Molecular weights of the RNA genome segments of a cytoplasmic polyhedrosis virus determined by a new comparative approach

VIROLOGY76, 210-216

(1977)

Molecular Weights of the RNA Genome Segments of a Cytoplasmic Polyhedrosis Virus Determined by a New Comparative Approach’ ERIC H. HARLEY,

M.R.C.U.C.T. Research Unit,

RIVA

RUBINSTEIN,z MICHELE DIANA LUTTON

LOSMAN,

AND

Protein Research Unit, Department of Chemical Pathology, and 2M.R.C.IU.C.T. Department of Bacteriology, University of Cape Town Medical School, Observatory Cape Town, Republic of South Africa Accepted

August

Virus 7925,

17,1976

A cytoplasmic polyhedrosis virus (CPV) from Heliothis armigera contains 10 doublestranded RNA genome segments which are all well resolved on polyacrylamide gel electrophoresis. The relationship between log molecular weight and electrophoretic mobility in polyacrylamide gels has been studied using 3ZP-labeled CPV RNA, and has been shown to deviate markedly from linearity at the higher molecular weight values. A new relationship between molecular weight and electrophoretic mobility for doublestranded RNA has therefore been defined and used to determine the absolute molecular weight of the CPV RNA by comparison with reovirus type 3 RNA. INTRODUCTION

A cytoplasmic polyhedrosis virus (CPV) isolated from Heliothis armigera (Rubinstein et al., 1975) has been shown to contain 10 well-spaced double-stranded RNA genome segments (Rubinstein et al., 1976). The most practical method for the determination of the molecular weights of the double-stranded RNA segments of multicomponent viruses is comparative polyacrylamide gel electrophoresis. This is the only method with sufficiently fine resolution to resolve species with small differences in molecular weight. Bishop et al. (1967) demonstrated a linear relationship between the log of the molecular weight and electrophoretic mobility for single-stranded RNA species. Since variations in secondary structure can give rise to deviations from linearity, electrophoresis is best performed in gels polymerized in denaturing solvents such as formaldehyde (Boedtker, 1971) and formamide (Staynov et al., 1972). It has 1 This work was reported in part at the 3rd International Congress for Virology, Madrid, Spain, September 1975 (Workshop 43). 210 Copyright All righta

0 1977 by Academic F’reas, Inc. of reproduction in any form reserved.

been assumed frequently that a similar linear relationship between log molecular weight and electrophoretic mobility applies to double-stranded RNA, and many determinations of the molecular weight of double-stranded viral RNA have been reported based on this relationship. However, the electrophoretic mobility characteristics of double-stranded RNA in polyacrylamide gels have been studied by Harley et al. (1973) and are very similar to those of double-stranded linear DNA, where the electrophoretic mobility approaches a lower limiting value (considerably greater than zero) at any fixed gel concentration, as the molecular weight increases. Consequently, this must throw doubt on the validity of a linear relationship between log molecular weight and electrophoretic mobility for doublestranded RNA. A new relationship between log molecular weight and electrophoretic mobility for double-stranded RNA in polyacrylamide gels has therefore been defined and shows a marked departure from linearity which becomes increasingly pronounced at molecular weight values greater than 106.

MOLECULAR

WEIGHT

Using this new relationship, the absolute molecular weight values for the genome segments of the CPV were determined by comparative gel electrophoresis. METHODS

Chemicals [5-3H]Uridine (29 Ci/mmol), 12-14Clthymidine (50 Ci/mol), and [32P]orthophosphate were purchased from the Radiochemical Centre, Amersham. Acrylamide and N,N’-methylene bisacrylamide were purchased from B.D.H. and recrystallized as described by Loening (1967). Scintillation fluid consisted of 40 g of Beckman Fluoralloy TLA (16 parts butyl PBD: 1 part PBBO) and 400 ml of Biosolve (Beckman) made up to 5 liters with toluene. Pancreatic ribonuclease type X1-A was obtained from Sigma. Actinomycin D was obtained from Merck Sharpe & Dohme, Inc. Production

of Viruses

CPV was isolated from and maintained in larvae of H. armigera as described previously (Rubinstein et al., 1975, 1976). 32PLabeled viral RNA was prepared by adding 32P, as orthophosphate, generally 100 PCilml on the day of addition, in 0.1 ml of water, to the diet surface of g-day-old larve (4 days postinfection) after 12 hr of starvation. Midguts were dissected out 72 hr later. Reovirus type 3 was a gift from Dr. D. W. Verwoerd; BHK cells were infected and the virus was harvested and purified as described by Bellamy et al. (1967). The CCL2 strain of HeLa cells for the propagation of poliovirus type 1 and the CCL 10 strain of BHK 21 cells were obtained from Flow Laboratories and propagated at 37” in monolayer culture in Falcon flasks. Media consisted of Eagle’s minimal essential medium buffered with bicarbonate and supplemented with 5% calf serum (Flow). It contained 100 units of penicillin and 100 pg of streptomycin per milliliter of medium. Cultures were monitored for mycoplasma contamination by methods described previously (Harley et al., 1970).

OF

CPV

Preparation

211

RNA

of Nucleic

Acids

To prepare 14C-labeled DNA, [214C]thymidine was added at a concentration of 0.2 &i/ml for 15 hr to HeLa monolayers and DNA was extracted as described previously (Harley et al., 1970). Double-stranded RNA from reovirus and CPV was extracted either by the method used for HeLa cell DNA or by the method of Scherrer and Darnell (1962). Similar yields and analytical profiles were obtained whichever method was used; omitting the heating stage in the latter method also gave similar results. “H-Labeled poliovirus RNA, including the RF (replicative form), was prepared from cells infected at a multiplicity of about 20 PFU and labeled from 3l/2 to 5l/2 hr after infection with 1 $Zi/ml of [5-3Hluridine in the presence of 2.5 pg/ml of actinomycin D (added at the time of infection). Total RNA was extracted by the hot ph&ol method of Scherrer and Darnell (1962), with the addition of Bentonite at a concentration of 1 mg/ml . Poliovirus RNA, and in some experiments CPV RNA, was digested with RNase by dissolving in 100 ~1 of 1 x SSC (0.15 M NaCl, 0,015 M Na citrate, pH 7.0) containing 1 pg/ml of pancreatic ribonuclease and by incubating for 30 min at 37”. The reaction was stopped by the addition of SDS to l%, and after addition of glycerol to 15%, the products of digestion were analyzed directly by polyacrylamide gel electrophoresis. The 3H-labeled replicative RNA of foot and mouth disease virus (FMDV-RF) was a gift from Dr. F. Brown and was prepared from virus-infected cells by phenol extraction followed by precipitation with 1.5 M NaCl to remove singlestranded RNA. The supernatant fraction was centrifuged in a sucrose density gradient, and the fractions sedimenting at approximately 20 S were used after alcohol precipitation. No ribonuclease was used in the preparation of FMDV-RF. Polyacrylamide Gel Electrophoresis This was performed in Tris-phosphate/ EDTA buffer (Loening, 1967) as described previously (Harley et al., 1973). Gels were scanned at 260 nm in a Varian Model 635

212

HARLEY

spectrophotometer and, where appropriate, sliced with a Mickle Laboratories gel slicer. For 32P estimation, slices 1 mm thick were put into vials containing 5 ml of water, incubated at 37” for at least 3 hr, and then counted by Cerenkov radiation. Gel slices containing 3H- or 14C-labeled RNA were incubated for 18 hr in 0.3 ml of 2 M ammonia solution at 37” in scintillation vials, before addition of 10 ml of scintillation fluid. If gels contained labeled DNA, the slices were given a preliminary incubation with 1 N HCl as described previously (Harley and White, 1973). RESULTS

The electrophoretic mobilities of the CPV RNA segments were obtained from 20 sets of measurements on 2.6% polyacrylamide gels run under similar conditions. Electrophoretic mobilities were calculated relative to segment one for increased relative precision and are listed in Table 1. From these values, absolute electrophoretie mobility was determined by coelectrophoresing CVP RNA with DNA. The absolute electrophoretic mobility of the latter at 2.6% acrylamide concentration has been determined previously (Harley et al., 1973). Reovirus type 3 RNA was chosen as a molecular weight standard since the average molecular weights of its three main genome segment groupings have been determined previously by a number of different methods, summarized by Bellamy et al. (1967) and Shatkin et al. (1968). Reovirus RNA and CPV RNA, therefore, were coelectrophoresed, and 260~nm scans of these gels are shown in Fig. 1. Using the known molecular weight values for either the L or M groups of reovirus doublestranded RNA segments, molecular weight values could then be assigned to specific values for absolute electrophoretic mobility on 2.6% gels. The S group has too wide a range of mobility values to justify its use in this way with the same degree of accuracy. In order to establish the relationship between molecular weight and electrophoretie mobility, CPV RNA generally labeled with 32P was prepared. This was ana-

&1AL. -’ TABLE

1

MOLECULAR WEIGHT AND ELECTROPHORETIC MOBILITIES OF WV RNA GENOME SEGMENTS 2.6% POLYACRYLAMIDE GEM Segmerit No.

Mobility relative to slowest segmerit begment 1)

SD

Absolute mobility kp’

Relative molecular weight ’ (%of set-’ x totaP*P 105 cpm)

1

1.00

-

1.92

2 3 4 5 6 7 8 9 10 Total

1.06 1.15 1.26 1.67 1.74 2.30 2.66 3.01 3.25

0.01 0.01 0.02 0.05 0.05 0.07 0.08 0.09 0.10

2.04 2.20 2.42 3.19 3.33 4.41 5.10 5.76 6.23

20.75 17.56 14.62 12.89 8.98 8.37 5.82 4.76 3.82 3.28 100.00

ON

SD

Molecular weight x 10-e

0.80 0.61 0.30 0.56 0.91 0.60 0.27 0.26 0.19 0.11

2.55 2.15 1.80 1.58 1.10 1.03 0.71 0.58 0.47 0.40 12.37

lyzed on a number of polyacrylamide gels, choosing durations of electrophoresis optimal for separation of either the slow-moving or the fast-moving RNA species. Two are representative electropherograms shown in Fig. 2. Quantitation of 32P in each peak gives a measure of the relative molecular weight of each segment. The combined results from a total of 12 gels are shown in Table 1. The resulting plot of log relative molecular weight versus electrophoretic mobility deviates markedly from linearity in the higher molecular weight region (Fig. 3). The group average molecular weight of the M band of reovirus double-stranded RNA, 1.3 x 106, can now be used to calibrate the ordinate of Fig. 3 in terms of absolute molecular weight. The M band RNA was chosen since its group average mobility is conveniently placed between CPV bands 4 and 5. As a check, a value for the group average molecular weight of the reovirus L band RNA can now be read from the calibrated curve, and is found to be 2.4 x 106. This is in good agreement with the values of 2.3 x lo6 (Bellamy et al., 1967) and 2.5 x lo6 (Shatkin et al., 1968) for the group average molecular weight of the L group of reovirus RNA derived by physicochemical means. To confirm by an alternative approach that the plot of log molecular weight ver-

MOLECULAR

WEIGHT

Direction

FIG. CPV

1. Polyacrylamide RNA. Electrophoresis

of

OF

CPV

migration

213

RNA

+

gel electropherogram of (a) CPV RNA, (bl reovirus type 3 RNA, was for 5 hr at 10 V/cm on 2.6% gels. - A,,, nm scans.

20

40

60

a0

50

Fraction

FIG. 2. Polyacrylamide gel electropherograms or (b) 8 hr at 10 V/cm. Electrophoresis was from

No.

7

ik‘0 90 il

of 32P-labeled CPV left to right. Genome

110

and (c) reovirus

.d

130

RNA electrophoresed for either (a) 4 hr segments are numbered consecutively.

HARLEY

ET AL.

1

FIG. 3. Plot of log molecular weight versus absolute electrophoretic mobility for the 10 doublestranded RNA genome segments of CPV (0) and the RF RNAs of FMDV (01 and poliovirus (0). The relative molecular weight of any segment is expressed as its percentage of the total 32P in all 10 genome segments. The bars express the limits of 2 x SD on each axis. Absolute molecular weights were derived by calibration with reovirus RNA as described in the text.

sus electrophoretic mobility for doublestranded RNA deviates from linearity at high molecular weight values, the electrophoretic mobilities of FMDV-RF and poliovirus-RF were determined. The molecular weights of the single-stranded RNA of both viruses have been estimated to be 2.6 x lo6 (Talbot and Brown, 1972; Granboulan and Girard, 1969). The molecular weights of the RFs will be double this value and, therefore, considerably greater than any of the CPV or reovirus RNA segments. The electrophoretic characteristics of FMDV-RF are typical of a fully doublestranded linear structure (Harley et ccl., 1973) as are those of polio RF (Harley and White, unpublished observations), which is in agreement with the observations of Granboulan and Girard (1969). Figure 4 shows the labeled RFs coelectrophoresed with DNA markers for the determination of absolute electrophoretic mobility. No RNase was used in the preparation of the FMDV-RF, but subsequent treatment did not alter its mobility relative to DNA. When the molecular weight and electrophoretic mobility values for these RFs are

20

40 Distance

60 Migrated

I)0

I

(mm)

FIG. 4. Polyacrylamide gel electropherograms of (a) W-labeled HeLa cell DNA + 3H-labeled FMDVRF RNA and (bl unlabeled DNA marker and 3Hlabeled RNA extracted from poliovirus-infected HeLa cells and treated with 1 pg/ml of pancreatic RNase for 30 min at 37” immediately before electrophoresis. Electrophoresis was for 4 hr at 10 V/cm (a), and 12.5 V/cm (b), respectively. l - 0, 3H disintegrations per minute (dpm); A- - - -A, i4C dpm; . . . . . , A,,, “,,, (arbitrary units).

included in Fig. 3, it can be seen that they fall on a predictable extrapolation of the curve relating CPV RNA molecular weights to electrophoretic mobility. The molecular weight values of the CPV RNA genome segments can be read from the calibrated ordinate and are given in Table 1. DISCUSSION

The double-stranded RNA genome segments of the CPV ofH. armigera all segregate well on gel electrophoresis, rendering accurate quantitation of each genome segment by 32Pincorporation a practical proposition. For many other multicomponent double-stranded RNA viruses, this approach would not be feasible because of clustering of the segments on electrophoresis. In addition, only in the reoviridae have the genome segments been shown to

MOLECULAR

WEIGHT

be present in equimolar proportions; this has been demonstrated by Millward and Graham (1970) for reovirus, and by Lewandowski and Millward (1971) for the CPV of Bombyx mori, which is serologitally related to the CPV of H. armigera (Rubinstein et al., 1976). This has enabled us to determine the molecular weights of the double-stranded RNA genome segments of this CPV by comparative gel electrophoresis, with reovirus RNA as a molecular weight standard. However, the 32P quantitation also has enabled us to define the predicted nonlinear relationship of the plot of log molecular weight versus electrophoretic mobility for double-stranded RNA at molecular weight values greater than about 106. Molecular weight determinations based on a straight line plot and using reovirus RNA (or any other set of double-stranded RNA species) as standards, therefore, will be subject to error, particularly at high molecular weight values. However, there is no reason why molecular weight determinations using comparative gel electrophoresis should not be used, provided a calibrated plot of this type is constructed in each laboratory with a suitable set of standards. Some candidate double-stranded RNA species for use as standards are being studied (Bozarth and Harley, 1976). The shape of the curve in Fig. 3 suggests that the electrophoretic mobility of doublestranded RNA asymptoses to a limiting value as the molecular weight increases. This behavior then would be similar to that shown for double-stranded linear DNA (Harley et al., 19731, except that the limiting electrophoretic mobility of RNA would be considerably less than that of DNA. Double-stranded RNA of higher molecular weight should be used to confirm whether this is the case, since this presumptive lower limiting value for the electrophoretic mobility of double-stranded linear RNA results, at the proximal end of gels electrophoresed under set conditions, in a “forbidden region” where no doublestranded linear species should be found. Variation in gel concentration between 2.4 and 4.0% does not alter the way in which the curve deviates from linearity

OF

CPV

215

RNA

above molecular weight values of about 106. Detailed analyses of the electrophoretic mobility of this CPV RNA over this range of gel concentrations have been published previously (Rubinstein et al., 1976). ACKNOWLEDGMENTS E. H. H. is in receipt of a University of Cape Town Staff Research Grant. We would like to thank Professor J. E. Kench for the facilities of his department and for critical review of the manuscript. REFERENCES BELLAMY, A. R., SHAPIRO, L., AUGUST, J. T., and JOKLIK, W. K. (1967). Studies on reovirus RNA. 1. Characterization of reovirus genome RNA. J. Mol. Biol. 29, l-17. BISHOP, D. H. L., CLAYBROOK, J. B., and SPIEGELMAN, S. (1967). Electrophoretic separation of viral nucleic acids on polyacrylamide gels. J. Mol. Biol. 26, 373-387. BOEDTKER, H. (1971). Conformation independent molecular weight determinations of RNA by gel electrophoresis. Biochim. Biophys. Actu 240, 448453. BOZARTH, R. F., and HARLEY, E. H. (1976). The electrophoretic mobility of double-stranded RNA in polyacrylamide gels as a function of molecular weight. Biochim. Biophys. Acta 432, 329-335. GRANBOULAN, N., and GIRARD, M. (1969). Molecular weight of poliovirus RNA. J. Virol. 4, 475-479. HARLEY, E. H., REES, K. R., and COHEN, A. (1970). HeLa cell nucleic acid metabolism. The effect of mycoplasma contamination. Biochim. Biophys. Actu 213, 171-182. HARLEY, E. H., and WHITE, J. S. (1973). The identification of the circular forms of SV40 DNA in whole infected cell preparations. Virology 52, 395 407. HARLEY, E. H., WHITE, J. S., and REES, K. R. (1973). The identification of different structural classes of nucleic acids by electrophoresis in polyacrylamide gels of different concentration. Biochim. Biophys. Acta 299, 253-263. LEWANDOWSKI, L. J., and MILLWARD, S. (1971). Characterization of the genome of cytoplasmic polyhedrosis virus. J. Virol. 7, 434-437. LOENING, U. E. (1967). The fractionation of high molecular weight RNA by polyacrylamide gel electrophoresis. Biochem. J. 102, 251-257. MILLWARD, S., and GRAHAM, A. F. (1970). Structural studies on reovirus: Discontinuities in the genome. Proc. Natl. Acad. Sci. USA 65, 422-429. RUBINSTEIN, R., HARLEY, E. H., LOSMAN, M., and LUTTON, D. (1976). The nucleic acids of viruses infecting Heliothis armigera. Virology 69, 323328. RUBINSTEIN, R., STANNARD, L., and POLSON, A.

216

HARLEY

(1975). Isolation of a cytoplasmic polyhedrosis virus by physical and immunological techniques. Prep. Biochem. 5, 79-90. SCHEBRER, K., and DARNELL, J. E. (1962). Sedimentation characteristics of rapidly labelled RNA from HeLa cells. Biochem. Biophys. Res. Commun. 7, 486-490. SHATKIN, A. J., SIPE, J. D., and LOH, P. (1968). Separation of ten reovirus genome segments by

ET

AL.

polyacrylamide gel electrophoresis. J. Viral. 2, 986-991. STAYNOV, D. Z., PINDER, J. C., and GRATZER, W. B. (1972). Molecular weight determination of nucleic acids by gel electrophoresis in nonaqueous solution. Nature New Biol. 235, 108-110. TALBOT, P., and BROWN, F. (1972). A model for footand-mouth disease virus. J. Gen. Viral. 15, 163170.