JOURNAL OF VIROLOGY, Dec. 1967, p. 1217-1223Copyright @ 1967 American Society for Microbiology

Vol. 1, No. 6Printed in U.S.A.

Isolation and Partial Characterization of Nucleic Acidof Influenza Virus

D. P. NAYAK AND M. A. BALUDA'Department of Medical Microbiology and Immunology, School ofMedicine, University of California,

Los Angeles, California 90024

Received for publication 5 September 1967

The principal ribonucleic acid (RNA) component isolated from purified equineinfluenza virus has an approximate sedimentation coefficient (S20ow) of 21S insucrose gradient containing 0.1 M NaCl. Three other components of 18S, 14S, and8S were also detected. All the RNA components have characteristics of single-stranded RNA. The average base composition of the principal RNA components iscytosine, 22.2; adenine, 22.9; guanine, 22.3; and uridine, 32.6. There was no qualitativedifference in the RNA isolated from noninfectious virus particles compared to thatfrom infectious virions.

Influenza virus, a myxovirus, exhibits certainunique characteristics not shared by other virusesof the group, e.g., Newcastle disease virus andmumps. These are pleomorphism of virions whichappear as spheres or filaments (5, 8), geneticrecombination (14, 23), multiplicity reactivation(13), incomplete virus production (26), andsensitivity of the early events of replication toultraviolet light, actinomycin D, and mitomycinC (6, 17). There is also a difference betweeninfluenza and parainfluenza viruses in the size ofthe viral genome. A large ribonucleic acid (RNA)molecule (495 or 57S) has been isolated fromNewcastle disease virus (11, 15), but there hasbeen disagreement about the size of the intactRNA genome isolated from influenza viruses;RNA of varying sizes ranging from 38S to 7Shave been reported (2, 10, 18, 22, 24). This studywas undertaken to characterize the RNA molecu-lar species isolated from purified influenza virions.The present report confirms the observations ofDuesberg and Robinson (12), who have recentlyshown that the RNA of influenza virus containsseveral RNA components of sizes ranging from 9Sto 18S.

MATERIALS AND METHODS

Viruses. The American (Miami) strain of equineinfluenza virus was used. To minimize the presenceof noninfectious virus, the virus stock was preparedin the following manner: three successive egg passageswere initiated intrachorioallantoically with 1 egginfectious unit (iu) per 10-day-old embryonated

chicken egg, and 100 iu from the third passage wasinoculated into each egg to make the final virus stock.Chorioallantoic fluid was harvested after 20 hr ofincubation at 37 C and contained 109 iu and 640hemagglutinin units/ml. Kimber farm strain crossK-137 chick embryonated eggs were used throughoutthis study.The BAI strain A of avian myeloblastosis virus

(AMV) was used in some experiments for comparisonof viral RNA's (19).

Hemagglutination and infectivity assays. The hemag-glutinin titers of influenza virus suspensions weredetermined with human type 0 erythrocytes accord-ing to the technique previously described (17).

Infectivity was assayed by making serial, 10-folddilution steps of a virus suspension and inoculating0.1 ml of each dilution step into each of four 10-day-old chick embryonated eggs. After 48 hr of incubation,each egg was tested for the presence of hemagglutininin the chorioallantoic fluid. The infectivity titer wascalculated from the highest dilution factor causinghemagglutinin formation and the fraction of eggswhich developed hemagglutinin at that dilution.

Preparation of radioactive viruses. To prepareradioactive influenza virions, 0.1 ml of virus suspen-sion containing 100 iu, and 1 mc Of 32PO4, or 200 ,ucof 3H-uridine ('H-UR; Schwarz Bio Research Inc.),were injected simultaneously into each embryonatedegg. After incubation at 38 C for 30 or 36 hr, theinfected eggs were chilled for 8 hr at 4 C; then chorio-allantoic fluid was harvested and used immediatelyfor isolation of virus.To prepare 3H-UR-labeled, incomplete (low in-

fectivity) influenza virions, monolayers of chickembryo fibroblasts in 120-mm plastic culture disheswere infected at an input multiplicity of 50 iu/cell(21) and incubated in 10 ml of modified Eagle'smedium containing 5% dialyzed calf serum plus100 uc of 3H-UR. After 16 hr of incubation in the

1217

IRecipient of American Cancer Society Award~PRA 34) for faecilty position support.

NAYAK AND BALUDA

presence of 3H-UR, the supernatant fraction washarvested and immediately used for virus purifica-tion. Each supernatant fraction contained 320 hemag-glutinin units and 105 iu in 10 ml.

Tritium-labeled AMV was prepared from super-natant fluids of leukemic chicken myeloblast culturesas described by Robinson and Baluda (19).

Virus purification. Influenza virus was purified by aprocedure similar to that used for AMV (4). Potas-sium citrate was added dropwise to fresh chorioal-lantoic fluid to a final concentration of 0.15 M. Thevirus was concentrated and partially purified by twosuccessive sedimentations to an interface between a65% sucrose-D20 cushion and a 15% sucrosesolution containing K-citrate (0.15 M). Finally, thevirus preparation was layered over a linear gradientof sucrose (15 to 70%) containing 0.1 M NaCl, 0.01 Mtris-(hydroxymethyl)aminomethane (Tris)-HCI (pH7.4), 0.001 M ethylenediaminetetraacetate (EDTA)(NTE) and centrifuged at 50,000 rev/min for 2 hrin a Spinco SW 50 rotor. Fractions were collectedfrom the bottom of the tube and analyzed for radio-activity, hemagglutinin, and infectivity. The fractionscontaining the virus were pooled and used for ex-traction of nucleic acid. The entire procedure wascarried out at 2 to 4 C.

3H-labeled influenza virus was also purified byhemadsorption and elution (10). Packed human redcells (type 0) were added to chorioallantoic fluidto a final concentration of 2% and kept at 4 C withoccasional shaking for 1 hr in the presence of colduridine (100 ,ug/ml). Red cells were sedimented at2,000 rev/min for 10 min and washed three times withice-cold phosphate-buffered saline containing colduridine. The virus was eluted with receptor-destroyingenzyme in phosphate-buffered saline (1:10) at 37 Cfor 2 hr and concentrated over a 65% sucrose-D20cushion. Viral nucleic acid was then isolated.

Extraction of viral RNA. Viral RNA was extractedby a procedure previously described (11). The frac-tions containing virus were pooled and diluted threetimes with NTE buffer (pH 8.5). Mercaptoethanolto 1.0%; yeast RNA, 200 pg; and sodium dodecylsulfate to 1%, were added successively to the viruspreparation. The mixture was kept at room tem-perature for 5 min; it was then extracted three times(6 min each) by shaking with cold phenol equilibratedwith NTE buffer (pH 8.5). The nucleic acid was pre-cipitated from the aqueous phase by adding sodiumacetate (pH 5.5) to 1.6% and 2 volumes of absoluteethyl alcohol. It was stored at -20 C.

Velocity sedimentation ofRNA in sucrose gradients.After two successive precipitations with ethyl alcohol,the viral RNA was dissolved in 0.2 ml of NTEbuffer (pH 7.4) and layered over a sucrose gradient(5 to 20%) containing NTE buffer, pH 7.4 (high-saltgradient). For running in low-salt gradients, the RNAwas dissolved in distilled water and the sucrosegradient contained only 0.0005 M Tris-HCI (pH 7.4)and 0.0005 M EDTA. After centrifugation at 49,000rev/min for 2 hr at 4 C in the SW 50 rotor, fractionswere collected from the bottom of each gradient byusing a Buchner piercing unit.

RESULTS

Properties of purified virus. Figure 1 shows thedistribution of hemagglutinin, infectivity, andradioactivity after the final step of purification of32P-labeled influenza virus by density equilibriumcentrifugation in a 15 to 70% linear sucrosegradient. The peaks of hemagglutinin, infectivity,and radioactivity coincided. A visible band alsocoincided with these peaks. The sharpness of theband indicates homogeneity of the virus prepara-tion. Infectivity per hemagglutinin at the peakwas approximately 106. The density of the sucrosegradient at the peak was 1.21 g/ml. Hemag-glutinin at the top of the gradient represents freehemagglutinin or hemagglutinin activity associ-ated with pieces of cell membrane or with dis-rupted virions.

Properties of viral nucleic acid. The fractionscontaining hemagglutinin and 32p radioactivityin the sucrose density gradient of Fig. 1 werepooled and nucleic acid was isolated as describedin Materials and Methods. The isolated 32p_labeled influenza viral RNA was analyzed by

3 5 7 9 13 15 17FRACTION NUMBER

- 2,000

- 1,500 3C)

r,

Z- 1,000 I

3

500

100

FIG. 1. Density equilibrium ultracentrifugation of22P4abeled influenza virus. After two sedimentationsto an interface between a 65% sucrose-D20 cushionand a 20% sucrose solution, the virus preparation wasdiluted with NTE buffer (pH 7.4) and I ml was layeredover 4 ml ofa preformed sucrose gradient (15 to 70%)containing NTE buffer (pH 7.4). After centrifugationin the Spinco SW 50 rotor at 50,000 rev/min for 2hr at 2 C, fractions were collected from bottom oftube. A sample ofeach fraction (25 i.diters) was dilutedin 10 ml ofBray's scintillationfluid (9) and assayedfor82p radioactivity (0). Another sample (10 gliters) wasused for assay of hemagglutinin (X), and a third sam-ple (10 pliters) was assayed for infectivity (0).

1218 J. VIROL.

NUCLEIC ACID OF INFLUENZA VIRUS

velocity sedimentation centrifugation (Fig. 2).Ribosomal RNA (28S) added as a marker wasdetected by absorbancy at 260 m,u. The major32P-labeled RNA component contained 95% ofthe total radioactivity and had a sedimentationcoefficient of approximately 18S by comparisonof positions with 28S ribosomal RNA. This waschecked further by using 18S ribosomal RNA asa marker. The peaks of the radioactivity of 32pviral RNA and of the optical density of 18S cellRNA were very close and often coincided. Theradioactivity profile of influenza viral RNA wasbroader than the optical density of 18S ribosomalRNA, and there was an occasional shoulder inthe 14S region. About 5 to 8% of the radioac-tivity was found sedimenting around 85 (Fig. 2).

In another experiment, influenza virions werepurified as in Fig. 1. Nucleic acid was isolatedfrom virus present in the three fractions that pre-cede the peak of radioactivity and compared withthe nucleic acid isolated from virus present in thepeak and the next two fractions. Both viral RNApreparations were analyzed by velocity sedimenta-

z

(n

z

0

<

600 -

300

LOWSALT

I,I §IlIl

'bR P

'd

0.0

--4

N

0)a

3-005 l1

5 9 13 17 21 25

FRACTION NUMBER

FIG. 2. Velocity sedimentation centtrifugationl of32P-labeled RNA isolated from purified influenza virus.32P-labeled influeniza viral RNA and unlabeled 285ribosomal RNA, used as sedimenttation marker, in0.2 ml ofNTE buffer (pH 7.4) were layered on top ofa5.0-ml linear sucrose gradienit (5 to 20%,7O) conttainingNTE buffer (pH 7.4) and centrifugedfor 2 hr at 49,000rev/min at 4 C in a Spinco SW 50 rotor. Fractions were

collected and samples of each were analyzed for ab-sorbancy at 260 mA and trichloroacetic acid precipita-ble radioactivity in presence of carrier yeast RNA(100 Mg). The arrow indicates the fractioni reached byinflueniza viral RNA when centrifuged in a sucrose

gradient (5 to 20c%s) containing only 0.0005 M Tris-HCI (pH 7.4) and 0.0005 M ethylenediaminetetracetate,and anialyzed in the same way as above. The solid linerepresents the absorbancy at 260 my, and the brokenline, 32p radioactivity.

tion in sucrose gradients and were found to beidentical. They were quantitatively and qualita-tively similar to the viral RNA shown in Fig. 2.To examine further the effect of the method of

purification of virus and of the extraction pro-cedure on the size of the isolated viral RNA, thevirus was purified by hemadsorption and elutionand RNA was extracted in presence of unlabeledchick embryo fibroblast cells. Again, the peak ofviral RNA coincided with the peak of 18S ribo-somal RNA. The optical density profile of ribo-somal RNA also indicated that the RNA extrac-tion procedure did not degrade the ribosomalRNA.

Finally, to eliminate any possible effect of puri-fication or of extraction procedure on the degra-dation of viral RNA, 32P-labeled equine influenzavirus and 3H-labeled AMV were purified simul-taneously. Influenza virus and AMV banded atdensities of 1.21 and 1.17 g/ml, respectively, inthe sucrose gradient as determined by infectivityand radioactivity. RNA was extracted, after mix-ing the fractions containing AMV or influenzavirus, and analyzed by sucrose velocity centrifuga-tion (Fig. 3). 3H-labeled AMV RNA, which hasa S20,W of 715 (19), has a sharp peak around frac-tion 8. The 32P-labeled influenza viral RNA has asharp peak at fraction 20, corrsponding to asedimentation coefficient of 21S by comparisonof its relative position to that of 71S AMV RNA

1000 -

750 -

I

z

500

F-

0

250 -

0 -

200

150

In0

-100

z

- 50

I ,-T5 9 13 17 21 25FRACTION NUMBER

FIG. 3. Sucrose gradient velocity centtrifugation ofequine influenza virus (EIV) RNA labeled with "2Pand AMV RNA labeled with 3H-uridine. EIV andAMVwere purified simultanieously and their RNA's were

extracted together, and centrifuged in sucrose velocitygradients at 36,000 rev/mim for 2.25 hr in the SpincoSW 39 rotor. 32p radioactivity (broken line) and 3Hradioactivity (solid line) were determined as in Fig. 2.

VOL. 1, 1967 1219

28S

NAYAK AND BALUDA

in the gradient. There is no 3"P-labeled RNA com-ponent heavier than 21S. Thus, it appears thatby all available criteria, the virus purificationand theRNA extraction should not have degradedinfluenza viral RNA.To determine more precisely the sedimentation

coefficient of influenza viral RNA and its homo-geneity, the influenza viral RNA was reprecipi-tated (Fig. 3, fractions 17-23) and run in asucrose gradient along with 'H-labeled 18S ribo-somal RNA for 3.5 hr at 50,000 rev/min in aSpinco SW 50 rotor at 4 C. The results are pre-sented in Fig. 4. The peaks of optical density andof radioactivity of 18S ribosomal RNA coincided,and the peak of 32p radioactivity of the influenzaviral RNA sedimented before the 18S ribosomalRNA. Influenza RNA has a S2o,W of 215 by com-paring its relative position with that of 18S ribo-somal RNA. The shoulder at 18S was not due tononspecific binding with 18S ribosomal RNA, asit was also present in the absence of 18S ribosomalRNA. In addition, there was a 145 componentwhich had not been resolved at the lower cen-trifugation speed (36,000 rev/min for 2.25 hr,Fig. 3).

0.125

0.100

0.075E0

(D

F-

o 0050

0.02 5

-100

80

60

-Q

40

20

0

3 5 7 9 11 13 15 17 19 21 23FRACTION NUMBER

FIG. 4. Sedimentation profile of equine influenzavirus (EIV) RNA and 18S ribosomal RNA in sucrosevelocity gradient. 32P-ElV RNA from Fig. 3 was

reprecipitated and analyzed along with 'H-labeled 18Sribosomal RNA in sucrose gradient as in Fig. 3for 3.5hr at 49,000 rev/min. 32p (X), 'H (0), radioactivityand absorbance at 260 m,u (0) determined from sam-

ples ofeach fraction.

TABLE 1. Base composition of influenza virus RNA4

Moles per centRNA prepn

C A G U

Viral RNA ....... 22.2 22.9 22.3 32.6(+ )(-i1 .0) (4-1.2) (-i0.8)

Ribosomal RNA28Sb ............ 28.0 18.7 33.3 20.018Sb............ 27.7 19.7 27.7 24.9

a Base composition of influenza virus RNA wasdetermined by using 32P-labeled 18S viral RNAisolated from purified virus (Fig. 2). 32P-labeledvirus was prepared after 36 hr of incubation in em-bryonated eggs as described in Materials and Meth-ods. Base composition was determined after alka-line hydrolysis and paper electrophoresis. The meanper cent and standard deviation was determinedfrom two separate preparations of viral RNA andfour separate electrophoresis runs. Ribosomal RNAwas labeled with P2p by exposure of noninfectedcells to 32p for 4 hr in absence of actinomycin D, andthe 28S and 18S fractions were isolated by velocitysedimentation in sucrose.

b Single determination

The rate of sedimentation of influenza viralRNA was found to depend on the ionic strengthin the sucrose gradient. In 10-1 M salt, the viralRNA moved faster than in 10-3 M salt. The posi-tion of influenza viral RNA centrifuged in asucrose gradient containing 0.0005 M Tris-HCI(pH 7.4) and 0.0005 M EDTA (10-3 M salt) isindicated by an arrow in Fig. 2. This reflects thedependence of the configuration of viral RNA onthe charge distribution in its vicinity, a propertycharacteristic of single-stranded RNA. Also,more than 97% of the 32P radioactivity in isolatedviral RNA was rendered trichloroacetic acid-soluble by ribonuclease treatment in 2 X SSC(saline-sodium citrate), a condition in whichdouble-stranded RNA is resistant to ribonucleasetreatment (3).To determine whether the 21S viral RNA is a

single, covalently linked polynucleotide or anaggregation of components of smaller molecularweight, 32P-labeled viral RNA was dissolved inbuffer (Tris-HCl, pH 7.4, 0.01 M; EDTA, 0.001M), heated in sealed ampoules at 100 C for 2 min,and quenched in ice water. Before heating, 28Sribosomal RNA was added as marker. The sedi-mentation coefficient of viral RNA after heatingwas very similar but slightly lower (17S) thanthat before heating. This could have been due toan irreversible effect of thermal denaturation.Similar changes in sedimentation coefficient ofRNA have been shown to be caused by heatingand changes in ionic strength (7). It appears then

J. VIROL.1220

NUCLEIC ACID OF INFLUENZA VIRUS

200Tv

v-1 5 10 I5 20

FRACTION NUMBER

FIG. 5. Velocity sedimentation centrifugation of 3H-labeled RNA isolated from purified incomplete virions.Incomplete influenza virions were produced and purifiedas described in Materials and Methods. After extrac-tion in presence ofyeast soluble RNA carrier, the 'H-labeled RNA and a 28S ribosomal RNA marker were

centrifuged in a sucrose velocity gradient at 50,000revlmin for 2.5 hr at 4 C in the Spinco SW 50 rotor.The position of the 28S ribosomal RNA marker wasdetermined by absorbancy at 260 mu. The distributionof 'H label was determined as in Fig. 2.

that the 21S viral RNA is covalently linked andnot an aggregation product of smaller compo-nents. The average base analysis of "2P-labeled 21Sinfluenza viral RNA is shown in Table 1. Thevalues are incompatible with Watson-Crickcomplementary base pairing and indicate singlestrandedness of this RNA. Influenza viral RNAhas a high uridylic acid content (33.6%) and a

relatively low guanine plus cytosine content(44.5%). A similar base composition for in-fluenza viral RNA has been reported by Agrawaland Bruening (2) and Duesberg and Robinson(12).RNA from incomplete virions. Virions having

low infectivity were produced in chick embryofibroblasts infected at high moi as described inMaterials and Methods. To examine the natureof the RNA content of such viruses, RNA was

extracted from viral hemagglutinin purified bythe same technique as that used for normalvirions, and analyzed by velocity sedimentationin a sucrose gradient. The results are presented inFig. 5. There is no qualitative difference betweenRNA of noninfectious virus particles comparedto that isolated from infectious virions. The rela-tive amounts of 21S, 18S, and 14S are also similarto those exhibited by RNA isolated from infec-tious virions (Fig. 3).

DISCussIoN

The largest RNA component isolated frompurified influenza virus has a S20, of 21S. This21S RNA does not appear to be the degradedproduct of a native viral genome of higher molecu-lar weight. There are two critical steps where tIeviral RNA might be damaged: (i) purification ofvirus and (ii) extraction of RNA from purifiedvirus. Virus purified by two different procedures,e.g., density-interface sedimentation followed bydensity equilibrium centrifugation, or by hemad-sorption and elution, yielded RNA's of identicalsize. Finally, simultaneous purification of equineinfluenza virus and AMV and extraction of RNAafter mixing purified equine influenza virus andAMV yielded high molecular weight AMV-RNAand 215 influenza viral RNA (Fig. 3).An RNA component with a S20,W of 21S,

along with other RNA components, has beenobserved in the RNA of influenza viruses by otherworkers (2, 12, 18). A heavy RNA componentof about 34S, found by Agrawal and Bruening(2) and Pons (18), seems to be an aggregate of215 RNA caused by divalent cations as shownby Duesberg and Robinson (12). RNA compo-nents smaller than 21S found by other workers(12) have also been observed in this study. Thesewere resolved after extended centrifugation. It isnot known whether these components representnative components present in the virion or de-graded product of 21S RNA. Davies and Barry(10) have obtained only 18S RNA from influenzavirus purified by hemadsorption and elution.However, their RNA preparation had not beencentrifuged for a long enough time to resolveother components if any were present.The viral nature of the 21S RNA isolated

from virus is evident. The base composition ofuniformly labeled 21S viral RNA is in no waysimilar to that of 18S ribosomal RNA (Table 1;2, 20). Evidence to be published elsewhere (D.P. Nayak, in preparation) shows that the RNAisolated from purified virus causes displacementof plus (viral) strands from duplexes of plus andminus (complementary) intracellular viral RNAstrands produced by annealing, and competes with

28S

I150 -

ro)I

.EU_ 100 -

z

0

50 -

0

VOL. 1, 1967 1221

NAYAK AND BALUDA

single-stranded viral RNA for annealing withcomplementary strands. Also, 32P-labeled viralRNA anneals with intracellular complementaryRNA. Finally, 21S RNA can be dissociatedfrom 18S ribosomal RNA after prolonged cen-trifugation (Fig. 3).The 21S viral RNA is single-stranded and

exists in coiled form, as is evident from its basecomposition, ribonuclease sensitivity, and de-pendence of the sedimentation rate on salt con-centration. It does not contain complementaryRNA, as was found by its inability to self-anneal.The 21S viral RNA is covalently linked and with-stands thermal denaturation to a large extent.However, following heat denaturation, all thehydrogen bonds cannot be restored and theoriginal helical form cannot be fully regained, asshown by its slightly lower sedimentation coeffi-cient (17S) after heating and quenching in lowsalt. Changes in S20,W caused by changes in thehelical configuration of MS2 RNA have beenreported by Bishop (7). He found that single-stranded MS2 RNA can exist in three differentforms having S20.W of 28S, 20S, or 12S, whichare dependent upon heat or salt treatment.A sedimentation coefficient of 21S for influ-

enza viral RNA was determined with two differ-ent markers, AMV-RNA having a S20,W of 71Sand 18S ribosomal RNA. Both of these RNAspecies have well-characterized sedimentationcoefficients which have been determined in theanalytical ultracentrifuge (19). A 21S single-stranded RNA would have a molecular weight ofabout 0.9 x 106 according to Spirin's formula(25). It is difficult to account for all the knownfunctions coded for by influenza virus RNA, e.g.,replication of RNA, synthesis of hemagglutinin,neuraminidase, and coat antigens in a single RNAmolecule of 21S. Therefore, one has to assumethat a virion contains more than one RNAmolecule, since the average estimation of 0.8%of RNA (1, 16) per virion corresponds to RNAhaving a molecular weight equivalent of 2 x 106daltons. A similar conclusion has been reachedby other workers (10, 12, 18). Hirst (14) postu-lated more than one RNA strand per virion toaccount for the high rate of genetic recombina-tion observed in influenza viruses. The resultspresented above also indicate that there are sev-eral RNA strands in an infectious virion. Themanner in which these RNA molecules are associ-ated in the virion is unknown.

There is no qualitative difference between theRNA isolated from incomplete virions and thatisolated from infectious virus. However, if therewere any quantitative difference in RNA con-tent in the noninfectious particle compared to

that in infectious virions, it could not be de-tected by this procedure.

ACKNOWLEDGMENTSWe thank A. F. Rasmussen, Jr., for encourage-

ment and support.This work was done under the sponsorship of the

Commission on Influenza of the Armed Forces Epi-demiological Board, and was supported by the U.S.Army Medical Research and Development Command,contract DA-49-193-MD-2495, and by Public HealthService research grant CA-10197 from the NationalCancer Institute.

LITERATURE CITED

1. ADA, G. L., AND B. T. PERRY. 1955. Infectivityand nucleic acid content of influenza virus.Nature 175:209-210.

2. AGRAWAL, H. 0., AND G. BRUENJNG. 1966.Isolation of high-molecular-weight, P32-labeledinfluenza virus ribonucleic acid. Proc. Natl.Acad. Sci. U.S. 55:818-825.

3. BALTIMORE, D. 1966. Purification and propertiesof poliovirus double-stranded ribonucleicacid. J. Mol. Biol. 18:421-428.

4. BALUDA, M. A. 1967. Subviral carcinogenesis,p. 19-35. First Intern. Symp. on TumorViruses, Aichi Cancer Center, Nagoya, Japan.

5. BANG, F. B. 1955. Morphology of viruses. Ann.Rev. Microbiol. 9:21-44.

6. BARRY, R. D. 1964. The effects of actinomycin Dand ultraviolet irradiation on the production offowl plague virus. Virology 24:563-569.

7. BISHOP, D. H. L. 1966. Physical properties ofsingle- and double-stranded coliphage ribo-nucleic acid. Biochem. J. 100:321-329.

8. BLOUGH, H. A. 1964. Ciba Found. Symp. CellularBiol. Myxovirus Infections, p. 120-151.

9. BRAY, G. A. 1960. A simple efficient liquidscintillator for counting aqueous solutions in aliquid scintillation counter. Anal. Biochem.1:279-285.

10. DAV1ES, P., AND R. D. BARRY. 1966. Nucleic acidof influenza virus. Nature 211:384-387.

11. DUESBERG, P. H., AND W. S. ROBINSON 1965.Isolation of the nucleic acid of Newcastledisease virus (NDV). Proc. Natl. Acad. Sci.U.S. 54:794-800.

12. DUESBERG, P. H., AND W. S. ROBINSON. 1967.On the structure and replication of influenzavirus. J. Mol. Biol. 25:383-405.

13. HENLE, W., AND 0. C. LIu. 1951. Studies onhost-virus interactions in the chick embryo-influenza virus system. J. Exptl. Med. 94:305-322.

14. HIRST, G. K. 1962. Genetic recombination withNewcastle disease virus, polioviruses, andinfluenza. Cold Spring Harbor Symp. Quant.Biol. 27:303-309.

15. KINGSBURY, D. W. 1966. Newcastle diseasevirus RNA. I. Isolation and preliminarycharacterization of RNA from virus particles.J. Mol. Biol. 18:195-203.

1222 J. VIROL.

NUCLEIC ACID OF INFLUENZA VIRUS

16. MILLER, H. K. 1956. The nucleic acid content ofinfluenza virus. Virology 2:312-320.

17. NAYAK, D. P., AND A. F. RASMUSSEN. 1966.Influence of mitomycin C on the replication ofinfluenza viruses. Virology 30:673-683.

18. PONS, M. W. 1967. Studies on influenza virusribonucleic acid. Virology 31:523-531.

19. ROBINSON, W. S., AND M. A. BALUDA. 1965.The nucleic acid from avian myeloblastosisvirus compared with the RNA from the Bryanstrain of Rous sarcoma virus. Proc. Natl. Acad.Sci. U.S. 54:1686-1692.

20. ROBINSON, W. S., A. PITKANEN, AND H. RUBIN.1965. The nucleic acid of the Bryan strain ofRous sarcoma virus: purification of the virusand isolation of the nucleic acid. Proc. Natl.Acad. Sci. U.S. 54:137-144.

21. Rorr, R., AND W. SCHAFER. 1964. Ciba Found.Symp. Cellular Biol. Myxovirus Infections,p. 27-42.

22. SCHXFER, W. 1963. Structure of some animalviruses and significance of their components.Bacteriol. Rev. 27:1-17.

23. SIMPSON, R. W. 1964. Ciba Found. Symp. Cellu-lar Biol. Myxovirus Infections, p. 187-206.

24. SOKOL, F., S. SCHRAMEK, AND J. SPONAR. 1963.Comparison of different preparations ofinfluenza virus ribonucleic acid. Biochem.Biophys. Res. Commun. 12:21-26.

25. SPIRIN, A. S. 1963. Some problems concerningthe macromolecular structure of ribonucleicacids. Progr. Nucleic Acid Res. 1:301-345.

26. VON MAGNUS, P. 1951. Propagation of the PR 8strain of influenza A virus in chick embryos.Acta Pathol. Microbiol. Scand. 28:278-293.

VOL. 1, 1967 1223