The polymerase chain reaction has potential for use in the detection of small amounts of human papillomavirus (HPV) viral nucleic acids present in clinical specimens. However, new HPV types for which no probes exist would remain undetected by using type-specific primers for the polymerase chain reaction before hybridization. Primers corresponding to highly conserved HPV sequences may be useful for detecting low amounts of known HPV DNA as well as new HPV types. Here we analyze a pair of primers derived from conserved sequences within the El open reading frame for HPV sequence amplification by using the polymerase chain reaction. The longest perfect homology among HPV sequences is a 12-mer within the first exon of E1M. A region of conserved amino acids coded by the El open reading frame allowed the detection of another highly conserved region about 850 base pairs downstream. Two 21-mers derived from these conserved regions were used to amplify sequences from all HPV DNAs used as templates. The amplified DNA was shown to be specific for HPV sequences within the El open reading frame. DNA from HPVs whose sequences were not available were amplified by using these two primers. HPV DNA sequences in clinical specimens could also be amplified with the primers.

A technique for selecting herpes simplex viruses expressing the viral thymidine kinase (TK+) from a population of predominantly TK-viruses was developed. This was accomplished by infecting TKcells and incubating the cultures under a liquid overlay medium containing methotrexate. Since the TKcells survive in this medium for only a limited period of time, it was necessary to add fresh uninfected TKcells 48 h after infection. The technique allowed the detection and quantitation of the TK+ virus fraction in mixtures of TK+ and TK-viruses where the TK+ fraction was present in frequencies as low as 10-. It was also used to estimate reversion frequencies and to obtain and analyze TK+ revertants from TKmutant strains of herpes simplex virus type 1.

No abstract

The nucleotide string TAAAACGAAAGT is the longest perfect homology shared by all sequenced human papillomavirus genomes. This nucleotide string, which was also found to be highly specific for human papillomavirus genomes, shares the same genomic position in all viral types (5' end of the E1 open reading frame) and putatively codes in every case for the same amino acids. One possible evolutionary model was used to estimate the probability of random occurrence of the nucleotide string in 10 human papillomavirus genomes. It assumed that the universal string had been subjected to the same mutation rate as the entire E1 open reading frame. The estimated probability was found to be very low, suggesting that the conservation of the string could not have resulted from random divergence and that its conservation among human papillomaviruses is likely to reflect the occurrence of biological constraints. It is speculated that this nucleotide string may be required to code for amino acids indispensable for the nuclear localization of E1-coded peptides or to bind cellular factors affecting viral replicative functions. Definitive evidence is expected to come from oligonucleotide-protein binding experiments and from site-directed mutagenesis of cloned HPV genomes. This motif, universal among human papillomaviruses, is being successfully used in the design of consensus primers from the early region.

No abstract