Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Oct 1.
Published in final edited form as: Environ Toxicol Pharmacol. 2024 Sep 19;111:104567. doi: 10.1016/j.etap.2024.104567

Significant Metabolic Alterations in Mouse Dams Exposed to an Environmental Mixture of Polychlorinated Biphenyls (PCBs) During Gestation and Lactation: Insights into PCB and Metabolite Profiles1

Xueshu Li 1, Youjun Suh 2, Rebecca Wilson 3, Pamela J Lein 3, Julia Y Cui 2, Hans-Joachim Lehmler 1
PMCID: PMC11499005  NIHMSID: NIHMS2025488  PMID: 39305941

Abstract

Polychlorinated biphenyls (PCBs) and metabolites are linked to developmental neurotoxicity, but their levels in the gestational and lactational environment remain unexplored. This study investigated the effects of dietary exposure to the Fox River Mixture (FRM) on serum levels of PCBs and their metabolites in female C57BL/6J mice. Mice were exposed to 0.1, 1.0, or 6.0 mg/kg body weight/day of FRM beginning two weeks before mating and throughout gestation and lactation. Serum samples collected from the dams at weaning were analyzed using gas chromatograph-mass spectrometry and nontarget liquid chromatography-high resolution mass spectrometry. Results showed complex and dose-dependent differences in PCB and metabolite profiles. Untargeted metabolomics revealed alterations in metabolites involved in glucuronidation. Network analysis suggested disturbances in heme and amino acid metabolism associated with higher chlorinated PCBs. These findings suggested that PCBs and metabolites present in the gestational and lactation environment of mice may contribute to developmental neurotoxicity in rodents.

Keywords: Polychlorinated biphenyls, Untargeted metabolomics, Metabolites, PCB metabolism, Exposure, Pathway

Graphical Abstract

graphic file with name nihms-2025488-f0001.jpg

1. Introduction

Polychlorinated biphenyls (PCBs) are a group of 209 synthetic organic chemicals manufactured for various industrial purposes, such as transformers, capacitors, hydraulic fluids, and plasticizers, due to their chemical stability, non-flammability, and insulating properties (Montano et al., 2022; Othman et al., 2022; Zarerad et al., 2023). Although commercial PCB production is banned worldwide, they are still generated as unintentional industrial byproducts (EPA, 2024). PCBs are widespread environmental pollutants and accumulate in the food chain (Montano et al., 2022; Othman et al., 2022; Zarerad et al., 2023). Exposure of the general population to PCBs occurs through ingestion of contaminated food and inhalation of contaminated air (Ampleman et al., 2015; Saktrakulkla et al., 2020). The International Agency for Research on Cancer classifies PCBs as human carcinogens (IARC, 2017). Other adverse health effects of PCBs include suppression of the immune system, disruption of the reproductive and endocrine systems, and adverse neurotoxic outcomes, such as neurodevelopmental toxicity (IARC, 2017; Montano et al., 2022; Pessah et al., 2019; Zarerad et al., 2023).

PCBs are oxidized by cytochrome P450 enzymes to hydroxylated PCBs (OH-PCBs) that are further transformed to di-hydroxylated derivatives, dechlorinated OH-PCBs, methyl sulfone PCBs (MeSO2-PCBs), and the corresponding sulfates, glucuronides, and methoxylated metabolites (Grimm et al., 2015). PCB metabolites have been detected in serum from human and animal studies (Duffel and Lehmler, 2024; Grimm et al., 2015; Quinete et al., 2017). Except for MeSO2-PCBs, the metabolites of PCBs are more polar and water-soluble, which facilitates their elimination from the body. Despite their increased aqueous solubility, some OH-PCBs are retained in the serum of mammals and humans (Bergman et al., 1994; Haraguchi et al., 2004). Like the parent compounds, PCB metabolites are associated with adverse health effects, such as oxidative stress, carcinogenicity, hormonal disruptions, and neurotoxicity (Grimm et al., 2015; Liu et al., 2020; Sethi et al., 2017).

Despite the growing evidence that PCB metabolites are also toxic, significant knowledge gaps remain regarding the PCB metabolites present in the systemic circulation of wildlife, laboratory animals, and humans. One study reported the presence of PCB sulfates, hydroxylated PCB sulfates (OH-PCB sulfates), hydroxylated MeSO2-PCBs, and PCB sulfonates in polar bear serum (Liu et al., 2018). In rats, the lack of uridine diphosphate glucuronosyltransferase (UGT) markedly altered the profiles of dihydroxylated and methyl sulfone metabolites; however, it is unknown if the lack of UGTs altered the circulating levels of PCB glucuronides (Haraguchi et al., 2004). Several PCB metabolites, including two hydroxylated PCB 11 sulfate isomers and a hydroxylated PCB 11 glucuronide, were detected in serum from female mice 24 h after acute oral exposure to PCB 11 (Zhang et al., 2021b).

The present study characterized the metabolite profile in mouse dams exposed to the Fox River Mixture (FRM), an environmental PCB mixture that models human exposure to PCB-contaminated fish (Kostyniak et al., 2005) and is implicated in neurotoxic outcomes following developmental exposures of rodent dams (Monaikul et al., 2017; Sable et al., 2006). As part of a developmental neurotoxicity study of the FRM, female mice were exposed to different doses of the FRM via the diet beginning two weeks before mating and continuing throughout gestation and lactation. Gas chromatography-tandem mass spectrometry (GC-MS/MS) and LC-HRMS analysis of serum samples collected at weaning of offspring at postnatal day 21 (P21) revealed complex PCB metabolite profiles in the dams, alterations in metabolites involved in the elimination of endogenous metabolites and xenobiotics via glucuronidation, and changes in heme metabolism associated with levels of higher chlorinated PCBs and PCB metabolites.

2. Materials and methods

2.1. Chemicals and materials

The nomenclature of PCBs and OH-PCBs is based on the Ballschmiter and Zell number of the PCB congeners, and the number of hydroxy (OH-) or methoxy groups (MeO-) is based on the numbering scheme of the respective PCB congeners (EPA, 2003; Maervoet et al., 2004). For details regarding the technical Aroclor mixtures used to prepare the FRM and other chemicals and materials used in this study, see the Supplementary Material. Deuterated PCB30 (d5-PCB30) and PCB204 were used as internal standards for the GC-MS/MS analysis of PCBs and OH-PCBs (MeO-PCBs). Potassium perfluorooctanesulfonate was used as an internal standard (volume corrector) for the LC-HRMS analysis; see details in the Supplementary Material. All data underlying this manuscript are available free of charge on Iowa Research Online (Li et al., 2024).

2.2. Animal exposure

All animal experiments followed the Guide for the Care and Use of Laboratory Animals of the National Research Council of the United States and were approved by the Institutional Animal Care and Use Committees of the University of Washington (Seattle, Washington, USA). Details regarding the animal studies are reported elsewhere following the ARRIVE guidelines (Wilson et al., 2024). Adult female C57BL/6J dams (8-10 weeks old) were exposed to the FRM at 0.1, 1.0, or 6.0 mg/kg/d in the diet beginning two weeks before mating and continuing throughout gestation and lactation (median exposure, 64 days, range 63 to 65 days of exposure). Control mice were exposed parallel to the vehicle (peanut oil in peanut butter). Dams were euthanized on P21, and tissues were collected to characterize the disposition of PCBs and PCB metabolites dams exposed to 0, 0.1, 1, and 6 mg/kg bw/d of the FRM. For a summary of the serum samples used for the chemical analyses, see Table S1.

2.3. Extraction of PCBs and OH-PCBs from serum samples for GC-MS/MS analysis

Serum samples (105 ± 32 mg, range: 36 to 159 mg) were mixed with a potassium chloride solution (4 mL, 1% in water), followed by the addition of 13C-PCB (10 ng each in hexane) and 13C-OH-PCB surrogate standards (10 ng each in methanol) to asses the extraction performance. Hydrochloric acid (1 mL of 6 M aqueous solution) and 2-propanol (5 mL) were added to each sample. Samples were extracted twice with a mixture of hexane and methyl tert-butyl ether (MTBE) (5 mL and 3 mL, 1:1, v/v), as described (Koh et al., 2016; Li et al., 2020; Marek et al., 2014). The combined extracts were washed with a potassium chloride solution (4 mL, 1% in water) and separated into PCBs and OH-PCBs using their physicochemical properties (Hovander et al., 2000; Hovander et al., 2006). Briefly, the OH-PCBs were separated from the PCBs by partitioning them into an aqueous-ethanolic solution of potassium hydroxide (2 mL, 0.5 M, 50% ethanol). The OH-PCB fraction was acidified with hydrochloric acid (0.5 mL of 2 M aqueous solution), and the OH-PCBs were extracted twice with hexane-MTBE (3 mL, 9:1, v/v). The OH-PCBs were derivatized with diazomethane in diethyl ether (0.5 mL, about 0.12 mmol) to form methoxylated PCBs. The PCB and OH-PCB fractions were cleaned up by solid-phase extraction with acidified silica gel, as described (Li et al., 2022a) and concentrated. d5-PCB30 and PCB204 (10 ng each isooctane) were added as internal standards (volume corrector) and extracts were transferred to amber glass autosampler vials for GC-MS/MS analysis. For additional details, see the Supplementary Material and Tables S2-7.

2.4. Gas chromatographic determinations

Both PCB and MeO-PCB samples were analyzed on an Agilent 7890A GC system coupled with an Agilent 7000D Triple Quad in the multiple reaction monitoring mode on an SPB-Octyl capillary column (50% n-octyl/50% methyl siloxane, 30 m length, 250 μm inner diameter, 0.25 μm film thickness; Sigma-Aldrich, St. Louis, MO, USA). For details of the GC-MS/MS method and quality assurance/quality control (QA/QC) data, see the Supplementary Material.

2.5. Extraction of PCB metabolites from serum for LC-HRMS analysis

Serum samples (105 ± 32 mg, range: 36 to 159 mg) were extracted for LC-HRMS analysis as described, with slight modifications (Li et al., 2022a); for details, see the Supplementary Material. Briefly, serum samples were mixed with milli-Q water and acetonitrile. Sodium chloride and anhydrous magnesium sulfate were added, and the acetonitrile phase was passed through a 3 mL HybridSPE cartridge (Millipore Sigma, Burlington, MA, USA) preloaded with anhydrous sodium sulfate and anhydrous magnesium sulfate and subsequently eluted with acetonitrile. The sample was prepared for LC-HRMS analysis in several evaporation, reconstitution, and centrifugation steps, as described in the Supplementary Material. The samples were reconstituted in an acetonitrile-water mixture (200 μL, 1:1, v/v) for LC-HRMS analysis.

2.6. LC-HRMS analysis ofPCB metabolites

Serum extracts were analyzed on a Q-Exactive Orbitrap Mass Spectrometry (Thermo Fisher Scientific, Waltham, MA, USA) equipped with a Vanquish Flex ultra-high-performance liquid chromatography (UPLC, Thermo Fisher Scientific) and an Acquity UPLC-C18 column (particle size: 1.7 particle size: 1.7 μm, 2.1 x 100 mm, Waters, Milford, MA, USA) at the High-Resolution Mass Spectrometry Facility of the University of Iowa. Details regarding the mobile phase and the instrument parameters are provided in the Supplementary Material. The LC-HRMS data were processed and analyzed with Xcalibur 4.1 (Thermo Fisher Scientific Inc, Waltham, MA, USA).

A PCB metabolite suspect screening list based on the PCB metabolism literature (Dhakal et al., 2012; Grimm et al., 2015; Li et al., 2021; Liu et al., 2018; Zhang et al., 2020) was used to identify PCB metabolites. Putative PCB metabolites were identified and integrated with Xcalibur software with a 5 ppm mass tolerance and 5 mass precision decimals (Li et al., 2021). The identification was confirmed using the chlorine isotope ratio of the molecular ion peak and, when observed, characteristic fragment ions; see Fig. S1 for representative chromatograms and mass spectra of PCB metabolites. The abundance of each PCB metabolite was expressed as the integrated area adjusted for the peak area of the volume corrector, potassium perfluorooctanesulfonate. For a summary of the abundance of each PCB metabolite and the QA/QC information, see the Supplementary Material (Table S8).

2.7. Untargeted metabolomics analysis

The LC-HRMS data were used to explore metabolomics changes in the serum of dams exposed to the FRM. The pre-processing of LC-HRMS data, including noise reduction, peak picking, feature identification, feature quantification, and feature alignment, was performed with the R package apLCMS (Yu et al., 2009). Further analysis was performed with MetaboAnalyst 6.0 (Pang et al., 2024). Partial least squares discriminant analysis (PLS-DA) was performed with all raw feature data for all exposure groups. Volcano plots comparing each PCB exposure group to the control group were generated using a threshold of 0.05 for both raw p-values and false discovery rates (FDR) and a fold change threshold of 2.

Pathway enrichment analyses were performed with the Functional Analysis module of MetaboAnalyst using a mass tolerance of 5 ppm, retention time in minutes, and a generic data source. Data were cleaned by excluding features with more than 50% missing data, and 40% of features were filtered based on the relative standard deviation (RSD/mean), yielding a list of 807 compounds, including parent compounds and their fragment ions and isomers. Data were normalized by sum, log10-transformed, and auto-scaled. The metabolites with Kyoto Encyclopedia of Genes and Genomes (KEGG) identifiers involved in the pathways were converted to the compound names with Metabolite ID Conversion function in MetaboAnalyst, resulting in a list with 354 annotated compounds. This annotation included several isomeric compounds for the same feature (same retention time and m/z value).

2.8. Network analysis

The final annotated feature list from the pathway analysis was refined for network analysis by retaining only the first isomeric compound when multiple annotations were present, resulting in a list of 205 metabolic features (Table S9). The relative abundance of the annotated compounds was obtained from the raw feature list by 1) selecting the monoisotopic ion when multiple ions were detected, 2) preferring [M-H]− over other ion forms of the same compound, such as [M-H2O-H]−, and 3) considering the retention time of the potential compounds. The network analysis was performed with this feature list and the PCB and OH-PCB levels in dam serum using xMWAS (Uppal et al., 2018). The relative standard deviation threshold was 0, the minimum non-missing sample ratio was 0.5, the missing values were treated as 0, the number of components for PLS analysis was 3, and the correlation threshold was 0.5. Otherwise, the default parameter settings were used. The results were visualized using Cytoscape (version 3.10.1) (Shannon et al., 2003).

2.9. Data analysis

Data are represented as the mean ± standard deviation and expressed as ng per gram serum weight (ng/g) (Tables S6-S7). The statistical analyses were conducted using the built-in one-way ANOVA and PLS-DA functions in MetaboAnalyst. The similarity coefficient (cos θ) was utilized to assess the similarity between PCB and OH-PCB profiles (Li et al., 2022a; Li et al., 2020). The value of cos θ ranges from 0 to 1, with 0 indicating complete dissimilarity and 1 indicating complete similarity between two profiles.

3. Results and Discussion

3.1. PCB congener profiles and PCB levels

The PCB congener profile of the FRM prepared for this study (Fig. S2) was comparable to earlier FRM preparations, as reported (Li et al., 2023; Wilson et al., 2024). The PCB congener profiles in serum differed from the FRM profile (cos θ from 0.63 to 0.93; Fig. S3). For example, many PCB congeners with one to three chorine substituents in the FRM were not detected in serum samples. The homolog composition shifted from mostly tetra-chlorinated congeners in the FRM to hexa-chlorinated PCBs in the serum (Fig. 1A1 vs. 1A2). Changes in the mass profiles based on dioxin-like vs. non-dioxin-like PCB congeners and PCB structure (e.g., the number of ortho-chlorine substituents and PCB class) were also observed when comparing the FRM to the mass profiles in the serum from PCB exposed dams (Fig. S4). Earlier studies have reported similar changes between the PCB congener profile in the dosing mixture and the PCB residue in serum and tissue (Hu et al., 2015; Kania-Korwel et al., 2005; Li et al., 2022b). These differences in the PCB congener profiles are likely due to the more rapid metabolism of lower chlorinated PCBs (Borlakoglu and Wilkins, 1993).

Fig. 1.

Fig. 1.

Open in a new tab

Heatmap-like plots of the (A1) mass profile of PCB homologs in FRM, (A2) levels of PCB homologs (ng/g) in serum, and (A3) levels of OH-PCB homologs (ng/g) in serum of dam exposed to FRM with different dosage (i.e., low dose, LD, n=5; 0.1 mg/kg; medium dose, MD, n=6; 1 mg/kg; high dose, HD, 6 mg/kg, n=6). These data indicate that lower chlorinated PCBs (Cl3 and Cl4) are more rapidly eliminated than high chlorinated PCBs (Cl5, Cl6 and Cl7). Hexa- and hepta-chlorinated PCBs were the major PCB homologs in the serum, compared to the tri- and tetra-chlorinated PCBs in the FRM to which dams were exposed. FRM. Tri- and tetra-chlorinated OH-PCBs were the major hydroxylated PCB metabolites. The partial least-square discriminant analysis (PLS-DA) of (B1) PCB and (B2) OH-PCB congener profiles, and the corresponding VIP score plots of the (C1) top 10 PCB congeners and (C2) top 10 OH-PCB congeners, revealed a separation of PCB but not OH-PCB profiles by exposure group. OH-PCBs were derivatized and analyzed as the corresponding MeO-PCBs. The PLS-DA was performed with MetaboAnalyst 6.0 (Pang et al., 2024). The data below the detection limits were imputed with LOQ/2, and the data were quantile normalized and log transformed before PLSDA.

PCB90+101+113, PCB95, and PCB110 were the major PCB congeners observed in the serum of low and medium-dose groups (Fig. S2). Higher chlorinated PCBs, such as PCB129+138+163, PCB153+168, and PCB180+193, were enriched relative to lower chlorinated PCBs in the high-dose group. A comparison of the average PCB profiles across exposure groups indicates that the low and medium-dose groups have similar profiles (cos θ = 0.93). In contrast, the PCB profile of serum from the high dose group differs from that of the low (cos θ = 0.63) and medium (cos θ = 0.82) dose groups. Comparing this study with a previously published study of female mice exposed sub-acutely to 6 vs. 30 mg/kg bw/d of the FRM, less pronounced dose-dependent differences in the PCB congener profiles were observed in conventional (cos θ = 0.92) and germ-free (cos θ = 0.92) respectively (Li et al., 2022b). This is likely due to the different dosing paradigms or, like other chemicals (Eke et al., 2023), differences in the disposition of PCBs in pregnant vs. non-pregnant mice. Partial least-square discriminant analysis (PLS-DA) of the PCB congener profile in the serum revealed a separation by dose effect along Principal Component 1 (PC1) (Fig. 2). Based on the variable importance in projection (VIP) scores, the separation along PC1 was driven by an increased abundance of higher chlorinated PCB congeners (e.g., PCB172, PCB196, and PCB202) at the high PCB dose.

Fig. 2.

Fig. 2.

Open in a new tab

General PCB metabolism pathway (Grimm et al., 2015; Li et al., 2021; Liu et al., 2018) showing the relative abundance of the seven classes of PCB metabolites identified using LC-HRMS in the serum of dams exposed to the FRM throughout gestations and lactation. The relative abundance is expressed as the peak area of the PCB metabolites per gram of serum adjusted for the peak area of the internal standard (PFOS). A, cytochrome P450 enzymes; B, glutathione S-transferase, non-enzymatic reactions, gamma-glutamyl transpeptidase, and cysteine S-conjugate beta-lyase; C, oxidation; D, epoxide hydrolase, dihydrodiol dehydrogenase; E, non-enzymatic reaction; F, catechol-O-methyltransferase (COMT); G, sulfotransferase. The chemical structures only show the functional groups without specifying the substituted positions.

The total PCB levels, ΣPCBs, in the dam serum of low, medium, and high dose groups were 150, 360, and 740 ng/g, respectively. The ΣPCB levels in the medium and high-dose groups were 2.4 and 4.9-fold higher than those in the low-dose group. The average levels of individual PCB congeners in the medium and high dose groups were 2.1 (range 0.4 to 6.0) and 6.0 (range 0.5 to 23.6) fold higher than those in the low dose group. For many PCB congeners, these fold differences are smaller than the 10- and 6-fold difference in the PCB dose, especially when comparing the low to the medium dose group. In studies using the same PCB doses and similar dosing paradigms, PCB95 blood levels in female mice showed a smaller-fold difference than the 10- and 6-fold difference in the low to medium to high PCB95 dose (Kania-Korwel et al., 2015; Kania-Korwel et al., 2012). The non-linear change in PCB blood levels is likely due to the induction of hepatic cytochrome P450 enzymes involved in the metabolism of PCBs (Lim et al., 2020; Robertson et al., 1984).

3.2. OH-PCB congener profiles and OH-PCB levels

The retention of some OH-PCBs in the blood of humans and other mammals was first reported by Bergman and co-workers (Bergman et al., 1994). A growing body of evidence implicates these and other OH-PCBs in adverse health outcomes, including developmental neurotoxicity (Grimm et al., 2015; Sethi et al., 2017). Because of the large number of 837 possible OH-PCBs, lack of available standards, and analytical challenges, many laboratory and epidemiological studies analyze only a small number of OH-PCBs in blood. We quantified 124 OH-PCB congeners as 114 peaks of co-eluting or single OH-PCB congeners (as the methylated derivatives) in the dams exposed to the FRM. Ninety-one single or co-eluting OH-PCBs were detected across all three dose groups (Fig. S5).

The OH-PCB congener profiles were similar across all three exposure groups (cos θ from 0.94 to 0.98; Fig. S3). The major homolog groups were tetra-chlorinated OH-PCBs, followed by tri- and penta-chlorinated OH-PCB congeners (Fig. 1A3). This homolog composition is consistent with the depletion of metabolically labile lower chlorinated PCBs (Grimm et al., 2015). Many OH-PCB congeners detected in mouse serum had ≤ 4 chlorinate substituents which, except for PCB28 metabolites, are rarely measured in animal or human studies. PCB28 metabolites, for example, 3-OH-PCB28 and 3'-OH-PCB28, have been reported in human plasma (Quinete et al., 2017). In addition, several higher chlorinated OH-PCB congeners that are present in human plasma (Bergman et al., 1994; Koh et al., 2016; Park et al., 2007), such as 4-OH-PCB107 and 4-OH-PCB146, were also detected in the dam serum.

The total OH-PCB (ΣOH-PCB) levels in the dam serum of low, medium, and high dose groups were 600, 2200, and 6000 ng/g, respectively. The ΣOH-PCB increased 3.6-fold from the low to the medium dose and 2.7-fold from the medium to the high dose. Analogous to the parent PCBs, these fold differences are smaller than the 10- and 6-fold difference in the PCB dose. However, the fold differences of individual OH-PCBs are more variable than the parent PCBs. The levels of individual OH-PCB congeners in the medium and high dose groups were 7.9 (range 0.1 to 70) and 11.4 (range 0.1 to 117) fold higher than those in the low dose group. For example, serum levels of 4'-OH-PCB107, a PCB metabolite found in human serum, were 3.9 and 31-fold higher in medium and high-dose groups than in the low-dose group. The serum levels of 4'-OH-PCB65, a major OH-PCB metabolite, were 3.3 and 6.7 higher in medium and high-dose groups than in the low-dose group. The variability of the OH-PCB metabolite levels across exposure groups is the result of congener-specific differences in PCB metabolism and, ultimately, the toxicokinetics.

The ΣOH-PCB serum levels in this study are much higher than the ΣOH-PCB levels observed in most human populations (Fängström et al., 2002). However, some individuals from highly PCB-exposed human populations have ΣOH-PCB levels within the range observed in the present study. For example, a study of pregnant Inuit women reported geometric mean maternal ΣOH-PCB levels of approximately 310 ng/g (110-1480 ng/g) (Dallaire et al., 2009). The comparisons between the ΣOH-PCB levels from this study and population-based studies need to be interpreted with caution. Our study measured a larger number of OH-PCB congeners than most human biomonitoring studies, including lower chlorinated OH-PCBs that are expected to be cleared more rapidly (Grimm et al., 2015). These lower chlorinated OH-PCBs may be significant serum metabolites following acute exposure in humans; however, they may not be detected in human biomonitoring studies because they are cleared rapidly from the systemic circulation.

Notably, the ΣOH-PCB serum levels are 4 to 8 times higher than the ΣPCB serum levels in the dams. In other animal studies, the ΣOH-PCB blood levels were higher than or comparable to the parent PCB levels in mice exposed repeatedly to different doses of PCB95 or PCB136 (Kania-Korwel et al., 2015; Kania-Korwel et al., 2017). In contrast, human biomonitoring studies consistently show lower ΣOH-PCB than ΣPCB levels. OH-PCB blood levels in humans frequently are 10–20% of the PCB blood concentration (Fängström et al., 2002; Sandau et al., 2000). For example, in a study from the United States, median ΣPCB and ΣOH-PCB serum levels in women from East Chicago, Illinois, were 0.41 vs. 0.11 ng/g, respectively (Marek et al., 2014). The same study reported a similar difference in PCB vs. ΣOH-PCB serum levels in women from Columbus Junction, Iowa.

3.3. Semi-targeted analysis of PCB serum metabolites

Using a published semi-targeted method, we tentatively identified 4, 7, and 7 classes of PCB metabolites in the serum from dams exposed to low, medium, and high doses of the FRM (Fig. 2), respectively. Metabolite classes detected in serum include OH-PCBs (C12H9-nClnO−, n=3-9, 18 analytes), PCB sulfates (C12H9-nClnSO4−, n=1-7; 18 analytes), PCB sulfonates (C12H9-nClnSO3−, n=3-6; 7 total analytes), hydroxylated PCB sulfates (OH-PCB sulfates, C12H9-nClnSO5−, n=1-7; 27 total analytes), methoxylated hydroxy PCBs (MeO-OH-PCBs, C13H11-nClnO2−, n=1-5, 9 analytes), methoxylated PCB sulfates (MeO-PCB sulfates, C13H11-nClnO5S−, n=1-5, 16 analytes), and methoxylated hydroxylated PCB sulfates (MeO-OH-PCB sulfates, C13H11-nClnO6S−, n=2-4, 8 analytes). The metabolite classes observed in the serum of mice exposed to the FRM are consistent with established PCB metabolism pathways (Li et al., 2021; Liu et al., 2018)(Fig. 2).

The relative levels of the metabolite classes followed the approximate rank order: PCB sulfate > OH-PCB sulfates > MeO-PCB sulfates > OH-PCBs > PCB sulfonates > MeO-OH-PCBs ~ MeO-OH-PCB sulfates. Consistent with the GC-MS/MS analysis, penta-chlorinated OH-PCBs were the major phase I metabolites, whereas tri- and tetra-chlorinated sulfated metabolites were the dominant phase II metabolites. No dihydroxylated, bisquinone, or polychlorinated dibenzofuran (PCDF) metabolites were found in the serum of dams exposed to the FRM. However, in the previous studies, both metabolite classes were detected in feces from female mice 24 h after oral exposure to the FRM (Fig. 2) (Li et al., 2021; Liu et al., 2018).

Compared to this study, a more significant number of PCB metabolites were observed in the feces of conventional (mice with a microbiome) or germ-free mice (mice without a microbiome) 24 h after oral exposure to the FRM (Fig. 3A) (Li et al., 2021). The major metabolites found in the feces of conventional female mice after 24 h of exposure to the FRM were OH-PCBs, not PCB sulfates. Moreover, dihydroxylated PCBs, bisquinones, and PCDF metabolites were observed in feces, but not in serum, from FRM-exposed conventional. The fecal PCB metabolite profiles in conventional mice differed from those observed in serum in this study (cos θ from 0.04 to 0.11; Fig. 3B). The fecal metabolite profiles of germ-free mice were more comparable to the serum profiles (cos θ from 0.75 to 0.81). The comparison of the PCB metabolite profiles between serum and feces from germ-free mice suggests that the profiles of PCB metabolites eliminated into the feces are comparable to the profiles of PCB metabolites in the systemic circulation but that PCB metabolites undergo further metabolism when bacteria are present in the gastrointestinal tract. Thus, unlike some OH-PCBs (Bergman et al., 1994)Many other PCB metabolites may not be selectively retained in the serum, a hypothesis that warrants attention considering their potential toxicity.

Fig. 3.

Fig. 3.

Open in a new tab

The mass profiles of PCB metabolites (A) and the similarity coefficient cos θ comparing the profiles between different experimental groups (B) in the serum of dams exposed to 0.1, 1, and 6 mg/kg bw/d of the FRM. Analogous metabolite profiles previously identified in fecal samples from female mice exposed to 6 or 30 mg/kg bw of the FRM for 24 hours are presented for comparison in panel A (Cheng et al., 2018; Li et al., 2021). “0” indicates data that are < 0.1 and blank fields indicate metabolites that were not detected. See mass profile data in Table S10.

3.4. Untargeted metabolomics

The LC-HRMS serum data were further analyzed using the untargeted metabolomics workflow implemented by MetaboAnalyst 6.0. (Pang et al., 2024) to explore if PCB exposure alters the serum metabolome in mice (Fig. 4). Partial least squares-discriminant analysis (PLS-DA) demonstrated separation of the metabolomic features of the four exposure groups along three PCs. Using a false discovery rate (FDR) threshold of 0.05, Volcano plots identified 92 features in the low-dose group, 342 in the medium-dose group, and 332 in the high-dose group that differed from the control animals. Pathway analysis revealed significant alterations in the "amino sugar and nucleotide sugar metabolism" and "ascorbate and aldarate metabolism” pathways across all three PCB exposure groups. The "pentose and glucuronate interconversions" pathway was also significantly affected in the medium and high exposure groups. Two metabolites, UDP-α-D-glucuronate and uridine diphosphate glucose, were significantly altered by PCB exposure in all three pathways. The serum levels of both metabolites increased with increasing PCB dose. Together, metabolomic changes suggest alterations in metabolic pathways involved in eliminating endogenous metabolites and xenobiotics via glucuronidation.

Fig. 4.

Fig. 4.

Open in a new tab

Untargeted metabolomics analysis of serum samples from dams exposed to peanut butter only (vehicle, n=4), low dose (0.1 mg/kg bw, n=6), medium dose (1 mg/kg bw, n=7), or high dose (6 mg/kg bw, n=8) of the FRM revealed dose-dependent effects of PCB exposure. (A) Partial least squares-discriminant analysis (PLSDA) showed a separation of the metabolomic features of the four exposure groups along three Principal Components. (B) Volcano plots with data from the (B1) low, (B2) medium, and (B3) high dose groups of 191, 496, and 474 features with a threshold of p = 0.05 (blue line) and 92, 342, and 332 features with an FDR threshold of 0.05 (orange line). (C) Pathway enrichment analyses identified 2, 3, and 3 affected metabolic pathways for the (C1) low, (C2) medium, and (C3) high dose groups, respectively. Three pathways altered by PCB exposure included the “amino sugar and nucleotide sugar metabolism”, “ascorbate and aldarate metabolism,” and “pentose and glucuronate interconversions” pathways. The data in the parenthesis indicate the significant hits, all hits, and the total numbers of empirical compounds in the pathway. (D) Examples of metabolites significantly altered by PCB exposure in the three pathways, including (D1) UDP-α-D-glucuronate and (D2) uridine diphosphate glucose. The pathway enrichment analysis and corresponding compound mapping and annotation were performed in the Functional Analysis module of MetaboAnalyst 6.0 (Pang et al., 2024). The statistical analyses were performed by one-way ANOVA as implemented by MetaboAnalyst. p-values for D1 and D2 are 1.62×10−6 and 1.03×10−6, and corresponding FRD are 2.88 ×10−5 and 1.97 ×10−5; * indicates significant differences with the Fisher post-hoc test.

PCBs and their metabolites increase UGT expression in rats (Horio et al., 1983; Kato et al., 2000; Vansell and Klaassen, 2002), which is expected to decrease and not increase UDP-α-D-glucuronate levels, as observed in this study. Limited information is available about the effects of PCBs on hepatic UGT expression. Unlike rats, exposure to the FRM did not alter the expression of UGTs, such as Ugt2b1, in the liver of conventional and germ-free female mice exposed to FRM (Li et al., 2022b). While some studies detected PCB glucuronides in rodents in vivo (Lucier et al., 1978; Zhang et al., 2021b), we did not observe any PCB glucuronide metabolites in the serum in this study. Therefore, the changes identified with the untargeted metabolomic analysis likely reflect PCB-mediated effects of endogenous metabolites subject to glucuronidation and not effects on the elimination of phenolic PCB metabolites. Because of the exploratory nature of our metabolomics analysis, further studies are needed to fully elucidate the impact of the FRM on the serum metabolome in pregnant and lactating mice.

3.5. Network analysis of PCBs, OH-PCBs, and untargeted endogenous metabolomics

The network analysis of serum PCBs, OH-PCBs, and endogenous metabolites identified four clusters (Fig. 5). Fifteen PCBs, 5 OH-PCBs, and 24 endogenous metabolites had strong and significant correlations in the network (∣r∣ > 0.4, p < 0.05). Several octa- and nona-chlorinated PCB congeners were positively correlated with four penta-chlorinated OH-PCBs and 4-OH-PCB202. These OH-PCBs are not formed from octa- and nona-chlorinated PCBs. The induction of drug-metabolizing enzymes by highly chlorinated PCB congeners in the mouse liver may explain the correlations observed in the network analysis.

Figure 5.

Figure 5.

Open in a new tab

A differential network analysis of the PCB, OH-PCB, and untargeted endogenous metabolite data identified four clusters. Fifteen PCB, 5 OH-PCB congeners, and 24 endogenous metabolites had strong and significant correlations in the network (∣r∣ > 0.4, p < 0.05). Panels show (A) the overall network of PCBs, OH-PCBs, and endogenous metabolites, (B) sub-network showing PCBs and OH-PCBs correlating with serum levels of heme and its intermediate, coproporphyrinogen III, and (C) sub-network showing correlations of 4'-OH-PCB93 with serum levels of endogenous metabolites, including metabolites in amino acid and pyrimidine metabolism pathways. The network correlation analyses were performed with xMWAS (Uppal et al., 2018). The visualization of the results was performed with Cytoscape (version 3.10.1). Oval node shapes represent PCBs and OH-PCBs and triangles show endogenous metabolites. The edge color indicates positive (red) and negative (blue) correlations.

The 15 PCB congeners identified by the network analysis correlate with the same 12 endogenous metabolites, as shown for PCB180+193 as example (Fig. 5B). These PCB congeners and several OH-PCBs showed negative correlations with heme and its intermediate, coproporphyrinogen III. PCBs and OH-PCBs can inhibit uroporphyrinogen decarboxylase, leading to the depletion of coproporphyrinogen III and heme (Kawanishi et al., 1983; Sano et al., 1985), and cause the oxidation of uroporphyrinogen and bilirubin (De Matteis et al., 2002). PCB exposure can also alter systemic iron homeostasis (Qian et al., 2015; Wang et al., 2013), suggesting an effect of PCBs on heme metabolism in FRM-exposed mice. These results suggest that higher chlorinated PCB disrupt metabolic pathways involved in heme synthesis and degradation.

In the network analysis, the cluster containing 4'-OH-PCB93 showed a negative correlation with deoxyuridine and thymidine, metabolites of the pyrimidine metabolism pathway, and 4-(2-aminophenyl)-2-4-dioxobuanoic acid, histidine, and 4-acetamidobutanoic acid, metabolites involved in amino acid metabolism (Fig. 5). A few other metabolites, such as methyl histidine and tyrosine, are also involved in amino acid metabolism. Both metabolites were negatively associated with several PCBs and OH-PCBs. Earlier metabolomics studies also found that PCBs affect amino acid metabolism pathways, such as tyrosine and pyrimidine metabolism pathways, in HepG2 cells in culture (Zhang et al., 2020; Zhang et al., 2021a; Zhang et al., 2022). There is also evidence that the FRM alters amino acid levels in female mice exposed to the FRM (Lim et al., 2020). Other rodent studies have also reported changes in serum amino acids following PCB exposure (Pikkarainen et al., 2019; Shi et al., 2012). The results indicate that highly persistent octa- and nona-chlorinated PCB congeners and some penta-chlorinated OH-PCBs affect amino acid and pyrimidine metabolism pathways in vivo.

4. Conclusion

The study provides comprehensive insights into the metabolism and systemic effects of PCBs in mouse dams, demonstrating dose-dependent alterations in PCB and PCB metabolite profiles and significant metabolic disruptions associated with higher chlorinated PCB congeners. The findings highlight the complexity of PCB biotransformation and its impact on endogenous metabolic pathways, emphasizing the need for further research to understand the implications of PCB exposure on PCB developmental neurotoxicity. The study underscores the importance of considering both parent compounds and their metabolites in assessing the toxicological impact of PCBs. The evidence of metabolic pathway disturbances, particularly in carbohydrate, heme, and amino acid metabolism, identifies potentially novel mechanisms underlying PCB-induced neurotoxicity. These results suggest that prenatal and early-life exposure to PCBs and their metabolites through maternal transfer could contribute to adverse developmental outcomes.

Supplementary Material

1
2

Highlights.

  • PCBs and metabolites were measured in serum from dams exposed to a PCB mixture

  • PCB but not OH-PCB serum profiles changed with the PCB dose

  • Serum contained complex sulfated and methylated PCB metabolite mixtures

  • PCB exposure altered serum levels of metabolites in the glucuronidation pathway

  • Higher chlorinated PCBs affected heme and amino acid metabolic pathways

Acknowledgments

We thank Alexander N. S. Slack and Binita Gautam for helping with the data processing, Drs. Rachel F. Marek and Keri C. Hornbuckle from the Analytical Core of Iowa Superfund Research Program for supporting the GC-MS/MS analyses, and Drs. Lynn M. Teesch and Vic R. Parcell from the High-Resolution Mass Spectrometry Facility at the University of Iowa for supporting the LC-HRMS analysis.

Funding

The work was supported by the National Institutes of Health [grant numbers R01ES014901, R01ES031098, and R01ES034691] and performed in facilities of the Environmental Health Sciences Research Center [P30ES005605] and the Iowa Superfund Research Program [P42ES013661]. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding agencies listed above.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1

Abbreviations: FDR, false discovery rate; FRM, Fox River Mixture; GC-MS/MS, gas chromatography-tandem mass spectrometry; KEGG, Kyoto Encyclopedia of Genes and Genomes; LC-HRMS, liquid chromatography-high resolution mass spectrometry; MeO-PCB, methoxylated PCB; MeSO2-PCB, methyl sulfone PCB; MTBE, methyl tert-butyl ether; OH-PCB, hydroxylated PCB; PCB, polychlorinated biphenyl; PLS-DA, partial least squares discriminant analysis; QA/QC, quality assurance/quality control; ΣOH-PCB, total OH-PCB levels; ΣPCBs, total PCB levels; UDP, uridine diphosphate; UGT, uridine diphosphate glucuronosyltransferase; UPLC, ultra-high-performance liquid chromatography; VIP, variable importance in projection.

CRediT authorship contribution statement

Xueshu Li: Conceptualization; Investigation; Data curation; Formal analysis; Methodology; Validation; Visualization; Writing – original draft; Writing – review & editing. Youjun Suh: Conceptualization; Investigation; Writing – review & editing. Rebecca Wilson: Conceptualization; Investigation; Writing – review & editing. Pamela J. Lein: Conceptualization; Funding acquisition; Project administration; Supervision; Writing - review & editing. Julia Y. Cui: Conceptualization; Funding acquisition; Project administration; Supervision; Writing – review & editing. Hans-Joachim Lehmler: Conceptualization; Formal analysis; Funding acquisition; Project administration; Supervision; Validation; Visualization; Writing – original draft; Writing – review & editing.

Declaration of generative AI and AI-assisted technologies in the writing process

During the preparation of this work the authors used ChatGPT and Grammarly to search the literature and improve the readability of the manuscript. After using these tools, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

Declaration of Competing Interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Pamela J. Lein was hired as an expert witness by lawyers representing a group of plaintiffs alleging they were harmed by exposure to PCBs in school air. In that capacity, she testified as an expert witness on PCB neurotoxicity. The defendant was Pharmacia, a successor company to Monsanto.

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Pamela J. Lein reports a relationship with Friedman Rubin (Seattle, WA) that includes: paid expert testimony. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The data used in the manuscript are freely available through Iowa Research Online at https://doi.org/10.25820/data.007305.

References

  1. Ampleman MD, Martinez A, DeWall J, Rawn DFK, Hornbuckle KC, Thorne PS, 2015. Inhalation and dietary exposure to PCBs in urban and rural cohorts via congener-specific measurements. Environ. Sci. Technol 49, 1156–1164. 10.1021/es5048039. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bergman A, Klassonwehler E, Kuroki H, 1994. Selective retention of hydroxylated PCB metabolites in blood. Environ. Health Persp 102, 464–469. 10.2307/3432042. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Borlakoglu JT, Wilkins JPG, 1993. Metabolism of dichlorobiphenyl, trichlorobiphenyl, tetrachlorobiphenyl, pentachlorobiphenyl and hexachlorobiphenyl by hepatic microsomes isolated from control animals and animals treated with Aroclor-1254, a commercial mixture of polychlorinated-biphenyls (PCBs). Comp. Biochem. Phys. C 105, 95–106. 10.1016/0742-8413(93)90064-R. [DOI] [PubMed] [Google Scholar]
  4. Cheng SL, Li XS, Lehmler HJ, Phillips B, Shen D, Cui JY, 2018. Gut microbiota modulates interactions between polychlorinated biphenyls and bile acid homeostasis. Toxicol. Sci 166, 269–287. 10.1093/toxsci/kfy208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Dallaire R, Muckle G, Dewailly E, Jacobson SW, Jacobson JL, Sandanger TM, Sandau CD, Ayotte P, 2009. Thyroid hormone levels of pregnant Inuit women and their infants exposed to environmental contaminants. Environ. Health Persp 117, 1014–1020. 10.1289/ehp.0800219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. De Matteis F, Dawson SJ, Pons N, Pipino S, 2002. Bilirubin and uroporphyrinogen oxidation by induced cytochrome P4501A and cytochrome P4502B - Role of polyhalogenated biphenyls of different configuration. Biochem. Pharmacol 63, 615–624. 10.1016/S0006-2952(01)00851-6. [DOI] [PubMed] [Google Scholar]
  7. Dhakal K, He XR, Lehmler HJ, Teesch LM, Duffel MW, Robertson LW, 2012. Identification of sulfated metabolites of 4-chlorobiphenyl (PCB3) in the serum and urine of male rats. Chem. Res. Toxicol 25, 2796–2804. 10.1021/tx300416v. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Duffel MW, Lehmler HJ, 2024. Complex roles for sulfation in the toxicities of polychlorinated biphenyls. Crit. Rev. Toxicol 54, 92–122. 10.1080/10408444.2024.2311270. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Eke AC, Gebreyohannes RD, Fernandes MFS, Pillai VC, 2023. Physiologic changes during pregnancy and impact on small-molecule drugs, biologic (monoclonal antibody) disposition, and response. J. Clin. Pharmacol 63, S34–S50. 10.1002/jcph.2227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. EPA, 2003. Table of polychlorinated biphenyls (PCB) congeners. https://www.epa.gov/pcbs/table-polychlorinated-biphenyl-pcb-congeners. Accessed November 11, 2024.
  11. EPA, 2024. Polychlorinated biphenyls (PCBs). Inadvertent PCBs. https://www.epa.gov/pcbs/inadvertent-pcbs. Accessed November 11, 2024. [Google Scholar]
  12. Fängström B, Athanasiadou M, Grandjean P, Weihe P, Bergman Å, 2002. Hydroxylated PCB metabolites and PCBs in serum from pregnant Faroese women. Environ. Health Perspect 110, 895–899. 10.1289/ehp.110-1240989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Grimm FA, Hu DF, Kania-Korwel I, Lehmler HJ, Ludewig G, Hornbuckle KC, Duffel MW, Bergman Å, Robertson LW, 2015. Metabolism and metabolites of polychlorinated biphenyls. Crit. Rev. Toxicol 45, 245–272. 10.3109/10408444.2014.999365. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Haraguchi K, Kato Y, Koga N, Degawa M, 2004. Metabolism of polychlorinated biphenyls by Gunn rats: Identification and serum retention of catechol metabolites. Chem. Res. Toxicol 17, 1684–1691. 10.1021/tx0498096. [DOI] [PubMed] [Google Scholar]
  15. Horio F, Kimura M, Yoshida A, 1983. Effect of several xenobiotics on the activities of enzymes affecting ascorbic-acid synthesis in rats. J. Nutr. Sci. Vitaminol 29, 233–247. [DOI] [PubMed] [Google Scholar]
  16. Hovander L, Athanasiadou M, Asplund L, Jensen S, Wehler EK, 2000. Extraction and cleanup methods for analysis of phenolic and neutral organohalogens in plasma. J. Anal. Toxicol 24, 696–703. 10.1093/jat/24.8.696. [DOI] [PubMed] [Google Scholar]
  17. Hovander L, Linderholm L, Athanasiadou M, Athanassiadis I, Bignert A, Fangström B, Kocan A, Petrik J, Trnovec T, Bergman Å, 2006. Levels of PCBs and their metabolites in the serum of residents of a highly contaminated area in eastern Slovakia. Environ. Sci. Technol 40, 3696–3703. 10.1021/es0525657. [DOI] [PubMed] [Google Scholar]
  18. Hu X, Adamcakova-Dodd A, Lehmler HJ, Gibson-Corley K, Thorne PS, 2015. Toxicity evaluation of exposure to an atmospheric mixture of polychlorinated biphenyls by nose-only and whole-body inhalation regimens. Environ. Sci. Technol 49, 11875–11883. 10.1021/acs.est.5b02865. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. IARC, 2017. Polychlorinated biphenyls and polybrominated biphenyls, https://publications.iarc.fr/131. Lyon, France. Accessed November 11, 2024. [Google Scholar]
  20. Kania-Korwel I, Barnhart CD, Lein PJ, Lehmler HJ, 2015. Effect of pregnancy on the disposition of 2,2′,3,5′,6-pentachlorobiphenyl (PCB 95) atropisomers and their hydroxylated metabolites in female mice. Chem. Res. Toxicol 28, 1774–1783. 10.1021/acs.chemrestox.5b00241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kania-Korwel I, Barnhart CD, Stamou M, Truong KM, El-Komy MHME, Lein PJ, Veng-Pedersen P, Lehmler HJ, 2012. 2,2′,3,5′,6-Pentachlorobiphenyl (PCB 95) and its hydroxylated metabolites are enantiomerically enriched in female mice. Environ. Sci. Technol 46, 11393–11401. 10.1021/es302810t. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kania-Korwel I, Hornbuckle KC, Peck A, Ludewig G, Robertson LW, Sulkowski WW, Espandiari P, Gairola CG, Lehmler HJ, 2005. Congener-specific tissue distribution of aroclor 1254 and a highly chlorinated environmental PCB mixture in rats. Environ. Sci. Technol 39, 3513–3520. 10.1021/es047987f. [DOI] [PubMed] [Google Scholar]
  23. Kania-Korwel I, Lukasiewicz T, Barnhart CD, Stamou M, Chung H, Kelly KM, Bandiera S, Lein PJ, Lehmler HJ, 2017. Editor's highlight: congener-specific sisposition of chiral polychlorinated biphenyls in lactating mice and their offspring: implications for PCB developmental neurotoxicity. Toxicol. Sci. 158, 101–115. 10.1093/toxsci/kfx071. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kato Y, Haraguchi K, Shibahara T, Shinmura Y, Masuda Y, Kimura R, 2000. The induction of hepatic microsomal UDP-glucuronosyltransferase by the methylsulfonyl metabolites of polychlorinated biphenyl congeners in rats. Chem. Biol. Interact 125, 107–115. 10.1016/S0009-2797(99)00168-4. [DOI] [PubMed] [Google Scholar]
  25. Kawanishi S, Seki Y, Sano S, 1983. Uroporphyrinogen decarboxylase - purification, properties, and inhibition by polychlorinated biphenyl isomers. J. Biol. Chem 258, 4285–4292. [PubMed] [Google Scholar]
  26. Koh WX, Hornbuckle KC, Marek RF, Wang K, Thorne PS, 2016. Hydroxylated polychlorinated biphenyls in human sera from adolescents and their mothers living in two US Midwestern communities. Chemosphere 147, 389–395. 10.1016/j.chemosphere.2015.12.113. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kostyniak PJ, Hansen LG, Widholm JJ, Fitzpatrick RD, Olson JR, Helferich JL, Kim KH, Sable HJK, Seegal RF, Pessah IN, Schantz SL, 2005. Formulation and characterization of an experimental PCB mixture designed to mimic human exposure from contaminated fish. Toxicol. Sci 88, 400–411. 10.1093/toxsci/kfi338. [DOI] [PubMed] [Google Scholar]
  28. Li XS, Hefti MM, Marek RF, Hornbuckle KC, Wang K, Lehmler HJ, 2022a. Assessment of polychlorinated biphenyls and their hydroxylated metabolites in postmortem human brain samples: age and brain region differences. Environ. Sci. Technol 56, 9515–9526. 10.1021/acs.est.2c00581. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Li XS, Lim JJ, Wang K, Prasad B, Bhatt DK, Cui JY, Lehmler HJ, 2022b. The disposition of polychlorinated biphenyls (PCBs) differs between germ-free and conventional mice. Environ. Toxicol. Phar 92. 10.1016/j.etap.2022.103854. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Li XS, Liu Y, Martin JW, Cui JY, Lehmler HJ, 2021. Nontarget analysis reveals gut microbiome-dependent differences in the fecal PCB metabolite profiles of germ-free and conventional mice. Environ. Pollut 268, 115726. 10.1016/j.envpol.2020.115726. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Li XS, Suh PY, Cui JY, Lehmler HJ, 2023. Dataset for characterization of the Fox River Mixture (FRM). University of Iowa (dataset). 10.25820/data.006235. [DOI] [Google Scholar]
  32. Li XS, Suh PY, Wilson R, Lein PJ, Cui JY, Lehmler HJ, 2024. Dataset for significant metabolic alterations in mouse dams exposed to an environmental mixture of polychlorinated biphenyls (PCBs) during gestation and lactation: insights into PCB and metabolite profiles. University of Iowa (Dataset). 10.25820/data.007305. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Li XS, Zhang CY, Wang K, Lehmler HJ, 2020. Fatty liver and impaired hepatic metabolism alter the congener-specific distribution of polychlorinated biphenyls (PCBs) in mice with a liver-specific deletion of cytochrome P450 reductase. Environ. Pollut 266. 10.1016/j.envpol.2020.115233. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Lim JJ, Li XS, Lehmler HJ, Wang DF, Gu HW, Cui JLY, 2020. Gut microbiome critically impacts PCB-induced changes in metabolic fingerprints and the hepatic transcriptome in mice. Toxicol. Sci 177, 168–187. 10.1093/toxsci/kfaa090. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Liu J, Tan Y, Song EQ, Song Y, 2020. A critical review of polychlorinated biphenyls metabolism, metabolites, and their correlation with oxidative stress. Chem. Res. Toxicol 33, 2022–2042. 10.1021/acs.chemrestox.0c00078. [DOI] [PubMed] [Google Scholar]
  36. Liu Y, Richardson ES, Derocher AE, Lunn NJ, Lehmler HJ, Li XS, Zhang YF, Cui JY, Cheng LH, Martin JW, 2018. Hundreds of unrecognized halogenated contaminants discovered in polar bear serum. Angew. Chem. Int. Ed 57, 16401–16406. 10.1002/anie.201809906. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Lucier GW, Mcdaniel OS, Schiller CM, Matthews HB, 1978. Structural requirements for accumulation of chlorinated biphenyl metabolites in fetal rat intestine. Drug Metab. Dispos 6, 584–590. [PubMed] [Google Scholar]
  38. Maervoet J, Covaci A, Schepens P, Sandau CD, Letcher RJ, 2004. A reassessment of the nomenclature of polychlorinated biphenyl (PCB) metabolites. Environ. Health Perspect 112, 291–294. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Marek RF, Thorne PS, DeWall J, Hornbuckle KC, 2014. Variability in PCB and OH-PCB serum levels in children and their mothers in urban and rural us communities. Environ. Sci. Technol 48, 13459–13467. 10.1021/es502490w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Monaikul S, Eubig P, Floresco S, Schantz S, 2017. Strategy set-shifting and response inhibition in adult rats exposed to an environmental polychlorinated biphenyl mixture during adolescence. Neurotoxicol. Teratol 63, 14–23. 10.1016/j.ntt.2017.08.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Montano L, Pironti C, Pinto G, Ricciardi M, Buono A, Brogna C, Venier M, Piscopo M, Amoresano A, Motta O, 2022. Polychlorinated biphenyls (PCBs) in the environment: occupational and exposure events, effects on human health and fertility. Toxics 10. 10.3390/toxics10070365. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Othman N, Ismail Z, Selamat MI, Kadir SHSA, Shibraumalisi NA, 2022. A review of polychlorinated biphenyls (PCBs) pollution in the air: where and how much are we exposed to? Int. J. Environ. Res 19. 10.3390/ijerph192113923. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Pang ZQ, Lu Y, Zhou GY, Hui FA, Xu L, Viau C, Spigelman AF, Macdonald PE, Wishart DS, Li SZ, Xia JG, 2024. MetaboAnalyst 6.0: towards a unified platform for metabolomics data processing, analysis and interpretation. Nucleic Acids Res. 10.1093/nar/gkae253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Park JS, Linderholm L, Charles MJ, Athanasiadou M, Petrik J, Kocan A, Drobna B, Trnovec T, Bergman Å, Hertz-Picciotto I, 2007. Polychlorinated biphenyls and their hydroxylated metabolites (OH-PCBs) in pregnant women from eastern Slovakia. Environ. Health Persp 115, 20–27. 10.1289/ehp.8913. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Pessah IN, Lein PJ, Seegal RF, Sagiv SK, 2019. Neurotoxicity of polychlorinated biphenyls and related organohalogens. Acta Neuropathol. 138, 363–387. 10.1007/s00401-019-01978-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Pikkarainen A, Lehtonen M, Håkansson H, Auriola S, Viluksela M, 2019. Gender- and dose-related metabolome alterations in rat offspring after in utero and lactational exposure to PCB 180. Toxicol. Appl. Pharm 370, 56–64. 10.1016/j.taap.2019.03.013. [DOI] [PubMed] [Google Scholar]
  47. Qian Y, Zhang SP, Guo WL, Ma J, Chen Y, Wang L, Zhao MR, Liu SJ, 2015. Polychlorinated biphenyls (PCBs) inhibit hepcidin expression through an estrogen-like effect associated with disordered systemic iron homeostasis. Chem. Res. Toxicol 28, 629–640. 10.1021/tx500428r. [DOI] [PubMed] [Google Scholar]
  48. Quinete N, Esser A, Kraus T, Schettgen T, 2017. PCB 28 metabolites elimination kinetics in human plasma on a real case scenario: Study of hydroxylated polychlorinated biphenyl (OH-PCB) metabolites of PCB 28 in a highly exposed German Cohort. Toxicol. Lett 276, 100–107. 10.1016/j.toxlet.2017.05.025. [DOI] [PubMed] [Google Scholar]
  49. Robertson LW, Parkinson A, Bandiera S, Lambert I, Merrill J, Safe SH, 1984. PCBs and PBBs - biologic and toxic effects on C57BL/6J and DBA /2J inbred mice. Toxicology 31, 191–206. 10.1016/0300-483x(84)90101-X. [DOI] [PubMed] [Google Scholar]
  50. Sable HJK, Powers BE, Wang VC, Widholm JJ, Schantz SL, 2006. Alterations in DRH and DRL performance in rats developmentally exposed to an environmental PCB mixture. Neurotoxicol. Teratol 28, 548–556. 10.1016/j.ntt.2006.06.005. [DOI] [PubMed] [Google Scholar]
  51. Saktrakulkla P, Lan T, Hua J, Marek RF, Thorne PS, Hornbuckle KC, 2020. Polychlorinated biphenyls in food. Environ. Sci. Technol 54, 11443–11452. 10.1021/acs.est.0c03632. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Sandau CD, Ayotte P, Dewailly E, Duffe J, Norstrom RJ, 2000. Analysis of hydroxylated metabolites of PCBs (OH-PCBs) and other chlorinated phenolic compounds in whole blood from Canadian Inuit. Environ. Health Perspect 108, 611–616. 10.2307/3434880. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Sano S, Kawanishi S, Seki YY, 1985. Toxicity of polychlorinated biphenyls with sepcial reference to porphyrin metabolism. Environ. Health Persp 59, 137–143. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Sethi S, Keil KP, Chen H, Hayakawa K, Li XS, Lin YP, Lehmler HJ, Puschner B, Lein PJ, 2017. Detection of 3,3′-Dichlorobiphenyl in human maternal plasma and its effects on axonal and dendritic growth in primary rat neurons. Toxicol. Sci 158. 401–411. 10.1093/toxsci/kfx100. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B, Ideker T, 2003. Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res. 13, 2498–2504. 10.1101/gr.1239303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Shi X, Wahlang B, Wei XL, Yin XM, Falkner KC, Prough RA, Kim SH, Mueller EG, McClain CJ, Cave M, Zhang X, 2012. Metabolomic analysis of the effects of polychlorinated biphenyls in nonalcoholic fatty liver disease. J. Proteome Res 11, 3805–3815. 10.1021/pr300297z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Uppal K, Ma CY, Go YM, Jones DP, 2018. xMWAS: a data-driven integration and differential network analysis tool. Bioinformatics 34, 701–702. 10.1093/bioinformatics/btx656. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Vansell NR, Klaassen CD, 2002. Increase in rat liver UDP-glucuronosyltransferase mRNA by microsomal enzyme inducers that enhance thyroid hormone glucuronidation. Drug Metab. Dispos 30, 240–246. 10.1124/dmd.30.3.240. [DOI] [PubMed] [Google Scholar]
  59. Wang L, Zhang SP, Lin RH, Li L, Zhang DQ, Li XH, Liu SJ, 2013. PCB-77 disturbs iron homeostasis through regulating hepcidin gene expression. Gene 532, 146–151. 10.1016/j.gene.2013.09.023. [DOI] [PubMed] [Google Scholar]
  60. Wilson RJ, Suh YP, Dursun I, Li XS, da Costa Souza F, Grodzki AC, Cui JY, Lehmler H-J, Lein PJ, 2024. Developmental exposure to the Fox River PCB Mixture modulates behavior in juvenile mice. Neurotoxicology 103, 146–161. 10.1016/j.neuro.2024.06.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  61. Yu TW, Park Y, Johnson JM, Jones DP, 2009. apLCMS-adaptive processing of high-resolution LC/MS data. Bioinformatics 25, 1930–1936. 10.1093/bioinformatics/btp291. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Zarerad E, Niksalehi K, Armandeh M, Sani MA, Ataei M, Mousavi T, Maghsoudi AS, Hassani S, 2023. Polychlorinated biphenyls: a review of recent updates on food safety and environmental monitoring, health and toxicological implications, and analysis. Mini-Rev. Med. Chem 23, 1390–1411. 10.2174/1389557523666221213091445. [DOI] [PubMed] [Google Scholar]
  63. Zhang CY, Flor S, Ruiz P, Dhakal R, Hu X, Teesch LM, Ludewig G, Lehmler HJ, 2020. 3,3′-Dichlorobiphenyl Is metabolized to a complex mixture of oxidative metabolites, including novel methoxylated metabolites, by hepG2 cells. Environ. Sci. Technol 54, 12345–12357. 10.1021/acs.est.0c03476. [DOI] [PMC free article] [PubMed] [Google Scholar]
  64. Zhang CY, Flor S, Ruiz P, Ludewig G, Lehmler HJ, 2021a. Characterization of the metabolic pathways of 4-chlorobiphenyl (PCB3) in hepG2 cells using the metabolite profiles of its hydroxylated metabolites. Environ. Sci. Technol 55, 9052–9062. 10.1021/acs.est.1c01076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Zhang CY, Klocke CR, Lein PJ, Lehmler HJ, 2021b. Disposition of PCB 11 in mice following acute oral exposure. Chem. Res. Toxicol 34, 988–991. 10.1021/acs.chemrestox.1c00067. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Zhang CY, Li XS, Flor S, Ruiz P, Kruve A, Ludewig G, Lehmler HJ, 2022. Metabolism of 3-chlorobiphenyl (PCB 2) in a human-relevant cell line: evidence of dechlorinated metabolites. Environ. Sci. Technol, 12460–12472. 10.1021/acs.est.2c03687. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2025488-supplement-1.docx (338.6KB, docx)
2
NIHMS2025488-supplement-2.xlsx (181.8KB, xlsx)

Data Availability Statement

The data used in the manuscript are freely available through Iowa Research Online at https://doi.org/10.25820/data.007305.

RESOURCES