OD2028 Final Report - Defra Science Search

Transcription

General Enquiries on the form should be made to:
Defra, Procurements and Commercial Function (Evidence Procurement Team)
E-mail: [email protected]
Evidence Project Final Report



Note
In line with the Freedom of Information
Act 2000, Defra aims to place the results
of its completed research projects in the
public domain wherever possible.
The Evidence Project Final Report is
designed to capture the information on
the results and outputs of Defra-funded
research in a format that is easily
publishable through the Defra website
An Evidence Project Final Report must
be completed for all projects.
This form is in Word format and the
boxes may be expanded, as appropriate.
ACCESS TO INFORMATION
The information collected on this form will
be stored electronically and may be sent
to any part of Defra, or to individual
researchers or organisations outside
Defra for the purposes of reviewing the
project. Defra may also disclose the
information to any outside organisation
acting as an agent authorised by Defra to
process final research reports on its
behalf. Defra intends to publish this form
on its website, unless there are strong
reasons not to, which fully comply with
exemptions under the Environmental
Information Regulations or the Freedom
of Information Act 2000.
Defra may be required to release
information, including personal data and
commercial information, on request under
the Environmental Information
Regulations or the Freedom of Information Act 2000. However,
Defra will not permit any unwarranted breach of confidentiality or
act in contravention of its obligations under the Data Protection
Act 1998. Defra or its appointed agents may use the name,
address or other details on your form to contact you in
connection with occasional customer research aimed at
improving the processes through which Defra works with its
contractors.
Project identification
1.
Defra Project code
2.
Project title
OD 2028
A generic tool box for the molecular epidemiology of
CTX-M bearing verotoxigenic E.coli & vaccine trial.
3.
Contractor
organisation(s)
Animal Health and Veterinray
Laboratories Agency (AHVLA)
Woodham lane
New Haw
Surrey
KT15 3NB
54. Total Defra project costs
(agreed fixed price)
5. Project:
EVID4 Evidence Project Final Report (Rev. 06/11) Page 1 of 29
£
905,939
st
start date ................
1 April 2009
end date .................
31 March 2012
st
6. It is Defra’s intention to publish this form.
Please confirm your agreement to do so. ...................................................................................... YES x NO
(a) When preparing Evidence Project Final Reports contractors should bear in mind that Defra intends that
they be made public. They should be written in a clear and concise manner and represent a full account
of the research project which someone not closely associated with the project can follow.
Defra recognises that in a small minority of cases there may be information, such as intellectual property
or commercially confidential data, used in or generated by the research project, which should not be
disclosed. In these cases, such information should be detailed in a separate annex (not to be published)
so that the Evidence Project Final Report can be placed in the public domain. Where it is impossible to
complete the Final Report without including references to any sensitive or confidential data, the
information should be included and section (b) completed. NB: only in exceptional circumstances will
Defra expect contractors to give a "No" answer.
In all cases, reasons for withholding information must be fully in line with exemptions under the
Environmental Information Regulations or the Freedom of Information Act 2000.
(b) If you have answered NO, please explain why the Final report should not be released into public domain
Executive Summary
7.
The executive summary must not exceed 2 sides in total of A4 and should be understandable to the
intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together
with any other significant events and options for new work.
The transmission of commensal enteric bacteria bearing Extended Spectrum Beta Lactamases (ESBLs)
through the food chain to humans is poorly understood. Recent studies in this, and other Defra funded
projects (OD2023 & VM2207), have demonstrated that similar plasmids bearing ESBLs can be found in
humans and animals suggesting possible linkage. Improved tools for the molecular typing of plasmids are
required to investigate such links, as are approaches for the removal of such bacteria from food producing
animals and the farm environment. This project has provided both improved tools for plasmid typing,
evaluated a vaccine for treatment of cattle with ESBL E. coli and described the persistence of such
organisms on a farm.
The first objective considered the persistence of an O26 verotoxigenic E. coli (VTEC) bearing a CTX-M
sequence type 14 ESBL on a large dairy farm. The farm was selected because it had previously been
positive for VTEC O26 CTX-M-14 ESBL E. coli. Two preliminary visits were made to this farm and were
unable to detect the O26 VTEC, but E. coli bearing the CTX-M-14 ESBL were found in abundance.
Subsequently a longitudinal epidemiological study was conducted where faecal and environmental
samples were collected to isolate presumptive E. coli and presumptive E .coli carrying CTX-M genes.
Isolates were characterised by pulsed field gel electrophoresis or by serotype. The presence of CTX-M
genes was confirmed (to group level) by real-time Light Cycler PCR and a selection of isolates was
analysed to determine the blaCTX-M gene sequence type. The proportion of faecal samples positive for E.
coli carrying CTX-M genes was consistently high at each visit, ranging from 78% to 100% of the samples
tested. The highest prevalence of CTX-M E. coli were obtained from faecal samples from high yield cows
(92%-100%) and individual calves (85%-100%). Pen swabs from animal pens and faecal samples from
the dry cow group had lower numbers of samples positive for CTX-M E. coli (47-83%). E. coli isolates
carrying CTX-M genes were widespread on the farm and comprised a diverse set of strain types by PFGE.
The CTX-M gene types identified were CTX-M-14 (~98% of isolates tested) and CTX-M-15 (~2%).
6
Approximately 48% of calves were shedding CTX-M E. coli in faeces at levels greater than 1x10 cfu/g.
Individual calves were found to be shedding persistently for up to 64 days (median 36 days). The high
degree of diversity in the PFGE strain types of CTX-M E. coli isolated from the farm suggest that the CTXM-14 plasmid is readily transmissible within the commensal E. coli population. Persistently high levels of
shedding of CTX-M positive E. coli suggest that there are unidentified drivers that are maintaining these
strain types in animals on parts of the farm.
Plasmids of veterinary origin were characterised by a range of molecular and phenotypic methods to
develop techniques with sufficient resolution to understand their movement on farms and through the food
chain. Plasmids bearing the CTX-M gene in E. coli isolated from cattle, chickens and turkeys in Great
Britain were analysed for replicon type and their ability to transfer plasmids. Most isolates from different
animal species transferred their plasmids with similar frequencies. The plasmid replicon type I1- was
most common and seen in 23%, 95% and 50% of the isolates tested from cattle, chickens and turkeys
respectively, whilst types F, FIA, FIB and K were common to isolates from cattle and turkeys only. This
study provides a baseline of the characteristics of CTX-M E. coli isolates from animals in Great Britain and
suggests that chicken and cattle CTX-M E. coli represent different populations. A panel of 25 plasmids
was assembled, their DNA extracted, purified and sequenced using a Roche 454 sequencer. The
sequences were assembled using Newbler software and BLAST was used to identify regions of similarity.
Many of the plasmids were similar to those described in the literature. Plasmid DNA sequence data was
compared using Mauve to determine both regions of homology and heterogeneity.
The original plan was to use this plasmid DNA sequence information for the discovery of molecular
markers and their translation to the micro-array format for typing. This approach has been modified to a
PCR and/or sequencing based approach following the sale of the AHVLA micro-array technology. Current
plasmid typing schemes are limited to incompatibility group (n=18) determination using a PCR based
replicon typing scheme. Multi locus sequence typing is available for 4 of these incompatibility groups. In
the present study a generic multi-locus sequence typing scheme was developed for ESBL plasmids and
combined with a high resolution PCR test for the presence of certain genes. The pMLST system was
based on the sequencing of 6 genes (parB, psiAB, tnpA, klcA, yfhA and dnameth) common to most ESBL
plasmids. For more detailed high resolution discrimination between plasmids of a particular type, such as
IncI-1γ, PCRs for the detection of unique molecular markers (typically five to six) were developed. This
new scheme was applied to 270 ESBL and non-ESBL plasmids. The pMLST data was presented as
phylogenetic trees for each locus and revealed grouping for plasmids bearing a CTX-M ESBL of a certain
group (e.g. 1 or 9). Further PCR analysis for molecular markers has provided further discrimination.
An autogenous vaccine was prepared from E. coli bearing the blaCTX-M gene and evaluated in cattle as a
potential control measure to reduce shedding and dissemination of these organisms into the food chain.
Calves (n=30) received either autogenous vaccine prepared from O33 E. coli bearing the CTX-M-14 gene
or placebo by intramuscular injection. After 2 weeks all calves were challenged by oral gavage with 1 x
10
10 cfu of O33 CTX-M-14 E. coli. Faecal samples were taken daily and blood samples every 2 weeks for
enumeration of total and CTX-M-14 E. coli and immunological assays respectively. The cumulative
number of total E. coli excreted at 7 days was significantly (p=0.006) lower in the vaccinated group than
placebo. However, there was no significant difference in the shedding of either CTX-M-14 or total E. coli
between vaccinated and placebo calves throughout the study period. Furthermore, there was no
significant difference in the cumulative excretion at 28 days for the surviving calves (p=0.3). The systemic
immune response to E. coli O33 antigen was tested by ELISA and was significantly higher (p<0.001) in
vaccinated than placebo calves. There was no significant difference in the mucosal immune response.
In conclusion, epidemiological studies at a single dairy farm have revealed the movement of plasmids
between E. coli of different PFGE and O antigen types. The development of the pMLST /molecular
marker scheme enables the movement of such plasmids to be traced with high resolution. This will allow
more detailed investigation of those plasmids found in both food producing animals and humans and
enable the application of focussed and relevant intervention measures. An autogenous vaccine for O33
CTX-M-14 E. coli showed little to no value for preventing the transmission of such organisms through the
food chain.
Project Report to Defra
8.
As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with details of
the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra
to publish details of the outputs to meet Environmental Information Regulation or Freedom of Information
obligations. This short report to Defra does not preclude contractors from also seeking to publish a full,
formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively
encourages such publications as part of the contract terms. The report to Defra should include:
 the objectives as set out in the contract;
 the extent to which the objectives set out in the contract have been met;
 details of methods used and the results obtained, including statistical analysis (if appropriate);
 a discussion of the results and their reliability;
 the main implications of the findings;
 possible future work; and
 any action resulting from the research (e.g. IP, Knowledge Exchange).
There were five main objectives to this work, as detailed below. Each of these will be discussed separately and then
drawn together in a final discussion at the end.
Objective 01/ Epidemiological investigation of CTX-M 14 gene in E. coli, gut commensals and ability to survive in
different farm environments.
Objective 02/ Detailed molecular characterisation of plasmids for marker selection.
Objective 03/ Production of a European database for CTX-M plasmid molecular data collation.
Objective 04/ Development and evaluation of an autogenous vaccine for E. coli bearing the CTX-M14 plasmid.
Objective 05/ Development of a generic molecular tool box for molecular typing of plasmids.
OBJECTIVE 01:- Epidemiological investigation of CTX-M 14 in farms and mathematical modelling
Introduction. The CTX-M family of extended spectrum beta lactamase (ESBL) enzymes (Bonnet, 2004; Livermore
rd
th
2008) have spread throughout Europe and the world and constitute a real threat to the efficacy of 3 and 4
generation cephalosporins for the treatment of hospital and community acquired infections (Bonnet, 2004; Canton et
al., 2008; Coque et al., 2012; Perez et al., 2007; Rodríguez-Baño et al., 2008). CTX-M ESBLs have been found in E.
coli present in cattle from European countries including France (Meunier at al, 2006), Germany (Guerra et al., 2012)
and Spain (Brinas et al., 2005). CTX-M ESBLs have also been found in E .coli present in cattle from other parts of
the world including Japan (Shiraki et al., 2004) and Hong Kong (Duan et al, 2006).
CTX-M ESBL E. coli carrying CTX-M-14 genes were first detected on a UK dairy farm in 2005 and 2006 (Liebana et
al, 2006; Teale et al., 2005). More recent passive surveillance studies in the regions of North Wales and the West
Midlands areas of the UK have confirmed that there are a number of dairy farms in this part of the country that have
CTX-M ESBLs present in E. coli isolates (Snow et al., 2011). CTX-M-15 genes have also been detected in E. coli
isolates from UK dairy farms (Horton et al., 2011; Snow et al., 2011; Watson et al., 2012).
Previous work at AHVLA has shown heifers and cows to be nearly eight times more likely to test positive for CTX-M
E. coli in the 10 days post-calving than in a similar period pre-calving (Watson and others 2012). This suggests the
calving environment is a potential reservoir of CTX-M E. coli.
4
Shedding levels (> 1x10 cfu/g) of CTX-M positive E. coli have also recently been reported in cattle from the UK
(Horton et al., 2011). The factors responsible for such levels of shedding in faeces are not well understood at
present. Therefore, the aim of this study was to gain a better understanding of: the temporal persistence of CTX-M
positive E. coli on a UK dairy farm, the extent of dissemination within the farm environment, the bacterial strain types
present on farm and the persistence of shedding density of E. coli bacteria carrying CTX-M ESBL genes from calves.
The data obtained are currently being used to inform a model for CTX-M E. coli transmission dynamics which it is
hoped will help to identify the most effective interventions in farm management practice that can be applied to
decrease the prevalence of CTX-M positive bacteria carried by cattle before they reach slaughter. The data obtained
were used to inform a model for CTX-M E. coli transmission dynamics in the bovine gut which investigated the effect
rd
th
of 3 and 4 generation cephalosporin use on CTX-M prevalence.
Milestones 01/01-01/02: Longitudinal epidemiology study & survivability in different environments
Two visits (11/08/09 and 19/10/09) to a farm in Northern England were initially made to establish the continued
presence of the CTX-M14 026 VTEC. We were unable to detect either CTX-M O26 VTEC or VTEC O26 at either
visit. Both slide agglutination methods (visit 1) and enrichment with Dynabeads (visit 2) were used. The somatic
antigen of 50 CTX-M positive and 50 CTX-M negative isolates was determined. The most abundant serotypes in the
50 ESBL positive isolates were O152 (n=23) and O106 (n=10). This objective was slightly modified to assess
plasmid transfer in E. coli rather the previous focus on O26 VTECs.
Following these first two visits, a total of six visits were made as summarised below.
Summary. A longitudinal study was conducted with 6 visits at two week intervals during August and October
2010. A total of 1409 samples were collected (including n=813 faecal samples from cows, n=312 faecal samples
from calves and n=284 environmental samples) and were cultured on two agars, one selective for E. coli and the
other selective mainly for E. coli carrying CTX-M genes. CTX-M gene presence was confirmed by real-time PCR
and gene sequencing on selected isolates. Strain types were characterised by PFGE and by serotype.
The proportion of faecal samples (pats) positive on selective agar for E. coli isolates carrying CTX-M genes was
consistently high at each visit, ranging from 78% to 100% of the samples tested (Table 1). E. coli isolates
carrying CTX-M genes comprised a diverse set of strain types by PFGE. The CTX-M gene types identified were
CTX-M-14 (~98% of isolates tested) and CTX-M-15 (~2%). Approximately 48% of calves were shedding CTX-M
6
E. coli in faeces at levels greater than 1x10 cfu/g.
The high degree of diversity in the PFGE strain types of CTX-M E. coli isolated from the farm suggest that the
plasmid is readily transmissible within the commensal E. coli population.
The high prevalence of CTX-M E. coli in calves is interesting. The third and fourth generation cephalosporin
antibiotics which could select for bacteria producing the plasmid are not commonly used in calves. However it is
possible that calves may be exposed to residues of these or other beta-lactam antibiotics through waste milk from
treated cows. Calves on this farm were fed waste milk, including milk from cows with mastitis. The first milk after
antibiotic treatment was not discarded.
Materials and methods for objective 1
Media and QC isolates. CHROMagar ECC (selective for Enterobacteriaceae) and CHROMagar CTX (mainly
selective for CTX-M positive Enterobacteriaceae (Randall et al., 2009) media were obtained from M-Tech Diagnostics
Ltd, UK.
Farm visits and sampling. A dairy farm known to be previously positive for VTEC O26 CTX-M14 ESBL E. coli was
recruited for this study. The farm was a commercial dairy unit made up of largely Holstein cattle, with approximately
250 lactating cows and 40 unweaned calves. Calves were moved from the holding at 3-4 months to a nearby heifer
rearing unit (under separate ownership) and stock returned to the main farm as calving heifers at approximately 2
years of age. Animals moved onto the rearing site were not sampled.
Six visits were made to the dairy farm between mid-August 2010 and the beginning of October 2010 at approximately
two week intervals. At the first visit the farmer was asked to complete a detailed questionnaire about cattle
movements, calving practices, feeding, housing, biosecurity and the farm environment. At each visit an animal
tracking sheet was completed detailing the movements of individual calves and cattle groups on the farm.
Management details including any veterinary treatments were recorded along with details of weather conditions on
and prior to the visit.
At the initial visit, fresh faecal samples were collected from 25 individual calves with priority given to the youngest
animals. Where possible the same calves were sampled at subsequent visits but where they were unavailable the
numbers were made up by sampling newly recruited calves. A sample size calculation indicated that 20 animals
would be sufficient in a group size of 250 animals to detect presence of the CTX-M plasmid at a prevalence of 15%
with 95% confidence.
In addition to the individually identified samples, at every visit up to 30 fresh faecal pat samples (100 g) were
collected from calves, 50 from high-yield cows, 50 from low-yield cows, and up to 40 from dry cows. Up to 49
environmental samples were collected using hand and boot swabs in buffered peptone water from general sites
around the farm including the calving box, drains, manure heap, pens and wild bird droppings and the same sites
were sampled on each visit (Table 1).
Isolation of CTX-M E. coli. The samples were screened for CTX-M positive E. coli by culturing (aerobically)
overnight at 37°C in buffered peptone water (BPW) to enrich the total bacterial population. Presumptive CTX-M
positive E. coli were then isolated from the BPW broth by plating a 10 µl aliquot onto CHROMagar CTX medium
(Randall et al., 2009) and incubating as before.
Estimation of E. coli counts from faeces .The number of E. coli present in faecal samples was obtained by using a
serial dilution method based on that described previously by Miles and Misra (1938). For those samples that had
overall counts of CTX-M E. coli below the detection limit (100 cfu/g), an aliquot (100 µl) of the enriched BPW broth
mixture (after culture at 37°C 18 hours) was plated onto CHROMagar CTX and incubated overnight. Samples that
had previously shown no growth after direct plating but did produce growth on CHROMagar CTX from enriched broth
were reported as CTX-M positive but below the detection limit (<100 cfu/g).
Bacterial identification. Matrix-assisted laser desorption ionization-time of flight (MALDI ToF) mass spectrometry
(Autoflex II, Bruker Daltonics Ltd, UK) together with the manufacturer’s Biotyper database identification software
(Mellman et al., 2008) was used to confirm the identity of most of the bacterial isolates producing blue / green
colonies on CHROMagar CTX (presumptive E. coli).
Confirmation of CTX-M gene group. The presence of CTX-M genes (identified to group level) in isolates from
CHROMagar CTX was confirmed by real-time light cycler PCR (LightCycler 2.0 System and the LightCycler Fast Start
DNA master SYBR Green I kit, Roche) as described previously (Horton et al., 2011).
Sequencing of blaCTX-M genes from selected isolates. A selection of CTX-M positive E. coli isolates from cattle
(n=300) were analysed to determine the blaCTX-M gene sequence type by using an ABI 3730 DNA Analyser (Applied
Biosystems, Warrington, UK) and by use of the ABI SeqScape program.
Strain type characterization. Pulsed field gel electrophoresis (PFGE) was used to characterize E. coli strain types
following macro restriction of bacterial DNA by Xba1, and by using the standard PulseNet USA protocol (Ribot et al.,
2006). There were 179 isolates selected for PFGE analysis. Strain types were designated on the basis of PFGE
profiles having greater than 75% similarity and strain type diversity was calculated using Simpson’s method
(Simpson, 1949).
Serotype characterization. A total of 67 presumptive E. coli isolates were selected for serotyping and most of these
(n=58) had also been characterised by PFGE. Serotyping was performed based on previously published methods
(Linton et al., 1979).
Statistical analysis A chi-squared procedure was used to assess the differences in the proportions of calves
shedding at medium, high or very high shedding densities between visits, and this was implemented using GraphPad
Prism version 4.00 software (GraphPad Software, San Diego California).
Results
Antibiotic treatments. Veterinary treatments were only recorded for the calves that were individually followed
throughout the study. The only veterinary treatment recorded throughout the study was a single calf treated for
pneumonia with Resflor (florfenicol) on visit 4.
The percentage of samples containing CTX-M positive E. coli. A total of 1409 samples were collected from a
variety of farm sources including both faecal samples and swabs from different locations around the farm over the six
visits. The percentage of samples positive for CTX-M E. coli are shown in Table 1 for each of the 6 visits. The
highest prevalence of CTX-M E. coli (Table 1) were obtained from faecal samples from high yield cows (92%-100%)
and individual calves (85%-100%). Pen swabs (P) from animal pens and faecal samples from the dry cow group
(DC) had lower numbers (Table 1) of samples positive for CTX-M E. coli (47-83%). The proportion of samples positive
for CTX-M E. coli from each location was similar at each visit. The data indicate that the CTX-M gene was present in
E. coli on the farm for at least a period of 78 days and that the bacteria carrying the blaCTX-M gene were widespread
throughout the farm.
Table 1. CTX-M E. coli status of samples from the farm at six visits
% of CTX-M positive samples (number of samples collected from each
visit)
Code
Sample description
Overall percentage
(n total)
1
2
3
4
5
6
Faecal pat samples from the high yield cow group
100 (50)
100 (50)
100 (50)
100 (50)
98 (50)
92 (50)
98 (300)
Individual faecal samples from identified calves
100 (25)
100 (24)
92 (24)
85 (27)
100 (20)
85 (20)
94 (140)
LY
Faecal pat samples from the low yield cow group
96 (50)
100 (50)
96 (50)
96 (50)
94 (50)
78 (50)
93 (300)
C
Faecal pat samples from the calf group
100 (30)
93 (30)
80 (30)
97 (30)
100 (29)
87 (23)
93 (172)
E
Environmental swabs (e.g calf pen run-off, yard, tractor, scraper
slats, farm driveway, bird droppings)
100 (7)
75 (8)
50 (8)
88 (8)
43 (7)
100 (8)
76 (46)
P
Pen swabs from group and individual pens, and calving box
83 (30)
67 (30)
47 (30)
67 (30)
77 (30)
83 (30)
71 (180)
Faecal pat samples from the dry cow group
65 (31)
65 (31)
75 (40)
83 (40)
55 (40)
55 (31)
67 (213)
Dust from calf shed
100 (4)
100 (4)
0 (4)
75 (4)
25 (4)
100 (4)
67 (24)
DW
Dirty water from drains
100 (2)
50 (2)
0 (2)
50 (2)
100 (2)
50 (2)
58 (12)
WC
Water course
0 (0)
0 (1)
0 (1)
100 (1)
0 (1)
0 (1)
20 (5)
Manure from main heap and undisturbed heap
0 (1)
33 (3)
33 (3)
0 (3)
0 (0)
0 (1)
18 (11)
100 (1)
0 (1)
0 (1)
0 (1)
0 (1)
0 (1)
17 (6)
HY
I
DC
D
M
WT
Communal water trough
1
Other results are summarised in the earlier summary.
Milestone 01/03: Mathematical modelling of plasmid transmission dynamics
Summary: In this section we propose a mathematical model specifically designed to model the within-animal
transmission of CTX-M genes between donor and recipient bacteria and to investigate the effect of a number of
rd
th
scenarios, in particular the use of 3 and 4 generation cephalosporins (hereafter referred to as the AMR scenario),
which kill off and inhibit the growth of bacteria that do not have the CTX-M gene. The model results suggest that the
use of cephalosporins could cause a large increase in CTX-M bearing E. coli, but once usage stops the levels in the
gut would return to normal after around 10 days. This could lead to significantly increased levels being excreted into
the environment, possibly increasing the risk of between-animal transmission of the CTX-M bearing E. coli on the
farm. To investigate this hypothesis, initial work was done on the construction of a between-animal transmission
model for CTX-M on a dairy farm, which used the within-animal model to estimate the number of CTX-M bearing E.
coli in the farm environment, due to shedding by infected animals. Due to time constraints and the complex dynamics
involved in transmission on the farm, particularly the differences between calves and adults, this work suffered from
significant data gaps and results were prone to high uncertainty and could not be validated at this time. However,
while we would caution against placing too much credence in the model results at this point, preliminary findings
rd
th
suggested that low 'reactive' usage of injectable antibiotics containing 3 and 4 generation cephalosporin's (data
from Defra project OD2031, currently puts this rate at about 18% per year) may not have much impact on the on-farm
prevalence of antimicrobial resistant bacteria. However, high, persistent usage on a regular basis (e.g. 5% per day)
could be of more concern. Further studies should be done to corroborate these findings, particularly with regards to
usage in calves.
Materials & Methods: The within-animal model incorporated transmission via conjugation as well as simulating
changes in the populations of bacteria in the gut due to ingestion events, growth, death and the effect of movement of
the donor bacteria through the gut (hereafter referred to as the 'decay rate'), modelled by estimating the time (in days)
for a 50% reduction of CTX-M bacteria populations, under bovine gut conditions. We parameterised the model based
on both the donor and recipient bacteria being strains of E. coli. The model parameters were estimated from
published literature, unpublished experimental studies and, where data were lacking, expert opinion. A full
description of the mathematical equations and parameter estimation can be found in Appendix A. Based on
comparing the deterministic model results with those of the experimental studies carried out in objective 04/01, we
extended the deterministic model to a stochastic one, where the inherent animal-level variation in the duration of
infection and concentration of bacteria during the latter stages of infection was incorporated. This is especially
2
important when levels excreted in faeces are generally below the limit of detection (10 cfu/g), but sporadic 'spikes' in
4
concentration of up to 10 cfu/g are observed, which we term 'increased shedding events'. Comparison of the
extended model results against these data suggested a good fit (see Appendix A). These types of dynamics are not
captured by deterministic models and while they arguably have little effect on asymptotic average values they could
play a pivotal role in modelling transmission of CTX-M bearing bacteria between hosts, as pockets of bacteria in the
environment that contain higher concentrations of CTX-M bearing E. coli could lead to higher initial doses and our
results suggest that higher initial doses can lead to longer periods of infection and higher levels excreted in the
faeces.
The between-animal model was a modified stochastic SIR (Susceptible-Infected-Recovered) model, which simulated
infection of animals on a dairy farm, primarily via ingestion of CTX-M bacteria from the environment (which are shed
by infected animals, as determined by the within-animal model). The model was parameterised based on published
literature, expert opinion and data collected as part of the longitudinal study (Milestone 01/01). A full description of the
mathematical equations and parameter estimation can be found in Appendix B.
The AMR scenario was developed based on discussions with experts at the AHVLA; it was assumed that
cephalosporins were given to the animal on day 1, this will cause a 90% reduction in E. coli that do not have the CTXM gene (i.e. the recipient bacteria) but not affect those with the CTX-M gene. It was further assumed that there is no
growth of the recipient bacteria during treatment and for 3 days afterwards and that treatment lasts one day. To
model the effect of cephalosporin usage in the between-animal model we used data collected as part of Defra project
rd
th
OD2031; on farms that used 3 or 4 generation cephalosporins, about 18% of animals were given injectable
antibiotics per year. No information was given on dosing regimes. When an animal was determined to be given
antibiotics, the within-animal AMR scenario was used; otherwise the baseline within-animal model was used.
Results: Figure 1 shows a comparison of the baseline model (i.e. no use of antibiotics) with the AMR scenario, where
it can be seen that CTX-M concentrations in the AMR scenario are considerably higher for about 10-15 days after
infection/ colonisation, but then return to similar levels to the baseline model.
Figure 1: Comparison of model predicted average concentrations (Log10) of bacteria in the bovine gut over time from
initial ingestion of CTX-M bearing E. coli, in the baseline model and AMR scenario.
A sensitivity analysis showed that the model results were insensitive to changes in the bacteria growth rates and
decay rates above 2.5 days. However a reduction in the decay rate to 1 day suggested that CTX-M bacteria would be
eradicated from the gut in about half the time; if this was reduced further to 0.5 days then the eradication time
reduced further. This effect was also seen in the AMR scenario, where there was also a more noticeable effect when
increasing the decay rate, with higher concentrations of CTX-M bacteria persisting for longer. Varying the
concentration during increased shedding events had a considerable effect on the concentrations of CTX-M over time,
with lower values reducing the overall average concentration of CTX-M bacteria and higher values increasing it
(although there was little effect in the AMR scenario until about 15 days, after the CTX-M concentrations started to
decrease). This suggests that increased shedding events may play an important role in transmission dynamics.
Preliminary results from the between-animal model suggested that under baseline conditions introduction of a CTX-M
infected animal could lead to subsequent, persistent infection of adults on the farm, but infection in most of these
2
animals would be at levels below the limits of detection by current tests (~10 cfu/g). The model results from an AMR
scenario, where animals on the farm had an 18% chance of being given injectable antibiotics per year (in which case
the AMR scenario from the within-animal model was used to model transmission), were similar to the baseline
scenario. However, a scenario where they were used on a more regular basis (5% chance of use per day), showed
an increase in prevalence both above and below the limit of detection.
It is hoped that this model can be refined in the future, using data from ongoing projects, such as OD2031, in order to
increase confidence in the results, particularly with regards to the transmission dynamics of calves. The within-animal
model could, assuming availability of data, be adapted for other bacteria, animal species and resistance genes.
OBJECTIVE 02:- Detailed molecular characterisation of plasmids for marker selection
Introduction. CTX-M ESBL plasmids from characterised E. coli solates from animals have been selected for
sequencing. The plasmids have been accurately size profiled using the S1 nuclease (Barton et al., 1985) method
which is best for large plasmids >40kbp) and the chloroform phenol approach described by Kado & Liu (1981) which
affords improved detection of small plasmids (<20 kbp). Profiling of the plasmids carrying the CTX-M gene has been
accomplished using both transconjugants or transformants and Southern blotting against a CTX-M gene. The CTX-M14 plasmids (n=15) from cattle are typically 70-95 kbp in size, however there are often other plasmids (up to 8)
carried by the isolates. Analysis of plasmids by Random Amplified Polymorphic DNA (RAPD) indicates a high degree
of homology for most CTX-M14 plasmids. A multiplex PCR for specific plasmid genes developed in VM2207 is being
applied to compare plasmid types. A preliminary panel (n=30) of CTX-M plasmids, various sequence and replicon
types, has been selected for 454 pyrosequencing at the VLA. This includes CTX-M1 IncI1 plasmids from poultry to
enable comparison with Dutch isolates which are now being found also in humans in the Netherlands, CTX-M14
plasmids from cattle and turkeys and CTX-M15 plasmids from cattle and humans. The plasmid from the index farm
(pCT) in Wales has been sequenced and annotated by the Sanger Center and this will be resequenced at the AHVLA
to validate our own in house methods.
Milestone 02/01 – Plasmid replicon typing and RFLP.
Conjugation rates. Prior to plasmid replicon typing, plasmids were transferred by conjugation to a Salmonella
Typhimurium recipient strain with cefotaxime selection. This was to select for CTX plasmids so that replicon typing
was performed on CTX-M plasmids and not on other plasmids if present. A one hour conjugation incubation time
period was chosen for all experiments, based on the preliminary study.
Table 2. Characteristics of cattle, chicken and turkey blaCTX-M E. coli isolates with respect to CTX-M sequence type
and plasmids.
Animal species
Characteristics of CTX-M strains from different animal species
Conjugation rate range (n)
Replicon types (n)
[mean / SD]
[% I1- and K]
9.09 x 10-8 to 1.22 x 10-4
A/C, F, FIA, FIB, I1- [23%], K
[5.44 x 10-5 / 1.18 x 10-4]
[10%], P, Y
(31)
(31)
CTX-M sequence (n)
types [% each type] b
1 [7.4%], 3 [0.5%], 14 [36%],
14B [11.3%], 15 b [37.5%], 20 [0.5%],
27 [0.5%], 28 [2.4%], 55 [1.5%], NT
[1.5%]
(202)
Cattle
Ra
70-9
Chickens
1 [73%], 3 [6.6%], 15 [17%], NT
[3.3%]
(27)
6.22 x 10-6 to 1.01 x 10-4
[1.00 x 10-4 / 1.08 x 10-4]
(20)
A/C, I1- [95%], P
(19)
90-
Turkeys
1 [36.5%], 14 [45.1 %], 15 [12.9%],
55 [1.0%] NT [5.3%] (93)
9.11 x 10-7 to 1.05 x 10-4 [6.77 x
10-5 / 1.61 x 10-4]
(32)
B/O, F, FIA, FIB, I1- [50%], K
[42%]
(30)
70-9
(n), number for a given animal species for a given test or characteristic. NT – Not tested. a - <70kb plasmid sizes not included. b – 23.2% of the
CTX-M 15 isolates were sequenced as CTX-M 15/28, but as CTX-M type 28 is not seen in the UK and further sequencing has shown such isolates
to be CTX-M 15, CTX-M 15/28 isolates are described as CTX-M 15 for convenience.
The CTX-M sequence type, conjugation rate, replicon type and range of plasmid sizes in isolates from cattle,
chickens and turkeys are presented in table 2. Conjugation experiments were performed on a subpanel of 31, 20 and
32 cattle, chicken and turkey isolates respectively. Of the cattle, chicken and turkey isolates, 60%, 75% and 64%
transferred their plasmids to recipient strains. However a number of isolates with CTX-M-1 (chicken, turkey), -3
(chicken), -14 (cattle, turkey) and -15 (cattle, chicken, turkey) sequence types were unable to conjugate using the
protocol described.
-8
-4
The relative conjugation frequencies where conjugation occurred, ranged from ~ 10 to ~ 10 for cattle isolates, from
-6
-4
-7
-4
~ 10 to 10 for chicken isolates and from 10 ~ to 10 for turkey isolates. Two turkey isolates which did not
conjugate against Salmonella Typhimurium (isolate 26R), did so with E.coli K12.
CTX genes, replicon types and plasmid sizes in isolates from different species. CTX-M sequence types in the 202 E.
coli from cattle were mainly those in CTX-M group 1 (47.3%) and group 9 (47.3%), of which the main sequence types
were CTX-M 15 (37.5%) and CTX-M 14 (36%) respectively (Tables 1 and 2). Less common CTX-M sequence types
in cattle isolates were 14B, 20, 27 and 55.
Table 3. Replicon types and plasmid sizes for specific CTX-M sequence types for cattle, chicken and turkey E. coli
CTX-M
Sequence
Type (n)
1 (32)
Number of isolates with replicon types
A/C
B/O
F
FIA
3
5
FIB
I1-
K
P
27
2
1
Y
Number of isolates with range of plasmid
sizes
70-90 kb 90-148
>148 kb
> one
kb
plasmid
1
31
3 (2)
14 (26)
15 (10)
20 (1)
55 (1)
2
1
9
1
10
10
1
2
1
3
5
5
16
2
1
12
3
14
7
1
1
1
Type 1 associated with I1-, 14 with incK and FIAor F and 15 with FIA.
2
RFLP. RFLP was not performed as the plasmid sequencing gave more information than could have been obtained by
RFLP
Milestone 02/02 – Plasmid sequencing. Several CTX-M plasmids with a variety of characteristics including different
CTX-M genes, different sizes, different sources and antimicrobial profiles were selected for sequencing and further
molecular analysis. To aid the downstream analysis and to ensure that only 1 plasmid was sequenced, whole DNA
containing plasmid DNA (pDNA) was extracted from the wild type strains and used to transform into a common
background (DH10b). Each transformant was confirmed by plasmid profile and PCR detection of CTX-M gene.
Different methods were employed to isolate purified pDNA from each transformant such as caesium chloride density
gradient centrifugation and several commercially available kits. However these methods were deemed unsuitable due
to issues with purity and integrity of the isolated pDNA. The large construct kit (Qiagen) was a superior method for the
isolation of pure pDNA of good integrity due to the inclusion of exonuclease step enabling selective and efficient
removal of gDNA as well as damaged plasmid DNA, thus reducing the levels of gDNA contamination to an
acceptable level and made the resulting purified large plasmid DNA suitable for 454 sequencing.
Plasmid sequencing was performed by 454 sequencing at the AHVLA central sequencing unit. Sequences were
assembled using Newbler software supplied by the 454 manufacturer. Due to the unknown composition and structure
of the plasmid sequences a reference plasmid sequence could not be used for assembly thus de novo assembly was
used, which generated basic information such as size of plasmids (range: 35-160kb) and the number of contigs
(range: 1-17) for each plasmid.
Milestone 02/03 – Plasmid bioinformatics. BLAST analysis was used to identify regions of similarity and difference
between the newly sequenced plasmids and those found in the publicly available NCBI database. There were
different degrees of homology (sequence coverage and identity) between the sequenced plasmids and those found in
the NCBI database (Table 4).
Table 4. Summary of Plasmid similarities (BLAST analysis)
AHVLA
strain
details
CTX-M
Group
ESBL111
G1
LR96
G9
TF2-53
G1
mN6S
G9
T91-2
G1
297
G9
T44-1
G1
ECR191
G1
TC2-47
G9
355
G1
ESBL428
G1
298
G2
ESBL488
G1
ESBL521
G1
426
G1
m356/LR3
39
360
G9
TC2-52
G1
mH19S
G9
LREC296
G1
LREC338
G1
LREC306
G9
356
G1
304
G9
ESBL292
G1
G9
Similar plasmids taken from international databases such as NCBI
(coverage/identity)
AM886293, pIP1206
(67/100)
AY458016, pC15-1a
(85/100)
AY509004, pSC138
(53/100)
FQ482074, pETN48
(65/100)
DQ364638, pNR1
(79/100)
EU219534, pOU7519
(53/100)
CP001118, pSL476_91
(64/100)
DQ017661, pNF1358
(67/100)
DQ017661, pNF1358
(81/100)
DQ017661, pNF1358
(88/100)
DQ115387, pOU1114
(73/100)
EU935740, pEK204
(64/100)
EU935740, pEK204
(77/100)
EU935740, pEK204
(87/100)
EU935740, pEK204
(94/100)
FJ876827, pKF3-140
(47/100)
FJ876827, pKF3-140
(88/100)
FN868832, pCT
(76/100)
FN868832, pCT
(98/100)
FN868832, pCT
(99/100)
GU256641, p3521
(68/100)
GU256641, p3521
(78/100)
GU371927,
pEC_Bactec (100/100)
GU371927,
pEC_Bactec (51/100)
GU371927,
pEC_Bactec (94/100)
HM355591, pHK01
(100/100)
JN087529, pHK08
(91/100)
JN232518, pHNDD81-1
(10/100)
AB021078, ColIb-P9
DNA (64/100)
CP001118, pSL476_91
(65/100)
CP001118, pSL476_91
(79/100)
CP001118, pSL476_91
(87/100)
EU219533, pSE34
(71/100)
DQ017661, pNF1358
(63/100)
DQ017661, pNF1358
(75/100)
CP001118, pSL476_91
(86/100)
AP011954, pR621a
(92/100)
EU935739, pEK499
(47/100)
AM886293, pIP1206
(62/100)
FJ386569, pO26-Vir
(76/100)
FJ386569, pO26-Vir
(84/99)
FJ386569, pO26-Vir
(86/99)
HE603110, pHUSEC411 (67/100)
FJ386569, pO26-Vir
(77/99)
EU935740, pEK204
(95/100)
JF274993, pSal8934a
(50/100)
EU935740, pEK204
(92/100)
FN594520, pEG356
(100/100)
HM355591, pHK01
(91/100)
JF267651, pSD_174
(8/100)
FJ876827, pKF3-140
(62/100)
AP000342, pR100
(79/100)
CP001125,
pCVM19633_110
(50/100)
DQ017661, pNF1358
(65/100)
AB021078, ColIb-P9
(65/100)
AB021078, ColIb-P9
(79/100)
AB021078, ColIb-P9
(87/100)
JF776874, pE001 (65/99)
GU371927, pEC_Bactec
(63/100)
(74/100)
DQ017661, pNF1358
(86/100)
(92/100)
FQ482074, pETN48
(45/100)
CU928148, p1ESCUM
(59/100)
GU256641, p3521
(74/100)
GU256641,
p3521(84/100)
GU256641, p3521
(85/100)
FJ386569, pO26-Vir
(60/99)
HE603110, pHUSEC41-1
(77/100)
AP011954, pR621a
(88/100)
AP011954, pR621a
(82/100)
JN087529, pHK08
(100/100)
FN594520, pEG356
(91/100)
Plasmid alignments using the software Mauve 2.3.1 (used to compare and align large sequences) clearly illustrated
the regions of sequence homology between newly sequenced plasmids and those found in the NCBI database. For
example, the plasmids from strains LREC296, LREC338 and LREC306 show high homology to plasmid pEC_Bactec
(GU371927). Although these plasmids share high sequence similarity their sizes vary, with the smallest being
LREC306 and the largest being LREC338 highlighting the changing nature of plasmid backbones (Figure 2).
Figure 2. Alignment of LREC296, LREC338 and LREC306 CTX-M plasmids to plasmid pEC_Bactec. The varying
colour blocks and lines between plasmids correspond to areas of sequence homology.
pEC_Bactec
LREC296
LREC338
LREC306
Further analysis of LREC338 shows it to be a combination of 2 previously described NCBI plasmids, pEC_Bactec and
pAPEC-02R (Figure 3) and is a good example of the recombination events that are possible between plasmids. The
pEC_Bactec was originally isolated from a horse, is from the IncI family of plasmids and only contains 2 resistance
genes (TEM and CTX-M15). In contrast, pAPEC-O2R was isolated from avian pathogenic E. coli, is from the IncFII
family of plasmids and contains multiple resistance genes excluding CTX-M (Johnson et al., 2005). Comparison of
the pMLST alleles shows that LREC338 contains 2 different copies of KlcA4, YfhA5 and DNAMeth6, illustrating the
likely recombination of 2 different plasmids.
Figure 3. Alignment of LREC338 to previously characterised plasmids to pEC_Bactec and C-O2-R. The varying
colour blocks and lines between plasmids correspond to areas of sequence homology.
pAPEC-O2-R
LREC338
pSal8934
pEC_Bactec
Milestone 02/04 – Plasmid transmission (See also Table 2 which gives conjugation rates). The ability of CTX-M
plasmids to transfer between strains was assessed by performing conjugation studies using a non-CTX-M containing
S. Typhimurium strain (S26) as the recipient. Eighty eight percent (37 out of 42) of the plasmids tested were
transferred into the Salmonella recipient by broth conjugation.
Five plasmids were not amenable to liquid conjugation, therefore these were tested for conjugation on solid surface.
Only one of these 5 plasmids was not transferred into Salmonella via solid media conjugation.
In summary, a total of 41 out of 42 plasmid strains had the ability to conjugate their plasmids into other bacteria, using
this as a mechanism to transfer resistant genes. Only 1 strain could not be transferred by conjugation.
Milestone 02/05 – Plasmid phenotypic characterisation - Effect of plasmid pCT on host metabolism (by Phenotype
Microarray)
The effect of plasmid pCT and various pCT mutant on metabolism were assayed using Biolog's phenotype microarray
(PM) system (Hayward, CA), allowing high throughput screening of bacterial respiration in response to a number of
compounds including metabolites and a variety of metabolic effectors and antimicrobials (Bochner et al., 2001). Four
panels were used including carbon (PM01-02), Nitrogen (PM03) and Phosphorus/Sulphur utilisation.
The presence of pCT in DH5a had mixed effects on respiration of several energy sources (carbon, nitrogen and
phosphorus). Presence of pCT in DH5a had a positive effect on respiration of L-Proline (carbon), N-Acetyl-DGlucosamine (nitrogen) and d-Amino-N-Valeric Acid (nitrogen) compared to DH5a without pCT. However pCT also
had negative effects on respiration of several energy sources including D-Galactose and a-Methyl-D-Galactoside
(carbon). The respiration in the presence of phosphorus containing substrates was compared between DH5α with
DH5αpCT. Significant differences (p<0.05) were observed for guanosine-2’ and 3’ monophosphates and cysteamineS-phosphate.
Mutations of genes associated with sigma factors (pCT3), Shufflon-specific DNA recombinase (pCT4) and Type IV
pilus major pilin subunit protein (pCT5) restored the ability of DH5a to utilise D-Galactose and a-Methyl-D-Galactoside
as carbon sources, L-cysteine as a nitrogen source, as well as D-Glucose-1-Phosphate and Cysteamine-SPhosphate as phosphorus sources. When pCT7 (traG: pilus assembly and DNA transfer) was also compared the
ability to utilise phosphorus was restored DH5αpCT7 and DH5α both respired with cysteamine-S-phosphorus but not
DH5αpCT.
Further studies investigated DH5α, C159/pCT cured (strains without plasmid), C159 cured + pCT (complemented)
and DH5α + pCT and C159 (both with pCT). Similar phenotypes were found regardless of the host strain but C159
cured + pCT did not revert to C159. Similar effects were observed with cysteamine-S-phosphate and inositol
hexaphosphate (results not shown). Other studies demonstrated that curing irreversibly changes the respiration with
substrates including dithiophosphate, thiophosphate, phosphor-glycolytic acid and D-mannose-6-phosophate. Many
of the effects of pCT curing were seen in phosphorus metabolism but were not fully resumed following reintroduction
of pCT, this is possibly an artefact of the curing process and has been observed in a previous curing studies of E. coli
O157 strains (Lim, 2010).
The negative effects of pCT on respiration of simple sugars such as D-galactose might potentially provide a simple
means to develop simple chromogenic screening media for it’s presence in bacteria. This was not investigated in the
project.
OBJECTIVE 3 – Production of CTX-M plasmid database
Due to the vast amount of sequence information generated, it was necessary to create a central database for storage
and searching of sequences. A FASTA file containing all the sequence information was constructed and an Excel file
that summarises the locations within the FASTA file of each plasmid sequence to allow easy retrieval of relevant
sequence data. A BLAST ESBL/CTX-M plasmid database containing plasmids identified during these studies as well
as plasmids identified with similarities that were found in the NCBI database was generated. This BLAST database
allows the user to rapidly and specifically search and identify plasmid DNA sequences of interest through a command
line prompt. The standalone BLAST (remote from the online NCBI BLAST) utility has been set up allowing users to
define a sequence query and BLAST a local database. Following full completion of AHVLA plasmid sequences, this
database will be added to a web interface to allow open access usage.
Using the BLAST 2 sequences algorithm, the FASTA file plasmid database can be queried using specific sequences
to generate BLAST output files containing coordinates of sequence matches, identity matches and sequence hits.
This database will allow users to identify common and unique plasmid sequences either within the database or in
future plasmid sequencing projects. The preliminary iterations of the database have enabled the identification of
plasmids markers as well as plasmid MLST alleles and aided development of the plasmid typing schemes described
in Objective 5.
OBJECTIVE 04 - Development and evaluation of an autogenous vaccine for E. coli bearing the CTX-M14 plasmid
Introduction. Vaccination has been widely trialled for the control of enterohaemorrhagic E. coli (EHEC) in cattle.
Most studies have focused on immunisation with bacterial proteins encoded by genes located within the locus of
enterocyte effacement (review see Svennerholm and Tobias, 2008). Various routes of administration for vaccination
of cattle have been assessed including subcutaneous (Potter et al., 2004), and intramuscular or rectal injection
(McNeilly et al., 2008). Intestinal immune response is of prime importance for protection against EHEC disease. A
commercial vaccine (Econiche) is now licensed in Canada for the control of VTEC E. coli O157. This vaccine is
based on secreted proteins which prevent the attachment of VTEC E. coli O157 to the GIT.
At a UK case farm the control of ESBL E. coli was attempted by administration of an autogenous vaccine to a
+
+
proportion of cows prior to calving prepared commercially from O33:K and O40:K CTX-M-14 E. coli. A subsequent
follow up visit by the AHVLA some months later suggested that there had been a significant (Chi= 40.85, p<0.0001)
reduction in the prevalence of CTX-M14 E. coli in milking cows from 57 (95% CI 42.4-68.8) to 3% (95% CI 0.4-11.5)
after vaccination. A reduction in the mortality of newborn calves, their scouring and of metritis in calving cows was
also noted. Clearly these data were interesting and may have indicated the efficacy of an autogenous vaccine
approach, although there was no control group for this intervention. Moreover many other factors at the farm also
changed after vaccination, in what was an uncontrolled on farm intervention.
The aim of the study in this part of OD2028 therefore was to investigate the protection provided by autogenous
vaccination to subsequent colonisation by homologous CTX-M-14 E. coli under carefully controlled conditions. If
successful, autogenous vaccination could be considered further to prevent further dissemination of CTX-M E. coli in
cattle.
Background. In all there were three experiments performed although the first two experiments were more
exploratory. For all studies the calves were initially screened before selection for absence of CTX-M E. coli in the
faeces using enrichment in buffered peptone water and ChromAgar CTX selective media (Randall et al., 2009).
10
Objective 04/01/01. In the first colonisation study calves (n=6; age 10 weeks) were dosed (1 x10 cfu; 3 calves per
strain) with either E. coli strain EC768/06 (O33:K+) or EC769/06 (O40:K+). Both of these strains carry a CTX-M14
gene and were recovered from a case farm in Wales and used to prepare and autogenous vaccine in 2006.
Interestingly, the EC769/06 strain was not recovered from the faeces after dosing and thus was not investigated
4
further. The EC768/06 strain however was recovered in faeces at levels of approximately 1 x 10 cfu/g faeces during
the 2 week study period after dosing. Pulse field gel electrophoresis was used to confirm that the dosed and
recovered strains were identical.
Objective 04/01/02. In the second study, colonisation of calves (n=10) with EC768/06 was maintained over a 10
4
week period with excretion rates of approximately 1 x 10 cfu/g faeces during that period.
Objective 04/01/02. An autogenous inactivated vaccine was prepared by Nationwide Laboratories (Leeds) from E.
coli EC768/06. Both control (carrier only) and vaccine (batch number ECOLI01) were supplied to the AHVLA. This
was then used in a study with 30 calves as below.
Summary of objective 04. Preliminary studies established a bovine colonisation and persistence model for E. coli
O33 bearing the blaCTX-M-14 plasmid (Horton et al., 2011). An autogenous vaccine study was then set up whereby
thirty 5-6 week old cross breed calves were randomly split into two groups, one receiving the vaccine and the other a
10
placebo intramuscularly. At 12 weeks of age all calves were orally challenged with 1 x 10 cfu of E. coli O33 and
faecal shedding monitored daily. Total E. coli and CTX-M E. coli shedding rates were determined using selective
media. Blood samples were taken throughout the study for immunological assays.
Results showed that the cumulative numbers of total E. coli excreted at 7 days was significantly lower in the
vaccinated group (p=0.006) however there was no significant difference observed between the vaccinated and
placebo calves in the number of days excretion of CTX-M E. coli or total E. coli. The systemic immune response to E.
coli O33 antigen was tested by ELISA throughout the study and the immune response was significantly higher
(p<0.001) in vaccinated calves. The mucosal immune response was also evaluated and showed no differences
between the vaccinated and unvaccinated calves.
Materials and methods for objective 04/02/02
Bacterial strains and preparation of inocula. CTX-M E. coli O33 strain EC768/06 was obtained from the Animal
Health and Veterinary Laboratories Agency (AHVLA; Weybridge United Kingdom) culture collection. This strain was
+
originally isolated from a farm with CTX-M E. coli (Liebana et al 2006) and was characterised as serotype O33:K with
a conjugative 65 MDa IncK plasmid bearing blaCTX-M14
In vivo calf study. All animals used in this study were screened for excretion of ESBL E. coli prior to any in vivo
studies by direct plating to CHROMagar CTX as described below. The efficacy of this vaccine was evaluated in two
groups of fifteen 5-6 week old cross breed calves. All animal studies were performed in accordance with the Animals
(Scientific Procedures) Act (1986) and were approved by the local Ethical Review Committee. After 1 week group one
received the prepared vaccine and group two a placebo (vehicle only) intramuscularly. Two further vaccinations (or
placebo) were administered to each group at two week intervals (McNeilly et al 2008).
10
Two weeks after the third vaccination (week 7), all calves were infected orally using an oral dosing gun with 1 x 10
cfu of the CTX-M positive E. coli challenge strain (EC768/06) suspended in 10 mls of PBS.
Rectal faecal samples were taken daily for the first 28 days post challenge to monitor the shedding of total E. coli and
CTX-M E. coli. After day 28, faecal samples were taken twice weekly for the remainder of the study at week 18.
Calves were blood sampled bi-weekly in order to evaluate the systemic immune response to the vaccine before,
during and after vaccination. Three animals from each group were randomly selected and euthanased at 8 and 12
weeks following primary vaccination for evaluation of challenge strain tissue tropism, all remaining animals were
euthanased at 18 weeks post primary vaccination.
Recovery and enumeration of bacteria from calves. The total E. coli (growth on CHROMagar ECC) and
presumptive CTX-M E. coli (growth on CHROMagar CTX) in faecal samples was determined. Faecal samples were
weighed to 1 g and re-suspended in 9 ml buffered peptone water (BPW) and ten-fold serial dilutions were plated
directly onto the respective agars. If no growth was observed from the direct plating of the faecal samples in PBS,
then cultures were enriched at 37C for 16-18 hours in buffered peptone water (BPW).
Post Mortem Examinations. At weeks 8 and 12 of the study (one and five weeks post infection) three calves from
both the vaccinated and placebo group were euthanased and detailed post mortem examinations were performed.
The tissues sampled were ileum, colon, caecum, rectum and anal-rectal junction. Samples were used for
microbiological assessment, and a section was fixed in 10 % (v/v) neutral buffered formalin for histology. The injection
site was removed and observed for pathology.
Serotyping. The somatic antigen of E. coli colonies from CHROMagar CTX was determined as previously described
(Horton et al, in-preparation)
CTX-M sequence type. A selection of presumptive CTX-M E. coli colonies from CHROMagar CTX were tested for
the presence of the blaCTX-M9 ESBL gene by PCR with sequencing primers as described previously (Sabate et al,
2002; Carattoli et al., 2008). The ~ 1 kb PCR products were analysed to determine the blaCTX-M gene sequence type
by using an ABI 3730 DNA Analyser (Applied Biosystems, Warrington, UK) and by use of the ABI SeqScape
program to determine CTX-M type against a reference library.
Characterisation of non-vaccine isolates. Isolates which grew on CHROMagar CTX and were therefore likely to
contain a CTX-M plasmid but were of a different serotype to the inoculum strain were tested for the presence of the
pCT plasmid using a multiplex PCR with primers for putative sigma factor as previously described (Cottell et al, 2011)
Pulse-field Gel Electrophoresis. A selection of presumptive CTX-M E. coli colonies from CHROMagar CTX were
typed by XbaI pulse-field gel electrophoresis (PFGE) according to the Pulse Net USA protocol for E. coli O157 (Ribot
et al 2006). Dendograms were produced using Bionumerics version 5.0.
Plasmid Profiling. A selection of E. coli colonies from CHROMagar CTX, of known serotype, PFGE profile and CTXM type were selected for plasmid profiling as previously described (Kado and Liu 1981).
ELISAs. Serum samples from the experimentally infected cattle before vaccination, after vaccination and after
challenge were analysed by ELISA for systemic immune response by measuring IgG and for mucosal immune
response by measuring IgA. A standard direct ELISA was used as previously described (Cawthraw et al, 1994).
Statistical Analysis. The generation of variables for comparison and analysis of statistical inference was performed
using STATA 10 software. The Wilcoxon rank sum test was used to examine the difference in median days of
excretion of CTX-M E. coli and total E. coli and a two sample T-test was used to compare the log10 cfu/ g faeces in
both the vaccinated and placebo groups. A lowess smoothing curve with scatter plot was produced to illustrate the
pattern of excretion in each group with time. A two tailed paired T-test was used to compare the immune responses
between the vaccinated and placebo groups over the course of the study.
Results
Shedding of bacteria and colonisation. The duration and level of excretion of CTX-M E. coli and total E coli, were
compared for vaccinated (n=15) and control (n=14) groups of calves, by analysis of certain summary measures
(Table 5). There was no difference in the number of days excretion of CTX-M E. coli or total E coli by either the
quantitative or qualitative culture method. There was no difference in quantifiable means such as: the maximum
excretion value per calf, the cumulative excretion value per calf, with the exception of the cumulative excretion of total
E coli at 7 days (p=0.006), but this was not supported by a statistically significant difference in the cumulative
excretion at 28 days for the surviving calves (p=0.3). There was therefore little or no evidence to support statistically
significant differences in excretion patterns between control and vaccinated groups.
Table 5. CTX-M E. coli Mean (SD) calf values for selected summary measures and the probability test results (p
value) for no difference between vaccinates and controls.
Summary
measure
Days
quantifiable
excretion
Days Pos
Enrichment
Maximum
excretion
(log cfu)
Total
Excretion
(log cfu)
Subgroup
Controls
n
Calves at
50 days
6.6 (3.93)
8
Calves at
50 days
All calves
at 7 days
35.3 (5.26)
Calves at
50 days
Vaccinates
n
P value
Test
11.3 (5.60)
9
0.06
WRS
8
35.7 (6.93)
9
0.8
WRS
5.4 (0.13)
14
5.1 (0.19)
15
0.3
t
20.1 (5.50)
8
35.2 (4.16)
9
0.05
t
WRS – Wilcoxon rank sum test; t – t-test.
Post mortem examinations were performed on two occasions throughout the study at weeks 8 and 12. The bacterial
counts showed no significant difference in colonisation rates between the two groups. Gross and histopathological
examination of the tissues collected at post mortem showed that there was no pathology associated with the
experimental E. coli infection in any of the tissues collected from any of the calves.
Characterisation of E. coli isolates recovered. A proportion of isolates recovered from faecal samples were
selected for O serotyping. Of these isolates selected 92% were established to be the same serotype as the O33 CTXM E. coli isolate used to infect the calves (EC768/06) and 13 isolates belonged to various other serotypes.
Please also see summary for other results.
Figure 4. Average daily faecal shedding rates of CTX-M E. coli O33 and total E. coli
1.0×10 6
5.2×10 5
CTX-M E. coli Vaccinated
CTX-M E. coli Control
Total E. coli Vaccinated
Total E. coli Control
Log2 cfu/ g Faeces
2.6×10 5
1.3×10 5
6.6×10 4
3.3×10 4
1.6×10 4
8.2×10 3
4.1×10 3
2.0×10 3
1.0×10 3
5.1×10 2
2.6×10 2
1.3×10 2
6.4×10 1
0
2
4
6
8
10
12
14
16
18
20
22
24
Days post infection with CTX-M-14 E. coli
Objective 5 – Development of a generic molecular toolbox for molecular typing of CTX-M plasmids.
Milestone 05/01 – Design and development of plasmid markers and milestone 05/02 – Use of plasmid
markers.
Plasmid biomarkers have currently been designed for 7 of the 9 individual plasmid sequences. The development of
biomarkers into a test format can be best shown with an example such as the plasmid CT07. The CT07 plasmid was
recovered from bovine E. coli and is an IncX plasmid which can currently not be typed using the standard PCR based
replicon typing (PBRT) scheme, it is 35,341bp in length and encodes the blaCTX-M-14 gene.
The plasmid is capable of both liquid and solid conjugation as determined using Salmonella Typhimurium 26R
recipient. After sequencing using the Roche 454, the CT07 plasmid was closed using specific primers to produce an
overlap on adjacent strands of at least 100bp. After the plasmid was closed it was put through BLASTn and RAST
annotation software for comparison with other plasmids/DNA and to provide automated annotation.
BLASTn searches of the CT07 plasmid showed homology to the sequenced plasmids pJIE143 from E. coli (Partridge
et al., 2011) with 93% similarity, pBS513_33 from Shigella boydii (CP001059.1) with 84% similarity and pCROD2
from Citrobacter rodentium (Petty et al., 2010) with 80% similarity. The plasmid pBS513_33 and pCROD2 lack any
known resistance genes, and pJIE143 carries the blaCTX-M-15. This gene is carried in the common transposition unit
ISEcp1-blaCTX-M-15-orf477 which is located between 1862-4832bp, downstream of the ‘pir’ replication initiation protein.
The blaCTX-M-14 gene in plasmid pCT07 is found in association with ISEcp1 which is 249 bp upstream, and the
downstream sequence is homologous with numerous insertion sequences (IS).
Comparison using MAUVE 2.3.1 software of plasmids pCT07, pJIE143, pBS512_33 and pCROD2 found conserved
and unique regions between plasmids. The regions identified are 1- Pir (replication initiation protein) which is found in
pCT07, pJIE143 and pBS512_33, 2- VirB5 (Pilx minor pilin) found in pCT07, pJIE143, pBS512_33 and pCROD2 and
3- HicA (toxin/antitoxin) present in pCT07, pJIE143, and pCROD2. These three regions together can be used to
identify the separate types of backbone whether it is similar to pJIE143, pBS512_33 or pCROD2, this is important as
currently a system to identify IncX plasmids is not in place. Two more regions have been identified which are unique
to the pCT07 plasmid 4- hypothetical gene and 5- S-Methylmethionine permease associated with the blaCTX-M-14, all of
the regions have been highlighted in figure 5.
Figure 5. Alignment of pCT07, pJIE143, pBS512_33 and pCROD2. The varying colour blocks and lines between
plasmids correspond to areas of sequence homology
PCR primers have been designed to detect the regions described and have been combined into two multiplex
reactions: one which will detect for the backbone of the plasmid and the other for the unique regions, as shown in
table 6.
Table 6. PCR primers to detect plasmid backbones and unique regions of backbones
Marker
1- Pir
2- VirB5
3- HicA
4- Hypothetical gene
5- S-Methylmethionine permease
Start
codon
822
20925
28212
10553
31316
Stop
codon
1595
21984
28561
11393
31786
Size
bp
774
1060
350
841
471
This PCR has been tested on pCT07 transformed into E. coli DH10b with cefotaxime selection and the transformants
have been shown to contain the plasmid as shown in figure 6. Current testing is being expanded to identify other
similar plasmids in field isolates.
Figure 6. Use of pCT07 markers to identify similar plasmids.
Biomarker sequences have also been identified for plasmids pCH01, pCH02, pCH03, pCT01, pCT06 and pT01 and
since CT01 is partially closed it is being used for biomarker design.
These plasmids all share homology as can been seen from the MAUVE representation in figure 7. These plasmids
are therefore ideal candidates for the development of biomarkers to allow the differentiation of such similar plasmids,
and as such stable genes or regions need to be identified. This has involved the closing these plasmids and
performing searches against NCBI using BLAST and RAST automated annotation.
Figure 7. Alignment and comparison of pCH01, pCH02, pCH03, pCT01, pCT06, pT01 and CT01 plasmids. The
varying colour blocks and lines between plasmids correspond to areas of sequence homology.
Plasmid specific biomarkers have been selected through the scoring of the BLAST searches, the less frequent a
marker the more suitable its use for biomarker selection, a list of the biomarkers is shown in Table 7.
Table 7. The strains, sources, CTX-M or ESBL type, size, replicon type and selected biomarkers with NCBI names
Strain
Animal/Source
Plasmid
CTX-M/ESBL
Replico
n Type
Size (bp)
Biomarker selection (names from
closest match to ncbi)
E. coli
Chicken
pCH01
CTX-M-3
IncAC
160,357
E. coli
Chicken
pCH02
CTX-M-1
IncI1-γ
75,805
E. coli
Chicken
pCH03
CTX-M-1
IncI1-γ
105,608
E. coli
Cattle
pCT06
TEM-1, CMY-2
IncI1-γ
91,709
E. coli
Cattle
pCT07
CTX-M-14
IncX
35,341
E. coli
Turkey
pT01
CTX-M-1
IncI1-γ
111,318
 AAC3-VI A1 protein
 Conserved hypothetical protein
 Hypothetical - Cytosine S methyl
transferase
 Hypothetical
 ISEcp1-CTX-M
 IS26
 Hypothetical
 Conserved hypothetical protein
 traX
 ISEcp1-CTX-M
 Hypothetical
 Hypothetical
 Hypothetical associated with CTXM
 pilJ-traD
 Protease SepA
 Hypothetical
 repZ
 parM-stB
 Hypothetical in cytoplasm
 ardA-Gp4
 Hypothetical protein
 finQ
 pir
 pilx5
 hicA
 Hypothetical
 S-Methylmethionine permease
(CTX-M)
 Hypothetical protein
 ltrA putative RNA direct DNA
polymerase / Rev transcriptase
 StbA
 IbfA
 Int1
 IS26 mobile element
.
These biomarkers have been screened against E. coli DH10b (the cell line used for transformations), without any
plasmids and with each of the sequenced plasmids. The results on the biomarker usages are shown in table 7below.
Green (0) shows the absence of a marker in that test, red (1) shows the presence of a marker in that test and purple
(1) highlights the plasmids for which the markers were designed; it can been seen from using a combination of these
markers that all plasmids can be differentiated (Table 8).
Table 8. Screening of biomarkers against the AHVLA sequenced plasmids, green (1) indicates the absence of the
biomarker, red (1) shows the presence, and purple (1) highlights the plasmid for which these biomarkers were
designed.
Biomarker
CH01B1
CH01B2
CH01B3
CH01B4
CH01B5
CH02B1
CH02B2
CH02B4
CH02B5
CH02B6
CH03A1
CH03A2
CH03A3
CH03A4
CH03A5
CH03A6
CT01F1
CT01F2
CT01F3
CT01F4
CT01F5
CT01F6
CT01F7
CT06A1
CT06A2
CT06A3
CT06A4
CT06A5
CT07B1
CT07B2
CT07B3
CT07B4
CT07B5
T01C1
T01C2
T01C3
T01C4
T01C5
T01C6
DH01
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
CH01
1
1
1
1
1
0
0
0
0
1
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
CH02
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
0
0
1
1
0
1
1
0
0
0
0
0
0
0
0
0
1
0
1
1
0
1
CH03
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
0
0
1
1
0
1
1
0
0
1
0
0
0
0
0
0
1
0
0
1
0
1
P01
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
0
0
1
1
0
1
1
0
0
1
0
0
0
0
0
0
1
0
1
1
1
1
Plasmids Screened
CB CT01 CT05 CT06
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
1
0
1
0
0
0
1
0
0
0
1
0
0
1
1
0
0
0
1
0
0
0
1
0
0
1
1
0
0
0
1
0
0
0
1
0
1
0
1
0
1
0
1
0
0
0
1
0
0
1
1
0
0
0
1
0
0
0
1
0
0
0
1
0
0
0
1
1
1
0
1
1
1
0
0
1
1
0
0
0
1
0
1
1
1
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
1
1
0
0
0
1
0
0
0
1
0
0
1
1
0
0
0
1
0
1
CT07
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
1
0
0
0
0
0
0
T01
0
0
0
0
0
1
1
1
1
1
1
1
1
1
1
1
1
0
1
1
1
1
1
1
0
0
1
0
0
0
0
0
0
1
1
1
1
1
1
T05
0
0
0
0
0
0
0
0
0
1
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
T06
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
0
0
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
TOTAL
1
1
1
1
1
6
5
5
5
8
5
5
9
5
6
6
5
2
3
5
5
2
7
7
2
1
6
1
1
1
1
1
1
5
3
4
5
5
6
CH = chicken, CT = cattle, T = turkey and P = pig, for biomarkers column the animal is denoted as described followed by a letter
indicating the contig and a numerical value for the biomarker. The total column shows the occurrence of that biomarker across the
sequenced plasmids tested.
The next phase of the biomarker development will be to refine the design of the markers to incorporate them into a
multiplex as with pCT07.
Current plasmid classification schemes are limited to (1) separating the major 18 different incompatibility groups using
a PCR-based replicon typing scheme (Caratolli et al., 2005) (however this has limitations as classification is based on
plasmids belonging to known incompatibility (Inc) groups and may thus fail if the replicons have diverged or are novel
(Carattoli et al., 2008), and (2) plasmid MLST typing schemes which are restricted to plasmids of only 4
incompatibility groups/replicon types (IncI1, IncH12, IncF, IncN) using between 2 and 5 commonly found plasmid
alleles (Plasmid MLST, 2012). Thus a plasmid typing scheme which distinguishes between the many different types
of ESBL plasmids is required.
The pMLST typing scheme utilises point mutation differences in specific sequences between the plasmids to
differentiate between the plasmid types. Thus sequences used to develop the pMLST sequences must contain a
degree of variation to enable good representation in the typing scheme. Using an initial library of 15 plasmids from the
AHVLA collection from different sources, CTX types, Inc types and of various sizes: 35-160kb, 6 alleles showing a
degree of sequence variations were selected for inclusion in the plasmid MLST scheme. The alleles were also
chosen on the basis that they were present in the majority of the plasmids initially sequenced. The selection of these
alleles has proven useful in classifying ESBL plasmids from the AHVLA collection and is applicable to plasmids in the
National Centre for Biotechnology Information (NCBI) database (Figures 8 and 9).
Figure 8. Distribution of pMLST alleles in 270 plasmids, including those possessing CTX-M gene. Green indicates
presence of allele and red indicates absence of allele.
Figure 9. Diversity and evolutionary relatedness of pMLST alleles. CTX-M group 1 plasmids are highlighted in red
and CTX-M group 9 plasmids are highlighted in blue.
PsiAB2
KlcA4
YfhA5
DNAMeth6
Several plasmid sequence types were shown to be specific to plasmids from specific sources, such as ST06 which
appears to be specific to ESBL plasmids found in pig isolates. However, as the database is populated the number
and type of plasmids within an ST type may change. The pMLST scheme developed also identified and grouped 92
non-CTX plasmids (34%). Only 6% of plasmids (8 CTX plasmids and another 8 ESBL) were not amenable to typing
using the pMLST loci selected. In order to further develop the pMLST scheme and include these untypable plasmids,
these plasmids were selected for complete sequencing to enable identification of new loci that may be incorporated
into future iterations of the pMLST scheme.
In conclusion development of biomarkers and pMLST typing scheme has enabled the identification and differentiation
of CTX-M and ESBL bearing plasmids. A multiplex PCR has been developed for the small untypable CTX-M-14
plasmid pCT07 after refinement of the biomarkers, which is now the next step for the analysis of other plasmids and
their biomarkers. These approaches can be used to identify similarities in plasmids, map their epidemiological spread
and monitor interventions which may reduce their spread and clinical impact. Additionally, these approaches will
permit targeted approaches to interrogate gene functions specific to identifying those genes that contribute to
potential fitness and virulence and the identification of ESBL plasmid biomarkers to further investigate the
epidemiology of antimicrobial resistance.
DISCUSSION
Bacteria with resistance to extented spectrum antibiotics (ESBLs), in particular the CTX-M class of ESBLs continue to
pose a threat to human health and remain a topic of international concern.
In recent years CTX-M-producing E. coli have been widely isolated from animals (Brinas et al., 2003; Jouini et al.,
2007; Machado et al., 2008; Randall et al, 2011; Shiraki et al., 2004; Smet et al., 2008; Yuan et al., 2009). In the UK
specifically, recent surveys have shown that cattle, chickens, turkeys and waste milk fed to cattle all harbour CTX-M
ESBLs with a predominance of types 1, 14, 14B and 15, although other types also occur (Horton et al., 2011; Randall
et al., 2011).
Chicken meat in the UK has been shown to be a source of E. coli with CTX-M-1 (Warren et al., 2008) whilst imported
chicken meat has been shown a source of E. coli with CTX-M group 2 and 8 ESBLs (Dhanji et al., 2010). Whilst at
present evidence suggests that in the UK CTX-M types from animals and meat are not those predominantly causing
disease in humans, in the Netherlands there is evidence of a high percentage of chicken meat samples having the
same CTX-M type as seen in human patients (Leverstein-van Hall et al., 2011) .
This study investigated the presence and spread of ESBLs in a dairy farm, the possible control of ESBLs by
vaccination and developed an pMLST /molecular marker scheme to enable the movement of ESBL plasmids to be
traced.
The first part of the study showed that faecal samples, pats and swabs from calves, cattle and parts (excluding dust,
water, and manure) of an ESBL E. coli positive farm investigated in a longitudinal study were positive for CTX-M E.
coli, with mean values for positive samples for all six visits for these samples of 67% to 98%. Approximately 48% of
6
calves shedding CTX-M E. coli in faeces at levels greater than 1x10 cfu/g. This suggests some ongoing selective
pressure on the farm for the persistence of ESBL E. coli and further studies as part of OD2031 may shed some light
on this.
With respect to the vaccination study, this showed that calves can excrete high levels of CTX-M E. coli in faeces for at
least 3 weeks post “infection”. Vaccination of calves with a crude autogenous vaccine, whilst leading to increased IgG
levels and decreased levels of total E. coli at one time point did not affect the numbers of CTX-M E. coli in faeces or
have an effect on plasmid transfer. The reduction in the prevalence of CTX-M-14 E. coli in milking cows on the farm
where the original autogenous vaccine was used (Liebana et al., 2006) may have been due to other factors which
changed during the same period such as feeding, movements of animals, reduced stocking density and use of other
antibiotics rather than the use of the vaccine alone. However, the increased IgG levels and decreased levels of total
E. coli suggest that further work could be worthwhile, perhaps to investigate intestinal immunity and administration of
colostrum from vaccinated dams.
Plasmids from previous studies at AHVLA were characterised by applying a range of molecular and phenotypic
techniques and further techniques were developed to give sufficient resolution to investigate the movement of
plasmids on farms and through the food chain.
Current plasmid typing schemes are limited to incompatibility group (n=18) determination using a PCR based replicon
typing scheme developed some years ago (Caratolli et al., 2005). Multi locus sequence typing is available for 4 of
these incompatibility groups (Plasmid MLST, 2012). In the work carried out in this project, a generic multi-locus
sequence typing scheme was developed for ESBL plasmids and combined with a high resolution PCR test for the
presence of certain genes. The pMLST system was based on the sequencing of 6 genes (parB, psiAB, tnpA, klcA,
yfhA and dnameth) common to most ESBL plasmids. For more detailed high resolution discrimination between
plasmids of a particular type, such as IncI-1γ, PCRs for the detection of unique molecular markers (typically five to
six) were developed. This new scheme was applied to 270 ESBL and non-ESBL plasmids. The pMLST data was
presented as phylogenetic trees for each locus and revealed grouping for plasmids bearing a CTX-M ESBL of a
certain group (e.g. 1 or 9). Further PCR analysis for molecular markers has provided further discrimination. This
plasmid toolbox could be potentially very useful in for example in determining whether the plasmids seen chickens
from the Netherlands were the same as those seen in humans (Leverstein-van Hall et al., 2011).
At present, suspect ESBLs from regional laboratories are checked for the presence of CTX-M, OXA, SHV and TEM
genes and if CTX-M positive, the CTX-M gene is sequenced. Isolates of interest are then tested for plasmid size and
for replicon type using the method of Caratolli (2005). This involves first transferring the plasmid of interest to a
recipient cell by conjugation or transformation, and for plasmid profiling, an extraction procedure followed by gel
separation, visualisation and sizing. For the replicon typing, a total of 18 PCR reactions have to be performed for
each isolate, and even though most of these are in multiplex reactions, it is still a time consuming and labour
intensive process. The plasmid MLST scheme and multiplex PCRs developed in this project provide a higher degree
of discrimination between plasmids for an equivalent or reduced effort.
In conclusion, all aspects of this project have been successfully addressed; the findings are summarised in the
executive summary. The work has led to seven posters at high profile scientific meetings, (one of which has not yet
been accepted), four papers in preparation and a further three proposed papers.
Possible future work leading on from this project is described below.
Possible future work:1. As part of work for OD2031, determine if waste milk fed to calves on the farm sampled as part of this
study contains cephalosporin antibiotics and investigate possible mitigation procedures. (This farm is
a suitable candidate for such study as the monitoring previously completed forms a basis for further
investigation).
2. Apply the multiplex PCR developed in this project to identify plasmid markers on all CTX-M ESBLs
recovered from routine scanning and other surveillance programmes.
3. Implement the plasmid MLST scheme developed in this project to further characterise CTX-M
plasmids for all CTX-M ESBLs isolated from scanning surveillance at regional laboratories.
4. Consider further vaccination studies with other types of vaccines, differing protocols and or with
different adjuvant.
References to published material
9.
This section should be used to record links (hypertext links where possible) or references to other
published material generated by, or relating to this project.
Posters published or submitted as part of this study
Manal AbuOun, Matthew Stokes, Hannah Preedy, Matthew Hayward, Guanghui Wu, Roberto La Ragione,
Nick Coldham & Martin Woodward. CTX-M Plasmid Sequencing (Part 1): Development of a Plasmid Multi
Locus Sequence Typing Scheme. ECCMID 2012 London.
Mark Arnold, Robin Simons, Lucy Snow, Nick Coldham, (2010). Mathematical modelling of plasmid
transmission dynamics. Maths and Molecules session at the AHVLA conference.
H.E. Reeves, S.B. Lotz, E. Kennedy, M.J. Woodward, N.G. Coldham and R.M. La Ragione. 2011.
Evaluation of an autogenous vaccine against Escherichia coli bearing the CTX-M14 plasmid. Poster at
AHVLA conference.
Brunton, L. A., Snow L.C., Warner R.G., Wearing H., Coldham N.G. Epidemiological investigation of the
ESBL CTX-M-14 gene in E. coli and other gut commensals in calves and its persistence in different farm
environments. RSHTM Research in Progress meeting 2010.
.
Horton RA, Wearing H, Randall LP, Brunton L, and Coldham NG Persistence, location and shedding
th
density of E. coli carrying CTX-M-14 on a UK dairy farm. 4 symposium on antimicrobial resistance in
th
th
animals and the environment (ARAE 2011), held in Tours, France (27 – 29 June 2011).
Matthew Stokes, Manal AbuOun, Hannah Preedy, Luke Randall, Guanghui Wu, Irene Freire Martin,
Roberto La Ragione, Mark D Fielder, Martin Woodward and Nick Coldham. ECCMID 2012 London.
Simons, R., Arnold, M., Randall, L., Coldham, N. (2012). A mathematical model to investigate the
transmission dynamics of CTX-M genes between E. coli in the bovine gut. ISVEE 2012. Maastricht, The
Netherlands (Oral presentation or poster or nothing).
Papers in preparation as part of this study
.
H. E. Reeves, S. B. Lotz, E. Kennedy, L. P. Randall, N. G. Coldham and R. M. La Ragione Evaluation of
an autogenous vaccine against Escherichia coli bearing the CTX-M-14 plasmid. In preparation for
Veterinary Microbiology.
Robert A. Horton, Heather Wearing, Luke P. Randall, Sam Chappell, Lucy A. Brunton, Richard Warner,
Daisy Duncan, Chris Teale and Nick G. Coldham. Persistence and environmental distribution of CTX-M E.
coli strain types on a UK dairy farm. In-preparation for JAC.
Simons, R. R. L., Randall, L. P., Coldham, N. G., Arnold, M. A model to investigate the transmission
dynamics of CTX-M genes between E. Coli in the bovine gut. In Preparation, may’be epidemiology and
infection.
M. O. Stokes, S Umur, G. Wu, H Preedy, M. D. Fielder N. G. Coldham and M AbuOun. Complete
Sequencing of pir IncX1a ISEcp1-blaCTX-M-14 plasmids from Escherichia coli ST10 and Enterobacter
cloaceae. In-preparation for AEM or JAC or AAC.
Proposed papers
Manal AbuOun, Matthew Stokes, Hannah Preedy, Matthew Hayward, Guanghui Wu, Roberto La Ragione,
Nick Coldham & Martin Woodward. CTX-M Plasmid Sequencing (Part 1): Development of a Plasmid Multi
Locus Sequence Typing Scheme. Proposed for high impact Journal - genetics based or AAC.
Stokes et al. Comparative genomics of veterinary and human IncI1y plasmids. Proposed for high impact
Journal - genetics based or AAC.
Stokes et al. Comparative genomics of veterinary and human A/C plasmids. Proposed for high impact
Journal - genetics based or AAC
General references
Barton, B. M., Harding, G. P. & Zuccarelli, A. J. (1995). A general method for detecting and sizing large
plasmids. Anal Biochem 226, 235–240.
Bochner BR, Gadzinski P, Panomitros E. Phenotype microarrays for high-throughput phenotypic testing
and assay of gene function. Genome Res. 2001 Jul;11(7):1246-55.
Bonnet R. Growing group of extended-spectrum B-lactamases: the CTX-M enzymes. Antimicrob Agents
Chemother 2004; 48: 1-14.
Brinas L, Moreno MA, Zarazaga M, Porrero C, Saenz Y, Garcia M, et al. Detection of CMY-2, CTX-M-14,
and SHV-12 beta-lactamases in Escherichia coli fecal-sample isolates from healthy chickens. Antimicrob
Agents Chemother 2003 Jun;47(6):2056-8.
Briñas L, Moreno MA, Teshager T et al. Monitoring and characterization of extended-spectrum betalactamases in Escherichia coli strains from healthy and sick animals in Spain in 2003. Antimicrob Agents
Chemother 2005; 49: 1262-4.
Cantón R, Novais A, Valverde A et al. Prevalence and spread of extended-spectru
producing Enterobacteriaceae in Europe. Clin Microbiol Infect 2008; 14: 144-53.
-lactamase-
Carattoli A, Bertini A, Villa L, Falbo V, Hopkins KL, Threlfall EJ. Identification of plasmids by PCR-based
replicon typing. J Microbiol Methods. Dec;63(3):219-28. Epub 2005 Jun 2. 2005.
Carattoli, A., Garcia- Fernandez, A., Varesi, P., Fortini, D., Geradi, S., Penni, A., Mancini, C., Giordano, A.
(2008) Molecular epidemiology of Escherichia coli producing extended- spectrum beta- lactamases
isolated in Rome, Italy. J. Clin Microbiol 46 103-108.
Coque TM, Baquero F, Canton R. Increasing prevalence of ESBL-producing Enterobacteriaceae in
Europe. Euro Surveill. 13 pii=19044. Available online from the Eurosurveillance Web site,
http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=19044.
Dhanji H, Murphy NM, Doumith M, Durmus S, Lee SS, Hope R, et al. Cephalosporin resistance
mechanisms in Escherichia coli isolated from raw chicken imported into the UK. J Antimicrob Chemother
2010 Dec;65(12):2534-7.
Duan RS, Sit TH, Wong SS et al. Escherichia coli producing CTX-M B-lactamases in food animals in Hong
Kong. Microb Drug Resist 2006; 12: 145-8.
Guerra B, Avsaroglu MD, Junker E et al. Detection and characterisation of ESBLs in German Escherichia
coli, isolated from animals, foods, and human origin between 2001-2006. Int J Antimicrob Agents 2012;
29: S270-S271.
Horton RA, Randall LP, Snary EL et al. Fecal carriage and shedding density of CTX-M extended-spectrum
B-lactamase-producing Escherichia coli in cattle, chickens, and pigs: implications for environmental
contamination and food production. Appl Environ Microbiol 2011; 77: 3715-9.
Leverstein-van Hall MA, Dierikx CM, Cohen SJ, Voets GM, van den Munckhof MP, van EssenZandbergen A, et al. Dutch patients, retail chicken meat and poultry share the same ESBL genes,
plasmids and strains. Clin Microbiol Infect 2011 Jun;17(6):873-80.
Johnson,T.J., Siek,K.E., Johnson,S.J. and Nolan,L.K. DNA sequence and comparative genomics of
pAPEC-O2-R, an avian pathogenic Escherichia coli transmissible R plasmid. Antimicrob. Agents
Chemother. 49 (11), 4681-4688 (2005)
Jouini A, Vinue L, Slama KB, Saenz Y, Klibi N, Hammami S, et al. Characterization of CTX-M and SHV
extended-spectrum beta-lactamases and associated resistance genes in Escherichia coli strains of food
samples in Tunisia. J Antimicrob Chemother 2007 Nov;60(5):1137-41.
Liebana E, Batchelor M, Hopkins KL et al. Longitudinal farm study of extended-spectrum B-lactamasemediated resistance. J Clin Microbiol 2006; 44 : 1630-4.
Lim Ji Youn, Hyun Joon La, Haiqing Sheng, Larry J. Forney and Carolyn J. Hovde. Influence of Plasmid
pO157 on Escherichia coli O157:H7 Sakai Biofilm Formation. Environ. Microbiol. 2010, 76(3):963.
Linton AH, Howe K, Sojka WJ et al. A note on the range of Escherichia coli O-serotypes causing clinical
bovine mastitis and their antibiotic resistance spectra. J Appl Bacteriol 1979; 46: 585-90.
Livermore DM. Defining an extended-
-lactamase. Clin Microbiol Infect 2008; 14 Suppl 1: 3-10.
Machado E, Coque TM, Canton R, Sousa JC, Peixe L. Antibiotic resistance integrons and extendedspectrum beta-lactamases among Enterobacteriaceae isolates recovered from chickens and swine in
Portugal. J Antimicrob Chemother 2008 Aug;62(2):296-302.
McNeilly, T.N., Naylor S.W., Mahajan, A., Mitchell, M.C., McAteer, S., Deane, D., Smith, D.G.E., Low,
J.C., Gally, D.L., Huntley, J.F. (2008) Escherichia coli O157:H7 colonisation in cattle following systemic
and mucosal immunization with purified H7 flagellin. Infect. and Immun. 76 2594-2602.
Mellmann A, Cloud J, Maier T et al. Evaluation of matrix-assisted laser desorption ionization-time-of-flight
mass spectrometry in comparison to 16S rRNA gene sequencing for species identification of
nonfermenting bacteria. J Clin Microbiol 2008; 46 : 1946-54.
Meunier D, Jouy E, Lazizzera C et al. CTX-M-1- and CTX-M-15-lactamases in clinical Escherichia
coli isolates recovered from food-producing animals in France. Int J Antimicrob Agents 2006; 28: 402-7.
Miles AA, Misra SS. The estimation of the bactericidal power of the blood. J Hyg 1938; 38 : 732-48.
Partridge SR, Ellem JA, Tetu SG, Zong Z, Paulsen IT, Iredell JR. Complete sequence of pJIE143, a pirtype plasmid carrying ISEcp1-blaCTX-M-15 from an Escherichia coli ST131 isolate.
Antimicrob Agents Chemother. 2011 Dec;55(12):5933-5.
Perez F, Endimiani A, Hujer KM et al. The continuing challenge of ESBLs. Curr Opin Pharmacol 2007; 7:
459-69.
Petty NK, Bulgin R, Crepin VF, Cerdeño-Tárraga AM, Schroeder GN, Quail MA, Lennard N, Corton C,
Barron A, Clark L, Toribio AL, Parkhill J, Dougan G, Frankel G, Thomson NR. The Citrobacter rodentium
genome sequence reveals convergent evolution with human pathogenic Escherichia coli. J Bacteriol. 2010
Jan;192(2):525-38.
Plasmid MLST 2012, http://pubmlst.org/plasmid/. Last accessed 20.03.2012.
Potter, A.A., Klashinsky, S., Li, Y., Frey, E., Townsend, H., Regan, D., Erickson, G., Hinkley, S.,
Klopfenstein, T., Moxley, R.A., Smith, D.R., Finlay, B.B. (2004) Decreased shedding of Escherichia coli
O157:H7 by cattle following vaccination with type III secreted proteins. Vaccine 22 362-369.
Randall LP, Kirchner M, Teale CJ et al. Evaluation of CHROMagar CTX, a novel medium for isolating
CTX-M-ESBL-positive Enterobacteriaceae while inhibiting AmpC-producing strains. J Antimicrob
Chemother 2009; 63: 302-8.
Randall LP, Clouting C, Horton RA, Coldham NG, Wu G, Clifton-Hadley FA, et al. Prevalence of
Escherichia coli carrying extended-spectrum beta-lactamases (CTX-M and TEM-52) from broiler chickens
and turkeys in Great Britain between 2006 and 2009. J Antimicrob Chemother 2011 Jan;66(1):86-95.
Ribot EM, Fair MA, Gautom R et al. Standardization of pulsed-field gel electrophoresis protocols for the
subtyping of Escherichia coli O157:H7, Salmonella, and Shigella for PulseNet. Foodborne Pathog Dis
2006; 3: 59-67.
Rodríguez-Baño J, Pascual A. Clinical significance of extendedInfect Ther 2008; 6: 671-83.
-lactamases. Expert Rev Anti
Sabaté, M., Navarro, F., Miró, E., Campoy, S., Mirelis, B., Barbé, J., Prats, G. (2002) Novel complex Sul1type integron in Escherichia coli carrying bla(CTX-M-9) Antimicrob Agents Chemother 46 4656-61
Shiraki Y, Shibata N, Doi Y et al. Escherichia coli producing CTX-M-2 beta-lactamase in cattle, Japan.
Emerg Infect Dis 2004; 10: 69-75.
Simpson EH. Measurement of Diversity. Nature 1949; 163: 688.
Smet A, Martel A, Persoons D, Dewulf J, Heyndrickx M, Catry B, et al. Diversity of extended-spectrum
beta-lactamases and class C beta-lactamases among cloacal Escherichia coli isolates in Belgian broiler
farms. Antimicrob Agents Chemother 2008 Apr;52(4):1238-43.
Smet,A., Van Nieuwerburgh,F., Vandekerckhove,T.T., Martel,A., Deforce,D., Butaye,P. and
Haesebrouck,F. Complete nucleotide sequence of CTX-M-15-plasmids from clinical Escherichia coli
isolates: insertional events of transposons and insertion sequences PLoS ONE 5 (6), E11202 (2010).
Snow LC, Wearing H, Stephenson B et al. Investigation of the presence of ESBL-producing Escherichia
coli in the North Wales and West Midlands areas of the UK in 2007 to 2008 using scanning surveillance.
Vet Rec 2011; 169: 656.
Svennerholm, A-M., Tobias, J. (2008) Vaccines against enterotoxigenic Escherichia coli. Expert Rev.
Vaccines 7 795-804.
Teale CJ, Barker L, Foster AP et al. Extended-spectrum beta-lactamase detected in E coli recovered from
calves in Wales. Vet Rec 2005; 156: 186-7.
Watson E, Jeckel S, Snow L et al. Epidemiology of extended spectrum beta-lactamase E. coli (CTX-M-15)
on a commercial dairy farm. Vet Microbiol 2012; 154: 339-46.
Yuan L, Liu JH, Hu GZ, Pan YS, Liu ZM, Mo J, et al. Molecular characterization of extended-spectrum
beta-lactamase-producing Escherichia coli isolates from chickens in Henan Province, China. J Med
Microbiol 2009 Nov;58(Pt 11):1449-53.

OD2028 Final Report - Defra Science Search

Transcription

Similar documents

ESBLs: From Soil to Suppuration

BioFire Diagnostics FilmArray® Documents

Donairs linked to illnesses

Our Multi-Hurdle Food Safety Program

Staphylococcus epidermidis and Escherichia coli Using Various

Cloning Of crylA(b) Gene Into pCAMBIA 1301

¥→→ General characters

3M™ Petrifilm™ Select E.coli Count Plate

E. coli - iGEM 2011