HsRec2/Rad51L1, a Protein Influencing Cell Cycle Progression, Has Protein Kinase Activity

Experimental Cell Research 254, 33– 44 (2000) doi:10.1006/excr.1999.4725, available online at http://www.idealibrary.com on

HsRec2/Rad51L1, a Protein Influencing Cell Cycle Progression, Has Protein Kinase Activity Pamela A. Havre, Michael Rice, Ronald Ramos, and Eric B. Kmiec 1 Department of Biological Sciences, University of Delaware, Newark, Delaware 19716

protein is induced by ionizing radiation, a feature of genes involved in recombinational repair, and causes a delay in G1 upon overexpression [7]. Several other groups have cloned the same gene naming it RAD51B [8] or R51H2 [9], respectively. The hREC2/RAD51L1 gene bears striking homology to the RAD51 family of genes, particularly within the so-called “A and B boxes” identified in RECA orthologues. Human Rad51 protein has been shown to catalyze strand transfer reactions in vitro, a hallmark reaction for RecA-like proteins [10]. In preliminary studies hsRec2/Rad51L1 protein was not found to promote similar reactions (H. Gamper, personal communication). The homozygous deletion of the RAD51 gene leads to embryonic lethality suggesting a more widespread role for this gene in DNA metabolism [11]. Since hsRec2/Rad51L1 appears not to catalyze in vitro DNA recombination reactions, yet is inducible by UV light and ionizing radiation, it is likely that the protein plays a more general role in a DNA damage response. There are a number of proteins that play a role in the progression from G1 to S phase, such as cyclins D, E, and A; cdk2; cdk4; cdk6; CAK [12]; p53; indirectly through its transactivation of p21 [13]; and other cdk inhibitors such as p16, p27, and p57 [12]. Hence, these proteins are likely to be part of a pathway responding to DNA damage through regulation of the cell cycle. Since the expression of hREC2/RAD51L1 is inducible by radiation [1] and its promoter contains sequence elements known to be in the regulatory regions of genes responding to DNA damage and cell proliferation [14], we searched for activities of the hsRec2/Rad51L1 protein that might correlate with the observed delay in G1 [7]. We report here that hsRec2/Rad51L1 exhibits protein kinase activity toward myelin basic protein and kemptide, a heptapeptide containing one phosphorylatable residue, a serine. In addition, it can phosphorylate GST– cyclin E, cdk2, and p53 which, by coimmunoprecipitation data, appears to be associated with hsRec2/Rad51L1. The hsRec2/Rad51L1 protein also contains limited homology to PKA catalytic domains I and XI. To our knowledge, this is the first report of protein kinase activity by a RAD51 analog.

Human Rec2/Rad51L1 is a member of the Rad51 family of proteins. Although recombinase activity, typical of this family, could not be established, its overexpression in mammalian cells has been shown to cause a delay in G1. Moreover, since hsRec2/Rad51L1 has been found to be induced by both ionizing and UV irradiation, it is likely that hsRec2/Rad51L1 is elevated following any DNA damage and causes a G1 delay to allow time for DNA repair to occur. Limited homology with catalytic domains X and XI of protein kinase A suggested that kemptide, an artificial substrate containing one phosphorylatable residue, a serine, might serve as a substrate for hsRec2/Rad51L1. Here, we report that hsRec2/Rad51L1 can phosphorylate kemptide, as well as myelin basic protein, p53, cyclin E, and cdk2, but not a peptide substrate containing tyrosine only. The finding that hsRec2/Rad51L1 exhibits protein kinase activity is a first step toward identifying a mechanism whereby this protein affects the cell cycle. © 2000 Academic Press

Key Words: cell cycle; G1 delay; protein kinase; p53; cyclins.

INTRODUCTION

As part of a cellular response to DNA damage, DNA synthesis must be suspended temporarily in order to prevent the replication of errors and the inheritance of mutations. The delay in DNA synthesis and DNA repair events is linked through the activity of certain regulatory proteins or factors. The identity of all of the proteins controlling these processes, however, has not been firmly established. Our laboratory recently isolated, cloned, and characterized a mammalian gene [1] based on sequence homology to the Ustilago maydis REC2 gene [2–5]. The fungal REC2 gene encodes a protein, whose expression is induced by radiation and regulated by cell cycle; controls progression through meiosis; and is involved in recombinational repair [6]. The hsRec2/Rad51L1 1 To whom correspondence and reprint requests should be addressed. Fax: (302) 831-8786. E-mail: [email protected].

33

0014-4827/00 $35.00 Copyright © 2000 by Academic Press All rights of reproduction in any form reserved.

34

HAVRE ET AL.

METHODS Isolation and purification of His-hsRec2/Rad51L1 in baculovirus and bacterial systems. The full-length hREC2/RAD51L1 cDNA was cloned into the expression vector, pAcHisA, for overexpression in a baculovirus system and purification utilizing a 6-histidine tag. For cloning, the hREC2/RAD51L1 expression cassette was cut with KpnI, the 39 protruding termini were removed with T4 polymerase, and the DNA was then digested with XbaI. The resulting fragment was ligated to pAcHisA using the SmaI and XbaI sites. Recombinant virus containing hREC2/RAD51L1 was purified and insect cells were infected by Dr. Z. Yu in the Baculovirus Expression Laboratory of the Kimmel Cancer Institute. Insect cell pellets from 2 liters of culture were suspended in 60 ml of 10 mM Tris–HCl, pH 7.5, 130 mM NaCl, 2% TX100, 2 mg/ml leupeptin and aprotinin and 1 mg/ml pepstatin, and sonicated on ice four times for 5 s each using a microtip at a 20% pulse (Branson sonifier 450). Debris was removed by centrifuging at 30,000g for 20 min. The clarified supernatant was divided between two 50-ml culture tubes and 1 ml of Ni–NTA agarose added to each tube for 1 h with rocking at 4°C. The unbound fraction was separated from the resin by a brief centrifugation and the resin was washed with 10 ml of 100 mM imidazole for 10 min on a rocker and centrifuged at 2000 rpm for 5 min. After a second 10-min wash with 500 mM imidazole, the slurry was transferred to a column and the effluent was discarded. The purified His-hsRec2/Rad51L1 was eluted with 1 M imidazole, pH 7.0 (the imidazole was allowed to remain on the column for 10 min before collection of eluate). The samples were then dialyzed overnight against 50 mM Tris–HCl, pH 7.4, 50 mM NaCl, 10% glycerol. For expression in Escherichia coli, the hREC2/RAD51L1 cDNAcoding region was excised from the mammalian expression vector pcDNA3 G8 by cleavage with XbaI, and the 59 protruding termini were filled in by the action of T4 polymerase, followed by cleavage with KpnI. The resulting fragment was ligated into the KpnI site and blunt-end ligated at the HindIII sites of a bacterial expression vector pBAD/HisC (Invitrogen Corp., San Diego, CA). The expression vector with hREC2/RAD51L1 cloned in frame with a hexahistidine tag was electrotransformed into LMG194 bacteria (Invitrogen Corp.). A 500-ml LB ampicillin culture was inoculated by a single colony and grown at 37°C. The culture was induced in log phase by 0.02% arabinose for 4 h and harvested by centrifugation at 8000g. The pellet was resuspended and lysed in 50 mM Hepes, pH 7.4, 0.1% NP-40 at 0°C. The lysate was clarified by centrifugation at 10,000g for 30 min and then added to a sealed column containing Talon metal affinity resin (Clontech, Palo Alto, CA) and agitated for 1 h. The column was then washed with 20 vol of 50 mM Hepes, pH 7.4, 1 mM PMSF, 0.5 M NaCl. The bound protein was eluted in 3 vol of the above wash buffer with 10 mM EDTA and collected in 1-ml fractions, and the purified His-hsRec2/Rad51L1 was dialyzed overnight against 50 mM Tris–HCl (pH 7.5), 50 mM NaCl, 10% glycerol, and stored at 280°C. No loss of activity has been observed for a period of over 6 months. Site-specific mutations were generated by multiple rounds of PCR with primers bearing the desired mutant base. Amplimers containing the altered sequence were expanded by a second round of PCR. For A box mutations, [K 3 L (114)], amplifications were carried out using primer pairs XhIF-KGR and KGF-HD3R. The plasmid vector containing the cDNA of hREC2/RAD51L1, LCK-hREC2, was used as the template for PCR. Products of XhIF-KGR and KGF-HD3R were mixed and used for the second round of PCR using XhIF-HD3R to generate the full-length open reading frame with the desired mutation. The final PCR product was digested by XhoI and HindIII (cloning sites XhoI and HindIII were incorporated in the primers) and ligated into His tag vector pBADHisC (Invitrogen, Inc.). Mutations in the B box, [D 3 V (210)], were generated using the same method with different primers. Primer pairs XhIF-DVR and DVF-HD3R were

used instead for the first-round PCR. To confirm the intended mutation and exclude other mutations, the final mutants (K 3 G mutant, D 3 V-mutant) were sequenced and analyzed using Sequencher 3.0 software. PCR primers used were XhIF 59 AGC TCG AGA ATG GGT AGC AAG AAA CTA AAA 39, HD3R 59 CCA AGC TTC TAG GAA TTT CCA TAG GCT TG 39, KGF 59 CAC AAC CTG GTG 39 DVF 59 CTT GTG ATT CTT GTC TCT GTT GCT TC 39, and DVR 59 GAA GCA ACA GAG ACA AGA ATC ACA AG 39. The Y 3 A (163) mutant was previously constructed and the protein isolated according to Havre et al. [7]. Protein kinase filter assays. Substrates were either kemptide or myelin basic protein (MBP) at a concentration of 0.2 mM unless otherwise noted, and approximately 1 mg of hsRec2/Rad51L1 was added. For both assays, the buffer contained 50 mM Tris–HCl, pH 7.5, 10 mM MgCl 2, 1 mM DTT. The second substrate, [g- 32P]ATP, was constant at 50 mM with a specific activity of 1972 cpm/pmol (kemptide) and 2980 cpm/pmol (MBP) except when renatured hsRec2/Rad51L1 was used, when specific activity was increased approximately 20-fold. [g- 32P]ATP was added to initiate the reaction, which was carried out at 30°C for the indicated time. At the end of the reaction, 20 ml was spotted on phosphocellulose disks, and the disks were washed twice with 10 ml per disk in 1% phosphoric acid and twice in distilled water. Filters were counted in a Wallac scintillation counter. Substrate without hsRec2/Rad51L1 added was used as a control and counts were subtracted to obtain a zero point. Renaturation of hsRec2/Rad51L1 following SDS gel electrophoresis. Approximately 10 mg hsRec2/Rad51L1 was mixed 1:1 with 23 gel loading buffer and run on a 10% ReadyGel (Bio-Rad Laboratories, Hercules, CA). The position of the hsRec2/Rad51L1 was determined by staining the detached outermost lanes with Coomassie blue. The remainder of the gel was equilibrated in 50 mM Tris–HCl, pH 7.4, 150 mM NaCl for 30 min prior to slicing the gel. Gel slices were shredded in an Eppendorf tube and incubated at least 4 h at room temperature in Rec2 buffer (50 mM Tris–HCl, pH 7.4, 50 mM NaCl, 10% glycerol) supplemented with 1% Triton X-100 to renature the protein. Gel fragments were removed using 0.22-mm filter units (Millipore, Bedford, MA). p53 phosphorylation. Human recombinant p53 (0.5 mg) (PharMingen, San Diego, CA) was incubated with or without hsRec2/ Rad51L1 in 50 mM Tris–HCl, pH 7.4, 10 mM MgCl 2, and 1 mM DTT at 30°C. The reaction was initiated by the addition of [g- 32P]ATP (25 mM ATP, 40 cpm/fmol). At the end of each time point, an equal volume of 23 loading buffer [15] was added and tubes were placed on ice until all tubes were collected. Samples were then heated at 100°C for 10 min and 13 ml were run on ReadyGels (Bio-Rad Laboratories) and transferred to nitrocellulose overnight prior to exposure to XOmat film (Kodak, Rochester, NY). Quantitation of p53 phosphorylation was obtained by an overnight exposure of the blot to a phosphorimager screen and scanning with a PhosphorImager (Molecular Dynamics, Sunnyvale, CA). The activity is expressed as phosphorimager units. Immunoprecipitations. Immunoprecipitation of hsRec2/Rad51L1 was carried out in 50 mM Tris–HCl, pH 7.4, 150 mM NaCl, 0.5% NP-40, 10 mg/ml aprotinin and leupeptin, and 1 mM PMSF. Following a 1-h preclearing step using normal mouse serum and protein A/G–Sepharose, 50 ml of three combined hybridoma supernatants was added followed by protein A/G–Sepharose. Each step was carried out for 1 h and after the final step, proteins bound to the Sepharose were pelleted and the Sepharose was washed three times in binding buffer followed by one wash with resuspension buffer (50 mM Tris–HCl, pH 7.4). For the GST– cyclin E assays with immunoprecipitates, no preclearing step was used, and cell supernatants from individual clones were used instead of combined supernatants from these three clones. Kinase assays were carried out using these immunoprecipitates as described above.

GENOME SURVEILLANCE PROCESSES IN MAMMALIAN CELLS In vitro association between p53 and hsRec2/Rad51L1. HsRec2/ Rad51L1 (5 mg) and 15 ml agarose–GST–p53 (Oncogene Sciences, Cambridge, MA) were added to 0.5 ml of binding buffer (10% glycerol, 50 mM Tris–HCl, pH 7.4, 0.1 mM EDTA, 1 mM DTT, 0.02% NP-40, 200 mM NaCl, 10 mg/ml aprotinin and leupeptin, and 20 mM PMSF). Following 1 h at room temperature, agarose–GST–p53 was pelleted and washed twice with buffer as above, using a higher concentration of detergent (0.1% NP-40), and once with 50 mM Tris–HCl, pH 7.4, 10 mM MgCl 2. cdk2/cyclin E protein kinase assay. GST– cyclin E was isolated from E. coli transformed with pGEX-2TcycE (kindly given to us by A. DeLuca in A. Giordano’s laboratory at Thomas Jefferson University) and purified using glutathione–Sepharose 4B (Pharmacia, Piscataway, NJ). The glutathione–sepharose GST– cyclin E was washed and then stored as a 1:1 slurry in 50 mM Tris–HCl, pH 7.4. For assays with cyclin E-bound cdk2, purified cdk2 (kindly given to us by A. Koff, Sloan Kettering, New York, NY) was incubated with cyclin E as described [16] and unbound cdk2 was removed by washing prior to storage as a 1:1 slurry. Kinase assays were carried out with the immobilized GST– cyclin E with or without bound cdk2 otherwise using the same conditions described for p53. Association of in vitro translated hsRec2/Rad51L1 with PCNA, p53, and cdc2. XbaI linearized pCMVhREC2/RAD51L1 was first transcribed in vitro (Ambion, Austin, TX) using 1 mg of the vector and then translated in vitro along with Xef1 mRNA included in the kit as a positive control. Reticulocyte lysates containing Xef1 or hsRec2/ Rad51L1 translation products labeled with [ 35S] methionine were incubated with 1.2 mg cell extract from HCT116 cells (50 mM Tris– HCl, pH 7.4, 120 mM NaCl, 0.5% NP-40, 20 mM PMSF, 2 mg/ml pepstatin, and 10 mg/ml leupeptin and aprotinin, MB) for 2 h, and then 10 mg of antibodies against PCNA, p53, or cdc2 was added for an overnight incubation. On the following day, protein A–Sepharose was added for 2 h, and pellets were washed four times with 500 ml MB. Pellets were suspended in 40 ml of sample buffer and boiled 10 min, and then 15 ml was run on a 10% gel and transferred to nitrocellulose to obtain a lower background, before exposure to X-Omat film (Kodak).

RESULTS

Purification of hsRec2/Rad51L1 from baculoviral and bacterial expression systems. The full-length hREC2/RAD51L1 cDNA was cloned into the expression vector pAcHisA, for overexpression in a baculovirus system and purification utilizing a 6-histidine tag at its amino terminus. Recombinant virus containing pAcHisA-hREC2/RAD51L1 was then purified and used to infect insect cells. For simplicity, this protein will be referred to as hsRec2/Rad51L1 instead of His-hsRec2/ Rad51L1. This preparation contains one major band when run on an SDS gel and stained with silver (Fig. 1A). Figure 1B illustrates the isolation of hsRec2/ Rad51L1 expressed in E. coli and purified by a histidine affinity column with a single band visible as judged by SDS–PAGE (polyacrylamide gel electrophoresis). Western blot analyses using monoclonal antibodies confirm that the protein band visualized in the SDS gel is hsRec2/Rad51L1 (data not shown). Protein kinase activity was determined initially using kemptide and MBP as substrates. MBP is a commonly used substrate to detect activity for both protein kinase C (PKC) and protein kinase A (PKA) whereas

35

FIG. 1. Purification of hsRec2/Rad51L1. (A) HsRec2/Rad52L1 produced in baculovirus was eluted from a Ni–NTA agarose column and run on a 10% SDS–acrylamide gel and stained with silver. Lane 1, prestained standards; lane 2, 0.036 mg; lane 3, 0.072 mg; lane 4, 0.180 mg hsRec2/Rad51L1. (B) HsRec2/Rad51L1 produced in E. coli was eluted from a Ni–NTA agarose column and run on a 10% SDS– acrylamide gel and stained with Coomassie blue. Lane 1, 0.3 mg; lane 2, 0.6 mg; lane 3, 1.2 mg hsRec2/Rad51L1 protein; lane 4, prestained SDS–PAGE standards, low range.

kemptide is a substrate for PKA. In the case of MBP, one serine and one threonine are preferentially phosphorylated by PKA, whereas two serines are phosphorylated by PKC and six residues (serines and threonines) can be phosphorylated by both kinases [17]. A time course is shown (Fig. 2A) illustrating that hsRec2/ Rad51L1 exhibits kinase activity and that incorporation of phosphate increased during the period assayed. Kemptide, is a heptapeptide (Leu-Arg-Arg-Ala-SerLeu-Gly) containing only one residue that can be phosphorylated and has been used extensively in PKA kinase assays. The results of a protein kinase assay using different concentrations of kemptide and 1 mg of hsRec2/Rad51L1 are shown in Fig. 2B. As can be seen from the direct plot of kemptide concentration versus phosphate incorporated per minute into kemptide, hsRec2/Rad51L1 exhibits saturable kinase activity toward this substrate. Using kemptide as substrate, which contains one residue (serine) that can be phosphorylated, 0.167 pmol of phosphate is incorporated per minute per microgram of hsRec2/Rad51L1 added.

36

HAVRE ET AL.

FIG. 2. Phosphorylation of proteins by hsRec2/Rad51L1. (A) Myelin basic protein (0.25 mM) was incubated for various times with hsRec2/Rad51L1 and [g- 32P]ATP, and an aliquot (half of the reaction mixture) was spotted on phosphocellulose filters and washed with phosphoric acid and water prior to counting. (B) Kemptide was used as a substrate instead of myelin basic protein and the kemptide concentration varied. The reaction was carried out for 60 min. (C) Various levels of purified hsRec2/Rad51L1 protein, expressed in E. coli were incubated with 1.0 mg of kemptide and [g- 32P]ATP. Counts per minute are shown after subtraction of background levels. (D) HsRec2/Rad51L1 renatured following SDS gel electrophoresis was assayed using MBP as substrate. Gel slices are numbered from bottom of gel.

GENOME SURVEILLANCE PROCESSES IN MAMMALIAN CELLS

Stated differently, this is 0.2 pmol of phosphate incorporated into kemptide in 30 min/pmol of hsRec2/ Rad51L1 added to the assay. In a similar fashion, purified hsRec2/Rad51L1 expressed in E. coli is active in phosphorylating kemptide in a dose-dependent fashion (Fig. 2C). Two substrates that were not phosphorylated by hsRec2/Rad51L1 were a tyrosine kinase substrate peptide containing one tyrosine, derived from the sequence surrounding the phosphorylation site in pp60 src (RRLIEDAEYAARG), and PHAS-I, which is a good substrate for mitogen-activated protein kinase (MAPK), p38, and PKC, but not PKA (data not shown). To further address the possibility of a contaminating kinase copurifying with hsRec2/Rad51L1, hsRec2/ Rad51L1 isolated from a baculovirus expression system and eluted from the Ni–NTA agarose column was run on an SDS gel which was cut into equal slices. Any protein present in each slice was renatured in dialysis buffer (see Methods) supplemented with 1% Triton X-100. The renatured samples were then assayed using MBP as substrate. Kinase activity coeluted with hsRec2/Rad51L1 as shown in the silver-stained gel run on duplicate samples (Fig. 2D). Taken together, these data indicate that the kinase activity is not derived from a contaminant protein. Additional support is presented in Fig. 3D (see below). Phosphorylation of p53 by hsRec2/Rad51L1. Since hsRec2/Rad51L1 phosphorylated two synthetic substrates, the next step was to check another protein substrate that could be phosphorylated by kinases PKA and PKC. p53 can be phosphorylated by both of these kinases and, furthermore, phosphorylation of one or more specific residues of p53 have been implicated in response to DNA damage and cell cycle delay [18 –20]. As an initial step, we asked whether hsRec2/Rad51L1 modifies p53, as a possible route for regulating its activity. Evidence that hsRec2/Rad51L1 and p53 were associated came from two experiments. First, hsRec2/ Rad51L1 protein was synthesized by in vitro translation using a reticulocyte lysate (see Methods) in the presence of [ 35S] methionine. Radiolabeled hsRec2/ Rad51L1 or Xef1 (used as a control) was incubated with a cell-free extract prepared from HCT116 cells. Monoclonal antibodies directed against PCNA, p53, or cdc2 were incubated with reticulocyte lysates containing in vitro translated hsRec2/Rad51L1 for 16 h. As seen in Fig. 3A, 35S-labeled hsRec2/Rad51L1, but not 35 S-labeled Xef1, was coimmunoprecipitated by all three antibodies. The results of this IP suggested hsRec2/Rad51L1 and p53 were associated in vitro. In a second assay, hsRec2/Rad51L1 was incubated with agarose GST–p53 and after 1 h the agarose beads were pelleted. After several stringent washes (see Methods), the remaining proteins in the pellets were visualized

37

by SDS–PAGE (Fig. 3B). The proteins hsRec2/ Rad51L1 and p53 are present together in the lane on the far right, further establishing that these two proteins associate in vitro. Since both PKA and PKC have been reported to phosphorylate p53 [21, 22], a protein kinase assay was carried out using purified recombinant p53 and hsRec2/Rad51L1. Shown in Fig. 3C (top) is an autorad from a gel on which reaction products from a kinase reaction were run. Phosphorylation of p53 (see arrow, Fig. 3C, top) was quantitated by phosphorimager analysis and expressed as phosphorimage units in Fig. 3C (bottom). Phosphorylation of p53 increases monotonically with time of incubation in the presence of hsRec2/ Rad51L1. Although we do not know the identity of the lower two bands on this overexposed gel, they are likely to be breakdown products of p53. As shown in Fig. 3C (bottom), quantitation of the p53 phosphorylation event confirms that an increasing amount of reaction occurs as a function of time. In addition to using hsRec2/Rad51L1 eluted from Ni– NTA agarose, which contains one major band (Fig. 1A), further purification was achieved by immunoprecipitation using monoclonal antibodies. The immunoprecipitated hsRec2/Rad51L1 was then assayed for protein kinase activity using p53 as a substrate (Fig. 3D). Lanes 3 and 5 (left) show that p53 is phosphorylated by hsRec2/ Rad51L1 before and after immunoprecipitation, respectively. Lane 2 contains no hsRec2/Rad51L1, and lane 4 contains a negative immunoprecipitation control to which no hsRec2/Rad51L1 was added. The gel on the right (Fig. 3E) confirms that the immunoprecipitation worked; two negative controls are shown, one without hsRec2/Rad51L1 (lane 3) and one without antibodies (lane 4). Lane 5 contains the immunoprecipitate when both antibodies and hsRec2/Rad51L1 were present and lane 2 contains 0.04 mg hsRec2/Rad51L1. Hence, under more stringent conditions, the phosphorylation of p53 by hsRec2/Rad51L1 was still observed. Phosphorylation of cyclin E and cdk2. Phosphorylation of cdk2 on thr14 or tyr15 or inhibition of thr16O phosphorylation could inactivate the enzyme, thereby preventing phosphorylation of Rb and the onset of S phase [23]. Furthermore, phosphorylation of one of the G1-specific cyclins, D or E, could shorten their halflives which would halt cell cycle progression [24, 25]. Therefore, two of these cell cycle proteins, cdk2 and cyclin E, were used as substrates for hsRec2/Rad51L1. GST– cyclin E was expressed in E. coli and purified on a glutathione–Sepharose column. In addition, cdk2 bound to cyclin was prepared following the procedure of Koff et al. [16]. Briefly, GST– cyclin E bound to glutathione–Sepharose was incubated with cdk2 for 30 min in the presence of 10 mM ATP and then washed in buffer without ATP prior to use in the assay. GST–

38

HAVRE ET AL.

GENOME SURVEILLANCE PROCESSES IN MAMMALIAN CELLS

39

FIG. 3—Continued

cyclin E and GST– cyclin E/cdk2 were used as a 1:1 slurry in 50 mM Tris–HCl, 10 mM MgCl 2. HsRec2/ Rad51L1 was added for 60 min at 30°C in the presence of radiolabeled [g- 32P]ATP and the reaction mixtures were processed for SDS–PAGE as described for p53 phosphorylation. The results are depicted in Fig. 4. Upon addition of hsRec2/Rad51L1, cyclin E became phosphorylated (lane 2). When bound cdk2 is present, cyclin E becomes phosphorylated even when hsRec2/ Rad51L1 is absent (cyclin E is phosphorylated by cdk2,

lane 3), but in the presence of hsRec2/Rad51L1, the phosphorylation of the cyclin increases significantly. To strengthen these findings, hsRec2/Rad51L1 eluted from the Ni–NTA agarose column was further purified by immunoprecipitation. For this IP, three different monoclonal antibodies were used. As shown by the silver-stained gel (Fig. 4B, top), each monoclonal antibody precipitated hsRec2/Rad51L1, albeit to varying degrees. Incubation with GST– cyclin E showed that phosphorylation was proportional to the amount

FIG. 3. Association of p53 with hsRec2/Rad51L1 and phosphorylation of p53 by hsRec2/Rad51L1. (A) In vitro translation of hsRec2/ Rad51L1 was carried out as described under Methods and 36 ml of reticulocyte lysate containing Xefl or hsRec2/Rad51L1 translation products labeled with [ 35S]methionine was preincubated with HCT116 cell lysate and then antibodies (Oncogene) to PCNA (NA03), p53 (OP03), or cdc2 (PC25) were added followed by protein A–Sepharose. Washed pellets were suspended in sample buffer, boiled, and run on a 10% SDS gel and then transferred to nitrocellulose. A 48-h exposure is shown. Lane 1, prestained standards; lanes 2 and 3, Xef1 and hsRec2/Rad51L1 translation products equivalent to 2.5 ml of reticulocyte lysate; lanes 4, 6, and 8 (Xef1) and lanes 5, 7, and 9 (hsRec2/Rad51L1) contain 15 of 40 ml of the immunoprecipitate. (B) Association of hsRec2/Rad51L1 with agarose–GST–p53. Lane 1, prestained standards; lane 2, 0.2 mg hsRec2/Rad51L1; lane 3, agarose–GST–p53 without hsRec2/Rad51L1; and lane 4, agarose–GST–p53 with hsRec2/Rad51L1. Arrows indicate position of GST–p53 and hsRec2/Rad51L1. (C) Phosphorylation of p53 by hsRec2/Rad51L1. (Top) Human recombinant p53 (0.5 mg) was incubated for 2– 60 min, with or without hsRec2/Rad51L1 at 30°C, as indicated. The reaction was terminated by the addition of 23 gel loading buffer and heated prior to loading on a 10% SDS gel, transferred to nitrocellulose, and exposed to X-Omat film. The main p53-phosphorylated band is indicated with an arrow (confirmed by Western blot). (Bottom) Phosphorylation of p53 (arrow, top) was quantitated by phosphorimage analysis and the activity is represented by phosphorimage units. (D) Phosphorylation of p53 using immunoprecipitated hsRec2/Rad51L1. Lane 1, standards; lane 2, 0.25 mg p53 and no hsRec2/Rad51L1; lane 3, 0.25 mg p53 and 0.1 mg hsRec2/Rad51L1; lane 4, 0.25 mg p53 and IP without hsRec2/Rad51L1; lane 5, 0.25 mg p53 and IP with hsRec2/Rad51L1. (E) Silver-stained gel containing immunoprecipitates used in the kinase assay described in D. Lane 1, standards; lane 2, 0.04 mg hsRec2/Rad51L1; lane 3, IP without hsRec2/Rad51L1 plus antibody; lane 4, IP with hsRec2/Rad51L1, no antibody; lane 5, IP with hsRec2/Rad51L1 plus antibody.

40

HAVRE ET AL.

FIG. 4. Phosphorylation of cdk2/cyclin E by hsRec2/Rad51L1. (A) Phosphorylation of cyclin E/cdk2 by hsRec2/Rad51L1. Lanes 1 and 2 contain GST– cyclin E without and with hsRec2/Rad51L1 as indicated; lanes 3 and 4 contain GST– cyclin E/cdk2 without and with hsRec2/Rad51L1 as indicated. The position of GST– cyclin E, hsRec2/Rad51L1 and cdk2 are indicated with arrows. (B) Immunoprecipitation (IP) of hsRec2/Rad51L1 using monoclonal antibodies. (Top) Silver stain of immunoprecipitates. Lane 1, prestained standards; lane 2, 0.04 mg hsRec2/Rad51L1; lane 3, protein A/G–Sepharose only; lanes 4 – 6, IP using 50 ml of cell supernatants from DF6IB10, DD10IIE11, and EF1KKF12, respectively. Heavy chain IgG (HC IgG), hsRec2/Rad51L1, and light chain IgG (LC IgG) are indicated by arrows. Kinase assay using glutathione–Sepharose GST– cyclin E as substrate. Lane 1, protein A/G–sepharose only; lane 2, 0.2 mg hsRec2/Rad51L1; lanes 3–5, IPs from DF6IB10, DD10IIE11, and EF1KKF12, respectively. GST– cyclin E, hsRec2/Rad51L1 and cdk2 are indicated by arrows.

of hsRec2/Rad51L1 immunoprecipitated (Fig. 4B, bottom). The overall lower level of phosphorylation most likely is a result of both substrate and kinase being immobilized or inhibition of hsRec2/Rad51L1 kinase activity by the monoclonal antibodies used. This evi-

dence provides further support that hsRec2/Rad51L1 is a kinase that phosphorylates cyclin E in vitro. As a final step in establishing the putative kinase activity of the hsRec2/Rad51L1 protein, we carried out site-directed mutagenesis. Since this protein lacks sub-

GENOME SURVEILLANCE PROCESSES IN MAMMALIAN CELLS

41

involved in ATPase activity nucleotide binding and protein phosphorylation. DISCUSSION

FIG. 5. Disruption of the “A box” sequence drastically reduces kinase activity. Reaction mixtures containing 0.5 mg of the indicated protein and 0.2 mM kemptide were incubated at 37°C for 60 min. The reactions were spotted on phosphocellulose disks and processed as described under Methods. Filters were counted in a Wallac scintillation counter and phosphorylation activity is presented as picomoles incorporated after subtraction of background cpm. Wild-type, normal protein containing no alteration; GKtQ 3 Q, protein containing a lysine to glutamic acid residue switch at position 114; LVILD 3 V, protein containing an aspartic acid to valine switch at position 214; FPRY 3 A, protein containing a tyrosine to alanine change at position 163. Statistical analyses were performed using Microsoft Excel III.

stantial regions of amino acids that represent “consensus kinase sequences” [see 26], we modified the two amino acids located in regions that bind ATP [27] and one site that has been chosen to be important to cell cycle regulation [7]. The first alteration took place at position 114 with a switch from lysine (K) to glutamic acid (Q). This location is within the so-called A box [see 1] while a second construct was produced harboring a glutamic acid to valine switch at position 210, the so-called B box [1]. A third change was at amino acid position 163 (tyrosine to alanine had been previously constructed) [7]. Proteins containing these individual mutations were expressed in the baculovirus system and purified as described for the experiments presented as Figs. 2B and 2C. Reactions were carried out in triplicate and the experiment performed three times. Figure 5 illustrates the results. While mutations in the B box or at position 163 had little effect on the kinase activity of hsRec2/Rad51L1, the A box mutation reduced activity 80%. Mixing protein samples containing the A box mutation with samples of the proteins bearing the B box mutation did not reduce B box mutant kinase activity (data not shown). These data suggest that the reduction in kinase activity observed with the A box mutant is not due to inhibitory substances in the sample. Hence, the A box may be a crucial sequence

This report reveals that the human Rec2/Rad51L1 protein has protein kinase activity based on several observations. The enzyme purified from baculovirus and E. coli exhibits kinase activity on two commonly used substrates, kemptide and MBP. Kinase activity was shown to be time-dependent, and using kemptide as a substrate, exhibits saturable kinase activity, indicating that phosphorylation is enzymatic. Phosphorylation of these substrates is observed even after the electrophoresed protein is excised from an SDS gel and renatured. Additionally, immunoprecipitated hsRec2/ Rad51L1 protein, using either combined or three different monoclonal antibodies, retained the capacity to phosphorylate p53 and GST– cyclin E, respectively. Finally, alteration of an amino acid within a region of the protein known to be important for ATP binding and hydrolysis disrupted kinase activity significantly. Although the GxGxxG motif [26] (a sequence known to bind ATP in the classic kinase mode) is absent in hsRec2/Rad51L1, the consensus ATP-binding domain characteristic of RecA-like proteins is present [1] and may provide this function. There are a few examples of protein kinases that do not contain the recognized domains of protein kinases. Two of these, the eukaryotic elongation factor-2 kinase [28] and myosin heavy chain kinase A [29], do not contain sequence motifs characteristic of kinase superfamilies, except for the ATPbinding domain which resides at the carboxyl instead of the amino terminus of the protein. In addition, several other protein kinases have been found to lack the “essential” GxGxxG motif, mikl [30], Vps15p [31], and the Cdk-activating kinase (CAK) from Saccharomyces cerevisiae [32], or to have no protein kinase consensus sequences at all [33, 34]. In contrast, hsRec2/Rad51L1 contains limited homology with catalytic domains X and X1 of PKA [35]. A partial alignment with these domains is shown in Table 1 along with BCR, another

42

HAVRE ET AL.

protein kinase that exhibits limited homology with these domains [36]. Previously, several characteristics of the human gene were identified. Among these is the induction of hREC2/RAD51L1 expression by ionizing radiation [1] and induction by UV light [14]. The downstream effect of elevated hREC2/RAD51L1 levels in the cell may be to retard cell cycle progression, particularly at the G1/S border [7]. Based on these observations, several proteins that participate in cell cycle regulation were considered as likely targets for the phosphorylation activity of hsRec2/Rad51L1 protein. The first, p53, plays a well-established role in cell-cycle regulation and a preliminary association between p53 and the regulation of hREC2/RAD51L1 expression has been demonstrated [14, 37]. In addition, the phosphorylation status controlling the activity of the cdk2/cyclin E complex during the G1/S transition is a critical element of the normal cycling time [see 38 for review]. Hence, p53, cdk2, and cyclin E were chosen as substrates for hsRec2/Rad51L1 kinase activity studies and found to be phosphorylated. A number of kinases have been reported to phosphorylate p53 in vitro [18, 21, 22, 39, 40] and to this list should now be added hsRec2/Rad51L1. Although several kinases have been reported to phosphorylate specific residues that enhance the binding of p53 to its consensus sequence [18, 20], others have shown that it is the phosphorylation of multiple residues [41] that govern p53 activation in vivo. Recently, a CAK complex consisting of CDK, -cycH-p36 (TFIIH), has been shown to phosphorylate p53 in vitro [20]. This group is part of the multisubunit complex comprising TFIIH [42] and includes DNA repair-associated proteins such as ERCC2 [43] and ERCC3 [44]. Since hsRec2/Rad51L1 is induced by UV light [14] and is a DNA repair protein [1, 8, 9], it is possible that the phosphorylation of p53 by hsRec2/Rad51L1 operates through a similar pathway. In addition, recent studies suggest that phosphorylation of residue 33 may play a role in the in vivo activation of p53 [19] and that the kinase responsible (CAK), which plays a role in cell cycle regulation, transcription, and DNA repair, links those three key pathways in the cell and accounts for genome stability. The finding that hsRec2/Rad51L1 may constitute another control point (since it is induced following DNA damage) will require additional experiments to substantiate. Interestingly, the hsRec2/Rad51L1 protein may itself be regulated by phosphorylation since it contains many potential phosphorylation sites; there are four minimal cdk consensus sites [45] and two consensus sites for phosphorylation by DNA PK [39]. In this report, our goal was to establish the kinase activity of hsRec2/Rad51L1 and to begin a survey of possible substrates whose modification might influence the observed cell cycle delay. We envision a preliminary scenario wherein DNA damage induces hREC2/

RAD51L1 expression [14] through a process invoking regulation by p53, with elevated levels of hsRec2/ Rad51L1 resulting in phosphorylation of key components modulating cell cycle progression. These modified proteins could then produce effects that lead to a block at the G1/S border. Since hREC2/RAD51L1 is expressed normally at low levels [1, 8, 9], it appears that the presence of hsRec2/Rad51L1 probably serves to alert the cell that genomic insult has occurred. The appropriate cell response would be to prevent propagation of mutations by a reduction in the rate of DNA synthesis. The significance of these gene, protein, and associated activities has recently been heightened by the identification of a specific 14t12 translocation in cases of uterine leiomyoma [46, 47]. The hREC2/RAD51L1 gene is translocated from chromosome 14 to 12 in these cases of cancer with one of the specific breakpoints occurring, most likely, in the promoter region [14] of the gene. A malfunctioning regulatory activity of a key protein may account, at least in part, for this oncogenesis. We thank Linda Miller and Song Mei Xu for purification of hsRec2/ Rad51L1 and the use of the monoclonal antibodies, Dr. Z. Yu for the expression of hsRec2/Rad51L1 in baculovirus, Anthony Rice for the graphics design, and Christina Johnson for manuscript preparation. This research was supported by NIH Grant R01 AR44692-01. We are grateful to our colleague, Allyson Cole-Strauss for her help in the mutagenesis assays.

REFERENCES 1.

Rice, M. C., Smith, S. T., Bullrich, F., Havre, P. A., and Kmiec, E. B. (1997). Isolation of human and mouse genes based on homology to REC2, a recombinational repair gene from the fungus Ustilago maydis. Proc. Natl. Acad. Sci. USA 94, 7417– 7422.

2.

Kmiec, E. B., and Holloman, W. K. (1982). Homologous pairing of DNA molecules promoted by a protein from Ustilago. Cell 29, 367–374.

3.

Kmiec, E. B., and Holloman, W. K. (1986). Homologous pairing of DNA molecules by Ustilago rec1 protein is promoted by sequences of Z-DNA. Cell 44, 545–554.

4.

Rubin, B. P., Ferguson, D. O., and Holloman, W. K. (1994). Structure of REC2, a recombinational repair gene of Ustilago maydis, and its function in homologous recombination between plasmid and chromosomal sequences. Mol. Cell. Biol. 14, 6287– 6296.

5.

Kmiec, E. B., Cole, A., and Holloman, W. K. (1994). The REC2 gene encodes the homologous pairing protein of Ustilago maydis. Mol. Cell. Biol. 14, 7163–7172.

6.

Ferguson, D. O., Rice, M. C., Rendi, M. H., Kotani, H., Kmiec, E. B., and Holloman, W. K. (1997). Interaction between Ustilago maydis REC2 and RAD51 genes in DNA repair and mitotic recombination. Genetics 145, 243–252.

7.

Havre, P. A., Rice, M. C., No¨e, A., and Kmiec, E. B. (1998). The human REC2/RAD51B gene acts as a DNA damage sensor by inducing G1 delay and hypersensitivity to ultraviolet irradiation. Cancer Res. 58, 4733– 4739.

GENOME SURVEILLANCE PROCESSES IN MAMMALIAN CELLS 8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

Albala, J. S., Thelen, M. P., Prange, C., Fan, W., Christensen, M., Thompson, L. H., and Lennon, G. G. (1997). Identification of a novel human RAD51 homolog, RAD51B. Genomics 46, 476 – 479. Cartwright, R., Dunn, A. M. P. J., Simpson, P. J., Tambini, C. E., and Thacker, J. (1998). Isolation of novel human and mouse genes of the recA/RAD51 recombination-repair gene family. Nucleic Acids Res. 26, 1653–1659. Sung, P. (1994). Catalysis of ATP-dependent homologous DNA pairing and strand exchange by yeast RAD51 protein. Science 265, 1241–1243. Tsuzuki, T., Fujii, Y., Sakumi, K., Tominaga, Y., Nakao, K., Sekiguchi, M., Matsushiro, A., Yoshimura, Y., and Morita, T. (1996). Targeted disruption of the Rad51 gene leads to lethality in embryonic mice. Proc. Natl. Acad. Sci. USA 93, 6236 – 6240. Morgan, D. O. (1997). Cyclin-dependent kinases: Engines, clocks, and microprocessors. Annu. Rev. Cell. Dev. Biol. 13, 261–291. Dulic, V., Kaufmann, W. K., Wilson, S. J., Tlsty, T. D., Lees, E., Harper, J. W., Elledge, S. J., and Reed, S. I. (1994). p53-dependent inhibition of cyclin-dependent kinase activities in human fibroblasts during radiation-induced G1 arrest. Cell 76, 1013– 1023. Peng, L., Rice, M. C., and Kmiec, E. B. (1998). Analysis of the human RAD51L1 promoter region and its activation by UV light. Genomics 54, 541. Laemmli, U. K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680 – 685. Koff, A., Giordano, A., Desai, D., Yamashita, K., Harper, J. W., Elledge, S., Nishimoto, T., Morgan, D. O., Franza, B. R., and Roberts, J. M. (1992). Formation and activation of a cyclin E-cdk2 complex during the G1 phase of the human cell cycle. Science 257, 1689 –1694. Kishimoto, A., Nishiyama, K., Nakanishi, H., Uratsuji, Y., Nomura, H., Takeyama, Y., and Nishizuka, Y. (1985). Studies on the phosphorylation of myelin basic protein by protein kinase C and adenosine 39, 59-monophosphate-dependent protein kinase. J. Biol. Chem. 260, 12492–12499. Shieh, S. Y., Ikeda, M., Taya, Y., and Prives, C. (1997). DNA damage-induced phosphorylation of p53 alleviates inhibition by MDM2. Cell 91, 325–334. Ko, L. J., Shieh, S. Y., Chen, X. B., Jayaraman, L., Tamai, Y., Taya, Y., Prives, C., and Pan, Z. Q. (1997). p53 is phosphorylated by CDK7-cyclin H in a p36(MAT1)-dependent manner. Mol. Cell. Biol. 17, 7220 –7229. Lu, H., Fisher, R. P., Bailey, P., and Levine, A. J. (1997). The CDK7-CYCH-p36 complex of transcription factor IIH phosphorylates p53, enhancing its sequences-specific DNA binding activity in vitro. Mol. Cell. Biol. 17, 5923–5934. Adler, V., Pincus, M. R., Minamoto, T., Fuchs, S. Y., Bluth, M. J., Brandt-Rauf, P. W., Friedman, F. K., Robinson, R. C., Chen, J. M., Wang, X. W., Harris, C. C., and Ronai, Z. (1997). Conformation-dependent phosphorylation of p53. Proc. Natl. Acad. Sci. USA 94, 1686 –1691. Baudier, J., Delphin, C., Grunwald, D., Khochbin, S., and Lawrence, J. J. (1992). Characterization of the tumor suppressor protein p53 as a protein kinase C substrate and a S100bbinding protein. Proc. Natl. Acad. Sci. USA 89, 11627–11631. Gu, Y., Rosenblatt, J., and Morgan, D. O. (1992). Cell cycle regulation of CDK2 activity by phosphorylation of Thr160 and Tyr15. EMBO J. 11, 3995– 4005. Diehl, J. A., Zindy, F., and Sherr, C. J. (1997). Inhibition of cyclin D1 phosphorylation on threonine-286 prevents its rapid

25.

26.

27.

28.

29.

30.

31.

32. 33.

34.

35.

36.

37.

38. 39.

40.

41.

43

degradation via the ubiquitin-proteasome pathway. Genes Dev. 11, 957–972. Won, K. A., and Reed, S. I. (1996). Activation of cyclin E/CDK2 is coupled to site-specific autophosphorylation and ubiquitindependent degradation of cyclin E. EMBO J. 15, 4182– 4193. Hanks, S. K., Quinn, A. M., and Hunter, T. (1988). The protein kinase family: Conserved features and deduced phylogeny of the catalytic domains [Review]. Science 241, 42–52. Aboussekhra, A., Chanet, R., Adijiri, A., and Fabre, F. (1992). Semidominant suppressors of Srs2 helicase mutations of Saccharomyces cerevisiae map in the RAD51 gene, whose sequence predicts a protein with similarities to procaryotic RecA proteins. Mol. Cell. Biol. 12, 3224 –3234. Ryazanov, A. G., Ward, M. D., Mendola, C. E., Pavur, K. S., Dorovkov, M. V., Wiedmann, M., Erdjument-Bromage, H., Tempst, P., Parmer, T. G., Prostko, C. R., Germino, F. J., and Hait, W. N. (1997). Identification of a new class of protein kinases represented by eukaryotic elongation factor-2 kinase. Proc. Natl. Acad. Sci. USA 94, 4884 – 4889. Futey, L. M., Medley, Q. G., Cote, G. P., and Egelhoff, T. T. (1995). Structural analysis of myosin heavy chain kinase A from Dictyostelium: Evidence for a highly divergent protein kinase domain, an amino-terminal coiled-coil domain, and a domain homologous to the beta-subunit of heterotrimeric G proteins. J. Biol. Chem. 270, 523–529. Lundgren, K., Walworth, N., Booher, R., Dembski, M., Kirschner, M., and Beach, D. (1991). mik1 and wee1 cooperate in the inhibitory tyrosine phosphorylation of cdc2. Cell 64, 1111–1122. Herman, P. K., Stack, J. H., DeModena, J. A., and Emr, S. D. (1991). A novel protein kinase homolog essential for protein sorting to the yeast lysosome-like vacuole. Cell 64, 425– 437. Kaldis, P., Sutton, A., and Solomon, M. J. (1996). The Cdkactivating kinase (CAK) from budding yeast. Cell 86, 553–564. Eichinger, L., Bomblies, L., Vandekerckhove, J., Schleicher, M., and Gettemans, J. (1996). A novel type of protein kinase phosphorylates actin in the actin-fragmin complex. EMBO J. 15, 5547–5556. Beeler, J. F., LaRochelle, W. J., Chedid, M., Tronick, S. R., and Aaronson, S. A. (1994). Prokaryotic expression cloning of a novel human tyrosine kinase. Mol. Cell. Biol. 14, 982–988. Hanks, S. K., and Quinn, A. M. (1991). Protein kinase catalytic domain sequence database: Identification of conserved features of primary structure and classification of family members. Methods Enzymol. 200, 38 – 62. Maru, Y., and Witte, O. N. (1991). The BCR gene encodes a novel serine/threonine kinase activity within a single exon. Cell 67, 459 – 468. Kuerbitz, S. J., Plunkett, B. S., Walsh, W. V., and Kastan, M. B. (1992). Wild-type p53 is a cell cycle checkpoint determinant following irradiation. Proc. Natl. Acad. Sci. USA 89, 7491– 7495. Morgan, S. E., and Kastan, M. B. (1997). p53 and ATM: Cell cycle, cell death, and cancer. Adv. Cancer Res. 71, 1–25. Lees-Miller, S. P., Sakaguchi, K., Ullrich, S. J., Appella, E., and Anderson, C. W. (1992). Human DNA-activated protein kinase phosphorylates serines 15 and 37 in the amino-terminal transactivation domain of human p53. Mol. Cell. Biol. 12, 5041– 5049. Milne, D. M., Campbell, L. E., Campbell, D. G., and Meek, D. W. (1995). p53 is phosphorylated in vitro and in vivo by an ultraviolet radiation-induced protein kinase characteristic of the c-Jun kinase, JNK1. J. Biol. Chem. 270, 5511–5518. Mayr, G. A., Reed, M., Wang, P., Wang, Y., Schweds, J. F., and Tegtmeyer, P. (1995). Serine phosphorylation in the NH2 ter-

44

HAVRE ET AL.

minus of p53 facilitates transactivation. Cancer Res. 55, 2410 – 2417. 42. Flores, O., Lu, H., and Reinberg, D. (1992). Factors involved in specific transcription by mammalian RNA polymerase II. Identification and characterization of factor IIH. J. Biol. Chem. 267, 2786 –2793. 43. Schaeffer, L., Moncollin, V., Roy, R., Staub, A., Mezzina, M., Sarasin, A., Weeda, G., Hoeijmakers, J. H., and Egly, J. M. (1994). The ERCC2/DNA repair protein is associated with the class II BTF2/TFIIH transcription factor. EMBO J. 13, 2388 – 2392. 44. Schaeffer, L., Roy, R., Humbert, S., Moncollin, V., Vermeulen, W., Hoeijmakers, J. H., Chambon, P., and Egly, J. M. (1993). Received May 11, 1999 Revised version received October 4, 1999

DNA repair helicase: A component of BTF2 (TFIIH) basic transcription factor. Science 260, 58 – 63. 45. Jallepalli, P. V., and Kelly, T. J. (1997). Cyclin-dependent kinase and initiation at eukaryotic origins: A replication switch? [Review]. Curr. Opin. Cell. Biol. 9, 358 –363. 46. Bhugra, B., Smolarek, T. A., Lynch, R. A., Meloni, A. M., Sandberg, A. A., Deaven, L., and Menon, A. G. (1998). Cloning of a breakpoint cluster region on chromosome 14 in uterine leiomyoma. Cancer Lett. 126, 119 –126. 47. Schoenmakers, E. F. P. M., Huysmans, C., and Van de Ven, W. J. M. (1999). Allelic knockout of novel splice variants of human recombination repair gene RAD51B in t(12:14) uterine leiomyomas. Cancer Res. 59, 19 –23.