JB
Home Help [Feedback] [For Subscribers] [Archive] [Search] [Contents]
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Other Versions of this Article:
JB.01481-06v1
188/24/8573    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sanders, L. H.
Right arrow Articles by Sutton, M. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sanders, L. H.
Right arrow Articles by Sutton, M. D.
Journal of Bacteriology, December 2006, p. 8573-8585, Vol. 188, No. 24
0021-9193/06/$08.00+0     doi:10.1128/JB.01481-06
Copyright © 2006, American Society for Microbiology. All Rights Reserved.

Role of Pseudomonas aeruginosa dinB-Encoded DNA Polymerase IV in Mutagenesis{triangledown}

Laurie H. Sanders,1,{dagger} Andrea Rockel,2,{dagger} Haiping Lu,2 Daniel J. Wozniak,2 and Mark D. Sutton1*

Department of Biochemistry, School of Medicine and Biomedical Sciences, University at Buffalo, State University of New York, 3435 Main Street, 140 Farber Hall, Buffalo, New York 14214,1 Department of Microbiology and Immunology, Wake Forest University School of Medicine, Medical Center Boulevard, Winston-Salem, North Carolina 271572

Received 19 September 2006/ Accepted 5 October 2006


   ABSTRACT
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Pseudomonas aeruginosa is a human opportunistic pathogen that chronically infects the lungs of cystic fibrosis patients and is the leading cause of morbidity and mortality of people afflicted with this disease. A striking correlation between mutagenesis and the persistence of P. aeruginosa has been reported. In other well-studied organisms, error-prone replication by Y family DNA polymerases contributes significantly to mutagenesis. Based on an analysis of the PAO1 genome sequence, P. aeruginosa contains a single Y family DNA polymerase encoded by the dinB gene. As part of an effort to understand the mechanisms of mutagenesis in P. aeruginosa, we have cloned the dinB gene of P. aeruginosa and utilized a combination of genetic and biochemical approaches to characterize the activity and regulation of the P. aeruginosa DinB protein (DinBPa). Our results indicate that DinBPa is a distributive DNA polymerase that lacks intrinsic proofreading activity in vitro. Modest overexpression of DinBPa from a plasmid conferred a mutator phenotype in both Escherichia coli and P. aeruginosa. An examination of this mutator phenotype indicated that DinBPa has a propensity to promote C->A transversions and –1 frameshift mutations within poly(dGMP) and poly(dAMP) runs. The characterization of lexA+ and {Delta}lexA::aacC1 P. aeruginosa strains, together with in vitro DNA binding assays utilizing cell extracts or purified P. aeruginosa LexA protein (LexAPa), indicated that the transcription of the dinB gene is regulated as part of an SOS-like response. The deletion of the dinBPa gene sensitized P. aeruginosa to nitrofurazone and 4-nitroquinoline-1-oxide, consistent with a role for DinBPa in translesion DNA synthesis over N2-dG adducts. Finally, P. aeruginosa exhibited a UV-inducible mutator phenotype that was independent of dinBPa function and instead required polA and polC, which encode DNA polymerase I and the second DNA polymerase III enzyme, respectively. Possible roles of the P. aeruginosa dinB, polA, and polC gene products in mutagenesis are discussed.


   INTRODUCTION
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Various endogenous and exogenous agents can damage DNA. To guard against this damage, organisms possess various mechanisms devoted to DNA repair, including base excision repair, nucleotide excision repair, mismatch repair (MMR), and homologous recombination (reviewed in reference 22). Despite the remarkable proficiency with which these repair functions identify and correct damaged DNA, lesions sometimes evade repair and must therefore be dealt with during DNA replication. In these cases, the DNA lesion is generally tolerated via postreplicative gap repair or translesion DNA synthesis (TLS). Postreplicative gap repair refers to a class of mechanisms in which the replication machinery utilizes the information in the nascent daughter strand to replicate past the lesion in a high-fidelity manner (reviewed in references 13, 20, and 22). In contrast, TLS involves direct replication over the site(s) of damage (reviewed in references 20, 22, and 56). Since replicative DNA polymerases (Pols) are unable to efficiently extend imperfect primer/template termini, TLS in general cannot be catalyzed by these Pols and is therefore dependent upon one or more specialized Pols capable of replicating over damaged DNA bases. Most Pols capable of participating in TLS belong to the Y family of DNA polymerases (reviewed in references 19, 21, 45, and 56). Y family Pols are typically distributive, lack proofreading activities, and possess fidelities lower than those of well-studied replicative Pols. Although Y family Pols, such as the human Pol {eta} (XP-V Pol), can replicate UV-induced DNA lesions with relatively high fidelity (reviewed in references 29, 30, 41, 42, 48, and 64), many replication-blocking lesions are noncoding. Thus, most TLS is inherently error prone.

As a consequence of its role in damage tolerance, TLS serves as a major pathway for mutagenesis. For example, following irradiation with UV light or exposure to certain chemical carcinogens, most induced mutagenesis in Escherichia coli depends on the umuDC-encoded Pol V (32, 52). E. coli Pol V is regulated as part of the SOS response. In this response, the LexA protein represses the transcription of roughly 40 different genes, including umuDC (12, 22). Following replication-blocking DNA damage, the E. coli RecA protein facilitates the autodigestion of LexA, leading to the transcriptional derepression of the LexA-regulated genes (38). TLS is also believed to significantly contribute to mutagenesis in eukaryotes (21, 22, 24). Furthermore, error-prone replication by various low-fidelity Pols is reported to contribute to immunoglobulin diversity via somatic hypermutation (51).

A capacity for DNA damage tolerance and repair is important for ensuring high-fidelity replication of the genetic material in all organisms. In addition, recent data suggest that DNA repair and mutagenesis may also play important roles in pathogenesis. For example, for Salmonella enterica and Listeria monocytogenes, the ability to repair DNA damage appears to be important for the initial stages of infection (8-10, 57, 58, 68). For Helicobacter pylori, the mutS gene product plays an important role in pathogenesis by protecting this pathogen from oxidative stress: mutS mutants of H. pylori exhibit reduced colonization in a mouse model of infection, presumably due to their inability to repair oxidative DNA damage (62). Recently, a striking correlation between mutagenesis and the persistence of Pseudomonas aeruginosa was reported (46). P. aeruginosa is a human opportunistic pathogen that chronically infects the lungs of cystic fibrosis (CF) patients and is the leading cause of morbidity and mortality of people afflicted with this disease. Oliver et al. (46) reported that, although P. aeruginosa isolates recovered from individuals with acute infections displayed normal low-level spontaneous mutation frequencies, roughly 20% of the isolates recovered from CF patients suffering from chronic P. aeruginosa infections displayed a hypermutable phenotype. This hypermutability resulted from a loss of MMR and correlated with an increased resistance of the P. aeruginosa strains to various commonly used antimicrobial agents (46).

P. aeruginosa isolates recovered from cystic fibrosis patients were reported to bear mutations in mucA that impair its function (40). The MucA protein of P. aeruginosa acts as an anti-{sigma} factor to regulate the expression of the exopolysaccharide alginate (40, 43). The finding that P. aeruginosa isolates recovered from CF patients possessed mutations inactivating mucA suggests that the mucA allele represents a "hot spot" for mutagenesis that confers an advantage in persistence. Nucleotide sequence analysis of the mucA allele in the different mucoid P. aeruginosa isolates indicated that most inactivating mutations corresponded to a –1 frameshift mutation within a run of five deoxyguanylic acid residues (40, 43, 49). Taken together, these results suggest a critically important role for mutagenesis in the persistence of P. aeruginosa infections. However, very little is known about how P. aeruginosa regulates TLS and mutagenesis.

As part of an effort to understand the role of Y family DNA Pols in mutagenesis in P. aeruginosa, particularly frameshift mutagenesis, we have cloned the dinB gene of P. aeruginosa and utilized a combination of genetic and biochemical approaches to characterize the recombinant P. aeruginosa DinB protein (DinBPa). Our results discussed below indicate that DinBPa is a distributive DNA polymerase that is apparently devoid of an intrinsic proofreading activity. Modest overexpression of DinBPa from a plasmid conferred a mutator phenotype in both E. coli and P. aeruginosa. Careful examination of this mutator phenotype indicated that DinBPa had a propensity to promote C->A transversions and –1 frameshift mutations within poly(dGMP) and poly(dAMP) runs. The characterization of lexA+ and {Delta}lexA::aacC1 P. aeruginosa strains indicated that the transcription of the dinB gene of P. aeruginosa was regulated as part of an SOS-like response. Moreover, the deletion of the dinB allele sensitized P. aeruginosa to the DNA-damaging agents nitrofurazone (NFZ) and 4-nitroquinoline 1-oxide (4-NQO), suggesting that DinBPa plays an important role in damage-induced TLS. Finally, P. aeruginosa displayed a modest UV-induced mutator phenotype that was independent of dinB function and instead required the polA and polC genes, which encode DNA polymerase I and the second DNA polymerase III enzyme, respectively. Possible roles of P. aeruginosa DNA polymerase I (polA), the second DNA polymerase III (polC), and DNA polymerase IV (dinB) in DNA mutagenesis are discussed.


   MATERIALS AND METHODS
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Bacteriological techniques. The E. coli and P. aeruginosa strains used in this study are described in Table 1. Strains MPAO1 and MPA50455 ({Delta}polA::ISphoA/hah), MPA17464 ({Delta}polB::ISlacZ/hah), and MPA34086 ({Delta}polC::ISphoA/hah) were obtained from the University of Washington Genome Center (27). ISphoA/hah and ISlacZ/hah transposon insertion within the polA, polB, or polC gene of P. aeruginosa was confirmed by diagnostic PCR using a protocol provided by Michael Jacobs of the University of Washington Genome Center. All strains were routinely grown in Luria-Bertani (LB) medium. When necessary, the following antibiotics were used at the indicated concentrations with E. coli strains: ampicillin, 150 µg/ml; kanamycin, 60 µg/ml; and rifampin (Rif), 100 µg/ml. When required, gentamicin and rifampin were each used at 100 µg/ml with P. aeruginosa strains, while tetracycline was used at 60 µg/ml, and carbenicillin was used at 250 µg/ml.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Bacterial strains and plasmid DNAs used in this study

 
The plasmid DNAs used in this study are described in Table 1. The dinB gene (including its native promoter) of P. aeruginosa was PCR amplified from P. aeruginosa PAO1 genomic DNA using primers dinB1 (5'-CGG GAT CCG AGT TCC ATC CGG TTC ACG-3') and dinB2 (5'-CCC AAG CTT GGG ACG CCG TGC TGA AGG CC-3'). The amplified DNA fragment was digested with BamHI and HindIII and cloned into pWSK29, resulting in pHL6. For the expression of DinBPa, a PCR amplification of plasmid pHL6 with primers dinB4 (5'-GGA ATT CCA TAT GCG GAA AAT CAT CCA TAT AGA CTG-3') and dinB5 (5'-CCC AAG CTT GAA CAA CCT GAG TTG TTC GT-3') was performed. The fragment was digested with NdeI and HindIII and cloned into pET29a (Novagen), resulting in plasmid pHL8, which was used for the overproduction of the hexahistidine-tagged, wild-type DinBPa. Three individual dinB missense mutations (D8A [aspartic acid position 8 changed to alanine], R49A, and D103A) were constructed in both pHL6 and pHL8 by site-directed mutagenesis using the QuikChange mutagenesis kit (Stratagene) (Table 1).

The lexA gene was PCR amplified from P. aeruginosa PAO1 genomic DNA using primers lexA5 (5'-CCC AAG CTT GCG CCG GAT CAC GC-3') and lexA6 (5'-CGG GAT CCA GGC GAC GAC ATG CA-3'). The resulting product was digested with HindIII and BamHI and cloned into the vector pET29a for expression with a C-terminally hexahistidine-tagged P. aeruginosa LexA protein (LexAPa), resulting in pAR88.

Construction of {Delta}lexA::aacC1 and {Delta}dinB::aacC1 P. aeruginosa strains. Standard reverse genetic techniques (26, 50) were used for the construction of P. aeruginosa dinB- and lexA-null mutants. For the {Delta}dinB mutant, plasmid pHL6 was digested with BamHI and HindIII and the dinB fragment was subcloned into pEX18Ap (26), resulting in pHL13. An ~200-bp fragment within the dinB coding sequence was removed by SalI cleavage of pHL13 and replaced with a gentamicin resistance cassette derived from SalI cleavage of pMS266 (5), generating pHL16. P. aeruginosa WFPA334 ({Delta}dinB::aacC1) was generated using standard mating and sucrose selection gene replacement strategies (26, 39) of wild-type dinB in P. aeruginosa PAO1 with {Delta}dinB::aacC1 from pHL16. A similar strategy was used to generate the lexA::aacC1 mutant, WFPA340. The lexA gene was amplified from PAO1 genomic DNA using primers lexA1 (5'-CGG GAT CCG CAG GAG GTC CTC CAG GGT-3') and lexA2 (5'-CCC AAG CTT TAT TCA GGC TCT GTG CTT GGC CC-3'), digested with BamHI and HindIII, and cloned into similarly digested pEX18Ap, resulting in pHL10. A gentamicin resistance cassette derived from SphI cleavage of pMS266 was cloned into the SphI site of pHL10, within the lexA coding sequence, generating pHL11. P. aeruginosa WFPA340 ({Delta}lexA::aacC1) was generated using standard mating and sucrose selection gene replacement strategies (26, 39) of wild-type lexA in P. aeruginosa PAO1 with {Delta}lexA::aacC1 from pHL11. The sulA gene is located immediately downstream of lexA; the expression of sulA may be affected by the lexA::aacC1 allele, as we did not have to disrupt sulA prior to constructing the {Delta}lexA strain.

Purification of P. aeruginosa and E. coli DinB proteins. The P. aeruginosa wild-type DinB protein was overproduced using E. coli BL21(DE3) harboring pHL8 grown at 30°C in LB supplemented with ampicillin. When the culture reached an optical density at 595 nm (OD595) of 0.8, IPTG (isopropyl-ß-D-thiogalactopyranoside) (Promega) was added to 50 µM and growth was continued for another 3 h. Cells were harvested by centrifugation, resuspended in buffer H (50 mM NaPO4 [pH 6.8], 500 mM NaCl, 10% glycerol, 20 mM 2-mercaptoethanol, and 1 mM imidazole), and lysed by double passage through a chilled French press. The soluble fraction was clarified by centrifugation at 4°C prior to chromatography. The extract containing the C-terminally hexahistidine-tagged DinBPa protein was first applied to a 5-ml Ni-nitrilotriacetic acid (NTA) agarose column (QIAGEN). Bound protein was eluted with buffer H containing 50 mM EDTA. Fractions containing DinBPa, identified by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), were pooled and dialyzed against buffer A (25 mM Tris-HCl [pH 7.0], 10% glycerol, and 20 mM 2-mercaptoethanol) containing 25 mM NaCl prior to chromatography using a 1-ml HiTrap heparin column (GE Healthcare) equilibrated in the same buffer. Bound protein was eluted from the heparin column by using a linear gradient of 2.5% to 100% buffer B (25 mM Tris-HCl [pH 7.0], 10% glycerol, 20 mM 2-mercaptoethanol, and 1 M NaCl). Fractions containing DinBPa were pooled, dialyzed, and applied to a Mono S HR 5/5 column (GE Healthcare) equilibrated in buffer A containing 25 mM NaCl. Bound protein was eluted using a linear gradient of 2.5% to 100% buffer B. The purity of the final material was judged to be >95% based on densitometric scanning of Coomassie-stained SDS-PAGE gels. DinB(D8A)Pa and DinB(R49A)Pa were purified using the same approach. The DinB(D103A)Pa mutant protein became insoluble during its purification. Purified E. coli DinB protein bearing a C-terminal hexahistidine tag was a generous gift from Daniel Jarosz and Graham Walker (Massachusetts Institute of Technology).

In vitro primer extension assay. Replication assays (10 µl) consisted of a 5'-[32P]ATP radiolabeled (using T4 polynucleotide kinase; Promega) synthetic 20-mer (5'-AC GCC TGT AGC ATT CCA CAG-3')/100-mer (5'-AA TCC CAT ACA GAA AAT TCA TTT ACT AAC GTC TGG AAA CTC GAC AAA ACT TTA GAT CGA AAC GCT AAC TAT GAG GGG TGT CTG TGG AAT GCT ACA GGC GT-3') DNA template (gel purified) (Fig. 1A) or the same 20-mer either alone or annealed to M13mp18 single-stranded DNA (ssDNA) (Fig. 1B) at a final concentration of 5 nM in replication buffer (20 mM Tris-HCl [pH 7.5], 8 mM MgCl2, 4% glycerol, 50 mM NaCl, 5 mM dithiothreitol, 40 µg/ml bovine serum albumin) containing 133 µM of either each individual or all four ultrapure deoxynucleoside triphosphates (dNTPs) (GE Healthcare), as indicated in the legend for Fig. 1. Reactions were initiated by the addition of DinBPa or the E. coli DinB protein (DinBEc) at the indicated concentrations (see the legend for Fig. 1) for 10 or 20 min at 37°C, as noted. Reactions were quenched with EDTA, heat denatured, and electrophoresed through 10% denaturing polyacrylamide gels. Radiolabeled products were visualized using a phosphorimager with a signal-intensifying phosphorimaging screen (Bio-Rad).


Figure 1
View larger version (51K):
[in this window]
[in a new window]

  		FIG. 1. In vitro DinBPa DNA polymerase activity. (A) In vitro primer extension assays were performed as described in Materials and Methods using the 20-mer/100-mer DNA template, a schematic of which is shown at the top of panel A. Specific conditions for each assay are indicated at the bottom of the figure. (B) Assays were as described for panel A but used either the 20-mer primer alone (lanes 1 and 2) or the 20-mer primer annealed to M13mp18 ssDNA (NEB) (see schematic at the top of panel B). Conditions for each assay are indicated. See the text for further details.

 
RNA harvest and real-time RT-PCR. Total RNA was harvested from P. aeruginosa strains PAO1 (lexA+) and WFPA340 ({Delta}lexA::aacC1) using the Ambion Ribo-Pure kit according to the manufacturer's instructions. For experiments investigating DNA damage-induced expression of dinB, cultures were grown to an OD600 of 0.6 and then treated with mitomycin C (MMC; 1 µg/ml) for 2 h. RNA was then harvested and quantified for real-time reverse transcription-PCR (RT-PCR) analysis. Following purity confirmation by gel electrophoresis, 100 ng of this RNA was used as a template in a one-step real-time RT-PCR (TaqMan one-step RT-PCR kit) with primers (dinBF [5'-GGC GGT GTC GAA GCA GAT-3'] and dinBR [5'-CGA AAG CGG CTC GAT CAG-3']) and a fluorescent probe specific for dinB mRNA (dinBP [5'-CAT GCG ATC TTC CGC GAT TAT ACC GA-3']). Reactions were performed and analyzed using an ABI Prism 7000 sequence detection system. For each RNA sample, dinB mRNA levels were normalized against rpoD mRNA levels, which were determined using primers rpoDF (5'-CCT GCC GGA GGA TAT TTC C-3') and rpoDR (5'-GAT CCC CAT GTC GTT GAT CAT-3') and a fluorescent probe specific for rpoD mRNA (rpoDP, 5'-ATC CGG AAC AGG TGG AAG ACA TCA TCC-3'). Once normalized against rpoD, data were presented as a severalfold increase over wild-type (PAO1) levels of dinBPa mRNA, which were set equal to 1.0.

Western blot analysis. Purified hexahistidine-tagged DinBPa derived from pHL8 was used to elicit antibodies from New Zealand White rabbits (Covance Research Products). Anti-DinBPa serum was adsorbed against extracts derived from WFPA334 ({Delta}dinB::aacC1) and subsequently used at a dilution of 1/10,000 as the primary antibody for the detection of DinBPa by Western blotting. Horseradish peroxidase-conjugated goat anti-rabbit immunoglobulin (Pierce) was used as the secondary antibody at a dilution of 1/10,000. Western blotting was performed using either whole-cell extracts (Fig. 2) or cell extracts (Fig. 3), and purified hexahistidine-tagged DinBPa protein as a standard. Whole-cell extracts were prepared as described previously (55). For the preparation of cell extracts, P. aeruginosa strains were grown to an OD600 of 0.6, 10 ml of each culture was centrifuged, and the pellets were resuspended in 1.0 ml of FBG (100 mM NaCl, 10 mM Tris-HCl [pH 8.0], 1 mM MgCl2, 5% glycerol). Samples were subjected to sonication and centrifuged for 15 min at 14,000 rpm. Protein concentration in the supernatant fraction was quantified by the bicinchoninic acid assay. Equivalent amounts of protein from extracts were separated by 12% SDS-PAGE and analyzed by Western blotting.


Figure 2
View larger version (35K):
[in this window]
[in a new window]

  		FIG. 2. Steady-state levels of DinBPa+, DinB(D8A)Pa, DinB(R49A)Pa, and DinB(D103A)Pa. Whole-cell extracts of E. coli strain FC1237 [relevant genotype, {Delta}(dinB-yafN)::kan (54)], bearing the indicated plasmids, were prepared and separated by SDS-PAGE, and blotting was performed using anti-DinBPa antibodies as described in Materials and Methods. Lanes 1 and 2 contain 1.0 and 0.1 µg, respectively, of the purified hexahistidine-tagged DinBPa protein.

 

Figure 3
View larger version (30K):
[in this window]
[in a new window]

  		FIG. 3. Expression of dinB is DNA damage regulated. (A) Real-time RT-PCR analysis of dinB mRNA from P. aeruginosa strains PAO1 (lexA+) and WFPA340 ({Delta}lexA::aacC1). PAO1+MMC represents strain PAO1 treated for 2 h with 1 µg/ml MMC before RNA harvest. WFPA340 is the mean of five replicate experiments (range, 3.05 to 11.52), and PAO1+MMC is the mean of nine replicate experiments (range, 2.35 to 10.36). All values are normalized against the levels of rpoD (see Materials and Methods). (B) Western blot showing DinBPa levels in P. aeruginosa strains PAO1 (lexA+), WFPA340 ({Delta}lexA::aacC1), and WFPA334 ({Delta}dinB::aacC1). Cell extracts were prepared and separated by SDS-PAGE, and blotting was performed using anti-DinBPa antibodies as described in Materials and Methods. Upper panel, DinB; lower panel, nonspecific cross-reactive band used for a normalizing control. Lane 1, purified His-tagged DinBPa; lane 2, PAO1 extract; lane 3, WFPA340 extract; lane 4, PAO1+MMC extract; lane 5, WFPA334 extract.

 
EMSA. Electrophoretic mobility shift assay (EMSA) experiments were performed generally as described previously (18). DNA fragments for dinBPa EMSA were labeled by PCR amplification of P. aeruginosa PAO1 genomic DNA with primers dinB1 (above) and dinB3 (5'-CCC AAG CTT CGA GGG CGG CAT AGA AAC AGT-3') and [{alpha}-32P]dCTP following previously published protocols (3). Primers umuprom1 (5'-GCG TCG TCG CCA GAA GG-3') and umuprom2 (5'-CGT AAT CTG CTG CCG GTG-3') were used to amplify the 5' regulatory sequences of E coli umuC using E. coli MG1655 genomic DNA. Cell extracts, purified LexAEc, or LexAPa was incubated with the DNA fragments in a 5-µl final volume for 30 min in binding buffer (10 mM HEPES-NaOH [pH 8.0], 10 mM Tris-HCl [pH 8.0], 5% glycerol, 50 mM NaCl, 1 mM EDTA, and 1 mM dithiothreitol). A 6% nondenaturing polyacrylamide gel was prerun for 1 h at 50 V at 4°C with electrophoresis buffer (25 mM Tris-HCl [pH 8.0], 250 mM glycine, 1 mM EDTA). Samples were loaded and resolved overnight at 50 V at 4°C. The gel was dried and visualized using a phosphorimager with a signal-intensifying phosphorimaging screen (Bio-Rad). LexAEc was kindly provided by John Little (University of Arizona). Hexahistidine-tagged LexAPa protein derived from JM109(DE3)(pAR88) was purified using Ni-NTA agarose resin (QIAGEN). Bound protein was eluted from Ni-NTA agarose using imidazole. LexAPa purified in this fashion contained two distinct bands of approximately 26 and 14 kDa, which likely represent mature and cleaved LexA, respectively.


   RESULTS
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
The P. aeruginosa dinB gene encodes a functional DNA polymerase (PolPa IV). Based on the Cluster of Orthologous Groups of the National Center for Biotechnology Information, the P. aeruginosa genome contains a single ortholog of E. coli umuC and dinB designated PA0923 (53). As PA0923 resembled E. coli dinB more closely than it did E. coli umuC, it was suggested to be a dinB (DNA Pol IV) ortholog. The predicted P. aeruginosa DinB protein (DinBPa) is 342 amino acids (aa) and exhibits 49% identity with DinBEc (340 aa). Moreover, residues D8, R49, and D103, which are essential for DinBEc activity (60) are completely conserved in DinBPa. In order to determine whether DinBPa possessed an intrinsic nucleotidyl transferase activity, the C-terminally hexahistidine-tagged protein was purified to apparent homogeneity and assayed in an in vitro primer extension assay (see Materials and Methods). Using a synthetic 20-mer primer annealed to a 100-mer as a template (20/100 template) (Fig. 1A), DinBPa was able to incorporate dNTPs with efficiencies comparable to that observed with DinBEc (Fig. 1A, compare lanes 1 to 3 and 9 to 11). Interestingly, in assays utilizing the same 20-mer primer annealed to ssM13 viral DNA as a template, significantly higher concentrations of DinBPa were required in order to visualize polymerase activity (Fig. 1, compare lanes 3 to 5 of panel B with lanes 9 to 11 of panel A), suggesting that it was inefficient at binding to the primer-template junction if challenged with extensive single-strand DNA. This is reminiscent of reports describing the DinBEc enzyme (61). Importantly, DinBPa nucleotidyl transferase activity was dependent upon the template DNA (Fig. 1B, compare lanes 1 and 2), indicating that it was a DNA polymerase and not a terminal transferase. The absence of primer degradation in the reaction containing only DinBPa and the 20/100 DNA substrate (Fig. 1A, lane 12) suggests that DinBPa, like DinBEc (Fig. 1A, lane 4), is devoid of an intrinsic 3'-to-5' exonuclease (proofreading) activity.

The accuracy of nucleotide incorporation by DinBPa in vitro on an undamaged template was investigated qualitatively by assessing the efficiency of the template-directed incorporation of each single dNTP. Using the 20/100 as a template, DinBPa efficiently catalyzed the incorporation of dATP opposite template dTMP (Fig. 1A, lane 13). In contrast, DinBPa did not incorporate dTTP, dGTP, or dCTP opposite template dTMP (Fig. 1A, lanes 14 to 16). In the presence of all four nucleotides, DinBPa extended the primer in what appeared to be a largely distributive manner (Fig. 1A, lane 11). Similar results were observed for DinBEc (Fig. 1A, lanes 3 and 5 to 8), consistent with previous reports (60).

Amino acid substitutions at residues D8, R49, and D103 of DinBEc were reported to severely impair its catalytic activity both in vitro and in vivo (60). Using site-directed mutagenesis, we replaced these corresponding residues in DinBPa with alanine. The DinB(D8A)Pa and DinB(R49A)Pa mutant proteins purified in a manner that was indistinguishable from that of the wild-type DinBPa. In contrast, the DinB(D103A)Pa mutant protein became insoluble during its purification and was therefore not evaluated for in vitro activity. Although no polymerase activity was detected with the DinB(D8A)Pa mutant protein in vitro (Fig. 1A, lanes 17 to 18), the DinB(R49A)Pa mutant protein retained detectable, albeit significantly reduced, polymerase activity (Fig. 1A, lanes 19 to 21).

Elevated expression of DinBPa confers a mutator phenotype in vivo. Overexpressed levels of DinBEc from a plasmid were reported to increase the frequency of mutagenesis in E. coli as measured using a Rifr assay (28, 34, 61, 66, 67). We therefore hypothesized that if DinBPa was similarly error prone, then plasmid-expressed DinBPa would likewise increase the frequency of Rifr in E. coli. The expression of dinB of P. aeruginosa from a low-copy-number plasmid (pHL6) conferred a 4.7-fold increase in the frequency of Rifr relative to the same strain bearing the pWSK29 control vector (Table 2). In contrast, neither plasmid-expressed DinB(D8A)Pa nor DinB(R49A)Pa enhanced the frequency of Rifr (Table 2). Given that the D8A and R49A mutants were expressed at steady-state levels comparable to that of the wild-type protein (Fig. 2), these findings indicate that the modest Rifr mutator phenotype conferred by ectopically expressed DinBPa was attributable to its intrinsic DNA polymerase activity. We also intended to test whether the DinB(D103A)Pa mutant protein was impaired for the Rifr mutator phenotype, but we determined that this mutant protein was present at a reduced steady-state level compared to that of the wild-type DinBPa or to those of the D8A and R49A mutants (Fig. 2). This finding, taken together with our difficulties in purifying DinB(D103A)Pa, suggested that this mutant protein was misfolded when expressed in E. coli. Therefore, DinB(D103A)Pa was not examined using this approach.


View this table:
[in this window]
[in a new window]

  		TABLE 2. Effect of P. aeruginosa dinB overexpression on the frequency of Rifr in E. coli

 
In order to determine the spectrum of mutations made by DinBPa, we utilized a lacZ->Lac+ reversion assay developed by Cupples and Miller (14, 15). This assay takes advantage of a series of isogenic strains bearing different mutant lacZ alleles. The ability of these strains to grow on minimal medium containing lactose as a sole carbon source requires that the lacZ allele (either a base substitution or a frameshift mutation) undergo a "true" reversion, as no other mutation will restore the Lac+ phenotype (14, 15). By comparing frequencies of lacZ->Lac+ reversion for transformants of these reporter strains bearing either the pWSK29 control vector or the pHL6, which expresses DinBPa+, we were able to determine the spectrum of mutations generated by ectopic expression of DinBPa. As summarized in Table 3, DinBPa clearly favored –1 frameshift mutations: the frequency of Lac+ reversion was increased 339- and 3-fold in the CC108 (lacZ[6G->5G]) and the CC111 (lacZ[7A->6A]) strains, respectively, relative to the pWSK29 control (P values of 0.001 and 0.036, respectively). Furthermore, DinBPa also increased by 117-fold the frequency of Lac+ reversion in the CC104 (lacZ[GCG->GAG]) strain (P value 0.056). Frequencies of Lac+ reversion observed with the other reporter strains bearing the DinBPa-expressing plasmid (pHL6) were comparable to those of the respective strains bearing pWSK29 (Table 3).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Abilities of plasmid-expressed DinBPa+, DinB(D8A)Pa, and DinB(R49A)Pa mutants to influence frequencies of lacZ->Lac+ reversion of different lacZ alleles

 
We next asked whether the DNA polymerase activity of DinBPa was necessary for increased lacZ->Lac+ reversion. Due to the fact that DinBPa+ enhanced lacZ->Lac+ reversion in the CC104 and CC108 strains more strongly than in the other strains examined (Table 3), we focused on these two strains for this experiment. As summarized in Table 3, the D8A mutant was unable to significantly alter the frequency of lacZ->Lac+ reversion in both the CC104 and the CC108 strains. Despite its inability to confer a mutator phenotype in E. coli as measured by Rifr (Table 2), the R49A mutant nonetheless enhanced the frequency of lacZ->Lac+ reversion in both the CC104 and CC108 strains by factors of 53- and 19-fold, respectively: these increases were statistically significant (P values of 0.012 and 0.005, respectively) and are consistent with the R49A mutant retaining partial catalytic activity (Fig. 1A, lanes 19 to 21). Taken together, these findings indicate that DinBPa is capable of catalyzing both misinsertion and frameshift mutations.

As DinBPa has the ability to confer a mutator phenotype in E. coli, we next investigated whether DinB contributes to mutagenesis in P. aeruginosa. To do this, we measured the frequency of Rifr of PAO1 bearing either a control plasmid (pUCP20T) or a plasmid directing expression of dinBPa (pAR101). As shown in Table 4, the overexpression of dinBPa resulted in a 2.6-fold increase in mutation frequency of PAO1.


View this table:
[in this window]
[in a new window]

  		TABLE 4. Effect of P. aeruginosa dinB overexpression on the frequency of Rifr in P. aeruginosa

 
LexA-mediated repression of dinBPa. As DinBPa is able to contribute to mutagenesis in P. aeruginosa when overexpressed (Table 4), we were interested in investigating the regulation of dinBPa. In E. coli and related bacteria, TLS is largely regulated by the SOS response (19, 21, 56). As part of this response, the LexA protein acts to repress the transcription of more than 40 different genes by binding to DNA sequences overlapping their promoter regions (12, 38). Following replication-blocking DNA damage, the LexA protein undergoes a RecA-facilitated autodigestion that serves to inactivate its DNA binding activity, leading to the transcriptional derepression of the SOS-regulated genes, including the three E. coli DNA Pols involved in TLS (Pol II, Pol IV, and Pol V). In an attempt to understand the regulation of dinBPa, we constructed a PAO1 {Delta}lexA::aacC1 mutant (WFPA340) by inserting a gentamicin resistance cassette (aacC1) within lexA coding sequences (see Materials and Methods). Real-time RT-PCR analysis was then performed with RNA derived from the PAO1 (lexA+) and the WFPA340 ({Delta}lexA::aacC1) strains using a dinB-specific probe and primers. Real-time RT-PCR experiments showed a 5.5-fold increase in dinB mRNA levels in WFPA340 compared with PAO1 dinB mRNA levels (Fig. 3A). We also tested whether DNA damage caused a similar up-regulation of dinB mRNA. Real-time RT-PCR analysis of MMC-treated PAO1 cells showed an increase in dinB mRNA levels similar to that of WFPA340 (Fig. 3A). We next addressed the effect of the {Delta}lexA::aacC1 allele on DinBPa protein levels. Cell-free protein extracts of PAO1 and WFPA340 were prepared, separated by SDS-PAGE, and processed by Western blotting using anti-DinBPa antibodies. A significant increase in DinBPa protein levels was observed in WFPA340 compared with PAO1 (Fig. 3B, compare lanes 2 and 3). When extracts were made from cultures of PAO1 treated with MMC, an increase in the steady-state levels of DinBPa was observed. These data support the hypothesis that LexAPa acts to repress dinBPa transcription and suggest that, like in E. coli, the transcription of dinBPa is regulated as part of an SOS-like response to DNA damage.

We next wished to determine whether LexA-dependent repression of dinBPa was due to a direct interaction of LexAPa with dinB. EMSAs were used in all DNA binding studies (Fig. 4). First, we examined whether purified LexAEc would bind a radiolabeled dinBPa promoter DNA. A dinBPa fragment extending 450 bp upstream of the coding sequence and containing a putative LexA binding site (5'-CTGT-N8-ACAG-3') located ~350 bp upstream of the predicted dinB start codon was used in these studies (pdinBPa). At all concentrations tested, LexAEc was unable to bind to dinBPa sequences (Fig. 4A), despite the fact that the LexAEc preparation readily recognized umuDCEc target DNA (Fig. 4B, lanes 2 to 3). However, a protein(s) present in extracts derived from P. aeruginosa PAO1 bound to the umuDCEc fragment (Fig. 4B, lane 4). Data supporting the hypothesis that this protein is LexAPa are (i) that no protein-DNA complex was formed when WFPA340 ({Delta}lexA::aacC1) extracts were tested (Fig. 4B, lane 5), while protein-DNA complexes were readily observed when wild-type P. aeruginosa extracts were tested with a radiolabeled dinBPa fragment (Fig. 4C, lanes 2 and 5), and (ii) that no binding was observed when the extracts were prepared from PAO1 cells treated with MMC (Fig. 4C, lanes 3 and 6) or from extracts prepared from {Delta}lexA::aacC1 cells (Fig. 4C, lanes 4 and 7).


Figure 4
View larger version (57K):
[in this window]
[in a new window]

  		FIG. 4. LexAPa and LexAEc DNA binding studies. (A) A radiolabeled dinBPa promoter DNA fragment (pdinBPa) was left untreated (lane 1) or incubated with increasing amounts of purified LexAEc (2.5 nM, 10 nM, 50 nM, 100 nM; lanes 2 to 5, respectively). (B) Radiolabeled umuDCEc promoter DNA (pumuDCEc) was incubated with the following extracts or purified LexAEc samples: lane 1, no protein; lane 2, 50 nM; lane 3, 100 nM; lane 4, 10 µg extract; lane 5, 10 µg extract. (C) A radiolabeled pdinBPa fragment was incubated with the following: lane 1, no protein; lane 2 to 4, 5.0 µg extract; lane 5 to 7, 10 µg extract; lane 8, no protein; lane 9 to 14, increasing amounts of LexAPa (100 ng, 500 ng, 750 ng, 1,000 ng, 1,500 ng, and 2,000 ng, respectively).

 
To verify the above findings, P. aeruginosa lexA was cloned and a hexahistidine-tagged form of LexAPa was generated (see Materials and Methods). When LexAPa was incubated with the 32P-radiolabeled dinB promoter DNA, a protein-DNA complex with a mobility similar to that seen with P. aeruginosa extracts was observed (Fig. 4C, lanes 9 to 14), indicating that purified LexAPa binds to dinBPa promoter sequences. This finding, when combined with the real time RT-PCR and DinBPa Western blot analysis (Fig. 3), strongly implies a role for LexAPa in directly repressing dinB, likely as part of an SOS-like response.

In light of our finding that the expression of dinBPa was LexA-regulated (Fig. 3 and 4), we next asked whether the disruption of lexAPa (lexA::aacC1), which results in elevated steady-state levels of DinBPa (Fig. 3B), conferred a mutator phenotype in P. aeruginosa. As shown in Table 4, the mutation frequency of P. aeruginosa strain WFPA340 ({Delta}lexA::aacC1) was indistinguishable from that of its isogenic lexA+ partner. Taken together, results discussed above indicate that DinBPa can contribute to mutagenesis when expressed at an elevated level (Table 4) and suggest that the inactivation of the LexAPa repressor alone is insufficient for DinBPa to contribute to mutagenesis under the conditions examined thus far.

Role of dinBPa in DNA damage-induced mutagenesis. We next asked whether DinB functions in induced mutagenesis in P. aeruginosa. Based on our findings that (i) dinBPa expression was regulated by LexA and induced by MMC treatment (Fig. 3 and 4) and (ii) elevated levels of dinBPa conferred a modest mutator effect in P. aeruginosa (Table 4), we hypothesized that MMC treatment might confer a mutator phenotype on PAO1 that was dependent, at least in part, on DinBPa. To test this hypothesis, we treated PAO1 (dinB+) and WFP3340 ({Delta}dinB::aacC1) with 1 µg/ml MMC for 2 h and then washed the cells and cultured them overnight in LB prior to plating appropriate dilutions onto both LB and LB-Rif plates to determine mutation frequencies. Despite the fact that MMC treatment induced the expression of DinBPa (Fig. 3), we failed to observe an MMC-induced mutator phenotype with PAO1 (3.1 ± 1.2 [untreated] versus 2.8 ± 0.83 [MMC treated] Rifr CFU per 108 survivors) or WFPA334 (6.4 ± 3.2 [untreated] versus 3.5 ± 2.3 [MMC treated] Rifr CFU per 108 survivors).

Reaction of nitrofurazone (NFZ) or 4-nitroquinilone 1-oxide (4-NQO) with DNA results primarily in N2-dG adducts (65). The dinB-encoded Pol IV of E. coli was recently reported to play an important role in protecting E. coli from the lethal effects of NFZ and 4-NQO by catalyzing largely accurate TLS over NFZ- and 4-NQO-induced N2-dG adducts (28). We therefore examined whether dinBPa played a similar role in protecting P. aeruginosa against the lethal effects of NFZ and 4-NQO by comparing the growth of PAO1 (dinB+) and WFPA334 ({Delta}dinB::aacC1) on LB plates supplemented with different concentrations of NFZ or 4-NQO. Consistent with the findings of Jarosz et al. (28), the inactivation of dinB sensitized E. coli to the potentially lethal effects of NFZ (Fig. 5A, compare growth of the dinB+ and {Delta}dinB strains at the 10–4 and 10–5 dilutions on plates containing 0 or 10 µM NFZ); however, in contrast to findings of Jarosz et al. (28), the inactivation of dinB did not appear to significantly sensitize E. coli to 4-NQO (Fig. 5C, compare growth of the dinB+ and {Delta}dinB strains at the 10–2 and 10–3 dilutions on plates containing 0 or 40 µM 4-NQO). The inactivation of dinB slightly sensitized P. aeruginosa to NFZ, although considerably higher concentrations of the NFZ were required in order to observe killing (Fig. 5A, compare growth of dinB+ and dinB strains at the 10–3, 10–4 and 10–5 dilutions on plates containing 0, 160, and 320 µM NFZ, respectively). In contrast to the E. coli {Delta}dinB control, the {Delta}dinB P. aeruginosa strain was significantly more sensitive to 4-NQO than was the isogenic dinB+ P. aeruginosa strain (Fig. 5C).


Figure 5
View larger version (46K):
[in this window]
[in a new window]

  		FIG. 5. Role of DinBPa in tolerating NFZ- and 4-NQO-induced DNA damage. Relative sensitivities to NFZ (A) or 4-NQO (C) of dinB+ and {Delta}dinB E. coli and P. aeruginosa strains. Cultures of E. coli strains RW118 (dinB+) and RM137 [{Delta}(dinB-yafN)::kan], and of P. aeruginosa strains PAO1 (dinB+) and WFPA334 ({Delta}dinB::accC1) were serially diluted in 0.8% NaCl and spotted on LB plates containing the indicated concentration of NFZ or 4-NQO. Plates were incubated overnight at 37°C prior to photographing. NFZ- and 4-NQO-induced mutation frequencies were measured by inoculating 5 ml of LB containing either N,N-dimethylformamide (DMF), 160 µM NFZ, or 320 µM NFZ (B) or either 160 µM or 320 µM 4-NQO (D) (each made freshly in DMF [NFZ] or ethanol [4-NQO]) with 0.1 ml of P. aeruginosa strains PAO1 or WFPA334 grown to an OD595 of ~0.5. Following growth for ~16 h at 37°C, cells were washed twice with 0.8% NaCl and appropriate dilutions were spread onto LB or LB-Rif agar plates. Mutation frequencies were calculated by dividing the number of Rifr CFU by the total number of viable cells for each strain. Control experiments indicated that the presence of DMF alone had no effect on the frequency of Rifr for either PAO1 or WFPA334. Mutation frequencies are expressed relative to those observed for strain PAO1 grown in the presence of DMF alone (2.9 ± 1.9 Rifr CFU per 108 CFU for the NFZ experiment), or in LB (4.9 ± 2.2 Rifr CFU per 108 CFU for the 4-NQO experiment), which were each set equal to 1.0. Results shown represent the averages of at least three independent experiments, each performed in duplicate. Error bars represent standard deviations. Differences in the frequencies of Rifr between strains PAO1 and WFPA334 grown in LB or in the presence of DMF alone (P value ≥ 0.100), relative to that grown in the presence of 4-NQO or NFZ (P values ≥ 0.220, respectively, relative to the DMF-alone or LB-only controls), were not statistically significant at the 95% confidence level based on the Student t test. Ec dinB+, dinBEc+; Pa dinB+, dinBPa+.

 
We next asked whether NFZ or 4-NQO treatment increased the frequency of Rifr in P. aeruginosa. Surprisingly, unlike with most TLS, which is error prone, dinB of E. coli is able to replicate NFZ- and 4-NQO-induced lesions in a largely accurate manner and hence DinBEc-mediated TLS over NFZ-induced lesions fails to result in an increased frequency of Rifr in E. coli (28). In order to determine whether NFZ-induced DNA lesions are tolerated in an accurate or an error-prone way by DinBPa, we compared the frequencies of Rifr of P. aeruginosa PAO1 (dinB+) and WFPA334 ({Delta}dinB::aacC1) cultures grown overnight in the presence of various concentrations of NFZ or 4-NQO. Although the strain bearing the {Delta}dinB::aacC1 allele grown in the presence of 160 or 320 µM NFZ consistently displayed a higher frequency of Rifr of about two- to threefold relative to the same strain grown in the absence of NFZ or the isogenic dinB+ strain PAO1 (Fig. 5B), this effect was not statistically significant (see the legend for Fig. 5). Similar results were observed with 4-NQO (Fig. 5D). Thus, we conclude that DinBPa protects P. aeruginosa against the potentially lethal effects of NFZ and 4-NQO by mediating largely accurate TLS over N2-dG adducts, as was reported for DinBEc (28).

Both polA (DNA polymerase I) and polC (the second DNA polymerase III enzyme), but not dinB, are required for UV-induced mutagenesis in P. aeruginosa. There are conflicting reports in the literature as to whether P. aeruginosa displays UV-induced mutagenesis (36, 44). In order to determine whether the prototrophic P. aeruginosa strain PAO1 displays UV-induced mutability, frequencies for spontaneous and UV-induced Rifr were measured. As summarized in Table 5, P. aeruginosa displayed a modest UV-induced mutator phenotype based on the approximately twofold increase in the frequency of Rifr of P. aeruginosa strain PAO1 following UV irradiation (P value of 0.041). Since P. aeruginosa lacks an E. coli umuC (Pol V) ortholog, we asked whether DinBPa played a role in UV-induced mutagenesis. Our finding that the frequencies of UV-induced Rifr were comparable in both the dinB+ (PAO1) and {Delta}dinB::aacC1 strains (WFPA334) indicated that DinBPa does not play an essential role in UV-induced mutagenesis in P. aeruginosa (Table 5). Moreover, our finding that the frequency of spontaneous Rifr in the WFPA334 strain was essentially indistinguishable from that of PAO1 suggests that DinBPa does not contribute significantly to spontaneous mutagenesis in P. aeruginosa under the conditions examined in this study (Table 5), consistent with what has been reported for dinB in P. putida (59).


View this table:
[in this window]
[in a new window]

  		TABLE 5. Spontaneous and UV-induced mutation frequencies in P. aeruginosa dinB+ and {Delta}dinB strains

 
Inspection of the PAO1 genome indicates that it encodes at least five distinct DNA polymerases. These include, in addition to the dinB-encoded Pol IV (PA0923), one Pol I (PA5493), one Pol II (PA1886), and two Pol IIIs (PA3640 and PA0669). Although Pol I has been reported to participate in mutagenesis in Bacillus subtilis (17), it does not play a role in mutagenesis in E. coli. In contrast, E. coli Pol II does participate in TLS under certain conditions (2, 4, 31) and the P. aeruginosa Pol II may likewise contribute to mutagenesis in P. aeruginosa. PA3640 is more closely related to the catalytic subunit of the E. coli replicase that is encoded by dnaE. This gene is presumably essential for the viability of P. aeruginosa (27), suggesting that PA3640 encodes the P. aeruginosa replicase. In contrast to PA3640, PA0669 is dispensable for viability in P. aeruginosa (27). Interestingly, both B. subtilis (16) and Caulobacter crescentus (23) also possess two Pol IIIs, encoded by the dnaE and polC genes and by the dnaE1 and dnaE2 genes, respectively. B. subtilis dnaE (35) and C. crescentus (23) dnaE2 are each DNA damage inducible and appear to be responsible for most damage-induced mutagenesis. We therefore used a collection of isogenic strains that were made available through the University of Washington Genome Center (27) to determine whether DNA Pol I (polA or PA5493), DNA Pol II (polB or PA1886), and/or the dispensable DNA Pol III (polC or PA0669) were involved in UV-induced mutagenesis. As shown in Fig. 6, UV-induced mutagenesis in P. aeruginosa was dependent upon both DNA Pol I (polA) and the second DNA Pol III (polC).


Figure 6
View larger version (21K):
[in this window]
[in a new window]

  		FIG. 6. UV-induced mutagenesis in P. aeruginosa. UV-induced mutagenesis was performed as described previously (36, 44) using the indicated isogenic P. aeruginosa strains expressing all five known P. aeruginosa DNA polymerases (MPAO1), or bearing a transposon insertion in the gene encoding either the DNA polymerase I (polA; MPA50455), DNA polymerase II (polB; MPA17464) or the second DNA polymerase III enzyme (polC; MPA34086). Results shown represent the averages of at least three independent experiments, each performed in duplicate. Error bars represent standard deviations. Based on the Student t test, differences between strains MPAO1 and MPA17464 are not statistically significant at the 95% confidence level (P value of 0.79), while those between strains MPAO1 and MPA50455 (P value of 0.03) as well as MPAO1 and MPA34086 (P value of 0.002) are statistically significant.

 

   DISCUSSION
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
As part of an effort to understand the mechanisms of mutagenesis in P. aeruginosa, we cloned the P. aeruginosa dinB gene and characterized the recombinant DinBPa protein using a combination of genetic and biochemical approaches. Results discussed above indicate that DinBPa possesses an intrinsic DNA polymerase activity in vitro without the requirement for additional partner proteins: DinBPa was distributive and lacked 3'-to-5' exonuclease proofreading activity (Fig. 1). Importantly, DinBPa was error prone when expressed from a plasmid in E. coli, enhancing both C->A transversion and –1 frameshift mutations (Table 3). Interestingly, DinBEc was previously reported to prefer these same mutations (33), although it was proficient at a wider spectrum of mutations, including transitions, transversions, and frameshift mutations that were not conferred by ectopic expression of DinBPa in this study (Table 3). As with E. coli, ectopic expression of DinBPa conferred a mutator phenotype in P. aeruginosa (Table 4). In contrast, the inactivation of lexAPa failed to affect the spontaneous mutation frequency of PAO1 as measured by Rifr (Table 4). Our finding that dinB protected P. aeruginosa against the lethal effects of NFZ and 4-NQO without significantly affecting the frequency of induced mutagenesis (as measured by Rifr) suggests that DinBPa promotes largely accurate TLS over N2-dG adducts (Fig. 5) as was recently reported for DinBEc (28).

In addition to our finding that DinBEc appeared more promiscuous than did DinBPa with respect to the types of mutations that it could confer, we also observed a striking difference in the behavior of equivalent mutant forms of DinBEc and DinBPa. Based on the high degree of identity (49%)/similarity (63%) between these two proteins, we hypothesized that amino acid substitutions previously described as impairing the catalytic DNA polymerase activity of DinBEc would likewise impair the activity of DinBPa. Substituting either D8, R49, or D103 with alanine was reported to severely impair DinBEc-mediated Lac+ reversion in vivo as well as catalytic activity of the purified protein in an in vitro primer extension assay, without noticeably affecting the stability of the recombinant protein (60). In striking contrast to these findings with DinBEc, the DinB(R49A)Pa mutant protein retained partial catalytic activity as measured both in vitro using a primer extension assay (Fig. 1) as well as in vivo using a lacZ->Lac+ reversion assay (Table 3). Based on the crystal structure of the Sulfolobus solfataricus Dpo4 dinB ortholog (37), residue R49 is located in the presumed DNA binding region and/or interacts with the incoming dNTP (37). Our finding that the R49A mutant DinBPa protein retained partial activity, while the R49A mutant DinBEc protein was reported to lack detectable activity (60), suggests that this residue (R49) may play different roles in these two enzymes. Moreover, our finding that the D103A DinBPa protein was poorly stable when expressed in E. coli (Fig. 2) similarly contrasts with previous reports describing the DinB(D103A)Ec mutant (60). Taken together, these findings suggest that residues R49 and D103 contribute differently to catalysis/folding in these two enzymes, perhaps due to subtle differences between the structures of their respective active sites.

In an attempt to understand the regulation of DinBPa, we generated a {Delta}lexA mutant and compared DinBPa levels with those of the wild type of P. aeruginosa. There were higher levels of DinBPa in the {Delta}lexA mutant, indicating a role for LexA in repressing DinBPa expression (Fig. 3). This was validated by a real time RT-PCR assay in which we observed a fivefold increase in dinB mRNA levels in WFPA340 compared with that in the wild-type P. aeruginosa strain. Moreover, experiments with MMC indicate that the expression of dinBPa is elevated in response to MMC-mediated DNA damage (Fig. 3). These results suggest that DinBPa is under the control of an SOS-like system which has not been well studied in P. aeruginosa. To directly test this possibility, recombinant LexAPa was generated and analyzed for binding to dinB regulatory sequences (Fig. 4). The fragment used in these binding assays contained a putative LexA box. It should be noted that, during the course of purification, we consistently recovered two distinct LexAPa species. LexAEc undergoes RecA-mediated autocatalytic cleavage during the SOS response (38). Therefore, it is likely that these two species correspond to the cleaved and uncleaved forms of LexAPa. This likelihood is supported by the molecular masses of the two proteins as well as by immunoblot analyses with anti-His antiserum (data not shown). We determined that purified LexAEc will not bind upstream of dinBPa at concentrations necessary to bind E. coli umuDC DNA (Fig. 4A and B). The binding of LexAPa to dinB requires significantly higher concentrations of protein than that required for LexAEc to bind its target DNAs. There are several possible explanations for these differences. First, it is possible that LexAPa exhibits overall weaker binding affinity for target sequences or that the dinBPa LexA box is of low affinity compared with those of other LexA-repressed genes. It is also possible that the two species of LexA in our preparations antagonize the binding activity, or that some DNA binding activity is lost during purification. Nevertheless, the Western blot and real-time RT-PCR data (Fig. 3), taken together with the LexA binding studies (Fig. 4), lead us to conclude that DinBPa is most likely under the direct control of LexAPa. The closely related Pseudomonas putida also encodes a DinB protein that is directly regulated by LexA (1). Our conclusion that dinBPa is regulated as part of an SOS-like response is further supported by the finding that the transcription of dinBPa was induced following exposure to hydrogen peroxide (47). On the other hand, there are conflicting reports in the literature as to whether ciprofloxacin (a fluoroquinolone), which is reported to induce the SOS response in Haemophilus influenzae and P. aeruginosa (11), induces dinBPa transcription: two independent microarray analyses failed to observe an effect of ciprofloxacin on dinBPa transcript levels (6, 7), while a third microarray analysis reported that ciprofloxacin induced dinBPa transcription in a LexA-independent fashion (11). Further work is necessary in order to better understand the effects that these antibiotics have on LexA-dependent regulation of dinB transcription.

Various mutations have been described that confer pathoadaptive value upon P. aeruginosa; the mucA gene is perhaps the best-studied locus in this regard. Although the mechanism(s) by which mucA acquires inactivating mutations is currently unknown, nucleotide sequence analysis of mucA alleles impaired for the regulation of AlgT have been previously described. Interestingly, a remarkably large fraction of these mucA alleles were reported to contain a –1 frameshift mutation located within a homopolymeric run of 5 deoxyguanylic acid residues (known as the mucA22 allele) (40). Thus, it is tempting to speculate that replication errors, such as those introduced by DinBPa, might underlie the mechanistic basis for the initial mutagenic events leading to mucA inactivation. Our finding that DinBPa efficiently promotes –1 frameshift mutations within homopolymeric runs of deoxyguanylic or deoxyadenylic acid residues (Table 3) suggests that this enzyme may play an important role in mucA inactivation. The finding that overexpressed levels of DinBPa resulted in a modest mutator phenotype is consistent with this model. Our results suggest that error-prone replication by DinBPa and/or P. aeruginosa Pols I and/or Pol III might also contribute to the hypermutable phenotype that has been described for some clinical isolates of P. aeruginosa recovered from individuals suffering from chronic infections. The finding that hypermutability in clinical P. aeruginosa isolates was largely the result of an MMR defect (46) is consistent with the idea that error-prone replication contributes to mucA mutagenesis. Thus, we suggest that, in the absence of MMR, replication errors introduced by Pols I, III, and IV would persist, leading to elevated levels of mutagenesis. Further genetic and biochemical analysis of DinBPa as well as the other P. aeruginosa Pols will address this possibility. Furthermore, in C. crescentus, two hypothetical genes, named imuA and imuB and found in the same operon as the dnaE2, appear to work in conjunction with dnaE2 to promote UV- and MMC-induced mutagenesis (23). An inspection of the PAO1 genome indicates that it carries an imuB ortholog (PA0670). Further work is necessary in order to determine whether imuBPa contributes to induced mutagenesis in P. aeruginosa.


   ACKNOWLEDGMENTS
 
We thank the two anonymous expert reviewers for their insightful comments, Sidney Kushner (University of Georgia, Athens) and Jeffrey Miller (University of California, Los Angeles) for plasmid DNAs and E. coli strains, Michael Jacobs (University of Washington Genome Center) for P. aeruginosa strains, Daniel Jarosz and Graham C. Walker (Massachusetts Institute of Technology) for purified C-terminally hexahistidine-tagged DinBEc protein, John Little (University of Arizona) for purified LexAEc protein, and the members of our two labs for numerous helpful discussions and suggestions.

This work was supported by Public Health Service grants HL58334 (D.J.W.) and GM66094 (M.D.S.).


   FOOTNOTES
 
* Corresponding author. Mailing address: Department of Biochemistry, School of Medicine and Biomedical Sciences, University at Buffalo, State University of New York, 3435 Main Street, 140 Farber Hall, Buffalo, NY 14214. Phone: (716) 829-3581. Fax: (716) 829-2661. E-mail: mdsutton{at}buffalo.edu. Back

{triangledown} Published ahead of print on 13 October 2006. Back

{dagger} These authors contributed equally. Back


   REFERENCES
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Abella, M., I. Erill, M. Jara, G. Mazon, S. Campoy, and J. Barbe. 2004. Widespread distribution of a lexA-regulated DNA damage-inducible multiple gene cassette in the Proteobacteria phylum. Mol. Microbiol. 54:212-222.[CrossRef][Medline]
  2. Al Mamun, A. A., and M. Z. Humayun. 2006. Escherichia coli DNA polymerase II can efficiently bypass 3,N4-ethenocytosine lesions in vitro and in vivo. Mutat. Res. 593:164-176.[Medline]
  3. Baynham, P. J., A. L. Brown, L. L. Hall, and D. J. Wozniak. 1999. Pseudomonas aeruginosa AlgZ, a ribbon-helix-helix DNA-binding protein, is essential for alginate synthesis and algD transcriptional activation. Mol. Microbiol. 33:1069-1080.[CrossRef][Medline]
  4. Becherel, O. J., and R. P. Fuchs. 2001. Mechanism of DNA polymerase II-mediated frameshift mutagenesis. Proc. Natl. Acad. Sci. USA 98:8566-8571.[Abstract/Free Full Text]
  5. Becker, A., M. Schmidt, W. Jager, and A. Puhler. 1995. New gentamicin-resistance and lacZ promoter-probe cassettes suitable for insertion mutagenesis and generation of transcriptional fusions. Gene 162:37-39.[CrossRef][Medline]
  6. Blázquez, J., J. M. Gomez-Gomez, A. Oliver, C. Juan, V. Kapur, and S. Martin. 2006. PBP3 inhibition elicits adaptive responses in Pseudomonas aeruginosa. Mol. Microbiol. 62:84-99.[CrossRef][Medline]
  7. Brazas, M. D., and R. E. Hancock. 2005. Ciprofloxacin induction of a susceptibility determinant in Pseudomonas aeruginosa. Antimicrob. Agents Chemother. 49:3222-3227.[Abstract/Free Full Text]
  8. Buchmeier, N. A., S. J. Libby, Y. Xu, P. C. Loewen, J. Switala, D. G. Guiney, and F. C. Fang. 1995. DNA repair is more important than catalase for Salmonella virulence in mice. J. Clin. Investig. 95:1047-1053.[Medline]
  9. Buchmeier, N. A., C. J. Lipps, M. Y. So, and F. Heffron. 1993. Recombination-deficient mutants of Salmonella typhimurium are avirulent and sensitive to the oxidative burst of macrophages. Mol. Microbiol. 7:933-936.[Medline]
  10. Cano, D. A., M. G. Pucciarelli, F. Garcia-del Portillo, and J. Casadesus. 2002. Role of the RecBCD recombination pathway in Salmonella virulence. J. Bacteriol. 184:592-595.[Abstract/Free Full Text]
  11. Cirz, R. T., B. M. O'Neill, J. A. Hammond, S. R. Head, and F. E. Romesberg. 2006. Defining the Pseudomonas aeruginosa SOS response and its role in the global response to the antibiotic ciprofloxin. J. Bacteriol. 188:7101-7110.[Abstract/Free Full Text]
  12. Courcelle, J., A. Khodursky, B. Peter, P. O. Brown, and P. C. Hanawalt. 2001. Comparative gene expression profiles following UV exposure in wild-type and SOS-deficient Escherichia coli. Genetics 158:41-64.[Abstract/Free Full Text]
  13. Cox, M. M., M. F. Goodman, K. N. Kreuzer, D. J. Sherratt, S. J. Sandler, and K. J. Marians. 2000. The importance of repairing stalled replication forks. Nature 404:37-41.[CrossRef][Medline]
  14. Cupples, C. G., M. Cabrera, C. Cruz, and J. H. Miller. 1990. A set of lacZ mutations in Escherichia coli that allow rapid detection of specific frameshift mutations. Genetics 125:275-280.[Abstract]
  15. Cupples, C. G., and J. H. Miller. 1989. A set of lacZ mutations in Escherichia coli that allow rapid detection of each of the six base substitutions. Proc. Natl. Acad. Sci. USA 86:5345-5349.[Abstract/Free Full Text]
  16. Dervyn, E., C. Suski, R. Daniel, C. Bruand, J. Chapuis, J. Errington, L. Janniere, and S. D. Ehrlich. 2001. Two essential DNA polymerases at the bacterial replication fork. Science 294:1716-1719.[Abstract/Free Full Text]
  17. Duigou, S., S. D. Ehrlich, P. Noirot, and M. F. Noirot-Gros. 2005. DNA polymerase I acts in translesion synthesis mediated by the Y-polymerases in Bacillus subtilis. Mol. Microbiol. 57:678-690.[CrossRef][Medline]
  18. Fernandez de Henestrosa, A. R., T. Ogi, S. Aoyagi, D. Chafin, J. J. Hayes, H. Ohmori, and R. Woodgate. 2000. Identification of additional genes belonging to the LexA regulon in Escherichia coli. Mol. Microbiol. 35:1560-1572.[CrossRef][Medline]
  19. Friedberg, E. C., W. J. Feaver, and V. L. Gerlach. 2000. The many faces of DNA polymerases: strategies for mutagenesis and for mutational avoidance. Proc. Natl. Acad. Sci. USA 97:5681-5683.[Free Full Text]
  20. Friedberg, E. C., A. R. Lehmann, and R. P. Fuchs. 2005. Trading places: how do DNA polymerases switch during translesion DNA synthesis? Mol. Cell 18:499-505.[CrossRef][Medline]
  21. Friedberg, E. C., R. Wagner, and M. Radman. 2002. Specialized DNA polymerases, cellular survival, and the genesis of mutations. Science 296:1627-1630.[Abstract/Free Full Text]
  22. Friedberg, E. C., G. C. Walker, and W. Siede. 1995. DNA repair and mutagenesis. ASM Press, Washington, D.C.
  23. Galhardo, R. S., R. P. Rocha, M. V. Marques, and C. F. Menck. 2005. An SOS-regulated operon involved in damage-inducible mutagenesis in Caulobacter crescentus. Nucleic Acids Res. 33:2603-2614.[Abstract/Free Full Text]
  24. Goodman, M. F. 2002. Error-prone repair DNA polymerases in prokaryotes and eukaryotes. Annu. Rev. Biochem. 71:17-50.[CrossRef][Medline]
  25. Ho, C., O. I. Kulaeva, A. S. Levine, and R. Woodgate. 1993. A rapid method for cloning mutagenic DNA repair genes: isolation of umu-complementing genes from multidrug resistance plasmids R391, R446b, and R471a. J. Bacteriol. 175:5411-5419.[Abstract/Free Full Text]
  26. Hoang, T. T., R. R. Karkhoff-Schweizer, A. J. Kutchma, and H. P. Schweizer. 1998. A broad-host-range Flp-FRT recombination system for site-specific excision of chromosomally-located DNA sequences: application for isolation of unmarked Pseudomonas aeruginosa mutants. Gene 212:77-86.[CrossRef][Medline]