Gene, 3 (1978) 97--112

97

© Elsevier/North-Holland Eiomedical Press, Amsterdam -- Printed in The Netherlands

CLONING OF RESTRICTION AND MODIFICATION GENES IN E. coli: THE HbaH SYSTEM FROM Haemopbilus baemolyticus (Recombinant DNA; restriction endonucleases; Pstl; DNA methylase; plasmids; gel electrophoresis; cleavage map)

MICHAEL B. MANN, R. NAGARAJA RAO and I-LO. SMITH

Department of Microbiology, The Johns Hopkins University School of Medicine, Baltimore, Maryland 21205 (U~q.4.) (Received October 20th, 1977) (Accepted January 5th, 1978)

SUMMARY

The genes for a Class II restriction-modification system (HhaII) from Haemophilus haemolyticus have been cloned in Escherichia coll. The vector used for cloning was plasmid pBR322 which confers resistance to tetracycline and ampicillin and contains a single endonuclease R. PstI site, (5')C-T-G-C-A!G (3'), in the ampicillin gene. The procedure developed by Bolivar et al. (1977) was used to form DNA recombinants. H. haemolyticus DNA was cleaved with PstI endonuclcase and poly(dC) extensions were added to the 3'~)H termini using terminal deoxynucleotidyl transferase. Circular pBR322 DNA was cleaved to linear molecules with PstI endonuclease and poly(dG) extensions were added to the 3'4)H termini, thus regenerating the PstI cleavage site sequence. Recombinant molecules, formed by annealing the two DNAs, were used to transfect a restriction and modification-deficient strain of E. coli (HB101 r-m-recA). Tetracycline-resistant clones were tested for acquisition of restriction phenotype (as measured by growth on plates seeded with phage ~cI. 0). A single phage-resistant clone was found. The recombinant plasmid, pDU0, isolated from this clone, had acquired 3 kilobases of additional DNA which could be excised with PstI endonuclease. In addition to the restriction function, cells carrying the plasmid expressed the HhaII modification function. Both activities have been partially purified by single-stranded D NA-agarose chromatography. The cloned HhaII restriction activity yields cleavage patterns identical to HinfI. A restriction map of the cloned DNA segment is presented.

INTRODUCTION

Although restriction enzymes have been extensively exploited in DNA re-

98

search, few have been adequately purified or obtained in sufficient quantity for biochemical and physical studies. This is due, in part, to the fact that many of these activities are found in b a ~ that require expensive media for growth and produce the enzymes in low yield. In several recent cases, yields of particular enzymes or proteins have been dramatically increased by gene cloning. To give two examples, 500-fold overproduction of DNA ligase has been obtained after induction of a hybrid laml~m lysogen construc*~l in ~ t r o (Panasenko et al., 1977), ~nd X repressor overproduction has been obtained by cloning the repressor gene in a multicopy E. eoU plasmid (Backman et al., 1976). These successes have encouraged us to undertake the cloning of a repo resentative class H restriction-modification (R-M) system. For this purpose, we have selected Haemophilus haemolyticus. This choice was motivated by our need for quantities of homogeneous HhaI endonuclease to use in studies of the site recognition process. We were also aware that a second class H endonuclease, HhaII (Mann, unpublished), exists in this strain. It is an isoschizomer of Hinfl which recognizes the sequence (5') G-A-N-T-C (Roberts, 1976), and is useful as a multicut enzyme for DNA sequencing work. A "shotgun" approach was used for the cloning. Total chromosomal DNA was cleaved with PstI endonuclease and inserted into the K coU plasmid, pBR322 (Bolivar et al., 1977), by annealing via a GC-extension procedure (Rougeon et al., 1975). After transfection into an r - m - r e v A EK1 host (HB101), tetracycline-resistant recombinant clones were tested for the acquisition of a restriction phenotype using phage ).. A single clone containing the HhaII restriction and modification genes was found and preliminary characterization is reported here. MATERIALS AND METHODS

Bacterial, plasmid, and phage strains. Haemophilus haemoly ticus ( ATCC 10014) is from the American Type Culture Collection, Rockville, Md., USA. The

E. coil K12 strain HB101 r - m - r e c A (from H.W. Boyer) was used as host for plasmid DNA transfections. E. coli K12 KL16-99r+m+recA was obtained from Roger McMacken. pBR322 (from H.W. Boyer) is a small (2.6.106 daltons) multicopy plasmid derived from ColE1 by Bolivar et al. (1977), for cloning procedures. It carries genes fo~"ampicillin and tetracycline resistance and contains single sites for HindIH, 8alI, BamHI, EcoRI, and PstI endonuclease cleavage, Insertions into the PstI site affect ampicillin resistance. Phage strains ),cI, ;~cI857 Sam7, ~80, h424i ~, P1, BF23, and T5 are from the laboratory collection. All E. coil strains were grown in L broth (1% tryptone, 0.5% yeast extract, 1% NaCI, pH 7.2). I-I. haemolyticus was grown in brain heart infusion (Difco) supplemented with NAD 2 #g/ml and heroin 10 pg/ml. Enzymes and reagents. Restriction endonucleases PstI, BamHI, HhaI, HaeIII, EcoRI, and Hin~III were from Biolabs, Beverly, Mass. Hinfl endonuclease was a gift of Dr. N. Musczycka. The bacterial source and site specificity of these en-

99

zymes are given by Roberts (1976). Terminal deoxynuc!eotidyl transferase was purchased from Product Research Associates, Pacific Palisades, Calif. 90272. [SH-methyl]S-adenosyl methionine (12 Ci/mmole) was from New England Nuclear. Single-stranded DNA agarose was prepared according to Amdt~ovin et al. (1975). Preparation of DNA. H. haemolyticus DNA was prepared by the procedure of Marmur (1961). Plasmid DNA was isolated by the procedure of Hirt (1967) and f ~ t h e r purified by ethidium bromide-CsC1 density gradient centrifugation (Clewell and Helinski, 1969). ~X174 RF1 DNA was isolated from E. coli H502 cells (Johnson and Sinsheimer, 1974) infected with the am3 mutant (Hutchison and Sinsheimer, 1966) according to the procedure of Johnson and Sinsheimer (1974). kcI857 Sam7 DNA was phenol-extracted from CsC1 purified phage (Thomas and Abelson, 1966). Construction of hybrid plasmid DNA. pBR322 DNA (25 pg)was cleaved to linear molecules with PstI endonuclease in a reaction mixture (0.25 ml) containing 50 mM NaCI, 7 mM MgCI2, 7 mM Tris--Cl (pH 7.6), 7 mM 2-mercaptoethanol, 100/Jg/ml autoclaved gelatin, and 50 units of enzyme. Incubation was for 1 h at 37°C. The DNA was then phenol-extracted, ethanol-precipitated, and redissolved in TE buffer (10 mM Tris--Cl, pH 7.6, 1 mM Na2EDTA) at 500 ~g/ml (Maxam and Gilbert, 1977). H. haemolyticus DNA (25/~g) was partially digested with 24 units of enzyme as above and then extracted and dissolved in TE buffer at 500 pg/ml. Homopolymer (dG) extensions were added to the PstI-generated linear pBR322 DNA molecules in a reaction mixture (60/~1) containing 100 mM Na cacodylate buffer (pH 7.8), 4 mM MgCI2, 0.2 mM dGTP, 7.5 ~ of DNA and 250 units of terminal transferase. Incubation was for I h at 37°C. The poly(dG) extensions were estimated to average about 30 nucleotides in length in preliminary experiments employing [3H]dGTP. Poly(dC) extensions were added to the PstI-generated H. haemolyticus DNA fragments in similar fashion and averaged 25 nucleotides in length. Annealing of the extended DNAs was can'ied out in a 2.5 ml volume containing 50 ~1 of each DNA preparation, 0.1 M NaCI, 10 mM Na2EDTA (pH 7.25). The total DNA concentration was 5 #g/ml. The mixture was incubated at 70°C for 10 min, cooled to 43°C over approx. 60 min and then held at 43°C overnight. The annealed DNA was concentrated by adding sodium acetate to 0.3 M and precipitating with 2.5 volumes of 95% ethanol. The mixture was chilled in a dry ice-ethanol bath for 10 min, then centrifuged at 10 000 × g for 10 min. The pellet was dissolved in 125/A of TE buffer at approx. 100/Jg/ml. Transfection and clone selection. Transfection of cells with annealed DNA was carried out using CaCl2-treated HB101 cells according to the procedure of Wensink et al. (1974). Annealed DNA (125/~1 containing 12.5 ~g of DNA) was diluted with 1.9 ml of 10 mM Tris-Cl (pH 7), 10 mM CaCI2,10 mM MgC12 and added to 4 ml of cells (approx. 5-109 cells/ml) in 50 mM CaCI2. The suspension was incubated at 0°C for 25 min, 37°C for 2 rain, and 23°C for 10 min. 20 m! of L broth was added, followed by incubation at 37°C for 30 rain with

100

gentle shaking. The 26 ml cell suspension was then mixed with 50 ml of soft agar (L broth containing 0.7% agar and 25 #g/ml of tetracycline) at 45°C. Aliquots (3.8 mi) were quickly pipetted onto L agar plates containing 25 #g/ml tetracycline. Plates were incubated 2 days at 37°C. (Colonies were too small to pick at 24 h.) Tetracycline-resistant colonies were picked into Microtiter plate wells (Weiss and Milcarek, 1974) filled with 50/d of L broth containing 25 #g/ml tetracycline. Plates were incubated at 37°C overnight. The array of microwell cuitvres was replicated with a multipronged device (Weiss and Mflcarek, 1974) onto 2 sets of agar plates, the f~st (without antibiotic) serving as a storage plate, and the second spread with 10 s XcI-0 phage. The plates were incubated overnight at 37°C and screened for surviving cell patches. Survivors were subcultured in 3 m! of L broth containing tetracycline. Aliquots (0.1 ml) of each broth culture were seeded in soft agar on L agar plates and spot-tested with various dilutions (100 , 10 -2 , 10 -4 , 10 -6) of high titer phage stocks (approx. 101° pfu/ml). For spotting, the entire array of phage dilutions was placed in a Microtiter plate and replicated onto the seeded soft agar layers with the multipronged device. Plates were incubated overnight at 37°C and then scored for confluent lysis or countable numbers of plaques within the inoculation spots, thus giving the approximate plating efficiency for each of the phages. These experiments were conducted under P3 physical containment and EK1 biological containment in compliance with the NIH Guidelines for recombinant DNA research (Federal Register, July 7, 1976). Restriction endonu¢lease digestions. Restriction endonuclease digestions were carried out in 20 #1 volumes containing 0°25 to 0.5 #g of DNA, approx. I unit of enzyme, and additional components, depending on the enzyme used, as follows: HindIII, BamHI, and PstI, 50 mM NaCI, 7 mM Tris--Cl (pH 7.6), 7 mM MgCI2,7 mM 2-mercaptoethanol; EcoRI, 100 mM Tris-Cl (pH 7.6), 50 mM NaCI, 5 mM MgCI2; HhaI, HhaII, and HinfI, 40 mM glycine--NaOH (pH 9.0), 7 rnM MgC12, 7 mM 2-mercaptoethanol; HaeIH, 7 mM NaCI, 7 mM Tris--C1 (pH 7.6), 7 mM MgCI2. Incubations were for 60 rain at 37°C. Reactions were terminated by addition of 2.5/~! of dye mixture (50% glycerol, 0.25% bromphenol blue, 0.25% xylene cyanol FF, 0.5% sodium dodecyl sulfate). Gel electrophoresis of DNA. Samples (5--20 ~1) of restriction enzyme digestion mixtures were loaded into I cm slots on 0.8 to 1.4% agarose slab gels (0.2 x 18 x 40 cm) and electrophoresed at 5 V/cm for specified times in a vertical apparatus, Electrophoresis buffer contained 15 g glycine, 15 ml I N NaC~H per liter. For separations on polyacrylamide, the lower 7 cm of the slab gelwas cast with 10% polyacrylamide and the remainder With 4% polyacrylamide using standard polymerizing recipes (DeWachter and Fiers, 1971). The buffer was 50 mM Tris-borate (pH 8~3), 1 mM EDTA. Electrophoresis was at 10 V/cm for 3 h with cooling by fan, Ethidium bromide staining and photography on Kodak Plus X film were as described by S h e e t al., 1973.

101

Enzyme purification. KJ12 cells (HB101 carrying the HhaII recombinant plasmid pDI10) were grown into stationary phase in L broth (3 liters) at 37°C with aeration. Cells (about 8 g) were harvested by centrifugation and resuspended in 12 ml of buffer (0.3 M NaC1, 10 mM Tris--Cl (pH 7.6), 5 mM 2-mercaptoethanol). Cells were disrupted by sonication at full intensity for 2 min. The suspension was centrifuged at 100 000 x g for 2 h at 4°C and the supernatant (15 ml) was recovered and adjusted to 5% glycerol. The bulk of contaminating nucleic acid was removed by chromatography on DEAESephadex. Supernatant was loaded onto a DEAE-Sephadex A-25-120 column (2.5 x 19 cm) equilibrated with buffer containing 0.3 M NaCI, 10 mM TrJs--Cl (pH 8.0), 1 mM Na2EDTA, 5 mM 2-mercaptoethanol and 5% glycerol. The !column was eluted at I ml/min with this buffer and fractions (6 ~nl) were collected. Fractions containing protein were pooled and diluted 3-fold with buffer lacking NaCI. To remove residual lower molecular weight nucleic acids, this material was loaded onto a DEAE cellulose (Whatman DE52) column (1 x 11 cm) and eluted with buffer containing 0.1 M NaCI. Fractions containing protein were again pooled and precipitated by the addition of solid ammonium sulfate to 70% saturation. After stirring at 4°C for 20 rain, the precipitate was collected by centrifugation at 14 000 x g for 15 rain. The veUet was dissolved in 5 ml of 10 mM sodium phosphate (pH 7.5), 1 mM Na2EDTA, 5 mM 2-mercaptoethanol, 5% glycerol. This was applied to a Sephadex G-50 column (2.5 x 30 cm) and eluted with the phospl'~te buffer. Protein-containing fractions were pooled (24 ml). A 12-ml sample was then chromatographed on a single-stranded DNA agarose column (2.5 x 20 cm) equilibrated with the phosphate buffer. Proteins were eluted with a 0 to 0.6 M NaCI gradient (500 ml) at 0.5 rnl/min. Fractions were assayed for DNA methylase and restriction endonuclease activities. Methylase assays were carried out in reaction mixtures (20/~1) containing 25 ~g sonicated DNA from HB101 cells carrying pBR322, 4 ~M [methyl-3H]S adenosylmethionine, 50 mM Tris--Cl (pH 7.5), 20 mM 2-mercaptoethanol, 10 mM EDTA, and 5/A of enzyme fraction. Incubation was at 37°C for 1.5 h. A 10-/A sample was then chromatographed on a I cm × 5 cm polyethyleneimine thin-layer strip with 1 M HCI and the lower half of the strip was counted to determine acid-precipitable radioactivity remaining at the origin (Kelly and Smith, 1970). Endonuclease assays were carried out in reaction mixtures (20/~1) containing 40 mM glycine--NaOH (pH 9), 7 mM MgCI2,7 mM 2-mercaptoethanol, 0.25/Jg )~cI857 Sam7 DNA, and 5 td of enzyme fraction for I h at 37°C. Reaction mixtures were then mixed with 2.5/A of dye mixture and analysed by 1.4% agarose gel electrophoresis. RESULTS

Construction of H. haemolyticus-pBR322 hybrid plasmid DNA. A modification of the "shotgun" approach of Clarke and Carbon et al. (1976) was used in order to obtain recombinants representing the entire genome. Fragments of

!02

322 DNA

H. hemolvticus DNA

PstI

F~.X

TdT + dGTP

TdT + dCTP

~-GG---6~r[

~rGc~---GG-~' 3;CC---CCAC~r~:

oc.o°

~CACC.-.CC-~

•

~./C--.CCACeTC CTGCAGG"-GG G G /C..-CCACGTC S-C"

Tronsfection end Introcellulor Repoir Elfish

,,~ . .

OSTSCae

pBR322 DNA

CTGCAGG...GGTGCAG ~/

.SACSTC.C-.-CC~CSTC Psi l~te

Fig. 1. Diagram of steps in the formation of H. haemo|yticus-pBR322 hybrid DNA. TdT is terminal deoxynucleotidyl transferase.

H.haemolyticus DNA suitable for cloning were obtained by partial digestion with PstI endonuclease. This method was chosen instead of mechanical shearing since PstI cleaves the sequence (5')C-T-G-C-A !G(3'~ to produce single stranded extensions, 4 residues in length.., having 3'-OH termini. Such termini are ideal substrates for homopolymer extension using terminal deoxynucleotidyl transferase. A partial, rather than a complete digest was used in the event the genes of interest should contain PstI cleavage sites. The chromosomal DNA fragments were then extended with poly(dC) and annealed to PstI-generated linear pBR322 DNA extended with poly(dG). The entire scheme is illustrated in Fig. 1. The choice of GC~xtension, rather than AT-extension, allowed for

103

the regeneration of PstI cleavage sites at the hybrid junctions (Bolivar et al., 1977). Selection o f a recombinant clone with restriction properties. Transfection of HB101 cells with recombinant DNA molecules was performed as described i n METHODS. The effectiveness of the cloning procedure was evaluated by testing the transfection efficiency of DNA samples taken at various steps in the procedure. Retention of tetracycline resistance was scored. Results are given in Table I. Conversion of circular pBR322 DNA to linear molecules by PstI digestion reduced transfection efficiency 650-fold. After extending the linear molecules with dG, a further 17-fold reduction was observed. However, following the annealing step, the tetracycline-resistant colony count rose 12.5-fold. Thus during the selection procedure, one could anticipate that more than 90% of the tetracycline-resistant colonies would be recombinants. A total of 12.5 ~g of recombinant DNA was used for transfection and yielded approx. 1700 tetracycline-resistant colonies. Of these, 1400 were picked and cultured in depression plates in preparation for selection of clones exhibiting a restriction-modification phenotype. Several types of selection based on either restriction or modification were considered. We adopted here a convenient one based on restriction. HB101 cells have an r - m - genotype and are readily killed by a variety of phages, but introduction of a new restrictionmodification system through transfection with a hybrid plasmid should lead to acquisition of phage res'Lstance. On this assumption, clones were initially screened for their ability to grow in the presence of a clear plaque mutant of phage ~. The micro-well cultures were replica plated onto agar plates spread with 10 s k cI. 0 phage. The majority gave rise to fiat, translucent, patches of dead cells. In 50 patches, there was some degree of growth, varying from a few individual colonies to confluent growth. Each of the 50 was subcultured in broth for use as plating bacteria in a second round of testing. Five different kinds of phage (p80, h424, P1, BF23, and TS), were diluted serially over a 10~-fold range and spot tested on lawns of each of the 50 clones. Only one of the clones, KJ10, caused a dramatic reduction in plating efficiency with all the phages. This clone was tetracycline-resistant and ampicillin-sensitive. TABLE I TRANSFECTION EFFICIENCY OF VECTOR AND HYBRID PLASMID DNA DNA

Tetracycline-resistant tra.nsfectants per ug of pBR322 DNA

2.24" 105 pBR322 circular molecules 350 pBR322 PstLcleaved linear molecules 20 pBR322 linear molecules with polydG extensions pBR322 dG-extended linear molecules annealed with 250 dC-extended H. haemolyticus DNA

104

Characterization o f the recombinant plasmid. Plasmid DNA was isolated from KJ10 cells and analysed by agarose gel electrophoresis (Fig. 2). It migrated more slowly than the parental plasmid, pBR322, as shown in gel slots b--d. The new plasmid was named pDI10. The pBR322 and pDI10 DNA preparations both contained significant amounts of nicked circular (Form II) and linear (Form III) DNA in addition to covalent circular (Form I) DNA. Some plasmid dimers were present in the pBR322 DNA as well. PstI digestion converted the various circular forms of pBR322 DNA into a single Form III band of about 4.0 kilobases (kb) (slot e). Similar cleavage of pDI10 DNA yielded two fragments, one migrating at the position of pBR322 Form III DNA, and the other migrating more rapidly at a position corresponding app r o x i m a l l y to 3 kb (slot f). Slot g shows a mixture of the cut DNAs. The

Fig. 2. Comparison of pBR322 and pDI10 DNAs by agarose gel electrophoresis. Restriction enzyme digests were performed as described in M ~ H O D S . Electrophoresis was on a 0.8% agarose slab gel run at 10 V/cm for 75 min. (a) HindHI digested ~ DNA. Fragment lengths are in kilobases (Murray and Murray, 1975). (b) pBR322 DNA. (c) pDI10 DNA. (d) mixture of (b) and (c). (e) PstI digested pBR322. (f) Pstl digested PDI10. (g) mixture of (e) and (f). (h) HindIH digested pBR322. (i) HindIII digested PDI10. (j) mixture of (i) and (j). In tract (b), the 3 lower bands are (from bottom to top) Form's I, III and II of monomer plasmid DNA while the 3 upper bands are Form's I, IH, II of dimer plasmid DNA.

105

a

b

c

d

e

Fig. 3. Susceptibility of pDI10 DNA to cleavage by HhaI and HinfI endonucleases. Digestions were performed as described in METHODS. Products were analysed by electrophoresis on a 4% polyacrylamide gel. (a) HhaI digested pDI10 DNA. Arrows indicate new fragments not present in pBR322 DNA. (b) Hhal digested pBR322 DNA. The a r r o w indicates the fragment which becomes altered as a result of insertion in the PstI site. (c) HinfI digested pDI10 DNA. (d) HinfI digested pBR322 DNA. (e) HinfI digested mixture of pDI10 and pBR322 DNAs.

cleavage pattern is consistent with the insertion of a 3 kb H. haemolyticus DNA fragment in pBR322 with regeneration of flanking PstI sites. HindIII endonuclease cleaved both plasmids only once producing Form III molecules migrating at clearly different positions (slots h--~). The approximate size of pDI10 DNA is 7 kb. In order to confirm that the pDI10 plasmid carries the gene conferring the restriction property, HB101 cells were transfected with the isolated pDI10

106

TABLE II RESTRICTION A N D MODIFICATION OF x BY THE HYBRID CLONE Efficiency of plating of x on strains Phage a

pBR322/HB101

pDI101HB101

E. coli K12 r+m +

k.0 x.pBR322/HB101 x- pDI10/HB101

1 1 1.1

1 . 2 - 1 0 -~ < 1 . 1 0 -4 1

1 . 5 - 1 0 -4 4 . 0 - 1 0 -4 4.1 • 10 -4

a X.0 was grown on HB101. x. p B R 3 2 2 / H B 1 0 1 and x- pDI101HB101 were obtained by picking single plaques from the respective cell lawns into I ml of TL broth and treating with chloroform.

D N A and selected for tetracycline resistance. Of numerous colonies, two were examined and both were indistinguishable from KJ10 with respect t o the restriction property. One of these new clones, KJ12, was selected for all subsequent studies. An estimate o f in vivo restriction activity in KJ12 was obtained by measuring efficiency o f plating (e.o.v.) with phage ~ as presented in Table II. KJ12 cells gave an e.o.p, of 10 -7 relative to p B R 3 2 2 / H B 1 0 1 cells. By way o f comparison, an r + m + strain of E. coli K12 gave an e.o.p, of 10 -4 . Having confirmed the presence of phage restriction in KJ12, it was expected that the corresponding modification system would also be present. This was demonstrated

c

A

X

e ~

El4 o.. u

~12 a) 4-

".-

E 0

0

-~

g6

0.6

"~

,,~

"o

I0(

o.5

. . . . .

8(

0.4

6(

o~ ~ u c

ffJ

~4 ~2 "

.~

8

a 4(

0.~

c 2(

0.1 ~

0

7

z

-0

" 19

~ 20

.--

21

~

?-2

~: 23

L. 24 25 Fraction

.

. . 26 27 number

28

= 29

~ 30

31

32

0

Fig. 4. Single-stranded DNA agarose chromatography of a protein extract from KJ12 cells. (®---I), [3H]methyl group incorporation, ( o - - - o ) , endonuclease activity as estimated by the extent of digestion of k DNA assayed by 1.4% agarose gel electrophoresis.

107

a b .c d

7z7

50(

427,417,411 30t 24( 20( 15| 139

[]

Fig. 5. Comparison of HhaII and HinfI cleavage patterns of ~X174 RFI DNA by polyacrylamide gel electrophoresis. (a) HhaII digest. (b) double digest with HhaII and HinfI. (c) HinfI digest. (d) HaeIII digest. Assignments of fragment lengths in base pairs were taken from the ~X sequence data of Sanger et al., 1977.

by analyzing phage progeny from a high titer infection of KJ12 cells.These phage plated with an e.o.p, of I on KJ12 cells,but were stillrestricted on E. coli K12 r+ m + cells.The modification was not specified by p B R 3 2 2 itself since phage grown on p B R 3 2 2 / H B 1 0 1 were fully restricted by KJ12. Identificationof the cloned R - M system as l-]haiL T w o class II R-M systems, Hhal (Roberts et al.,1976) and HhaII (Mann, unpublished), are k n o w n to

108

efmt in the H. haemolyticus strain. The clone could be c ~ g either of these systems. To determine which was involved, pDI10 DNA was cleaved with restriction enzymes b ~ n g these two specificities. Hinfl, an isoschizomer of HhaII, was substituted for HhaH enzyme in this test because of its availability. As shown in Fig. 3, pDI10 DNA is susceptible to HhaI cleavage (slots a and b), but is resistant to HinfI digestion (slots c and d), a l t h o ~ the parental plasmid, pBR322, contains numerous sites for both endonucleases. We conclude that pDI10 DNA has acquired the HhaH modification pattern. Presence o f HhalI restriction endonuclease and DNA methylase activities in extrocts o f KJ12 cells. A protein extract was obtained from 3 liters of KJ12 cells as d e s c r i ~ , L-~ ~'~'HODS. The extract was fractionated on a column of single-stTanded DNA ~.~tr~se. Column fIactions were analysed for DNA methylase and restrictio~ endonuclease activities (Fig. 4). The substrate for methylase activity was sonic~ted DNA from the isogenic parent of KJ12, pBR322/HB101. This substrate was selected so as to detect only the cloned methylase activity, since other endogenous methylases are known to be present (Mafinus and Monks, 1973). A peak of methylase activity was eluted at approx. 0.15 M NaCI. Endonuclease activity eluted at about 0.45 M NaCI. Both enzymes exhibited the expected specificity. The endonuclease specificity was demonstrated by its cleavage pattern of ~X174 RFI DNA (Fig. 5). The methylase, when used to exhaustively methylate ~X174 RFI DNA, rendered the DNA resistant to cleavage by Endo R. HhaH (data not shown). ~o~X

pBR322 DNA H. hemolylicusDNA

135

somm

_~:

Fig. 6, A cleavage map of pDX10. Data for the pBR322 pozt~on of the map aze from Bolivar et al., 1977. The numbers indicate kilobases,

109

A cleavage map o f PDIIO DNA. Single and double digests of pDI10 DNA were obtained with the restriction enzymes BamHI, EcoRI, PstI, and HindIII The lengths of digestion products were analysed by gel electrophoresis (data not shown) to yield the cleavage map shown in Fig. 6. There are 2 PstI cleavage sites located at the extremes of the inserted H. haemolyticus DNA segment, and 2 BamHI sites plus a single EcoRI site within the insert. DISCUSSION

Historically, the construction of recombinant DNA molecules has involved the use of two principal methods for the joining of DNA fragments; annealing of DNA termini extended with complementary homopolymers (Jackson et al., 1972; Lobban and Kaiser, 1973), and enzymatic ligation of the cohesive termini generated by certain restriction enzymes (Mertz and Davis, 1972). With the extension procedure, self annealing i~ inhibited while hybrid formation by joining of complementary homopolymer extensions is strongly enhanced. However, recovery of inserted fragments from vector DNA is not readily accomplished. On the other hand, the procedure based on ligation of cohesive restriction termini produces hybrid molecules at relatively low efficiency due to cyclization and other non-productive joinings, but restriction sites are regenerated, thus facilitating subsequent recovery of inserts. The method used here (Bolivar et al., 1977) retains the advantages of both procedures. The 3' single-stranded termini generated by PstI cleavage are efficient primers for terminal transfersse and addition of dG residues to the cleaved vector DNA regenerates the PstI cleavage sites. Likewise, PstI-digested chromosomal DNA is efficiently extended with dC residues, whereas fragments produced by mechanical shear would require treatment with X exonuclease to ensure that all 3' termini are exposed for addition (Lobban and Kaiser, 1972). Annealing of the complementary extensions gives, in theory, only hybrid molecules and these appear to be repaired efficiently in vivo to give intact PstI sites at the junctions between vector and inserted DNA. Selection for recombinant clones carrying an R-M system can be based either on detection of new methylation or new restriction properties. We chose screening by phage restriction because the host cell is an r - m - derivative and the pBR322 plasmid vector imparts no new restriction activity. Thus, the appearance of a restriction phenotype after introducing recombinant plasmids should be an indication of expression of R-M system genes carried by the inserted DNA. In our hands, individual recombinants could easily be scored for restriction on plates seeded with phage. It ~hould also be feasible to enrich for r + clones by phage infection in liquid culture, provided such clones do not have an intrinsic growth disadvantage as is manifested, for example, by some clones in the Clarke and Carbon (1976) col!~ction. An alternative selection procedure based on methylation has not yet been explored. This involves growing recombinant clones (either individually or in mass liquid culture), extracting the plasmid DNA, and digesting with the

110 relevant restriction enzyme. Only fully methylated plasmid molecules, i.e., those carrying the methylase gene, should survive; and on re-transfection, recombinant clones should be highly enriched for methylase gene recombinants. This scheme c o ~ allow cloning of the methylase gene alone, whereas the restriction procedure selects only for recombinants ~ i n g both genes. Several conditions must generally be met if one is to achieve success in cloning of a particular R-M system using our procedure. First, the r + and m + genes must be closely enough linked to be included on a single fragment of DNA of a few million daltons. Second, it must be possible to introduce the genes into a host cell without lethal effects, i.e., the modification activity must express itself well in advance of the restriction activity. Third, the genes must be expressed and confer the classical restriction and modification phenotype. Fourth, the genes should not contain an inordinate number of PstI sites, since this would greatly lower the probability of obtaining partial digestion products bearing these genes.'The HhaII system satisfies these conditions, but other R-M systems may not, especially with regard to the location of PstI sites. However, in such cases, other cloning procedures may prove satisfactory. In regard to the second condition, it has been shown for the phage Pl~pecified R-M system that modification is expressed more rapidly than restriction when cells are newly infected (Arber, 1974). This sequential action could be a fairly general property of R-M systems, allowing transfer to other cells. Sequential expression could be most simply achieved if methylase were to act as a positive regulator of restriction gene expression. The insert in pDI10 is 3 kb in length, an amount sufficient to code for about 120 000 daltons of protein. We have demonstrated two chromatographicaily separable enzymatic activities specified by the cloned D NA segment that may possibly account for most of the coding capacity. The absence of internal PstI cleavage sites in the cloned insert also simplifies its recovery for possible studies of gene arrangement, selective in vitro mutation, and even base sequence analysis as a means of deducing amino acid sequence. Transplantation of the genes into a viral vector may permit study of regulation and sequential expression on entry into a new host. Of greater interest is development of overproducing strains. While KJ12 appears to possess multiple gene copies and more abundant enzyme than in the cell of origin, it seems likely that with proper manipulation, a higher yielding clone can be developed. ACKNOWLEDGEMENTS We are i n d e b t ~ to Dr. H.W. Boyer for supplying the piasmid vector and a cleavage map prior to publication. We thank Dr. N. Musczycka for the gift of HinH enzyme. This work was supported by National Science Foundation Grant No. PCM74-19985 and by National ~ c e r Institute Grant No. CA16519. M.B.M. was the recipient of a Public Health S ~ i c e Research Fellowship Award No. AI01731.

111

REFERENCES

Arber, W., DNA modification and restriction, in Cohn, W.E. (Ed.), Progress in Nucleic Acids Research and Molecular Biology, Vol. 14, Academic Press, New York, 1974, pp. 1--34. Arndt-Jovin, D.J., Jovin, T.M., B~hr, W., Frischauf, A.-M. and I~arquardt, M., Covalent attachment of DNA to agarose. Improved synthesis and use in affinity chromatography, Eur. J. Biochem., 54 (1975) 411--418. Backman, K., Ptashne, M. and Gilbert, W., Construction of plasmids carrying the cI gene of bacteriophage k, Proc. Natl. Acad. Sci. USA, 73 (1976) 4174--4178. Bolivar, F., Rodriguez, R.L., Greene, P.J., Betlach, M.C., Heynecker, H.L., Boyer, H.W., Crosa, J.H. and Falkow, S., Construction and characterization of new cloning vehicles, H. A multipurpose cloning system, Gene, 2 (1977) 95--113. Clarke, L. and Carbon, J., A colony bank containing synthetic colE1 hybrid plasmid representative of the entire E. coU genome, Cell, 9 (1976) 91--99. Cleweli, D.B. and Helinski, D.R., Supercoiled circular DNA-protein complex in Escherichia coli: purification and induced conversion to an open circular DNA form, Proc. Natl. Acad. Sci. USA, 62 (1969) 1159--1166. DeWachter, R. and Fiers, W., Fractionation of RNA by electrophoresis on polyacrylamide gel slabs, in Grossman, L. and Moldave, K. (Ecls.), Methods in Enzymology, Vol. XXI, Academic Press, New York, 1971, pp. 167--178. Hirt, B., Selective extraction of polyoma DNA from infected mouse cell cultures, J. Mol. Biol., 26 (1967) 365--369. Hutchison, C.A. HI and Sinsheimer, R.L., The process of infection with bacteriophage ~X 174, X. Mutations in a ~X lysis gene, J. Mol. Biol., 18 (1966) 429--447. Jackson, D.A., Symons, R.H. and Berg, P., Biochemical method for inserting new genetic information into DNA of simian virus 40: Circular SV40 DNA molecules containing lambda phage genes and the galactose operon of Escherichia coli, Proc. Natl. Acad. Sci. USA, 69 (1972) 2904--2909. Johnson, P.H. and Sinsheimer, R.L., Structure of an intermediate in the replication of bacteriophage ~ X174 deoxyribonucleic acid: the initiation site for DNA replication, J. Mol. Biol., 83 (1974) 47--61. Kelly, T.J. Jr., and Smith, H.O., A restriction enzyme from Hemophilus influenzae, II. Base sequence of the recognition site, J. Mol. Biol., 51 (1970) 393--409. Lobban, P.E. and Kaiser, A.D., Enzymatic end-to-end joining of DNA molecules, J. Mol. Biol., 78 (1973) 453--472. Marinus, M.G. and Morris, N.R., Isolation of deoxyribonucleic acid methylase mutants of Escherichia coli K12, J. Bacteriol., 14 (1973) 1143--1150. Marmur, J., A procedure for the isolation of deoxyribonucleic acid from microorganisms, J. Mol. Biol., 3 (1961) 208--218. Maxam, A.M. and Gilbert, W., A new method for sequencing DNA, Proc. Natl. Acad. Sci. USA, 74 (1977) 560--564. Mertz, J.E. and Davis, R.W., Cleavage of DNA by RI restriction endonuclease generates cohesive ends, Proc. Natl. Acad. Sci. USA, 69 (1972) 3370--3374. Murray, K. and Murray, N.E., Phage lambda receptor chromosomes for DNA fragments made with restriction endonuclease IH of Haemophilus influenzae and restriction endonuclease I of Escherichia coli, J. Mol. Biol., 98 (1975) 551--564. Panasenko, S.M., Cameron, J.R., Davis, R.W. and Lehman, I.R., Five-hundred fold overproduction of DNA ligase after induction o~ a hybrid lambda lysogen constructed in vitro Science, 196 (1977) 188--189. Roberts, R.J., Restriction endonucleases, CRC Crit. Rev. Biochem., 4 (1976) 123--164. Roberts, R.J., Myers, P.A., Morrison, A. and Murray, K., A specific endonuclease from Haemophilus haemolyticus. J. Mol. Biol., 103 (1976) 199--208.

112

Rougeon, F., Kourilsky, P. and Maeh, B., Insertion of a rabbit #-globin gene sequence into an E. coli plasmid, Nuel. Acids Res., 2 (1975) 2365--2378. Sanger, F., Air, G.M., Barrell, B.G., Brown, N.L., Coulson, A.R., Fiddes, J.C., Hutebdson, C~A. IH, Slocombe,P.M. and Smith, M., Nucleotide sequence of bacteriophage ~X174 DNA, Nature, 265 (1977) 687--695, ..... Sharp, P.A., Sugden, B. and Sambrook, J., Detection . t two restriction endonucleeso activities in l-I~mophilus parain.ffuenzae ~ analytical agarose~ethidium bromide eleetrophoresis, Biochemistry, 12 (1973) 3055--°o0(~ . Thomas, C.A. Jr. and Abelson, J., The isolation and characterization of DNA from bacteriophage, in Cantoni, G . L and Davies, D.R. (Eds.), Procedures in Nucleic Acid Research, Harper and Row, New York, 1966, pp. 553--561. Weiss, B. and Milcarek, C., Mass screening for mutants w i ~ rJNeses by microassay techniques, in Grossman, L. and Moldave, K. (Eds.), 1Vi~thodsin Enzymology, Vol. =.XIX, Academic Press, New York, 1974, pp. 180--196. Wensink, P.C., Finnegan, D.J., Donelson, J.E. and Hogness, D.S., A system for mapping DNA sequences in the chromosomes of Drosophila melanogaster, Cell, 3 (1974) 315--325. Communicated by A. Skalka.