Cockayne syndrome B protein stimulates apurinic endonuclease 1 activity and protects against agents that introduce base excision repair intermediates

Abstract

The Cockayne syndrome B (CSB) protein--defective in a majority of patients suffering from the rare autosomal disorder CS--is a member of the SWI2/SNF2 family with roles in DNA repair and transcription. We demonstrate herein that purified recombinant CSB and the major human apurinic/apyrimidinic (AP) endonuclease, APE1, physically and functionally interact. CSB stimulates the AP site incision activity of APE1 on normal (i.e. fully paired) and bubble AP-DNA substrates, with the latter being more pronounced (up to 6-fold). This activation is ATP-independent, and specific for the human CSB and full-length APE1 protein, as no CSB-dependent stimulation was observed with Escherichia coli endonuclease IV or an N-terminal truncated APE1 fragment. CSB and APE1 were also found in a common protein complex in human cell extracts, and recombinant CSB, when added back to CSB-deficient whole cell extracts, resulted in increased total AP site incision capacity. Moreover, human fibroblasts defective in CSB were found to be hypersensitive to both methyl methanesulfonate (MMS) and 5-hydroxymethyl-2'-deoxyuridine, agents that introduce base excision repair (BER) DNA substrates/intermediates.

Figure 1.

CSB associates with APE1. Interaction between CSB and APE1 in indirect ELISA. Mouse anti-APE1 (panel A) or anti-CSB (panel B) was coated to the well. Purified APE1 (30 fmol, 1 ng) or CSB (6 fmol, 1 ng) was then bound to the appropriate antibody. These complexes were next incubated with different concentrations of CSB (panel A) or APE1 (panel B) as indicated. Ethidium bromide (EtBr) or DNase I was added where indicated. Values are means ± standard errors (n = 3). (C) CSB and Ape1 are in a common complex in human cell extracts. Lysates from CS1AN cells transfected with the ECFP-CSB plasmid were immunoprecipitated (IP) with either ECFP (left) or APE1 (right) antibodies. The IPs were then analyzed by immunoblotting (IB) with either anti-CSB or anti-APE1 antibodies. Lysates of 10% of the amount used for IP were loaded as input comparisons. As controls, lysates IPed with rabbit IgG only were loaded (IgG Control), and IPs with anti-ECFP were performed on extracts from ECFP-only expressing cells (Vector Control).

Figure 2.

CSB stimulates APE1 endonuclease activity, independent of ATP. (A) APE1 incision of 42F fully paired duplex substrates with or without CSB. Reactions contained 10 fmol of APE1 and 200 fmol of 42F-11: 42 comp duplex DNA, either alone or with increasing amounts of CSB (10, 30, 100 and 300 fmol). CSB alone (300 fmol) was also assayed. (B) APE1 incision of a 42-nt bubble substrate with or without CSB. Reactions were carried out as above, except with 30 fmol of APE1 and the 42F-11:42F bubble comp duplex substrate. AP endonuclease activity is the amount of radiolabeled F-DNA substrate (S) converted to the shorter incised DNA product (P) in 10 min. In panels A and B, values in graphs represent the average and standard deviation of a least three independent data points. The molar ratio of APE1 to CSB is indicated. (C) CSB stimulation of APE1 activity is not affected by ATP. Reactions were performed as above in panel B. ATP (0.5 mM) was added in the reactions indicated. No enzyme, as well as APE1 and CSB (300 fmol) alone, reactions are shown. Percentage of S converted to P is denoted under each lane.

Figure 3.

CSB specifically activates full-length human APE1. (A) APE1 stimulation is dependent on CSB. Reactions were carried out as in Figure 2. Immunodepleted supernatant (ID) or immunoprecipitated (IP) CSB fractions were prepared as described in Materials and Methods section. Percentage of intact substrate (S) converted to incised product (P) is denoted under each lane, and compared with the effects of purified, recombinant CSB protein. (B) CSB stimulation is specific for human APE1. Reactions were performed with E. coli AP endonuclease IV (0.1 fmol; ∼3 pg) and increasing amounts of CSB (0.01, 0.03, 0.1, 0.3, 1, 3, 10, 30, 100 and 300 fmol). (C) Incision activity of truncated APE1 protein is not activated by CSB. Truncated APE1 lacking the first 29 amino acids (5 fmol) was incubated with increasing concentrations of CSB (1, 5, 20, 50 and 100 fmol) as in Figure 2. Representative gel images are shown in panels A, B and C. In panels B and C, a full-length APE1 (30 fmol) and CSB (300 fmol) control reaction was simultaneously performed. Percentage of S converted to P is denoted.

Figure 4.

Global AP site repair in CSB-V and CSB-WT cells. (A) Steady-state AP site levels in total chromosomal DNA from CSB-V and CSB-WT cells. AP sites measured per 10⁶ nucleotides (nt) are shown. Values represent the average and standard deviation of three independent measurements. (B) Total AP endonuclease activity of CSB-V and CSB-WT whole cell extracts. Shown is a representative gel of reactions performed at the indicated concentration of CSB-V or CSB-WT whole cell extract (μg/10 μl) with 34F duplex substrates (44). NE = no enzyme control. Percentage of intact substrate (S) converted to incised product (P) is denoted under each lane. (C) AP site incision kinetics.Three microgram/10 μl of the indicated extract was incubated for the time specified, and the percentage of S converted to P was determined. Plotted is the average and standard deviation of at least three data points. (D) Recombinant CSB activates AP site incision in whole cell extracts. Extracts (0.2 μg) were prepared from CSB-V cells and assayed for AP site incision efficiency, without or with supplemented recombinant CSB protein (10, 30, 100, 300 fmol).

Figure 5.

CSB deficient cells exhibit hypersensitivity to MMS and HmdU. (A) MMS sensitivity. CSB-V, CSB-E646Q, and CSB-WT cells were exposed to MMS at different concentrations (indicated) for 1 h, and subsequently incubated for 10 days in drug-free medium. The surviving fraction of cells (% survival) was calculated by dividing the number of colonies in treated dishes by those counted in the untreated control. (B) HmdU sensitivity. The indicated cell lines were exposed to HmdU at different concentrations (denoted) for 24 h and plated in complete media. The colony formation ability was determined after 10 days. Values shown in both panels represent the mean and standard deviation of six independent data points.