Synapsis of recombination signal sequences located in cis and DNA underwinding in V(D)J recombination

Abstract

V(D)J recombination requires binding and synapsis of a complementary (12/23) pair of recombination signal sequences (RSSs) by the RAG1 and RAG2 proteins, aided by a high-mobility group protein, HMG1 or HMG2. Double-strand DNA cleavage within this synaptic, or paired, complex is thought to involve DNA distortion or melting near the site of cleavage. Although V(D)J recombination normally occurs between RSSs located on the same DNA molecule (in cis), all previous studies that directly assessed RSS synapsis were performed with the two DNA substrates in trans. To overcome this limitation, we have developed a facilitated circularization assay using DNA substrates of reduced length to assess synapsis of RSSs in cis. We show that a 12/23 pair of RSSs is the preferred substrate for synapsis of cis RSSs and that the efficiency of pairing is dependent upon RAG1-RAG2 stoichiometry. Synapsis in cis occurs rapidly and is kinetically favored over synapsis of RSSs located in trans. This experimental system also allowed the generation of underwound DNA substrates containing pairs of RSSs in cis. Importantly, we found that the RAG proteins cleave such substrates substantially more efficiently than relaxed substrates and that underwinding may enhance RSS synapsis as well as RAG1/2-mediated catalysis. The energy stored in such underwound substrates may be used in the generation of DNA distortion and/or protein conformational changes needed for synapsis and cleavage. We propose that this unwinding is uniquely sensed during synapsis of an appropriate 12/23 pair of RSSs.

FIG. 1.

Schematic representation of the facilitated ligation assay and possible outcomes resulting from RAG-mediated paired complex formation on RSSs located in cis or in trans. At the top is shown the linear IS95 substrate (total length, 263 bp), which has cohesive ends generated by digestion with XbaI. Synapsis in cis and trans facilitates formation of monomer circles and linear dimers, respectively, upon ligation with T4 DNA ligase. In this and subsequent figures, filled and open triangles represent the 12- and 23-RSSs, respectively.

FIG. 2.

HMG2 is essential for RAG-mediated coupled cleavage of IS95. Coupled cleavage reactions were performed in 2.5 mM Mg²⁺ with constant amounts of linear IS95 substrate, MBP-RAG1, and GST-RAG2, while HMG2 was added in increasing amounts in lanes 5 to 15. Lanes 1 to 4, control reactions lacking various components as indicated above the lanes; lanes 5 to 15, HMG2 at 2.5, 3.75, 5, 10, 15, 22.5, 30, 60, 90, 120, and 200 nM, respectively. (Inset) Quantitated data displayed as percent efficiency of cleavage versus HMG2 concentration (see Materials and Methods). Open symbols, cleavage at one RSS; closed symbols, cleavage at both RSSs.

FIG. 3.

HMG2 induces formation of distinct topoisomeric circles in a circularization assay. The ring closure assay was performed with the IS95 substrate in the presence of increasing concentrations of HMG2. All ligation reactions were performed in the presence of 200 U of T4 DNA ligase (NEB) at 25°C for 2 min, with a DNA concentration of 3 nM and 2.5 mM Mg²⁺. Lane 1, no HMG2; lane 10, no T4 DNA ligase; lanes 2 to 9, HMG2 at 0.175, 0.35, 1.75, 3.5, 10.5, 21, 70, and 175 nM, respectively. LM, linear monomeric 263-bp IS95 substrate; LD, linear dimers of IS95; MC, monomeric circles. Sizes of molecular mass markers are indicated in base pairs on the right.

FIG. 4.

Characterization of ligation products. (A) Ligation products generated in the presence of ethidium bromide have mobilities similar to those induced by HMG2. A ring closure assay was performed using 3 nM IS95 substrate in the presence of increasing concentrations of ethidium bromide. Lane 1, no T4 DNA ligase; lane 2, T4 DNA ligase alone; lanes 3 and 4, ligation in the presence of 21 and 35 nM HMG2; lanes 5 to 19, ethidium bromide at 6.67, 13.34, 20, 26.7, 40, 66.7, 133.4, 267, 400, 533, and 677 ng/ml and 1.34, 2.67, 3.34, and 4 μg/ml, respectively. LM, linear monomeric 263-bp IS95 substrate; LD, linear dimers of IS95; MC, monomeric circles. Sizes of molecular mass markers are indicated in base pairs on the right. (B) Analysis of products generated by ligation of IS95 in the presence of HMG2. Substrates (as indicated at top) were treated withvarious enzymes as indicated above the lanes. Lane 1, linear IS95; lane 2, ligation of IS95 (3 nM) in the presence of 22.5 nM HMG2; lanes 3 and 4, 3 nM IS95 linear substrate digested with ClaI (0.5 U for 30 min [lane 3] or 1 U for 2 h [lane 4]); lanes 5 and 6, 3 nM IS95 linear substrate digested with Exo III (0.5 U for 15 min [lane 5] or 1 U for 2 h [lane 6]); lane 7, purified Lk25 topoisomer (band a); lanes 8 and 9, 3 nM Lk25 digested with ClaI as in lanes 3 and 4; lane 10, Lk25 digested with Exo III (1 U for 2 h); lane 11, purified Lk24 topoisomer (band b); lanes 12 and 13, 3 nM Lk24 digested with ClaI as in lanes 3 and 4; lane 14, Lk24 digested with Exo III (1 U for 2 h); lane 15, purified LD of IS95; lane 16, 3 nM LD digested with ClaI (1 U for 2 h); lanes 17 and 18, 3 nM LD digested with Exo III as in lanes 5 and 6.

FIG. 5.

The fast-migrating topoisomer generated by ligation of IS95 in the presence of HMG2 is unwound by one helical turn with respect to the relaxed 263-bp circle with an Lk of 25. Lane 1, linear monomer 263-bp IS95; lanes 2 to 5, purified band b (Lk24) untreated or treated with wheat germ topoisomerase I (TI wg), calf thymus topoisomerase I (TI thy), E. coli topoisomerase I (TI E. coli), or DNase I (D), as indicated above the lanes; lanes 7 to 11, purified band a (Lk25) untreated or treated as for Lk24, as indicated above the lanes. Sizes of molecular mass markers are indicated in base pairs on the left.

FIG.6.

Effect of RAG1 on the products of ligation. (A) RAG1 by itself does not influence ligation. Ring closure assays were performed in the presence of 3 nM IS95 equilibrated with increasing amounts of MBP-RAG1-D708A prior to ligation. Lanes 1 to 3, control reactions lacking various components, as indicated above the lanes; lanes 4 to 18, MBP-RAG1-D708A at 1.25, 2.5, 3.75, 5, 7.5, 10, 12.5, 15, 30, 45, 60, 90, 120, 150, and 300 nM, respectively (all lanes [4 to 18] lack HMG2). Sizes of molecular mass markers are indicated in base pairs on the left. (B) RAG1 inhibits ligation in the presence of HMG2. Ring closure assays were performed in the presence of 3 nM IS95 equilibrated with 22.5 nM HMG2 and increasing amounts of MBP-RAG1-D708A prior to ligation. Lanes 1 to 3, control reactions lacking various components, as indicated above the lanes; lanes 4 to 19, MBP-RAG1-D708A at 1.25, 2.5, 3.75, 5, 7.5, 10, 12.5, 15, 30, 45, 60, 90, 120, 150, 300, and 900 nM, respectively (all lanes [4 to 19] also contained 22.5 nM HMG2). Sizes of molecular mass markers are indicated in base pairs on the left. LM, linear monomeric 263-bp IS95 substrate; LD, linear dimers of IS95; MC, monomeric circles. The inset shows the quantitation of the data from panels A and B, represented graphically as the fraction circularization efficiency (see Materials and Methods) versus the concentration of MBP-RAG1-D708A.

FIG. 7.

Synapsis of RSSs located in cis, as indicated, by facilitated ligation. (A) Effect of increasing GST-RAG2 concentration in a ligation mixture containing constant amounts of linear IS95 substrate (3 nM), HMG2 (22.5 nM), and MBP-RAG1-D708A (6 nM). Lanes 1 to 4, control reactions lacking various components, as indicated above the lanes; lanes 5 to 19, GST-RAG2 at 1.38, 2.77, 5.55, 6.94, 8.5, 10, 12.50, 16.6, 20.8, 25, 33, 41.65, 50, 66.67, or 83.3 nM, respectively. LT, linear trimer of IS95; other abbreviations are as for Fig. 3. Sizes of molecular mass markers are indicated in base pairs on the left. (B) Effect of RAG2 concentration on substrate circularization. Quantitated data from gels such as that inpanel A are represented as the fraction efficiency of circularization (see Materials and Methods) versus GST-RAG2 concentration. The main graph shows results for substrates containing two RSSs as well as for TN263, which lacks RSSs. The inset shows results for substrates containing a single RSS compared to IS95 and TN263. Data points represent average values obtained from at least three determinations. Data were analyzed statistically, and a small number of data points were excluded from calculations of average values as described in Materials and Methods. Error bars denote the extreme maximal and minimal values obtained from individual determinations, not the standard deviation. Differences in the values obtained were statistically significant for the following pairs of substrates in the range of RAG2 concentrations from 7 to 22 nM: IS95 versus TN263, P < 0.006; invIS111 versus TN263, P < 0.018; IS95-12/12 versus TN263, P < 0.044; IS95-23/23 versus TN263, P < 0.005; IS95 versus IS95-12/12, P < 0.044; IS95 versus IS95-23/23, P < 0.024; invIS111 versus invIS111-12/12, P < 0.049.

FIG. 8.

Effect of RAG2 concentration on coupled cleavage of RSSs located in cis. (A) Effect of increasing GST-RAG2 concentration on cleavage of linear IS95 substrate (3 nM) in reaction mixtures containing HMG2 (22.5 nM) and MBP-RAG1-D708A (6 nM). Lanes 1 to 4, control reactions lacking various components, as indicated above the lanes; lanes 5 to 19 GST-RAG2 at 1.38, 2.77, 5.55, 6.94, 8.5, 12.50, 16.6, 25, 33, 41.7, 50, 58.3, 66.7, 75, and 83.3 nM, respectively. SC, product of single cleavage at either of the RSSs; DC, double cleavage product. The asterisk indicates a band created by nicking of the substrate at both RSSs. Sizes of molecular mass markers are indicated in base pairs on the left. (B) Effect of RAG2 concentration on coupled cleavage. Quantitated data from gels such as that of panel A are represented as the percent efficiency of double cleavage (see Materials and Methods) versus GST-RAG2 concentration. Data points represent average values obtained from at least three data sets, while error bars denote the extreme maximal and minimal values obtained from individual determinations.

FIG. 9.

Kinetics of RAG-mediated synapsis of cis RSSs. Complete ligation reactions were performed by adding a mixture containing 200 U of T4 DNA ligase, 12.5 nM GST-RAG2, and 6 nM MBP-RAG1-D708A to a solution containing 3 nM IS95 DNA in the presence of 2.5 mM MgCl₂ at 25°C, followed by quenching with 16 mM EDTA at the appropriate time intervals. For control reactions, GST-RAG2 was omitted from the reaction mixture. The main figure represents quantitation of monomer circle and linear multimer formation with time. In the case of circle formation, the efficiency of ligation expresses the ratio between the amount of circle formed by ligation versus the amount of unligated monomer, while in the case of multimers the efficiency expresses the ratio between the amount of linear dimer and trimer generated versus the amount of unligated monomer. (Inset) Ratio of underwound Lk24 to relaxed Lk25 product plotted versus time. Data points represent average values obtained from at least three data sets (except for circle formation with TN263, which represents a single data set), while error bars denote the extreme maximal and minimal values obtained from individual determinations.

FIG.10.

Underwinding enhances cleavage by the RAG proteins. (A) Reactions were performed for 2 h at 37°C in the presence of 2.5 mM MgCl₂, 3 nM DNA, 6 nM MBP-RAG1, 12.5 nM GST-RAG2, and increasing concentrations of HMG2. Lanes 1 to 10, Lk24 IS95 circular substrate; lanes 11 to 20, Lk25 IS95 relaxed circle. Reaction mixtures in lanes 1 to 10 and 11 to 20 contained the following concentrations of HMG2: 0, 2.5, 5, 10, 15, 22.5, 30, 60, 120, and 200 nM, respectively. (Inset) Quantitation of the amount of double cleavage, expressed as the percent efficiency of cleavage versus HMG2 concentration (see Materials and Methods). Dark boxes, Lk24 substrate; open diamonds, Lk25 substrate. Sizes of molecular size markers are indicated in base pairs on the right. (B) Kinetics of RAG-mediated coupled cleavage with linear, relaxed circular, and underwound circular substrates. Reactions were performed at 37°C in the presence of 2.5 mM MgCl₂, 3 nM DNA, 6 nM MBP-RAG1, 12.5 nM GST-RAG2, and 120 nM HMG2 for the indicated length of time. Lanes 1 to 6, linear IS95 substrate; lanes 7 to 13, underwound Lk24 IS95 circular substrate; lanes 14 to 20, relaxed Lk25 IS95 circular substrate. Molecular size markers are the same as those used for panel A. (Inset) Quantitation of the amount of double cleavage, expressed as the percent efficiency of cleavage versus time. Data points represent average values obtained from three data sets, while error bars denote the extreme maximal and minimal values obtained from individual determinations.