Methods
Friends of the Earth conducted these peer-reviewed studies in collaboration with researchers at University of California at Berkeley, University of California at San Francisco, Commonweal Biomonitoring Resources Center, and Health Research Institute. We wanted to know whether eating an organic diet could reduce levels of detectable pesticides and pesticide breakdown products in participants’ bodies.
We secured oversight for protection of human subjects through Western Institutional Review Board. The study was published in the journal Environmental Research.
We chose four families in Atlanta, Baltimore, Minneapolis and Oakland to participate in the study. These families reported that they did not typically eat an organic diet. In total, 16 people participated in the study.
The study lasted 12 days. During the first phase, each family ate only conventional (non-organic) foods and beverages. During the second phase, each family ate only organic foods and beverages. Each participant collected a daily urine sample during the 12 days of the study and kept a food diary.
To ensure that the families were able to eat a completely organic diet, organic food was provided to them in two ways: 1) participants were asked to compile a list of all groceries they would need for a week, and research assistants purchased organic foods from this list and delivered the groceries to the participants’ homes; and 2) dinners during the organic week were prepared with all organic foods by a licensed chef or caterer and delivered to study participants by the research assistants. All organic foods were provided free of charge to the families.
The samples were analyzed for by laboratories at the University of California at San Francisco, Québec National Institute of Public Health, and Health Research Institute.
For a full discussion of methods, see Organic Diet Intervention Significantly Reduces Urinary Pesticide Levels in U.S. Children and Adults in Environmental Research, a peer-reviewed scientific journal.
University of California at San Francisco Lab Methodology
Urine samples were analyzed for 18 pesticide analytes, including nine specific organophosphate, neonicotinoid, pyrethroid, fungicide and herbicide analytes and three non-specific pyrethroid analytes.
Quantification of the nine pesticide-specific analytes and three non-specific pyrethroid analytes was performed using liquid-chromatography-tandem mass spectrometry (LC-MS/MS) on an Agilent LC 1260-AB Sciex 5500 system. The urine specimens (1 mL) were prepared for LC-MS/MS analysis by solid phase extraction (SPE) using Waters Oasis WAX cartridges (10mg, 30um, 1cc). The compounds were ionized in the negative mode using electrospray ionization (ESI) and monitored by multiple reaction monitoring. Each compound was monitored using two transitions (see Supplementary Information S1) along with 2,4-D-d3 and Cloth-d3 as internal standards). Each batch of samples was injected in duplicates and run alongside calibration standards that were run at the beginning, between the duplicate sample injections and after the sample set. Additionally, low and high spiked quality control (QC) samples were run for each analysis batch. A batch run was accepted if both QC samples were within 20 percent of their target values and had coefficients of variation (CV) ≤ 20 percent.
Québec National Institute of Public Health
Urine samples were analyzed for six non-specific OP dialkyl phosphate (DAPS) analytes.
To quantify the six non-specific OP pesticide DAP metabolites (DEP, DETP, DEDTP, DMP, DMTP and DMDTP), 0.1 mL of the urine specimens were enriched with internal standards (DEP-13C4, DETP-13C4, DEDTP-13C4, DMP-d6, DMTP-d6, DMDTP-d6). After adding 1 mL of acetonitrile and 200 mg of potassium carbonate, the samples were derivatized with 15 µL of pentafluorobenzyl bromide (PFBBr) at 70°C for 2 hours. The derivatized products were extracted with 7 mL of a mixture of dichloromethane:hexane (8:92), mixed 15 minutes and centrifuged 5 minutes at 1500 rpm. The solvent was then evaporated to dryness, taken up in 500 µL of dichloromethane:hexane (20:80) and analyzed for pesticide metabolites on an Agilent 6890 Network gas chromatograph (GC) (Agilent Technologies; Mississauga, Ontario, Canada) coupled to a Waters Quattro Micro GC mass spectrometer in tandem (MS/MS) (Waters; Milford, Massachusetts). The GC was fitted with an Agilent 30 m HP-5MS column (0.25 mm i.d., 0.25 µm film thickness) to the MS/MS. The internal reference materials used to control the quality of the analyses were the non-certified reference material ClinChek (Urine Level 1; RECIPE Chemicals; Munich, Germany) and three in-house reference materials (low, medium, high QC) prepared by the Centre de Toxicologie du Québec (CTQ), Institut National de Santé Publique du Québec (INSPQ). The overall quality and accuracy of the analytical method was monitored by the interlaboratory program as the German External Quality Assessment Scheme (G-EQUAS; Erlangen, Germany).
The labs tested for the following chemicals and did not detect them in any participant urine samples: 5-Hydroxy-Imidacloprid, 5-OH-Thiabendazole (a breakdown product of thiabendazole, used in agriculture as a fungicide), Iprodione (a fungicide and nematicide), and Boscalid (a fungicide). This could mean that participants were not exposed to these chemicals. However, all of the pesticides these chemicals represent are used in significant quantities and are frequently found as residues on food (see USDA Pesticide Data Program.) Because the science of studying pesticides in people’s bodies is still evolving, it could also mean that the methods the labs used were not sensitive enough to detect these chemicals or that the specific chemicals studied are not the correct compound to look for. For example, this study found the neonicotinoid pesticide clothianidin in every participants’ urine but did not find the neonicotinoid breakdown product for imidacloprid (5-Hydroxy-Imidacloprid) despite the fact that imidacloprid is far more widely used in agriculture.
Health Research Institute Lab Methodology
Analysis of glyphosate and AMPA in urine specimens was performed at Health Research Institute, Fairfield, IA using an isotope dilution methodology accredited according to ISO/IEE 17025 and to the US EPA CLIA program. Separations were carried out using a Shimadzu Nexera X2 ultra high-performance liquid chromatograph (Shimadzu Scientific Instruments, Columbia, MD, USA) linked to a Bio-Rad (Hercules, CA, USA) Cation-H guard column (30 mm by 4.6 mm). Mass spectrometry was carried out using a QTRAP 5500 triple quadrupole instrument from AB Sciex (Framingham, MA, USA).
Quantification was carried out using a previously described method (Jensen et al., 2016) modified to achieve greater sensitivity. The modifications lowered the limit of quantification for glyphosate and AMPA from 0.100 to 0.050 ng/ml and the limit of detection from 0.023 to 0.020 ng/ml for glyphosate and from 0.033 to 0.013 ng/ml for AMPA. In brief, isotopically labeled internal standards of glyphosate (13C,15N) (Cambridge Isotope Laboratories, Andover, MA, USA) and AMPA (13C, 15N, D2) (Sigma-Aldrich, St Louis, USA) were added to urine samples, which were also adjusted to 0.05% formic acid. Samples were centrifuged at 14,800 rpm for 10 min to sediment particulates and were transferred to polypropylene vials for LC-MS/MS analysis. Samples were corrected for dilution using specific gravity using a BlueTooth enabled refractometer (ATAGO, Tokyo, Japan).
Quality control (QC) procedures included the following: (1) analysis of certified reference material (LGC, Lancashire, UK)) diluted into control urine at moderate concentrations (~0.4 ng glyphosate/ml and ~0.80 ng AMPA/ml) at the beginning and end of the run, and at low concentrations (~0.04 ng glyphosate/ml and ~0.08 ng/ml AMPA) at the beginning and end of each run and following every 10 samples during the run, with ± 20% agreement between measured and declared values; (2) matrix-matched calibration curve for glyphosate and AMPA (certified reference materials from Sigma-Aldrich, St. Louis, USA) from 0.025 ng/ml to 50.0 ng/ml placed at the beginning of each run, with a repeat of the 2.5 and 0.25 ng/ml points at the end of the run, with ± 20% agreement between measurements; (3) duplicate analysis of 10% of samples and repeat of analyses in cases where duplicates diverged by more than 30%; (4) verification that the raw counts of the highest and lowest points in the calibration curve met criteria for between-run consistency for both glyphosate and AMPA.
For samples with glyphosate and AMPA concentrations below the LOD, concentrations were set at the LOD divided by the square root of 2 (Hornung and Reed, 1990). For samples with glyphosate and AMPA concentrations below the LOQ but above the LOD, concentrations were set at 50% of the LOQ.