Replication of mouse hepatitis virus: negative-stranded RNA and replicative form RNA are of genome length.

(1)

JOURNAL P. 487-492 0022-538X/82/110487-06$02.00/0

Copyright01982,American Society forMicrobiology

Vol.44,No.2

Replication

of Mouse

Hepatitis

Virus:

Negative-Stranded

RNA

and

Replicative

Form RNA Are

of

Genome Length

MICHAEL M.C. LAI,* CHRISD. PATTON,ANDSTEPHEN A.STOHLMAN

Departmentof Microbiologyand Departmentof Neurology,* UniversityofSouthernCaliforniaSchool of

Medicine,LosAngeles, California90033

Received 10 May1982/Accepted 16July 1982

Therearesevenvirus-specific mRNA speciesin mousehepatitis virus-infected

cells (Lai etal., J. Virol. 39:823-834, 1981). In this study, weexamined

virus-specific negative-stranded RNA to determine whether there are corresponding

multiplenegative-strandedRNAs. Intracellular RNAfrommousehepatitis

virus-infected cells was separated by agarose gel electrophoresis, transferred to

nitrocellulose membranes, and hybridized to positive-stranded genomic 60S

[32P]RNA. Onlyasingle RNAspecies ofgenomic size wasdetectedunder these

conditions.This RNA was negative stranded. Nonegative-stranded subgenomic

RNA wasdetected. Wealso studieddouble-stranded replicative-formRNAin the

infected cells. Only onereplicative-formof genomic size was detected. When the

double-strandedRNAisolated withoutRNasetreatment wasanalyzed, again only

one RNAspecies of genomic sizewasdetectable.Furthermore,mostof the

virus-specific mRNAs couldbe releasedfrom this RNA species uponheating. These

resultssuggestthatallofthe mousehepatitis virus-specificRNAs aretranscribed

from a singlespeciesofnegative-stranded RNAtemplateofgenomic size.

Mousehepatitis virus (MHV), acoronavirus,

contains a single, positive-stranded genomic

RNAwithamolecularweightof5.4x

106(7,

8).

This RNA can be divided into at least seven

genetic regions (5) whichencode three structural

proteins,pp5O,gp23, andgp90/180,andpossibly

severalnonstructuralproteins (11).Theinfected

cells contain six virus-specific subgenomic

mRNAs and a genomic mRNA (2, 5, 9, 13).

These mRNAs have a nested-set structure in

which the sequence of each mRNA is included

withinthe next-larger mRNA. Furthermore, the

sequenceofeach mRNAcorresponds tothe

3'-endofthegenomicRNAand extends into the 5'-endforadistance corresponding tothe size of

eachindividual mRNA species (5).Thus, all of

the mRNAsshare identicalsequences at the

3'-end but presumably contain different 5'-half

sequences. We have also found that all ofthe

mRNAs share at leastfive nucleotidesatthe

5'-endsandthatsomemRNAs contain an

oligonu-cleotide which is not present in the genomic

RNA(6). These datasuggest that the synthesis

of MHV mRNAs involves a complex

mecha-nism, possibly an unusual form of RNA splicing

whichtakesplace in the cytoplasm (6).

Recently, we have detected two RNA

poly-merase activities in MHV-infected cells, one

detected early (1 hpostinfection [p.i.]) and the

other detected late (6 h p.i.). These two

poly-meraseshave different enzymatic requirements

andproperties (1). Furthermore, the RNA

prod-ucts of the early polymerase are of

negative-strand polarity, compared with the genomic

RNA, whereas those of the late polymeraseare

of positive-strand polarity (P. R. Brayton,

M. M. C.Lai, and S. A. Stohlman,unpublished

data). Therefore, MHV RNA replicatesthrough

a negative-stranded RNA intermediate which

serves as the template for the synthesis of

mRNAs. Since multiple subgenomic mRNAs

havebeendetected,it is of interesttodetermine whether the negative-stranded RNA

intermedi-ateisasingle genomicspeciesorincludes

multi-ple subgenomic and genomic species

corre-sponding tothepositive-stranded RNA. In this

study, weexamined thisissuebydirectly

exam-iningthenegative-strandedRNAspeciesand the

double-stranded replicative-form (RF) RNA in

MHV-infectedcells. We found thatonlya

nega-tive-stranded genomicRNAintermediate could

be detected.

MATERIALS ANDMETHODS

Viruses and cells.MHVstrain A59(MHV-A59)was used throughout. In some experiments, strain JHM wasalso used. These two strains have beendescribed previously (8). Virus was grown in DBT cells in Dulbeccominimalessentialmedium(DMEM) supple-mentedwith1% fetal calf serum.

PreparationofIntrcellular virus-specificRNA. Prep-aration ofintracellular[32P]RNAfrom MHV-infected cells has been described (5, 6). Briefly, L-2 cellswere 487

on November 10, 2019 by guest

http://jvi.asm.org/

(2)

preincubated with DMEM containing 1/10 the normal concentration of phosphate and 0.5% fetal calf serum for 10 to 12 h before infection. After virus adsorption, the cells were incubated in phosphate-free DMEM containingtwice thenormalconcentrationofvitamins and amino acids, 2 ,ug of actinomycin D per ml, 1% dialyzedfetal calf serum, and 250 ,Ci of32Pi (ICN Pharmaceuticals) per ml. At 7 to 9 h p.i., the cells were chilledonice andlysedwith a buffer containing 10 mM Tris (pH 8.5), 60 mM NaCl, 1 mM EDTA, and 0.5% Triton N-101. Nuclei were removed by centrifugation at1,800xg for 5min. Sodium dodecyl sulfate (SDS) wasadded to the supernatant fluid to a final volume of 1% SDS,and thesupernatantfluid was thenextracted withchloroform-phenol (1:1). The RNA was precip-itatedwith 2 volumes of ethanol.

Forpreparation of unlabeledintracellularRNA, the same protocol wasfollowedexcept that theinfected cellswere grown inunmodifiedDMEM supplemented with 1% fetal calf serum throughout infection.

Preparation ofdouble-sandedRNA. The intracellu-lar [32P]RNAextractedas described above was precip-itated by sedimentation at 15,000 xg for 15min.The RNA was resuspended in NTE buffer (0.1 M NaCl, 0.01 MTris-hydrochloride [pH7.4],0.001 MEDTA). NaClwas added toobtain a finalconcentrationof 2 M, and the solution was left at 0 to4°Cfor 48 h. The RNA was then sedimented at 13,000 rpm for 30min. The supernatantwasdiluted with an equal volume of water and precipitated with 2 volumes of ethanol. For diges-tionwithRNase, the RNA waspelleted at 12,000 rpm for15 minand suspended in low-salt buffer (0.01 M Tris-hydrochloride [pH 7.4], 0.001 M EDTA), and SSC buffer was added to the solution to a final concentrationof 2x SSC (1xSSC is 0.15 MNaCl plus 0.015 Msodium acetate). The RNA was digested with RNase A (20,ug/ml)at37°C for 60 min. The RNA was then extracted with phenol-chloroform (1:1) and pre-cipitatedwith 2volumesof ethanol.

Agarosegelelectrophoresisof RNA.Electrophoretic analysis of RNA was generally performed in 1% agarose gelsmade in REbuffer,pH 8.1(50 mM boric acid,5mMsodiumborate,1 mMEDTA, and 10 mM sodium sulfate (5, 6), except when RNA was to be transferredtonitrocellulose filters (seebelow). Elec-trophoresiswas at90 Vfor5.5 h.After electrophore-sis, the wet gel was wrapped with cellophane and autoradiographed.

Forextraction ofRNAfrom the agarose gel, the agarose fractioncontainingRNAwashomogenizedin abuffer containing0.3 M NaCl, 0.01 M Tris-hydro-chloride [pH7.4],0.001MEDTA,and0.5% SDS and continuously agitatedat roomtemperature for 12 h. After the gel pieces were removed, the RNA was

precipitatedwith 2 volumes of ethanol.

RNA blottingand fflterhybridization. Intracellular RNA was analyzed by a modification of the filter hybridizationmethod of Thomas (15). Briefly, intra-cellular virus-specific RNA was denatured in 1 M

glyoxal-50%odimethylsulfoxide(DMSO)-10mM sodi-umphosphate buffer (pH 6.8)at50°Cfor1h(10).The RNA was then electrophoresed ina1% agarosegel made in 10 mMphosphate buffer(pH6.8) at90 Vfor 5.5 h.Afterelectrophoresis,the RNAwastransferred to nitrocellulose membrane filters by the procedure described by Thomas (15). Hybridization was per-formed in10mlofbuffercontaining50%oformamide,

5x SSC, 50 mM sodium phosphate (pH 6.8),250,ugof sonicated, denatured salmon sperm DNA per ml, 0.02%each bovineserumalbumin, Ficoll, and poly-vinylpyrrolidone,250 ,ugof rRNA perml,10% dextran sulfate, and 5 x 106 cpm of 60S MHV [32P]RNA (specific activity, S x 106 to 10 x 106 cpm/,ug). Hybridizationwasperformedat42°Cfor 3days. The RNAblots were washed with fourchanges of room-temperature 2x SSC containing 0.1% SDS and then with two changes of50°C 0.1x SSCcontaining 0.1% SDS. They were thenexposed to Kodak XRfilm at -70°C with anintensifyingscreen.

RESULTS

Characterization of negative-stranded RNA in

MHV-infected cells. We first characterized the

negative-stranded RNA in MHV-infected cells

by a direct approach. The total cytoplasmic

RNA of MHV-A59-infected L-2 cells was

ex-tracted at 7to 9 hp.i., when the virus-specific

RNA synthesis was at its maximum level (5).

Thetotal RNAwasthendenatured with

DMSO-glyoxalby the procedure of McMasterand

Car-michael(10)andseparated byelectrophoresison a1%agarosegel. Aparallelcontrol experiment with MHV-A59-specific intracellular [32P]RNA analyzed under identical conditions revealed

that there were atleast six virus-specific RNA

species (Fig. 1, lane A). These species have been

extensively studied and represent

positive-strandedgenomicandsubgenomic virus-specific

mRNAs (5, 6). The unlabeled RNA was

trans-ferred from the agarose gel to a nitrocellulose

membrane and hybridizedto60S[32P]RNA

ex-tracted from the purified MHV-A59 virion (7).

The viral genomic RNA contains

only

positive-stranded RNA (7, 8); therefore, this 2P-labeled

probewould detectonly negative-strandedRNA

species. Onlyone RNA band, correspondingin

size tothe viriongenomic RNA, was detected

(Fig. 1, lane B). Therewas occasionally some

faint radioactivity in the lower part of the gel;

however, it was not reproducible and did not representdiscrete RNAbands.

Thisobservationsuggeststhat there isonlya single negative-stranded RNA species in the

MHV-infectedcells and that it has

approximate-ly thesamesizeasthatof thegenomic RNA.We

couldnotruleoutthepossibilitythat therewere

very small amounts of negative-stranded

sub-genomic RNA; however, this possibility seems

veryunlikely, sincethe amountof smaller

posi-tive-stranded subgenomic mRNAs far exceeded

that of genomic RNA (Fig. 1, lane A) (4, 9).

Therefore, our data suggest that all the

subge-nomic and genomic mRNAs are transcribed

from a negative-stranded genomic RNA

tem-plate. Since the RNA band detected by this

procedurewasbroad (Fig.1, lane B), wecould not ruleout the presence of several species of

negative-strandedRNAwithverysimilar

molec-ularweights.Alternatively,theheterogeneity of

http://jvi.asm.org/

(3)

MOUSE HEPATITIS VIRUS REPLICATION 489

A

B

--2

--3

-45

--

-'6---7

--FIG. 1. Electrophoretic analysis o

positive- and negative-strandedRNAs cellular[32P]RNA from MHV-infecte' denatured and separatedby electropt

agarose gel. After electrophoresis, tl exposed to Kodakfilm directly. Lan intracellular RNAwasdenatured with

(10) and separated by electrophoresis lane A. After electrophoresis, the R ferredtoanitrocellulose filter andhyl genomic [32P]RNA by publishedproce filterwas washed, dried, andexpose

withanintensifyingscreen.

solublefractionwassmallsingle-stranded RNA

which could not be precipitated with high salt

concentrations. Similar observations have been

madefor the RF RNA of influenza virus(3).The

solublefractionwasdigested with RNase A(20

,ug/ml)

in 0.3 MNaCltoremove

contaminating

single-stranded RNA speciessowecouldstudy

the RF RNA. The resulting RNase-resistant

RNAwasthenanalyzedbyelectrophoresison a

1%agarosegel. This RNAwasasinglespecies

with an electrophoretic mobility only slightly

slowerthan that of the genomic RNA (Fig. 2,

lane A). This RNAwasextractedfrom thegel,

denatured with DMSO-glyoxal (10), and

ana-lyzed by agarose gel electrophoresis. A single

genomic RNA specieswasdetected(Fig. 2, lane

B). The electrophoretic mobility of this RNA

was slightly faster than or sometimes

indistin-ooc guishable from that of the double-stranded RF

_-LO RNA (data not shown). It was unclear which

secondarystructure rendered theseRNAs

elec-trophoretically similar. Atthe bottomof thegel,

there was some faint dispersed radioactivity

which might represent the remaining portions,

after RNase A digestion, of the various RNA

species bound to the genomic RNA (compare

with Fig. 3, lane C; see below). This result

showed thatthedouble-strandedRFRNAin the MHV-infected cells consisted of a single

genomic species. Although it was still possible

f MHV-specific thatsomedouble-stranded subgenomic RFRNA

celnls

wAas

heat

might

have a

configuration

which rendered it

ioresis on a1% inseparable from the genomic

RF

RNA under

he wet gel was our electrophoretic condition, this possibility e B, Unlabeled seems quite unlikelyin view of the fact thatthe

glyoxal-DMSO size of the smallest RNA is only-1 thatof the asdescribed for genomic RNA (5, 9, 13). This result, together LNA was trans- with the result obtainedfrom the filter

hybridiza-bridized toviral tionanalysis of negative-stranded RNA,

strong-tuo

Koda)

film

ly

suggeststhat the

template

RNAfor all of the

MHV mRNAs is a

single

negative-stranded

ge-nomic RNA species.

Characterizationofapartfally double-stranded

the RNAscouldhave beenanartifact of

extrac-tion,assimilarheterogeneitywasnoticed for the

positive-stranded genomic RNA(Fig. 1, lane A).

Characterization ofdouble-stranded RF RNA.

To confirm the interpretation derived from the

analysis of negative-stranded RNA, we studied

the double-stranded RF RNA involved in the

synthesis ofMHV mRNA.Virus-specific

intra-cellular [32P]RNA was precipitated with 2 M

NaCl at0to4°C for48h. The double-stranded

RNA in the soluble fraction was found to be

enriched(Table 1). However, 24% of this

frac-tion at most was RNase resistant. This value

was consistently obtained in repeated

experi-ments. The reason for the failure to obtain a

higherpercentageof double-strandedRNAwas

notclear. Presumably,therestof the 2 M

NaCl-TABLE 1. RNaseresistance of various forms of MHVintracellular RNA'

Sourceof RNA _resistance% RNase

2MNaCl(soluble) 24

2MNaCl(precipitate) 2

RI RNA(Fig. 2, lane A) 100

RI RNA(Fig. 3, lane B) 61

RNAno. 6(Fig. 3, lane C) 9 RNAno. 7(Fig. 3, lane C) 9

a RNA was dissolved in 0.5 ml ofasolution contain-ing0.3 MNaCl, 0.03 M sodium acetate, 0.01 M Tris-hydrochloride (pH 7.4), 1 mM EDTA, and 20 Fg of RNase Aper ml and incubatedat37°Cfor 60 mnu. The amountof RNase-resistant RNAwas determined by precipitation with trichloroacetic acid.

VOL. 44,

http://jvi.asm.org/

[image:3.491.47.240.75.374.2] [image:3.491.250.447.524.629.2]

(4)

the majority of the MHV-A59-specific mRNA

species became detectable (Fig.3, lane C) (5, 6),

suggesting that the majorityof the mRNAswere

associated with a genomic RNA complex. Among the heat-dissociated RNAs,thegenomic RNA and the subgenomic RNAs no.4and no. 5 were not distinct. Thereasonfor the failure to

detect distinct bands of these RNAspecieswas

notclear. The mostlikely explanation wasthat

these RNAs were degraded during isolation.

RNAsno.1andno.4 wereusuallymoredifficult

todetect undermostoftheanalyticalconditions

(5). This resultsuggeststhatatleastsomeof the

virus-specific mRNAs in MHV-infected cellsare

bound to genomic RNA complex which might

represent anRI.Themolecularweight ofthis RI

would be difficult to determine because the

secondarystructureof the RNA wouldinterfere

with electrophoretic mobility under these

condi-tions. Thisobservation and the results obtained

from the study ofnegative-stranded RNA

sug-A

B

C

FIG. 2. Electrophoretic analysis of double-strand-ed RF RNA. Lane A, Double-stranded RF RNA (arrow) isolated from MHV-infected cells and then digestedwith RNasewasanalyzed byelectrophoresis on1%agarosegelasdescribedin thelegendtoFig.1. LaneB, TheRNA band in lane Awaseluted from the gel,denatured withglyoxal-DMSO, andanalyzed by

agarose gel electrophoresis (arrow) as described for

laneA.

RI. To better understand the RNA replicative intermediate (RI) involved in the synthesis of MHVRNA,wefurtheranalyzed the RNA in the

2 MNaCl-soluble fraction without RNase treat-ment.Anysingle-stranded RNA partially bound

tothe negative-stranded RNA could be

identi-fiedelectrophoretically. This RNA fractionwas

analyzed by agarose gel electrophoresis. There

wasonlyoneRNAspecies detected underthese conditions, and it was in the region of the gel

correspondingtothegenomic RNA (Fig. 3,lane B). This RNA hadamigrationratesimilartothat of RNase-treated RF RNA (Fig. 3, lane A). Again, itwasunclear whichsecondary structure rendered these RNAspecieselectrophoretically indistinguishable. No discrete subgenomicRNA

was detected even when no RNase digestion was performed. When this RNA species was

heatdenaturedandanalyzed byelectrophoresis,

[image:4.491.86.210.79.378.2]

-*7

FIG. 3. Electrophoretic analysis ofpartially dou-ble-stranded RI RNA. RNA obtained fromthe 2 M NaCI-soluble fraction without RNase digestion was

analyzed by electrophoresis on a 1% agarose gel. Double-stranded RF RNA (as described for Fig. 2, laneA)wasusedas acontrol(lane A).LaneB,Native RI RNA. LaneC,RI RNAheatedat100°Cfor 1min beforeelectrophoresis.

A

B

_ i

_b

.. _

.

-_ _:

g

w

:^

... ..

.:

:.. :: ss s

I

http://jvi.asm.org/

[image:4.491.303.401.307.599.2]

(5)

MOUSE

gest that this RI consists of various mRNA

species associatedwith anegative-stranded ge-nomic RNA.

To further understand the structure of the

various RNA species detected in the

MHV-infected cells, we determined the RNase

resist-anceof each RNA species after the species was

elutedfromtheagarose gel. The

double-strand-ed RF RNA wascompletely resistant to RNase

digestion (Table 1). In contrast, RNA isolated

directlyfromthe 2 M NaCl-solublefractionwas

only 61% resistant to RNase. This result

con-firmed thehypothesis that the latter form

repre-sented an RNA RI which contained

single-stranded RNA tails. In contrast, the RNA

species releasedfrom the RIby heat treatment

were sensitive to RNase digestion, consistent

with the idea that they represented

single-stranded RNAspecies.

DISCUSSION

From thestudy of thevirus-specific

negative-stranded RNA and double-negative-stranded RF RNA in

MHV-infected cells, we concluded that there

was only one negative-stranded genomic RNA

speciesand thatthis species servedasthe

tem-plate for the synthesis of multiple

positive-stranded mRNAs. Therefore, thereplication of

MHV RNA canbe summarized asin

Fig.

4. In

this scheme, the positive-stranded genomic

RNAfromincomingvirus is firsttranscribedby

the early RNA polymerase (1) into a

negative-stranded genomic RNA. This process

presum-ably involvesfaithfulcopying ofallof the

geno-mic sequences, except for the

3'-polyadenylic

acid sequences. However, no details of the

structureofthe negative-stranded RNA,except

foritssize, arecurrentlyknown. Thenextstep inreplication isthe synthesis ofmultiple

geno-mic and subgenomic mRNAs. This step

appar-ently involves transcription

by

the late RNA

polymerase (1) ofthefull-length

negative-strand-ed RNA. This isaverycomplexstep, and very

few details about itarecurrentlyknown.

Tran-scriptional mapping data have shown that the

UV targetsizeofeach mRNA is thesame asits

physical size (4),suggestingthat each mRNAis

transcribed individually at its own initiation

pointrather that beingderived fromagenomic

RNA. Furthermore, as we have shown previ-ously,thesynthesisofsubgenomicmRNAsmay

involve some forms ofRNAprocessing(6). One

of these processes could be involvement ofRNA

splicing (6; unpublished data). Itisnotclear how

and whenRNAprocessingoccursduringmRNA

synthesis.

The second unanswered question is whether

all of the mRNAsaresynthesizedfrom thesame

negative-strandedRNAmolecule. Thefindingin

(i.)

5,

U

early polymerase

5'3'

.10

I

3959 3'

3p5'

5' 3' 35 53

5' 3' FIG. 4. Proposed scheme for MHV RNA replica-tion.This scheme assumes that nascent RNAstrands areimmediately displaced byother newlysynthesized strands, so that there are multiple single-stranded RNA tails in the RI. The shorter positive-stranded RNAtails complementary to the 5'-end of the nega-tive-strandedRNAtemplatein the RI representnewly synthesizedsubgenomic mRNAs. The solid rectangles atthe 5'-ends of the positive-stranded RNAs represent the presumptive leader sequences (6; unpublished data). The number of RNA strands in the RI is arbitrarily assigned. No evidence has been obtained that the nascent positive strands contain the leader sequences. (a) Semiconservative mechanism for mRNAsynthesis; (b)conservative mechanism.

this study that no double-stranded subgenomic

RF RNA was detected after RNase digestion

suggests that the entire negative-strandedRNA

template is protected by binding to mRNAs.

Thus, any singlemolecule of negative-stranded

template RNA is probably used for the synthesis

of more than one mRNA species and also of

genomicRNA(Fig.4,model a). It is notknown,

however, whetherallof themRNAs are

synthe-sized on the same RNAtemplate

simultaneous-ly. Analternative possibility is that the

MHV-specific negative-stranded RNA is always

present as adouble-stranded RNA. In that case,

mRNAS would be synthesized by a

conserva-tive mechanism (Fig. 4,model b) as opposed to a

semiconservative one (model a). At the present time, our data did not enable us to determine

whether the negative-stranded MHV RNA was

everpresent as freemolecules. Our data also did

notenable us to rule out thepossibility that there

were very small amounts ofnegative-stranded

http://jvi.asm.org/

[image:5.491.259.447.73.310.2]

(6)

subgenomic RNA. However, if this was the

case, the rates of synthesis of the smaller

mRNAs must have been very fast, since the

amount of smaller subgenomic RNAs far

ex-ceeded thatofgenomicRNA.

This replication scheme raises an interesting

question: how is the synthesis of various

mRNAs regulated? It hasbeen shown that the

smaller mRNAs (no. 7 and no. 6) arefar more

abundant than the rest of the mRNAs (4, 5, 9).

Furthermore, the relative rate of synthesis of

most of mRNAs is constant throughout viral

replication (9).Therefore,there mustbe a

mech-anism which regulates the frequency of tran-scription of each mRNA species on the

negative-stranded RNA template.This remains one of the

outstanding unansweredquestions regarding the

replication scheme of MHV RNA.

It should be noted that ourfindingof a single

negative-stranded genomic RNA is

compati-blewithpreviously reported UV transcriptional

mapping data (4). In that study, UV light was

administered toMHV-infected cells at6 h p.i.,

at which time synthesis of negative-stranded

RNA should have already been completed.

Therefore, the dataobtained in that study

per-tain only to the synthesis of positive-stranded

RNA. Our present findings also suggest that if

RNA splicing or another modification was

in-volved in MHV RNA synthesis, it most likely

took placeduring thesynthesis ofmRNA, rather

than during the synthesis of negative-stranded

RNA. Therefore, late RNA polymerase, rather than early RNA polymerase (1), may possess

properties which areresponsible for such RNA

processing.

Our present findings thus distinguish the

mechanism of RNA synthesis for coronavirus

from that for other positive-stranded-RNA

vi-ruses, e.g., togavirus and picornavirus. In the

case oftogaviruses, several genomic and

sub-genomicRF RNAshavebeen detected(12, 14).

There is asubgenomic RF RNAcorresponding

tothesubgenomic mRNAofalphaviruses.

Fur-thermore, the RI involved in the synthesis of

alphavirus subgenomic mRNAcontainsa

nega-tive-stranded genomic template which is

sensi-tive to RNase atthe site of initiation of

subge-nomic mRNA(12). Thisappeared not to be the

case with coronaviruses, which use a unique

mechanism ofRNA synthesis. Further

under-standing of coronavirus RNAsynthesiswill

re-quirethe study ofthe sequences ofthe mRNAs

and the detailedstructureofthe RI. Such

experi-ments are in progress.

ACKNOWLEDGMENTS

We thank Todd Lasman forexcellenttechnical assistance and Raymond Mitchell for assistancein manuscript prepara-tion.

This work was supported in part by grant PCM4507 from the National Science Foundation,grantRG1449-A-1from the National Multiple Sclerosis Society, andPublic Health Ser-vice grants NS 15079, AI 19244, and CA 16113 from the NationalInstitutes of Health.

LITERATURECITED

1.Brayton, P. R., M. M. C. Lai, C. D. Patton, and S. A. Stohlman.1982.CharacterizationoftwoRNApolymerase activities induced by mouse hepatitis virus. J. Virol. 42:847-853.

2. Cheley,S., V. L. Morris, M. J.Cupples,and R.Anderson. 1981. RNA and peptidehomology amongmurine cdrona-viruses.Virology 115:310-321.

3. Content, J., and P. H. Duesberg. 1971. Base sequence differencesamong theribonucleic acidsof influenza virus. J.Mol.Biol.62:273-285.

4. Jacobs, L., W.J.M.Spasan,M.C. Horzinek, and B. A. M. van derZeIjst.1981.Synthesis of subgenomicmRNA'sof mousehepatitis virusisinitiated independently: evidence fromUVtranscriptionmapping. J. Virol.39:401-406. 5. Lai, M. M. C., P. R. Brayton, R. C. Armen, C. D. Patton,

C.Pugh,andS. A.Stobhnan.1981.Mousehepatitisvirus A59: mRNA structure and genetic localization of the sequencedivergence fromahepatotropic strainMHV-3. J. Virol.39:823-834.

6. Lai, M. M. C., C. D. Patton, and S. A.Stohlman. 1982. Further characterization ofmRNA's ofmousehepatitis virus:presenceof common5'-end nucleotides. J. Virol. 41:557-565.

7. Lai, M. M. C., and S. A.Stobhlman.1978. RNA of mouse hepatitisvirus. J. Virol. 26:236-242.

8. Lai, M. M. C., and S. A. Stoblman. 1981.Comparative analysis ofRNA genomesofmousehepatitis viruses.J. Virol. 38:661-670.

9. Leibowltz, J.L., K. C.Wilbelmsen,andC. W. Bond. 1981. The virus specific intracellular species oftwo murine coronaviruses: MHV-A59 and MHV-JHM. Virology 114:39-51.

10. McMaster, G. K., andCarmichael,G.G.1977.Analysis ofsingle-anddouble-stranded nucleic acidson polyacryl-amideand agarose gels by using glyoxal and acridine orange. Proc. Natl. Acad.Sci. U.S.A. 74:4835-4838. 11. Siddell,S., H. Wege, A.Barthel,andV. terMeulen.1981.

Coronavirus JHM:intracellularproteinsynthesis.J. Gen. Virol. 53:145-155.

12. Simmons,D.T., and J. H. Straus. 1972.Replication of Sindbisvirus. II. Multiple forms of double-stranded RNA isolated from infected cells. J. Mol. Biol.71:615-631. 13. Spaan,W.J.M.,P.J. M.Rotter,M.C.Horznek,and

B. A. M.van derZeUst.1981.Isolationandidentification of virus-specific mRNAs in cells infected with mouse

hepatitis (MHV-A59).Virology 108:424-434.

14. Strauss,3.H., andE.G. Straws.1977.Togaviruses,p. 111-166. InD. P.Nayak (ed.),The molecularbiologyof animalviruses, vol.1MarcelDekker, Inc.,New York. 15. Thomas,P.s. 1980.Hybridizationof denatured RNAand

small DNAfragmentstransferredtonitrocellulose. Proc. Natl.Acad. Sci.U.S.A. 77:5201-5205.

http://jvi.asm.org/