Disposition and Metabolomic Effects of 2,2’,5,5’-Tetrachlorobiphenyl in Female Rats Following Intraperitoneal Exposure

Abstract

The disposition and toxicity of lower chlorinated PCBs (LC-PCBs) with less than five chlorine substituents have received little attention. This study characterizes the distribution and metabolomic effects of PCB 52, an LC-PCB found in indoor and outdoor air, three weeks after intraperitoneal exposure of female Sprague Dawley rats to 0, 1, 10, or 100 mg/kg BW. PCB 52 exposure did not affect overall body weight. Gas chromatography-tandem mass spectrometry (GC-MS/MS) analysis identified PCB 52 in all tissues investigated. Hydroxylated, sulfated, and methylated PCB metabolites, identified using GC-MS/MS and nontarget liquid chromatography-high resolution mass spectrometry (Nt-LCMS), were primarily found in the serum and liver of rats exposed to 100 mg/kg BW. Metabolomic analysis revealed minor effects on L-cysteine, glycine, cytosine, sphingosine, thymine, linoleic acid, orotic acid, L-histidine, and erythrose serum levels. Thus, the metabolism of PCB 52 and its effects on the metabolome must be considered in toxicity studies.

1. Introduction

Polychlorinated biphenyls (PCBs), a class of 209 industrial chemicals, were manufactured and sold as complex mixtures of over 150 individual PCB congeners. Because of their physicochemical properties, they were used in various commercial applications, including insulating fluids in transformers, building materials, adhesives, and resins (ATSDR, 2000; Faroon and Ruiz, 2016; Markowitz and Rosner, 2018). The industrial production of PCBs has been banned worldwide because of their environmental persistence, long-range transport in the environment, the potential to bioaccumulate and biomagnify in aquatic and terrestrial food chains, and environmental and human health concerns (Stockholm-Convention, 2019). PCBs have been released into the environment and can be found, for example, in soils, sediments, water sources, wildlife, and air (Hu et al., 2011; Marek et al., 2017). In addition, PCBs continue to be inadvertently produced by various industrial processes and can be found, for example, in certain paints and adhesives used in kitchen cabinets (Anezaki and Nakano, 2013; Herkert et al., 2018; Hu and Hornbuckle, 2010).

PCBs and their metabolites are present in human tissues because of the widespread environmental contamination with PCBs (Ampleman et al., 2015; Li et al., 2022; Sethi et al., 2019). Exposure of the general population to PCBs primarily occurs via contaminated foods like fish, meats, and dairy (Saktrakulkla et al., 2020; Schecter et al., 2010). In addition, growing evidence demonstrates that inhalation of indoor or outdoor air represents a current route of human PCB exposure (Ampleman et al., 2015; Harrad et al., 2006; Kaifie et al., 2019; Meyer et al., 2013). Of particular concern is the presence of PCBs in the indoor air of schools (Egsmose et al., 2016; Marek et al., 2017). Therefore, exposures to PCBs represent a current human health concern (Agus et al., 2022; Hammel et al., 2023). PCB exposure has been linked to cancer and non-cancer endpoints (Carlson et al., 2022; 2016). Importantly, laboratory and population-based studies implicating higher chlorinated PCBs (i.e., congeners with > 4 chlorine substituents) and their metabolites in neurotoxic outcomes across the lifespan (Bullert et al., 2021; Gaum et al., 2014; Klocke and Lein, 2020; Pessah et al., 2010). Lower chlorinated PCBs have not been studied as extensively for their neurotoxic effects as compared to their association with cancer and other outcomes (Deen et al., 2023; Hammel et al., 2023; Kofoed et al., 2021).

PCB 52, a tetrachlorinated biphenyl, is a major constituent of technical PCB mixtures (Frame, 1997; Frame et al., 1996) and has been widely detected in air and human samples (Grimm et al., 2015b; Hammel et al., 2023). This tetrachlorinated PCB congener is metabolized via an arene oxide intermediate and by direct insertion of oxygen into a C-H bond to 3-OH-PCB 52, 4-OH-PCB 52, and 4,5-diOH-PCB 52 by rat liver microsomes (Borlakoglu et al., 1991; Borlakoglu and Wilkins, 1993; Forgue et al., 1979). CYP2A6 is the cytochrome P450 (CYP) isoform that forms 4-OH-PCB 52 in humans (Shimada et al., 2016). The CYPs involved in the metabolism of PCB 52 in rodents are unknown; however, PCB 52 is likely oxidized by rats CYP2B1, similar to PCB 95 and PCB 136 (Lu et al., 2013; Lu and Wong, 2011). OH-PCB 52 metabolites can undergo further oxidation, glucuronidation, and sulfation, resulting in complex metabolite mixtures (Grimm et al., 2015b). Although monohydroxylated PCB 52 metabolites have not been detected in human samples, the corresponding PCB 52 sulfate metabolites have been recently detected in human populations (Zhang et al., 2022b).

PCB metabolites can exhibit toxicity, including endocrine disruption and carcinogenic and neurologic effects. For example, OH-PCBs interact with estrogen receptors as either agonists or antagonists, with some OH-PCBs being more effective agonists than estradiol based on their interaction with the estrogen receptors (Grimm et al., 2015b; Tehrani and Van Aken, 2014). 4-OH-PCB 52 was found to have a low affinity for the thyroid hormone receptor and display low estrogenic activity in vitro (Kitamura et al., 2005). PCBs and OH-PCBs can be further metabolized to potentially toxic PCB sulfates by sulfotransferase enzymes (SULT) (Grimm et al., 2015b). 4-OH-PCB 52 and the corresponding sulfate metabolite are inhibitors of SULT2A1, while the PCB 52 sulfate is an inhibitor of SULT1E1 (Parker et al., 2018). In addition, OH-PCB 52 and its PCB 52 sulfate displayed toxicity in the low micromolar range in two neural cell lines, the N27 and SH-SY5Y cell lines (Rodriguez et al., 2018). PCB 52 can also be metabolized to redox reactive quinone metabolites, as demonstrated by the presence of quinonoid-derived protein adducts in the liver and brain of PCB 52 exposed Sprague-Dawley rats (Lin et al., 2000). PCB quinones and other PCB metabolites can lead, for example, to gene mutations, inhibition of nuclear protein topoisomerase, and a reduction in telomerase activity (Grimm et al., 2015b; McLean et al., 1996; Song et al., 2008).

This exploratory study characterizes the disposition of PCB 52 and its metabolites in female Sprague Dawley rats exposed to different doses of PCB 52 using both targeted and nontargeted approaches and assesses changes in endogenous metabolites implicated in PCB neurotoxicity. Our results reveal minor the transformation of PCB 52 to a complex metabolite mixture in rats.

2. Materials and Methods

2.1. Chemicals

The sources of the test compound, PCB 52, and the analytical standards are described in the Supplementary Material. Unique chemical identifiers are summarized in Table S1. Pesticide grade solvents, including hexane and dichloromethane, and mass spectrometry grade solvent, such as acetonitrile, were purchased from Fisher Scientific.

2.2. Animal exposure

All experiments were conducted with the approval of the Institutional Animal Care and Use Committee of the University of Iowa. Female Sprague Dawley (SD) rats, weighing 75–100 g at 4–5 weeks old, were supplied by Harlan Sprague-Dawley/Envigo (Indianapolis, Indiana, USA). The animals were provided a standard rodent chow and water ad libitum. After a week of acclimation, rats were administered a single intraperitoneal (IP) injection of corn oil (vehicle control) or PCB 52 in corn oil at PCB 52 doses of 1 mg/kg, 10 mg/kg, or 100 mg/kg BW (body weight). While these doses are higher than the Tolerable Daily Intake (TDI) of PCBs for humans of 1.6 μg per day for an 80 kg individual (Faroon, 2003), the use of high doses and short study periods is a well-established approach for generating data useful for risk assessment purposes (Doull, 2003). Three weeks after PCB 52 administration, all rats were euthanized using carbon-dioxide asphyxiation followed by thoracotomy. After euthanasia, whole blood was obtained from the heart and allowed to rest at room temperature for at least 30 min, followed by centrifuge at 2,500 g for 10 min at 4°C to prepare serum. The serum was then aliquoted into glass vials. The livers, spleen, and thymus were excised en bloc, flash-frozen in liquid nitrogen. All samples were stored at −80°C until further analysis.

2.3. Targeted GC-MS/MS analysis of PCB 52 and metabolites

2.3.1. Extraction of PCB 52 and metabolites from tissues

PCB 52 and its hydroxylated metabolites were extracted from the adipose (109 ± 13 mg range: 95 – 139 mg), brain (648 ± 77 mg range: 460 – 760 mg), and liver (509 ± 4 mg range: 503 –515 mg) using the “Jensen” extraction method (Jensen et al., 2003). Briefly, tissues were homogenized in 3 mL of isopropanol with 1 mL of diethyl ether, as described (Kania-Korwel et al., 2008b; Wu et al., 2015). Recovery standards (PCB 77, 20 ng in isooctane; 4’-OH-PCB 159, 20 ng in methanol) were added to each sample. The samples were capped, inverted for 5 min, and centrifuged at 1,690 g for 5 min. Next, the supernatant was transferred to a new glass tube containing 5 mL of 0.1 M phosphoric acid in 0.9% NaCl solution. The tissue pellets were resuspended with 1 mL of isopropanol and 2.5 mL hexane-diethyl ether (9:1, v/v), followed by vortexing. Samples were then inverted and centrifuged as described above, and the supernatant was combined with the previous extract. Extracts were inverted and centrifuged as described above to facilitate phase separation. The organic phase was transferred to a new vial while the aqueous phase was re-extracted with 3 mL of the hexane-diethyl ether solution (9:1, v/v). The combined organic phases were concentrated under a slow nitrogen stream.

Five drops of methanol and 0.5 mL of diazomethane (about 5 mmol of diazomethane) in diethyl ether were added to the tissue extracts (Kania-Korwel et al., 2008b). The samples were stored at 4° C for at least 3 h to methylate hydroxylated PCB 52 metabolites. After derivatization, excess diazomethane was evaporated under a gentle stream of nitrogen, and samples were loaded onto glass solid phase extraction (SPE) cartridges containing 0.2 g of activated silica gel (bottom) and 2 g of acidified silica gel (silica gel: H₂SO₄, 2 :1, w/w; top). Analytes were eluted with 10 mL of dichloromethane, and the extracts were concentrated to near dryness under a gentle nitrogen stream. The tissue extracts were treated with 4 mL of concentrated sulfuric acid for further lipids removal. The organic phase was transferred to a new vial, concentrated to near dryness, and samples were spiked with the internal standard (PCB 204, 20 ng in isooctane) for gas chromatographic analysis.

2.3.2. Extraction of PCB 52 and metabolites from serum for GC-MS/MS analysis

Serum (342 ± 92 mg range: 93 – 477 mg) samples were processed using a previously validated extraction protocol (Li et al., 2019). Briefly, 3 mL of aqueous 1% KCl solution was added to each serum sample, followed by the recovery standards (PCB 77, 20 ng in isooctane; 4’-OH-PCB 159, 20 ng in methanol). Subsequently, 1 mL of 6 M HCl, 5 mL of 2-propanol, and 5 mL of a hexanes-MTBE mixture (1:1 v/v) were added to each sample, vortexing between each solvent addition. The samples were inverted and centrifuged to facilitate separation between the organic and aqueous phases. The organic phases were transferred to a new glass tube, while the aqueous phase was re-extracted with 3 mL of hexanes. The combined organic phase was washed with 3 mL of aqueous 1% KCl and concentrated under a gentle stream of nitrogen. The extracts were diluted with 0.5 mL, derivatized with diazomethane as described above for tissue extracts, and concentrated with a gentle nitrogen stream. The internal standard (PCB 204, 20 ng in isooctane) was added before the gas chromatographic analysis. Quality assurance samples were analyzed in parallel with all samples, as described for the tissue samples.

2.3.3. GC-MS/MS analysis of PCB 52 and its metabolites

PCB 52 and its mono-hydroxylated metabolites (as methylated derivatives) were quantified using the multiple reaction monitoring setting (MRM) on an Agilent 7890 A GC system equipped with an Agilent 7000 Triple Quad and an Agilent 7693 autosampler (GC-MS/MS) from Agilent Technologies, Inc. (Santa Clara, CA, USA). Separations were performed with an SPB-Octyl capillary column (30 m length, 25 mm inner diameter, 0.25 μm film thicknesses: Supelco, Bellefonte, Pennsylvania, USA). Samples were injected in the solvent vent injection mode with a helium (carrier gas) flow of 0.75 mL/min and nitrogen as the collision gas. The temperature program was set as follows: Initial temperature of 45 °C, hold for 2 minutes, 100 °C/min to 75 °C, hold for 5 mins, 15 °C/min to 150 °C, hold for 1 minute, 2.5 °C/min to 280 °C, and final hold of 5 minutes. Precursor and product ions of all analytical standards and the corresponding collision energy are summarized in Table S2.

2.3.4. GC-MS/MS analysis quality assurance and quality control

All PCB 52 analyses followed validated standard operating procedures (SOPs) to extract and quantify PCB 52 and its hydroxylated metabolites. Surrogate recovery standards, PCB 77 and 4’-OH-PCB 159, were added to each sample to assess the precision and reproducibility of the PCB and OH-PCB measurements. The levels of PCB 52 and its primary metabolite, 4-OH-PCB 52, were corrected based on the recoveries of the corresponding recovery standard (Table S3). In parallel, the precision and recoveries of the analyses were assessed by spiking sample blanks and matched blank tissue matrices with known amounts of PCB 52 and 4-OH-PCB 52 (Table S4). Detection limits were determined from the method blanks or blank tissues analyzed in parallel with the samples (Table S5).

2.4. Nontarget liquid chromatography-high resolution mass spectrometric analysis (Nt-LCMS)

2.4.1. Extraction of PCB 52 and metabolites from serum or tissues

Brain (213 ± 4 mg, range: 203 – 217 mg), liver ( 249 ± 2 mg, range: 245 – 254 mg), and serum samples (127 ± 57 mg, range: 51 – 282 mg) were processed using a QuEChERS method, as described with modifications (Li et al., 2022; Li et al., 2021). Briefly, brain or liver tissue was homogenized using a TissueRuptor from Qiagen (Germantown, Maryland, USA) in 2 mL of water, followed by adding the surrogate standard, F-OH-PCB 3 (50 ng in acetonitrile). Similarly, serum was added to 2 mL of water and vortexed, followed by the addition of the surrogate standards (50 ng each). After vortexing, 4 mL of acetonitrile was added. Then, a mixture of 200 mg of NaCl and 800 mg of MgSO₄ was added, and samples were capped and shaken vigorously. Samples were then inverted for 5 minutes and centrifuged (1811 g) for 5 minutes to facilitate phase separation.

Hybrid phospholipid solid phase extraction (HybridSPE) cartridges (100 mg/3 mL, Sigma-Aldrich, St. Louis, Missouri, USA) were modified by adding 3 g of a MgSO₄:Na₂SO₄ mixture (1:1, w:w) and preconditioned with 3 mL of acetonitrile. The acetonitrile layer from the QuEChERS extraction was then passed through the HybridSPE cartridge. The aqueous phase was re-extracted with 1 mL of acetonitrile, vortexed, and centrifuged for 5 minutes. The organic layer from the second extraction step was also passed through the HybridSPE cartridge, and the metabolites were eluted using 3 mL of acetonitrile. Next, the samples were evaporated to dryness at 35 °C using a Savant SpeedVac SPD210 (ThermoFisher Scientific, Waltham, Massachusetts, USA). Next, the samples were reconstituted in 300 μL of acetonitrile, vortexed, and centrifuged for 5 minutes. The supernatants were transferred to a microcentrifuge tube and spiked with perfluorooctanesulfonic acid potassium salt (50 ng). The solvent was again evaporated to dryness at 35 °C using the SpeedVac. The resulting extracts were reconstituted with 200 μL of water:acetonitrile (1:1, v/v) and vortexed. Finally, the samples were kept at −20 °C for at least 30 minutes, centrifuged repeatedly at 4 °C for 10 minutes at 16,000 g, and stored at −80 °C before Nt-LCMS analysis.

2.4.2. Liquid chromatography-high resolution mass spectrometry analysis

Serum and tissue extracts were analyzed on a Q-Exactive Orbitrap Mass Spectrometer (ThermoFisher Scientific) with a Vanquish Flex ultra-high-performance liquid chromatograph (ThermoFisher Scientific) with an ACQUITY UPLC-C18 column (particle size: 1.7 μm, 2.1 × 100 mm, Waters, Milford, Massachusetts, USA) at the High-Resolution Mass Spectrometry Facility of the University of Iowa (Zhang et al., 2020). Water and acetonitrile were used as mobile phase A and B, with a flow rate of 0.3 mL/min. The pressure of the chromatographic system ranged from 4000 to 8000 psi. The UHPLC gradient program was as follows: starting at 5% B, held for 1 min, increased linearly to 95% B, held for 3 min, and returning to 5% B, with a hold for 4 min before the next injection. The injection volume was 2 μL. The Q-Exactive Orbitrap Mass Spectrometry was operated in the negative polarity mode. The spray voltage and current were 2472 V and 18.2 μA. The flow rate of sheath and auxiliary gas were 48 and 2 mL/min, respectively. The capillary and auxiliary temperature were 256 °C and 413 °C, respectively. The analyses were performed in the full scan mode with a range of 85 to 1000 m/z. The autogain control target setting was 1×106, the full scan resolution setting was 70,000 and maximum interval time (IT) was 200 ms.

2.4.3. Nt-LCMS analysis quality assurance and quality control

Solvent (50% water/50% acetonitrile) and method blanks were used in the Nt-LCMS to monitor carryover. No OH-PCB metabolites were detected in these blank samples. The average recoveries of the surrogate standard (i.e., 3-F-PCB3 sulfate) were 74 ± 18% (range 45–130%, N = 47).

2.4.4. Processing and visualization of Nt-LCMS data

The chromatographic peaks of suspected PCB metabolites were extracted from the acquired data (as .raw files) with Thermo Xcalibur (version 4.1, Thermo Fisher Scientific, USA). The chromatogram extraction was performed with a m/z tolerance of 5 ppm, mass precision decimals of 5, and a smoothing factor of 7. The isotopic pattern of chlorine was used as an important factor for confirming putative chlorine compounds, as described (Li et al., 2022). The chromatographic and mass spectrometric data were exported, and figures were prepared with GraphPad Prism (version 9.4.1, GraphPad Software, USA).

2.5. Targeted metabolomics of serum samples

2.5.1. Sample preparation

Targeted metabolomic analyses were performed by the Metabolomics Core Facility of the Fraternal Order of Eagles Diabetes Research Center at the University of Iowa (Iowa City, Iowa, USA) using standard protocols (Tompkins et al., 2019). Briefly, the serum samples were extracted in ice-cold 1:1 methanol/acetonitrile containing a mixture of nine internal standards (d₄-Citric Acid, ¹³C₅-Glutamine, ¹³C₅-Glutamic Acid, ¹³C₆-Lysine, ¹³C₅-Methionine, ¹³C₃-Serine, d₄-Succinic Acid, ¹³C₁₁-Tryptophan, d₈-Valine; Cambridge Isotope Laboratories, 1 μg/mL each). The ratio of extraction solvent to sample volume was 18:1. The samples were then incubated at −20°C for 1 h, followed by a 10-minute centrifugation. Supernatants were transferred to fresh tubes. Pooled quality control (QC) samples were prepared by adding an equal volume of each sample to a fresh 1.5 mL microcentrifuge tube. Processing blanks were created by adding extraction solvent to microcentrifuge tubes. Serum samples, QC samples, and processing blanks were evaporated using a Savant SpeedVac SPD210 (ThermoFisher). The resulting dried extracts were derivatized using methyoxyamine hydrochloride (MOC) and N,O-bis(trimethylsilyl)trifluoroacetamide (TMS) (Sigma-Aldrich, St. Louis, MO, USA). The dried extracts were reconstituted in 60 μL of 11.4 mg/mL MOC in anhydrous pyridine (VWR International, Radnor, PA, USA), vortexed for 5 min, and heated for 1 hour at 60°C. Next, 40 μL TMS was added to each sample, and samples were vortexed for 1 minute before heating for 30 minutes at 60°C. The serum, QC, and blank samples were immediately analyzed using GC-Orbitrap MS.

2.5.2. Targeted analysis of serum sample using GC-Orbitrap MS

Gas chromatographic separation was conducted on a ThermoFisher Scientific Trace 1300 GC with a TraceGold TG-5SilMS column (0.25 μm film thickness; 0.25 mm ID; 30 m length; ThermoFisher Scientific), as described (Tompkins et al., 2019). The injection volume of 1 μL was used for all samples. The gas chromatograph was operated in split mode with the following settings: 20:1 split ratio; split flow: 24 μL/min, purge flow: 5 mL/min, Carrier mode: Constant Flow, Carrier flow rate: 1.2 mL/min). The inlet temperature was 250°C. The oven temperature gradient was: 80°C for 3 minutes, ramped at 20°C/minute to a maximum temperature of 280°C, held for 8 minutes. The injection syringe was washed 3 times with methanol and 3 times with pyridine between samples. Metabolites were detected using a ThermoFisher Scientific Q-Exactive Plus mass spectrometer. The data were acquired from 3.90 to 21.00 minutes in EI mode (−70eV) in full scan (m/z 56.7–850) at 60K resolution. Metabolite profiling data were analyzed using Tracefinder 4.1, utilizing standard verified peaks and retention times.

2.5.3. Analysis of targeted serum metabolomics data

Targeted metabolomic data were collected with a target list containing 94 metabolites. Metabolites with > 50 % missing values were removed from further analysis. The remaining missing values were imputed with the one-half lowest intensity detected for those metabolites across all samples. Metabolites detected with a relative standard deviation > 30 % in multiple injected QC samples were removed from the dataset. As a result, 81 metabolites were included for further data processing. The data were normalized by the sum of total metabolite intensity and log2 transformed. Data variability and group clustering were visualized with score plots in principal component analysis (PCA) and partial least squares-discriminant analysis (PLS-DA). Metabolites that drive the sample clustering among exposure groups were indicated with the variable importance in projection (VIP) values in the PLS-DA. Further univariate analysis was performed with a linear regression model using the LinearModelFit function in the R package NormalizeMets (De Livera et al., 2018) to generate raw p-values reflecting the correlation between metabolite intensities and dose levels. Raw p-values were adjusted for the false discovery rate with the Benjamini and Hochberg method (Benjamini and Hochberg, 1995). Analyses were performed with MetaboAnalyst 4.0 (Chong et al., 2018) or R (version 4.1.2). No pathway analysis was performed since only a limited number of metabolites (9 out of 81) were obtained with raw p-values < 0.05.

2.6. Dopamine and metabolites analysis

Brains were isolated and nigrostriatal tissue collected for further analysis. Tissue was homogenized in ice-cold 0.3 M perchloric acid and stored at −80°C. Prior to HPLC analysis, the homogenate was centrifuged at 10,000 g for 10 min and supernatant injected into the HPLC for neurochemistry analysis. The weight of tissue collected was used to normalize the pmol of dopamine and dopamine metabolites detected to mg of tissue. Dopamine and its metabolites, 3,4-dihydroxyphenylacetic acid (DOPAC) and 3,4-dihydroxyphenylacetaldehyde (DOPAL), were analyzed via an Agilent 1100 Series capillary HPLC system with an ESA Coulochem III coulometric electrochemical detector (Enayah et al., 2018). Separation was achieved with a Phenomenex Synergi C18 column (2 × 150 mm, 40 Å) using an isocratic mobile phase (50 mM citric acid, 1.8 mM sodium heptane sulfonate, 0.2% trifluoroacetic acid, 2% acetonitrile, pH = 3.0) at 200 μL/min. For electrochemical detection of catechol-containing compounds, the guard cell, electrode 1, and electrode 2 were set to +350, −150, and +200 mV, respectively.

2.7. Statistical analysis

Data are presented as means ± standard deviation. The differences among means were compared using one-way or two-way analysis of variance (ANOVA) with Tukey’s post hoc method testing in GraphPad Prism (version 9.0.0.), as appropriate. P-values less than 0.05 were considered statistically significant. The original data underlying this study are openly available through the Iowa Research Online repository at DOI: 10.25820/data.006225.

3. Results and Discussion

3.1. General toxicity

Exposure to PCB 52 had no significant impact on the overall body weight of the adolescent rats (Fig. 1a). In contrast, other studies have shown that PCB exposure increases mammalian body weights during critical growth phases, such as adolescence (Birnbaum, 1994; Corey et al., 1996). This increase is due to altered hormone homeostasis, inflammation of organs, and increasing organ weights affecting overall body weight following PCB exposure. PCB 52 exposure did not affect spleen and thymus weights (Figs. S1 and S2), while the bodyweight-adjusted livers of animals exposed to 10 mg/kg BW of PCB 52 were significantly smaller than the vehicle group (Fig. 1b). In an earlier study, bodyweight-adjusted liver weights from male rats exposed to a single IP dose (150 mg/kg BW) of PCB 52 where not significantly different than their control counterparts (Table S6) (Parkinson et al., 1983). In contrast, exposure to many PCB mixtures or higher chlorinated PCBs typically results in an enlarged liver (ATSDR, 2000; Parkinson et al., 1983).

Fig. 1.

PCB 52 exposure did not affect body weights but altered relative liver weights following intraperitoneal (IP) exposure of female rats to different doses of PCB 52. (A) Growth curve of body weights throughout the 3-week study, measured at 5 different time points post IP injection of PCB 52. (B) Body weight adjusted liver weights. A significant decrease in liver weights was measured 3 weeks post IP injection of PCB 52 for rats exposed to 10 mg/kg BW. Group means were compared using one-way ANOVA followed by Tukey’s pairwise comparisons. A p-value < 0.05 was considered statistically significant.

3.2. PCB tissue distribution

PCB 52 is a potentially neurotoxic PCB congener that, based on congener-specific PCB analyses, is frequently detected in air and human serum form the United States (summarized in (Grimm et al., 2015b). PCB 52 was present in all tissues investigated, including the brain, three weeks following the IP injection (Fig. 2, Table S7). Except for the high dose group, PCB 52 levels followed the rank order adipose > liver ~ brain ~ blood across all three exposure groups. In the high-dose group, the mean PCB 52 levels in the liver were higher than in the brain and blood. Disposition studies with PCB 95, a PCB congener structurally related to PCB 52, show a similar rank order in mice, with adipose ≫ liver > brain > blood (Kania-Korwel et al., 2015; Kania-Korwel et al., 2012). In the present study, PCB 52 levels were 50 to 100 times higher in the adipose tissue than in all other tissues investigated. The high PCB 52 levels in adipose tissue, the primary storage site of PCBs (Kania-Korwel et al., 2008a), are due to its high triglyceride content compared to the other tissues (McCann et al., 2021; Seelbach et al., 2010; Tan et al., 2004). PCB 52 was also detected in the liver, the primary site of oxidative PCB metabolism, in part because the blood vessels of the peritoneum drain directly into the liver, contributing to a sustained release of PCB 52 into the liver (Al Shoyaib et al., 2019).

Fig. 2.

PCB 52 and its mono-hydroxylated metabolite levels were detected in (a) adipose, (b) brain, (c) liver, and (d) serum. Levels expressed as ng/g wet weight are log-transformed to facilitate comparison across tissues with 1 (orange), 10 (blue), and 100 (purple) mg/kg BW, respectively. Data are reported as mean ± standard deviation (N = 3 to 5). ND, not detected.

As ligands for the transport protein transthyretin (TTR) (ATSDR, 2000; Chauhan et al., 2000; Grimm et al., 2015b; 2016), PCBs may cross the blood-brain barrier bound to TTR. However, as lipophilic compounds, PCBs are more likely partition into the brain by passive diffusion. Based on studies of the partitioning of PCBs into lipid bilayers, PCB 52 is expected to cross the blood-brain barrier (Totland et al., 2016). Indeed, PCB 52 was present in the brain (Fig. 2, Table S7). PCB 52 levels in the brain were about 50 to 100 times lower than the adipose tissue across all three exposure groups. The extractable lipid content in the mouse brain, expressed as a percent of tissue wet weight, is only 8-fold lower than the lipid content in the adipose tissue (9.3% vs. 72%, respectively) and, thus, cannot explain the drastic difference in PCB 52 levels in both tissues (Wu et al., 2015). Lower PCB levels in the brain than in adipose tissue have been observed in other animal studies and result from the different lipid composition of the brain compared to the adipose tissue (ATSDR, 2000). For example, a composition-based model predicting the PCB disposition in mice poorly approximated the distribution of PCBs from the blood into the brain, most likely because published partition coefficients do not reflect the partitioning of PCBs into brain lipids (Li et al., 2020).

PCB 52 levels in all tissues increased dose-dependent (Fig. 2, Table S7). The tissue levels did not reflect the 10-fold differences between doses. The mean PCB 52 levels were approximately 4.0 (serum) to 7.0 (liver) times higher in the tissues from the 10 mg/kg BW than the 1 mg/kg BW dose group. The ratio of the mean PCB 52 tissue levels ranged from only 2.9 (brain) to 5.6 when comparing the 100 mg/kg BW to the 10 mg/kg BW dose group. Other studies have also reported that the PCB tissue ratios between different dose groups do not reflect the fold difference of the dose. For example, female mice exposed orally for 39 days to different doses of PCB 95 showed a 2.5- to 4-fold increase in PCB 95 tissue levels despite a 6-fold difference in the PCB 95 dose (Kania-Korwel et al., 2012). These differences likely reflect the transient induction of hepatic CYPs involved in the oxidative metabolism of PCBs, as suggested by the earlier PCB 95 disposition study in mice. Because CYP2B enzyme activity in the rat liver is induced by PCB 52 exposure (Parkinson et al., 1983), enzymes involved in the metabolism of structurally related PCBs (Lu and Wong, 2011), it is likely that CYPs involved in PCB 52 were transiently induced, thus increasing the elimination of PCB 52 at the higher dose levels. Alternatively, growth dilution could explain the fold differences between exposure groups (Hansen and Welborn, 1977); however, this explanation is unlikely because we did not observe differences in the growth rate (Fig. 1a).

3.3. Tissue distribution of monohydroxylated PCB 52 metabolites

PCBs with a 2,5-chlorine pattern, like PCB 52, are susceptible to CYP-catalyzed direct electrophilic addition of oxygen to a C-H bond or epoxide formation, ultimately resulting in the formation of isomeric OH-PCBs (Grimm et al., 2015b). Evidence from laboratory and population-based studies demonstrates that these OH-PCBs, like the parent compounds, are also neurotoxic (Grimm et al., 2015b; Ludewig et al., 2008; Tehrani and Van Aken, 2014). OH-PCB isomers may have different modes of action, as shown, for example, by their activity at the level of the ryanodine receptor (Niknam et al., 2013; Sethi et al., 2019). Therefore, we determined the levels of OH-PCBs in all four target tissues using GC-MS/MS (Fig. 2).

We detected two metabolites, X1-PCB 52 and 4-OH-PCB 52, in several tissues. The unknown metabolite, X1-PCB 52, is probably 3-OH-PCB 52 (2,2’,5,5’-tetrachlorobiphenyl-3-ol). This assignment is based on the slightly shorter relative retention time of X1-PCB 52 than 4-OH-PCB 52 (relative retention times of 0.788 and 0.80, respectively). Similarly, the retention times of meta vs. para hydroxylated OH-PCBs of multi-ortho chlorinated PCB congeners follow the order 1,2-shift << meta < para hydroxylated metabolite (Sundström and Jansson, 1975; Uwimana et al., 2018). Moreover, 3-OH-PCB 52 was formed by phenobarbital-induced rat liver microsomes suggesting a role of CYP2B enzymes in the oxidative metabolism of PCB 52 (Lin et al., 2000). However, studies using an authentic analytical standard are needed to confirm this tentative identification of X1-PCB 52. Levels of OH-PCBs across all tissues were lower compared to PCB 52 levels, consistent with an earlier disposition study with other PCB congeners in rats (Dhakal et al., 2014).

X1-PCB 52 was present in the adipose tissue from the high-dose group but below the LOD in all other exposure groups (Fig. 2). 4-OH-PCB 52 was detected only in adipose tissue from one animal in the high-dose group. Previous work from our group also did not detect OH-PCBs in the adipose of PCB-exposed mice (Kania-Korwel et al., 2015; Kania-Korwel et al., 2012; Wu et al., 2020). In contrast, other PCB metabolites have been discovered in other organisms, such as MeSO₂-PCB metabolites in the blubber of gray seals (Jensen et al., 1979). MeSO₂-PCBs were also some of the first PCB metabolites identified in human adipose tissue from a woman exposed via the occupational handling of PCB-containing capacitors (Yoshida and Nakamura, 1977). Primarily higher chlorinated PCB metabolites have been reported previously in human liver and adipose samples (Guvenius et al., 2002). These results suggest that some PCB metabolites, including OH-PCBs, can be present at low levels in the adipose and, in addition to the parent PCBs, may contribute to toxic outcomes. In a recent study in which human preadipocytes were treated with PCB 52 or 4-OH-PCB 52 in vitro, we found that 4-OH-PCB 52 caused both different and considerably more changes in the number of differentially expressed genes as compared to PCB 52 (Gourronc et al., 2023), suggesting that OH-PCBs are more biologically active and potentially more toxic to adipose than the parental PCB. Similarly, metabolites of other environmental contaminants, such as polycyclic aromatic hydrocarbons (PAHs), can be more toxic than the parent compounds (Lam et al., 2018).

In this study, 4-OH-PCB 52 and X1-PCB 52 were detected in the livers of animals from the 10 mg/kg BW (mean levels of 0.15 and 0.28 ng/g, respectively) and 100 mg/kg BW (mean levels of 0.57 and 0.85 ng/g, respectively) exposure groups. OH-PCBs have been previously identified in the livers of mice and their offspring orally exposed subacutely and subchronically to PCB 95, a PCB congener structurally related to PCB 52 (Kania-Korwel et al., 2012; Kania-Korwel et al., 2017). Interestingly, OH-PCB 95 metabolite levels were 4 to 900-fold higher in these earlier studies than in this study. These differences in the OH-PCB levels are likely due to the different routes of exposure, the dosing paradigms, and structural differences. In human liver samples, OH-PCB levels (∑OH-PCBs between 7 to 175 ng/g lipids) were about 10 to 1000-fold higher than in the present study (Guvenius et al., 2002).

We found estimated mean X1-PCB 52 levels in the serum of 0.07 ng/kg BW for the 10 mg/kg BW and 0.69 ng/g for the 100 mg/kg BW exposure group. In addition, 4-OH-PCB was present in the serum of animals with mean levels of 0.14, 0.70, and 2.27 ng/g for animals exposed to 1, 10, or 100 mg/kg BW PCB 52, respectively (Table S7). Serum PCB metabolites are routinely assessed in rodents (Stamou et al., 2015; Zhang et al., 2021), wildlife (Quinete et al., 2014), and humans (Koh et al., 2016a; Koh et al., 2016b; Quinete et al., 2017; Quinete et al., 2014). Another study with rural and urban participants identified 58 OH-PCBs at mean levels between 0.05 to 0.08 ng/g in the serum of both adults and adolescents (Koh et al., 2016a; Koh et al., 2016b). Previously in humans, 38 OH-PCB metabolites have been reported in plasma, many of which have five or more chlorine substitutes (Hovander et al., 2002). More work reported total median levels of OH-PCBs in different human populations between 0.0002 and 1.6 ng/g wet weight in serum/plasma (Quinete et al., 2014). In addition, lower chlorine PCB metabolites, such as OH-PCB 28 metabolites, are present in human plasma, with total OH-PCB levels ranging from 0.03 to 27.7 ng/mL plasma (Quinete et al., 2017). These total OH-PCB levels are comparable to the OH-PCB 52 levels detected in the serum in this study (Table S7); however, OH-PCB 52 metabolites have not been reported in human serum to date, most likely because the corresponding analytical standards are not commercially available.

X1-PCB 52 and 4-OH-PCB 52 were also observed in the brain of one animal. The presence of hydroxylated PCB 52 metabolites in the brain has not been reported previously. However, a single cysteinyl adduct of a 1,2-benzoquinone of PCB 52 was detected in the brain of rats exposed to PCB 52 (Lin et al., 2000). Several other studies have reported the presence of OH-PCBs in the mammalian brain. One study identified radiolabeled 4-OH-PCB 107 in the forebrain and cerebellum of Wistar dams and their offspring (Meerts et al., 2002). Our group has identified OH-PCB 95 metabolites at low levels in the brain (11 to 14 ng/g tissue across doses from 0.1 to 6.0 mg/kg body weight per day) of C57Bl/6 mice exposed subchronically to different PCB 95 doses (Kania-Korwel et al., 2012). Several OH-PCBs were present in the brains of dogs exposed intraperitoneally to a synthetic PCB mixture (Nomiyama et al., 2019). In the Japanese macaque (Macaca fuscata), OH-PCBs were present in the fetal brain (Nomiyama et al., 2020). Another study did not detect OH-PCBs in the mouse brain, possibly because of the comparatively high detection limit of the analytical method used in these studies. For example, no OH-PCB 136 metabolites were detected in the brain of mice exposed to PCB 136 (Wu et al., 2015). Recently, OH-PCB metabolites of several lower chlorinated PCBs were detected in postmortem human brain tissue (Li et al., 2022).

3.4. Nt-LCMS analysis of PCB 52 conjugates

Complex mixtures of PCB 52 metabolites, including PCB sulfate and other conjugates, may be formed in PCB 52 exposed rats. These metabolites, for example, cysteinyl adducts of PCB quinone metabolites, may also be present in the brain and can be neurotoxic (Grimm et al., 2015b). Because the metabolism of PCB 52 in rats has not been fully characterized, we employed an Nt-LCMS method to semiquantitatively identify PCB 52 metabolites present in target tissues and elucidate the metabolic pathway of PCB 52 in rats. This Nt-LCMS method has been used previously to characterize PCB metabolite profiles in cell culture (Zhang et al., 2020) and animal models (Li et al., 2021). The Nt-LCMS analyses of liver and serum samples revealed several PCB 52 metabolites (Table 1). Several metabolites were only detected in one animal exposed to the high PCB 52 dose. This animal also displayed higher PCB 52 and OH-PCB 52 levels than animals from the same exposure group. Metabolites identified in both tissues included mono-hydroxylated PCB 52 metabolites and the corresponding sulfates. We also observed several methoxylated and dechlorinated metabolites, consistent with previous PCB metabolism studies in cells in culture (Zhang et al., 2020) and in the urine of rabbits exposed to PCB 153 through diet (Hutzinger et al., 1974), the feces of mice, rats, and quails after consumption of PCB 95 (Sundström and Jansson, 1975), or through metabolism of PCB 3 following intraperitoneal exposure (Safe et al., 1975). Importantly, no PCB 52 conjugates were detected in the brain and adipose tissue, including the cysteinyl adduct of a PCB 52 quinone metabolite reported previously (Lin et al., 2000). It is important to emphasize that, due to a 1,2 shift of PCB 52 epoxide intermediates, metabolites may not have retained the 2,2’,5,5’-tetrachloro substitution pattern (Guroff et al., 1967).

Table 1.

Several PCB 52 metabolite classes were detected by LC-Orbitrap MS in tissues from female rats exposed to 100 mg/kg BW PCB 52 via I.P. injection.

Class No	Metabolitesa	Retention time (min)	Formula	Normalized intensityb	[M-H]−		Accurate mass difference, ppmc	Confidence level
					Calculated (Da)	Measured (Da)
Liver
1.1		7.71	C12H5OC14	0.18 ± 0.20	304.91000	304.90991	−0.2952	3
1.2	Tetra-CB sulfate	6.34	C12H5Cl4O4S	0.22 ± 0.34	384.86681	384.86694	0.3378	2
		6.41		0.19 ± 0.29	384.86681	384.86691	0.2598	3
1.3	diOH-Tetra-CB	6.63	C12H5Cl4O2	0.29	320.90491	320.90485	−0.1870	3
		5.42		0.35 ± 0.54	400.86173	400.86176	0.0748	2
1.4	OH-Tetra-CB sulfate	5.62	C12H5O5Cl4S	0.12 ± 0.13	400.86173	400.86172	−0.0249	2
		6.77		0.74	400.86173	400.86145	−0.6985	2
1.5	MeO-OH-T etra-CB	7.86	C13H7O3Cl4	0.09	334.92056	334.9205	−0.1791	3
		5.28		0.05	430.87229	430.87070	−3.6902	2
1.6	MeO-OH-Tetra-CB sulfate	5.47	C13H7O6Cl4S	0.01	430.87229	430.87235	0.1393	2
		5.57		0.02	430.87229	430.87238	0.2089	2
		5.67		0.01	430.87229	430.87188	−0.9516	2
2.1	OH-Tri-CB sulfate	6.01	C12H6Cl3O5S	0.12	366.90070	366.90070	0.0000	2
2.2	MeO-Tri-CB sulfate	5.98	C13H8Cl3O5S	0.04	380.91635	380.91669	0.8926	2
2.3	MeO-diOH-T ri-CB	6.49	C13H8Cl3O3	0.02	316.95445	316.95352	−2.9342	3
2.4	MeO-OH-Tri-CB sulfate	5.47	C13H8Cl3O6S	0.42± 0.20	396.91127	396.91125	−0.0504	2
Serum
3.1		7.70	C12H5OC14	1.10	304.91000	304.90942	−1.9022	3
3.2	Tetra-CB sulfate	6.34	C12H5Cl4O4S	1.04	384.86681	384.86713	0.8315	2
3.3	OH-Tetra-CB sulfatea	6.41	C12H5O5Cl4S	1.47 ± 2.07	400.86173	400.86133	−0.9979	2
		6.63		0.63 ± 0.97	400.86173	400.86066	−2.6693	2
3.4	MeO-OH-Tetra-CB	5.42	C13H7O3Cl4	0.79	334.92056	334.91989	−2.0005	3
3.5	MeO-OH-Tetra-CB sulfate	5.26	C13H7O6Cl4S	0.07	430.87229	430.87260	0.7195	2
		5.47		0.02	430.87229	−	−	2
4.1	OH-Tri-CB sulfate	6.02	C12H6Cl3O5S	0.03	366.90070	366.90036	−0.9267	2
4.2	MeO-OH-Tri-CB sulfate	5.06	C13H8C13O6S	0.10	396.91127	396.90991	−3.4265	2
		5.47		0.37	396.91127	396.91052	−1.8896	2

^aMetabolite also detected in 10 mg/kg PCB 52 exposed animals

^bSome metabolites were only detected in one animal

^cparts per million, ppm = (measured -calculated)/calculated*1000000

CB, chlorobiphenyl

3.4.1. Identification of PCB 52 metabolites in the liver by Nt-LCMS

The Nt-LCMS analysis revealed the presence of several PCB metabolites in the liver (Fig. 3B, Table 1; see also Figs. S3 to S12). For example, two PCB 52 sulfate compounds were detected in the liver, consistent with the presence of two OH-PCB isomers in the GC-MS/MS analysis. OH-PCBs undergo sulfation to PCB sulfates by rat SULT in gene family 1, more specifically SULT1A1 (Liu et al., 2006; Liu et al., 2009); however, the formation of PCB 52 sulfates in rats in vivo has not been reported previously. Moreover, other studies have reported the presence of other PCB sulfates in the liver of rodents exposed to lower chlorinated PCBs (Dhakal et al., 2014; Zhang et al., 2021). The Nt-LCMS analysis also identified one dihydroxylated OH-PCB metabolite in the liver of PCB 52 exposed rats, consistent with earlier in vitro and in vivo studies reporting the formation of 2,2’,5,5’-tetrachlorobiphenyl-3,4-diol. Dihydroxylated PCB metabolites can be formed from the monohydroxylated metabolites in secondary oxidation reactions (Grimm et al., 2015b). Our analysis detected several di-OH-PCB-derived metabolites, including three OH-PCB sulfates and one tetrachlorinated MeO-OH-PCB metabolite. Finally, we tentatively identified four tetrachlorinated MeO-OH-PCB sulfates, metabolites derived from tri-hydroxylated PCBs (Fig. S8). These metabolites of PCB 52 are in agreement with earlier metabolism studies with LC-PCBs in vitro (McLean et al., 1996), cells in culture (Zhang et al., 2020), in rats in vivo (Grimm et al., 2015b; Sundström and Jansson, 1975), and humans (Zhang et al., 2022b). The formation of di- and tri-hydroxylated PCB 52 metabolites and their conjugates is a concern because these hydroxylated metabolites can cause toxic effects through intracellular oxidative stress mechanisms (Song et al., 2008).

Fig. 3.

Several tetrachlorinated PCB metabolites were detected by LC-Orbitrap MS in the liver of a rat exposed to 100 mg/kg BW of PCB 52, including (A) two PCB sulfates ([C₁₂H₅Cl₄O₄S]⁻, m/z 384.86681), (B) one diOH-PCB ([C₁₂H₅Cl₄O₂]⁻, m/z 320.90491), (C) three OH-PCB sulfate ([C₁₂H₅O₅Cl₄S]⁻, m/z 400.86173), and (D) one MeO-OH-PCB ([C₁₃H₇O₃Cl₄]⁻, m/z 334.92056) peaks. The presence of the (E) PCB sulfate at 6.35 min, (F) diOH-PCB at 6.63 min, (G) OH-PCB sulfate at 5.44 min, and (H) MeO-OH-PCB at 7.87 min was confirmed based on the accurate masses and isotopic pattern of isotope ions corresponding to a tetrachlorinated PCB metabolite. The LC-Orbitrap MS analysis was performed in the negative polarity mode. m/z values are reported for the monoisotopic ion. More information, including the MS data of other tetrachlorinated metabolites detected in the same extract, are included in the supporting information.

Along with tetrachlorinated metabolites, we also observed trichlorinated PCB metabolites in the liver of PCB 52 exposed rats, including one OH-PCB sulfate and one MeO-OH-PCB sulfate (Fig. 4; see also Figs. S9 to S12). This observation indicates a metabolism pathway that results in the dehalogenation of the biphenyl moiety. The loss of chlorine from metabolizing PCB 95 was observed in feces extracts of rats (Sundström and Jansson, 1975). In addition, we tentatively identified one trichlorinated MeO-PCB sulfate (Fig. S10) and one trichlorinated MeO-diOH-PCB (Fig. S11). Similar dechlorinated PCB 2 and PCB 11 metabolites were also formed by HepG2 cells, a human hepatocarcinoma cell line. Metabolism studies with PCB 2 and selected OH-PCB 2 metabolites demonstrated that these diOH-PCB-derived metabolites are formed by the displacement of a Cl group adjacent to an OH-group by OH (Zhang et al., 2022a).

Fig. 4.

Select trichlorinated PCB metabolites detected by LC-Orbitrap MS in the liver of PCB 52 exposed rats including (A) one OH-PCB sulfate ([C₁₂H₆C₁₃O₅S]⁻, m/z 366.90070), (B) one MeO-OH-PCB sulfate ([C₁₃H₈Cl₃O₆S]⁻, m/z 396.91127) peaks. The presence of the (C) trichlorinated OH-PCB sulfate at 6.00 min and (D) trichlorinated MeO-OH-PCB sulfate at 5.99 min was confirmed based on the accurate masses and isotopic pattern of isotope ions corresponding to a trichlorinated PCB metabolite. The LC-Orbitrap MS analysis was performed in the negative polarity mode. More information found in the supporting information.

3.4.2. Identification of PCB 52 metabolites in the serum by Nt-LCMS

The Nt-LCMS analysis identified several compounds in the serum of animals exposed to PCB 52, including two OH-PCB 52 sulfates and one MeO-OH-PCB 52 (Fig. 5; see also Figs. S13 to S18). In addition, one PCB 52 sulfate and one MeO-OH-PCB 52 sulfate were tentatively identified by Nt-LCMS. The PCB sulfate formed from 4-OH-PCB 52 has been detected in human serum (Zhang et al., 2022b). Like the dechlorinated PCB metabolites identified in the liver, the serum also contained two trichlorinated MeO-OH-PCB sulfates, providing further evidence that dechlorinated PCB metabolites are formed in rats in vivo (Hutzinger et al., 1974; Sundström and Jansson, 1975). Additional preclinical studies and human biomonitoring studies are needed to confirm the formation of dechlorinated PCB metabolites in humans.

Fig. 5.

Select PCB metabolites were detected by LC-Orbitrap MS in the serum of PCB 52 exposed rats, including (A) two tetrachlorinated OH-PCB sulfate ([C₁₂H₅O₅Cl₄S]⁻, m/z 400.86173), (B) one tetrachlorinated MeO-OH-PCB ([C₁₃H₇O₃Cl₄]⁻, m/z 334.92056), and (C) two trichlorinated MeO-OH-PCB sulfates ([C₁₃H₈C₁₃O₆S]⁻, m/z 396.90991) peaks. The presence of the (D) OH-PCB sulfate at 5.43 min, (E) MeO-OH-PCB at 7.86 min, and (F) trichlorinated MeO-OH-PCB sulfate at 5.06 min was confirmed based on the accurate masses and isotopic pattern of isotope ions corresponding to a tetrachlorinated PCB metabolite for the monoisotopic ion). The accurate masses of two high-abundance isotope ions at (F) 5.06 match the theoretical accurate mass and isotopic pattern of a trichlorinated compound. The LC-Orbitrap MS analysis was performed in the negative polarity mode, as described in the Supporting Information. More information in the supporting information.

3.4.3. Metabolism pathway of PCB 52 in rats

Based on the Nt-LCMS analysis results, we propose the metabolic pathway for PCB 52 in rats shown in Fig. 6. Briefly, PCB 52 is oxidized to mono-hydroxylated OH-PCB metabolites by CYPs. The OH-PCB metabolites are metabolized to the corresponding sulfates by sulfotransferase (SULT) or oxidized by CYPs to di- and tri-hydroxylated PCB metabolites. These metabolites undergo further SULT or catechol-O-methyltransferase (COMT) catalyzed biotransformation to diverse sulfated and methoxylated PCB 52 metabolites. Also, OH-PCB metabolites undergo chlorine displacement to form dechlorinated dihydroxylated PCB metabolites that undergo further biotransformation to sulfate conjugates and methoxylated PCBs. It is also possible that OH-PCB sulfate metabolites are formed via the oxidation of PCB 52 sulfates. Consistent with this metabolism pathway, 4-PCB 11 sulfate is metabolized to further biotransformation products in rats exposed intravenously to 4-PCB 11 sulfate (Grimm et al., 2015a). Several metabolites, including MeO-PCBs and di-OH-PCBs, may also be formed by deconjugation reactions. For example, human hepatic microsomal sulfatases catalyze the deconjugation of PCB sulfates to the corresponding OH-PCBs (Duffel et al., 2021). Overall, the PCB 52 metabolism pathway outlined in Fig. 6 is consistent with published in vitro and in vivo studies (Grimm et al., 2015b). However, as reported previously, we did not observe the formation of glutathione or cysteine adducts of PCB 52 quinones (Lin et al., 2000). These differences may be due to the different routes of exposure (intraperitoneal vs. oral gavage) or the different analytical methods.

Fig. 6.

Proposed PCB 52 metabolic pathway in rats based on the metabolites detected in the targeted GC-MS/MS and the Nt-LCMS analyses, see text for details. CYP, cytochrome P450 enzyme; SULT, sulfotransferase; COMT, catechol-O-methyltransferase.

3.5. Changes in serum metabolome of PCB 52 exposed female rats

We investigated changes in endogenous metabolic pathways in the serum of female Sprague Dawley rats following exposure to PCB 52 to assess if PCB 52 and metabolite levels are potentially associated with changes in specific metabolic pathways. The targeted metabolomic analysis identified only 9 significantly altered cell metabolites out of the 81 metabolites included in the data analysis (Fig. 7), thus making it challenging to study associations between PCB levels and metabolomic outcomes. These results suggest that PCB 52 may affect some metabolic functions in vivo. However, because the metabolomics analyses were performed in samples taken three weeks after PCB 52 exposure, further research is necessary to fully understand the effects of PCB 52 exposure on serum metabolome over time and after repeated exposure.

Fig. 7.

Targeted metabolomic analysis suggested a small number of serum metabolites associated with acute IP exposure to PCB 52 doses ranging from 0 to 100 mg/kg body weight. (a) Volcano plot with p-values from an univariate analysis identified seven and two metabolites negatively or positively associated with PCB 52 exposure, respectively. For bar plots of these significantly associated metabolites, see Fig. 8. Multivariate analysis, including (b) principal component analysis (PCA) and (c) partial least squares-discriminant analysis (PLS-DA), revealed reasonable data variability and potential metabolites that drive the dose-dependent clustering, as indicated by the (d) variable importance in projection (VIP) values in the PLS-DA. For information regarding the metabolomic analysis, see Materials and Methods.

L-cysteine, cytosine, sphingosine, orotic acid, and erythrose levels decreased dose-dependent following PCB 52 exposure. A decrease in L-cysteine levels has been implicated in male reproductive maladaptive responses following PCB exposure in rats (Mega et al., 2022). The reduction in sphingosine suggests alterations in gap-junction intracellular communication via changes in sphingolipid levels. PCB-mediated effects on sphingolipid levels in the liver decrease gap-junction intracellular communication, a phenomenon implicated in tumor regulation (Pierucci et al., 2017). Finally, the PCB mixture Phenoclor DP6 reduced the incorporation of the orotic acid-derived nucleotides thymine and cytosine into the nuclei of liver cells in PCB-exposed rats (Narbonne, 1979). The effects of PCB exposure on erythrose have not been reported previously. However, a recent study reported that PCBs increased levels of an erythrose phosphate, D-erythrose-4-phosphate, in the serum of cats, potentially affecting the pentose pathway (Nomiyama et al., 2022).

Glycine and L-histidine levels showed a dose-dependent increase following PCB 52 exposure (Fig. 8). Effects of PCB 52 and structurally related PCBs on these two metabolites in vivo have not been reported previously. Furthermore, PCB 52 significantly affected thymine and linoleic acid levels in the serum of female rats; however, no dose-response relationship was observed for either metabolite. The observation that PCB 52 exposure affected linoleic acid metabolism is not surprising because PCB congeners with multiple ortho chlorine substituents are known to significantly change the fatty acid composition, particularly in the rat liver (Azaïs-Braesco et al., 1990; Matsusue et al., 1997).

Fig. 8.

Univariate analysis with a linear regression model selected nine dose-dependent serum metabolites in rats exposed via IP injection to PCB 52 with doses ranging from 0 to 100 mg/kg body weight. Data are reported as mean ± standard deviation. Intensities were normalized with the total intensity of each sample.

3.6. Changes in striatal dopamine and metabolites after PCB 52 exposure in female rats

Several individual PCB congeners and complex PCB mixtures influence the dopaminergic system in vivo. Briefly, studies with rat brain synaptosomes demonstrate that ortho-substituted PCBs significantly inhibit dopamine uptake into vesicles (Mariussen et al., 2001; Mariussen and Fonnum, 2001). Studies in cells in culture reported conflicting effects of individual, ortho-chlorinated PCB congeners on dopamine levels, consistent with different, structure-dependent mechanisms by which PCBs affect cellular dopamine metabolism (Enayah et al., 2018; Uwimana et al., 2019). In vivo, dopamine levels were significantly decreased in several brain regions of pig-tailed macaques following oral exposure to Aroclor 1016 or Aroclor 1260 (Seegal et al., 1994; Seegal et al., 1990). Another study reported decreased dopamine transporters (DAT) protein levels and dopamine uptake in mice exposed to Aroclor 1254 and Aroclor 1260 (Caudle et al., 2006). In contrast, a 4-week oral exposure in rats to a PCB mixture containing PCB 52 increased striatal dopamine levels 2 weeks post-exposure (Lee et al., 2012).

Here, we investigated if PCB 52 exposure affects the levels of dopamine and its metabolites, DOPAL and DOPAC. We observed no significant differences in dopamine, DOPAL, or DOPAC levels in the striatum of PCB 52 exposed animals (Fig. 9). No changes in either dopamine or DOPAC levels were reported by a study that exposed rats for 91 days to an inhaled mixture of PCBs (Wang et al., 2022). It is possible that the effects of PCB 52 on dopamine homeostasis in the rat brain were transient and, therefore, could not be detected three weeks after PCB 52 exposure. Moreover, PCBs or their metabolites may differentially affect the expression of dopaminergic genes, such as the dopamine receptor, in dopaminergic brain regions (Lesmana et al., 2014; Liberman et al., 2020), thus masking changes in dopamine homeostasis in our study. Finally, the metabolism of PCB 52 to hydroxylated metabolites may be necessary to alter dopamine levels in the female rat brain.

Fig. 9.

Dopamine and metabolite concentrations from the striatum of PCB 52 exposed female rats. Concentrations are plotted on a log₁₀ scale and reported in pmol per mg of striatal tissue processed. Dopamine (DA), 3,4-Dihydroxyphenylacetic acid (DOPAC), or 3,4-dihydroxyphenylacetaldehyde (DOPAL) concentrations were not significantly different 3 weeks post IP injection of different doses of PCB 52. Group means were compared using two-way ANOVA followed by Tukey’s pairwise comparisons. A p-value < 0.05 was considered statistically significant.

4. Conclusions

Because only limited studies of the disposition and metabolomic effects of PCB 52 are available, this exploratory study characterized the disposition of PCB 52 and its metabolites in target tissues following intraperitoneal exposure of female rats to doses ranging from 1 to 100 mg/kg BW PCB 52. Three weeks after PCB 52 dosing, PCB 52 remained in adipose, brain, liver, and serum. Several PCB 52 metabolite mixtures were detected in the liver and serum of animals from the 100 mg/kg BW dose groups, with some metabolites only being detected in a single animal. Based on these results, female rats can metabolize PCB 52 to complex PCB metabolite mixtures and likely eliminate the metabolites quickly. The study also found that PCB 52 had minimal effects on certain measurements, such as body and organ weights, serum metabolites, and brain dopamine levels. These effects may be due to the fast elimination of PCB 52. Because PCB 52 is a major contaminant in indoor air, additional work will be needed to determine how age, sex, time of exposure, and exposure route, particularly inhalation, affect PCB 52 disposition and toxicity. Further, it will be important to determine the long-term effects of PCB52 exposure and how this may act synergistically with other factors, such as exposure to other toxicants and diet, to cause disease.

Highlights.

• PCB 52 was present in adipose, brain, liver, and serum 3 weeks after PCB exposure

• Liver and serum contained hydroxylated, sulfated, and methylated PCB 52 metabolites

• Metabolomics analysis revealed minor changes in endogenous serum metabolites

• Levels of dopamine and its metabolites in the brain were not affected by PCB 52

Abbreviations:

IP

intraperitoneal

PCB

polychlorinated biphenyl

BW

body weight

Nt-LCMS

Nontarget; liquid chromatography-high resolution mass spectrometric

GC-MS/MS

Gas chromatograph mass spectrometric

UPLC-QTof-MS

Ultra-performance liquid chromatography-quadrupole time-of-flight mass spectrometric

QC

quality control

PCA

principle component analysis

ANOVA

analysis of variance

DA

dopamine

DOPAC

3,4-Dihydroxyphenylacetic acid

DOPAL

3,4-dihydroxyphenylacetaldehyde

PLS-DA

partial least squares-discriminant analysis (PLS-DA)

VIP

variable importance in projection

CYP

cytochrome P450 enzyme

SD

Sprague-Dawley

MOC

methyoxyamine hydrochloride

TMS

N,O-Bis(trimethylsilyl)trifluoroacetamide

SPE

solid phase extraction

SULT

sulfotransferase

COMT

catechol-O-methyltransferase

ND

non-detect