Insights into microbial contamination and antibiotic resistome traits in pork wholesale market: An evaluation of the disinfection effect of sodium hypochlorite

Abstract

Chlorine and its derivatives, such as sodium hypochlorite (NaClO) and chlorine dioxide, are frequently employed
as disinfectants throughout the pork supply chain in China. Nevertheless, the extensive use of NaClO has the
potential to cause the creation of ’chlorine-tolerant bacteria’ and accelerate the evolution of antibiotic resistance.
This study evaluated the efficacy of NaClO disinfection by examining alterations in the microbiome and resistome of a pork wholesale market (PWM), and bacteria isolation and analysis were performed to validate the
findings. As expected, the taxonomic compositions of bacteria was significantly different before and after
disinfection. Notably, Salmonella enterica (S. enterica), Salmonella bongori (S. bongori), Escherichia coli (E. coli),
Klebsiella pneumoniae (K. pneumoniae), and Pseudomonas aeruginosa (P. aeruginosa) were observed on all surfaces,
indicating that the application of NaClO disinfection treatment in PWM environments for pathogenic bacteria is
limited. Correlations were identified between antibiotic resistance genes (ARGs) associated with aminoglycosides
(aph(3′’)-I, aph(6′)-I), quinolone (qnrB, abaQ), polymyxin (arnA, mcr-4) and disinfectant resistance genes (emrA/
BD, mdtA/B/C/E/F). Furthermore, correlations were found between risk Rank I ARGs associated with amino-
glycoside (aph(3′)-I), tetracycline (tetH), beta_lactam (TEM-171), and disinfectant resistance genes (mdtB/C/E/F,
emrA, acrB, qacG). Importantly, we found that Acinetobacter and Salmonella were the main hosts of disinfectant
resistance genes. The resistance mechanisms of the ARGs identified in PWM were dominated by antibiotic
deactivation (38.7%), antibiotic efflux (27.2%), and antibiotic target protection (14.4%). The proportion of
genes encoding efflux pumps in the PWM resistome increased after disinfection. Microbial cultures demonstrated
that the traits of microbial contamination and antibiotic resistane were consistent with those observed by
metagenomic sequencing. This study highlights the possibility of cross-resistance between NaClO disinfectants
and antibiotics, which should not be ignored.

Full text

Journal of Hazardous Materials 468 (2024) 133811
Available online 16 February 2024
0304-3894/© 2024 Elsevier B.V. All rights reserved.
Insights into microbial contamination and antibiotic resistome traits in pork
wholesale market: An evaluation of the disinfection effect of
sodium hypochlorite
Xingning Xiao
a
, Miao He
a
,
b
, Lingyan Ma
a
, Wentao Lv
a
, Kang Huang
c
, Hua Yang
a
, Yanbin Li
d
,
Likou Zou
e
,
**
, Yingping Xiao
a
,
*
, Wen Wang
a
,
*
a
State Key Laboratory for Managing Biotic and Chemical Threats to the Quality and Safety of Agro-products, MOA Laboratory of Quality & Safety Risk Assessment for
Agro-products (Hangzhou), Institute of Agro-product Safety and Nutrition, Zhejiang Academy of Agricultural Sciences, Hangzhou, China
b
College of Food and Pharmaceutical Sciences, Ningbo University, Ningbo, China
c
Biological Systems Engineering, Washington State University, Pullman, USA
d
Department of Biological & Agricultural Engineering, University of Arkansas, Fayetteville, AR 72701, USA
e
College of Resources and Environment, Sichuan Agricultural University, Chengdu, China
HIGHLIGHTS GRAPHICAL ABSTRACT
•Taxonomic compositions of bacterial
were changed before and after
disinfection.
•The application of NaClO disinfection
treatment on pathogenic bacteria was
limited.
•The Acinetobacter and Salmonella were
main hosts of disinfectant resistance
genes.
•Signicant correlations were found be-
tween ARGs and disinfectant resistance
genes.
•The proportion of genes encoding efux
pumps increased in the PMW
disinfection.
ARTICLE INFO
Keywords:
Pork wholesale market
NaClO
Chlorine tolerance
ABSTRACT
Chlorine and its derivatives, such as sodium hypochlorite (NaClO) and chlorine dioxide, are frequently employed
as disinfectants throughout the pork supply chain in China. Nevertheless, the extensive use of NaClO has the
potential to cause the creation of ’chlorine-tolerant bacteria’ and accelerate the evolution of antibiotic resistance.
This study evaluated the efcacy of NaClO disinfection by examining alterations in the microbiome and
* Correspondence to: 198 Shiqiao Road, Hangzhou 310021, China.
** Correspondence to: 211 Huimin Road, Chengdu 611130, China.
E-mail addresses: zoulikou@sicau.edu.cn (L. Zou), xiaoyp@zaas.ac.cn (Y. Xiao), ww_hi1018@163.com (W. Wang).
Contents lists available at ScienceDirect
Journal of Hazardous Materials
journal homepage: www.elsevier.com/locate/jhazmat
https://doi.org/10.1016/j.jhazmat.2024.133811
Received 28 December 2023; Received in revised form 6 February 2024; Accepted 14 February 2024

Journal of Hazardous Materials 468 (2024) 133811
2
Metagenomics
Antimicrobial resistance
resistome of a pork wholesale market (PWM), and bacteria isolation and analysis were performed to validate the
ndings. As expected, the taxonomic compositions of bacteria was signicantly different before and after
disinfection. Notably, Salmonella enterica (S. enterica), Salmonella bongori (S. bongori), Escherichia coli (E. coli),
Klebsiella pneumoniae (K. pneumoniae), and Pseudomonas aeruginosa (P. aeruginosa) were observed on all surfaces,
indicating that the application of NaClO disinfection treatment in PWM environments for pathogenic bacteria is
limited. Correlations were identied between antibiotic resistance genes (ARGs) associated with aminoglycosides
(aph(3
′
’)-I, aph(6
′
)-I), quinolone (qnrB, abaQ), polymyxin (arnA, mcr-4) and disinfectant resistance genes (emrA/
BD, mdtA/B/C/E/F). Furthermore, correlations were found between risk Rank I ARGs associated with amino-
glycoside (aph(3
′
)-I), tetracycline (tetH), beta_lactam (TEM-171), and disinfectant resistance genes (mdtB/C/E/F,
emrA, acrB, qacG). Importantly, we found that Acinetobacter and Salmonella were the main hosts of disinfectant
resistance genes. The resistance mechanisms of the ARGs identied in PWM were dominated by antibiotic
deactivation (38.7%), antibiotic efux (27.2%), and antibiotic target protection (14.4%). The proportion of
genes encoding efux pumps in the PWM resistome increased after disinfection. Microbial cultures demonstrated
that the traits of microbial contamination and antibiotic resistane were consistent with those observed by
metagenomic sequencing. This study highlights the possibility of cross-resistance between NaClO disinfectants
and antibiotics, which should not be ignored.
1. Introduction
Food safety pertaining to livestock is of utmost importance, given the
intricate interplay between animals and humans. Approximately 58% of
human infections are classied as zoonotic, indicating that they can
cause disease in both human and animals [1]. The issue of food safety
stems from an intricate network, including the
human-animal-environment interactions and the points at which they
meet. This encompasses the entire process, starting with farming and
slaughter, extending to retail, and culminating with cooking. Human
activities, including industrial and agricultural activities, discharge a
variety of pollutants (i.e., potential pathogenic bacteria, antibiotic
resistant bacteria and other contaminants) to the environment, leading
to potential health risks from the One Health perspective [2–4].
Disinfectants are extensively employed across various domains,
encompassing food processing, healthcare, household, agriculture, ani-
mal husbandry, and environment contexts [5]. Recent statistics indicate
that the global sales of surface sanitizers in 2020 amounted to $4.5
billion, representing a signicant 30% year-on-year growth [6]. Chlo-
rination is highly effective against numerous bacteria, however, it can
lead to the formation of toxic, mutagenic, and/or carcinogenic disin-
fection by-products (DBPs) and residual chlorine throughout the disin-
fection process [7]. The disinfectant agents such as chlorine, iodine and
peroxygens that oxidize various chemical groups (amino, sulfhydryl and
thiol) associated with lipids, proteins and nucleic acids, thus disrupting
major cytoplasmic membrane function, enzyme function and DNA
synthesis. Chlorine-based and iodine-based compounds and peracetic
acid have been associated with membrane damage presumably through
protein oxidation [8].
NaClO is a chlorine-based disinfectant that has been used extensively
because of its wide-ranging antibacterial efcacy against many patho-
gens [9,10]. The World Health Organization (WHO) suggests using a
1000 mg/L NaClO solution to sanitize household surfaces and disinfect
utility gloves [11]. In food production facilities, the use of NaClO is
prevalent in the treatment of various surfaces and equipment [12].
Nevertheless, during the sanitizing procedure, bacteria can endure and
proliferate despite the presence of residual disinfectants, thereby earn-
ing the designation of being tolerant [13,14]. In addition, evidence
shows that disinfectants can lead to the emergence of cross-resistance to
clinically important antibiotic compounds [15]. It has been observed
that disinfectants have the potential to exacerbate the dissemination of
antibiotic resistance genes (ARGs). This is attributed to the generation of
selective pressure, which increases the likelihood of horizontal gene
transfer (HGT) within and between bacterial genera [16,17]. In partic-
ular, the persistence of antibiotic resistance in environmental bacteria,
despite the absence of direct antibiotic usage, has raised concerns
regarding the development of cross-resistance between chlorine and
antibiotics [18,19].
In China, approximately 70–80% of agro-food is distributed through
wholesale markets [20,21]. Wholesale markets serve as intermediaries
between slaughterhouses and small-scale retailers in the pork supply
chain, as illustrated in Fig. 1. Hog carcasses are halved and transported
to wholesale markets. After the commencement of trading in the market,
the corpses are transported by carts from vehicles to stalls, where they
are later divided into portions and sold to retailers. The environment in
which raw meat is sold and handled can serve as a potential medium for
transmission of resistance. One study investigated the surface hygiene
levels of wooden chopping boards used for processing meat in Hong
Kong and detected several common resistance genes [22]. Typically, the
PWMs are open during the late hours of the night and undergo disin-
fection procedures upon closure in the early morning (Figs. 1A and 1B).
Disinfection serves to eliminate pathogenic microorganisms from the
environment, curbing cross-contamination, and enhancing the safety of
pork food to a certain extent [23]. In the pork wholesale marketplaces in
China, a widely used disinfectant called ’84 disinfection’ has been
extensively utilized to sanitize various contact surfaces, including the
ground, vehicle wheels, and carts. This disinfectant comprises NaClO
and surface- active agents [24]. However, the potential effects of NaClO
disinfection on microbial contamination and antibiotic resistome re-
mains to be evaluated.
In this study, we investigated microbial contamination and antibiotic
resistome traits before and after disinfection in a PWM environment
using metagenomic sequencing, and bacteria isolation and analysis were
conducted to validate the ndings (Fig. 1C). This study aimed to gain
insight into the inuence of NaClO disinfection on the risk of pathogenic
bacteria and ARG transmission. The main scientic investigations
involved two crucial aspects: (1) assessing the inuence of NaClO
disinfectant on bacterial communities and ARGs; and (2) examining the
potential presence of cross-resistance between NaClO disinfectant
resistance genes and ARGs.
2. Materials and methods
2.1. Sampling
Samples were collected from a wholesale pork market in Hangzhou,
China, which processes an estimated 10,000 hogs daily. The sampling
procedure was conducted within a single day of market activity to en-
sures temporal and spatial microbial consistency, leading to the acqui-
sition of a total of 104 samples. As shown in Table 1, 62 samples were
collected during the trading period from various sources, including
external transport vehicles (WA, n=6), local transport vehicles (WB,
n=6), carts (WC, n=12), pork hooks (WD, n=3), meat (WE, n=10),
knives (WF, n=10), chopping boards (WG, n=10), and oors (WH,
n=5). In this market, following the transaction closure, a regular
disinfection operation was implemented. This involved spraying NaClO
X. Xiao et al.

Journal of Hazardous Materials 468 (2024) 133811
3
solution with a chlorine concentration of 500 mg/L over the surfaces of
the surroundings and facilities. After market disinfection, 42 samples
were collected from external transport vehicles (BA, n=6), local
transport vehicles (BB, n=6), carts (BC, n=12), pork hooks (BD,
n=3), chopping boards (BF, n=10), and oors (BH, n=5). Meat and
knives were sampled at night during the trading period. The time lapse
between the collection of the environmental samples before and after
disinfection was approximately 12 h.
Sampling was performed on surfaces with suitable accessibility such
as oors and chopping boards, covering an area of approximately 1 m
2
.
The process entailed systematically wiping the surfaces, beginning with
horizontal strokes, followed by vertical strokes, and nally diagonal
strokes. A swab was spun between each stroke to ensure thorough
sampling. In cases where it was not feasible to swab an entire surface
area of 1 m
2
, for example, with pork hooks or knives, individual units
such as a single pork hook or knife were swabbed. Samples were
collected using sterile absorbent gauze pre-moistened with phosphate-
buffered saline (1 ×PBS, pH 7.2). Single-use disposable gloves were
Fig. 1. Flow charts of this study. (A) Pork wholesale markets conduct trade operations from 12 pm to 8 am the following morning. (B) Disinfection with NaClO after
closing at 9 am. (C) Culture-dependent and independent characterization.
X. Xiao et al.

Journal of Hazardous Materials 468 (2024) 133811
4
used during sampling and the gloves were changed after each sample
was collected to avoid cross-contamination. The swabs collected for the
study were immersed in tubes containing 15 mL of PBS and stored in a
cooling box with ice packs. The transportation of the samples to the
laboratory was completed within a maximum of 2 h to ensure prompt
processing.
2.2. Culture independent methods
2.2.1. DNA extraction
Microbial DNA of all 42 samples was extracted using a E.Z.N.A.®
DNA Kit (Omega Bio-tek, Norcross, GA, USA) according to the manu-
facturer’s instructions. Twenty-four samples were selected before the
environment was disinfected, including the following samples: three
external transport vehicles (WA), three local transport vehicles (WB),
three carts (WC), three pork hooks (WD), three meats (WE), three knives
(WF), three chopping boards (WG), and three oors (WH). The 18
remaining samples were collected after disinfection and included three
external transport vehicles (BA), three local transport vehicles (BB),
three carts (BC), three pork hooks (BD), three chopping boards (BF), and
three oors (BH). Swab samples were individually placed into tubes
containing 15 mL PBS and centrifuged at 10,000 rpm for 5 min. The
supernatants were passed through 0.22 µm polycarbonate membranes
(GE Osmonics, Minnetonka, MN, USA) and preserved at −80 ◦C until
extraction. DNA concentrations were determined using a Nanodrop One
spectrophotometer (Thermo Fisher Scientic, Waltham, MA, USA) and
DNA quality was veried by 1% agarose gel electrophoresis. Samples
collected in triplicate from each site were used for DNA extraction and
metagenomic sequencing [25].
2.2.2. Metagenomic sequencing
Metagenomic sequencing libraries were obtained from Shanghai
Biozeron Biological Technology Co. Ltd. (Shanghai, China). All samples
were sequenced using an Illumina NovaSeq 6000 (Illumina, San Diego,
California, USA). Adaptor contaminants and low-quality reads were
removed using the Trimmomatic software (V 0.32). Bwa2 was used to
remove the host DNA. The reference genome was the Susscrofa genome
assembly Sscrofa11.1. Reads with host genome contamination and low-
quality data are called clean data and were used for further analyes [26].
2.2.3. Metagenome sequencing data analysis
For taxonomic classication, the reads were cleaned using Kraken2
(v2.0.7) with the NCBI taxonomic ID and a customized complete
genome k-mer database [27]. The classication results were further
passed through Bracken 2.0 [28]. For ARG quantication, the reads
were used as inputs to the ARG-analysis pipeline ARGs-OAP v2.0, inte-
grating the detection of ARGs using the SARG v2.0 reference database,
as described in the studies conducted by Yang et al. [29] and Yin et al.
[30]. ARGs were normalized to the number of 16 S rRNA genes
expressed as copies per 16 S rRNA gene. To measure the abundance of
mobile genetic elements (MGEs), an antibiotic resistance gene-oriented
assembly pipeline was used, with the reference database replaced by a
recently released MGE database [31]. The threshold for gene annotation
was greater than 90% and the matching length between the reads and
the reference sequence was greater than 25 amino acids.
The alpha diversity index of the microbial communities in the sam-
ples was calculated using the R package vegan according to the anno-
tated species abundance table obtained at the species level, including
the Chao1 and Shannon indices. The alpha diversity index of ARGs in the
samples was examined to see the effect of disinfection on the composi-
tion of the resistome. Principal coordinate analysis (PCoA) based on
Bray–Curtis metrics was used to determine the beta-diversity of the
bacterial community distribution among groups. A stacked column of
microbial communities and ARG abundances was generated using
GraphPad Prism software (version 8.0). Heatmaps of the correlation
between ARG and disinfectant-resistance gene abundance were gener-
ated using the corrplot package in R software (version 3.4.1). The
Spearman correlation coefcient of the ARG types was calculated to
analyze the correlation. Network visualization was performed using
Cytoscape 3.8.2 and Gephi 0.9.2. Sankey plots generated by the RAW-
Graphs visualization program were used for visualization to represent
the abundance results for different types of risk rank ARGs [32]. Raw
metagenomic sequencing data were deposited in the National Center for
Biotechnology Information (NCBI) database under the project accession
number PRJAN954132.
2.3. ARGs health risk evaluation
The health risk assessment of ARGs was evaluated according to the
Table 1
Prevalence, antibiotic and NaClO resistance of bacteria taken from different sampling sites.
Sampling
time
Sampling sites Symbol No. of
samples
E. coli Salmonella
Prevalence
(%)
a
MDR
(%)
b
NaClO tolerance
(%)
Prevalence
(%)
a
MDR
(%)
b
NaClO tolerance
(%)
Night External transport
vehicle
WA 6 2 (33.3%) 2 (100%) 1 (50%) 0 (0%) _ _
Local transport
vehicle
WB 6 1 (16.67%) 1 (100%) 0 (0%) 0 (0%) _ _
Carts WC 12 6 (50%) 6 (100%) 0 (0%) 0 (0%) _ _
Pork hook WD 3 0 (0%) _ _ 0 (0%) _ _
Meat WE 10 7 (70%) 7 (100%) 4 (57.1%) 1 (10%) 1 (100%) 1 (100%)
Chopping board WF 10 9 (90%) 9 (100%) 2 (22.2%) 1 (10%) 1 (100%) 1 (100%)
Knives WG 10 5 (50%) 4 (80%) 4 (80%) 2 (20%) 2 (100%) 2 (100%)
Floor WH 5 2 (40%) 2 (100%) 0 (0%) 1 (20%) 1 (100%) 1 (100%)
Total — 62 32 (51.6%) 31
(96.9%)
11(34.4%) 5 (8.1%) 5 (100%) 5 (100%)
Day External transport
vehicle
BA 6 0 (0%) _ _ 1 (16.67%) 1 (100%) 1 (100%)
Local transport
vehicle
BB 6 1 (16.67%) 1 (100%) 0 (0%) 0 (0%) _ _
Carts BC 12 5 (41.67%) 5 (100%) 1 (20%) 0 (0%) _ _
Pork hook BD 3 0 (0%) _ _ 0 (0%) _ _
Chopping board BF 10 4(40%) 3 (75%) 0 (0%) 2 (20%) 2 (100%) 2 (100%)
Floor BH 5 0 (0%) _ _ 0 (0%) _ _
Total — 42 10 (23.8%) 9 (90%) 1 (10%) 3(7.1%) 3(100%) 3(100%)
a
MDR: multidrug resistance was dened as resistance to at least three antibiotics;
b
NaClO tolerance: tolerance to NaClO was dened as MIC >128 mg.
X. Xiao et al.

Journal of Hazardous Materials 468 (2024) 133811
5
criteria proposed by Zhang et al. [33] according to these perspectives:
(1) enrichment in human-associated environments; (2) gene mobility;
(3) presence in host pathogenicity. ARGs were classied into four ranks
ARGs: (1) Rank IV (lowest risk, Q4): ARGs that do not meet the rst
criterion; (2) Rank III (Q3): ARGs that meet the rst, but not the second;
(3) Rank II (Q2): ARGs that meet the rst and second but not the third;
(4) Rank I (the highest risk, Q1): ARGs that meet all three criteria.
2.4. Culture dependent methods
To further verify the metagenome results, representative isolates
(Escherichia coli (E. coli) and Salmonella) were recovered from 104
samples to evaluate the efcacy of the NaClO disinfectant in the PWM
environment. Additionally, the isolates of E. coli and Salmonella were
used to explore cross-resistance between NaClO disinfectant and anti-
biotics using microbial culture methods.
2.4.1. Microbial isolation
To isolate and identify E. coli, the following methods were employed:
Gram staining, assessment of growth characteristics of nutrients, Mac-
Conkey and eosin methylene blue agar (Merck, Darmstadt, Germany),
and biochemical testing, including sugar fermentation, indole produc-
tion, methyl red test, Voges–Proskauer test, and citrate utilization test.
The isolates were subjected to molecular validation using PCR targeting
the phoA gene. This process involved the use of a primer set specic for
E. coli as previously described [34]. To facilitate the isolation and
identication of Salmonella, samples were selectively enriched in Tet-
rathionate and Rappaport–Vassiliadis Soya broth (Hopebio, Qingdao,
China). Next, the suspensions were applied to xylose lysine Tergitol 4
agar (XLT4, Beckton Dickinson, Franklin Lakes, NJ, USA). Isolates with
typical Salmonella phenotypes were further conrmed using API 20E test
strips (bioMerieux, Marcy-l
′
Etoile, France). The molecular verication
of the presumptive isolates was conducted by amplifying the invA gene,
following the methodology reported in a previous study by Zhang et al.
[35].
2.4.2. NaClO tolerance determination
The broth microdilution method was employed to determine the
minimum inhibitory concentrations (MIC) of NaClO against the bacte-
rial isolates. The quality control strains employed in this study were
E. coli ATCC 29522 and S. enteritidis CVCC 1806. Bacterial cultures were
generated by reconstituting 3 to 5 individual colonies that had been
incubated overnight on trypticase soy agar (TSA, Becton Dickinson)
plates. These colonies were suspended in 3 mL of 0.9% saline solution,
achieving a turbidity comparable to that of a 0.5 McFarland standard.
Subsequently, these suspensions were further diluted at a ratio of 1:100
in Mueller Hinton broth (MH, Becton Dickinson). Stock NaClO solutions
with a chlorine concentration of 56.8 mg/mL were obtained from San-
gon Biotech (Shanghai, China). These solutions were prepared by dilu-
tion with sterile Milli-Q water obtained from PALL (Buckinghamshire,
UK). The chlorine concentrations were determined using a Palintest
ChlorSense meter (Gateshead, UK). Preliminary tests revealed that the
MIC of E. coli ATCC 29522 and S. enteritidis CVCC 1806 towards NaClO
was determined to be 128 mg/L. The tolerance to NaClO was dened as
>128 mg/L, as previously published by Luo et al. [36]. The MIC ex-
periments for NaClO tolerance were conducted using 96-well microtiter
test plates. Each well contained 100
μ
L of NaClO solution and was
inoculated with 100
μ
L of suspended bacterial cultures, resulting in a
nal inoculum density of 5 Log CFU/mL per well [15]. The plates were
sealed utilizing a perforated plate seal and subjected to incubation at a
temperature of 37 ◦C for a duration of 24 h. The MIC values of NaClO
were documented as the concentration at which no discernible growth
was seen. Positive controls consisted of control wells containing bacteria
without the addition of NaClO, whereas negative controls were repre-
sented by wells without MH broth. The experiment was conducted in
triplicates on separate occasions.
2.4.3. Antibiotic susceptibility testing
Bacterial suspensions derived from plate colonies, as described in
Section 2.4.2, were used to inoculate 96-well plates containing anti-
biotic solutions, using the broth microdilution technique. The com-
mercial Gram-negative antibiotic susceptibility testing panel used in this
study (Biofosun, Fosun Diagnostics, Shanghai, China) consisted of
various antibiotics with their respective resistance breakpoints. These
included aztreonam (ATM ≥16
μ
g/mL), fosfomycin (FOS ≥256
μ
g/
mL), cefotaxime (CTX ≥4
μ
g/mL), meropenem (MEM ≥4
μ
g/mL),
amikacin (AMK ≥64
μ
g/mL), gentamicin (GEN ≥16
μ
g/mL), colistin
(CS ≥2
μ
g/mL), ceftiofur (CEF ≥8
μ
g/mL), ciprooxacin (CIP ≥1
μ
g/
mL), sulfamethoxazole (T/S ≥4/76
μ
g/mL), tetracycline (TET ≥16
μ
g/
mL), tigecycline (TIG ≥8
μ
g/mL), and orfenicol (FFC ≥16
μ
g/mL).
The 2019 Clinical and Laboratory Standards Institute (CLSI) established
breakpoints for each antibacterial medication. Bacteria that exhibited
sensitivity to all 13 drugs were categorized as drug sensitive, whereas
those that showed resistance to one or two antibiotics were designated
as drug resistant. The term ‘multidrug resistance” is operationally
dened as the ability to withstand the effects of a minimum of three
antibiotics, as stated by Mahros et al. [37].
2.5. Statistical analysis
The data were processed using Ofce 2019 software. SPSS 19.0
software was used to analyze differences and correlations between
groups; and the statistical results are presented as the mean ±standard
error (M ±SE). Multivariate analysis of variance (MANOVA) (or one-
way analysis of variance (ANOVA)) was conducted to conrm the
observed differences. Differences were considered statistically signi-
cant at p<0.05 [38].
3. Results and discussion
3.1. Effects of NaClO disinfection on the composition of microbial
communities in PWM
Alpha and beta diversity analyses were conducted to evaluate the
impact of NaClO disinfection on the microbial community in PWM. No
signicant difference was observed in the mean Chao1 index (Fig. 2D),
whereas a decreased mean Shannon index (p<0.001) was found
following NaClO disinfection (Fig. 2A), indicating that the treatment did
not affect the overall species richness but decreased the microbial di-
versity. Interestingly, both Shannon and Chao1 indices for local vehicles
increased following disinfection (Figs. 2B and 2E). PCoA revealed that
the microbial structures in the majority of samples were in close prox-
imity to each other, with the exception of pork hook samples, which
formed a distinct cluster separate from the other samples (Fig. 2C).
Disinfection signicantly effected the microbiological composition of
external transport vehicles, local transport vehicles, carts, pork hooks,
chopping boards, and oors (Fig. S1). It is noteworthy that there were no
signicant differences in the microbiological composition of pork, knife,
and chopping board samples before and after disinfection (p>0.1),
indicating potential cross-contamination (Fig. 2F). Besides, the bacterial
taxonomy on pork meat was highly positively correlated with that of
chopping boards (r =0.99, p<0.05), suggesting that these two sample
types were potential sources of bacterial cross-contamination (Fig. S2).
The results showed that meat portioning played a signicant role in the
transmission of bacteria. Cobo-Díaz et al. [39] provided evidence that
cutting activities constitute the primary sources of harmful microor-
ganisms within the processing environment, showing an observed in-
crease in bacterial resistance to sanitizers, with wooden surfaces
exhibiting a higher level of resistance than stainless steel surfaces. The
study conducted by Xiao et al. [40] provided empirical evidence that a
wooden surface exhibits greater roughness than a steel surface, as
determined by differential interferometry.
The application of NaClO modied the taxonomic compositions in
X. Xiao et al.

Journal of Hazardous Materials 468 (2024) 133811
6
Fig. 2. (A) Shannon index of bacterial ora before (Group W) and after disinfection (Group B). * ** p<0.001. (B) Shannon index of bacterial ora in each sampling
site. * p<0.05, * * p<0.01. ns, no signicant difference. (C) Principal coordinate analysis at each sample type before and after disinfection. (D) Chao1 index of
bacterial ora before (Group W) and after disinfection (Group B). ns, no signicant difference. (E) Chao1 index of bacterial ora in each sampling site. * p<0.05,
* ** p<0.001. ns, no signicant difference. (F) Principal coordinate analysis in meat, chopping boards, and knives before and after disinfection. (G) Relative
abundance of bacterial taxa at the genus level. Bar plot representing the relative abundance of the 17 most relevant bacterial genera (average total abundance >5%).
(H) Relative abundance of pathogenic bacteria at the species level.
X. Xiao et al.

Journal of Hazardous Materials 468 (2024) 133811
7
PWM (Fig. 2G and Fig. S3). As shown in Fig. 2G, seventeen genera were
identied as the most abundant colonizers. Among these, Acinetobacter,
Psychrobacter, Bradyrhizobium, and Moraxella, with average abundances
of 17.1%, 16.5%, 6.8% and 5.3%, respectively, were dominant. After
disinfection, a signicant decrease (p<0.05) was observed in the
relative abundance of Psychrobacter and Bradyrhizobium, with a signi-
cant decrease in Psychrobacter on chopping boards and Bradyrhizobium
on the pork hooks. In contrast, a signicant increase (p<0.05) in the
relative abundance after disinfection was observed for Acinetobacter,
Pseudomonas, Shewanella, with the increase being particularly marked
for Acinetobacter on external transport vehicles and for Pseudomonas and
Shewanella on chopping boards (Fig. 2G). Shewanella has been linked to
food poisoning through its production of tetrodotoxins [41].
This study focused specically on the frequency of pathogenic bac-
teria in the PWM microbiome, as these bacteria pose a risk to human
health [42]. Eleven pathogenic bacteria were identied, as shown in
Fig. 2H. Acinetobacter baumannii (A. baumannii) (3.3%), S. enterica
(1.2%), and E. coli (0.2%) were the most common pathogenic bacteria in
the PWM environment. There was an increase in the abundance of
pathogenic bacteria in the PWM environment, except for the chopping
boards, as indicated by an increase from 1.9% to 4.5%. These results
suggest that the application of NaClO disinfection treatment on patho-
genic bacteria in a PWM environment is limited. It was worth
mentioning that S. enterica, S. bongori, E. coli, K. pneumoniae, and
P. aeruginosa can be observed in all surfaces (Fig. 2H). Among these,
S. enterica, S. bongori, E. coli, and K. pneumoniae belong to the
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0
WA BA WB BB WC BC WD BD WE WF BF WG WH BH
Antibiotic deactivation Cellular protection
Efflux pump
Other/Unknown
Regulator
Resistance mechanism
Carts Chopping board
External vehicle
Floor
Hook Knives
Local vehicle Meat
****
**
**
ns
**
200
250
300
350
ARG numbers
WA BA WB BB WC BC WD BD WE WF BF WG WH BH
Carts Chopping board
External vehicle Floor
Hook Knives
Local vehicle Meat
(A)
***
*
ns
****
ns
1.0
2.0
3.0
ARG abundance
(copies/16S rRNA genes)
WA BA WB BB WC BC WD BD WE WF BF WG WH BH
0
Carts
External vehicle Floor
Hook
es
Local vehicle
Meat Chopping board Knives
(B)
**
ns
**
ns
ns
0
2
4
6
WA BA WB BB WC BC WD BD WE WF BF WG WH BH
MGE abundance
(copies/16S rRNA genes)
Carts
Chopping board
External vehicle
Floor
Hook
Knives
Local vehicle
Meat
BH
OXA-244
OXA-370
rmtB
KPC-16
PER-2
aac(6')-30-aac(6')-Ib'
OXA-325
TEM-148
IND-5
JOHN-1
Type BFBDBCBBBA
mecI
TEM-104
TEM-130
TEM-90
aac(6')-31
OXA-18
ACT-1
tet37
IND-4
AER-1
dfrA22
CTX-M-15
CfxA3
CMY-55
CfxA2
OXA-145
dfrA25
ACT-16
viomycin phosphotransferase
OCH-2
mycinamicin-resistance protein myrB
DNA-binding transcriptional regulator gadX
FONA_6
OXA-5
sul3
Type
Type
Aminoglycoside
Beta_lactam
Sulfonamide
Tetracycline
Trimethoprim
Unclassified
0
1
2
3
4
WA WB WC WD WE WF WG WH
(E) (F)
(D)
(C)
Fig. 3. (A) Detected ARG subtype counts at the sample sites. * p<0.05, * * p<0.01, * ** p<0.001. ns, no signicant difference. (B) Box plot showing the total
ARGs type abundance (copy/16 S rRNA gene) at the sample sites. * p<0.05, * ** p<0.001. ns, no signicant difference. (C) Distribution of MGE abundance across
sampling sites. * p<0.05, * * p<0.01. ns, no signicant difference. (D) Stacked bar plot showing antibiotic resistance mechanisms in each group. (E) Heatmap
showing the elimination of ARGs after disinfection (average total abundance >10%). (F) Heatmap showing the emergence of ARGs after disinfection (average total
abundance >10%).
X. Xiao et al.

Journal of Hazardous Materials 468 (2024) 133811
8
Enterobacteriaceae family and are potential causes of foodborne diseases
and strong signs of contamination after processing [43]. According to
Centers for Disease Control and Prevention, the Enterobacteriaceae
family is composed of gram-negative bacteria with an outer membrane
acting as a barrier against disinfectants [44]. P. aeruginosa has been
documented as a causative agent of skin and wound infections, bacter-
aemia, pneumonia, and has been found to be correlated with an
increased risk of mortality [45]. P. aeruginosa cells damaged by chlorine
disinfectant can enhance antibiotic resistance through increasing the
expression of efux pump [46]. MexEF-OprN efux pump plays a key
role in generation of antibiotic resistance in P. aeruginosa following
chlorine exposure. Hou et al. [47] demonstrated that the expression of
GPx and SOD was signicantly up-regulated in P. aeruginosa after NaClO
exposure as antioxidant enzymes are used to protect bacteria from ROS.
The environment stress not only triggered a positive antioxidant defense
against oxidative damages but also prepared P. aeruginosa to employ a
passive defense via dormancy against disinfectants [48]. It is necessary
to replace the existing widely used disinfectants by developing new
disinfectants or alternatives. Photocatalytic disinfection as an emerging
greener technology, gained much attention due to its high efciency,
strong sterilizing ability, and environmental friendly non-toxic
by-product generation. A novel solar/TiO2/chlorine drinking water
treatment system was developed to enable disinfection of bacterial
contaminants, which can potentially replace chlorination in conven-
tional drinking water treatment [49].
3.2. Effect of disinfection on the composition of the PWM resistome
A total of 659 ARG subtypes belonging to 20 types were detected in
the PWM environmental samples (Fig. S4 and S5), with total abundances
ranging from 0.37 to 2.02 copies/16 S rRNA. A total of 564 and 559
ARGs were identied before and after disinfection, respectively
(Fig. S4). The order of decreasing relative abundance of ARGs was as
follows: multidrug (33.9%), aminoglycosides (16.1%), tetracycline
(14.9%), beta lactam (9.6%), and bacitracin (7.7%) (Fig. S5). No sig-
nicant difference was observed for the mean of Chao1 and Shannon
indexs (Fig. S6A and S6C). Both the Shannon index and Chao1 index for
pork hook decreased following the disinfection (Fig. S6B and S6D). The
surfaces of the external and local transport vehicles exhibited elevated
levels of ARGs in terms of both number and abundance after disinfection
(Fig. 3A-3B). The disinfection of pork hooks with NaClO greatly reduced
the number and abundance of ARGs and MGE (Fig. 3A-3 C). No major
differences were observed for chopping boards in terms of ARG numbers
and MGE abundance (Fig. 3A and 3 C). No signicant differences were
observed on oors for ARG and MGE abundances (Fig. 3B-3 C). The ARG
subtypes were further identied for their mechanisms of resistance, and
these subtypes covered four different resistance mechanisms. The ARG
resistance mechanisms were dominated by antibiotic deactivation
(40.1% before disinfection, 36.7% after disinfection), antibiotic efux
(26.6% before disinfection, 28.1% after disinfection), and antibiotic
target protection (15.2% before disinfection, 13.5% after disinfection).
The proportion of genes encoding efux pumps in the PWM resistome
increased after disinfection (Fig. 3D). In total, 100 ARGs were removed
by chlorination (Table S1). Disinfection may remove ARGs such as mecl,
aac(6
′
)−31, tet37, dfrA22, or sul3 (Fig. 3E). None of the OXA-244, TEM-
148 or rmtB genes were detected before disinfection (Fig. 3F). Yan et al.
[19] investigated the infuence of disinfection protocols on resistance
genes within an isolated enclosure environment and reported that none
of the AcrE, marA, mdtB, or mdtC genes were detected in the soil before
disinfection. Chlorine disinfection not only affects bacterial cells, but it
can also compromise the integrity of extracellular ARGs, although it is
less effective in reducing ARG abundance than inactivating bacterial
cells [50,51]. Lin et al. [52] detected 125 unique ARGs subtypes before
chlorination treatment and found that 119 ARG subtypes decreased by
7.49 ×10
4
to 3.92 ×10
7
copies/100 mL, whereas 6 ARGs subtypes
were enriched by 1.09 to 10.90 fold after chlorination. According to
another study on the effects of chlorine disinfection on the occurrence
and concentration of both extracellular and intracellular ARGs in a
full-scale wastewater treatment, chlorine disinfection can increase the
abundance of both intracellular (7.8 folds) and extracellular (3.8 folds)
ARGs [53]. Other studies have also indicated that chlorination could
signicantly reduce the relative abundance signicantly in reclaimed
wastewater but could not reduce the relative abundance of ARGs in
biolms [54,55].
3.3. Host microbe of disinfectant resistance gene distribution in the PWM
environment
Network inference was employed to investigate the co-occurrence
patterns between disinfectant resistance genes and genera, as well as
between disinfectant resistance genes and pathogenic bacteria. At genus
level, the results revealed that Acinetobacter was signicant hosts of
disinfectant resistance genes because we found the bacteria strongly
associated with most disinfectant resistance genes in samples (Fig. 4 A).
Among pathogenic bacteria at the species level, Salmonella spp. were the
main hosts of disinfectant resistance genes (Fig. 4B). We further plotted
the relative abundance of the bacterial versus the abundance of ARGs
and MGEs from all samples to directly determine the association be-
tween the bacterial and overall resistome outcomes. The scatterplots
showed a association between Acinetobacter and ARG and MGE abun-
dance (Fig. S7). This is particularly relevant as Acinetobacter is respon-
sible for hospital-acquired infections caused by multidrug-resistant
isolates [56], and some reports have previously concluded that raw meat
represents a reservoir of multidrug-resistant Acinetobacter strains [57,
58]. In previous investigations conducted by Ibrahim et al. [59], Liu
et al. [1], Xu et al. [60], and Yang et al. [61], it was observed that Sal-
monella played a signicant role as hosts for ARGs in livestock meat. In
the study of Luo et al. [62] also reported that Acinetobacter and Salmo-
nella was observed as the dominant antibiotic-resistant bacteria and
obtained the most ARGs diversity. These ndings align closely with the
outcomes of the present study. The ndings of this study provided more
evidence supporting the notion that the progression of bacterial com-
munities during disinfection treatment had a signicant role in the
variation of ARGs, particularly in relation to pathogens harboring ARGs
[63,64]. As shown in Fig. 4 C, there were correlations found between
disinfectant resistance genes (emrA/BD, mdtA/B/C/E/F) and ARGs
related to aminoglycosides (aph(3
′
’)-I, aph(6
′
)-I), quinolones (qnrB,
abaQ), and polymyxins (arnA, mcr-4). These ndings align with earlier
studies that have reported on the existence of cross-resistance between
disinfectants and antibiotics [65,66]. Xiong et al. [67] reported that
disinfection resistance gene (qacE) was specically associated with
tetracycline resistance (tetA, tetB, tetM), and aminoglycoside resistance
(aadA, aphA-1). Some researchers have linked increased chlorine toler-
ance with the presence of antimicrobial resistance (AMR) determinants
in bacteria [68], and have described chlorine and antibiotic
cross-resistance [69]. According to Karumathil et al. [70], exposure to
chlorine had the potential to induce bacterial resistance to chloram-
phenicol, sulfonamide, and β-lactam antibiotics. We also investigate the
change of fteen disinfectant resistance genes following the disinfection
[13,19]. All of the 15 genes could be detected in samples before and
following NaClO disinfection. The abundance of disinfectant resistance
genes in external transport vehicles, carts, oors increased after disin-
fection. The abundance of disinfectant resistance genes in pork hook
decreased after disinfection, especially for mdtB/C (Fig. 4D). The
abundance of disinfectant resistance genes in samples were low ( <0.15
copies per 16 S rRNA) and used abundance to directly characterize the
risk of disinfectant resistance genes might inaccuration.
3.4. The health risk assessment of the ARGs
A methodology based on omics was employed to conduct a health
risk evaluation of the antibiotic resistome, resulting in the identication
X. Xiao et al.

Journal of Hazardous Materials 468 (2024) 133811
9
of 647 high risk (Q1, Q2) ARGs (Fig. S8). It was observed that ARGs with
high risk were primarily found in aminoglycosides, β-lactams, tetracy-
cline, kasugamycin, and trimethoprim resistance. A total of 113 Rank I
(Q1) genes were identied and Psychrobacter was the predominant host
for Q1 genes in the samples (Fig. 5 A). As shown in Fig. 5B, correlations
were identied between high-risk ARGs associated with aminoglycoside
(aph(3
′
)-I), tetracycline (tetH), beta_lactam (TEM-171), and disinfectant
resistance genes (mdtB/C/E/F, emrA, acrB, qacG). The use of biocidal
products for disinfection is the corner stone of infection prevention and
control in health care, the food industry and home hygiene settings. The
use of biocidal products to reduce infection risk is an integral element of
combatting the spread of antibiotic microbial resistance. This study
employed metagenomic sequencing analysis to elucidate the charac-
teristics of ARGs. The use of new research tools has allowed us to un-
derstand that biocide-led cross-resistance to different chemistries [8].
Besides, it was noteworthy that out of the detected high abundance of
ARGs related to aminoglycosides and tetracycline, only 9 were classied
as risk ARGs. This suggested that certain high abundance ARGs may
exhibit limited mobility and provide a lower risk.
A total of 647 potential risk ARGs, including 113 Rank I (Q1) risk
ARGs such as qacEΔ1, dfrA1, tetM, ermB, and sul1, were identied uti-
lizing an omics-based methodology for evaluating the risk associated
with ARGs. These risk ARGs (Q1) are linked to “current threats” that
have the highest potential to introduce multidrug resistant pathogens
and “future threats” that can transfer to pathogens as new resistance
forms. Human-associated-enrichment, gene mobility, and host patho-
genicity should be considered in ARGs risk assessment [16]. In our
analysis, we discovered the qacEΔ1(0.0038 copies per 16 S rRNA) gene
as the risk Rank I ARGs and disinfectant resistance gene. The qacEΔ1
gene have been reported in clinical isolates of Gramnegative bacilli, and
it has been suggested that the increase in both antibiotic and antiseptic
resistance in this organism is related to the presence of this gene [71].
QacEΔ1 gene encodes efux proteins that belong to a small multidrug
resistance (SMR) protein family containing distinct sub-sets of amino
acid residues involved in substrate recognition and binding. Proteins
which are then integrated in the cytoplasmic membrane via trans-
membrane segments, effectively pumping cytotoxic antimicrobial drug
elements out of the cells [72]. Obviously, the risk of ARGs in the envi-
ronmental resistomes often assessed by indicators such as abundance,
propensity for lateral transmission and ability of ARGs to be expressed in
Fig. 4. (A) Co-occurrence patterns of disinfectant resistance genes (green dots) and dominant genera (red dots). (B) Co-occurrence patterns of disinfectant resistance
genes (green dots) and pathogenic bacteria (red dots) at the species level. (C) Correlation heatmap of disinfectant resistance genes and ARGs. * p<0.05, * * p<0.01,
* ** p<0.001. (D) Distribution of disinfectant resistance genes at each sampling sites.
X. Xiao et al.

Journal of Hazardous Materials 468 (2024) 133811
10
(A)
(B)
**
*
**
*
*
**
**
**
**
*
*
**
*
* **
**
**
**
*
**
**
*
**
**
*
**
**
**
*
*
mdtP
emrB
TolC
EmrB
emrD
mdtA
mdtC
acrF
emrA
acrB
mdtB
mdtE
mdtF
qacG
lnuB
O
ermF
aadA
cat
arr
sul2
catB
tetH
fosA
mdt
qnrS
qnrB
tetV
Disinfectant related genes
Rank ARG
Multidrug
Quinolone
Rifamycin
Fosfomycin
Sulfonamide
Chloramphenicol
Aminoglycoside
Beta_lactam
Tetracycline
MLS
Trimethoprim
rank
Aminoglycoside
rank
rank
rank
Bacitracin
Beta_lactam
Bleomycin
Chloramphenicol
Fosfomycin
Kasugamycin
MLS
Multidrug
Polymyxin
Quinolone
Rifamycin
Sulfonamide
Tetracycline
Trimethoprim
Vancomycin
Psychrobacter
Moraxella
Pseudomonas
Corynebacterium
Macrococcus
Acinetobacter
Shewanella
Kocuria
Myroides
Micrococcus
Deinococcus
Paracoccus
Mycolicibacterium
Chryseobacterium
Bradyrhizobium
Rhodopseudomonas
Mesorhizobium
Rhodanobacter
WA
WB
WC
WD
WE
WF
WG
WH
BA
BB
BC
BD
BF
BH
Rank gene
ARG type
Genue
Sample site
Fig. 5. (A) Sankey diagram indicating the co-occurrence of risk rank genes with antibiotics, potential hosts, and sample sites (B) Correlation analysis of disinfectant
resistance genes and Rank I ARGs.
X. Xiao et al.

Journal of Hazardous Materials 468 (2024) 133811
11
pathogens. Only using abundance to directly characterize the risk of
ARGs would overestimate the situation [73]. Some high abundance
ARGs may have low mobility, which inhibits their spreads in the envi-
ronment and shows low risk [74].
3.5. Culture dependent results consistent with metagenome sequencing
In order to validate the ndings of the metagenomic analysis,
representative isolates were selected from the samples to assess the
effectiveness of NaClO in the PWM environment. In addition, the iso-
lates were utilized to further investigate the cross-resistance between
NaClO disinfection and antibiotics. According to the data presented in
Table 1, the prevalence of E. coli and Salmonella during market trans-
actions was determined to be 51.6% and 8.1% respectively. After the
completion of trading and disinfection processes, these levels decreased
to 23.8% and 7.1% respectively. The pathogenic bacteria, E. coli or
Salmonella, were isolated in all surfaces except pork hooks.The effec-
tiveness of disinfectant is dependent on factors such as materials, protein
load [75]. The wooden chopping board surface exhibits greater rough-
ness and can signicantly increase the thickness and compactness of the
biolm, thereby increasing the resistance to disinfectants [76].”
Achieving and maintaining good hygiene is also important for the
effectiveness of disinfectants. In unhygienic conditions, high protein
load has almost no effect on the mortality rate of Gram-negative bac-
teria, but it might reduce the efcacy against Gram-positive bacteria
[77].
The three sampling sites with the highest levels of contamination
were chopping boards (80%), meat (80%), and knives (70%) (Table 1).
These ndings suggest the possibility of bacterial cross-contamination
during meat portioning. Chopping boards represent complex microbial
ecosystems in which several factors that may favor the emergence and
spread of AMR converge (i.e., they are environments with high humidity
and contain dense microbial populations exposed to high contents of
organic matter). This micro-environment provides ideal conditions for
biolm formation and HGT [39]. Previous study has been reported that
the overuse of antibiotics as therapeutic, metaphylactic, or prophylactic
agents in intensive rearing of food production animals, linked to the
cross-contamination of meat with antibiotic-resistant bacteria during
evisceration and other dressing activities at slaughterhouses, may result
in the introduction of antibiotic-resistant microbes in meat processing
plants [78].
All Salmonella isolates displayed tolerance to NaClO, whereas E. coli
isolates exhibited a tolerance rate of 28.6%. The Salmonella and E. coli
isolates exhibited a 100% and 95.2% incidence of multidrug resistance,
respectively. as shown in Table 1. The majority of these isolates were
found to be resistant to tetracycline at a rate of 96% and to colistin at a
rate of 94%. Subsequently, we conducted non-parametric Spearman
correlation tests to assess the potential link between NaClO and anti-
biotic resistance by comparing their MIC values. The study revealed
noteworthy favorable associations between NaClO and aminoglyco-
sides, quinolone, polymyxin resistance, as indicated in Table S2. Similar
observations from metagenome sequencing were found between disin-
fectant resistance genes and ARGs related to aminoglycosides, quino-
lone, polymyxin. The correlation analysis is a statistical measure often
used in studies to show an association between variables. The limitations
of correlation analysis when were its assumption of a linear association
and its sensitivity to the range of observations. The value of the corre-
lation coefcient is inuenced by measurement error. Additional
culturing procedures needed to be conducted to validate the ndings of
correlation analysis [79].
4. Conclusions
Our study represents an initial attempt to assess the efcacy of NaClO
disinfection in mitigating microbial contamination in PWM environ-
ments. This evaluation included a comprehensive examination of both
the microbiome and resistomes, thereby enhancing our understanding of
the transmission of pathogenic bacteria, ARGs, and disinfectant resis-
tance genes. The abundance of pathogenic bacteria increased in the
PWM environment. The results suggest that the application of NaClO
disinfection in PWM environments is limited. Moreover, the abundance
of disinfectant resistance genes in external transport vehicles, carts, and
oors increased after disinfection, which raises the risk of further
adaptive resistance. A 100% occurrence of tolerance to NaClO at a MIC
greater than 128 mg/L was observed in Salmonella isolates, whereas
E. coli isolates exhibited a tolerance rate of 28.6%. The Salmonella and
E. coli isolates exhibited a 100% and 95.2% incidence of multidrug
resistance, respectively. Correlations were identied between risk Rank
I ARGs and aminoglycoside, tetracycline, beta-lactam, and disinfectant
resistance genes. These data highlight the potential for cross-resistance
between NaClO and antibiotics. Future research should include studies
on environmental reservoirs to identify the sources of ARGs and to
elucidate their routes to (or from) wildlife from the perspective of One
Health.
CRediT authorship contribution statement
Li Yanbin: Investigation, Conceptualization. Yang Hua: Investiga-
tion, Conceptualization. Xiao Yingping: Investigation. Ma Lingyan:
Methodology, Investigation. He Miao: Writing – review & editing,
Methodology, Investigation. Huang Kang: Methodology, Investigation.
Lv Wentao: Investigation. Zou Likou: Investigation, Conceptualization.
Xiao Xingning: Writing – original draft, Project administration, Meth-
odology, Investigation, Formal analysis, Conceptualization. Wang Wen:
Writing – review & editing, Supervision, Project administration,
Conceptualization.
Declaration of Competing Interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Data Availability
The metagenomic sequencing raw data has been deposited in the
National Center for Biotechnology Information (NCBI) database under
the project accession numbers PRJAN954132.
Acknowledgements
This research was supported by the Natural Science Founding of
China (32302233), Key Research and Development Program of Zhejiang
Province (2022C02049), Key Research and Development Program of
Ningbo (2022Z178), Ministry of Agriculture and Rural Affairs
(14234017), Walmart Foundation (UA2020-152, UA2021-247), and
State Key Laboratory for Managing Biotic and Chemical Threats to the
Quality and Safety of Agro-products (2021DG700024-KF202104).
Appendix A. Supporting information
Supplementary data associated with this article can be found in the
online version at doi:10.1016/j.jhazmat.2024.133811.
References
[1] Liu, Z., Klumper, U., Shi, L., Ye, L., Li, M., 2019. From pig breeding environment to
subsequently produced pork: comparative analysis of antibiotic resistance genes
and bacterial community composition. Front Microbiol 10, 43.
[2] Liu, C., Chen, J., Shan, X., Yang, Y., Song, L., Teng, Y., et al., 2023. Meta-analysis
addressing the characterization and risk identication of antibiotics and antibiotic
resistance genes in global groundwater. Sci Total Environ 860, 160513.
X. Xiao et al.

Journal of Hazardous Materials 468 (2024) 133811
12
[3] Wang, M., Ren, P., Liu, H., Dai, X., 2021. Investigating antibiotics, antibiotic
resistance genes in soil, groundwater and vegetables in relation to agricultural eld
- applicated with lincomycin mycelial residues compost. Sci Total Environ 777,
146066.
[4] Gao, F.Z., Zou, H.Y., Wu, D.L., Chen, S., He, L.Y., Zhang, M., et al., 2020. Swine
farming elevated the proliferation of Acinetobacter with the prevalence of
antibiotic resistance genes in the groundwater. Environ Int 136, 105484.
[5] Exner, M., Bhattacharya, S., Gebel, J., Goroncy-Bermes, P., Hartemann, P.,
Heeg, P., et al., 2020. Chemical disinfection in healthcare settings: critical aspects
for the development of global strategies. GMS Hyg Infect Control 15.
[6] Lewis, D., 2021. COVID-19 rarely spreads through surfaces. So why are we still
deep cleaning? Nature 590, 26–28.
[7] Yan, M., Xu, C., Li, C., Feng, Y., Duan, J., Zhao, K., et al., 2023. Effects of
environmental disinfection on microbial population and resistance genes: a case
study of the microecology within a panda enclosure. Environ Res 235, 116662.
[8] Maillard, J.-Y., Pascoe, M., 2024. Disinfectants and antiseptics: mechanisms of
action and resistance. Nat Rev Microbiol 22, 4–17.
[9] Sato, Y., Ishihara, M., Nakamura, S., Fukuda, K., Kuwabara, M., Takayama, T.,
et al., 2019. Comparison of various disinfectants on bactericidal activity under
organic matter contaminated environments. Biocontrol Sci 24, 103–108.
[10] Li, Y., Wang, H., Xu, C., Sun, S.H., Xiao, K., Huang, X., 2022. Two strategies of
stubborn biofouling strains surviving from NaClO membrane cleaning: EPS
shielding and/or quorum sensing. Sci Total Environ 838, 156421.
[11] Parry, A.H., Wani, H.A., Yaseen, M., 2021. Management of a diagnostic radiology
department amid Coronavirus disease-19 (COVID-19) pandemic. Indian J Radiol
Imaging 31, S192–S195.
[12] Xiao, X., Bai, L., Wang, S., Liu, L., Qu, X., Zhang, J., et al., 2022. Chlorine tolerance
and cross-resistance to antibiotics in poultry-associated salmonella isolates in
China. Front Microbiol 12.
[13] Xiao, X., Bai, L., Wang, S., Liu, L., Qu, X., Zhang, J., et al., 2021. Chlorine tolerance
and cross-resistance to antibiotics in poultry-associated salmonella isolates in
China. Front Microbiol 12, 833743.
[14] Sato, Y., Ishihara, M., Nakamura, S., Fukuda, K., Kuwabara, M., Takayama, T.,
et al., 2019. Comparison of various disinfectants on bactericidal activity under
organic matter contaminated environments. Biocontrol Sci 24.
[15] Wu, G., Yang, Q., Long, M., Guo, L., Li, B., Meng, Y., et al., 2015. Evaluation of agar
dilution and broth microdilution methods to determine the disinfectant
susceptibility. J Antibiot 68, 661–665.
[16] Zhang, T., Ding, H., Chen, L., Zhang, S., Wu, P., Xie, K., et al., 2021.
Characterization of chilled chicken spoilage using an integrated microbiome and
metabolomics analysis. Food Res Int 144, 110328.
[17] Rolbiecki, D., Paukszto, Ł., Krawczyk, K., Korzeniewska, E., Sawicki, J.,
Harnisz, M., 2023. Chlorine disinfection modies the microbiome, resistome and
mobilome of hospital wastewater – A nanopore long-read metagenomic approach.
J Hazard Mater 459.
[18] Tong, C., Hu, H., Chen, G., Li, Z., Li, A., Zhang, J., 2021. Chlorine disinfectants
promote microbial resistance in Pseudomonas sp. Environ Res 199, 111296.
[19] Yan, M., Xu, C., Li, C., Feng, Y., Duan, J., Zhao, K., et al., 2023. Effects of
environmental disinfection on microbial population and resistance genes: a case
study of the microecology within a panda enclosure. Environ Res 235.
[20] Xiao, X., Wang, W., Zhang, J., Liao, M., Rainwater, C., Yang, H., et al., 2021.
A quantitative risk assessment model of Salmonella contamination for the yellow-
feathered broiler chicken supply chain in China. Food Control 121.
[21] Zhou, J., Jin, Y., Liang, Q., 2022. Effects of regulatory policy mixes on traceability
adoption in wholesale markets: food safety inspection and information disclosure.
Food Policy 107.
[22] Lo, M.Y., Ngan, W.Y., Tsun, S.M., Hsing, H.L., Lau, K.T., Hung, H.P., et al., 2019.
A eld study into Hong Kong’s wet markets: raised questions into the hygienic
maintenance of meat contact surfaces and the dissemination of microorganisms
associated with nosocomial infections. FRONT MICROBIOL 10, 2618.
[23] Yi, Z., Qiu, M., Xiao, X., Ma, J., Yang, H., Wang, W., 2024. Quantitative
characterization and dynamics of bacterial communities in ready-to-eat chicken
using high-throughput sequencing combined with internal standard-based absolute
quantication. Food Microbiol 118, 104419.
[24] Wu, J., Xiang, J., Zeng, Q., Liu, Q., Li, Y., Sun, Y., et al., 2023. Mitigation of the
negative impacts of 84 disinfectant exposure for phenol treatment by an electro-
enhanced SBR: Performance, oxidative stress, iron homeostasis and antibiotic
resistance genes. J Water Process Eng 55.
[25] Zhu, M., Li, Y., Wang, L., Zhang, W., Niu, L., Hu, T., 2023. Unraveling antibiotic
resistomes associated with bacterial and viral communities in intertidal mudat
aquaculture area. J Hazard Mater 459.
[26] Bolger, A.M., Lohse, M., Usadel, B., 2014. Trimmomatic: a exible trimmer for
Illumina sequence data. Bioinformatics 30, 2114–2120.
[27] Valenzuela-Gonz´
alez, F., Martinez-Porchas, M., Villalpando, E., Vargas-Albores, F.,
2016. Studying long 16S rDNA sequences with ultrafast-metagenomic sequence
classication using exact alignments (Kraken). J Microbiol Methods 122.
[28] Mcardle, A., Kaforou, M., 2020. Sensitivity of shotgun metagenomics to host DNA:
abundance estimates depend on bioinformatic tools and contamination is the main
issue. Access Microbiol 2.
[29] Yang, Y., Jiang, X.-T., Chai, B., Ma, L., Li, B., Zhang, A.N., et al., 2016. ARGs-OAP:
Online analysis pipeline for antibiotic resistance genes detection from
metagenomic data using an integrated structured ARG-database. Bioinforma (Oxf,
Engl) 32.
[30] Yin, X., Jiang, X.-T., Chai, B., Li, L., Yang, Y., Cole’s, J.R., et al., 2018. ARGs-OAP
v2.0 with an expanded SARG database and hidden markov models for
enhancement characterization and quantication of antibiotic resistance genes in
environmental metagenomes. Bioinforma (Oxf, Engl) 34.
[31] Parnanen, K., Karkman, A., Hultman, J., Lyra, C., Bengtsson-Palme, J., Larsson, D.
G.J., et al., 2018. Maternal gut and breast milk microbiota affect infant gut
antibiotic resistome and mobile genetic elements. Nat Commun 9, 3891.
[32] M. Mauri, T. Elli, G. Caviglia, G. Uboldi, M. Azzi. RAWGraphs: A Visualisation
Platform to Create Open Outputs[M]. 2017: 1–5.
[33] Zhang, A.N., Gaston, J.M., Dai, C.L., Zhao, S., Poyet, M., Groussin, M., et al., 2021.
An omics-based framework for assessing the health risk of antimicrobial resistance
genes. Nat Commun 12, 4765.
[34] Yuan, Y., Xu, W., Zhai, Z., Shi, H., Luo, Y., Chen, Z., et al., 2009. Universal primer-
multiplex PCR approach for simultaneous detection of Escherichia coli, Listeria
monocytogenes, and Salmonella spp. in food samples. J Food Sci 74, M446–M452.
[35] Zhang, L., Fu, Y., Xiong, Z., Ma, Y., Wei, Y., Qu, X., et al., 2018. Highly prevalent
multidrug-resistant salmonella from chicken and pork meat at retail markets in
Guangdong, China. Front. Microbiol. 9, 2104.
[36] Luo, L.W., Wu, Y.H., Yu, T., Wang, Y.H., Chen, G.Q., Tong, X., et al., 2021.
Evaluating method and potential risks of chlorine-resistant bacteria (CRB): a
review. Water Res 188, 116474.
[37] Mahros, M., Abd-Elghany, S., Sallam, K., 2021. Multidrug-, methicillin-, and
vancomycin-resistant Staphylococcus aureus isolated from ready-to-eat meat
sandwiches: an ongoing food and public health concern. INT J FOOD MICROBIOL
346, 109165.
[38] Ma, L., Tao, S., Song, T., Lyu, W., Li, Y., Wang, W., et al., 2024. Clostridium
butyricum and carbohydrate active enzymes contribute to the reduced fat
deposition in pigs. iMeta.
[39] Cobo-Díaz, J.F., Alvarez-Molina, A., Alexa, E.A., Walsh, C.J., Mencía-Ares, O.,
Puente-G´
omez, P., et al., 2021. Microbial colonization and resistome dynamics in
food processing environments of a newly opened pork cutting industry during 1.5
years of activity. Microbiome 9.
[40] Xiao, X., Pang, H., Wang, W., Fang, W., Fu, Y., Li, Y., 2018. Modeling transfer of
vibrio parahaemolyticus during peeling of raw shrimp: bacterial transfer in shrimp
peeling…. J Food Sci 83.
[41] Wang, D., Wang, Y., Huang, H., Lin, J., Xiao, D., Kan, B., 2013. Identication of
tetrodotoxin-producing Shewanella spp. from feces of food poisoning patients and
food samples. Gut Pathog 5, 15.
[42] Qiu, M., Xiao, X., Xiao, Y., Ma, J., Yang, H., Jiang, H., et al., 2022. Dynamic
changes of bacterial communities and microbial association networks in ready-to-
eat chicken meat during storage. Foods 11, 3733.
[43] Nyenje, M.E., Odjadjare, C.E., Tanih, N.F., Green, E., Ndip, R.N., 2012. Foodborne
pathogens recovered from ready-to-eat foods from roadside cafeterias and retail
outlets in alice, Eastern Cape Province, South Africa: public health implications. Int
J Environ Res Public Health 9, 2608–2619.
[44] Emamjomeh, M., Mohd Hashim, A., Abdul-Mutalib, N., Khairil Mokhtar, N.,
Mustapha, N.A., Maeda, T., et al., 2023. Proling bacterial communities and
foodborne pathogens on food-associated surface following contact with raw beef,
chicken and pork using 16S amplicon metagenomics. Food Control 149, 109698.
[45] Streeter, K., Katouli, M., 2016. Pseudomonas aeruginosa: a review of their
Pathogenesis and Prevalence in Clinical Settings and the Environment, Infection.
Epidemiol Med 2, 25–32.
[46] Karumathil, D., Yin, H., Kollanoor Johny, A., Venkitanarayanan, K., 2014. Effect of
chlorine exposure on the survival and antibiotic gene expression of multidrug
resistant acinetobacter baumannii in water. Int J Environ Res Public Health 11,
1844–1854.
[47] Hou, A.-M., Yang, D., Miao, J., Shi, D.-Y., Yin, J., Yang, Z.-W., et al., 2019. Chlorine
injury enhances antibiotic resistance in Pseudomonas aeruginosa through over
expression of drug efux pumps. Water Res 156, 366–371.
[48] Zhang, S.-P., Feng, H.-Z., Wang, Q., Quan, S.-W., Yu, X.-Q., Tao, X., et al., 2021.
Proteomic analysis reveals the mechanism of different environmental stress-
induced tolerance of Pseudomonas aeruginosa to monochloramine disinfection.
J Hazard Mater 417.
[49] Lung, C.W., Zheng, Z., Lo, I.M.C., 2023. Solar-driven photocatalytic chlorine
activation for the simultaneous degradation of pharmaceuticals and personal care
products and the inactivation of Escherichia coli in drinking water. Chemosphere
311, 137019.
[50] Zhang, M., Chen, S., Yu, X., Vikesland, P., Pruden, A., 2019. Degradation of
extracellular genomic, plasmid DNA and specic antibiotic resistance genes by
chlorination. Front Environ Sci Eng 13, 38.
[51] Stange, C., Sidhu, J.P.S., Toze, S., Tiehm, A., 2019. Comparative removal of
antibiotic resistance genes during chlorination, ozonation, and UV treatment. Int J
Hyg Environ Health 222, 541–548.
[52] Lin, W., Zhang, M., Zhang, S., Yu, X., 2016. Can chlorination co-select antibiotic-
resistance genes? Chemosphere 156, 412–419.
[53] Liu, S.-S., Qu, H.-M., Yang, D., Hu, H., Liu, W.-L., Qiu, Z.-G., et al., 2018. Chlorine
disinfection increases both intracellular and extracellular antibiotic resistance
genes in a full-scale wastewater treatment plant. Water Res 136, 131–136.
[54] Baquero, F., Martinez, J.L., Canton, R., 2008. Antibiotics and antibiotic resistance
in water environments. Curr Opin Biotechnol 19, 260–265.
[55] Zhang, G., Guan, Y., Zhao, R., Feng, J., Huang, J., Ma, L., et al., 2020. Metagenomic
and network analyses decipher proles and co-occurrence patterns of antibiotic
resistome and bacterial taxa in the reclaimed wastewater distribution system.
J Hazard Mater 400.
[56] Oliveira, D.M.P.D., Forde, B.M., Kidd, T.J., Harris, P.N.A., Schembri, M.A.,
Beatson, S.A., et al., 2020. Antimicrobial resistance in ESKAPE pathogens. Clin
Microbiol Rev 33 e00181-00119.
X. Xiao et al.

Journal of Hazardous Materials 468 (2024) 133811
13
[57] Lupo, A., Vogt, D., Seiffert, S.N., Endimiani, A., Perreten, V., 2014. Antibiotic
resistance and phylogenetic characterization of acinetobacter baumannii strains
isolated from commercial raw meat in Switzerland. J Food Prot 77, 1976–1981.
[58] Carvalheira, A., Casquete, R., Silva, J., Teixeira, P., 2017. Prevalence and
antimicrobial susceptibility of Acinetobacter spp. isolated from meat. INT J FOOD
MICROBIOL 243, 58–63.
[59] Ibrahim, R.A., Cryer, T.L., La, S.Q., Basha, E.A., Good, L., Tarazi, Y.H., 2019.
Identication of Escherichia coli from broiler chickens in Jordan, their
antimicrobial resistance, gene characterization and the associated risk factors. BMC
Vet Res 15, 159.
[60] Xu, Z., Wang, M., Zhou, C., Gu, G., Liang, J., Hou, X., et al., 2020. Prevalence and
antimicrobial resistance of retail-meat-borne Salmonella in southern China during
the years 2009-2016: the diversity of contamination and the resistance evolution of
multidrug-resistant isolates. INT J FOOD MICROBIOL 333, 108790.
[61] Yang, Y., Li, L., Huang, F., Hu, X., Cao, X., Mi, J., et al., 2022. The fate of antibiotic
resistance genes and their association with bacterial and archaeal communities
during advanced treatment of pig farm wastewater. Sci Total Environ 851, 158364.
[62] Luo, T., Dai, X., Chen, Z., Wu, L., Wei, W., Xu, Q., et al., 2023. Different
microplastics distinctively enriched the antibiotic resistance genes in anaerobic
sludge digestion through shifting specic hosts and promoting horizontal gene
ow. Water Res 228, 119356.
[63] Jia, S., Wu, J., Ye, L., Zhao, F., Li, T., Zhang, X.-X., 2019. Metagenomic assembly
provides a deep insight into the antibiotic resistome alteration induced by drinking
water chlorination and its correlations with bacterial host changes. J Hazard Mater
379.
[64] Yang, Y., Li, T., Liu, P., Li, H., Hu, F., 2022. The formation of specic bacterial
communities contributes to the enrichment of antibiotic resistance genes in the soil
plastisphere. J Hazard Mater 436.
[65] Tong, C., Hu, H., Chen, G., Li, Z., Aifeng, L., Zhang, J., 2021. Chlorine disinfectants
promote microbial resistance in Pseudomonas sp. Environ Res 199, 111296.
[66] Guo, N., Wang, Y., Tong, T., Wang, S., 2018. The fate of antibiotic resistance genes
and their potential hosts during bio-electrochemical treatment of high-salinity
pharmaceutical wastewater. Water Res 133, 79–86.
[67] Xiong, L., Sun, Y., Shi, L., Yan, H., 2019. Characterization of antimicrobial
resistance genes and class 1 integrase gene in raw meat and aquatic product, fresh
vegetable and fruit, and swine manure in southern China. Food Control 104,
240–246.
[68] Beattie, R.E., Skwor, T., Hristova, K.R., 2020. Survivor microbial populations in
post-chlorinated wastewater are strongly associated with untreated hospital
sewage and include ceftazidime and meropenem resistant populations. Sci Total
Environ 740, 140186.
[69] Adesoye, M., Olaniran, A., 2022. Does chlorination promote antimicrobial
resistance in waterborne pathogens? mechanistic insight into co-resistance and its
implication for public health. Antibiotics 11, 564.
[70] Karumathil, D.P., Yin, H.B., Kollanoor-Johny, A., Venkitanarayanan, K., 2014.
Effect of chlorine exposure on the survival and antibiotic gene expression of
multidrug resistant Acinetobacter baumannii in water. Int J Environ Res Public
Health 11, 1844–1854.
[71] Gomaa, F., Helal, Z., Khan, M., 2017. High Prevalence of blaNDM-1, blaVIM, qacE,
and qacEΔ1 Genes and Their Association with Decreased Susceptibility to
Antibiotics and Common Hospital Biocides in Clinical Isolates of Acinetobacter
baumannii. Microorganisms 5.
[72] Chen, S., Fu, J., Zhao, K., Yang, S., Li, C., Penttinen, P., et al., 2023. Class 1
integron carrying qacEΔ1 gene confers resistance to disinfectant and antibiotics in
Salmonella. Int J Food Microbiol 404.
[73] Manaia, C.M., 2017. Assessing the risk of antibiotic resistance transmission from
the environment to humans: non-direct proportionality between abundance and
risk. Trends Microbiol 25, 173–181.
[74] Martinez, J.L., Coque, T.M., Baquero, F., 2015. What is a resistance gene? Ranking
risk in resistomes. Nat Rev Microbiol 13, 116–123.
[75] Wang, Y., Li, S., Gong, X., Chen, Q., Ji, G., Liu, Y., et al., 2020. Characterization of
RaeE-RaeF-RopN, a putative RND efux pump system in Riemerella anatipestifer.
Vet Microbiol 251, 108852.
[76] Zhu, Z., Shan, L., Zhang, X., Hu, F., Zhong, D., Yuan, Y., et al., 2021. Effects of
bacterial community composition and structure in drinking water distribution
systems on biolm formation and chlorine resistance. Chemosphere 264, 128410.
[77] Tong, C., Hu, H., Chen, G., Li, Z., Li, A., Zhang, J., 2021. Disinfectant resistance in
bacteria: mechanisms, spread, and resolution strategies. Environ Res 195, 110897.
[78] Capita, R., Alonso-Calleja, C., 2013. Antibiotic-resistant bacteria: a challenge for
the food industry. Crit Rev Food Sci Nutr 53, 11–48.
[79] Janse, R., Hoekstra, T., Jager, K., Zoccali, C., Tripepi, G., Dekker, F., et al., 2021.
Conducting correlation analysis: important limitations and pitfalls. Clin Kidney J
14.
X. Xiao et al.