MG13

Molec. gen. Genet. 151, 319-326 (1977)

© by Springer-Verlag 1977

Mechanism of Conjugation and Recombination in Bacteria XVI Single-Stranded Regions in Recipient Deoxyribonucleic Acid during Conjugation in Escherichia coli K-12 Hanna Bialkowska-Hobrzafiska and Wtadystaw J.H. Kunicki-Goldfinger Department of General Microbiology, Institute of Microbiology, Warsaw University, Nowy Swiat 67, 00-046 Warsaw, Poland

Summary. The formation of mating pairs between F- and Hfr cells resulted in increased sensitivity of recipient deoxyribonucleic acid (DNA) to singlestrand-specific $1 nuclease, from 3.6% to 23.5% after 30 rain conjugation. A comparable amount of single strand regions in the DNA of mated wild type and recA mutant cells was detected. 10 rain of conjugation resulted in almost the same amount of single-strand recipient DNA as 30 rain of continuous transfer of donor DNA. Also the transfer of plasmid DNA from, F+recA strain led to the occurrence of single-strand recipient DNA. In similar experiments with Hfr tra mutant no such effect was observed. We conclude that alterations in the secondary structure of recipient DNA are induced in the early phases of conjugation associated with the formation of mating pairs and/or initiation of transfer donor DNA.

Introduction The molecular mechanism of recombination following bacterial conjugation is at present only partially understood and the initiation of genetic recombination remains one of the most obscure aspects of this process. Previous molecular studies on recombination in the E. coli conjugation system have supplied physical evidence supporting at least the temporary existence of a single-stranded structure of entering donor DNA (Piekarowicz and Kunicki-Goldfinger, 1968; Cohen et al., 1968 ; Rupp and Ihler, 1968 ; Vielmetter, Bonhoeffer and Schiitte, 1968). In the Bacillus subtilis and Haemophilus influenzae transformation systems the initiation of recombination seems to be connected with the conversion of transforming DNA (doublestrand) to a completely single-stranded form (e.g. Piechowska and Fox, 1971 ; Davidoff-Abelson and Dubnau, 1973b) or to DNA with single-stranded regions

(Sedgwick and Setlow, 1976), respectively, and also with the existence of local single-stranded regions in the recipient DNA of both organisms (Harris and Barr, 1969, 1971 ; LeClerk and Setlow, 1975). According to the hypotheses of Kunicki-Goldfinget (1968, 1971), and Curtiss (1969), initiation of genetic exchange, following conjugal transfer, requires activation mechanisms which could entirely or in part alter the secondary and tertiary structure of both parental molecules of DNA. The experiments described here were designed to investigate the physical state of recipient DNA prior to and/or during integration of donor DNA. Of particular interest was whether recipient singlestrand DNA occurs as a result of conjugation. The basic method for the determination of the secondary DNA structure involved the use of nuclease $I preparation. Nuclease St from Aspergillus oryzae is strongly specific for single-stranded DNA in mixtures of both single and double strands (Ando, 1966; Godson, 1973; Vogt, 1973; John, Johnson and Bonner, 1974). It has recently been found that nuclease $1 attacks double-stranded DNA in regions where single strands occur due either to local melting, or formation of "loops" or at fi'ee ends adjacent to single strand nicks in the duplex, because of both endonucleolytic and exonucleolytic activity (Sutton, 1971; Germond, Vogt and Hirt, 1974; Miller and Wetmur, 1975). This enzyme thus appears to be the best suited for the study of any alterations in the double helical structure of DNA.

Materials and Methods Bacterial Strains and Bacteriophage. All bacterial strains employed in this work are derivatives of Escherichia coli K-12. Their relevant properties are listed in Table 1. Laboratory stock of male-specific phage f2 was used.

320

H. Biatkowska-Hobrzafiska and W.J.H. Knnicki-Goldfinger: Single-Stranded Regions in F - D N A of E. coli K-12

Table 1. Strains of Escherichia coli K-12 Strain

Mating type

Relevant genotype

Source, references

H64

HfrH

thi thy str s

S.J. A1ichanian

P721

HfrC

met tra str s

R.A. Skurray (Cooke, Meynell and Lawn, 1970)

W1655

F+

metB str s

(Lederberg and Lederberg, 1953)

AB2463/A

F+

as AB2463

This laboratory by a cross W1655 x AB2463

108

F-

met pro thy str R P.H. Pritchard

ABl157

F-

thr leu thiproA his argE str R

A.J. Clark (Willetts and Clark, 1969)

AB2463

F

as ABl157 but recA13

A.J. Clark (Willetts and Clark, 1969)

The gene symbols used are those recommended by Taylor and Trotter (1972) Media. Minimal liquid and solid medium (MD) was prepared according to Davis (1950). The following final concentrations of growth supplements and drugs were added to this medium whenever necessary: D L t h r e o n i n e - 20 gg/ml, all other aminoacids (L-form) ( R e a n a l ) - 1 0 g g / m l , vitamins (Nutritions Biochemical C o r p . ) - 1 ~tg/ml, thymine (Koch Light Laboratories L T D ) - 30 ~tg/ ml, Difco C a s a m i n o a c i d s - 0 . 2 % (only to liquid medium), streptomycin ( P o l f a ) - 2 0 0 gg/ml (for selection of recombinants). For radioisotope labeling the M D medium was supplemented with 3Ht h y m i d i n e - l . 5 gCi/0.018 gg/1 ml (from UV VVR Czechoslovakia). For thymine independent strains M D medium containing 3H-thymidine was supplemented with 250 gg adenosine (Schuchardt) per ml. Preparation o f Bacterial Cultures. A single colony of each of the appropriate strains was subcultured (5 ml of medium in 100 ml flask) and incubated overnight at 37 ° . Logarithmic cultures were prepared by dilution of an 16-18 h non-areated culture 20-fold in fresh, warm appropriately enriched M D medium (about 50 ml in 100 ml flask) and then incubated with shaking for 2 ~ . 5 h to optical density (OD) of 0.25 at 550 nm, corresponding to viable count of about 2 x 108 cells/ml. The donor cell suspension was always grown with gentle shaking to O D at least 0.05 units higher than that of F cell suspension. Labeling of Recipient Cells with 3H-thymidine. F - (ABl157 or AB2463) cells were grown for 16-18 h at 37 ° in M D medium containing 3H-thymidine with specific activity 1.5 gCi/0.018 gg/ l ml. After overnight incubation, cells were diluted 1:20 in fresh radioactive medium and incubated for about 2.5 h. Bacteria were then separated from the radioactive medium by filtration through coli 5 membrane filters (Warszawska Wytw6rnia Surowic i Szczepionek) and washed several times with M D solution without any supplements with the exception of cold thymidine at a final concentration of 50 gg/ml. Bacteria were then resuspended in appropriately enriched prewarmed M D medium and adjusted to approximately 2 x 108 cells/ml. Recipient cells prepared in this way were mated with unlabeled Hfr cells as described in the next section. Mating Procedure. According to the proposal of Hayes (1957) parental cultures from about the middle part of the logarithmic phase

of growth were mated at ratio 1 : 1 or with a slight excess of donor cells, in non radioactive medium (20 ml volume in 300 ml erlenmeyer flask). Mating was carried out in a water-bath at 37 ° for appropriate time without shaking. If F + was used as a donor strain, incubation was carried out at 42 ° as suggested by Achtman, Willets and Clark (1971) to increase the effectivity of transfer. The conjugation samples were withdrawn, cooled in an ice-bath and mating pairs were disrupted with the use of Waring blendor (2 min at 60,000 rpm). The cell suspension was immediately diluted five-fold in Tris-HC1 ~ buffer, p H 8 or 0.15 M NaC1 plus 0.1 M E D T A , p H 8, at 4 ° C, depending on further preparation of cell lysates. Preparation of Lysates. Two methods were used for the preparation of cell lysates. In the first method the cell pellets were resuspended in one-fourth the original volume of Tris-HC1, pH 8, buffer and then treated at 56 ° with 1% SDS (Koch Light Laboratories) for 10 min. The mixture was incubated with 0.2 m g / m l of nuclease-free Pronase (Calbiochem) at 40 ° for 2 h. The lysates obtained this way were used without any additional treatment as a substrate in S~ nuclease reaction. In the modification of the Piechowska and Fox (1971) method cells were washed once with cold 0.15 M NaC1, 0.1 M E D T A , pH 8 and twice with a solution of a 0.15 M NaC1, 0.01 M E D T A , pH 8. The washed cells were suspended in one-fourth their original volume of the latter solvent and then 10 mg]ml lysozyme (Reanal) was added. The mixture was incubated at 2 ° for 30 min with several numbers of 1 min interruptions at 37 °. Lysis was completed by addition of Sarkosyl N L (Ciba-Geigy) to a final concentration of 1% and heating for 10 min at 56 °. The lysate was then incubated for 2 h at 40 ° with 1 mg/ml of nuclease-free Pronase (Calbiochem). NaC1 was then added to a concentration of 4 M and samples were heated at 70 ° for 20 min. The lysate was cooled and dialysed overnight against two changes of 0.3 M NaC1, p H 7. $1 Nuclease Purification. Takadiastase powder (Sankyo Co. Ltd., Tokyo) from A. oryzae was used as a source of endonuclease S~ and was stored at 2 ° until use. The enzyme was partially purified by a one-step fractionation procedure on a diethylaminoethyl-celulose column ( W h a t m a n DE-53) according to the method of Ando (1966). Fractions which exhibited more than 90% digestion of denatured calf t h y m u s D N A (Sigma) but no more t h a n 5% activity on double stranded D N A , were pooled and kept at - 2 0 ° C. The S~ nuclease preparation was stable for at least 4 m o n t h s under these conditions. SI Nuclease Reaction. The final method adopted from Crosa, Brenner and Falkow (1973) for nuclease assay was as follows. The standard reaction mixture, containing in 1 ml: native or thermally denatured radioactive D N A in cell lysate (corresponding to 1.3 ml sample of bacterial culture), about 20 gg of thermal denatured calf t h y m u s D N A (Sigma), 0.03 M sodium acetate buffer, pH 4.6, 0.1 m M ZnSO4, 0.2 M NaC1, $1 nuclease 800 U, was incubated at 55 ° for 30 min. The reaction was terminated by cooling in ice water, followed by addition of an equal volume of ice-cold 10% TCA. Degradation of single stranded D N A was measured by reduction in acid precipitable radioactivity. The control assays of enzyme activity were done in the above conditions but about 10 ~tg/ ml E. coli a H - D N A isolated by modification of Piechowska-Fox method and purified by Sephadex G-100 filtration (Ayad, 1972) was substituted for the cell lysate. One unit of nuclease $1 activity (U) is defined according to Ando (1966) as the a m o u n t of enzyme that degrades 1 ~tg denatured calf thymus D N A (Sigma) to T C A 1 Abbreviations used: Tris, tris(-hydroxymethyl)aminomethane; E D T A , ethylenediaminetetraacetic acid; SDS, sodium dodecyl sulphate; TCA, trichloroacetic acid

H. Biatkowska-Hobrzafiska and W.J.H. Kunicki-Goldfinger: Single-Stranded Regions in F - D N A of E. coli K-12

321

solubility in the standard conditions. The qnantity of T C A soluble fraction of D N A was determined by the method of Dische (1955).

Table 2. Comparison of the action of nuclease $1 on various pre-

Assay of Radioactivity Insoluble in Cold TCA. To standard reaction mixture with $1 nnclease, an equal volume of ice-cold 10% T C A and 0.i ml 1% bovine serum albumin solution as a carrier were added. After storage for 14-16 h at 4 °, the precipitate was centrifuged and dissolved in 0.1 ml 8% K O H . The samples were transferred whole to 22 m m diameter W h a t m a n 17 M cellulose discs (Carrier and Setlow, 1971), each tube being washed twice with 0.1 ml of distilled water. The dry filters were washed once with cold 10% T C A for 30 min, twice with 5% T C A for 15 min and then twice with 95% ethanol for 15 rain, all steps at 4 °. The filters were dried, taken into 5 ml of toluene-based scintillator (40 ml Permafluor I "Packard Instrument C o m p a n y " p e r liter of toluene), and the radioactivity measured in Beckman LS-355 Liquid Scintillation Counter.

Substrate preparation"

parations of native (DS) and heat-denatured (SS) E. coli D N A TCA-precipitable material after Sl-sen-

30min incubation (3H cpm)

sitive fraction (%)

without S1 nuclease

with 81 nuclease

mean-+s.e, b

mean_+ s.e. b

Heating cells with DS SDS and digestion of pronase SS

23,583_+3.4%

22,582_+ 1.4%

4.2

24,511 _+0.9%

563 + 4 . 7 %

97.7

Modification the Piechowska-Fox method

DS

16,983-+0.9%

15,274_+0.1%

6.5

SS

16,983-+ 1.5%

177+5.2%

99.0

As above but DS D N A purified by Sephadex G100 filtration (control) SS

1489_+1.5%

1431 ___0.6%

3.9

Denaturation of DNA. For thermal denaturation a small volume of cell lysate or D N A was heated in boiling water for 20 min and then rapidly chilled in ice.

2778_+4.3%

99_+0.1%

96.4

Results

All the substrates were prepared from 3H-labeled recipient cells. For details see Materials and Methods b These values were always determined at least in triplicate. The value of the standart error never exceeded 5%

Usefulness of the St Nuclease for the Analysis of the Secondary Structure of E. coli DNA in Cell Lysates In preliminary experiments the substrate specificity of nuclease $1 was analysed with regard to native and heat-denatured DNA to determine whether the extent of hydrolysis depends on the method of substrate preparation. The results obtained with a long experimental series, shown in Table 2, imply that after partial purification (DEAE-cellulose step), nuclease $1 specifically degrades, under the assay conditions, over 95% of the single-stranded material, this being independent of the method of cell lysate preparation. The observed small amount (about 5%) of double-stranded hydrolysis activity approximates that reported recently (Crosa, Brenner and Falkow, 1973; John, Johnson and Bonner, 1974). This indicates, moreover, that no artifactual changes in structure of native DNA take place when extracted by modification of the Piechowska and Fox (1971) method and that used in our laboratory. To make our study more complete, we checked S~ nuclease digestion of denatured 14C-DNA E. coli purified by Sephadex G-100 (Ayad, 1972) in the presence of undenaturated cell lysates, containing 3H-DNA, obtained by the procedures employed in this work. We found, that about 95% of the 14C-DNA was degraded by $I nuclease in reaction mixture with recipient cell lysate prepared by the Piechowska-Fox method. $1 nuclease degraded about 80% of the t*C-DNA if recipient cell lysate obtained according to the method used earlier in our laboratory was present. It is possible that the difference in the degree of $1 nuclease digestion may reflect

a residual activity of DNA binding (unwinding) protein in the latter sample (Sigal et al., 1972; Molineux, Friedman and Gefter, 1974) which reduces the activity of $1 endonuclease (Molineux, Gefter, 1975). Basing on the results summarized herein as well as the recent findings concerning the characteristic activity of endonuclease $1 (Sutton, 1971; Germond, Vogt and Hirt, 1974; Barth and Grinter, 1975; Miller and Wetmur, 1975) we think it justified to regard the S~ reaction as an objective probe for secondary DNA structure analysis.

Structure of Recipient DNA after Uptake of Donor (Hfr) DNA. To detect any effect of conjugation upon the physical state of recipient DNA, several crosses with recipient F - A B 1157 labeled with 3H-thymidine prior to mating and unlabeled donor Hfr H64 in non radioactive mating medium were made. This system exhibited a good yield of recombinants (always more than 10% Thr Leu Str ~ recombinants after 90 min of mating). The employed ratio of Hfr to F- greater than 1 : 1 ensured maximum engagement of recipient cells in the uptake of donor DNA without occurrence of lethal zygosis. In additional experiments, not reported here, we find that in such conditions a number of recipient cells during 90 min conjugation remains constant. Lysates of mating samples were prepared by heating in the presence of SDS and pronase (for details see Materials

322

H. Biatkowska-Hobrzafiska and W.J.H. Kunicki-Goldfinger: Single-Stranded Regions in F

and Methods). The manner of preparation of cell lysate used here ensures good recovery of recipient DNA and avoids a decrease of nuclease activity compared to purified E. coli DNA (see Table 2). The average results of $1 nuclease analysis presented in Table 3 clearly suggest the existence of single-stranded regions in native molecule of recipient DNA during conjugation, shown by the presence of $1 nuclease degradable material. The relatively greatest alteration of secondary structure of recipient DNA seems to be associated with 45 rain of conjugation, after which a decrease of the amount of S~-nuclease-sensitive fraction of recipient DNA is observed. To augment the recovery of single-stranded recipient material we used Piechowska-Fox (1971) extraction procedure (in a slight, modification). According to the suggestion of these authors and Davidoff-Abelson and Dubnau (1973 a) lysozyme demonstrates a protective action for single-stranded DNA at the time of lysis. Table 3 summarizes a comparison of results obtained with samples treated either by the method previously used in our laboratory or that of Piechowska and Fox. The latter procedure of nucleic acid preparation allows an apparent greater recovery and better reproducibility of results than the former one. After 30 min of conjugation, 23.5% of single-stranded recipient DNA was detected (with the first procedure for lysates only 16%). Thus the Piechowska-Fox procedure was used in further assays of the fate of recipient DNA.

Structure of recA Recipient DNA after Uptake of Donor (Hfr) DNA

Table 3. Recovery of Sl-sensitive fraction of recipient D N A during conjugation with Hfr Time of mating (min)

0 15 30 45 60 9o

S~-sensitive fraction of recipient D N A " (%) Method of cell lysis: Heating cells with SDS and digestion of pronase

Modification of the Piechowska-Fox method

Mated HfrxF-

U n m a t e d mixture b Mated Hfr+FHfrxF

6.8 7.6 16.0 18.6 16.2 9.8

3.6

23.5

4.2

11.3

Sl-sensitive fraction was determined as in Table 2 b D o n o r and recipient cells grown separately and mixed immediately prior to the cell lysis after 30, 90 min of incubation were taken as a control All experimental details are as described in the Materials and Methods

Table 4. Sl-sensitivity of recA recipient D N A after mating with Hfr

Time of mating (rain)

Sl-sensitive fraction a (%) U n m a t e d mixture b Hfr+F-

Mated Hfr x F -

30 45 90

7.0 0.9 4.8

17.7 11.7 1.6

a, b

The observed single-strand regions were initially assumed to be presynaptic recombination structures. To verify this hypothesis and F-AB2463 strain carrying a recA13 mutation was chosen. The recA mutant shows the lack of covalent association of donor and recipient DNA (Paul and Riley, 1974). The results of several experiments, similar to those previously described, indicate the appearance of nuclease S1sensitive DNA fraction in the recA recipient during conjugation, with maximum after 30 min, followed by return to the control level after 90 min (Table 4). The level of the single-strand regions in the recA mutant seems not to differ from this in wild-type strain (see Table 3). These findings were confirmed by data (not presented) on the buoyant density CsC1 gradient analysis of sheared recipient DNA (Biatkowska-Hobrzafiska, 1976). This suggests that the generation of singlestrand segments in the conjugational recipient DNA may be considered to be the initial stage of recombination.

D N A of E. coli K-12

see legend to Table 3

For substrate preparation the modification of the Piechowska-Fox method was used

The Nature of the Signals Initiating the Change of Recipient DNA Structure According to a number of authors the receptiveness of the female cells (Novotny, Knight and Brinton, 1968; Kunicki-Goldfinger, 1968) and initiation of the transfer of donor DNA (Jacob, Brenner and Cuzin, 1963 ; Kunicki-Goldfinger, 1968 ; Ou, 1975) are governed by some sort of signals in the early stages of conjugation. The change of the recipient DNA structure may be stimulated by such early signals or may depend on other succeeding stages of conjugation. It has thus been attempted to examine the influence of the preliminary stages of conjugation on the size of the single-strand recipient DNA fraction. Prelabeled F-AB1157 was mated for 10 rain with

H. Biatkowska-Hobrzaflska and W.J.H. Kunicki-Goldfinger: Single-Stranded Regions in F - D N A of E. coli K-12

non radioactive Hfr H64 in non radioactive medium. Conjugation was interrupted by the addition of high concentration of phage t2 (450 particles per Hfr cell) according to Novotny, Knight and Brinton (1968). After 5 min the sample was diluted five-fold in conjugation medium supplemented with 1 mg/ml streptomycin and shaken for a further 5 min at 37 °. Afterwards, post-incubation samples of the recipient cells were withdrawn, the lysates prepared, and then incubated with nuclease $1 in the same conditions as described in Materials and Methods. The described procedure resulted in complete elimination of donor cells and inhibition of transfer without affecting the number o f recipient cells (data not presented). It can be shown (Table 5) that 10 min conjugation followed by 20 min incubation gives final quantitative alterations of recipient DNA structure resembling those obtained after 30 min continuous transfer (see Table 3). This may imply that the initial stages of conjugation stimulate generation of single-stranded regions of recipient DNA and that the further accumulation of homologous donor DNA seems to have no effect. Similar results could also be obtained by the uptake of plasmid DNA only, by recipient cell. This possibility was tested with F+AB2463/A recA donor strain which was not capable of chromosome mobilization, hence only F factor DNA could be transferred. The number of Met "recombinants" after 90 min conjugation between F+2463/A and F- 108 was equal to the quantity of observed revertants in F-108 strain, whereas the number of sexductants reached 60% (data not presented). The absence of the transfer of chromosomal markers as well as the demonstrated by Vapnek and Rupp (1970) circularization of plasmid DNA after synthesis of the complementary strand hinders post-conjugational recombination. The appearance of a detectable, although relatively lower quantity of Sl-sensitive DNA in the recipient chromosome is hence interpreted as associated with the early phases of conjugation (Table 6). The mating given in Table 7 was carried out to check whether the presence of donor cells alone is sufficient for initiation of the recipient DNA alterations (Goldfarb et al., 1973). Strain F - AB 1157 labeled prior to conjugation was mated with non radioactive Hfr P721 tra mutant devoid of sex pili and not able to transfer donor material (Cooke, Meynell and Lawn, 1970). Further procedure were as described previously. The results indicate that when no mating pairs are formed the recipient DNA shows no single strand regions. In summary, we believe we have shown that the alterations in the secondary structure of recipient DNA are induced in the early phases of conjugation.

323

Table 5. Sl-sensitivity of recipient D N A after initiation of transfer by Hfr D N A

Time of mating (min)

Time of post-incubation of F after elimination (10min) of Hfr (min)

Sl-sensitive fraction (%)

0 0 10 l0 10 10 10

0 20 5 20 35 50 65

-0.4 5.8 9.9 20.7 15.4 10.4 8.1

Mixtures of mated and u n m a t e d control cells were lysed according to the modification of Piechowska-Fox method. For other details see the text. S~-sensitive fraction was determined as in Table 2

Table6. S~-sensitivity of recipient D N A during conjugation with F+recA donor strain

Time of mating (min)

Sl-sensitive fraction" (%) U n m a t e d mixture b F++ F

Mated F+x F -

30 45

0.9 1.0

10.0 13.8

For ~,b and further notes see Table 3

Table7. Sl-sensitivity of recipient D N A during conjugation with Hfr tra m u t a n t strain

Time of mating (min)

30 45

S~-sensitive fraction a (%) U n m a t e d mixture b Hfr+F-

Mated HfrxF

-0.8 3.2

1.8 " 2.4

For "' b and further notes see Table 3

It is not yet clear whether contact between partners' cells or initiation of transfer donor DNA (or combination of these phenomena) trigger the generation of single-stranded regions in recipient DNA. More detailed studies on these problems are currently under way.

Discussion

According to the model of post-conjugational genetic recombination in bacteria proposed by Kunicki-

324

H. Biatkowska-Hobrzafiskaand W.J.H. Kunicki-Goldfinger:Single-StrandedRegionsin F- DNA of E. coli K-12

Goldfinger (1968, 1971) and Wtodarczyk and Kunicki-Goldfinger (1970a) the uptake of donor singlestranded DNA by the recipient chromosome should trigger succeeding events involving breakage followed by copying. The indispensability of DNA synthesis for the formation of functional recombinant structures was shown by Piekarowicz, Wtodarczyk and Kunicki-Goldfinger (1968), Wtodarczyk and KunickiGoldfinger (1970b), Bednarska et al. (1972), Wtodarczyk et al. (1974). The model assumed DNA synthesis to be preceded by invasion of recipient doublestranded DNA by donor single-stranded DNA, transient triplex formation, followed by partial unwinding of the endogenous duplex and breakage of the recipient chromosome. A similar model for recombination was proposed by Boon and Zinder (1971). Progress in this field in recent years has been reviewed by several authors (e.g. Radding, 1973; Hotchkiss, 1974; Chipchase, 1976) but no experimental data contradicting our model have been published in this period. The most recent paper by Enea and Zinder (1976) shows that the breakage and copying model is not applicable for a special case of recombination in fl bacteriophage. It is not, however, obvious that postconjugational recombination in bacteria should be carried out by the same mechanism which is operative in phage recombination. To seek further empirical facts which could verify our model it appeared worthwhile to check whether single-strand fragments of the recipient DNA may be found during conjugation. The results presented here indicate that the formation of mating pairs followed by DNA transfer or at least initiation of DNA transfer causes the appearance of single-stranded segments in the post-conjugational recipient DNA. The amount of single-stranded fraction increases to 23.5 percent of the total recipient DNA 30 rain after the beginning of mating (Table 3). It then slowly decreases to reach the initial level after 90 rain. This has been confirmed by an independent method based on fractionation of sheared recipient DNA in Cs C1 buoyant density gradient (BiatkowskaHobrzafiska and Kunicki-Goldfinger, 1977). The fact that single-strand material was detectable in that case only with sheared recipient DNA shows explicitly the local character of observed alterations in the secondary structure of the recipient DNA. We believe that observed single-strand regions in the post-conjugational female cell may reflect initial recombination events because: (i) single-strand regions were well recoverable from recipient mated with Hfr but poorly recoverable from unmated recipient cells or mated with Hfr tra mutant (Tables 3, 7), (ii) the quantity of single-stranded regions in the recA m u t a n t - w h i c h shows lack of the conversion of joint molecules to covalently linked recombinant molecules (Paul and

Riley, 1974) have been similar to this in wild type (Table 4). Our findings are consistent with LeClerk and Setlow's (1975) observations on formation of singlestranded regions in DNA of competent H. influenzae cells and its importance for early steps in recombination. Recently, Gillen and Clark (1974) proposed a possible presynaptic role for Exo V basing on the studies of McKay and Linn (1974) which suggest the reversible opening up of single-stranded regions of native DNAs for base pairing interaction. This prompted us to test whether a recB reeC mutant is active in the generation 0f $1 nuclease sensitive regions (study in progress). Initiation events of genetic recombination have been successfully demonstrated in vitro by Holloman et al. (1975), who observed the uptake of homologous single-stranded fragments by superhelical DNA (RFI of bacteriophage~X174). Such superhelical DNA has been observed in the folded chromosome of E. coli which is arranged in numerous (about 50) supertwisted loops (Worcel and Burgi, 1972; Pettijohn and Hecht, 1973). Superhelical DNA structures are characterized by a positive free energy (Bauer and Vinograd, 1971 ; Beerman and Lebowitz, 1973; Wang, 1974) which facilitates any process leading to partial denaturation of the duplex. According to the proposed model, the invasion of the recipient superhelical DNA by a donor singlestranded DNA triggers the succeeding stage of the recombinational process i.e. partial unwinding of the recipient DNA. This prediction of the model has been verified by the results presented in this paper. Generation of single-stranded fragments seems to be a general characteristic of initiation of recombinational events. Such an assumption is supported by Observations of alteration in DNA during early stages of meiosis in yeast (Jacobson et al., 1975; Simchen and Friedman, 1975). Ensuing copying of the fourth strand results in the formation of a four-stranded intermediate. Such an intermediate has recently been found in rec mutants of Bacillus subtilis (KShnlein and Hutchinson, 1976). Preliminary results received in our laboratory suggest that in recA and probably in recB recC mutant of E. coli K-12, the "replicative" stage following unwinding of the recipient double-stranded DNA is for some reasons inhibited, and donor single strand first accumulates and is later degraded. These results will be reported in a subsequent paper. Acknowledgments. The authors are indebted to Dr. H. Okazaki

Sankyo Co., Ltd., Tokyo,Japan, for generous supply of Sankyo Takadiastase. This research was carried out within the Project 09.3.1, the Polish Academyof Sciences.

H. Biatkowska-Hobrzafiska and W.J.H. Kunicki-Goldfinger: Single-Stranded Regions in F

References Achtman, M., Willetts, N., Clark, A.J.: Beginning a genetic analysis of conjugational transfer determined by the F factor in E. coil by isolation and characterization of transfer-deficient mutants. J. Bact. 106, 529-538 (1971) Ando, T.: A nuclease specific for heat-denatured DNA isolated from a product of Aspergillus oryzae. Biochim. biophys. Acta (Amst.) 114, 158-168 (1966) Ayad, S.R.: Techniques of nucleic acid fractionation. First ed. London-New York-Sydney-Toronto: Wiley-Interscience 1972 Bauer, W., Vinograd, J. : The use of intercalative dyes in the study of closed circular DNA. In: Prog. Mol. Subcell. Biol. (Hahn, F.E., ed.), Vol. 2, pp. t81-215. New York: Springer 1971 Barth, P.T., Grinter, N.J. : Assay of deoxyribonucleic acid homology using a single-strand-specific nuclease at 75 ° C. J. Bact. 121, 434M4I (1975) Beerman, T.A., Lebowitz, J.: Further analysis of the altered secondary structure of superhelical DNA. Sensitivity to methylmercuric hydroxide a chemical probe for unpaired bases. J. molec. Biol. 79, 451M70 (1973) Bednarska, M., Wtodarczyk, M , Kraczkiewicz, A., Knnicki-Goldfinger, Wt.: The role of DNA synthesis in the formation of post-conjugational recombinant structure in E. coli K-12. Bull. Acad. pol. Sci., Ser. Sci. Biol. 20, 627~634 (1972) Biatkowska-Hobrzafiska, H. : Structure of recipient DNA after uptake of donor DNA during conjugation in E. coli K-12. Ph. D. Thesisl University of Warsaw (1976) Biatkowska-Hobrzafiska, H., Kunicki-Goldfinger, Wt.: Mechanisrn of conjugation and recombination in bacteria XVII. Further evidence for single-stranded regions in recipient DNA during conjugation in Escherichia coil K-12. Acta microbiol, pol. 26 (1977) (in press) Boon, T., Zinder, N.D.: Genotypes produced by individual recombination events involving bacteriophage fl. J. molec. Biol. 58, 133-151 (1971) Carrier, W.L, Setlow, R.B. : Paper strip method for assaying gradient fractions containing radioactive macromolecules. Anal. Biocl-/em. 43, 427432 (1971) Chipchase, M. : Recombination by strand assimilation and strand crossover. J. theor. Biol. 57, 249-279 (1976) Cohen, A., Fisher, W.D., Curtiss, R., Adler, H.I.: The properties of DNA transferred to minicells during conjugation. Cold Spr. Harb. Syrup. quant. Biol. 33, 635-641 (1968) Cooke, M., Meynell, E., Lawn, A.M. : Mutant Hfr strains defective in transfer: restoration by F-like and I-like de-repressed R factors. Genet. Res. 16, 101-112 (1970) Crosa, J.H., Brenner, D.J., Falkow, S.: Use of a single-strand specific nuclease for analysis of bacterial and plasmid deoxyribonucleic acid homo- and heterodnplexes. J. Bact. 115, 904-911 (1973) Curtiss, R. : Bacterial conjugation. Ann. Rev. Microbiol. 23, 69-136 (1969) Davidoff-Abelson, R., Dubnau, D. : Conditions affecting the isolation from transformed cells of Bacillus subtilis of high-molecular-weight single-stranded DNA of donor origin. J. Bact. 116, 146-153 (1973a) Davidoff-Abelson, R., Dubnau, D.: Kinetic analysis of the products of donor deoxyribonucleate in transformed cells of Bacillus subtilis. J. Bact. 116, 154 162 (1973b) Davis, B.D.: Mutants of E. coli requiring methionine or vitamin B12. J. Bact. 60, 17 28 (t950) Dishe, Z.: Color reactions of nucleic acid components. In: The nucleic acids (Chargaff, E., Davidson, J.N., ed.), Vol. 1, pp. 285-305. New York: Academic Press 1955 Enea, V., Zinder, N.D.: Heteroduplex DNA: A recombinational

DNA of E. coli K-12

325

intermediate in bacteriophage fl. J. molec. Biol. 101, 25-38 (1976) Germond, J.E., Vogt, V.M., Hirt, B.: Characterization of the single-strand-specific nuclease $1 activity on double-stranded supercoiled polyoma DNA. Europ. J. Biochem. 43, 591~500 (1974) Gillen, J.R., Clark, AJ. : The recE pathway of bacterial recombination. In: Mechanisms in recombination (Grell, R.F., ed.), pp. t23 136. New York: Plenum 1974 Godson, G.N.: Action of the single-stranded DNA specific nuclease $1 oll double-stranded DNA. Biochim. biophys. Acta (Amst.) 308, 59~i7 (1973) Goldfarb, D.M., Goldberg, GJ., Chernin, LS., Gukova, L.A., Avdienko, J.D., Kuznetsova, B.N., Kushner, J.Ch.: A protein produced by male strains of E. coli K12 which increases the yield of recombinants in conjugation: Its nature and mode of action. Molec. gen. Genet. 120, 211-226 (1973) Harris, WJ., Barr, G.C.: Some properties of DNA in competent Bacillus subtilis. J. molec. Biol. 39, 245 255 (1969) Harris, W.J., Barr, G.C. : Structural features of DNA in competent Bacillus subtilis. Molec. gen. Genet. 113, 316-330 (1971) Hayes, W. : The kinetics of the mating process in Escherichia coll. J. gem Microbiol. 16, 97-119 (1957) Holloman, W.K., Wiegand, R., Hoessli, C., Radding. Ch.M. : Uptake of homologous single-stranded fragments by superhelical DNA: A possible mechanism for initiation of genetic recombination. Proc. nat. Acad. Sci. (Wash.) 72, 2394-2398 (1975) Hotchkiss, R.D.: Models of genetic recombinatiom Ann. Rev. Microbiol. 28, 445~468 (1974) Jacob, F., Brenner, S., Cuzin, F.: On the regulation of DNA replication in bacteria. Cold Spr. Harb. Symp_ quant. Biol. 28, 329-348 (1963) Jacobsen, G.K., Pifion, R., Esposito, R.E., Esposito0 M.S. : Singlestrand scissions of chromosomal DNA during commitment to recombination at meiosis. Proc. nat. Acad. Sci. (Wash.) 72, 1887-1891 (1975) John, T.St., Johnson, J.D, Bonner, J.: Degradation of duplex DNA by S t nuclease from Aspergiltus. Biochem. biophys. Res. Commun. 57, 240-247 (1974) K6hnlein, W., Hutchinson, F. : A four-stranded DNA from B. subtills which may be an intermediate in genetic recombination. Molec. gen. Genet. 144, 323-331 (1976) Kunicki-Goldfinger, W.J.H. : Mechanism of bacterial conjugation and recombination; a tentative model. Acta microbiol, pol. 17, 147-180 (1968) Kunicki-Goldfinger, WJ.H.: Single-strandedness of donor DNA and triplex formation as a prerequisite of post-conjugational recombination. In: Recent Adv. Microbiol. (P~rez-Miravete, A., Pel/tez, D., ed.), pp. 291-303. Mexico: Int. Congr. Microbiol., Assoc. Mexicana Microbiol. 1971 LeClerk, J.E., Setlow, J.K. : Single-strand regions in the deoxyribonucleic acid of competent Haemophilus influenzae. J. Bact. 122, 1091-1102 (1975) Lederberg, E.M., Lederberg, J.: Genetic studies of lysogenicity in interstrain crosses in E. coll. Genetics 38, 51~54 (1953) MacKay, V, Linn, S.: The mechanism of degradation of duplex deoxyribonucleic acid by the recBC enzyme of Escherichia coll. J. biol. Chem. 249, 4286~,294 (1974) Miller, S.J., Wetmur, J.G.: Physical properties of endonuclease S~ digestion products of DNA renaturation intermediates. Biopolymers 14, 309-317 (1975) Molineux, I J_, Friedman, S., Gefter, M.L.: Purification and properties of the Escherichia coli DNA-unwinding protein. J. biol. Chem. 249, 6090-6098 (1974) Molineux, I.J., Gefter, M.L.: Properties of the Escherichia coli DNA-binding (unwinding) protein interaction with nucleolytic enzymes and DNA. J. molec. Biol. 98, 811-825 (I975)

326

H. Biatkowska-Hobrzaflska and W.J.H. Kunicki-Goldfinger: Single-Stranded Regions in F - DNA of E. coli K-12

Novotny, Ch., Knight, W.S., Brinton, Ch.C., Jr.: Inhibition of bacterial conjugation by RNA and DNA male-specific bacteriophages. J. Bact. 95, 314-326 (1968) Ou, J.T. : Mating signal and DNA penetration deficiency in conjugation between male Escherichia eoli and minicells. Proc. nat. Acad. Sci. (Wash.) 72, 3721 3725 (1975) Paul, A.V., Riley, M. : Joint molecule formation following conjugation in wild type and mutant Eseherichia coli recipients. J. molec. Biol. 82, 35-56 (1974) Pettijohn, D.E., Hecht, R.M. : RNA molecules bound to the folded bacterial genome stabilize DNA folds and segregate domains of spreading. Cold Spr. Harb. Syrup. quant. Biol. 38, 31~41 (1973) Piechowska, M., Fox, M.S. : Fate of transforming deoxyribonucleate in Bacillus subtilis. J. Bact. 108, 680~689 (1971) Piekarowiez, A., Kunicki-Goldfinger, W.J.H. : Mechanism of conjugation and recombination in bacteria. IV. Single strandedness of donor DNA in mating bacteria. Acta microbiol, pol. 17, 135-146 (1968) Piekarowicz, A., Wtodarczyk, M., Kunicki-Goldfinger, W.J.H.: Mechanism of conjugation and recombination in bacteria. I. The role of DNA synthesis in Hfr and F - cells during mating in E. coli K-12. Molec. gen. Genet. 101, 131-139 (1968) Radding, Ch.M. : Molecular mechanisms in genetic recombination. Ann. Rev. Genet. 7, 87-111 (1973) Rupp, W.D., Ihler, G. : Strand selection during bacterial mating. Cold Spr. Harb. Symp. quant. Biol. 33, 647-650 (1968) Sedgwick, B., Setlow, J.K. : Single-stranded regions in transforming DNA after uptake by competent Haemophilus influenzae. J. Bact. 125, 588-594 (1976) Sigal, N., Delius, H., Kornberg, T., Gefter, M.L., Alberts, B.: A DNA-unwinding protein isolated from Eseheriehia coli: its interaction with DNA and with polymerases. Proc. nat. Acad. Sci. (Wash.) 69, 3537-3541 (1972) Simchen, G., Friedmann, A. : Structure of DNA molecules in yeast meiosis. Nature (Lond.) 257, 64-66 (1975) Sutton, W.D.: A crude nuclease preparation suitable for use in DNA reassociation experiments. Biochim. biophys. Acta (Amst.) 240, 522 531 (1971)

Taylor, A.L., Trotter, C.D. : Linkage map of Escheriehia coli strain K-12. Bact. Rev. 36, 504 524 (1972) Vapnek, D., Rupp, W.D. : Asymmetric segregation of the complementary sex-factor DNA strands during conjugation in Eseherichia coli. J. molec. Biol. 53, 287-303 (1970) Vielmetter, W., Bonhoeffer, F., Schtitte, A. : Genetic evidence for transfer of a single strand during bacterial conjugation. J. molec. Biol. 37, 81-86 (1968) Vogt, V.M. : Purification and further properties of single-strandspecific nuclease from Aspergillus oryzae. Europ. J. Biochem. 33, 192-200 (1973) Wang, J.C.: Interactions between twisted DNAs and enzymes: The effects of superhelical turns. J. molec. Biol. 8"7, 797-816 (1974) Willetts, N.S., Clark, A.J. : Characteristics of some multiply recombination-deficient strains of Eseherichia coli. J. Bact. 100, 231-239 (1969) Wtodarczyk, M., Kunicki-Goldfinger, W.J.H. : Mechanism of conjugation and recombination in bacteria. XII. Susceptibility of donor DNA to 3ZP-decay during conjugation. The new model of recombination involving single-strandedness of donor DNA and triplex formation. Acta microbiol, pol. Ser. A 2(19), 57-74 (1970a) Wtodarczyk, M., Bednarska, M., Kraczkiewicz, A., Zi6tkowska, J., Kunicki-Goldfinger, W.J.H. : Mechanism of conjugation and recombination in bacteria. XIV. Timing of the recombination and of the succeeding replication of recombinant chromosome. Acta microbiol, pol. Ser. A 6(23), 3748 (1974) Wtodarczyk, M., Kunicki-Goldfinger, W.J.H. : Mechanism of conjugation and recombination in bacteria. IX. The role of DNA synthesis in post-conjugational genetic recombination. Molec. gen. Genet. 106, 263-273 (1970b) Worcel, A., Burgi, E. : On the structure of the folded chromosome of Eseheriehia coli. J. molec. Biol. 71, 127-147 (1972)

Communicated by W. Gajewski Received December 13, 1976