Investigating the Role of STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis: Insights from In Vitro and In Vivo Studies Ying Xia Department of Chemistry and Molecular Biology Gothenburg, Sweden 2024 Investigating the Role of STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis: Insights from In Vitro and In Vivo Studies YING XIA ISBN: 978-91-8069-805-4 (PRINT) ISBN: 978-91-8069-806-1 (PDF) Cover Art: The Secret Garden of the Liver, designed by Ying Xia. © Ying Xia Department of Chemistry and Molecular Biology University of Goteborg Sweden Printed in Borås, Sweden 2024 Printed by Stema Specialtryck AB ii To my dear parents, iv ABSTRACT Metabolic dysfunction-associated steatotic liver disease (MASLD) has emerged as a leading cause of chronic liver disease worldwide, affecting approximately 30% of the adult population. Up to 20% of individuals with MASLD progress to metabolic dysfunction-associated steatohepatitis (MASH), which is characterized by hepatic inflammation, fibrosis, and cell damage in the form of ballooning degeneration and apoptosis, in addition to hepatic fat infiltration. Patients with MASH are at high risk of developing cirrhosis, liver failure, and hepatocellular carcinoma (HCC), which is one of the most fatal and fastest-growing cancers. Thus, understanding the molecular pathogenesis of MASLD, and the signals that trigger the transition of simple steatosis to MASH and MASH-related HCC, are of high clinical importance. Our previous studies have identified several STE20-type kinases including STK25, MST3, and MST4 as important regulators of MASLD/MASH susceptibility. Here, we examined the possible role of STE20 kinases TAOK1 and TAOK3 in hepatocellular lipotoxicity and MASLD. We found that the expression levels of TAOK1 and TAOK3 in human liver biopsies positively correlated with key hallmarks of MASLD (i.e., hepatic steatosis, inflammation, and ballooning). The subcellular localization of TAOK1 and TAOK3 in human and mouse hepatocytes was confined to intracellular lipid droplets. Knockdown of TAOK1 or TAOK3 alleviated lipotoxicity in cultured human hepatocytes by accelerating lipid catabolism (mitochondrial β-oxidation and triacylglycerol secretion), suppressing lipid anabolism (fatty acid influx and lipogenesis), and mitigating oxidative/endoplasmic reticulum stress; the opposite changes were detected in TAOK1- or TAOK3-overexpressing cells. However, in contrast to the in vitro observations, genetic deficiency of TAOK3 in obese mice failed to inhibit diet- induced liver steatosis, inflammation, fibrosis, or systemic metabolic disturbances. Interestingly, the hepatic mRNA abundance of several TAOK3-related kinases, which have been previously implicated in MASLD development, was elevated in Taok3–/– vs. wild-type mice. We also investigated the role of STE20-type kinase STK25 in the initiation and progression of MASH- related HCC. Analysis of publicly available databases and in-house cohorts revealed that STK25 expression in human liver biopsies positively correlated with the incidence and severity of HCC. We found that the in vitro silencing of STK25 in human hepatoma cells suppressed proliferation, migration, and invasion, with efficacy comparable to that achieved by anti-HCC drugs sorafenib or regorafenib. STK25 knockout in human hepatoma cells also blocked tumor formation and growth in a subcutaneous xenograft mouse model. Furthermore, pharmacologic inhibition of STK25 with antisense oligonucleotides, either globally across all peripheral tissues or specifically in hepatocytes, efficiently mitigated the development and exacerbation of hepatocarcinogenesis in a mouse model of MASH- driven HCC. Collectively, this thesis suggests that (1) TAOK1 and TAOK3 are critical regulatory nodes controlling hepatocellular lipotoxicity; (2) the lack of hepatic phenotype in Taok3 knockout mice in vivo may be attributable to the liver-specific compensation response for the genetic loss of TAOK3 by related STE20-type kinases; and (3) antagonizing STK25 signaling is a promising therapeutic strategy for the prevention and treatment of HCC in the context of MASH. Keywords: Metabolic dysfunction-associated steatotic liver disease, metabolic dysfunction-associated steatohepatitis, hepatocellular carcinoma, STE20-type kinases, lipid droplets, antisense oligonucleotide treatment i SAMMANFATTNING PÅ SVENSKA Leverförfettning, även kallat “metabolic dysfunction-associated steatotic liver disease” (MASLD), är idag den vanligaste orsaken till kronisk leversjukdom i världen och drabbar cirka 30% av den vuxna befolkningen. Av dessa utvecklar cirka 20% den mer allvarliga varianten av sjukdomen, så kallad ”metabolic dysfunction-associated steatohepatitis” (MASH), som utöver fettinlagring även inkluderar inflammation, fibros och cellskador på levern. Patienter med MASH har också ökad risk för att utveckla cirros, leversvikt och hepatocellulär cancer (HCC), som är en av de dödligaste cancerformerna som dessutom ökar mest i världen. Trots den höga kliniska relevansen vet man inte idag varför vissa utvecklar MASLD samt går från enkel fettackumulering till MASH och MASH-inducerad HCC. Våra tidigare studier har visat att STE20-typ-kinaserna STK25, MST3 och MST4 har en viktig roll i utvecklingen av MASLD/MASH. I denna studie har vi undersökt om även två andra STE20-kinaser, TAOK1 och TAOK3, styr regleringen av ackumulering av fett i levern. Vi fann att TAOK1- och TAOK3-nivåerna i humana leverbiopsier var positivt korrelerade med de viktigaste kännetecknen för MASLD (dvs. fettackumulering, inflammation och cellskador), samt att de är lokaliserade till lipiddroppar. Genom att minska nivåerna av TAOK1 eller TAOK3 i leverceller uppnåddes en ökad förbränning och utsöndring av fett, samtidigt som upptaget och syntesen minskade, vilket resulterade i minskad oxidativ och ER-stress. Ökade nivåer av TAOK1 eller TAOK3 visade motsatt effekt på lipidinlagring och metabolisk stress. Dock observerades inte dessa förändringar hos överviktiga genmodifierade möss som saknar TAOK3 då de utvecklade fettinlagring, inflammation och fibros i lever på samma nivå som kontrollmössen. Däremot noterades förhöjt genuttryck av flera av de andra STE20-kinaser som vi visat är inblandade i utvecklingen av MASLD i mössen med genetisk avsaknad av TAOK3. Vi undersökte även STE20-kinaset STK25 i utvecklingen av MASH-inducerad HCC, där analyser av allmänna databaser och patientkohorter visade en positiv korrelation mellan nivåerna av STK25 och incidensen samt svårighetsgraden av HCC. Lägre nivåer av STK25 i humana leverceller ledde till minskad proliferation, migration och invasion med samma effektivitet som HCC-läkemedlen sorafenib och regorafenib. Vi visade även att levercancerceller som saknar STK25 inte bildar tumörer i en xenograftmodell, samt att farmakologisk hämning av STK25 effektivt mildrade utvecklingen av HCC i möss med MASH. Sammanfattningsvis föreslår denna avhandling att (1) TAOK1 och TAOK3 är kritiska regulatorer av de skador fettackumulering kan orsaka i lever; (2) genetisk avsaknad av TAOK3 i möss inte ger några metabola effekter, vilket kan bero på ökade nivåer av relaterade STE20-kinaser; och (3) farmakologisk hämning av STK25 är en lovande terapeutisk strategi för att motverka och behandla MASH-inducerad HCC. Keywords: Metabolic dysfunction-associated steatotic liver disease, metabolic dysfunction-associated steatohepatitis, hepatocellulär cancer, STE20-typ-kinaserna, lipiddroppar, antisensoligonukelotider ii LIST OF PAPERS This thesis is based on the following papers that will be referred to their Roman numerals: I. Xia Y, Caputo M, Cansby E, Anand SK, Sütt S, Henricsson M, Porosk R, Marschall HU, Blüher M, and Mahlapuu M. (2021). STE20-Type Kinase TAOK3 Regulates Hepatic Lipid Partitioning. Molecular Metabolism, 54:101353. doi: 10.1016/j.molmet.2021.101353. II. Xia Y, Andersson E, Anand SK, Cansby E, Caputo M, Kumari S, Porosk R, Kilk K, Nair Y, Marschall HU, Blüher M, and Mahlapuu M. (2023). Silencing of STE20-Type Kinase TAOK1 Confers Protection against Hepatocellular Lipotoxicity via Metabolic Rewiring. Hepatology Communication, 17;7(4):e0037. doi: 10.1097/HC9.0000000000000037. III. Xia Y, Andersson E, Caputo M, Cansby E, Sedda F, Font-Gironès F, Ruud J, Kurhe Y, Hallberg B, Marschall HU, Asterholm IW, Romeo S, Blüher M, and Mahlapuu M. (2023). Knockout of STE20-Type Kinase TAOK3 Does not Attenuate Diet-Induced NAFLD Development in Mice. Molecular Medicine, 20;29(1):138. doi: 10.1186/s10020-023-00738-y. IV. Xia Y, Caputo M, Andersson E, Asiedu B, Zhang J, Hou W, Amrutkar M, Cansby E, Gul N, Gemmink A, Myers C, Aghajan M, Booten S, Hoy AJ, Romeo S, Härtlova A, Lindahl P, Ståhlberg A, Schaart G, Hesselink M, Peter A, Murray S, and Mahlapuu M. (2024). Therapeutic Potential of the Inhibitors of STE20-Type Kinase STK25 in Metabolically Induced Hepatocellular Carcinoma Prevention and Treatment. In Manuscript. iii CONTRIBUTION REPORT Paper I. Xia Y generated the bulk of the results, analyzed the data, edited the manuscript, revised the article critically for important intellectual content, and approved the final version of the article to be published. Paper II. Xia Y generated the bulk of the results, analyzed the data, wrote and edited the manuscript, revised the article critically for important intellectual content, and approved the final version of the article to be published. Paper III. Xia Y generated the bulk of the results, analyzed the data, wrote and edited the manuscript, revised the article critically for important intellectual content, and approved the final version of the article to be published. Paper IV. Xia Y generated the bulk of the results, analyzed the data, wrote and edited the manuscript, revised the article critically for important intellectual content, and approved the final version of the article to be published. iv PUBLICATIONS NOT INCLUDED IN THE THESIS I. Kurhe Y, Caputo M, Cansby E, Xia Y, Kumari S, Anand SK, Howell BW, Marschall HU, and Mahlapuu M. (2021). Antagonizing STK25 Signaling Suppresses the Development of Hepatocellular Carcinoma Through Targeting Metabolic, Inflammatory, and Pro-Oncogenic Pathways. Cellular Molecular Gastroenterology Hepatology, 13(2):405-423. doi: 10.1016/j.jcmgh.2021.09.018. II. Anand SK, Caputo M, Xia Y, Andersson E, Cansby E, Sima Kumari S, Henricsson M, Porosk R, Keuenhof K, Höög JL, Nair S, Marschall HU, Bluher M, and Mahlapuu M. (2022). Inhibition of MAP4K4 Signaling Initiates Metabolic Reprogramming to Protect Hepatocytes from Lipotoxic Damage. Journal of Lipid Research, 63(7):100238. doi: 10.1016/j.jlr.2022.100238. III. Cansby E, Kumari S, Caputo M, Xia Y, Porosk R, Robinson J, Wang H, Olsson, B-M, Vallin J, Grantham J, Soomets U, Svensson L T, Sihlbom C, Marschall HU, Edsfeldt A, Goncalves I, and Mahlapuu M. (2022). Silencing of STE20-Type Kinase STK25 in Human Aortic Endothelial and Smooth Muscle Cells Is Atheroprotective. Communications Biology, 19;5(1):379. doi.org/10.1038/s42003-022-03309-9. IV. Mahlapuu M, Caputo M, Xia Y, Cansby E. (2022). GCKIII Kinases in Lipotoxicity: Roles in NAFLD and Beyond. Hepatology Communications, 6(10): 2613–2622. doi.org/10.1002/hep4.2013. V. Caputo M, Xia Y, Anand SK, Cansby E, Andersson E, Marschall HU, Königsrainer A, Peter A, and Mahlapuu M. (2023). STE20-Type Kinases MST3 and MST4 Promote the Progression of Hepatocellular Carcinoma: Evidence from Human Cell Culture and Expression Profiling of Liver Biopsies. FASEB J, 37(8):e23105. doi: 10.1096/fj.202300397RR. VI. Caputo M, Andersson E, Xia Y, Hou W, Cansby E, Erikson M, Lind E D, Hallberg B, Amrutkar M, and Mahlapuu M. (2024). Genetic Ablation of STE20-Type Kinase MST4 Does Not Alleviate Diet-Induced MASLD Susceptibility in Mice. International Journal of Molecular Science, 19;25(4):2446. doi: 10.3390/ijms25042446. v CONTENT ABBREVIATIONS .............................................................................................................................. vii 1. INTRODUCTION .............................................................................................................................. 1 1.1 Metabolic Dysfunction-Associated Steatotic Liver Disease ........................................................ 1 1.2 Hepatic Lipid Metabolism and Dysfunction in MASLD ......................................................... …3 1.3 Progression from MASLD to HCC.......................................................................................... …7 1.3.1 Pathogenesis of MASH-associated HCC ............................................................................ 7 1.3.2 Genetic alterations in MASH-related HCC....................................................................... 10 1.3.3 Treatment of MASH-driven HCC ..................................................................................... 10 1.4 STE20-Type Kinases TAOK1, TAOK3, and STK25 ............................................................ …11 1.4.1 Overview of STE20-type kinases ..................................................................................... 11 1.4.2 Function of STE20-type kinases TAOK1 and TAOK3 .................................................... 11 1.4.3 Function of STE20-type kinase STK25 ............................................................................ 15 2. AIM ................................................................................................................................................... 21 3. METHODS ....................................................................................................................................... 23 3.1 Ethical Statements ...................................................................................................................... 23 3.2 Human Subjects ......................................................................................................................... 23 3.2.1 Patient cohorts .................................................................................................................. 23 3.2.2 Data collection from public databases ............................................................................. 24 3.3 Cell Culture ................................................................................................................................ 24 3.3.1 Cells ................................................................................................................................ 24 3.3.2 Transient transfections and incubation with anti-HCC drugs ......................................... 25 3.3.3 Assessment of lipid metabolism ..................................................................................... 25 3.3.4 Measurement of carbohydrate metabolism ..................................................................... 26 3.3.5 Evaluation of tumorigenicity .......................................................................................... 27 3.4 Animal Experiments .................................................................................................................. 28 3.4.1 Mouse models ................................................................................................................. 28 3.4.2 In vivo tests ..................................................................................................................... 29 3.4.3 Biochemical assays ......................................................................................................... 30 3.5 Histological, Immunohistochemical, and Immunofluorescence Analysis ................................. 30 3.6 Reverse Transcription Quantitative PCR ................................................................................... 31 3.7 Western Blot .............................................................................................................................. 31 3.8 Statistical Analysis ..................................................................................................................... 32 4. RESULTS AND DISCUSSION ....................................................................................................... 33 5. FUTURE PERSPECTIVES .............................................................................................................. 41 5.1 Investigation of the Metabolic Effects of Genetic Ablation of TAOK1 in Mice ....................... 41 5.2 Benefit of Liver-Specific Pharmacological Inhibitors of Human STK25 to Mitigate the Initiation and/or Exacerbation of MASH-Related HCC ............................................. ... .................................. 42 ACKNOWLEDGEMENT .................................................................................................................... 44 REFERENCES ..................................................................................................................................... 46 vi ABBREVIATION 4-HNE, 4-hydroxynonenal; 8-oxoG, 8-oxoguanine; α-SMA, α-smooth muscle actin; ACC, acetyl-CoA carboxylase; AFP, α-fetoprotein; ATGL, adipose triglyceride lipase; ALT, alanine aminotransferase; ASO, antisense oligonucleotide; AST, aspartate aminotransferase; BAT, brown adipose tissue; BCA, body composition analysis; CHOP, C/EBP-homologous protein; CIDE, cell death-inducing DFFA-like effector; CPT1, carnitine palmitoyltransferase 1; DAG, diacylglycerol; DEN, diethylnitrosamine; DHE, dihydroethidium; DNL, de novo lipogenesis; EMT, epithelial-mesenchymal transition; ER, endoplasmic reticulum; ERK1/2, extracellular signal-regulated kinase 1/2; eWAT, epididymal white adipose tissue; G3P, glycerol-3-phosphate; GalNAc, N-acetylgalactosamine; GCK, germinal center kinase; GOLGA2, golgin subfamily A member 2; GTT; glucose tolerance test; HCC; hepatocellular carcinoma; H&E, hematoxylin and eosin; 1H-MRS, single-proton magnetic resonance spectroscopy; HSC, hepatic stellate cell; IHH, immortalized human hepatocyte; IR, insulin resistance; ITT, insulin tolerance test; JNK, c-Jun N-terminal kinase; vii LATS1/2, large tumor suppressor kinase 1/2; LPS, lipopolysaccharide; MAS, MASLD activity score; MASH, metabolic dysfunction-associated steatohepatitis; MASLD, metabolic dysfunction-associated steatotic liver disease; MO25, mouse protein 25; NTC, non-targeting control; PDCD10, programmed cell death 10; PEX5, peroxisomal biogenesis factor 5; PMP70, peroxisomal membrane protein 70 kDa; PPAR, peroxisome proliferator-activated receptor; ROS, reactive oxygen species; SNP, single nucleotide polymorphism; sWAT, subcutaneous white adipose tissue; sh, short-hairpin; si, small interfering; STAT3, signal transducer and activator of transcription 3; STK25, serine/threonine protein kinase 25; TAG, triacylglycerol; TAOK1, thousand and one kinase 1; TAOK3, thousand and one kinase 3; VLDL, very low-density lipoprotein; YAP, yes-associated protein. viii 1 INTRODUCTION 2 Ying Xia 1. INTRODUCTION 1.1 Metabolic Dysfunction-Associated Steatotic Liver Disease Epidemiology Metabolic dysfunction-associated steatotic liver disease [MASLD; previously referred to as non- alcoholic fatty liver disease (NAFLD)], represents a hepatic manifestation of metabolic syndrome and is emerging as the leading cause of chronic liver disease worldwide1-4. MASLD is defined by the presence of hepatic steatosis (identified via imaging or biopsy) in the context of co-existing cardiometabolic risk factors like elevated BMI, insulin resistance, hypertension, or dyslipidemia, without significant alcohol consumption history5,6. MASLD is broadly classified into two subtypes: metabolic dysfunction-associated steatotic liver (MASL), the non-progressive form of MASLD; and metabolic dysfunction-associated steatohepatitis [MASH; formerly known as non-alcoholic steatohepatitis (NASH)], the progressive form of MASLD, which in addition to hepatic steatosis is characterized by local inflammation and cell damage (ballooning), carrying an increased risk of developing hepatocellular carcinoma (HCC) (Figure 1)7-12. Figure 1. Disease progression of MASLD to HCC. In the context of obesity, excessive lipids accumulate in the liver causing MASLD. About 20% of patients with MASL further progress to MASH. Patients with MASH are, in turn, at high risk of developing HCC. There is a close link between MASLD and several metabolic conditions or disturbances, with respective prevalence among MASLD patients being 23% for type 2 diabetes, 51% for central obesity, 69% for dyslipidemia, and 43% for metabolic syndrome13. Consequently, global MASLD prevalence has increased from 25.3% (21.6 to 29.3%) in 1990-2006 to 38.0% (33.7 to 42.5%) in 2016-2019 (p<0.001) in parallel to rising rates of obesity14. Currently, MASLD is estimated to afflict over one billion individuals globally, with prevalence ranging from 25.1% in Western Europe to 44.4% in Latin America (Figure 2)14. The geographical variation in the incidence of MASLD among different ethnic groups is perhaps partly secondary to a distinct frequency of genetic risk variants associated with the disease (e.g., rs738409 in PNPLA3) and may also reflect different obesity and overweight rates15-17. Figure 2. Prevalence of MASLD according to global regions data collected from 1990 to 2019. Figure adapted from Younossi et al14. 1 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis Pathogenesis The development and progression of MASLD are driven by multifactorial mechanisms. Initially, the “two-hit” hypothesis was proposed, suggesting that excessive hepatic lipid accumulation (the first hit) sensitizes the liver to further damage through inflammatory cascades and fibrogenesis (the second hit)18. However, it became rapidly evident that this view is too simplistic to recapitulate the complexity of the human MASLD. The “two-hit” hypothesis has now been replaced by a “multi-hit” model that incorporates multiple interrelated processes, including lipotoxicity, adipose tissue dysfunction, and the microbiome against a background of genetic and environmental factors (Figure 3). Figure 3. The “Multi-hit” hypothesis for the development of MASLD. FFA, free fatty acid. Figure adapted from Buzzetti et al19. Insulin resistance (IR) is a key contributor to the development of MASLD, as it enhances hepatic de novo lipogenesis (DNL) and facilitates adipose tissue lipolysis, resulting in an increased flux of fatty acids to the liver20,21. Additionally, IR induces adipose tissue dysfunction, leading to altered production and secretion of adipokines and inflammatory cytokines21,22. Skeletal muscle IR, one of the earliest defects linked to metabolic syndrome and prediabetes, can also contribute to the development of MASLD via increasing hepatic DNL and hypertriglyceridemia by diverting ingested glucose away from skeletal muscle glycogen synthesis and into the liver7. The resultant hepatic lipid overload leads to liver steatotoxicity, which further activates mitochondrial dysfunction with oxidative stress, generation of reactive oxygen species (ROS), and endoplasmic reticulum (ER) stress6. Moreover, dysbiosis of the gut microbiota results in the production of fatty acids in the bowel and augmented intestinal permeability and, thus, exacerbates fatty acid absorption and contributes to a meta-inflammatory state characterized by up-regulated release of pro-inflammatory cytokines such as IL-6 and TNF-α23. In individuals predisposed by genetic factors or epigenetic modifications, all these factors collectively contribute to hepatocyte fat accumulation and liver inflammatory environment, ultimately leading to the development of MASLD. 2 Ying Xia Management Due to the limited knowledge of disease pathogenesis, until 14 Mar 2024, no licensed pharmacotherapy was available for MASLD/MASH and clinical recommendations focused mainly on weight loss- directed lifestyle interventions24. Substantial evidence indicates that weight loss not only decreases liver fat content but also leads to histological resolution of MASH, and almost half of patients with 10% weight loss will have fibrosis regression25. However, lifestyle modifications remain challenging for most patients, with only 10% of subjects managing to achieve a 10% weight loss even in controlled clinical trial settings26. Once MASH patients develop decompensated cirrhosis, the only treatment option is liver transplantation. According to data from the European Liver Transplant Registry (ELTR) and the United Network for Organ Sharing (UNOS) databases, MASLD and MASH have become the fastest growing indications for liver transplantation in the Europe and USA27,28. Although short- and mid-term overall survival rates for patients undergoing liver transplantation due to MASH-related cirrhosis are comparable to those for other liver disease indications, patients with MASH-related cirrhosis face higher waiting list mortality and are at an elevated risk for recurrent liver disease and cardiometabolic complications in the long-term post-transplantation29. Recently, resmetirom (previously known as MGL-3196), which is a liver-targeted thyroid hormone receptor (THR)-β selective drug, has been approved by the FDA for the treatment of adults with non- cirrhotic MASH with moderate to advanced fibrosis30. However, resmetirom is effective only in <30% of patients at one-year follow-up, and no long-term data are currently available30. There are two additional drug candidates currently in the late stages of the clinical program (e.g., phase 3): lanifibranor [an oral pan-peroxisome proliferator-activated receptor (PPAR) agonist] and semaglutide [a subcutaneous glucagon-like peptide 1 (GLP-1) receptor agonist], which are expected to reach a decision on a possible approval within a few years31. No MASH-targeted pharmacotherapy can, as of now, be recommended for the cirrhotic stage. 1.2 Hepatic Lipid Metabolism and Dysfunction in MASLD Lipid anabolism Lipid droplets are ubiquitous cellular organelles responsible for storing neutral lipids for subsequent utilization during energy deficit6,32,33. Hepatocytes, the functional unit of the liver parenchyma, serve as a major site for lipid metabolism, including the uptake, esterification, oxidation, and secretion of fatty acids in response to nutritional and hormonal signals (Figure 4). 3 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis Figure 4. A working model of lipid anabolism and catabolism. GDK, diacylglycerol kinase. Figure adapted from Huh et al34. Fatty acids in the liver originate from dietary sources (comprising 15-30% of the liver fatty acid pool), hepatic DNL (accounting for up to 30% of liver fatty acids during feeding), and the recycling of fatty acids released into the circulation from peripheral adipose tissue during fasting35. Free fatty acids enter hepatocytes via protein transporters such as fatty acid transporter proteins (FATPs), fatty acid binding proteins (FABPs), and fatty acid translocase (FAT; also known as CD36)36. Upon uptake, a major branchpoint in the metabolism of fatty acids is their partitioning between esterification into triacylglycerols (TAGs) and oxidation. Numerous long-chain acyl-CoA synthetases (ACSLs) catalyze the conversion of fatty acids to their respective CoAs, affecting their downstream fates. In the liver, ACSL5 is required as a critical isoform contributing to TAG synthesis37, whereas ACSL4 is thought to regulate arachidonic acid metabolism, which in turn affects phospholipid composition and very-low- density lipoprotein (VLDL) secretion38. In addition to the delivery from circulation, fatty acids are also produced within hepatocytes via DNL, where acetyl-CoA carboxylase (ACC) and fatty acid synthase (FAS) are responsible for carrying out the committed steps to generate acyl-CoAs39. In most mammalian cells, the glycerol-3-phosphate (G3P) pathway contributes to over 90% of the total TAG synthesis40. Here, the initial step involves the esterification of long-chain acyl-CoA to G3P, catalyzed by mitochondrial and microsomal G3P acyltransferase (GPAT) enzymes. The resultant lysophosphatidic acid (LPA) is then acylated by acylglycerol-3-phosphate acyltransferases (AGPATs) in the ER membrane to form phosphatidic acid (PA). Subsequently, PA is converted into diacylglycerol (DAG), followed by the acylation of DAG into TAG via DAG acyltransferases (DGATs). The 4 Ying Xia acyltransferase enzymes in this pathway possess various isoforms with distinct substrate specificities, dictating the downstream fatty acid composition. For instance, DGAT1 preferentially channels exogenous fatty acids into TAGs, whereas DGAT2 is more selective to saturated and monounsaturated FAs (MUFAs) derived from DNL41-43. In addition to TAGs, hepatocytes have a substantial capacity to synthesize and store cholesterol esters via acyl CoA:cholesterol acyltransferase (ACAT) enzymes, ACAT1 and ACAT2 (also known as sterol O-acyltransferase 1 and 2, respectively)35. The process of packaging neutral lipids, such as TAGs, into lipid droplets is only beginning to be understood. A current model suggests that lipid droplets form in ER, where neutral lipids accumulate within the leaflet of the membrane bilayer. Upon reaching a critical concentration, phase separation occurs to generate lipid lenses. These lenses expand and distort the ER bilayer membrane until they form a certain size, initiating the budding of lipid droplets into the cytoplasm. Newly formed small nascent lipid droplets may expand through a fusion process that involves cell death-inducing DFFA- like effectors (CIDEs). Under normal conditions, CIDEB is the most abundantly expressed isoform; however, during fasting and steatosis, the expression of CIDEA and CIDEC is increased to facilitate lipid droplet fusion44,45. In addition to fusion, it is hypothesized that lipid droplets may expand through lipid synthesis occurring directly on their surfaces or at newly formed bridges between the lipid droplets and the ER membrane32. The monolayer membrane of lipid droplets houses various enzymes critical for TAG synthesis, including acyl-CoA synthetases and the acyltransferases GPAT4 and DGAT2, which facilitate the expansion of existing lipid droplets46. Lipid catabolism Fatty acids can be released from the neutral lipid pool for metabolic fuel. This mobilization primarily occurs through canonical lipolysis, wherein a series of lipases (adipose triglyceride lipase [ATGL], hormone-sensitive lipase [HSL], and monoglyceride lipase [MGL]) sequentially act on the surface of lipid droplets, reducing TAGs into free fatty acids32. Within the past decade, selective autophagy (also called lipophagy) has emerged as an alternative mechanism for hepatic lipid consumption. Here, lipid droplets are first engulfed by a membrane bilayer to form an autophagosome, which then fuses with a degradative lysosome to form an autolysosome. Within the autolysosome, TAGs are hydrolyzed by lysosomal acid lipase (LAL) to free fatty acids32. Free fatty acid released from hepatocellular lipid droplets can undergo oxidation or secretion as VLDL- TAG. There are three main routes for fatty acid oxidation: mitochondrial β-oxidation, peroxisomal β- oxidation, and ω-oxidation. Mitochondrial β-oxidation is the predominant oxidative pathway, responsible for the oxidation of short-chain (C20), which are subsequently oxidized in mitochondria once their chain has been shortened47. Very-long-chain fatty acids can also be metabolized by the cytochrome P450 4A ω-oxidation system to dicarboxylic acids48. VLDL formation commences with the synthesis of apolipoprotein-B100 (apoB100)49 and is followed by two stages of lipidation50,51. While the exact mechanism of VLDL transport out of the liver is not fully elucidated, it is known that apoB100 exits the ER in COPII vesicles52. TANGO1 and TALI (Mia2- cTAGE5 fusion) facilitate the export of VLDL from the ER by promoting the fusion of the ER-Golgi intermediate compartment membrane with ER exit sites53,54. Ultimately, VLDL particles are secreted into the plasma where they are taken up by other organs for energy49. Lipid droplet proteins The surface of lipid droplets contains various proteins involved in lipid metabolism, membrane trafficking, and protein degradation. Depending on their subcellular location, these proteins can be divided into two classes: ER bilayer proteins and cytosolic proteins. ER bilayer proteins localize within the ER bilayer and are characterized by the presence of a hydrophobic hairpin membrane-embedded domain. This structural feature facilitates the translocation of these proteins between the ER bilayer and the monolayer of lipid droplets by an ER-lipid droplet membrane bridge55. In the liver, this class of proteins encompasses TAG biosynthetic enzymes such as PNPLA3, DGAT2, ACSL3, and GPAT46. Cytosolic proteins are localized in the cytosol and target the lipid droplet surface through a hydrophobic domain. Within this class are CTP cytidylyltransferase (CCT), perilipins (PLINs), hypoxia-inducible factor-2α (HIF-2α), and CIDE6. Changes of lipid metabolism in MASLD and MASH At the molecular level, the initial accumulation of lipid droplets in the liver may represent an adaptive response to an imbalance between fatty acid anabolism and catabolism. In obese or overweight individuals (defined by BMI≥23 kg/m2 in Asians and ≥25 kg/m2 in non-Asians), the liver primarily compensates for the overspill of non-esterified fatty acids from insulin-resistant adipose tissue. In this context, excessive free fatty acids enhance hepatic TAG storage and gluconeogenesis, when the capacity of hepatocytes to oxidize these substrates or to export them as VLDL is exhausted. Notably, although adipose tissue-derived fatty acids are the predominant source of hepatocellular TAGs, hepatic DNL rates are also markedly elevated in obese MASLD patients, further exacerbating intrahepatic lipid accumulation56. Moreover, MASLD can also develop in lean individuals (defined by BMI<23 kg/m2 in Asians and <25 kg/m2 in non-Asians)57, and in this case, similarly to obese MASLD subjects, IR contributes to MASLD directly by increasing hepatic DNL and indirectly by enhancing the delivery of free fatty acids to the liver via decreased inhibition of lipolysis in the fat depots58. Following neutral lipid synthesis and lens formation, cytoplasmic lipid droplet biogenesis may either proceed with 6 Ying Xia budding towards the cytosol (lipid with an inverted shape) or remain embedded in the ER bilayer (lipids with a cone shape, such as DAGs). In the latter scenario, proteins and enzymes cannot access lipid droplets to regulate their neutral lipid incorporation, mobilization, or their protein composition. Therefore, embedded lipid droplets fail to remove unfolded and misfolded proteins from the ER membrane, leading to ER stress, ROS overproduction, and further exacerbating excessive lipid droplet accumulation in the liver6. The presence of hepatocellular injury (ballooning) due to dysregulated lipid metabolism is a crucial concept in understanding the drivers of MASH6. Ballooned hepatocytes exhibit increased release of fatty acids from the core of lipid droplets, activating c-Jun N-terminal kinase (JNK) signaling pathways that induce cellular stress, inflammation, apoptosis, and mitochondrial dysfunction59. Furthermore, the JNK pathways phosphorylate PPARγ that, in a vicious cycle, suppress fatty acid β-oxidation with consequent exacerbation of lipotoxicity and hepatic inflammation59. Additionally, the release of fatty acids from the core of lipid droplets leads to the accumulation of DAGs, which contributes to IR by activating protein kinase C (PKC) and protein kinase D (PKD) that, in turn, stimulate DNL and increase the flux of fatty acids from adipose tissue to the liver60. Numerous studies have linked the susceptibility and severity of MASLD to the proteins that bind the surface of lipid droplets. For instance, PNPLA3 hydrolyzes TAGs and catalyzes the transfer of the polyunsaturated fatty acids from di- and tri-acylglycerols to phosphocholinesin within hepatocytes; however, a polymorphism characterized by the substitution of isoleucine to methionine in position 148 (p.I148M) results in a loss of function61. PNPLA3 p.I148M is not degraded by the ubiquitin proteasome system and autophagy and therefore, leads to retention of TAGs and polyunsaturated fatty acid-enriched lipid droplets, priming accumulation of liver fat62. A recent investigation has also revealed that the hepatocytes expressing the PNPLA3 p.I148M variant undergo lipid remodeling, featured by a reduction in polyunsaturated fatty acids and an elevation in ceramides. This lipid composition alternation subsequently activates inflammatory cytokines, leading to hepatic cell injury63. Moreover, PNPLA3 expression in liver biopsies from MASH patients positively correlates both with fibrosis stage and hepatic α-smooth muscle actin (α-SMA) level, independently of the genotype64. In addition, PLIN2, which coats lipid droplet membranes and prevents lipolysis, is implicated in the development of liver steatosis and IR65. In MASLD, PLIN2 is upregulated and is associated with the hepatic accumulation of ceramides66. 1.3 Progression from MASLD to HCC 1.3.1 Pathogenesis of MASH-associated HCC HCC is the third leading cause of cancer death globally and the malignancy that displays the steepest increase in both incidence and mortality in the Western world67-69. Despite the application of various locoregional and systemic therapies, most individuals with HCC develop disease progression with a low 5-year survival rate of 20% and a high recurrence rate of about 80%70. Recently, the trend in HCC 7 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis occurrence has shifted epidemiologically from subjects with virus-related liver disease to those with MASLD/MASH4,70,71. The initial association between MASH and HCC was documented in 2001, and to date, MASH has been reported in several studies as the most rapidly rising underlying etiology of HCC67,68,72,73. Globally, approximately 10% of all HCC cases are attributed to MASLD/MASH, with the higher estimates (>20%) reported in studies from the USA, UK, India, Germany, and the Middle East, and lower estimates (1-2%) from China and Japan11. Advanced age, male sex, and the presence of MASH-related cirrhosis are pivotal risk factors predisposing individuals with MASLD to develop HCC. Additionally, lifestyle, genetic susceptibilities, metabolic dysfunction, altered microbiome composition, and the severity of MASLD itself significantly affect the propensity for HCC progression in this patient cohort (Figure 5). Figure 5. Risk factors for MASLD-related HCC. HDL, high-density lipoprotein; LDL, low-density lipoprotein. Figure adapted from Yip et al74. As stated above, in the context of MASLD/MASH, IR in adipose tissue results in enhanced release of free fatty acids and delivery to the liver, facilitating excessive hepatic lipid accumulation and steatotoxicity. Within hepatocytes, the enhanced lipotoxicity can generate pathogenic drivers of carcinogenesis, particularly oxidative DNA damage, which has been found higher in MASH-HCC patients than MASH subjects without HCC75. As a substrate for oxidant generation, hepatic steatosis can also result in chronic low-grade inflammation, triggering increased levels of IL-6, leading to activation of signal transducer and activator of transcription 3 (STAT3), which stimulates hepatocyte proliferation and malignant transformation76,77. In addition, hepatic mitochondrial dysfunction induced by fat overload is known to force a higher degree of fatty acid oxidation to peroxisomes, resulting in exaggerated oxidative and ER stress, which, in a vicious cycle, promotes liver inflammation, fibrosis, 8 Ying Xia and hepatocellular apoptosis and proliferation, thereby driving the progression from MASLD to HCC78- 80. Patients with MASH-driven HCC display increased expression of the hepatocellular cholesterol biosynthetic enzyme squalene epoxidase (SQLE), which elevates NADP+ and cholesterol ester levels, thereby epigenetically silencing the tumor suppressor phosphatase and tensin homolog (PTEN), thus promoting cell proliferation and oxidative stress81. Consistently, the use of terbinafine, a clinically approved antifungal medication to suppress SQLE, has been shown to mitigate tumorigenesis in both cultured HCC cells and animal models of HCC81. In addition, a recent study has demonstrated that ND- 654, a liver-specific inhibitor of ACC (an enzyme promoting lipid storage by both stimulating lipogenesis and inhibiting lipid oxidation), improves survival of HCC-bearing rats by inhibiting tumor proliferation82. Autophagy, a lysosomal degradative pathway that plays a crucial role in lipid catabolism as described above, is also essential for maintaining cellular health in hepatocytes. Currently, the experimental evidence suggests that the role of autophagy in the occurrence and development of HCC is dependent on the context of hepatocytes. Within dysplastic hepatocytes, basal autophagy acts as a tumor suppressor by removing newly damaged mitochondria and mutated cells and thus maintaining genomic stability83. However, once the tumor is established, autophagy stimulated by oxidative stress activates the transcriptional activity of nuclear factor erythroid 2-related factor (NRF2) to generate an antioxidant response, which can promote survival not only of steatotic hepatocytes but also of tumor cells84. Furthermore, autophagy inhibitor chloroquine has been demonstrated to exert a tumor-suppressive effect in diethylnitrosamine (DEN)-induced HCC rat model in the tumor-forming stage, whereas it displays a tumor-promoting effect in the dysplastic stage85. The increased stiffness of the liver, resulting from the deposit of extracellular matrix, can create a conducive environment for tumor cell proliferation. The accrued scar matrix acts as a reservoir for bound growth factors, including insulin-like growth factors, which may augment the survival of pre- neoplastic hepatocytes, thereby fostering tumor initiation and progression86. Nevertheless, a lower prevalence of cirrhosis has been reported in MASH-driven HCC compared to HCC associated with other etiologies such as viral hepatitis and excessive alcohol consumption87. The precise mechanisms driving the development of MASH-related HCC in the absence of cirrhosis remain elusive; however, they may be linked to the status of liver fibrosis. Although the majority of studies on HCC in the context of MASH emphasize dysregulated pathways in hepatocytes, other resident and infiltrating liver cells also contribute to hepatocarcinogenesis. For instance, toll-like receptor 4 (TLR4) signaling in macrophages and stellate cells induces inflammasome activation, leading to pro-inflammatory and fibrogenic responses88. Stellate cell activation and subsequent fibrogenesis can further aggravate tumorigenesis, including the generation of factors that promote hepatocyte growth and survival as well as tumor evasion7. 9 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis 1.3.2 Genetic alterations in MASH-related HCC Genetic predispositions contributing to HCC development in individuals with MASLD remain insufficiently explored. To date, the best validated genetic risk of HCC is the PNPLA3 variant (PNPLA3 rs738409, c.444 C>G p.I148M), which, based on EASL clinical practice guidelines, is associated with HCC development in obese individuals and in patients with histologically proven MASLD89-91. Previous studies have also identified the rs641738 variant of MBOAT7 as exacerbating the risk of HCC in non-cirrhotic Italian people with MASH92, and a variant of TM6SF2 (rs58542926) that promotes MASH with an independent effect on HCC risks93. A recent investigation additionally links GCKR (rs1260326) to a 1.8-fold increased risk of MASH-associated HCC compared with the general population94. It’s noteworthy that no single genetic variant is capable of adequate risk stratification in HCC, whereas combining numerous variants in polygenic risk scores is an attractive approach. 1.3.3 Treatment of MASH-driven HCC The management of HCC has witnessed significant advancements due to the first Barcelona Clinic Liver Cancer (BCLC) classification in 199995. As the heterogeneous BCLC-C category, encompassing the majority of potentially treatable patients with MASH-related HCC, continues to expand, its value as an aid for treatment stratification has been diminished96. Since 2007, the FDA has licensed twelve anti-HCC drugs, including five multi-kinase inhibitors and seven immune checkpoint blockade therapies; however, these approaches show less effect in patients with MASH-driven HCC94,97-99. Moreover, as previously mentioned, patients with MASH-associated HCC have a high prevalence of comorbidities (such as cardiovascular disease) that might preclude the implementation of curative therapies, particularly surgical procedures. Currently, a multitude of small molecules targeting various pathways, such as anti-inflammatory, antioxidant, anti-fibrotic, anti-lipid metabolic, microbiota- pertaining, and immunomodulatory mechanisms, have been developed to impede the progression of MASH-related HCC (Figure 6). While none are currently approved by the FDA, they hold promise for hindering MASH-HCC progression. Figure 6. Mechanistic classes of example agents against MASH-driven HCC. Figure adapted from Wang et al100. 10 Ying Xia 1.4 STE20-Type Kinases TAOK1, TAOK3, and STK25 1.4.1 Overview of STE20-type kinases Kinases are enzymes that catalyze covalent modification of proteins by attaching phosphate groups (from ATP or GTP) mainly to serine, threonine, and/or tyrosine residues101. The human kinome features a large branch of STE20-type kinases, named after the founding member yeast Sterile20 kinase involved in the mating pathway102. Mammalian STE20-type kinases (~30 described to date) are distinguished by a high degree of homology within the catalytic domain, and they fall into two subfamilies: the p21- activated kinases (PAKs) with a C-terminal kinase domain and an N-terminal p21 GTPase-binding domain and the germinal center kinases (GCKs), which have an N-terminal kinase domain and lack GTPase-binding domains (Figure 7)103. Functionally, STE20-type kinases have been demonstrated to regulate a broad range of biological processes, including cell differentiation, proliferation, apoptosis, motility, polarity, metastasis, angiogenesis, and stress responses104-112. Figure 7. Phylogenetic tree of STE20-type kinases. Figure adapted from Strange et al113. 1.4.2 Function of STE20-type kinases TAOK1 and TAOK3 Structure and expression pattern of TAOK1 and TAOK3 In the search for mammalian orthologs of STE20-type kinases, the TAO kinases, belonging to GCKVIII subfamily, have been discovered. The first member, thousand and one kinase 1 (TAOK1; also known as MAP3K16 or PSK2), was identified in the rat by fishing the cDNA library with a degenerate STE20 kinase probe114. It is named “thousand and one” because 1001 amino acids are encoded by the TAOK1 gene. Subsequently, a second TAO kinase TAOK2 (also known as MAP3K17 or PSK1) was characterized115. The third member, human TAOK3 (also known as MAP3K18, JIK, or DPK), was later identified by its association with EGFR kinase substrate 8 (EPS8) in an expression library assay116. In this thesis, we focus only on TAOK1 and TAOK3. In humans, TAOK1 is located on chromosome 17q at position 11.2, while TAOK3 is on 12q24.23. They share similar domain structures despite the difference in the amino acid length. At the N-terminal end of these two kinases, a highly conserved kinase domain (88.6% similarity in amino acid sequence) 11 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis is followed by a serine-rich domain. The C-terminal half of TAOK1 and TAOK3 contains two and three coiled-coil regions, respectively (Figure 8). TAOK1 and TAOK3 are ubiquitously expressed, with the highest levels of TAOK1 observed in the brain and TAOK3 in myeloid/lymphoid tissues117. A recent TAOK1 phosphoregulatory network analysis has proposed Ser9 and Ser181 as autophosphorylation and kinase activity-associated phosphosites and Ser421, the most frequently detected phosphosite in TAOK1, as a significant regulatory phosphosite involved in the maintenance of genome integrity118. The activation of TAOK3 is correlated with its phosphorylation at Ser324 within the kinase domain119. Figure 8. (A) Diagram of domain structure of human TAOK1 and TAOK3. (B) AlphaFold2.0-predicted protein structures of human TAOK1 and TAOK3. Structures are color- coded based on the confidence score of prediction. AlphaFold produces a per-residue confidence score (plDDT) between 0 and 100. KD: kinase domain; S: serine-rich domain; CC: coiled-coil regions. Figure adapted from Fang et al120. Signaling pathways and cellular physiologies mediated by TAOK1 and TAOK3 The TAOK subfamily of kinases has been established as upstream regulators of the p38 MAPK cascades, JNK signaling, and Hippo pathways to control multiple cellular processes, including cytoskeleton organization, apoptosis, stress response, inflammation, and cell proliferation (Figure 9). Figure 9. Kinase cascades regulated by TAOK1 and TAOK3. Violet arrows: upstream stimuli; Turquoise arrows: canonical pathways; orange arrows: alternative pathways in the Hippo cascades; black arrows: nuclear translocation/exportation. Figure adapted from Fang et al120. 12 Ying Xia TAOK1 and TAOK3, through their activity as MAP3Ks, have been reported to activate the p38 MAPK pathways, which are involved in the regulation of several cytosolic cofactors and nuclear transcription factors121. Here, in the context of chemical and environmental insults such as ionizing and ultraviolet radiation, stimulated TAOK1 and TAOK3 can phosphorylate MKK3 (at Ser189/Thr193) and MKK6 (at Ser207/Thr211), which then phosphorylate p38 kinases (at Thr180/Tyr182) in response to DNA damage114. It is also proposed that TAOK1 and TAOK3 serve as intermediaries linking specific G protein-coupled receptors (GPCRs) to the p38 MAPK pathway122. In addition to MKK3/6 in the p38 MAPK cascade, TAOK1 is found to phosphorylate MKK4 (at Ser257/Thr261) and MKK7 (at Ser271/Thr275), further phosphorylating the JNK proteins (at Thr183/Tyr185), which are involved in cell proliferation, differentiation, apoptosis, and inflammatory responses59. The involvement of TAOK3 in the JNK pathway is somewhat perplexing. Tassi et al116 initially reported that TAOK3 inhibits basal JNK activity and reduces its phosphorylation in response to human epidermal growth factor in COS7 (monkey kidney fibroblast-like cells). In contrast, a study by Zhang et al123 found that TAOK3 activates JNK in NIH3T3 (embryonic mouse fibroblast cells). Our studies (Paper I and II of the current thesis)124,125 have shown that the phosphorylation of JNK (Thr183/Tyr185) is significantly lower in TAOK1- or TAOK3-deficient immortalized human hepatocytes (IHHs), both under basal culture conditions and after exposing cells to oleic acid. Intriguingly, TAO1 (TAOK1 ortholog in Drosophila) is found to phosphorylate and activate Hippo pathway, which then functions to restrict cell proliferation in Drosophila126. Similarly, studies in HEK293 (human embryonic kidney cells) have demonstrated that mammalian TAOK1 and TAOK3 can either phosphorylate mammalian sterile-20-like kinase 1/2 (MST1/2) at residues Thr183 on MST1 and Thr180 on MST2, or bypass MST1/2 and phosphorylate large tumor suppressor kinase 1/2 (LATS1/2) directly at residue Thr1079 on LATS1 and Thr1041 on LATS2, which leads to phosphorylation of yes-associated protein (YAP; at Ser127) and transcriptional coactivator with PDZ-binding motif (TAZ; at Ser89), thereby inhibiting YAP/TAZ-mediated transcription in the nucleus and in turn limiting cell proliferation126-128. Additionally, Meng et al129 has further reported that knockout of MAP4K4/6/7 significantly blocks TAOK1-induced YAP phosphorylation in HEK293 cells, suggesting that TAOK1 may also act through MAP4K4/6/7 to activate LATS1/2. However, our studies (Paper I and II of the current thesis)124,125 found that the in vitro silencing of TAOK1 or TAOK3 does not affect the phosphorylation of YAP in IHHs. In addition to regulating the aforementioned signaling pathways, TAOK1 and TAOK3 are reported to interact with other cytosolic proteins and be involved in additional physiological processes. For instance, TAOK1 and TAOK3 are implicated in the regulation of inflammation. Specifically, TAOK1 negatively regulates IL-17-mediated proinflammatory cytokine production in a kinase activity- independent manner via interaction with IL-17RA through its C-terminal domain and inhibition of MAPKs and nuclear factor κB (NF-κB)130. In contrast, TAOK1 is also found to increase the lipopolysaccharide (LPS)-induced production of pro-inflammatory cytokines, including IL-6, TNF-α, 13 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis and IL12p40, via interaction with tumor necrosis factor receptor-associated factor 6 (TRAF6) through its C-terminal domain and activation of extracellular signal-regulated kinase 1/2 (ERK1/2)131. TAOK3 has been reported to function as a negative regulator of inflammation and proinflammatory responses in macrophages132. Additionally, TAOK1 can induce substantial changes in the cytoskeleton through its ability to regulate three major cytoskeletal components: microtubules, actin, and septins133-135. TAOK1 is activated during mitosis with increased levels of phosphorylation at the regulatory site Ser181; once activated, TAOK1 localizes to the cytoplasm and spindle poles, where it plays a critical role in regulating mitotic cell rounding and spindle positioning136. TAOK1 activation in H1299 (human non- small-cell lung carcinoma cells) also leads to cell contraction, membrane blebbing, cleavage of Rho kinase 1 and caspase 3, and activates the JNK pathway to induce apoptosis137. Role of TAOK1 and TAOK3 in liver lipid metabolism and MASLD development Our previous studies have identified TAOK1 and TAOK3 as hepatic lipid droplet-binding proteins, based on a global proteomic analysis of lipid droplet fraction isolated from livers of high-fat diet-fed mice138,139. Consistently, we here observed that TAOK1 and TAOK3 were concentrated on the surface of intrahepatocellular lipid droplets in cultured human hepatocytes and in liver sections from high-fat diet-fed mice as demonstrated by immunofluorescence microscopy, as described in Paper I and Paper II of the current thesis, respectively124,125. In the same studies, we found that the hepatic mRNA levels of TAOK1 and TAOK3 were significantly and positively correlated with liver lipid content and MASLD activity score (MAS; composed of histological scores for liver steatosis, lobular inflammation, and hepatocellular ballooning) in human liver biopsies124,125. Furthermore, we showed that the in vitro silencing of TAOK1 or TAOK3 in human hepatocytes reprogrammed cellular metabolism by stimulating lipid catabolism (mitochondrial β-oxidation and TAG efflux) and inhibiting lipid anabolism (fatty acid influx and lipogenesis), collectively lowering ectopic fat storage within intrahepatocellular lipid droplets124,125. Notably, both the rate of canonical lipolysis and lipophagy, facilitating lipid mobilization from the lipid droplet for β-oxidation and secretion, were significantly increased in TAOK1-deficient hepatocytes125. In parallel with the reduced fat accumulation, we discovered markedly lowered incidences of oxidative/ER stress in hepatocytes where TAOK1 or TAOK3 was knocked down124,125. The in vivo significance of TAOK1 in MASLD development remains unknown. Previous studies have shown that TAOK3 is involved in childhood obesity, with low methylation of the CpG locus Cg17627898 in whole blood samples from 78 obese patients, and a 1% increase in TAOK3 methylation multiplicatively reducing the likelihood of obesity by 0.91 (95% CI: 0.86-0.97)140. Here, Paper III of the current thesis demonstrated that, in contrast to the in vitro observations, mice with genetic ablation of Taok3 gene and their wild-type littermates developed a similar degree of high-fat diet-induced liver steatosis, inflammation, and fibrosis, and no difference was observed in lipotoxic damage of adipose tissue, kidney, or skeletal muscle when comparing the two genotypes141. In addition, TAOK3 deficiency 14 Ying Xia had no impact on body weight or composition, food consumption, locomotor activity, or systemic glucose or insulin homeostasis in obese mice141. The lack of hepatic phenotype in Taok3–/– mice in vivo may be attributable to the liver-specific compensation response for the genetic loss of TAOK3 by related STE20-type kinases, including TAOK1, MST3, MST4, STK25, and MAP4K4, which were up- regulated in the livers of knockout mice and have been previously implicated to increase the risk of MASLD susceptibility125,138,139,142-149. In contrast, Maes et al found that elimination of TAOK3 in mice promoted high-fat diet-induced systemic glucose intolerance and insulin resistance150. It is possible that the differences in the genetic background of mouse stains studied, and the composition of the high-fat diets used, have contributed to this variability in results. 1.4.3 Function of STE20-type kinase STK25 Domain Structure of STK25 Serine/threonine-protein kinase 25 (STK25; also known as YSK1 or SOK1) belongs to the GCKIII subfamily of STE20-type kinases together with MST3 (also known as STK24) and MST4 (also known as STK26 or MASK)151. In humans, STK25 is located on chromosome 2q37.3. The human STK25 gene consists of 12 exons, with exon 1 encoding a 5’ untranslated region, while exons 2-12 encode an mRNA that translates into a 426-residue protein (Figure 10)152,153. STK25 is highly conserved in mice and humans (94% and 99% identity on gene and protein level, respectively). Figure 10. Domain structure and sequence of human STK25. (A) Diagram of domain structure of human STK25. (B) AlphaFold2.0-predicted protein structures of human STK25. (C) Sequence of human STK25. Figure adapted from Mahlapuu et al151. At the N-terminal end of STK25, a short variable sequence is followed by a highly conserved catalytic domain derived from exons 3-7 and the 5’ region of exon 8154. This 251-residue catalytic domain encompasses the ATP-binding site [GXGX(F)GX16K] and a protein substrate binding site, which also contains the STE20 signature peptide (GTPFWMAPE)155. Molecular modeling and crystallographic studies have revealed that the catalytic region of STK25 displays a typical structure of protein kinase 15 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis domain, with a smaller N-lobe and a larger C-lobe, and the substrate-binding cleft placed between them155. Notably, the lysine residue (Lys49) and the threonine residue (Thr174) within the kinase domain have been identified as instrumental for the catalytic activity of STK25151. The regulatory domain of STK25 lies at the C-terminal to the catalytic domain and carries a putative bipartite nuclear localization signal at its N-terminal end, which is relatively well-conserved156. This sequence is followed by a variable region of 133 residues encoded by the 3’ region of exon 8 and exons 9-12, which is predicted to be highly disordered and is suggested to interact with various signaling molecules and regulatory proteins155. The regulatory regions are also likely to participate in homodimerization of STK25 although the exact residues involved in these interactions are not clearly determined157. Molecular modeling has confirmed that the regulatory domain of STK25 is highly disordered155. However, intrinsically disordered proteins can adopt ordered conformations upon interaction with appropriate ordered macromolecular ligands. In signaling proteins, such disordered regions may function as “hubs” within signaling networks through these interactions. Activation, interactions, and inhibitors of STK25 Several inducers of oxidative stress, such as H2O2 and menadione, have been shown to activate STK25152. In contrast, STK25 activation was not observed in response to growth factors, osmotic stress, heat shock, or cytokines152. An increase in the concentration of cytosolic-free Ca2+ due to oxidative stress and chemical anoxia has also been shown to activate STK25158. In addition to activation by upstream signal(s), STK25 appears to be activated by constitutive autophosphorylation of Thr174; however, whether autophosphorylation occurs by a trans or a cis mechanism is unclear155. Mouse protein 25 (MO25) operates as a critical scaffolding subunit of the LKB1 tumor suppressor protein kinase complex and has evolved as a master activator of several STE20 kinases159. A previous yeast two-hybrid (Y2H) screen of a human placenta library with MO25 as “bait” identified MST3 and MST4 as interactants, but not STK25160. However, a subsequent structural study revealed that MO25 can tightly interact with the N-lobe, the αE helix, and the WxF motif of STK25, forming a four-site interface to promote its activation161. Binding assays using purified recombinant proteins discovered an interaction between the coiled-coil structure of Golgin subfamily A member 2 (GOLGA2; also known as GM130) and residues 270-302 of STK25 (encompassing the conserved N-terminal of the regulatory domain)162. In addition, STK25 is a component of multiprotein assemblies named STRIPAK complexes where striatin family members bind STK25 via the stabilizing scaffold programmed cell death 10 (PDCD10; also known as CCM3 or TFAR15) and orchestrate the dephosphorylation and inactivation of STK25 by STRN-associated phosphatase 2A (PP2A)155. To identify novel interacting proteins of STK25, in Paper I, we performed a genome-wide Y2H screen of a cDNA library derived from primary human hepatocytes using full-length human STK25 fused to the LexA or Gal4 DNA binding domain as bait. The screen revealed a total of 11 potential interaction partners of STK25, including previously identified interactors described above (i.e., GOLGA2, PDCD10, and MO25)124. Interestingly, when 16 Ying Xia using the Gal4-STK25 as bait, we also pinpointed TAOK3 as the binding partner for STK25, although the specific interaction site remains unknown124. To date, targeting STK25 with small molecules has few precedents. For example, J60 has been used as a ligand to STK25 in conjunction with a co-crystal structure determination of the adapter molecule MO25161. Among natural products, staurosporin, a broad-range kinase inhibitor, has been used as a test standard in inhibiting STK25163. Recently, Kiyeleko et al. developed a series of “first-in kind” STK25 inhibitors by attaching a p-N-pyrrolidinosulphonamide group to the arylamine group164. However, it’s noteworthy that, antisense oligonucleotide (ASO) therapeutics, which are used in Paper IV of the current thesis, are today generally recognized to have advantages compared to small molecule candidate drugs in terms of improved selectivity165. Role of STK25 in liver lipid portioning and MASLD development To the best of my knowledge, we are the only group to date that has reported on the function of STK25 in liver lipid metabolism. Similarly to TAOK1 and TAOK3, we found that STK25 is exclusively localized around lipid droplets, as shown by global proteomic analysis of the lipid droplet fraction isolated from steatotic livers of obese mice as well as using immunofluorescence microscopy of human hepatocytes and mouse liver sections138,139,144,166. We observed a significant positive correlation between STK25 transcript abundance in human liver biopsies and all three individual lesions of the MAS, total MAS, as well as liver fat content measured by magnetic resonance spectroscopy (1H-MRS)144,145. Moreover, mRNA levels of STK25 were about 3-fold higher in subjects with MAS≥5 (defines definite MASH) compared with participants with MAS≤4 (defines simple steatosis or borderline MASH)145. In addition, our recent studies have revealed that the in vitro silencing of STK25 in human hepatocytes markedly reduced intracellular lipid accumulation, which was mediated by increased β-oxidation and VLDL-TAG secretion (i.e., output), combined with decreased TAG synthesis (i.e., input)144. Importantly, STK25-deficient hepatocytes were also substantially protected against oxidative and ER stress138,145, which as stated above, are considered the key triggers of disease progression from simple liver steatosis to MASH (Figure 11). Reciprocally, a substantial increase in lipid deposition as well as oxidative/ER stress was detected in human hepatocytes overexpressing STK25 (Figure 11)144,145. 17 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis Figure 11. Proposed mechanism for the regulation of liver lipotoxicity by STK25. LD, lipid droplet. Figure adapted from Amrutkar et al144. We have further examined the in vivo role of STK25 in the development of MASLD and MASH. We found that, when challenged with MASH-inducing diets, the livers from Stk25 knockout mice and mice treated with Stk25-targeting ASOs, displayed substantial suppression of integral MASH features compared with wild-type livers: along with less steatosis, we observed attenuation of oxidative stress, inflammation, and fibrosis, combined with reduced hepatocellular damage (Figure 12)146-148. Reciprocally, we discovered that diet-induced MASH is markedly aggravated in Stk25-overexpressing transgenic vs. wild-type mice145,166. Figure 12. A schematic model: STK25 antagonism protects mice against diet- induced MASH. Figure adapted from Mahlapuu et al151. In a study of 430 participants from the Tübingen Family study, genotyping for 11 tagging single nucleotide polymorphisms (SNPs) of the STK25 gene revealed that three SNPs (rs34506685, rs34751932, rs4675810) were nominally associated and one SNP (rs6757649) was significantly associated with liver fat content. Homozygous minor allele carriers of rs34751932 and rs4675810 had 1.4-fold and 1.7-fold increased liver fat content, respectively, while carriers of rs6757649 and rs34506685 had 35% and 31% decreased liver fat content, respectively, compared to homozygous major allele carriers147. 18 Ying Xia Role of STK25 in hepatocarcinogenesis As demonstrated by previous investigations167,168 and Paper IV of the current thesis, hepatic STK25 expression was positively correlated with the incidence and severity of human HCC. In the same studies, STK25 knockdown and knockout human hepatoma cells using short-hairpin (sh)RNA and CRISPR/Cas9 editing, respectively, significantly suppressed tumor formation and growth in xenograft mouse models167,168. We169, and other groups167,168, have also found that the in vitro silencing of STK25 reduced proliferation, migration, and invasion of human hepatocarcinoma cells, which was accompanied by a lower expression of the markers of epithelial-mesenchymal transition (EMT) and augmented autophagic flux. In Paper IV, we further discovered that the inhibitory effect of STK25 knockdown in proliferation, migration, and invasion was comparable to that achieved by anti-HCC drugs sorafenib or regorafenib. Reciprocally, the exacerbated proliferation and invasiveness were observed in STK25-overexpressing HCC cells168. We have additionally shown that the genetic depletion of STK25 in mice alleviated liver tumor growth by protecting against hepatic steatotoxicity and inactivating STAT3, ERK1/2, and p38 signaling pathways169 – the key pathways also implicated in human HCC170-172. In Paper IV, we further demonstrated that pharmacologic suppression of STK25 with ASOs, either globally across all peripheral tissues or specifically in hepatocytes, efficiently mitigated the development and exacerbation of hepatocarcinogenesis in a mouse model of MASH- driven HCC when ASO treatment was initiated at the different stages of disease trajectory. Extrahepatic role of STK25 Many reports have revealed that STK25 is not only involved in hepatic lipid partitioning and liver tumorigenesis but also possesses extrahepatic functions. For instance, our previous studies have found that high-fat-fed Stk25–/– mice accumulated less ectopic fat in skeletal muscle, kidney, and vascular wall, and had healthier adipose tissue, compared with wild-type littermates148,173-175. The role of STK25 in extrahepatic tumors is controversial. On the one hand, STK25 was highly expressed in prostate cancer tissues and has been implicated in prostate carcinogenesis176. Under the condition of hypoxia, STK25 also accelerated the invasiveness of human melanoma cells through the activation of matrix metalloproteinase 2 (MMP2)177. On the other hand, STK25-induced inhibition of aerobic glycolysis via GOLPH3-mTOR pathway suppressed cell proliferation of colorectal cancer cells178. STK25 is also involved in the nervous system development. Specifically, STK25 indirectly promoted the phosphorylation of the brain microtubule-binding protein Tau, hyperphosphorylation of which is associated with several neurodegenerative disorders179. STK25 also can phosphorylate cerebral cavernous malformation 2 (CCM2) and initiate death signaling in medulloblastoma cells180. 19 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis 20 Ying Xia 2 AIM STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis 2 Ying Xia 2. AIM The overall objective of this thesis is to examine the impact of lipid droplet-binding STE20-type kinases TAOK1, TAOK3, and STK25 in the regulation of liver lipid metabolism and hepatocarcinogenesis. The specific aims of the three papers included in this thesis are: Paper I: to investigate the role of TAOK3 protein in the control of hepatocellular lipotoxicity using in vitro investigations in cultured human hepatocytes Paper II: to determine the role of TAOK1 in the regulation of hepatocellular lipid partitioning using in vitro studies in cultured human hepatocytes Paper III: to elucidate the in vivo role of TAOK3, which we have previously reported to regulate hepatocellular lipotoxicity in vitro, in the development of diet-induced MASLD and systemic insulin resistance in mice Paper IV: to examine the role of STK25, a critical molecular driver of the hepatocellular lipotoxic milieu and MASH susceptibility, in the initiation and progression of MASH-related HCC 21 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis 22 Ying Xia 3 METHODS STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis 16 Ying Xia 3. METHODS 3.1 Ethical Statements The studies involving human participants were approved by the Ethics Committee of the University of Leipzig, Germany (363-10-13,122,010 and 159-12-21,052,012) or by the Ethics Committee of the University of Tübingen, Germany (368/2012BO2), and carried out in accordance with the Declaration of Helsinki. All patients provided written informed consent before enrolling in the studies. The mice used in this study received humane care in accordance with the National Institutes of Health (NIH; Bethesda, MD) recommendations outlined in the Guide for the Care and Use of Laboratory Animals. All the in vivo experiments were conducted in compliance with the guidelines approved by the local Ethics Committee for Animal Studies at the Administrative Court of Appeals in Gothenburg, Sweden (approval number 5.8.18-20238/2020 for xenograft model and 5.8.18-17285/2018 for MASLD and MASH-related HCC models). 3.2 Human Subjects 3.2.1 Patient cohorts In Papers I-III, the mRNA expression levels of TAOK1, TAOK2, and TAOK3 were quantified in interoperative liver biopsies collected from Caucasian individuals (men, n=35; women, n=27) who underwent laparoscopic abdominal surgery for Roux-en-Y bypass (n=12), sleeve gastrectomy (n=9), or elective cholecystectomy (n=41) at Leipzig University Hospital (Leipzig, Germany). The participants fulfilled the following inclusion criteria: (1) men and women, age >18 years; (2) indication for elective laparoscopic or open abdominal surgery; (3) BMI between 18 and 50 kg/m2; (4) abdominal MRI feasible; and (5) signed written informed consent. The exclusion criteria for liver biopsy donors were: (1) significant acute or chronic inflammatory disease or clinical signs of infection; (2) C-reactive protein (CrP) >10 mg/dl; (3) type 1 diabetes and/or antibodies against glutamic acid decarboxylase (GAD) and islet cell antibodies (ICA); (4) systolic blood pressure >140 mmHg and diastolic blood pressure >95 mmHg; (5) clinical evidence of cardiovascular or peripheral artery disease; (6) thyroid dysfunction; (7) alcohol or drug abuse; and (8) pregnancy. Total body fat was measured by dual X-ray absorptiometry (DEXA) and liver fat was measured by 1H-MRS181. A small liver biopsy was obtained during the surgery (between 08:00 and 10:00 hours after an overnight fast), immediately snap-frozen in liquid nitrogen, and stored at –80°C. Histological features were blindly evaluated by two specialized hematopathologists in hematoxylin and eosin (H&E)- and Oil Red O-stained liver sections using the well-validated MAS and fibrosis staging score182. In Paper IV, STK25 mRNA expression was measured in interoperative liver biopsies collected from 262 Caucasian individuals (men, n=157; women, n=105) undergoing surgery at the Department of General, Visceral, and Transplant Surgery at the University Hospital of Tübingen (Tübingen, Germany). Protein analysis was carried out in a subgroup of the entire study cohort consisting of 133 subjects (men, n=87; women, n=46) for whom there was sufficient tissue available. After food 23 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis withdrawal overnight, liver biopsies were taken from normal, non-diseased tissue determined by the pathologist during surgery, immediately snap frozen in liquid nitrogen, and stored at –80°C. The participants tested negative for viral hepatitis and had no liver cirrhosis. 3.2.2 Data collection from public databases In Paper IV, three HCC datasets (GSE25097, GSE14520, and GSE36376) were downloaded from GEO database. Additionally, we accessed the whole transcriptome sequencing (RNA-seq) data from the HCC Project in the Cancer Genome Atlas (TCGA)183 and the Genotype-Tissue Expression Portal (GTEx)184. The GEO, TCGA, and GTEx databases are publicly accessible resources and written informed consent was obtained from the patients prior to data collection. 3.3 Cell Culture 3.3.1 Cells Immortalized human hepatocytes (IHHs; a gift from B. Staels, the Pasteur Institute of Lille, University of Lille Nord de France, Lille, France) were originally obtained from healthy liver tissue removed surgically from a 59-year-old man and immortalized by stable transfection with SV40 large T antigen- expressing plasmid185. HepG2 cells (LGC Standards, Teddington, UK) were derived from the liver tissue of a 15-year-old male Caucasian with a well-differentiated human hepatoblastoma and retained characteristics of quiescent hepatocytes186. HepG2-NTCP cells (a kind gift from S. Urban, Department of Infectious Diseases, University Hospital Heidelberg, Heidelberg, Germany) derive from HepG2 cells and express the human sodium taurocholate co-transporting polypeptide (NTCP)187. Huh7 (JCRB Cell Bank, Tokyo, Japan) is a well-differentiated hepatocyte-derived carcinoma cell line that was originally taken from a liver tumor in a 57-year-old Japanese male188. Hep3B (ATCC, Manassas, VA) is a well- differentiated hepatocyte-derived carcinoma cell line derived from liver tissue obtained from an 8-year- old black patient with liver cancer188. SNU-475 (ATCC) is a poorly differentiated hepatocyte-derived carcinoma cell line taken from a Korean patient prior to cytotoxic therapy189. LX-2 (Millipore, Burlington, MA) was originally established from primary hepatic stellate cells (HSCs) that were transfected with the large SV40 T-antigen in passage 4190. THP-1 (ATCC) designates a spontaneously immortalized monocyte-like cell line, derived from the peripheral blood of a childhood case of acute monocytic leukemia191. For comprehensive details on the culture conditions of the aforementioned cell lines, refer to the Materials and Methods sections of Papers I-IV. Primary mouse hepatocytes were isolated from male Taok3–/– and wild-type mice applying a collagenase perfusion method192 and maintained in William’s E medium (Invitrogen, Carlsbad, CA) supplemented with 0.28 mol/l sodium ascorbate (Sigma-Aldrich, St. Louis, MO), 0.1 mmol/l sodium selenite (Sigma-Aldrich), 100 mg/ml penicillin and 100 U/ml streptomycin (Gibco, Paisley, UK), 3 g/l glucose (Sigma-Aldrich), and 26 U/l human recombinant insulin (Actrapid Penfill; Novo Nordisk, 24 Ying Xia Bagsværd, Denmark). Cryopreserved primary human hepatocytes (M00995-P) were purchased from BioIVT, Westbury, NY, and were cultured following the manufacturer’s protocol. 3.3.2 Transient transfections and incubation with anti-HCC drugs For RNA interference, primary mouse hepatocytes were transfected with mouse Taok3 small interfering (si)RNA (s232238; Invitrogen), mouse Taok2 siRNA (M-059829-01; Dharmacon, Lafayette, CO), or non-targeting control (NTC) siRNA (4390843; Invitrogen) using Lipofectamine RNAiMax (Thermo Fisher Scientific, Waltham, MA). Human hepatocytes and liver nonparenchymal cells were transfected with human TAOK3 siRNA (a pool of s27994, s27995, and s27996; Invitrogen), human TAOK1 siRNA (M-004846-03; Dharmacon, Lafayette, CO), human TAOK2 siRNA (s17865; Invitrogen), human STK25 siRNA (s20570; Ambion, Austin, TX), or NTC siRNA (SIC001; Sigma-Aldrich) using Lipofectamine RNAiMax. In Papers I and II, for overexpression, cultured human hepatocytes were transfected with human MYC- tagged TAOK3 expression plasmid (EX-A8822-M43; GeneCopoeia, Nivelles, Belgium), human MYC- tagged TAOK1 expression plasmid (EX-T7024-M43; GeneCopoeia), or an empty control plasmid (EX- NEG-M43; GeneCopoeia) using Lipofectamine 2000 (Thermo Fisher Scientific). In Paper IV, the cells were treated with 5 μmol/l sorafenib (Cayman Chemical), 5 μmol/l regorafenib (Selleck Chemicals, Houston, TX), or vehicle control (DMSO; Invitrogen) for 24 h prior to harvest. The dosages of drugs were selected to be equivalent to the steady‐state plasma concentrations of clinically effective doses. In all in vitro experiments in Papers I-IV, cells were incubated with or without 25-100 µmol/l oleic acid (Sigma-Aldrich) for 48 h before harvest. 3.3.3 Assessment of lipid metabolism In Papers I-III, to visualize the neutral lipids, cells were fixed with phosphate-buffered formaldehyde (Histolab Products, Gothenburg, Sweden), and stained with Bodipy 493/503 (Invitrogen; the dye passing through the cell membrane into the cell and localizing the polar lipids in the cell to specifically stain the lipid droplets) or Oil Red O (Sigma-Aldrich; the azo dye based on the use of a lysochrome for the visualization of neutral TAGs and lipids). Fatty acid uptake was quantified using the Quencher-Based Technology (QBT) Fatty Acid Uptake Assay Kit (Molecular Devices, San Jose, CA) according to the manufacturer’s recommendations. Briefly, cells were cultured in a 96-well black-wall/clear-bottom plate and were serum-deprived for 1 h, followed by the addition of QBT fatty acid-loading buffer to each well. Kinetic readings were started immediately with the SpectraMax iD3 multi-mode microplate reader, using bottom read settings at 490/525 nm excitation/emission. To assess lipogenesis, incorporation of [3H] glucose and [3H] oleic acid (both from PerkinElmer Life and Analytical Sciences, Waltham, MA) into TAG was measured. In brief, cells were incubated with 25 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis medium containing [3H] glucose or [3H] oleic acid, and then collected for lipid extraction, followed by lipid separation by thin-layer chromatography on silica gel plates. Radiolabeled TAGs were detected by iodine vapor and quantified by a scintillation counter (LS6500 Multipurpose Scintillation Counter; Beckman Coulter, Providence, RI). To analyze lipolysis rate, the TAG hydrolase activity was determined in total cell lysates using [3H]- triolein as the substrate. The total amount of triolein hydrolyzed was adjusted to the protein concentration of the cellular homogenate to yield specific lipase activity. In addition, the protein abundance of ATGL, the first lipase in the canonical lipolysis pathway, was measured by Western blot as described below. To measure the secretion of TAG into the medium, cells were cultured in 12-well plates and first incubated with pulse medium [Complete William’s E medium containing 0.5 μCi/ml [3H] oleic acid, 360 μmol/l oleic acid and 1% (vol/vol) fatty acid-free BSA] for 8 h, followed by incubation with chase medium [Complete William’s E medium supplemented with 30% (vol/vol) fatty acid-free BSA] for an additional 8 h. The chase medium was collected for lipid extraction, prior to lipid separation by thin- layer chromatography on silica gel plates. Radiolabeled TAGs were detected by iodine vapor and quantified by a liquid scintillation counter. β-oxidation is the process by which fatty acids are broken down in the mitochondria to generate acetyl- CoA, which will be fed into the tricarboxylic acid cycle to generate the high-energy molecule ATP, CO2, and H2O. To determine the rate of β-oxidation, cells were incubated in the presence of [9,10- 3H(N)]-palmitic acid as substrate, and [3H]-labeled water was quantified as the product of free fatty acid oxidation by a scintillation counter144. 3.3.4 Measurement of carbohydrate metabolism Glycogen content was determined by a coupled enzyme assay using the Glycogen Assay Kit (Sigma- Aldrich) which uses a single working reagent that combines the enzymatic breakdown of glycogen and the detection of glucose in one step. The absorbance was measured at 570 nm using the SpectraMax iD3 multi-mode microplate reader. To measure hepatocellular glycogenolysis, cells were incubated in DMEM (Gibco) without glucose, glutamine, or pyruvate but containing 0.24 mmol/l 3-isobutyl-1-methylxanthine (IBMX; a cAMP pathway activator) for 30 min. The released glucose content in the medium was then measured using the Amplex Red Glucose/Glucose Oxidase Assay Kit (Invitrogen). In this assay, glucose oxidase reacts with D-glucose to form d-gluconolactone and H2O2. In the presence of horseradish peroxidase (HRP), the H2O2 then reacts with the Amplex Red reagent (10-acetyl-3,7-dihydroxyphenoxazine) in a 1:1 stoichiometry to generate the red-fluorescent oxidation product resorufin, which has fluorescence excitation and emission maxima of approximately 571 nm and 585 nm, respectively. To determine the rate of gluconeogenesis, cells were cultured in glucose production assay medium (glucose- and phenol red-free DMEM containing 1 mmol/l sodium pyruvate, 20 mmol/l sodium lactate, 26 Ying Xia and 15 mmol/l HEPES) for 4 h; the released glucose content in the medium was then measured as described above. To evaluate glycolysis, cells were transfected in the Seahorse XF Cell Culture Microplate (Agilent Technologies, Santa Clara, CA) as described above. The next day, culture medium was changed to the XF Base Medium (Agilent Technologies) supplemented with 10 mmol/l glucose, 1 mmol/l sodium pyruvate, and 2 mmol/l glutamine for 1 h. Glycolytic flux was measured before and after injecting a mixture of rotenone (0.5 μmol/l; a complex I inhibitor) and antimycin A (0.5 μmol/l; a complex III inhibitor) followed by an injection of 2-deoxy-D-glucose (0.5 μmol/l), using the Seahorse XFe96 Extracellular Flux Analyzer (Agilent Technologies). Basal glycolysis and compensatory glycolysis were calculated using the Seahorse Glycolysis Rate Assay Report Generator (Agilent Technologies). To measure glucose uptake, cells were preincubated in glucose-free DMEM (Gibco) for 2 h. After 1.5 h, insulin (100 nmol/l; Actrapid Penfill; Novo Nordisk) was added to the wells to stimulate glucose uptake. Cells were washed with PBS before adding D-[2-3H] glucose and 2-deoxy-D-glucose. Glucose transport was stopped after 10 min by adding phloretin. Cells were lysed in 0.2 mol/l NaOH, and radioactivity was quantified in a scintillation counter. 3.3.5 Evaluation of tumorigenicity Cell viability was analyzed using the CellTiter-Blue Cell Viability Assay (Promega, Stockholm, Sweden) according to the manufacturer’s protocol. The assay is based on the ability of living cells to convert a redox dye (resazurin) into a fluorescent end product (resorufin), whereas nonviable cells rapidly lose metabolic capacity and thus do not generate a fluorescent signal. The proliferation was measured by the Click-iT EdU Proliferation Assay for Microplates Kit (Thermo Fisher Scientific). In this assay, the nucleoside analog EdU (5-ethynyl-2’-deoxyuridine) is added to live cells and incorporated into DNA during active DNA synthesis. The incorporated EdU contains an alkyne group which is covalently joined to the azide group present in HRP using a highly specific copper-catalyzed covalent reaction. Amplex UltraRed Reagent is then added and its conversion by HRP into a highly fluorescent product is recorded using a fluorescence microplate reader at 568/585 nm excitation/emission. The Apoptosis/Necrosis Detection Kit (Abcam, Cambridge, UK) was applied to monitor the initial/intermediate stages of apoptosis by staining with Apopxin Green for phosphatidylserine. The activation of caspase 3 and caspase 7 was determined using the Caspase-Glo 3/7 Assay Kit (Promega) following the manufacturer’s protocol. For migration assay, cells were seeded in 200 μl of serum-free medium into the upper chambers of a transwell and 600 μl of the respective medium with 10% FBS was added in the bottom chambers of the transwells for a chemotactic gradient. After incubation at 37°C, the cells on the top side of the membranes were removed with cotton swabs, and cells that had migrated to the bottom side were fixed and stained with 0.1% crystal violet. The inserts were cleaned in tap water and left to dry before imaging. To assess invasion, the transwells were coated with 100 μl of Matrigel before the experiment. 27 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis Images were acquired using a Zeiss Axio Observer microscope with the ZEN Blue software. The labeled area was quantified by using the ImageJ software. To assess EMT, cells were processed for immunofluorescence with anti-N-cadherin (a mesenchymal marker) or anti-E-cadherin (an epithelial marker) antibodies as described below. 3.4 Animal Experiments 3.4.1 Mouse models MASLD model. In Paper III, Taok3 knockout mice (on the C57BL/6J background) were purchased from the Jackson Laboratory (Bar Harbor, ME). Male knockout mice and their wild-type littermates were weaned at 3 weeks of age and housed 3 to 5 per cage in a temperature-controlled (21°C) facility with a 12-h light/dark cycle and ad libitum access to chow and water. From the age of 6 weeks, the mice were fed a pelleted high-fat diet (45 kcal% fat; D12451, Research Diets, New Brunswick, NJ); the body weights were recorded and blood was collected for measurement of glucose and insulin at different time points, 24-h urine was obtained from custom-made Perspex restraint cages at the age 20 weeks, and various in vivo tests were carried out as described below. At the age of 24 weeks, mice were killed by cervical dislocation under isoflurane anesthesia after 4 h of fasting. Blood was collected by heart puncture. Liver, epididymal white adipose tissue (eWAT), and subcutaneous white adipose tissue (sWAT) were weighed. Liver, eWAT, brown adipose tissue (BAT), kidney, and gastrocnemius skeletal muscle were collected for histological and immunofluorescence microscopy analysis and/or snap frozen in liquid nitrogen and stored at –80°C for analysis of protein and gene expression and biochemical assays as described below. Xenograft model. In Paper IV, a total of 1×107 cells (STK25-depleted or wild-type HepG2 generated by CRISPR/Cas9 editing) were suspended in 100 µl of PBS and injected subcutaneously into the upper flanks of 5-week-old BALB/c nude mice (Janvier Labs, Le Genest-Saint-Isle, France). Viability of cells was checked with Trypan Blue Stain 0.4% (Invitrogen) and differed neither between the two groups, nor before and after injection. Following inoculation, mice were fed a high-fat diet (60 kcal% fat, D12492; Research Diets). Tumor dimensions were inspected every 3-5 days using calipers and tumor volume was calculated by the formula V=(length × width2)/2. Once xenografts reached a size of ~12- 14 mm in the maximal diameter (on day 48 post-injection), the mice were killed and tumors were excised and photographed. MASH-related HCC model. In Paper IV, C57BL/6J mice (Charles River, Sulzfeld, Germany) were housed 3 to 5 per cage in a temperature-controlled (21°C) facility with a 12-h light/dark cycle and ad libitum access to chow and water. To induce HCC in the context of MASH, a single intraperitoneal injection of DEN (25 mg/kg; N0258; Sigma-Aldrich) was performed in 14-day-old mice. Starting from four weeks after DEN injection, mice were fed a high-fat diet (60 kcal% fat) for 30 weeks. At 6 weeks of age, weight-matched mice were randomly divided into three experimental cohorts which were injected intraperitoneally with Stk25 ASO (50 mg/kg/wk), control ASO (50 mg/kg/wk), N- 28 Ying Xia acetylgalactosamine (GalNAc)-conjugated Stk25 ASO (12.5 mg/kg/wk), GalNAc-control ASO (12.5 mg/kg/wk), or PBS twice weekly for the last 12, 21, or 30 weeks of high-fat diet feeding. The body weights were recorded weekly, blood was collected for determination of fasting glucose and insulin at several time points during the study, and glucose tolerance test (GTT) and insulin tolerance test (ITT) were carried out as described below. At the age of 36 weeks, mice were killed by cervical dislocation under isoflurane anesthesia after 4 h of food withdrawal. To investigate the cell proliferation, mice were subjected to a single intraperitoneal injection with BrdU (100 mg/kg; Sigma-Aldrich) 2 h before killing. Blood was obtained by cardiac puncture for assessment of alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels. Livers were weighed, and each lobe of the liver was photographed. The liver tumor parameters were evaluated as previously reported169. Liver tissues were processed for flow cytometry, harvested for histological and immunohistochemical/ immunofluorescence analysis, or snap frozen in liquid nitrogen and stored at –80°C for examination of protein and gene expression and biochemical assessments as described below. 3.4.2 In vivo tests Body composition and indirect calorimetry. In Paper III, body composition analysis (BCA) of total, lean, and fat body mass was carried out by time-domain nuclear magnetic resonance (TD-NMR) with the Minispec LF110 Analyzer (Bruker Corporation, Rheinstetten, Germany). Energy expenditure was assessed using an indirect calorimeter chamber (INCA; SOMEDIC, Hörby, Sweden) as previously described193. Basal daily food intake was determined as the average of duplicate readings taken over 2 consecutive days. Locomotor activity. In Paper III, activity was measured by the open-field test. Mice were placed into the center of a chamber (25 × 25 × 25 cm) to allow free exploration. Locomotor activity was recorded for 15 min during the dark phase of the day on 3 consecutive days and analyzed using the EthoVision XT software (v8.5; Noldus, Wageningen, Netherlands). Glucose and insulin tolerance. In Papers III and IV, after 4 h of morning fast194, mice received an intraperitoneal injection with glucose (1 g/kg; Sigma-Aldrich) or human recombinant insulin (2 U/kg; Actrapid Penfill) for GTT or ITT, respectively. Blood was taken from the tail tip at 0, 15, 30, 60, 90, and 120 min post-injection to determine glucose concentrations using an Accu-Chek glucometer (Roche Diagnostics, Basel, Switzerland). Tissue-specific glucose uptake. In Paper III, mice were injected with human recombinant insulin (0.5 U/kg; Actrapid Penfill) and [14C]-2-deoxy-D-glucose (50 μCi; PerkinElmer Life and Analytical Sciences) intravenously after withholding food for 4 h. Blood samples for the measurement of glucose and 14C content were obtained from the tail vein at 0, 3, 6, 10, 15, 20, 30, 40, and 60 min post-injection. After the last blood sampling, the mice were killed and different types of skeletal muscle (extensor digitorum longus, soleus, gastrocnemius, and quadriceps), liver, heart, brain, eWAT, sWAT, and BAT were dissected and weighed. Samples were then placed into 500 μl of 1 mol/l NaOH (Sigma-Aldrich) 29 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis and incubated for 1 h at 60°C to homogenize the tissue, prior to neutralization with 500 μl of 1 mol/l HCl (Sigma-Aldrich). 200 μl of homogenized sample was added to 1 ml of 6% perchloric acid (Sigma- Aldrich), followed by centrifugation at 13,000 × g for 2 min at 4°C. 800 μl of supernatant was collected and radioactivity was measured using a liquid scintillation counter. Tissue-specific glucose uptake was calculated by dividing the tissue 14C content with the integrated glucose-specific activity and normalized to the tissue weight195. 3.4.3 Biochemical assays In Papers III and IV, the fasting plasma insulin was assessed using the Ultrasensitive Mouse Insulin ELISA Kit (Crystal Chem, Downers Grove, IL). Glycogen content in the liver and gastrocnemius skeletal muscle tissues was measured using the Glycogen Assay Kit (Sigma-Aldrich) as described above. TAG content in the liver and kidney tissues was determined with the Triglyceride Colorimetric Assay Kit (Cayman Chemical, Ann Arbor, MI). The levels of reduced and oxidized glutathione were determined in the kidney lysates with the GSH-Glo Glutathione Assay Kit (Promega). Urinary albumin and creatinine concentrations were assessed by the Mouse Albumin ELISA Kit and the Creatinine Assay Kit (both from Abcam), respectively. Plasma ALT and AST concentrations were assessed with the Mouse Alanine Aminotransferase ELISA Kit and the Mouse Aspartate Aminotransferase ELISA Kit (both from MyBioSource, San Diego, CA), respectively. All biochemical assays were performed in duplicate. 3.5 Histological, Immunohistochemical, and Immunofluorescence Analysis Liver and eWAT tissues were fixed in 4% (vol/vol) phosphate-buffered formaldehyde (Histolab Products), embedded in paraffin, and sectioned. Paraffin sections were stained with H&E (Histolab Products) for morphological analysis (the hematoxylin stains cell nuclei a purplish blue while eosin stains the extracellular matrix and cytoplasm pink). The MAS was performed on H&E-stained liver sections following the Kleiner/Brunt criteria as previously described146. To examine the degree of fibrosis, liver paraffin sections were stained with Picrosirius Red (Histolab Products) and counterstained with Fast Green (Sigma-Aldrich) to stain both collagen type I and type III, or processed for immunofluorescence with anti-collagen IV or anti-fibronectin (an extracellular matrix protein) antibodies. To determine oxidative stress, paraffine sections and cultured hepatocytes were stained with dihydroethidium (DHE; Life Technologies, Grand Island, NY) to measure superoxide radical (O •−2 ) formation, or processed for immunofluorescence with anti-8-oxoguanine (8-oxoG), anti-4- hydroxynonenal (4-HNE), or anti-E06 antibodies to quantify oxidative DNA damage, lipid peroxidation products, and oxidized phospholipids, respectively. To measure ER stress, paraffine sections and cultured hepatocytes were processed for immunofluorescence with anti-KDEL (the four amino acid motif that functions as an ER retention signal that retrieves proteins from the Golgi back to the ER by COPI vesicles) and anti-CHOP (an indicator of ER stress-induced cell death) antibodies. To assess 30 Ying Xia peroxisomal activity, cultured hepatocytes were processed for immunofluorescence with anti- peroxisomal biogenesis factor 5 (PEX5; a peroxisome biogenesis marker) and anti-peroxisomal membrane protein 70 kDa (PMP70; a peroxisomal membrane marker) antibodies. Apoptotic cells were identified by using the HRP-DAB TUNEL Assay Kit (Abcam), which detects DNA fragmentation by labeling the 3’-hydroxyl termini during apoptosis. Liver and gastrocnemius muscle tissues were embedded in optimal cutting temperature (OCT) mounting medium (Histolab Products) and frozen in liquid nitrogen, followed by cryosectioning. Liver cryosections were stained with Bodipy 493/503 (Invitrogen) or Oil Red O (Sigma-Aldrich) as demonstrated above. Liver cryosections and cultured hepatocytes were also stained with MitoTracker Green (covalently binding mitochondrial matrix proteins) or MitoTracker Red (monitoring mitochondrial membrane potential) to quantify mitochondrial content or activity, respectively. Gastrocnemius muscle cryosections were stained with Nile Red (Sigma-Aldrich) for detection of polar membrane lipids or subjected to enzymatic activity assays as previously described196. For detailed information about histology and antibodies used for immunohistochemistry/ immunofluorescence, see corresponding Material and Methods sections and supplementary tables of the Papers I-IV. All the images were acquired using a Zeiss Axio Observer microscope with the ZEN Blue software and quantified using the ImageJ software (1.47v; National Institutes of Health, Bethesda, MD). 3.6 Reverse Transcription Quantitative PCR (RT-qPCR) RNA was isolated from tissue samples and cultured hepatocytes with the EZNA Total RNA Kit (Omega Bio-Tek, Norcross, GA) or the RNeasy Lipid Tissue Mini Kit (used for the eWAT and BAT from mice as well as for human liver biopsies; Qiagen, Hilden, Germany). cDNA was synthesized using the High- Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific). Relative quantification was performed with the CFX Connect Real-Time System (Bio-Rad, Hercules, CA) or the QuantStudio 6 Flex Real-Time PCR System (Thermo Fisher Scientific). The relative quantities of the target transcripts were calculated from duplicate samples after normalization of the data to the endogenous control, 18S rRNA (used for tissue samples, cultured human hepatocytes, and human liver tissues in Papers I-III), RSP13 (used for human liver tissues in Paper IV), or β-actin (used for mouse liver tissues in Paper IV). TaqMan probes and assays for RT-qPCR technique have been used and analyzed according to the Minimum Information for Publication of Quantitative Real-Time PCR Experiments guidelines197 in all papers of this thesis work. 3.7 Western Blot Cells or homogenized tissue samples were lysed in lysis buffer optimized to maintain protein stability, prevent degradation, and inhibit phosphatase activity. Protein concentration was measured in duplicates using the Bicinchoninic Acid Protein Assay Kit (Thermo Fisher Scientific). Proteins were separated by 31 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis size using gel electrophoresis and subsequently transferred to a membrane. To ensure that the transfer of proteins to the membrane was complete, the membrane was incubated with 0.5% (wt./vol.) Ponceau S solution. Target proteins were detected using specific antibodies, with endogenous controls, such as actin, glyceraldehyde 3-phosphate dehydrogenase (GAPDH), or vinculin, included in most analyses. For detailed information about the antibodies and dilutions used to measure the protein abundance, see the corresponding supplementary tables of the Papers I-IV. 3.8 Statistical Analysis Sample distribution was assumed normal with equal variance. Unpaired 2-tailed Student’s t-test or one- way analysis of variance (ANOVA) followed by Tukey’s post hoc test were applied when comparing two or more groups, respectively. Differences were considered statistically significant at p<0.05. The Spearman’s correlation analysis was performed to determine the relationship between TAOK1, TAOK2, and TAOK3 expression in human liver biopsies and hepatic lipid content, MAS, and fibrosis score, as well as between the protein abundance of STK25 and phospho-STK25. All statistical analyses were carried out using SPSS statistics (24v; IBM Corporation, Armonk, NY). In Paper IV, the HCC samples extracted from the TCGA database were divided into high- and low- expressing groups based on the median expression value of STK25 in all samples. The survival data of HCC patients from the TCGA database was evaluated using the “survival” R package (statistical analysis of survival data) and “survminer” R package (visualization)198 for the prognostic analysis according to the Kaplan-Meier method. 32 Ying Xia 4 RESULTS AND DISCUSSION STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis 18 Ying Xia 4. RESULTS AND DISCUSSION In this section, the main results of Papers I, II, III, and IV are summarized and discussed. For further details, see the full Papers provided at the end of the thesis book. Paper I STE20-Type Kinase TAOK3 Regulates Hepatic Lipid Partitioning Our previous studies have identified TAOK3 as a lipid droplet-associated protein as shown by global proteomic analysis performed on the lipid droplet fraction from steatotic livers of high-fat-fed mice138,139. It is important to note that the proteins isolated by this approach do not necessarily represent bona fide lipid droplet proteins residing primarily or exclusively on the droplets, since the method fails to effectively deplete the membrane-bound cellular organelles that closely associate with lipid droplets, including the ER, mitochondria, peroxisomes, and endosomes199,200. To this end, we here employed immunofluorescence microscopy to confirm the close association of TAOK3 with lipid droplets, visualized by adipose differentiation-related protein (ADRP; also known as adipophilin or perilipin-2) staining, both in cultured human hepatocytes and mouse liver sections. The primary driver of MASLD is an imbalance in hepatic lipid metabolism, leading to the accumulation of intrahepatocellular fat that then fuels oxidative and ER stress, local inflammation, cell damage, and subsequent fibrogenesis in the liver201,202. In this study, we found that overexpression of TAOK3 in human hepatocytes enhanced DNL (i.e., input) and reduced -oxidation and TAG secretion (i.e., output), resulting in aggravated fat storage within lipid droplets, and the opposite effect was observed in TAOK3-deficient hepatocytes (Figure 13). Consistently, oxidative and ER stress was substantially exacerbated or suppressed in human hepatocytes where TAOK3 was overexpressed or knocked down, respectively (Figure 13). In line with these results, we observed that TAOK3 transcript levels in human liver biopsies were positively correlated with the key lesions of MASLD (i.e., histological scoring for hepatic steatosis, inflammation, and ballooning), liver fat content, and histological fibrosis score. Mechanistically, we demonstrated that the silencing of TAOK3 inhibited JNK signaling in human hepatocytes. This is interesting in light of recent reports revealing that decreased mitochondrial fat oxidation in liver steatosis and MASH is initiated by the activation of JNK, which can phosphorylate mitochondrial proteins203. Conversely, hepatocyte-specific suppression of the JNK pathway has been shown to enlarge mitochondria and elevate mitochondrial -oxidation, protecting the mice from liver steatosis204. Thus, reduced JNK phosphorylation may have contributed to a significant increase in mitochondrial activity observed in TAOK3-deficient hepatocytes. The present study does have some limitations. First, all the in vitro experiments in this report were performed using immortalized human hepatocyte cell lines, which may not be representative of in vivo conditions. To this end, further investigations using mouse models and human primary hepatocytes are performed in Paper III to further characterize the role of TAOK3 in liver lipid metabolism and MASLD 33 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis development. Second, we were unable to delineate the primary vs. the secondary changes in response to modifying the abundance of TAOK3 in hepatocytes. For example, alterations in oxidative and ER stress can be caused by increased or decreased lipid storage in cells where TAOK3 is overexpressed or silenced, respectively; however, TAOK3 may also have a direct impact on metabolic stress response. Together, this investigation emphasizes the importance of lipid droplet-binding proteins in general, and TAOK3 in specific, in the control of intrahepatocellular fat storage and warrants further studies to explore whether the inhibition of TAOK3 signaling can prevent MASLD development and/or progression by antagonizing hepatic steatosis in vivo. Figure 13. A working model of the function of TAOK3 in regulating liver lipotoxicity. Overexpression of TAOK3 in hepatocytes stimulates lipid droplet anabolism through increased TAG synthesis and inhibits lipid droplet catabolism through suppressed β-oxidation and VLDL-TAG secretion, augmenting oxidative and ER stress. Although not investigated in our study, these alterations in TAOK3-overexpressing cells may subsequently lead to hepatic inflammation and fibrogenesis. The opposite changes in intrahepatocellular lipid storage and oxidative/ER stress are seen when TAOK3 is knocked down. 34 Ying Xia Paper II Silencing of STE20-Type Kinase TAOK1 Confers Protection against Hepatocellular Lipotoxicity via Metabolic Rewiring Similarly to TAOK3, the closely related STE20-type kinase TAOK1 has been identified as a component of hepatocellular lipid droplet proteome138,139, suggesting a potential role in regulating liver steatosis and MASLD development. In this study, we investigated the association between TAOK1 expression and MASLD severity in human liver biopsies and characterized its mechanism of action in cultured human hepatocytes. We observed that TAOK1 mRNA expression in human liver biopsies was positively correlated with the key hallmarks of MASLD (i.e., histological scoring for hepatic steatosis, inflammation, and ballooning). Importantly, we showed that the silencing of TAOK1, similarly to TAOK3 knockdown, reprogrammed cellular metabolism by stimulating lipid catabolism (mitochondrial β-oxidation and TAG efflux) and inhibiting lipid anabolism (fatty acid influx and lipogenesis), collectively lowering ectopic fat storage within intrahepatocellular lipid droplets (Figure 14), and the opposite changes were detected in TAOK1- overexpressing cells. Consistently, both the rate of canonical lipolysis and lipophagy, facilitating lipid mobilization from the lipid droplets for β-oxidation and secretion, were significantly enhanced in TAOK1-deficient hepatocytes (Figure 14). In parallel with the reduced fat accumulation, we observed markedly lower incidences of oxidative/ER stress in hepatocytes where TAOK1 was knocked down. This finding is interesting considering recent evidence demonstrating that oxidative/ER stress are key factors, that trigger MASLD progression from simple steatosis toward MASH as well as further aggravation to HCC7. Mechanistically, we found that the silencing of TAOK1 significantly diminished the abundance of ACC protein in human hepatocytes. This provides a plausible mechanism underlying the protection against ectopic fat storage observed in TAOK1-deficient hepatocytes since the enzymatic product of ACC, malonyl-CoA, is an intermediate of lipogenesis and also represses β-oxidation205. Furthermore, we detected elevated levels of ATGL protein in hepatocytes where TAOK1 was knocked down, which is expected to impact on the increased lipid utilization by enhancing canonical lipolysis rate206-208. In parallel, we discovered that the silencing of TAOK1 in hepatocytes suppressed phosphorylation of ERK and JNK, which in addition to regulating liver steatosis as described in Paper I, are critical signaling components stimulating proliferation, migration, and invasion of MASH-driven HCC170. Consistently, we observed lower proliferative, migratory, and invasive capacity as well as EMT in TAOK1-deficient human hepatoma cells. We found that, in addition to hepatocytes, TAOK1 was abundantly expressed in liver non-parenchymal cells, including HSCs and macrophages. This is noteworthy in light of evidence showing that macrophages and HSCs are susceptible to lipotoxic damage characterized by excessive fat storage and oxidative stress, and causally contribute to MASH initiation and progression by stimulating 35 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis inflammation and fibrogenesis209,210. However, in this study, we did not find any indication that the silencing of TAOK1 would reduce lipid storage or oxidative stress in liver non-parenchymal cells. The limitations described in Paper I of the current thesis are also applicable to this study as the setup is very similar. In conclusion, this study demonstrates for the first time that TAOK1 kinase modulates hepatocellular lipid homeostasis and, through control of the lipid channeling between anabolic and catabolic pathways, its deficiency breaks the vicious cycle of excessive lipid storage and oxidative/ER stress within hepatocytes, warranting further studies to address its potential role in MASLD. Figure 14. A working model of the function of TAOK1 in regulating hepatocellular lipotoxicity. The silencing of TAOK1 in hepatocytes inhibits lipid anabolism through suppressing fatty acid uptake and TAG synthesis, stimulates lipid catabolism through facilitating β-oxidation and VLDL-TAG secretion, and alleviates oxidative and ER stress. Mechanistically, the rate of canonical lipolysis and lipophagy, which both enhance lipid mobilization from the lipid droplets for fatty acid oxidation and TAG secretion, are stimulated by TAOK1 knockdown. Furthermore, the silencing of TAOK1 decreases ACC protein abundance, which is expected to both reduce lipogenesis and augment β-oxidation, as well as increases canonical ATGL lipase levels. FFA, free fatty acid; LD, lipid droplet. 36 Ying Xia Paper III Knockout of STE20-Type Kinase TAOK3 Does not Attenuate Diet-Induced NAFLD Development in Mice On the basis of our previous research, which reveals the importance of TAOK3 in the control of hepatocellular lipid partitioning in vitro (Paper I), we here employed the genetic model of high-fat diet- fed Taok3 whole-body knockout mice to decipher the possible in vivo role of this protein in metabolically triggered MASLD development and the regulation of systemic glucose and insulin homeostasis. We found that TAOK3 deficiency had no effect on body weight or composition, food intake, or locomotor activity of obese mice. Consistently, Taok3–/– mice and wild-type controls exhibited a comparable degree of high-fat diet-induced liver steatosis, inflammation, and fibrosis. No difference was observed in lipotoxic damage of adipose tissue, kidney, or skeletal muscle when comparing the two genotypes, and the systemic glucose and insulin homeostasis was also unaffected by depletion of TAOK3 (Figure 15). The observed discrepancies between the results in Taok3–/– mice in this study and our previous experiments with TAOK3-deficient human hepatocytes (as detailed in Paper I) do not seem to be attributable to species-specific differences since we found reduced lipid content and lower oxidative damage even in cultured mouse hepatocytes where TAOK3 was silenced by siRNA. Consistently, TAOK3 is highly conserved in mice and humans with an amino-acid sequence identity of 96%. An alternative explanation for the lack of phenotypic changes in Taok3 knockout mice may involve compensatory mechanisms by other STE20-type kinases in response to the genetic ablation of TAOK3. Numerous studies have documented inconsistencies between the phenotypic outcomes of genetic knockouts and gene knockdowns, suggesting that such disparities may arise from a genetic compensation mechanism that involves the transcriptional up-regulation of related genes, which can substitute for the function of the mutated gene in the former but not the latter211-213. In line with these previous reports, we found a significant upregulation in the mRNA expression of several STE20-type kinases including TAOK1, MST3, MST4, STK25, and MAP4K4, which have been identified as critical mediators of hepatocellular lipotoxic milieu125,138,142-145,214, in the livers of Taok3–/– vs. wild-type mice but not in mouse or human hepatocytes where TAOK3 was knocked down in vitro. This suggests that the absence of TAOK3 throughout development in Taok3 knockout mice may trigger compensatory mechanisms to mitigate the gene loss, whereas in hepatocytes treated with Taok3 siRNAs, the gene function is likely suppressed before such compensatory networks can be established. Remarkably, while the abundance of all the TAOK3-related STE20-type kinases analyzed in this study was increased in the livers of Taok3 knockout mice, no alterations were found in the transcript levels of any of these proteins in extrahepatic tissues (adipose tissues, kidney, or skeletal muscle) when comparing Taok3–/– mice and wild-type controls. Thus, transcriptional compensation appears to be organ-specific, potentially reflecting the varying functional significance of individual STE20 kinases across different tissues. 37 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis There are some limitations associated with this study. First, this investigation was confined to high-fat- fed mice, and the impact of TAOK3 was not examined in other contexts, such as HCC models or under different dietary challenges. This restriction limits the generalizability of our findings to other pathological conditions and metabolic stressors. Second, this research focused exclusively on metabolic organs, potentially overlooking the effects of TAOK3 deficiency in other tissues. Future studies by incorporating diverse dietary models and examining a broader range of organs are warranted to fully elucidate the physiological and pathological functions of TAOK3 in vivo. Together, this report demonstrates that genetic deficiency of TAOK3 in mice fails to mitigate the development of diet-induced MASLD and has no impact on the progression of systemic glucose intolerance or insulin resistance in the context of obesity. The lack of hepatic phenotype in Taok3 knockout mice in vivo, in contrast to the protective effect against hepatocellular lipotoxicity observed by gene knockdown in vitro, may be attributable to the liver-specific compensation response for the genetic loss of TAOK3 by related STE20-type kinases. Figure 15. Schematic illustration of metabolic responses in individual organs as well as at the whole-body level in Taok3–/– mice and wild-type littermates. OXPHOS, oxidative phosphorylation. 38 Ying Xia Paper IV Therapeutic Potential of STE20-Type Kinase STK25 Inhibition for the Prevention and Treatment of Metabolically Induced Hepatocellular Carcinoma MASH-associated HCC remains a significant clinical challenge due to the aggressive nature of the disease and its poorly understood molecular basis. Our recent studies have identified the liver lipid droplet-decorating protein STK25 as a critical driver of the hepatocellular lipotoxic milieu and MASH susceptibility138,144-148. In light of these observations, we now hypothesize that STK25 antagonists could represent an effective therapeutic strategy for the prevention and/or treatment of HCC in the context of MASH. In this study, we observed the protein levels of STK25, as well as the abundance of its active form phospho-STK25 (Thr174), were increased in human MASH-driven HCC compared with healthy liver tissue. In addition, STK25 knockout in human hepatoma cells blocked tumor formation and growth in a xenograft mouse model. Our previous investigations have shown that the whole-body genetic ablation of STK25 protects mice from hepatocarcinogenesis in the context of MASH169. However, STK25 is ubiquitously expressed, and using a model of global Stk25−/− mice presents an intrinsic limitation of STK25 being depleted throughout the development and in all body locations. To this end, we here assessed the in vivo effect of two classes of Stk25-targeting ASOs: GalNAc-conjugated Stk25 ASO (selectively delivered to hepatocytes) and the parent unconjugated Stk25 ASO (broadly distributed to peripheral organs without penetrating the blood-brain barrier). Both ASO types were administered in a mouse model, where MASH-related HCC was triggered by combining an injection with procarcinogen DEN and a high-fat diet feeding, at different phases of the disease trajectory. We found that the treatment with GalNAc- Stk25 ASO achieved similar effectiveness in reducing HCC tumor formation and growth compared to 4-fold higher doses of Stk25 ASO, suggesting that the selective inhibition of STK25 in hepatocytes is sufficient to mitigate metabolically triggered HCC in mice. In agreement with our earlier studies, we discovered that the in vitro silencing of STK25 in human hepatoma cells suppressed proliferation, migration, and invasion169, and the effect was similar to that achieved by incubating cells with sorafenib or regorafenib, which are widely used anti-HCC drugs in current clinical practice. Notably, anti-tumorigenic activity was significantly enhanced when sorafenib or regorafenib treatment was combined with STK25 knockdown. These results suggest that STK25 and sorafenib/regorafenib likely operate in different signaling pathways, and further investigations are warranted to validate the potential benefits of STK25 inhibitors as an adjunctive therapy to sorafenib and regorafenib, as well as in HCC tumors which are resistant to the current first-line standard of care. Intriguingly, we detected marked changes in mitochondrial ultrastructure in STK25-depleted human hepatoma cells, including about a 1.5-fold increase in mitochondrial area and perimeter as well as a nonuniform arrangement of cristae, suggesting a shifted equilibrium towards fusion and a higher bioenergetic efficiency of mitochondria215,216. Consistently, we found elevated protein levels of 39 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis mitochondrial fusion markers, higher mitochondrial membrane potential, and enhanced mitochondrial β-oxidation in STK25-deficient hepatocytes144,146. In parallel with alterations in mitochondrial dynamics, we discovered a suppression of peroxisomal biogenesis in STK25-deficient hepatoma cells. Together, these data indicate that regulating the homeostatic balance between mitochondrial and peroxisomal fatty acid oxidation likely contributed to the mitigation of hepatocarcinogenesis by STK25 inhibitors. Collectively, this study, which combines in vivo analyses in mouse models with expression profiling in human liver biopsies and in vitro assessments in cultured human hepatoma cells, demonstrates that antagonizing STK25 signaling mitigates the development and progression of MASH-driven HCC (Figure 16). Future clinical studies are required to test if these results will translate into therapeutic benefits in patients. Figure 16. Schematic illustration demonstrating the mechanism of ASO- mediated knockdown of Stk25 in DEN+HFD- induced HCC model. HFD, high-fat diet. 40 Ying Xia 5 FUTURE PERSPECTIVES STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis 26 Ying Xia 5. FUTURE PERSPECTIVES 5.1 Investigation of the Metabolic Effects of Genetic Ablation of TAOK1 in Mice In Paper II of the current thesis, we have provided the first evidence that TAOK1 is involved in the regulation of lipid metabolism in human cultured hepatocytes125. Based on these previous findings, we would now like to further investigate the in vivo role of TAOK1 in liver steatotoxicity and whole-body energy metabolism. In the frames of this future project, the hypothesis that depletion of TAOK1 in mice protects against metabolic dysfunction caused by chronic exposure to dietary lipids will be tested. Whole-body and liver-specific Taok1 knockout mice will be generated by crossbreeding the conditional Taok1fl/fl mice (have been ordered from Cyagen Biosciences, Santa Clara, CA) with mice expressing Cre recombinase under the Sox2 (Jackson Laboratory, Bar Harbor, ME) or Alb (Jackson Laboratory) promoters, respectively (all on the C57BL/6 J background). Depletion of exons 6 and 7 and the flanking introns will be verified using RT-qPCR and Western blot analysis. From the age of 6 weeks, the mice will be challenged with a pelleted high-fat diet (45 kcal% fat; D12451; Research Diets, New Brunswick, NJ) for 18 weeks. The body weights and food intake will be recorded weekly, and circulating glucose and insulin will be monitored monthly (see Figure 16 for a schematic overview of the experimental design). Locomotor activity will be measured by the open-field test. Energy expenditure will be assessed using an indirect calorimeter chamber. Urine samples will be collected to determine albumin, creatinine, and sodium levels. GTT and ITT will be carried out to evaluate glucose and insulin homeostasis. BCA of total, lean, and fat body mass will be conducted by time-domain nuclear magnetic resonance (TD-NMR) with the Minispec LF110 Analyzer. At the age of 24 weeks, mice will be killed by cervical dislocation under isoflurane anesthesia after 4 hours of fasting. Blood will be obtained by heart puncture for ALT and AST activity measurements. Liver, eWAT, and sWAT will be weighed. Liver, eWAT, sWAT, BAT, kidney, and gastrocnemius skeletal muscle will be collected for histological and immunofluorescence microscopy analysis and/or snap frozen in liquid nitrogen and stored at –80°C for analysis of protein and gene expression and biochemical assays, to characterize the metabolic profile of Taok1 knockout mice and wild-type littermates challenged with a high-fat diet. Figure 16. Schematic illustration of the experimental design. HFD, high-fat diet; IF, immunofluorescence; W, week 41 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis 5.2 Benefit of Liver-Specific Pharmacological Inhibitors of Human STK25 to Mitigate the Initiation and/or Exacerbation of MASH-Related HCC In Paper IV of the current thesis, we examined the impact of pharmacological STK25 antagonism in a mouse model using the ASOs targeting mouse Stk25. Despite the high conservation of STK25 between mice and humans (94% and 99% identity at the gene and protein levels, respectively), further research employing ASOs targeting the human STK25 sequence in mice, where the human STK25 gene has replaced the mouse counterpart, is necessary for better prediction of the efficacy in clinical settings. To develop human STK25-targeting ASOs, the screening for human STK25 ASOs could be performed to select the most potent ASO sequences, followed by the introduction of GalNAc conjugation to direct the ASOs to hepatocytes. Functional cellular assays can then be conducted to assess the ability of the GalNAc-STK25 ASOs in suppressing lipotoxic MASH milieu and tumorigenicity in human HCC cell lines. In these cells, intracellular fat accumulation should be quantified using fluorescent microscopy, while the reduction in tumorigenic potential would be evaluated by measuring cell proliferation through EdU incorporation and migration/invasion capacity via transwell assay. The in vivo efficacy of the selected GalNAc-STK25 ASOs could further be investigated in a genetically engineered humanized mouse model, where the human STK25 gene has replaced the mouse gene. To replicate the development of MASH-driven HCC in humanized STK25 knock-in mice, a single injection of chemical procarcinogen DEN can be combined with feeding a high-fat/high-cholesterol diet (Research Diets, New Brunswick, NJ). To determine treatment effects, mice can be intraperitoneally injected with selected GalNAc-STK25 ASOs or nontargeting control ASO twice per week, with the ASO administration being initiated in separate cohorts of mice at different timepoints of disease progression. Throughout the experiment, food intake, body weight, and plasma AFP, ALT, and AST levels should be monitored weekly. At the age of 9 months, mice would be killed to compare the HCC tumor number and size between the groups. To evaluate the role of STK25 in tumor proliferation, the mice should be BrdU pulsed 2 h prior to sacrifice; liver sections can then be immunostained with anti- BrdU antibody. In addition, immunofluorescence analysis with anti-PCNA and anti-Ki67 antibodies would be informative to identify proliferating cells in liver sections. Western blot could also be performed to examine the hepatic levels of AFP, YAP, GRP78, and EpCAM, whose abundance correlates with HCC grade and poor patient prognosis. This comprehensive study would provide valuable insights into the therapeutic potential of STK25 antagonisms in the context of MASH-driven HCC and may inform future clinical strategies. 42 Ying Xia 43 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis ACKNOWLEDGEMENT This thesis represents a milestone of a 4-year of work, experience, and opportunities. It has been a fantastic journey! Looking back to the past four years, I am overwhelmed with a mix of emotions and memories. It would not have been possible to complete this thesis without the support and encouragement of many wonderful individuals. I would like to take this moment to express my deepest gratitude to everyone who contributed to this. My main supervisor, Margit Mahlapuu, who was always available, late night and last minute, for guidance, encouragement, and inspiration. Your deep knowledge and catching enthusiasm were why I decided to embark on this journey in the first place. I admit I struggled in the beginning but now coming out with flying colors. Thank you for believing in me and cheering me on when my motivation was lacking. I really appreciated the helpful meetings, whenever I was stuck in my projects. This thesis simply could not have happened without you. My co-supervisor, Emmelie Cansby. Thanks for your constant support, enthusiasm for research, and willingness to teach me both fundamental principles and lab techniques, which have been inspiring, energizing, and kept me going forward. I felt safe knowing I could always rely on you for support and advice. When I first joined the lab, you left me your thesis with the words, “Looking forward to reading yours in a few years!” Well, here we are — time flies! Mara Caputo, my best partner, my sweetie, my queen of the lab. Thanks for being a “lab mom” to me, thanks for teaching me how to handle troubles not only in the lab but also in life, thanks for the happiness when we hang out and share gossip, thanks for your patience and constitutive encouragement, thanks for the days at EBM together which form the backbone of the results herein. Thank you for being a crucial part of my doctoral journey and, I hope, my life beyond. Emma Andersson, thanks for the always lovely smile, thanks for the accompanying and encouragement whenever I felt down, thanks for saying “be kind” when Mara tried to make fun of me, thanks for thoughtful comments for my papers and figures even though I didn’t adopt all of them, thanks for being a translator when I am confused with Swedish documents. Auld Lang Syne! Jingjing Zhang, thanks for sharing your experimental thoughts and ideas, I enjoyed collaborating with you. Bernice Asiedu, thanks for your kind support in ASO project, your beautiful smile is something that will stay in the back of my mind for good. Sofía Marcano, thanks for believing in me to be a good mentor even when I didn’t, and thanks for your hard work on ASO project. Sumit Kumar Anand, thanks for your guidance with injections and earmarking, and I really miss the joyful times when you were here. Thanks to all the past and present co-workers in Margit Mahlapuu Lab, for your friendship and for all the discussions that we had in the lab. 44 Ying Xia I would like to extend my sincere thanks to my friends who have supported me a lot. Lu, thanks for your kindness, your companionship during the quarantine in Beijing, and the memorable journeys we had in Prague, Budapest, and the Netherlands. Kexin, thanks for the delicious BBQ and hotpot, and thanks for the wonderful adventure in Tenerife. Carmen, thanks for your enthusiasm and I really wish I could have joined more parties with you, my fault! Anna, thanks for the interesting chats in the office, and thanks for caring and encouraging me. I would like to thank all the senior researchers — Marc, Julie, Markus, Peter, Anne, Åsa, Daniel, Anders, Kaisa, and Beidong — for a great scientific environment. Special thanks to my examiner, Per, for your caring, warmth, support, and all the helpful suggestions during my studies. I gratefully acknowledge the assistance from our administrative and technical staff, Valida, Peter, Lars, Bruno, Dan, Lotta, and Linda, thank you all for your help and support. All co-authors, for your excellent collaborations and all your hard work, I could not have written this thesis without you. I would like to thank everyone I have interacted with over the years who made my studies at Lundberg and Natrium more enjoyable. Yovana, Delaney, James, Uros, August, Jon, Shuwen, Annie, Monika, Sunniva, Silvana, Suélen, Nena, Io, Alfred, Emma, Jakub, Yuan, and Tomás. Thank you all for the happiness shared at parties, fika, and retreats. I would also like to thank all my remote friends and relatives who have given me the strength to handle all difficulties. Special thanks to my boyfriend, Wei Huang, for your unwavering encouragement and support. Best of luck with your own doctoral studies! 最后, 我要感谢我的父母. 你们是我的骄傲, 所 以我也希望成为你们的骄傲. 感谢你们多年来对我的鼓励和支持, 而我也终将结束在外的漂泊, 回到你们身边. I would like to end with the last sentence from my favourite book The Count of Monte Cristo, and I hope it can encourage everyone who is reading this thesis: “All human wisdom is contained in these two words, wait and hope.” 45 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis RERERENCES 1. Feng G, et al. Recompensation in cirrhosis: unravelling the evolving natural history of nonalcoholic fatty liver disease. Nat Rev Gastroenterol Hepatol, (2023). 2. Wong VW, Ekstedt M, Wong GL, Hagstrom H. Changing epidemiology, global trends and implications for outcomes of NAFLD. J Hepatol 79, 842-852 (2023). 3. Targher G, Corey KE, Byrne CD, Roden M. The complex link between NAFLD and type 2 diabetes mellitus - mechanisms and treatments. Nat Rev Gastroenterol Hepatol 18, 599-612 (2021). 4. Rinella ME, et al. A multi-society Delphi consensus statement on new fatty liver disease nomenclature. J Hepatol, (2023). 5. Powell EE, Wong VW, Rinella M. Non-alcoholic fatty liver disease. Lancet 397, 2212-2224 (2021). 6. Scorletti E, Carr RMM. A new perspective on NAFLD: Focusing on lipid droplets. Journal of Hepatology 76, 934-945 (2022). 7. Loomba R, Friedman SL, Shulman GI. Mechanisms and disease consequences of nonalcoholic fatty liver disease. Cell 184, 2537-2564 (2021). 8. Ioannou GN. Epidemiology and risk-stratification of NAFLD-associated HCC. J Hepatol 75, 1476-1484 (2021). 9. Vitale A, et al. Epidemiological trends and trajectories of MAFLD-associated hepatocellular carcinoma 2002-2033: the ITA.LI.CA database. Gut 72, 141-152 (2023). 10. Sheka AC, Adeyi O, Thompson J, Hameed B, Crawford PA, Ikramuddin S. Nonalcoholic Steatohepatitis: A Review. JAMA 323, 1175-1183 (2020). 11. Llovet JM, et al. Nonalcoholic steatohepatitis-related hepatocellular carcinoma: pathogenesis and treatment. Nat Rev Gastroenterol Hepatol 20, 487-503 (2023). 12. Singal AG, Kanwal F, Llovet JM. Global trends in hepatocellular carcinoma epidemiology: implications for screening, prevention and therapy. Nat Rev Clin Oncol 20, 864-884 (2023). 13. Younossi ZM, Koenig AB, Abdelatif D, Fazel Y, Henry L, Wymer M. Global Epidemiology of Nonalcoholic Fatty Liver Disease-Meta-Analytic Assessment of Prevalence, Incidence, and Outcomes. Hepatology 64, 73-84 (2016). 14. Younossi ZM, Golabi P, Paik JM, Henry A, Van Dongen C, Henry L. The global epidemiology of nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH): a systematic review. Hepatology 77, 1335-1347 (2023). 15. Romeo S, et al. Genetic variation in PNPLA3 confers susceptibility to nonalcoholic fatty liver disease. Nat Genet 40, 1461-1465 (2008). 16. Rich NE, et al. Racial and Ethnic Disparities in Nonalcoholic Fatty Liver Disease Prevalence, Severity, and Outcomes in the United States: A Systematic Review and Meta-analysis. Clinical Gastroenterology and Hepatology 16, 198-210 (2018). 17. Yip TC, et al. Geographical similarity and differences in the burden and genetic predisposition of NAFLD. Hepatology 77, 1404-1427 (2023). 18. Peverill W, Powell LW, Skoien R. Evolving Concepts in the Pathogenesis of NASH: Beyond Steatosis and Inflammation. International Journal of Molecular Sciences 15, 8591-8638 (2014). 46 Ying Xia 19. Buzzetti E, Pinzani M, Tsochatzis EA. The multiple-hit pathogenesis of non-alcoholic fatty liver disease (NAFLD). Metabolism 65, 1038-1048 (2016). 20. Driessen S, et al. Metabolic dysfunction-associated steatotic liver disease and the heart. Hepatology, (2023). 21. Bo T, et al. Hepatic selective insulin resistance at the intersection of insulin signaling and metabolic dysfunction-associated steatotic liver disease. Cell Metab 36, 947-968 (2024). 22. James DE, Stöckli J, Birnbaum MJ. The aetiology and molecular landscape of insulin resistance. Nature Reviews Molecular Cell Biology 22, 751-771 (2021). 23. Aron-Wisnewsky J, et al. Gut microbiota and human NAFLD: disentangling microbial signatures from metabolic disorders. Nat Rev Gastroenterol Hepatol 17, 279-297 (2020). 24. Do A, Kuszewski EJ, Langberg KA, Mehal WZ. Incorporating Weight Loss Medications Into Hepatology Practice for Nonalcoholic Steatohepatitis. Hepatology 70, 1443-1456 (2019). 25. Younossi ZM, Corey KE, Lim JK. AGA Clinical Practice Update on Lifestyle Modification Using Diet and Exercise to Achieve Weight Loss in the Management of Nonalcoholic Fatty Liver Disease: Expert Review. Gastroenterology 160, 912-918 (2021). 26. Vilar-Gomez E, et al. Weight Loss Through Lifestyle Modification Significantly Reduces Features of Nonalcoholic Steatohepatitis. Gastroenterology 149, 367-378 e365; quiz e314-365 (2015). 27. Cotter TG, Charlton M. Nonalcoholic Steatohepatitis After Liver Transplantation. Liver Transpl 26, 141-159 (2020). 28. Adam R, et al. 2018 Annual Report of the European Liver Transplant Registry (ELTR) - 50- year evolution of liver transplantation. Transpl Int 31, 1293-1317 (2018). 29. Lonardo A, Mantovani A, Petta S, Carraro A, Byrne CD, Targher G. Metabolic mechanisms for and treatment of NAFLD or NASH occurring after liver transplantation. Nat Rev Endocrinol 18, 638-650 (2022). 30. Harrison SA, et al. A Phase 3, Randomized, Controlled Trial of Resmetirom in NASH with Liver Fibrosis. New England Journal of Medicine 390, 497-509 (2024). 31. Ciardullo S, Muraca E, Vergani M, Invernizzi P, Perseghin G. Advancements in pharmacological treatment of NAFLD/MASLD: a focus on metabolic and liver-targeted interventions. Gastroenterol Rep (Oxf) 12, goae029 (2024). 32. Mashek DG. Hepatic lipid droplets: A balancing act between energy storage and metabolic dysfunction in NAFLD. Molecular Metabolism 50, (2020). 33. Mathiowetz AJ, Olzmann JA. Lipid droplets and cellular lipid flux. Nature Cell Biology 26, 331-345 (2024). 34. Huh JY, Saltiel AR. Roles of IkappaB kinases and TANK-binding kinase 1 in hepatic lipid metabolism and nonalcoholic fatty liver disease. Exp Mol Med 53, 1697-1705 (2021). 35. Gluchowski NL, Becuwe M, Walther TC, Farese RV. Lipid droplets and liver disease: from basic biology to clinical implications. Nature Reviews Gastroenterology & Hepatology 14, 343- 355 (2017). 36. Sinha RA, Singh BK, Yen PM. Direct effects of thyroid hormones on hepatic lipid metabolism. Nature Reviews Endocrinology 14, 259-269 (2018). 37. Bu SY, Mashek DG. Hepatic long-chain acyl-CoA synthetase 5 mediates fatty acid channeling between anabolic and catabolic pathways. Journal of Lipid Research 51, 3270-3280 (2010). 47 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis 38. Singh AB, Kan CFK, Kraemer FB, Sobel RA, Liu JW. Liver-specific knockdown of long-chain acyl-CoA synthetase 4 reveals its key role in VLDL-TG metabolism and phospholipid synthesis in mice fed a high-fat diet. American Journal of Physiology-Endocrinology and Metabolism 316, E880-E894 (2019). 39. Batchuluun B, Pinkosky SL, Steinberg GR. Lipogenesis inhibitors: therapeutic opportunities and challenges. Nature Reviews Drug Discovery 21, 283-305 (2022). 40. Coleman RA, Lewin TM, Muoio DM. Physiological and nutritional regulation of enzymes of triacylglycerol synthesis. Annual Review of Nutrition 20, 77-103 (2000). 41. Wurie HR, Buckett L, Zammit VA. Diacylglycerol acyltransferase 2 acts upstream of diacylglycerol acyltransferase 1 and utilizes nascent diglycerides and de novo synthesized fatty acids in HepG2 cells. FEBS J 279, 3033-3047 (2012). 42. Qi J, et al. The use of stable isotope-labeled glycerol and oleic acid to differentiate the hepatic functions of DGAT1 and -2. J Lipid Res 53, 1106-1116 (2012). 43. Villanueva CJ, et al. Specific role for acyl CoA:Diacylglycerol acyltransferase 1 (Dgat1) in hepatic steatosis due to exogenous fatty acids. Hepatology 50, 434-442 (2009). 44. Xu WY, et al. Differential Roles of Cell Death-inducing DNA Fragmentation Factor-α-like Effector (CIDE) Proteins in Promoting Lipid Droplet Fusion and Growth in Subpopulations of Hepatocytes. Journal of Biological Chemistry 291, 4282-4293 (2016). 45. Langhi C, Baldán A. CIDEC/FSP27 Is Regulated by Peroxisome Proliferator-Activated Receptor Alpha and Plays a Critical Role in Fasting- and Diet-Induced Hepatosteatosis. Hepatology 61, 1227-1238 (2015). 46. Wilfling F, et al. Triacylglycerol Synthesis Enzymes Mediate Lipid Droplet Growth by Relocalizing from the ER to Lipid Droplets. Developmental Cell 24, 384-399 (2013). 47. Vergani L. Fatty Acids and Effects on In Vitro and In Vivo Models of Liver Steatosis. Curr Med Chem 26, 3439-3456 (2019). 48. Bellanti F, Villani R, Facciorusso A, Vendemiale G, Serviddio G. Lipid oxidation products in the pathogenesis of non-alcoholic steatohepatitis. Free Radic Biol Med 111, 173-185 (2017). 49. Tiwari S, Siddiqi SA. Intracellular Trafficking and Secretion of VLDL. Arteriosclerosis Thrombosis and Vascular Biology 32, 1079-1086 (2012). 50. Alexander CA, Hamilton RL, Havel RJ. Subcellular-Localization of B Apoprotein of Plasma Lipoproteins in Rat-Liver. Journal of Cell Biology 69, 241-263 (1976). 51. Boren J, Rustaeus S, Olofsson SO. Studies on the Assembly of Apolipoprotein-B-100- Containing and Apolipoprotein-B-48-Containing Very-Low-Density Lipoproteins in Mca- Rh7777 Cells. Journal of Biological Chemistry 269, 25879-25888 (1994). 52. Gusarova V, Brodsky JL, Fisher EA. Apolipoprotein B100 exit from the endoplasmic reticulum (ER) is COPII-dependent, and its lipidation to very low density lipoprotein occurs post-ER. J Biol Chem 278, 48051-48058 (2003). 53. Santos AJ, Nogueira C, Ortega-Bellido M, Malhotra V. TANGO1 and Mia2/cTAGE5 (TALI) cooperate to export bulky pre-chylomicrons/VLDLs from the endoplasmic reticulum. J Cell Biol 213, 343-354 (2016). 54. Pitman JL, Bonnet DJ, Curtiss LK, Gekakis N. Reduced cholesterol and triglycerides in mice with a mutation in Mia2, a liver protein that localizes to ER exit sites. J Lipid Res 52, 1775- 1786 (2011). 48 Ying Xia 55. Jacquier N, Choudhary V, Mari M, Toulmay A, Reggiori F, Schneiter R. Lipid droplets are functionally connected to the endoplasmic reticulum in Saccharomyces cerevisiae. J Cell Sci 124, 2424-2437 (2011). 56. Huttasch M, Roden M, Kahl S. Obesity and MASLD: Is weight loss the (only) key to treat metabolic liver disease? Metabolism, 155937 (2024). 57. Das K, Chowdhury A. Lean NASH: distinctiveness and clinical implication. Hepatol Int 7 Suppl 2, 806-813 (2013). 58. Smith GI, et al. Insulin resistance drives hepatic de novo lipogenesis in nonalcoholic fatty liver disease. Journal of Clinical Investigation 130, 1453-1460 (2020). 59. Garg R, Kumariya S, Katekar R, Verma S, Goand UK, Gayen JR. JNK signaling pathway in metabolic disorders: An emerging therapeutic target. European Journal of Pharmacology 901, (2021). 60. Kolczynska K, Loza-Valdes A, Hawro I, Sumara G. Diacylglycerol-evoked activation of PKC and PKD isoforms in regulation of glucose and lipid metabolism: a review. Lipids Health Dis 19, 113 (2020). 61. Yuan L, Terrrault NA. PNPLA3 and nonalcoholic fatty liver disease: towards personalized medicine for fatty liver. Hepatobiliary Surg Nutr 9, 353-356 (2020). 62. BasuRay S, Wang Y, Smagris E, Cohen JC, Hobbs HH. Accumulation of PNPLA3 on lipid droplets is the basis of associated hepatic steatosis. Proc Natl Acad Sci U S A 116, 9521-9526 (2019). 63. Banini BA, et al. Identification of a Metabolic, Transcriptomic, and Molecular Signature of Patatin-Like Phospholipase Domain Containing 3-Mediated Acceleration of Steatohepatitis. Hepatology 73, 1290-1306 (2021). 64. Bruschi FV, Tardelli M, Herac M, Claudel T, Trauner M. Metabolic regulation of hepatic PNPLA3 expression and severity of liver fibrosis in patients with NASH. Liver Int 40, 1098- 1110 (2020). 65. Carr RM, Ahima RS. Pathophysiology of lipid droplet proteins in liver diseases. Exp Cell Res 340, 187-192 (2016). 66. McManaman JL, et al. Perilipin-2-null mice are protected against diet-induced obesity, adipose inflammation, and fatty liver disease. J Lipid Res 54, 1346-1359 (2013). 67. Huang DQ, et al. Hepatocellular carcinoma surveillance - utilization, barriers and the impact of changing aetiology. Nature Reviews Gastroenterology & Hepatology, 797-809 (2023). 68. Singal AG, Kanwal F, Llovet JM. Global trends in hepatocellular carcinoma epidemiology: implications for screening, prevention and therapy. Nat Rev Clin Oncol, 864-884 (2023). 69. Johnson P, Zhou Q, Dao DY, Lo YMD. Circulating biomarkers in the diagnosis and management of hepatocellular carcinoma. Nat Rev Gastroenterol Hepatol 19, 670-681 (2022). 70. Vogel A, Meyer T, Sapisochin G, Salem R, Saborowski A. Hepatocellular carcinoma. Lancet 400, 1345-1362 (2022). 71. Huang DQ, Singal AG, Kono Y, Tan DJH, El-Serag HB, Loomba R. Changing global epidemiology of liver cancer from 2010 to 2019: NASH is the fastest growing cause of liver cancer. Cell Metab 34, 969-977 e962 (2022). 72. Friedman SL. We’ve hit the iceberg – the threat of NASH-related HCC arrives. J Hepatol 60, 1-2 (2014). 49 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis 73. Ioannou GN. Epidemiology and risk-stratification of NAFLD-associated HCC. Journal of Hepatology 75, 1476-1484 (2021). 74. Yip TCF, Lee HW, Chan WK, Wong GLH, Wong VWS. Asian perspective on NAFLD- associated HCC. Journal of Hepatology 76, 726-734 (2022). 75. Tanaka S, et al. Increased hepatic oxidative DNA damage in patients with nonalcoholic steatohepatitis who develop hepatocellular carcinoma. Journal of Gastroenterology 48, 1249- 1258 (2013). 76. Grohmann M, et al. Obesity Drives STAT-1-Dependent NASH and STAT-3-Dependent HCC. Cell 175, 1289-1306 e1220 (2018). 77. Naugler WE, et al. Gender disparity in liver cancer due to sex differences in MyD88-dependent IL-6 production. Science 317, 121-124 (2007). 78. Fromenty B, Roden M. Mitochondrial alterations in fatty liver diseases. Journal of Hepatology 78, 415-429 (2023). 79. Francque S, et al. Nonalcoholic steatohepatitis: the role of peroxisome proliferator-activated receptors. Nature Reviews Gastroenterology & Hepatology 18, 24-39 (2021). 80. Longo M, Meroni M, Paolini E, Macchi C, Dongiovanni P. Mitochondrial dynamics and nonalcoholic fatty liver disease (NAFLD): new perspectives for a fairy-tale ending? Metabolism 117, 154708 (2021). 81. Liang JQ, et al. Dietary cholesterol promotes steatohepatitis related hepatocellular carcinoma through dysregulated metabolism and calcium signaling. Nature Communications 9, (2018). 82. Lally JSV, et al. Inhibition of Acetyl-CoA Carboxylase by Phosphorylation or the Inhibitor ND-654 Suppresses Lipogenesis and Hepatocellular Carcinoma. Cell Metabolism 29, 174-182 (2019). 83. Liu L, Liao JZ, He XX, Li PY. The role of autophagy in hepatocellular carcinoma: friend or foe. Oncotarget 8, 57707-57722 (2017). 84. Xu D, et al. The Role of Nrf2 in Liver Disease: Novel Molecular Mechanisms and Therapeutic Approaches. Front Pharmacol 9, 1428 (2018). 85. Sun K, et al. Paradoxical role of autophagy in the dysplastic and tumor-forming stages of hepatocarcinoma development in rats. Cell Death Dis 4, e501 (2013). 86. Zhang DY, Friedman SL. Fibrosis-dependent mechanisms of hepatocarcinogenesis. Hepatology 56, 769-775 (2012). 87. Takahashi Y, Dungubat E, Kusano H, Fukusato T. Pathology and Pathogenesis of Metabolic Dysfunction-Associated Steatotic Liver Disease-Associated Hepatic Tumors. Biomedicines 11, (2023). 88. Anstee QM, Reeves HL, Kotsiliti E, Govaere O, Heikenwalder M. From NASH to HCC: current concepts and future challenges. Nature Reviews Gastroenterology & Hepatology 16, 411-428 (2019). 89. Liu YL, et al. Carriage of the PNPLA3 rs738409 C >G polymorphism confers an increased risk of non-alcoholic fatty liver disease associated hepatocellular carcinoma. J Hepatol 61, 75-81 (2014). 90. Trépo E, et al. Association Between the (rs738409 C>G) Variant and Hepatocellular Carcinoma: Evidence From a Meta-Analysis of Individual Participant Data. Hepatology 59, 2170-2177 (2014). 50 Ying Xia 91. European Association for the Study of the Liver. Electronic address eee, European Association for the Study of the L. EASL Clinical Practice Guidelines: Management of hepatocellular carcinoma. J Hepatol 69, 182-236 (2018). 92. Donati B, et al. MBOAT7 rs641738 variant and hepatocellular carcinoma in non-cirrhotic individuals. Sci Rep 7, 4492 (2017). 93. Yang J, et al. PNPLA3 and TM6SF2 variants as risk factors of hepatocellular carcinoma across various etiologies and severity of underlying liver diseases. Int J Cancer 144, 533-544 (2019). 94. Llovet JM, et al. Nonalcoholic steatohepatitis-related hepatocellular carcinoma: pathogenesis and treatment. Nature Reviews Gastroenterology & Hepatology 20, 487-503 (2023). 95. Llovet JM, Brú C, Bruix J. Prognosis of hepatocellular carcinoma: The BCLC staging classification. Seminars in Liver Disease 19, 329-338 (1999). 96. Dyson J, et al. Hepatocellular cancer: The impact of obesity, type 2 diabetes and a multidisciplinary team. Journal of Hepatology 60, 110-117 (2014). 97. El-Khoueiry AB, Hanna DL, Llovet J, Kelley RK. Cabozantinib: An evolving therapy for hepatocellular carcinoma. Cancer Treatment Reviews 98, (2021). 98. Kudo M. Durvalumab Plus Tremelimumab: A Novel Combination Immunotherapy for Unresectable Hepatocellular Carcinoma. Liver Cancer 11, 87-93 (2022). 99. Pfister D, et al. NASH limits anti-tumour surveillance in immunotherapy-treated HCC. Nature 592, 450-456 (2021). 100. Wang YM, et al. Pharmacological therapy of metabolic dysfunction-associated steatotic liver disease-driven hepatocellular carcinoma. Frontiers in Pharmacology 14, (2024). 101. Baier A, Szyszka R. Compounds from Natural Sources as Protein Kinase Inhibitors. Biomolecules 10, (2020). 102. Manning G, Whyte DB, Martinez R, Hunter T, Sudarsanam S. The protein kinase complement of the human genome. Science 298, 1912-1934 (2002). 103. Strange K, Denton J, Nehrke K. Ste20-type kinases: evolutionarily conserved regulators of ion transport and cell volume. Physiology (Bethesda) 21, 61-68 (2006). 104. Dan I, Watanabe NM, Kusumi A. The Ste20 group kinases as regulators of MAP kinase cascades. Trends Cell Biol 11, 220-230 (2001). 105. Zhao ZS, Manser E. PAK family kinases: Physiological roles and regulation. Cell Logist 2, 59- 68 (2012). 106. Wu A, Jiang X. p21-Activated kinases as promising therapeutic targets in hematological malignancies. Leukemia, (2021). 107. Kumar R, Sanawar R, Li X, Li F. Structure, biochemistry, and biology of PAK kinases. Gene 605, 20-31 (2017). 108. Radu M, Semenova G, Kosoff R, Chernoff J. PAK signalling during the development and progression of cancer. Nat Rev Cancer 14, 13-25 (2014). 109. Ye DZ, Field J. PAK signaling in cancer. Cell Logist 2, 105-116 (2012). 110. Whale A, Hashim FN, Fram S, Jones GE, Wells CM. Signalling to cancer cell invasion through PAK family kinases. Front Biosci (Landmark Ed) 16, 849-864 (2011). 111. Delpire E. The mammalian family of sterile 20p-like protein kinases. Pflugers Arch 458, 953- 967 (2009). 51 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis 112. Ling P, Lu TJ, Yuan CJ, Lai MD. Biosignaling of mammalian Ste20-related kinases. Cell Signal 20, 1237-1247 (2008). 113. Strange K, Denton J, Nehrke K. Ste20-type kinases: Evolutionarily conserved regulators of ion transport and cell volume. Physiology 21, 61-68 (2006). 114. Hutchison M, Berman KS, Cobb MH. Isolation of TAO1, a protein kinase that activates MEKs in stress-activated protein kinase cascades. Journal of Biological Chemistry 273, 28625-28632 (1998). 115. Chen Z, Hutchison M, Cobb MH. Isolation of the protein kinase TAO2 and identification of its mitogen-activated protein kinase/extracellular signal-regulated kinase kinase binding domain. Journal of Biological Chemistry 274, 28803-28807 (1999). 116. Tassi E, Biesova Z, Di Fiore PP, Gutkind JS, Wong WT. Human JIK, a novel member of the STE20 kinase family that inhibits JNK and is negatively regulated by epidermal growth factor. Journal of Biological Chemistry 274, 33287-33295 (1999). 117. Byeon S, Yadav S. Pleiotropic functions of TAO kinases and their dysregulation in neurological disorders. Science Signaling 17, (2024). 118. Priyanka P, et al. A global phosphosite-correlated network map of Thousand And One Kinase 1 (TAOK1). International Journal of Biochemistry & Cell Biology 170, (2024). 119. Raman M, Earnest S, Zhang K, Zhao YM, Cobb MH. TAO kinases mediate activation of p38 in response to DNA damage. Embo J 26, 2005-2014 (2007). 120. Fang CY, Lai TC, Hsiao M, Chang YC. The Diverse Roles of TAO Kinases in Health and Diseases. International Journal of Molecular Sciences 21, (2020). 121. Martinez-Limon A, Joaquin M, Caballero M, Posas F, de Nadal E. The p38 Pathway: From Biology to Cancer Therapy. Int J Mol Sci 21, (2020). 122. Chen Z, Raman M, Chen L, Lee SF, Gilman AG, Cobb MH. TAO (thousand-and-one amino acid) protein kinases mediate signaling from carbachol to p38 mitogen-activated protein kinase and ternary complex factors. J Biol Chem 278, 22278-22283 (2003). 123. Zhang WP, et al. Cloning of DPK, a novel dendritic cell-derived protein kinase activating the ERK1/ERK2 and JNK/SAPK pathways. Biochem Bioph Res Co 274, 872-879 (2000). 124. Xia Y, et al. STE20-type kinase TAOK3 regulates hepatic lipid partitioning. Molecular Metabolism 54, (2021). 125. Xia Y, et al. Silencing of STE20-type kinase TAOK1 confers protection against hepatocellular lipotoxicity through metabolic rewiring. Hepatology Communications 7, (2023). 126. Boggiano JC, Vanderzalm PJ, Fehon RG. Tao-1 Phosphorylates Hippo/MST Kinases to Regulate the Hippo-Salvador-Warts Tumor Suppressor Pathway. Developmental Cell 21, 888- 895 (2011). 127. Poon CL, Lin JI, Zhang X, Harvey KF. The sterile 20-like kinase Tao-1 controls tissue growth by regulating the Salvador-Warts-Hippo pathway. Dev Cell 21, 896-906 (2011). 128. Plouffe SW, et al. Characterization of Hippo Pathway Components by Gene Inactivation. Mol Cell 64, 993-1008 (2016). 129. Meng Z, et al. MAP4K family kinases act in parallel to MST1/2 to activate LATS1/2 in the Hippo pathway. Nat Commun 6, 8357 (2015). 130. Zhang Z, et al. TAOK1 negatively regulates IL-17-mediated signaling and inflammation. Cell Mol Immunol 15, 794-802 (2018). 52 Ying Xia 131. Zhu LX, et al. TAOK1 positively regulates TLR4-induced inflammatory responses by promoting ERK1/2 activation in macrophages. Molecular Immunology 122, 124-131 (2020). 132. Poirier A, Wu CY, Hincapie AM, Martinez-Cordova Z, Abidin BM, Tremblay ML. TAOK3 limits age-associated inflammation by negatively modulating macrophage differentiation and their production of TNFα. Immunity & Ageing 20, (2023). 133. Timm T, et al. MARKK, a Ste20-like kinase, activates the polarity-inducing kinase MARK/PAR-1. Embo J 22, 5090-5101 (2003). 134. Johne C, Matenia D, Li XY, Timm T, Balusamy K, Mandelkow EM. Spred1 and TESK1 - Two new interaction partners of the kinase MARKK/TAO1 that link the microtubule and actin cytoskeleton. Molecular Biology of the Cell 19, 1391-1403 (2008). 135. Byeon S, et al. Proteomic Identification of Phosphorylation-Dependent Septin 7 Interactors that Drive Dendritic Spine Formation. Front Cell Dev Biol 10, 836746 (2022). 136. Wojtala RL, Tavares IA, Morton PE, Valderrama F, Thomas NSB, Morris JDH. Prostate- derived Sterile 20-like Kinases (PSKs/TAOKs) Are Activated in Mitosis and Contribute to Mitotic Cell Rounding and Spindle Positioning. Journal of Biological Chemistry 286, 30161- 30170 (2011). 137. Zihni C, Mitsopoulos C, Tavares IA, Ridley AJ, Morris JD. Prostate-derived sterile 20-like kinase 2 (PSK2) regulates apoptotic morphology via C-Jun N-terminal kinase and Rho kinase- 1. J Biol Chem 281, 7317-7323 (2006). 138. Nerstedt A, et al. Lipid droplet-associated kinase STK25 regulates peroxisomal activity and metabolic stress response in steatotic liver. Journal of Lipid Research 61, 178-191 (2020). 139. Cansby E, et al. Protein kinase MST3 modulates lipid homeostasis in hepatocytes and correlates with nonalcoholic steatohepatitis in humans. Faseb J 33, 9974-9989 (2019). 140. Huang RC, et al. Genome-wide methylation analysis identifies differentially methylated CpG loci associated with severe obesity in childhood. Epigenetics-Us 10, 995-1005 (2015). 141. Xia Y, et al. Knockout of STE20-type kinase TAOK3 does not attenuate diet-induced NAFLD development in mice. Molecular Medicine 29, (2023). 142. Caputo M, et al. STE20-Type Protein Kinase MST4 Controls NAFLD Progression by Regulating Lipid Droplet Dynamics and Metabolic Stress in Hepatocytes. Hepatol Commun 5, 1183-1200 (2021). 143. Anand SK, et al. Inhibition of MAP4K4 signaling initiates metabolic reprogramming to protect hepatocytes from lipotoxic damage. J Lipid Res 63, 100238 (2022). 144. Amrutkar M, et al. Protein kinase STK25 controls lipid partitioning in hepatocytes and correlates with liver fat content in humans. Diabetologia 59, 341-353 (2016). 145. Amrutkar M, et al. STK25 is a critical determinant in nonalcoholic steatohepatitis. Faseb J 30, 3628-3643 (2016). 146. Cansby E, et al. Targeted Delivery of Stk25 Antisense Oligonucleotides to Hepatocytes Protects Mice Against Nonalcoholic Fatty Liver Disease. Cell Mol Gastroenterol Hepatol 7, 597-618 (2019). 147. Nuñez-Durán E, et al. Serine/Threonine Protein Kinase 25 Antisense Oligonucleotide Treatment Reverses Glucose Intolerance, Insulin Resistance, and Nonalcoholic Fatty Liver Disease in Mice. Hepatology Communications 2, 69-83 (2018). 148. Amrutkar M, et al. Genetic Disruption of Protein Kinase STK25 Ameliorates Metabolic Defects in a Diet-Induced Type 2 Diabetes Model. Diabetes 64, 2791-2804 (2015). 53 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis 149. Caputo M, et al. Silencing of STE20-type kinase MST3 in mice with antisense oligonucleotide treatment ameliorates diet-induced nonalcoholic fatty liver disease. Faseb Journal 35, (2021). 150. Maes B, et al. STE20 kinase TAOK3 regulates type 2 immunity and metabolism in obesity. J Exp Med 220, (2023). 151. Mahlapuu M, Caputo M, Xia Y, Cansby E. GCKIII kinases in lipotoxicity: Roles in NAFLD and beyond. Hepatology Communications 6, 2613-2622 (2022). 152. Pombo CM, Bonventre JV, Molnar A, Kyriakis J, Force T. Activation of a human Ste20-like kinase by oxidant stress defines a novel stress response pathway. Embo J 15, 4537-4546 (1996). 153. Osada S, et al. YSK1, a novel mammalian protein kinase structurally related to Ste20 and SPS1, but is not involved in the known MAPK pathways. Oncogene 14, 2047-2057 (1997). 154. Delpire E. The mammalian family of sterile 20p-like protein kinases. Pflugers Archiv- European Journal of Physiology 458, 953-967 (2009). 155. Sugden PH, McGuffin LJ, Clerk A. SOcK, MiSTs, MASK and STicKs: the GCKIII (germinal centre kinase III) kinases and their heterologous protein-protein interactions. Biochemical Journal 454, 13-30 (2013). 156. Lee WS, Hsu CY, Wang PL, Huang CYF, Chang CH, Yuan CJ. Identification and characterization of the nuclear import and export signals of the mammalian Ste20-like protein kinase 3. Febs Letters 572, 41-45 (2004). 157. Zalvide J, et al. The CCM3-GCKIII partnership. Histology and Histopathology 28, 1265-1272 (2013). 158. Pombo CM, Tsujita T, Kyriakis JM, Bonventre JV, Force T. Activation of the Ste20-like oxidant stress response kinase-1 during the initial stages of chemical anoxia-induced necrotic cell death - Requirement for dual inputs of oxidant stress and increased cytosolic [Ca2+]. Journal of Biological Chemistry 272, 29372-29379 (1997). 159. Filippi BM, et al. MO25 is a master regulator of SPAK/OSR1 and MST3/MST4/YSK1 protein kinases. Embo J 30, 1730-1741 (2011). 160. ten Klooster JP, et al. Mst4 and Ezrin Induce Brush Borders Downstream of the Lkb1/Strad/Mo25 Polarization Complex. Developmental Cell 16, 551-562 (2009). 161. Hao Q, et al. Structural insights into regulatory mechanisms of MO25-mediated kinase activation. J Struct Biol 186, 224-233 (2014). 162. Preisinger C, et al. YSK1 is activated by the Golgi matrix protein GM130 and plays a role in cell migration through its substrate 14-3-3ζ. Journal of Cell Biology 164, 1009-1020 (2004). 163. Karaman MW, et al. A quantitative analysis of kinase inhibitor selectivity. Nat Biotechnol 26, 127-132 (2008). 164. Kiyeleko S, et al. Targeting non-alcoholic fatty liver disease: Design, X-ray co-crystal structure and synthesis of 'first-in-kind' inhibitors of serine/threonine kinase25. Bioorg Med Chem Lett 75, (2022). 165. Crooke ST, Baker BF, Crooke RM, Liang XH. Antisense technology: an overview and prospectus. Nature Reviews Drug Discovery 20, 427-453 (2021). 166. Amrutkar M, et al. Protein kinase STK25 regulates hepatic lipid partitioning and progression of liver steatosis and NASH. FASEB J 29, 1564-1576 (2015). 167. Zhang YC, et al. STK25 enhances hepatocellular carcinoma progression through the STRN/AMPK/ACC1 pathway. Cancer Cell International 22, (2022). 54 Ying Xia 168. Jiang J, Zheng Y, Chen FR, Dong L, Guo XY. Activation of YAP1 by STK25 contributes to the progression of hepatocellular carcinoma. Tissue & Cell 76, (2022). 169. Kurhe Y, et al. Antagonizing STK25 Signaling Suppresses the Development of Hepatocellular Carcinoma Through Targeting Metabolic, Inflammatory, and Pro-Oncogenic Pathways. Cell Mol Gastroenterol Hepatol 13, 405-423 (2022). 170. Min L, He B, Hui L. Mitogen-activated protein kinases in hepatocellular carcinoma development. Semin Cancer Biol 21, 10-20 (2011). 171. He G, et al. Hepatocyte IKKbeta/NF-kappaB inhibits tumor promotion and progression by preventing oxidative stress-driven STAT3 activation. Cancer Cell 17, 286-297 (2010). 172. Calvisi DF, et al. Ubiquitous activation of Ras and Jak/Stat pathways in human HCC. Gastroenterology 130, 1117-1128 (2006). 173. Cansby E, et al. Depletion of protein kinase STK25 ameliorates renal lipotoxicity and protects against diabetic kidney disease. JCI Insight 5, (2020). 174. Sutt S, et al. STK25 regulates oxidative capacity and metabolic efficiency in adipose tissue. J Endocrinol 238, 187-202 (2018). 175. Cansby E., et al. Silencing of STE20-Type Kinase STK25 in Human Aortic Endothelial and Smooth MuscleCells is Atheroprotective. Atherosclerosis thrombosis and vascular biology, (2022). 176. Zhang HY, Ma X, Peng SH, Nan X, Zhao HS. Differential expression of MST4, STK25 and PDCD10 between benign prostatic hyperplasia and prostate cancer. International Journal of Clinical and Experimental Pathology 7, 8105-8111 (2014). 177. Lee MH, Kundu JK, Choi BY. p21 Regulation by hYSK1 Activates SP-1 Transcription Factor and Increases MMP-2 Expression under Hypoxic Conditions. International Journal of Molecular Sciences 20, (2019). 178. Wu F, et al. STK25-induced inhibition of aerobic glycolysis via GOLPH3-mTOR pathway suppresses cell proliferation in colorectal cancer. Journal of Experimental & Clinical Cancer Research 37, (2018). 179. Matsuki T, et al. Identification of as a Genetic Modifier of Tau Phosphorylation in Mutant Mice. Plos One 7, (2012). 180. Costa B, et al. STK25 Protein Mediates TrkA and CCM2 Protein-dependent Death in Pediatric Tumor Cells of Neural Origin. Journal of Biological Chemistry 287, 29285-29289 (2012). 181. Kannt A, et al. Association of nicotinamide-N-methyltransferase mRNA expression in human adipose tissue and the plasma concentration of its product, 1-methylnicotinamide, with insulin resistance. Diabetologia 58, 799-808 (2015). 182. Kleiner DE, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology 41, 1313-1321 (2005). 183. Cancer Genome Atlas Research N, et al. The Cancer Genome Atlas Pan-Cancer analysis project. Nat Genet 45, 1113-1120 (2013). 184. Consortium GT. The Genotype-Tissue Expression (GTEx) project. Nat Genet 45, 580-585 (2013). 185. Samanez CH, et al. The human hepatocyte cell lines IHH and HepaRG: models to study glucose, lipid and lipoprotein metabolism. Arch Physiol Biochem 118, 102-111 (2012). 55 STE20-Type Kinases in Liver Lipid Metabolism and Hepatocarcinogenesis 186. Javitt NB. Hep-G2 Cells as a Resource for Metabolic Studies - Lipoprotein, Cholesterol, and Bile-Acids. Faseb J 4, 161-168 (1990). 187. Oswald A, Chakraborty A, Ni Y, Wettengel JM, Urban S, Protzer U. Concentration of Na(+)- taurocholate-cotransporting polypeptide expressed after in vitro-transcribed mRNA transfection determines susceptibility of hepatoma cells for hepatitis B virus. Sci Rep 11, 19799 (2021). 188. Hsu CM, et al. Emodin inhibits the growth of hepatoma cells: Finding the common anti-cancer pathway using Huh7, Hep3B, and HepG2 cells. Biochem Bioph Res Co 392, 473-478 (2010). 189. Park JG, et al. Characterization of Cell-Lines Established from Human Hepatocellular- Carcinoma. Int J Cancer 62, 276-282 (1995). 190. Xu L, et al. Human hepatic stellate cell lines, LX-1 and LX-2: new tools for analysis of hepatic fibrosis. Gut 54, 142-151 (2005). 191. Tsuchiya S, Yamabe M, Yamaguchi Y, Kobayashi Y, Konno T, Tada K. Establishment and Characterization of a Human Acute Monocytic Leukemia-Cell Line (Thp-1). Int J Cancer 26, 171-176 (1980). 192. Cansby E, Nerstedt A, Amrutkar M, Duran EN, Smith U, Mahlapuu M. Partial hepatic resistance to IL-6-induced inflammation develops in type 2 diabetic mice, while the anti- inflammatory effect of AMPK is maintained. Mol Cell Endocrinol 393, 143-151 (2014). 193. Cansby E, et al. Increased expression of STK25 leads to impaired glucose utilization and insulin sensitivity in mice challenged with a high-fat diet. FASEB J 27, 3660-3671 (2013). 194. Ayala JE, et al. Standard operating procedures for describing and performing metabolic tests of glucose homeostasis in mice. Dis Model Mech 3, 525-534 (2010). 195. Vallerand AL, Perusse F, Bukowiecki LJ. Cold exposure potentiates the effect of insulin on in vivo glucose uptake. Am J Physiol 253, E179-186 (1987). 196. Chursa U, et al. Overexpression of protein kinase STK25 in mice exacerbates ectopic lipid accumulation, mitochondrial dysfunction and insulin resistance in skeletal muscle. Diabetologia 60, 553-567 (2017). 197. Bustin SA, et al. The MIQE Guidelines: Minimum Information for Publication of Quantitative Real-Time PCR Experiments. Clin Chem 55, 611-622 (2009). 198. Liu J, et al. An Integrated TCGA Pan-Cancer Clinical Data Resource to Drive High-Quality Survival Outcome Analytics. Cell 173, 400-416 e411 (2018). 199. Su W, et al. Comparative proteomic study reveals 17beta-HSD13 as a pathogenic protein in nonalcoholic fatty liver disease. Proc Natl Acad Sci U S A 111, 11437-11442 (2014). 200. Krahmer N, et al. Protein correlation profiles identify lipid droplet proteins with high confidence. Mol Cell Proteomics 12, 1115-1126 (2013). 201. Sanyal AJ. Past, present and future perspectives in nonalcoholic fatty liver disease. Nat Rev Gastro Hepat 16, 377-386 (2019). 202. Friedman SL, Neuschwander-Tetri BA, Rinella M, Sanyal AJ. Mechanisms of NAFLD development and therapeutic strategies. Nat Med 24, 908-922 (2018). 203. Win S, Than TA, Min RWM, Aghajan M, Kaplowitz N. c-Jun N-Terminal Kinase Mediates Mouse Liver Injury Through a Novel Sab (SH3BP5)-Dependent Pathway Leading to Inactivation of Intramitochondrial Src. Hepatology 63, 1987-2003 (2016). 56 Ying Xia 204. Leboucher GP, et al. Stress-Induced Phosphorylation and Proteasomal Degradation of Mitofusin 2 Facilitates Mitochondrial Fragmentation and Apoptosis. Mol Cell 47, 547-557 (2012). 205. Batchuluun B, Pinkosky SL, Steinberg GR. Lipogenesis inhibitors: therapeutic opportunities and challenges. Nat Rev Drug Discov 21, 283-305 (2022). 206. Jha P, et al. Role of adipose triglyceride lipase (PNPLA2) in protection from hepatic inflammation in mouse models of steatohepatitis and endotoxemia. Hepatology 59, 858-869 (2014). 207. Reid BN, et al. Hepatic overexpression of hormone-sensitive lipase and adipose triglyceride lipase promotes fatty acid oxidation, stimulates direct release of free fatty acids, and ameliorates steatosis. J Biol Chem 283, 13087-13099 (2008). 208. Turpin SM, et al. Adipose triacylglycerol lipase is a major regulator of hepatic lipid metabolism but not insulin sensitivity in mice. Diabetologia 54, 146-156 (2011). 209. Kazankov K, et al. The role of macrophages in nonalcoholic fatty liver disease and nonalcoholic steatohepatitis. Nat Rev Gastro Hepat 16, 145-159 (2019). 210. Schuster S, Cabrera D, Arrese M, Feldstein AE. Triggering and resolution of inflammation in NASH. Nat Rev Gastro Hepat 15, 349-364 (2018). 211. De Souza AT, et al. Transcriptional and phenotypic comparisons of Ppara knockout and siRNA knockdown mice. Nucleic Acids Res 34, 4486-4494 (2006). 212. He L, et al. XIAP Knockdown in Alcohol-Associated Liver Disease Models Exhibits Divergent in vitro and in vivo Phenotypes Owing to a Potential Zonal Inhibitory Role of SMAC. Front Physiol 12, 664222 (2021). 213. Ievlev V, et al. Krt14 and Krt15 differentially regulate regenerative properties and differentiation potential of airway basal cells. Jci Insight 8, (2023). 214. Cansby E, et al. Protein kinase MST3 modulates lipid homeostasis in hepatocytes and correlates with nonalcoholic steatohepatitis in humans. FASEB J 33, 9974-9989 (2019). 215. Giacomello M, Pyakurel A, Glytsou C, Scorrano L. The cell biology of mitochondrial membrane dynamics. Nature Reviews Molecular Cell Biology 21, 204-224 (2020). 216. Quintana-Cabrera R, Scorrano L. Determinants and outcomes of mitochondrial dynamics. Molecular Cell 83, 857-876 (2023). 57