Cholangiocarcinoma progression depends on the uptake and metabolization of extracellular lipids

Hepatology. 2022;76:1617–1633. | 1617 wileyonlinelibrary.com/journal/hep ORIGINAL ARTICLE Cholangiocarcinoma progression depends on the uptake and metabolization of extracellular lipids Mikel Ruiz de Gauna1 | Francesca Biancaniello2,3 | Francisco GonzálezRomero1 | Pedro M. Rodrigues2,4,5 | Ainhoa Lapitz2 | Beatriz GómezSantos1 | Paula Olaizola2 | Sabina Di Matteo2,3 | Igor Aurrekoetxea1,6 | Ibone Labiano2 | Ane NievaZuluaga1 | Asier BenitoVicente7,8 | María J. Perugorria2,4 | Maider ApodakaBiguri1 | Nuno A. Paiva2 | Diego Sáenz de Urturi1 | Xabier Buqué1 | Igotz Delgado1 | César Martín7,8 | Mikel Azkargorta9 | Felix Elortza4,9 | Diego F. Calvisi10 | Jesper B. Andersen11 | Domenico Alvaro3 | Vincenzo Cardinale12 | Luis Bujanda2,4 | Jesús M. Banales2,4,5,13 | Patricia Aspichueta1,4,6 1Faculty of Medicine and Nursing, Department of Physiology, University of the Basque Country (UPV/EHU), Leioa, Spain 2Department of Liver and Gastrointestinal Diseases, Biodonostia Health Research Institute, Donostia University Hospital, University of the Basque Country (UPV/EHU), San Sebastian, Spain 3Department of Translational and Precision Medicine, “Sapienza” University of Rome, Rome, Italy 4National Institute for the Study of Liver and Gastrointestinal Diseases (CIBERehd, Carlos III Health Institute), Madrid, Spain 5IKERBASQUE, Basque Foundation for Science, Bilbao, Spain 6Biocruces Bizkaia Health Research Institute, Cruces University Hospital, Barakaldo, Spain 7Department of Molecular Biophysics, Biofisika Institute (University of Basque Country and Consejo Superior de Investigaciones Científicas (UPV/EHU, CSIC), Leioa, Spain 8Department of Biochemistry and Molecular Biology, University of the Basque Country (UPV/EHU), Leioa, Spain 9Proteomics Platform, CIC bioGUNE, BRTA (Basque Research and Technology Alliance), ProteoRedISCIII, CIBERehd, Bizkaia Science and Technology Park, Derio, Spain 10Institute of Pathology, University of Regensburg, Regensburg, Germany 11Biotech Research & Innovation Centre (BRIC), Department of Health and Medical Sciences, University of Copenhagen, Copenhagen, Denmark 12Department of MedicoSurgical Sciences and Biotechnology, "Sapienza” University of Rome, Rome, Italy 13Department of Biochemistry and Genetics, School of Sciences, University of Navarra, Pamplona, Spain Received: 3 May 2021 | Revised: 17 December 2021 | Accepted: 17 December 2021 DOI: 10.1002/hep.32344 Mikel Ruiz de Gauna and Francesca Biancaniello share first authorship. Jesús M. Banales and Patricia Aspichueta share senior authorship. Abbreviations: ACADM, acylCoA dehydrogenase medium chain; CCA, cholangiocarcinoma; CE, cholesteryl ester; CFSE, carboxyfluorescein succinimidyl ester; CL, cholesterol; eCCA, extrahepatic cholangiocarcinoma; FA, fatty acid; FABP, fatty acid– binding protein; FAO, fatty acid oxidation; FASN, fatty acid synthase; iCCA, intrahepatic cholangiocarcinoma; LPL, lipoprotein lipase; NHC, normal human cholangiocyte; Nicd1, NOTCH1 intracellular cytoplasmic domain; PC, phosphatidylcholine; PCNA, proliferating cell nuclear antigen; PG, prostaglandin; TG, triglyceride. Correspondence Jesús M. Banales, Department of Liver and Gastrointestinal Diseases, Biodonostia Health Research Institute – Donostia University Hospital, Paseo del Dr. Begiristain s/n, E20014, San Sebastian, Spain. Email: [email protected] Abstract Background and Aims: Cholangiocarcinoma (CCA) includes a heterogeneous group of biliary cancers with a dismal prognosis. We investigated if lipid metabolism is disrupted in CCA and its role in tumor proliferation. This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited and is not used for commercial purposes. © 2022 The Authors. Hepatology published by Wiley Periodicals LLC on behalf of American Association for the Study of Liver Diseases Downloaded from http://journals.lww.com/hep by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWn YQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/23/2025 1618 | CATABOLISM OF EXTRACELLULAR LIPIDS PROMOTES CCA PROGRESSION INTRODUCTION Cholangiocarcinoma (CCA) includes a diverse group of biliary malignant tumors and represents the second most common primary liver cancer.[1] According to the anatomical location, these cancers are classified as intrahepatic (iCCA), perihilar (pCCA), or distal (dCCA) tumors, although pCCA and dCCA were previously considered as extrahepatic (eCCA).[2] CCA is a rare cancer globally (0.3– 6 cases per 100,000 people); however, its incidence and associated mortality have been significantly rising over the last few decades.[1,2] Currently, the only treatment with curative intent is the surgical resection of the tumor. Nevertheless, patients with CCA are usually asymptomatic until late, unresectable stages of the disease. This highlights the need to Patricia Aspichueta, Department of Physiology, University of the Basque Country, Barrio Sarriena s/n, E48940, Leioa, Spain. Email: patri[email protected] Funding information This work was supported by “Ayudas para apoyar grupos de investigación del sistema Universitario Vasco” (IT97116 to PA), MCIU/AEI/FEDER, UE (2018095134B100 to PA and by the University of Basque Country COLAB20/01 to PA; Spanish Carlos III Health Institute (ISCIII) (FIS PI15/01132, PI18/01075, PI21/00922, and Miguel Servet Program CON14/00129 and CPII19/00008 to JMB; FIS PI14/00399, PI17/00022 and PI20/00186 to MJP; Sara Borrell [CD19/00254 to PMR]) cofinanced by “Fondo Europeo de Desarrollo Regional” (FEDER); CIBERehd (ISCIII) to JMB, MJP, PMR, PA and LB); “Diputación Foral Gipuzkoa” (DFG15/010, DFG16/004 to JMB and 2020CIEN00006701 to PMR), Department of Health of the Basque Country (2019111024 to MJP, 2017111010 to JMB, and 2020111077 to JMB and PA), “Euskadi RIS3” (2016222001, 2017222014, 2018222029, 2019222054, 2020333010 to JMB), BIOEF (Basque Foundation for Innovation and Health Research: EiTB Maratoia BIO15/CA/016/BD to JMB) and Department of Industry of the Basque Country (Elkartek: KK2020/00008 to JMB); La Caixa Scientific Foundation (HR1700601 to JMB). “Fundación Científica de la Asociación Española Contra el Cáncer” (AECC Scientific Foundation, to JMB). AMMFThe Cholangiocarcinoma Charity (EU/2019/ AMMFt/001, to JMB and PMR). MRDG was funded by “Fundación Científica de la Asociación Española Contra el Cáncer” (AECC de Bizkaia), MJP was funded by the Spanish Ministry of Economy and Competitiveness (MINECO: “Ramón y Cajal” Program RYC201517755), IL, AL and FGR by the Basque Government (PRE_2016_1_0152, PRE_2018_2_0195 and PRE 2020 2 02500, respectively), ANZ and BGS by the UPV/EHU, ABV by “Programa de especialización de Personal Investigador Doctor” at the UPV/EHU (20192020) and MA by the MCIU/AEI/FEDER. The funding sources were not involved in the study design, data collection and analysis, decision to publish, or preparation of the article Approach and Results: The in vitro and in vivo tumorigenic capacity of five human CCA cell lines was analyzed. Proteome, lipid content, and metabolic fluxes were evaluated in CCA cells and compared with normal human cholangiocytes (NHC). The Akt1/NOTCH1 intracellular cytoplasmic domain (Nicd1)- driven CCA mouse model was also evaluated. The proteome of CCA cells was enriched in pathways involved in lipid and lipoprotein metabolism. The EGI1 CCA cell line presented the highest tumorigenic capacity. Metabolic studies in high (EGI1) versus low (HUCCT1) proliferative CCA cells in vitro showed that both EGI1 and HUCCT1 incorporated more fatty acids (FA) than NHC, leading to increased triglyceride storage, also observed in Akt1/Nicd1driven CCA mouse model. The highly proliferative EGI1 CCA cells showed greater uptake of verylowdensity and HDLs than NHC and HUCCT1 CCA cells and increased cholesteryl ester content. The FA oxidation (FAO) and related proteome enrichment were specifically upregulated in EGI1, and consequently, pharmacological blockade of FAO induced more pronounced inhibition of their tumorigenic capacity compared with HUCCT1. The expression of acylCoA dehydrogenase ACADM, the first enzyme involved in FAO, was increased in human CCA tissues and correlated with the proliferation marker PCNA. Conclusions: Highly proliferative human CCA cells rely on lipid and lipoprotein uptake to fuel FA catabolism, suggesting that inhibition of FAO and/ or lipid uptake could represent a therapeutic strategy for this CCA subclass. Downloaded from http://journals.lww.com/hep by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWn YQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/23/2025 | 1619 HEPATOLOGY determine risk factors, implement screening policies, and investigate accurate diagnostic methods and therapeutic strategies for this cancer. However, the high heterogeneity of CCAs extremely compromises the finding of common and effective treatments for all patients but opens a door to precisiontargeted therapies.[2] CCA subclasses differ not only on their anatomical location but also on their histological features,[3,4] risk factors,[5] putative cell of origin,[6] and mutational landscape.[7– 9] In this sense, two different molecular subclasses (i.e., proliferative or inflammatory) have been proposed for iCCA based on gene expression profiling.[7] A comparable genomic and molecular profiling for eCCA was also performed, proposing four biological subclasses for eCCA (i.e., proliferative, mesenchymal, metabolic, or immune).[8] Metabolic reprogramming is a hallmark of cancer, and targeting metabolism has been proposed for cancer therapy.[10] Cancer cells need a great amount of energy and biomaterials for the abnormal cell growth and division that characterizes the disease. Therefore, they adapt their mechanisms of nutrient uptake, energy production, and biosynthesis of complex molecules to suit these demands.[11] Genetic, epigenetic, and molecular alterations in tumor cells modulate their metabolic state. Moreover, differences in the microenvironment and the supply of nutrients and oxygen also contribute to the metabolic heterogeneity of the tumor.[11] The cancerspecific metabolic adaptability also contributes to the spread and survival to treatments.[11] Thus, there is a need to identify specific metabolic rewiring of each cancer type. Regarding lipid metabolism, cancer cells require higher amount of lipids mainly for the generation of new membranes and for energy production. Consequently, most cancer types are characterized by increased de novo lipogenesis through overexpression of lipogenic enzymes, such as fatty acid synthase (FASN) or acetylCoA carboxylase (ACC). However, they can also proliferate relying on extracellular lipid sources.[11] Little is known about the rewiring of lipid metabolism in CCA.[12,13] However, previous data showed a remarkable downregulation of FASN and ACC levels in human iCCA tumors compared with surrounding nontumorous tissue. Fasn expression was found also diminished in different mouse CCA models.[14– 16] Consistently, Fasn silencing did not abrogate CCA development in the Akt1/NOTCH1 intracellular cytoplasmic domain (Nicd1) and Akt1/Nras models.[16] Therefore, although still unclear, these data suggest that the uptake of exogenous lipids, rather than the de novo biosynthesis, should preferentially play a predominant role in CCA progression. The main aim of the present study was to gain further insights into the identification of the source, metabolic fate, and role of lipids in the proliferation of CCA cells. MATERIALS AND METHODS Animal models To promote the generation of iCCA in mice, 8week old male wildtype mice (mixed background C57BL/6J and 129/Sv) were subjected to hydrodynamic tail vein injection of plasmids, as described previously and detailed in Supporting Materials and Methods.[14,15] For the xenograft animal models, CCA cells were subcutaneously injected in flanks of immunodeficient CD1 nude mice as described in Supporting Materials and Methods. Etomoxir (30 mg/kg, MedChemExpress) was administered every 2 days for 32 days as described in Supporting Materials and Methods. The Ethics Committee of the University of the Basque Country (UPV/EHU) (CEEA M20/2019/60) or the Biodonostia Health Research Institute (CEEA2110/ OH21027) approved all the procedures. Human cohorts Gene expression microarray data from the Copenhagen cohort including 104 CCA surgical specimens,[17] and the Jusakul cohort including 118 CCA surgical specimens[18] were used for gene expression analysis and correlation. Besides, 23 pairs of matched tumor stroma and epithelium were obtained.[17] 182 tissue samples from human iCCAs collected at the Medical University of Regensburg (Regensburg, Germany) were used for the immunohistochemistry studies. Institutional Review Board approval was provided by the local Ethical Committee of the Medical University of Regensburg (approval # 171015101) in compliance with the Helsinki Declaration. Informed consent was obtained from all individuals. No donor organs were obtained from executed prisoners or other institutionalized persons. Cells Human CCA cell lines (EGI1, TFK1, WITT, HUCCT1 and TKKK) were used in the experiments, as well as primary cultures of normal human cholangiocytes (NHC; i.e., NHC2, NHC3, NHCSS, and NHC324). NHCs were isolated from normal liver tissue specimens as previously described.[19– 21] Additional information has been added in Supporting Materials and Methods. Metabolic fluxes Oleate and palmitate uptake The analysis of oleate and palmitate uptake was performed in vitro, in the EGI1 and HUCCT1 CCA cell lines, Downloaded from http://journals.lww.com/hep by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWn YQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/23/2025 1620 | CATABOLISM OF EXTRACELLULAR LIPIDS PROMOTES CCA PROGRESSION Downloaded from http://journals.lww.com/hep by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWn YQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/23/2025 | 1621 HEPATOLOGY and in the normal NHC3 cholangiocytes as control, as described previously[22] and detailed in Supporting Materials and Methods. Lipoprotein isolation, labeling, and uptake Lipoprotein uptake was determined in NHC3, EGI1, and HUCCT1 cells by fluorescent labeling and flow cytometry, as described in Supporting Materials and Methods. [3H]- oleate incorporation into lipids The analysis was performed ex vivo, in fresh mice liver pieces, and in vitro, in NHC3, EGI1, and HUCCT1 cells as previously described.[23] Lipids from liver or cell samples were extracted and separated. Additional information can be found in Supporting Materials and Methods. [14C]- palmitate oxidation The analysis of FA oxidation (FAO) rate was performed ex vivo, in fresh mice liver pieces, and in vitro, in NHC3, EGI1, and HUCCT1 cells, as described previously.[24,25] Additional information can be found in Supporting Materials and Methods. [3H]- oleate blood clearance and liver uptake The analysis was performed in vivo in the Akt1/Nicd1driven CCA mouse model using untreated animals as control. Animals were injected with 1 μCi [3H]- oleate. Blood samples were collected and liver uptake of [3H]- oleate was measured as detailed in Supporting Materials and Methods. Lipid quantification Lipids from liver samples and from NHC3, EGI1, and HUCCT1 cells were quantified as described in Supporting Materials and Methods. Statistical analysis Data are represented as mean ± SEM. Normal distribution assessments were carried out with the ShapiroWilk test. Differences between groups were analyzed with a twotailed Student’s t test, a Tukey’s multiple comparison test, Wilcoxon test, or with a twoway ANOVA test. Association between two variables was assessed by Pearson correlation coefficient or by Spearman test. Significance was defined as p < 0.05. Results were statistically analyzed using GraphPad Prism version 8.01 software (San Diego, CA, USA) and SPSS 22 software (IBM, Ehningen, Germany). RESULTS CCA cells are enriched in proteins involved in lipid and lipoprotein metabolism To assess the relevance of lipid metabolism in CCA, the proteomic profiles of 5 different human CCA cell lines (HUCCT1, TKKK, EGI1, TFK1, and WITT) and 4 primary cultures of NHCs (NHC2, NHC3, NHCSS, NHC324)[19– 21] were determined. The analysis of the proteome clearly differentiated CCA and NHC cells (Figure 1A,B and Figure S1A) and indicated that 2066 proteins were upregulated and 1148 downregulated in CCA compared with NHC. Enrichment analysis showed that the most represented biological processes related to those dysregulated proteins in CCA versus NHC cell cultures were “regulation of nucleic acid metabolism,” “metabolism,” and “energy pathways.” Among the 415 proteins inside the “metabolism” biological process, the most represented specific biological pathway was “metabolism of lipids and lipoproteins” (Figure 1C). Validation of proteins involved in different lipid metabolic pathways by immunoblotting showed that levels of the acylCoA synthetase long chain family member 5 (ACSL5), which participates in the activation of fatty acids (FAs) to acylCoA (Figure S2A,E), and levels of the FAbinding protein 5 (FABP5), relevant for the malignant progression of CCA[26] and involved in FA uptake, intracellular transport, and intracellular metabolism (Figure S2B,E), were increased in CCA cells when compared with NHC (Figure S2A,B). Regarding lipoprotein lipase (LPL) and CD36, the levels of both proteins decreased in CCA cell lines compared with most NHC, except in the EGI1 CCA cell line (Figure S2C– E). CCA cells display different proliferative and migration capacities Given the different origin and mutational profiles of the human CCA cell lines (HUCCT1, TKKK, EGI1, TFK1, and FIGURE 1 The profile of proteins involved in lipid metabolism is dysregulated in CCA. (A) Volcano plot showing differentially expressed proteins between CCA and NHCs. (B) Heatmap showing differentially expressed proteins between CCA and NHCs. (C) Enrichment analysis of biological processes was performed with differentially expressed proteins between NHC primary cultures and CCA cell lines (up) and with metabolismrelated, differentially expressed proteins between NHC and CCA cells. Only pathways with p < 0.05 are shown (hypergeometric test). The number of proteins categorized in each process/pathway is displayed next to the name. Downloaded from http://journals.lww.com/hep by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWn YQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/23/2025 1622 | CATABOLISM OF EXTRACELLULAR LIPIDS PROMOTES CCA PROGRESSION WITT),[27] we functionally characterized them by measuring the proliferation and migration rates in baseline conditions. EGI1 cells displayed the highest growth rate in a subcutaneous xenograft mouse model (Figure 2A), leading to increased tumor size (Figure 2A,B) and tumor weight (Figure 2B). A flow cytometrybased proliferation assay (carboxyfluorescein succinimidyl ester [CFSE]) also pinpointed EGI1 as the most proliferative CCA cell line in vitro (Figure 3A). We additionally performed a wound healing assay to compare the migration properties of the cells. Among the CCA cell lines, EGI1 cells displayed the highest migration capacity (Figure 3B), whereas no significant differences were found among the other cell lines (Figure 3B). The uptake of extracellular free FAs, and very lowand highdensity lipoproteins, is increased mostly in highly proliferative CCA cells The results mentioned above showed that EGI1 is the most proliferative cell line, with the highest migration and tumorigenic capacity, whereas the HUCCT1 cell line is among the less proliferative ones. Thus, taking into account the evidence suggesting a role of extracellular lipids on CCA progression,[16] the uptake of exogenous free FAs and lipoproteins was analyzed in these CCA cell lines and compared with NHC (i.e., NHC3). FIGURE 2 In vivo tumorigenic capacity of human CCA cell lines. (A) Tumor volume was measured 7, 14, 21, 28, 36, and 42 days after subcutaneous injection of EGI1, WITT, TFK1, HUCCT1, or TKKK CCA cells in the xenograft mouse model. Representative images of tumors in mice are shown below (n = 57). (B) Tumor weight was measured 42 days after injection of EGI1, WITT, TFK1, HUCCT1, or TKKK cells in the subcutaneous xenograft mouse model. Representative images of resected tumors (left) and quantification (right) are shown. Values are means ± SEM. Statistical analysis was determined by Tukey’s multiple comparison test and by twoway ANOVA test. Significant differences are denoted as **p < 0.01, ***p < 0.001 Downloaded from http://journals.lww.com/hep by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWn YQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/23/2025 | 1623 HEPATOLOGY Notably, the uptake of extracellular FAs (oleic acid and palmitic acid) was increased in both EGI1 and HUCCT1 cell lines compared with NHC3 cells (Figure 4A). Differences were also observed between tumor cell lines, as EGI1 cells presented higher uptake compared with HUCCT1 cells (Figure 4A). To study lipoprotein uptake, VLDLs, LDLs, and HDLs from serum samples were isolated, purified, and fluorescentlylabeled. The results showed that VLDL and HDL uptake was markedly increased in EGI1 compared with HUCCT1 or NHC3 cells, whereas LDL uptake was decreased significantly only in HUCCT1 cells compared with both EGI1 and NHC3 cells (Figure 4B). As cholesteryl ester (CE) is a major component of HDLs, we investigated whether the increased uptake of lipoproteins resulted in differences in the CE concentration in the highly proliferative and migrative EGI1 CCA cells. As expected, the CE content was higher in EGI1 cells than NHC3 or HUCCT1 cells (Figure 4C), closely resembling the observed changes in HDL uptake. Nevertheless, the increased CE levels in EGI1 cells were not linked to changes in free cholesterol (CL) levels (Figure S3A). The same profile was observed in the CCA lesions from Akt1/Nicd1 mice in which liver CE content was found elevated (Figure S3B), whereas CL remained unaltered (Figure S3B). Given that most circulating lipoproteins in mice are HDL[28] and that liver is a main organ involved in HDL uptake,[29] total serum CL and serum HDLCL levels were quantified. Accordingly, lower serum levels of total CL and of HDLCL were observed in Akt1/Nicd1 mice compared with control mice (Figure S3C), suggesting an increased liver uptake of serum HDLs in the cancer model. The endogenous CE synthesis from exogenous oleic acid (oleate) in CCA cell lines (Figure 4D) and in liver tumors from Akt1/Nicd1 mice (Figure S3D) remained unchanged compared with corresponding normal controls. The de novo synthesis of CL and CE also remained unaltered (Figure S3D), suggesting all together that the increased HDL uptake is a source of CEs in specific CCA cell subclasses. Rewiring of glycerolipid metabolism in CCA cells In hepatocytes, an increased uptake of lipids usually leads to increased storage of triglycerides (TGs). Given that the uptake of FAs and the TGrich lipoprotein VLDL was increased, mainly in the EGI1 cells, the glycerolipid content and the intracellular metabolic fluxes that regulate the esterification of oleate into complex lipids, as an indicator of glycerolipid synthesis, were analyzed. The results showed that the TG content was higher in both EGI1 and HUCCT1 CCA cells than in normal cholangiocytes (NHC3) (Figure 5A). Consistent with the increased FA uptake in EGI1 compared with HUCCT1 (Figure 4A), the fluxes that regulate the synthesis of TGs were also higher in the EGI1 CCA cell line (Figure 5A). We[30] and others[31] have previously demonstrated that phosphatidylcholine (PC) might be a source of TGs; thus, the PC content was also measured. The results showed that PC levels were decreased in both CCA cell lines (Figure 5B), whereas PC synthesis was unchanged (Figure 5B). The analysis of liver TG and PC content and the metabolic fluxes that regulate their synthesis in the Akt1/ Nicd1 CCA mouse model (Figure 6A,B) showed there was a faster clearance of circulating FAs in mice with iCCA lesions, compared with control mice (Figure 6C). In concordance, FA uptake by the liver was increased in the Akt1/Nicd1 mice compared with healthy mice (Figure 6D). According to the results obtained in the cell lines (Figure 5A), TG levels and TG synthesis were also increased in the mouse CCA lesions when compared with the control mice (Figure 6E). Metabolomic analysis revealed an evident increase in a high number of TG species (Figure 6E). This was linked again with decreased PC content (Figure 6F). Furthermore, in this in vivo iCCA model, liver PC species were decreased, whereas liver PC synthesis was found increased compared with their normal controls (Figure 6D), which suggests the activation of mechanisms involved in PC catabolism, such as phospholipases or SM synthases, that might contribute to the increased FA pool,[30,31] and/ or the generation of substrates for prostaglandin (PG) synthesis, altogether promoting proliferation. PC is required for the synthesis of sphingomyelin (SM) through the action of the SM synthase (SMS) using ceramide as a substrate. SM, at the same time, can be hydrolyzed by sphingomyelinases (SMases), yielding a ceramide and a phosphorylcholine (Figure S4A). Levels of ceramides were significantly decreased in both CCA cell lines compared with NHC3, with more prominent alterations in HUCCT1 cells (Figure S4B). On the contrary, SM levels were increased only in HUCCT1 compared with both NHC3 and EGI1 (Figure S4C). These changes in the lipidome were in concordance with the marked decrease in the acid SMase activity (the most abundant SMase in cholangiocytes[32] in both EGI1 and HUCCT1, and the increase in the SMS activity only in the HUCCT1 cell line (Figure S4D). The results here suggest that this remodeling in SM metabolism might not be the major metabolic pathway responsible for the observed decreased PC content in both EGI1 and HUCCT1 CCA cell lines (Figure 5B), pointing out the potential involvement of other phospholipases.[30,31] FAO promotes the hyperproliferation of specific CCA cell subclasses So far, the data showed that the increased uptake of extracellular free FAs, VLDLs and HDLs results in the accumulation of neutral lipids (i.e., TGs and CEs), likely leading to an intracellular pool of FAs that can be used Downloaded from http://journals.lww.com/hep by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWn YQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/23/2025 1624 | CATABOLISM OF EXTRACELLULAR LIPIDS PROMOTES CCA PROGRESSION as an energy source through FAO. These results are also supported by our data (Figure S5A,B) and previous reports[12– 16] that showed that de novo lipogenesis is not activated in CCA, thus confirming a more prominent role of FA uptake in promoting the accumulation of FAs. Downloaded from http://journals.lww.com/hep by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWn YQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/23/2025 | 1625 HEPATOLOGY To ascertain whether the catabolism of FAs was altered in the human CCA cell lines, complete oxidation of [14C]- palmitate into CO2 was measured. We found that the FAO rate was markedly increased in the EGI1 CCA cells as compared with the rest of the CCA cell lines or to the NHC (Figure 7A). Curiously, FAO rate decreased in HUCCT1 cells as compared with NHC3 cells suggesting other sources, different from lipids, as major energy substrates (Figure 7A). Thus, both glucose and glutamine uptake and complete oxidation into CO2 were measured. The results showed that only HUCCT1 presented a significant upregulation of glucose uptake compared with NHC3 (Figure S6A). However, complete glucose oxidation into CO2 did not change in the tumor cell lines (Figure S6B). Regarding glutamine, the results showed that its uptake remained unaltered in the CCA cell lines (Figure S6C), whereas the complete oxidation into CO2 was upregulated in HUCCT1 compared with NHC3 and EGI1, pointing glutamine as a relevant source to fuel the tricarboxylic acid (TCA) cycle and energy production in HUCCT1 CCA cells. Considering the differences in FAO rates between EGI1 and HUCCT1 cells come along with differential proliferation rates (Figure 3A), we decided to investigate the contribution of FAO to the proliferation capacity of EGI1 and HUCCT1 CCA cell lines. As predicted, etomoxir, a recognized inhibitor of FAO,[33– 35] effectively blocked FAO in both EGI1 and HUCCT1 cells, reaching similar oxidation values (Figure S7A). Noteworthy, FAO blockage with etomoxir induced a more pronounced inhibition of proliferation in the highly proliferative CCA cell line EGI1 than in the less proliferative HUCCT1 (Figure 7B), suggesting that EGI1 relies more on FAO to proliferate and grow. Similarly, in a subcutaneous xenograft mouse model, etomoxir administration decreased tumor growth in both EGI1 and HUCCT1 CCA cell lines (Figure 7C), more remarkably in the most proliferative one (EGI1). CCA cells with a stem celllike phenotype are more reliant on oxidative phosphorylation.[36] Accordingly, our results showed that the mRNA levels of the stemness markers EPCAM, ITGA6, CD133 and CD44 were markedly increased in the most proliferative, highly lipiddependent EGI1 cells compared with both NHC3 and HUCCT1 (Figure S7B). Thus, these results suggest that increased lipid uptake fuels the increased mitochondrial oxidation observed in the CCA stemlike cells. Lipophagy or autophagy of cellular lipid droplets can also have a role in cancer metabolic reprogramming.[37] Given the increased FAO in the most proliferative CCA cell line, protein levels of the autophagy markers ATG5 and ATG7, as well as LC3B lipidation, were mainly increased in the most proliferative EGI1 CCA cell line compared with both HUCCT1 and NHC3 (Figure S7C), which suggests increased autophagy fluxes in CCA cells, but particularly in the EGI1 CCA cell line. Next, we compared the proteomic profile of both CCA cell lines (EGI1 and HUCCT1) and 1409 proteins were differentially expressed. Among them, 1026 were upregulated in EGI1 compared with HUCCT1 and 383 in HUCCT1 compared with EGI1 (Figure S8). Enrichment analysis on biological pathways was performed with the 211 differentially expressed proteins categorized into the “metabolism” biological process. From those, 145 showed higher expression in EGI1 than in HUCCT1, and 66 in HUCCT1 than in EGI1. The enrichment analysis considering only pathways with a p value smaller than 0.0001 indicated that differentially expressed proteins were highly enriched in processes such as “metabolism of lipids and lipoproteins,” “mitochondrial FA βoxidation,” “the citric acid (TCA) cycle,” and “respiratory electron transport, ATP synthesis and heat production,” which usually upregulate when FAO rate increases (Figure 7D). Most of the proteins inside these pathways showed higher expression in EGI1 than in HUCCT1 (Figure 7D). As mentioned, EGI1 cells are more dependent on FAO for proliferation than HUCCT1, potentially explaining their higher proliferative capacity. Among the proteins involved in FAO showcasing increased levels in EGI1 when compared with HUCCT1, acylCoA dehydrogenase medium chain (ACADM), the first enzyme involved in FAO stood out. Validation by immunoblotting confirmed a higher expression of ACADM in CCA cell lines compared with NHC, particularly in EGI1 (Figure 7E). Accordingly, the immunostaining analysis of ACADM in 182 iCCA samples (Figure 8A) showed that the nontumor liver tissue presented a strong cytoplasmic ACADM staining in hepatocytes, whereas biliary epithelial cells exhibited faint or absent immunoreactivity for ACADM (Figure 8A, upper panels). On the contrary, 168 out of 182 (92.31%) iCCA samples showed robust immunostaining for ACADM in tumor (T) areas (Figure 8A, lower panels). In addition, ACADM expression was upregulated in the tumor epithelium compared with matched tumor stroma FIGURE 3 In vitro proliferative and migration capacity of human CCA cell lines. (A) In vitro proliferation rates of NHC3, EGI1, WITT, TFK1, HUCCT1, and TKKK CCA cell lines were determined at 48 h by flow cytometry measuring loss of fluorescence of CFSElabeled cells. Representative histograms (lightcolored histograms, at 24 h; darkcolored histograms, at 72 h) (left) and quantification (right) are shown. Results are expressed as relative to proliferation at 24 h of each cell line. (B) In vitro migration rates of NHC3, EGI1, WITT, TFK1, HUCCT1, and TKKK CCA cell lines were measured at 12 h by a wound healing assay. Results are expressed as relative to time 0. Representative images of wound healing areas (up) and quantification (down) are shown. Values are means ± SEM. Statistical analysis was determined by Tukey’s multiple comparison test. Significant differences are denoted as ***p < 0.001 Downloaded from http://journals.lww.com/hep by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWn YQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/23/2025 1632 | CATABOLISM OF EXTRACELLULAR LIPIDS PROMOTES CCA PROGRESSION according to their mutational and genetic expression profile.(7,8). A “proliferation class,” present in both iCCA and eCCA, has been defined. Our results indicate that highly proliferative CCA cells are characterized by higher lipid consumption, and treatments that affect the entry or consumption of lipids could have therapeutic value for CCA tumors with these characteristics. In summary, highly proliferative CCA cells are strongly lipiddependent, as shown by their upregulated lipid and lipoprotein uptake and catabolism for proliferation (Figure 8E). This observation underscores the potential relevance of targeting lipid metabolism for the treatment of specific subtypes of CCA. ACKNOWLEDGMENTS We thank Jose Antonio Lopez from the Department of Physiology, Faculty of Medicine and Nursing UPV/ EHU, for his help. This article is based on work from the COST Action CA18122 European Cholangiocarcinoma Network supported by COST (European Cooperation in Science and Technology) www.cost.eu. CONFLICT OF INTEREST Authors disclose no conflicts related to this study. AUTHOR CONTRIBUTIONS Mikel Ruiz de Gauna, Francesca Biancaniello, Jesús M. Banales, and Patricia Aspichueta designed the study. Mikel Ruiz de Gauna, Francesca Biancaniello, Francisco GonzálezRomero, Pedro M. Rodrigues, Ainhoa Lapitz, Sabina Di Matteo, Igor Aurrekoetxea, Ibone Labiano, Ane NievaZuluaga, Asier BenitoVicente, María J. Perugorria, Maider ApodakaBiguri, Diego Sáenz de Urturi, Beatriz GómezSantos, Xabier Buqué, Igotz Delgado, Mikel Azkargorta, Felix Elortza, Jesper B. Andersen, Luis Bujanda, Jesús M. Banales, and Patricia Aspichueta performed experiments and investigations. César Martín, Diego F. Calvisi, Jesper B. Andersen, Domenico Alvaro, Vincenzo Cardinale, Luis Bujanda, Jesús M. Banales, and Patricia Aspichueta designed experimental protocols. Mikel Ruiz de Gauna, Jesús M. Banales, and Patricia Aspichueta wrote the paper, and all authors contributed to editing. ORCID Jesper B. Andersen https://orcid.org/0000-0003-1760-5244 Patricia Aspichueta https://orcid.org/0000-0002-3553-1755 REFERENCES 1. Banales JM, Cardinale V, Carpino G, Marzioni M, Andersen JB, Invernizzi P, et al. Cholangiocarcinoma: current knowledge and future perspectives consensus statement from the European Network for the Study of Cholangiocarcinoma (ENSCCA). Nat Rev Gastroenterol Hepatol. 2016;13(5):261– 80. 2. Banales JM, Marin JJG, Lamarca A, Rodrigues PM, Khan SA, Roberts LR, et al. Cholangiocarcinoma 2020: the next horizon in mechanisms and management. Nat Rev Gastroenterol Hepatol. 2020;17(9):557– 88. 3. Nakanuma Y, Kakuda Y. Pathologic classification of cholangiocarcinoma: new concepts. Best Pract Res Clin Gastroenterol. 2015;29(2):277– 93. 4. Aishima S, Oda Y. Pathogenesis and classification of intrahepatic cholangiocarcinoma: different characters of perihilar large duct type versus peripheral small duct type. J Hepatobiliary Pancreat Sci. 2015;22(2):94– 100. 5. Clements O, Eliahoo J, Kim JU, TaylorRobinson SD, Khan SA. Risk factors for intrahepatic and extrahepatic cholangiocarcinoma: a systematic review and metaanalysis. J Hepatol. 2020;72(1):95– 103. 6. Cardinale V, Renzi A, Carpino G, Torrice A, Bragazzi MC, Giuliante F, et al. Profiles of cancer stem cell subpopulations in cholangiocarcinomas. Am J Pathol. 2015;185(6):1724– 39. 7. Sia D, Hoshida Y, Villanueva A, Roayaie S, Ferrer J, Tabak B, et al. Integrative molecular analysis of intrahepatic cholangiocarcinoma reveals 2 classes that have different outcomes. Gastroenterology. 2013;144(4):829– 40. 8. Montal R, Sia D, Montironi C, Leow WQ, EstebanFabró R, Pinyol R, et al. Molecular classification and therapeutic targets in extrahepatic cholangiocarcinoma. J Hepatol. 2020;73(2):315– 27. 9. Lowery MA, Ptashkin R, Jordan E, Berger MF, Zehir A, Capanu M, et al. Comprehensive molecular profiling of intrahepatic and extrahepatic cholangiocarcinomas: potential targets for intervention. Clin Cancer Res. 2018;24(17):4154– 61. 10. Hanahan D, Weinberg RA. Hallmarks of cancer: the next generation. Cell. 2011;144(5):646– 74. 11. DeBerardinis RJ, Chandel NS. Fundamentals of cancer metabolism. Sci Adv. 2016;2(5):e1600200. 12. Satriano L, Lewinska M, Rodrigues PM, Banales JM, Andersen JB. Metabolic rearrangements in primary liver cancers: cause and consequences. Nat Rev Gastroenterol Hepatol. 2019;16(12):748– 66. 13. Pastore M, Lori G, Gentilini A, Taddei ML, Di Maira G, Campani C, et al. Multifaceted aspects of metabolic plasticity in human cholangiocarcinoma: an overview of current perspectives. Cells. 2020;9(3):596. 14. Fan B, Malato Y, Calvisi DF, Naqvi S, Razumilava N, Ribback S, et al. Cholangiocarcinomas can originate from hepatocytes in mice. J Clin Invest. 2012;122(8):2911– 5. 15. Calvisi DF, Wang C, Ho C, Ladu S, Lee SA, Mattu S, et al. Increased lipogenesis, induced by AKTmTORC1RPS6 signaling, promotes development of human hepatocellular carcinoma. Gastroenterology. 2011;140(3):1071– 83. 16. Li L, Che L, Tharp KM, Park HM, Pilo MG, Cao D, et al. Differential requirement for de novo lipogenesis in cholangiocarcinoma and hepatocellular carcinoma of mice and humans. Hepatology. 2016;63(6):1900– 13. 17. Andersen JB, Spee B, Blechacz BR, Avital I, Komuta M, Barbour A, et al. Genomic and genetic characterization of cholangiocarcinoma identifies therapeutic targets for tyrosine kinase inhibitors. Gastroenterology. 2012;142(4):1021– 31. 18. Jusakul A, Cutcutache I, Yong CH, Lim JQ, Huang MN, Padmanabhan N, et al. Wholegenome and epigenomic landscapes of etiologically distinct subtypes of cholangiocarcinoma. Cancer Discov. 2017;7(10):1116– 35. 19. Banales JM, Sáez E, Úriz M, Sarvide S, Urribarri AD, Splinter P, et al. Upregulation of microRNA 506 leads to decreased Cl−/ HCO3− anion exchanger 2 expression in biliary epithelium of patients with primary biliary cirrhosis. Hepatology. 2012;56(2):687– 97. 20. Urribarri AD, MunozGarrido P, Perugorria MJ, Erice O, MerinoAzpitarte M, Arbelaiz A, et al. Inhibition of metalloprotease hyperactivity in cystic cholangiocytes halts the development of polycystic liver diseases. Gut. 2014;63(10):1658– 67. 21. MerinoAzpitarte M, Lozano E, Perugorria MJ, EsparzaBaquer A, Erice O, SantosLaso Á, et al. SOX17 regulates cholangiocyte differentiation and acts as a tumor suppressor in cholangiocarcinoma. J Hepatol. 2017;67(1):72– 83. Downloaded from http://journals.lww.com/hep by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWn YQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/23/2025 | 1633 HEPATOLOGY 22. Buqué X, Cano A, MiquilenaColina ME, GarcíaMonzón C, Ochoa B, Aspichueta P. High insulin levels are required for FAT/CD36 plasma membrane translocation and enhanced fatty acid uptake in obese Zucker rat hepatocytes. Am J Physiol Endocrinol Metab. 2012;303(4):E504– 14. 23. Aspichueta P, Pérez S, Ochoa B, Fresnedo O. Endotoxin promotes preferential periportal upregulation of VLDL secretion in the rat liver. J Lipid Res. 2005;46(5):1017– 26. 24. Huynh FK, Green MF, Koves TR, Hirschey MD. Measurement of fatty acid oxidation rates in animal tissues and cell lines. Methods Enzymol. 2014;542:391– 405. 25. GonzálezRomero F, Mestre D, Aurrekoetxea I, O'Rourke CJ, Andersen JB, Woodhoo A, et al. E2F1 and E2F2mediated repression of CPT2 establishes a lipidrich tumorpromoting environment. Cancer Res. 2021;81(11):2874– 87. 26. Nakagawa R, Hiep NC, Ouchi H, Sato Y, Harada K. Expression of fattyacidbinding protein 5 in intrahepatic and extrahepatic cholangiocarcinoma: the possibility of different energy metabolisms in anatomical location. Med Mol Morphol. 2020;53(1):42– 9. 27. Scherer D, Dávila López M, Goeppert B, Abrahamsson S, González Silos R, Nova I, et al. RNA sequencing of hepatobiliary cancer cell lines: data and applications to mutational and transcriptomic profiling. Cancers (Basel). 2020;12(9):2510. 28. Shiomi M, Koike T, Ishi T. Genetically modified animal models for lipoprotein research. In: Frank S, Kostner G, ed. Lipoproteins - Role in Health and Diseases. IntechOpen; 2012:533– 60. 29. Lewis GF, Rader DJ. New insights into the regulation of HDL metabolism and reverse cholesterol transport. Circ Res. 2005;96(12):1221– 32. 30. MartínezUña M, VarelaRey M, Cano A, FernándezAres L, Beraza N, Aurrekoetxea I, et al. Excess Sadenosylmethionine reroutes phosphatidylethanolamine towards phosphatidylcholine and triglyceride synthesis. Hepatology. 2013;58(4):1296– 305. 31. van der Veen JN, Lingrell S, Vance DE. The membrane lipid phosphatidylcholine is an unexpected source of triacylglycerol in the liver. J Biol Chem. 2012;287(28):23418– 26. 32. Uhlén M, Fagerberg L, Hallström BM, Lindskog C, Oksvold P, Mardinoglu A, et al. Tissuebased map of the human proteome. Science. 2015;347(6220):1260419. 33. Hernlund E, Ihrlund LS, Khan O, Ates YO, Linder S, Panaretakis T, et al. Potentiation of chemotherapeutic drugs by energy metabolism inhibitors 2deoxyglucose and etomoxir. Int J Cancer. 2008;123(2):476– 83. 34. Chen T, Wu G, Hu H, Wu C. Enhanced fatty acid oxidation mediated by CPT1C promotes gastric cancer progression. J Gastrointest Oncol. 2020;11(4):695– 707. 35. Xu AN, Wang B, Fu J, Qin W, Yu T, Yang Z, et al. Dietinduced hepatic steatosis activates Ras to promote hepatocarcinogenesis via CPT1α. Cancer Lett. 2019;442:40– 52. 36. Raggi C, Taddei ML, Sacco E, Navari N, Correnti M, Piombanti B, et al. Mitochondrial oxidative metabolism contributes to a cancer stem cell phenotype in cholangiocarcinoma. J Hepatol. 2021;74(6):1373– 85. 37. Maan M, Peters JM, Dutta M, Patterson AD. Lipid metabolism and lipophagy in cancer. Biochem Biophys Res Commun. 2018;504(3):582– 9. 38. Kreuzaler P, Panina Y, Segal J, Yuneva M. Adapt and conquer: metabolic flexibility in cancer growth, invasion and evasion. Mol Metab. 2020;33:83– 101. 39. Butler LM, Perone Y, Dehairs J, Lupien LE, de Laat V, Talebi A, et al. Lipids and cancer: emerging roles in pathogenesis, diagnosis and therapeutic intervention. Adv Drug Deliv Rev. 2020;159:245– 93. 40. Buranrat B, Senggunprai L, Prawan A, Kukongviriyapan V. Simvastatin and atorvastatin as inhibitors of proliferation and inducers of apoptosis in human cholangiocarcinoma cells. Life Sci. 2016;153:41– 9. 41. Liu Z, Alsaggaf R, McGlynn KA, Anderson LA, Tsai HT, Zhu B, et al. Statin use and reduced risk of biliary tract cancers in the UK Clinical Practice Research Datalink. Gut. 2019;68(8):1458– 64. 42. Halama A, Kulinski M, Dib SS, Zaghlool SB, Siveen KS, Iskandarani A, et al. Accelerated lipid catabolism and autophagy are cancer survival mechanisms under inhibited glutaminolysis. Cancer Lett. 2018;430:133– 47. 43. Zuurbier CJ, Bertrand L, Beauloye CR, Andreadou I, RuizMeana M, Jespersen NR, et al. Cardiac metabolism as a driver and therapeutic target of myocardial infarction. J Cell Mol Med. 2020;24(11):5937– 54. 44. Yizhak K, Gaude E, Le Dévédec S, Waldman YY, Stein GY, van de Water B, et al. Phenotypebased cellspecific metabolic modeling reveals metabolic liabilities of cancer. eLife. 2014;3:e03641. 45. Lupien LE, Bloch K, Dehairs J, Traphagen NA, Feng WW, Davis WL, et al. Endocytosis of very lowdensity lipoproteins: an unexpected mechanism for lipid acquisition by breast cancer cells. J Lipid Res. 2020;61(2):205– 18. 46. Lo PK, Yao Y, Zhou Q. Inhibition of LIPG phospholipase activity suppresses tumor formation of human basallike triplenegative breast cancer. Sci Rep. 2020;10(1):8911. 47. Seo J, Jeong DW, Park JW, Lee KW, Fukuda J, Chun YS. Fattyacidinduced FABP5/HIF1 reprograms lipid metabolism and enhances the proliferation of liver cancer cells. Commun Biol. 2020;3(1):638. 48. Li Z, Agellon LB, Allen TM, Umeda M, Jewell L, Mason A, et al. The ratio of phosphatidylcholine to phosphatidylethanolamine influences membrane integrity and steatohepatitis. Cell Metab. 2006;3(5):321– 31. 49. Urman JM, Herranz JM, Uriarte I, Rullán M, Oyón D, González B, et al. Pilot multiomic analysis of human bile from benign and malignant biliary strictures: a machinelearning approach. Cancers. 2020;12(6):1644. 50. Wu T. Cyclooxygenase2 and prostaglandin signaling in cholangiocarcinoma. Biochim Biophys Acta. 2005;1755(2):135– 50. SUPPORTING INFORMATION Additional supporting information may be found in the online version of the article at the publisher’s website. How to cite this article: Ruiz de Gauna M, Biancaniello F, GonzálezRomero F, Rodrigues PM, Lapitz A, GómezSantos B, et al. Cholangiocar cinoma progression depends on the uptake and metabolization of extracellular lipids. Hepatology. 2022;76:1617– 1633. https://doi. org/10.1002/hep.32344 Downloaded from http://journals.lww.com/hep by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWn YQp/IlQrHD3i3D0OdRyi7TvSFl4Cf3VC1y0abggQZXdtwnfKZBYtws= on 01/23/2025