Chinese hamster apurinic ⁄ apyrimidinic endonuclease (chAPE1) expressed in sf9 cells reveals that its endonuclease activity is regulated by phosphorylation Mandula Borjigin 1 , Bobbie Martinez 2 , Sarla Purohit 2 , Gaudalupe de la Rosa 2 , Pablo Arenaz 2 and Boguslaw Stec 3 1 Department of Chemistry, Bowling Green State University, OH, USA 2 Department of Biological Sciences, Department of Chemistry, University of Texas, El Paso, USA 3 Sanford-Burnham Medical Research Institute, La Jolla, CA, USA Introduction The mammalian apurinic ⁄ apyrimidinic endonuclease (APE1) is a multifunctional protein that plays an essential role in DNA repair and gene regulation [1]. In particular, it is a critical component of the base excision repair pathway, which is employed to repair damaged DNA. The base excision repair pathway is initiated by spontaneous or enzymatic N-glycosidic bond cleavage creating an abasic site in DNA [2]. Aba- sic sites in DNA alter genetic information and hinder normal cellular activity, posing a major threat to the integrity of the DNA molecule and the survival of the cell [3–5]. The importance of APE1 is also underscored by the fact that homozygous knockout mice are embryonic lethal [6]. The mechanism of its prominent Keywords apurinic endonuclease; caseine kinase phsphorylation; DNA repair; enzyme kinetics; ICP; regulation by phosphorylation Correspondence B. Stec, Sanford-Burnham Medical Research Institute, 10901 N. Torrey Pines Rd, La Jolla, CA 92037, USA Fax: 858 795 5225 Tel: 858 795 5257 E-mail: bstec@burnham.org M. Borjigin, Department of Chemistry, 144 Overman Hall, Bowling Green State University, Bowling Green, OH 43403, USA Fax: 419 372 8088 Tel: 419 372 8088 E-mail: dman@bgsu.edu (Received 7 June 2010, revised 30 August 2010, accepted 10 September 2010) doi:10.1111/j.1742-4658.2010.07879.x Apurinic ⁄ apyrimidinic endonuclease (APE), an essential DNA repair enzyme, initiates the base excision repair pathway by creating a nick 5¢ to an abasic site in double-stranded DNA. Although the Chinese hamster ovary cells remain an important model for DNA repair studies, the Chinese hamster APE (chAPE1) has not been studied in vitro in respect to its kinetic characteristics. Here we report the results of a kinetic study per- formed on cloned and overexpressed enzyme in sf9 cells. The kinetic parameters were fully compatible with the broad range of kinetic parame- ters reported for the human enzyme. However, the activity measures depended on the time point of the culture. We applied inductivity coupled plasma spectrometry to measure the phosphorylation level of chAPE1. Our data showed that a higher phosphorylation of chAPE1 in the expression host was correlated to a lower endonuclease activity. The phosphorylation of a higher activity batch of chAPE1 by casein kinase II decreased the endonuclease activity, and the dephosphorylation of chAPE1 by lambda phosphatase increased the endonuclease activity. The exonuclease activity of chAPE1 was not observed in our kinetic analysis. The results suggest that noticeable divergence in reported activity levels for the human APE1 endonuclease might be caused by unaccounted phosphorylation. Our data also demonstrate that only selected kinases and phosphatases exert regula- tory effects on chAPE1 endonuclease activity, suggesting further that this regulatory mechanism may function in vivo to turn on and off the function of this important enzyme in different organisms. Abbreviations APE, apurinic ⁄ apyrimidinic endonuclease; ChAPE1, Chinese hamster apurinic ⁄ apyrimidinic endonuclease; CK I, casein kinase I; CK II, casein kinase II; ICP, inductivity coupled plasma. 4732 FEBS Journal 277 (2010) 4732–4740 ª 2010 The Authors Journal compilation ª 2010 FEBS endonuclease activity is that the enzyme incises the phosphodiester backbone 5¢ next to the abasic site (cleaving P-O-3¢ bond), leaving a 3¢-OH and a 5¢ deoxy- ribose phosphate [7]. Other important functions are duplex-specific 3¢–5¢ exonuclease activity, 3¢-repair phosphodiesterase activity, 3¢-phosphatase activity and RNase H activity [8–11]. Although many mammalian APEs were studied, the APE1 from Chinese hamster ovary cells was not stud- ied in vitro, despite being an important model for DNA repair mechanisms [12]. This is an important enzyme and Chinese hamster APE (chAPE1) should provide additional data that can bridge the gap between mouse and human models. There are quite noticeable discrepancies in reports concerning two major catalytic (endonuclease and 3¢–5¢ exonuclease) activities reported for several species. There is a sub- stantial spread in the level of endonuclease activity reported for human APE1, with K m ranges from 3.4 to 200 nm, k cat from 1.38 to 10 s )1 and k cat ⁄ K m from 0.05 to 0.5 nm )1 Æs )1 [13–18]. There is also controversy with regard to the 3¢–5¢ exonuclease activity of human APE1, for which robust activity has been reported [19,20], a much lower level ( 100–10 000-fold lower) than its endonuclease activity [13,21,22] or no measur- able activity [23–25]. For instance, the murine APE had approximately the same level of 3¢–5¢ exonuclease activities as its endonuclease activity [26,27] and the bovine and rat APE1 expressed in the bacterial cell do not exhibit 3¢–5¢ exonuclease activity [28,29]. Here we report the results of studies performed on the recombinant protein (chAPE1) using a steady-state kinetics method with radiolabeled substrates and an electrophoretic gel assay. The kinetic constants obtained it this study fell into the expected range, tak- ing into account the identity level compared with human enzyme (92%) and mouse enzyme (94%). However, we noticed significant variation from batch to batch of the enzyme, which was reminiscent of the abovementioned results. We hypothesize that the phos- phorylation might be responsible for a broad range of activity levels. We also speculate that the results obtained for chAPE1 might have full relevance to the results obtained for highly homologous mammalian endonucleases. The phosphorylation level of chAPE1 was quantita- tively analyzed by measuring the phosphate amount of protein using inductivity coupled plasma (ICP) spec- trometry. The endonuclease activity rate constant of the differentially phosphorylated naturally expressed chAPE1 was obtained by performing a steady-state kinetics analysis. To further verify the phosphorylation effects on chAPE1 endonuclease activity, the protein was subject to casein kinase I (CK I) or casein kinase II (CK II) and dephosphorylated with lambda phos- phatase or alkaline phosphatase. Their effects were quantified by performing the endonuclease assay and the kinetic parameters were obtained by fitting the data into a Michaelis–Menten model. We did not detect perceivable exonuclease activity of chAPE1 in our study. Results Overexpression of chAPE1 in the sf9 cell line and its purification The expression level of chAPE1 in insect cells infected by the recombinant baculovirus with a multiplicity of infection of eight reached the plateau at 48 h and declined after 72 h postinfection. The western blot showed the expression level in the selected time course (Fig. 1). The protein was purified using a Ni-NTA col- umn and a size exclusion column and the histidine tag was cleaved with enterokinase; the native protein appeared as a band at  35.5 kDa (Fig. 2). The purity was > 90%, as judged by gel electrophoresis. Fig. 1. Expression profile of chAPE1 in sf9 cells. Western blot autoradiograph: Lanes 1–8 correspond to the time points of chA- PE1 expression after infection by recombinant virus. The time points are 0, 6, 12, 18, 24, 48, 60 and 72 h, respectively. Fig. 2. SDS ⁄ PAGE of chAPE1 stained with Coommassie Blue. In the left-hand lane are protein markers; the lanes to the right are dif- ferent concentrations of chAPE1 purified from sf9 cells by running Ni-NTA and Sepherose 75 columns. M. Borjigin et al. Phosphorylation controls chAPE1 activity FEBS Journal 277 (2010) 4732–4740 ª 2010 The Authors Journal compilation ª 2010 FEBS 4733 Different endonuclease activity levels at different time points of the expression We carried out the endonuclease activity screen for chAPE1 from different time points (24, 48 and 72 h) of three different batches of sf9 cell culture. Because of the convenience and reliability of the kinetic parameter for either the first or pseudo-first order reaction scheme, we measured K obs of the catalytic activity, using 100 nm abasic DNA and 5 nm enzyme. The density counts of the product and the substrate at each time point were quantified from the gel auto- radiograph using the Phosphoimager software, quan- tity one. A typical gel image of the chAPE1 endonuclease catalysis is shown in Fig. 3A. The activ- ity level also peaked at the 24 h postinfection expres- sion time point, with a 1.8-fold higher activity than at the 72 h time point and 1.4-fold higher than at 48 h (Fig. 3B). Effects of phosphorylation on endonuclease activity of chAPE1 We initiated the investigation of the phosphorylation effects by measuring the phosphate amount of chAPE1 from the nine samples studied above. The estimated number of phosphorylated residues in the chAPE1 sample was 9.6 (at 24 h), 15.4 (at 48 h) and 18.0 (at 72 h) (Table 1). The phosphorylation level correlated quite well with the endonuclease activity level mea- sured above. The higher the level of phosphorylation of chAPE1 the lower the endonuclease activity, and conversely the lower the phosphorylation the higher the activity level. In order to validate this statement, we phosphorylated the batch of chAPE1 (24 h time point sample with the highest activity), measured its activity level and dephosphorylated the same sample of chAPE1 to reverse the phosphorylation effect. The chAPE1 harvested at the 24 h time point was phosphorylated with either CK I or CK II and dephosphorylated with lambda phosphatase or alkaline phosphatase to measure its endonuclease activity. CK II decreased the rate constant ( K obs ) of chAPE1 by 6.2-fold and lambda phosphatase elevated the activity by 2.1-fold, whereas CK I and alkaline phosphatase did not affect the activity level (Fig. 4). When the A B Fig. 3. APE activity of chAPE1 and the initial screen at several time points. (A) APE catalytic activity was shown at the designated time points of the reaction. The top bands are the substrate and the bottom bands are the product. (B) The batches at the 24 h time point had an activity level 1.4-fold higher than that at 48 h and 1.8-fold higher than that at 72 h. Table 1. The phosphorylation state of the recombinant chAPE1 at three different time points of expression. The recombinant chAPE1 has 44 potential phosphorylation sites (serine, threonine, tyrosine). The protein contains 11 sulfur atoms (in methionine, cystine), and its molecular mass is 35.5 kDa. The mole number of chAPE1 was calculated using mass divided by the molecular mass. The mole number of sulfur in the sample was also calculated and was used as the reference (or standard). The number of phosphorus atoms in a chAPE1molecule was calculated by dividing the mole number of phosphorus by the mole number of the protein. The R 2 values in the linear regression of the standard curves were higher than 0.9998. Time Concentration of chAPE1 lgÆmL )1 Phosphorus measured lgÆmL )1 Sulfur measured lgÆmL )1 Number of phosphorus atoms in a chAPE1 molecule 24 h Sample 1 252.6 2.14 2.53 9.6 ± 0.2Sample 2 2.07 2.46 Sample 3 2.16 2.55 48 h Sample 1 250.4 3.47 2.63 15.4 ± 0.4Sample 2 3.29 2.45 Sample 3 3.37 2.59 72 h Sample 1 249.1 3.86 2.42 17.97 ± 0.2Sample 2 3.91 2.47 Sample 3 3.96 2.52 Phosphorylation controls chAPE1 activity M. Borjigin et al. 4734 FEBS Journal 277 (2010) 4732–4740 ª 2010 The Authors Journal compilation ª 2010 FEBS chAPE1 phosphorylated by either CK I or CK II was dephosphorylated with lambda phosphatase, the endo- nuclease activity was restored to the highest level (Fig. 4). Steady-state kinetic studies of chAPE1 endonuclease activity We carried out the endonuclease kinetic analysis on phosphorylated or dephosphorylated chAPE1, using a Michaelis–Menten model. The dephosphorylation of chAPE1 by lambda phosphatase increased the endonu- clease activity (k cat ⁄ K m ) by 16.7-fold relative to the activity of chAPE1 phosphorylated by CK II. The kinetic parameters for chAPE1 phosphorylated by CK II were k cat = 0.58 ± 0.02 s )1 , K m = 81 ± 10.59 nm and k cat ⁄ K m = 7.2 · 10 )3 nm )1 Æs )1 . The parameter values for chAPE1 dephosphorylated by lambda phosphatase were k cat = 5.67 ± 0.21 s )1 , K m =48± 6.98 nm and k cat ⁄ K m = 0.12 nm )1 Æs )1 (Fig. 5). These results, along with the K obs values from the initial screen and ICP data, show that phosphorylation regulates chAPE1 endonuclease activity in vitro and that chAPE1 expressed in sf9 cell has a different level of phosphorylation at different time points of expression. In the same manner as the initial endonuclease assay, exonuclease activity was tested for untreated chAPE1, phosphorylated and dephosphorylated chA- PE1 with a much higher enzyme concentration (100 nm), as described in the experimental proce- dures section. No detectable exonuclease activity was observed up to 720 s (Fig. 6) in any of these conditions. Discussion Here we studied the APE from the Chinese hamster ovary cell, an important model for DNA repair. The Chinese hamster and its cell lines have been the para- digm for DNA repair research at cellular and gene reg- ulation levels for several decades [30–36] and its APE (chAPE1) gene was cloned a few years ago [37]. We cloned the cDNA of chAPE1, expressed it in insect cell line sf9 and examined the activity of the enzyme in vi- tro. We investigated the levels of the two main cata- lytic activities, 3¢–5¢ exonuclease activity and the Fig. 4. Phosphorylation effects on the APE activity of chAPE1. The rate constant K obs was calculated using the equation ln([S t ] ⁄ [S o ]) = )K obs *t for first or pseudo-first order reactions. 24 h: chAPE1 expressed at the 24 h time point. CK I, CK II, LP and CIP: chAPE1 activity treated with CK I, II, lambda phosphatase and alka- line phosphatase, respectively. CK I + LP, CK II + LP: chAPE1 trea- ted with CK I or CK II first and then dephosphorylated with lambda phosphatase before the activity assay. CK II decreased the chAPE1 activity level by 6.2-fold and lambda phosphatase increased the activity by 2.1-fold, and also restored the activity level of CK II-inhib- ited chAPE1. AB Fig. 5. Endonuclease activity of chAPE1. (A) Michaelis–Menten analysis of the activity of chAPE1 phosphorylated by CK II with k cat ⁄ K m = 7.2 · 10 )3 nM )1 Æs )1 . (B) Michaelis–Menten analysis of the activity of chAPE1 dephosphorylated by lambda phosphatase previously phosphorylated by CK II, k cat ⁄ K m = 0.12 nM )1 Æs )1 . Fig. 6. Exonuclease activity of chAPE1 and exonuclease III (autora- diograph of the gel image). 100 n M of double-stranded DNA oligo substrate (5¢-GTCACCGTCATACGACTC-3¢, complementary strand was not shown, and both strands were labeled with P 33 isotope), 100 n M chAPE1 and 10 nM Escherichia coli exonuclease III as a control were used in this particular assay. M. Borjigin et al. Phosphorylation controls chAPE1 activity FEBS Journal 277 (2010) 4732–4740 ª 2010 The Authors Journal compilation ª 2010 FEBS 4735 endonuclease activity. We did not detect exonuclease activity under our experimental conditions. Endonucle- ase activity varied from batch to batch, but remained within the broad range obtained for other mammalian APE1 enzymes and especially human APE1 [13–18]. However, activity varied with the time of the culture. In order to resolve this variation, we investigated our hypothesis that the activity of the enzyme can be controlled by phosphorylation. Indeed, we detected such a control and were able to narrow the range of possible phosphatases and kinases that can potentially control it. The enzyme kinetic assays performed on chAPE1 phosphorylated by CK II and dephosphorylated by lambda phosphatase using a Michaelis–Menten model revealed the kinetic parameter values of chAPE1. The phosphorylation efficiencies of both CK I and CK II on chAPE1 were fully comparable (data not shown) with that previously reported [38]. We observed that CK I had no effect on chAPE1 endonuclease activity, which is consistent with the findings of Yacoub et al. [38]. However, the level of inhibition of chAPE1 activ- ity by CK II is not as complete as that of the human APE1 activity reported in [38]. The difference might be attributed to the species or expression host difference. Alkaline phosphatase did not alter the activity level of the phosphorylated chAPE1, whereas lambda phos- phatase increased the catalytic activity of chAPE1. These data imply that CK I and CK II target different amino acids in chAPE1 and lambda phosphatase may work on a common set of amino acid residues with CK II. In particular, we were interested in the similarities and differences to the human APE1, for which signifi- cantly divergent activities have been reported. The evi- dence accumulated over many years shows that phosphorylation regulates the human APE1 endonu- clease and redox activities in vitro [38–40]. Several groups have investigated the effects of phosphoryla- tion on the endonuclease and redox activities of human APE1. Although the results of their reports were somewhat conflicting, the consensus conclusion was that phosphorylation regulates the two important functions of human APE1 [38–40], which might switch on and off different functions at different physiologi- cal conditions. Although the phosphorylation of human APE1 by CK II was reported to have no effect [39] to complete inhibition [38] on its APE activity, redox activity was enhanced by CK II treat- ment [39]. In addition to the in vitro evidence of the regulation of human APE1 endonuclease activity by kinases, a very recent report showed that human APE1 is phosphorylated by another regulatory pro- tein, cdk5, which reduces the endonuclease activity of APE1. This in vivo experiment carried out on mice demonstrated that phosphorylation of APE1 at Thr 232 reduces its APE activity, resulting in an accumula- tion of DNA damage and contributing to neuronal death [12]. The increased phosphorylation of APE1 was also observed in post-mortem brain tissue from patients with Parkinson’s disease and Alzheimer’s dis- ease, suggesting a potential link between APE1 phos- phorylation and the pathogenesis of neurodegenerative diseases [12]. In our study of the phosphorylation of chAPE1, we determined quantitatively the extent of phosphoryla- tion by using a novel and accurate analytical tool. We were fortunate to have access to state of the art equip- ment, an ICP spectrometer, an instrument with pico to nanomolar sensitivity in measuring trace elements from aqueous solutions. We were successful in measuring the absolute amount of phosphorus and sulfur from micromolar protein samples with significant accuracy. Similar measurements were carried out with this instrument to determine the same elements in vegetable oils and beef [41] and other trace elements bound to proteins [42]. This technique offers significant advanta- ges over traditionally used methods. There are several more traditional methods of direct measurement of phosphorylation. The most popular involves the incubation of whole cells with radiola- beled 32 P-orthophosphate, the generation of cellular extracts, separation of proteins by SDS ⁄ PAGE and exposure to film for phosphoimaging [43]. A clear drawback of this method is labor expense and the use of radioactive isotopes and the difficulty of eliminat- ing the background presence of a natural phosphate source in the culture medium. Another specific tech- nique is the use of phosphate-specific antibodies. This technique can be used for an immunoassay to deter- mine the phosphorylation amount [44]. The main caveat in successfully utilizing a phosphor-specific antibody technique is the specificity and affinity of the antibody for the phosphoprotein of interest. The most accurate and powerful technique for determining and sequencing the phosphoproteins is MS [45]. However, there are also several difficulties with the analysis of phosphoproteins by this technique. First, signals from phosphopeptides are generally weaker, as they are negatively charged and poorly ionized by electrospray MS (it is performed in the positive mode). Second, it can be difficult to observe the signals from low-abun- dance phosphoproteins of interest in the high back- ground of abundant nonphosphorylated proteins [45]. One of the more innovative uses of this technique was developed by McKenzie & Strauss [46]. However, this Phosphorylation controls chAPE1 activity M. Borjigin et al. 4736 FEBS Journal 277 (2010) 4732–4740 ª 2010 The Authors Journal compilation ª 2010 FEBS technique was used to measure the phophorylation efficiency of kinases with a radioactive phosphate compound. In summary, the level of phosphorylation of chAPE1 in sf9 cells varied at different time points of expression, which correlated well with its endonuclease activity. The observation was confirmed by a kinetic assay of the phosphorylated and dephosphorylated chAPE1. The in vitro catalytic activity tests also dem- onstrated that different regulatory proteins (kinases and phosphatases) have different effects on chAPE1. These results suggest that the different functions of the multifunctional chAPE1 are switched on and off by regulatory proteins at different stages of the cell life, which might also provide a plausible explanation for the reported discrepancies in endonuclease activity level of the human APE1 in the literature. Our final finding suggests that chAPE1 may not have exonucle- ase activity regardless of its phosphorylation state. This implies that further studies into exonuclease activ- ity of human APE1 in the phosphorylated and ⁄ or dephosphorylated state are warranted. This important regulatory effect of phosphorylation has not been explored exhaustively in other mammalian APE1. More studies in this area with human enzyme are also warranted. Experimental procedures Overexpression of chAPE1 in the sf9 cell line using the baculovirus system The cDNA developed in our laboratory was subcloned into a baculovirus transfer vector (pBlueBacHis2B; Invitrogen, Carlsbad, CA, USA) using PCR amplification with a pair of designed primers. The sequences of the primers were 5¢-GAAGATCTAAGCGTGGGAAGAGAGCG-3¢ and 5¢-GGGGTACCAGGTGTAAGTTACTTCAGCAG-3¢ (MWG Biotech, Ebersberg, Germany). The insect cell line sf9 was cotransfected with the pBlueBacHis2B construct and Bac-N-Blue AcMNPV viral DNA (Invitrogen), fol- lowed by an agarose overlay plaque assay to select the recombinant virus. High titer virus stocks (up to 1.3 · 10 8 plaque forming unitsÆmL )1 ) were generated in a suspension sf9 cell culture. Large-scale protein expression was carried out in 500 mL suspension culture with viral stock at a mul- tiplicity of infection of eight. The cells were harvested and lysed by sonication followed by centrifugation. The supernatant was applied to Ni-NTA affinity and S75 Sepherose size exclusion columns to purify the protein under native conditions. The N-terminal ( 4 kDa) HisTag linker was cleaved with enterokinase (EKmax; Invitrogen) and the native chAPE1 was isolated from the Tag and enterokinase using a nickel affinity column and an EKaway resin column. Protein purity was tested using SDS ⁄ PAGE followed by Coommassie blue staining; the concentration was determined using the Brad- ford protein assay (Bio-Rad, Hercules, CA, USA). Initial endonuclease activity screen Fifty milliliters of sf9 cell culture were removed from 500 mL suspension culture at 24, 48 and 72 h postinfection with the recombinant baculovirus. chAPE1 was purified and quantified from the 50 mL cell culture aliquots, using the same procedure as described above. The enzyme at three time points of expression from three different batches of culture was prepared. An abasic DNA substrate was made by annealing a 5¢ P 33 -labeled, tetrahydrofuran-containing oligonucleotide (5¢-GTCACCGTC FTACGACTC-3¢) with its complementary oligonucleotide (MWG Biotech) in 50 mm HEPES buffer, pH 7.5, 50 mm KCl, 0.1 m m EDTA by heating in a 95 °C water bath and cooling down at room temperature within 2 h. The endonuclease reaction was carried out in a total vol- ume of 30 lL containing 100 nm DNA substrate and 5 nm chAPE1 in 50 mm HEPES ⁄ KOH (pH 7.5), 50 mm KCl, 0.1 mm EDTA and 5 mm MgCl 2 at room temperature [26]. Aliquots of 3 lL of reaction mix were transferred into 3 lL stop solution containing 85 mm EDTA at designated time points to quench the reaction. Subsequently, the reaction products were resolved in a DNA sequencing gel (15% polyacrylamide gel containing 8 m urea). The gels were dried, exposed to K screen (BioRad) and the image and the band intensities measured and quantified with a BioRad PhosphoImager FX and quantity one software. All enzyme kinetic experiments were repeated in triplicate and K obs was calculated using the formula ln([S t ] ⁄ [S o ]) = )K obs *t, which applies to first or pseudo-first order reac- tions. Here, [S o ]=[S t ]+[P t ]; [S t ] is the uncleaved (or intact) substrate concentration, [P t ] is the product concen- tration at time point t,[S o ] is the initial (or total) substrate concentration. The quantitation proceeded through measur- ing the intensity of both bands and normalizing them by adding both bands and taking the fractional intensity belonging to the particular (substrate, product) band. Such measured intensity of bands was used to construct K obs plots and obtaining full Michaelis–Menten substrate satura- tion curves. Steady-state kinetic studies on the endonuclease activity of chAPE1 and Michaelis–Menten analysis In order to determine the kinetic parameters of the endonu- clease activity of chAPE1, 5 nm chAPE1 was mixed with various concentrations (10–400 nm) of abasic DNA M. Borjigin et al. Phosphorylation controls chAPE1 activity FEBS Journal 277 (2010) 4732–4740 ª 2010 The Authors Journal compilation ª 2010 FEBS 4737 substrate in 30 lL of reaction under the standard condi- tion, and the subsequent procedures of the assay were the same as described above. The Michaelis–Menten analysis was carried out with the SigmaPlot enzyme kinetic module to calculate the parameter values. ICP analysis to measure phosphate amount Because the sensitivity of ICP for sulfur and phosphorus is in the subnanomolar scale, it is quite appropriate to measure accurately the amounts of these elements in a protein sample of micromolar concentration. An Optima 4300 DV ICP spectrometer (Perkin Elmer, Boston, MA, USA) was used with the following parameters: nebulizer backpressure 258.0 kPa, nebulizer flow 0.80 LÆmin )1 , wave- lengths for P 213.617 nm, S 180.669 nm. The nine protein preparations from three time points of three different batches of chAPE1 were dialyzed in 50 mm Tris ⁄ HCl, pH 8.0, and their concentration adjusted to 250 lgÆ mL )1 in 3 mL volume. The standard curve was constructed with solutions of known concentration of phosphorus and sulfur and other elements for references, in parts per billion concentration. The absolute concentration of sulfur and phosphorus was determined based on the conversion of the spectrum intensity to a concentration value. The phosphorus in the protein samples was accurately calculated using the ratio of the molarity of the protein and the measured molarity of phosphorus, and sulfur was used as the reference control for the calculation. Each sample was measured in triplicate and the error was calculated using Microsoft Excel. Phosphorylation of chAPE1 with CK I or CK II Both kinases and buffers were obtained from New England Biolabs (Beverly, MA, USA). Five pmol of chAPE1 was phosphorylated in a 50 lL reaction volume using 5 units of CK I or CK II. The CK II reaction condition was 0.1 mm ATP, 0.6 lCiÆlL )1 [c )33 P]ATP, 20 mm Tris ⁄ HCl (pH 7.5), 50 mm KCl and 10 mm MgCl 2 . The CK I reaction condition was 0.1 m m ATP, 0.6 lCiÆlL )1 [c )33 P]ATP, 50 mm Tris ⁄ HCl (pH 7.5), 5 mm dithiothreitol and 10 mm MgCl 2 . The reaction mix was incubated at 30 °C for 45 min, and the efficiency of phosphorylation was assessed in SDS ⁄ PAGE followed by imaging and quantification by means of BioRad PhosphoImager FX and quantity one software. Dephosphorylation of chAPE1 with alkaline phosphatase or lambda phosphatase chAPE1 (2 pmol) was dephosphorylated using alkaline phosphatase (0.5 unit) or lambda phosphatase (5 units) in 20 lL reaction mix at 30 °C for 45 min. The alkaline phos- phatase reaction was carried out in 100 mm NaCl, 50 mm Tris ⁄ HCl (pH 7.9), 10 mm MgCl 2 and 1 mm dithiothreitol. The conditions for lambda phosphatase were 50 mm Tris ⁄ HCl (pH 7.5), 0.1 mm Na 2 EDTA, 5 mm dithiothreitol, 0.01% Brij 35 surfactant and 2 mm MnCl 2 . Exonuclease assay The standard exonuclease assay was conducted using the same procedures as for the APE activity, except for the reg- ular double-stranded DNA oligo substrate, of which the 5¢ of each strand was labeled with the P 33 isotope. The concentration of the enzymes (100 nm chAPE1 and 10 nm Escherichia coli exonuclease III) and 10–400 nm DNA substrate were used. Here, Escherichia coli exonuclease III (New England Biolabs) was used as a positive control. Acknowledgements We thank Dr Siddhartha Das for helpful discussion on the subjects and thank Drs Jorge Gardea-Torresdey and Jose Peralta for their help in performing experi- ments using the ICP and providing necessary reagents. This work was partially supported by NIH grants GM08012, RR008124 and the NSF grant HRD9701775 to Dr P. Arenaz. References 1 Evans AR, Limp-Foster M & Kelley MR (2000) Going APE over ref-1. Mutat Res 461, 83–108. 2 Manoharan M, Mazamder A, Ransom SC, Gerlt JA & Bolton PH (1988) Mechanism of UV endonuclease V cleavage of abasic sites determined by 13C labeling. J Am Chem Soc 110, 2690–2691. 3 Taylor AF & Weiss B (1982) Role of exonuclease III in the base excision repair of uracil-containing DNA. J Bacteriol 151, 351–357. 4 Schaaper RM & Leob LA (1981) Depurination causes mutations in SOS-induced cells. Proc Natl Acad Sci USA 78, 1773–1777. 5 Shearman CW & Leob LA (1977) Depurination decreases fidelity of DNA synthesis in vitro. Nature 270, 537–538. 6 Wilson DM III & Thompson LH (1997) Life without DNA repair. Proc Natl Acad Sci USA 94, 12754–12757. 7 Lindahl T, Karrans P & Wood R (1997) DNA excision repair pathways. Curr Opin Genet Dev 7 , 158–169. 8 Richardson C & Kornberg A (1964) A deoxyribonucleic acid phosphatase-exonuclease from Escherichia coli. I. Purification of the enzyme and characterization of the phosphatase activity. J Biol Chem 239, 242–250. 9 Richardson C, Lehman I & Kornberg A (1964) A deoxyribonucleic acid phosphatase-exonuclease from Phosphorylation controls chAPE1 activity M. Borjigin et al. 4738 FEBS Journal 277 (2010) 4732–4740 ª 2010 The Authors Journal compilation ª 2010 FEBS Escherichia coli. II. Characterization of the exonuclease activity. J Biol Chem 239, 251–258. 10 Demple B, Johnson A & Fung D (1986) Exonuclease III and endonuclease IV remove 3¢ blocks from DNA synthesis primers in H 2 O 2 -damaged Escherichia coli. Proc Natl Acad Sci USA 83, 7731–7735. 11 Bernelot-Moens C & Demple B (1989) Multiple DNA repair activities for 3¢-deoxyribose fragments in Escheri- chia coli. Nucleic Acids Res 17, 587–600. 12 Huang E, Qu D, Zhang Y, Venderova K, Haque ME, Rousseaux MW, Slack RS, Woulfe JM & Park DS (2010) The role of Cdk5-mediated apurinic ⁄ apyrimidinic endonuclease 1 phosphorylation in neuronal death. Nat Cell Biol 12, 563–571. 13 Wilson DM III, Takeshita M, Grollman AP & Demple B (1995) Incision activity of human apurinic endonucle- ase (Ape) at abasic site analogs in DNA. J Biol Chem 270, 16002–16007. 14 Mol CD, Izumi T, Mitra S & Tainer JA (2000) DNA- bound structures and mutants reveal abasic DNA bind- ing by APE1 DNA repair and coordination. Nature 403, 451–456. 15 Mckenzie JA & Strauss PR (2001) Oligonucleotides with bistranded abasic sites interfere with substrate binding and catalysis by human apurinic ⁄ apyrimidinic endonuclease. Biochemistry 40, 13254–13261. 16 Strauss PR, Beard WA, Patterson TA & Wilson SH (1997) Substrate binding by human apurinic ⁄ apyrimidi- nic endonuclease indicates a Briggs-Haldane mecha- nism. J Biol Chem 272, 1302–1307. 17 Lucas JA, Masuda Y, Bennett RAO, Strauss NS & Strauss PR (1999) Single-turnover analysis of mutant human apurinic ⁄ apyrimidinic endonuclease. Biochemis- try 38, 4958–4964. 18 Izumi T, Schein CH, Oezguen N, Feng Y & Braun W (2004) Effects of backbone contacts 3¢ to the abasic site on the cleavage and the product binding by human apu- rinic ⁄ apyrimidinic endonuclease (APE1). Biochemistry 43, 684–689. 19 Yoshita A & Ueda T (2003) Human AP endonuclease possesses a significant activity as major 3¢–5¢ exonucle- ase in human leukemia cells. Biochem Biophys Res Com- mun 310, 522–528. 20 Chou K & Cheng Y (2002) An exonucleolytic activity of human apurinic ⁄ apyrimidinic endonuclease on 3¢ mispaired DNA. Nature 415, 655–659. 21 Barzilay G, Walker LJ, Robson CN & Hickson ID (1995) Site-directed mutagenesis of the human DNA repair enzyme HAP1: identification of residues impor- tant for AP endonuclease and RNase H activity. Nucleic Acids Res 23, 1544–1550. 22 Suh D, Wilson DM III & Povirk LF (1997) 3¢-phospho- diesterase activity of human apurinic ⁄ apyrimidinic endonuclease at DNA double-strand break ends. Nucleic Acids Res 25, 2495–2500. 23 Demple B, Herman T & Chen DS (1991) Cloning and expression of APE, the cDNA encoding the major human apurinic endonuclease: definition of a family of DNA repair enzymes. Proc Natl Acad Sci USA 88, 11450–11454. 24 Robson CN & Hickson ID (1991) Isolation of cDNA clones encoding a human apurinic ⁄ apyrimidinic endo- nuclease that corrects DNA repair and mutagenesis defects in E. coli xth (Exonuclease III) mutants. Nucleic Acids Res 19, 5519–5523. 25 Lebedeva NA, Khodyreva SN, Favre A & Lavrik OI (2003) AP endonuclease 1 has no biologically significant 3¢-5 exonuclease activity. Biochem Biophys Res Commun 300, 182–187. 26 Seki S, Ikeda S, Watanabe S, Hatsushika M, Tsutsui K, Akiyama K & Zhang B (1991) A mouse DNA repair enzyme (APEX nuclease) having exonuclease and apuri- nic ⁄ apyrimidinic endonuclease activities: purification and characterization. Biochim Biophys Acta 1079, 57– 64. 27 Seki S, Akiyama K, Watanabe S, Hatsushika M, Ikeda S & Tsutsui K (1991) cDNA and deduced amino acid sequence of a mouse DNA repair enzyme (APEX nuclease) with significant homology to Escherichia coli exonuclease III. J Biol Chem 266, 20797–20802. 28 Robson CN, Milne AM, Pappin DJC & Hickson ID (1991) Isolation of cDNA clones encoding an enzyme from bovine cells that repair oxidative DNA damage in vitro: homology with bacterial repair enzyme. Nucleic Acids Res 19, 1087–1092. 29 Huq I, Wilson TM, Kelley MR & Deutsch WA (1995) Expression in Escherichia coli of a rat cDNA encoding an apurinic ⁄ apyrimidinic endonuclease. Mutat Res 337, 191–199. 30 Berquist BR, Singh DK, Fan J, Kim D, Gillenwater E, Kulkarni A, Bohr VA, Ackerman EJ, Tomkinson AE & Wilson DM III (2010) Functional capacity of XRCC1 protein variants identified in DNA repair-defi- cient Chinese hamster ovary cell lines and the human population. Nucleic Acids Res 38, 5023–5035. 31 Pan Y, Yuan D, Zhang J, Xu P, Chen H & Shao C (2009) Cadmium-induced adaptive response in cells of Chinese hamster ovary cell lines with varying DNA repair capacity. Radiat Res 171, 446–453. 32 Biversta ˚ l A, Johansson F, Jenssen D & Erixon K (2008) Cyclobutane pyrimidine dimers do not fully explain the mutagenicity induced by UVA in Chinese hamster cells. Mutat Res 648, 32–39. 33 Taccioli GE, Cheng H-L, Varghese AJ, Whitmore G & Alt FW (1994) A DNA repair defect in Chinese hamster ovary cells affects V(D)J recombination similarly to the murine mid mutation. J Biol Chem 269, 7439–7442. 34 Boiteux S & le Page F (2001) Repair of 8-oxoguanine and Ogg1-incised apurinic sites in a CHO cell line. Prog Nucleic Acid Res Mol Biol 68, 95–105. M. Borjigin et al. Phosphorylation controls chAPE1 activity FEBS Journal 277 (2010) 4732–4740 ª 2010 The Authors Journal compilation ª 2010 FEBS 4739 35 Gille JJ, van Berkel CG & Joenje H (1993) Mechanism of hyperoxia-induced chromosomal breakage in Chinese hamster cells. Environ Mol Mutagen 22, 264–270. 36 Lehmann AR (1985) Use of recombinant DNA tech- niques in cloning DNA repair genes and in the study of mutagenesis in mammalian cells. Mutat Res 150 , 61–67. 37 Purohit S & Arenaz P (1999) Molecular cloning, sequence and structure analysis of hamster apuri- nic ⁄ apyrimidinic endonuclease (chApel) gene. Mutat Res 435, 215–224. 38 Yacoub A, Kelley M & Deutsch WA (1997) The DNA repair activity of human redox ⁄ repair APE ⁄ Ref-1 is inactivated by phosphorylation. Cancer Res 57, 5457– 5459. 39 Fritz GB & Kina B (1999) Phosphorylation of the DNA repair Ape1 ⁄ ref1 by CK II affects redox regula- tion of AP-1. Oncogene 18, 1033–1040. 40 Hsieh MM, Hegde V, Kelley MR & Deutsch WA (2001) Activation of APE ⁄ Ref-1 redox activity is medi- ated by reactive oxygen species and PKC phosphoryla- tion. Nucleic Acids Res 29, 3116–3122. 41 Ostrovsky M, Loukianova T, Hilligoss D & Wee P. Sulphur and Phosphorus Analysis in Vegetable Oil and Beef Tallow for Biodiesel Production Using the Optima Inductively Coupled Plasma-Optical Emission Spectrome- ter, http://las.perkinelmer.com/FocusAreas/Mkt+Sus- tainable+Energy/Biofuels_Application_Notes.htm. 42 Ma R, McLeod CW, Tomlinson K & Poole RK (2004) Speciation of protein-bound trace elements by gel elec- trophoresis and atomic spectrometry. Electrophoresis 25, 2469–2477. 43 de Graauw M, Hensbergen P & van de Water B (2006) Phospho-proteomic analysis of cellular signaling. Elec- trophoresis 27, 2676–2686. 44 Czernik AJ, Girault JA, Nairn AC, Chen J, Snyder G, Kebabian J & Greengard P (1991) Production of phos- phorylation state-specific antibodies. Methods Enzymol 201, 264–283. 45 Mann M, Ong SE, Gronborg M, Steen H, Jensen ON & Pandey A (2002) Analysis of protein phosphorylation using mass spectrometry: deciphering the phosphoprote- ome. Trends Biotechnol 20, 261–268. 46 Mckenzie JA & Strauss PR (2003) A quantitative method for measuring protein phosphorylation. Anal Biochem 313, 9–16. Phosphorylation controls chAPE1 activity M. Borjigin et al. 4740 FEBS Journal 277 (2010) 4732–4740 ª 2010 The Authors Journal compilation ª 2010 FEBS . Chinese hamster apurinic ⁄ apyrimidinic endonuclease (chAPE1) expressed in sf9 cells reveals that its endonuclease activity is regulated by phosphorylation Mandula. apurinic ⁄ apyrimidinic endonuclease; ChAPE1, Chinese hamster apurinic ⁄ apyrimidinic endonuclease; CK I, casein kinase I; CK II, casein kinase II; ICP, inductivity