PPII Abstract, 2003

Pharmacogenetics of Phase II Drug Metabolizing Enzymes

Goals

The goal of this Pharmacogenetic Research Network (PGRN) program is the discovery and functional characterization of genetic polymorphisms and intragene haplotypes involving genes encoding phase II, conjugating, drug-metabolizing enzymes - with a special emphasis on enzymes that catalyze methylation and sulfation. To accomplish these goals, a genotype-to-phenotype strategy is applied that begins with the resequencing - using dye primer chemistry - of genes encoding phase II drug-metabolizing enzymes using DNA samples from subjects of differing, defined ethnicity, followed by a determination in mammalian cells of cellular phenotypes for all nonsynonymous coding single nucleotide polymorphisms (cSNPs) observed as well as polymorphisms within the 5'-flanking regions (5'-FRs) that might influence transcription. Phenotypes associated with these SNPs and intragene haplotypes are also confirmed - whenever possible - by performing genotype-phenotype correlation analyses with human tissue biopsies obtained during clinically-indicated surgery. Large numbers of human biopsy samples - together with a long history of experience with studies of the biochemistry, molecular biology and regulation of conjugation (phase II) enzymes are among the strengths of this PGRN program. Also included among those strengths is the presence at Mayo of large numbers of patients with virtually all known diseases who are available to test clinically-relevant pharmacogenetic hypotheses.

Progress

Application of this research strategy has already made it possible to resequence 20 genes encoding phase II drug-metabolizing enzymes - most often using 60 DNA samples from African-American and 60 DNA samples from Caucasian-American subjects that are available from the Coriell Institute. We are now also including 60 DNA samples from Han Chinese-American and 60 samples from Mexican-American subjects that only recently have became available from the Coriell Institute. All of these gene resequencing data have been submitted to PharmGKB.

The Mayo Bioinformatics group led by Dr. Chute has not only submitted the Mayo data to PharmGKB, but the Mayo group has also assisted other PRGN Centers with data submission. They have also developed a Bioinformatics Supplement involving several other PGRN Centers which is designed to develop a database ontology suitable for use with pharmacogenetic data. This group also performed database mining studies that have resulted in the identification of previously unknown human genes encoding sulfotransferase and small molecule methyltransferase enzymes - thus creating additional "target genes" for study.

Functional genomic studies of nonsynonymous cSNPs have been completed for 8 genes, and - as a result - a total of 22 variant allozymes have been characterized. Those data have been submitted to PharmGKB, and that submission process has also served as a useful model for the formatting and submission of "cellular phenotypic" data for other drug-metabolizing enzymes. The cellular phenotypes studied have included determinations of basal levels of enzyme activity and immunoreactive protein after the transfection of mammalian cells - corrected for transfection efficiency - as well as apparent Km values for both cosubstrates for the conjugation reaction. These data have demonstrated clearly that the most common mechanism by which function was altered involved a decrease in the level of enzyme protein - as a result of accelerated protein degradation in those cases in which we have studied the underlying mechanism. We have also discovered and functionally characterized polymorphisms within promoters for a smaller number of genes. These resequencing and functional genomic studies were made possible as a result of close collaboration between Dr. Wieben's and Dr. Weinshilboum's laboratories.

Our functional genomic studies of nonsynonymous cSNPs also led directly to experiments conducted by the Computational Biology group led by Dr. Pang and - more recently - the Protein Structure group led by Dr. Yee at Case Western Reserve University. Both of those laboratories are exploring the structural basis for the functional effects of nonsynonymous cSNPs which we have studied - as a possible result of their impact on protein folding or misfolding. Dr. Pang has used a homology modeling approach, and Dr. Yee is determining x-ray crystal structures for proteins selected on the basis of Mayo functional genomic data. The results of these studies have been or will be deposited in the Protein Data Bank. Finally, our phase II enzyme pharmacogenetic studies have already resulted in clinical collaborations at Mayo and elsewhere to test clinically relevant hypotheses involving breast cancer, pancreatic cancer, rheumatoid arthritis and allergy.

Conclusions

In summary, the Mayo PGRN program is utilizing a genotype-to-phenotype pharmacogenetic research strategy to discover and functionally characterize genetic polymorphisms and intragene haplotypes in genes that encode phase II (conjugating) drug-metabolizing enzymes - followed by the application of that information to test clinically relevant hypotheses. During the initial 3 years of funding, use of this approach has resulted in the publication of 16 manuscripts and an equal number of abstracts. This approach has clearly proven its value for the rapid accumulation and deposit of data in PharmGKB as well as the generation of mechanistic hypotheses which can be tested in greater depth by the Mayo Functional Genomic, Computational Biology and Protein Structure Groups - with all of these groups functioning within a highly integrated and focused Pharmacogenetics Research initiative.