Nucleotide sequence of the herpes simplex virus type 2 (HSV-2) thymidine kinase gene and predicted amino acid sequence of thymidine kinase polypeptide and its comparison with the HSV-1 thymidine kinase gene

158

Biochimica etBiophysicaActa, 741 (1983) 158-170 Elsevier

BBA91279

NUCLEOTIDE SEQUENCE OF THE HERPES SIMPLEX VIRUS TYPE 2 (HSV-2) THYMIDINE KINASE GENE AND PREDICTED AMINO ACID SEQUENCE OF THYMIDINE KINASE POLYPEPTIDE AND ITS COMPARISON WITH THE HSV-I THYMIDINE KINASE GENE SAUL KIT, MALON KIT, HAMIDA QAVI, DAVID TRKULA and HARUKI OTSUKA

Division of Biochemical Virology, Baylor College of Medicine, Houston, TX 77030 (U.S.A.) (Received April 19th, 1983) (Revised manuscript received June 15th, 1983)

Key words: Thymidine kinase gene; Nucleotide sequence," Amino acid sequence prediction; (Herpes simplex virus)

To analyze the boundaries of the functional coding region of the HSV-2(333) thymidine kinase gene (TK gene), deletion mutants of hybrid plasmid pMAR401 H2G, which contains the 17.5 kbp Bglll-G fragment of HSV-2 DNA, were prepared and tested for capacity to transform LM(TK-) cells to the thymidine kinase-positive phenotype. These studies showed that hybrid plasmids containing 2.2-2.4 kbp sublragments of HSV-2 Bglll-G DNA transformed LM(TK-) cells to the thymidine kinase-positive phenotype and suggested that the region critical for transformation might be less than 2 kbp. That the activity expressed in the transformants was HSV-2 thymidine kinase was shown by experiments with type-specific enzyme-inhibiting rabbit antisera and by disc-polyacrylamide gel electrophoresis analyses. DNA fragments of the HSV-2 TK gene were subcloned in phage M13mp9 and Ml3mp8. A sequence of 1656 bp containing the entire coding region of the TK gene and the flanking sequences was determined by the dideoxynucleotide chain termination method. Comparisons with the HSV-I(C! 101) TK gene revealed that Pstl, Pvull, and EcoRl cleavage sites had homologous locations as did promoter, translational start and stop, and polyadenylation signals. Extensive homology was observed in the nucleotide sequence preceding the ATG translational start signal and in portions of the coding region of the genes. Comparisons of the predicted amino acid sequences of the HSV-I and HSV-2 thymidine kinase polypeptides revealed that both were enriched in alanine, arginine, glycine, leucine, and proline residues and that clear, but interrupted homology existed within several regions of the polypeptide chains. Stretches of 15-30 amino acid residues were identical in conserved regions. The possibility is suggested that domains containing some of the conserved amino acid sequences might have a role in substrate binding and as major antigenic determinants.

Introduction The salvage pathway enzyme, thymidine kinase, is encoded by many herpesviruses [1]. These include herpes simplex virus type 1 (HSV-1), which Abbreviations: HSV, herpes simplex virus; TK gene, thymidine kinase gene; dThd, thymidine; BrdUrd, 5-bromodeoxyuridine; tet s, tetracycline sensitivity; amp r, ampicillin-resistant; OMK, owl-monkey kidney. 0167-4781/83/$03.00 © 1983 Elsevier Science Publishers B.V.

is usually transmitted nonvenereally and causes keratoconjunctivitis, gingivostomatitis, and recurrent herpes labialis, and HSV-2, which is most often transmitted venereally and causes primary and recurrent genitial herpes and generalized herpetic infections of neonates [2]. Herpesvirus thymidine kinases have an important role in the pathobiology of herpes infections [3-12] and they are primary targets for many of the nucleoside analogs, which are currently being used in the

159 treatment of herpes infections [13-16]. Herpesvirus thymidine kinases differ from the principal host cell thymidine kinase in that they catalyze the phosphorylation of deoxycytidine as well as thymidine. They also recognize nucleoside analogs as good substrates and selectively phosphorylate them [13,15]. Herpesvirus thymidine kinases have the ability to replace ATP with CTP as phosphate donor and they are less sensitive to feedback inhibition by dTTP than host cell thymidine kinases. On the other hand, herpesvirus thymidine kinases exhibit individual differences. The HSV-1 and HSV-2 thymidine kinases share cross-reacting antigenic determinants, but they can be distinguished from each other and from other viral or cellular thymidine kinases by using serially diluted, typespecific rabbit antisera. HSV-1 thymidine kinase is less thermolabile than HSV-2 thymidine kinase and has a greater affinity for arabinosylthymine, bromovinyldeoxyuridine, and acycloguanosine [13-19]. The latter difference partially explains the greater sensitivity of type 1 than type 2 herpesvirus replication to these nucleoside analogs. Certain HSV-1 mutants induce thymidine kinase activities with altered substrate specificities. The mutated enzymes phosphorylate dThd and its analogs, but lack deoxycytidine and acyclognanosine phosphorylating activity. Similarly, pseudorabies virus and bovine herpesvirus type 1 differ from HSV in that they induce thymidine kinase activities which lack deoxycytidine or acycloguanosine phosphorylating activities [20-24]. Fundamental to an understanding of the similarities and differences between HSV-1, HSV-2, and other herpesvirus-induced thymidine kinases is knowledge of the nucleotide and amino acid sequences of the genes and enzyme polypeptides. The HSV-1 TK gene has previously been cloned [25-27], translated in Escherichia coli [28], and sequenced [29,30]. We have now cloned DNA fragments containing the HSV-2 TK gene, analyzed the sequences required for biochemical transformation of LM(TK-) cells, and determined the nucleotide sequence of the HSV-2 TK gene. An amino acid sequence has been predicted from the nucleotide sequence. The data to be presented demonstrate that the HSV-1 and HSV-2 TK genes share considerable nucleotide sequence homology and indicate that amino acid sequences have been

conserved within several regions of the polypeptide chains. A preliminary report of this work was presented at the annual meeting of the American Society for Microbiology in New Orleans, March 6-11, 1983. Material and Methods

Cells. CV-1 and owl-monkey kidney (OMK) cells were grown in 8-oz prescription bottles in Eagle's minimal essential medium (Autopow, Flow Laboratories Inc., Rockville, MD) supplemented with 10% fetal calf serum. Bromodeoxyuridine (BrdUrd)-resistant mouse LM(TK-) cells [31] were grown in the same medium plus 25 #g/ml BrdUrd, except that BrdUrd was omitted from the medium in the passage immediately preceding an experiment. All biochemically transformed cell lines were grown in Eagle's minimal essential medium (Autopow) plus 10% donor calf serum and HATG (10 - 4 M hypoxanthine/10 -6 M aminopterin/4 • 10 -5 M thymidine/10 -5 M glycine). E. coli K-12 strain RR1 ( F - p r o leu thi lacY rpsL hsdR hsdM) was grown in medium 1 (10 g bactotryptone/5 g yeast extract/5 g NaC1 per liter) or in complete M9 medium (supplemented with 1.5% casamino acids/0.5% glucose/0.2% glycerol) and thiamine at 40 #g/ml [26-28]. Virus. The HSV-2(333) used in these experiments was obtained from William Rawls in 1973 and had been passaged four times in CV-1 cells, plaque-purified in OMK cells, and then passaged only five more times in CV-1 or OMK cells prior to the start of these experiments. Working pools of HSV-2(333) were prepared by infecting CV-1 or OMK cells at low multiplicity. Titrations were carried out in CV-1 or OMK monolayers with an agar overlay containing Eagle's minimal essential medium (Autopow) supplemented with 10% fetal calf serum and 0.04% protamine sulfate. Preparation of viral and plasmid DiVAs. Herpesvirus DNAs were isolated from infected cultures by the Triton X-100 method of Pignatti et al. [33] and purified by centrifugation in CsCI density gradients [26,32]. Plasmid DNAs were isolated from transformed bacteria and purified by sucrose gradient centrifugation [26]. Biochemical transformation experiments. LM(TK-) mouse fibroblast cells were biochemi-

160

cally transformed to the thymidine kinase-positive phenotype with either superhelical or restriction nuclease-cleaved plasmid DNAs plus carrier (5 /~g/dish) LM(TK-) cell DNA, using at least three Falcon tissue culture dishes (60 mm) per experimental group and at least two D N A concentrations per experiment [26-28,32]. Biochemically transformed cells were propagated in selective growth medium containing HATG. Cytosol extracts were prepared and assayed for thymidine kinase activity, using ATP-Mg 2÷ as the phosphate donor and [3 H]thymidine as the nucleoside acceptor. Serological studies on the capacity of antiHSV-2 IgG (No. 656) to inhibit the thymidine kinase activities of virus-infected and transformed cells were carried out as described [12,20,22,26-28]. Ligation of HSV-2 DNA fragments to plasmid pMAR401 DNA and isolation of a plasmid encoding the HSV-2 TK gene. HSV-2 DNA (16 /xg) was cleaved with BgllI, the DNA fragments were fractionated by centrifugation in a sucrose gradient (10-40%, w/v), and portions of each fraction were analyzed by agarose (1%) gel electrophoresis. Aliquots of the fractions with DNA fragments corresponding in size to HSV-2 BglII-G, which is known to encode the HSV-2 TK gene [34-37], were ligated to BgllI-cleaved plasmid pMAR401 (25 ng) [32]. The ligation products were used to transform E. coli K12 strain RR1 as described [26-28], and amp r, tet s colonies were isolated. Plasmid DNAs were prepared from the transformed bacteria, cleaved with BgllI, and analyzed by agarose gel electrophoresis. A candidate plasmid, designated pMAR401-H2G, contained a 17.5 kbp BgllI fragment, corresponding to the HSV-2 BgllI-G fragment. The restriction map shown in Fig. la was constructed from analyses of single and double restriction nuclease digests of pMAR401-H2G. A series of deletion mutants were derived from plasmid pMAR401-H2G by standard procedures [12,26-28,32], as summarized in Table I. Restriction maps of the derivative plasmids pDT4, pCT4, and pDM33, which efficiently transform LM(TK-) cells to the thymidine kinase-positive phenotype, are depicted in Fig. la and b. Subcloning of HSV-2 TK gene fragments in M13 vectors. Fragments of the HSV-2 TK gene derived from pDT4, pDM33, pCT4, and pCT415 were

subcloned in the replicative forms (RF) of phage M13mp8 and M13mp9 DNA [39-41]. This method permits the cloning of the same restriction fragment in both possible orientations. DNA sequence analyses. Single-stranded phage DNA was prepared from mixtures of phage-infected E. coli K-12 JM103 and freshly growing uninfected bacteria [39-41]. DNA sequencing was carried out by the dideoxynucleotide chain termination method [45], using single-stranded M13mp9 (or M13mp8) subclones of HSV-2 TK gene DNA as templates, the 17mer-synthetic primer (Collaborative Research, Lexington, MA), either [a- 32P]dATP or [ a- 32P]dTTP as the labeled substrate, Mg 2+, the appropriate unlabeled dNTPs and dideoxynucleoside triphosphates, and E. coil DNA polymerase (Klenow fragment, Bethesda Research Lab, Gaithersburg, MD). Reaction products were precipitated with ethanol, dissolved in 10 #1 of 90% formamide/30 mM N a O H / 1 0 mM EDTA/0.3% Bromophenol blue/0.3% xylene cyanol, heated for 1 min at 90°C, and loaded onto standard 8% sequencing gels [39,45]. Results

Biochemical transformation of LM(TK ) cells by hybrid plasmids The HSV-2 TK gene is contained within a 4 kbp BgllI/HindlII fragment, which maps at about 0.284-0.314 on the HSV-2 genome [34-37,42,4648]. The HSV-2 and HSV-1 TK genes are colinear. The polarity of transcription is from 0.314 to 0.284 map units [44]. To delimit more precisely the boundaries of the HSV-2 DNA fragment required for the functional expression of the TK gene in the biochemical transformation assay, the 17.5 kbp BglII-G fragment of HSV-2 DNA was initially cloned at the unique BgllI site of plasmid pMAR401 [32], recombinant plasmids containing deletions in the BglII-G fragment of HSV-2 DNA were prepared (Table I), and transfection experiments were carried out in LM(TK-) cells. As expected, hybrid plasmid pMAR401-H2G (Fig. la) efficiently transformed LM(TK-) cells to the thymidine kinase-positive phenotype, verifying that HSV-2 BglII-G encoded TK gene. Likewise, plasmid pDT4, which was derived from pMAR401-H2G by deleting the pMAR401-H2G

161

TABLE I DERIVATION OF PLASMIDS CONTAINING HSV-2 THYMIDINE KINASE GENE FRAGMENTS +, yes; - , no. Name of plasmid

Transforms LM(TK-) cells to thymidine kinasepositive phenotype

pMAR401

pMAR401-H2G

+

pDT4 and pDT3 pMK25 and pMK38

+ -

pMK30 pCT4

+

pCT45 pCT4 dl Eco pCT415 pDM33

+ + +

pDM330 pDM33 dlKpn

Description of plasmids

Deletion mutant of pMAR4 lacking 2.6 kbp KpnI fragment, pMAR4 was obtained by insertion of a 9 kbp HindIII fragment of Herpesvirus tamarinus (MarHV) into the HindIII site of pBR322, pMAR4 transforms LM(TK-) cells to the thymidine kinase-positivephenotype, but pMAR401 does not. 17.5 kbp BglII-G fragment of HSV-2(333) inserted at unique BgllI site of pMAR401 Delete 18.5 kbp BarnHI sequences from pMAR401-H2G (map units 3.8-22.3) Similar to pDT4 except that BamHI sequence of pDT4 (map units 6.0-9.3) is inverted in pMK25 and pMK38 BamHI sequence of pDT4 (map units 6.0-9.3) inserted at the BamHI site of pBR322 BglII/XhoI fragment of pDT4 (map units 5.5-8.7) ligated to the large (4.1 kbp) BamHI/SalI fragment of pBR322 Delete SmaI sequences (map units 1.5-3.55) from pCT4 Delete EcoRI fragment (about 0.5 kbp) from pCT4 Delete Sinai sequences (map traits 2.6-3.55) from pCT4 Insert 2.5 kbp ClaI fragment of pCT4 (map units 0.5-3.0) at the ClaI site of pBR322. Fig. lb shows a single KpnI site at 0.9 map units; later studies revealed that there were two KpnI sites, 213 bases apart. Delete 1.8 kbp EcoRV fragment from pDM33 Delete 0.2 kbp KpnI fragment from pDM33

sequences mapping from 3.8 to 22.4, efficiently transformed LM(TK-) cells, but plasmid pMK30, which lacks the BglII/BamHI sequence of pMAR401-H2G mapping at 0.0-0.5 kbp, and plasmids pMK25 and pMK38, in which the BamHI fragment is inverted relative to the same fragment of pDT4 (6.0-9.3 map units) did not transform LM(TK-) cells (Tables I and II). Plasmids pCT4 and pDM33 (Fig. lb), and plasmid pCT415, which is derived from pCT4, transformed LM(TK-) cells, but plasmids pCT45 and pDM330 did not, showing that the sequence required for functional expression of HSV-2 TK gene mapped from about the ClaI to SmaI sites (0.5-2.6 map units) on pCT415. Cleavage of pDT4 with EcoRI reduced, but did not eliminate transforming activity (Table II). The nucleotide sequence analyses to be described later demonstrated that the EcoRI site was 62 bp 3' to the ClaI site and was part of a 'CAAT' transcriptional regulatory sequence [29,30,49-51].

To evaluate further the significance of the sequences 5' to the EcoRI site, a deletion mutant, designated pCT4 dlEco, was prepared. Plasmid pCT4 dlEco lacks the EcoRI fragment (0.0-0.5 map units) of pCT4 (Table I, Fig. 1). Plasmid pCT4 dlEco transformed LM(TK-) cells to the thymidine kinase-positive phenotype, but with low efficiency (Table II), consistent with the hypothesis that sequences near the EcoRI site of HSV TK genes may contain transcriptional control signals [43,50]. The nucleotide sequence analyses predicted that there were two KpnI sites separated by 213 bp close to the PstI-SmaI cleavage sites of pDM33 (Fig. 1). This was verified by molecular hybridization experiments in which a 32p-labeled by nick-translation pDM33 probe was hybridized to KpnI-cleaved HSV-2(333) D N A and to KpnI-cleaved pCT4 DNA. The 32P-labeled probe hybridized to a small 213 bp KpnI fragment derived from the HSV-2 and pCT4 DNAs, as ex-

162 B(~HI

KPnl SmQI

HindIII

/

EcoRI Pvull ClOl

7.~ 6.6

HindlIC~,.

~PIstl \| L

Pvull BOtnHi l#"

5.8

f.v ~Am+o]-H,20 Kpnl.._~1.0 ~omHi . . ~ Og5,

£EcoRI~ Io1 ' . . . ~ (

~

s -2 ~ (,27.8

~'~ ~,

....

t~. 16.1 17.5

kb)

R

1~.~\ 17.8"~l

sol 1 ~ , . 2 ^ .

,

~o'., \ \

".\ " ~

Hint]i ~ . 2

"

. . . . .

I

/

~,,JI °° '~

/ /

5oli

Pst~ ~

EcoRIp, '~I~'" " ~ , J N ~ " ~

Orl

Pvul

Socl

Pvut[HI)el

I

~x+.... p°°, ]' ~' 'BomHl HincIIII

Sect

BglIJ/Bo~!

EcORI

Cl01,

HI.Ill PstI,PvuII,Clol, EcoRI EcOR~ CloI,EcoRI

°"IL;

(

)

,,,

J

3~4 BomHIECORV Clol, H|ndlll

Mru I Fig. 1. Restriction maps of: (a) recombinant plasmids pMAR401-H2G and pDT4; and (b) plasmids pCT4 and pDM33. Curved bars, solid lines and wavy lines denote pBR322, HSV-2 DNA, and marmoset herpesvirus DNA sequences, respectively. Procedures used to derive smaller plasmids from pMAR401-H2G are summarized in Table I. pMAR401-H2G contains a 17.5 kbp HSV-2 BglII-G fragment inserted at the unique Bg/lI site of pMAR401. The ampicillin-resistant (ampr) gene of pBR322 is shown, as is the approximate location of the HSV TK gene. The polarity of transcription of the HSV TK gene is clockwise for each of the plasmids [42-44].

163 TABLE II BIOCHEMICAL TRANSFORMATION OF LM(TK-) CELLS BY RECOMBINANT PLASMIDS CONTAINING HERPESVIRUS THYMIDINE KINASE GENES Plasmid pAGO contains a 2 kb PvulI fragment which encodes the HSV-1 TK gene [25,26] and was included as a positive control in these experiments. As a negative control, LM(TK-) cells (5-105 cells/dish) were incubated overnight at 37°C and were transfected with LM(TK-) DNA (5 #g/dish), but no plasmid. Thymidine kinase-positive transformants were not obtained. Exp. No.

Source of DNA

1

pMAR401 H2 G (HindlII cut) pAG0 (PoulI cut)

40 20

211 37

2

pDT4 ( EcoRI cu0 pDT4 (BgllI + ClaI cut) pDT4 (EcoRI + ClaI cut) pDT4 (PstI cut) pDT4 (EcoRI+ KpnI cut) pDT4 (Bg/II + KpnI cut) pAG0 ( PoulI cut)

33 33 33 33 33 33 20

18 54 7 0 0 0 34

3

pMK25 (superhelical) pMK30 (superhelical) pAG0 ( Pvull cut)

25 or 100 25 or 100 25

0 0 91

4

pAG0 (PvulI cut) pDM33 (superhelical) pDM33 (HinclI cut) pDM330 (superhelical) pCT4 (superhelical) pCT4 (superhelical) pCT45 (superhelical) pCT415 (superhelical) pCT415 (superhelical)

25 63 63 63 or 125 125 63 63 or 125 63 125

83 259 270 0 196 > 300 0 62 144

5

pDM33 (HinclI cut) pCT4 dl Eco (superhelical) pDM33 dl Kpn (superhelical)

250 250 63,125, or 250

pected (data not shown). Two KpnI sites close to the PstI-SmaI cleavage sites of HSV-2 T K gene have previously been described [36,47]. A KpnI deletion mutant, designated p D M 3 3 dlKpn, was prepared. This m u t a n t failed to biochemically transform L M ( T K - ) cells, confirming the data of Table II, Exp. 2, which suggested that sequences near the KpnI site were essential for transforming activity [47].

Characterization of the thymidine kinase activity expressed in cells transformed by recombinant plasmids To verify that the thymidine kinase activity expressed in cells biochemically transformed by

ng DNA/dish

Average number transformed colonies per dish

174 3 0

recombinant plasmids p C T 4 and p D T 4 was HSV2-specific, enzyme inhibition studies with antiHSV-2 thymidine kinase rabbit antisera (diluted 1 : 1 2 ) were carried out [20]. These experiments showed that the thymidine kinase activity in extracts of the cells transformed b y p C T 4 and p D T 4 was inhibited 76-95% b y rabbit sera No. 656, which inhibits the enzyme activity of HSV-2-infected L M ( T K - ) cells b y about 90% and the thymidine kinase activity of HSV-l-infected cells b y only 20%. Sera No. 656 does not inhibit the thymidine kinase activity of uninfected m a m malian cells (data not shown). I n other experiments, D N A was isolated from cells transformed by p C T 4 dlEco, digested with

164

b Q e~4

T I ~ V ~ I - A T ] O N OF ~ V 2 T K A 3 ~C~BT~ROCT T C ~ ( ~ T i B ~ P ( K : I ~ C

TYR LEU NET m . Y g(ER MET THR PRO ~ . N T~C CTC ~ O ( m ~ ~VaC ~TO ACC c c c C ~ TSC T T C ~ T B ~ C C ~ T ~ T C C I C ¥ 8 T T C ( I ~ I T T T ~ T a ~ C C S T B T C

C

¢¢A V~L LEU ¢¢A P~ v~L ~t_A LEU PIE~r P~O P ~ ~CC GT(~ T T S GR:O ' r T c e~TO OCC CTC ~ T ~ CCC CCO

~ ¢¢24

Tt4R A L ~ PRO I~..y T H R ABN LEU V~IL L E U ~ L Y V A L L E U PRO ~LU ~I,.A BILl,/ HIS ~I-¢I AISp k R U CA(: I~:c ~ COl: ACC ~ c s CCC ~ ~C(~ ~ C T ~ ~ T C CTB S ~ T eT(~ C T T CCI~ alA6 ~ . c ~

(~ AT

TTT~CATQTCTTTIq~TCC~q~I~:IT~(~C~P~CTC~4BTTTTI~BTTEI~ICII~J~TTC~T C

~B4 ~C~MB'rC~IICOBCCCEIA~TC~TTCeCAT~TT~MBTII~C~C~ITIIICCT~T~C4k T

CTG ~ C

~

C~C C~,P~ CSC (:CB

B L U ~RB L E U A g P L E U A L A MET LEU ~ R I~.A ILE C e 6 C T ~ e ~ C ¢ T O QC(~ ~TO CTO TCC ~ICC a T T C~C

1044

~ V ~ L T y R A~P L £ U L E U ~ L A A~N T H R V N - ¢d~(~ T Y R t.EU m-N A~qG O*.V m.V AR~ T~P A~S #/CO (ITS ~ TAC CT~I CWS C ~ {SOC ~ A6S T e ~ C~T ~TC T~J[: IB~T CTA CTC []M~C

243

~C~TCATP,:~J~IT~TTCC~C~A~TC~T(~IT~SC~TT~q~CTCCC~[~CCTCTCOB~CIB~CTT~T 323

~q~C~T

~I*A ~ CaT ~ (

MET R~-A ~ H I S N*.A S L Y B L N EIt.N H I E N - A P R O t ~ . A P H I I L Y ~I¢.N N ~ A N..A R T E E ~ T T C T CA(:; BCC QeC C4tA C ~ ~ eC(ll CCT ~ T T C ~ S T CP~ OCT ~ T lILY P%'~OTI4R ~ ~ CCT RCC K

II-V q ~LA A L A II~R H PR¢I ~I~R H i l l ~ ~ ~ C C(W: ~ GCS "P~C COT CCT ~ CAT C~C C ~

SLY ~M.A ~ ~

~L.A w,.¥ ~ R LEU PRO ARG ILE SLU ~ s = T H e L E O PHE A L A L E U PH~ ~RG VHt. PRO tZLU u E u ~CG ~ T C T CTS CCC CBC ATC ~ i ~ f~Ce C T G T T T BCC CTB T T C CGC S I T C(:C ~ CTS

~.V ~

LIEU ALA PRO ASN ~IL.y ASP L E U T Y R H 1 6 I L E pHIE A L A T R P VAL L E U ~SP ViCA_ L E U ~ - A ~ P OTC TTI3 GCC 6k4V~ CTG B4CC SCC ~AC GG(~ SAC T T S TAC CAC R T T T T T (]CC TGG GTC T T G ~

444

|~4

m.U i~.A ~ E4.V A ~ ~ ~ L U LI[U PRO THI~ L E U LIEU ~ V~tl- TYR I L K ~ TC(: ~ E~C CE~ ~ E ~ T CCG ~ C T 8 CCC ~ C T 6 CTQ COB ~ T T T A T A T A ~

~ CCC

~04

t21B4

H I E E L y V A L E L Y L Y S Tt4R TMR T H R ~ R (~L.A I~-kl L E U N E T ~K.U ~(..A L.EU S L Y PRO ~ q ~ ~ P ~ STIS ~ ~ ACC ~ C ~,CC TCC ~CO ~ CTIB ~T[~ ~ ~ C CTEI ~ ~ ~ ~ 564 ~T

1104 I~LU A m > T R P S L Y ARB L E U THF~ SLY vmw. A L A A L ~ At_A THI~ ARG P~O ~BP PIIO Ot.u ASP e~_v ~ T C s c c BCB O c ~ ~ c e C¢C CCC ~ CCC ( ~ t ~ GPMS ~ T84~l ~ C~I~ C T 8 P~C~

] L E V ~ L TyF( V~.. P~S ~BLU P~O MET THIR T y R T R P ~..N VAt. Lieu l I L Y ~ - A A~I'C ~11"c TA~ ETC ( ~ ~ CC~ R T 6 ~CT T~C T I ~ ~ GT9 c'r9 ~ ~

g ~ R (~.U TPIR T(~c ~ ~

~24

1344 ~ ~kt-A L E U L E U ~ i..~U T P ~ g~._A S L Y MET I L E PRO T H R A ~ VAL T H ~ T H R ~q_h SLY ~ER (~ (]cc CTG T T G C~C CTC ~ C S(~C (~SS A T S ATC CCA ACC CSC ~XC ~CA ACC QCC ~ e ~ TCC 1.04

L £ U THR ~ I L ~ T Y R ASN T~4R B L N H I S ~ R ~ L E U ~ P C T e ~C~ ¢ ~ C A T e X~C ~ ~CS C ~ C~C COT C T e ~ C

~ ~ L Y ~t_U I L E ~ R A L A S L Y SLU CeC ~ C ~ A T A TCS ~ C 0 ~

684

(~1(~6 ~K~G 13TS 8 T H A T S ~4~C ~

~ LIEU I-EU PR~ I ' ~ T H I S L E U PHIE V A L L E U ASP T y R ASP e L N SER pRO VAt. ~ L Y CVS ~ R ~ C~IC CTC C T T CCG ATG CAT C T A T T T GTC CTS ~ T TAC GAT C/US TCB CCC ~ C ~ TaT cs~

[ L E ~t_A 6 L U I L E ARG ~SP L E U ~ . A ~R(3 T H ~ F'~E H L A RRS ~ U V ~ ~ L Y ~ y Qcc ~o A T A C a C O A ¢ C T G GCO C a C ACG T T X ~ C C CGC aHG ~ T G ~ Q a ~

arc

V~ eve aTX Z~

TT~

t~b4

~I~C CAG A T H ~ A

ATe ~C

ACE CCT T H T ~ICB ~

~

~C

744 ~ L A V ~ L L E U ~J-A PRO H I S I L E GI.y S L Y ~,LU A L A VAL ~ L Y PRO ~ L N ALA PRO PRO ~ ALA ~A~C (~TT T T S ~ T CCT CAT ATC ~ (~ ~ SCT QTS 13~C CCS C ~ q BCC CCG CCC CCG SO4 L E U THR L E U VP~L pHE ASP A~I~ H I S PRO I L E A L A E~ER LEU LiEu C y s T y R pRO ~4_A ~ L A ARG CTC ~CC CT~" OT T T T C SAC Ca(] C~.C CCT A T e ~ C I"CC CTG CTG TmC TAC CCG ( ~ C ~C(] C ~

A~C~4~AAGCCC~A~GGA~(3GCCTcCCGGC~'~T~BG~AT~'~d~A~C~`(s~T~=~4~GBCATTGTTGTCBTGTeG p 1544 T~T~TCcA~A~icG~B6QGGTTCQGG0CCA~cTGGCACC~TAT~A~ACccCAC~A~AAACG~GC~GCCBAT~CG 1624 ~CCtXBTTtTC~CXCTGTnCtCetTSTBCtTT X~

t~OLECtJLAR ~ l ~ M /

O1: T H I S PROTEIN IB

40~1S.3

Fig. 2. Nucleotide sequence of the HSV-2(333) TK gene and its flanking regions. The complement of the transcribed strand is shown. Locations of the putative 'CAP=T, Goldberg-Hogness, and polyadenylation signals are indicated by 'C', 'T', and 'P', respectively. The predicted amino acid sequence of the TK gene polypeptide is also given. The sequence begins at the third nucleotide (G) of a PstI site

(CTGCAG).

either BamHI, BgllI, or EcoRI, electrophoresed in agarose gels, and hybridized to a 32p-labeled by nick-translation pCT4 dlEco probe. The results indicated that pCT4 dlEco DNA was integrated in high molecular weight fragments of the biochemically transformed ceils at three to five sites (data not shown).

DNA sequence of HSV-2(333) TK gene The nucleotide sequence of a 1656 bp region of HSV-2 DNA that contains the entire TK gene is shown in Fig. 2. The sequence was determined by the dideoxynueleotide method of Sanger [45], utilizing the strategy shown in Fig. 3. For more than 90% of the gene, the sequence was confirmed by analyses of both complementary strands. The sites for restriction enzyme cleavage which were located by standard mapping techniques (Fig. 1) are also predicted by the sequence. Additional restriction nuclease sites predicted from the sequence are summarized in Table III. Fig. 3 shows that PstI, PoulI, and EcoRI cleavage sites on the

HSV-2(333) TK gene have homologous locations on the HSV-I(C1 101) TK gene [29,47]. Comparison of the HSV-2(333) TK gene and the HSV-I(C1 101) TK gene sequences also revealed that both contained at homologous locations Goldberg-Hogness (TATTAA) and 'CAAT' (CATTGGCGAATTC) sequences similar to the conserved sequences observed at specific distances outside the transcribed regions of many other eukaryotic and viral genes. Evidence that these sequences function as HSV promoters has been presented by McKnight [43,50] and Read and Summers [51]. The 5' end of the HSV-I(C1 101) TK gene mRNA had previously been located 31 nucleotides downstream from the 'TATTA~ k' sequence and 85 nucleotides from the 'CAAF' sequence [29]. Kozak [52] has proposed that eukaryotic ribosomes usually initiate protein synthesis at the AUG closest to the 5' end of an mRNA and that ~XXAUGG is a favored sequence for eukaryotic initiation sites. A sequence with these characteris-

165 TABLE IIl RESTRICTION ENDONUCLEASE CLEAVAGE SITES PREDICTED FROM THE NUCLEOTIDE SEQUENCE

. . .6oo o. . .~oo. . ,oo 12

=

, ~ o ,o~ o ~,~ ~ o

-

2 4 3

d

15

~ bs

-

',leo (bp)

21

~19 24

27

26

2O 7

..................

~

5

6

' ',,'0o' ,,o 'o ,2oo

' 17

i

22 25 2"~ L L

........................................

~

f8 ~

•

29

9

..................................

Restriction nucleas¢ sites shown in the HSV-2 TK gene sequence of Fig. 3 arc: PstI (1, 238, 1080), EcoRI (145), PvuIl (25, 534), ClaI (83), BamHI (458), EcoRV (668), KpnI (862, 1075), Ball (985, 1021), Sinai (1001), and Stul (1485). There are no cleavage sites predicted from the nucleotide sequences for the following enzymes:

8 28

Fig. 3. Strategy for DNA sequence determination of HSV-2(333) TK gene. Arrows indicate the fragments subcloned in phage M13mp9 or M13mp8 [39-41]. Numbers above each arrow designate specific subfragments. The length of the arrow (solid plus dotted line) shows the position of the subfragment on the HSV-2 TK gen¢. The solid line indicates the portion of the fragment (5' to 3') actually sequenced, DNA fragments to be used for subcloning were isolated by standard procedures. For example, DNA fragment No. 1 is the EcoRl/BamHl fragment isolated from plasmid pDT4 and subcloned in phage M13mp9. DNA fragments No. 4 and 11-16 axe Mspl fragments derived from the 0.9 kb EcoRi/KpnI fragment of pDM33 and subcloned in M13mp9. DNA fragments No. 17, 19, 21, and 27 are SaulIIA fragments subcloned in M13mp9. Fragments No. 18, 20, 23, 24, 25, and 26 are Mspl fragments derived from the 0.9 kbp Kpnl/SstI (SacI) fragment of pDM33 and subcloned in M13mp9. Fragments No. 13 and 22 are the complements of fragments No. 12 and 21, respectively, and were subcloned in M13mp8. The restriction map of the HSV-I(C1 101) TK gene is shown at the top of the figure to emphasize homologies between the HSV-1 and HSV-2 TK genes [29]. The physical maps of the HSV-1 and HSV-2 TK genes are shown in the I L orientation.

tics occurs at n u c l e o t i d e 333 o f the HSV-2(333) T K gen¢ at a p o s i t i o n h o m o l o g o u s to the p u t a t i v e A U G t r a n s l a t i o n a l start sequence o f the HSV-I(C1

101) T K gene [29] (Figs. 2 and 3). From the A T G at nucleotide 333 of the HSV-2 T K gene, there is an open reading frame of 1125 nucleotides followed.by a T A G terminator codon [53]. Forty-eight nucleotides beyond the T A G is the polyadenylation signal, AATAAAA, which is repeated again seven n u c l e o t i d e s later. T e r m i n a t o r and AATAAAA polyadenylation signals are located at homologous positions on the HSV-I(C1 101) T K gene [29] (Fig. 3). The predicted amino acid sequence of the HSV-2 thymidine kinase polypeptide is shown in Fig. 2. The open sequence can code for a protein containing 375 amino acids with

AvrlI BstelIC DdelD HgiaIA HincIIC HinfIC Pvu I XhoI

Bcll BstelID HaelB HgiaIB HincIID HinflD Sac I

BglII DdelA HaeIIA HgiaIC HindIII HpaI Sal I

BsteIIA

BsteIIB

DdelB HaeIIC

DdeIC

XhoIIB

XhoIIC

XhoIID

HaeIID HincIIA HincIIB HinfIA HinfIB MboIIB Mstl Sph I Xba I

Endonucleas¢

Location of first nuclcotide in sequence (nuclcotid¢ number)

AluI AvaIA AvaIB AvalC

8, 26, 101

Bgll

675, 744, 891 105, 434, 887, 1571 279, 548, 941, 947, 983, 1019, 1258, 1310, 1421 64, 179, 217, 348, 394, 405, 440, 545, 553, 611, 674, 780, 800, 854, 890, 908, 965, 986, 1022, 1113, 1226, 1277, 1486, 1569, 1614 22, 175, 211, 360, 409, 531, 557, 703, 858, 927, 1085, 1142, 1 162, 1355, 1425, 1427, 1555 74, 346, 462, 676, 798, 821, 852, 931, 960, 1002, 1115, 1366, 1396, 1492 459, 1053, 1320, 1373, 1618 452, 659, 1132 84, 149, 221, 817, 1183, 1340

EcoRIIA EcoRlIB HaelII

HhaI

Mspl SauIIIA SaclI TaqI

1001

181, 579, 1 152, 1215 305

a m o l e c u l a r weight o f 40 318. This value c o m p a r e s f a v o r a b l y with previous estimates b a s e d o n glycerol gradient centrifugation experiments and sodium d o d e c y l s u l f a t e - p o l y a c r y l a m i d e gel electrophoresis analyses o f l a b e l e d p o l y p e p t i d e s , which suggested that HSV-1 a n d HSV-2 t h y m i d i n e kinases a r e c o m p o s e d o f 40 k D a p o l y p e p t i d e subunits [54,55]. T h e m o l e c u l a r weight o f the HSV(C1 101) thymid i n e k i n a s e p o l y p e p t i d e p r e d i c t e d f r o m the n u c l e o t i d e sequence b y W a g n e r et al. [29] was 40 894.

166 A COMPARISON OF HSV2TKA AND HSVITK3 THE UPPER SEQUENCE IS HSVITK3 THE LOWER SEQUENCE IS HSV2TKA

192 202 212 222 232 242 252 262 CTGC~BS~SCTTC~TGSCSCASCTECTTCATCCCCGTSGCCCGTTGCTCSCGTTTGCTG~CGGTGTCCCC(38~qG CTGC~CTTCAGGGASTB(3CGC~CTGCTTCATGCCCGTGGTCCGCTGTTCGCGTTTGCTGGCCBTGTCCCC(N3A~

10

20

30

40

50

bO

70

BO

272 282 292 302 512 322 332 342 A~ATATATTT8C~TGTCTTTAGTTCT~T~T~C~C~ACCCC~CCC~GC~TCTTGTCATT~C8AATTC8AACACGCAG A~TCGATTTGCATSTCTTTAECTCCASGATS~CSCACACACCTCCCAACGTTTTGTCRTTGGCGRATTCSAACACBCAG 90 IOO 110 120 130 140 150 l&O 3~2 362 372 382 392 402 412 422 AT(3C~TCaG~(3C~8C~C~GTCCC~GTCCACTTC~CATATTARGGTG~C8CGTGT8GCCTCG~AC~CC~AGCGA~CCTG AT~CA~T~T~8C88C8CGGC~CG~GGTCC~CTT~GC~TATT~G~T~AC(3CGCET~CCTC8~CA~CGA~CGACCCTG 170 180 190 2OO 210 220 230 240 432 442 CAGCSACCCGCTTAACRGCGTCAAC CAGCGACCCSCTCRTCA~CGTCAGA 250 260

457 467 477 487 497 507 517 527 AGC~TG~C~C~TCTT8GT~C~T~3~d~CTCCC~C~CCTCTTT~CAA~CGCCTTGT~Gk~C~CGTAT~CTTCGT~C AGCGTT~ACA~TCCT~3TG~CGTT(3~WiCT~CC~C~CTCTC~8C8~AC~CCTT~TA8A~GC~GGT~T~CTTCTCAC 274 284 294 304 314 324 334 344 537 547 557 567 577 ~87 597 b07 CCCTGCCATC~C~C~C8TCT8CGTT~Ek~CC~8~CT8C8C~TTCTC~C~8CC~T~(3CA~CC8AC6T~CG~CGTT8C~CCC GCC~CC~C~C~C8C8CCT8C~TTC8~TC~CT8CTC~T8CG~8C8~CCT~CC8ACG~CC~C~C8~CGTCCC~TCC 354 364 374 384 394 404 414 424 bit b27 b37 b47 &57 667 &77 TCSCCGSCAGC~qG~N~CCACG(3~STCCGCCT~ARATGCCCACGCTACTSCG~GTTTATATASACGGT TP~3CCATCGCC~QCCTCCE~CCCQC~TCCGGAGCTGCCC.qCGCTGCTGCGGBTTTATAT~AC~(3A 434 444 454 464 474 484 494

b89 b99 709 719 729 739 749 C(3GTCCTCAC~38~TBGSG~CC~CC~CCACSCAACTSCTBGTGGCCCTGGGTTCGCSCEACGRTATC8

509

519

529

539

549

559

569

745 755 765 775 785 795 805 815 CGCGCGAC~TATCSTCTRCSTACCCBAQCC~T~TT~CTGGCABGTGCTGGQGGCTTCCGABACAATCGCG~ACATC

565

575

585

595

605

615

625

b~5

825 B3~ 845 Q55 8~5 875 885 895 TAC~CCRC~i~ACCBCCTCBA~CA~T~4iGATATCBG~CGGG8A~E~csBCBGTGGTAATGACAAGCGCCCAGATAAC Ti{~[:~i~AC~C~A~C~TCT~q~C~::~(~A(3~T~TCG~C~8~C~GCG~T~TRAT~C~AGCGCCC~GATA~C 645 b55 665 ~75 b~ ~95 705 7i5 905 915 925 935 ~45 955 ~b5 ~75 ~T~RT~CCTTAT~CCGT~i~CG~K:;Q~cGTTCTTGCTCCTCATGTCGGGGGGG~GGCTGGGAGTTCACAT~CCCCGC ~T~A~CTTAT~C~GC:G~iC~qCQ~GTTTT~3~TCCT~ATRTCGGGGG~GAG~CTGTG~GCCC~A~CCCC~ 725 735 745 755 7~5 775 785 79~

Fig. 4. Comparison of HSV-I(C1 101) and HSV-2(333) T K gene nudeotide sequences on the noncoding strands. The top sequence is that of the HSV-I(C1 101) TK gene, with the nucleofides numbered from a PstI site, as in reference [29]. The bottom sequence in each p ~ r is that of the HSV-2(333) TK gene (Figs. 2 and 3).

167

Comparison of nucleotide sequences of HSV.I and HSV-2 TK genes Figure 4 shows that nucleotides 1-263, which are at the 5' end of the HSV-2(333) TK gene and include the 'CAAT' and 'Goldberg-Hogness' promoter signals, have over 91% sequence homology to the corresponding promoter region of the HSVI(Cll01) TK gene. The data confirm the observations of Reyes et al. [47], who reported that 161 of 175 nucleotides at the 5' end of the HSV-2(333) TK gene (nucleotides 71-246 of Fig. 2) were completely homologous in position and sequence with those in the HSV-1 (MP) TK gene. Figure 4 also shows that strong sequence homology exists in the coding regions of the HSV-I(C1 101) and HSV2(333) TK genes• HSV-2(333) TK' gene sequences from 469-499 and 556-795, respectively, exhibited 97 and 89% homology to HSV-I(CI 101) TK gene (Fig. 4). HSV-2(333) TK gene nucleotides from 265-382 and 521-569 (Fig. 4), and from 790-1000, 1000-1114, and 1182-1300, respectively, exhibited 83, 83, 85, 85, and 90% homology to HSV-I(CI

101) TK gene (data not shown). Throughout the coding regions of these two genes, homologous stretches of 10-20 nucleotides were common.

Comparison of the predicted amino acid sequences o] HS V-1 and HSV-2 thymidine kinase polypeptides Wagner et al. [29] have reported that the amino acid composition of the HSV-I(CI 101) thymidine kinase polypeptide, as predicted from the nucleotide sequence, is unusual, being high in alanine, arginine, glycine, and proline. The sequence data shown in Fig. 2 likewise predict that the HSV2(333) thymidine kinase polypeptide is unusually high in alanine, arginine, glycine, and proline. Out of 375 amino acids, the HSV-2(333) thymidine kinase polypeptide had 52, 33, 32, and 31 alanine, arginine, glycine, and proline residues, respectively. These four amino acids can be specified by codons containing only guanine and cytosine, suggesting that the choices of these codons and amino acids may represent adaptations to the high G + C content (67-69%) of HSV TK gene DNAs. Two

J

~.°

a s

f

l^a

J

J

a.

f

N~

f

"-

r~.

'51

'181

'l~i

'2~i

'2~I

'~i

"3'

Fig. 5. Representation of amino acid sequencehomologyfor HSV-I(CI 101) and the HSV-2(333)thymidine kinase polypeptides predicted from the nucleotidesequences. The predicted amino acid sequenceof the HSV-I(C1101) thymidinekinase polypeptideis taken from the data of Wagneret al. [29]. The amino acid residuesare numbered sequentiallyfor the HSV-1and HSV-2thymidine kinase polypeptideson the abcissaand ordinate, respectively.Each dot represents a match of three consecutiveamino acids.

168 TABLE IV PREDICTED AMINO ACID COMPOSITION OF HSV-2 THYMIDINE KINASE POLYPEPTIDE Amino acid residueNo, of residues per No. of residues per HSV-2 thymidine kinaseHSV-1 dThd kinase molecule molecule [29] Ala Arg Ash Asp Cys Gin Glu Gly His Ile Leu Lys Met Phe Pro Set Thr Trp Tyr Val

52 33 6 21 2 13 17 32 10 13 43 1 11 8 31 16 26 4 11 25

46 28 8 19 4 19 14 30 12 16 41 4 13 8 30 17 28 5 12 22

cysteine residues were predicted for the HSV-2 thymidine kinase polypeptide compared with four for the HSV-1 thymidine kinase polypeptide (Table IV). It is noteworthy that only one lysine residue (codons AAA and AAG) was predicted for the HSV-2(333) thymidine kinase polypeptide (Fig. 2) compared with four for the HSV-I(C1 101) thymidine kinase polypeptide [29]. Figure 5 shows clear but limited homology within several regions of the HSV-1 and HSV-2 thymidine kinase polypeptide chains. The homology is particularly apparent for amino acid residues 47-59, 79-148, 153-211,216-252, 285-304, and 308-341 of the HSV-2 thymidine kinase polypeptide. Discussion

This study describes the cloning of DNA fragments and the determination of the nucleotide sequence of a 1656 bp region containing the entire HSV-2 TK gene. The nucleotide region essential for expression of the HSV-2 TK gene in the bio-

chemical transformation assay was identified by studies on the transforming activities of a series of deletion mutants derived from plasmids pMAR401-H2G, pDT4 and pCT4. The biochemical transformation assays showed that nucleotide sequences 5' to a ClaI site (plasmid pDM33) or 3' to a Sinai site (plasmid pCT415) were dispensable for efficient transformation. The nucleotides 62 bases downstream from the ClaI site appeared to be important, but not essential, since plasmid pCT4 dlEco, which lacks these sequences, had low transforming activity. These observations on the sequences required for efficient transformation are in general agreement with recent studies by Reyes et al. [47], except that the latter investigators observed that cleavage of plasmids containing the HSV-2 TK gene with ClaI and EcoRI inactivated the gene. It may be noted that Kit et al. [26] have previously shown that plasmid pMH1, which contains a 1900 bp EcoRI to PvulI fragment of the HSV-1 TK gene and has 5' flanking sequences similar to pCT4 dlEco, efficiently transformed LM(TK-) cells to the thymidine kinase-positive phenotype. Indeed, Kit et al. [27,28] have described biochemical transformation by a promoterless plasmid (pMHll0), emphasizing that the site of integration of HSV DNA fragments in cellular DNA may affect the expression of the HSV TK gene and that cellular nucleotide sequences flanking the coding portion of the gene may sometimes substitute for viral nucleotide sequences. From the nucleotide sequence analyses, it was possible to identify putative 'CAAT' and 'TATTAA' promoter sequences, an ATG translational start and a TAG translational stop codon, and a polyadenylation signal. Restriction nuclease sites predicted from the nucleotide sequence were in agreement with those mapped by conventional procedures on the plasmids. As anticipated by Reyes et al. [47], a StuI site was found close to the TAG and polyadenylation signals (Figs. 2 and 3). The nucleotide sequence analyses re~ealed extensive homology throughout the coding regions of the HSV-I(C1 101) and HSV-2(333) TK genes. These observations attest to the common evolutionary origin of these thymidine kinase genes and suggest that recombination across the TK genes may sometimes occur in mixed HSV-1 and HSV-2 infections.

169 Knowledge of the three-dimensional structure of thymidine kinase enzymes is desirable to elucidate their catalytic and allosteric properties [56-58], but a crystallographic structure or even a direct amino acid sequence determination of any thymidine kinase is remote. As an interim measure, comparisons based on the amino acid sequences predicted from nucleotide sequences can be useful [59,60]. In this work, we report the predicted amino acid sequence of the HSV-2(333) thymidine kinase polypeptide. The results indicate a clear but limited homology within several regions of the HSV-1 and HSV-2 thymidine kinase polypeptides (Fig. 5). Comparisons of two HSV-1 strains have shown that there are 19 nucleotide changes within the coding portion of the strain CI 101 and MP TK genes, but, of these, only seven resulted in amino acid differences in the two HSV-1 strains [29,30]. As expected, the differences between the thymidine kinase polypeptide of HSV2(333) and HSV-I(CI 101) strains are greater, even in major conserved regions of the thymidine kinase polypeptide. The conserved sequences are probably important for the conformation and catalytic activities of the type 1 and type 2 HSV thymidine kinase activities. Previous investigations have demonstrated that antigenic determinants are surface features of proteins, frequently found on regions of a molecule that have an unusually high degree of exposure to solvents, and that antigenic determinants can be predicted at the point of greatest local hydrophilicity in an amino acid sequence [61]. In the present study, computer analyses indicated that strongly hydrophilic regions occurred at amino acid residues 107-113 and 210-230 of the HSV-2 thymidine kinase polypeptide. In these regions, 6 out of 7 and 16 out of 20 amino acid residues, respectively, were the same in the type 2 and type 1 H S V thymidine kinase polypeptides. It should be possible to verify whether these sequences are antibody-binding sites by synthesizing the indicated regions chemically, testing their activities in an appropriate immune assay, and by raising antisera against the synthetic determinants [61,62]. Additional studies may also disclose whether any of the domains containing conserved amino acid sequences resemble those previously identified as nucleotide binding sites [56-60].

Several HSV-1 and HSV-2 strains with mutations in the TK gene which alter their substrate affinities are available in our laboratory. The isolation and analyses of additional HSV-1 mutants and of other herpesvirus TK genes are planned as approaches towards understanding how the structures of herpesvirus TK genes influence their expression.

Addendum After this work was submitted for publication, we learned that a manuscript on the HSV-2 TK gene sequence was 'in press' (M.A. Swain and D.A. Galloway (1983) J. Virol. 46, 1045-1050). It was requested that we compare sequences with Dr. Galloway so as to minimize discrepancies. Comparisons revealed that the two sequences differed in seven of the 1656 nucleotides. After a careful reexamination of our radioautograms, our sequence was amended. The sequence of Fig. 2 differs from that of Swain and Galloway in that they insert a C after nucleotide Number 264 and three C's after nucleotide Number 1141. The latter change adds a proline to the predicted amino acid sequence. It is interesting to note that the TK gene sequences of two HSV-2(333) subclones which were probably separated only 5-10 years ago appear to show four insertions/deletions of which three are a triplet in the coding region.

Acknowledgements This investigation was aided by USPHS Grants CA-06656-20 and K6-AI-2352-20 from the National Cancer Institute and the Institute of Allergy and Infectious Diseases, respectively. Recombinant DNA experiments were performed in a P2 facility in accordance with National Institutes of Health guidelines.

References 1 Kit, S. (1979) in Pharmacologyand Therapeutics(Sartorelli, A.C. and Shugar, D., eds.), Vol. d, pp. 501-585, Pergamon Press, Ltd., Oxford 2 Juel-Jensen, B.E. and Mm:.callum,F.O. (1972) in Herpes Simplex,Varicella and Zoster, P. 188, J.B. LippincottCompany, Philadelphia 3 Field, H.J., Anderson, J.R. and Wildy, P. (1982) J. Gen. Virol. 59, 91-99

170 4 Field, H.J. and Darby, G. (1980) Antimicrobiol. Agents Chemother. 17, 209-216 5 Field, H.J. and Wildy, P. (1978) J. Hyg. Camb. 81,262-277 6 Hill, T.J., Field, H.J. and Blyth, W.A. (1975) J. Gen. Virol. 28, 341-353 7 Klein, R.J., DeStefano, E., Brady, E. and Friedman-Kien, A.E. (1980) Arch. Virol. 65, 237-246 8 Klein, R.J., Friedman-Kien, A.E. and DeStefano, E. (1981) Infect. Immun. 34, 693-701 9 Price, R.W. and Khan, A. (1981) Infect. Immun. 34, 571-580 10 Tenser, R.B. and Dunstan, M.E. (1979) Virology 99, 417-422 11 Tenser, R.B., Miller, R.L. and Rapp, F. (1979) Science 205, 915-917 12 Kit, S., Qavi, H., Dubbs, D.R. and Otsuka, H. (1983) J. Med. Virol. 12, 25-36 13 Fyfe, J.A., Keller, P.M., Furman, P.A., Miller, R.L. and Elion, G.B. (1978) J. Biol. Chem. 253, 8721-8727 14 DesCamps, J., Sehgal, R.K., DeClercq, E. and Allaudeen, H.S. (1982) J. Virol. 43, 332-336 15 Allaudeen, H.S., DesCamps, J., Sehgal, R.K. and Fox, J.J. (1982) L Biol. Chem. 257, 11879-11882 16 Ashton, W.T., Karkas, J.D., Field, A.K. and Tolman, R.L. (1982) Biochem. Biophys. Res. Commun. 108, 1716-1721 17 Cheng, Y.-C., Dutschman, G., Fox, J.J., Watanabe, K.A. and Machida, H. (1981) Antimicrobiol. Agents Chemother. 20, 420-423 18 Gentry, G.A., Allen, G.P., Holton, R., Nevins, R.B., McGowan, J.J. and Veerisetty, V. (1983) Intervirology 19, 67-76 19 Reefschlager, L, Wutzler, P., Thiel, K.-D., Barwolf, D., Langen, P., Sprossig, M. and Rosenthal, H.A. (1982) Antiviral Res. 2, 255-262 20 Leung, W.-C., Dubbs, D.R., Trkula, D. and Kit, S. (1975) J. Virol. 16, 486-497 21 Weinmaster, G.A., Misra, V., McGuire, R., Babiuk, L.A. and DeClerq, E. (1982) Virology 118, 191-201 22 Kit, S., Jorgensen, G.N., Dubbs, D.R., Chan, S.-K. and Leung, W.-C. (1976) Virology 69, 179-190 23 Darby, G., Field, H.J. and Salisbury, S.A. (1981) Nature 289, 81-83 24 Larder, B.A. and Darby, G. (1982) J. Virol. 42, 649-658 25 Colbere-Garapin, F., Chousterman, S., Horodniceanu, F., Kourilsky, P. and Garapin, A.-C. (1979) Proc. Natl. Acad. Sci. U.S.A. 76, 3755-3759 26 Kit, S., Otsuka, H., Qavi, H., Trkula, D., Dubbs, D.R. and Hazen, M. (1980) Nucl. Acids Res. 8, 5233-5253 27 Kit, S., Qavi, H., Hazen, M., Trkula, D. and Otsuka, H. (1981) Virology 113, 452-464 28 Kit, S., Otsuka, H., Qavi, H. and Kit, M. (1981) Gene 16, 287-295 29 Wagner, M.J., Sharp, J.A. and Summers, W.C. (1981) Proc. Natl. Acad. Sci. U.S.A. 78, 1441-1445 30 McKnight, S.L. (1980) Nucl. Acids Res. 8, 5949-5964 31 Kit, S., Dubbs, D.R., Piekarski, L.J. and Hsu, T.C. (1963) Exp. Cell Res. 31,297-312

32 Otsuka, H., Hazen, M., Kit, M., Qavi, H. and Kit, S. (1981) Virology 113,196-213 33 Pignatti, P.F., Cassai, E., Meneguzzi, G., Chenciner, N. and Milanesi, G. (1979) Virology 93,260-264 34 McDougall, J.K., Masse, T.H. and Galloway, D.A. (1980) J. Virol. 33, 1221-1224 35 Galloway, D.A. and Swain, M. (1980) Gene 11,253-257 36 Van den Berg, F.M. and Walboomers, J.M.M. (1980) Intervirology 14, 217-222 37 Van den Berg, F.M. and Walboomers, J.M.M. (1981) Intervirology 16, 184-190 38 Southern, E.M. (1975) J. Mol. Biol. 98, 503-517 39 Messing, J. (1981) in The Third Cleveland Symposmm on Macromolecules: Recombinant DNA (Walton, A., ed), pp. 143-153, Elsevier, Amsterdam 40 Messing, J. and Vieira, J. (1982) Gene 19, 269-276 41 Vieira, J. and Messing, M. (1982) Gene 19, 259-268 42 HaUiburton, I.W., Morse, L.S., Roizman, B. and Quinn, K.E. (1980) J. Gen. Virol. 49, 235-253 43 McKnight, S.L. and Kingsbury, R. (1982) Science 217, 316-324 44 Smiley, J.R., Wagner, M.J., Summers, W.P. and Summers, W.C. (1980) Virology 102, 83-93 45 Sanger, F., Nicklen, S. and Coulson, A.R. (1977) Proc. Natl. Acad. Sci. U.S.A. 74, 5463-5467 46 Galloway, D.A., Fenoglio, C.M. and McDougall, J.K. (1982) J. Virol. 41,686-691 47 Reyes, G.R., Jeang, K.-T. and Hayward, G.S. (1982) J. Gen. Virol. 62, 191-206 48 Minson, A., Bell, S.E. and Bastow, K. (1982) J. Gen. Virol. 58, 127-138 49 Breathnach, R. and Chambon, P. (1981) Annu. Rev. Biochem. 50, 349-383 50 McKnight, S.L. (1982) Cell 31, 355-365 51 Read, G.R. and Summers, W.C. (1982) Proc. Natl. Acad. Sci. U.S.A. 79, 5215-5219 52 Kozak, M. (1981) Nucl. Acids Res. 9, 5233-5252 53 Bienz, M., Kubli, E., Kohli, J., de Henau, S., Huez, G., Marbaix, G. and Grosjean, H. (1981) Nucl. Acids Res. 9, 3835-3849 54 Kit, S. (1976) Mol. Cell. Biochem. 11,161-182 55 Kit, S., Jorgensen, G.N., Dubbs, D.R., Trkula, D. and Zaslavsky, V. (1978) Intervirology 9, 162-172 56 Bryant, T.N., Watson, H.C. and Wendell, P.L. (1974) Nature 247, 14-17 57 Blake, C.C.F. and Evans, P.R. (1974) J. Mol. Biol. 84, 585-601 58 Rossmann, M.G., Moras, D. and Olsen, K.W. (1974) Nature 250, 194-199 59 Wootton, J.C. (1974) Nature 252, 542-546 60 Chou, P.Y. and Fasman, G.D. (1978) Annu. Rev. Biochem. 47, 251-276 61 Hopp, T.P. and Woods, K.R. (1981) Proc. Natl. Acad. Sci. U.S.A. 78, 3824-3828 62 Prince, A.M., Ikram, H. and Hopp, T.P. (1982) Proc. Natl. Acad. Sci. U.S.A. 79, 579-582