Portal de Periódicos da CAPES

Hepatocyte-specific perturbation of NAD+ biosynthetic pathways in mice induces reversible nonalcoholic steatohepatitis–like phenotypes

2021; Elsevier BV; Volume: 297; Issue: 6 Linguagem: Inglês

10.1016/j.jbc.2021.101388

1083-351X

Morten Dall, Anna S. Hassing, Lili Niu, Thomas S. Nielsen, Lars R. Ingerslev, Karolina Sulek, Samuel A.J. Trammell, Matthew P. Gillum, Romain Barrès, Steen Larsen, Steen Seier Poulsen, Matthias Mann, Cathrine Ørskov, Jonas T. Treebak,

Pancreatic function and diabetes

Nicotinamide phosphoribosyltransferase (NAMPT) converts nicotinamide to NAD+. As low hepatic NAD+ levels have been linked to the development of nonalcoholic fatty liver disease, we hypothesized that ablation of hepatic Nampt would affect susceptibility to liver injury in response to diet-induced metabolic stress. Following 3 weeks on a low-methionine and choline-free 60% high-fat diet, hepatocyte-specific Nampt knockout (HNKO) mice accumulated less triglyceride than WT littermates but had increased histological scores for liver inflammation, necrosis, and fibrosis. Surprisingly, liver injury was also observed in HNKO mice on the purified control diet. This HNKO phenotype was associated with decreased abundance of mitochondrial proteins, especially proteins involved in oxidoreductase activity. High-resolution respirometry revealed lower respiratory capacity in purified control diet–fed HNKO liver. In addition, fibrotic area in HNKO liver sections correlated negatively with hepatic NAD+, and liver injury was prevented by supplementation with NAD+ precursors nicotinamide riboside and nicotinic acid. MS-based proteomic analysis revealed that nicotinamide riboside supplementation rescued hepatic levels of oxidoreductase and OXPHOS proteins. Finally, single-nucleus RNA-Seq showed that transcriptional changes in the HNKO liver mainly occurred in hepatocytes, and changes in the hepatocyte transcriptome were associated with liver necrosis. In conclusion, HNKO livers have reduced respiratory capacity, decreased abundance of mitochondrial proteins, and are susceptible to fibrosis because of low NAD+ levels. Our data suggest a critical threshold level of hepatic NAD+ that determines the predisposition to liver injury and supports that NAD+ precursor supplementation can prevent liver injury and nonalcoholic fatty liver disease progression. Nicotinamide phosphoribosyltransferase (NAMPT) converts nicotinamide to NAD+. As low hepatic NAD+ levels have been linked to the development of nonalcoholic fatty liver disease, we hypothesized that ablation of hepatic Nampt would affect susceptibility to liver injury in response to diet-induced metabolic stress. Following 3 weeks on a low-methionine and choline-free 60% high-fat diet, hepatocyte-specific Nampt knockout (HNKO) mice accumulated less triglyceride than WT littermates but had increased histological scores for liver inflammation, necrosis, and fibrosis. Surprisingly, liver injury was also observed in HNKO mice on the purified control diet. This HNKO phenotype was associated with decreased abundance of mitochondrial proteins, especially proteins involved in oxidoreductase activity. High-resolution respirometry revealed lower respiratory capacity in purified control diet–fed HNKO liver. In addition, fibrotic area in HNKO liver sections correlated negatively with hepatic NAD+, and liver injury was prevented by supplementation with NAD+ precursors nicotinamide riboside and nicotinic acid. MS-based proteomic analysis revealed that nicotinamide riboside supplementation rescued hepatic levels of oxidoreductase and OXPHOS proteins. Finally, single-nucleus RNA-Seq showed that transcriptional changes in the HNKO liver mainly occurred in hepatocytes, and changes in the hepatocyte transcriptome were associated with liver necrosis. In conclusion, HNKO livers have reduced respiratory capacity, decreased abundance of mitochondrial proteins, and are susceptible to fibrosis because of low NAD+ levels. Our data suggest a critical threshold level of hepatic NAD+ that determines the predisposition to liver injury and supports that NAD+ precursor supplementation can prevent liver injury and nonalcoholic fatty liver disease progression. Nonalcoholic fatty liver disease (NAFLD) is a spectrum of diseases ranging from simple steatosis to nonalcoholic steatohepatitis (NASH). It is a major challenge for global health care (1Loomba R. Sanyal A.J. The global NAFLD epidemic.Nat. Rev. Gastroenterol. Hepatol. 2013; 10: 686-690Crossref PubMed Scopus (1066) Google Scholar), and as the number of patients increases because of the ongoing obesity pandemic, we need to understand the molecular events that occur from steatosis to steatohepatitis to prevent progression to end-stage liver diseases. Hepatic NAD+ levels decrease in livers of obese rodents (2Gariani K. Menzies K.J. Ryu D. Wegner C.J. Wang X. Ropelle E.R. Moullan N. Zhang H. Perino A. Lemos V. Kim B. Park Y.K. Piersigilli A. Pham T.X. Yang Y. et al.Eliciting the mitochondrial unfolded protein response by nicotinamide adenine dinucleotide repletion reverses fatty liver disease in mice.Hepatology. 2016; 63: 1190-1204Crossref PubMed Scopus (201) Google Scholar, 3Choi S.-E. Fu T. Seok S. Kim D.-H. Yu E. Lee K.-W. Kang Y. Li X. Kemper B. Kemper J.K. Elevated microRNA-34a in obesity reduces NAD+ levels and SIRT1 activity by directly targeting NAMPT.Aging Cell. 2013; 12: 1062-1072Crossref PubMed Scopus (167) Google Scholar, 4Zhang Z.-F. Fan S.-H. Zheng Y.-L. Lu J. Wu D.-M. Shan Q. Hu B. Troxerutin improves hepatic lipid homeostasis by restoring NAD+-depletion-mediated dysfunction of lipin 1 signaling in high-fat diet-treated mice.Biochem. Pharmacol. 2014; 91: 74-86Crossref PubMed Scopus (54) Google Scholar, 5Uddin G.M. Youngson N.A. Doyle B.M. Sinclair D.A. Morris M.J. Nicotinamide mononucleotide (NMN) supplementation ameliorates the impact of maternal obesity in mice: Comparison with exercise.Sci. Rep. 2017; 7: 15063Crossref PubMed Scopus (35) Google Scholar, 6Zhou C.C. Yang X. Hua X. Liu J. Fan M.B. Li G.Q. Song J. Xu T.Y. Li Z.Y. Guan Y.F. Wang P. Miao C.Y. Hepatic NAD+ deficiency as a therapeutic target for non-alcoholic fatty liver disease in ageing.Br. J. Pharmacol. 2016; 173: 2352-2368Crossref PubMed Scopus (92) Google Scholar, 7Wang X. Zhang Z.-F. Zheng G.-H. Wang A.-M. Sun C.-H. Qin S.-P. Zhuang J. Lu J. Ma D.-F. Zheng Y.-L. The inhibitory effects of purple sweet potato color on hepatic inflammation is associated with restoration of NAD+ levels and attenuation of NLRP3 inflammasome activation in high-fat-diet-treated mice.Molecules. 2017; 22: 1315Crossref Scopus (31) Google Scholar), suggesting NAD+ metabolism is a potential target to prevent NAFLD progression. Treatment with NAD+ precursors such as nicotinamide riboside (NR) resolves steatosis and diet-induced liver damage in rodents (2Gariani K. Menzies K.J. Ryu D. Wegner C.J. Wang X. Ropelle E.R. Moullan N. Zhang H. Perino A. Lemos V. Kim B. Park Y.K. Piersigilli A. Pham T.X. Yang Y. et al.Eliciting the mitochondrial unfolded protein response by nicotinamide adenine dinucleotide repletion reverses fatty liver disease in mice.Hepatology. 2016; 63: 1190-1204Crossref PubMed Scopus (201) Google Scholar, 6Zhou C.C. Yang X. Hua X. Liu J. Fan M.B. Li G.Q. Song J. Xu T.Y. Li Z.Y. Guan Y.F. Wang P. Miao C.Y. Hepatic NAD+ deficiency as a therapeutic target for non-alcoholic fatty liver disease in ageing.Br. J. Pharmacol. 2016; 173: 2352-2368Crossref PubMed Scopus (92) Google Scholar, 8Canto C. Houtkooper R.H. Pirinen E. Youn D.Y. Oosterveer M.H. Cen Y. Fernandez-Marcos P.J. Yamamoto H. Andreux P.A. Cettour-Rose P. Gademann K. Rinsch C. Schoonjans K. Sauve A.A. Auwerx J. The NAD(+) precursor nicotinamide riboside enhances oxidative metabolism and protects against high-fat diet induced obesity.Cell Metab. 2012; 15: 838-847Abstract Full Text Full Text PDF PubMed Scopus (679) Google Scholar), and although more studies are needed, clinical evidence is in line with these observations (9Dollerup O.L. Christensen B. Svart M. Schmidt M.S. Sulek K. Ringgaard S. Stødkilde-Jørgensen H. Møller N. Brenner C. Treebak J.T. Jessen N. A randomized placebo-controlled clinical trial of nicotinamide riboside in obese men: Safety, insulin-sensitivity, and lipid-mobilizing effects.Am. J. Clin. Nutr. 2018; 108: 343-353Crossref PubMed Scopus (109) Google Scholar). Thus, improving liver NAD+ metabolism may have beneficial effects on liver health in humans. The majority of the NAD+ pool in the liver is derived from tryptophan via the de novo synthesis pathway and from nicotinamide (10Liu L. Su X. Quinn W.J. Hui S. Krukenberg K. Frederick D.W. Redpath P. Zhan L. Chellappa K. White E. Migaud M. Mitchison T.J. Baur J.A. Rabinowitz J.D. Quantitative analysis of NAD synthesis-breakdown fluxes.Cell Metab. 2018; 27: 1067-1080.e5Abstract Full Text Full Text PDF PubMed Scopus (166) Google Scholar). Nicotinamide phosphoribosyltransferase (NAMPT) converts nicotinamide and phosphoribosyl pyrophosphate to nicotinamide mononucleotide (NMN) and pyrophosphate (11Revollo J.R. Grimm A.A. Imai S. The NAD biosynthesis pathway mediated by nicotinamide phosphoribosyltransferase regulates Sir2 activity in mammalian cells.J. Biol. Chem. 2004; 279: 50754-50763Abstract Full Text Full Text PDF PubMed Scopus (729) Google Scholar). The NMN adenylyltransferases (NMNAT1–3) then catalyze the condensation of NMN with ATP to generate NAD+ (12Berger F. Lau C. Dahlmann M. Ziegler M. Subcellular compartmentation and differential catalytic properties of the three human nicotinamide mononucleotide adenylyltransferase isoforms.J. Biol. Chem. 2005; 280: 36334-36341Abstract Full Text Full Text PDF PubMed Scopus (355) Google Scholar). NMN can also be generated from NR through phosphorylation by NR kinases 1 and 2 (13Bieganowski P. Brenner C. Discoveries of nicotinamide riboside as a nutrient and conserved NRK genes establish a Preiss-Handler independent route to NAD+ in fungi and humans.Cell. 2004; 117: 495-502Abstract Full Text Full Text PDF PubMed Scopus (439) Google Scholar), and nicotinic acid (NA) can be converted to NAD+ through the Preiss–Handler pathway (14Preiss J. Handler P. Biosynthesis of diphosphopyridine nucleotide: I. Identification of intermediates.J. Biol. Chem. 1958; 233: 488-492Abstract Full Text PDF PubMed Google Scholar). A recent study showed that decreased NAMPT expression is associated with NASH in humans (15Liao S. He H. Zeng Y. Yang L. Liu Z. An Z. Zhang M. A nomogram for predicting metabolic steatohepatitis: The combination of NAMPT, RALGDS, GADD45B, FOSL2, RTP3, and RASD1.Open Med. 2021; 16: 773-785Crossref Scopus (1) Google Scholar). To investigate the causal relationship between impaired NAD+ metabolism and NAFLD, we generated hepatocyte-specific Nampt knockout (HNKO) mice (16Dall M. Trammell S.A.J. Asping M. Hassing A.S. Agerholm M. Vienberg S.G. Gillum M.P. Larsen S. Treebak J.T. Mitochondrial function in liver cells is resistant to perturbations in NAD + salvage capacity.J. Biol. Chem. 2019; 294: 13304-13326Abstract Full Text Full Text PDF PubMed Scopus (14) Google Scholar). We observed that HNKO mice have 50% reduced liver NAD+ content, but this did not affect hepatic fatty acid oxidation or respiratory capacity. This was unexpected, as a previous study reported that knockout of Nampt impaired fatty acid oxidation and decreased mitochondrial oxygen consumption (17Peek C.B. Affinati A.H. Ramsey K.M. Kuo H.-Y. Yu W. Sena L.A. Ilkayeva O. Marcheva B. Kobayashi Y. Omura C. Levine D.C. Bacsik D.J. Gius D. Newgard C.B. Goetzman E. et al.Circadian clock NAD+ cycle drives mitochondrial oxidative metabolism in mice.Science. 2013; 342: 1243417Crossref PubMed Scopus (417) Google Scholar). Other studies suggest that manipulation of NAMPT renders the liver more susceptible toward hepatic lipid accumulation, as both treatment with the NAMPT inhibitor FK866 and overexpression of a dominant-negative Nampt mutation induces susceptibility to steatosis development (6Zhou C.C. Yang X. Hua X. Liu J. Fan M.B. Li G.Q. Song J. Xu T.Y. Li Z.Y. Guan Y.F. Wang P. Miao C.Y. Hepatic NAD+ deficiency as a therapeutic target for non-alcoholic fatty liver disease in ageing.Br. J. Pharmacol. 2016; 173: 2352-2368Crossref PubMed Scopus (92) Google Scholar, 18Wang L.-F. Wang X.-N. Huang C.-C. Hu L. Xiao Y.-F. Guan X.-H. Qian Y.-S. Deng K.-Y. Xin H.-B. Inhibition of NAMPT aggravates high fat diet-induced hepatic steatosis in mice through regulating Sirt1/AMPKα/SREBP1 signaling pathway.Lipids Health Dis. 2017; 16: 82Crossref PubMed Scopus (47) Google Scholar). Because an inducible knockout of Nampt was shown to impair liver regeneration (19Mukherjee S. Chellappa K. Moffitt A. Ndungu J. Dellinger R.W. Davis J.G. Agarwal B. Baur J.A. Nicotinamide adenine dinucleotide biosynthesis promotes liver regeneration.Hepatology. 2017; 65: 616-630Crossref PubMed Scopus (55) Google Scholar), we hypothesized that HNKO mice would be more likely to develop liver injury if the liver was more severely challenged. To determine whether HNKO mice had increased susceptibility to develop components of NASH, we challenged HNKO mice and WT littermates with a low-methionine, choline-free 60% high-fat diet (MCD) and evaluated the effects on liver function and metabolic parameters. Collectively, our data suggest the presence of a specific threshold for hepatic NAD+ levels that determines the predisposition for developing liver injury. Our data also suggest a zonation pattern in liver injury induced by knockout of Nampt, suggesting a spatial requirement for NAD+ levels across the liver lobule. In addition, liver injuries induced by subthreshold NAD+ levels were found to be associated with a decreased abundance of NAD+-dependent oxidoreductases and impaired maximal respiratory capacity. Key features of the liver phenotype were reversible when liver NAD+ levels were restored through supplementation with NAD+ precursors. Our findings may have implications for liver health in humans. To determine if lack of Nampt would induce liver injury, we challenged female HNKO mice with an MCD diet for 3 weeks (Fig. 1A). Lean mass and plasma alanine aminotransferase (ALT) activity were unaffected in WT mice, but HNKO mice lost 1.5 g of lean mass (Fig. 1B; p < 0.01) and had a three-fold increase in plasma ALT activity in response to the MCD diet (Fig. 1C; p < 0.01). This indicated liver injury in HNKO mice. Body weight (BW) and fat mass were not affected by MCD feeding (Fig. S1A and B). Hepatic NAD+ content was ∼66% lower in purified control diet (PD)-fed HNKO mice compared with WT littermates, but no additional decrease following MCD diet feeding was observed (Fig. S1C; p < 0.01). NADP+ levels were ∼20% lower in PD-fed HNKO mice compared with WT (p < 0.01), and a decrease in NADP+ was observed for both genotypes following MCD diet feeding (Fig. S1D; p < 0.05). NADPH levels were lower in PD-fed HNKO mice compared with WT littermates, and MCD feeding decreased NADPH levels in WT mice but not HNKO mice (Fig. S1E; p < 0.05). HNKO mice developed periportal fibrosis, immune cell infiltration, bile duct proliferation, and focal necrosis (Fig. 1D) and showed lower accumulation of hepatic triglyceride from the MCD diet challenge than WT mice (Fig. 1E). Remarkably, portal inflammation, necrosis, and fibrosis were also observed in HNKO mice on the PD, and scores were not significantly different between the MCD group and the control group (Fig. 1, F–H). However, there was no decrease in lean mass and no increase in plasma ALT activity in the PD HNKO group. HNKO livers appeared to have more proliferating cells (Ki-67, Fig. 1I) and had a significant increase in leukocyte infiltration in the portal areas for both diet groups (CD45, Fig. 1, I and J). Myofibroblast activation and bile duct proliferation was observed in HNKO mice regardless of diet (smooth muscle actin (SMA) and cytokeratin-19 [CK-19] Fig. 1, K and L). Overall, HNKO mice were more susceptible to damage from the MCD diet but developed liver fibrosis, inflammation, and bile duct proliferation even on the control diet. To determine temporal patterns in the development of liver injury in PD-fed HNKO mice, 9-week-old WT and HNKO mice were fed PD for 3, 6, 12, and 21 days (Fig. 2A). Fibrosis was present after 3 days (Fig. 2, B and C; p < 0.05), although the response was heterogeneous until the 12-day time point where all HNKO mice exhibited a positive fibrosis score (p < 0.01). Portal inflammation and necrosis varied between HNKO mice across the time course (Fig. 2, D and E). Liver NAD+ levels were decreased by 66% in HNKO mice and did not change over time (Fig. 2F; p < 0.01). In both humans and mice, NAFLD and NASH affect the plasma proteome, and content of plasma biomarkers may predict NAFLD/NASH development (20Niu L. Geyer P.E. Wewer Albrechtsen N.J. Gluud L.L. Santos A. Doll S. Treit P.V. Holst J.J. Knop F.K. Vilsbøll T. Junker A. Sachs S. Stemmer K. Müller T.D. Tschöp M.H. et al.Plasma proteome profiling discovers novel proteins associated with non-alcoholic fatty liver disease.Mol. Syst. Biol. 2019; 15e8793Crossref PubMed Scopus (81) Google Scholar). To investigate whether the plasma biomarker profile of HNKO mice was similar to other models of NAFLD and NASH, we analyzed the plasma proteome using MS. We identified that the highest number of differentially abundant proteins between genotypes were observed at day 6 and 12 (Fig. 2G, File S1). Eight proteins were differentially abundant in plasma between genotypes at all time points and were all increased in HNKO plasma compared with plasma from WT mice (Fig. 2H). Increased serum amyloid A1 abundance was previously reported in mice fed a high-fat diet for 6 months (20Niu L. Geyer P.E. Wewer Albrechtsen N.J. Gluud L.L. Santos A. Doll S. Treit P.V. Holst J.J. Knop F.K. Vilsbøll T. Junker A. Sachs S. Stemmer K. Müller T.D. Tschöp M.H. et al.Plasma proteome profiling discovers novel proteins associated with non-alcoholic fatty liver disease.Mol. Syst. Biol. 2019; 15e8793Crossref PubMed Scopus (81) Google Scholar), and polymeric immunoglobulin receptor was identified as a candidate plasma biomarker for human NAFLD and cirrhosis (20Niu L. Geyer P.E. Wewer Albrechtsen N.J. Gluud L.L. Santos A. Doll S. Treit P.V. Holst J.J. Knop F.K. Vilsbøll T. Junker A. Sachs S. Stemmer K. Müller T.D. Tschöp M.H. et al.Plasma proteome profiling discovers novel proteins associated with non-alcoholic fatty liver disease.Mol. Syst. Biol. 2019; 15e8793Crossref PubMed Scopus (81) Google Scholar). These data indicate the presence of fibrosis in HNKO mice already after 3 days on the PD, and they confirm the relevance of previously identified biomarker proteins for NAFLD. To determine potentially underlying mechanisms for PD-associated liver injury, we assessed time course dependent gene expression changes using RNA-Seq. Similarly to the observed changes in the plasma proteome, the largest number of differentially expressed genes in the liver between HNKO and WT was observed at day 6, where more than 2000 genes were differentially expressed (Fig. 3A, File S2). About 239 genes were found to be differentially expressed at all time points, and Gene Ontology (GO) enrichment analysis for "molecular function" revealed enrichment of terms such as calcium ion binding, cell adhesion molecule binding, and extracellular matrix structural constituent for this group, suggesting the presence of organ remodeling and extracellular matrix deposition at all investigated time points (Fig. 3B). We noted an increased expression of several markers for oval cells and cholangiocytes (21Sato K. Marzioni M. Meng F. Francis H. Glaser S. Alpini G. Ductular reaction in liver diseases: Pathological mechanisms and translational significances.Hepatology. 2019; 69: 420-430Crossref PubMed Scopus (116) Google Scholar), including Epcam, Ncam1, Spp1, Sox9, and Krt7 in HNKO livers throughout the time course (Fig. S2, A–E). Similarly, we observed a decreased expression of albumin after 6 days (Fig. S2F). It has been suggested that cholangiocytes and oval cells can contribute to liver regeneration through transdifferentiation when hepatocyte proliferation is impaired (21Sato K. Marzioni M. Meng F. Francis H. Glaser S. Alpini G. Ductular reaction in liver diseases: Pathological mechanisms and translational significances.Hepatology. 2019; 69: 420-430Crossref PubMed Scopus (116) Google Scholar). Hence, the increased expression of these markers suggests that ductular reaction and fibrosis may be a compensatory event in HNKO livers to support liver regeneration. To investigate whether the major transcriptional changes manifested as changes to the proteome, we quantified the proteome of HNKO and WT livers from the same samples. In contrast to the transcriptomics analysis, the largest number of differentially abundant proteins was observed at day 3 (Fig. 3C and File S3). About 92 proteins were differentially abundant at all four time points, and "molecular function" GO terms enriched for these proteins included structural molecule activity, extracellular matrix structural constituent, and extracellular matrix binding (Fig. 3D). Thus, these proteomics data confirm a continuous damage/regeneration response in HNKO mice fed a PD. Because fibrosis was observed after only 3 days of PD feeding, it was likely that fibrosis was present before PD feeding. To test this idea and further evaluate whether the micronutrient and macronutrient composition affected fibrosis development in HNKO mice, we fed 8 to 13-week-old HNKO and WT mice four different diets with a different NAD+ precursor content for 3 weeks (Fig. 4A). We tested PD, a fortified breeding diet (FBD; the standard chow in our animal facility), a standard breeding diet (SBD), and a standard maintenance diet (SMD). Fortified diets have a higher content of vitamins (including NA) compared with standard diets, whereas breeding diets have a higher content of fat and amino acids (including tryptophan) compared with maintenance diets (Fig. 4B). All mice started on FBD following weaning and until the beginning of the experiment. We observed a main effect of diet for hepatic NAD+ content (Fig. 4C; p < 0.05). Post hoc testing revealed a tendency toward increased liver NAD+ levels for FBD-fed groups compared with other diets, though this was not statistically significant (SBD: p = 0.07; SMD: p = 0.1; PD: p = 0.14). NAD+ levels were lower in HNKO liver for all diet groups (Fig. 4C, p < 0.01). Fibrosis and portal inflammation scores were significantly higher in HNKO mice for all diets, and necrosis scores were significantly increased in SBD-fed HNKO mice compared with SBD-fed WT mice (Fig. 4, D–F and I, p < 0.05). We observed a tendency toward lower fibrosis scored in FBD-fed HNKO mice compared with SBD-fed HNKO mice (Fig. 4D; p = 0.05) and significantly lower necrosis scores in FBD-fed HNKO mice compared with PD-fed and SBD-fed HNKO mice (Fig. 4F; p < 0.05). Notably, in several FBD-fed HNKO mice, fibrosis was absent, suggesting that the higher NA content protected some HNKO mice from fibrosis development (Fig. 4D). In contrast, fibrosis did not seem to be dependent on tryptophan intake, as fibrosis developed in SBD-fed HNKO mice. When fibrotic areas in sections were quantified, we saw that HNKO livers had significantly larger areas stained for fibrosis compared with WT animals across all diet groups (Fig. 4G; p < 0.01). A significant and negative correlation between stained fibrosis area and hepatic NAD+ content was observed in HNKO mice (Fig. 4H; Pearson r2 = 0.36; p < 0.01). Collectively, although necrosis in the HNKO livers are dependent on the dietary NAD+ precursor content, fibrosis was present irrespective of the diet and correlated negatively with hepatic NAD+ content. Fibrosis in HNKO mice was associated with decreased hepatic NAD+ content, and we therefore hypothesized that increasing hepatic NAD+ content could alleviate this phenotype. NR and NA are precursors of NAD+ that do not depend upon NAMPT for NAD+ synthesis. To investigate if fibrosis in HNKO mice could be prevented by NR, HNKO and WT mice were fed PD with or without NR in the drinking water for 3 weeks (Fig. 5, A and B). Hepatic NAD+ content in HNKO mice was three-fold higher with NR (Fig. 5C; p < 0.01), and they had significantly lower fibrosis scores, portal inflammation scores, and necrosis scores compared with the control group (Fig. 5, D–F; p < 0.05). To test whether supplementation with smaller amounts of NAD+ precursor had similar beneficial effects, we fed HNKO mice a PD with an NA content of 75 mg/kg for 3 weeks (Fig. 5, G and H). The additional supplementation of NA to the PD increased NAD+ levels two-fold in HNKO livers (Fig. 5I; p < 0.01). NA supplementation did not affect fibrosis scores (Fig. 5J) but significantly reduced portal inflammation scores (Fig. 5K; p < 0.01), necrosis scores (Fig. 5L; p < 0.01) and ductular reaction scores (Fig. 5M; p < 0.01). Together, these data demonstrate that increasing liver NAD+ levels attenuates liver damage in HNKO mice, and they also suggest the presence of a threshold for NAD+ under which liver damage occurs in HNKO mice. Bile duct–ligated mice present a similar phenotype to HNKO mice with a strong zonation of fibrosis in the portal areas of the lobuli (22Tag C.G. Sauer-Lehnen S. Weiskirchen S. Borkham-Kamphorst E. Tolba R.H. Tacke F. Weiskirchen R. Bile duct ligation in mice: Induction of inflammatory liver injury and fibrosis by obstructive cholestasis.J. Vis. Exp. 2015; https://doi.org/10.3791/52438Crossref PubMed Scopus (181) Google Scholar). Therefore, we investigated whether hepatic NAD+ deficiency was associated with cholestasis by measuring bile flow in HNKO and WT mice. WT and HNKO mice were fed PD for 6 weeks, and half of the mice received NR in their drinking water after the first 3 weeks (Fig. 6A). This would show whether NR reverses the phenotype induced by decreased NAD+ precursor intake. Interestingly, we found bile flow to be slightly increased in HNKO mice (Fig. 6B), possibly because of the enlargened bile ducts. Bile flow was not corrected by NR supplementation suggesting that the HNKO phenotype is not associated with cholestasis. NAD+ content in HNKO mice was 75% lower than for WT animals, and NR supplementation caused a 10-fold increase in hepatic NAD+ content in HNKO mice (Fig. 6C; p < 0.01). HNKO livers had a significantly lower content of ATP, which was not corrected by NR (Fig. 6D; p < 0.01). Rather, NR caused a decrease in hepatic ATP content regardless of genotype (p < 0.01). Three of seven HNKO mice had a fibrosis score of 0 following NR treatment (Fig. 6, E and F; p = 0.06). No significant change in necrosis scores was observed (Fig. 6G), but portal inflammation was significantly decreased in the NR-supplemented HNKO group (Fig. 6H; p < 0.01). Thus, liver injury in the HNKO mice is not caused by cholestasis, and NR effectively reduces liver inflammation. Next, we mapped the liver proteome to identify NR-associated changes in HNKO mice. Major differences in protein abundance were observed between WT and HNKO in the control group and between control and NR-treated HNKO mice (Fig. 6I and File S4). Remarkably, only few differentially abundant proteins were altered by genotype in the NR groups and by NR in WT mice. This clearly indicates that NR has only minor effects in livers of WT mice but that it reverses the proteome profile in the HNKO liver to resemble that of WT mice. Furthermore, this clearly highlights lack of NAD+ as a direct mechanism for the observed phenotype. GO enrichment analyses of proteins with significantly altered abundance between genotypes in the control group were enriched for "molecular function" terms associated with cell motility and fibrosis, such as cytoskeletal protein binding and extracellular matrix structural constituent (Fig. S3A). GO enrichment analysis for "cellular component" showed not only also an enrichment of extracellular matrix–associated terms, such as extracellular region, but also of mitochondria-associated terms such as mitochondrial membrane and mitochondrial envelope (Fig. S3B). Hence, the proteomics analysis suggested altered mitochondrial protein content in PD-fed HNKO mice. GO analysis of hepatic proteins with a significantly altered abundance following NR treatment in HNKO mice showed an enrichment of molecular function terms, such as oxidoreductase activity, actin binding, and oxidoreductase activity acting on CH-OH group of donors using NAD or NADP as acceptor (Fig. 7A). We observed a significant enrichment of cellular component terms, including mitochondrial envelope, mitochondrial membrane, and actin cytoskeleton, suggesting that NR attenuated cell motility and/or cytoskeletal rearrangements, while also affecting mitochondrial protein abundance. To further explore the cellular processes that were affected in HNKO livers, we annotated the proteins from the oxidoreductase activity term according to cofactor and/or cellular function, based on entries in the database UniProt (23Consortium T.U. Bateman A. Martin M.-J. Orchard S. Magrane M. Agivetova R. Ahmad S. Alpi E. Bowler-Barnett E.H. Britto R. Bursteinas B. Bye-A-Jee H. Coetzee R. Cukura A. Da Silva A. et al.UniProt: The universal protein knowledgebase in 2021.Nucleic Acids Res. 2021; 49: D480-D489Crossref PubMed Scopus (748) Google Scholar). This classification revealed that a high number of these proteins use NAD(H) or NADP(H) as cofactors (Fig. 7C). A heatmap representation of these proteins showed a clear overall decrease in the abundance of these proteins in livers of HNKO mice, which was normalized by NR treatment (Fig. 7D). This was also the case when all differentially abundant proteins from the oxidoreductase activity term were plotted in a heatmap (Fig. S4A). Hence, NR treatment rescued hepatic NAD+-dependent oxidoreductase protein abundance. As NR supplementation altered the abundance of many mitochondrial proteins in HNKO mice, we hypothesized that the HNKO phenotype could be associated with lower mitochondrial content. We observed a tendency toward decreased mitochondrial NAD+ levels in HNKO mice fed a PD diet compared with WT littermates (Fig. 8A, p = 0.05); although the mitochondrial to total NAD+ ratio was increased in HNKO mice (Fig. 8B; p < 0.05). Hence, the decreased abundance of mitochondrial proteins was not