Mutations in cytomegalovirus ul97 and dna polymerase: structural models of the effects on drug resistance

Alejo Erice, Thomas Grys, Hank Balfour, Jessica Bell, Ellis Bell

Research output: Contribution to journalArticlepeer-review

Abstract

Cytomegalovirus is a major cause of opportunistic infection and death in immunocompromised patients. Drugs used to treat CMV including foscarnet and ganciclovir whose targets both appear to be the DNA Polymerase. Ganciclovir is first phosphorylated by the UL97 gene product. While numerous mutations of both gene products giving rise to drug resistance have been described, structural models of these proteins or a biophysical basis for drug action or drug resistance is missing. Homology modeling of the DNA Polymerase, based on known polymerase three dimensional structures and the nucleotide sequences of 26 clinical isolates allows for a structure to be computed which provides a rational interpretation of both known mutations affecting drug resistance and for other mutations which are observed in the DNA sequence. Mutations cluster in three structurally distinct regions of the model: near the replication fork; near the nucleotide binding domain; and near the nuclease domain. Using a similar approach with the UL97 gene product it appears that this protein may be a multifunctional protein with both a kinase domain and a domain towards the C terminal which is highly homologous to various viral RNA Polymerase/3 subunits. Several functional domains in the protein have been modeled which provide a structural basis for ganciclovir resistance, and allow for putative assignment of the normal active site of this protein which appears to be a tyrosine dependent kinase.

Original languageEnglish (US)
Pages (from-to)A1438
JournalFASEB Journal
Volume11
Issue number9
StatePublished - 1997

ASJC Scopus subject areas

  • Biotechnology
  • Biochemistry
  • Molecular Biology
  • Genetics

Fingerprint

Dive into the research topics of 'Mutations in cytomegalovirus ul97 and dna polymerase: structural models of the effects on drug resistance'. Together they form a unique fingerprint.

Cite this