Grantee Research Project Results
Final Report: Validation of Diesel Exhaust Biomarkers
EPA Grant Number: R832097Title: Validation of Diesel Exhaust Biomarkers
Investigators: Zhang, Junfeng , Lioy, Paul J. , Stern, Alan , Kipen, Howard , Zhang, Lin , Fiedler, Nancy , Ohman-Strickland, Pamela , Laumbach, Robert
Institution: Environmental and Occupational Health Sciences Institute , University of Medicine and Dentistry of New Jersey , New Jersey Department of Environmental Protection
Current Institution: Environmental and Occupational Health Sciences Institute , New Jersey Department of Environmental Protection , University of Medicine and Dentistry of New Jersey
EPA Project Officer: Aja, Hayley
Project Period: May 1, 2005 through April 30, 2008
Project Amount: $572,497
RFA: Application of Biomarkers to Environmental Health and Risk Assessment (2004) RFA Text | Recipients Lists
Research Category: Air Quality and Air Toxics , Air Toxics , Air
Objective:
Health risk associated with diesel exhaust (DE) has been a concern of the EPA and the general public for decades, but is still poorly characterized, partially due to the lack of DE-specific markers or biomarkers of exposure. The overall goal of this study is to examine whether urinary amino-PAHs can serve as biomarkers of DE exposure, because amino-PAHs are metabolites of nitro-PAHs that are considered to be specifically emitted by diesel engines. Based on the relative abundance of their parent nitro-PAHs in DE, the following amino-PAHs in the urine excretion were targeted in this study: 1-aminopyrene, 1-aminonaphthalene, 2- aminonaphthalene, and 3-aminobenzanthrone. We recruited and retained 54 healthy nonsmoking adults who participated in this study as study subjects. Each of the study subjects underwent two 1-hour-long controlled exposure sessions: one being an atmosphere of diluted DE at a particle concentration of ~300 μg/m3, and the other being an atmosphere with filtered ambient air (as the control session). A urine void had been collected right before a subject started each exposure session; and the subject was asked to collect all urine voids within the 24 hours following exposure with designated containers. Urine samples were analyzed for the target amino-PAHs.
Specifically aims of this study include : (1) to optimize an amino-PAH analysis method to increase the sensitivities and recoveries; (2) to determine how long urinary amino-PAH concentrations reach maximum levels after the controlled one-hour DE exposure; (3) to quantify inhalation exposure to the parent nitro-PAHs of each of the target biomarkers during the entire time frame of urinary monitoring and minimize potential interferences from exposures that may occur before and after the controlled exposure session; (4) to estimate the fraction of inhaled nitro-PAHs converted into amino-PAH metabolites which are excreted in the urine; and (5) to assess inter-individual biomarker variability, with respect to several physiological factors such as gender, age, and body mass index.
Summary/Accomplishments (Outputs/Outcomes):
Aim 1: Development/Optimization of Analytical Method for Amino-PAHs
Previously published methods for analyzing amino-PAHs are based on GC-MS analysis of urine sample extracts (Seidel, et al., 2002; Weiss, et al., 2000; Grimmer, et al., 2000). Because amino-PAHs are not volatile enough for GC-MS analysis, they have to be first derivatized with pentafluoropropionic anhydride for the GC-MS analysis. In this study, we developed a method that uses an HPLC-fluorescence technique. The use of HPLC, instead of GC, enables the elimination of the derivatization step and consequently increases method recoveries. The use of fluorescence results in increased sensitivities. The method is described in an early publication resulting from this study (Laumbach, et al., 2009). Briefly, the method is described as follows.
Shortly after urine collection, the acetylated urinary amino-PAH conjugates were hydrolyzed with concentrated HCl (1 ml HCl 12M + 10 ml urine) at 80oC in a shaking bath for 1 hour. The mixture was then kept cool on an ice bath before extraction. During the extraction process, the pH of the urine-HCl mixture was first adjusted to 7 – 8 by adding ~ 1.1 ml of 10 M NaOH and a small quantity of concentrated acetic acid. The resulting solution was then extracted twice with 5 ml of dichloromethane by centrifugation at 2500 g for 10 min. The two dichloromethane extracts were combined in a glass tube and evaporated to near dryness using nitrogen at room temperature. The residue was rinsed from the glass tube wall and made into a final solution of 1 ml with acetonitrile. The final solution was filtered through a 0.2 µm PVDF Liquid Filter into a 1.8-ml amber vial that was sealed with a Teflon septum for HPLC analysis. The HPLC system used included a 600E System Controller, a 717 Auto Sampler, and a 2474 Fluorescence Detector (Waters Co., Milford, MA). An Ascentis RP-Amide column (4.6mm x 250mm) (Supelco Co., Bellefonte, PA) was used. The mobile phase involved the use of two solutions: Solution A = 50 % acetonitrile in water, and Solution B = 100 % acetonitrile, with, linear gradient from 100% A to 100% B in 30 min at a flow rate of 1.0 ml/min. The injection volume was 20 µL. The fluorescence detector was set at an excitation wavelength of 254 nm and an emission wavelength of 425 nm. An external standard of amino-PAHs was used to generate calibrations curves, all of which had a near-zero intercept and a R2 > 0.99. This method had a detection limit of 0.02 ng/ml for 1 aminopyrene and similar detection limits for the other amino-PAHs. The method had high reproducibility (with RSD <10% for repeated analyses) and high recovery (>85%), defined as ratio of the concentration determined from the HPLC analysis to the concentration spiked into a real urine sample.
Our method had sensitivity, precisions (reproducibility), and recovery for the target amino-PAH compounds similar to or better than the published methods. Among more than 700 samples analyzed in this study, 97% were above the detection limit for 1-aminonaphthalene, 99% for 2-aminonaphthalene, 76% for 1-aminopyrene, and 90% for 3-aminobenzanthrone.
In this study, we also measured nitro-PAHs (those of parent compounds of the target amino-PAHs) in the air during the DE exposure session and attempted to measure nitro-PAHs during the clean air session and during the 24 hours prior to and following an exposure session. Table 1 shows precisions, sensitivities, and recoveries of the methods.
Table 1. Analytical precisions, detection limits, and method recoveries of the method used in the present study for the three classes of compounds
Target Compounds |
Precision (% RSD)1 |
Detection Limit (ng/mL)2 |
Recovery (%)3 |
|
Amino-PAH:
|
1-Aminonaphthalene 2-Aminonaphthalene 1-Aminopyrene 3-Aminobenzanthrone 1 & 2-Nitronaphthalene 1-Nitropyrene 3-Nitrobenzanthrone |
4.72 7.31 10.1 12.6 9.20 6.33 10.1 |
0.04 0.09 0.02 0.08 0.15 0.06 0.52 |
82.6 85.9 88.3 85.2 89.4 91.7 87.9 |
Nitro-PAH:
|
1Precision: expressed as the coefficient of variation (CV) of repeated injection (%) (N =6).
2Detection Limit: expressed as 3 times the standard deviation of the lab blank or a low-concentration calibration standard (N=8).
3Recovery (%): expressed as the ratio (%) of the concentration measured to the concentration spiked into a real urine sample (concentration in the original urine was subtracted in the calculation).
Aim 2: Determination of Time of Maximum Excretion (Tmax)
In bio-monitoring of exposures, urine sampling represents a more convenient and less-invasive alternative to blood sampling; however, less work has been published on methodologies for characterizing the time course of excretion and the determination of the time of maximum excretion from urine samples. We compared two methods of characterizing the urine excretion profile and estimating the time of maximum excretion: non-compartmental analysis versus a nonlinear pharmacokinetic modeling. We examined these methodologies using both simulated data and the experimental data of 1-aminopyrene concentrations collected in this study. Simulated data demonstrated that the use of nonlinear modeling techniques to estimate pharmacokinetic parameters was more likely to estimate the true time of maximum excretion compared to the non-compartmental approach. Our analysis of observed concentrations of 1-aminopyrene, from 723 urine samples (see Table 2), led to a hypothesis that there are two subgroups of the subjects in terms of the timing of their 1-aminopyrene excretion. Results showed that approximately 70% of the subjects had a median time of maximum excretion (Tmax) of 5.9 hours, while 30% of the subjects may have had maximum excretion times longer than 24 hours. This finding is presented in a manuscript under review and a doctoral dissertation (Huyck, 2010).
Table 2. Summary of subjects and urine samples
|
Number |
Total # of subjects screened Total # of number subjects successfully studied # of male subjects # of female subjects Total # of subjects who failed initial health screening or dropped out Total # of urine samples collected Total # of valid urine samples # of valid urine samples for DE exposure # of valid urine samples for clean air control Total # of invalid/flagged urine samples # of invalid urine samples due to subjects drop out # of urine samples without creatinine concentration |
71 54 32 22 17 763 723 367 356 40 31 9 |
Aim 3: Quantification of Inhalation Exposure to Nitro-PAHs
During 1-hour DE exposure sessions (in the exposure chamber), we measured the nitro-PAHs concentrations as follows based on measurements from 6 sessions (n=6): the sum of 1-nitronaphthalene and 2-nitronaphthalene at 0.73 ± 0.39 ng/m3, 1-nitropyrene at 1.40 ± 1.08 ng/m3, and 3-nitrobenzanthrone at 0.73 ± 0.30 ng/m3. In contrsat, all measurements (n=4) made during 1-hour clean air sessions (in the exposure chamber) yelded deteable concentrations of target nitro-PAHs.
Our original study protocol includes the monitoring of personal air for 24-hours before and after each exposure or control session by having subjects carrying a personal sampling pump for 24 hours before and after each exposure session,. The purpose of this monitoring is to capture any diesel exposure occurring beyond the 1-hour controlled exposure session. After having analyzed 53 of 160 air samples collected, we discontinued the sample analysis and further collection of personal air samples, because none of the samples showed concentrations of any target nitro-PAHs above their detection limits. This does not necessarily mean that nitro-PAH exposures outside of the 1-hour control exposure were all negligible, because our personal sampling method, using a pump with a small sampling flow rate (4 l/min), did not have sufficient sensitivity to capture DE exposure events such as staying on or near diesel-powered buses as reflected from time-activity diaries.
Aim 4: Estimation of Amino-PAHs Excretions in Relation to DE Exposure
Because of practical constraints, the study protocol did not require the collection of the whole volume of each urine voids. This makes the calculation of "absolute" mass excreted impossible. Coupling with the fact that we were not able to capture possible DE exposure events outside of the 1-hour controlled exposure, our calculation to estimate the fraction of inhaled nitro-PAHs that were converted into amino-PAHs generated some not-so-useful results. Therefore, we focused on alternative approaches to examining whether the 1-hour controlled DE exposure resulted in increased urinary concentrations of amino-PAHs, compared to the 1-hour controlled clean air exposure.
Another challenge is that our urine collection scheme follows natural voids, instead of a fixed schedule. Hence, we aggregated urine concentrations into 2-hour intervals across all subjects. Our analysis focused on average concentration of urine samples collected in the 24 hours post clean air and DE exposures weighted by the interval of time between valid urine samples (see Method details in Laumbach, et al., 2009).
Box-plots were used to examine the general pattern of amino-PAHs concentrations over time following exposure. Median concentrations following the DE exposure were up to 5.1 times higher than median concentrations following the clean air exposure for 1-aminopyrene, 1.7 times for 1-aminonaphthalene, and 2.2 times for 2-aminonaphthalene, and 8.4 times for 3-aminobenzanthrone.
From histograms of the time-weighted concentrations following the DE exposure and the clean air exposure, it is clear that the distribution shifted towards higher 1-aminopyrene concentrations following the DE exposure compared to the distribution of concentrations following clean air exposure. Following the DE exposure, < 40% of the subjects had 1-aminopyrene concentrations ≤ 100 ng/g creatinine, compared to 90% of the subjects with average time-weighted 1-aminopyrene concentration ≤ 100 ng/g creatinine following the clean air exposure. Similar patterns were observed for the other target amino-PAHs.
Because the data distributions are highly skewed, we used the signed rank test to compare the time-weighted concentrations following the DE exposures and those following the clean air exposure. The difference was significant for each of the target amino-PAHs (P < 0.0001 for 1-aminopyrene, P < 0.0058 for 1-aminonaphthalene, P < 0.0001 for 2-aminonaphthalene, and P=0.0001 for 3-aminobenzanthrone). In the case of 1-aminopyrene, in fact, 33 out of 38 subjects (86.8%, 95% CI: 73.1, 97.6%) had higher time-weighted concentrations following the DE exposure than following the clean air exposure. This is reflected in the large difference in medians of the time-weighted concentrations for the 38 subjects between the two exposure conditions (138.7 ng/g for the DE and 21.7 ng/g for the clean air). Because the data were highly skewed, the difference in mean time-weighted concentrations were smaller (with large standard deviations): 324.0 ± 442.4 ng/g following the DE exposure and 234.4 ± 852.9 ng/g following the clean air exposure.
We also made comparisons of changes in 1-aminopyrene concentrations from baseline to either first void or the peak (i.e., the highest concentrations among all the post-exposure spot urines). The signed rank tests revealed significant differences between the DE and clean control exposures as measured by the change from baseline to either of these post-exposure spot urine samples (p=0.027 for first voids; p=0.0026 for peak-concentration voids).
Aim 5: Assessment of Inter-person Variability in Relation to Demographic Variables
We observed a large inter-person variability, in both the concentration of urinary 1-aminopyrene and the time course of appearance in the urine following the controlled 1-hour DE exposure. This suggests the need to explore subject variables that may affect conversion of inhaled 1-nitropyrene to urinary excretion of 1-aminopyrene.
We used linear mixed-effects models to explore the inter-individual variability with respect to gender, age, and body mass index (BMI). The time-weighted-average concentrations of the target amino-PAHs were calculated for DE and clean air exposure. Then differences in time-weighted-average concentrations between DE and clean air exposure were regressed against gender, age and BMI in order to determine whether these predictors explain inter-person variability in exposure effects. It was found that the effect of exposure type on each of the target amino-PAHs was not modified by gender, age or BMI.
Conclusions:
Data analyses performed so far lead to the following conclusions:
- The HPLC-fluorescence technique we used in the present study was sensitive enough to quantify the four urinary amino-PAHs in spot urine samples of non-occupationally exposed volunteers, with and without following a controlled 1-hour diesel exhaust (DE) exposure.
- Differences between pre- and post-exposure urinary amino-PAH concentrations were greater after exposure to DE compared to the clean air control.
- Based on the analysis of the 1-aminopyrene data, we suggest the presence of two distinct subgroups of the subjects in terms of the timing of their amino-PAH excretion following the 1-hour controlled DE exposure. The majority of the subjects appeared to have the maximum excretion within several hours (e.g., the median time of maximum 1-aminopyrene excretion was 5.9 hours), while a smaller fraction (e.g., ~ 30% in the case of 1-aminopyrene execration) may have had maximum excretion times longer than 24 hours.
- Large inter-individual variation in the time courses of amino-PAHs in the urine may limit the utility of spot samples for quantifying individual exposure to nitro-PAHs markers of DE exposure, but this variability may be less problematic when employing such data to compare exposure levels in populations with differential exposures.
- It is recommended to examine whether steady-state urinary amino-PAH concentrations are higher in routinely exposed individuals (e.g., people who live or work close to DE sources).
References:
Grimmer G, Dettbarn G, Seidel A, Jacob J. Detection of carcinogenic aromatic amines in the urine of non-smokers. Science of the Total Environment 2000;247:81-90.Seidel A, Dahmann D, Krekeler H, and Jacob J. Biomonitoring of polycyclic aromatic compounds in the urine of mining workers occupationally exposed to diesel exhaust. International Journal of Hygiene and Environmental Health 2002;204(5-6):333-338.
Weiss T and Angerer J. Simultaneous determination of various aromatic amines and metabolites of aromatic nitro compounds in urine for low level exposure using gas chromatography-mass spectrometry. Journal of Chromatography 2000;247:81-90.
Laumbach R., Tong J., Zhang L., Ohman-Strickland P., Stern A., Fiedler N., Kipen H., Kelly-McNeil K., Lioy P., Zhang J. Quantification of 1-aminopyrene in human urine after a controlled exposure to diesel exhaust. Journal of Environmental Monitoring 2009;11:153-159.
Huyck S. Modeling Times of Maximum Biomarker Excretion. Doctoral Dissertation, School of Public Health, UMDNJ and Rutgers University, Piscataway, New Jersey, January 2010.
Journal Articles on this Report : 2 Displayed | Download in RIS Format
Other project views: | All 4 publications | 2 publications in selected types | All 2 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Huyck S, Ohman-Strickland P, Zhang L, Tong J, Xu XU, Zhang J. Determining times to maximum urine excretion of 1-aminopyrene after diesel exhaust exposure. Journal of Exposure Science and Environmental Epidemiology 2010 (May 5-Advance online publication). |
R832097 (Final) |
Exit |
|
Laumbach R, Tong J, Zhang L, Ohman-Strickland P, Stern A, Fiedler N, Kipen H, Kelly-McNeil K, Lioy P, Zhang J. Quantification of 1-aminopyrene in human urine after a controlled exposure to diesel exhaust. Journal of Environmental Monitoring 2009;11(1):153-159. |
R832097 (Final) R832144 (Final) R832515 aka R832098 (2008) |
Exit |
Supplemental Keywords:
air toxics, diesel exposure, 1-aminopyrene, amino-PAHs, nitro-PAHs, RFA, Health, Scientific Discipline, Air, particulate matter, Environmental Chemistry, Health Risk Assessment, Risk Assessments, Environmental Monitoring, ambient air quality, atmospheric particulate matter, particulates, air toxics, atmospheric particles, chemical characteristics, ambient air monitoring, airborne particulate matter, environmental risks, inner city, air pollution, diesel exhaust, PAH, aerosol composition, atmospheric aerosol particles, human exposure, biomarkerProgress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.