Escherichia coli is the leading cause of urinary tract infection, one of the most common bacterial infections in humans. Despite this, a genomic perspective is lacking regarding the phylogenetic distribution of isolates associated with different clinical syndromes. Here, we present a large-scale phylogenomic analysis of a spatiotemporally and clinically diverse set of 907 E. coli isolates, including 722 uropathogenic E. coli (UPEC) isolates. A genome-wide association approach identifies the (P-fimbriae-encoding) papGII locus as the key feature distinguishing invasive UPEC, defined as isolates associated with severe UTI, i.e., kidney infection (pyelonephritis) or urinary-source bacteremia, from non-invasive UPEC, defined as isolates associated with asymptomatic bacteriuria or bladder infection (cystitis). Within the E. coli population, distinct invasive UPEC lineages emerged through repeated horizontal acquisition of diverse papGII-containing pathogenicity islands. Our findings elucidate the molecular determinants of severe UTI and have implications for the early detection of this pathogen.

Recurrent urinary tract infection (rUTI) remains a major problem for many women and therefore the pursuit for genomic and phenotypic traits which could define rUTI has been ongoing. The present study applied a genomic approach to investigate recurrent urinary tract infections by comparative analyses of recurrent and non-recurrent Escherichia coli isolates from general practice. From whole-genome sequencing data, phylogenetic clustering and genomic traits were studied on a collection of isolates which caused recurrent infection compared to non-recurrent isolates. In addition, genomic variation between the 1st and following infection was studied on a subset of the isolates. Evidence of limited adaptation between the recurrent infections based on single nucleotide polymorphism analyses with a range of 0–13 non-synonymous single nucleotide polymorphisms (SNPs) between the paired isolates. This included an overrepresentation of SNPs in metabolism genes. We identified several genes which were more common in rUTI isolates, including nine fimbrial genes, however, not significantly after false-discovery rate. Finally, the results show that recurrent isolates of the present dataset are not distinctive by variation in the core genome, and thus, did not cluster distinct from non-rUTI isolates in a SNP phylogeny.

Linezolid is used as first-line treatment of infections caused by vancomycin-resistant Enterococcus faecium . However, resistance to linezolid is increasingly detected. The aim of the present study was to elucidate the causes and mechanisms for the increase in linezolid-resistant E. faecium at Copenhagen University Hospital – Rigshospitalet. We therefore combined patient information on linezolid treatment with whole-genome sequencing data for vancomycin- or linezolid-resistant E. faecium isolates that had been systematically collected since 2014 (n=458). Whole-genome sequencing was performed for multilocus sequence typing (MLST), identification of linezolid resistance-conferring genes/mutations and determination of phylogenetically closely related strains. The collection of E. faecium isolates belonged to prevalent vancomycin-resistant MLST types. Among these, we identified clusters of closely related linezolid-resistant strains compatible with nosocomial transmission. We also identified linezolid-resistant enterococcus isolates not genetically closely related to other isolates compatible with de novo generation of linezolid resistance. Patients with the latter isolates were significantly more frequently exposed to linezolid treatment than patients with related linezolid-resistant enterococcus isolates. We also identified six patients who initially carried a vancomycin-resistant, linezolid-sensitive enterococcus, but from whom vancomycin-resistant, linezolid-resistant enterococci (LVRE) closely related to their initial isolate were recovered after linezolid treatment. Our data illustrate that linezolid resistance may develop in the individual patient subsequent to linezolid exposure and can be transmitted between patients in a hospital setting.

Hospitalization and treatment with antibiotics increase the risk of acquiring multidrug-resistant bacteria due to antibiotic-mediated changes in patient microbiota. This study aimed to investigate how broad- and narrow-spectrum antibiotics affect the gut microbiome and the resistome in antibiotic naïve patients during neurointensive care. Patients admitted to the neurointensive care unit were treated with broad-spectrum (meropenem or piperacillin/tazobactam) or narrow-spectrum antibiotic treatment (including ciprofloxacin, cefuroxime, vancomycin and dicloxacillin) according to clinical indications. A rectal swab was collected from each patient before and after 5–7 days of antibiotic therapy (N = 34), respectively. Shotgun metagenomic sequencing was performed and the composition of metagenomic species (MGS) was determined. The resistome was characterized with CARD RGI software and the CARD database. As a measure for selection pressure in the patient, we used the sum of the number of days with each antibiotic (antibiotic days). We observed a significant increase in richness and a tendency for an increase in the Shannon index after narrow-spectrum treatment. For broad-spectrum treatment the effect was more diverse, with some patients increasing and some decreasing in richness and Shannon index. This was studied further by comparison of patients who had gained or lost >10 MGS, respectively. Selection pressure was significantly higher in patients with decreased richness and a decreased Shannon index who received the broad treatment. A decrease in MGS richness was significantly correlated to the number of drugs administered and the selection pressure in the patient. Bray–Curtis dissimilarities were significant between the pre- and post-treatment of samples in the narrow group, indicating that the longer the narrow-spectrum treatment, the higher the differences between the pre- and the post-treatment microbial composition. We did not find significant differences between pre- and post-treatment for both antibiotic spectrum treatments; however, we observed that most of the antibiotic class resistance genes were higher in abundance in post-treatment after broad-spectrum treatment.