ABSTRACT: Analyzing the organochlorine pesticide (OCP) content in milk is relevantbecause it helps evaluate the quality of milk that reaches the consumer's table and also helps identify the geographical areas where there is a high possibility of contamination. Accordingly,thispilot project was aimed at determining the extent of contamination by OCP residues and their metabolites in fresh cow's milk in Mojuí dos Campos and Belterra, Pará, using solid-phase extraction (SPE) and gas chromatography coupled with electron capture detection (GC-ECD). The OCPs evaluated in this study weredichlorodiphenyltrichloroethane (DDT) and its metabolites (dichlorodiphenyldichloroethylene (DDE) and dichlorodiphenyldichloroethane (DDD)), hexachlorocyclohexane (HCH) (α, β, γ, and δ), endosulfanα and β, andendosulfansulfate. None of the thirty bovine milk samples analyzedshowed any contamination with OCPs.TheOCP content was within the limit of quantification of the method. The results of this study add to the existing knowledge on the quality ofbovine milk produced in these locations. However, further researchon other environmental matricesis required to confirm the results obtained in this study.
Keywords: Contamination, DDT, SPE, GC-ECD, lipids.
ANIMAL PRODUCTION
Levels of organochlorine contaminants in natural bovine milk marketed in Mojuí dos Campos and Belterra, Brazil
Received: 30 November 2021
Accepted: 31 March 2022
Pesticides are used to control pests and pathogenic organisms that compromise agricultural production. However, improper application of pesticides is responsible for environmental contamination and associated public health hazards. These substances reach humans via different environmental routes and could pose health issues (Laabs, Amelung, Pinto, & Zech, 2002, Flores, Ribeiro, Neves, & Queiroz, 2004, Alves, Antoniosi Filho, Oliveira, & Furtado, 2010). Among pesticides, organochlorine compounds (OCPs) were introduced to improve agricultural production. However, OCPs persist for a long time in the environment and are highly liposoluble. This causes indirect contamination by OCP residues in food, mainly those of animal origin, via the contaminated soil, air, and water (Mello & Silveira, 2012; Polder, Skaare, Skjerve, Loken, & Eggesb, 2009; Azeredo et al., 2008; Wilson & Tisdell, 2001). Despite the prohibition on theiruse for agricultural purposes since the 1980s, OCPstend to remain in places where they have been banned for many years or even where they have never been used, because they can be carried by atmospheric transport and spread by contaminated water or even food from other regions where their use is allowed (Çok et al., 2012; Azeredo et al., 2008; Zhong, Xu, Chai, & Mao, 2003).
OCPs include chlorinated diphenylmethane derivatives, such asdichlorodiphenyltrichloroethane (DDT) and its metabolites (dichlorodiphenyldichloroethylene (DDE), dichlorodiphenyldichloroethane (DDD), and methoxychlor, as well as hexachlorobenzene (BHC), hexachlorocyclohexanes (α-HCH, β-HCH, γ-HCH, and δ-HCH), cyclodienes (aldrin, dieldrin, endrin, chlordane, nonachlor, heptachlor, and heptachlor-epoxide), and chlorinated hydrocarbons (dodecachlor, toxaphene, and chlordecone) (D’Amato, Torres, & Malm, 2002).Foods rich in animal fat are potential sources of these persistent contaminants that bioaccumulate via the food chain (Guan et al., 2009; Gasull et al., 2010; Tsiplakou, Anagnostopoulos, Liapis, Haroutounian, & Zervas, 2010, Rêgo et al., 2019).
Milk is essential for a healthy and balanced diet and is rich in mineral salts (mainly calcium salts), lipids, carbohydrates, proteins, and vitamins (Pereira, 2014). However, milk can be contaminated by pesticides, primarily via pastures and feed exposed to such compounds, and by the improper use of veterinary products that defy good agricultural practices.
In the 1970s, the northern territory was developed by the Brazilian government, including road construction, habitation, and land-use projects, mainly along the Cuiabá-Santarém Highway (BR-163) and Belém-Brasília Highway (BR-153) (Walker, Defries, Vera-Diaz, Shimabukuro, & Venturieri, 2010). Agricultural production, especially that of soy and corn, in these regions has in creased (Walker et al., 2010). However, in the municipalities of Santarém and Belterra, State of Pará, no mechanized production of soy was reported before 1990 (Barros, Lopes, Laurent, & Coelho, 2020). The introduction of crops that are not well adapted to local conditions predisposes them to compete with native vegetation, and also in creases their susceptibility to fungi, insects, and other pests. This situation forces farmers to use a wide variety of pesticides, including both allowed and banned substances, such as OCPs, which are still found in clandestine markets (Waichman, 2008; Flores et al., 2004).
Torres et al. (2002) detected low concentrations (average value of 100 μg kg-1) of DDT in soil sourced from the Amazon region. This was attributed to its rampant use in sanitary campaigns to combat malaria. The continuous quantification of OCPs in the soil is essential for monitoring contamination sites and tracking the illegal use of these compounds (Villa, Oliveira, & Nogueira, 2011). The presence of pesticide residues and their metabolites in milk poses a risk to consumer health. Bovine milk has been used in some countries as an indicator of the presence and persistence of chemical substances in agriculture and the environment (Lemes, Kussumi, & Rocha, 2004, Luzardo et al., 2012; Avancini, Silva, Rosa, Sarcinelli, & Mesquita, 2013, Rêgo et al., 2019).The main objective of this study was to determine the contamination level of OCPs and their metabolite residues, as well as the fatty acid content in fresh cow's milk from Mojuí dos Campos and Belterra, Pará, via solid-phase extraction (SPE) and gas chromatography coupled with electron capture detection(GC-ECD).
All the solvents used were of HPLC grade, and other reagents, viz. acetone, acetonitrile, n-hexane, methanol, and ethyl acetate (J.T. Baker, Aparecida de Goiânia, Brazil), Milli-Q® water, Extran® MA Neutral 1% (Merck, Darmstadt, Germany), were of high purity.
Thirty samples of fresh commercial bovine milk were collected, 15 from Mojuí dos Campos (02º10'17”S and 56º44'42”W) and 15 from Belterra (02º38'11”S and 54º56'14”W), Brazil (Figure 1), between September and November 2017. It is noteworthy that the dairy herds were relatively small (no more than 20 animals), and these farms were in the vicinity of crop production areas that mainly practiced monoculture (soy and corn).
The samples were thawed in a water bath at 37°C for 20 min., and 1 mL aliquots were placed in graduated tubes. Subsequently, 10 mL of 2:4:4 (v/v) ethyl acetate/methanol/acetone solution was added to each sample. The resulting mixture was homogenized for 1 min in a vortex (Maximix II, Thermo lyne Type 37600, São Paulo, Brazil) and then subjected to ultrasound (Branson® 5210, Chicago, USA) for 20 min. The mixture was then centrifuged (Excelsa Baby II centrifuge, Model 206-R; Nova tecnica, Piracicaba, Brazil) for 15 min. at 2000 rpm. To the supernatant, 10 mL of ultrapure water was added, and the mixture was then passed through a preconditioned (2×1 mL each of n-hexane, ethyl acetate, methanol, and ultrapure water) SPE-PACK® C18 column (Sigma-Aldrich, St. Louis, USA). The mixture was passed through the column at a flow rate of approximately 6 mL min-1. Finally, the column was washed twice with 1 mL of an aqueous solution of 25% acetonitrile, after which the column was vacuum-dried, and the OCPs were eluted with 2 mL of n-hexane.
The elute from the C18 column was separated into two 1-mL aliquots; one aliquot was used to determine the lipid content of the sample, and the other was passed through a Florisil® column preconditioned with 10 mL each of dichloromethane, ethyl acetate, and 15% acetone/ethyl acetate solution. The pesticides were then eluted with 10 mL n-hexane and 5 mL acetone solution in 15% ethyl acetate. The total eluted volume was reduced to 1 mL using a low-pressure rotary evaporator (Rotavapor R144, Büchi Switzerland; Integrator HP 7673A). Subsequently, 100 μL of the internal standard, DCN (1,2-dichloronaphthalene), was added at a concentration of 108 ng mL-1.
Since organochlorine compounds are lipophilic, they may be concentrated in the lipid fraction, making it important to determine the fat content of the milk samples. Fatty content analysis was performed during the final phase of the sample extraction. After the pesticides retained in the column were eluted with 2 mL n-hexane, 1 mL of the eluate was dried in a rotary evaporator under low pressure (Rotavapor R144, Büchi, Switzerland; Integrator HP 7673A) to determine the lipid content. All analyses were repeated in triplicates.
For quantitative analysis, a GC-ECD (Varian CP-3800; USA) with Workstation 5.0 (software for chromatographic data proce5ssing) was used, with a fused silica capillary column of 30 mm length, 0.32 mm ID, and 0.25 µm film thickness (OV-5, Ohio Valley Specialty Chemical, OHIO, USA). The temperature ramp from the oven to the column was programmed as follows: 150°C for 1 min. and 150 to 250°C (6°C min.-1) for 5 min. The carrier gas was N2 (99.999% pure) at a flow rate of 1.2 mL min.-1. The injector was operated at 250°C in the split less mode. The detector temperature was 300°C. The qualitative analysis for the confirmation of the identified analytes was performed using a gas chromatograph coupled to a mass spectrometer (GC-MS; TSQ 8000 Thermo Scientific MS 3000; USA) with a triple-quadrupole mass analyzer operating in full-scan mode, equipped with Workstation software for processing the chromatographic data, using a fused silica capillary column (DB 05) of 30 mm length, 0.32 mm ID, and 0.25 µm film thickness. The temperature ramp from the oven to the column was programmed as 80°C for 3 min. and 80 to 280°C (10°C min.-1) for 3.5 min. The carrier gas was helium (99.999% pure) set at a flow rate of 1.0 mL min.-1. The injector was operated at 250°C in the splitless mode. The temperature of the quadrupole and interface was 260°C, and that of the ion source was 275°C.
To identify the analytes by GC-ECD, a standard solution containing all analytes was injected, and their respective retention times were determined. For quantification, an external calibration curve was constructed, and quantification was based on the integrated areas of each analytical signal (chromatographic peaks). Calibration curves were constructed to quantify each substance, with the injection of diluted solutions (2-100 ng mL-1) of the analyte mixture. A Pearson's correlation index of 99% (r=0.99) was set to fit the calibration curve. Selectivity was analyzed according to the analytical signal resolution, which is related to the retention times (RTs) of the compounds. The existing resolution between various compounds was evaluated by analyzing the peak areas. The RTs were obtained from the largest peak areas present in the chromatograms of the replicates of the standard pesticide solutions. Only tenanalytes were separated without coelution (Table 1). By this method, the OCPs that could be separated despite the proximity of the RTs were α-HCH (7.7 min.), β-HCH (8.5 min.), γ-HCH (Lindane) (8.7 min.), δ-HCH (9.6 min.), op’-DDE (13,7 min.), endosulfan α (14.0 min.), pp'-DDE (14.7 min.), op'-DDD (14.9 min.), endosulfan β (15.9 min.), and pp'-DDD (16.0 min.). Endosulfan sulfate and op'-DDT coeluted at 17.0 minutes each (Figure 2). The percentage recovery was calculated by processing a sample of bovine milk with 20 ng of each OCP.
The average fat content (%, m/m) in the milk samples collected from Belterra and Mojuí dos Campos were 2.23 and 2.27%, respectively. Both samples showed significant variations in the fat content, ranging from 0.20 to 8.20% in the samples from Mojuí dos Campos and from 0.46 to 4.91% in the samples from Belterra ((Table 2). The fat content of fresh bovine milk typically varies from 3.5 to 6.0% (Fangmeier, Helfenstein, & Oliveira, 2015). The differences in the lipid content within each group may be due to the varying milk composition, which is affected by lactation periods, feed, management, etc. The fat content of milk is associated with several factors, including food, climate, and the variety of cattle. Durr (2004) reported that the presence of trans-fatty acids in the diet or produced in the cow’s rumen dramatically decreases the fat content of milk. This decrease occurs when there is a reduction in the ruminal pH. It is noteworthy that the water (pH 4.4-5.6) and soil (pH 4.9-5.2) in the region of Belterra and Mojuí are acidic (Aguiar, Peleja, & Sousa, 2014; Silva, Silva, Veloso, Dantas, & Sacramento, 2018; Sacramento et al., 2019).
The limits of quantification associated with this method for the evaluated OCPs varied from0.03 to 0.05 ng g-1 (Table 3). The study of linearity involves constructing calibration curves for varying standard concentrations as per the concentration range determined for each pesticide. The calibration curve allows parameters, such as the correlation coefficient (r), to be obtained (Paschoal et al., 2008; Barbosa, 2012). Accordingly, linearity was assessed using the internal calibration curve method in the range from 2 to 100 ng mL-1 of the standard solution containing all the analytes under study. The Pearson correlation index for accepting the calibration curve was 99% (r = 0.99), implying that the internal calibration curve showed analyte signals proportional to the signal areas. The accuracy of the chromatographic method was evaluated by analyzing the analyte recovery and matrix effects. This method is used to assesses the extraction yield, i.e. the mass fraction of the analyte in the sample present in the final extract. For the analysis of organic compounds, the accuracy was measured using recovery tests.
The recovery of analytes is defined by the analysis of samples fortified with known quantities, but the parameter generally depends on the concentration; therefore, it must be evaluated within the range of concentrations expected for the sample (Kurz, 2007; Leite, 2011). The recoveries obtained for all the OCPs ranged from48 to 91% (Table 4). Complex matrix recoveries of around 70-130% are expected, implying that some samples showed lower than desirable recovery values, possibly due to losses during the sample preparation. Unfortunately, the recovery of six non-polar analytes could not be evaluated, perhapsdue to the polarity of the solvent (methanol) used during the elution stage of the SPE. Owing to the low polarity of these compounds, the solvent may not have enough strength (polarity) to elute them from the C18octadecyl silica SPE Pack® column cartridge, thus retaining them in the cartridge (Halimah et al., 2008).
None of the samples analyzed showed any contamination by the OCPs within the limits of detection and quantification of the method. This study contributes to the knowledge regarding the quality of fresh bovine milk from the municipalities of Mojuí dos Campos and Belterra. The results of this work were compared with other similar studies carried out in different places across the world where DDT and other OCPs were widely used. Most samples from such locations showed contamination by OCPs (Table 5).
The results of this study are similar to that from the São Paulo State (Lemes et al., 2004); however, it differs from those conducted at Rio Grande do Sul and Mato Grosso do Sul by Heck et al. (2007) and Avancini et al. (2013), respectively. Further, the results of this work is also unlike those reported by John et al. (2000) in India, Zhong (2003) in China, Darko et al. (2008) in Ghana, Hernández, Vidal, and Marrugo (2010) in Colombia, Kampire et al. (2011) in Kampala, Gutiérrez, et al. (2012) in Mexico and Luzardo et al. (2012) in Spain. Lemes et al. (2004) did not detect any OCPs or any of their metabolites in 73 samples of commercial bovine milk in the state of São Paulo. Avancini et al. (2013) evaluated organochlorine compounds in bovine milk in the state of Mato Grosso do Sul, and of the total samples analyzed, more than 90% were contaminated with at least one OCP residue. In Rio Grande do Sul, Heck et al. (2007) analyzed samples of fresh pasteurized and ultra-pasteurized (or ultra-high-temperature pasteurized) milk. In all theanalyzed samples, α-HCH, lindane, aldrin, HCB, pp'-DDE, op'-DDD, pp'-DDD, and op'-DDT), and PCB (congeners 10, 28, 52, 138, and 180) were found.
None of the bovine milk samples analyzed in this study showed any contamination by OCPs. This may be due to the absence of pesticide residues in the feed consumed by the cattle. Although the results obtained indicate no OCP contamination in bovine milk, more studies are required for a more comprehensive understanding of the OCP contamination in the studied region. It is important to highlight the relevance of milk-monitoring studies in assessing the presence of pesticide residues in foodstuff.
This work was supported by Project 88881.159143/2017-01 funded by the Coordination of Improvement of Higher Education Personnel (CAPES) and the Amazonia Foundation for Studies and Research in Pará (FAPESPA). The authors also thank two anonymous reviewers for their constructive comments and suggestions. We also would like to thank Editage (www.editage.com) for English language editing.
* Author for correspondence. E-mail: pstjunior@yahoo.com.br