Presence of antimicrobial resistance in coliform bacteria from hatching broiler eggs with emphasis on ESBL/AmpC-producing bacteria

ABSTRACT

Antimicrobial resistance is recognized as one of the most important global health challenges. Broilers are an important reservoir of antimicrobial resistant bacteria in general and, more particularly, extended-spectrum β-lactamases (ESBL)/AmpC-producing Enterobacteriaceae. Since contamination of 1-day-old chicks is a potential risk factor for the introduction of antimicrobial resistant Enterobacteriaceae in the broiler production chain, the presence of antimicrobial resistant coliform bacteria in broiler hatching eggs was explored in the present study. Samples from 186 hatching eggs, collected from 11 broiler breeder farms, were inoculated on MacConkey agar with or without ceftiofur and investigated for the presence of antimicrobial resistant lactose-positive Enterobacteriaceae, particularly, ESBL/AmpC-producers. Escherichia coli and Enterobacter cloacae were obtained from the eggshells in 10 out of 11 (10/11) sampled farms. The majority of the isolates were recovered from crushed eggshells after external decontamination suggesting that these bacteria are concealed from the disinfectants in the egg shell pores. Antimicrobial resistance testing revealed that approximately 30% of the isolates showed resistance to ampicillin, tetracycline, trimethoprim and sulphonamides, while the majority of isolates were susceptible to amoxicillin-clavulanic acid, nitrofurantoin, aminoglycosides, florfenicol, neomycin and apramycin. Resistance to extended-spectrum cephalosporins was detected in eight Enterobacteriaceae isolates from five different broiler breeder farms. The ESBL phenotype was confirmed by the double disk synergy test and bla_SHV–12, bla_TEM–52 and bla_ACT–39 resistance genes were detected by PCR. This report is the first to present broiler hatching eggs as carriers and a potential source of ESBL/AmpC-producing Enterobacteriaceae for broiler chicks.

KEYWORDS:

• Broiler hatching eggs
• Enterobacteriaceae
• antibiotic
• ESBL
• AmpC
• disinfectant

Introduction

Antimicrobial resistance is recognized as one of the most important global health challenges. The introduction of extended-spectrum cephalosporins (ESCs) improved the treatment options in both veterinary and human medicine (Pfeifer et al., 2010). However, resistance to these antimicrobials, mainly conferred by extended-spectrum β-lactamases (ESBL) and AmpC β-lactamases, has recently gained importance and is considered as an emerging public health concern (Reich et al., 2013). ESBL- and AmpC-producing Enterobacteriaceae were first associated with hospitals and institutional care centres in humans, but are now increasingly present in the community (Livermore et al., 2007), in food-producing animals (Dierikx et al., 2013a), and retail meat (Kola et al., 2012), suggesting organism or gene exchange between different reservoirs (Leverstein-van Hall et al., 2011).

Several studies have reported that the gastrointestinal tract of healthy broilers (Smet et al., 2008) and layer hens (Chauvin et al., 2013) may be an important reservoir for antimicrobial-resistant bacteria in general and, more particularly, ESBL/AmpC-producers. These birds represent a potential risk for public health, particularly via broiler meat (Kola et al., 2012) or table eggs (Rasheed et al., 2014).

The spread of antimicrobial-resistant bacteria from broiler parent birds to broilers and subsequently to chicken meat has been recently shown (Dierikx et al., 2013b). A number of factors are incriminated in the introduction of antimicrobial-resistant bacteria to the broiler production chain (Hiroi et al., 2012; Schwaiger et al., 2013). Faeces, litter and even dust may act as transmission sources of ESBL/AmpC-producing bacteria (Laube et al., 2014). One-day-old chicks are also shown to be a major risk factor for the introduction of ESBL/AmpC-producing Escherichia coli in the broiler production chain (Dierikx et al., 2013b).

Breeder flocks and commercial hatcheries could be the first steps in the contamination of 1-day-old chicks by ESBL/AmpC-producing bacteria. However, to the best of our knowledge, no previously published data demonstrated the origin of this contamination. In this context, understanding the role of broiler hatching eggs in the carriage of resistant bacteria is of fundamental importance.

In the current experiments, we therefore investigated the occurrence of antimicrobial-resistant, more specifically, ESC-resistant, lactose-positive Enterobacteriaceae in broiler hatching eggs and characterized their susceptibility to specific antimicrobial agents.

Materials and methods

Egg sampling

The eggshell surface, the crushed decontaminated eggshell and the egg contents of 186 broiler hatching eggs that were not disinfected on-site, collected from 11 broiler breeder farms (nine from Belgium, one from the Netherlands, and one from France), were examined individually as described by Jones et al. (2002) and Chousalkar et al. (2010), with slight modifications.

In short, each intact egg was submerged in 50 ml of nutrient broth No. 2 (Oxoid, UK) for 10 min in a sterile container. After removing the egg and shaking the nutrient broth, 10 ml was transferred to a sterile 15 ml Falcon tube and incubated aerobically overnight at 37°C without shaking.

For the isolation of bacteria from eggshell pores, the eggs were first dipped in a 75% ethanol solution for 5 min to eliminate any bacteria present on the surface of the shell, and allowed to air dry in a biosafety cabinet for 5–10 min. Each egg was, then, cracked open with a sterile implement into a sterile container. The inside of the eggshells was washed with sterile phosphate-buffered saline to remove the adhering egg albumen. Egg shell and shell membranes were transferred to a 50 ml Falcon tube and crushed with a sterile laboratory tool. Twenty-five millilitres of nutrient broth No. 2 (Oxoid) was added to the 50 ml Falcon tube. The tubes were then incubated overnight at 37°C, without shaking.

The egg contents collected in the sterile containers were aseptically mixed and 3 ml of internal egg content was inoculated with 7 ml of nutrient broth No. 2 (Oxoid) and incubated overnight at 37°C, without shaking, for further processing.

Enterobacteriaceae isolation

After overnight incubation, 10 µl of each sample (egg surface, eggshell crushed after disinfection, and egg content), was plated out on MacConkey agar No. 3 (Oxoid). After overnight aerobic incubation at 37°C, from each sample, one lactose-positive colony was identified and purified for further identification and susceptibility testing.

In addition, 200 µl of each sample (egg surface, eggshell crushed after disinfection, and egg content), was plated out on MacConkey agar No. 3 (Oxoid) supplemented with 8 µg ml⁻¹ ceftiofur (CFTIO). After overnight aerobic incubation at 37°C, all presumed Enterobacteriaceae were purified for further identification, susceptibility testing and ESBL/AmpC characterization.

Bacterial identification

Strains were provisionally identified using standard biochemical methods (Markey et al., 2013). The specific tests used were: glucose/lactose fermentation, gas production and H₂S production on Kligler iron agar (Oxoid), aesculin hydrolysis (bile aesculin agar; Oxoid); motility, indole production and the ornithine degradation on motility-indol-ornithine media (Oxoid) (Callens et al., 2015). Final identification was obtained using matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS); a Bruker Daltonics Biotyper instrument on ethanol-washed cell pellets extracted with formic acid and acetontirile solutions was used as recommended by the manufacturer (Bruker Daltonics Inc., Billerica, MA, USA).

Isolates confirmed as being Escherichia coli were serotyped by agglutination in micro-titre plates with monospecific antisera towards 24 different somatic O antigens: O1, O2, O5, O6, O8, O9, O11, O12, O14, O15, O17, O18, O20, O35, O36, O45, O53, O78, O81, O83, O102, O103, O115, O116, purchased from the E. coli Reference Laboratory of the University of Santiago de Compostela (Lugo, Spain), according to the procedure of Blanco et al. (1996).

Antimicrobial susceptibility testing

Antimicrobial susceptibility testing was carried out on all obtained isolates by the Kirby Bauer disk diffusion test using the Neo-Sensitabs tablets (Rosco, Diagnostica, Taastrup, Denmark) with interpretation of inhibition zones according to the veterinary breakpoints, as described by the manufacturer. Briefly, after standardization of the inoculum in phosphate-buffered saline to 0.5 McFarland (Densimat; Biomérieux, Marcy l’Etoile, France), the inoculum was triple streaked (60° rotation of the round Petri dishes within streaks) on Mueller–Hinton II agar (Oxoid), and antimicrobial tablets were then brought onto the medium by means of a dispenser within 15 min after inoculation and were immediately aerobically incubated for 16–20 h at 35°C. E. coli ATCC® 25922 and E. coli ATCC^® 35218 were used as internal quality control strains.

The following antimicrobials were tested: ampicillin (AM, 10 µg), amoxicillin-clavulanic acid (AMC, 20/10 µg), CFTIO, 30 µg, florfenicol (FFC, 30 µg), flumequine (FLUM, 30 µg), apramycin (APRA, 40 µg), enrofloxacin (ENRO, 10 µg), tetracycline (TET, 30 µg), gentamicin (GEN, 10 µg), neomycin (NEO, 120 µg), spectinomycin (SPECT, 200 µg), nitrofurantoine (200 µg), sulphonamides (SULF, 240 µg) and trimethoprim (TR, 5 µg) (Neo-Sensitabs; Rosco Diagnostica, Taastrup, Denmark).

Isolates resistant to CFTIO (≤17 mm) were screened for ESBL production and AmpC production using the modified double disk synergy-tests (Jarlier et al., 1988), and for susceptibility to imipenem (IMI, 15 µg). For the double disk synergy-tests, disks containing CFTIO (30 µg) and cefquinome (30 µg) were used at a distance of 20 mm of a disk with AMC (20/10 µg). In addition, a cefoxitin disk (30 µg) was added to this panel to detect the AmpC phenotype. Isolates resistant to AMC (≤27 mm) and classified as intermediate or resistant to cefoxitin (≤17 mm) were suspected to be AmpC-producers (Dierikx et al., 2012).

Detection of β-lactamase genes

DNA was extracted from all phenotypically suspected ESBL/AmpC-producing Enterobacteriaceae by inoculating a single colony of each isolate from a blood agar plate into1 ml of nutrient broth. After overnight aerobic incubation at 37°C, cells were harvested by centrifugation at 875g for 5 min. The pellet was resuspended in 1 ml of HPLC water, and cells were lysed by heating at 95°C for 5 min. Cellular debris was removed by centrifugation at 15,000g for 5 min. PCRs for the detection of genes encoding TEM-, SHV-, CTX-M-, OXA-1, DHA-, ACC-, ACT-, MIR-, MOX-, FOX- and CMY-type enzymes were carried out as previously described (Pérez-Pérez & Hanson, 2002; Eckert et al., 2004; Dierikx et al., 2012).

Briefly, PCR was performed with a final volume of 25 µl. The primers used for amplification are listed in Table S1 (presented in supplementary data). The primer pair denoted as ACT-21F (5′-ATGATGAAAAAATCTCTTTGC-3′) and ACT-21R (5′-TTACTGTAGCGCGTCGAGGATA-3′) was designed from the alignment of the reference sequence for the bla ACT-21 of Enterobacter hormaechei obtained from GenBank database (http://www.ncbi.nlm.nih.gov) (Genbank access number: KF526118).

Each mixture contained 12.5 µl PCR master mix (BioMix™, Bioline), 1.5 µl of each primer (10 µM, Integrated DNA Technologies), 7.5 µl of HPLC water and 2 µl template DNA. Strains B4-25, B4-75, B5-33, B1-54, B4-16 containing genes encoding CTX-M-2, CTX-M-1, CTX-M-9, TEM-52, SHV-12/CMY-2, respectively, were used as positive controls (Smet et al., 2008). Subsequently, the PCR amplicons were sequenced with primers designed to flank the entire bla gene (Table S1, presented in Supplementary data). Finally, the obtained nucleotide sequences were compared with BLASTN to previously described bla genes stored in GenBank (http://www.ncbi.nlm.nih.gov/BLAST/).

Disinfectant susceptibility testing

Three agents commonly used for egg disinfection in hatcheries were selected for minimum inhibitory concentration (MIC) determination: formaldehyde (FOR, 39% wt/vol in H₂O), glutaraldehyde (GLU, 50% wt/vol in H₂O) and hydrogen peroxide (H₂O₂, 31.1% wt/vol in H₂O). FOR was obtained from VWR international BVBA/SPRL (Leuven, Belgium), GLU was obtained from Fluka AG (Chemische Fabrik, Ch-9470 Buchs) and the H₂O₂ from Merck Schuchardt OHG (Hohenbrunn, Germany).

The MICs for all obtained isolates were determined with the broth microdilution assay according to the Clinical and Laboratory Standards Institute (2013) standards, as previously described (Oosterik et al., 2014). Twofold dilutions of the disinfectants were added to test tubes with cation-adjusted Mueller–Hinton broth (Oxoid N.V.) after which 50 µl of the disinfectant dilution was transferred to the wells of a sterile plastic 96-well microdilution plate. Bacterial suspensions were adjusted to a concentration of 10⁶CFU/ml. Fifty microliters of these suspensions were added to the wells of the microdilution plate, yielding a total suspension volume of 100 µl. The plates were incubated aerobically for 24 h at 35 ± 2°C and MICs were read afterwards. The MIC was defined as the lowest concentration of disinfectant where no visible growth of the isolates could be seen. E. coli ATCC® 25922 was used as quality control strain.

Statistical analysis

Chi-square test was performed to compare the prevalence of the coliform isolates in the tested hatching broiler eggs and the percentages of the total obtained isolates from eggshell surface and crushed eggshell. Differences were considered significant at P < 0.05. All statistical analyses were performed with the R software (R.2.8.0).

Results

Coliform isolates, E. coli serotyping, and antimicrobial susceptibility

Out of 186 broiler hatching eggs collected from 11 breeder farms, 30 presumptive lactose-positive colonies were isolated from the non-supplemented MacConkey agar. E. coli was most frequently identified (n = 27) followed by Enterobacter cloacae (n = 3) (P < 0.05). The isolates were obtained from 16% (30/186) of the tested eggs: 12% (22/186) were isolated from the crushed decontaminated eggshell whereas only 4% (8/186) were isolated from the eggshell surface (P < 0.05). No lactose-positive Enterobacteriaceae were isolated from any of the egg contents (Table 1).

Table 1. Prevalence of lactose-positive Enterobacteriaceae isolates from broiler hatching eggs by source.

Source	No. of samples analysed	No. of E. coli isolates (%)	No. of E. cloacae isolates (%)	Total (%)
Shell crushed	186	19 (10.2)	3 (1.6)	22 (11.8)
Shell surface	186	8 (4.3)	0	8 (4.3)
Content	186	0	0	0
Total	186 eggs	27 (14.5)	3(1.6)	30 (16.1)

From 27 E. coli isolates, 18 could not be serotyped with the currently used antisera. From the typed isolates, the most prevalent serogroup was O2 (n = 5), followed by O8 (n = 3) and O1 (n = 1).

All coliform (30) isolates were tested for antimicrobial susceptibility. A summary of the results of disk diffusion tests is shown in Table 2. Resistance to FFC, NEO, SPECT and APRA was not observed. Resistance to CFTIO, AMC, FLUM, GEN, nitrofurantoin and ENRO was low and ranged between 3.3% and 13.3%. Resistance rates to TR, TET, AM and sulphonamides was moderate and ranged between 26.7% and 33.3%.

Table 2. Antimicrobial susceptibility percentage with numbers in parenthesis of lactose-positive Enterobacteriaceae isolates from broiler hatching eggs.

Antibiotic	Antimicrobial susceptibility % (N)
	S	I	R
CFTIO	93.3 (28)	0 (0)	6.7 (2)
AM^a	63.0 (17)	3.7 (1)	33.3 (9)
AMC^a	92.6 (25)	0 (0)	7.4 (2)
FFC	96.7 (29)	0 (1)	0 (0)
TET	66.7 (20)	3.3 (1)	30 (9)
FLUM	83.3 (25)	3.3 (1)	13.3 (4)
ENRO	83.3 (25)	3.3 (1)	13.3 (4)
GEN	10.0 (3)	86.7 (26)	3.3 (1)
NEO	76.7 (23)	23.3 (7)	0 (0)
SPECT	100.0 (30)	0 (0)	0 (0)
Nitrofurantoin	96.7 (29)	0 (0)	3.3 (1)
Sulphonamide	66.7 (20)	0 (0)	33.3 (10)
APRA	93.3 (28)	6.7 (2)	0 (0)
TR	73.3 (22)	0 (0)	26.7 (8)

^aE. cloacae was not included to calculate the percentage of susceptibility to this antibiotic.

ESC-resistant Enterobacteriaceae isolates

Eight (8) isolates resistant to ESC were obtained. Six isolates were obtained using the CFTIO supplemented medium, while two isolates were obtained using a non-supplemented MacConkey agar. The ESC resistant isolates were recovered from 5 of 11 analyzed farms and from 7/186 (3.8%) of the total tested eggs; one egg carried two isolates showing distinct antimicrobial resistance profiles. The ESC-resistant isolates were mainly obtained from the crushed decontaminated eggshell, while only one was isolated from the eggshell surface and none from the egg contents (Table 3). Out of the eight ESC-resistant isolates, seven isolates were identified as E. cloacae and one as Escherichia fergusonii. Six of eight CFTIO-resistant Enterobacteriaceae isolates (S47, S08R, S01, S02R, S20R, and S39R) were also resistant to the fourth generation cephalosporins (cefquinome) and some of the other tested non-β-lactam antimicrobial agents. No imipenem resistance was detected (Table 3).

Table 3. Genetic and phenotypic traits of CFTIO-resistant Enterobacteriaceae obtained from broiler hatching eggs.

Isolate	Farm no.	Egg no.	Source	Identification	ESBL/AmpC phenotype	β-Lactamase genes	Additional resistance phenotype
S12R	2	1	Shell crushed	E. cloacae	AmpC	ACT-39	TET, GEN^a, SUL
S47	2	7	Shell crushed	E. cloacae	ESBL	TEM-1, SHV-12	TET^a
S08R	3	3	Shell surface	E. cloacae	AmpC	Not detected	TET^a, APRA^a
S14R	5	4	Shell crushed	E. cloacae	AmpC	ACT-39	FFC, TET, NIT,
S01	6	2	Shell crushed	E. cloacae	ESBL/AmpC	TEM-1, SHV-12, ACT-39	TET, GEN^a, SUL, TR
S02R	6	2	Shell crushed	E. cloacae	ESBL/AmpC	TEM-1, SHV-12, ACT-39	TET, GEN^a, SUL
S20R	9	2	Shell crushed	E. fergusonii	ESBL	TEM-52	FFC^a, TET, GEN,NEO^a,SUL,TR
S39R	9	10	Shell crushed	E. cloacae	AmpC	ACT-39	/

Note: TET: tetracycline, GEN: gentamicin, SUL: sulphonamide, APRA: ampramycin, NIT: nitrofurantoine, NEO: neomycin, TR: trimethoprim, FFC: florfenicol.

^aStrain intermediately susceptible to this antibiotic.

The PCR and sequencing results are listed in Table 3. Three E. cloacae isolates (S47, S01 and S02R) carried a bla_TEM–1 and bla_SHV–12 gene while bla_TEM–52 was only detected in E. fergusonii isolate S20R. Five of the seven E. cloacae isolates (S12R, S14R, S01, S02R and S39R) harboured an ACT-39 enzyme (GenBank accession number: KU884289. Additionally, none of the screened β-lactamase genes could be found in S08R (Table 3).

Disinfectant MICs

The MICs determined by broth microdilution assay related to the 36 obtained isolates are presented in Figure 1.

Figure 1. (a) MIC of the recovered isolates from broiler hatching eggs determined for GLU per genus. (b) MIC of the recovered isolates from broiler hatching eggs determined for H₂O₂ per genus. (c) MIC of the recovered isolates from broiler hatching eggs determined for FOR per genus.

The ranges of MIC values of the retrieved isolates and quality control strains were similar to MIC ranges described earlier (Oosterik et al., 2014). Since no wild type cut-off values are yet available for disinfectants, acquired resistance was assumed when MIC values showed a bimodal, multimodal or tailing distribution. The distribution of the MIC values for GLU showed a unimodal distribution (Figure 1(a)). The distribution of the MIC values for H₂O₂ showed a bimodal distribution with the majority of isolates showing MIC values of 32–64 µg ml⁻¹ and three isolates showing a MIC value of 2 µg ml⁻¹ (Figure 1(b)). Considering previously published data (Oosterik et al., 2014), the isolates showing MIC values of 32–64 µg ml⁻¹ are probably wild type strains, while the isolates showing a MIC value of 2 µg ml⁻¹ are probably isolates with increased susceptibility. It was therefore concluded that there was no acquired resistance towards both GLU and H₂O₂. The MICs of FOR show a monomodal distribution, with possibly tailing towards the higher MIC values which might suggest low level acquired resistance in isolates with a MIC of 128 µg ml⁻¹ (Figure 1(c)).

Discussion

As previously published, a well-documented case report suggested that parental birds may represent an important reservoir of pathogenic E. coli strains and subsequent colibacillosis in newly hatched broiler chicks (Petersen et al., 2006). In the current screening of 186 eggs E. coli isolates belonged to serotypes O2, O8 and O1, known to be associated with avian colibacillosis (Schouler et al., 2012).

Using coliforms and ESBL-producing Enterobacteriaceae as indicator bacteria for antimicrobial resistance, strains were isolated from hatching eggs from almost all sampled broiler breeder farms (10/11). These results show that hatching eggs could be a potential source of transmission of antimicrobial-resistant bacteria or antimicrobial resistance genes in the broiler production chain. Indeed, it has been shown by Mevius et al. (2009) and Dierikx et al. (2013b) that 1-day-old chicks are often colonized by ESBL/AmpC-producing Enterobacteriaceae.

Antimicrobial resistance towards antimicrobial classes frequently used in veterinary medicine was tested on 30 lactose-positive Enterobacteriaceae isolates. Approximately one third of these isolates showed resistance to AM, TET, TR and sulphonamides. On the contrary, most isolates were susceptible to CFTIO, AMC, nitrofurantoin, the aminoglycosides, FFC, NEO and APRA. The presence of a small percentage of fluoroquinolone resistant Enterobacteriaceae isolates is noteworthy, but not entirely surprising considering the frequent detection of quinolone resistance in Enterobacteriaceae isolated from poultry (Vanni et al., 2014). SPECT was the only antimicrobial agent against which 100% susceptibility was detected. These results are in concert with data from several previous studies revealing that resistance to aminopenicillins, TETs and sulphonamides is common among E. coli strains isolated from poultry (Persoons et al., 2010) or commercial table eggs (Musgrove et al., 2006).

Resistance to ESCs, recovered in eight Enterobacteriaceae isolates, was found in five of the 11 analyzed farms. To the best of our knowledge, this report is the first to describe broiler hatching eggs as carriers and therefore as a potential source of ESBL-producing Enterobacteriaceae in broiler chicks. Previous research in laying hens (Schwaiger et al., 2008) or commercial table eggs (Musgrove et al., 2006) all reported E. coli isolates susceptible to ESCs. Molecular characterization of the ESC-resistant isolates showed that the resistance determinants were conferred by mobile elements and therefore could easily be horizontally transferred between strains, species or even genera (de Been et al., 2014). Half of these isolates harboured genes encoding ESBLs (SHV-12 and TEM-52) and most of E. cloacae isolates (5/7) harboured bla_ACT–39, a new variant of the ACT AmpC β-lactamase. SHV-12 and TEM-52 are among the most frequently detected ESBLs in Enterobacteriaceae responsible for infections in humans (Coque et al., 2008).

A relevant finding is the fact that the Enterobacteriaceae were mainly isolated from crushed eggshells even after extensive external decontamination. This could be related to the particular constitution of the eggshell itself, making the bacteria unreachable by the disinfection agents in the egg shell pores. The obtained Enterobacteriaceae isolates were probably concealed in the pores of the egg shells from the disinfection step. This is in agreement with previously reported results showing that the shell and membrane crush method recovers microorganisms from the pores and membranes that are missed by swabbing or rinsing methods (Musgrove et al., 2005). The bacterial contamination of the egg could occur while passing through the cloaca, or immediately after lay from the breeding and hatching environment, especially during the first 30–60 s, when the eggs are most vulnerable to bacterial penetration before the cuticle hardens and effectively caps the pores (Montgomery et al., 1999).

It is also possible that acquired resistance against disinfecting agents could lead to persistence of bacteria in or on egg shells after decontamination. However, in the current investigations, none of the investigated isolates showed clear features of acquired resistance to the three commonly used disinfectants in hatcheries. In addition, commercial disinfectants are often a mixture of active compounds, thereby hampering the selection of acquired resistance (Jones et al., 2000).

The phenomenon of bacterial concealment in the shell pores may be of importance in the field; (antimicrobial-resistant) bacteria that survive this process may colonize 1-day-old chicks and spread throughout the hatchery.

In conclusion, the present results revealed that broiler hatching eggs can be carriers of broad-spectrum β-lactamase-producing Enterobacteriaceae. The origin of this contamination remains speculative and further studies are warranted. The phenomenon of bacteria concealment revealed in the current results increases the hatching eggs as a potential factor in antibiotic resistance dissemination.

Acknowledgements

The authors would like to thank Dr P. Wattiau, Coda-Cerva Brussels, Belgium for the identification of all isolates by MALDI-TOF MS and the serotyping of the E. coli isolates.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

F. Haesebrouck http://orcid.org/0000-0002-1709-933X

Additional information

Funding

The authors thank the Belgian Federal Public Service of Health, Food Chain Safety and Environment for the funding of Ilias Chantziaras [RT-09/05].