Multiple JC virus genomes from one patient.

JC virus was previously isolated from the urine and from diseased brain of a patient with progressive multifocal leukoencephalopathy. The DNAs of these isolates, MAD-7 and MAD-8 respectively, were shown in this study to be mixtures of several different full-length genomes. By differences in their restriction endonuclease cleavage patterns, the genomes were categorized into two types. Type I DNA was defined as that typified by the prototype virus, MAD-1; Type II, that typified by viral DNA molecularly cloned from brain tissue of the MAD-7/8 patient (BrC-8 DNA). MAD-7 DNA was an approximately equal mixture of Type I and Type II; MAD-8 DNA appeared to be homogeneously Type II. These DNAs were molecularly cloned for further studies. Of 17 clones of recombinant MAD-7 (rMAD-7) DNA, nine were Type I and were identical to each other and to MAD-1 DNA. The other eight clones comprised five different species of Type II, none of which was identical to MAD-8 or to BrC-8 DNA. Among 17 recombinant MAD-8 (rMAD-8) clones there were two which contained Type I DNA, which had not been detected in the original viral DNA. The remaining rMAD-8 clones were of Type II. The differences among all these DNAs were primarily in the 'extra' PvuII sites and in the insertions and deletions in the hypervariable regulatory region of the genome (0.67 to 0.72 map unit). Possible explanations for multiple genomes in separate isolates from the same patient are discussed. It is concluded that the observed variation originated in vivo and not during isolation in vitro.