The mechanism of Mus81-Mms4 cleavage site selection distinguishes it from the homologous endonuclease Rad1-Rad10

Abstract

Mus81-Mms4 and Rad1-Rad10 are homologous structure-specific endonucleases that cleave 3' branches from distinct substrates and are required for replication fork stability and nucleotide excision repair, respectively, in the yeast Saccharomyces cerevisiae. We explored the basis of this biochemical and genetic specificity. The Mus81-Mms4 cleavage site, a nick 5 nucleotides (nt) 5' of the flap, is determined not by the branch point, like Rad1-Rad10, but by the 5' end of the DNA strand at the flap junction. As a result, the endonucleases show inverse substrate specificity; substrates lacking a 5' end within 4 nt of the flap are cleaved poorly by Mus81-Mms4 but are cleaved well by Rad1-10. Genetically, we show that both mus81 and sgs1 mutants are sensitive to camptothecin-induced DNA damage. Further, mus81 sgs1 synthetic lethality requires homologous recombination, as does suppression of mutant phenotypes by RusA expression. These data are most easily explained by a model in which the in vivo substrate of Mus81-Mms4 and Sgs1-Top3 is a 3' flap recombination intermediate downstream of replication fork collapse.

FIG. 1.

Purification and substrate preference of Rad1-Rad10 and Mms4-Mus81. (A) Five hundred nanograms of Rad1-10 and 250 ng of Mms4-Mus81 were subjected to sodium dodecyl sulfate-15% polyacrylamide gel electrophoresis and visualized by silver staining. The positions of the individual subunits His6-Rad1 (160 kDa), Rad10 (27 kDa), His6-Mms4 (80 kDa), and Mus81 (72 kDa) are shown. Molecular size markers (lanes M) are indicated in kilodaltons. (B) Increasing concentrations of Rad1-10 (0, 18, 35, 70, and 140 nM) or Mms4-Mus81 (0, 0.5, 1.2, 2.5, and 5.0 nM) were incubated with 50 nM (each) ³²P-labeled simple Y and 3′ flap substrates, and the products were resolved on a 10% native polyacrylamide gel prior to autoradiography. (C) The data in panel B were quantitated and are presented as percent maximum (Max) hydrolysis for Rad1-Rad10 (left) and Mus81-Mms4 (right). (D) Sixty-seven nanomolar Rad1-10 (left) or 5 nM Mus81-Mms4 (right) was incubated in the presence of a single ³²P-labeled substrate (50 nM). Eighteen-microliter aliquots were removed at the indicated times and analyzed as described above, and the product contained inthe aliquot was plotted as a function of time. Simple Y and 3′ flap substrates were assembled from oligonucleotides *888/1128 and *888/1128/994, respectively. The asterisks indicate 5′ ³²P labeling.

FIG. 2.

Inverse activities of Mms4-Mus81 and Rad1-Rad10. (A) Mms4-Mus81 (0, 0.2, 0.4, 0.8, 1.6, or 3.2 nM) was incubated in the presence of a 0.1 nM concentration of the indicated substrate, and the products were analyzed on a 10% native polyacrylamide gel. Gap refers to the number of unpaired nucleotides 3′ of the flap. (B) Duplicate reactions were incubated with (+) or without (−) 70 nM Rad1-10 endonuclease in the presence of a 50 nM concentration of the indicated gapped substrate, as illustrated in panel A. (C) Following quantitation, the product obtained with 1.6 nM Mms4-Mus81 in panel A and 70 nM Rad1-Rad10 in panel B was plotted as a function of substrate gap size. Substrates were assembled as follows: 3′ flap, 0-bp gap (*888/1128/994), 1-bp gap (*888/1128/1038), 3-bp gap (*888/1128/1039), or 5-bp gap (*888/1128/1040); simple Y, 25-bp gap (*888/1128).

FIG. 3.

The cleavage site of Mus81-Mms4 is sequence independent. (A) 3′ flap junctions were incubated in the presence (+) or absence (−) of Mus81-Mms4, and the products were resolved by sequencing gel electrophoresis alongside a chemical sequencing ladder of the labeled oligonucleotide. As indicated, the Maxam-Gilbert ladder of products runs faster than corresponding fragments terminating in a 3′ OH. (B) Sequences, structures, and cleavage sites of the substrates used in the experiment whose results are shown in panel A.

FIG. 4.

Upstream determinants for Mus81-Mms4 cleavage. Various 3′ flap (left) or gapped duplex (right) substrates were designed with the indicated flap (Y) or upstream (X) sizes given in nucleotides. Substrates were incubated with (+) or without (−) Mus81-Mms4, and the products were resolved on a 10% sequencing gel. Substrates were assembled as follows: 3′ flap with a 24-nt branch (*888/1128/994) or 20- to 0-nt branches (*1078-1084/1128/994); gapped-duplex substrates (*1115-1119/1128/994). A minus sign indicates a lane in which the untreated substrate remained in the well. The values on the left and right are numbers of nucleotides. Enz, enzyme; oligo, oligonucleotide; T, template.

FIG. 5.

Downstream determinants of Mus81-Mms4 cleavage. Various duplex (A) or 3′ flap (B) substrates were designed with the indicated downstream oligonucleotides (Z). Substrates were incubated with (even-numbered lanes) or without (odd-numbered lanes) Mus81-Mms4, and the products were resolved on a 10% sequencing gel. (C) RF (*888/1128/992/994) or Y-junction (*888/1128/1127) substrates with identical sequences were assembled and incubated at 0.2 nM with 75, 150, 300, 600, 1,200 pM Mus81-Mms4. Reaction mixtures were analyzed by native gel electrophoresis. Reaction substrates and products are illustrated. Enz, enzyme.

FIG.6.

MUS81-MMS4 functions downstream of homologous recombination. (A) S. cerevisiae strains with the indicated genotypes were concentrated to an optical density at 600 nm of 3.0 and serially diluted 10-fold, and approximately 5-μl volumes were spotted on YPD (38) plates containing the indicated drug. Plates were photographed following 3 days of growth at 30°C. WT, wild type. (B) The indicated double and triple mutants were constructed by genetic crosses between appropriately marked strains, some of which carried a complementing SGS1/URA3 plasmid (34). Following sporulation and tetrad dissection, the viability of the meiotic segregants was determined by attempting to eliminate the SGS1/URA3 plasmid by growth on medium containing 5-fluoroorotic acid. wt, wild type. (C, part a) Yeast cells with the indicated genotypes were spread on YPD plates and irradiated with UV light, and viability was determined following 3 days of growth at 30°C. (C, parts b to f) Yeast cells carrying an empty vector (pRS415) or the RusA expression plasmid (pKR6980) were spread on the appropriate selective plates and treated as in the experiment whose results are shown in part a. For CPT sensitivity testing, yeast cells were grown in selective medium containing CPT at 5 μg/ml. After the indicated times, aliquots of cells were removed, washed once, and plated on selective plates lacking CPT. Viability was then determined following 3 days of growth at 30°C. WT, wild type.

FIG. 7.

Mechanism of Mus81-Mms4 cleavage and its functional overlap with Sgs1-Top3. (A) Mus81-Mms4 (transparent gray dimer) recognizes the 5′ end of the DNA at a flap junction and nicks the flap strand 5 nt upstream within the DNA duplex (gray arrow). Conditions a through d illustrate Mus81-Mms4 activity on various substrates as described in the text. (B) Model of DNA repair following fork stalling or a CPT-induced DSB illustrating the formation of a 3′ flap substrate following SDSA. This substrate is appropriate for cleavage by Mus81-Mms4 to create a 5-nt gap (left). Alternatively, Sgs1-Top3 is proposed to process this substrate by displacing the downstream strand and annealing the newly synthesized flap DNA (right). indep., independent.