ANSES: in E. coli biofilms, a month of exposure to TMN or benzalkonium chloride at the MIC increased the number of rifampicin-resistant variants, and TMN selected mutations in lipopolysaccharide biosynthesis genes and colanic acid overproduction
On 2 November 2026 the French Agency for Food, Environmental and Occupational Health & Safety (ANSES) described research by its Fougères Laboratory: Escherichia coli biofilms were exposed for a month to four biocidal active substances used in food-processing disinfectants, and rifampicin-resistant variants were then counted. Two of them — N-(3-aminopropyl)-N-dodecylpropane-1,3-diamine (TMN) and benzalkonium chloride — increased their number; the results were published in npj Antimicrobials and Resistance in May 2026.
- Biofilms of eight E. coli strains — reference strain MG1655, a chicken isolate from the INRAE collection and six strains from the pork and poultry industries held by the French National Reference Laboratory for Antimicrobial Resistance — were grown for 72 h at 20 °C in microplates and then, over four one-week cycles, given fresh medium with a biocide at its MIC: benzalkonium chloride 6.25 mg/L, polyhexamethylene biguanide (PHMB) 1.25 mg/L, TMN 6.67 mg/L or sodium hypochlorite 100 mg/L (control: water). Each week resistant variants were counted on TSA with 20 mg/L rifampicin, separately from the biofilm and from the liquid above it.
- TMN increased the number of rifampicin-resistant variants in the biofilms of seven of the eight strains and benzalkonium chloride in five; PHMB and sodium hypochlorite had no such effect, and hypochlorite at its MIC completely eradicated the biofilms of three strains within seven days. In bacteria from the liquid above the biofilm, the general effect of TMN and benzalkonium chloride was not seen — the authors regard it as a biofilm-specific response.
- Under TMN, variants with mutations in genes linked to lipopolysaccharide (LPS) biosynthesis were selected in the biofilms of five of the eight strains, without mutations in rpoB and with low-level rifampicin resistance — 20–60 mg/L, compared with 40 to over 200 mg/L in rpoB mutants. Most of them had an altered LPS profile, a more negative surface charge (zeta potential, e.g. −24 instead of −19.9 mV), lower membrane permeability and more biofilm matrix with colanic acid; in biofilms they survived TMN at 400 mg/L (60 × MIC) better than their parental strains, while in planktonic suspension there was no difference.
- Mutations were identified by whole-genome sequencing (Illumina) and confirmed by Sanger sequencing, the lineage of the variants by ERIC-PCR; LPS was examined by SDS-PAGE with silver staining, surface charge by zeta potential, permeability by a crystal violet uptake assay, matrix by Congo red fluorescence and the protein response by DIA-PASEF proteomics. An earlier study by the same team (Frontiers in Cellular and Infection Microbiology, 2023) showed that PHMB raises gentamicin resistance in E. coli biofilms without genetic selection — the bacteria lost it once exposure stopped; according to ANSES the risk is particularly relevant where surfaces are disinfected very frequently, i.e. in the food industry and in hospitals.
Source: ANSES — Shared resistance mechanisms for biocides and antibiotics (2.11.2026)