VH7 _ _‘ LIBRARY Michigan Staw University This is to certify that the thesis entitled A COMPARISON OF SOME BIOLOGICAL AND BIOCHEMICAL CHARACTERISTICS OF SEVERAL TOBACCO RINGSPOT VIRUS ISOLATES presented by Rose Gergerich has been accepted towards fulfillment of the requirements for Ph.D. Botany and degree in Plant Pathology ['57qu C, vflmf/ Major professor Date December 10, 1980 0-7639 F933 I994 OVERDUE FINES: 25¢ per day per item RETURNING LIBRARY MATERIALS: Place in book return to remove change from circulation records A COMPARISON OF SOME BIOLOGICAL AND BIOCHEMICAL CHARACTERISTICS OF SEVERAL TOBACCO RINGSPOT VIRUS ISOLATES By Rose Gergerich A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Botany and Plant Pathology 1980 ABSTRACT A COMPARISON OF SOME BIOLOGICAL AND BIOCHEMICAL CHARACTERISTICS OF SEVERAL TOBACCO RINGSPOT VIRUS ISOLATES By Rose Gergerich Tobacco ringspot virus (TRSV) is an RNA-containing plant virus which has a wide host range, including both herbaceous and' woody plants. This dissertation reports a study of the biological diversity among seven isolates of TRSV which were obtained from different herbaceous and woody plants. The purpose of this research was to identify TRSV isolates which were unique in the symptoms they induced in selected herbaceous hosts, and subsequently to compare some of the biochemical properties of these selected isolates. Seven different isolates of TRSV were collected from grape, cherry, two cultivars of blueberry, tobacco, soybean, and watermelon. These isolates were characterized on the basis of: (l) the symptoms they induced in herbaceous indicator plants, (2) sensitivity to thermal inactivation, and (3) serological properties. Three TRSV isolates which showed distinct antigenic differences and distinct differences in the symptoms they induced in the herbaceous indicator plants were selected for further comparisons. The three isolates selected for further study were isolates from blueberry, tobacco, Rose Gergerich and soybean. One of these three isolates was shown to be different in the overall charge on its virus particles as indicated by differ- ences in mobility in an electric field. lfl_yjtrg_synthesis of the protein products coded by the RNA of the three isolates of TRSV made it possible to look for isolate-dependent differences in the molecular weight of these protein products. The RNA from the three isolates of TRSV was extracted and used to direct protein synthesis in a wheat embryo cell-free protein synthetic system. Polyacrylamide gel electro- phoresis was used to separate and characterize the radioactively labeled protein products of jn_yjtrg_synthesis. The number of in 31339 protein products was found to be the same, regardless of the TRSV isolate used as the source of RNA in the cell-free protein synthetic system. However, an isolate-dependent difference in the molecular weight of one of the jg_yjtrg_protein products was demon- strated. Evidence obtained by jg_yjtrg_translation of the small RNA of TRSV indicated that the protein product which showed the isolate- dependent molecular weight difference was coded by the large RNA piece of the TRSV genome. ACKNOWLEDGMENTS I express my gratitude and appreciation to Dr. D. C. Ramsdell and Dr. James Asher who provided guidance and the opportunity to learn and grow throughout this course of study. I also acknowledge the advice and effort of Dr. Gary Hooper, Dr. Fritz Rottman, and Dr. Harry Murakishi who served on my guidance committee. The support of the Michigan Blueberry Growers Association is gratefully acknowledged. I thank all of the peOple of the Department of Botany and Plant Pathology for their friendship and assistance throughout the period of my studies at Michigan State University. To my husband Joseph I extend a special thanks for his encouragement and guidance during my graduate studies. ii TABLE OF CONTENTS LIST OF TABLES . LIST OF FIGURES INTRODUCTION LITERATURE Serology . . Electrophoretic Mobility . Nematode Transmissibility Host Range . . Symptom Expression . . Additional Proteins Found in the Virus Particle Viral- -Specific Proteins Found in Infected Tissue . Viral- -Specific Replicases Viral- -Specific Proteases . Viral- -Specific Proteins in Plant Virus Inclusion Bodies. MATERIALS AND METHODS Virus Isolates Symptom Expression on Herbaceous Indicator Plants Thermal Inactivation Tests Virus Purification . Serology . . Antisera Sources . Spur Formation Tests Cross- -Absorption Tests . . Electrophoretic Mobility of whole Virus Particles RNA Extraction . . . . Gel Electrophoresis of Purified TRSV- RNA . . Separation of TRSV- RNA on Sucrose Density Gradients . Infectivity of Purified TRSV- RNA . . . . . Detection of Poly (A) Tracts in TRSV- RNA . Translation of TRSV- RNA in the Wheat Embryo Cell- Free System . . . Page vi Page Analysis of Translation Products by SOS-Polyacrylamide Gel Electrophoresis . . . . 36 Dual- Label Experiments to Detect Small Differences in Molecular Heights of Protein Products Produced by Different Isolates of TRSV . . . . . . . . . . 38 RESULTS . . . . . . . . . . . . . . . . . . 39 Symptom Expression on Herbaceous Indicator Plants . . . 39 Thermal Inactivation . . . . . . . . . . . . . 50 Virus Purification . . . . . . . . . . . . . . 52 Serology . . . . . . . . . . . . . . 55 Spur Formation Tests . . . 55 Selection of Three TRSV Isolates for Further Study . . 59 Cross- Absorption Tests . . . . 59 Electrophoretic Mobility of Whole Virus Particles . . . 62 RNA Extraction and Characterization . . . . . . . . 65 Infectivity of Purified TRSV- RNA . . . . . . 72 Separation of TRSV- RNA on Sucrose Density Gradients . . . 72 Detection of Poly(A) Tracts in TRSV- RNA. . . . 78 Translation of TRSV- RNA in the Wheat- -Embryo Cell- Free System . . . . 83 Analysis of Translation Products by SDS- Polyacrylamide Gel Electrophoresis . . . . 86 Dual- Label Experiments to Detect Small Differences in Molecular Heights of Protein Products Produced by Different Isolates of TRSV . . . . . . . . 93 Protein Products Made by RNA- 2 of TRSV . . . . . . . l07 DISCUSSION . . . . . . . . . . . . . . . . . lll BIBLIOGRAPHY . . . . . . . . . . . . . . . . 123 iv LIST OF TABLES Table Page l. Herbaceous host reactions to seven isolates of tobacco ringspot virus (TRSV) . . . . . . . . 4O 2. Thermal inactivation of tobacco ringspot virus (TRSV) isolates . . . . . . . . . 51 3. Spur precipitin lines formed when antisera to isolates of tobacco ringspot virus (TRSV) were reacted with homologous and heterologous antigens in double gel-diffusion tests . . . . . . . . . . . 56 4. Cross-absorption tests for three isolates of tobacco ringspot virus (TRSV) . . . . . . . . . . . 6l 5. Electrophoretic mobility of whole virus for three isolates of tobacco ringspot virus (TRSV) . . . . 64 LIST OF FIGURES Figure I. ‘National Pickling' cucumber cotyledons four days after inoculation with three different isolates of tobacco ringspot virus (TRSV) showing differential reactions . . . . 2. 'National Pickling' cucumber seedlings four days after inoculation with three different isolates of tobacco ringspot virus (TRSV) showing differential reactions . . . . . . . - . . . 3. Nicotiana tabacum 'Havana 423' plants inoculated with three different isolates of tobacco ringspot virus (TRSV) showing differential reactions . . 4. Vigna unguiculata Cowpea 'SR' plants inoculated with three different isolates of tobacco ringspot virus (TRSV) showing differential reactions . . 5. Sucrose density-gradient centrifugation of 'Jersey' blueberry isolate of TRSV purified from cucumber cotyledons . . . . 6. Ouchterlony two-dimensional double-diffusion tests showing confluent precipitin lines . 7. Ouchterlony two-dimensional double-diffusion tests showing a reaction of non-identity . . 8. Agarose-formaldehyde gel electrophoresis of total RNA from the tobacco isolate of TRSV 9. Agarose-formaldehyde gel electrophoresis of total RNA from the 'Jersey' blueberry isolate of TRSV . lO. Agarose-formaldehyde gel electrophoresis of total RNA from the soybean isolate of TRSV vi Page 42 44 46 48 53 57 57 66 68 7O Figure ll. Sucrose density-gradient fractionation of TRSV-RNA l2. Analysis by agarose-formaldehyde gel electrophoresis of unfractionated TRSV-RNA and RNA-2 fractionated by two successive cycles of sucrose density gradient centrifugation . l3. Oligo(dT)-cellulose chromatography of TRSV-RNA . l4. Electrophoresis of RNA preparations from Oligo(dT)- cellulose chromatography in agarose-formaldehyde denaturing gels . . . . . . 15. Incorporation of 3H-valine into protein in the wheat embryo extract in response to added TRSV-RNA . ++ . . . . . . . 16. Mg requ1rement for am1no ac1d 1ncorporat1on into protein directed by TRSV-RNA in the wheat embryo extract . l7. K+ ion requirement for amino acid incorporation into pro- tein directed by TRSV-RNA in the wheat embryo system l8. Kinetics of 3H-valine incorporation into protein directed by TRSV-RNA from the 'Jersey' blueberry isolate in the wheat embryo system . . . l9. SDS-polyacrylamide gel electrophoresis of in vitro translation products using RNA from the tobacco isolate of TRSV 20. SDS-polyacrylamide gel electrophoresis of in vitro translation products using RNA from the soybean isolate of TRSV . 21. SDS-Polyacrylamide gel electrophoresis of in vitro translation products using RNA from the 'Jersey' blueberry isolate of TRSV 22. SDS-polyacrylamide gel electrophoresis of in vitro translation products of TRSL RNA from the tobacco isolate and the 'Jersey' blueberry isolate 23. SDS- polyacrylamide gel electrophoresis of in vitro translation products of TRSV- RNA from the tobacco isolate and the soybean isolate . . . vii Page 73 76 79 81 84 87 89 91 94 96 98 TO] 103 Figt 24 25 Figure Page 24. SDS-polyacrylamide gel electrophoresis 0f jn_vitro translation products of TRSV-RNA from the 'Jersey' blueberry isolate and the soybean isolate . . . . 105 25. SOS-polyacrylamide el analysis of the products of translation of (a? unfractionated TRSV-RNA, and (b) RNA-2 of TRSV 109 viii INTRODUCTION An important characteristic of any natural population is the diversity that exists within that population. It is the differ- ential selection of variants within a natural population that leads to evolution. The occurrence of new variants in a group of plant viruses is of both practical and theoretical importance to the plant virologist. Newly arisen variants which induce different symptoms in the host may damage crops to a greater extent, may circumvent established control measures, and may make breeding for host resistance to infection more complicated. Variants which differ greatly in virulence, host range, or vector specificity must have some changes in their biochemical structure to account for these changes in their biological proper- ties. By defining the biochemical changes leading to variants which accompany differences in biological properties, we will begin to understand these viral properties which are important in disease propagation and regulation. The related strains of a virus that infect different hosts have probably arisen, through mutation, from some common ancestral type. The strains which successfully infect new host plants are those mutants which are better adapted to these new hosts. Various examples are known where particular host plants allow the selective multiplication of certain strains of a virus when inoculated with a mixture of strains (26,59). In nature the selective multiplication of newly arisen variants in a host plant may account for the survival and eventual predominance of these new variants. Tobacco ringspot virus (TRSV), a member of the nepovirus group (19), has a broad natural host range, and exists as a number of serologically distinct strains (16,27). The natural host range of tobacco ringspot virus includes many economically important crops such as tobacco, blueberry, soybean, grape, cherry, and other annual and perennial crops. It also occurs in weeds associated with many of these crops (l9). Although the existence of several serological strains of TRSV has been reported in the literature (16,27,45,50, 51,60), there has been no attempt to do an in-depth comparison of these strains. Biochemical comparisons of plant virus strains in the past have dealt with the biochemistry of the molecules which make up the virus particle, namely the coat protein and the nucleic acid. An examination of the literature shows that many properties of plant viruses are associated with the composition of the coat protein of the virus. However, some examples in the literature indicate that there are other important properties of plant viruses which are not associated with the viral coat protein. This is not surprising when one considers that plant virus genomes code for the synthesis of several other virus-specific proteins which appear during the course of infection and multiplication in the plant. The following literature review will consider: (a) the evidence which indicates that viral coat protein determines a set of important plant viral properties, (b) the evidence that other viral functions are deter- mined by proteins other than the viral coat protein, and (c) the types of viral-specific proteins which have been identified in plant virus infection. This review of the literature will be limited to plant viruses which have RNA genomes, and which have a simple protein shell consisting of one or two polypeptides. The terminology used throughout this dissertation will be as defined below. Isolate: a variant of a virus which was collected from a particular host, i.e., the 'Jersey' blueberry isolate of TRSV was collected from a blueberry of the 'Jersey' variety. Multipartite genome: a genome which consists of more than one piece of genetic material, i.e., TRSV has a genome which consists of two pieces of RNA, hence TRSV has a bipartite genome. The RNAs of TRSV will be referred to as RNA-l (2.2 x l06 daltons) and RNA-2 (l.4 x l06 daltons). LITERATURE REVIEW Like many other plant viruses, TRSV has a simple coat protein consisting of one type of polypeptide. The molecular weight of this polypeptide has been variously reported to be 57,000 daltons (36), or a tetramer of a monomer of 13,000 daltons (5). Following is a review of the biochemical and biological characteristics of plant viruses which have been shown to be a function of virus coat protein. Serology The antigenicity of a plant virus particle depends on the composition of the coat protein of the virus. Von Sengbusch (56) and van Regenmortel (68) compared the serological behavior of tobacco mosaic virus (TMV) strains which differed in one or more amino acids in the protein coat polypeptide. They found that only eight of the twenty amino-acid replacements resulted in a change in the antigenic properties of the virus. The antigenic properties of a virus are a function of the coat protein, but not all coat protein changes result in a change in serological identity. Serologically indistinguishable viruses do not necessarily produce the same symptoms on a selected herbaceous host. Gooding (16) classified one hundred naturally-occurring isolates of TRSV from tobacco as belonging to one of four serological groups. He found no correlation between isolates within a strain and the symptoms they caused in tobacco. Some isolates of a serological strain caused mild symptoms; other serologically indistinguishable isolates caused severe symptoms. Even though it was possible to determine the serological relationships of these isolates it was not possible to use serological grouping to predict the effect of the isolate on selected host plants. Either serology was not capable of detecting biologically significant differences in the coat protein, or the differences in the symptoms induced in different host plants were due to some other change in the virus. Electrophoretic Mobility The electrophoretic mobility of a virus particle is the rate of migration of that virus particle in an electric field at a particular pH. This property of a virus depends on the amino acid composition of the coat protein, and also on the three-dimensional structure of the virus particle as it affects the availability of ionizing groups of the amino acids. The virus particles of different serological strains of a virus sometimes differ in their electro- phoretic mobilities, a property which reflects the surface charge density of the virus particle (13). Extensive studies done with chemically-induced mutants of TMV in which the amino acid composition of the coat protein of each mutant was known, have shown that changes in electrophoretic mobility are always a function of changes in the amino acid composition of the coat protein (56). However, the converse was not true. Many mutants which differed in their amino acid composition, did not vary in their electrophoretic mobilities. In order to produce a change in electrophoretic mobility, an amino acid substitution must result in a net difference in charge of the virus particle surface. This may occur either by changing the conformation of the protein coat, and thereby exposing or masking charged groups, or by directly substituting amino acids with a different charge. Electrophoretically variant strains of brome mosaic virus (BMV) have been isolated and used in combination with wild-type BMV for pseudorecombinant genetic analysis (28). Pseudorecombination is an experimental technique in which the separated segments of the multipartite virus genome, by themselves uninfectious, are brought together with complementary genome segments (from a different strain) to yield a progeny virus with a hybrid genome. BMV has a multi- partite genome consisting of four RNA pieces. Using separated RNA components of wild-type and variant strains of BMV in various combi- nations, Lane and Kaesberg (28) were able to demonstrate that the electrophoretic mobility of these strains was inherited through RNA-3 of the virus. RNA-3 of BMV has since been shown to contain the gene for coat protein (57). Therefore, in this case, as with the chemically induced, well characterized mutants of TMV, electro- phoretic mobility is thought to be solely a function of the coat protein of the virus, and not dependent on the nucleic acid which is found inside of the coat protein shell of the virus particle. Nematode Transmissibility Studies of serological strains of raspberry ringspot virus, tobacco rattle virus, and tomato black ring virus, have shown that nematode transmissibility is determined by the serological make-up of these viruses (22,23,21). This is not surprising considering that transmission by the nematode vector seems to involve the reversible association of virus particles with specific surfaces in the food canal of the nematode (64). Since plant viruses do not multiply in their nematode vectors (13), the interaction between TRSV and its nematode vector (Xiphinema americanum Cobb) is probably limited to the physical contact between the virions and portions of the vector, and is therefore presumed to be a function of the protein coat of the virus. Pseudorecombinant work with raspberry ringspot virus (a nepovirus like TRSV with a genome of two separate RNA molecules) has shown that nematode vector specificity is determined by the piece of the RNA genome that carries the gene for the virus coat protein (22). Similar pseudorecombinant work done with tomato black ring virus, another nepovirus, showed that transmissibility by the nematode vector was correlated with the antigenic specificity of the virus, and was determined by the RNA segment which contains the coat protein cistron (21). Thus, nematode transmissibility seems to be a characteristic of a virus which is determined by the properties of the protein coat of the virus particle. For viruses such as TRSV the coat protein probably reflects only about 3 to 14% of the viral genome coding capacity. Thus, current ideas on relationships among strains of TRSV, based on properties dependent on the coat protein (serological relation- ships, nematode transmissibility, electrophoretic mobility, amino acid composition of the coat protein) are based rather heavily on a relatively small segment of the viral genome. Biological properties such as symptom expression or host specificity, which involve an interaction between several of the virus-coded proteins and the host plant, can be expected to reflect differences between strains that no physical or chemical methods applied to the virus coat protein could detect. Following are several examples where host specificity and symptom expression have been shown to be independent of coat protein function. Host Range The host range of different strains of the same virus are usually quite similar. A strain of a virus which has acquired the ability to infect a new host usually retains the original host range as well. The host range of a virus strain has been reported in some cases to be independent of the coat protein properties of the virus strain (6,20,37). The host range of four strains of TMV was shown to be independent of the coat protein of these strains by the use of a technique called genome masking. Genome masking is an experi- mental technique in which the RNA and coat protein components of a virus are separated and then brought together with complementary components (from a different strain) to yield a reconstituted virus. When the RNA from one of four naturally-occurring host specific strains of TMV was reconstituted with the coat protein from any of the other three strains, the host specificity of the reconstituted virus particle was the same as the host specificity of the strain from which the RNA had been isolated (37). The result of this experiment suggested that it is not the coat protein of the infecting virus particle, but the genetic information con- tained in the RNA which determines host specificity. The determination of host range by a factor other than coat protein was also demonstrated by Dahl and Knight (6) who studied twelve nitrous acid-induced mutants of tomato atypical mosaic virus (a tomato mosaic strain of TMV). A host range mutant was found which could no longer infect tomato, even though the parent virus readily infected tomato. Amino acid analysis of the coat protein of this mutant demonstrated no change in the amino acid content of its coat protein as compared to the coat protein of the parent virus. Genome masking experiments using two strains of tomato atypi- cal mosaic virus, one of which was not capable of infecting tomato, showed that the coat protein of the reconstituted virus did not affect the ability of the virus to infect tomato. The determination of host range in this case was not a function of the coat protein of the infecting virus particle, but rather was determined by the nucleic acid of the infecting particle. Analysis of pseudorecombinants made with four naturally- occurring strains of raspberry ringspot virus strongly suggests that the virus coat protein, which is coded by the smaller of the 10 two viral segments, does not determine either the ability to invade Phaseolus vulgaris systemically or the ability to infect 'Lloyd George' raspberry (20). In addition, the virulence of the pseudorecombinants seemed to be determined mostly by the larger RNA segment which does not code for coat protein. The above examples show that host range differences among strains of a virus are, in some cases, not due to properties of the coat protein, but rather are dependent on some other function of the RNA content of the virion. Whether the RNA determines host range through effects on the replication and translation of the RNA, or through the proteins (other than coat protein) which it codes, remains to be determined. Symptom Expression The symptoms expressed by a plant infected with a particular virus strain are a readily detectable characteristic of that virus strain. Unique symptoms produced by a particular variant of a virus may be due to changes in the coat protein of the virus, but there are several examples in the literature in which coat protein is not the only determinant for differences in symptom expression among virus isolates. Tobacco rattle virus (TRV) has an RNA genome consisting of two RNA pieces distributed between short and long virus particles. The coat protein is specified by the short particle RNA (53). Ghabrial and Lister (12) separated the long and short virus par- ticles from two strains of TRV which had distinctive coat protein 11 composition and characteristic symptom expression in several herbaceous host plants. They mixed the long and short particles from the two strains of TRV to produce new reconstituted viruses. The symptoms produced by the reconstituted viruses clearly demon- strated that both the long and the short particles of TRV play a role in symptom expression. The coat protein gene of the short- particle RNA does not completely specify symptom expression in TRV. Some property of the long-particle RNA of TRV influences symptom expression in the host plant. Cowpea mosaic virus (CPMV) has an RNA genome consisting of 6 daltons and 2.3 x 106 two RNA species of molecular weight 1.5 x 10 daltons, and has a protein coat composed of two different-sized polypeptides. The available genetic evidence suggests that the coat proteins are coded by the small RNA of CPMV (38,43,72). Thongmeearkom and Goodman (66) characterized two naturally-occurring variants of CPMV as having different antigenic properties, and as inducing different symptoms in cowpea plants. Analysis of pseudo- recombinants derived from heterologous mixtures of the RNAs from these two variants agrees with other evidence that the antigenic specificity (determined by the coat protein) is a function of the small RNA. However in these same pseudorecombinant studies the authors determined that the symptoms expressed by cowpea plants were a function of the large RNA of the reconstituted virus. Similar results were seen in pseudorecombinant studies with four naturally-occurring variants of raspberry ringspot virus (RRSV) (20). The severity of symptoms induced on selected herbaceous 12 hosts by reconstituted viruses made from heterologous mixtures of the large and small RNAs from these four strains was shown to be a function of the large RNA. The large RNA of RRSV does not code for the coat protein of RRSV. Some coat protein polypeptides have functions in addition to their role of coat protein. The coat protein of alfalfa mosaic virus (AMV) is thought to be necessary for translation of RNA-4 from AMV (69). A function of the coat protein in post-translational cleavage of large primary translation products is also possible. Such a function has been demonstrated for the coat protein of RNA phages (69). Even though the coat protein of a virus may have multiple functions, the symptoms that a virus induces in a host are not always a function of coat protein. Other genetic determinants of the virus are responsible for the symptoms induced in the host plant. What types of viral-specific proteins (other than coat protein) might be synthesized which could account for these effects on the host? Additional Proteins Found in the Virus Particle It has been shown that a polypeptide of about 4,000 daltons molecular weight is attached to the 5' end of the two RNA pieces of the TRSV genome (35). Similar genome proteins have been found on the RNA pieces of the cowpea mosaic (CPMV) genome (62), and the RNAs of several animal viruses (24,30). All of the genome proteins 13 reported thus far in the literature have a molecular weight of 4,000 to 10,000 daltons (14). The precise function of these genome proteins is not known, but infectivity of TRSV-RNA is abolished when the genome proteins are removed by proteases (35). In contrast, the genome protein of CPMV-RNA is not necessary for the infectivity of the RNA (62). Furthermore, this genome protein has no effect on the jg_yitrg translational efficiency of the RNA in the wheat embryo system. The molecular weight of the protein products of in_vitrg translation of CPMV-RNA were not affected by the removal of the genome protein (62). Because the genome protein of TRSV is necessary for infec- tivity of the RNA, it would be of interest to know the effects of the TRSV genome proteins on translation. 'This has not been reported. The lack of a function of genome proteins in the translation of CPMV suggests a replication or virus assembly function, but no evidence to support these possibilities is available. The genome proteins of CPMV and poliovirus have been shown to be coded by the viral genome (7,14). Differences in the molecular weights of the genome proteins of two strains of CPMV have been demonstrated (7), but the contribution of the genome proteins to differences in the CPMV strains has not been established. Viral-Specific Proteins Found in Infected Tissue In contrast to the coat protein and genome proteins which are structurally part of the viral particle, there are viral-specific proteins which are produced during viral reproduction in the host, 14 but which are not incorporated into the virus particle. These proteins are made, jn_yiyg, in much smaller amounts than the coat proteins, and are difficult to detect in the presence of host pro- teins. The existence of several proteins which are coded by the viral genome, but which are not present in the virus particle, has been demonstrated in infected host tissue, but the function of these proteins has not been firmly established. Synchronously infected protoplasts have been used in the study of the appearance of virus-specific proteins during BMV infection. Sakai et a1. (52) demonstrated the synthesis of four virus-specific proteins, one of which was the coat protein, in BMV-infected protoplasts. BMV has a multipartite genome consisting of four RNA molecules separately encapsidated by a simple coat pro- tein consisting of one type of polypeptide. The size of the four proteins synthesized in infected protoplasts accounted for virtually all of the RNA coding capacity of BMV. One of the proteins was identified as the coat protein of BMV, but the function of the other three proteins is not known. When the four RNAs of BMV were trans- lated in the wheat embryo cell-free translating system, four proteins were produced of approximately the same size as those produced jg_yi!g in infected protoplasts (57). It has been proposed that 4 daltons one of these four proteins of molecular weight 3.5 x 10 functions as a subunit of the viral replicase (18), but there is little evidence to substantiate this. Pseudorecombinant studies with the genomes of nitrous acid-induced mutants of BMV have been useful in associating specific 15 virus properties with each of the four RNA species of BMV (2). Since the gene product of each of these RNAs has been identified by jg_yitrg_translation studies it is possible to assign specific virus properties to these gene products. Many of the nitrous acid- induced mutant properties were associated with changes in the coat protein gene, but changes in local and primary lesion appearances were due to mutations in two RNA species which did not code for coat protein. Unfortunately, the function of the gene products of these two RNAs is not yet known. The evidence from in_vivo and jn_vitro studies with BMV indicates that three proteins, in addition to coat protein, are necessary for successful infection with BMV. Variation in any one of these four viral-specific proteins might account for strain variation in BMV. Structural variation in the RNA that resulted in changes in the efficiency of translation, replication, or coating of the virus particle could also account for the variation in these nitrous acid-induced mutants of BMV. The following discussion con- siders evidence for the production and function of several kinds of virus-coded proteins which are necessary in plant virus infection. Viral-Specific Replicases Plant viruses whose single-stranded RNAs can serve directly as messenger RNA are thought to initiate viral replication by trans- lation of viral RNA to produce proteins which are required for replication of the viral RNA. Ribophages, bacteriophages in which single-stranded RNA serves as a messenger RNA, code for a polypeptide 16 which replaces one of the subunits of the bacterial host replicase, thus making the replicase specific for the ribophage RNA (63). There is not strong evidence for the synthesis of a plant virus-coded replicase, or of a replicase subunit. Varying degrees of viral- specific replicase activity have been demonstrated in infected plant tissue. It has been possible to extract, partially purify, and demonstrate jn_yjtrg_activity of an RNA replicase in plants infected with TMV (73), BMV (17), turnip yellow mosaic virus (39), cucumber mosaic virus (34), TRSV (41), and other plant viruses. Estimates of the molecular weights of these viral replicases range from 130,000 to 150,000 daltons (41). Most of these polymerases from infected plants exhibited little or no template specificity for the plant viral RNA. However, Sela and Haushner (55) purified TMV replicase from TMV-infected plants and showed that TMV-RNA was the only RNA template that was capable of stimulating incorporation into acid- insoluble product. Contrary to the evidence of Sela and Hauschner, there is recent evidence that indicates that RNA replicase activity is already present in healthy plant tissue, and that virus infection of a plant merely increases the activity of normal plant cell RNA- dependent RNA replicase (4). Even if RNA replicase activity is present in healthy plant tissue, some protein coded by the viral genome might function to increase the activity of the plant RNA replicase. Although it has been shown that replicases can selec- tively transcribe viral RNA, the determinants for this specificity are unknown at this time. 17 Viral-Specific Proteases A virus-specific protein of cowpea mosaic virus (CPMV) has been shown to function as a proteolytic enzyme in post-translational cleavage of other proteins coded by CPMV-RNA (43). The CPMV protease is coded by the large RNA of CPMV, and cleaves a precursor protein product of the small RNA of CPMV. Pelham (43) was not able to determine which of the two protein products (32,000 or 170,000 dalton) of the large RNA was responsible for the post-translational protease activity. This is the first report to date of protein- precursor processing in a plant virus system, but there are many examples of post-translational cleavage in entero- and picorna- viruses in animals (42,29). Although post-translational cleavage has not been demonstrated in TRSV, it is interesting to note the similarities between TRSV, CPMV, and the animal viruses that have been shown to have post-translational cleavage. Not only do these viruses all possess genome-linked proteins, but they all have been shown to have a 3'-terminal polyadenylate segment on their RNA genomes (35). These parallels suggest an underlying similarity in the way the genomes of these viruses function, and also suggest that TRSV may depend on viral-specific proteases for its replication. Viral-Specific Proteins in Plant Virus Inclusion Bodies Tubular inclusion bodies containing single rows of virus particles have been observed by electron microsc0py in the cytoplasm of undifferentiated cells infected with TRSV (70). Although similar in form to the inclusion bodies induced by turnip yellow mosaic 18 virus, optical diffraction experiments showed that the surface structure of the inclusions induced by these two viruses are not the same. This suggests that the tubular inclusion bodies induced by a virus have a specific structure and are characteristic of the virus or virus type which induces it. It is not known whether the proteins of the tubular inclusions of TRSV are virus-specific proteins. The function of the tubular inclusion bodies in TRSV infection is not known, but various researchers have speculated that they might be involved in cell-to-cell movement of the virus, or in virus replication (70,49). Because so little is known about the nature of the tubular inclusion bodies induced by TRSV infection, it is of interest to examine the well-characterized proteins of the pinwheel inclusions present in plant cells infected with viruses belonging to the potato virus Y grouping. Dougherty and Hiebert (9) have identified several translation products of tobacco etch virus (TEV) RNA as being serologically identical to the structural proteins of the pinwheel inclusions of TEV. Purcifull et a1. (44) partially puri- fied the inclusion bodies from plants infected with five poty- viruses, and made antisera to these partially purified inclusion bodies. These authors demonstrated that the inclusion bodies from different potyviruses were antigenically distinct, and suggested that the antigenic variation of the pinwheel virus proteins might be a good characteristic to use in the classification of potyviruses. Variation among many isolates of a plant virus is a function of differences in viral-coded proteins of these isolates. In the 19 past, comparisons and characterizations of viral isolates have been made on the basis of one viral-coded product, namely the coat protein. This is explained by the relative ease of purifying coat protein compared to other viral-coded proteins. Though often easily made, these comparisons have been shown to be of limited use in predicting the biological behavior of virus isolates (16), probably because the comparisons are based on only one of the proteins coded by the viral genome. This dissertation reports a method for comparing virus isolates which involves a comparison of several of the viral-coded proteins. The wheat embryo cell-free protein synthetic system was used to synthesize the proteins coded by TRSV-RNA. Using RNA from different isolates of TRSV as messenger RNA in the wheat embryo cell-free protein synthetic system, viral-specific protein products were synthesized and compared. This method allows the investigator the ability to compare directly all of the viral-specific proteins produced by a given viral isolate in the wheat embryo cell-free protein synthetic system. Seven isolates of TRSV were chosen for study. They were characterized by the symptoms they induced in selected herbaceous plants, and by their coat protein antigenic properties. 0f the seven isolates,three were markedly different in the reactions they induced in the herbaceous host plants and were selected for further comparisons. The RNA from these three isolates was extracted and used to direct protein synthesis in a wheat embryo cell-free protein synthetic system. The in_vitro protein products coded by the RNA 20 of the three isolates of TRSV were analyzed by SDS-polyacrylamide gel electrophoresis in order to detect differences in the molecular weights of the jn_vitro products. MATERIALS AND METHODS Virus Isolates The following isolates of tobacco ringspot virus (TRSV) were used in this study: Cherry isolate from Dr. Richard Stace-Smith (Agriculture Canada Research Station, Vancouver, BC V6T1X2), origi- nally isolated from sweet cherry (Prunus avium L.) (60); Tobacco isolate, Serogroup 38, from Dr. Guy Gooding, Jr. (North Carolina State University, Raleigh, NC 27607), originally isolated from flue-cured tobacco (Nicotiana tabaccum L.) (16); Watermelon (Citrullus vulgaris Schrad.) isolate from Dr. Donald deZeeuw (Department of Botany and Plant Pathology, Michigan State University, East Lansing, MI 48824); Soybean (Glycine max (L.) Merr) isolate, obtained from cotyledons of infected soybean seeds received from Dr. Richard Lister (Purdue University, West Lafayette, IN 47907); 'Riesling' grape isolate from Dr. J. K. Uyemoto (Kansas State Uni- versity, Manhattan, KS 66506); 'Jersey' blueberry (Vaccinium corymbosum L.) isolate from Dr. Donald Ramsdell (Department of Botany and Plant Pathology, Michigan State University, East Lansing, MI 48824) (45); 'Stanley' blueberry isolate obtained from 'Stanley' highbush blueberry, by triturating frozen infected blossoms in 2.0% nicotine in water, and inoculating it to Cucumis sativus 'National Pickling' seedlings. 21 22 All isolates were increased in Chenopodium quinoa Willd. Single lesions were excised and inoculated back to C, quinoa, Two such successive single lesion transfers insured that a virus isolate was not contaminated by other viruses, or by strains of the same. virus. Since the soybean isolate did not form local lesions on Q, quinoa, Cucumis sativus 'National Pickling' was used for making single lesion transfers of this isolate. The infected plants obtained from this first inoculation to a large number of plants were frozen at -20°C, and aliquots of this frozen tissue were used as inoculum for production of virus for further experimentation. Using this‘approach, the virus preparations used in the different experiments resembled the original isolate as much as possible. Symptom Expression on Herbaceous Indicator Plants The following herbaceous plant species were inoculated separately with the seven isolates of TRSV: Chenopodium quinoa Willd., Chenopodium amaranticolor Coste & Reyn., Cucumis sativus L. 'National Pickling,‘ Nicotiana tabaccum L. 'Havana 423,‘ Vigna unguiculata (Cowpea SR) (8), Phaseolus vulgaris 'Prince,‘ Nicotiana clevelandii x Nicotiana glutinosa (Christie's hybrid), and Glycine max_(L.) Merr. 'Hark.‘ Young, vigorously growing plants were dusted with 320-mesh carborundum and rub-inoculated with infectious sap obtained by triturating infected cucumber cotyledons in 0.01 M sodium phosphate buffer, pH 7.0. Test plants were grown in the greenhouse under fluorescent lights for a 14-hour day length at a temperature between 18°C and 30°C depending upon the season. 23 Sap from plants which did not show symptoms was inoculated to g, quinoa or tested serologically to detect symptomless infection. Thermal Inactivation Tests Crude extracts of virus were made by triturating infected cucumber cotyledons in glass-distilled water, and then straining the sap through cheesecloth. These crude extracts were placed in thin- walled glass tubes and heated for 10 minutes in a waterbath of the desired temperature and then cooled on ice. After heat treatment the crude extracts were tested for infectivity by rub-inoculation to C, guingathich had been dusted with 320-mesh carborundum. Plants which did not develop symptoms were tested serologically for asymptomatic infection. Virus Purification Two different methods were used to obtain the purified virus used in various experiments. The first method was adapted from Stace—Smith et a1. (61). Cucumber seedlings which were infected with TRSV were homogenized in a buffer of 0.05 M boric acid (containing 0.1% thioglycollate and 0.1% diethyldithiocarbamate, adjusted to pH 6.8 with NaOH) in a proportion of l kilogram of tissue to 1 liter of buffer. After the sap was squeezed through a double layer of cheesecloth, it was frozen at -20°C and thawed slowly at 4°C. After centrifugation at 10,000xg for 30 minutes, the super- natant was dialyzed for 16-24 hours against 5 volumes of ammonium .sulfate, 200 g/l. The dialyzate was centrifuged at 10,000xg for 30 minutes, and the virus was then pelleted out of the supernatant 24 by centrifugation in a Beckman No. 30 rotor at 28,000 rpm at 4°C for 2 hours. The virus-containing pellets were suspended overnight at 4°C in a buffer of 0.01 M Tris, 0.1 N NaCl, pH 7.0. The resulting virus suspension was clarified in a Beckman No. 40 rotor at 15,000 rpm for 10 minutes. The supernatant was then centrifuged in a Beckman No. 40 rotor at 38,000 rpm for 90 minutes, and pellets were resuspended as before. This virus preparation was then layered onto a 0-30% linear-log sucrose gradient (3) in 0.01 M sodium phosphate buffer at pH 7.0 and centrifuged in a Beckman SW 41 rotor at 38,000 rpm for 90 minutes at 4°C. The sucrose density gradients were frac- tionated from the top of the gradient tube, and scanned at 254 nm using an ISCO density-gradient fractionator (Instrument Specialties Co., Inc., Lincoln, NE 68505). 0f the three peaks that resulted, designated top (T), middle (M), and bottom (8), the M and B peaks were collected, diluted 1:3 (v/v) with 0.01 M sodium phosphate buffer, pH 7.0, and pelleted by centrifugation for 5 hours in a Beckman No. 30 rotor at 28,000 rpm at 4°C. The virus-containing pellets were resuspended in 0.01 M sodium phosphate, pH 7.0 buffer and stored at 4°C until further use. The second purification method was adapted from the method of Rezaian and Francki (48). Infected 'National Pickling' cucumber seedlings were homogenized with 1 ml of 0.1 M sodium phosphate buffer, pH 7.0, and 1 ml of chloroform per gram of tissue. The homogenate was centrifuged at 10,000xg for 30 minutes and the upper aqueous buffer phase was collected. Polyethylene glycol, molecular weight 6000 (Fischer Scientific Co., Carbowax 6000), and NaCl were 25 added to the buffer phase to a final concentration of 6% (w/v) and 0.3 M, respectively, and the mixture was stirred and left at 0°C for 30 minutes. The precipitate was pelleted by centrifugation at 10,000xg for 10 minutes and resuspended overnight at 4°C in 0.1 M phosphate buffer, pH 7.0 with 0.01 M EDTA. The virus preparation was purified further by alternate high- and low-speed centrifugation, and sucrose density-gradient fractionation as described above for the first method of virus purification. Serology Antisera Sources Antisera to the TRSV isolates used in this study were either gifts from other investigators or were produced in our laboratory. Antisera to the cherry, soybean, 'Riesling' grape, and 'Jersey' blue- berry isolates were kindly supplied by the researchers who supplied these virus isolates. Antisera to the tobacco, watermelon, and 'Stanley' blueberry isolates were made in the following manner. The antigen used for injection consisted of virus which had been purified through sucrose density gradients as described in virus purification methods. Female New Zealand white rabbits were injected intra- muscularly at weekly intervals with one volume (1-2 mls) of purified virus (1-3 mg/injection) emulsified with an equal volume of Freund's adjuvant (Difco Products Co., Detroit, MI 48232). For the first injection Freund's complete adjuvant was used; all subsequent injections were done using Freund's incomplete adjuvant. Serum was obtained 10 days after the second injection, or if the antibody 26 titer was not high enough at this time, serum was taken 10 days after the final injection. All antisera were titered against their homologous virus isolates using the double diffusion test in agar plates (1). The gel consisted of 0.8% agarose (Sigma Type I Agarose, Sigma Chemical Co., St. Louis, MO 63178), 0.85% NaCl, and 0.15% sodium azide. All antisera were stored either 1yophilized or as whole serum at -20°C. Spur Formation Tests One of the criteria used to identify serological strains was spur formation tests (40). Spur formation tests are carried out in an agarose gel. Wells are cut into the gel and filled with the reactants. The antigens and antibodies diffuse into the gel from the wells, and form a precipitin line where they meet in optimal proportions. Two antigen (virus) preparations are paired and tested against an antiserum made to one of the antigens. The three reactants are put in wells arranged at the corners of an equilateral triangle. Two reactions are possible: (a) when the antigens are identical, the precipitin lines formed between each antigen and the antiserum will join smoothly (Figure 6), and (b) when the antigens have some determinants in common the precipitin line formed between each antigen and the antiserum will join, but the antibodies specific to the homologous antigen will diffuse through the heterologous precipitin band to form a spur (Figure 7). The agarose medium used in the spur formation tests was the same as that used to determine antisera titers (see section on 27 antisera production). A six-hole gel pattern was cut by using a Grafar gel cutter (Grafar Co., Detroit, MI 48238). The wells were 3.5 mm in diameter and spaced 10 mm apart center to center. In these studies antigen-containing wells surrounded a central antiserum-containing well. The virus used as the antigen in these tests was prepared from infected 'National Pickling' cucumber seed- lings by homogenizing seedlings in chloroform (5 m1 chloroform/g tissue) containing 0.01 g sodium ascorbate/g of tissue, and breaking the resulting emulsion by centrifugation at 10,000 x g for 30 minutes. The virus-containing supernatant was made 0.1% (w/v) with sodium azide and used directly as test antigen. This crude virus prepara- tion contained plant host comtaminants. To insure that these con- taminants were not responsible for the test results, some of the results were verified using purified virus preparations as the antigen. The various antisera were diluted with a 0.85% (w/v) NaCl solution to a concentration which produced a sharp precipitin band when reacted with the various test antigen preparations in agar double gel-diffusion tests. A dilution of between 1:8 and 1:32 was used for all spur formation tests; at these dilutions there was no reaction to healthy plant sap with any of the antisera. Each antiserum was tested for spur formation between its homologous antigen and all other heterologous virus antigens. All tests were repeated at least twice. 28 Cross-Absorption Tests Reciprocal absorption tests were used in addition to spur formation tests as a criterion in establishing serological differ- ences among the isolates of TRSV. In this test an isolate-specific antiserum was reacted with heterologous virus, and all antibodies which cross-reacted with the heterologous virus were removed. The titer of residual antibodies to homologous virus in the cross- absorbed antiserum is used to quantify the serological differences among the virus isolates (33). Most antisera were fully absorbed with the addition of 16 to 32 parts of heterologous virus prepara- tions (the antigen was prepared as for spur formation tests) to one part of undiluted antiserum. Absorption was carried out by mixing the antiserum with the absorbing virus preparation, incubating for 3 to 4 hours at 37°C and then overnight at 4°C. The antibody- antigen complexes were removed by centrifugation in a Beckman No. 40 rotor at 15,000 rpm for 10 minutes at 4°C. To determine if the cross—absorption was complete each absorbed antiserum was tested for residual antibodies to the absorbing virus in double gel-diffusion tests. If absorption was not complete the procedure was repeated using a higher proportion of absorbing antigen to antiserum. Cross-absorbed antisera were tested for residual antibodies to homologous virus using double gel-diffusion tests. Results were reported as the reciprocal of the titer of absorbed antiserum to homologous virus over the reciprocal of the titer of unabsorbed antiserum to homologous virus. 29 Electrophoretic Mobility of Whole Virus Particles The electrophoretic mobility of whole virus particles was determined by the method of Tremaine & Wright (67). Electrophoresis was carried out in a slab gel consisting of 0.75% agarose (Sigma Type I Agarose) in 0.02 M Tris, 0.02 M sodium dibasic phosphate, adjusted to the desired pH with citric acid. Gels were cast by pouring 5 m1 of molten agarose solution onto a plexiglass plate (2.5 cm x 13.5 cm x 0.5 cm) and allowing the gels to cool at room temperature before cutting 1.5 mm diameter wells for the samples. Three equidistant wells were cut in the center of the gel on a line running the width of the gel. Virus samples consisted of approximately 15 ug of purified virus in 0.005 M sodium phosphate buffer pH 7.0. Electrophoresis was carried out in a Gelman Deluxe Electrophoretic Chamber (Gelman Instr. Co., Ann Arbor, MI 48106) using a BioRad Model 400 Power Pack (BioRad Laboratories, Richmond, CA 94804) at 150 volts, 3.5 to 7 milliamps, for 5 hours. Electrophoresis tray buffer consisted of 0.02 M Tris, 0.02 M sodium dibasic phosphate, adjusted to the desired pH with citric acid. Following electrophoresis the gels *were stained for 2 hours in a solution of 5% acetic acid, 5% glycerol, and 0.0125% Coomasie blue. Gels were destained in a solution of 5% acetic acid and 10% glycerol until background was reduced to a minimum. 30 RNA Extraction Ribonuclease degradation of RNA was kept to a minimum by heating all glassware to 225°C prior to use. If "baking" was not possible, glassware was soaked in a dichromate solution or in 0.1 N NaOH before use. Phenol and ethanol were redistilled, and stored at -20°C prior to use. RNA was extracted from TRSV using a modification of the method developed by Dr. James Asher, Jr. (Department of Zoology, Michigan State University, East Lansing, MI 48824). The virus preparations used as sources of viral RNA were purified through 2 cycles of sucrose density gradients (as described under virus purification). Only the middle and bottom components which contain RNA were used in the RNA extraction procedure. Virus suspensions (about 1 mg/ml) in 0.01 M sodium phosphate buffer, pH 7.0 were thoroughly mixed with 2 volumes of 5% SDS, and 2 volumes of 80% phenol (v/v with distilled water). This solution was heated, with stirring, at 65°C for 5 minutes, and then cooled on ice. This high- low temperature treatment breaks up any hybrids in the RNA molecules and forces any DNA into the phenol phase. Chloroform (2 volumes) was added, and the solution was stirred vigorously for 5 minutes at room temperature. The addition of chloroform stabilizes the inter- face between the two layers, and also helps remove the RNA from the protein precipitate. The phases were separated by centrifugation in an IEC (International Equipment Co., Needham Hts., MA 02194) No. 870 rotor at 10,000 rpm, 0°C, for 15 minutes. The phenol phase was extracted with a mixture of 1.5 volumes of EDTA-acetate buffer 31 and 0.5 volumes of 5% $05. The combined aqueous phases were extracted by mixing well with 1 volume of 80% phenol, and then adding one volume of chloroform with thorough mixing. The phases were separated by centrifugation (as above). This extraction procedure of the aqueous phase was repeated three times to remove any residual protein in the aqueous phase. Nucleic acids were precipitated from the aqueous phase overnight at -20°C after the addition of 1/15 volume of 3 M sodium acetate pH 6.0 and 2.5 volumes 95% ethanol. After centrifugation at 10,000 rpm in the IEC No. 870 rotor for 15 minutes at -20°C, the pellet was washed with 10 m1 of 0.1 M sodium acetate in 66% ethanol to remove any residual phenol and other insoluble material. The washed pellet was air dried to remove residual ethanol. The nucleic acid precipitate was then washed thoroughly 3 times with a solution of 10 mls of 3 M sodium acetate, pH 6.0 and 0.25 ml 0.2 M EDTA. This washing procedure removes DNA, small RNA molecules, and polysaccharides from the pellet. The remaining pellet was solubilized in 12.5 ml of 0.1 M sodium acetate, 5 mM EDTA at pH 6.0. The nucleic acids were then precipitated by the addition of 2.5 volumes of 95% ethanol at -20°C overnight. After centrifugation the pellet was air dried, solubil- ized in glass-distilled water, and stored at -80°C until use. RNA concentrations were determined spectrophotometrically assuming 0.1% 260nm = 22- E 32 Gel Electrophoresis of Purified TRSV RNA RNA preparations were analyzed for purity and integrity on agarose-formaldehyde denaturing gels according to the method described by Lehrach et a1. (31). Ribosomal RNA from chick embryo muscle (a gift from Dr. James Asher, Jr., Department of Zoology, Michigan State University, East Lansing, MI 48824) and brome mosaic virus (BMV) RNA were used as molecular weight markers for RNA molecular weight deter- minations. Gels were cast in quartz tubes, and pre-electrophoresced at 4°C at 2 milliamps/gel for 60 minutes. Electrophoresis was carried out at 4°C at 2 milliamps/gel for 3 to 4 hours, after which gels were immediately scanned at 254 nm using an ISCO (Instrument Specialties Co., Lincoln, NE 68505) Model 1310 gel scanner. Separation of TRSV-RNA on Sucrose Density Gradients The small RNA of TRSV (RNA-2) was separated from total TRSV-RNA by sedimentation in 0 to 32.5% linear-log sucrose density gradients (3) made up in 0.15 M sodium chloride and 0.015 M sodium citrate buffer pH 7.0 with ribonuclease—free sucrose (Schwarz/Mann, Orangeburg, NY 10962). Prior to separation on sucrose gradients the RNA samples were denatured for 5 minutes at 65°C in 0.075 M sodium chloride, 0.0075 M sodium citrate, pH 7.0. Unfractionated, denatured TRSV-RNA (200 ug/Beckman SW 41 rotor tube) was separated by centrifugation for 6 hours at 38,000 rpm, 4°C, and fractionated using an ISCO density fractionator. The peak—containing fractions were pooled, alcohol precipitated, and fractionated again on 33 sucrose density gradients. The resulting RNA preparations were analyzed for completeness of separation using agarose-formaldehyde gel electrophoresis as described in the foregoing section on RNA extraction. Infectivity of Purified TRSV-RNA The infectivity of various RNA preparations was tested by rub-inoculation onto primary leaves of Vigna unguiculata (Cowpea 'SR') which had been previously dusted with 320-mesh carborundum. RNA preparations were diluted in 0.1 M sodium phosphate buffer, pH 7.5, and inoculation was accomplished by putting drops of the RNA solution onto the leaves, and then rubbing gently with a spatula. Test plants were grown in the greenhouse under fluorescent lights for a 14-hour day length at a temperature between 18°C and 30°C depending on the season. Lesions were counted 4 to 8 days following inoculation, and results were calculated as the mean number of lesions on ten or more cowpea leaves. Detection of Poly (A) Tracts in TRSV RNA The presence of poly (A) tracts on the 3' ends of eucaryotic messenger RNAs and some plant and viral RNAs has been a useful char- acteristic for separating these RNAs from poly (A) negative, non- messenger RNAs. When this research was initiated it was not known *whether the RNA of TRSV contained poly (A) tracts. Since the presence of poly (A) tracts on TRSV-RNA had potential for use in 34 separating TRSV-RNA from contaminating RNAs, the poly (A) containing properties of TRSV were investigated. 01igo(dT)-cellulose (Type 2 from Collaborative Research Inc., Waltham, MA 02154) was used to detect the presence of poly (A) tracts in the RNA of TRSV. Unfractionated TRSV-RNA was made 0.4% in SDS, heated to 65°C for 2 minutes, and then quickly chilled in an ice bath. This RNA solution was diluted with 1 volume of the following solution: 0.02 M Tris-HCl pH 7.4, 2 mM EDTA, and 0.24 M NaCl. This solution of RNA in 'high salt' was then passed over an oligo(dT)-cellulose column which had been equilibrated in 0.01 M Tris-HCl, pH 7.4, 1 mM EDTA, 0.2% SDS, and 0.12 M NaCl. Poly (A) RNA binds to the oligo(dT)-cellulose under these 'high salt' condi- tions, but RNA which does not contain poly (A) tracts, or which contains only short tracts of poly (A), does not bind. The sample was washed onto the column and the column was eluted with additional 'high salt' buffer until the eluate fractions did not contain any RNA as determined by spectrophotometric analysis at 260 nm. The column was then eluted with 'low salt' buffer (0.01 M Tris-HCl, pH 7.4, 1 mM EDTA, 0.2% SDS), and the eluate was collected in 1 ml fractions. The fractions from the column which contained RNA, as determined by spectrophotometric analysis, were pooled and precipi- tated with 1/15 volume of 3 M sodium acetate, pH 6.0, and 2.5 volumes of 95% ethanol at -20°C overnight. After centrifugation the pellet was air dried, dissolved in 100 pl of distilled water, and stored at -80°C until use. All RNA-containing fractions were analyzed on agarose-formaldehyde gels as described above. 35 Translation of TRSV-RNA in the Wheat Embryo Cell-Free System Wheat embryos were extracted from wheat seeds (Triticum aestivum L. 'Ionia'), and the dialyzed wheat embryo extract was prepared according to the method of Marcus et a1. (32). RNA (1-2 pg) was added to 50 p1 of reaction mixture containing 20 pl of dialyzed wheat embryo extract, 45 mM potassium chloride, 2.7-3.2 mM magnesium acetate (Mg++ concentration optimum varied slightly with RNA prepara- tion), 0.04 mM GTP, 1.4 mM ATP, 3 mM dithiothreitol, 35 mM Tris- acetate (pH 8.0), 10 mM creatine phosphate, 2.7 pg creatine phos- 3 (3H-valine: specific activity phate kinase, 14C- or H—valine 1.2 Ci/mmole, c‘4 -valine: specific activity 246 mCi/mmole, New England Nuclear, Boston, MA 02118), a mixture of the remaining 19 amino acids to 0.034 mM, and 0.5 mM spermine tetrahydrochloride. Spermine tetrahydrochloride was added to prevent premature termina- tion of protein synthesis. After 45 minutes of incubation at 30°C, 1.4 pg of aurin tricarboxylic acid/50 pl of reaction mix was added, and the reaction was terminated after an additional 15 minutes of incubation. Aurin tricarboxylic acid was added to prevent any further initiation of protein synthesis. The reaction was terminated by spotting 50 pl of the reaction mix onto Whatman #1 filter paper discs and drying under a heat lamp. The synthetic ability of different reaction mixes was evaluated by measuring the amount of radioactively labeled, hot trichloroacetic acid-precipitable protein produced. After boiling the filter paper in 5% trichloroacetic acid (TCA), 10 mM valine 36 for 15 minutes the filter papers were rinsed 3 times with hot 5% TCA, 10 mM valine. The filter papers were then washed 2 times with 50% ether:50% ethanol, to remove residual TCA. Two final washes with 100% ether were done to remove the ethanol. After air drying, the filters were placed in counting vials with 0.8 ml Protosol mix (10 ml toluene: 9 ml Protosol (New England Nuclear, Boston, MA 02118): 1 ml distilled water), and heated at 60°C for 1.5 hours. When cool, 5 ml of counting fluid (1 liter toluene: 6.0 gm PPO: 0.15 gm dimethyl POPOP) was added and the vials were counted for radioactivity in a Searle Isocap/300 Model 6868 liquid scintillation counter. Experiments were carried out to determine the optimum concentrations of K+, Mg++, and TRSV-RNA necessary for maximum incorporation of radioactivity into TCA-precipitable counts by the wheat embryo system. Analysis of Translation Products by iSDS-Polyacrylamide Gel Electrophoresis Translation products were analyzed by SOS-polyacrylamide gel electrophoresis by the method of Fairbanks et a1. (10). Molec- ular weights of the labeled jg_yitgg products were calculated by assuming a linear relationship between log molecular weight and mobility using the following unlabeled protein markers: cytochrome c, myoglobin, pepsin, ovalbumin, TRSV coat protein, bovine serum albumin, and phosphorylase a (molecular weights of 12,400, 17,200, 34,600, 43,000, 54,000, 68,000 and 94,000 daltons, respectively). 37 TRSV coat protein was prepared from whole virus by heating purified virus in dissociation buffer (0.05% B-mercaptoethanol, 0.05% SDS, 2 M urea in 0.05 M sodium phosphate buffer, pH 7.2) at 100°C for 90 seconds. The dissociated virus was then dialyzed extensively against the electrophoresis chamber buffer (0.04 M Tris, 0.02 M sodium acetate, 0.002 M EDTA pH 7.4) to which had been added 1% B-mercaptoethanol. The translation products were characterized by electro- phoresis on 5.6% polyacrylamide- 0.21% bisacrylamide cylindrical gels (7.5 cm long) containing 1% SDS. The products (75 p1 3H-valine mix) were mixed with 9 pl of 20% SDS, 2 pl of B-mercaptoethanol, and 89 p1 distilled water. Standard protein solutions, except for the TRSV coat protein, were prepared in a similar manner using 20 pl of the protein solution (1 mg/ml), 9 p1 of 20% SDS, 2 pl of B-mercaptoethanol, and 135 pl distilled water. All samples for electrophoresis were denatured by heating to 100°C for 3 minutes in a boiling water bath. After denaturation, 50 pl aliquots of the samples were mixed with 2 pl of pyronin y (0.025 g/50 ml distilled water) and one drop of glycerol, and separated by electrophoresis at 8 milliamps per gel for approximately one hour. Following electrophoresis the gels were stained with Coomasie blue (1.25 g Coomasie blue in 454 m1 of 50% methanol and 46 ml glacial acetic acid) for 12-16 hours, and then electrophoretically destained at 8 milliamps per gel until background was reduced to a minimum. The gels containing the radioactively labeled protein products were sliced into l-mm segments, mixed with 0.8 ml 38 Protosol mix, and heated at 60°C for 1.5 hours. After the addition of 5 ml of counting fluid, the gel segments were counted for radio- activity in a Searle Isocap/300 Model 6868 liquid scintillation counter. Stained gels containing the molecular weight standards were scanned at 590 nm using a Gilford Model 2410-5 Linear Gel Transport System (Gilford Medical Instruments Co., Oberlin, OH 44074). The logarithm of the molecular weights of the protein standards were plotted against the relative mobility of these standards, and the resulting graph was used to estimate the molecular weights of the ig_yjt§g_protein products. Dual—Label Experiments to Detect Small Differences in Molecular Weights of ’Protein Products Produced by Different Isolates of TRSV Labeled protein products were made by translating the RNA from different isolates of TRSV in the wheat embryo system as described above. The protein products of some isolates were labeled with 14C-valine; protein products of other isolates were 3 labeled with H-valine. Samples for SOS-polyacrylamide gel electrophoresis were prepared by mixing the 3H- and 14C- labeled protein products from different isolates of TRSV, and then separated by electrophoresis as described earlier. The gels were sliced and 14 3 the radioactivity due to C and H was determined for each slice using a Searle Isocap/300 Model 6868 liquid scintillation counter. RESULTS Symptpm Expression on Herbaceous Indicator Plants Reactions of the various herbaceous indicator plants to infection by TRSV are given in Table 1. The virus isolates are listed in approximate order of virulence as judged by symptom severity on the herbaceous hosts. The symptoms expressed by a particular host, such as local lesion formation on inoculated leaves, were tabulated at the same time for all virus isolates. In some cases the difference in symptom expression between isolates was only a difference in the time of appearance or the severity of the symptoms produced. For instance, the local lesions produced by the soybean and 'Jersey' blueberry isolates on the inoculated cotyledons of Cucumis sativus L. 'National Pickling' are quite unique when the isolates are compared at 4 days post-inoculation (see Table l and Figure 1L. However, if one compares the local lesions of the soybean isolate at 8 days post-inoculation with those produced by the 'Jersey' blueberry isolate at 4 days post- inoculation, the difference between the isolates is less pronounced. Some examples of the differences in symptom expression on the various indicator plants are given in Figures 2, 3, and 4. The isolates of TRSV that were used in this study differed in their ability to infect various herbaceous hosts. For example, when the soybean isolate was inoculated to Nicotiana tabacum 'Burley' type, only 1/6 of the inoculated plants became infected. In contrast, the 39 hila‘ A, , _. , . .:‘— wad 4i) .w_44oe 044ocopgu u :4 4:4omu u o 4:044aELoL_cE use m:_4::um wcm_ u :4 mmuoamo:_c 0.4ocumc n «z 4m:o_mm_ u_uoco_zu u 4o 4m_mocooc .e:_m> u z> ”xoanm_u n ma wmcoemm_ o_4ocoo: u 42 4mm>om_ cm4a_:uoc_ - 44o .mm>am_ 36: co m_44oe .4_ :8 z> u;m__m 54?: 4Q uu:*um_uc4 mm>am_ :0: co 4: oEOm z4_3 to .44 co z> e4_: 48 uue_4m_ee_ .o4a_omp ouuo404 mm osmm .42 :44: mm>ow_ 38: .2) ;u_: 44 co 4U Sue,am_oe_ .mo>om_ ouow—ow_gu be mo .z> saw: 4— co 4: .uw4oguwc we 40: mo>am_ to: 4:4 .xwmcow. mo 62am .mmsomp muawpoepc4 we as .z> zu_3 4. CO 48 uue_um_e=_ .mw>mm_ rm: we m:__44os u_smam»m new a:_ luczum .44 :o 44 open .mm>om_ to: co mew—44oz owewumxw a oe_4e=4m .44 no 48 .mumpom_ ouuanou mo aEam .mo>ow_ so: no 42 a 48 J_Eoum»m ;4_: .mewueaam .44 so 48 .ouo_omw xgcoao:_n .xmmcma. ma asam .44 co 44 :_ m4mocuo: o: 4:4 .xmmcma. no 65cm .4u a 42 ;4_3 mw>om~ 30: we mcwu==4m wsmcuxw .44 co 4: .mm>ew_ .a:_=:o4 :o m_moco_go 42a__m 7: co meouasxm oz .mua_om_ accuses—n .xm—caem. mm oEem .x4 :4_3 mm>am_ pouwae .44 no 42 a 4g uc_oa:_a .24 cum: mm>om_ _co_aa .44 co mcowmmp oz .x4 ;4_: mo>uo_ .oo_aa .4. co 4: 4c_oacwa .mo .ouwaa o: 4:4 .mum—om_ accmno:_n .ammcma. we 65mm .mo _ao_ac .24 :44; mw>cmp .mo_aa .44 :o 40 ucm 42 .xemr. mme m:_uxwm m=>_4cm m_E:u:u .mcwucaum ecu c.4405 owemumxm 4;o_Pm .44 co «commmp o_4_mw> oz .o paaucu>o mmxou op we we poupaa .44 :o 42 .o —a:4co>m “mama o 4a mo Fuu_nu .44 :o 42 .o .m:u:m>m .mxcu c an on puu_ao .44 no 42 u=_oa:wa .o Penucm>u ”mama or an ma pou_qa .44 no 42 pp~5m .o poaucu>w mmxcu e an on _au.aa .44 co 4: .o pozucm>m umzou e 4e ma pou_ao .44 co 42 «men. apa>w4opmm ccmnxom xgcozu accwnm:_n .xopco4m. ouuono» oauca .uc_—mm_¢. co—msgauo: accmaw:_a .xmmcmw. uo_ou_4:ekmam .u mo:_:a e:_uoaocm;u .. .. 1:, ,1. 1 h. ..l 1.111111lI-111.n.lfl u h Ifliuhcbhw .114." u .115 . J .1 u 1 n 1.. . r1 kahufllulunnlx 430: a u 11 J H.1u.u|u.1.flfl1.E ".1" n1 ,.;. . .11fl11111. 4“"H111111I11 «.331611"... o4c_om4 macw> co wucaom . a. . ...II.IlIn1.1IBnd..1.4.lhlrIl-n:n|"uuflzu.4. ...”... ”at” .A>mm»4 macw> uoqmm:_c 0008404 we mm4a_om_ ce>mm cu mmcow4oamc «mo: mzomuoocmz--.4 m4m

« z> “xeono.e a mo "meepmep p.4eceee u 4: “mo>ee— ewes—:uee, u 4.e .mm>eo— so: we m_moceme wsom .44 no 44 epe_mw> oz .e4e—em_ ee—wscmue: we eEem .mm>eep so: Le memocee: .44 :e 40 wu¢e_uoec4 .emeoemec e_se4mzm e: .44 ce «2 e4epesouew creep oeom .me _au_aa .44 co 4: ezoee e;e_4 .mm>ee— rm: :4 memeceee .mo pee_ee .44 ee 4U ueewum_a .mo _ee_ee .mc_m> mee_e acmeeocom 44 go 44 .m4e_em_ accwzu we eEem .z4 meow .44 so 44 em; 4Lee aucpum_b —_oEm .44 ee 44 Lm_—esm .m4epem4 Acumeoe—e .xemcoe. we eEem .Le__esm 44 :e 44 4:4 .eeepem_ weecm .mevpmewz. me esem .14 .44 co mcwecon uue_um_ee¢ ‘4’; 44 ems gene emce4 .z495w.44 eo mcewme— em; seen .24 a mo .ee_ee .44 :e meeueme ceeca :44: mm em; seen emce4 e.ee_4emcc_ ewe—Eeuoszm .emscee_ee me>eo_ 3e: 4:4 .euepem_ weeem .ac_pme_z. we eEem .m4e_em_ meeem .me_—mmw¢. me eeem .me>eep to: co meeemoevc e_4eLe—;e .44 ee 44 oz .me>eo— 3e: ee 12 eea 42 eeeeee_a .44 ee 4: ue_oac4a .emeeemec e45e4mzm ee .44 ee 44 emeuw_o .mm>ew— 3m: :o 42 P—esm .44 co :2 exoca uee_umwec_ 1111 .mwewmnt111111 m_comm=> m:_ewmeza .mm. amazee cue—.4333... new; .:o_4eee:_ mm_ee4osxm .e4e_em_ eueeee4 we usem .omcoomec e_Ee4mzm e: .44 :0 «z exepm .emeeemec u_eeumzm e: .44 ee 44 aceeee .eeLe—eu eeh .zz eeoece mee44Le4 1m4e weep cup: 434 .Aomcma. me esem .me—eEm eee acme new. «2 4:4 ..zemcoe. me esom .mmeeemoc umeoemzm e: .Le4e3e_e :4 EE o.1~ .44 ce :2 xeepm eeoezem agngu xccweeope .zepeeum. oueeeeb mecca .oewpme_z. eepwsceue: Accweeepe .zwmcoe. ,mwmemflmxmmm .z x ew0c_4omm .m .mme nWw>ex. , Eeeeeeu eee_4eewz LII.» .1. 1.11 111.1111 F F N. «HM 11 1111.111 h..l1L1.. n14 ..»111..1DHA1..,IH..IL..n..l.u..lh...l:11.1 mace—e 4me: i h h .11..H1.. .Hrh IFL. n u. ...u 1.. . .1 "in 9 1.111111. 11111.1 ”grillifl.“ EH1! lflrlvfifihhxvr .1.) 1L. I e4e_em4 mocw> we meceem :1 01.111.11th 1...“.fila.x.ll.hnl .z.e=oo4--.4 44mwpeemew Lee eczemme ems“ use .me_ so empeee .mceueemeeeu emcwmme ozu we meueews op Lee emuem; wee: meeeepzuee geeseeee eepemmew emueceuwca Eecw Lopez eeppwem_e1mme~m cw mueecuxm mecw>e m\o M\N M\m M\m M\m M\m m\m m\m N\N eo_eezmea3 M\o M\o M\m M\m M\m m\m M\m m\m N\N seemee M\o M\m M\m m\m M\m m\m m\m M\m N\N aaaem .meepmeem. M\o M\m M\~ M\N M\m M\m m\m m\m N\N seeaemspa .Ameeaem. M\o M\m M\m M\m M\m m\m M\m M\m «\N seemnaapn .sameae. - m\o M\o M\_ M\N M\N M\N M\m N\N camasom 8-- M\o M\o M\m M\m m\m M\m M\m N\~ oeeaeoe m.~e oe m.em mm m.Nm om m.ee me o meaPOmH maew> AooV eceueceesmk cowee>wueee4 e.mmuepemw A>mmhv mec_> ueemmewg eeeeeeu we :e_ee>wuuec_ Peacock-1.~ m4mm cowpeELeL Leom eze o + o + o o o emmnsom + o o + o o o :ePeELmuez o + o . o o o oouaaoe + o o o o o o seemem=_n .smmeme. o + o + o o o seeeemapn .»m_eaem. o + o + o o o agate .mc__maam. o + 0 5+ e “1.0 o seemee cemezem cepmscmuez eoeeeeh zccmeee_m zccmemefim omega accwzo emummh muepemH .zemcee. .zepeeem. .mcwpmewx. >mmh meemepecmpwz mea_om4 >mme Lo azeameee< .memeu eewmeeePe-me m—eeee cw mcemwuee meemepegeumz ece meemepesez new: eeueemc ego: A>mzpv “commewe eeeeeep we wepeFemw eu exemwpee ems: emecee mmewp ewuweweege seem-1.m 44m

meemePeEez e4 Eecemwpee emecemeeee we Louwh 61 mec_> meemeFese; e4 Eecemwuee eeeLemee Le Leewh u cowpeecue omm\mmv emmup mmupv mmnp eeemee» sneezem accmeeepm emm\e4v emm P e_”4v e444 .smmeme. camnsom emm\ee emm 4 «one mmup camasom oeuaeoe zeceeeepm em~\e_v emmu_ an_v e4“_ .xmmeme. oeeaaoe zcceeeepm N_m\m~_ Npmu_ mNF _ mm”, sneezem .zemcme. zLLmemepm N_m\ee N_mu_ eenp «map oeuaaoe .xmmeme. eeeeem uoz mecw> mecw> cowpeeeme< opeFemH emeceme< ea memecw> eeeceme< meemePeEez m:_zeppem >mzh new: Ezeemwue< zewzz Leuwh ee Loewe ee Leuwh Eecem cewuepwo eeeceme< we eeweeecd Eecemwpc< 1_ee< eeeceme< Eecemwue< 515-111-1115 I III-II .A>mzhv mecw> peemmc_c eeeeeeu we meuepemw mecca Lee mummy cowuocomnm1mmOLU11.¢ m4m .uzmwgz a oewesmc» we eezues ea meweceuee .ze meewceeceee we euezemeze upmeewe Espeem z ~o.o1mwcu z No.o cw mmegeme &~.o er eoc_scmuooe m-op x mm.o- m-op x mN.o - m-o_ x 44.0 + seeaaazpm .Aememe. m-o_ x m_.o- m-o~ x o_.o + m-e_ x ep.o + camezom m-op x m_.o- m-o_ x o4.o + m-oe x eP.o + oceanoe o.m In o.m In 0.5 In mHMPOmH m=e_> Le oeceem Ap1eem F1:9, Nsev zuwpweez ewpmeezoecuue—m a.z>mme4 mecw> peemmcwc eeeeeep we meeepemw emcee Lem mecw> upon: me ape—wees ewuegezeecueepw11.m m4m~ O ... 1r" -1 2 C: I— ———> m y» N 1 J 1 1 10. 51‘. '0. “1 '1 O O O O O O (mu 1793) aouoqiosqv Depth (cm) 70 FIGURE 10.--Agarose-formaldehyde gel electrophoresis of total RNA from the soybean isolate of TRSV. Samples of RNA were separated by electrophoresis at 4°C at 2 milliamps/gel for 3 to 4 hours, after which the gels were scanned at 254 nm. The migration position of ribosomal RNAs from chick embryo muscle, which were used as molecular weight markers, is indicated by the arrows. 71 N f-CD Q. J o9 ‘; “"10 m .9" J r? m ‘: 9 '5 3 "° -N 1r c... 1- 4. «s 41 -. 0° 0 O O O 0 (ulu v93) aouoqiosqv Depth (cm) 72 blueberry, and soybean isolates of TRSV are shown in Figures 8, 9, and 10. Infectivity of Purified TRSV-RNA The retention of biological activity in the preparations of unfractionated TRSV-RNA was considered to be an important measure of the integrity of the viral RNA. Purified preparations of TRSV-RNA which had been diluted to l, 10, and 100 pg/ml in 0.1 M sodium phosphate buffer, pH 7.5, were tested for infectivity by inoculation to opposite, unifoliate leaves of Vigna unguiculata (Cowpea 'SR'). Since the specific activities of different RNA preparations were determined at different times, under varying greenhouse conditions, it was not possible to compare directly the biological activity of each of the RNA preparations. However, all of the preparations, at 10 pg/ml, induced the formation of an average of 49-350 local lesions/ cowpea leaf (at least 10 leaves were inoculated/RNA preparation). Separation of TRSV-RNA on Sucrose Density Gradients Centrifugation of 0-32.5% linear-log sucrose gradients was used for the separation of RNA-2 (1.2 x 106 daltons) from total TRSV-RNA. Heat denaturation of the RNA preparations prior to sucrose density gradient centrifugation was necessary to eliminate the presence of a third peak in sucrose gradients which sedimented just ahead of RNA-1 (2.2 x 10° daltons). The preparative centrifugation of heat-denatured TRSV-RNA (200 pg/Beckman SW41 rotor tube) gave a profile with two RNA peaks (Figure 11a). Fractions containing 73 FIGURE 11.--Sucrose density-gradient fractionation of TRSV— RNA. Fractions corresponding to the RNA-2 peak in Figure 11a were pooled, as indicated by brackets, concentrated by ethanol precipitation, and refractionated from sucrose gradients to give the profile in Figure 11b. The RNA-2 peak was again collected and concentrated by ethanol precipitation. Absorbanco at 254 nm 2.0 1.0 1.0 74 RNA 1 1 1 4 6 Depth (cm) 75 the RNA-2 peak were collected as indicated by the brackets in Figure 11a, followed by ethanol precipitation of the RNA and recentri- fugation on 0 to 32.5% linear-log sucrose density gradients. The profile of the fractionated RNA in the second sucrose density gradient is given in Figure 11b. The peak containing the small RNA was again collected (as indicated by brackets in Figure 11b), ethanol precipitated, and analyzed for purity and intactness by agarose- formaldehyde gel electrophoresis. The fact that RNA-l was always present in relatively low concentrations compared to RNA-2, made it impossible to adequately separate RNA-1 from total TRSV-RNA. RNA-2, fractionated by two cycles of sucrose gradients, was compared to unfractionated TRSV-RNA by agarose-formaldehyde gel electrophoresis, and by its infectivity in cowpea plants. The unfractionated TRSV-RNA showed two peaks typical of RNA 1 + 2 of TRSV in agarose-formaldehyde gels (Figure 12a). Analysis of the purified RNA-2 on agarose-formaldehyde gels (Figure 12b) showed that this RNA had been successfully separated from RNA-1 by sucrose density gradient fractionation, and the resulting RNA was covalently intact. The purified RNA-2 preparation was not infectious when inoculated onto cowpea plants at a concentration of 100 pg/ml. The preparation of unfractionated RNA from which the RNA-2 was purified produced 62 lesions/cowpea leaf when inoculated at a concentration of 10 pg/ml. This purified preparation of RNA-2 was used for analysis of jp_yitpp translation in the wheat embryo cell-free system. 76 FIGURE 12.--Analysis by agarose-formaldehyde gel electrophoresis of unfractionated TRSV-RNA (a), and RNA-2 fractionated by two successive cycles of sucrose density gradient centrifugation (b). Samples of RNA were separated by electrophoresis at 4°C at 2 milliamps/gel for 3 to 4 hours, after which the gels were scanned at 254 nm. Absorbanco at 254 nm 77 0.5 0.25 " 1 1 °§ RllA 1 0.25 J J 1 1 1 l 1 2 3 4 5 6 Depth(cm) 78 Detection of Poly(A) Tracts in TRSV-RNA The use of oligo(dT)-cellulose column chromatography was investigated as a possible tool for further purification of TRSV-RNA. A large proportion of TRSV-RNA was bound to oligo(dT)-cellulose in buffers of high ionic strength. This suggests that TRSV-RNA molecules contain poly(A) tracts. The binding of TRSV-RNA to an oligo(dT)-cellulose column is demonstrated by the elution profile given in Figure 13. In this experiment 100 pg of denatured TRSV-RNA from the tobacco isolate, was chromatographed on an oligo(dT)- cellulose column 5 cm long and 1.5 cm in diameter as described under Materials and Methods. Upon chromatography of total TRSV-RNA about 65% of the RNA was bound and then eluted, and about 35% was not bound to the oligo(dT)-cellulose column. The two RNA species that make up the genome of TRSV were both bound to oligo(dT)-cellulose under high ionic strength conditions. Agarose-formaldehyde gel electrophoresis of the RNA eluted under high and low ionic strength conditions showed that both RNA species of TRSV were bound equally under high ionic conditions. Figure 14 shows the agarose formaldehyde gel profiles of unfractionated TRSV- RNA (prior to chromatography on oligo(dT)-cellulose), and profiles of the RNA-containing fractions eluted under high and low ionic strength conditions. Electrophoresis of the gel whose profile is shown in Figure 14a was carried out at a different time than those in Figure 14b and 14c. Only 25% of the 00260 absorbing material that was eluted with 'high salt' buffer precipitated with ethanol. When the material which was precipitated by ethanol was analyzed by 79 FIGURE 13.--Oligo(dT)-cellulose chromatography of TRSV-RNA. Poly (A)' RNA and Poly (A)+ RNA, in fractions 3-4, and 12-14, respectively, were ethanol precipitated and analyzed using agarose- formaldehyde gel electrophoresis. 80 I .1 1' Elution 101 1 1 0.8 - H 0.7 t 0.6 ' 0.5 " 0.4 F 0.3 r 0.2 r 0.1- A260 /m| o "1' ‘ * J O 5 IO 15 20 Frociion Number 81 FIGURE 14.--Electrophoresis of RNA preparations from oligo(dT)- cellulose chromatography in agarose-formaldehyde denaturing gels. (a) Unfractionated RNA from TRSV prior to oligo(dT)-cellulose chromatography, (b) Fraction of TRSV-RNA preparation eluted with 'high salt' buffer, (c) Fraction of TRSV-RNA preparation eluted with '1ow salt' buffer. Each gel contained 1-4 pg of RNA, estimated spectrophotometrically. Samples of RNA were separated by electrophoresis at 2 milliamps/gel for 3 to 4 hours, after which gels were scanned at 254 nm. Absorbance at 254 nm 82 RNA 2 RNAT - Migration 83 electrophoresis (0.25 00260 absorbing units/gel) there was little evidence of discrete RNA species (Figure 14b). The 00260 absorbing material which had been eluted with 'low salt' buffer was also etmnml precipitated and subsequently analyzed by gel electrophoresis (0.8 00260 absorbing units/gel). Two discrete RNA species were evi- dent in the fraction eluted with '1ow salt' buffer (Figure 14c). Since the relative proportions of RNA-l and RNA-2 remained the same after chromatography on oligo(dT)-cellulose, it was concluded that both genome RNAs of TRSV are polyadenylated in roughly equal amounts, and that oligo(dT)-cellulose chromatography could be used as a puri- fication step in the extraction of TRSV-RNA. However, since it was possible to obtain relatively pure TRSV-RNA without oligo(dT)— cellulose chromatography this step was not included in the routine RNA purification procedure. Translation of TRSV-RNA in the Wheat-Embryo Cell-Free System Addition of TRSV-RNA to the wheat embryo cell-free protein synthetic system stimulated the incorporation of 3H-valine into acid-precipitable material 50 to lOO-fold, showing that TRSV-RNA is a relatively efficient messenger RNA. These results are similar to those of Fritsch et a1. (11) for translation of tomato black ring virus in the wheat embryo system. When a range of concentrations of RNA were tested, incorporation was essentially linear up to a con- centration of 2 pg RNA/50 pl of reaction mix (Figure 15). It has been found that low levels of plant virus messenger RNAs give improved completion of polypeptide chains in the wheat embryo system 84 FIGURE 15.--Incorporation of 3H-valine into protein in the wheat embryo extract in response to added TRSV-RNA. Concentrations of Mg++ and K+ were 3.2 mM and 45 mM, respectively. 85 m.~ 2183 14:23:52.3 <2: 0d mé o; d a _ ow cc cm cm 03. cup c-Ol X (Wdal leA'sH 86 (57). Therefore, in subsequent experiments TRSV-RNA concentrations of 1-2 pg/SO pl incubation mix were used, a level which was less than saturation, but which gave adequate incorporation for analysis of 1p yitpg products. The magnesium ion optimum for 3H-valine incorporation into acid-insoluble material varied for different preparations of TRSV- RNA, but was usually between 2.7 and 3.2 mM (Figure 16). The magnesium ion concentration was adjusted to the optimum concentration .¢U_W,M1 for each RNA preparation. The potassium ion level for maximum 3H- valine incorporation was found to be 45 mM for TRSV-RNA (Figure 17). Since there was little variation in optimum potassium requirements for different RNA preparations, this concentration of potassium was used in all synthesis reactions. The time kinetics of 3H-valine incorporation in the wheat embryo system were determined. Translation of unfractionated TRSV- RNA directed valine incorporation for at least 70 minutes (Figure 18). The size of the jp_yitpg products translated after different incuba- tion times was not found to increase with increasing incubation time (data not shown). This experiment indicates that the routine incuba- tion time of 45 minutes was adequate to obtain the largest possible translation products. Analysis of Translation Products by SDS- Polyacrylamide Gel Electrophoresis Analysis in SOS-polyacrylamide gels of the 3H-valine labeled products of TRSV-RNA revealed 10 to 11 bands of protein with a maxi- mum molecular weight of about 133,000 daltons (Figure 19). When TRSV 87 FIGURE 16.-—Mg++ requirement for amino acid incorporation directed by TRSV-RNA in the wheat embryo extract. Acid-insoluble incorporation was assayed in a 50-pl subsample as described in Materials and Methods. Concentrations of K+ and RNA were 45 mM and 1 pg/50 p1, respectively. 88 22:: 528.533 .192 m c m - fl 4 cw m.. cm mm on e-Ol x (was) Ian-9H 89 FIGURE 17.--—K+ ion requirement for amino acid incorporation directed by TRSV-RNA in the wheat embryo system. Acid-insoluble incorporation was assayed in a SO-pl subsample as described in Materials and Methods. Concentrations of Mg++ and RNA were 3.2 mM and 1 pg/50 p1, respectively. 90 our AEEV Cczflzcmucau+ v. cm om cc on cow 4 u now imw now 1mm new 8.011: (de) 1211 ‘CH 91 FIGURE 18.--Kinetics of 3H-valine incorporation directed by TRSV-RNA from the 'Jersey' blueberry isolate in wheat embryo system. Reaction mixture was made 45 mM K+, 3.2 mM Mg++, and RNA concentration was 2 pg/50 pl of reaction mix. Acid-insoluble incorporation wasassayed in a 50 pl subsample as described in Materials and Methods. 92 l J 30 40 50 60 70 Time (min) 1 20 1 10 100*- l o o o a: co m 3.011: (wdp) 12A - Hg 60— 500 93 coat protein subunit was used as a molecular weight standard it migrated to a position similar to one of the 1p.yjt:9_protein products, but this product was not made in large amounts as might be expected of coat protein since large amounts of virus coat protein are syntheized j_ vivo. Tomato black ring virus, which is a nepovirus like TRSV, has been shown to synthesize an overwhelming amount of coat protein 1_ vivo (11). The number of protein products that were made when the RNA from three different isolates of TRSV was used as messenger RNA in the wheat embryo cell-free translation system was the same regardless of the source of the RNA (Figures 19-21). However, the relative amounts of various proteins differed from one RNA preparation to another. Dual-Label Experiments to Detect Small Differences in Molecular Weights of Protein Products Produced by Different Isolates of TRSV Molecular weight estimates of proteins by SOS-polyacrylamide gel electrophoresis are only accurate, at best, to within 15% (71). This is true even when the proteins are separated by electrophoresis at the same time, but in different gels (71). Therefore, it is very difficult to detect a difference in the molecular weight of two proteins by SOS-polyacrylamide gel electrophoresis in cylindrical gels unless the molecular weight difference is greater than 10%. For this reason a dual-label experiment was conducted in which the protein 14 products from one isolate of TRSV, which had been labeled with C- valine, and protein products from another isolate labeled with 94 FIGURE 19.-—SDS-polyacrylamide gel electrophoresis of in yitrp translation products using RNA from the tobacco isolate of TRSV. Electrophoresis was done at 8 milliamps/gel for 60 to 90 minutes. The migration position of molecular weight markers is indicated by the arrows. Migrationis from left to right. In .8632 82m .80 00 on On mv O¢ on On mm ON 9 O. m o 95 d 00. l ICON - oon Loco - com 1 00m 00» 1 00m 1 - 00m 000. l l + + + * 00: go a... .26 4mm 08. l (wdp) IDA-H, 96 FIGURE 20.--SDS-polyacrylamide gel electrophoresis of 1p yitgg translation products using RNA from the soybean isolate of TRSV. Electrophoresis was done for 60 to 90 minutes at 8 milliamps/gel. The migration position of molecular weight markers is indicated by arrows. Migration is from left to right. 97 mm 00 mm On 04v 9» mm on mm ON 9 IIIIL— .2232 8.6 to O. m 0 M q A 4 05 — 4 a: - u a 4 4 .30 4 — 4 4mm 4 — _ 4 oo. 4 com 1 com I cow 1 00m 1 com L 004- 4 00m 1 00m 1 000. 1 CO: (de) IDA-H, 98 FIGURE 21.--SDS-polyacrylamide gel electrophoresis of in xjtrg translation products using RNA from the 'Jersey' blueberry isolate of TRSV. Electrophoresis was done for 60 to 90 minutes at 8 milliamps/gel. The migration position of molecular weight markers is indicated by the arrows. Migration is from left to right. 99 .3532 8.5 Bo mm8mmOmm¢O¢mm0mm~0NQQmo q q d d d a q q u d 1 q q 1 000 4000. 000. 1 000m 1 000m 1 1000n 1 00mm 1 000¢ L 00m¢ 0000 a a .— 4 o 1 88 26 o .6 «mm 8...:932 l lOO 3H-valine, were mixed together and separated by electrophoresis on the same SOS-polyacrylamide cylindrical gel. The gels were sliced and the radioactivity from the 3H-valine products, and from the 14C- valine products was then determined for each gel slice. The protein profiles for the set of proteins from one isolate (counted as 3H) was then plotted on the same graph as the protein 14C). This type of profile from the second isolate (counted as 'within-gel' comparison using a dual-label was done with the protein products of the 'Jersey' blueberry, soybean, and tobacco isolates of TRSV (Figures 22 to 24). Each comparison was made at least three times, but only one graph of each comparison is shown here (Figures 22-24). A close examination of Figure 22 shows that the protein profiles of the 'Jersey' blueberry and the tobacco isolate are similar except for the two protein peaks that co-migrate with the phosphorylase-a and the cytochrome-c markers. On replicate gels the migration difference between the protein which co-migrated with the phosphorylase-a marker was not evident. Also it was difficult to compare the migration of some of these higher molecular weight pro- teins because the protein peaks were relatively small. When the protein products of the soybean and tobacco isolates were compared in the same gel using a dual-label there were no con- sistent differences in the migration of the polypeptides from the different TRSV isolates (Figure 23). However, a comparison of the products of the soybean isolate and the 'Jersey' blueberry isolate (Figure 24) showed a consistent molecular weight difference in the protein product which migrated slightly ahead of the cytochrome-c 101 FIGURE 22.--SDS-polyacrylamide gel electrophoresis of in_yitrg translation products of TRSV-RNA from the tobacco isolate (-—J and the 'Jersey' blueberry isolate (---). jn_vitrg protein products were made in the wheat embryo system using 3H-valine when translating the RNA from the 'Jersey' blueberry isolate, and 14C-valine when translating the RNA from the tobacco isolate. The protein products from these two isolates were then mixed, denatured, and separated electrophoretically in the same gel for 60 to 90 minutes at 8 milliamps/ gel. The migration position of the molecular weight markers is indicated by the arrows. Gels were sliced and counted simultaneously for 14C and 3H radioactivity. Migration is from left to right. (—)Wd0 0,. O O O O O O y (p 0 O N V CD 0 In G V n N — 0 0 j I I I I I I > 3’2 N +— ‘ F E > \ U) o g E 3 r g * . o o .‘n , 3 3‘ 3" 3 g --"’ d 8 is P u:-—---"' q ' . ‘ ------ *-. ' v > o"‘ u-i 0 ' ”-..! 8 x" g: .E ’o’-"”‘ -‘ 8 E “'- -~ -D "o-‘- . 's... 2» ‘2 a Q"" -—-u E .2“ Pa" :13” ‘ g o -..“;- ... ‘-"-p K g ‘--------- q 9 -.‘---- t e . m o l s 1 L 1 1 1 1 1 Ho 0 O O O O 8 § § 8 9 8 ° .‘8 ~ 4. ‘4 O . ... M r . 100m. 1.: u . ooou ( .4 4 4 4 4 4 4 4 o 28 4.834490 <8 3.82.4 1.... >9: 1 >94» 536m 100% >mE 8033' 1 100mm 105 FIGURE 24.--SDS-polyacrylamide gel electrophoresis in in vitro translation products of TRSV-RNA from the 'Jersey' blueberry isolate (~——l and the soybean isolate (---). In vitro protein products were made in the wheat embryo system using 3H—valine when translating the RNA from the soybean isolate, and 14C-valine when translating the RNA from the 'Jersey' blueberry isolate. The protein products from these two isolates were then mixed, denatured, and separated electro- phoretically in the same gel for 60 to 90 minutes at 8 milliamps/gel. The migration position of the molecular weight markers is indicated by the arrows. 14C Gels were sliced and counted simultaneously for and 3H radioactivity. Migration is from left to right. 106 (—)wao-o,. .00. 0mm 0 co tr 0V0 00h 00h 0mm .3452 345 .8 N4. .5 mm 8 04v Nm cm 0. 0 4 1 i- 4 4 4 4 4 4 O \ >1 > .4 r 4/ I .4 00¢ ... I 4 r a .. 000 ... ..4 a. .40 c. I rt. 2. 00m. H 4. NJ. 3 \ m as. J “a. .- d r 4 4 I. W I .4 . 000. s ) .... 4 4 1.1 - 4 4 4 4 .— 4 4 88 26 38 4.438 426 48 24.4854 >mm... c0346."). 1 1.. 2mm... I >mm... b.3020 .Fflo?l|1 00cm I 000m 107 marker. The protein product of the tobacco isolate is slightly larger than the corresponding protein product of the 'Jersey' blue- berry isolate. In the comparison between the 'Jersey' blueberry isolate and the soybean isolate, this protein product also differed in molecular weight, the protein product of the soybean isolate being slightly larger than that of the 'Jersey' blueberry isolate. Although the size of this protein product was smaller for the ’Jersey' blueberry isolate when compared with both the soybean and tobacco isolate product, there was no detectable difference in the molecular weight of this protein product when the soybean and tobacco isolates were compared (Figure 23). The in_yit:g protein products of the different isolates of TRSV were sometimes strikingly different in the relative amounts of the proteins synthesized. For instance, a comparison of the profiles of the protein products from the 'Jersey' blueberry isolate and the soybean isolates of TRSV (Figure 24) shows that more of the smaller molecular weight products were synthesized when RNA from the 'Jersey' blueberry was used as messenger RNA in the wheat embryo cell-free synthetic system. The shifts in the protein product profiles to larger or smaller molecular weight proteins did not appear to be a function of the TRSV isolate used as the source of RNA. Protein Products Made by RNA-2 of TRSV When RNA-2 of TRSV was translated in the wheat embryo synthetic system, the translation efficiency of RNA-2 was about three times that for the unfractionated TRSV-RNA from which the 108 small RNA had been isolated, even though equal amounts of RNA were used. The SDS-polyacrylamide gel profiles of protein products made using unfractionated TRSV-RNA, and using RNA-2 are given in Figure 25. The unfractionated RNA used in this protein synthesis was from the same RNA preparation from which the RNA-2 was isolated. One important feature of Figure 25 is that the largest translation product of RNA-2 has a molecular weight of about 130,000. A product of this size represents between 8l and 93% of the total coding capacity of RNA-2 of TRSV. Another important aspect of the profiles of protein products from unfractionated TRSV—RNA and the RNA-2 of TRSV is the absence, in the RNA-2 profile, of the protein product which migrates just ahead of the standard marker cytochrome-c. It appears that RNA-2 is capable of inducing the synthesis of the same polypeptides as unfractionated TRSV-RNA with the exception of one polypeptide of molecular weight about l0,500 daltons. 109 FIGURE 25.--SDS-polyacrylamide gel analysis of the products of translation of (a) unfractionated TRSV-RNA, and (b) RNA-2 of TRSV. Electrophoresis was carried out for 60 to 90 minutes at 8 milliamps/gel. The migration position of molecular weight markers is indicated by the arrows. Migration is from left to right. 3l-l--Vol (dpm) 110 IGOOP 1400!- 1200 - 1000- 800 - 600- 400- 200- Tracking a? cytochaome c D‘ye 7000 - 6000 - 5000- 4000- 3000 - 2000 - 1000- l l 1 Tracking BSA Ctochromec ’ 1 ”l" 20 30 40 Gel Slice Number 70 DISCUSSION TRSV isolates can be clearly differentiated using several biological and physical-chemical criteria. This seems to be especially true for the isolates from 'Jersey' blueberry and from soybean which were found to be quite different from the other TRSV isolates in a number of properties studied. The seven isolates of TRSV used in this study showed a range of virulence on the herbaceous indicator plants tested. The soybean isolate generally produced the mildest symptoms, and the 'Jersey' blueberry isolate produced the most severe symptoms. It is of interest to speculate on the natural selection pressures which might have caused the difference in virulence seen in these two isolates of TRSV. The occurrence of TRSV in commercial plantings of soybean is thought to be mostly due to the use of infected soy- bean seed (58). TRSV-infected soybean seeds germinate and grow into infected seedlings, which produce seed that may be TRSV-infected. The perpetuation of this disease cycle is dependent on the survival of the TRSV-infected seedling to a point where it produces viable seed. An extremely virulent isolate of TRSV would stunt and debili- tate the soybean seedling to such an extent that production of infected seed would be unlikely. Relatively mild isolates of TRSV would be more likely to survive and reproduce in commercial soybean 111 112 production. 0n the other hand, transmission of TRSV in woody plants, such as blueberries, is thought to occur by nematode transmission, or through the use of infected propagating material; seed transmission is of little importance (64). The selection pressure on a TRSV iso- late which is transmitted by nematodes from one plant to another, or by man through propagation, would be quite different from the selec- tion pressure on a seed-borne TRSV isolate, and would not necessarily select for variants which produced a mild, non-lethal disease in the host plant. Although there was only a small difference in the thermal inactivation temperatures of the seven isolates of TRSV (Table 2), it is interesting to note the correlation between the resistance to thermal inactivation and the virulence of the TRSV isolates on herbaceous hosts. The 'Jersey' blueberry, 'Riesling' grape, and watermelon isolates were the most resistant to inactivation by heating and were also the most virulent on most of the herbaceous plants tested. In contrast, the soybean isolate, which induced the mildest symptoms on the herbaceous hosts, was one of the isolates most sensitive to thermal inactivation. The results of the spur formation tests (Table 3) indicate that the coat protein of the 'Jersey' blueberry isolate has at least one antigenic determinant which is not present on the other TRSV isolates. The watermelon isolate and the soybean isolate also appear to have unique antigenic sites in their coat proteins when compared to some of the other TRSV isolates. The lack of spur formation in the remaining isolates of TRSV indicates that these 113 isolates are not different in any surface antigenic determinants, at least that can be detected by spur formation tests. The interpretation of the spur formation test results can best be explained by analyzing the comparison between the 'Jersey' blueberry isolate and the two isolates from soybean and cherry. Spurs were formed between the 'Jersey' blueberry antigen and the cherry antigen. Spurs were also formed between the 'Jersey' blue- berry antigen and the soybean antigen when reacted with the 'Jersey' blueberry antiserum. This means that the 'Jersey' blueberry isolate has an antigenic site not shared with the cherry antigen, and that the 'Jersey' blueberry isolate has an antigenic site not shared by the soybean isolate. The possibility exists that the cherry isolate possesses an antigenic site not present on the 'Jersey' blueberry isolate. No spur was formed between the cherry antigen and the 'Jersey' blueberry antigen when reacted with the cherry antiserum. Therefore, all the antigenic sites found on the cherry isolate are also found on the 'Jersey' blueberry isolate. The 'Jersey' blueberry antigen possesses an antigenic site not present on the soybean isolate. Again, the possibility exists that the soybean isolate possesses an antigenic determinant not present on the 'Jersey' antigen. A spur was formed between the soybean antigen and the 'Jersey' antigen when reacted with the soybean antiserum. Therefore the soybean isolate has an antigenic site which is not shared by the 'Jersey' blueberry isolate. This is in contrast to the cherry isolate which (as explained in the 114 preceding paragraph) possesses only antigenic sites which are also present on the 'Jersey' blueberry isolate. From this example it is evident that an interpretation of spur formation tests to determine the serological relationship between isolates must include two tests for spur formation, i.e., both antisera to each of the isolates must be tested against the paired antigens. Nhen dual comparisons such as this are made, it becomes evident that even though two isolates (such as the cherry and the soybean isolate) appear to be similar in that they both lack an antigenic determinant not present on a third isolate (such as the 'Jersey' blueberry isolate), they may also differ in that one of the two isolates may possess an antigenic determinant not present on the third isolate. The results of the cross-absorption tests (Table 4) show that the blueberry isolate possesses at least one antigenic determinant which is not present on the soybean or tobacco isolates; this is in agreement with the results of the spur formation tests. The results of the cross-absorption tests also indicate that the tobacco isolate possesses at least one antigenic determinant not present on the soybean isolate. This conflicts with the results of the spur-formation tests which showed that the tobacco isolate did not possess any unique antigenic sites over those found on the soybean isolate. The reason for this discrepancy is not clear, but it may be due to the difference in the medium in which the antibody-antigen reaction takes place in these two tests. In the spur formation tests, the reaction takes place in an agar-gel II. .54. - 115 matrix, whereas in the cross-absorption tests the reaction takes place in an aqueous solution. The 'Jersey' blueberry isolate and the watermelon isolate of TRSV both possess coat proteins which are serologically unique com- pared to the other TRSV isolates. These two isolates were also shown to be the most virulent on the different herbaceous hosts tested. It is not known whether the unique coat protein properties of these two isolates are responsible for their increased virulence in the herbaceous test plants. Viruses which differ from one another in the composition of their coat proteins, as evident by serology, do not necessarily differ in the overall charge on their virus particles. When the electrophoretic mobility of the 'Jersey' blueberry isolate was compared to that of the soybean and tobacco isolates, it was evident that the virus particles of the 'Jersey' blueberry isolate were more negative in their overall charge than the virions of the soybean or tobacco isolates. Evidence from infection studies done with plant viruses in protoplasts indicates that the overall charge on the virus particle may be important in determining the virus-cell interaction during infection (25). The charge difference on the virions of the 'Jersey' blueberry isolate may be a contributing factor in the increased virulence of this isolate, especially if the charge of the virus particle is critical in determining cell- virion interactions in the plant. Although differences in electrophoretic mobiltiy have been used to characterize the strains of several viruses, such as BMV (28) and TMV (56), there is no 116 evidence to indicate that the overall charge of the virus particle influences its ability to cause disease in a host plant. The RNA of TRSV is theoretically capable of coding for about seven times the amount of protein found in the coat protein of TRSV. For this reason, I synthesized the total protein products of TRSV- RNA in wheat embryo system, and then compared the in vitrg_protein products of three different isolates of TRSV. This allowed me to determine if there were protein differences, other than coat protein differences, which might be affecting biological properties of these TRSV isolates. The results presented in this thesis for in_vitrg_transla- tion of TRSV-RNA in the wheat embryo system show that the RNA-2 of TRSV codes for a polypeptide of about l30,000 daltons. This accounts for roughtly 88% of the coding capacity of the RNA-2 of TRSV. There are as many as 9 other smaller polypeptides which are also produced when RNA-2 is translated in the wheat embryo system. Since the 130,000 dalton product accounts for about 88% of the coding capacity of RNA-2, the smaller polypeptides either represent proteolytic cleavage products of the 130,000 dalton product, or products from incomplete peptide synthesis. The addition of spermine and aurin tricarboxylic acid to the jg_yjt§9_translation mix should have minimized the formation of incomplete synthesis of polypeptides. An analysis of the protein products of tomato black ring virus (TBRV) (ll), a nepovirus like TRSV, showed that the RNA-2 of TBRV was translated into a protein product of 160,000 daltons molecular weight. This represents an almost complete translation 117 product of the RNA-2 of this virus. As with other nepoviruses the RNA-2 is thought to code for coat protein (11,20), and if this is so, then the 130,000 dalton protein product of RNA-2 of TRSV must be processed by proteolytic cleavage to produce the 54,000 dalton coat protein. Post—translational cleavage has been demonstrated in cowpea mosaic virus, a virus which resembles TRSV in many other respects (35). The existence of post-translational cleavage of the protein products of TRSV-RNA could be demonstrated by inhibiting protease activity through the addition of amino acid analogs or protease inhibitors to the wheat embryo translation system. There does not seem to be a primary translation product of the RNA-l (2.2 x 105 daltons) of TRSV. Although it was not possible to separate RNA-l from total TRSV-RNA, an examination of the protein products made with total TRSV-RNA shows a protein of maximum molec- ular weight of about 130,000 daltons, i.e., the primary translation product of RNA-2. When one compares the translation products made using RNA-2 (Figure 25), it is evident that the translation products of total TRSV-RNA contain only one additional protein product of about 10,500 daltons molecular weight. It appears that this 10,500 dalton product is coded by RNA-1 of TRSV when a mixture of the two RNAs from TRSV is translated in the wheat embryo system. This result must be interpreted with caution since RNA-1 has not yet been separated from total TRSV-RNA, and translated by itself in the wheat embryo system. 118 Schwinghamer and Symons (54) found that translation of unfractionated cucumber mosaic virus RNA gave essentially the same protein product profile as that found when RNA-4 (one of 4 RNA species of CMV) was translated alone. The translation products of the other 3 RNA species were produced only when these RNA species were translated separately in the wheat embryo system. By analogy, it may be that it is necessary to separate RNA-1 from total TRSV-RNA to produce the primary translation product of RNA-l in the wheat embryo system. In contrast to our results with TRSV, the translation of unfractionated TBRV-RNA (11) yielded a large protein of molecular weight about 200,000 daltons, much larger than any of the proteins produced when we translated unfractionated TRSV in the wheat embryo system. This 200,000 dalton protein product of TBRV may represent the primary translation product of the large RNA of TBRV. A cor- responding protein representing the primary translation of the RNA-l of TRSV was not detected when unfractionated TRSV-RNA was translated in the wheat embryo system. It appears that the 10,500 dalton pro- tein product was the only protein product made by the RNA-1 of TRSV when RNA-1 and RNA-2 were translated together in the wheat embryo system. Analysis of SOS-polyacrylamide gels separating proteins synthesized jg_vjt§9_showed considerable variation in the relative amounts of the protein products when different RNA preparations were used as messenger RNA. There are several possible explana- tions which might account for shifts in the protein product profiles. 119 One possibility is that the RNA from some isolates of TRSV has properties which allow selective translation of certain proteins, resulting in a greater proportion of small or large molecular weight polypeptides. However, since two different RNA preparations from the same isolate of TRSV were capable of inducing the jn_vitro synthesis of polypeptides whose profiles were quite different (compare the soybean isolate product profiles in Figure 20 and Figure 23), this possibility seems unlikely. A second possible explanation for the shifts in the relative amounts of the in vitrg_protein products is that the RNA is trans- lated to produce a primary translation product which is subsequently cleaved to produce smaller molecular weight polypeptides. Variation in the amount of proteolytic cleavage of this primary translation product could account for changes in the relative amounts of the different protein products. Post-translational proteolytic cleavage has been demonstrated to be a function of proteases coded by the viral RNA, and also a function of endogenous proteases present in the jg_vjtrg_protein synthetic system (43). Marcus, et a1. (32) and Dr. James Asher (personal communication) both indicated that the amount of endbgenous message in the wheat extract varies depend- ing on the pH of the homogenate during the preparation of the wheat embryo extract. A higher pH resulted in release of the endogenous wheat embryo messenger RNA. If this endogenous messenger RNA is translated to produce an active protease, then varying amounts of endogenous messenger RNA could lead to varying amounts of proteolytic cleavage of other ig_vitro protein products. 120 A third possible explanation for the shifts in the jn_vitrg protein product profiles involves the possibility that the RNA of TRSV has several initiation sites for protein synthesis. The avail- ability of these sites for initiation may be dependent on secondary RNA structure, or on the presence of other factors which enhance or inhibit initiation at these sites (47). A subtle change in the secondary structure of the RNA, or in the concentration of these other factors which have been shown to affect initiation, could affect the availability of certain initiation sites for protein synthesis, and result in reduced or enhanced synthesis of a par- ticular protein product. Selective degradation by ribonuclease could also lead to changes in the secondary structure of the RNA which might influence the availability of different initiation sites on the RNA. A comparison of the molecular weights of the translation products from three isolates of TRSV showed that the molecular weight of the 10,500 dalton protein product was isolate-dependent. This protein product was detectably smaller when the protein products of the 'Jersey' blueberry isolate were compared to the protein products of either the tobacco or the soybean isolates (Figures 22-24). This isolate-dependent molecular weight difference in the 10,500 dalton protein product is a result of a difference in the RNA-l of the TRSV isolates, since our evidence indicates that the 10,500 dalton product is coded by RNA-l. This is in contrast to the differences which are a function of coat protein differences and which are coded by RNA-2 of TRSV. 121 The 'Jersey' blueberry isolate of TRSV is more virulent than the soybean or tobacco isolates of TRSV. The 'Jersey' blue- berry isolate also differs from the soybean and tobacco isolates of several properties which are a function of the coat protein of the virus. Since current evidence indicates that the genetic information for coat protein is found on RNA-2 of TRSV, these coat protein- dependent properties are probably a function of RNA-2 of the TRSV genome. In addition to the coat protein differences between these isolates, there is also an isolate dependent difference in the molecular weight of a gene product from RNA-1 of these isolates. It is not known which of the gene product differences, those from RNA-l or those from RNA-2, is responsible for the increased virulence of the 'Jersey' blueberry isolate of TRSV. Pseudorecombinant analysis has been used to determine which of the two RNAs of raspberry ringspot virus (RRSV) is responsible for various properties such as severity of systemic symptoms in various herbaceous hosts, serological specificity, and nematode transmissibility (20). In experiments with four naturally occurring strains of RRSV, virus hybrids were made by mixing RNA-l and RNA-2 preparations from different strains. The genetic determinant for symptom severity in the herbaceous hosts was shown to be carried by RNA-1, the large RNA of RRSV, which does not carry the genetic information for coat protein. An analagous pseudorecombinant experi- ment with the isolates of TRSV in this study would reveal which of the 2 RNA species of the 'Jersey' blueberry isolate is responsible for the increased virulence of this isolate. 122 The manner in which the biological properties of viruses are determined by the virus genome is an important problem for plant virologists. To solve this problem it is necessary to know the number of proteins that a virus is capable of producing and also what the functions of these viral-coded proteins are. In this research I have identified some of the biological properties of several isolates of TRSV, and have also identified some of the unique biochemical properties of these TRSV isolates. It remains to be determined which of these unique biochemical properties that have been identified are responsible for the biological properties of these TRSV isolates. BIBLIOGRAPHY 123 10. BIBLIOGRAPHY Ball, E. M. 1974. Serological tests for the identification of plant viruses. American Phytopathological Society. St. Paul, Minnesota. 31 pg. Bancroft, J. B., and L. C. Lane. 1973. Genetic analysis of ; cowpea chlorotic mottle virus and brome mosaic virus. Journal of General Virology 19:381-389. Brakke, Myron K., and Nancy van Pelt. 1970. Linear-log sucrose gradients for estimating sedimentation coefficients of plant viruses and nucleic acids. Analytical Biochemistry 38: 56-64. I P Allie-'1 " Chifflot, S., P. Sommer, D. Hartman, C. Stussi-Garaud, and L. Hirth. 1980. Replication of alfalfa mosaic virus RNA: Evidence for a soluble replicase in healthy and infected tobacco leaves. Virology 100:91-100. Chu, P. W. G., and R. I. B. Francki. 1979. The chemical sub- unit of tobacco ringspot virus coat protein. Virology 93:398-412. Dahl, 0., and C. Knight. 1963. Some nitrous acid-induced mutants of tomato atypical mosaic virus. Virology 21:580-586. Daubert, S. A., and G. Bruening. 1979. Genome-associated proteins of comoviruses. Virology 98:246-250. deZeeuw, D. J., and R. A. Crum. 1963. Inheritance of resistance to tobacco ringspot and cucumber mosaic viruses in black cowpea crosses. Phytopathology 53:337-340. Dougherty, W. G., and E. Hiebert. 1980. Translation of poty- virus RNA in a rabbit reticulocyte lysate: Identification of nuclear inclusion proteins as products of tobacco etch virus RNA translation and cylindrical inclusion protein as a product of the potyvirus genome. Virology 104:174-182. Fairbanks, G., T. Steck, and D. F. Wallach. 1971. Electro- phoretic analysis of the major polypeptides of the human erythrocyte membrane. Biochemistry 10:2606-2617. 124 11. 12. 13. 14. 15. 16. 17. 18. 19. 21. 22. 23. 125 Fritsch, C., M. A. Mayo, and A. F. Murant. 1978. Translation of satellite RNA of tomato black ring virus jn_vitro and in tobacco protoplasts. Journal of General Virology 40:587-593. Ghabrial, S. A., and R. M. Lister. 1973. Coat protein and symptom specification in tobacco rattle virus. Virology 5221-12. * Gibbs, A. J., and 8. Harrison. 1976. Plant Virology, The Principles. Edward Arnold (Publishers) Ltd., London. 292 pg. Golini, F., A. Nomoto, and E. Wimmer. 1978. The genome-linked protein of picornaviruses. Virology 89:112-118. Gooding, G. V., Jr. Personal communication. Gooding, G. V., Jr. 1970. Natural serological strains of tobacco ringspot virus. Phytopathology 60:708-713. Hadidi, A., and H. Fraenkel-Conrat. 1973. Characterization and specificity of soluble RNA polymerase of brome mosaic virus. Virology 52:363-372. Hariharasubramanian, V., A. Hadidi, 8. Singer, and H. Fraenkel- Conrat. 1973. Possible identification of protein in brome mosaic virus infected barley as a component of viral RNA polymerase. Virology 54:190-198. Harrison, 8. 0. 1977. CMI/AAB Descriptions of Plant Viruses No. 185, Nepovirus group. Commonwealth Mycological Institute and the Association of Applied Biologists, Kew, Surrey, England. Unpaged. Harrison, 8. 0., A. F. Murant, M. A. Mayo, and I. M. Roberts. 1974. Distribution of determinants for symptom production, host range and nematode transmissibility between two RNA components of raspberry ringspot virus. Journal of General Virology 22:233-247. Harrison, 8. D., and A. F. Murant. 1977. Nematode trans- nfissibflity of pseudo-recombinant isolates of tomato black ring virus. Annals of Applied Biology 86:209-212. Harrison, B. 0., W. M. Robertson, and C. E. Taylor. 1974. Specificity of retention and transmission of viruses by nematodes. Journal of Nematology 6:155-164. Hoof, H. A. van. 1968. Transmission of tobacco rattle virus by Trichordorus species. Nematologica 14:20-24. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 126 Hruby, D. E., and W. K. Roberts. 1978. Encephalomyocarditis virus RNA. III. Presence of a genome-associated protein. Journal of Virology 25:413-415. Jarvis, N. P., and H. H. Murakishi. 1980. Infection of proto- plasts from soybean cell culture with southern bean mosaic and cowpea mosaic viruses. Journal of General Virology 48:365-376. Johnson, J. 1947. Virus attenuation and mutation. (Abstr.) Phytopathology 37:12. Kahn, R. P., H. A. Scott, and R. L. Monroe. 1962. Eucharis ' mottle strain of tobacco ringspot virus. Phytopathology 52: . 1211-1216. Lane, L. C., and P. Kaesberg. 1971. Multiple genetic components in bromegrass mosaic virus. Nature New Biology 232240-43. Lawrence, C., and Thach, R. 1975. Identification of a viral protein involved in post-translational maturation of the encephalomyocarditis virus capsid precursor. Journal of Virology 15:918-928. “-611“?- Lee, Y. F., A. Nomoto, B. M. Detjen, and E. Wimmer. 1977. A protein covalently linked to poliovirus genome RNA. Pro- ceedings of the National Academy of Science, USA 74:59-63. Lehrach, H., 0. Diamond, J. M. Wozney, and H. Boedtker. 1977. RNA molecular weight determinations by gel electrophoresis under denaturing conditions, a critical re-examination. Biochemistry 16:4743-4751. Marcus, A., D. Efron, and D. P. Weeks. 1974. The wheat embryo cell-free system, pp. 749-753. Ig_K. Moldave and L. Grossman (eds), Methods in Enzymology, Vol. XXX, Nucleic Acids and Protein Synthesis, Part F. Academic Press, New York and London, 808 pg. Matthews, R. E. F. 1949. Studies on potato virus X. 11. Criteria of relationships between strains. Annals of Applied Biology 36:460-474. May, J., J. Gilliland, and R. Symons. 1970. Properties of a plant virus RNA polymerase in particulate fractions of cucumber infected with cucumber mosaic virus. Virology 41:653-664. 35. 36. 37. 38. 39. 40. 41. 43. 44. 45. 46. 127 Mayo, M. A., H. Barker, and B. 0. Harrison. 1979. Evidence for a protein covalently linked to tobacco ringspot virus RNA. Journal of General Virology 43:735-740. Mayo, M. A., A. F. Murant, and B. 0. Harrison. 1971. New evidence on the structure of nepoviruses. Journal of General Virology 12:175-178. McRitchie, J. J., and L. J. Alexander. 1963. Host-specific Lycopersicon strains of tobacco mosaic virus. Phytopathology 53:394-398. Moore, 8. J., and H. A. Scott. 1971. Properties of a strain of bean pod mottle virus. Phytopathology 61:831-835. Mouches, C., C. Bove, and J. M. Bove. 1974. Turnip yellow mosaic virus-RNA replicase: partial purification of the enzyme from the solubilized enzyme-template complex. Virology 58:409-423. Ouchterlony, O. 1964. Gel-diffusion techniques, pp. 55-78. In J. F. Ackroyd (ed.), Immunological methods. Blackwell, Oxford, England. 628 pg. Peden, K. W., J. T. May, and R. H. Symons. 1972. A comparison of two plant virus-induced RNA polymerases. Virology 47: 498-501. Pelham, H. 1978. Translation of encephalomyocarditis virus RNA jn_vitro yields an active proteolytic processing enzyme. European Journal of Biochemistry 85:457-462. Pelham, H. 1979. Synthesis and proteolytic processing of cowpea mosaic virus proteins in reticulocyte lysates. Virology 96:463-477. Purcifull, D. E., E. Hiebert, and J. G. McDonald. 1973. Immuno- chemical specificity of cytoplasmic inclusions induced viruses in the potato y group. Virology 55:275-279. Ramsdell, D. C. 1978. A strain of tobacco ringspot virus associated with a decline disease of 'Jersey' highbush blue- berry. Plant Disease Reporter 62:1047-1051. Randles, J. W., B. 0. Harrison, A. F. Murant, and M. A. Mayo. 1977. Packaging and biological activity of the two essential RNA species of tomato black ring virus. Journal of General Virology 36:187-193. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 128 Revel, M., and Y. Groner. 1978. Post-transcriptional and translational controls of gene expression in eucaryotes. Annual Review of Biochemistry 47:1079-1126. Rezaian, M. A., and R. I. B. Francki. 1973. Replication of tobacco ringspot virus. Virology 56:238-249. Roberts, 1. M., and B. 0. Harrison. 1970. Inclusion bodies and tubular structures in Chenopodium amaranticolor plants infected with strawberry latent ringspot virus. Journal of General Virology 7:47-54. Rush, M. C., and G. V. Gooding, Jr. 1970. The occurrence of tobacco ringspot virus strains and tomato ringspot virus in hosts indigenous to North Carolina. Phytopathology 60: 1756-1760. Rush, M. C., G. V. Gooding, Jr., and D. E. Ellis. 1968. Wild Rubus spp. as natural hosts of tobacco ringspot virus. Abstr.) Phytopathology 58:1065. Sakai, F., J. R. O. Dawson, and J. W. Watts. 1979. Synthesis of proteins in tobacco protoplasts infected with brome mosaic virus. Journal of General Virology 42:323-328. Sanger, H. L. 1969. Functions of the two particles of tobacco rattle virus. Journal of Virology 3:304-312. Schwinghamer, M. W., and R. H. Symons. 1975. Fractionation of cucumber mosaic virus RNA and its translation in a wheat embryo cell-free system. Virology 63:252-262. Sela, I., and A. Hauschner. 1975. Isolation and characteriza- tion of a TMV-RNA dependent enzyme from TMV-infected tobacco leaves. Virology 64:284-288. Sengbusch, P. von, and H. G. Wittman. 1965. Serological and physiochemical properties of the wild strains and two mutants of tobacco mosaic virus with the same amino acid exchange in different positions of the protein chain. Biochemical and Biophysical Research Communications 18: 780-787. Shih, D. S., and P. Kaesberg. 1976. Translation of the RNAs of brome mosaic virus: the monocistronic nature of RNA 1 and RNA 2. Journal of Molecular Biology 103:77-88. 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 129 Sinclair, J. B., and M. C. Shurtleff. 1975. Compendium of soy- bean diseases. Published by the American Phytopathological Society, Inc., St. Paul, Minnesota. 69 pg. Solberg, R. A., and J. G. Bald. 1963. Distribution of a natural and an alien form of tobacco mosaic virus in the shoot apex of Nicotiana glauca Groh. Virology 21:300-308. Stace-Smith, R., and A. J. Hansen. 1974. Occurrence of tobacco ringspot virus in sweet cherry. Canadian Journal of Botany 52:1647-1651. Stace-Smith, R., M. E. Reichmann, and N. S. Wright. 1965. Purification and properties of tobacco ringspot virus and two RNA-deficient components. Virology 25:487-494. Stanley, J., P. Rotteir, J. W. Davies, P. label, and A. vanKammen. 1978. A protein linked to the 5' termini of both RNA components of the cowpea mosaic virus genome. Nucleic Acids Research 5:4505-4522. Stavis, R. L., and J. T. August. 1970. The biochemistry of RNA bacteriophage replication. Annual Review of Biochemistry 39:527-560. Taylor, C. E., and W. M. Robertson. 1969. The location of raspberry ringspot and tomato black ring viruses in the nematode vector, Longidorus elongatus (de Man). The Annals of Applied Biology 64:233. Taylor, C. E., and W. M. Robertson. 1975. Acquisition, reten- tion and transmission of viruses by nematodes, pp. 253- 276. .In F. Lamberti, C. E. Taylor, and J. W. Seinhorst (eds.), Nematode vectors in plant disease. Plenum Press, London and New York, 460 pg. Thongmearkom, P., and R. M. Goodman. 1978. Complementation and pseudorecombination between ribonucleic acids from two natural isolates of cowpea mosaic virus (severe subgroup). Virology 85:75-83. Tremaine, J. H., and N. 5. Wright. 1967. Cross-reactive anti- bodies in antisera to two strains of southern bean mosaic virus. Virology 31:481-488. van Regenmortel, M. H. V. 1967. Serological studies on naturally occurring strains and chemically induced mutants of tobacco mosaic virus. Virology 31:467-480. 69. 70. 71. 72. 73. 130 Van Vloten-Doting, L., and E. M. J. Jaspars. 1977. Plant covirus systems: Three component systems. Ig_H. Fraenkel- Conrat, and R. R. Wagner (eds.), Comprehensive Virology 11, Regulation and Genetics of Plant Viruses. Plenum Press, New York and London, 348 pg. Walkey, D. G. A., and M. J. W. Webb. 1970. Tubular inclusion bodies in plants infected with viruses of the NEPO type. Journal of General Virology 7:159-166. Weber, K., and Osborn, M. 1969. The reliability of molecular weight determinations by dodecyl sulfate-polyacrylamide gel electrophoresis. Journal of Biological Chemistry 244:4406- 4412. Wood, H. A. 1972. Genetic complementation between the two nucleo-protein components of cowpea mosaic virus. Virology 49:592-598. Zaitlin, M., C. T. Duda, and M. A. Petti. 1973. Replication of tobacco mosaic virus. V. Properties of the bound and solubilized replicase. Virology 53:300-311. "1.1 11111117111111.1411