Human Rad54 protein stimulates human Mus81-Eme1 endonuclease

Abstract

Rad54, a key protein of homologous recombination, physically interacts with a DNA structure-specific endonuclease, Mus81-Eme1. Genetic data indicate that Mus81-Eme1 and Rad54 might function together in the repair of damaged DNA. In vitro, Rad54 promotes branch migration of Holliday junctions, whereas the Mus81-Eme1 complex resolves DNA junctions by endonucleolytic cleavage. Here, we show that human Rad54 stimulates Mus81-Eme1 endonuclease activity on various Holliday junction-like intermediates. This stimulation is the product of specific interactions between the human Rad54 (hRad54) and Mus81 proteins, considering that Saccharomyces cerevisiae Rad54 protein does not stimulate human Mus81-Eme1 endonuclease activity. Stimulation of Mus81-Eme1 cleavage activity depends on formation of specific Rad54 complexes on DNA substrates occurring in the presence of ATP and, to a smaller extent, of other nucleotide cofactors. Thus, our results demonstrate a functional link between the branch migration activity of hRad54 and the structure-specific endonuclease activity of hMus81-Eme1, suggesting that the Rad54 and Mus81-Eme1 proteins may cooperate in the processing of Holliday junction-like intermediates during homologous recombination or DNA repair.

Fig. 1.

hRad54 stimulates hMus81 endonuclease activity. nHJ (20 nM, molecules) were incubated with hRad54 (120 nM, or indicated otherwise) or its dilution buffer. DNA cleavage was initiated by addition of hMus81 (2.5 nM or as indicated otherwise). DNA cleavage was carried out for 20 min or as indicated otherwise. (A) The kinetics of nHJ resolution. The vertical arrow on the nHJ scheme indicates the hMus81 incision site. The arrows at the right of the gel indicate migration of the DNA substrate and the cleavage product. (B) Graphical representation of the data from A. (C) The effect of hRad54 concentration. (D) The effect of hMus81 concentration. The data are the mean of at least 3 measurements, and the error bars represent the SE. *, ³²P-label at the 5′ DNA end.

Fig. 2.

hRad54 stimulates hMus81 endonuclease activity on various DNA substrates. (A) hRad54 (120 nM) was incubated with DNA substrates (20 nM, molecules), hMus81 (5 nM, or 10 nM in the case of HJ) was added to initiate DNA cleavage. ³²P-label was in oligonucleotide 369, common for all DNA substrates. Note that cleavage of forked DNA would produce 5′-tailed DNA. (B) Experimental scheme to test hRad54 stimulation of hMus81 on mobile PX-junctions. The PX-junction was designed in such way that branch migration can proceed only in 1 direction of movement, shown by the block arrow. The zig-zag line denotes the poly(dT)₃₀ region. Single base pair heterologies are denoted by AT and GC. (C) PX-junction (20 nM, molecules) cleavage by hMus81 (5 nM) alone (lanes 1–4), in the presence of hRad54 (120 nM) (lanes 5–7), or branch migration by hRad54 (120 nM) (lanes 8–10). In lanes 5–7, hRad54 was added first, followed by immediate addition of hMus81. In lane 11 (denoted 5^spon), hRad54 was replaced with dilution buffer. The arrows at the sides of the gels indicate migration of DNA substrates and products of PX cleavage or branch migration. *, ³²P-label at the 5′ DNA end.

Fig. 3.

Yeast Rad54 inhibits hMus81 resolution activity. hRad54 (120 nM, lanes 3 and 6) or yRad54 (120 nM, lane 4) proteins were incubated with PX-junctions (20 nM, molecules), then hMus81 (5 nM, lanes 2–4) or the hMus81AA mutant (5 nM, lanes 5 and 6) were added to the reaction mixture. In lanes 2 and 5, hRad54 protein was replaced with dilution buffer. *, ³²P-label at the 5′ DNA end.

Fig. 4.

Effect of the order of protein addition on hMus81 stimulation by hRad54. (A) In lane 3, hRad54 (120 nM) and hMus81 (5 nM) were mixed and then added to PX-junctions (20 nM, molecules) followed by a 20-min incubation. In lanes 4 and 5, hRad54 was incubated with PX-junctions for 5 and 10 min, respectively, then hMus81 was added and the cleavage was carried out for 20 min. In lanes 6 and 7, hRad54 was added to an on-going DNA cleavage by hMus81 at 5 and 10 min after initiation, and the reactions were continued for 15 and 10 min, respectively. In lane 2, hRad54 was replaced with dilution buffer, and hMus81 was incubated with PX-junction for 20 min. (B) Graphical representation of the data from A. The data are the mean of at least 3 measurements, and the error bars represent the SE. *, ³²P-label at the 5′ DNA end.

Fig. 5.

Effect of nucleotide cofactors on the stimulation of hMus81 by hRad54. (A) hRad54 (120 nM) was incubated with nHJ (20 nM, molecules) for 5 min in the presence of ATP (bars 1 and 2), ATPγS (bars 5 and 6), or ADP (bars 7 and 8) or in the absence of a nucleotide cofactor (bars 3 and 4). DNA cleavage was initiated by addition of hMus81 (5 nM) and carried out for 20 min. The ATP regeneration system was included only in the reactions containing ATP. Magnesium acetate concentrations in the presence of ATP, absence of ATP, and presence of ATPγS or ADP were 7.5, 3, and 5 mM, respectively, which were optimal for each reaction. (B) hRad54 protein (120 nM) was incubated with nHJ (20 nM, molecules) for 2 min in the absence of a nucleotide cofactor. Then, the reaction mixture was divided between 5 test tubes. In the first, DNA cleavage was initiated by addition of hMus81 (5 nM, lane 3). In the second, incubation was continued for another 5 min, and then hMus81 was added (lane 4). In the third, fourth, and fifth, ATP, ATPγS, and ADP, respectively, were added and incubation was continued for 5 min, followed by addition of hMus81 (lanes 5–7). After addition of hMus81, all reactions were carried out for 20 min. hMus81 cleavage activity, in the absence of hRad54 and nucleotide cofactors, is shown in lane 2. (C) Graphical representation of the data from A. The data are the mean of 3 measurements, and the error bars represent the SE. *, ³²P-label at the 5′ DNA end.

Fig. 6.

Effect of hRad54 on hMus81 cleavage specificity on mobile PX-junctions. The conditions of DNA cleavage by hMus81 were identical to those described in the legend of Fig. 2C. (A) The reactions were carried out either in the absence or presence hRad54 (120 nM) (lanes 3 and 4, respectively) for 5 min and the DNA products were analyzed in a 12% denaturing polyacrylamide gel (19:1). Migration markers were produced by cleavage of oligonucleotides 71 and 169 at A+G positions (lanes 1 and 5, respectively). (B) The cleavage sites (denoted by “q” through “z”) are indicated by arrows. Main cleavage sites are designated by capital letters (R, S, and T). *, ³²P-label at the 5′ DNA end.