Requirement of DNA Polymerase Activity of Yeast Rad30 Protein for Its Biological Function*

The RAD30 gene of Saccharomyces cerevisiae encodes a DNA polymerase, Polη. The Rad30 protein shares homology with the yeast Rev1 and the Escherichia coli DinB and UmuC proteins. Although these proteins contain several highly conserved motifs, only Rad30 has been shown to possess a DNA polymerase activity. To determine whether the DNA polymerase activity of Rad30 was essential for its biological function, we made a mutation in the highly conserved SIDE sequence in Rad30, in which the aspartate and glutamate residues have each been changed to alanine. The mutant Rad30 protein lacks the DNA polymerase activity, and the mutant gene does not complement the rad30Δ mutation. These findings indicate that DNA polymerase activity is indispensable for the biological function of RAD30.


From the Sealy Center for Molecular Science, University of Texas Medical Branch, Galveston, Texas 77555-1061
The RAD30 gene of Saccharomyces cerevisiae encodes a DNA polymerase, Pol. The Rad30 protein shares homology with the yeast Rev1 and the Escherichia coli DinB and UmuC proteins. Although these proteins contain several highly conserved motifs, only Rad30 has been shown to possess a DNA polymerase activity. To determine whether the DNA polymerase activity of Rad30 was essential for its biological function, we made a mutation in the highly conserved SIDE sequence in Rad30, in which the aspartate and glutamate residues have each been changed to alanine. The mutant Rad30 protein lacks the DNA polymerase activity, and the mutant gene does not complement the rad30⌬ mutation. These findings indicate that DNA polymerase activity is indispensable for the biological function of RAD30.
UV-induced DNA damage presents a block to the DNA replication machinery. To maintain the continuity of the DNA during replication, UV lesions encountered by the replication machinery are circumvented by both error-free and error-prone means. In the yeast Saccharomyces cerevisiae, genes in the RAD6 epistasis group function in the replication of DNA-containing lesions generated by UV light and by other DNA damaging agents. Mutations in the RAD6 and RAD18 genes confer extreme sensitivity to UV light, and these mutants are defective in postreplicative bypass of UV-damaged DNA and in UV-induced mutagenesis (1). Rad6, a ubiquitin-conjugating enzyme, exists in vivo in a complex with Rad18, a DNA-binding protein (2,3). How Rad6-Rad18 protein-dependent ubiquitination promotes error-free and mutagenic postreplicative bypass is not known.
Of the genes in the RAD6 epistasis group, REV1, REV3, and REV7 are required for mutagenic bypass of UV damage, and yeast lacking any of these genes is nonmutable by UV light (1). The Rev3 and Rev7 proteins form DNA polymerase , which shows limited ability to bypass a cis-syn thymine-thymine dimer (4). Rev1 is a deoxycytidyl transferase that can incorporate a dCMP residue opposite an abasic site (5). The RAD5 gene is required for error-free postreplicative bypass of UV lesions, and it encodes a DNA-dependent ATPase (6,7). The RAD30 gene affects an alternate pathway of error-free bypass of UV lesions, and the rad5⌬ rad30⌬ double mutant exhibits a synergistic increase in UV sensitivity over either single mutant (8). Rad30 shares homology with the yeast Rev1 protein and with the Escherichia coli DinB and UmuC proteins (8,9). We have recently shown that RAD30 encodes a novel eukaryotic DNA polymerase, named Pol, which has the unique ability to efficiently replicate a cis-syn thymine-thymine dimer-containing template, and it inserts two A residues across from the dimer (10).
The presence of biochemical activity in a protein does not necessarily imply the requirement of that activity in the biological function of the protein. For example, the deoxycytidyl transferase activity of Rev1 seems to have no role in the bypass of UV-damaged DNA templates (5). The S. cerevisiae Rad3 protein and its human counterpart, XPD, both possess DNA helicase activity (11,12), and they are required for nucleotide excision repair and for RNA polymerase II transcription. Mutational inactivation of the DNA helicase activity of these proteins, however, impairs only the repair function and not the transcription function (13)(14)(15). Thus, the DNA helicase activities of Rad3 and XPD are required for DNA repair but not for transcription. Here, we examine whether the DNA polymerase activity of Rad30 is necessary for its role in damage bypass. For this purpose, we altered the aspartate and glutamate residues present in the highly conserved domain of serine, isoleucine, aspartate, and glutamate (SIDE) in Rad30 to alanines. The resulting Rad30 mutant protein lacks DNA polymerase activity, and this mutation inactivates the biological function of RAD30. Thus, the DNA polymerase activity of Rad30 is indispensable for its role in damage bypass.

MATERIALS AND METHODS
Yeast Strains and Plasmids-Genetic studies were done using yeast strain EMY74.7 (MATa his3⌬-100, leu2-3, 112, trp1⌬, ura3-52) and its derivatives. Genomic deletions of RAD genes were generated by the gene replacement method. To generate the rad30⌬ mutation, 1.4-and 1.1-kilobase PCR 1 products corresponding to the 5Ј-and 3Ј-flanking regions of the RAD30 gene, respectively, were directionally cloned into pUC19. The URA3 "geneblaster fragment," which contains the yeast URA3 gene flanked by the duplicated Salmonella typhimurium hisG gene (16), was then inserted in between these PCR products. The rad30 deletion generating plasmid, pR30.2, when digested with the restriction enzyme EcoRI, releases a 6.4-kilobase fragment, which when introduced into yeast deletes nucleotides from position ϩ42 to position ϩ1800 of the 1896-nucleotide RAD30 open reading frame. To generate the rad5⌬ mutation, yeast strains were transformed with the rad5⌬ generating plasmid pBJ22 digested with XbaI as described (6). Deletions were confirmed by PCR analysis of genomic DNA. Loss of the URA3 gene was selected for by plating strains on medium containing 5-flouro-orotic acid.
Mutation of the RAD30 Gene and Plasmid Constructions-To generate the rad30 Asp 155 3 Ala, Glu 156 3 Ala mutation, we first isolated the wild type RAD30 gene from the yeast genome by gap repair. The wild type gene was then cloned into pUC19, generating plasmid pBJ579. To create the DE 3 AA mutation, the MORPH site-specific mutagenesis kit (5 Prime 3 3 Prime, Inc., Boulder, CO) was employed using the mutagenic oligonucleotide 5Ј-GTCGAAAGGGCGACTATTGCTGCAG-TATTTCTTGATTTGGG-3Ј, which contains the codons for amino acids AA instead of DE at positions 155 and 156, respectively, in RAD30. To increase the yield of mutant-containing plasmid, pBJ579 was grown in dut ung E. coli in medium containing uracil and was used as template for DNA synthesis from the mutagenic primer. The resulting DNA was ligated and introduced into mutS Ϫ E. coli strain BMH 71-18, which cannot repair mismatches and degrades the uracil-containing template. Plasmid DNA containing the rad30 DE 3 AA mutation was isolated, and the presence of the mutation was confirmed by sequencing. The DNA fragment containing the rad30 mutation was then used to replace the wild type fragment in the pUC19-derived pBJ579, generating plasmid pBJ639. To overexpress the mutant protein, the rad30 Ala 155 -Ala 156 mutant gene was fused in-frame with the glutathione S-transferase gene under control of the galactose-inducible phosphogylcerate kinase promoter, generating plasmid pBJ643. Plasmids pBJ640 and pBJ646 contain the wild type and rad30 Ala 155 -Ala 156 mutant genes, respectively, in the low copy CEN plasmid YCplac111. These constructs contain approximately 800 nucleotides of 5Ј-flanking RAD30 sequence.
Purification of rad30 Ala 155 -Ala 156 Mutant Protein-Rad30 Ala 155 -Ala 156 mutant protein was purified from the protease-deficient yeast strain BJ5464 as described previously for the wild type Rad30 protein (10).
UV Sensitivity and UV Mutagenesis-Yeast strains were grown to mid-exponential phase in selective medium, washed, sonicated to disperse cell clumps when necessary, and resuspended in sterile distilled water to a density of 2 ϫ 10 8 cells/ml. Cell suspensions were diluted, spread onto the appropriate medium, and irradiated at a dose rate of 1 J/m 2 /s. Plates were incubated in the dark, and colonies were counted after 3-5 days. For UV-induced mutagenesis, cells were processed as described above, and appropriate dilutions were plated on synthetic complete medium for viability determinations and on synthetic medium lacking arginine but supplemented with canavanine for determination of can1 r mutation frequencies.

Conserved Domains in Rad30 and Related Proteins-Rad30
shares significant homology with the S. cerevisiae Rev1 protein and the E. coli UmuC and DinB proteins (8,9). Previous studies have indicated that Rev1 is a deoxycytidyl transferase that transfers a dCMP residue to the 3Ј-end of a DNA primer in a template-dependent reaction (5) and that Rad30 is a DNA polymerase that can efficiently replicate DNA containing a thymine-thymine dimer (10). The E. coli UmuC and UmuDЈ proteins promote damage bypass by DNA polymerase III, but the mechanism of their action remains to be elucidated (17,18). DinB is required for the untargeted mutagenesis of unirradiated phage grown in pre-UV irradiated E. coli cells (19).
Alignment of the amino acid sequences of Rad30 and its related proteins indicates the presence of five conserved motifs, I-V (Fig. 1). Of particular interest are motif I, which contains a conserved aspartate residue flanked by conserved hydrophobic residues on both sides, and motif III, which contains the highly conserved sequence SIDE. Even though the Rad30 family of proteins shares no sequence homology with any other known prokaryotic or eukaryotic DNA polymerases, motif III resembles motif C, which is common to the entire polymerase family and which also contains an invariant aspartate residue and another highly conserved acidic residue (20). Motif I of the Rad30 family may be analogous to motif A, which also is common to the entire polymerase family and which contains an invariant aspartate residue (20). Mutational studies of these three conserved acidic residues in motifs A and C of various polymerases have indicated a crucial role of these acidic residues in catalysis (20).
The Rad30 Ala 155 -Ala 156 Mutant Protein Lacks DNA Polymerase Activity-Because of the high degree of conservation of the SIDE sequence present in Rad30 at residues 153-156 and because of the similarity of this domain to motif C of other polymerases, we changed the Asp 155 and Glu 156 residues of Rad30 to alanines. Similar to the previously reported work for the wild type Rad30 protein, the mutant rad30 gene was fused in frame downstream of the GST gene expressed in yeast from the galactose-inducible phosphogylcerate kinase promoter, and the resulting fusion protein was purified from a protease-deficient yeast strain harboring the GST-Rad30 mutant plasmid pBJ643. During purification, the mutant protein displayed the same chromatographic properties as the wild type protein, and in SDS-polyacrylamide gel electrophoresis, both proteins exhibit identical electrophoretic mobility ( Fig. 2A). With all four DNA substrates S-1, S-2, S-3, and S-4, in which a 75-nucleotide oligomer template has been annealed to 40-, 41-, 42-, and 43-nucleotide 5Ј 32 P-labeled oligomer primers, respectively, the wild type Rad30 protein carries out extensive DNA synthesis in the presence of all four dNTPs (Fig. 2B, lanes 1-4). In striking contrast to the robust DNA polymerase activity of the wild type Rad30 protein, the Rad30 Ala 155 -Ala 156 mutant protein shows no DNA polymerase activity (Fig. 2B, lanes 5-8), even when four times as much mutant protein was used as the wild type protein. In fact, the mutant protein was unable to add even a single nucleotide onto the primer strand of any of the DNA substrates. Thus, the rad30 Ala 155 -Ala 156 mutation completely inactivates the DNA polymerase activity of Rad30.
The rad30 Ala 155 -Ala 156 Mutation Inactivates the Biological Function of Rad30 -Previous genetic studies have indicated a role of Rad30 in error-free bypass of UV lesions. The rad5⌬ rad30⌬ double mutant exhibits a synergistic increase in UV sensitivity compared with the rad5⌬ and rad30⌬ single mutants, and the rate of UV-induced reversion of the trp1-1 allele is greatly enhanced in the double mutant over that in either single mutant (8). These observations have suggested that Rad5 and Rad30 constitute alternate Rad6-Rad18-dependent pathways for the error-free bypass of UV-damaged DNA templates.
To determine whether Rad30 DNA polymerase activity was required for UV damage bypass, the ability of the rad30 Ala 155 -Ala 156 mutation to complement the rad30⌬ mutation was exam- ined. The wild type or the mutant gene was expressed in yeast from the native RAD30 promoter on a low copy CEN plasmid. As expected, the wild type RAD30 gene fully complemented the UV sensitivity of the rad30⌬ mutation and restored UV survival of the rad30⌬ rad5⌬ strain to the rad5⌬ level (data not shown). The rad30 Ala 155 -Ala 156 mutation, however, was unable to complement the UV sensitivity of either the rad30⌬ strain or the rad5⌬ rad30⌬ strain (Fig. 3A). To determine the effect of the rad30 Ala 155 -Ala 156 mutation on UV mutagenesis, we examined the rate of forward mutations at the CAN1 S locus. As shown in Fig.  3B, the incidence of UV-induced can1 r mutations was higher in the rad30⌬ strain than in the wild type or in the rad5⌬ strain. Up to the UV dose of 7.5 J/m 2 , the frequency of can1 r mutations was about the same in the rad5⌬ rad30⌬ strain as in the rad30⌬ strain, but at 10 J/m 2 , the rad5⌬ rad30⌬ strain displays a sharp rise in the frequency of can1 r mutations. Introduction of the rad30 Ala 155 -Ala 156 mutant gene in the rad30⌬ strain or in the rad5⌬ rad30⌬ strain had no effect on the incidence of UV-in-duced can1 r mutations. As expected, the wild type RAD30 gene lowered the frequency of UV-induced mutations in the rad30⌬ strain to the wild type level and in the rad5⌬ rad30⌬ strain to the rad5⌬ level (data not shown).
In summary, inactivation of the DNA polymerase activity of Rad30 causes complete loss of the biological function of this protein, thus indicating the requirement of this DNA polymerase activity in damage bypass. Our studies also suggest the possibility that the aspartate 155 and glutamate 156 residues present in the highly conserved sequence SIDE in Rad30 play an important role in catalysis. It is possible that these residues are part of the active site of the enzyme and that they coordinate the binding of divalent metal ions.