Comparison of Antimicrobial Resistance of Thermophilic Campylobacter Isolates from Conventional and Organic Turkey Meat in Germany

Abstract

The objective of this study was to compare the prevalence and antimicrobial resistance (AMR) rates of Campylobacter spp. isolated from conventional and organic turkey meat sold at retail in Germany. Samples of conventional (N = 527) and organic (N = 245) fresh turkey meat without skin were collected at retail markets throughout Germany and tested for Campylobacter spp.. Campylobacter isolates were tested for resistance to six antimicrobials (gentamicin, streptomycin, ciprofloxacin, nalidixic acid, erythromycin, and tetracycline) using broth microdilution. Prevalence of Campylobacter spp. was higher in organic (32.7%) than in conventional (19.4%) turkey meat. The proportion of fully susceptible isolates was lower in Campylobacter coli (6.8%) than in Campylobacter jejuni (33.9%) and higher in isolates from organic (38.4%) than from conventional production (17.4%). Overall, resistance rates were the highest to ciprofloxacin, nalidixic acid, and tetracycline. Resistance to erythromycin was only observed in C. coli and resistance to gentamicin was absent. Overall, resistance rates to tetracycline and fluoroquinolones were higher in isolates from conventional (60.9% and 78.9%) than from organic meat (32.9% and 58.9%, respectively). However, this significant difference was only observed for C. jejuni, but not for C. coli. Further studies are needed to identify the reasons for the differences in the association of production type of turkeys with AMR in the different Campylobacter spp. and the critical parameters for the reduction of AMR in Campylobacter from turkey meat.

Introduction

Antimicrobial resistance (AMR) is one of the major challenges to public and global health. The contribution of antimicrobial use in farm animals to AMR issues on the medical side is constantly under debate. Campylobacteriosis is the most frequent foodborne bacterial zoonotic infection observed in humans in Germany and Europe (EFSA and ECDC, 2018b).

Recently, the European Food Safety Authority (EFSA), the European Centre for Disease Prevention and Control (ECDC), and the European Medicines Agency (EMA) have pointed out that resistance to fluoroquinolones in human Campylobacter infections is associated with antimicrobial use in animals in the country (ECDC et al., 2017). The likely reason is that Campylobacter infections in humans are mainly attributable to contaminated meat products and a lack of adequate kitchen hygiene. Human beings are only transient hosts to the zoonotic pathogen. Therefore, potential person-to-person spread of the bacteria is limited to these short periods. Selection pressure owing to antimicrobial use in humans can also only be effective while the bacteria colonize the intestines of humans.

Using source attribution models, human campylobacteriosis has repeatedly been attributed predominantly to Campylobacter from poultry (Kittl et al., 2013; Rosner et al., 2017). The two main Campylobacter spp. associated with human campylobacteriosis in Germany and other countries are Campylobacter jejuni and Campylobacter coli, with a clear dominance of C. jejuni (EFSA and ECDC, 2018b; RKI, 2019). Of the two, C. coli has been reported to be more frequently resistant to antimicrobials than C. jejuni (EFSA and ECDC, 2019).

In Europe, organic production is regulated by Council Regulation (EC) No. 834/2007 (European Council, 2007). According to this regulation, antimicrobial use in organically raised poultry is restricted to necessary cases and under strict conditions. Although separate data on antimicrobial use in organic poultry farms is not available for Germany, it is assumed that antimicrobial use in organic poultry is lower than in conventional poultry. In line with that, previous studies observed lower resistance of Campylobacter spp. to tetracycline and fluoroquinolones in turkeys from organic farms than in turkeys from conventional farms (Luangtongkum et al., 2006; El-Adawy et al., 2015). The same has been reported for Campylobacter spp. from conventional and organic pigs in France, but not in Sweden (Kempf et al., 2017).

In this study, we investigated whether the difference in AMR between Campylobacter spp. from conventional and organic turkeys is also reflected in isolates from turkey meat. This had been under debate as in poultry slaughterhouses cross-contamination is a major issue and therefore the slaughter process might be prone to diminish the differences observed in turkeys on farms (Normand et al., 2008; Pacholewicz et al., 2015). Therefore the objective of this study was to test the two hypotheses, (1) whether AMR is lower in Campylobacter isolates from organic than from conventional turkey meat and (2) whether AMR is higher in C. coli than in C. jejuni in both production sectors.

Materials and Methods

Study design

The study was carried out over the whole year in 2018 in the framework of a national monitoring program set up by the authorities of the Federal States, the Federal Office for Consumer Protection and Food Safety (BVL) and the German Federal Institute for Risk Assessment (BfR) in Berlin, Germany. The structure of the monitoring is regulated by the General Administrative Regulation on the recording, analysis, and publication of data on the occurrence of zoonoses and zoonotic agents along the food chain (BMEL, 2012).

The number of samples to be collected was distributed across the country based on the human population in the respective federal states. The sample size was calculated to allow for an accuracy of ±5% at an assumed prevalence of 50% leading to a target of 384 samples per population because the framework of the monitoring sample are examined for various bacterial species with diverse expected prevalences. However, as there were concerns regarding the availability of organic turkey meat at the retail outlets it was expected that for the organic meat the full sample size would not be achieved. Only one sample per production batch was to be collected per sampling point. However, because of decentralized sampling and primary isolation it cannot be excluded that samples from different retail points contained the same production badge. The primary purpose of the monitoring was the determination of the exposure of consumers in Germany to zoonotic bacteria in the food chain, namely Campylobacter spp. Samples were collected by the veterinary officials or their staff in retail markets all over Germany and transported under cooled conditions (2–8°C) to the laboratory within 24 h.

Microbiological analysis and susceptibility testing

Samples were tested in the regional laboratories of the federal states according to ISO 10272-1:2017, procedure B and 25 g of meat were used as test portion (ISO, 2017). In brief, the meat samples were 1:10 (w/v) diluted in Preston broth and incubated for 24 ± 2 h at 41.5°C ± 1°C under microaerobic atmosphere. After this enrichment step, a loop of ∼10 μL was spread on modified charcoal cefoperazone deoxycholate agar and incubation was performed for another 44 ± 4 h under the same conditions. At least one typical or suspect Campylobacter colony was confirmed using either phenotypic tests as described in ISO 10272-1:2017 or matrix-assisted laser desorption/ionization time-of flight or PCR methods, which were in-house validated in the respective regional laboratories. According to ISO 10272-1:2017, if the first colony was negative, up to four more suspect colonies were checked unless the sample was considered negative for thermophilic Campylobacter spp.

Isolates of Campylobacter spp. were sent to the National Reference Laboratory (NRL) for Campylobacter at the BfR in Berlin, Germany, where Campylobacter spp. were species-differentiated by real-time PCR (Mayr et al., 2010; BVL, 2013) and isolates were tested for AMR according to the prescriptions given in Commission Implementing Decision (CID) 2013/652/EU (European Commission, 2013). For this purpose, strains were subcultured on Columbia blood agar for 24 ± 2 h at 42°C under microaerobic atmosphere (5% O₂, 10% CO₂, rest N₂). Broth microdilution susceptibility testing was performed according to M45-A (Clinical and Laboratory Standards Institute [CLSI], 2015) and VET06 (CLSI, 2017) with the in-house validated modification of use of fetal calf serum instead of lysed horse blood in the culture medium for improved readability of Campylobacter growth. Cation-supplemented Mueller–Hinton broth (TREK Diagnostic Systems, United Kingdom) supplemented with 5% fetal calf serum (PAN-Biotech, Germany) was inoculated with 2–8 × 10⁵ colony-forming units/mL using bacteria grown on Columbia blood agar. Minimum inhibitory concentrations (MICs) were determined using the European standardized microtiter plate format EUCAMP2 (TREK Diagnostic Systems).

Antimicrobials tested included nalidixic acid, ciprofloxacin, gentamicin, tetracycline, erythromycin, and streptomycin. Incubation was performed for 44 ± 4 h at 37°C under microaerobic atmosphere. MICs (mg/L) were semi-automatically analyzed using the Sensititre Vizion system and the SWIN-Software (TREK Diagnostic Systems). In principle, the Sensititre Vizion apparatus has an integrated camera and a mirror, reflecting and recording a translucent picture from the microtiter plates. The resulting photographs can be visually read and the MIC data are automatically stored and can be exported using the Thermo Scientific SWIN Software. MIC were evaluated according to the epidemiological cutoff values (ECOFF) published by the European Committee for Antimicrobial Susceptibility Testing (EUCAST; https://mic.eucast.org/Eucast2) and laid down in the CID 2013/652/EU.

Statistical analysis

Sample metadata and results of the primary isolation of the bacteria in food were submitted to a national database run by the BVL. Data on AMR were stored in the Laboratory information management system of the BfR. Both data were combined using MS Excel, based on the sample identifiers. Isolates from turkey meat arriving at the BfR that had no corresponding data in the BVL database were excluded from the analysis in this study. However, if isolates from positive samples did not arrive at the BfR or could not be subcultured upon arrival, data were not excluded from the prevalence estimate to avoid bias that might be caused by the selective exclusion of positive samples. For negative samples metadata on the samples were exclusively derived from the BVL database.

Prevalence of Campylobacter spp. in the samples was calculated as the percentage of positive samples among the tested samples. If both species, C. jejuni and C. coli, were identified in the same sample, this sample was only considered as one positive sample.

All isolates were tested against six antimicrobials. The percentage of resistant isolates was calculated as the percentage of isolates with an MIC for the respective antimicrobial above the ECOFF. The proportion of isolates that was susceptible to all antimicrobials was calculated as an indicator for the degree of AMR beyond the test range of the individual antimicrobial. This has recently been recommended by the European Agencies (Cormican et al., 2017). The proportion of all tests with the outcome “resistant” was also calculated as an additional indicator of the degree of AMR in the isolates. The prevalence of resistant isolates was calculated as the proportion of the tested samples that harbored resistant Campylobacter spp. The proportion of positive samples for Campylobacter spp. was compared using chi-square test. The comparison of AMR between the two bacterial species was carried out by logistic regression analyses using resistance to a specific antimicrobial as the binary outcome and the bacterial species (C. coli vs. C. jejuni) and the source of the meat (conventional vs. organic) as fixed factors. Federal state was first included in the analyses but it proved to be a nonsignificant factor in all analyses and was, therefore, removed from the models. We did separate logistic regression analyses for the individual antimicrobials. The results showed significant differences between the species for some antimicrobials. Therefore, separate logistic regression models for the bacterial species were calculated additionally, to see if the association of source of meat with AMR could be confirmed for both species. The analysis was carried out using SPSS (IBM SPSS Statistics, Vs. 21). Level of significance was set at alpha = 0.05.

Results

In total 527 samples of conventional and 245 samples of organic turkey meat were tested for the presence of Campylobacter spp. The proportion of positive samples was significantly higher in organic (80/245, 32.7%) than in conventional meat (102/527, 19.4%, p < 0.01). A total of 171 isolates was sent to the NRL, was subcultured, species-differentiated, and tested for AMR. Six isolates (five organic, one conventional) were identified as Campylobacter lari and not further analyzed. Of the remaining 165 isolates, 92 were from samples of conventional meat and 73 from samples of organic meat (Table 1).

Table 1.

Antimicrobial Resistance of Campylobacter jejuni and Campylobacter coli from Conventional and Organic Turkey Meat in Germany in 2018

Bacterial species	Campylobacter jejuni				Campylobacter coli				Campylobacter spp.^a
Production type of meat	Conventional		Organic		Conventional		Organic		Conventional		Organic
	N	%	N	%	N	%	N	%	N	%	N	%
No. of isolates	70		51		22		22		92		73
No. of tests^b	420		306		132		132		552		438
Resistant results^c	142	33.8	58	19.0	61	46.2	55	41.7	203	36.8	113	25.8
Gentamicin	0	0.0	0	0.0	0	0.0	0	0.0	0	0.0	0	0.0
Streptomycin	2	2.9	2	3.9	4	18.2	1	4.5	6	6.5	3	4.1
Ciprofloxacin	54	77.1	24	47.1	18	81.8	19	86.4	72	78.3	43	58.9
Nalidixic acid	48	68.6	23	45.1	18	81.8	19	86.4	66	71.7	42	57.5
Erythromycin	0	0.0	0	0.0	3	13.6	1	4.5	3	3.3	1	1.4
Tetracycline	38	54.3	9	17.6	18	81.8	15	68.2	56	60.9	24	32.9
Fully susceptible^d	14	20.0	27	52.9	2	9.1	1	4.5	16	17.4	28	38.4

Sum of Campylobacter jejuni and Campylobacter coli.

No. of isolates × no. of antimicrobials tested.

No. and percent of test results that indicated resistance.

Susceptible to all six tested antimicrobials.

Overall, isolates of C. coli were less frequently fully susceptible (3/44, 6.8%) than isolates of C. jejuni (41/121, 33.9%, p < 0.001). Moreover, isolates from organic meat were less frequently resistant to at least one of the tested substances than those from conventionally produced turkey meat (p < 0.01). In line with that, more test results indicated resistance across all antimicrobials in isolates from conventional meat (36.8%) than in isolates from organic meat (25.8%). This was specifically true for the antimicrobials tetracycline, ciprofloxacin, and nalidixic acid. However, resistance to ciprofloxacin was similar in isolates of C. coli from organic and conventional meat (86.4% vs. 81.8%). In C. jejuni resistance to ciprofloxacin, nalidixic acid, and tetracycline was significantly less frequent in isolates from organic meat (Table 1). Resistance to erythromycin was only observed in C. coli from both production types. There was no significant difference between the resistance rates to erythromycin and streptomycin in C. coli from organic meat (4.8% and 4.5%, respectively) and from conventional meat (13.6% and 18.2%, respectively). None of the isolates showed resistance to gentamicin.

Table 2 shows the results of the logistic regressions. The probability of full susceptibility was significantly higher for C. jejuni than for C. coli (odds ratio [OR] = 8.7) and for isolates from organic than from conventional meat (OR = 3.6). Gentamicin was not included in the logistic regression models for the individual substances as no isolate was resistant to gentamicin. For erythromycin this logistic regression could also not be validly calculated, as there was no resistant C. jejuni isolate. For the other four antimicrobials, C. coli were significantly more likely to be resistant to antimicrobials with ORs between 3.3 for ciprofloxacin and 6.5 for tetracycline. Isolates from conventional meat were more likely to be resistant to tetracycline (OR = 4.3), nalidixic acid (OR = 2.1), and ciprofloxacin (OR = 2.8), but not to streptomycin.

Table 2.

Results of Logistic Regression on the Risk of Resistance in Campylobacter spp. from Conventional and Organic Turkey Meat in Germany 2018

		Coefficient	Standard error	Wald	Degrees of freedom	p-value	OR	95% CI for OR
		Coefficient	Standard error	Wald	Degrees of freedom	p-value	OR	Lower	Upper
Susceptible^a	C. c./C. j.^b	2.158	0.644	11.228	1	<0.001	8.65	2.45	30.56
	Conv/org^c	1.288	0.387	11.076	1	<0.001	3.63	1.70	7.74
	Constant	−0.015	0.275	0.003	1	0.96	0.99
Streptomycin	C. c./C. j.	1.374	0.701	3.842	1	0.050	3.95	1.00	15.61
	Conv/org	0.604	0.738	0.670	1	0.413	1.83	0.43	7.76
	Constant	−3.765	0.729	26.673	1	<0.001	0.023
Tetracycline	C. c./C. j.	1.873	0.436	18.439	1	<0.001	6.51	2.77	15.30
	Conv/org	1.470	0.372	15.587	1	<0.001	4.35	2.10	9.02
	Constant	−1.373	0.321	18.307	1	<0.001	0.25
Nalidixic acid	C. c./C. j.	1.410	0.460	9.381	1	0.002	4.10	1.66	10.10
	Conv/org	0.750	0.346	4.713	1	0.030	2.12	1.08	4.17
	Constant	−0.073	0.267	0.074	1	0.786	0.93
Ciprofloxacin	C. c./C. j.	1.205	0.467	6.653	1	0.010	3.34	1.34	8.33
	Conv/org	1.031	0.359	8.260	1	0.004	2.80	1.39	5.66
	Constant	0.035	0.268	0.017	1	0.898	1.04

Susceptible = 0, ≥1 resistance = 1.

C. j. = C. jejuni = 0, C. c. = C. coli = 1.

Org = Organic meat = 0, Conv = Conventional meat = 1.

CI, confidence interval; OR, odds ratio.

Separate logistic regression for the two Campylobacter species confirmed the differences between conventional and organic meat for C. jejuni. However, for C. coli there were no significant differences in AMR between isolates from conventional and organic meat.

Table 3 shows the distribution of MIC in isolates from conventional and organic turkey meat. For every drug/bacteria combination, there is a minimum of two dilution steps between the highest wild-type MIC and the lowest non-wild-type (resistant) MIC.

Table 3.

Minimum Inhibitory Concentrations of Six Antimicrobials for Campylobacter jejuni and Campylobacter coli from Organic and Conventional Turkey Meat

Antimicrobial	Population (N)	MIC (mg/L)
Antimicrobial	Population (N)	0.13	0.25	0.5	1	2	4	8	16	32	64	128	256
Ciprofloxacin	Campylobacter coli organic (22)	3	0	0	0	0	0	6	10	3
	C. coli conventional (22)	4	0	0	0	0	0	6	7	5
	Campylobacter jejuni organic (51)	26	1	0	0	0	0	11	10	3
	C. jejuni conventional (70)	16	0	0	0	0	0	25	25	4
Erythromycin	C. coli organic				19	2	0	0	0	0	0	0	1
Erythromycin	C. coli conventional				15	3	0	1	0	0	1	0	2
	C. jejuni organic				51	0	0	0	0	0	0	0	0
	C. jejuni conventional				69	1	0	0	0	0	0	0	0
Gentamicin	C. coli organic	0	0	20	2	0	0	0	0	0
Gentamicin	C. coli conventional	0	1	14	7	0	0	0	0	0
	C. jejuni organic	0	7	43	1	0	0	0	0	0
	C. jejuni conventional	0	21	46	3	0	0	0	0	0
Nalidixic acid	C. coli organic				0	0	3	0	0	0	11	8
Nalidixic acid	C. coli conventional				0	0	3	1	0	0	11	7
	C. jejuni organic				1	2	22	3	0	0	3	20
	C. jejuni conventional				2	3	16	1	0	0	4	44
Streptomycin	C. coli organic		0	0	7	14	0	0	0	1
Streptomycin	C. coli conventional		0	0	5	11	2	0	0	4
	C. jejuni organic		0	6	27	16	0	0	0	2
	C. jejuni conventional		0	12	41	15	0	0	0	2
Tetracycline	C. coli organic			6	0	1	0	0	0	0	2	13
Tetracycline	C. coli conventional			4	0	0	0	0	0	0	3	15
	C. jejuni organic			42	0	0	0	0	1	0	4	4
	C. jejuni conventional			31	1	0	0	3	1	3	12	19

Number of isolates not inhibited by the highest test concentration are at the far right of each line in italics. Vertical lines within the frame indicate epidemiological cutoff values as fixed in Commission Implementing Decision 2013/652/EU.

MIC, minimum inhibitory concentration.

Combining prevalence of Campylobacter spp. in the samples and the proportion of resistant isolates among the tested isolates, the likelihood of an isolate that was resistant to at least one antimicrobial in a meat sample was numerically higher in organic (45/245 samples, 18.4%) than in conventional turkey meat (76/527 samples 14.4%) (OR = 1.3, 95% confidence interval = 0.9–2.0), but the difference was not significant.

Discussion

To our knowledge, this is the first report on the comparison of AMR in thermophilic Campylobacter spp. from conventional and organic turkey meat. The results of our study confirm that bacteria on meat from organic farming are less resistant than those from meat from conventional turkey production for thermophilic Campylobacter spp.. However, the significant differences were observed in C. jejuni but not in C. coli. Our results also highlight that Campylobacter was more frequently present in organic than in conventional turkey meat. This is in contrast to a previous study in the United States that did not find any Campylobacter spp. in organic turkey meat. However, the number of samples in that study was very limited (Noormohamed and Fakhr, 2014). The differences in prevalence of Campylobacter spp. in the meat could either be associated with differences in the colonization rate of the animals at slaughter or with differences in slaughter hygiene. However, information on the slaughterhouses of origin and the origin or colonization status of the birds slaughtered were not collected for the meat samples investigated.

In the monitoring according to CID 2013/652/EU, C. coli was more prevalent than C. jejuni in cecal content samples from turkeys in Germany in 2014 (55.1% of all isolates) and 2018 (60.7%) (BVL, 2016, 2019), which was in line with the ratio found in organic turkeys (El-Adawy et al., 2015). In 2016, it was 45.7% (BVL, 2017). In contrast, in our study, C. jejuni was far more prevalent on meat in both populations (76.1% and 70.9%, respectively), which is in line with the results on meat samples from previous years (BVL, 2016, 2017).

The absence of resistance to gentamicin reflects the overall resistance situation in Campylobacter isolates from food and animals observed in recent years (BVL, 2017, 2018; El-Adawy et al., 2015; Varga et al., 2019; EFSA and ECDC, 2020). However, higher resistance rates to gentamicin were observed in poultry meat in the United States (Noormohamed and Fakhr, 2014).

The differences in AMR rates observed between C. coli and C. jejuni support previous reports on AMR in Campylobacter spp. from turkeys, other animals, and food sources that did not consider production type (Kashoma et al., 2014; BVL, 2017; EFSA and ECDC, 2018a; Varga et al., 2019). This indicates that the bacterial species also contributes to the AMR differences and a general analysis of AMR in Campylobacter spp. neglecting the individual species is of limited use. The reasons for the differences in AMR between the species are not known.

The proportion of resistant isolates of C. jejuni was significantly higher in the conventional isolates, whereas differences were not significant for C. coli. This was reflected in the proportion of isolates that were susceptible to all antimicrobials and likewise in the proportion of all test results across all antimicrobials that were “resistant”. The latter was independent of the inclusion of both (fluoro-) quinolones as individual tests or a combined evaluation of the two substances (data not shown). The difference between C. jejuni and C. coli with respect to the association of production type with AMR was most obvious for the resistance to ciprofloxacin and nalidixic acid. The analysis of this association is hampered by the limited number of C. coli isolates in both origins.

A difference in AMR of Campylobacter spp. from conventional and organic turkeys has previously been postulated (Luangtongkum et al., 2006; El-Adawy et al., 2015). El Adawy et al. (2015) tested isolates from organically raised turkeys in Germany and compared the result with those of the national monitoring program that was part of the same framework used in our study (El-Adawy et al., 2015). Compared with our study, they found similar resistance rates for C. jejuni to ciprofloxacin and nalidixic acid (43.9% vs. 47.1% in our study) in organic production, but higher resistance rates for tetracycline (48.7% vs. 17.6% in our study). For C. coli, resistance rates in their study were lower than in our study with resistance to ciprofloxacin at 72.7% to nalidixic acid at 74.5% and to tetracycline at 48.2% versus 90.5, 90.5, and 66.7% in our study, respectively. Resistance to the other antimicrobials was similar. The differences can have a number of reasons. El Adawy et al. (2015) used the plate format EUCAMP that was used in the European Union until, 2013, whereas we used the current format EUCAMP2. However, the two plate formats do not differ much. Test conditions were similar and cutoffs identical. Moreover, wild-type and non-wild-type isolates were separated by at least two dilution steps for each antimicrobial (Table 3). Therefore, major changes in the method might have been required to change the classification of isolates. El Adawy et al. (2015) pointed out that there were substantial differences between farms. Farm effects can neither be excluded nor analyzed with our data as meat samples cannot be traced back to the farms of origin.

Moreover, between 2012 and 2018 there was a substantial reduction of antimicrobial use in the German livestock sector that probably had an effect on the overall level of AMR (Wallmann et al., 2019). The older report from the United States by Luangtongkum et al. (2006) found even greater differences between isolates from organic and conventional farms, but in their study use of antimicrobials was completely banned in the organic farms, which is different from the European regulations.

Because of a low overall number of isolates that were resistant to gentamicin, streptomycin, and erythromycin, differences in AMR between the two meat origins could not be expected. However, it is surprising that the differences in resistance to fluoroquinolones that were observed in C. jejuni were not observed in C. coli. In line with that, no significant difference was observed in resistance to fluoroquinolones in C. coli from conventional and organic pigs in Sweden and France (Kempf et al., 2017). Thus, it might be suitable to investigate fitness enhancement associated with fluoroquinolone resistance in C. coli as it was previously studied in C. jejuni. If fluoroquinolone resistance is associated with a fitness advantage also in the absence of antimicrobials, a reduction of fluoroquinolone use cannot be expected to result in a reduction of AMR to fluoroquinolones. However, fitness issues in C. coli and C. jejuni related to fluoroquinolone resistance have been shown to depend on a number of factors (Zeitouni and Kempf, 2011).

Bacteria that can be found on meat at retail mostly originate from cross-contamination of bacteria harbored by the animals and spread to the carcass during the slaughter process (Normand et al., 2008, Vossenkuhl et al., 2014). Therefore, it was expected that isolates from turkey meat, similar to those from turkeys on the farm, would reflect differences in antimicrobial use during the life of the animals.

Although the metadata to samples in a national monitoring program is limited, the results are in line with the assumption of lower antimicrobial use on organic farms. Likewise, other aspects of organic farming such as reduced stocking density or differences in the provided feed may have contributed to the observed reduced antibiotic resistance of the Campylobacter isolates (Murphy et al., 2018). Factors contributing to the differences in AMR beyond differences in antimicrobial use and factors influencing the need for antimicrobial use during fattening remain to be investigated in further studies (Murphy et al., 2018).

On the contrary, resistance of C. coli from organic turkey meat was still high. We do not know where the turkey chicks raised on the organic farms came from. It is possible that they were bought from the same hatchery and therefore had a similar primary bacterial colonization population. To investigate this in longitudinal on farm studies would be required.

Whereas resistance to tetracycline was 4.3 times more likely in isolates from meat of conventional turkeys as compared with isolates from organic turkey meat, the OR was only 2.1–2.8 for nalidixic acid and ciprofloxacin, respectively (Table 2).

In Germany, fluoroquinolones are licensed for the use in poultry. A recent government report has highlighted that the use of these drugs is higher in turkeys than in broilers or in cattle and pigs. Next to penicillins and polymyxins they are the most frequently used drugs in turkeys (Flor et al., 2019). Currently it is not known, whether that also holds true for organic farming. In 2016, resistance of C. jejuni from turkeys to ciprofloxacin was very high (≥75%) in all European Member States reporting data to EFSA (EFSA and ECDC, 2018a) with lower values only reported for the United Kingdom (37.4%). A recent Canadian study found increasing rates of resistance to fluoroquinolones in Campylobacter from turkeys despite low reported use of these drugs in Canada (Agunos et al., 2019). It was suggested that fluoroquinolone resistance in C. jejuni was linked to certain sequence types, which are disseminated among countries (Kovac et al., 2015). Most other resistances exert a fitness cost on the bacterium.

However, the underlying mechanism of fluoroquinolone resistance in Campylobacter, a point mutation in the subunit A of the essential enzyme gyrase, can convey a fitness benefit, depending on the genetic background of the strains (Luo et al., 2005). Thus, use of fluoroquinolones at farm might exert a long-term effect on circulating fluoroquinolone-resistant Campylobacter. As a consequence, reduction of the use of this class of antimicrobials may not be as effective in reducing resistance, which might be reflected in only smaller differences in resistant rates to ciprofloxacin than to tetracycline between isolates from conventional or organic meat. Tetracyclines, on the contrary, are less frequently used in German turkey production than fluoroquinolones and AMR to tetracycline in turkeys has been decreasing recently (Flor et al., 2019).

Conclusion

The results confirm that C. coli is more likely to be resistant to antimicrobials than C. jejuni. Overall, AMR was lower in isolates from organic than from conventional turkey meat primarily because of difference in antibiotic resistance found in C. jejuni. Differences were largest for tetracycline but also observed for ciprofloxacin and nalidixic acid. On the contrary, Campylobacter spp. were more frequently isolated from organic than from conventional turkey meat. This might either be caused by more frequent colonization of organic animals or by differences in slaughter hygiene. Further studies are needed to elucidate in more detail the reasons for the differences in contamination rates of meat and resistance of isolates between organic and conventional turkey meat.

Footnotes

Acknowledgments

The authors gratefully acknowledge the support of the authorities of the federal states and the regional laboratories for providing the data and the isolates. The excellent technical support of the staff of the National Reference Laboratory for Campylobacter is also gratefully acknowledged.

Disclosure Statement

No competing financial interests exist.

Funding Information

The work was carried out as part of duties of the respective institutions and funded by their regular budget.

References

Agunos

, Gow

, Leger

, Carson

, Deckert

, Bosman

, Loest

, Irwin

, Reid-Smith

. Antimicrobial use and antimicrobial resistance indicators-integration of farm-level surveillance data from broiler chickens and turkeys in British Columbia, Canada. Front Vet Sci, 2019; 6:131.

BMEL. General administrative regulation on the recording, analysis and publication of data on the occurrence of zoonoses and zoonotic agents in the food chain (AVV Zoonosen Lebensmittelkette). Last amended by administrative regulation on June 19 2017 (in German). (BAnz AT 23.06.2017 B2). Bundesanzeiger, 2012; N.27.

BVL. Examination of food—Detection of Campylobacter spp. in minced meat—real-time PCR method. Parameter: Campylobacter spp. (in German). In Amtlichen Sammlung von Untersuchungsverfahren nach § 64 LFGB. Bundesamt für Verbraucherschutz und Lebensmittelsicherheit (ed.). Berlin: Beuth Verlag, 2013.

BVL. Reports on food safety—Zoonoses-Monitoring 2014 (in German). Berlin: Bundesamt für Verbraucherschutz und Lebensmittelsicherheit, 2016.

BVL. Reports on food safety—Zoonoses-Monitoring 2016 (in German). Berlin: Bundesamt für Verbraucherschutz und Lebensmittelsicherheit, 2017.

BVL. Reports on food safety—Zoonoses – Monitoring 2017 (in German). Berlin: Bundesamt für Verbraucherschutz und Lebensmittelsicherheit, 2018.

BVL. Reports on food safety—Zoonoses-Monitoring 2018 (in German). Berlin: Bundesamt für Verbraucherschutz und Lebensmittelsicherheit, 2019.

[CLSI] Clinical and Laboratory Standards Institute. Methods for Antimicrobial Dilution and Disk Susceptibility Testing of Infrequently Isolated or Fastidious Bacteria. Wayne, Pennsylvania: Clinical and Laboratory Standards Institute, 2015.

CLSI. Methods for Antimicrobial Dilution and Disk Susceptibility Testing of Infrequently Isolated or Fastidious Bacteria Isolated from Animals. Wayne, Pennsylvania: Clinical and Laboratory Standards Institute, 2017.

10.

Cormican

, Hopkins

, Jarlier

, et al. ECDC, EFSA and EMA Joint scientific opinion on a list of outcome indicators as regards surveillance of antimicrobial resistance and antimicrobial consumption in humans and food-producing animals. EFSA J, 2017; 15:70.

11.

ECDC

EFSA

, EMA. ECDC/EFSA/EMA second joint report on the integrated analysis of the consumption of antimicrobial agents and occurrence of antimicrobial resistance in bacteria from humans and food-producing animals. Joint Interagency Antimicrobial Consumption and Resistance Analysis (JIACRA) Report. EFSA J, 2017; 15:4872.

12.

EFSA

ECDC

. The European Union summary report on antimicrobial resistance in zoonotic and indicator bacteria from humans, animals and food in 2016. EFSA J, 2018a;16:5182.

13.

EFSA

ECDC

. The European Union Summary Report on Trends and Sources of Zoonoses, Zoonotic Agents and Food-borne Outbreaks in 2017. EFSA J, 2018b;16:5500.

14.

EFSA, ECDC. The European Union summary report on antimicrobial resistance in zoonotic and indicator bacteria from humans, animals and food in 2017. EFSA J, 2019; 17:5598.

15.

EFSA, ECDC. The European Union summary report on antimicrobial resistance in zoonotic and indicator bacteria from humans, animals and food in 2017/2018. EFSA J, 2020; 18:6007.

16.

El-Adawy

, Ahmed

, Hotzel

, Tomaso

, Tenhagen

, Hartung

, Neubauer

, Hafez

. Antimicrobial susceptibilities of Campylobacter jejuni and Campylobacter coli recovered from organic turkey farms in Germany. Poult Sci, 2015; 94:2831–2837.

17.

European Commission. Commission Implementing Decision of 12 November 2013 on the monitoring and reporting of antimicrobial resistance in zoonotic and commensal bacteria, 2013/652/EU. Official Journal of the European Union, 2013; L303 (14.11.2013):26–39.

18.

European Council. Council Regulation (EC) No 834/2007 of 28 June 2007 on organic production and labelling of organic products and repealing Regulation (EEC) No. 2092/91. Official Journal of the European Union, 2007; L 189 (20.7.2007):1–23.

19.

Flor

, Käsbohrer

, Kaspar

, Tenhagen

B-A

, Wallmann

. Contributions of the working group on antimicrobial resistance to the evaluation of the 16^th amendment of the German drug act – Chapter 1: Development of antimicrobial sales and use and treatment frequency. (in German). Berlin: Bundesministerium für Ernährung und Landwirtschaft, 2019.

20.

ISO. ISO_10272-1:2017. Microbiology of the Food Chain - Horizontal Method for Detection and Enumeration of Campylobacter spp.—Part 1: Detection Method. Geneva, Switzerland: International Organization for Standardization, 2017.

21.

Kashoma

, Kumar

, Sanad

, Gebreyes

, Kazwala

, Garabed

, Rajashekara

. Phenotypic and genotypic diversity of thermophilic Campylobacter spp. in commercial turkey flocks: A longitudinal study. Foodborne Pathog Dis, 2014; 11:850–860.

22.

Kempf

, Kerouanton

, Bougeard

, Nagard

, Rose

, Mourand

, Osterberg

, Denis

, Bengtsson

. Campylobacter coli in organic and conventional pig production in France and Sweden: Prevalence and antimicrobial resistance. Front Microbiol, 2017; 8:955.

23.

Kittl

, Heckel

, Korczak

, Kuhnert

. Source attribution of human Campylobacter isolates by MLST and fla-typing and association of genotypes with quinolone resistance. PLos One, 2013; 8:e81796.

24.

Kovac

, Cadez

, Stessl

, Stingl

, Gruntar

, Ocepek

, Trkov

, Wagner

, Smole Mozina

. High genetic similarity of ciprofloxacin-resistant Campylobacter jejuni in central Europe. Front Microbiol, 2015; 6:1169.

25.

Luangtongkum

, Morishita

, Ison

, Huang

, McDermott

, Zhang

. Effect of conventional and organic production practices on the prevalence and antimicrobial resistance of Campylobacter spp. in poultry. Appl Environ Microbiol, 2006; 72:3600–3607.

26.

Luo

, Pereira

, Sahin

, Lin

, Huang

, Michel

, Zhang

. Enhanced in vivo fitness of fluoroquinolone-resistant Campylobacter jejuni in the absence of antibiotic selection pressure. Proc Natl Acad Sci U S A, 2005; 102:541–546.

27.

Mayr

, Lick

, Bauer

, Tharigen

, Busch

, Huber

. Rapid detection and differentiation of Campylobacter jejuni, Campylobacter coli, and Campylobacter lari in food, using multiplex real-time PCR. J Food Prot, 2010; 73:241–250.

28.

Murphy

, Carson

, Smith

, Chapman

, Marrotte

, McCann

, Primeau

, Sharma

, Parmley

. Factors potentially linked with the occurrence of antimicrobial resistance in selected bacteria from cattle, chickens and pigs: A scoping review of publications for use in modelling of antimicrobial resistance (IAM.AMR Project). Zoonoses Public Health, 2018; 65:957–971.

29.

Noormohamed

, Fakhr

. Prevalence and antimicrobial susceptibility of Campylobacter spp. in Oklahoma conventional and organic retail poultry. Open Microbiol J, 2014; 8:130–137.

30.

Normand

, Boulianne

, Quessy

. Evidence of cross-contamination by Campylobacter spp. of broiler carcasses using genetic characterization of isolates. Can J Vet Res, 2008; 72:396–402.

31.

Pacholewicz

, Swart

, Schipper

, Gortemaker

, Wagenaar

, Havelaar

, Lipman

. A comparison of fluctuations of Campylobacter and Escherichia coli concentrations on broiler chicken carcasses during processing in two slaughterhouses. Int J Food Microbiol, 2015; 205:119–127.

32.

RKI. Annual report on notifiable diseases for 2018 (in German). Berlin: Robert Koch-Institut, 2019.

33.

Rosner

, Schielke

, Didelot

, Kops

, Breidenbach

, Willrich

, Golz

, Alter

, Stingl

, Josenhans

, Suerbaum

, Stark

. A combined case-control and molecular source attribution study of human Campylobacter infections in Germany, 2011–2014. Sci Rep, 2017; 7:5139.

34.

Varga

, Guerin

, Brash

, Slavic

, Boerlin

, Susta

. Antimicrobial resistance in Campylobacter jejuni and Campylobacter coli isolated from small poultry flocks in Ontario, Canada: A two-year surveillance study. PLoS One, 2019; 14:e0221429.

35.

Vossenkuhl

, Brandt

, Fetsch

, Kasbohrer

, Kraushaar

, Alt

, Tenhagen

. Comparison of spa types, SCCmec types and antimicrobial resistance profiles of MRSA isolated from turkeys at farm, slaughter and from retail meat indicates transmission along the production chain. PLoS One, 2014; 9:e96308.

36.

Wallmann

, Bode

, Heberer

. Recording of sales data of antimicrobials in Germany 2018. Analysis of data provided according to DIMDI-AMV on 2017 and comparison with data from previous years. (In German). Deutsches Tierärzteblatt, 2019; 67:1082–1090.

37.

Zeitouni

, Kempf

. Fitness cost of fluoroquinolone resistance in Campylobacter coli and Campylobacter jejuni . Microb Drug Resist, 2011; 17:171–179.