Developing a Genetic System in Deinococcus radiodurans for Analyzing Mutations

Corresponding author: Jeffrey H. Miller, Immunology and Molecular Genetics and the Molecular Biology Institute, University of California, 609 Charles E. Young Dr. #1602, Los Angeles, CA 90095., jhmiller{at}mbi.ucla.edu (E-mail)

Communicating editor: S. T. LOVETT

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have applied a genetic system for analyzing mutations in Escherichia coli to Deinococcus radiodurans, an extremeophile with an astonishingly high resistance to UV- and ionizing-radiation-induced mutagenesis. Taking advantage of the conservation of the ß-subunit of RNA polymerase among most prokaryotes, we derived again in D. radiodurans the rpoB/Rif^r system that we developed in E. coli to monitor base substitutions, defining 33 base change substitutions at 22 different base pairs. We sequenced >250 mutations leading to Rif^r in D. radiodurans derived spontaneously in wild-type and uvrD (mismatch-repair-deficient) backgrounds and after treatment with N-methyl-N'-nitro-N-nitrosoguanidine (NTG) and 5-azacytidine (5AZ). The specificities of NTG and 5AZ in D. radiodurans are the same as those found for E. coli and other organisms. There are prominent base substitution hotspots in rpoB in both D. radiodurans and E. coli. In several cases these are at different points in each organism, even though the DNA sequences surrounding the hotspots and their corresponding sites are very similar in both D. radiodurans and E. coli. In one case the hotspots occur at the same site in both organisms.

AS we continue to explore the vast diversity of microorganisms growing in extreme environments, we need to develop genetic systems to study biological processes and to help interpret the information from genome-sequencing projects. But how can we carry out genetic studies in organisms with no characterized genetic systems? One approach is to adapt methods from studying mutations that have been developed in well-studied microorganisms and to derive them again in the new organism of interest. We have recently completed the development of a system for analyzing base substitutions in Escherichia coli based on sequencing rpoB mutations that generate the rifampicin-resistant (Rif^r) phenotype (GARIBYAN et al. 2003 ). This system extends the previous work of several investigators (OVCHINNIKOV et al. 1983 and references therein; JIN and GROSS 1988 ; SEVERINOV et al. 1993 ) and, together with other recent work (RANGARAJAN et al. 1997 ; REYNOLDS 2000 ; PETERSEN-MAHRT et al. 2002 ; WOLFF et al. 2004 ), monitors at least 77 mutations at 37°. It has been used to analyze >1500 mutations derived from a series of mutators and mutagens, as well as from untreated wild-type controls (GARIBYAN et al. 2003 ; KIM et al. 2003 ; WOLFF et al. 2004 ). This work has revealed several prominent hotspots in the spontaneous spectrum and different hotspots in each of the mutagen-induced sets. Because the mutations that result in Rif^r are clustered within two small regions of rpoB, they can be analyzed by using only two primer pairs for amplification and sequencing. The rpoB-encoded ß-subunit of RNA polymerase is highly conserved among prokaryotes (MUSSER 1995 ; CAMPBELL et al. 2001 ). Rif^r mutants have been analyzed in a number of microorganisms, including several pathogens (see review by MUSSER 1995 ), and the mutations resulting in Rif^r have been determined. The mutations cluster in the same region as those for E. coli and mostly affect the corresponding residues. Thus, the rpoB/Rif^r system offers an opportunity for developing a genetic system to analyze mutations even in organisms that have had little genetic analysis. To evaluate the feasibility of applying this system to other bacteria, we initiated an investigation of mutagenesis in Deinococcus radiodurans R1 (BATTISTA and RAINEY 2001 ), a species within one of the deeply branching phyla of the domain Bacteria. Genetic methods available for the study of D. radiodurans are relatively primitive (BATTISTA 1997 ), but since the regions of rpoB containing the sites for the mutations leading to Rif^r have close to 80% amino acid identity between E. coli and D. radiodurans (Fig 1), we felt that we could construct a mutagenesis assay system for use in D. radiodurans similar to that constructed for E. coli. D. radiodurans is recognized for its ability to tolerate the lethal and mutagenic effects of DNA damage, exhibiting unusually high resistance to ionizing radiation and ultraviolet (UV) light (MOSELEY and MATTINGLY 1971 ; UDUPA et al. 1994 ), but the biochemical details of the response of D. radiodurans to DNA damage are poorly understood. Being able to analyze the specificity of mutators and mutagens with a system similar to the E. coli rpoB/Rif^r system may lead to insights into the nature of mutagenesis and repair in this organism.

View larger version (37K): In this window In a new window Download PPT slide   Figure 1. Homologies in the portion of the rpoB-encoded ß-subunits altered in Rifr mutants in D. radiodurans (D. rad.) and E. coli (E. coli).

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Bacterial strains:
We used D. radiodurans strain R1, ATCC 13939, as the wild type (ANDERSON et al. 1956 ). We constructed a uvrD derivative, NS3113, as detailed below.

Construction of pNS1165:
A PCR fragment encoding the uvrD gene (DR1775) of D. radiodurans R1 was amplified directly from purified R1 chromosomal DNA using a pair of primers derived from the published sequence of the R1 genome (http://www.tigr.org/tigr-scripts/CMR2/GenomePage3.spl?database=gdr). Primers uvrDLP (5'-TCACGCCTAGCCCAACTTCCTCT-3') and uvrDRP1 (5'-TACAGGATCGCCATCTCCGACCA-3') were designed for amplification of the first 1165 bp of the uvrD coding sequence. This PCR fragment was inserted directly into the vector pGEM-T (Promega, Madison, WI) to generate the construct pNS1165. The insert was sequenced and found to be identical to that of locus DR1775 in The Institute of Genomic Research database.

Construction of NS3113:
An in vitro transposition protocol (EARL et al. 2002 ) developed specifically for use in D. radiodurans was used to disrupt the uvrD coding sequence. Twenty nanograms of purified, circular pGTC101, a derivative of pGPS3 carrying a transposon that is functional in D. radiodurans, was combined with commercially available TnsABC* transposase (New England Biolabs, Beverly, MA) and pNS1165 in a 4:1 molar ratio. The transposition reaction mixture was transformed by heat shock into 5 x 10⁵ CFU (colony-forming units) of DH5MCR. Successful transposon insertions into the target were selected by plating the transformed cells onto Luria broth medium containing 25 µg/ml chloramphenicol. Fifty of the chloramphenicol-resistant colonies were picked and the plasmids they carried were isolated. These plasmids were digested with a combination of ApaI and PstI to release the gene of interest from the vector. Digestions were separated on 1% agarose gel and stained to confirm that the transposon had inserted into uvrD.

One microgram of ApaI-linearized plasmid was added to competent cultures of D. radiodurans R1 (1 x 10⁷ CFU/ml). After an 8-hr incubation, 300 µl of the transformation mixture was plated onto TGY agar plates containing 5 µg/ml chloramphenicol. Thirty-six colonies were used to inoculate TGY broth containing 5 µg/ml chloramphenicol and cultures were grown to stationary phase. One hundred microliters of this broth culture was used to inoculate TGY broth containing 10 µg/ml chloramphenicol and cultures were grown to stationary phase. Dilutions were plated on TGY agar containing 10 µg/ml chloramphenicol. Transposon insertions into uvrD were verified using PCR. The set of primers designed to amplify uvrD, uvrDLP and uvrDRP1, was combined with a primer (primer S, 5'-ATAATCCTTAAAAACTCCATTTCCACCCCT-3') that anneals within the transposon as described previously. The 1165-bp fragment corresponding to the amplified uvrD sequence could not be detected when all three primers were present. DNA sequencing using the uvrDRP1 primer established that the transposon inserted between nucleotides 886 and 887 of the uvrD coding sequence. The strain containing the disruption was designated NS3113. Since the disruption of uvrD would result in a mutator phenotype, we used the Rif^r assay to test for a mutator character. The frequency of Rif^r mutants in NS3113 was determined to be 12 times that of the wild type, R1, indicating that strain NS3113 is a mutator.

Isolating Rif^r mutants:
Spontaneous Rif^r mutants were obtained by inoculating 5-ml cultures with 100–300 freshly growing cells of R1 and growing for 48 hr on a rotor at 37° to saturation. Dilutions of the cultures were plated on TGY plates to determine the cell titer. The cultures were also concentrated 10-fold by centrifugation and 100 µl was plated on TGY containing 50 µg/ml rifampicin (Sigma Chemical, St. Louis) to determine the frequency of Rif^r mutations. Mutant frequencies were determined, and the median frequency () from a set of cultures (29) was used to calculate the mutation rate (µ) per replication by the method of DRAKE 1991 , using the formula µ = /ln Nµ, where N is the number of cells in the culture. Ninety-five percent confidence limits were determined according to DIXON and MASSEY 1969 . Once rifampicin-resistant colonies were obtained, either spontaneously or through the use of a mutagen (see protocols below), they were purified on TGY plates and incubated for 48 hr at 37° and then double picked onto TGY and TGY + rifampicin plates to confirm the Rif^r phenotype. Colonies from these TGY plates were used to inoculate cultures for DNA isolation and subsequent sequencing.

D. radiodurans genomic DNA:
Genomic DNA was isolated from saturated cultures. Briefly, using the Invitrogen (Carlsbad, CA) DNAzol protocol, cells were pelleted, resuspended in 500 µl of 95% ethanol, and incubated at room temperature for 5 min to remove the outer membrane. The ethanol-stripped cells were collected by centrifugation at 4° for 5 min at 10,000 x g and resuspended in 200 µl of TE buffer (10 mM Tris, 1 mM EDTA, pH 8.0). The stripped cells were incubated with 3 µl of 50 mg/ml lysozyme (Sigma Chemical) at room temperature for 8 min, followed by another incubation for 1 min with 1 ml of DNAzol reagent (Invitrogen). The cells were centrifuged at 4° for 10 min at 10,000 x g, and 1 ml of the supernatant was transferred to a new tube with 500 µl of 100% EtOH. The samples were mixed by inverting and kept at room temperature for 2 min and then centrifuged at 4° for 10 min at 10,000 x g to precipitate the DNA. The DNA pellet was washed twice with 200 µl of 75% EtOH and allowed to air dry for 30 min. When the ethanol had evaporated, 50 µl of TE buffer was added to the tube and the DNA was allowed to dissolve overnight at room temperature.

Sequencing the rpoB gene for mutations:
Once the chromosomal DNA was isolated, one of two primer pairs was used to amplify the DNA for direct sequencing. Most of the mutations occurred in the region obtained with 5'-AAACTGTGCCGATGGTGGAC-3' (5' position 1058) and 5'-TAGCTCACGCGGCCATTCAC-3' (5' position 1945). The rest of the Rif^r mutations were found using the primer pair 5'-TCTTTCCCATCGACGAGTCC-3' (5' position 173) and 5'-CACGATGGGGCGGTTGTT-3' (5' position 1224). The PCR reaction included 1x PCR buffer (Bio-Rad, Hercules, CA), 50 pmol of each PCR primer, 2 mM MgCl₂ (Bio-Rad), 40 nmol dNTP, 2.5% formamide, 1.5 units Taq DNA polymerase (Bio-Rad), 1 µl of DNA, and double-distilled H₂O. The DNA was denatured at 95° for 4 min, amplified for 30 cycles of 95° for 30 sec, 57° for 30 sec, and 72° for 1 min and extended for 7 min at 72°. PCR products were purified with the MinElute PCR purification kit (QIAGEN, Valencia, CA) and manually sequenced with the SequiTherm EXCEL II DNA sequencing kit (Epicentre Technologies, Madison, WI) using one of the two sequencing primers, 5'-CATGCTGCTCGGCAACCC-3' (5' position 1221) and 5'-TGATTCACAAAGACACTGGCGT-3' (5' position 323), respectively.

N-methyl-N'-nitro-N-nitrosoguanidine mutagenesis:
Six tubes of 5 ml TGY broth were inoculated with R1 and aerated for 48 hr at 37°. The cultures were pooled together into an Erlenmeyer flask and diluted 1:1 with TGY. A 1-mg/ml N-methyl-N'-nitro-N-nitrosoguanidine (NTG) solution was made in 1:1 acetone and 0.1 M NaCitrate buffer (pH 5.5). Two microliters of culture was aliquoted into tubes, and NTG was added to a final concentration of 0, 30, or 100 µg/ml. After a 90-min incubation in a 37° water bath, the mutagenized cultures were washed three times: first with 5 ml of 0.01 M MgSO₄, followed by 5 ml of TGY, and finally with 2 ml of TGY (MILLER 1992 ). To determine the percentage of cells surviving exposure to NTG, 50 µl of a 10^-4 dilution of each culture was plated onto TGY and incubated at 37° for 48 hr. Outgrowth cultures were made by adding 500 µl of each mutagenized culture to 5 ml TGY and incubating for 48 hr at 37°. The outgrowth cultures were concentrated 10-fold and 100 µl was plated on TGY + rifampicin plates to yield Rif^r colonies for sequencing. Two Rif^r colonies from each of the 10, 30, and 100 µg/ml NTG TGY + rifampicin plates were sequenced. The experiment was repeated with more cultures at a concentration of 100 µg/ml NTG.

5-Azacytidine mutagenesis:
An overnight culture was diluted and used to seed overnight cultures with 100–300 cells in TGY with 100 µg/ml of 5-azacytidine (5AZ) and grown for 48 hr on a rotor (MILLER 1992 ). The mutational frequency was determined by plating 100 µl of a 10^-5 dilution of each culture on TGY and 50 µl directly on TGY + rifampicin.

uvrD Rif^r mutants:
D. radiodurans uvrD Rif^r mutants were obtained in the same manner as spontaneous mutants, with the exception that each overnight culture was inoculated with a single uvrD colony instead of seeding the cultures with 100–300 cells.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Isolation of Rif^r mutants and sequence analysis of rpoB mutations:
We isolated Rif^r mutants at 37° on TGY plates containing 50 µg/ml rifampicin (see MATERIALS and METHODS) that occurred spontaneously in D. radiodurans wild-type or uvrD strains or that were induced by NTG or 5AZ. Table 1 shows the mutation frequencies and rates that we obtained. Although NTG was a potent mutagen for D. radiodurans, ethyl methanesulfonate (EMS) failed to mutagenize this organism at all. Also, although the cytosine analog 5AZ gave positive results, the cytosine analog zebularine, as well as the adenine analog 2-aminopurine, failed to mutagenize D. radiodurans. We isolated one mutant per culture and prepared DNA for sequence analysis of the relevant regions of the rpoB gene (see MATERIALS AND METHODS). Table 2 shows the results for >250 mutations in the D. radiodurans rpoB gene that result in the Rif^r phenotype, including 185 spontaneous mutations derived in a wild-type background, 33 NTG-induced mutations, and 19 5AZ-induced mutations, as well as 17 spontaneous mutations occurring in a uvrD background. From these data we can already define 33 different base substitution mutations at 22 sites (base pairs). Each of the 6 base substitutions can be monitored at a set of 3–7 sites.

Spontaneous mutations—deletion and base substitution hotspots:
We found that 35 of the 185 (19%) spontaneous mutations detected in the wild-type background are deletions of 9 bp at or adjacent to a 7-bp direct repeat separated by 2 bp and probably are the result of slipped mispairing stimulated at this site (see Fig 2). We detected three different deletions, although one of these (type III in Fig 2) is represented by only 1 occurrence, while type I and type II deletions are represented by 19 and 15 occurrences, respectively (see also Table 2). Deletion type II (Fig 2) is of the exact form found for many deletions in both E. coli and other organisms (see FARABAUGH et al. 1978 ; ALBERTINI et al. 1982 and references therein). Deletion type I is shifted just 1 bp in one direction. In E. coli, one or two small deletions in rpoB that result in Rif^r mutants have been reported (JIN and GROSS 1988 ), but these are relatively rare compared with the base substitutions that result in Rif^r in that organism.

View larger version (8K): In this window In a new window Download PPT slide   Figure 2. Deletions in rpoB resulting in the Rifr phenotype in D. radiodurans. The letters in boldface type indicate a 7-bp sequence repeat separated by 2 bp. See Table 2 for relative frequencies of these deletions.

In addition to the two deletion hotspots mentioned above, the spontaneous spectrum of rpoB mutations in D. radiodurans shows three base substitution hotspots at positions 1273 and 1303 (G:C A:T) and 1259 (A:T C:G). Together, these three hotspots account for 95 of the 150 base substitutions (63%) found among the spontaneous mutations in rpoB, even though these are distributed among 33 sites. A comparison of these hotspots with those in E. coli rpoB reveals that the respective hotspots involve different sites, as shown in Fig 3 (see also Table 2 and GARIBYAN et al. 2003 ), with one exception. Mutations of G:C A:T at 1303 in D. radiodurans and the corresponding 1576 in E. coli both represent hotspots. Fig 4 shows the DNA sequence alignments surrounding each of the hotspot sites. In three cases, both E. coli and D. radiodurans have identical nearest neighbors at the respective sites and yet dramatically different rates in two of these examples. In one case (E. coli 1547), the sequence identity is more extensive. We consider reasons for different hotspots in the DISCUSSION.

View larger version (34K): In this window In a new window Download PPT slide   Figure 3. Comparison of relative mutation frequencies in rpoB in D. radiodurans and E. coli. The numbers above the 0 line indicate the base number of the mutational site in D. radiodurans and the numbers below the line are the corresponding bases in E. coli. The D. radiodurans sample is from 150 base substitution mutations in D. radiodurans (see Table 2), while the E. coli sample is from 294 mutations (GARIBYAN et al. 2003 ; WOLFF et al. 2004 ). The total D. radiodurans mutations are distributed among 33 identified sites, and the E. coli mutations are distributed among 77 identified sites.

View larger version (44K): In this window In a new window Download PPT slide   Figure 4. DNA sequences and homologies surrounding the mutational hotspots in D. radiodurans and E. coli. The underlined base is the hotspot site (see also Table 2 and Fig 3).

Mutations resulting from mutators and mutagens:
We examined the mutational spectrum in a uvrD strain (Table 2). Here, all the mutations are transitions, with G:C A:T mutations predominating over A:T G:C mutations, although the spectrum is dominated by a single hotspot at position 1435. We have also employed two mutagens, NTG and 5AZ. It is clear from Table 2 that these agents induce specifically G:C A:T transitions in the case of NTG and G:C C:G transversions in the case of 5AZ in D. radiodurans and thus have the same specificity as found in E. coli and other organisms (CUPPLES and MILLER 1989 ).

Altered amino acids in RNA polymerase ß in Rif^r mutants:
The amino acid residues affected in Rif^r mutants are strikingly similar in E. coli and D. radiodurans, as detailed in Fig 5. However, only one site has been detected in D. radiodurans that has not yet been detected in E. coli. In E. coli, in which far more extensive studies have been carried out (JIN and GROSS 1988 ; SEVERINOV et al. 1993 ; GARIBYAN et al. 2003 ; WOLFF et al. 2004 ), alterations at 27 different residues can result in the Rif^r phenotype, whereas so far 15 such residues have been found in D. radiodurans.

View larger version (18K): In this window In a new window Download PPT slide   Figure 5. Amino acid changes in rpoB that lead to the Rifr phenotype in D. radiodurans and E. coli.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

In this work, we have successfully applied a genetic system developed in E. coli to D. radiodurans, a species that separated from the rest of the bacterial family tree early in evolution (BATTISTA and RAINEY 2001 ). Although assays measuring the frequency of forward mutation to rifampicin (TEMPEST and MOSELEY 1982 ), streptomycin (KERSZMAN 1975 ), and trimethoprim (SWEET and MOSELEY 1974 ) resistance have been described for D. radiodurans, currently no genetic system allows the analysis of mutations generated in wild-type and repair-deficient mutants of D. radiodurans. Therefore, we used the rpoB/Rif^r system recently developed in E. coli (GARIBYAN et al. 2003 ) to monitor base substitutions. Because the rpoB-encoded ß-subunit of RNA polymerase is highly conserved among most prokaryotes, it should be possible to develop a detailed rpoB/Rif^r system in an organism without sophisticated genetic tools. Investigators from numerous laboratories have analyzed Rif^r mutants from a variety of pathogens, including Mycobacterium tuberculosis (TELENTI et al. 1993 ; MUSSER 1995 ), M. smegmatis (KARUNAKARAN and DAVIES 2000 ), M. leprae (CAMBAU et al. 2002 ), M. kansasii (KLEIN et al. 2001 ), Bacillus anthracis (VOGLER et al. 2002 ), B. cereus (VOGLER et al. 2002 ), Rhodococcus equi (FINES et al. 2001 ), Legionella pneumophila (NEILSEN et al. 2000 ), Neisseria meningitides (STEFANELLI et al. 2001 ), Streptococcus pyogenes (AUBRY-DAMON et al. 2002 ), Staphylococcus aureus (AUBRY-DAMON et al. 1998 ), and Helicobacter pylori (HEEP et al. 2000 ). In all of these cases, the rpoB mutations affect residues corresponding to those altered in E. coli Rif^r mutants, although a complete catalog of possible mutations has not been attempted in these studies.

Table 2 shows the results of sequencing >250 mutations leading to Rif^r in D. radiodurans derived spontaneously in wild-type and uvrD (mismatch-repair-deficient) backgrounds, and after treatment with N-methyl-N'-nitro-N-nitrosoguanidine and 5-azacytidine. We have defined 33 base change substitutions at 22 different sites (base pairs). This allows us to monitor the A:T G:C change at 3 sites, the G:C A:T change at 7 sites, the A:T T:A change at 5 sites, the A:T C:G change at 5 sites, the G:C T:A change at 6 sites, and the G:C C:G change at 7 sites.

Analysis of the rpoB/Rif^r system provides the first detailed description of mutagenesis in D. radiodurans and lays the groundwork for studies of the mechanisms that this species uses to avoid mutation. D. radiodurans has an astonishing ability to avoid UV- and ionizing-radiation-induced mutagenesis (SWEET and MOSELEY 1974 ; KERSZMAN 1975 ; TEMPEST and MOSELEY 1982 ). For example, even when cultures are exposed to 700 J/m² UV light, a dose that introduces 5000 thymine-containing pyrimidine dimers per genome in exposed cells and that kills 90% of the irradiated population (MOSELEY 1983 ), there is no evidence of UV-induced mutagenesis in this species (TEMPEST and MOSELEY 1982 ). We anticipate that the introduction of the rpoB/Rif^r system will encourage further investigation of mutagenesis and DNA repair in D. radiodurans.

Several aspects of the results found in this study are worth noting. The spontaneous mutations reveal a deletion hotspot centered near sequence repeats that very probably serve as substrates for the type of slipped mispairing events that generate deletions in other organisms (FARABAUGH et al. 1978 ; ALBERTINI et al. 1982 ). Although deletions in rpoB do not usually generate viable Rif^r strains, the in-frame deletion of 9 bp removes three amino acids that alter the ß-subunit of RNA polymerase enough to become resistant to inactivation by rifampicin, but not enough to affect function. The specificities of NTG and 5AZ are the same in D. radiodurans as found for E. coli and other organisms (see, for instance, CUPPLES and MILLER 1989 ). What is interesting, and not easy to explain, is the failure to detect mutagenesis with EMS, even though NTG is a potent mutagen for D. radiodurans. Also, neither the cytosine analog zebularine (MCCORMACK et al. 1980 ) nor the adenine analog 2-aminopurine causes mutations in D. radiodurans, whereas they do in E. coli (see CUPPLES and J. H. MILLER 1989 ; LEE et al. 2004 ). Finally, the rpoB mutations in the D. radiodurans uvrD strain predominate at a single G:C A:T site, whereas the major hotspot in E. coli rpoB for mismatch-repair-deficient mutations is an A:T G:C mutation.

There are prominent base substitution hotspots in rpoB in both D. radiodurans and E. coli (Table 2; GARIBYAN et al. 2003 ), although with one exception they occur at different sites (see Fig 3). What makes this remarkable is that the DNA sequences surrounding the hotspot sites are very similar, as shown in Fig 4. For instance, the E. coli spontaneous hotspot (31 occurrences among 298 mutations) at position 1714 (A:T C:G) has a DNA sequence identical to that of the corresponding D. radiodurans site 1441 for 10 bp on one side and the nearest neighbor on the other (Fig 3), yet the D. radiodurans site has only two occurrences of A:T C:G mutations among 150 base substitutions. Similarly, the G:C A:T mutation at position 1273 in D. radiodurans is a hotspot, whereas the corresponding change at 1546 is not in E. coli, despite sharing the same 6 bp on one side and the same nearest neighbor on the other. As part of the same sequence segment, the AT G:C change at position 1547 is the most prominent hotspot in E. coli (40 occurrences among 298 mutations), but no mutations have been found at the corresponding position (1274) in D. radiodurans. Normally, one would not be able to make conclusions regarding this site in D. radiodurans, since without any occurrences recorded it might be that the A:T G:C change there does not result in a mutation that yields Rif^r cells. However, a look at Fig 5 shows that the aspartic acid residue specified by the codon involving 1547 in E. coli and 1274 in D. radiodurans is an amino acid (residue 516 in E. coli; 425 in D. radiodurans) at which many different exchanges yield Rif^r. In E. coli, where many more mutations have been generated, all five changes at this site have been detected and shown to result in Rif^r, whereas so far four of these five changes have already been found in D. radiodurans. It is highly probable that the remaining change (to glycine) will also yield a Rif^r cell and that the zero occurrences of an A:T G:C change at 1274 in D. radiodurans simply reflect failure to induce the mutation at a detectable rate with the current sample size and not a failure of the resulting change to generate Rif^r colonies. Position 1259 is a hotspot for the A:T C:G transversion in D. radiodurans, but not in E. coli. Although this site is part of a 5-bp homology and a 12 of 14-bp homology (Fig 4), the corresponding sites do not have identical nearest neighbors on one side. The lack of a nearest neighbor in an otherwise homologous stretch does not prevent both 1303 in D. radiodurans and 1576 in E. coli (25 occurrences among 298 mutations) from being hotspots for the G:C A:T transition. That these two sites are hotspots despite being in different organisms is remarkable. It is not clear whether the other sites that are hotspots in one organism but not in the other reflect a requirement for a more extensive sequence environment as a determinant of mutation rate or whether mutation rates at all sites are really organism specific. Additional experiments with engineered sequences are required to answer these questions.

We see no reason why the rpoB/Rif^r system cannot be applied to other genetically intractable bacterial species. The only requirements for implementing this system are that the investigator knows the rpoB gene sequence and has the ability to isolate Rif^r mutants from the species of interest. It should now be possible to extend the detailed analysis of spontaneous mutagenesis to a large number of diverse species and to explore how the specifics of this process compare among prokaryotes that evolved to occupy different ecological niches.

	ACKNOWLEDGMENTS

The authors gratefully acknowledge Nicole C. Shank for constructing strain NS3113. J.R.B. is supported by Department of Energy grant DE-FG02-01ER63151. J.H.M. is supported by National Institutes of Health grant ES0110875.

Manuscript received September 3, 2003; Accepted for publication October 16, 2003.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ALBERTINI, A. M., M. HOFER, M. P. CALOS, and J. H. MILLER, 1982  On the formation of spontaneous deletions: the importance of short sequence homologies in the generation of large deletions. Cell 29:319-328.[CrossRef][Medline]

ANDERSON, A. W., H. C. NORDON, R. F. CAIN, G. PARRISH, and D. DUGGAN, 1956  Studies on a radio-resistant micrococcus. Isolation, morphology, cultural characteristics, and resistance to gamma radiation. Food Technol. 10:575-578.

AUBRY-DAMON, H., C. J. SOUSSY, and P. COURVALIN, 1998  Characterization of mutations in the rpoB gene that confer rifampicin resistance in Staphylococcus aureus. Antimicrob. Agents Chemother. 42:2590-2594.[Abstract/Free Full Text]

AUBRY-DAMON, H., M. GALINMAND, G. GERBAUD, and P. COURVALIN, 2002  rpoB mutation conferring rifampicin resistance in Streptococcus pyogenes.. Antimicrob. Agents Chemother. 46:1571-1573.[Abstract/Free Full Text]

BATTISTA, J. R., 1997  Against all odds: the survival strategies of Deinococcus radiodurans.. Annu. Rev. Microbiol. 51:203-224.[CrossRef][Medline]

BATTISTA, J. R., and F. A. RAINEY, 2001 The Deinococcaceae, pp. 395–403 in Bergey's Manual of Systematic Bacteriology, Vol. 1, edited by G. M. GARRITY. Springer-Verlag, New York.

CAMBAU, E., P. BONNAFOUS, E. PERANI, W. SOUGAKOFF, and B. JI et al., 2002  Molecular detection of rifampicin and ofloxacin resistance for patients who experience relapse of multibacillary leprosy. Clin. Infect. Dis. 34:39-45.[CrossRef][Medline]

CAMPBELL, E. A., N. KORZHEVA, A. MUSTAEV, K. MURAKAMI, and S. NAIR et al., 2001  Structural mechanism for rifampicin inhibition of bacterial RNA polymerase. Cell 104:901-912.[CrossRef][Medline]

CUPPLES, C. and J. H. MILLER, 1989  A set of lacZ mutations in Escherichia coli allows rapid detection of each of the six base substitutions. Proc. Natl. Acad. Sci. USA 86:5345-5349.[Abstract/Free Full Text]

DIXON, W. J., and F. J. MASSEY, JR., 1969 Introduction to Statistical Analysis. McGraw-Hill, New York.

DRAKE, J. W., 1991  A constant rate of spontaneous mutation in DNA-based microbes. Proc. Natl. Acad. Sci. USA 88:7160-7164.[Abstract/Free Full Text]

EARL, A. M., S. K. RANKIN, K. P. KIM, O. N. LAMENDOLA, and J. R. BATTISTA, 2002  Genetic evidence that the uvsE gene product of Deinococcus radiodurans R1 is a UV damage endonuclease. J. Bacteriol. 184:1003-1009.[Abstract/Free Full Text]

FARABAUGH, P. J., U. SCHMEISSNER, M. HOFER, and J. H. MILLER, 1978  Genetic studies of the lac repressor. VII. On the molecular nature of spontaneous hotspots in the lacl gene of Escherichia coli. J. Mol. Biol. 126:847-863.[CrossRef][Medline]

FINES, M., S. PRONOST, K. MAILLARD, S. TAOUJI, and R. LECLERCQ, 2001  Characterization of mutations in the rpoB gene associated with rifampicin resistance in Rodococcus equi isolated from foals. J. Clin. Microbiol. 39:2784-2787.[Abstract/Free Full Text]

GARIBYAN, L., T. HUANG, M. KIM, E. WOLFF, and A. NGUYEN et al., 2003  Use of the rpoB gene to determine the specificity of base substitution mutations on the Escherichia coli chromosome. DNA Repair 2:593-608.[CrossRef][Medline]

HEEP, M., S. ODENBREIT, D. BECK, J. DECKER, and E. PROHASKA et al., 2000  Mutations at four distinct regions of the rpoB gene can reduce the susceptibility of Helicobacter pylori to rifamycin. Antimicrob. Agents Chemother. 44:1713-1715.[Abstract/Free Full Text]

JIN, D. J. and C. A. GROSS, 1988  Mapping and sequencing of mutations in the Escherichia coli rpoB gene that lead to rifampicin resistance. J. Mol. Biol. 202:45-58.[CrossRef][Medline]

KARUNAKARAN, P. and J. DAVIES, 2000  Genetic antagonism and hypermutability in Mycobacterium smegmatis.. J. Bacteriol. 182:3331-3335.[Abstract/Free Full Text]

KERSZMAN, G., 1975  Induction of mutation to streptomycin resistance in Micrococcus radiodurans.. Mutat. Res. 28:9-14.[CrossRef][Medline]

KIM, M., T. HUANG, and J. H. MILLER, 2003  Competition between MutY and mismatch repair at A • C mispairs in vivo. J. Bacteriol. 185:4626-4629.[Abstract/Free Full Text]

KLEIN, J. L., T. J. BROWN, and G. L. FRENCH, 2001  Rifampicin resistance in Mycobacterium kansasii is associated with rpoB mutation. Antimicrob. Agents Chemother. 45:3056-3058.[Abstract/Free Full Text]

LEE, G., E. WOLFF, and J. H. MILLER, 2004  Mutagenicity of the cytidine analog zebularine in Escherichia coli. DNA Repair 3:155-161.[Medline]

MCCORMACK, J. J., V. E. MARQUEZ, P. S. LIU, D. T. VISTICA, and J. S. DRISCOLL, 1980  Inhibition of cytidine deaminase by 2-oxopyrimidine riboside and related compounds. Biochem. Pharmacol. 29:830-832.[CrossRef]

MILLER, J. H., 1992 A Short Course in Bacterial Genetics: A Laboratory Manual and Handbook for Escherichia coli and Related Bacteria. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

MOSELEY, B. E., 1983  Photobiology and radiobiology of Micrococcus (Deinococcus) radioduran.. Photochem. Photobiol. Rev. 7:223-275.

MOSELEY, B. E. and A. MATTINGLY, 1971  Repair of irradiation transforming deoxyribonucleic acid in wild type and a radiation-sensitive mutant of Micrococcus radiodurans.. J. Bacteriol. 105:976-983.[Abstract/Free Full Text]

MUSSER, J. M., 1995  Antimicrobial resistance in mycobacteria: molecular genetic insights. Clin. Microbiol. Rev. 8:496-514.[Abstract]

NEILSEN, K., P. HINDERSSON, N. HOIBY, and J. M. BANGSBORG, 2000  Sequencing of the rpoB gene in Leionella pneumophila and characterization of mutations associated with rifampicin resistance in the legionellaacaeae. Antimicrob. Agents Chemother. 44:2679-2683.[Abstract/Free Full Text]

OVCHINNIKOV, Y. A., G. S. MONASTYRSKAYA, S. O. GURIEV, N. F. KALININA, and E. D. SVERDLOV et al., 1983  RNA polymerase rifampicin resistance mutations in Escherichia coli: sequence changes and dominance. Mol. Gen. Genet. 190:344-348.[CrossRef][Medline]

PETERSEN-MAHRT, S. K., R. S. HARRIS, and M. S. NEUBERGER, 2002  AID mutates E. coli, suggesting a DNA deamination mechanism for antibody diversification. Nature 418:99-103.[Medline]

RANGARAJAN, S., G. GUDMUNDSSON, Z. QIU, P. L. FOSTER, and M. F. GOODMAN, 1997  Esherichia coli DNA polymerase II catalyzes chromosomal and episomal DNA synthesis in vivo. Proc. Natl. Acad. Sci. USA 94:946-951.[Abstract/Free Full Text]

REYNOLDS, M. G., 2000  Compensatory evolution in rifampicin-resistant Escherichia coli.. Genetics 156:1471-1481.[Abstract/Free Full Text]

SEVERINOV, K., M. SOUSHKO, A. GOLDFARB, and V. NIKIFOROW, 1993  Rifampicin region revisited. J. Biol. Chem. 268:14820-14825.[Abstract/Free Full Text]

STEFANELLI, P., C. FAZIO, G. LA ROSA, C. MARIANELLI, and M. MUSCILLO et al., 2001  Rifampicin-resistant meningococci causing invasive disease: detection of point mutations in the rpoB gene and molecular characterization of the strains. J. Antimicrob. Chemother. 47:219-222.[Abstract/Free Full Text]

SWEET, D. M. and B. E. MOSELEY, 1974  Accurate repair of ultraviolet-induced damage in Micrococcus radiodurans.. Mutat. Res. 23:311-318.[CrossRef][Medline]

TELENTI, A., P. IMBODEN, F. MARCHESI, D. LOWRIE, and S. COLE et al., 1993  Detection of rifampicin resistance mutations in Mycobacterium tuberculosis.. Lancet 341:647-650.[CrossRef][Medline]

TEMPEST, P. R. and B. E. MOSELEY, 1982  Lack of ultraviolet mutagenesis in radiation-resistant bacteria. Mutat. Res. 104:275-280.[CrossRef][Medline]

UDUPA, K. S., P. A. O'CAIN, V. MATTIMORE, and J. R. BATTISTA, 1994  Novel ionizing radiation-sensitive mutants of Deinococcus radiodurans.. J. Bacteriol. 176:7439-7446.[Abstract/Free Full Text]

VOGLER, A. J., J. D. BUSCH, S. PERCY-FINE, C. TIPTON-HUNTON, and K. L. SMITH et al., 2002  Molecular analysis of rifampicin resistance in Bacullus anthracis and Bacillus cereus.. Antimicrob. Agents Chemother. 46:511-513.[Abstract/Free Full Text]

WOLFF, E., M. KIM, K. HU, H. YANG, and J. H. MILLER, 2004  Polymerases leave fingerprints: analysis of the mutational spectrum in Escherichia coli rpoB to assess the role of Pol IV in spontaneous mutation. J. Bacteriol. in press.

This article has been cited by other articles:

				A. M. Marcobal, D. A. Sela, Y. I. Wolf, K. S. Makarova, and D. A. Mills Role of Hypermutability in the Evolution of the Genus Oenococcus J. Bacteriol., January 15, 2008; 190(2): 564 - 570. [Abstract] [Full Text] [PDF]

				K. Zeibell, S. Aguila, V. Yan Shi, A. Chan, H. Yang, and J. H. Miller Mutagenesis and Repair in Bacillus anthracis: the Effect of Mutators J. Bacteriol., March 15, 2007; 189(6): 2331 - 2338. [Abstract] [Full Text] [PDF]

				R. S. Galhardo, R. P. Rocha, M. V. Marques, and C. F. M. Menck An SOS-regulated operon involved in damage-inducible mutagenesis in Caulobacter crescentus Nucleic Acids Res., May 10, 2005; 33(8): 2603 - 2614. [Abstract] [Full Text] [PDF]