Insulin post-transcriptional regulation via PARP12-mediated ADP- ribosylation

Insulin post-transcriptional regulation via PARP12-mediated ADP- ribosylation | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Insulin post-transcriptional regulation via PARP12-mediated ADP- ribosylation Soumyadeep Sarkar, Fangjia Li, Youngki You, Emily C. Elliott, and 17 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6473437/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract ADP-ribosylation is a common modification that occurs in proteins and nucleic acids, regulating many cellular processes ranging from DNA repair to inflammatory signaling. ADP-ribosylation plays an important role in cancer biology, infectious diseases, and obesity, but its role in the development of type 1 diabetes is not well understood. Here, we studied the role of ADP-ribosyltransferase PARP12 in type 1 diabetes development. PARP12 expression is highly induced in human islets treated with pro-inflammatory cytokines or β cells from diabetic donors. Proteomics analysis of MIN6 insulin-producing cells identified that the RNA machinery is regulated by PARP12 during inflammation. PARP12 also ADP-ribosylates 150 mRNAs, including the insulin mRNA. This mRNA ADP-ribosylation in turn modifies transcript localization and halts translation. Overall, our data identified a role for PARP12 in ADP-ribosylation and translation halting of mRNAs, which may affect insulin production during insulitis. Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Type 1 diabetes (T1D) is a devastating disease that reduces life expectancy by 11 years for men and 13 years for women [ 1 ]. The disease currently affects approximately 8.4 million people worldwide [ 2 ]. Treatment is lifelong and based on controlling the blood glucose levels with diet, exercise, and insulin administration. Despite advances in blood glucose control, the disease leads to complications such as cardiovascular and renal diseases, which account for the reduction in life expectancy. Mechanistically, T1D is an autoimmune disease with characteristic insulitis pathology [ 3 ] marked by infiltration of immune cells and apoptosis of insulin-producing pancreatic β cells mediated by pro-inflammatory cytokines and chemokines [ 4 ]. However, the highly complex cellular signaling transduced by the cytokines and chemokines is only partially characterized. ADP-ribosylation plays a central role in regulating DNA repair, inflammatory signaling and anti-viral response [ 5 ], but its role in insulitis is not well understood. This modification is characterized by the addition of adenosine diphosphate (ADP)-ribose units to proteins and nucleic acid by ADP-ribosyltransferases (PARPs or ADRTs) with nicotinamide-adenosine dinucleotide (NAD) as a donor [ 6 , 7 ]. ADP-ribosylation can also modify RNAs[ 8 ], but has only been characterized in bacteria, where it plays a role in antiviral response[ 9 ]. ADP-ribosylation occurs as single units (mono-ADP-ribosylation or MARylation) or chains (poly-ADP-ribosylation or PARylation) [ 7 ]. ADP-ribosylation is reversible and can be removed by hydrolases such as poly(ADP-ribose)glycohydrolase (PARG) and ADP-ribosylhydrolases [ 10 ]. In DNA repair, ADP-ribosylation labels damaged chromatin regions, recruiting the repair machinery. In immune cells, ADP-ribosylation enhances the activity of key transcription factors in inflammation, such as STATs, NF-κB, and NFATc3 [ 11 – 13 ]. Pro-inflammatory cytokines, such as IFN-γ, induce the expression of PARP9-14 which in turn regulate antiviral responses by targeting viral RNAs to degradation, viral replication, and protein translation [ 5 ]. Related to T1D, PARP1 deletion prevents the development of diabetes induced by streptozotocin in mice [ 14 – 16 ]. We have recently shown that the ADP-ribosylhydrolase ARH3, which removes ADP-ribosylation from serine residues [ 17 ], mediates the reduction of cytokine-induced apoptosis and production of chemokine CXCL9 by omega-3 fatty acids in MIN6 insulin-producing cells [ 18 , 19 ]. Mechanistically, omega-3 fatty acids induce the degradation of SUZ12, a component of the histone methylation polycomb PCR2, leading to a reduction in histone methylation and increasing the expression of ARH3. However, it is unknown if the other 16 mammalian PARPs or 3 hydrolases have functions in T1D development. Here, we investigated the role of PARP12 in insulitis owing to its expression profile in previous omics datasets [ 19 – 22 ]. We performed genomic analysis to determine mechanistically how the expression of PARP12 is regulated by pro-inflammatory cytokines in insulitis. We next performed a combination of RNA silencing and proteomics to determine possible functions of PARP12. This investigation identified PARP12 as an RNA ADP-ribosyltransferase, playing a role in post-transcriptional regulation of inflammation. Material and Methods Data reanalysis Protein relative abundances (fold changes) of all 17 mouse and human PARPs were extracted along with their statistical significance from previously published proteomics datasets from our laboratory. The model systems included mouse insulinoma cell line MIN6 treated with cytokine cocktail CT1: IFN-γ, IL-1β and TNFα for 24 h [ 19 ], human islets treated with CT2: IFN-γ and IL-1β for 24 h [ 21 ], and human insulin-producing cell line EndoC-βH1 treated with CT2: IFN-γ and IL-1β for 48 h [ 22 ] or CT3: IFN-α for 8 h and 48 h [ 20 ] ( ESM Table S1 ). For single-cell RNA seq analysis, data from human islets were downloaded from the Human Pancreas Analysis Program (HPAP) portal ( https://hpap.pmacs.upenn.edu/ ) and processed with Cell Ranger (v6.1.2) as previously described [ 23 ] (donor characteristics in ESM Table S2 ). After cell type annotation, we retrieved β cells from donors normoglycemic (non-diabetic, ND), positive for 1 (AAB1+) or multiple autoantibodies (AAB2+), and with T1D. In total, we analyzed 6,423 β cells from 15 ND, 5,751 from 9 AAB1+, 4,977 from 2 AAB2+, and 2,013 from 11 T1D donors. Differential gene expression analysis for β cells between T1D, AAB2 + and AAB1 + vs ND was performed using the FindMarkers function from Seurat, employing the Wilcoxon rank-sum test. P-values were adjusted using the Bonferroni correction method to account for multiple comparisons for all the genes ( ESM Table S3 ). Chromatin features (ATAC-seq and H3K27ac CHIP-seq) and gene expression (RNA-seq), along with annotated regulatory elements in human pancreatic islets treated or untreated with CT2 (IFN-γ and IL-1β) for 48 h, were obtained from our previous publication [ 22 ]. The datasets were lifted over from hg19 to hg38 genome assemblies. We selected regulatory elements within a 40 kb window centered on the PARP12 locus and divided them into three groups: Promoter/SRE (n = 1, PARP12 promoter), Distal/Opening_IRE (n = 2, increased accessibility and H3K27ac upon cytokines), and Distal/SRE (n = 7, without after cytokine exposure). Motif enrichment analysis was conducted using AME from the MEME suite (version 5.4.1) [ 24 ], testing for enrichment of cytokine-regulated TF motifs (i.e., transcription factors whose expression is upregulated upon cytokine exposure) across the three groups of regulatory elements, using shuffled sequences as a control. Cell culture MIN6 cells were a gift from the Yamagata lab and were grown in DMEM containing 4.5 g/L each of D-glucose and L-glutamine, 10% FBS, 100 units/mL penicillin, 100 µg/mL streptomycin, and 50 mM 2-mercaptoethanol maintained at 37 ºC in a 5% CO 2 atmosphere. For PARP12 knockdown experiments, cells were treated at 80% confluency using Lipofectamine RNAiMAX (Invitrogen, Cat# 13778150) with SMARTpool ON-TARGETplus non-targeting siRNA (Dharmacon, cat#D-001810-10-20) or siRNA targeting Parp12 (Dharmacon, cat# L-065127-01-0020), with concomitant treatment with CT2 (100 ng/mL IFN-γ: R&D, cat#485-MI-100, 10 ng/mL TNF-α: R&D, Cat#410-MT-010, and 5 ng/mL IL-1β: R&D, cat #401-ML-005) for 24 h. For pulse-chase experiments with stable isotope labeling by/with amino acids in cell culture (SILAC), cells were treated in DMEM containing heavy 13 C 6 15 N 2− lysine (Thermo Fisher Scientific, Cat# A33969). Label-free proteomics analysis The MIN6 cell pellets were resuspended in 50 mM Tris-HCl, 8 M urea, and 10 mM dithiothreitol, and incubated for 1 h at 37 ºC with shaking at 800 rpm. Subsequently, 400 mM iodoacetamide was added to a final concentration of 40 mM, and the samples were incubated for 1 h in the dark at room temperature. The samples were then diluted 8-fold with 50 mM Tris-HCl, and 1 M CaCl 2 was added to a final concentration of 1 mM. Proteins were digested overnight at room temp using trypsin at an enzyme-to-protein ratio of 1:50. The digested peptides were desalted using C18 cartridges (Discovery, 50 mg, Sulpelco) and dried in a vacuum centrifuge. Peptides were analyzed using a Q Exactive Plus mass spectrometer (Thermo Scientific) as previously described[ 18 ]. The data were processed with MaxQuant software (v.1.5.5.) [ 25 ] using the mouse reference proteome database from UniProt Knowledge Base (downloaded on 02-18-2022, 17082 entries). Protein N-terminal acetylation and oxidation of methionine were set as variable modifications, and cysteine carbamidomethylation was set as a fixed modification. The software's default mass shift tolerance was used. Only fully tryptic-digested peptides were considered, with up to two missed cleavage sites allowed per peptide. Protein quantification was performed using the intensity-based absolute quantification (iBAQ) method [ 26 , 27 ]. The data was normalized across samples based on total protein level, followed by substituting 1/4th of the lowest value from the dataset as missing values and log2 transformed. Statistically significant proteins (p ≤ 0.05) were determined using Student’s t -test. Network analysis of the statistically significant proteins was done with String database (V12.0) [ 28 ]. Immunofluorescence imaging MIN6 cells were cultured in 35 mm culture dishes (MatTek, Cat#P35G-1.5-10-C) and remained covered with PBS (Gibco, Cat#10010023) throughout the imaging process. They were permeabilized with 0.5% Triton X-100 for 30 min, followed by blocking with 1× PBS containing 5% goat serum and 0.3% Triton X-100 for 1 h at room temperature. The sections were incubated in 1:500 dilution of monoclonal rabbit anti-insulin IgG (Cell Signaling Technology, 3014), PARP12 (G bioscience, ITN2133-50u-555) and G3BP (Thermo Fisher, Cat#CL488-66486) overnight at 4 ºC. The primary antibody was thoroughly washed with PBS followed with incubation in 1:1000 dilution goat anti-rabbit IgG (H + L) highly cross-adsorbed secondary antibody, Alexa Fluor Plus 800 (Thermo Fisher Scientific). For the fliFISH analysis, each primary FISH probe was composed of about 20 nucleotides complementary to the target mRNA and 28 nucleotides complementary to the secondary probe (Integrated DNA Technologies) [ 29 ] ( ESM table S4 ). The properties of the probes, such as Tm, DG, and hybridization efficiency, were evaluated [ 30 ] and their targeting specificity was confirmed with BLAST ( https://blast.ncbi.nlm.nih.gov/Blast.cgi ). FISH probe solution was prepared by incubating 0.5 µM of each primary probe and 5 µM secondary probe (conjugated Alex 647 dye in both oligonucleotide termini) in 50 M Tris-HCl, pH 7.9, containing 100 M NaCl, 10 mM MgCl 2 and 1 mM dithiothreitol at 85 ºC for 3 min, followed by a gradual cool down to room temperature to hybridize the primary and secondary probes. Cell slides were immersed in 70% ethanol overnight at 4 ºC to permeabilize cell membranes. The slides were incubated with the FISH probe solution diluted to a final concentration of 4 nM of each probe in hybridization buffer (10% dextran sulfate, 2 mM vanadyl-ribonucleoside complex, 0.02% BSA, 2 × SSC, 10% formamide) overnight incubation at 37 ºC. Slides were thoroughly rinsed with buffer containing 2 × SSC and 10% formamide. The cell nuclei were stained with 0.05% DAPI. Imaging was carried out using a Zeiss LSM 710 confocal laser scanning microscope, captured using a 100× oil immersion objective (NA 1.47). Laser excitation was set to 633 nm for Ins2 (Alexa Fluor 647), 542 nm for PARP12 antibody (Alexa Fluor 555), and 405 nm for DAPI. Fluorescence emission was detected using spectral settings of 650–700 nm for Alexa Fluor 647, 570–620 nm for Alexa Fluor 555, and 420–480 nm for DAPI. Key imaging parameters, including detector gain, laser power, and pinhole size, were optimized to minimize photobleaching while ensuring a high signal-to-noise ratio for each channel. Image processing was performed using FIJI/ImageJ, applying only linear adjustments to brightness and contrast without modifying pixel intensity values. Prediction of RNA binding to PARP12 zinc finger motifs The interaction strength of Ins2 pre-mRNA with PARP12 was calculated using the RPISeq program ( http://pridb.gdcb.iastate.edu/RPISeq/index.html ). For PARP12, we used sequences from three zinc finger (ZnF1-3) domains in the following combinations: ZnF1 and 2, ZnF2 and 3, and ZnF1, 2 and 3. The interaction strength was predicted in 50-nucleotide segments of Ins2 pre-mRNA with 10-nucleotide overlapping in both ends. Prediction of the PARP12 structure The PARP12 structure predicted with AlphaFold (identifier AF-Q8BZ20-F1) was downloaded from Uniprot Knowledgebase and analyzed with Discovery Studio Visualizer v21. The Ins2 linear RNA oligonucleotide (ACAGGCAUGCAACCCCUGCCACCUG) with the highest predicted binding was positioned on PARP12, based on the ZnF motif orientations similar to RNA-bound protein structures of PDB, 3D2S and 7C07. The geometry of the RNA-bound PARP12 model was optimized using Discovery Studio Visualizer. ADP-ribosylated RNA pulldown MIN6 cells were harvested using Tri reagent (Zymo Research, R2050-1-200) and the RNA was extracted using RNA Clean & Concentrator™-5 kit (Zymo Research, R1014), following manufacturer instructions. To test the specificity of the ADP-ribosylation-binding AF1521 macrodomain, an RNA aliquot was pre-treated with 2 nM SARS-CoV2 macrodomain MAC1 at room temp for 1 h in hydrolase reaction buffer (250 mM HEPES, 750 mM NaCl, 10 mM β-mercaptoethanol, 0.05 mM Triton X100, 5 mM MgCl 2 , 1X RNAsecure) to remove ADP-ribosylation. Equal amount of total RNA from each sample was incubated with either 20 µL of Af1521 macrodomain conjugated to magnetic beads (Tulip Biolabs, 2426) or control (naked or mutated Af1521) beads in 600 µL RNA pulldown buffer (10 mM Tris-HCl, 0.6 M NaCl, 0.1% NP40, 1X RNA secure) overnight at 4°C. After overnight incubation, the beads were gently washed with RNA pulldown buffer three times, and 50 µL of Tri reagent was added to elute RNA from the beads, followed by RNA Clean & Concentrator™-5 extraction. Quantitative real-time PCR analysis Extracted total and Af1521-enriched RNAs were quantified using nanodrop and expression level of genes were measured in StepOnePlus RTPCR systems using QuantiNova™ SYBR Green RT-PCR reagent (Qiagen, Cat. No. / ID: 208154). Pre-designed mouse primers were ordered from Millipore sigma (Cat# KSPQ12012) and IDT ( EMS Table S5 ). The average CT value of two housekeeping genes Rlp13 and Nono was used to calculate the fold change of genes using Livak’s method [ 18 , 31 ]. RNAseq analysis RNAs eluted post-incubation with AF1521 and Control (mutant AF1521) beads were processed for cDNA synthesis using SMARTer® Universal Low Input RNA Kit (Takara, cat#634938) according to instruction manual, followed by template library prep using NEBNext® Ultra™ II FS DNA Library Prep Kit for Illumina (NEB, cat#E7805S). Single-read sequencing of the barcoded cDNA libraries with a read length of 150 were performed on NextSeq 550 plateform using NextSeq500/550 High Output v2.5 150 cycles (cat#20024907). Quality control of RNA-seq data was performed using FastQC (Version 0.12.0) ( https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ ). Read trimming was performed using BBDuk ( https://github.com/BioInfoTools/BBMap/blob/master/sh/bbduk.sh ). Reads were aligned to the mouse genome GRCm39 using STAR[ 32 ] and mapped to genome features using the featureCounts program from the subread package ( https://subread.sourceforge.net/ ). To account for technical variability raw counts were batch corrected with ComBat-Seq[ 33 ] with default parameters prior to differential expression analysis with DESeq2[ 34 ]. Heavy isotope-labeled lysine proteomics PARP12 siRNA cells were treated with CT1 in heavy isotope-labeled lysine SILAC media for 24 h. Samples were submitted to proteomics analysis as described above with the exception that the digestion was done with endoproteinase LysC instead trypsin. Raw files were processed in MaxQuant (v.2.5.1.0), using mouse reference protein datasets downloaded from UniprotKB (2023-03-01, 21,949 entries). The multiplicity was set to 2 considering light and heavy 13 C 6 15 N 2− lysine labels. Searches were performed with oxidation of methionine and protein N-terminal acetylation as variable modifications and cysteine carbamidomethylation as fixed modification. The quantification parameters included re-quantification and matching between runs with a matching and aligning time windows of 1 and 20 min, respectively. All remaining parameters were kept as the default settings. Intact insulin analysis by mass spectrometry Pellets of 300,000 cells pulsed-chased in SILAC media were lysed in 10 µL of water. Aliquots of 2 µL were spotted onto poly-lysine coated indium-tin oxide slides (Bruker Daltonics, Billerica, MA). After drying, slides were serially washed for 30 s each with 70% ethanol, 100% ethanol, and 50 mM ammonium acetate. The spotted cells were prepared by spraying 5% acetic acid in 50% ethanol and subsequently applying 15 mg/mL 2,5-dihydroxyacetophenone as matrix in 90% acetonitrile with 0.2% trifluoroacetic acid using an M5 sprayer (HTX Technologies, Chapel Hill, NC)[ 35 ]. Slides were analyzed on an orbitrap mass spectrometer (Q Exactive HF, Thermo Fisher Scientific, Bremen, DE) retrofit with ultrahigh mass range (UHMR) boards and an EP-MALDI source (Spectroglyph LLC, Kennewick, WA). Briefly, the instrument was operated under custom privileges licenses, and mass spectra were collected with transients of 1.024 s (480k mass resolution at m/z 200) and a m/z range of 2000 to 20000. The raw data and position files were imported into SCiLS Lab (v.2021c; Bruker Daltonics, Bremen, DE) using automatic import and processing settings. Maximum intensity values were extracted with root mean square normalization across conditions. Data visualization and statistical analysis Microsoft Excel was used to perform basic data analysis with GraphPad Prism 9 (Version 9.4.1) was used for statistical analysis and data visualization. Perseus-Max-Quant was used to create Heatmaps. Pathway analysis for Figs. 5 and 6 was performed using the core analysis setting of Ingenuity Pathway Analysis software using the Reactome database. Results PARP12 regulation by cytokines in islets and beta cells. To investigate the possible roles of PARPs in insulitis, we re-analyzed published proteomics data from human islets, EndoC-βH1 cells and MIN6 cells treated with different combination of pro-inflammatory cytokines (CT1: IFN-γ, IL-1β and TNFα; CT2: IFN-γ and IL-1β; CT3: INFα)[ 19 – 22 ]. This re-analysis revealed that ADP-ribosyltransferases PARP9, 10, 12, and 14 are consistently upregulated by the pro-inflammatory cytokines in all 5 datasets (Fig. 1 A). We investigated the expression of PARPs in β cells from single-cell RNAseq data from donors at different stages of T1D development, determined by the number of detected islet autoantibodies and disease onset. The expression levels of Parps 1, 4, 5B, 6, 7, 8, 9, 11, 12, and 14 were significantly higher in β cells of individuals with T1D than those with islet autoantibodies compared to the absence of diabetes or non-diabetic donors (Fig. 1 B). Together, the results showed that the expression levels of PARP9, 10, 12, and 14 were increased both in vitro models and donors with T1D. Notably, Parp12 abundance was increased in the phospholipase Pla2g6 knockdown MIN6 cells[ 18 ]. Further, Adprhl2 (ARH3 protein) knockdown causes 52.3% reduction in PARP12 expression by the cytokine cocktail [ 18 ]. ARH3 expression is regulated by Pla2g6 via the release of omega-3 fatty acids [ 18 ]. To test if PARP12 is also regulated by omega-3 fatty acids, we treated MIN6 cells with cytokine cocktail CT1 in the presence of the omega-3 fatty acids, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). The EPA and DHA increased the cytokine-induced Parp12 expression by 54% and 62%, respectively (Fig. 1 C). Based on these results, we hypothesized that PARP12 has a role in cytokine-mediated signaling and β-cell stress and selected it for further characterization. To investigate the regulation of the gene encoding PARP12, we examined our previously published histone H3K27Ac enrichment profiles, RNA-seq and ATAC-seq data of CT2-treated human islets[ 22 ]. We detected a cytokine-induced enrichment of the active histone modification H3K27Ac upstream of the PARP12 promoter region (Fig. 1 D). The higher H3K27Ac levels matched with an increased chromatin accessibility measured by ATAC-seq (Fig. 1 D). As expected, these cytokine-induced chromatin changes correlated with increased transcription activity, measured by RNA-seq data (Fig. 1 D). To determine potential transcription factors involved in regulation of Parp12 gene expression, we studied the sequence composition of the gene locus. We found that the Parp12 gene promoter and distal regulatory elements bear binding motifs for POU2F2, JUNB, NFKB1, IRF1, IRF2, IRF4, IRF8, and IRF9 (Fig. 1 E). Among those, IRF9 binding motif sequence in the promoter region of the gene was apparent, prompting functional analyses. Knockdown of Irf9 in MIN6 cells reduced the cytokine cocktail CT1-mediated induction of Parp12 by 52% (Fig. 1 F), confirming the role of this transcription factor in regulating Parp12 gene expression. Overall, these data identified PARP12 as a cytokine-regulated PARP during insulitis in islets and that its gene expression is regulated by IRF9. Proteomics of PARP12 knockdown MIN6 cells To gain knowledge on PARP12 function, we performed proteomic analysis of cytokine cocktail CT1-treated MIN6 cells with or without Parp12- targeting siRNA. As a control, we observed 62% and 85% knockdown efficiency of both Parp12 mRNA and its encoded protein levels in the CT1-treated group, respectively (Fig. 2 A). Our label-free proteomic analysis identified 4898 proteins, of which 683 proteins were regulated by the cytokine cocktail (Fig. 2 B, ESM Table S6 ). Moreover, 63 of the cytokine-regulated proteins were significantly affected by PARP12 (Fig. 2 C), which were submitted to a network analysis. The network analysis identified a cluster of interacting proteins related to transcription and post-transcription regulation: IRGM1, IRF9, PARP10, ZC3HAV1 and OAS3 (Fig. 2 D). In the cytokine-treated groups, the production of these proteins was partially or completely dependent on PARP12 (Fig. 2 E). Together, the proteomics analysis indicates that PARP12 regulates proteins related to transcription and post-transcription regulation. PARP12 post-transcriptionally regulates insulin 2 To investigate the possibility of PARP12 being a post-transcriptional regulator, we analyzed the proteins that were affected by the knockdown of Parp12 siRNA in cells treated with CT1. We found 94 proteins with a significant increase in cells treated with Parp12 siRNA and CT1 compared to parental cells treated with CT1 (Fig. 3 A). Insulin 2 (INS2) was one of the proteins whose abundance was downregulated by PARP12 (Fig. 3 B). However, its mRNA level was not significantly altered (Fig. 3 C), supporting that PARP12 post-transcriptionally regulates insulin production. We next performed in situ hybridization of Insulin 2 (Ins2) mRNA in MIN6 parental and PARP12 knockdown cells treated with cytokines. The cells were also immunostained for PARP12, which showed a perinuclear localization and increased abundance in cells treated with cytokines (Fig. 3 D). As expected, Parp12 siRNA reduced its protein abundance in both parental and siRNA cells (Fig. 3 D). Under cytokine treatment, Ins2 mRNA distribution moves from the perinuclear region and became diffused in the cytosol (Fig. 3 D). In Parp12 siRNA cells, Ins2 mRNA remained in perinuclear localization, showing that its release into the cytosol is dependent on PARP12 (Fig. 3 D). Overall, these data showed that PARP12 is a post-transcriptional regulator that releases the Ins2 mRNA from the perinuclear region to a diffused distribution in the cytosol. PARP12 is a potential RNA ADP-ribosyltransferase As PARP12 is an ADP-ribosyltransferase, we hypothesized that PARP12 ADP-ribosylates the Ins2 mRNA to downregulate its translation. To investigate this possibility, we examined the PARP12 structures, including its catalytic domain (PDB ID: 6V3W) and an AlphaFold prediction (identifier AF-Q8BZ20-F1) of the remaining protein composed of two WWE (tryptophan-tryptophan-glutamate) domains, and four zinc finger domains (ZnF) (Fig. 4 A). We assembled the Ins2 mRNA onto this predicted structure based on the catalytic site and the ZnF domains of PARP12. These four ZnF domains are predicted to form a groove that could accommodate RNA molecules (Fig. 4 A). Of these four, ZnF motifs 1, 2 and 3 are predicted to bind to the RNA and were therefore used to identify possible Ins2 mRNA regions that might interact with PARP12. The Ins2 mRNA was divided into 27 oligonucleotide sequences of 50 nucleotides with 10 overlapping nucleotides between oligonucleotides. The oligonucleotide with the highest prediction corresponded to a region in intron 2 of insulin 2 mRNA (Fig. 4 B). We then positioned the oligonucleotide with the highest predicted binding affinity onto the PARP12 structure. This prediction analysis showed that the PARP12 structure can accommodate RNA molecules in proximity to its catalytic site (Fig. 4 A), indicating this enzyme may ADP-ribosylate transcripts such as Ins2 . To determine if the Ins2 mRNA gets ADP-ribosylated, we extracted RNAs from MIN6 cells treated or not with CT1 followed by pull-down of the ADP-ribosylated RNAs with AF1521 ADP-ribosylation binding macrodomain and performed qPCR (Fig. 4 C). Ins2 mRNA was amplified in both untreated and CT1-treated cells (Fig. 4 D). As a control, pulldowns with beads alone without AF1521 showed only trace amplification (Fig. 4 D). To test the specificity of AF1521 beads, we pretreated RNA aliquots with SARS-CoV2 macrodomain MAC1 to remove ADP-ribosylation (Fig. 4 C). This treatment led to a 94–99% reduction in Ins2 mRNA amplification (Fig. 4 D), confirming that the AF1521 specifically enriches for ADP-ribosylated RNAs. Together, these data show that PARP12 is a potential RNA ADP-ribosyltransferase and that the Ins2 mRNA is ADP-ribosylated. Landscape of PARP12 ADP-ribosylated RNAs To determine if PARP12 is an RNA ADP-ribosyltransferase and to identify its substrates, we enriched ADP-ribosylated RNAs from PARP12 knockdown cells treated with cytokines CT1 with AF1521 beads and submitted them to RNAseq. By quantifying RNAs enriched in AF1521 vs. control beads, 1381 RNAs were found to be ADP-ribosylated. From these, 150 RNAs were significantly downregulated in PARP12 knockdown cells (Fig. 5 A, ESM table S7 ), therefore are potential PARP12 substrates. A function enrichment analysis of the PARP12 substrates showed top enrichment of genes related to respiratory electron transport, cellular responses to stimuli, metabolism of proteins and RNAs, immune system and cell cycle (Fig. 5 B, ESM table S8 ). PARP12 knockdown led to a 63% reduction in insulin 2 ADP-ribosylation in cells treated with CT1 but no significant changes were observed in untreated cells (Fig. 5 C). These results show that ADP-ribosylation is a major RNA modification and that PARP12 is an RNA ADP-ribosyltransferase that modifies many transcripts, including the insulin 2 mRNA. PARP12 suppresses the translation of specific proteins We next tested if PARP12-mediated ADP-ribosylation of RNAs has a role in regulating protein translation. We treated PARP12 siRNA MIN6 cells with CT1 and pulse-chased them with heavy isotope-labeled lysine for 24 h to label nascent proteins and analyzed by proteomics. The analysis identified 187 proteins whose translation was affected by PARP12 in untreated and CT1-treated cells (Fig. 6 A, ESM table S9 ). A functional-enrichment analysis revealed that 19 pathways categories, with top enriched pathways from Cellular response to stimuli, metabolism of protein and RNA, were influenced by PARP12 (Fig. 6 B, ESM table S10 ). As the proteomics analysis did not yield to detectable insulin peptides, we measured insulin 2 by intact protein mass spectrometry analysis. Insulin 2 translation is repressed by the CT1 treatment, but this repression is alleviated in Parp12 siRNA cells (Fig. 6 C). These results show that PARP12 regulates the translation of many proteins, including repressing insulin 2 production. Discussion Our data showed that PARP12 expression is induced by pro-inflammatory cytokines in human islets and insulin-producing cell lines. PARP12 is also elevated in β cells during T1D development. Mechanistically, cytokines induce PARP12 gene expression by histone acetylation, chromatin opening and activation of the transcription factor IRF9. IRF9 activated by interferon signaling, has an essential role in the expression of interferon-stimulated genes (ISGs) during antiviral responses [ 36 ]. Indeed, PARP12 is an antiviral factor that suppresses Zika virus by targeting its non-structural proteins NS1 and NS3 to proteasomal degradation [ 37 ]. PARP12 also impedes viral replication by targeting viral RNAs in stress granules, protein-RNA aggregates, stalling the translation [ 38 – 40 ]. Here, we show that PARP12 blocks insulin 2 translation by ADP-ribosylating and translocating its transcript from the perinuclear region to the cytosol under cytokine influence. PARP12 overexpression causes global translational repression by associating with cellular translational machinery and potentially targeting mRNA to stress granules [ 39 ]. E. coli has a similar machinery that ADP-ribosylates and halts the translation of phage transcripts as part of an antiviral response [ 9 ]. Conversely, coronavirus ADP-ribosylhydrolase MAC1 counteracts the protective effect of PARP12[ 41 ]. At least under in vitro conditions, we show that MAC1 efficiently cleaves PARP12-mediated mRNA ADP-ribosylation, suggesting a counteracting mechanism for ADP-ribosylation-mediated translation halting. Our data show that a large portion of the cell transcriptome is ADP-ribosylated, but only a fraction is mediated by PARP12. Other mammalian PARPs ADP-ribosylate RNAs, but specific substrates remained unknown [ 8 , 42 ]. Therefore, it is possible that the ADP-ribosylation of mRNAs is a much broader translation regulation mechanism. Besides targeting mRNAs, ADP-ribosylation represses translation by PARP16-mediated ADP-ribosylation of ribosomal protein subunits [ 43 ]. Our structural analysis predicts that PARP12 has four ZnF motifs that form a groove that accommodates the RNA chain. ZnF motifs can bind to DNA, RNA and ADP-ribosylation [ 44 ]. Similar ZnF motifs are also present in PARP13, also known as zinc finger antiviral protein (ZAP). Despite not displaying ADP-ribosylation activity, PARP13 binds to GC motifs in viral RNAs and target them to degradation [ 37 ]. Indeed, the insulin 2 mRNA region bearing three GC motifs was predicted to have the highest binding to PARP12, suggesting a possible similar binding specificity compared to PARP13. The structural analysis also identified two ADP-ribosylation-binding WWE domains, which are present in other PARPs and some E3 ubiquitin ligases [ 37 ]. PARP12 WWE domains help its recruitment to stress granules via poly-ADP-ribose signaling catalyzed by PARP1 [ 38 ]. Combined these structural insights align with our observation that PARP12 acts as a translation inhibitor under pro-inflammatory conditions, potentially binding RNA with its ZnF motifs and utilizing WWE domains for cellular localization and function. The reduction in insulin production by PARP12 might aggravate insulin availability during T1D development. At this stage, pro-inflammatory cytokines such as interferons, which induce PARP12 expression, are major regulators of the insulitis process [ 45 ]. Our findings might also have potential implications in aggravation of infection in type 2 diabetic patients, who are already at higher risk for complications. Notably, SARS-CoV-2 infections are associated with a robust production of pro-inflammatory cytokines and insulin depletion [ 46 , 47 ]. It is possible that the insulin depletion is caused, at least in part, by ADP-ribosylation of its mRNA and consequent block in translation. However, the involvement of PARP12 in this process still needs to be investigated. In conclusion, we identified PARP12 as a cytokine-induced RNA ADP-ribosyltransferase that halts protein translation. Specifically, PARP12 targets insulin mRNA and ADP-ribosylating it and thereby reducing insulin production. This mechanistic insight underpins a potential physiological impact during T1D development, where reduced insulin availability can exacerbate the disease progression. Moreover, our findings offer a broader perspective on the role of PARP12 in inflammatory and antiviral responses, suggesting its involvement in the complex interplay between cytokine signaling and translational control. Declarations Data availability The proteomics data were deposited into the MassIVE repository, a member of the ProteomeXchange, under accession numbers/user MSV000097349 and MSV000097350 (accession information below). The ADP-ribosylated RNA sequencing data is available at NCBI Geo under accession number GSE292484. Title: Insulin post-transcriptional regulation via PARP12-mediated ADP-ribosylation (global label-free) MassIVE accession: MSV000097349 https://massive.ucsd.edu/ProteoSAFe/dataset.jsp?task=93584fb8b1df4ad8b5f998c7fafb4ab0 For now, manuscript reviewers can access the data via FTP using: Server: massive.ucsd.edu User: MSV000097349 Password: Insulin6433 Title: Insulin post-transcriptional regulation via PARP12-mediated ADP-ribosylation (pulse-chase SILAC) MassIVE accession: MSV000097350 https://massive.ucsd.edu/ProteoSAFe/dataset.jsp?task=778b2027fd574d039b0ecd293259e851 Manuscript reviewers can access the data via FTP using: Server: massive.ucsd.edu User: MSV000097350 Password: Insulin6432 Competing interests The authors declare that they have no competing interests. Acknowledgments The authors would like to thank the technical assistance and support of Mr. Armando Puente (University of Chicago), Dr. Mikhail Belov (Spectroglyph, LLC), Mr. Gordon Anderson and Mr. Chris Anderson (GAA Custom Electronics, LLC), Drs. Kyle Fort, Maria Reinhardt-Szyba, and Alexander Makarov (Thermo Fisher Scientific). Part of the work was performed in the Environmental Molecular Sciences Laboratory, a U.S. DOE national scientific user facility at Pacific Northwest National Laboratory (PNNL) in Richland, WA. Battelle operates PNNL for the US Department of Energy under contract DE-AC05-76RLO01830. Author Contributions Statement Conceptualization: SS, ESN. Investigation, data curation, formal analysis and validation: all authors. Visualization: SS, FL, YY, HK, MR-R, XY. Methodology: SS, FL, YY, KJZ, LMM, HK, AKL, RGM, ESN. Project administration, supervision and resources: DLE, AKLL, LP, LS, TOM, RGM, ESN. Funding acquisition: DLE, MR-R, LP, LS, TOM, RGM, ESN. Writing – original draft: SS, FL, YY, KJZ, HK, MR-R, XY, JRE, HM, ESN. Writing – review & editing: all authors. Funding This work was supported by the Catalyst Award from the Human Islet Research Network (HIRN) (to E.S.N) (via U24 DK104162) and by National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) grants R01 DK138335 (to ESN, and TOM), U01 DK127505 (to LS and ESN), U01 DK127786 (to RGM, DLE, and TOM), R01 DK060581 (to RGM), R01 DK105588 (to RGM), and P30 DK020595 (to RGM). The development of ADP-ribosylated RNA sequencing assay was conducted under the Laboratory Directed Research and Development Program at PNNL (to ESN). This work was also supported by Spanish Ministry of Science and Innovation PID2023-151556OB-I00, CNS2024-154742, “la Caixa” Foundation, LCF-PR-HR24-00150 (to LP) and the 2024 EFSD/Lilly Young Investigator Research Award (MR-R). References Livingstone SJ, Levin D, Looker HC, Lindsay RS, Wild SH, Joss N, Leese G, Leslie P, McCrimmon RJ, Metcalfe W, et al: Estimated life expectancy in a Scottish cohort with type 1 diabetes, 2008-2010. JAMA 2015, 313: 37-44. Gregory GA, Robinson TIG, Linklater SE, Wang F, Colagiuri S, de Beaufort C, Donaghue KC, International Diabetes Federation Diabetes Atlas Type 1 Diabetes in Adults Special Interest G, Magliano DJ, Maniam J, et al: Global incidence, prevalence, and mortality of type 1 diabetes in 2021 with projection to 2040: a modelling study. Lancet Diabetes Endocrinol 2022, 10: 741-760. Eizirik DL, Pasquali L, Cnop M: Pancreatic beta-cells in type 1 and type 2 diabetes mellitus: different pathways to failure. Nat Rev Endocrinol 2020, 16: 349-362. DiMeglio LA, Evans-Molina C, Oram RA: Type 1 diabetes. Lancet 2018, 391: 2449-2462. Fehr AR, Singh SA, Kerr CM, Mukai S, Higashi H, Aikawa M: The impact of PARPs and ADP-ribosylation on inflammation and host-pathogen interactions. Genes Dev 2020, 34: 341-359. Zarkovic G, Belousova EA, Talhaoui I, Saint-Pierre C, Kutuzov MM, Matkarimov BT, Biard D, Gasparutto D, Lavrik OI, Ishchenko AA: Characterization of DNA ADP-ribosyltransferase activities of PARP2 and PARP3: new insights into DNA ADP-ribosylation. Nucleic Acids Res 2018, 46: 2417-2431. Suskiewicz MJ, Prokhorova E, Rack JGM, Ahel I: ADP-ribosylation from molecular mechanisms to therapeutic implications. Cell 2023, 186: 4475-4495. Munnur D, Bartlett E, Mikolcevic P, Kirby IT, Rack JGM, Mikoc A, Cohen MS, Ahel I: Reversible ADP-ribosylation of RNA. Nucleic Acids Res 2019, 47: 5658-5669. Vassallo CN, Doering CR, Laub MT: Anti-viral defence by an mRNA ADP-ribosyltransferase that blocks translation. Nature 2024, 636: 190-197. Mashimo M, Kato J, Moss J: Structure and function of the ARH family of ADP-ribosyl-acceptor hydrolases. DNA Repair (Amst) 2014, 23: 88-94. Nie Y, Nirujogi TS, Ranjan R, Reader BF, Chung S, Ballinger MN, Englert JA, Christman JW, Karpurapu M: PolyADP-Ribosylation of NFATc3 and NF-kappaB Transcription Factors Modulate Macrophage Inflammatory Gene Expression in LPS-Induced Acute Lung Injury. J Innate Immun 2021, 13: 83-93. Hassa PO, Hottiger MO: The functional role of poly(ADP-ribose)polymerase 1 as novel coactivator of NF-kappaB in inflammatory disorders. Cell Mol Life Sci 2002, 59: 1534-1553. Iwata H, Goettsch C, Sharma A, Ricchiuto P, Goh WW, Halu A, Yamada I, Yoshida H, Hara T, Wei M, et al: PARP9 and PARP14 cross-regulate macrophage activation via STAT1 ADP-ribosylation. Nat Commun 2016, 7: 12849. Andreone T, Meares GP, Hughes KJ, Hansen PA, Corbett JA: Cytokine-mediated beta-cell damage in PARP-1-deficient islets. Am J Physiol Endocrinol Metab 2012, 303: E172-179. Burkart V, Wang ZQ, Radons J, Heller B, Herceg Z, Stingl L, Wagner EF, Kolb H: Mice lacking the poly(ADP-ribose) polymerase gene are resistant to pancreatic beta-cell destruction and diabetes development induced by streptozocin. Nat Med 1999, 5: 314-319. Masutani M, Suzuki H, Kamada N, Watanabe M, Ueda O, Nozaki T, Jishage K, Watanabe T, Sugimoto T, Nakagama H, et al: Poly(ADP-ribose) polymerase gene disruption conferred mice resistant to streptozotocin-induced diabetes. Proc Natl Acad Sci U S A 1999, 96: 2301-2304. Abplanalp J, Leutert M, Frugier E, Nowak K, Feurer R, Kato J, Kistemaker HVA, Filippov DV, Moss J, Caflisch A, Hottiger MO: Proteomic analyses identify ARH3 as a serine mono-ADP-ribosylhydrolase. Nat Commun 2017, 8: 2055. Sarkar S, Deiter C, Kyle JE, Guney MA, Sarbaugh D, Yin R, Li X, Cui Y, Ramos-Rodriguez M, Nicora CD, et al: Regulation of beta-cell death by ADP-ribosylhydrolase ARH3 via lipid signaling in insulitis. Cell Commun Signal 2024, 22: 141. You Y, Sarkar S, Deiter C, Elliott EC, Nicora CD, Mirmira RG, Sussel L, Nakayasu ES: Reduction of Chemokine CXCL9 Expression by Omega-3 Fatty Acids via ADP-Ribosylhydrolase ARH3 in MIN6 Insulin-Producing Cells. Proteomics 2025, 25: e202400053. Colli ML, Ramos-Rodriguez M, Nakayasu ES, Alvelos MI, Lopes M, Hill JLE, Turatsinze JV, Coomans de Brachene A, Russell MA, Raurell-Vila H, et al: An integrated multi-omics approach identifies the landscape of interferon-alpha-mediated responses of human pancreatic beta cells. Nat Commun 2020, 11: 2584. Nakayasu ES, Syed F, Tersey SA, Gritsenko MA, Mitchell HD, Chan CY, Dirice E, Turatsinze JV, Cui Y, Kulkarni RN, et al: Comprehensive Proteomics Analysis of Stressed Human Islets Identifies GDF15 as a Target for Type 1 Diabetes Intervention. Cell Metab 2020, 31: 363-374 e366. Ramos-Rodriguez M, Raurell-Vila H, Colli ML, Alvelos MI, Subirana-Granes M, Juan-Mateu J, Norris R, Turatsinze JV, Nakayasu ES, Webb-Robertson BM, et al: The impact of proinflammatory cytokines on the beta-cell regulatory landscape provides insights into the genetics of type 1 diabetes. Nat Genet 2019, 51: 1588-1595. Elliott EC, Sakar S, Bramer L, Burnet M, Kim Y-M, Yi X, Estevao IL, Rewers M, Cambronne XA, Vehik K, et al: Free fatty acids accelerate β-cell death in type 1 diabetes. medRxiv 2024 : 2024.2009.2016.24313433. McLeay RC, Bailey TL: Motif Enrichment Analysis: a unified framework and an evaluation on ChIP data. BMC Bioinformatics 2010, 11: 165. Tyanova S, Temu T, Cox J: The MaxQuant computational platform for mass spectrometry-based shotgun proteomics. Nat Protoc 2016, 11: 2301-2319. Krey JF, Wilmarth PA, Shin JB, Klimek J, Sherman NE, Jeffery ED, Choi D, David LL, Barr-Gillespie PG: Accurate label-free protein quantitation with high- and low-resolution mass spectrometers. J Proteome Res 2014, 13: 1034-1044. Schwanhausser B, Busse D, Li N, Dittmar G, Schuchhardt J, Wolf J, Chen W, Selbach M: Global quantification of mammalian gene expression control. Nature 2011, 473: 337-342. Snel B, Lehmann G, Bork P, Huynen MA: STRING: a web-server to retrieve and display the repeatedly occurring neighbourhood of a gene. Nucleic Acids Res 2000, 28: 3442-3444. Li F, Hu D, Dieter C, Ansong C, Sussel L, Orr G: Single Molecule-Based fliFISH Validates Radial and Heterogeneous Gene Expression Patterns in Pancreatic Islet beta-Cells. Diabetes 2021, 70: 1117-1122. Yilmaz LS, Parnerkar S, Noguera DR: mathFISH, a web tool that uses thermodynamics-based mathematical models for in silico evaluation of oligonucleotide probes for fluorescence in situ hybridization. Appl Environ Microbiol 2011, 77: 1118-1122. Livak KJ, Schmittgen TD: Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 2001, 25: 402-408. Dobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR: STAR: ultrafast universal RNA-seq aligner. Bioinformatics 2013, 29: 15-21. Zhang Y, Parmigiani G, Johnson WE: ComBat-seq: batch effect adjustment for RNA-seq count data. NAR Genom Bioinform 2020, 2: lqaa078. Love MI, Huber W, Anders S: Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol 2014, 15: 550. Zemaitis KJ, Veličković D, Kew W, Fort KL, Reinhardt-Szyba M, Pamreddy A, Ding Y, Kaushik D, Sharma K, Makarov AA, et al: Enhanced Spatial Mapping of Histone Proteoforms in Human Kidney Through MALDI-MSI by High-Field UHMR-Orbitrap Detection. Analytical Chemistry 2022, 94: 12604-12613. Paul A, Tang TH, Ng SK: Interferon Regulatory Factor 9 Structure and Regulation. Front Immunol 2018, 9: 1831. Li L, Zhao H, Liu P, Li C, Quanquin N, Ji X, Sun N, Du P, Qin CF, Lu N, Cheng G: PARP12 suppresses Zika virus infection through PARP-dependent degradation of NS1 and NS3 viral proteins. Sci Signal 2018, 11 . Catara G, Grimaldi G, Schembri L, Spano D, Turacchio G, Lo Monte M, Beccari AR, Valente C, Corda D: PARP1-produced poly-ADP-ribose causes the PARP12 translocation to stress granules and impairment of Golgi complex functions. Sci Rep 2017, 7: 14035. Welsby I, Hutin D, Gueydan C, Kruys V, Rongvaux A, Leo O: PARP12, an interferon-stimulated gene involved in the control of protein translation and inflammation. J Biol Chem 2014, 289: 26642-26657. Lee S, Lee YS, Choi Y, Son A, Park Y, Lee KM, Kim J, Kim JS, Kim VN: The SARS-CoV-2 RNA interactome. Mol Cell 2021, 81: 2838-2850 e2836. Kerr CM, Parthasarathy S, Schwarting N, O'Connor JJ, Pfannenstiel JJ, Giri E, More S, Orozco RC, Fehr AR: PARP12 is required to repress the replication of a Mac1 mutant coronavirus in a cell- and tissue-specific manner. J Virol 2023, 97: e0088523. Weixler L, Feijs KLH, Zaja R: ADP-ribosylation of RNA in mammalian cells is mediated by TRPT1 and multiple PARPs. Nucleic Acids Res 2022, 50: 9426-9441. Challa S, Khulpateea BR, Nandu T, Camacho CV, Ryu KW, Chen H, Peng Y, Lea JS, Kraus WL: Ribosome ADP-ribosylation inhibits translation and maintains proteostasis in cancers. Cell 2021, 184: 4531-4546 e4526. Cassandri M, Smirnov A, Novelli F, Pitolli C, Agostini M, Malewicz M, Melino G, Raschella G: Zinc-finger proteins in health and disease. Cell Death Discov 2017, 3: 17071. Coomans de Brachene A, Alvelos MI, Szymczak F, Zimath PL, Castela A, Marmontel de Souza B, Roca Rivada A, Marin-Canas S, Yi X, Op de Beeck A, et al: Interferons are key cytokines acting on pancreatic islets in type 1 diabetes. Diabetologia 2024, 67: 908-927. Reiterer M, Rajan M, Gomez-Banoy N, Lau JD, Gomez-Escobar LG, Ma L, Gilani A, Alvarez-Mulett S, Sholle ET, Chandar V, et al: Hyperglycemia in acute COVID-19 is characterized by insulin resistance and adipose tissue infectivity by SARS-CoV-2. Cell Metab 2021, 33: 2174-2188 e2175. Lucas C, Wong P, Klein J, Castro TBR, Silva J, Sundaram M, Ellingson MK, Mao T, Oh JE, Israelow B, et al: Longitudinal analyses reveal immunological misfiring in severe COVID-19. Nature 2020, 584: 463-469. Additional Declarations No competing interests reported. Supplementary Files ESMfile1.xlsx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6473437","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":457716395,"identity":"e549554f-2c00-41ac-83ca-751d0d239d0d","order_by":0,"name":"Soumyadeep Sarkar","email":"","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Soumyadeep","middleName":"","lastName":"Sarkar","suffix":""},{"id":457716396,"identity":"59ab648e-f10a-41fb-82ad-841fafce82aa","order_by":1,"name":"Fangjia Li","email":"","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Fangjia","middleName":"","lastName":"Li","suffix":""},{"id":457716397,"identity":"45e6739a-dafa-4353-b68f-fdce9a35b9a8","order_by":2,"name":"Youngki You","email":"","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Youngki","middleName":"","lastName":"You","suffix":""},{"id":457716398,"identity":"8d63846a-2365-409a-a044-1f540adae3e1","order_by":3,"name":"Emily C. Elliott","email":"","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Emily","middleName":"C.","lastName":"Elliott","suffix":""},{"id":457716399,"identity":"758c8f46-370d-43ce-8d9e-2f21cd5c2729","order_by":4,"name":"Kevin J. Zemaitis","email":"","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Kevin","middleName":"J.","lastName":"Zemaitis","suffix":""},{"id":457716400,"identity":"5a4563e1-9a27-4a5e-927f-22c4d646e5fb","order_by":5,"name":"Hyeyoon Kim","email":"","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Hyeyoon","middleName":"","lastName":"Kim","suffix":""},{"id":457716401,"identity":"7792e3da-a25a-464e-98c0-4a18b1930944","order_by":6,"name":"Lye Meng Markillie","email":"","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Lye","middleName":"Meng","lastName":"Markillie","suffix":""},{"id":457716402,"identity":"ff91c810-3da8-4330-bd0a-1b549c278771","order_by":7,"name":"Jeremiah Traeger","email":"","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Jeremiah","middleName":"","lastName":"Traeger","suffix":""},{"id":457716403,"identity":"630dc413-bc74-4a86-bada-f94edadae04b","order_by":8,"name":"Samantha M. Powell","email":"","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Samantha","middleName":"M.","lastName":"Powell","suffix":""},{"id":457716404,"identity":"c1a91451-f7e8-4856-86f3-c22785e3caae","order_by":9,"name":"Mireia Ramos-Rodriguez","email":"","orcid":"","institution":"University Pompeu Fabra","correspondingAuthor":false,"prefix":"","firstName":"Mireia","middleName":"","lastName":"Ramos-Rodriguez","suffix":""},{"id":457716405,"identity":"b32bdbf1-3c0f-4d24-a8ee-349dabf39b0f","order_by":10,"name":"Xiaoyan Yi","email":"","orcid":"","institution":"Université Libre de Bruxelles (ULB)","correspondingAuthor":false,"prefix":"","firstName":"Xiaoyan","middleName":"","lastName":"Yi","suffix":""},{"id":457716406,"identity":"5924d97f-9643-451b-8f5a-21c0860f9411","order_by":11,"name":"Jacob R. Enriquez","email":"","orcid":"","institution":"The University of Chicago","correspondingAuthor":false,"prefix":"","firstName":"Jacob","middleName":"R.","lastName":"Enriquez","suffix":""},{"id":457716407,"identity":"7111fe51-b43e-44d1-b66d-4f5b5560a6f2","order_by":12,"name":"Marina Gritsenko","email":"","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Marina","middleName":"","lastName":"Gritsenko","suffix":""},{"id":457716408,"identity":"54807e36-00bb-44b0-9181-366429ee5ff3","order_by":13,"name":"Hugh Mitchell","email":"","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Hugh","middleName":"","lastName":"Mitchell","suffix":""},{"id":457716409,"identity":"6a5fa255-3b42-4afd-b152-139af631f86f","order_by":14,"name":"Decio L. Eizirik","email":"","orcid":"","institution":"Université Libre de Bruxelles (ULB)","correspondingAuthor":false,"prefix":"","firstName":"Decio","middleName":"L.","lastName":"Eizirik","suffix":""},{"id":457716410,"identity":"4236dad7-52c4-4115-a086-7db26ff78c55","order_by":15,"name":"Anthony K. L. Leung","email":"","orcid":"","institution":"Bloomberg School of Public Health, Johns Hopkins University","correspondingAuthor":false,"prefix":"","firstName":"Anthony","middleName":"K. L.","lastName":"Leung","suffix":""},{"id":457716411,"identity":"f8209ef1-8012-4127-94ad-84935a8d3fee","order_by":16,"name":"Lorenzo Pasquali","email":"","orcid":"","institution":"University Pompeu Fabra","correspondingAuthor":false,"prefix":"","firstName":"Lorenzo","middleName":"","lastName":"Pasquali","suffix":""},{"id":457716412,"identity":"40b4f6ba-bb4b-4c5d-be21-95f31f822774","order_by":17,"name":"Lori Sussel","email":"","orcid":"","institution":"University of Colorado Anschutz Medical Center","correspondingAuthor":false,"prefix":"","firstName":"Lori","middleName":"","lastName":"Sussel","suffix":""},{"id":457716413,"identity":"1f00e4bb-a496-4dc1-9919-a4e29cf66a73","order_by":18,"name":"Thomas O. Metz","email":"","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"O.","lastName":"Metz","suffix":""},{"id":457716414,"identity":"942e695f-52ab-4b1d-acb6-960598935ebd","order_by":19,"name":"Raghavendra G. Mirmira","email":"","orcid":"","institution":"The University of Chicago","correspondingAuthor":false,"prefix":"","firstName":"Raghavendra","middleName":"G.","lastName":"Mirmira","suffix":""},{"id":457716415,"identity":"e1a145e4-8ef3-45a6-b832-b155d6d34322","order_by":20,"name":"Ernesto S. Nakayasu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABD0lEQVRIiWNgGAWjYHADNgjFDyIeEFSdgKRFsgEmQLQWgwMEtPBLnz348OcPGwZz9mOJn3lq6hI33+499iCBoVYOrBcLkOzLSzbmSUhjsOxJOyzNc+xw4rY759INEhiOG+PSYnCGx0yaIeEw0D3pDdI8bAcSt93IMZNIYDiWOLMBuxb7MzzmP3+AtJx/3vyb5x/QYTMIaDHg4TFj4AFpuZF2TJq3jTlxgwRYS01iPw7vS5zhMZbmSUvjMbjxLM1ybt9h4xl3zgC1GBww5scVYj08hh9/2NjIGZxPM77x5ludbP/sHjOJDxV1cmw4tMAAD4hgApMSYAcfJqABChh/wLUw1BGnZRSMglEwCkYCAACerFkN9VQDUAAAAABJRU5ErkJggg==","orcid":"","institution":"Pacific Northwest National Laboratory","correspondingAuthor":true,"prefix":"","firstName":"Ernesto","middleName":"S.","lastName":"Nakayasu","suffix":""}],"badges":[],"createdAt":"2025-04-17 16:23:20","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6473437/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6473437/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":83220452,"identity":"adbf04ef-aafb-483e-bc90-1bec9668d33c","added_by":"auto","created_at":"2025-05-21 10:16:18","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":352690,"visible":true,"origin":"","legend":"\u003cp\u003eExpression analysis of PARP12.\u003cstrong\u003e \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eQuantification of PARP protein abundances in proteomics re-analysis of human islets[21], EndoC-βH1 cells[20, 22] and MIN6 cells[18] treated with combinations of pro-inflammatory cytokines: (CT1: IFN-γ, IL-1β and TNFα; CT2: IFN-γ and IL-1β; CT3: INFα). Asterisks represent significant expression of the respective protein compared to no cytokine-treated control. (\u003cstrong\u003eB\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eQuantification of PARP gene expression in β cells in single-cell RNA-seq data downloaded from the Human Pancreas Analysis Program (HPAP) data portal (\u003ca href=\"https://hpap.pmacs.upenn.edu/\"\u003ehttps://hpap.pmacs.upenn.edu/\u003c/a\u003e). The data were collected from pancreata from non-diabetic donors or donors at different stages of type 1 diabetes development: AAB1+ - individuals seropositive for 1 islet autoantibody, AAB2+ - individuals seropositive for multiple islet autoantibodies, beta – β cells, T1D – type 1 diabetes. We analyzed 6,423 β cells from 15 ND, 5,751 β cells from 9 AAB1+, 4,977 β cells from 2 AAB2+, and 2,013 β cells from 11 T1D donors. Asterisks represent significant gene expression compared to non-diabetic control. (\u003cstrong\u003eC\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eEffect of omega-3 fatty acids, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA), on the expression of \u003cem\u003eParp12 \u003c/em\u003egene induced by CT1 in MIN6 cells (n=3-4, 2way ANOVA, Holm-Šídák's multiple comparisons test). (\u003cstrong\u003eD\u003c/strong\u003e) View of the PARP12 locus regulation in human islets treated with CT2, depicted by histone acetylation (H3K27Ac) ChIP-seq, ATAC-seq and RNA-seq data[22]. (\u003cstrong\u003eE\u003c/strong\u003e) Genomic motif analysis of putative transcription factors that regulate PARP12 expression. (\u003cstrong\u003eF\u003c/strong\u003e) Regulation of \u003cem\u003eParp12 \u003c/em\u003egene expression by the transcription factor IRF9.\u003cstrong\u003e \u003c/strong\u003e\u003cem\u003eParp12\u003c/em\u003e mRNA level in MIN6 cells treated with CT1 in combination with NT (Non-target), \u003cem\u003eIrf9 \u003c/em\u003esiRNA (si\u003cem\u003eIrf9\u003c/em\u003e) (n=4, 2way ANOVA, Holm-Šídák's multiple comparisons test). Outlier data points were identified and removed using Grubb’s test for single outliers with p\u0026lt;0.3.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6473437/v1/97458f727cbab2d26676c3af.png"},{"id":83221542,"identity":"57585057-574f-4893-ab55-9200825ac503","added_by":"auto","created_at":"2025-05-21 10:24:18","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":444513,"visible":true,"origin":"","legend":"\u003cp\u003eProteomics analysis of MIN6 cells exposed to Parp12 siRNA (siParp12) and treated with the pro-inflammatory cytokine cocktail CT1 (IFN-γ, IL-1β and TNFα). (\u003cstrong\u003eA\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003ePARP12 expression analysis\u003cstrong\u003e \u003c/strong\u003ein \u003cem\u003eParp12\u003c/em\u003e siRNA MIN6 cells treated with CT1. Samples were measured by qPCR (Top panel) and proteomics analyses (Bottom panel) (n=4, 2way ANOVA, Holm-Šídák's multiple comparisons test). (\u003cstrong\u003eB\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eHeatmap representing changes in the global\u003cstrong\u003e \u003c/strong\u003eproteome of \u003cem\u003eParp12\u003c/em\u003e siRNA MIN6 treated with CT1 (n=4, Student’s \u003cem\u003et\u003c/em\u003e-test). (\u003cstrong\u003eC\u003c/strong\u003e) Subset of the heatmap of panel B that proteins are regulated by both CT1 and \u003cem\u003eParp12\u003c/em\u003e siRNA. (\u003cstrong\u003eD\u003c/strong\u003e) A network by the String database of proteins found in panel C. (\u003cstrong\u003eE\u003c/strong\u003e) Abundance profile of proteins in the network of panel D (n=4, 2way ANOVA, Holm-Šídák's multiple comparisons test). Outlier data points were identified and removed using Grubb’s test for single outliers with p\u0026lt;0.3.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6473437/v1/3c911fd4fc6e989894bf941d.png"},{"id":83221543,"identity":"a1f910e0-eb1f-496c-866b-d397e0ee8ff3","added_by":"auto","created_at":"2025-05-21 10:24:18","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":275278,"visible":true,"origin":"","legend":"\u003cp\u003eRegulation of insulin 2 by PARP12.\u003cstrong\u003e \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e) Abundance of 94 proteins regulated by PARP12 in MIN6 cells treated with the pro-inflammatory cytokine cocktail CT1 (IFN-γ, IL-1β and TNFα) and measured in the proteomics analysis (Mann-Whitney test). (\u003cstrong\u003eB\u003c/strong\u003e) Abundance profile of insulin 2 protein in the proteomics analysis (n=3-4, 2way ANOVA, Holm-Šídák's multiple comparisons test) (\u003cstrong\u003eC\u003c/strong\u003e) qPCR quantification of \u003cem\u003eIns2 \u003c/em\u003emRNA in \u003cem\u003eParp12\u003c/em\u003e siRNA MIN6 cells treated with CT1 (n=4, 2way ANOVA, Holm-Šídák's multiple comparisons test). (\u003cstrong\u003eD\u003c/strong\u003e) Fluorescence imaging analysis of \u003cem\u003eIns2\u003c/em\u003emRNA (red) in \u003cem\u003eParp12\u003c/em\u003e siRNA MIN6 cells treated with CT1. In addition to Ins2 mRNA, PARP12 protein was stained by immunofluorescence (green) and nuclei with DAPI (cyan).\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6473437/v1/b578ea082c1f6e0cd4486c23.png"},{"id":83220454,"identity":"84f30a07-ed59-4438-a06a-e23dc7f0ba96","added_by":"auto","created_at":"2025-05-21 10:16:18","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":391944,"visible":true,"origin":"","legend":"\u003cp\u003eRT-PCR amplified \u003cem\u003eIns2\u003c/em\u003e intron2 pre-mRNA from MIN6 cells treated with -/+ si\u003cem\u003eParp12 \u003c/em\u003ein CT1 treated group. (\u003cstrong\u003eA\u003c/strong\u003e) Predictive complex structure of PARP12 and \u003cem\u003eIns2\u003c/em\u003e mRNA. Four zinc finger motifs of PARP12 are highlighted by Corey–Pauling–Koltun representation. The yellow section of mRNA represents the potential ADPribosylation site as near the catalytic domain of PARP12. (\u003cstrong\u003eB\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003e\u0026nbsp;The prediction of RNA-protein interactions was conducted using RPISeq. The random forest values for 27 segments of the \u003cem\u003eIns2\u003c/em\u003epre-mRNA were calculated against three different combinations of zinc finger domains of PARP12. Predictions with probabilities greater than 0.5 were considered positive. (\u003cstrong\u003eC\u003c/strong\u003e) Scheme representing ADP-ribosylated RNA pulldown experimental design. (\u003cstrong\u003eD\u003c/strong\u003e) ADP-ribosylated\u003cem\u003e \u003c/em\u003eRNA\u003cem\u003e \u003c/em\u003ewith/without MAC1 treatment were pulldown using AF1521 (AF) or naked-control beads (CB).\u003cem\u003e \u003c/em\u003eAmplification was calculated against 0.3% Input and represented as relative enrichment to AF.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6473437/v1/2db1af4a4cb416d280f8d322.png"},{"id":83221546,"identity":"ec2b7b88-e696-48b2-b21a-9ecc18644654","added_by":"auto","created_at":"2025-05-21 10:24:19","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":393771,"visible":true,"origin":"","legend":"\u003cp\u003ePARP12 RNA substrates.\u003cstrong\u003e \u003c/strong\u003e(\u003cstrong\u003eA\u003c/strong\u003e) Heat map representing significant ADP-ribosylated epi-transcriptome dependent on PARP12. (\u003cstrong\u003eB\u003c/strong\u003e)\u003cstrong\u003e \u003c/strong\u003eBubble plot represents significant pathway categories with multiple sub-pathway terms enriched by the identified PARP12 RNA substrates from A. Pathway enrichment was done using Ingenuity Pathway Analysis software and the Reactome database as a reference. (\u003cstrong\u003eC\u003c/strong\u003e) Foldchange of ADP-ribosylated \u003cem\u003eIns2\u003c/em\u003e transcript compared to NoCT1-NT group. (n=3, 2way ANOVA, Holm-Šídák's multiple comparisons test).\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-6473437/v1/5e9524b2d3fb7a22ff30b9a3.png"},{"id":83220457,"identity":"592876b0-1407-441c-9d9e-5f5df94dac3e","added_by":"auto","created_at":"2025-05-21 10:16:18","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":443376,"visible":true,"origin":"","legend":"\u003cp\u003eProtein translational effect of PARP12. (\u003cstrong\u003eA\u003c/strong\u003e) heatmap showing significant change in abundance of newly synthesized proteins due to PARP12 knockdown identified by bottom-up proteomics post-heavy lysine pulse-chase experiment in MIN6 cells with -/+ si\u003cem\u003eParp12 \u003c/em\u003etreated with CT1. (\u003cstrong\u003eB\u003c/strong\u003e) Bubble plot represents significant pathway categories enriched by PARP12-dependent proteins identified by post-heavy lysine pulse-chase experiment from A. Pathway enrichment was done by Ingenuity Pathway Analysis software, using the Reactome database as reference. (\u003cstrong\u003eC\u003c/strong\u003e) Abundance profile of nascent insulin 2 (INS2) protein.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-6473437/v1/7b21099963d925de08f54e5c.png"},{"id":104401934,"identity":"f6e30b4c-34da-4051-986f-1243ce4e0e75","added_by":"auto","created_at":"2026-03-11 12:13:56","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":5092896,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6473437/v1/1c0ce020-77f4-446a-8884-a2afdb472c22.pdf"},{"id":83220474,"identity":"2a3a298e-d4d7-41d9-be5d-873f955c2a43","added_by":"auto","created_at":"2025-05-21 10:16:19","extension":"xlsx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":9809744,"visible":true,"origin":"","legend":"","description":"","filename":"ESMfile1.xlsx","url":"https://assets-eu.researchsquare.com/files/rs-6473437/v1/d284a6169eab549fceae1cf4.xlsx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Insulin post-transcriptional regulation via PARP12-mediated ADP- ribosylation","fulltext":[{"header":"Introduction","content":"\u003cp\u003eType 1 diabetes (T1D) is a devastating disease that reduces life expectancy by 11 years for men and 13 years for women [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. The disease currently affects approximately 8.4\u0026nbsp;million people worldwide [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Treatment is lifelong and based on controlling the blood glucose levels with diet, exercise, and insulin administration. Despite advances in blood glucose control, the disease leads to complications such as cardiovascular and renal diseases, which account for the reduction in life expectancy. Mechanistically, T1D is an autoimmune disease with characteristic insulitis pathology [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e] marked by infiltration of immune cells and apoptosis of insulin-producing pancreatic β cells mediated by pro-inflammatory cytokines and chemokines [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. However, the highly complex cellular signaling transduced by the cytokines and chemokines is only partially characterized.\u003c/p\u003e \u003cp\u003eADP-ribosylation plays a central role in regulating DNA repair, inflammatory signaling and anti-viral response [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e], but its role in insulitis is not well understood. This modification is characterized by the addition of adenosine diphosphate (ADP)-ribose units to proteins and nucleic acid by ADP-ribosyltransferases (PARPs or ADRTs) with nicotinamide-adenosine dinucleotide (NAD) as a donor [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. ADP-ribosylation can also modify RNAs[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e], but has only been characterized in bacteria, where it plays a role in antiviral response[\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. ADP-ribosylation occurs as single units (mono-ADP-ribosylation or MARylation) or chains (poly-ADP-ribosylation or PARylation) [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. ADP-ribosylation is reversible and can be removed by hydrolases such as poly(ADP-ribose)glycohydrolase (PARG) and ADP-ribosylhydrolases [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. In DNA repair, ADP-ribosylation labels damaged chromatin regions, recruiting the repair machinery. In immune cells, ADP-ribosylation enhances the activity of key transcription factors in inflammation, such as STATs, NF-κB, and NFATc3 [\u003cspan additionalcitationids=\"CR12\" citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. Pro-inflammatory cytokines, such as IFN-γ, induce the expression of PARP9-14 which in turn regulate antiviral responses by targeting viral RNAs to degradation, viral replication, and protein translation [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eRelated to T1D, PARP1 deletion prevents the development of diabetes induced by streptozotocin in mice [\u003cspan additionalcitationids=\"CR15\" citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. We have recently shown that the ADP-ribosylhydrolase ARH3, which removes ADP-ribosylation from serine residues [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], mediates the reduction of cytokine-induced apoptosis and production of chemokine CXCL9 by omega-3 fatty acids in MIN6 insulin-producing cells [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Mechanistically, omega-3 fatty acids induce the degradation of SUZ12, a component of the histone methylation polycomb PCR2, leading to a reduction in histone methylation and increasing the expression of ARH3. However, it is unknown if the other 16 mammalian PARPs or 3 hydrolases have functions in T1D development.\u003c/p\u003e \u003cp\u003eHere, we investigated the role of PARP12 in insulitis owing to its expression profile in previous omics datasets [\u003cspan additionalcitationids=\"CR20 CR21\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. We performed genomic analysis to determine mechanistically how the expression of PARP12 is regulated by pro-inflammatory cytokines in insulitis. We next performed a combination of RNA silencing and proteomics to determine possible functions of PARP12. This investigation identified PARP12 as an RNA ADP-ribosyltransferase, playing a role in post-transcriptional regulation of inflammation.\u003c/p\u003e"},{"header":"Material and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eData reanalysis\u003c/h2\u003e \u003cp\u003eProtein relative abundances (fold changes) of all 17 mouse and human PARPs were extracted along with their statistical significance from previously published proteomics datasets from our laboratory. The model systems included mouse insulinoma cell line MIN6 treated with cytokine cocktail CT1: IFN-γ, IL-1β and TNFα for 24 h [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e], human islets treated with CT2: IFN-γ and IL-1β for 24 h [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e], and human insulin-producing cell line EndoC-βH1 treated with CT2: IFN-γ and IL-1β for 48 h [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e] or CT3: IFN-α for 8 h and 48 h [\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e] (\u003cb\u003eESM Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eFor single-cell RNA seq analysis, data from human islets were downloaded from the Human Pancreas Analysis Program (HPAP) portal (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://hpap.pmacs.upenn.edu/\u003c/span\u003e\u003cspan address=\"https://hpap.pmacs.upenn.edu/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e) and processed with Cell Ranger (v6.1.2) as previously described [\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e] (donor characteristics in \u003cb\u003eESM Table S2\u003c/b\u003e). After cell type annotation, we retrieved β cells from donors normoglycemic (non-diabetic, ND), positive for 1 (AAB1+) or multiple autoantibodies (AAB2+), and with T1D. In total, we analyzed 6,423 β cells from 15 ND, 5,751 from 9 AAB1+, 4,977 from 2 AAB2+, and 2,013 from 11 T1D donors. Differential gene expression analysis for β cells between T1D, AAB2\u0026thinsp;+\u0026thinsp;and AAB1\u0026thinsp;+\u0026thinsp;vs ND was performed using the \u003cem\u003eFindMarkers\u003c/em\u003e function from Seurat, employing the Wilcoxon rank-sum test. P-values were adjusted using the Bonferroni correction method to account for multiple comparisons for all the genes (\u003cb\u003eESM Table S3\u003c/b\u003e).\u003c/p\u003e \u003cp\u003eChromatin features (ATAC-seq and H3K27ac CHIP-seq) and gene expression (RNA-seq), along with annotated regulatory elements in human pancreatic islets treated or untreated with CT2 (IFN-γ and IL-1β) for 48 h, were obtained from our previous publication [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. The datasets were lifted over from hg19 to hg38 genome assemblies. We selected regulatory elements within a 40 kb window centered on the PARP12 locus and divided them into three groups: Promoter/SRE (n\u0026thinsp;=\u0026thinsp;1, PARP12 promoter), Distal/Opening_IRE (n\u0026thinsp;=\u0026thinsp;2, increased accessibility and H3K27ac upon cytokines), and Distal/SRE (n\u0026thinsp;=\u0026thinsp;7, without after cytokine exposure). Motif enrichment analysis was conducted using AME from the MEME suite (version 5.4.1) [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e], testing for enrichment of cytokine-regulated TF motifs (i.e., transcription factors whose expression is upregulated upon cytokine exposure) across the three groups of regulatory elements, using shuffled sequences as a control.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eCell culture\u003c/h3\u003e\n\u003cp\u003eMIN6 cells were a gift from the Yamagata lab and were grown in DMEM containing 4.5 g/L each of D-glucose and L-glutamine, 10% FBS, 100 units/mL penicillin, 100 \u0026micro;g/mL streptomycin, and 50 mM 2-mercaptoethanol maintained at 37 \u0026ordm;C in a 5% CO\u003csub\u003e2\u003c/sub\u003e atmosphere. For PARP12 knockdown experiments, cells were treated at 80% confluency using Lipofectamine RNAiMAX (Invitrogen, Cat# 13778150) with SMARTpool ON-TARGETplus non-targeting siRNA (Dharmacon, cat#D-001810-10-20) or siRNA targeting \u003cem\u003eParp12\u003c/em\u003e (Dharmacon, cat# L-065127-01-0020), with concomitant treatment with CT2 (100 ng/mL IFN-γ: R\u0026amp;D, cat#485-MI-100, 10 ng/mL TNF-α: R\u0026amp;D, Cat#410-MT-010, and 5 ng/mL IL-1β: R\u0026amp;D, cat #401-ML-005) for 24 h. For pulse-chase experiments with stable isotope labeling by/with amino acids in cell culture (SILAC), cells were treated in DMEM containing heavy \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e6\u003c/sub\u003e\u003csup\u003e15\u003c/sup\u003eN\u003csub\u003e2\u0026minus;\u003c/sub\u003elysine (Thermo Fisher Scientific, Cat# A33969).\u003c/p\u003e\n\u003ch3\u003eLabel-free proteomics analysis\u003c/h3\u003e\n\u003cp\u003eThe MIN6 cell pellets were resuspended in 50 mM Tris-HCl, 8 M urea, and 10 mM dithiothreitol, and incubated for 1 h at 37 \u0026ordm;C with shaking at 800 rpm. Subsequently, 400 mM iodoacetamide was added to a final concentration of 40 mM, and the samples were incubated for 1 h in the dark at room temperature. The samples were then diluted 8-fold with 50 mM Tris-HCl, and 1 M CaCl\u003csub\u003e2\u003c/sub\u003e was added to a final concentration of 1 mM. Proteins were digested overnight at room temp using trypsin at an enzyme-to-protein ratio of 1:50. The digested peptides were desalted using C18 cartridges (Discovery, 50 mg, Sulpelco) and dried in a vacuum centrifuge. Peptides were analyzed using a Q Exactive Plus mass spectrometer (Thermo Scientific) as previously described[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. The data were processed with MaxQuant software (v.1.5.5.) [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e] using the mouse reference proteome database from UniProt Knowledge Base (downloaded on 02-18-2022, 17082 entries). Protein N-terminal acetylation and oxidation of methionine were set as variable modifications, and cysteine carbamidomethylation was set as a fixed modification. The software's default mass shift tolerance was used. Only fully tryptic-digested peptides were considered, with up to two missed cleavage sites allowed per peptide. Protein quantification was performed using the intensity-based absolute quantification (iBAQ) method [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. The data was normalized across samples based on total protein level, followed by substituting 1/4th of the lowest value from the dataset as missing values and log2 transformed. Statistically significant proteins (p\u0026thinsp;\u0026le;\u0026thinsp;0.05) were determined using Student\u0026rsquo;s \u003cem\u003et\u003c/em\u003e-test. Network analysis of the statistically significant proteins was done with String database (V12.0) [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e].\u003c/p\u003e\n\u003ch3\u003eImmunofluorescence imaging\u003c/h3\u003e\n\u003cp\u003eMIN6 cells were cultured in 35 mm culture dishes (MatTek, Cat#P35G-1.5-10-C) and remained covered with PBS (Gibco, Cat#10010023) throughout the imaging process. They were permeabilized with 0.5% Triton X-100 for 30 min, followed by blocking with 1\u0026times; PBS containing 5% goat serum and 0.3% Triton X-100 for 1 h at room temperature. The sections were incubated in 1:500 dilution of monoclonal rabbit anti-insulin IgG (Cell Signaling Technology, 3014), PARP12 (G bioscience, ITN2133-50u-555) and G3BP (Thermo Fisher, Cat#CL488-66486) overnight at 4 \u0026ordm;C. The primary antibody was thoroughly washed with PBS followed with incubation in 1:1000 dilution goat anti-rabbit IgG (H\u0026thinsp;+\u0026thinsp;L) highly cross-adsorbed secondary antibody, Alexa Fluor Plus 800 (Thermo Fisher Scientific).\u003c/p\u003e \u003cp\u003eFor the fliFISH analysis, each primary FISH probe was composed of about 20 nucleotides complementary to the target mRNA and 28 nucleotides complementary to the secondary probe (Integrated DNA Technologies) [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e] (\u003cb\u003eESM table S4\u003c/b\u003e). The properties of the probes, such as Tm, DG, and hybridization efficiency, were evaluated [\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e] and their targeting specificity was confirmed with BLAST (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://blast.ncbi.nlm.nih.gov/Blast.cgi\u003c/span\u003e\u003cspan address=\"https://blast.ncbi.nlm.nih.gov/Blast.cgi\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). FISH probe solution was prepared by incubating 0.5 \u0026micro;M of each primary probe and 5 \u0026micro;M secondary probe (conjugated Alex 647 dye in both oligonucleotide termini) in 50 M Tris-HCl, pH 7.9, containing 100 M NaCl, 10 mM MgCl\u003csub\u003e2\u003c/sub\u003e and 1 mM dithiothreitol at 85 \u0026ordm;C for 3 min, followed by a gradual cool down to room temperature to hybridize the primary and secondary probes. Cell slides were immersed in 70% ethanol overnight at 4 \u0026ordm;C to permeabilize cell membranes. The slides were incubated with the FISH probe solution diluted to a final concentration of 4 nM of each probe in hybridization buffer (10% dextran sulfate, 2 mM vanadyl-ribonucleoside complex, 0.02% BSA, 2 \u0026times; SSC, 10% formamide) overnight incubation at 37 \u0026ordm;C. Slides were thoroughly rinsed with buffer containing 2 \u0026times; SSC and 10% formamide. The cell nuclei were stained with 0.05% DAPI.\u003c/p\u003e \u003cp\u003eImaging was carried out using a Zeiss LSM 710 confocal laser scanning microscope, captured using a 100\u0026times; oil immersion objective (NA 1.47). Laser excitation was set to 633 nm for \u003cem\u003eIns2\u003c/em\u003e (Alexa Fluor 647), 542 nm for PARP12 antibody (Alexa Fluor 555), and 405 nm for DAPI. Fluorescence emission was detected using spectral settings of 650\u0026ndash;700 nm for Alexa Fluor 647, 570\u0026ndash;620 nm for Alexa Fluor 555, and 420\u0026ndash;480 nm for DAPI. Key imaging parameters, including detector gain, laser power, and pinhole size, were optimized to minimize photobleaching while ensuring a high signal-to-noise ratio for each channel. Image processing was performed using FIJI/ImageJ, applying only linear adjustments to brightness and contrast without modifying pixel intensity values.\u003c/p\u003e\n\u003ch3\u003ePrediction of RNA binding to PARP12 zinc finger motifs\u003c/h3\u003e\n\u003cp\u003eThe interaction strength of \u003cem\u003eIns2\u003c/em\u003e pre-mRNA with PARP12 was calculated using the RPISeq program (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttp://pridb.gdcb.iastate.edu/RPISeq/index.html\u003c/span\u003e\u003cspan address=\"http://pridb.gdcb.iastate.edu/RPISeq/index.html\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). For PARP12, we used sequences from three zinc finger (ZnF1-3) domains in the following combinations: ZnF1 and 2, ZnF2 and 3, and ZnF1, 2 and 3. The interaction strength was predicted in 50-nucleotide segments of \u003cem\u003eIns2\u003c/em\u003e pre-mRNA with 10-nucleotide overlapping in both ends.\u003c/p\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003ePrediction of the PARP12 structure\u003c/h2\u003e \u003cp\u003eThe PARP12 structure predicted with AlphaFold (identifier AF-Q8BZ20-F1) was downloaded from Uniprot Knowledgebase and analyzed with Discovery Studio Visualizer v21. The \u003cem\u003eIns2\u003c/em\u003e linear RNA oligonucleotide (ACAGGCAUGCAACCCCUGCCACCUG) with the highest predicted binding was positioned on PARP12, based on the ZnF motif orientations similar to RNA-bound protein structures of PDB, 3D2S and 7C07. The geometry of the RNA-bound PARP12 model was optimized using Discovery Studio Visualizer.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eADP-ribosylated RNA pulldown\u003c/h3\u003e\n\u003cp\u003eMIN6 cells were harvested using Tri reagent (Zymo Research, R2050-1-200) and the RNA was extracted using RNA Clean \u0026amp; Concentrator\u0026trade;-5 kit (Zymo Research, R1014), following manufacturer instructions. To test the specificity of the ADP-ribosylation-binding AF1521 macrodomain, an RNA aliquot was pre-treated with 2 nM SARS-CoV2 macrodomain MAC1 at room temp for 1 h in hydrolase reaction buffer (250 mM HEPES, 750 mM NaCl, 10 mM β-mercaptoethanol, 0.05 mM Triton X100, 5 mM MgCl\u003csub\u003e2\u003c/sub\u003e, 1X RNAsecure) to remove ADP-ribosylation. Equal amount of total RNA from each sample was incubated with either 20 \u0026micro;L of Af1521 macrodomain conjugated to magnetic beads (Tulip Biolabs, 2426) or control (naked or mutated Af1521) beads in 600 \u0026micro;L RNA pulldown buffer (10 mM Tris-HCl, 0.6 M NaCl, 0.1% NP40, 1X RNA secure) overnight at 4\u0026deg;C. After overnight incubation, the beads were gently washed with RNA pulldown buffer three times, and 50 \u0026micro;L of Tri reagent was added to elute RNA from the beads, followed by RNA Clean \u0026amp; Concentrator\u0026trade;-5 extraction.\u003c/p\u003e\n\u003ch3\u003eQuantitative real-time PCR analysis\u003c/h3\u003e\n\u003cp\u003eExtracted total and Af1521-enriched RNAs were quantified using nanodrop and expression level of genes were measured in StepOnePlus RTPCR systems using QuantiNova\u0026trade; SYBR Green RT-PCR reagent (Qiagen, Cat. No. / ID: 208154). Pre-designed mouse primers were ordered from Millipore sigma (Cat# KSPQ12012) and IDT (\u003cb\u003eEMS Table S5\u003c/b\u003e). The average CT value of two housekeeping genes Rlp13 and Nono was used to calculate the fold change of genes using Livak\u0026rsquo;s method [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e].\u003c/p\u003e \u003cdiv id=\"Sec11\" class=\"Section2\"\u003e \u003ch2\u003eRNAseq analysis\u003c/h2\u003e \u003cp\u003eRNAs eluted post-incubation with AF1521 and Control (mutant AF1521) beads were processed for cDNA synthesis using SMARTer\u0026reg; Universal Low Input RNA Kit (Takara, cat#634938) according to instruction manual, followed by template library prep using NEBNext\u0026reg; Ultra\u0026trade; II FS DNA Library Prep Kit for Illumina (NEB, cat#E7805S). Single-read sequencing of the barcoded cDNA libraries with a read length of 150 were performed on NextSeq 550 plateform using NextSeq500/550 High Output v2.5 150 cycles (cat#20024907). Quality control of RNA-seq data was performed using FastQC (Version 0.12.0) (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.bioinformatics.babraham.ac.uk/projects/fastqc/\u003c/span\u003e\u003cspan address=\"https://www.bioinformatics.babraham.ac.uk/projects/fastqc/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Read trimming was performed using BBDuk (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://github.com/BioInfoTools/BBMap/blob/master/sh/bbduk.sh\u003c/span\u003e\u003cspan address=\"https://github.com/BioInfoTools/BBMap/blob/master/sh/bbduk.sh\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). Reads were aligned to the mouse genome GRCm39 using STAR[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e] and mapped to genome features using the featureCounts program from the subread package (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://subread.sourceforge.net/\u003c/span\u003e\u003cspan address=\"https://subread.sourceforge.net/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e). To account for technical variability raw counts were batch corrected with ComBat-Seq[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e] with default parameters prior to differential expression analysis with DESeq2[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e].\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec12\" class=\"Section2\"\u003e \u003ch2\u003eHeavy isotope-labeled lysine proteomics\u003c/h2\u003e \u003cp\u003ePARP12 siRNA cells were treated with CT1 in heavy isotope-labeled lysine SILAC media for 24 h. Samples were submitted to proteomics analysis as described above with the exception that the digestion was done with endoproteinase LysC instead trypsin. Raw files were processed in MaxQuant (v.2.5.1.0), using mouse reference protein datasets downloaded from UniprotKB (2023-03-01, 21,949 entries). The multiplicity was set to 2 considering light and heavy \u003csup\u003e13\u003c/sup\u003eC\u003csub\u003e6\u003c/sub\u003e\u003csup\u003e15\u003c/sup\u003eN\u003csub\u003e2\u0026minus;\u003c/sub\u003elysine labels. Searches were performed with oxidation of methionine and protein N-terminal acetylation as variable modifications and cysteine carbamidomethylation as fixed modification. The quantification parameters included re-quantification and matching between runs with a matching and aligning time windows of 1 and 20 min, respectively. All remaining parameters were kept as the default settings.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec13\" class=\"Section2\"\u003e \u003ch2\u003eIntact insulin analysis by mass spectrometry\u003c/h2\u003e \u003cp\u003ePellets of 300,000 cells pulsed-chased in SILAC media were lysed in 10 \u0026micro;L of water. Aliquots of 2 \u0026micro;L were spotted onto poly-lysine coated indium-tin oxide slides (Bruker Daltonics, Billerica, MA). After drying, slides were serially washed for 30 s each with 70% ethanol, 100% ethanol, and 50 mM ammonium acetate. The spotted cells were prepared by spraying 5% acetic acid in 50% ethanol and subsequently applying 15 mg/mL 2,5-dihydroxyacetophenone as matrix in 90% acetonitrile with 0.2% trifluoroacetic acid using an M5 sprayer (HTX Technologies, Chapel Hill, NC)[\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Slides were analyzed on an orbitrap mass spectrometer (Q Exactive HF, Thermo Fisher Scientific, Bremen, DE) retrofit with ultrahigh mass range (UHMR) boards and an EP-MALDI source (Spectroglyph LLC, Kennewick, WA). Briefly, the instrument was operated under custom privileges licenses, and mass spectra were collected with transients of 1.024 s (480k mass resolution at \u003cem\u003em/z\u003c/em\u003e 200) and a \u003cem\u003em/z\u003c/em\u003e range of 2000 to 20000. The raw data and position files were imported into SCiLS Lab (v.2021c; Bruker Daltonics, Bremen, DE) using automatic import and processing settings. Maximum intensity values were extracted with root mean square normalization across conditions.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec14\" class=\"Section2\"\u003e \u003ch2\u003eData visualization and statistical analysis\u003c/h2\u003e \u003cp\u003eMicrosoft Excel was used to perform basic data analysis with GraphPad Prism 9 (Version 9.4.1) was used for statistical analysis and data visualization. Perseus-Max-Quant was used to create Heatmaps. Pathway analysis for Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e5\u003c/span\u003e and \u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e6\u003c/span\u003e was performed using the core analysis setting of Ingenuity Pathway Analysis software using the Reactome database.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cem\u003ePARP12 regulation by cytokines in islets and beta cells.\u003c/em\u003e \u003c/p\u003e \u003cp\u003eTo investigate the possible roles of PARPs in insulitis, we re-analyzed published proteomics data from human islets, EndoC-βH1 cells and MIN6 cells treated with different combination of pro-inflammatory cytokines (CT1: IFN-γ, IL-1β and TNFα; CT2: IFN-γ and IL-1β; CT3: INFα)[\u003cspan additionalcitationids=\"CR20 CR21\" citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. This re-analysis revealed that ADP-ribosyltransferases PARP9, 10, 12, and 14 are consistently upregulated by the pro-inflammatory cytokines in all 5 datasets (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). We investigated the expression of PARPs in β cells from single-cell RNAseq data from donors at different stages of T1D development, determined by the number of detected islet autoantibodies and disease onset. The expression levels of \u003cem\u003eParps\u003c/em\u003e 1, 4, 5B, 6, 7, 8, 9, 11, 12, and 14 were significantly higher in β cells of individuals with T1D than those with islet autoantibodies compared to the absence of diabetes or non-diabetic donors (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). Together, the results showed that the expression levels of PARP9, 10, 12, and 14 were increased both \u003cem\u003ein vitro\u003c/em\u003e models and donors with T1D. Notably, \u003cem\u003eParp12\u003c/em\u003e abundance was increased in the phospholipase \u003cem\u003ePla2g6\u003c/em\u003e knockdown MIN6 cells[\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. Further, \u003cem\u003eAdprhl2\u003c/em\u003e (ARH3 protein) knockdown causes 52.3% reduction in PARP12 expression by the cytokine cocktail [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. ARH3 expression is regulated by \u003cem\u003ePla2g6\u003c/em\u003e via the release of omega-3 fatty acids [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. To test if PARP12 is also regulated by omega-3 fatty acids, we treated MIN6 cells with cytokine cocktail CT1 in the presence of the omega-3 fatty acids, eicosapentaenoic acid (EPA) and docosahexaenoic acid (DHA). The EPA and DHA increased the cytokine-induced \u003cem\u003eParp12\u003c/em\u003e expression by 54% and 62%, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e1\u003c/span\u003eC). Based on these results, we hypothesized that PARP12 has a role in cytokine-mediated signaling and β-cell stress and selected it for further characterization.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo investigate the regulation of the gene encoding PARP12, we examined our previously published histone H3K27Ac enrichment profiles, RNA-seq and ATAC-seq data of CT2-treated human islets[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e]. We detected a cytokine-induced enrichment of the active histone modification H3K27Ac upstream of the \u003cem\u003ePARP12\u003c/em\u003e promoter region (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e1\u003c/span\u003eD). The higher H3K27Ac levels matched with an increased chromatin accessibility measured by ATAC-seq (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e1\u003c/span\u003eD). As expected, these cytokine-induced chromatin changes correlated with increased transcription activity, measured by RNA-seq data (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e1\u003c/span\u003eD). To determine potential transcription factors involved in regulation of \u003cem\u003eParp12\u003c/em\u003e gene expression, we studied the sequence composition of the gene locus. We found that the \u003cem\u003eParp12\u003c/em\u003e gene promoter and distal regulatory elements bear binding motifs for POU2F2, JUNB, NFKB1, IRF1, IRF2, IRF4, IRF8, and IRF9 (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e1\u003c/span\u003eE). Among those, IRF9 binding motif sequence in the promoter region of the gene was apparent, prompting functional analyses. Knockdown of \u003cem\u003eIrf9\u003c/em\u003e in MIN6 cells reduced the cytokine cocktail CT1-mediated induction of \u003cem\u003eParp12\u003c/em\u003e by 52% (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e1\u003c/span\u003eF), confirming the role of this transcription factor in regulating \u003cem\u003eParp12\u003c/em\u003e gene expression.\u003c/p\u003e \u003cp\u003eOverall, these data identified PARP12 as a cytokine-regulated PARP during insulitis in islets and that its gene expression is regulated by IRF9.\u003c/p\u003e \u003cdiv id=\"Sec16\" class=\"Section2\"\u003e \u003ch2\u003eProteomics of PARP12 knockdown MIN6 cells\u003c/h2\u003e \u003cp\u003eTo gain knowledge on PARP12 function, we performed proteomic analysis of cytokine cocktail CT1-treated MIN6 cells with or without \u003cem\u003eParp12-\u003c/em\u003etargeting siRNA. As a control, we observed 62% and 85% knockdown efficiency of both \u003cem\u003eParp12\u003c/em\u003e mRNA and its encoded protein levels in the CT1-treated group, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Our label-free proteomic analysis identified 4898 proteins, of which 683 proteins were regulated by the cytokine cocktail (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003eB, \u003cb\u003eESM Table S6\u003c/b\u003e). Moreover, 63 of the cytokine-regulated proteins were significantly affected by PARP12 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003eC), which were submitted to a network analysis. The network analysis identified a cluster of interacting proteins related to transcription and post-transcription regulation: IRGM1, IRF9, PARP10, ZC3HAV1 and OAS3 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003eD). In the cytokine-treated groups, the production of these proteins was partially or completely dependent on PARP12 (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e2\u003c/span\u003eE). Together, the proteomics analysis indicates that PARP12 regulates proteins related to transcription and post-transcription regulation.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec17\" class=\"Section2\"\u003e \u003ch2\u003ePARP12 post-transcriptionally regulates insulin 2\u003c/h2\u003e \u003cp\u003eTo investigate the possibility of PARP12 being a post-transcriptional regulator, we analyzed the proteins that were affected by the knockdown of \u003cem\u003eParp12\u003c/em\u003e siRNA in cells treated with CT1. We found 94 proteins with a significant increase in cells treated with \u003cem\u003eParp12\u003c/em\u003e siRNA and CT1 compared to parental cells treated with CT1 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003eA). Insulin 2 (INS2) was one of the proteins whose abundance was downregulated by PARP12 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003eB). However, its mRNA level was not significantly altered (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003eC), supporting that PARP12 post-transcriptionally regulates insulin production. We next performed \u003cem\u003ein situ\u003c/em\u003e hybridization of \u003cem\u003eInsulin 2 (Ins2)\u003c/em\u003e mRNA in MIN6 parental and PARP12 knockdown cells treated with cytokines. The cells were also immunostained for PARP12, which showed a perinuclear localization and increased abundance in cells treated with cytokines (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). As expected, \u003cem\u003eParp12\u003c/em\u003e siRNA reduced its protein abundance in both parental and siRNA cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). Under cytokine treatment, \u003cem\u003eIns2\u003c/em\u003e mRNA distribution moves from the perinuclear region and became diffused in the cytosol (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). In \u003cem\u003eParp12\u003c/em\u003e siRNA cells, \u003cem\u003eIns2\u003c/em\u003e mRNA remained in perinuclear localization, showing that its release into the cytosol is dependent on PARP12 (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e3\u003c/span\u003eD). Overall, these data showed that PARP12 is a post-transcriptional regulator that releases the \u003cem\u003eIns2\u003c/em\u003e mRNA from the perinuclear region to a diffused distribution in the cytosol.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003ePARP12 is a potential RNA ADP-ribosyltransferase\u003c/h2\u003e \u003cp\u003eAs PARP12 is an ADP-ribosyltransferase, we hypothesized that PARP12 ADP-ribosylates the \u003cem\u003eIns2\u003c/em\u003e mRNA to downregulate its translation. To investigate this possibility, we examined the PARP12 structures, including its catalytic domain (PDB ID: 6V3W) and an AlphaFold prediction (identifier AF-Q8BZ20-F1) of the remaining protein composed of two WWE (tryptophan-tryptophan-glutamate) domains, and four zinc finger domains (ZnF) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). We assembled the \u003cem\u003eIns2\u003c/em\u003e mRNA onto this predicted structure based on the catalytic site and the ZnF domains of PARP12. These four ZnF domains are predicted to form a groove that could accommodate RNA molecules (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Of these four, ZnF motifs 1, 2 and 3 are predicted to bind to the RNA and were therefore used to identify possible \u003cem\u003eIns2\u003c/em\u003e mRNA regions that might interact with PARP12. The \u003cem\u003eIns2\u003c/em\u003e mRNA was divided into 27 oligonucleotide sequences of 50 nucleotides with 10 overlapping nucleotides between oligonucleotides. The oligonucleotide with the highest prediction corresponded to a region in intron 2 of insulin 2 mRNA (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003eB). We then positioned the oligonucleotide with the highest predicted binding affinity onto the PARP12 structure. This prediction analysis showed that the PARP12 structure can accommodate RNA molecules in proximity to its catalytic site (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003eA), indicating this enzyme may ADP-ribosylate transcripts such as \u003cem\u003eIns2\u003c/em\u003e.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eTo determine if the \u003cem\u003eIns2\u003c/em\u003e mRNA gets ADP-ribosylated, we extracted RNAs from MIN6 cells treated or not with CT1 followed by pull-down of the ADP-ribosylated RNAs with AF1521 ADP-ribosylation binding macrodomain and performed qPCR (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). \u003cem\u003eIns2\u003c/em\u003e mRNA was amplified in both untreated and CT1-treated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). As a control, pulldowns with beads alone without AF1521 showed only trace amplification (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003eD). To test the specificity of AF1521 beads, we pretreated RNA aliquots with SARS-CoV2 macrodomain MAC1 to remove ADP-ribosylation (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003eC). This treatment led to a 94\u0026ndash;99% reduction in \u003cem\u003eIns2\u003c/em\u003e mRNA amplification (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e4\u003c/span\u003eD), confirming that the AF1521 specifically enriches for ADP-ribosylated RNAs. Together, these data show that PARP12 is a potential RNA ADP-ribosyltransferase and that the \u003cem\u003eIns2\u003c/em\u003e mRNA is ADP-ribosylated.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eLandscape of PARP12 ADP-ribosylated RNAs\u003c/h2\u003e \u003cp\u003eTo determine if PARP12 is an RNA ADP-ribosyltransferase and to identify its substrates, we enriched ADP-ribosylated RNAs from PARP12 knockdown cells treated with cytokines CT1 with AF1521 beads and submitted them to RNAseq.\u0026nbsp;By quantifying RNAs enriched in AF1521 vs. control beads, 1381 RNAs were found to be ADP-ribosylated. From these, 150 RNAs were significantly downregulated in PARP12 knockdown cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e5\u003c/span\u003eA, \u003cb\u003eESM table S7\u003c/b\u003e), therefore are potential PARP12 substrates. A function enrichment analysis of the PARP12 substrates showed top enrichment of genes related to respiratory electron transport, cellular responses to stimuli, metabolism of proteins and RNAs, immune system and cell cycle (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e5\u003c/span\u003eB, \u003cb\u003eESM table S8\u003c/b\u003e). PARP12 knockdown led to a 63% reduction in insulin 2 ADP-ribosylation in cells treated with CT1 but no significant changes were observed in untreated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). These results show that ADP-ribosylation is a major RNA modification and that PARP12 is an RNA ADP-ribosyltransferase that modifies many transcripts, including the insulin 2 mRNA.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003ePARP12 suppresses the translation of specific proteins\u003c/h2\u003e \u003cp\u003eWe next tested if PARP12-mediated ADP-ribosylation of RNAs has a role in regulating protein translation. We treated PARP12 siRNA MIN6 cells with CT1 and pulse-chased them with heavy isotope-labeled lysine for 24 h to label nascent proteins and analyzed by proteomics. The analysis identified 187 proteins whose translation was affected by PARP12 in untreated and CT1-treated cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e6\u003c/span\u003eA, \u003cb\u003eESM table S9\u003c/b\u003e). A functional-enrichment analysis revealed that 19 pathways categories, with top enriched pathways from Cellular response to stimuli, metabolism of protein and RNA, were influenced by PARP12 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e6\u003c/span\u003eB, \u003cb\u003eESM table S10\u003c/b\u003e). As the proteomics analysis did not yield to detectable insulin peptides, we measured insulin 2 by intact protein mass spectrometry analysis. Insulin 2 translation is repressed by the CT1 treatment, but this repression is alleviated in \u003cem\u003eParp12\u003c/em\u003e siRNA cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e6\u003c/span\u003eC). These results show that PARP12 regulates the translation of many proteins, including repressing insulin 2 production.\u003c/p\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eOur data showed that PARP12 expression is induced by pro-inflammatory cytokines in human islets and insulin-producing cell lines. PARP12 is also elevated in β cells during T1D development. Mechanistically, cytokines induce PARP12 gene expression by histone acetylation, chromatin opening and activation of the transcription factor IRF9. IRF9 activated by interferon signaling, has an essential role in the expression of interferon-stimulated genes (ISGs) during antiviral responses [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. Indeed, PARP12 is an antiviral factor that suppresses Zika virus by targeting its non-structural proteins NS1 and NS3 to proteasomal degradation [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. PARP12 also impedes viral replication by targeting viral RNAs in stress granules, protein-RNA aggregates, stalling the translation [\u003cspan additionalcitationids=\"CR39\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eHere, we show that PARP12 blocks insulin 2 translation by ADP-ribosylating and translocating its transcript from the perinuclear region to the cytosol under cytokine influence. PARP12 overexpression causes global translational repression by associating with cellular translational machinery and potentially targeting mRNA to stress granules [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. \u003cem\u003eE. coli\u003c/em\u003e has a similar machinery that ADP-ribosylates and halts the translation of phage transcripts as part of an antiviral response [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. Conversely, coronavirus ADP-ribosylhydrolase MAC1 counteracts the protective effect of PARP12[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. At least under \u003cem\u003ein vitro\u003c/em\u003e conditions, we show that MAC1 efficiently cleaves PARP12-mediated mRNA ADP-ribosylation, suggesting a counteracting mechanism for ADP-ribosylation-mediated translation halting. Our data show that a large portion of the cell transcriptome is ADP-ribosylated, but only a fraction is mediated by PARP12. Other mammalian PARPs ADP-ribosylate RNAs, but specific substrates remained unknown [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. Therefore, it is possible that the ADP-ribosylation of mRNAs is a much broader translation regulation mechanism. Besides targeting mRNAs, ADP-ribosylation represses translation by PARP16-mediated ADP-ribosylation of ribosomal protein subunits [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eOur structural analysis predicts that PARP12 has four ZnF motifs that form a groove that accommodates the RNA chain. ZnF motifs can bind to DNA, RNA and ADP-ribosylation [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e]. Similar ZnF motifs are also present in PARP13, also known as zinc finger antiviral protein (ZAP). Despite not displaying ADP-ribosylation activity, PARP13 binds to GC motifs in viral RNAs and target them to degradation [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. Indeed, the insulin 2 mRNA region bearing three GC motifs was predicted to have the highest binding to PARP12, suggesting a possible similar binding specificity compared to PARP13. The structural analysis also identified two ADP-ribosylation-binding WWE domains, which are present in other PARPs and some E3 ubiquitin ligases [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e]. PARP12 WWE domains help its recruitment to stress granules via poly-ADP-ribose signaling catalyzed by PARP1 [\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e]. Combined these structural insights align with our observation that PARP12 acts as a translation inhibitor under pro-inflammatory conditions, potentially binding RNA with its ZnF motifs and utilizing WWE domains for cellular localization and function.\u003c/p\u003e \u003cp\u003eThe reduction in insulin production by PARP12 might aggravate insulin availability during T1D development. At this stage, pro-inflammatory cytokines such as interferons, which induce PARP12 expression, are major regulators of the insulitis process [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e]. Our findings might also have potential implications in aggravation of infection in type 2 diabetic patients, who are already at higher risk for complications. Notably, SARS-CoV-2 infections are associated with a robust production of pro-inflammatory cytokines and insulin depletion [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e, \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]. It is possible that the insulin depletion is caused, at least in part, by ADP-ribosylation of its mRNA and consequent block in translation. However, the involvement of PARP12 in this process still needs to be investigated.\u003c/p\u003e \u003cp\u003eIn conclusion, we identified PARP12 as a cytokine-induced RNA ADP-ribosyltransferase that halts protein translation. Specifically, PARP12 targets insulin mRNA and ADP-ribosylating it and thereby reducing insulin production. This mechanistic insight underpins a potential physiological impact during T1D development, where reduced insulin availability can exacerbate the disease progression. Moreover, our findings offer a broader perspective on the role of PARP12 in inflammatory and antiviral responses, suggesting its involvement in the complex interplay between cytokine signaling and translational control.\u003c/p\u003e"},{"header":"Declarations","content":" \u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe proteomics data were deposited into the MassIVE repository, a member of the ProteomeXchange, under accession numbers/user MSV000097349 and MSV000097350 (accession information below). The ADP-ribosylated RNA sequencing data is available at NCBI Geo under accession number GSE292484.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTitle: Insulin post-transcriptional regulation via PARP12-mediated ADP-ribosylation (global label-free)\u003c/p\u003e\n\u003cp\u003eMassIVE accession: MSV000097349\u003c/p\u003e\n\u003cp\u003ehttps://massive.ucsd.edu/ProteoSAFe/dataset.jsp?task=93584fb8b1df4ad8b5f998c7fafb4ab0\u003c/p\u003e\n\u003cp\u003eFor now, manuscript reviewers can access the data via FTP using:\u003c/p\u003e\n\u003cp\u003eServer: \u0026nbsp; \u0026nbsp;massive.ucsd.edu\u003c/p\u003e\n\u003cp\u003eUser: \u0026nbsp; \u0026nbsp; \u0026nbsp;MSV000097349\u003c/p\u003e\n\u003cp\u003ePassword: Insulin6433\u003c/p\u003e\n\u003cp\u003eTitle: Insulin post-transcriptional regulation via PARP12-mediated ADP-ribosylation (pulse-chase SILAC)\u003c/p\u003e\n\u003cp\u003eMassIVE accession: MSV000097350\u003c/p\u003e\n\u003cp\u003ehttps://massive.ucsd.edu/ProteoSAFe/dataset.jsp?task=778b2027fd574d039b0ecd293259e851\u003c/p\u003e\n\u003cp\u003eManuscript reviewers can access the data via FTP using:\u003c/p\u003e\n\u003cp\u003eServer: \u0026nbsp; \u0026nbsp;massive.ucsd.edu\u003c/p\u003e\n\u003cp\u003eUser: \u0026nbsp; \u0026nbsp; \u0026nbsp;MSV000097350\u003c/p\u003e\n\u003cp\u003ePassword: Insulin6432\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that they have no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgments\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors would like to thank the technical assistance and support of Mr. Armando Puente (University of Chicago), Dr. Mikhail Belov (Spectroglyph, LLC), Mr. Gordon Anderson and Mr. Chris Anderson (GAA Custom Electronics, LLC), Drs. Kyle Fort, Maria Reinhardt-Szyba, and Alexander Makarov (Thermo Fisher Scientific). Part of the work was performed in the Environmental Molecular Sciences Laboratory, a U.S. DOE national scientific user facility at Pacific Northwest National Laboratory (PNNL) in Richland, WA. Battelle operates PNNL for the US Department of Energy under contract DE-AC05-76RLO01830.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions Statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConceptualization: SS, ESN. Investigation, data curation, formal analysis and validation: all authors. Visualization: SS, FL, YY, HK, MR-R, XY. Methodology: SS, FL, YY, KJZ, LMM, HK, AKL, RGM, ESN. Project administration, supervision and resources: DLE, AKLL, LP, LS, TOM, RGM, ESN. Funding acquisition: DLE, MR-R, LP, LS, TOM, RGM, ESN. Writing – original draft: SS, FL, YY, KJZ, HK, MR-R, XY, JRE, HM, ESN. Writing – review \u0026amp; editing: all authors.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by the Catalyst Award from the Human Islet Research Network (HIRN) (to E.S.N) (via U24 DK104162) and by National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) grants R01 DK138335 (to ESN, and TOM), U01 DK127505 (to LS and ESN), U01 DK127786 (to RGM, DLE, and TOM), R01 DK060581 (to RGM), R01 DK105588 (to RGM), and P30 DK020595 (to RGM). The development of ADP-ribosylated RNA sequencing assay was conducted under the Laboratory Directed Research and Development Program at PNNL (to ESN). This work was also supported by Spanish Ministry of Science and Innovation PID2023-151556OB-I00, CNS2024-154742, “la Caixa” Foundation, LCF-PR-HR24-00150 (to LP) and the 2024 EFSD/Lilly Young Investigator Research Award (MR-R).\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eLivingstone SJ, Levin D, Looker HC, Lindsay RS, Wild SH, Joss N, Leese G, Leslie P, McCrimmon RJ, Metcalfe W, et al: \u003cstrong\u003eEstimated life expectancy in a Scottish cohort with type 1 diabetes, 2008-2010.\u003c/strong\u003e \u003cem\u003eJAMA \u003c/em\u003e2015, \u003cstrong\u003e313:\u003c/strong\u003e37-44.\u003c/li\u003e\n\u003cli\u003eGregory GA, Robinson TIG, Linklater SE, Wang F, Colagiuri S, de Beaufort C, Donaghue KC, International Diabetes Federation Diabetes Atlas Type 1 Diabetes in Adults Special Interest G, Magliano DJ, Maniam J, et al: \u003cstrong\u003eGlobal incidence, prevalence, and mortality of type 1 diabetes in 2021 with projection to 2040: a modelling study.\u003c/strong\u003e \u003cem\u003eLancet Diabetes Endocrinol \u003c/em\u003e2022, \u003cstrong\u003e10:\u003c/strong\u003e741-760.\u003c/li\u003e\n\u003cli\u003eEizirik DL, Pasquali L, Cnop M: \u003cstrong\u003ePancreatic beta-cells in type 1 and type 2 diabetes mellitus: different pathways to failure.\u003c/strong\u003e \u003cem\u003eNat Rev Endocrinol \u003c/em\u003e2020, \u003cstrong\u003e16:\u003c/strong\u003e349-362.\u003c/li\u003e\n\u003cli\u003eDiMeglio LA, Evans-Molina C, Oram RA: \u003cstrong\u003eType 1 diabetes.\u003c/strong\u003e \u003cem\u003eLancet \u003c/em\u003e2018, \u003cstrong\u003e391:\u003c/strong\u003e2449-2462.\u003c/li\u003e\n\u003cli\u003eFehr AR, Singh SA, Kerr CM, Mukai S, Higashi H, Aikawa M: \u003cstrong\u003eThe impact of PARPs and ADP-ribosylation on inflammation and host-pathogen interactions.\u003c/strong\u003e \u003cem\u003eGenes Dev \u003c/em\u003e2020, \u003cstrong\u003e34:\u003c/strong\u003e341-359.\u003c/li\u003e\n\u003cli\u003eZarkovic G, Belousova EA, Talhaoui I, Saint-Pierre C, Kutuzov MM, Matkarimov BT, Biard D, Gasparutto D, Lavrik OI, Ishchenko AA: \u003cstrong\u003eCharacterization of DNA ADP-ribosyltransferase activities of PARP2 and PARP3: new insights into DNA ADP-ribosylation.\u003c/strong\u003e \u003cem\u003eNucleic Acids Res \u003c/em\u003e2018, \u003cstrong\u003e46:\u003c/strong\u003e2417-2431.\u003c/li\u003e\n\u003cli\u003eSuskiewicz MJ, Prokhorova E, Rack JGM, Ahel I: \u003cstrong\u003eADP-ribosylation from molecular mechanisms to therapeutic implications.\u003c/strong\u003e \u003cem\u003eCell \u003c/em\u003e2023, \u003cstrong\u003e186:\u003c/strong\u003e4475-4495.\u003c/li\u003e\n\u003cli\u003eMunnur D, Bartlett E, Mikolcevic P, Kirby IT, Rack JGM, Mikoc A, Cohen MS, Ahel I: \u003cstrong\u003eReversible ADP-ribosylation of RNA.\u003c/strong\u003e \u003cem\u003eNucleic Acids Res \u003c/em\u003e2019, \u003cstrong\u003e47:\u003c/strong\u003e5658-5669.\u003c/li\u003e\n\u003cli\u003eVassallo CN, Doering CR, Laub MT: \u003cstrong\u003eAnti-viral defence by an mRNA ADP-ribosyltransferase that blocks translation.\u003c/strong\u003e \u003cem\u003eNature \u003c/em\u003e2024, \u003cstrong\u003e636:\u003c/strong\u003e190-197.\u003c/li\u003e\n\u003cli\u003eMashimo M, Kato J, Moss J: \u003cstrong\u003eStructure and function of the ARH family of ADP-ribosyl-acceptor hydrolases.\u003c/strong\u003e \u003cem\u003eDNA Repair (Amst) \u003c/em\u003e2014, \u003cstrong\u003e23:\u003c/strong\u003e88-94.\u003c/li\u003e\n\u003cli\u003eNie Y, Nirujogi TS, Ranjan R, Reader BF, Chung S, Ballinger MN, Englert JA, Christman JW, Karpurapu M: \u003cstrong\u003ePolyADP-Ribosylation of NFATc3 and NF-kappaB Transcription Factors Modulate Macrophage Inflammatory Gene Expression in LPS-Induced Acute Lung Injury.\u003c/strong\u003e \u003cem\u003eJ Innate Immun \u003c/em\u003e2021, \u003cstrong\u003e13:\u003c/strong\u003e83-93.\u003c/li\u003e\n\u003cli\u003eHassa PO, Hottiger MO: \u003cstrong\u003eThe functional role of poly(ADP-ribose)polymerase 1 as novel coactivator of NF-kappaB in inflammatory disorders.\u003c/strong\u003e \u003cem\u003eCell Mol Life Sci \u003c/em\u003e2002, \u003cstrong\u003e59:\u003c/strong\u003e1534-1553.\u003c/li\u003e\n\u003cli\u003eIwata H, Goettsch C, Sharma A, Ricchiuto P, Goh WW, Halu A, Yamada I, Yoshida H, Hara T, Wei M, et al: \u003cstrong\u003ePARP9 and PARP14 cross-regulate macrophage activation via STAT1 ADP-ribosylation.\u003c/strong\u003e \u003cem\u003eNat Commun \u003c/em\u003e2016, \u003cstrong\u003e7:\u003c/strong\u003e12849.\u003c/li\u003e\n\u003cli\u003eAndreone T, Meares GP, Hughes KJ, Hansen PA, Corbett JA: \u003cstrong\u003eCytokine-mediated beta-cell damage in PARP-1-deficient islets.\u003c/strong\u003e \u003cem\u003eAm J Physiol Endocrinol Metab \u003c/em\u003e2012, \u003cstrong\u003e303:\u003c/strong\u003eE172-179.\u003c/li\u003e\n\u003cli\u003eBurkart V, Wang ZQ, Radons J, Heller B, Herceg Z, Stingl L, Wagner EF, Kolb H: \u003cstrong\u003eMice lacking the poly(ADP-ribose) polymerase gene are resistant to pancreatic beta-cell destruction and diabetes development induced by streptozocin.\u003c/strong\u003e \u003cem\u003eNat Med \u003c/em\u003e1999, \u003cstrong\u003e5:\u003c/strong\u003e314-319.\u003c/li\u003e\n\u003cli\u003eMasutani M, Suzuki H, Kamada N, Watanabe M, Ueda O, Nozaki T, Jishage K, Watanabe T, Sugimoto T, Nakagama H, et al: \u003cstrong\u003ePoly(ADP-ribose) polymerase gene disruption conferred mice resistant to streptozotocin-induced diabetes.\u003c/strong\u003e \u003cem\u003eProc Natl Acad Sci U S A \u003c/em\u003e1999, \u003cstrong\u003e96:\u003c/strong\u003e2301-2304.\u003c/li\u003e\n\u003cli\u003eAbplanalp J, Leutert M, Frugier E, Nowak K, Feurer R, Kato J, Kistemaker HVA, Filippov DV, Moss J, Caflisch A, Hottiger MO: \u003cstrong\u003eProteomic analyses identify ARH3 as a serine mono-ADP-ribosylhydrolase.\u003c/strong\u003e \u003cem\u003eNat Commun \u003c/em\u003e2017, \u003cstrong\u003e8:\u003c/strong\u003e2055.\u003c/li\u003e\n\u003cli\u003eSarkar S, Deiter C, Kyle JE, Guney MA, Sarbaugh D, Yin R, Li X, Cui Y, Ramos-Rodriguez M, Nicora CD, et al: \u003cstrong\u003eRegulation of beta-cell death by ADP-ribosylhydrolase ARH3 via lipid signaling in insulitis.\u003c/strong\u003e \u003cem\u003eCell Commun Signal \u003c/em\u003e2024, \u003cstrong\u003e22:\u003c/strong\u003e141.\u003c/li\u003e\n\u003cli\u003eYou Y, Sarkar S, Deiter C, Elliott EC, Nicora CD, Mirmira RG, Sussel L, Nakayasu ES: \u003cstrong\u003eReduction of Chemokine CXCL9 Expression by Omega-3 Fatty Acids via ADP-Ribosylhydrolase ARH3 in MIN6 Insulin-Producing Cells.\u003c/strong\u003e \u003cem\u003eProteomics \u003c/em\u003e2025, \u003cstrong\u003e25:\u003c/strong\u003ee202400053.\u003c/li\u003e\n\u003cli\u003eColli ML, Ramos-Rodriguez M, Nakayasu ES, Alvelos MI, Lopes M, Hill JLE, Turatsinze JV, Coomans de Brachene A, Russell MA, Raurell-Vila H, et al: \u003cstrong\u003eAn integrated multi-omics approach identifies the landscape of interferon-alpha-mediated responses of human pancreatic beta cells.\u003c/strong\u003e \u003cem\u003eNat Commun \u003c/em\u003e2020, \u003cstrong\u003e11:\u003c/strong\u003e2584.\u003c/li\u003e\n\u003cli\u003eNakayasu ES, Syed F, Tersey SA, Gritsenko MA, Mitchell HD, Chan CY, Dirice E, Turatsinze JV, Cui Y, Kulkarni RN, et al: \u003cstrong\u003eComprehensive Proteomics Analysis of Stressed Human Islets Identifies GDF15 as a Target for Type 1 Diabetes Intervention.\u003c/strong\u003e \u003cem\u003eCell Metab \u003c/em\u003e2020, \u003cstrong\u003e31:\u003c/strong\u003e363-374 e366.\u003c/li\u003e\n\u003cli\u003eRamos-Rodriguez M, Raurell-Vila H, Colli ML, Alvelos MI, Subirana-Granes M, Juan-Mateu J, Norris R, Turatsinze JV, Nakayasu ES, Webb-Robertson BM, et al: \u003cstrong\u003eThe impact of proinflammatory cytokines on the beta-cell regulatory landscape provides insights into the genetics of type 1 diabetes.\u003c/strong\u003e \u003cem\u003eNat Genet \u003c/em\u003e2019, \u003cstrong\u003e51:\u003c/strong\u003e1588-1595.\u003c/li\u003e\n\u003cli\u003eElliott EC, Sakar S, Bramer L, Burnet M, Kim Y-M, Yi X, Estevao IL, Rewers M, Cambronne XA, Vehik K, et al: \u003cstrong\u003eFree fatty acids accelerate \u0026beta;-cell death in type 1 diabetes.\u003c/strong\u003e \u003cem\u003emedRxiv \u003c/em\u003e2024\u003cstrong\u003e:\u003c/strong\u003e2024.2009.2016.24313433.\u003c/li\u003e\n\u003cli\u003eMcLeay RC, Bailey TL: \u003cstrong\u003eMotif Enrichment Analysis: a unified framework and an evaluation on ChIP data.\u003c/strong\u003e \u003cem\u003eBMC Bioinformatics \u003c/em\u003e2010, \u003cstrong\u003e11:\u003c/strong\u003e165.\u003c/li\u003e\n\u003cli\u003eTyanova S, Temu T, Cox J: \u003cstrong\u003eThe MaxQuant computational platform for mass spectrometry-based shotgun proteomics.\u003c/strong\u003e \u003cem\u003eNat Protoc \u003c/em\u003e2016, \u003cstrong\u003e11:\u003c/strong\u003e2301-2319.\u003c/li\u003e\n\u003cli\u003eKrey JF, Wilmarth PA, Shin JB, Klimek J, Sherman NE, Jeffery ED, Choi D, David LL, Barr-Gillespie PG: \u003cstrong\u003eAccurate label-free protein quantitation with high- and low-resolution mass spectrometers.\u003c/strong\u003e \u003cem\u003eJ Proteome Res \u003c/em\u003e2014, \u003cstrong\u003e13:\u003c/strong\u003e1034-1044.\u003c/li\u003e\n\u003cli\u003eSchwanhausser B, Busse D, Li N, Dittmar G, Schuchhardt J, Wolf J, Chen W, Selbach M: \u003cstrong\u003eGlobal quantification of mammalian gene expression control.\u003c/strong\u003e \u003cem\u003eNature \u003c/em\u003e2011, \u003cstrong\u003e473:\u003c/strong\u003e337-342.\u003c/li\u003e\n\u003cli\u003eSnel B, Lehmann G, Bork P, Huynen MA: \u003cstrong\u003eSTRING: a web-server to retrieve and display the repeatedly occurring neighbourhood of a gene.\u003c/strong\u003e \u003cem\u003eNucleic Acids Res \u003c/em\u003e2000, \u003cstrong\u003e28:\u003c/strong\u003e3442-3444.\u003c/li\u003e\n\u003cli\u003eLi F, Hu D, Dieter C, Ansong C, Sussel L, Orr G: \u003cstrong\u003eSingle Molecule-Based fliFISH Validates Radial and Heterogeneous Gene Expression Patterns in Pancreatic Islet beta-Cells.\u003c/strong\u003e \u003cem\u003eDiabetes \u003c/em\u003e2021, \u003cstrong\u003e70:\u003c/strong\u003e1117-1122.\u003c/li\u003e\n\u003cli\u003eYilmaz LS, Parnerkar S, Noguera DR: \u003cstrong\u003emathFISH, a web tool that uses thermodynamics-based mathematical models for in silico evaluation of oligonucleotide probes for fluorescence in situ hybridization.\u003c/strong\u003e \u003cem\u003eAppl Environ Microbiol \u003c/em\u003e2011, \u003cstrong\u003e77:\u003c/strong\u003e1118-1122.\u003c/li\u003e\n\u003cli\u003eLivak KJ, Schmittgen TD: \u003cstrong\u003eAnalysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method.\u003c/strong\u003e \u003cem\u003eMethods \u003c/em\u003e2001, \u003cstrong\u003e25:\u003c/strong\u003e402-408.\u003c/li\u003e\n\u003cli\u003eDobin A, Davis CA, Schlesinger F, Drenkow J, Zaleski C, Jha S, Batut P, Chaisson M, Gingeras TR: \u003cstrong\u003eSTAR: ultrafast universal RNA-seq aligner.\u003c/strong\u003e \u003cem\u003eBioinformatics \u003c/em\u003e2013, \u003cstrong\u003e29:\u003c/strong\u003e15-21.\u003c/li\u003e\n\u003cli\u003eZhang Y, Parmigiani G, Johnson WE: \u003cstrong\u003eComBat-seq: batch effect adjustment for RNA-seq count data.\u003c/strong\u003e \u003cem\u003eNAR Genom Bioinform \u003c/em\u003e2020, \u003cstrong\u003e2:\u003c/strong\u003elqaa078.\u003c/li\u003e\n\u003cli\u003eLove MI, Huber W, Anders S: \u003cstrong\u003eModerated estimation of fold change and dispersion for RNA-seq data with DESeq2.\u003c/strong\u003e \u003cem\u003eGenome Biol \u003c/em\u003e2014, \u003cstrong\u003e15:\u003c/strong\u003e550.\u003c/li\u003e\n\u003cli\u003eZemaitis KJ, Veličković D, Kew W, Fort KL, Reinhardt-Szyba M, Pamreddy A, Ding Y, Kaushik D, Sharma K, Makarov AA, et al: \u003cstrong\u003eEnhanced Spatial Mapping of Histone Proteoforms in Human Kidney Through MALDI-MSI by High-Field UHMR-Orbitrap Detection.\u003c/strong\u003e \u003cem\u003eAnalytical Chemistry \u003c/em\u003e2022, \u003cstrong\u003e94:\u003c/strong\u003e12604-12613.\u003c/li\u003e\n\u003cli\u003ePaul A, Tang TH, Ng SK: \u003cstrong\u003eInterferon Regulatory Factor 9 Structure and Regulation.\u003c/strong\u003e \u003cem\u003eFront Immunol \u003c/em\u003e2018, \u003cstrong\u003e9:\u003c/strong\u003e1831.\u003c/li\u003e\n\u003cli\u003eLi L, Zhao H, Liu P, Li C, Quanquin N, Ji X, Sun N, Du P, Qin CF, Lu N, Cheng G: \u003cstrong\u003ePARP12 suppresses Zika virus infection through PARP-dependent degradation of NS1 and NS3 viral proteins.\u003c/strong\u003e \u003cem\u003eSci Signal \u003c/em\u003e2018, \u003cstrong\u003e11\u003c/strong\u003e.\u003c/li\u003e\n\u003cli\u003eCatara G, Grimaldi G, Schembri L, Spano D, Turacchio G, Lo Monte M, Beccari AR, Valente C, Corda D: \u003cstrong\u003ePARP1-produced poly-ADP-ribose causes the PARP12 translocation to stress granules and impairment of Golgi complex functions.\u003c/strong\u003e \u003cem\u003eSci Rep \u003c/em\u003e2017, \u003cstrong\u003e7:\u003c/strong\u003e14035.\u003c/li\u003e\n\u003cli\u003eWelsby I, Hutin D, Gueydan C, Kruys V, Rongvaux A, Leo O: \u003cstrong\u003ePARP12, an interferon-stimulated gene involved in the control of protein translation and inflammation.\u003c/strong\u003e \u003cem\u003eJ Biol Chem \u003c/em\u003e2014, \u003cstrong\u003e289:\u003c/strong\u003e26642-26657.\u003c/li\u003e\n\u003cli\u003eLee S, Lee YS, Choi Y, Son A, Park Y, Lee KM, Kim J, Kim JS, Kim VN: \u003cstrong\u003eThe SARS-CoV-2 RNA interactome.\u003c/strong\u003e \u003cem\u003eMol Cell \u003c/em\u003e2021, \u003cstrong\u003e81:\u003c/strong\u003e2838-2850 e2836.\u003c/li\u003e\n\u003cli\u003eKerr CM, Parthasarathy S, Schwarting N, O\u0026apos;Connor JJ, Pfannenstiel JJ, Giri E, More S, Orozco RC, Fehr AR: \u003cstrong\u003ePARP12 is required to repress the replication of a Mac1 mutant coronavirus in a cell- and tissue-specific manner.\u003c/strong\u003e \u003cem\u003eJ Virol \u003c/em\u003e2023, \u003cstrong\u003e97:\u003c/strong\u003ee0088523.\u003c/li\u003e\n\u003cli\u003eWeixler L, Feijs KLH, Zaja R: \u003cstrong\u003eADP-ribosylation of RNA in mammalian cells is mediated by TRPT1 and multiple PARPs.\u003c/strong\u003e \u003cem\u003eNucleic Acids Res \u003c/em\u003e2022, \u003cstrong\u003e50:\u003c/strong\u003e9426-9441.\u003c/li\u003e\n\u003cli\u003eChalla S, Khulpateea BR, Nandu T, Camacho CV, Ryu KW, Chen H, Peng Y, Lea JS, Kraus WL: \u003cstrong\u003eRibosome ADP-ribosylation inhibits translation and maintains proteostasis in cancers.\u003c/strong\u003e \u003cem\u003eCell \u003c/em\u003e2021, \u003cstrong\u003e184:\u003c/strong\u003e4531-4546 e4526.\u003c/li\u003e\n\u003cli\u003eCassandri M, Smirnov A, Novelli F, Pitolli C, Agostini M, Malewicz M, Melino G, Raschella G: \u003cstrong\u003eZinc-finger proteins in health and disease.\u003c/strong\u003e \u003cem\u003eCell Death Discov \u003c/em\u003e2017, \u003cstrong\u003e3:\u003c/strong\u003e17071.\u003c/li\u003e\n\u003cli\u003eCoomans de Brachene A, Alvelos MI, Szymczak F, Zimath PL, Castela A, Marmontel de Souza B, Roca Rivada A, Marin-Canas S, Yi X, Op de Beeck A, et al: \u003cstrong\u003eInterferons are key cytokines acting on pancreatic islets in type 1 diabetes.\u003c/strong\u003e \u003cem\u003eDiabetologia \u003c/em\u003e2024, \u003cstrong\u003e67:\u003c/strong\u003e908-927.\u003c/li\u003e\n\u003cli\u003eReiterer M, Rajan M, Gomez-Banoy N, Lau JD, Gomez-Escobar LG, Ma L, Gilani A, Alvarez-Mulett S, Sholle ET, Chandar V, et al: \u003cstrong\u003eHyperglycemia in acute COVID-19 is characterized by insulin resistance and adipose tissue infectivity by SARS-CoV-2.\u003c/strong\u003e \u003cem\u003eCell Metab \u003c/em\u003e2021, \u003cstrong\u003e33:\u003c/strong\u003e2174-2188 e2175.\u003c/li\u003e\n\u003cli\u003eLucas C, Wong P, Klein J, Castro TBR, Silva J, Sundaram M, Ellingson MK, Mao T, Oh JE, Israelow B, et al: \u003cstrong\u003eLongitudinal analyses reveal immunological misfiring in severe COVID-19.\u003c/strong\u003e \u003cem\u003eNature \u003c/em\u003e2020, \u003cstrong\u003e584:\u003c/strong\u003e463-469.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-6473437/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6473437/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eADP-ribosylation is a common modification that occurs in proteins and nucleic acids, regulating many cellular processes ranging from DNA repair to inflammatory signaling. ADP-ribosylation plays an important role in cancer biology, infectious diseases, and obesity, but its role in the development of type 1 diabetes is not well understood. Here, we studied the role of ADP-ribosyltransferase PARP12 in type 1 diabetes development. PARP12 expression is highly induced in human islets treated with pro-inflammatory cytokines or β cells from diabetic donors. Proteomics analysis of MIN6 insulin-producing cells identified that the RNA machinery is regulated by PARP12 during inflammation. PARP12 also ADP-ribosylates 150 mRNAs, including the insulin mRNA. This mRNA ADP-ribosylation in turn modifies transcript localization and halts translation. Overall, our data identified a role for PARP12 in ADP-ribosylation and translation halting of mRNAs, which may affect insulin production during insulitis.\u003c/p\u003e","manuscriptTitle":"Insulin post-transcriptional regulation via PARP12-mediated ADP- ribosylation","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-05-21 10:16:14","doi":"10.21203/rs.3.rs-6473437/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"0f98ea5b-6aa8-4a72-85a3-5b923a0c9fb0","owner":[],"postedDate":"May 21st, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-03-05T01:24:22+00:00","versionOfRecord":[],"versionCreatedAt":"2025-05-21 10:16:14","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-6473437","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6473437","identity":"rs-6473437","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}