Environmental reservoirs of antibiotic resistance genes

Antibiotic resistance in the environment: soil, sediments, water bodies Environment acts as an reservoir for antibiotic resistance genes: -associated with antibiotic biosynthesis clusters - in closely related non-producers - in unrelated non-producers indigenous soil bacteria - in unrelated non-producers exotic bacteria = pathogens/commensals added to soil •Potential for selection of resistance -pollution •HGT of resistance genes- mobilome •Pathogens can survive in soil

-Acquire integrons/plasmids -Act as source of antibiotic resistance

Flow of antibiotic resistance genes, antibiotics and pathogens in the environment

Hospital Soil

People

Water Food

Clinical

antibiotics

Antibiotics produced

by soil bacteria

Veterinary

antibiotics

Farm

animals

Pollutants

Disinfectants

Surfactants

Studying the environmental gene pool: Antibiotic resistance

• Cultivation from environmental samples: Isolation and PCR • Metagenomics- expression screening

• Gene capture: mobilomes

•Integrons -gene cassettes- gene capture •Transposons •Plasmids- exogenous isolation – host capture •Phage

Studying the environmental gene pool: Antibiotic resistance

Resistance gene expression can vary and up regulation commonly occurs

Co-selection results from linkage of several resistance

genes, resistance to pollutants, heavy metals, disinfectants, detergents

Horizontal gene transfer, adaptive genes form part of the mobilome:

•Integrons -gene cassettes- gene capture •Transposons •Plasmids •Phage

No

n-p

roduce

rs

Str

epto

myci

n

No

n-p

roduce

rs

Gen

tam

icin

N

on

-pro

duce

rs

Tet

racy

clin

e

aac(3)-I

aac(3)-II/VI

aac(3)-III/IV

aac(6’)-II/Ib

ant(2”)-I

aph(2”)-I

aph3

aph6-Id

ant3

adenylase

aph6-Ic

aph6-Ic (deg)

tetA

tetB

tetC

tetD

tetE

tetG

tetH

tetK

tetL

tetM

tetO

tetT

Pro

du

cers

Str

epto

myci

n

strA

aphD

strB1

stsC

Soil Rhizosphere Manure Sewage Seawater

Soil Rhizosphere Manure Sewage Seawater

Reservoirs of antibiotic resistance genes in diverse environments: RESERVOIR survey Prevalence

Antibiotic class General

behaviour

Sewage

sludge

River

water

Groundwat

er

Drinking

water

Fish Soil Crops Example compounds

monitored

Chloramphenicol impersistent/

mobile - X - - - - -

2,4-

diaminopyridines

persistent/

immobile

X X - trimethoprim

Fluoroquinolones persistent/

immobile X X - -

ciprofloxacin, norfloxacin,

ofloxacin

-lactams impersistent

mobile - X X X - - -

amoxicillin, cloxacillin,

dicloxacillin, methicillin,

nafcillin, oxacillin, penicillin

G, penicillin V

Macrolides

slightly

persistent/

slightly mobile

X - - - -

azithromycin,

clarithromycin, lincomycin,

roxithromycin, spyramycin,

tylosin

Sulfonamides persistent/

mobile X -

sulfamethoxazole,

sulfadiazine, sulfamerazine,

sulfamethazine, sulfapyridine

Tetracyclines persistent/

immobile - X X

chlortetracycline,

doxycycline, oxytetracycline,

tetracycline

Occurrence of antibiotics in the natural environment, fish, crops and drinking water from published studies

A tick means that it has been monitored for and detected and a cross means that it has been monitored for and not detected. No entry means that no monitoring has been done yet (Alistair Boxall)

Application of sewage sludge to land: what is the impact on antibiotic resistance soil?

Sewage treatment and disposal

Schematic map of the complex class 1 integron carrying the blaCTX-M-14 gene on

plasmid pAJE0508 gene on plasmid pAJE0508

Bae et al., AAC, Aug 2007, 3017-19

Sampling polluted environments: Diagram of mill and reed bed system

Reed bed system for mill effluent

remediation oceans-ESU Ltd

CTAB cetyltrimethylammonium chloride DTDMAC ditallowdimethylammonium chloride

Relative resistance to DTDMAC and CTAB at 50 µg/ml

Gaze et al., AAC, May 2005, 1802-07

Incidence of int11, qaE, and qacEΔ : Reed bed and river sediment

Gaze et al., AAC, May 2005, 1802-07

Integron and qac gene prevalence: comparison of environmental samples

0

1

2

3

4

5

6

7

8

9

RB SS PS CW

sample site

pre

vale

nce (

%)

intI1

qacE∆1

qacE

qacG

qacH

Gaze et al., 2011. Impacts of anthropogenic activity on the ecology of class 1 integrons and integron-associated genes in the environment. ISME Journal.

RB, Reed bed sediment from textile mill - QAC pollution only SS, Fully digested sewage sludge - QAC pollution and antibiotic residues

Pig slurry - antibiotic residues only CW, Fallowed agricultural soil - no QACs or antibiotics

• Under strong QAC selection IS elements were maintained in the 5’ region of class 1 intergons where they acted as powerful promoters for gene cassette expression

• 90 million tons animal faecal slurry added to UK soils per year

Bailey-Byrne K, Gaze WH, Zhang L, Hawkey P, Wellington EMH. 2011. Class 1 integron and class 2 integron diversity in agricultural soil amended with pig slurry. AEM, Jan;77(2):684-7.

class 1 integron prevalence following pig manure application to land

0

0.002

0.004

0.006

0.008

0.01

0.012

0.014

0.016

pre-

application

day 1 day 21 day 90 day 289

days after slurry application

pre

vale

nce %

attI1 integrase binding site

aadA9 streptomycin/spectinomycin resistance gene, 99% blast homology with aadA9 from Corynebacterium glutamicum

qacEΔ1 quaternary ammonium compound resistance gene, 98% blast homology with E.coli

sulI sulphonamide resistance gene, 98% blast homology with Salmonella enterica

intI1

500bp

P2

5’conserved region/attI1

aadA9

750bp

59bp element

qacEΔ1 sulI

3’conserved region

450bp 6000bp

Schematic diagram of class 1 integron from Arthrobacter aritaii (strain C361), putatively carried on

a transferable plasmid

Byrne-Bailey et al. 2009. Appl Environ Microbiol. 77, 684-7.

Bailey-Byrne et al., 2011. Antimicrob.Agents Chemotherap. 53, 696-702.

Resistance to β-lactam antibiotics- culture independent

Reed Bed

soil

Sewage

Cake

1 Month

Cake

Applied

Control

Soil

Grass

Land Soil

FYM

Applied

Grass Soil

No. of clones 400000 386000 500000 170000 630000 210000

Average insert size

(Kb)

4.64 4.12 4.40 3.70 2.85 3.71

Clones with inserts

(%) 65 65 85 50 75 85

Coverage (Gb) 0.63 1.59 1.87 0.32 1.53 1.47

No of cefotaxime

resistance 0 2 1 0 0 0

No of ceftazidime

resistance 2 1 0 0 0 0

No of imipenem

resistance 1 ? 2 ? 0 0 0 0

No of amp res

clones 4 ? 5? 1 ? 0 0 0

Hit rate 1/80 1/150 1/900 0 0 0

Waste water treatment plants as a reservoir for antibiotic resistance

Waste Water treatment plants Hotspot for Horizontal Gene Transfer (HGT) as waste received from various sources

Little is known about the impacts of effluent further downstream in the river or the possible role of co-selection of antibiotic resistant determinants via quaternary ammonium compounds (QACs) (Gaze et al., AAC 2005, ISMEJ 2011)

Case study of blaCTX-M prevalence

3rd generation cephalosporin (3GC) resistance necessitates the use of last resort antibiotics

Extended Spectrum β-lactamases (ESBLs)

blaCTX-M is most dominant. CTX-M-15 is the major dominant enzyme type in the UK, and is reported worldwide

blaCTX-M is carried on several plasmids.

Genetic context can vary and is important in understanding the dissemination of blaCTX-M

P

ISEcP1 blaCTX-M-15 ORF477 48bp

P

24bp ISEcP1 remnant

blaCTX-M-15 ORF477

48bp

IS26

•3 sediment cores were taken at 3 x 500m intervals below (DS) and above (US) Finham sewage works on the River Sowe •Samples taken a year apart in late 2009 and early 2011 •Cultivation on Chromocult and PCR screening 3rd generation cephalosporin (3GC) resistance gene abundance and diversity

WWTP effluent acts as an input and or/selects for mobile antibiotic resistance determinants

Resistance Quotients Coliforms

4.48 x 105 coliforms / g DS 2.07 x 105 coliforms / g US

* P<0.05

0

5

10

15

20

25

Res

ista

nce

pre

vale

nce

/ (

%)

Antibiotic selection

DS

USDownstream and upstream of WWTP 2009 and 2011

*

*

* *

* * *

Flow of resistance genes into the rivers: Waste Water treatment plants

Amos et al., 2013

3GC resistance gene analysis

A subset of E. coli and other Enterobacteriaceae were taken from 2011 samples for further analysis

Sequencing of blaCTX-M revealed all belonged to the genotype blaCTX-M-15

0

10

20

30

40

50

60

70

80

90

100

CTX-M TEM SHV intI1

Pre

vale

nce

/ (

%)

Gene

DS

US

708 blaCTX-M carrying presumptive coliforms / g DS 141 blaCTX-M carrying presumptive coliforms / g US

Amos et al., 2013

blaCTX-M genetic context

P

ISECP1 blaCTX-M-15 ORF477 48bp

blaCTX-M was carried on 13 genetic contexts, including the ‘international genetic context’ Eleven were novel, 8 were found DS, 3 were found US and 2 were found DS and US, simultaneously

Different genetic contexts were carried on different plasmids and one genetic context could be seen on multiple plasmids

0

10

20

30

40

50

60

FIA FIB FIIA HI2 A/C I1/IY IncK

Pre

va

len

ce

Plasmid rep type

DS

US

Amos et al., 2013

Mobilisation of blaCTX-M-15

One DS E. coli carrying FIA, one DS E. coli carrying HI2. One US E. coli carrying FIB and HI2. One DS C. Freundii carrying FIB + K and one DS C. Freundii carrying FIB and I1/IY

• CTX-M-15 is carried throughout a wide range of genetic contexts and plasmids

• Contexts were seen in human pathogens, including several novel genetic contexts

• The environment may mobilise CTX-M-15 between plasmids and species and WWTP effluent may drive this process

Amos et al., 2013

IS26 tnpa 716 bp

CTX-M-15 875 bp

ORF477 151 bp

47 bp spacer

ISeCP1 tnpa 181 bp

256 bp, ISeCP1 IR CTX-M-15 promoter, spacer

CTX-M-15 875bp

IS26 tnpa 716 bp

124 bp spacer Is26 IRL 80 bp, IS26 IR CTX-M-15 promoter, spacer

CTX-M-15 875 bp

CTX-M-15 875 bp

Introduction or selection or both?

More detailed typing of the E. coli (MLST) was used to determine if E. coli were of the same origin upstream and downstream Upstream the E. coli sequence types were mainly uncharacterized (80 %), indicative of a more environmental origin Downstream the sequences types were split between the clinically familiar ST131, ST167, ST3103, ST1421 and environmental STs

0

0,5

1

1,5

2

2,5

3

3,5

4

DS1 DS2 DS3 US1 US2 US3

Pre

va

len

ce /

(%

)

Sample site

intI1

qacE∆1

qacE

qacH

Integron prevalence based on real time PCR data

Down stream of WWTP Upstream of WWTP

Collaboration with Wallingford CEH, meta-data available 13 sites samples every 3 months for a year: analysed for integron prevalence and 3GC resistance counts

Contribution of WWTP effluent to integron levels in a whole river system

River Thames catchment area:

Integron prevalence

0

0,5

1

1,5

2

2,5

3

3,5

Inte

gro

n P

reva

len

ce /

(%

)

Sample site

May

August

February

Significant difference between summer months (May and August, and Winter months November and February

P = 0.004 t-test

November

Evaluating the impact of WWTPs – model

development

Output WWTP only

Explained 49 % of variance: R2 adjusted (0.49) P < 0.01

0.5 2.5 1.5

-2.0

0.0

-1.5

-1.0

-0.5

0.0

0.5

1.0 2.0

Act

ual

lo

g in

tegr

on

pre

vale

nce

Predicted log integron prevalence

Explained 82. 9 % of variance : R2 adj (0.83) P < 0.01

WWTP: Main positive driver Coniferous Woodland -ve:Rainfall Neutral Grassland: Rainfall Rough Grassland - ve Acid Grassland : Season Heather Grassland - ve: Season Inland rock : Season Urban : Season Suburban : Season

-1.5 0.5 -0.5

-2.0

-2.0

-1.5

-1.0

-0.5

0.0

0.5

-1.0 0.0

actu

al lo

g in

tegr

on

pre

vale

nce

Predicted log integron prevalence

All metadata included

Model validated using 6 samples from independent river basin : Could predict integron prevalence within a 20 % range of actual integron value

The connectivity of potential sources of antibiotic-resistant bacteria

Acknowledgements

University of Warwick Dr William Gaze Dr Greg Amos Dr Andrew Mead Dr Lihong Zhang Dr Leonides Calvo-Bado Helen Green Abigail Carter Shruthi Sankaranaryanan

University of Birmingham Professor Peter Hawkey Claire Murray University of York Professor Alistair Boxall

CONCLUSIONS

Both resistance genes and integrons recovered in bacteria indigenous to soil and water environments

Enteric bacteria survive in soil and water, can transfer genes to both G+ and G- indigenous bacteria

Integrons and ESBLs present in environmental metagenomes

Pollutants, sewage, WWTP effluent associated with increased resistance- anthropogenic effects

QUESTIONS • What are the risks of environmental exposure? • Which mitigation strategies are possible? • How can efficiency of interventions be measured?