Effects of antibiotics and metals on lung and intestinal microbiome dysbiosis after sub-chronic lower-level exposure of air pollution in ageing rats

Abstract

We investigated the effects of antibiotics, drugs, and metals on lung and intestinal microbiomes after sub-chronic exposure of low-level air pollution in ageing rats. Male 1.5-year-old Fischer 344 ageing rats were exposed to low-level traffic-related air pollution via whole-body exposure system for 3 months with/without high-efficiency particulate air (HEPA) filtration (gaseous vs. particulate matter with aerodynamic diameter of ≤2.5 µm Abbreviations: AQG, Air Quality Guidelines; BALF, Bronchoalveolar lavage; CD, Cluster of differentiation; CO, Carbon monoxide; COPD, Chronic obstructive pulmonary disease; FEF 25-75 , forced expiratory flow at 25~75% of the FVC; FEV 20 , forced expiratory volume in 20 ms; FVC, forced vital 2 Lung function Metals Microbiome PM 2.5 (PM 2.5) pollution). Lung functions, antibiotics, drugs, and metals in lungs were examined and linked to lung and fecal microbiome analyses by high-throughput sequencing analysis of 16 s ribosomal (r)DNA. Rats were exposed to 8.7 μg/m 3 PM 2.5 , 10.1 ppb NO 2 , 1.6 ppb SO 2 , and 23.9 ppb O 3 in average during the study period. Air pollution exposure decreased forced vital capacity (FVC), peak expiratory flow (PEF), forced expiratory volume in 20 ms (FEV 20), and FEF at 25~75% of FVC (FEF 25-75). Air pollution exposure increased antibiotics and drugs (benzotriazole, methamphetamine, methyl-1 H-benzotriazole, ketamine, ampicillin, ciprofloxacin, pentoxifyl-line, erythromycin, clarithromycin, ceftriaxone, penicillin G, and penicillin V) and altered metals (V, Cr, Cu, Zn, and Ba) levels in lungs. Fusobacteria and Verrucomicrobia at phylum level were increased in lung microbiome by air pollution, whereas increased alpha diversity, Bacteroidetes and Proteobacteria and decreased Firmicutes at phylum level were occurred in intestinal microbiome. Lung function decline was correlated with increasing antibiotics, drugs, and metals in lungs as well as lung and intestinal microbiome dysbiosis. The antibiotics, drugs, and Cr, Co, Ca, and Cu levels in lung were correlated with lung and intestinal microbiome dysbiosis. The lung microbiome was correlated with intestinal microbiome at several phylum and family levels after air pollution exposure. Our results revealed that antibiotics, drugs, and metals in the lung caused lung and intestinal microbiome dysbiosis in ageing rats exposed to air pollution, which may lead to lung function decline.

Full text

Ecotoxicology and Environmental Safety 246 (2022) 114164
Available online 13 October 2022
0147-6513/© 2022 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Effects of antibiotics and metals on lung and intestinal microbiome
dysbiosis after sub-chronic lower-level exposure of air pollution in
ageing rats
Vincent Laiman
a
,
b
,
1
, Yu-Chun Lo
c
,
1
, Hsin-Chang Chen
d
, Tzu-Hsuen Yuan
e
, Ta-Chih Hsiao
f
,
Jen-Kun Chen
g
, Ching-Wen Chang
h
, Ting-Chun Lin
i
, Ssu-Ju Li
i
, You-Yin Chen
c
,
h
,
i
,
Didik Setyo Heriyanto
b
, Kian Fan Chung
j
, Kai-Jen Chuang
k
,
l
, Kin-Fai Ho
m
, Jer-Hwa Chang
n
,
o
,
*
,
Hsiao-Chi Chuang
n
,
p
,
q
,
r
,
**
a
International Ph.D. Program in Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
b
Department of Anatomical Pathology, Faculty of Medicine, Public Health, and Nursing, Universitas Gadjah Mada – Dr. Sardjito Hospital, Yogyakarta, Indonesia
c
Ph.D. Program in Medical Neuroscience, College of Medical Science and Technology, Taipei Medical University, Taipei, Taiwan
d
Department of Chemistry, College of Science, Tunghai University, Taichung, Taiwan
e
Department of Health and Welfare, College of City Management, University of Taipei, Taipei, Taiwan
f
Graduate Institute of Environmental Engineering, National Taiwan University, Taipei, Taiwan
g
Institute of Biomedical Engineering & Nanomedicine, National Health Research Institutes, Miaoli, Taiwan
h
Industrial Ph.D. Program of Biomedical Science and Engineering, National Yang Ming Chiao Tung University, Taipei, Taiwan
i
Department of Biomedical Engineering, National Yang Ming Chiao Tung University, Taipei, Taiwan
j
National Heart and Lung Institute, Imperial College London, London, UK
k
School of Public Health, College of Public Health, Taipei Medical University, Taipei, Taiwan
l
Department of Public Health, School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan
m
School of Public Health and Primary Care, the Chinese University of Hong Kong, Hong Kong
n
School of Respiratory Therapy, College of Medicine, Taipei Medical University, Taipei, Taiwan
o
Division of Pulmonary Medicine, Departments of Internal Medicine, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan
p
Division of Pulmonary Medicine, Department of Internal Medicine, Shuang Ho Hospital, Taipei Medical University, New Taipei City, Taiwan
q
Cell Physiology and Molecular Image Research Center, Wan Fang Hospital, Taipei Medical University, Taipei, Taiwan
r
Graduate Institute of Medical Sciences, College of Medicine, Taipei Medical University, Taipei, Taiwan
ARTICLE INFO
Edited by Dr Yong Liang
Keywords:
Air pollution
Antibiotics
ABSTRACT
We investigated the effects of antibiotics, drugs, and metals on lung and intestinal microbiomes after sub-chronic
exposure of low-level air pollution in ageing rats. Male 1.5-year-old Fischer 344 ageing rats were exposed to low-
level trafc-related air pollution via whole-body exposure system for 3 months with/without high-efciency
particulate air (HEPA) ltration (gaseous vs. particulate matter with aerodynamic diameter of ≤2.5 µm
Abbreviations: AQG, Air Quality Guidelines; BALF, Bronchoalveolar lavage; CD, Cluster of differentiation; CO, Carbon monoxide; COPD, Chronic obstructive
pulmonary disease; FEF
25–75
, forced expiratory ow at 25~75% of the FVC; FEV
20
, forced expiratory volume in 20 ms; FVC, forced vital capacity; GMD, Geometric
mean diameter; HEPA, High-efciency particulate air; ICP-MS, Inductively coupled plasma-mass spectrometry; LDA, Linear discriminant analysis; LEfSe, Linear
discriminant analysis effect size; NO
2
, Nitrogen dioxide; O
3
, Ozone; PCA, principal component analysis; PCOA, Principal coordinate analysis; PCR, Polymerase chain
reaction; PEF, peak expiratory ow; PM
2.5
, Particulate matter with an aerodynamic diameter of ≤2.5 µm; PNC, Particle number concentration; RH, Relative hu-
midity; rDNA, ribosomal DNA; SO
2
, Sulfur dioxide; UniFrac, Unique fraction; UPLC-MS/MS, Ultra-performance liquid chromatography - tandem mass spectrometer;
WHO, World Health Organization.
* Correspondence to: School of Respiratory Therapy, College of Medicine, Taipei Medical University, 250 Wuxing Street, Taipei 11031, Taiwan.
** Correspondence to: Inhalation Toxicology Research Lab (ITRL), School of Respiratory Therapy, College of Medicine, Taipei Medical University, 250 Wuxing
Street, Taipei 11031, Taiwan.
E-mail addresses: vincentharun29@gmail.com (V. Laiman), aricalo@tmu.edu.tw (Y.-C. Lo), hsinchang@thu.edu.tw (H.-C. Chen), thyuan@go.utaipei.edu.tw
(T.-H. Yuan), tchsiao@gmail.com (T.-C. Hsiao), jkchen@nhri.org.tw (J.-K. Chen), love6xup@gmail.com (C.-W. Chang), jimmy601521@gmail.com (T.-C. Lin),
louislee0722@gmail.com (S.-J. Li), irradiance@so-net.net.tw (Y.-Y. Chen), didik_setyoheriyanto@mail.ugm.ac.id (D.S. Heriyanto), f.chung@imperial.ac.uk
(K.F. Chung), kjc@tmu.edu.tw (K.-J. Chuang), kfho@cuhk.edu.hk (K.-F. Ho), m102094030@tmu.edu.tw (J.-H. Chang), chuanghc@tmu.edu.tw (H.-C. Chuang).
1
These authors contributed equally to this work.
Contents lists available at ScienceDirect
Ecotoxicology and Environmental Safety
journal homepage: www.elsevier.com/locate/ecoenv
https://doi.org/10.1016/j.ecoenv.2022.114164
Received 5 July 2022; Received in revised form 5 September 2022; Accepted 5 October 2022

Ecotoxicology and Environmental Safety 246 (2022) 114164
2
Lung function
Metals
Microbiome
PM
2.5
(PM
2.5
) pollution). Lung functions, antibiotics, drugs, and metals in lungs were examined and linked to lung and
fecal microbiome analyses by high-throughput sequencing analysis of 16 s ribosomal (r)DNA. Rats were exposed
to 8.7
μ
g/m
3
PM
2.5
, 10.1 ppb NO
2
, 1.6 ppb SO
2
, and 23.9 ppb O
3
in average during the study period. Air
pollution exposure decreased forced vital capacity (FVC), peak expiratory ow (PEF), forced expiratory volume
in 20 ms (FEV
20
), and FEF at 25~75% of FVC (FEF
25–75
). Air pollution exposure increased antibiotics and drugs
(benzotriazole, methamphetamine, methyl-1 H-benzotriazole, ketamine, ampicillin, ciprooxacin, pentoxifyl-
line, erythromycin, clarithromycin, ceftriaxone, penicillin G, and penicillin V) and altered metals (V, Cr, Cu, Zn,
and Ba) levels in lungs. Fusobacteria and Verrucomicrobia at phylum level were increased in lung microbiome by
air pollution, whereas increased alpha diversity, Bacteroidetes and Proteobacteria and decreased Firmicutes at
phylum level were occurred in intestinal microbiome. Lung function decline was correlated with increasing
antibiotics, drugs, and metals in lungs as well as lung and intestinal microbiome dysbiosis. The antibiotics, drugs,
and Cr, Co, Ca, and Cu levels in lung were correlated with lung and intestinal microbiome dysbiosis. The lung
microbiome was correlated with intestinal microbiome at several phylum and family levels after air pollution
exposure. Our results revealed that antibiotics, drugs, and metals in the lung caused lung and intestinal
microbiome dysbiosis in ageing rats exposed to air pollution, which may lead to lung function decline.
1. Introduction
Air pollution is linked to a variety of respiratory diseases, including
chronic obstructive pulmonary disease (COPD), asthma, and pneumonia
(Li and Liu, 2021). In the case of COPD, for example, while tobacco
smoking is the leading cause, many other environmental factors,
including air pollution, play a role in its pathogenesis (He et al., 2017).
Air pollution in urban areas can be attributed to a variety of emission
sources, including heavy transport and mechanical workshops, vehic-
ular emissions, and worn vehicle tires (Zeb et al., 2022). The major
components of air pollution invading human lungs during respiration
are particles and gases (Jheng et al., 2021; Li and Liu, 2021). Particulate
matter with an aerodynamic diameter of ≤2.5 µm (PM
2.5
) can deposit
deep into the respiratory tract through inhalation and were suggested as
causative agents associated with adverse respiratory health (Carvalho
et al., 2011). Therefore, investigating health effects of air pollution,
particularly PM
2.5
is an important issue for human health.
Notably, existence for antibiotics in various environmental phase
including the air and soil has been identied (Dai et al., 2022). Antibi-
otics are used to prevent and treat bacterial infections, and they enter
the environment through a variety of routes, including direct entry from
the animal to the agriculture eld, pharmaceutical factories, and animal
grazing (Muhammad et al., 2019). Antibiotics can rmly adhere to clay
soil and sediments in a variety of environmental conditions, entering
and reaching the groundwater (Loghin et al., 2020; Muhammad et al.,
2019). These antibiotics were then redistributed to other environmental
phases such as soil and air via complex mechanisms of adsorption,
diffusion, transformation, and dispersion (Dai et al., 2022; Li et al.,
2021b). Hamscher and colleagues demonstrated that airborne antibiotic
residues were dispersed into the environment and persisted for an
extended period of time (Hamscher et al., 2003). The main concern for
the antibiotics released into the environment is the development of
antibiotic resistance bacteria and that its continuous exposure may
impose hazardous effects to the human body (Hu et al., 2018; Sessink
et al., 2019). Additionally, previous studies have also shown possible
contamination of drugs in the environment (Kuhn et al., 2019; Wright
et al., 2020). Methamphetamine, for example, is commonly manufac-
tured in clandestine drug laboratories and aerosols are released into the
surrounding atmosphere during the manufacturing processes (Kuhn
et al., 2019). Benzotriazole, a widely used corrosion inhibitor, was also
reported to be found in public places at an average concentration of
4.97 ng/m
3
(Xue et al., 2016). Airborne residues can settle on surfaces
and be transported via air transfer (Wright et al., 2020). However,
adverse health effects of exposure to environmental antibiotics and
drugs in the lung are rarely reported.
Our previous study characterized the components of PM
2.5
and
discovered the presence of trace metals in PM
2.5
(Laiman et al., 2022).
Further analysis revealed that PM
2.5
containing metals originated from
various sources, including secondary aerosol, biomass burning,
industry, airborne soil, and trafc. The trace metals in PM
2.5
was re-
ported to cause lung injury by catalyzing the formation of oxidants in the
lung (Saldarriaga-Nore˜
na et al., 2009). Previous studies of rats exposed
to ambient PM
2.5
concentration of 19.7
μ
g/m
3
in northern Taiwan
showed declining lung function (Chuang et al., 2020; Jheng et al., 2021).
Air pollution, including PM
2.5
, therefore, have crucial roles in lung
function deterioration. Studies aimed at better elucidation of the path-
ogenic processes due to air pollution exposure, particularly PM
2.5
, are
still required.
In recent years, it has been demonstrated that lung contains diverse
microbial community (microbiome), which maintains normal homeo-
stasis and was signicantly altered in variety of pulmonary disorders
(Ramsheh et al., 2021; Tian et al., 2022). For example, COPD patients
showed different lung microbiome than healthy individuals, and the
microbiome further shifted during COPD acute exacerbation episodes
(Healy et al., 2021; Mammen and Sethi, 2016). Previous study
comparing COPD and healthy individuals found that air pollution,
specically PM
2.5
, signicantly altered the lower airway microbiome
composition, which was correlated with decreased lung function (Wang
et al., 2019). Interestingly, the intestinal microbiome was also reported
to be a signicant modulator of pulmonary inammatory diseases
(Bowerman et al., 2020). The study reported that fecal microbiome of
COPD patients was distinct from that of healthy individuals, and the
changes also correlate with reduced lung function. Concomitantly,
particulate matter present in ambient air pollution also altered the
murine intestinal microbiome, which was linked to important health
problems, including asthma and COPD (Healy et al., 2021; Mutlu et al.,
2018; Zheng et al., 2020). Dysbiosis of the lung and intestinal micro-
biomes, therefore, has signicant adverse impacts on human health.
Currently, the relationships of antibiotics, drugs, and metals with the
lung and intestinal microbiomes, as well as the lung functions by air
pollution are not well understood. Although associations of host
microbiomes with lung disease have been investigated, the effects of
antibiotics, drugs, and metals on lung and intestinal microbiome dys-
biosis after air pollution exposure remain unclear. This study aimed to
investigate the effects of antibiotics, drugs, and metals on the lung
function and lung and intestinal microbiomes of ageing rats by sub-
chronic exposure to low-level ambient air pollution.
2. Materials and methods
2.1. Ageing animal model
Male 1.5-year-old Fischer 344 ageing rats (equivalent to 45-year-old
humans) (National Laboratory Animal Center, Taipei, Taiwan) were
housed at constant temperature of 22 ±2 ◦C and relative humidity (RH)
of 55% ±10% with 12:12-h light: dark cycle (Chang et al., 2022; Sen-
gupta, 2013). This study was conducted in compliance with the Animal
and Ethics Review Committee of the Laboratory Animal Center at Taipei
V. Laiman et al.

Ecotoxicology and Environmental Safety 246 (2022) 114164
3
Medical University (Taipei, Taiwan; IACUC: LAC-2019–0424). Ageing
rats were used in this study to increase the lung susceptibility to develop
lung disease which is comparable to that in humans (Fukuchi, 2009).
2.2. Exposure to ambient air pollution
Ageing rats were continually exposed to ambient unconcentrated
trafc-related air pollution for 3 months (24 h/day) using whole-body
exposure system equipped with and without high-efciency particu-
late air (HEPA) ltration (gaseous (n =10) vs. PM
2.5
pollution (n =9)).
Simultaneously, another group of rats was housed in the animal center
and supplied with HEPA-ltered clean air (control group (n =11)). The
air used in the whole-body exposure system was sourced from a nearby
highway and expressway in a trafc-heavy urban area (Taipei, Taiwan;
25◦1′5.2176’’N, 121◦32’17.8548’’E). The whole-body exposure system
and monitoring methods were described in detail in section S1 and
Fig. S1 in Supplementary Information (SI). Rat necropsies were per-
formed, and bronchoalveolar lavage (BALF), lung tissues and fecal
samples were collected after 3 months of exposure as previously
described (Li et al., 2009; Wang et al., 2018).
2.3. Lung function examination
Testing for the forced vital capacity (FVC), tidal volume, peak
expiratory ow (PEF), and forced expiratory volume in 20 ms (FEV
20
)
were performed in this study. Additionally, data for forced expiratory
ow at 25~75% of the FVC (FEF
25–75
) and the ratio of the FEV
20
to FVC
(FEV
20
/FVC) were presented in accordance with the Buxco pulmonary
maneuvers protocol (Hohlfeld et al., 2004; Jheng et al., 2021). The lung
function testing methods were described in section S2 in SI.
2.4. Ultra-performance liquid chromatography - tandem mass
spectrometer (UPLC-MS/MS)
Levels of benzotriazole, methamphetamine, methyl-1 H-benzo-
triazole, ketamine, ampicillin, ciprooxacin, pentoxifylline, erythro-
mycin, clarithromycin, ceftriaxone, penicillin G, and penicillin V were
determined in BALF using an ACQUITY UPLC System coupled with an
AI-4000 triple-quadrupole mass spectrometer (Danaher Corporation,
Washington, D.C., USA) equipped with an electrospray ionization (ESI)
source operated in positive mode. Multiple reaction monitoring (MRM)
mode was used for qualifying and quantitating target analytes by
applying nitrogen gas for collision. UCLP-MS/MS acquisition and
quantication was performed by Analytes (Sciex, Framingham, MA,
USA). Details of the analysis methods were described in section S3 in SI.
2.5. Inductively coupled plasma-mass spectrometry (ICP-MS)
Lung samples were freeze-dried using a manifold freeze-dryer
(UNISS, New Taipei City, Taiwan). ICP-MS (Agilent 7500, CA, USA)
was used to determine nine metal concentrations: beryllium (Be), cal-
cium (Ca), vanadium (V), chromium (Cr), manganese (Mn), cobalt (Co),
copper (Cu), Zinc (Zn), and barium (Ba). Details of the analysis methods
were described in section S4 in SI.
2.6. Microbiotic DNA preparation and analysis
To extract DNA of lung bacteria of each rat, the left lobe of the lung
was harvested under sterile conditions, and 10 mg fresh sample was used
in QIAamp DNeasy Blood & Tissue Kits (Qiagen, Hilden, Germany). For
each rat, at least two fecal samples were collected using sterilized
microtubes. All fecal samples were immediately stored at −80 ◦C. Fecal
samples of each rat were mixed, and we obtained 220 mg of the mixed
sample for further DNA extraction. The V3-V4 region of the 16 S ribo-
somal (r)DNA gene was amplied with specic primers, including Illu-
mina sequencing adapters and sample-specic barcodes, and sequenced
on an Illumina MiSeq sequencer. The universal primers 341 F (5′-
CCTACGGGNGGCWGCAG-3′) and 805 R (5′-GACTACHVGGGTATC-
TAATCC-3′) with Illumina overhang adapter sequences in the forward
(5′-TCGTCGGCAGCGTCAGATGTGTATAAGAGACAG-3′) and reverse
(5′-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAG-3′) primers were
used to amplify the V3-V4 highly variable region of the 16 S rDNA gene
sequence. The DNA extraction, 16 S rDNA gene amplication,
sequencing, and analysis methods were described in section S5 in SI.
2.7. Statistical analysis
For comparisons among multiple values, one-way analysis of vari-
ance (ANOVA) with Tukey’s post-hoc test was used. Alpha diversity
indices were calculated using estimated richness function from the
phyloseq package. Beta diversity was calculated by unweighted unique
fraction (UniFrac) principal coordinate analysis (PCoA). Microbiotic
enrichment analysis between groups was conducted using the linear
discriminant analysis (LDA) effect size (LEfSe) method, with alpha set to
0.05 (using the Kruskal-Wallis and Wilcoxon tests) and a logarithmic
LDA score of ≥2 (Segata et al., 2011); the analysis was visualized as
cladogram using GraPhlAn (Asnicar et al., 2015). Spearman’s correla-
tion coefcients were used to examine correlations of (1) lung functions
with the lung and intestinal microbiomes, (2) lung functions with anti-
biotics, drugs, and metals in the lung, (3) antibiotics and drugs with the
lung and intestinal microbiomes, (4) metals with the lung and intestinal
microbiomes, and (5) the lung microbiome with the intestinal micro-
biome. Visualization of Spearman’s correlation was done with heatmap
by RStudio (vers. 4.1.1) for macOS. Principal component analysis (PCA)
was applied to evaluate the effects of antibiotics, drugs, metals, lung and
intestinal microbiomes of lung functions in ageing rats. PCA was per-
formed by scikit-learn library of the Python software package (Pedre-
gosa et al., 2011). Statistical analyses were performed using GraphPad
vers. 9 for macOS. The level of signicance was set to p <0.05.
3. Results
3.1. Characterization of air pollution exposure
Daily distributions of PM
2.5
mass concentrations and particle number
concentrations (PNC) are shown in Fig. 1A. The PM
2.5
mass concentra-
tion was 8.7 ±4.2
μ
g/m
3
, and the geometric mean diameter (GMD) was
64.5 ±6.8 nm during the study period (Fig. 1B). The PNC was 6460.5 ±
2086.2 particles/cm
3
(Fig. 1B). Gaseous pollutants concentrations were
0.3 ±0.1 ppm for carbon monoxide (CO), 1.6 ±0.4 ppb for sulfur di-
oxide (SO
2
), 10.6 ±4.2 ppb for nitrogen dioxide (NO
2
), and 24.7 ±10.5
ppb for ozone (O
3
) (Fig. 1B). The temperature and RH during the study
period were 29.1 ±2.4 ºC and 73.4% ±4.7%, respectively.
3.2. Lung functions decline by air pollution
Rats exposed to PM
2.5
had signicantly decreased FVC compared to
HEPA group (p<0.05) (Fig. 2). The PEF was signicantly decreased in
the HEPA group compared to control and PM
2.5
groups (p<0.05).
PM
2.5
-exposed rats showed signicant decrease in FEV
20
compared to
the control and HEPA groups (p<0.05). Moreover, the FEF
25–75
was
signicantly decreased in the PM
2.5
group compared to control group
(p<0.05). However, there were no signicant differences in tidal vol-
ume or FEV
20
/FVC among the three exposure groups.
3.3. Increasing antibiotics and drugs by air pollution
Antibiotics and drugs concentrations in the BALF are shown in
Table 1. Concentrations of the 12 antibiotics and drugs (benzotriazole,
methamphetamine, methyl-1 H-benzotriazole, ketamine, ampicillin,
ciprooxacin, pentoxifylline, erythromycin, clarithromycin, ceftriax-
one, penicillin G, and penicillin V) were signicantly increased in HEPA
V. Laiman et al.

Ecotoxicology and Environmental Safety 246 (2022) 114164
4
group compared to control group (p<0.05). The 5 highest concentra-
tion of antibiotics and drugs found in HEPA group were ampicillin
(10.60 ng/mL), methyl-1 H-benzotriazole (9.31 ng/mL), benzotriazole
(9.25 ng/mL), ceftriaxone (8.97 ng/mL), and clarithromycin (8.94 ng/
mL). The 12 identied antibiotics and drugs were also signicantly
increased in PM
2.5
group compared to HEPA and control groups
(p<0.05). The 5 highest concentrations of antibiotics and drugs found
in PM
2.5
group were methyl-1 H-benzotriazole (41.28 ng/mL), meth-
amphetamine (36.10 ng/mL), penicillin V (31.97 ng/mL),
clarithromycin (31.17 ng/mL), and pentoxifylline (30.82 ng/mL).
3.4. Metals in the lungs
Metal concentrations in the lungs are shown in Table 2. Concentra-
tions of Be and Co were similar across the three groups. V (28.7
μ
g/g), Cr
(27.3
μ
g/g), and Ba (65.7
μ
g/g) concentrations were higher in HEPA
group, followed by control and PM
2.5
groups. Ca (297.7
μ
g/g) and Mn
(4.2
μ
g/g) concentrations were higher in control group than in HEPA
A.
B.
29/7
1/8
4/8
7/8
10/8
13/8
16/8
19/8
22/8
25/8
28/8
31/8
3/9
6/9
9/9
12/9
15/9
18/9
21/9
24/9
27/9
0
10
20
30
Date
PM
2.5
Concentrations ( g/m
3
)
CO SO
2
NO
2
O
3
0
1
2
3
0
20
40
60
80
CO (ppm) /SO
2
(ppb)
NO
2
(ppb) /O
3
(ppb)
PM2.5
Fig. 1. (A) Daily distributions of particulate matter with an aerodynamic diameter of ≤2.5 µm (PM2.5) mass concentrations and particle number concentrations
(PNCs). (B) Characteristics of the PM2.5 and gaseous pollution.
Fig. 2. Lung function examination including the forced vital capacity (FVC), tidal volume, peak expiratory ow (PEF), forced expiratory volume at 20 ms (FEV
20
),
FEF at 25~75% of the FVC (FEF
25–75
), and the ratio of FEV
20
and FVC (FEV
20
/FVC). * p<0.05.
V. Laiman et al.

Ecotoxicology and Environmental Safety 246 (2022) 114164
5
and PM
2.5
groups. Meanwhile, PM
2.5
group had the highest Cu con-
centration (47.1
μ
g/g), followed by HEPA and control groups. Zn con-
centration was higher in HEPA group (48.5
μ
g/g), followed by PM
2.5
and control groups. However, none of these differences in metal con-
centrations documented among the three groups was signicant.
3.5. Air pollution altered lung bacterial compositions
Alpha diversity was analyzed to compare lung microbiome richness
among the control, HEPA, and PM
2.5
groups (Fig. 3). As shown in
Fig. 3A, the richness estimator, Chao1, and beta diversity did not
signicantly differ among the three groups. At the phylum level, all
samples from the control, HEPA, and PM
2.5
groups contained three
major bacterial phyla: Proteobacteria, Bacteroidetes, and Firmicutes.
Proteobacteria accounted for nearly 90% of the total abundance in the
three groups (Fig. 3B). The phyla Fusobacteria and Verrucomicrobia
showed signicantly increased relative abundances in the PM
2.5
group
than the control group (p<0.05) (Fig. 3 C). In the family-level LEfSe
analysis, the relative abundance of Bacillaceae in HEPA and PM
2.5
group
was signicantly decreased compared to control group (p<0.05)
(Fig. 3D). However, the relative abundance of Atopobiaceae was
signicantly higher in PM
2.5
group, while it was zero in both control and
HEPA groups (p<0.05). Meanwhile, the relative abundances of
Akkermansiaceae and Fusobacteriaceae in PM
2.5
group were signi-
cantly increased compared to control group (p<0.05).
3.6. Air pollution altered the intestinal bacterial composition and diversity
Fig. 4 shows the intestinal bacterial diversity and composition among
the three groups. The HEPA group exhibited higher alpha diversity
compared to control and PM
2.5
groups (Fig. 4A). The PCoA plot shows
that the three groups were separated with 24.2% and 12.9% variation
explained by the PC1 and PC2 principal components, respectively. At the
phylum level, all samples from the three groups contained three major
bacterial phyla (Fig. 4B): Bacteroidetes, Firmicutes, and Proteobacteria.
In Fig. 4C, relative abundance of Bacteroidetes was signicantly
increased in PM
2.5
group compared to control group (p<0.05). How-
ever, the relative abundance of Firmicutes in PM
2.5
group was signi-
cantly decreased compared to control group (p<0.05). In addition, the
relative abundance of Proteobacteria in the HEPA group was signi-
cantly higher compared to control and PM
2.5
groups (p<0.05). We
calculated the Firmicutes/Bacteroidetes ratio to consider intestinal
dysbiosis. The ratio in the PM
2.5
group was signicantly decreased than
that of the control group (p<0.05) (Fig. 4D). In the family-level LEfSe
analysis, the relative abundances of Enterococcaceae and Burkholder-
iaceae in the HEPA group were signicantly increased than those of
control and PM
2.5
groups (p<0.05). Meanwhile, the relative abundance
of Bacteroidaceae in HEPA and PM
2.5
groups was signicantly increased
compared to control group (p<0.05). Relative abundance of Rike-
nellaceae in the PM
2.5
group was signicantly increased compared to
control and HEPA group (p<0.05). Additionally, the relative abun-
dance of Barnesiellaceae in the PM
2.5
group was signicantly increased
compared to control group (p<0.05).
3.7. Correlations of lung functions with lung and intestinal microbiome
Fig. 5A shows correlations between the lung functions with the lung
and intestinal microbiome. In lung microbiome, FEV
20
were negatively
correlated with both Fusobacteria in phylum level and Fusobacteriaceae
in family level (p<0.05). In intestinal microbiome, FVC and tidal vol-
ume was positively correlated with Actinobacteria (p<0.05). FEV
20
was most negatively correlated with Bacteroidaceae while positively
correlated with Firmicutes (p<0.05). FEV
25–75
was most negatively
correlated with Barnesiellaceae (p<0.05). FEV
20
/FVC was negatively
correlated with Enterococcaceae (p<0.05). PEF was most negatively
correlated with Burkholderiaceae (p<0.05).
3.8. Correlations of lung functions with antibiotics, drugs, and metals in
the lung
We analyzed correlation of the lung functions with antibiotics, drugs,
and metals in the lung (Fig. 5B). Both FEV
20
and FEV
25–75
were nega-
tively correlated with all antibiotics and drugs in the lung (p<0.05).
Additionally, FEV
20
/FVC was negatively correlated with Ketamine
(p<0.05). In the correlation of lung functions and metals, FEV
20
/FVC
and PEF were positively correlated with Co in the lung (p<0.05).
3.9. Correlations of antibiotics and drugs with the lung and intestinal
microbiome
Correlations of the antibiotics and drugs with the lung and intestinal
microbiome are shown in Fig. 5 C. In lung microbiome, Fusobacteria
was most positively correlated with erythromycin (p<0.05). Syn-
ergistetes and Verrucomicrobia in the phylum level and Akkermansia-
ceae in family level was most positively correlated with clarithromycin
(p<0.05). Both Atopobiaceae and Fusobacteriaceae are positively
correlated with all antibiotics and drugs in lungs (p<0.05). Bacillaceae,
Table 1
Antibiotic and drug concentrations in the bronchoalveolar lavage of ageing rats
among control, high-efciency particulate air (HEPA) and particulate matter
with an aerodynamic diameter of ≤2.5 µm (PM
2.5
) groups.
Control (ng/
mL)
HEPA (ng/
mL)
PM
2.5
(ng/mL)
Benzotriazole 0.10 ±0 9.25 ±2.49* 30.61 ±2.26*,
**
Methamphetamine 0.14 ±0.06 7.05 ±2.38* 36.10 ±4.45*,
**
Methyl-1 H-
benzotriazole
0.20 ±0.12 9.31 ±1.57* 41.28 ±1.87*,
**
Ketamine 0.08 ±0 8.40 ±3.24* 30.29 ±3.61*,
**
Ampicillin 0.12 ±0.07 10.60 ±2.65* 27.18 ±3.08*,
**
Ciprooxacin 0.03 ±0 5.37 ±1.40* 26.44 ±2.85*,
**
Pentoxifylline 0.09 ±0 6.21 ±1.54* 30.82 ±3.36*,
**
Erythromycin 0.15 ±0.08 4.86 ±0.57* 27.14 ±1.29*,
**
Clarithromycin 0.03 ±0 8.94 ±2.46* 31.17 ±4.60*,
**
Ceftriaxone 0.06 ±0 8.97 ±1.59* 30.03 ±3.21*,
**
penicillin G 0.10 ±0 4.89 ±1.13* 28.18 ±2.79*,
**
penicillin V 0.27 ±0.09 8.87 ±0.54* 31.97 ±1.57*,
**
* Signicantly different compared to control group at p < 0.05; ** Signicantly
different compared to HEPA group at p < 0.05.
Table 2
Metal concentrations in the lungs of ageing rats among control, high-efciency
particulate air (HEPA) and particulate matter with an aerodynamic diameter of
≤2.5 µm (PM
2.5
) groups.
Control (
μ
g/g) HEPA (
μ
g/g) PM
2.5
(
μ
g/g)
Be 4.1 ±0.2 4.1 ±0.3 4.1 ±0.3
Ca 298 ±122 293 ±83 281 ±168
V 22.2 ±14.3 28.7 ±11.9 19.8 ±14.0
Cr 23.9 ±7.4 27.3 ±7.0 23.6 ±8.3
Mn 4.2 ±6.5 3.8 ±2.3 4.1 ±3.1
Co 1.1 ±0.4 1.1 ±0.4 1.0 ±0.4
Cu 33.8 ±11.2 44.7 ±16.7 47.1 ±14.2
Zn 22.1 ±0.0 48.5 ±20.9 28.4 ±13.8
Ba 51.0 ±58.0 65.7 ±60.8 40.3 ±48.2
V. Laiman et al.

Ecotoxicology and Environmental Safety 246 (2022) 114164
6
however, was most negatively correlated with erythromycin (p<0.05).
In intestinal microbiome, Bacteroidetes was most positively correlated
with methyl-1 H-benzotriazole (p<0.05). Firmicutes, however, was
negatively correlated with all antibiotics and drugs in lung (p<0.05). In
the family level, Bacteroidaceae, Rikenellaceae, and Barnesiellaceae
were positively correlated with all antibiotics and drugs (p<0.05).
3.10. Correlations of metals with the lung and intestinal microbiomes
We analyzed correlations between examined metals and the lung
microbiome (Fig. 5D). Cr was negatively correlated with Atopobiaceae
while Co was negatively correlated with Verrucomicrobia and Akker-
mansiaceae of the lung microbiome (p<0.05). Next, we analyzed cor-
relations between the examined metals and the intestinal microbiome
(Fig. 5D). Ca was negatively correlated with Barnesiellaceae while Co
was most negatively correlated with Bacteroidetes (p<0.05). Positive
correlation was found between Cr and Enterococcaceae, Co and Firmi-
cutes, and Cu and Patescibacteria (p<0.05).
3.11. Correlations between the lung and intestinal microbiomes
Fig. 5E shows correlations between the lung and intestinal micro-
biomes. Negative correlations were observed between lung Actino-
bacteria and intestinal Proteobacteria as well as lung Bacillaceae and
intestinal Bacteroidaceae. Lung Epsilonbacteraeota was most negatively
correlated intestinal Fusobacteria while positively correlated with in-
testinal Patescibacteria (p<0.05). Both intestinal Rikenellaceae and
Barnesiellaceae were most positively correlated with the lung Atopo-
biaceae (p<0.05). The eigenvector plot demonstrates that the PCA
model had an excellent ability to distinguish samples between the three
groups (Fig. 5F). The PCA with the rst three principal components were
selected, accounting for 61% of the total variance. The rst principal
component (PC1), second principal component (PC2), and third prin-
cipal component (PC3) accounted for 38%, 12% and 11%, respectively.
4. Discussion
The signicance of this study is that we investigated the effects of
antibiotics, drugs, and metals on lung and intestinal microbiomes in
ageing rats. The novelty of this study is that we identied antibiotics,
drugs, and metals in the lungs after air pollution exposure were corre-
lated with lung function decline as well as dysbiosis of the lung and
intestinal microbiomes of ageing rats. The main ndings of this study
were: (1) air pollution signicantly increased antibiotic and drug con-
centration in the lung of ageing rats, (2) air pollution caused lung
function decline that correlated with increasing antibiotics, drugs, and
metals in lungs, (3) air pollution caused lung and intestinal microbiomes
imbalance that correlated with increasing antibiotics, drugs, and metals
in lungs, and (4) the lung microbiome was correlated with the intestinal
microbiome at several phylum and family levels after air pollution
exposure.
The average PM
2.5
concentration (8.7
μ
g/m
3
) in urban area of
northern Taiwan during the study period was in line with previous
studies for which values ranged 11.4–12.0
μ
g/m
3
in Taiwan (Ho et al.,
2018; Lee et al., 2021). The PM
2.5
concentration in our study was
0.6-times lower than the PM
2.5
standard concentration (mean 15
μ
g/m
3
in 24 h) from Air Quality Guidelines (AQG) by World Health Organi-
zation (WHO) in 2021 (WHO, 2021). PNC average of 6460.5
Fig. 3. (A) Lung microbiome analysis of alpha diversity and beta diversity among the three groups. (B) Composition of lung microbiome at the phylum level. (C)
Fusobacteria and Verrucomicrobia compositions among the three groups. (D) Compositions of the Bacillaceae, Atopobiaceae, Akkermansiaceae, and Fusobacter-
iaceae at the family level. * p<0.05.
V. Laiman et al.

Ecotoxicology and Environmental Safety 246 (2022) 114164
7
particles/cm
3
was observed in our study, which was relatively lower
than levels from previous studies in Taipei City ranging from 14,250 to
11,257 particles/cm
3
(Cheng et al., 2014; Shih et al., 2018). Gaseous
pollutants, including CO, SO
2
, NO
2
, and O
3
, were close to the recom-
mended 24-hour exposure level by WHO AQG (WHO, 2021). CO and
SO
2
levels were both 0.1-times lower than the WHO AQG (4 mg/m
3
and
40
μ
g/m
3
, respectively). The NO
2
concentration was 0.8-times of the
WHO recommendation (25
μ
g/m
3
), while O
3
was 0.5-times of the WHO
recommendation (100
μ
g/m
3
). The GMD of PM
2.5
was 64.5 nm in our
study, which was consistent with a previous report of 55.8 nm in Taipei
City (Shih et al., 2018). Taken together, ageing rats were sub-chronically
exposed to relatively lower air pollution during the study period.
Air pollution exposure signicantly reduced FVC, PEF, FEV
20
, and
FEF
25–75
in ageing rats in our study. Previous studies among elderly in
Taiwan reported that long term exposure to PM
2.5
mainly decrease the
FVC, which indicated restrictive lung disorder or loss of lung volume
(Chen et al., 2019; Moore, 2012). We observed that air pollution
decreased the PEF although the difference was not statistically signi-
cant. An increase in 24-hour PM
2.5
mean concentration was also asso-
ciated with decrease in morning and evening PEF in previous study on
patients with asthma, which was associated with airow limitation
(Pothirat et al., 2015; Yamazaki et al., 2011). Similar ndings with
decreased FEF
25–75
and FEV
20
/FVC in rats were reported after 6 months
of trafc-related air pollution exposure with an average ambient PM
2.5
of 19.7
μ
g/m
3
(Jheng et al., 2021). FEV
20
and FEF
25–75
are medium to
small airway function tests, which were the primary site of increased
resistance in early stages of airway obstruction (Jheng et al., 2021;
McFadden and Linden, 1972). These ndings indicated that air pollution
exposure particularly in the presence of PM
2.5
decreased the lung
function due to airow limitation in ageing rats.
Air pollution increased antibiotics and drugs levels in lungs of rats,
especially in the PM
2.5
group. Previous study discovered vancomycin,
ceftriaxone, and piperacillin (0.01–5 ng/L, 0.02–0.06 ng/L, and
0.02–1.39 ng/L, respectively) in the air by stationary and personal air
samplers in hospitals of Hungary and Sweden (Sessink et al., 2019).
Hamscher and colleagues discovered tylosin, tetracyclines, sulfametha-
zine, and chloramphenicol in dust samples with total concentrations of
up to 12.5 mg/kg collected from the animal house for two decades
(Hamscher et al., 2003). Evidence for antibiotic presence and trans-
mission was also reported in various environments. For example, 6 an-
tibiotics (oxytetracycline, tetracycline, noroxacin, ooxacin,
erythromycin, and amoxicillin) were found in the air, seawater, soil,
sediment, and groundwater of the Yangtze River in China (Dai et al.,
2022). They discovered that antibiotics in the air, groundwater, and soil
phases were inuenced and redistributed by antibiotics in the soil phase,
which could be achieved through complex mechanisms of adsorption,
diffusion, transformation, and dispersion (Dai et al., 2022; Li et al.,
2021b). An average concentration of 0.36 ng/m
3
of benzotriazole was
also previously reported to be found in parking garages in New York
(Xue et al., 2016). There are currently no published threshold limit
values indicating what levels are considered safe for air contamination
of antibiotics and drugs due to limited studies available (Sessink et al.,
2019). This suggest that antibiotics and drugs can be found in the air and
is further found in larger quantities in combination with PM
2.5
. Our
Fig. 4. (A) Intestinal microbiome analysis of alpha diversity and beta diversity among the three groups. (B) Compositions of the intestinal microbiome at the phylum
level. (C) Bacteroidetes, Firmicutes, and Proteobacteria compositions among the three groups. (D) Ratio of Firmicutes and Bacteroidetes and compositions of the
Enterococcaceae, Burkholderiaceae, Bacteroidaceae, Rikenellaceae, and Barnesiellaceae at the family level. * p<0.05.
V. Laiman et al.

Ecotoxicology and Environmental Safety 246 (2022) 114164
8
ndings further revealed that these accumulated antibiotics and drugs in
the lungs were occurred by inhalation of air pollution in the urban area.
Next, we observed alterations in V, Cr, Cu, Zn, and Ba concentrations
in the lungs following air pollution exposure although the changes were
not statistically signicant. A previous study observed increased Cu
levels in the lungs by exposure to 4 mg/mL of suspended atmospheric
PM
2.5
via tracheal perfusion for 12 weeks in rats (Li et al., 2015). Cu was
linked to vehicle brake friction and gasoline fuel through combination of
dense vehicular trafc enhancing the grinding of brake-lining derived
copper particulates, which may contribute to Cu enrichment (Manalis
et al., 2005; Ntziachristos et al., 2007). Zn, for example, was strongly
linked to diesel fuel and vehicle tire wear (Lin et al., 2005). Similarly,
our previous study found trace metals in PM
2.5
including V, Cr, Cu, and
Zn in an urban area in Taichung, Taiwan (Laiman et al., 2022). Similar
ndings reported the presence of V, Cr, and Ba in PM
2.5
in Zhongshan
station, Taipei, Taiwan near business areas with the highest population
and vehicle densities (Wang et al., 2020). The WHO air quality guide-
lines and the US environmental protection agency (EPA) regulatory
guideline set the permissible limit for several heavy metals in the
ambient air (Morakinyo et al., 2021). For example, in the summer, the
permissible limit for Cr and Cu in the ambient air were 0.309
μ
g/m
3
and
0.2
μ
g/m
3
, respectively. As the metals detected in this study were of
lung-deposited metals, further proling and analysis of metal composi-
tion in PM
2.5
may be needed to determine the metal levels in ambient
air. Nevertheless, our ndings indicated that these metals could accu-
mulate in the lungs as a result of trafc emissions.
We observed that alpha diversity and beta diversity did not show
signicant changes between control and air pollution-exposed ageing
rats. Previous study discovered an increase in beta diversity in the mice
lung microbiome after 6-months exposure of 16.2 mg/kg PM
2.5
via
intratracheal instillation (Li et al., 2020). Our ndings suggest that air
pollution exposure did not shift the microbiome richness and separation
of the microbiome diversity in the lungs. Next, the major bacterial phyla
in the lung microbiome were the Proteobacteria, Bacteroidetes, and
Firmicutes and were similar in abundance across the groups. Previous
study using C57Bl/6 mice treated with different diet and exposed to
either ltered air or mixture of gasoline and diesel engine exhaust for 30
days showed no signicant alterations in Firmicutes and Bacteroidetes
between the groups (Daniel et al., 2021). Additionally, these bacteria
account for more than 95% of the total abundance in the lung micro-
biome (Li et al., 2020; Wu et al., 2020). We discovered that exposure to
air pollution increased Fusobacteria and Veruccomicrobia in phylum
level. When compared to healthy subjects, the relative abundance of
Bacteroidetes and Fusobacteria was signicantly lower in COPD pa-
tients, whereas Firmicutes, Proteobacteria, and Actinobacteria were
signicantly higher (Ramsheh et al., 2021). The study also showed that
some genera in Proteobacteria, for example Haemophilus, and Veillo-
nella of Firmicutes did not differ between the COPD patients and healthy
Fig. 5. (A) Correlations heatmap between the lung functions and the lung and intestinal microbiome. (B) Correlations heatmap between the lung functions and the
antibiotics, drugs, and metals in the lung. (C) Correlations heatmap between antibiotics and drugs in the lung and the lung and intestinal microbiome. (D) Corre-
lations heatmap between metals in the lung and the lung and intestinal microbiome. (E) Correlations heatmap between the lung microbiome and the intestinal
microbiome. The depth of the color indicates strength of correlation coefcient (blue: positive correlation; red: negative correlation). Size of the point indicates the
correlation signicance. (F) Principal component analysis (PCA) between antibiotics, drugs, metals, lung and intestinal microbiomes of lung functions in control,
high-efciency particulate air (HEPA) ltration, and particulate matter with an aerodynamic diameter of ≤2.5 µm (PM
2.5
) rat groups.
V. Laiman et al.

Ecotoxicology and Environmental Safety 246 (2022) 114164
9
subjects. Although the microbiome prole in lung of human differs from
that of our air pollution-exposed rats, this suggests that air pollution can
also disrupt lung microbiome, and possibly leading to lung disease. At
the family level, air pollution exposure increased the Atopobiaceae,
Akkermansiaceae, and Fusobacteriaceae and decreased the Bacillaceae.
Air pollution contains particles, gases, and also variety of microbial
species (Qin et al., 2020). This could explain certain species, for example
the Atopobiaceae in our study, were discovered in the lungs of rats
exposed to air pollution but not in the control group. Exposure to
biomass fuel and motor vehicle exhaust (with respective 24.7 and
1.4 mg/m
3
PM
2.5
) for up to 7 months also altered the microbial
composition and pulmonary immunologic homeostasis in rat lungs (He
et al., 2017; Li et al., 2017). The lung microbiome was found to interact
with the airway immune system (Wu et al., 2017). Several phyla, for
example the Proteobacteria, were negatively associated with neutrophil
inltration, whereas the Actinobacteria was positively associated with
B-cell inltration. Rats exposed to 3 months of trafc-related air pollu-
tion caused signicant lung injury with increased alveolar spaces and
macrophage inltration (Jheng et al., 2021). Therefore, the lung
microbiome and airway immune system may act as intermediaries of
interactions between air pollution and lung injury, which should be
investigated in the future.
We observed an increase in the microbiome diversity and changes in
intestinal microbiome species in ageing rats after air pollution exposure.
An increase in microbiome diversity was also reported in mice exposed
to 16.3
μ
g/m
3
ambient PM
2.5
for 3 weeks (Mutlu et al., 2018).
Discrimination of the beta diversity was also observed between control
group and air pollution exposed groups in our study. Mice exposed to
276.2
μ
g/m
3
concentrated PM
2.5
exhibited clear lung microbiome
discrimination previously (Wang et al., 2018). We also discovered
perturbation in Bacteroidetes, Firmicutes, and Proteobacteria, three
major bacterial phyla in the intestinal microbiome. Intestinal bacterial
phyla composition alteration was also reported in mice exposed to
16.3
μ
g/m
3
PM
2.5
for 3 weeks; in particular, the Firmicutes became less
abundant, while the Bacteroidetes became more abundant (Mutlu et al.,
2018). The Firmicutes/Bacteroidetes ratio was signicantly decreased
after air pollution exposure in our study. The Firmicutes/Bacteroidetes
ratio was a dysbiosis marker, with an increase in the ratio associated
with obesity and a decrease in the ratio associated with inammatory
disease (Garshick et al., 2021; Stojanov et al., 2020). At the family level,
we observed increases of the Enterococcaceae, Burkholderiaceae, Bac-
teroidaceae, Rikenellaceae, and Barnesiellaceae upon air pollution
exposure in ageing rats. In various inammatory-sustained conditions,
alterations in intestinal microbiota composition occurred and in-
teractions between microbiome and intestinal cells are essential to shape
the immune system (Grigg and Sonnenberg, 2017; Singanayagam et al.,
2021). These ndings suggest that air pollution can result in intestinal
microbiome shifting and may be linked to decreased in lung function.
The lung functions were correlated with changes in lung and intes-
tinal microbiome in ageing rats. Altered lung microbiome composition
and impaired lung innate immunity were previously reported to be
correlated with worsening lung function (Healy et al., 2021; Rangelov
and Sethi, 2014). The dysbiosis in the intestinal microbiome was also
linked to immune response changes and development of lung disease
(Zhang et al., 2020). For example, intestinal microbiome transplanted
from COPD patients induced lung mucus hypersecretion in mice,
resulting in acceleration of lung function decline (Li et al., 2021a). We
further found that lung function decline were correlated with increasing
antibiotics and drugs in lung. Additionally, antibiotics and drugs in lung
were correlated with lung and intestinal microbiomes in the phylum and
family level. Administration of aerosol vancomycin and neomycin was
previously shown to alter the commensal lung microbiome (Le Noci
et al., 2018). Disruption in microbiome homeostasis was also reported to
allow some pathogens that normally enter the lower respiratory tract via
micro-aspiration or inhalation to overgrow (Sethi and Murphy, 2008;
Tian et al., 2022). These bacteria can cause inammation and lung
damage, which allows more bacteria to proliferate, resulting in a vicious
cycle of chronic infection that contributes to progressive lung damage
and function loss (Sethi and Murphy, 2008). These ndings suggested
that decreased in lung functions were correlated with changes in lung
and intestinal microbiome, which may be linked to the presence of an-
tibiotics and drugs in the lung. On the other hand, we found that the lung
function improvement were correlated with metals in lung. Metals in the
lung were also correlated with changes in the phylum and family com-
positions of the lung and intestinal microbiomes in ageing rats. Trace
metals are essential to all life forms, with changes in host metal avail-
ability affected the bacterial diversity in vivo (Andrei et al., 2020; Dostal
et al., 2013). Metals are essential for the growth and survival of
microbiome in healthy lungs, such as Bacteroidetes, Prevotella, and
Streptococcus (Healy et al., 2021). The lung resident microbiome also
requires iron and other metals to maintain their metabolism and struc-
tural functions. However, more research into how these metals
contribute to each species found in our study is required. Together, our
ndings suggested that while metals can improve certain microbiomes
and lung function, they ultimately lose out to other factors such as an-
tibiotics and drugs.
We observed strong correlations between the lung and the intestinal
microbiomes at phylum and family levels in ageing rats after air pollu-
tion exposure. Association between the intestinal and lung microbiomes
was previously reported, with essential involvement of mesenteric
lymphatic system transporting intact bacteria, including fragments and
metabolites, to other organs, such as lungs, to favor specic microbiome
(Enaud et al., 2020). For example, in COPD patients, the fecal micro-
biome was signicantly different from that of healthy controls, which
was associated with altered systemic metabolism and immunity, as well
as reduced lung function (Bowerman et al., 2020). Our ndings sug-
gested that air pollution exposure altered microbiome via lung-gut axis,
which lead to lung function decline. We also observed that rats exposed
to air pollution tended to have microbiome composition cluster sepa-
rately. Previous study also showed that mice exposed to 8, 16, and 24
weeks of concentrated ambient PM
2.5
and ltered air demonstrated
separate microbiome cluster (Xie et al., 2022). Our ndings suggest that
the antibiotics, drugs, and metals in lung can signicantly impact the
lung and intestinal microbiome, leading to lung function decline.
There are some limitations in our work. The effects of other chem-
icals in PM
2.5
such as organics were not determined in this study. Thus,
alterations in metals in the lungs by external air pollution exposure or
internal metal balance were unclear in this study. Inammatory re-
sponses in lung and intestinal samples were not examined in this work.
Future research may be required to determine how the gut-lung axis
interacts with the specic microbiota identied in this study, which may
aid in a better understanding each microbiota role in shaping
homeostasis.
5. Conclusions
In conclusion, the antibiotics, drugs, and metals in the lung caused
dysbiosis in the lung and intestinal microbiomes of ageing rats after sub-
chronic exposure of low-level air pollution, which led to lung function
decline. The deposition of antibiotics, drugs, and metals in lung with the
lung function decline demonstrated distinct separate microbiome clus-
ter. Despite the limitations listed above, this study provides motivation
for additional research to dene limits for antibiotics, drugs, and metals
in air pollution. Further understanding of the patho-mechanisms by
which antibiotics, drugs, and metals inuence microbiome changes and,
as a result, lung impairment will aid in disease understanding and the
development of novel preventive strategies. Our ndings suggest that
urban air pollution, particularly PM
2.5
, could be an important determi-
nant of lung impairment and shifts in the lung and intestinal micro-
biomes in respiratory disease.
V. Laiman et al.

Ecotoxicology and Environmental Safety 246 (2022) 114164
10
Ethics approval
All animal protocols were prepared in accordance with the Guide for
the Care and Use of Laboratory Animals and were approved (IACUC: LAC-
2019-0424) by the Laboratory Animal Center at Taipei Medical Uni-
versity (Taipei, Taiwan).
Funding
This study was funded by the Ministry of Science and Technology of
Taiwan (108-2314-B-038-093 and 109-2314-B-038-093-MY3) and Tai-
pei Medical University (DP2-111-21121-01-T-01-02).
Authors’ contributions
All authors contributed substantially to the concept and design of the
study, drafting of the article, and critically revising the manuscript for
important intellectual content. All authors have read and approved the
nal version of the manuscript for publication.
CRediT authorship contribution statement
Hsin-Chang Chen, Yu-Chun Lo, Jer-Hwa Chang, and Hsiao-Chi
Chuang: Conceptualization, Methodology, Software. Vincent Laiman,
Ching-Wen Chang, Tzu-Hsuen Yuan, Jen-Kun Chen, Ta-Chih Hsiao,
Ting-Chun Lin, and Ssu-Ju Li: Data curation, Writing – original draft.
You-Yin Chen: Visualization, Investigation. Jer-Hwa Chang and
Hsiao-Chi Chuang: Supervision. Kai-Jen Chuang and Kin-Fai Ho:
Software, Validation. Didik Setyo Heriyanto, Kian Fan Chung, and
Hsiao-Chi Chuang: Writing – review & editing.
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 datasets used and/or analyzed during the current study are
available from the corresponding author on reasonable request.
Acknowledgements
The authors wish to thank Xiao-Yue Chen for technical assistance
with this research. We also would like to acknowledge the technological
and analytical support provided by the TMU Core Laboratory of Human
Microbiome.
Appendix A. Supporting information
Supplementary data associated with this article can be found in the
online version at doi:10.1016/j.ecoenv.2022.114164.
References
Andrei, A., et al., 2020. Cu homeostasis in bacteria: the ins and outs. Membranes 10.
Asnicar, F., et al., 2015. Compact graphical representation of phylogenetic data and
metadata with GraPhlAn. PeerJ 3.
Bowerman, K.L., et al., 2020. Disease-associated gut microbiome and metabolome
changes in patients with chronic obstructive pulmonary disease. Nat. Commun. 11.
Carvalho, T.C., et al., 2011. Inuence of particle size on regional lung deposition – what
evidence is there? Int. J. Pharm. 406, 1–10.
Chang, J.-H., et al., 2022. Air pollution-regulated E-cadherin mediates contact inhibition
of proliferation via the hippo signaling pathways in emphysema. Chem. Biol.
Interact. 351.
Chen, C.-H., et al., 2019. The effects of ne and coarse particulate matter on lung
function among the elderly. Sci. Rep. 9.
Cheng, Y.H., et al., 2014. Correlations between black carbon mass and size–resolved
particle number concentrations in the Taipei urban area: a ve–year long–term
observation. Atmos. Pollut. Res. 5, 62–72.
Chuang, H.-C., et al., 2020. Alteration in angiotensin-converting enzyme 2 by PM1
during the development of emphysema in rats. ERJ Open Res. 6.
Dai, C., et al., 2022. Simulation and risk assessment of typical antibiotics in the multi-
media environment of the Yangtze River Estuary under tidal effect. Environ. Sci.
Pollut. Res.
Daniel, S., et al., 2021. Trafc generated emissions alter the lung microbiota by
promoting the expansion of Proteobacteria in C57Bl/6 mice placed on a high-fat
diet. Ecotoxicol. Environ. Saf. 213.
Dostal, A., et al., 2013. Low iron availability in continuous in vitro colonic fermentations
induces strong dysbiosis of the child gut microbial consortium and a decrease in
main metabolites. FEMS Microbiol. Ecol. 83, 161–175.
Enaud, R., et al., 2020. The gut-lung axis in health and respiratory diseases: a place for
inter-organ and inter-kingdom crosstalks. Front. Cell. Infect. Microbiol. 10.
Fukuchi, Y., 2009. The aging lung and chronic obstructive pulmonary disease: similarity
and difference. Proc. Am. Thorac. Soc. 6, 570–572.
Garshick, M.S., et al., 2021. Reshaping of the gastrointestinal microbiome alters
atherosclerotic plaque inammation resolution in mice. Sci. Rep. 11.
Grigg, J.B., Sonnenberg, G.F., 2017. Host-microbiota interactions shape local and
systemic inammatory diseases. J. Immunol. 198, 564–571.
Hamscher, G., et al., 2003. Antibiotics in dust originating from a pig-fattening farm: a
new source of health hazard for farmers? Environ. Health Perspect. 111, 1590–1594.
He, F., et al., 2017. Exposure to ambient particulate matter induced COPD in a rat model
and a description of the underlying mechanism. Sci. Rep. 7.
Healy, C., et al., 2021. Nutritional immunity: the impact of metals on lung immune cells
and the airway microbiome during chronic respiratory disease. Respir. Res. 22.
Ho, W.-Y., et al., 2018. Application of positive matrix factorization in the identication
of the sources of PM2.5 in Taipei city. Int. J. Environ. Res. Public Health 15.
Hohlfeld, J.M., et al., 2004. Keratinocyte growth factor transiently alters pulmonary
function in rats. J. Appl. Physiol. 96, 704–710.
Hu, J., et al., 2018. Metagenomic proling of ARGs in airborne particulate matters
during a severe smog event. Sci. Total Environ. 615, 1332–1340.
Jheng, Y.-T., et al., 2021. Prolonged exposure to trafc-related particulate matter and
gaseous pollutants implicate distinct molecular mechanisms of lung injury in rats.
Part. Fibre Toxicol. 18.
Kuhn, E.J., et al., 2019. Household contamination with methamphetamine: knowledge
and uncertainties. Int. J. Environ. Res. Public Health 16.
Laiman, V., et al., 2022. Contributions of acidic ions in secondary aerosol to PM2.5
bioreactivity in an urban area. Atmos. Environ. 275.
Le Noci, V., et al., 2018. Modulation of pulmonary microbiota by antibiotic or probiotic
aerosol therapy: a strategy to promote immunosurveillance against Lung Metastases.
Cell Rep. 24, 3528–3538.
Lee, S.-H., et al., 2021. Three month inhalation exposure to low-level PM2.5 induced
brain toxicity in an Alzheimer’s disease mouse model. PLoS One 16.
Li, J., et al., 2020. PM2.5 exposure perturbs lung microbiome and its metabolic prole in
mice. Sci. Total Environ. 721.
Li, N., et al., 2009. The adjuvant effect of ambient particulate matter is closely reected
by the particulate oxidant potential. Environ. Health Perspect. 117, 1116–1123.
Li, N., et al., 2017. Exposure to ambient particulate matter alters the microbial
composition and induces immune changes in rat lung. Respir. Res. 18.
Li, N., et al., 2021a. Gut microbiota dysbiosis contributes to the development of chronic
obstructive pulmonary disease. Respir. Res. 22.
Li, Q., et al., 2015. The preferential accumulation of heavy metals in different tissues
following frequent respiratory exposure to PM2.5 in rats. Sci. Rep. 5.
Li, X., Liu, X., 2021. Effects of PM2.5 on chronic airway diseases: a review of research
progress. Atmosphere 12.
Li, Z., et al., 2021b. Occurrence, sources and fate of pharmaceuticals and personal care
products and articial sweeteners in groundwater. Environ. Sci. Pollut. Res. 28,
20903–20920.
Lin, C.-C., et al., 2005. Characteristics of metals in nano/ultrane/ne/coarse particles
collected beside a heavily trafcked road. Environ. Sci. Technol. 39, 8113–8122.
Loghin, F., et al., 2020. Antibiotics in the environment: causes and consequences. Med.
Pharm. Rep. 93, 231–240.
Mammen, M.J., Sethi, S., 2016. COPD and the microbiome. Respirology 21, 590–599.
Manalis, N., et al., 2005. Toxic metal content of particulate matter (PM10), within the
Greater Area of Athens. Chemosphere 60, 557–566.
McFadden, E.R., Linden, D.A., 1972. A reduction in maximum mid-expiratory ow rate.
Am. J. Med. 52, 725–737.
Moore, V.C., 2012. Spirometry: step by step. Breathe 8, 232–240.
Morakinyo, O.M., et al., 2021. Health risk analysis of elemental components of an
industrially emitted respirable particulate matter in an urban area. Int. J. Environ.
Res. Public Health 18.
Muhammad, J., et al., 2019. Antibiotics in poultry manure and their associated health
issues: a systematic review. J. Soils Sediment. 20, 486–497.
Mutlu, E.A., et al., 2018. Inhalational exposure to particulate matter air pollution alters
the composition of the gut microbiome. Environ. Pollut. 240, 817–830.
Ntziachristos, L., et al., 2007. Fine, ultrane and nanoparticle trace element
compositions near a major freeway with a high heavy-duty diesel fraction. Atmos.
Environ. 41, 5684–5696.
Pedregosa, F., et al., 2011. Scikit-learn: machine learning in python. JMLR 12,
2825–2830.
Pothirat, C., et al., 2015. Peak expiratory ow rate as a surrogate for forced expiratory
volume in 1 second in COPD severity classication in Thailand. Int. J. Chronic Obstr.
Pulm. Dis. 10, 1213–1218.
V. Laiman et al.

Ecotoxicology and Environmental Safety 246 (2022) 114164
11
Qin, N., et al., 2020. Longitudinal survey of microbiome associated with particulate
matter in a megacity. Genome Biol. 21.
Ramsheh, M.Y., et al., 2021. Lung microbiome composition and bronchial epithelial gene
expression in patients with COPD versus healthy individuals: a bacterial 16S rRNA
gene sequencing and host transcriptomic analysis. Lancet Microbe 2, e300–e310.
Rangelov, K., Sethi, S., 2014. Role of Infections. Clin. Chest Med. 35, 87–100.
Saldarriaga-Nore˜
na, H., et al., 2009. Characterization of trace metals of risk to human
health in airborne particulate matter (PM2.5) at two sites in Guadalajara, Mexico.
J. Environ. Monit. 11.
Segata, N., et al., 2011. Metagenomic biomarker discovery and explanation. Genome
Biol. 12.
Sengupta, P., 2013. The laboratory rat: relating its age with human’s. Int. J. Prev. Med. 4,
624–630.
Sessink, P.J.M., et al., 2019. Reduction of contamination with antibiotics on surfaces and
in environmental air in three european hospitals following implementation of a
closed-system drug transfer device. Ann. Work Expo. Health 63, 459–467.
Sethi, S., Murphy, T.F., 2008. Infection in the pathogenesis and course of chronic
obstructive pulmonary disease. N. Engl. J. Med. 359, 2355–2365.
Shih, C.-H., et al., 2018. Chronic pulmonary exposure to trafc-related ne particulate
matter causes brain impairment in adult rats. Part. Fibre Toxicol. 15.
Singanayagam, A., et al., 2021. Comprehensive proling of the gut microbiota in patients
with chronic obstructive pulmonary disease of varying severity. PLoS One 16.
Stojanov, S., et al., 2020. The inuence of probiotics on the rmicutes/bacteroidetes
ratio in the treatment of obesity and inammatory bowel disease. Microorganisms 8.
Tian, Z., et al., 2022. Dynamic alterations in the lung microbiota in a rat model of
lipopolysaccharide-induced acute lung injury. Sci. Rep. 12.
Wang, L., et al., 2019. Airway microbiome is associated with respiratory functions and
responses to ambient particulate matter exposure. Ecotoxicol. Environ. Saf. 167,
269–277.
Wang, W., et al., 2018. Exposure to concentrated ambient PM2.5 alters the composition
of gut microbiota in a murine model. Part. Fibre Toxicol. 15.
Wang, Y.-S., et al., 2020. Explore regional PM2.5 features and compositions causing
health effects in Taiwan. Environ. Manag. 67, 176–191.
WHO, 2021. WHO Global Air Quality Guidelines: Particulate Matter (PM2.5 and PM10),
Ozone, Nitrogen Dioxide, Sulfur Dioxide and Carbon Monoxide. World Health
Organization, Geneva.
Wright, J., et al., 2020. Current practices underestimate environmental exposures to
methamphetamine: inhalation exposures are important. J. Expo. Sci. Environ.
Epidemiol. 31, 45–52.
Wu, B.G., et al., 2017. Lung microbiota and its impact on the mucosal immune
phenotype. Microbiol. Spectr. 5.
Wu, X., et al., 2020. Exposure to trafc-related PM2.5 pollutants signicantly affect the
diversity and quantity of lung microbiota in a rat model. IOP Conf. Ser. Earth
Environ. Sci. 601.
Xie, S., et al., 2022. Exposure to concentrated ambient PM2.5 (CAPM) induces intestinal
disturbance via inammation and alternation of gut microbiome. Environ. Int. 161.
Xue, J., et al., 2016. Occurrence of benzotriazoles (BTRs) in indoor air from Albany, New
York, USA, and its implications for inhalation exposure. Toxicol. Environ. Chem. 99,
402–414.
Yamazaki, S., et al., 2011. Effect of hourly concentration of particulate matter on peak
expiratory ow in hospitalized children: a panel study. Environ. Health 10.
Zeb, B., et al., 2022. Variation in coarse particulate matter (PM10) and its
characterization at multiple locations in the semiarid region. Front. Environ. Sci. 10.
Zhang, D., et al., 2020. The cross-talk between gut microbiota and lungs in common lung
diseases. Front. Microbiol. 11.
Zheng, P., et al., 2020. The impact of air pollution on intestinal microbiome of asthmatic
children: a panel study. BioMed. Res. Int. 2020, 1–13.
V. Laiman et al.