Copyright ( 1971 AmericanSociety for Microbiology Printedin U.S.A.
Mechanism of Synthesis of Vaccinia
Virus
Double-Stranded Ribonucleic
Acid
In Vivo
and In
Vitro
CLARENCE COLBY,1 CHRISTINE JURALE, AND JOSEPH R. KATES
Department of Biology, University of California, San Diego, La Jolla, California 92037, andDepartment of Chemistry, University of Colorado, Boulder,Colorado80302
Received forpublication2September 1970
The synthesis of vaccinia virus double-stranded ribonucleic acid (RNA) in
infected HeLa cellswassensitiveto actinomycin D, suggestingthat a
deoxyribonu-cleic acid dependent reaction is involved. Some double-stranded RNA was made
inthepresenceofcytosinearabinoside in infected cells. Double-stranded and
com-plementary RNAweresynthesized in vitro byusing vacciniacores.Thesetwo
obser-vations indicate thatsomeofthe double-strandedRNAis readfrom"early"genes.
Thedouble-stranded RNA synthesized in vitro had thesamepropertiesasthatmade
invivo. Atleast 70% of the double-stranded RNA madein vivowasin
ribonuclease-resistant form prior to sodium dodecyl sulfate-phenol extraction. In addition,
there was a complementary RNA in infected cells which could be converted to
double-stranded RNA by annealing.
Vaccinia is a large deoxyribonucleic acid
(DNA)-containing virus which replicates in the
cytoplasm of infected cells (8) and which
stimu-lates thosecellstosynthesize the antiviralprotein,
interferon (6). Field et al. (5) postulated that a
DNAribonucleic acid (RNA) hybrid might be the
inducer of interferon in cells infected withaDNA
virus; however, DNA-RNA-like synthetic
poly-nucleotideswerefoundnot toinduceinterferon in
chick embryo fibroblasts (2).
Recently, Colby and Duesberg (3) reported finding ribonuclease-resistant RNA in vaccinia
virus-infected chick cells. The RNA was
charac-terizedasdouble-strandedRNAand itwasshown
to be virus specific; i.e., it was hybridized with
vaccinia virusDNA(3, 4). ThisRNAisanactive
inducerof interferon (3).
Thepresenceof double-stranded RNA in cells
infected with a DNA virus is not restricted to
eucaryotic cells. B0vre and Szybalski (1) found
complementary RNA synthesized from the b2
region of coliphagex,and Juraleetal. (9) found
virus-specific double-stranded RNA in
T4-in-fected E.coli.Thus, themechanism of synthesis of
the vaccinia virus double-stranded RNA may
reflectagenerally interesting biological
phenome-non.
MATERIALS AND METHODS
HeLa cells were grown in monolayers in 32-oz
(ca. 900ml) prescription bottlesin 40 ml of Eagle's
S
Senior Dernham Postdoctoral Fellow of theAmerican Cancer Society. Present address: Division of Biological Sciences,
Uni-versityofConnecticut, MicrobiologySection, Storrs, Conn. 06268.
71
medium containing 5% calf serum treated for 1 hr
at 56 C. Vaccinia virus (WR strain) was grown in HeLa cells and purified as described by Joklik (7). Confluent monolayers were infected with 100 virus particles per cell for 1 hr in 1.0 ml of medium.
Radioactive RNA was prepared bydecanting the
medium and incubating the monolayers for 15 min with lOO,Ci of 3H-uridine (New England Nuclear
Corp.) in 5 mlofmedium. Theradioactive medium
was removed, and thecellswere washedwith 25 ml of ice-cold 0.15 M NaCI solution. The monolayers werethen treatedwith1.5mlofice-cold0.01 MNaCl, 0.01 M tris(hydroxymethyl)aminomethane
(Tris)-hy-drochloride(pH7.4), 0.001 MMgCl2andallowed to
stand onicefor 3 min. The cells were removed from
thesurface of theglassbyscraping and disruptedby
treatment with arotatingDounce homogenizer. The nuclei were removed by centrifugation, and sodium
dodecyl sulfate(SDS) and Pronasewereadded to the
cytoplasmic extract to final concentrations of 0.5%
(w/v) and 0.5 mg/ml, respectively. This extract was
incubatedfor 30 min at 37 C. The cytoplasmic nucleic
acids were purified by three extractions with phenol
followed by ethanol precipitation.
Double-stranded RNA was prepared exactly as
described previously (3). DNA and single-stranded
RNA weredigestedwith deoxyribonuclease and ribo-nuclease,repectively, and, after removing the nucleases
by three phenol extractions, the remaining nucleic acids and oligonucleotides were applied to a 6% agarosecolumn (3).
TheRNAwascharacterizedas anRNAduplex by
the following criteria: (i) nuclease sensitivity, (ii)
thermaldenaturation, (iii)Cs2SO4equilibrium density gradient centrifugation, (iv) induction of interferon in chick embryo cells, (v) self-annealing after removal
ofpossible contaminating DNA strands (3, 4). It is
important to note that only the material which is
on November 11, 2019 by guest
http://jvi.asm.org/
phenol-SDS-purified, resistant to digestion with deoxyribonuclease and ribonuclease, and excluded froma6%agarosecolumn will bereferredtobelowas vaccinia virus double-stranded RNA.
RNAwas synthesized in vitro from vaccinia virus
cores asdescribed previously (10, 11). The RNAwas synthesized in reaction mixtures which routinely contained the followingcomponents, in0.4 ml: 2 X 101" cores,0.01M2-mercaptoethanol, 0.005M
MgC92,
0.05 M Tris-hydrochloride (pH 8.4), 1 ,umole ofadenosine triphosphate, 0.5 ,umole each of cytosine triphosphate (CTP) and guanosine triphosphate (GTP), 0.02 ,umole of uridine triphoshate (UTP),
and2uCi of3H-UTP. When 3H-CTPor3H-GTP was used asthelabeled substrate, UTPwas at0.5 jAmole per 0.4 ml and the labeled nucleotide was at 0.02 ,umoleper0.4 ml and 21ACiper0.4ml.Incubationwas for 25 min at 37 C. Purification of the RNA after synthesis was by phenol extraction, and ethanol
precipitationwas aspreviously described (2).
RESULTS
Effect of inhibition of RNA synthesisonthe
syn-thesis of double-stranded RNA. Since vaccinia
virusreplicatesin thecytoplasm of infected cells,
onecanfollowviral RNAsynthesis with 15-min
pulses of 3H-uridine. Vaccinia messenger RNA
(mRNA) and a small amount of transfer RNA
arethe only RNA species in thecytoplasm that
arelabeled undertheseconditions (8).
Thedouble-stranded vaccinia virus RNA could
arise by two different mechanisms. One strand
could be transcribed from vacciniaDNA, andan
RNA-dependent RNA polymerase could catalyze
the synthesis of the complementary strand of
RNA. Actinomycin D wouldnot beexpected to
inhibitthesynthesis ofthedouble-strandedRNA
for this mechanism. Alternatively, both strands
might be copied from complementary regions of
thevaccinia DNA, in which caseactinomycin D
should inhibit thesynthesis of the double-stranded
RNA.
Actinomycin D (10 ,g/ml) was added to the
culture medium ofvaccinia virus-infected HeLa
cells 4 hr after infection. At 5 hrafterinfection,
3H-uridine-labeled cytoplasmic RNA (15-min
pulse)waspreparedasdescribed above. Controls
included 3H-RNA prepared from infected and
uninfected HeLa cells not treated with
actino-mycin D.
The results presented in Table 1 indicate that
therewas avery smallamountof 3H-ribonuclease
resistantRNAin60-minpulse-labeled uninfected
HeLa cells, confirming the results previously
re-ported by Montagnier (12). Infection with
vac-cinia causeda35-fold increasein
3H-ribonuclease-resistant RNA. Treatment of vaccinia
virus-infectedHeLa cells withactinomycin D resulted
in a 95% inhibition of RNA synthesis as
meas-ured by the incorporation of 3H-uridine. The
synthesis of vaccinia virus-directed
double-strandedRNAwas also inhibited more than 90%.
Inother experiments when vaccinia mRNA
syn-thesis was inhibited 98% by actinomycin D, the
synthesis ofdouble-stranded RNA was inhibited
95 %. These results aresimilar to those previously
obtained with vaccinia-infected chick embryo
cells (4) and suggest that the DNA-dependent
mechanism is the one of choice.
Effect ofinhibitionof DNAsynthesis on the
syn-thesis ofdouble-stranded RNA. At aconcentration
of 10 ,ug/ml, cytosine arabinoside inhibits host
cell and vaccinia virus DNA synthesis by 99%
(13). Only early vaccinia virus mRNA is
tran-scribed in the presence of cytosine arabinoside
(13). It was previously suggested (4) that the
double-strandedRNAis a"late" productofviral intracellular biosynthesis. Ifthis is true, onewould
expectcytosinearabinosidetoabolishthe
synthe-sis of vacciniadouble-strandedRNA.
Cytosine arabinoside (10,ug/ml) wasaddedto
the culture medium of HeLa cells immediately
after infection with vaccinia virus. At 5 hrafter
infection, the cultures were pulse-labeled with
3H-uridine, andthe total cytoplasmic RNA and
double-stranded RNA were prepared. The
amount of 3H-labeled vaccinia mRNA in the
cytosine arabinoside-treatedcells was reduced to
one-third of that in the untreated infected cells
(Table 1). However, of the RNA which was
labeled, a significant proportion was
double-stranded. Thus, it appears that some of the
TABLE 1. Effects ofinhibitors of niucleic acid
synz-thesis on double-stranded RNA
3H counts/min3H counts/min Treatmenta inphenol after nucleases extract andagarose
HeLa, 60min, labeling 4.6 X 105 195 total nucleic acids
HeLa, 15 min, labeling 2.8 X 104 35
cytoplasmic extract
HeLa + vaccinia 4.5 X 105 5,300
HeLa+ vaccinia + acti- 2 X 104 520 nomycin
(10,ug/ml),
4to 5 hrpostinfection
HeLa+vaccinia + cyto- 1.5 X 105 820 sine arabinoside(lO,:g/
ml),0 to5postinfection
a Forthefirst treatment, uninfected HeLacells
werelabeled for 1 hrwith25
tsCi
of3H-uridine.Forthe remaining treatments, cultures were labeled
for 15 min with
lOO1
uCi of 3H-uridine and cyto-plasmic RNA was prepared. All RNA prepara-tions were purified, subjected to nucleasediges-tion,andeluted froma6%agarosecolumnas
pre-viouslv described (3).
on November 11, 2019 by guest
http://jvi.asm.org/
[image:2.491.261.455.427.586.2]vaccinia double-stranded RNA is made from early genes.
Synthesis of double-stranded and complementary
RNA in vitro. Vaccinia virions contain an RNA
polymerase activity which directs the synthesisof
early viralmRNA (10,11).Table2showsthatthe
RNA made in vitro containsapproximatelyequal
amounts of the three nucleotides uridine
mono-phosphate, cytidine monophosphate, and
guano-sine monophosphate. 3H-ATP was not used
be-causeoftheformation ofpolyA(10).Someofthe
RNA made in vitro was resistant to a mixtureof
pancreatic and T1 ribonuclease at a high salt
concentration (Table 2). However, when we
sub-jected the RNA to annealing conditions, there
was asixfold increase in the level of
ribonuclease-resistant RNA. Finally, thermal denaturation
rendered both annealed and nonannealed RNA
susceptible toribonuclease degradation.
The level of ribonuclease-resistant 3H-RNA
afterannealing varied from3to 10%of thetotal synthesis in different experiments. The annealing
reaction isdependenton theconcentrationofthe
input RNA. Figure 1A shows that the level of
annealed ribonuclease-resistant 3H-RNA
de-creased with second-order kinetics as the
3H-RNAwas diluted. Figure 1B shows that the
an-nealingreaction was alsotime-dependent.
We carried out agarose chromatography of
annealed and nonannealed ribonuclease-treated
TABLE 2. Effect of annealinig on the amount of
ribonuclease-resistant RNAG
3H-UTPb 3H-CTP 3H-GTP Treatment (counts/ (counts/ (counts/
min) min) min)
A. Trichloroacetic acid 151,589 113 ,984 109,644
ppt
B. Ribonuclease 2,432 1,931 1,539
C. Anneal ribonuclease 14,271 11,825 8,674
D. Melt ribonuclease 206 217 208
aPurification of theRNAaftersynthesiswasby
phenol extraction and ethanol precipitation, the
RNA wasresuspended in2X SSC (SSC = 0.15 M
NaClplus0.15 Msodiumcitrate), and thesamples
were treated as follows. (A) The 5% trichloro-acetic acid-insoluble 3H-RNA was determined.
(B) Ribonuclease was treated with 10mg of
pan-creatic ribonuclease plus 3pg of T1 ribonuclease
per mlat 37 Cfor 15 min in2X SSC, and the 5% trichloroacetic precipitable counts per minute
weredetermined. (C) The RNA wasannealed for
6 hr at 65 Cin2X SSC andtreated as in B. (D)
Samplesdiluted to O.1X SSC beforeor after
an-nealing (shown here for after annealing) were
heatedto 100C, cooled, andtreatedasinB.
bAbbreviations: UTP, uridine triphosphate;
CTP, cytidine triphosphate; GTP, guanosine
triphosphate.
z
-._
0
cs
.U)
40
cc I 0 z
cr
E 0. 0 t)
z
C
l
I
0 0
0
0
800
6001
4001
800
600
400
200
2.5 5.0 7.5 10 15 25
Dilution
B
IS
1 2 3 4 5 6 7 8 9
Time of Annealing (Hr.) FIG. 1. Concentration and time dependentce of the annealingreaction. (A)RNA waspreparedanidpurified
in the standardmannerby using vaccinia virus cores (10). The RNA was diluted as indicated and
an-nealedfor 6 hrat 65 Cin 2X SSC (0.3 M NaCI,
0.03M trisodium, citrate). After annealing, the RNA
was treatedwith ribonucleaseasdescribed in Table 2, and the 5% trichloroacelic acid insoluble 3Hcounts perminuteweredeterminied. (B) Samples of RNAwere
annealedas in (A) for different periods oftime, and treated with ribonuclease; the 5% trichloroacetic
acid-precipitable 3H counts per minuite were determinied
3H-RNAaspreviouslydescribed(3).Aportion of
the 3H-RNA was excluded from the agarose
column in both cases, indicating the presence
of high-molecular-weight ribonuclease-resistant
RNA with and without annealing. Only 0.074%
of the3H-RNA which hadnoprior annealingwas
excluded from the column, whereas the amount
excluded increased to 0.34% for 3H-RNA which
wasannealedbeforeribonucleasetreatment.
Table 3 indicatesthat the excluded RNA was
resistant to both ribonuclease and ribonuclease
plusdeoxyribonuclease, butitbecamecompletely
sensitive to ribonuclease afterthermal
denatura-tion.Themelting temperatureofthe
ribonuclease-resistant RNA excluded by the agarose column
A
0
.
Is
-1
on November 11, 2019 by guest
http://jvi.asm.org/
[image:3.491.251.444.80.384.2] [image:3.491.49.239.393.503.2]TABLE 3. Nuclease sensitivity ofpurified ribonuclease-resistant RNAa
Treatment 3H-counts/min Per centoftotal
A. Trichloroacetic acidppt.. 1,604
B. Ribonuclease... 1,542 93.7
C. Ribonuclease +
deoxy-ribonuclease... 1,437 82.5
D. Melt and ribonuclease... 74 4.6
aAnnealedRNA waspreparedandsubjectedto
ribonuclease digestion and agarose
chromatog-raphy. The excluded RNA was dialyzed versus 0.01 MTris (pH 7.4),MNaCI.Samplesweretreated
asfollows. (A) Five percent trichloroacetic
acid-precipitablecounts per minute were determined.
(B) Ribonucleasewas treated with 20,gof
pan-creatic ribonucleaseplus2pgofT,RNase per ml
in 0.1 M Tris (pH7.4), 0.2 M NaCl for 20 minat37 C. Then treatmentproceededasin A. (C) AsinB,
but the solutionwas made 0.004 MMgCl2and 50
,pg
of DNase per ml(ribonuclease-free,Worthing-ton) wasadded for 20 minat37C. (D)Samplewas
heatedto100 C for 15 min,cooled, and treatedas
in B.
was determined as previously described (3). A
sharptransition withaTm of 76 Cwas observed
(Fig. 2).
Sucrose gradient sedimentation in 15 to 30%
sucrose was carried out on the
ribonuclease-resistantRNA excluded byagarose, indicating a
rangefrom7.5 to 10.5Sforcoredouble-stranded
RNA (Fig. 3). This value is in good agreement
with the sedimentationcoefficient ofthevaccinia
double-strandedRNAsynthesizedin vivo
(3).
Effect ofannealing on the amount ofvaccinia
virus 3H-double-stranded RNA made in vivo. Our
results with thevaccinia RNA
synthesized
invitroprompted us to reinvestigate the kinetics of
appearanceof thevaccinia virus double-stranded
RNA intheinfected cell. Confluent
monolayers
ofHeLacells were infected with vaccinia virus. At
1,3, and5hrafter
infection,
cultureswerepulse-labeled with100
,uCi
of3H-uridine,
and thetotalcytoplasmic nucleic acids were
prepared.
Afterremoving the
contaminating
DNAby digestion
with
deoxyribonuclease (100
,ug/ml,
60 min,37C),each3H-cytoplasmicRNA
preparation
waspurified by three phenol extractions and three
ethanol precipitations. The
purified
RNA wasdivided into two
equal
samples.
From one,double-stranded RNA was
prepared
as above.The other wasallowedtoself-annealfor 6 hr and
then
double-stranded
RNA wasprepared.
Theresults are
presented
in Table 4. It is clear thatcomplementary RNA is made
throughout
theinfection
cycle.
Is there double-stranded RNA inside the cell?
It maybeargued that only complementaryRNA
exists inside the cells and that the SDS-phenol
extraction catalyzes theannealingof theRNA to
give helical structures. This argument was tested
by the following experiment. 3H-cytoplasmic
RNA was prepared as above. Before extraction
with SDS and phenol, NaCl and ribonuclease
wereadded to final concentrationsof 0.25 M and
100
,g/ml,
respectively. After incubating themixture at 37 C for 60
min,
double-strandedRNA wasprepared by the standard SDS-phenol,
nucleasedigestion, agarosechromatography tech-nique. As a control, duplicate cultures were
infected at the same time and double-stranded
RNAwasprepared as usual. At least 70% of the
double-stranded RNA purified by this technique
was in a form that is resistant to ribonuclease
100
.
z
co75
w
50
c)
z
-
40
60
80
100
TEMPERATURE
OC
FIG. 2. Effect of temperature on
ribonuclease-re-resistantRNA. RNA excludedby agaroseas inTable2
was dialyzed versus 0.02 M Tris (pH 7.4), 0.01 M
NaCl. Samples were made 0.001 M
ethylenediamine-tetraaceticacid and heatedtotheindicatedtemperature
for10min,andrapidlycooledonice.Eachsamplewas
made 0.2M NaCI and treated with ribonucleaseasin
Table 2. The trichloroacetic acid-insoluble RNA was determined.
CVi
on November 11, 2019 by guest
http://jvi.asm.org/
[image:4.491.264.457.276.573.2]16
0
12
k.)
4
23S 16S 4S
4 8 12 16 20
FRACTION NO.
FIG. 3. Sucrose gradient analvsis of ribonuclease-resistant RNA excludedfrom 6% agarose. The RNA was run ona 15 to30% (w/v) sucrose gradient
con-taining
0.1%70
sodiumdodecylsulfate, 0.1 MNaCI, and 0.01 M ethylenediaminetetraacetic acidfor 16 hr at27,000 rev/min at 25 C in a Spinco SW27 rotor. Bacillus subtilis RNA wasusedas a marker.
TABLE4. Effect ofanniealing on the amount of
WH-double-stranded RNA made in vivo
3Hcounts/min Hcounts/min Hcounts/min Timeafter afterphenoland afterribonu- after annealing,
infection(hr) deoxyribo- cleaseand ribonuclease,
nuclease agarose and agarose
1 1.0 X 105 560 800
3 2.0 X 105 2,400 3,600
5 1.8 X 105 2,440 4,600
before extraction with
phenol (Table 5).
Con-sideringtheseverityoftheribonucleasedigestion, it is
likely
that most, ifnotall,
of thedouble-stranded RNA isolated by this technique is
helical RNA inside the cell.
DISCUSSION
We find thatthe synthesis ofvaccinia
double-stranded RNA is inhibited
by
actinomycin D.Therefore, we conclude thatthis RNA arises via
a DNA-dependent reactionmechanism.
Ribonuclease resistant RNA synthesized in
vitro by vaccinia cores possesses properties
similar to those of ribonuclease-resistant RNA
isolated from vaccinia virus-infected cells and
T4phage-infectedbacteria (3, 4, 9). It is
interest-ing to note that the increase in
ribonuclease-resistant RNA after annealing corresponds to a
similar phenomenon observed for the in vivo
vaccinia RNA isolated throughout infection.
Since cores synthesize only "early" RNA (11), we may regard these results as proof that vaccinia virus does produce complementary RNA early
ininfection. Ourresults with cytosine
arabinoside-treated cells offer further confirmation of this
idea.
Itisimportant to note that the annealing of the
ribonuclease-resistant RNA made in vitro is both concentration- and time-dependent. These results
support the modelinvolving the transcription of
complementary regions of vaccinia DNA and rule
out thepossibilitythat theribonucleaseresistance
could be theresultof a folding back of a
single-stranded molecule to give a lengthy hairpin
structure.
TABLE5. Effect ofpretreatment with ribonuclease before phenol extraction ont double-stranded
RNA
'Hcounts/min 3H counts/min Per-Treatment inphenol after nucleases cent-extract andagarose aeo
Standard
proce-dure... 3.2 X 1063.7 X 103 1.16 Pretreatment with
ribonuclease 2 X 104 2.7 X
103
0.84CONVERGENT TRANSCRIPTION
- ~ ~ ~~~~~~~VACCINIA
C
COMPLEMENTARY
1 I RNA
1 INTRACELLULAR
111~ ds-RNA
1RNase
PURIFIED ds-RNA
DIVERGENT TRANSCRIPTION
5'1 ,^5'
3 < - L --~~~~~~ 31 1
3R
RNase
FIG. 4. Models ofconvergent and divergenit trani-scription. RNA is transcribedfromboth strandsof the vacciniaDNAsuch that molecules with complementary regionsaresynthesized. Someofthese molecules form the appropriate base pairs andare converted to intra-cellular double-stranded RNA with single-stranded
regions (Sw20 = 9 to22S). Duringpurification of the
double-stranded RNA, thesingle-stranded portions are removedby ribonuclease digestion. Inthecase of con-vergent transcription, the terminal sequences of the complementary RNA are found in double-stranded RNA, whereas the initial sequences ofcomplementary
RNA become double-stranded RNA in the case of divergent transcription. ds-RNA, double-stranded
RNA. IC',
on November 11, 2019 by guest
http://jvi.asm.org/
[image:5.491.42.237.67.280.2] [image:5.491.248.443.267.519.2] [image:5.491.45.240.362.462.2]Duesberg and Colby
(4) reported
that theintracellular
form of the double-stranded RNA isa heterogeneous
population
of molecules withsedimentation coefficients
ranging
from9to22S.Treatment with ribonuclease converts these
molecules into a homogeneous population of
double-stranded
molecules of 9 to lOS.Thus,
the
intracellular
formappearstobeapopulation
of
molecules
sharing
common sizedouble-stranded cores and
having single-stranded
por-tionsof various lengths.Thesedata
coupled
withthe above arguments that the double-stranded
RNA is
synthesized
via aDNA-dependent
reac-tionmechanismareconsistent withthepattern of
convergent transcription
recently suggested by
B0vre and
Szybalski (1)
for the b2region
ofcoliphage x.
Alternatively,
the vacciniadouble-stranded RNA could arise from a pattern of
divergent
transcription (Fig.
4).
Both modelsinvolve the
transcription
ofcomplementary
regions of theDNA;
they
differfrom each otherwith respect to the promoter sites or
regions
ofinitiation of
transcription.
It should beem-phasized that neithermodel demands the
simul-taneous transcription of the
complementary
sequences. Indeed, the
finding
ofcomplementary
RNA in the
cytoplasm
argues againstsimul-taneoustranscription.
It was
previously
shownthat thepurified
vacciniadouble-strandedRNAisavery potentinducerof
interferon in chick cells
(3).
The demonstrationthatatleast 70%ofthepurified double-stranded
RNA is in a ribonuclease-resistant form in the
cytoplasm
of infected cellssuggestsmorestrongly
that this
population
of molecules isresponsible
for the induction of interferon in cells infected with
vaccinia virus.
ACKNOWLEDGMENTS
This investigation was supported by Public Health Service grants CA-10802 from the National Cancer Institute and IROlA-10841-02 from the National Institute for Allergy and InfectiousDiseases.
LITERATURECITED
1. Bovre,K. E., andW. Szybalski. 1969. Patterns of convergent andoverlappingtranscription within theb2 region of coli-phageX.Virology 38:614-626.
2. Colby, C., andM.J.Chamberlin. 1969. Thespecificity of inter-feroninduction in chickembryo cells by helical RNA. Proc. Nat. Acad.Sci.U.S.A.63:160-167.
3. Colby, C., andP. H.Duesberg. 1969.Double-stranded RNA invaccinia virus infectedcells. Nature (London) 222:940-944.
4. Duesberg, P. H., and C.Colby. 1969. Onthe biosynthesis and structure of double-stranded RNA in vaccinia virus-in-fected cells.Proc.Nat. Acad. Sci. U.S.A.64:396-403. 5. Field, A. K., G. P. Lampson, A. A., Tytell, M. M. Nemes, and
M.R.Hilleman. 1967. Inducers of interferon and host re-sistance.IV.Double-stranded replicative formRNA
(MS2-RF-RNA) fromE.coliinfectedwith MS2-coliphage. Proc. Nat.Acad. Sci. U.S.A. 58:2102-2108.
6. Glasgow, L.A., andK.Habel.1962. Theroleof interferonin vaccinia virus infection ofmouse embryo tissue culture. J. Expt.Med. 115:503-512.
7.Joklik,W.K. 1962. Thepreparation and characterizationof highly purified radioactively labelled poxvirus. Biochem. Biophys. Acta 61:290-301.
8.Joklik,W. K. 1968.In H.Frankel-Conrat (ed.), Molecular basis of virology, p. 576. Reinhold PublishingCo., New York.
9.Jurale, C., J. R. Kates, and C. Colby. 1970. Isolation of double-strandedRNAfromT4phage infected cells. Na-ture (London) 226:1027-1;29.
10.Kates, J. R., and J.Beeson. 1970. Ribonucleic acid synthesis in vaccinia virus. I.Themechanism of synthesis and re-leaseofRNAinvacciniacores.J. Mol. Biol. 50:1-18. 11.Kates, J. R.,andB.McAuslan. 1967.Messenger RNA
syn-thesis by a"coated" viralgenome. Proc. Nat.Acad. Sci. U.S.A. 57:314-320.
12. Montagnier, L. 1968.Presenced'un acide ribonucleique en doublechaine dans des cellulesanimales. C.R. H.Acad. Sci. Ser.D267:1417-1420.
13. Oda,K.S., and W. K.Joklik. 1967. Hybridizationand sedi-mentation studieson"early" and "late" vaccinia messen-gerRNA.J. Mol.Biol.27:395-419.