- Volume 12, Issue 2, 1971

Ten strains of avian influenza A viruses, naturally pathogenic for birds and grown in chick embryo allantois were compared by haemagglutination-inhibition and neuraminidase-inhibition tests, using chicken antisera. The viruses segregated into different cross-reaction groups in the two tests, emphasizing the need to include an assessment of neuraminidase antigens in the taxonomy of influenza A viruses. Four major groups of viruses were observed in the haemagglutination-inhibition test but several very low titre cross-reactions were constantly found outside these groups. Some of these reactions, not all of which were two-way, corresponded to anti-neuraminidase cross-reactions. The results of neuraminidase-inhibition tests, using a largely automated technique and a human serum fraction as substrate, confirmed our previous results obtained with a smaller substrate. The virus neuraminidases could be separated into five separate antigen groups, without cross-reactions between the groups.

Previous work on the serological relationships of avian influenza A viruses provided an opportunity to assess the protective effects of anti-neuraminidase anti-body in a single species system. Three viruses naturally lethal for the chicken (A/dutch, A/turkey/england/63 and A/tern/south africa/61) were selected so that two (A/dutch and A/turkey/england/63) had an antigenically similar haemagglutinin, two (A/turkey/england/63 and A/tern/southafrica/61) had a similar neuraminidase and two (A/dutch and A/tern/south africa/61) were unrelated by surface antigens. Each virus was used to immunize a group of 7-week-old chickens using one dose of killed virus followed by one dose of live virus. Each group was then divided into three subgroups and each subgroup challenged with one of the viruses. The results of cross-challenge with live virus were assessed objectively on mortality. (1) Immunization with any of the viruses protected the birds completely against the homologous virus. (2) Immunization protected the birds completely against a heterologous virus possessing a similar haemagglutinin. (3) Immunization gave partial protection against a heterologous virus possessing a similar neuraminidase and this protection was significantly greater (P = 0.01) than unimmunized controls when birds immunized with A/turkey/england/63 virus were challenged with A/tern/south africa/61. In birds immunized with A/tern/south africa/61 and challenged with A/turkey/england/63 the protection was approaching significance (P < 0.1 > 0.05). The results of testing the sera from the immunized birds confirmed the seriological relationships of the viruses.

In the absence of added template, the DNA polymerase of the feline sarcoma virus particle appeared to exhibit two-phase kinetics—a rapid reaction for about 4 min., followed by a slower reaction. Equilibrium sedimentation in Cs2SO4 density gradients showed the formation of RNA-DNA complexes within 12 sec., followed by the formation of DNA free of an RNA template. Virtually all of the DNA product formed at 4 min. and later was free of RNA template. About half of the 60 min. DNA product was double-stranded, since it was resistant to digestion by exonuclease I and eluted from hydroxyapatite columns at the salt concentration characteristic for duplex DNA. After denaturation, the 15, 30 and 60 min. DNA product sedimented at 6s and formed a broad band in CsCl gradients with an average buoyant density of about 1.725. Calf thymus DNA stimulated DNA polymerase activity of detergent-treated virus particle about 10- to 20-fold, indicating the presence of DNA-dependent as well as RNA-dependent DNA polymerase activity within the virus particle.

During serial passage of mouse embryo fibroblasts in vitro it was found that an established line of cells was more susceptible to murine sarcoma virus transformation than early passages of the cells from which it was derived. An explanation can be made in terms of the growth rates of the cells involved, although primary cells were more susceptible to murine sarcoma virus than would be expected from their rate of growth. Non-dividing (amitotic) cells were not susceptible to murine sarcoma virus transformation.

Bovine syncytial virus was found to replicate in BHK 21 cells. A thin-section study of the morphogenesis of the virus in BHK 21 cells indicated that virus precursor is formed in the nucleus and, in some cells at least, is liberated into the cytoplasm by breakdown of the nuclear envelope. Internal components were found in the nucleus and the cytoplasm and also embedded in the chromosomes of dividing nuclei. Virus internal components gained their envelopes by budding through one of a variety of cytoplasmic membranes, including the plasma membrane. Some stages of the morphogenesis of bovine syncytial virus in BESp cells, and of a type 1 simian foamy virus in BHK 21 cells, are also illustrated. It is concluded that bovine syncytial virus, a feline agent and simian foamy agents all belong to the same virus group.

The proteins of purified preparations of reovirus type 3 have been examined using polyacrylamide gel electrophoresis. Preparations of cell-associated virus grown with foetal calf serum show three major peaks, λ, µ, σ, and several minor peaks. Preparations of released virus grown without foetal calf serum show only the three major peaks, suggesting that the minor peaks are not virus proteins.

Analysis of preparations of released virus labelled with both [14C]lysine and [3H] leucine shows that each of three major peaks contains more than one polypeptide chain. The molecular weights of the polypeptides are: 140,000 to 150,000 for the λ group, 75,000 to 85,000 for µ and 36,000 to 44,000 for σ, as determined by migration in 5% sodium dodecyl sulphate-polyacrylamide gels. There are at least eight polypeptide chains distributed as two in peak λ, three in peak µ, and three in peak σ.

Top component particles isolated during purification of virus show the same three protein peaks as demonstrated in complete virus. Virus cores produced by trypsin or chymotrypsin treatment of the virus lack µ-group polypeptides and some σ polypeptides.

Radioactive glucosamine and fucose are not incorporated into any of the three groups of virus particle proteins during the growth cycle. The content of amino sugar or methylpentose in purified virus is less than 0.1% and that of neutral carbohydrates less than 1%. These results indicate that purified preparations of reovirus type 3 contain little if any carbohydrate. Such purified preparations agglutinate ox red blood cells.

A virus preparation, obtained from the mycelium of Penicillium chrysogenum, strain Q176, was purified by isopycnic centrifugation on linear gradients of potassium tartrate. Double-stranded RNA, prepared from the virus particles, revealed three components on electrophoresis on polyacrylamide gels.

The serum interferon levels in mice after intravenous injection of free virus RNA were higher and earlier than for purified virus particles, containing an equivalent amount of RNA. After reaching their maxima, the interferon activities declined at similar rates to very low levels after 45 hr for free RNA and 39 hr for virus particles.

The virulence of several early and serologically indistinguishable strains of the four original isolations of Semliki Forest virus was defined quantitatively through the response of mice, guinea-pigs and rabbits to intraperitoneal, respiratory and intracerebral administration of virus. There was a clear gradient of virulence against a corresponding gradient in the efficiency of the hosts’ protective responses. When the responsiveness of the host was critically balanced against the virulence of the invading virus a transitional state developed in which the groups of infected test animals showed a dual response in which some died and some were protected. This may have been due to the presence in some virus samples of subpopulations of extreme virulence and avirulence. Regardless of the ultimate response being death, incapacitation or benign protection, the efficiency of initial invasion and replication of virus was uniformly high for virus strains of extreme virulence or avirulence.

We have provided a specification of the virulence of virus and of the efficiency of the hosts’ protective responses which may be of value in the standardization of other studies on virus heterogeneity and the induction of the immune state.

Electron-microscopic studies of the morphology and development of a coronavirus (linder strain), isolated in human foetal diploid lung cells from a case of upper respiratory illness, revealed virus particles of diameter 80 to 160 nm. They were initially observed 12 hr after infection. Extracellular and intracytoplasmic virus concentration increased sharply at 18 hr and reached a maximum at 24 hr. The number of particles decreased slightly at 48 hr and by 72 hr many cells were lysed. The particles formed by budding into the cisternae of the endoplasmic reticulum or into an inclusion composed of tubular structures resembling part of the complete virus particle. There were cytoplasmic inclusions of dense particles within a granular matrix and surrounded by a double membrane. The release of particles by lysis is illustrated. Extracellular virus was specifically tagged with ferritin-labelled antibody.

The longer of the two types of particle of the cam strain of tobacco rattle virus is associated with mitochondria in the cells of Nicotiana clevelandii Gray leaves (Harrison & Roberts, 1968), and the RNA-producing defective isolate, cam/df, causes gross structural abnormalities in mitochondria (Harrison, Stefanac & Roberts, 1970). Thus mitochondria may play a special role in the synthesis or assembly of tobacco rattle virus.

In plant tissue, chloramphenicol is reputed to inhibit protein synthesis by chloroplast, and possibly by mitochondrial but not by cytoplasmic ribosomes, whereas cycloheximide is claimed to inhibit protein synthesis by cytoplasmic but not chloroplast ribosomes (Ellis, 1969; Boulter, 1970). We have therefore tested the effects of these and some other compounds on the accumulation of tobacco rattle virus, to see whether the results would shed any fresh light on the possible role of organelles in the replication of the virus.