THE JOURNAL OF BIOLOGICAL CHEMISTRY 0 1990 by The American Society for Biochemistry and Molecular Biology, Inc.

Vol. 265, No. 30, Issue of October 25, pp. 18311-18317,199O Printed in U.S. A.

Properties of Overexpressed Phage T5 D15 Exonuclease SIMILARITIES WITH ESCHERICHIA COLI DNA POLYMERASE 15’-3’ EXONUCLEASE*

(Received for publication, March 16, 1990)

Jon R. Sayers and Fritz EcksteinS From the Man-Planck Institut fiir Experimentelle Medizin, Abteilung Chemie, Hermann-Rein-Strasse 3, 03400 Gtittingen, Federal Republic of Germany

The D15 gene of the bacteriophage T5, thought to encode an exonuclease, was cloned into an Ml3 phage on a 1344-base pair fragment. The deduced amino acid sequence of 291 residues (Kaliman, A.V., Krutilina, A. I., Kryukov, V. M., and Bayev, A. A. (1986) FEBS Lett. 195, 61-64) shows a high degree of homology with the first 320 amino acid residues of Escherichia coli DNA polymerase I, the region containing the en- zyme’s 5’-3’ exonuclease activity. Recombinant Ml3 phage DNA was manipulated by oligonucleotide-di- rected mutagenesis to enable subcloning into a high efficiency expression vector, allowing the production of large amounts of enzyme for physical characteriza- tion and crystallization trials. The enzyme was puri- fied to homogeneity. The purified enzyme is active on both native and heat-denatured DNA and shows no endonuclease activity on either double-stranded closed-circular or nicked DNA. The enzyme is also able to degrade some oligonucleotides in a manner which depends not only on the nucleotide sequence but also on the state of hybridization of the potential substrate. The mode of action of this enzyme is similar to, al- though not identical to that of the 5’-3’ exonuclease activity of E. coli DNA polymerase I.

Infection of Escherichia coli cells with bacteriophage T5 results in an increase in deoxyribonuclease activity (Stone and Burton, 1962) which is manifested in viuo as the destruc- tion of host cell DNA (Lanni, 1969). Phage T5 specifies several endonucleases, four of which have been isolated by Rogers and Rhoades (1976). A T5-specified exonuclease was first purified from phage-infected cells by Paul and Lehman (1966) and later characterized as a 5’-3’ exonuclease by Frenkel and Richardson (1971). The latter workers showed that although the exonuclease releases some mononucleotides and dinucleotides, small oligonucleotides are the major hy- drolysis products. Moyer and Rothe (1977) later isolated the exonuclease from phage-infected cells and demonstrated an associated endonuclease activity which required DNA with a pre-existing nick as a substrate.

The D15 gene is thought to encode the nuclease described by Paul and Lehman (1966) and Frenkel and Richardson (1971). The nuclease is thought to be involved in the bacte- riophage T5 transcription-replication enzyme complex (Ficht and Moyer, 1980). The D15 gene product is required for transcription of late T5 genes (Chinnadurai and Mc-

* The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked “aduertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

$ To whom correspondence should be addressed.

Corquodale, 1973) and for normal DNA replication (Frenkel and Richardson, 1971). The exonuclease has been likened to the 5’-3’ exonuclease activity of E. coli DNA polymerase I, which is localized to amino acid l-298 of the latter enzyme (Kelly and Joyce, 1983). However, the D15 exonuclease is not thought to be the enzyme responsible for destruction of the host cell DNA. This latter function has been ascribed to the gene Al product (Lanni, 1969).

We wished to clone and overexpress the T5 D15 exonucle- ase to enable investigations into the properties of the pure enzyme and to produce large amounts of protein for crystal- lization experiments. The cloning of the D15 gene was re- ported recently, although attempts to overexpress the enzyme were unsuccessful (Kaliman et al., 1986).

MATERIALS AND METHODS’

RESULTS

The cloning strategy used for the isolation and subcloning of the T5 D15 exonuclease gene is shown in Fig. 1. First a 1344-bp’ HueIII/PstI fragment was cloned from T5 viral DNA into the SmaI/PstI sites of M13mp18 DNA. Single-stranded recombinant DNA (designated M13T5) was isolated for se- quencing and site-directed mutagenesis. The cloned fragment carrying the DlS-coding region contained a 300 bp sequence upstream of the D15 native ribosomal-binding site and prob- able promotor region.

To circumvent any deleterious effects of the D15 gene’s native promotor (or the 5’-flanking region) in the proposed expression vector we introduced a Sac1 site between the native promotor sequence and the D15 ribosomal-binding site region (Fig. 2). This was accomplished using the phosphorothioate- based site-directed mutagenesis procedure (Sayers et al., 1987). This allowed the coding sequence to be readily sub- cloned into the tightly repressed but inducible expression vector pDOC55 (O’Connor and Timmis, 1987).

The newly introduced Sac1 site and the already present HpaI site which flank the D15 open reading frame were used to excise the target sequence. The D15 gene was cloned under the control of the APL. As can be seen from SDS-PAGE gel (Fig. 3) heat induction of the over-expressing cell line M72( X, pJON45) gave yields of induced protein which were easily

1 The “Materials and Methods” are presented in miniprint at the end of this paper. Miniprint is easily read with the aid of a standard magnifying glass. Full size photoco$es are included in the microfilm edition of the Journal that is available from Waverly Press.

’ The abbreviations used are: bp, base pair; dA,, polydeoxyadenylic acid; DTT, dithiothreitol: PAGE. Dolvacrvlamide ael electrouhoresis: RBS, riboiomal binding site; RFI; i1, and iI DNA iefer to supercoiled circular, open (nicked) circular and linear double-stranded DNA, respectively; SDS, sodium dodecyl sulfate; HPLC, high performance liquid chromatography; pol I, polymerase I.

18311

by guest on May 31, 2018

http://ww

w.jbc.org/

Dow

nloaded from

18312 Properties of Cloned T5 015 Exonucleme

detectable by eye. Scanning laser densitometry of this Coo- massie Blue-stained SDS-PAGE gel showed that the induced protein accounted for approximately 8% of stained soluble cell protein.

The induced enzyme was found in the soluble cell extract which was first subjected to treatment with polyethylenimine to precipitate nucleic acid. This step was carried out at inter- mediate ionic strength which avoided loss of protein during the precipitation. The clarified supernatant was then sub- jected to an ammonium sulfate treatment to precipitate the protein while leaving excess polyethylenimine in solution. The resuspended pellet was then dialyzed in a low salt buffer allowing the fortuitous precipitation of the T5 exonuclease in partially purified form. Scanning laser densitometry of the Coomassie Blue-stained SDS-PAGE shown (Fig. 3) revealed that the T5 exonuclease was 28% pure by this stage. The precipitated proteins were taken up in buffer of intermediate ionic strength and further purified by chromatography on phosphocellulose and DEAE ion exchange resins. The phos- phocellulose eluate was found to be greater than 90% pure by laser densitometry of a Coomassie Blue-stained SDS-PAGE. The DEAE fraction (fraction IV) was found to be homoge- neous on analytical SDS-PAGE even when overloaded (Fig. 3). Table I outlines the purification procedure and indicates the approximate specific activities obtained during the pro- cedure.

The over-expressed enzyme migrates on an SDS gel with a mobility consistent with the calculated molecular mass of 33.4 kDa. Amino acid sequencing of the first 20 amino acid residues revealed that the protein had lost the first methionine residue in vivo. Otherwise the sequence was as deduced previously (Kaliman et al., 1986). We found that the protein has a tendency to precipitate in low ionic strength buffers. The protein could also be precipitated isoelectrically in the range

M13T5

* 1344 bps

c Hpa I

(Hae III) Pstl

I I-

1’ \ Native Promotor D15 Open Readlng Frame

676 bps

FIG. 1. Schematic diagram showing the cloning of the D15 gene. The 1344-bp HaeIII/PstI fragment carrying the D15 open reading frame was cloned between the SmaI and PstI sites of M13mp18. Bracketed sites were destroyed during construction of the recombinant phage M13T5.

-35

GTTGACAICT TAATGAGTCG

AGAAAGACAT GGAAATCGTA

GCTACATCAT GCATCTGTTG

pH 4-5. The enzyme is absorbed by both DEAE and phos- phocellulose ion exchangers at pH 7.5-8.0. However, the enzyme is not absorbed by cation exchangers such as sulfpro- pyl and carboxymethyl containing resins under these condi- tions which indicates that phosphocellulose is acting as an affinity medium for this enzyme under these conditions. Throughout the purification procedure the presence of a sulfhydryl reagent was required to maintain activity during storage. However, the absence of such a reagent in assay buffers had no significant effect on the rate of enzymatic reaction (data not shown). No loss of activity was observed in samples of enzyme stored at -20 “C in 50% glycerol (v/v) for several months.

Enzymatic Properties of the Purified Enzyme-Recombi- nant T5 D15 exonuclease had a specific activity in excess of lo6 units/mg (Table I) when assayed in the absence of added salt with native calf thymus DNA as substrate. This activity corresponds to approximately 550 nmol of released nucleo- tide/s/mg of enzyme. Thus, the turnover number in nucleo- tides converted to acid-soluble products is 18 s-l. The enzyme was stimulated slightly by added KC1 or NaCl (75 mM, 1.4- and 1.8-fold stimulation, respectively) in the standard assay buffer which contained 25 mM potassium glycinate (data not shown). The effect of KC1 and NaCl on the enzyme were compared using ethanolamine-glycine substituted for potas- sium glycinate in the standard assay. KC1 was slightly stim- ulatory in the range 25-150 mM (1.4-fold increase in activity at 75 mM) while similar concentrations of NaCl were not stimulatory (data not shown). The enzyme was also able to hydrolyze heat-denatured DNA as effectively as native DNA (data not shown).

We then investigated the action of the enzyme on pUC19 RF11 DNA (2686 bp), nicked in one strand only. Less than one-quarter molar equivalent of T5 exonuclease (12 ng) con- verted the nicked pUC19 (2.5 pg) to a fully gapped species (buffer conditions: 40-~1 reaction, 25 mM potassium glycinate buffer, pH 9.3, 10 mM MgC12, 60 mM KCl). However, electro- phoretic analysis of this gapping process showed that the DNA appeared as a diffuse smeared-out streak which even- tually reformed as a discrete band of fully gapped product after 20 min (results not shown). Calculation of the rate of nucleotide hydrolysis under such conditions gives a value of approximately 10 nucleotides/enzyme molecule s-‘.

The rate of this gapping reaction with nicked pUC19 is readily controlled by the use of high salt buffer conditions which slows down the enzymatic reaction. Thus, we were able to obtain intermediately gapped plasmid DNA, gapped at the rate of approximately 300 bases/min by using a slight molar excess of enzyme under conditions of high ionic strength. These results are shown in Fig. 4. Faint bands with low electrophoretic mobility can be also seen in this gel photo which appear to be concatameric plasmids, which parallel the behavior of the main species present.

The purified enzyme is free of double-stranded endonucle- ase activity. The incubation of a large molar excess of enzyme

-10

TACCCATATG GTGTTAAATA

ATTGGTTTAT TTGAGCCGTG

AGGACTTAAT TAAATAATG..

FIG. 2. Nucleotide sequence immediately upstream of the D15 open reading frame. The possible native promotor sequences upstream of the D15 open reading frame are boxed. The Shine-Delgarno sequence and initiator methionine codon are underlined. The sequence GAAATC (doubly underlined) was changed to the Sac1 recognition sequence GAGCTC by site-directed mutagenesis (see “Materials and Methods”).

by guest on May 31, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Properties of Cloned T5 D15 Exonuclease 18313

1234567

FIG. 3. SDS-PAGE showing the induction and purification of the T5 D15 exonuclease. SDS-PAGE was performed under reducing conditions in a 10% polyacrylamide gel. Molecular weight markers (M,), phosphorylase b (94,000), bovine serum albumin (64,000), ovalbumin (43,000), carbonic anhydrase (30,000), soybean trypsin inhibitor (20,100), and Lu-lactalbumin (14,400) (lane I). Total protein from JM83(X) cells carrying pJON45 before induction (lane 2). Total protein from heat-induced JM83(X) cells carrying pJON45 (lane 3). Fraction I, soluble proteins from heat induced cells (lane 4). Fraction II, precipitated proteins resuspended and applied to phos- phocellulose (lane 5). Phosphocellulose purified, Fraction III (lane 6) and the DEAE-purified protein, Fraction IV (lane 7). Fractions I-IV are defined under “Materials and Methods.”

TABLE I Purification of cloned T5 exonuclease

The fraction numbers (I-IV). are defined in the text.

FIG. 5. Agarose gel analysis showing the absence of double- stranded endonuclease activity associated with T5 exonucle- ase. Aliquots of 2 pg of double-stranded, closed-circular M13mp18 DNA (10 ~1 of 25 mM potassium glycinate, pH 9.3, 10 mM MgCl,) were incubated for 30 min at 37 “C. The reactions were stopped by the addition of an equal volume of stop mix and half of each sample analyzed by electrophoresis on a 1% agarose gel. The photograph shows the result of incubation with no enzyme, 20 ng, 40 ng, 100 ng, or 1 pg of T5 exonuclease (lanes 1-5, respectively). A marker of BstEII digest of phage lambda DNA is shown with 8454-, 7242-, 6369-, 5686-, 4822-, 4324, 3675, 2323-, and 1919-bp fragments vizable (lane 44).

Fraction Total protein Total activity Specific activity

w units” units/mg Crude extract (I) 800 3 x 10’ 3.8 x lo4 First precipitate (II) 30 2.5 x 10’ 8.3 x lo” Phosphocellulose (III) 14 1.2 x 10’ 8.6 x lo” DEAE (IV) 10 1.3 x 10’ 1.3 x 10”

” One unit releases 1 nmol of acid soluble nucleotides at 37 “C in 30 min under standard assay conditions (see “Materials and Meth- ods”).

RF II

RF III

0 2 4 6 8 10 RFI

FIG. 4. Agarose gel electrophoresis analysis of the gapping of nicked double-stranded plasmid DNA. Circular nicked plas- mid DNA (4 pg) was incubated at 37 “C with T5 exonuclease (50 ng) in 50 ~1 of 100 mM KYHPO,, pH 9.3, 50 mM NaCl, 10 mM MgC12. Samples (4-12 11) were removed and terminated by the addition of an equal volume of stop mix at the times indicated above the lanes (min). The samples were then analyzed on a 2% agarose gel. A marker of closed circular double-stranded plasmid (RF I) is shown. The terms RF II and RF III denote nicked and linearized pUC19 DNA (2686 bp), respectively.

with double-stranded, closed circular plasmid DNA revealed no significant loss of double-stranded, closed circular material as monitored by agarose gel electrophoresis (Fig. 5). At the highest concentration of enzyme tested, the plasmid DNA exhibits a significant shift in mobility (Fig. 5, lane 5). This shift is presumably due to the large amount of protein used in the assay binding to the DNA (approximately a 70-fold molar excess of enzyme over M13mp18 DNA).

The enzyme degraded oligonucleotides very slowly com-

pared with high molecular weight DNA. However, the diges- tion of dTls was easily monitored by reverse-phase HPLC. Coinjection experiments showed that one of the major prod- ucts of the reaction was the tetramer dT, (results not shown). We then over-digested several 5’-‘“P-labeled oligonucleotides with the enzyme in order to examine the end products ob- tained by such over digestions using analytical PAGE. The results obtained from the digestion of dTIP-lR are shown in Fig. 6. In this case the enzyme produces tri- (40%), tetra- (30%), and pentamers (25%) from single-stranded dT1g.,R. However, mono- (50%), di- (22%), and trinucleotides (23%) were produced when labeled dTIP-IR was digested in the pres- ence of unlabeled polydeoxyadenosine. Very similar results were obtained with the oligonucleotide dT1,. The distribution of products also followed a similar pattern when 5’-labeled dAls was digested with the enzyme. Thus, labeled trimers, tetramers, and pentamers were produced from dAls alone, and monomers, dimers, and trimers were produced in the presence of the unlabeled complementary oligonucleotide dTls. The results obtained using different oligonucleotides are summa- rized in Table II. Furthermore the T5 exonuclease seemed unable to digest dT, even upon extended incubation. While dT, was cleaved by the enzyme, it was digested approximately lo-loo-fold more slowly than a mixture of dT12-18. As dTs persists in the reaction products obtained by the digestion of larger oligothymidylates (in the absence of oligo(dA)) it may be assumed that this too is a very poor substrate for the enzyme. The large excess of enzyme employed and the reten- tion of the ayP label as nucleotide in these experiments indi- cates that the enzyme is free of contaminating phosphatase activity.

The D15 exonuclease was unable to cleave dGls unless it was annealed to d&. The latter oligonucleotide, however, was cleaved to give trimers, tetramers, and pentamers as the major products whether present alone or annealed to dG1,. The results show that the mode of action of the enzyme on short oligonucleotides is dependent both on the base sequence and whether or not the oligonucleotide is double- or single- stranded.

by guest on May 31, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Properties of Cloned T5 015 Exonuclease 18314

WPT),

18

12

6

2

1234567

r-1 & “!

FIG. 6. Autoradiograph of analytical PAGE analysis of the complete digestion of 5’-32P-laheled oligonucleotides dTlz-,s with T5 exonuclease. The size of the markers are indicated at the left side of the figure. Aliquots of dTIP.le (2 x lo-” mol of nucleotide) were reacted with either zero, 3 X lo-“, 3 X lo-‘“, 3 X lo-“, 3 X lo-” mol of T5 exonuclease at 37 “C for 5 min (lanes l-5, respectively). Markers of 5’-“‘P-labeled dTZ and dTs (lanes 6 and 7), respectively. Details as described under “Materials and Methods.”

TABLE II Action of T5 exonuclease on various homopolymeric oligonucleotides

The approximate percentages of labeled products (where indicated) were determined by scanning video densitometry of photographic negatives obtained by autoradiography of 20% analytical PAGE gels. For conditions see “Materials and Methods.” *, indicates the 5’-32P- labeled oligonucleotide.

Oligonucleotide(s) substrates Major labeled products

d*T,,-,, d*T,s d*A,s

E&A,, d*T,,.,n/& d*T,s/d&, d*A,s/dT,s d*G,s d*Gs d*C,s/dG,s d’G,JdC,s

3-mer (40%), 4-mer (30%), 5-mer (25%) 3-mer (40%), 4-mer (30%), 5-mer (25%) 3-mer (40X), 4-mer (30%), 5-mer (25%) 3-mer, 4-mer dTMP, 3-mer, 2-mer dTMP (50%), 3-mer (23%), P-mer (22%) dTMP (50%), 3-mer (23%), 2-mer (22%) dAMP (50%), 3-mer (23%), 2-mer (22%) Not degraded 3-mer, 4-mer, 5-mer 3-mer, 4-mer, 5-mer dGMP. 3-mer

The apparent size of the products obtained on incubating 5’ “*P-labeled mixed sequence oligonucleotides with the D15 exonuclease is shown in Table III. The apparent size of the products was determined relative to labeled oligo(dT) of known lengths. The effect of base sequence on electrophoretic mobility was not taken into account, although the presence of faint bands corresponding to minor labeled failure se- quences present as contaminants within the substrate assisted in product analysis. However, it is clear that the pattern of products released from the mixed sequence oligonucleotides differs markedly from sequence to sequence. These patterns also differ from those obtained by degrading any of the homopolymers discussed above.

Sequence Homologies with E. coli DNA Polymerase I-The

TABLE III Action of T5 ewonuclease on various mixed sequence oligonucleotides

The major labeled products of over-digestion of the specified oli- gonucleotides with excess T5 exonuclease. For conditions see “Ma- terials and Methods.” The approximate percentages of labeled prod- ucts (where indicated) were determined by scanning video densitom- etry of photographic negatives obtained by autoradiography of 20% analytical PAGE gels.

Oligonucleotide substrate (5’-“‘P Apparent size of major labeled) labeled products”

5’- dCACAACTCGAAAGCATCC P-mer, 5-mer 5 ‘ - dGGTAAATCAAACCAGAATC 2-mer, 5-mer, 6-mer 5'-dAGGGTTTTCCCAGTCACG I-mer 5'-dTCCCGGGTACCGAGCTCT 1-mer, 2-mer, 3-mer

‘Apparent size based on comparison with the migration of 1-6- mer dTMP markers.

PO11 1 . . . . . ..MVQ......IPQNPLILVDGSSYLYRAYHAFFFLTNSAGEPTG 37 ::: . I::IIl... :I I I.. .I :

T5 EXO 1 MSKSWGKFIEEEEAEMASRRNLMIVDGTNLGFRFKH......NNSKKPFA 44

38 AMYGiLNMLRSLIMQYKPTHAAWiDAKGKT.FR;ELFE"YKSHi....P 82 I :. :.I1 I ..:..:I::.III. II I :..(I::1

45 SSY..VSTIQSLAKSY.SARTTIVLGDKGKSVFRLEHLPEYKGNRDQKY~ 91

83 PMPDiLRAQIEPLHiMVK.AMGL . . . . ..PLLAVSGVEADDVIGTLAREAE 126 . ...: :I I.: :I I::1 I :.:.IIIllI: : :.: :

92 QRTEEEKALDEQFFEYLKDAFELCKTTFPTFTIRGVEADDmYIVKLIG 141

127 KAGR&.LIST.GDiDMAQLVTPNITLINTMTNTILGPEEWNKYGV.PP 173 I IIII II.1 1:1..:. :. I. :...I

142 HLYDHVWLISTDGDWDT..LLTDKVSRFSFTTRRE;HLR~~YER~NVDDV 189

174 ELIIDFLALMGDSSDNIPGVPGVGEKTAQALLQGLGGLDTLY~E~EKIAG 223 I :I.: I:III :IlI.II.I:I.I : .::.::I.: I : I..

190 EQFISLKAIMGDLGDNIRGVEGIGAKRGYNIIREFGNV..L....DIIDQ 233

224 LSFRGAKTMAAKLEiNKEVAYLSYiLATIKTDVEiELTCEQLEVi?QPAAE 273 I.:.1 I:...1: : I

234 LPLPGKQKYI&LNASEELLFRNLIL..... I:I. I ::. .:.::

.VDLPTYC..VDAIAAVGQ 275

274 ELLGLFKKYEFKRWTADVEAGKWLQAKGAKPAAKPQETSVADEA~EVTAT 323 ::I: ::I 1: I: :I::

276 DVLD........KFTKDI.....LE..............IAEQ*...... 292

FIG. 7. Sequence alignment of E. coli DNA polymerase I (upper sequences) with T5 exonuclease. The upper sequences are from E. coli DNA polymerase I, residues l-323 (Poll), the lower sequences are from the T5 D15 exonuclease (T5 Exe). The symbols I, :, and . indicate identical, similar, and not dissimilar amino acids, respectively.

DNA sequence determined by Kaliman et al. (1986) was translated into its amino acid sequence. The amino acid sequence alignment between E. coli DNA polymerase I and T5 exonuclease are shown in Fig. 7. The overall sequence identity at the amino acid level is 28% with 54% similarity.

DISCUSSION

The bacteriophage T5 D15 exonuclease gene was originally cloned and sequenced by Kaliman and co-workers (1986) and was shown to contain an open reading frame coding for a 291- amino acid polypeptide. These authors failed to identify any sequences typical of an E. coli promoter region. They were also unsuccessful in attempts to over-express the enzyme. However we noticed, on inspection of the DNA sequence, a perfect consensus -35 sequence (TTGACA) positioned 17 bases upstream of a possible -10 sequence (CATATG) situ- ated some 70 bp upstream of the D15 ribosomal-binding site sequence (Fig. 2).

The tightly repressed, yet inducible pDOC55 expression plasmid was chosen for the over-production of the possibly toxic exonuclease. The pDOC55 vector carries two convergent promoters flanking a multiple cloning site, namely the phage lambda leftward promotor (Remaut et al.,1983) and the luc promotor (PLac). A gene may then be introduced into the multiple cloning site between the two convergent promotors such that it is under control of the XPL. Any leakage from

by guest on May 31, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Properties of Cloned T5 D15 Exonuclease 18315

this promoter is effectively suppressed by the convergent PLEA. The latter may be induced with isopropyl-1-thio-P-D-galac- topyranoside for maximal repression. This system has been used to clone the EcoRI restriction endonuclease gene in cells lacking the restriction modification methylase (O’Connor and Timmis, 1987). Cells carrying the pDOC55 plasmid derivative containing the D15 gene (pJON45) grew normally in rich media compared with cells carrying the parent plasmid. How- ever, using similar methodologies to those presented here we have also cloned the D15 gene under the control of the less tightly repressed hybrid trp-lacpromotor (DeBoer et al., 1983). Cells carrying such a construct grew relatively poorly and so the pDOC55 derivative was used in this work.

The cloned, homogeneous T5 enzyme has a turnover num- ber of approximately lo-30 nucleotides s-l at 37 “C! under the conditions reported herein. This compares with the value of 0.2 nucleotides s-’ at 30 “C! reported by Joannes et al. (1985) for the enzyme purified from virus-infected cells. However, their enzyme was not homogeneous, and these workers deter- mined the turnover number at pH 8. Paul and Lehman (1966) determined that the enzyme has a pH optimum of 9.3 in glycinate buffer and is less active at pH 8. The suboptimal buffer and heterogeneous enzyme preparation used by Joannes et al. (1985) probably accounts for the discrepancy between their results and the findings presented here. The observation that NaCl and KC1 stimulate the enzyme activity in a manner which is dependent upon the buffer used is probably due to the different ionic strengths of the two buffers. Under conditions that allowed direct comparison of the effect of the two salts (ethanolamine/glycine buffer as opposed to potassium glycinate), KC1 was found to be a better stimulant than NaCl. This finding indicates that the mecha- nism of this stimulation is not simply dependent upon total ionic strength.

The D15 nuclease is thought to be part of the bacteriophage T5 transcription-replication enzyme complex (Ficht and Moyer, 1980). The enzyme is required for transcription of late T5 genes (Chinnadurai and McCorquodale, 1973) and for normal DNA replication (Frenkel and Richardson, 1971). The exonuclease has been likened to the 5’-3’ exonuclease activity of E. coli DNA polymerase I which is localized to amino acids 1-298 of this enzyme (Kelly and Joyce, 1983). A recent report of amino acid sequence homology between these two proteins has been presented in schematic form (Leavitt and Ito, 1989). However, no amino acid sequence alignments were shown. The amino acid sequence comparison presented here (Fig. 7) shows extensive homology between the two proteins. The most striking features are the shared six amino acid sequence GVEADD (residues 126-131 of the T5 enzyme) and the four amino acid sequence LIST (1499152 of the T5 enzyme). Amino acids 197-215 of the T5 enzyme and 182-200 of E. coli DNA pol I share 12 identical and 3 similar residues. A search for the sequence motif GVEADD revealed that phage T4 gene 47 (Gram and Ruger, 1985) contains the closely related se- quence GVKADD (amino acids 12-17). The T4 protein is thought to be an exonuclease involved in an enzyme complex which may be responsible for generating single-stranded DNA regions (Prashad and Hosoda, 1972). A further amino acid comparison of the T5 and T4 enzymes disappointingly showed little additional similarity.

Early studies on the 5’-3’ exonuclease of E. coli DNA polymerase I showed that oligothymidylate is degraded to thymidylic acid as the sole product (Kelly et al., 1969). The same study also showed that when this substrate is annealed to polydeoxyadenylate the oligothymidylate is digested to give the dinucleotide (21%) and the mononucleotide (79%) as

products. These results show that the D15 exonuclease reacts in a similar although not identical manner to the 5’-3’ exonuclease of E. coli DNA polymerase I. Furthermore, T5 exonuclease appears to display some base selectivity when cleaving oligonucleotides as shown by the patterns of labeled product formation with different mixed-sequence oligonucle- otide substrates.

Lundquist and Olivera (1982) have shown that displaced single-stranded overhangs are transiently produced and de- stroyed during nick translation by E. coli DNA polymerase I. They suggest that polymerization leads to strand displace- ment of the 5’ end of the nick by the growing 3’ terminus. The 5’ overhang so produced is then hydrolyzed to yield a nick by the 5’-3’ exonuclease activity. The mode of action of T5 D15 exonuclease in uiuo may well be analogous to that of the 5’-3’ exonuclease activity of E.coli DNA pol I. The T5- induced DNA polymerase (gene D9) has extensive sequence homology to E.coli DNA polymerase I (Leavitt and Ito, 1989) and the T5 D15 exonuclease has homology with the amino terminus of E.coli DNA pol I. Thus, it may be inferred that together the T5 D15 and D9 gene products fulfill a similar role in phage T5 as that carried out by E. coli DNA polymerase I in uninfected E. coli. This raises the question why should the T5 phage synthesize two separate polypeptides to fulfil the function of the host DNA polymerase I, if it does indeed do the same job? Perhaps, as the phage’s 5’-3’ exonuclease activity is not chemically bound to it’s polymerase, it is possible that the exonuclease is also involved in other func- tions. The presence of single-stranded regions has been de- tected in replicating T5 DNA (Carrington and Lunt, 1973) and may be important in the formation of the transcription- replication complex (Ficht and Moyer, 1980). Single-stranded regions may also be involved in the regulation of late gene transcription (Chinnadurai and McCorquodale, 1973). Thus, it is possible that the D15 exonuclease is involved not only in nick translation but also in the production of single-stranded DNA regions.

The D15 exonuclease may prove a useful tool in manipu- lation of DNA, as under appropriate conditions it is possible to degrade DNA in a synchronous manner. The D15 enzyme is a 5’-3’ exonuclease and as such it forms the complement to E. coli exonuclease III which is a 3’-5’ exonuclease. The latter has been used for several applications such as deletion cloning (Henikoff, 1984), targeted random mutagenesis (Shor- tle et al., 1982), and oligonucleotide site-directed mutagenesis (Taylor et al., 1985b; Nakamaye and Eckstein, 1986). Indeed, the over-expressed D15 exonuclease has been used success- fully in the plasmid mutagenesis method recently described by Olsen and Eckstein (1990), and as the 5’-3’ exonuclease in oligonucleotide site-directed mutagenesis (Sayers et al., 1988). Furthermore, the D15 exonuclease has the advantage over exonuclease III and T7 exonuclease in that it is able to degrade either single-stranded or duplex DNA. In contrast the latter two enzymes are essentially inactive on single- stranded DNA (Brutlag and Kornberg, 1972; Kerr and Sa- dowski, 1972).

The enzyme preparation presented here has been scaled up severalfold allowing the production of large amounts of ho- mogeneous protein allowing crystallization trials to begin.

Acknowledgments-We would like to thank Dr. J.-M. Saucier, Laboratoire de Biochemie et Enzymologie, Institut Gustave Roussy, Villejuif, France for the kind gift of a small sample of the T5 exonuclease purified from phage-infected cells. We also thank Dr. Hartmut Kratzin for amino acid sequencing, Dr. C. D. O’Connor for the pDOC55 expression plasmid, Anja Wendler for expert technical assistance and Drs. Hans-Heinrich Forster and Michael Zimmer for their help with amino acid sequence comparisons. We are also in-

by guest on May 31, 2018

http://ww

w.jbc.org/

Dow

nloaded from

18316 Properties of Cloned T5 D15 Exonucleme

debted to Drs. David Olsen and Gerald Gish for their helpful com- ments and suggestions.

REFERENCES

Bradford, M. M. (1976) Anal. Biochem. 72, 248-254 Brutlag, D., and Kornberg, A. (1972) J. Biol. Chem. 247, 241-248 Carrington, J. M., and Lunt, M. R. (1973) J. Gen. Virol. l&91-109 Chinnadurai, G., and McCorquodale, D. J. (1973) Proc. N&l. Acad.

Sci. U. S. A. 70,3502-3505 Chung, C. T., Niemela, S. L., and Miller, R. H. (1989) Proc. N&l.

Acad. Sci. U. S. A. 86,2172-2175 DeBoer, H. A., Comstock, L. J., and Vasser, M. (1983) Proc. Natl.

Acad. Sci. U. S. A. 80, 21-35 Devereux, J., Haeberli, P., and Smithies, 0. (1984) Nucleic Acids Res.

12,387-395 Ficht, T. A., and Moyer, R. W. (1980) J. Biol. Chem. 255, 7040-7048 Fraser, M. J. (1980) Methods Enzymol. 65, 255-263 Frenkel, G. D., and Richardson, C. C. (1971) J. Biol. Chem. 246,

4848-4852 Gram, H., and Riiger, W. (1985) EMBO J. 4,257-264 Hayward, G. S., and Smith, M. G. (1972) J. Mol. Biol. 63, 383-395 Henikoff, S. (1984) Gene (Am&) 28,351-359 Joannes, M., Saucier, J.-M., and Jacquemin-Sablon, A. (1985) Bio-

chemistry 24,8043-8049 Joyce, C. M., and Grindly, N. D. F. (1983) Proc. Natl. Acad. Sci. U.

S. A. 80,1830-1836 Kaliman, A. V., Krutilina, A. I., Kryukov, V. M., and Bayev, A. A.

(1986) FEBS Lett. 195, 61-64 Kelley, W. S., and Joyce, C. M. (1983) J. Mol. Biol. 164, 529-560 Kelly, R. B., Atkinson, M. R., Huberman, J. A., and Kornberg, A.

(1969) Nature 224, 495-501 Kerr, C., and Sadowski, P. D. (1972) J. Biol. Chem. 247,311-318 King, P. V., and Blakesley, R. V. (1986) Focus 8, l-3 Laemmli, U. K. (1970) Nature 227,680-685 Lanni, Y. T. (1969) J. Mol. Biol. 44, 173-183

Leavitt, M. C., and Ito, J. (1989) Proc. N&l. Acad. Sci. U. S. A. 86, 4465-4469

LeClerc, J. E., Istock, N. L., Saran, B. R., and Allan, R., Jr. (1984) J. Mol. Biol. 180, 217-237

Lipman, D. J., and Pearson, W. R. (1988) Proc. N&l. Acad. Sci. U. S. A. 85, 2444-2448

Ludwig, J., and Eckstein, F. (1989) J. Org. Chem. 54, 631-635 Lundquist, R. C., and Olivera, B. M. (1982) Cell 31,53-60 Maniatis, T., Fritch, E. F., and Sambrook, K. (1982) Molecular

Cloning: a Laboratory Manual, Cold Spring Harbor Laboratory, Cold Spring Harbor, NY

Moyer, R. W., and Rothe, C. T. (1977) J. Viral. 24, 177-193 Nakamaye, K. L., and Eckstein, F. (1986) Nucleic Acids Res. 14,

9679-9698 Needleman, S. B., and Wunsch, C. D. (1970) J. Mol. Biol. 48, 443-

453 O’Connor, C. D., and Timmis, K. N. (1987) J. Bacterial. 169,4457-

4462 Olsen, D. B., and Eckstein, F. (1990) Proc. Natl. Acad. Sci. U. S. A.

87,1451-1455 Ott, J., and Eckstein, F. (1987) Biochemistry 26, 8237-8241 Paul, A. V., and Lehman, I. R. (1966) J. Biol. Chem. 241,3441-3451 Prashad, N., and Hosoda, J. (1972) J. Mol. Biol. 70,617-635 Remaut, E., Stanssons, P., and Fiers, W. (1983) Nucleic Acids Res.

11,4677-4688 Rogers, S. G., and Rhoades, M. (1976) Proc. Natl. Acad. Sci. U. S. A.

73,1576-1580 Sayers, J. R., and Eckstein, F. (1989) in Protein Function: a Practical

Approach (Creighton, T. E., ed) pp. 279-295, IRL Press, Oxford Sayers, J. R., Schmidt, W., and Eckstein, F. (1988) Nucleic Acids Res.

16,791-802 Shortle, D., Grisafi, P., Benkovic, S. J., and Botstein, D. (1982) Proc.

Natl. Acad. Sci. U. S. A. 79, 1588-1592 Stone, A. B., and Burton, K. (1962) Biochem. J. 85, 600-606 Taylor, J. W., Schmidt, W., Cosstick, R., Okruszek, A., and Eckstein,

F. (1985a) Nucleic Acids Res. 13, 8749-8764 Taylor, J. W., Ott, J., and Eckstein, F. (1985b) Nucleic Acids Res.

13,8765-8685

by guest on May 31, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Properties of Cloned T5 015 Exonuclease

by guest on May 31, 2018

http://ww

w.jbc.org/

Dow

nloaded from

J R Sayers and F EcksteinEscherichia coli DNA polymerase I 5'-3' exonuclease.

Properties of overexpressed phage T5 D15 exonuclease. Similarities with

1990, 265:18311-18317.J. Biol. Chem. 

  http://www.jbc.org/content/265/30/18311Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/265/30/18311.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on May 31, 2018

http://ww

w.jbc.org/

Dow

nloaded from