Mutants of sindbis virus II. Characterization of a maturation-defective mutant, ts103

Mutant ts103 is a minute-plaque former which grows very slowly at any temperature and produces, under optimal conditions, virus yields of 3–10% of those of the parental HR strain. It is slightly temperature sensitive with growth. It reverts at a frequency of 10^(−7) to 10^(−8) to large-plaque, temperature-independent strains and thus is possibly two mutations removed from the large-plaque HR strain of Sindbis virus. The very slow rate of virus production by ts103 appears to be due to a defect in the final stages of virus maturation, the budding of nucleocapsids through the plasma membrane to produce infectious virus. RNA synthesis after infection by the mutant appears to be normal. Nucleocapsids are produced in ts103-infected cells in amounts comparable to that produced in HR-infected cells, although a significant fraction of the mutant nucleocapsids sediment more slowly than HR capsids. Viral hemagglutinin appears in the cell surface earlier in ts103 infection than in HR infection. Electron microscopy of cells infected by ts103 reveals the presence of large amounts of nucleocapsids apparently in the process of budding. Yet the release of mature virus is delayed and the final yield of virus is much reduced in ts103 infection. Furthermore, ts103 budding occurs almost exclusively in virus-specific processes which are quite different in appearance from those found after infection by other strains of Sindbis virus. The virus-like particles produced during infection of chick embryo fibroblasts by this mutant have been examined by sedimentation velocity, isopycnic centrifugation, thermal inactivation at 56°, acrylamide-gel electrophoresis of viral RNA and viral protein, and electron microscopy. Some particles are produced during ts103 infection, which cosediment with Sindbis HR virus at 280 S, and are indistinguishable from HR by a number of other criteria, including isopycnic density, protein composition, sedimentation coefficient of the nucleocapsid, size of RNA, and specific infectivity. However, these 280 S ts103 particles are more sensitive to thermal inactivation at 56° than HR virions. Most of the virus-specific particles produced during ts103 infection sediment faster than 280 S and are heterogeneous in structure. These particles contain more than one nucleocapsid in a single envelope, and the tightness of packing of the nucleocapsids in the envelope is different from particle to particle. This leads to variability in isopycnic density, with particles denser or less dense than HR virus, as well as particles of HR density. In addition, this pleomorphism means that particles containing the same number of nucleocapsids may differ in sedimentation coefficient. These rapidly sedimenting ts103 particles contain small numbers of the nucleocapsids which sediment more slowly than capsids from HR virus, but most of the nucleocapsids are indistinguishable from those of HR. The protein composition of these multicored particles is very similar to that of the HR strain (as examined by acrylamidegel electrophoresis) although they do appear to contain slightly higher ratios of nucleocapsid protein relative to the glycoproteins. These multicored particles are fully infectious with a specific infectivity approaching 100%. The data are all consistent with the following hypothesis: ts103 has an altered nucleocapsid protein. During budding, ts103 capsids interact less strongly with viral glycoproteins in the cell surface than is the case for HR infection. This weakened binding results in a very slow rate of maturation and the production of a large fraction of multiploid particles. In addition some misassembled capsids arise which are unable to mature into normal-sized virions