Clostridioides difficile infection (CDI) is among the most common health care-associated infections in the United States, linked to approximately 30,000 deaths annually (1). Antibiotic exposure is a risk factor for CDI, and resistance to commonly used antimicrobials may contribute to the evolving epidemiology of C. difficile (1). The Centers for Disease Control and Prevention (CDC) performs surveillance through the Emerging Infections Program (EIP) to monitor CDI in the United States (https://www.cdc.gov/hai/eip/cdiff-tracking.html). We announce a panel of 30 community- and health care-associated EIP CDI isolates, representing the 10 most prevalent PCR ribotypes (RTs) collected in 2016, available through the CDC & FDA Antibiotic Resistance Isolate Bank (AR Isolate Bank, https://wwwn.cdc.gov/ARIsolateBank/).

Strain typing was performed using a standardized high-resolution capillary gel-based PCR ribotyping protocol (2) with an in-house curated standard profile library and 100% similarity threshold (BioNumerics 6.6.11; Applied Maths). Multiplex real-time PCR (3) was used to detect genes encoding cytotoxins A and B (tcdA, tcdB) and binary toxin (cdtA, cdtB). Deletions in anti-sigma factor tcdC were assessed by PCR fragment (3) and whole-genome sequencing (WGS) analysis. For WGS, genomic DNA was extracted from colonies cultured overnight on CDC anaerobic blood agar with vitamin K, hemin, and 5% sheep blood at 37°C under anaerobic conditions from 10% skim milk glycerol stocks. DNA was isolated using the Promega Maxwell 16-cell low elution volume DNA purification kit and the Maxwell 16 MDx instrument (Madison, WI). DNA was sheared using the Covaris ME220 focused ultrasonicator (Woburn, MA), and indexed libraries were prepared using the NuGEN Ovation ultralow system version 2 assay kit (San Carlos, CA) and the PerkinElmer Zephyr G3 next-generation sequencing (NGS) workstation (Waltham, MA). Libraries were analyzed using the standard-sensitivity NGS fragment analysis kit and fragment analyzer system (Agilent Technologies, Santa Clara, CA). Sequencing was performed using the MiSeq reagent kit version 2 (500 cycles) and MiSeq system (Illumina, San Diego, CA), generating 2 × 250-bp paired-end reads. Sequence data were analyzed using the QuAISAR-H pipeline (https://github.com/DHQP/QuAISAR_singularity, version 1.0.2, default settings, accessed May 2019), including removal of phiX reads using BBDuk (BBMap version 37.87, sourceforge.net/projects/bbmap/), removal of adaptors and read trimming using Trimmomatic version 0.36 (4), and de novo assembly using SPAdes version 3.13.0 (5). Sequence types were assigned using the software package MLST version 2.16 (6) and the pubMLST C. difficile multilocus sequence typing scheme (7). Antimicrobial resistance (AR) genes were identified using a nonredundant combined database of the ResFinder (last updated 1 May 2019 and accessed on 20 May 2019) and ARG-ANNOT (last updated in May 2018 and accessed on 20 May 2019) AR databases; 98% sequence identity and 90% sequence coverage were used. A full description of custom scripts and publicly available tools and versions utilized by QuAISAR-H is available at the GitHub site. Assembled genomes were submitted to the NCBI Prokaryotic Genome Annotation Pipeline for annotation. The Clinical and Laboratory Standards Institute (CLSI) agar dilution method (8) was used to determine MICs for clindamycin, ceftriaxone, meropenem, metronidazole, moxifloxacin, and vancomycin. The results are summarized in Table 1.

The 10 most prevalent RTs collected in 2016 were 027, 106, 002, 014, 020, 015, 056, 054, 019, and 078. All isolates harbored tcdA and tcdB, and 10/30 (33%) contained cdtA and cdtB. The six RT027 isolates contained an 18-bp deletion and a deletion at nucleotide position 117 in tcdC, resulting in a premature stop codon characteristic of the BI/NAP1/RT027 epidemic strain (9, 10). The two RT078 isolates contained a 39-bp deletion and C184T point mutation in tcdC, resulting in a premature stop codon associated with this hypervirulent ribotype (11).

No isolates displayed elevated MICs to vancomycin based on the CLSI epidemiological cutoff value (wild type, ≤2 μg/ml; non-wild type, ≥4 μg/ml), and all isolates were susceptible to metronidazole (12). No acquired genes conferring resistance to vancomycin or metronidazole were detected. A total of 28 isolates contained genes belonging to the vanG-like gene cluster, which is highly conserved among C. difficile strains, but their presence is not indicative of vancomycin resistance (13). Recently, investigators suggested that mutations in the regulatory genes vanS and vanR may be associated with vancomycin resistance in RT027 isolates (14). No RT027 isolates in this panel contained known mutations in vanS or vanR; however, 3/3 RT002 and 1/3 RT014 (AR-1090) isolates contained the VanS Thr349Ile mutation described by Shen et al. (14) but had a vancomycin MIC of 0.5 μg/ml (Table 1). Five isolates with a clindamycin MIC of >16 μg/ml harbored ermB genes, which are associated with macrolide-lincosamide-streptogramin B resistance (15, 16). Two of these isolates also contained the cfrC gene, which is associated with resistance to phenicols, lincosamides, oxazolidinones, and streptogramins (17, 18). While all isolates contained the wild-type gyrB allele, seven with a moxifloxacin MIC of ≥8 μg/ml harbored an amino acid substitution (Thr82Ile) in the gyrA gene product associated with fluoroquinolone resistance (13).

The “Clostridioides difficile EIP 2016” panel represents the first publicly available resource for highly characterized isolates, and additional panels may be included to provide a growing source of contemporary isolates.