Various organisms have at least one enzyme that degrades the RNA of RNA-DNA hybrids. Such hybrids result in vivo from transcription and are often associated with DNA replication, even replication of retroviruses such as HIV. As far as we know, these ribonucleases H (RNases H) fall into two classes based on primary amino acid sequence similarity. Our studies have revealed that the well-characterized Escherichia coli RNase HI has homologs in many different species, including the human and the mouse. We know that, in both sequence and function, these mammalian proteins resemble the RNase H1 of Saccharomyces cerevisiae; they exhibit double-stranded RNA-binding activity in addition to the RNase H activity. Similarly, the bacterial RNase HII protein has counterparts in eukaryotes. RNases H perform essential cellular functions, and it is thus important to know how and by what means these enzymes are synthesized. For example, DNA drugs employed in oligonucleotide-based antisense therapy rely on endogenous RNases H to degrade certain disease-causing mRNAs (RNAs synthesized at inappropriate times or locations). The ability to increase RNase H activity in target cells could make these antisense DNAs more effective drugs. In a similar vein, certain types of drugs targeted to inhibit the RNase H activity of HIV reverse transcriptase could also inhibit the cellular enzymes, leading to undesired effects. Over the past year, we have made considerable progress in examining the regulation and activity of RNase H2 of S. cerevisiae and in learning more about the details of antisense DNA oligonucleotides. We previously reported that transcription of the RNH2L gene fluctuates as cells progresses through the cell cycle. We noted that a DNA element upstream of the gene is relatively rare in S. cerevisiae, occurring 110 times in the entire genome and only 29 times in a manner in which it can positively regulate transcription of the adjacent gene. The DNA element contains two overlapping DNA sequences recognized by the transcription factors responsible for expression in the S- and G2/M phases of the cell cycle, and we have demonstrated that both sequences are used to aid transcription of the RNH2L gene. Alterations in the DNA element sequence manifest themselves in modified expression patterns of other genes as well as in the expression of the RNH2L gene itself. Perhaps some of the other 28 genes with this overlapping promoter element are also regulated in a manner similar to RNH2L. The RNase H2Lp protein purified from S. cerevisiae has high levels of RNase H activity, yet the same protein expressed in E. coli is inactive. We find that other polypeptides copurify with the active form of the enzyme, suggesting that the active enzyme comprises multiple subunits. Cell-cycle regulation of the RNH2L gene expression, together with the differential expression due to the overlapping DNA sites, suggests that the protein may have different subunits at different stages of the cell cycle and that the subunits may participate in either DNA replication or DNA repair. The results indicate that induction or expression of the RNase H2Lp may not be sufficient for increasing RNase H activity associated with this polypeptide. In fact, when overexpression of the RNase H2Lp occurs, we have found only a modest increase in RNase H activity. To understand more fully how RNA-DNA hybrids are recognized by RNases H and thereby move us closer to antisense-based therapies, we have made several novel modifications to the DNA component. One type of modification added various substituents at the 3'-end of the DNA oligonucleotide. We observed differences in the site of cleavage of the RNA of the RNA-DNA hybrid, which revealed more information about the interaction of the RNase H with the RNA-DNA hybrid.