Download Co-selection may explain high rates of ciprofloxacin non

Journal of Medical Microbiology (2013), 62, 1743–1746
DOI 10.1099/jmm.0.062729-0
Co-selection may explain high rates of ciprofloxacin
non-susceptible Escherichia coli from retail poultry
reared without prior fluoroquinolone exposure
Paul Robert Ingram,1,2 Benjamin A. Rogers,3 Hanna E. Sidjabat,3
Justine S. Gibson4 and Timothy J. J. Inglis1,2
1
Correspondence
Department of Microbiology, PathWest Laboratory, QEII Medical Centre, Hospital Avenue,
Nedlands, Perth, WA, Australia
Paul Robert Ingram
[email protected]
2
School of Pathology and Laboratory Medicine, University of Western Australia, Nedlands, WA,
Australia
3
University of Queensland, UQCCR, Brisbane, QLD, Australia
4
University of Queensland, School of Veterinary Science, Gatton, QLD, Australia
Received 9 May 2013
Accepted 11 August 2013
Australia has never permitted fluoroquinolone use in food-producing animals. We examined local
retail poultry for contamination with fluoroquinolone non-susceptible Escherichia coli, then
explored the hypothesis that their presence may be due to co-selection of resistance
determinants. Between August and November 2010, samples from 30 locally produced,
uncooked retail poultry carcasses from four different processing centres underwent selective
enrichment culture for ciprofloxacin non-susceptible E. coli. Their chromosomal- and plasmidmediated resistance determinants were characterized, and phylogenetic analysis and
transformation experiments were performed. Unexpectedly, we found nine (30 %) of our small
collection of poultry samples carried fluoroquinolone non-susceptible E. coli of which nearly half
possessed aac(6’)-Ib-cr, a novel plasmid-mediated gene encoding an aminoglycoside acetylating
enzyme that also confers fluoroquinolone resistance. All nine isolates were co-resistant to
amoxicillin, gentamicin, tetracycline and trimethoprim/sulfamethoxazole – all antibiotic classes that
are registered for use in poultry reared for food production within Australia. Their unique
phylogenetic relatedness suggested clonal dissemination driven by non-fluoroquinolone selective
pressures. aac(6’)-Ib-cr was successfully transformed and selected for using non-fluoroquinolone
antibiotic pressure. Vertical and perhaps horizontal co-selection may be contributing to the
emergence of fluoroquinolone resistance in poultry and could play a similar role in the human
setting. This suggests that preservation of the usefulness of fluoroquinolones may require more
than just restriction of their use in isolation from other interventions.
INTRODUCTION
Antimicrobial use in food-producing animals combined
with faecal contamination from animals during processing
of carcasses may contribute to transmission of resistant
pathogens and/or resistance determinants from animals to
humans (Cheng et al., 2012; JETACAR, 2009). This is of
particular concern for fluoroquinolones, which are critical
to human health yet are one of the classes of antimicrobial
most commonly used in veterinary medicine (Forcella
et al., 2010). Australia is in the unique position of never
having permitted fluoroquinolone use in food-producing
animals (JETACAR, 2009). Preliminary findings from
our group suggested the prevalence of fluoroquinolone
Abbreviation: rep-PCR, repetitive element palindromic-PCR.
062729 G 2013 SGM
non-susceptible Escherichia coli in Western Australian retail
poultry samples was significantly higher than the 0 %
previously reported in 2007 (Barlow & Gobius, 2008). In
the absence of selective pressure due to fluoroquinolone
use, we explored the hypothesis that this may be due to coselection of resistance determinants.
METHODS
Between August and November 2010, 30 locally produced, uncooked
retail poultry carcasses were obtained from four Western Australian
processing centres and rinsed with 500 ml peptone water. Fifty
millilitres of rinse fluid was added to 50 ml double-strength E. coli
broth and incubated at 44 uC overnight. A 500 ml sample of this broth
was spread evenly onto MacConkey agar containing 1 mg ciprofloxacin ml21 and incubated overnight at 37 uC. Presumptive E. coli were
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 19:35:34
Printed in Great Britain
1743
P. R. Ingram and others
randomly selected and then identified using standard methods. MIC
values were determined by Etest (AB Biodisk; bioMérieux) and
interpreted using Clinical and Laboratory Standards Institute criteria
(CLSI, 2010). For all ciprofloxacin non-susceptible isolates, PCR and
sequencing of gyrA, gyrB, parC and parE genes were performed as
described previously (Komp Lindgren et al., 2003). A multiplex PCR
was used to screen for isolates with qnrA, qnrB and qnrS (Robicsek
et al., 2006b). PCR was used to detect isolates containing qepA1,
qepA2 and aac(6’)-Ib (Wang et al., 2003). The aac(6’)-Ib-cr variant (a
novel plasmid-mediated gene encoding an aminoglycoside acetylating
enzyme that also confers fluoroquinolone resistance) was detected by
sequencing the aac(6’)-Ib gene (Park et al., 2006). Phylogenetic
grouping was determined with multiplex PCR (Clermont et al., 2000).
Genetic relatedness was investigated using repetitive element
palindromic-PCR (rep-PCR) (DiversiLab; bioMérieux). Plasmid
mobility was determined by transformation. Plasmids purified by
alkaline lysis were electroporated into E. coli DH10B using a
previously described method, modified by use of selective medium
containing 4 mg gentamicin l21 (Sidjabat et al., 2009).
RESULTS AND DISCUSSION
Ciprofloxacin non-susceptible E. coli were isolated from
nine (30 %) of the 30 poultry carcasses (originating from
three of the four processing plants). The ciprofloxacin MIC
for all nine isolates was .32 mg l21. Each isolate
demonstrated an identical susceptibility profile, namely
non-susceptibility to amoxicillin, tetracycline, trimethoprim/sulfamethoxazole and gentamicin, but susceptibility
to ceftriaxone and ceftazidime. All the isolates had the same
chromosomal mutations known to be associated with
fluoroquinolone non-susceptibility including those within
gyrA (S83L, D83N) and parC (S80I, E84G), but not gyrB or
parE. Four isolates possessed the aac(6’)-Ib-cr variant. No
qnr or qepA plasmid-mediated quinolone resistance was
detected. Two ciprofloxacin non-susceptible isolates (WA
CH1, WA CH2) were selected for transformation experiments (Table 1). These results demonstrated that genes
encoding amoxicillin, tetracycline and trimethoprim/
sulfamethoxazole resistance were co-located on the same
plasmid as genes encoding aminoglycoside resistance; these
genes could be passed horizontally between E. coli strains
and then be selected out in the presence of an aminoglycoside. When present, the aac(6’)-Ib-cr variant was also
shown to be co-located on the same plasmid. This variant
was subject to the same horizontal transfer and selection
processes in the absence of a fluoroquinolone. All nine
isolates belonged to phylogenetic group D and showed a
high degree of genetic relatedness (Fig. 1). All nine
appeared unrelated to four ciprofloxacin-susceptible isolates from the same cohort of retail poultry samples.
Fluoroquinolone non-susceptible E. coli have been detected
from poultry in countries in which fluoroquinolone use is
currently, or has previously been permitted in the rearing
of food-producing animals. This includes the USA
(Johnson et al., 2007), Europe (0–4 % of isolates) (Gyles,
2008), Iran (42 %) (Moniri & Dastehgoli, 2005) and
Nigeria (55 %) (Fortini et al., 2011), although, to our
knowledge, aac(6’)-Ib-cr has only ever been detected in
retail poultry from China (Huang et al., 2009). Thus, our
finding of a comparatively high rate (30 %) of ciprofloxacin non-susceptible E. coli [nearly half of which possessed
the aac(6’)-Ib-cr gene] in poultry without prior fluoroquinolone exposure was unexpected.
Contamination of carcasses with human faecal flora from
poultry processors is unlikely since all nine isolates had a
unique antibiotic susceptibility profile that is dissimilar to
isolates causing community-onset E. coli human infections
in Australia (Pearson et al., 2007). All nine were genetically
closely related and hence unlikely to have resulted from
contamination at three separate poultry processing centres.
Surreptitious veterinary use of fluoroquinolones is unlikely
because Australia has strict regulatory control over antibiotic
use in animals (JETACAR, 2009). A national study of antibiotic
consumption demonstrated no evidence of quinolone use in
stock feed between 1992 and 1997 (JETACAR, 2009).
A more plausible explanation for our findings is co-selection
due to the use of non-fluoroquinolone antimicrobials in
Table 1. Antimicrobial MIC testing and molecular characterization of quinolone resistance determinants in the parent and
transformant isolate
Study isolate
WA CH2
WA CH1
Parent
Transformant
Parent
Transformant
.16
.256
.256
.32
.32
.16
.256
.256
.32
,0.002
.16
.256
.256
.32
.32
.16
.256
.256
.32
0.047
Yes
Yes
No
No
Yes
No
Yes
Yes
Yes
No
Yes
Yes
21
Phenotype (MIC, mg l )
Gentamicin
Amoxicillin
Tetracycline
Trimethoprim/sulfamethoxazole
Ciprofloxacin
Genotype
gyrA (S83L, D83N), parC (S80I, E84G)
aac(6’)-Ib
aac(6’)-Ib-cr
1744
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 19:35:34
Journal of Medical Microbiology 62
Fluoroquinolone non-susceptible E. coli in poultry
Diversilab v3.4
PC
#178
80
85
90
95
Similarity (%)
Key
Sample ID
1
WA CH25
2
WA CH14
3
WA CH5
4
WA CH3
5
WA CH27
6
WA CH26
7
WA CH2
8
WA CH1
9
WA_CHC_17
10
WA CH42
11
WA CH38
12
WA CH30
13
WA CH28
100
Fig. 1. Dendrogram showing results of rep-PCR typing of the nine ciprofloxacin non-susceptible E. coli isolates (WA CH1, -2,
-3, -5, -14, -25, -26 and -27 and WA_CHC_17) compared to four ciprofloxacin-susceptible isolates (WA CH28, -30, -38 and
-42) from the same cohort of retail poultry samples.
food-producing animals (Fortini et al., 2011; Gyles, 2008).
All nine isolates demonstrated co-resistance to antibiotic
classes that are registered for therapeutic or prophylactic
use in poultry reared for food production within Australia
(JETACAR, 2009). Co-selection may be driven by one of
two mechanisms. Firstly, a single resistance mechanism [e.g.
aac(6’)-Ib-cr or a multi-drug efflux pump] may confer
resistance to multiple antibiotic classes concurrently.
Secondly, two different antibiotic resistance determinants
(usually co-located) may be vertically or horizontally
passed between bacterial isolates. The unique genetic
relatedness of all nine isolates suggests that dissemination
of chromosomal quinolone resistance determinants is a
clonal phenomenon (vertical co-selection), although the
origins of this clone are unknown. Our transformation
experiments demonstrate that dissemination of plasmidmediated quinolone resistance determinants such as
aac(6’)-Ib-cr may also occur in the context of nonfluoroquinolone selective pressure (horizontal co-selection). Although acquisition of aac(6’)-Ib-cr confers only a
minor rise in fluoroquinolone MIC, it may promote
subsequent mutations that confer high-level resistance
(Robicsek et al., 2006a). Plasmid heterozygosity may
explain why only some isolates appeared to carry the
aac(6’)-Ib-cr variant (Novick & Richmond, 1965).
http://jmm.sgmjournals.org
Co-selection has been proposed as an explanation for the
emergence of fluoroquinolone resistance in Gram-negative bacteria in both humans (Park et al., 2006) and
animals (Gyles, 2008). However, in these studies, the
relative contribution of co-selection versus direct selective
pressure has been difficult to discern in the context of
active fluoroquinolone use. As our study was performed
in a country that has never permitted fluoroquinolone use
in food-producing animals, our results suggest that
vertical and perhaps horizontal co-selection may be
contributory forces behind the emergence of fluoroquinolone resistance. This necessitates a re-evaluation of the
mechanisms contributing to this problem, and our
attempts to address it. If co-selection does explain our
findings, then efforts to preserve the long-term future of
fluoroquinolones, whether in animals or humans, may
require more than just restriction of their use in isolation
from other interventions.
ACKNOWLEDGEMENTS
We wish to thank PathWest Food and Waters scientific staff for their
assistance in processing of the retail poultry specimens. This study
was supported by internal funding.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 19:35:34
1745
P. R. Ingram and others
REFERENCES
Komp Lindgren, P., Karlsson, A. & Hughes, D. (2003). Mutation rate
Barlow, R. & Gobius, K. (2008). Pilot Survey for Antimicrobial Resistant
and evolution of fluoroquinolone resistance in Escherichia coli isolates
from patients with urinary tract infections. Antimicrob Agents
Chemother 47, 3222–3232.
Bacteria in Australian Food. Queensland, Australia: Food Science Australia,
for the Australian Government Department of Health and Ageing.
Moniri, R. & Dastehgoli, K. (2005). Fluroquinolone resistant
successful regulation, Australia. Emerg Infect Dis 18, 1453–1460.
Escherichia coli isolated from healthy broilers with previous exposure
to fluoroquinolones: is there a link? Microb Ecol Health Dis 17, 69–
74.
Clermont, O., Bonacorsi, S. & Bingen, E. (2000). Rapid and simple
Novick, R. P. & Richmond, M. H. (1965). Nature and interactions
determination of the Escherichia coli phylogenetic group. Appl Environ
Microbiol 66, 4555–4558.
of the genetic elements governing penicillinase synthesis in
Staphylococcus aureus. J Bacteriol 90, 467–480.
CLSI (2010). Performance Standards for Antimicrobial Susceptibility
Testing: 20th Informational Supplement M100-S20. Wayne, PA:
Clinical and Laboratory Standards Institute.
Park, C. H., Robicsek, A., Jacoby, G. A., Sahm, D. & Hooper, D. C.
(2006). Prevalence in the United States of aac(69)-Ib-cr encoding a
Cheng, A. C., Turnidge, J., Collignon, P., Looke, D., Barton, M. &
Gottlieb, T. (2012). Control of fluoroquinolone resistance through
Forcella, C., Alessiani, A., Perilli, M., Zilli, K., Di Giannatale, E. &
Amicosante, G. (2010). Characterization of quinolone resistance in
Escherichia coli strains of animal origin from Italy. J Chemother 22,
165–168.
ciprofloxacin-modifying enzyme. Antimicrob Agents Chemother 50,
3953–3955.
Pearson, J., Turnidge, J., Franklin, C., Bell, J. & Australian Group on
Antimicrobial Resistance (2007). Prevalence of antimicrobial
Fortini, D., Fashae, K., Garcı́a-Fernández, A., Villa, L. & Carattoli, A.
(2011). Plasmid-mediated quinolone resistance and b-lactamases in
resistances in common pathogenic Enterobacteriaceae in Australia,
2004: report from the Australian Group on Antimicrobial Resistance.
Commun Dis Intell Q Rep 31, 106–112.
Escherichia coli from healthy animals from Nigeria. J Antimicrob
Chemother 66, 1269–1272.
Robicsek, A., Strahilevitz, J., Jacoby, G. A., Macielag, M., Abbanat, D.,
Hye Park, C., Bush, K. & Hooper, D. C. (2006a). Fluoroquinolone-
Gyles, C. L. (2008). Antimicrobial resistance in selected bacteria from
poultry. Anim Health Res Rev 9, 149–158.
modifying enzyme: a new adaptation of a common aminoglycoside
acetyltransferase. Nat Med 12, 83–88.
Huang, S. Y., Dai, L., Xia, L. N., Du, X. D., Qi, Y. H., Liu, H. B., Wu, C. M.
& Shen, J. Z. (2009). Increased prevalence of plasmid-mediated
Robicsek, A., Strahilevitz, J., Sahm, D. F., Jacoby, G. A. & Hooper,
D. C. (2006b). qnr prevalence in ceftazidime-resistant Enterobac-
quinolone resistance determinants in chicken Escherichia coli isolates
from 2001 to 2007. Foodborne Pathog Dis 6, 1203–1209.
teriaceae isolates from the United States. Antimicrob Agents
Chemother 50, 2872–2874.
JETACAR (2009). The Use of Antibiotics in Food-Producing Animals:
Sidjabat, H. E., Paterson, D. L., Adams-Haduch, J. M., Ewan, L.,
Pasculle, A. W., Muto, C. A., Tian, G. B. & Doi, Y. (2009). Molecular
Antibiotic-Resistant Bacteria in Animals and Humans. Report of the
Joint Expert Technical Advisory Committee on Antibiotic Resistance
(JETACAR). Canberra: Biotext.
epidemiology of CTX-M-producing Escherichia coli isolates at a
tertiary medical center in western Pennsylvania. Antimicrob Agents
Chemother 53, 4733–4739.
Johnson, J. R., Sannes, M. R., Croy, C., Johnston, B., Clabots, C.,
Kuskowski, M. A., Bender, J., Smith, K. E., Winokur, P. L. & Belongia,
E. A. (2007). Antimicrobial drug-resistant Escherichia coli from
Wang, M., Tran, J. H., Jacoby, G. A., Zhang, Y., Wang, F. & Hooper,
D. C. (2003). Plasmid-mediated quinolone resistance in clinical
humans and poultry products, Minnesota and Wisconsin, 2002–
2004. Emerg Infect Dis 13, 838–846.
isolates of Escherichia coli from Shanghai, China. Antimicrob Agents
Chemother 47, 2242–2248.
1746
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Fri, 12 May 2017 19:35:34
Journal of Medical Microbiology 62