- Email: [email protected]
Signal joint formation is inhibited in murine scid preB cells and fibroblasts in substrates with homopolymeric coding ends
Molecular Immunology 36 (1999) 551±558
www.elsevier.com/locate/molimm
Signal joint formation is inhibited in murine scid preB cells and ®broblasts in substrates with homopolymeric coding ends Tianhe Sun a, 1, Uthayashanker R. Ezekiel 1, 2, a, 2, Leslie Erskine b, Ryan Agulo a, Grazyna Bozek a, David Roth b, Ursula Storb a,* a
Department of Molecular Genetics and Cell Biology, University of Chicago, 920 East 58th Street, Chicago, IL 60637, USA Department of Microbiology and Immunology, Howard Hughes Medical Institute, Baylor College of Medicine, Houston, TX 77030, USA
b
Received 24 December 1998; received in revised form 15 March 1999; accepted 15 March 1999
Abstract During B and T lymphocyte development, immunoglobulin and T cell receptor genes are assembled from the germline V, (D) and J gene segments (Lewis, S.M., 1994. The mechanism of V(D)J joining: lessons from molecular, immunological and comparative analyses. Adv. Immunol. 56, 27±150). These DNA rearrangements, responsible for immune system diversity, are mediated by a site speci®c recombination machinery via recognition signal sequences (RSSs) composed of conserved heptamers and nonamers separated by spacers of 12 or 23 nucleotides (Lewis, S.M., 1994. The mechanism of V(D)J joining: lessons from molecular, immunological and comparative analyses. Adv. Immunol. 56, 27±150). Recombination occurs only between a RSS with a 12mer spacer and a RSS with a 23mer spacer (Lewis, S.M., 1994. The mechanism of V(D)J joining: lessons from molecular, immunological and comparative analyses. Adv. Immunol. 56, 27±150). RAG1 and RAG2 proteins cleave precisely at the RSS-coding sequence border leading to ¯ush signal ends and coding ends with a hairpin structure (Eastman, M., Leu, T., Schatz, D., 1996. Initiation of V(D)J recombination in vitro obeying the 12/23 rule. Nature 380, 85±88; Roth, D.B., Menetski, J.P., Nakajima, P.B., Bosma, M.J., Gellert, M., 1992. V(D)J recombination: broken DNA molecules with covalently sealed (hairpin) coding ends in scid mouse thymocytes. Cell 983±991; Roth, D.B., Zhu, C., Gellert, M., 1993. Characterization of broken DNA molecules associated with V(D)J recombination. Proc. Natl. Acad. Sci. USA 90, 10,788±10,792; van Gent, D., McBlane, J., Sadofsky, M., Hesse, J., Gellert, M., 1995. Initiation of V(D)J recombination in a cell-free system. Cell 81, 925± 934). Signal ends join, forming a signal joint. The hairpin coding ends are opened by a yet unknown endonuclease, and are further processed to form the coding joint (Lewis, S.M., 1994. The mechanism of V(D)J joining: lessons from molecular, immunological and comparative analyses. Ad. Immunol. 56, 27±150.) The murine scid mutation has been shown to aect coding joints, but much less signal joint formation. In this study we demonstrate that the murine scid mutation inhibits correct signal joint formation when both coding ends contain homopolymeric sequences. We suggest that this ®nding may be due to the function of the SCID protein as an assembly component in V(D)J recombination. # 1999 Elsevier Science Ltd. All rights reserved. Keywords: Immunoglobulin gene rearrangement; Severe combined immunode®ciency; Murine scid
1. Introduction
* Corresponding author. Tel.: +773-702-4440; fax: +773-7023172. E-mail address: [email protected] (U. Storb) 1 This author contributed equally to this study. 2 Current address: Department of Neurology, Washington University Medical Center, P.O. Box 8111, St Louis, MO 63110, USA.
Recent identi®cation of genes and proteins involved in double strand break repair (DSBR) in mammalian systems has elucidated the connection between DSBR and V(D)J recombination (Boubnov et al., 1995; Finnie et al., 1995; Pergola et al., 1993; Rathmell and Chu, 1994; Taccioli et al., 1993, 1994b). The DNA dependent serine/threonine protein kinase holoenzyme
0161-5890/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 1 6 1 - 5 8 9 0 ( 9 9 ) 0 0 0 5 3 - X
552
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
Fig. 1. Signal joint plasmid SJ-G10T10. The construction of the plasmid is described in Materials and Methods. The plasmid contains the polyoma origin of replication (Py ori), the polyoma T antigen gene (Py T), and a prokarotic origin of replication (pUC ori). The black dots stand for the promoters of the respective genes. The nucleotides in bold are the heptamer and the nonamer sequences of the RSSs. The small vertical arrows show the ligation sites of the compatible cohesive ends, which are generated by the two indicated restriction enzymes during the plasmid construction.
(DNA-PK) is composed of the catalytic subunit, DNA -PKcs/SCID, encoded by the XRCC7 gene and a heterodimeric Ku protein composed of a p70 subunit and a p86 subunit encoded by XRCC5 (Blunt et al., 1995; Taccioli et al., 1994a, b; Zdzienicka, 1995). Inactivation of the genes encoding p70 or p86 aects both coding and signal joint formation, but the murine scid mutation which eliminates the kinase function (Kirchgessner et al., 1995; Danska et al., 1996) aects mainly coding joint formation (Bosma and Carroll, 1991; Lieber et al., 1988). Hairpin coding ends accumulate in the thymus of scid mice (Roth et al., 1992; Zhu and Roth, 1994). On the basis of the biochemical properties of Ku proteins and DNA-PKcs, several possible functions have been proposed for their involvement in V(D)J recombination (Jeggo et al., 1995; Roth et al., 1995): (1) The Ku proteins bind to the hairpin coding ends and the signal ends thereby protecting them from nucleases. Subsequent binding of DNA-PKcs results in phosphorylation of Ku leading to its dislocation from
the hairpin coding ends and allowing access to the hairpin opening enzyme (Jeggo et al., 1995; Roth et al., 1995). (2) DNA-PK regulates other protein components involved in V(D)J recombination by its phosphorylating activity. (3) Because of its enormous size, the DNA-PKcs (450 Kda) acts as an alignment factor or a scaold on which the V(D)J recombinase and substrate DNA assemble (Jeggo et al., 1995; Roth et al., 1995). We have previously studied the in¯uence of homopolymeric coding ends on V(D)J recombination in normal preB cells (Ezekiel et al., 1995, 1997). Both coding and signal joint formation are strongly aected in frequency and quality by certain homopolymeric coding ends. In this study we have analyzed the formation of signal joints in scid preB cells and scid ®broblasts expressing the Rag1 and Rag2 proteins, using recombination substrates with homopolymeric coding ends which do not aect signal joint formation in normal preB cells.
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
2. Materials and methods 2.1. Cell lines The scid preB cell line S33 was a gift of M. Bosma (Schuler et al., 1986). The scid ®broblast cell line sc3T3 has been described (Han et al., 1997). 2.2. Plasmid constructions The signal joint plasmids SJ-G10Wt, -WtT10, A10Wt, -WtC10, -G10T5, and -G5T5 were all derived from the coding joint construct CJ-G10C10 (Ezekiel et al., 1995) by reversing the 12mer spacer and 23mer spacer rearrangement signal sequences (RSS) in two steps. In the ®rst step the plasmid was cut by Nhel and Sall and the 12mer RSS and adjacent sequences were replaced by an oligonucleotide cassette containing a 23mer RSS and the relevant coding end. The top strand of the oligonucleotide with the wild type coding end is 5'CTAGCGGTTTTTGTACAGCCAGACAG TGGGGATCCACCACTGTGGTGGACGTTCG3 ' and the bottom strand is 5 'TCGAC GAACGTCCACCACAGTGGTGGATCCCCACT GTCTGGCTGTACAAAAACCG3 '. The underlined wild type coding end sequences are changed to the respective homopolymeric sequences and used to make the intermediate constructs. In the second step these intermediate constructs were cut by Spel and Xhol and an oligonucleotide cassette containing a 12mer RSS and the respective coding end was inserted to get the signal joint constructs. The top strand of the oligonucleotide with the wild type coding end sequence is 5 'CTAGTGAGTATCCTCCACAGTGATAGATCTC TGAACAAAAACC3 ' and the bottom strand is 5 'TCGAGGTTTTTGTTCAGAGATCTATCACTGT GGAGGATACTCA3 '. The signal joint plasmid SJ-G10T10 is shown in Fig. 1. 2.3. Transient transfection in preB cells and ®broblasts Transient transfection in scid preB cells S33 was carried out as described (Ezekiel et al., 1995). Transient transfection of ®broblasts was performed as previously described (Sadofsky et al., 1993). Brie¯y, SC3T3 cells were transiently transfected with 5 mg of plasmid substrate (SJWT or SJG10T10), 2.1 mg of full length Rag-1 expression vector (pJH548), and 2.5 mg full length Rag-2 expression vector (pJH549) using the calcium phosphate method. The DNA was harvested from the cells 48 h after transfection as described previously (Steen et al., 1996; Steen et al., 1997). 2.4. Recombination assay The recombination assay was done as described ear-
553
lier (Ezekiel et al., 1995; Fig. 1). In brief, the plasmid substrates contain two bacterial drug resistance marker genes, chloramphenicol acetyl transferase and neomycin phosphotransferase, which confer resistance to chloramphenicol (Cm) and kanamycin (Km), respectively. The neomycin phosphotransferase promoter (PKm) is separated from its coding region by prokaryotic transcription terminators (T1T2) that are ¯anked on either side by a coding end and a RSS. Upon rearrangement in V(D)J recombination competent cells (preB or Rag1/2 expressing ®broblasts), the two RSSs join which leads to the deletion of the V and J coding ends and the terminators. Plasmid DNA is recovered from preB cells or ®broblasts and digested with Dpnl. Replication of plasmid molecules in the preB cells or ®broblasts causes loss of their dam methylation and these replicated molecules are refractory to Dpnl digestion. After Dpnl digestion, the plasmids are transformed into Escherichia coli. Only the recombined plasmids confer resistance to Km. All Km resistant colonies are further checked by polymerase chain reaction (PCR) assay for rearrangement. The percent recombination is calculated from the ratio of Cm plus Km double resistant to Cm resistant colonies, always using the PCR corrected value for the double resistant colonies. 2.5. PCR assay for signal joints In order to detect signal joints, the DNA harvested from the scid ®broblasts was subjected to 24 cycles of PCR using primers to the Km promoter (pKm: 5'CCTCTGGTAAGGTTGGGAAG-3 ') and the Km coding region (neo: 5'-AGCCGAATAGCCTCTCCA CCCAAG-3 ') (Ezekiel et al., 1995). A portion of the PCR products was digested with the restriction enzyme ApaLl, as described (Roth et al., 1993). PCR products were separated on 6% polyacrylamide gels, transferred to nylon membranes, and probed with a 32P-end labeled oligonucleotide which hybridizes to a sequence between the Km primer and the 23 RSS (DR285: 5 'CTGGATGGCTTTCTTGCCGCCAAGG3 '). 3. Results 3.1. Signal joint formation in scid preB cells We and others have shown that coding end nucleotide composition aects V(D)J recombination in wild type cells (Boubnov et al., 1993; Ezekiel et al., 1995; Ezekiel et al., 1997; Gerstein and Lieber, 1993). Furthermore, a particular RAG1 mutation is sensitive to coding end nucleotides (Sadofsky et al., 1995). In this study we have tested the eect of coding end composition on the function of SCID/DNA-PKcs in signal
554
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
Fig. 2. Recombination frequencies for signal joint (SJ) constructs in a scid preB cell line. The ®rst column shows the signal joint constructs, the second column shows the number of experiments, and the last column shows the percent recombination. indicates that not even a single recombinant was found for SJ-G10T10 and SJ-A9C10 in the scid preB cell line. For SJ-G10T10 at least 0.7 106 total bacterial cells were screened for recombinants. Constructs are named by the nucleotide composition of the top strand. For example, plasmid substrate SJ-G10T10 was so named because the 12mer RSS (open triangle) coding end contains ten G nucleotides and the 23mer RSS (closed triangle) coding end has ten T nucleotides. In SJ-Wt (wild type), the ten nucleotides at the 12mer RSS coding end represent Vk and the ten nucleotides at the 23mer RSS coding end represent Jk (Engler and Storb, 1987). Recombination frequency in a normal preB cell line (ABC1.28.13.8) (Ezekiel et al., 1995) follows: SJWt=0.14520.048; SJ-G10T10=0.09420.022; SJ-A9C10=0.02620.009.
joint formation. We transfected a scid preB cell line with extrachromosomal signal joint constructs containing homopolymeric coding ends at the 12mer and 23mer RSS (SJ-G10T10 and its inverse, SJ-A9C10, see Fig. 2) and a construct containing ten nucleotides of wild type coding ends (SJ-Wt) (the Wt coding ends are derived from Vk and Jk sequences; Ezekiel et al., 1995). The orientation of the RSSs is such that upon V(D)J recombination, a signal joint is formed in the plasmid. All three of these constructs have similar rearrangement frequencies in normal preB cells (Ezekiel et al., 1995; see legend Fig. 2). Strikingly, the two homopolymeric constructs do not give detectable rearrangement in scid preB cells, although the wild type
construct gives a similar recombination frequency as in normal preB cells (Fig. 2). Thus, the presence of certain homopolymeric sequences at both coding ends eliminates signal joint formation in cells with the scid mutation. To understand to what extent homopolymeric coding ends in¯uence signal joints, additional constructs containing one homopolymeric coding end and one wild type coding end were tested in scid preB cells (SJG10Wt, -WtT10, -A10Wt, -WtC10, see Fig. 2). These four constructs give recombination frequencies similar to the construct with two wild type coding ends. The result indicates that a single wild type coding end at
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
555
Fig. 3. Transformation of cell extracts recovered from scid ®broblasts. The table shows the transformation results of scid ®broblast cell extracts from the transfections of either SJ-G10T10 or SJ-WT constructs. Four transfections are shown, two for each construct. For the cell extract from each transfection, two or three independent transformations were performed. The `total colonies' stands for the number of single drug resistant 1 bacteria (Chloramphenicol ) in that speci®c transformation. The double drug resistant bacteria were further con®rmed by colony PCR using pkm and neo primers and the numbers are shown in the `PCR positive' column. These recombined plasmids were further analyzed by restriction digestion (ApaLl) and sequencing (for the ApaLl uncut only, i.e. imperfect joints). Three categories were applied to these recombined plasmids. The `perfect joints' stands for the plasmid products which can be digested by ApaLl. The `imperfect joints' stands for the ones with a few nucleotides deletion at the RSS regions. The `non-V(D)J joints' refers to the ones with some deletions away from the RSS regions and thus unrelated to V(D)J recombination. These deletions presumably occurred in the bacteria and deleted the two terminators, giving rise to double drug resistant bacterial clones and shorter PCR products whose sizes are matched with the V(D)J recombined ones. The V(D)J recombination frequency is the ratio of the numbers of the total V(D)J joints (including both perfect and imperfect joints) and the total colonies.
either RSS rescues the homopolymeric eect on signal joint formation in murine scid preB cells. To determine how the number of homopolymeric nucleotides at both coding ends aects signal joint formation, we made two additional constructs (Fig. 2). Construct SJ-G10T5 contains ten G nucleotides at the 5 ' end of the 12mer RSS and ®ve A's at the 5 ' end of the 23mer RSS. Construct SJ-G5T5 contains ®ve G's and A's at the 5' ends of 12 and 23mer RSSs, respectively. Both these constructs give recombination frequencies similar to the wild type construct and the constructs with homopolymeric sequences at one coding end (Fig. 2). Thus, homopolymeric sequences aect signal joint formation in scid cells only when they are present at both coding ends, and when both are longer than ®ve nucleotide pairs. In normal cells the majority of the signal ends join perfectly. The perfect signal joint creates an ApaLl/ HgiAl restriction site (Lieber et al., 1988). Previous studies have shown that in scid preB cells 50% of the signal joints (Lieber et al., 1988) and in scid ®broblasts (expressing RAG1 and RAG2) 80% of the signal joints are perfect (Taccioli et al., 1994a). To test the ®delity of the joints, we digested the recombinant plasmids from SJ-Wt, -WtC10, -A10Wt, -WtT10, and -G5T5 with ApaLl. SJ-Wt gave 65% perfect joints (64/98). The constructs with homopolymeric ends gave the following percentages of perfect joints: SJWtC10=84.5% (82/97); SJ-A10Wt=94% (17/18); SJWtT10=92% (24/26); and SJ-G5T5=85% (68/90). The small increase in perfect joints derived from constructs with homopolymeric ends is probably not biologically signi®cant. The results suggest that the coding
ends do not in¯uence the ®delity of the signal joint formation when they permit signal joints to be formed. 3.2. Signal joint formation in scid ®broblasts Normal and scid ®broblasts expressing Rag1 and Rag2 proteins are competent to carry out V(D)J recombination (Han et al., 1997). To determine whether the defective signal joint formation described above was preB cell speci®c, we transfected the Wt and G10T10 substrates into scid ®broblasts. DNAs harvested from two dierent transfections of both substrates (experiments 3 and 4) were tested for signal joints by transformation of E. coli (Fig. 3). Signal joints were obtained with the Wt substrate in all ®ve separate transformation assays. The recombination frequency was on average 28 times lower than in the scid preB cells. This low frequency may be due to the poor transfection eciency of the scid ®broblast cell line (not shown). Three recombinants with the Wt substrate were found to be unrelated to V(D)J recombination (Fig. 4). They arose presumably during plasmid replication in E. coli. They are not included in the recombination frequency calculation. Two such recombinants were also found with the G10T10 substrate (Fig. 4). However, not a single, even imperfect, signal joint was obtained with the G10T10 substrate (Fig. 3). Thus, homopolymeric coding ends interfere with signal joint formation in ®broblasts as well as in preB cells. In an attempt to determine if any signal joints were formed with the G10T10 substrate, plasmid DNAs recovered from the scid ®broblasts were analyzed for
556
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
Fig. 4. Imperfect signal joint sequences in scid ®broblasts. Wt3, Wt4 (GT3, GT4): SJ-Wt (SJ-G10T10) plasmids recovered from two ®broblast transfection Nos. 3 and 4; I, II and III are dierent bacterial transformations. The sequences in the top line are part of the two RSS sequences, with the coding ends (not shown) in between. The heptamer of each RSS is underlined. Ctgry, category 1: imperfect joint plasmids; 2: non-V(D)J joint plasmids (see Fig. 3 legend). Italic WT or (G)10 stands for complete coding end sequences and 20 nucleotides beyond (maximinal sequences read) found next to the RSS sequence in the corresponding experiment. A hyphen stands for a nucleotide deletion at the corresponding site. Two plasmids with `' have 34 nucleotides deleted at the 23mer RSS sites.
signal joints by a highly sensitive, non-quantitative PCR assay. Signal joints were detected with both the SJWt and SJG10T10 substrates (Fig. 5). Transfections were repeated four times with similar results. As men-
tioned above, a perfect signal joint generates an ApaLl restriction site. There appear to be a higher percentage of correct Wt signal joints formed with the PCR assay than with the bacterial transformation assay. This is presumably due to the non-quantitative behavior of the PCR assay; correct joints may be more readily ampli®ed. Analysis of the digested products with the G10T10 substrate showed that about 90% of the signal joints were ApaLl sensitive (Fig. 5) indicating that they are authentic, intact signal joints. Since the sensitive PCR assay we used did not allow a quantitative comparison of the levels of signal joints with the Wt and G10T10 substrates, these data only show that if signal joints are formed with the mutant substrate they can be correct.
4. Discussion
Fig. 5. PCR analysis of plasmid DNA recovered from transiently transfected scid ®broblasts. Signal joints are detected with both SJWt and SJG10T10 substrates (lanes 2 and 4). Most of these signal joints are perfect as indicated by their sensitivity to ApaLl digestion (lanes 3 and 5). Lane 1 contains radiolabeled markers (small fragments in the `1 kb' ladder, Gibco).
V(D)J recombination proceeds in two phases, cleavage at the RSSs, followed by processing of the coding and signal ends into coding and signal joints (Lewis, 1994). The cleavage and processing of the joints appears to occur in a DNA-protein complex (Eastman et al., 1996; Ezekiel et al., 1995; Hiom and Gellert, 1998; Zhu et al., 1996). After cleavage all four DNA ends (two coding end hairpins and two blunt signal ends) are apparently held together in this complex (Eastman et al., 1996; Ezekiel et al., 1995; Hiom and
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
Gellert, 1998). Presence of hairpin coding ends and full length signal ends in Ku86 de®cient mice indicates that Ku86 is not required for cleavage and also may not be involved in the protection of the coding and signal ends (Zhu et al., 1996). Ku86 and DNA-PKcs may instead by required for further processing of coding and signal ends (Zhu et al., 1996). The defective signal joint formation with homopolymeric coding ends described here may be due to lack of processing of the signal ends, rather than lack of cleavage. This possibility is supported by the ®nding that RAG1 and RAG2 proteins alone are sucient for cleavage at RSS in vitro (McBlane et al., 1995) and the G10T10 substrates form signal joints at the same frequency as wild type constructs in wild type cells (Ezekiel et al., 1995). The results with the homopolymer containing substrates suggest that the nucleotide composition of the coding ends aects the formation of signal joints in murine scid cells. Murine scid cells completely lack NA-PK activity; in assays with ®broblasts, the scid cells had no detectable activity, while the wild type cells had at least 250 fold higher levels (Blunt et al., 1995). Murine scid cells are competent to form signal joints with wild type substrates (Lieber et al., 1988). This and recent experiments with mice with a targeted disruption of the kinase domain of the scid gene suggest that the kinase activity is not required for the signal joint formation (Taccioli et al., 1998; Gao et al., 1998; Bogue et al., 1998). However, in scid mice and mice that are null for DNA-PKcs, the eciency of signal joint formation is decreased and a high proportion of signal joints are aberrant in that the RSSs are not joined perfectly end-to-end (Lieber et al., 1988; Gao et al., 1998; Bogue et al., 1998). DNA-PKcs may be important by recruiting the joining components to the post-cleavage complex (Bogue et al., 1998). This recruitment function of DNA-PKcs may be required to a much greater degree in the presence of homopolymeric coding ends. The function may depend on the kinase activity. Alternatively, as has been postulated previously (Blunt et al., 1995; Roth et al., 1995), the scid protein may in addition have a scaold function. The scid function may be critical for the processing of signal joints in constructs with homodecamers at both coding ends which may occur at a much slower rate than that of constructs with wild type coding ends. The few, mostly correct, signal joints that can be seen with the G10T10 substrate using a highly sensitive PCR assay support the notion that the scid protein is not absolutely essential, but required for high eciency formation of signal joints associated with unusual coding ends. The dierential signal joint formation in the pentameric and decameric substrates is interesting. Signal joints form readily when only ®ve nucleotide pairs are
557
homopolymers. Apparently, protein interactions can take place when only ®ve nucleotide pairs are homopolymers. This may suggest, that the protein interactions with the post-cleavage complex may not occur with, or may not be aected by the hairpins, but that the important binding site may be the double stranded DNA beyond the hairpins. Alternatively, the overall DNA structure of the stem-loop of the coding end hairpins may be perturbed and may in¯uence signal-end joining when ten homopolymeric nucleotide pairs are present at both coding ends. The substrates with homopolymeric coding ends will be an interesting tool in the complete unraveling of the protein/DNA interactions during V(D)J recombination.
Acknowledgements We thank P. Engler and R. Heuertz for critically reading the manuscript. This work was supported by NIH Grant AI24780. U.R.E. was the recipient of a postdoctoral fellowship from the Cancer Research Institute. R.A. was supported by a NIH Research Supplement for Underrepresented Minority Undergraduates.
References Blunt, T., Finnie, N., Taccioli, G., Smith, G., Demengeot, J., Gottlieb, T., Mizuta, R., Varghese, A., Alt, F., Jeggo, P., Jackson, S., 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±823. Bogue, M.A., Jhappan, C., Roth, D.B., 1998. Analysis of variable (diversity) joining recombination in DNA-dependent protein kinase (DNA-PK)-de®cient mice reveals DNA-PK-independent pathways for both signal and coding joint formation. Proc. Acad. Natl. Sci. USA 95, 15559±15564. Bosma, M.J., Carroll, A.M., 1991. The scid mouse mutant: de®nition, characterization, and potential uses. Ann. Rev. Immunol. 9, 323±344. Boubnov, N.V., Wills, Z.P., Weaver, D.T., 1993. V(D)J recombination coding junction formation without DNA homology: processing of coding termini. Mol. Cell. Biol. 13, 6957±6968. Boubnov, N., Hall, K., Wills, Z., Lee, S., He, D., Benjamin, D., Pulaski, C., Band, H., Reeves, W., Hendrickson, E., Weaver, D., 1995. Complementation of the ionizing radiation sensitivity, DNA end binding, and V(D)J recombination defects of doublestrand break repair mutants by the p86 Ku autoantigen. Proc. Natl. Acad. Sci. USA 92, 890±894. Danska, J., Holland, D., Mariathasan, S., Williams, K., Guidos, C., 1996. Biochemical and genetic defects in the DNA-dependent protein kinase in murine scid lymphocytes. Mol. Cell. Biol. 16, 5507±5517. Eastman, M., Leu, T., Schatz, D., 1996. Initiation of V(D)J recombination in vitro obeying the 12/23 rule. Nature 380, 85±88. Engler, P., Storb, U., 1987. High-frequency deletional rearrangements of immunoglobulin kappa gene segments introduced into a pre-B cell line. Proc. Natl. Acad. Sci. USA 84, 4949±4953. Ezekiel, U., Engler, P., Stern, D., Storb, U., 1995. Asymmetric pro-
558
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
cessing of coding ends and the eect of coding end nucleotide composition on V(D)J recombination. Immunity 2, 381±389. Ezekiel, U., Sun, T., Bozek, G., Storb, U., 1997. The composition of coding joints formed in V(D)J recombination is strongly aected by the nucleotide sequence of the coding ends and their relationship to the recombination signal sequences. Mol. Cell. Bio. 17, 4191±4197. Finnie, N., Gottlieb, T., Blunt, T., Jeggo, P., Jackson, S., 1995. DNA-dependent protein kinase activity is absent in xrs-6 cells: implications for site-speci®c recombination and DNA doublestrand break repair. Proc. Natl. Acad. Sci. USA 92, 320±324. Gao, Y., Chaudhuri, J., Zhu, C., Davidson, L., Weaver, D., Alt, F., 1998. A targeted DNA-PKcs-null mutation reveals DNA-PKindependent functions for ku in V(D)J recombination. Immunity 9, 367±376. Gerstein, R.M., Lieber, M.R., 1993. Coding end sequence can markedly aect the initiation of V(D)J recombination. Genes and Development 7, 1459±1469. Han, J-O., Steen, S.B., Roth, D.B., 1997. Ku86 is not required for protection of signal ends or for formation of nonstandard V(D)J recombination products. Mol. Cell. Biol. 17, 2226±2234. Hiom, K., Gellert, M., 1998. Assembly of a 12/23 paired signal complex: a critical control point in V(D)J recombination. Mol. Cell 1, 1011±1019. Jeggo, P., Taccioli, G., Jackson, S., 1995. Menage a trois: double strand break repair, V(D)J recombination and DNA-PK. Bioessays 17, 949±957. Kirchgessner, C., Patil, C., Evans, J., Cuomo, C., Fried L, Carter, T., Oettinger, M., Brown, J., 1995. DNA-dependent kinase (p350) as a candidate gene for the murine SCID defect. Science 267, 1178±1182. Lewis, S.M., 1994. The mechanism of V(D)J joining: lessons from molecular, immunological and comparative analyses. Adv. Immunol. 56, 27±150. Lieber, M.R., Hesse, J.E., Lewis, S., Bosma, G.C., Rosenberg, N., Mizuuchi, K., Bosma, M.J., Gellert, M., 1988. The defect in murine severe combined immunode®ciency: joining of signal sequences but not coding segments in V(D)J recombination. Cell 55, 7±16. McBlane, F., van Gent, D., Ramsden, D., Romeo, C., Cuomo, C., Gellert, M., Oettinger, M., 1995. Cleavage at a V(D)J recombination signal requires only RAG1 and RAG2 proteins and occurs in two steps. Cell 83, 387±395. Pergola, F., Zdzienicka, M.Z., Lieber, M.R., 1993. V(D)J recombination in mammalian cell mutants defective in DNA double-strand break repair. Mol. Cell. Biol. 13, 3464±3471. Rathmell, W.K., Chu, G., 1994. A DNA end-binding factor involved in double-strand break repair and V(D)J recombination. Mol. Cell. Biol. 14, 4741±4748. Roth, D.B., Menetski, P.B., Nakajima, P.B., Bosma, M.J., Gellert, M., 1992. V(D)J recombination: broken DNA molecules with
covalently sealed (hairpin) coding ends in scid mouse thymocytes. Cell 70, 983±991. Roth, D.B., Zhu, C., Gellert, M., 1993. Characterization of broken DNA molecules associated with V(D)J recombination. Proc. Natl. Acad. Sci. USA 90, 10,788±10,792. Roth, D., Lindahl, T., Gellert, M., 1995. How to make ends meet. Current Biology 5, 496±498. Sadofsky, M.J., Hesse, J.E., McBlane, J.F., Gellert, M., 1993. Expression and V(D)J recombination activity of mutated RAG-1 proteins. Nucleic Acids Research 21, 5644±5650. Sadofsky, M., Hesse, J., van Gent, D., Gellert, M., 1995. RAG1 mutations that aect the target speci®ty of V(D)J recombination: a possible direct role of RAG1 in site recognition. Genes and Developments 9, 2193±2199. Schuler, W., Weiler, I.J., Schuler, A., Philips, R.A., Rosenberg, N., Mak, T.W., Kearney, J.F., Perry, R., Bosma, M.J., 1986. Rearrangement of antigen receptor genes is defective in mice with severe combined immune de®ciency. Cell 46, 963±972. Steen, S.B., Gomelsky, L., Roth, D.B., 1996. The 12/23 rule is enforced at the cleavage step of V(D)J recombination in vivo. Genes to Cells 1, 543±553. Steen, S.B., Gomelsky, L., Speidel, S.L., Roth, D.B., 1997. Initiation of V(D)J recombination in vivo: role of recombination signal sequences in formation of single and paired double-strand breaks. EMBO J. 16, 2656±2664. Taccioli, G.E., Rathbun, G., Olt, E., Stamato, T., Jeggo, P.A., Alt, F.W., 1993. Impairment of V(D)J recombination in double-strand break repair mutants. Science 260, 207±210. Taccioli, G.E., Cheng, H-L., Varghese, A.J., Whitmore, G., Alt, F.W., 1994a. A DNA repair-defect in chinese hamster ovary cells aects V(D)J recombination similarly to the murine scid mutation. J. of Bio. Chem. 269, 7439±7442. Taccioli, G.E., Gottlieb, T.M., Blunt, T., Priestly, A., Demengeot, J., Mizuta, R., Lehmann, A.R., Alt, F.W., Jackson, S.P., Jeggo, P.A., 1994b. Ku80: Product of the XRCC5 gene and its role in DNA repair and V(D)J recombination. Science 265, 1442±1445. Taccioli, G., Amatucci, A., Beamish, H., Gell, D., Xiang, X., Torres, Arzayus M., Priestley, A., Jackson, S., Marshak Rothstein, A., Jeggo, P.A., Herrera, L., 1998. Targeted disruption of the catalytic subunit of the DNA-PK gene in mice confers severe combined immunode®ciency and radiosensitivity. Immunity 9, 355±366. Zdzienicka, M., 1995. Mammalian mutants defective in the response to ionizing radiation-induced DNA damage. Mutation Research 336, 202±215. Zhu, C., Roth, D.B., 1994. Characterization of coding ends in thymocyte of scid mice: implications for the mechanism of V(D)J recombination. Immunity 2, 101±112. Zhu, C., Bogue, M.A., Lim, D-S., Hasty, P., Roth, D.B., 1996. Ku86-De®cient mice exhibit severe combined immunode®ciency and defective processing of V(D)J recombination intermediates. Cell 86, 379±389.
www.elsevier.com/locate/molimm
Signal joint formation is inhibited in murine scid preB cells and ®broblasts in substrates with homopolymeric coding ends Tianhe Sun a, 1, Uthayashanker R. Ezekiel 1, 2, a, 2, Leslie Erskine b, Ryan Agulo a, Grazyna Bozek a, David Roth b, Ursula Storb a,* a
Department of Molecular Genetics and Cell Biology, University of Chicago, 920 East 58th Street, Chicago, IL 60637, USA Department of Microbiology and Immunology, Howard Hughes Medical Institute, Baylor College of Medicine, Houston, TX 77030, USA
b
Received 24 December 1998; received in revised form 15 March 1999; accepted 15 March 1999
Abstract During B and T lymphocyte development, immunoglobulin and T cell receptor genes are assembled from the germline V, (D) and J gene segments (Lewis, S.M., 1994. The mechanism of V(D)J joining: lessons from molecular, immunological and comparative analyses. Adv. Immunol. 56, 27±150). These DNA rearrangements, responsible for immune system diversity, are mediated by a site speci®c recombination machinery via recognition signal sequences (RSSs) composed of conserved heptamers and nonamers separated by spacers of 12 or 23 nucleotides (Lewis, S.M., 1994. The mechanism of V(D)J joining: lessons from molecular, immunological and comparative analyses. Adv. Immunol. 56, 27±150). Recombination occurs only between a RSS with a 12mer spacer and a RSS with a 23mer spacer (Lewis, S.M., 1994. The mechanism of V(D)J joining: lessons from molecular, immunological and comparative analyses. Adv. Immunol. 56, 27±150). RAG1 and RAG2 proteins cleave precisely at the RSS-coding sequence border leading to ¯ush signal ends and coding ends with a hairpin structure (Eastman, M., Leu, T., Schatz, D., 1996. Initiation of V(D)J recombination in vitro obeying the 12/23 rule. Nature 380, 85±88; Roth, D.B., Menetski, J.P., Nakajima, P.B., Bosma, M.J., Gellert, M., 1992. V(D)J recombination: broken DNA molecules with covalently sealed (hairpin) coding ends in scid mouse thymocytes. Cell 983±991; Roth, D.B., Zhu, C., Gellert, M., 1993. Characterization of broken DNA molecules associated with V(D)J recombination. Proc. Natl. Acad. Sci. USA 90, 10,788±10,792; van Gent, D., McBlane, J., Sadofsky, M., Hesse, J., Gellert, M., 1995. Initiation of V(D)J recombination in a cell-free system. Cell 81, 925± 934). Signal ends join, forming a signal joint. The hairpin coding ends are opened by a yet unknown endonuclease, and are further processed to form the coding joint (Lewis, S.M., 1994. The mechanism of V(D)J joining: lessons from molecular, immunological and comparative analyses. Ad. Immunol. 56, 27±150.) The murine scid mutation has been shown to aect coding joints, but much less signal joint formation. In this study we demonstrate that the murine scid mutation inhibits correct signal joint formation when both coding ends contain homopolymeric sequences. We suggest that this ®nding may be due to the function of the SCID protein as an assembly component in V(D)J recombination. # 1999 Elsevier Science Ltd. All rights reserved. Keywords: Immunoglobulin gene rearrangement; Severe combined immunode®ciency; Murine scid
1. Introduction
* Corresponding author. Tel.: +773-702-4440; fax: +773-7023172. E-mail address: [email protected] (U. Storb) 1 This author contributed equally to this study. 2 Current address: Department of Neurology, Washington University Medical Center, P.O. Box 8111, St Louis, MO 63110, USA.
Recent identi®cation of genes and proteins involved in double strand break repair (DSBR) in mammalian systems has elucidated the connection between DSBR and V(D)J recombination (Boubnov et al., 1995; Finnie et al., 1995; Pergola et al., 1993; Rathmell and Chu, 1994; Taccioli et al., 1993, 1994b). The DNA dependent serine/threonine protein kinase holoenzyme
0161-5890/99/$ - see front matter # 1999 Elsevier Science Ltd. All rights reserved. PII: S 0 1 6 1 - 5 8 9 0 ( 9 9 ) 0 0 0 5 3 - X
552
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
Fig. 1. Signal joint plasmid SJ-G10T10. The construction of the plasmid is described in Materials and Methods. The plasmid contains the polyoma origin of replication (Py ori), the polyoma T antigen gene (Py T), and a prokarotic origin of replication (pUC ori). The black dots stand for the promoters of the respective genes. The nucleotides in bold are the heptamer and the nonamer sequences of the RSSs. The small vertical arrows show the ligation sites of the compatible cohesive ends, which are generated by the two indicated restriction enzymes during the plasmid construction.
(DNA-PK) is composed of the catalytic subunit, DNA -PKcs/SCID, encoded by the XRCC7 gene and a heterodimeric Ku protein composed of a p70 subunit and a p86 subunit encoded by XRCC5 (Blunt et al., 1995; Taccioli et al., 1994a, b; Zdzienicka, 1995). Inactivation of the genes encoding p70 or p86 aects both coding and signal joint formation, but the murine scid mutation which eliminates the kinase function (Kirchgessner et al., 1995; Danska et al., 1996) aects mainly coding joint formation (Bosma and Carroll, 1991; Lieber et al., 1988). Hairpin coding ends accumulate in the thymus of scid mice (Roth et al., 1992; Zhu and Roth, 1994). On the basis of the biochemical properties of Ku proteins and DNA-PKcs, several possible functions have been proposed for their involvement in V(D)J recombination (Jeggo et al., 1995; Roth et al., 1995): (1) The Ku proteins bind to the hairpin coding ends and the signal ends thereby protecting them from nucleases. Subsequent binding of DNA-PKcs results in phosphorylation of Ku leading to its dislocation from
the hairpin coding ends and allowing access to the hairpin opening enzyme (Jeggo et al., 1995; Roth et al., 1995). (2) DNA-PK regulates other protein components involved in V(D)J recombination by its phosphorylating activity. (3) Because of its enormous size, the DNA-PKcs (450 Kda) acts as an alignment factor or a scaold on which the V(D)J recombinase and substrate DNA assemble (Jeggo et al., 1995; Roth et al., 1995). We have previously studied the in¯uence of homopolymeric coding ends on V(D)J recombination in normal preB cells (Ezekiel et al., 1995, 1997). Both coding and signal joint formation are strongly aected in frequency and quality by certain homopolymeric coding ends. In this study we have analyzed the formation of signal joints in scid preB cells and scid ®broblasts expressing the Rag1 and Rag2 proteins, using recombination substrates with homopolymeric coding ends which do not aect signal joint formation in normal preB cells.
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
2. Materials and methods 2.1. Cell lines The scid preB cell line S33 was a gift of M. Bosma (Schuler et al., 1986). The scid ®broblast cell line sc3T3 has been described (Han et al., 1997). 2.2. Plasmid constructions The signal joint plasmids SJ-G10Wt, -WtT10, A10Wt, -WtC10, -G10T5, and -G5T5 were all derived from the coding joint construct CJ-G10C10 (Ezekiel et al., 1995) by reversing the 12mer spacer and 23mer spacer rearrangement signal sequences (RSS) in two steps. In the ®rst step the plasmid was cut by Nhel and Sall and the 12mer RSS and adjacent sequences were replaced by an oligonucleotide cassette containing a 23mer RSS and the relevant coding end. The top strand of the oligonucleotide with the wild type coding end is 5'CTAGCGGTTTTTGTACAGCCAGACAG TGGGGATCCACCACTGTGGTGGACGTTCG3 ' and the bottom strand is 5 'TCGAC GAACGTCCACCACAGTGGTGGATCCCCACT GTCTGGCTGTACAAAAACCG3 '. The underlined wild type coding end sequences are changed to the respective homopolymeric sequences and used to make the intermediate constructs. In the second step these intermediate constructs were cut by Spel and Xhol and an oligonucleotide cassette containing a 12mer RSS and the respective coding end was inserted to get the signal joint constructs. The top strand of the oligonucleotide with the wild type coding end sequence is 5 'CTAGTGAGTATCCTCCACAGTGATAGATCTC TGAACAAAAACC3 ' and the bottom strand is 5 'TCGAGGTTTTTGTTCAGAGATCTATCACTGT GGAGGATACTCA3 '. The signal joint plasmid SJ-G10T10 is shown in Fig. 1. 2.3. Transient transfection in preB cells and ®broblasts Transient transfection in scid preB cells S33 was carried out as described (Ezekiel et al., 1995). Transient transfection of ®broblasts was performed as previously described (Sadofsky et al., 1993). Brie¯y, SC3T3 cells were transiently transfected with 5 mg of plasmid substrate (SJWT or SJG10T10), 2.1 mg of full length Rag-1 expression vector (pJH548), and 2.5 mg full length Rag-2 expression vector (pJH549) using the calcium phosphate method. The DNA was harvested from the cells 48 h after transfection as described previously (Steen et al., 1996; Steen et al., 1997). 2.4. Recombination assay The recombination assay was done as described ear-
553
lier (Ezekiel et al., 1995; Fig. 1). In brief, the plasmid substrates contain two bacterial drug resistance marker genes, chloramphenicol acetyl transferase and neomycin phosphotransferase, which confer resistance to chloramphenicol (Cm) and kanamycin (Km), respectively. The neomycin phosphotransferase promoter (PKm) is separated from its coding region by prokaryotic transcription terminators (T1T2) that are ¯anked on either side by a coding end and a RSS. Upon rearrangement in V(D)J recombination competent cells (preB or Rag1/2 expressing ®broblasts), the two RSSs join which leads to the deletion of the V and J coding ends and the terminators. Plasmid DNA is recovered from preB cells or ®broblasts and digested with Dpnl. Replication of plasmid molecules in the preB cells or ®broblasts causes loss of their dam methylation and these replicated molecules are refractory to Dpnl digestion. After Dpnl digestion, the plasmids are transformed into Escherichia coli. Only the recombined plasmids confer resistance to Km. All Km resistant colonies are further checked by polymerase chain reaction (PCR) assay for rearrangement. The percent recombination is calculated from the ratio of Cm plus Km double resistant to Cm resistant colonies, always using the PCR corrected value for the double resistant colonies. 2.5. PCR assay for signal joints In order to detect signal joints, the DNA harvested from the scid ®broblasts was subjected to 24 cycles of PCR using primers to the Km promoter (pKm: 5'CCTCTGGTAAGGTTGGGAAG-3 ') and the Km coding region (neo: 5'-AGCCGAATAGCCTCTCCA CCCAAG-3 ') (Ezekiel et al., 1995). A portion of the PCR products was digested with the restriction enzyme ApaLl, as described (Roth et al., 1993). PCR products were separated on 6% polyacrylamide gels, transferred to nylon membranes, and probed with a 32P-end labeled oligonucleotide which hybridizes to a sequence between the Km primer and the 23 RSS (DR285: 5 'CTGGATGGCTTTCTTGCCGCCAAGG3 '). 3. Results 3.1. Signal joint formation in scid preB cells We and others have shown that coding end nucleotide composition aects V(D)J recombination in wild type cells (Boubnov et al., 1993; Ezekiel et al., 1995; Ezekiel et al., 1997; Gerstein and Lieber, 1993). Furthermore, a particular RAG1 mutation is sensitive to coding end nucleotides (Sadofsky et al., 1995). In this study we have tested the eect of coding end composition on the function of SCID/DNA-PKcs in signal
554
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
Fig. 2. Recombination frequencies for signal joint (SJ) constructs in a scid preB cell line. The ®rst column shows the signal joint constructs, the second column shows the number of experiments, and the last column shows the percent recombination. indicates that not even a single recombinant was found for SJ-G10T10 and SJ-A9C10 in the scid preB cell line. For SJ-G10T10 at least 0.7 106 total bacterial cells were screened for recombinants. Constructs are named by the nucleotide composition of the top strand. For example, plasmid substrate SJ-G10T10 was so named because the 12mer RSS (open triangle) coding end contains ten G nucleotides and the 23mer RSS (closed triangle) coding end has ten T nucleotides. In SJ-Wt (wild type), the ten nucleotides at the 12mer RSS coding end represent Vk and the ten nucleotides at the 23mer RSS coding end represent Jk (Engler and Storb, 1987). Recombination frequency in a normal preB cell line (ABC1.28.13.8) (Ezekiel et al., 1995) follows: SJWt=0.14520.048; SJ-G10T10=0.09420.022; SJ-A9C10=0.02620.009.
joint formation. We transfected a scid preB cell line with extrachromosomal signal joint constructs containing homopolymeric coding ends at the 12mer and 23mer RSS (SJ-G10T10 and its inverse, SJ-A9C10, see Fig. 2) and a construct containing ten nucleotides of wild type coding ends (SJ-Wt) (the Wt coding ends are derived from Vk and Jk sequences; Ezekiel et al., 1995). The orientation of the RSSs is such that upon V(D)J recombination, a signal joint is formed in the plasmid. All three of these constructs have similar rearrangement frequencies in normal preB cells (Ezekiel et al., 1995; see legend Fig. 2). Strikingly, the two homopolymeric constructs do not give detectable rearrangement in scid preB cells, although the wild type
construct gives a similar recombination frequency as in normal preB cells (Fig. 2). Thus, the presence of certain homopolymeric sequences at both coding ends eliminates signal joint formation in cells with the scid mutation. To understand to what extent homopolymeric coding ends in¯uence signal joints, additional constructs containing one homopolymeric coding end and one wild type coding end were tested in scid preB cells (SJG10Wt, -WtT10, -A10Wt, -WtC10, see Fig. 2). These four constructs give recombination frequencies similar to the construct with two wild type coding ends. The result indicates that a single wild type coding end at
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
555
Fig. 3. Transformation of cell extracts recovered from scid ®broblasts. The table shows the transformation results of scid ®broblast cell extracts from the transfections of either SJ-G10T10 or SJ-WT constructs. Four transfections are shown, two for each construct. For the cell extract from each transfection, two or three independent transformations were performed. The `total colonies' stands for the number of single drug resistant 1 bacteria (Chloramphenicol ) in that speci®c transformation. The double drug resistant bacteria were further con®rmed by colony PCR using pkm and neo primers and the numbers are shown in the `PCR positive' column. These recombined plasmids were further analyzed by restriction digestion (ApaLl) and sequencing (for the ApaLl uncut only, i.e. imperfect joints). Three categories were applied to these recombined plasmids. The `perfect joints' stands for the plasmid products which can be digested by ApaLl. The `imperfect joints' stands for the ones with a few nucleotides deletion at the RSS regions. The `non-V(D)J joints' refers to the ones with some deletions away from the RSS regions and thus unrelated to V(D)J recombination. These deletions presumably occurred in the bacteria and deleted the two terminators, giving rise to double drug resistant bacterial clones and shorter PCR products whose sizes are matched with the V(D)J recombined ones. The V(D)J recombination frequency is the ratio of the numbers of the total V(D)J joints (including both perfect and imperfect joints) and the total colonies.
either RSS rescues the homopolymeric eect on signal joint formation in murine scid preB cells. To determine how the number of homopolymeric nucleotides at both coding ends aects signal joint formation, we made two additional constructs (Fig. 2). Construct SJ-G10T5 contains ten G nucleotides at the 5 ' end of the 12mer RSS and ®ve A's at the 5 ' end of the 23mer RSS. Construct SJ-G5T5 contains ®ve G's and A's at the 5' ends of 12 and 23mer RSSs, respectively. Both these constructs give recombination frequencies similar to the wild type construct and the constructs with homopolymeric sequences at one coding end (Fig. 2). Thus, homopolymeric sequences aect signal joint formation in scid cells only when they are present at both coding ends, and when both are longer than ®ve nucleotide pairs. In normal cells the majority of the signal ends join perfectly. The perfect signal joint creates an ApaLl/ HgiAl restriction site (Lieber et al., 1988). Previous studies have shown that in scid preB cells 50% of the signal joints (Lieber et al., 1988) and in scid ®broblasts (expressing RAG1 and RAG2) 80% of the signal joints are perfect (Taccioli et al., 1994a). To test the ®delity of the joints, we digested the recombinant plasmids from SJ-Wt, -WtC10, -A10Wt, -WtT10, and -G5T5 with ApaLl. SJ-Wt gave 65% perfect joints (64/98). The constructs with homopolymeric ends gave the following percentages of perfect joints: SJWtC10=84.5% (82/97); SJ-A10Wt=94% (17/18); SJWtT10=92% (24/26); and SJ-G5T5=85% (68/90). The small increase in perfect joints derived from constructs with homopolymeric ends is probably not biologically signi®cant. The results suggest that the coding
ends do not in¯uence the ®delity of the signal joint formation when they permit signal joints to be formed. 3.2. Signal joint formation in scid ®broblasts Normal and scid ®broblasts expressing Rag1 and Rag2 proteins are competent to carry out V(D)J recombination (Han et al., 1997). To determine whether the defective signal joint formation described above was preB cell speci®c, we transfected the Wt and G10T10 substrates into scid ®broblasts. DNAs harvested from two dierent transfections of both substrates (experiments 3 and 4) were tested for signal joints by transformation of E. coli (Fig. 3). Signal joints were obtained with the Wt substrate in all ®ve separate transformation assays. The recombination frequency was on average 28 times lower than in the scid preB cells. This low frequency may be due to the poor transfection eciency of the scid ®broblast cell line (not shown). Three recombinants with the Wt substrate were found to be unrelated to V(D)J recombination (Fig. 4). They arose presumably during plasmid replication in E. coli. They are not included in the recombination frequency calculation. Two such recombinants were also found with the G10T10 substrate (Fig. 4). However, not a single, even imperfect, signal joint was obtained with the G10T10 substrate (Fig. 3). Thus, homopolymeric coding ends interfere with signal joint formation in ®broblasts as well as in preB cells. In an attempt to determine if any signal joints were formed with the G10T10 substrate, plasmid DNAs recovered from the scid ®broblasts were analyzed for
556
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
Fig. 4. Imperfect signal joint sequences in scid ®broblasts. Wt3, Wt4 (GT3, GT4): SJ-Wt (SJ-G10T10) plasmids recovered from two ®broblast transfection Nos. 3 and 4; I, II and III are dierent bacterial transformations. The sequences in the top line are part of the two RSS sequences, with the coding ends (not shown) in between. The heptamer of each RSS is underlined. Ctgry, category 1: imperfect joint plasmids; 2: non-V(D)J joint plasmids (see Fig. 3 legend). Italic WT or (G)10 stands for complete coding end sequences and 20 nucleotides beyond (maximinal sequences read) found next to the RSS sequence in the corresponding experiment. A hyphen stands for a nucleotide deletion at the corresponding site. Two plasmids with `' have 34 nucleotides deleted at the 23mer RSS sites.
signal joints by a highly sensitive, non-quantitative PCR assay. Signal joints were detected with both the SJWt and SJG10T10 substrates (Fig. 5). Transfections were repeated four times with similar results. As men-
tioned above, a perfect signal joint generates an ApaLl restriction site. There appear to be a higher percentage of correct Wt signal joints formed with the PCR assay than with the bacterial transformation assay. This is presumably due to the non-quantitative behavior of the PCR assay; correct joints may be more readily ampli®ed. Analysis of the digested products with the G10T10 substrate showed that about 90% of the signal joints were ApaLl sensitive (Fig. 5) indicating that they are authentic, intact signal joints. Since the sensitive PCR assay we used did not allow a quantitative comparison of the levels of signal joints with the Wt and G10T10 substrates, these data only show that if signal joints are formed with the mutant substrate they can be correct.
4. Discussion
Fig. 5. PCR analysis of plasmid DNA recovered from transiently transfected scid ®broblasts. Signal joints are detected with both SJWt and SJG10T10 substrates (lanes 2 and 4). Most of these signal joints are perfect as indicated by their sensitivity to ApaLl digestion (lanes 3 and 5). Lane 1 contains radiolabeled markers (small fragments in the `1 kb' ladder, Gibco).
V(D)J recombination proceeds in two phases, cleavage at the RSSs, followed by processing of the coding and signal ends into coding and signal joints (Lewis, 1994). The cleavage and processing of the joints appears to occur in a DNA-protein complex (Eastman et al., 1996; Ezekiel et al., 1995; Hiom and Gellert, 1998; Zhu et al., 1996). After cleavage all four DNA ends (two coding end hairpins and two blunt signal ends) are apparently held together in this complex (Eastman et al., 1996; Ezekiel et al., 1995; Hiom and
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
Gellert, 1998). Presence of hairpin coding ends and full length signal ends in Ku86 de®cient mice indicates that Ku86 is not required for cleavage and also may not be involved in the protection of the coding and signal ends (Zhu et al., 1996). Ku86 and DNA-PKcs may instead by required for further processing of coding and signal ends (Zhu et al., 1996). The defective signal joint formation with homopolymeric coding ends described here may be due to lack of processing of the signal ends, rather than lack of cleavage. This possibility is supported by the ®nding that RAG1 and RAG2 proteins alone are sucient for cleavage at RSS in vitro (McBlane et al., 1995) and the G10T10 substrates form signal joints at the same frequency as wild type constructs in wild type cells (Ezekiel et al., 1995). The results with the homopolymer containing substrates suggest that the nucleotide composition of the coding ends aects the formation of signal joints in murine scid cells. Murine scid cells completely lack NA-PK activity; in assays with ®broblasts, the scid cells had no detectable activity, while the wild type cells had at least 250 fold higher levels (Blunt et al., 1995). Murine scid cells are competent to form signal joints with wild type substrates (Lieber et al., 1988). This and recent experiments with mice with a targeted disruption of the kinase domain of the scid gene suggest that the kinase activity is not required for the signal joint formation (Taccioli et al., 1998; Gao et al., 1998; Bogue et al., 1998). However, in scid mice and mice that are null for DNA-PKcs, the eciency of signal joint formation is decreased and a high proportion of signal joints are aberrant in that the RSSs are not joined perfectly end-to-end (Lieber et al., 1988; Gao et al., 1998; Bogue et al., 1998). DNA-PKcs may be important by recruiting the joining components to the post-cleavage complex (Bogue et al., 1998). This recruitment function of DNA-PKcs may be required to a much greater degree in the presence of homopolymeric coding ends. The function may depend on the kinase activity. Alternatively, as has been postulated previously (Blunt et al., 1995; Roth et al., 1995), the scid protein may in addition have a scaold function. The scid function may be critical for the processing of signal joints in constructs with homodecamers at both coding ends which may occur at a much slower rate than that of constructs with wild type coding ends. The few, mostly correct, signal joints that can be seen with the G10T10 substrate using a highly sensitive PCR assay support the notion that the scid protein is not absolutely essential, but required for high eciency formation of signal joints associated with unusual coding ends. The dierential signal joint formation in the pentameric and decameric substrates is interesting. Signal joints form readily when only ®ve nucleotide pairs are
557
homopolymers. Apparently, protein interactions can take place when only ®ve nucleotide pairs are homopolymers. This may suggest, that the protein interactions with the post-cleavage complex may not occur with, or may not be aected by the hairpins, but that the important binding site may be the double stranded DNA beyond the hairpins. Alternatively, the overall DNA structure of the stem-loop of the coding end hairpins may be perturbed and may in¯uence signal-end joining when ten homopolymeric nucleotide pairs are present at both coding ends. The substrates with homopolymeric coding ends will be an interesting tool in the complete unraveling of the protein/DNA interactions during V(D)J recombination.
Acknowledgements We thank P. Engler and R. Heuertz for critically reading the manuscript. This work was supported by NIH Grant AI24780. U.R.E. was the recipient of a postdoctoral fellowship from the Cancer Research Institute. R.A. was supported by a NIH Research Supplement for Underrepresented Minority Undergraduates.
References Blunt, T., Finnie, N., Taccioli, G., Smith, G., Demengeot, J., Gottlieb, T., Mizuta, R., Varghese, A., Alt, F., Jeggo, P., Jackson, S., 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±823. Bogue, M.A., Jhappan, C., Roth, D.B., 1998. Analysis of variable (diversity) joining recombination in DNA-dependent protein kinase (DNA-PK)-de®cient mice reveals DNA-PK-independent pathways for both signal and coding joint formation. Proc. Acad. Natl. Sci. USA 95, 15559±15564. Bosma, M.J., Carroll, A.M., 1991. The scid mouse mutant: de®nition, characterization, and potential uses. Ann. Rev. Immunol. 9, 323±344. Boubnov, N.V., Wills, Z.P., Weaver, D.T., 1993. V(D)J recombination coding junction formation without DNA homology: processing of coding termini. Mol. Cell. Biol. 13, 6957±6968. Boubnov, N., Hall, K., Wills, Z., Lee, S., He, D., Benjamin, D., Pulaski, C., Band, H., Reeves, W., Hendrickson, E., Weaver, D., 1995. Complementation of the ionizing radiation sensitivity, DNA end binding, and V(D)J recombination defects of doublestrand break repair mutants by the p86 Ku autoantigen. Proc. Natl. Acad. Sci. USA 92, 890±894. Danska, J., Holland, D., Mariathasan, S., Williams, K., Guidos, C., 1996. Biochemical and genetic defects in the DNA-dependent protein kinase in murine scid lymphocytes. Mol. Cell. Biol. 16, 5507±5517. Eastman, M., Leu, T., Schatz, D., 1996. Initiation of V(D)J recombination in vitro obeying the 12/23 rule. Nature 380, 85±88. Engler, P., Storb, U., 1987. High-frequency deletional rearrangements of immunoglobulin kappa gene segments introduced into a pre-B cell line. Proc. Natl. Acad. Sci. USA 84, 4949±4953. Ezekiel, U., Engler, P., Stern, D., Storb, U., 1995. Asymmetric pro-
558
T. Sun et al. / Molecular Immunology 36 (1999) 551±558
cessing of coding ends and the eect of coding end nucleotide composition on V(D)J recombination. Immunity 2, 381±389. Ezekiel, U., Sun, T., Bozek, G., Storb, U., 1997. The composition of coding joints formed in V(D)J recombination is strongly aected by the nucleotide sequence of the coding ends and their relationship to the recombination signal sequences. Mol. Cell. Bio. 17, 4191±4197. Finnie, N., Gottlieb, T., Blunt, T., Jeggo, P., Jackson, S., 1995. DNA-dependent protein kinase activity is absent in xrs-6 cells: implications for site-speci®c recombination and DNA doublestrand break repair. Proc. Natl. Acad. Sci. USA 92, 320±324. Gao, Y., Chaudhuri, J., Zhu, C., Davidson, L., Weaver, D., Alt, F., 1998. A targeted DNA-PKcs-null mutation reveals DNA-PKindependent functions for ku in V(D)J recombination. Immunity 9, 367±376. Gerstein, R.M., Lieber, M.R., 1993. Coding end sequence can markedly aect the initiation of V(D)J recombination. Genes and Development 7, 1459±1469. Han, J-O., Steen, S.B., Roth, D.B., 1997. Ku86 is not required for protection of signal ends or for formation of nonstandard V(D)J recombination products. Mol. Cell. Biol. 17, 2226±2234. Hiom, K., Gellert, M., 1998. Assembly of a 12/23 paired signal complex: a critical control point in V(D)J recombination. Mol. Cell 1, 1011±1019. Jeggo, P., Taccioli, G., Jackson, S., 1995. Menage a trois: double strand break repair, V(D)J recombination and DNA-PK. Bioessays 17, 949±957. Kirchgessner, C., Patil, C., Evans, J., Cuomo, C., Fried L, Carter, T., Oettinger, M., Brown, J., 1995. DNA-dependent kinase (p350) as a candidate gene for the murine SCID defect. Science 267, 1178±1182. Lewis, S.M., 1994. The mechanism of V(D)J joining: lessons from molecular, immunological and comparative analyses. Adv. Immunol. 56, 27±150. Lieber, M.R., Hesse, J.E., Lewis, S., Bosma, G.C., Rosenberg, N., Mizuuchi, K., Bosma, M.J., Gellert, M., 1988. The defect in murine severe combined immunode®ciency: joining of signal sequences but not coding segments in V(D)J recombination. Cell 55, 7±16. McBlane, F., van Gent, D., Ramsden, D., Romeo, C., Cuomo, C., Gellert, M., Oettinger, M., 1995. Cleavage at a V(D)J recombination signal requires only RAG1 and RAG2 proteins and occurs in two steps. Cell 83, 387±395. Pergola, F., Zdzienicka, M.Z., Lieber, M.R., 1993. V(D)J recombination in mammalian cell mutants defective in DNA double-strand break repair. Mol. Cell. Biol. 13, 3464±3471. Rathmell, W.K., Chu, G., 1994. A DNA end-binding factor involved in double-strand break repair and V(D)J recombination. Mol. Cell. Biol. 14, 4741±4748. Roth, D.B., Menetski, P.B., Nakajima, P.B., Bosma, M.J., Gellert, M., 1992. V(D)J recombination: broken DNA molecules with
covalently sealed (hairpin) coding ends in scid mouse thymocytes. Cell 70, 983±991. Roth, D.B., Zhu, C., Gellert, M., 1993. Characterization of broken DNA molecules associated with V(D)J recombination. Proc. Natl. Acad. Sci. USA 90, 10,788±10,792. Roth, D., Lindahl, T., Gellert, M., 1995. How to make ends meet. Current Biology 5, 496±498. Sadofsky, M.J., Hesse, J.E., McBlane, J.F., Gellert, M., 1993. Expression and V(D)J recombination activity of mutated RAG-1 proteins. Nucleic Acids Research 21, 5644±5650. Sadofsky, M., Hesse, J., van Gent, D., Gellert, M., 1995. RAG1 mutations that aect the target speci®ty of V(D)J recombination: a possible direct role of RAG1 in site recognition. Genes and Developments 9, 2193±2199. Schuler, W., Weiler, I.J., Schuler, A., Philips, R.A., Rosenberg, N., Mak, T.W., Kearney, J.F., Perry, R., Bosma, M.J., 1986. Rearrangement of antigen receptor genes is defective in mice with severe combined immune de®ciency. Cell 46, 963±972. Steen, S.B., Gomelsky, L., Roth, D.B., 1996. The 12/23 rule is enforced at the cleavage step of V(D)J recombination in vivo. Genes to Cells 1, 543±553. Steen, S.B., Gomelsky, L., Speidel, S.L., Roth, D.B., 1997. Initiation of V(D)J recombination in vivo: role of recombination signal sequences in formation of single and paired double-strand breaks. EMBO J. 16, 2656±2664. Taccioli, G.E., Rathbun, G., Olt, E., Stamato, T., Jeggo, P.A., Alt, F.W., 1993. Impairment of V(D)J recombination in double-strand break repair mutants. Science 260, 207±210. Taccioli, G.E., Cheng, H-L., Varghese, A.J., Whitmore, G., Alt, F.W., 1994a. A DNA repair-defect in chinese hamster ovary cells aects V(D)J recombination similarly to the murine scid mutation. J. of Bio. Chem. 269, 7439±7442. Taccioli, G.E., Gottlieb, T.M., Blunt, T., Priestly, A., Demengeot, J., Mizuta, R., Lehmann, A.R., Alt, F.W., Jackson, S.P., Jeggo, P.A., 1994b. Ku80: Product of the XRCC5 gene and its role in DNA repair and V(D)J recombination. Science 265, 1442±1445. Taccioli, G., Amatucci, A., Beamish, H., Gell, D., Xiang, X., Torres, Arzayus M., Priestley, A., Jackson, S., Marshak Rothstein, A., Jeggo, P.A., Herrera, L., 1998. Targeted disruption of the catalytic subunit of the DNA-PK gene in mice confers severe combined immunode®ciency and radiosensitivity. Immunity 9, 355±366. Zdzienicka, M., 1995. Mammalian mutants defective in the response to ionizing radiation-induced DNA damage. Mutation Research 336, 202±215. Zhu, C., Roth, D.B., 1994. Characterization of coding ends in thymocyte of scid mice: implications for the mechanism of V(D)J recombination. Immunity 2, 101±112. Zhu, C., Bogue, M.A., Lim, D-S., Hasty, P., Roth, D.B., 1996. Ku86-De®cient mice exhibit severe combined immunode®ciency and defective processing of V(D)J recombination intermediates. Cell 86, 379±389.