Chronic Aroclor 1260 Exposure Alters the Mouse Liver Proteome, Selenoproteins, and Metals in Steatotic Liver Disease

Abstract

The prevalence of metabolic dysfunction-associated steatotic liver disease (MASLD) continues to increase due in part to the obesity epidemic and to environmental exposures to metabolism disrupting chemicals. A single gavage exposure of male mice to Aroclor 1260 (Ar1260), an environmentally relevant mixture of non-dioxin-like polychlorinated biphenyls (PCBs), resulted in steatohepatitis and altered RNA modifications in selenocysteine tRNA 34 weeks post-exposure. Unbiased approaches identified the liver proteome, selenoproteins, and levels of 25 metals. Ar1260 altered the abundance of 128 proteins. Enrichment analysis of the liver Ar1260 proteome included glutathione metabolism and translation of selenoproteins. Hepatic glutathione peroxidase 4 (GPX4) and Selenoprotein O (SELENOO) were increased and Selenoprotein F (SELENOF), Selenoprotein S (SELENOS), Selenium binding protein 2 (SELENBP2) were decreased with Ar1260 exposure. Increased copper, selenium (Se), and zinc and reduced iron levels were detected. These data demonstrate that Ar1260 exposure alters the (seleno)proteome, Se, and metals in MASLD-associated pathways.

Graphical Abstract

1. Introduction

Polychlorinated biphenyls (PCBs) are metabolism disrupting chemicals (MDC) that persist in the environment and promote metabolic disease including metabolic dysfunction-associated steatotic liver disease (MASLD) (reviewed in (Wahlang et al., 2019)). MASLD may progress from steatosis to steatohepatitis (metabolic dysfunction-associated steatohepatitis (MASH)), in which liver injury and inflammation are detected with or without fibrosis, cirrhosis, and hepatocellular carcinoma (HCC) (Cave et al., 2016). Although PCB manufacture was banned in the U.S. in 1979, exposure to PCBs is considered “ubiquitous” for the U.S. population (Curtis et al., 2021). Human PCB exposure is primarily by dietary intake from contaminated food, e.g., fish and dairy products (reviewed in (Montano et al., 2022)). The biological effects of PCBs are considered to be mediated by receptor-based modes of action. Coplanar, or “dioxin-like” (DL) PCBs activate the aryl hydrocarbon receptor (AHR), while the non-coplanar, or “non-dioxin-like” (NDL) PCBs activate nuclear receptors, e.g., constitutive androstane receptor (CAR) and pregnane X receptor (PXR) (Safe et al., 1985) and inhibit epidermal growth factor receptor (EGFR) signaling (Hardesty et al., 2021). PCB mixtures were sold under the trade name Aroclor followed by four digits in which the first two indicate that the mixture is composed of chlorinated biphenyls and the last two digits indicate the percent chlorine by weight (Master et al., 2002). Aroclor1260 (Ar1260) is a mixture of highly chlorinated, NDL PCBs. Due to the high thermodynamic stability of these congeners, Ar1260 reflects the PCB bioaccumulation patterns by mass in human tissue (Wahlang et al., 2014a).

We recently reported that male C57Bl/6J mice receiving a single oral dose of 20 mg/kg Ar1260 developed, at the 34^th week (wks.) post-Ar1260, liver steatosis and fibrosis with increased serum AST and liver Col1a1 (collagen type 1 alpha 1 chain) transcript abundance (Piell et al., 2023). The human environmental relevance of the 20 mg/kg Ar1260 dose was confirmed by exposure assessment in mouse liver (Hardesty et al., 2019; Wahlang et al., 2014b) to model the upper end of observed human exposure in our translational human cohort, the Anniston Community Health Survey (ACHS) (Cave et al., 2022; Clair et al., 2018), based on PCB partitioning ratios observed in NTP studies (2006). We reported 12 RNA epitranscriptomic changes in the livers of mice at the 34^th wks. post-Ar1260 exposure compared to control mice Piell et al., 2023). Using a comprehensive bioinformatics approach, we identified cell pathway networks related to MAFLD that were associated with the epitranscriptome changes and observed a reduction in the levels of the hub protein NRF2 in the Ar1260-exposed livers, which we suggested would impair hepatic detoxification and anti-oxidative stress responses.

Among the tRNA modifications we reported, i6A (N6-isopentenyl-adenosine) was increased and mcm5U (5-methoxy-carbonylmethyluridine) was decreased (Piell et al., 2023). mcm5U and i6A are specifically located in the anti-codon loop of selenocysteine (Sec, the 21^st amino acid) transfer RNA (Sec-tRNA, also called tRNA^[Ser]Sec) (Moustafa Mohamed et al., 2001). The essential trace element selenium (Se) is ingested in food as various organic Se species (Marschall et al., 2016). The liver is the central organ for Se regulation (Burk and Hill, 2015). Se is co-translationally incorporated into Sec which is part of the active site in selenoprotein antioxidant enzymes (Lai et al., 2011). There are 24- and 25- Se-dependent proteins (selenoproteins) in rodents and humans, respectively (Zhang et al., 2020b). Selenoprotein synthesis uniquely depend on Sec-tRNA. Although there is only one gene encoding Sec-tRNA (n-TUtca2 in mouse), there are two Sec-tRNAs isoforms with either mcm5U or mcm5Um at position 34 (Sengupta et al., 2008), and then invariably i6A at position 37 in the anticodon loop, Y at 55, and m1A at 58 (Carlson et al., 2018). i6A is required for Sec insertion into selenoproteins at UGA, a process requiring a special elongation factor eEFsec (Eukaryotic Elongation Factor, Selenocysteine-TRNA Specific, Eefsec in mouse) and SBP2 (Secisbp2, SECIS binding protein 2 in mouse) (Moustafa Mohamed et al., 2001; Peng et al., 2021).

PCB exposures do not affect hepatic total glutathione concentrations in rodents (Twaroski et al., 2001). However, hepatic Se levels in rats were reduced two wks. after a single i.p. injection of 100 μmol/kg of the DL PCB77 (Twaroski et al., 2001) or 5 μmol/kg of the DL PCB126 (Klaren et al., 2015). Dietary Se supplementation prevented a PCB126 exposure-dependent decrease in the activities of selenoproteins glutathione peroxidase (GPX) and thioredoxin (TRX) reductase in rat liver (Lai et al., 2011). High Se (0.15 or 0.60 ppm) for seven wks directly after weaning increased GPX and TRX activity and plasma triglycerides in mice (Speckmann et al., 2017). To our knowledge, no one has evaluated the impact of Ar1260 exposures on liver Se levels or selenoprotein expression. To address this gap, we used an untargeted proteomics approach to identify protein changes in the livers of male mice treated with the same experimental procedure as previous study (9), i.e., a single gavage of Ar1260 at 20 mg/kg and examination at the 34^th wks. post-exposure. Pathway enrichment analysis of the proteome revealed glutathione metabolism as the top pathway in the livers of the Ar1260-exposed mice. In addition, we evaluated the levels of 25 metals and identified significant increases in Se, zinc (Zn), and copper (Cu) and reduced iron (Fe) with Ar1260 exposure that are associated with pathways and transcription factors regulating the hepatic pathways altered by Ar1260 exposure.

2. Materials and Methods

2.1. Materials

Aroclor 1260 (Ar1260) was purchased from AccuStandard (New Haven, CT, USA). Ar1260 was a commercial heavily chlorinated PCB mixture containing ~ 99% non-coplanar, di-ortho substituted PCBs that have 5, 6, 7, or 8 chlorines (see Supplementary Table 1 in (Wahlang et al., 2014a)).

2.2. Animal Studies

The experimental design is modeled in Supplementary Figure 1 as reported in (Piell et al., 2023). The animal protocol was ratified by the University of Louisville Institutional Animal Care and Use Committee (Piell et al., 2023). Adult male C57BL/6 mice (8 wks. old) were purchased from Jackson Laboratory and randomized into two groups (n = 10 control and n = 30 Ar1260 exposure) (Piell et al., 2023). All mice were fed ad libitum a normal, control diet with 20%, 69.8%, and 10.2% of total calories from protein, carbohydrate, and fat, respectively (TekLad 06416, Envigo, Indianapolis, IN, USA) throughout the study (Piell et al., 2023). At 10 wks. of age, the mice in each group were given either corn oil or Aroclor1260 (20 mg/kg) via a one-time oral gavage as reported in our previous study (Piell et al., 2023). This concentration was based on previous studies showing that Ar1260 acts as a ‘second hit’ in high-fat diet (HFD)-fed mice to induce steatohepatitis in a chronic (12 wks.) mouse diet + PCB exposure model of human MASLD (Jin et al., 2020; Shi et al., 2019). Although Ar1260 contains the PCB congeners that are reflective of human adipose bioaccumulation patterns (Wahlang et al., 2014a), it did not activate mouse AHR at environmentally relevant doses (Wahlang et al., 2014b). The dose of Ar1260 in this study is the same as in our previous studies (Jin et al., 2020; Piell et al., 2023; Shi et al., 2019; Wahlang et al., 2014b). The total experimental time was 36 wks. and post-exposure time of single gavage with Ar1260 was 34 wks (Supplementary Figure 1). At the end of 36^th wks., the mice were fasted for ~ 6 h prior to euthanasia and liver samples were harvested as described (Piell et al., 2023; Shi et al., 2019).

2.3. RNA isolation

Mouse liver was preserved in RNAlater Stabilization Solution (ThermoFisher, Waltham, MA) immediately following excision. Total RNA was isolated as previously described (Klinge et al., 2021; Piell et al., 2023). In summary, liver tissue was washed in PBS, and total RNA was isolated using the Qiagen (Germantown, MD) RNeasy Mini Kit according to manufacturer’s protocol. Total RNA concentration and quality were measured using a NanoDrop spectrophotometer and a Qubit Fluorometer (ThermoFisher, Rockford, IL, USA).

2.4. LC-MS/MS analysis of RNA modifications and data processing

All liver RNA samples (n = 10 control + 30 Ar1260-exposed) were analyzed in random order on a Thermo Orbitrap Fusion Lumos Tribrid mass spectrometer coupled with a Thermo Vanquish™ HPLC system (Thermo Fisher) as previously described (Piell et al., 2023). Xcalibur software (v4.5, Thermo Fisher) was used to identify and quantify the LC-MS data. Quan Browser, a build-in component in Xcalibur was used to quantify the nucleosides using the peak area of daughter ion. Nucleoside identification was achieved by matching the experimental value of retention time, parent ion m/z, and the m/z of the most abundant daughter ion to the corresponding data of the standard (Piell et al., 2023). The threshold for the retention time difference and m/z variation window were set as ≤ 3 s and ≤ 5 ppm, respectively.

2.5. Sample Preparation for Proteomics Analysis

Proteins were extracted from liver tissues in RIPA buffer supplemented with protease and phosphatase inhibitors using a bead homogenizer and protein amounts were quantitated by BioRad DC protein assay (Bio-Rad Laboratories, Inc, Hercules, CA.) (Piell et al., 2023). Protein lysates (200 μg) were diluted to 4% SDS plus 1X HALT™ protease/phosphatase inhibitors. Samples (100 μg) were trypsinized for proteomic analyses with an S-Trap™ mini column (Protifi, Fairport, NY, USA) according to the manufacturer’s recommendations. After drying the S-Trap eluate, each sample was dissolved in LC-MS grade water and a NanoDrop 2000 (ThermoFisher Scientific, Waltham, MA, USA) was used to estimate peptide concentration by absorbance at 205 nm (extinction coefficient 31(mL·cm)/mg).

2.6. Liquid Chromatography

The samples were resuspended in 2 % acetonitrile/0.1 % formic acid (0.1 μg/μL) and 250 ng injected onto a 300 μm × 5 mm, 5 μm PepMap100 C18 trap cartridge heated at 30°C (ThermoFisher Scientific) before elution onto a 75 μm × 15 cm, 3 μm, 100Å PepMapTM RSLC C18 EASY-spray separating column heated at 40°C. Peptides were separated using a Dionex Ultimate3000 RSLCnano system (ThermoFisher) at 200 nL/min with an 90 min 5 % to 35 % acetonitrile (0.1 % v/v formic acid) gradient. An EASY-spray source (ThermoFisher) was used to position the emitter of the separating column 1mm from the ion transfer capillary of the mass spectrometer. The ion transfer capillary temperature and spray voltage of the mass spectrometer were set at 320°C and 1.8 kV, respectively.

2.7. Mass Spectrometry

An Orbitrap Exploris480 mass spectrometer (ThermoFisher) was used to collect data from the LC eluate. A Full MS – ddMS2 method with a 3sec cycle time was created in Xcalibur v4.5.445.18 (ThermoFisher) operating in positive polarity. Scan event one obtained an MS1 scan (60,000 resolution, Normalized AGC target of 100%, scan range 350–1400 m/z). Scan event two obtained dd-MS2 scans (7,500 resolution, Normalized AGC target of 50%) on ions with charge states from 2–6 and a minimum intensity of 8,000 until the cycle time was complete.

2.9. Proteome Data Analysis

Proteome Discoverer v2.5.0.400 (ThermoFisher) was used to analyze the data collected by the mass spectrometer. In the processing step, the database used in SequestHT was the 1/24/2023 version of the UniprotKB canonical Mus musculus sequences (Proteome ID UP000000589). Trypsin (KR|P) digestion with up to two missed cleavages was assumed with the dynamic modifications Oxidation (M), Acetyl (Protein N-term), Met-los (Protein N-term), and Met-loss+Acetyl (Protein N-term); and the static modification Carbamidomethyl (C). Precursor and fragment mass tolerances were 10 ppm and 0.02 Da, respectively.

In the consensus step, proteins were quantified from the summed abundances of all high confidence unique and razor precursor ion intensities. Samples were normalized to total peptide amount and scaled to 100 on all average. Proteins were grouped by the strict parsimony principle. Peptides and proteins were accepted at 1% FDR for high confidence or 5% for medium confidence based on the q-value. A proteins text file was exported from the consensus workflow result of Proteome Discoverer for curation in Microsoft Excel (Supplementary Methods). Primary proteomic data will be shared through the Mass Spectrometry Interactive Virtual Environment (MassIVE, https://massive.ucsd.edu/ProteoSAFe/static/massive.jsp) hosted by the Center for Computational Mass Spectrometry at the University of California, San Diego. Data will be shared from MassIVE into the ProteomeXchange Consortium data sharing site (http://www.proteomexchange.org/). The data will be made public upon manuscript publication.

Hepatic proteins identified in minimally three of the five individual livers examined and that had significance abundance were imported into MetaCore software (Clarivate Analytics, Philadelphia, PA) for the following analyses: enrichment by pathway maps, enrichment by process networks, gene ontology (GO) process, enrichment by protein function (EPF), and interaction by protein function (IPF), and transcription factor associated function.

2.8. Western blots

Liver tissues were homogenized, sonicated, and supernatants collected after 21,000 × g centrifugation. Protein concentrations determined (BioRad DC protein assay, Bio-Rad Laboratories) and samples stored as previously described (Petri et al., 2023b). 50 μg of protein were separated on 10–15% SDS-PAGE gels and transferred to PVDF membranes (Bio-Rad Laboratories). Membranes were blocked and incubated overnight at 4°C with primary antibodies: GPX4 (#67763–1-Ig, Protein Tech, Rosemont, IL, USA); Selenoprotein S (SELENOS, SELS, # 15591–1-AP, Protein Tech); and α-tubulin (Fisher Scientific # MS581P1). Membranes were washed with TBS-Tween followed by incubation with anti-mouse (#7076S) or anti-rabbit (#7074S) (Cell Signaling Technology, Danvers, MA, USA) secondary antibodies. Membranes were incubated with Clarity Western ECL (Bio-Rad) and imaged on a Bio-Rad ChemiDoc^™ XRS+ System with Image Lab^™ Software (Bio-Rad). Blots were stained with Ponceau S and Amido Black for additional quantification.

2.9. Metals analysis

50–100 mg of each mouse liver sample were homogenized in RIPA buffer supplemented with protease and phosphatase inhibitors and centrifuged at 21,000 × g for 10 min. at 4°C. Protein concentrations in the supernatant were determined using the Bio-Rad DC protein-assay (Bio-Rad Laboratories Inc., Hercules, CA). 5 mg protein was brought to a final volume of 200 μl with nuclease-free H₂O. Each liver protein sample was digested in 800 μl 70% nitric acid (trace metal grade, Fisher Scientific Cat# A509-P500) and 200 μl H₂O₂ (Sigma Cat# 95321) at 85°C for 4 hours. Digested samples were transferred to a chemical hood, allowed to cool to room temperature, and diluted to a final 5% nitric acid with deionized water (Milli Q system). Each sample was filtered by 45 μm filter.

To measure the metal content in liver samples, Agilent 7800 Inductively Coupled Plasma Quadrupole Mass Spectrometer (ICP-MS) (Agilent Technologies, Japan) was used. The Agilent 7800 ICP-MS was optimized by performance checks with 1 ppb tuning solution and the assay program was auto tuned by 10 ppb tuning solution (Agilent Cat#5188–6564). The autosampler SPS 4 was used for sample introduction. A 25 metals calibration standard was purchased from Inorganic Ventures (Cat# IV-STOCK-50) and serial metal standard dilutions were made with same acid matrix of samples. Internal standards were purchased from Agilent (Cat# 5188–6525). During sample injection, internal standards including bismuth, indium, lithium, scandium, terbium and yttrium were mixed with each sample for drift correction and accuracy improvement. The assay program was run by Agilent MassHunter software with He mode and each sample was read three times for final mean value. Data were analyzed quantile normalization to normalize the intensity data after logarithm transformation with base 2. Then, the LIMMA/Moderated t-test was used to compare the vehicle control to Ar1260 exposure. The p value is the adjustment for the raw p-value using Benjamini-Hochberg method for multiple testing of chemicals (Benjamini and Hochberg, 1995).

3. Results

3.1. Effects of Ar1260 exposure on the hepatic proteome

We recently reported that the top pathway from MetaCore enrichment by pathway maps in our analysis of epitranscriptome and mRNA changes in the livers from control diet-fed male mice 34 wks. after a single oral gavage of a mixture of NDL PCBs Ar1260 (20mg/kg) was “Response to hypoxia and oxidative stress” (Piell et al., 2023). Integrated network analysis of epitranscriptomic modifications identified a NRF2 (gene Nfe2l2) pathway and we reported that NRF2 protein was reduced in the mouse livers after chronic Ar1260 exposure (Piell et al., 2023). We also identified changes in the abundance of RNA modifications mcm5U and i6A (Piell et al., 2023), located specifically in the anticodon loop of Sec-tRNA which is required for the synthesis of selenoprotein synthesis (Moustafa Mohamed et al., 2001).

To evaluate the impact of these Ar1260 exposure-mediated alterations in transcript abundance and RNA chemical modifications, we used an unbiased approach to identify proteins altered after the long-term exposure Ar1260 using mass spectrometry. Of the 2,156 proteins detected in the liver samples (Supplementary Table 1), the abundance of 128 proteins was significantly different between control and Ar1260-exposed mice (Figure 1A and 1B, Supplementary Table 2). MetaCore enrichment analysis was performed to identify pathways and networks specific to these liver protein changes with chronic Ar1260 exposure (Supplementary Tables 3 and 4). The top significant pathway was “Glutathione metabolism” and the fifth significant pathway was “Translation_(L)-selenoaminoacids incorporation in proteins during translation” (Supplementary Table 3). We identified seven selenoproteins and three Sec synthesis proteins in the liver proteomes of both control and Ar1260-exposed mice (Table 1, Supplementary Table 5) with three significantly different proteins with Ar1260 exposure (Table 1, Figure 1C). Glutathione peroxidase 4 (GPX4) and Selenoprotein O (SELENOO) were increased and Selenoprotein F (SELENOF) was decreased with Ar1260 exposure (Figure 1C). A western blot confirmed higher GPX4 protein levels in the Ar1260- exposed liver samples (Figure 2). We also observed a reduction in Selenium binding protein 2 (SELENBP2) with Ar1260 exposure (Figure 1B).

Figure 1: Long term Ar1260 exposure alters the hepatic proteome.

A) Volcano plot of all 2,715 proteins in the mouse liver plots showing significance (y-axis) versus log2 fold change in protein abundance between Ar1260-exposed vs. control livers. B) Significant alterations in 128 hepatic proteins after Ar1260 exposure were graphed in a volcano plot. C) Volcano plot of the 7 seleonoproteins and 4 Sec synthesis machinery proteins identified in the proteome analysis. The dotted line separates the significantly altered proteins (red dots) with log2FC > 2 or < 2. Black circles are unaltered proteins.

Table 1: Seven Selenoproteins and four Sec synthesis proteins were identified in the livers of the normal fed control and Ar1260 exposed mice.

These proteins were detected in at least three of the five individual livers in each group examined (Supplementary Table 5).

Accession	Description	# Unique Peptides	Gene Symbol	Abundance Ratio Adj. P-Value * P < 0.05	Log2 fold-change
P97364	Selenide, water dikinase 2	15	Sephs2	0.864	−0.242
Q9ERR7	Selenoprotein F	2	Selenof	0.014 *	−0.894
A4FUU9	Selenoprotein O (Fragment)	5	Selenoo	0.048 *	0.196
O70325	Phospholipid hydroperoxide glutathione peroxidase	8	Gpx4	0.033 *	0.872
P11352	Glutathione peroxidase 1	21	Gpx1	0.732	0.217
A0A0M3HEQ0	thioredoxin-disulfide reductase	5	Txnrd2	0.804	−0.676
Q9JMH6	Thioredoxin reductase 1, cytoplasmic	9	Txnrd1	0.888	−0.125
Q6P6M7	O-phosphoseryl-tRNA(Sec) selenium transferase	2	Sepsecs	0.915	0.135
Q3UGH6	Tr-type G domain-containing protein	6	Eefsec	0.920	0.055
Q8BH69	Selenide, water dikinase 1	2	Sephs1	0.925	0.103
A0A0R4J069	Selenocysteine lyase	7	Scly	0.969	−0.302

Figure 2: GPX4 was increased in Ar1260-exposed mouse livers.

Liver homogenates were prepared from the liver samples from control or Ar1260-exposed mice. GPX4 was quantified relative to Ponceau S staining and normalized to the values in the five control liver samples. A two-tailed t-test was performed ** p < 0.01.

As seen previously in our integrated analysis of transcript (gene, mRNA-seq) and global epitranscriptomic changes with chronic Ar1260 exposure (Piell et al., 2023), the top significant network identified in this proteome analysis was “Response to hypoxia and oxidative stress” (Supplementary Table 4). The top two GO (gene ontology) processes were “small molecule metabolic process” and “cellular detoxification” (Supplementary Table 6). Enrichment by Protein Function (EFP) analysis was performed and revealed that 24 of the 128 Ar1260-regulated proteins were enzymes (Supplementary Table 7).

Interaction by protein function (IFP) analysis (Vazquez et al., 2003) was performed in MetaCore for the 128 proteins regulated by Ar1260 exposure. This analysis identified 68 transcription factors, 19 enzymes, 11 kinases, 7 ligands, and 5 receptors, and 42 ‘other’ proteins overconnected with Ar1260 exposure (Supplementary Table 8). The over-connected interactions by z-score are shown in Figure 3. Thyroid hormone receptor beta (THRβ) showed the highest overconnection among the transcription factors. Reduced local hepatic thyroid hormone activity and THRβ expression have been reported in human MASH (Krause et al., 2018). Resmetirom, an oral, once-daily, liver-targeted THR-β selective agonist was reported to be safe and well-tolerated in a phase 3 clinical trial demonstrating reduced LDL-C, apoB, hepatic fat, and liver stiffness in adults with presumed MASH (Harrison et al., 2023) and was recently (3/14/2024) approved by the FDA for treatment of NASH (MASH) patients who have progressed to fibrosis. Three transcription factors: Glucocorticoid receptor (GR), SRY-Box Transcription Factor 17 (SOX17), and ETS Proto-Oncogene 1 (ETS1) (Figure 3 and Supplementary Table 8) and one enzyme, protein arginine methyltransferase 1 (PRMT1), were in common with our previous HFD-fed Ar1260 liver proteome analysis (Jin et al., 2020). None of the other enzymes, ligands, kinases, or proteases in the current analysis of theAr1260-exposed livers were identified in the IFP analysis of the HFD-fed, Ar1260-exposed mouse liver proteome (Jin et al., 2020). The low overlap in IFP suggests that diet and time after initial oral gavage of Ar1260 impacts hepatic proteome response to Ar1260 exposure.

Figure 3: Effect of Ar1260 on liver protein function.

Interaction by protein function (IFP) analysis was performed in MetaCore for the 128 liver proteins differentially expressed in Ar1260-exposed mice with the top proteins shown with the corresponding z-scores shown in the heatmap.

Methionine adenosyltransferase 1A (MAT1A) had the top z-score among enzymes. MAT1A is exclusively expressed in the liver and is the terminal enzyme in the synthesis of S-adenosylmethionine (SAM), the methyl donor for DNA, RNA, and proteins as well as glutathione synthesis, that is reduced in chronic liver disease (Mato et al., 2013). Immunoglobulin heavy constant gamma 3 (IGHG3) was the top z-score in ligands (Figure 3) and is produced by B lymphocytes which are recruited to the liver, activated, and release pro-inflammatory IL-6 and TNFα, that promote inflammation and fibrogenesis in MAFLD progression to MASH (reviewed in (Barrow et al., 2021)). IGHG3 (Immunoglobulin heavy constant gamma 3) transcript was identified in differentiated plasma cells in human liver by single cell RNA sequencing (MacParland et al., 2018). IGHG3 was more highly expressed in extracellular vesicles isolated from blood serum of patients with alcoholic hepatitis (AH) compared to those with MASLD (Nguyen et al., 2021). Dual‐Specificity Tyrosine Phosphorylation–Regulated Kinase 3 (DYRK3) was the top z-score in kinases (Figure 3). DYRK3 was reduced in HCC and acts as a tumor suppressor by inhibiting the de novo purine synthesis (Ma et al., 2019)

Using a cut-off of detection of a protein in minimally three of the five individual liver samples/exposure group, nine proteins were uniquely detected in Ar1260-exposed livers and 15 proteins were uniquely detected in the control group livers (Tables 1 and 2). For the nine hepatic proteins uniquely detected in the Ar1260-exposed mice, the top pathway map was “Androstenedione and testosterone biosynthesis and metabolism” for UGT1A9 (UDP Glucuronosyltransferase Family 1 Member A9, Supplementary Table 9) in the data and the top process network was “Response to hypoxia and oxidative stress” for GSTM4 (Glutathione S-Transferase Mu 4, Supplementary Table 10). UGT1A9 protein was reduced in human MASH (Hardwick et al., 2013). GSTM4 transcript was reduced in the livers of patients with steatosis alone or steatosis + inflammation versus obese controls (Younossi et al., 2005). For the fifteen liver proteins detected only in the control diet-fed, vehicle control mice, the top pathway map was “Development_Epigenetic and transcriptional regulation of oligodendrocyte precursor cell differentiation and myelination” for QKI (Quaking Homolog, KH Domain RNA Binding, Supplementary Table 11) and the top process network was “Protein folding_ER and cytoplasm” for SCO2 (Synthesis Of Cytochrome C Oxidase 2, Supplementary Table 12). These data suggest that Ar1260 exposure in mice altered the liver proteome in pathways related to human MASH.

Table 2: Nine proteins uniquely identified in livers of mice 34 wks. after a single oral gavage of Ar1260.

These proteins were detected in at least three of the five individual Ar1260-exposed mouse livers examined and in none of the control mouse livers.

Accession	Description and role(s) in liver and liver diseases	# Unique Peptides	Gene Symbol
P62838	Ubiquitin-conjugating enzyme E2 D2 – ubiquitin-mediated proteolysis; UBE2D2 was increased in human NASH liver samples (Dahlhoff et al., 2014).	1	Ube2d2
E9QA46	Peroxisomal assembly protein PEX3-essential for peroxisome membrane assembly and mutations, frameshifts, or deletions cause severe metabolic disease (Fujiki et al., 2012). PEX3 is phosphorylated by CaMMK2 (calcium/calmodulin-dependent protein kinase kinase 2) resulting in larger lipid droplets accumulating in steatosis (Stork et al., 2022)	1	Pex3
Q8JZL3	Thiamine-triphosphatase Thtpa transcript abundance was increased in the livers of HFD-fed male C7Bl/6J mice (Soltis et al., 2017)	1	Thtpa
Q3URM1	Caseinolytic peptidase B protein homolog – acts as tissue-specific mammalian mitochondrial chaperone that may play a role in mitochondrial protein homeostasis (Santagata et al., 1999)	1	Clpb
Q8R5I6	Glutathione S-transferase Mu 4 – an oxidative stress-related gene Induced by 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) and Aroclor1254 co-exposure in ApoE KO mice (Shan et al., 2015). Liver transcript levels were up-regulated by the herbicides Glyphosate and/or 2,4-D in male C57Bl/6J mice (Romualdo et al., 2023).	2	Gstm4
A0A0R4J1R7	4a-hydroxytetrahydrobiopterin dehydratase PCBD2 interacts directly with hepatocyte nuclear factor 1β (HNF1β) shuttling between the cytoplasm and the nucleus, acting as a coactivator to regulate the ability of HNF1β to regulate transcription (Tholen et al., 2021).	2	Pcbd2
Q62452	UDP-glucuronosyltransferase 1A9 Induced by TCDD and PCB126 in male C57Bl/6 mouse liver (Buckley and Klaassen, 2009).	3	Ugt1a9
Q9EQI8	39S ribosomal protein L46, mitochondrial Is considered a ‘house-keeping gene’ for mouse liver (Mazin et al., 2022; Yarushkin et al., 2018).	3	Mrpl46
Q8BTZ7	Mannose-1-phosphate guanyltransferase beta is a key enzyme in the glycosylation pathway, which catalyzes the synthesis of GDP-mannose from mannose-1-phosphate and guanosine triphosphate (Franzka et al., 2022). More than 50 mutations in the GMPPB gene have been identified in patients with congenital disorders of glycosylation affecting multiple tissues especially the nervous system, muscles, and intestines (Liu et al., 2021). Increased in mouse liver after 90 consecutive days of intragastric administration with titanium dioxide nanoparticles (TiO2 NPs) (Liu et al., 2021).	5	Gmppb

None of the changes in the 128 significantly altered proteins in the Ar1260 exposed versus control liver proteome were observed at the transcript level in the mRNA seq transcriptome (Piell et al., 2023) from the same liver samples. However, since we observed a decrease in mouse Selenos (Selenoprotein S) transcript abundance in livers from Ar1260-exposed mice, although no Selenoprotein S (SELS) in the mouse proteome, we performed western blots to examine Selenoprotein S. Western blot confirmed reduced SELENOS liver protein with Ar1260 exposure (Figure 4). None of the protein changes in the livers of the control (normal) diet-fed, Ar1260 exposed mice identified here were detected in proteome data from the livers of HFD-fed, Ar1260 (12 wks., same dose)-exposed mice (Jin et al., 2020). These data suggest that diet before and after Ar1260 exposure regulates the liver proteome.

Figure 4: Selenoprotein S (SELENOS) was decreased in Ar1260-exposed mouse livers.

Liver homogenates were prepared from the liver samples from control or Ar1260-exposed mice. SELS was quantified relative to Amido Black staining and normalized to the values in the five control liver samples. A two-tailed Mann-Whitney was performed. * p = 0.0159

3.2. Effects of Ar1260 exposure on liver metal levels

GO analysis of the Ar1260-altered proteome using the web-based g:Profiler tool (https://biit.cs.ut.ee/gprofiler/) (Raudvere et al., 2019) identified “catalytic function” and “selenium binding” as the top GO molecular function process (Supplementary Figure 2). Ten metals are considered “essential” for liver homeostasis including Fe, Cu, and Zn functioning in redox reactions enzymes, metabolism, and in transcription (Jomova et al., 2022; Kardos et al., 2018). To our knowledge, no one has examined the impact of Ar1260 exposure on hepatic metals. The levels of three metals: Cu, Fe, and Zn and the microelement Se (Table 4) showed significant differences in liver samples between mice with and without Ar1260-exposure with Fe, reduced and Se, Cu, and Zn increased.

Table 4: Metal analysis in liver samples by ICP-MS.

Five liver samples from normal diet-fed, vehicle control mice and ten liver samples from Ar1260-exposed mice were analyzed. Data were analyzed quantile normalization to normalize the intensity data after logarithm transformation with base 2. Then, the LIMMA/Moderated t-test was used to compare the vehicle control to Ar1260 exposure. The p value is the adjustment for the raw p-value using BH method for multiple testing of chemicals (Benjamini and Hochberg, 1995). Positive and negative values mean increased and decreased levels, respectively.

Metal	Estimate logFC	adj. p-value
Cu	0.0152	0.0000
Fe	−0.1238	0.0001
Se	0.2170	0.0000
Zn	0.0456	0.0000

4. Discussion

Although PCBs are ‘legacy contaminants’, they persist in the environment and are consumed in foods including fish and dairy products (reviewed in (Montano et al., 2022)). Human PCB exposures interact with diet and are associated with metabolic dysfunction and MASLD (Cave et al., 2022; Clair et al., 2018; Pavuk et al., 2023; Pavuk et al., 2019; Petriello et al., 2022). However, the precise mechanisms by which diet and PCB exposures result in steatohepatitis remain unclear. We recently reported specific diet and PCB exposure-induced changes in the global transcriptome of mouse liver (Klinge et al., 2021; Piell et al., 2023). For mice on a normal diet and exposed to Ar1260, we found that the abundance of specific RNA modifications, Am and m6A, were associated with reduced NRF2 (Nfe2l2) and NFATC4 (Nfatc4) proteins which would be expected to impair anti-oxidant stress responses and increase fatty acid accumulation in the liver (Piell et al., 2023). In addition, we reported that i6A was increased and mcm5U was decreased (Petri et al., 2023c), because these modifications are specifically located in the anti-codon loop of Sec-tRNA (Moustafa Mohamed et al. 2001), we hypothesized that liver selenoprotein abundance may be altered in response to Ar1260 exposure. Here we used an unbiased approach to evaluate the liver proteomic profile of these same mice. We report that the abundance of 128 proteins, including seven selenoproteins and four Sec synthesis proteins, was altered by Ar1260 exposure.The pathways identified from the altered proteome analysis include “Glutathione Metabolism” and “Translation (L)-selenoaminoacids incorporation in proteins during translation”. Based on these terms and GO terms “catalytic function and “selenium binding, we used metallomics analysis and observed increases in Se, Cu, Zn and a decrease in Fe in the livers of the Ar1260-exposed mice.

The decrease in hepatic mcm5U abundance in RNA isolated from the Ar1260-exposed mice is in agreement with a study in mice showing a reciprocal relationship between i6A and mcm5U in tRNA isopentenyltransferase 1 (Trit1)-hepatocyte specific knockout mice (Fradejas-Villar et al., 2021). However, Ar1260 exposure did not affect hepatic Trit1 transcript or protein abundance. TRIT1 is the writer for the i6A mark; however, i6A abundance was not rate-limiting for selenoprotein expression, although translation of Sephs2 (Selenophosphate Synthetase 2), Selenop (Selenoprotein P), and Txnrd1 (Thioredoxin reductase 1) were reduced in the livers of the Trit1-hepatocyte specific knockout mice (Fradejas-Villar et al., 2021). This study noted that a deficiency of selenoproteins induced NRF2 (Fradejas-Villar et al., 2021) and earlier studies showed that knockout of nuclear-encoded tRNA selenocysteine 2 (gene n-TUtca2 in mouse, protein TRSP) increased liver phase II detoxifying enzyme gene expression in mice (Sengupta et al., 2008). Although NRF2 (mouse gene Nfe2l2) was not detected in this proteome analysis, we recently reported that Ar1260 exposure reduced NRF2 liver protein levels in mice exposed to Ar1260 (Piell et al., 2023) and mice exposed to HFD (Hardesty et al., 2019).

The liver is the central organ for Se regulation and the synthesis of selenoproteins requires unique RNA elements and proteins. We found that Ar1260 exposure increased liver Se in mice. The Sec insertion sequence (SECIS) element in the 3′-untranslated region (3′-UTR) is required for recruitment of Sec-tRNA (N-TUtca2), in a complex with Eefsec (eukaryotic elongation factor, selenocysteine-tRNA-specific, EFSEC or SELB) and SECIS binding protein 2 (Secisbps, SBP2) to insert Sec in response to a UGA codon (Burk and Hill, 2015; Labunskyy et al., 2014). There was no change in n-TUtca2, Eefsec, or Secisbp2 transcript abundance with Ar1260 exposure (Piell et al., 2023). An increase in Secisbp2l (SECIS binding protein 2-like, protein SBP2L) transcript in Ar1260-exposed mouse liver (Piell et al., 2023). SBP2L binds SECIS elements in Sec-protein mRNAs (The UniProt, 2021). SBP2L was not detected in our mouse proteome analysis.

Of the 24 mouse selenoprotein genes, only glutathione peroxidase (Gpx3) transcript abundance was increased in Ar1260-exposed livers and Selenoprotein S (Selenos, protein SELS) transcript was decreased in Ar1260-exposed livers (Piell et al., 2023). In the present study, GPX4 and SELENOO (SELO) were increased and SELENOF was decreased. The increases in GPX4 and SELENOO were associated with the “Glutathione Metabolism” and “Translation_(L)-selenoaminoacids incorporation in proteins during translation” enrichment pathways identified. Increased GPX4 protein was confirmed by western blot. GPX4 protects cells against lipid peroxidation and inhibits ferroptosis (Wu et al., 2021; Yang et al., 2014).

Here, reduced SELENOF protein was observed in the Ar1260-exposed livers, despite not identifying Selenof in the mouse transcriptome (Piell et al., 2023). Selenof KO mice (C57Bl/6) showed enhanced HFD (45% fat)-induced hepatic steatosis and reduced SELENBP2 (Zheng et al., 2020), as seen in the current study. Previously, we reported a decrease in mouse Selenos hepatic transcript abundance in Ar1260-exposed mice (Piell et al., 2023) and here we demonstrated a reduction in SELENOS (SELS) protein in the same livers. In contrast, Selenof KO mice showed increased SELENOS (SELS) protein in liver (Li et al., 2022). We note that MASH livers have lower SELENOS transcript expression compared to healthy controls (Day et al., 2021). Thus, the potential interplay among selenoproteins in liver in response to Ar1260 exposure requires further study.

We detected a decrease in SELENBP2 protein, previously known as acetaminophen-binding protein (AP56), a cytosolic xenobiotic receptor (Giometti et al., 2000; Mattow et al., 2006). We did not detect a difference in Selenbp2 transcript abundance in the Ar1260 exposed liver relative to controls (Piell et al., 2023), suggesting that Ar1260 regulation of SELENBP2 protein may be post-transcriptional. Since miRNAs post-transcriptionally regulate protein abundance, we examined which miRs regulate SELENBP2 with one result: miR-24–3p repressed SELENBP2 translation in mouse liver (Zhang et al., 2020a). However, miR-24–3p levels were not affected by Ar1260-exposure in HFD-fed mouse liver (Petri et al., 2022) and miR-24–3p not identified among altered miRNAs found in our RNA-seq analysis of the control or Ar1260-exposed livers used here, although miRNA-seq was not performed (Piell et al., 2023). Future studies will examine the mechanism by which Ar1260 exposure reduces hepatic SELENBP2 levels.

To our knowledge, this is the first examination of metals in the livers of rodents exposed to Ar1260, a predominantly NDL PCB mixture. We report increased Se, Cu, and Zn whereas Fe was decreased in the livers of Ar1260-exposed mice. Previously, a single i.p. injection of male Sprague-Dawley (SD) rats with a 1 μmol/kg dose of DL PCB126 resulted a decrease in Se and Zn and an increase in Cu after 2 wks. (Lai et al., 2010). Similarly, a single i.p. injection of male SD rats with a 5 μmol/kg (1.63 mg/kg) dose of DL PCB126 resulted in a time-dependent increase in hepatic Cu between days 3 – 12, increases in hepatic Zn at 9 hr. post-injection and again day 3, and an increase in hepatic Se at 9 hr., but decrease at day 12 (Klaren et al., 2015). The study also reported a non-significant decrease in hepatic Fe after days 6 – 12 (Klaren et al., 2015). In another study, female SD rats received multi i.p. injections of 100 μmol/kg DL PCB77, which was not detected in Ar1260 (Wahlang et al., 2014a), twice weekly for 1, 2 or 3 wks. (Twaroski et al., 2001). The authors reported a time- (accumulated dose) dependent decrease in hepatic Se levels to almost half of control levels by week 3, while male SD rats showed a significant decrease in hepatic Se (Twaroski et al., 2001). Although these limited studies did not demonstrate specific patterns of DL PCBs exposures on hepatic essential metals due to the differences in exposure doses, frequency (single vs multi exposure), and sexes, increased hepatic Cu and decreased hepatic Se levels showed apparent similarity among the three studies. Here we demonstrated that a single oral exposure of mice to Ar1260 also resulted in increased hepatic Cu, consistent with the findings in SD rats exposed to DL PCBs (Klaren et al., 2015; Qian et al., 2015; Twaroski et al., 2001).

Few studies in mice have investigated the effects of exposure to PCBs on liver Fe metabolism. A single i.p. injection of PCB126 (20 or 70 mg/kg) significantly reduced hepatic Fe and increased serum Fe in female C57Bl/6 mice at 48 hr. (Qian et al., 2015). In contrast, four i.p. injections of PCB126 (1 or 5 mg/kg at 2, 3, 4, and 5 wks.), which resulted in steatohepatitis, increased hepatic Fe and reduced serum Fe at the 6^th wk. in male C57Bl/6 mice (Kim et al., 2022). Another study administered Ar1260 (10 or 20 mg/kg) i.p. at 2, 3, 4, and 5 wks., which induced hepatic steatosis and inflammation, but Fe was not measured (52). Therefore, no model similar to that used in our current study is available to directly compare the effects of Ar1260 and DL PCBs on mouse liver metals. Here, decreased hepatic Fe levels in Ar1260 exposed mice were similar to the findings of a small decrease in hepatic Fe in female C57Bl/6 mice 48 hr after i.p. injection of PCB126 (Qian et al., 2015). Taken together, we suggest that exposure of rats and mice to DL PCBs and NDL PCBs caused significant changes of hepatic essential metals (Cu, Fe, Zn, and Se). Further examination of dose-, time-, sex-, species-dependent responses to PCBs on essential metal homeostasis remains further explored.

In human MASH livers, altered levels of essential elements including Se, Cu, Zn, and Fe have been reported (reviewed in (Day et al., 2021; Ma et al., 2022)) and dyshomeostasis of essential metals, particularly increased Cu or the ratio of Cu to Zn, has been extensively discussed as important mechanisms for the liver diseases and cancer development (Lin et al., 2006; Liu et al., 2023; Tamai et al., 2020). For example, patients with Wilson disease, caused by inactivation of the Cu transporter ATP7B that results in increased liver Cu liver, have steatosis, inflammation, fibrosis, and liver failure (Gottlieb et al., 2022). It is difficult to directly compare changes in essential metals and elements by PCB exposures in this mouse model of MASLD with human studies. There are few studies examining metals and elements in livers from human MASLD or MASH patients. Livers from patients with MASLD showed low Cu (Aigner et al., 2008) and high Fe (Younossi et al., 1999), findings opposite of the current study in Ar1260-exposed mice. Most human studies examine blood, serum, or plasma elements in MASLD patients. High blood Se was associated with MASLD while MASH was associated with low blood Se (Liu et al., 2022). Patients with MASLD showed lower serum Zn (Ito et al., 2020).

In the present study, hepatic Fe levels were reduced along with increased Se and GPX4, suggesting that the livers of the mice exposed to Ar1260 for 34 wks. demonstrated ferroptosis resistance. We also observed increased hepatic Cu. Exogenous Cu was reported to increase GPX4 protein levels along with ubiquitination and aggregation, promoting ferroptosis under certain conditions (Xue et al., 2023). The increase in GPX4 protein agrees with other reports: GPX4 protein was increased ~ 15–25% in C57Bl/6 mouse liver with MCD (methionine and choline deficient) diet-induced steatosis for 8–12 wks. (Lu et al., 2021) and GPX4 is elevated in MASH (Day et al., 2021). SELENOO localizes to mitochondria and participates in redox reactions (Han et al., 2014). SELENOO transcript abundance was lower in MASH compared to healthy controls (Day et al., 2021) and lower in HepG2 and Huh7 human hepatoma cells compared to normal human hepatocytes (Guariniello et al., 2015). In contrast with these findings, we demonstrated the increased expression of SELENOO in the liver of mice at 34 wks. after a single exposure to Ar1260. We suggest that increased GPX4 and SELENOO with Ar1260 exposure are likely hepatic protective measures.

In the present study, the biological role of the increase in hepatic Zn after Ar1260 exposure in male mice observed here is unknown. In SD rats, Zn status and dietary supplementation did not affect PCB126 hepatic toxicity (Klaren et al., 2016). In the present study, we speculate that the Ar1260-exposed mouse liver may be ferroptosis-resistant due to increased expression of GPX4 and a lack of free iron and apoptotic resistance due to the increased Zn and Zn-metallothionein. However, the chronic pathogenic metabolism and carcinogenic processes caused by PCB exposures and increased reactive metals such as Cu, which causes liver cancer development as reported in human studies (Deen et al., 2022; Donato et al., 2021; Ludewig and Robertson, 2013), was seen in some mice in our recent long-term Ar1260 study (Head et al., 2023).

Limitations

Despite the strengths of the current study, e.g., high sample numbers and unbiased analysis, the identification of 128 proteins differentially expressed in Ar1260-exposed mice 34 wks. after a single oral gavage of 20 mg/kg is a small change in the liver proteome. We eliminated proteins from analysis in which less than three liver samples in each of the two groups (control or Ar1260-exposed) showed values. Further, as reported for our study in HFD-fed mice exposed to Ar1260 (Klinge et al., 2021; Petri et al., 2023b; Petri et al., 2022; Petri et al., 2023c), and in other models of steatotic liver disease (Herranz et al., 2023), the transcript abundance does not reflect protein abundance likely due to multiple mechanisms and analysis issues (Fortelny et al., 2017; Poverennaya et al., 2017), including epitranscriptomic and epigenetic pathways regulating post-transcriptional processing and protein synthesis (Kan et al., 2022; Petri et al., 2023a).

Conclusions

In summary, this study identified 128 liver proteins, including changes in seven selenoproteins and four Sec synthesis proteins, regulated by exposure of mice to environmentally relevant PCB mixture, Ar1260. Major findings were the regulation of selenoproteins and SELENBP2 and the increase in liver Se, Cu, and Zn and the reduction in Fe in these mice. The lack of concordance between transcript and protein abundance in this Ar1260-exposure-mediated model of MAFLD in mice on a normal diet suggest that epigenetic and epitranscriptomic regulation of proteins in pathways associated with liver homeostasis and pathology require further investigation.

Table 3: Fifteen proteins uniquely identified in livers of control mice that were not detected in the livers of the Ar1260-exposed mice.

These proteins were detected in at least three of the five individual livers examined.

Accession	Description and role(s) in liver and liver diseases	# Unique Peptides	Gene Symbol
H7BXC3	Triosephosphate isomerase: catalyzes the interconversion of dihydroxyacetone phosphate (DHAP) and d-glyceraldehyde-3-phosphate (G3P) during glycolysis and gluconeogenesis TPI1 is reduced in HCC and its overexpression in HCC cell lines reduced cell proliferation, invasion, and migration and induced cell cycle arrest (Jiang et al., 2017).	1	Tpi1
Q80UT3	Tsfm protein (Fragment): TSFM activates mitochondrial elongation factor Tu. Mutation in TSFM results in severe infantile liver failure (Vedrenne et al., 2012). Lysozyme f3 – Lysozyme P, a glycosyl hydrolase;	1	Tsfm
A0A077S9N1	Lyz1 transcript is a maker for Kupffer cells and monocytes/macrophages and is elevated in MAFLD mouse liver after high fructose and HFD (Su et al., 2021).	1	Lyz1
Q9CWW6	Peptidyl-prolyl cis-trans isomerase NIMA-interacting 4 PIN4 protein was decreased in the livers of APOE3 (steatosis and fibrosis) versus APOE4 mice fed a HFD + 0.2% cholesterol (Huebbe et al., 2023).	1	Pin4
Q8BUR8	Steroid 5-alpha reductase C-terminal domain-containing protein – converts testosterone to dihydrotestosterone or 3α-androstenediol. Srd5a1 ko mice have reduced clearance of corticosterone without primary histological abnormality in the adrenal gland (Livingstone et al., 2014). Srd5a1 ko mice show no change in liver weight (males or females) (Windahl et al., 2011).	1	Srd5a1
Q9QYS9	KH domain-containing RNA-binding protein QKI is an RNA reader protein that binds specific RNA and is involved in pre-mRNA splicing, circRNA formation, and in mRNA export, stability and translation (Neumann et al., 2022). QKI is upregulated in HCC (Han et al., 2019). QKI serves as a coactivator for the PPARβ-RXRα complex, which controls the transcription of lipid-metabolism genes in oligodendrocytes (Zhou et al., 2020). Adipocyte-specific deletion of QKI in mice prevented HFD-induced liver steatosis and reduced hepatic TG content (Lu et al., 2020).	1	Qki
O35457	C-C chemokine receptor-like 2 - is a non-functional G-protein coupled receptor for the adipokine chemerin but does not transduce any signals to the cell. It may dampen inflammation by binding chemokines and reducing signaling (Del Prete et al., 2013). Hepatic CCRL2 transcript levels are positively associated with MASH (Zimny et al., 2017).	1	Ccrl2
Q3V250	Protein-serine/threonine kinase – located in the mitochondrial matrix Pdk3 is expressed in hepatocytes and was activated by liver overload and suppressed the activity of AKT, mTORC1, and mTORC2 thus curtailing insulin signaling, lipid synthesis, and lipid accumulation in liver (Mayer et al., 2019).	1	Pdk3
G3UW40	Mutated in colorectal cancers	1	Mcc
Q9QWR8	Alpha-N-acetylgalactosaminidase - removes terminal alpha-N-acetylgalactosamine residues from glycolipids and glycopeptides	2	Naga
P58044	Isopentenyl-diphosphate Delta-isomerase 1 Idi1 transcript levels were lower in rat liver regeneration than control livers (Xu et al., 2008). Idi1 was down-regulated at the hepatic transcript level by a 1-wk. high cholesterol (0.5%) diet in male and female C57Bl/6 mice (Maxwell et al., 2003).	2	Idi1
Q8CAS9	Protein mono-ADP-ribosyltransferase PARP9 The activity of PARP enzymes is regulated by lipid molecules including oxidized cholesterol derivatives, steroid hormones or bile acids and PARPs regulate lipid homeostasis (Szántó et al., 2021). PARP9 and PARP14 regulate the expression of the LDL receptor and apolipoproteins in macrophages (Szántó et al., 2021). PARP9 promotes macrophage responses to IFNγ (Iwata et al., 2016).	2	Parp9
E9Q616	AHNAK nucleoprotein (desmoyokin) Identified as a phosphoprotein in female C57Bl/6 mouse liver (Jin et al., 2004). Ahnak KO mice are resistant to HFD-induced obesity and hepatic steatosis because FGF21 and PPARα are upregulated in the absence of Ahnak expression (Kim et al., 2021).	3	Ahnak
Q8VCL2	Protein SCO2 homolog, mitochondrial SCO2 is copper chaperone required for Cytochrome c oxidase (COX) activity in Complex IV of the mitochondrial respiratory chain (Hill et al., 2017). Sco2 knockout mice show increased hepatic steatosis, elevated serum and liver TG and serum cholesterol (Hill et al., 2017).	4	Sco2

Highlights.

• Aroclor 1260 exposure altered mcm5U and i6A in Sec-tRNA

• Aroclor 1260 exposure altered 128 liver proteins

• Selenoproteins GPX4 and SELENBP2 were increased

• Selenoproteins SELENOS and SELENOF were reduced

• Liver selenium, copper, and zinc were increased and iron was reduce