Direct Fluconazole Disk Susceptibility Testing for Candida glabrata-Positive Blood Cultures

Optimizing fungal DNA extraction and purification for Oxford Nanopore untargeted shotgun metagenomic sequencing from simulated hemoculture specimens

Long-read metagenomics provides a promising alternative approach to fungal identification, circumventing methodological biases, associated with DNA amplification, which is a prerequisite for DNA barcoding/metabarcoding based on the primary fungal DNA barcode (Internal Transcribed Spacer (ITS) region). However, DNA extraction for long-read sequencing-based fungal identification poses a significant challenge, as obtaining long and intact fungal DNA is imperative. Comparing different lysis methods showed that chemical lysis with CTAB/SDS generated DNA from pure fungal cultures with high yields (ranging from 11.20 ± 0.17 µg to 22.99 ± 2.22 µg depending on the species) while preserving integrity. Evaluating the efficacy of human DNA depletion protocols demonstrated an 88.73% reduction in human reads and a 99.53% increase in fungal reads compared to the untreated yeast-spiked human blood control. Evaluation of the developed DNA extraction protocol on simulated clinical hemocultures revealed that the obtained DNA sequences exceed 10 kb in length, enabling a highly efficient sequencing run with over 80% active pores. The quality of the DNA, as indicated by the 260/280 and 260/230 ratios obtained from NanoDrop spectrophotometer readings, exceeded 1.8 and 2.0, respectively. This demonstrated the great potential of the herein optimized protocol to extract high-quality fungal DNA from clinical specimens enabling long-read metagenomics sequencing.

IMPORTANCE

A novel streamlined DNA extraction protocol was developed to efficiently isolate high molecular weight fungal DNA from hemoculture samples, which is crucial for long-read sequencing applications. By eliminating the need for labor-intensive and shear-force-inducing steps, such as liquid nitrogen grinding or bead beating, the protocol is more user-friendly and better suited for clinical laboratory settings. The automation of cleanup and extraction steps further shortens the overall turnaround time to under 6 hours. Although not specifically designed for ultra-long DNA extraction, this protocol effectively supports fungal identification through Oxford Nanopore Technology (ONT) sequencing. It yields high molecular weight DNA, resulting in longer sequence fragments that improve the number of fungal reads over human reads. Future improvements, including adaptive sampling technology, could further simplify the process by reducing the need for human DNA depletion, paving the way for more automated, bioinformatics-driven workflows.

Rapid Antifungal Susceptibility Testing Based on Single-Cell Metabolism Analysis Using Stimulated Raman Scattering Imaging

Rapid antifungal susceptibility testing (AFST) is urgently needed in clinics to treat invasive fungal infections with the appropriate antifungal drugs and to slow the emergence of antifungal resistance. However, current AFST methods are time-consuming (24-48 h) due to the slow growth of fungal cells and the methods not being able to work directly for clinical samples. Here, we demonstrate rapid AFST by measuring the metabolism in single fungal cells using stimulated Raman scattering imaging and deuterium probing. Distinct metabolic responses were observed in Candida albicans to different classes of antifungal drugs: while the metabolism was inhibited by amphotericin B, it was stimulated by azoles (fluconazole and voriconazole) and micafungin. Accordingly, we propose metabolism change as a biomarker for rapid AFST. The results were obtained in 4 h with 100% categorical agreement with the gold standard broth microdilution test. In addition, a protocol was developed for direct AFST from positive blood cultures. This method overcomes the limitation of slow growth in conventional methods and has the potential for the rapid diagnosis of candidemia and other clinical fungal infections.

Impact of the Disk Diffusion Test on Fluconazole De-Escalation in Patients with Candidemia

Disk diffusion (DD) is a simple antifungal susceptibility method for Candida. This study examined the impact of fluconazole DD testing on antifungal de-escalation. We enrolled patients with candidemia whose Candida isolates were tested for fluconazole susceptibility using DD between January 2019 and January 2020. The historical controls were patients with candidemia who underwent fluconazole susceptibility testing using the broth microdilution (BMD) method. Clinical data including antifungal therapy were analyzed. In total, 108 patients were enrolled. Most baseline characteristics were comparable between the groups. C. tropicalis was the predominant isolate (54.6%), followed by C. albicans (17.6%). The rates of antifungal de-escalation within 72 h were 25.9 and 9.3% in the DD and BMD groups, respectively (p = 0.023). The median time to de-escalation was 3 days in the DD group, versus 6 days in the BMD group (p = 0.037). The 14-day mortality rate and antifungal cost tended to be lower in the DD group. There were no differences in the length of hospital stay and treatment-related complications between the two groups. The agreement between the DD and BMD results was 90%. DD testing can be substituted for BMD to enhance antifungal de-escalation and antifungal stewardship.