Targeting the 16S rRNA Gene by Reverse Complement PCR Next-Generation Sequencing: Specific and Sensitive Detection and Identification of Microbes Directly in Clinical Samples

Abstract

The detection and accurate identification of bacterial species in clinical samples are crucial for diagnosis and appropriate antibiotic treatment. To date, sequencing of the 16S rRNA gene has been widely used as a complementary molecular approach when identification by culture fails. The accuracy and sensitivity of this method are highly affected by the selection of the 16S rRNA gene region targeted. In this study, we assessed the clinical utility of 16S rRNA reverse complement PCR (16S RC-PCR), a novel method based on next-generation sequencing (NGS), for the identification of bacterial species. We investigated the performance of 16S RC-PCR on 11 bacterial isolates, 2 polymicrobial community samples, and 59 clinical samples from patients suspected of having a bacterial infection. The results were compared to culture results, if available, and to the results of Sanger sequencing of the 16S rRNA gene (16S Sanger sequencing). By 16S RC-PCR, all bacterial isolates were accurately identified to the species level. Furthermore, in culture-negative clinical samples, the rate of identification increased from 17.1% (7/41) to 46.3% (19/41) when comparing 16S Sanger sequencing to 16S RC-PCR. We conclude that the use of 16S RC-PCR in the clinical setting leads to an increased sensitivity of detection of bacterial pathogens, resulting in a higher number of diagnosed bacterial infections, and thereby can improve patient care. IMPORTANCE The identification of the causative infectious pathogen in patients suspected of having a bacterial infection is essential for diagnosis and the start of appropriate treatment. Over the past 2 decades, molecular diagnostics have improved the ability to detect and identify bacteria. However, novel techniques that can accurately detect and identify bacteria in clinical samples and that can be implemented in clinical diagnostics are needed. Here, we demonstrate the clinical utility of bacterial identification in clinical samples by a novel method called 16S RC-PCR. Using 16S RC-PCR, we reveal a significant increase in the number of clinical samples in which a potentially clinically relevant pathogen is identified compared to the commonly used 16S Sanger method. Moreover, RC-PCR allows automation and is well suited for implementation in a diagnostic laboratory. In conclusion, the implementation of this method as a diagnostic tool is expected to result in an increased number of diagnosed bacterial infections, and in combination with adequate treatment, this could improve clinical outcomes for patients.

Keywords: 16S rRNA; NGS; RC-PCR; bacterial identification; clinical diagnostics; infectious disease; molecular diagnostics; reverse complement PCR.

FIG 1

16S rRNA gene RC-PCR technology. The schematic is adapted from data reported previously by Kieser et al. (13) and Coolen et al. (14). (A) In the master mix, two types of oligonucleotides are present, one of which contains a unique dual index (UDI), a sequence adapter, and a universal sequence. The second one is the RC primer, which contains an extension blocker, a universal sequence, and a reverse complement of the 16S rRNA gene target. During PCR, after the annealing of the universal sequences, a 16S rRNA gene-specific PCR primer is formed. (B) Regular PCR will be performed. (C) Amplicons will be formed, which are compatible with NGS using the Illumina platform. (D) The RC-PCR is performed on two separate plates, plates A and B. This is to increase sensitivity and minimize chimera formation during PCR. (E) The 16S rRNA RC-PCR design consists of 6 primer pairs covering the V1-V6 and V9 regions of the 16S rRNA gene, covering ±84% of the 16S rRNA gene. See also the supplemental methods and Table S4 in the supplemental material for more details about the design.

FIG 2

Representation of RC-PCR results for the mock community and clinical samples. (A and B) Representation of the results for laboratory-derived microbial community samples (A) and the results for the commercial microbial community standard (B). The sizes of the bars are based on the calculated abundances. (C) Abundance heatmap of species detected in 14 heart valves with a positive culture result. (D) Abundance heatmap of culture-positive and culture-negative clinical samples (n = 45). Abundance is calculated as the relative percentage of fragments of a given hit compared to all quality-filtered fragments. The added value of the automated taxonomic information retrieved from the RC-PCR Classifier is that it uses the curated SILVA database and taxonomy, enabling straightforward interpretation compared to NCBI BLAST performed using 16S Sanger sequencing.

FIG 3

(A) Comparison of identification results for culture-positive and culture-negative clinical samples using 16S Sanger sequencing and 16S RC-PCR. (B) Comparison of the identification results by 16S Sanger sequencing and 16S RC-PCR for clinical samples (n = 45) based on sample type.