Penicillin Binding Protein Substitutions Co-occur with Fluoroquinolone Resistance in ‘Epidemic’ Lineages of Multi Drug-Resistant Clostridioides difficile

Clostridioides difficile remains a key cause of healthcare-associated infection, with multi-drug-resistant (MDR) lineages causing high mortality (≥20%) outbreaks. Cephalosporin treatment is a long-established risk factor, and antimicrobial stewardship a key control. A mechanism underlying raised cephalosporin MICs has not been identified in C. difficile, but among other species resistance is often acquired via amino acid substitutions in cell wall transpeptidases (penicillin binding proteins, PBPs). Here, we investigated five C. difficile transpeptidases (PBP1-5) for recent substitutions. Previously published genome assemblies (n=7096) were obtained, representing sixteen geographically widespread lineages, including healthcare-associated MDR ST1(027), ST3(001) and ST17(018). Recent amino acid substitutions were found within PBP1 (n=50) and PBP3 (n=48), ranging from 1-10 substitutions per genome. β-lactam MICs were measured for closely related pairs of wild-type and PBP substituted isolates separated by 20-273 SNPs. Recombination-corrected, dated phylogenies were constructed to date substitution acquisition. Key substitutions such as PBP3 V497L and PBP1 T674I/N/V emerged independently across multiple lineages. They were associated with extremely high cephalosporin MICs; 1-4 doubling dilutions >wild-type up to ≤1506μg/ml. Substitution patterns varied by lineage and clade, showed geographic structure, and notably occurred post-1990, coincident with the acquisition of gyrA/B substitutions conferring fluoroquinolone resistance. In conclusion, recent PBP1 and PBP3 substitutions are associated with raised cephalosporin MICs in C. difficile. The co-occurrence of resistance to cephalosporins and fluoroquinolones hinders attempts to understand their relative importance in the dissemination of epidemic lineages. Further controlled studies of cephalosporin and fluoroquinolone stewardship are needed to determine their relative effectiveness in outbreak control. IMPORTANCE Fluoroquinolone and cephalosporin prescribing in healthcare settings have triggered outbreaks of high-mortality, multi-drug resistant C. difficile infection. Here, we identify a mechanism of acquired cephalosporin resistance in C. difficile, comprising amino acid substitutions in two cell-wall transpeptidase enzymes (penicillin binding proteins). The higher the number of substitutions, the greater the impact on phenotype. Dated phylogenies revealed that resistance to both cephalosporins and fluoroquinolones was co-acquired immediately before clinically important, outbreak strains emerged. PBP substitutions were geographically structured within genetic lineages, suggesting adaptation to local antimicrobial prescribing. Antimicrobial stewardship of cephalosporins and fluoroquinolones is an effective means of C. difficile outbreak control. Genetic changes conferring resistance likely impart a ‘fitness-cost’ after antibiotic withdrawal. Our study identifies a mechanism that may explain the contribution of cephalosporin stewardship to resolving outbreak conditions. However, due to the co-occurrence of cephalosporin and fluoroquinolone resistance, further work is needed to determine the relative importance of each.

likely impart a 'fitness-cost' after antibiotic withdrawal. Our study identifies a mechanism 58 that may explain the contribution of cephalosporin stewardship to resolving outbreak 59 conditions. However, due to the co-occurrence of cephalosporin and fluoroquinolone 60 resistance, further work is needed to determine the relative importance of each.

INTRODUCTION
The frequency of each amino acid substitution was recorded per lineage (Table 2). This   (Table S1). 151 Among the fourteen clinically important lineages, PBP substitutions occurred relatively 152 rarely in the absence fluoroquinolone resistance (Figure 2). In this respect ST2(014/020)   The choice of isolates from ST1(027), ST17(018), ST3(001) and ST42(106), was based on 165 low numbers of SNP differences between wild type and PBP substituted genomes ( Figure   166 3A). In total, ten different PBP substitutions were represented, three in PBP3 and seven in 167 PBP1 ( Figure 3B), including the four most frequently identified substitutions. 168 Isolates containing PBP substitutions showed increased cephalosporin and carbapenem MICs 169 relative to wild-type, but their penicillin MICs were unchanged ( Figure 3A). The greatest 170 increases in cephalosporin MICs were associated with the highest numbers of substitutions, 171 for example cefuroxime MIC increased from 376 to 1506μg/ml in ST3(001) (four 172 substitutions) and ST17(018) (five substitutions). Cephradine MIC increased from 36 to 173 239μg/ml in the latter. Intriguingly, the wild-type ancestors of these four PBP substituted 174 9 lineages had cefotaxime MICs which were still higher than the four lineages which have not 175 yielded PBP substituted strains; cefuroxime 128µg/ml vs 376µg/ml and cefotaxime 128µg/ml 176 vs 256µg/ml.  Table 2). Then further PBP substitutions followed, yielding a variety of patterns. Among 188 lineages well known for epidemic spread, the initial PBP3 V497L substitution occurred 189 simultaneously with fluoroquinolone resistance (Figures 4-7). One notable exception was the   The almost total absence of non-synonymous SNPs within each lineage suggested that PBP 210 substitutions accumulate by the step-wise fixation of de novo point mutations in response to 211 β-lactam selection, rather than the import of novel variants by horizontal genetic exchange.

305
PBP substitutions occurred without fluoroquinolone resistance in reasonable numbers of 306 genomes of only three toxigenic lineages studied; ST2(020/014), ST54(012), and ST42(106) 307 (USA) ( Table S1). The latter was of interest since it recently exceeded the prevalence of  The cephalosporin MICs for PBP substituted strains were extremely high for certain 317 antibiotics (for example >512µg/ml for cefotaxime, and up to 1506µg/ml for cefuroxime, to 1,345μg/ml have been reported (72) and so the potential exists for C. difficile to be 322 exposed in vivo to cephalosporin concentrations reaching the MICs measured here. Resistant 323 bacteria can also be selected experimentally at antimicrobial concentrations up to several 324 hundred-fold below lethal levels (73)(74)(75)(76)(77). However, the overall contribution made by such 325 'sub-MIC selection' to resistance in clinically important bacteria is unknown (78).

329
This suggests a wild-type phenotype which favours the acquisition of chromosomal SNPs 330 which raise cephalosporin and fluoroquinolone MICs. As discussed above, the baseline 331 cephalosporin MICs for MDR-yielding lineages were higher than those lacking MDR strains 332 ( Figure 3A). This potentially favours survival of MDR-yielding lineages in low 333 cephalosporin concentrations, allowing selection of PBP substitutions. An alternative, 334 mechanism might be a hypermutator phenotype, as in S. pneumoniae (73).

335
In summary, our findings identify a role for cephalosporin selection in the evolution of 336 epidemic CDI lineages. Specific regional prescribing practises may determine the locally 337 predominant epidemic strains, potentially explaining the marked international variation in C.  indicate genetic lineages, identified by the notation ST1(027) (sequence type-1 (PCR-359 ribotype-027)). these loci (PBP1-5, gyrA, gyrB, rpoB and ermB) was assigned a number (Table S1) and can 367 be downloaded at https://pubmlst.org/organisms/clostridioides-difficile/ (44, 85). Newly 368 extracted gene sequences were queried against this database and the allele numbers were 369 recorded for each genome, together with the substitutions relevant to AMR (Table S1).          Raw data, including the identity of each substitution are shown in Table 2.

882
The plots extend from 0.9 to 1.