Comparison of tet(X4)-containing contigs assembled from metagenomic sequencing data with plasmid sequences of isolates from a cohort of healthy subjects

ABSTRACT Recently discovered tet(X) gene variants have provided new insights into microbial antibiotic resistance mechanisms and their potential consequences for public health. This study focused on detection, analysis, and characterization of Tet(X4)-positive Enterobacterales from the gut microbiota of a healthy cohort of individuals in Singapore using cultivation-dependent and cultivation-independent approaches. Twelve Tet(X4)-positive Enterobacterales strains that were previously obtained from the cohort were fully genome-sequenced and comparatively analyzed. A metagenomic sequencing (MS) data set of the same samples was mined for contigs that harbored the tet(X4) resistance gene. The sequences of tet(X4)-containing contigs and plasmids sequences were compared. The presence of the resistance genes floR and estT (previously annotated as catD) was detected in the same cassette in 10 and 12 out of the 12 tet(X4)-carrying plasmids, respectively. MS detected tet(X4)-containing contigs in 2 out of the 109 subjects, while cultivation-dependent analysis previously reported a prevalence of 10.1%. The tet(X4)-containing sequences assembled from MS data are relatively short (~14 to 33 kb) but show high similarity to the respective plasmid sequences of the isolates. Our findings show that MS can complement efforts in the surveillance of antibiotic resistance genes for clinical samples, while it has a lower sensitivity than a cultivation-based method when the target organism has a low abundance. Further optimization is required if MS is to be utilized in antibiotic resistance surveillance. IMPORTANCE The global rise in antibiotic resistance makes it necessary to develop and apply new approaches to detect and monitor the prevalence of antibiotic resistance genes in human populations. In this regard, of particular interest are resistances against last-resort antibiotics, such as tigecycline. In this study, we show that metagenomic sequencing can help to detect high abundance of the tigecycline resistance gene tet(X4) in fecal samples from a cohort of healthy human subjects. However, cultivation-based approaches currently remain the most reliable and cost-effective method for detection of antibiotic-resistant bacteria.

to whole-genome sequencing and comparative genomic analysis to determine their antibiotic resistance genes, plasmid sequences, and possible transmission clusters.This approach has allowed detailed insights into the genomic structure of many MDR bacteria by numerous studies, especially for the recently emerged tet(X) family-mediated tigecycline-resistant Enterobacterales (1,2).On the other hand, shotgun metagenomic sequencing (MS) characterizes the microbial communities in clinical and environmen tal samples via an unbiased culture-independent approach in which the total DNA of the samples is extracted and sequenced.Subsequently, the antibiotic resistance genes of interest can be further analyzed using the contigs assembled from metage nomic sequencing data to determine their presence in the samples, as well as the associated plasmid types and host species.Shotgun metagenomic sequencing may therefore complement culture-based whole-genome sequencing approaches for rapid identification of MDR bacteria, especially when cultivation of bacteria is difficult, or a high-throughput screen is required.
The recently identified tet(X) family tigecycline resistance genes are variants of the initially described tet(X) (3).These new variants share 95% sequence identity with tet(X), which also confers high-level resistance to last-resort antibiotics such as tigecy cline (4), posing a serious threat to public health (5).Among the various emerging tet(X) variants, tet(X4) has been identified in animals, healthy individuals, and patients in multiple provinces of China and other regions (6), and its successful transmission could be attributed to conjugative plasmids and ISCR2-mediated transposition (7).We previously reported that the prevalence of Tet(X4)-producing Enterobacterales in the gut microbiota of healthy individuals in Singapore is 10.1% and analyzed the sequences of two IncI1-type plasmids (p2EC1-1 and p94EC-2) that carry tet(X4) (1).Here, we further sequenced and characterized additional 12 tigecycline-resistant Enterobacterales strains isolated from human fecal samples in Singapore.We show that tet(X4) is associated with a diverse range of plasmid types and hosts and is potentially co-transferred with florfenicol resistance gene floR and tylosin resistance gene estT.The latter has only recently been characterized as a serine-dependent macrolide esterase (8,9).We further leveraged on recently published high-quality metagenomic sequence data for the same fecal samples collected from the cohort to evaluate if contigs assembled from metage nomic sequencing data could reveal tet(X4) plasmid sequences (1,10).Our findings suggest that metagenomic sequencing could complement culture-based surveillance for MDR bacteria if they are present at high abundance in clinical samples.

Sample collection and DNA extraction
The collection and DNA extraction of fecal samples have been described previously (1,10).In brief, feces from 109 individuals aged 48-76 years old of the Singapore Integrative Omics Study were collected in 2018 using a BioCollector (BioCollective) kit, according to the manufacturer's instructions.Fecal samples were handled in a Coy anaerobic chamber containing N 2 (75%), CO 2 (20%), and H 2 (5%) gas mixture.Homogenized samples were transferred to 50-mL screw-cap tubes prior to storage at -80°C.The QIAamp Power Fecal Pro DNA kit was used to extract gDNA for genomic (2 × 2 mL pure culture, OD 600 = 0.17) and metagenomic (fecal material, ~0.5 g) sequencing.DNA for genomic sequencing was further purified using a Qiagen Genomic-tip 20/G kit as described in the manufacturer's protocol (Qiagen, Germany).Cells from cultures were concentrated at 10,000 × g for 15 min before DNA extraction.DNA was quantified using a Qubit v.1.0fluorometer with a broad-range assay kit (Life Technologies) and a NanoDrop-2000 (Thermo Fisher Scientific).

Metagenomic sequencing assembly and analysis
MS contigs are derived from the Singapore Platinum Metagenomes Project (SPMP) (10), which was conducted on DNA extracted from the same fecal samples that were also used for the cultivation-based analysis.Contigs containing the tet(X4) gene were identified using BLAST, and subsequent verification was performed using ResFinder with default settings (16,20).

CFU counting
Colony-forming unit (CFU) counting experiment was done for our previous study (1).Briefly, frozen fecal samples were weighed and inoculated into Luria broth, followed by incubation at 37°C with 200-rpm shaking for 3 h.The fecal suspensions were then serially diluted in 0.9% NaCl and spotted onto MacConkey agar plates supplemented with 2-mg/L eravacycline dihydrochloride.The CFUs were enumerated after incubation at 37°C for 18 h, and the results were normalized to CFU per gram of input fecal sample.

GenBank accession numbers
MS short and long reads can be found under BioProject number PRJEB49168, and genomes sequences can be found under BioProject number PRJNA599529.

Evaluation of shotgun metagenomic sequencing in detection of tet(X4)-car rying plasmids
In total, 11 fecal samples contain tet(X4)-positive Enterobacterales, and the tet(X4)-carry ing plasmid sequences were analyzed in this study (Fig. 1) and in our previous study (1).To assess if shotgun metagenomic sequencing could detect tet(X4)-carrying plasmids, we further screened the contigs assembled from shotgun metagenomic sequencing for tet(X4).Interestingly, we found that tet(X4)-harboring contigs can only be detected in two fecal samples (subject SPMP-39 and SPMP-94).The sizes of the tet(X4)-harboring contigs (14-33 kbp) were shorter than the plasmids (101-134 kbp).A comparison of the tet(X4)-harboring contigs with the plasmid sequences revealed high homology of the contigs to the plasmid sequences (Fig. 2).This finding indicates that shotgun metage nomic sequencing may potentially aid in the detection of tet(X4) and its surrounding genetic environment.
Enterobacterales is often present in low abundance in the human gut, which may potentially result in lower sensitivity for the detection of its associated antibi otic resistance genes when using shotgun metagenomic sequencing.We therefore wondered if the detection of tet(X4)-carrying contigs from shotgun metagenomic sequencing data is related to the abundance of the tet(X4)-positive Enterobacterales in the fecal samples.Interestingly, out of the three tet(X4)-harboring contigs identified, two were detected in subject SPMP-94, who uncoincidentally has a much higher CFU count -by four orders of magnitude-than subject SPMP-39 and the other nine samples for which MS failed to detect tet(X4)-containing contigs (Fig. 2c).Thus, these results suggest that shotgun metagenomic sequencing could detect tet(X4)-harboring plasmids when the bacteria containing the plasmid are present in high abundance in clinical samples.
Taken together, we report that tet(X4) is associated with a broad range of plasmids and host bacteria in the gut of healthy Singaporeans and is closely associated with florfenicol resistance gene floR and tylosin resistance gene estT.By comparing the contigs assembled from shotgun metagenomic sequencing, we show that this approach could complement culture-based detection of tet(X4) plasmids in human fecal samples when present at higher abundance.Further optimization is required if metagenomic sequencing should be used to discover MDR from clinical and environmental samples.However, selective cultivation currently remains the most reliable and cost-effective approach for detection of antibiotic-resistant bacteria.