In partial fulfilment of the Degree Master of Science.
Repeated therapeutic dosing selects macrolide-resistant Campylobacter spp. in a turkey facility
Article first published online: 30 APR 2010
© 2010 The Authors. Journal compilation © 2010 The Society for Applied Microbiology
Journal of Applied Microbiology
Volume 109, Issue 4, pages 1379–1388, October 2010
How to Cite
Logue, C.M., Danzeisen, G.T., Sherwood, J.S., Thorsness, J.L., Mercier, B.M. and Axtman, J.E. (2010), Repeated therapeutic dosing selects macrolide-resistant Campylobacter spp. in a turkey facility. Journal of Applied Microbiology, 109: 1379–1388. doi: 10.1111/j.1365-2672.2010.04765.x
- Issue published online: 16 SEP 2010
- Article first published online: 30 APR 2010
- 2010/0256: received 10 February 2010, revised 2 April 2010 and accepted 23 April 2010
Aims: This study assessed the effects of the therapeutic use of Tylan® in a large-scale turkey production facility on the selection of macrolide-resistant Campylobacter.
Methods and Results: A flock of production turkeys (c. 30 000 birds) was followed from brooding to slaughter, and the effects of macrolide application was assessed in one half of the flock from finishing stage to final product and compared against the control barn where no macrolide was used. Overall, Campylobacter prevalence in turkeys was almost 100% by 4 weeks of age. When Campylobacter prevalence was assessed in relation to treatment, high levels of macrolide resistance were evident in this group following treatment, with Campylobacter coli becoming the dominant strain type. Over time, and in the absence of a selection agent, the population of resistant strains decreased suggesting that there was a fitness cost associated with macrolide resistance carriage and persistence. Macrolide resistance was detected in the control barn at a very low level (four isolates recovered during the study), suggesting that the creation or selection of macrolide-resistant Campylobacter was correlated with the treatment regime used. Molecular analysis of a selection of macrolide-resistant Campylobacter recovered was assessed using PCR, RFLP and sequence analysis of the 23S rRNA. The majority of isolates displaying high-level macrolide resistance (>256 μg ml−1) possessed an A2075G transition mutation in the 23S rRNA and the CmeABC efflux pump.
Conclusions: These studies suggest that macrolide resistance can be promoted through the application of treatment during the grow-out phase and once established in a production facility has the potential to persist and be transferred to final product.
Significance and Impact of the Study: The study highlights the prudent use of antimicrobials in treatment of disease in poultry. Of significance is the presence of macrolide-resistant Campylobacter in poultry production and finished product as a consequence of macrolide usage.
Campylobacter is one of the most commonly isolated foodborne pathogens, being responsible for c. 2·4 million illnesses in the United States annually (CDC 2006). An important risk factor associated with foodborne Campylobacter infection is the consumption of undercooked or improperly prepared poultry products (Friedman et al. 2004). In cases where a Campylobacter infection warrants treatment with antimicrobials, the drugs of choice belong to the macrolide and fluoroquinolone classes of antibiotics (Skirrow and Blaser 2000). The prevalence of macrolide resistance in Campylobacter has been reported at a relatively low level, although the antimicrobial has gained increased attention in the light of the emergence of macrolide-resistant Campylobacter in food production animals (Ge et al. 2003; Cui et al. 2005; Lunangtongkum et al. 2006; Luangtongkum et al. 2009). Macrolide use in poultry is relatively uncommon, and treatment with this agent is usually linked with mycoplasma disease (Jordan et al. 1999).
Macrolide antimicrobials act by inhibiting polypeptide elongation by binding to the ribosome of prokaryotes and blocking the E site of the ribosome (Poulsen et al. 2000; Schlunzen et al. 2001; Franceschi et al. 2004). Both erythromycin (Ery) and tylosin bind the prokaryotic ribosome at the same location on the 23S rRNA subunit, suggesting that resistance to one antimicrobial will probably confer resistance the other (Schlunzen et al. 2001). Two primary mechanisms associated with macrolide resistance in Campylobacter spp. are the CmeABC multidrug efflux system and point mutations in the 23S rRNA genes (Lin et al. 2002; Pumbwe and Piddock 2002; Mamelli et al. 2003; Vacher et al. 2003; Payot et al. 2004). The CmeABC efflux pump is typically associated with low-level macrolide resistance (16 > MIC < 128 mg ml−1) (Payot et al. 2004; Pumbwe et al. 2004; Cagliero et al. 2005; Akiba et al. 2006).
Macrolide resistance is also associated with point mutations of the 23S rRNA (Sigmund et al. 1984; Yan and Taylor 1991). Hydrogen bonding can occur between erythromycin and adenine residues at Escherichia coli 23S rRNA bases 2058 and 2059; a nucleotide change at either position disrupts hydrogen bonding and binding of the erythromycin thus conferring resistance (Schlunzen et al. 2001). High-level resistance to macrolides (>128 mg ml−1) has been observed in studies in which one of the two 23S rRNA mutations has been identified and linked to resistance (Payot et al. 2004; Gibreel et al. 2005; Corcoran et al. 2006). In Campylobacter, an A2075G transition and an A2074C transversion mutation corresponding to bases 2058 and 2059 in E. coli 23S rRNA (Vacher et al. 2003) have been identified as the site of action of erythromycin (Schlunzen et al. 2001; Payot et al. 2004; Gibreel et al. 2005; Corcoran et al. 2006). Gibreel et al. (2005) demonstrated that the mutations of the 23S rRNA were relatively stable and found only one strain of seven examined reverted to erythromycin susceptible after 55 subcultures.
Currently, there is some concern that antibiotic-resistant Campylobacter could be selected at the farm level through the therapeutic or subtherapeutic application of antimicrobials (Aarestrup and Wegener 1999; van den Bogaard and Stobberingh 2000). The effects of macrolide use in broiler production have found different effects (Ladely et al. 2007; Lin et al. 2007). Lin et al. (2007) reported that medicated treatment of broilers with tylosin transiently reduced the level of erythromycin-resistant Campylobacter in birds but did not result in the selection of resistant strains; however, treatment at subtherapeutic levels did promote the emergence of erythromycin-resistant strains. In contrast, Ladely et al. (2007) reported erythromycin resistance among Campylobacter isolates recovered from broilers administered therapeutic or subtherapeutic doses of tylosin. The authors also noted that resistance was detected at a higher frequency in Campylobacter coli than Campylobacter jejuni. In each of these studies, birds were studied under controlled conditions and exposed to Campylobacter either through seeding a group using infected birds or by oral feeding. In turkey production, concern for the use of therapeutic or subtherapeutic macrolide antibiotics that could select for erythromycin-resistant (EryR) Campylobacter species is of concern as the lifespan of these birds is longer leading to a greater period of time for the creation or generation of macrolide-resistant strains which could have the potential to enter into the food chain. The effects of macrolides on the selection of naturally occurring Campylobacter in poultry have not been assessed. This study examined the effect of macrolide use in turkeys on the selection of EryRCampylobacter species occurring naturally in a flock and assessed the nature of resistance in recovered strains. Erythromycin resistance in Campylobacter of turkeys treated with the macrolide Tylosin®, which is approved for the treatment of infectious sinusitis caused by Mycoplasma gallisepticum in commercial turkeys (Jordan et al. 1999; Shryock 2000; Elanco 2010), was assessed.
Materials and methods
Research study location
This study was carried out in cooperation with a turkey production facility located in ND. All analysis and sampling of birds were carried out under the approval of NDSU’s Institutional and Animal and Care and Use Committee (IACUC). The project personnel worked with the farm to identify an incoming flock which was selected as the cohort for the study; the birds were monitored weekly during brooding, finishing and for the duration of the treatment.
Therapeutic administration of medicated water to turkeys in the production setting
A single flock of c. 30 000 (broad breasted white) turkeys was selected for use in this study. The flock was physically separated into two halves (containing c. 15 000 birds each) during the brooding phase (i.e. the first 4 weeks of development) in the brooder barn. Once the poults had reached 4 weeks of age, each half of the brooder barn was transferred into a separate finishing barn where the flock remained until slaughter. One half of the flock was designated as the untreated (control) group and received nonmedicated water, the other half was designated the treated group and received tylosin (Tylan® Soluble; Elanco Animal Health, Greenfield, IN, USA) treatments during three separate dosing periods. All birds were supplied the same standard feed and rations during the study and had a 12 h : 12 h day/night light cycle. Tylosin was first administered to the treated group 3 weeks after faecal cultures were confirmed positive for the presence of Campylobacter species. Treatment periods lasted 3 days each as recommended by the manufacturer at a dosage 0·53 g l−1, approved for the treatment of commercial turkeys. During treatments, medicated water was provided as the only water source to ensure the consumption of at least 60 mg tylosin per kg body weight as instructed by the manufacturer.
The treated group of turkeys was scheduled for the administration of Tylan® Soluble during three separate dosing periods, between weeks 5 and 6; 12 and 13; 16 and 17. Three dosing periods were used to simulate antibiotic treatment of a recurring infection. Each dosing period lasted for 3 days, according to the manufacturer’s recommended duration of treatment. Tylan® Soluble was administered at a concentration as described previously. Farm sample collection began on 24 March 2007 and ceased on 30 July 2007.
A total of 3252 samples were collected during this study as follows. At week 0, caecal swabs (n = 200) were taken to test for the presence of Campylobacter in the newly hatched poults. Faecal samples were collected on a weekly basis to recover Campylobacter isolates from turkeys: for weeks 1 through 5, 100 samples were collected in the control and treated barns; for weeks 6, 7 and 10, 50 samples were collected in the control barn and 100 in the treated barn; for week 11 and 12, 50 samples per barn were collected, for weeks 13 through 18, 50 samples were collected in the control barn and 100 in the treated barn. Samples were not collected during weeks 8 and 9.
Environmental swabs of the brooder barn before placement (n = 20) and prior to placing in the finisher barns (n = 20 for control and n = 20 treated) were collected; samples were also collected from the finisher barns after the birds were removed for processing (n = 16 for each barn). SpongeSicles (Biotrace International, Bridgend, UK) were used to swab a 100 cm2 area of the floors, walls, water feeders and grain feeders.
Samples were collected at the processing facility where the turkeys were processed to final product. One hundred postchill carcass swabs were collected from carcasses using methods described previously (Logue et al. 2003a,b). Caecal contents were collected from birds at pre-evisceration (n = 100) by squeezing contents of the caecum into Whirl-Pack® bags (Nasco, Fort Atkinson, WI). Samples were collected from each flock that was processed, i.e. control and treated birds. Finally, samples of the transport trucks were collected onsite pre (n = 5) and postwash, (n = 5) using Spongilces to swab an area of c. 100 cm2.
A total of 3252 samples were collected from the farm (faecal and environmental) and processing plant (cloacal contents, carcass swabs and truck swabs). The collection consisted of 1810 samples recovered from treated birds and 1350 from the control group.
Faecal swabs, and swabs dipped in the collected caecal contents were subject to both direct plating and enrichment to detect Campylobacter spp. Faecal swabs were placed in 4·0 ml volumes of Preston broth containing growth supplements (CM 067/SR 48, SR 117 and SR 232; Oxoid, Hampshire, UK) and vortexed. To duplicate Campylobacter charcoal desoxycholate agar (CCDA; Oxoid) plates supplemented with CM 739 and SR 155 (Oxoid) (Musgrove et al. 2001; Thorsness et al. 2008), 0·5 ml volumes of the broth were transferred. Preston enrichment broth tubes containing the remainder of the sample were incubated at 42°C for 48 h; following enrichment, the samples were struck to CCDA. All direct plated and enrichment sample plates were incubated microaerophilically at 42°C for 48 h. Suspect Campylobacter colonies on the CCDA plates with a silver grey or creamy grey morphology were selected and struck to blood agar and incubated at 42°C for 48 h. When suspect Campylobacter were not detected on the direct plates, they were picked from plates following enrichment.
Both environmental and carcass swabs were processed by adding 30 ml of buffered peptone water to the swab and homogenizing the sample in a stomacher (Stomacher® 400 Circulator; Seward, Norfolk, UK) for 90 s. To 2 ml of double-strength Preston broth, 2 ml of the homogenate was transferred and incubated at 42°C for 48 h. The enrichment samples were struck to CCDA and processed as described previously.
Owing to size of the collection, c. 25% of all isolates recovered at each time interval collected were chosen for further speciation and analysis of resistance; thus, the collection for characterization consists of 425 isolates representing 425 samples (i.e. one isolate per positive sample).
The species of Campylobacter isolates recovered during the study was determined using multiplex PCR as previously described (Cloak and Fratamico 2002; Thorsness et al. 2008) (Table 1). Quality control strains used were Camp. jejuni ATCC 33560, Camp. jejuni NCTC 11168 and Camp. coli ATCC 33559.
|Gene||Primer sequence||Amplicon size (bp)||Species||Reference|
|cadF||F 5′-TTGAAGGTAATTTAGATATG-3′ R 5′-CTAATACCTAAAGTTGAAC-3′||400||Campylobacter coli, Campylobacter jejuni||Cloak and Fratamico (2002); Konkel et al. (1999)|
|ceuE||F 5′-ATGAAAAAATATTTAGTTTTTGCA-3′ R 5′-ATTTTATTATTTGTAGCAGCG-3′||894||Camp. coli||Cloak and Fratamico (2002); Gonzalez et al. (1997)|
|C-1||F 5′-CAAATAAGTTAGAGGTAGAATGT-3′ R 5′-GGATAAGCACTAGCTAGCTAGCTGAT-3′||160||Camp. jejuni||Cloak and Fratamico (2002); Winters and Slavik (1995)|
|cmeA||F 5′-TAGGCGCGTAATAGTAAATAAAC-3′ R 5′-ATAAAGAAATCTGCGTAAATAGGA-3′||500||Olah et al. (2006)|
|cmeB||F 5′-AGGCGGTTTTGAAATGTATGTT-3′ R 5′-TGTGCCCGCTGGGAAAAG-3′||444||Olah et al. (2006)|
|cmeC||F 5′-AGATGAAGCTTTTGTAAATT-3′ R 5′-TATAAGCAATTTTATCATTT-3′||500||Fakhr and Logue (2007)|
|Campy-23S||F 5′AATTGATGGGGTTAGCATTAGC-3′ R 5′-CAACAATGGCTCATATACAACTGG-3′||316||Vacher et al. (2003)|
Antimicrobial susceptibility test
MIC values for Ery resistance were determined using the agar dilution method as described by the Clinical and Laboratory Standards Institute (CLSI) (CLSI 2006). Campylobacter jejuni ATCC 33560 was used for quality control purposes, with an Ery quality control MIC range of 1–4 μg ml−1. Ery MIC values ≥32 μg ml−1 were considered resistant, while values ≤8 μg ml−1 were deemed susceptible based on CLSI guidelines (Clinical and Laboratory Standards Institute 2006). Ery test concentrations used in the study ranged from 0 to 256 μg ml−1; Ery was obtained from Sigma Chemical Co., St Louis, MO, USA.
Sequence analysis of 23S rRNA gene
A 316-bp fragment of the 23S rRNA gene was amplified by PCR using gene-specific primers (Table 1) (Vacher et al. 2003). PCR parameters used were the following: 94°C for 5 min, 30 cycles of 95°C for 1 min, 49·8°C for 1 min, 72°C for 1 min followed by final extension at 72°C for 10 min. The PCR products were purified using a QIAquick PCR purification kit (Qiagen, Valencia, CA, USA) and subsequently sent to the DNA Sequencing Facility at Iowa State University. Sequencing was carried out on an Applied Biosystems 3730xl DNA Analyzer (Applied Biosystems, Foster City, CA, USA). Sequence data were assembled using the Lasergene Software Suite (DNASTAR, Madison, WI, USA), and the assembled sequences were aligned using the online application MultAlin (Corpet 1988) (http://multalin.toulouse.inra.fr/multalin/multalin.html).
Restriction fragment length polymorphism (RFLP) analysis
RFLP was used to analyse the 316-bp fragment of 23S rRNA. The 316-bp fragment of the 23S rRNA was amplified as previously described (Vacher et al. 2003) (Table 1). Restricted products were run on a precast 10% TBE polyacrylamide gel (Invitrogen, Carlsbad, CA, USA) and stained using EtBr. The gels were visualized using an UVP Biospectrum imager (UVP, Upland, CA). Campylobacter jejuni NCTC 11168 was used as the control strain.
Detection of the genes associated with the CmeABC efflux pump was carried out as described previously. Briefly, DNA was extracted from cultures gown for 48 h on blood agar using the single cell lysing buffer technique (Marmur 1961). All PCR reactions were carried out in 50 μl reaction volumes, using parameters and primers as described previously (Olah et al. 2006; Fakhr and Logue 2007) (Table 1). Amplified product was subjected to horizontal gel electrophoresis in 2% agarose using E gels (Invitrogen). The gel cassettes were loaded onto the Mother E base device and run for 12 min. Gel images were captured using an Alpha Innotech imaging system and the images aligned using E editor software (Invitrogen). Campylobacter jejuni NCTC 11168 was used as the positive control for the PCRs.
A Kruskal–Wallis nonparametric test was performed using the statistical software MiniTab 14 (MiniTab, Inc., State College, PA, USA). The Kruskal–Wallis test was used to compare the distribution of Campylobacter species between treated and control groups. A two proportion test was used to compare the proportion of Camp. jejuni and Camp. coli farm isolates. An α of 0·05 was used in all statistical analyses performed.
A total of 3252 samples were collected from the farm (faecal and environmental) and processing plant (cloacal contents, carcass swabs and truck swabs). The collection consisted of 1810 samples recovered from treated birds and 1350 from the control group. Of the 1810 samples collected from the treated group, 975 suspect Campylobacter spp. were recovered; in the control group, 680 suspect Campylobacter were recovered for a total collection of 1655 suspect isolates. No Campylobacter were recovered from the environmental swabs collected in the barns before or after placement. Owing to the size of the collection, 25% of the suspect isolates (one isolate representing a sample) collected at each time interval was chosen for further analysis and characterization.
Results of isolate identification on 425 isolates chosen for further study were grouped as 231 Camp. jejuni, 188 Camp. coli and 6 isolates that were identified as mixed cultures of both Camp. jejuni and Camp. coli. When broken down by stage of isolation, 362 isolates were associated with the farm and 63 were recovered from the processing plant (see Table 2).
|Campylobacter jejuni||Campylobacter coli||Mixed||Camp. jejuni||Camp. coli||Mixed|
Antimicrobial susceptibility against the macrolide erythromycin was tested for all isolates using the agar dilution method (Clinical and Laboratory Standards Institute 2006). A break point of ≥32 μg ml−1 was considered resistant (Clinical and Laboratory Standards Institute 2006). Immediately after the first dose of tylosin in the treatment group, erythromycin-resistant Campylobacter were detected in faecal samples recovered from the barn. Overall, 103 isolates from the treated group were found to be highly resistant to erythromycin with an MIC >256 μg ml−1. The remaining isolates displayed MICs ranging from 0·5 to 8 μg ml−1. In the time period prior to the first treatment, the MICs for erythromycin in the treated group ranged from 0·5 to 2 μg ml−1. On three separate occasions, macrolide-resistant strains were recovered from the control group – a total of four isolates were recovered; three with an MIC >256 μg ml−1 and one with an MIC of 64 μg ml−1. Aside from these four isolates, no other resistant Campylobacter strains were detected in the control group, and MICs for erythromycin ranged from 0·5 to 4 μg ml−1. Figure 1 shows the prevalence of erythromycin resistance in the control and treated groups for the duration of the study. Of the erythromycin-resistant isolates recovered from the control and treated groups, 92 were identified as Camp. coli and 15 as Camp. jejuni.
Detection of the 23S rRNA mutation
The PCR fragment of the 23S rRNA was amplified as previously described and the amplified product sequenced. The sequences obtained were aligned using the MultAlign sequence alignment tool. Of the 103 isolates sequenced, 85 harboured an A2075G mutation, two had an A2074C mutation and the remaining 16 isolates had sequence that aligned with the wild-type sequence. Most of the isolates displaying the wild-type sequence were identified as Camp. jejuni (14 isolates) and two as Camp. coli, both isolates identified as possessing the A2074C mutation were Camp. coli, and of the isolates identified as possessing the A2075G mutation, 84 were identified as Camp. coli, and one was identified as Camp. jejuni. Visual analysis of the sequence chromatograms from the sequenced strains indicated that the A2074C mutation observed in the two Camp. coli isolates was present in all three copies of the of the 23S rRNA gene in one strain of Camp. coli but not in the other Camp. coli as peak heights of the C at that position were triple that of the A peak, in the second strain, peak height of the C was relatively similar to the A suggesting only one copy of the mutation was present in this strain (Caldwell et al. 2008).
RFLP analysis of a small selection of the isolates that were identified as macrolide resistant and subjected to sequence analysis confirmed the presence of mutations in the 23S rRNA at positions 2074 and 2075. Figure 2a,b show the restriction patterns observed for the 316-bp fragment when restricted using BsaI and BceAI. BsaI failed to digest the 316-bp fragment in wild-type strains and strains harbouring an A2074C mutation (Fig. 2a), while a 201-bp and 115-bp fragment was observed in isolates harbouring an A2075G mutation (lane 6) (Vacher et al. 2003). An additional band of 316 bp was evident in one strain (lane 5; Fig. 2a) suggesting a heterozygous strain type. Such strains have been reported as having 23S rRNA genes containing both mutated (A2075G) and nonmutated copies of the gene in resistant isolates.
Restriction analysis of the 316-bp fragment using BceAI cut at two locations on wild-type DNA and isolate DNA containing an A2075G mutation, yielding 251-, 41- and 24-bp fragments (Fig. 2b; lane 6). In strains containing the A2074C mutation, fragments were evident at 251 and 98 bp in addition to the 24-, 41- and 153-bp fragments (lane 3) (Vacher et al. 2003), this strain was considered heterozygous for both mutations. An additional band was also evident in this same strain (lane 5; Fig. 2b) at 65 bp and in the other lanes and was considered to be unrestricted product of the 41- and 24-bp fragments (lanes 2–6).
Detection of the CmeABC efflux pump
All of the erythromycin-resistant isolates were subjected to further molecular analysis to determine the nature of the resistance observed. A total of 103 isolates were tested for the presence of the efflux pump genes cmeABC using standard PCR protocols described previously (Olah et al. 2006; Fakhr and Logue 2007). Four isolates are missing from this analysis as they were not recoverable from freezer stock. The cmeA gene was detected in 3 Camp. coli and 14 Camp. jejuni; cmeB was detected in 101 isolates tested (one Camp. coli and one Camp. jejuni were negative for the gene) and cmeC was detected in all 103 isolates tested.
This study found that the Campylobacter became established in the study flocks at or near 100% prevalence by 4 weeks of age. Natural contamination of the birds with Campylobacter was probably related to the environment and may have also been influenced by carry-over from previous flocks (Peterson and Wedderkopp 2001; Lee et al. 2005). One study has suggested that the source of these organisms to subsequent flocks remains relatively unknown (Lee et al. 2005). While the facility participating in this study typically operated an ‘all in all out’ procedure, contamination of new birds was probably linked to the previous flock rotating through the brooder and finisher barns. A similar effect was observed in this facility in an earlier study (Thorsness et al. 2008) where the brooder house appeared to be linked to contamination of incoming/rotating flocks moving to finisher barns.
Initial identification of the Campylobacter species found in the control and treated barns up to the time of treatment found there was no significant difference in the prevalence of Camp. jejuni or Camp. coli (P > 0·05). Once tylosin was introduced into the ‘treatment’ group, however, the prevalence of Campylobacter species in this barn differed significantly (P < 0·004) from that of the control barn with Camp. coli becoming the dominant strain type detected. This data suggested that tylosin selects for Camp. coli when applied in a production situation; a similar effect was reported by Ladely et al. (2007) in broilers and contrasts significantly with earlier studies of the same facilities where the only Campylobacter species detected was Camp. jejuni (Thorsness et al. 2008). The current study was carried out c. 1 year after the study of Thorsness et al. suggesting that the birds can easily become infected with other Campylobacter species once the facility is fully established. Studies by other authors (Humphrey et al. 2005; Lee et al. 2005) suggest other species of Campylobacter can dominate in production systems – indeed earlier studies where we (Logue et al. 2003a) examined the prevalence of strain types in processed slaughterline birds found that Camp. coli was the dominant species among all isolates tested.
Susceptibility testing against the macrolide erythromycin found that the majority of the isolates in the treated group displayed high erythromycin resistance at greater than 256 μg ml−1. Such a high level of resistance is usually associated with various mechanisms including alterations in the DNA (usually associated with the 23S rRNA) or efflux-mediated resistance (CmeABC) (Gibreel and Taylor 2006; Payot et al. 2006; Caldwell et al. 2008; Hao et al. 2009; Ladely et al. 2009; Luangtongkum et al. 2009). This study found that the dominant mutation among the highly resistant strains of Campylobacter characterized was linked to the presence of an A2075G transition mutation of the 23S rRNA. Some rarer A2074C transversion mutations were also observed in two isolates one of which one appeared to have mutations in the three copies of the 23S rRNA genes, this was evident based on the peak height of the mutation (height of the C peak compared to the A peak), while the other isolate appeared to only have one copy of the mutation. Others have used sequence chromatograms as an indication of the number of mutations in the 23S rRNA (Caldwell et al. 2008). In general, isolates in our study with high macrolide resistance levels (>256 μg ml−1) were positive for both mutations in the 23S rRNA and the presence of the CmeABC efflux pump. Lower resistance observed in one strain was linked to the presence of the efflux pump genes only (this strain displayed an MIC of 64 μg ml−1).
Of interest in this study was the very low level of naturally occurring macrolide resistance in the nontreated group of birds aside from four strains that were found to possess resistance (three at >256 μg ml−1 and one at 64 μg ml−1). These individual isolates were all recovered from faecal samples of birds on three different sampling occasions (being recovered 2–3 weeks apart) suggesting that naturally occurring macrolide resistance may exist in the control flock albeit it an extremely low level. Aside from these occurrences, the relative absence of macrolide resistance in the control flock supports the theory of the influence of macrolides on the creation and selection of macrolide-resistant Campylobacter as a result of treatment.
The presence of the multidrug efflux pump CmeABC was confirmed by PCR analysis of the three genes of the efflux pump –cmeA, cmeB and cmeC. Of the 103 isolates tested, 101 were positive for the cmeB gene and all were positive for the cmeC gene; however, detection of the cmeA gene was much lower with only 14 of the Camp. jejuni and 3 Camp. coli being positive for the gene; all remaining isolates were negative. Our success in detection of the cmeC in all isolates tested was because of the use of new primers designed to amplify a conserved region of the cmeC gene (Fakhr and Logue 2007). The cmeABC data would appear to suggest that almost all isolates harboured a multidrug efflux pump, but the prevalence of the cmeA gene appeared to be low suggesting some significant genetic variation in the gene. We have seen similar effects previously with the outer membrane gene cmeC, suggesting that significant variation can occur in the structure of the outer membrane portion of the pump (Fakhr and Logue 2007), and this would appear to have some potential bearing on the periplasmic fusion protein CmeA also. This effect was also noted in another study from our lab assessing campylobacters from slaughterline carcasses (Olah et al. 2006). Regardless, further investigation into these strains to assess the genetic variation in the sequence of the cmeA gene is warranted and how this low detection rate is influenced by the Campylobacter species detected and exposure to macrolides.
Mutations in the 23S rRNA were also confirmed by RFLP analysis which displayed restriction patterns showing evidence of the A2075G and A2074C mutations; in addition, there was also evidence of some strains being heterozygous for mutations associated with A2075G and A2074C (Vacher et al. 2003). RFLP was carried out to compare data against sequence analysis data; however, sequence analysis was found to be superior providing results in a significantly shorter time frame without the need for multistep restriction and gel analysis. Our results confirm those of the earlier study of Vacher et al. (2003) and Caldwell et al. (2008) who found that macrolide resistance was most often associated with the A2075G point mutation of the 23S rRNA leading to high-level macrolide resistance. Of interest in our study also was the detection of the rarer A2074C transversion mutation which is also linked with high levels of macrolide resistance (Hao et al. 2009) and evidence that among the three copies of the 23S rRNA for A2074C and A2075G mutants that not all three copies contained the mutation.
Overall, external selective pressure associated with Tylosin® treatment had a significant effect on the selection of macrolide-resistant Campylobacter strains from grow-out flocks of turkeys. In between treatment regimes, the prevalence of macrolide-resistant Campylobacter strains appeared to drop in favour of nonresistant strains, suggesting that there was a high fitness cost associated with the possession of 23S rRNA mutations, an effect that others have also noted (Caldwell et al. 2008; Hao et al. 2009). This data would appear to suggest that in the absence of pressure, the macrolide-resistant strains do not compete well with susceptible strain types (Hao et al. 2009). Our data also demonstrated the presence of the rare A2074C mutation but its low level of detection may also be explained by the fitness cost associated with its carriage and its poor ability to compete with susceptible strains. Further work is currently ongoing to assess if the effects observed in this study are also evident in Campylobacter recovered from turkeys subjected to a subtherapeutic dosing strategy of macrolides.
This study was funded by the USDA CSREES NIFSI award # 2005-5110-03273. The authors gratefully acknowledge the assistance of Ellen Lutgen, Christy Dockter, Aneesa Noormohamed and Mohamed Fakhr with technical aspects of the study.
- 1999) The effects of antibiotic usage in food animals on the development of antimicrobial resistance of importance for humans in Campylobacter and Escherichia coli. Microbes Infect 1, 639–644. and (
- 2006) Interaction of CmeABC and CmeDEF in conferring antimicrobial resistance and maintaining cell viability in Campylobacter jejuni. J Antimicrob Chemother 57, 52–60. , , and (
- 2000) Epidemiology of resistance to antibiotics. Links between animals and humans. Int J Antimicrob Agents 14, 327–335. and (
- 2005) Involvement of the CmeABC efflux pump in the macrolide resistance of Campylobacter coli. J Antimicrob Chemother 56, 948–950. , , and (
- 2008) Development, stability, and molecular mechanisms of macrolide resistance in Campylobacter jejuni. Antimicrob Agents Chemother 52, 3947–3954. , and (
- CDC (2006). FoodNet Surveillance Report for 2004 (Final Report). FoodNet; Centers for Disease Control and Prevention.
- Clinical and Laboratory Standards Institute (2006). Methods for antimicrobial dilution and disk susceptibility testing of infrequently isolated or fastidious bacteria. Approved Guideline M45-A. Clinical and Laboratory Standards Institute, Wayne, PA.
- 2002) A multiplex polymerase chain reaction for the differentiation of Campylobacter jejuni and Campylobacter coli from a swine processing facility and characterization of isolates by pulsed-field gel electrophoresis and antibiotic resistance profiles. J Food Prot 65, 266–273. and (
- 2006) An investigation of the molecular mechanisms contributing to high-level erythromycin resistance in Campylobacter. Int J Antimicrob Agents 27, 40–45. , , and (
- 1988) Multiple sequence alignment with hierarchical clustering. Nucleic Acids Res 16, 10881–10890. (
- 2005) Prevalence and antimicrobial resistance of Campylobacter spp and Salmonella serovars in organic chickens from Maryland retail stores. Appl Environ Microbiol 71, 4108–4111. , , and (
- Elanco (2010) posting date. Tylan Soluble. http://elms.xh1.lilly.com/tylan_soluble_label.pdf (accessed 25 Jan 2010).
- 2007) Sequence variation in the outer membrane protein-encoding gene cmeC, conferring multidrug resistance among Campylobacter jejuni and Campylobacter coli strains isolated from different hosts. J Clin Microbiol 45, 3381–3383. and (
- 2004) Macrolide resistance from the ribosome perspective. Curr Drug Targets Infect Disord 4, 177–191. , , and (
- 2004) Risk factors for sporadic Campylobacter infection in the United States: a case–control study in FoodNet sites. Clin Infect Dis 38(Suppl. 3), S285–S296. , , , , , , , et al. (
- 2003) Antimicrobial resistant Campylobacter species from retail raw meats. Appl Environ Microbiol 69, 3005–3007. , , , , , and (
- 2006) Macrolide resistance in Campylobacter jejuni and Campylobacter coli. J Antimicrob Chemother 58, 243–255. and (
- 2005) Macrolide resistance in Campylobacter jejuni and Campylobacter coli: molecular mechanism and stability of the resistance phenotype. Antimicrob Agents Chemother 49, 2753–2759. , , , , , and (
- 1997) Specific identification of the enteropathogens Campylobacter jejuni and Campylobacter coli by using a PCR test based on the ceuE gene encoding a putative virulence determinant. J Clin Microbiol 35, 759–763. , , , and (
- 2009) 23S rRNA mutation A2074C conferring high-level macrolide resistance in Campylobacter jejuni. Microb Drug Resist 15, 239–244. , , , , and (
- 2005) Prevalence and subtypes of ciprofloxacin-resistant Campylobacter spp. in commercial poultry flocks before, during and after treatment with fluoroquinolones. Antimicrob Agents Chemother 49, 690–698. , , , , , , and (
- 1999) The comparison of an aqueous preparation of tilmicosin with tylosin in the treatment of Mycoplasma gallisepticum infection of turkey poults. Avian Dis 43, 521–525. , , and (
- 1999) Identification of the enteropathogens Campylobacter jejuni and Campylobacter coli based on the cadF virulence gene and its product. J Clin Microbiol 37, 510–517. , , , and (
- 2007) Development of macrolide-resistant Campylobacter in broilers administered subtherapeutic or therapeutic concentrations of tylosin. J Food Prot 70, 1945–1951. , , , , and (
- 2009) 23S rRNA gene mutations contributing to macrolide resistance in Campylobacter jejuni and Campylobacter coli. Foodborne Path Dis 6, 91–98. , , , and (
- 2005) Strain persistence and fluctuations of multiple antibiotic resistant Campylobacter coli colonizing turkeys over successive production cycles. Foodborne Pathog Dis 2, 103–110. , , , , and (
- 2002) CmeABC functions as a multidrug efflux system in Campylobacter jejuni. Antimicrob Agents Chemother 46, 2124–2131. , and (
- 2007) Effect of macrolide usage on emergence of erythromycin-resistant Campylobacter isolates in chickens. Antimicrob Agents Chemother 51, 1678–1686. , , , , and (
- 2003a) The incidence of Campylobacter spp. on processed turkey from processing plants in the midwestern United States. J Appl Microbiol 95, 234–241. , , , and (
- 2003b) The incidence of antimicrobial-resistant Salmonella spp on freshly processed poultry from US Midwestern processing plants. J Appl Microbiol 94, 16–24. , , , and (
- 2009) Antibiotic resistance in Campylobacter: emergence, transmission and persistence. Future Microbiol 4, 189–200. , , , , and (
- 2006) Effect of conventional and organic production practices on the prevalence of antimicrobial resistance of Campylobacter spp. in poultry. Appl Environ Microbiol 72, 3600–3607. , , , , and (
- 2003) A phenylalanine-arginine beta-naphthylamide sensitive multidrug efflux pump involved in intrinsic and acquired resistance of Campylobacter to macrolides. Int J Antimicrob Agents 22, 237–241. , , and (
- 1961) A procedure for the isolation of deoxyribonucleic acid from micro-organisms. J Mol Biol 3, 208–218. (
- 2001) Detection of Campylobacter spp. in ceca and crops with and without enrichment. Poult Sci 80, 825–828. , , , and (
- 2006) Prevalence of the Campylobacter multi-drug efflux pump (CmeABC) in Campylobacter spp. isolated from freshly processed Turkeys. Food Microbiol 23, 453–460. , , and (
- 2004) Relative contribution of target gene mutation and efflux to fluoroquinolone and erythromycin resistance, in French poultry and pig isolates of Campylobacter coli. Int J Antimicrob Agents 23, 468–472. , , , , and (
- 2006) Mechanisms of fluoroquinolone and macrolide resistance in Campylobacter spp. Microbes Infect 8, 1967–1971. , , , , and (
- 2001) Evidence that certain clones of Campylobacter jejuni peaks during successive broiler flock rotations. Appl Environ Microbiol 67, 2739–2745. and (
- 2000) Inhibition of the ribosomal peptidyl transferase reaction by the mycarose moiety of the antibiotics carbomycin, spiramycin and tylosin. J Mol Biol 304, 471–481. , and (
- 2002) Identification and molecular characterisation of CmeB, a Campylobacter jejuni multidrug efflux pump. FEMS Microbiol Lett 206, 185–189. and (
- 2004) Expression of the efflux pump genes cmeB, cmeF and the porin gene porA in multiple-antibiotic-resistant Campylobacter jejuni. J Antimicrob Chemother 54, 341–347. , , and (
- 2001) Structural basis for the interaction of antibiotics with the peptidyl transferase centre in eubacteria. Nature 413, 814–821. , , , , , , and (
- 2000) Growth promotion and feed antibiotics. In Antimicrobial Therapy in Veterinary Medicine, 3rd edn ed. Prescott, J.F., Baggot, J.D. and Walker, R.D. pp. 735–743. Ames, IA: Iowa State Press. (
- 1984) Antibiotic resistance mutations in 16S and 23S ribosomal RNA genes of Escherichia coli. Nucleic Acids Res 12, 4653–4663. , and (
- 2000) Clinical aspects of Campylobacter infection. In Campylobacter, 2nd edn ed. Nachamkin, I. and Blaser, M.J. pp. 69–88. Washington, D.C: ASM Press. and (
- 2008) Baseline Campylobacter prevalence at a new turkey production facility in North Dakota. J Food Prot 71, 2295–2300. , , , and (
- 2003) PCR-restriction fragment length polymorphism analysis for detection of point mutations associated with macrolide resistance in Campylobacter spp. Antimicrob Agents Chemother 47, 1125–1128. , , and (
- 1995) Evaluation of a PCR based assay for specific detection of Campylobacter jejuni in chicken washes. Mol Cell Probes 9, 307–310. and (
- 1991) Characterization of erythromycin resistance in Campylobacter jejuni and Campylobacter coli. Antimicrob Agents Chemother 35, 1989–1996. and (