Fate of polyoma virus in mouse and human cells
Frost, Eric H.E.
Our aim was to investigate some early events in the polyoma virus (Py) - cell interaction, namely the liberation of the input viral genome from its shell, using virus labelled in its protein moiety. It was apparent from the start that biologically meaningful data could only be obtained if relatively low multiplicity of infection were used, requiring radioactive virus preparations of high specifie activity. We therefore developed a modified iodination technique using chloramine T. This was characterized by the use of a lower concentration of chemicals, addition of bovine serum albumin (BSA) in the viral sample to be iodinated, and a rapid virus purification after iodination allowing its use promptly after labelling. Using this modified technique, we labelled virus with either 125I or 131I to specifie activities 100 times greater than that which can be obtained with tritiated amino acids, without affecting the sedimentation coefficient, buoyant density or hemagglutinating activity of Py. The effect on plaque formation was slight. Adding dimethyl sulfoxide, (DMSO) to the chemical iodination mixture or using enzymatic iodination (lactoperoxidase) rather than chemical iodination did not significantly help preserve infectivity, while reducing the specifie activity of the virus preparation. Analysis by polyacrylamide gel electrophoresis (PAGE) suggested that no viral polypeptide could be iodinated more effectively using chloramine T rather than lactoperoxidase. This is surprising in view of the concept of a virus core and could mean that all Py polypeptides are exposed to the surface of the virion. In an appendix we report some evidence suggesting that some of the histone-like proteins of Py might be located outside the viral capsid. Cytoplasmic extracts from mouse embryo and Hep2 cells infected with radioactlvely labelled virus were analysed by velocity sedimentation. This revealed the existence of a newly recognized DNA-protein complex or subviral particle, termed FB, with unique properties; DNA/protein ratio superior to that of the virion, with reduced amounts of VP4 -6; sedimentation coefficient of 120 - 150 S; stable in the presence of IM NaC1, but unstable in 2.7M CsC1. A differently sedimenting species, DFC, was tentively identified in cytoplasmic extracts. It has the following characteristics: reduced amounts of VP4-6 and little or no DNA when conpared with virions: sedimentation coefficient of about 185 S; stable in the presence of IM NaC1 but unstable in 2.7M CsC1. Input viral DNA was found in the nucleus, and occasionally in the cytoplasm, of infected mouse embryo cells. It was apparently free of protein when analysed in sucrose gradients containing 1M NaC1. This "free nuclear DNA" never represented more than 1.5% of the Py DNA molecules present in the virus preparation used to infect the cells. This figure may be of significance, since the ratio of physical particles to infectious units is reportedly of 100 for Py (10). We propose a decapsidation scheme for Py (see Sketch p. 79). Virions enter the cell by viropexis. Decapsidation starts by loss of VP4 - 6, generating FB. The next step is transfer of DNA to the nucleus with a simultaneous reorganization of the capsid proteins into DFC. The viral DNA in the nucleus then permits infection to continue. Analysis of temperature sensitive mutants of Py has confirmed this scheme. The mutant ts-3 has an altered decapsidation pattern in which FB and DFC are found in reduced amounts, while "free nuclear DNA" is barely detectable. Although other possibilities remain, these observations are consistent with our hypothesis that FB and DFC are not mere degradation products of the input virus but result from a decapsidation process which is essential for the continuation of viral infection. While the early mutant ts-a and wild type were found to behave similarly as expected, ts-10 and ts-1260, two late complementing mutants were decapsidated far more readily while ts-3, as already mentioned, is hardly subjected to decapsidation. We can thus conclude that these last three mutants have defects in structural proteins rendering the virion more or less susceptible to uncoating. In human cells, similar subviral species were observed, although in différent proportions to those found in mouse cells. This shows that our proposed decapsidation scheme could also apply to Hep2 cells, but does not explain why these cells are non-permissive for Py.