POLYCYCLIC AROMATIC HYDROCARBON (PAH)

POLYCYCLIC AROMATIC HYDROCARBON (PAH) IN COOKED (CHARRED) RICE: LEVELS, RISK ASSESSMENT AND SOURCE CHARACTERISATION Essumang D. K. Department of Chemistry, University of Cape Coast, Cape Coast. Ghana *Corresponding Author’s E-mail: [email protected]; [email protected] (Submitted on 16th July, 2014 and accepted on 22th December, 2014) Abstract The presence of polycyclic aromatic hydrocarbons (PAHs) as carcinogens in food has been known since the 1950s. Hundreds of individual PAHs may be formed or released as a result of incomplete combustion or pyrolysis of organic matter (heat generated food toxicant). The amount of PAHs present in cooked foods depends on the duration of the cooking process, the temperature of the heat source, the distance of the food material from the heat source and the presence of fat or oil in the food. This research seeks to determine the levels of PAHs introduced into cooked rice (“kanzou” or charred) through some common cooking methods in Ghana using the gas chromatography with flame-ionization detection (GC/FID). The result showed the presence of eighteen PAHs including some carcinogens such as benzo(a)anthracene, benzo(b)fluoranthene dibenz(a,h)anthracene and benzo(e)pyrene in almost all the samples analyzed. The levels detected were relatively higher than accepted by the European Commission but risk analysis showed low levels of risk as compared to U.S EPA accepted risk of 1.0E-06 to 1.0E-04. Source characterization of PAHs showed combustion of oil and rice at an unregulated high temperature as the main source of PAHs in the cooked rice. Key Words: Polycyclic Aromatic Hydrocarbon, kanzou, charred rice, risk assessment, source characterization Introduction The primary objective of food processing operation is to improve the quality of the foodstuff to make them palatable for human consumption. Nevertheless, some processing operations do induce the formation of materials that are potentially toxic and harmful to humans (heat generated food toxicant). Thus, toxic substances may be formed by the interaction between any endogenous and/or exogenous food components or their derivatives or between these substances and outside agents, such as oxygen. Chemical degradation may also occur due to heat, light etc., and other agents and in turn may result in the formation of toxic compounds (Deshpande, 2002). The presence of polycyclic aromatic hydrocarbons (PAHs) as carcinogens in food has been known since 1950s. These chemicals have also been found in uncooked vegetable oils, charred meat, flour, bread, and marine life in contaminated waters (Furihata and Matsushima, 1986; Simko, 2002). At least, about 18 mutagenic and/or carcinogenic PAHs have also been found in uncooked vegetables, fruits, cereals, and vegetable oils (Concon, 1988; Deshpande, 2002; Alomirah, 2010). The amount of PAH present in cooked foods depends on the duration of cooking, the distance of materials from the heat source, whether the melted fat is allowed to drop into the heat source or hot surfaces, etc. The amounts of PAHs in foods vary from 0 to 400µg/kg (Furihata and Matsushima, 1986). Essumang, D. K. (2014). Polycyclic aromatic hydrocarbon (pah) in cooked (charred) rice: levels, risk assessment and source characterization. Journal of Basic & Applied Sciences, 1 (2), 63-83

Polycyclic aromatic hydrocarbon contamination of food can originate from different sources among which the sources of major importance are the environment and food processing operations. Processing procedures, such as smoking, drying roasting, baking, frying, barbecuing/grilling and cooking of food are a recognized major source of contamination of PAH. Data reported in the literature on PAH in smoked foods are highly variable. The main reason for such discrepancies is the differences in the procedures used for smoking. The content of 12 PAHs in smoked fishery products from modern smoking kilns with external smoke generation and procedures that remove high-boiling compounds such as PAH and particles potentially containing PAH have been compared with products from traditional smoking kilns where the smoke is generated in direct contact with the product. The average benzo (a) pyrene concentration determined for the traditional kilns was 1.2µg/kg and 0.1µg/kg for the modern kills (Codex Agenda Item 17h). Levels of individual PAHs, as high as 200µg/kg in food, have been found in smoked fish and meat. PAH formation during charcoal grilling is shown to be dependent upon the fat content of the meat, the duration of cooking and the temperature used. For example a heavily barbecued lamb sausage contained 14µg/kg of the sum of six PAH, considered by European Union Scientific Committee on Food (EUSCF, 2002) to be carcinogenic and mutagenic. When foods are subjected to high temperatures (>200 – 300oC), pyrogenic compounds are produced. Heating of fats and oils to high temperatures can result in oxidation and polymerization reactions, the products may be harmful to humans. The formation of PAH is only significant at high temperatures. Temperatures between 400 and 1000 oC produces significant amount of PAHs (Toth and Pothast, 1984; Concon, 1988; Deshpande, 2002). At this temperature, a significant amount of charred or tarry products is formed. They are most typically formed from pyrolysis of fats and oils at temperatures above 400oC. This can occur if portions of the food encounter very hot surfaces. Benzo(a)pyrene (3,4benzpyrene) has been identified as the active component of charred material with benzo(b)fluoranthene, dibenz(a,h)anthracene, dibenz(a,i)pyrene and dibenz(a,h)pyrene occurring at various levels (Badger, 1962; Deshpande, 2002). Health effects due to PAHs and other contaminant in food has been discussed extensively in recent times (Shen, et al. 2008) which include growth retardation, low birth weight, small head circumference, low IQ, damaged DNA, in unborn children and disruption of endocrine systems, such as estrogen, thyroid and steroids. Skin changes (thickening, darkening, and pimples) and reproductive related effects such as early menopause due to destruction of ovum and cancers (breast cancer) have been identified (Davis, et al., 1987; Boström et al., 2002; ATSDR, 2007; Gray, 2008; Shen, et al. 2008) Humans are basically exposed to PAHs through a) the respiratory tract by inhalation of PAHcontaining matter such as cigarette smoke, vehicle exhaust, PAH contaminated air emitted from certain industries or by the burning of wood for heating, etc., b) the digestive tract following intake of PAH-containing foodstuffs (e.g., fried and charcoal-grilled meat) and PAH contaminated vegetables and crops grown close to areas with intense traffic, etc., and c) the skin following contact with substances such as petroleum products (e.g., soot, pitch, and tars) (Boström et al., 2002). The emphasis of the digestive tract PAH exposure to humans over the years has mostly focused on meat and smoked fish products (Furihata and Matsushima, 1986). However, the method of certain food preparations may induce PAHs into the food (pyrolysis of the food) as a result of cooking at

Essumang, D. K. (2014). Polycyclic aromatic hydrocarbon (pah) in cooked (charred) rice: levels, risk assessment and source characterization. Journal of Basic & Applied Sciences, 1 (2), 63-83

high temperatures above 400 oC (Toth and Pothast, 1984; Concon, 1988; Deshpande, 2002). An example is the cooking of rice, a staple food in Ghana. “Jollof - rice” one of the most popular rice foods made from rice, tomatoes, meat or fish and oil (Wilson, 1972) is usually cooked at very high (unregulated) temperature and that result in part becoming charred (at the base of the saucepan). This portion of rice is known in Ghana as “kanzou” or “under” and because it is cheap it is heavily patronized by most people. This research seeks to determine the levels of PAHs introduced into this portion of the cooked rice through some common cooking methods in Accra, Ghana and to assess possible cancer and non cancer risk. Materials and Method Sampling Ten (10) samples each were collected from popular cooked rice vendors in Accra, Cape Coast, Takoradi and Agona Swedru and composited. About 500g of rice samples were taken in aluminum foil and stored in a refrigerator. Four samples comprising uncooked, cooked rice, charred (burnt part) cooked plain and jollof rice of same brand was prepared using “non-stick cooking utensil”. These four samples were also collected in aluminum foil, stored in a refrigerator and used as control samples (Grimmer, et al 1975). The table in Appendix I give the samples and their codes. Procedure About 20 g of the thawed sample from the refrigerator was homogenized and weighed into a labeled round bottom flask. About 0.1mL of an internal standard (2,2-di-naphthyl and 3,6dimethylphenonthrene mixture) was added to the weighed sample in a flask. Also, 100 mL of 2M methanolic-KOH was then added and saponified for 4 hours. After saponification, the condenser was flushed with 25 mL of 2M methanolic KOH into the sample (Kazerouni 2001). The solution was cooled and transferred into a 500 mL separating funnel with the aid of a glass funnel fitted with a glasswool. 25 mL of methanol-water (4:1) was added to the residue in the flask, swirled and added to the solution in the separator. About 40 mL cyclohexane was added to the filtrate in the separator, mixed gently and layers were allowed to separate. The lower layer was collected into a second separating funnel and the extraction process repeated with another 40 mL cyclohexane. The lower KOH solution was set aside and the cyclohexane extracts were combined in the first separator. 40 mL methanol: water (1 + 1) was added to the cyclohexane extracts; gently shook to wash and the layers were allowed to separate. The lower layer was drained and discarded later. About 80 mL N, N-Dimethylformamide (DMF)-Water was added to the cyclohexane extract shook gently and layers allowed separating. The lower DMF-water layer was collected into a second separating funnel. Extraction with cyclohexane was repeated with another 40 mL DMF-water, allowed to separate, the lower DMF-water extract drained and added to previous extract 120 mL of water and 80ml cyclohexane were added to the combined DMF-water extract, shook gently and layers allowed to separate with the lower DMF-water as waste. The cyclohexane extract was released onto an anhydrous sodium sulphate (previously dried at 550 oC for an hour) in a glass funnel plugged with glass wool. An amount of dichloromethane (DCM) was used to rinse the sodium sulphate and the filtrate added to the extract in a Turbo Vap II tubes. A drop of iso-octane was added to the extract as a keeper. Nitrogen gas was used to evaporate the extract at Essumang, D. K. (2014). Polycyclic aromatic hydrocarbon (pah) in cooked (charred) rice: levels, risk assessment and source characterization. Journal of Basic & Applied Sciences, 1 (2), 63-83

30oC to a volume of about 0.5 mL. The bond elute column was then conditioned with 3.0 mL methanol followed by 3.0 mL DCM-Pentane solution. The extract was made to pass through the bond elute column drop wise and eluted with 0.5 mL DCM-Pentane solution. Then the column was filled with 3 mL DCM-Pentane solution. The elution was repeated with the solvent. A drop of iso-octane was added to the eluted solution and nitrogen gas was used to reduce the volume to 0.5 mL at 30 oC. The reduced solution was quantitatively transferred into a vial and stored for GC-FID analysis. Instrumental Analysis The final extracts, quality control and blanks were automatically injected on the gas chromatograph in a splittless mode. The conditions of the chromatographic system were as follows: Column: Agilent 19091S– 436, HP-5MS, 60m x 250µm.x 0.25µm, Column head pressure: 24.87 psi, Column flow: 1.4 mL/min, injection volume used was 1.0 µL, Make-up flow: 45.0 mL/min., H2 flow: 40.0ml/min, Air flow: 300 mL/min, Purge flow: 75.0 mL/min., Purge time: 1.0 min, Oven temperature programme: initial temperature 60oC, hold 2 min, rate 12oC/min to 250oC, rate , A 3oC/min to 310oC, hold for 10 min, Injector temperatures 280oC, Detector, temperature: 350oC, Volume injected: 2 µL Gas chromatography analysis separates all of the components in a sample and provides a representative spectral output. The sample vials containing the extracts were arranged on a platform at the injection chamber and the injection of 2 µL was automatically done by the instrument. Quality Control Approximately 0.04 mL of a Certified Reference Material (CRM) 2260 (NIST) of concentration 60 µg/mL was added to 100 mL of 2M methanolic – KOH solution in a round bottom flask and treated as was done for the samples. The final concentrate of 0.5 mL was expected to have a concentration of 5µg/mL of the standard added. This was done to ensure that the method was effective and reliable. Table 1: Percentage Recovery Using NIST Certified Reference Material Compound Name

Average µg/mL

True Value µg/mL

Relative Error (%)

% Recovery

Naphthalene 2- Methylnaphthalene 1-Methylphthalene Biphenyl 2,6-Dimethylnaphthalene Acenaphthylene Acenaphthene 2,3,5-Trimethylnaphthalen Fluorene Phenanthrene Anthracene 1-Methylphenanthrene

14.2 14.8 14.7 14.9 14.8 14.8 14.8 14.7 14.8 14.8 14.9 14.7

15.0 15.0 15.0 15.0 15.0 15.0 15.0 15.0 15.0 15.0 15.0 15.0

-5.3 -1.3 -2.0 -0.7 -1.3 -1.3 -1.3 -2.0 -1.3 -1.3 -0.7 -2.0

95 99 98 99 99 99 99 98 99 99 99 98

IS 3,6-DMP

10.0

10.0

0.0

100


Fluoranthene Pyrene Benzo(a)anthracene Chrysene IS BB-Biphenyl Benzo(b)fluoranthene Benzo(k)fluoranthene Benzo(e)pyrene

14.8 14.2 14.5 14.5 10.0 15.2 14.1 14.5

15.0 15.0 15.0 15.0 10.0 15.0 15.0 15.0

-1.3 -5.3 -3.3 -3.3 0.0 1.3 -6.0 -3.3

99 95 97 97 100 101 94 97

Benzo(a)pyrene Perylene Indeno(1,2,3cd)pyrene Dibenz(a,h)anthracene Benzo(g,h,i)perylene

14.5 14.6 14.8 14.1 14.3

15.0 15.0 15.0 15.0 15.0

-3.3 -2.7 -1.3 -6.0 -4.7

97 97 99 94 95

Results from the NIST reference material showed high recovery of PAH values that ranged from 94% to 101% with an average PAH recovery of 98%. The use of these recovery results which were from certified NIST reference material was used to assess the efficiency of extraction of PAH from the charred rice sample matrix is highly contentious since the two are of different matrices. However, the values could be used to establish the reliability of the extraction system as well as the efficiency of the GC/FID instrument since that was the only CRM available at the time of the analysis. Calculation of Carcinogenic Risk Human health evaluation computerized software-RISC 4.02 (USEPA, 1989) was used in the evaluation of the cancer and non–cancer risk assessment. Carcinogenic risks are estimated as the incremental probability of an individual developing cancer over a lifetime as a result of exposure to the potential carcinogen. This risk is referred to as the individual excess lifetime cancer risk (IELCR) or just carcinogenic risk. Published values of chemical carcinogenic toxicity (slope factor) are used to calculate risk from the Lifetime Average Daily Dose (LADD): IELCRij = SFij LADDij

(1)

Where IELCRij = individual excess lifetime cancer risk for chemical i exposure route i [dimensionless], SFij = slope factor for chemical i exposure route j [mg/kg-d]-1, LADDij = lifetime average daily dose for chemical i exposure route j [mg/kg-d]. Calculation of Hazard Index Human health evaluation computerized software-RISC 4.02 (USEPA, 1989) was used in the evaluation of the cancer and non–cancer risk assessment. The potential for non-carcinogenic effects was evaluated by comparing an exposure level over the exposure duration (maximum of 70 years) with a reference dose derived for a similar exposure period. This ratio of exposure to toxicity for an


individual pathway and chemical is called a hazard quotient. The hazard quotients are usually added across all chemicals and routes to estimate the hazard index. Some, however, will argue that it is more appropriate to only sum the hazard quotients for chemicals that affect the same target organ (e .g. liver or blood). The non-cancer hazard quotient assumes that there is a level of exposure below which it is unlikely that even sensitive populations would experience adverse health effects (USEPA, 1989).

Results and Discussion The precision and suitability of the method to the measuring equipment used, was initially established using the certified reference material. This was done by using the reference material as a sample and treated the same way as the real samples. The percentage recoveries were then calculated. The method verification and sample results are tabulated below (table 1) above. PAH Concentration in Rice The rice used as a control in this research work showed various levels of PAH. Notable among them were 2-Methylnaphthalene with concentration 0.16 mg/kg, 2, 3, 5 – Trimethylnaphthalene with 0.37 mg/kg, pyrene with 0.24 mg/kg, benzo (b) fluoranthene with 0.66 mg/kg and dibenz (a, h) anthracene with 2.54 mg/kg concentrations. Those that showed high values which included Benzo (b) fluoranthene concentration almost doubled from 0.66 mg/kg in C1 to 1.2 mg/kg in C2 (table 3). This may be due to the cooking temperature. Dibenz (a,h) anthracene was however not detected in C2, though it was detected in C1, this may either be due to pretreatment of rice before cooking or temperature differences during cooking. The control, C3 (table 3) was partially burnt rice which was cooked in a “non-stick” cooking utensil. Levels of detection followed trend as that of C2 but had elevated levels for Benzo (g.h.i) perylene with concentration of 0.58 mg/kg compared to 0.30 mg/kg in C2 (table 3). The level of PAHs in the jollof rice (C4) of the same brand of rice used as control showed a similar trend as that of C3 but recorded 0.28 mg/kg of Dibenz (a,h) anthracene which was not detected in C2 and C3. Possibly this may be due to the vegetable oil used in preparing the jollof and the temperature factor (Alomirah, 2010). Two PAH classified as a strong carcinogen, namely benzo (b) fluoranthene and dibenz (a, h) fluoranthene were detected in the uncooked control sample (C1). This confirms similar work by Liu and Korenega (2001) which demonstrate levels ranging from 0.05 to 0.08 mg/kg dry weight in unpolished and polished rice. Liu and Korenega (2001) concluded that PAHs in polished rice was far lower than the unpolished rice which was due according to them to the purification stages (Liu and Korenega 2001; Kazerouni 2001). In this study, the average distribution of PAH in the rice samples from vendors ranged from the lowest of none detected for chrysene, biphenyl, benzo(k)fluoranthene, benzo(a)pyrene, perylene, indeno (1,2,3-cd)pyrene to the highest 4.2 mg/kg of 2-methylnaphthalene. The mean results from


table 2 shows that, 2-methylnaphthalene and benzo (g,h,i) perylene are the most predominant PAHs in the charred rice (fig 1). Mean Levels of PAHs in Control (uncooked and cooked rice) and Charred rice from Vendors 4.5 4

Concentration in mg/kg

3.5

Control - Uncooked Control - Charred Vendors - charred

3 2.5 2 1.5 1 0.5

2-

M Na et ph 1- hyln tha M le a et p n hy hth e ln a ap len 2, 6ht e ha D im le et ne B hy ln iphe a Ac pht nyl en ha 2, 3, ap len 5e A ht Tr im ce hyle et na n hy ph e ln t ap hen ht e ha le n F Ph luo e en re n a 1nt e M A hr et hy nth ene lp he rac na en nt e Fl h uo re ra ne nt Be he nz ne o( A) Py A n re th ne Be ra nz ce o( Be B) Chr ne F nz lu yse o( o K) ran ne Fl uo the Be ra ne nz nth o e Be (E) ne nz Py re o( In A) ne de Py no re (1 ne D ,2 Pe ib ry en ,3 z( -Cd len Be A,H )P e nz )A yre n n o( G thra e ,H ,I) cen Pe e ry le ne

0

Types of PAHs

Fig. 1: Mean levels of control (Uncooked and cooked) rice and samples from Vendors Despite the fact that the concentrations of PAH from the various samples are comparable, some of the PAH were more predominant with some specific rice samples though some others either decreased or increased across the vendors (table 3). The least total PAH concentration of 2.8 mg/kg (S6) which is less charred, whereas the highest total PAH concentration of 12.9 mg/kg (S10) which was totally charred from table 3. A close observation of the above results shows that the use of very high temperature resulted in high PAH levels. Compounds with PAH molecular weights ranging from 252–276 (i.e. Benzo (e) pyrene, benzo (b) fluoranthene, perylene, dibenz(a,h)anthracene, 2,3,5trimethylnaphthalene and benzo(g,h,i) perylene) apart from 2-methylnaphthalene recorded the highest levels (Badger, 1962; Deshpande, 2002; Kanchanamayoon and Tatrahun, 2009)

Sample S1, a jollof rice, sourced from a sales point in Teshie contained most of the PAH under investigation. Two of the strong carcinogens, namely benzo (b) fluoranthene and dibenz (a, h) anthracene had values 0.92 mg/kg and 1.95 mg/kg respectively (table 3). The range of PAH concentration was between 2, 3, 5trimethynaphthalene (1.03 mg/kg) > benzo (g, h i) pyrylene (0.87 mg/kg) >dibenz (a, h) anthracene (0.87 mg/kg) etc. (table3). Sample S3 also recorded similar trend of concentration ranging from