Skip to main content
NCBI home page
Microbiology Spectrum logoLink to Microbiology Spectrum
. 2023 Sep 29;11(5):e02220-23. doi: 10.1128/spectrum.02220-23

Molecular characterization of highly prevalent Escherichia coli and Escherichia marmotae resistant to extended-spectrum cephalosporins in European starlings (Sturnus vulgaris) in Tunisia

Meriem Souguir 1,2, Pierre Châtre 2, Antoine Drapeau 2, Pauline François 2, Sana Azaiez 1, Sana Ncir 1,2, Jean-Yves Madec 2, Wejdene Mansour 1, Marisa Haenni 2,✉
Editor: Po-Yu Liu3
PMCID: PMC10581222  PMID: 37772831

ABSTRACT

European starlings are widespread migratory birds that have already been described as carrying bacteria resistant to extended-spectrum cephalosporins (ESC-R). These birds are well known in Tunisia because they spend the wintertime in this country and are hunted for human consumption. The goal of our study was to estimate the proportion of ESC-R in these birds and to characterize the collected isolates using whole-genome sequencing. Results showed that 21.5% (42/200) of the birds carried either an extended-spectrum beta-lactamase (ESBL) or an acquired AmpC gene. Diverse bla CTX-M genes were responsible for the ESBL phenotype, bla CTX-M-14 being the most prevalent, while only bla CMY-2 and one bla CMY-62 were found in AmpC-positive isolates. Likewise, different genetic determinants carried these resistance genes, including IncHI2, and IncF plasmids for bla CTX-M genes and IncI1 plasmids for bla CMY-2 genes. Three chromosomally encoded bla CTX-M-15 genes were also identified. Surprisingly, species identification revealed a large proportion (32.7%) of Escherichia marmotae isolates. This species is phenotypically indistinguishable from Escherichia coli and has obviously the same capacity to acquire ESC-R genes. Our data also strongly suggest that at least the IncHI2/pST3 plasmid can spread equally between E. coli and E. marmotae. Given the potential transmission routes between humans and animals, either by direct contact with dejections or through meat preparation, it is important to closely monitor antimicrobial resistance in European starlings in Tunisia and to set up further studies to identify the sources of contamination of these birds.

IMPORTANCE

The One Health concept highlighted knowledge gaps in the understanding of the transmission routes of resistant bacteria. A major interest was shown in wild migratory birds since they might spread resistant bacteria over long distances. Our study brings further evidence that wild birds, even though they are not directly submitted to antibiotic treatments, can be heavily contaminated by resistant bacteria. Our results identified numerous combinations of resistance genes, genetic supports, and bacterial clones that can spread vertically or horizontally and maintain a high level of resistance in the bird population. Some of these determinants are widespread in humans or animals (IncHI2/pST3 plasmids and pandemic clones), while some others are less frequent (atypical IncI1 plasmid and minor clones). Consequently, it is essential to be aware of the risks of transmission and to take all necessary measures to prevent the proportions of resistant isolates from increasing uncontrollably.

KEYWORDS: E. marmotae, starlings, Tunisia, ESBL, AmpC, plasmids, One Health, spread

INTRODUCTION

Antimicrobial resistance (AMR) is a complex phenomenon that knows no barriers and is threatening humans, animals, and the environment worldwide. Enterobacterales resistant to extended-spectrum cephalosporins (ESC-R) and carbapenems (CP) are among the most significant issues. AMR has long been overlooked in wildlife (1) but, for the last decade, clinically relevant AMR bacteria—including ESC-R and CP-R—have been isolated from various wild bird species on all continents, including vultures, gulls, buzzards, or red kites (2 – 4). The presence of resistance genes and antibiotic-resistant bacteria in wildlife is most probably an indicator of anthropogenic contamination rather than resistance selection since wild animals are usually not directly exposed to antibiotics (5). Consequently, wild birds can be both reservoirs and spreaders of resistance genes or bacteria of human origin. Moreover, migratory birds, through their movements over long distances, can be partly responsible for the worldwide transmission of resistance genes (6).

Starlings (Sturnus vulgaris or European starling) are wild birds native to Europe and South-East Asia that have later reached to North America (7), Australia, New Zealand, and South Africa (8). They are great travelers, migrating over long distances (1,000–1,500 km) searching for food in winter (8). Starlings are so widespread and problematic that they were classified among the three worst invasive birds on the World Conservation Union List (9, 10). They can be potential reservoirs of bacteria as they have been recurrently described as carrying Campylobacter jejuni isolates (11, 12). Cefotaxime- and ciprofloxacin-resistant Escherichia coli were also reported from European starlings in Canada, and one ESC-R SHV-12-producing E. coli was reported from a Spotless starling in Spain (13 – 15).

Tunisia is on the migratory route of starlings that are traveling from Europe for the winter season before leaving in spring. During this 6-month period, starlings are traditionally hunted for human consumption and sold alive on local markets. While consumption of meat produced in intensive farms (such as poultry meat) where animals are treated with antibiotics has been widely studied, the risk of consuming wild animal meat has been much less evaluated. Our study thus aimed at isolating and characterizing ESC-R and CP-R Enterobacterales, in order to determine whether these birds carry resistant bacteria, which could be further transmitted through the food chain or by fecal contamination of domestic animals and the environment.

RESULTS

Prevalence and genetic diversity of ESBL/AmpC-producing isolates

Over the 200 samples tested, corresponding to 200 birds, 42 (42/200, 21.5%) presented growth on plates containing cefotaxime, while none of them grew on imipenem-containing plates. One isolate was picked per plate/bird, except for one plate that presented two colony morphologies, leading to a final collection of 43 Enterobacterales.

Isolates were identified as E. coli (n = 27, 62.8%) and Escherichia marmotae (n = 16, 37.2%). All E. marmotae isolates belonged to ST133, except for two presenting a common unknown sequence type (ST) (Table S1; Fig. 1). ST133 isolates were genetically linked and clustered according to their geographical origin; isolates from Gabès were nearly identical with 0–4 single-nucleotide polymorphism (SNP) differences, but differed by 42 SNPs from the Bizerte isolate and by 192–198 SNPs from the Zaghouan isolates (Table S2; Fig. S1). Only one isolate from Gabès (#60301) was genetically closer to the Bizerte isolate (#60314) than from the other isolates from Gabès. On the contrary, E. coli isolates belonged to 20 different sequence types, including 1 unknown and 19 already described STs: ST5451 was found in three different samples, while ST126, ST162, ST973, and ST1177 were each found in two samples. Three STs belonged to pandemic clones, namely ST38, ST155, and ST162.

Fig 1.

Fig 1

Open in a new tab

SNP phylogeny of the E. coli and E. marmotae isolates using iTOL. Two isolates (#60314 and #60289) presented both the bla CTX-M-14 and the bla CMY-2 gene. The bla CTX-M-14 gene was located on an IncHI2/pST3 plasmid in both isolates, while the bla CMY-2 gene was located on an IncI1 plasmid in isolate #60289 and on an undetermined genetic element in isolate #60314.

ESBL and AmpC resistance genes

Twenty-three isolates (12 E. coli and 11 E. marmotae; 23/200, 11.5%) displayed an extended-spectrum beta-lactamase (ESBL) phenotype, and 22 (16 E. coli and 6 E. marmotae; 11%) an AmpC phenotype, among which two isolates presented a mixed ESBL/AmpC phenotype (Table S1). The ESBL phenotype was due to the presence of the bla CTX-M-14 gene in all E. marmotae isolates, while it was conferred by the bla CTX-M-15 (n = 4), bla CTX-M-27 (n = 4), bla CTX-M-14 (n = 3), and bla CTX-M-1 (n = 1) genes in E. coli isolates (Table S1). AmpC-positive isolates systematically displayed the bla CMY-2 gene, except for one E. coli isolate that presented the bla CMY-62 gene. One E. coli and one E. marmotae co-harbored the bla CTX-M-14 and bla CMY-2 genes. The 14 bla CTX-M-14-positive isolates originated from the three cities where birds were caught (Gabès, Bizerte, and Zaghouan), while 20/21 of the bla CMY-2-positive isolates originated from Zaghouan.

Antibiotic susceptibility testing and associated resistance genes

All ESBL-producing isolates presented resistance to non-β-lactam antibiotics, while those only displaying the bla CMY-2 gene were susceptible to all non-β-lactam antibiotics tested (Table S1). E. marmotae isolates presented globally less resistance to non-β-lactams than E. coli. Streptomycin was the most frequently found resistance in both E. coli and E. marmotae isolates (51.2%), followed by tetracyclines, sulfonamides, and trimethoprim (25.6%, 23.3%, and 11.6%, respectively). Resistance proportions were less than 5% to enrofloxacin (2.3%) and gentamicin (4.7%) (Table 1). Aminoglycoside resistance was principally due to the co-occurrence of the aph(6)-Id and aph(3″)-Ib genes (n = 22), in both E. coli (n = 11) and E. marmotae (n = 11) (Table S1). The fosA3 gene conferring high-level resistance to fosfomycin was also found in both E. coli (n = 3) and E. marmotae (n = 11). On the contrary, genes conferring resistance to sulfonamides [sul1 (n = 3), sul2 (n = 7), and sul3 (n = 1)], trimethoprim [drfA1 (n = 2) and drfA17 (n = 3)], tetracyclines [tet(A), n = 9] as well as quinolones [qnrS1 (n = 1)] were only identified in E. coli. Finally, the mcr-1 gene present in the ST38 E. coli isolate was truncated by the ISApl1, explaining why colistin resistance was not observed.

TABLE 1.

Antimicrobial susceptibility phenotypes of all 43 E. coli and E. marmotae isolates characterized in this study a

Escherichia coli (n = 27) Escherichia marmotae (n = 16) Total (n = 43)
n % n % n %
Amoxicillin 27 100.0 16 100.0 43 100.0
Amoxicillin + clavulanic acid 21 77.8 6 37.5 27 62.8
Ceftiofur 27 100.0 16 100.0 43 100.0
Cefquinome 12 44.4 13 81.25 25 58.1
Ceftazidime 24 88.9 6 37.5 30 69.8
Cefoxitin 17 63.0 5 31.25 22 51.2
Ertapenem 0 0.0 0 0.0 0 0.0
Sulfonamides 9 33.3 1 6.25 10 23.3
Trimethoprim 5 18.5 0 0.0 5 11.6
Nalidixic acid 7 25.9 0 0.0 7 16.3
Enrofloxacin 1 3.7 0 0.0 1 2.3
Colistin 0 0.0 0 0.0 0 0.0
Chloramphenicol 4 14.8 0 0.0 4 9.3
Florfenicol 2 7.4 0 0.0 2 4.7
Tetracycline 10 37.0 1 6.25 11 25.6
Gentamicin 2 7.4 0 0.0 2 4.7
Streptomycin 12 44.4 10 62.5 22 51.2
Kanamycin 0 0.0 0 0.0 0 0.0
Tobramycin 2 7.4 0 0.0 2 4.7
Netilmicin 0 0.0 0 0.0 0 0.0
Apramycin 0 0.0 0 0.0 0 0.0
Amikacin 0 0.0 0 0.0 0 0.0
Open in a new tab
a

Tests were performed using disc diffusion, except for colistin, for which the minimum inhibitory concentration was determined using broth microdilution.

Characterization of the genetic determinants carrying ESBL/AmpC genes

Combined analysis of Southern blots, short-read, and long-read sequences proved that the plasmid-borne bla CTX-M genes were carried by large IncHI2 or IncF plasmids. The bla CTX-M-15 gene was carried on the chromosome in three isolates, and on an undetermined genetic determinant in the fourth one. All four bla CTX-M-15 genes were preceded by the ISEcp1 element and the two chromosomally encoded genes were followed by a tryptophan synthase beta chain gene in a genetic environment similar to the one described by Guenther et al. in the ST38 IMT37356 isolate (16). Alternatively, bla CMY-2 genes were carried either on IncI1 plasmids or on the chromosome.

bla CTX-M-14-carrying plasmids

The bla CTX-M-14 gene, which was found both in E. coli (n = 3, including two clonal ST126 isolates with no SNP difference) and E. marmotae (n = 11) isolates, was systematically carried by IncHI2/ST3 plasmids and co-harbored the fosA3, aph(6)-Id, and aph(3″)-Ib genes. Three plasmids were fully sequenced, namely p60283-CTX-M-14 (190,524 bp), p60289-CTX-M-14 (199,906 bp), and p60303-CTX-M-14 (201,374 bp). They were nearly identical with a coverage >97% and presented a GC content of 45.3% (Fig. 2A). The bla CTX-M-14 and fosA3 genes were identified in the IS26-ΔISEcp1-bla CTX-M-14-ΔIS903B-fosA3-orf1-Δorf2-IS26 genetic context, which is identical to the type V fosA3 surrounding region also identified on an IncHI2/pST3 plasmid by Yang et al. (17). Downstream were the aph(6)-Id and aph(3″)-Ib genes (also named strA-strB), preceded by an IS903B-ΔIS1133 element identical to the one described by Jarocki et al., which was identified on an E. coli IncHI2/pST3 plasmid isolated from a piglet in Australia (pF2_18C_HI2; accession number: CP043545.1) (18). The three bla CTX-M-14-carrying plasmids exhibited 99% identity and a 94%–97% coverage with this pF2_18C_HI2 plasmid and with p280_128 (accession number: CP045449.1; Salmonella enterica subsp. serovar Schwarzengrund isolates collected from a poultry in Brazil) (19). Both plasmids lacked the bla CTX-M-14 and fosA3 genes, but p280_12888 displayed a bla CTX-M-2 gene (Fig. 3A).

Fig 2.

Fig 2

Open in a new tab

Visualization of the long-read sequenced plasmids carrying the bla CTX-M-14 (A), bla CTX-M-27 (B), bla CTX-M-1 (C), and the bla CMY-2 (D) genes using the PROKSEE tool.

Fig 3.

Fig 3

Open in a new tab

Comparison of plasmids carrying the bla CTX-M-14 (A), bla CTX-M-27 (B), bla CTX-M-1 (C), and the bla CMY-2 (D) genes with the closest plasmids found in the publically available databases using the PROKSEE tool.

bla CTX-M-27-carrying plasmids

The bla CTX-M-27 gene was found in four E. coli isolates, including the two clonal ST5451 ones (0 SNP difference). It was carried on IncF/F2:A-:B10 plasmids in three isolates and on the chromosome in the fourth one. In this latter isolate, the bla CTX-M-27 gene was preceded by an IS908B element and the isolate additionally presented the aadA1, aph(6)-Id, aph(3″)-Ib, tet(A), sul3, floR, and dfrA1 resistance genes. The two fully sequenced p60294-CTX-M-27 (113,851 bp) and p60304-CTX-M-27 (109,448 bp) were highly similar (>99% identity, GC content around 51.9%) and co-harbored the aph(6)-Id, aph(3″)-Ib, tet(A), and sul2 resistance genes (Fig. 2B). In both plasmids, the bla CTX-M-27 gene was flanked by two copies of the IS26, while the downstream IS26 and a third IS26 surrounded the sul2-aph(3″)-Ib-aph(6)-Id-tet(A) genes. Comparisons with the NCBI database showed that both plasmids shared >95% identity and 99.9% coverage with two plasmids (OX359170.1 and OX359164.1) found in E. coli strains isolated from human urine samples in a hospital in Norway (Fig. 3B). These two plasmids contained the same resistance genes found in the plasmids of the present study as well as an additional sul2 gene and shared the same genetic organization with flanking IS26 elements.

bla CTX-M-1-carrying plasmid

The bla CTX-M-1 gene was carried by a large (251,934 bp) IncHI2/ST4 plasmid named p60280-CTX-M-1 which co-harbored the bla TEM-1B, mph(B), two copies of aadA1, dfrA1, catA1, tet(A), and a truncated mcr-1 gene (Fig. 2C). The bla CTX-M-1 gene was preceded by an ISEcp1 insertion sequence, but not located close to other resistance genes. p60280-CTX-M-1 shared 99.9% sequence identity and 87% coverage with the 240,662 bp 180-PT54 plasmid (CP015833.1) which was found in an E. coli O157 strain isolated from a human fecal sample with diarrhea in 2012 in the United Kingdom (20). The genetic backbones of these two plasmids were similar but they differed in the resistance region, since only the dfrA1 and tet(A) genes were shared (Fig. 3C).

bla CMY-2-carrying plasmids

The 21 bla CMY-2 genes and the unique bla CMY-62 gene were identified in six E. marmotae ST133 and 16 E. coli belonging to 15 different STs (two isolates belonged to ST162). The bla CMY-62-positive ST5451 differed by 17 SNPs from the two bla CTX-M-27-positive ST5451 isolates, while the E. marmotae ST133 were identical (0–5 SNPs) when originating from the same city and differed by nearly 200 SNPs when originating from Zaghouan or Bizerte (Table S2). Isolates were all devoid of additional resistance genes, except 16 isolates which carried the chromosomally encoded mdf(A) gene, and two (one E. coli and one E. marmotae) which carried the aph(6)-Id, aph(3″)-Ib, and fosA3 genes, together with the bla CTX-M-14 gene, on an additional IncHI2 plasmid. Seventeen bla CMY-2 genes and the bla CMY-62 gene were carried on ca. 90 kbp IncI1 plasmids, three bla CMY-2 genes were chromosomally encoded (#60286, #60307, and #60320), while the genetic determinant of the last bla CMY-2 gene could not be determined (Table S1; Fig. 1). All IncI1 plasmids from E. coli were untypable since they lacked the repI1 gene, while those from E. marmotae presented an unknown repI1 allele close to repI1_12. The four other pMLST alleles were identical in all E. marmotae and E. coli isolates and belonged to the 29/15/11/3 allelic profile.

Three IncI1 plasmids were fully sequenced, which showed 99% identity over their total length and a GC content of 49.4% (Fig. 2D). The AmpC gene was found on an ISEcp1-bla CMY-2- blc-sugE fragment (3,735 bp) integrated in the finQ gene, a fragment also observed in Illumina sequences from other bla CMY-2-positive isolates. These plasmids shared 99% identity over up to 90% of their length with p113k (CP025339.1), pR18.0409_97k (CP100709.1), and pCMY2_DNA (LC019731.1). p113k and pR18.0409_97k were both isolated from Salmonella enterica serovar Typhimurium, the first one from a sick pig (Sus scrofa domesticus) (21) and the second one from a human stool sample, while pCMY2_DNA was found in an E. coli from human stool (22) (Fig. 3D). These three plasmids carried a bla CMY-2 gene, presented a pMLST closely related to our starling isolates (1/4/15/11/3 assigned to pST217 for p113k; 2/3/15/11/3 for pCMY2_DNA; and 1/4/15/11/2 assigned to pST52 for pR18.0409_97k) and all originated from Taiwan.

DISCUSSION

ESC-R E. coli have spread in all One Health sectors including wildlife (23), so that they have become a key indicator in the global monitoring of AMR and a marker of environmental contamination related to human activity (24 – 26). ESBL-producing E. coli isolates were first described from wild birds in 2006 in Portugal (27) and numerous cases from a wide variety of bird species have been reported since then throughout the world (28). In this study, we observed a high prevalence of ESC-R Escherichia spp. isolates (21.5%) in European starlings, due to the presence of both ESBL-conferring (11.5%) and AmpC-conferring (11.0%) resistance genes. However, the origin of these resistance genes and resistant bacteria remains unknown; indeed, they might have been acquired in Tunisia as well as on the birds’ migratory routes. This is the first description of ESC-R-positive European starlings in North Africa, after their report in North America where 4% of the birds tested were ESC-resistant (13), and the detection of one SHV-12-positive E. coli in a Spotless starling in Spain (15). The carriage of ESC-R determinants in these widespread migratory birds is of concern and further studies are needed to decipher the routes of contamination, including through meat consumption, and the risk of contamination for humans and animals in contact. Also, other factors that play an important role in the evolution of the bird microbiota, such as diet, sex, age, feeding habits, and geographical location—which were not recorded here and are often overlooked—should be taken into account in future studies.

Here, 62.8% of the isolates were identified as E. coli (n = 27). These isolates were genetically highly diverse since they belonged to 20 different STs, indicating the absence of the large-scale spread of one or two successful clones among these birds that live in large groups. Only ST38, ST155, and ST162 belonged to pandemic clones. The remaining 37.2% (n=16) were identified as E. marmotae, among which 14/16 belonged to ST133. These isolates mainly clustered according to their city of origin, differing by 0–5 SNPs when originating from the same city and by 45 or 198 SNPs when originating from different cities (Table S2). This strongly suggests inter-individual transmissions and micro-evolutions at the city level. Only one E. marmotae from Gabès (#60313) clustered with the unique isolate collected from Bizerte (#60314); this genetic proximity cannot be explained since the two cities are 475 km apart. E. marmotae, renamed in 2015 after its host Marmota himalayana (29), was formerly known as E.coli cryptic clade V, a clade especially associated with birds (30). This new species, which is phenotypically identical to E. coli, has since then been reported as a bla CTX-M-1 carrier from an Alpine marmot in a Belgian zoo (31), and as a bla CTX-M-32 carrier from a red deer in Poland (32); these reports, in addition to the starlings’ samples here, suggest that wild animals are preferential hosts of E. marmotae. A recent publication analyzing the genomes of 41 E. marmotae isolates suggested that this species can be pathogenic for humans but is still carrying few resistance determinants, even if sporadic cases of ESBL- and carbapenemase-producing isolates were reported (33). Our study revealed that E. marmotae can be a frequent carrier of AmpC- and ESBL-conferring genes that can be shared with E. coli isolates present in the same niche.

The ESBL phenotype was due to the presence of four different genes (bla CTX-M-1, bla CTX-M-15, bla CTX-M-14, and bla CTX-M-27) but only bla CTX-M-14 was identified in E. marmotae. The bla CTX-M-14 gene was carried by an IncHI2/ST3 plasmid in all E. coli and E. marmotae isolates, suggesting that this plasmid can spread easily and equally in both Escherichia species. This plasmid was also geographically widespread since it has been isolated in all three cities where birds were captured, even though it was much less present in Zaghouan. The backbone of this IncHI2 plasmid has already been described in E. coli and Salmonella Schwarzengrund isolates displaying different resistance genes. The resistance gene modules aph(6)-Id-aph(3″)-Ib and bla CTX-M-14-fos3 described here were each flanked by insertion sequences and displayed genetic organizations already identified in other non-identical IncHI2/pST3 plasmids, indicating the plasticity of the resistance region. The presence of the fosA3 gene is of concern since fosfomycin is one of the last resort antibiotics to treat carbapenemase-producing Gram-negative bacteria. The fosA3 gene has often been associated with ESBL-conferring genes (34), including on IncHI2 plasmids in the same genetic context as described here in both E. coli and E. marmotae (17, 35), but it has only been reported twice in wild birds: in a German black kite and a frigate bird in a pristine Brazilian atoll (36, 37). Its presence in migratory birds living in large colonies favoring close contacts and genetic transfers should thus be monitored.

The second genetic determinants shared between E. coli and E. marmotae were the IncI1/bla CMY-2 plasmids, which have only been identified in Zaghouan. These plasmids only carried the bla CMY-2 gene with no additional gene conferring resistance to non-β-lactam antibiotics, as often seen in IncI1/pST2 and pST12 subtypes (38). The ISEcp1-bla CMY-2- blc-sugE fragment, which is a common genetic environment for bla CMY-2, was always found in the finQ gene (38, 39). All plasmids were genetically highly similar, except that those identified in E. coli lacked the repI1 allele. Given the rare allelic profile identified in both E. coli and E. marmotae for the four other genes of the pMLST scheme, we can hypothesize that the IncI1 plasmid with the complete pMLST profile is the common ancestor, which then underwent recombination and excision before spreading in E. coli. Interestingly, plasmids showing the closest pMLST types all originated from humans or pigs in Taiwan, except two originating from a dog in the UK and from an unknown origin in Sweden.

In conclusion, our study revealed a surprisingly high prevalence of ESC-R isolates in European starlings in Tunisia. This was mostly due to the epidemic success of the bla CTX-M-14/IncHI2/pST3 mostly in Bizerte and Gabès, and bla CMY-2/IncI1 plasmids exclusively in Zaghouan. In addition, a few more sporadic resistance determinants were found, such as bla CTX-M-27/IncF/F2:A-:B10, bla CTX-M-1/IncHI2/ST4, and the chromosomally encoded bla CTX-M-15 genes. Our results also highlighted the importance of E. marmotae—and notably the ST133 lineage—as an ESC-R commensal bacterial species in wild birds, in coherence with other reports of E. marmotae in wildlife. The absence of data on the resistance genes/plasmids found in humans, animals, and the environment of these three cities, and especially in the olive farms that are seasonally invaded by starlings, is a limitation of this study—as well as a potential perspective for further studies, since the real risk of AMR transmission cannot be assessed.

MATERIALS AND METHODS

Wild bird sampling and bacterial isolation

Wild birds (n = 200) were purchased alive on the Sousse market, Tunisia, between January and February 2022 during the legal hunting period. Birds were caught in Bizerte, Zaghouan, or Gabès (Fig. S1), and kept in aviaries for as short a time as possible, without being fed. The intestine of each bird was collected immediately after the death of the bird using a sterile scalpel and stored at −20°C. Intestinal content was emptied in 10 mL of Trypto-casein soy broth (Biokar), homogenized and incubated for 18–24 h at 37°C. Overnight cultures were inoculated on selective MacConkey agar plates supplemented with cefotaxime or imipenem (2 mg/L), for the detection of ESC- or CP-resistant Enterobacterales. One colony per morphology and per plate was picked up. Identification was performed using API20E galleries (bioMérieux).

Antimicrobial susceptibility testing

Susceptibility testing was performed on all 43 non-duplicate E. coli and E. marmotae using the disc diffusion method on Mueller-Hinton agar, according to the guidelines and clinical breakpoints of the Antibiogram Committee of the French Society for Microbiology (CA-SFM; www.sfm-microbiologie.org). The E. coli ATCC 25922 strain was used as quality control. A total of 16 β-lactam (amoxicillin, piperacillin, ticarcillin, amoxicillin/clavulanic acid, piperacillin/tazobactam, ticarcillin/clavulanic acid, cefalotin, cefuroxime, cefotaxime, ceftiofur, ceftazidime, cefoxitin, cefepime, cefquinome, aztreonam, and ertapenem) and 14 non-β-lactam (tetracycline, kanamycin, tobramycin, gentamicin, amikacin, apramycin, netilmicin, streptomycin, florfenicol, chloramphenicol, sulfonamides, trimethoprim, nalidixic acid, and enrofloxacin) antibiotics was tested. ESBL producing Enterobacterales were detected using the Double Disc Synergy Test. Minimum inhibitory concentrations were determined by broth microdilution for colistin, according to the European Committee for Antimicrobial Susceptibility Testing.

Illumina short-read sequencing and data analyses

DNA was extracted using the NucleoSpin Microbial DNA extraction kit (Macherey-Nagel, Hoerdt, France) and sequencing was performed on a NovaSeq 6000 instrument (Illumina, San Diego, CA, USA). Quality control of the reads was performed using FastQC and low-quality sequences were trimmed using Trimmomatic v0.39. De novo assembly was performed using Shovill v1.0.4 and the quality of assemblies was assessed using QUAST v5.0.2 (Table S3). Identification was performed using Kraken (https://github.com/DerrickWood/kraken), STs according to Achtman’s MLST scheme, resistance genes and virulence factors were determined using the CGE online tools (http://www.genomicepidemiology.org/) MLSTFinder v2.0, ResFinder v4.1, and VirulenceFinder 2.0.3, while replicon content and plasmid formula were identified using PlasmidFinder 2.0.1 and pMLST 2.0. Serotypes were determined using SeroTypeFinder2.0. Detection of the phylogenetic groups of each E. coli isolate was performed in silico using Clermontyping online tool (http://clermontyping.iame-research.center/). The phylogenetic analysis was performed using Roary v.3.11.0 as already described (40). Pairwise SNP distances were calculated from core genome alignments generated by Roary using snp-dists (https://github.com/tseemann/snp-dists) (Table S2). Visualizing was performed using iTOL v6.5.2 (https://itol.embl.de/).

MinION long-read sequencing

MinION long-read sequencing libraries were prepared according to the Oxford Nanopore Technologies using the native barcoding expansion kit (EXP-NBD104) and the ligation sequencing kit (SQK-LSK109). Sequencing was performed on a MinION sequencer using a SpotON Mk 1 R9 version flow cell (FLO-MIN106D). Assembly of both Illumina and Nanopore reads was performed using Unicycler. The assembled contigs were annotated using Bakta (Web version 1.7.0/DB: 5.0.0) (41).

Characterization of plasmids

Molecular characterization

The genetic determinants carrying the ESBL/AmpC genes were detected by Southern blot on I-Ceu1- or S1-digested DNA Pulsed-Field Gel Electrophoresis (PFGE) as already described (42), using the DIG DNA Labeling and Detection Kit (Roche Diagnostics, Meylan, France) according to the manufacturer’s instructions.

In silico characterization

The PROKSEE server (https://proksee.ca/) was used to generate high-quality navigable maps of each circular plasmid described and the visualization of genetic environment of chromosomal genes (43). Circular comparison between the plasmids belonging to the same incompatibility group and the closest plasmids found in the NCBI database was carried out on the PROKSEE server using BLAST analysis (BLAST+ 2.12.0).

ACKNOWLEDGMENTS

This work was funded by the French Agency for Food, Environmental and Occupational Health Safety (ANSES) and by the PHC-Utique/Campus France (AMR1Health, grant no. 46213QA).

Contributor Information

Marisa Haenni, Email: marisa.haenni@anses.fr.

Po-Yu Liu, Taichung Veterans General Hospital, Taichung, Taiwan .

DATA AVAILABILITY

The project was deposited in DDBJ/EMBL/GenBank under the BioProject accession number PRJNA976065.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/spectrum.02220-23.

Fig. S1. spectrum.02220-23-s0001.tif.

Map of the sampling.

Click here for additional data file. (2.2MB, tif)
DOI: 10.1128/spectrum.02220-23.SuF1
Table S1. spectrum.02220-23-s0002.xlsx.

List and characteristics of all E. coli and E. marmotae strains.

Click here for additional data file. (45.6KB, xlsx)
DOI: 10.1128/spectrum.02220-23.SuF2
Table S2. spectrum.02220-23-s0003.xlsx.

SNP distances between all E. coli and E. marmotae isolates.

Click here for additional data file. (31.5KB, xlsx)
DOI: 10.1128/spectrum.02220-23.SuF3
Table S3. spectrum.02220-23-s0004.xlsx.

Quality controls of all E. coli and E. marmotae data.

Click here for additional data file. (19.3KB, xlsx)
DOI: 10.1128/spectrum.02220-23.SuF4

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.

REFERENCES

  • 1. Dolejska M, Literak I. 2019. Wildlife is overlooked in the epidemiology of medically important antibiotic-resistant bacteria. Antimicrob Agents Chemother 63:e01167-19. doi: 10.1128/AAC.01167-19 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Blanco G, López-Hernández I, Morinha F, López-Cerero L. 2020. Intensive farming as a source of bacterial resistance to antimicrobial agents in sedentary and migratory vultures: implications for local and transboundary spread. Sci Total Environ 739:140356. doi: 10.1016/j.scitotenv.2020.140356 [DOI] [PubMed] [Google Scholar]
  • 3. Vittecoq M, Laurens C, Brazier L, Durand P, Elguero E, Arnal A, Thomas F, Aberkane S, Renaud N, Prugnolle F, Solassol J, Jean-Pierre H, Godreuil S, Renaud F. 2017. VIM-1 carbapenemase-producing Escherichia coli in gulls from southern France. Ecol Evol 7:1224–1232. doi: 10.1002/ece3.2707 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Guenther S, Aschenbrenner K, Stamm I, Bethe A, Semmler T, Stubbe A, Stubbe M, Batsajkhan N, Glupczynski Y, Wieler LH, Ewers C. 2012. Comparable high rates of extended-spectrum-beta-lactamase-producing Escherichia coli in birds of prey from Germany and Mongolia. PLoS One 7:e53039. doi: 10.1371/journal.pone.0053039 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Ramey AM, Ahlstrom CA. 2020. Antibiotic resistant bacteria in wildlife: perspectives on trends, acquisition and dissemination, data gaps, and future directions. J Wildl Dis 56:1–15. [PubMed] [Google Scholar]
  • 6. Wang J, Ma Z-B, Zeng Z-L, Yang X-W, Huang Y, Liu J-H. 2017. The role of wildlife (wild birds) in the global transmission of antimicrobial resistance genes. Zool Res 38:212. doi: 10.24272/j.issn.2095-8137.2017.024 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Hofmeister NR, Werner SJ, Lovette IJ. 2021. Environmental correlates of genetic variation in the invasive European starling in North America. Mol Ecol 30:1251–1263. doi: 10.1111/mec.15806 [DOI] [PubMed] [Google Scholar]
  • 8. Linz GM, Homan HJ, Gaulker SM, Penry LB, Bleier WJ. 2007. European starlings: a review of an invasive species with far-reaching impacts. Managing vertebrate invasive species 24 [Google Scholar]
  • 9. Bodt LH, Rollins LA, Zichello JM. 2020. Contrasting mitochondrial diversity of European starlings (Sturnus vulgaris) across three invasive continental distributions. Ecol Evol 10:10186–10195. doi: 10.1002/ece3.6679 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Lowe S, Browne M, Boudjelas S, Poorter M. 2002. 100 of the world’s worst invasive alien species, a selection from the global invasive species database. The Invasive Species Specialist Group (ISSG) a specialist group of the Species Survival Commission (SSC) of the World Conservation Union (IUCN). [Google Scholar]
  • 11. Cabe PR. 2021. European Starlings (Sturnus vulgaris) as vectors and reservoirs of pathogens affecting humans and domestic livestock. Animals (Basel) 11:466. doi: 10.3390/ani11020466 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Colles FM, McCarthy ND, Howe JC, Devereux CL, Gosler AG, Maiden MCJ. 2009. Dynamics of Campylobacter colonization of a natural host, Sturnus vulgaris (European starling). Environ Microbiol 11:258–267. doi: 10.1111/j.1462-2920.2008.01773.x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Chandler JC, Anders JE, Blouin NA, Carlson JC, LeJeune JT, Goodridge LD, Wang B, Day LA, Mangan AM, Reid DA, Coleman SM, Hopken MW, Bisha B. 2020. The role of European starlings (Sturnus vulgaris) in the dissemination of multidrug-resistant Escherichia coli among concentrated animal feeding operations. Sci Rep 10:11978. doi: 10.1038/s41598-020-68365-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Medhanie GA, Pearl DL, McEwen SA, Guerin MT, Jardine CM, Schrock J, LeJeune JT. 2017. Spatial clustering of Escherichia coli with reduced susceptibility to cefotaxime and ciprofloxacin among dairy cattle farms relative to European starling night roosts. Zoonoses Public Health 64:204–212. doi: 10.1111/zph.12296 [DOI] [PubMed] [Google Scholar]
  • 15. Alcalá L, Alonso CA, Simón C, González-Esteban C, Orós J, Rezusta A, Ortega C, Torres C. 2016. Wild birds, frequent carriers of extended-spectrum β-lactamase (ESBL) producing Escherichia coli of CTX-M and SHV-12 types. Microb Ecol 72:861–869. doi: 10.1007/s00248-015-0718-0 [DOI] [PubMed] [Google Scholar]
  • 16. Guenther S, Semmler T, Stubbe A, Stubbe M, Wieler LH, Schaufler K. 2017. Chromosomally encoded ESBL genes in Escherichia coli of ST38 from Mongolian wild birds. J Antimicrob Chemother 72:1310–1313. doi: 10.1093/jac/dkx006 [DOI] [PubMed] [Google Scholar]
  • 17. Yang X, Liu W, Liu Y, Wang J, Lv L, Chen X, He D, Yang T, Hou J, Tan Y, Xing L, Zeng Z, Liu JH. 2014. F33: A-: B-, IncHI2/ST3, and IncI1/ST71 plasmids drive the dissemination of fosA3 and blaCTX-M-55/-14/-65 in Escherichia coli from chickens in China. Front Microbiol 5:688. doi: 10.3389/fmicb.2014.00688 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Jarocki VM, Reid CJ, Chapman TA, Djordjevic SP. 2019. Escherichia coli ST302: genomic analysis of virulence potential and antimicrobial resistance mediated by mobile genetic elements. Front Microbiol 10:3098. doi: 10.3389/fmicb.2019.03098 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19. Galetti R, Filho R, Ferreira JC, Varani AM, Sazinas P, Jelsbak L, Darini ALC. 2021. The plasmidome of multidrug-resistant emergent Salmonella serovars isolated from poultry. Infect Genet Evol 89:104716. doi: 10.1016/j.meegid.2021.104716 [DOI] [PubMed] [Google Scholar]
  • 20. Cowley LA, Dallman TJ, Fitzgerald S, Irvine N, Rooney PJ, McAteer SP, Day M, Perry NT, Bono JL, Jenkins C, Gally DL. 2016. Short-term evolution of Shiga toxin-producing Escherichia coli O157:H7 between two food-borne outbreaks. Microb Genom 2:e000084. doi: 10.1099/mgen.0.000084 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Hong YP, Wang YW, Huang IH, Liao YC, Kuo HC, Liu YY, Tu YH, Chen BH, Liao YS, Chiou CS. 2018. Genetic relationships among multidrug-resistant Salmonella enterica serovar Typhimurium strains from humans and animals. Antimicrob Agents Chemother 62:e00213-18. doi: 10.1128/AAC.00213-18 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Lin Y-T, Pan Y-J, Lin T-L, Fung C-P, Wang J-T. 2015. Transfer of CMY-2 cephalosporinase from Escherichia coli to virulent Klebsiella pneumoniae causing a recurrent liver abscess. Antimicrob Agents Chemother 59:5000–5002. doi: 10.1128/AAC.00492-15 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Guenther S, Ewers C, Wieler LH. 2011. Extended-spectrum beta-lactamases producing E. coli in wildlife, yet another form of environmental pollution? Front Microbiol 2:246. doi: 10.3389/fmicb.2011.00246 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Ewbank AC, Fuentes-Castillo D, Sacristán C, Esposito F, Fuga B, Cardoso B, Godoy SN, Zamana RR, Gattamorta MA, Catão-Dias JL, Lincopan N. 2022. World health organization critical priority Escherichia coli clone ST648 in magnificent frigatebird (Fregata magnificens) of an uninhabited insular environment. Front Microbiol 13:940600. doi: 10.3389/fmicb.2022.940600 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Haenni M, Dagot C, Chesneau O, Bibbal D, Labanowski J, Vialette M, Bouchard D, Martin-Laurent F, Calsat L, Nazaret S, Petit F, Pourcher A-M, Togola A, Bachelot M, Topp E, Hocquet D. 2022. Environmental contamination in a high-income country (France) by antibiotics, antibiotic-resistant bacteria, and antibiotic resistance genes: status and possible causes. Environ Int 159:107047. doi: 10.1016/j.envint.2021.107047 [DOI] [PubMed] [Google Scholar]
  • 26. WHO. 2021. WHO integrated global surveillance on ESBL-producing E. coli using a “One Health” approach: implementation and opportunities. World Health Organization. https://apps.who.int/iris/handle/10665/340079. [Google Scholar]
  • 27. Costa D, Poeta P, Sáenz Y, Vinué L, Rojo-Bezares B, Jouini A, Zarazaga M, Rodrigues J, Torres C. 2006. Detection of Escherichia coli harbouring extended-spectrum beta-lactamases of the CTX-M, TEM and SHV classes in faecal samples of wild animals in Portugal. J Antimicrob Chemother 58:1311–1312. doi: 10.1093/jac/dkl415 [DOI] [PubMed] [Google Scholar]
  • 28. Palmeira JD, Cunha MV, Carvalho J, Ferreira H, Fonseca C, Torres RT. 2021. Emergence and spread of cephalosporinases in wildlife: a review. Animals (Basel) 11:1765. doi: 10.3390/ani11061765 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Liu S, Jin D, Lan R, Wang Y, Meng Q, Dai H, Lu S, Hu S, Xu J. 2015. Escherichia marmotae sp. nov., isolated from faeces of Marmota himalayana. Int J Syst Evol Microbiol 65:2130–2134. doi: 10.1099/ijs.0.000228 [DOI] [PubMed] [Google Scholar]
  • 30. Clermont O, Gordon DM, Brisse S, Walk ST, Denamur E. 2011. Characterization of the cryptic Escherichia lineages: rapid identification and prevalence. Environ Microbiol 13:2468–2477. doi: 10.1111/j.1462-2920.2011.02519.x [DOI] [PubMed] [Google Scholar]
  • 31. De Witte C, Vereecke N, Theuns S, De Ruyck C, Vercammen F, Bouts T, Boyen F, Nauwynck H, Haesebrouck F. 2021. Presence of broad-spectrum beta-lactamase-producing Enterobacteriaceae in zoo mammals. Microorganisms 9:834. doi: 10.3390/microorganisms9040834 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Nüesch-Inderbinen M, Tresch S, Zurfluh K, Cernela N, Biggel M, Stephan R. 2022. Finding of extended-spectrum beta-lactamase (ESBL)-producing Enterobacterales in wild game meat originating from several European countries: predominance of Moellerella wisconsensis producing CTX-M-1, November 2021. Euro Surveill 27:2200343. doi: 10.2807/1560-7917.ES.2022.27.49.2200343 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Sivertsen A, Dyrhovden R, Tellevik MG, Bruvold TS, Nybakken E, Skutlaberg DH, Skarstein I, Kommedal Ø. 2022. Escherichia marmotae - a human pathogen easily misidentified as Escherichia coli. Microbiol Spectr 10:e0203521. doi: 10.1128/spectrum.02035-21 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Zurfluh K, Treier A, Schmitt K, Stephan R. 2020. Mobile fosfomycin resistance genes in Enterobacteriaceae-an increasing threat. MicrobiologyOpen 9:e1135. doi: 10.1002/mbo3.1135 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Freitag C, Michael GB, Li J, Kadlec K, Wang Y, Hassel M, Schwarz S. 2018. Occurrence and characterisation of ESBL-encoding plasmids among Escherichia coli isolates from fresh vegetables. Vet Microbiol 219:63–69. doi: 10.1016/j.vetmic.2018.03.028 [DOI] [PubMed] [Google Scholar]
  • 36. Ewbank AC, Fuentes-Castillo D, Sacristán C, Cardoso B, Esposito F, Fuga B, de Macedo EC, Lincopan N, Catão-Dias JL. 2022. Extended-spectrum β-lactamase (ESBL)-producing Escherichia coli survey in wild seabirds at a pristine atoll in the southern Atlantic ocean, Brazil: first report of the O25b-ST131 clone harboring blaCTX-M-8. Sci Total Environ 806:150539. doi: 10.1016/j.scitotenv.2021.150539 [DOI] [PubMed] [Google Scholar]
  • 37. Villa L, Guerra B, Schmoger S, Fischer J, Helmuth R, Zong Z, García-Fernández A, Carattoli A. 2015. IncA/C plasmid carrying bla NDM-1 , bla C MY-16 , and fosA3 in a Salmonella enterica serovar Corvallis strain isolated from a migratory wild bird in Germany. Antimicrob Agents Chemother 59:6597–6600. doi: 10.1128/AAC.00944-15 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38. Pietsch M, Irrgang A, Roschanski N, Brenner Michael G, Hamprecht A, Rieber H, Käsbohrer A, Schwarz S, Rösler U, Kreienbrock L, Pfeifer Y, Fuchs S, Werner G, RESET Study Group . 2018. Whole genome analyses of CMY-2-producing Escherichia coli isolates from humans, animals and food in Germany. BMC Genomics 19:601. doi: 10.1186/s12864-018-4976-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Hansen KH, Bortolaia V, Nielsen CA, Nielsen JB, Schønning K, Agersø Y, Guardabassi L. 2016. Host-specific patterns of genetic diversity among IncI1-Igamma and IncK plasmids encoding CMY-2 beta-lactamase in Escherichia coli isolates from humans, poultry meat, poultry, and dogs in Denmark. Appl Environ Microbiol 82:4705–4714. doi: 10.1128/AEM.00495-16 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40. Al-Mir H, Osman M, Drapeau A, Hamze M, Madec JY, Haenni M. 2021. Spread of ESC-, carbapenem- and colistin-resistant Escherichia coli clones and plasmids within and between food workers in Lebanon. J Antimicrob Chemother 76:3135–3143. doi: 10.1093/jac/dkab327 [DOI] [PubMed] [Google Scholar]
  • 41. Schwengers O, Jelonek L, Dieckmann MA, Beyvers S, Blom J, Goesmann A. 2021. Bakta: rapid and standardized annotation of bacterial genomes via alignment-free sequence identification. Microb Genom 7:000685. doi: 10.1099/mgen.0.000685 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Saidani M, Messadi L, Chaouechi A, Tabib I, Saras E, Soudani A, Daaloul-Jedidi M, Mamlouk A, Ben Chehida F, Chakroun C, Madec JY, Haenni M. 2019. High genetic diversity of Enterobacteriaceae clones and plasmids disseminating resistance to extended-spectrum cephalosporins and colistin in healthy chicken in Tunisia. Microb Drug Resist 25:1507–1513. doi: 10.1089/mdr.2019.0138 [DOI] [PubMed] [Google Scholar]
  • 43. Grant JR, Enns E, Marinier E, Mandal A, Herman EK, Chen C, Graham M, Van Domselaar G, Stothard P. 2023. Proksee: in-depth characterization and visualization of bacterial genomes. Nucleic Acids Res 51:W484–W492. doi: 10.1093/nar/gkad326 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Fig. S1. spectrum.02220-23-s0001.tif.

Map of the sampling.

Click here for additional data file. (2.2MB, tif)
DOI: 10.1128/spectrum.02220-23.SuF1
Table S1. spectrum.02220-23-s0002.xlsx.

List and characteristics of all E. coli and E. marmotae strains.

Click here for additional data file. (45.6KB, xlsx)
DOI: 10.1128/spectrum.02220-23.SuF2
Table S2. spectrum.02220-23-s0003.xlsx.

SNP distances between all E. coli and E. marmotae isolates.

Click here for additional data file. (31.5KB, xlsx)
DOI: 10.1128/spectrum.02220-23.SuF3
Table S3. spectrum.02220-23-s0004.xlsx.

Quality controls of all E. coli and E. marmotae data.

Click here for additional data file. (19.3KB, xlsx)
DOI: 10.1128/spectrum.02220-23.SuF4

Data Availability Statement

The project was deposited in DDBJ/EMBL/GenBank under the BioProject accession number PRJNA976065.

Articles from Microbiology Spectrum are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES