https://orcid.org/0000-0002-0780-3004
Figures
Abstract
The transcriptomic profile in a biofilm model of ribotypes (RT) 001 and 027 associated with recurrent Clostridioides difficile infection (R-CDI) and not associated with recurrent (NR)-CDI was analyzed to identify genes that may favor the recurrence. Twenty strains were selected, 10 RT001 and 10 RT027. From each ribotype, 5 were R-CDI and 5 NR-CDI. Biofilm and nonadherent cells were prepared from each clinical isolate, and the RNA was extracted. RNA samples were pooled in 8 combinations implying ribotype, recurrence, and biofilm formation. Each pool was separately labeled with Cy3 dye and hybridized on a microarray designed for this study. Slides were scanned, analyzed, and gene expression was compared between unique and grouped pools using the Student’s t-test with Benjamini-Hochberg correction when appropriate. Validation was carried out by qRT-PCR for selected genes. Results: After comparisons of differentially expressed genes from both ribotypes of R-CDI strains (nonadherent cells vs. biofilm) and both ribotypes in biofilm (R-CDI vs. NR-CDI), we found 3 genes over-expressed and 1 under-expressed in common (adj. p ≤ 0.05). Overexpressed genes were CAJ70148 (a putative dehydrogenase), CAJ68100 (a secretion type II system protein from the GspH (pseudopilins) family), and CAJ69725 (a putative membrane protein); under-expressed was CAJ68151 (a segregation and condensation protein A). Because CAJ70148, CAJ68100, CAJ69725 and CAJ68151 were differentially expressed in biofilm in strains associated with R-CDI, they may support the biofilm favoring the recurrence of CDI. However, further studies will be needed for poorly studied genes.
Citation: Rubio-Mendoza D, Córdova-Fletes C, Martínez-Meléndez A, Morfín-Otero R, Maldonado-Garza HJ, Garza-González E (2023) Transcriptomic analysis of biofilm formation in strains of Clostridioides difficile associated with recurrent and non-recurrent infection reveals potential candidate markers for recurrence. PLoS ONE 18(8): e0289593. https://doi.org/10.1371/journal.pone.0289593
Editor: Rajesh P. Shastry, Yenepoya University, INDIA
Received: February 6, 2023; Accepted: July 23, 2023; Published: August 3, 2023
Copyright: © 2023 Rubio-Mendoza et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This research was funded by Mexico’s National Council for Science and Technology, CONACYT, grant 284042. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Clostridioides difficile is an opportunistic, gram-positive anaerobic bacillus responsible for the most common cause of diarrhea. C. difficile infection (CDI) is the most common healthcare-associated diarrhea and an important cause of morbidity and mortality [1]. In Mexico, an increase in the incidence of CDI was reported from 2012 to 2015, with a predominance of C. difficile ribotype (RT) 027 [2], which in addition to RT001, is the most common ribotype in this country [3]. When CDI recurs within the first 8 weeks of resolution of CDI symptoms, the episode is defined as recurrent CDI (R-CDI) [4, 5]. Several factors increase the risk of R-CDI, including successive episodes of disease (up to 65%) [1, 4], a previous history of CDI, and infection with the hypervirulent and epidemic strain C. difficile RT027 [6–9].
Some virulence factors have been associated with R-CDI, including sporulation and biofilm formation [10, 11]. Among sporulation evidence, mice infected with C. difficile spo0A knockout mutant treated with oral vancomycin did not develop recurrence over 100 days of follow-up. In contrast, mice infected with C. difficile wild type strains developed R-CDI on the third day of follow-up, suggesting that spores may be a persistence factor [12].
Several studies have shown that C. difficile produces a biofilm that may protect the spores from antibiotics, including vancomycin [11, 13–16]. Moreover, increased expression of sporulation pathway genes (sigH, spo0A) was found in biofilm of C. difficile [17].
Biofilm formation has been associated with R-CDI, but the underlying transcriptomic patterns are largely unknown. Therefore, we analyzed the transcriptomic profile in an in vitro biofilm model of C. difficile RT001 and RT027 associated with recurrent CDI (R-CDI) and non-recurrent CDI (NR-CDI) to identify a gene expression signature that may favor the recurrence.
Material and methods
Ethics statement
This study was performed with approval from the Ethics Committee of the Hospital Universitario Dr. José Eleuterio González (approval number BI22-00003). The need for informed consent was waived because the study did not involve any physical intervention and was solely using the routinely collected clinical isolates. All data was analyzed anonymously.
Strain collection, ribotyping and selection of clinical isolates
Clinical isolates were obtained from the Laboratory of Molecular Microbiology strain collection. Clinical isolates were recovered from patients with CDI or R-CDI, according to Surawicz et al. [4].
Bacterial DNA was extracted using the phenol-chloroform method [18]. As previously described, the intergenic region between 16S and 23S of rRNA genes was amplified by polymerase chain reaction (PCR) [19]. Strains C. difficile ATCC BAA-1805 (RT027) and C. difficile ATCC 9689 (RT001) were used as controls.
Twenty clinical isolates were selected: 10 from RT001 and 10 from RT027. From each ribotype, 5 isolates were from patients with R-CDI and 5 from patients with NR-CDI. Eighteen isolates were collected at the Hospital Civil of Guadalajara “Fray Antonio Alcalde” and two at the University Hospital Dr. José Eleuterio González).
Biofilm production
Biofilm was prepared as previously reported [13], with some modifications. From a 7-day growth in Schaedler agar supplemented with 5% blood in anaerobic conditions (from each isolate), a 0.5 McFarland suspension was adjusted and diluted 1: 100 in brain-heart infusion broth supplemented with 0.5% yeast extract and 0.1% L-cysteine (BHIS), 200 μl from each suspension were deposited on wells of a 96-well microplate (Sarstedt, Germany). Plates were incubated at 37°C in anaerobiosis, and after 72 h, the medium was discarded. Plates were washed thrice with 200 μl of distilled water; the biofilm was fixed with 200 uL of methanol for 15 min and washed thrice, dried for 1 h at 60°C and stained with 100 μl Hucker’s crystal violet solution for 15 min at room temperature. Next, each well was washed four times with 200 μl distilled water and left to dry overnight at room temperature. Finally, 100 μl 99% ethanol (Sigma-Aldrich, USA) was added and left for 30 min. The optical density at 595 nm was measured in a multimode reader Cytation 1 (BioTek Instruments, Inc., USA). C. difficile ATCC BAA-1805 (RT027, strong biofilm producer) and C. difficile ATCC 9689 (RT001, strong biofilm producer) were used as controls. Assays for each clinical isolate were repeated eight times, and the BHIS broth was included as a blank. Biofilm production was classified according to criteria described by Vukovic D et al. [20].
Biofilm and non-adherent cells for transcriptomic assays
Biofilm was prepared, as described in the previous section, but after incubation for 72 h, the medium was carefully discarded by sucking it through the cell wall. Eight replicates of each strain were pooled by scraping the bottom of the well with the tip of a micropipette, mixing with suction and ejecting with the micropipette, and depositing them in an Eppendorf tube. This mixture was analyzed as a biofilm.
For non-adherent cells obtention, biofilm was prepared as described, with incubation of 24 h instead of 72 h. After 24 h, 100 μl of the non-adherent cells in the supernatant of eight wells were carefully taken with a micropipette, grouped in an Eppendorf tube and analyzed as nonadherent cells.
RNA extraction
The RNA was extracted from each sample (biofilm or non-adherent cells, 8 wells from the plate grouped). Each sample was centrifuged at 14,000 rpm for 5 min and incubated with 200 μL Tris-HCl (10 mM, pH 8) and 50 mg/L of lysozyme at 37°C for 1 h. After incubation, proteinase K (10 mg/L) was added, and the tubes were incubated at 55°C for 1 h. RNA was extracted using the Trizol Max Bacterial RNA Isolation kit (Thermo Fisher Scientific, USA) and resuspended in 50 μL RNase-free water.
Pools for microarray analysis
Eight pools of RNA were prepared with the RNA from each of the 20 strains (biofilm or nonadherent cells, 40 RNAs in total). Each pool included 5 strains with the same characteristics and conditions to minimize the variation of one single strain.
Pools of RNA were prepared according to a format that included 3 conditions: biofilm or nonadherent cells, ribotype (001 or 027), and recurrence or not of CDI.
Each of the eight pools comprised RNA from five strains with the same characteristics as follows: pool 1: nonadherent, RT001, NR-CDI; pool 2: nonadherent, RT001, R-CDI; pool 3: nonadherent, RT027, NR-CDI; pool 4: nonadherent, RT027, R-CDI; pool 5: biofilm-RT001-NR, CDI, pool 6: biofilm, RT001, R-CDI; pool 7: biofilm, RT027, NR-CDI, and pool 8: biofilm, RT027, R-CDI.
The concentration of RNA from each pool was adjusted to 25 ng/μL with equal contribution of each sample.
Microarray design and processing of the microarrays
Expression microarrays were designed for this study from the complete transcriptome of the strains C. difficile 630 (RT012) (Gen Bank: GCA_000009205), C. difficile ATCC 9689 (RT001) (Gen Bank: GCA_001077535), and C. difficile R20291 (RT027) (Gen Bank: GCA_000027105). The designed microarray was arranged in an 8 × 15K microarray format (Agilent Technologies, Inc., USA) in which 3,858 common genes in all strains 192 genes were unique for the strain C. difficile ATCC 9689; 249 genes, for C. difficile R20291, were included. The Agilent one-color, microarray-based exon analysis low input quick amp WT labeling protocol (Agilent Technologies, Inc., USA) was followed according to the manufacturer’s instructions [21].
Each pool was separately labeled with Cy3 dye and hybridized on the designed microarray and shortly after being washed. Slides were scanned using the SureScan microarray scanner (Agilent Technologies, Inc., USA). Subsequently, data were extracted using the Agilent feature extraction software (Agilent Technologies, Inc., USA) [21].
Microarrays bioinformatics analysis
The data were processed using the gene expression analysis platform v.0.4 software (also known as GEAP) [22]. Quantile type normalization was performed with the Linear Models for Microarrays Data (LIMMA) tool, and the quality analysis was conducted with arrayQualityMetrics.
Gene expression of all eight pools was compared with LIMMA using the Student’s t-test with a statistical significance of ≤0.01 and a logarithmic fold change (LogFC) of 1.5. For multiple comparisons, the Benjamini-Hochberg correction was used.
Differentially expressed genes (DEGs) between pools were visualized with a heat map using GraphPad Prism v9.3.0. Software. Functional annotation was performed with the DAVID Knowledgebase Query Tool [23], and National Center for Biotechnology Information; protein interaction and enriched pathways were evaluated in the Gene Ontology (GO) and Encyclopedia of Genes Kyoto Genes and Genomes (KEGG) in String and ShinyGO v0.75 online tool [24].
Validation of microarray results by qRT-PCR
Microarray results were validated by qRT-PCR on some selected genes using primers designed in this study (z 1). Two housekeeping genes (rpsJ and rrs) were used for normalization [25]. Retrotranscription was performed using 1 μL of SuperScript III 200U/μL (Invitrogen, USA), 4 ng random primers (Invitrogen, USA), 400 μM dNTP mix, and 125 ng total RNA from each of the pools for a final volume of 25 μL; the manufacturer recommended conditions were followed [26].
For qRT-PCR, 1 U Taq DNA polymerase (Invitrogen, USA), 400 μM dNTPs, 3.5 mM MgCl2, 1× NH4 reaction buffer, 3.125 μL 12.5× SYBR Green I (Sigma Aldrich, USA), 200 nM of each primer (except for rrs, 50 nM), and 160 ng cDNA were added for a final volume of 25 μL. qPCR conditions were as follows: 1 cycle at 95°C for 2 min, then 35 cycles at 95°C for 30 s, 60–62°C (Table 1) for 30 s, and 72°C for 25 s. Reactions were carried out using a CFX96 Touch Real-Time PCR Detection System thermal cycler (Bio-Rad, Korea). The change in the expression of each target gene was determined using the Delta-delta Ct Pfaffl relative quantification method [27], and the results were expressed as the mean relative expression ± standard deviation.
Results
Clinical isolates and analysis
All 027 strains were strong biofilm producers. Among RT 001 clinical isolates, five isolates were strong producers, 3 were moderate producers, and 2 clinical isolates were no biofilm producers. Biofilm production and ribotyping results are shown in Fig 1 and S1 Table.
Line 1: molecular weight marker. Line 2: Control strain ATCC 9689 (RT001). Line 13: ATCC BAA 1805 (RT027). Lines 3–12 and 14–23: clinical isolates.
Microarrays quality and analysis
The eight pools yielded RNAc and specific activity greater than reference parameters (˃0.825 vs. 1.8195 μg and 15 vs. 20.3365 pmol Cy3 per μg of RNAc), respectively. Gene expression of unique pools and combination of pools was compared.
Pool 1 (nonadherent, RT001, NR-CDI) vs. pool 5 (biofilm, RT001, NR-CDI)
When we compared nonadherent cells vs. biofilm among RT001, NR-CDI strains, 149 genes were found to be overexpressed and 47 underexpressed from biofilm (Fig 2; S2 Table). The higher overexpression was detected for CAJ70248 (LogFC 2.909), which encodes an exosporium collagen-like glycoprotein BclA3. Conversely, the most underexpressed was CAJ67148 (LogFC, −1.955), which encodes an ABC-type transport system. Enriched pathways on gene ontology (GO) were associated with arginine metabolism, ethanolamine, organonitrogen compounds, small molecules, amines, the phosphotransferase system (PTS), and the transmembrane, carbohydrate, and organic substance transport pathways (Fig 3; S3 Table).
RNA from nonadherent cells (RT001 and 027) associated with R-CDI and NR-CDI (eight pools) were labeled with Cy3 dye and hybridized on expression microarray. Slides were scanned and gene expression were compared with LIMMA using the Student’s t-test. Only genes with FClog ≥ 1.5, ≤ -1.5 were considered; genes out of parameters are shown in black; overexpressed genes are in green and yellow colors; underexpressed genes are in blue.
Protein interaction and enriched pathways of DEGs found in nonadherent cells, and biofilm of RT001 and 027 or independent of ribotype, associated with R-CDI and NR-CDI were evaluated in the Gene Ontology (GO), Encyclopedia of Genes Kyoto Genes and Genomes (KEGG) in String and ShinyGO v0.75 online tool. FDR = <0.05.
Pool 2 (nonadherent, RT001, R-CDI) vs. pool 6 (biofilm, RT001, R-CDI)
The differential expression analysis between pool 2 and 6 revealed 56 overexpressed genes and 18 underexpressed from biofilm (Fig 2; S4 Table). The higher overexpression was detected in the CAJ67274 gene (LogFC, 2.268), which encodes for a dependent enzyme of the alanine/ornithine racemase family, whereas the most underexpressed gene was CAJ70208 (LogFC, −3.188), which encodes an uncharacterized protein.
The most enriched pathways in GO were associated with the transport of carbohydrates, organic substances, and transmembrane carbohydrates. Additional enriched pathways were associated with metabolism among phosphoenolpyruvate-dependent sugar PTS and small molecule catabolic process (Fig 3; S3 Table).
Pool 3 (nonadherent, RT027, NR-CDI) vs. pool 7 (biofilm, RT027, NR-CDI)
In this comparison, 45 genes were overexpressed, and 25 were underexpressed from biofilm (Fig 2; S5 Table). The most overexpressed gene was CAJ67274 (LogFC, 2.268), which encodes a dependent enzyme of the alanine/ornithine racemase family, and the least underexpressed gene was CAJ70208 (LogFC, −3.188), which encodes an uncharacterized protein. Two enriched pathways were detected, one associated with sporulation (genes, 4; FDR, 0.0010) and the other related to ascorbate and aldarate metabolism (KEGG; genes, 2; FDR, 0.0216) (Fig 3; S3 Table).
Pool 4 (non-adherent, RT027, R-CDI) vs. pool 8 (biofilm, RT027, R-CDI)
When comparing DEGs in pools 4 and 8, 42 genes were overexpressed, and 22 were underexpressed in biofilm (Fig 2; S6 Table). The most overexpressed gene was CAJ69029 (LogFC, 2.776), which encodes a sporulation membrane protein YtaF, and the more underexpressed gene was CAJ67779 (LogFC, −2.52), which encodes a hypothetical protein. Moreover, a GO-enriched sporulation pathway (Gene 3; FDR 0.0411) was detected (Fig 3; S3 Table).
Differentially expressed genes in biofilm in recurrent and non-recurrent strains (ribotype 001)
From the results obtained from the comparisons of Pool 1 (nonadherent, RT001, NR-CDI) vs. pool 5 (biofilm, RT001, NR-CDI) and Pool 2 (nonadherent, RT001, R-CDI) vs. pool 6 (biofilm, RT001, R-CDI), heat maps were built, and 189 genes were unique for biofilm in non-recurrent strains, 8 genes were common, while 64 genes were unique for recurrent strains (50 genes overexpressed and 14 underexpressed) (Fig 2; S7–S9 Tables). The most overexpressed gene was CAJ67269 (LogFC, 1.505), which encodes a 2-amino-4-ketopentanoate thiolase alpha subunit, and the least underexpressed gene was CAJ70208 (LogFC, −3.188), which encodes an uncharacterized protein. Most of the enriched pathways in gene ontology were associated with transmembrane carbohydrates and general transport and other metabolites such as carbohydrates and organic substances, as well as a PTS metabolism pathway and pathways associated with the regulation of the cellular process and regulation of transcription (Fig 3; S3 Table).
Among 64 unique expressed genes in recurrent strains, 16 genes encode uncharacterized proteins, with two out of them having similarity to phage proteins phiC2 (CBE03994; query cover, 97%; identity, 96.24%) and phiCD211 (CAJ69823; query cover, 96%; identity, 75%) (S10 Table). No other similarities were detected.
Differentially expressed genes in biofilm in recurrent and non-recurrent strains (ribotype 027)
From the results obtained from the comparisons of pool 3 (nonadherent, RT027, NR-CDI) vs. pool 7 (biofilm, RT027, NR-CDI) and pool 4 (nonadherent, RT027, R-CDI) vs. pool 8 (biofilm, RT027, R-CDI), heat maps were built and, where determined that 35 DEGs were unique for biofilm in non-recurrent strains, 35 genes were common, and 26 genes were unique for recurrent strains (S11–S13 Tables).
For the latter, 7 were overexpressed, and 19 were underexpressed (Fig 2; S11 Table). The most overexpressed gene was CAJ67765 (LogFC, 1.91), which encodes a sporulation membrane protein YtaF, and the least underexpressed gene was CAJ69479 (LogFC, −2.52), which encodes an uncharacterized protein. No enriched pathways were found.
In addition, 11 genes encoded uncharacterized proteins. Amino acid sequences were searched using Blastp, and one of them had similarity to an amidohydrolase 3 (CBE02518; query cover, 78%; identity, 99.71%); two of them had similarity with a phage tail tube protein (CBE04002; query cover, 100%; identity, 100%) and a protein containing the C-terminal domain of phage (CAJ67456; query cover, 100%; identity, 98.79%) (S14 Table).
Differentially expressed genes in biofilm in recurrent strains (ribotypes 001 and 027)
From the comparison between the 64 genes unique for recurrent strains of RT001 and 26 genes unique for recurrent strains of RT027, 63 genes were unique in RT001, 25 genes were unique in RT027, and 1 gene, CAJ69276, was common and overexpressed in both ribotypes (Fig 2). This gene encoded a probable membrane protein CAJ69276 (RT001: LogFC, 1.581; RT027: LogFC, 1.626). No protein homology was found in Blastp databases.
Pools 1 (nonadherent, RT001, NR-CDI) and 3 (nonadherent, RT027, NR-CDI) vs. pools 5 (biofilm-RT001, NR-CDI) and 7 (biofilm, RT027, NR-CDI)
When comparing DEGs in both ribotypes of non-recurrent strains (nonadherent cells vs. biofilm), 13 genes were overexpressed and 4 underexpressed (Fig 4; S15 Table) in non-recurrent biofilm strains. After correction for multiple comparisons, no significant differences were observed.
Data from DEGs on nonadherent cells and biofilm (R-CDI and NR-CDI strains independent of ribotype) was compared with LIMMA using the Student’s t-test and Benjamini-Hochberg correction for multiple comparisons. Only genes with adj.p = <0.01 and FClog ≥ 1.5, ≤ -1.5 were considered; genes out of parameters are shown in black; overexpressed genes are in green and yellow colors; underexpressed genes are in blue colors.
Pools 2 (nonadherent, RT001, R-CDI) and 4 (nonadherent, R027, R-CDI) vs. pools 6 (biofilm, RT001, R-CDI) and 8 (biofilm, RT027, R-CDI)
Comparing both ribotypes of recurrent strains (nonadherent cells vs. biofilm), we observed that 23 genes were overexpressed and 3 underexpressed (Fig 4; S16 Table) in recurrent biofilm strains. After Benjamini-Hochberg correction, genes were overexpressed: CAJ70148 (a putative dehydrogenase), CAJ68100 (putative protein), CAJ69725 (putative membrane protein), and one was underexpressed; CAJ68151 (a segregation and condensation protein A) (adj. p = 0.05).
Differentially expressed genes from both ribotypes in biofilm (non-recurrent and recurrent strains)
Finally, we compared DEGs in both ribotypes from biofilm (i.e., R-CDI vs. NR-CDI) and observed that 12 genes were unique in non-recurrent strains, 21 genes were unique in recurrent strains, and 4 genes were common (Fig 4; S17 and S18 Tables). For the latter, 18 were overexpressed, and 3 were underexpressed (Fig 4; S17 Table). After Benjamini-Hochberg correction, 3 genes were overexpressed (CAJ70148, a putative dehydrogenase: LogFC, 3.767; CAJ68100, putative protein: LogFC, 2.986; and CAJ69725, putative membrane protein: LogFC, 4.222) and 1 gene was underexpressed (CAJ68151, segregation and condensation protein A; LogFC, −2.877) (adj. p = ≤ 0.05) (S17 Table).
Among 21 unique expressed genes in recurrent strains, 11 genes encoding no characterized proteins were searched using Blastp, and 5 homologies were detected: similar to a type II secretion system (CAJ68100; query cover, 97%; identity, 46.11%), similar to a putative dehydrogenase (CAJ70148; query cover, 100%; identity, 100%) and a lactonase (CAJ70148; query cover, 100%; identity, 98.96%), similar to transposase (CBE06724; query cover, 100%; identity, 75.44%), and a similar to a 5-bromo-4-chloroindolyl phosphate hydrolysis family protein (CAJ69224; query cover, 100%; identity, 95.10%).
Validation of the results of the microarrays by RT-qPCR
Five overexpressed genes were selected to perform the validation of the results of the microarrays by means of RT-qPCR (LogFC, ≥2), including genes in the sporulation pathway and CAJ70248 (associated with the recurrence of C. difficile R20291 [RT027]) (14). The overexpression of genes evaluated by qPCR is like what was found in microarrays. Therefore, the results are validated (S1 Fig).
Discussion
Biofilm formation and changes in the genome of C. difficile have been associated with recurrence in CDI. However, the transcriptomic profile has been poorly studied, and the mechanisms underlying recurrence have not yet been completely understood. In this study, we analyzed the transcriptomic profile in an in vitro biofilm model of the C. difficile RT001 and RT027 ribotypes associated with R-CDI and NR-CDI. After multiple comparisons among groups involving features such as ribotype, biofilm formation, and recurrence of infection, we found several genes differentially expressed.
According to our results, four genes were differentially expressed (adj. p ≤ 0.05) in recurrence (pools 1 and 3 vs. pools 5 and 7) and biofilm (pools 5 and 7 vs. pools 6 and 8) evaluations. These genes correspond to a putative dehydrogenase (CAJ70148)—with high sequence homology to the YncE family protein and a lactonase—, a secretion type II system protein from the GspH (pseudopilins) family (CAJ68100), a putative membrane protein (CAJ69725), and the segregation and condensation protein A (CAJ68151); the first three were consistently overexpressed while the remaining underexpressed over comparisons. Taken together, differential expression of these genes appears to be essential to bacterial survival and communication functions.
The specific expression signature pointed to genes related to survival and biofilm maintenance/formation. However, the role of biofilm as a significant virulence and/or recurrence factor was rather contrasting among comparisons (i.e., in strains with biofilm formation but no recurrence). This is consistent with the notion that more than one mechanism (e.g., sporulation) is needed for the recurrence [28].
The putative dehydrogenase (CAJ70148) has a high sequence homology to the YncE family protein and a lactonase.
The yncE gene encodes the YncE protein identified in Escherichia coli [29]. YncE is an essential protein composed of a seven-bladed β-propeller [30, 31]. Analysis of subcellular fractions showed that YncE is mostly located in the periplasm [30]. Also, YncE has been identified as an antigen in patients whose bile ducts contain Salmonella enterica serovar Typhi, and it has been proposed that it might play a role in bacterial survival in the biliary tract [32].
CAJ70148 is also a potential lactonase; these enzymes have been shown to inhibit bacterial quorum sensing [33, 34] by hydrolysis of the ester bond in the lactone ring of acylated homoserine lactones (AHLs) [33, 35].
Some bacterial pathogens produce AHLs as signaling molecules in response to population density [36, 37]. Thus, by degrading AHLs, lactonases may curb quorum sensing to reduce the capability of bacterial pathogens to trigger virulent expression, which may inhibit the production of virulence factors and biofilm formation. However, since Gram-positive bacteria use auto-inducer peptides other than lactones [38], thus, a potential lactonase overexpression here is intriguing.
CAJ68100 encodes a type II secretion system (T2SS) protein from the GspH family. Proteins of this family participate in the assembling of the pseudopilus. This system is crucial for the survival of pathogen Gram-negative bacteria through the transport of toxins and several enzymes, including proteases, lipases and carbohydrate-active enzymes, to the cell surface or extracellular space of these bacteria [39]. Because secretion systems are virulence and pathogenicity factors [40], overexpression of CAJ68100 in our study could be linked to both the recurrence and biofilm formation events in the ribotypes evaluated here.
CAJ69725 is known to be a putative membrane protein. However, there is no additional information. As mentioned above, survival success often depends on extracellular communication, and this membrane protein’s overexpression may be involved.
Noticeably, when both ribotypes of recurrent strains (biofilm vs. nonadherent cells) and both ribotypes from biofilm (R-CDI vs. NR-CDI) were compared, CAJ68151 was underexpressed. CAJ68151 (segregation and condensation protein A) has a role in chromosomal partition during cell division and was underexpressed in the biofilm of and in R-CDI strains. We speculate that this fact could be a result of a “stationary state” where persistent cells limit their metabolism (slow-growing) or forfeit propagation in the presence of lethal threats [41]. This landscape describes biofilm features and may explain recurrence in bacteria such as Escherichia coli [41]. Thus, it is tempting to visualize the down expression of this gene as a recurrence driver in C. difficile.
The bioinformatic analysis revealed that the most enriched pathways on biofilm recurrent strains were related to the transport of carbohydrates and organic substances, transmembrane carbohydrates transport, and the PTS (Fig 3; S3 Table). The PTS is the main bacterial system for carbohydrate assimilation [42] and has been associated with the absorption and metabolism of carbohydrates, carbon, nitrogen, and phosphorus, as well as the production of toxins and chemotaxis [43]. Members of this system were either overexpressed or underexpressed in our model, which may provide advantages to C. difficile. In fact, some bacteria, such as Enterococcus faecalis, can coordinate their metabolism to escape from the immune system [44], decreasing the metabolism of carbohydrates (mannose and fructose) by the PTS1 pathway and changing primary carbon resources for macrophages impairing the ability of these to clear intracellular E. faecalis; consequently, it decreases the polarization M1 and reduces nitric oxide production affecting immune response [45]. Furthermore, the downregulation of metabolism pathways (e.g., PTS, carbon catabolism, tricarboxylic acid pathway, and pentose phosphate pathway) [46] can cause an ATP depletion associated with persistent bacteria biofilm formation in S. aureus [47]. In addition, Kulecka et al. determined that groups of genes related to metabolism pathways, such as oxidative phosphorylation, are common in the genome of recurrence-related strains of C. difficile [48].
Despite the statistical limitations in our microarray study, as several genes were underestimated after the Benjamini-Hochberg correction, the expression signature pointed to genes related to maintaining biofilm, survival, and likely favor recurrence.
Therefore, it deserves further attention, and studies with more biological replicates are guaranteed.
Within DEGs without statistical significance in microarrays, CAJ70248 was overexpressed in recurrent strains on nonadherent cells (RT001) (S3 Table). This protein encodes a protein that forms hair-like structures of the exosporium of C. difficile 20291 (RT027) and participates in adhesion [49] to fibronectin (α5β1 subunit), vitronectin [50], and integrin (αvβ1 subunit) to be internalized in colonocytes. This process may facilitate the immune response and antibiotic therapy evasion, thus favoring R-CDI [10].
Similarly, CAJ67607 (coding the SpoVAC protein), CAJ66946 (SpoIIID transcriptional regulator of sporulation protein), and CAJ67428 (CotJA spore coat–associated protein) (Fig 2; S4 Table (on biofilm in RT001 recurrent strains), were also overexpressed. In the case of RT027 recurrent strains, genes such as CAJ68046 (stage III sporulation protein AA) were overexpressed, while CAJ67607 (SpoVAC) and CAJ69029 (sporulation membrane protein YtaF) were underexpressed (Fig 2; S6 Table). In addition to sporulation genes, some virulence factors were detected to be differentially expressed. For example, flgB—which encodes a flagellar basal body rod protein—was underexpressed on biofilm in R-CDI strains (RT027) (Fig 2; S6 Table). Another virulence factor found in the present study is CD630_26040, which encodes a fragment of ADP-ribosyltransferase CdtAB [51–53]. This gene was underexpressed on biofilm in R-CDI strains (RT027) compared with nonadherent cells (Fig 2; S6 Table). The presence of genes encoding C. difficile toxin (CDT) can predict a recurrent episode in humans, and a combination of the detection of tcdA, tcdB, and CDT genes results in an odds ratio of 3.1 (95% confidence interval, 2.97–3.33) for R-CDI [54].
In our study, RT-qPCR experiments were performed to validate differential expressions in the microarrays. Genes were selected based on their expression and presence in recurrence.
Our results are the basis for further experiments, including some gene modifications- like gene deletion, gene knockdown or over-expression- to identify the role recurrence.
In Bacillus spp. and other clostridia, germination begins when a germinating molecule interacts with the germinative (Ger) receptor. C. difficile spores do not share homologs of the gerA, gerB, and gerK sprout receptors commonly recognized [55].
In C. difficile, the first step of germination is the binding of germinant to subtilin-like serine protease (CSP)-like proteins encoded at the cspBAC locus, where the cspBA and cspC genes that code for the CspBA and CspC are key regulators of germination signal transduction [56]. CspC functions as a receptor for bile salts. In addition, it transmits the signal through protein-protein interactions to CspB. Overexpression of cspC and cspAB has been found in a biofilm model of C.difficile 630Δerm growth at 37°C in tryptone yeast extract supplemented with 0.1% sodium thioglycolate [57]. In our study, the cspC gene was not overexpressed, possibly due to the lack of thioglycolate in the medium used (BHIS).
The use of a pool of samples was included in this study to minimize the effects of individual patients and different strains (including the level of biofilm formation). We included the same patient characteristics, as this is the common denominator for the preparation of the pools.
It has been reported that in CDI-R, 70 to 81% is due to the same strain, and 15 to 30% is due to a different strain [58, 59]. Unfortunately, we have no strains collected in the initial CDI and the recurrent CDI to define whether the strain is the same. The use of pools is intended to minimize this effect.
Some ribotypes are frequently associated with recurrence, including RT012 [60], RT078, RT244 [61], and RT027 [7–9]. In contrast, other ribotypes have no clear association with recurrence, including RT001 RT106, RT017 and 018 [62–64]. We selected a ribotype clearly associated with recurrence and a second ribotype not clearly associated with recurrence. Our results provide an overview of the transcriptomic differences among strains associated with CDI-R and CDR-NR in a 3-day biofilm, depending on the ribotype.
The inclusion of other ribotypes will be of interest to clearly define the importance of ribotypes in the recurrence of infection.
In our study, CAJ70148, CAJ68100, CAJ69725 and CAJ68151 were differentially expressed in biofilm in strains associated with R-CDI. A significant limitation in the results is that some of these genes have not yet been studied, and the proteins they encode and the pathways in which they participate are unknown. Multiple investigations are required to determine the function of the unknown genes to elucidate the exact mechanisms.
In conclusion, genes related to survival, virulence, and sporulation were differentially expressed, which suggests an essential role in biofilm formation and recurrence. Moreover, our model further supports the notion that other biological processes besides biofilm formation are involved in the recurrence. Finally, as long as we characterize unknown proteins, we will better understand the adaptations of C. difficile in R-CDI.
Supporting information
S1 Table. Classification of biofilm production in C. difficile.
https://doi.org/10.1371/journal.pone.0289593.s001
(DOCX)
S2 Table. Differentially expressed genes in NR-CDI strains, RT001 (Pool 1, nonadherent, RT001, NR-CDI vs. Pool 5, biofilm, RT001, NR-CDI).
https://doi.org/10.1371/journal.pone.0289593.s002
(DOCX)
S3 Table. Pathways enriched in gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG).
https://doi.org/10.1371/journal.pone.0289593.s003
(DOCX)
S4 Table. Differentially expressed genes in R-CDI strains, RT001 (Pool 2, nonadherent, RT001, R-CDI vs. Pool 6, biofilm, RT001, R-CDI).
https://doi.org/10.1371/journal.pone.0289593.s004
(DOCX)
S5 Table. Differentially expressed genes in NR-CDI strains, RT027 (Pool 3, nonadherent, RT027, NR-CDI vs. Pool 7, biofilm, RT027, NR-CDI).
https://doi.org/10.1371/journal.pone.0289593.s005
(DOCX)
S6 Table. Differentially expressed genes in R-CDI strains, RT027 Pool 4, non-adherent, RT027, R-CDI vs. Pool 8, biofilm, RT027, R-CDI).
https://doi.org/10.1371/journal.pone.0289593.s006
(DOCX)
S7 Table.
Unique genes differentially expressed on biofilm R-CDI, RT001 strains. Pool 1 (nonadherent, RT001, NR-CDI) vs. Pool 5 (biofilm, RT001, NR-CDI) and Pool 2 (nonadherent, RT001, R-CDI) vs. Pool 6 (biofilm, RT001, R-CDI).
https://doi.org/10.1371/journal.pone.0289593.s007
(DOCX)
S8 Table. Unique genes differentially expressed on biofilm NR-CDI, RT001 strains.
Pool 1 (nonadherent, RT001, NR-CDI) vs. Pool 5 (biofilm, RT001, NR-CDI) and Pool 2 (nonadherent, RT001, R-CDI) vs. Pool 6 (biofilm, RT001, R-CDI).
https://doi.org/10.1371/journal.pone.0289593.s008
(DOCX)
S9 Table. Common genes differentially expressed on biofilm NR-CDI and R-CDI, RT001 strains.
https://doi.org/10.1371/journal.pone.0289593.s009
(DOCX)
S10 Table. Identification of unique unidentified proteins by Blastp of C. difficile biofilm on biofilm R-CDI strains of RT001.
Pool 1 (nonadherent, RT001, NR-CDI) vs. Pool 5 (biofilm, RT001, NR-CDI) and Pool 2 (nonadherent, RT001, R-CDI) vs. Pool 6 (biofilm, RT001, R-CDI).
https://doi.org/10.1371/journal.pone.0289593.s010
(DOCX)
S11 Table. Unique genes differentially expressed in biofilm of R-CDI RT027 strains.
Pool 3 (nonadherent, RT027, NR-CDI) vs. Pool 7 (biofilm, RT027, NR-CDI) and Pool 4 (nonadherent, RT027, R-CDI) vs. Pool 8 (biofilm, RT027, R-CDI).
https://doi.org/10.1371/journal.pone.0289593.s011
(DOCX)
S12 Table. Unique genes differentially expressed on biofilm NR-CDI, RT027 strains.
Pool 3 (nonadherent, RT027, NR-CDI) vs. Pool 7 (biofilm, RT027, NR-CDI) and Pool 4 (nonadherent, RT027, R-CDI) vs. Pool 8 (biofilm, RT027, R-CDI).
https://doi.org/10.1371/journal.pone.0289593.s012
(DOCX)
S13 Table. Common genes differentially expressed on biofilm NR-CDI and R-CDI, RT027 strains.
Pool 3 (nonadherent, RT027, NR-CDI) vs. Pool 7 (biofilm, RT027, NR-CDI) and Pool 4 (nonadherent, RT027, R-CDI) vs. Pool 8 (biofilm, RT027, R-CDI).
https://doi.org/10.1371/journal.pone.0289593.s013
(DOCX)
S14 Table. Identification of unique unidentified proteins by Blastp in C. difficile biofilm of R-CDI, RT027 strains.
Pool 3 (nonadherent, RT027, NR-CDI) vs. Pool 7 (biofilm, RT027, NR-CDI) and Pool 4 (nonadherent, RT027, R-CDI) vs. Pool 8 (biofilm, RT027, R-CDI).
https://doi.org/10.1371/journal.pone.0289593.s014
(DOCX)
S15 Table. Differentially expressed genes in RT001 and RT027, NR-CDI strains (nonadherent cells vs. biofilm).
Pool 1 (nonadherent, RT001, NR-CDI) and Pool 3 (nonadherent, RT027, NR-CDI) vs. Pool 5 (biofilm, RT001, NR-CDI) and Pool 7 (Biofilm, RT027, NR-CDI).
https://doi.org/10.1371/journal.pone.0289593.s015
(DOCX)
S16 Table. Differentially expressed genes in RT001 and RT027, R-CDI strains (nonadherent vs. biofilm).
Pool 2 (nonadherent, RT001, R-CDI) and Pool 4 (nonadherent, R027, R-CDI) vs. Pool 6 (biofilm, RT001, R-CDI) and Pool 8 (biofilm, RT027, R-CDI).
https://doi.org/10.1371/journal.pone.0289593.s016
(DOCX)
S17 Table. Unique differentially expressed genes biofilm R-CDI, independent of ribotype.
https://doi.org/10.1371/journal.pone.0289593.s017
(DOCX)
S18 Table. Unique differentially expressed genes biofilm NR-CDI, independent of ribotype.
https://doi.org/10.1371/journal.pone.0289593.s018
(DOCX)
S1 Fig. Graph of relative expression of the 8 pools of interest.
Bonferroni post hoc p≤ 0.01.
https://doi.org/10.1371/journal.pone.0289593.s019
(JPG)
S1 Raw image. Ribotyping of Clostridioides difficile strains.
Line 1: molecular weight marker. Line 2: Control strain ATCC 9689 (RT001). Line 13: ATCC BAA 1805 (RT027). Lines 3–12 and 14–23: clinical isolates. Fig 1 was created from this image. Image was captured using a BioRad Molecular Imager ChemiDoc XRS instrument.
https://doi.org/10.1371/journal.pone.0289593.s020
(PDF)
Acknowledgments
We thank MSc Ana Zarazua for her help with the microarray analysis and Lucy Acevedo for her technical assistance.
References
- 1. McDonald LC, Gerding DN, Johnson S, Bakken JS, Carroll KC, Coffin SE, et al. Clinical Practice Guidelines for Clostridium difficile Infection in Adults and Children: 2017 Update by the Infectious Diseases Society of America (IDSA) and Society for Healthcare Epidemiology of America (SHEA). Clin Infect Dis. 2018;66(7):e1–48. pmid:29462280
- 2. Dávila LP, Garza-González E, Rodríguez-Zulueta P, Morfín-Otero R, Rodríguez-Noriega E, Villar-Compte D, et al. Increasing rates of Clostridium difficile infection in Mexican hospitals. Brazilian J Infect Dis. 2017;21(5):530–4. pmid:28641091
- 3. Martínez-Meléndez A, Tijerina-Rodríguez L, Morfin-Otero R, Camacho-Ortíz A, Villarreal-Treviño L, Sánchez-Alanís H, et al. Circulation of Highly Drug-Resistant Clostridium difficile Ribotypes 027 and 001 in Two Tertiary-Care Hospitals in Mexico. Microb Drug Resist. 2018;24(4):386–92. pmid:29485939
- 4. Surawicz CM, Brandt LJ, Binion DG, Ananthakrishnan AN, Curry SR, Gilligan PH, et al. Guidelines for diagnosis, treatment, and prevention of clostridium difficile infections. Am J Gastroenterol. 2013;108(4):478–98. pmid:23439232
- 5. Cho JM, Pardi DS, Khanna S. Update on Treatment of Clostridioides difficile Infection. Mayo Clin Proc. 2020;95(4):758–69. pmid:32247350
- 6. Clements A, Magalhães R, Tatem A, Paterson D, Riley T. Clostridium difficile PCR ribotype 027: assessing the risks of further worldwide spread. Lancet Infect Dis. 2010;10(6):395–404. pmid:20510280
- 7. Marsh JW, Arora R, Schlackman JL, Shutt KA, Curry SR, Harrison LH. Association of relapse of Clostridium difficile disease with BI/NAP1/027. J Clin Microbiol. 2012;50(12):4078–82. pmid:23052318
- 8. Abou Chakra CN, Pepin J, Sirard S, Valiquette L. Risk factors for recurrence, complications and mortality in Clostridium difficile infection: A systematic review. PLoS One. 2014;9(6) e98400.
- 9. Krishna Rao, Higgins Peter YV. An Observational Cohort Study of Clostridium difficile Ribotype 027 and Recurrent Infection. mSphere. 2018;3(3):1–8.
- 10. Castro-Córdova P, Mora-Uribe P, Reyes-Ramírez R, Cofré-Araneda G, Orozco-Aguilar J, Brito-Silva C, et al. Entry of spores into intestinal epithelial cells contributes to recurrence of Clostridioides difficile infection. Nat Commun. 2021;12(1):1–18.
- 11. Normington C, Moura IB, Bryant JA, Ewin DJ, Clark E V., Kettle MJ, et al. Biofilms harbour Clostridioides difficile, serving as a reservoir for recurrent infection. npj Biofilms Microbiomes. 2021;7(1). pmid:33547298
- 12. Deakin LJ, Clare S, Fagan RP, Dawson LF, Pickard DJ, West MR, et al. The Clostridium difficile spo0A gene is a persistence and transmission factor. Infect Immun. 2012;80(8):2704–11.
- 13. Dawson LF, Valiente E, Faulds-pain A, Donahue EH, Wren BW. Characterisation of Clostridium difficile Biofilm Formation, a Role for Spo0A. 2012;7(12):e50527. pmid:23236376
- 14. Timmaraju A, Balasubramanian K, Shakih S, David F, Sapi E. Biofilm formation by Clostridium difficile. J Bacteriol. 2013;195(October):1–20.
- 15. Semenyuk EG, Laning ML, Foley J, Johnston PF, Knight KL, Gerding DN, et al. Spore formation and toxin production in Clostridium difficile biofilms. PLoS One. 2014;9(1):e87757.
- 16. Pantaléon V, Monot M, Eckert C, Hoys S, Collignon A, Janoir C, et al. Clostridium difficile forms variable biofilms on abiotic surface. Anaerobe. 2018;53:34–7. pmid:29859742
- 17. Tijerina-Rodríguez L, Villarreal-Treviño L, Baines SD, Morfín-Otero R, Camacho-Ortíz A, Flores-Treviño S, et al. High sporulation and overexpression of virulence factors in biofilms and reduced susceptibility to vancomycin and linezolid in recurrent Clostridium [Clostridioides] difficile infection isolates. PLoS One. 2019;14(7):1–14. pmid:31365590
- 18. Sambrook J, Russell DW. Purification of nucleic acids by extraction with phenol:chloroform. CSH Protoc. 2006 Jun;2006(1):pdb.prot4455.
- 19. Bidet P, Petit J. Development of a new PCR-ribotyping method for Clostridium di ¤ cile based on ribosomal RNA gene sequencing. FEMS Microbiol Lett. 1999;175:261–6.
- 20. Vukovic D, Bonaventura GDI, Djukic S, Ruzicka F. Quantification of biofilm in microtiter plates: overview of testing conditions and practical recommendations for assessment of biofilm production by staphylococci. Acta Pathol Microbiol Immunol Scand APMIS. 2007;(115):891–9. pmid:17696944
- 21. Technologies A. Agilent Two-Color Microarray-Based Exon Analysis Low Input Quick Amp WT Labeling Protocol. 2010;(November).
- 22. Guimarães I, Recamonde-Mendoza M, Feltes B. Gene Expression Analysis Platform (GEAP): A highly customizable, fast, versatile and ready-to-use microarray analysis platform. Genet Mol Biol. 2022;20210077:1–15.
- 23. Sherman BT, Huang DW, Tan Q, Guo Y, Bour S, Liu D, et al. DAVID Knowledgebase: A gene-centered database integrating heterogeneous gene annotation resources to facilitate high-throughput gene functional analysis. BMC Bioinformatics. 2007;8:1–11.
- 24. Ge SX, Jung D, Jung D, Yao R. ShinyGO: A graphical gene-set enrichment tool for animals and plants. Bioinformatics. 2020;36(8):2628–9. pmid:31882993
- 25. Denève C, Bouttier S, Dupuy B, Barbut F, Collignon A, Janoir C. Effects of subinhibitory concentrations of antibiotics on colonization factor expression by moxifloxacin-susceptible and moxifloxacin-resistant Clostridium difficile strains. Antimicrob Agents Chemother. 2009;53(12):5155–62. pmid:19805572
- 26. Invitrogen C. SuperScript III Reverse Transcriptase (200U/μL) #18080–085. Manual. 2004;(18080):1–4.
- 27. Pfaffl M. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 2001;1(29):45. pmid:11328886
- 28. Taggart MG, Snelling WJ, Naughton PJ, la Ragione RM, Dooley JSG, Ternan NG. Biofilm regulation in Clostridioides difficile: Novel systems linked to hypervirulence. PLoS Pathog. 2021;17(9):1–20. A pmid:34499698
- 29. McHugh JP, Rodríguez-Quiñones F, Abdul-Tehrani H, Svistunenko DA, Poole RK, Cooper CE, et al. Global iron-dependent gene regulation in Escherichia coli: A new mechanism for iron homeostasis. J Biol Chem. 2003;278(32):29478–86.
- 30. Baba-Dikwa A, Thompson D, Spencer NJ, Andrews SC, Watson KA. Overproduction, purification and preliminary X-ray diffraction analysis of YncE, an iron-regulated Sec-dependent periplasmic protein from Escherichia coli. Acta Crystallogr Sect F Struct Biol Cryst Commun. 2008;64(10):966–9. pmid:18931447
- 31. Li J, Xiong K, Zou L, Chen Z, Wang Y, Hu X, et al. Lytic Action of the Truncated yncE Gene in Escherichia coli. Curr Microbiol. 2016;72(4):390–6. pmid:26687463
- 32. Charles RC, Sultana T, Alam MM, Yu Y, Wu-Freeman Y, Bufano MK, et al. Identification of Immunogenic Salmonella enterica Serotype Typhi Antigens Expressed in Chronic Biliary Carriers of S. Typhi in Kathmandu, Nepal. PLoS Negl Trop Dis. 2013;7(8):1–8. pmid:23936575
- 33. Bergonzi C, Schwab M, Naik T, Daudé D, Chabrière E, Elias M. Structural and Biochemical Characterization of AaL, a Quorum Quenching Lactonase with Unusual Kinetic Properties. Sci Rep. 2018;8(1):19–21.
- 34. Gunaratnam S, Millette M, McFarland L V., DuPont HL, Lacroix M. Potential role of probiotics in reducing Clostridioides difficile virulence: Interference with quorum sensing systems. Microb Pathog. 2021;153(June 2020):104798. pmid:33609647
- 35. Bergonzi C, Schwab M, Naik T, Elias M. The Structural Determinants Accounting for the Broad Substrate Specificity of the Quorum Quenching Lactonase GcL. ChemBioChem. 2019;20(14):1848–55. pmid:30864300
- 36. Papenfort K, Bassler BL. Quorum sensing signal-response systems in Gram-negative bacteria. Nat Rev Microbiol. 2016;14(9):576–88. pmid:27510864
- 37. Yang M, Meng F, Gu W, Li F, Tao Y, Zhang Z, et al. Effects of Natural Products on Bacterial Communication and Network-Quorum Sensing. Biomed Res Int. 2020;2020:8638103. pmid:32596389
- 38. Verbeke F, De Craemer S, Debunne N, Janssens Y, Wynendaele E, Van de Wiele C, et al. Peptides as quorum sensing molecules: Measurement techniques and obtained levels in vitro and in vivo. Front Neurosci. 2017;11(APR):1–18.
- 39. Korotkov K V, Sandkvist M. Architecture, funtion, and substrates of the type II secretion system. EcoSal Plus. 2019;8(2):10.1128
- 40. Altayb HN, Elbadawi HS, Baothman O, Kazmi I, Alzahrani FA, Nadeem MS, et al. Genomic Analysis of Multidrug-Resistant Hypervirulent (Hypermucoviscous) Klebsiella pneumoniae Strain Lacking the Hypermucoviscous Regulators (rmpA/rmpA2). Antibiotics. 2022;11(596):1–18. pmid:35625240
- 41. Lewis K. Persister Cells and the Riddle of Biofilm Survival. Biochemistry. 2005;70(2):327–36. pmid:15807669
- 42. Antunes A, Martin-Verstraete I, Dupuy B. CcpA-mediated repression of Clostridium difficile toxin gene expression. Mol Microbiol. 2011;79(4):882–99.
- 43. Janoir C, Denève C, Bouttier S, Barbut F, Hoys S, Caleechum L, et al. Adaptive strategies and pathogenesis of clostridium difficile from In vivo transcriptomics. Infect Immun. 2013;81(10):3757–69. pmid:23897605
- 44. Zhai W, Wu F, Zhang Y, Fu Y, Liu Z. The immune escape mechanisms of Mycobacterium Tuberculosis. Int J Mol Sci. 2019;20(2):340. pmid:30650615
- 45. Wei L, Xia F, Wang J, Ran S, Liang Y, Zhou W, et al. Carbohydrate Metabolism Affects Macrophage-Mediated Killing of Enterococcus faecalis. mSystems. 2021;6(5):1–17. pmid:34491082
- 46. Loss G, Simões PM, Valour F, Cortês MF, Gonzaga L, Bergot M, et al. Staphylococcus aureus Small Colony Variants (SCVs): News From a Chronic Prosthetic Joint Infection. Front Cell Infect Microbiol. 2019;9(October):1–12. pmid:31696062
- 47. Conlon BP, Rowe SE, Gandt AB, Nuxoll AS, Donegan NP, Zalis EA, et al. Persister formation in Staphylococcus aureus is associated with ATP depletion. Nat Microbiol. 2016;1(5):16051.
- 48. Kulecka M, Waker E, Ambrozkiewicz F, Paziewska A, Skubisz K, Cybula P, et al. Higher genome variability within metabolism genes associates with recurrent Clostridium difficile infection. BMC Microbiol. 2021;21(1):1–10.
- 49. Mora-Uribe P, Miranda-Cárdenas C, Castro-Córdova P, Gil F, Calderón I, Fuentes JA, et al. Characterization of the adherence of Clostridium difficile spores: The integrity of the outermost layer affects adherence properties of spores of the epidemic strain R20291 to components of the intestinal mucosa. Front Cell Infect Microbiol. 2016;6(SEP):1–16.
- 50. Paredes-Sabja D, Sarker MR. Adherence of Clostridium difficile spores to Caco-2 cells in culture. J Med Microbiol. 2012;61:1208–18. pmid:22595914
- 51. Riegler M, Sedivy R, Pothoulakis C, Hamilton G, Zacherl J, Bischof G, et al. Clostridium difficile toxin B is more potent than toxin A in damaging human colonic epithelium in vitro. J Clin Invest. 1995;95(5):2004–11. pmid:7738167
- 52. Rocha MFG, Maia MET, Bezerra LRPS, Lyerly DM, Guerrant RL, Ribeiro RA, et al. Clostridium difficile toxin A induces the release of neutrophil chemotactic factors from rat peritoneal macrophages: Role of interleukin- 1β, tumor necrosis factor alpha, and leukotrienes. Infect Immun. 1997;65(7):2740–6.
- 53. Martin-Verstraete I, Peltier J, Dupuy B. The regulatory networks that control Clostridium difficile toxin synthesis. Toxins (Basel). 2016;8(5):1–24. pmid:27187475
- 54. Stewart DB, Berg A, Hegarty J. Predicting Recurrence of C. difficile Colitis Using Bacterial Virulence Factors: Binary Toxin Is the Key. J Gastrointest Surg. 2013;17(1):118–25.
- 55. Chilton CH, Pickering DS, Freeman J. Microbiologic factors affecting Clostridium difficile recurrence. Clin. Microbiol. Infect. 2018;24(5): 476–482.
- 56. Lawler AJ, Lambert PA, Worthington T Revised Understanding of Clostridioides difficile spore Germination. Trends Microbiol. 2020;28(9):744–752.
- 57. Poquet I, Saujet L, Canette A, Monot M, Mihajlovic J, et al. Clostridium difficile Biofilm: Remodeling Metabolism and Cell Surface to Build a Sparse and Heterogeneously Aggregated Architecture. Front Microbiol. 2018 S;9:2084.
- 58. Eyre DW, Babakhani F, Griffiths D, Seddon J, Del Ojo Elias C, Gorbach SL, et al. Whole-genome sequencing demonstrates that fidaxomicin is superior to vancomycin for preventing reinfection and relapse of infection with Clostridium difficile. J Infect Dis. 2014;209(9):1446–51.
- 59. Mac Aogáin M, Moloney G, Kilkenny S, Kelleher M, Kelleghan M, et al. Whole-genome sequencing improves discrimination of relapse from reinfection and identifies transmission events among patients with recurrent Clostridium difficile infections. J Hosp Infect. 2015;90(2):108–16.
- 60. Krutova M, Matejkova J, Drevinek P, Kuijper EJ, Nyc O; study group. Increasing incidence of Clostridium difficile ribotype 001 associated with severe course of the infection and previous fluoroquinolone use in the Czech Republic, 2015. Eur J Clin Microbiol Infect Dis. 2017;36(11):2251–2258.
- 61. Rounds J, Strain J. Bezlotoxumab for Preventing Recurrent Clostridium difficile Infections S D Med. 2017;70(9):422–423.
- 62. Magnusson C, Wullt M, Löfgren S, Iveroth P, Akerlund T, Matussek A. Ribotyping of Clostridium difficile strains associated with nosocomial transmission and relapses in a Swedish County. APMIS. 2013;121(2):153–7.
- 63. Neely F, Lambert ML, Van Broeck J, Delmée M. Clinical and laboratory features of the most common Clostridium difficile ribotypes isolated in Belgium. J Hosp Infect. 2017;95(4):394–399.
- 64. Hung YP, Huang IH, Lin HJ, Tsai BY, Liu HC, Liu HC, et al. Predominance of Clostridium difficile Ribotypes 017 and 078 among Toxigenic Clinical Isolates in Southern Taiwan. PLoS One. 2016;11(11):e0166159.