Appendix 1: Materials & Methods
Materials & Methods section of the Pollution in People report
The ten participants in our sample were selected for diversity in occupations, geography, age, gender, and race. Each participant was asked to complete an exposure assessment questionnaire and provide information about their residences, occupations, diet, and potential toxic exposures. Samples were taken primarily in September and October of 2005. Nurses and a certified phlebotomist collected blood samples into vacutainers; after clotting, serum was obtained by centrifuging tubes and pouring off or pipetting serum into storage vials. One vacutainer of whole blood was maintained for each participant for lead testing. Participants provided first morning void urine samples for phthalate testing and mid-morning urine samples for pesticide and arsenic testing. Hair samples for mercury analysis were cut from the base of the scalp (or beyond, if necessary); in one case, chest hair was collected due to insufficient head hair.
PFC, PBDE, and Phthalate Analysis
AYXS Analytical Services (Victoria, BC) analyzed serum samples for perfluorinated compounds and PBDEs and urine samples for phthalates. Below are the laboratory’s methods, in brief.
PBDEs: EPA Method 1614, an HRGC/HRMS method using an isotope dilution internal standard quantification. The laboratory reported blank-corrected levels for BDE-209 due to detections in blank samples.
Phthalate monoesters: Each separate urine sample was spiked with a suite of 13C mono-ester surrogate standards and with an enzyme to release the mono-esters from their glucoronated form. The sample was extracted on NEXUS SPE cartridges, eluted, and analyzed by liquid chromatography tandem mass spectrometry (LC/MS/MS). The laboratory reported blank-corrected levels for MEHP due to detections in blank samples.
PFCs: For each individual, a 2 mL serum sample was homogenized with 0.25M sodium carbonate and 0.5M tetrabutylammonium hydrogen sulfate to disrupt cells and form respective ion pairs. The aqueous homogenate was then extracted with methyl tert-butyl ether (MTBE). The MTBE solution was reduced to dryness and reconstituted in 50:50 water/methanol. Analysis was performed using LC/MS/MS with a Genesis C8 (2.1 mm x 50 mm, 4 µm) chromatographic column. Quantification was performed by standard curve using perfuorononanoic acid or perfluorododecanoic acid as the internal standard.
Pesticide, Arsenic, Lead, and PCB Analysis
Pacific Toxicology Labs (Los Angeles, CA) and National Medical Services (Willow Grove, PA) analyzed urine samples for organophosphate pesticides, 1-napthol, 2,5,6-TCP, 2,4-D, and arsenic; serum for organochlorine pesticides and PCBs; and whole blood for lead. Below are the laboratories’ methods, in brief.
Organochlorine pesticides: A hexane extract of serum was concentrated and analyzed by electron capture capillary gas chromatography using the internal standard method.
PCBs: The Webb-McCall method was used, in which PCBs were extracted from de-proteinized serum with 1:1 hexane/ethyl ether. PCBs were separated from organochlorine pesticides and biogenic material by chromatography on silica gel using hexane as eluent. PCB concentrations in the eluent were determined by electron capture gas chromatographic analysis using Webb-McCall mean weight percent factors and the internal standard method.
Organophosphate pesticide metabolites: Freeze-dried urine samples were derivatized with a benzyltoyltriazine reagent to produce benzyl derivatives of alkylphosphate metabolites. A saturated salt solution was added to the tubes and the benzyl derivatives were extracted with cyclohexane and analyzed by gas chromatography with flame photometric detection.
Arsenic: Arsenic was reduced to the trivalent form with potassium iodide and extracted into toluene. After a back-extraction into aqueous HNO3, the arsenic was measured by graphite furnace atomic absorption spectophotometry.
Lead: Lead in whole blood was measured on graphic furnace atomic absorption spectroscopy. Blood samples were mixed with antifoam agent and Triton X-100 prior to analysis. The sample was atomized at 2800º C in a graphite tube and light absorbed was measured by passing monochromatic light from a cathode tube. A calibration curve was generated from known standards and the values for unknown samples were extrapolated.
3,5,6-TCP and 2,4-D: These compounds were measured in urine by gas chromatography/mass spectrometry (GC/MS). The samples were hydrolyzed in acid to release the compounds from conjugates. After extraction in butyl chloride, the compounds were derivatized to form trimethylsilyl derivatives prior to analysis in the GC/MS in the SIM mode. A calibration curve was generated from known standards.
Two samples were analyzed by National Medical Services due to changes in procedures at Pacific Toxicology . For 3,5,6-TCP, urine samples were made acidic with HCl and heated to convert any conjugated compound into the un-conjugated form. Samples were extracted with an organic solvent that was tested on an Agilent 5890 gas chromatograph with an electron capture detector. The analytical column was a 30-meter x 0.32 mm ID with a 0.25 micron DB-1 film, and it ran on a temperature program from 100ºC to 300ºC at a rate of 20ºC per minute. Calibrators at concentrations of 50, 100, 500 and 1000 ppb and controls at 75 and 750 ppb were run to calibrate and control each analytical batch.
For 2,4-D, urine samples were made acidic with HCl and extracted with an organic solvent, back-extracted into a sodium hydroxide solution that was made acidic and extracted with an organic solvent. Extractions were analyzed on an Agilent 6890 gas chromatograph with an electron capture detector. The analytical column was a 30-meter x 0.32 mm ID with a 0.25 micron DB-5 film, and it ran on a temperature program from 140ºC to 280ºC to at a rate of 10ºC per minute. Calibrators at concentrations of 20, 50, 100, 200, 400, and 600 ppb, and controls at 50 and 300 ppb were run to calibrate and control each analytical batch.
Carbamates: Carbaryl in urine was measured by GC/MS as the metabolite 1-napthol. The samples were hydrolyzed in acid to release the compounds from conjugates. After extraction in butyl chloride, the compounds were derivatized to form trimethylsilyl derivatives prior to analysis in the GC/MS in the SIM mode. A calibration curve was generated from known standards.
Again, two samples were analyzed by National Medical Services due to changes in procedures at Pacific Toxicology. A 0.5-mL aliquot of urine was buffered with saturated ammonium chloride and an internal standard (methyl-phenobarbital) was added. This was extracted with a mixed organic solvent that was evaporated to dryness and reconstituted with the HPLC mobile phase. Reconstituted extractions were analyzed on a Zorbax Stable Bond C-18, 4.6 x 150 mm analytical column with an Optimize 3mm cartridge C-18 precolumn. The mobile phase consisted of 24% CH3CN in 0.05 M KH2PO4, 0.007 M 1-heptanesulfonate sodium, and 0.01 M triethylamine (adjusted to pH 3.0 with H3PO4). Detection was by UV at 210 nm. Calibrators for both carbaryl and 1-naphthol at concentrations of 100, 400, 1000, and 4000 ppb, and controls at 400 and 2000 ppb, were run to calibrate and control each analytical batch.
Mercury analysis in hair was conducted by Brooks Rand (Seattle, WA). Hair samples were homogenized by cutting the hair into small pieces. The hair was transferred to clean glass jars. Sufficient acetone to cover each sample was added, and the jars were capped and shaken for 30 seconds. After allowing the hair to settle, the acetone was pipetted off. The samples were then cleaned three times with Triton TX-100 washes, followed by rinses with deionized water and filtration. The hair samples were then transferred to clean, dry, 40 mL glass vials and dried in a 40º C oven overnight. Two aliquots of quartz wool were homogenized according to the same procedure and analyzed along with the samples. Homogenized and cleaned hair samples were prepared and analyzed in accordance with the Appendix to EPA Method 1631. Hair samples were digested with a nitric/sulfuric acid solution and further oxidized with bromine monochloride. All samples were then analyzed with stannous chloride reduction, gold amalgamation, and cold vapor atomic fluorescence spectroscopy (CVAFS) detection using a BRL Model III CVAFS Mercury Analyzer. All sample results for low-level mercury analysis were blank corrected.
For phthalates, PFCs, PBDEs, DDE, and PCBs, medians were calculated setting non-detectable values at the detection limit divided by the square root of two. Total PBDEs were calculated in the same manner. Medians were not calculated for arsenic, lead, organophosphate pesticides, or carbaryl because of the relatively high number of participants with undetectable levels. The influence of diet on pesticide levels was analyzed by computing the Spearman correlation between (1) the number of organophosphate and/or carbaryl metabolite detections per week and (2) the number of organic meals per week. Spearman’s correlation coefficient was calculated, and it was significantly different from zero at p=0.028, indicating a statistically significant correlation.