Study population
The study group included patients aged ≥ 18 years, who suffered from diarrhea, were hospitalized at the Baruch Padeh Medical Center Poriya, Israel, between January 2016 and April 2018, and were diagnosed with CDI via samples routinely collected and tested at the Clinical Microbiology Laboratory. The study was approved by the Helsinki Ethics Committee of the hospital (POR-0085-15) and every subject signed a consent form before enrollment.
All CDI cases were confirmed by stool examination for toxigenic C. difficile, identifying three targets: toxin B, binary toxin, and tcdC deletion using the GeneXpert C. difficile PCR assay (Cepheid, Sunnyvale, CA, USA).
Bacterial Isolation And Identification
All stool samples were inoculated on a selective growth medium – chromID™ C. difficile (CDIF) (bioMérieux, Durham, NC) and then incubated at 37 °C, under anaerobic conditions, for 48 h. C. difficile colonies appear asymmetric and black-colored. Final identification was performed by matrix-assisted laser desorption ionization-time of flight (MALDI-TOF) mass spectrometry, using the Bruker Biotyper system (Bruker Daltonics, Bremen, Germany) [32].
Multi-locus Sequence Typing
Total genomic DNA was extracted from all bacterial isolates using the QIAamp DNA Mini Kit (QIAGEN GmbH, Germany), according to the manufacturer’s instructions. MLST was performed in two stages, as previously described [33]:
A. Amplification
PCR amplicons for 7 housekeeping genes were obtained for each isolate using the primers shown in Table 1, and the qPCRBIO SyGreen Blue Mix Hi-ROX kit, on a real-time polymerase chain reaction (PCR) device (BioRad CFX96 Real-Time Detection System). Amplification conditions were: 95 °C for 2 min to ensure polymerase activation, followed by 40 cycles of [95 °C for 5 s, and 60–65 °C for 20-30s]. The amplification products were characterized by melting curve analysis subsequent to the amplification run.
B. Sequencing and sequence type (ST) determination
Prior to sequencing, the PCR products were purified and their nucleotide sequences on each DNA strand were determined using the amplification primers [33].
Genotyping was performed with an ABI PRISM® 310 Genetic Analyzer (Applied Biosystems), using the capillary electrophoresis method. The reaction mix (per sample) consisted of 1 µl PCR product, 12 µl Hi-Di formamide, and 0.5 µl GeneScan™ 500 ROX™ size standard. The data were analyzed using ChromasLite v2.01 and Sequencher v5.1.
Following sequencing of the 7 housekeeping genes found in each specific strain, the allelic numbers of each gene and the STs were assigned using the PubMLST C. difficile database (http://pubmlst.org/cdifficile/). The ST number was assigned for each specific combination of alleles. Using phylogenetic analysis, the sequences of the genes were used to assess the evolutionary relationship of the different strains, enabling the identification of clades, i.e., phylogenetic lineages.
Toxin detection
A. Toxin Gene Detection By Pcr
A multiplex real-time PCR assay was designed to target and detect the cdtA (binary toxin), tcdA and tcdB (toxin A and B, repsectively) genes in a single PCR run using specific primers [34]. For this purpose, 1 µl of the DNA sample tubes containing 24 µL reaction mixture (0.8 µM of each primer, 0.4 µM of each fluorophore probe, 6 mM MgCl2, 200 µM dNTPs, 1 IU Super-Hot Taq polymerase, and 1 × PCR buffer). The assay was run on the Bio-Rad CFX96 Real-Time System, as previously described. The cycler conditions were denaturation of the pre-amplified templates at 95℃ for 15 min, followed by 45 cycles of denaturation at 95℃ for 15 sec, and annealing and extension at 60℃ for 60 sec [34].
B. Toxin detection using a chromatographic immunoassay
The presence of C. difficile toxins in fecal samples was determined using the CerTest Clostridium difficile GDH + Toxin A + B combo card test kit (Certest Biotec, S.L, Spain), according to the manufacturer’s protocol. This is a chromatographic immunoassay for the simultaneous qualitative detection of Clostridium difficile antigens toxin A, toxin B, and glutamate dehydrogenase (GDH), an enzyme produced in large quantities by all strains [35].
Antibiotic susceptibility testing
Antimicrobial susceptibility testing (AST) was performed using the Etest quantitative technique, in order to determine the minimum inhibitory concentration (MIC) values, i.e., the minimal concentration (µg/mL) of a given antibiotic that inhibits the growth of a particular bacterium under specific experimental conditions. Once isolated, C. difficile colonies were suspended in thioglycollate broth medium (Becton Dickinson, Heidelberg, Germany) until 1.0 McFarland turbidity. Using a sterile cotton swab, the inoculum was spread on Brucella blood agar growth medium (Hy Laboratories, Rehovot, Israel). A gradient Etest strip (bioMérieux, Durham, NC) of the following antibiotics was added to the plates: vancomycin, metronidazole, moxifloxacin, and tigecycline. The plates were incubated under anaerobic conditions, at 37 °C, for 24 h. MIC was determined for each antibiotic. The interpretation of test results (sensitive/resistant) was in accordance with the European Committee on Antimicrobial Susceptibility Testing (EUCAST) recommendations [36].
Disease Severity Scoring And Demographic Data Collection
Disease severity was scored using the severity score index (SSI), developed by Velazquez-Gomez et al. [37]. This method, each of the following criteria are assigned one point: altered mental status, abdominal pain or distention, tachycardia (≥ 110 beats/min), hypotension (mean arterial pressure < 65 mmHg), fever (≥ 38.3 °C), leukocytosis (WBC > 20,000) or leucopenia (WBC < 1500), or increased band forms (> 10% bands); ascites or colitis documented by imaging studies; hypoalbuminemia (< 2.5 mg/dL); admission or transfer to an intensive care unit. CDI with an SSI of 1–3 is defined as a mild disease, 4–6 is considered moderate, and ≥ 7 is considered severe.
The following demographic data were retrospectively collected from patient medical records: gender, age, community versus nosocomial-acquired CDI, and death during hospitalization.
Gut Microbiome
A. DNA extraction
DNA was extracted from stool samples using the Power Soil DNA Isolation Kit (MoBio,USA), according to the manufacturer’s instruction, with an initial step of bead-beating for 2 min.
B. DNA amplification
The variableV4 region of the bacterial 16S rRNA gene was amplified by PCR, using the 515F and 806R barcoded primers, according to the Earth Microbiome Project protocol [38].
A unique barcode sequence was applied to each sample to distinguish between them. For each PCR tube, the following materials were added: 2 µl 806R (reverse, 10 µM) primer, 2 µl 515F (forward, 10 µM) primer, 17 µl ultra-pure water, 25 µl Primestar max PCR mix (Takara, USA), and 4 µl DNA. The reaction conditions were: 30 cycles of denaturation (98 °C for 10 sec), annealing (55 °C for 5 sec), and extension (72 °C for 20 sec), with final elongation at 72 °C (for 1 min).
C. DNA purification
The PCR products were purified using AMPure magnetic beads (Beckman Coulter, USA), according to the manufacturer’s protocol [39].
D. DNA quantification and pooling
Purified DNA was quantified using the Qubit dsDNA HS assay (Thermofisher, USA), according to the manufacturer’s protocol [40]. Following the measurement, samples were pooled at equal concentrations (50 ng/µl).
E. Pooling purification
Prior to sequencing, the DNA was further purified; 2% E-Gel agarose was inserted in an E-Gel PowerBase device. DNA fragments were cut and purified from the gel, using NucleoSpin® Gel and PCR Clean-up kit (Macherey-Nagel, Germany), according to the manufacturer’s protocol.
F. Sequencing
Purified DNA products were sequenced using the Illumina MiSeq platform, at the Genomic Center, Azrieli Faculty of Medicine, BIU, Israel.
G. Analysis
Data analysis was performed using QIIME2 [41]. To this end, paired-end sequences were joined, sequence reads were demultiplexed by per-sample barcodes, and Illumina-sequenced amplicon read errors were corrected using the Divisive Amplicon Denoising Algorithm (DADA2) [42]. Alpha and beta diversity calculations, describing diversity within and between samples, respectively, was performed using a feature table (species) containing features observed in at least 35 samples (50%), and on samples containing at least 8000 sequences. In addition, a phylogenetic tree was generated. For beta diversity, principal coordinate analysis (PCoA) was performed using both weighted and unweighted UniFrac distances [43]. Evenness and richness, two alpha diversity parameters, were calculated using Faith’s Phylogenetic Diversity and Pielou’s Evenness measures [44, 45].
In addition, LEfSe was performed [46], which identifies the features that are significantly different between samples according to relative abundances.