summary writing

Rawmeat-baseddietforpetsaneglectedsourceofhumanexposuretoSalmonellaandpathogenicEscherichiacoliclonescarryingmcr2019toJanuary2020.pdf

Home >Biology homework help >summary writing

1www.eurosurveillance.org

Research

Raw meat-based diet for pets: a neglected source of human exposure to Salmonella and pathogenic Escherichia coli clones carrying mcr, Portugal, September 2019 to January 2020

Marisa Ribeiro-Almeida1,2,3 , Joana Mourão1,2,4 , Mafalda Magalhães1,2,5 , Ana R Freitas1,2,6 , Carla Novais1,2 , Luísa Peixe1,2 , Patrícia Antunes1,2,5

1. UCIBIO – Applied Molecular Biosciences Unit, Laboratory of Microbiology, Department of Biological Sciences, Faculty of Pharmacy, University of Porto, Porto, Portugal

2. Associate Laboratory i4HB - Institute for Health and Bioeconomy, Faculty of Pharmacy, University of Porto, Porto, Portugal 3. School of Medicine and Biomedical Sciences, University of Porto (ICBAS-UP), Porto, Portugal 4. Centre for Innovative Biomedicine and Biotechnology (CIBB), University of Coimbra, Coimbra, Portugal 5. Faculty of Nutrition and Food Sciences, University of Porto, Porto, Portugal 6. 1H-TOXRUN, One Health Toxicology Research Unit, University Institute of Health Sciences (CESPU-CRL), Gandra, Portugal Correspondence: Patrícia Antunes ([email protected])

Citation style for this article: Ribeiro-Almeida Marisa, Mourão Joana, Magalhães Mafalda, Freitas Ana R, Novais Carla, Peixe Luísa, Antunes Patrícia. Raw meat-based diet for pets: a neglected source of human exposure to Salmonella and pathogenic Escherichia coli clones carrying mcr, Portugal, September 2019 to January 2020. Euro Surveill. 2024;29(18):pii=2300561. https://doi.org/10.2807/1560-7917.ES.2024.29.18.2300561

Article received on 16 Oct 2023 / Accepted on 04 Mar 2024 / Published on 02 May 2024

Background: The pet industry is expanding worldwide, particularly raw meat-based diets (RMBDs). There are concerns regarding the safety of RMBDs, especially their potential to spread clinically relevant antibiotic- resistant bacteria or zoonotic pathogens. Aim: We aimed to investigate whether dog food, including RMBD, commercially available in Portugal can be a source of Salmonella and/or other Enterobacteriaceae strains resistant to last-line antibiotics such as colistin. Methods: Fifty-five samples from 25 brands (21 inter- national ones) of various dog food types from 12 sup- pliers were screened by standard cultural methods between September 2019 and January 2020. Isolates were characterised by phenotypic and genotypic methods, including whole genome sequencing and comparative genomics. Results: Only RMBD batches were contaminated, with 10 of 14 containing poly- clonal multidrug-resistant (MDR) Escherichia coli and one MDR Salmonella. One turkey-based sample con- tained MDR Salmonella serotype 1,4,[5],12:i:- ST34/ cgST142761 with similarity to human clinical isolates occurring worldwide. This Salmonella exhibited typical antibiotic resistance (blaTEM + strA-strB + sul2 + tet(B)) and metal tolerance profiles (pco + sil + ars) associ- ated with the European epidemic clone. Two sam- ples (turkey/veal) carried globally dispersed MDR E. coli (ST3997-complexST10/cgST95899 and ST297/ cgST138377) with colistin resistance (minimum inhibitory concentration: 4 mg/L) and mcr-1 gene on IncX4 plasmids, which were identical to other IncX4 circulating worldwide.

Conclusion: Some RMBDs from European brands avail- able in Portugal can be a vehicle for clinically relevant MDR Salmonella and pathogenic E. coli clones carrying genes encoding resistance to the last-line antibiotic colistin. Proactive actions within the One Health context, spanning regulatory, pet-food industry and consumer levels, are needed to mitigate these public health risks.

Introduction The pet industry has evolved in recent decades due to increasing pet populations, stronger human–pet bonds and demand for high-quality pet food products [1,2]. Processed pet food manufactured with various pro- cessing methods (e.g. grinding, cooking, extrusion and dehydration) has traditionally been considered micro- biological safe and nutritionally suitable for feeding pets [1,3]. However, since some pet owners consider unprocessed food healthier, raw meat-based diets (RMBDs) for dogs have gained popularity [1,2,4]. The RMBDs are mainly composed of uncooked or minimally processed meat, bones and organs, with freezing as the primary treatment, and are considered to be more natural than conventional processed pet food [1,5]. Nevertheless, the scientific evidence supporting RMBD benefits is scarce, and many veterinary professional organisations (e.g. the World Small Animal Veterinary Association) and international public health agencies (e.g. the United States (US) Centers for Disease Control and Prevention (CDC)) view them as potential health hazards for both animals and humans [1,5]; awareness of this issue appears less evident in Europe [6]. The

2 www.eurosurveillance.org

safety concerns associated with RMBDs are related to the potential contamination of raw ingredients with zoonotic pathogenic bacteria and parasites [1,3,4]. Such contamination could lead to the spread of these pathogens to both pets and humans cohabitating with pets, through direct contact with the pet or its feed, or indirectly through contact with contaminated house- hold surfaces or hands during feed preparation.

In the European Union (EU), legal requirements for the use of animal by-products and derived prod- ucts not intended for human consumption are estab- lished, including those to produce processed or raw pet food, helping to ensure microbiological safety [7]. Nevertheless, since 2020, there have been more than 20 notifications or recalls of pet food and RMBD in the EU due to the detection of zoonotic pathogens, particu- larly Salmonella and pathogenic Escherichia coli [8], and also cases of human infections with Salmonella and Shiga toxin-producing E. coli (STEC) linked to exposure to RMBDs [9-11]. Several studies have also established a correlation between the microbiota of pets and their owners, including the presence of antibiotic-resistant strains, with pet food as a potential source [12,13]. However, certain antibiotic-resistant bacteria and genes of public health concern, such as the mcr gene conferring resistance to the last-line antibiotic colistin, have not been extensively studied in pet food and RMBDs [1,14-16]. Consequently, these antibiotic-resist- ant strains and genes have not been recognised as notable food safety issues in the context of the pet food industry [6]. To address this knowledge gap, we aimed to investigate the occurrence of and further characterise Salmonella and other Enterobacteriaceae resistant to critical antibiotics, such as colistin, in dog

food, including RMBDs, that is available in stores in Portugal to investigate if they represent a possible source of these hazards to public health.

Methods

Sampling strategy We visited physical locations such as major supermar- kets and pet stores in the Porto metropolitan area and conducted an online search to gather information on the primary canine food types and brands commercially accessible in Portugal. Over 5 months (September 2019 to January 2020), 55 dog food samples (22 wet, 14 raw- frozen, eight dry, seven treats and four semi-moist), corresponding to 50 different dog food items (four food types were acquired 2–3 times) and 25 brands commer- cialised in Portugal, were collected from 12 retail stores (eight supermarkets, three specialised stores and one veterinary clinic) in the Porto region; further details about the samples are appended in Supplementary Table S1. Most samples were obtained from brands marketed globally, including in the EU (21/25). The 14 raw-frozen dog food samples were a combination of fruits, vegetables and different types of meat (muscle/ viscera). They were categorised into two groups based on the main meat type: poultry (n = 8; chicken, turkey, duck, goose) and ruminant (n = 6; veal, steer, deer) samples. The RMBDs were from the two international brands available in Portuguese stores (brand A types produced in the EU and brand B in the United Kingdom (UK)). The samples were processed according to their type, as previously described [17]. For the pre-enrich- ment step at 37 °C for 16–18 h, 25 g of each sample were homogenised (2 min in a Stomacher blender) with 1:10 buffered peptone water (BPW).

What did you want to address in this study and why? Raw meat-based diets (RMBDs) are increasingly popular among pet owners. However, the potential role of RMBDs has been neglected as a new source of bacteria resistant to last-resort antibiotics which could affect people co-living with pets. We wanted to analyse if dog food from brands sold in the European Union represents a possible source of Salmonella or other bacteria resistant to the important antibiotic colistin.

What have we learned from this study? Conventionally processed pet food is a safer option than RMBDs. This is because RMBDs of European brands can carry multidrug-resistant bacteria, including globally disseminated pathogenic Salmonella and E. coli harbouring genes encoding resistance to colistin, an antibiotic critically important for human medicine. These hazards are frequent in food-animal production and are causing infections in humans worldwide.

What are the implications of your findings for public health? The detection in RMBDs of a predominant pandemic Salmonella clone and pathogenic E. coli carrying mobile colistin resistance genes may pose a potential risk of human exposure. This can occur through handling of pet food and/or environmental release by pets. These findings indicate a need for proactive actions involving the pet industry, food safety agencies, and pet owners to mitigate risks for public health.

KEY PUBLIC HEALTH MESSAGE

3www.eurosurveillance.org

Detection of non-typhoidal Salmonella Salmonella detection was performed using the International Standard Organisation [18] method for foodstuffs. Briefly, after the pre-enrichment, 0.1 mL and 1 mL of the BPW were transferred to Rappaport– Vassiliadis medium with Soya (RVS) and Muller– Kauffmann tetrathionate-novobiocin (MKTTn) broths, respectively, for selective enrichment (RVS at 41.5 °C for 24 h and MKTTn at 34–38 °C for 24 h). These broths were then streak-plated on xylose lysine deoxycholate agar and CHROMagar Salmonella Plus. Presumptive Salmonella colonies recovered from both selective agar plates (up to five colonies per plate) were confirmed by biochemical tests (e.g. API-20 E, bioMérieux, Marcy l’Etoile, France), by agglutination with Salmonella O poly antisera and serogroup-specific antisera (Becton Dickinson, New Jersey, US) and by PCR for invA gene detection and Salmonella serotypes of particular concern in the EU (Enteritidis, Typhimurium and 1,4,[5],12:i:-) [19].

Screening of mcr-carrying Enterobacteriaceae After BPW enrichment, 100 µL and 10 µL were spread on Tryptone Bile X-glucuronide agar plates (TBX) and Simmons citrate agar + inositol (SCAi) with and without colistin (3.5 mg/L) and incubated (TBX at 37 °C for 24 h; SCAi at 37 °C for 48 h) for E. coli and Klebsiella spp. detection, respectively. From each plate, between one and five colonies of each morphotype were spread on a CLED medium for further identification by matrix- assisted laser desorption-ionisation-time of flight mass spectrometry (MALDI-TOF VITEK MS, bioMérieux) and standard PCRs for E. coli and K. pneumoniae [20]. Colistin resistance genes (mcr-1 to mcr-5 and mcr-6 to mcr-9) were identified in E. coli, K. pneumoniae and S. enter- ica isolates using a multiplex PCR published previously [21]. Amplified simplex PCR products were purified using the NZYGelpure kit (NZYTech, Lisbon, Portugal) and sequenced at Eurofins Genomics (Konstanz, Germany).

Phenotypic and genotypic characterisation of Enterobacteriacea We used disk diffusion to test susceptibility to the fol- lowing antibiotics: amoxicillin 10 µg, amoxicillin/cla- vulanic acid 30 µg, cefepime 30 µg, cefoxitin 30 μg, ceftazidime 30 μg, cefotaxime 30 μg, meropenem 10 µg, ciprofloxacin 5 μg, pefloxacin 5 μg (only for Salmonella) nalidixic acid 30 µg, gentamicin 10 μg, streptomycin 10 µg, kanamycin 30 µg, tobramycin 10 μg, chloram- phenicol 30 μg, tetracycline 30 μg, sulfonamides 300 µg, trimethoprim 5 µg and fosfomycin 200 µg (only for E. coli). Colistin minimum inhibitory concentration (MIC) was determined by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) reference cation-adjusted Mueller–Hinton broth microdilution method [22]. Escherichia coli ATCC 25922 served as the control. Interpretation followed the EUCAST guidelines [23], and for nalidixic acid and tetracycline, we used the Clinical and Laboratory Standards Institute clini- cal breakpoints [24]. Multidrug resistance (MDR) was

defined as resistance to antibiotics from three or more different families. Phylogenetic groups (PhG) of E. coli were determined using a standard multiplex PCR [20]. In addition, Shiga toxin-producing E. coli (STEC) was assessed by PCR for stx1 and stx2 virulence genes [25].

Whole genome sequencing for characterisation of Salmonella and mcr-1-carrying Escherichia coli We selected one isolate per sam- ple of Salmonella and mcr-1-posi- tive Enterobacteriaceae for WGS. We extracted DNA using the Wizard Genomic DNA purification kit (Promega Corporation, Madison, US) and measured its concentration with a Qubit 3.0 Fluorometer (Invitrogen, Thermo Fisher Scientific, Massachusetts, US). The HiSeq (2 × 151 bp) Illumina platform (Illumina, San Diego, US) was used for sequencing at Eurofins Genomics. Raw reads quality was assessed with FastQC v0.11.9 (http://www.bioinformatics.babraham.ac.uk/ projects/fastqc) using default parameters. High-quality reads were then de novo assembled using SPAdes v3.15.5 (https://github.com/ablab/spades) within Unicycler v0.5.0 (https://github.com/rrwick/Unicycler). Assembly quality and completeness were assessed with QUAST v5.0.2 (https://quast.sourceforge.net) and BUSCO v5.0.0 (https://github.com/WenchaoLin/ BUSCO-Mod), respectively. Draft genomes were anno- tated on the RAST server (https://rast.nmpdr.org). For metal tolerance genes search (arsRD2A2BCA1D1- arsR1HD1A1A2CBA3D2R2, pcoGE1ABCDRSE2, silESRCF- BAGP, terFEDCBAZ-terY3Y2XY1W and merEDACPTR), we used ABRicate v1.0.1 (https://github.com/ tseemann/abricate) with an in-house database. We used tools from the Centre for Genomic and Epidemiology (http://www.genomicepidemiology. org) to evaluate E. coli and Salmonella antibiotic resistance genes (ResFinder v4.1, https://cge.food. dtu.dk/services/ResFinder) or known mutations (PointFinder v4.1, https://bitbucket.org/genomicepi- demiology/pointfinder_db.git), virulence genes (only for E. coli, VirulenceFinder v2.0, https://cge.food. dtu.dk/services/VirulenceFinder), plasmid replicons (PlasmidFinder v2.1, https://cge.food.dtu.dk/services/ PlasmidFinder), plasmid typing (pMLST v2.0, https:// cge.food.dtu.dk/services/pMLST) and Multilocus Sequence Typing (MLST v2.0, https://cge.food.dtu.dk/ services/MLST). Salmonella serotypes were confirmed with the Salmonella In Silico Typing Resource (SISTR) (https://github.com/phac-nml/sistr_cmd) and E. coli PhGs were validated using ClermontTyper (http:// clermontyping.iame-research.center).

For confirmation of mcr-1 gene location and hypothetical plasmid reconstructions, we used the MOB-recon tool v3.1.0 from the MOB-suite package (https://github. com/phac-nml/mob-suite). If the mcr-1 gene was iden- tified in a plasmid by MOB-recon or on the same contig as the replicon, it was considered plasmid-associated. The PLSDB-plasmid database (https://ccb-microbe.

4 www.eurosurveillance.org

cs.uni-saarland.de/plsdb) was used for comparative genomic analysis. Alignment of mcr-1-carrying plasmids with closely related IncX4 ones was conducted using the BRIG tool (v0.95) (https://github.com/happykhan/ BRIG).

Comparative genomic analysis of Salmonella and mcr-1-carrying Escherichia coli We conducted a comparative genomic analysis using core-genome MLST (cgMLST) between our isolates and genomes queried from Enterobase as well as the hierarchical clustering of cgMLST (HierCC) (https:// enterobase.warwick.ac.uk). These strains were used to develop a minimum-spanning tree using GrapeTree (https://achtman-lab.github.io/GrapeTree/ MSTree_holder.html) and MSTreeV2. Metadata of the included Salmonella and E. coli isolates were retrieved from Enterobase (isolate name, cgST, country, year, source). In addition, we conducted a search of antibiotic resistance, metal tolerance and virulence genes as described in the previous section.

Statistical analysis Occurrence rates and antibiotic-resistant E. coli variations across food types were assessed using Fisher’s exact test (α = 0.05). The 95% confidence intervals (CI) for proportions were calculated using Wilson CI. Both analyses were computed using Prism v 9.1.1 (GraphPad, Boston, Massachusetts, US).

Results

Detection and characterisation of Salmonella In our study of 55 pet food samples (41 processed and 14 raw), only raw samples tested positive for Gram-negative bacteria, including the zoonotic pathogen Salmonella, along with bioindicators E. coli and Klebsiella pneumo- niae (Table 1). We detected Salmonella in one of the raw samples (7%; 95% CI: 1.3–31.5; n = 1/14), a raw- frozen batch (EU, brand A), predominantly containing turkey (Table 1).

Six Salmonella isolates, all identified as Salmonella enterica serotype 1,4,[5],12:i:- (Typhimurium monophasic variant) were successfully isolated from the same sample. They exhibited the MDR ASSuT profile conferring resistance to amoxicillin (A, encoded by the bla TEM gene), streptomycin (S, strA-strB), sul- phonamides (Su, sul2), and tetracycline (T, tet(B)). In addition, they had an integrative and conjugative ele- ment (ICE) with copper/silver (pco, sil) and arsenic (ars) tolerance clusters/operons (Table 2), typical of the widespread clinically relevant ‘European clone’ [26].

The S. 1,4,[5],12:i:- strain belonged to sequence type ST34 (Table 2), which is commonly observed in Europe, as evidenced by the data available on EnteroBase (https://enterobase.warwick.ac.uk/species/index/ senterica). Using the EnteroBase cgMLST scheme, the pet food isolate was classified as cgST142761, which grouped into a distinct cluster (Hierarchical Clustering- HierCC HC5–142761 group) among the globally dis- persed ST34 clone (Figure 1A). This isolate exhibited

Table 1 Distribution of Salmonella enterica, Escherichia coli and Klebsiella pneumoniae among samples of commercially available raw-frozen dog food, Portugal, September 2019–January 2020 (n = 14)

Food type Sample number Branda Main ingredientsb

(% of the most prevalent ingredients) Collection date Bacteria detected

S. enterica E. coli (mcr-1) K. pneumoniae

Ruminant- based

6 A Veal (59%); salmon (20%); vegetables;

salmon oil

Sept 2019 − + − 46 Nov 2019 − + − 53 Jan 2020 − + − 18

A Steer (79%); vegetables; salmon oil Oct 2019 − + +

45 Nov 2019 − + − 26 B Deer (80%); vegetables; fruit Nov 2019 − + (mcr-1) −

Poultry- based

15 A Chicken (60%); lamb (19%); vegetables Oct 2019 − + − 16

A Chicken (60%); veal (19%); vegetables Oct 2019 − + +

55 Jan 2020 − + − 17

A Turkey (60%); lamb (20%); vegetables Oct 2019 − + −

54 Jan 2020 + + − 25 B Duck (80%); vegetables; fruit Nov 2019 − + +

51 B Turkey (50%); goose (30%); vegetables; fruit Jan 2020 − + +

52 B Turkey (40%); salmon (20%); white fish (20%); vegetables Jan 2020 − + (mcr-1) −

−: not detected; +: detected. a Raw food samples were from either of two international brands distributed worldwide, here randomly designated as A and B. b The most prevalent ingredient in each food type is shown in bold.

5www.eurosurveillance.org

Ta bl

e 2

G en

om ic

c ha

ra ct

er is

at io

n of

S al

m on

el la

e nt

er ic

a, a

nd th

e m

cr -1

-c ar

ry in

g Es

ch er

ic hi

a co

li is

ol at

es re

co ve

re d

fr om

sa m

pl es

o f c

om m

er ci

al ly

a va

ila bl

e ra

w -f

ro ze

n do

g fo

od , P

or tu

ga l,

Se pt

em be

r 2 01

9– Ja

nu ar

y 20

20 (n

= 3

)

Is ol

at e

na m

e Sp

ec ie

s Br

an d

(m ai

n in

gr ed

ie nt

) Se

ro ty

Ph G

M LS

cg M

LS Ta

An tim

ic ro

bi al

re si

st an

ce g

en es

; c hr

om os

om al

ut at

io ns

m cr

-1 lo

ca tio

PL -In

c (k

b) Vi

ru le

nc e

ge ne

sb Ac

qu ire

d m

et al

to le

ra nc

e ge

ne s

R5 4_

S1 00

Sa lm

on el

en te

ri ca

A (t

ur ke

y) 1,

4, [5

],1 2:

i:-

N A

ST 34

cg ST

14 27

bl a TE

M -1

B, a ac

(6 ’)-

Ia a,

s tr

A, s

tr B,

s ul

2, te

t( B)

N A

NS si

lE RS

CF BA

G P,

p co

G E1

AB CD

RS E2

, a rs

RD 2A

2B CA

1D 1,

m er

ED AC

PT R

R2 6_

1_ 10

0 Es

ch er

ic hi

a co

li B

(d ee

r) O

15 3:

H3 7

ST 29

cg ST

19 31

bl a TE

M -1

B, b la

CA RB

-2 , a

ad A1

, a ad

A2 , c

m lA

1, m

cr -

1. 1,

s ul

3, d

fr A1

6; g

yr A

p. S8

3L , g

yr A

p. D8

7N ,

pa rC

p .S

80 I

PL -X

4 (3

3) ga

d, lp

fA ,o m pT

, t er

ar sR

1B CR

R5 2_

8 Es

ch er

ic hi

a co

li B

(t ur

ke y)

O 10

1: H3

A ST

39 97

cg ST

19 31

bl a TE

M -1

B,m cr

-1 .1

, s tr

A, s

tr B,

te t(

B) , m

df (A

); gy

rA p

.S 83

L, g

yr A

p. D8

7N , p

ar C

p. S8

0I PL

-X 4

(3 3)

cm a,

c va C,

e ts C,

g ad

,h ly F,

h ra

, i ro N,

is s,

o m pT

, t er

C, tr aT

ar sR

1B C

AP EC

: a vi

an p

at ho

ge ni

c Es

ch er

ic hi

a co

li; c

gM LS

T: c

or e

ge no

m e

m ul

til oc

us s

eq ue

nc e

ty pi

ng ; E

xP EC

: e xt

ra in

te st

in al

p at

ho ge

ni c

Es ch

er ic

hi a

co li;

M LS

T: m

ul til

oc us

s eq

ue nc

e ty

pi ng

; N A:

n ot

a pp

lic ab

le ; N

S: n

ot s

ea rc

he d;

P hG

: E . c

ol i p

hy lo

ge ne

tic g

ro up

; P L:

p la

sm id

. a c

gM LS

T ob

ta in

ed fr

om E

nt er

ob as

e (h

tt ps

:/ /e

nt er

ob as

e. w

ar w

ic k.

ac .u

k) fo

r b ot

h sp

ec ie

s. b T

he v

ir ul

en ce

g en

es a

ss oc

ia te

d w

ith E

xP EC

a re

h ig

hl ig

ht ed

in b

ol d,

w hi

le th

os e

as so

ci at

ed w

ith A

PE C

ar e

un de

rl in

ed ; g

en es

th at

b el

on g

to b

ot h

ar e

in di

ca te

d in

b ol

d an

d un

de rl

in e.

6 www.eurosurveillance.org

a close genetic relationship (HC5) with 21 isolates of human origin (Czechia, France and UK; 2018–2022), with no apparent clustering based on European geo- graphical location (Figure 1A); Supplementary Table S2 contains further strain details.

Detection and characterisation of colistin- resistant Enterobacteriaceae We detected E. coli (n = 59 isolates; none STEC) in all raw-frozen food samples (100%; 95% CI: 79–100; n = 14/14). Four of them were also contaminated with K. pneumoniae (n = 5 isolates) (Table 1). A considerable percentage (71%; 95% CI: 45–88; n = 10/14) of the samples carried MDR E. coli isolates, independently of the main ingredients and PhGs (Table 3). Antibiotic resistance rates were similar between samples with poultry or ruminant-based ingredients (p > 0.05). More than half of the samples contained at least one E. coli with resistance to amoxicillin (79%; 95% CI: 52–92; n = 11/14 samples), ciprofloxacin (50%; 95% CI: 27–73; n = 7/14), nalidixic acid (57%; 95% CI: 33–79;

n = 8/14), streptomycin (71%; 95% CI: 45–88; n = 10/14), tetracycline, sulphonamides or trimethoprim (64%; 95% CI: 39–84; n = 9/14 each); For further antibiotic resistance details, we refer to Supplementary Figure S1.

Colistin-resistant E. coli isolates (n = 4) were present in two batches (14%; 95% CI: 4–40; n = 2/14) from the same pet food brand (UK, brand B), one with deer and the other with turkey as the main ingredient (Table 3). All iso- lates carried the mcr-1 gene with MIC = 4 mg/L and were recovered in TBX medium supplemented with colistin. They were co-resistant to amoxicillin, streptomycin, ciprofloxacin, nalidixic acid, streptomycin and sulphonamides. These mcr-1-carrying E. coli isolates belonged to B1-ST297 and A-ST3997-ST10 complex- Cplx (Table 2). The ST297 (cgST193137), classified as an extraintestinal pathogenic E. coli (ExPEC), shared a common HC100 cluster (Hierarchical Clustering-HierCC HC100–3512 group) with 162 genomes and the lowest number of allelic differences with 13 genomes from globally dispersed sources and countries (Germany,

Figure 1 Phylogenetic trees of Salmonella Typhimurium and Escherichia coli isolates from raw pet food samples, Portugal, 2020 (n = 3) and related available genomes on EnteroBase up to 8 June 2023 (n = 22 Salmonella Typhimurium, n = 20 E. coli)

57575757

39393939

4444 ST95899

ST227583

ST95936

ST193139

China (n = 4) Portugal (n = 1) Czechia (n = 1)

Country

5555

4444

3333 3333

3333

2222

1111

1111 1111

1111

1111 ST163262

ST247147

ST245660

ST243925

ST282169 ST244490

ST142932 ST150646

ST163786

ST262929

ST180660

ST201463

ST150634

ST181237

ST142761

France (n = 16) United Kingdom (n = 4) Czechia (n = 1) Portugal (n = 1)

Country

92929292

80808080

63636363

62626262

60606060

50505050

44444444

41414141

40404040 40404040

30303030

2222

1111ST138377

ST167644

ST96576

ST208541

ST213010

ST59463 ST218183

ST190921

ST45894

ST45901

ST29145

ST191345

ST193137

Netherlands (n = 2) Germany (n = 3) Uganda (n = 3) Ecuador (n = 2) Portugal (n = 2) Italy (n = 1) Kenya (n = 1)

Country

A. Grape tree of Salmonella Typhimurium isolates and the monophasic variant S. 1,4[5],12:i:- ST34/cgST142761 (HC5-142761)

C. Grape tree of the E. coli ST3997/cgST193139 (HC50-57741) isolates

B. Grape tree of E. coli ST297/cgST193137 (HC100-3512) isolates; inset: pet food isolates and related genomes

The core genome minimum spanning tree was created within the EnteroBase pipeline using the MSTreeV2 algorithm and GrapeTree tool. The cgST is indicated in each node. The yellow circle highlights the pet food isolate of each Grape tree. For the geographical analysis, the cgST was annotated using the country data (the number of genomes by country is indicated within parentheses). The scale bar corresponds to the number of cgMLST allelic differences.

7www.eurosurveillance.org

Italy, the Netherlands, Ecuador, Kenya and Uganda; 2014–2021), including from a Portuguese poultry farm (Figure 1B); Supplementary Table S3 provides further strain details. The ST3997-ST10 Cplx (cgST193139) presented virulence genes associated with avian pathogenic E. coli (APEC) and shared a single cluster at the HC50 level with five isolates from humans in Europe (Czechia, 2020) and Asia (China, 2017) (Figure 1-C); for further strain details we refer to Supplementary Table S4. Whole genome sequencing revealed that in both E. coli strains from the same pet food brand, the mcr-1.1 gene was located on similar IncX4 plasmids (99.89% identity). These plasmids shared a common genetic environment near the mcr-1 cassette, con- tained the pap2 gene (membrane-associated lipid phosphatase) and lacked the ISApI1 element [16]. Comparative genomics revealed that these IncX4 plas- mids were similar to others (MOB-recon; mash dis- tance: 0.000780658–0.00126265) and are circulating among diverse hosts (humans, pig, poultry) and the

environment in many different countries, including Portugal (Figure 2); Supplementary Table S5 provides further plasmid details.

Discussion This study investigated the presence and character- istics of Salmonella and other Enterobacteriaceae in 55 dog food samples, with a focus on colistin- resistant strains. These samples comprised various types of meat and were obtained from different suppliers and international brands in Portugal. We found Enterobacteriaceae, including Salmonella and MDR isolates, only in samples from raw pet food, in contrast to a parallel study in the same samples [17], where Enterococcus spp. was detected across all sample types, including dry and wet. Current regulations in the EU propose counting Enterobacteriaceae (and including Salmonella detection) as a hygiene criterion for all categories of pet foods [7]. Numerous studies have demonstrated contamination levels exceeding the

Table 3 Characteristics of Escherichia coli recovered from samples of commercially available raw-frozen dog food, Portugal, September 2019–January 2020 (n = 59 isolates)

Food type Sample ID (brand)

Main ingredientsa (% of the most prevalent ingredients)

PhG (number of isolates) Antibiotic resistance profileb mcr-1 gene

detection

MDR (sample

ID)c

Ruminant- based

6, 46, 53 (A)

Veal (59%); salmon (20%); vegetables; salmon oil

A (2) AML (STR, SUL, TMP) − + (53) B1 (8) − − − B2 (2) − − − D (1) − − −

18, 45 (A) Steer (79%); vegetables; salmon oil

A (2) (AML, STR, CHL, TET, SUL, TMP) − + (45) B1 (4) (AML, STR, NAL, TET) − + (18)

26 (B) Deer (80%); vegetables; fruit A (2) TET − −

B1 (4) (AML, AMC, CIP, NAL, STR, CHL, COL, TET, SUL, TRP) + +

Poultry-based

15 (A) Chicken (60%); lamb (19%); vegetables B1 (4) AML, TET, SUL, TMP (AMC, CHL,

CIP, NAL, STR, KAN) − +

16, 55 (A) Chicken (60%); veal (19%); vegetables

A (10) (AML, CIP, NAL, STR, TET, SUL, TMP) − +

C (2) AML, CIP, NAL, STR, TET, SUL, TMP (FOX) − + (55)

17, 54 (A) Turkey (60%); lamb (20%); vegetables

B1 (2) − − −

D (2) STR, TET, SUL, TMP (AML, CIP, NAL) − + (17)

25 (B) Duck (80%); vegetables; fruit A (1) − − −

B1 (1) − − −

51 (B) Turkey (50%); goose (30%); vegetables; fruit

A (1) NAL, TET − −

B1 (7) CIP, NAL, TET (AML, GEN, STR, KAN, TOB, CHL, SUL, TRP) − +

52 (B) Turkey (40%); salmon (20%); white fish (20%); vegetables

A (3)

AML, AMC, STR, TET (CIP, NAL, KAN, GEN, COL, SUL, TRP)

AML, AMC, CIP, NAL, GEN, STR,

TET, SUL, TRP

+ +

B1 (1) − +

−: not detected; +: detected; AMC: amoxicillin + clavulanic acid; AML: amoxicillin; CHL: chloramphenicol; CIP: ciprofloxacin; COL: colistin; FOX: cefoxitin; GEN: gentamicin; ID: identification number; KAN: kanamycin; MDR: multidrug resistance; NAL: nalidixic Acid; PhG: E. coli Phylogenetic Group; STR: streptomycin; SUL: sulphonamides; TET: tetracycline; TOB: tobramycin; TRP: trimethoprim.

a The most prevalent ingredient in each food type is represented in bold. b Variable presence of antibiotic resistance is presented in brackets. (−) indicates isolates susceptible to all tested antibiotics. c The sample ID number is indicated only when more than one food sample type was analysed.

8 www.eurosurveillance.org

EU limits (i.e. Salmonella: absence in 25 g and Enteroba cteriaceae < 5 × 103 CFU/g) in RMBDs [4,5,14,15,27]. Our results strongly suggest that conventionally processed pet food is a safer option, emphasising the critical role of heat treatment in pet food production for effectively mitigating microbiological hazards [1,2].

Although the overall prevalence of Salmonella in the RMBDs samples in this study was low (one of 14 batches produced in the EU had unsatisfactory microbiological quality), other studies from Europe also detected Salmonella: 4% of raw pet food samples in Switzerland, 20% in the Netherlands and 71% in Italy [14,16,27]. The Salmonella enterica detected in this study was of the serotype 1,4,[5],12:i:- and belonged to ST34, which has emerged as the predominant pandemic genotype in recent decades, particularly in food animal production and human infections in the EU [26,28]. Their MDR features (ASSuT + ICE) may have facilitated the adaptation of this serotype to environ- ments with extensive usage of antibiotics and heavy metals, such as pig and poultry farms [26,28,29], whose raw animal by-products are the sources of the pet food industry. Since food animals are asympto- matic carriers of Salmonella, these bacteria can spread easily at slaughterhouses through cross-contamination events between flocks or animal by-products or at pet

food production plants, in various types of meat, ani- mal species and geographical places of origin [6,19,30]. Notably, we showed genetic similarities between S. 1,4,[5],12:i:- from RMBDs and public genomes from human clinical cases from different European coun- tries, suggesting a role of raw pet food as a potential vehicle for the transmission of this serotype consid- ered of human health significance in the EU and car- rying a MDR profile. Some studies consistently show a significant difference in Salmonella excretion in faeces between dogs fed with RMBDs and those fed with dry food, highlighting the microbiological risk associated with RMBDs [3,31,32]. This risk extends not only to dogs but also to pet owners handling RMBD and dog faeces, as well as to the environment, as documented by recent Salmonella outbreaks where WGS confirmed a connection between pets, pet food and human disease [10,11,33,34].

A high percentage of our samples carried MDR E. coli isolates, regardless of the raw food types tested, similar to a recent parallel study focused on MDR Enterococcus in dog food in Portugal [14]. Resistance to commonly used veterinary antibiot- ics such as β-lactams, fluoroquinolones, tetracycline and sulphonamides was especially pronounced, mir- roring trends seen in other European studies on pet

Figure 2 Circular maps of mcr-1-harbouring IncX4 plasmids in Escherichia coli isolated from Portuguese pet food, September 2019– January 2020 (n = 2) and closely related ones from different sources and geographical regions

Accession number

NZ_CP016550.1

Source (host) Country

NZ_CP069661.1

NZ_KX894453.1

CP042970.1

NZ_MK172815.1

NZ_MK875285.1

This study

Faeces

(Homo sapiens)

Netherlands

Mash distance

0.00126265 (R26_1_100)

0.00210772 (R52_8)

Wound

(Homo sapiens)

Poland 0.00175922 (R26_1_100)

0.000705841 (R52_8)

Faeces

(Sus scrofa)

Italy 0.00152214 (R26_1_100)

0.000780658 (R52_8)

Raw milk

cheese

Egypt 0.00146992 (R26_1_100)

0.000780658 (R52_8)

Urine

(Homo sapiens)

Russia 0.00136596 (R26_1_100)

0.000855821 (R52_8)

Chichen meat

from Brazil

(Gallus gallus)

South Korea 0.0011089 (R26_1_100)

0.000385581 (R52_8)

Raw-frozen

dog food

Portugal 0.00113443 (R26_1_100)

Raw-frozen

dog food

Portugal 0.00113443 (R52_8)

Homo sapiens

Alignment of R26_1_100 of 33789 bp and R52_8 of 32534 bp (Portugal, 2020) against others. Black inner ring: plasmid used as a reference for the alignment; name and size of the reference indicated in the middle of the panel. Genes are represented by black arrows (indicating the position and direction of transcriptional open-reading frames) in the outer circle, with mcr-1.1 and pap genes highlighted in red.

9www.eurosurveillance.org

food samples of diverse origins [14,35]. The use of these antibiotics, particularly in poultry production, has been associated with increased E. coli resistance rates [29], suggesting that raw meat-based ingredients might be introducing MDR strains in pet food, which then can persist until they reach humans and their pets [32]. While the percentage of samples containing MDR and mcr-carrying E. coli isolates was relatively low, in line with findings from other studies [14,16], it underscores the importance of employing antibiot- ics judiciously within the livestock industry. This is needed to curb the co-selection of genes conferring resistance to colistin, a “highest priority critically important antimicrobial” for human medicine among various bacterial species [20,21]. In fact, E. coli ST297 (ExPEC) and ST3997-ST10 Cplx (APEC) identified in this study have been detected worldwide in various animal, food, environmental and human sources, and have been linked to numerous human infections (https:// enterobase.warwick.ac.uk/species/index/ecoli), which highlights their capacity to be transmitted to humans through the food chain. In Portugal, the MDR E. coli ST297 lineage is predominant in many food sources [20,36] and has now been detected in raw pet food in our study. Meanwhile, in Asia, E. coli ST3997-ST10 Cplx isolates found in poultry, the environment and humans also carried mcr-1 associated with diverse plasmid backgrounds [37]. Notably, both E. coli strains obtained in this study from the same pet food brand carried the mcr-1.1 gene on similar IncX4 plasmids. These findings, along with the similarity of these plasmids to globally distributed ones, suggest possible cross- contamination events and/or diverse origins of pet food contamination arising from ingredients or human handling at the production plant.

In this study, RMBDs were identified as a potential vehicle of MDR zoonotic-related pathogenic bacteria, with some carrying genes such as mcr-1 conferring resistance to last-line antibiotics. Despite the EU›s efforts to reduce antibiotics such as colistin in livestock production and the successful colistin restrictions on EU farms [20,21], the introduction of colistin-resistant bacteria through imported animal by-products (e.g. from non-EU countries with different antibiotic prac- tices and regulations), raw vegetables (common in most samples) or wildlife (e.g. deer as the main ingre- dient in one batch with mcr-carrying E. coli) cannot be excluded. Continuous vigilance is essential to address these potential pathways and mitigate the spread of antibiotic-resistant bacteria. Furthermore, most man- ufacturers do not provide information on food safety practices (e.g. handwashing, safe handling) for han- dling raw pet food [38], including on labels found on the raw pet food samples obtained for this study. Appropriate hygiene measures and safe handling prac- tices should be observed when dealing with pets and raw pet food to mitigate the risk of MDR bacterial infec- tions in humans.

Finally, we acknowledge the study’s limitations. Firstly, the results should be interpreted considering our con- venience-based sampling strategy, which exclusively captured dog food types and brands available on the Portuguese pet food market, primarily in four cities in the Porto metropolitan area. Consequently, the results may not be extrapolated to pet food products avail- able from other suppliers. Secondly, the small sample size and the uneven distribution among suppliers and various canine food items may have introduced unin- tended selection bias. Moreover, additional studies, encompassing brands available in every region in the world, along with local risk assessment investigations, are required to discern the broader implications of pet food on public health.

Conclusion This study demonstrates that RMBDs from European brands available in Portugal can be a vehicle for MDR clinically-relevant Salmonella and E. coli carrying genes encoding resistance to the last-resort antibiotic colistin. Promoting awareness of potential risks linked to RMBDs and providing guidance to pet owners on proper handling and feeding practices are crucial steps in minimising potential health risks. Identifying environmental transmission routes of pathogenic and MDR bacteria to pet food and the continuous microbiological monitoring (pathogens and antibiotic- resistant bacteria/genes) of ingredients and processes used in the fast-growing pet food industry needs to be addressed in future One Health studies to mitigate public health risks.

Ethical statement Ethical approval was not required for this study because the samples were exclusively from dog food acquired in retail stores.

Funding statement This work was supported by the Applied Molecular Biosciences Unit - UCIBIO which is financed by national funds from FCT - Fundação para a Ciência e a Tecnologia [UIDP/04378/2020 and UIDB/04378/2020, by the Associate Laboratory Institute for Health and Bioeconomy-i4HB [LA/P/0140/2020] and by the AgriFood XXI I&D&I project [NORTE-01- 0145-FEDER-000041] co-financed by the European Regional Development Fund (ERDF) through NORTE 2020 (Programa Operacional Regional do Norte 2014/2020). Marisa Ribeiro- Almeida was supported by a PhD fellowship from FCT (SFRH/ BD/146405/2019). Joana Mourão (2021.03416.CEECIND) and Ana R Freitas (CEECIND/02268/2017) were supported by national funds through FCT/MCTES in the context of the Scientific Employment Stimulus. The authors are greatly in- debted to all the financing sources.

Data availability Raw reads from E. coli strains R26_1_100 (Uberstrain num- ber ESC_YA6262AA) and R52_8 (Uberstrain number ESC_ YA9779AA), as well as from Salmonella enterica Typhimurium monophasic variant 1,4,[5],12:i:- strain R54_S100 (Uberstrain

10 www.eurosurveillance.org

number SAL_LB4229AA), were deposited at Enterobase (htt- ps://enterobase.warwick.ac.uk).

Preprint A pre-print version of our manuscript, titled “Raw meat- based diet for pets: a neglected source of human exposure to Salmonella and mcr-carrying Escherichia coli in Europe” by Ribeiro-Almeida, Marisa, et al., has been deposited on SSRN, and the corresponding DOI for the preprint is http:// dx.doi.org/10.2139/ssrn.4532439, as posted on 9 August 2023.

Acknowledgements The authors would like to thank Lina Cavaco, Yang Wang, Alessandra Carattoli, and Maria Borowiak for providing the mcr-positive controls.

Conflict of interest None declared.

Authors’ contributions Marisa Ribeiro-Almeida: Methodology, Software, Formal analysis, Investigation, Visualisation, Writing – origi- nal draft, Writing – review & editing. Joana Mourão: Methodology, Software, Formal analysis, Investigation, Visualisation, Writing – review & editing. Mafalda Magalhães: Investigation, Visualisation, Writing – original draft. Ana R Freitas: Formal analysis, Writing – review & edit- ing. Carla Novais: Formal analysis, Writing – review & edit- ing, Funding acquisition. Luísa Peixe: Supervision, Funding acquisition, Writing – review & editing. Patrícia Antunes: Conceptualisation, Methodology, Software, Formal analysis, Investigation, Supervision, Writing – original draft, Writing – review & editing, Funding acquisition, Project administration.

References 1. Davies RH, Lawes JR, Wales AD. Raw diets for dogs and cats: a

review, with particular reference to microbiological hazards. J Small Anim Pract. 2019;60(6):329-39. https://doi.org/10.1111/ jsap.13000 PMID: 31025713

2. Freeman LM, Chandler ML, Hamper BA, Weeth LP. Current knowledge about the risks and benefits of raw meat-based diets for dogs and cats. J Am Vet Med Assoc. 2013;243(11):1549-58. https://doi.org/10.2460/ javma.243.11.1549 PMID: 24261804

3. Runesvärd E, Wikström C, Fernström L-L, Hansson I. Presence of pathogenic bacteria in faeces from dogs fed raw meat-based diets or dry kibble. Vet Rec. 2020;187(9):e71. https://doi. org/10.1136/vr.105644 PMID: 32054718

4. Solís D, Toro M, Navarrete P, Faúndez P, Reyes-Jara A. Microbiological quality and presence of foodborne pathogens in raw and extruded canine diets and canine fecal samples. Front Vet Sci. 2022;9:799710. https://doi.org/10.3389/ fvets.2022.799710 PMID: 35923819

5. Vecchiato CG, Schwaiger K, Biagi G, Dobenecker B. From nutritional adequacy to hygiene quality: A detailed assessment of commercial raw pet-food for dogs and cats. Animals (Basel). 2022;12(18):2395. https://doi.org/10.3390/ani12182395 PMID: 36139257

6. European Pet Food Industry (FEDIAF). Guide to good practice for the manufacture of safe pet foods. Brussels: FEDIAF; 2018. Available from: http://europeanpetfood.org/wp-content/ uploads/2022/03/FEDIAF_Safety_Guide_February_2018_ online.pdf

7. Europe Commission. Commission Regulation (EU) No 142/2011 implementing Regulation (EC) No 1069/2009 of the European Parliament and of the Council laying down health rules as regards animal by-products and derived products not intended

for human consumption and implementing Council Directive 97/78/EC as regards certain samples and items exempt from veterinary checks at the border under that Directive. Official Journal of the European Union. 26.2.2011:L 54/1. Available from: http://www.eur-lex.europa.eu/LexUriServ/LexUriServ.do ?uri=OJ:L:2011:054:0001:0254:EN:PDF

8. Rapid Alert System for Food and Feed (RASFF). RASFF Window. Brussels: European Commission. [Accessed: 12 Feb 2024]. Available from: https://webgate.ec.europa.eu/rasff-window/ screen/search

9. Kaindama L, Jenkins C, Aird H, Jorgensen F, Stoker K, Byrne L. A cluster of Shiga toxin-producing Escherichia coli O157:H7 highlights raw pet food as an emerging potential source of infection in humans. Epidemiol Infect. 2021;149(e124):e124. https://doi.org/10.1017/S0950268821001072 PMID: 33955833

10. Russini V, Corradini C, Rasile E, Terracciano G, Senese M, Bellagamba F, et al. A familiar outbreak of monophasic Salmonella serovar Typhimurium (ST34) involving three dogs and their owner’s children. Pathogens. 2022;11(12):1500. https://doi.org/10.3390/pathogens11121500 PMID: 36558834

11. Jones JL, Wang L, Ceric O, Nemser SM, Rotstein DS, Jurkovic DA, et al. Whole genome sequencing confirms source of pathogens associated with bacterial foodborne illness in pets fed raw pet food. J Vet Diagn Invest. 2019;31(2):235-40. https://doi.org/10.1177/1040638718823046 PMID: 30663530

12. Hamame A, Davoust B, Cherak Z, Rolain JM, Diene SM. Mobile colistin resistance (mcr) genes in cats and dogs and their zoonotic transmission risks. Pathogens. 2022;11(6):698. https://doi.org/10.3390/pathogens11060698 PMID: 35745552

13. Menezes J, Moreira da Silva J, Frosini SM, Loeffler A, Weese S, Perreten V, et al. mcr-1 colistin resistance gene sharing between Escherichia coli from cohabiting dogs and humans, Lisbon, Portugal, 2018 to 2020. Euro Surveill. 2022;27(44):2101144. https://doi.org/10.2807/1560-7917. ES.2022.27.44.2101144 PMID: 36330821

14. Nüesch-Inderbinen M, Treier A, Zurfluh K, Stephan R. Raw meat-based diets for companion animals: a potential source of transmission of pathogenic and antimicrobial-resistant Enterobacteriaceae. R Soc Open Sci. 2019;6(10):191170. https://doi.org/10.1098/rsos.191170 PMID: 31824726

15. Nilsson O. Hygiene quality and presence of ESBL-producing Escherichia coli in raw food diets for dogs. Infect Ecol Epidemiol. 2015;5(1):28758. https://doi.org/10.3402/iee. v5.28758 PMID: 26490763

16. van Bree FPJ, Bokken GCAM, Mineur R, Franssen F, Opsteegh M, van der Giessen JWB, et al. Zoonotic bacteria and parasites found in raw meat-based diets for cats and dogs. Vet Rec. 2018;182(2):50. https://doi.org/10.1136/vr.104535 PMID: 29326391

17. Finisterra L, Duarte B, Peixe L, Novais C, Freitas AR. Industrial dog food is a vehicle of multidrug-resistant enterococci carrying virulence genes often linked to human infections. Int J Food Microbiol. 2021;358(109284):109284. https://doi. org/10.1016/j.ijfoodmicro.2021.109284 PMID: 34144837

18. International Organization for Standardization (ISO). ISO 6579–1:2017. Microbiology of the food chain Horizontal method for the detection, enumeration and serotyping of Salmonella. Part 1: Detection of Salmonella spp. Geneva: ISO; 2022. Available from: https://www.iso.org/standard/56712.html

19. Mourão J, Rebelo A, Ribeiro S, Peixe L, Novais C, Antunes P. Atypical non-H2S-producing monophasic Salmonella Typhimurium ST3478 strains from chicken meat at processing stage are adapted to diverse stresses. Pathogens. 2020;9(9):701. https://doi.org/10.3390/pathogens9090701 PMID: 32859122

20. Ribeiro S, Mourão J, Novais Â, Campos J, Peixe L, Antunes P. From farm to fork: Colistin voluntary withdrawal in Portuguese farms reflected in decreasing occurrence of mcr-1-carrying Enterobacteriaceae from chicken meat. Environ Microbiol. 2021;23(12):7563-77. https://doi.org/10.1111/1462-2920.15689 PMID: 34327794

21. Mourão J, Ribeiro-Almeida M, Novais C, Magalhães M, Rebelo A, Ribeiro S, et al. From farm to fork: Persistence of clinically relevant multidrug-resistant and copper-tolerant Klebsiella pneumoniae long after colistin withdrawal in poultry production. Microbiol Spectr. 2023;11(4):e0138623. https:// doi.org/10.1128/spectrum.01386-23 PMID: 37428073

22. European Committee on Antimicrobial Susceptibility Testing (EUCAST). Recommendations for MIC determination of colistin (polymyxin E)—as recommended by the joint CLSI-EUCAST Polymyxin Breakpoints Working Group. Växjö: EUCAST. [Accessed: 5 Jan 2023]. Available from: https://www.eucast. org/fileadmin/src/media/PDFs/EUCAST_files/General_ documents/Recommendations_for_MIC_determination_of_ colistin_March_2016.pdf

11www.eurosurveillance.org

23. European Committee on Antimicrobial Susceptibility Testing (EUCAST). Breakpoint tables for interpretation of MICs and zone diameters, version 13.0. Växjö: EUCAST. [Accessed: 5 Jan 2022]. Available from: https://www.eucast.org/fileadmin/ src/media/PDFs/EUCAST_files/Breakpoint_tables/v_13.0_ Breakpoint_Tables.pdf

24. Clinical and Laboratory Standards Institute (CLSI). M100. Performance standards for antimicrobial susceptibility testing. 32nd ed. Wayne: CLSI; 2022. Available from: https://clsi.org/ media/wi0pmpke/m100ed32_sample.pdf

25. Fujioka M, Otomo Y, Ahsan CR. A novel single-step multiplex polymerase chain reaction assay for the detection of diarrheagenic Escherichia coli. J Microbiol Methods. 2013;92(3):289-92. https://doi.org/10.1016/j. mimet.2012.12.010 PMID: 23270615

26. Mourão J, Marçal S, Ramos P, Campos J, Machado J, Peixe L, et al. Tolerance to multiple metal stressors in emerging non-typhoidal MDR Salmonella serotypes: a relevant role for copper in anaerobic conditions. J Antimicrob Chemother. 2016;71(8):2147-57. https://doi.org/10.1093/jac/dkw120 PMID: 27118781

27. Bottari B, Bancalari E, Barera A, Ghidini S, Gatti M. Evaluating the presence of human pathogens in commercially frozen, biologically appropriate raw pet food sold in Italy. Vet Rec. 2020;187(7):e50. https://doi.org/10.1136/vr.105893 PMID: 32430390

28. Campos J, Mourão J, Peixe L, Antunes P. Non-typhoidal Salmonella in the pig production chain: a comprehensive analysis of its impact on human health. Pathogens. 2019;8(1):19. https://doi.org/10.3390/pathogens8010019 PMID: 30700039

29. European Food Safety AuthorityEuropean Centre for Disease Prevention and Control. The European Union Summary Report on Antimicrobial Resistance in zoonotic and indicator bacteria from humans, animals and food in 2018/2019. EFSA J. 2021;19(4):e06490. PMID: 33868492

30. Zeng H, De Reu K, Gabriël S, Mattheus W, De Zutter L, Rasschaert G. Salmonella prevalence and persistence in industrialized poultry slaughterhouses. Poult Sci. 2021;100(4):100991. https://doi.org/10.1016/j.psj.2021.01.014 PMID: 33610890

31. Finley R, Ribble C, Aramini J, Vandermeer M, Popa M, Litman M, et al. The risk of salmonellae shedding by dogs fed Salmonella-contaminated commercial raw food diets. Can Vet J. 2007;48(1):69-75. PMID: 17310625

32. Groat EF, Williams NJ, Pinchbeck G, Warner B, Simpson A, Schmidt VM. UK dogs eating raw meat diets have higher risk of Salmonella and antimicrobial-resistant Escherichia coli faecal carriage. J Small Anim Pract. 2022;63(6):435-41. https://doi. org/10.1111/jsap.13488 PMID: 35191029

33. Centers for Disease Control and Prevention (CDC). Outbreak of multidrug-resistant Salmonella infections linked to contact with pig ear pet treats. Washington, D.C.: CDC; 2019. Available from: https://www.cdc.gov/salmonella/pet-treats-07-19/index. html

34. Centers for Disease Control and Prevention (CDC). Multidrug- resistant salmonella infections linked to raw turkey products. Washington, D.C.: CDC; 2019. Available from: https://www.cdc. gov/salmonella/reading-07-18/index.html

35. Bacci C, Vismarra A, Dander S, Barilli E, Superchi P. Occurrence and antimicrobial profile of bacterial pathogens in former foodstuff meat products used for pet diets. J Food Prot. 2019;82(2):316-24. https://doi.org/10.4315/0362-028X.JFP-18- 352 PMID: 30688534

36. Campos J, Gil J, Mourão J, Peixe L, Antunes P. Ready-to- eat street-vended food as a potential vehicle of bacterial pathogens and antimicrobial resistance: An exploratory study in Porto region, Portugal. Int J Food Microbiol. 2015;206:1- 6. https://doi.org/10.1016/j.ijfoodmicro.2015.04.016 PMID: 25910073

37. Hadjadj L, Riziki T, Zhu Y, Li J, Diene SM, Rolain J-M. Study of mcr-1 gene-mediated colistin resistance in Enterobacteriaceae isolated from humans and animals in different countries. Genes (Basel). 2017;8(12):394. https://doi.org/10.3390/ genes8120394 PMID: 29257080

38. Bulochova V, Evans EW. Exploring food safety perceptions and self-reported practices of pet owners providing raw meat– based diets to pets. J Food Prot. 2021;84(5):912-9. https://doi. org/10.4315/JFP-20-338 PMID: 33428742

License, supplementary material and copyright This is an open-access article distributed under the terms of the Creative Commons Attribution (CC BY 4.0) Licence. You may share and adapt the material, but must give appropriate credit to the source, provide a link to the licence and indicate if changes were made.

Any supplementary material referenced in the article can be found in the online version.

This article is copyright of the authors or their affiliated in- stitutions, 2024.