In lymphocyte ontogeny, T and B cell precursors rearrange their antigen receptor or immunoglobulin gene loci to form diverse cell surface receptors by a site-specific recombination process, V(D)J recombination. While V(D)J recombination has been analyzed in detail as a developmentally regulated process in vivo, much less is known about the biochemical processes which drive V(D)J recombination (for reviews, see Lieber(1991, 1992), Gellert(1992), Schatz et al.(1992), Taccioli et al.(1992), Jessberger(1994)).

Several proteins have been detected to bind and possibly bend DNA sequences which contain the V(D)J signal sequences (Aguilera et al., 1987; Li et al., 1989; Matsunami et al., 1989; Andrews et al., 1993; Wu et al., 1993b). A direct function in recombination, however, could not yet be assigned to these proteins.

Two genes (recombination activating genes (RAG-1 and RAG-2) were identified which, following transfer into fibroblasts, mediate V(D)J recombination of plasmid DNA substrates (Schatz et al., 1989; Oettinger et al., 1990). The crucial role of these genes has been demonstrated in mice which carry a homozygous mutation in either the RAG-1 (Mombaerts et al., 1992) or the RAG-2 (Shinkai et al., 1992) gene. RAG-deficient lymphocyte precursors do not initiate the rearrangement of their antigen receptor genes and remain arrested at early stages of development. However, it is still unknown, whether the RAG genes directly participate in the recombination reaction or whether they are signal factors which indirectly trigger the activation of a recombination machinery.

Although not essential for V(D)J rearrangement, the only defined enzyme which has been shown to take part in the reaction is terminal deoxynucleotidyltransferase (Landau et al., 1987; Komori et al., 1993), which creates the N-region diversity by adding nucleotide monophosphates to the coding ends at the V to J and J to D junctions.

Taken together, the enzymology of V(D)J joining is still largely unknown. Similarly, little is known about the mammalian enzymatic activities which are involved in DNA double-strand break repair through end-joining or homologous recombination (West, 1992).

Another phenotype similar to that one observed in RAG-deficient mice is found in the well characterized Severe Combined Immune Deficiency syndrome (SCID) (for review, see Bosma and Carroll(1991)). SCID thymocytes and early B cells are developmentally arrested at early, immature stages. In contrast to the RAG deficiencies, in SCID lymphocytes V(D)J joining is initiated, but no functional coding joints are formed (Lieber et al., 1988b). Large, unrepaired deletions around the V(D)J recombination sites have been observed in SCID cells. Signal joint formation occurs, but is also altered such that a high percentage of imprecise signal joints are formed (Lieber et al., 1988b). The SCID mutation is thought to affect a late step in the V(D)J joining pathway (Malynn et al., 1988), probably the resolution of reaction intermediates in coding joint formation (Roth et al., 1992a, 1992b). Very recently, evidence was obtained for the identification of the SCID gene as the gene for DNA-dependent protein kinase (Carter et al., 1990; Lees-Miller et al., 1990). ()()

A number of investigators have described another defect associated with the SCID mutation: an elevated sensitivity to ionizing radiation and other agents which generate DNA double-strand breaks (DSBs). ()SCID fibroblasts, myeloid, intestinal crypt, and bone marrow cells are not able to repair DNA double-strand breaks as efficiently as wild-type cells (Fulop and Philips, 1990; Biedermann et al., 1991; Hendrickson et al., 1991; Pergola et al., 1993; Taccioli et al., 1993). These results have been interpreted as indications for a possible link between DSB repair and V(D)J recombination.

Two other genetic complementation groups of non-lymphocyte hamster cell lines have been described which also display a deficiency in both, the repair of x-ray-induced DSBs and V(D)J recombination (Pergola et al., 1993; Taccioli et al., 1993). Cells of the xrs group (group 5) lack a DNA end-binding factor identified as protein Ku (Paillard and Strauss, 1991; Getts and Stamato, 1994; Rathmell and Chu, 1994; Smider et al., 1994; Taccioli et al., 1994). Ku protein is composed of an 86- and a 70-kDa subunit, binds to a variety of DNA structures, and is associated with DNA-dependent protein kinase (Gottlieb and Jackson, 1993). SCID and the V3 hamster cell line constitute one other complementation group (group 9), and the third group is represented by XR-1 (group 4).

There are at least two ways in which a cell might repair DNA double-strand breaks. One is end-joining, the religation of DNA ends of various structures without the possibility to repair deletions or insertions at the site of the break. The alternative is the repair of DSBs by recombination with the homologous allele. Homologous recombinational repair, although more complex than rejoining of the ends, is the only known way to accurately repair deletions or gaps, which might have been generated by nucleolytic removal of deoxynucleotides at the DNA ends of a DSB (Szostak et al., 1983; Haber, 1992). It should be noted that despite a possible factor-sharing, homologous, nonhomologous, and V(D)J recombination remain three distinct reactions with unique and independent features.

Recently, a cell-free assay for intermolecular DNA recombination was developed (Jessberger and Berg, 1991). This assay, named DNA transfer assay (DTA), directly measures the transfer and stable incorporation of donor plasmid DNA into recipient plasmid DNA in vitro. In the set up reported here, the recipient DNA molecules bear a deletion either with or without a double-strand break (DSB) at that site. Transfer of homologous donor DNA is an essential feature of all pathways of homologous recombination as described by the double-strand break repair model (Szostak et al., 1983), the single-strand annealing model (Lin et al., 1990a, 1990b), or the Meselson-Radding model (Meselson and Radding, 1975). Transfer of DNA strands may occur also during heterologous recombination including end-joining.

In the DTA reaction, tritium-labeled donor DNA becomes covalently incorporated into the digoxigeninylated recipient DNA. The recipient DNA is subsequently recovered through binding to affinity beads. The radioactivity, originating from the donor DNA, and present after transfer in the recipient DNA on the affinity beads, allows the direct measurement of the DNA transfer between the two DNAs (Jessberger and Berg, 1991). The DTA measures the sum of all the recombination events leading to a stable strand transfer (see ``Experimental Procedures'' and ``Discussion''). Depending on the nature of the DNA substrates and protein fractions, these reactions include, but are not limited to, crossover and non-crossover pathways of homologous recombination (Jessberger and Berg, 1991; Jessberger et al., 1993), as well as end-joining, heterologous (Jessberger and Berg, 1991), and sequence-specific reactions. ()

The DTA has been used to purify and characterize a high molecular weight protein complex (RC-1; recombination complex 1) from fetal calf thymus, which catalyzes DNA recombinational repair of gaps and deletions in homologous plasmid DNA substrates (Jessberger et al., 1993).

An in vitro assay measuring the activity of nuclear extracts to mediate intermolecular DNA transfer should facilitate the molecular analysis of mutations in the enzymatic machinery which affect DSB repair such as the SCID mutation. The recombination deficiency of SCID is reproduced in the cell-free system, where nuclear extracts prepared from SCID lymphocytes and fibroblasts show significantly decreased activity. A protein (SRSP; Scid Recombination Stimulatory Protein) has been purified from normal mouse thymus which stimulates the DTA activity specifically in SCID thymocyte and proB cell nuclear extracts. Highly purified SRSP appears as a protein of approximately 72 kDa in SDS-polyacrylamide gel electrophoresis. SRSP has no obvious functional relationship to a variety of DNA metabolic enzymes.

EXPERIMENTAL PROCEDURES

Mice and Cells

Antibodies

Isolation of Thymocyte Subpopulations

Subsequently, enriched DN populations were further purified into CD3CD4CD8 thymocytes by fluorescence-activated cell sorting (see below).

Immunofluorescence Staining and Cell Sorting

The bone marrow-derived pro-, pre-, immature, and mature B cells were sorted as follows: bone marrow was deprived of sIg B cells by the use of sheep anti-mouse Ig-coated Dyna beads (Dynal A.S., Oslo, Norway) as recommended by the supplier. The remaining cells were stained with the FITC-labeled monoclonal antibody RA3-6B2 (anti-B220; Pharmingen) and phycoerythrin-coupled Acky (anti-c-kit; McCormack et al.(1989)), and separated into B220, c-kit proB cells and B220, c-kit preB cells. For sorting of immature and mature B cells, total bone marrow cells were stained with FITC-labeled goat anti-mouse IgM (Southern Biotechnology Associates, Birmingham, AL) and the phycoerythrin-coupled monoclonal antibody 1.19 (anti-IgD). Immature sIgM, IgD B cells and mature sIgM, IgD, and the proB and preB cells were sorted using the FACStar Plus cell sorter.

Growth and Differentiation of Stromal Cells and IL-7-dependent preB Cell Lines

Preparation of Nuclear Extracts

For preparation of nuclear extracts from organs and tissues, the material (between 0.5 and 15 g) was washed once in buffer A (1500 rpm centrifugation at 4 °C for 3 min.), homogenized in a blender to produce a cell suspension, again washed in buffer A (2500 rpm centrifugation at 4 °C for 5 min), and the cells were resuspended in buffer B. They were then treated as described above, except that the volumina had to be adopted respectively. The protein concentrations were between 0.1 and 0.5 mg/ml. The tissues and organs were derived from strain- and age-matched mice.

The nuclear extracts did contain low amounts of nucleic acids (<2%) as judged by optical density measurements at 260 and 280 nm (Warburg and Christian, 1941), presumably mostly ribonucleic acids. As often experienced for nuclear salt extracts from mammalian cells, no major double-strand DNA-degrading activities were present.

DNA Transfer Assay (DTA)

The DNA substrates, their modification by digoxigeninylation, or in vivo labeling with [H]thymidine, the preparation of antidigoxigenin polyacrylamide beads, and the bacterial strains used in these studies have been described before (Jessberger and Berg, 1991). For most experiments, the donor DNA was uniformly H-labeled pSV2neo supercoiled plasmid DNA, the recipient DNA was a deletion derivative of pSV2neo, bearing a deletion of 70 bp or 248 bp in length (pSV2neo70 or -248). Unless otherwise noted, both DNAs were used in their circular form in equal amounts of 0.1 µg of each substrate. For some experiments, the recipient DNA has been linearized at the site of the deletion by treatment with XhoI restriction endonuclease.

The DTA measures the stable transfer of radioactively labeled donor DNA into the recipient DNA. Tritium-labeled donor DNA was incubated in the nuclear extract (0.03-1.2 µg of protein) with digoxigeninylated recipient DNA in a solution containing 10 mM EPPS, pH 7.8, 15 mM ammonium sulfate, 5 mM MgCl, 1 mM dithiothreitol, 0.1 mM spermidine, 0.1 mM concentration of each of the four dNTPs, and 1 mM ATP.

The reaction was incubated for 30 min at 37 °C, terminated by the addition of SDS to 0.05% and EDTA to 80 mM, and the DNA was re-extracted by phenol:chloroform (1:10). Heating of the protein-free DNA in the presence of SDS and EDTA to 70 °C or 95 °C did not significantly affect the results. The recipient DNA was then bound to anti-digoxigenin antibodies covalently attached to polyacrylamide beads. After extensive washing with phosphate-buffered saline, 0.05% Tween 20, the amount of radioactivity associated with the recipient DNA was determined by counting the bead-bound DNA in a scintillation counter. DTA activity is expressed as percentage of input H radioactivity stably transferred into the recipient DNA on the beads. The DTA measures the total of all DNA strand transfer reactions, which take place in a given sample. The complete repair of deletions and gaps in the recipient DNA has been shown by polymerase chain reaction analysis, where crossover and non-crossover products of homologous recombination have been observed (Jessberger and Berg, 1991; Jessberger et al., 1993). Recombination between nonhomologous DNA substrates, end-joining, and recombination stimulated by specific DNA sequences can also be measured in the assay. Thus, if any one of the above-mentioned recombination reactions comprises a significant proportion of the total strand transfer events in a particular set up, the assay will detect a deficiency in that pathway.

DNA Polymerase, Topoisomerase, and Nuclease Assays

X-ray Survival Assay

Purification of SRSP

Other Methods

RESULTS

Recombination Activity in Normal and Mutant Thymocytes

Figure 1: Cell-free DTA activity on circular plasmid DNA substrates, pSV2neo70 and pSV2neo, in thymus extracts from normal, RAG-2, and SCID mice. Various amounts of nuclear extract protein were incubated in the standard DTA reaction for 30 min at 37 °C. The activity is expressed as percentage of input H radioactivity being transferred into the bead-bound recipient DNA.

Both normal and RAG-2 thymus extracts show linear dose-response curves for the recombination activity (Fig. 1). In marked contrast, SCID-derived nuclear extracts did only weakly catalyze the reaction over the range of protein concentrations tested. The background in this assay was 0.1% H counts/min (50-100 cpm).

To control the functional (enzymatic) integrity of the protein preparations, the extracts were analyzed in DNA polymerase assays (see ``Experimental Procedures''). All three extracts showed nearly identical levels of total DNA polymerase activity (0.33 ± 0.05 pmol incorporation of dNMP per 0.5 µg of protein extract). In addition, the DNA topoisomerase I and II activities were tested and found to be similar in all three extracts (data not shown). In addition, nuclease assays were performed and demonstrated very low double-strand endo- and exonuclease activities in the nuclear extracts described here (data not shown). Comparable assays have been carried out with all nuclear extracts described in this communication.

The use of recipient DNA which had been linearized at the site of the deletion (pSV2neo70-XhoI) did not yield significantly different results: 34% counts/min in normal and 5% counts/min in SCID extracts (1.2 µg of protein each) compared with 46% counts/min and 4.5% counts/min for circular substrates.

Recipient DNA substrates with or without homology to the pSV2neo donor DNA were compared in the DTA using normal and SCID thymus extracts (Table 2). Two homologous (pSV2neo70 and -248) and three heterologous (RF form of M13mp10, X174, and ) recipient DNA substrates were used. Overall, wild-type extracts are on average 19-fold more active than SCID extracts. Both normal and SCID extracts showed a 2- to 3-fold preference for homologous DNA substrates.

The cytoplasmic fraction of the extracts was also tested for recombination activity but found inactive in every case.

These results show a specific deficiency in recombination in the SCID thymus-derived cell extracts as determined by the DTA.

Recombination Activity in Thymocyte Subsets Representing Distinct Stages of Development

To analyze whether lymphocyte precursors undergoing V(D)J recombination simultaneously express elevated levels of strand transfer activity, nuclear extracts were prepared from purified TCR (CD3-negative) DN, DP, and CD4 SP wild-type thymocytes (see ``Experimental Procedures''). Subsequently, equal amounts of nuclear protein extracts, 0.1 µg each, were tested in the recombination reaction (Fig. 2). The most active extract was derived from DN cells, followed by the DP subset, while extracts prepared from the most mature SP cells showed the lowest activity. Data in Fig. 2represent one of three experiments.

Figure 2: Cell-free DTA activity in sorted thymocyte populations. Thymocytes from normal mice were sorted into the CD4/CD8 double negative (DN), double positive (DP), and single positive (SP) populations, and their nuclear extracts, 0.1 µg of protein each, were tested in the standard DTA reaction. See ``Experimental Procedures'' for details.

Hence, there is a directly measurable DNA transfer activity in nuclear extracts with the highest levels at the earliest stage and decreasing activity with progression of development. However, SCID thymocytes, which are of the DN phenotype, did not yield any significant recombination activity (Fig. 1; Table 2).

Recombination Activity during B Lymphocyte Development

Figure 3: Cell-free DTA activity in sorted B cells from normal, SCID, and RAG-2 mice. The sorting, extract preparation, and DTA procedures are described under ``Experimental Procedures.'' From each extract, 50 ng of protein was tested.

B cell development in both RAG-2 and SCID mice is arrested at the proB cell stage (B220IgM). To distinguish whether the observed DNA recombination activity in early B cell precursors is correlated with active V(D)J rearrangements or early development, and to look for a recombination deficiency in SCID B cells, proB cells were purified by cell sorting from SCID and RAG-2 mice. Subsequently, each nuclear extract was analyzed for its activity in the DTA (Fig. 3). proB cells from RAG-2 mice demonstrate an activity roughly comparable to normal proB cells. In contrast, SCID-derived proB cells are inefficient in performing the recombination reaction.

V(D)J recombination is initiated at the proB stage. To further test possible correlations between site-specific V(D)J joining and the DTA reaction, we analyzed B cell differentiation in a cell culture system. To this end, B cell precursors were cultured on stromal cells in the presence of IL-7 which allows the maintenance of proB cells arrested in an early stage of differentiation (Rolink et al., 1993). These proB cells are V(D)J recombination inactive and only weakly express the RAG genes. Withdrawal of IL-7, however, causes the induction of differentiation including an up-regulation of RAG-1 and RAG-2 expression, a stop in proliferation, and activation of V(D)J rearrangement. To prevent cell death by IL-7 withdrawal, early B cell progenitors from bcl-2 transgenic mice were used (Rolink et al., 1993).

Nuclear extracts were prepared from cells prior to (day 0) and at several days after induction of differentiation following IL-7 withdrawl from proB cell cultures (days 1 to 5) and tested for cell-free DNA recombination activity (Fig. 4). In two experiments, different amounts of protein (0.21 µg and 0.12 µg, respectively) have been used. The day 0 extracts had an activity comparable to those observed with primary T and B cell extracts (Fig. 1, Fig. 2, and Fig. 3). We observed an increase in DTA activity after induction with a peak at day 2. The activity was enhanced about 7-fold compared to day 0, and decreased afterwards to starting levels. This induction of cell-free recombination activity temporally correlates with the induction of V(D)J joining and up-regulation of the expression of various genes, including RAG-1 and RAG-2, as measured in the same cell culture system (Rolink et al., 1993).

Figure 4: Cell-free DTA activity in the proB cell line bcl-2-5 (Rolink et al., 1993). Extracts were prepared before and after induction of differentiation by withdrawal of IL-7 and 0.21 µg (Exp. 1) or 0.12 µg (Exp. 2) of protein per reaction was tested for activity in the DTA system.

Thus, with ongoing differentiation of normal B cells from proB cells through preB cells, immature B cells to mature B cells, the activity decreases. This parallels the activity pattern seen throughout thymocyte development. Both profiles correlate kinetically roughly with site-specific V(D)J recombination activity in lymphocyte ontogeny in vivo (Kronenberg et al., 1986; Lieber et al., 1987; Rolink and Melchers, 1993). However, the observed DNA transfer activity in early lymphocyte extracts is not by itself induced by V(D)J rearrangement since early thymocytes as well as early B cells from rearrangement-incompetent RAG-2-deficient mice express wild-type levels of recombination activity on the homologous DNA substrates used in the DTA (see ``Discussion'').

Recombination Activity in Wild-type and SCID-derived Fibroblast Cell Extracts

Figure 5: A, x-ray irradiation survival assay on CB.17 and CB.17scid fibroblast cells. The assay was performed as described in Biedermann et al.(1991). B, DTA kinetic of nuclear extracts, 0.5 µg of protein per reaction, prepared from CB.17 and CB.17scid fibroblast cells.

A Protein Which Restores DNA Recombination in SCID Lymphocyte Extracts

Table 1describes the purification of SRSP (see ``Experimental Procedures''). The final protein fraction (Fraction V) is shown in Fig. 6as a silver-stained major protein band of approximately 72 kDa after reducing SDS-polyacrylamide gel electrophoresis. The specific activity is defined as x-fold stimulation per µg of protein added to the inactive SCID thymus nuclear extract. If some low DTA activity was present in a preparation of SRSP itself (Fractions III and IV), this activity was deducted before calculating the specific stimulatory activity.

Figure 6: SDS-polyacrylamide gel electrophoresis of SRSP. 100 ng of SRSP, Fraction V, were loaded on a 7.5% polyacrylamide-SDS gel, electrophoresed under standard conditions, and silver-stained.

About 10 µg of SRSP, Fraction V, could be obtained from 25 mg of nuclear protein extract (Fraction I). The presence of high recombinational activity in Fractions I and II prevents the calculation of specific stimulatory activity, total units, and recovery at these steps. The increase in specific activity therefore is certainly higher than the calculated 18-fold for steps III to V, the recovery lower than 12%.

The effect of SRSP on various nuclear extracts is described in Table 3. The protein, Fraction V, by itself is not active (No. 15). SCID thymus nuclear extract (0.4 µg), low by itself (No. 1), can be stimulated 12-fold with 100 ng of SRSP (No. 2) reaching about 50% of the wild-type thymus nuclear extract (0.4 µg) level (No. 12). Similarly, 0.17 µg of SCID proB cell extract can be stimulated 8-fold by the addition of 50 ng of SRSP (Nos. 3 and 4). The DN thymocyte extract, the most active thymocyte subpopulation, cannot be stimulated and is even weakly inhibited by the addition of SRSP (Nos. 5 and 6) The SP thymocyte extract, very low on its own, could be weakly stimulated 2.5-fold (Nos. 9 and 10) The extract prepared from DP thymocytes remained at its medium level of activity with no or a weak stimulation by SRSP (Nos. 7 and 8). Normal thymus extract, consisting to >80% of DP thymocytes, is not stimulated by SRSP (Nos. 11 and 12), as is the previously purified recombination complex RC-1 (Jessberger et al., 1993), which is even inhibited by various amounts of SRSP (Nos. 13, 14, and data not shown).

The kinetics of stimulation of SCID thymus extracts by SRSP (Fig. 7) reveals linearity over at least 1 h. At the 30-min time point, the activity reaches approximately 2.5% counts/min, which equals approximately 40% of the wild-type thymus extract activity (see Fig. 1) obtained with 0.25 µg of protein. The stimulatory effect of SRSP was not altered if heterologous or linearized recipient DNA has been used (data not shown).

Figure 7: Kinetics of the stimulatory effect of SRSP on SCID thymus extracts. 0.3 µg of SCID thymus extract was incubated with 75 ng of SRSP under standard DTA conditions, and the reaction was stopped after various times of incubation.

In control experiments, the ability of a series of different purified DNA metabolic enzymes and other proteins to stimulate the recombination-deficient SCID thymus extracts was tested (data not shown). Among the enzymes tested were nucleases (exonuclease I (Escherichia coli), exonuclease III (E. coli), exonuclease V (Micrococcus luteus), gene 6 exonuclease (T7), exonuclease, phophodiesterase (Crotalus adamanteus), S1 nuclease, P1 nuclease, the cruciform DNA cutting endonuclease VII (T4), and RNases A and H), DNA polymerases (polymerase I and Klenow polymerase (E. coli), , , , (calf thymus), DNA helicases (B and D, calf thymus), topoisomerases (I and II, calf thymus), and other proteins (terminal deoxynucleotidyltransferase, T4 DNA ligase, recA and SSB (E. coli), T4 gene 32 protein, proliferating cell nuclear antigen (calf thymus), and RF-C (calf thymus)). These proteins were added to the reaction mixture individually in at least five different concentrations. In addition, several combinations of the proteins were tested. In addition, two protein kinases, casein kinase II (tyrosine kinase) and protein kinase A (Ser/Thr kinase) were added to the SCID extract.

However, none of these proteins exhibited a significant stimulatory effect on the reaction, and none of them was active in the DTA on its own (data not shown). This indicates specificity of the SRSP stimulatory effect and it might suggest, that SRSP is functionally different from those enzymes and proteins. It also demonstrates the need for a special set of enzymatic activities, e.g. RC-1 (Jessberger et al., 1993), to perform the recombination reaction as measured with the DTA.

In Western blotting experiments, purified SRSP is not recognized by polyclonal anti RAG-1 or anti RAG-2 antibodies (data not shown).

Limited amino acid sequence data obtained from the N terminus of SRSP showed that SRSP is distinct from the 72-kDa subunit of protein Ku (Reeves and Sthoeger, 1989; data not shown). Monoclonal and anti-Ku p80 or polyclonal anti-Ku p70 antibodies do not react with SRSP in Western blot experiments (data not shown).

Preincubation of the DNA substrates in the reaction mixture with SRSP but with the nuclear extract being added at some time points after preincubation failed to change the outcome. Similarly, preincubation with nuclear extract but without SRSP and subsequent addition of SRSP did not affect the results. Taken together, changes in the order of addition of individual protein components of the reaction, with all components present in the final mixture, did not exert a significant influence on the stimulatory activity.

DISCUSSION

We have used a cell-free system, the DNA transfer assay (DTA; Jessberger and Berg(1991), Jessberger et al.(1993)), to study the biochemistry of recombination on plasmid or phage DNA substrates mediated by wild-type and mutant mammalian nuclear extracts. In these experiments, SCID lymphocytes and fibroblasts were found to be inefficient to perform the DTA-measured recombination reaction. In contrast, extracts from wild-type lymphocytes and fibroblasts as well as from RAG-2 lymphocytes showed significantly higher recombination activity in the assay.

The assay measures a strand transfer reaction between two double-stranded DNA substrates. The products are stable in that they do not depend on the continuous presence of protein and divalent cations, are insensitive to SDS, EDTA, and phenol-chloroform treatments, and are heat-resistant. Because of this stability and since complete reaction products were observed among the products bound to the beads (Jessberger and Berg, 1991; Jessberger et al., 1993), the DTA reaction indicates the presence of advanced strand transfer intermediates and complete recombination products. The routinely used substrates are homologous plasmid DNAs, one of which bearing a deletion to be repaired. They can be recombined by homologous or nonhomologous (end-joining) processes if the required activities were present in a given protein fraction. Crossover and non-crossover products of homologous recombination have been demonstrated earlier by polymerase chain reaction analysis (Jessberger and Berg, 1991; Jessberger et al., 1993). In addition to homologous, heterologous DNA substrates were also used in this study (Table 2). Preferred recombination of substrates bearing specific sequences, e.g. class switch regions, has also been observed in a related assay with certain, partially purified extract fractions. Enzymatic activities, e.g. DNA polymerase and ligase, considered essential to repair deletions and gaps, were identified as components of the purified recombination complex 1 (RC-1, Jessberger et al.(1993)). RC-1 preferentially catalyzes recombination between homologous DNA substrates.

Cell-free assays, which, like the DTA, measure any type of stable intermolecular DNA recombination reaction, should reflect a recombination deficiency in cell extracts, if the deficient pathway constitutes a significant portion of the total recombination activity present in a particular extract.

The recombination defect associated with the SCID mutation is of bivalent nature: V(D)J rearrangement and DNA DSB repair are affected (Lieber et al., 1988a, 1988b; Fulop and Phillips, 1990; Biedermann et al., 1991; Hendrickson et al., 1991; Roth et al., 1992a, 1992b; Pergola et al., 1993; Taccioli et al., 1993). In V(D)J joining, one of the later steps in the site-specific recombination process might be aberrant (Malynn et al., 1988), possibly the resolution of hairpin-like intermediates (Roth et al., 1992a, 1992b). For DSB repair, either end-joining or repair through homologous recombination with the intact allele could be nonfunctional. A defect in all pathways of homologous recombination seems less likely since sequences of 70-bp homology linked to the coding ends supported the otherwise disrupted formation of coding joints on extrachromosomal substrates in SCID cells (Lieber et al., 1988b). It is, however, conceivable, that a different, perhaps minor pathway for homologous recombination has been utilized in that case. A variety of such pathways has been demonstrated for eukaryotes (Szostak et al., 1983; Lin et al., 1990a, 1990b; Haber, 1992).

End-joining, as the lower fidelity alternative to homologous recombinational repair, is a candidate DSB repair pathway to be affected by the SCID mutation (Harrington et al., 1992; Staunton and Weaver, 1994). Coding joint formation in normal V(D)J recombination does not require homologies at the coding ends and may thus have similarities to nonhomologous, end-joining recombination reactions. There is, however, no difference between various normal and SCID cells in random integration of linear DNA into the genome (Staunton and Weaver, 1994), in intramolecular end-joining, and in intermolecular end-joining of transfected plasmid DNA substrates (Harrington et al., 1992).

Our comparison of several homologous and nonhomologous recipient DNAs (Table 2) revealed that the SCID extracts are deficient in performing the DTA reaction with both types of substrates. This suggests a general deficiency in DNA recombination in the SCID extracts. The data could be indicative of a defect in a key enzyme or protein, which should be involved in more than one pathway of DNA recombination. In the DTA system, normal and SCID extracts recombine homologous DNAs 2- to 3-fold better than heterologous substrates.

A protein, SRSP, has been purified from normal mouse thymus, which specifically stimulates DNA recombination in otherwise recombination-inefficient nuclear extracts prepared from SCID lymphocytes. The specificity of SRSP for stimulating exclusively SCID-derived extracts and neither wild-type nuclear extracts nor RAG-2 extracts nor the recombination complex RC-1 (Table 3), makes some relationship to the SCID mutation likely. The SCID mutation very likely is a mutation in the DNA-dependent protein kinase gene. DNA-dependent protein kinase is a serine/threonine kinase, which is dependent on double-strand DNA ends. DNA-dependent protein kinase in vitro phosphorylates a number of proteins including p53, Sp1, Tau, hsp90, SV40 large T antigen, human Oct-1 and Oct-2, c-Myc, and Ku (Lees-Miller and Anderson, 1989; Lees-Miller et al., 1990; Jackson et al., 1990; Gottlieb and Jackson, 1993; Wu et al., 1993a). DNA-dependent protein kinase is associated with both subunits of the Ku protein (Gottlieb and Jackson, 1993) and may be involved in DNA metabolism (for review, see Gottlieb and Jackson(1994)). SRSP might be part of a signal cascade downstream of DNA-dependent protein kinase. It might itself be a protein kinase which is inactive in SCID cells. Alternatively, a phosphorylated, wild-type SRSP could be directly involved in the enzymatic machinery mediating the recombination reaction.

Based on the purification data, SRSP appears to be an abundant protein. Since the recombination deficiency can be measured in the cell-free system using a number of different DNA substrates, SRSP might function in several pathways of recombination.

None of a large series of DNA metabolic enzymes including different endo- and exonucleases, DNA polymerases, DNA helicases, DNA ligase, two protein kinases (protein kinase A and casein kinase II), was able to restore recombinational activity in the SCID thymus extracts by analysis in the DTA. Thus, SRSP should functionally differ from these control enzymes. The interaction of SRSP with the recombination machinery and other proteins, possibly with DNA-dependent protein kinase, and/or with the DNA substrates, therefore, might be very specific. SRSP fails to stimulate the recombination complex RC-1 (Table 3), which does not contain SRSP. One possible explanation would be that SRSP is an activator of a recombination machinery, but cannot act any more on an already active complex like RC-1. The amount of SRSP needed to generate a large stimulatory effect (100 ng; Table 2) is relatively high. However, the stoichiometric requirements among the three reaction partners, DNA substrates, recombination enzymes, and SRSP, are not known. Furthermore, the purification procedure needed to isolate SRSP may yield an only partially active protein fraction, demanding the use of relatively high amounts of SRSP per reaction. The preparation of SRSP does contain at least two other minor proteins, visible as very weak bands in a silver-stained gel at a position below SRSP (Fig. 6). These two proteins, however, are not very likely to be involved in the stimulatory effect, since they decrease during purification. The purification data (Table 1) for SRSP from wild-type thymus reveals a high abundance of this protein in normal cells.

SCID cell extracts were not fully complemented by addition of SRSP, but rather to 40% to 60% of wild-type activity levels. Several explanations are conceivable, e.g. that another factor is still missing or inactive, that the optimal complementation conditions have not yet been determined, or that too much of SRSP became inactivated during its isolation. It is also possible that SRSP functionally substitutes for another protein, which has not been found in the extracts.

The role of a protein like SRSP might not be limited to intermolecular DNA transfer, since the SCID mutation markedly affects also V(D)J recombination. Our data, therefore, support the view, originally based on cell irradiation survival assays (Fulop and Philips, 1990; Biedermann et al., 1991; Hendrickson et al., 1991; Pergola et al., 1993; Taccioli et al., 1993), that repair of DSBs and V(D)J joining might share one or more factors. As outlined above, recent data, obtained in a comparison of various mutant hamster ovary cell lines which are defective in DSB repair, also identified a parallel defect in V(D)J recombination in two hamster cell complementation groups (xrs and XR-1; Pergola et al.(1993) and Taccioli et al.(1993)).

Evidence for a link between the two recombination reactions comes also from our experiments on the regulation of cell-free DTA activity throughout lymphocyte development. The analysis of sorted normal thymocytes and B cell progenitors representing various stages of development revealed peaks of recombination activities at the earliest stages: DN thymocytes and proB cell nuclear extracts are most active ( Fig. 2and Fig. 3). The activity then decreases with further maturation. This distribution of DTA activities on homologous DNA substrates parallels V(D)J joining activities throughout development. DN thymocytes and proB cells are most active in V(D)J recombination, and the following stages, DP and preB, still perform V(D)J rearrangements (Kronenberg et al., 1986; Lieber et al., 1987; Rolink and Melchers, 1993). Furthermore, V(D)J rearrangement induced by growth factor withdrawal in a proB cell line (Rolink et al., 1993) leads to a parallel induction of DTA activity.

These experiments provide evidence for a temporal correlation between V(D)J rearrangements and DTA-measured recombination in various lymphocytes from normal mice.

SCID lymphocyte progenitors are arrested in development at the CD4/CD8 DN or proB cell stages, corresponding in phenotypes to wild-type mouse early T and B cells. RAG-2 lymphocytes, although arrested at a comparable early stage and phenotypically similar, are different from the SCID cells in that they do not even initiate V(D)J joining. Although wild-type DN RAG-2 and SCID thymocytes represent phenotypically the same early stages in development, only SCID, but not wild-type DN or RAG-2-derived extracts were recombination-deficient by DTA. The question than arises of whether the observed DTA deficiency of the SCID lymphocyte extracts is caused by the SCID mutation itself, or rather by developmental changes and consecutive secondary events, like a proliferation block or a commencement of apoptosis in SCID lymphocytes.

The lower DTA activity in extracts prepared from SCID fibroblasts, however, argues for a direct relationship between the SCID mutation and the deficiency of SCID extracts in the DTA.

Is DTA activity coupled to cell proliferation? SCID lymphocytes may not proliferate at a rate comparable to wild-type early lymphocytes. However, proB cells, in which DTA and V(D)J rearrangements were induced, do not, or only very slowly, proliferate. The cells also cannot be poisoned with hydroxyurea (Rolink et al., 1993). Similarly, the SCID-derived fibroblasts, although faster proliferating than the wild-type fibroblasts, were lower in DTA activity (Fig. 5). Thus, the cell-free recombination reaction, like V(D)J recombination in vivo (Rolink and Melchers, 1993), apparently does not depend on cell proliferation. In vivo studies on transfected V(D)J plasmid DNA substrates also provided evidence for the independence of V(D)J recombination and replication (Hsieh et al., 1991).

Some SCID lymphocytes might enter apoptosis, a process which could cause low DTA activities. The wild-type CD4/CD8 DP thymocytes, however, are apoptotic, but their extracts, unlike SCID extracts, cannot be comparably stimulated in the DTA by SRSP (Table 3). In addition, enzyme assays for DNA polymerase, topoisomerases, and nucleases did not reveal any significant differences between normal, RAG-2, and SCID thymus nuclear extracts ( Fig. 1and data not shown). Such differences may be expected, however, in apoptotic thymocytes.

Unlike SCID lymphocytes, the RAG-2-deficient lymphocytes do not initiate V(D)J recombination, but have been found to be almost normal in the cell-free recombination activity. This underscores the specificity of the observed recombination deficiency for SCID extracts. Thus, it seems, that for the DTA-measured recombination activity, an active V(D)J recombination machinery is not necessary.

FOOTNOTES

*

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore by hereby marked ``advertisement'' in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

§

To whom correspondence and reprint requests should be addressed. Tel.: 41-61-605-1289; Fax: 41-61-605-1364.

()

C. Kirchgessner and J. M. Brown, personal communication.

()

S. P. Jackson and P. Jeggo, personal communication.

()

The abbreviations used are: DSB, double-strand break; TLCK, N-p-tosyl-L-lysine chloromethyl ketone; DTA, DNA transfer assay; TCR, T cell receptor; RC, recombination complex; FITC, fluorescein isothiocyanate; bp, base pair(s); FPLC, fast protein liquid chromatography; EPPS, 4-(2-hydroxyethyl)-1-piperazinepropanesulfonic acid.

()

M. Wabl and R. Jessberger, submitted for publication.

ACKNOWLEDGEMENTS

REFERENCES

1. Aguilera, R. J., Akira, S., Okazaki, K., and Sakano, H. (1987) Cell 51, 909-917 [CrossRef][Medline] [Order article via Infotrieve]
2. Andrews, R., Halligan, N. L., and Halligan, B. D. (1993) Biochem. Biophys. Res. Commun. 193, 139-145 [CrossRef][Medline] [Order article via Infotrieve]
3. Biedermann, K. A., Sun, J., Giaccia, A. J., Tosto, L. M., and Brown, J. M. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 1394-1397 [Abstract/Free Full Text]
4. Bosma, M. J., and Carroll, A. M. (1991) Annu. Rev. Immunol. 9, 323-350 [CrossRef][Medline] [Order article via Infotrieve]
5. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254 [CrossRef][Medline] [Order article via Infotrieve]
6. Carter, T., Vancurova, I., Sun, I., Willard, L., and Delon, S. (1990) Mol. Cell. Biol. 10, 6460-6471 [Abstract/Free Full Text]
7. Ceredig, R., Lowenthal, J. W., Nabholz, M., and MacDonald, H. R. (1985) Nature 314, 98-100 [CrossRef][Medline] [Order article via Infotrieve]
8. Fulop, G. M., and Phillips, R. A. (1990) Nature 347, 479-482 [CrossRef][Medline] [Order article via Infotrieve]
9. Gellert, M. (1992) Annu. Rev. Genet. 22, 425-446
10. Getts, R. C., and Stamato, T. D. (1994) J. Biol. Chem. 269, 15981-15984 [Abstract/Free Full Text]
11. Gottlieb, T. M., and Jackson, S. P. (1993) Cell 72, 131-142 [CrossRef][Medline] [Order article via Infotrieve]
12. Gottlieb, T. M., and Jackson, S. P. (1994) Trends Biochem. Sci. 19, 500-503 [CrossRef][Medline] [Order article via Infotrieve]
13. Haber, J. E. (1992) Curr. Opin. Cell Biol. 4, 401-412 [CrossRef][Medline] [Order article via Infotrieve]
14. Harrington, J., Hsieh, C.-L., Gerton, J., Bosma, G., and Lieber, M. R. (1992) Mol. Cell. Biol. 12, 4758-4768 [Abstract/Free Full Text]
15. Hendrickson, E. A., Qin, X.-Q., Bump, E. A., Schatz, D. G., Oettinger, M., and Weaver, D. T. (1991) Proc. Natl. Acad. Sci. U. S. A. 88, 4061-4065 [Abstract/Free Full Text]
16. Hsieh, C.-L., McCloskey, P., Radany, E., and Lieber, M. R. (1991) Mol. Cell. Biol. 11, 3972-3977 [Abstract/Free Full Text]
17. Jackson, S. P., MacDonald, J. J., Lees-Miller, S., and Tjian, R. (1990) Cell 63, 155-165 [CrossRef][Medline] [Order article via Infotrieve]
18. Jessberger, R. (1994) The Immunologist 2, 201-207
19. Jessberger, R., and Berg, P. (1991) Mol. Cell. Biol. 11, 445-457 [Abstract/Free Full Text]
20. Jessberger, R., Podust, V., Hübscher, U., and Berg, P. (1993) J. Biol. Chem. 268, 15070-15079 [Abstract/Free Full Text]
21. Komori, T., Okada, A., Stewart, V., and Alt, F. W. (1993) Science 261, 1171-1173 [Abstract/Free Full Text]
22. Kronenberg, M., Siu, G., Hood, L. E., and Shastri, N. (1986) Annu. Rev. Immunol. 4, 529-591 [CrossRef][Medline] [Order article via Infotrieve]
23. Landau, N., Schatz, D. G., Rosa, M., and Baltimore, D. (1987) Mol. Cell. Biol. 7, 3237-3243 [Abstract/Free Full Text]
24. Lees-Miller, S. P., and Anderson, C. W. (1989) J. Biol. Chem. 264, 17275-17280 [Abstract/Free Full Text]
25. Lees-Miller, S. P., Chen, Y. R., and Anderson, C. W. (1990) Mol. Cell. Biol. 10, 6472-6481 [Abstract/Free Full Text]
26. Li, M., Morzycka-Wroblewska, E., and Desiderio, S. V. (1989) Genes & Dev. 3, 1801-1813
27. Lieber, M. R. (1991) FASEB J. 5/14, 2934-2944
28. Lieber, M. R. (1992) Cell 70, 873-876 [CrossRef][Medline] [Order article via Infotrieve]
29. Lieber, M. R., Hesse, J., Mizuuchi, K., and Gellert, M. (1987) Genes & Dev. 1, 751-761
30. Lieber, M. R., Hesse, J., Mizuuchi, K., and Gellert. M. (1988a) Proc. Natl. Acad. Sci. U. S. A. 85, 8588-8592 [Abstract/Free Full Text]
31. Lieber, M. R., Hesse, J. E., Lewis, S., Bosma, G., Rosenberg, N., Mizuuchi, K., Bosma, M. J., and Gellert, M. (1988b) Cell 55, 7-16 [CrossRef][Medline] [Order article via Infotrieve]
32. Lin, F.-L. M., Sperle, K., and Sternberg, N. (1990a) Mol. Cell. Biol. 10, 103-112 [Abstract/Free Full Text]
33. Lin, F.-L. M., Sperle, K., and Sternberg, N. (1990b) Mol. Cell. Biol. 10, 113-119 [Abstract/Free Full Text]
34. Malynn, B. A., Blackwell, K. T., Fulop, G. M., Rathburn, G. A., Furley, A. J., Ferrier, P., Heinke, B. L., Phillips, R. A., Yancopoulos, G. D., and Alt, F. W. (1988) Cell 54, 453-460 [CrossRef][Medline] [Order article via Infotrieve]
35. Matsunami, N., Hamaguchi, Y., Yamamoto, Y., Kuze, K., Kangawa, K., Matsuo, H., Kawaichi, M., and Honjo, T. (1989) Nature 342, 934-937 [CrossRef][Medline] [Order article via Infotrieve]
36. McCormack, W. T., Tjoelker, L. W., Carlson, L. M., Petryniak, B., Barth, C. F., Humphries, E. H., and Thompson, C. B. (1989) Cell 56, 785-791 [CrossRef][Medline] [Order article via Infotrieve]
37. Meselson, M. S., and Radding, C. M. (1975) Proc. Natl. Acad. Sci. U. S. A. 72, 358-361 [Abstract/Free Full Text]
38. Mombaerts, P., Iacomini, J., Johnson, R. S., Herrup, K., Tonegawa, S., and Papaioannou, V. E. (1992) Cell 68, 869-877 [CrossRef][Medline] [Order article via Infotrieve]
39. Oettinger, M. A., Schatz, D. G., Gorka, C., and Baltimore, D. (1990) Science 248, 1517-1523 [Abstract/Free Full Text]
40. Ogawa, M., Matsuzaki, Y., Nishikawa, S., Hayashi, S.-I., Kunisada, T., Sudo, T., Kina, T., Nakauchi, H., and Nishikawa, S.-I. (1991) J. Exp. Med. 174, 63-71 [Abstract/Free Full Text]
41. Paillard, S., and Strauss, F. (1991) Nucleic Acids Res. 19, 5619-5624 [Abstract/Free Full Text]
42. Pergola, F., Zdzienicka, Z., and Lieber, M. R. (1993) Mol. Cell. Biol. 13, 3464-3471 [Abstract/Free Full Text]
43. Rathmell, W. K., and Chu, G. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 7623-7627 [Abstract/Free Full Text]
44. Reeves, W. H., and Sthoeger, Z. M. (1989) J. Biol. Chem. 264, 5047-5052 [Abstract/Free Full Text]
45. Rolink, A., and Melchers, F. (1993) Curr. Opin. Immunol. 5, 207-217 [CrossRef][Medline] [Order article via Infotrieve]
46. Rolink, A., Grawunder, U., Haasner, D., Strasser, A., and Melchers, F. (1993) J. Exp. Med. 178, 1263-1270 [Abstract/Free Full Text]
47. Roth, D. B., Menetski, J. P., Nakajima, P. B., Bosma, M. J., and Gellert, M. (1992a) Cell 70, 983-991 [CrossRef][Medline] [Order article via Infotrieve]
48. Roth, D. B., Nakajima, P. B., Menetski, J. P., Bosma, M. J., and Gellert, M. (1992b) Cell 69, 41-53 [CrossRef][Medline] [Order article via Infotrieve]
49. Sarmiento, M., Glasebrook, A. L., and Fitch, F. W. (1980) J. Immunol. 125, 2665-2672 [Abstract]
50. Schatz, D. G., Oettinger, M. A., and Baltimore, D. (1989) Cell 59, 1035-1048 [CrossRef][Medline] [Order article via Infotrieve]
51. Schatz, D. G., Oettinger, M. A., and Schlissel, M. S. (1992) Annu. Rev. Immunol. 10, 359-383 [Medline] [Order article via Infotrieve]
52. Shinkai, Y., Rathbun, G., Lam, K.-P., Oltz, E. M., Stewart, V., Mendelsohn, M., Charron, J., Datta, M., Young, F., Stall, A. M., and Alt, F. (1992) Cell 68, 855-867 [CrossRef][Medline] [Order article via Infotrieve]
53. Smider, V., Rathmell, W. K., Lieber, M., and Chu, G. (1994) Science 266, 288-291 [Abstract/Free Full Text]
54. Staunton, J. E., and Weaver, D. T. (1994) Mol. Cell. Biol. 14, 3876-3883 [Abstract/Free Full Text]
55. Szostak, J. W., Orr-Weaver, T. L., Rothstein, R. J., and Stahl, F. W. (1983) Cell 33, 25-35 [CrossRef][Medline] [Order article via Infotrieve]
56. Taccioli, G. E., Rathburn, G., Shinkai, Y., Oltz, E., Cheng, H., Whitmore, G., Stamato, T., Jeggo, P., and Alt, F. (1992) Curr. Top. Microbiol. Immunol. 182, 107-114 [Medline] [Order article via Infotrieve]
57. Taccioli, G. E., Rathburn, G., Oltz, E., Stamato, T., Jeggo, P. A., and Alt, F. (1993) Science 260, 207-210 [Abstract/Free Full Text]
58. Taccioli, G. E., Gottlieb, T. M., Blunt, T., Priestley, A., Demengot, J., Mizuta, R., Lehman, A. R., Alt, F. W., Jackson, S. P., and Jeggo, P. (1994) Science 265, 1442-1445 [Abstract/Free Full Text]
59. Warburg, O., and Christian, W. (1941) Biochem. Z. 310, 384-421
60. West, S. C. (1992) Annu. Rev. Biochem. 61, 603-640 [CrossRef][Medline] [Order article via Infotrieve]
61. Wu, J. M., Chen, Y., An, S., Perruccio, L., Abdel-Ghany, M., and Carter, T. H. (1993) Biochem. Biophys. Res. Commun. 193, 13-18 [CrossRef][Medline] [Order article via Infotrieve]
62. Wu, L.-C., Mak, C.-H., Dear, N., Boehm, T., Foroni, L., and Rabbits, T. H. (1993) Nucleic Acids Res. 21, 5067-5073 [Abstract/Free Full Text]

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

This Article Abstract Full Text (PDF) Submit a Letter to Editor Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Jessberger, R. Articles by Rodewald, H.-R. Search for Related Content PubMed PubMed Citation Articles by Jessberger, R. Articles by Rodewald, H.-R. Social Bookmarking                      What's this?