Modulatory effects of CNNM4 on protein- l -isoaspartyl- O -methyltransferase repair function during alcohol-induced hepatic damage

ORIGINAL ARTICLE Modulatory effects of CNNM4 on protein-L-isoaspartylO-methyltransferase repair function during alcohol-induced hepatic damage Irene González-Recio 1 |Naroa Goikoetxea-Usandizaga 1,2 | Claudia M. Rejano-Gordillo 1,3,4 |Carolina Conter 1 | Rubén Rodríguez Agudo 1 |Marina Serrano-Maciá 1 | Leidy Estefanía Zapata-Pavas 1 |Patricia Peña-Sanfélix 1 | Mikel Azkargorta 5 |Félix Elortza 5 |José María Herranz 6 | Álex Guillamon Thiery 7 |Armando Raúl Guerra-Ruiz 8 |Ramiro Jover 2,9 | Unai Galicia-Garcia 4 |César Martín 4 |Ute Schaeper 10 | Teresa C. Delgado 1 |Irene Díaz-Moreno 11 |Antonio Díaz Quintana 12 | Daniela Buccella 13 |Rubén Nogueiras 14 |JosepMaria Argemi 15,16 | Matías A. Ávila 6 |Jordi Gratacós-Ginès 7 |Paula Iruzubieta 17 | Elisa Pose 7 |Ramón Bataller 16 |Javier Crespo 17 | Luis Alfonso Martínez-Cruz 1 |María Luz Martínez-Chantar 1,2 1 Liver Disease Lab, Center for Cooperative Research in Biosciences (CIC bioGUNE), Basque Research and Technology Alliance (BRTA), Derio, Spain 2 Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (CIBERehd), Carlos III National Health Institute, Madrid, Spain 3 Department of Biochemistry and Molecular Biology, Faculty of Sciences, University of Extremadura, University Institute of Biosanitary Research of Extremadura (INUBE), Badajoz, Spain 4 Biofisika Institute (UPV/EHU, CSIC), UPV/EHU Science Park, and Department of Biochemistry and Molecular Biology, Faculty of Science and Technology, University of the Basque Country UPV/EHU, Leioa, Spain 5 Proteomics Platform, Center for Cooperative Research in Biosciences (CIC bioGUNE), Basque Research and Technology Alliance (BRTA), Derio, Spain 6 Hepatology Programme, CIMA, Idisna, Universidad de Navarra, Pamplona, Spain 7 Hospital Clinic of Barcelona, University of Barcelona, IDIBAPS, Barcelona, Catalonia, Spain 8 Clinical Biochemistry, Hospital Universitario Marqués de Valdecilla, Santander, Spain 9 Experimental Hepatology Joint Research Unit, IIS Hospital La Fe & Dep. Biochemistry, University of Valencia, Valencia, Spain 10 Silence Therapeutics GmbH, Berlin, Berlin, Germany 11 Institute for Chemical Research (IIQ), Scientific Research Centre “Isla de la Cartuja”(cicCartuja), University of Seville-CSIC, Seville, Spain 12 Departamento de Bioquímica Vegetal y Biologia Molecular, Facultad de Biología, Universidad de Sevilla, Sevilla, Spain 13 Department of Chemistry, New York University, New York, New York, USA 14 Molecular Metabolism Lab, Center for Research in Molecular Medicine and Chronic Diseases (CIMUS), Av. Barcelona, Campus Vida, Santiago de Compostela University, Santiago de Compostela, Spain 15 Hepatology Program, Centro de Investigación Médica Aplicada (CIMA), Liver Unit, Clinica Universidad de Navarra (CUN), Instituto de Investigación de Navarra (IdisNA), University of Navarra, Pamplona, Spain 16 Division of Gastroenterology, Hepatology, and Nutrition, Center for Liver Diseases, Pittsburgh Liver Research Center, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania, USA 17 Gastroenterology and Hepatology Department, Marqués de Valdecilla University Hospital, Clinical and Translational Digestive Research Group, IDIVAL, Santander, Spain Abbreviations: ALD, alcohol-associated liver disease; CNNM4, Cyclin M4; ER, endoplasmic reticulum; PCMT1, L-isoaspartate O-methyltransferase; SAMe, S-adenosylmethionine; WT, wild type Supplemental Digital Content is available for this article. Direct URL citations are provided in the HTML and PDF versions of this article on the journal's website, www.hepjournal.com. Received: 23 March 2024 | Accepted: 17 October 2024 DOI: 10.1097/HEP.0000000000001156 388 | www.hepjournal.com Hepatology. 2025;82:388–404 Abstract Background and Aims: Alcohol-associated liver disease (ALD) is a leading cause of liver-related mortality worldwide, with limited treatment options beyond abstinence and liver transplantation. Chronic alcohol consumption has been linked to magnesium (Mg 2+ ) deficiency, which can influence liver disease progression. The mechanisms underlying Mg 2+ homeostasis dysregulation in ALD remain elusive. This study aimed to investigate the role of the Mg 2+ transporter Cyclin M4 (CNNM4) in ALD by analyzing its expression patterns in patients with ALD and preclinical animal models. Approach and Results: In this study, CNNM4 is upregulated in the liver of both patients with ALD and animal models. CNNM4 overexpression triggers Mg 2+ homeostasis dysregulation, linked to ALD progression. We propose a novel therapeutic approach for ALD treatment using N-acetylgalactosamine silencing RNA technology to specifically modulate Cnnm4 expression in the liver, improving mitochondrial function and alleviating endoplasmic reticulum stress. Notably, silencing Cnnm4 restores protein isoaspartyl methyltransferase (PCMT1) activity, essential for repairing ethanol-induced protein damage. Enhancing mitochondrial activity through Cnnm4-dependent mechanisms increases S-adenosylmethionine levels, crucial for PCMT1 function, highlighting the interconnected roles of mitochondrial health and protein homeostasis in ALD treatment. Conclusions: These findings shed light on the dysregulation of Mg 2+ homeostasis in ALD, providing a promising therapeutic approach targeting CNNM4. N-acetylgalactosamine siCnnm4 therapy boosts the repair processes of ethanol-damaged proteins through the upregulation of PCMT1 activity. Keywords: alcohol, ALD, hepatotoxicity, magnesium INTRODUCTION Alcohol-associated liver disease (ALD), one of the most common liver diseases worldwide,[1]accounts for up to 60%–80% of liver-related mortality in Europe,[2] being the first cause of liver transplantation in the Western world.[3]ALD encompasses a broad spectrum of liver diseases, ranging from steatosis to alcoholassociated hepatitis (AAH) and cirrhosis.[4]Approximately 2% of patients with cirrhosis develop primary HCC.[1] Hypomagnesemia has been extensively studied in individuals with ALD,[5]with lower serum Mg 2+ levels linked to higher alcohol concentrations.[6]Studies suggest that higher Mg 2+ intake may reduce the risk of death from liver diseases, especially in alcohol users and those with hepatic steatosis.[7]Herein, we studied the role of Mg 2+ homeostasis and the contribution of any Mg 2+ transporter in this cation homeostasis dysregulation in ALD. Simón et al[8]identified Cyclin M4 (CNNM4) as a key regulator of Mg 2+ homeostasis in fatty liver, and its ------------------------------------------------------------------------------------------------------- This is an open access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal. Copyright © 2024 The Author(s). Published by Wolters Kluwer Health, Inc. Correspondence María Luz Martínez-Chantar, Liver Disease Lab, Center for Cooperative Research in Biosciences (CIC bioGUNE), Basque Research, Derio, Spain. Email: [email protected] Luis Alfonso Martínez-Cruz, Technology Alliance (BRTA), Parque Científico Tecnológico de Bizkaia building 801A, 48160 Derio, Biscay Spain. Email: amartin[email protected] MODULATORY ROLE OF CNNM4 IN REGULATING PROTEIN-L-ISOASPARTYL METHYLTRANSFERASE ACTIVITY IN ALD | 389 3.0 (A) (C) (E) (F) (D) ** *** *** *** Cnnm1 Cnnm2 Cnnm3 Cnnm4 Trpm6 Trpm7 Magt1 Mmgt1 Mrs2 1 1 1 1 1 1 1 1 1 0.71 1.27 0.75 2.18 1.16 0.67 0.68 1 0.67 CNNM1 mRNA expression levels (fold-change) mRNA expression levels (fold-change) CNNM3 CNNM2 CNNM4 TRPM6 TRPM7 MAGT1 MMGT1 MRS2 Healthy 1 1 1 1 1 1 1 1 1 1.36 2.5 3 2 1 2.0 1.5 1.0 1.36 0.81 2.15 0.73 0.6 0.8 1.94 0.83 1.14 1.16 1.06 1.93 0.7 0.66 0.87 1.46 0.98 1.5 *1.41 *** ** *** * *** *** *** *** *** *** *** * ** *** *** *** ** 1.49 1.07 1.09 0.63 1.03 1.26 1.25 1.15 1.41 1.16 1.69 0.48 0.81 0.97 1.33 0.99 Mg2+ serum levels (mg/dl) 2.5 2.0 1.5 1.0 Healthy pre-cirrhosis compensated-cirrhosis decompensated-cirrhosis AH (B) 1.5 Mg2+ serum levels (fold-change) 1.0 0.5 0.0 Ctrl EtOH * 50 CNNM4 (fold-change) 40 20 10 30 0 Ctrl EtOH *** Expl_AH SevAH Control 50 µm 50 µm Etanol Ctrl CNNM4 EtOH NonSev AH Early AH Healthy CNNM4 ALD patients 150 CNNM4 (fold-change) 100 50 0 Healthy ALD patients * 390 | HEPATOLOGY overexpression is linked to disease progression in NAFLD. González-Recio et al[9]identified that CNNM4 expression levels are induced in DILI, dysregulating Mg 2+ homeostasis by acting as an Mg 2+ extruder. Silencing Cnnm4 appeared as a new therapeutic option to ameliorate endoplasmic reticulum (ER) stress and mitochondrial dysfunction. ER stress is crucial in ALD since the accumulation of misfolded proteins alters essential biological functions.[10] Besides, nonenzymatic modification of asparagine and aspartate residues to atypical forms of isoaspartyl also disrupts protein homeostasis. This process activates L-isoaspartate O-methyltransferase (PCMT1), which specifically methylates the isoaspartyl residue formed by the spontaneous deamidation of asparagine using the molecule S-adenosylmethionine (SAMe). PCMT1 activity prevents damaged protein accumulation and regulates liver metabolism.[11]Modified protein accumulation with atypical isoaspartyl residues in ALD could be related to insufficient activity of PCMT1 and ER stress.[11]Herein, we showed that reduced PCMT1 levels in patients with ALD were linked to increased hepatic CNNM4 levels. Targeting Cnnm4 with RNA therapeutics in the mouse model of chronic and binge ethanol feeding (NIAAA model)[12]ameliorated liver injury, reduced ER stress, and restored PCMT1 activity, avoiding aberrant isoaspartyl modifications. Overall, our data show a link between Mg 2+ homeostasis mediated by CNNM4, and PCMT1 repair activity in both patients with ALD and preclinical models, suggesting a new therapeutic approach to treat patients with ALD. METHODS Human samples Samples from healthy donors Healthy human serum samples from nonobese donors without liver lesions (Marqués de Valdecilla University Hospital) were used to determine Mg 2+ levels (n =85), measure CNNM4 expression levels in liver biopsies (n =5), and measure PCMT1 serum levels (n =5) to compare with patients with ALD. The study was approved by the Research Ethics Committee of IDIVAL Cantabria. Samples from patients with ALD Serum samples from patients with different stages of ALD: 18 patients with pre-cirrhotic conditions defined by any degree of fibrosis on the METAVIR scale (F1–F3), 40 patients with compensated cirrhosis, 47 patients with decompensated cirrhosis, and 59 patients with alcoholassociated hepatitis (AH) (Hospital Clinic of Barcelona) were used for magnesium level determination. In a subgroup from this cohort (10 patients with pre-cirrhotic conditions, 9 with compensated cirrhosis, 11 with decompensated cirrhosis, and 10 with AH), serum samples were used to measure PCMT1 levels. Clinical characteristics are described in Supplemental Table S1, http://links.lww.com/HEP/J632. Moreover, derived liver samples for RNAseq studies were obtained from the Human Biorepository Core from the NIH-funded international InTeam consortium (7U01AA021908-05). Patients with early alcohol-associated steatohepatitis from Cliniques Universitaires Saint-Luc. All patients included gave written informed consent, and the research protocols were approved by the local Ethics Committees and by the central Institutional Review Board of the University of North Carolina at Chapel Hill. Patients were selected according to different clinically relevant stage groups: (1) patients with early AH, who were nonobese with high alcohol intake, and presented mild elevation of transaminases and histologic criteria of steatohepatitis (early AH, N =12); (2) patients with histologically confirmed AH who were biopsied before any treatment (AH, N = 18), and (3) explants from patients with AH who underwent early transplantation following a well-defined protocol[13](exAH, N =11). These groups were compared with fragments of nondiseased human livers (N =10). Patients with malignancies were excluded from the study. Clinical characteristics are described in Supplemental Table S2, http://links.lww.com/HEP/J632. The study was approved by the Ethics Committee of the Medical University of Graz and performed in accordance with the Declaration of Helsinki. Indeed, 21 samples from FIGURE 1 Disturbances of Mg 2+ homeostasis and hepatic CNNM4 overexpression in ALD. (A) Mg 2+ serum levels in patients with ALD: precirrhosis (n =20), compensated cirrhosis (n =42), decompensated cirrhosis (n =47), and AH (n =59), and compared with healthy donors (n =85). (B) Mg 2+ serum levels in mice fed with the NIAAA alcohol feeding model and compared to controls. (C) mRNA expression levels of the mammalian magnesium transporters in patients with ALD: early AH (n =12), nonsevere AH (NonSev AH) (n =11), severe AH (SevAH) (n =15), and AH explants (Expl_AH) (n =11) and compared to healthy donors (n =10). (D) mRNA expression levels of the mammalian magnesium transporters mice fed with the NIAAA alcohol feeding model (n =5) and compared to controls (n =5). (E) Liver IHC of CNNM4 in mice fed with the NIAAA alcohol feeding model (n =5) and compared to controls (n =5). The fold-change was calculated by comparing the EtOH versus the control group. (F) Liver IHC of CNNM4 in patients with ALD (n =21) and compared to healthy donors (n =5). The fold-change was calculated by comparing patients with ALD versus the Healthy group. Data are shown as mean ±SEM. *p<0.05; **p<0.01; ***p<0.001 (Student test). The scale bar corresponds to 50 µm. Abbreviations: AH, alcohol-associated hepatitis; ALD, alcohol-associated liver disease; CNNM4, Cyclin M4; IHC, immunohistochemistry. MODULATORY ROLE OF CNNM4 IN REGULATING PROTEIN-L-ISOASPARTYL METHYLTRANSFERASE ACTIVITY IN ALD | 391 explants undergoing urgent liver transplantation for ALD were used for immunohistochemical staining for CNNM4 expression (Marqués de Valdecilla University Hospital). A detailed characterization is summarized in Supplemental Table S3, http://links.lww.com/HEP/J632. All patients included gave written informed consent. The study was approved by the Research Ethics Committee of IDIVAL Cantabria. Animal model for ALD Male adult (3-month-old) C57BL/6 mice were used as wild-type (WT) mice and were treated according to the mouse model of chronic and binge ethanol feeding (the NIAAA model).[12] RESULTS Disturbances of Mg 2+ homeostasis and hepatic CNNM4 overexpression in ALD Hypomagnesemia was evaluated in a cohort of patients with ALD across varying stages: pre-cirrhosis, compensated cirrhosis, decompensated cirrhosis, and AH. These levels were compared with a healthy control group. A notable decline in serum Mg 2+ concentrations was observed at all disease stages (Figure 1A, Supplemental Table S1, http://links.lww.com/HEP/J632), corroborating that Mg 2+ homeostasis is compromised in ALD.[14]The NIAAA mouse model,[12]which replicates human ALD pathology, also exhibited reduced serum magnesium levels compared to control-fed mice (Figure 1B). To understand whether any mammalian Mg 2+ transporter was modulated in ALD, we analyze their mRNA expression in liver samples from a cohort of patients with ALD at different disease stages: early AH, nonsevere AH, severe AH, and AH explants (Figure 1C, Supplemental Table S2, http://links.lww.com/HEP/ J632). A significant upregulation of membrane magnesium transporter 1 (MMGT1) and Cyclin M4 (CNNM4) levels was correlated with the ALD severity. Subsequent analysis in the NIAAA model (Figure 1D) revealed an exclusive elevation of Cnnm4. CNNM4 upregulation may contribute to the vulnerability to hepatic damage in ALD. Consistent with this, isolated hepatocytes from WT animals exposed to EtOH exhibited increased Cnnm4 mRNA levels (Supplemental Figure S1A, http://links.lww.com/HEP/J632) while no changes were identified in KCs (Supplemental Figure S1B, http://links. lww.com/HEP/J632). Histological and western blot analysis in the liver of mice fed the NIAAA model revealed enhanced CNNM4 expression in ethanol-fed mice compared to controls (Figure 1E, Supplemental Figure S1C, http://links.lww. com/HEP/J632). Importantly, in liver biopsies from ALDdiagnosed patients, immunohistochemistry staining demonstrated a significantly elevated CNNM4 level in patients with chronic ALD relative to healthy individuals (Figure 1F, Supplemental Table S3, http://links.lww. com/HEP/J632). Notably, CNNM4 expression was higher in liver biopsies of patients who continued alcohol consumption compared to those who abstained (Supplemental Figure S1D, http://links.lww.com/HEP/ J632, Supplemental Table S3, http://links.lww.com/ HEP/J632). However, no significant expression differences were discernible between lean and obese patients with ALD (Supplemental Figure S1E, http:// links.lww.com/HEP/J632, Supplemental Table S3, http://links.lww.com/HEP/J632). Finally, a positive correlation was established between CNNM4 levels and the ALT enzyme in the patient cohort (Supplemental Figure S1F, http://links.lww.com/HEP/J632, Supplemental Table S3, http://links.lww.com/HEP/J632). These findings illustrate that hepatic CNNM4 expression is significantly elevated in ALD and is, intriguingly, associated with the disease progression in both clinical and preclinical models. The absence of Cnnm4 ameliorates the hepatotoxicity caused by ALD in primary hepatocytes Given that CNNM4 is overexpressed in ALD and that this pathology is characterized by cell death, Cnnm4 was silenced in primary hepatocytes treated with EtOH for 12, 24, and 36 hours. Cell death was prevented in Cnnm4-silenced hepatocytes, as shown by the TUNEL assay (Figure 2A, Supplemental Figure S2A, http:// links.lww.com/HEP/J632). N-acetylgalactosamine (GalNAc) silencing RNA was employed for the effective silencing of Cnnm4[8,9](Supplemental Figure S3A, http://links.lww.com/HEP/J632). Considering the important role of Mg 2+ in this pathology, we studied if Mg 2+ could mitigate ethanolrelated liver damage. Thus, hepatocytes were administered 5 mM and 20 mM Mg 2+ with EtOH for 12, 24, and 36 hours; however, the TUNEL assay indicated no protective effect (Supplementary Figure 3B, http://links. lww.com/HEP/J632). Indeed, ethanol-induced oxidative stress and ROS production were unaffected by CNNM4 overexpression but significantly reduced by siCnnm4 treatment during 12, 24, and 36 hours (Figure 2B, Supplemental Figure S3C, http://links.lww.com/HEP/ J632), which also normalized ATP levels and mitochondrial membrane potential for 12 hours (Supplemental Figures S3D, E, http://links.lww.com/HEP/J632). CNNM4 acts as a Mg 2+ extruder in the liver in other pathologies.[8,9]Thus, we investigated intracellular and extracellular Mg 2+ levels in primary hepatocytes treated with EtOH for 12 hours and GalNAc siCnnm4. The 392 | HEPATOLOGY relative intracellular Mg 2+ levels were determined using mitochondrial (Mag-S-TPP-AM)[8,15]and nontargeted (Mg-S-AM)[8]labeling (Figure 2C). Mitochondrial and cytosolic Mg 2+ levels were reduced under EtOH, and GalNAc siCnnm4 enabled the recovery of mitochondrial Mg 2+ levels (Figure 2C). Indeed, GalNAc siCnnm4 restores extracellular Mg 2+ to normal levels (Figure 2D), suggesting that CNNM4 acts as an Mg 2+ extruder in the liver under ethanol. Nonetheless, ER Ca 2+ release capacity was evaluated in primary hepatocytes with fura-2, using the Grynkiewicz method.[16]Ca 2+ release under thapsigargin TUNEL (% positive nuclei) 10 12h 24h 12h Ctrl TUNEL EtOH EtOH+GaINAc 8 4 2 6 0 Ctrl EtOH EtOH+GaINAc *** *** Mitochondrial Mg2+ levels (nM) 20 15 5 10 0 Ctrl EtOH EtOH+GaINAc Citosolic Mg2+ levels (nM) 6 4 2 0 Ctrl EtOH EtOH+GaINAc TUNEL (% positive nuclei) 50 40 20 10 30 0 Ctrl EtOH EtOH+GaINAc *** *** ** ** * * * 36h TUNEL (% positive nuclei) 35 30 20 15 5 10 25 0 Ctrl EtOH EtOH+GaINAc ** *** (A) MitoSOX (fold change) *** *** ** *** 2.5 12h 2.0 1.0 0.5 1.5 0.0 Ctrl EtOH EtOH+GaINAc Ctrl EtOH EtOH+GaINAc Ctrl EtOH EtOH+GaINAc 24h 36h 12h 24h 36h (B) (C) 24h Ctrl TUNEL EtOH EtOH+GaINAc 36h Ctrl TUNEL EtOH EtOH+GaINAc 50 µm 50 µm 50 µm Extracellular Mg2+ levels (fold-change) 1.2 1.0 0.2 0.6 0.4 0.8 0.0 Ctrl EtOH EtOH+GaINAc (D) * ER Ca2+ release capacity with thapsigargin (a.u.) 300 Ctrl EtOH EtOH+GaINAc 200 100 0 Ctrl Representative ER Ca2+ release capacity with Thapsigargin (a.u) 300 200 100 0 0 50,016 100,016 150,062 200,078 EtOH EtOH+GaINAc * *** * ** * ER tracker (fold change) 20 15 10 5 0 t (s) Ctrl EtOH GaINAc Ctrl EtOH GaINAc Ctrl EtOH GaINAc (E) (F) FIGURE 2 The absence of Cnnm4 ameliorates the hepatotoxicity caused by ALD in primary hepatocytes. (A, B) In WT primary hepatocytes under EtOH for 12, 24, and 36 hours and treated with GalNAc siCnnm4 or an unrelated control compared to a Ctrl: (A) cell death by TUNEL; (B) the mitochondrial ROS by MitoSOX. (C–E) In WT primary hepatocytes under EtOH for 12 hours and treated with GalNAc siCnnm4 or an unrelated control compared to a Ctrl: (C) Relative intracellular Mg 2+ determination by mitochondrial-specific labeling and cytosolic-specific labeling; (D) extracellular Mg 2+ levels; (E) Ca 2+ release capacity by ER with thapsigargin; (F) ER tracker red staining in WT primary hepatocytes under EtOH for 12, 24, and 36 hours and treated with GalNAc siCnnm4 and compared with a Ctrl. Data are shown as mean ±SEM. *p<0.05; **p<0.01; ***p <0.001 (Student test). Quadrupled were used for experimental conditions. Abbreviations: ALD, alcohol-associated liver disease; CNNM4, Cyclin M4; Ctrl, control group; GalNAc, N-acetylgalactosamine; WT, wild type. MODULATORY ROLE OF CNNM4 IN REGULATING PROTEIN-L-ISOASPARTYL METHYLTRANSFERASE ACTIVITY IN ALD | 393 Relative mRNA expression Cnnm4 (fold-change) 5 4 2 1 3 0 Ctrl EtOH EtOH+GaINAc * *** *** ** ** ** ** *** *** *** *** *** *** *** ** CNNM4 (fold-change) 70 60 50 40 20 10 30 0 Ctrl EtOH EtOH+GaINAc *** ** Mg2+ serum levels (fold-change) 2.0 1.5 1.0 0.5 0.0 Ctrl EtOH EtOH+GaINAc Mg2+ urine levels (fold-change) 1.5 1.0 0.5 0.0 Ctrl EtOH EtOH+GaINAc Sudan Red (fold-change) 50 40 20 30 10 0 EtOH Ctrl EtOH+GaINAc HNE (fold-change) 4 3 2 1 0 EtOH Ctrl EtOH+GaINAc F4/80 (fold-change) 30 20 10 0 EtOH Ctrl EtOH+GaINAc Cell apoptosis (fold-change) 15 10 5 0 EtOH Ctrl EtOH+GaINAc CNNM4 protein expression normalized (a.u.) 2.5 2.0 1.0 1.5 0.5 0.0 ER Ctrl EtOH Ctrl EtOH (A) (C) (E) (F) (D) (B) Ctrl CNNM4 EtOH EtOH + GaINAc 100 µm Ctrl Sudan Red EtOH Control EtOH 90kDa 90kDa 90kDa 37kDa ER CNNM4 CANX EtOH + GaINAc HNEF4/80 CELL APOPTOSIS 100 µm 100 µm 100 µm 100 µm Control EtOH Mitochondria Mitochondria CNNM4 VDAC 394 | HEPATOLOGY stimuli was impaired in cells treated with EtOH for 12 hours but restored by GalNAc siCnnm4 (Figure 2E). EtOH stimulated ATF4 expression and phosphorylation of EIF2alpha, whereas their expressions were reduced under GalNAc siCnnm4 (Supplemental Figure S4A, http://links.lww.com/HEP/J632). In this sense, GalNAc siCnnm4 decreased the ER stress markers expression, such as Atf6 and Chop, at different times and tended to decrease Grp78 levels (Supplemental Figure S4B, http:// links.lww.com/HEP/J632). To further investigate the ER perturbations associated with alcohol, we labeled EtOHtreated primary hepatocytes with ER tracker staining for 12, 24, and 36 hours. Fluorescence images showed a reduced number of red-stained cells upon GalNAc siCnnm4 (Figure 2F, Supplemental Figure S4C, http:// links.lww.com/HEP/J632), while CNNM4 overexpression did not decrease the staining cells (Supplemental Figure S4D, http://links.lww.com/HEP/J632). Finally, we evaluated the aggresomes of misfolded proteins in primary hepatocytes under EtOH for 12 hours (Supplemental Figure S4E, http://links.lww.com/HEP/J632). GalNAc siCnnm4 significantly reduces misfolded proteins associated with EtOH damage (Supplemental Figure S4E, http://links.lww.com/HEP/J632). Overall, targeting Cnnm4 by GalNAc siCnnm4 ameliorates alcohol-induced cell death, mitochondrial dysfunction, and ER stress in liver cells and restores mitochondrial and extracellular Mg 2+ levels to basal. Targeting Cnnm4 in preclinical ALD mouse models reduces lipid accumulation and inflammation response In the NIAAA model,[12]GalNAc siCnnm4 was used to investigate its potential benefit. Cnnm4 was the only Mg 2+ transporter overexpressed in the liver of ALD mice, and it was specifically silenced by GalNAc siCnnm4 (Figures 3A, B) (Supplemental Figures S5A, B, http://links.lww.com/HEP/J632). Reduced Cnnm4 expression was associated with less liver injury as measured by ALT and AST levels (Supplemental Figure S5C, http://links.lww.com/HEP/J632). Since alcohol consumption and severity are linked to a decrease in Mg 2+ content in the liver,[17]serum Mg 2+ concentration was significantly decreased in mice fed a NIAAA diet, whereas GalNAc siCnnm4 treatment returned it to near normal levels (Figure 3C). In contrast, urine Mg 2+ content increased in mice after EtOH, but it was also restored by siCnnm4 (Figure 3D). These findings imply that CNNM4 modulated Mg 2+ homeostasis disrupted by alcohol-induced liver injury. Moreover, mice fed NIAAA diets had a decrease in serum triglycerides (Supplemental Figure S5D, http:// links.lww.com/HEP/J632), but an increase in liver lipid content by Sudan Red (Figure 3E). Importantly, silencing Cnnm4 normalized these parameters (Supplemental Figure S5D, http://links.lww.com/HEP/J632 and Figure 3E). Furthermore, siCnnm4 reversed lipid peroxidation activation in mice fed the NIAAA diet by 4hydroxy-2-nonenal staining, accompanied by a decreased inflammatory response by F4/80 staining (Figure 3E) and reduced mRNA expression of inflammatory markers (Supplemental Figure S5E, http://links. lww.com/HEP/J632). In addition, myeloperoxidase activity, linked to neutrophil activation,[18]was reduced with GalNAc siCnnm4 in mice fed with the NIAAA alcohol feeding model (Supplemental Figure S5F, http:// links.lww.com/HEP/J632). Cell death was lower in the liver of GalNAc siCnnm4–treated mice assessed by ApopTag Peroxidase In Situ Apoptosis Detection Kit (Figure 3E). Moreover, no changes were observed in the intestinal permeability of mice on the NIAAA diet in the siCnnm4 group compared to the ethanol group, confirming that the GalNAc silencing RNA therapy did not alter the intestinal permeability caused by ethanol (Supplemental Figure S5G, http://links.lww.com/HEP/ J632). In addition, there are certain pathways responsible for the excessive production of ROS in acute and chronic ethanol-induced injury.[19]CYP2E1 protein levels remained upregulated in the liver of mice fed with the NIAAA alcohol feeding model w/wo GalNAc siCnnm4 (Supplemental Figure S6A, http://links.lww. com/HEP/J632). However, CYP2E1 activity decreased markedly (Supplemental Figure S6B, http://links.lww. com/HEP/J632) in primary hepatocytes treated with EtOH and upon GalNAc siCnnm4. NADP + /NADPH ratio increased under EtOH in primary hepatocytes (Supplemental Figure S6C, http://links.lww.com/HEP/J632). Indeed, the NAD + /NADH ratio was restored under GalNAc siCnnm4 in primary hepatocytes (Supplemental Figure S6D, http://links.lww.com/HEP/J632). Regarding acetaldehyde pathway, GalNAc siCnnm4 was able to FIGURE 3 Targeting Cnnm4 in preclinical ALD mice models reduces lipid accumulation and inflammation response. (A–F) WT mice treated by NIAAA model. (A) mRNA levels of Cnnm4. (B) Liver IHC of CNNM4. The fold-change was calculated by comparing EtOH and EtOH + GalNAc groups versus the control group. (C) Magnesium levels in serum. (D) Magnesium levels in urine. (E) Lipid content assessed by Sudan red staining in the liver. Lipid peroxidation by HNE staining in the liver. Inflammation by F4/80 staining in liver sections. Cell death by ApopTag assay in liver sections. (F) Cellular location of CNNM4 in liver tissue. Protein expression was measured in mitochondria and ER locations. VDAC and CANX were used as a loading control. Data are shown as mean ±SEM. *p<0.05; **p<0.01; ***p<0.001 (Student test). The scale bar corresponds to 50 µm. Abbreviations: CANX, Calnexin; Cnnm4, Cyclin M4; GalNAc, N-acetylgalactosamine; HNE, 4-hydroxy-2-nonenal; IHC, immunohistochemistry; VDAC, voltage-dependent anion channel; WT, wild type. MODULATORY ROLE OF CNNM4 IN REGULATING PROTEIN-L-ISOASPARTYL METHYLTRANSFERASE ACTIVITY IN ALD | 395 (A) (B) (D) Translation Gene ontology -log (P value) Keggs Pathways (%) Metabolic pathways Insulin signaling pathway Glutathione metabolism Drug metabolism-cytochrome P450 membrane cytoplasm extracellular exosome mitochondrion cytosol endoplasmic reticulum membrane basolateral plasma membrane intracellular membranebounded organelle membrane cytoplasm extracellular exosome mitochondrion mitochondrial inner membrane nucleus cytosol Inflammatory mediator regulation of TRP channels EtOH vs EtOH+GalNAc EtOH EtOH+GalNAc Positive regulation of apoptotic process Positive regulation of protein dephosphorylation Cytoplasmic translation 02468 CP238 20 18 16 14 12 PCMT1 PDS5A DAG1 TFR1 PTPRF SYAM HEXB NU4M RCC2 CXB2 ASM3A CO4A1 MK01 SAP3 VAC14 WDR82 KAPCA CCD25 PICAL NDRG1 TTL12 PEDF FLOT1 KC1A CP26A GP180 TIM29 MIPEP FA98B IMDH2 URAD RBM25 ALDR CATB FKBP5 ADCK5 LC7L3 CA050 XRCC6 RT06 PTN11 P85A CDD VATB2 GSTM3 THIOM ENPP1 CP4AA TMX3 Oxidation-reduction process Gene ontology -log (P value) EtOH + GalNAc vs EtOH steroid metabolic process cellular response to oxidative stress S-adenosylmethionine metabolic process lipid metabolic process cellular response to drug response to endoplasmic reticulum stress protein folding 02468 (C) Cellular component (%) EtOH GalNAc vs EtOH Cell component (%) EtOH vs EtOH GalNAc FIGURE 4 Targeting Cnnm4 in the NIAAA model induced a repair response in the liver. (A–D) Proteomics analysis by LC-MS/MS in the liver of mice fed with the NIAAA alcohol feeding model. (A) GO process analysis for the regulated genes in EtOH versus EtOH + GalNAc and EtOH + GalNAc versus EtOH. (B) Kegg pathways in GalNAc siCnnm4 upon EtOH. (C) Cellular component in EtOH + GalNAc versus EtOH and EtOH versus EtOH + GalNAc. (D) The top 50 upregulated and downregulated specific proteins regulated by GalNAc siCnnm4. Abbreviations: Cnnm4, Cyclin M4; GalNAc, N-acetylgalactosamine; GO, gene ontology; LC-MS/MS, liquid chromatography tandem mass spectrometry. 396 | HEPATOLOGY CONFLICTS OF INTEREST Ute Schaeper owns stock in and is employed by Silence Therapeutics. Ramón Bataller consults for Novo Nordisk, GSK, and Boehringer Ingelheim. He is on the speakers’bureau for AbbVie and Gilead. Javier Crespo consults for and received grants from Gilead and AbbVie. He consults for BMS. He received grants from MSD. The remaining authors have no conflicts to report. ORCID Irene González-Recio https://orcid.org/0000–0003– 1432–8974 Naroa Goikoetxea-Usandizaga https://orcid.org/ 0000–0003–4569–028X Claudia M. Rejano-Gordillo https://orcid.org/0000– 0003–4586–5975 Carolina Conter https://orcid.org/0000–0002–5999– 8667 Rubén Rodríguez Agudo https://orcid.org/0000– 0001–5731–5052 Marina Serrano-Maciá https://orcid.org/0000–0003– 4183–6384 Leidy Estefanía Zapata-Pavas https://orcid.org/0009– 0003–3433–7699 Patricia Peña-Sanfélix https://orcid.org/0009–0004– 4227–9633 Mikel Azkargorta https://orcid.org/0000–0001–9115– 3202 Félix Elortza https://orcid.org/0000–0001–8839–5438 Álex Guillamon Thiery https://orcid.org/0009–0009– 6903–8633 Armando Raúl Guerra-Ruiz https://orcid.org/0000– 0001–8896–8611 Ramiro Jover https://orcid.org/0000–0002–4914– 5804 Unai Galicia-Garcia https://orcid.org/0000–0003– 1480–3261 César Martín https://orcid.org/0000–0002–4087– 8729 Ute Schaeper https://orcid.org/0000–0003–2556– 8391 Teresa C. Delgado https://orcid.org/0000–0001– 9204–581X Irene Díaz-Moreno https://orcid.org/0000–0002– 5318–7644 Antonio Díaz Quintana https://orcid.org/0000–0001– 8973–8009 Daniela Buccella https://orcid.org/0000–0002–2266– 9026 Rubén Nogueiras https://orcid.org/0000–0002–9976– 9930 JosepMaria Argemi https://orcid.org/0000–0003– 1696–7753 Matías A. Ávila https://orcid.org/0000–0001–6570– 3557 Jordi Gratacós-Ginès https://orcid.org/0000–0003– 1181–7238 Paula Iruzubieta https://orcid.org/0000–0001–9476– 1801 Elisa Pose https://orcid.org/0000–0002–0224–7686 Ramón Bataller https://orcid.org/0000–0002–1119– 7799 Javier Crespo https://orcid.org/0000–0001–8248– 0172 Luis Alfonso Martínez-Cruz https://orcid.org/0000– 0002–5856–9377 Maria Luz Martínez-Chantar https://orcid.org/0000– 0002–6446–9911 REFERENCES 1. Avila MA, Dufour JF, Gerbes AL, Zoulim F, Bataller R, Burra P, et al. Recent advances in alcohol-related liver disease (ALD): Summary of a Gut round table meeting. Gut. 2020;69:764–80. 2. Benedé-Ubieto R, Estévez-Vázquez O, Guo F, Chen C, Singh Y, Nakaya HI, et al. An experimental DUAL model of advanced liver damage. Hepatol Commun. 2021;5:1051–68. 3. Terrault NA, Francoz C, Berenguer M, Charlton M, Heimbach J. Liver transplantation 2023: Status report, current and future challenges. Clin Gastroenterol Hepatol. 2023;21:2150–66. 4. Mitra S, De A, Chowdhury A. Epidemiology of non-alcoholic and alcoholic fatty liver diseases. Transl Gastroenterol Hepatol. 2020; 5:16. 5. Winrich EJ, Tiwari H, Gala KS, Royer AJ, Parajuli D, Vatsalya V. Characterization of hypomagnesemia in alcoholic hepatitis patients and its association with liver injury and severity markers. J Clin Med. 2023;12:2968. 6. Petroianu A, Barquete J, De Almeida Plentz EG, Bastos CMC, Maia DJ. Acute effects of alcohol ingestion on the human serum concentrations of calcium and magnesium. J Int Med Res. 1991; 19:410–3. 7. Wu G, Yang Q, Yu Y, Lin S, Feng Y, Lv Q, et al. Taurine inhibits Kupffer cells activation induced by lipopolysaccharide in alcoholic liver damaged rats. Adv Exp Med Biol. 2017;975 pt 2: 789–800. 8. Simón J, Goikoetxea-Usandizaga N, Serrano-Maciá M, Fernández-Ramos D, Sáenz de Urturi D, Gruskos JJ, et al. Magnesium accumulation upon cyclin M4 silencing activates microsomal triglyceride transfer protein improving NASH. J Hepatol. 2021;75: 34–45. 9. González-Recio I, Simón J, Goikoetxea-Usandizaga N, SerranoMaciá M, Mercado-Gómez M, Rodríguez-Agudo R, et al. Restoring cellular magnesium balance through Cyclin M4 protects against acetaminophen-induced liver damage. Nat Commun. 2022;13:6816. 10. Xia SW, Wang ZM, Sun SM, Su Y, Li ZH, Shao JJ, et al. Endoplasmic reticulum stress and protein degradation in chronic liver disease. Pharmacol Res. 2020;161:105218. 11. Kharbanda KK, Mailliard ME, Baldwin CR, Sorrell MF, Tuma DJ. Accumulation of proteins bearing atypical isoaspartyl residues in livers of alcohol-fed rats is prevented by betaine administration: Effects on protein-L-isoaspartyl methyltransferase activity. J Hepatol. 2007;46:1119–25. 12. Bertola A, Mathews S, Ki SH, Wang H, Gao B. Mouse model of chronic and binge ethanol feeding (the NIAAA model). Nat Protoc. 2013;8:627–37. 13. Mathurin P. Early liver transplantation for acute alcoholic hepatitis: We can’t say no. J Hepatol. 2021;75:718–22. 14. Liang Y, Harris FL, Brown LAS. Alcohol induced mitochondrial oxidative stress and alveolar macrophage dysfunction. Biomed Res Int. 2014;2014:371593. 15. Ostapowicz G, Fontana RJ, Schiødt FV, Larson A, Davern TJ, Han SH, et al. Results of a prospective study of acute liver failure MODULATORY ROLE OF CNNM4 IN REGULATING PROTEIN-L-ISOASPARTYL METHYLTRANSFERASE ACTIVITY IN ALD | 403 at 17 tertiary care centers in the United States. Ann Intern Med. 2002;137:947–54. 16. Grynkiewicz G, Poenie M, Tsien RY. A new generation of Ca2+ indicators with greatly improved fluorescence properties. J Biol Chem. 1985;260:3440–50. 17. Rink EB. Magnesium deficiency in alcoholism. Alcohol Clin Exp Res. 1986;10:590–4. 18. Tjondro HC, Ugonotti J, Kawahara R, Chatterjee S, Loke I, Chen S, et al. Hyper-truncated Asn355and Asn391-glycans modulate the activity of neutrophil granule myeloperoxidase. J Biol Chem. 2021;296:100144. 19. Cederbaum AI. CYP2E1 - Biochemical and toxicological aspects and role in alcohol-induced liver injury. Mt Sinai J Med. 2006;73: 657–72. 20. Gilbert JM, Fowler A, Bleibaum J, Clarke S. Purification of homologous protein carboxyl methyltransferase isozymes from human and bovine erythrocytes. Biochemistry. 1988;27:5227–33. 21. Rodríguez-Agudo R, Goikoetxea-Usandizaga N, Serrano-Maciá M, Fernández-Tussy P, Fernández-Ramos D, Lachiondo-Ortega S, et al. Methionine Cycle Rewiring by Targeting miR-873-5p Modulates Ammonia Metabolism to Protect the Liver from Acetaminophen. Antioxidants (Basel). 2022;11:897. 22. Gao B, Bataller R. Alcoholic liver disease: Pathogenesis and new therapeutic targets. Gastroenterology. 2011;141: 1572–85. 23. Grochowski C, Blicharska E, Baj J, Mierzwińska A, Brzozowska K, Forma A, et al. Serum iron, magnesium, copper, and manganese levels in alcoholism: A systematic review. Molecules. 2019;24:1361. 24. Rylander R, Mégevand Y, Lasserre B, Amstutz W, Granbom S. Moderate alcohol consumption and urinary excretion of magnesium and calcium. Scand J Clin Lab Invest. 2001;61:401–5. 25. Seitz HK, Meier P. The role of acetaldehyde in upper digestive tract cancer in alcoholics. Transl Res. 2007;149:293–7. 26. Malhi H, Kaufman RJ. Endoplasmic reticulum stress in liver disease. J Hepatol. 2011;54:795–809. 27. Altamura S, Schaeper U, Dames S, Löffler K, Eisermann M, Frauendorf C, et al. SLN124, a GalNAc-siRNA conjugate targeting TMPRSS6, efficiently prevents iron overload in hereditary haemochromatosis type 1. HemaSphere. 2019;3:6–10. 28. Lu Y, Cederbaum AI. CYP2E1 and oxidative liver injury by alcohol. Free Radic Biol Med. 2008;44:723–38. 29. Lu SC, Martínez-Chantar ML, Mato JM. Methionine adenosyltransferase and S-adenosylmethionine in alcoholic liver disease. J Gastroenterol Hepatol. 2006;21 suppl 3:S61–4. How to cite this article: González-Recio I, Goikoetxea-Usandizaga N, Rejano-Gordillo CM, Conter C, Rodríguez Agudo R, Serrano-Maciá M, et al. Modulatory effects of CNNM4 on protein-Lisoaspartyl-O-methyltransferase repair function during alcohol-induced hepatic damage. Hepatology. 2025;82:388–404. https://doi.org/10.1097/ HEP.0000000000001156 404 | HEPATOLOGY