Mol Gen Genet (1981) 183:314-317 © Springer-Verlag 1981

Deletion Map of the Escherichia coli K-12 dnaB Gene R o b e r t A. Sclafani 1'2. a n d James A. Wechsler 1 ** 1 Department of Biology, University of Utah, Salt Lake City, Utah 84112, USA 2 Department of Biological Sciences, Columbia University New York, New York 10027, USA

Summary. Twenty-four dnaB alleles have been ordered by deletion m a p p i n g of the Escherichia coli K-12 dnaB gene. The position of the alleles had no linear correlation with the k n o w n phenotypes of dnaB m u t a t i o n : fast shut-off, slow shut-off but immediate change in rate, and GroP. The relevance of the deletion m a p to future studies of the dnaB protein is discussed.

The dnaB gene product of Escherichia coli K-12 is required for D N A replication of the chromosome, of 2 bacteriophage and of 0X174 (Wechsler 1978; F a n g m a n and Feiss 1969; Steinberg and D e n h a r d t 1968). M u t a t i o n s in dnaB are the most c o m m o n dna-ts m u t a t i o n s and all but one are elongation-defective (Table 1). Some dnaB m u t a n t s are non-permissive for 2 even u n d e r conditions permissive for the cell - the G r o P phcnotype (Georgopoulos and Herskowitz 1971). In addition, m a p p i n g of the various m u t a n t s has led to the suggestion that there may be a r e c o m b i n a t i o n a l hot-spot within the gene (Schendel 1977). Because of the multiple phenotypes a n d the question of a possible hot-spot, a fine-structure m a p of the gene was of interest. We have constructed such a m a p using deletion mapping. Deletions were isolated on a 2-dnaB + specialized transducing phage using standard techniques, A 2ci857 dnaB + transducing phage, m a d e by Dr. Jeffrey Glassberg, was obtained from C. Georgopoulos and has been partially characterized (Sclafani et al. 198l). A 2ci60 dnaB + derivative was constructed by mixed infection of 594 with 2ci60 and 2ci857 dnaB + at 30C, Since 2ci857 makes only turbid plaques at 30C, and 2c160 c a n n o t grow on the G r o P strain, clear plaques formed on a G r o P (dnaB) m u t a n t under these conditions should be produced by recombin a t i o n to form a 2ci60 dnaB + phage. Clear plaques were picked, purified and the resultant phage shown to plaque efficiently on other G r o P strains and on a dnaB: : T n l 0 insertion mutant, RS116. The phage transduced all of the dnaB conditional-lethal m u t a n t s listed in Table 1 to dnaB +. (Also a 2 c I + dnaB + phage was isolated and used to form lysogens of all of the dnaB mutants. These lysogens all grew at restrictive temperature). The 2c160 dnaB + transducing phage will be referred to as 2dnaB + in the remainder of this report. RS116 is a d n a B : : T n l O (Plbac) strain which c a n n o t support *

Present Address: Department of Genetics, University of Wash-

ington, Seattle, Washington 98195, USA ** Present Address and offprint request to: The Public Health Research

Institute, 455 First Avenue, New York, N.Y. 10016, USA

0026-8925/81/0183/0314/$01.00

Table 1. Characteristics of dnaB mutants

Strain

dnaB

allele

165/59 FA21 El07 E391 BT1037 BT43 165/70 RSl16 E125 RS149 PC~ RSll7 BT454 RS162 BT1071 FA22 E279 GRll PC8 BT500 E194 E368 E173 GR7

59 21 107 391 1037 43 70 518::Tn10 125 266 6 567 :: TnlO 454 252 1071 22 279 42 8 500 194 368 173 558

Cessation a of DNA synthesis

GroP b

References ~

immediate rate change immediate rate change immediate immediate immediate immediate immediate immediate immediate immediate immediate slow* immediate immediate immediate immediate immediate immediate slow immediate slow n.d.

+ A + X (A) + + (A) + + + + + + + (A) A + + + A + B

[1] [4] [6] [10] [6] [10] [9] [1] [2] [13] [6] [14] [3] [11] [14] [5] [13] [1] [1] [8] [14] [9] [4] [6] [10] [3] [12] [5] [1] [6] [10] [6] [6] [I0] [7]

a Immediate=synthesis stops completely within five minutes at the non-permissive temperature; rate change =immediate change in rate of synthesis but complete cessation takes longer; slow=substantial residual synthesis; n.d. = n o t determined; *=initiation kinetics b + =plates 2~ +, ~ A and 2~B as well as a dnaB + strain; A=plates )~zA or 2~B but not 2~+; B=plates 2~B but not 2~A or 2~z+; (A)-plates 2~A or ,:07cBbetter than 2~ + but not as well as dnaB + parent; X=plates 2~A or 2~ +, but not 2~B; + + =The only dnaB mutant which supports growth of 2 at the nonpermissive temperature; - = PI bac lysogens are completely restrictive for 2 DNA replication at any temperature e References [1] Bonhoeffer and Schaller 1965; [2] Bonhoeffer 1966; [3] Kohiyama eta1. 1966; [4] Fangman and Novick 1968; [5] Carl 1970; [6] Wechsler and Gross 1971 ; [7] Georgopoulos and Herskowitz 1971 ; [8] Beyersman et al. 1971; Zyskind and Smith 1977; [9] Wechsler et al. 1973; [10] Lark and Wechsler 1975; [11] D'Ari etal. 1975; Ogawa 1975; [12] Schendel 1976 ; [13] Sclafani, Wechsler and Schuster (1981); [14] Sclafani and Wechsler (1981 a, b)

315 125 266 6 567

70 21

59 I

107 I

391 I

1037 43 I

I

518

I

I

I

454 252

8 /,2

1071 5 0 0 I

I

I

194 llmlm 368 m173m m5 5 8 I I I I

• 55-1

== 71-2 • 41-1

I

"

i 62-2 | 152-1 | 40-1

I

1 38-2

I I

" i,

I 49-1

I lOl-2 | 121-1

Bill

| 46-2,

aml|

mlII IIII BIIIIII

,

45-1 | 13-1, 25-2 ~| 110-2,134-1 | 138-1,120-2, 96 -2 ~ ~= B,,= 8-1, 36-1

growth, and no mutant 2 which can grow on this host can be isolated even after heavy mutagenesis (Sclafani et al. 1981). The 2dnaB ÷ specialized transducing phage can supply dnaB ÷ protein for itself during lytic growth and, therefore, makes plaques on R S l l 6 . Deletions of this phage were isolated as EDTA-resistant plaques on a wild-type strain according to the method of Parkinson and Huskey (1971). Each deletion was an independent isolate from a single plaque of ,~dnaB +. 2dnaB ÷ like its parent, ,~cI60 is sensitive to 0.6 m M E D T A in tryptone plates - the efficiency of plating is reduced by 1-5 x 10 -4. A large number of independent EDTA-resistant phage were screened for the ability to plate on R S l l 6 ; those that could not had presumably lost dnaB ÷ function. Phage which had lost functional dnaB ÷ activity were purified, rechecked, and used to make lysates. One hundred and fifty independent 2AdnaB phages were isolated in this manner. Several of these were further analyzed by restriction enzyme digestion and were shown to have lost D N A from the chromosomal insert in ,~dnaB + (data not shown; R.A. Sclafani, Ph.D. Thesis, Columbia University, 1981). Lambda lysogens of each of the E. coli K-12 dnaB mutants listed in Table 1 were isolated. One-tenth ml of a log phase culture of each of these 2 + lysogens was plated on nutrient media, and the various )~AdnaB phage were then spotted ( ~ 109 p.f.u.'s/spot) onto the plate. After 48 hours of incubation at the restrictive temperature, colonies appeared within the spots except in those cases in which the wild-type genetic material, corresponding to the mutant dnaB sequence, had been deleted. All spot tests were done two or more times on different days to ensure reproducibility. The results of the deletion mapping are shown in Fig. 1. Of the 24 dnaB alleles mapped, two are insertion mutations, two are suppressed nonsense mutations, and all others are missense mutations. Most of the alleles were separable from each other, though there was some apparent clustering toward the central region of the map. No mutational hot-spots were clearly evident. The dnaB ÷ specialized tranducing phage did not contain any other identifiable chromosomal marker (malB, ubiA, and uvrA were checked) and, thus, we could not orient the map with respect to outside markers. Because all of the deletions which overlap alleles 368, 173 and 558 also overlap the 21 other mutations, the placement of these alleles could equally well be to the extreme left rather than on the extreme right as shown.

Fig. 1. Deletion map of dnaB. The extent of material deleted is shown by the bars; numbers adjacent to the bars refer to independent deletions. Numbers above the line are allele numbers (see Table 1), Italicized numbers are GroP. Solid underline denotes minor uncertainty. Broken underline means these alleles could be at opposite end of gene. Box refers to more significant ambiguity

Alleles dnaB279 and dnaB22 proved difficult to order unequivocally as each gave low and somewhat variable numbers of recombinants with some deletions (152-1, 38-2 and 49-1); however, we have interpreted the data on the basis of recombination frequencies between each allele and these deletions as indicative of their proximity to the ends of the deletions, and have ordered them accordingly. The relative positions of the members of two pairs, alleles i07 and 391 and alleles 1037 and 43 are also based on differences in recombination frequencies with particular deletions (55-1, 62-2, and 101-2) however, in these cases distinction was between no recombination and some recombination. The assigned order is, therefore, probably correct though not absolutely certain. The order, dnaB42, dnaBTO, dnaB391, differs from the sequence determined by Schendel using F's but is in agreement with his transductional data (Schendel 1977). Those dnaB alleles which exhibit a GroP phenotype (Table 1 ; italicized numbers on the map) are scattered throughout the locus. Though the single GroPB mutant mapped, allele 558, maps at the very end of the gene, the significance of this position is uncertain; attempts to position several other GroP mutants strongly suggested that they were double mutants (Sclafani, unpublished). There were also some GroP ÷ dnaB mutants which appeared to have more than one lesion (data not shown). The few mutations which are leaky in that, though D N A synthesis does not cease immediately, the rate of synthesis changes quickly after a shift to restrictive temperature, also have no particular correlation with map position. Though these alleles are well to either side of the center of the map, all, but allele 173, are bracketed by immediate stop mutants. The single initiation-defective phenotype dnaB allele, dnaB252, lies in the central portion of the map amidst many elongation defective mutations and in the same deletion segment as two of these. We have suggested elsewhere that the dnaB252 allele inactivates only one function of the dnaB product and, specifically, the ability to complex with dnaC protein (Sclafani and Wechsler 1981a). The dnaB252 product is the only dnaB mutant product know to retain the associated DNA-dependent ATPase activity (Lanka et al. 1978). It is quite possible that some, or even all, of the elongation-defective dnaB alleles also inactivate the initiation function of the product, but as the elongation defect would necessarily be phenotypically dominant, there is, as yet, no way to ascertain if this be the case.

316 The amber mutations, dnaB125 and dnaB266, are not separated by the deletions obtained. They cannot be at the same site, however, as in the same genetic background, these two alleles exhibit several distinct phenotypes: 1) a dnaB125 supD strain is G r o P - while a dnaB266 supD strain is GroP +, 2) the GroP phentotype of dnaB266 supE is dependent on the state of the rpsL gene - rpsL + are G r o P + and rpsL are G r o P - , and 3) dnaB266 supF strains are inefficiently lysogenized by Plbae and lysogens, when obtained, grow poorly at 42 ° C whereas dnaB125 supFare lysogenized as efficiently as the dnaB + parent and growth is normal in the 25 ° C to 42 ° C range (Ogawa 1975; Sclafani and Wechsler 1981b). Of the four mutations that require suppression by P l b a c prophage for survival - the two ambers when in sup + strains and the two TnlO insertions - three lie in one segment and one in the neighboring segment of the map. It is possible that production of the amino-terminal segment of the dnaB protein is a minimal requirement for P l b a c supression. Alternatively, the apparent clustering of these mutations could be for trivial reasons. The d n a B 5 6 7 : : T n l 0 allele almost certainly results in production of the leading portion of the protein or, via an insertion caused promoter, of the terminal section of the product. This conclusion is based on the observed negative complementation of dnaB + by dnaB567::TnlO in the presence of Plbac (Sclafani et al. 1981). The various phenotypes, in addition to the gross inhibition of D N A synthesis, exhibited by dnaB mutants presumably reflect defects in one or more of the biochemical properties associated with the dnaB protein: 1) the DNA-independent and/or D N A dependent ATPase activities (Wickner et al. 1974), 2) the formation of homomultimers apparently required to generate the true active form of the protein (Lark and Wechsler 1975, RehaKrantz and Hurwitz 1978; Ueda etal. 1978), 3) the binding to dnaCprotein (Wickner and Hurwitz 1975; Schuster et al. 1977), 4) the interaction with the 2 bacteriophage P gene product during 2 replication (Georgopoulos and Herskowitz 1971 ; Klinkert and Klein 1979), 5) the promotion of dnaG primase a c t i v i t y (McMacken et al. 1977; Wickner 1978), and 6) a function in D N A replication independent of dnaG action (Staudenbauer et al. 1979; Sclafani and Wechsler 1981 c). The initial conclusion drawn from the deletion map is that the known properties conferred by different dnaB alleles do not provide evidence for specific domains in the primary protein structure. This conclusion is not surprising in spite of attempts to derive a unifying theory for multifunctional proteins based on such domains (Kirschner and Bisswanger 1976). The three dimensional conformation of a protein depends on secondary, tertiary and, in oligomeric proteins, on quaternary structure. Thus, functional domains, when they can be differentiated, need not be a reflection of primary sequence. Furthermore, the degree of multifunctionality of the dnaB protein may well be less than it appears from the catalog of its characteristics. The interactions with dnaC product and with 2 P gene product may by nearly identical as 2 initiation requires P protein and not dnaC protein, whereas E. eoli initiation requires dnaC protein. Similarly, the promotion of dnaG activity and the action independent of dnaG may be closely coupled. The function of the ATPase activities could be separate or could be necessary for one or all of the other properties. There are few examples in which a series of dnaB alleles have been analyzed with respect to a particular property other than defective D N A replication. Though GroP mutations are not clustered, all dnaB mutations may be defective in interacting with the 2 P product but only some, the GroP mutants, so

defective as to exhibit the defect at 30 ° C. Similarly, as mentioned earlier, all dnaB alleles may affect the initiation process but this characteristic would be obscured by the elongation defect. The deletion map should serve as a useful guide for choosing particular mutants or mutant proteins for further studies to clarify the specifi c properties of the dnaB protein. Acknowledgements. We are grateful to Costa Georgopoulos for suggesting the use of 2, to Sandy Parkinson for helpful technical comments, and to Ira Herskowitz for bacterial strains. We thank Gordon Lark and C.I. Davern for their hospitality. These experiments were conducted by R.A. Sclafani in partial fulfillment of the Ph.D. requirements of Columbia University. This research was supported by Public Health Service Research Grant No. GM25476 from the National Institute of General Medical Sciences.

References

Bonhoeffer F (1966) DNA transfer and D NA synthesis during bacterial conjugation. Z Verebungsl 98 : 141-149 Bonhoeffer F, Schaller H (1965) A method for selective enrichment of mutants based on the high UV sensitivity of DNA containing 5-bromouracil. Biochem Biophys Res Commun 20:93 97 Fangman WL, Feiss M (1969) Fate of 2DNA in a bacterial host defective in DNA synthesis (FA22) J Mol Biol 44:103 116 Fangman WL, Novick A (1968) Characterization of two bacterial mutants with temperature-sensitive synthesis of DNA. Genetics 60:1-17 Georgopoulos C, Herskowitz I (1971) Escherichia coli mutants blocked in lambda DNA synthesis. In : Hershey AD (ed) The Bacteriophage lambda. Cold Spring Harbor Laboratory, Cold Spring Harbor NY, pp 553-564 Kirschner K, Bisswanger H (1976) Multifunctional proteins. Annu Rev Biochem 45:143 166 Klinkert J, Klein A (1979) Cloning of the replication gene P of bacteriophage lambda: effects of increased P-protein synthesis on cellular and phage DNA replication. Mol Gen Genet 171:219-227 Kohiyama M, Cousin D, Ryter A, Jacob F (1966) Mutants thermosensibles D'Escherichia coli K-12. Ann Inst Pasteur 110 :465-486 Lanka E, Geschke B, Shuster H (1978) Escherichia coli dnaB mutant defective in DNA initiation: Isolation and properties of the dnaB protein. Proc Natl Acad Sci USA 75 : 799-803 Lark KG, Wechsler JA (1975) DNA replication in dnaB mutants of E. coli: gene product interaction and synthesis of 4S pieces. J Mol Biol 92:145-163 McMacken R, Ueda K, Kornberg A (1977) Migration of Escherichia coli dnaB protein on the template DNA strand as a mechanism in initiating DNA replication. Proc Natl Acad Sci USA 74:41904194 Ogawa T (1975) Analysis of the dnaB function of Escherichia coli K12 and the dnaB-like function of P1 prophage. J Mol Biol 94:327340 Parkinson JS, Huskey RJ (1971) Deletion mutants of bacteriophage lambda. I. Isolation and initial characterization. J Mol Biol 56: 369384 Reha-Krantz LJ, Hurwitz J (1978) The dnaB gene product of Escherichia coli. I. purification, homogeneity, and physical properties. J Biol Chem 253:4043-4050 Schendel PF (1976) Identification of a genetic locus affecting chromosome stability and cellular survival in a dnaB mutant J Bacteriol 127 : 698-705 Schendel PF (1977) Genetic analysis of the dnaB region of Escherichia coli K-12: The identification of a recombinational hot spot. Mol Gen Genet 156:281-287 Schuster H, Schlicht M, Lanka E, Mikolajczyk M, Edelbluth C (1977) DNA synthesis in an Escherichia coli dnaB dnaC mutant. Mol GenGenet 151:11 16 Sclafani RA, Wechsler JA (1981a) Deoxyribonucleic acid initiation

317 mutation dnaB252 is suppressed by elevated dnaC + gene dosage. J Baeteriol 146:418-421 Sclafani RA, Wechsler JA (1981b) dnaB125, a dnaB nonsense mutation. J Bacteriol 146 : 1170-1173 Sclafani RA, Weehsler JA (1981 c) DNA replication intermediates synthesized by lysates of dnaB, dnaG, and dnaB dnaG mutants in vitro. Mol Gen Genet 182:95 08 Sclafani RA, Wechsler JA, Schuster H (1981) The isolation and characterization of Eseherichia coli dnaB:: TnlO insertion mutations. Mol Gen Genet 182:112 118 Staudenbauer WL, Scherzinger E, Lanka E (1979) Replication of the colicin E1 plasmid in extracts of Escherichia coli: uncoupling of leading strand from lagging strand synthesis Mol Gen Genet 177 : 113-120 Steinberg RA, Denhardt DT (1968) Inhibition of synthesis of 0X174 DNA in a mutant host thermosenstive for DNA synthesis. J Mol Biol 37 : 525-528 Ueda K, McMacken R, Kornberg A (1978) dnaB protein of Escherichia eoli. J Biol Chem 253:261-269

Wechsler JA (1978) The genetics of E. coli DNA replication. In: Molineux I, Kohiyama M (eds) DNA synthesis: present and future. Plenum Press, New York, pp 49-70 Wickner SH (1978) DNA replication proteins of Escherichia coli and phage 2. Cold Spring Harbor Syrup Quant Biol 43:303-310 Wickner S, Hurwitz J (1975) Interaction of Escherichia coli dnaB and dnaC(D) gene products in vitro. Proc Natl Acad Sci USA 72:921925 Wickner S, Wright M, Hurwitz J (1974) Association of DNA-dependent and -independent ribonueleoside triphosphate activities with dnaB gene product of Escherichia coli. Proc Natl Acad Sci USA 71:783-787

C o m m u n i c a t e d by W.H. Boyer

Received June 17, 1981