In vivo fluconazole pharmacodynamics and resistance development in a previously susceptible Candida albicans population examined by microbiologic and transcriptional profiling

Abstract

Antimicrobial drug resistance can limit the ability to effectively treat patients. Numerous factors have been proposed to impact the development of antimicrobial resistance, including those specific to the drug and the dosing regimen. The field of investigation that examines the relationship between dosing regimen and outcome is termed antimicrobial pharmacokinetics and pharmacodynamics. Our prior in vivo investigations examined the relationship between fluconazole pharmacodynamics and the modulation of isogenic resistant and susceptible Candida albicans populations in a mixed-inoculum design (1). The goal of the current studies was to examine the impact of fluconazole pharmacodynamics on resistance emergence from a susceptible parent population over time using a murine systemic-candidiasis model. Both microbiologic and transcriptional endpoints were examined during the evolution of cell populations. As in our previous investigation, the more frequently administered dosing regimen prevented the emergence of a resistant cell phenotype. Conversely, dosing regimens that produced prolonged sub-MIC concentrations were associated with resistance development. The studies also demonstrated a striking relationship between fluconazole pharmacodynamic exposures and the mRNA abundance of drug resistance-associated efflux pumps. Global transcriptional profiling of cell populations during the progressive emergence of a resistance phenotype provides insight into the mechanisms underlying this complex physiologic process.

FIG. 1.

(Top) Burden of viable C. albicans with a resistance phenotype (based on growth on fluconazole-containing media) isolated from mouse kidneys following treatment with one of eight fluconazole treatment regimens. Each bar (A through J) indicates 72 h of therapy. Each bar represents data from two mice. (Bottom) Total burden of viable C. albicans based on growth on SDA plates. The error bars represent standard deviations.

FIG. 2.

Stability of the drug resistance phenotypes of three strains (J1, J4, and J6) following passage of the strains in vitro in the absence of fluconazole (Fluc). The black bars represent the total viable cell count. The gray bars represent viable growth of cells with a drug-resistant phenotype. The error bars represent standard deviations.

FIG. 3.

(a) In vitro growth of C. albicans in RPMI broth as estimated by the OD₆₀₀ measurement over time of the parent, C. albicans K1, and each of eight evolved isolates following 30 days of different fluconazole exposures in vivo. Each bar represents mean data from two independent cultures. (b) In vivo burden of the parent, C. albicans K1, and each of eight evolved isolates after 24 h. Each bar represents mean data from two mice. The error bars represent standard deviations. (c) Stabilities of the drug resistance phenotypes of three strains (J1, J4, and J6) following passage of the strains in vivo in the absence of fluconazole. Each bar indicates 72 h of growth in the mice (the bars from left to right represent consecutive periods of time). The top graph represents viable growth of cells exhibiting a resistant phenotype based on growth on fluconazole-containing agar. The bottom graph represents total viable-organism recovery based on growth on SDA plates. Each bar represents mean data from two mice.

FIG. 4.

(a) Categories and percentages of C. albicans genes up-regulated in the evolved cell population J4 compared to the parent, K1. (b) Categories and percentages of C. albicans genes down-regulated in the evolved cell population J4 compared to the parent, K1.

FIG. 5.

Relationship between fluconazole %T > MIC, 24-h AUC/MIC, and C_max/MIC and the burden of growth of C. albicans in kidneys of mice with a resistant phenotype based on growth on fluconazole-containing agar. Each bar indicates a 72-h in vivo drug exposure. Each bar represents mean data from two mice. The error bars represent standard deviations. R² is the coefficient of determination.

FIG. 6.

Relationship between in vivo fluconazole %T > MIC, 24-h AUC/MIC, and C_max/MIC and the in vitro MIC of the cell population after each 72-h treatment period. Each symbol represents the MIC of the entire isolated population after each 72-h period. The MIC value is the mean from three independent experiments.

FIG. 7.

Relationship between in vivo fluconazole %T > MIC, 24-h AUC/MIC, and C_max/MIC and the expression of CDR1 and CDR2 in C. albicans from infected mice. Each bar represents the change (n-fold) in mRNA abundance of an archived cell population relative to the beginning C. albicans K1 population. The bars represent the means and standard deviations from three biological replicates using quantitative RT-PCR. The RT-PCR assay was performed in triplicate for each biological replicate.