Vol. 15, No. 6 Printed in U.S.A.

JOURNAL OF VIROLOGY, June 1975, p. 1342-1347 Copyright 0 1975 American Society for Microbiology

Characterization of Bovine Viral Diarrhea Virus RNA RANDALL PRITCHETT, JARUE S. MANNING, AND YUAN CHUNG ZEE* Department of Veterinary Microbiology and Department of Bacteriology, University of California, Davis, California 95616

Received for publication 18 October 1974

RNA extracted from isopycnically banded [3H]uridine-labeled bovine viral diarrhea virus with sodium dodecyl sulfate was resolved into one major and two minor components by both sedimentation analysis and electrophoresis in polyacrylamide gels. The major RNA component was estimated to have a 38S sedimentation coefficient. The minor RNA components were estimated to have S values of 31 and 24. The approximate molecular weights were calculated to be 3.22 x 106 (38S), 2.09 x 106 (31S), and 1.22 x 106 (24S). A single broad peak of radioactivity, maximum at 24S, was obtained when sedimentation was conducted under conditions of low ionic strength. All three RNA components were found to be susceptible to digestion with RNase. The presence of multiple RNA components in heterogeneous populations of infectious virus is discussed.

Bovine viral diarrhea virus (BVDV), the causative agent of viral diarrhea-mucosal disease in cattle (12, 20), is ether sensitive (7) and contains RNA as its genetic material (7). The BVD virion is reported to be approximately spherical in shape, to possess an envelope with few prominant surface projections, and to have a spherically shaped central core (6, 8, 9, 15). An uncommonly wide range of infectious particle size has been reported, i.e., 30 to greater than 100 nm (6, 8, 9, 15). This heterogeneity in the infectious particle size has been demonstrated by rate-zonal centrifugation in glycerol gradients (8). In addition, buoyant density-infectivity determinations conducted on various strains of BVDV indicate that infectious particles are also heterogeneous with respect to density (5, 9, 11). In light of the heterogeneity in both the size and density of the infectious particles of BVDV, it was of interest to determine if the nucleic acid of such particles also reflected this heterogeneity. In this paper we report on the finding of multiple RNA components in isopycnically banded BVDV preparations. The relationship of this finding with respect to the heterogeneity of infectious particles is discussed. MATERIALS AND METHODS Virus and cells. The propagation and assay of BVDV, strain NADL, in fetal bovine bone marrow cells were described earlier by Parks et al. (11). Vero cells were grown as described previously (4). Buffers. TEN buffer consisted of 0.1 M Tris-hydrochloride (pH 7.4), 0.1 M NaCl, and 0.001 M EDTA. SSC (2x) buffer contained 0.15 M NaCl and 0.15 M

sodium citrate, pH 7.7. AES buffer consisted of 0.01 M sodium acetate, 0.001 M EDTA, and 0.1 M NaCl which was adjusted to pH 5.1 by the addition of 0.01 M acetic acid. Extraction buffer contained 4 gg of dextran sulfate per ml, 0.14 M NaCl, 0.01 M Trishydrochloride (pH 7.2), and 0.0015 M MgCl2. Electrophoresis buffer consisted of 0.04 M Tris-acetate (pH 7.2), 0.001 M EDTA, and 0.2% sodium dodecyl sulfate (SDS). Chemicals. [5-3HJuridine (28 Ci/mmol) and carrier-free [32P ]orthophosphorous were obtained from Schwarz/Mann, Van Nuys, Calif. SDS, acrylamide, N,N'-methylene-bisacrylamide, and N,N,N'N'-tetramethylenediamine were purchased from Eastman Organic Chemical Co., Rochester, N.Y. Omnifluor and Protosol were obtained from New England Nuclear, El Cerrito, Calif. Bovine RNase was obtained from Gallard-Schlesinger Chemical Mfg. Corp. (Carle Place, Long Island, N.Y.), and bovine pancreatic DNase was obtained from Calbiochem (San Diego, Calif.). Radioisotope labeling of viral and cellular RNA. Labeling of viral RNA was accomplished by the addition of [HH]uridine (25 uCi/ml) in Eagle minimal essential medium plus 5% fetal calf serum to cell cultures at 8 h after infection. Extracellular virus was harvested at 72 h postinfection. Uninfected cell cultures were similarly labeled. For labeling of cellular RNA, partially confluent monolayers of Vero cells were incubated for 30 h at 37 C in phosphate-free balanced salt solution to which had been added [2P ]orthophosphate (20 ,Ci/ml) and 5% dialyzed fetal calf serum. Purification of virus. The methods of concentration and partial purification by isopycnic banding were slightly modified from those described previously (11). Briefly, extracellular fluids harvested from infected, [3H]uridine-labeled cultures were clarified by centrifugation at 5,900 x g for 15 min at 4 C. Twenty milliliters of the clarified fluid was layered 342

VOL. 15, 1975

1343

BOVINE VIRAL DIARRHEA VIRUS RNA

onto a 25 to 50% (wt/wt) sucrose discontinuous gradient and was centrifuged in a Spinco SW25.1 rotor at 25,000 rpm for 2 h at 4 C. Sucrose density gradients were prepared using TEN buffer. One lightscattering band developed at the supernatant-25% sucrose interface (upper) and another at the 25 to 50% sucrose interface (lower). The upper band, which was found to contain the majority of infectivity, was diluted in TEN buffer, layered onto a continuous 15 to 40% (wt/wt) sucrose gradient and centrifuged in an SW25.1 rotor at 25,000 rpm for 20 h at 4 C. One-milliliter fractions were collected and an aliquot from each fraction was assayed for infectivity and acid-precipitable radioactivity (11). Fractions from the region of the gradient where the peaks of infectivity and radioactivity were coincident were pooled and stored at -70 C. Such samples served as the stock of isopycnically banded virus used in subsequent studies. Uninfected bovine bone marrow cell cultures, labeled with [H ]uridine to serve as a control, were frozen and thawed three times and processed by identical procedures served as controls. Isolation of viral and celiular RNA. Viral RNA was obtained from isopycnically banded virus. Virus was pelleted by centrifugation at 50,000 rpm for 90 min in a Spinco SW50L rotor at 4 C and RNA was extracted by suspending the virus pellet in AES buffer containing 1% SDS and 4.0 ,g of dextran sulfate per ml. The suspension was incubated at 37 C for 30 min and then heated to 60 C for 10 min. For preparation of cellular RNA (2, 18) cultures of Vero cells which had been incubated in the presence of [82P]orthophosphate were washed three times with Eagle minimal essential medium and cells scraped into extraction buffer. Cells were lysed by the addition of NP-40 to a concentration of 0.5%. The cytoplasmic extract obtained after low-speed centrifugation (600 x g) was treated with a one-tenth volume of 10% SDS to liberate the RNA. Such SDStreated cytoplasmic extracts were assayed for acidprecipitable radioactivity and stored at - 70 C. Velocity sedimentation of viral and cellular RNA. Sedimentation of viral and cellular RNA was performed on linear 5 to 20% (wt/wt) sucrose gradients in an SW50L rotor at 50,000 rpm for 90 min at 18 C. Gradients were prepared with AES buffer, pH 5.1. After piercing a hole in the bottom of the centrifuge tube, fractions of approximately 0.25 ml each were obtained by bottom puncture on filter paper strips for radioassay of acid-insoluble radioactivity (16). Samples were counted in a Nuclear Chicago Mark HI liquid scintillation spectrometer. Correction was made for 32p spillover into the 3H channel. Polyacrylamide gel electrophoresis. Electrophoresis in polyacrylamide gels was performed essentially according to the methods described elsewhere (10, 15) except that acrylamide (2.4%) was cross-linked with 0.12% N',N'-methylene bisacrylamide in place of ethylene diacrylate. Gels were polymerized with 0.08% ammonium persulfate in glass tubes (0.6 by 10 cm) closed at the lower end with a double thickness of dialysis tubing. Samples to be subjected to electrophoresis were adjusted to approximately 10% glycerol

just prior to application onto the gel column. A current of 2.5 mA/gel was applied for 15 min, after which time it was increased to 5.0 mA/gel and maintained for 5.75 h. When electrophoresis was completed, gels were left in their glass tubes and frozen by slowly emersing them into -70 C ethanol. Methods used in the slicing of frozen low-concentration polyacrylamide gels have been described earlier (17). Gel slices (1.6 mm) were collected in scintillation vials. Ten milliliters of solution consisting of 3% Protosol in an Omnifluor-toluene counting fluid were added and the vials were incubated at 37 C on a shaker for 16 h to allow elution of the RNA. Samples were then counted in a liquid scintillation spectrometer.

RESULTS Buoyant density characterization of BVDV. Tritiated uridine-labeled virus was isopycnically banded and gradient fractions

I0

-J

5.5 ,

I-

,-

I-.

5.0>

C.)

N

C.)

L&

U-

4.5 Z

a-

.-J

C-,

z

4.0 CD0

W'

-J

Top 5

20 15 10 FRACTION NUMBER

25

Bottom

FIG. 1. Isopycnic-zonal centrifugation of [3H]uridine-labeled BVDV. Virus concentrated onto a 25% sucrose interface was diluted in TEN buffer, layered on top of a linear 15 to 40% sucrose gradient, and centrifuged at 25,000 rpm for 20 h at 4 C using a Spinco SW25.1 rotor. Fractions were collected and the radioactivity (0) and infectivity (-) of each fraction were determined. Radioactivity of uninfected culture fluids (A) processed in parallel was also determined. The solution density (-) was determined by weighing 100-ul aliquots.

1344

J. VIROL.

PRITCHETT, MANNING, AND ZEE

8

16

7

14

X6 C~L

10

-4

8

12

C)

cr

3

6

2

4

2

5

10

15

20

5

10

15

20

FRACTION NUMBER FIG. 2. Velocity sedimentation of BVDV RNA. [3Hjuridine-labeled viral RNA was extracted from isopycnically banded virus with 1% SDS, layered onto a 5 to 20% (wt/wt) linear sucrose gradient and centrifuged in a Spinco SW50L rotor at 50,000 rpm for 90 min at 18 C. Gradients were fractionated and acid-insoluble radioactivity was determined. (A) Aliquots of gradient fractions were diluted 10-fold with 2x SSC containing 0.003 M MgCI2 and incubated for I h at 37 C with either 20 ,g of RNase (0) per ml or 50 Asg of DNase (0) per ml. (B) BVDV RNA control; no enzymatic treatment.

assayed for radioactivity and infectivity. Good correlation was observed between the infectivity and acid-precipitable radioactivity profiles as is illustrated in Fig. 1. Although there was a minor peak of radioactivity at a density of 1.145 g/cm 3, the major peak of radioactivity corresponded with that of infectivity at a density of 1.12 g/cm3. When uninfected [3H]uridine-labeled cultures were processed in parallel, a minor peak of radioactive material was found at a density of 1.145 g/cm3 (Fig. 1); however, only insignificant amounts of radioactivity were present at a density of 1.12 g/cm3. Such evidence suggests that the material banding at 1.145 g/cm3 is of cellular origin. Velocity sedimentation of BVDV RNA. [3H]uridine-labeled RNA which had been extracted from isopycnically banded virus with SDS was analyzed by velocity sedimentation in linear 5 to 20% sucrose gradients. The radioactive material was resolved into one major and two minor components (Fig. 2). The sedimentation coefficient of the major rapidly sedimenting component was estimated to be 38S with respect to cosedimented 32P-labeled Vero cell ribosomal (18S and 28S) RNA markers. The slower sedimenting components had S values of 31 and 24. The approximate molecular weights were

of the 31S and 24S components were calculated by the method of Spirin (19) to be 3.22 x 106 and 1.22 x 10, respectively. Analysis of the nucleic acid of BVDV was conducted under conditions of low ionic strength by preparing sucrose solutions in buffer containing 0.001 M NaCl rather than the standard 0.1 M concentration. The results, illustrated in Fig. 3, show an altered sedimentation pattern as compared to that seen at higher ionic strength. Instead of the multiple sedimenting components, there was only a single rather broad peak of radioactivity that was maximum at approximately 24S. Sensitivity of BVDV RNA to enzymatic digestion. To further determine the nature of the viral nucleic acid, the radioactive material contained in fractions from a velocity sedimentation gradient was tested for sensitivity to enzymatic digestion. Each fraction was divided into equal portions. One portion was incubated with RNase and the other with DNase. These two samples and an untreated BVDV RNA sample were centrifuged on rate-zonal gradients. As shown in Fig. 2, it was found that the [3H ]uridine-labeled components were completely digested by the action of RNase but were resistant to the action of DNase. The pattern of

VOL. 15, 1975

1345

BOVINE VIRAL DIARRHEA VIRUS RNA

N

10 x

CL

0-

L-

z

;z

I

c= I

10

20 30 40 SLICE NUMBER

50

FIG. 4. Polyacrylamide gel electrophoreogram of

[aH]uridine-labeled RNA from BVDV. Viral RNA

15 5 10 FRACTION NUMBER

20

FIG. 3. Effect of low ionic strength on the sedimentation pattern of BVDV RNA. Extraction procedures, gradients, and centrifugation were as described for Fig. 2, except that sucrose solutions were prepared in AES buffer containing 0.001 M NaCI. '2P-labeled Vero cell ribosomal RNA was centrifuged on a separate gradient prepared with standard AES buffer and the relative positions of the 28S and 18S RNA are indicated by vertical arrows. the RNA sedimentation after DNase treatment was the same as the pattern for untreated BVDV RNA (Fig. 2B), which shows that the pattern obtained after DNase digestion was not due to RNase contamination. Polyacrylamide gel electrophoresis of BVDV RNA. A portion of the RNA sample utilized in the sedimentation studies was also analyzed by electrophoresis in 2.4% polyacrylamide gels. Radioactivity profiles of gels after electrophoresis consistently revealed the presence of three distinct components (Fig. 4). The migration mobilities of the viral RNA compo-

extracted as described for Fig. 2 was subjected to co-electrophoresis with 32P-labeled Vero cell ribosomal RNA in 2.4% acrylamide gels. Relative positions of the 28S and 18S ribosomal RNA markers are indicated by vertical arrows.

nents estimated by electrophoresis of Vero cell ribosomal [32P ]RNA were 38S, 31S, and 24S, which is in agreement with the values determined for their sedimentation rates. Some viral preparations contained nucleic acid having electrophoretic properties similar to that of 28S and 18S.

DISCUSSION Coincident peaks of infectivity and radioactivity at a solution density of 1.12 g/cm3 were obtained after isopycnic centrifugation of concentrated [3H]uridine-labeled BVDV. Such an observation indicates that the radiolabel had been incorporated into viral RNA. There was also a minor peak of radioactivity observed at a solution density of 1.145 g/cm3. When [3H ]uridine-labeled uninfected control cultures were processed in parallel, a peak of radioactivity was again found at 1.145 g/cm9, but only an insignificant amount was present at 1.12 g/cm'. Therefore, it was concluded that the material

1346

PRITCHETT, MANNING, AND ZEE

banding at 1.145 g/cm3 was of cellular origin. The RNA extraced from isopycnically banded BVDV was resolved into three components by both velocity sedimentation and polyacrylamide gel electrophoresis. An RNA species with a sedimentation rate of 38S corresponding to a mol wt of 3.22 x 106 represented the major component. Two minor components were also separated, one of 2.09 x 106 mol wt (31S) and another of 1.22 x 106 mol wt (24S). All three RNA components were found to be susceptible to RNase digestion which suggests that none are double stranded in structure. We have observed that SDS-extracted BVDV RNA sediments as a single broad peak at approximately 24S under conditions of low ionic strength. A dependence of sedimentation rate on ionic strength has been shown to be characteristic of single-stranded RNA (1). It was not experimentally determined, however, whether the conversion of the S value was due to disaggregation or due to an extensive conformational change. As mentioned previously, available evidence indicates that the size of the infectious particle of BVDV is heterogeneous. It is of interest, therefore, to find that the RNA extracted from partially purified infectious preparations is also heterogeneous. A possible explanation is that particles containing incomplete genome equivalents (i.e., less than 3.22 x 106 mol wt) are responsible for the observed heterogeneity in the RNA obtained from a population of viruses. Infection of cells with incomplete but genetically complementing virus particles could account for the heterogeneity in the size of the infectious particle genome. Alternatively, the high-molecular-weight component of RNA may represent a polyploid genome. If such were the case, particles containing lower-molecularweight RNA (presumably one genome equivalent) could be smaller in size yet still be infectious. The present experimental data is not sufficient to distinguish between these two alternatives. The observation that the molecular weights of the two minor RNA components added together result in a value approximating that of the major component would seem to support either hypothesis. Several other possible explanations for the occurrence of multiple RNA components in isopycnically banded BVDV have been considered. (i) The minor RNA components may have resulted from random degradation of the 38S component during extraction; however, random degradation would more likely have resulted in the appearance of a broad sedimentation pattern rather than distinct peaks. (ii) The RNA

J. VIROL.

may have been digested by RNase which would result in a similarly broad sedimentation pattern; however, this was not observed and therefore considered unlikely. (iii) The multiple RNA species may represent a segmented genome analogous to that of influenza virus (13, 14), although this does not seem likely since infectious RNA has been extracted from BVDV (3). The 31S and 24S species of BVDV RNA may represent incomplete viral nucleic acid or perhaps degradation products of 38S RNA. The data presented herein indicate that the RNA extracted from an isopycnically banded population of infectious BVDV is heterogeneous. The origin of this heterogeneity is not known. Additional experimentation will be required to determine the precise size of the infectious BVDV genome as well as the origin and properties of the various species of viral RNA detected in these studies. LITERATURE CITED 1. Boedtker, H. 1960. Configurational properties of tobacco mosaic virus RNA. J. Mol. Biol. 3:171-188. 2. Borun, T. W., M. D. Schartt, and E. Robbins. 1967. Preparation of mammalian polyribosomes with the detergent Nonidet P-40. Biochim. Biophys. Acta 149:302-304. 3. Diderholm, H., and Z. Dinter. 1966. Infectious RNA derived from bovine virus diarrhea virus. Zentralbl. Bakteriol. Parisitenkd., Infektionskr. Hyg. Abt. I. Orig. 201:270-272. 4. Earley, E., P. H. Peralta, and K. M. Johnson. 1967. A plaque neutralization method for arboviruses. Proc. Soc. Exp. Biol. Med. 125:741-747. 5. Fernelius, A. L. 1968. Characterization of bovine viral diarrhea viruses. I. Determination of buoyant density. Arch. Gesamte Virusforsch. 25:211-218. 6. Hafez, S. M., M. K. Perzoldt, and E. Reczko. 1968. The morphology of bovine virus. Acta Virol. 12:471-473. 7. Hermodsson, S., and Z. Dinter. 1962. Properties of the bovine virus diarrhea virus. Nature (London) 194:893-894. 8. Horzinek, M., J. Maess, and R. Laufs. 1971. Studies on the substructure of togaviruses. II. Analysis of equine arteritis, rubella, bovine viral diarrhea, and hog cholera viruses. Arch. Gesamte Virusforsch. 33:306-318. 9. Maess, J., and E. Reczko. 1970. Electron optical studies of bovine viral diarrhea-mucosal disease virus (BVDV). Arch. Gesamte Virusforsch. 30:39-46. 10. Manning, J. S., F. L. Schaffer, and M. E. Soergel. 1972. Correlation between murine sarcoma virus buoyant density, infectivity and viral RNA electrophoretic mobility. Virology 49:804-807. 11. Parks, J. B., R. F. Prtichett, and Y. C. Zee. 1972. Buoyant density of bovine viral diarrhea virus. Proc. Soc. Exp. Biol. Med. 140:594-598. 12. Peter, C. P., D. E. Tuler, and F. K. Ramsey. 1967. Characteristics of a condition following vaccination with bovine virus diarrhea vaccine. J. Am. Vet. Med. Assoc. 150:46-52. 13. Pons, M. W. 1967. Studies on influenza virus ribonucleic acid. Virology 31:523-531. 14. Pons, M. W., and G. K. Hirst. 1968. Polyacrylamide gel electrophoresis of influenza virus RNA. Virology 34:385-388. 15. Ritchie, A. E., and A. L. Fernelius. 1969. Characteriza-

VOL. 15, 1975

BOVINE VIRAL DIARRHEA VIRUS RNA

tion of bovine viral diarrhea viruses. V. Morphology of characteristic particles studied by electron microscopy. Arch. Gesamte Virusforsch. 28:369-389. 16. Schaffer, F. L., A. J. Hackett, and M. E. Soergel. 1968. Vesicular stomatitis virus RNA: complementarity between infected cell RNA and RNA's from infectious and autointerfering viral fractions. Biochem. Biophys. Res. Commun. 31:685-692. 17. Schaffer, F. L., M. E.- Soergel, and D. C. Straube. 1971. Electrophoretic analysis of ribosomal and viral ribonucleic acids with a simple technique for slicing low-concentration polyacrylamide gels. Appl. Microbiol. 22:538-545.

1347

18. Soergel, M. E., and F. L. Schaffer. 1972. Ribosomal ribonucleic acids of cultured cells: a preliminary survey of differences among mammalian species detectable by polyacrylamide gel electrophoresis. In Vitro 8:13-18. 19. Spirin, A. S. 1963. Some problems concerning the macromolecular structure of ribonucleic acids, p. 301-345. In J. N. Davidson and W. E. Cohn (ed.), Progress in nucleic acid research, vol. 1. Academic Press Inc., New York. 20. Thomson, R. G., and M. Savan. 1963. Studies on virus diarrhea and mucosal disease of cattle. Can. J. Comp. Med. Vet. Sci. 72:217-224.

Characterization of bovine viral diarrhea virus RNA.

RNA extracted from isopycnically banded [3-H]uridine-labeled bovine viral diarrhea virus with sodium dodecyl sulfate was resolved into one major and t...
762KB Sizes 0 Downloads 0 Views