Nse2, a component of the Smc5-6 complex, is a SUMO ligase required for the response to DNA damage

Abstract

The Schizosaccharomyces pombe SMC proteins Rad18 (Smc6) and Spr18 (Smc5) exist in a high-M(r) complex which also contains the non-SMC proteins Nse1, Nse2, Nse3, and Rad62. The Smc5-6 complex, which is essential for viability, is required for several aspects of DNA metabolism, including recombinational repair and maintenance of the DNA damage checkpoint. We have characterized Nse2 and show here that it is a SUMO ligase. Smc6 (Rad18) and Nse3, but not Smc5 (Spr18) or Nse1, are sumoylated in vitro in an Nse2-dependent manner, and Nse2 is itself autosumoylated, predominantly on the C-terminal part of the protein. Mutations of C195 and H197 in the Nse2 RING-finger-like motif abolish Nse2-dependent sumoylation. nse2.SA mutant cells, in which nse2.C195S-H197A is integrated as the sole copy of nse2, are viable, whereas the deletion of nse2 is lethal. Smc6 (Rad18) is sumoylated in vivo: the sumoylation level is increased upon exposure to DNA damage and is drastically reduced in the nse2.SA strain. Since nse2.SA cells are sensitive to DNA-damaging agents and to exposure to hydroxyurea, this implicates the Nse2-dependent sumoylation activity in DNA damage responses but not in the essential function of the Smc5-6 complex.

FIG. 1.

Nse2 has homology to SUMO ligases. (A) Sequence alignment of the N-terminal sequences of S. pombe Nse2 (aa 1 to 237 or 250) and Pli1 (aa 1 to 375 or 727) with those of S. cerevisiae Siz1 (aa 1 to 420 or 750) and Mms21 (1 to 253 or 267), created by use of the ClustalW program. Dark shading, identical amino acids; light shading, conserved residues, *, conserved Cys and His residues (C195 and H197 in Nse2). (B) Percentages of identity between S. cerevisiae Mms21 and Siz1 and S. pombe Pli1 and Nse2. The first figure in each pair is the percent identity along the full length of the proteins, and the figure in parentheses is the percent identity between RING-finger-like domains.

FIG. 2.

Nse2 has SUMO ligase activity. (A) Expression of Nse2 in E. coli. ORF SPAC16A10.06c, which was amplified by PCR and cloned into pGEX, was expressed in BL21 cells, purified by the use of glutathione-Sepharose (lane 2), and compared to extracts from cells transformed with the empty pGEX vector (lane 1). (B) Pmt3 sequence requirements. The results of an in vitro sumoylation assay to test three forms of Pmt3 (Pmt3.GG, Pmt3.GG,K30R, and Pmt3) are shown. Hus5 (0.05 μg/μl) was used in all assays. The products were analyzed by Western blotting with anti-Pmt3 antisera. The species migrating at about 35 and 50 kDa are likely SUMO chains and reaction intermediates. The arrow indicates the junction of stacking and separating gels. (C) Nse2 is itself sumoylated. An in vitro sumoylation assay was performed with ³⁵S-labeled Nse2 as a substrate and with 0.05 μg of Hus5/μl. The products were separated by SDS-PAGE and detected with a phosphorimager. The species migrating at 37 kDa is a nonspecific band. (D) Sumoylation of Nse2 occurs predominantly on the C-terminal part of the protein in vitro. An in vitro sumoylation assay was performed with ³⁵S-labeled N- and C-terminal fragments of Nse2 as indicated. Nse2-N, aa 1 to 178; Nse2-C, aa 114 to 250. The products were detected as described above. *, modified forms.

FIG. 3.

Nse2 is present in high-molecular-mass complexes. A wild-type cell extract was analyzed by gel filtration on a Superdex 200 column and Western blotted with antisera as indicated.

FIG. 4.

Smc6 (Rad18) and Nse3 are sumoylated in an Nse2-dependent manner in vitro, and Smc6 (Rad18) is sumoylated in vivo. (A to D) ³⁵S-labeled proteins were tested with in vitro sumoylation assays using 0.05 μg of Hus5/μl and 0 or 0.2 μg of Nse2 or Pli1/μl, as indicated. (A) Smc6 (Rad18); (B) Smc5 (Spr18); (C) Nse1; (D) Nse3. (E) Smc6 (Rad18) is sumoylated in vivo. Ni²⁺ pull-down assays were performed with cell extracts from cells transformed with pREP42MH (empty vector) (lanes 1, 3, 5, and 7) or pREP42MH-Rad18 (Smc6) (lanes 2, 4, 6, and 8). TCA, total cell extract controls. Western analysis was conducted with anti-Myc or anti-Pmt3 antisera as indicated. (F) Sumoylation of Smc6 (Rad18) expressed at wild-type levels increases after exposure to MMS. Lysates (containing 50 mg of total protein) prepared from a Myc-tagged Smc6 (Rad18) strain and an untagged control with or without exposure to MMS (0.01%) were incubated overnight at 4°C with an anti-Myc antibody that had been previously cross-linked to protein G-Sepharose beads. The beads were washed extensively, and bound proteins were eluted by incubation with 100 mM glycine, pH 2.3, separated by SDS-PAGE, and analyzed by Western blotting (WB) with anti-Myc and anti-Pmt3 antibodies, as indicated. Lanes 1, 2, 5, 7, and 8, Myc-tagged Rad18 (Smc6) strain; lanes 3, 4, 6, 9, and 10, wild type (untagged Rad18 [Smc6]). Lanes 5 and 6 are the same as lanes 7 and 9, but with longer exposure times. *, modified forms.

FIG. 5.

Mutation of C195 and H197 results in a loss of sumoylating activity in vitro and in vivo. (A) Nse2.SA protein is unable to form SUMO chains in vitro. An in vitro sumoylation assay was performed in the absence of an added target protein, with 0.05 μg of Hus5/μl. The products were detected by Western analysis with anti-Pmt3 antisera. The arrow indicates the junction of stacking and separating gels. (B) Nse2.SA does not facilitate sumoylation of Smc6 (Rad18) in vitro. An in vitro sumoylation assay was performed with ³⁵S-labeled Smc6 (Rad18) and 0.05 μg of Hus5/μl. The products were detected with a phosphorimager. (C) Nse2.SA is not sumoylated in vitro. An in vitro sumoylation assay was performed with an ³⁵S-labeled wild-type or mutant (Nse2.SA) Nse2 protein. The products were detected as described for panel B. (D) Western analysis of total cell extracts probed with anti-Pmt3 antisera (top) or anti-tubulin antisera (bottom). Lane 1, wild type (sp.011); lane2, rad31.d; lane 3, hus5.17; lane 4, hus5.62; lane 5, pli1.d; lane 6, nse2.SA; lane 7, nse2.CH. (E) nse2.SA cells have reduced levels of sumoylated Smc6 (Rad18). Extracts of wild-type (nse2.CH) (lanes 3 and 4) and mutant (nse2.SA) (lanes 1 and 2) strains harboring either pREP41MH-Rad18 (Smc6) (lanes 1 and 3) or the empty pREP41MH vector (lanes 2 and 4) were bound to nickel beads under denaturing conditions, and the bound proteins were analyzed by immunoblotting with either anti-Myc or anti-Pmt3 antibodies as indicated. *, modified forms.

FIG. 6.

nse2.SA cells are sensitive to DNA-damaging agents. (A) UV survival analysis. (B) IR survival analysis. (C) HU and MMS sensitivities. Five-microliter samples of 10-fold dilutions of exponentially growing cultures (5 × 10⁶/ml) were plated onto yeast extract-agarose plates with supplements as indicated. (D) PFGE. DNAs from wild-type, nse2.SA, and rad18.X cells before and after exposure to 450 Gy of IR were analyzed by PFGE. (E and F) Epistasis analysis with rad18.T2 (E) and rhp51.d (F) cells. (G) Analysis of DNA damage checkpoint. Cells that were synchronized by lactose gradients were incubated at 30°C after no treatment (▪), 200 Gy of IR (⧫), or 400 Gy of IR (▴). Samples were taken at 20-min intervals and fixed in methanol. They were stained with DAPI and calcofluor, and the percentages of cells that passed mitosis were assessed.