Search

Cell-free preparations from encapsulated Artemia salina embryos readily translate poly(U), but there is no endogenous protein synthesis or translation of added natural mRNA until development resumes upon incubation in saline. High-salt-washed 80S ribosomes and highspeed supernatant (cytosol) were prepared from undeveloped eggs and from eggs allowed to develop to a stage prior to hatching. Endogenous protein synthesis occurs only with ribosomes from developed embryos, whether with undeveloped or developed cytosol. This is mainly elongation of preformed polypeptide chains, for it is largely resistant to edeine, an inhibitor of chain initiation. Edeinesensitive translation of added natural mRNA occurs with both developed and undeveloped ribosomes but requires developed cytosol. Translation of brome mosaic virus RNA is not much, if at all, further stimulated by high-saltwash of developed ribosomes, but that of globin mRNA is markedly enhanced. Levels of the chain initiation factor EIF-1 are the same before and after development. These results are consistent with the view that resumption of developemtn is triggered by transcription, with ensuing translation of the resulting messengers to yield, among other proteins, mRNA-recognizing initiation factors of the IF-3 type which are partly free in the cytosol and partly ribosome-bound. The data also suggest that different factors may be involved in the translation of brome mosaic virus RNA and globin mRNA by this system.

The self-assembly of Brome mosaic virus protein has been studied as a function of pH and ionic strength at 22°. At ionic strengths above 0.25, the dissociated protein (dimers of the subunit) polymerized quantitatively below pH 5 into empty capsids or PTC (pseudo top component) in a fully reversible manner. The threshold of reassociation was displaced to higher pH values with increasing ionic strength. Once formed, PTC particles could as well be dissociated into ribbons of protein at low pH, but at low ionic strength only; in the latter configuration, BMV protein was unable to form shells upon reverting to initial conditions. Filamentous structures also formed from dissociated protein when reassociation was performed at low ionic strength; these ribbons too could hardly, if ever, give rise to PTC particles upon shifting to conditions promoting shell formation. Here again, an increase in ionic strength at constant pH resulted in a higher degree of polymerization, suggesting a major role for hydrophobic forces in the formation of organized structures from BMV protein subunits.

The RNA from bacteriophage Qβ can be translated by cell-free extracts from wheat embryos. This translation, by 80 S ribosomes, occurs at a low magnesium ion concentration. Three products are synthesized which coelectrophorese with Qβ proteins synthesized in Escherichia coli extracts. The smallest of these has been identified as coat protein. Although the polycistronic bacteriophage message is translated with fidelity, the efficiency is much less than when the monocistronic brome mosaic virus coat protein message is translated.

The ability of potato spindle tuber viroid (PSTV) RNA to act as a messenger RNA for protein synthesis has been investigated in several cell-free protein synthesizing systems. Extracts from wheat embryo, commercial wheat germ, Escherichia coli and Bacillus stearothermophilus , which are known to translate RNAs from viruses that do not infect these organisms, are unable to synthesize any specific proteins directed by PSTV. Amino acid incorporation is not stimulated much above the control without PSTV. This stimulation is less than 1% of the activity due to Qβ and brome mosaic virus RNAs, which can be translated into specific products under the same conditions. We tentatively conclude that this form of PSTV RNA is not a messenger for protein synthesis.

We have studied a number of plant viral RNAs by gel electrophoresis under denaturing conditions. Assuming a linear relationship between electrophoretic mobility and log (molecular weight)−1 the following mean molecular weights were calculated for the RNAs of: cucumber mosaic virus, 1.21, 1.06, 0.79, 0.33 × 106; brome mosaic virus, 1.19, 1.07, 0.80, 0.275 × 106; cowpea chlorotic mottle virus, 1.20, 1.07, 0.81, 0.25 × 106; cowpea mosaic virus, 2.02, 1.37 × 106; tobacco rattle virus, 2.37, 0.84 × 106; tobacco mosaic virus, 2.07 × 106. The significance of these data is discussed in relation to molecular weight values for the same RNAs published by others.

Summary Neither tobacco mesophyll protoplasts nor tobacco leaves could be infected with wild type brome mosaic virus (BMV wt) or its RNA, but both were infected when inoculated with 50 µg/ml or more of a variant of BMV (BMV V5). This is about 100 × the amount required for efficient infection by cowpea chlorotic mottle virus (CCMV). Poly-l-ornithine stimulated the infection of protoplasts with BMV V5, but was not essential. BMV V5. RNA inoculum caused infection of protoplasts under conditions similar to those needed for their infection with CCMV-RNA.

Coat protein subunits from cowpea chlorotic mottle virus, brome mosaic virus, and broad bean mottle virus have been mixed in various combinations and ratios and added to RNA to form spherical particles with mixed protein coats. These particles, some of which are infective and resistant to snake venom phosphodiesterase, are immunoelectrophoretically distinct from nucleoproteins with any single type of coat protein.

Molecular weights of virus protein subunits of cowpea chlorotic mottle virus (CCMV), broad bean mottle virus (BBMV), and brome mosaic virus (BMV) were determined by means of polyacrylamide gel electrophoresis. Values of 19,200 for CCMV, 16,400 for BBMV, and 16,200 for BMV were obtained. Amino acid compositions of the three viruses have been determined and compared with values reported earlier. Based on these data CCMV, BBMV, and BMV were found to possess 179, 152, and 150 amino acid residues per protein subunit, respectively. The CCMV subunit on trypsin treatment gave one component with a molecular weight of 16,400 and 154 residues. Unlike CCMV in which only certain lysyl and arginyl bonds in the N-terminal portion of the protein are hydrolyzed by trypsin, BBMV did not show any selective response and some of the virus particles were completely digested under the same conditions. Thus the BBMV protein subunits are much more susceptible to tryptic action than those of CCMV. Chymotrypsin treatment of BBMV hydrolyzed some virus particles completely while the protein of a small fraction was only partially hydrolyzed, suggesting the presence of peptide bonds susceptible to attack by enzyme in either terminal of the molecule. Trypsin treatment of BMV produced at least four components. A stable spherical particle of 16 nm diameter (36 S) with an amino acid composition different from that of the virus and containing 6% RNA was isolated after gel chromatography and density gradient centrifugation. However, a precipitate that formed during the digestion was found to contain three components with molecular weights of 11,000, 9700, and 8800. Chymotrypsin treatment under similar conditions had no effect on this virus.

Although the wheat curl mite, Aceria tulipae, reared on wheat plants infected with brome mosaic virus (BMV) failed to transmit the virus to wheat plants, the virus was detected in high concentrations in these mites (“exposed mites”) by infectivity, specific serological reaction, and electron microscopy of mite extracts. Exposed mites maintained the virus in their bodies in an infective and morphologically unaffected state for at least 6 days when kept on virus-free wheat plants. Thin sections of exposed mites revealed infection of a large part of the alimentary system (midgut and a part of foregut) with the virus. BMV virions, in crystalline arrays, were packed in a large number of membrane bound, ball-like structures in the gut cells and uniformly filled the gut lumen. Cytopathic effects such as degeneration of microvilli, reduction in the number and size of mitochondria, and disintegration of the endoplasmic reticulum were noted in severely infected gut cells where almost all the cytoplasm was replaced by masses of virus. Virions were also found in the body cavity in different parts of the mite's body. No viruslike particles were found in salivary glands or nervous tissue, or in virusfree mites examined. When virusfree mites were allowed a 15–16 hr acquisition access period on BMV infected plants, held subsequently on healthy plants, and were examined immediately after and at different subsequent times, a clearly discernible increase in virus content of the midgut cells was observed. BMV infection in the gut cells by day 7 after the start of acquisition access showed an abundance of large, multilayered virus crystals and aggregates of unarranged virions embedded in the cytoplasm. Nurse cells lining the genital canal were also infected by the virus. Cylindrical- or spindle-shaped inclusion bodies composed of long, thin filaments occurred in the gut lumen of A. tulipae that had a 15–16 hr acquisition access to virus-infected plants. BMV appears to multiply locally in the gut tissue of the mites, indicating that transmissibility and specificity of a virus to an arthropod vector may not always be a function of virus multiplication in the arthropod. Possible reasons for inability of A. tulipae to transmit BMV are discussed.

Crude sap from Moore barley infected with brome mosaic virus (BMV) and purified BMV produced two and sometimes three precipitation lines when diffused against an antiserum to purified virus in the Ouchterlony double-diffusion test. Antigens, which sedimented more slowly than infectious virus, were detected after density-gradient centrifugation at pH 7.0 and 8.0 of crude sap from BMV-infected plants and of BMV previously purified by density-gradient centrifugation, a result indicating that the virus dissociated during density-gradient centrifugation. The stability of crude and purified virus preparations decreased as pH was increased from pH 5.0 to 8.0, whether stability was measured by loss of infectivity or by the increase in concentration of breakdown products. Materials in boiled extracts of healthy barley stabilized the virus. The stabilization could not be detected at pH 5.0, but it became increasingly greater as pH was increased from pH 6.0 to 8.0. Purified virus dissociated into two subunits of uniform size. These subunits, antigens 1 and 2, were noninfectious; apparently they were nucleoproteins. Noninfectious antigens from above the infectious zone in density-gradient columns were concentrated by high speed centrifugation and, when analyzed by the Ouchterlony test, formed precipitation lines identical to those formed by purified virus. It is suggested on the basis of immunoelectrophoretic analysis of crude sap and purified virus preparations that the virus might have dissociated in the agar gel to produce these antigens.

Purified brome mosaic virus was stabilized at pH 7 by 10 −2 , 10 −3 , and 10 −4 M MgCl 2 and by 10 −2 M CaCl 2 , and at pH 6 by 10 −2 M MgCl 2 , the only concentrations tested. The virus was less stable at pH 7 in 0.1 M NaCl or 0.1 M potassium phosphate buffer than in 0.01 M buffer only, whereas at pH 6 it was more stable at the higher salt concentration. Sodium ascorbate, sodium sulfite, sodium ethylenediamine tetraacetate, and sodium oxalate decreased the stability of the virus at pH 7. The virus was more stable in tris(hydroxymethyl)aminomethane buffers than in phosphate. Healthy barley sap stabilized purified brome mosaic virus, but the extent of stabilization was not consistent from experiment to experiment. Infectivity assays were used to determine the concentration of surviving virus, but the stabilization by magnesium could also be shown by analytical density-gradient centrifugation. Frozen virus survived better in 0.01 M Tris-citrate pH 6.5 buffer with 0.001 M MgCl 2 and 0.1 M glycine and in 0.01 M acetate pH 6.0 with 0.001 M MgCl 2 and 0.1 M glycine than in other buffers tested.

Cowpea chlorotic mottle virus (CCMV) sediments as a single species but is heterogeneous when centrifuged in CsCl. The heterogeneity is not due to degradation. The original interpretation of the nucleoprotein distribution in CsCl was based on the assumption that two populations of particles of equal densities but different diffusion rates formed in CsCl, the more diffuse band being composed of monomers and the less of tetramers (Bancroft et al. 1968). However, in light of the recently demonstrated behaviour of CCMV-RNA in polyacrylamide gels (Fowlks & Young, 1970; Bancroft, 1971) and the correlation between band density and RNA size found by Lane & Kaesberg (1971) for brome mosaic virus which has a similar RNA distribution to that of CCMV, a reconsideration of the explanation for the behaviour of CCMV in CsCl was thought to be appropriate and is now presented. CCMV was grown, purified and assayed as described (Bancroft et al. 1968).

A comparison was made of the mobility rates of 13 strains of tobacco mosaic virus by a zonal electrophoresis technique using phosphate buffer at pH 6, 7, and 8 and using 2 gradients of sucrose. The common type strain, AC-3, was used as an internal standard. All of the laboratory isolates and two naturally occurring strains migrated similarily to AC-3; four naturally occurring strains were different. The difference between the groups of the slow-moving (unlike AC-3) and fast-moving (like AC-3) strains was distinct. A Tiselius electrophoresis technique was used with several strains to make a comparison with the data from the zonal technique. Serological analysis resolved the strains into two groups also, but all naturally occurring strains were differentiated from the AC-3 strain. Three different buffer solutions at pH 6, 7, and 8 were used to compare the mobility rates of four plant viruses. The pH of the solution caused little change in migration, but the buffer effected the rates. Both Tris and veronal buffer produced faster migration of the viruses than did phosphate buffer. In all buffers, brome mosaic virus failed to show much migration from the origin.

Broad bean mottle virus (BBMV) is a spherical plant virus belonging to the same group as brome mosaic virus (BMV) and cowpea chlorotic mottle virus (CCMV) (Bancroft, 1970). BMV and CCMV have recently been shown to have divided genomes (Lane & Kaesberg, 1971; Bancroft, 1971) contained in virus particles which sediment as a single species. The RNA from BBMV, when examined in the ultracentrifuge, has been shown to be heterogeneous (Kodama & Bancroft, 1964), consisting of a 17S infectious species, and a 10S species (Bancroft et al. 1968) which was considered to be a degradation product about one-third the size of the large piece. As with the other members of this group, further heterogeneity of the fastest sedimenting species has been found on electrophoresis of the RNA on polyacrylamide gels (Fowlks & Young, 1970).

Brome mosaic virus (BMV) and cowpea chlorotic mottle virus (CCMV) are spherical plant viruses which have recently been shown to possess divided genomes (Lane & Kaesberg, 1971; Bancroft, 1971) even though the virus particles sediment as single species. BMV contains four species of RNA with molecular weights of 1.09 × 106, 0.99 × 106, 0.75 × 106 and 0.28 × 106 and CCMV contains four species with molecular weights of 1.15, 1.0, 0.85 and 0.32 × 106 (Fowlks & Young, 1970). The species are named 1, 2, 3 and 4, respectively. Species 1 + 2 + 3 are required to initiate BMV infections (Lane & Kaesberg, 1971) but CCMV seems only to require species 1 + 2, although the addition of species 3 enhances lesion production (Bancroft, 1971). Species 4 is needed by neither virus for infection. I wished to determine if a virus possessing some properties of both viruses could be obtained by mixing BMV species 1 + 2 with CCMV species 3 prior to biological assay.

Biochemistry @ Copyright 1973 by the American Chemical Society Volume 12, Number 2 January 16,1973 Fidelity of Translation of Satellite Tobacco Necrosis Virus Ribonucleic Acid in a Cell-Free Escherichia coli Systemt R. Rice* and H. Fraenkel-Conrat The polypeptides synthesized in an E. coli cell-free system under the direction of the small RNA of the satellite tobacco necrosis virus were compared with the authentic coat protein of this virus. Many of the tryptic peptides obtained from the two types of proteins moved identically upon high- resolution cation exchange column chromatography, but distinct qualitative and quantitative differences between the two peptide patterns were also evident. Gel electrophoresis in 9 M urea at pH 4.0 showed that the biosynthesized material was not homogeneous in regard to charge; some of it showed a ABSTRACT: E scherichia coli cell-free systems synthesize recognizable proteins when programmed with messengers from the small RNA bacteriophages (Stavis and August, 1970). These extracts generally respond better to exogenous messenger RNA, particularly in quantitative respects, than do incorporation systems from plant and animal sources. The genetic code, possibly including initiation (Stewart et at., 1971 ; Housman et a/., 1970; Marcus et a/., 1970) and termination (Goldstein et al., 1970) codons, seems to be identical in the animal, plant, and bacterial kingdoms. Hence, E. coli extracts have been used with animal and plant viral RNA messengers to investi- gate translation in these heterologous systems and to study the viral gene products in vitro. The most direct test to establish whether accurate translation of a viral messenger occurs has been to ascertain whether the viral coat protein is synthesized in vitro since other gene prod- ucts have not been as readily identifiable. Some studies have indicated considerable similarity between translation products and viral coat protein. For example, with the RNA from the t From the Department of Molecular Biology and Virus Laboratory, University of California, Berkeley, California 94720. Receired August 18, 1972. This investigation was supported by U. S. Public Health Service Grant No. G M 01389 from the National Institute of General Medical Sciences, and Grant No. GB 6209 from the National Science Foundation. R . Rice was a recipient of a National Science Foundation Graduate Fellowship. * Address correspondence to : Institute of Marine Resources, Uni- versity of California, Davis, Calif. 95616. mobility similar to that of the authentic coat protein. Gel electrophoresis in the presence of sodium dodecyl sulfate demonstrated the presence of a wide range of molecular weight species in the biosynthesized protein ; the largest, representing about 5 % of the total, coincided with authentic virus coat pro- tein. It is concluded that much of the information on satellite tobacco necrosis virus RNA is translated into proper amino acid sequences in the E. coli system, but that only very little complete coat protein, and that probably carrying a terminal formyl-methionyl group, is synthesized. satellite of tobacco necrosis virus (STNV)' (Clark et a/., 1965) and the similarly small top a component of alfalfa mosaic virus (van Ravenswaay Classen et a/,, 1967), two-dimensional peptide maps have been reported as showing nearly complete coincidence of radioactive spots from the translation products and ninhydrin-positive spots from the coat proteins. In con- trast, earlier studies with TMV-RNA had indicated no clear evidence for coat-related material being synthesized in citro (Aach et a/., 1964; Schwartz, 1967). Studies with brome mosaic virus RNA (Stubbs and Kaesberg, 1967) and poliomyelitis virus RNA (Rekosh et al., 1970) gave somewhat ambiguous results and the recent claim that avian myeloblastosis virus RNA directs the synthesis of viral antigens (Siegert et a/., 1972) must await substantiation. In view of these widely varying results, it was our intention to reinvestigate one of the reports indicating positive results. We chose to characterize the translation products of STNV- RNA using more discriminating techniques which have be- come available since the original report appeared. The RNA of STNV is a convenient messenger for this kind of work in that it is small (0.4 X lo6) and can code maximally for only one small protein in addition to the viral coat protein (Roy et at., 1969). Since the completion of this work, an interesting study of the translation of STNV-RNA in E. coli and wheat embryo 1 Abbreviations used are: STNV, satellite tobacco necrosis virus; TMV, tobacco mosaic virus; TPCK, ~-(l-tosylamido-2-phenyl)ethyl chloromethyl ketone. BIOCHEMISTRY V O , L . 1 2 , NO.

Different components of brome mosaic virus (BMV) RNA, either separate or mixed, stimulated similarly the activity of soluble BMV-RNA polymerase. When unfractionated RNA was heated, the various compone

Component four of brome mosaic virus RNA is a structural component of the virus but is not required for infection. Sequence analysis of oligonucleotides resulting from pancreatic ribonuclease digestion of the brome mosaic virus RNA's indicates that the nucleotide sequence of component four occurs in brome mosaic virus RNA component three but not in components one and two.

Summary The RNA polymerase activity of a cell-free particulate fraction of barley leaves infected with brome mosaic virus was investigated in the presence of actinomycin D and EDTA. The RNA products were fractionated according to their solubility in 2 m-LiCl. Both LiCl-soluble and LiCl-insoluble fractions of the RNA synthesized during a 2 min. pulse of [3H]-UTP were largely resistant to RNase in high salt concentration; the radioactive RNA of the insoluble fraction had sedimentation properties expected for replicative intermediates. After a 2 min. chase, LiCl-insoluble radioactivity was essentially RNase-sensitive in high salt, and sedimented in sucrose gradients in association with all three components of brome mosaic virus RNA, provided protective exogenous RNA was added throughout the experimental procedure. The overall results suggest that the in vitro synthesis of the RNAs found in virus particles occurs by a continuous process resulting possibly from recycling activity of the polymerase.

The molecular weights of four well-characterized single-stranded viral RNAs (turnip yellow mosaic virus-RNA, bacteriophage MS2-RNA, and cucumber mosaic virus-RNA components B and A) were obtained by sedimentation velocity ultracentrifugation after reaction with formaldehyde. The logarithms of their molecular weights were linearly related to the electrophoretic migration distances of the non-formylated RNAs in 2.4% polyacrylamide gels. This relationship was used to estimate the molecular weights of the RNAs from tobacco ringspot virus, southern bean mosaic virus, carnation mottle virus, tobacco necrosis virus, and brome mosaic virus by electrophoresis, and to compare them with the values obtained by means of formylation and sedimentation velocity analysis.

A considerable number of plant viruses produce in infected plants more than one nucleoprotein component. Typical examples are tobacco rattle virus (TRV) , alfalfa mosaic virus (AMV), viruses of the cowpea mosaic virus (CPMV) group, tobacco streak virus (TSV) and brome mosaic virus (BMV). It has been shown in recent years for some of these muItiparticulate viruses that the genetic information necessary for virus multiplication is di­ vided among two or more nucleoprotein components. Discovery of plant viruses with a divided genome has changed the con­ cept of the nature of viruses. The classical concept implied that the mature virus, the ultimate phase of viral development, was a single particle, the vir­ ion, containing the genome of the virus. This is of course still true for many plant viruses, but no longer for all. Several important groups of plant viruses have been recognized to have a genome divided among two or more physi­ cally separable particles. A certain heterogeneity occurs in practically all purified preparations of plant viruses. In preparations of tobacco mosaic virus (TMV) for example, rodshaped particles with a modal length of 300 m,u predominate, but smaller and longer rods occur also. The presence of such particles has been explained by possible fragmentation of 300 mft particles during purification, faulty syn­ thesis of particles, or, in case of longer particles, by aggregation. Such hetero­ geneity is quite normal in preparations of rodshaped and threadlike viruses. Similarly, preparations of spherical viruses may contain some particles with fragmented RNA. This type of particle we shall regard as defective; it con­ sists of viral structural proteins and fragments of the viral genome. It is possi­ ble that these defective particles have some biological significance, and that they can interfere with virus replication. The infectivity of TMV, for exam­ ple, is enhanced by the presence of short particles derived from the virus in the inoculum (23, 37). In these cases it is a question of occurrence of defec­ tive particles, besides nondefective homOlogous virus particles which contain a complete genome necessary for multiplication (36). Heterogeneity in such virus preparations does not reflect functional heterogeneity. This is different from the multiparticulate viruses we will discuss here. Plant viruses with a divided genome are characterized by physical hetero

From brome mosaic virus-infected barley leaves an RNA-dependent RNA polymerase activity was isolated by solubilization with Triton X-100. This enzyme, active in the presence of actinomycin D, was stimulated about 5-fold by the presence of an RNA and the four nucleoside triphosphates. The RNAs of brome mosaic virus and the closely related cowpea chlorotic mottle virus were the most effective, but some activity was also shown by certain other viral nucleic acids and polyribonucleotides. Enzyme activity was optimal at pH 8.0 and at 32 mM Mg2+ concentration and depressed by increasing monovalent ion concentration. The enzyme activity could be precipitated with ammonium sulfate, and it could be separated from the bulk of the plant components by sucrose gradient centrifugation where it moved with 5–7 S. However, its further purification was vitiated by its great instability.

Summary We have assigned mutations in cowpea chlorotic mottle virus and the related brome mosaic virus to specific RNA components. Coat protein changes, as well as associated ‘maturation faults’ and systemic symptom changes, are located on RNA species 3; local and primary lesion alterations are found on either RNA species 2 or 3. Changes in nucleoprotein component ratio are associated either with the coat protein changes on RNA species 3 or with primary lesion alterations on species 2. Local lesion temperature sensitivity is found on RNA species 1.

THE ability of RNA from several plant viruses to bind a specific amino acid in a manner like tRNA is now well established. All members of the tymovirus group so far examined contain RNA which can bind valine1–3; all members of the bromovirus group bind tyrosine (ref. 4 and T.C.H., in preparation), and tobacco mosaic virus (TMV) RNA binds histidine5. The ubiquity of the reaction, and the similarity of the esterification to that of aminoacylation of tRNA indicates that a significant biological function for the property exists. A tRNA-like function such as that of a virus-specific tRNA, or a missense tRNA seems plausible. Although there is evidence that turnip yellow mosaic virus (TYMV) RNA, or a fragment containing the 3′ terminus of TYMV RNA, may donate valine to nascent polypeptides6, it has been established that tyrosine, bound to brome mosaic virus (BMV) RNA, is not donated to peptidyl material in a cell-free system from wheat embryo when directed by viral, plant, or synthetic messenger RNA (ref. 7).

CELL-FREE systems which are capable of accurate translation of exogenous mRNA have proved invaluable tools for studying the mechanism of protein synthesis both in prokaryotes1 and, more recently, in eukaryotes2. Since the availability of suitable mRNAs is often a limiting factor in such studies, RNA-containing viruses have been widely used as sources of RNA for translation in bacterial, plant and animal cell-free systems; for example, the bacteriophages f2 (ref. 3) and Qβ (ref. 4), the plant viruses brome mosaic virus5 and satellite tobacco necrosis virus6 and the picornaviruses, encephalomyocarditis virus7 and mengo virus8. But, although picornavirus RNA is actively translated in animal cell-free systems, it is not an ideal messenger since its large size9 and single polypeptide chain initiation sit10, 11 hinder complete translation in vitro12. As alternative sources of mRNA for animal cell-free systems, we have studied the plant viruses brome mosaic virus (BMV) and tobacco rattle virus (TRV). Here we report the efficient and accurate synthesis of the coat proteins of these viruses in a mammalian cell-free system directed by the viral RNAs.

Summary The RNA of broad bean mottle virus (BBMV), cowpea chlorotic mottle virus (CCMV), and cucumber mosaic virus (CMV) bound tyrosine, as does brome mosaic virus (BMV) RNA. Other amino acids were not bound. Comparison of the efficiency of synthetase enzymes from Escherichia coli, bean, wheat, and Saccharomyces cerevisiae in aminoacylation of RNA from these viruses, and from tobacco mosaic virus (TMV) and turnip yellow mosaic virus (TYMV), showed that the plant enzymes were effective in tyrosylation of RNA from bromoviruses and CMV, whereas the yeast and bacterial enzymes were ineffective. The E. coli enzyme did not esterify TMV RNA with histidine, and the bean enzyme was poor in this ability. All enzymes were able to catalyse valine binding to TYMV RNA. However, 40 mm-KCl inhibited valine binding to TYMV RNA by the bacterial enzyme while its ability to catalyse binding by tRNA was not affected.

In both 1.5 M NaCl and 1 M CaCl 2 , brome mosaic virus (BMV) particles are unaffected at low pH but are degraded above a well defined pH value. The degradation into protein subunits and RNA occurs in two steps in both NaCl and CaCl 2 . First, the RNA is released without collapse of the protein shell: the products obtained are RNA and artificial top component (ATC); in a second step, the ATC formed undergoes degradation into protein subunits upon further pH rise. The main difference between the action of 1 M CaCl 2 and 1.5 M NaCl is that, with CaCl 2 , no intermediates of degradation were found between virus and ATC, while in NaCl, some particles underwent progressive degradation, yielding intermediate forms together with reassociation products and contaminated protein subunits.

The products synthesized in vitro by the detergent solubilized BMV-RNA polymerase with complete brome mosaic virus-RNA were found after phenol extraction to be 60–70% ribonuclease-resistant and soluble in 2 M LiCl, and to behave as double-stranded RNA upon hydroxyapatite chromatography. Gel electrophoresis indicated that the approximate molecular weight of the main and first product to be synthesized was relatively small with lesser amounts of a larger and a smaller molecular weight product. Replacement of complete BMV-RNA in the in vitro assay reaction mixture with either BMV-RNA components 1 + 2, 3, or 4, also resulted in the predominant production of the small molecular weight product, especially when components 3 or 4 were used. Higher molecular weight products were not observed when component 4 was used; however, such products were formed when component 3 was utilized. The products synthesized in vitro by the particle-bound enzyme were about 70–85% resistant to RNase and the major component had a relatively high molecular weight. The product was the first to be synthesized, and its labeling increased with increasing incubation time. Addition of complete BMV-RNA to the bound enzyme reaction mixture did not change the nature of the products.

Four major RNAs (A, A′, B, D) and one minor RNA (C) have been isolated by polyacrylamide gel electrophoresis from purified brome mosaic virus. All RNA components were stable when treated with heat, DMSO, and formamide, thus excluding the possibility of hidden breaks. The mixture A + A′ + B was infectious. Analysis of the replicative forms (RF) isolated from infected plants showed that only the three largest RNAs (A, A′, B) had their own RFs. 32 P incorporation showed that the viral RNAs were synthesized at different rates. The specific activities of A + A′ and D were identical and were always higher than that of B. The minor component C was rapidly labeled but very little was encapsulated. The origin and significance of RNAs C and D are discussed.

The expression of the transforming ability of Haemophilus influenzae DNA was investigated after its encapsidation by the coat protein of two different plant viruses, Brome Mosaic Virus (BMV) and Alfalfa Mosaic Virus (AMV). The influence of the encapsidation on the various steps of the transformation process was studied, as well as the protection of the DNA molecule inside of the DNA-protein complex particles against nucleolytic attack.

The behavior at different salt and acidic pH levels of protein isolated from three spherical plant viruses was investigated. Broad bean mottle virus protein did not aggregate into recognizable organized structures. Protein from brome mosaic virus formed capsids similar to those of the virus at a pH and salt-dependent efficiency with best yields at pH 5.0 in 0.2 M NaCl. Best yields for cowpea chlorotic mottle virus protein capsids occurred at pH 5.0 in 0.1 M NaCl or from pH 3.5 to 5.0 in 0.2 M NaCl. Protein from the latter virus also formed double-shelled and rosettelike particles in 0.2 M NaCl from pH 5.3 to 5.7. Narrow tubes were made at pH levels of 6.0 and higher, and the tubes were accompanied by T = 1 and T = 3 particles if ribonu-clease-digested RNA was present. In the absence of NaCl in 0.01 M acetate buffer at pH 4.0 and 4.5, laminar and platelike aggregates were found with CCMV protein. The production and some of the properties of the various forms are described.

Brome mosaic virus degraded to RNA in 0.02 M tris buffer, 0.001 M EDTA, pH 7–9, containing bentonite and 0.1 M NaCl. The virus coat protein was apparently adsorbed by the bentonite. In bentonite in buffer without NaCl, brome mosaic virus degraded to three nucleoproteins, each containing one piece of RNA. The largest nucleoprotein was infective. Tobacco mosaic virus degraded to RNA and protein in the above buffer with bentonite, but degraded only to short, variable length rods when 0.1 M NaCl was added. Tobacco mosaic virus coat protein apparently was not adsorbed by the bentonite. Tobacco mosaic virus RNA prepared by bentonite treatment and density-gradient centrifugation had about 20% the specific infectivity of RNA in intact virions.

RNA extracted from purified barley stripe mosaic virus (BSMV) was analyzed by density gradient centrifugation and polyacrylamide gel electrophoresis. All strains tested had a prominent 21.3 S RNA, which was separated into two components by gel electrophoresis. Most preparations of North Dakota 18 and Argentina Mild strains had an additional 19.5 S component, which was resolved into one or two RNA species by gel electrophoresis. Some preparations also contained variable amounts of two or three minor RNA components migrating slowly in gels; these components sedimented at about 30 S in sucrose gradients. The two 19.5 S components were more variable than other components, and occasionally one or both disappeared during virus subculture. Double-stranded RNA isolated from infected plants migrated more slowly in polyacrylamide gels than did single-stranded RNA isolated from purified virions, but both types of RNA from the same strain had the same number of major RNA components. The RNAs of BSMV migrated more slowly in gels than expected from their sedimentation rates if brome mosaic virus RNAs and barley ribosomal RNAs were used as standards, which suggests that BSMV RNAs and the standard RNAs differ in secondary configuration. Mixtures of the separated RNA components were more infectious than single components, which indicates that more than one RNA component was necessary to establish infection.

Barley plants infected with both barley stripe mosaic virus (BSMV) and brome mosaic virus (BMV) had more severe symptoms than plants infected with either virus alone, although the viruses were present in lower concentrations than in singly infected plants. To determine whether BMV protein encapsidated BSMV RNA in vivo , assay plants were inoculated with BMV virions purified from doubly infected plants. These assay plants later contained a small number of BSMV virions in addition to a high concentration of BMV virions. The BSMV virions presumably resulted from infection by BSMV RNA encapsidated by BMV protein.

RECENT studies have shown that the RNA (mol wt 2 × 106) of turnip yellow mosaic virus (TYMV) possesses at its 3′ end a structure similar to that of a tRNA. This is deduced from the fact that in the presence of valyl-tRNA synthetase, the 3′ end of TYMV RNA can be acylated with valine1,2. This RNA is also recognized by the tRNA nucleotidyltransferase3 and by the peptidyl-tRNA hydrolase2,4. Similar observations have been reported with tobacco mosaic virus (TMV) RNA which can be charged with histidine5,6 or with either methionine or serine7, and brome mosaic virus (BMV) RNA with tyrosine (Hall, Shih and Kaesberg, in the press).

The four RNAs of brome mosaic virus induce substantial incorporation of amino acids into protein when used as messengers in a cell-free protein-synthesizing system derived from wheat embryo. RNA 4 is highly efficient as a monocistronic messenger for the viral coat protein. Acetate, derived from acetyl coenzyme A, is incorporated into the product made in vitro. Although RNA 3 also contains the coat-protein cistron, it induces synthesis mostly of a protein larger than coat protein. RNAs 1 and 2 also induce the synthesis of substantial amounts of protein other than coat protein. However, an equimolar mixture of RNAs 3 and 4 or of 1, 2, 3, and 4 induces synthesis of coat protein almost exclusively. This result suggests that the coat-protein cistron, when present as a monocistronic messenger, inhibits translation of all other viral messages.

Four relatively stable, purified plant viruses in water-in-oil emulsions and in phosphorylated hesperidin solutions were used to immunize rabbits. Minimal quantities of antigen, 1.2–2.6 mg per milliliter per injection, were given once a week for 3–4 weeks by intramuscular or subcutaneous routes. Serum samples were collected periodically and were titrated by the complement fixation and by agar double diffusion techniques. The water-in-oil adjuvant produced serum of higher titer than the phosphorylated hesperidin adjuvant when tobacco mosaic virus and barley stripe mosaic virus were used as antigens. Both adjuvants produced sera of equally good titers when brome mosaic virus and southern bean mosaic virus were used as antigens. Under the conditions employed for immunization, the phosphorylated hesperidin adjuvant was never superior to the water-in-oil emulsions. The advantages of using the phosphorylated hesperidin adjuvant and the water-in-oil adjuvant with plant virus antigens are discussed.

A protein of 34,500 daltons was detected by means of SDS-gel electrophoresis in brome mosaic virus-infected barley leaves. This protein, synthesized only at an early stage of infection, was found to be associated with cell fractions rich in particulate bound BMV-RNA polymerase activity. After solubilization of this activity by Triton X-100 treatment, this protein was found only in the sucrose gradient fractions also containing polymerase activity.

Summary The molecular weights of the protein subunits of Odontoglossum ringspot virus, belladonna mottle virus and brome mosaic virus were estimated by the equilibrium centrifugation method of Yphantis to be 17,300, 20,700 and 17,200, respectively. Solutions of the virus protein subunit monomers were prepared by suspending the virus particles in neutral buffers containing 6m-urea and Cleland's reagent; preliminary tests with well-characterized proteins showed that the urea solution decreased the partial specific volume of the proteins by about 0.02 to 0.03 g./ml. In solvents without urea brome mosaic virus dimer subunits had a molecular weight of about 40,000.

Incubation of tyrosine with the RNA from brome-mosaic-virus under conditions favourable for aminoacylation of tRNA resulted in binding of tyrosine to the viral RNA. Sucrose-density-gradient centrifugation showed that tyrosine binds to three and probably to all four of the viral RNA components. The bound amino acid was readily released in mild alkaline solutions. Of 20 amino acids tested, only tyrosine was found to bind. The maximum extent of binding observed was 0.58mol of tyrosine/mol of brome-mosaic-virus RNA.

The reassembly of cowpea chlorotic mottle virus, brome mosaic virus, and broad bean mottle virus was examined under a variety of conditions. The experimental basis is presented for optimal reassembly occurring in terms of infectivity and sedimentation behavior when protein in 1 M NaCl near neutrality and RNA are mixed stoichiometrically prior to dialysis at about 4° against near-neutral low ionic strength solutions containing at least 5 × 10 −3 M MgCl 2 or various polyamines. The consequences of omitting divalent cations, lowering pH levels, and varying the ratios of protein and RNA are described. The reassembly process is discussed in terms of the conditions investigated and related particularly to the properties of natural cowpea chlorotic mottle virus.

The arrangement of the morphological units of cowpea chlorotic mottle virus (CCMV), which has a dry diameter of 250 A, is consistent with that expected for a 32-unit structure, the units being situated on the vertices of a rhombic triacontahedron. The hydrodynamic behaviour of CCMV is pH dependent. The particles “swell” near neutrality and in doing so become susceptible to attack by nucleases. After treatment with pancreatic ribonuclease, ellipsoidal particles of 210 A diameter and 280 A length, small icosahedral particles of 160 A diameter, double-shelled particles of 340 A, and long tubes with a diameter of 160 A and rounded ends are produced. All these particles are also found after treatment with takadiesterase T1 and in addition, small spherical particles with a double shell with an outer diameter of 250 A, and tubes with a diameter of 250 A which may enclose 160 A tubes, are produced. The production of these new forms are described and discussed and the products are related to model structures. In addition, a possible reconstitution system for CCMV and brome mosaic virus is outlined.