VIROLOGY 169, 346-353 (1989)

The Minimal Transforming Fragment of HSV-2 mtrlll Can Functionas a Complex Promoter Element

CLINTON JONES'

Department of Microbiology, University of Mississippi Medical Center, 2500 North State Street, Jackson, Mississippi 39216-4505

Received September 28, 1988 ; accepted November 29, 1988

The minimal transforming fragment (486 TF) of HSV-2 mtrlll (0 .567-0 .570 map units) is composed of two distinctand non-overlapping promoter elements when linked to bacterial CAT genes . A 230-nucleotide fragment of 486 TF,SaA-Ho, was active as a promoter element in primate cells but not rodent cells . A 173-nucleotide fragment, Smal-Pstl, was active in both primate and rodent cells. The 486 TF did not compete for limiting cellular factors required todrive the CAT gene under control of the SV40 early promoter/enhancer . However, gel-retardation assays suggest thatunique factors exist in cells transformed by HSV-2 which specifically recognized regions of 486 TF . These results arediscussed with respect to HSV-2-mediated transformation . ©19a9Academlc Press, Inc .

INTRODUCTION

Two distinct regions of herpes simplex virus 2 (HSV-2) induce transformation of cultured cells in transfec-lion assays using cloned viral DNA fragments (Fig . 1) .One sequence located between 0 .585 and 0 .63 mapunits (m .u.), designated mtrll (morphological transform-ing region), transforms BaIbC/3T3 cells (Reyes et al.,1979), NIH 3T3 cells, and rat embryo fibroblasts (Gallo-way et al ., 1981, 1984) . A second sequence between0.42 and 0 .58 m .u . (Bglll-C), designated mtrlll, can fullytransform Syrian hamster embryo cells (SHE) . The left-hand ; of Bg/ll-C immortalizes diploid cells while theright-hand 's, BamHl-E fragment, induces the tumori-genic conversion of established cell lines (Jariwalla etal ., 1983, 1986). The minimal transforming region ofBamHl-E has been mapped to a 486-bp Pstl-Sail frag-ment, 0.567-0.570 m .u ., using an established rat cellline (Rat-2) (Jones et al., 1986). The 486-bp minimaltransforming fragment of HSV-2 mtrlll (designated 486TF) is located near the amino terminus of the virus-en-coded large subunit of ribonucleotide reductase . Ex-pression of viral proteins is not apparently required toinitiate or maintain the transformed state since an in-tact viral gene is not contained within 486 TF . Howeverstem-loop structures, alternating purine/pyrimidinemotifs, inverted repeats, and purine- or pyrimidine-richelements are clustered in 486 TF . Thus, multiple repeti-tive motifs, which resemble control elements in recom-bination and gene activation, apparently mediate thetransformation process .

In this report, it is demonstrated that two fragmentsderived from 486 TF can function as transcriptionalpromoter elements when fused to the bacterial gene

To whom requests for reprints should be addressed .

0042-6822/89 $3.00copyright 01989 by Academic Press, Inc .All rights of reproduction in any form reserved.

346

coding for chloramphenicol acetyltransferase gene(CAT). The promoter for the HSV-2-encoded large sub-unit of ribonucleotide reductase (RRA), 0 .42-0 .51, canalso efficiently cis-activate the CAT gene . In addition,an 83-nucleotide fragment derived from 486 TF, Hpal-Smal, cis-repressed the two promoter activities. Evi-dence is also presented which demonstrates the RRApromoter, but not 486 TF, can compete for limiting cel-lular factors required to trans-activate the SV40 earlypromoter/enhancer . Gel-retardation assays revealedthat factors in the nucleus of mammalian cells specifi-cally recognize 486 TF . When binding of 486 TF withRat -2 nuclear extracts was compared to extracts de-rived from Rat-2 cells transformed by HSV-2, qualitativeand quantitative differences were seen . These resultsare discussed with respect to HSV-2-induced morpho-logical transformation and expression of HSV-2 genes .

MATERIALS AND METHODS

Cells and media

CV-1 (monkey fibroblasts) were obtained from Dr .Robert Su . COS-7 (SV40 transformed monkey fibro-blasts) was obtained from the ATCC. Rat-2 (rat fibro-blasts) and HSV-2-transformed Rat-2 cells (183 F3RTI,183F8RT1, 778) were obtained from Dr . Raxit Jariwalla .All cells were grown in Earle's modified Eagle's me-dium (Sigma) supplemented with 5% fetal bovine se-rum, penicillin (10 U/ml), and streptomycin (100 µg/ml) .

Eucaryotic cell transfections

Cells were transfected with 10 tg DNA (exceptwhere noted otherwise) using the CaPO, precipitationprocedure (Graham and Van der Eb, 1973) .

FUNCTIONAL ANALYSIS OF HSV-2 MINIMAL TRANSFORMING FRAGMENT

347

CAT assays

Forty hours after transfection, cells were washedthree times with phosphate-buffered saline and a celllysate was prepared by three freeze-thaw cycles in0.25 MTris (pH 7 .8) . Cell debris was pelleted by centrif-ugation and protein concentration of the cell extractmeasured (Bradford, 1976) . The cellular supernatantsadjusted to 25 jig protein were assayed for CAT activityin a reaction mixture containing 0 .2 pCi [ 14C]chloram-phenicol (45 mCi/mmol), 100 mM Tris-HCI (pH 7 .8),0 .5 mM acetyl-CoA (Gorman et al ., 1982). After incuba-tion for the indicated time at 37°, chloramphenicol wasextracted with 1 ml ethyl acetate and the enzymaticproducts were analyzed by thin-layer chromatography(TLC) on silica gel in the solvent system CHCI 3/MeOH(95/5). TLC plates were autoradiographed at room tem-perature for 48 hr. The amount of radioactivity associ-ated with [ 14C]CM and the various forms of acetylated[ 14C]CM in TLC plates was measured with a System400 Imaging Scanner (Bioscan) .

Construction of HSV-2 CAT plasmids

The plasmid pSV2CAT contains the SV40 early pro-moter and enhancer 5'to the bacterial gene encodingchloramphenicol acetyltransferase (CAT) (Gorman etal., 1982 ; Laimins et al., 1982) . The plasmid pSVOCATdoes not contain a promoter or enhancer at the 5' ter-minus of the CAT gene . pSVOCAT contains uniqueHindlll and BamHl sites 5' and 3' respectively to theCAT gene. These plasmids were obtained from Dr.Bruce Howard .To assess the promoter activity of HSV-2 BamHl-E

fragments in mammalian cells, 486 TF, 826-bp Pstl-SaIIA fragment of Pstl-C, and Pstl-B fragment wereeach fused to the CAT gene in pSVOCAT . Recombi-nant molecules containing these fragments have beenpreviously described (Jones et al., 1986) . The respec-tive fragments were purified on polyacrylamide gels,made blunt end by the Klenowfragment, and ligated toHindlll linkers . The 486-bp transforming fragment wasinserted in both orientations at the unique Hindlll siteof pSVOCAT and the resulting plasmids designatedpOTF/1 and pOTF/4 . Orientation of the 486-bp frag-ment with respect to the CAT gene was determined bySail-BamHl restriction digestion . The promoter regionof the HSV-2-encoded RRA is contained within the Pstl-B fragment of BamHl-E (Swain and Galloway, 1986) .Pstl-B fragment of HSV-2 was inserted in both orienta-tions at the unique Hindlll site of pSVOCAT and theresulting plasmids were designated pBOCAT/24 andpBOCAT/31 . Orientation of Pstl-B with respect topSVOCATwas mapped by Sacl digestion . The 826-nu-cleotide Sall fragment of Pstl-C was inserted at the Hin-

dill site of pSVOCAT and designated p0 .8CAT/A andpO .8CAT/B. A summary of the HSV-2 BamHl-E locusand relative CAT constructs is presented in Figs . 1and 2 .

Restriction fragments of 486 TF were purified in poly-acrylamide gels as described above and after additionof Hindlll linkers cloned into the unique Hindlll site ofpSVOCAT. A summary of these constructs is pre-sented in Fig . 3 . The orientation of these fragmentswith respect to CAT was deduced by restriction en-zyme mapping using single sites in the respective frag-ment and the unique BamHl site in pSVOCAT .

Preparation of nuclear extracts

Nuclear extracts were prepared from mammaliancells according to the procedure of Dignam et al.,1983 . Extracts were frozen in liquid nitrogen and sub-sequently stored at -1 10° .

DNA probes

The respective fragments of 486 TF were obtainedfrom the plasmids pOSS/1, pOSH/1, pOSP/2 (Fig . 3) .The inserts were released by digestion with Hindlll andsubsequently 3'-end-labeled with Klenow fragment . Toobtain the Hpal-Smal fragment, the insert of pOSS/1was digested with Hpal . All probes were purified byelectrophoresis in 3 .5-7 .0% polyacrylamide gels andsubsequently electroeluted, phenol-extracted twice,ether-extracted, and ethanol-precipitated .

0.4

0s

osMAP UNIT 1IiII

E3

C IN1~Tranaform .

Trensfarm(mmll)

Immq

tumor ize(.rill)

9emHl BIH1H'I 0IYIJI M IEI T IA

486 TF

.RNA

Protein

UL = us

1

FIG . 1 . Structural organization of HSV-2 BamHl-E . Map coordi-nates along the HSV-2 genome are indicated above the linear map .Restriction enzyme map of BamHl-E was previously reported (Joneset al_ 1986) . Position of the mRNA and 144-kb large subunit of ribo-nucleotide reductase protein are indicated (Swain and Galloway,1986) . The 486 transforming fragment (TF) of BamHl-E is indicatedby the solid box .

348

CLINTON JONES

Binding assays by gel-shift analysis

A.

Binding reactions were performed according toGreen et al, (1987) in a final volume of 20

ulcontaining

the following components : 0.5-1 ng end-labeled frag-ment (10,000-25,000 cpm), 45 mM KCI, 25 mMHEPES (pH 7 .6), 1 .0 mM EDTA, 0 .5 mM DTT, and 3pg poly(dl :dC) . Binding reactions were initiated by theaddition of 1-4 ul (1-4 ug protein) of nuclear extract .After 20 min at 20°, 3 ul loading buffer was added (60%glycerol, 10 mM Tris (pH 7 .5), 1 mM EDTA, 0 .003%bromphenol blue). The reaction mix was run on a 1 %agarose gel in 10 TIM Tris (pH 7.5), 1 mM EDTA at 4°and the buffer recirculated (Green et al., 1987). The gelwas dried and autoradiographed at-70° .

RESULTS

cis-Activation of CAT gene by HSV-2 mtrlllsequences

To test the ability of 486 TF to activate transcription,486 TF was inserted next to the CAT gene of pSVO-CAT, a plasmid lacking promoter/enhancer elements,COS-7 cells were transfected with these constructs(pOTF/1 and pOTF/4), and 40-hr posttransfection lev-els of CAT activity were measured (Gorman et al .,1982). The 486 TF cis-activated the promoter/en-hancer minus CAT plasmid pSVOCAT (Fig . 2). Further-more, the promoter activity of 486 TF was only slightlyaffected by orientation . Thus, 486 TF functioned as abidirectional promoter element in COS-7 cells . In con-trast, the adjacent Sall fragment of Pstl-C did not func-tion as a promoter . These findings demonstrate thatspecific sequences in 486 TF can cis-activate tran-scription and that cis-activation was not merely due toGC-rich (HSV-2) DNA at the 5' terminus of the CATgene .The promoter activity of 486 TF was compared to

that of the HSV-2 Pstl-B fragment . Promoter se-quences for the HSV-2-encoded large subunit of ribo-nucleotide reductase are contained in the Pstl-B frag-ment of BamHl-E (Fig . 2) (Swain and Galloway, 1986) .As expected, Psrl-B also functioned as a promoter inCOS-7 cells when inserted at the 5' terminus of theCAT gene in pSVOCAT (pBOCAT/24, pBOCAT/31)(Fig . 2) . Overall, the promoter activity of Pstl-B was two-to fourfold less than 486 TF promoter constructs butwas significantly higher than in cells transfected withpSVOCAT .

Localization of promoter activity in 486 TF

Further experiments were undertaken to identify thefunctionally essential components of promoter activity .To this end, small restriction fragments of 486 TF were

HSV-2 Bam-E

Sic ISa l c'

Hp alPll

III

BamH I

I t

1"1' BamMIPat : FlABICDT EBen

E I

A

I D I C l

B

486,FPRA

Sacl

,'--~

alI]1

B

pall

C

Pt1 `-bririrrrnmrr~I~

sell

B .

% AcelaterCa

72

es

02

FIG . 2. Promoter activity of mtrlll DNA fragments . (A) Restrictionenzyme map of HSV-2 BamHl-E and approximate start site for tran-scription of viral encoded ribonucleotide reductase . (B) Promoter ac-tivity of BamHI-E fragments. The respective restriction fragments ofBamHl-E were inserted into the unique Hindlll site of pSVOCAT asdescribed under Materials and Methods . The resulting HSV-2/CATplasmids were transfected into monkey cells (COS-7) . Forty hourspost-transfection cell lysates were prepared and subsequently as-sayed for CAT activity in the presence of [ 14 C] chloramphenicol (CM)for 60 min at 37° . Acetylated forms of CM (ac-CM) were separatedfrom CM by thin-layer chromatography . The levels of ac-CM is ex-pressed as a percentage of the total radioactivity present in the vari-ous forms of CM . The direction of CAT transcription is indicated bythe solid arrow . The direction of the promoter for large subunit ofribonucleotide reductase in Psrl-8 is designated by the dashed ar-row. Restriction enzyme sites are H(Hindlll), B(BamHl), P(Pstl),S(Sall), and Sacl(Sacl) .

inserted into the unique Hindlll site of pSVOcat . Pro-moter activities were measured in transfected CV-1 orRat-2 cells. The Sall-Hpal fragment, 230 nucleotides,was an efficient promoter element with slight orienta-tion preferences (pOSH/1, pOSH/2) . The activity ofSail-Hpal was approximately three- to fivefold higherthan intact 486 TF . However, this fragment displayedlittle or no activity in Rat-2 cells . In addition, this frag-ment possessed high promoter activity in human fi-broblasts but little or no activity in two mouse cell lines,NIH 3T3 and JB-6 (data not shown) . Thus, the Sall-Hpal fragment functioned as a promoter element in pri-mate cells .

The Smal-Psrl fragment of 486 TF, 173 bp, alsofunctioned as a promoter element in CV-I cells and Rat-

HSV-2insert

PlaamlaDesignation

CATConstruct

Psac pOBCAT/ASal A

010 pnaunesa1A

48. TF 9OTF1

ae TF poTF!

sadPvle pe0G4T124 CAT

Ps4B ptOCnr,al -1 sa;CAT

pmcAT CAT6" - .

CAT 6pSVOGT

FUNCTIONAL ANALYSIS OF HSV-2 MINIMAL TRANSFORMING FRAGMENT

349

2 cells (Fig . 3 ; pOSP/1, pOSP/2) . This fragment dis-played sharp orientation preferences with promoter ac-tivity dependent on the Smal site being adjacent to CATcoding sequences (pOSP/1). Furthermore, promoteractivity of Smal-Pstl (pOSP/1) was significantly higherthan intact 486 TF .A Hpal-Smal fragment, 83 nucleotides, significantly

reduced the promoter activity of the two promoter ele-ments (pOSS/1, pOSS/2, pOHP/1, pOHP/2) . Evenwhen the Hpal-Smal region was oriented such that it

Plasmid

48611

CAT construct

%Acetylated CMnest nation Insert

CV-1

Rat-2

was not proximal to the 5' terminus of CAT coding se-quences (Fig . 3; pOSS/1), a pronounced negativeeffect was seen . This small fragment virtually elimi-nated promoter activity of Smal-Pstl fragment. How-ever, given the orientation preferences of Smal-Pstlpromoter activity this was predictable . Taken together,these results indicate that 486 TF can function as acomplex transcriptional regulatory element composedof two non-overlapping promoters and a "silencer" re-gion .

FIG . 3 . Promoter activity of 486 TF subclones . Small restriction fragments were inserted into the unique HindlIl site of pSVOCAT . The respec-tive constructs were transfected into either monkey (CV-1) or rat cells (rat-2) and CAT activity was measured 40 hr post-transfection for 30 minat 37°. The levels of ac-CM are expressed as percentages of the total radioactivity present in the various forms of CM . CAT transcription is fromleft to right . Restriction enzyme sites are P(Pstl), Sm(Smal), Hp(Hpal), S(Sall), H(Hindllp, and B(BamHl) . The values given are the average of threeindependent experiments.

P Sm Hp S/H cat

POTF/1 486W 4.9±2 6.1±3

DOSS/1 SALI-SwJI

H/Sm Hp S/H

cat

3 .4±2 0.7±0.4

DOSS/2 SII-SAI

H/S Hp Sm/H cat

1.7±1 0.5±0.3

POSH/1 SALI-*AI

H/Hp S/H cat

F 25.9±8 3.5!1

H/S Hp/H cat

POSH/2 SALI-HPAI 14.1±5 2.1±1

POHP/1 HPAI-PsTI

H/P Son. Hp/H cat

0.5±0.2 0.5+0 .3 I

/1-

POP/2 HPAI-PsTI

H/Hp SmuP/H

I

cat

/ 0.8±0.5 0.4+0 .2

POSP/1 SrwI-PSTI

H/P Sm/H cat

17 .5±5 16 .7±71 //

POSP/2 SwtI-PsTI

H/Sm

P/H cat

4.8±2 1.5±0 .7

PSVOCAT

cat

// 0.3±0 .2 0.2±0 .1

SV40

11 cat

PSV2cATE.promJenh,

//

L49 .5±12 48.8~18

RSVCAT

RSVLTR

Icat

/b 36.4±9 35.7±12

3 50

CLINTON JONES

486 TF does not compete with the SV40 promoterfor transcription factors in vivo

Numerous investigators have identified cellular fac-tors that interact with viral transcriptional regulatory se-quences. For example, the cellular transcription factorSpI binds to GC-rich sequences within the SV40 21 -bprepeat (Dynan and Tjian, 1983), AP-1 specifically rec-ognizes sequences in SV40 72-bp repeats (Lee et a/ .,1987), and AP-2 binds to unique domains in SV40 72-bp repeats and early promoter (Mitchell et al ., 1987) .Thus, it appears that multiple cellular factors can inter-act with viral promoters to induce the expression of vi-ral genes .To determine if SV40 early promoter/enhancer and

HSV-2 promoter sequences utilized the same tran-scription factors, the in vivo competition assay ofScholer and Gruss (1984) was used . In this assay thetest promoter (SV40), fused to the CAT gene, is co-transfected with increasing amounts of the competingpromoter without CAT (HSV-2) . The total amount ofDNA used to transfect cells is kept constant by the ad-dition of unrelated DNA sequences, in this case pUC-19. The results (Fig . 4) demonstrate that 486 TF didnot readily compete with the SV40 promoter for limitingfactors in Rat-2 or CV-I . In sharp contrast, the ribonu-cleotide reductase promoter (Pstl-B) significantly re-duced the amount of CAT synthesis driven by the SV40promoter. As expected, the SV40 promoter was aneffective competitor of itself . This finding implied thatunique cellularfactors trans-activated 486 TF, whereasthe promoter of ribonucleotide reductase apparentlyutilizes one or more cellular factors that were neces-sary for SV40 promoter activity .

Assay for 486 TF promoter-binding factor

The principal difficulty in detecting a protein that hasspecificity for a given DNA sequence in crude extractsis the presence of other DNA-binding proteins . How-ever, a gel electrophoresis assay is capable of doingthis (Garner and Revzin, 1981 ; Fried and Crothers,1981). This technique involves analysis of protein-DNA binding by gel electrophoresis under conditionswhere binding of a protein to DNA will slow its migra-tion and yield a discrete band . Such a band is visualizedby suppressing nonspecific protein binding to the la-beled DNA probe with random DNA sequence . Poly(dl :dC) (1500 jug/ml) was used in these studies as heterol-ogous competitor DNA and is an extremely usefulblocking agent of nonspecific DNA-protein interac-tions (Singh et al., 1986) . When fragments derived from486 TF were end-labeled and incubated with nuclearextracts derived from mammalian cells, specific bind-ing of nuclear factors to these fragments occurred as

10090go

A

0-0-0

0

0-0-0

o

B

5

20ug Cwietitor DNA

FIG . 4 . Promoter 486 TF and SV40 promoter/enhancer do not com-pete for limiting cellular transcription factors . Two micrograms ofpSV2CAT DNA (Gorman et al., 1982) was transfected in Rat-2 (A) orCV-1 (B) cells with increasing amounts of plasmids which containedeither 486 TF (0, pUTF/1) or HSV-2 Pstl-B (0, pUB-13), or the SV40Hindlll C fragment (A, pSVC) . pSVC contains the intact SV40 en-hancer/promoter sequences (Hindlll C) and has been previously de-scribed (Jones and Su, 1987) . The total DNA concentration wasmaintained at 22 µg by the addition of pUC/19 DNA . Cell lysateswere prepared and CAT activity was assayed 40 hr after transfection.The 100% control was pSV2CAT without any competitor exceptpUC-19 .

judged by the gel retardation assay (Fig . 5). Factorspresent in nuclear extracts prepared from CV-I or COS-7 cells formed multiple DNA-protein complexes withthe Sall-Hpal fragments of 486 TF (Fig . 5A) . In con-trast, equivalent amounts of nuclear extracts preparedfrom Rat-2 cells (normal or HSV-2 transformed) dis-played diminished binding to the fragment . This corre-lated with diminished promoter activity of Sall-Hpalfragment in rodent cells (see Fig . 3). The "silencer'' re-gion, Hpal-Smal, and the orientation-specific promoterelement, Smal-Pstl, of 486 TF also formed specificcomplexes with factors in nuclear extracts (Figs . 5Band 5C). When binding reactions of these fragmentswere compared using nuclear extracts derived fromnormal Rat-2 cells and HSV-2-transformed cells, quali-tative and quantitative differences were observed . Twoadditional transformation specific bands were ob-served using the "silencer" region as a probe and oneadditional transformation specific band was observedif the orientation-specific promoter element was usedas a probe . Nuclear extracts derived from two otherRat-2 cell lines transformed by HSV-2 have been usedin gel-retardation assays and transformation-specific

70

60

5040 .\30

20

10

2

s

10

20

0 2

FUNCTIONAL ANALYSIS OF HSV-2 MINIMAL TRANSFORMING FRAGMENT

DlI

A

B

CV-1

DNA:protein complexes were observed (data notshown). Furthermore, an additional major band wasobserved in CV-1 but not COS-7 cells when the ''si-lencer" region was used as a probe . These results indi-cated that transcriptional regulatory fragments of 486TF formed specific DNA-protein complexes with pro-teins in nuclear extracts .

To define the binding of nuclear factors to fragmentsderived from 486 TF, reactions were performed with100-fold excess of unlabeled homologous probe orhigher concentrations of MgCl 2 . The results of theseexperiments are shown in Fig . 6 . Addition of excessnonradioactive homologous DNA eliminated bandswhich migrated with retarded mobilities for each re-spective probe . This implied that specific sequencesin the respective fragments are recognized by nuclearfactors . Addition of MgCl2 to the reactions had only mi-nor effects on proteins binding to either the Hpal-Smalprobe or the Smal-Pstl probe. However, MgCl 2 had aquantitative and qualitative effect on binding reactionswhich contained the Sall-Hpal fragment . This may im-ply that Mg" ions stimulate binding of nuclear factorsto this fragment, or that a class of factors require Mg"to specifically bind DNA . A similar effect by MgCl 2 onbinding of nuclear factors to the polyoma virus en-hancer has been described (Bohnlein and Gruss,1986) . In summary, these results strongly suggest thatthe DNA:protein complexes observed were due to a

cas-7 RAT RAT

His I

SmslB

486TF

FIG . 5 . Interaction of nuclear factors from mammalian cells with 486 TF . Nuclear extracts were prepared from the respective cell lines by themethod of Dignam . Probes of 486 TF were end-labeled by the Klenow fragment and subsequently purified in a polyacrylamide gel . The respectiveend-labeled probes (0 .5-1 .0 ng) were incubated (20 min, 20°) with the indicated concentration (pg) of nuclear extract in the presence of poly(dl :dC) (3 µg) . A gel retardation assay was carried out to separate DNA :protein complexes Monkey extract DNA :protein complexes (- .). Rat-2extract DNA:protein complex (.-). HSV-2/rat-2 specific DNA :protein complex (4) . Probe A was derived from the Sall-Hpal fragment of 486 TF .Probe B was derived from the Hpal-Smal fragment of 486 TF. Probe C was derived from Smal-Psrl fragment of 486 TF .

CRSV

CV-1 COS-7 RAT RAT

PgtlC

361

specific interaction between several cellular factorsand 486 TF .

DISCUSSION

The data presented indicate 486 TF can function asa complex transcriptional element . Two separate pro-moter elements and a ''silencer" region were identified(Figs. 3 and 7) . Nuclear factors specifically interactedwith the individual regulatory components of 486 TF .Differential binding to regions of 486 TF was observedwhen normal extracts from rat cells were compared tothose from HSV-2-transformed cells . Thus, 486 TF maybe a promoter element which is trans-activated by aunique class of transcriptional factors .The promoter activity of 486 TF was significantly

lower than two smaller subclones of this fragment (Fig .7) . These smaller subclones did not contain the 83-nu-cleotide "silencer" region which is located betweenHpal and Smal. Even when the "silencer" sequencewas not adjacent to the CAT coding sequence, the pro-moter activity of Sall-Hpal fragment was lower(pOSS1 ; Fig 3) . Thus, the Hpal-Smal region of 486 TF"silenced'' or"repressed" the promoter activity of 486TF. It is tempting to speculate that proteins which spe-cifically bind to the "silencer'' mediate this processsince factors in nuclear extracts specifically bind thisfragment. The DNA sequence of the "silencer" con-

35 2

CLINTON JONES

tains a 12-base direct repeat, a 9-base element withdyad symmetry, and two regions which have the poten-tial to form z-DNA (Jones et al., 1986). Whether theseelements play a role in "silencer" function is being in-vestigated .

Nuclear factors specifically interacted with both pro-moter elements of 486 TF . Furthermore, reduced bind-ing activity of factors in rodent cells to the Sall-Hpalfragment correlated well with the diminished promoteractivity of this fragment in rodent cells . Transformation-specific DNA :protein complexes were observed whenboth the Hpal-Smal fragment (''silencer") and Smal-Pstl fragment were used as probes . This may indicatethat HSV-2-transformed cells have novel DNA bindingproteins or proteins with altered affinities for DNA . Thesequence of 486 TF (Jones et al., 1986) contains sev-eral Spl binding sites as well as numerous repetitiveelements. However, there is not a consensus Gold-berg/Hogness box (TATAA) . Thus, it should prove in-teresting to determine where the nuclear factors inter-act with 486 TF and their role in promoter activity and/or transformation .The finding that 486 TF contains two distinct pro-

moter elements is surprising since RRA is the onlyknown transcript which maps to this region (Swain andGalloway, 1986 ; references therein) . It is possible thatthese promoter activities, in the context of the HSV-2genome, are cryptic. However, it also is conceivable

1 2 3 4A

Sal II

1 2C3 4 1 2 3 4

B

Hpa I Sma I

Pst IIIIA

B

C

FIG . 6. Biochemical requirements of DNA protein interactions .End-labeled fragments were incubated with 1 fig nuclear extract de-rived from CV-1 cells . Probe A was derived from the SaIl-Hpal frag-ment of 486TF . Probe B was Hpal-Smal . Probe C was Smal-Pstl . (1)Normal binding reaction . (2) Normal binding reaction plus 100-foldexcess unlabeled homologous fragment . (3) Normal binding reactionplus 1 mm MgCl, . (4) Normal binding reaction plus 10 mm MgCl, .

P

Sm Hp

S

Prnmter

"Silence"°

P-1-

t

tcraneformatto,, SPeciftcbindin g by nuclear P*at.tne

FIG . 7. Schematic representation of transcriptional regulatory com-ponents of 486 TF . Restriction enzyme sites are S(Sall), Hp(Hpal),Sm(Smal), and P(Pstl) .

that other small or less abundant transcripts originatein 486 TF and transcription is regulated by one of thesepromoter elements . Within 486 TF, there are severalsmall open reading frames (data not shown) . Clearly itwill be of great interest to examine the role of thesepromoter activities during the lytic infection cycle .

The mechanism of HSV-2-mediated transformation,in particular by 486 TF, is not clear. Numerous studieshave clearly indicated that expression of viral genes isnot necessary (reviewed by McNab, 1987) . This studyclearly indicated that 486 TF could stimulate the ex-pression of a reporter gene, CAT . Perhaps 486 TF me-diates transformation, in part, by altering the expres-sion of cellular genes . It is noteworthy that a transform-ing fragment of cytomegalovirus can also function as atranscriptional enhancer (Galloway et al ., 1985). Cur-rent studies are aimed at identifying cellular geneswhich are differentially expressed in transformed cellsand determining whether 486 TF promoter activitiesplay a role in this process .

ACKNOWLEDGMENTS

This work was supported by grants from NIH (R29-CA47872,2-507-RR05386). I thank Drs . G. Chinchar, G . Gentry, andA. Razzaque for helpful discussions and a critical reading of themanuscript, L . Devine for the manuscript preparation, and D . Dalefor artwork .

REFERENCES

BOHNL[IN, E ., and Gauss, P . (1986) . Interaction of distinct nuclearproteins with sequences controlling the expression of polyomavi-rus early genes . MoL Cell. Biol. 6, 1401-1411 .

BRADFORD, M . M . (1976) . A rapid and sensitive method for the quanti-tation of microgram quantities of protein utilizing the principle ofprotein-dye binding . Anal. Biochem.72, 248-264 .

DIGNAM, J . D ., LEROVITZ, R . M ., and ROEDER, R . G . (1 983) . Accuratetranscription initiation by RNA polymerase II in a soluble extractfrom isolated mammalian nuclei . Nucleic Acids Res. 1 1, 1475-1489 .

DYNAN, W . S ., and TnAN, R . (1983) . The promoter-specific transcrip-tion factor SpI binds to upstream sequences in the SV40 early pro-moter. Cell 35, 79-87 .

FRIED, M ., and CROTHER5, D . M . (1981) . Equilibria and kinetics of lacrepressor-operator interactions by polyacrylamide gel electropho-resis . Nucleic Acids Res. 9, 6505-6525 .

GALLOWAY, D . A., BuoNAGUnE, F . M ., BRANDT, C . R., and McDOUGALL,

FUNCTIONAL ANALYSIS OF HSV-2 MINIMAL TRANSFORMING FRAGMENT

J . K . (1984) . Small fragments of herpesvirus DNA with transformingactivity contain insertion-like structures . Proc. Natl. Aced. Sci. USA81,4736-4740 .

GALLOWAY, D . A ., BUONAGURE, F . M ., BRANDT, C . R ., and MCDOUGALL,1 . K . (1985) . Herpes simplex virus and cytomegalovirus : unconven-tional DNA tumor viruses . In "DNA Tumor Viruses, Cancer Cells",(M. Botchan, T . Grodzicker, and P . A . Sharp, Eds .), Vol . 4, pp . 355-361 . Cold Spring Harbor Laboratory, New York .

GALLOWAY, D. A., and McDOUGALL, 1 . K. (1981) . Transformation ofrodent cells by a cloned DNA fragment of herpes simplex virustype 2 .1. Virol. 38, 749--760 .

GARNER, M . M ., and REVZIN, A. (1981) . A gel electrophoresis methodfor quantifying the binding of proteins to specific DNA regions . Ap-plications to components of the E . cell lactose operon regulatorycomponents . Nucleic Acids Res . 9, 3047-3060 .

GORMAN, C. M., MOFFAT, L . F ., and HOWARD, B . H . (1982) . Recombi-nant genomes which express chloramphenicol acetyltransferasein mammalian cells . Mol . Cell. Biol. 2, 1044-1051 .

GRAHAM, T . L., and VAN DER Es, A . 1 . (1973). A new technique for theassay of infectivity of human adenovirus 5 DNA . Virology 52, 456-467 .

GREEN, P . J ., KAY, S . A., and CHUA, N .-H . (1987) . Sequence-specificinteractions of a pea nuclear factor with light responsive elementsupstream of the rbc5-3A gene . EMBOJ. 6, 2543-2549 .

JARIWALLA, R . J ., AURELIAN, L ., and Ts'o, P . O . P . (1983) . Immortaliza-tion and neoplastic transformation of normal diploid cells by de-fined cloned DNA fragments of herpes simplex virus type 2 . Proc.Nell. Acad. Sci. USA 80, 5902-5906 .

JARIWALLA, R . J ., TANCZOS, B ., JONES, C ., ORTIZ, J ., and SALIMI-LOPEZ,S . (1986). DNA amplification and neoplastic transformation medi-ated by a herpes simplex fragment containing cell related se-quences.Proc .NatI.Aced. Sci.USA 83,1738-1742 .

JONES, C ., ORTIZ, J ., and JARIWALLA, R . J . (1986) . Localization and com-parative nucleotide sequence analysis of the transforming domain

353

in herpes simplex virus DNA containing repetitive genetic ele-ments . Proc. Nail. Aced. Sci. USA 83, 7855-7859 .

JONES, C ., and So, R . T . (1987) . Association of viral and plasmid DNAwith the nuclear matrix during productive infection . BBA 910, 52-62 .

LAIMws, L. A ., KHOURY, G ., GORMAN, C ., HOWARD, B ., and CRUSE, P .(1982) . Host specific activation of transcripts by tandem repeatsfrom simian virus 40 and Moloney murine sarcoma virus . Proc .Nat!. Aced. Sci. USA 79, 6453-6457 .

LEE, W., MITCHELL, P ., and TJIAN, R . (1987) . Purified transcription fac-tor AP-1 interacts with TPA inducible enhancer elements, Cell 49,741-752 .

MONAD, J . C . M . (1987) . Herpes simplex virus and human cytomega-lovirus: Their role in morphological transformation and genital can-cers . l. Gen. Viral. 68, 2625-2550 .

MITCHELL, P .1 ., WANG, C ., and TTIAN, R . (1987) . Positive and negativeregulation of transcription in vitro: Enhancer binding protein AP-2is inhibited by SV40 T-antigen . Cell 50, 847-861 .

REYES, G . R., LAFEMINA, R ., HAYWARD, S . D., and HAYWARD, G . S .(1979) . Morphological transformation by DNA fragments of humanherpesviruses: Evidence for two distinct transforming regions inherpes simplex virus types 1 and 2 and lack of correlation withbiochemical transfer of the thymidine kinase gene . Cold SpringHarborSymp . Quent. Biol. 44, 629-641 .

SCHOLER, H . R ., and Gauss, P . (1984) . Specific interaction betweenenhancer-containing molecules and cellular components . Cell 36,403-411 .

SINGH, H ., SEN, R ., BALTIMORE, D ., and SHARP, P. A . (1986) . A nuclearfactor that binds to a conserved sequence motif in transcriptionalcontrol element of immunoglobulin genes. Nature (London) 319,154-157 .

SWAIN, M . A ., and GALLOWAY, D . A. (1986) . Herpes simplex virus spe-cifies two subunits of ribonucleotide reductase encoded by 3'-co-terminal transcripts . J. Virol . 57, 802-813 .