NorUDCA promotes degradation of α1-antitrypsin mutant Z protein by inducing autophagy through AMPK/ULK1 pathway

Alpha-1 Antitrypsin (α1AT) Deficiency is a genetic disease in which accumulation of α1AT mutant Z (α1ATZ) protein in the ER of hepatocytes causes chronic liver injury, liver fibrosis, and hepatocellular carcinoma. No effective medical therapy is currently available for the disease. We previously found that norUDCA improves the α1AT deficiency associated liver disease by promoting autophagic degradation of α1ATZ protein in liver in a mouse model of the disease. The current study unravels the novel underlying cellular mechanism by which norUDCA modulates autophagy. HTOZ cells, modified from HeLa Tet-Off cells by transfection with the resulting pTRE1-ATZ plasmid and expressing mutant Z proteins, were studied in these experiments. The role of norUDCA in inducing autophagy, autophagy-mediated degradation of α1ATZ and the role of AMPK in norUDCA-induced autophagy were examined in the current report. NorUDCA promoted disposal of α1ATZ via autophagy-mediated degradation of α1ATZ in HTOZ cells. Activation of AMPK was required for norUDCA-induced autophagy and α1ATZ degradation. Moreover, mTOR/ULK1 was involved in norUDCA-induced AMPK activation and autophagy in HTOZ cells. Our results provide novel mechanistic insights into the therapeutic action of norUDCA in promoting the clearance of α1ATZ in vitro and suggest a novel therapeutic approach for the treatment of α1ATZ deficiency disease and its associated liver diseases.


Introduction
Alpha-1 Antitrypsin (α1AT) Deficiency is a genetic disease, which is caused by homozygosity for the Z mutant of α1AT and occurs in 1 out 2,000-5,000 live births in North America [1]. The mutant Z protein is encoded by the α1AT mutant Z (α1ATZ) gene, a substitution of lysine for glutamate at residue 342. The protein product adopts a polymerized conformation and

Cell culture
HTOZ cells were modified from HeLa Tet-Off cells by transfection with the resulting pTRE1-ATZ plasmid (Teckman et al., 2001). HTOZ/M were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% of fetal bovine serum (FBS) and Pen/Strep, at 5% CO2, 100% humidity and 37˚C. Cells cultured in medium containing doxycycline (DOX, 40ng/ml) turns off expression of α1ATZ proteins. 60-70% confluent cells were used in experiments. In some experiments, Z expression was turned on 24 hours before intervention and sometimes it was simultaneous depending on interaction of drugs and the amount of intracellular protein accumulated required for the readout, for example, depending on if the experiment is to show a prevention or a reduction. When expression is turned on, new mRNA reaches a steady state in approximately 24h, which roughly corresponds to steady state levels of new ATZ monomer protein levels in the ER. However, progressive accumulation of ATZ insoluble aggregates (polymers) continues to increase over several days and is somewhat variable between parent cell lines. The ATZ soluble and insoluble protein fractions have different intracellular half-lives as previously shown [24]. mRNA expression remains stable for [30][31][32][33][34][35][36][37][38][39][40] days then begins to decline as lines age and are passed. Passing cells alters the soluble/insoluble ATZ protein ratios, given the different half-lives of the pools and depending on the age of the cells. Cells were appropriately treated with norUDCA as indicated.

RNA extraction and Real-time PCR
Total RNAs were extracted by TRI 1 Reagent (Sigma, MO, USA) from HTOZ/M cells and treated with DNase I before the synthesis of the first strand of cDNA. Real-time PCR were performed as we previously described using SYBR Green Supermix [25]. mRNA levels are expressed as fold changes after normalization with glyceraldehyde-3-phosphate dehydrogenase (GAPDH), as described by Schmittgen et al. [26]. The primers for human are as follows:

Plasmids, transient transfection and LC3 puncta analyses
The plasmid pEGFP-LC3 (Addgene plasmid 21073) was purchased from Addgene (Cambridge, MA, USA). pEGFP-LC3 contained a fragment (738 bp) of the gene LC3. The cells were fed with fresh medium or treated with norUDCA at appropriate dosages in fresh medium for 1 hour prior to transfection. For transfection, semi-confluent cells in p60 dishes were transiently transfected with a total of 3 μg DNA per dish, using the Lipofectamine1 2000 reagent (Invitrogen Corp., Carlsbad, CA), and following the protocol provided by the manufacturer. Cells were fixed with 4% paraformaldehyde at 0, 16 and 24 hours after transfection and observed under inverted fluorescent microscope (Leica DMI 4000 B). The 10-12 areas were randomly chosen and images were taken under 20X magnification. Total living, LC3-positive and puncta-positive cells were counted, based on morphological and fluorescent criteria, and then calculated for percentage of LC3 positive and puncta positive cells, respectively. Only fluorescent cells were included for puncta percentage analysis. Of this population, cells having at least three morphologically discrete fluorescent puncta qualified to be counted as puncta positive cells. The number of puncta positive cells was divided by the Total LC3-GFP positive cells for calculating the percentage.

Statistical analysis
Differences between means were evaluated using an unpaired two sided Student's t test (P<0.05 considered as significant). Where appropriate, comparisons of multiple treatment conditions with controls were analyzed by ANOVA with the Dunnett's test for post hoc analysis.

Exogenous norUDCA enhances degradation of α1ATZ in HTOZ cells
Our previous data in an in vivo model showed that norUDCA prevented hepatic accumulation of α1AT mutant Z protein polymers in juvenile animals, reversed existing accumulation in adult animals, and reduced liver injury. To test the role of norUDCA in a reduction of accumulation of α1AT mutant Z proteins in an in vitro system, HTOZ cells that express mutant Z protein (Fig 1A), were treated with norUDCA at 200 μM for indicated periods (Fig 1B), or at different concentrations (0-1000μM) for 24 hours (Fig 1C) in 10% FBS DOX-free medium. NorUDCA significantly reduced the steady-state levels of α1AT in a time-dependent manner, and to some degree dose-dependently (Fig 1). NorUDCA had no impact on α1AT mRNA levels ( Fig 1D and 1E), indicating that norUDCA did not affect α1ATZ synthesis. Next, we sought to explore whether norUDCA affect the soluble and insoluble fractions of α1ATZ in HTOZ cells. The soluble α1ATZ represents newly synthesized polypeptides, which are known to fold inefficiently during biogenesis and to be retained in the ER. α1ATZ polypeptides in the ER that escape degradation by the ERAD pathway aggregate (polymerize) into insoluble masses targeted for degradation by autophagy. To test differences in the degradation of these two pools by norUDCA, HTOZ cells were treated with norUDCA at 200 μM for 24 hours, soluble and insoluble α1ATZ were isolated for western blotting analysis. This published technique is known to efficiently separate the insoluble aggregates from soluble monomeric α1ATZ molecules [5,24]. The insoluble fraction is then denatured, which breaks the aggregates (also called polymers) back into monomeric α1ATZ, as they are not bound by covalent bonds. NorUDCA remarkably decreased the steady-state levels of α1ATZ in both the insoluble and soluble fractions in vitro ( Fig 1F). Similar results were found in the long-term treated HTOZ cells ( Fig 1G). Taken together, these data demonstrate that norUDCA reduces accumulation of α1ATZ in HTOZ cells in vitro.

NorUDCA increased autophagy in HTOZ cells
It is well known that misfolded proteins can be degraded through autophagy. To address underlying mechanisms by which norUDCA reduces accumulation of α1ATZ, we test the role of norUDCA in enhancing autophagy, specifically its effect on isoform conversion of autophagosomal membrane-specific protein LC3, an indicator of autophagosome formation, in the HTOZ cell line. Cells were treated with norUDCA at different concentrations or 200μM for different time points as indicated in doxycycline-free DMEM. Whole cell lysates were prepared and analyzed with western blotting analysis. The ratio of LC3-II to LC3-I was increased in both time-dependent manner (2A) and dose-dependent manner (2B) in HTOZ cell line ( Fig  2). Also, norUDCA enhanced LC3 puncta formation, another indicator of autophagic activity, in a dose-dependent manner in HTOZ cells after transfection of pEGFP-LC3 plasmids ( Fig  2B). LC3 is found in the membrane of the autophagic vacuoles and increased labeling is shown to be proportional to increased vacuole formation in various systems.
Next, we tested if norUDCA affects expression of genes related to autophagy. HTOZ or HTOM cells were treated with norUDCA at different concentrations for 24 hours in doxycycline-free DMEM. Autophagy-related gene 5 (ATG5) is necessary for autophagy due to its role in autophagosome elongation and p62 is commonly used as a marker to study autophagic flux. Both are major genes relevant to autophagy. Real-time PCR analysis showed that nor-UDCA increased ATG5 mRNA levels both in HTOZ and HTOM cells (Fig 2C and 2D) and western blotting analysis showed that role of norUDCA in increasing ATG5 while reducing p62 protein abundance in HTOZ cells (2D). These data indicate the role of norUDCA in inducing autophagy in vitro. Collectively, these data demonstrate that norUDCA induces autophagy in vitro.

NorUDCA increases degradation of insoluble α1ATZ Via induction of autophagy
To determine if norUDCA-induced autophagy is involved in insoluble α1ATZ degradation, HTOZ cells, cultured in 10% FBS dox-free DMEM up to 80% confluence, were pretreated with chloroquine (CQ), commonly used pharmacological autophagy inhibitor, at different concentrations (0, 5, 10 μM) for 1 hour before addition of norUDCA at 200 μM for an additional 24 hours. Insoluble and soluble α1ATZ was separated for western blotting analysis. As expected in Fig 3A, norUDCA induced a decrease in levels of insoluble α1ATZ in HTOZ cells, which was inhibited by CQ, indicating that norUDCA-induced degradation of α1ATZ was reversed by the blockade of autophagy in vitro.
To confirm effect of norUDCA on α1ATZ degradation involved in enhanced autophagy, we determined its effect on α1ATZ levels in ATG5-KD (atg5 knock-down) HTOZ cells. ATG5 -KD HTOZ cell line was established by using ATG5 shRNA plasmids to knock down atg5 (autophagy-deficient) as described in Materials and Methods. Importantly, the efficacy of ATG5 knock-down is more than 90% (Fig 3B). Expression of shATG5 blocks autophagy at a proximal step by preventing the formation of the ATG5-ATG12 complex, which is required for the generation of autophagosomes. Knock-down of ATG5 resulted in an increase of insoluble fraction of α1ATZ in ATG5-KD HTOZ cells, compared to ATG5 expressing (ATG5-WT) HTOZ cells (Fig 3C). Further, ATG5 HTOZ and ATG5-KD HTOZ cells were treated with/out norUDCA at 200 μM for 24 hours, respectively. NorUDCA induced a decrease in levels of insoluble α1ATZ in the ATG5-WT HTOZ cells but not in the ATG5-KD HTOZ cells, shown in Fig 3D. NorUDCA also induced a decrease in levels of soluble α1ATZ in both ATG5-WT and ATG5-KD cells. Collectively, the data indicate that norUDCA enhances removal of insoluble α1ATZ by autophagy and has an independent effect on the removal of soluble α1ATZ by mechanism(s) that do not involve the conventional autophagic pathway. Moreover, these data suggest that norUDCA-induced removal of insoluble α1ATZ could be eliminated by blockage of autophagy.
Next, we explored whether ERAD is related to norUDCA-caused reduction of insoluble α1ATZ in HTOZ cell line. The cells, cultured in 10% FBS DOX minus DMEM up to 80% confluence, were pre-incubated with MG132 (30 μM) for 1 hour and followed by treatment of nor-UDCA (200 μM) for additional 24 hours. Insoluble and soluble α1ATZ was separated for western blotting analysis. Cells that were incubated with MG132 alone served as control to validate that the proteasome was inhibited. As shown in Fig 3E, norUDCA decreased levels of insoluble α1ATZ, which could not be restored by blockage of proteasomal pathway, indicating that norUDCA-mediated decrease in levels of insoluble α1ATZ is independent of proteasomal pathway. Taken together, norUDCA induces degradation of insoluble α1ATZ via autophagic pathway in HTOZ cells. Our previous work shows soluble α1ATZ to be degraded by ERAD 8 , which we did not further investigate here.

Activation of AMPK mediates norUDCA-induced autophagy and α1ATZ degradation
AMPK is a sensor of energy in mammalian cells. It has been reported that AMPK regulates autophagy in several cell types [27][28]. Therefore, we sought to examine whether AMPK mediates norUDCA-caused autophagy and α1ATZ polymer reduction, and whether the activator of AMPK, AICAR would initiate this effect. To test this hypothesis, HTOZ cells were treated with norUDCA at 200 μM as indicated time points or at different concentrations (0-500 μM) for 1 hour. Phosphorylation of AMPK (Thr172) was determined by western blotting. As shown, norUDCA increased levels of phosphorylation of AMPK in both the time-dependent manner (Fig 4A) and dose-dependent manner (Fig 4B). NorUDCA-induced AMPK could be dose-dependently eliminated by Compound C, a well-established pharmacological inhibitor of AMPK (Fig 4C), showing that norUDCA can activate AMPK in vitro.
Next, we tested whether AICAR, mimics the effect of norUDCA and induces autophagy in HTOZ cells. The cells were treated with AICAR at different concentrations (0-1.0 mM) for 24 hours. As shown in Fig 4D, AICAR increased LC3-II protein abundance. Moreover, data from quantitative real-time PCR indicated that AICAR increased mRNA levels of ATG5 while slightly decreased p62 mRNA levels, the major autophagy related genes, in HTOZ cells after exposure to AICAR at 1.0 mM for 24 hours (Fig 4E). To establish the role of AMPK in mediating norUDCA-induced autophagy, HTOZ cells were pretreated with/without compound C (20 μM) for 1 hour prior to addition of norUDCA (200 μM) or AICAR (1.0 mM) for 24 hours. Western blotting analyses indicated that norUDCA increased the ratio LC3-II/GAPDH and LC3-II/LC3-I, which was abrogated by compound C (Fig 4F) in vitro. These data demonstrate that activation of AMPK mediates norUDCA-induced autophagy in HTOZ cells.
Further, we tested the role of AMPK in mediating norUDCA-induced α1ATZ polymer reduction in HTOZ cells. The cells were pretreated with/without compound C (20 μM) for 1 hour prior to addition of norUDCA (200 μM) or AICAR (1.0 mM) for 24 hours. Isolation of α1ATZ polymer and monomer was performed for western blotting analyses. As shown in Fig  4G, norUDCA, mimic AICAR (the 2nd and 7th lane from the left), reduced α1ATZ polymers and monomers (the 3rd and 8th lane from the left), which was repressed by compound C (the 4th and 9th lane from the left). There was no change between control (The 1st lane) and compound C alone (the last lane). These results suggest that AMPK signaling is involved in norUDCA-induced α-ATZ reduction. Taken together, our results demonstrate that activation of AMPK mediates norUDCA-induced autophagy and α1ATZ degradation in HTOZ cells in vitro.

Involvement of mTOR signal in norUDCA-induced AMPK activation and autophagy
To further address mechanisms by which norUDCA enhances degradation of α1ATZ proteins via AMPK activation in HTOZ cells, we evaluated the downstream of AMPK signaling related to autophagy, such as mTOR, a serine/threonine protein kinase that involves in autophagy, and the signals upstream of mTOR, including AKT. The HTOZ cells were treated with nor-UDCA at different concentrations for 1 hour. As shown by western blotting analysis, nor-UDCA decreased phosphorylation of mTOR Ser2448 (Fig 5A), indicating that mTOR might be involved in norUDCA-induced AMPK activation. To establish the role of mTOR in nor-UDCA-mediated AMPK activation, HTOZ cells were pretreated with AICAR or compound C Novel therapeutic bile acid norUDCA promotes autophagy via AMPK for 1 hour and subsequently treated with or without norUDCA at 200 μM for an additional 1 hour. As shown in western blotting analysis, norUDCA decreased phosphorylation levels of mTOR Ser2448 (Fig 5B), which was suppressed by compound C and synergistically enhanced by AICAR, suggesting that norUDCA modulates mTOR via AMPK signaling pathway. However, norUDCA did not alter the levels of AKT phosphorylation (Fig 5C and 5D). Collectively, our data demonstrates that mTOR is involved in norUDCA-induced AMPK activation.

Activation of ULK1 is needed For norUDCA-induced reduction of insoluble fraction of α1ATZ Via AMPK activation
Unc-51 like autophagy activating kinase 1 (ULK1) is an important protein in autophagy and plays a critical role in early steps of autophagosome biogenesis. AMPK upregulates ULK1 activity via AMPK-dependent phosphorylation [11,29] and hence is essential in induction of autophagy [30]. To further address mechanisms by which norUDCA enhances autophagy and degradation of α1ATZ proteins in HTOZ cells, we sought to evaluate whether norUDCA activates ULK1 via AMPK signaling pathway, and whether ULK1 mediates norUDCA-induced reduction of insoluble fraction of α1ATZ in HTOZ cells. The HTOZ cells were treated with norUDCA at different concentrations for 1 hour. As shown by western blotting analysis, nor-UDCA increased phosphorylation of ULK1 at Ser555 site (Fig 6A), also increased phosphorylation of ULK1 at Ser317, Ser777 sites and decreased Ser757 (Fig 6B), indicating that norUDCA activates ULK1, which might be involved in norUDCA-induced autophagy.
To establish the role of ULK1 in norUDCA-mediated AMPK activation, HTOZ cells were pretreated with AICAR or compound C for 1 hour and subsequently treated with or without norUDCA at 200 μM for an additional 1 hour. As shown in western blotting analysis, nor-UDCA increased phosphorylation of ULK1 Ser555, which was suppressed by compound C (Fig 6C), suggesting that norUDCA induces ULK1 via activation of AMPK signaling pathway.
To further establish the role of ULK1 in norUDCA-mediated clearance of insoluble fraction of α1ATZ, HTOZ cells were pretreated with SBI-0206965 (10 mM), ubiquitous inhibitor of ULK1, for 1 hour and followed by treatment with or without norUDCA at 200 μM for an additional 24 hours. Insoluble and soluble α1ATZ was separated for western blotting analysis. As shown in western blotting analysis, norUDCA decreased insoluble fraction of α1ATZ, which was diminished by SBI-0206965 (Fig 6D), suggesting that norUDCA-induced ULK1 is involved in the removal of insoluble α1ATZ. Collectively, our data demonstrates that nor-UDCA induces autophagy and reduces accumulation of α1ATZ by modulating AMPK/ULK1 signaling pathway.

Discussion
In homozygous ZZ individuals, the classic form of α1AT deficiency, mutant Z protein synthesized in the liver misfolds, aggregates and accumulates in hepatocyte ER, leading to cell dysfunction, susceptibility to cell death, and liver injury [2,31]. Liver transplantation is the only curative approach for patients with severe α1AT deficiency disease. In the current work, we demonstrate that norUDCA reduced accumulation of α1ATZ and promoted degradation of aggregates (polymers) of α1ATZ by inducing autophagy via activation of AMPK/mTOR/ ULK1 signaling pathway.
NorUDCA is a derivate of UDCA that is a minor constituent of human bile. Both nor-UDCA and UDCA indicate the therapeutic potential against cholestatic and metabolic disorders [15,[19][20][21]32]. Since norUDCA is only effectively conjugated with glycine or taurine, it has specific physicochemical and therapeutic properties distinct from UDCA [21]. Our previous work showed that norUDCA ameliorated α1AT deficiency associated liver disease in a mouse model. Here, HTOZ cell line was applied to the experiments as previously described [8]. While not a liver cell line, several past publications have established this cell line as useful and have shown that it faithfully recapitulates the intracellular processing and degradation of mutant Z protein. In fact, every aspect of Z protein intracellular retention, trafficking and degradation that has been possible to test in cell culture, mouse liver and human liver have shown the same results. Here we found that norUDCA lowered the steady-state and polymers of Novel therapeutic bile acid norUDCA promotes autophagy via AMPK α1ATZ, but did not alter synthesis of α1AT in HTOZ cell line (Fig 1), which is indicative that norUDCA effectively reduced accumulation of α1ATZ by promoting degradation in an in vitro system. This is the first study showing the role of norUDCA on the intracellular mechanisms of α1ATZ degradation. These data confirm the hypothesis that norUDCA enhanced degradation of α1ATZ, leading to disposal of the burden of α1ATZ, and appeal to us to further address the underlying mechanisms.
Wild type α1AT ("M" allele) is predominantly synthesized in liver and rapidly secreted into the blood to protect host tissues from the enzyme neutrophil elastase during periods of inflammation. The Z mutation causes the freshly produced α1AT polypeptide chain to be misfolded and retained within ER of hepatocytes, resulting in hepatic damage. For mammalian cells, misfolded Z protein can be targeted by autophagy and ERAD [5][6][7][8]. We previously reported that the proteasome plays an important role in ER degradation of α1ATZ in many cell types including hepatocytes [8]. Our laboratory and others have shown the responsibility of autophagy for efficient disposal of ER-retained α1ATZ [5]. Therefore, targeting autophagy and proteasome pathways could be an effective approach to removal of α1ATZ. As an explanation for nor-UDCA-induced degradation of α1ATZ in HTOZ cells (Fig 1), we found that norUDCA induced autophagy in HTOZ cells (Fig 3), which is required for norUDCA-mediated degradation of polymers of α1ATZ other than proteasome pathway (Fig 4). NorUDCA also decreased levels of soluble α1ATZ in both wild-type and atg5-deficient HTOZ cells, suggesting that nor-UDCA enhances removal of insoluble α1ATZ by autophagy and has an independent action on the removal of soluble α1ATZ by a mechanism that is not involved in the autophagic pathway.
Autophagy is important for balancing sources of energy at critical times in development and in response to nutrient stress [33]. Usually, autophagy is activated under cellular stress and nutrient deprivation and is regulated by cell signaling pathways, including PI3K/AKT, mTOR and AMPK [10][11][12]. Our data showed that norUDCA-activated AMPK was required for norUDCA-induced autophagy and reduced polymers of α1ATZ (Fig 5). AMPK is a Serine/ Threonine kinase that plays a role in cellular energy homeostasis and activated by a rise in the ratio of AMP/ATP within the cell upon depletion of ATP [34]. In addition, AMPK kinase activity is upregulated by LKB1 [35,36] and Ca2+/calmodulin-dependent protein kinase kinase-β (CaMKKβ) [37,38]. Although the exact mechanism by which norUDCA activates AMPK is not clear, it can be explained by the possibility of norUDCA in increase of Ca 2+ [39] in the cells. Additional experiments are needed to address this issue.
This study also unraveled that norUDCA increased phosphorylation of ULK1 at Serine (Ser) 317, Ser555 and Ser777 sites, and suppressed phosphorylation of mTOR, which was reversed by blockade of AMPK signaling pathway, but norUDCA did not affect phosphorylation of AKT (Fig 6), showing the involvement of mTOR and ULK1 in this process. ULK1 is the initiator of autophagy, which has multiple phosphorylation sites, including Ser317 & Ser777 [40], Ser555 [41,42] and Ser757 [43]. The former three are AMPK-dependent and the last is associated with mTOR. The direct interaction between AMPK and ULK1 positively regulates ULK1 activity through AMPK-dependent phosphorylation [11] and is essential for autophagy induction [30]. Bach et al. confirmed Ser555 as a major AMPK-dependent phosphorylation [41]. Based on the current observations, we propose that the mechanism by which norUDCA promotes degradation of polymers of α1ATZ is to induce autophagy by activating AMPK, and subsequently by phosphorylating and activating ULK1 and inhibiting mTOR signaling pathway, leading to the reduction of accumulation of α1ATZ in hepatocytes in vitro (Fig 7). Future work will examine if norUDCA affects ERAD and other degradation pathways of soluble α1ATZ.
Increasing evidence shows that autophagy can exert as a promising target to treat a number of pathological conditions, such as neurodegeneration [44] and aging [45]. Increased mTOR activity is associated with development of Alzheimer's disease (AD) in vitro and in vivo models [46], and with Huntington disease in animals [47] and humans [48]. We previously reported that rapamycin, an inhibitor of mTOR and activator of autophagy, showed its ability to reduce α1ATZ protein accumulation and ameliorate other markers of liver injury in PiZ mice [49]. However, not all the α1AT aggregates in the cell were fully degraded, for reasons that are still not clear, and an unusual dosing scheme was required to see benefits in the mouse model. Given that these mechanistic and toxicity questions remain, human studies of rapamycin have not commenced [50]. In contrast to other drugs previously proposed for α1AT deficiency [7,49], norUDCA has the main advantage of being safe and well tolerated and is currently undergoing further clinical development in humans [16,51]. Our findings could be rapidly translated into a clinical trial. In conclusion, our results in this report shed novel insights into mechanisms of norUDCA in the reduction of accumulation of α1ATZ in vitro and provide a therapeutic candidate for the treatment of α1AT deficiency disease and its associated liver diseases. Novel therapeutic bile acid norUDCA promotes autophagy via AMPK