Article
The acute effects of fine particulate matter constituents on blood inflammation and coagulation Cong Liu, Jing Cai, Renjie Chen, Liping Qiao, Hongli Wang, Wenxi Xu, Huichu Li, Zhuohui Zhao, and Haidong Kan Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.7b00312 • Publication Date (Web): 16 Jun 2017 Downloaded from http://pubs.acs.org on June 18, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 42
Environmental Science & Technology
The acute effects of fine particulate matter constituents on blood inflammation and coagulation
Authors: Cong Liu1, †, Jing Cai1, 2, †, Liping Qiao3, Hongli Wang3, Wenxi Xu4, Huichu Li1, Zhuohui Zhao1, Renjie Chen1, 2,*, Haidong Kan1, 5,*
†
These authors contributed equally to this work.
Affiliations: 1. School of Public Health, Key Lab of Public Health Safety of the Ministry of Education and Key Lab of Health Technology Assessment of the Ministry of Health, Fudan University, Shanghai 200032, China; 2. Shanghai Key Laboratory of Meteorology and Health, Shanghai 200030, China; 3. State Environmental Protection Key Lab of the Formation and Prevention of Urban Air Pollution Complex, Shanghai Academy of Environmental Sciences, Shanghai 200233, China; 4. Huangpu District Center for Disease Control and Prevention, Shanghai 200023, China; 5. Key Laboratory of Reproduction Regulation of National Population and Family Planning Commission, Shanghai Institute of Planned Parenthood
ACS Paragon Plus Environment
Environmental Science & Technology
Page 2 of 42
Research, Institute of Reproduction and Development, Fudan University, Shanghai 200032, China
*Address correspondence to: Dr. Haidong Kan, P.O. Box 249, 130 Dong-An Road, Shanghai 200032, China. Tel/fax: +86 (21) 5423 7908. E-mail:
[email protected]; Dr. Renjie Chen, P.O. Box 249, 130 Dong-An Road, Shanghai
200032,
China.
Tel/fax:
+86
(21)
[email protected].
ACS Paragon Plus Environment
5423
7908.
E-mail:
Page 3 of 42
Environmental Science & Technology
1
Abstract
2
Limited evidence is available on the effects of various fine particulate matter
3
(PM2.5) constituents on blood inflammation and coagulation. We examined
4
the associations between 10 constituents and 10 circulating biomarkers in a
5
panel of 28 urban residents with 4 repeated measurements in Shanghai,
6
China.
7
single-constituent models, the constituent-PM2.5 joint models, and the
8
constituent-residual models to evaluate the associations between PM2.5
9
constituents and 8 inflammatory biomarkers (fibrinogen, C-reactive protein,
10
monocyte chemoattractant protein-1, tumor necrosis factor-α, interleukin-1b,
11
intercellular adhesion molecule-1, P-selectin, vascular cell adhesion
12
molecule-1) and 2 coagulation biomarkers (plasminogen activator inhibitor-1
13
and soluble CD40 ligand). We found robust associations of organic carbon
14
(OC), elemental carbon (EC), nitrate (NO3-), and ammonium (NH4+) with at
15
least 1 of 8 inflammatory markers. On average, an interquartile range
16
increase in the four constituents corresponded to increments of 50%, 37%, 25%
17
and 26% in inflammatory biomarkers, respectively. Only sulfate (SO42-) or
18
NH4+ was robustly associated with coagulation markers (corresponding
19
increments: 23% and 20%). Our results provided evidence that some
20
constituents in PM2.5 (OC, EC, NO3-, SO42- and NH4+) might play crucial roles
21
in inducing systematic inflammation and coagulation, but their roles varied by
22
the selected biomarkers.
Based
on
the
linear
mixed-effect
models,
ACS Paragon Plus Environment
we
fitted
the
Environmental Science & Technology
23
Keywords: fine particulate matter; chemical constituent; inflammation;
24
coagulation; biomarker; panel study
25 26
Word count: abstract (200 words) + text (3800 words) + 2 tables (600 words)
27
and 4 figures (2400 words) = 7000 words
28
ACS Paragon Plus Environment
Page 4 of 42
Page 5 of 42
Environmental Science & Technology
29
Introduction
30
Associations between short-term exposure to fine particulate matter (PM2.5)
31
and
32
worldwide.1-3 PM2.5 has very complex chemical compositions and it was thus
33
crucial to determine which constituents dominate the effects of PM2.5 on the
34
cardiovascular system.1,
35
effects of specific constituents are scarce and limited to a small fraction of
36
constituents, especially in developing countries.5, 6
cardiovascular
diseases
4
(CVDs)
have
been
well
documented
However, investigations of the cardiovascular
37
An increasing number of studies have attempted to elucidate the time
38
courses during which PM2.5 exposure causes adverse cardiovascular
39
outcomes. These studies have focused on the effects of sub-daily PM2.5
40
exposure on clinical or subclinical outcomes, such as cardiac arrest,
41
myocardial infarction, ST-segment depression, arrhythmia, fibrillation, and
42
increased blood pressure.7 Systemic inflammation and hypercoagulability are
43
two common mechanisms among a number of possible biological pathways
44
whereby PM2.5 adversely affects the cardiovascular system,8-10 but such
45
evidence has been limited with regard to time course. Our previous studies
46
have demonstrated the acute effects of sub-daily PM2.5 exposure on an array
47
of relevant biomarkers,11-13 but little knowledge is available on the sub-daily
48
exposure to various PM2.5 constituents.
49
As the largest developing country in the world, China is facing enormous
50
public health challenges due to severe air pollution problems and the heavy
ACS Paragon Plus Environment
Environmental Science & Technology
51
burden of CVDs. We therefore designed this longitudinal study in Shanghai,
52
China, to explore the short-term associations of PM2.5 constituents on blood
53
inflammation and coagulation, and further, to deduce which constituents are
54
most deleterious to the cardiovascular system. Elderly patients with chronic
55
obstructive pulmonary disease (COPD) were selected because they are
56
hypothesized to be susceptible to the adverse cardiovascular effects of air
57
pollutants as they have a higher deposition of particles and an inherent
58
inflammatory state.6
59 60
Material and methods
61
Study Design and Participants
62
We initially recruited 30 volunteers from a community-based registry of COPD
63
patients in Shanghai. The sample size was determined to be comparable with
64
previous panel studies.11, 12, 14, 15 Two patients were excluded because they
65
took medication due to exacerbation of COPD condition during the study
66
period. Details on the subject recruitment and study design have been
67
described in our previous publication.16 Briefly, all COPD patients were
68
diagnosed by physicians. We only included the stable patients with
69
mild-to-moderate COPD in this study according to the classification of the
70
Global Initiative for Chronic Obstructive Lung Disease based on the baseline
71
spirometry test, and we excluded those who were current active or passive
72
smokers (living with a current smoker), consumed any alcohol, or had severe
ACS Paragon Plus Environment
Page 6 of 42
Page 7 of 42
Environmental Science & Technology
73
comorbidities or inflammatory diseases. All the participants had a predicted
74
forced expiratory volume 1 (FEV1) ≥ 50% and an FEV1/forced vital capacity
75
(FVC) < 0.70. Six weekly follow-up visits were scheduled from May 27 to July
76
5, 2014, but we only arranged 4 blood collection appointments at 1-week
77
intervals due to the subjects’ refusal to have more blood drawn. For each
78
patient, blood collection was scheduled for the consecutive 4 weeks at the
79
same time (1:30 p.m. to 2:30 p.m.) on the same day of week to control for
80
possible circadian rhythms. Data on individual characteristics (such as age,
81
gender, height, weight, educational attainment, income, medication use, and
82
history of chronic morbidities) were collected at baseline. The study protocol
83
was approved by the Institutional Review Board of the School of Public
84
Health of Fudan University, and written informed consent form was obtained
85
from each subject.
86
Blood collection and lab analysis
87
During each follow-up, venous peripheral blood samples (5 ml) were drawn
88
by a certified nurse using coagulant vacuum tubes and then were rapidly
89
separated into serum and plasma by centrifugation at 4,000 rpm for 10
90
minutes within 20 minutes of collection. Serum samples were transported
91
directly to our laboratory and stored at -80°C before analysis.
92
We analyzed 10 circulating biomarkers associated with particulate air
93
pollution in at least 1 panel study.11, 14, 16 These included: 1) 8 biomarkers of
94
inflammation, including fibrinogen, C-reactive protein (CRP), P-selectin,
ACS Paragon Plus Environment
Environmental Science & Technology
95
monocyte chemoattractant protein (MCP)-1, interleukin-1b (IL-1β), tumor
96
necrosis factor (TNF)-a, intercellular adhesion molecule-1 (ICAM-1), and
97
vascular cell adhesion molecule-1 (VCAM-1); and 2) 2 biomarkers of
98
coagulation, soluble CD40 ligand (sCD40L) and plasminogen activator
99
inhibitor-1 (PAI-1). These biomarkers were measured by a commercial
100
Millipore MILLIPLEX MAP human cytokine/chemokine kit (Millipore Corp.,
101
Billerica, MA), which is based on the Luminex xMAP technology. The level for
102
each biomarker was simultaneously quantified using the MAGPIX system and
103
xPONENT software 134 (Luminex, Austin, TX). The lower limits of
104
quantitation (LLOQ) of the biomarkers varied from 0.01 pg/ml to 1.00 pg/ml.
105
Measurements lower than the LLOQ (8.5%) were replaced by half of the
106
LLOQ. All biomarker tests were performed under the same conditions
107
according to the manufacturer’s instructions, and all results were within the
108
quality control ranges.
109
Environmental Data
110
During the study period (from May 27 to July 5, 2014), we obtained real-time
111
(hourly) concentrations of PM2.5 and its constituents from a fixed-site monitor,
112
which was located on the rooftop of a 5-story building at the Shanghai
113
Academy of Environmental Sciences (approximately 4 km away from the
114
community). The two sites were mostly surrounded by commercial properties
115
and residential dwellings, and were not in the direct vicinity of main roadways,
116
industrial pollution, or other local pollution sources. The mass concentration
ACS Paragon Plus Environment
Page 8 of 42
Page 9 of 42
Environmental Science & Technology
117
of PM2.5 was measured by an online particulate monitor (FH 62 C14 series,
118
Thermo Fisher Scientific, Inc.) equipped with a verified PM2.5 cyclone using
119
beta attenuation techniques. Organic carbon (OC) and elemental carbon (EC)
120
were measured using a semi-continuous OC/EC analyzer (model RT-4,
121
Sunset Laboratory, Inc.) equipped with a PM2.5 cyclone and an upstream
122
parallel-plate organic denuder (Sunset Laboratory Inc.). The concentrations
123
of 8 major water-soluble inorganic ions, including chlorine (Cl−), nitrate (NO3−),
124
sulfate (SO42−), ammonium (NH4+), sodium (Na+), potassium (K+), magnesium
125
(Mg2+), and calcium (Ca2+), were measured by a commercial instrument for
126
online monitoring of aerosols and gases (MARGA, model ADI 2080, Applikon
127
Analytical B.V.). The quality assurance/quality control procedures were
128
routinely conducted, including maintenance/cleaning for this instrument as
129
well as calibrations for air flow rate, mass foil, and temperature/pressure. The
130
time resolution was 1 hour for each sample, with 45 min of sampling and 15
131
min of analysis. The principle and operation of this instrument have been
132
provided in detail elsewhere.17-19
133
Daily mean temperature and mean relative humidity were collected from
134
the Shanghai Meteorological Bureau to allow for the adjustment of weather
135
conditions. We also collected hourly concentrations of gaseous pollutants,
136
including sulfur dioxide (SO2), nitrogen dioxide (NO2), ozone (O3), and carbon
137
monoxide (CO), from one fixed-site national monitoring station, which is 1.8
138
km away from the community.
ACS Paragon Plus Environment
Environmental Science & Technology
139
Statistical Analyses
140
Environmental and health data were linked by the time of blood sampling. All
141
biomarker measurements were natural log-transformed to improve the
142
normality before statistical analyses. We applied the linear mixed-effect
143
model to evaluate the associations between biomarkers and PM2.5. In the
144
basic model, PM2.5 and its components were incorporated one at a time as
145
the fixed-effect terms. We also incorporated several covariates as fixed-effect
146
terms: (1) an indicator variable of “week” of blood collections to exclude any
147
unknown weekly time trends; (2) an indicator variable of “day of the week” to
148
control for the potential day-of-week effects; (3) the moving average of mean
149
temperature and relative humidity on the current day and previous 3 days to
150
adjust for the confounding effects of weather conditions; and (4) individual
151
characteristics, such as age, gender, body mass index, education, and the
152
history of morbidities. Finally, a random intercept was introduced to account
153
for the within-subject correlations due to repeated measurements. To fully
154
capture the time-lag patterns in the effects of PM2.5 and its various
155
constituents, we fit the above models using multiple separate intervals
156
preceding the blood draw: 0 to 6 h, 7 to 12 h, 13 to 24 h, 0 to 24 h (lag 0 day),
157
25 to 48 h (lag 1 day), 49 to 72 h (lag 2 days) and 3 to 7 days.
158
In addition to the basic single-constituent model described above, we
159
built a “constituent-PM2.5 joint model” with the adjustment of total PM2.5 mass
160
to account for potential confounding by PM2.5 and other constituents that
ACS Paragon Plus Environment
Page 10 of 42
Page 11 of 42
Environmental Science & Technology
161
co-vary with PM2.5. However, it usually leads to underestimation of the effects
162
for a specific constituent due to the over-adjustment with respect to the strong
163
correlations with a constituent and PM2.5.20 We thus further fitted a
164
“constituent-residual model”, which has the advantage of eliminating
165
confounding and extraneous variation by total PM2.5, as well as collinearity
166
with the remaining constituents. In this model, we first obtained the residual of
167
each constituent by establishing a linear regression model between total
168
PM2.5 and the constituent, and then introduced the residual into the basic
169
model replacing this constituent. The constituent residual can be regarded as
170
a crude measure of the independent contribution of a constituent to the
171
effects of PM2.5 after excluding its collinearity of the remaining constituents.
172
To test the robustness of our results on the adjustment for concomitant
173
exposure to gaseous pollutants, we performed a sensitivity analysis by
174
including
175
single-constituent models individually.
4
gaseous
pollutants
(CO,
NO2, SO2
and
O3 )
in
the
176
The statistical tests were two-sided, and values of P < 0.05 were
177
considered statistically significant. All models were performed using R
178
software (Version 3.3.0, R Foundation for Statistical Computing, Vienna,
179
Austria) with the “lme4” package. The estimates for blood biomarkers were
180
calculated as the percent changes and their 95% confidence intervals (CIs)
181
associated with an interquartile range (IQR) increase in PM2.5 concentrations.
182
ACS Paragon Plus Environment
Environmental Science & Technology
183
Results
184
Descriptive Statistics
185
We obtained all the scheduled blood samples for 28 subjects. Details of the
186
descriptive characteristics of the participants have been provided in our
187
previous publication.16 Briefly, on average, the participants were 64 years old
188
with a body mass index of 24.7 kg/m2. Twelve patients had comorbid
189
hypertension, and they all had a regular intake of antihypertensive
190
medications. According to self-reported questionnaires, none of the subjects
191
participated in strenuous physical activities, had an exacerbation of COPD,
192
took anti-COPD medication, or traveled out of the central urban areas of
193
Shanghai 3 days before the scheduled blood collection.
194
We tested the levels of 10 biomarkers in a total of 112 blood samples.
195
Table 1 provides the summary statistics of 8 inflammatory biomarkers and 2
196
coagulation biomarkers. There are considerable variations of these cytokines
197
in and between subjects.
198
Table 2 provides the descriptive statistics on the daily average
199
concentrations of PM2.5 constituents, weather variables and gaseous
200
pollutants. There are no missing hourly data for PM2.5, but a small fraction
201
(about 5%) of missing data in the hourly measurements of some metal ions.
202
The 24-h mean concentrations of PM2.5 before the scheduled blood collection
203
varied substantially from 14.4 to 105.1 µg/m3, with an average of 38.4 µg/m3,
204
which is much higher than the World Health Organization Air Quality
ACS Paragon Plus Environment
Page 12 of 42
Page 13 of 42
Environmental Science & Technology
205
Guidelines (20 µg/m3).21 SO42- accounted for the largest proportion of PM2.5
206
(32% on average), followed by NO3- (25%), OC (18%), and NH4+ (16%).
207
In general, there were weak to high correlations among PM2.5
208
constituents (SI Table S1). For instance, there were weak correlations
209
between Cl- and Mg2+, Ca2+, and K+ (Pearson r: 0.05-0.21), but there were
210
strong correlations between OC and EC (Pearson r=0.97). We did not
211
observe large variations in weather conditions during the study period, but
212
they were moderately-to-strongly correlated with PM2.5 constituents. For
213
example, temperature was mildly or moderately positively correlated with OC,
214
EC, Cl-, NO3-, and NH4+ (Pearson r: 0.03-0.47) and strongly positively
215
correlated with other water-soluble ions (Pearson r: 0.72-0.83, for Na+, K+,
216
Mg2+, and Ca2+). Relative humidity was negatively correlated with most PM2.5
217
constituents.
218
Regression Results
219
Figure 1 illustrates the lag patterns of percent changes in 10 blood
220
biomarkers associated with an IQR increase in PM2.5 total mass. We
221
observed significantly positive associations between PM2.5 and most
222
biomarkers within 24 hours. These associations occurred within 0 to 6 hours
223
and became strongest between 13 and 24 hours, but attenuated greatly and
224
lost statistical significance at lag 1 day and longer lag days (data not shown).
225
This kind of lag pattern was not appreciably changed in most associations
226
between each biomarker and constituent, regardless of the statistical
ACS Paragon Plus Environment
Environmental Science & Technology
227
significance (SI Figure S11-S20). We therefore used the exposure averaged
228
at lags of 0 to 24 hours to capture almost all effects caused by PM2.5 in our
229
main analyses. An IQR (27.4 µg/m3) increase in total PM2.5 was significantly
230
associated with increments of 22%, 14%, 6.6%, 4.5%, 12%, 16%, 12%, 8.7%,
231
and 27% in serum levels of fibrinogen, CRP, MCP-1, TNF-α, ICAM-1,
232
P-selectin, VCAM-1, PAI-1, and sCD40L, respectively.
233
Figure 2 presents the percent changes in 10 cytokines per an IQR
234
increase in various constituents at 0 to 24 hours (lag 0 day) in the
235
single-constituent model. We observed significantly positive associations of
236
all PM2.5 constituents, except K+ and Mg2+, with at least one cytokine. OC and
237
EC were consistently associated with all 8 inflammatory biomarkers. On
238
average, an IQR increase in OC and EC corresponded to 50% and 37%
239
increments in these biomarkers, respectively. SO42- was associated with the 2
240
coagulation biomarkers, and an IQR increase resulted in 11% increments in
241
PAI-1 and 34% increments in sCD40L.
242
The magnitude of the associations between PM2.5 constituents (lag 0 day)
243
and cytokines attenuated appreciably, and the 95% CIs became larger in
244
constituent-PM2.5 joint models and constituent-residual models (Figures 3 and
245
4). We found relatively robust associations of OC, EC, NO3- and NH4+ on at
246
least 3 inflammation markers. Only SO42- or NH4+ was relatively robustly
247
associated with the 2 coagulation markers. Using CRP as an example, the
248
graphic abstract illustrated the associations with five constituents (OC, EC,
ACS Paragon Plus Environment
Page 14 of 42
Page 15 of 42
Environmental Science & Technology
249
NO3-, SO42-, and NH4+) in all 3 models (see Table of Contents).
250
In the sensitivity analyses, by controlling for gaseous pollutants (SI
251
Figure S1-S10), the associations between constituents and cytokines were
252
almost unchanged when adjusting for O3. After controlling for NO2, SO2, and
253
CO, the associations of NO3-, SO42-, and NH4+ decreased slightly and
254
became less significantly associated with almost all cytokines. The
255
associations of OC and EC with inflammatory cytokines were strengthened
256
and statistical significance was lost for some cytokines when adjusting for
257
SO2 and CO. The associations of the 5 constituents on coagulation were also
258
strengthened after controlling for NO2 and SO2, but became insignificant for
259
NO3-, SO42-, and NH4+ in some cases.
260 261
Discussion
262
This study provided a relatively comprehensive analysis of the short-term
263
associations of PM2.5 chemical constituents (2 carbonaceous fractions and 8
264
inorganic ions) on blood biomarkers of inflammation and coagulation. We
265
found significant associations between PM2.5 and all cytokines, and these
266
associations were restricted within 24 hours. We further identified some
267
constituents, including OC, EC, SO42, NO3-, and NH4+, have more robust
268
associations with blood inflammation or coagulation than the remaining 5
269
constituents. Our findings were generally insensitive to the adjustment for
270
gaseous pollutants.
ACS Paragon Plus Environment
Environmental Science & Technology
271
Abundant human evidence has demonstrated that short-term inhalation of
272
PM2.5 would result in elevations of circulating biomarkers of inflammation and
273
coagulation.1,
274
PM2.5 on the concurrent day and most biomarkers we selected. The effects of
275
PM2.5 occurred within 0 to 6 hours, became strongest within 13 to 24 hours,
276
and disappeared after 24 hours. Previous studies have also reported that the
277
acute effects of PM2.5 were restricted on the current day or sub-day after
278
exposure.23,
279
biomarkers. For example, we estimated that an IQR increase in 24-h average
280
PM2.5 concentrations was associated with increments of 22%, 14%, 6.6%,
281
4.5%, 12%, 16%, 12%, 8.7% and 27% in fibrinogen, CRP, MCP-1, TNF-a,
282
ICAM-, P-selectin, VCAM-1, PAI-1, and sCD40L, respectively. These results
283
were generally comparable to previous estimates. For example, in our
284
previous panel study with a crossover design, we found that an IQR increase
285
(64 µg/m3) in PM2.5 concentrations was associated with significant increases
286
of 16.1% and 71.3% in MCP-1 and sCD40L, respectively.15 A panel study
287
among healthy young students in Beijing observed a significant increase of
288
7.1% in TNF-α per an IQR (63.4 µg/m3) increase in PM2.5.14 Another panel
289
study in the US reported a 7.6% increase in CRP associated with an IQR
290
(19.6 µg/m3) increase in PM2.5 concentrations.25 The similar lag patterns in
291
the effects of constituents with PM2.5 total mass suggested that various
292
constituents of PM2.5 have similar time courses from entering the body to
9, 22
24
We observed significantly positive associations between
The magnitude of associations varied slightly among
ACS Paragon Plus Environment
Page 16 of 42
Page 17 of 42
Environmental Science & Technology
293
potential production of effects.
294
Nevertheless, previous studies on the acute effects of PM2.5 constituents
295
on systemic inflammation and coagulation were limited and inconsistent. OC
296
and EC are two major components in PM2.5 total mass. We found that they
297
were independently associated with inflammatory biomarkers, but not with
298
coagulation biomarkers. Similarly, a panel study on the 2008 Beijing Olympics
299
demonstrated significant increases in inflammatory cytokines (fibrinogen,
300
sCD40L, etc.) associated with EC and OC among healthy young adults.26
301
Another study among a panel of COPD patients in Germany reported
302
increased levels of fibrinogen by exposure to EC and OC.6 The independent
303
effects of OC and EC on the cardiovascular system were also broadly
304
supported by time-series or long-term studies.27-29
305
Soluble ions (such as SO42-, NO3-, and NH4+) typically constitute the
306
majority of PM2.5 mass. We found independent associations between NO3-
307
and/or NH4+ with inflammatory biomarkers and between SO42- or NH4+ and
308
coagulation biomarkers, which were also comparable to previous findings.30
309
For example, a panel study in Taiwan reported that both SO42- and NO3- were
310
positively associated with CRP, fibrinogen, and PAI-1 in single-pollutant
311
models, but only the association between SO42- and fibrinogen and PAI-1
312
remained significant when controlling for PM2.5.9 SO42- was also robustly
313
associated with sCD40L in a panel of healthy young adults surrounding the
314
Beijing Olympics.26 Another panel study among healthy young adults in
ACS Paragon Plus Environment
Environmental Science & Technology
315
Beijing only demonstrated significant associations of TNF-α with SO42- and
316
NO3- in single-constituent models. However, the associations of SO42- and
317
NO3- were null or inverse with other inflammatory biomarkers.14 In our
318
previous study, we demonstrated significant effects of PM2.5 and its
319
constituents of SO42-, NH4+, OC, and EC on an indicator of airway
320
inflammation.16 The independent cardiovascular effects of the 3 components
321
of PM2.5 (such as SO42-, NO3-, and NH4+) were also supported by other
322
time-series or long-term studies.29, 31, 32
323
Our findings may have implications for developing air pollution abatement
324
strategies to maximize public health benefits. As mentioned above, we
325
observed independent associations of carbonaceous components and
326
several soluble ions with circulating biomarkers, which may reflect the public
327
health importance from one or a set of sources.33 We found the independent
328
associations of OC, EC, SO42-, NO3-, and NH4+, rather than Cl-, Na+, K+, Mg2+,
329
and Ca2+, suggesting the relative importance of fossil combustion and
330
biomass burning that merit further investigations against sea salt and
331
wind-blown dust.34-36 However, potential differential measurement errors
332
across constituents may lead to challenges in interpreting these results.
333
There were few data available concerning the intra-city spatial distribution of
334
PM2.5 constituents in China. Therefore, it may still be plausible that the
335
observed stronger effects of combustion-related constituents (OC, EC, SO42-,
336
NO3-, and NH4+) might be attributable to lesser extent of exposure
ACS Paragon Plus Environment
Page 18 of 42
Page 19 of 42
Environmental Science & Technology
337
measurement errors in that they are enriched more in the finer size range of
338
PM2.5 size distribution and thus are more spatially uniformly distributed in the
339
city. In contrast, the non-significant associations of constituents (Cl-, Na+,
340
Mg2+, and Ca2+) with biomarkers might be explained by the larger
341
measurement errors due to their closer relations with sea salt and wind-blown
342
dust, which are less uniformly distributed within the city.
343
Because temperature is an important confounder when evaluating the
344
health effects of air pollutants,30, 37 the different correlations of constituents
345
with temperature may be helpful to partly explain the differentiated
346
associations between constituents and biomarkers. In this analysis, we
347
analyzed the associations between temperature and biomarkers using the
348
same models with PM2.5 constituents and found almost null or non-significant
349
associations. For OC, EC, and NO3-, which are weakly or moderately
350
correlated with temperature, their associations with biomarkers may be not
351
substantially confounded by temperature. For Na+, K+, Mg2+, Ca2+, which are
352
strongly correlated with temperature, their non-significant associations with
353
biomarkers might actually reflect the weak associations between temperature
354
and
355
correlations in modifying the effects of constituents on adverse health
356
outcomes merited further investigation because temperature and PM2.5
357
constituents were not measured at the individual level.
358
biomarkers.
Nonetheless,
the
roles
of
temperature-constituent
It remains unclear how PM2.5 constituents affect cardiovascular function.
ACS Paragon Plus Environment
Environmental Science & Technology
359
Our results supported the hypotheses that short-term exposure to PM2.5 and
360
some of its constituents was significantly associated with increments of CRP,
361
TNF-α, MCP-1, ICAM-1, VCAM-1, sP-selectin, sCD40L, and PAI-1. These
362
cytokines are well-established biomarkers of blood inflammation and
363
coagulation that is heavily involved in the development of a number of
364
adverse cardiovascular outcomes.38-42 Our findings indicated that some
365
constituents may be primarily responsible for the blood inflammation and
366
coagulation caused by PM2.5, which may aid in further investigations, for
367
example, on the genetic and epigenetic mechanisms whereby PM2.5
368
constituents affect biomarkers.
369
Our study has several strengths. First, we obtained real-time
370
concentrations of PM2.5 constituents, which allowed us to explore their
371
sub-daily effects and time courses. Second, the longitudinal panel design with
372
repeated-measures allowed the study subjects to serve as their own controls
373
and thus increased the statistical power. Third, we comprehensively
374
examined the effects of various PM2.5 constituents on a series of circulating
375
biomarkers, which avoided potential publication bias. Our results provided
376
abundant evidence linking air pollution with CVDs.
377
However, our results should be treated with caution because of the
378
following limitations. First, exposure measurement errors are inevitable
379
because all exposure data (including air pollutants and weather conditions)
380
were obtained from a nearby fixed-site monitor. Second, the sample size of
ACS Paragon Plus Environment
Page 20 of 42
Page 21 of 42
Environmental Science & Technology
381
the present study is relatively small, and some important associations might
382
have been underestimated or missed. Third, as all the participants were
383
elderly COPD patients, the generalizability of our results was limited, but the
384
impacts were not large because they were all stable patients with
385
mild-to-moderate COPD without any medications. Fourth, because of the
386
limitations of our instruments, we failed to evaluate the effects of metals,
387
which may also cause systemic inflammation and coagulation.14
388
In summary, this panel study added to the existing evidence that
389
short-term exposure to particulate air pollution could result in significant
390
increase in circulating biomarkers of inflammation and coagulation in China.
391
Furthermore, some chemical constituents in PM2.5, for instance, OC, EC,
392
SO42-, NO3-, and NH4+, might play crucial roles in inducing the systemic
393
inflammation and coagulation, but their roles varied according to the selected
394
biomarkers. Further investigations with a larger sample size, personal
395
exposure measurements, and more comprehensive measurements of PM2.5
396
constituents are needed to replicate our findings and characterize the
397
pathophysiological pathways whereby PM2.5 affect the cardiovascular system.
398
ACS Paragon Plus Environment
Environmental Science & Technology
399
Supporting Information
400
Table S1. Pearson correlation coefficients between 24-h average (lag 0 day)
401
concentrations of PM2.5 constituents, weather conditions and gaseous
402
pollutants.
403 404
Figure S1-Figure S10. Percent changes in 10 biomarkers associated with an
405
interquartile range increase in 24-h average (lag 0 day) concentrations of
406
PM2.5 constituents after adjusting for gaseous pollutants in 2-pollutant models.
407
Abbreviations as in Table 1.
408 409
Figure S11-Figure S20. Percent changes in the 10 blood biomarkers
410
associated with an interquartile range increase in sub-daily concentrations of
411
PM2.5 constituents. Abbreviations as in Table 1.
412 413
Author contributions
414
CL and JC performed the statistical analysis and drafted the manuscript. RC
415
and HK revised the manuscript. LQ and HW collected the environmental data.
416
WX, HL and AZ collected the health data. HK and RC designed the study and
417
takes responsibility for the integrity of the data and the accuracy of the data
418
analysis.
419 420
Notes
ACS Paragon Plus Environment
Page 22 of 42
Page 23 of 42
Environmental Science & Technology
421
The authors declared that they had no competing financial interests.
422 423
Acknowledgements
424
The authors appreciated the contributions of all volunteers in this study. The
425
study was supported by the Public Welfare Research Program of National
426
Health and Family Planning Commission of China (201502003), National
427
Natural Science Foundation of China (91643205 and 81502774), China
428
Medical Board Collaborating Program (13-152), and Cyrus Tang Foundation
429
(CTF-FD2014001).
430
ACS Paragon Plus Environment
Environmental Science & Technology
431
References
432
(1) Brook, R. D.; Rajagopalan, S.; Pope, C. A., 3rd; Brook, J. R.; Bhatnagar,
433
A.; Diez-Roux, A. V.; Holguin, F.; Hong, Y.; Luepker, R. V.; Mittleman, M. A.;
434
Peters, A.; Siscovick, D.; Smith, S. C., Jr.; Whitsel, L.; Kaufman, J. D.,
435
Particulate matter air pollution and cardiovascular disease: An update to the
436
scientific statement from the American Heart Association. Circulation. 2010,
437
121, (21), 2331-2378.
438
(2) Pope, C. A., 3rd; Dockery, D. W., Health effects of fine particulate air
439
pollution: lines that connect. J. Air Waste Manage. 2006, 56, (6), 709-742.
440
(3) Sun, Q.; Hong, X.; Wold, L. E., Cardiovascular effects of ambient
441
particulate air pollution exposure. Circulation. 2010, 121, (25), 2755-2765.
442
(4) Valavanidis, A.; Fiotakis, K.; Vlachogianni, T., Airborne particulate matter
443
and human health: toxicological assessment and importance of size and
444
composition of particles for oxidative damage and carcinogenic mechanisms.
445
J. Environ. Sci. Heal. C. 2008, 26, (4), 339-362.
446
(5) Anderson, G. B.; Krall, J. R.; Peng, R. D.; Bell, M. L., Is the relation
447
between ozone and mortality confounded by chemical components of
448
particulate matter? Analysis of 7 components in 57 US communities. Am J
449
Epidemiol. 2012, 176, (8), 726-732.
450
(6) Hildebrandt, K.; Ruckerl, R.; Koenig, W.; Schneider, A.; Pitz, M.; Heinrich,
451
J.; Marder, V.; Frampton, M.; Oberdorster, G.; Wichmann, H. E.; Peters, A.,
452
Short-term effects of air pollution: a panel study of blood markers in patients
ACS Paragon Plus Environment
Page 24 of 42
Page 25 of 42
Environmental Science & Technology
453
with chronic pulmonary disease. Part. Fibre Toxicol. 2009, 6, 25.
454
(7) Burgan, O.; Smargiassi, A.; Perron, S.; Kosatsky, T., Cardiovascular
455
effects of sub-daily levels of ambient fine particles: a systematic review.
456
Environ. Health. 2010, 9, 26.
457
(8) Lippi, G.; Favaloro, E. J.; Franchini, M.; Guidi, G. C., Air pollution and
458
coagulation testing: a new source of biological variability? Thromb. Res. 2008,
459
123, (1), 50-54.
460
(9) Chuang, K. J.; Chan, C. C.; Su, T. C.; Lee, C. T.; Tang, C. S., The effect of
461
urban air pollution on inflammation, oxidative stress, coagulation, and
462
autonomic dysfunction in young adults. Am. J. Respir. Crit. Care Med. 2007,
463
176, (4), 370-376.
464
(10)
465
M.; Kleinman, M. T.; Vaziri, N. D.; Longhurst, J.; Sioutas, C., Air pollution
466
exposures and circulating biomarkers of effect in a susceptible population:
467
clues to potential causal component mixtures and mechanisms. Environ.
468
Health Perspect. 2009, 117, (8), 1232-1238.
469
(11) Chen, R.; Zhao, Z.; Sun, Q.; Lin, Z.; Zhao, A.; Wang, C.; Xia, Y.; Xu, X.;
470
Kan, H., Size-fractionated particulate air pollution and circulating biomarkers
471
of inflammation, coagulation, and vasoconstriction in a panel of young adults.
472
Epidemiology (Cambridge, Mass.). 2015, 26, (3), 328-336.
473
(12)
474
Kan, H., Particulate air pollution and circulating biomarkers among type 2
Delfino, R. J.; Staimer, N.; Tjoa, T.; Gillen, D. L.; Polidori, A.; Arhami,
Wang, C.; Chen, R.; Zhao, Z.; Cai, J.; Lu, J.; Ha, S.; Xu, X.; Chen, X.;
ACS Paragon Plus Environment
Environmental Science & Technology
475
diabetic mellitus patients: the roles of particle size and time windows of
476
exposure. Environ. Res. 2015, 140, 112-118.
477
(13)
478
Chen, X.; Zhou, Y.; Xu, Y.; Kan, H., Associations Between Air Quality
479
Changes and Biomarkers of Systemic Inflammation During the 2014 Nanjing
480
Youth Olympics: A Quasi-Experimental Study. Am. J. Epidemiol. 2017, 1-7.
481
(14)
482
Qin, Y.; Zheng, C.; Hao, Y.; Guo, X., Chemical constituents of ambient
483
particulate air pollution and biomarkers of inflammation, coagulation and
484
homocysteine in healthy adults: a prospective panel study. Part. Fibre Toxicol.
485
2012, 9, 49.
486
(15)
487
H.; Xu, X.; Ha, S.; Li, T.; Kan, H., Cardiopulmonary benefits of reducing indoor
488
particles of outdoor origin: a randomized, double-blind crossover trial of air
489
purifiers. J. Am. Coll. Cardiol. 2015, 65, (21), 2279-2287.
490
(16)
491
Wang, H.; Zhao, Z.; Xu, X.; Hu, H.; Kan, H., Fine Particulate Matter
492
Constituents, Nitric Oxide Synthase DNA Methylation and Exhaled Nitric
493
Oxide. Environ. Sci. Technol. 2015, 49, (19), 11859-11865.
494
(17)
495
Huang, G., Insights into summertime haze pollution events over Shanghai
496
based on online water-soluble ionic composition of aerosols. Atmos. Environ.
Li, H.; Zhou, L.; Wang, C.; Chen, R.; Ma, X.; Xu, B.; Xiong, L.; Ding, Z.;
Wu, S.; Deng, F.; Wei, H.; Huang, J.; Wang, H.; Shima, M.; Wang, X.;
Chen, R.; Zhao, A.; Chen, H.; Zhao, Z.; Cai, J.; Wang, C.; Yang, C.; Li,
Chen, R.; Qiao, L.; Li, H.; Zhao, Y.; Zhang, Y.; Xu, W.; Wang, C.;
Du, H.; Kong, L.; Cheng, T.; Chen, J.; Du, J.; Li, L.; Xia, X.; Leng, C.;
ACS Paragon Plus Environment
Page 26 of 42
Page 27 of 42
Environmental Science & Technology
497
2011, 45, (29), 5131-5137.
498
(18)
499
Geng, F., Consecutive transport of anthropogenic air masses and dust storm
500
plume: Two case events at Shanghai, China. Atmos. Res. 2013, 127, 22-33.
501
(19)
502
and Aalto P. P., Semi-continuous gas and inorganic aerosol measurements at
503
a Finnish urban site: comparisons with filters, nitrogen in aerosol and gas
504
phases, and aerosol acidity. Atmos. Chem. Phys. 2012, (12), 5617-5631.
505
(20)
506
A.; Suh, H. H.; Gold, D. R.; Mittleman, M. A., Modeling the association
507
between particle constituents of air pollution and health outcomes. Am. J.
508
Epidemiol. 2012, 176, (4), 317-326.
509
(21)
510
nitrogen dioxide and sulfur dioxide: global update 2005: summary of risk
511
assessment. 2006.
512
(22)
513
Ohman-Strickland, P.; Hu, M.; Philipp, C.; Diehl, S. R.; Lu, S. E.; Tong, J.;
514
Gong, J.; Thomas, D.; Zhu, T.; Zhang, J. J., Association between changes in
515
air pollution levels during the Beijing Olympics and biomarkers of
516
inflammation and thrombosis in healthy young adults. JAMA. 2012, 307, (19),
517
2068-2078.
518
(23)
Wang, L.; Du, H.; Chen, J.; Zhang, M.; Huang, X.; Tan, H.; Kong, L.;
Makkonen U., V. A., Mäntykenttä J., Hakola H., Keronen P., Vakkari V.,
Mostofsky, E.; Schwartz, J.; Coull, B. A.; Koutrakis, P.; Wellenius, G.
WHO, WHO Air quality guidelines for particulate matter, ozone,
Rich, D. Q.; Kipen, H. M.; Huang, W.; Wang, G.; Wang, Y.; Zhu, P.;
XU, M. M.; JIA, Y. P.; LI, G. X.; LIU, L. Q.; MO, Y. Z.; JIN, X. B.; PAN,
ACS Paragon Plus Environment
Environmental Science & Technology
519
X. C., Relationship between ambient fine particles and ventricular
520
repolarization changes and heart rate variability of elderly people with heart
521
disease in Beijing, China. Biomed. Environ. Sci. 2013, 26, (8), 629-637.
522
(24)
523
Hwang, J. S.; Hsu, S. H.; Chao, H.; Chuang, K. J.; Chou, C. C.; Wang, Y. N.;
524
Ho, C. C.; Su, T. C., Association of short-term exposure to fine particulate
525
matter and nitrogen dioxide with acute cardiovascular effects. Sci. Total
526
Environ. 2016, 569-570, 300-305.
527
(25)
528
N. L.; Wilson, W. E.; Eatough, D. J., Ambient particulate air pollution, heart
529
rate variability, and blood markers of inflammation in a panel of elderly
530
subjects. Environ. Health Perspect. 2004, 112, (3), 339-345.
531
(26)
532
Wang, Y.; Lu, S. E.; Ohman-Strickland, P.; Diehl, S.; Hu, M.; Tong, J.; Gong, J.;
533
Thomas, D., Cardiorespiratory biomarker responses in healthy young adults
534
to drastic air quality changes surrounding the 2008 Beijing Olympics. Re.s
535
Rep. Health Eff. Inst. 2013, (174), 5-174.
536
(27)
537
particle components and health--a systematic review and meta-analysis of
538
epidemiological time series studies of daily mortality and hospital admissions.
539
J. Expo. Sci. Environ. Epidemiol. 2015, 25, (2), 208-214.
540
(28)
Wu, C. F.; Shen, F. H.; Li, Y. R.; Tsao, T. M.; Tsai, M. J.; Chen, C. C.;
Pope, C. A., 3rd; Hansen, M. L.; Long, R. W.; Nielsen, K. R.; Eatough,
Zhang, J.; Zhu, T.; Kipen, H.; Wang, G.; Huang, W.; Rich, D.; Zhu, P.;
Atkinson, R. W.; Mills, I. C.; Walton, H. A.; Anderson, H. R., Fine
Ostro, B.; Lipsett, M.; Reynolds, P.; Goldberg, D.; Hertz, A.; Garcia, C.;
ACS Paragon Plus Environment
Page 28 of 42
Page 29 of 42
Environmental Science & Technology
541
Henderson, K. D.; Bernstein, L., Long-term exposure to constituents of fine
542
particulate air pollution and mortality: results from the California Teachers
543
Study. Environ. Health Perspect. 2010, 118, (3), 363-369.
544
(29)
545
and multisite time-series analysis of the differential toxicity of major fine
546
particulate matter constituents. Am. J Epidemiol. 2012, 175, (11), 1091-1099.
547
(30)
548
D.; Sheppard, L.; Simpson, C. D.; Szpiro, A. A., National Particle Component
549
Toxicity (NPACT) initiative report on cardiovascular effects. Res. Rep. Health
550
Eff. Inst. 2013, (178), 5-8.
551
(31)
552
constituents and cardiopulmonary mortality in a heavily polluted Chinese city.
553
Environ. Health. Perspect. 2012, 120, (3), 373-378.
554
(32)
555
Associations between long-term exposure to chemical constituents of fine
556
particulate matter (PM2.5) and mortality in Medicare enrollees in the eastern
557
United States. Environ. Health. Perspect. 2015, 123, (5), 467-474.
558
(33)
559
research mean? Am. J. Resp. Crit. Care. 2011, 183, (1), 4-6.
560
(34)
561
Wang,
562
Beijing—concentration, composition, distribution and sources. Atmos.
Levy, J. I.; Diez, D.; Dou, Y.; Barr, C. D.; Dominici, F., A meta-analysis
Vedal, S.; Campen, M. J.; McDonald, J. D.; Larson, T. V.; Sampson, P.
Cao, J.; Xu, H.; Xu, Q.; Chen, B.; Kan, H., Fine particulate matter
Chung, Y.; Dominici, F.; Wang, Y.; Coull, B. A.; Bell, M. L.,
Vedal, S.; Kaufman, J. D., What does multi-pollutant air pollution
Sun, Y.; Zhuang, G.; Wang, Y.; Han, L.; Guo, J.; Dan, M.; Zhang, W.; Z.;
Hao,
Z.,
The
air-borne
particulate
ACS Paragon Plus Environment
pollution
in
Environmental Science & Technology
563
Environ. 2004, 38, (35), 5991-6004.
564
(35)
565
Zotter, P.; Shen, R. R.; Schafer, K.; Shao, L.; Prevot, A. S.; Szidat, S., Source
566
Apportionment of Elemental Carbon in Beijing, China: Insights from
567
Radiocarbon and Organic Marker Measurements. Environ. Sci. Technol.
568
2015, 49, (14), 8408-8415.
569
(36)
570
M., Investigation of sources of atmospheric aerosol at a hot spot area in
571
Dhaka, Bangladesh. J. Air Waste Manage. 2005, 55, (2), 227-240.
572
(37)
573
Associations between outdoor temperature and markers of inflammation: a
574
cohort study. Environ. Health. 2010, 9, 42.
575
(38)
576
molecular mechanisms and clinical implications. Clin. Sci (Lond). 2009, 117,
577
(3), 95-109.
578
(39)
579
P-selectin, interleukin-6, and tissue factor in diabetes mellitus: relationships to
580
cardiovascular disease and risk factor intervention. Circulation. 2004, 109,
581
(21), 2524-2528.
582
(40)
583
P-selectin and cardiovascular disease. Eur. Heart J. 2003, 24, (24),
584
2166-2179.
Zhang, Y. L.; Schnelle-Kreis, J.; Abbaszade, G.; Zimmermann, R.;
Begum, B. A.; Biswas, S. K.; Kim, E.; Hopke, P. K.; Khaliquzzaman,
Halonen, J. I.; Zanobetti, A.; Sparrow, D.; Vokonas, P. S.; Schwartz, J.,
Niu, J.; Kolattukudy, P. E., Role of MCP-1 in cardiovascular disease:
Lim, H. S.; Blann, A. D.; Lip, G. Y., Soluble CD40 ligand, soluble
Blann, A. D.; Nadar, S. K.; Lip, G. Y., The adhesion molecule
ACS Paragon Plus Environment
Page 30 of 42
Page 31 of 42
Environmental Science & Technology
585
(41)
586
system and vascular disease. Curr. Vasc. Pharmacol. 2006, 4, (2), 101-116.
587
(42)
588
necrosis factor-alpha, biologic agents and cardiovascular risk. Lupus. 2005,
589
14, (9), 780-784.
Nicholl, S. M.; Roztocil, E.; Davies, M. G., Plasminogen activator
Sarzi-Puttini, P.; Atzeni, F.; Doria, A.; Iaccarino, L.; Turiel, M., Tumor
590
ACS Paragon Plus Environment
Environmental Science & Technology
591
Page 32 of 42
Table 1. Summary of the health indicators over the study period. Biomarkers
Mean
SD
Min
Median
Max
Fibrinogen, ng/ml
0.96
0.53
0.3
0.86
3.4
CRP, mg/L
4.5
6.4
0.3
1.9
16.9
MCP-1, pg/ml
596
180
263
566
1302
TNF-α, pg/ml
15.0
18.6
3.6
10.9
114.5
IL-1β, pg/ml
4.4
3.3
0.0
5.1
9.2
ICAM-1, ng/ml
105
36
38
100
206
P-selectin, ng/ml
42.7
17.3
23.3
38.5
106.5
VCAM-1, ng/ml
383
102
213
357
705
PAI-1, ng/ml
126
44
57
119
283
sCD40L, µg/ml
3.6
2.5
0.3
3.0
11.2
Blood inflammation
Blood coagulation
592
Definition of abbreviations: SD = standard deviation; IQR = interquartile range;
593
CRP = C-reactive protein; MCP-1= monocyte chemoattractant protein-1;
594
TNF-α= tumor necrosis factor-α; IL-1β= interleukin-1β; ICAM-1= intercellular
595
adhesion molecule-1; VCAM-1= vascular cell adhesion molecule-1; PAI-1=
596
plasminogen activator inhibitor-1; and sCD40L= soluble CD40 ligand.
597
ACS Paragon Plus Environment
Page 33 of 42
Environmental Science & Technology
598
Table 2. Descriptive statistics on the 24-h average ambient air pollutants,
599
PM2.5 chemical constituents, and weather variables for the study participants
600
over the study period. Variables
Mean
SD
Min
Median
Max
IQR
Total mass
44.4
25.9
14.4
41.6
105.1
27.4
Cl-
0.69
0.38
0.11
0.66
1.34
0.63
NO3-
9.41
6.38
2.43
8.40
24.97
7.44
SO42-
13.65
7.98
2.81
11.67
34.32
8.38
NH4+
6.40
4.01
0.90
5.38
15.08
5.40
Na+
0.09
0.13
0.00
0.00
0.43
0.16
K+
0.22
0.32
0.00
0.02
0.95
0.41
Mg2+
0.23
0.14
0.08
0.21
0.50
0.23
Ca2+
1.88
1.32
0.55
1.33
4.66
2.28
OC
7.59
3.53
3.78
6.96
13.29
7.94
EC
2.01
0.82
0.78
1.97
3.28
1.61
Temperature (℃)
24.7
1.7
22.7
24.2
27.8
3.2
Relative humidity (%)
68.5
12.5
45.0
72.3
84.0
25.3
49.5
13.3
30.8
46.9
70.0
21.7
PM2.5 (µg/m3)
Weather a
Gaseous pollutants (µg/m3) NO2
ACS Paragon Plus Environment
Environmental Science & Technology
Page 34 of 42
SO2
9.6
5.9
3.2
7.7
20.8
13.3
O3
84.9
31.6
29.9
75.9
143.5
49.5
CO
0.81
0.19
0.49
0.83
1.20
0.30
601
Definition of abbreviations: SD= standard deviation; IQR= interquartile range.
602
a
603
and previous 3 days.
Data are presented as the average of weather conditions on the present day
ACS Paragon Plus Environment
Page 35 of 42
Environmental Science & Technology
604 605
Figure 1. Percent changes (mean and 95% confidence intervals) in blood biomarkers associated with an interquartile range
606
increase in PM2.5 mass concentration using different lag periods. Abbreviations as in Table 1.
ACS Paragon Plus Environment
Environmental Science & Technology
607 608
Figure 2. Percent changes (mean and 95% confidence intervals) in blood
609
biomarkers associated with an interquartile range increase in 24-h average
610
(lag 0 days) concentrations of PM2.5 constituents in the single-constituent
611
model. Label abbreviations: (A) Fibrinogen; (B) CRP, C-reactive protein; (C)
ACS Paragon Plus Environment
Page 36 of 42
Page 37 of 42
Environmental Science & Technology
612
MCP-1, monocyte chemoattractant protein-1; (D) TNF-α, tumor necrosis
613
factor-α; (E) IL-1β, interleukin-1β; (F) ICAM-1, intercellular adhesion
614
molecule-1; (G) P-selectin; (H) VCAM-1, vascular cell adhesion molecule-1; (I)
615
PAI-1, plasminogen activator inhibitor-1; (J) sCD40L, soluble CD40 ligand.
616
ACS Paragon Plus Environment
Environmental Science & Technology
617 618
Figure 3. Percent changes (mean and 95% confidence intervals) in blood
619
biomarkers associated with an interquartile range increase in 24-h average
620
(lag 0 days) concentrations of PM2.5 constituents in the constituent-PM2.5 join
621
model. Label abbreviations: (A) Fibrinogen; (B) CRP, C-reactive protein; (C)
ACS Paragon Plus Environment
Page 38 of 42
Page 39 of 42
Environmental Science & Technology
622
MCP-1, monocyte chemoattractant protein-1; (D) TNF-α, tumor necrosis
623
factor-α; (E) IL-1β, interleukin-1β; (F) ICAM-1, intercellular adhesion
624
molecule-1; (G) P-selectin; (H) VCAM-1, vascular cell adhesion molecule-1; (I)
625
PAI-1, plasminogen activator inhibitor-1; (J) sCD40L, soluble CD40 ligand.
626
ACS Paragon Plus Environment
Environmental Science & Technology
627 628
Figure 4. Percent changes (mean and 95% confidence intervals) in blood
629
biomarkers associated with an interquartile range increase in 24-h average
630
(lag 0 days) concentrations of PM2.5 constituents in the constituent-residual
631
model. Labels abbreviations: (A) Fibrinogen; (B) CRP, C-reactive protein; (C)
ACS Paragon Plus Environment
Page 40 of 42
Page 41 of 42
Environmental Science & Technology
632
MCP-1, monocyte chemoattractant protein-1; (D) TNF-α, tumor necrosis
633
factor-α; (E) IL-1β, interleukin-1β; (F) ICAM-1, intercellular adhesion
634
molecule-1; (G) P-selectin; (H) VCAM-1, vascular cell adhesion molecule-1; (I)
635
PAI-1, plasminogen activator inhibitor-1; (J) sCD40L, soluble CD40 ligand.
636
ACS Paragon Plus Environment
Environmental Science & Technology
637
638 639
Table of Contents or Abstract Figure
640
ACS Paragon Plus Environment
Page 42 of 42
