Positive Role of the Mammalian TBPIP/HOP2 Protein in DMC1-mediated Homologous Pairing*

In meiosis, homologous recombination preferentially occurs between homologous chromosomes rather than between sister chromatids, which is opposite to the bias of mitotic recombinational repair. The TBPIP/HOP2 protein is a factor that ensures the proper pairing of homologous chromosomes during meiosis. In the present study, we found that the purified mouse TBPIP/HOP2 protein stimulated homologous pairing catalyzed by the meiotic DMC1 recombinase in vitro. In contrast, TBPIP/HOP2 did not stimulate homologous pairing by RAD51, which is another homologous pairing protein acting in both meiotic and mitotic recombination. The positive effect of TBPIP/HOP2 in the DMC1-mediated homologous pairing was only observed when TBPIP/HOP2 first binds to double-stranded DNA, not to single-stranded DNA, before the initiation of the homologous pairing reaction. Deletion analyses revealed that the C-terminal basic region of TBPIP/HOP2 is required for efficient DNA binding and is also essential for its homologous pairing stimulation activity. Therefore, these results suggest that TBPIP/HOP2 directly binds to DNA and functions as an activator for DMC1 during the homologous pairing step in meiosis.

In meiosis, homologous recombination preferentially occurs between homologous chromosomes rather than between sister chromatids, which is opposite to the bias of mitotic recombinational repair. The TBPIP/HOP2 protein is a factor that ensures the proper pairing of homologous chromosomes during meiosis. In the present study, we found that the purified mouse TBPIP/ HOP2 protein stimulated homologous pairing catalyzed by the meiotic DMC1 recombinase in vitro. In contrast, TBPIP/HOP2 did not stimulate homologous pairing by RAD51, which is another homologous pairing protein acting in both meiotic and mitotic recombination. The positive effect of TBPIP/HOP2 in the DMC1-mediated homologous pairing was only observed when TBPIP/ HOP2 first binds to double-stranded DNA, not to singlestranded DNA, before the initiation of the homologous pairing reaction. Deletion analyses revealed that the C-terminal basic region of TBPIP/HOP2 is required for efficient DNA binding and is also essential for its homologous pairing stimulation activity. Therefore, these results suggest that TBPIP/HOP2 directly binds to DNA and functions as an activator for DMC1 during the homologous pairing step in meiosis.
During meiosis, two successive rounds of nuclear division, meiosis I and meiosis II, are promoted with a single round of DNA replication. As a result, diploid cells produce haploid gametes in eukaryotes. In the cell division at meiosis I, homologous chromosomes are aligned, and a high level of recombination occurs between homologous chromosomes but not between sister chromatids. This preferential recombination between homologous chromosomes, called homologous recom-bination, ensures their correct segregation at meiosis I through the formation of chiasmata, which physically connect homologous chromosomes.
Homologous recombination is initiated by double strand break (DSB) 1 formation by SPO11 at the initiation sites for recombination (1)(2)(3). Then a single-stranded DNA (ssDNA) tail derived from a DSB site invades the homologous doublestranded DNA (dsDNA) to form a heteroduplex. This strand invasion step, called homologous pairing, is the key step in homologous recombination. In Escherichia coli, the RecA protein catalyzes the homologous pairing step (4,5). Two eukaryotic homologues of RecA, the RAD51 and DMC1 proteins, which are conserved from yeast to human, have been identified (6 -9) and have been shown to catalyze homologous pairing in vitro (10 -15). The RAD51 gene is expressed in both mitotic and meiotic cells, but the DMC1 gene functions only in meiotic cells. Disruption of the Rad51 gene results in early embryonic lethality in mice (16,17). In chicken DT40 cells, the RAD51 gene disruption causes the accumulation of spontaneous chromosomal breaks and significantly reduces the recombination frequency between sister chromatids (18,19). In contrast, disruption of the Dmc1 gene does not cause lethality in mice, but the dmc1 knock-out mice are sterile with an arrest of gametogenesis in the first meiotic prophase (20,21). Therefore, DMC1 is a meiosis-specific homologous pairing protein that may be a central player in meiosis-specific events such as the homologous pairing between homologous chromosomes but not between sister chromatids.
In addition to the DMC1 gene, genetic studies with Saccharomyces cerevisiae have identified the HOP2 gene, which is essential for proper homologous chromosome pairing and recombination during meiosis (22). The HOP2 homologues of Schizosaccharomyces pombe and Arabidopsis thaliana, meu13 and AHP2, respectively, were also identified as meiosis-specific factors that facilitate the pairing of homologous chromosomes and recombination (23,24). The hop2 deletion mutant in S. cerevisiae fails to sporulate due to a uniform arrest at the pachytene stage of meiosis I with most of the chromosomes engaged in synapsis with nonhomologous partners (22). This meiotic cell cycle arrest by the hop2 mutation is bypassed by the dmc1 mutation, indicating that the meiotic arrest of hop2 is due to the action of the Dmc1 protein (25). This genetic interaction between HOP2 and DMC1 suggests that the Hop2 protein functions in the strand invasion step, which is promoted by the Dmc1 protein, during meiosis. The mammalian HOP2 homologue, TBPIP, was first identified as a factor interacting with TBP-1, which binds to the human immunodeficiency virus, type 1 Tat protein (26,27). The TBPIP/hop2 knock-out mice also display the meiotic cell cycle arrest due to the failure of DSB repair (28).
In the present study, we demonstrated that the mammalian TBPIP/HOP2 protein is an activator that specifically stimulates the homologous pairing catalyzed by DMC1. The homologous pairing stimulation by the TBPIP/HOP2 protein was only observed when TBPIP/HOP2 binds dsDNA, not ssDNA, before the initiation of homologous pairing by DMC1. Deletion analyses of TBPIP/HOP2 showed that the C-terminal basic region, required for efficient DNA binding, is essential for the homologous pairing stimulation activity of TBPIP/HOP2. This is the first biochemical evidence that TBPIP/HOP2 functions with the meiosis-specific homologous pairing protein DMC1.

EXPERIMENTAL PROCEDURES
Overexpression and Purification of the Mouse and Human TBPIP/ HOP2 Proteins-The mouse and human TBPIP/Hop2 genes (26,27) were ligated with the pET-15b expression vector (Novagen) at the NdeI-BamHI sites. The proteins were overexpressed in the E. coli strain BL21-CodonPlus(DE3)-RIL (Stratagene) as N-terminal His 6 -tagged proteins. The cells were grown in 10 liters of LB medium containing 100 g/ml ampicillin and 34 g/ml chloramphenicol at 30°C. At the logarithmic phase of growth (A 600 ϭ 0.6), the TBPIP/HOP2 expression was induced with 50 M isopropyl-1-thio-␤-D-galactopyranoside (final concentration). Cells were harvested after an overnight incubation at 18°C and were lysed by sonication in buffer A (50 mM Tris-HCl buffer (pH 8.0) containing 0.5 M NaCl, 10% glycerol, and protease inhibitors (Complete EDTA-free, Roche Applied Science)) on ice. The cell lysates were centrifuged at 27,700 ϫ g for 20 min, and the supernatants were gently mixed by the batch method with 4 ml of Ni-NTA-agarose beads (Qiagen) for 1 h at 4°C. The Ni-NTA-agarose beads with the His 6 -tagged mouse TBPIP/HOP2 protein were packed into an Econo column (Bio-Rad) and were washed with 30 column volumes of buffer A containing 15 mM imidazole at a flow rate of about 0.3 ml/min. The beads with the His 6 -tagged human TBPIP/HOP2 protein were washed with buffer A containing 50 mM imidazole. The His 6 -tagged mouse and human TB-PIP/HOP2 proteins were eluted in a 15-column volume linear gradient from 15 mM (or 50 mM for the human TBPIP/HOP2 protein) to 400 mM imidazole in buffer A. The His 6 tag was uncoupled from the mouse and human TBPIP/HOP2 proteins (mTBPIP/HOP2 and hTBPIP/HOP2, respectively) by digestions with 1 unit of thrombin protease (Amersham Biosciences)/mg of mTBPIP/HOP2 and with 3 units of thrombin protease/mg of hTBPIP/HOP2. The mTBPIP/HOP2 and hTBPIP/HOP2 proteins were immediately dialyzed against buffer B (20 mM Tris-HCl buffer (pH 8.0) containing 0.2 M KCl, 2 mM 2-mercaptoethanol, 0.25 mM EDTA, and 10% glycerol) and buffer C (20 mM Tris-HCl buffer (pH 8.0) containing 0.4 M KCl, 2 mM 2-mercaptoethanol, 0.25 mM EDTA, and 10% glycerol), respectively, for more than 12 h at 4°C. The proteins were loaded onto a 6-ml heparin-Sepharose (Amersham Biosciences) column previously equilibrated with either buffer B (for mTBPIP/ HOP2) or buffer C (for hTBPIP/HOP2). The column was washed with 20 column volumes of buffer B or C, and the proteins were eluted with a 20-column volume linear gradient from 0.2 to 1.2 M KCl in buffer B (mTBPIP/HOP2) or from 0.4 to 1.4 M KCl (hTBPIP/HOP2) in buffer C. The peak fractions of mTBPIP/HOP2 were dialyzed against buffer B, and those of hTBPIP/HOP2 were dialyzed against buffer D (20 mM Tris-HCl buffer (pH 8.0) containing 0.5 M KCl, 2 mM 2-mercaptoethanol, 0.25 mM EDTA, and 10% glycerol). Protein concentrations were determined using a Bio-Rad protein assay kit with bovine serum albumin as the standard.
Overexpression and Purification of the Human DMC1 and RAD51 Proteins-The human RAD51 protein was purified as described previously (29). The human DMC1 gene was inserted into the pET-15b plasmid (Novagen) at the NdeI-BamHI sites, and the protein was overexpressed in the E. coli strain BL21-CodonPlus(DE3)-RIL (Stratagene) as an N-terminally His 6 -tagged protein (30). The cells were grown in 10 liters of LB medium containing 100 g/ml ampicillin and 34 g/ml chloramphenicol at 30°C. When the A 600 of the culture was between 0.4 and 0.6, 1 mM isopropyl-1-thio-␤-D-galactopyranoside (final concentration) was added to induce protein expression. Cells were harvested after an overnight incubation and were lysed by sonication in buffer F (50 mM Tris-HCl buffer (pH 8.0) containing 0.5 M NaCl, 10 mM 2-mercaptoethanol, 10% glycerol, and protease inhibitors (Complete EDTA-free, Roche Applied Science)) on ice. The lysates were centrifuged at 27,700 ϫ g for 20 min, and the supernatants were gently mixed by the batch method with 4 ml of Ni-NTA-agarose beads (Qiagen) for 1 h at 4°C. The protein-bound beads were packed into an Econo column (Bio-Rad) and were washed with 30 column volumes of buffer F containing 5 mM imidazole. The DMC1 protein was eluted in a 20-column volume linear gradient from 5 to 300 mM imidazole in buffer F. The His 6 tag was uncoupled from DMC1 by a digestion with 3 units of thrombin protease (Amersham Biosciences)/mg of DMC1, and the protein was immediately dialyzed against buffer G (50 mM Tris-HCl buffer (pH 8.0) containing 0.2 M KCl, 0.5 mM EDTA, 10 mM 2-mercaptoethanol, and 10% glycerol) for more than 12 h at 4°C. The protein was loaded onto a 4-ml heparin-Sepharose (Amersham Biosciences) column previously equilibrated with buffer G. The column was washed with 20 column volumes of buffer G, and the protein was eluted with a 20-column volume linear gradient of 0.2 to 1.2 M KCl in buffer G. The peak fractions of DMC1 were dialyzed against buffer G. The protein concentration was determined using a Bio-Rad protein assay kit with bovine serum albumin as the standard.
DNA Substrates-In the D-loop formation assay, alkaline treatment of the cells harboring the plasmid DNA was avoided to prevent the dsDNA substrates from undergoing irreversible denaturation. Instead the cells were gently lysed using sarkosyl as described previously (29). The pGsat4 DNA was created by inserting a 198-base pair fragment of the human ␣-satellite sequence into the pGEM-T Easy vector (Promega).
For the ssDNA substrate used in the D-loop assay, the following high pressure liquid chromatography-purified oligonucleotide was purchased from Roche Applied Science: SAT-1 (50-mer, 5Ј-ATT TCA TGC TAG ACA GAA GAA TTC TCA GTA ACT TCT TTG TGC TGT GTG TA-3Ј). The 5Ј ends of the oligonucleotides were labeled with T4 polynucleotide kinase (New England Biolabs) in the presence of [␥- 32  In this assay, a 4.3-fold excess amount of ssDNA (molecule) was used relative to the amount of dsDNA (molecule). After incubations at 37°C for the indicated times, the reactions were terminated by the addition of 0.5% SDS and 700 g/ml proteinase K (Roche Applied Science) followed by an incubation at 37°C for 15 min. The 6-fold loading dye was added, and the products were resolved by 1% agarose gel electrophoresis in 0.5ϫ TBE buffer (90 mM Tris base, 64.6 mM boric acid, and 2 mM EDTA) at 3.3 V/cm for 2 h and were visualized by autoradiography of the dried gel.

RESULTS
Purification of the Mouse and Human TBPIP/HOP2 Proteins-To study the function of TBPIP/HOP2, we overexpressed mTBPIP/HOP2 in the E. coli BL21(DE3) Codon Plus (Stratagene) strain as a fusion protein with an N-terminal hexahistidine tag (His 6 tag) containing a cleavage site for thrombin protease. The His 6 -tagged mTBPIP/HOP2 protein was expressed by induction with isopropyl-1-thio-␤-D-galactopyranoside and was purified by chromatography on a Ni 2ϩ -chelating column (Qiagen). The His 6 tag was uncoupled with thrombin protease (Amersham Biosciences) from the mTBPIP/HOP2 portion, and mTBPIP/HOP2 was further purified by heparin-Sepharose column chromatography (Amersham Biosciences) (Fig. 1A).
The human TBPIP/HOP2 protein, which shares about 90% amino acid sequence identity with mTBPIP/HOP2, was also overexpressed and was purified by a similar method as that used for mTBPIP/HOP2. The purified human TBPIP/HOP2 preparation contained about 10% of a lead-through product as judged by SDS-polyacrylamide gel electrophoresis analysis. Furthermore degradation products (about 10%) were detected in the fraction containing the purified human TBPIP/HOP2 protein (data not shown), suggesting that the human TBPIP/ HOP2 protein is less stable than mTBPIP/HOP2. Therefore, we used mTBPIP/HOP2 in the subsequent biochemical analyses.
The DNA Binding and Homologous Pairing Activities of the TBPIP/HOP2 Protein-The purified mTBPIP/HOP2 protein was tested for its ssDNA and dsDNA binding abilities. As shown in Fig. 1, B and C, mTBPIP/HOP2 bound to both ssDNA and dsDNA. These DNA binding characteristics are similar to those of the homologous pairing proteins such as RAD51, RAD52, and DMC1. Then we tested the homologous pairing activity of mTBPIP/HOP2. To do so, we used the D-loop formation assay, which was used to detect the homologous pairing activity of the human and yeast DMC1 proteins (13,15). In this assay, a single-stranded 50-mer oligonucleotide and superhelical dsDNA were used as substrates, and the D-loop was formed as a product of the homologous pairing reaction ( Fig. 2A). As reported previously (13), the homologous pairing activity of the human DMC1 protein was detected in the D-loop formation assay (Fig. 2, B and C, lane 1). On the other hand, the same amount of mTBPIP/HOP2 (5 M) itself formed a trace amount of D-loop, which was only detected when the gel was overex- posed (Fig. 2, B and C). Therefore, mTBPIP/HOP2 shows very little homologous pairing activity in the D-loop formation assay as compared with DMC1.
The TBPIP/HOP2 Protein Enhances the DMC1-mediated Homologous Pairing-We next tested whether mTBPIP/HOP2 affects homologous pairing by DMC1. We used the human DMC1  5, 8, 11, and 14). The reactions were continued for 0 min (lanes 3, 4, 5, 9, 10, and 11) and 10 min (lanes 1, 2, 6, 7,  8, 12, 13, and 14). Lane 1 indicates a negative control experiment without protein, and lane 2 indicates a control experiment with DMC1 alone. C, the D-loop assay with RAD51 and mTBPIP/HOP2. The reactions were conducted in the same manner as those in A except that the human RAD51 protein was used instead of DMC1. Reaction orders are indicated on the top of C. When mTBPIP/HOP2 was first incubated with ssDNA, RAD51 was added to the reaction mixture 5 min before reaction initiation by the addition of dsDNA (lane 1). Lane 3 indicates the negative control experiment without protein, and lanes 4 and 5 indicate control experiments with RAD51 alone and DMC1 alone, respectively. D, the D-loop assay with RecA and mTBPIP/HOP2. A 32 P-labeled single-stranded 50-mer oligonucleotide (1 M) was incubated with 5 M RecA (lanes 2, 4, 6, and 8) in the presence of 1 mM ATP (lanes 2, 3, and 4) 3, 4, 7, and 8). Reactions were conducted for 3 min after initiation. protein, which shares 97% amino acid identity with the mouse DMC1 protein. When mTBPIP/HOP2 (5 M) was incubated with superhelical dsDNA before mixing with DMC1 (5 M) and ssDNA for the initiation of the homologous pairing reaction, the D-loop yield was significantly increased (Fig. 3A, lane 2) as compared with the reaction without mTBPIP/HOP2 (lane 3). The D-loop formation promoted by DMC1 was inhibited in the presence of ADP (Fig. 3B, lanes 9 -14) because DMC1 catalyzes D-loop formation in an ATP-dependent manner (13,15). As shown in Fig. 3B (lanes 3-8), the D-loop formation promoted by DMC1 and mTBPIP/HOP2 was also inhibited in the presence of ADP. On the other hand, the mTBPIP/HOP2 protein contains neither the ATP-binding motif nor the ATPase activity (data not shown). These results further confirmed that DMC1 is the catalytic center for homologous pairing in the presence of DMC1 and mTBPIP/HOP2. When mTBPIP/HOP2 was mixed with ssDNA before the addition of DMC1, the D-loop formation by DMC1 was signifi-cantly inhibited (Fig. 3A, lane 1). Therefore, mTBPIP/HOP2 only enhanced the DMC1-mediated homologous pairing reaction when it was bound to dsDNA, not ssDNA, before the initiation of the homologous pairing reaction.
Intriguingly mTBPIP/HOP2 did not enhance homologous pairing by the human RAD51 protein, another eukaryotic RecA homologue. Although the homologous pairing activity of the human RAD51 protein alone is too weak to be detected by the D-loop formation assay (31,32), mTPBIP/HOP2 did not enhance the RAD51-mediated D-loop formation to a detectable level (Fig. 3C). In addition, mTPBIP/HOP2 significantly inhibited the RecA-mediated D-loop formation when it was incubated with superhelical dsDNA before mixing with RecA (5 M) and ssDNA (Fig. 3D). These results indicate that mTBPIP/HOP2 does not enhance either RAD51-or RecA-dependent homologous pairing. Instead it may specifically function in the DMC1-dependent homologous pairing, which is probably essential for recombination between homologous chromosomes in meiosis. The mTBPIP/HOP2 Protein Does Not Inhibit D-loop Dissociation-In the present study, we found that the human DMC1 protein formed D-loops in 5-10 min and dissociated them after 10 min (Fig. 4, A and C). Under the same reaction conditions (5 M RecA, 1 M 50-mer ssDNA, and 30 M dsDNA), RecA formed D-loops in 1 min and quickly dissociated them (Fig. 4, D and E).  1, 2, 4, and 6) or 1 mM AMP-PNP (lanes 3, 5, and 7). human RAD52 proteins (29,35). When mTBPIP/HOP2 was incubated with dsDNA before DMC1 and ssDNA were added, the D-loop yield at 10 min was increased about 2.9-fold relative to that promoted by DMC1 alone, but the D-loop dissociation was still observed (Fig. 4, B and C). These results indicate that mTBPIP/HOP2 stimulates the DMC1-mediated homologous pairing but does not inhibit the D-loop dissociation by DMC1.
During the D-loop formation reaction, about 30% of the dsDNA, corresponding to about 7% of the ssDNA, was incorporated into D-loops by RecA within 1 min (Fig. 4, D and E). On the other hand, DMC1 converted only about 8% of the dsDNA, corresponding to about 1.8% of the ssDNA, to D-loops (Fig. 4, A and E) in the presence of ATP at 5 or 10 min. These results indicate that the ATP-dependent homologous pairing ability of DMC1 is about 25% of that of RecA at the optimal reaction times.

AMP-PNP Significantly Enhances the DMC1-dependent Homologous Pairing and Inhibits D-loop Dissociation-
The Dloop formation by the yeast Dmc1 protein is reportedly enhanced by AMP-PNP, which is a non-hydrolyzable analogue of ATP (15). Consist with this previous observation, AMP-PNP significantly stimulated the D-loop formation by the human DMC1 protein. After a 10-min reaction with AMP-PNP, DMC1 formed about 5-fold more D-loops than in the experiment with ATP (Fig. 5, A and B). This D-loop yield with DMC1 and AMP-PNP is almost the same as that with RecA and ATP.
Then we tested whether the mTBPIP/HOP2 protein stimulates the DMC1-dependent homologous pairing in the presence of AMP-PNP. As shown in Fig. 5, A and B, mTBPIP/HOP2 enhanced more than double the D-loop yield by DMC1 alone in the presence of AMP-PNP (Fig. 5, A, lanes 5 and 7, and B).
These results indicate that AMP-PNP is a better cofactor than ATP for homologous pairing by DMC1 and does not inhibit the TBPIP/HOP2 activity during homologous pairing.
In the presence of ATP, DMC1 promoted D-loop formation and dissociation (Fig. 4); however, time course experiments revealed that the D-loop dissociation by DMC1 (with mTBPIP/ HOP2) did not occur in the presence of AMP-PNP (Fig. 5, C and  D). These results indicate that the D-loop formation (homologous pairing) by DMC1 requires ATP or AMP-PNP binding but not ATP hydrolysis, and its dissociation strongly requires ATP hydrolysis.
In the present study, we found that mTBPIP/HOP2 stimulated the homologous pairing catalyzed by DMC1, which is a meiosis-specific homologous pairing protein. The stimulation of homologous pairing by mTBPIP/HOP2 only occurred when it binds dsDNA first. When mTBPIP/HOP2 first binds to ssDNA, the DMC1-mediated homologous pairing was rather inhibited. These results suggest that, during meiotic recombination, TB-PIP/HOP2 binds to the donor dsDNA, which does not contain DSBs, and DMC1 is assembled on the recipient ssDNA tails at DSB sites independently. Actually the yeast Hop2 protein binds to chromatin before the loading of the Rad51 and Dmc1 proteins, and this Hop2 localization is observed even in the absence of DSBs (22,37). These facts are consistent with a model in which TBPIP/HOP2 first binds to dsDNA in the DMC1-mediated homologous pairing during meiosis (28).
As described in this study, DMC1 catalyzes homologous pairing between ssDNA and superhelical dsDNA, but its ability to form D-loops is quite weak as compared with that of RecA. Although this low level of D-loop formation by the human DMC1 protein is consistent with previous observations with the human and yeast DMC1 proteins (13,15), a better reaction for the DMC1-dependent homologous pairing is required to evaluate ancillary factors, such as TBPIP/HOP2, which may function with DMC1 during homologous pairing. To develop better reaction conditions for the DMC1-dependent homologous pairing, we performed the D-loop formation assay with AMP-PNP, a non-hydrolyzable analogue of ATP, because AMP-PNP is known as a better cofactor than ATP for D-loop formation by the yeast Dmc1 protein (15). Consistent with this previous observation, we found that AMP-PNP enhanced the D-loop yield by the human DMC1 protein about 5-fold. In the presence of AMP-PNP, DMC1 formed almost the same amount of Dloops as did RecA with ATP. Then we tested whether mTBPIP/ HOP2 activates homologous pairing by DMC1 in the presence of AMP-PNP and found that the mTBPIP/HOP2 protein signif-icantly enhanced the DMC1-dependent homologous pairing in the presence of AMP-PNP. These results indicate that the mTBPIP/HOP2 protein stimulates the Dmc1-dependent homologous pairing, and even the basal level of homologous pairing by DMC1 is enhanced in the presence of AMP-PNP.
These biochemical observations are consistent with the results from previous genetic studies, which suggested that TB-PIP/HOP2 functions together with DMC1, probably in the homologous pairing step (25,28). Therefore, we conclude that TBPIP/HOP2 is a factor that positively functions with DMC1 in the strand invasion step of homologous pairing. However, we cannot eliminate the possibility that the DMC1 and TBPIP/ HOP2 proteins may have another function to ensure the proper pairing of homologous chromosomes during meiosis because the homologous pairing ability of DMC1 is weakly detected in the presence of ATP as compared with that of RecA. Further analyses will be required to clarify this issue.
Recently we determined the crystal structure of the human DMC1 octameric ring, a functional form of DMC1, and revealed the structural basis for octameric ring formation (30). Although the monomeric structures of DMC1, RAD51, and RecA are very similar (30, 39 -41), RAD51 and RecA form helical filaments, which are considered to be their functional form. In the present study, we found that TBPIP/HOP2 specifically activated homologous pairing by DMC1 but not by RAD51 or RecA. The TBPIP/HOP2 protein may specifically function with a homologous pairing protein that forms an octameric ring rather than a helical filament.
The yeast Hop2 protein reportedly forms a complex with the Mnd1 protein, which has been identified as a multicopy suppressor of a temperature-sensitive hop2 mutant allele (37). The MND1 gene is also required for meiotic recombination (38). The mnd1-null mutant exhibits a strikingly similar phenotype to that of the hop2-null mutant (37,38). This genetic evidence suggests that the HOP2 and MND1 proteins function as a complex to promote meiotic chromosome pairing, and the HOP2-MND1 complex may be a cofactor for the RAD51 and DMC1 recombinases. In the present study, we showed that the mammalian TBPIP/HOP2 protein stimulates the DMC1-mediated homologous pairing, indicating that the TBPIP/HOP2 subunit in the HOP2-MND1 complex possesses the ability to activate homologous pairing by the meiosis-specific DMC1 protein.
It would be intriguing to study how the DMC1-TBPIP/HOP2mediated homologous pairing functions to discriminate homologous chromosomes from nonhomologous chromosomes during meiosis. The MND1 protein, like TBPIP/HOP2, is highly conserved from yeast to human. Analyses of the biochemical activities of the MND1 protein and the HOP2-MND1 complex will be the next central issue to reveal the molecular mechanism of the meiosis-specific homologous pairing between homologous chromosomes.