Mapping of RNA- temperature-sensitive mutants of Sindbis virus: complementation group F mutants have lesions in nsP4.

(1)

0022-538X/89/031194-09$02.00/0

Mapping of RNA-

Temperature-Sensitive Mutants of Sindbis Virus:

Complementation Group F Mutants Have Lesions in nsP4

YOUNGS. HAHN,' ARASH GRAKOUI,2 CHARLES M. RICE,2 ELLEN G. STRAUSS,' AND

JAMES H.

STRAUSS'*

Division of Biology, CaliforniaInstitute of Technology, Pasadena, California 91125,1 andDepartment of Microbiology and

Immunology, Washington

University,

St. Louis, Missouri 631102

Received 24 October 1988/Accepted 28 November 1988

Temperature-sensitive (ts)mutantsof Sindbis virusbelongingtocomplementationgroupF, ts6,tsllO, and ts118,aredefectivein RNAsynthesisatthenonpermissivetemperature. cDNA clones of thesegroupFmutants, aswellasofts+revertants, have beenconstructed. To assign thetsphenotype toaspecific regioninthe viral

genome,restriction fragmentsfromthemutantcDNA cloneswereused toreplacethe corresponding regions

ofthefull-lengthcloneTotol101 of Sindbis virus. Thesehybridplasmidsweretranscribed in vitroby SP6 RNA polymerasetoproduceinfectious transcripts, andthe virusrecoveredwastested fortemperature sensitivity. After thetslesion of eachmutantwasmappedtoaspecific regionof 400 to 800nucleotides by thisapproach,

thisregion ofthecDNA clonesofboth thets mutantandts+ revertantswassequencedin order todetermine

the precisenucleotide changeand amino acid substitution responsible for each mutation. Rescuedmutants,

which have a uniform background except for one or two defined changes, were examined for viral RNA synthesis and complementationtoshow that thephenotypesobservedwerethe resultof the mutationsmapped.

ts6andtsllOhadasinglebasesubstitution innsP4,resultinginreplacementofGlybyGlu atposition 153or position 324, respectively.It is of interestthat nsP4contains theGly-Asp-Aspmotifcharacteristic ofanumber ofviralreplicases, andthis, togetherwith the fact that all RNAsynthesis ints6-infected cellsand,toalesser

extent, in tsllO-infected cells shut off when the cells were shifted from a permissive to a nonpermissive

temperature, suggeststhatnsP4 isthe viruspolymerase.ts118was adouble mutant. It containedasinglebase

substitution in nsP2, resulting inreplacementofValbyAla atposition425 thatresulted in theformation of minute plaques, but not in a reduction in the plaque number atthe nonpermissive condition. The second change, asubstitution of Glnby Argints118atresidue 93 in nsP4, had littleapparentphenotype onitsown,

but incombination with thechangeinnsP2 led toatsphenotype. Thus,in eachcasethe mutationresponsible forthe temperaturesensitivityof the three knowncomplementationgroupF mutantslayin nsP4. Inaddition, theresult withtsll8suggeststhatnsP2and nsP4mayinteractwitheachother inacomplex.

Sindbis virusis a well-studied memberofthe alphavirus family. Its genome isa single molecule of plus-strand RNA 11,703nucleotides in length thatis capped at the 5' end and

polyadenylated at the 3' end (35). During replication, the

parental49Splus-strand RNA istranscribed into a comple-mentary minus strand which serves as a template for the

synthesis of both 49S plus-strandgenomic RNA and a 26S subgenomic RNA. Nonstructural polypeptides are trans-latedfromthe genomic 49S RNA as two polyprotein precur-sors that are processed by cotranslational or

posttransla-tionalcleavage into fournonstructuralproteins, callednsPl,

nsP2, nsP3, and nsP4, which are required for RNA

replica-tion (11). Three structural polypeptides are produced by

processing of a polyprotein precursor translated from the

subgenomic26S mRNA.

Large numbers oftemperature-sensitive (ts) mutants of the HR strain of Sindbis virus have been isolated and characterized (3, 30, 34). Mutants may be defective in RNA

replication (RNA- mutants) or in the production of the structuralproteins

(RNA'

mutants) andhave been grouped

by complementation into four RNA- groups (A, B, F, and

G) and three RNA' groups (C, D, and E) (4, 30). Repre-sentative mutant-revertant pairs from RNA' groups have been analyzed by sequence analysis, and there is an excel-lent correlation between specific nucleotide changes and phenotypes (1, 10, 15). None of the RNA- mutant defects

*Corresponding author.

has been rigorously assigned to specific nonstructural

pro-teinsorRNA sequences.Thesemutantspresumably contain

tslesions in the viral nonstructural proteins whichfunction to replicateviral RNA.

A full-length cDNA clone of Sindbis virus has been

constructed that can be transcribed in vitro by SP6 RNA

polymerasetoproduceinfectiousfull-length transcripts (22).

Viruses produced from in vitro transcripts are identical to Sindbis virus and showstrain-specificphenotypesreflecting

the source ofRNAusedforcDNA synthesis (17, 22). This full-lengthclonecanbe usedtostudyinterestingphenotypes

ofSindbis virus.

Wehaveusedthisapproachtodefineprecisely the

muta-tions responsible for the ts phenotypes of Sindbis virus

complementation

groupF mutants. Mutants ts6, tsllO, and tsll8 of complementation group F are defective in RNA

synthesis atthenonpermissivetemperature. The

best-char-acterized member, ts6,ceases all viral RNAsynthesis after a shiftfrom permissive to nonpermissiveconditions, and it

has been postulated that ts6 has a defect in theelongation activity of the replicase (2, 14, 26). cDNA clones of these mutants,aswellasof ts+ revertants, have been constructed, and restriction fragments from the mutant cDNA clones were used to replace the corresponding regions of a

full-length clone of Sindbis virus. These plasmids were tran-scribed in vitroby SP6 RNA polymeraseto produce infec-tioustranscripts, which were then testedfor ts phenotype. The viruses recovered from these transcripts have been

1194

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

characterized.Together with sequence analysis of the cDNA

clones, these experiments have defined the mutations re-sponsible for the group F mutants.

MATERIALSAND METHODS

Virus stocks, growth, and purification. Mutant ts6,

ob-tained originally from B. Burge, was isolated from the HR strain of Sindbis virus following mutagenesis with

nitroso-guanidine(3). Mutants tsllO and sll8 were isolated froma

small-plaque strain ofHR Sindbis virus following

mutagen-esis with nitrous acid (30). Revertants were isolated by

plaquingmutantstocksat 30 and40°C. Asinglevirus plaque

of a ts+ revertant was picked from the 40°C plate, and the

virus was eluted into 1 ml of Eagle medium containing 5% fetal calf serum. This revertant plaque was used to infect primary chicken cells at 40°C, and the resulting stocks, following plaqueassay at 30and40°C,wereused asinfecting stocks forRNApreparation. Viruseswere grown inprimary

orsecondary chicken embryo fibroblastsandharvested10 to 20 h afterinfection, depending on the mutant. Mutants and revertants were grown at 30 or 40°C, respectively. Viral RNAwas isolated as described before (24).

cDNA cloning. cDNA

synthesis

forts6, tsllO,and

tsll8,

as well as for their revertants, followed the procedure of Okayama and Berg (20). A

primer complementary

to a sequence nearthe startcodonof

capsid protein (nucleotides

[nt]

7642 to 7661ofthegenomic RNA) wasmade;this

primer

also contained the

recognition

site for XbaI restriction

endonuclease at its 5' end. This

primer

was used for first-strandsynthesis,andsecond-strand

synthesis

wasdonewith EscherichiacoliDNApolymerase I,E.coliRNase

H,

andE.

coliDNA ligase.

Phosphorylated

EcoRIlinkerswere

ligated

to the double-stranded cDNA to facilitate later

digestion

with XbaI (see below). The double-stranded cDNA was

divided into two

portions

for

cloning.

For the 5'

library,

the AccI (nt83)-SacII (nt2771)

fragment

of the

cDNA,

which encodes nsPl and the N-terminal half of

nsP2,

was cloned into Kahn5. Kahn5 isa

plasmid

containing

acDNA copyof

the 5' terminus ofthe Sindbis virus genome in Proteus

1,

a vector consisting ofthe replicon and

P-lactamase

genes of pBR322 and an SP6 RNA

polymerase

promoter

(22;

H. V. Huang and C. M. Rice,

unpublished).

The 3'

library

was

constructed by cloning the

BglII

(nt2268)-XbaI (nt7662)

fragment ofthecDNA,

encoding

the C-terminal half ofnsP2 and allofnsP3andnsP4, into

plasmid pMT21,

an

ampicillin-resistant

cloning

vectorderivedfrom

pBR322 (the

XbaI site is notpresent in this viral RNAbut was introduced

by

the

primer, as notedabove).

Construction of

hybrid

genomes.

Hybrid

genomes were

produced by

replacing

restriction

fragments

inSindbisvirus clone TotollOl

(22)

with the

corresponding

regions

from

cDNA clones derived from the mutantsor theirrevertants

(17). Details of restriction sites used are included in the

figure legends.

Full-length

hybrid

plasmids

that contained

oneofthree

nonoverlapping

intervals

(A,

B,

and

C)

fromthe mutantssubstituted intoTotollOl werefirstconstructed for

gross mapping. Plasmids with interval A contained the sequence of the mutant from the

SspI

(ntSO4)

to the ClaI (nt2713)site inTotollOl.

(TotollO1

contains

approximately

13,638

nucleotides;

numbering

begins

fromthe first nucleo-tideoftheSindbis virus

genome.)

Since the

SspI

site is not

unique, ashuttle vector,

irnsP12

(S.

A.ChervitzandC. M.

Rice,

unpublished),

containing

the Sacl

(ntl3552;

a site in the vectorupstreamofthe SP6

promoter)

toEcoRV

(nt2750)

region of

TotollOl

cloned in

IAN7

(18)

was

digested

with

SspI (nt504)

and ClaI

(nt2713)

and

ligated

with the

corre-sponding fragment

of 5' cDNA of the ts mutant. The

SacI-ClaI

fragment

ofthe

resulting

clone was then cloned into TotollOl which had been cut with Sacl and ClaI and treated with calf intestinal alkaline

phosphatase.

Interval B

plasmids

contained the sequence ofthe mutant from ClaI

(nt2713)

to

SpeI

(nt5262)

inTotollO1 andwasconstructed

by

replacing

this

fragment

inTotollO1 with the

corresponding

fragment

fromthe3' cDNA

library

ofthetsmutant. Interval

region

C

plasmids

contained the

SpeI

(nt5262)

to AatII

(nt7999)

region

of the ts mutant in TotollOl and was

constructed

by digesting

the 3' cDNA

library

of the ts mutantwith

SpeI

(nt5262)

and BamHI

(nt7334)

and

cloning

into shuttle vector

TnsP34,

which is a

nAN7

derivative

containing

the PvuII

(nt5160)

to NcoI

(nt8038)

fragment

of TotollOl

(obtained

from H. V.

Huang).

The

SpeI

toAatII

fragment

of the

resulting

cloneswasthen usedto

replace

the

corresponding fragment

ofTotollOl.

For fine

mapping

of the B

region, plasmids

were

con-structed that contained three

overlapping

subregions

re-ferred to as

Bi,

B2,

and B3.

Subregions

Bi,

covering

the

region

ClaI

(nt2713)-AvrII

(nt4280),

and

B3,

covering

the AvrII

(nt4280)-SpeI

(nt5262) region,

were cloned

directly

into TotollOl.

Subregion B2,

covering

the AvaI

(nt3546)-BamHI

(nt4633)

region,

wasobtainedfrom theshuttlevector

Kahn5B,

consisting

of the ClaI

(nt2713)

to EcoRI

(nt5869)

fragment

of Sindbis virus subcloned into Kahn5. Three clones

containing

overlapping

subregions

Cl, C2,

and C3 were constructed for fine

mapping

of the C

region

with

lTnsP34.

Fragment

SpeI

(nt5262)-HindIII

(nt6267),

PstI

(nt5824)-HpaI

(nt6919),

orNsiI

(nt6461)-BamHI

(nt7334)

of

thetsmutantwascloned into

TrnsP34,

and the

SpeI

(nt5262)-AatII

(nt7999)

fragment

wasusedto

replace

the

correspond-ing

fragment

in

TotollOl.

In vitro

transcription

and transfection. RNA

transcripts

were

synthesized

in vitro with

SP6

RNA

polymerase,

using

supercoiled

plasmid template

or

plasmid

DNA

digested

with

the

appropriate

restrictionendonuclease forthe

production

of runoff

transcripts,

as described

previously

(22).

The

resulting

transcripts

were transfected into confluent

mono-layers

of

secondary

chicken cells in 35-mm multiwell tissue

culture

plates,

andthe

phenotype

oftherecoveredviruswas tested.

Plaques

were

quantitated by overlaying

the

monolay-erswith 2mlof1% agarose in

Eagle

medium

containing

2%

fetal calf serum, followed

by

incubation at 30 and

40°C.

Plaques

were visualized

by

staining

with neutral red or

crystal

violet after incubation for36to40 hat

40°C

orfor60

to72 h at

300C.

Analysis

of viralRNA

synthesis.

Chicken

embryo

fibroblast

monolayers (60-mm

plate)

wereinfected with Sindbis virus HRor tsmutantsorrecombinantvirusesrecoveredfrom the

hybrid

cDNA clones at a

multiplicity

of 50 PFU/cell in

phosphate-buffered

saline

(PBS)

(6)

containing

1%fetal calf

serumand

dactinomycin

(ActD)

(1

,ug/ml)

andincubated at 30or

40°C

for 1 h. At the end of the

adsorption

period,

the

inocula were

removed,

and the cells were washed with

warmed medium and incubated at 30 or

40°C

in

Eagle

medium

containing

3%fetal calfserumand ActD

(1

,ug/ml).

For theshiftfrom 30to

400C,

at3.5 h

postinfection

(p.i.)

one

set of

30°C

plates

was washed once with warmed medium

lacking

ActD;

warmed medium

containing

ActD

(1

,g/ml)

wasthen

added,

and the

plates

wereshiftedto

40°C.

At 10 h

p.i.

(30°C),

6 h

p.i. (400C),

or8 h

p.i.

(after

theshiftto

40°C),

cells wereharvested.

Theamountof viral RNA

present

was

quantitated

by

the

cytoplasmic

dot

hybridization

method ofWhite and Bancroft

on November 10, 2019 by guest

http://jvi.asm.org/

(3)

NONSTRUCTURAL PROTEINS

I _ I _ rl%et I n MA IL

ns P2

II

CIa Ava AvrII

2713 3546 4280

3,

I ns P3 I ns F4 _

IT I I I I I I

BamHISpe I Pst Hind IIINsi HpaI BamHI

4633 5262 5824 6267 6461 6919 7334

l Toto:ts6A, Toto:tsl1 OA

I Toto:ts6B, Toto:ts11 OB 7777I| Toto:ts6C, Toto:tsl 1OC

Toto:ts6Cl, Toto:tsl1OC1

Toto:ts6C2, Toto:tsl10C2 Toto:ts6C3, Toto:tsl10C3

Toto:tsl 18A Toto:tsl18B Toto:tsl18C Toto:tsl18B1 Toto:tsl 18B2 Toto:tsl 18 B3 Toto:tsl18BC Toto:tsl18BC1 Toto:tsl18 BC2 Toto:tsl18BC3

FIG. 1. Construction ofhybridgenomes.Aschematic of thenonstructural-protein-coding regionofSindbisvirus cDNAcloneTotollOl (22) is shown together with anumberof restriction sites and theirpositions (nt) inthe Sindbis virusgenome numbered fromthe 5' end accordingtoStraussetal.(33). Translated regionsareshownastheopenboxes,in which thenamesof thevariousproteinsareindicated. Nontranslated regionsareshownas asingle line. The hatched boxes show the location of restriction fragmentsincloneTotollOl thatwere replaced with thecorrespondingrestriction fragments from thetsmutantsortheirrevertants.Thenamesusedtorefertothesehybridclones

areshownattheright.

(37). EqualnumbersofcellswerewashedwithcoldPBSand then lysed with 1% Nonidet P-40in TE buffer(10mM Tris chloride[pH 7.5],0.1 mMEDTA).Nucleiwerepelleted,and the supernatantwastreated with 14.8%formaldehyde in 1x

SSC (150 mM sodium chloride, 15 mM sodium citrate) at 60°Cfor 15min.RNAsamples(10to20,ul)wereblottedonto the nitrocellulose membranes and probed with 32P-labeled minus-strand RNA from the region of thegenome (the 26S region) encoding the structural proteins. This probe was transcribedwith SP6 RNA polymerasefromacDNA clone

ofSindbisvirusthatcontained the structural protein region onlyinsertedinan invertedsense downstream froman SP6 promoter. Relative amounts of RNAwere determined bya beta-scanning counter. All results were corrected for the amount ofincorporation into mock-infected plates, which was between 0.5 and 1% of the incorporation into cells infectedwiththe parental strain of Sindbis virus.

Alternatively, RNA synthesis following the shift was

assayed byexamining the incorporation of [3H]uridine into infectedcells. Following infectionat30°C, cellswereshifted at3hp.i.to40°C andlabeled with [3H]uridine(20,uCi/ml) in thepresence ofActD from 3.5 to8 h p.i. Monolayers were

then washedwith PBS and lysed with 0.5 ml of 2% sodium dodecyl sulfate, 50 ,ul wasprecipitated with trichloroacetic

acid,and the incorporated radioactivity was quantitated by

liquid scintillation counting.

Complementation analysis. Complementation tests were

performed as described by Strauss et al. (30) but 35-mm multiwellplatesandamultiplicityofinfectionof 20 PFU/cell for each mutant were used. A complementation index was

calculated as theyieldfrom the mixed infection dividedby the sum of the yields following infection by each parent

alone. A complementation index was calculated separately foreach mutant intests in which the two mutants differed markedly in plaque size. The absolute magnitude of the complementation index isdependentontheyieldofparental viruses (i.e., the extent of leakage of the parents), as

complementationisalwaysinefficient(4, 5, 30), not exceed-ing1 to10%of the wild-typeyield, andin the case oftsll8

only one-way complementation couldbe demonstrated

be-causeofrelativelyhighyields of tsll8at40°C. RESULTS

Construction of recombinantplasmids. Inordertolocalize thetsmutations ofts6, tsllO,andtsl18,weconstructedand

tested a number ofrecombinant plasmids. The constructs

areillustrated inFig. 1. Ineachcasesmall(873 to 2,584 nt) 5'

Sspl 504

I

17,7777.77,771

V/77777TI

V,17122=1

I

on November 10, 2019 by guest

http://jvi.asm.org/

[image:3.612.75.552.79.414.2]

(4)

Constructs tested for phenotype nonpermissivetemperature

Recombinant Fragment Phenotypeb Location of clonea replaced(nt) mutation(nt)

Toto:ts6A 504-2713 wt 5824-6267

Toto:ts6B 2713-5262 wt

Toto:ts6C 5262-7334 ts

Toto:ts6Cl 5262-6267 ts

Toto:ts6C2 5824-6919 ts

Toto:ts6C3 6461-7334 wt

Toto:tsllOA 504-2713 wt 6461-6919

Toto:tsllOB 2713-5262 wt

Toto:tsllOC 5262-7334 ts

Toto:tsllOCl 5262-6267 wt

Toto:tsllOC2 5824-6919 ts

Toto:tsllOC3 6461-7334 ts

Toto:tsll8A 504-2713 wt 2713-3546,

Toto:tsll8B 2713-5262 (ts) 5824-6267

Toto:tsll8Bl 2713-4280 (ts) Toto:tsll8B2 3546-4633 wt

Toto:tsll8B3 4280-5262 wt

Toto:tsll8C 5262-7334 wt

Toto:tsll8BC 2713-7334 ts

Toto:tsll8BC1 2713-6267 ts

Toto:tsll8BC2 2713-5262, ts 5824-6919

Toto:tsll8BC3 2713-5262, (ts) 6461-7334

aSeeFig. 1.

bRNAtranscriptsweretransfectedontocellsat30or40°C,and the plaque titerwasdeterminedasdescribedin Materials and Methods.wt,Wildtype.

(ts),Partiallyts,inthatplaquesizebutnotplaquenumber is reducedat40°C.

restriction fragments in the Sindbis virus cDNA clone TotollO1, from which infectious RNAcanbetranscribed in vitro with SP6 RNA polymerase (22), were replaced with

cDNA from a mutant or its revertant. In this figure, the restriction sites used to construct the hybrid genomes and

theirnumbering from the 5' end of the RNA (33) are also

shown. The genomic region encoding the nonstructural proteins was first divided into three large nonoverlapping regions, A, B, and C, forgrossmapping. For fine mapping, regionsBand Cwereeachsubdivided into three overlapping

regions (Bi, B2, and B3 andCl, C2, and C3). The 5' 444 nt and the 3' 265 nt of the coding region for the nonstructural proteins, which were not covered by hybrid genome con-structions, were sequenced in each case to ensure that no changes hadoccurred within these regions.

RNAwastranscribedin vitro from therecombinant

plas-midswithSP6 RNApolymeraseandtransfectedonto mono-layers ofchicken cells. Monolayers were incubated under agarose at 30 or 40°C to determine whether the virus recovered ineach casewas temperature sensitive.

Localization of the mutations ints6, tsllO, andtsll8. The

results obtainedwith the constructs tested are summarized

in Table 1. For ts6,ofthe three large interval replacement clones (Toto:ts6A, Toto:ts6B, and Toto:ts6C) tested, plas-midsToto:ts6A andToto:ts6Bgaverisetovirus that exhib-ited wild-type growth at the nonpermissive temperature, while ts virus was obtained from plasmid Toto:ts6C. This localized thetsmutation tothe interval nt5262to nt7334of the genome. Plasmids Toto:ts6Cl, Toto:ts6C2, and Toto: ts6C3, containing three smaller intervals in the C region, werethenconstructedandtested. Plasmids Toto:ts6Cl and Toto:ts6C2gaverisetotsvirus,whereasToto:ts6C3 didnot

TABLE 2. Plaquemorphology and RNA synthesis by tsll8at 40 and 30°C

Relative RNA Plaquesize Efficiency synthesisb

* ~~~~~~~~of

Virus plaquinga Shift

400C 300C (40OC/30°C) 400C 300C to 400C TotollOl Large Large 2.6 x 10-1 1.00 1.00 1.00 tsll8 Minute Small 5.0 x 10-5 0.07 0.70 0.32 Toto:tsll8Bl Minute Small 4.0 x 10-1 0.28 0.99 0.58 Toto:tsll8C Large Large 5.0 x 10-1 0.71 0.93 0.84 Toto:tsll8BC Minute Small 5.0 x 10-5 0.08 0.73 0.48 Toto:tsll8BC*c Minute Small 2.6 x 10-5 0.08 0.76 0.44 tsll8R Large Large 5.4 x 10-1 0.70 1.14 0.81

a Plaquetiter at400C dividedbythatat300C.

bViral RNAsynthesis (relativeto that byTotollOl)wasassayedby dot hybridization following infectionat 40 or 30°C orafterashiftto40°C following infectionat30°C,asdescribedinMaterialsandMethods.

'Toto:tsll8BC* containsthe Bfragmentfromtsll8and the Cfragment fromtsll8R.

(Table 1). Thus, ts6 has one or more mutations in the region between SpeI (nt5262) and HpaI (nt6919); if only a single mutation is involved,itmustlie in the region of overlap ofCl andC2, between PstI (nt5824) and HindIII (nt6267), which is

located neartheN terminus of nsP4.

FortsllO, virusrecovered from plasmid Toto:tsllOC was ts, whereas virus from Toto:tsllOA and Toto:tsllOB was not. Plasmids Toto:tsllOCl, Toto:tsllOC2, and Toto:

tsllOC3werethenconstructed and testedfor fine mappingof

the intervalregionC oftsllO. ts viruses were obtained from

recombinant plasmids Toto:tsllOC2 and Toto:tsllOC3,

whereas plasmidToto:tsllOCl gave rise to wild-type virus. Fromthisweconclude that tsllO has one or more mutations in theregion between nt5824and nt7334 of the genome and that ifa single mutation is involved, it must lie between nt6461and nt6919ofthe genome.This region is also in nsP4.

Mappingoftsll8 suggestedthatitwasadoublemutantin which one mutation was located in the nsP4 region, as was thecaseforts6andtsllO,and thesecondmutationwasina different region. Viruses from plasmids Toto:tsll8A and

Toto:tsll8C were apparently wild type, whereas that from

plasmid Toto:tsll8B was partially ts (Tables 1 and 2). This

partial temperature sensitivity manifested itselfasachange

inplaque sizefromsmallplaquesat 30°C tominute plaques

at40°C, althoughthenumberofplaquesat30and40°Cwere the same. RNA synthesis at 40°C was reduced relative to that in TotollOl virus (Table 2). When construct Toto:

tsll8BC was tested, the virus once

again

formed small

plaques at 30°C, whereas at 40°C minute plaques were

formed,with theplaquenumberbeingreducedby4ordersof

magnitude (Table 2), as wasthe casefor theparentaltsll8. Furthermore, RNA synthesis at40°C was reduced to very low levels(Table 2). Thus,weconcludethattsll8 isadouble mutant in which a mutation in the B

region

results in

formation ofminuteplaquesat

40°C

butnochangein

plaque

number and a second mutation in the C region,

although

havinglittle apparentphenotypeonits own, whencombined

with the change in the B region, results in red4ction of

plaquenumber(thusbeingscored as tsin

plaque

assays).

Fine mapping of the two tsll8 mutations was done

by

constructing Toto:tsll8Bl, -B2, and -B3

(Fig. 1),

and the

change

responsible

forformation of minute

plaques

at

40°C

mapped toregion

Bi

(Table 1).

Similarly,

the

change

in the C region

that,

when combined with the B

region change,

on November 10, 2019 by guest

http://jvi.asm.org/

(5)

H R

ts6

ts6R

5' NONSTRUCTURAL PROTEINS

-i

nsPl

I

nsP2

I

nsP3

I

nsP4

I I I I

1 2 3 4 5 6 7 (kb)

425 93 153 324

Val GlnGly Gly

I

I.I

I

U

2953

II I

A G G

6046 6226 6739

3'

Gly - Glu

A Gly G

Gly-_ Glu

ts1I 10 I

--A Gly ts11 0 R

---ValI-_ Ala t s118

C

ts1 18R

Gln-*_ Arg

G

Val GIn-_ Arg

G

U G

FIG. 2. Localization oftsmutations.Aschematic of thenonstructural-protein-coding regionof Sindbis virus is shown. Belowareshown sequencing schematics forHRSindbis virus (33),theparentalstrain from which thetsmutants wereisolated(34),and formutantsts6, tsllO, andtsll8andtheirrevertants. Sequencedregionsareshownassolid lines. Anychangefrom theHR sequenceonthe first line is indicated. Where nochangesareshown, the sequence is identicaltothatofHR.Nucleotidesarenumbered from the 5' end of theRNA;amino acids arenumbered fromthe Nterminus ofeachprotein.

resulted in reduction in plaque count was mapped to the

region ofoverlap in Cl and C2 between nt5824 and nt6267. Sequence analysis of ts6, tsllO, and tsll8 and their rever-tants. Inordertodefine thetslesionsofts6,tsllO,and tsll8,

regionsshownby themapping experimentstocontainthets lesions were sequenced by the chemical method ofMaxam and Gilbert (19) as modified by Smith and Calvo (29), with cDNAclones ofmutants aswellasof theirrevertants. The sequences obtained are shown inFig. 2.

ts6hadasingle base substitutionin theregion sequenced. Comparingthe ts6 sequence with that of its revertant, and

from the results in Table 1, we found that the mutation responsible for temperature sensitivity in ts6 was a change of GtoAatnt6226,which led to the replacement of Gly (GGG) atposition 153 ofnsP4by Glu (GAG). Inthe revertant, the

changed nucleotide reverted to the original nucleotide, re-storing the parental amino acid. tsllO also had only one

changethroughout the sequenced region. The change was G toA at nt6739, resulting in the change of Gly (GGG) to Glu

(GAG) atposition 324 of nsP4. In the tsllO revertant, this

nucleotidereverted to the original nucleotide.

tsll8 had asinglebasesubstitution in the region between nt2713 and nt3546. The change was U to C at nt2953,

resulting in the change from Val (GUG) to Ala (GCG) at

position 425 of nsP2. As discussed earlier, this change

resulted inreductioninplaque sizebutnotnumberat

40°C.

Inthe tsll8 revertant,this nucleotide revertedtothe

original

nucleotide.

Thesecondchangeints118wasfoundtobeachange ofA toGatposition 6046, resultinginaGlntoArgsubstitutionat

position93of nsP4. Thischange, combinedwith thechange

innsP2, rendered the virus tsin thatplaquenumberaswell asplaquesizewerereducedat40°C.Therevertantselected, tsll8R,retained thechangeinnsP4, which,asnoted

earlier,

had little apparent phenotype on its own. An additional

construct was made and tested in order to show that this

change was in fact

responsible

in part for temperature

sensitivity. The tsll8RC region was combined with Toto: tsll8BtoproduceToto:tsll8BC*, whichupon assaywasts,

formingtiny plaques at40°C in reduced numbers (Table 2).

Characterization ofthe tsmutations rescuedfromts6,tsllO, and tsll8. As a control to establish that the mutations

mapped and defined here were theonesresponsible for the

phenotypesdescribedpreviouslyfor these mutations, and in order to establish the phenotype of these mutations in a uniform background, recombinant viruses containing a de-fined region from each of the ts mutants in a TotollOl backgroundwerestudied.Monolayersweretransfected with dilutions of RNA transcribed from recombinant plasmids Toto:ts6C2, Toto:tsllOC2,

Toto:tsll8Bl,

Toto:tsll8C,

on November 10, 2019 by guest

http://jvi.asm.org/

[image:5.612.94.521.71.417.2]

(6)

[image:6.612.51.293.95.207.2]

TABLE 3. Efficiency ofplaqueformation and RNA synthesis by ts6 and tsllOat40and 30'C'

Titer Relative RNA

(PFU/ml) Efficiency synthesis

Virus ofplaquing Shift

400C 300C (40OC/30°C) 40°C 30'C to

400C

TotollOl 5.2 x 108 2.0 x 109 2.6 x 10-1 1.00 1.00 1.00 ts6 2.0 x 106 1.4 x 109 1.4 x 10-3 0.04 0.38 0.02 Toto:ts6C2 2.7 x 105 8.0 x 109 3.4 x 10-5 0.02 0.46 0.05

tsllO 4.0 x 104 3.0 x 109 1.3 x 10-5 0.05 0.72 0.16

Toto:tsllOC2 1.2 x 104 2.1 x 109 5.7 x 10-6 0.05 0.64 0.12

aSeeTable2, footnotesaand b.

Toto:tsll8BC, and Toto:tsll8BC* and incubated at 30and 40°C, withresults similartothose shown in Table 1.Asingle plaque of each viruswas isolated from the 30°Cplate and a

stockgrownat30°C, and these plaque-purified virus stocks

were characterized further. To start, titers of these

recom-binantviruses weredeterminedat 30and 40°C to ascertain the relative efficiency of plaque formation, and the results areshown in Tables 2 and 3. The virus stocks derived from

these infectious transcripts clearly showed thetemperature sensitivity of the parentalmutantin thecaseofts6andtsllO,

and theefficiencyofplaqueformationat40°C relativetothat at 30°C was low (Table 3) (the relatively high apparent reversionrateof ts6 is duetouseofastock thathadnotbeen recently plaque purified; during passage, revertants are am-plified inmost tsmutant stocks). However, asnotedearlier,

the virus recovered from Toto:tsll8Bl formed minute plaques at40°C, and the number of PFU was only slightly reduced (Table 2). The amino acid change of Valto Alain nsP2 isthusresponsible for the minute-plaque phenotype at

thenonpermissivetemperaturebutnotfor thereduced

num-berofplaques.ThemutationinnsP4 (constructToto:tsll8C) had littleeffectonitsown,butincombination with the nsP2 change(constructsToto:tsll8BC andToto:tsll8BC*) it led toapronounced decrease in theefficiency of plaque

forma-tionat40°C (Table 2).

To examine these viruses further, RNA synthesis was analyzed afterinfectionat30°C, afterinfection at40°C, and at40°Cfollowingashift from30°C.Theparentalmutantsts6, tsllO,and tsll8wereincluded,as wasvirusrecovered from

cloneTotollO1. TotalviralRNAsynthesiswasanalyzed by

thecytoplasmic hybridizationmethodofWhiteandBancroft (37) using 32P-labeled minus-strand RNA transcribed from the structural protein region as aprobe. Thevalues deter-mined, relative to those forTotollO1 virus, are shown in Tables 2 and 3. The synthesis of minus-strand RNA was significantlylessthan thatofplus-strand RNA,andtherefore the amount ofplus-strand RNA detected by hybridization wasassumedto bethetotal viralRNA. Followinginfection at 40°C, RNA synthesis by the viruses recovered from Toto:ts6C2 andToto:tsllOC2wasreducedtoalevel similar to that seen after infection by their respective parents.

However, the virus recoveredfromToto:tsll8Bl showed a higherlevelof RNAsynthesisthan theparentaltsll8 atthe nonpermissivetemperature(Table 2).Thisisconsistent with theobservationthat the mutation in nsP2givesrise tominute plaques at 40°C but does not reduce plaquing efficiency. RNAsynthesis by virus fromToto:tsll8C, containingonly thechangeinnsP4,wasonly slightlyreducedat40°Crelative to that by TotollO1 virus and was the same as that by tsll8R. However, RNA synthesis by the double mutants Toto:tsll8BC and Toto:tsll8BC* at 40°C was low and

TABLE 4. Complementation analysis' Complementation index

Virus Group A GroupB Group G GroupF (ts24) (tsll) (tsl8) ts6 tsllO tsll8

Toto:ts6C2 53 48 55 0.4 b

Toto:tsllOC2 67 42 42 3 0.3

-Toto:tsll8B1 12 7 7 4 0.2

Toto:tsll8BC 16 104 4 1 0.2

aComplementation indicesshown are theyieldfrom mixedly infected cells divided by the sum of the yields from singly infected cells. For tsll8 constructs,thecomplementation indicesare one way, asdescribedin Mate-rialsand Methods.

b-, Notdetermined.

identical to that of the parental tsll8, and thus the nsP4 mutationatposition93 markedly reduced RNA synthesisat thenonpermissive temperature whenit was combined with

the nsP2 change.

Themutantsand the constructs werealso tested for RNA

synthesis afterestablishing infectionat30°C andshifting to 40°C, for comparison with the results of Keranen and Kaariainen (14) and Sawicki et al. (26) for ts6 (see also

reference2). ts6 makes verylittle RNA after a shift to 40°C, asfoundby Sawickietal. (26), and thevirus recoveredfrom

Toto:ts6C2 exhibited the same phenotype (Table 3). tsllO also made little RNA after a shift to 40°C, as did Toto:

tsllOC2(Table 3). Fromamoredetailedstudyofthekinetics of cessation of RNA synthesis in ts6-infected cells after a

shift up (14, 26), and from studies of ts6 replication com-plexes in vitro (2), it was concluded that the elongation of

RNA chains, as opposed to initiation, was ts in ts6, and it waspostulatedthat the F groupfunction, here shown to lie in nsP4, definedthe RNA polymerase. Thetsll8 constructs

such as Toto:tsll8BC made significant amounts of RNA

after a shift, much more RNA than when infection and

incubationweredone at40°C continuously(Table2).Thus,

thereplicase complexes of tsll8, onceformedat 30°C,are

active afterashift to 40°C,in contrast to those specified by

tsllO and ts6.

It is worth noting that the assay used for these shift experiments examines total virus plus-strand RNA in the infectedcellsat8 hp.i. afterashiftat3.5 h. The results make clear that the RNA present at 3.5 h p.i. at 30°C does not

contribute

significantly

tothe RNApoolat8h,althoughthe

replicase enzymes needed for an essentially full yield of virus RNA are present. As a control for these results, we

repeatedthe shift experiments and examined labeled RNA made between 3.5and 8 h p.i. in the presenceofActD and

[3H]uridine.

The results were similarto those

presented

in Tables 2and 3.

Complementation analysis of rescued mutants. We also

examined the ability ofthese virusesto

complement

repre-sentative ts mutants from the three other

complementation

groups of RNA- mutants. The

complementation

indices shown in Table4demonstrate that theviruses derivedfrom

Toto:ts6C2

andToto:tslIOC2complementedtheotherthree

complementation

groups of RNA- mutants, as did the

parental viruses, but did not

complement

their

parental

viruses or each other, in agreement with

tsll8B1

and Toto:tsll8BC is

complicated

because ofthe high titer of virus

produced

at

40°C (the

virus from Toto:

tsll8B1

was only

marginally

ts, asnoted

earlier,

and virus

on November 10, 2019 by guest

http://jvi.asm.org/

(7)

1200 HAHN ET AL.

from Toto:tsll8BC also leaked at 40°C compared with the other mutants studied). In these cases thecomplementation indices shown are one-way indices. Because viruses from Toto:tsll8Bl and Toto:tsll8BC formed minute plaques, it waspossible to distinguish these plaques from thoseformed by the other ts mutants used. The complementation indices shown are the yield of large plaques in mixed infection divided by the yield of the large-plaque parents during single infection. Even so, complementation by Toto:tsll8Bl was marginal, and we cannot assign it to a complementation group, although it did seem tocomplement all of the other mutants tried. Analysis of the doublemutant Toto:tsll8BC showed that it did not complement its parental virus or ts6 andonly poorly complemented ts18 in groupG,but

comple-mented better the A mutant ts24 andquite well the Bmutant tsll. Thus, except for the marginal complementation of Toto:tsll8BC with tsl8, the resultswith therescued muta-tions are in good agreement withpreviouscomplementation

results (30).

DISCUSSION

Inthis report we have localized themutationsresponsible for the ts phenotype ofcomplementation group F mutants. The mutations in ts6 andtsllOwere mapped to nsP4. Each of them had a single base substitution, resulting in replace-ment of Gly by Glu at positions 153 and 324 of nsP4, respectively. Analysis of these mutations in a TotollO1

backgroundin avariety of ways,including RNA synthesisat

40°Candabilitytocomplement other RNA- mutants, dem-onstrated thatthese mutationsare infactthose responsible forthe tsphenotype and the mutantphenotypes previously

described in the literature. tsll8 turned out to bea double mutant. It had a defect in nsP2 (a Val to Ala change at

position425)whichonlypartiallydisabled itat40°C,

result-ing in production of small plaques and reduced RNA

syn-thesis, although the plaque numberwas unchanged. A sec-ond mutation in nsP4 combined with this change in nsP2 resulted in true temperature sensitivity, in that the plaque titer as wellas plaque size was reduced at 40°C and RNA synthesis after infection at 40°C was reduced to the level

characteristic of RNA- mutants. It is of interest that this

change in nsP4 alone had little apparent phenotype, but

becauseit is the one responsible for temperature sensitivity,

leadingtoreduced plaque titer in combination with the nsP2

change,thedouble mutant ts118 complements ts6 andtsllO

asagroup F mutant. Thus, mutations in nsP4 lead to group Fcomplementation behavior.

The experiments of Fuller and Marcus (8) in which the complementation groups were ordered by the relative rate of UVinactivationof their ability to complement gave an order for the complementation groups of NH2-G-A-B/F-COOH. Mutations in B and Fcould not be precisely localized, but both occurred downstream of group G and group A. Thus, the UV mapping data are consistent with our results that group F mutants contain defects innsP4. However, mapping of the other complementation groups of Sindbis virus ts RNA- mutants, currently being carried out, indicates that the Fuller and Marcus (8) order is not correct (Y. S. Hahn, unpublished observations).

Keranen and Kaariainen (14) and Sawicki et al. (26)

demonstrated that ts6 ceased genomic, subgenomic, and

minus-strand RNA synthesis upon ashift from the permis-sive to nonpermispermis-sive temperature and postulated that there wasa ts lesion in the elongationcomponent of the replicase. Recently, Barton et al. (2) extended such studies to in vitro

SIN SF RR ONN MID

SIN SF RR

ns P4 E:ts6

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~I

170

NYPTVASYQITDEYDAYLDMVDGTVACLDTATFCPAKLRS ---SDS---R---C ---SES ---R---C ---V---SES---R---N-S----

---T---SES---R-A---S----E:tsll0

ns P4 t 336

PGTKHTEERPKVQVIQAAEPLATAYLCGIHRELVRRLTAV

---N--

---K--ONN

---N--MID

---N--ns P2 A:tsll8

SIN QKVNENPLYAITSEHVNVLLTRTEDRLVWKTLQGDPWIKQ SF ---PA---A---V RR

ONN

---PS---N---S---V ---P---GK-T----S---I

ns P4 R:tsll8 106

;* 106~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

SIN KVENQKAITTERLLSGLRLYNSATDQPEC YKITYPKPLY SF ----M--TVVD--T--A---TG-DVGRIPT-AVR--R-V-RR ----M--VIID--KD-A-T-LTEQSEKIPT-VSK--R-V-ONN ----M--TIIH--KE-C---LASETPRVPS-RVT--A-I-MID ----M--EVID--LG-AK-FVTP-TDCRY

VTHKH---M-FIG. 3. Aminoacidsequencesofproteins from fivealphaviruses in theregionnearthe group F mutations.Sequencedataarefrom the followingsources: Sindbis virus(SIN), Straussetal. (33); Semliki

Vorest

virus(SF), Takkinen (36);RossRivervirus (RR),Faragheret al.(7)andStraussetal.(31);O'Nyong-nyong virus (ONN), Strauss

etal.(31)andunpublished data; Middelburg virus (MID), Strausset al.(32).

studies of replication complexes isolated from ts6-infected cellsandconcludedthatsuch transcription complexeswere ts inelongation. In contrast,othermutantsexaminedretain activity at 40°C ifreplication complexes areformed first at

30°C (14). Theseresults, togetherwith theresultspresented here, suggest that nsP4 is an RNA polymerase or the

elongation component of the alphavirus replicase. In this regard, it is of considerable interest that nsP4 contains the

Gly-Asp-Aspsequenceandsurrounding hydrophobicamino acids that have been found to be present in the replicase

proteins of several other RNA viruses (13, 23).

Themutations in ts6,tsllO, and tsll8 are shown in Fig. 3, in which the sequences of up to five alphaviruses are

compared in the regions affected. It is readily evident that theglycines affected in nsP4 of ts6 and tsllO are conserved in allalphaviruses sequenced to date and that each is found within a long stretch of highly conserved amino acids. In view of this conservation in sequence, it is evident that

changes in the sequence, such as in ts6 and tsllO, might affect function. The nsP4 polypeptide is composed of 610 amino acids and is onaverage 71 to 74% conserved among

alphaviruseswhich havebeen examined (31). However, this

conservation is not uniform throughout the protein; amino acids1 to125from theNterminus(including the location of thechangeintsll8)are lesshighlyconserved, as are amino J. VIROL.

on November 10, 2019 by guest

http://jvi.asm.org/

[image:7.612.322.560.75.399.2]

(8)

acids 550 to 604 near the C terminus. The canonical se-quence Gly-Asp-Asp, flanked by

hydrophobic

amino

acids,

is found at residues 464 to 466. This sequence, which has been found in a number ofRNA-dependent RNAreplicases

(13; see alsodiscussioninreference23),is wellseparated (on

thelinear sequence) from either the ts6 or tsllO lesion. The change ofVal to Ala in nsP2 of tsll8 is found in a domain thatis wellconservedamongalphaviruses, although conservation is not absolute. In

particular,

the Val affected in ts118 is not totally conserved, being replaced by Thr in

O'Nyong-nyong virus. The Gln to Arg

change

in nsP4 of ts118affectsanamino acidin adomainthatis notconserved (Fig. 3). The Gln is

replaced by Gly, Glu, Pro,

or

Asp

in

different alphaviruses. Thus, the failure of this

change

to have much effect by itself is perhaps not

surprising.

The

pronounced effect of this substitution in combination with

thechangein nsP2 suggests that nsP2 and nsP4 mayinteract

to form a functional complex, although other

explanations

for such a synergistic effectare

possible.

In Sindbisvirus, translation ofnsP4

requires

readthrough

ofanopal termination codonsothat nsP4is

underproduced

relativetonsPl, nsP2, and nsP3(11,

16, 32).

Theactive form ofnsP4 may be the

polypeptide

nsP34, which accumulates

during

Sindbis virus

infection,

whereas little or no nsP4 is detected. The

finding

that nsP4 may be the viral RNA

polymerase, basedonresultswith

complementation

group F mutants and the presence of motifs within this

protein

that are shared with other RNA

polymerases

ofanimal viruses

(13), is then reminiscent ofthe control ofvirus

replicases

in

other systems in which

readthrough

ofaterminationcodon

is required to produce the

polymerase.

In tobacco mosaic virus,readthrough ofanambercodonis

required (9, 21),

and

the

readthrough

portion

ofthe

protein

is

homologous

tothat

ofthe Sindbis virus

protein (12).

Similarly,

in the

retrovi-ruses, translation of the reverse

transcriptase

requires

readthrough ofanambercodonorframe

shifting

toeliminate

an amber codon(25, 27,

28).

Wepresume that

regulation

of the amount of RNA

polymerase

produced

is

important

during

theviral life

cycle.

Ashasbeen

pointed

out,

however,

Semliki Forest virus and

O'Nyong-nyong

virus lack this

termination codon(31,

36),

and

regulation

ofthe

activity

of thepolymerase seems tobedifferent forthese

alphaviruses.

ACKNOWLEDGMENTS

The expert technical assistance ofE. M. Lenches is

gratefully

acknowledged. We thank H. V. Huang and S. A. Chervitz for plasmid DNAs.

This work was supported by Public Health Service grants Al 10793, Al 20612, and Al 24134 from the National Institutes of Health. C.M.R. is aPewscholarin the biomedical sciences.

LITERATURE CITED

1. Arias, C., J. R. Bell, E. M.Lenches, E. G. Strauss,and J.H. Strauss.1983.Sequenceanalysisoftwomutantsof Sindbis virus defective in theintracellular transport of their

glycoproteins.

J. Mol. Biol. 168:87-102.

2. Barton, D.J., S. G.Sawicki,and D. L. Sawicki. 1988. Demon-stration in vitrooftemperature-sensitive

elongation

of RNA in Sindbisvirusmutantts6. J. Virol. 62:3597-3602.

3. Burge, B. W., and E. R. Pfefferkorn. 1966. Isolation and characterization of conditional-lethal mutants of Sindbis virus. Virology 30:204-213.

4. Burge, B. W., andE. R. Pfefferkorn. 1966.

Complementation

betweentemperature-sensitivemutants of Sindbisvirus. Virol-ogy30:214-223.

5. Burge, B. W., and E. R. Pfefferkorn. 1967.

Temperature-sensitive mutants of Sindbis virus: biochemical correlates of

complementation.

6. Dulbecco, R., and M.

Vogt.

1954. One step

growth

curve of

Western

equine encephalomyelitis

virusonchick

embryo

cells grown in vitroand

analysis

of virus

yields

from

single

cells. J.

Exp. Med.99:183-189.

7. Faragher, S. G., A.D. J. Meek, C. M. Rice, and L. Dalgarno. 1988. Genome sequences ofa mouse-avirulent and a

mouse-virulentstrain ofRossRivervirus.

Virology

163:509-526. 8. Fuller, F. J., and P. I. Marcus. 1980. Sindbis virus. I. Gene

order oftranslation in vivo.

Virology

107:441-451.

9. Goelet, P.,G.P.Lomonssoff,P.J.G.Butler,M. E.Akam,M.J.

Gait,andJ.Karn.1982.Nucleotidesequenceof tobaccomosaic virusRNA. Proc. Natl. Acad. Sci. USA 79:5818-5822. 10. Hahn,C. S., E.G.Strauss, andJ. H.Strauss. 1985.

Sequence

analysis

of threeSindbis virusmutants

temperature-sensitive

in the

capsid

autoprotease. Proc. Natl. Acad. Sci. USA 82:4648-4652.

11.

Hardy,

W.R.,andJ.H.Strauss.1988.

Processing

the

nonstruc-tural

polyproteins

of Sindbis virus:

study

ofthekineticsinvivo

by

using

monospecific

antibodies. J. Virol. 62:998-1007. 12. Haseloff, J.,P. Goelet, D. Zimmern,P.

Ahlquist,

R.

Dasgupta,

and P. Kaesberg. 1984.

Striking

similarities in amino acid sequence among nonstructural

proteins

encoded

by

RNA

vi-ruses that have dissimilar

genomic

organization.

Proc. Natl.

Acad. Sci. USA81:4358-4362.

13. Kamer,G.,and P. Argos. 1984.

Primary

structural

comparison

of

RNA-dependent polymerases

from

plant,

animal and bacte-rial viruses. Nucleic AcidsRes. 12:7269-7282.

14. Keranen, S., and L. Kaariainen. 1979. Functional defects of RNA(-) ts mutants of Sindbis and Semliki Forestviruses. J. Virol.32:19-29.

15.

Lindqvist,

B.H.,J.DiSalvo,C. M.Rice,J.H.Strauss,and E. G. Strauss. 1986. Sindbis virus mutant ts2O of

complementation

group E contains a lesion in

glycoprotein

E2.

Virology

151:

10-20.

16. Lopez,S.,J.R.Bell,E.G.Strauss,andJ.H.Strauss. 1985. The nonstructural

proteins

of Sindbis virus as studied with an

antibody

specific

for the C terminus of the nonstructural

readthrough

polyprotein. Virology

141:235-247.

17.

Lustig,

S., A. C. Jackson, C. S. Hahn, D. E. Griffin, E.G.

Strauss, and J.H. Strauss. 1988. Molecular basis of Sindbis virusneurovirulence inmice. J. Virol. 62:2329-2336.

18. Lutz, C. T., W. M. Hollifield,B.Seed, J.M. Davie,and H. V.

Huang. 1987.

Syoinx

2A: an

improved

X

phage

vector

designed

for

screening

DNA libraries

by

recombination in vivo. Proc. Natl. Acad. Sci. USA84:4379-4383.

19. Maxam,A. M.,and W.Gilbert. 1980.

Sequencing

end-labeled DNAwith

base-specific

chemical

cleavages.

Methods

Enzymol.

65:499-560.

20.

Okayama,

H., and P.

Berg.

1982.

High-efficiency cloning

of

full-length

cDNA. Mol. Cell.Biol. 2:161-170.

21. Pelham, H. R. B. 1978.

Leaky

UAG termination codon in tobacco virus RNA. Nature

(London)

272:469-471.

22. Rice, C. M., R. Levis, J. H.Strauss, andH. V. Huang. 1987. Production of infectious RNA

transcripts

from Sindbis virus cDNA clones:

mapping

oflethalmutations,rescueofa

temper-ature-sensitive marker, and in vitro

mutagenesis

to generate definedmutants.J. Virol.61:3809-3819.

23. Rice,C.M.,E.G.Strauss,andJ.H.Strauss. 1986.Structure of the flavivirus genome, p. 35-90. In S.

Schlesinger

and M.

Schlesinger

(ed.), The

Togaviridae

and Flaviviridae. Plenum

Publishing Corp.,

NewYork.

24. Rice, C. M., andJ. H. Strauss. 1981.

Synthesis,

cleavage

and

sequenceanalysisof DNAcomplementarytothe 26Smessenger RNAofSindbisvirus. J. Mol. Biol. 150:313-340.

25. Rice,N.R.,R.M.

Stephens,

A.Burny,andR. V.Gilden. 1985. The gag and pol genes of bovine leukemia virus: nucleotide sequenceand

analysis. Virology

142:357-377.

26. Sawicki,D. L.,S. G.Sawicki, S. Keranen,and L. Kaariainen. 1981.

Specific

Sindbis virus-coded function for minus-strand RNA

synthesis.

J. Virol. 39:348-358.

27. Schwartz, D. E., R. Tizard, and W. Gilbert. 1983. Nucleotide sequenceofRous sarcomavirus. Cell32:853-869.

on November 10, 2019 by guest

http://jvi.asm.org/

(9)

28. Shinnick, T.M., R. A. Lerner, and J. G. Sutcliff. 1981. Nucle-otide sequence of Moloney murine leukaemia virus. Nature

(London)293:543-548.

29. Smith,D. R., and J. M. Calvo.1980.Nucleotidesequenceofthe

E. coligenecodingfor dihydrofolatereductase. Nucleic Acids

Res. 8:2255-2274.

30. Strauss, E. G., E.M. Lenches, andJ. H.Strauss. 1976. Mutants ofSindbis virus. I. Isolation and partial characterization of89

newtemperature-sensitive mutants. Virology74:154-168. 31. Strauss, E. G., R. Levinson, C. M. Rice, J. Dalrymple, and J. H.

Strauss. 1988. Nonstructural proteins nsP3 and nsP4 ofRoss Riverand O'Nyong-nyong viruses: sequence and comparison

with those of other alphaviruses. Virology 164:265-274. 32. Strauss, E. G., C. M. Rice, and J. H.Strauss. 1983.Sequence

coding forthe alphavirus nonstructural proteins is interrupted by an opal termination codon. Proc. Natl. Acad. Sci. USA 80:5271-5275.

33. Strauss, E. G., C. M. Rice, and J. H. Strauss. 1984. Complete nucleotide sequence of the genomic RNA of Sindbis virus.

Virology 133:92-110.

34. Strauss, E.G., and J. H. Strauss. 1980. Mutants ofalphaviruses: geneticsandphysiology,p.393-426.InR. W.Schlesinger(ed.),

Thetogaviruses. AcademicPress, NewYork.

35. Strauss, E. G., and J. H. Strauss. 1986. Structure andreplication of thealphavirus genome, p. 35-90. InS. Schlesingerand M.

Schlesinger (ed.), The Togaviridae and Flaviviridae. Plenum Publishing Corp., New York.

36. Takkinen, K. 1986. Complete nucleotidesequenceof the

non-structural proteingenesof Semliki Forestvirus. Nucleic Acids Res.14:5667-5682.

37. White, B. A., and F. C. Bancroft. 1982. Cytoplasmic dot hybridization. Simple analysis of relative mRNA levels in multiple small cell or tissue samples. J. Biol. Chem. 257: 8569-8572.