Characterization of Clostridioides difficile DSM 101085 with A-B-CDT+ Phenotype from a Late Recurrent Colonization

Abstract

During the last decades, hypervirulent strains of Clostridioides difficile with frequent disease recurrence and increased mortality appeared. Clostridioides difficile DSM 101085 was isolated from a patient who suffered from several recurrent infections and colonizations, likely contributing to a fatal outcome. Analysis of the toxin repertoire revealed the presence of a complete binary toxin locus and an atypical pathogenicity locus consisting of only a tcdA pseudogene and a disrupted tcdC gene sequence. The pathogenicity locus shows upstream a transposon and has been subject to homologous recombination or lateral gene transfer events. Matching the results of the genome analysis, neither TcdA nor TcdB production but the expression of cdtA and cdtB was detected. This highlights a potential role of the binary toxin C. difficile toxin in this recurrent colonization and possibly further in a host-dependent virulence. Compared with the C. difficile metabolic model strains DSM 28645 (630Δerm) and DSM 27147 (R20291), strain DSM 101085 showed a specific metabolic profile, featuring changes in the threonine degradation pathways and alterations in the central carbon metabolism. Moreover, products originating from Stickland pathways processing leucine, aromatic amino acids, and methionine were more abundant in strain DSM 101085, indicating a more efficient use of these substrates. The particular characteristics of strain C. difficile DSM 101085 may represent an adaptation to a low-protein diet in a patient with recurrent infections.

Keywords: Clostridioides difficile; Clostridium difficile; R20291 resequencing; binary toxin; fermentation profile; pathogenicity locus.

Fig. 1.—

Transposons, transposon-like elements (A), and prophages of C. difficile DSM 101085 (B). For tree construction, analysis of both transposons and phages, VICTOR was used (Meier-Kolthoff and Göker 2017). All pairwise comparisons of the nucleotide sequences were conducted using the Genome-BLAST Distance Phylogeny method (Meier-Kolthoff et al. 2013) under settings recommended for prokaryotic viruses (Meier-Kolthoff and Göker 2017). The resulting intergenomic distances were used to infer a balanced minimum evolution tree with branch support via FASTME including SPR postprocessing (Lefort et al. 2015) for formula D0. Branch support was inferred from 100 pseudo-bootstrap replicates each. Trees were visualized with FigTree.

Fig. 2.—

Genomic features of strain DSM 101085. (A) Comparison of PaLocs and flanking regions. The PaLocs of strain DSM 28645 (630Δerm), DSM 27147 (R20291), and DSM 101085 were genomically compared. (B) Annotations of the PaLoc-associated transposon-like sequence in DSM 101085. (C) Annotations of the PaLoc-associated transposon-like sequence according Dingle et al. (2014). (D) Comparison of CdtLocs and its flanking regions. The CdtLocs of strain DSM 28645 (630Δerm), DSM 27147 (R20291), and DSM 101085 were genomically compared.

Fig. 3.—

Quantification of toxins. Strains DSM 101085, DSM 27147 (R20291), and DSM 28645 (630Δerm) were analyzed for the presence of the toxins TcdA (light green) and TcdB (dark green) by ELISA, as well as cdtA (light purple) and cdtB (dark purple) by qPCR.

Fig. 4.—

Comparison of logarithmized normalized peak areas between C. difficile DSM 101085 and the model strains. (A) Metabolites found in cell extracts after cultivation in casamino acids in comparison to the model strains DSM 28645 (630Δerm) and (B) DSM 27147 (R20291) are shown. All strains were simultaneously grown until about 1/2 OD_max. Values represent the average of four independent experiments. Error bars represent the standard deviation between the three experiments. Metabolites with an FC > 3 and a P value of >0.05 in a nonparametric Wilcoxon–Mann–Whitney test using the Benjamini–Hochberg correction are labeled. Detailed data are available as supplementary table 3, Supplementary Material online.

Fig. 5.—

Specific formation of fermentation products and substrate usage. The relative abundances in the culture supernatant of four individual cultivations compared with the average of abundances in cultivations of the reference strains. Light green: >1.5-fold more abundant compared with one reference strain; dark green: >1.5-fold more abundant compared with both reference strains; light red: >1.5-fold less abundant compared with one reference strain; and dark red: >1.5-fold less abundant compared with both reference strains. Gray substrates were not used by one of the strains. Used methods for relative quantifications of each compound are listed in supplementary table 3, Supplementary Material online.