Oxidative potential and cancer risk of ambient particles in an urban coastal area of China: chemical characterization, source apportionment, and respiratory deposition

Abstract

Particulate matter (PM) continues to be a significant contributor to the deterioration of air quality and is known to cause both acute and chronic health effects in exposed populations. The health effects associated with PM depend on their size, chemical composition, and sources. For instance, the chemical composition of PM, including polycyclic aromatic hydrocarbons (PAHs), are recognized for their highly carcinogenic properties. The assessment of health risks associated with PM is increasingly focused on measuring oxidative potential (OP) rather than mass concentration. OP has emerged as a highly promising metric in this regard. The study focuses on two specific aspects: the cancer risk associated with PAHs and the OP induced by the chemical compositions of PM. The study commenced with a comprehensive year-round field campaign for aerosol sampling in Ningbo, a coastal city in China. The field campaign focused on PM with an aerodynamic diameter of ≤2.5 μm (PM2.5), 2.5‒10 μm (PM10), and size fractionated PM. The study involved detailed chemical characterization, analysis of source contributions to PAHs lung cancer risk and OP, and modeling the deposition of OP in various regions of the human respiratory tract.
The risk assessment of PAHs documented in the literature has predominantly concentrated on the 16 priority PAHs, inadvertently neglecting other PAHs that possess a high cancer risk. Consequently, there is a risk of underestimating the overall risk level. Additionally, in this study, we utilized the World Health Organization (WHO) and Environmental Protection Agency (EPA) unit risk methods to estimate the Lifetime Excess Cancer Risk (LECR) and address the errors frequently reported in the literature regarding these methods. The results of the lung cancer risk assessment conducted in this study revealed that the inclusion of highly carcinogenic PAHs, such as 7H-benzo[c]fluorene and various dibenzopyrene derivatives (e.g., dibenzo[a,h]pyrene, dibenzo[a,l]pyrene, and dibenzo[a,e]pyrene), increased the risk of lung cancer by four-fold. This significant increase in LECR suggests that future investigations should consider incorporating more highly carcinogenic PAHs into the risk assessment framework to accurately estimate the lung cancer risk posed by PAHs. The traditional approach for estimating the lung cancer risk associated with PAHs is the use of the component-based potency factor, as adopted by the EPA. In contrast, the WHO approach requires the use of benzo[a]pyrene (BaP) as a marker for the complex mixture. Consequently, employing the WHO unit risk and component-based potency factor approach resulted in significant overestimation of the lung cancer risk. In our study, we utilized the EPA unit risk and component-based potency factor approach to estimate the LECR of 16 and 20 PAHs, resulting in values of 5.1×10-7 and 2.23×10-6, respectively. However, when employing the WHO unit risk and component-based potency factor approach, we observed a 14-fold increase in the LECR estimates for the same set of 16 and 20 PAHs (7.45×10-6 and 32.4×10-6, respectively). By adopting the appropriate WHO approach, we estimated the LECR to be 3.11×10-6, which closely aligns with the LECR estimated for the 20 PAHs using the EPA approach. Our findings revealed that incorporating additional PAHs in the component-based potency factor approach enhanced the comparability of EPA estimations with those of the WHO. This, in turn, led to a more precise assessment of the lung cancer risk associated with PAH exposure. Moreover, in our study, we identified five sources associated with PAHs in our study domain. These sources include natural gas combustion, vehicular exhaust, coal combustion, biomass burning, and the volatilization of urban fuel. Among these sources, our finding indicated that natural gas combustion, coal combustion, and vehicular exhaust emissions make significant contributions to the LECR. Therefore, implementing measures to reduce the levels of PAHs from these three sources would significantly impact the reduction of LECR and improve the overall public health of the population in Ningbo, China.
The investigation of the OP induced by the water- and methanol-soluble fractions of PM2.5 was conducted using a comprehensive year-round field campaign. Two acellular assays, namely Dithiothreitol (DTT) and Ascorbic acid (AA), were employed in this investigation. This study took into account diurnal and seasonal variations, focusing on identifying the key factors that drive the OP, such as chemical species and sources. The OP induced by water-soluble fractions of PM2.5 in both DTT and AA assays were higher at nighttime (0.39 ± 0.10 nmol min-1m-3 and 0.31 ± 0.22 nmol min-1m-3, respectively) than during the day (0.25 ± 0.13 nmol min-1m-3 and 0.23 ± 0.09 nmol min-1m-3, respectively). The OP induced by methanol-soluble fractions were consistently higher during nighttime compared to daytime. The higher OP induced by PM2.5 during the night can be explained by the fact that aerosols during this time were more photochemically aged, as evidenced by low BaA/Chr and LMW-PAHs ratios. Moreover, the bivariate polar plots revealed a strong association between nighttime OP and marine/sea spray from the Ningbo-Zhoushan port and East China Sea, which is in line with our findings that nighttime aerosols are more acidic compared to daytime aerosols. The elevated aerosol acidity at night contributed to high levels of OP, as it enhanced the dissolution of trace metals in the aerosols. However, we reached the conclusion that photochemically aged aerosols were the primary contributor to the elevated levels of OP at night, based on the weak-to-moderate correlation found between nighttime OP and trace metals. The OP induced by water- and methanol-soluble fractions of PM2.5 exhibited seasonal variations. The water-soluble fractions in the DTT assay showed higher OP in winter (1.31 ± 0.49 nmol min-1 m-3), followed by summer (1.22 ± 0.19 nmol min-1 m-3), autumn (1.19 ± 0.26 nmol min-1 m-3), and spring (1.00 ± 0.37 nmol min-1 m-3). The higher OP in winter was attributed to a strong correlation with trace metals, specifically Fe and Cu. During the summer, the AA assay detected higher OP in both the water- and methanol-soluble fractions. This was attributed to the photochemical aging process and the high aerosol acidity levels observed in the summer aerosols. The Positive Matrix Factorization (PMF) model was employed to identify six sources of PM2.5 in Ningbo. These sources encompass industrial emissions, biomass burning, secondary aerosols, sea spray/marine emissions, vehicular emissions, and road dust. During the field campaign, the three dominant sources contributing to the mass concentration of PM2.5 were industrial emissions, secondary aerosols, and vehicular emissions. The bivariate polar plots clearly demonstrated that the primary local sources of industrial emissions, which have a significant impact on PM2.5 pollution in Ningbo, are the industrial facilities situated in the Wangchun industrial zone, as well as the Zhenhai and Beilun industrial parks. The plots indicated that vehicular emissions originated from the freeways and highways that surrounded our study area. In addition, the plots demonstrated that the precursor gases of secondary aerosol - namely SO2, NO2, and VOCs - originate from local industrial facilities and vehicular traffic. The analysis of air mass trajectories revealed that long-range transport from Inner Mongolia, the East China Sea, Northern China, and Taiwan contributed to the levels of PM2.5 in Ningbo. To estimate the contribution of six sources to PM2.5-induced OP, this study used Multiple Linear Regression (MLR) analysis. The findings emphasize the variability in source contributions to both PM2.5 mass concentration and OP, depending on the assay and extraction solvent used for OP measurement. Targeting mass concentration alone, as currently done in PM2.5 pollution regulation, would not sufficiently reduce health risks. Thus, future regulatory efforts should incorporate both the source contributions of mass concentration and OP to effectively mitigate risks associated with PM2.5.
The health risk of six size-fractionated PM (ultrafine: ≤0.49, 0.49‒0.95 µm, accumulation: 0.95‒1.5, 1.5‒3 µm, coarse: 3‒7.2, ≥7.2 µm), and PM2.5 and PM10 were compared by modeling their OP deposition in the various regions of the human respiratory tract. This began with a comparative OP assessment of water- and methanol-soluble fractions of these particles in DTT and AA assays. The OP induced by water-and-methanol-soluble PM size fractions measured by DTT assay exhibited unimodal size distribution with peak concentration in accumulation particles (0.95‒1.5 µm). The high OP induced by water-and methanol-soluble fraction of 0.95‒1.5 µm was dominated by industrial emissions and vehicular traffic. In contrast, in AA assay, the OP exhibited trimodal size distribution with the concentration peaked in coarse particles, followed by ultrafine particles, and accumulation particles. The high OP of coarse particles in AA assay was attributed to industrial emissions, vehicular traffic, and marine/sea spray. The OP induced by water- and methanol-soluble fractions of PM10 was consistently higher than that of PM2.5 in both DTT and AA assay. The OP induced by PM10 exhibited a strong correlation with Cu, Fe, WSOC, and quinones. The high concentration of Cu and Fe in the presence of quinones causes the synergistic effect, thereby elevating the OP levels of PM10. The bivariate polar plots revealed vehicular traffic, and road dust are potential sources contributing to the high OP induced by water-and methanol-soluble fractions of PM10. In all OP measurements, the DTT assay consistently demonstrated high OP in the water-soluble fractions, indicating its higher responsiveness to a wide range of water-soluble chemical species, including trace metals, high molecular weight (HMW) quinones, and humic-like substances (HULIS). In contrast, the AA assay exhibited higher OP in the methanol-soluble fractions, suggesting its high sensitivity to water-insoluble chemical species that were extracted in methanol. This study has demonstrated, for the first time, that the DTT assay is the most effective method for assessing the OP of water-soluble fractions, while the AA assay is best suited for evaluating the OP of methanol-soluble fractions. The Multiple-path particle dosimetry (MPPD) model was employed to simulate the deposition of OP in the Extrathoracic (ET), Tracheobronchial (TB), and Pulmonary (PL) regions of the respiratory tract. The patterns of OP deposition in the DTT and AA assays showed similarities. Both assays measured similar levels of OP deposition for ultrafine particles. However, the DTT assay demonstrated higher values for accumulation particles, whereas the AA assay exhibited higher levels for coarse particles. These findings hold significant implications for assessing the health risks associated with ambient particles across various size ranges. Moreover, this study investigated for the first time a potential link between OP and the particle deposition dose in various regions of the respiratory tract. The results indicated that this relationship varies depending on the type of assay used. In the case of the DTT assay, a weak correlation was observed. The DTT assay is sensitive to a broader range of water-soluble chemical species present in particle doses within our study area. Conversely, no correlation was observed in the AA assay, suggesting that the substances causing OP, as detected by AA, may not significantly contribute to the dose of each PM size fraction collected in our study domain. These findings indicated that the particle deposition dose, which depends on the particle concentration in the ambient air and exposure conditions, is not a suitable metric for predicting OP or the potential health risk posed by ambient particles.

Date of Award	15 Mar 2024
Original language	English
Awarding Institution	University of Nottingham
Supervisor	Jun He (Supervisor), Collin E. Snape (Supervisor) & Qingjun Guo (Supervisor)