNA ligase IV stimulated by complex formation with XRCC4 protein in mammalian cells. Nature 388:492.[Medline]
  17. Grawunder, U., D. Zimmer, P. Kulesza, M. R. Lieber. 1998. Requirement for an interaction of XRCC4 with DNA Ligase IV for wild-type V(D)J recombination and DNA double-strand break repair in vivo. J. Biol. Chem. 273:24708.[Abstract/Free Full Text]
  18. Grawunder, U., D. Zimmer, S. Fugmann, K. Schwarz, M. R. Lieber. 1998. DNA ligase IV is essential for V(D)J recombination and DNA double-strand break repair in human precursor lymphocytes. Mol. Cell 2:477.[Medline]
  19. Bosma, M. J., A. M. Carroll. 1991. The scid mouse mutant: definition, characterization, and potential uses. Annu. Rev. Immunol. 9:323.[Medline]
  20. Schuler, W., I. J. Weiler, A. Schuler, R. A. Phillips, N. Rosenberg, T. W. Mak, J. F. Kearney, R. P. Perry, M. J. Bosma. 1986. Rearrangement of antigen receptor genes is defective in mice with severe combined immune deficiency. Cell 46:963.[Medline]
  21. Malynn, B. A., T. K. Blackwell, G. M. Fulop, G. A. Rathbun, A. J. W. Furley, P. Ferrier, L. B. Heinke, R. A. Phillips, G. D. Yancopoulos, F. W. Alt. 1988. The scid defect affects the final step of the immunoglobulin VDJ recombinase mechanism. Cell 54:453.[Medline]
  22. Schuler, W., N. R. Ruetsch, M. Amsler, M. J. Bosma. 1991. Coding joint formation of endogenous T cell receptor genes in lymphoid cells from scid mice: unusual P-nucleotide additions in VJ-coding joints. Eur. J. Immunol. 21:589.[Medline]
  23. Fish, S. M., M. J. Bosma. 1994. Abnormal deletions in the T-cell receptor {delta} locus of mouse thymocytes. Mol. Cell. Biol. 14:4455.[Abstract/Free Full Text]
  24. Kirchgessner, C. U., C. K. Patil, J. W. Evans, C. A. Cuomo, L. M. Fried, T. Carter, M. A. Oettinger, M. J. Brown. 1995. DNA-dependent kinase (p350) as a candidate gene for the murine SCID defect. Science 267:1178.[Abstract/Free Full Text]
  25. Blunt, T., N. J. Finnie, G. E. Taccioli, G. C. Smith, J. Demengeot, T. M. Gottlieb, R. Mizuta, A. J. Varghese, F. W. Alt, P. A. Jeggo, S. P. Jackson. 1995. Defective DNA-dependent protein kinase activity is linked to V(D)J recombination and DNA repair defects associated with the murine scid mutation. Cell 80:813.[Medline]
  26. Blunt, T., D. Gell, M. Fox, G. E. Taccioli, A. R. Lehmann, S. P. Jackson, P. A. Jeggo. 1996. Identification of a nonsense mutation in the carboxyl-terminal region of DNA-dependent protein kinase catalytic subunit in the scid mouse. Proc. Natl. Acad. Sci. USA 93:10285.[Abstract/Free Full Text]
  27. Danska, J., D. P. Holland, S. Mariathasan, K. M. Williams, C. J. Guidos. 1996. Biochemical and genetic defects in the DNA-dependent protein kinase in murine scid lymphocytes. Mol. Cell. Biol. 16:5507.[Abstract]
  28. Zhu, C., D. B. Roth. 1995. Characterization of coding ends in thymocytes of scid mice: implications for the mechanism of V(D)J recombination. Immunity 2:101.[Medline]
  29. Paull, T. T., M. Gellert. 1999. Nbs1 potentiates ATP-driven DNA unwinding and endonuclease cleavage by the Mre11/Rad50 complex. Genes Dev. 13:1276.[Abstract/Free Full Text]
  30. Carney, J. P., R. S. Maser, H. Olivares, E. M. Davis, M. Le Beau, J. R. Yates, L. Hays, W. F. Morgan, J. H. Petrini. 1998. The hMre11/hRad50 protein complex and Nimegen breakage syndrome: linkage of double-strand break repair to the cellular DNA damage response. Cell 93:477.[Medline]
  31. Khanna, K. K., S. P. Jackson. 2001. DNA double-strand breaks: signaling, repair and the cancer connection. Nat. Genet. 27:247.[Medline]
  32. Besmer, E., J. Mansilla-Soto, S. Cassard, D. J. Sawchuk, G. Brown, M. Sadofsky, S. M. Lewis, M. C. Nussenzweig, P. Cortes. 1998. Hairpin coding end opening is mediated by RAG1 and RAG2 proteins. Mol. Cell 2:817.[Medline]
  33. Shockett, P. E., D. G. Schatz. 1999. DNA hairpin opening mediated by the RAG1 and RAG2 proteins. Mol. Cell. Biol. 19:4159.[Abstract/Free Full Text]
  34. Ma, Y., U. Pannicke, K. Schwarz, M. R. Lieber. 2002. Hairpin opening and overhang processing by an Artemis/DNA-dependent protein kinase complex in nonhomologous end joining and V(D)J recombination. Cell 108:781.[Medline]
  35. Taccioli, G. E., A. G. Amatucci, H. J. Beamish, D. Gell, X. H. Xiang, M. I. T. Arzayus, A. Priestly, S. P. Jackson, A. M. Rothstein, P. A. Jeggo, V. L. M. Herrera. 1998. Targeted disruption of the catalytic subunit of the DNA-PK gene in mice confers severe combined immunodeficiency and radiosensitivity. Immunity 9:355.[Medline]
  36. Nakajima, P. B., J. P. Menetski, D. B. Roth, M. Gellert, M. J. Bosma. 1995. V-D-J rearrangements at the T cell receptor {delta} locus in mouse thymocytes of the {alpha}{beta} lineage. Immunity 3:609.[Medline]
  37. Chomczynski, P.. 1992. One-hour downward alkaline capillary transfer blotting of DNA and RNA. Anal. Biochem. 201:134.[Medline]
  38. Feinberg, A. P., B. Vogelstein. 1983. A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132:6.[Medline]
  39. Carroll, A., M. J. Bosma. 1991. T-lymphocyte development in scid mice is arrested shortly after the initiation of T-cell receptor {delta} gene recombination. Genes Dev. 5:1357.[Abstract/Free Full Text]
  40. Nakajima, P. B., M. J. Bosma. 1997. Characterization of excised DNA intermediates associated with V(D)J recombination at the T cell receptor {delta} locus. Mol. Cell. Biol. 17:2631.[Abstract]
  41. Hardy, R. R., C. E. Carmack, S. A. Shinton, J. D. Kemp, K. Hayakawa. 1991. Resolution and characterization of pro-B and pre-pro-B cell stages in normal mouse bone marrow. J. Exp. Med. 173:1213.[Abstract/Free Full Text]
  42. Mueller, P. R., B. Wold. 1989. In vivo footprinting of a muscle specific enhancer by ligation mediated PCR. Science 246:780.[Abstract/Free Full Text]
  43. Schlissel, M. S.. 1998. Structure of nonhairpin coding-end DNA breaks in cells undergoing V(D)J recombination. Mol. Cell. Biol. 18:2029.[Abstract/Free Full Text]
  44. Takagaki, Y., N. Nakanishi, I. Ishida, O. Kanagawa, S. Tonegawa. 1989. T cell receptor-{gamma} and -{delta} genes preferentially utilized by adult thymocytes for the surface expression. J. Immunol. 142:2112.[Abstract]
  45. Schuler, W., A. Schuler, G. G. Lennon, G. C. Bosma, M. J. Bosma. 1988. Transcription of unrearranged antigen receptor genes in scid mice. EMBO J. 7:2019.[Medline]
  46. Brown, M. L., Y. Chang. 2000. Metabolism of recombination coding ends in scid cells. J. Immunol. 164:4135.[Abstract/Free Full Text]
  47. Ramsden, D. A., M. Gellert. 1995. Formation and resolution of double-strand break intermediates in V(D)J rearrangement. Genes Dev. 9:2409.[Abstract/Free Full Text]
  48. Livak, F., D. G. Schatz. 1997. Identification of V(D)J recombination coding end intermediates in normal thymocytes. J. Mol. Biol. 267:1.[Medline]
  49. Nadel, B., A. J. Feeney. 1995. Influence of coding-end sequence on coding end processing in V(D)J recombination. J. Immunol. 155:4322.[Abstract]
  50. Boubnov, N. V., Z. P. Wills, D. T. Weaver. 1993. V(D)J recombination coding junction formation without DNA homology: processing of coding termini. Mol. Cell. Biol. 13:6957.[Abstract/Free Full Text]
  51. Gauss, G. H., M. R. Lieber. 1996. Mechanistic constraints on diversity in human V(D)J recombination. Mol. Cell. Biol. 16:258.[Abstract]
  52. Wilbert, W., J. Weber-Arden, D. Kabelitz, B. Arden. 1997. TCR-{delta} gene rearrangement and selection during fetal thymocyte development. J. Immunol. 159:3338.[Abstract]
  53. Chang, L. M., F. J. Bolum. 1986. Molecular biology of terminal transferase. CRC Crit. Rev. Biochem. 21:27.[Medline]
  54. Bogue, M. A., C. Jhappan, D. B. Roth. 1998. Analysis of variable (diversity) joining recombination in DNA dependent protein kinase (DNA-PK)-deficient mice reveals DNA-PK-independent pathways for both signal and coding joint formation. Proc. Nat. Acad. Sci. USA 95:15559.[Abstract/Free Full Text]
  55. Chien, Y., M. Iwashima, D. A. Wettstein, K. B. Kaplan, J. F. Elliott, W. Born, M. M. Davis. 1987. T-cell receptor {delta} gene rearrangements in early thymocytes. Nature 330:722.[Medline]
  56. Harfst, E., S. Cooper, S. Neubauer, L. Distel, U. Grawunder. 2000. Normal V(D)J recombination in cells from patients with Nimengen breakage syndrome. Mol. Immunol. 37:915.[Medline]
  57. Stewart, G. S., R. S. Maser, T. Stankovic, D. A. Bressan, M. I. Kaplan, N. G. Jaspers, A. Raams, P. J. Byrd, J. H. Petrini, A. M. Taylor. 1999. The DNA double-strand break repair gene hMRE11 is mutated in individuals with an ataxia-telangiectasia-like disorder. Cell 99:577.[Medline]
  58. Yarnell Schultz, H., M. A. Landree, J. X. Qiu, S. B. Kale, D. B. Roth. 2001. Joining-deficient RAG1 mutants block V(D)J recombination in vivo and hairpin opening in vitro. Mol. Cell 7:65.[Medline]
  59. Qiu, J. X., S. B. Kale, H. Yarnell Schultz, D. B. Roth. 2001. Separation-of-function mutants reveal critical roles for RAG2 in both the cleavage and joining steps of V(D)J recombination. Mol. Cell 7:77.[Medline]
  60. Lin, J. M., M. A. Landree, D. B. Roth. 1999. V(D)J recombination catalyzed by mutant RAG proteins lacking consensus DNA-PK phosphorylation sites. Mol. Immunol. 36:1263.[Medline]
  61. Moshous, D., I. Callebaut, R. de Chasseval, B. Corneo, M. Cavazzana-Calvo, F. Le Deist, I. Tezcan, O. Sanal, Y. Bertrand, N. Philippe, et al 2001. Artemis, a novel DNA double-strand break repair/V(D)J recombination protein, is mutated in human severe combined immune deficiency. Cell 105:177.[Medline]
  62. Perkins, E. J., A. Nair, D. O. Cowley, T. Van Dyke, Y. Chang, D. A. Ramsden. 2002. Sensing of intermediates in V(D)J recombination by ATM. Genes Dev. 16:159.[Abstract/Free Full Text]
  63. Lieber, M. R., J. E. Hessie, S. Lewis, G. C. Bosma, K. Mizuuchi, M. J. Bosma, M. Gellert. 1988. The defect in murine severe combined immune deficiency: joining of signal segments but not coding segments in V(D)J recombination. Cell 55:7.[Medline]
  64. Harrington, J. H., C. L. Hsieh, J. Gerton, G. Bosma, M. R. Lieber. 1992. Analysis of the defect in DNA end joining in the murine scid mutation. Mol. Cell. Biol. 12:4758.[Abstract/Free Full Text]
  65. Grawunder, U., E. Harfst. 2001. How to make ends meet in V(D)J recombination. Curr. Opin. Immunol. 13:186.[Medline]
  66. Chen, L., K. Trujillo, P. Sung, A. E. Tomkinson. 2000. Interactions of the DNA ligase IV-XRCC4 complex with DNA ends and the DNA-dependent protein kinase. J. Biol. Chem. 275:26196.[Abstract/Free Full Text]
  67. Critchlow, S. E., R. P. Bowater, S. P. Jackson. 1997. Mammalian DNA double-strand break repair protein XRCC4 interacts with DNA ligase IV. Curr. Biol. 7:588.[Medline]
  68. Leber, R., T. W. Wise, R. Mizuta, K. Meek. 1998. The XRCC4 gene product is a target for and interacts with the DNA-dependent protein kinase. J. Biol. Chem. 273:1794.[Abstract/Free Full Text]
  69. Modesti, M., J. E. Hesse. 1999. DNA binding of XRCC4 protein is associated with V(D)J recombination but not with stimulation of DNA ligase IV activity. EMBO J. 18:2008.[Medline]
  70. Biedermann, K. A., J. Sun, A. J. Giaccia, L. M. Tosto, M. Brown. 1991. scid mutation in mice confers hypersensitivity to ionizing radiation and a deficiency in DNA double-strand break repair. Proc. Nat. Acad. Sci. USA 88:1394.[Abstract/Free Full Text]
  71. Fulop, G. M., R. A. Phillips. 1990. The scid mutation in mice causes a general defect in DNA repair. Nature 347:479.[Medline]
  72. Hendrickson, E. A., X.-Q. Qin, E. A. Bump, D. G. Schatz, M. Oettinger, D. T. Weaver. 1991. A link between double-strand break-related and V(D)J recombination: the scid mutation. Proc. Nat. Acad. Sci. USA 88:4061.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
K. Kiefer, P. B. Nakajima, J. Oshinsky, S. H. Seeholzer, M. Radic, G. C. Bosma, and M. J. Bosma
Antigen Receptor Editing in Anti-DNA Transitional B Cells Deficient for Surface IgM
J. Immunol., May 1, 2008; 180(9): 6094 - 6106.
[Abstract] [Full Text] [PDF]

Home page
 Vet PatholHome page
L. E. Perryman
Molecular Pathology of Severe Combined Immunodeficiency in Mice, Horses, and Dogs
Vet. Pathol., March 1, 2004; 41(2): 95 - 100.
[Abstract] [Full Text] [PDF]

Home page
 JEMHome page
G. C. Bosma, J. Kim, T. Urich, D. M. Fath, M. G. Cotticelli, N. R. Ruetsch, M. Z. Radic, and M. J. Bosma
DNA-dependent Protein Kinase Activity Is Not Required for Immunoglobulin Class Switching
J. Exp. Med., December 2, 2002; 196(11): 1483 - 1495.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nakajima, P. B.
Right arrow Articles by Bosma, M. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nakajima, P. B.
Right arrow Articles by Bosma, M. J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS