Health and carcinogenic risk evaluation for cohorts exposed to PAHs in petrochemical workplaces in Rawalpindi city (Pakistan)

This study presents the analyses of urinary biomarkers (1-OHPyr, α- and β-naphthols) of polycyclic aromatic hydrocarbons (PAHs) exposure and biomarkers of effect (i.e. blood parameters) in petroleum-refinery workers (RFs) and auto-repair workers (MCs). Exposed subjects had higher concentrations of white blood cell (WBC) count than control subjects (CN) subjects (5.31 × 103 μL−1 in exposed vs. 5.15 × 103 μL−1 in CN subjects), while the biomarker of oxidative DNA damage (8-OHdG) was significantly higher in MCs. The exposure among these two cohorts could be influenced by the ambience of the workplaces; in fact, MCs’ shops are relatively damp and enclosed workplaces in comparison with the indoor environment of refineries. PAHs in the dust samples from mechanical workshops probably originated from mixed sources (traffic exhaust and petroleum spills), while the incremental lifetime cancer risk (ILCR) for MCs showed moderate-to-low cancer risk from exposure to dust-bound PAHs. The study shows that increasing PAH exposure can be traced in MC workstations and needs to be investigated for the safety of public health.


Introduction
Polycyclic aromatic hydrocarbons (PAHs), which are well-known mutagens, teratogens (Kennish 1997), and endocrine disrupting chemicals (Augulyte et al. 2009), are primarily formed during the incomplete combustion of organic materials, and are also released during industrial activities. PAHs have also been detected at low levels in cigarette smoke motor vehicle emissions, and in some foods such as char-broiled meat, vegetables, fruit, and cereals. Occupational exposure to PAHs may also occur in workers breathing in exhaust fumes such as auto-repair mechanics, street vendors, motor vehicle drivers as well as those involved in mining, metal working, and oil refining. If exposed to PAHs, the harmful effects that may occur largely depend on the way people are exposed. Various studies on workers that breathed in or had a long-term contact with PAHs have suggested that PAHs may cause not only breathing problems, chest pain, irritation, and coughing lung but also kidney, pancreatic, prostate, or skin cancer.
According to the International Agency for Research on Cancer (IARC), some PAHs are carcinogenic to humans. In particular, occupational exposure to PAHs may occur in workers continuously inhaling the fumes of used gasoline engine oil (UGEO) and volatile fraction (mist), such as auto-repair workers (Kamal et al. 2011). Grimmer et al. (1982) showed that carcinogenic contents are present in both used and virgin engine oils, and in fact, they evidenced that UGEO induces local tumor after prolonged application in mice, while PAHs containing more than three rings were responsible for more than 70 % of carcinogenicity of UGEO. Long-term petroleum or crude oil exposure is also related to other human health effects in addition to tumors (Malins & Ostrander 1994), such as blood disorders (Yamato et al. 1996), reproductive problems (Eisler 1987), nephrotoxicity (Vyskocil & Cizkova 1996;Ezejiofor et al. 2014), reduced growth (Oliver et al. 1993), and morphological abnormalities (Kennish 1997).
In developing countries such as Pakistan, there is lack of awareness among workers regarding workplace chemicals and other related hazards. In fact, there is a large number of workers in Pakistan who are self-employed and work as auto-mechanics, and do not use personal-protective equipment to minimize the risk of skin cancer and respiratory tract diseases. Some of these workers deal with auto-spare parts; they work in unhygienic conditions, and they come into contact with UGEO on daily basis. This study aimed to conduct a comparative analysis of PAHs exposure among auto-repair station and petrol-refinery workers and to evaluate impact of occupational exposure to petrochemicals on selected biochemical parameters. Moreover, the carcinogenic risk from ingestion, inhalation, and ingestion of dust-bound PAHs was also evaluated.

Study area and hot spots description
Rawalpindi is a rapidly growing city in the Pothohar region of northern Punjab (Pakistan), located only 14 km south from the capital city of Islamabad and 275 km to the North West of Lahore city. The growth rate of the urban areas of Rawalpindi city is 3.39 % (Shabbir & Ahmad 2010), and due to the low literacy rate in the country and increasing unemployment, the trend of self-employment is often followed by young generation. There are several mismanaged and self-established workplaces (i.e. auto-mechanic and spare part shops) on main roads, amid narrow streets, and residential areas. These hot spots not only release large quantity of untreated petrochemical residues into the environment, but they also pose health risk to the surrounding population. These hot spots are mainly located in the Murree road and its surrounding areas in Rawalpindi; thus, exposed human subjects and workplace dust samples were respectively selected and collected in the area (see Plate S1-Supplementry material S1).

Selection of subjects and their self-certification of health status
Sampling was done at random between car mechanics (MCs, n = 25), including diesel engine mechanics, motor-truck mechanics, engine-repair mechanics, differential and brakes repairers, general automobile service-station mechanics and refinery workers (RFs, n = 30) located in the district Rawalpindi. Participants for an age-matched nonoccupationally exposed group (CNs, n = 34) were selected from non-chemical-related occupations of the same city.
All participants filled a short questionnaire to provide confidential information on their occupation. The information included the work/exposure hour/day (h/d), work experience, and types of chemicals used in the workplace. Moreover, the health and socio-demographic status, such as age, height, body mass index (BMI), education, history of known medical problems, and lifestyle factors, including frequency and amount of smoking were also documented. The inclusion criteria for workers were that the subject should be in good health, non-smoker, do not take any kind of medication, work or be exposed 6 h/d and six d/week, and have a working experience of ≥5 years. Same criteria (except that they were non-occupationally exposed) were used for the recruitment of the CN group. All study subjects were recruited with informed written consent under an approved protocol from the ethical review committee of the Quaid-I-Azam University Islamabad (Pakistan). The subjects, who were former smoker or recently recruited staff/worker, those on some kind of medication, and having the extreme body mass index (BMI < 17 and > 30 kg/m 2 ), were excluded from the analyses.
The subjects were also asked for the prevalence of any respiratory symptoms. A respiratory disturbance was characterized by having one or more of the following symptoms: coughing, sneezing, fatigue, upper respiratory congestion, and rhinitis, while headache symptoms included dizziness, fatigue, and headache. Subjects experiencing these symptoms only once in a month were ranked as no or low, with more than once in two weeks were ranked as medium, while with more than one symptom in a week were ranked as high. Similarly, headache and other work-related health symptoms (if any) reported by subjects were also documented.

Reagents
All solvents were pesticide grade and purchased by Supelco (Bellefonte, PA, USA) and tested for contaminants before use. Standard PAHs mixture (EPA) was commercially purchased from Supelco (Bellefonte, PA, USA). Benzo(e)pyrene and coronene were purchased from Alltech (Deerfield, USA). Silica (100-200 mesh) and sodium sulfate (Na 2 SO 4 ) were purchased from Merck (Darmstadt, Germany). Sodium sulfate was heated for 12 h at 450°C to remove any organic matter and kept at 120°C until use.

Blood sampling
Post-shift blood samples of workers and non-occupationally exposed subjects were collected on the same day during sampling. From each subject, 3-4 ml blood was collected in disposable vacutainors tubes (without anticoagulant for serum and with EDTA for whole blood). The blood samples were always withdrawn by a trained technician. The blood specimen was immediately shipped to the analytical laboratory and were kept refrigerated until analysis, except for the analysis of hematologic parameters and superoxide dismutase (SOD) activity, which was performed the same day. The gelvacutainors were centrifuged for 20 min to separate serum from blood cells within one hour after blood collection. The separated serum samples were transferred to Eppendorftubes with an identification code and kept refrigerated. All biological samples were sent to an accredited bio-medical chemistry laboratory for selected biochemical analysis after proper labeling/tagging each sample with accurate information.

Urinary 8-OHdG assay
Before the analyses of 8-OHdG, the thawed urine samples were centrifuged at 3000 g for 5 min to collect the supernatant, which was further diluted with the diluents provided with the kit and vortexed to mix well. The solution was re-centrifuged, to collect supernatant which was further used for the quantification of 8-OHdG, according to the instruction provided by manufacturer's of competitive enzyme-linked immunosorbent assay kit (ELISA; Cell Bio-labs Inc. San Diego CA). A logarithmic standard curve was used to interpolate the results; the results were further adjusted with urinary creatinine values and presented as 8-OHdG ng mg-Cr −1 .
Dust sample collection, preparation, extraction, cleanup, and analysis A total of 19 soil/dust samples were collected from automobile repair stations/small shops located on the main roads and slum areas of Rawalpindi city, using dustpans and plastic brushes (new pair for each site) by gentle sweeping motion to collect fine particulates. Collection of pure dust was often not possible, and instead, the samples comprised dust particles with small factions of soil in most of the cases. All the shops falling within an area of 5 m were considered a single sampling site, and soil/dust samples from about five different work areas were combined into one composite sample. Soil/dust samples were collected in refinery (n = 17) belonged to seven different location of same refinery; each sample was a composite of three soil/dust samples from the same site collected in three consecutive days. In order to protect all samples from sunlight exposure, aluminum foil was used to wrap each soil/dust sample separately. After that, the samples were sealed in zip-locked polyethylene bags and stored until analysis.
The details of sample preparation and instrumental analysis have been described in detail previously (Martellini et al. 2012;Kamal et al. 2014). All the prepared samples were analyzed using a Hewlett-Packard 6890 gas chromatography-mass spectrometer (GC-MS), equipped with a 5973 MSD and a HP-5MS capillary column (J&W Scientific, Folsom, CA, USA, 30 m, 0.25 mm I.D., 0.25 mm film thickness). Compound identification was based on the MSD database (NIST, 98) and GC retaining time of PAHs standards. The MSD was operated in selected ion monitoring mode (SIM).

ILCR assessment and model and parameters
The incremental lifetime cancer risk (ILCR), which represents the estimated increase in the lifetime cancer risk due to exposure to a particular carcinogen over 70 years of life time (Meiners & Yandle 1995), was calculated using the following equations for ingestion, dermal contact, and inhalation pathways, where ΣTEQ = sum of toxic equivalent concentrations, that is, the sum of BaP equivalent concentrations for each PAH calculated by multiplying the individual PAH concentration by its corresponding TEF value (Nisbet & LaGoy 1992;Fang et al. 2006;Cincinelli et al. 2007); IngR = ingestion rate; CSF = carcinogenic slope factor, which was based on the cancer-causing ability of benzo(a)pyrene; EF = outdoor exposure frequency; ED = outdoor exposure duration; BW = body weight (kg); AT = average life time; CF = conversion factor; PEF = particle emission factor for BaP; AT (h) = average life time; SA = workers exposed skin area; AF (soil) = dermal adherence factors; ABS = fraction of contaminant absorbed dermally from soil (unitless and specific to contaminant).

Statistical analyses
Data distribution was tested using Kolmogorov-Smirnov (K-S) test. The comparative analysis of urinary PAHs biomarkers (being non-normally distributed) was performed using the Mann-Whitney's U-test, whereas, the independent sample t-test was used for comparative analysis of demographic parameters including age, BMI weight, height, and daily work hour, the data are presented in median, mean ± standard deviation, and minimum and maximum values, while the categorical/nominal data are tabulated as frequencies (percentage). The comparison of nominal/ordinal parameters was performed using Chi-square analysis. All other statistical analyses were performed using SPSS 20 and Ms-Excel 2010. Urine dilution was corrected by adjusting urinary biomarkers with urinary creatinine, and values were expressed in μmol/mol-creatinine (μmol mol-Cr −1 ).

Results and discussion
The occupationally exposed and non-exposed groups recruited in this study were age matched, with almost similar work experience and BMI. Table 1 summarizes the sociodemographic information of the subjects, with the comparative analyses revealing long working hours and low life status of MC subjects as compared to CN (p < 0.05). However, the RFs workers had the highest educational background as compared to CNs and MCs (who are self-employed), showing that in the refinery sector, generally technical staff with sound educational background is recruited. On the contrary, MCs do not have a technical education in their work sector, because they have acquired their expertise working as apprentices since their childhood.
Environmental pollution is considered one of main cause of muscular problems and headache symptoms (Pick et al. 2002), and there are some additives in the gasoline which are known to cause neuropathies and psychosis (Epstein & Selber 2002). Taking into account these aspects, the self-reported health symptoms, including headache and respiratory disturbances, were examined in connection with exposure to petroleum mist in the work environment. Naphthalene, which is present at high concentrations in the petroleum-contaminated workplaces (Kamal et al. 2011Klasing & Brodberg 2013), causes numerous respiratory effects, especially damages in ciliated and Clara cells of the bronchiolar epithelium have been reported in the mouse model (OEHHA 2001). Headache symptoms were more experienced by RFs and MCs than CN subjects, while a more positive response for respiratory effects was reported by RFs.

A comparative overview of biomarkers of exposure and effect
In this study, we analyzed biomarkers of PAHs exposure and effect to evaluate exposure outcomes and observed different results in biochemical parameters of the subjects compared with results obtained in a previous investigation on exposure of workers to pyrogenic sources of PAHs (Kamal et al. 2014). In particular, most of the comparison between CN and exposure groups was non-significant. It was generally observed that the DNA damage (as indicated by urinary 8-OHdG concentration) was quite high in both groups exposed to petroleum products/fumes in occupational environments. Moreover, MCV increased more in MCs than RFs (94.6 fL than 92.5 fL, respectively). Similarly, packed cell volume (PCV) was also higher in MCs than RFs (47.8 % in MCs and 45.4 % in RFs). Overall hematocrit of exposed subjects was higher than CNs (46.5 % vs. 45 % respectively, p < 0.05). Both these symptoms indicated significant Table 1. Comparative analyses of demographic parameters, of exposed (RFs and MCs) and control (CN) groups.
The exposure of RFs to PAHs is rarely reported in the past literature. Even if it is expected that new oil should contain lower content of PAHs, huge exposure to volatile organic compounds (evidenced from high petroleum odor in workplaces) is expected in refineries. The petrogenic sources are known to emit lower molecular weight (MW) PAHs like naphthalene (Klasing & Brodberg 2013), as well as refined petroleum products (Granella & Clonfero 1991). Therefore, αand β-naphthols, which are representative of volatile and gas phase (inhalable) PAHs at low levels (Kim et al. 1999), were also analyzed. Moreover, the combustion of fuel, vehicular emission, and oil refining are all sources of PAHs emissions (Baek 1991). The PAHs which are semi-volatile and have low MW (≤ 206), such as pyrene, can emerge from incomplete combustion of organic materials and also from petrogenic sources, including evaporation of petroleum products and leakages of such oil (Zhang & Tao 2009;Ma et al. 2010). The results provide an indication that MCs were more exposed to PAHs than RFs. This was also evidenced from the higher concentration of urinary 1-OHPyr (median 1.02 μmol mol-Cr −1 ) in the exposed group than in CN subjects (0.62 μmol mol-Cr −1 ); 1-OHPyr was also high in RFs, but the difference vs. MCs was non-significant. The difference of exposure among these two cohorts seems to be influenced by the ambience of the workplace; in fact, it was noticed that MC-workshops are relatively damp and enclosed places as compared to the environment of refineries, where the workers have to perform the routine tasks in an indoor environment for most of the time. MCs come in contact with body parts of automobile engines and handling tools, and they smear most of the body parts with petrochemical fractions. Ventilation in the workplace is also a key factor that influences the indoor concentration significantly (Kamal et al. 2011). 1-OHPyr is a very useful biomarker of exposure to air born mixture of PAHs and has been used in numerous biomonitoring studies in the past. The survey of such workplaces has shown that the MCs work in a highly hazardous situation, that is, they do not use self-protective equipment and their hands are routinely smeared with UGEO, which facilitates cutaneous absorption of PAHs. A previous study from same country has reported that MCs are exposed to volatile PAHs in their workplace and the main exposure among MCs occurs through UGEO on daily basis (Kamal et al. 2011;. In both exposed groups, we did not observe much effect of occupational exposure on blood parameters and any sound association of any blood parameter and PAHs biomarkers. Despite the PCV (47.8 %), RBC (5.42 × 10 6 μL −1 ) and WBC counts (5.4 × 10 3 μL −1 ) were exclusively higher in MCs than CN subjects (Table 2). In general, the exposed subjects had higher level of 8-OHdG; in particular, it was significantly higher in MCs (48 ng mg-Cr −1 ) than CNs (25.8 ng mg-Cr −1 ) (p < 0.001). All other parameters were not so much different among groups, and despite relatively increased blood parameters in MCs, the values were within the normal ranges. Since the risk of exposure was also high among MCs, they work without self-protective equipment , and therefore, they are highly exposed to genotoxins present UGEO, including petroleum residues and carcinogenic PAHs which are accumulated in the lubricating oil over time. Table 2. Comparative analyses of biochemical parameters, among exposed (MCs and RFs) and control (CN) groups.

Min-Max
*TEF for individual PAHs relative to BaP as reported by Nisbet and LaGoy (1992), except TEF values for Cor and BeP which were adopted from Malcom and Dobson (1994). a P = probability value, significant at α < 0.05.
The diagnostic ratios (Table 3) and the score plot ( Figure 2 and Figure S3) showed dominance of petrogenic sources in RF samples, whereas predominance of petroleum combustion and mixed sources in MC soil/dust samples. The ratios suggest dominance of petrogenic sources and traces of petroleum combustion, however, in MCs soil/dust; the petroleum combustion and other petrogenic sources were major contributors. The Σ456-PAHs/ΣPAHs ratios < 0.40 is an indicator of petrogenic source while it represents petroleum and other biomass combustion burning when it is > 0.5, and in the case of MC soil/dust samples, the ratio shows a mixed PAH source between petrogenic emission and petroleum burning origins, whereas a predominant petroleum source in RFs (Biache et al. 2014 and reference therein).
The ratio BaP/(BeP + BaP) provides information on the photodegradation of PAHs in ambient air, because BaP degrades faster than BeP in the atmosphere. Thus, a ratio value < 0.5 indicates that PAHs have undergone aging and photodegradation, as shown by the ratio of the MC soil/dust samples. The ratio of IP/(IP + BghiP) presumes that PAHs ratios < 0.2 indicate petroleum sources; ratios in the range ≥ 0.2 and ≤ 0.5 as petroleum combustion sources; and > 0.5 as grass, wood, and coal combustion sources (Biache et al. 2014 and reference therein). In the case of both MC and RF soil/dust samples, the ratio has been found ranging between 0.13 and 0.19, which indicated petroleum as a major contributing source. The Ant/(Ant + Phe) ratio presumes that ratios > 0.1 indicate combustion origin, while ratios < 0.1 indicate PAHs source to be of petroleum origin (Yunker et al. 2002;Mannino & Orecchio 2008). This ratio ranged between 0.08 and 0.28 in RFs samples, which indicated mixed petroleum and petroleum combustion signature, and was significantly lower than that found in MC soil/dust samples, and the ratio indicated combustion of petroleum as dominant source. The Fl/(Pyr + Fl) ratio presumes that ratios < 0.4 also indicate petroleum sources, while ratios in the range ≥ 0.4 and ≤ 0.5 indicate petroleum combustion. Ratios < 0.4 for both MC and RFs soil/ dust samples indicated petrogenic source. According to BaP/BghiP ratio, a value < 0.6 indicates non-traffic emission, while > 0.6 may be related with traffic emission. This ratio ranging between 0.2 and 0.88 in MC soil/dust samples showed some signatures of traffic exhaust emissions (Moyo et al. 2013 and reference there in). LMPAHs/HMPAHs ratios also indicated that RFs PAHs were mostly derived from petrogenic sources (i.e. > 1) while that of MC soil/dust samples (< 1) showed prevalence of some traces of pyrogenic sources (Zhang et al. 2008).

Evaluation of ILCR
The ILCR approach was used to estimate PAH exposure risk for adults (18-70 years) and children (5-17 years, considered as child-labors in MC working areas) via ingestion, inhalation, and dermal contact pathways (Table 4). The results showed that dermal contact and ingestion pathways contributed more than inhalation route to PAHs exposure in MCs. ILCR-values of dermal route of exposure ranged from 4.69 × 10 −6 to 9.39 × 10 −4 and from 3.44 × 10 −6 to 6.88 × 10 −4 for adults and children, respectively, whereas the ILCRvalues for ingestion route of exposure varied from 2.53 × 10 −6 to 5.06 × 10 −4 and from 1.85 × 10 −6 to 3.71 × 10 −4 for adults and children, respectively. It is worth noting that children working as apprentices in auto-repair work areas are at high risk of soil/dust-bound PAHs exposure via dermal contact and ingestion routes because their high frequency of hand-to-mouth activities as well as their immune and nervous system vulnerability (Maertens et al. 2008). The total cancer risk due to PAHs exposure from all the routes ranged between 7.22 × 10 −6 and 1.45 × 10 −3 and from 5.29 × 10 −6 to 1.06 × 10 −3 for adults and children, respectively. According to United States Environmental Protection Agency (US EPA) criteria, which reported a ILCR value of 10 −4 and/or greater as moderate-high cancer risk, all ILCR-values reported for RF soil/dust samples (lower than 10 −6 ) suggested no significant risk for any individual or combined exposure pathway risk for RF workers (as proposed by the US EPA (2005).

Conclusions
In this study, we assessed the potential exposure of workers to PAHs in chemical work places via dermal, ingestion, and inhalation pathways. All participants, in particular the self-employed in auto-mechanical workshops, were found to be working without any personal protective equipment and regular use of coveralls and gloves when handling petrochemicals during daily working hours. In comparison with RF workplaces, exposure to aromatic solvents, driven by UGEO and residues of petroleum products, was much more prevalent in the ambience of MCs. In fact, RFs were found to be under low carcinogenic risk. The hazardous occupational environments have been demonstrated to be potential risks to workers' health, causing mild impact on in blood parameters. Exposure to PAHs and other toxic chemicals in the work environments can be reduced considerably by the use of personal protections, work practice controls, and specific regulations.