Studies on Persistent Infections of Tissue Culture VI. Reversible Changes in Newcastle Disease Virus Populations as a Result of Passage in L Cells or Chick Embryos

Populations of the Victoria strain of Newcastle disease virus (NDV), reisolated from persistently infected L-cell cultures and passed twice in the embryonated hen's egg (NDV(L-E-2)), were found to differ strikingly from the original, chick embryo-adapted virus (NDV(o)). After exposure of L cells to NDV(o) at high multiplicities of infection, all cells became abortively infected; they produced only small aggregates of viral antigen and few, if any, infectious virus particles, but they yielded large amounts of interferon. No cytopathic effects (CPE) were noted, and the cultures survived readily as viral carriers. In contrast, NDV(L-E-2) yielded under similar conditions large quantities of viral antigen and infectious virus particles, but no detectable interferon, and the cultures were rapidly destroyed. This change in "virulence" was at least partially reversible by further serial passages of NDV(L-E-2) in chick embryos, as was evident from a consecutive decrease in CPE with a concomitant increasingly rapid recovery of the L-cell cultures, gradually diminishing yields of infectious viral progeny, and the returning of a capacity to induce interferon synthesis. Thus, NDV(L-E-16) resembled NDV(o) in many aspects, except for a less striking reduction in its ability to replicate in L cells. Although a selection of viral variants under the given sets of conditions has not been entirely excluded, the establishment of "avirulence" appears to be largely explained by a gradual accumulation of noninfectious, interferon-inducing components in the course of serial passages in the embryonated hen's egg, and the acquisition of "virulence" by a loss of these components. The evidence is as follows. (i) By a step-wise decrease in the dose of virus and restriction of the analyses to the first infectious cycle, a multiplicity of infection was ultimately reached for all "avirulent" populations at which infected cells produced normal yields of infectious viral progeny; i.e., the interferon-inducing components were diluted to noneffective levels. The lowest multiplicity which resulted in a measurable reduction in infectious virus replication was also the last one to induce detectable interferon synthesis. (ii) All viral clones derived from "avirulent" populations behaved like NDV(L-E-2) rather than like the parent viral suspensions, except that some of them elicited small amounts of interferon in L cells. The interferon-inducing components were reduced or lost in the cloning procedures. The nature of the interferon-inducing components has not been established. These components, which were neutralized by rabbit sera against "virulent" NDV(L-E-2) populations, may represent largely inactive or incomplete virus particles; however, the infectious virus-hemagglutinin ratios of "avirulent" populations were mostly of an order similar to those of "virulent" populations. The interferon-inducing components aborted the infectious process in cells simultaneously invaded by infectious virus particles. The implications of these findings are discussed.