DNA repair in Escherichia coli: Identification of the ... - Semantic Scholar

Proc. Natl Acad. Sci. USA Vol. 79, pp. 5616-5620, September 1982 Genetics

DNA repair in Escherichia coli: Identification of the uvrD gene product (recombinant DNA/corA/UV sensitivity/mutators/metE)

VALERIE F. MAPLES AND SIDNEY R. KUSHNER Department of Molecular and Population Genetics, University of Georgia, Athens, Georgia 30602

Communicated by Norman H. Giles, June 14, 1982

MATERIALS AND METHODS Materials. Reagents were obtained from the following sources: CsCl (technical grade), Penn Rare Metals; [a-32P]dATP (400 Ci/mmol) and [mS]methionine (1,400 Ci/mmol), Amersham (1 Ci = 3.7 X 101° becquerels); agarose, FMC; acrylamide, NN'-methylenebisacrylamide and N,N,N',N'-tetramethylethylenediamine, Bio-Rad; chloramphenicol, tetracycline, kanamycin, ampicillin, and lysozyme, Sigma; protein molecular weight standards, Pharmacia. Other chemicals were of analytical grade. Restriction endonucleases were purchased from either New England BioLabs or Bethesda Research Laboratories. E. coli DNA polymerase I was obtained from Boehringer Mannheim. Phage T4 DNA ligase was the generous gift of D. Prasher and M. Bittner. The Klenow fragment of E. coli DNA polymerase I was obtained from New England Nuclear. Bacterial Strains and Plasmids. The relevant genotypes and origins of the plasmids and bacterial strains used are listed in Tables 1 and 2. Bacterial nomenclature conforms to the suggestions ofBachmann and Low (13). SK4702 carries the original dominant uvrD3 allele. SK729 contains uvrD3 plus a second site mutation, designated uvr-253. Thus the mutations in SK729 [uvrD3 uvr-253, in accordance with the nomenclature of Siegel and Race (7)] lead to phenotypic UV sensitivity but are complemented by the F' SKF103 (uvrD+) plasmid. GW2001, a UVsensitive strain of E. coli containing a Tn5 insertion near metE, was obtained from G. Walker. Genetic Techniques. Cells were grown in Luria (L) broth (14) or K medium (15). For solid medium 2% agar was added. Appropriate antibiotics were added to either liquid or solid medium in the following concentrations: ampicillin, 20 ,ug/ml; tetracycline, 20 ,ug/ml; kanamycin, 50 ,g/ml; chloramphenicol, 20 ,ug/ml. Quantitative UV survival analysis was carried out as described by Kushner (14). Doses were determined with a no. 16569 GE germicidal UV intensity meter. The assay for lactose-utilizing (Lac') recombinants was as outlined by Zieg et aL (8). Strains were transformed with various plasmids by using the procedure of Kushner (16), except that CaCl2 was 15 mM instead of 50 mM when transforming corAl strains. corAl strains were unable to grow on rich medium supplemented with 100 mM CaCl2 (17). DNA Preparation. SKF103 plasmid DNA was prepared from 10-liter cultures of SK745 by the method of Davis and Vapnek (18). Other plasmid DNAs were prepared by the technique of Ish-Horowicz and Burke (19). Rapid plasmid screens used the procedure of Meagher et al. (9), except that the ether extraction step was omitted and the ethanol pellet was washed three times with cold (-20°C) 66% (vol/vol) ethanol. Radioactively labeled plasmid DNA was prepared by nick-translation with E. coli DNA polymerase I as described by Rigby et al. (20).

ABSTRACT A 2.9-kilobase (kb) Pvu II DNA fragment that contains the uvrD gene of Escherichia coli K-12 has been cloned in both low-copy and multiple-copy plasmid vehicles. The low-copy uvrD plasmid (pVMK49) complements a variety of uvrD, uvrE, and recL mutations. In contrast, the same strains carrying the 2.9kb fragment in a multiple-copy plasmid (pVMK45) remain sensitive to ultraviolet light (UV). Additionally, pVMK45 transformants of wild-type E. coli are sensitive to UV and methyl methanesulfonate and appear to be recombination deficient. The cloned uvrD gene does not complement the dominant uvrD3 allele. The 2.9-kb Pvu II insert in these plasmids encodes a single 76,000-dalton protein, which, on the basis of insertional inactivation experiments with the TnlOOO transposon, must be the uvrD gene product. These data confirm earlier genetic analysis which suggested that recL, uvrE, and uvrD were all allelic. The direction of transcription of the uvrD gene has also been determined.

A series of mutations affecting spontaneous mutation frequency, sensitivity to ultraviolet light (UV), host cell reactivation, and sensitivity to alkylating agents has been independently identified and mapped between ilv and metE at minute 84.5 on the Escherichia coli genome. Ogawa et aL isolated a mutant with the dominant uvrD3 allele, which produced increased sensitivity to UV and methyl methanesulfonate (1). Siegel isolated a strain of E. coli that showed increased spontaneous mutation frequency and enhanced UV sensitivity. The mutation responsible for these defects was shown to map near metE and was originally designated mutU4 (2). At about the same time, Smirnov and Skavronskaya reported mutants (called uvrE) which had very similar properties (3). Additionally, Horii and Clark, using a different selection procedure, identified the recL locus after mutagenesis of a recombination-proficient strain of E. coli that was genotypically recB21 recC22 sbcB15 (4). Subsequent genetic analysis has suggested that uvrE, recL, mutU, and uvrD mutations are all alleles of the same gene (5-7). In order to facilitate the identification of the uvrD gene product and to prove that all the mutations described above mapped in the same locus, the uvrD gene was cloned in both low-copy and multiple-copy vehicles. As described in this communication, a 2.9-kilobase (kb) Pvu II fragment when present in a lowcopy plasmid complements uvrE, recL, and recessive uvrD mutations. Multiple-copy plasmids carrying the same fragment will not complement these same mutations, and, in fact, caused increased sensitivity to UV and methyl methanesulfonate in wild-type strains of E. coli. Analysis of minicells carrying the fragment has led to the identification ofthe 76,000-dalton uvrD protein. The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U. S. C. §1734 solely to indicate this fact.

Abbreviations: kb, kilobase(s); Lac+, lactose-utilizing. 5616

Genetics:

Proc. Natl. Acad. Sci. USA 79 (1982)

Maples and Kushner

5617

Table 1. Bacterial strains

Strain no. GW2001 P678-54 SK234 SK239 SK246 SK729 SK745 SK2333 SK3440 SK3442 SK3444 SK3451 SK3952 SK4351

Hi

+ + 4 4 4 4 4 4 4 4 4 4 4 4

SK4352

Hi

SK4702 SR393 * lacBK1 =

Hi

arg + + E3 E3 E3

Hi Hi Hi Hi Hi Hi Hi Hi

+

his

uvrD Tn5 + + + 3 3 + 3

recL

uvr + + + + 253 253

uvrE + + + 254 + +

+

+ 152 + + + + + + + + + + +

+

corA + + + + + + +

SKF103 pVMK18 pBR325 pVMK32 pVMK45

lacBKi ilvDM88 lacBKi metE46

Source or derivation G. Walker (9) (8) (8) (8) This laboratory This laboratory This paper This paper This paper This paper This laboratory This laboratory pVMK32 transformant of SK729 pBR325 transformant of SK729 This laboratory K. Smith

-

+

3

253

4

3

253

+

+

+

lacBK1 metE46

4 +

3 3

+ +

+ +

+ +

+ +

lacBKi metE46 leuB19 thyA36

Tn5

+

1 1 + + 1 +

lacBK1 metE46

pVMK32 pBR325

lacMS286080dIIlacBK1 (8).

Table 2. Plasmids

SKF103 pTR262 pVMK18 pVMK32 pVMK35 pVMK45 pVMK49 pVMK50 pVMK54pVMK58

leu pro thr minA lacBKI lacBKi lacBKi lacBKi metE46 lacBKi ilvDM88 metE46 recA56 lacBK1 metE46 lacBKi metE46 lacBK1 metE46 lacBKi metE46

+ + + + +

+ + +

incubation, enzymes were heat inactivated if possible. If not, the mixture was extracted with phenol and DNA was precipitated with ethanol. DNA pellets were resuspended in ligase buffer with 0.5 mM spermidine added. Blunt-end ligation mixtures also contained 1 mM hexamminecobalt(III) chloride and 1 mM ATP. Ligation mixtures were incubated overnight at 150C. Construction of Recombinant Plasmids. SKF103 plasmid DNA and pTR262 plasmid DNA (12) were simultaneously digested to completion with HindIII. The DNAs were dialyzed against ligase buffer prior to adding T4 DNA ligase. The ligated mixture was used to transform SK729 and tetracycline-resistant colonies were selected. Out of 18 tetracycline-resistant transformants 1 (SK2333) no longer required methionine for growth. The plasmid (pVMK18) isolated from this strain was shown to contain a 27-kb HindIII fragment which will be described in more detail elsewhere. pVMK18 was digested with Sal I and the fragments were ligated into the Sal I site of pBR325 (10). Chloramphenicol-re-

pDP3

Plasmid

+ + + + + + + +

253

Restriction and Ligation. Restriction endonuclease digestions and T4 DNA ligase reactions were performed in the buffers recommended by Bethesda Research Laboratories. After

Plasmid pBR325

Other markers*

Genotype bla+ tet+ cat+ kant

tra' ilv+ uvrD+ met+ cro+ tet' udp+ met' corA+ uvrDi cat+ bla+ corA+ uvrD' kan+ corA' uvrD+ bla+ uvrD+ kan+ uvrD' kan+ uurD' bla+ uvrD255:: TnlOOO uvrD259: :TnlOOO

Derivation or source (10) Derived from pDF41 (11) by D. Prasher (this laboratory) (6) (12) This paper This paper This paper This paper This paper This paper This paper

sistant transformants of SK3952 (corAl) were replica plated to media containing 100 mM CaC12. Plasmid DNA isolated from transformants phenotypically CorA' and UV-sensitive were shown to contain a 12-kb insert that had six Pvu II sites (pVMK32, see Fig. 1A). pVMK35 was obtained by ligating the 12-kb Sal I fragment from pVMK32 into the low-copy plasmid pDP3 [a derivative of pDF41 (11) that contains a single Sal I site, the kanamycin-resistance determinant and an F origin of replication]. Either partial or complete digestions of pVMK32 with Pvu II were performed and the fragments were religated. Plasmid DNA (pVMK45) isolated from one ampicillin-resistant UV-sensitive transformant contained a 2.9-kb E. coli fragment from pBR325 kb

6

2

4 PI

A

uvrD

corA

6 SI

8

lo

12

14

16

18

P Hp PPs ....II

Hp PI . CP lo

Bg P

S

lI I

5,7.--

B C

5~7 -8 kb

131 9

14A

Ps HpXBs III 11 k5 6 Tn5

1j.9

15,.4

BXII

BI EI 5,4

159

164

uvrD

FIG. 1. Restriction endonuclease cleavage map of the corA-uvrD region. The map was determined by using both agarose and polyacrylamide gel analysis. Solid lines indicate the locations of the pBR325 cloning vehicle and the corA and uvrD coding sequences. The direction of transcription of the uvrD gene is indicated by the arrow and was derived from the data described in Fig. 4. Abbreviations for restriction endonucleases are as follows: B, BamHI; Bg, Bgl II; Bs, BstEII; C, Cla I; E, EcoRI; Hp, Hpa I; P, Pvu II; Ps, Pst I; X, Xor II. (A) Restriction map of pVMK32. There are no Kpn I, Xba I, or Xho I sites in pVMK32. There is one Sph I site, but its precise location is not known. (B) Detailed restriction map of the 2.9-kb Pvu II fragment contained in pVMK45. pVMK45 is not cleaved by Sph I, Sst I, Ava I, or Nru I. (C) Locations of the TnIOOO and Tn5 insertions. Numbers refer to the plasmids generated by the specific insertion (see Table 2).

'5618

Proc. Nad Acad. Sci. USA 79 (1982)

Genetics: Maples and Kushner

pVMK32 (Fig. 1B) and the 2.6-kb Pvu II fragment of pBR325, which carries the gene for ampicillin resistance and the origin of replication. To clone the 2.9-kb Pvu II fragment of pVMK45 in the low-copy pDP3 vehicle, DNA was first digested with Sal I and the ends were filled in by using the Klenow fragment of DNA polymerase I (21). This vehicle was then ligated with Pvu II-digested pVMK45 DNA, regenerating two Sal I sites. Transformants of SK729 (uvrD3 uvr-253) that became kanamycin-resistant and UV-resistant were shown to contain the 2.9-kb fragment in either orientation relative to the cloning vehicle (pVMK49, pVMK50). TnlOOO insertions in pVMK45 were generated as described by Guyer (22). Restriction Endonuclease Mapping. Restriction endonuclease fragments were analyzed on 0.8%, 1.0%, or 1.2% agarose slab gels in either TEA (23) or TA buffer (40 mM Tris1HCl, pH 7.8/20 mM NaOAc/2 mM EDTA). DNADNA hybridization experiments were carried out as described by Schweizer et al. (24). Detection of Plasmid-Encoded Proteins in E. coli Minicells. E. coli strain P678-54 was transformed with various plasmids. Minicells were isolated as described by Roozen et aL (25) but labeled with [3S]methionine and analyzed on sodium dodecyl sulfate/polyacrylamide gels by the method of Meagher et al

(9). RESULTS Cloning of the uvrD Gene. When a mixture of SKFI03 plasmid DNA (metE+ uvrD' ilv+) and pTR262 plasmid DNA was digested to completion with HindIII, treated with phage T4 DNA ligase, and used to transform SK729 (metE46 uvrD3 uvr253), one methionine-independent transformant was obtained. The plasmid (pVMK18) contained in this strain (SK2333) was shown to carry a 27-kb HindIII fragment that complemented udp, metE, and corA mutants (data not shown). A comparison of previously obtained genetic mapping data (6) and the physical map of pVMK18 indicated that a 12-kb Sal I fragment should contain both corA and uvrD genes. Accordingly, a complete Sal I digest of pVMK18 was ligated with Sal I-digested pBR325 DNA and used to transform SK3952 (corAl). Plasmid DNA isolated from chloramphenicol-resistant CorA+ transformants contained the predicted 12-kb Sal I fragment. This plasmid (pVMK32) complemented corA but not the uvrD3 uvr-253 mutations found in SK729; i.e., the transformants remained UV sensitive. In addition, pVMK32 transformants of a uvrD+ strain (SK3952) became UV sensitive. In order to determine ifthe increased UV sensitivity ofwildtype strains resulted from increased levels of the uvrD+ gene product, the 12-kb Sal I fragment was transferred into the lowcopy cloning vehicle pDP3. The resulting recombinant, called pVMK35, was shown by replica plating to complement corAl and uvrD3 uvr-253 (also see Fig. 3). Recessive uvrD, uvrE, and recL mutations were also complemented by a2.9-kb Pvu II fragment derived from pVMK32 and ligated into the low-copy vehicle pDP3. These results were obtained when the fragment was in either orientation relative to the cloning vehicle (pVMK49, pVMK50; data not shown). However, a multiplecopy plasmid containing the same 2.9-kb Pvu II fragment (pVMK45) did not relieve the UV sensitivity ofauvrD3 uvr-253 strain and indeed rendered a uvrD+ strain sensitive to UV (see Fig. 3). Restriction Mapping of the corA-uvrD Region. pVMK32 was digested with a variety of restriction enzymes both singly and in combination. Additionally, a series of subclones (derived from pVMK32 with Pvu II, Bgl II, EcoRI, BamHI, and BstEII) was generated to help in localizing the corA and uvrD genes. The restriction map and locations of the corA and uvrD genes

within the 12-kb Sal I fragment are shown in Fig. 1. When pVMK45 plasmid DNA (Fig. lB) was used as a hybridization probe with Sal I -and Pvu II digests of wild-type E. coli DNA, DNA from uvrD, uvrE, and recL mutants, and SKF103 plasmid DNA, fragments identical in size were detected (data not shown). The sizes of the Pvu II fragments observed with DNA from GW2001 indicated the presence of the Tn5 transposon. Its location was mapped more precisely with additional restriction enzymes and is indicated in Fig. 1C. Phenotypic Properties of Strains Carrying Recombinant Plasmids. SK239 (uvrE254), SK234 (recL152), and SK246 (uvrD3 uvr-253) were transformed with the low-copy plasmid pVMK35 (carrying the 12-kb Sal I fragment). As a control pDP3 plasmid DNA was used separately to transform the same strains. Kanamycin-resistant transformants were purified and their sensitivity to UV was quantitatively determined. As shown in Fig. 2, SK239, SK234, and SK246 carrying pDP3 were very UV sensitive, but these same strains carrying pVMK35 were as UV resistant as the uvrD+ control, SK3440. Similar results were obtained when SK234, SK239, and SK246 harbored either pVMK49 or pVMK50 (the 2.9-kb Pvu II fragment inserted into pDP3) (data not shown). The plasmids pVMK35, pVMK49, and pVMK50 did not complement the dominant uvrD3 allele in SK4702 (transformants remained UV sensitive) (data not shown). Plasmids consisting of either the 12-kb Sal I (pVMK32) or the 102

102

lo'

lo1

-

100

101 10 ._.

i

10-1.

U G)

W-

102

10-2

.10-2

.

10-3

10-4

,

0

10

20

10-4

30

UV fluence, J/m2 FIG. 2. UV survival of uvrD, uvrE, and recL strains carrying the cloned uvrD gene in a low-copy plasmid. e, SK234 (recL152) containing pBR325; o, SK234 containing pVMK35; m, SK239 (uvrE254) containingpBR325; +,SK239 containing pVMK35; A, SK246 (uvrD3 uvr-253) containingpBR325; A, SK246 containingpVMK35; o, SK3440 (uvrD+) containing pBR325.

Genetics:

Maples and Kushner

Proc. Natl. Acad. Sci. USA 79 (1982)

2.9-kb Pvu II (pVMK45) fragments inserted into the multiplecopy vehicle pBR325 showed no qualitative evidence of complementation ofany ofthe recL, uvrE, or uvrD mutants tested. However, as shown in Fig. 3, SK729 (uvrD3 uvr-253) carrying pVMK45 was quantitatively slightly more resistant to UV than was the comparable pBR325-transformed control. Also, a wildtype strain (SK3952) transformed with pVMK45 was almost as sensitive to UV as was the uvrD3-uvr-253 strain containing the same plasmid (Fig. 3). Controls of SK729 (uvrD3 uvr-253) and SK3952 (uvrD+) transformed with pBR325 gave the expected UV survival curves (Fig. 3). Strains with mutations in recL, uvrE, and certain uvrD alleles have been shown previously to carry out increased levels of genetic recombination between two duplicated lactose operons (8). As shown in Table 3, SK4352 (uvrD3 uvr-253/pBR325) produced 4 times as many Lac' recombinants as did appropriate control strains (SK3952 and SK3440). SK3451, carrying Tn5 inserted into the uvrD gene, displayed a similar increase in the number of Lac' recombinants. The previously untested dominant uvrD3 allele did not cause an increase in Lac' recombinant formation, and in fact a reproducible 50% reduction was observed. Unexpectedly, the presence of multiple copies ofthe uvrD' gene in wild-type strains [SK3442(pVMK32) and SK3444(pVMK45)] produced a dramatic reduction in the numbers of Lac' recombinants. Additionally, pVMK45 transformants of SK729 (uvrD3 uvr-253) showed markedly reduced levels 102

5619

Table 3. Lactose recombination proficiency in various uvrD strains Strain no. SK3952 SK3440 SK3442 SK3444 SK4352 SK4351 SK3451 SK4702

Genotype uvrD+ uvrD' uvrD' uvrD+ uvrD3 uvr-253 uvrD3 uvr-253 uvrD254::Tn5 uvrD3

Plasmid

pBR325 pVMK32(uvrDr ) pVMK45(uvrD+) pBR325 pVMK45(uvrD+) -

Lac+ recombinants per plate 78 72 6 6 310 14 291 39

of Lac' recombinants even though they remained sensitive to UV (see Fig. 3). Identification of the uvrD Protein. pVMK45, as well as several derivatiVes with TnlOOO insertions in the 2.9-kb Pvu II fragment (Fig. 1C), was used to transform the minicell-producing strain P678-54. The autoradiograph of the [3S]methionine-labeled proteins is shown in Fig. 4. Lane 4 represents minicells containing pVMK45. Four major polypeptide species were evident, with molecular masses of 76,000, 32,000, 28,000, and 16,000 daltons. The 32,000- and 28,000-dalton species represent the ,B-lactamase protein encoded by pBR325, whereas the 16,000-dalton peptide seems to have arisen as a fusion product of pBR325 and the cloned E. coli fragment, because this polypeptide was not observed when the same 2.9-kb fragment was inserted into pBR322 or pDP3 (data not shown). Lanes 2, 3, 6, 1

2

3

4

5

6

7

lol

.:.

100 'MP

4-.

bD

--OM

r.

S'4

.-low-

lo-,

0 U, Go

1-

cJ4

10-2

0

10

20

30 UV

40

50

fluence, J/m2

FIG. 3. UV survival of various strains carrying the cloned uvrD in a multiple-copy plasmid. o, SK3440 (uvrD+/pBR325); o, SK3444 (uvrD+/pVMK45); +, SK3451 (uvrD3 uvr-253/pVMK45); A, SK3452 (uvrD3 uvr-253/pBR325). gene

_

FIG. 4. Autoradiograph of polypeptides synthesized in minicells by pVMK45 and its derivatives containing the TnlOOO transposon. Minicells were labeled with [35Slmethionine and analyzed by electrophoresis on sodium dodecyl sulfate/12.5% polyacrylamide gels. The autoradiograph was exposed for 2 days. The locations of the TnlOOO insertions in plasmids pVMK54-56 and pVMK58 are shown in Fig. 1C. The two proteins of 40,500 and 40,000 daltons appear to be encoded by the TnlOOO insertion sequence. Lane 1, P678-54 minicells containing no plasmid; lanes 2-4 and 6 and 7, P678-54 minicells containing, respectively, pVMK58, pVMK54, pVMK45, pVMK55, and pVMK56; lane 5, protein standards: phosphorylase b, 94,000 daltons; bovine serum albumin, 67,000 daltons; ovalbumin, 43,000 daltons; carbonic anhydrase, 30,000 daltons; soybean trypsin inhibitor, 20,100 daltons; a-lactalbumin, 14,400 daltons. Radioactive ink was used to mark the positions of the stained protein standards.

5620

Genetics: Maples and Kushner

and 7 represent proteins encoded from a set ofpVMK45 derivatives containing unique Tnl000 insertions. These plasmids do not make wild-type E. coli UV sensitive, presumably due to inactivation of the uvrD gene by the TnlO00 insertion. In all cases the intense band at 76,000 daltons disappeared. (The faint band at 76,000 daltons that remained was present in all lanes and was therefore presumed to be of minicell origin.) In lanes 2, 6, and 7 new polypeptides of 73,000, 55,000, and 50,000 daltons, respectively, appeared. This result indicated that the uvrD gene is transcribed from right to left as shown in the map in Fig. 1. DISCUSSION The results reported above indicate that the uvrD gene of E. coli K-12 is encoded on a 2.9-kb Pvu II DNA fragment that, when present on low-copy plasmids, complements known recessive uvrD, uvrE, and recL mutations (Fig. 2). This evidence confirms previous genetic analysis (6) and recent experiments by Oeda et al. (26) with a ApuvrD' phage that indicate these mutants represent a single locus. The fact that the 2.9-kb Pvu II fragment encodes a single 76,000-dalton protein species further supports this hypothesis. Because premature termination ofthis protein by TnlO00 insertions into the multiple-copy plasmid pVMK45 (Fig. 4) eliminated the increase in UV sensitivity observed in wild-type strains, we conclude that the 76,000-dalton protein is the translational product of the uvrD gene. Of interest is the fact that uvrD, uvrE, and recL mutants remained UV sensitive when the fragment was present in multiple copies (see Fig. 3 and ref. 26) but at the same time lost their hyper-Rec phenotype (Table 3). Furthermore, the presence of multiple copies of the fragment in wild-type strains led to enhanced sensitivity to UV (Fig. 3) and a dramatic reduction in the cells' ability to form Lac' recombinants (Table 3). Additionally, the cells became very sensitive to methyl methanesulfonate and defective in Pl-mediated transduction (data not shown). These results suggest that overproduction of the uvrD+ gene product inhibits the normally existing DNA repair and recombination systems of the bacteria. The phenotypic properties of strains carrying multiple copies of the uvrD gene product are somewhat similar to those of the strain carrying the dominant uvrD3 mutation (SK4702), which is sensitive to UV and methyl methanesulfonate. As shown in Table 3, however, in the uvrD3 (SK4702) strain the number of Lac' recombinants is not as reduced as in strains carrying pVMK45 (SK3444). Furthermore, SK4702 (uvrD3) is more UV sensitive than is SK3444 (uvrD+/ pVMK45) (data not shown). Thus it is not clear whether the UV sensitivity observed in SK3444 (uvrD+/pVMK45) arises because the increase in the -level of the wild-type uvrD+ protein mimics the dominant uvrD3 protein. It should be noted that when the cloned uvrD3 uvr-253 alleles are present in multiple copies, wild-type strains are as UV sensitive as is SK3444(uvrD+/ pVMK45) (data not shown). This finding indicates a possible

Proc. Natl. Acad. Sci. USA 79 (1982)

difference in the cause of the phenotypic properties associated with the various uvrD mutants versus the overproduction ofthe wild-type uvrD' protein. The results described above suggest that the uvrD protein plays some regulatory role in the repair of DNA damage. The isolation ofthe gene should now permit the identification of the biological function of this protein. Many of the seemingly contradictory. genetic results should be more easily understood once the biochemical activities of the protein are known. The authors thank A. Easton for his many helpful suggestions and D. Prasher for the gift of plasmid pDP3. This work was supported in part by Grants GM00048 and GM27997 from the National Institutes of Health to S.R.K. 1. Ogawa, H., Shimada, K. & Tomizawa, J. (1968) Mol Gen. Genet. 101, 227-244. 2. Siegel, E. C. (1973)J. Bacteriol 113, 145-160. 3. Smirnov, G. B. & Skavronskaya, A. G. (1971) Mol. Gen. Genet. 113, 217-221. 4. Horii, Z. I. & Clark, A. J. (1973) J. Mol. Biol 80, 327-344. 5. Smirnov; G. B. & Abdukhalykova, G. F. (1976) Sov. Genet. 12, 475-481. 6. Kushner, S. R., Shepherd, J., Edwards, G. & Maples, V. (1978) in DNA Repair Mechanisms, eds. Friedberg, E. C. & Fox, C. F. (Academic, New York), pp. 251-254. 7. Siegel, E. C. & Race, H. M. (1981) Mutat. Res. 83, 49-59. 8. Zieg, J., Maples, V. F. & Kushner, S. R. (1978)J. Bacteriol 134, 958-966. 9. Meagher, R. B., Tait, R. C., Betlach, M. & Boyer, H. W. (1977) Cell 10, 521-536. 10. Bolivar, F. (1978) Gene 4, 121-136. 11. Kahn, M., Kolter, R., Thomas, C., Figurski, D., Meyer, R., Remant, E. & Helinski, D. (1979) Methods Enzymnot 68, 268-280. 12. Roberts, T. M., Swanberg, S. F., Poteete, A., Riedel, G. & Backman, K. (1980) Gene 12, 123-127. 13. Bachmann, B. J. & Low, K. B. (1980) Microbiol Rev. 44, 1-56.

14. Kushner, S. R. (1974)J. Bacteriol 120, 1213-1218.

15. Weigle, J., Meselson, M. & Paigen, K. (1959) J. Mol Biol 1, 379-386. 16. Kushner, S. R. (1978) in Genetic Engineering, eds. Boyer, H. W. & Nicosia, S. (Elsevier/North-Holland, Amsterdam), pp. 17-23. 17. Park, M. H., Wong, B. B. & Lusk, J. E. (1976)J. Bacteriol 126, 1096-1103. 18. Davis, R. & Vapnek, D. (1976)J. Bacteriol. 121, 1148-1155. 19. Ish-Horowicz, D. & Burke, J. F. (1981) Nucleic Acids Res. 9, 2989-2998. 20. Rigby, P. W. J., Dieckmann, M., Rhodes, C. & Berg, P. (1977) J. Mol. Biol. 113, 237-251. 21. Shah, D. M., Hightower, R. C. & Meagher, R. B. (1982) Proc. Natl Acad. Sci. USA 79, 1022-1026. 22. Guyer, M. S. (1978)J. Mol Biol 126, 347-365. 23. Helling, R. B., Goodman, H. M. & Boyer, H. W. (1974)J. Virol 14, 1235-1244. 24. Schweizer, M., Case, M. E., Dykstra, C. C., Giles, N. H. & Kushner, S. R. (1981) Gene 14, 23-32. 25. Roozen, K. J., Fenwick, R. G., Jr., & Curtiss, R., III (1971) J. Bacteriol 107, 21-33. 26. Oeda, K., Horiuchi, T. & Sekiguchi, M. (1981) Mol Gen. Genet 184, 191-199.