Reduction of Chemokine CXCL9 Expression by Omega-3 Fatty Acids via ADP-Ribosylhydrolase ARH3 in MIN6 Insulin-Producing Cells
Funding: This work was supported by the Catalyst Award from the Human Islet Research Network (to E.S.N) (via U24 DK104162) and by the National Institute of Diabetes and Digestive and Kidney Diseases grants R01 DK138335 (to E.S.N.), U01 DK127505 (to L.S. and E.S.N.), R01 DK125360 (to L.S.), U01 DK127786 (to R.G.M.), R01 DK060581 (to R.G.M.).
Notice: Manuscript Authored by Battelle Memorial Institute Under Contract Number DE-AC05-76RL01830 with the US Department of Energy. The US Government retains and the publisher, by accepting this article for publication, acknowledges that the US Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so for US Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan: (http://energy.gov/downloads/doe-public-access-plan)
ABSTRACT
Type 1 diabetes (T1D) results from the autoimmune destruction of the insulin-producing β cells of the pancreas. Omega-3 fatty acids protect β cells and reduce the incidence of T1D, but the mechanism is poorly understood. We have shown that omega-3 fatty acids reduce pro-inflammatory cytokine-mediated β-cell apoptosis by upregulating the expression of the ADP-ribosylhydrolase ARH3. Here, we further investigate the β-cell protection mechanism of ARH3 by performing siRNA analysis of its gene Adprhl2 in MIN6 insulin-producing cells, subsequent treatment with a cocktail of the pro-inflammatory cytokines IL-1β + IFN-γ + TNF-α, followed by proteomics analysis. ARH3 regulated proteins from several pathways related to the nucleus (splicing, RNA surveillance, and nucleocytoplasmic transport), mitochondria (metabolic pathways), and endoplasmic reticulum (protein folding). ARH3 also regulated the levels of proteins related to antigen processing and presentation, and the chemokine-signaling pathway. We further studied the role of ARH3 in regulating the chemokine CXCL9. We found that ARH3 reduces the cytokine-induced expression of CXCL9, which is dependent on omega-3 fatty acids. In conclusion, we demonstrate that omega-3 fatty acids regulate CXCL9 expression via ARH3, which may have a role in protecting β cells from immune attack thereby preventing T1D development.
Significance of the Study: Omega-3 fatty acids have a variety of health benefits. In type 1 diabetes, omega-3 fatty acids reduce the islet autoimmune response and the disease development. Here, we studied the pathways regulated by the adenosine diphosphate (ADP)-ribosylhydrolase ARH3, a protein whose expression is regulated by omega-3 fatty acids. We showed that ARH3 reduces the expression of chemokines in response to omega-3 fatty acids. This represents an anti-inflammatory mechanism of omega-3 fatty acids that might be involved with protection against type 1 diabetes development.
1 Introduction
Type 1 diabetes affects approximately 8.4 million people worldwide and is caused by the autoimmune destruction of the insulin-producing β cells of the pancreas [1, 2]. This autoimmune response is characterized by infiltration of immune cells to the pancreatic islets, production of pro-inflammatory cytokines and chemokines, and circulating autoantibodies against islet proteins [3]. The immune cells, in conjunction with cytokines and chemokines, induce apoptosis of β cells [3]. However, the signaling networks that control this autoimmune attack are not completely understood.
Protein post-translational modifications, such as phosphorylation, acetylation, and ubiquitination, are major regulators of signal transduction in cells [4]. ADP-ribosylation is a modification on which adenosine diphosphate (ADP)-ribose is transferred to proteins, RNAs, and DNAs by ADP-ribosyltransferases (PARPs or ADRTs) using nicotinamide-adenosine dinucleotide (NAD) as a donor [5]. ADP-ribosylation can occur as single units (mono-ADP-ribosylation or MARylation) or chains (poly-ADP-ribosylation or PARylation) [5]. ADP-ribosylation also regulates many processes of the immune response. For instance, several PARPs play a central role in antiviral responses. ADP-ribosylation is a key player in stress granule formation, which sequesters viral and cellular RNAs into cytosolic aggregates and halts their translation [6]. Moreover, ADP-ribosylation enhances the activity of transcription factors involved in inflammatory responses, such as STATs, NF-κB, and NFATc3, upregulating the production of cytokines and chemokines [7-9].
Only a few functions are known for ADP-ribosylation in T1D development. For instance, deletion of PARP1 protects mice from developing diabetes induced by streptozotocin [10-12]. Moreover, various ADP-ribosyltransferases, including PARP3, -9, -10, -12, and -14, are strongly regulated by pro-inflammatory cytokines in pancreatic islets [13]. Adding to the previously known functions, we have found that the ADP-ribosylhydrolase ARH3, which cleaves ADP-ribosylation on serine residues [14], suppresses apoptosis in MIN6 cells [13]. The mechanism includes the degradation of SUZ12, a component of the histone methylation polycomb PCR2, induced by the omega-3 fatty acids, eicosapentaenoic acid (EPA), and docosahexaenoic acid (DHA). SUZ12 degradation reduces histone methylation, increasing the expression of ARH3. The increase in ARH3 expression presumably protects cells by hydrolyzing ADP-ribosylation involved in pro-apoptotic signaling [13]. As mentioned, very little is known about the signaling regulated by ARH3.
Here, we studied the role of ARH3 in regulating pro-inflammatory cytokine signaling. We performed proteomics of ARH3 knockdown MIN6 cells treated with or without a cocktail of pro-inflammatory cytokines IL-1β + IFN-γ + TNF-α. We chose these cytokines because their possible role in T1D development and being models of insulitis in vitro [15, 16]. IL-1β and TNF-α activate the MAP kinase/NF-κB pathways while IFN-γ primarily activates the JAK-STAT pathway, but they all contribute to apoptosis. IL-1β induces the expression of pro-inflammatory cytokines and chemokines, while IFN-γ mainly regulates the antigen presentation and pathogen defense. TNF-α balances the expression of genes related to inflammation, apoptosis, and cell survival. Together, these cytokines have an additive effect and orchestrate a comprehensive cellular immune response [16, 17]. In addition to the proteomics experiments, we also performed ELISA and qPCR to validate a protein regulated by ARH3 and to test the influence of omega-3 fatty acids on this regulation.
2 Material and Methods
2.1 MIN6 β Cell Line Culture and Treatment
MIN6 cells were cultured in DMEM containing 10% FBS, 50 µM 2-mercaptoethanol, and 1% penicillin-streptomycin and maintained at 37°C in 5% CO2 atmosphere. For knockdown experiments, cells were transfected using Lipofectamine RNAiMAX (Invitrogen) with SMARTpool ON-TARGETplus non-targeting (NT) siRNA (Cat# D-001810-10-50, Dharmacon) as control or siRNA targeting the ARH3 gene Adprhl2 (Cat# L-051819-01-0020, Dharmacon). To achieve robust knockdown of ARH3, cells were transfected a second time with NT or ARH3 siRNA 24 h after the first transfection in the proteomic experiment. In experiments utilizing cytokines, cells were transfected one time with NT or ARH3 siRNA. After 48 h transfection, cells were treated for 24 h with 100 ng/mL IFN-γ, 10 ng/mL TNF-α, and 5 ng/mL IL-1β. This combination and concentration of cytokines was based on the work by Andersen et al., which showed to have an additive effect and to induce an optimum kinase signaling response [17]. In experiments utilizing omega-3 fatty acids, cells were treated with the cocktail of the three cytokines in combination 80 µM DHA (docosahexaenoic acid) or EPA (eicosapentaenoic acid) dissolved in ethanol.
2.2 Proteomic Analysis
For proteomics analysis, cells cultured in 6-well plate (approximately 1 million cells) were treated as described above and rinsed with PBS before harvesting. Cells were dissolved in 50 mM NH4HCO3 containing 8 M urea and 10 mM dithiothreitol. After incubating for 1 h at 37°C with shaking at 800 rpm, 400 mM iodoacetamide was added to a final concentration of 40 mM, and the mixture was incubated for another hour in the dark at room temperature. The reaction mixture was diluted 8 folds with 50 mM NH4HCO3, and 1 M CaCl2 was added to a final concentration of 1 mM. Proteins were digested for 3 h at 37°C using trypsin at 1:50 enzyme:protein ratio. Digested peptides were desalted by solid-phase extraction using C18 cartridges (Discovery, 50 mg, Supelco) and dried in a vacuum centrifuge.
Peptides were analyzed on a Waters NanoAquity UPLC system coupled with a Q-Exactive mass spectrometer (Thermo Scientific), as previously described in detail [13]. Data were processed with MaxQuant software (v1.6.14.0) [18] by matching against the mouse reference proteome database from Uniprot Knowledge Base (downloaded on August 31, 2020). Searching parameters included protein N-terminal acetylation and oxidation of methionine as variable modifications, and carbamidomethylation of cysteine residues as fixed modification. Mass shift tolerance was used as the default setting of the software. Only fully tryptic-digested peptides were considered, allowing up to two missed cleaved sites per peptide. Identifications were filtered at 1% false-discovery rate in both peptide-spectrum match and protein group levels.
Quantitative information was extracted using the LFQ function of MaxQuant and analyzed in Perseus [19]. We filtered the data to include proteins that were detected in at least 66% of the replicates within at least one group. The abundance of ARH3 was also assessed using the intensity-base absolute quantification (iBAQ) function of MaxQuant to provide an absolute quantification, as the LFQ performs better for relative quantifications (fold changes). The data were normalized by the total intensity of the proteins in the samples. Missing values were imputed by using half of the minimum value. Proteins were considered differentially abundant with a p ≤ 0.05 using Student's t-test between groups. For functional-enrichment analysis, DAVID [20] analysis was used by querying the differentially abundant proteins against the entire mouse genome as background ontologies were considered enriched with a p ≤ 0.05 using Fisher's exact test. The network of immune proteins was plotted with VANTED [21].
2.3 ELISA
The levels of CXCL9 secreted to the media by MIN6 cells were quantified by Mouse DuoSet ELISA kit (R&D Systems). Polystyrene 96-well plates were coated with 100 µL of 200 ng/mL unconjugated goat anti-mouse CXCL9 antibody in PBS at room temperature overnight. Plates were washed 3× with 400 µL washing buffer and then blocked with 1% BSA for 1 h. After blocking, plates were washed 3× with 400 µL washing buffer. Cell culture supernatants were diluted 1:125 with reagent diluent and incubated for 2 h at room temperature, along with standards. Plates were washed 3× with 400 µL washing buffer and incubated with biotinylated goat anti-mouse CXCL9 antibody for 2 h at room temperature. Plates were washed 3× with 400 µL washing buffer and incubated with streptavidin-HRP incubation for 20 min at room temperature. Plates were washed 3× with 400 µL washing buffer, developed with tetramethylbenzidine and H2O2, and read at 570 nm in a plate reader (BioTek, USA).
2.4 Quantitative Real-Time PCR Analysis
Cells were harvested using Tri reagent (Zymo, Cat# R2050-1-200) and total mRNA was extracted using the RNA Clean & Concentrator-5 (Zymo, Cat# R1014). Total RNA quantity was measured using nanodrop and qRT-PCR assays were performed using QuantiNova SYBR Green RT-PCR Kit (Qiagen, Cat# ID: 208154) in StepOnePlus Real-Time PCR System (Applied Biosystems). Expression levels of Adprhl2 (R: CAAACTTCTGTACATCTTGGAC & F: AGAAACTCCTGAATCCCAAG) & Cxcl9 (R: GTTTGATCTCCGTTCTTCAG & F: GAGGAACCCTAGTGATAAGG) were normalized to two housekeeping gene Nono (R: CATACTCATACTCAAAGGAGC & F: CTTCTTGCTGACTACATTTCC) & Rpl13a (R: CAGGTAAGCAAACTTTCTGG, F: CCTATGACAAGAAAAAGCGG), as previously described [22, 23].
3 Results
3.1 ARH3-Regulated Proteins in MIN6 Cells
To investigate the roles of ARH3 in β-cells, we performed siRNA analysis of its gene Adprhl2 (siARH3) in MIN6 insulin-producing cells for 24 h, followed by a 24 h treatment with a cocktail of pro-inflammatory cytokines (IL-1β + IFN-γ + TNF-α) and performed proteomics analysis (Figure 1A). The analysis resulted in the identification and quantification of 4749 proteins (Table S1). The abundance of ARH3 was assessed by extracting the intensity-based absolute quantification (iBAQ) values with the MaxQuant software, which revealed a reduction of 82% in the ARH3 protein abundance (Figure 1B), similar to previously observed [24]. We first analyzed the impact of siARH3 in cells without cytokine treatment. The comparative analysis revealed that 159 and 161 proteins were up and downregulated with siARH3, respectively (Figure 1C). Enrichment analysis of the subcellular localization revealed that the cytoplasm, nucleus, mitochondrion, and endoplasmic reticulum are the most common locations of proteins regulated by ARH3 (Figure 1D). Further, a functional-enrichment analysis revealed an enrichment of proteins from a variety of pathways (Figure 1E). In agreement with the subcellular localization, oxidative phosphorylation, Parkinson's disease, prion disease, amyotrophic lateral sclerosis, Huntington's disease, and thermogenesis pathways (Figure 1E) have the mitochondrion as one of their main components. In addition to oxidative phosphorylation and thermogenesis, the siARH3 regulated two other metabolic pathways, folate and pyrimidine metabolisms (Figure 1E). This indicates that ARH3 might have a role in regulating the mitochondrion and cellular metabolism. Similar enrichment of pathways was observed in nuclear proteins, including the spliceosome, thermogenesis, and circadian entrainment pathways (Figure 1E) that are associated at least in part with this organelle. Protein processing in the endoplasmic reticulum (ER), Parkinson's disease, prion disease, amyotrophic lateral sclerosis, and Huntington's disease are pathways associated with the endoplasmic reticulum (Figure 1E). Overall, the functional-enrichment analysis indicates that ARH3 regulates pathways associated with the mitochondrion, nucleus, and ER.
3.2 ARH3-Regulated Proteins in Cytokine-Treated Cells
We subsequently studied the impact of ARH3 in regulating proteins in cytokine-treated β-cells by comparing the parental and siARH3 MIN6 cells (Figure 1A). Confirming previous findings [13], ARH3 abundance increased by 49% with the cytokine treatment (Figure 1B). Similar to the untreated cells, siARH3 reduced the abundance of ARH3 in 74% in cytokine-treated cells (Figure 1B). Our analysis of MIN6 cells that were treated with the cytokine cocktail showed that 158 and 359 proteins (517 total) were significantly up and downregulated by siARH3, respectively (Figure 2A). A functional-enrichment analysis showed a similar trend of the cytosol, nucleus, mitochondrion, and ER being the most abundant cellular localizations of the ARH3-regulated proteins in cytokine-treated cells (Figure 2B). However, the nucleus had a noticeable increase in the number of ARH3-regulated proteins, an increase from 44 (13.7% of the proteins assigned to specific cellular localizations) (Figure 1C) to 219 proteins (42% of the proteins assigned to specific cellular localizations) (Figure 2B). Mitochondrial and ER had a similar proportion of proteins assigned to specific cellular localizations in both untreated (mitochondria–13.7%, ER–11.5%) and cytokine-treated cells (mitochondria–13.4%, ER–12.3%) (Figures 1C and 2B). The increase in nuclear proteins regulated by ARH3 after cytokine-treatment was also reflected at the pathway-enrichment level. The spliceosome and thermogenesis pathways were similarly regulated by ARH3 in untreated and cytokine-treated cells, although the number of regulated proteins increased after cytokine treatment (Figures 1D and 2C). In cells treated with cytokines, additional nuclear pathways were enriched among the ARH3-regulated proteins, including nucleocytoplasmic transport, mRNA surveillance, and viral carcinogenesis pathways (Figure 2C). As histones are major ARH3 substrates [25], we looked at the levels of these proteins in the proteomics analysis. We found that the siARH3 reduced the levels of histone H1.5 in the cells treated with cytokines and the levels of histone H1.2 and H2B in untreated cells (Figure 3). Together, these results show that in the presence of cytokines, ARH3 regulates a variety of proteins involved in nuclear processes, such as splicing, nucleocytoplasmic transport, and mRNA surveillance.
3.3 Regulation of Immune Proteins by Both Cytokine and ARH3 Signaling
We next focused on the proteins that were significantly regulated by the cytokine cocktail and were further regulated by siARH3. Comparing the non-target siRNA sample treated versus untreated with cytokines led to a total of 621 differentially abundant proteins. When comparing these 621 proteins with the 517 proteins that were significant in siARH3 samples, a total of 128 proteins were commonly regulated by the cytokine treatment and siARH3 (Figure 4A). The functional-enrichment analysis revealed that these proteins were enriched in a variety of immune-related pathways, such as antigen processing and presentation (including proteasome), chemokine signaling pathways, and infections (Epstein–Barr virus, human cytomegalovirus, human immunodeficiency virus, toxoplasmosis, Kaposi sarcoma-associated herpesvirus and viral carcinogenesis) (Figure 4B). To further investigate the regulation and relationship of these 128 proteins, we extracted the interaction networks from the String database. We found 18 inter-connected proteins from the chemokine signaling pathway and immune system pathway (Figure 4C). The levels of the transcription factors Stat1 and Stat3 were strongly induced by the cytokine treatment, but to significantly less extent in siARH3 cells (Figure 4C). A similar trend was observed for the proteins involved in antigen processing and presentation, H2-K1, B2m, Tap2, Psmb8, Psmc2, Psmb10, and Hsp90ab1, which were all induced by the cytokine treatment with a significantly less extent in siARH3 cells (Figure 4C). This is possibly in response to their transcription factors Stat1 and Stat3 regulation. Conversely, the levels of the autophagy receptor Sqstm1, inducible nitric oxide synthase Nos2, and chemokine Cxcl9 were induced by cytokine treatment and further upregulated in siARH3 cells (Figure 4C). These results show that ARH3 regulates cytokine-mediated signaling, including the regulation of its main transcription factors Stat1 and Stat3, along with other immune proteins.
We further studied the possible mechanisms of how ARH3 regulates the expression of the chemokine CXCL9. We treated MIN6 cells with different combinations of IFN-γ, IL-1β, and TNFα to determine the contribution of each of these cytokines in the production of CXCL9 by ELISA. None of the cytokines by themselves induced CXCL9 production (Figure 5A). Only combinations contained IFN-γ induced CXCL9 production with all three cytokines together inducing the maximum CXCL9 production (Figure 5A). Production of CXCL9 is further increased in siARH3 cells (Figure 5B). As we previously demonstrated that ARH3 expression is regulated by omega-3 fatty acids [13], we investigated whether this regulation would also influence the expression of CXCL9. MIN6 cells were treated with the omega-3 fatty acids, EPA and DHA, in combination with the cytokine cocktail for 8 h. The levels of CXCL9 transcript were measured by qPCR. The expression of CXCL9 was induced by the cytokine cocktail, which was reduced by 38.6% and 44.4% by EPA and DHA, respectively (Figure 5C). We observed similar regulation at the secreted CXCL9 level, as measured by ELISA. At 8 h post-treatment, CXCL9 was only detected after cytokine treatment, which was reduced by 89.6% and 61.9% by EPA and DHA, respectively (Figure 5D). After 24 h of cytokine treatment, the levels of CXCL9 increased by 46 folds in media, which was reduced by 65.4% and 35.7% by EPA and DHA, respectively (Figure 5E). To test if the omega-3-mediated reduction of CXCL9 production was dependent on ARH3, siARH3 cells were pre-treated with EPA and DHA for 48 h and then treated with the cytokine cocktail. EPA reduced the production of CXCL9 in 71.5% in cells treated with non-targeting siRNA and in 61.9% in siARH3 cells (Figure 5F). DHA reduced the production of CXCL9 in 31.0% in cells treated with non-targeting siRNA, which was completely abolished in siARH3 cells (Figure 5F). These results show that ARH3 downregulates the expression/production of CXCL9, which is mediated by ARH3 and omega-3 fatty acids.
4 Discussion
Here we investigate the roles of ARH3 in MIN6 cells by proteomics analysis. Our data show that ARH3 regulates pathways associated with mitochondrion, nucleus, and ER. This supports previous observations that ARH3 hydrolyses ADP-ribosylation on both mitochondrial and nuclear proteins [26, 27]. The regulation of mitochondrial proteins by ARH3 was accompanied by enrichment in metabolic pathways in both untreated and cytokine-treated MIN6 cells. However, the regulation of specific metabolic pathways by ARH3 has not been reported and will require additional studies. The regulation of mitochondrial proteins by ARH3 may play a role in apoptosis. We have previously demonstrated that this protein reduces apoptosis in MIN6 cells [13], and the mitochondrion is a main organelle involved in this process. In addition, ADP-ribosylation has also been described to play an important role in apoptosis [28]. In the ADP-ribosylation database ADPriboDB 2.0 [29], several apoptotic signaling proteins have been listed as being ADP-ribosylated, including FADD, Casp8, BID, BCL2, BCL2L1, Bax, Cycs, Diablo, BIRC2, and BIRC6. More recently, the apoptotic proteins API5, ACIN1, CCAR1, PARW, AATF, and CIAPIN1, were found to be ARH3 substrates [25]. However, the strongest evidence is that ARH3 regulates apoptotic signaling by preventing the release of apoptosis-inducing factor AIF from mitochondria [27]. Therefore, further investigation is required to gain a better understanding of the ARH3 functions in the mitochondrion. Our data also show that ARH3 regulates a variety of proteins from pathways associated with the ER. To the best of our knowledge, no functions have been associated with ARH3 in this organelle.
Our proteomics data also show that nuclear ARH3-regulated proteins were enriched in RNA-related pathways, such as splicing, nucleocytoplasmic transport, and mRNA surveillance. ARH3 hydrolyzes ADP-ribosylation from RNA chains [14, 30], but its function is still poorly understood. However, ADP-ribosylation itself is well understood to regulate RNA biology, including all forementioned pathways [31]. Therefore, it is possible that ARH3 is a major regulator of ADP-ribosylation-mediated RNA regulation in cells. Our data also show that ARH3 regulates proteins involved in gene expression, such as histones and the transcription factors Stat1 and Stat3. The reduction of Stat1 and Stat3 in siARH3 cells might be related to the downregulation of antigen processing and presentation proteins. Stat1 and Stat3 are transcription factors of antigen processing and presentation proteins [32], and they were regulated at similar levels. The reduction in histone levels in siARH3 β-cells might be a measurement artifact of the reduced detectability of the ADP-ribosylated peptide in the mass spectrometer. ADP-ribosylation in histones is well described and ARH3 is a major regulatory enzyme of this process [25]. ADP-ribosylation of histones are associated with chromatin opening and increase in gene expression, including cytokine and chemokine genes [28]. Therefore, ARH3-regulation of histone ADP-ribosylation could be associated with CXCL9 expression. The inflammatory transcription factor NF-κB activity is enhanced by PARP1, not by ADP-ribosylation of itself [8]. It is possible that NF-κB activity is increased by ADP-ribosylation of other proteins, such as histones. Another possibility is that ARH3 is removing ADP-ribosylation from Stat1 and repressing its activity [33], as this transcription factor is a major activator of CXCL9 gene expression in β cells [7, 9].
We also showed that reduction of cytokine-induced CXCL9 expression by ARH3 involves omega-3 fatty acids. Omega-3 fatty acids reduce the risk of developing islet autoimmunity in humans by 55% [34] and diabetes in mice by 60% [13]. We previously demonstrated that omega-3 fatty acids improve the expression of ARH3 by reducing the level of Suz12, a component of the histone methylation polycomb PCR2 [13]. Omega-3's are anti-inflammatory fatty acids that regulate the expression of pro-inflammatory cytokines and chemokines by engaging the G protein-coupled receptor GPR120, inhibiting the activity of the pro-inflammatory transcription factor NF-κB, and activating the anti-inflammatory transcription factor peroxisome proliferator-activated receptor γ [35]. The signal transduction of the GPR120 involves the sequestration of the NF-κB activator Tab1, but this scenario is not completely understood [35].
In conclusion, this paper found that ARH3 regulates a variety of proteins and pathways, in particular, the ones related to the cell mitochondria and nuclei. We show that ARH3 reduces the expression of the chemokine CXCL9 in response to cytokine signaling, which is mediated by omega-3 fatty acids.
4.1 Limitations of the Study
In this study, we focus our efforts on proteins regulated by ARH3. As ARH3 is an ADP-ribosylhydrolase, it is likely that this regulation also involves the cleavage of specific ADP-ribosylation sites. The analysis of ADP-ribosylation sites is challenging due to their sizes (it can form polymers of many units), being labile in MS/MS analysis, and stability during storage [36, 37]. However, recent developments in enrichment and depolymerization have improved the ADP-ribosylation proteome analysis [25, 38, 39]. Such techniques open opportunities for further study and characterize the omega-3 protective signaling that is transduced via ARH3. Another limitation of our study is that the p values are not adjusted. We tested adjusting the p values by Benjamini–Hochberg method, but the correction was too stringent and removed almost all the significant hits, including the validated protein CXCL9.
Acknowledgments
The authors thank Drs. Charles Ansong and Charanya Muralidharan for their insightful inputs. Part of the work was performed in the Environmental Molecular Sciences Laboratory, a U.S. Department of Energy (DOE) national scientific user facility at Pacific Northwest National Laboratory (PNNL) in Richland, WA. Battelle operates PNNL for the DOE under contract DE-AC05-76RLO01830.
Conflicts of Interest
The authors declare no conflicts of interest.
Open Research
Data Availability Statement
Proteomics data were deposited into Pride repository (www.ebi.ac.uk/pride) under accession number PXD021501.
