Thomas Greenfield Mitchell
Associate Professor Emeritus in Molecular Genetics and Microbiology

Among patients with AIDS, leukemia or other cancers, organ or bone marrow transplants, and similar immunocompromising risk factors, the incidence of opportunistic mycoses and the number of different fungal pathogens are increasing dramatically. For many of these fungi, the definition of a species and the recognition of pathogen are highly problematic. Conventional methods of identification are based on morphological and physiological characteristics and are often time-consuming, difficult to interpret, and inconsistent. This laboratory is using DNA-based methods to (i) identify fungal pathogens, (ii) resolve taxonomic issues, (iii) facilitate epidemiological studies, (iv) recognize strains with clinically relevant phenotypes, such as resistance to antifungal drugs, (v) elucidate the origin(s) of diversity and the population genetics of the major pathogens, and (vi) explore functional genomics to identify virulence factors. We have developed reliable methods to genotype strains and are analyzing gene sequences to clarify the phylogeny of controversial taxa.

To conduct rigorous population studies of Candida albicans , we developed single-locus markers based on polymorphisms of PCR products. Genotypic frequencies and segregation patterns at these loci have confirmed that C. albicans is diploid and suggest that some form of recombination occurs in this "asexual" yeast. To investigate whether separate populations of C. albicans exist in disparate geographical locations, we compared strains collected from healthy and HIV-infected persons in U.S. and Brazil. Although a number of different genotypes were recognized at each location, the same multilocus genotype was prevalent among the clinical isolates, indicating a remarkable homogeneity among these populations.

We are using DNA-based methods to compare global isolates of Cryptococcus neoformans from patients with AIDS and other sources, to analyze the distribution and relatedness of strains, to identify genotypes of clinical importance, and to create linkage map of this pathogen. To determine the source of C. neoformans in patients, we developed a genetic markers to investigate the structure of clinical and environmental populations. With analysis of quantitative trait loci, specific genotypes will be identified that represent clones that have significantly diverged with respect to clinically relevant phenotypes, including susceptibility to antifungal drugs and the expression of virulence factors. We are investigating genomic evolution and phenotypic variation in natural populations of C. neoformans . These approaches will correlate genotypes with pathobiological phenotypes, leading to beneficial and predictive information about the epidemiology, diagnosis and prognosis of cryptococcosis in patients with AIDS.

Current Appointments & Affiliations

Contact Information

Some information on this profile has been compiled automatically from Duke databases and external sources. (Our About page explains how this works.) If you see a problem with the information, please write to Scholars@Duke and let us know. We will reply promptly.