Thomas G. Mitchell, PhD
Associate Professor
Department of Molecular Genetics and Microbiology
Director, Molecular Mycology and Pathogenesis Training Program (MMPTP)

Overview. The AIDS epidemic and the expanded use of immunosuppressive therapies have accelerated a global increase in the incidence of systemic fungal infections. The most prevalent opportunistic mycoses are caused by species of Candida, Aspergillus, and Cryptococcus neoformans. As a result of this mycological crisis, there is a pressing need to improve the accuracy and speed of the diagnosis, to identify the sources of individual cases and outbreaks, to develop fungal targets for intervention, and to understand the genetics and distribution of populations of pathogenic fungi. Hence, there is a crucial need for rapid and accurate methods to (1) identify fungal pathogens, (2) determine the origin of a mycotic infection, (3) resolve the status of problematic species, (4) track the transmission of strains involved in nosocomial mycoses, (5) recognize strains with clinically important phenotypes, such as the expression of virulence factors or resistance to antifungal drugs, and (6) clarify the origin(s) of diversity and the population genetics of the major pathogens. The Medical Mycology Research Laboratory has developed DNA-based approaches to address these issues and focused on Candida species and C. neoformans.

C. albicans. To conduct rigorous population studies of Candida albicans, we developed methods for genetic typing, such as single-locus markers based on restriction fragment polymorphisms of amplified products (PCR-RFLP). Genotypic frequencies at several of these co-dominant, single-copy loci deviated significantly from random predictions, suggesting the existence of a mechanism for recombination. These studies confirmed that C. albicans is diploid and that substantial heterozygosity exists in natural populations.

C. neoformans. The epidemiological and evolutionary relationships among isolates of C. neoformans from nature and from patients are not well understood. Isolates from a variety of sources are being analyzed using defined genetic markers, such as PCR-RFLP, amplified fragment length polymorphisms, and single nucleotide polymorphisms identified by sequencing genes and microsatellites. These studies are designed to elucidate patterns of genetic variation, recombination, and migration among populations and to identify specific genotypes that represent lineages that have significantly diverged with respect to clinically relevant phenotypes.

Research Strategy. Regardless of the method used to assess virulence, most species of pathogenic microorganisms exhibit strain variation in virulence. There are three paradigmatic approaches to investigate microbial pathogenicity: (1) Researchers can focus on a small number of genes in a small number of laboratory strains, mutants, and constructs to isolate and elucidate a particular protein or pathway that is essential for virulence. This approach has been most successful for pathogens in which virulence factors are well-established and confined to a few genes. In fungi, pathogenicity is polygenic. While several properties are clearly essential for the virulence of a given fungal species, they are not necessarily sufficient. (2) Other methods, such as differential display, in vivo expression technology, and more recently, genomics and microarrays, have been used to identify multiple genes whose expression is regulated under conditions of infection. These studies also usually involve one or a few strains of the pathogen. (3) Our alternative approach is to analyze well-defined populations composed of numerous isolates to discover genotypes or lineages that are associated with enhanced (or decreased) pathogenicity and important clinical attributes, such as tissue tropisms or antifungal drug resistance. For example, we discovered that samples of C. albicans from individuals with HIV are dominated by a small number signature multilocus genotypes. In addition to predicting the pathogenetic potential of a clinical isolate, this information benefits other researchers who focus on particular strains and genes because it is crucial that they study isolates that represent the most clinically relevant strains. With the expanding sequence databases of pathogenic yeasts, we will be able to develop many more markers and refine the power of multilocus genotypes to discriminate among strains that differ in their pathobiology, and ultimately, to identify the responsible genes.