Molecular epidemiology of clinical Cryptococcus neoformans strains from India

Abstract

Little is known about the molecular epidemiology of the human pathogenic fungus Cryptococcus neoformans in India, a country now in the midst of an epidemic of AIDS-related cryptococcosis. We studied 57 clinical isolates from several regions in India, of which 51 were C. neoformans var. grubii, 1 was C. neoformans var. neoformans, and 5 were C. neoformans var. gattii. This strain set included 18 additional sequential isolates from 14 patients. Strains were characterized phenotypically by measuring the polysaccharide capsule and by determining the MICs of standard antifungals. Molecular typing was performed by a PCR-based method using the minisatellite-specific core sequence (M13), by electrophoretic karyotyping, by restriction fragment length polymorphisms with the C. neoformans transposon 1 (TCN-1), and by URA5 DNA sequence analysis. Overall, Indian isolates were less heterogeneous than isolates from other regions and included a subset that clustered into one group based on URA5 DNA sequence analysis. In summary, our results demonstrate (i) differences in genetic diversity of C. neoformans isolates from India compared to isolates from other regions in the world; (ii) that DNA typing with the TCN-1 probe can adequately distinguish C. neoformans var. grubii strains; (iii) that TCN-1 sequences are absent in many C. neoformans var. gattii strains, supporting previous studies indicating that these strains have a limited geographical dispersal; and (iv) that human cryptococcal infection can be associated with microevolution of the infecting strain and by simultaneous coinfection with two distinct C. neoformans strains.

FIG. 1.

Electrophoretic karyotype patterns of Indian C. neoformans strains. The majority of strains exhibit pattern I or III. (SC is a Saccharomyces cerevisiae chromosomal DNA molecular weight marker).

FIG. 2.

RFLP patterns with the TCN-1 probe. Pattern B was the most common pattern for Indian strains. X denotes a C. neoformans var. gattii strain (AS 2557) that did not hybridize with the TCN probe. T is the only Australian C. neoformans var. gattii strain (AS1337) of 12 that hybridized with the TCN probe. (MW, 1-kb and 100-bp size markers.) The other letters denote RFLP patterns in C. neoformans strains.

FIG. 3.

(Left) A phylogenetic tree of 42 URA5 DNA sequences from C. neoformans serotype A and serotype D isolates was constructed using the neighbor-joining algorithm (p-distance method) and rooted to ATCC strain 24064. Each number over a branch indicates the percentage of 1,000 bootstrap replicates that support the phylogenetic branch. The bootstrap values of <40% are not shown. Note that Indian strains fall into separate clusters: one is more closely related to the New York strains, whereas the other strains cluster with the Brazilian strains. (Right) Bayesian PP, 98% identifies a similar group but keeps I57 and I14 as a separate group.

FIG. 4.

Differences in karyotype and RFLP pattern are shown between I114 and I117 and between I19-SM and I19-MC. Panel A represents coinfection with a serotype A strain (I117) (karyotype I, RFLP O pattern) and a serotype D strain (I114) (karyotype IX, RFLP A pattern). Panel B represents karyotype and RFLP changes between an SM variant (karyotype IV, RFLP N pattern) and an MC variant of I19 (karyotype III, RFLP M pattern). The immunofluorescence staining of the polysaccharide capsule of the two colony variants revealed differences in binding by MAb 12A1 to the SM (C) and MC (D) polysaccharide capsules. URA5 sequence analysis and sugar assimilation were identical.