Environ HealthEnviron HealthEnvironmental Health1476-069XBioMed Central2375884337008271476-069X-12-4710.1186/1476-069X-12-47ResearchA panel study of occupational exposure to fine particulate matter and changes in DNA methylation over a single workday and years worked in boilermaker weldersKileMolly L1molly.kile@oregonstate.eduFangShona2sfang@hsph.harvard.eduBaccarelliAndrea A2abaccare@hsph.harvard.eduTarantiniLetizia3letizia.tarantini@unimi.itCavallariJennifer4cavallari@uchc.eduChristianiDavid C2dchris@hsph.harvard.eduOregon State University, College of Public Health and Human Sciences, 15 Milam, Corvallis, OR 97331, USAHarvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115, USACenter of Molecular and Genetic Epidemiology, Department of Environmental and Occupational Health, University of Milan, Milan, ItalyUniversity of Connecticut, School of Medicine, Community Medicine & Health Care, 263 Farmington Avenue, Farmington, CT 06030, USA201311620131247477102012462013Copyright © 2013 Kile et al.; licensee BioMed Central Ltd.2013Kile et al.; licensee BioMed Central Ltd.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Background

Exposure to pollutants including metals and particulate air pollution can alter DNA methylation. Yet little is known about intra-individual changes in DNA methylation over time in relationship to environmental exposures. Therefore, we evaluated the effects of acute- and chronic metal-rich PM2.5 exposures on DNA methylation.

Methods

Thirty-eight male boilermaker welders participated in a panel study for a total of 54 person days. Whole blood was collected prior to any welding activities (pre-shift) and immediately after the exposure period (post-shift). The percentage of methylated cytosines (%mC) in LINE-1, Alu, and inducible nitric oxide synthase gene (iNOS) were quantified using pyrosequencing. Personal PM2.5 (particulate matter with an aerodynamic diameter ≤ 2.5 μm) was measured over the work-shift. A questionnaire assessed job history and years worked as a boilermaker. Linear mixed models with repeated measures evaluated associations between DNA methylation, PM2.5 concentration (acute exposure), and years worked as a boilermaker (chronic exposure).

Results

PM2.5 exposure was associated with increased methylation in the promoter region of the iNOS gene (β = 0.25, SE: 0.11, p-value = 0.04). Additionally, the number of years worked as a boilermaker was associated with increased iNOS methylation (β = 0.03, SE: 0.01, p-value = 0.03). No associations were observed for Alu or LINE-1.

Conclusions

Acute and chronic exposure to PM2.5 generated from welding activities was associated with a modest change in DNA methylation of the iNOS gene. Future studies are needed to confirm this association and determine if the observed small increase in iNOS methylation are associated with changes in NO production or any adverse health effect.

DNA methylationPM2.5iNOSWeldersLINE-1AluBoilermakers
Background

Boilermakers are skilled welders who cut and weld metal plates. They use a variety of welding technologies including oxyacetylene gas torches, gas tungsten arc welding, shielded metal arc welding, or gas metal arc welding. The extreme heat produced by welding generates a complex mixture of gases, aerosols and particulate matter. This molten mixture condenses into ultrafine and fine particulate matter with an aerodynamic diameter ≤ 2.5 μm (PM2.5) that is easily inhaled [1]. Welding fumes are rich in metal oxides and can contain antimony, beryllium, cadmium, chromium, cobalt, copper, iron, lead, manganese, mercury, molybdenum, nickel, vanadium, and zinc [2,3] although the exact composition of welding fumes will depend on many factors including base metal characteristics and type of welding.

Epidemiological studies show that chronic exposure to welding fumes are associated with respiratory health effects including cardiovascular disease, asthma, bronchitis, lung function changes, and increased risk of lung cancer [4-10]. While the mechanisms linking welding fume exposure to adverse health outcomes is not fully understood, previous studies implicate several biological pathways for welding fume toxicity including oxidative stress [11], systemic inflammation [12-14], and alterations in cardiac autonomic responses [4,15,16]. It is also possible that welding fumes alter epigenetic mechanisms and subsequently gene expression. Data from previous studies shows that many of the metals detected in welding fumes including nickel, lead, cadmium, chromium, beryllium, and arsenic are associated with altered DNA methylation [17-27]. There is also evidence that exposure to particulate matter air pollution is associated with altered DNA methylation [28-31]. Changes in DNA methylation have been observed in subjects with cardiovascular disease and cancer [32]. Thus, it is possible that epigenetic changes may provide a potential mechanism by which welding fumes are linked to adverse biological effects.

We, therefore, hypothesized that exposure to particulate matter generated from welding activities can alter DNA methylation. This hypothesis was tested in boilermakers who were previously recruited for a panel study designed to evaluate the cardiopulmonary health effects of welding fumes [4,15]. We used existing data to explore the relationship between exposure to PM2.5 and DNA methylation over time because individuals provided blood samples collected both pre- and post-work shift, had personal PM2.5 occupational exposure measurements, and detailed work histories. Specifically, we hypothesized that exposure to PM2.5 over a single work shift would be associated with changes to DNA methylation, as well as, from cumulative exposure based on years worked as a welder.

We employed a candidate gene approach and measured DNA methylation in short interspersed nucleotide elements (Alu) and long interspersed nucleotide elements (LINE-1). These two repetitive elements make up approximately 55% of the human genome and are heavily methylated to suppress their expression [33]. Repetitive elements are activated during conditions of cellular stress and de-methylation of LINE-1 and Alu elements increase their activity as retrotransposable sequences [34-36]. Additionally, we quantified DNA methylation in the inducible nitric oxide synthase gene (iNOS, also known as NOS2, Genbank accession number AF017634) which is involved in the production of nitric oxide and plays an important role in a variety of cardio-pulmonary processes including asthma [12], chronic obstructive pulmonary disease [37], and cardiovascular health [38].

MethodsParticipant selection

The Institutional Review Board at the Harvard School of Public Health approved the study protocol and informed consent was obtained prior to participation. The details of participant selection and study recruitment have been described previously [4,15]. This analysis used samples collected from 38 individuals who had archived whole blood samples available for DNA extraction. Briefly, boilermaker construction workers were monitored on a high-exposure welding day and low-exposure non-welding days at the Union welding school in 2003 and 2008 as part of a panel study on welding fumes and cardiovascular risk. On non-welding days, participants performed office work in a large enclosed break room adjacent to the welding room. Per the initial study design, boilermakers were allowed to participate on multiple occasions in each observation period. We observed that two boilermakers participated in both 2003 and 2008 and three boilermakers participated twice on different days in 2008. Not all participants opted to provide a post-shift blood sample. Therefore, this analysis utilized 40 blood samples collected in the morning prior to the start of a work-shift (pre-shift) and 38 blood samples were collected in the afternoon after all work activities had been completed (post-shift). Also, twenty-four individuals provided two blood samples on the same observation day (pre- and post-shift). The most common type of welding was manual metal arc welding on mild (manganese alloys) and stainless steel (chromium and nickel alloys) bases. All welding was performed in a room outfitted with 10 workstations, each with local exhaust ventilation.

Data collection

A modified American Thoracic Society questionnaire was used to collect information on socio-demographic information, smoking history, occupational history and medical history [39].

Exposure assessment

All individuals wore personal gravimetric particulate samplers over the duration of the work shift using a KTL cyclone (GK2.05SH, BGI Incorporated, Waltham, MA) with an aerodynamic diameter ≤2.5 μm used in line with a pump drawing 3.5 L/min of air. The inlet tubing was secured to the participant’s shoulder in the breathing zone area. Exposure to PM2.5 was measured over a six-hour welding shift or equivalent non-welding period as previously described by Cavallari et al. [4]. All samples were blank corrected. Two individuals were missing PM2.5 data. Air quality monitoring was performed on the same day as blood sample collection.

DNA methylation

Whole blood was collected in the morning before any welding activity (pre-shift) and in the afternoon at the end of the work shift (post-shift) and frozen at -80°C. Blood samples were collected at the same times each day to account for the potential influence of circadian rhythm. DNA was extracted using Puregene DNA solutions (Qiagen, Valencia, CA, USA). DNA methylation was quantitatively measured on bisulfite-treated DNA using PCR-Pyrosequencing following the methods described by Tarantini et al. [28]. The primers and sequence that was analyzed for each marker are presented in Table 1. Samples were analyzed in duplicate and the average methylation, expressed for each DNA locus as the percentage of methylated cytosines (%mC) over the sum of methylated and unmethylated cytosine, was used in the statistical analysis. The coefficient of variation for each DNA locus was 0.009, 0.006, and 0.008 for LINE-1, iNOS, and Alu, respectively. All samples were subjected to a quality control check incorporated in the software which evaluates the bisulfite conversion rate of any cytosine not followed by a guanine. Five pre-shift and four post-shift blood samples failed pyrosequencing quality control for iNOS and were subsequently excluded from the analysis.

Primers and location of CpG sites that were quantified by pyrosequencing

Sequence ID
Forward Primer
Reverse Primer
Sequencing Primer
Sequence analyzed
 (5′ to 3′)(5′ to 3′)(5′ to 3′) 
Global methylation analysis
 Alu
Biotin-TTTTTATTAAAAATATAAAAATT
CCCAAACTAAAATACAATAA
AATAACTAAAATTACAAAC
G/AC/TG/AC/TG/ ACCACCA
 LINE-1
TTTTGAGTTAGGTGTGGGATATA
Biotin-AAAATCAAAAAATTCCCTTTC
AGTTAGGTGTGGGATATAGT
TTC/TGTGGTGC/TGTC/TG
Gene-specific methylation analysis
 iNOSTTGGATGGTATGGGGTGAGTATBiotin-TACCCAATCCCCTCATCAAGTGTGTTTATAATTTTGTAGC/TGAGTC/TGAAAATTGAGGTTTC/TGG

Nucleotides where DNA methylation was measured are in bold text.

Statistical analysis

The average %mC was calculated for 3 CpG sites in LINE-1, 3 CpG sites in Alu, and 4 CpG sites in iNOS. Linear mixed effects regression models with a repeated statement to account for repeated measurements on the same subjects were used to evaluate the relationship between %mC and PM2.5; as well as, the relationship between %mC and the number of years each individual worked as a boilermaker. Separate models were constructed for LINE-1, Alu, and iNOS. In the models that looked at cross-shift PM2.5 exposure, pre-shift blood %mC for each gene was included as a predictor in the models to control for individual factors that may influence methylation. Whereas, pre-shift blood %mC was not included in the models that evaluated the relationship between the number of years a person reported working as a boilermaker and methylation. All models were adjusted for smoking (yes/no), wearing a respirator (yes/no), and age. Since age was highly collinear with the number of years a person worked as a boilermaker (σspearman=0.57), we used the residuals from a regression model that had age as the dependent variable and years worked as a boilermaker to adjust for age (ageRS) in models that looked at the association between methylation and chronic exposure. All analyses were performed with SAS version 9.2.

Results

This panel study consisted of 38 adult males who provided a total of 54 measurement events (Table 2). The average age of the participants at first enrollment was 35.6 years (range: 21.3 to 61.0 years) and they had worked an average of 6.8 years as a boilermaker (range: 1–35 years). The majority of participants reported their race as white (85.3%) and 31.6% reported that they currently smoked. In 13 instances during sampling, a respirator was used by the worker. The mean work shift exposure for PM2.5 over the 54 measurement events was 1.06 mg/m3 (median: 0.52 mg/m3, range: 0.02-3.41 mg/m3, average duration of exposure, 303.2 minutes; SD = 84.1 minutes; range: 61–339 minutes). Eighty-two percent of the blood samples were collected from individuals who welded during the day of collection and 18% were collected from individuals who were present at the union hall but did not participate in any welding activities. The average percentage of methylated cytosines at baseline for Alu, LINE-1 and iNOS was 25.5%, 85.3%, and 97.5%, respectively (Table 3).

Description of selected characteristics in 38 boilermakers at the time of their first recruitment

 NMean (SD) or PercentRange
Cohort characteristics
Age (years)a
34
36.0 (12.0)
21.3-61.0
Males
38
100%
 
Race
 
 
 
 Caucasian
31
83.3%
 
 African-American
5
11.1%
 
 Hispanic
2
5.6%
 
Current smoker
 Yes
12
30.6%
 
 No
26
69.4%
 
Years worked as boilermakerb
36
6.8 (9.4)
1 - 35
Exposure characteristicsc
 
 
Median (IQR) or Percent
PM2.5 exposure (mg/m3)c
54
0.52 (1.34)
0.02-3.41
Weld day samplesd
 
 
 
 Yes
44
81.5%
 
 No
10
18.5%
 
Used respirator
 
 
 
 Yes
13
24.1%
 
 No4175.9% 

aAge at initial participation in the study. Missing age = 2.

bMissing self-reported number of years working as a boilermaker = 2.

cDescriptive statistics based on 54 observation days.

dMissing personal PM2.5 measurements = 2.

Description of DNA methylation for the repeated blood sample measurements

 
Baseline
Post-shift
 NMeanSDMinMaxNMeanSDMinMax
Alu (%mC)
40
25.5
0.3
25.0
26.0
38
25.5
0.4
23.8
25.9
LINE-1 (%mC)
40
85.3
0.9
82.0
87.3
38
85.4
0.6
84.3
86.9
iNOS (%mC)3697.50.895.299.23497.30.696.298.7

Linear mixed effect models evaluated the association between PM2.5 exposure across the work shift and methylation in blood samples collected post-shift (Table 4). We observed that PM2.5 exposure was associated with increased methylation within the promoter region of the iNOS gene (β = 0.25, SE: 0.11, p-value = 0.04) in models that adjusted for iNOS methylation measured pre-shift, current smoking status, age, and respirator usage. When the analysis was restricted to only those 25 individuals who actually performed welding activities on the observation day, the association between PM2.5 and iNOS methylation was marginally greater (β = 0.35, SE: 0.11, p-value = 0.006). This suggested that individuals who were directly exposed to welding fume PM2.5 had a stronger response compared to individuals who were exposed to background PM2.5 within the practice hall. When the cohort was further restricted to exclude the five individuals who wore respirators on the observation day, the association between PM2.5 and iNOS methylation was fractionally greater (β = 0.38, SE: 0.14, p-value = 0.02) which further supported the association between metal-rich welding fume PM2.5 and modest increased methylation in the promoter region of the iNOS gene. Overall, these results suggested that for every 1 mg/m3 increase in PM2.5, iNOS methylation increased an average of 0.25% (95% CI: 0.01 – 0.48%). No association was observed between PM2.5 and Alu (β = 0.05, SE: 0.07, p-value = 0.47) or LINE-1 (β = −0.12, SE: 0.10, p-value = 0.28).

Association between PM2.5 exposure across the work shift (mg/m3) and methylation of Alu, LINE-1 and iNOS measured in whole blood collected post-shift

  Unadjusted regression1 Adjusted regression2
 
N
β
SE
p-value
N
β
SE
p-value
Global DNA Methylation
 
 
 
 
 
 
 
 
 Alu (%5mC)
37
0.02
0.07
0.73
35
0.05
0.07
0.47
 LINE-1 (%5mC)
37
-0.05
0.10
0.61
35
-0.12
0.10
0.28
Gene specific DNA Methylation
 
 
 
 
 
 
 
 
iNOS (%5mC)290.220.100.04270.250.110.04

1 Adjusted for DNA methylation in the sample pre-shift.

2 Adjusted for DNA methylation in the sample pre-shift, currently smoking (yes), age, and wearing a respirator.

To evaluate the association between DNA methylation and chronic exposure to PM2.5, linear mixed effect models evaluated the relationship between the number of years each participant worked as a boilermaker and DNA methylation (Table 5). These models suggested that chronic exposure to welding fumes, as characterized by years worked as a boilermaker, were associated with small increases in methylation of the promoter region within the iNOS gene (β = 0.03, SE: 0.01, p-value = 0.03) after adjusting for current smoking status and ageRS. No association between years worked as a boilermaker and methylation in Alu (β = -0.004, SE: 0.004, p-value = 0.28) or LINE-1 (β = 0.02, SE: 0.02, p-value = 0.28) was observed. We further restricted this analysis to only pre-shift samples to avoid the potential for acute work shift exposures to influence the observed associations and observed similar effects (Table 6).

Association between years as a boilermaker and methylation of Alu, LINE-1 and iNOS measured in all whole blood samples

  Unadjusted regression Adjusted regression1
Global DNA Methylation
N
β
SE
p-value
N
β
SE
p-value
 Alu (%5mC)
77
-0.002
0.004
0.66
73
-0.01
0.005
0.05
 LINE-1 (%5mC)
77
0.005
0.01
0.64
73
0.01
0.01
0.28
Gene specific DNA Methylation
 
 
 
 
 
 
iNOS (%5mC)700.020.010.006660.030.010.02

1 Adjusted for currently smoking (yes), white blood cell count, and age.

Association between years as a boilermaker and methylation of Alu, LINE-1 and iNOS measured in only in pre-shift whole blood samples

  Unadjusted regression Adjusted regression1
Global DNA Methylation
N
β
SE
p-value
N
β
SE
p-value
 Alu (%5mC)
42
-0.001
0.004
0.13
40
-0.02
0.005
0.002
 LINE-1 (%5mC)
42
0.02
0.02
0.24
40
0.03
0.02
0.15
Gene specific DNA Methylation
 
 
 
 
 
 
iNOS (%5mC)380.030.010.007360.040.020.02

1 Adjusted for currently smoking (yes) and age.

Considering that smoking was common in this population and might be an effect modifier of iNOS methylation, an interaction term between years worked as a boilermaker and current smoking was also evaluated in the model. This interaction term was not significant which indicated that there was no effect modification by smoking status (βinteraction = 0.10, SE: 0.09, p-value = 0.31).

Discussion

We observed that occupational exposure to fine particulate matter generated from welding activities measured across a single work-shift was associated with modest increases in DNA methylation in the promoter region of the inducible nitric oxide synthase gene in whole blood DNA. The number of years worked as a boilermaker was also associated with modest increased DNA methylation in the iNOS gene. However, we did not observe any relationship between acute exposure to welding fume PM2.5 or years worked and DNA methylation in two repeated elements, Alu and LINE-1. These findings suggest that methylation of the iNOS gene may be mutable over short periods of time.

DNA methylation regulates gene expression with increased methylation leading to gene silencing. Biologically, it is plausible that iNOS could be influenced by exposure to PM2.5 generated from welding activities. Animal models show that acute exposure to welding fumes produce reactive oxygen species and inflammatory cytokines including tumor necrosis factor-alpha and interleukin-1 beta [5,40]. In turn, these immunologic and inflammatory factors stimulate inducible nitric oxide synthase which produces nitric oxide (NO) which can constrict vascular beds, produce hypertensive responses, regulate nonspecific host defenses, and modulate inflammation [41,42]. These effects appear to depend on the PM composition. For instance, PM derived from urban sources diminished NO production, whereas, respirable cristobalite, fine particulate matter with enriched polycyclic aromatic hydrocarbons, aerosols derived from wildfire smoke, and traffic-generated PM increased NO production [43-45]. There is also evidence that welding fume PM2.5 influences NO production. Prior epidemiological research by our group showed that the fractional concentration of expired breath nitric oxide decreased after five days of exposure to welding-related PM2.5 and that the decreases in expired NO were associated with the soluble metal fraction in the PM2.5[46]. This inhibition in exhaled NO would be consistent with diminished expression of genes involved in the synthesis of NO.

Other environmental epigenetic studies have also investigated the association between particulate matter air pollution and DNA methylation in the iNOS gene. For instance, Tarantini et al. reported that particulate matter with aerodynamic diameters < 10 μm (PM10) was associated with a 0.61% decrease in methylation in the iNOS gene measured in whole blood DNA across a 3-work day period in a cohort of 63 foundry workers [28]. Another study reported that 7-day average ambient PM2.5 was associated with decreased iNOS methylation (0.30% decrease in iNOS per 5 μg/m3 increase in ambient PM2.5) in DNA extracted from buccal cells collected from 940 children who resided in the Los Angeles area [47]. Both of these studies observed an inverse association between increased particulate matter exposure and decreased DNA methylation in the iNOS gene. Whereas, our study observed a positive association between increased PM2.5 and increased DNA methylation in the iNOS gene. It is not clear why the direction of the effect differs between our study and the two previous studies but it is worth noting we measured different CpG sites within the promoter region of the iNOS gene and perhaps different CpG sites even within the same promoter region respond differently to environmental stimuli. Also, the composition of the particulate matter likely differed between the studies. It is also possible there was an uncontrolled confounding by smoking and/or second hand smoke exposure. As such, it is difficult to draw direct comparisons between the three studies.

We recognize that there are several limitations to our study. Methylation is tissue specific and we used DNA extracted from whole blood which is a mixture of different white blood cell types. Ultrafine particles diffuse across the alveoli and enter the bloodstream where they exert a direct effect on white blood cells by triggering an immune response. Since we did not measure white blood cell counts or measure DNA methylation in specific white blood cell sub-populations we cannot rule out that the increased iNOS DNA methylation was produced by systemic inflammation from inhaling fine particulate matter [13,14,48]. Another limitation of the study was that we did not collect samples that would allow for measurement of gene expression or production of nitric oxide to determine if the methylation changes in iNOS had any biological impact.

However, our study had several strengths including personal measurements of PM2.5 across the work shift which minimizes the potential for misclassification of exposure. We used pyrosequencing which is a very sensitive method and provides a quantitative measurement of DNA methylation. It is also worth noting that the blood samples were stored for several years in a cold storage facility at −80°C prior to analysis which likely reduced DNA yield but not DNA quality.

Conclusions

We observed a positive relationship between occupational exposure to fine particulate matter generated from welding activities and a modest increase in DNA methylation in the promoter region of the iNOS across a single work shift. We also observed a positive association between the number of years a person reported working as a boilermaker and increased DNA methylation in the iNOS gene. These results suggest that DNA methylation may be altered by PM2.5 over a short period of time. Further studies are needed that collect multiple measurements over longer periods of time to more fully understand the dynamic relationship between welding fume exposures and DNA methylation. These studies should also measure blood cell mixture to determine if the changes in DNA methylation are true epigenetic effects or a function of inflammation.

Abbreviations

Alu: A family of short interspersed elements; CpG: Cytosine-guanine nucleotide sequence; iNOS: Inducible nitric oxide synthase; LINE-1: Long interspersed nucleotide elements; LOD: Limit of detection; mC: Methylated cytosines; NO: Nitric oxide; PM2.5: Fine particulate matter with an aerodynamic diameter <2.5 μm; PM10: Fine particulate matter with an aerodynamic diameter <10 μm; PM: Particulate matter; SD: Standard deviation.

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

MLK – conceived the study, conducted the data analysis, and drafted the manuscript. SF – conducted the sample collection and exposure assessment, participated in the design of the study and helped with the manuscript preparation. AB – participated in the DNA methylation analsysis and helped with the manuscript preparation. LT – performed the DNA methylation analysis.JC – conducted the sample collection and exposure assessment, participated in the design of the study, and helped with the manuscript preparation.DCC –conceived of the study, and participated in its design and coordiation, and helped with the manuscript preparation. All authors read and approved the final manuscript.

Acknowledgments

This work was supported by US NIH Grants (Funding ES00002, NIEHS K01ES017800, NIEHS ES009860, T42 OH008416, and T32 ES 07069) and HSPH-NIEHS Center Pilot Grant/New Investigator funding. Additional research support was provided by the Flight Attendants Medical Research Institute. SF was supported by the American Heart Association.

GlinsmannPWRosenthalFSEvaluation of an aerosol photometer for monitoring welding fume levels in a shipyardAm Ind Hyg Assoc J19854639139510.1080/152986685913950303880193AntoniniJMHealth effects of weldingCrit Rev Toxicol2003336110310.1080/71361103212585507LaohaudomchokWCavallariJMFangSCLinXHerrickRFChristianiDCWeisskopfMGAssessment of occupational exposure to manganese and other metals in welding fumes by portable X-ray fluorescence spectrometerJ Occup Environ Hyg2010745646510.1080/15459624.2010.48526220526948CavallariJMEisenEAFangSCSchwartzJHauserRHerrickRFChristianiDCPM2.5 metal exposures and nocturnal heart rate variability: a panel study of boilermaker construction workersEnviron Health200873610.1186/1476-069X-7-3618613971AntoniniJMLeonardSSRobertsJRSolano-LopezCYoungSHShiXTaylorMDEffect of stainless steel manual metal arc welding fume on free radical production, DNA damage, and apoptosis inductionMol Cell Biochem2005279172310.1007/s11010-005-8211-616283511PalmerKTMcNeill LoveRMPooleJRCoggonDFrewAJLinakerCHShuteJKInflammatory responses to the occupational inhalation of metal fumeEur Respir J20062736637310.1183/09031936.06.0005320516452594WangZPLarssonKMalmbergPSjogrenBHallbergBOWrangskogKAsthma, lung function, and bronchial responsiveness in weldersAm J Ind Med19942674175410.1002/ajim.47002606037892825BeachJRDennisJHAveryAJBromlyCLWardRJWaltersEHStentonSCHendrickDJAn epidemiologic investigation of asthma in weldersAm J Respir Crit Care Med19961541394140010.1164/ajrccm.154.5.89127548912754FangSCCassidyAChristianiDCA systematic review of occupational exposure to particulate matter and cardiovascular diseaseInt J Environ Res Public Health201071773180610.3390/ijerph704177320617059RobinsonCSternFHalperinWVenableHPetersenMFrazierTBurnettCLalichNSalgJSesititoJFingerhutMAssessment of mortality in the construction industry in the United States, 1984–1986Am J Ind Med199528497010.1002/ajim.47002801057573075KimJYProutyLAFangSCRodriguesEGMagariSRModestGAChristianiDCAssociation between fine particulate matter and oxidative DNA damage may be modified in individuals with hypertensionJ Occup Environ Med2009511158116610.1097/JOM.0b013e3181b967aa19786898AndreadisAAHazenSLComhairSAAErzurumSCOxidative and nitrosative events in asthmaFree Radic Biol Med20033521322510.1016/S0891-5849(03)00278-812885584KimJYChenJCBoycePDChristianiDCExposure to welding fumes is associated with acute systemic inflammatory responsesOccup Environ Med20056215716310.1136/oem.2004.01479515723880WangZXNeuburgDLiCSuLKimJYChenJCChristianiDCGlobal gene expression profiling in whole-blood samples from individuals exposed to metal fumesEnviron Health Perspect200511323324110.1289/ehp.113-a23315687063FangSCCavallariJMEisenEAChenJCMittlemanMAChristianiDCVascular function, inflammation, and variations in cardiac autonomic responses to particulate matter among weldersAm J Epidemiol200916984885610.1093/aje/kwn40519153215MagariSRHauserRSchwartzJWilliamsPLSmithTJChristianiDCAssociation of heart rate variability with occupational and environmental exposure to particulate air pollutionCirculation200110498699110.1161/hc3401.09503811524390AritaACostaMEpigenetics in metal carcinogenesis: nickel, arsenic, chromium and cadmiumMetallomics2009122222810.1039/b903049b20461219BaccarelliABollatiVEpigenetics and environmental chemicalsCurr Opin Pediatr20092124325110.1097/MOP.0b013e32832925cc19663042BelinskySASnowSSNikulaKJFinchGLTellezCSPalmisanoWAAberrant CpG island methylation of the p16(INK4a) and estrogen receptor genes in rat lung tumors induced by particulate carcinogensCarcinogenesis20022333533910.1093/carcin/23.2.33511872642CostaMKleinCBNickel carcinogenesis, mutation, epigenetics, or selectionEnviron Health Perspect1999107A43843910.1289/ehp.99107a43810464080JiangGXuLSongSZhuCWuQZhangLEffects of long-term low-dose cadmium exposure on genomic DNA methylation in human embryo lung fibroblast cellsToxicology2008244495510.1016/j.tox.2007.10.02818077075KileMLBaccarelliAHoffmanETarantiniLQuamruzzamanQRahmanMMahiuddinGMostoaGHsuehYMWrightROChristianiDCPrenatal arsenic exposure and DNA methylation in maternal and umbilical cord leukocytesEnviron Health Perspect20121201061106610.1289/ehp.110417322466225KleinCBSuLBowserDLeszczynskaJChromate-induced epimutations in mammalian cellsEnviron Health Perspect200211073974312426123LeeYWBrodayLCostaMEffects of nickel on DNA methyltransferase activity and genomic DNA methylation levelsMutat Res Genet Toxicol Environ Mutagen199841521321810.1016/S1383-5718(98)00078-3WrightROSchwartzJWrightRJBollatiVTarantiniLParkSKHuHSparrowDVokonasPBaccarelliABiomarkers of lead exposure and DNA methylation within retrotransposonsEnviron Health Perspect201011879079510.1289/ehp.090142920064768ZhangJZhangJLiMJWuYJFanYYZhouYTanLLShaoZHShiHJMethylation of RAR-beta 2, RASSF1A, and CDKN2A Genes Induced by Nickel Subsulfide and Nickel-carcinogenesis in RatsBiomed Environ Sci20112416317121565688ZhousZHLeiYXWangCXAnalysis of aberrant methylation in DNA repair genes during malignant transformation of human bronchial epithelial cells induced by cadmiumToxicol Sci201212541241710.1093/toxsci/kfr32022112500TarantiniLBonziniMApostoliPPegoraroVBollatiVMarinelliBCantoneLRizzoGHouLFSchwartzJBertazziPABaccarelliAEffects of particulate matter on genomic DNA methylation content and iNOS promoter methylationEnviron Health Perspect200911721722219270791TarantiniLHouLBonziniMApostoliPPegoraroVSchwartzJEffects of exposure to metal-rich air particles on DNA methylationEpidemiology200920S127S128BaccarelliAWrightROBollatiVTarantiniLLitonjuaAASuhHHZanobettiASparrowDVokonasPSSchwartzJRapid DNA methylation changes after exposure to traffic particlesAm J Respir Crit Care Med200917957257810.1164/rccm.200807-1097OC19136372LiuJBallaneyMAl-alemUQuanCJinXPereraFChenLCMillerRLCombined inhaled diesel exhaust particles and allergen exposure alter methylation of T helper genes and IgE production in vivoToxicol Sci2008102768118042818RobertsonKDDNA methylation and human diseaseNat Rev Genet2005659761016136652BirdAPCpG-rich islands and the function of DNA methylationNature198632120921310.1038/321209a02423876KochanekSRenzDDoerflerWDNA methylation in the Alu sequences of diploid and haploid primary human cellsEMBO J199312114111518384552Gardiner-GardenMFrommerMCpG islands in vertebrate genomesJ Mol Biol198719626128210.1016/0022-2836(87)90689-93656447ChalitchagornKShuangshotiSHourpaiNKongruttanachokNTangkijvanichPThong-ngamDVoravudNSriuranpongVMutiranguraADistinctive pattern of LINE-1 methylation level in normal tissues and the association with carcinogenesisOncogene2004238841884610.1038/sj.onc.120813715480421RahmanIAdcockIMOxidative stress and redox regulation of lung inflammation in COPDEur Resp J20062821924210.1183/09031936.06.00053805TsutsuiMShimokawaHOtsujiYYanagiharaNPathophysiological relevance of NO signaling in the cardiovascular system: Novel insight from mice lacking all NO synthasesPharmacol Ther201012849950810.1016/j.pharmthera.2010.08.01020826180FerrisBGEpidemiology standardization project (American Thoracic Society)R Rev Respir Dis19781181120AntoniniJMMurthyGGKBrainJDResponses to welding fumes: Lung injury, inflammation, and the release of tumor necrosis factor-alpha and interleukin-1 betaExp Lung Res19972320522710.3109/019021497090873689184789NathanCInducible nitric oxide synthase: what difference does it make?J Clin Invest19971002417242310.1172/JCI1197829366554MoncadaSHiggsAThe L-arginine-nitric oxide pathwayN Engl J Med19933292002201210.1056/NEJM1993123032927067504210JalavaPISalonenROHalinenAIPenttinenPPennanenASSillanpaaMSandellEHillamoRHirvonenMRIn vitro inflammatory and cytotoxic effects of size-segregated particulate samples collected during long-range transport of wildfire smoke to HelsinkiToxicol Appl Pharmacol200621534135310.1016/j.taap.2006.03.00716678872LindbomJGustafssonMBlomqvistGDahlAGudmundssonASwietlickiELjungmanAGWear particles generated from studded tires and pavement induces inflammatory reactions in mouse macrophage cellsChem Res Toxicol20072093794610.1021/tx700018z17516662KimJYWandMPHauserRMukherjeeSHerrickRFChristianiDCAssociation of expired nitric oxide with occupational particulate exposureEnviron Health Perspect200311167668012727593KimJYHauserRWandMPHerrickRFAmarasiriwardenaCJChristianiDCThe association of expired nitric oxide with occupational particulate metal exposureEnviron Res20039315816610.1016/S0013-9351(03)00017-312963400SalemMTByunHMLurmannFBretonCVWangXEckelSPGillilandFDGenetic and epigenetic variations in inducible nitric oxide synthase promoter, particulate pollution, and exhaled nitric oxide levels in childrenJ Allergy Clin Immunol201212923223910.1016/j.jaci.2011.09.03722055874FangSCEisenEACavallariJMMittlemanMAChristianiDCAcute changes in vascular function among welders exposed to metal-rich particulate matterEpidemiology20081921722510.1097/EDE.0b013e31816334dc18300696