Search

ABSTRACT Targeted high-throughput sequencing (HTS) has revolutionized the way we look at bacterial communities. It can be used for the species-specific detection of bacteria as well as for the determination of the microbiome and resistome and can be applied to samples from almost any environment. However, the results of targeted HTS can be influenced by many factors, which poses a major challenge for its use in clinical diagnostics. In this study, we investigated the impact of the DNA extraction method on the determination of the bacterial microbiome and resistome by targeted HTS using principles from metrology and diagnostics such as repeatability and analytical sensitivity. Sputum samples spiked with Acinetobacter baumannii, Klebsiella pneumoniae, and Pseudomonas aeruginosa at three different concentrations (103–106 cells/mL) were used. DNA was extracted from each sample on 2 separate days in three replicates each using three different extraction methods based on cetrimonium bromide, magnetic beads, and silica membranes. All three spiked bacteria were detected in sputum, and the DNA extraction method had no significant effect on detection. However, the DNA extraction method had significant effects on the composition of the microbiome and the resistome. The sequencing results were repeatable in the majority of cases. The silica membrane-based DNA extraction kit provided the most repeatable results and the highest diversity of the microbiome and resistome. Targeted HTS has been shown to be a reliable tool for determining the microbiome and resistome; however, the method of DNA extraction should be carefully selected to minimize its impact on the results.IMPORTANCEHigh-throughput sequencing (HTS) is one of the crucial new technologies that gives us insights into previously hidden parts of microbial communities. The DNA extraction method is an important step that can have a major impact on the results, and understanding this impact is of paramount importance for their reliable interpretation. Our results are of great value for the interpretation of sputum microbiome and resistome results obtained by targeted HTS. Our findings allow for a more rational design of future microbiome studies, which would lead to higher repeatability of results and easier comparison between different laboratories. This could also facilitate the introduction of targeted HTS in clinical microbiology for reliable identification of pathogenic bacteria and testing for antimicrobial resistance (AMR). As AMR is a major threat to public health, the improved methods for determining AMR would bring great benefits to both the healthcare system and society as a whole.

Detection and quantification of DNA biomarkers relies heavily on the yield and quality of DNA obtained by extraction from different matrices. Although a large number of studies have compared the yields of different extraction methods, the repeatability and intermediate precision of these methods have been largely overlooked. In the present study, five extraction methods were evaluated, using digital PCR, to determine their efficiency in extracting DNA from three different Gram-negative bacteria in sputum samples. The performance of two automated methods (GXT NA and QuickPick genomic DNA extraction kit, using Arrow and KingFisher Duo automated systems, respectively), two manual kit-based methods (QIAamp DNA mini kit; DNeasy UltraClean microbial kit), and one manual non-kit method (CTAB), was assessed. While GXT NA extraction kit and the CTAB method have the highest DNA yield, they did not meet the strict criteria for repeatability, intermediate precision, and measurement uncertainty for all three studied bacteria. However, due to limited clinical samples, a compromise is necessary, and the GXT NA extraction kit was found to be the method of choice. The study also showed that dPCR allowed for accurate determination of extraction method repeatability, which can help standardize molecular diagnostic approaches. Additionally, the determination of absolute copy numbers facilitated the calculation of measurement uncertainty, which was found to be influenced by the DNA extraction method used.