Appendix A -cdn.com



Appendix A: Air pollutants measured and population characteristics of the RAPTES studyTable A.1. 48 individual air pollutants measured on site in the RAPTES study. Air pollutantAcronymMean (sd)UnitsNitrogen monoxideNO37.25 (20.48)ppbNitrogen dioxide NO220.96 (6.88)ppbNitrogen oxides NOx 16.29 (14.82)ppbOzone O3 19.47 (6.86)ppbParticulate matter with a diameter smaller than 10 ?m PM10 41.75 (24.89)ug/m3Particulate matter with a diameter smaller than 2.5 ?mPM2.5 27.27 (20.68)ug/m3Particle number concentration PNC29.34 (24.12)10^3/cm3absorption coefficient of PM10Absorption (PM10)3.43 (2.32)--absorption coefficient of PM2.5Absorption (PM2.5)3.27 (2.22)--Elemental carbon determined in PMcoarseEC (PMcoarse)0.24 (0.19)ug/m3Elemental carbon determined in PM2.5 EC (PM2.5)3.01 (2.25)ug/m3Endotoxin determined in PM10 Endotoxin (PM10)5.13 (10.68)EU/m3NO3 determined in PMcoarseNO3 (PMcoarse)1.44 (1.11)ug/m3NO3 determined in PM2.5NO3 (PM2.5)7.44 (9.02)ug/m3Organic carbon determined in PMcoarse OC (PMcoarse) 1.61 (1.11)ug/m3Organic carbon determined in PM2.5 OC (PM2.5) 1.60 (1.72)ug/m3Oxidative Potential AA assay PM10a OP AA (PM10)79.30 (95.16)nmol AA/s/m3Oxidative Potential AA assay PM2.5a OP AA (PM2.5)48.10 (58.15)nmol AA/s/m3Oxidative Potential DTT assay PM10b OP DTT (PM10)2.74 (1.70)nmol DTT/min/m3Oxidative Potential DTT assay PM2.5b OP DTT (PM2.5)2.43 (1.48)nmol DTT/min/m3Oxidative Potential ESR assay PM10c OP ESR (PM10)9637.63 (11924.13)AU/m3Oxidative Potential ESR assay PM2.5c OP ESR (PM2.5)4386.09 (4918.21)AU/m3Sulfate determined in PMcoarse SO4 (PMcoarse)0.40 (0.18)ug/m3Sulfate determined in PM2.5 SO4 (PM2.5 )4.27 (4.45)ug/m3Aluminum determined in PMcoarse Al (PMcoarse )71.80 (39.67)ng/m3Aluminum determined in PM2.5 Al (PM2.5)74.71 (65.98)ng/m3Barium determined in PMcoarse Ba in PMcoarse) 12.35 (9.86)ng/m3Barium determined in PM2.5 Ba (PM2.5) 32.30 (14.55)ng/m3Copper determined in PMcoarse Cu (PMcoarse) 21.25 (17.15)ng/m3Copper determined in PM2.5 Cu (PM2.5) 19.66 (16.83)ng/m3Iron determined in PMcoarse Fe (PMcoarse) 5.41 (4.29)ng/m3Iron determined in PM2.5 Fe (PM2.5) 4.00 (3.45)ng/m3Manganese determined in PMcoarse Mn (PMcoarse)12.04 (8.59)ng/m3Manganese determined in PM2.5 Mn (PM2.5) 11.82 (8.67)ng/m3Nickel determined in PMcoarse Ni (PMcoarse)1.04 (0.91)ng/m3Nickel determined in PM2.5 Ni (PM2.5) 5.61 (9.31)ng/m3Lead determined in PMcoarse Pb (PMcoarse) 2.24 (1.34)ng/m3Lead determined in PM2.5 Pb (PM2.5) 7.04 (5.14)ng/m3Antimony determined in PMcoarse Sb (PMcoarse)4.23 (4.40)ng/m3Antimony determined in PM2.5 Sb (PM2.5) 2.71 (3.00)ng/m3Strontium determined in PMcoarse Sr (PMcoarse) 2.06 (0.83)ng/m3Strontium determined in PM2.5 Sr (PM2.5) 1.51 (1.40)ng/m3Titanium determined in PMcoarse Ti (PMcoarse) 2.43 (1.46)ng/m3Titanium determined in PM2.5 Ti (PM2.5) 1.94 (1.24)ng/m3Vanadium determined in PMcoarse V (PMcoarse)0.54 (0.25)ng/m3Vanadium determined in PM2.5 V (PM2.5) 2.98 (2.39)ng/m3Zinc determined in PMcoarse Zn (PMcoarse) 67.64 (153.49)ng/m3Zinc determined in PM2.5 Zn (PM2.5) 273.25 (523.04)ng/m3a Oxidative potential measured by Ascorbate (AA) depletion.b Oxidative potential measured by electron spin resonance.c Oxidative potential measured by dithiothreitol.Table A.2. Population characteristics and baseline (t=0) FENO, lung function, and acute phase proteins in blooda,b.Characteristic ValueValueAge (years) 22 (19-26)Female, n (%)21 (68%)Body Mass Index (kg/m2) 22.3 (17.0-32.0)FENO (ppb)c 15.9 (5-61)FEV1 (L)c 3.86 (2.57-5.51)FVC (L)c 4.68 (2.73-6.70)CRP (mg/L)c1.00 (0.10–14.48)Fibrinogen (g/L)3.02 (1.43–5.19)IL-6 (ng/mL)c1.49 (0.27-8.30)Platelet counts (109/L)267.79 (130.00–416.00)vWF (% of normal)c89.41 (37.72–199.64)tPA/PAI-1 complex (ng/mL)c2.81 (0.08–27.39)a Results reproduced from (Strak et al. 2012, 2013).b Unless otherwise stated, values are mean (range) or geometric means (ranges) for the biomarkers.c FEno: concentration of NO in exhaled breath, FEV1: forced expiratory volume in 1 second, FVC: Forced Vital Capacity, CRP: c-reactive protein, IL-6: interleukin 6, vWF: von Willebrand Factor, tPA/PAI-1 complex: tissue plasminogen activator / plasminogen activator inhibitor-1 complex.References:Strak M, Hoek G, Godri KJ, Gosens I, Mudway IS, van Oerle R, et al. 2013. Composition of PM affects acute vascular inflammatory and coagulative markers - the RAPTES project. PLoS One 8:e58944.Strak M, Janssen NAH, Godri KJ, Gosens I, Mudway IS, Cassee FR, et al. 2012. Respiratory health effects of airborne particulate matter: the role of particle size, composition, and oxidative potential-the RAPTES project. Environ. Health Perspect. 120:1183–9.Appendix B: Assessment of the human blood metabolome.Sample preparationAn aliquot of 50 microL of serum samples was extracted with 250 microL ice-cold methanol (Sigma Aldrich catalogue number - 14262?FLUKA). After adding the methanol, samples were shaken and kept on ice for 1 hour. Then, they were centrifuged at 13000 rpm for 10 minutes at 4 °C. A 200 microL of the supernatant was transferred to 4 aliquots and dried in Speed-Vac and stored at -20 °C until analysis. One aliquot was then reconstituted in 30 microL of 50% methanol-water, centrifuged at 13000 rpm for 2 minutes and 20 microL was then transfer to a QSERT vial (Sigma-Aldrich catalogue number 29392-U?SUPELCO). UHPLC-QTOF analysisSamples were analyzed with a UHPLC-QTOF-MS system (Agilent Technologies) consisting of a 1290 Binary LC system, a Jet Stream electrospray ionization (ESI) source, and a 6550 QTOF mass spectrometer. Autosampler tray was kept refrigerated and 1 uL of the sample solution was injected on an ACQUITY UPLC HSS T3 column (2.1 × 100mm, 1.8 μm; Waters). Column temperature was 45 °C and mobile phase flow rate 0.4 ml/min, consisting of ultrapure water and LC-MS grade methanol, both containing 0.1 % (v/v) of formic acid. The gradient profile was as follows: 0–6 min: 5% → 100% methanol, 6–10.5 min: 100% methanol, 10.5–13 min: 5% methanol.The mass spectrometer was operated in positive polarity using the following conditions: drying gas (nitrogen) temperature 175 °C and flow 12 L/min, sheath gas temperature 350 °C and flow 11 L/min, nebulizer pressure 45 psi, capillary voltage 3500 V, nozzle voltage 300 V, and fragmentor voltage 175 V. Data acquisition was performed using 2 GHz extended dynamic range mode across a mass range of 50–1200. Scan rate was 1.67 Hz and data acquisition was in centroid mode. Continuous mass axis calibration was performed by monitoring two reference ions throughout the runs (m/z 121.050873 and m/z 922.009798). Data was acquired using MassHunter Acquisition B.05.01 (Agilent Technologies). The analytical run was initiated with priming injections of a QC sample to achieve stable instrument response, following by study samples, intervened with a QC sample to monitor instrument performance and sample stability. Data processingRaw data preprocessing was performed using Qualitative Analysis B.06.00, DA Reprocessor, and Mass Profiler Professional 12.1 software ((Agilent Technologies). Recursive feature finding was employed to find compounds as singly charged proton adducts, by using data from all study samples. The initial processing of the data was performed using an MFE algorithm set to small molecules, and threshold for mass and chromatographic peak heights were 900 and 6000 counts, respectively. A single mass peak was considered a feature if neutral mass could be calculated and peak spacing tolerance for isotope peaks was 0.0025 m/z plus 7 ppm, with the isotope model set to common organic molecules. Only singly charged ions were included. After the initial feature finding, the compounds that existed in at least 5 samples were combined into a single list, using 0.1 min retention time window and 15 ppm + 2 mDa mass window for alignment. The resulting list was used as a target for the recursive feature extraction of the data, which was performed using an Agilent FBF algorithm with match tolerance for the compound mass and retention time set at ±10 ppm and ±0.05 min. Multiply charged ions were excluded, ion species were limited to M+ and [M+H]+ and a filter for the chromatographic peak height set at 5000 counts. The resulting files were then merged using the same alignment settings as above, to generate the final data matrix.Appendix C: Visual confirmation associations air pollutants and metabolic featuresIn Figures S1-S4 we show patterns across sampling locations in levels of air pollutants and robustly, associated metabolic features. Locations are ordered by the location specific median air pollutant level. In case multiple metabolic features were robustly associated to an air pollutant (n>1 in the title of the plot), boxplots were generated for the mixture of the independent metabolic features (the variables were concatenated). For each plot we estimated the trend for the metabolic feature concentration across the four study locations ranked by their median air pollutant level and its significance level. These plots illustrate that for most air pollutant – metabolic feature associations the exposure contrast between sampling locations contributed to the observed associations. Positive associations generally resulted in a positive slope across sampling locations and negative associations resulted in a negative slope. In addition to contrast between locations there was also contrast in air pollutants within locations, which was incorporated in our main analysis. 6372764222750B00B6391812974975A00A6347361692275B00B633095464820A00AFigure C.1. Patterns across sampling locations in levels of air pollutants and robustly, positively associated metabolic features measured at t9. In panel A we show boxplots of the air pollutant concentrations at each location. In panel B we show boxplots of the change in concentration from t0 to t9 of the corresponding metabolic features. In case multiple metabolic features were robustly associated to an air pollutant (n>1 in the title of the plot), boxplots were generated for the mixture of the independent metabolic features. Boxplots were ordered by location specific median air pollutant level. Label x-axis: Con=Urban background site, Far=Farm location, HIA=Continuous traffic location, Mar=Stop-and-go traffic location. Label y-axis exposure plots: units correspond to Supplemental Material I, Table S1. Label y-axis metabolite plots: log-transformed peak areas. Slope and p-value are from a simple linear regression using the exposure ranking of the sampling location as explanatory variable. We consider a p value <0.1 for the trend estimate noteworthy.8985252931795A00A8940801649095B00B8966204179570B00B893181421640A00AFigure C.2. Patterns across sampling locations in levels of air pollutants and robustly, negatively associated metabolic features measured at t9. In panel A we show boxplots of the air pollutant concentrations at each location. In panel B we show boxplots of the change in concentration from t0 to t9 of the corresponding metabolic features. In case multiple metabolic features were robustly associated to an air pollutant (n>1 in the title of the plot), boxplots were generated for the mixture of the independent metabolic features. Boxplots were ordered by location specific median air pollutant level. Label x-axis: Con=Urban background site, Far=Farm location, HIA=Continuous traffic location, Mar=Stop-and-go traffic location. Label y-axis exposure plots: units correspond to Supplemental Material I, Table S1. Label y-axis metabolite plots: log-transformed peak areas. Slope and p-value are from a simple linear regression using the exposure ranking of the sampling location as explanatory variable. We consider a p value <0.1 for the trend estimate noteworthy.13757813164205B00B1377686944880A00AFigure C.3. Patterns across sampling locations in levels of air pollutants and robustly, positively associated metabolic features measured at t25. In panel A we show boxplots of the air pollutant concentrations at each location. In panel B we show boxplots of the change in concentration from t0 to t25 of the corresponding metabolic features. In case multiple metabolic features were robustly associated to an air pollutant (n>1 in the title of the plot), boxplots were generated for the mixture of the independent metabolic features. Boxplots were ordered by location specific median air pollutant level. Label x-axis: Con=Urban background site, Far=Farm location, HIA=Continuous traffic location, Mar=Stop-and-go traffic location. Label y-axis exposure plots: units correspond to Supplemental Material I, Table S1. Label y-axis metabolite plots: log-transformed peak areas. Slope and p-value are from a simple linear regression using the exposure ranking of the sampling location as explanatory variable. We consider a p value <0.1 for the trend estimate noteworthy.14122403085465B00B1414409866140A00AFigure C.4. Patterns across sampling locations in levels of air pollutants and robustly, negatively associated metabolic features measured at t25. In panel A we show boxplots of the air pollutant concentrations at each location. In panel B we show boxplots of the change in concentration from t0 to t25 of the corresponding metabolic features. In case multiple metabolic features were robustly associated to an air pollutant (n>1 in the title of the plot), boxplots were generated for the mixture of the independent metabolic features. Boxplots were ordered by location specific median air pollutant level. Label x-axis: Con=Urban background site, Far=Farm location, HIA=Continuous traffic location, Mar=Stop-and-go traffic location. Label y-axis exposure plots: units correspond to Supplemental Material I, Table S1. Label y-axis metabolite plots: log-transformed peak areas. Slope and p-value are from a simple linear regression using the exposure ranking of the sampling location as explanatory variable. We consider a p value <0.1 for the trend estimate noteworthy.Appendix D: Correlations between metabolic featuresFigure D.1. Density plot of Pearson correlation coefficients between differences in metabolic features t9 – t0 (median r = 0.03; solid line) and t25-t0 (median r = 0.53; dashed line). Appendix E: Mediation analysesTable E.1. Evidence for mediation of a significant air pollutant - health marker association by metabolic features measured at t9aMI (Da)exposureboutcomec,dp-value exposure outcome associationp-value for evidence mediation via metabolite178.0415NOFEV1 (t25)0.0200.032178.0415NOFEV1 (t9)6.810*e-40.066178.0415NOxFEV1 (t25)0.0070.022178.0415NOxFEV1 (t9)0.0010.082272.1554OP AA (PM10)FEV1 (t9)0.0330.186272.1554Absorption (PM10)FEV1 (t9)0.0030.024272.1554Absorption (PM25)FEV1 (t9)0.0030.038133.0504Absorption (PM25)FEV1 (t9)0.0030.382573.7178Cu (PMcoarse)FEV1 (t25)0.0480.308772.0059Cu (PMcoarse)FEV1 (t25)0.0480.270772.0059Cu (PMcoarse)FEV1 (t9)0.0140.782178.0415Cu (PMcoarse)FEV1 (t9)0.0140.064178.0415Cu (PMcoarse)FEV1 (t25)0.0480.012606.3815Sb (PMcoarse)FEV1 (t9)0.0120.342772.0059Sb (PMcoarse)FEV1 (t25)0.0410.256772.0059Sb (PMcoarse)FEV1 (t9)0.0120.846a Table includes metabolic features that were associated to both an air pollutant and a health marker. In case no association between the air pollutant and the health marker was observed, metabolic features were excluded. We included exposure-health effect associations in mediation analysis if the p-value was <0.05, we considered a p value <0.1 for the indirect effect noteworthy.b Trace elements, endotoxin, and absorption determined in particles with a diameter <2.5 ?m (PM2.5), <10 ?m (PM10), and between 2.5-10 ?m (PMcoarse); OP AA ( oxidative potential measured by the extent of ascorbate depletion), OP ESR (oxidative potential measured by electron spin resonance). c Concentration of NO in exhaled breath (FENO), Forced Vital Capacity (FVC), forced expiratory volume in 1 second (FEV1), interleukin 6 (IL-6), tissue plasminogen activator / plasminogen activator inhibitor-1 complex (tPA/PAI-1), Von Willebrand factor (VWF), c-reactive protein (CRP).d t9, determined two hours after exposure; t25, determined eighteen hours after exposure.Table E.2. Evidence for mediation of a significant air pollutant - health marker association by metabolic features measured at t25aMI (Da)exposureboutcomec,dp-value exposure outcome associationp-value for evidence mediation via metabolite302.1696OP ESR (PM2.5)FEV1 (t9)0.0240.242431.323OP ESR (PM2.5)FEV1 (t9)0.0240.432620.4016OP ESR (PM2.5)FEV1 (t9)0.0240.984418.31SO4 (PMcoarse)FEV1 (t9)0.0400.240424.8007SO4 (PMcoarse)FEV1 (t9)0.0400.658427.7952SO4 (PMcoarse)FEV1 (t9)0.0400.694431.808SO4 (PMcoarse)FEV1 (t9)0.0400.534447.3283SO4 (PMcoarse)FEV1 (t9)0.0400.598458.3098SO4 (PMcoarse)FEV1 (t9)0.0400.570471.8211SO4 (PMcoarse)FEV1 (t9)0.0400.810474.3145SO4 (PMcoarse)FEV1 (t9)0.0400.710483.3439SO4 (PMcoarse)FEV1 (t9)0.0400.598493.8345SO4 (PMcoarse)FEV1 (t9)0.0400.722498.3647SO4 (PMcoarse)FEV1 (t9)0.0400.948498.8651SO4 (PMcoarse)FEV1 (t9)0.0400.554512.854SO4 (PMcoarse)FEV1 (t9)0.0400.770514.3698SO4 (PMcoarse)FEV1 (t9)0.0400.656515.8453SO4 (PMcoarse)FEV1 (t9)0.0400.556523.8418SO4 (PMcoarse)FEV1 (t9)0.0400.668548.3765SO4 (PMcoarse)FEV1 (t9)0.0400.830549.3962SO4 (PMcoarse)FEV1 (t9)0.0400.808552.8643SO4 (PMcoarse)FEV1 (t9)0.0400.754553.3675SO4 (PMcoarse)FEV1 (t9)0.0400.888562.7231SO4 (PMcoarse)FEV1 (t9)0.0400.634571.4084SO4 (PMcoarse)FEV1 (t9)0.0400.550574.3921SO4 (PMcoarse)FEV1 (t9)0.0400.530596.7477SO4 (PMcoarse)FEV1 (t9)0.0400.602601.4219SO4 (PMcoarse)FEV1 (t9)0.0400.908610.9052SO4 (PMcoarse)FEV1 (t9)0.0400.608639.9266SO4 (PMcoarse)FEV1 (t9)0.0400.642653.9403SO4 (PMcoarse)FEV1 (t9)0.0400.628661.9418SO4 (PMcoarse)FEV1 (t9)0.0400.592674.471SO4 (PMcoarse)FEV1 (t9)0.0400.986684.4522SO4 (PMcoarse)FEV1 (t9)0.0400.142696.9832SO4 (PMcoarse)FEV1 (t9)0.0400.942706.4708SO4 (PMcoarse)FEV1 (t9)0.0400.204711.9881SO4 (PMcoarse)FEV1 (t9)0.0400.558714.4672SO4 (PMcoarse)FEV1 (t9)0.0400.420770.4774SO4 (PMcoarse)FEV1 (t9)0.0400.272775.5642SO4 (PMcoarse)FEV1 (t9)0.0400.924510.2516SO4 (PM2.5)Fibrinogen (t25)0.0320.026a Table includes metabolic features that were associated to both an air pollutant and a health marker. In case no association between the air pollutant and the health marker was observed, metabolic features were excluded. We included exposure-health effect associations in mediation analysis if the p-value was <0.05, we considered a p value <0.1 for the indirect effect noteworthy. b Trace elements, endotoxin, and absorption determined in particles with a diameter <2.5 ?m (PM2.5), <10 ?m (PM10), and between 2.5-10 ?m (PMcoarse); OP AA ( oxidative potential measured by the extent of ascorbate depletion), OP ESR (oxidative potential measured by electron spin resonance). c t9, determined two hours after exposure; t25, determined eighteen hours after exposure.d Concentration of NO in exhaled breath (FENO), Forced Vital Capacity (FVC), forced expiratory volume in 1 second (FEV1), interleukin 6 (IL-6), tissue plasminogen activator / plasminogen activator inhibitor-1 complex (tPA/PAI-1), Von Willebrand factor (VWF), c-reactive protein (CRP).Appendix F: Variability in metabolic features over timeThe design of RAPTES (3 repeated blood measurements within 24 hours for each individual, and repeated pre-exposure blood samples for each individual) provided us with the opportunity to explore the variability of the metabolic features within individuals over time (assuming most metabolic features were affected by air pollution). We assessed the variability by calculating the intraclass correlation coefficient (ICC), a measure of the reproducibility of a measurement when it is randomly repeated for the same individual. We assessed three different ICC’s: reflecting the variability of measured metabolic feature intensities at t0 from day to day (ICCt0), reflecting the variability of metabolic feature intensities measured at t0 and t9 on the same day (ICCt0_t9), and reflecting the variability of metabolic feature intensities measured at t0 and t25 on the same day (ICCt0_t25). 3878580101600b00b1685290102870a00a-7341-76003882445134675d00d1686560133350c00c3876730181555f00f1680845172720e00eFigure F.1. Histograms showing the distribution of the Intraclass Correlation Coefficient (ICC) for all metabolic features, calculated for variation across and within sampling days. Panels a-c: ICC reflecting day-to-day variation in metabolic feature intensities measured at t0, t9, and t25. Panels d-f: ICC reflecting within day variation (from t0 to t9, from t0 to t25, from t9 to t25) in metabolic feature intensities. Median ICCs were 0.41 (panel a), 0.41 (panel b), 0.42 (panel c), 0.51 (panel d), 0.51 (panel e), 0.52 (panel f).Appendix G. Identification of tyrosine, hypoxanthine, and guanosineFor confirmation of the compounds, a representative sample was reanalyzed together with a standard solution prepared in methanol/water. Standards were from IROA Technologies’ Mass Spectrometry Metabolite Library of Standards (IROA Technologies, Bolton, MA). The analysis of the samples and standards was performed in one uninterrupted analytical run. Injection volume was 0.5 ?L for both standards and samples and the same acquisition method was employed as described in the methods section. Chromatograms and spectra were drawn using Find by Formula algorithm in Agilent MassHunter Qual B.06.00. Product ion spectra of the ion of interest were acquired with 1.3 Da isolation widths from both the sample and standard. Matching of the retention times, isotope patterns and product ion spectra were based on visual inspection of the data provided in this appendix. Figure. G.1. Chromatograms, isotope patterns, and product ion spectra of the three identified compounds. The boxes drawn on top of the measured ions represent isotope patterns calculated for the elemental composition of the analyte. In the product ion spectra, the precursor ion is indicated with a rectangular dot above the mass peak. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download