Molecular architecture of the mouse DNA polymerase alpha-primase complex.

The DNA polymerase alpha-primase complex is the only enzyme that provides RNA-DNA primers for chromosomal DNA replication in eukaryotes. Mouse DNA polymerase alpha has been shown to consist of four subunits, p180, p68, p54, and p46. To characterize the domain structures and subunit requirements for the assembly of the complex, we constructed eukaryotic polycistronic cDNA expression plasmids expressing pairwise the four subunits of DNA polymerase alpha. In addition, the constructs contained an internal ribosome entry site derived from poliovirus. The constructs were transfected in different combinations with vectors expressing single subunits to allow the simultaneous expression of three or four of the subunits in cultured mammalian cells. We demonstrate that the carboxyl-terminal region of p180 (residues 1235 to 1465) is essential for its interaction with both p68 and p54-p46 by immunohistochemical analysis and coprecipitation studies with antibodies. Mutations in the putative zinc fingers present in the carboxyl terminus of p180 abolished the interaction with p68 completely, although the mutants were still capable of interacting with p54-p46. Furthermore, the amino-terminal region (residues 1 to 329) and the carboxyl-terminal region (residues 1280 to 1465) were revealed to be dispensable for DNA polymerase activity. Thus, we can divide the p180 subunit into three domains. The first is the amino-terminal domain (residues 1 to 329), which is dispensable for both polymerase activity and subunit assembly. The second is the minimal core domain (residues 330 to 1279), required for polymerase activity. The third is the carboxyl-terminal domain (residues 1280 to 1465), which is dispensable for polymerase activity but required for the interaction with the other three subunits. Taken together, these results allow us to propose the first structural model for the DNA polymerase alpha-primase complex in terms of subunit assembly, domain structure, and stepwise formation at the cellular level.

Induction of mutation in mouse FM3A cells by N4-aminocytidine-mediated replicational errors.

To explore the potential use of a nucleoside analog, N4-aminocytidine, in studies of cellular biology, the mechanism of mutation induced by this compound in mouse FM3A cells in culture was studied. On treatment of cells in suspension with N4-aminocytidine, the mutation to ouabain resistance was induced. The major DNA-replicating enzyme in mammalian cells, DNA polymerase alpha, was used to investigate whether the possible cellular metabolite of N4-aminocytidine, N4-aminodeoxycytidine 5'-triphosphate (dCamTP), can be incorporated into the DNA during replication. Using [3H]dCamTP in an in vitro DNA-synthesizing system, we were able to show that this nucleotide analog can be incorporated into newly formed DNA and that it can serve as a substitute for either dCTP or dTTP. dCamTP in the absence of dCTP maintained the activated calf thymus DNA-directed polymerization of deoxynucleoside triphosphates as efficiently as in its presence. Even in the presence of dCTP, dCamTP was incorporated into the polynucleotide. When dCamTP was used as a single substrate in the poly(dA)-oligo(dT)-directed polymerase reaction, it was incorporated into the polynucleotide fraction. The extent of incorporation was 4% of that of dTTP incorporation when dTTP was used as a single substrate. Even in the presence of dTTP, dCamTP incorporation was observed. A copolymer containing N4-aminocytosine residues was shown to incorporate guanine residues opposite the N4-aminocytosines. However, we were unable to observe adenine incorporation opposite N4-aminocytosine in templates. These cell-free experiments show that an AT-to-GC transition can take place in the presence of dCamTP during DNA synthesis, strongly suggesting that the mutation induced in the FM3A cells by N4-aminocytidine is due to replicational errors.

Incubation at the nonpermissive temperature induces deficiencies in UV resistance and mutagenesis in mouse mutant cells expressing a temperature-sensitive ubiquitin-activating enzyme (E1).

In temperature-sensitive (ts) mutants of mouse FM3A cells, the levels of mutagenesis and survival of cells treated with DNA-damaging agents have been difficult to assess because they are killed after their mutant phenotypes are expressed at the nonpermissive temperature. To avoid this difficulty, we incubated the ts mutant cells at the restrictive temperature, 39 degrees C, for only a limited period after inducing DNA damage. We used ts mutants defective in genes for ubiquitin-activating enzyme (E1), DNA polymerase alpha, and p34(cdc2) kinase. Whereas the latter two showed no effect, E1 mutants were sensitized remarkably to UV light if incubated at 39 degrees C for limited periods after UV exposure. Eighty-five percent of the sensitization occurred within the first 12 h of incubation at 39 degrees C, and more than 36 h at 39 degrees C did not produce any further sensitization. Moreover, while the 39 degrees C incubation gave E1 mutants a moderate spontaneous mutator phenotype, the same treatment significantly diminished the level of UV-induced 6-thioguanine resistance mutagenesis and extended the time necessary for expression of the mutation phenotype. These characteristics of E1 mutants are reminiscent of the defective DNA repair phenotypes of Saccharomyces cerevisiae rad6 mutants, which have defects in a ubiquitin-conjugating enzyme (E2), to which E1 is known to transfer ubiquitin. These results demonstrate the involvement of E1 in eukaryotic DNA repair and mutagenesis and provide the first direct evidence that the ubiquitin-conjugation system contributes to DNA repair in mammalian cells.

Studies of nematode TFIIE function reveal a link between Ser-5 phosphorylation of RNA polymerase II and the transition from transcription initiation to elongation.

The general transcription factor TFIIE plays important roles in transcription initiation and in the transition to elongation. However, little is known about its function during these steps. Here we demonstrate for the first time that TFIIH-mediated phosphorylation of RNA polymerase II (Pol II) is essential for the transition to elongation. This phosphorylation occurs at serine position 5 (Ser-5) of the carboxy-terminal domain (CTD) heptapeptide sequence of the largest subunit of Pol II. In a human in vitro transcription system with a supercoiled template, this process was studied using a human TFIIE (hTFIIE) homolog from Caenorhabditis elegans (ceTFIIEalpha and ceTFIIEbeta). ceTFIIEbeta could partially replace hTFIIEbeta, whereas ceTFIIEalpha could not replace hTFIIEalpha. We present the studies of TFIIE binding to general transcription factors and the effects of subunit substitution on CTD phosphorylation. As a result, ceTFIIEalpha did not bind tightly to hTFIIEbeta, and ceTFIIEbeta showed a similar profile for binding to its human counterpart and supported an intermediate level of CTD phosphorylation. Using antibodies against phosphorylated serine at either Ser-2 or Ser-5 of the CTD, we found that ceTFIIEbeta induced Ser-5 phosphorylation very little but induced Ser-2 phosphorylation normally, in contrast to wild-type hTFIIE, which induced phosphorylation at both Ser-2 and Ser-5. In transcription transition assays using a linear template, ceTFIIEbeta was markedly defective in its ability to support the transition to elongation. These observations provide evidence of TFIIE involvement in the transition and suggest that Ser-5 phosphorylation is essential for Pol II to be in the processive elongation form.

The second-largest subunit of the mouse DNA polymerase alpha-primase complex facilitates both production and nuclear translocation of the catalytic subunit of DNA polymerase alpha.

DNA polymerase alpha-primase is a replication enzyme necessary for DNA replication in all eukaryotes examined so far. Mouse DNA polymerase alpha is made up of four subunits, the largest of which is the catalytic subunit with a molecular mass of 180 kDa (p180). This subunit exists as a tight complex with the second-largest subunit (p68), whose physiological role has remained unclear up until now. We set out to characterize these subunits individually or in combination by using a cDNA expression system in cultured mammalian cells. Coexpression of p68 markedly increased the protein level of p180, with the result that ectopically generated DNA polymerase activity was dramatically increased. Immunofluorescence analysis showed that while either singly expressed p180 or p68 was localized in the cytoplasm, cotransfection of both subunits resulted in colocalization in the nucleus. We identified a putative nuclear localization signal for p180 (residues 1419 to 1437) and found that interaction with p68 is essential for p180 to translocate into the nucleus. These results indicate that association of p180 with p68 is important for both protein synthesis of p180 and translocation into the nucleus, implying that p68 plays a pivotal role in the newly synthesized DNA polymerase alpha complex.

HHR23B, a human Rad23 homolog, stimulates XPC protein in nucleotide excision repair in vitro.

A protein complex which specifically complements defects of XP-C cell extracts in vitro was previously purified to near homogeneity from HeLa cells. The complex consists of two tightly associated proteins: the XPC gene product and HHR23B, one of two human homologs of the Saccharomyces cerevisiae repair gene product Rad23 (Masutani et al., EMBO J. 13:1831-1843, 1994). To elucidate the roles of these proteins in "genome-overall" repair, we expressed the XPC protein in a baculovirus system and purified it to near homogeneity. The recombinant human XPC (rhXPC) protein exhibited a high level of affinity for single-stranded DNA and corrected the repair defect in XP-C whole-cell extracts without extra addition of recombinant HHR23B (rHHR23B) protein. However, Western blot (immunoblot) experiments revealed that XP-C cell extracts contained excess endogenous HHR23B protein, which might be able to form a complex upon addition of the rhXPC protein. To investigate the role of HHR23B, we fractionated the XP-C cell extracts and constructed a reconstituted system in which neither endogenous XPC nor HHR23B proteins were present. In this assay system, rhXPC alone weakly corrected the repair defect, while significant enhancement of the correcting activity was observed upon coaddition of rHHR23B protein. Stimulation of XPC by HHR23B was found with simian virus 40 minichromosomes as well as with naked plasmid DNA and with UV- as well as N-acetoxy-2- acetylfluorene-induced DNA lesions, indicating a general role of HHR23B in XPC functioning in the genome-overall nucleotide excision repair subpathway.

Identification and characterization of XPC-binding domain of hHR23B.

hHR23B was originally isolated as a component of a protein complex that specifically complements nucleotide excision repair (NER) defects of xeroderma pigmentosum group C cell extracts in vitro and was identified as one of two human homologs of the Saccharomyces cerevisiae NER gene product Rad23. Recombinant hHR23B has previously been shown to significantly stimulate the NER activity of recombinant human XPC protein (rhXPC). In this study we identify and functionally characterize the XPC-binding domain of hHR23B protein. We prepared various internal as well as terminal deletion products of hHR23B protein in a His-tagged form and examined their binding with rhXPC by using nickel-chelating Sepharose. We demonstrate that a domain covering 56 amino acids of hHR23B is required for binding to rhXPC as well as for stimulation of in vitro NER reactions. Interestingly, a small polypeptide corresponding to the XPC-binding domain is sufficient to exert stimulation of XPC NER activity. Comparison with known crystal structures and analysis with secondary structure programs provided strong indications that the binding domain has a predominantly amphipathic alpha-helical character, consistent with evidence that the affinity with XPC is based on hydrophobic interactions. Our work shows that binding to XPC alone is required and sufficient for the role of hHR23B in in vitro NER but does not rule out the possibility that the protein has additional functions in vivo.

Two human homologs of Rad23 are functionally interchangeable in complex formation and stimulation of XPC repair activity.

XPC-hHR23B protein complex is specifically involved in nucleotide excision repair (NER) of DNA lesions on transcriptionally inactive sequences as well as the nontranscribed strand of active genes. Here we demonstrate that not only highly purified recombinant hHR23B (rhHR23B) but also a second human homolog of the Saccharomyces cerevisiae Rad23 repair protein, hHR23A, stimulates the in vitro repair activity of recombinant human XPC (rhXPC), revealing functional redundancy between these human Rad23 homologs. Coprecipitation experiments with His-tagged rhHR23 as well as sedimentation velocity analysis showed that both rhHR23 proteins in vitro reconstitute a physical complex with rhXPC. Both complexes were more active than free rhXPC, indicating that complex assembly is required for the stimulation. rhHR23B was shown to stimulate an early stage of NER at or prior to incision. Furthermore, both rhHR23 proteins function in a defined NER system reconstituted with purified proteins, indicating direct involvement of hHR23 proteins in the DNA repair reaction via interaction with XPC.

Complementation of defective translesion synthesis and UV light sensitivity in xeroderma pigmentosum variant cells by human and mouse DNA polymerase eta.

Defects in the human gene XPV result in the variant form of the genetic disease xeroderma pigmentosum (XP-V). XPV encodes DNA polymerase eta, a novel DNA polymerase that belongs to the UmuC/DinB/Rad30 superfamily. This polymerase catalyzes the efficient and accurate translesion synthesis of DNA past cis-syn cyclobutane di-thymine lesions. In this report we present the cDNA sequence and expression profiles of the mouse XPV gene and demonstrate its ability to complement defective DNA synthesis in XP-V cells. The mouse XPV protein shares 80.3% amino acid identity and 86.9% similarity with the human XPV protein. The recombinant mouse XPV protein corrected the inability of XP-V cell extracts to carry out DNA replication, by bypassing thymine dimers on template DNA. Transfection of the mouse or human XPV cDNA into human XP-V cells corrected UV sensitivity. Northern blot analysis revealed that the mouse XPV gene is expressed ubiquitously, but at a higher level in testis, liver, skin and thymus compared to other tissues. Although the mouse XPV gene was not induced by UV irradiation, its expression was elevated approximately 4-fold during cell proliferation. These results suggest that DNA polymerase eta plays a role in DNA replication, though the enzyme is not essential for viability.

Sulfolobus solfataricus P2 DNA polymerase IV (Dpo4): an archaeal DinB-like DNA polymerase with lesion-bypass properties akin to eukaryotic poleta.

Phylogenetic analysis of Y-family DNA polymerases suggests that it can be subdivided into several discrete branches consisting of UmuC/DinB/Rev1/Rad30/Rad30A and Rad30B. The most diverse is the DinB family that is found in all three kingdoms of life. Searches of the complete genome of the crenarchaeon Sulfolobus solfataricus P2 reveal that it possesses a DinB homolog that has been termed DNA polymerase IV (Dpo4). We have overproduced and purified native Dpo4 protein and report here its enzymatic characterization. Dpo4 is thermostable, but can also synthesize DNA at 37 degrees C. Under these conditions, the enzyme exhibits misinsertion fidelities in the range of 8 x 10(-3) to 3 x 10(-4). Dpo4 is distributive but at high enzyme to template ratios can synthesize long stretches of DNA and can substitute for Taq polymerase in PCR. On damaged DNA templates, Dpo4 can facilitate translesion replication of an abasic site, a cis-syn thymine-thymine dimer, as well as acetyl aminofluorene adducted- and cisplatinated-guanine residues. Thus, although phylogenetically related to DinB polymerases, our studies suggest that the archaeal Dpo4 enzyme exhibits lesion-bypass properties that are, in fact, more akin to those of eukaryotic poleta.

Cloning and characterization of the 5'-upstream sequence governing the cell cycle-dependent transcription of mouse DNA polymerase alpha 68 kDa subunit gene.

We have isolated the genomic DNA fragment spanning the 5-end and the first four exons encoding the 68 kDa subunit (p68) of the mouse DNA polymerase alpha-primase complex [corrected]. The p68 promoter region lacks TATA and CAAT boxes, but contains a GC-rich sequence, two palindrome sequences and two putative E2F-binding sites [corrected]. A series of transient expression assays using a luciferase reporter gene indicated that a region from nucleotide position -89 to -30 (-89/-30) with respect to the transcription initiation site is crucial for basal transcription of the p68 gene in proliferating NIH 3T3 cells. In particular, part of the GC-rich sequence (-57/-46) and the palindrome (-81/-62) elements were necessary for promoter activity, both of which share homology with the E-box sequence. Gel mobility shift assays using NIH 3T3 nuclear extracts revealed that the upstream stimulatory factor, known as an E-box-binding protein, binds to these sites. Moreover, we observed binding of E2F to two sites near the transcription initiation site (-11/-3 and +9/+16). A transient luciferase expression assay using synchronized NIH 3T3 cells in G(0)phase revealed that these E2F sites are essential for transcription induction of the p68 gene after serum stimulation, but are dispensable for basal transcription. These results indicate that growth-dependent regulation of transcription of the mouse p68 and p180 genes is mediated by a common factor, E2F; however, basal transcription of the genes, interestingly, is regulated by different transcription factors.

The human homolog of Saccharomyces cerevisiae Mcm10 interacts with replication factors and dissociates from nuclease-resistant nuclear structures in G(2) phase.

Mcm10 (Dna43), first identified in Saccharomyces cerevisiae, is an essential protein which functions in the initiation of DNA synthesis. Mcm10 is a nuclear protein that is localized to replication origins and mediates the interaction of the Mcm2-7 complex with replication origins. We identified and cloned a human cDNA whose product was structurally homologous to the yeast Mcm10 protein. Human Mcm10 (HsMcm10) is a 98-kDa protein of 874 amino acids which shows 23 and 21% overall similarity to Schizosaccharomyces pombe Cdc23 and S. cerevisiae Mcm10, respectively. The messenger RNA level of HsMcm10 increased at the G(1)/S-boundary when quiescent human NB1-RGB cells were induced to proliferate as is the case of many replication factors. HsMcm10 associated with nuclease-resistant nuclear structures throughout S phase and dissociated from it in G(2) phase. HsMcm10 associated with human Orc2 protein when overexpressed in COS-1 cells. HsMcm10 also interacted with Orc2, Mcm2 and Mcm6 proteins in the yeast two-hybrid system. These results suggest that HsMcm10 may function in DNA replication through the interaction with Orc and Mcm2-7 complexes.

Characterization of chromosome aberrations induced by incubation at a restrictive temperature in the mouse temperature-sensitive mutant tsFT20 strain containing heat-labile DNA polymerase alpha.

tsFT20 cells derived from a mouse mammary carcinoma cell line, FM3A, which has temperature-sensitive DNA polymerase alpha activity (Y. Murakami, H. Yasuda, H. Miyazawa, F. Hanaoka, and M. Yamada, Proc. Natl. Acad. Sci. USA, 82:1761-1765, 1985) were rapidly committed to death after temperature upshift to 39 degrees C. tsFT20 cells synchronized in S phase were more sensitive to the restrictive temperature than exponentially growing cells. In order to gain insight into the processes from the interruption of DNA synthesis to cell death, we analyzed chromosome aberrations induced in tsFT20 cells which had been incubated for 2 or 4 h at the restrictive temperature and then cultured at the permissive temperature. The majority of metaphase cells showed extensive chromosome aberrations such as chromatid gaps, breaks, and exchanges; chromosome pulverizations; their mixed types; and ring chromosomes. Analyses with the use of cell synchronization and autoradiography revealed that chromosome aberrations were induced only in the cells which synthesized DNA during incubation at 39 degrees C. We classified the chromosome aberrations into five types: gap or break type; exchange type; pulverization type; complex type; and ring type. The temporal order of the appearance of these types of chromosome aberrations was found to be the above described order. It was further found that cycloheximide dramatically repressed the induction of chromosome aberrations, and metaphases with many chromosome aberrations exhibited a large number of sister chromatid exchanges. These results indicate that abnormal cessation of DNA replication in tsFT20 cells at the restrictive temperature due to the inactivation of DNA polymerase alpha results in cell death via induction of double-strand breaks which lead to chromosome aberrations as well as sister chromatid exchanges.

Genomic structure, chromosomal localization and identification of mutations in the xeroderma pigmentosum variant (XPV) gene.

The xeroderma pigmentosum variant (XP-V) is one of the most common forms of this cancer-prone syndrome. XP groups A through G are characterized by defective nucleotide excision repair, whereas the XP-V phenotype is proficient in this pathway. The XPV gene encodes DNA polymerase eta, which catalyzes an accurate translesion synthesis, indicating that the XPV gene contributes tumor suppression in normal individuals. Here we describe the genomic structure and chromosomal localization of the XPV gene, which includes 11 exons covering the entire coding sequence, lacks a TATA sequence in the upstream region of the transcription-initiation, and is located at the chromosome band 6p21.1-6p12. Analyses of patient-derived XP-V cell lines strongly suggested that three of four cell lines carried homozygous mutations in the XPV gene. The fourth cell line, XP1RO, carried heterozygous point mutations in the XPV gene, one of which was located at the splice acceptor site of exon 2, resulting in the omission of exon 2 from the mature mRNA. These findings provide a basis for diagnosis and therapy of XP-V patients.

The c-myb proto-oncogene product binds to but does not activate the promoter of the DNA polymerase alpha gene.

The myb proto oncogene product (c-Myb) is a transcriptional regulator and its expression and function are tightly linked to the cellular entry into S phase and DNA synthesis. It has been shown [Venturelli, D., Travali, S. & Calabretta, B. (1990). Proc. Natl. Acad. Sci. USA, 87, 5963-5967] that inhibition of T-cell proliferation by a myb antisense oligomer is accompanied by down-regulation of DNA polymerase alpha expression. To examine whether the transcription of the DNA polymerase alpha gene is directly regulated by c-Myb, we have identified and characterized the 5' regulatory region of the human DNA polymerase alpha gene. Two major and several minor transcription start sites were identified by nuclease S1 mapping. DNA sequence analysis showed that the promoter region contains no TATA box, one CCAAT box and putative Ap-1, AP-2 and E2F binding sites. In DNAase I footprinting, the bacterially expressed c-Myb protected six sites in the 5' flanking region of the human DNA polymerase alpha gene. However, c-Myb did not activate the DNA polymerase alpha gene promoter in a co-transfection assay. Our results suggest that an unknown factor(s) is required for the c-Myb-induced activation of the DNA polymerase alpha gene promoter, or c-Myb does not directly activate this promoter.

Abrogation of p53-mediated transactivation by SV40 large T antigen.

p53 is known to bind specifically to the 44-bp human DNA sequence in an immunoprecipitation assay. We show here that the transcription of the reporter CAT gene linked with the herpesvirus thymidine kinase (tk) promoter containing the 44-base sequence is enhanced by mouse wild-type but not mutant-type p53 in F9 and p53-null Saos-2 cells. The p53-mediated transactivation was dramatically abrogated by introduction of SV40 large T antigen (SVLT) in Saos-2 cells in which p53 was clearly associated with SVLT. Furthermore, the p53-SVLT complex did not bind to the 44-base sequence at all. Thus, SVLT sequesters the transactivation function of the wild-type p53 by inhibiting the binding of p53 to the 44-base sequence. This is good evidence to show 'loss of functions' in the product of a tumor-suppressor oncogene by a dominant oncogene product at a molecular level.

DNA repair synthesis in HeLa cell lysate.

In order to study the mechanism of DNA repair, we established an in vitro system for repair (unscheduled) DNA synthesis. HeLa cells synchronized at G2-G1 phase were irradiated with ultraviolet light in the presence of two DNA replication inhibitors, hydroxyurea and 1-beta-D-arabinofuranosyl cytosine (araCyt), to reduce the replicative DNA synthesis as much as possible. Hypotonic treatment of the cells was followed by gentle homogenization, and the resulting cell lysate was incubated with [3H]dTTP. The lysate system required all four dXTPs and Mg2+, but required no ATP. The incorporation of [3H]dTTP was dependent on the dose of ultraviolet light, was linear for 2 min, and reached the maximum at 5 min. The presence of hydroxyurea and araCyt during in vivo incubation was necessary for in vitro DNA synthesis. Accumulation of single-strand breaks was observed under these conditions, and this could explain the very high incorporation of [3H]dTTP in this system.

Transcription of the catalytic 180-kDa subunit gene of mouse DNA polymerase alpha is controlled by E2F, an Ets-related transcription factor, and Sp1.

We have isolated a genomic DNA fragment spanning the 5'-end of the gene encoding the catalytic subunit of mouse DNA polymerase alpha. The nucleotide sequence of the upstream region was G/C-rich and lacked a TATA box. Transient expression assays in cycling NIH 3T3 cells demonstrated that the GC box of 20 bp (at nucleotides -112/-93 with respect to the transcription initiation site) and the palindromic sequence of 14 bp (at nucleotides -71/-58) were essential for basal promoter activity. Electrophoretic mobility shift assays showed that Sp1 binds to the GC box. We also purified a protein capable of binding to the palindrome and identified it as GA-binding protein (GABP), an Ets- and Notch-related transcription factor. Transient expression assays in synchronized NIH 3T3 cells revealed that three variant E2F sites near the transcription initiation site (at nucleotides -23/-16, -1/+7 and +17/+29) had no basal promoter activity by themselves, but were essential for growth-dependent stimulation of the gene expression. These data indicate that E2F, GABP and Sp1 regulate the gene expression of this principal replication enzyme.

A new Drosophila ultraviolet light-damaged DNA recognition endonuclease that selectively nicks a (6-4) photoproduct site.

We have previously described the purification of an ultraviolet light (UV) damage-specific DNA-binding protein from Drosophila melanogaster, designated D-DDB P1 [Nucleic Acids Res., 23 (1995) 2600-2607]. Here, we obtained highly purified D-DDB P1 from Drosophila Kc cells, and we found that D-DDB P1 is also a nuclease. D-DDB P1 can selectively bind to pyrimidine (6-4) pyrimidone photoproducts, and in the presence of Mg++, D-DDB P1 can catalyze an incision immediately on the 3' and 5' sides of the (6-4) photoproduct site.

Error rate and specificity of human and murine DNA polymerase eta.

We describe here the error specificity of mammalian DNA polymerase eta (pol eta), an enzyme that performs translesion DNA synthesis and may participate in somatic hypermutation of immunoglobulin genes. Both mouse and human pol eta lack intrinsic proofreading exonuclease activity and both copy undamaged DNA inaccurately. Analysis of more than 1500 single-base substitutions by human pol eta indicates that error rates for all 12 mismatches are high and variable depending on the composition and symmetry of the mismatch and its location. pol eta also generates tandem base substitutions at an unprecedented rate, and kinetic analysis indicates that it extends a tandem double mismatch about as efficiently as other replicative enzymes extend single-base mismatches. This ability to use an aberrant primer terminus and the high rate of single and double-base substitutions support the idea that pol eta may forego strict shape complementarity in order to facilitate highly efficient lesion bypass. Relaxed discrimination is further indicated by pol eta infidelity for a wide variety of nucleotide deletion and addition errors. The nature and location of these errors suggest that some may be initiated by strand slippage, while others result from additional mechanisms.