Mismatch Repair of DNA Replication Errors Contributes to Microevolution in the Pathogenic Fungus Cryptococcus neoformans

Abstract

The ability to adapt to a changing environment provides a selective advantage to microorganisms. In the case of many pathogens, a large change in their environment occurs when they move from a natural setting to a setting within a human host and then during the course of disease development to various locations within that host. Two clinical isolates of the human fungal pathogen Cryptococcus neoformans were identified from a collection of environmental and clinical strains that exhibited a mutator phenotype, which is a phenotype which provides the ability to change rapidly due to the accumulation of DNA mutations at high frequency. Whole-genome analysis of these strains revealed mutations in MSH2 of the mismatch repair pathway, and complementation confirmed that these mutations are responsible for the mutator phenotype. Comparison of mutation frequencies in deletion strains of eight mismatch repair pathway genes in C. neoformans showed that the loss of three of them, MSH2, MLH1, and PMS1, results in an increase in mutation rates. Increased mutation rates enable rapid microevolution to occur in these strains, generating phenotypic variations in traits associated with the ability to grow in vivo, in addition to allowing rapid generation of resistance to antifungal agents. Mutation of PMS1 reduced virulence, whereas mutation of MSH2 or MLH1 had no effect on the level of virulence. These findings thus support the hypothesis that this pathogenic fungus can take advantage of a mutator phenotype in order to cause disease but that it can do so only in specific pathways that lead to a mutator trait without a significant tradeoff in fitness.IMPORTANCE Fungi account for a large number of infections that are extremely difficult to treat; superficial fungal infections affect approximately 1.7 billion (25%) of the general population worldwide, and systemic fungal diseases result in an unacceptably high mortality rate. How fungi adapt to their hosts is not fully understood. This research investigated the role of changes to DNA sequences in adaption to the host environment and the ability to cause disease in Cryptococcus neoformans, one of the world's most common and most deadly fungal pathogens. The study results showed that microevolutionary rates are enhanced in either clinical isolates or in gene deletion strains with msh2 mutations. This gene has similar functions in regulating the rapid emergence of antifungal drug resistance in a distant fungal relative of C. neoformans, the pathogen Candida glabrata Thus, microevolution resulting from enhanced mutation rates may be a common contributor to fungal pathogenesis.

Keywords: microevolution; mismatch repair; pathogens.

FIG 1

Two clinical isolates of C. neoformans display an elevated mutation rate due to mutations in the MSH2 gene. (A) Spontaneous 5-FOA-resistant colonies in a standard wild-type (KN99α) isolate, the C23 and C45 clinical isolates, and the C23 and C45 isolates with a wild-type copy of MSH2 introduced at an ectopic location. (B) Quantitative assessment of mutation rates in the five strains using fluctuation analysis of the spontaneous resistance to 5-FOA. The Lea-Coulson method of the median was used to estimate the number of mutations from the observed values of mutants from 20 independent parallel cultures.

FIG 2

Deletion of three MMR components elevates mutation rates. (A) Spontaneous 5-FU-resistant colonies in the wild-type, msh2Δ, msh2Δ MSH2⁺, mlh1Δ, mlh1Δ MLH1⁺, pms1Δ, and pms1Δ PMS1⁺ strains. (B and C) Quantitative assessment of mutation rates using fluctuation analysis based on spontaneous resistance to 5-FOA (B) and 5-FU (C). The Lea-Coulson method of the median was used to estimate the number of mutations from the observed values of mutants across 20 independent parallel cultures.

FIG 3

MMR mutants show an increased proportion of single-base-pair G-to-A transitions and an increased proportion of deletions in multiple types of homopolymeric tracts. (A) Schematic representation of the URA5 gene indicating the locations and types of spontaneous mutations generated in the 5-FOA-resistant isolates derived from the wild-type (WT), msh2Δ, mlh1Δ, and pms1Δ strains. Insertions are indicated with black arrows, deletions with gray arrows, transitions as solid lines, and transversions as dashed lines. Mutations are as follows: red, G/T; black, G/C; pink, C/A; green, G/A; blue, T/C; orange, C/T; purple, A/G; gray, A/T. (B) Percentages of URA5 insertions (black), deletions (dark gray), transversions (white), and transitions (light gray) in the WT, msh2Δ, mlh1Δ, and pms1Δ 5-FOA-resistant isolates. MMR mutants showed an increased proportion of single-base-pair transition mutations. (C) Percentage of transition types in URA5 from the WT, msh2Δ, mlh1Δ, and pms1Δ 5-FOA-resistant isolates. The transition mutations are indicated as follows; black, T to C; dark gray, C to T; white, G to A; light gray, A to G. MMR mutants show an increased proportion of single-base-pair G/A transition mutations. (D) Schematic representation of the FUR1 gene indicating the locations and types of spontaneous mutations generated in the WT, msh2Δ, mlh1Δ, and pms1Δ 5-FU-resistant isolates. Insertions are indicated with black arrows, deletions with gray arrows, and transitions as solid lines. Colors of transition mutations are as follows: G/A, blue; T/C, orange; C/T; purple, A/G. The three homopolymeric tracts in FUR1 are indicated as red boxes. (E) Graph showing the percentages of FUR1 with no mutations (white), large deletions (black), small deletions (hatched gray), small insertions (checkered gray), deletions in homopolymeric tracts (light gray), insertions in homopolymeric tracts (medium gray), and both deletions and insertions in homopolymeric tracts (dark gray) in the WT, msh2Δ, mlh1Δ, and pms1Δ 5-FU-resistant isolates. MMR mutants showed an increased proportion of deletions in homopolymeric tracts. (F) Graph of the percentages of FUR1 deletions in the (A)₆ homopolymer (light gray), deletions in the (A)₆ homopolymers and (T)₁₄ homopolymers (medium gray), deletions in the (C)₇ homopolymer (dark gray), deletions in the (C)₇ and (T)₁₄ homopolymers (spotted gray), insertions in the (C)₇ homopolymer (hatched gray), and insertions in the (C)₆ homopolymer and deletions in the (T)₁₄ homopolymer (white) in the WT, msh2Δ, mlh1Δ, and pms1Δ 5-FU-resistant isolates. MMR mutants showed an increased proportion of deletions in multiple types of homopolymeric tracts.

FIG 4

Deletion of MMR components causes only minor sensitivity to DNA-damaging agents and oxidative stress. Strains were cultured overnight in YPD medium, 10-fold serially diluted, and plated onto YPD medium with or without the following stress agents: 0.04% methyl methanesulfonate (MMS), 0.06% ethidium bromide solution (EtBr), 0.25 mM menadione, 0.25 mM paraquat dichloride hydrate (paraquat), 0.4 mM tert-butyl hydroperoxide solution (tBOOH), 5 mM hydrogen peroxide solution (H₂O₂). One set of plates was exposed to UV light (120 J m⁻²). Plates were incubated at 28°C for 2 days.

FIG 5

Disruption of the mismatch repair pathway leads to rapid microevolution of new phenotypes. Comparisons of the original wild-type strain and the msh2Δ, mlh1Δ, and pms1Δ mutants with 3 strains derived from independent passaging for approximately 600 generations with population bottlenecks every 90 generations were performed. (A) Strains grown at 37°C or at 28°C for 2 days on 5 mM hydrogen peroxide solution (H₂O₂), 0.25 mM paraquat dichloride hydrate (paraquat), 1.5 mg/ml calcofluor white (CFW), and 5 mg/ml Congo red (CR). (B) Melanization on l-DOPA medium. These experiments were repeated multiple times with consistent results.

FIG 6

Deletion of MMR components leads to rapid resistance to antifungal agents. (A) Fluconazole MICs determined by Etests. (B) Fluconazole Etests of the wild-type strain and the msh2Δ, mlh1Δ, and pms1Δ mutants showing spontaneously arising fluconazole-resistant colonies in the zone of clearing. (C) Quantification of the number of spontaneously resistant colonies arising on 72 µg/ml fluconazole (16× MIC) media. Asterisks indicate statistical significance determined using a two-tailed Student’s t test (***, P < 0.0005; **, P < 0.005). (D) Spontaneously resistant colonies arising on 72 µg/ml fluconazole differed in size. (E) Quantification of the number of spontaneously resistant colonies arising on 4.8 µg/ml amphotericin B (32× MIC). Asterisks indicate statistical significance determined using a two-tailed Student’s t test (*, P < 0.05). (F) Spontaneously resistant colonies arising on 4.8 µg/ml amphotericin B differed in size.

FIG 7

Two clinical mismatch repair mutants display increased spontaneous resistance to antifungal agents. (A) Fluconazole Etests of the C23 and C45 clinical isolates, showing MIC and spontaneously arising fluconazole-resistant colonies in the zone of clearing. (B) Quantification of the number of spontaneously resistant colonies arising on 72 µg/ml fluconazole (16× MIC). Asterisks indicate statistical significance determined using a two-tailed Student’s t test (***, P < 0.0005; *, P < 0.05). Lower panels illustrate colony formation on fluconazole.

FIG 8

Deletion of PMS1, but not MSH2 or MHL1, results in a reduction in C. neoformans virulence. (A) Percent survival of wild-type (WT), MMR mutant, and complementation strains in a murine inhalation virulence assay. (B) CFU levels measured from brain and lung tissue of mice at sacrifice. Asterisks indicate statistical significance determined using a two-tailed Student’s t test (*, P < 0.05).