Search

Herpesvirus genomes are often characterized by the presence of direct and inverted repeats that delineate their grouping into six structural classes. Class D genomes consist of a long (L) segment and a short (S) segment. The latter is flanked by large inverted repeats. DNA replication produces concatemers of head-to-tail linked genomes that are cleaved into unit genomes during the process of packaging DNA into capsids. Packaged class D genomes are an equimolar mixture of two isomers in which S is in either of two orientations, presumably a consequence of homologous recombination between the inverted repeats. The L segment remains predominantly fixed in a prototype (P) orientation; however, low levels of genomes having inverted L (I(L)) segments have been reported for some class D herpesviruses. Inefficient formation of class D I(L) genomes has been attributed to infrequent L segment inversion, but recent detection of frequent inverted L segments in equine herpesvirus 1 concatemers [Virology 229 (1997) 415-420] suggests that the defect may be at the level of cleavage and packaging rather than inversion. In this study, the structures of virion and concatemeric DNA of another class D herpesvirus, bovine herpesvirus 1, were determined. Virion DNA contained low levels of I(L) genomes, whereas concatemeric DNA contained significant amounts of L segments in both P and I(L) orientations. However, concatemeric termini exhibited a preponderance of L termini derived from P isomers which was comparable to the preponderance of P genomes found in virion DNA. Thus, the defect in formation of I(L) genomes appears to lie at the level of concatemer cleavage. These results have important implications for the mechanisms by which herpesvirus DNA cleavage and packaging occur.

To evaluate the utility of tetracycline gene regulation in the study of human cytomegalovirus gene functions, expression of luciferase under the control of tetracycline-regulatable promoters was studied following transient plasmid transfections and from within recombinant human cytomegalovirus genomes. The tetracycline-regulatable promoter PhCMV*-1 contains sequences from the human cytomegalovirus ie1/ie2 promoter and seven upstream tet operator sites which bind the activator protein tTA only in the absence of tetracycline (Gossen and Bujard (1992). Proc. Natl. Acad. Sci. USA 89, 5547–5551). Two modifications of PhCMV*-1 were also studied: P1129, in which the tet operator sites were reduced from seven to one; and P1125, in which human cytomegalovirus sequences were replaced by adenovirus major late promoter and terminal deoxynucleotidyltransferase initiator sequences. In transient assays, PhCMV*-1 and P1125 exhibited modest differential regulation but were strongly activated by viral infection. P1129 exhibited less viral activation and narrower regulation. In the viral genome, PhCMV*-1 exhibited regulation up to 7-fold during late times of infection, whereas P1125 displayed nearly 100-fold regulation. Regulation of P1125 was fully reversed within 12 to 24 h of adding or removing tetracycline. These results suggest that P1125 may provide sufficient conditional expression to effectively regulate human cytomegalovirus late genes.

The 230-kb human cytomegalovirus genome is among the largest of the known viruses. Experiments to determine the genetic determinants of attenuation, pathogenesis, and tissue tropism are underway; however, a lack of complete sequence data for multiple strains and substantial problems with genetic instability during in vitro propagation create serious complications for such studies. For example, recent findings suggest that common laboratory strains Towne and AD169 passaged in cultured human fibroblasts are missing up to 15 kb of genetic information relative to clinical isolates. To establish standard, genetically stable genomes that can be sequenced, disseminated, and repeatedly reconstituted to produce virus stocks, we have undertaken to clone two variants of Towne, designated Towne long and Towne short (referred to asTownevarRIT3) (A. Skaletskaya et al., 2001 , Proc. Natl. Acad. Sci. USA 98, 7829–7834), and the pathogenic strain Toledo into bacterial artificial chromosomes (BACs). We further demonstrate the ease with which mutagenesis can be achieved by deleting 13.5 kb from the Toledo genome using a PCR-based technique.