Replication of tobacco ringspot virus

VIROLOGY

66.

238-219 (1973)

Replication 1. Detection

of Tobacco

of a Low Molecular

Weight

from Infected RI. A. REZAIAN Department

Ringspot

AND

Virus

Double-Stranded

RNA

Plants R. I. B. FRANCKI

of Planf Pathology, llYaite Agricullural Research InstiMe, L’niversity of Adelaide, South A~sf~alia Accepted July

26, 1973

Concentration of tobacco ringspot virus (TRSV) in cotyledons of cucumber and primary leaves of French bean increased linearly between 2 and 5 days after inoculation but decreased thereafter. The proportion of middle (M) to bottom (B) component in TRW preparations from cucumber cotyledons did not vary significantI) between 3 and 12 days after inoculation. High molecular weight double-stranded R.N.4 ids-RNA) with the espected properties of either replicative form or replicative intermediate was not detected in TRW-infected plants. However, a polydisperse population of low molecular weight da-R.NA with nucleotide sequences complementary to TRSV-RNA was isolated. Although the function of this viru. s-specific dx-RNA remains obscure, it may be involved in the synthesis of TR,SV. The TRSV-specific ds-RNA could be isolated in high yields only from leaves in which virus was increasing and its concentration in leaf tissues was positively correlated with the concentration of virus-induced RN.4. dependent RN.4 pnlymerase. INTRODUCTION

Purified preparations of tobacco ringspot’ virus (TRSV) contain three types of polghedral part.icles, top (T), middle (AI), and bottom (B) components with sedimentation coefficient,s of 53, 94. and 1128 S (State-Smith et al., 1965; Randles and Francki, 1965). The three types of particles are indistinguishable in size (about 2S-30 nm) and protein COIW position but contain 0, 2S, and 42 % RN-4, respectively (State-Smith et al., 1965). -II component particles each contain a single R.N4 molecule cJf approximately 1.2 X 10fi daltons (RNA,). Some B component particles cont,ain a single RNA molecule of approximately 2.2 X lo6 daltons (RNA,), whereas others appear to enclose two molecules of RNA, (Diener and Schneider, 1966). R.N% of JI cannot8 be distinguished from RNA1 of B component by either sucrose density-gradient centrifugation (Diener and 238 Copyright ~a 19i3 by Academic Press, Inc. All rights of reproduction in any form reserved.

Schneider, 1966) or polyacrylamidc-gel elect.rophoresis (Murant ef al., 1972). The function and mode of s\-nthesis of t,he various TRSV component par&clrs and t#heir RNAs is unknown. It has been shown that only B component particles are infectious (Diener and Schneider, 1966; Francki, 1972) and infectivity of viral RN-4 has been correlated w&h the presence of RN&i? (Dicner and Schneider, 1966). However. it appears that the infectivit#y of RNA? can be increased by addit,ion CJf RNA, to the inoculum (Harrison et a,l., 1972). It is not known if nucleotide sequences of RNA, from 11 and B differ and if they have any common sequences with RN.4, . One approach to the problem of comparing nucleotidr wqwnces of RNA species isolated from TRS1’ preparations would be to isolat,e from virus-infected plants, do&h-stranded RNA (ds-RNA) such as rcplicativc form

DS-RNA

ASSOCIATED

(RF) and replicative intermediate (RI) which have been shown to be present in plants infected by several RNA viruses (Ralph, 1969). As one of the strands of RF and RI are known t,o contain nucleotide sequences complementary to viral RNA, they could be useful for comparing the sequences of various species of viral RNA by RNA-RNA annealing techniques as used by Content and Duesberg (1971) with influenza viral RN-4, and van Kammen (1971) wit,h cowpea mosaic viral RNA. Although experiments described in this paper failed to demonstrate the presence of RNA species with the expected properties of either RF or RI, TRSV-specific ds-RNA of low molecular weight has been isolated from virus-infected cucumber cotyledons and French bean primary leaves.

WITH

TRSV

INFECTION

239

tion technique according to Hariharasubramanian et al. (1970). Virus yield was about 400 mg/lOO g of leaf tissue. Determinatim of TRSV concentratim in leaves. Virus was precipitated with PEG from a known weight of plant material processed as above, resuspended in 0.1 AI phosphate buffer, pH 7, and subjectsed to sucrose density-gradient centrifugation. The gradients were analysed with an ISCO apparatus and t,he virus concentration n-as determined from the peak areas measured with a planimeter. Preparation of ,viral RNA. Purified preparat,ions of TRSV and TMV were dissociated into protein and RNA by the single-phase phenol-sodiumdodecyl sulfate (SDS) method (Diener and Schneider, 1968) and the prot#ein was removed by t,wo extractsions with % volume of water-saturated phenol for 1 min and centrifugation at 5000g for 10 min. RNA MATERIALS AND METHODS was precipit#ated from the buffer phase by I’z&s isolates. TRSV originally isolated addition of 2 volumes of cold 95% ethanol. from GbacFi&s (Randles and Francki, 1965) The precipitate n-as left at -15°C for at and a tobacco mosaic virus (TMV) strain of least 3 hr, recovered by centrifugation and unknown origin (Crowley et al.? 1969) were washed wit,h either cold ethanol or acetone used. Both surfaces of cucumber cot,yledons and then with ether; it was dried under or primary leaves of French beans were inoc- vacuum and resuspended in the required ulated mechanically, and the plants were buffer. The two-phase phenol extraction step maintained in a constant temperature room was omitt#ed if the viral RNA was required for analysis by sucrose density-gradient cenat 35 f 2” under fluorescent lights providtrifugation. ing continuous illumination of 350450 ft-c. Preparation of 14C-labeleclTR2Wan.d TMVVirus pu,ri$cation. TRSV was purified by RX’A. Plants in -l-inch pots were transferred a combination of polyet,hylene glvcol (PEG) to a large desiccator immediately aft,er inocuprecipitation a.nd ultracentrifugatlon (Atchilation wit,h virus. 14C0, was released in the son, 1971). French bean leaves or cucumber cotyledons infect,ed for 5 days were homoge- desiccator by introducing H$O, int#o a small beaker containing about 2.5 mCi of 14Cnized with 1 ml of 0.1 M phosphat,e buffer, pH 7, and 1 ml of chloroform per gram of labelled sodium bicarbonate. Five days later leaf t,issue and cent#rifuged at 10,OOOgfor 30 the plants were harvested, virus w-as purified and the RNA isolat,ed as described above. min. PEG and NaC1 were added to the buffer phase to a final concentration of 6 % and 0.3 The specific activity of viral RNA was usuAl, respectivel.y, and the extract was left, at ally about 1400 cpm/pg. Extracfim of total leaf n,urleic acids. Up to 0” for 30 min. The precipitat,e was pelleted by centrifugation at, 10,OOOgfor 10 min and 100 g of plant material was used to ext,ract nucleic acids from healthy or virus-infected resuspended in 0.1 31 phosphate buffer, pH 7, with 0.01 M EDTA and left overnight’ at plant,s. In earlier experiments freshly harvested tissue was homogenized with TNE 4”. Virus was sedimented by ultracentrifugation at 160,OOOgfor 50 min and resuspended buffer (0.1 A1 Tris.HCl, 0.1 d NaCl, 0.01 Af EDTA, pH 7) (Jackson et al., 1971) and in 0.1 AI phosphat,e buffer, pH 7, and clariTNE-saturated phenol containing 0.1% 8fied by centrifugation at 10,OOOgfor 10 min. (2 ml of each reagent per Yield of virus was usually about 20 mg/lOO g hydroxyquinoline gram of leaf tissue). In later experiments the of tissue. material was frozen in liquid nitrogen prior TMV was purified by the PEG precipita-

to extraction with TNE buffer containing 1 % SDS and phenol containing 0.1 ‘7, 8-h\-droxvquinoline. The slurry n-as shaken fm a.5 n& centrifuged at .5OOOyfor 10 min and the buffer phase was reext,racted twice more with half a volume of the phenol reagent,. The RNA was precipitated by addition of 2 volumes of chilled et,hanol and kept at -15% for at least, 3 hr. The precipitated RNA wa.s sedimented by centrifugation at 5OOOqfor 10 min, washed wit,h acetone followed by et.her, dried under vacuum and resuspended in STE buffer (0.1 ~11NaCl, 0.0.5 31 Tris.HCl, 0.001 II1 EDTA, pH 6.85) (Jackson et a.l., 1971). Sa.lt fractionation. of nucleic acid prepamtions. NaCl was added t,o nucleic acid preparat,ions in STE buffer to a final concentration of 1.5 M and kept at - 15°C for at, least, 3 hr (Bishop and Koch, 1967). After slow thawing, t.he preparat,ion was centrifuged at 5OOO~q for 20 min at -1” and the salt-insoluble RNA species were recovered in the pellet. Two volumes of et,hanol were added to the supernatant, to precipitate the salt,-soluble nucleic acids (DNA, transfer RNA (tRNA), 5 S ribosomal RNA (rRNA), and in the case of preparations from virus-infected tissues, ds-RNA). TCJfree the salt,-soluble fraction from contaminating polysaccharides, t,he nucleic acids were further extracted wit,h 2-methoxyethanol and precipitaCed with cetyltriethylammonium bromide (CTAB) (Ralph and Bergquist , 1967). Where it. was required t,o remove DNA from a nucleic acid preparation, 10 pg/ml of DNase free from ribonuclease (RNase) (Sigma DN-EP, eleclrophoretically purified) was added to buffer containing 0.01 dI RlgCl? and the mixture was incubated at 30” for 30 min. To remove single-st,randed RNA (ss-R.KA), the preparation was incubated at 37” for 30 min with 2 pgjml of pancreatic RNase A (Sigma Chemical Co.) in buffer containing not, less t,han 0.2 M NaCl and 0.01 dl EDTA. Following enzyme treatmcnt, the preparations were incubated with 10 pg./:ml of Pronase (Sigma Type VI fungal protease) for 30 min at 37” and extract,ed with one-half volume of phenol reagent, and the remaining nucleic acids were precipitated with ethanol as described above.

I~nbellir~~~of cucunder cotyledorl R&\-d with zwidiue. Excised cucumber cotyledons were placed in small Petri dishes, incisions wrre made about 2 mm apart from the midrib parallcl to t.he veins (Zait,lin et al., 1963) and 0.2 ml of water coutaiuing 2.8 &i llC-uridine n-as applied to each healt,hy and 2-1 &i 3Huridine to each virus-infected cotyledon. After -l hr t,he co@ledons were rinsed several t,imes with dist,illrd water and the nucleic acids were isolat,ed as described above. Separatiort of uucleic acids by pobyacrylarnicle-gel electrophoresis. Composite 2.4 0; or 1.8 4”opolyacrylamide and O..i %#agarose gels were prepared in either 7.5 or 11 cm long plexiglass tubes (6 mm internal diamet,er), respect,ively. The SDS-buffer system (Loening, 1969) n-as used and 1.82, gels were sliced immediately but, 2.4 Co gels were frozen at - 15” before cutting into a series of slices 1.4 mm thick with a mult#ibladed apparatus. Individual slices were placed in glass scintillation vials and 0.5 ml of a 9: 1 mixture of NCS (Amcrsham:‘Searle Corporation) and water was added to each vial. Aft,er incubation overnight, at -Is”, scintillation fluid consisting of 0.5 ‘% PPO and 0.01 %#POPOP in tolucne was added (10 ml per vial) and the radioacbivity was determined in a Packard scintillation spectrometer. When it, was required to recover RNA from t,he gels, each gel slice was pulverized with a fine glass rod in 0.5 ml of SSC buffer (0.15 J1 NaCl and 0.015 31 sodium citrat,e) containing .5 pg Pronasc in 10 X 1 cm test tube and incubat,ed at room temperature for 5 hr. The gel fragments were removed bJ cenbrifugation and t,he clutcd RN,4 was n-ithdrawn with a micros\-ringe. R.VA-RSA h ybridizatioli. techniques. The ds-RNA preparation to be hybridized, suspended in 150-2.50 ~1 of SSC containing 2 pg of Pronase, was incubated for 30 min at 37” and then denatured by heating for 10 min at 100”. The ss-RNA to be hybridized was added and the samples transferred to a n-at,er bat,h at 85”, which was then allowed to cool to 37” over a period of about S hr. RNase was then added t#o a concentration of 50 pg//ml and the samples were incubated at, 37” for 30 min. The preparations were spot,ted on disks of Whatman No. 3 filter paper and extracted 13 times in 5% TC*4,

DS-R.NA

ASSOCIATED

twice in 80% ethanol and once in ether before being placed in scintillat,ion vials for radioact#ivity determinations (Byfield and Scherbaum, 1966). RNA-clependen

t RSA

polymerase

TRSV 05

241

INFECTION

ra

assay.

Part,ially purified enzyme from a known weight of freshly harvested cucumber cot,+ dons was prepared and enzyme activity assayed as described by May and Symons (1971). Density-gradient ce?ltrifugation. Purified virus preparat#ions were analysed on 5 to 25 % linear sucrose density gradients buffered with 0.02 M phosphate, pH 7.0, in an SW 50 Spinco rotor. Similar gradient.s were used for the analysis of RNA preparations, but the sucrose solutions used for making gradients were autoclaved at 15 psi for 15 min. Spectrophotmetry. Virus and RNA concentrations were determined using extinction coefficients (Ey.$) at 260 nm of 25 for RNA, 7.0 for purified TRSV (blurant et al., 1972) and 3.0 for TPtlV. RESULTS

TRST’ Mu.ltiplicafio~~.

WITH

in Cu,cwnber Cotyledons

TRSV was detected by sucrose densit,ygradient centrifugation (see Materials and 3Iethods) in cot#plcdons of cucumber plants 2 days after inoculation and t,hereaft,er virus increase was linear until day 5 after inoculation. In most experiments the amount of virus in the cotyledons declined after reaching a maximum. All three TRSV components, T, RI and B, were invariably detected in virus preparaCons from cotyledons harvested at various t,imes after infection (Fig. 1). However, in some preparations, two additional components, Ba and B, sedimenting ahead of B, were also detected (Fig. lb). The presence of BY in TRSV preparat,ions has also been reported by :La,dipo and de Zoeten (1972), who found that it was infect,ious. In our experiments, a preparation of B? particles wit.h an optical density of 0.25 at 260 nm produced a mean of 89 lesions per half-leaf, whereas a preparation of B particles of the same optical clensity produced 156 lesions per half-leaf when inoculated to 12 leaves of cow-pea plants. Both preparations had ulkaviolet absorption spectra characteristic of

DEPTH

(cm)

FIG. 1. Sucrosedensity-gradient centrifugation of TRSV purified from cucunlber cotyledons. One hundred milligrams of virus was layered on a 5 to 25qb sucrose gradient, centrifuged at. 50,000 rpm for 30 min and analysed by an ISCO fractionator and recorded at 251 nm. Virus preparations containing only a trace of component B2 (aj and one containing relatively high concentrations of Bz and Ba components (bj; a and h are from two independent experiments.

nucleoprotein, but the 260,/280 nm ratio of the B:! preparation was 1.55 and that of B preparation 1.62. Polyhedral particles charact#eristic of TRSV (Chambers et al., 1965) were observed in negatively stained preparat,ions of both B, and B examined in the electron microscope. RNA isolated from TRSV preparations containing B, and subject,ed to sucrose density-gradient centrifugation contained two species of R.NA with sedimentation rates indistinguishable from t.hose of RNA isolated from virus preparations devoid of B, and B3 . Taking the sqg, of T, 31, and B to be 53, 94, and 1% S, rcsprctively (State-Smith et al., 1965), it n-as calculated that the s20nof Bi and B, n-as 180 and 210 S! respectively (Fig. lb). These values are very close to the expected saOnof the dimer and trimer of B (Rlarltham, 1962). These observations support the view that B:! and Ba are aggregates of B component particles. Schneider and Diener (1966) reported that in TRSV-infect,ed bean plants the proportion of RI to B components was higher at early

T.4BLE PROPORTIONS CUCUMBER

Day after inoculation

OF V.\RIOUS COTI-LEDONS

1

TRSV

COMPONENTS IN PURIFIED VIRC~ HARVESTED AT V.\RIOUS TIME INTERWLP

_-~

T;of

F’REP.LR.~TION~ FROM .WTER INFECTION

each cowonentn

Experiment

1

Experiment

RI

B

1 2

0* 13.2

0 79.9

0 6.9

3 4 5

11.0 10.6 ll.B

83.5 84.5 s3.4

3 .5 4.9 5 0

6 i

11.0

9 12

9.9 -

B?

iv

B c

B2 -

12.6

75.1

i3.7

12.4

14.5 -

77.8 77.2

11.1 10.3

-

83.5

5.5

11.8 -

84.8

5.3 -

11.1 12.6

a Calculated on the basis of areas under the peaks recorded 60 = not detectable. c - = not determined.

2

at 254 nm w&h the IX0

apparatus.

stages of infection. However, in our experiments with TRSV-infected cucumber cotyledons, the ratio of M to B components appears to be remarkably constant during all stages of infection (Table 1). In some experiments the amount of Bf component was almost as high as that of 11 (Table 1). Properties of RNA Isolated from TRST’ Preparations

Sucrose density-gradient profiles of RNA isolated from TRSV preparations show t,he presence of t#wo species of RNA (Fig. 2). The ratio of the amount of the slower sedimenting species (R.NA1) to that of the faster sedimenting one (RNA2) was remarkably constant in preparations from virus isolated at various times after inoculation (Fig. 2). The amount of RN-41 was always considerably greater than that of RNA2 although in all our virus preparat,ions the amount of B component was always greater than that of RI (cf. Figs. 1 and 2). This, together wit#h the fact that both RNA, and RNA? were isolated from preparations of purified B component, confirms the conclusion reached by Diener and Schneider (1966) that R/I component particles contain one RNA, molecule each, whereas B component particles each contain either one molecule of RNA? or t,wo molecules of RNA1 .

DEPTH

CC”,,

FIG. 2. Sucrose density-gradient centrifugation of TRW-RNB. RNA was prepared by the singlephase phenol-SDS procedure resuspended in 0.02 di phosphate buffer, pH 7. RNA from 100 pg virus was layered on 5 to 25?0 sucrose gradient in 0.02 M phosphate buffer, pH 7, and centrifuged at 50,000 rpm for 3 hr at lo. 811 samples of virus were harvested from the same batch of virus-infected plants.

In our experiments, both RNA, and RNA? sediiented as relatively homogeneous peaks in sucrose density gradients (Fig. 2). No obvious signs of degradation were detected in RNA preparations isolat.ed from virus obtained from plants at late stages of infection as was observed in virus from bean plants by Schneider and Diener (1968). Detection. of Virus-SpeciJc cls-R:VA irb TRSFInfected Cucumber Cotyledons At various times after inoculation, 14C02 labelled TRSV-RNA was used to det,ect

DS-RNA

BSSOCIATED

WITH

TRSV

INFECTION

243

falls very rapidly, as previously observed by Peden et al. (1972), and is soon followed by a similar decline in the level of the ds-RNA. In all subsequent experiments, TRSV-specific ds-RNA preparat,ions were obtained from cucumber cotyledons 3 days after inoculation to ensure a high yield of the ds-RNA. The ds-RNA suspended in 1 X SSC buffer was unaffect#ed by RNase or DNase but became sensitive t,o RNase after heat denaturation or on suspension in 0.01 X SSC buffer (Table 2). These observations confirm that the virus-specific ds-RNA was indeed a double-stranded polyribonucleot8ide. Sedimentation am? Electrophoretic Properties of TRSI’-Specific ds-RSA

FIG. 3. Correlation of TRSV multiplication with RN&dependent RNA polymerase activity and virus specific ds-RNA synthesis in cucumber cotyledons. Virus and polymerase activity were estimated as described in Materials and Methods; and TRSV specific ds-RN-4 by annealing ‘IClabeled viral-RNA wit.h the salt-soluble RNaseand DNase-treated nucleic acid preparations extracbed from infected cotyledon t.issue. The amount of annealing was det.ermined using the equation ds-RN.4 = (2a X ss-RNA)/(lOO - a) (see Results for details). All results are expressed as percentage of the maximum recorded during the course of the experiment (on day 5 TRSV concentration in cot,yledon tissue was 681 pg/g fresh weight; on day 3 ds-RNA concentration was 54 PP/d~

ds-RNA with base sequences complementary to viral RN.1 by annealing the TRSV-RNA with salt soluble and nuclease-treabed nucleic acid preparations from virus-infected cucumber cotyledons. We determined at each time of sampling, the amount of virus, the virusspecific ds-RNA and the RNA-dependent RNA polymerase activity (Fig. 3). Results of this experiment indicate that there is a rapid increase in both the amount of virusspecific ds-RNA and polymerase activity just prior to, and during the rapid synthesis of virus. However, as soon as virus synthesis ceases? the level of the polymerase activity

Cells infect,ed by many RNA viruses whose mode of multiplicat,ion has been adequately investigated, have been shown to contain species of ds-RNA defined as replicative form (RF) and replicative intermediate (RI) (Bishop and Levintow, 1971). The RF has been shown to be relatively &able to RNase in buffers of moderate ionic strengt#h and the RI to assume t,he characteristics of RF on RNase treatment (Ralph, 1969). In all cases investigated, the RF appears to be a duplex of the viral RNA and hence has a molecular weight twice that of the RNA (Bishop and Levintow, 1971). In planning our initial investigaCions of t,he TRSV-specific ds-RNA detected in virus-infected cucumber cotgledons, we expected t,hat it would have properties characteristic of an RF. We also considered that t,here could be two species of RF in TRSV-infected tissues, one corresponding to t,he viral RNA1 and one to RNA2 as has been demonstrat,ed in plants infected with cowpea mosaic virus, a multicomponent virus with T, RI, and B particles somewhat similar to TRSV (van Griensvcn and van Kammen, 1969). However, our experiments failed to detect ds-RNA species of the expected molecular weight#s, approximately 4.4 X lo6 (RF corresponding t.o RNA2) and 2.4 X lo6 daltons (RF corresponding to RNAI) when TRSV-infected cucumber cotyledons were incubated with labeled uridine. The most sensitive method of det,ecting incorporat,ion of label into TRSV specific RNA species during short exposures was by

Treatment

14C-lat~elled TRSV-RN1

of ds-RNAb hefore annealing

Experiment

Untreated (control) RNase in 1 X MC RNase in 0.01 X SSC Heat denatured Heat denatured + RNase 1 x SSC DNase

in

annealed to ds-RN;\‘1

1

Experiment

2

Cpm

“; of control

Cpm

(,‘;, of control

372 52 9 369 T9

100 91.2 2.4 99.2 21.2

870 850 17 848 215

100 97.7 2 0 9i ..i ‘4.i

3.X

95.7

St%

99.4

Q Annealing procedure as described in Materials and hlethods. b Following RNase (20 pg’mlj or DNase (50 pg;mlj treatment of unlabeled salt-soluble leaf nucleic acid preparations, RNase was added to all samples which had not been treated and a phenol extraction was carried out immediately. The buffer phase was recovered, nucleic acid precipitated with ethanol, resuspended in 1 X SSC and annealed to K-labeled TR.SV-RNA.

labelling healthy cucumber cotyledons n-it,h Y-uridine, TRSV-infected ones with 3H-uridine and then coelectrophoresing the isolated RN& in polracrylamide gels. Results of such an experiment using 1.S % polyacrylamide gel (Fig. 4) show that virt.ually no label was incorporat,ed into RNA species with molecular weight,s in excess of 2.8 X lo6 daltons (fractions l-30 in Fig. 4). Incorporation of 14C-uridine into 25 and 1S S rRNA of healt,hy cotyledons and 3H-uridinc into t,hose of diseased ones was readily detected (Fig. 1). In addit,ion, by plotting the ratio of 3H-uridine:14C-uridine detected along t,he polyacrylamide gels, it, n-as possible to detect the mcorporation of label into t#hree species of TRW-specific RNA (peaks a, b, and e in t.he inset to Fig. 4). In a subsequent experiment , by coelectrophoresing RNA from purified TRW with RNA from healthy cucumber cotyledons, it n-as found that, peaks a and b were in the expected positions of the viral RNAs, RNA, , and RN& , respectively. The troughs c and d in Fig. 4 are in positions where 23 and 16 S rRNA from chloroplasts should migrate. The troughs suggest, that TRSV infection reduces t,he rate of synt#hesis of chloroplast rRNA as has been shown t,o occur in plants infected with several ot’her viruses (Hirai and Wildman, 1969; Randles and Colemaq 1970; Mohamed and Randles, 1972). The TRSV-specific RNA species migrating as peak e in Fig. 4 was even more readily

FIG. 4. Polyacrylamide gel-electrophoresis of total nucleic acids from healthy and TRSVinfected cucumber cotyledons, labeled with I%and “H-uridine, respectively. Escised cotyledons infected 60 hr previously and healt.hy controls (5 of each) were labeled for 4 hr as described in hlaterials and Methods. Samples of RNA from healthy and diseased plants were mixed and coelectrophoresed in 1.8:; polya.crylamide, 0.57; agarose gel at room temperature and 8 V/cm for 4 hr. Radioactivity of 3H and “C were determined in two counting channels and corrected for W cross-over (6.2?,). The inset, shows ratios of 3H to “C. (Ratios were not, calculated when 3H samples counted at less than 100 cpm.)

detected when the RNA preparations were coelectrophoresed in a 2.4 % polyacrylamide gel (Fig. 5). This RNA species has also been det,ect,ed on polyacrylamide gels by direct,

DS-RNA

ASSOCL4TED

WITH

TKSV

INFECTION

FIG. 5. Coelectrophoresis of uridine labeled total nucleic acids cotyledons in 2.4y0 polyacrylamide, 0.57~ agarose gel. Electrophoresis details as described in Fig. 4.

FIG. 6. Electrophoretic mobility of TRSVspecific ds-RNA as determined by annealing to IaC-labeled TRSV-RNA (1000 cpm/assay). Saltsoluble nucleic acid fract,ion from unlabeled virus-infected cucumber cotyledons (2 g fresh weight) was prepared, and electrophoresis was carried out as described in Fig. 5. The gel was removed and sliced, and each slice was eluted in 1 x 8X and the eluate used for annealing to ‘4Clabeled TRPV-RNA. Nucleic acid fraction from uninfected plants failed to anneal to X-labeled TRSV-RNA (see Table 3 for results of similar experiments).

with toluidine blue (Rezaian and Francki, in preparation). Its salt solubility and ability to anneal with 14C-labeled TRSV-RnTA is illustrated in Fig. 6. When nucleic acid preparations from TRW infected cucumber cot#yledons were subjected to sucrose density-gradient. centrifugation under conditions in which 18 S staining

243

from healthy and TRSV infected was carried out for 3 hr and other

FIG. 7. Sucrose density-gradient cenkifugation of TRSV specific ds-RNA as determined b> annealing to ‘%-labeled TRSV-RNA. The saltsoluble nucleic acid fraction from virus-infected cucumber cotyledons (2 g fresh weight) was prepared as described in Fig. i. RNA in 1 X SSC was layered on a 5 to 25yb sucrose densit,y-gradient in the same buffer and centrifuged as described in Fig. 2. Fractions (250 ~1) were collected and each fraction was annealed to 1500 cpm of 14C-labeled TRSV-RNA (similar results were obtained when the salt-soluble fraction of a nucleic acid preparat.ion from TRSV-infected cot,yledons were subjected to sucrose density gradient centrifugation).

sedimented about half-may down the tube, only fractions near the meniscus contained material which annealed with 14CIstwlcd _.~ RNA (Fig. __.____.. \ u 71. It. was also shown rRNA

246

R.EZAIAN

AND

that t,he electrophoretic mobility of the virusspecific RNA was not significantlv altered by RNase t,reatment in 1 X SSC. Similar RNA preparations from healthy cucumber cotyledons failed to anneal to 14C-labeled TRSVRNA either after RNase treat,ment or after the removal of ss-R.N-4 species by salt precipitation. It is concluded from these experiment.s that the virus-specific RNA det,ected as peak e in Figs. 4 and 5 is double-stranded and consists of molecules very much smaller than t,hose expected for RF. It seems unlikely that the ds-RN-4 detected in TRSV-infected cucumber cotyledons was a degradat,ion product of larger molecules as most of t.he nucleic acid preparations were not subjected to enzymat,ic treatments (Figs. -1-8). Furthermore, when t,he ext.raction procedure was modified, dsRNA with virt#ually ident,ical properties was detected. Neither the addition of 1% diethyl pyrocarbonate (Solymosy et al., 1968) nor 0.1% bentonite (Fraenkel-Conrat, et al., 1961) to the extraction buffer changed the properties of the virus-specific ds-RNA isolated. In another experiment, the t.ime of the initial phenol-SDS extraction was reduced to only 10 min and another extraction buffer, containing 0.1 X Tris. HCl, 0.05 a1 NaCl, 0.005 64 EDTA, 0.05 X sodium tet#raborate, 1% ascorbic acid, 1%’ SDS and 0.1 Cr,,bentonite, pH 7.6 (Bockstahler, 1967), was used; but again, the properties of the isolated virus-specific ds-RNA were similar to those already reported. Bearls I,nfectecl with TRSV ad TdlT’ TRSV and a bean strain of TRIV have been shown to multiply well in primary leaves of French beans (Crowley et al., 1969). Since the RF of TRIV has been characterized in some detail and shown to consist of a molctule of I\IW approximately 4 X lo6 daltons (Jackson et al., 1971), we wished to determine whether undegraded TMV-RF could be isolated from infected bean leaves by t,he same t,echniques used for the isolation of t.he virus-specific ds-RNA from TRSVinfected t,issue. The multiplication curve of TRSV in primary leaves of beans was very similar to that, in cucumber cotyledons and virus-spe-

FRANCKI

FIG. 8. Sucrose density-gradient cenkfugation of TRW and TMV specific ds-RNAs isolated from primary leaves of French beans as deterhomologous mined by annealing to W-labeled RNA. The salt-soluble nucleic acid fractions from 15 g fresh weight of leaves infected with each virus were further purified by 2-methoxyethanol extraction and CTAB precipitation and subjected to sucrose density-gradient centrifugation. Fra.ctions from gradient.s were annealed to the respective ‘%-labeled viral RNAs as described in Fig. 7.

cific ds-RNA n-as readily detected in nucleic acid preparations from infect,ed bean leaves (Table 3). After centrifugat,ion in sucrose density-gradients, the distribution of TRSVspecific ds-RN,4 isolated from bean leaves was similar to that isolated from cucumber cotyledons (Fig. 8). However, TJIVspecific ds-RNA from the same host sedimentcd as a peak between 12 and 13 S, which is approximately that observed for TMV-RF and that expected for a ds-RNA molecule of approximately 4 X lo6 daltons (Burdon et al., 1964). The broadness of t#hepeak is probably due to the nonlinear increase in the amount of 14Clabeled TRIV-RNA hybridized with t,he increase in t#he amount of RF (see next section for explanation.) Estimatiorls of T’ims-Speci$c ds-R.l’A centmtiom in Infected Leaves

Con-

From theoretical considerations it can be deduced that when a given weight of labeled viral ss-RNA is exposed to annealing condit#ions with double the weight of unlabeled

IX-RNA

ASSOCIATED

WITH TABLE

DETECTION WITH

&-RN;\

347

INFECTION

3

IND ESTIK~TION OF ds-RNA IN TRSV .IND TMV INFECTED PL.LNTS NUCLEOTIDE SEQUENCES COMPLEMENTARY TO VIR.LL RNA

isolated froma

0.5 g TRSV-infected cotyledons 10 g TRSV-infected leaves 10 g TMVinfected leaves

TKSV

Homologous ss-viral RNA added to each annealing assay

cucumber primary

bean

primary

bean

Viral RN.L\b annealed (cpm) to: -.

Cpm

i%

H

D

4300

3.8

56

3800

1600

1.3

118

884

0.33

2250

1.6

54

1403

0.64

u ds-RNA in salt-soluble fraction of leaf nucleic acid preparations was further methoxyethanol and CTA procedure (see Materials and Methods). b Preparations from healthy plants (H) and TRSV-infected (U) were used. c Calculations were made using the equation ds-RNA = (2~ X ss-RNA)/100 details).

virus-specific ds-R.NB, -50c”o of the labeled ss-RNA will anneal into double-stranded strucbures. Thus, if a known n-eight of labeled ss-RNA is added to an unknown weight, of ds-R.NA and the mixture melt,ed and reannealed to equilibrium, the weight of ds-RNA in the mixture can be calculated from the eqwtion ds-RNA

= “a X ss-RNA 100 -

Calculatedc concentration of dsRN-4 per gram infected (D) tissue (rg)

a

where a is the percentage of ss-RNA annealing. Since it is unlikely that RNA melting and reannealing is 100% efEcient, calculations of ds-RNA concentrations by this met,hod proba.bly result in underestimations; however, the:; can serve as an approximation. Using the above equation, n-e have calculated the approximate concentrations of TRSV-specific ds-RNA reached in virusinfected leaves. The concent,ration of virusspecific ds-RNA in cucumber cotyledons three days aft.er inoculalion, when it reaches its maximum (viz. Fig. 3), ranged from 8 pg per gram to 54 Mg per gram fresh n-eight of leaf mat,erial. Similar calculations derived from data on beans infected with TRSV and TRIV are summarized in Table 3. DISCUSSION

Under our experimental condit.ions, the multiplication of TRSV in both inoculated

34

purified

by the 2-

a (see Results

for

cucumber cotyledons (Fig. 3) and primary bean leaves is very rapid. Similar TRSV multiplication curves have been reportsed by Ladipo and de Zoeten (1972) when virus was assayed by infect#ivity. Reasons for t,he decrease of virus concentration in inoculated leaves remain unknown, but one possibility is that some virus is translocated out of the leaves to other parts of the plant,, such as root t,ips as suggest,ed by Atchison and Francki (197‘2). From their studies of t.wo TRSV st,rains in French bean, Schneider and Diener (1966) concluded that although M and B component? were synthesized concurrently, RI component accumulated more rapidly than B during early st,ages of infecCon and that later this t,rend was reversed. In our studies the ratio of nI to B component8 remained remarkably constant both during the period of virus synthesis and after t,he net, increase in virus concentration had ceased (Table 1). We are unable to explain why in some of our experiment,s dimers and trimers of virus particles were detected (Fig. 1). Ladipo and de Zoeten (1972) concluded t,hat. presence of B? component was seasonal, but this can hardly explain our results, as all experiments were carried out under the same controlled condit#ions. In contrast t,o results reported by Schneider and Diener (1968), RNA from TRSV purified from leaves at, late stages of infection showed no significant signs of deg-

218

11F:Z.4Ir\N

.4Nl)

sadation (Fig. 2). Diff errnces in properties of TRW a.sshown in our experiments and those of others (Diener and Schneider, 1966; Schneider and Diener, 1968) may IX a reflrcCon of strain differences. The low molecular weight ds-RNA which we have shown to be present, in TRSVinfected leaves, alt#hough very heterogeneous and probably a population of several distinct RNA species, is unlikely to be a degradation product of either RF or RI for t,he following reasons: (1) several metshods of RNA extract#ion resultsed in t,he isolation of ds-RNA with essent,ially identical propert#ies, and (2j the same t#echniques used for extraction of RNA from TRIV-infected plank resulted in the isolation of RF (Fig. 5). It appears that the presence of TRSVspecific ds-RNA in infected cucumber cotyledons is associated with virus synthesis because: (1) The appearance and increase of ds-R.NA concentration in leaves follows closely the rise and fall in the concentration of the virus-induced RNA-dependent RNA polymerase (Fig. 3) which is very similar to that reported by Peden et al. (1972), and (2) increase in ds-RNA concentration precedes the rapid accumulation of TRSV in infected cot,yledons and it,s concentrat#ion drops very rapidly as soon as virus concentrat,ion reaches a maximum (Fig. 3). The almost complete disappearance of ds-RNA from infected tissue soon after virus accumulation ceases indicates t#hat’ the ds-RN-4 is unlikely to be a “dead-end” product. Although at present the function of the TRSVspecific ds-RNA in infect,ed tissues remains unknown, it, seems that it may be involved in virus synthesis. The high concentrations of t.he ds-RNA reached in infected cucumber cotyledons has enabled us to isolak sufficient, quantities of the material for characterization without resorting to isotopic t#echniques and the results of these studies will be published in due course. ACKNOWLEIXhlENTS We thank Dr. R. H. Symons for generous gift.s of ol-3?P-guanosine triphosphate and yeast RN.4, his advice on polymerase assays and access to his unpublished results; Dr. J. W. Randles for helpful discussions and advice on polyacrylamide-gel electrophoretic met,hods; and Mrs. L. Wichman

Ff:AN(‘KI for the line drawings. One of us (hI.h.It.) is a postgraduate scholar supported by the Iranian hlinistry of Science and Higher Edlration. The project was also supported by the .~llstralian Research (-irants Commit,tee. HEFERFNCFS 1

II

ATCHISON, B. A. (1971). Studies on the invasion of meristematic tissue of French beans (Pha.seolrts oulgaris L. cv“Hawkesbury Wonder”) by tobacco ringspot virus. Ph.l>. Thesis, University of -4delaide. ATCHISON, B. A., and FII.~NCZ;I, 1~. I. B. (19i2j. The source of tobacco ringspot virus in root,-tip tissue of bean plants. Physiol. Plar~t Pathol. 2, 1Olhll. BISHOP, J. M., and KOCH, G. (1967). Purification and characterization of poliovirus-induced infectious double-stranded ribonucleic acid. J. Biol. Chem. 242, 1736-li-l3. BISHOP, J. hl., and LI:~INT~XV, L. (1971). Replicat.ive forms of viral RNA. Structure and function. Progr. Med. Viral. 13, l-82. BOCECSTIHLER, L. E. (1967). Biophysical studies on double-stranded RNAfrom turnip yellow mosaic virus-infected plants. Mol. Gear. Get/et. 100, 337348. BURDON, R. H., BILLETER, h1. A., WXISS~~.LNN, C., WIRNER, R. C., OCHO.\, S., and KNIGHT, C. A. (1964). Replication of viral RNA. V. Presence of a virus-specific double-stranded RNA in leaves infected with t,obacco mosaic virus. Proc.

.Vai.

;Ica.d.

Sci.

IrS.

22, X8-775.

BBFIELD, J. E., and SCHERRIUM~ 0. H. (1966). 9 rapid radioassay technique for cellular suspensions. Arial. Biocheltl. 17, 434-143. CH.IMBERS, T. C., FR.\NCBI, R. I. B., and I~.\NDI.EB, J. W. (1965). The fine structure of (.;ladiolus virus. Virology 25, 1521. CONTENT, J., and DUESBERG, P. H. (lsilj. Base sequence differences among the ribonucleic acids acids of influenza virus. J. MO/. Biol. 62, 27% 285. CRO~LEP, N. C., D.YVISON, E. M., FR.\NCE;I, R. I. B., and Owusu, G. K. (1969). Infect.ion of bean root-meristems by tobacco ringspot virus. T’irology

39, 322-330.

DIENF:R, T. O., and SCHNEIDER, I. R. (1966). The two component.s of tobacco ringspot virus nucleic acid: origin and properties. Virology 29, loo-105. DIENER, T. O., and SCHNEIDER, I. R. (1968). Virus degradation and nucleic acid release in single-phase phenol systems. Arch. Riochem. Biophys. 121, 401-412. FRIFZNIXL-CONR.IT, H., SINGER, B., and TSUGIT.~, 4. (1961). Purification of viral RNA by means of bentonite. Virology 14, 54-58.

IS-RNA

ASSOCIATED

FR.\NCICI, R. I. B., (1972). Purification of viruses. In “Principles and Techniques in Plant. Virology” (C. I. Kado and H. 0. Sgrawal, eds.), pp. 295-335. Van Nostrand-Reinholds, Princeton, New Jersey. HIRIH.~R.~ISUIIRB~~.~NIIN, T'., ZAITLIN, M., and hGEL, A. (1970). A temperature-sensitive mutant of T&IV with unstable coat protein. vifo/og,y 40, 579-589. HARRISON, B. D., hIun.INT, ii. F., and Mayo, hl. A. (l972j. Evidence for two funct,ional RNA species in raspberry ringspot virus. J. Gen. Viral. 16, 33S348. HIR.U, -4., and wrmnrm, 8. cr. (1969j. Effect of TMV multiplication on RNA and protein synthesis in tobacco chloroplasts. I’irology 38, 7382. SIEGEL, A. J.~CKSON, A.O.,PIIITCHELL, D.M.,and (1971j. Replication of tobacco mosaic virus. I. Isolation and charact,erization of doublestranded forms of rihonucleic acid. Virology 45, 182-191. IADIPO, J. L., and ~6 ZOETEN, G. A. (1972). Influence of host and seasonal variation on the components of tobacco ringspot virus. Phytopathology 62, 195-201. LOENING, U. E:. (1969). The determination of the molecular weight of ribonucleic acid by polyacrylamide-gel electrophoresis. The effects of changes in cl,nformation. Biochpnl. J. 113, 131138. hhRsH.m, R. (1962). The analytical ult,racentrifuge as a tool for the investigation of plant viruses. -~&~urr. TYirrts Res. 9, 2-l-270. M.\r, J. T., and SYnroNs, R. H. (19il). Specificity of the cucumber mosaic virus-induced RNA polymerase for RNA and polynucleotide templates. T’irology 44, 517-526. MOHAMED, N. A., and R.YNDLKS, J. W. (1972). Effect, of tomato spotted wilt virus on ribosomes, ribonucleic acids and Fraction I protein in Nicofia,lcc tabaclc,n leaves. Physiol. Plarct Pathol. 2, 235-245. bIUK.\NT, A. F., MAYO, hl. A., H.IRRISON, B. D., and GOOLD, R. A. (1972). Properties of virus and R.N-4 components of raspberry ringspot. virus. J. GPII. I'irol. 16, 327-338. PEDI-:N, K. W. C., hI.iy, J. T., and SYMONS, R. H. (1972j. A comparison of two plant. virus-induced RNA polymerases. T’irology -li, 498, 501.

WITH

TRSV

INFECTION

249

RALPH, R. K. (1969). Double-stranded viral RNA. Advm. Virus Res. 15, 61-158. RALPH, R. K., and BERGQUIST, P. L. (1967). Separation of viruses into components. Z~L “hlethods in Virology” (Ii. Maramorsch and H. Koprowski, eds.), Vol. 2, pp. 463-545. RANDLES, J. W., and COLEMIN, D. F. (19iO). Loss of ribosomes in Sicotiatla glutinosa L. infect,ed with lettuce necrotic yellows virus. Virology 41, 459-464. R.\NDLE:S, J. W., and FRINCKI, R. I. B. (1965). Some propert.ies of a t.obacco ringspot virus isolate from South Australia. rlust. J. Biol. Sci. 18, 979-986. SCHNEIDER, I. R., and DIENER? T. 0. (1966j. The correlat,ion between the proportions of virusrelated products and the infect.ious component during the synthesis of tobacco ringspot virus. Virology 29, 92-99. SCHNEIDER, I. R., and DIENER, T. 0. (1968j. I,, rliso and in vitro decline of specific infectivit,y of tobacco ringspot, virus correlat,ed with nucleic acid degradation. T,‘iiologly 35, 150-157. So~mrosr, F., FEDORCKX, I., GULTIS, A., F.IRK\S, G. L.? and EHRENIIEILG, L. (1968). A new method based on the use of diethyl pyrocarbonat,e m a nuclease inhibitor for the extraction of undegraded nucleic acid from plant, tissues. Eltr. J. Biochem. 5, 520-527. STICK-SMITH, R., KEI~HMANN, M.E.,and WRIGHT, N. S. (1965). Purification and properties of tobacco ringspot virus and two RNA-deficient components. l’irology 25, 487-494. VAN GRIENSVEN, L. J. L. I)., and KIN Iikninrb:N, 4. (1969). The isolation of ribonuclease-resistant RNA induced by cowpea mosaic virus: evidence for two double-stranded RN-4 components. J. Gen. Viral. 4, 423%$28. VIN K.UIhIEN, .4. (1971j. COWpea InOSaiC VirUS, un virus au genome divisC. Physiol. ITeg. 9, 4i9485. ZAITLIN, M., SPENCER? I)., and WHITFELD, P. R. (1968). Studies on the intracellular sit.e of tobacco mosaic virus assembly. It! “Proceedings of the Int,ernational Symposium on Plant Biochemical Regulation in Viral and Ot,her Dispp. 91-103. Kyoritsu Printing ease or Injury,” Co., Tokyo, Japan.