Escherichia coli resistant to the highest priority critically important fluoroquinolone or 3rd and 4th generation cephalosporin antibiotics persist in pigsties

Abstract

Antimicrobial resistance threatens human and animal health, with antimicrobial usage being a key driver of selection, transmission, and spread of resistant bacteria. Livestock represents a potential reservoir for human transmission, leading authorities to restrict veterinary usage of fluoroquinolones and certain cephalosporins. However, growing evidence indicates that the corresponding resistance determinants can be retained even in the drugs' absence. To obtain data on the magnitude and dynamics of this phenomenon in pig farming, we quantitatively and qualitatively assessed fluoroquinolone- and cephalosporin-resistant Escherichia coli in Thuringian pigsties practicing a closed management system to minimize the impact of externally introduced strains. Pooled fecal samples from consecutive fattening runs at one conventional and two organic farms and from 25 piglet groups from another conventional farm were collected over 16 months and screened for E. coli on plates containing enrofloxacin, ceftiofur, or cefquinome. Resistant bacteria were isolated on all farms; their counts varied strongly but were generally higher in piglets and declined with increasing animal age. Phylogenetic comparison of 393 isolates was performed via multiple-locus variable number tandem repeat analysis (MLVA) to follow strain dynamics and persistence. The isolates displayed large phylogenetic heterogeneity, featuring 52 different MLVA patterns. Still, conserved MLVA patterns indicated long-term persistence of specific strains in each farm's environment. This suggests that resistant strains appear well-adapted to the particular farm and its management practices, implying that, beyond restricting usage, further measures, including, e.g., consideration of the type of resistance as well as its persistence and transmission dynamics, will be indispensable to reduce the antimicrobial resistance load in pork production.IMPORTANCEAntimicrobial resistance (AMR) represents a global threat to human and animal health, with animals considered a reservoir for transmission of AMR to humans. Because antimicrobial usage is a driver for resistance, one approach to decrease the AMR burden is to reduce its usage. However, this can, but does not necessarily, lead to lower AMR prevalence. German and EU legislation restrict the use of fluoroquinolones and certain cephalosporins, substance classes designated as highest priority critically important antimicrobials for human medicine, in animal husbandry. Longitudinal sampling of organic and conventional farms in Thuringia for resistance to these antibiotic classes revealed that certain resistant Escherichia coli strains can persist in the farm environment over extended time periods. These strains displayed farm specificity, indicating adaptation to the particular farm and its management practices, so that their elimination might be difficult, requiring either procedures acting generally against Enterobacterales or targeted action against the specific strains.

Keywords: Escherichia coli; antimicrobial resistance; cephalosporins; fecal organisms; fluoroquinolones; persistence; pig farming.

Fig 1

Proportion of cephalosporin-/fluoroquinolone-resistant γ-proteobacteria in all γ-proteobacteria per age group and farm. Representation of all determined percentage values of resistant bacteria on the selective plates as box-and-whisker plots; value points slightly scattered horizontally for better visibility; antennas enclose all values within 1.5 times the interquartile range. In order to be able to logarithmically represent values of pooled fecal samples without growth (measured value 0), the value 0.001 was added to all percentage values of resistant CFU. Values above 100% were truncated at 100% (n = 3).

Fig 2

Average percentage of resistant bacteria disaggregated for selective plate and farm. Representation of all percentage values determined for resistant bacteria per selective plate and farm as box-and-whisker plots; value points are slightly scattered horizontally for better visibility; antennas enclose all values within 1.5 times the interquartile range. Values above 100% were excluded (n = 3, all on farm C2). GE = Gassner agar supplemented with 4 µg/mL enrofloxacin; GF = Gassner agar supplemented with 4 µg/mL ceftiofur; GQ = Gassner agar supplemented with 8 µg/mL cefquinome. The confidence intervals (95%) for the respective median values are presented in Table S4. Significant differences between the average percentage values of the different farms for the selective plates are not shown in the figure, as they would interfere with its clarity, but are presented in Table S3.

Fig 3

A + B: Percentage values of resistant bacteria on GE, GF, and GQ plates of all fecal samples from farms O1 (A) and C2 (B) compared between selective plate types. For a detailed description of the structure of the heatmaps, see Fig. S1A-D; GE = Gassner agar supplemented with 4 µg/mL enrofloxacin; GF = Gassner agar supplemented with 4 µg/mL ceftiofur; GQ = Gassner agar supplemented with 8 µg/mL cefquinome; FR = fattening run. The corresponding respective absolute CFU counts are given in Table S2. “0%” corresponds to less than 0.01% of the colonies detected on the Gassner agar plate without antibiotics.

Fig 4

Phylogeny tree calculated from the MLVA profiles determined for the 393 strains isolated. CI–CIII = Three clades, in which exemplary MLVA clusters have been named (letters “a” to “h”). The comparison of MLVA profiles is based on the Dice coefficient. Clusters were created with unweighted pair group method with arithmetic mean.

Fig 5

Examples of animal group-persistent (A) and farm-persistent (B) MLVA profiles. For a detailed structure of the heatmaps, see Fig. S1A-D; MP = MLVA profile (indicated on the left); FR = fattening run; gray color = at least one isolate from this pooled fecal sample with corresponding MP detected; farms marked in color; – = no cephalosporin-resistant/fluoroquinolone-resistant E. coli were isolated from this pooled fecal sample.