This article was downloaded by: [University of Newcastle (Australia)] On: 08 September 2014, At: 08:26 Publisher: Taylor & Francis Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK
Food Additives & Contaminants: Part A Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tfac20
Determination of polycyclic aromatic hydrocarbons in Italian milk by HPLC with fluorescence detection a
a
A.M. Girelli , D. Sperati & A.M. Tarola a
b
Department of Chemistry, Sapienza University of Rome, P.le A. Moro 5, I-00185 Rome, Italy
b
Laboratory of Commodity Science, Department Management, Sapienza University of Rome, Via Castro Laurenziano 9, I-00161 Rome, Italy Accepted author version posted online: 13 Jan 2014.Published online: 03 Mar 2014.
To cite this article: A.M. Girelli, D. Sperati & A.M. Tarola (2014) Determination of polycyclic aromatic hydrocarbons in Italian milk by HPLC with fluorescence detection, Food Additives & Contaminants: Part A, 31:4, 703-710, DOI: 10.1080/19440049.2013.878959 To link to this article: http://dx.doi.org/10.1080/19440049.2013.878959
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Food Additives & Contaminants: Part A, 2014 Vol. 31, No. 4, 703–710, http://dx.doi.org/10.1080/19440049.2013.878959
Determination of polycyclic aromatic hydrocarbons in Italian milk by HPLC with fluorescence detection A.M. Girellia*, D. Speratia and A.M. Tarolab a
Department of Chemistry, Sapienza University of Rome, P.le A. Moro 5, I-00185 Rome, Italy; bLaboratory of Commodity Science, Department Management, Sapienza University of Rome, Via Castro Laurenziano 9, I-00161 Rome, Italy
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
(Received 27 September 2013; accepted 18 December 2013) The presence of polycyclic aromatic hydrocarbons (PAHs) in Italian commercial milk samples is reported. The study was carried out on lactating (cow and goat) and plant (rice, soya, oat) milk. The quantitative determination involved liquid– liquid extraction of PAHs, a pre-concentration and determination by HPLC using a fluorescence detector. The recovery of analytes was in the range of 70–115%. The precision of the method was found to be between 6% and 24%. The detection limit ranged from 0.66 to 33.3 µg l–1 corresponding to 0.03–1.66 µg kg–1 milk (wet weight), at a signal-to-noise ratio of 3, depending on the compound. By this procedure, the levels of more volatile PAHs (two to three aromatic rings) were confirmed in 34 commercial milk and three plant milk samples, whereas benzo[a]pyrene was found only in five pasteurised milk samples at a mean concentration of 0.17 µg kg–1 milk. These results provide evidence that PAH levels are influenced by heat treatments and skimming processes of milk production. Keywords: PAHs; cow milk; plant milk; HPLC-FLD
Introduction Polycyclic aromatic hydrocarbons (PAHs) are a group of ubiquitous contaminants with significant toxicity (Agency for Toxic Substances and Disease Registry 2009), potential mutagenic and carcinogenic properties (International Agency for Research on Cancer 2010). Sixteen of them are even included in the list of priority pollutants by the USEPA (2006). They are present in all environmental components and may cause harmful effects near to and distant from their source (Feidt et al. 2000). The principal source of human exposure is the ingestion of contaminated food in which PAHs may accumulate in the lipid phase. Indeed the presence of fused aromatic rings favours migration through the food chain into hydrophobic compartments. Therefore, a primary dietary human food, like milk, which contains a high proportion of lipids (triacylglycerols, phospholipids, fatty acids and sterols), may be contaminated by PAHs. In fact environmental factors related to the rearing system (fodder and potential contaminated soil) and ruminant’s source of exposure (ingestion during grazing, drinking of contaminated water and inhalation of contaminated air) are the main source of PAHs intake from lactating ruminants (Lutz et al. 2006; Lapole et al. 2007). In addition food processing carried out at high temperature can cause the combustion of organic matter with a recombination of free-radicals formed, at first, in low molecular mass PAHs and then in higher molecular mass PAHs (Simoneit 2002). *Corresponding author. Email: [email protected] © 2014 Taylor & Francis
In the case of lactose intolerance by individuals, alternative products like “plant milks” and derivatives such as soy, rice and oat milk are frequently consumed. These products that are lactose-free that resemble milk in both appearance and nutrition. Milk and plant milk could always be considered interesting biomarkers for assessing the exposure to environmental contamination and for human risk evaluation. For this reason we report here a comparison of PAH levels in pasteurised and UHT commercial milks (whole, semi-skimmed and skimmed) and in samples of plant milk which are different in their production processes and their fat content. Pasteurised milk is the most consumed milk in Europe, produced with a mild heat treatment (72°C for a minimum of 15 s) according to a process of high temperature short time (HTST), to eliminate the original bacteria, but not the spores that can germinate after opening the milk box. Ultra-high-temperature treatment (UHT) can destroy both bacteria and spores, since the milk is heated to a temperature of at least 135°C. So any dangerous bacteria are removed (Lewis & Deeth 2008). Semi-skimmed and skimmed milk, which contain, respectively, 1.7% fat and a content between 0% and 0.5% fat, are formed by skimming cream (high-fat) from whole milk. The separation process is generally carried out in a centrifugal separator which is based on the fact that when liquids of different specific gravities resolve around the same centre at the same distance with the same angular velocity, a greater centrifugal force is exerted on the heavier liquid than on the lighter one. Milk can be regarded as two
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
704
A.M. Girelli et al.
liquids of different specific gravities: the serum and the fat. Taking into account that in fresh raw whole milk fat is in the form of natural emulsified globules surrounded by a native stabilising membrane, the milk fat globule membrane (MFGM) and that in milk submitted to severe mechanical treatments (such as in homogenisation or skimming process) and heat treatment, the native MFGM can be damaged. This results in a reduction in milk fat globule size and in the alteration of interfacial and electrostatic properties (Michalski et al. 2002). There are various data in the literature concerning the levels of PAHs in milk samples from different countries, such as Taiwan (Tay-Lung et al. 2010), Kuwait (Husain et al. 1997), Puerto Rico (Andrews et al. 2003), Spain (Aguinaga et al. 2007), France (Grova, Feidt, Crepineau, et al. 2002), Japan (Kishikawa et al. 2003) and also from the Italian region of Calabria (Naccari et al. 2011), and some studies demonstrated the transfer of low-molecular PAHs with fewer than five rings from feed to goats (Grova, Feidt, Laurent, et al. 2002; Costera et al. 2009), cows (Kan et al. 2003), and human milk (Kishikawa & Kuroda 2009). In all these papers milk contamination from PAHs was reported, but it is not clear if differences in concentrations (1 ng kg–1 to 100 μg kg–1 milk) and in PAH distribution can be attributed to the different environmental pollution exposure or to the method of analysis. The aim of this study was to determine the levels of the more extensively studied 16 priority pollutant PAHs (Environmental Protection Agency 2006) in milk samples from Italy. In particular, samples of whole, semi-skimmed and skimmed pasteurised and UHT cow’s milk were examined to evaluate the role of food processing, such as heating and skimming treatments, on PAH formation. Samples of some “plant milk” such as rice, soya and oat were also investigated since they are actually healthy alternatives to cow’s milk.
Table 1.
NA ACL AC FL PHE AN FA PY B[a]A CHR B[b]FA B[k]FA B[a]P DB[ah]A B[ghi]P I[cd]P
Materials and methods Chemicals PAHs pure reference standard solution in CH3CN: MeOH = 9:1 (v:v) (EPA TCL PAH Mix) (concentrations are reported in Table 1) and all 16 standard PAHs (naphthalene; acenaphtylene; acenaphtene; fluorene; phenanthrene; anthracene; fluoranthene; pyrene; benz[a] anthracene; chrysene; benzo[k]fluoranthene; benzo[b] fluoranthene; benzo[a]pyrene; indeno[1,2,3-cd]pyrene; dibenz[ah]anthracene; and benzo[ghi]perylene) were purchased from Supelco (Milan, Italy). Acetonitrile (99.9%), water for HPLC, cyclohexane ACS (>99.5%) Hexane ACS (99%), ethanol (96%) NaOH were obtained from Sigma-Aldrich (Milan, Italy); while pentane was obtained from Carlo Erba (Milan, Italy). ISO discTm filters PTFE 4 mm × 0.45 µm were from Supelco.
HPLC system The HPLC system consisted of a liquid chromatographic system LC module (Jasco PU1580; Milan, Italy), a 7125 injector (Rheodyne, Cotati, CA, USA) with a 20 µl sample loop, a 110 series fluorescence detector (Agilent, Rome, Italy) and an HP Chemstation integrator. Separation of PAHs was achieved by a column C-18 (150 × 4.6 mm ID; particle size 5 µm) (Prosphere 300; Alltech, Rome, Italy) with a gradient elution programme using solvent A (1% CH3CN in aqueous solution) and solvent B (99% CH3CN in aqueous solution). The gradient elution was programmed as follows: 40% B to 50% B linearly (0– 20 min), 50% B to 100% B (20–37 min), 100% B (37– 42 min) and 100% B to 40% B (42–45 min). The flow rate was set at 1 ml min–1, at RT. Under these conditions 16 PAHs could be separated satisfactorily within 45 min. The
Amounts of individual PAH in reference standard (A), diluted 1:1000 (B) and 1:5000 (C) solutions. PAH
Solution A (µg ml–1)
Solution B (µg ml–1)
Solution C (µg ml–1)
Naphtalene Acenaphtylene Acenaphtene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benzo[a]anthracene Chrysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[a]pyrene Dibenzo[ah]anthracene Benzo[ghi]perylene Indeno[1,2,3-cd]pyrene
500 500 1000 100 40 20 50 100 50 50 20 20 50 200 80 50
0.500 0.500 1.009 0.100 0.040 0.020 0.050 0.100 0.050 0.050 0.020 0.020 0.050 0.200 0.080 0.050
0.100 0.100 0.200 0.020 0.008 0.004 0.010 0.020 0.010 0.010 0.004 0.004 0.010 0.040 0.016 0.010
Food Additives & Contaminants: Part A Table 2. PAHs.
Detection wavelength programme for the separation of
Time (min)
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
0.00 20.00 23.50 26.00 28.50 35.60 42.50
λex (nm)
λem (nm)
280 250 254 280 265 280 364
340 360 400 435 390 410 500
detection wavelength programme to determine individual PAHs is shown in Table 2.
Sampling Thirty-seven milk samples comprising 29 cow’s milk, two goat’s milk and three plant milk (rice, soya and oat) were obtained from a local supermarket while three of cow’s milk (one raw, one whole and one semi-skimmed) were provided by a local farm. The 29 samples of cow’s milk were made up of 18 pasteurised milk (seven whole, seven semi-skimmed and four skimmed) and 11 UHT milk (four whole, four semi-skimmed and three skimmed). After the homogenisation with a stainless steel blender, all the samples were kept at –20°C until analysis.
Sample preparation Milk samples (2 g) in a 10-ml reaction vial with a screw cap were saponified with 4.0 ml of NaOH ethanolic solution 0.4 M. The mixture was placed in a water bath at 60°C for 30 min and cooled to RT. Then 2.0 ml of cyclohexane were added and the mixture was vortexed for 5 min. The supernatant layer was recovered and transferred into a vial. The samples were re-extracted, as previously described with 2 ml of cyclohexane. A total of 3.5 ml of the combined extracts was successively taken and evaporated to dryness under nitrogen. The residue was then recovered with 100 µl of acetonitrile and filtered with a PTFE membrane filters of 0.45 µm (Sigma Aldrich, Milan, Italy). An aliquot of 20 µl of this solution was injected into the HPLC system. A reagent blank, constituted by 2 ml of deionised water, was simultaneously performed by the same procedure above described, with each series of samples. For PAH quantifications eight standard working solutions at various concentrations were prepared by appropriate dilution of aliquots of the stock solution in acetonitrile in order to prepare the calibration curve and to calculate the LODs and LOQs for all compounds.
705
Results and discussion Optimisation of extraction conditions In a preliminary study a liquid–liquid extraction of an aqueous standard solution was examined by using n-pentane, n-hexane and cyclohexane as solvents. The standard solution was obtained by diluting 100 µl of solution C containing PAHs in the range 0.02–1 µg ml–1, as indicated in Table 1, to 2 ml of deionised water. The extractability of PAHs, expressed as recovery percentage, were examined with two aliquots of 2 ml of solvent (Figure 1). It appeared that cyclohexane, even if being less volatile than n-hexane and with an evaporation time greater than pentane, has the best extraction efficiency of PAHs, with the exception of naphthalene. Thus, cyclohexane was chosen for further experimentation. Successively, taking into account that the extraction efficiency depends not only on partition coefficients, temperature and ionic force, but also on the extraction volume, a study with a variable volume of cyclohexane (2, 4, 6 ml) was undertaken in triplicate analysis (Figure 2). The different recovery of PAHs was due to the different extraction efficiency and to the different volatility of each compound. Indeed, by increasing the volume of organic phase, an increase in the time of cyclohexane evaporation from extracts and a greater loss of volatile analyte could be obtained. Consequently, PAHs with low molecular mass (two to three rings) (AC and PHE) presented 4 ml as the preferred extraction value but for those with more than four aromatic rings (from FA to I[cd]P) a value of 2 ml was optimal. To avoid any decrease in extractability for ACE and PHE, a value of 4 ml as the total volume was chosen. To optimise successively the number of repeated extractions, a study was undertaken with the standard solution C (Table 1) by varying the number of extractions (one extraction with 4 ml, two extractions with 2 ml and four extractions with 1 ml)
Figure 1. Effect of extraction solvent on the recovery of compounds, expressed as [PAHs]found/[PAHs]spiked (%).
706
A.M. Girelli et al. In the saponification step, an NaOH concentration of 0.4 M was fixed in accordance with Kishikawa et al. (2003). The solvent was 100% EtOH since a decrease in the level of ethanol and an increase of water not influenced PAH extraction but caused an increase in the separation time of the organic layer (data not shown).
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
HPLC-FLD method
Figure 2. PAHs.
The performance of the HPLC-FLD method was evaluated by establishing the quality parameters. These parameters were determined using standard solutions; the results obtained are shown in Table 3. For each compound, the linearity of the calibration curve, obtained with eight calibration points by plotting analyte area versus concentration, was confirmed by Mandel’s fitting text (Mandel 1964) in conjunction with an evaluation of the plot. All the investigated calibration curves were characterised by high coefficients of determination (R2 > 0.99) and with variation coefficients (CV%) < 10% with the exception of indeno[1,2,3-cd]pyrene. LOD and LOQ, defined as the concentration of the analyte that produced a signal-tonoise ratio of 3 and 10, respectively, were higher than 0.20 and 0.66 µg l–1 and were then texted experimentally by spiking blank samples at such levels. It appeared that high values of LOD and LOQ for indeno[1,2,3-cd]pyrene occurred since it presented the weakest fluorescence. In order to verify the accuracy and precision of the analytical procedure, recovery experiments were carried out by adding 100 µl of standard PAHs solutions appropriately diluted to 2 g of a pasteurised milk sample rich in fat (3.6%). For each level, five different spiked samples, as well as unspiked controls, were prepared and each of these samples was analysed in triplicate according to the analytical procedure. Recoveries were calculated from the differences between the spiked and unspiked samples in the total amount of each compound for each level. The mean
Effect of cyclohexane volume on the peak area of
Figure 3. Effect of the number of extractions on the recovery of compounds, expressed as [PAHs]found/[PAHs]spiked (%).
but maintaining fixed the total volume of cyclohexane (Figure 3). The maximum recovery percentage for almost all the PAHs was obtained for two extractions, therefore the number of extractions was set at two. Table 3. Correlation coefficients, linearity ranges, LODs and LOQs. PAHs Naphtalene Acenaphtene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz[a]anthracene Chrysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[a]pyrene Dibenzo[ah]anthracene Benzo[ghi]perylene Indeno[1,2,3-cd]pyrene
R2
Linearity (μg l–1)
LOD (µg l–1)
LOQ (µg l–1)
0.9998 0.9998 0.9509 0.9963 0.9999 0.9930 0.9984 0.9962 0.9970 0.9938 0.9965 0.9948 0.9973 0.9969 0.9944
2.1–500 5.3–300 5.0–30 1.7–40 0.66–20 4.2–15 7.0–100 1.5–30 1.5–50 6.7–20 0.83–20 2.1–50 2.0–200 4.6–80 33–50
0.63 1.60 1.50 0.35 0.20 1.25 2.10 0.45 0.45 2.00 0.25 0.63 0.60 1.40 10
2.10 5.3 5.00 1.70 0.66 4.20 7.0 1.50 1.50 6.70 0.83 2.10 2.00 4.60 33.3
Food Additives & Contaminants: Part A Table 4. Mean recoveries and coefficients of variation (CV) of polycyclic aromatic hydrocarbons (PAHs) added to whole pasteurised milk.
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
PAHs Naphtalene Acenaphtene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz[a]anthracene Chrysene Benzo[b] fluoranthene Benzo[k] fluoranthene Benzo[a]pyrene Dibenzo[ah] anthracene Benzo[ghi]perylene Indeno[1,2,3-cd] pyrene
Spiking level (µg kg–1)
Recovery (%)
CV (%)
1.25 2.50 0.25 0.10 0.05 0.12 0.25 0.12 0.12 0.50
15 43 76 70 100 110 89 60 115 74
25 23 9 7 22 6 23 14 24 10
0.05
97
19
0.12 0.50
88 74
25 13
0.20 0.20
69 80
18 16
recovery percentages and coefficients of variation are shown in Table 4 at the lowest level of PAHs spiked to milk samples. It appeared that the recoveries were related to PAH volatility: low values were observed for the more volatile compounds such as acenaphtene and, above all, naphthalene. This was correlated to their loss during the step of cyclohexane evaporation. In addition to the verification of the absence of any matrix effects, the data obtained for the same pasteurised milk sample by the external standard method (ESM) were compared with those obtained by the standard addition method (SAM). In this last procedure the milk sample was divided in four portions and for each were added known amounts of the standard mixture of PAHs prior to saponification. This procedure was designed to determine the content of an analyte in a sample, inherently taking into account the recovery of the analytical procedure and also compensating for any matrix effect. The calibration curves for all PAHs evidenced a coefficient of determination between 0.9912 and 0.9997, and slopes similar to those obtained for each compound by the external calibration method with the exception of benz[a]anthracene and benzo[ghi]perylene. This confirmed the absence of a matrix effect. The PAHs found in this sample of pasteurised milk were: fluorene, phenanthrene, fluoranthene, pyrene, anthracene and benzo[a]pyrene. The amounts obtained by ESM and SAM were also very similar, with the exception of fluorene, even if SDs from external standard were slightly lower than those from the addition of the standard
707
method. The PAH levels determined by SAM were 0.15 ± 0.03 µg kg–1 (fluorene), 1.26 ± 0.10 µg kg–1 (phenanthrene), 0.18 ± 0.04 µg kg–1 (fluoranthene), 0.18 ± 0.47 µg kg–1 (pyrene), 0.13 ± 0.18 µg kg–1 (anthracene), and 0.18 ± 0.15 µg kg–1 (benzo[a]pyrene), whereas by ESM they were 0.35 ± 0.41, 1.42 ± 0.54, 0.08 ± 0.02, 0.24 ± 0.02, 0.33 ± 0.01 and 0.13 ± 0.01 µg kg–1, respectively.
Determination of PAHs in milk samples The chromatographic method used in this study to quantify PAHs in milk samples was shown be adequate (Figure 4). Only seven of the 15 compounds researched were detected: acenaphtene, fluorene, phenanthrene, anthracene, fluoranthene, pyrene and benzo[a]pyrene (Table 5), even if acenaphtene presented high uncertainties. The distribution of PAHs in milk shows the absence of high molecular weight compounds with the exception of benzo[a]pyrene in some pasteurised whole milk with concentrations ranging between 0.13 and 0.2 µg kg–1 milk. These values are lower than 1 µg kg–1, which is the maximum level fixed in infant milk (EC 835/2011). The distribution of PAHs can be ascribed to their different absorption during digestion (Grova et al. 2006; Lapole et al. 2007) and to the animal exposure to PAHs by inhalation of atmospheric particulates or intake of dietary food such as drinking water and contaminated feed (Tay-Lung et al. 2010).
Figure 4. Spectrofluorimetric chromatograms of (A) PAHs standard solution B: NA (1), AC (3), FL (4), PHE (5), AN (6), FA (7), PY (8), CHR (9), B[a]A (10), B[k]F (11), B[b]F (12), B [a]P (13), DB[ah]A (14), B[ghi]P (15), I[cd]P (16); and (B) whole pasteurised milk sample. Acenaphtylene (2) is not detectable in fluorescence.
708
A.M. Girelli et al.
Table 5. Levels of PAHs, expressed as mean ± SD (µg kg–1 milk) of the examined samples of pasteurised and UHT cow’s milk from popular Italian brands by external standard method. Pasteurised milk
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
Whole (N = 7) PAHs
Mean ± SD
Acenaphtene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benzo[a]pyrene Total PAHs
0.28 0.34 3.23 0.13 1.11 0.67 0.10 5.86
± ± ± ± ± ± ± ±
0.42 0.23 2.18 0.04 0.97 0.46 0.09 1.11
Semi-skimmed (N = 7)
Skimmed (N = 4)
Range
Mean ± SD
Range
Mean ± SD
Range
n.d.–1.01 n.d.–0.58 1.31–6.02 0.09–0.19 0.29–2.93 0.27–1.43 n.d.–0.20 2.6–10.2
0.93 ± 1.23 0.67 ± 0.45 8.84 ± 5.79 0.28 ± 0.18 4.52 ± 3.22 1.47 ± 1.83 n.d. 16.7 ± 3.36
n.d.–2.99 n.d.–0.43 3.95–18.52 0.12–0.53 1.65–9.01 n.d.–4.91 n.d. 6.32–35.6
1.08 ± 2.16 0.28 ± 0.21 13.24 ± 4.61 0.55 ± 0.43 7.63 ± 6.06 3.68 ± 2.98 n.d. 26.6 ± 5.13
n.d.–4.32 n.d.–0.49 7.56–18.70 0.20–1.17 3.61–16.55 1.72–8.20 n.d. 14.3–47.3
UHT milk Whole (N = 4) Acenaphtene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benzo[a]pyrene Total PAHs
n.d. 2.49 ± 4.32 8.52 ± 5.21 1.29 ± 1.42 4.35 ± 2.43 1.92 ± 1.77 n.d. 18.6 ± 2.92
n.d. n.d.–10.19 0.28–14.68 0.27–4.67 2.08–8.22 n.d.–4.52 n.d. 13.8 ± 29.4
Semi-skimmed (N = 4) n.d. n.d. 10.05 ± 2.04 0.28 ± 0.04 5.04 ± 1.36 2.91 ± 0.60 n.d. 18.3 ± 4.14
n.d. n.d. 7.72–11.51 0.23–0.31 3.63–6.35 2.39–3.56 n.d. 13.9 ± 21.1
Skimmed (N = 3) n.d. n.d. 9.52 ± 3.65 0.28 ± 0.11 4.82 ± 2.05 3.18 ± 1.43 n.d. 17.8 ± 3.87
n.d. n.d. 5.36–12.17 0.15–0.34 2.53–6.49 1.89–4.72 n.d. n.d.
Note: n.d., Not detectable.
From a comparison of the levels of PAHs in pasteurised and UHT milk samples from popular Italian brands (Table 5) it is evident that the heat treatment used in processing techniques of milk influences PAH formation, as reported by Naccari et al. (2011). Indeed, whole (ΣPAHs = 5.86 µg kg–1 milk) and semi-skimmed (ΣPAHs = 16.7 µg kg–1 milk) pasteurised milk present lower PAH concentrations than UHT samples (ΣPAHs = 18.6 and 18.3 µg kg–1 milk, respectively). Different effects of heat treatment are evident for skimmed milk samples with 0.3% fat levels (pasteurised: ΣPAHs = 26.6 µg kg–1 milk; UHT: ΣPAHs = 17.8 µg kg–1 milk). The influence of milk fat levels on the quantities of PAHs has not been widely studied (Kishikawa et al. 2003; Del Bubba et al. 2005) and so it influence is not yet clear. Relating to UHT milk samples studied in this paper, levels of PAHs in whole, semi-skimmed and skimmed milk are very similar even if a slight, not very significant decrease in PAHs was observed as a function of the total fat content. In the case of pasteurised milk, it was evident that the total content of PAHs significantly increased as the content of fat decreased (3.7% in whole, 1.7% in semi-skimmed milk and 0.3% in skimmed milk) (Table 5). These results are different from those reported by Kishikawa et al. (2003) for milk produced and commercialised in Japan (in which samples with more fats provided higher levels of
PAHs) and from the findings on human milk (Del Bubba et al. 2005) in which no relationship was found between PAHs and fat concentration. From the available data, it is not possible to explain these different findings since the formation of PAHs can be influenced by too many factors such as raw milk quality, different methods of milk treatment and production, environmental pollution, etc. In addition, it is not possible to exclude the influence of centrifugal force in skimming separators on the size of the fat globules (Iametti et al. 1997) and, consequently taking into account the fact that PAHs are dissolved in fats, on the release and levels of these pollutants in the milk samples. The increase of PAHs evidenced in pasteurised milk samples from popular brands (Table 5) was not shown for milk samples from a local farm. In fact in this case a significant reduction in PAH levels during the skimming process was obtained: 15.45 ± 0.53, 9.74 ± 0.32 and 6.68 ± 0.29 µg kg–1 in raw milk, whole and semi skimmed milk, respectively. This then would lead to a debate about the actual operations that occur in the milk-processing plants. It would be necessary to extend traceability for any products through all stages of production, processing and distribution. In this study three plant milk, namely soya, rice and oat, were examined (Table 6). For comparison, Figure 5 reports the results, expressed as mean values, with those obtained from whole cow’s pasteurised, cow’s UHT and goat’s UHT milk samples.
Food Additives & Contaminants: Part A
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
Table 6. Levels of PAHs (means ± SD µg kg–1 milk of four independent determinations) in plant milk samples by external standard method. PAHs
Soya milk
Phenanthrene Anthracene Fluoranthene Pyrene
6.98 0.29 5.28 2.51
± ± ± ±
1.84 0.08 1.72 0.69
Rice milk 6.55 0.26 3.01 2.07
± ± ± ±
0.12 0.03 0.31 0.18
Oat milk 8.90 0.38 5.62 2.46
± ± ± ±
0.50 0.01 0.10 0.38
Figure 5. Comparison among levels of PAHs, expressed as μg kg–1, of whole cow pasteurised and UHT, plant and goat milk samples from Italy.
It appeared that in all types of milk analysed the PAHs present at the highest levels were phenanthrene and fluoranthene; intermediate values were found for pyrene and fluorene; and lower concentrations for anthracene, acenaphtene and benzo[a]pyrene. Particularly acenaphtene and benzo[a]pyrene were detected only in some pasteurised milk. The absence of PAHs with more than four aromatic cycles in milk samples is in accordance of the findings reported for samples from Japan (Kishikawa et al. 2003), Kuwait (Husain et al. 1997), France (Grova, Feidt, Crepineau, et al. 2000; Grova et al. 2002), and Spain (Aguinaga et al. 2007). This is in accordance with the fact that the occurrence of PAH in foods is influenced by their physicochemical characteristics such as the very poor aqueous solubility that determines their capacity for distribution between different environmental compartments, the volatility that influences the transportation of PAHs in the atmosphere, and the lipophilicity that causes their accumulation in lipid tissue of plants and animals (Agency for Toxic Substances and Disease Registry 2009; Moon et al. 2012). PAH will not tend to accumulate in plant tissues with a high water content and limited transfer from the soil to root vegetables will occur (EC
709
2002). In addition, PAHs are strongly adsorbed in the organic fraction of soils and do not penetrate deeply into most soils, therefore limiting both leaching to groundwater and availability for uptake by plants. Consequently, the surface of vegetables can concentrate low molecular mass PAH mainly through surface adsorption (Kuipers et al. 2010). Their deposition could be obtained on fields also far from any contaminants sources since they can be transported over a longer distance than heavy compounds. Taking into account also that only PAHs with fewer than five rings are transferred from feed to cow’s milk as native compounds (Kan et al. 2003) while the other compounds are transferred as metabolites (Lutz et al. 2006), it is possible to hypothesise that the major source of PAHs for lactating species is the feed.
Conclusion This study has provided a sensitive and reliable analytical method for the determination of PAHs in milk samples. The calibration curves presented excellent linearity ranges (one to two orders of magnitude), excellent R2 (0.9930–0.9998) and very low LOQ (0.03–1.66 µg kg–1 milk). These were found to be lower than the values obtained by HPLC-FLD (Lutz et al. 2006; Moroles et al. 2007; Naccari et al. 2011) and GC-MS (Husain et al. 1997; Aguinaga et al. 2007). In addition, good recoveries (70–115%) were obtained with the exception of naphthalene and acenaphtene. The results obtained in this study provide evidence that the levels of carcinogenic PAHs in milk samples produced in Italy are not alarming, except for the non-carcinogenic PAHs which were high in various samples. However, the fact that we find the more volatile PAHs, which are not classified carcinogens to humans, is not to suggest that the milk samples tested are completely harmless to health since a massive accumulation in the body can lead, for example, to adverse effects in reproductive capacity (Mattison & Nightingale 1980). Therefore, B[a]P cannot be considered a suitable marker of exposure to PAHs in milk samples, as has already stated by the CONTAM Panel (European Food Safety Authority EFSA 2008). In addition, it is confirmed that the operations that occur in milk processing plants such as heat treatment and skimming processes influence PAH levels. Consequently, it would be necessary to extend the traceability for any products through all stages of production, processing and distribution.
Acknowledgments The authors are grateful for the support from University Sapienza of Rome, Italy.
710
A.M. Girelli et al.
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
References Agency for Toxic Substances and Disease Registry. 2009. Case Studies in Environmental Medicine Toxicity of Polycyclic Aromatic Hydrocarbons (PAHs) [Internet; cited 2009 Jul 1]. Available from: http://www.atsdr.cdc.gov/csem/pah/docs/ pah.pdf Aguinaga N, Campillo N, Vinas P, Hernandez-Cordoba M. 2007. Determination of 16 polycyclic aromatic hydrocarbons in milk and related products using solid-phase microextraction coupled to gaschromatography-mass spectrometry. Anal Chim Acta. 596:28–290. Andrews K, Schaum J, Schuda L. 2003. Method evaluation to measure persistent bioaccumulative toxic pollutants in cow milk. Organohalogen Compounds. 63. Costera A, Feidt C, Dziurla MA, Monteau F, Le Bizec B, Rychen G. 2009. Bioavailability of polycyclic aromatic hydrocarbons (PAHs) from soil and hay matrices in lactating goats. J Agric Food Chem. 57:5352–5357. Del Bubba M, Zanieri L, Galvan P, Donzelli GP, Checchini LO, Lepri L. 2005. Determination of polycyclic aromatic hydrocarbons (PAHs) and total fats in human milk. Ann Chim. 95:629–641. EC. 2011. Commission Regulation (EC) No 835/2011 of 19 August amending Regulation (EC) No 1881/2006 as regards maximum levels for polycyclic aromatic hydrocarbons in foodstuffs. Off J Eur Union. L 54:215/4–8. Environmental Protection Agency. 2006. [Internet]. Available from: http://water.epa.gov/scitech/methods/cwa/pollutantsbackground.cfm European Commission. 2002. Polycyclic aromatic hydrocarbons – occurrence in foods, dietary exposure and health effects [Internet]. Available from: http://ec.europa.eu/food/fs/sc/scf/ out154_en.pdf European Food Safety Authority. 2008. Polycyclic aromatic hydrocarbons in food. Scientific opinion of the panel on contaminants in the food chain. EFSA J. 724:1–114. Feidt C, Grova N, Laurent C, Rychen G, Laurent F. 2000. Le transfert des micropollulants organiques dans la chain alimentaire. Etat et perspectives de recherche. Oleagineux Corps Gras Lip. 5:431–435. Grova N, Feidt C, Crepineau C, Laurent F, Lafargue PE, Hachimi A, Rychen G. 2002. Detection of polycyclic hydrocarbon levels in milk collected near potential contamination sources. J Agric Food Chem. 50:4640–4642. Grova N, Feidt C, Laurent C, Rychen G. 2002. Milk, urine and feces excretion kinetics in lactating goats after an oral administration of 14C polycyclic aromatic hydrocarbons. Int Dairy J. 12:1025–1031. Grova N, Laurent C, Feidt C, Rychen G, Laurent F, Lichtfouse E. 2000. Gas chromatography-mass spectrometry study of polycyclic hydrocarbons in grass and milk from urban and rural farms. J Mass Spectrum. 6:457–460. Grova N, Rychen G, Monteau F, Le Bizec B, Feidt C. 2006. Effect of oral exposure to polycyclic aromatic hydrocarbons on goat’s milk contamination. Agron Sustain Dev. 26:195–199. Husain A, Naeemi E, Dashti B, Al-Omirah H, Al-Zenki S. 1997. Polycyclic aromatic hydrocarbons in food products originating from locally reared animals in Kuwait. Food Addit Contam. 14:295–299. Iametti S, Versuraro L, Tregna S, Giangiacomo R, Bonomi F. 1997. Surface properties of the fat globule in treated creams. Int Dairy J. 7:375–380.
International Agency for Research on Cancer. 2010. Some nonheterocyclic polycyclic aromatic hydrocarbons and some related exposures. IARC Monogr Eval Carcinog Risks Hum. 92:1–853. Kan CA, Traag WA, Hoogenboom LAP. 2003. Voorkomen van PAK’s in voer, omgeving van dieren, melk en zuivelproducten alsmede een oriënterende studie in melkvee. Animal Sciences Group Wageningen UR, Report nr. 03/ 0027745. Available from: http://library.wur.nl/way/bestanden/clc/1873114.pdf Kishikawa N, Kuroda N. 2009. Evaluation of organic environmental pollutans detected in human milk. J Health Sci. 55:1–10. Kishikawa N, Wada M, Kuroda N, Akiyama S, Nakashima K. 2003. Determination of polycyclic aromatic hydrocarbons in milk samples by high-performance liquid chromatography with fluorescence detection. J Chrom B. 789:257–264. Kuipers J, Oostdijk J, Schulz C, Ruthenschroer A. 2010. Detecting polycylic aromatic hydrocarbons in food [Internet; cited 2010 Aug]. Available from: http://www.foodquality.com/details/ article/842341/Detecting_Polycylic_Aromatic_Hydrocarbons_ in_Food.html?tzcheck=1 Lapole D, Rychen G, Grova N, Monteau F, Le Bizec B, Feidt C. 2007. Milk and urine excretion of polycyclic aromatic hydrocarbons and their hydroxylated metabolites after a single oral administration in ruminants. J Dairy Sci. 90:2624–2629. Lewis MJ, Deeth HC. 2008. Heat treatment of milk. In: Tamine AY, editor. Milk processing and quality management. Chichester: SDT, Wiley-Blackwell; p. 168–204. Lutz S, Feidt C, Monteau F, Rychen G, Le Bizec B, Jurianz S. 2006. Effect of exposure to soil-bound polycyclic aromatic hydrocarbons on milk contaminations of parent compounds and their monohydroxylated metabolites. J Agric Food Chem. 54:263–268. Mandel J. 1964. The statistical analysis of experimental data. New York (NY): Wiley & Sons. Mattison DR, Nightingale MR. 1980. The biochemical and genetic characteristics of murine ovarian aryl hydrocarbon (Benzo[a]pyrene) hydroxylase activity and its relationship to primordial oocyte destruction by polycyclic aromatic hydrocarbons. Toxicol Appl Pharm. 56:399–408. Michalski MC, Michel F, Sainmont D, Briard V. 2002. Apparent ζ-potential as a tool to assess mechanical damages to the milk fat globule membrane. Coll Surf B Biointer. 23:23–30. Moon HB, Lee DH, Lee YS, Kannan K. 2012. Occurrence and accumulation patterns of polycyclic aromatic hydrocarbons and synthetic musk compounds in adipose tissues of Korean females. Chemosphere. 86:485–490. Moroles N, Ulloa HJ, Castro-Rios R, Villarelal EG, Barbarin-Castillo JM, Salazar-Cavazos M, Waksman-De Torres N. 2007. A comparison of the performance of two chromatographic and three extraction techniques for the analysis of PAHs in sources of drinking water. J Chromatogr Sci. 45:57–62. Naccari C, Cristani M, Giofrè F, Ferrante M, Siracusa L, Trombetta D. 2011. PAHs concentration in heat-treated milk samples. Food Res Int. 44:716–724. Simoneit BRT. 2002. Biomass burning a review of organic tracers for smoke from incomplete combustion. Appl Geochem. 17:129–162. Tay-Lung C, Chiu-Jung L, Meei-Fang C. 2010. Comparison of liquid–liquid extraction and solid phase extraction for the determination of polycyclic aromatic hydrocarbons in the milk of Taiwan. J Taiwan Inst Chem Eng. 41:178–183.
Food Additives & Contaminants: Part A Publication details, including instructions for authors and subscription information: http://www.tandfonline.com/loi/tfac20
Determination of polycyclic aromatic hydrocarbons in Italian milk by HPLC with fluorescence detection a
a
A.M. Girelli , D. Sperati & A.M. Tarola a
b
Department of Chemistry, Sapienza University of Rome, P.le A. Moro 5, I-00185 Rome, Italy
b
Laboratory of Commodity Science, Department Management, Sapienza University of Rome, Via Castro Laurenziano 9, I-00161 Rome, Italy Accepted author version posted online: 13 Jan 2014.Published online: 03 Mar 2014.
To cite this article: A.M. Girelli, D. Sperati & A.M. Tarola (2014) Determination of polycyclic aromatic hydrocarbons in Italian milk by HPLC with fluorescence detection, Food Additives & Contaminants: Part A, 31:4, 703-710, DOI: 10.1080/19440049.2013.878959 To link to this article: http://dx.doi.org/10.1080/19440049.2013.878959
PLEASE SCROLL DOWN FOR ARTICLE Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and should be independently verified with primary sources of information. Taylor and Francis shall not be liable for any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of the Content. This article may be used for research, teaching, and private study purposes. Any substantial or systematic reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http:// www.tandfonline.com/page/terms-and-conditions
Food Additives & Contaminants: Part A, 2014 Vol. 31, No. 4, 703–710, http://dx.doi.org/10.1080/19440049.2013.878959
Determination of polycyclic aromatic hydrocarbons in Italian milk by HPLC with fluorescence detection A.M. Girellia*, D. Speratia and A.M. Tarolab a
Department of Chemistry, Sapienza University of Rome, P.le A. Moro 5, I-00185 Rome, Italy; bLaboratory of Commodity Science, Department Management, Sapienza University of Rome, Via Castro Laurenziano 9, I-00161 Rome, Italy
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
(Received 27 September 2013; accepted 18 December 2013) The presence of polycyclic aromatic hydrocarbons (PAHs) in Italian commercial milk samples is reported. The study was carried out on lactating (cow and goat) and plant (rice, soya, oat) milk. The quantitative determination involved liquid– liquid extraction of PAHs, a pre-concentration and determination by HPLC using a fluorescence detector. The recovery of analytes was in the range of 70–115%. The precision of the method was found to be between 6% and 24%. The detection limit ranged from 0.66 to 33.3 µg l–1 corresponding to 0.03–1.66 µg kg–1 milk (wet weight), at a signal-to-noise ratio of 3, depending on the compound. By this procedure, the levels of more volatile PAHs (two to three aromatic rings) were confirmed in 34 commercial milk and three plant milk samples, whereas benzo[a]pyrene was found only in five pasteurised milk samples at a mean concentration of 0.17 µg kg–1 milk. These results provide evidence that PAH levels are influenced by heat treatments and skimming processes of milk production. Keywords: PAHs; cow milk; plant milk; HPLC-FLD
Introduction Polycyclic aromatic hydrocarbons (PAHs) are a group of ubiquitous contaminants with significant toxicity (Agency for Toxic Substances and Disease Registry 2009), potential mutagenic and carcinogenic properties (International Agency for Research on Cancer 2010). Sixteen of them are even included in the list of priority pollutants by the USEPA (2006). They are present in all environmental components and may cause harmful effects near to and distant from their source (Feidt et al. 2000). The principal source of human exposure is the ingestion of contaminated food in which PAHs may accumulate in the lipid phase. Indeed the presence of fused aromatic rings favours migration through the food chain into hydrophobic compartments. Therefore, a primary dietary human food, like milk, which contains a high proportion of lipids (triacylglycerols, phospholipids, fatty acids and sterols), may be contaminated by PAHs. In fact environmental factors related to the rearing system (fodder and potential contaminated soil) and ruminant’s source of exposure (ingestion during grazing, drinking of contaminated water and inhalation of contaminated air) are the main source of PAHs intake from lactating ruminants (Lutz et al. 2006; Lapole et al. 2007). In addition food processing carried out at high temperature can cause the combustion of organic matter with a recombination of free-radicals formed, at first, in low molecular mass PAHs and then in higher molecular mass PAHs (Simoneit 2002). *Corresponding author. Email: [email protected] © 2014 Taylor & Francis
In the case of lactose intolerance by individuals, alternative products like “plant milks” and derivatives such as soy, rice and oat milk are frequently consumed. These products that are lactose-free that resemble milk in both appearance and nutrition. Milk and plant milk could always be considered interesting biomarkers for assessing the exposure to environmental contamination and for human risk evaluation. For this reason we report here a comparison of PAH levels in pasteurised and UHT commercial milks (whole, semi-skimmed and skimmed) and in samples of plant milk which are different in their production processes and their fat content. Pasteurised milk is the most consumed milk in Europe, produced with a mild heat treatment (72°C for a minimum of 15 s) according to a process of high temperature short time (HTST), to eliminate the original bacteria, but not the spores that can germinate after opening the milk box. Ultra-high-temperature treatment (UHT) can destroy both bacteria and spores, since the milk is heated to a temperature of at least 135°C. So any dangerous bacteria are removed (Lewis & Deeth 2008). Semi-skimmed and skimmed milk, which contain, respectively, 1.7% fat and a content between 0% and 0.5% fat, are formed by skimming cream (high-fat) from whole milk. The separation process is generally carried out in a centrifugal separator which is based on the fact that when liquids of different specific gravities resolve around the same centre at the same distance with the same angular velocity, a greater centrifugal force is exerted on the heavier liquid than on the lighter one. Milk can be regarded as two
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
704
A.M. Girelli et al.
liquids of different specific gravities: the serum and the fat. Taking into account that in fresh raw whole milk fat is in the form of natural emulsified globules surrounded by a native stabilising membrane, the milk fat globule membrane (MFGM) and that in milk submitted to severe mechanical treatments (such as in homogenisation or skimming process) and heat treatment, the native MFGM can be damaged. This results in a reduction in milk fat globule size and in the alteration of interfacial and electrostatic properties (Michalski et al. 2002). There are various data in the literature concerning the levels of PAHs in milk samples from different countries, such as Taiwan (Tay-Lung et al. 2010), Kuwait (Husain et al. 1997), Puerto Rico (Andrews et al. 2003), Spain (Aguinaga et al. 2007), France (Grova, Feidt, Crepineau, et al. 2002), Japan (Kishikawa et al. 2003) and also from the Italian region of Calabria (Naccari et al. 2011), and some studies demonstrated the transfer of low-molecular PAHs with fewer than five rings from feed to goats (Grova, Feidt, Laurent, et al. 2002; Costera et al. 2009), cows (Kan et al. 2003), and human milk (Kishikawa & Kuroda 2009). In all these papers milk contamination from PAHs was reported, but it is not clear if differences in concentrations (1 ng kg–1 to 100 μg kg–1 milk) and in PAH distribution can be attributed to the different environmental pollution exposure or to the method of analysis. The aim of this study was to determine the levels of the more extensively studied 16 priority pollutant PAHs (Environmental Protection Agency 2006) in milk samples from Italy. In particular, samples of whole, semi-skimmed and skimmed pasteurised and UHT cow’s milk were examined to evaluate the role of food processing, such as heating and skimming treatments, on PAH formation. Samples of some “plant milk” such as rice, soya and oat were also investigated since they are actually healthy alternatives to cow’s milk.
Table 1.
NA ACL AC FL PHE AN FA PY B[a]A CHR B[b]FA B[k]FA B[a]P DB[ah]A B[ghi]P I[cd]P
Materials and methods Chemicals PAHs pure reference standard solution in CH3CN: MeOH = 9:1 (v:v) (EPA TCL PAH Mix) (concentrations are reported in Table 1) and all 16 standard PAHs (naphthalene; acenaphtylene; acenaphtene; fluorene; phenanthrene; anthracene; fluoranthene; pyrene; benz[a] anthracene; chrysene; benzo[k]fluoranthene; benzo[b] fluoranthene; benzo[a]pyrene; indeno[1,2,3-cd]pyrene; dibenz[ah]anthracene; and benzo[ghi]perylene) were purchased from Supelco (Milan, Italy). Acetonitrile (99.9%), water for HPLC, cyclohexane ACS (>99.5%) Hexane ACS (99%), ethanol (96%) NaOH were obtained from Sigma-Aldrich (Milan, Italy); while pentane was obtained from Carlo Erba (Milan, Italy). ISO discTm filters PTFE 4 mm × 0.45 µm were from Supelco.
HPLC system The HPLC system consisted of a liquid chromatographic system LC module (Jasco PU1580; Milan, Italy), a 7125 injector (Rheodyne, Cotati, CA, USA) with a 20 µl sample loop, a 110 series fluorescence detector (Agilent, Rome, Italy) and an HP Chemstation integrator. Separation of PAHs was achieved by a column C-18 (150 × 4.6 mm ID; particle size 5 µm) (Prosphere 300; Alltech, Rome, Italy) with a gradient elution programme using solvent A (1% CH3CN in aqueous solution) and solvent B (99% CH3CN in aqueous solution). The gradient elution was programmed as follows: 40% B to 50% B linearly (0– 20 min), 50% B to 100% B (20–37 min), 100% B (37– 42 min) and 100% B to 40% B (42–45 min). The flow rate was set at 1 ml min–1, at RT. Under these conditions 16 PAHs could be separated satisfactorily within 45 min. The
Amounts of individual PAH in reference standard (A), diluted 1:1000 (B) and 1:5000 (C) solutions. PAH
Solution A (µg ml–1)
Solution B (µg ml–1)
Solution C (µg ml–1)
Naphtalene Acenaphtylene Acenaphtene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benzo[a]anthracene Chrysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[a]pyrene Dibenzo[ah]anthracene Benzo[ghi]perylene Indeno[1,2,3-cd]pyrene
500 500 1000 100 40 20 50 100 50 50 20 20 50 200 80 50
0.500 0.500 1.009 0.100 0.040 0.020 0.050 0.100 0.050 0.050 0.020 0.020 0.050 0.200 0.080 0.050
0.100 0.100 0.200 0.020 0.008 0.004 0.010 0.020 0.010 0.010 0.004 0.004 0.010 0.040 0.016 0.010
Food Additives & Contaminants: Part A Table 2. PAHs.
Detection wavelength programme for the separation of
Time (min)
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
0.00 20.00 23.50 26.00 28.50 35.60 42.50
λex (nm)
λem (nm)
280 250 254 280 265 280 364
340 360 400 435 390 410 500
detection wavelength programme to determine individual PAHs is shown in Table 2.
Sampling Thirty-seven milk samples comprising 29 cow’s milk, two goat’s milk and three plant milk (rice, soya and oat) were obtained from a local supermarket while three of cow’s milk (one raw, one whole and one semi-skimmed) were provided by a local farm. The 29 samples of cow’s milk were made up of 18 pasteurised milk (seven whole, seven semi-skimmed and four skimmed) and 11 UHT milk (four whole, four semi-skimmed and three skimmed). After the homogenisation with a stainless steel blender, all the samples were kept at –20°C until analysis.
Sample preparation Milk samples (2 g) in a 10-ml reaction vial with a screw cap were saponified with 4.0 ml of NaOH ethanolic solution 0.4 M. The mixture was placed in a water bath at 60°C for 30 min and cooled to RT. Then 2.0 ml of cyclohexane were added and the mixture was vortexed for 5 min. The supernatant layer was recovered and transferred into a vial. The samples were re-extracted, as previously described with 2 ml of cyclohexane. A total of 3.5 ml of the combined extracts was successively taken and evaporated to dryness under nitrogen. The residue was then recovered with 100 µl of acetonitrile and filtered with a PTFE membrane filters of 0.45 µm (Sigma Aldrich, Milan, Italy). An aliquot of 20 µl of this solution was injected into the HPLC system. A reagent blank, constituted by 2 ml of deionised water, was simultaneously performed by the same procedure above described, with each series of samples. For PAH quantifications eight standard working solutions at various concentrations were prepared by appropriate dilution of aliquots of the stock solution in acetonitrile in order to prepare the calibration curve and to calculate the LODs and LOQs for all compounds.
705
Results and discussion Optimisation of extraction conditions In a preliminary study a liquid–liquid extraction of an aqueous standard solution was examined by using n-pentane, n-hexane and cyclohexane as solvents. The standard solution was obtained by diluting 100 µl of solution C containing PAHs in the range 0.02–1 µg ml–1, as indicated in Table 1, to 2 ml of deionised water. The extractability of PAHs, expressed as recovery percentage, were examined with two aliquots of 2 ml of solvent (Figure 1). It appeared that cyclohexane, even if being less volatile than n-hexane and with an evaporation time greater than pentane, has the best extraction efficiency of PAHs, with the exception of naphthalene. Thus, cyclohexane was chosen for further experimentation. Successively, taking into account that the extraction efficiency depends not only on partition coefficients, temperature and ionic force, but also on the extraction volume, a study with a variable volume of cyclohexane (2, 4, 6 ml) was undertaken in triplicate analysis (Figure 2). The different recovery of PAHs was due to the different extraction efficiency and to the different volatility of each compound. Indeed, by increasing the volume of organic phase, an increase in the time of cyclohexane evaporation from extracts and a greater loss of volatile analyte could be obtained. Consequently, PAHs with low molecular mass (two to three rings) (AC and PHE) presented 4 ml as the preferred extraction value but for those with more than four aromatic rings (from FA to I[cd]P) a value of 2 ml was optimal. To avoid any decrease in extractability for ACE and PHE, a value of 4 ml as the total volume was chosen. To optimise successively the number of repeated extractions, a study was undertaken with the standard solution C (Table 1) by varying the number of extractions (one extraction with 4 ml, two extractions with 2 ml and four extractions with 1 ml)
Figure 1. Effect of extraction solvent on the recovery of compounds, expressed as [PAHs]found/[PAHs]spiked (%).
706
A.M. Girelli et al. In the saponification step, an NaOH concentration of 0.4 M was fixed in accordance with Kishikawa et al. (2003). The solvent was 100% EtOH since a decrease in the level of ethanol and an increase of water not influenced PAH extraction but caused an increase in the separation time of the organic layer (data not shown).
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
HPLC-FLD method
Figure 2. PAHs.
The performance of the HPLC-FLD method was evaluated by establishing the quality parameters. These parameters were determined using standard solutions; the results obtained are shown in Table 3. For each compound, the linearity of the calibration curve, obtained with eight calibration points by plotting analyte area versus concentration, was confirmed by Mandel’s fitting text (Mandel 1964) in conjunction with an evaluation of the plot. All the investigated calibration curves were characterised by high coefficients of determination (R2 > 0.99) and with variation coefficients (CV%) < 10% with the exception of indeno[1,2,3-cd]pyrene. LOD and LOQ, defined as the concentration of the analyte that produced a signal-tonoise ratio of 3 and 10, respectively, were higher than 0.20 and 0.66 µg l–1 and were then texted experimentally by spiking blank samples at such levels. It appeared that high values of LOD and LOQ for indeno[1,2,3-cd]pyrene occurred since it presented the weakest fluorescence. In order to verify the accuracy and precision of the analytical procedure, recovery experiments were carried out by adding 100 µl of standard PAHs solutions appropriately diluted to 2 g of a pasteurised milk sample rich in fat (3.6%). For each level, five different spiked samples, as well as unspiked controls, were prepared and each of these samples was analysed in triplicate according to the analytical procedure. Recoveries were calculated from the differences between the spiked and unspiked samples in the total amount of each compound for each level. The mean
Effect of cyclohexane volume on the peak area of
Figure 3. Effect of the number of extractions on the recovery of compounds, expressed as [PAHs]found/[PAHs]spiked (%).
but maintaining fixed the total volume of cyclohexane (Figure 3). The maximum recovery percentage for almost all the PAHs was obtained for two extractions, therefore the number of extractions was set at two. Table 3. Correlation coefficients, linearity ranges, LODs and LOQs. PAHs Naphtalene Acenaphtene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz[a]anthracene Chrysene Benzo[b]fluoranthene Benzo[k]fluoranthene Benzo[a]pyrene Dibenzo[ah]anthracene Benzo[ghi]perylene Indeno[1,2,3-cd]pyrene
R2
Linearity (μg l–1)
LOD (µg l–1)
LOQ (µg l–1)
0.9998 0.9998 0.9509 0.9963 0.9999 0.9930 0.9984 0.9962 0.9970 0.9938 0.9965 0.9948 0.9973 0.9969 0.9944
2.1–500 5.3–300 5.0–30 1.7–40 0.66–20 4.2–15 7.0–100 1.5–30 1.5–50 6.7–20 0.83–20 2.1–50 2.0–200 4.6–80 33–50
0.63 1.60 1.50 0.35 0.20 1.25 2.10 0.45 0.45 2.00 0.25 0.63 0.60 1.40 10
2.10 5.3 5.00 1.70 0.66 4.20 7.0 1.50 1.50 6.70 0.83 2.10 2.00 4.60 33.3
Food Additives & Contaminants: Part A Table 4. Mean recoveries and coefficients of variation (CV) of polycyclic aromatic hydrocarbons (PAHs) added to whole pasteurised milk.
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
PAHs Naphtalene Acenaphtene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benz[a]anthracene Chrysene Benzo[b] fluoranthene Benzo[k] fluoranthene Benzo[a]pyrene Dibenzo[ah] anthracene Benzo[ghi]perylene Indeno[1,2,3-cd] pyrene
Spiking level (µg kg–1)
Recovery (%)
CV (%)
1.25 2.50 0.25 0.10 0.05 0.12 0.25 0.12 0.12 0.50
15 43 76 70 100 110 89 60 115 74
25 23 9 7 22 6 23 14 24 10
0.05
97
19
0.12 0.50
88 74
25 13
0.20 0.20
69 80
18 16
recovery percentages and coefficients of variation are shown in Table 4 at the lowest level of PAHs spiked to milk samples. It appeared that the recoveries were related to PAH volatility: low values were observed for the more volatile compounds such as acenaphtene and, above all, naphthalene. This was correlated to their loss during the step of cyclohexane evaporation. In addition to the verification of the absence of any matrix effects, the data obtained for the same pasteurised milk sample by the external standard method (ESM) were compared with those obtained by the standard addition method (SAM). In this last procedure the milk sample was divided in four portions and for each were added known amounts of the standard mixture of PAHs prior to saponification. This procedure was designed to determine the content of an analyte in a sample, inherently taking into account the recovery of the analytical procedure and also compensating for any matrix effect. The calibration curves for all PAHs evidenced a coefficient of determination between 0.9912 and 0.9997, and slopes similar to those obtained for each compound by the external calibration method with the exception of benz[a]anthracene and benzo[ghi]perylene. This confirmed the absence of a matrix effect. The PAHs found in this sample of pasteurised milk were: fluorene, phenanthrene, fluoranthene, pyrene, anthracene and benzo[a]pyrene. The amounts obtained by ESM and SAM were also very similar, with the exception of fluorene, even if SDs from external standard were slightly lower than those from the addition of the standard
707
method. The PAH levels determined by SAM were 0.15 ± 0.03 µg kg–1 (fluorene), 1.26 ± 0.10 µg kg–1 (phenanthrene), 0.18 ± 0.04 µg kg–1 (fluoranthene), 0.18 ± 0.47 µg kg–1 (pyrene), 0.13 ± 0.18 µg kg–1 (anthracene), and 0.18 ± 0.15 µg kg–1 (benzo[a]pyrene), whereas by ESM they were 0.35 ± 0.41, 1.42 ± 0.54, 0.08 ± 0.02, 0.24 ± 0.02, 0.33 ± 0.01 and 0.13 ± 0.01 µg kg–1, respectively.
Determination of PAHs in milk samples The chromatographic method used in this study to quantify PAHs in milk samples was shown be adequate (Figure 4). Only seven of the 15 compounds researched were detected: acenaphtene, fluorene, phenanthrene, anthracene, fluoranthene, pyrene and benzo[a]pyrene (Table 5), even if acenaphtene presented high uncertainties. The distribution of PAHs in milk shows the absence of high molecular weight compounds with the exception of benzo[a]pyrene in some pasteurised whole milk with concentrations ranging between 0.13 and 0.2 µg kg–1 milk. These values are lower than 1 µg kg–1, which is the maximum level fixed in infant milk (EC 835/2011). The distribution of PAHs can be ascribed to their different absorption during digestion (Grova et al. 2006; Lapole et al. 2007) and to the animal exposure to PAHs by inhalation of atmospheric particulates or intake of dietary food such as drinking water and contaminated feed (Tay-Lung et al. 2010).
Figure 4. Spectrofluorimetric chromatograms of (A) PAHs standard solution B: NA (1), AC (3), FL (4), PHE (5), AN (6), FA (7), PY (8), CHR (9), B[a]A (10), B[k]F (11), B[b]F (12), B [a]P (13), DB[ah]A (14), B[ghi]P (15), I[cd]P (16); and (B) whole pasteurised milk sample. Acenaphtylene (2) is not detectable in fluorescence.
708
A.M. Girelli et al.
Table 5. Levels of PAHs, expressed as mean ± SD (µg kg–1 milk) of the examined samples of pasteurised and UHT cow’s milk from popular Italian brands by external standard method. Pasteurised milk
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
Whole (N = 7) PAHs
Mean ± SD
Acenaphtene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benzo[a]pyrene Total PAHs
0.28 0.34 3.23 0.13 1.11 0.67 0.10 5.86
± ± ± ± ± ± ± ±
0.42 0.23 2.18 0.04 0.97 0.46 0.09 1.11
Semi-skimmed (N = 7)
Skimmed (N = 4)
Range
Mean ± SD
Range
Mean ± SD
Range
n.d.–1.01 n.d.–0.58 1.31–6.02 0.09–0.19 0.29–2.93 0.27–1.43 n.d.–0.20 2.6–10.2
0.93 ± 1.23 0.67 ± 0.45 8.84 ± 5.79 0.28 ± 0.18 4.52 ± 3.22 1.47 ± 1.83 n.d. 16.7 ± 3.36
n.d.–2.99 n.d.–0.43 3.95–18.52 0.12–0.53 1.65–9.01 n.d.–4.91 n.d. 6.32–35.6
1.08 ± 2.16 0.28 ± 0.21 13.24 ± 4.61 0.55 ± 0.43 7.63 ± 6.06 3.68 ± 2.98 n.d. 26.6 ± 5.13
n.d.–4.32 n.d.–0.49 7.56–18.70 0.20–1.17 3.61–16.55 1.72–8.20 n.d. 14.3–47.3
UHT milk Whole (N = 4) Acenaphtene Fluorene Phenanthrene Anthracene Fluoranthene Pyrene Benzo[a]pyrene Total PAHs
n.d. 2.49 ± 4.32 8.52 ± 5.21 1.29 ± 1.42 4.35 ± 2.43 1.92 ± 1.77 n.d. 18.6 ± 2.92
n.d. n.d.–10.19 0.28–14.68 0.27–4.67 2.08–8.22 n.d.–4.52 n.d. 13.8 ± 29.4
Semi-skimmed (N = 4) n.d. n.d. 10.05 ± 2.04 0.28 ± 0.04 5.04 ± 1.36 2.91 ± 0.60 n.d. 18.3 ± 4.14
n.d. n.d. 7.72–11.51 0.23–0.31 3.63–6.35 2.39–3.56 n.d. 13.9 ± 21.1
Skimmed (N = 3) n.d. n.d. 9.52 ± 3.65 0.28 ± 0.11 4.82 ± 2.05 3.18 ± 1.43 n.d. 17.8 ± 3.87
n.d. n.d. 5.36–12.17 0.15–0.34 2.53–6.49 1.89–4.72 n.d. n.d.
Note: n.d., Not detectable.
From a comparison of the levels of PAHs in pasteurised and UHT milk samples from popular Italian brands (Table 5) it is evident that the heat treatment used in processing techniques of milk influences PAH formation, as reported by Naccari et al. (2011). Indeed, whole (ΣPAHs = 5.86 µg kg–1 milk) and semi-skimmed (ΣPAHs = 16.7 µg kg–1 milk) pasteurised milk present lower PAH concentrations than UHT samples (ΣPAHs = 18.6 and 18.3 µg kg–1 milk, respectively). Different effects of heat treatment are evident for skimmed milk samples with 0.3% fat levels (pasteurised: ΣPAHs = 26.6 µg kg–1 milk; UHT: ΣPAHs = 17.8 µg kg–1 milk). The influence of milk fat levels on the quantities of PAHs has not been widely studied (Kishikawa et al. 2003; Del Bubba et al. 2005) and so it influence is not yet clear. Relating to UHT milk samples studied in this paper, levels of PAHs in whole, semi-skimmed and skimmed milk are very similar even if a slight, not very significant decrease in PAHs was observed as a function of the total fat content. In the case of pasteurised milk, it was evident that the total content of PAHs significantly increased as the content of fat decreased (3.7% in whole, 1.7% in semi-skimmed milk and 0.3% in skimmed milk) (Table 5). These results are different from those reported by Kishikawa et al. (2003) for milk produced and commercialised in Japan (in which samples with more fats provided higher levels of
PAHs) and from the findings on human milk (Del Bubba et al. 2005) in which no relationship was found between PAHs and fat concentration. From the available data, it is not possible to explain these different findings since the formation of PAHs can be influenced by too many factors such as raw milk quality, different methods of milk treatment and production, environmental pollution, etc. In addition, it is not possible to exclude the influence of centrifugal force in skimming separators on the size of the fat globules (Iametti et al. 1997) and, consequently taking into account the fact that PAHs are dissolved in fats, on the release and levels of these pollutants in the milk samples. The increase of PAHs evidenced in pasteurised milk samples from popular brands (Table 5) was not shown for milk samples from a local farm. In fact in this case a significant reduction in PAH levels during the skimming process was obtained: 15.45 ± 0.53, 9.74 ± 0.32 and 6.68 ± 0.29 µg kg–1 in raw milk, whole and semi skimmed milk, respectively. This then would lead to a debate about the actual operations that occur in the milk-processing plants. It would be necessary to extend traceability for any products through all stages of production, processing and distribution. In this study three plant milk, namely soya, rice and oat, were examined (Table 6). For comparison, Figure 5 reports the results, expressed as mean values, with those obtained from whole cow’s pasteurised, cow’s UHT and goat’s UHT milk samples.
Food Additives & Contaminants: Part A
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
Table 6. Levels of PAHs (means ± SD µg kg–1 milk of four independent determinations) in plant milk samples by external standard method. PAHs
Soya milk
Phenanthrene Anthracene Fluoranthene Pyrene
6.98 0.29 5.28 2.51
± ± ± ±
1.84 0.08 1.72 0.69
Rice milk 6.55 0.26 3.01 2.07
± ± ± ±
0.12 0.03 0.31 0.18
Oat milk 8.90 0.38 5.62 2.46
± ± ± ±
0.50 0.01 0.10 0.38
Figure 5. Comparison among levels of PAHs, expressed as μg kg–1, of whole cow pasteurised and UHT, plant and goat milk samples from Italy.
It appeared that in all types of milk analysed the PAHs present at the highest levels were phenanthrene and fluoranthene; intermediate values were found for pyrene and fluorene; and lower concentrations for anthracene, acenaphtene and benzo[a]pyrene. Particularly acenaphtene and benzo[a]pyrene were detected only in some pasteurised milk. The absence of PAHs with more than four aromatic cycles in milk samples is in accordance of the findings reported for samples from Japan (Kishikawa et al. 2003), Kuwait (Husain et al. 1997), France (Grova, Feidt, Crepineau, et al. 2000; Grova et al. 2002), and Spain (Aguinaga et al. 2007). This is in accordance with the fact that the occurrence of PAH in foods is influenced by their physicochemical characteristics such as the very poor aqueous solubility that determines their capacity for distribution between different environmental compartments, the volatility that influences the transportation of PAHs in the atmosphere, and the lipophilicity that causes their accumulation in lipid tissue of plants and animals (Agency for Toxic Substances and Disease Registry 2009; Moon et al. 2012). PAH will not tend to accumulate in plant tissues with a high water content and limited transfer from the soil to root vegetables will occur (EC
709
2002). In addition, PAHs are strongly adsorbed in the organic fraction of soils and do not penetrate deeply into most soils, therefore limiting both leaching to groundwater and availability for uptake by plants. Consequently, the surface of vegetables can concentrate low molecular mass PAH mainly through surface adsorption (Kuipers et al. 2010). Their deposition could be obtained on fields also far from any contaminants sources since they can be transported over a longer distance than heavy compounds. Taking into account also that only PAHs with fewer than five rings are transferred from feed to cow’s milk as native compounds (Kan et al. 2003) while the other compounds are transferred as metabolites (Lutz et al. 2006), it is possible to hypothesise that the major source of PAHs for lactating species is the feed.
Conclusion This study has provided a sensitive and reliable analytical method for the determination of PAHs in milk samples. The calibration curves presented excellent linearity ranges (one to two orders of magnitude), excellent R2 (0.9930–0.9998) and very low LOQ (0.03–1.66 µg kg–1 milk). These were found to be lower than the values obtained by HPLC-FLD (Lutz et al. 2006; Moroles et al. 2007; Naccari et al. 2011) and GC-MS (Husain et al. 1997; Aguinaga et al. 2007). In addition, good recoveries (70–115%) were obtained with the exception of naphthalene and acenaphtene. The results obtained in this study provide evidence that the levels of carcinogenic PAHs in milk samples produced in Italy are not alarming, except for the non-carcinogenic PAHs which were high in various samples. However, the fact that we find the more volatile PAHs, which are not classified carcinogens to humans, is not to suggest that the milk samples tested are completely harmless to health since a massive accumulation in the body can lead, for example, to adverse effects in reproductive capacity (Mattison & Nightingale 1980). Therefore, B[a]P cannot be considered a suitable marker of exposure to PAHs in milk samples, as has already stated by the CONTAM Panel (European Food Safety Authority EFSA 2008). In addition, it is confirmed that the operations that occur in milk processing plants such as heat treatment and skimming processes influence PAH levels. Consequently, it would be necessary to extend the traceability for any products through all stages of production, processing and distribution.
Acknowledgments The authors are grateful for the support from University Sapienza of Rome, Italy.
710
A.M. Girelli et al.
Downloaded by [University of Newcastle (Australia)] at 08:26 08 September 2014
References Agency for Toxic Substances and Disease Registry. 2009. Case Studies in Environmental Medicine Toxicity of Polycyclic Aromatic Hydrocarbons (PAHs) [Internet; cited 2009 Jul 1]. Available from: http://www.atsdr.cdc.gov/csem/pah/docs/ pah.pdf Aguinaga N, Campillo N, Vinas P, Hernandez-Cordoba M. 2007. Determination of 16 polycyclic aromatic hydrocarbons in milk and related products using solid-phase microextraction coupled to gaschromatography-mass spectrometry. Anal Chim Acta. 596:28–290. Andrews K, Schaum J, Schuda L. 2003. Method evaluation to measure persistent bioaccumulative toxic pollutants in cow milk. Organohalogen Compounds. 63. Costera A, Feidt C, Dziurla MA, Monteau F, Le Bizec B, Rychen G. 2009. Bioavailability of polycyclic aromatic hydrocarbons (PAHs) from soil and hay matrices in lactating goats. J Agric Food Chem. 57:5352–5357. Del Bubba M, Zanieri L, Galvan P, Donzelli GP, Checchini LO, Lepri L. 2005. Determination of polycyclic aromatic hydrocarbons (PAHs) and total fats in human milk. Ann Chim. 95:629–641. EC. 2011. Commission Regulation (EC) No 835/2011 of 19 August amending Regulation (EC) No 1881/2006 as regards maximum levels for polycyclic aromatic hydrocarbons in foodstuffs. Off J Eur Union. L 54:215/4–8. Environmental Protection Agency. 2006. [Internet]. Available from: http://water.epa.gov/scitech/methods/cwa/pollutantsbackground.cfm European Commission. 2002. Polycyclic aromatic hydrocarbons – occurrence in foods, dietary exposure and health effects [Internet]. Available from: http://ec.europa.eu/food/fs/sc/scf/ out154_en.pdf European Food Safety Authority. 2008. Polycyclic aromatic hydrocarbons in food. Scientific opinion of the panel on contaminants in the food chain. EFSA J. 724:1–114. Feidt C, Grova N, Laurent C, Rychen G, Laurent F. 2000. Le transfert des micropollulants organiques dans la chain alimentaire. Etat et perspectives de recherche. Oleagineux Corps Gras Lip. 5:431–435. Grova N, Feidt C, Crepineau C, Laurent F, Lafargue PE, Hachimi A, Rychen G. 2002. Detection of polycyclic hydrocarbon levels in milk collected near potential contamination sources. J Agric Food Chem. 50:4640–4642. Grova N, Feidt C, Laurent C, Rychen G. 2002. Milk, urine and feces excretion kinetics in lactating goats after an oral administration of 14C polycyclic aromatic hydrocarbons. Int Dairy J. 12:1025–1031. Grova N, Laurent C, Feidt C, Rychen G, Laurent F, Lichtfouse E. 2000. Gas chromatography-mass spectrometry study of polycyclic hydrocarbons in grass and milk from urban and rural farms. J Mass Spectrum. 6:457–460. Grova N, Rychen G, Monteau F, Le Bizec B, Feidt C. 2006. Effect of oral exposure to polycyclic aromatic hydrocarbons on goat’s milk contamination. Agron Sustain Dev. 26:195–199. Husain A, Naeemi E, Dashti B, Al-Omirah H, Al-Zenki S. 1997. Polycyclic aromatic hydrocarbons in food products originating from locally reared animals in Kuwait. Food Addit Contam. 14:295–299. Iametti S, Versuraro L, Tregna S, Giangiacomo R, Bonomi F. 1997. Surface properties of the fat globule in treated creams. Int Dairy J. 7:375–380.
International Agency for Research on Cancer. 2010. Some nonheterocyclic polycyclic aromatic hydrocarbons and some related exposures. IARC Monogr Eval Carcinog Risks Hum. 92:1–853. Kan CA, Traag WA, Hoogenboom LAP. 2003. Voorkomen van PAK’s in voer, omgeving van dieren, melk en zuivelproducten alsmede een oriënterende studie in melkvee. Animal Sciences Group Wageningen UR, Report nr. 03/ 0027745. Available from: http://library.wur.nl/way/bestanden/clc/1873114.pdf Kishikawa N, Kuroda N. 2009. Evaluation of organic environmental pollutans detected in human milk. J Health Sci. 55:1–10. Kishikawa N, Wada M, Kuroda N, Akiyama S, Nakashima K. 2003. Determination of polycyclic aromatic hydrocarbons in milk samples by high-performance liquid chromatography with fluorescence detection. J Chrom B. 789:257–264. Kuipers J, Oostdijk J, Schulz C, Ruthenschroer A. 2010. Detecting polycylic aromatic hydrocarbons in food [Internet; cited 2010 Aug]. Available from: http://www.foodquality.com/details/ article/842341/Detecting_Polycylic_Aromatic_Hydrocarbons_ in_Food.html?tzcheck=1 Lapole D, Rychen G, Grova N, Monteau F, Le Bizec B, Feidt C. 2007. Milk and urine excretion of polycyclic aromatic hydrocarbons and their hydroxylated metabolites after a single oral administration in ruminants. J Dairy Sci. 90:2624–2629. Lewis MJ, Deeth HC. 2008. Heat treatment of milk. In: Tamine AY, editor. Milk processing and quality management. Chichester: SDT, Wiley-Blackwell; p. 168–204. Lutz S, Feidt C, Monteau F, Rychen G, Le Bizec B, Jurianz S. 2006. Effect of exposure to soil-bound polycyclic aromatic hydrocarbons on milk contaminations of parent compounds and their monohydroxylated metabolites. J Agric Food Chem. 54:263–268. Mandel J. 1964. The statistical analysis of experimental data. New York (NY): Wiley & Sons. Mattison DR, Nightingale MR. 1980. The biochemical and genetic characteristics of murine ovarian aryl hydrocarbon (Benzo[a]pyrene) hydroxylase activity and its relationship to primordial oocyte destruction by polycyclic aromatic hydrocarbons. Toxicol Appl Pharm. 56:399–408. Michalski MC, Michel F, Sainmont D, Briard V. 2002. Apparent ζ-potential as a tool to assess mechanical damages to the milk fat globule membrane. Coll Surf B Biointer. 23:23–30. Moon HB, Lee DH, Lee YS, Kannan K. 2012. Occurrence and accumulation patterns of polycyclic aromatic hydrocarbons and synthetic musk compounds in adipose tissues of Korean females. Chemosphere. 86:485–490. Moroles N, Ulloa HJ, Castro-Rios R, Villarelal EG, Barbarin-Castillo JM, Salazar-Cavazos M, Waksman-De Torres N. 2007. A comparison of the performance of two chromatographic and three extraction techniques for the analysis of PAHs in sources of drinking water. J Chromatogr Sci. 45:57–62. Naccari C, Cristani M, Giofrè F, Ferrante M, Siracusa L, Trombetta D. 2011. PAHs concentration in heat-treated milk samples. Food Res Int. 44:716–724. Simoneit BRT. 2002. Biomass burning a review of organic tracers for smoke from incomplete combustion. Appl Geochem. 17:129–162. Tay-Lung C, Chiu-Jung L, Meei-Fang C. 2010. Comparison of liquid–liquid extraction and solid phase extraction for the determination of polycyclic aromatic hydrocarbons in the milk of Taiwan. J Taiwan Inst Chem Eng. 41:178–183.
