Assessment of Polychlorinated Biphenyls and Their Hydroxylated Metabolites in Postmortem Human Brain Samples: Age and Brain Region Differences

Exposure to polychlorinated biphenyls (PCBs) and their hydroxylated metabolites (OH-PCBs) has been implicated in neurodevelopmental disorders. However, the distribution of PCBs and OH-PCBs in the human brain has not been characterized. This study investigated the age-, sex-, and brain region-specific distribution of all 209 PCBs using gaschromatography–tandem mass spectrometry (GC–MS/MS) in neonatal (N = 7) and adult (N = 7) postmortem brain samples. OH-PCB analyses were performed by GC–MS/MS (as methylated derivatives) and, in a subset of samples, by nontarget liquid chromatography high-resolution mass spectrometry (Nt-LCMS). Fourteen higher chlorinated PCB congeners were observed with a detection frequency >50%. Six lower chlorinated PCBs were detected with a detection frequency >10%. Higher chlorinated PCBs were observed with higher levels in samples from adult versus younger donors. PCB congener profiles from adult donors showed more similarities across brain regions and donors than younger donors. We also assess the potential neurotoxicity of the PCB residues in the human brain with neurotoxic equivalency (NEQ) approaches. The median ΣNEQs, calculated for the PCB homologues, were 40-fold higher in older versus younger donors. Importantly, lower chlorinated PCBs made considerable contributions to the neurotoxic potential of PCB residues in some donors. OH-PCBs were identified for the first time in a small number of human brain samples by GC–MS/MS and Nt-LCMS analyses, and all contained four or fewer chlorine.


Chemicals and materials S3
Extraction procedure of PCBs and OH-PCBs from brain tissues for gas chromatographytandem mass spectrometry (GC-MS/MS) analysis S3

Targeted GC-MS/MS determination of PCBs and OH-PCBs S5
Quality assurance/quality control for the GC-MS/MS analysis S5 Chromatographic separation and instrument parameters for the Nt-LCMS analysis of OH-PCBs
S8 Figure S2. Chromatograms on the SPB-Octyl column showing the presence of possible unknown mono-to tetra-chlorinated OH-PCB metabolites in the brain from selected donors.
Chemicals and materials. 13 C-labeled PCBs ( was re-extracted with 1 mL of isopropanol and 2.5 mL of hexane and diethyl ether (9:1, vol/vol).
The combined organic extracts were washed with 5 mL of phosphoric acid (0.1 M in 0.9% aqueous sodium chloride), and the aqueous phase was re-extracted with 1 mL of hexane and diethyl ether S4 (9:1, v/v).
The combined extract, which contained both PCBs and OH-PCBs, was concentrated to near dryness under a gentle stream of nitrogen. The extract was reconstituted in 4 mL of hexane, and 2 mL of a potassium hydroxide solution (0.5 M in water-ethanol, 1:1, v/v) was added. After inversion for 5 min and centrifugation at 1,378 g for 3 min, the organic phase was separated, and the bottom aqueous phase was re-extracted with 3 mL of hexane. The combined organic phase contained the PCBs. To extract the OH-PCBs, the aqueous phase was acidified with 0.5 mL of hydrochloric acid (2 M) and extracted twice with 4 mL and 3 mL of hexane and methyl t-butyl ether (9:1, vol/vol).
This combined organic phase contained the OH-PCBs.
The PCB fraction was concentrated to ~ 0.5 mL under a gentle stream of nitrogen and passed through a glass cartridge filled with 2 g of acidified silica gel (silica gel and concentrated sulfuric acid, 2:1, w/w) with 0.2 g activated silica gel at the bottom and prewashed with 3 mL of hexane.
PCBs were eluted from the cartridge with 14 mL of hexane. The eluent was concentrated to 4 mL and treated with 2 mL of concentrated sulfuric acid. After inversion for 5 min and centrifugation at 1,378 g for 3 min, the organic phase was collected, and the sulfuric acid layer was re-extracted with 3 mL of hexane. The combined organic extract was concentrated to about 50 µL and transferred to a glass autosampler vial with an insert. The sample was spiked with the internal standards (d5-PCB 30 and PCB 204; 5 ng each in hexane) prior to analysis.
The OH-PCB fraction was concentrated to about 1 mL, and 5 drops of methanol were added.
As described previously, the OH-PCBs were derivatized with diazomethane to the corresponding methoxylated PCBs. [4][5][6] After derivatization, the OH-PCB fraction was concentrated to about 0.5 mL and passed through a cartridge as described above for PCBs, except that 14 mL dichloromethane was used as eluent. The solvent was exchanged to hexane by concentrating to 0.5 mL and diluting with 3 mL of hexane three times. The extract was treated with sulfuric acid as  Table SS1 below. 6,8 Quality assurance/quality control for the GC-MS/MS analysis. The brain samples were extracted and analyzed using a rigorous quality assurance/quality control (QA/QC) program Surrogate recovery standards (i.e., representative PCB or OH-PCB congeners) were spiked into every sample immediately prior to extraction for the primary purpose of assessing and correcting for analytical losses during sample workup and assess the precision and reproducibility of the extraction across the entire study. PCB surrogate standards included the ten 13 C-labeled mono-to deca-chlorinated PCBs (one from each Cl homolog) listed in Table S3. Average PCB recovery rates, including the range of recoveries and relative standard deviation for each standard, are provided in Table S3. OH-PCB surrogate standards were selected based on their commercial availability and included seven 13 C-labeled OH-PCBs from the di, tri, tetra, penta, hexa, and two hepta chlorinated homologs, see Table S3. Average OH-PCB recovery rates, including the range of recoveries and relative standard deviation for each standard, are provided in Table S3.
Based on our earlier studies, deuterated PCB 30 and PCB 204 were selected as representative congeners to normalize instrument response for PCB and MeO-PCB congeners to concentrations in the calibration standard. These internal quantification standards were spiked into every sample immediately prior to instrument analysis.
Standard Reference Materials SRM 1957 (organic contaminants in non-fortified human serum, National Institute of Standards and Technology) was used to assess the accuracy. The SRM was selected because no suitable SRM or laboratory reference material is currently available for human brain tissues. A comparison of the measured and certified PCB and published OH-PCB levels in this SRM is provided in Table S6. Moreover, a SRM 1957 sample was extracted with every sample S7 batch to evaluate the reproducibility of the analysis. The results from all SRM samples extracted as part of this study are also summarized in Table S6.   Table S1. Amy: amygdala, BA19: Broadman area 19, CTX: cortex, PFC: prefrontal cortex.    Figure S8. An unknown tetrachlorinated OH-PCB was detected by GC-MS/MS and Nt-LCMS Orbitrap in the amygdala from a 1-day old female donor (donor 2018-13, see Table S1). (A) GC-MS/MS chromatogram with MRM transition (m/z) of 321.9 → 278.9 showing a peak corresponding to a tetrachlorinated OH-PCB (as methoxylated PCBs) in the extract from an amygdala sample from this donor. The method blank, and reference standard are shown for comparison. Nt-LCMS Orbitrap analysis of the same brain region from the same donor also showed a single tetrachlorinated OH-PCB peak, indirectly confirming the presence of a tetrachlorinated OH-PCB in this brain sample.     The short names of the MeO-PCB congeners are based on the PCB congener table of the US EPA. The number indicating the position of the methoxy group is based on the numbering scheme of the respective PCB congener, as proposed previously. 33