Characteristics of tobacco rattle virus

Summary

• 1. Inocula of TRV produce two types of primary infections: a) Local lesions containing infectious long and non-infectious short virus particles i.e. particulate or “complete” TRV (C-TRV), b) local lesions containing proteinless or “free” infectious RNA of the long particles i.e. “defective” TRV (D-TRV). Both types of lesions are morphologically indistinguishable and their quality needs to be determined in biological tests. The effect of dilution on the occurrence of both types of lesions was studied using artificial mixtures of highly purified long and short particles of a German isolate. The total number of all the lesions produced is directly proportional to the concentration of the infectious long particles present in the inoculum, thus resembling a single hit curve. But, the corresponding curves of the relative numbers of either C-TRV or D-TRV lesions resemble two-hit curves. The percentage of C-TRV lesions decreases by the square of the dilution, whereas the percentage of D-TRV lesions increases in a complementary way.

• 2. Highly purified long and short particles of TRV were used separately as well as in mixtures for inoculation, and the nature of the resulting local lesions was tested. About 94–100% of all the lesions produced by long particles alone were D-TRV lesions. The short particles did not produce any lesions nor did they multiply detectably. When short particles were added to preparations of homologous long particles neither the morphology nor the absolute number of lesions produced by the latter was affected. But the proportion of C-TRV lesions increased from zero to unity with increasing concentration of the short particles.

• 3. When non-infectious short particles or their RNA were added to inocula prepared from continuous cultures of homologous “proteinless” D-TRV particulate C-TRV was found to be synthesized in a large number of the resulting local lesions. Such C-TRV cultures as obtained by homologous complementation resembled the parent isolate in all of their properties.

• 4. Heterologous complementation was carried out using two strains of TRV (TRV-GER and TRV-USA) which are clearly characterized by several genetic markers such as particle size, lesion morphology and coat protein specificity. When the proteinless D-TRV of one strain is complemented with the short particles of the heterologous strain or their RNA, a “mixed” C-TRV is obtained. This “mixed” virus is composed of the long particles resembling the ones of the strain from which the D-TRV originates and of the short particles of the corresponding heterologous strain as used for complementation. The morphology of the local lesions produced by the “mixed” TRV is determined by the RNA of the long particles. Therefore, they resemble the lesions of the C-TRV strain from which the complemented D-TRV originates. Serological tests showed, however, that the “mixed” virus is only distantly related antigenically to this C-TRV isolate. But the “mixed” C-TRV is serologically indistinguishable from the strain from which the complementing short particles were derived.

• 5. If taken together the four types of evidence presented show, that in the TRV system a full cycle of viral replication is based on the complementing interaction of the RNA of an infectious long and a non-infectious short particle. The infectious “long” RNA appears to be specialized in early functions of viral replication only. It is able to initiate the synthesis of infectious “long” progeny RNA but this RNA remains uncoated by a viral coat protein. The non-infectious “short” RNA, however, appears to be specialized only in late functions and carries the coat protein gene of TRV. For its own replication it is dependent on the replicating “long” RNA. The implications of these findings are discussed.