Membrane Binding of Input Arbovirus Ribonucleic Acid: Effect of Interferon or Cycloheximide

(1)

Membrane Binding

of

Input

Arbovirus Ribonucleic

Acid: Effect

of

Interferon

or

Cycloheximide

ROBERT M. FRIEDMAN AND T. SREEVALSAN

Laboratory of Pathology, NationalCancerInstitute,NationalInstitutesofHealth,

Bethesda, Maryland 20014,and Department of Microbiology, SchoolofMedicine andDentistry,Georgetown University, Washington,D.C.20007

Received for publication 9 March 1970

By 1hr afterinfection, 36%of input Semliki Forest virus ribonucleic acid(RNA) which was cellassociatedwas found in amembrane structure. This structure had

many similarities to the membrane-associated replication complex (MRC) which had previously been identified in arbovirus infections. Interferontreatment didnot

affect theassociation of viral RNA with the MRCstructure,butcycloheximide treat-mentinhibited it.

In a previously published report, the fate of

input SemlikiForestvirus (SFV) ribonucleic acid

(RNA) wasstudied. Under normalconditions, a

portion ofthe42Svirion RNA (7) was, by 1 hr

afterinfectionofchick cells, alteredto 30 and 20S

viral RNA forms. The latter was ribonuclease

resistant. In cells treated with interferon before

infection, or with cycloheximide or puromycin

during

infection,

viral RNA probably was

un-coated,asevidencedby the formationof26S viral

RNA,aformneverfound in the virion, but little

or no 20S

ribonuclease-resistant

RNA was formed. It wasconcluded thatallthreetreatments

were similar in that they inhibited the

transcrip-tion ofinput viral RNA, thereby preventing the

formation of an RNA complementary tothe

in-put strand (6).

The results of a more detailed study are

re-portedinthispaper. The

input

SFVRNArapidly

became associated with a membranous

cyto-plasmicstructure which hadmany of the

proper-ties of a replication complex, a structure which hasbeen

previously

described inSindbisvirusand

poliovirus infections (9, 15). The replication

com-plex was membrane

associated,

as was that of

Sindbis virus

(15).

A difference wasnoted,

how-ever, in the effects of treatment with

cyclohexi-mide during the course of infection or with interferon before infection. In

cycloheximide-treated cells, entry ofviral RNA into the

mem-brane-associatedreplication complex (MRC)was

decreased, whereas in interferon-treated cells no

effectonthisstepwasnoted.

MATERIALS AND METHODS

Viruses. Pools of SFV and chick cell monolayer

cultures were prepared by previously described

methods (8). An SFVpoolwasprepared in the

pres-enceof 10,uCi/mleachoftritiatedadenosine (19Ci/

mmole) and tritiated uridine (20 Ci/mmole). The viruswas partially purified bythe methodofCheng

(1) with theaddition ofaterminal bandingina15to 60% sucrosedensity gradient.Radioactivitywas con-centratedintwofractions at adensity of1.20g/cm3. Infectivity [22 X 10"1 plaque-forming units (PFU)/ ml] washighest in these fractions. Theinfectivityto

radioactivity ratiowas2.4 X 10' PFU per count per min.Allreportedresultswereconfirmedatleastonce

with another pool ofpurified 3H-SFV and, in most

cases,withapool ofthecloselyrelatedSindbis virus. Infection procedure.Monolayers of chickcells

con-tainingabout 2.5 X 107cellswereincubatedat37C

for1 hrwith 0.5

;tg

ofactinomycinD perml and then

for 1 hrat4 Cwith 0.1 mlof3H-SFVat avirus:cell multiplicity of2:1. After this period, 5 mlofwarm

Eagle'smediumwith10%fetalcalfserum wasadded

for1hrat37C.

CeU fractionation. Theprocedureemployed forcell

fractionation was based on previously described methods (15). After 1 hr at 37 C, the infected cells

were washed five times with iced 0.85% NaCl and

scraped intoasmallvolumeofthesaline.After

sedi-mentation at800 X gin thecold,thecellswere

re-suspended in 1 ml of 0.01 M tris(hydroxymethyl)-aminomethane (Tris), pH 7.2, and 0.001 M

ethylene-diaminetetraacetic acid (EDTA) andallowedtoswell

for30minbeforehomogenizationinaDounce

homog-enizer. After homogenization, the cell extract was

centrifuged at 800 X g for 5 min, the fluidwas

re-moved, andthesedimentwasresuspendedin 1 mlof Tris-EDTA and again sedimented. The supematant fluidswere combined and sedimentedat 10,000 X g for 15min. The supernatant from this sedimentation

was the postmitochondrial fraction. The pellet was

gently resuspended in 1.0mlofTris-EDTA, and the differential sedimentation procedure was repeated.

The postmitochondrial supematants werecombined,

and thefinalpelletwasresuspendedin0.5mlof Tris-169

on November 11, 2019 by guest

http://jvi.asm.org/

(2)

EDTA. Thesuspensionswerethenlayeredovera

dis-continuous sucrose gradientin Tris-EDTA prepared

as follows: 0.4 ml of 50% sucrose, 0.4 ml of 45% sucrose,0.4 ml of 42.5% sucrose, 1.1 ml of 40%

sucrose, 1.1 mlof25% sucrose, and 1.1 mlof15%

sucrose. Sedimentation in the SW-50 head wasfor 30 min at 131,000 X g. The gradients were collected,

and radioactivity was estimated by previously

de-scribed methods (7).

RNA was extractedfromthe mitochondrial frac-tion by an sodium dodecyl sulfate (SDS)-phenol

methodpreviously described(7).Itwassedimentedin a6 to 30% sucrosegradient for 1 hrat300,000 X g

inanSW-65 rotor.

Interferon. Partially purified interferon was pre-pared anddonatedbyKarl Fantes (3). The

prepara-tion contained 10,000 international units of chick interferonand 179 ,ugofproteinper ml.

Incorporation of radioactivity. Incorporation of radioactivityfrom viral RNA intocell fractionswas

estimated by precipitation and washing with 2.5%

perchloric acid. The washed precipitates were then

solubilized with 0.33 N NaOH and counted. Protein

contentwasestimated bythemethodofLowryetal.

(10).

Sarkosylfractionation of the mitochondrial fraction. ThemethodofTremblayetal. (16) wasemployedto

isolate a cell membrane-viral RNA complex as

follows:2.5mlof15%sucrosein 0.01 M Tris(pH 7.2),

0.1MKCl,and0.01MMgCl2waslayeredover2.5ml

of40% sucroseinthe same

buffer,

and 0.1 mlof the

mitochondrial fraction was layered above the 15%

sucrose. Sodium lauroyl sarcosinate was added to

themitochondrial fractionto afinalconcentration of 0.1%,and the upperlayerwasgentlymixed.Thetube was immediately centrifuged in an SW-50 rotor at

51,000 X g for 30 min. The band of

magnesium-Sarkosyl crystals which collectedattheinterface ofthe sucrose solutions was harvested by puncturing the sideofthe tube.

RESULTS

Membraneassociatedinputviral RNA. Tostudy

the distribution of

radioactivity

after infection

with 3H-SFV, chickcell

monolayers

wereinfected

with a partially purified pool of 3H-SFV as

de-scribed. After1 hrat37C, thecells werewashed

and fractionated into nuclear and cytoplasmic extracts. Only 16% of the radioactivity applied

was adsorbed to the cells, even atthe low

mul-tiplicityofinfection and smallvolumeused(Table

1). Of theradioactivity associated with thecells,

about 90% waspresent in the cytoplasmic

frac-tion. Of this radioactivity, 36% was associated

with the mitochondrial fraction and 64%with the postmitochondrial fraction.

Whenthe mitochondrialandpostmitochondrial

fractions were analyzed in discontinuous 15 to

50% sucrosegradients(Fig. 1),mostofthe

radio-activity

present in themitochondrial fractionwas

[image:2.492.258.452.91.207.2]

found to sediment at a density of 1.18 (41%

TABLE 1. Distribution of radioactivity after infec-tion of chick fibroblasts with tritiated Semliki

Forest virus (3H-SFV)

Acid-pre-Determination cipitable_counts recovered

Total radioactivity added... 38,300

Washedoff cells... 31,200 Adsorbedto cell... 6,120

Nuclearfraction... 680

Cytoplasmicfraction... 5,300

Postmitochondrial fraction... 3,300

Mitochondrial fraction... 1,850

sucrose).Very little radioactivity was found at the

topof thegradient (Fig. 1A).On theother hand,

in the postmitochondrial fraction the reverse

situationwasseen. Most oftheradioactivity was

present at the top ofthe gradient, and a relatively minorfraction was present at its midpoint (Fig.

1B).

We were interested in thenature ofthe

radio-activity in the mitochondrial fraction and

sedi-mentationat adensityof1.18 since, later in

infec-tion, this region of the gradient in extracts of

Sindbis virus-infected cells has been shownto

con-tainan MRC with the replicative intermediate

form of viral RNAand the viral RNA-dependent

polymerase(15). The radioactivityofthis fraction

(Fig. IA) was probably not due to unecipsed

virus or to nucleocapsids with theirouter

mem-branecoatremoved. The peak of viral infectivity,

although notsharp, did notcoincide withthat of

the radioactivity (Fig. IA). Also, although the

uncoated viral RNA is completely ribonuclease

resistant, that of the fraction under study was

only partially resistant (Fig. 2). Moreover, when

artificial mixtures were prepared from the

cyto-plasm of SFV-infected cells and 3H-SFV,

3H-SFV core or nucleocapsid, or the purified viral

RNA forms (42, 26, and 20S RNA), only the

3H-SFV showed any tendency to adhere to the

mitochondrial fraction (Fig. 3); however, that

3H-SFV which did adhere tended to band at a

higher density than did the bulk ofthe

radioac-tivity examined in Fig. IA.Thenucleocapsidand

purifiedviral RNA formsremainedwiththe

post-mitocondrial fraction.

These resultssuggested to us that thebulk of

theradioactivity sedimentingwith the

mitochon-drial fraction was not due to

unecipsed

virus.

Extraction of radioactive RNA fromthis fraction

tended to confirm this notion (Fig. 4). The

ex-tracted RNA was polydisperse with a peak at

about 20S. It was partially resistant to

ribonu-cleasetreatment. OftheRNA forms ofSFV,these

properties are most similar to those of the

on November 11, 2019 by guest

http://jvi.asm.org/

(3)

U.

CL

2,

200

r-C)

mnr

CO

20=

c),

:0n 0rtl

mc~ O ou

0 10 20 30

[image:3.492.254.451.69.219.2]

FRACTiON NUMBER

FIG. 1. Analysisin a discontinuous sucrose density

gradient of cytoplasmic radioactivity after infection

with 3H-SFV. Gradient fractions from the (A)

mito-chondrial or (B) postmitochondrial fraction were

analyzed for radioactivity and, in the case of the

mitochondrial fraction, infectivity. The top of the gradient in this andsubsequentfiguresis to the right.

tially double-stranded replicative intermediate (RI) form (5). No significant amount of 42S RNA, the RNA of the virion,wasfound. These

results do notrule outthe possibility that some

26Ssingle-strandedviral RNAwaspresentinthe mitochondrialfraction since the method of

analy-sis employed does not separate this form

com-pletely from the RI form (7). The bulk ofthe RNA, however, would appear to be in the RI form.

Tostudy thekinetics of the entry of viral RNA into the MRC, RNA was extracted from the

mitochondrial fraction ofinfected cells atearlier

time periods. Cells were washed five times after

virusadsorptionat4C,andmediumat37 Cwas

added. Mitochondrialfractions wereprepared at

5 and 20 min afterwarming. RNAwasthen

ex-tracted from these fractions and analyzed on sucrose density gradients. The results (Fig. 5)

showed that 5 min after warming, a peak

ap-pearedat 42S, correspondingtothe RNA of the

virion. Some slower sedimenting material was

alreadypresent. At20minafterwarming,mostof

- 150 C

: f~~~rec/ed 0-0--0 X

0 10 20 30

FRACTION NUMBER

F1G. 2. Ribonuclease resistance of radioactivity associated with the mitochondrial fraction after infec-tion with 3H-SFV. A mitochondrial fraction was pre-pared as in Fig. 1. One-half of the fraction was sedi-mented as in Fig. 1, and the rest was treated with ribo-nuclease (1 /2g/ml, 20min, 0 C) before sedimentation.

2-.150 ~~I T

-

100

<vs

LI

3/-SSFV

"Core"

iooneot-oO-o

-0 10 20 30

FRACTION NUMBER

FIG. 3. Sedimentation in themitochondrial fraction of radioactivity .from artificial mixtures containing 3H-SFV or 3H-SFV cores or 3H-SFV-RNA forms (42, 26, or 20SRNA). Preparations containing these were mixed with a cytoplasmic extract from SFV-infected chick cellsafter homogenization. Mitochondrial fractions were then prepared as in Fig. 1 and analyzed on 15 to 50% discontinuoussucrose density gradients.

NVo

radioactivity was found to be associated with the mitochondrial fractions from the mixtures of cytoplasm with 3H-SFV cores (as shown) or 3H-SFV-RNAforms (notshown).

the labeled RNA was of the slower sedimenting species, probably, as at 1 hr after warming (Fig. 4), in the RI form. This result suggested that some 42S RNA rapidly became membrane-asso-ciated and was then converted to a replicating

RNAstructure.

Finally, additional results indicated that the radioactive fraction found in Fig. 1 was not due to uncoated 3H-SFV. In infection of chick cells with radioactive infectious RNA of the closely related Sindbis virus, the radioactivity of the

on November 11, 2019 by guest

http://jvi.asm.org/

[image:3.492.254.449.297.415.2]

(4)

28S 18S

/0

,O Rlbo ucleose-frea/ed

C'

1 .

ICP0

O/d E 00

i0o ,o OooX°> '

I T-'T -T

20 FRACTION NUMBER

FIG. 4. Analysis ofRNA extractedfrom the

mito-chondrial fraction of3H-SFV-infectedcells. Cellswere

infected with 3H-SFV, and after I hr mitochondrial

fractions were preparedasdescribed in the legendto

Fig. 1. RNA was extracted in 0.1 M NaCl by an

SDS-phenol methodandanalyzed (300,000X g,1hr)

on6to30%sucrosedenisity gradients.Thedesignations

18S and 28S specify the peaks ofadded ribosomal RNA markers. One-half ofthemitochondrialfraction

was treatedin0.1Am NaCl with I ,ug ofriboniuclease permlfor20minat37 Cbeforesedimentation.

z 100

0

28S I8S z 75

uJ

z

,0

8

2

50s

LU

z ₀ ₈ _H;S

[image:4.492.52.247.64.250.2]

FRACTION NUMBER

FIG. 5. Kinetics ofappearance ofviral RNAforms in the mitochondrialfractionzof 3H-SFV-infectedcells. Cellswereinfectedasdescribed inthelegendtoFig.4.

After 5 or 20 min, mitochondrial extracts were

pre-pared and RNAwasextractedfromthese.Analysisby sucrose density gradients wasalso carried out as

de-scribedinthelegendtoFig.4.Symbols: *,RNA

ex-tractedafter5min; 0,RNAextractedafter20 min.

mitochondrial fraction sedimented in the same mannerastheresult showninFig.IA(T. Sreeval-san,inpreparation). Also, analysisofthespecific activity of thevirus employed in theexperiment

shown in Fig. 1A (see above) demonstrated that

the titer of infectious virus recovered could not

account for the level ofradioactivity seen in the fractions containingthehighestlevels oftritium. Therefore,the association of viral RNAwith a specific cell fraction (Fig. IA) was probably not due toan artifactual interaction of viral RNA with cell membranesduringthepreparationof cy-toplasmic extracts since mixing of viral RNA forms with such extracts failed to cause such a

specific association. Additional evidence in favor

of the idea thatinputviralRNAwaspresentin an MRCwasdetermined in experiments employing

Sarkosyl magnesium precipitation ofmembrane

fractions.Ithas beendemonstrated that functional

bacterialdeoxyribonucleic acid (DNA) is associ-ated withamembrane and the complex may be

precipitated by magnesium with Sarkosyl (16).

Theprecipitation isspecificformembrane-bound functionalDNA,asDNAor RNAforms

artifici-ally mixed with cellextracts are notprecipitated.

When the mitochondrial fraction of cells infected

with3H-SFVwas precipitated with Sarkosyl and

magnesiumandtheprecipitate bandedat a15and

40% sucrose interface, most of theradioactivity

recoveredwasfound in theinterface band (Table 2). The counts present in the postmitochondrial

fraction or in artificial mixtures of viral RNA

forms and cell sap did not band atthe interface

whensubjectedtothesametreatment.

Theresults discussedsofarsuggested that, 1 hr

after productive infection with 3H-SFV, a small percentageof the

input

viral RNAwasspecifically

associated with a membrane component of the

mitochondrial fraction.By 1 hr afterwarming to

37C, mostof this RNA wasprobably inthe RI form. This membrane component had, therefore,

someof thecharacteristics ofaviralMRC. Effectofcycloheximideorinterferon on

associa-tionininputRNAwithcellmembrane. Tostudythe

effect of cycloheximide or interferon on input

viral RNA, chick cells were treated with 1,000

units/ml of

partially

purified chick interferon

beforeinfectionwith 3H-SFV, or, with 100

Ag

of

cycloheximide perml during infection. The dose

of interferon employedinhibited virusgrowthby

more than 200-fold; the dose of cycloheximide

inhibited protein synthesis by94%. The effect of

these treatments on the distribution of cyto-plasmicradioactivitywasstudied (Table 3). In the control cells, the distribution of cytoplasmic

radioactivitybetween themitochondrial and

post-mitochondrial fractionswassimilartothat shown in Table 1. When thespecificactivityofthe

radio-activity in these fractions was studied, however,

morethan2.5 times the activity of the postmito-chondrial fraction was present in the mitochon-drial fraction. Inthe cells treated with

interferon,

the distribution of

radioactivity

was

approxi-matelythesame asin thecontrol. Incells treated

120

0 100

Z 80

X 60

4z

(-- 40)

20

0

on November 11, 2019 by guest

http://jvi.asm.org/

[image:4.492.55.246.372.505.2]

(5)

TABLE 2. Distribution of radioactivity after sedi-mentation of a sodium lauroyl

sarcosinate-treated mitochondrial fraction

Acid-pre-Determination cipitable_counts recovered

Total radioactivity added... 1,055

Top component.. 90

[image:5.492.252.443.61.353.2]

Interface precipitate... 816

TABLE 3. Distribution ofcytoplasmic radioactivity after infection of chick fibroblasts with

tritiated Semliki Forest virus

Ratio of Total Specific specific radioactivity activities

fraction acti - (counts

(Mito-Cytoplasmic

frraction

aecotvit

pe min chondrial

(ecounret

_(counts/pepg fraction: of postmito-mm) protein) chondrial fraction)

Control

Mitochondrial ... 8,339 11.9 2.6:1 Postmitochondrial. 17,596 4.65

Interferon-treated

Mitochondrial... 8,425 14.0 2.4:1 Postmitochondrial. 13,443 5.85

Cycloheximide-treated

Mitochondrial... 3,288 4.50 0.66:1

Postmitochondrial.. 24, 726 6.82

withcycloheximide during infection, however, the bulk of the specific activity was in the

postmito-chondrialfraction.

These findingswereconfirmed insucrose

den-sity sedimentation studies on the mitochondrial

fraction (Fig. 6). Again, the control containeda

peak ofradioactivity sedimentingat adensity of

1.18 (Fig. 6A).Asexpected, the interferon-treated

cells contained a similar, but usually an even

sharper, peak, asshowninFig.6B. On theother

hand, in cells treated with cycloheximide during

infection, thepeak atthis densitywasvery much

decreased (Fig. 6C). Therefore, inputviral RNA

appearedto enterareplication complex (RC) -like structure after interferon treatment, butthis step

was inhibited in the presence ofcycloheximide.

Onepointof interestwasthenatureofthe viral RNA present in the RC-like structure in

inter-feron-treated cells. As shown in Fig. 4, under

ordinary conditions the predominant RNA

species present in the mitochondrial fraction of control cells after 1 hr was the RI form. It has

also been shown that, in cells treated with high

concentrations of interferon, little or no

ribo-nuclease-resistant RNA was formed from input

LdJ

a-z

n

z

0

U

A. Control

I

I10 20

FRACTION NUMBER

30

FIG. 6. Effectofinterferon orcycloheximide treat-ment on association ofinput 3H-SFV with

mitochon-drialfractionis. (A) Untreated cells; (B) cells treated

with 1,000 unitsof interferon permlfor 14 hrbefore

infection; (C)cellstreated with 100,ug ofcycloheximide

per ml during infectiont. The cells were infectedwithl

3H-SFV and, after I hr, mitochondrialfractions were

prepared and analyzed on discontinuous 15 to 50%

sucrosedensitygradients asdescribed in thelegendto

Fig. 1.

SFV RNA (6).Thepresenceof viralRNA inthe MRCstructureofinterferon-treated cells seemed

paradoxical. RNAwas,therefore,extracted from the mitochondrial fraction of interferon-treated cells. Theresults(Fig. 7)showthat themajor peak

of RNA presentwasin the 42S form, inmarked

contrast tothepredominant205 peakinthe

con-trol cells (Fig. 4). In addition, as expected, little or noribonuclease-resistant RNA could be

dem-onstrated in the mitochondrial fraction of the interferon-treated cells (Fig. 7). In this respect, the interferon-treated cells after 1 hr resembled

the control cellsveryearlyafterwarmingto37 C

(Fig. 5).

DISCUSSION

The results indicated that inuntreated,infected cells a small fraction of the input 3H-SFV RNA entersanMRCstructure. Most, if notall,ofthis

VOL.

173

oo.-n

on November 11, 2019 by guest

http://jvi.asm.org/

[image:5.492.42.238.161.386.2]

(6)

100

C-w 80

c 60

2 40

:D

-o-oQ'0O-°- °~oo_ oo- °'

_-0 10 20

[image:6.492.62.254.66.240.2]

FRACTION NUMBER

FIG. 7. Analysis ofRNA extracted from the

mito-chondrial fraction of 3H-SFV-infected cells which had

been treated with interferon. Cells were infected with

3H-SFV after 14 hr oftreatment with 1,000 units of interferon per ml. Mitochondrial fractions were

pre-pared, and RNA was extractedfrom them and

ana-lyzedasinFig. 4.

RNA is in the RI form (5), suggesting that the

structure is indeed associated with virus RNA synthesis.Later ininfection, functionalviralRNA

polymerase is present in a very similar structure

(15).

Results from cycloheximide-treated cells sug-gest that entry ofinputRNA into this structure

requires protein synthesis, since this viral func-tionisinhibited. Onthe otherhand, no such

in-hibition was seen in interferon-treated cells, but

the viral RNA present in the MRC structure

remained for themostpart inthe 42S form pres-entin the virion.

The resultswith interferon indicate that,

what-everthe site ofits actionis, this must liebeyond thestepinwhichtheinputRNAisintegrated into the MRC structure. Interferon treatment has no

effectonviraluncoating (2),exceptinthecaseof

vacciniavirusin which thisispartially dependent

on viral protein synthesis (11). Since treatment

with 1,000 units of interferonper ml appears to

block translation of input RNA in the

SFV-infected cell (4), the integration step probably

doesnotrequire theviral RNApolymerase. This conclusion was strengthened by findings with a

temperature-sensitive mutant of the closely

re-latedSindbisvirus, TS-6. The temperature-sensi-tive step ofTS-6 isrelatedto viralRNA

polym-eraseproduction, but TS-6 RNAhasbeen found toentertheRC-likestructure atpermissive(27 C)

andnonpermissive (39 C) temperatures. Atboth

temperatures, however, cycloheximide treatment

blocks this step (T. Sreevalsan, in preparation).

In all cases studied, therefore, the action of

cycloheximidewas to inhibitviral RNAentryin the MRCstructure; thenatureofthe protein

re-quired for this function is ofsome interest. The

currentconceptofinterferon action is that

trans-lation of viral, butnotcellular,messengerRNAis

blocked ininterferon-treatedcells (14). The pres-entresults, taken together,suggest thatitmaybe

thesynthesis ofacellprotein which is requiredfor

theintegration ofinput viral RNA into the

RC-like structure. If the latter hypothesis is correct, someofthe properties ofthis cell proteincan be

deduced. The protein would not be one

specifi-cally induced by virus infection, since the virus replicates quite well in actinomycin D-treated cells. Also the protein must have a fairly rapid

turnover rate,aconclusion suggestedby the result

obtained with cycloheximide. The existence ofa

host proteincontrolling virus replication in C-M cells persistently infected with mumps virus was

recently reported. This protein alsoseemstohave the properties noted above (13).

An alternative explanation for our

observa-tions would be that the integration step requires viral protein synthesis andthat cycloheximide isa more efficientinhibitor of virus protein synthesis

than isinterferon. In the lattercase,residual virus

proteinsynthesis ininterferon-treatedcellswould

be responsible forthe integrationstep. Interferon hasbeen showntoinhibitsome stepsinthe

syn-thesis of viral RNA more effectively than others (12).

LITERATURE CITED

1. Cheng,P. Y. 1961. Purification, size, and morphology ofa

mosquito-borne animal virus, Semliki Forest virus.Virology 14:124-131.

2. DeSomer, P., A. Prinzie,P.Denys, Jr.,and E.Schonne.1962. Mechanism of action ofinterferon. I. Relationship with viralribonucleic acid.Virology 16:63-70.

3. Fantes, K.H. 1965.Furtherpurificationof chickinterferon. Nature(London)207:1298.

4. Friedman,R. M. 1968. Inhibition ofarbovirusprotein syn-thesisbyinterferon. J. Virol.2:1081-1085.

5. Friedman,R.M. 1968.Replicative intermediate ofan arbo-virus. J. Virol. 2:547-552.

6. Friedman,R.M.,K. H.Fantes,H. B.Levy,andW.B.Carter.

1967. Interferon actionon parental Semliki Forest virus

ribonucleic acid.J. Virol.1:1168-1173.

7. Friedman, R. M., H.B.Levy,andW.B. Carter. 1966.

Replica-tion of Semliki Forest virus.Three forms of viral RNA pro-ducedduringinfection. Proc. Nat. Acad. Sci.U.S.A. 56: 440-446.

8. Friedman, R. M., and J. A.Sonnabend. 1965. Inhibitionof inteferon action by puromycin.J. Immunol. 95:696-703. 9. Girard, M.,D.Baltimore,and J.E.Darnell.1967. The

polio-virus replication complex: site forsynthesis ofpoliovirus

RNA. J. Mol. Biol. 24:59-74.

10. Lowry,0.H.,N.J.Rosebough,A.L.Farr,and R.J.Randall.

1951.Proteinmeasurementwith the Folinphenolreagent. J. Biol. Chem. 193:265-275.

11. Magee, W. E.,S. Levine,0.V.Miller,andR. D.Hamilton.

on November 11, 2019 by guest

http://jvi.asm.org/

(7)

1968.Inhibitionbyinterferonof theuncoatingof vaccinia virus.Virology 35:505-511.

12. Mecs, E., J.A.Sonnabend,E.M.Martin,and K. H.Fantes.

1967. The effect ofinterferononthesynthesis of RNA in chickcells infected with Semliki Forest virus. J. Gen. Virol. 1:25-40.

13. Northrop,R.L.1969.EffectofpuromycinandactinomycinD

onapersistentmumpsvirusinfection in vitro. J. Virol. 4: 133-140.

14. Sonnabend,J.A.,andR. M. Friedman. 1967.Mechanismsof interferonaction,p.202-231. In N. B. Finter (ed.), Inter-ferons. North-Holland Publishing Co., Amsterdam. 15. Sreevalsan, T., and F. H. Yin. 1969. Sindbis virus-induced viral

ribonucleic acid polymerase. J. Virol.3:599-604. 16.Tremblay,G.Y., M. J.Daniels, and M.Schaechter. 1969.

Iso-lation ofa cell membrane-DNA-nascent RNA complex

from bacteria.J. Mol. Biol.40:65-76.