Hemolytic Uremic Syndrome Outbreak in Adults in France, 2025

Unbiased portrait of a LEE-negative STEC outbreak in France: design, strain features, and epidemiologic context

coli National Reference Center identified an outbreak of hemolytic uremic syndrome (HUS) affecting adults, caused by Shiga toxin–producing E.

coli that lacked the locus of enterocyte effacement (LEE).

The investigation recognized a cluster of HUS cases linked to a distinctive STEC lineage, with documentation extending to two additional European cases in Scotland and Belgium.

coli infection cases, plus 5 probable or possible cases identified by in-house PCR assays, and two supplementary international cases.

The pattern of cases and laboratory confirmation supported a cohesive outbreak series rather than sporadic, unrelated infections.

A substantial 134-kilobase plasmid within the strain housed enterotoxin genes estb-STb2 and eltAB, consistent with a hybrid pathotype combining Shiga toxin–producing and enterotoxigenic properties.

The strain lacked both the eae gene (LEE structural virulence) and the usual LEE-associated elements, yet demonstrated virulence features via alternative loci.

The outbreak strain displayed a unique repertoire of virulence determinants and capsule characteristics, which may bear on transmissibility and host range.

• In January 2025, the French E.
• The outbreak comprised 18 confirmed E.
• Whole-genome sequencing characterized the outbreak strain as an O77 g:K92:H18 serotype, placed within phylogroup D, and carrying the Shiga toxin 2 genetic variant stx2d-073-C165-02.
• Epidemiologic tracing implicated raw-milk cheese as the contamination source.

This combination marks a departure from the classic LEE-positive, highly adherent STEC strains commonly associated with HUS.

This serotypic feature potentially intersects with host immune recognition and could influence clinical manifestations across populations, including the observed lack of pediatric cases in this outbreak.

The isolates were categorized as sequence type ST69 within phylogroup D, indicating a clonal lineage with limited allelic diversity among outbreak-related samples.

Genome-level analyses included phylogroup determination, core-genome multilocus sequence typing (cgMLST) and hierarchical clustering (HC), single-nucleotide polymorphism (SNP) scrutiny, and plasmid architecture characterization.

A broad virulence gene search was conducted by constructing a curated LA (late annotation) database to align presence/absence of gene features across the outbreak genomes, with careful de-redundancy using CD-HIT.

Resistance determinants were also surveyed via Abricate against ResFinder, though the report does not specify the presence or absence of antimicrobial resistance genes in the published excerpt.

• The isolate family identified in this outbreak is distinguished by its LEE-negative status, with the absence of eae and the conventional LEE region, yet harboring a Shiga toxin gene profile dominated by stx2d.
• The strain demonstrated a capsule phenotype described as K92, which has known cross-immunogenic properties with Neisseria meningitidis group C.
• The outbreak strain was assigned to O77 g:H18, a serogroup designation that involves a composite interpretation of O antigen typing and H antigen status, further refined through phenotypic serotyping and capsular antigen analysis.
• Each outbreak isolate underwent comprehensive genomic assessment: Illumina short-read sequencing for all isolates, with three isolates subjected to hybrid assembly using long-read Oxford Nanopore data (GridION MinION) to procure closed chromosomal and plasmid configurations.
• The virulence gene landscape extended beyond stx2d, with plasmid-borne enterotoxin genes estb-STb2 and eltAB contributing to the pathogenic potential.

Positive colonies underwent confirmation via the eazyplex EHEC assay for stx1/stx2/eae/ehxA genes.

Antimicrobial susceptibility testing was performed using disk diffusion on cultured isolates, and preparations were stored at –80°C for downstream analyses.

Hybrid assemblies were employed for select isolates to achieve circularized chromosomes and plasmids.

The analysis used Panaroo for core-genome alignment and pairsnp for SNP delineation, with IQ-TREE deployed for phylogenetic inferences.

This approach sought to identify stool samples positive for outbreak-associated elements even when culture success was limited.

• The France NRC employed a multilayered diagnostic workflow to identify STEC cases: initial DNA extraction and multiplex PCR targeting stx1, stx2, and eae, followed by culture on selective media including CHROMagar STEC, MacConkey CT, and chromID CPSO.
• In addition to standard STEC detection, researchers performed targeted serogroup testing for a panel of ten major STEC serogroups with in-house multiplex PCR.
• Whole-genome sequencing pipelines integrated both short-read (Illumina) and long-read (Oxford Nanopore) data to achieve high-quality assemblies, with downstream analyses spanning phylogeny, core-genome alignment, SNP-based distance calculations, and plasmid reconstruction.
• Phylogenetic context was established by situating the outbreak strains within Enterobase hierarchical clusters (HC), specifically HC5, and comparing core-genome SNP profiles to determine clonal relatedness (median SNP differences around 4 among related isolates).
• In attempting to broaden case identification beyond culture-confirmed isolates, the investigators designed a set of in-house quadruplex PCR assays targeting plasmid-borne virulence markers (tpx region and eltA) and chromosomal targets (wzy for O77 serogroup, neuS for K92 capsule).
• To reconcile discrepancies in O antigen determination among serogroups, researchers leveraged phenotypic serotyping in Spain to confirm O77 group assignment and used the K1/neisseria meningitidis B antisera to refine capsule typing, ensuring congruence between genomic inference and phenotypic serology.

An additional 5 probable or possible cases were detected through the in-house quadruplex PCR strategy, including patients without recoverable isolates but with molecular signals linked to the outbreak genotype.

Pediatric involvement was notably absent in the outbreak cohort, despite the capacity of STEC-associated disease to affect children more commonly in France’s historical surveillance of HUS.

• The confirmed case count at the close of the study period included 18 culture-confirmed STEC infections during the outbreak window.
• The spectrum of illness included adults with HUS, a severe and potentially fatal complication associated with STEC.
• Two extra cases outside France—one in Scotland and another in Belgium—were linked to the same outbreak strain, underscoring cross-border transmission or introduction of the clonal lineage into neighboring regions.
• The authors emphasize that the HUS cases occurred predominantly in elderly adults hospitalized with HUS in continental France, aligning with the observation of a LEE-negative STEC with stx2 (not eae or ehxA) features as the etiologic driver in this setting.

In France, reported HUS incidence in children surpasses adult rates by roughly an order of magnitude, highlighting a pronounced pediatric burden under typical circumstances.

The typical virulence gene constellation includes Shiga toxin type 2 variants, especially 2a or 2d, often carried on phage-associated elements, and a chromosomal pathogenicity island known as LEE, including the eae gene encoding intimin.

The emergence of an outbreak linked to a LEE-negative STEC with stx2 and a plasmid-encoded enterotoxin repertoire signals a distinct virulence ecology.

The association with a K92 capsule adds another layer of potential interaction with host defenses and immune recognition.

The presence of the 134-kb plasmid with estb-STb2 and eltAB adds to the unique virulence plasmid content of this clone.

• HUS remains the most severe complication associated with STEC infections, defined by a triad that includes thrombocytopenia, mechanical hemolytic anemia with schistocytes, and renal injury.
• Historically, French HUS cases from STEC infections have often clustered around serogroups O26, O80, and O157, with lesser frequency attributable to O121, O145, O55, and O103.
• The LEE-negative subset of STEC typically lacks the chromosomal eae locus, which differentiates their adhesion strategies and pathogenic mechanisms from LEE-positive lineages.
• The authors note that the outbreak strain belongs to a novel cluster within HC5 and displays clonal relatedness across isolates, with core-genome SNP analyses showing tight clustering (median SNP distance around 4) among outbreak-associated genomes.

This approach demonstrates the value of molecular surveillance augmentation in detecting atypical STEC outbreaks.

This observation may inform future surveillance targeting and informs considerations about which virulence determinants to monitor in non-traditional STEC lineages.

• The detection framework employed by the NRC (France) combined routine STEC screening with targeted virulence and serogroup assays, enabling identification of a non-traditional outbreak strain that would have been missed by relying solely on standard diagnostics targeting eae or common O-serogroups.
• The integration of culture-independent PCR assays targeting plasmid- and chromosomal loci allowed broader case ascertainment by capturing stool specimens positive for outbreak markers even when culture yield was poor or when isolates could not be recovered on selective media.
• The genomic confirmation of a single clonal entity across multiple domains (France and two European countries) emphasizes the importance of cross-border data sharing and harmonized molecular typing in tracing dissemination pathways and confirming outbreak scope.
• The identification of a cross-immunogenic capsule (K92) and the absence of eae in the outbreak strain suggest that standard eae-centric risk stratification might overlook certain LEE-negative strains with significant virulence potential.

It also does not present numerical incidence rates beyond general context comparing pediatric and adult HUS incidence, nor does it provide age-specific or comorbidity-adjusted risk estimates for this outbreak.

Consequently, mechanistic assertions about pathogenesis remain inferential, grounded in genomic and epidemiologic association rather than direct experimental causality.

The long-term epidemiology and potential reservoirs remain to be fully characterized.

The extent to which asymptomatic carriage contributed to transmission pathways is not described.

• The provided report excerpt does not explicitly enumerate all clinical outcomes, such as the precise proportion of patients progressing to renal failure, duration of hospital stay, or long-term sequelae.
• While the source identifies two additional European cases associated with the same strain, it does not detail the exact epidemiologic links, travel histories, or food consumption data for those cases, limiting the reconstruction of transmission networks beyond the contaminated raw-milk cheese hypothesis for the French cases.
• The report describes a novel O77 g:H18 lineage with specific plasmid-encoded virulence determinants, but functional assays or in vivo data assessing the virulence contribution of the plasmid-encoded enterotoxins in the context of a LEE-negative background are not detailed.
• The study mentions the existence of related strains in international databases since 2005, suggesting global emergence, but it does not enumerate the geographic distribution, historical isolates, or evolutionary dynamics beyond cluster membership and SNP distances.
• Uncertainties persist about the scope of asymptomatic carriage or subclinical infection within the population, given that stool samples were collected from patients suspected of HUS or experiencing diarrhea, as well as at-risk groups.
• Although cross-border cases imply a transnational footprint, the report does not provide comprehensive information on regulatory or public health actions undertaken in Scotland or Belgium, such as recalls, product tracebacks, or cheese production chain interventions, limiting evaluation of the immediate public health response.

The absence of the eae gene in this lineage highlights the need to consider non-LEE virulence determinants and plasmid-encoded factors in diagnostic and surveillance schemas.

• This outbreak underscores the necessity of maintaining vigilance for LEE-negative STEC strains as potential causes of HUS, particularly in older adults where the clinical presentation may differ from the conventional pediatric-centric narrative.
• The characterization of a K92 capsule–bearing O77 g:H18 strain with stx2d and an effector-deficiency profile demonstrates the potential for atypical serotype and virulence gene constellations to drive significant disease in populations often considered at lower risk for HUS.