George K. Hirst

Isolation and characterization of the ribonucleoprotein of influenza virus

Abstract Treatment of the WSN strain of influenza virus with the non-ionic detergent nonidet P-40, followed by velocity gradient centrifugation, yielded a ribonucleoprotein (RNP) with a sedimentation constant of about 38 S. This 38 S RNP, which consisted of 10% RNA and 90% protein, contained the five pieces of RNA previously found by extracting the virus with phenol-SDS. The RNA in the 38 S complex, although less susceptible to RNase digestion than was 20 S RNA, could be completely degraded by prolonged treatment with the enzyme. The 20 S RNA could be obtained from the 38 S RNP by treating the RNP with polyvinylsulfate (PVS). The PVS replaced the RNA on the protein and the PVS-protein complex sedimented at 38 S. The protein portion of the RNP appears to be a single component as shown by gel electrophoresis. Electron micrographs are presented which show the structure of the RNP and which support conclusions drawn from the biochemical data.

Genetic recombination among influenza viruses: I. Cross reactivation of plaque-forming capacity as a method for selecting recombinants from the progeny of crosses between influenza A strains☆

Abstract Crosses carried out in vitro between a UV-inactivated influenza virus (strain WSN or a plaque-type recombinant thereof) and an active strain have yielded plaque-type recombinants of the following influenza A serotypes: WS, Melbourne, PR8, FM-1, S-1, Jap/305/57, and Swine. The same methods have failed to yield plaque-type recombinants with the following influenza A strains: Asian RI 5 + (filamentous strain of A 2 ), NWS (Stuart-Harris strain of neurotropic WS), strain “N” (nonpathogenic strain of fowl plague virus), and a strain of equine influenza. In addition the following myxoviruses gave negative results: influenza B (four strains), influenza C (strain 1233), hemadsorption virus, type 2 (strain Sendai), Newcastle disease virus (Hitchner strain), and mumps virus (strain Enders). Plaque-type recombinants were usually recombinant for a number of other characteristics, and when seven characters were examined a wide variety of combinations were found. The recombinants were stable on passage. In certain crosses the irradiated parent type was frequently rescued, and usually with no evidence of recombination in the several markers tested. The WSN strain was unique among influenza A strains in acting as a donor of plaque-forming capacity in recombination experiments, and this property, once conferred on other strains, was readily exchangeable in further experiments with diverse influenza A serotypes.

Temperature-sensitive mutants of influenza a virus: Isolation of mutants and preliminary observations on genetic recombination and complementation

Abstract The WSN strain of influenza A virus grows equally well in chick embryo fibroblasts (CEF) at temperatures between 34° and 39.5°. Stocks of this strain contain about 0.5% recoverable spontaneous mutants that are unable to grow or form plaques at 39.5° and that are unrestricted in growth at 34°. Chemical mutagenesis increases the proportion of these temperature-sensitive ( ts ) mutants by a factor of 3 to 4. Thirty-five induced and 6 spontaneous ts mutants which plaque at 39.5° with an efficiency of 10 −4 or less were selected for further study. High frequency recombination (about 5–20%) occurs with certain pairs of WSN ts mutants and recombinant progeny are characterized by their ability to produce distinct plaques in CEF cultures at both 34° and 39.5°. Primary yields from certain mixed infections carried out with ts mutants at restrictive temperatures show enhanced levels of parent type ts virus that in some cases represent 500-fold increases above the combined yields of ts virus obtained in single-infection controls. This enhancement effect is regarded as positive evidence for the occurrence of complementation. The WS strain of influenza A virus, which does not show a temperature restriction at 39.5° but which forms very faint plaques in CEF cultures at all temperatures of incubation, can interact with various WSN ts mutants to give a high proportion of recombinants that form distinct plaques at both 34° and 39.5°. Predictions are made regarding the behavior of WSN ts mutants in pairwise crosses based on present knowledge of the physical characteristies of the influenza genome.

Mechanism of influenza recombination: I. Factors influencing recombination rates between temperature-sensitive mutants of strain WSN and the classification of mutants into complementation-recombination groups

Mixed infection of chick embryo fibroblasts with ts mutants of WSN virus resulted in a number of reeombinant-yielding cells far in excess of what was predicted in terms of the adsorbed multiplicity of the plaque-forming virus employed. Some pairs of mutants had a strong tendency to form mixed aggregates which gave rise to recombinants by a single hit process. Reeombinant-yielding cells were also generated by a two-hit mechanism, in which case the yield of mixedly infected cells (recombinant yielders) exceeded the expectation (based on the number of plaque-forming particles adsorbed) by a factor of 30–60 times. These results reinforce the previously proposed conclusion that non-plaque-forming particles are the principal source of recombinants in mixed infection. The two-hit recombinant-yielding cells were obtained at exceedingly low virus inputs and were obtained in cells that were incubated solely at the restrictive temperature for the ts mutants employed. This undoubtedly means that the initial steps leading to recombinant production were the result of complementation, followed at maturation by an exchange of RNA pieces which in some cases gave wild-type particles. These results strongly suggest that non-plaque-forming virus has the capacity to enter cells and generate some syntheses such as the formation of messenger RNA and viral structural proteins. They also suggest that two non-plaque-forming particles, each bearing a mutation on a different RNA piece, may by means of complementation and recombination, give rise to wild-type virus. On the basis of complementation-recombination patterns a number of mutant strains have been classified into eight groups. Members of each group give high frequency complementation-recombination with each of the other groups. The physiological basis of the classification is being investigated.

Mechanism of influenza recombination: II. Virus aggregation and its effect on plaque formation by so-called noninfective virus☆

Abstract Normal wild-type WSN virus stocks contain a few small aggregates which can be partially separated out by velocity centrifugation and which have an infective capacity per unit of RNA which may be 10–20 times that of a single nonaggregated particle. Aggregates formed by addition of nucleohistone to the virus also have a much enhanced infective capacity. On the basis of these observations, together with experiments on recombination, it has been postulated that the assembly of RNA pieces at virus maturation is a random process and as a consequence most particles mature with an incomplete set, and hence are unable to produce infective virus. If two non-plaque-forming particles with a complete set of RNA pieces between them enter a cell at the same point, a process of complementation can ensue, followed by maturation of particles which may contain a full complement of RNA pieces and be fully infectious.

The single- and double-stranded RNA's and the proteins of incomplete influenza virus☆

Abstract The development of von Magnus virus (vMv) with a strain of influenza virus was investigated from the standpoint of the characteristics of its RNA, both the single-stranded within the particle and the double-stranded from the infected cells. The proteins on the particles were also examined. The amount of radioactively labeled RNA that could be isolated from vMv particles of the second and third passages decreased over that of the first passage. The amount of labeled double-stranded RNA which could be isolated from the cells also declined sharply in subsequent vM passages while the decline of labeled coat protein was not so marked. The most significant qualitative change was the absence of the fifth peak of single-stranded RNA from the virus of the second von Magnus passage.

Mixed Infection of HeLa Cells with Polioviruses Types 1 and 2.

Abstract Poliovirus obtained from mixedly infected HeLa cells (types 1 and 2) contains particles that are neutralizable by both type 1 and type 2 specific antisera. This alteration in phenotype is not transmitted to progeny virus and is believed to be due to phenotypic mixing. The progeny from most mixedly infected cells shows a high degree of phenotypic mixing (frequently 100%), and in some such cells there was no tangible evidence of the presence of one of the infecting genotypes. Mass yields from mixed infections produced the greatest proportion of phenotypic mixing in the first newly maturing virus. When HeLa cells are mixedly infected the amount of doubly neutralizable virus produced, the number of mixedly infected cells, and the number of cells producing phenotypically mixed virus all exceed theoretical expectations based on virus input by a factor of 2 to 14. Two factors are discussed which may contribute to this excess: one is the possible presence of HeLa cells which adsorb virus poorly or not at all and the second is a possible activating effect of plaque-forming on non-plaque-forming virus.

Experimental Production of Combination Forms of Virus.∗ IV. Mixed Influenza A-Newcastle Disease Virus Infections

Summary Mixed infection with the viruses of influenza A and Newcastle disease virus was found to result in the formation of some particles which had antigenically mixed surfaces, i.e., which contained components from both parent types. The phenotypic change was transitory and did not appear in the progeny of such a particle. It is suggested that this phenomenon is due to phenotypic mixing.

The experimental production of combination forms of virus: VI. Reactivation of influenza viruses after inactivation by ultraviolet light

Abstract A neurotropic and a non-neurotropic strain of influenza virus were used in reactivation experiments. Either strain, when inactivated by heat or ultraviolet light, was reactivable by the other strain. The inactive and the reactivating strains could be given simultaneously or the latter could be inoculated as long as 16 hours later in the egg and still obtain revival of the inactivated serotypes. Undiluted passage virus could not be reactivated by fully active virus. When M− was reactivated by W+ about half the reactivated strains were recombinant for virulence. This contrasts markedly with the M progeny from a cross between active W+ and M− and suggests that radiation damage may be repaired by recombination of genes between active and inactive particles. The stability of damaged virus in the egg also suggests that such particles do not progress far in the normal pattern of multiplication until an active agent is added.

The experimental production of combination forms of virus. V. Alterations in the virulence of neurotropic influenza virus as a result of mixed infection.

Abstract Mixed infection of the allantoic sac with W + (virulent WSN) and M − (avirulent Melbourne) viruses under conditions of a massive inoculum of a 1:1 M − W + ratio yielded only parent-type particles and serotypic heterozygotes. The heterozygotes on re-entering a cell segregated, yielding either M − and W + or M − and W − , each pair half the time. Specific anti-M serum neutralized these heterozygotes. Mixed infection with M + and W − viruses also yielded heterozygotes, and the segregants from them were M + and W + almost 100% of the time. With both pairs of infecting viruses, the recombinations were very asymmetric, since W virus showed changes with a much higher frequency than M virus. In two instances, M virus was isolated from a mixed M − -W + infection which had the ability to recombine with W − and give W + . These two strains were not virulent in mice by direct test. This and other evidence indicated that changes in neurotropic virulence were due to multiple factors.