Skip Navigation
Skip to contents

JLC : Journal of Liver Cancer

OPEN ACCESS
SEARCH
Search

Articles

Page Path
HOME > J Liver Cancer > Volume 26(2); 2026 > Article
Original Article
SINEUP-mediated enhancement of HNF4α improves metabolic function in Huh7-based bioengineered hepatic microtissues
Zohreh Hashemian1,2orcid, Morteza Zarrabi3orcid, Masoumeh Nouri3orcid, Abbas Piryaei4,5orcid, Ali Mohammad Moradi6,7orcid, Mohssen Nassiri Toosi6orcid, Sara Taleahmad8*orcid, Mustapha Najimi9,10*orcid, Massoud Vosough1,11*orcid
Journal of Liver Cancer 2026;26(2):298-316.
DOI: https://doi.org/10.17998/jlc.2026.05.13
Published online: August 21, 2026

1Department of Regenerative Medicine, Cell Science Research Center, Royan Institute for Stem Cell Biology and Technology, Academic Center for Education, Culture and Research, Tehran, Iran

2Department of Applied Cell Sciences, Faculty of Basic Sciences and Advanced Medical Technologies, Royan Institute, ACECR, Tehran, Iran

3Research and Development Department, Royan Stem Cell Technology Company, Tehran, Iran

4Department of Biology and Anatomical Sciences, School of Medicine, Shahid Beheshti University of Medical Sciences, Tehran, Iran

5Urogenital Stem Cell Research Center, Research Institute for Urology and Nephrology, Shahid Beheshti University of Medical Sciences, Tehran, Iran

6Liver Transplantation Research Center, Tehran University of Medical Sciences, Tehran, Iran

7Hepatobiliary Surgery and Liver Transplantation Division, Department of General Surgery, Imam Khomeini Hospital Complex, Tehran, Iran

8Department of Stem Cells and Developmental Biology, Cell Science Research Center, Royan Institute for Stem Cell Biology and Technology, ACECR, Tehran, Iran

9Vrije Universiteit Brussel, Liver Therapy & Evolution Team, In Vitro Toxicology and Dermato-Cosmetology (IVTD) Faculty of Medicine and Pharmacy, Brussels, Belgium

10Private University of Marrakesh, Marrakesh, Morocco

11Experimental Cancer Medicine, Institution for Laboratory Medicine, Karolinska Institute, Stockholm, Sweden

Corresponding author: Massoud Vosough, Department of Regenerative Medicine, Cell Science Research Center, Royan Institute for Stem Cell Biology and Technology, ACECR, Banihashem St., Resalat Ave., Tehran 1665659911, Iran, E-mail: masvos@yahoo.com
Corresponding author: Mustapha Najimi, Vrije Universiteit Brussel, Liver Therapy & Evolution Team, In Vitro Toxicology and Dermato-Cosmetology (IVTD) Faculty of Medicine and Pharmacy, Laarbeeklaan 103, Jette, Brussels 1090, Belgium, E-mail: Mustapha.najimi@vub.be
Corresponding author: Sara Taleahmad, Department of Stem Cells and Developmental Biology, Cell Science Research Center, Royan Institute for Stem Cell Biology and Technology, ACECR, Banihashem St., Resalat Ave., Tehran 1665659911, Iran, E-mail: sara_taleahmad@yahoo.com
* These three authors contributed equally to this work.
• Received: March 10, 2026   • Revised: April 29, 2026   • Accepted: May 13, 2026

Copyright © 2026 by The Korean Liver Cancer Association

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/4.0/) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 844 Views
  • 57 Downloads
prev
  • Background/Aims
    Human liver cell-based drug-screening platforms require stable hepatocyte identity and metabolic competence, but Huh7 cells show reduced hepatic function. We investigated whether SINEUP-mediated enhancement of hepatocyte nuclear factor 4 alpha (HNF4α) could improve the metabolic performance of Huh7-based three-dimensional hepatic microtissues.
  • Methods
    Decellularized liver extracellular matrix-derived microparticles were incorporated into three-dimensional microtissues containing Huh7 cells, human umbilical vein endothelial cells, and Wharton’s jelly mesenchymal stem cells using AggreWell technology. Experimental microtissues were generated with Huh7 cells expressing a SINEUP-based long noncoding RNA targeting HNF4α and were compared with control microtissues.
  • Results
    Enhanced HNF4α expression upregulated hepatic markers including HNF4α and albumin, decreased alpha-fetoprotein and CDH2, increased CDH1 expression, suppressed glycolytic genes, modulated lipid metabolism, and increased cytochrome P450-related gene expression. Albumin and fibrinogen secretion, urea synthesis, and glycogen storage increased, whereas alpha-fetoprotein secretion, lactate production, and cell migration decreased.
  • Conclusions
    SINEUP-mediated enhancement of HNF4α partially restored hepatocyte-like metabolic and functional properties in Huh7-based hepatic microtissues, supporting their potential use as an in vitro platform for drug discovery and toxicity screening.
In recent years, high-throughput drug screening platforms have mainly relied on human hepatocyte-based in vitro cell models. These models are capable of accurately simulating human liver physiology, particularly metabolic pathways involved in detoxification and drug clearance, lipid metabolism, glucose homeostasis, and cytochrome P450 (CYP) enzymes activities.1,2 Traditionally, two-dimensional (2D) monolayer hepatocyte cultures have been widely used due to their simplicity and compatibility with automated screening workflows. However, 2D systems are not reliable models for mimicking the complex structure of the native tissue’s three-dimensional (3D) microenvironment in culture. In 2D culture, CYP expression and functionality are reduced, leading to decreased metabolic performance. This metabolic shift diminishes the predictive power of these systems for assessing drug toxicity and efficacy.1,3,4 Therefore, scaffold-free or scaffold-based 3D platforms such as hepatic spheroids, organoids, and microtissues, have emerged as alternative modalities. These 3D models support cell-cell and cell-matrix interactions, oxygen and nutrient gradients, somewhat similar to liver tissue,5,6 and demonstrate enhanced accuracy in detecting drug-induced liver injury (DILI). This approach provides considerably better predictive performance compared to other conventional 2D monolayer culture systems.7
A major challenge in designing physiologically relevant liver drug screening models is the type of functional hepatocytes and their sources.8 Primary human hepatocytes (PHHs) are considered the gold standard cells in pharmaceutical research due to their intrinsic functionality and strong CYP activity. However, PHHs have certain limitations, including a lack of sufficient donors, rapid dedifferentiation in ex vivo culture conditions, limited proliferation capacity, donor dependent variations, and ethical/technical constraints.9 Hepatocyte-like cells (HLCs), derived from pluripotent stem cells, are available as innovative alternatives. However, their practical application faces major challenges such as high cost, incomplete maturation, insufficient CYP activity, and unstable hepatocytic phenotypes. These issues have limited their broad use in drug discovery.8,10
Cancer cell-derived hepatic cell lines provide an unlimited source of cells for large-scale drug screening. However, cancer cells exhibit profound metabolic differences compared to normal hepatocytes, including increased glycolysis (Warburg effect), dysregulated lipid metabolism, reduced gluconeogenic potential, and suppressed CYP activities in hepatoma cell lines.11-13 These differences limit their predictive value for drug metabolism and toxicology.
Given these challenges, restoring the metabolic functions of liver cancer cells could be an innovative option. Hepatocyte nuclear factor 4 alpha (HNF4α) is the key master regulator in hepatic differentiation, induction of genes involved in glucose and lipid metabolism, maintenance of the epithelial phenotype of cells, and boosting the detoxification capacity of hepatocytes.11,13 Downregulation of HNF4α expression is an important driver in hepatocarcinogenesis and a major obstacle to achieving metabolically functional hepatocytes in vitro.11 Therefore, restoring HNF4α levels might be an effective approach to switch poorly differentiated hepatocellular carcinoma cells toward well-differentiated ones with improved metabolic functions.14-16
SINEUP technology, which employs a class of long noncoding RNAs (lncRNA) that enhance the translation of specific target mRNAs, offers a unique opportunity to increase HNF4α production without any genetic manipulation. By increasing ribosomal recruitment to endogenous target mRNAs mediated by specific lncRNA secondary structures, SINEUPs can selectively augment protein synthesis in a controlled and physiologically relevant manner.1,17 Applying this innovative strategy to enhance HNF4α translation in liver cancer cells could reprogram the metabolic activity of liver cancer cells, promoting their differentiation and functions, and improving their utility for drug screening applications. 18
While HNF4α-mediated hepatocyte differentiation has been widely studied, most existing approaches rely on gene overexpression or genome editing strategies.19 In contrast, SINEUP technology enables selective enhancement of endogenous mRNA translation without altering the genomic sequence, offering a more controlled way to modulate protein expression. In addition, the application of this RNA-based approach in a multicellular 3D liver microenvironment has not been explored so far.20 Therefore, in this study, we set out to examine whether SINEUP-mediated upregulation of HNF4α could induce metabolic reprogramming in hepatoma-derived cells within a biomimetic 3D system. In this study, we generated 3D liver microtissues consisting of a carcinoma cell line (Huh7), human umbilical vein endothelial cells (HUVEC), and human Warton jelly mesenchymal stem cells (WJ-MSCs) along with microparticles derived from sheep liver extracellular matrix (LEM). In the experimental group, Huh7 cells were transformed using a specific lncRNA targeting HNF4α. Our aim was to evaluate whether increased HNF4α translation via SINEUP technology could reprogram cancerous hepatic microtissues toward a more metabolically functional phenotype. Subsequently, we performed transcriptional, proteomic, and functional assays on the modified microtissues, to determine whether this approach could enhance the metabolic capacity of the lab-made hepatic microtissues as an innovative and cost-effective candidate platform for the primary screening of drugs and toxicology studies.
Ethical approval statement
All experimental procedures in this study were conducted in accordance with the ethical guidelines set by the Ethics Committee of Royan Institute (approval No. IR.ACECR.ROYAN. REC.1402.131).
Plasmid preparation and cell line generation
The SINEUP-based antisense lncRNA (AS-lncRNA) HNF4α construct was kindly provided by Shokouhian et al.,18 as previously described in their publication. This construct contains the antisense binding domain (BD) corresponding to nucleotides −40 to +32 of the HNF4α mRNA, fused to a 10-nucleotide linker and cloned upstream of the effector domain (ED) region within the pDUAL/ED plasmid backbone. In our study, plasmid extraction was performed using a standard MiniPrep protocol (GeneJET Plasmid Miniprep Kit, K0502, Thermo Fisher Scientific, Waltham, MA, USA) prior to transfection.1
For stable transfection, Huh7 cells (CVCL_0336) were seeded in 6-well plates at a density of 500,000 cells per well and transfected with the AS-lncRNA HNF4α expression plasmid using X-tremeGENETM HP DNA Transfection Reagent (6366236001; Roche, Mannheim, Germany). The DNA-to-reagent was 3:2, according to the manufacturer’s protocol. After 24 hours, the culture medium (high-glucose DMEM [Dulbecco’s modified eagle medium] [11965092; Thermo Fisher Scientific] supplemented with 10% fetal bovine serum [FBS, 16140-071; Thermo Fisher Scientific] and 1× GlutaMAX [35050-061; Thermo Fisher Scientific]) was replaced with fresh medium. After 72 hours, selection was initiated by adding 700 μg/mL G418 (11811031; Thermo Fisher Scientific), which was previously determined to be the optimal selection dose. Cells were maintained under selection for approximately 7 days, with medium replacement every 48 hours, until all non-transfected cells were eliminated and a stable resistant population emerged. The resulting polyclonal stable pool was expanded and subsequently cryopreserved in a freezing medium consisting of 90% FBS and 10% dimethyl sulfoxide (D8418; Thermo Fisher Scientific) for long-term storage and future experiments.
To verify successful stable transfection, the expression level of the AS-lncRNA HNF4α was assessed in the G418-resistant Huh7 cell population. Total RNA was extracted from both transfected and non-transfected Huh7 cells using TRIzolTM Reagent (15596018 Invitrogen; Thermo Fisher Scientific), and complementary DNA (cDNA) was synthesized using a reverse transcription kit (A101162, Pars Tous, Mashhad, Iran) according to the manufacturer’s instructions. Quantitative polymerase chain reaction (qPCR) was performed to measure the expression of AS-lncRNA HNF4α using SYBR Green Master Mix (A3254022, Ampliqon, Odense, Denmark). Gene expression levels were normalized to the internal housekeeping gene β-actin (ACTB), and relative expression fold changes were calculated using 2−ΔΔCt method.
Preparation of hepatic microtissues
Fresh sheep liver tissue was cut into 2–3 mm pieces and washed several times with distilled water (DW) to remove blood residues. Tissue fragments were incubated in 1% sodium dodecyl sulfate (SDS, 436143; Sigma-Aldrich, St. Louis, MO, USA) at 4°C on a magnetic stirrer for 3 days with daily replacement of fresh 1% SDS solution.
Next, the tissues were rinsed thoroughly using DW and then incubated in 0.5 mM CaCl2 for 2 hours at 4°C under stirring to neutralize and remove residual SDS. The tissues were washed with DW and subsequently incubated in phosphate-buffered saline (PBS) for 2 days. The tissue samples were then treated with 1% Triton X-100 (108643; Merck, Burlington, MA, USA) for 30 minutes at 4°C, followed by washing in PBS for 2 hours and an additional 1 hour washing in DW. The decellularized LEM was frozen, lyophilized using a Christ lyophilizer (Alpha 1–2 LDplus; Sigma Laborzentrifugen GmbH, Osterode am Harz, Germany). Then, the LEM milled into a fine powder, and sterilized via ultraviolet irradiation.21,22
For histological evaluation, both native liver tissue and harvested LEM were dehydrated, embedded in paraffin, sectioned, and mounted on glass slides. Sections were stained with hematoxylin and eosin (H&E) to show decellularization and with Masson’s trichrome (MT) to evaluate the collagen content preservation. Comparative analysis was performed between native liver tissue and decellularized extracellular matrix (ECM).
Biochemical characterization was conducted to assess the LEM components. DNA content was measured using a Nano-Drop spectrophotometer (NanoDrop 2000; Thermo Scientific), following total DNA extraction. Furthermore, total collagen and glycosaminoglycan (GAG) content were assessed using colorimetric assay kits KHPO96 and KGAG96, respectively (Kiazist, Tehran, Iran) according to the manufacturer’s instructions. LEM powder was dissolved at a concentration of 20 mg/mL in 0.5 M acetic acid (64-19-7; Thermo Fisher Scientific) containing 1% pepsin (P6887; Merck), and the mixture was stirred at 4°C for 48 hours for complete enzymatic digestion. The final solution was neutralized on ice using 5 M NaOH under continuous stirring status.
For micro-particle (MP) formation, 1 mL of the neutralized LEM hydrogel was added on a dropwise manner into 80 mL of Corn oil supplemented with 100 μL Tween-80 (P1754; Thermo Fisher Scientific) as a surfactant. The mixture was homogenized at 16,000 rpm for 3 minutes. Crosslinking was performed by adding 5 mL of 2% glutaraldehyde (G5882; Thermo Fisher Scientific), followed by stirring at 0°C for 1 hour at 700 rpm, and then maintaining the mixture under continuous stirring at room temperature for 48 hours.
The emulsion was transferred to a 15 mL Falcon tube, centrifuged to remove the oil phase, and washed with 10 mL of 70% ethanol by vortexing and discarding the supernatant. The harvested MPs were subsequently filtered through a 40-μm filter using 70% ethanol and then centrifuged. To neutralize residual cross-linker, MPs were incubated in 10 mg/mL glycine (20 mL) for 30 minutes, centrifuged, washed with 70% ethanol, and then filtered through a 5-μm filter. Finally, the purified MPs were resuspended in deionized water and stored at 4°C.
Morphological assessment and size analysis of the obtained MPs were initially performed using light microscopy (IX71; Olympus, Tokyo, Japan), followed by high-resolution imaging using scanning electron microscopy (SEM, EVO 40; Zeiss, Oberkochen, Germany). Particle size distribution was analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA).
To evaluate the cytotoxicity of the MPs, Huh7 and human foreskin fibroblast (HFF) cells were exposed to various concentrations of MPs for 48 h, and cell viability was quantified based on MTS assay (MTS Cell Proliferation Assay Kit; Karmania Pars Gene, Tehran, Iran) and a microplate reader (51118650; Thermo Fisher Scientific).

Cell culture

Human WJ-MSCs were obtained at passage 5 from the Royan Stem Cell Bank (Tehran, Iran). Cells were cultured in high-glucose DMEM supplemented with 10% FBS and 2 mM Gluta-MAXTM and incubated at 37°C in a humidified incubator with 5% CO2.
HUVECs were isolated from fresh umbilical cords. The umbilical vein was flushed thoroughly with PBS, followed by enzymatic digestion using Collagenase Type IV (17104-019; Thermo Fisher Scientific) for 30 minutes at 37°C. The enzyme activity was neutralized with culture medium containing 10% FBS, and the suspension was centrifuged to collect the cell pellet. HUVECs were cultured in Royan Endothelial Growth Medium (REGM 0404; Royan Institute, Tehran, Iran) supplemented with 20% FBS and 1% GlutaMAXTM, 1% non-essential amino acids (NEAA, 11140035; Thermo Fisher Scientific), and 1% penicillin/streptomycin (Pen/Strep, 15070-063; Thermo Fisher Scientific). Cells at passages 3–5 were used for microtissue formation experiments.

Hepatic microtissues formation

Liver microtissues were generated by combining LEM-derived MPs with Huh7, HUVEC, and WJ-MSC cells. MPs were mixed with the cell suspension at a 2:1 cell-to-MP ratio, and the three cell types were incorporated at a physiological liver-mimicking ratio of 10:7:2 (Huh7:HUVEC:MSC). The resulting cell–MP mixture was gently homogenized and transferred into AggreWellTM 400 plates (34831; StemCell Technologies, Vancouver, Canada). Plates were centrifuged at 180 g for 3 minutes to promote uniform sedimentation of cells and MPs into the microwells and then incubated under optimized culture medium to facilitate controlled microtissue assembly. Two types of microtissues were generated: the experimental (EXP) group, formed using HNF4α-enhanced Huh7 cells, and the control (CTRL) group, formed using parental non-transfected Huh7 cells. Fully formed microtissues from both groups were subsequently used for viability evaluation, morphological characterization, molecular analyses, and functional assays.
Microtissues characterization

Cell viability assay

Cell viability within the microtissues was assessed using a Live/Dead Viability/Cytotoxicity Kit (L3224; Invitrogen, Carlsbad, CA, USA). Microtissues from both CTRL and EXP groups were collected following 48 hours post seeding. The microtissues were washed gently with PBS, and incubated with Calcein-AM (live cell indicator) and ethidium homodimer-1 (dead cell indicator) according to the manufacturer’s protocol. Samples were incubated at 37°C for 30 minutes in the dark, and then transferred onto glass-bottom imaging dishes. Live (green) and dead (red) cells were visualized and imaged using an inverted fluorescence microscope (IX71; Olympus) equipped with fluorescein isothiocyanate and Texas Red filter sets. Representative fields were captured for each condition, and viability was qualitatively compared between the EXP and CTRL microtissues.

Microtissues size distribution analysis

MT size and morphology were evaluated using phase-contrast microscopy. Fully formed microtissues from both groups were imaged at low magnification. The diameter of individual microtissues was measured using ImageJ software (ver. 1.2.4; SCR_003070). For each group, at least 100 microtissues were analyzed to obtain an average size distribution profile.
Gene expression profile analysis
Total RNA was extracted after 48 hours from CTRL and EXP microtissues using TRIzolTM reagent (15596018; Thermo Fisher Scientific), following the manufacturer’s instructions. RNA concentration and purity were assessed using a spectrophotometer. cDNA was synthesized from the purified RNA using a reverse transcription kit (A101162; Pars Tous) under optimized reaction conditions recommended by the supplier. In addition to microtissues, a human non-tumoral liver specimen was included as a physiological reference control. qRT-PCR was then performed to evaluate the expression pattern of genes involved in multiple functional categories, including differentiation markers (HNF4α, ALB, and AFP), adhesion/epithelial-mesenchymal transition (EMT) markers (CDH1 and CDH2), glycolytic enzymes (HK2, LDHA, and GAPDH), gluconeogenic regulators (G6PC, GCK, and PCK1), lipid metabolism (APOA1, APOB, SOAT1, and SOAT2), and some CYP enzymes (CYP3A4, CYP2C9, CYP2C19, CYP1A2, CYP2B6, and CYP7A1). qPCR reactions were carried out using a SYBR Green Master Mix on the StepOnePlusTM Real-time PCR System (Applied Biosystems, Foster City, CA, USA). All qPCR data were normalized to β-actin (ACTB) expression as the housekeeping control, and relative expression levels were calculated using the 2−ΔΔCt method, enabling direct comparison between EXP and CTRL microtissues.
Protein production analysis

Immunofluorescence staining

Protein production in CTRL and EXP microtissues were evaluated using immunofluorescence staining. microtissues were fixed in 4% paraformaldehyde (A11313.22; Thermo Fisher Scientific), washed with PBS, and permeabilized using 0.1% Triton X-100, followed by blocking with 3% bovine serum albumin (BSA) (9048-46-8; Thermo Fisher Scientific). Samples were incubated overnight at 4°C with primary antibodies against HNF4α (mouse monoclonal, ab41898; Abcam, Cambridge, UK) at dilution of 1:200, GLUT1 (rabbit polyclonal, ab115730; Abcam) at dilution of 1:200, GLUT2 (rabbit polyclonal, ab54460; Abcam) at dilution factor of 1:200, pH 9, N-cadherin (rabbit polyclonal, 13116S; BD Biosciences, Milpitas, CA, USA) at dilution factor of 1:400, pH 6, E-Cadherin (mouse monoclonal, sc-8426; Santa Cruz Biotechnology, Dallas, TX, USA) at dilution factor of 1:200, pH 6, and CYP3A4 (mouse monoclonal, sc-53850; Santa Cruz Biotechnology) at dilution factor of 1:200, pH 6.
After washing, samples were incubated for 1 hour at room temperature with relevant secondary antibodies Alexa Fluor 488 goat anti-mouse immunoglobulin G (IgG, Cat# A-11003; Invitrogen) 1:1,000 and Alexa Fluor 488 goat anti-rabbit IgG (Cat# A-11008/A-11035; Invitrogen) 1:1,000 in the dark. Nuclei were counterstained with DAPI, and microtissues were mounted on glass-bottom dishes. Micrographs were acquired using a fluorescence microscope (BX51; Olympus) equipped with appropriate filter sets.

Western blot analysis

Protein lysates were prepared after 72 hours from microtissues using RIPA buffer supplemented with protease inhibitors (A32963; Thermo Fisher Scientific). Equal amounts of proteins were separated via SDS-PAGE and transferred onto PVDF membranes. The membranes were blocked with 5% non-fat milk and incubated overnight at 4°C with primary antibodies against HNF4α (mouse monoclonal, ab41898; Abcam) at dilution of 1:1,000, GLUT1 (rabbit polyclonal, ab115730; Abcam) at dilution of 1:2,000, GLUT2 (rabbit polyclonal, ab54460; Abcam) at dilution factor of 1:1,000, N-Cadherin (rabbit polyclonal, 13116S; BD Biosciences) at dilution factor of 1:1,000, E-Cadherin (mouse monoclonal, SC8426; Santa Cruz Biotechnology) at dilution factor of 1:1,000, GAPDH (rabbit monoclonal, 2118L; GeneTex, Irvine, CA, USA) at dilution factor of 1:5,000, and β-actin (mouse monoclonal, IG60008-1; Invitrogen) at dilution factor of 1:2,000 which was used as the loading control.
After washing, membranes were incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies: goat anti-mouse IgG-HRP (32230; Invitrogen) at dilution factor of 1:10,000 or goat anti-rabbit IgG-HRP (A0545; Sigma-Aldrich) at dilution factor of 1:10,000. Protein bands were visualized using an enhanced chemiluminescence detection system (SuperSignalTM West Pico Plus, Thermo Fisher Scientific), and band intensities were semi-quantified using ImageJ. The densities were normalized to ACTB and compared between EXP and CTRL groups.
Functional assays

Evaluation of protein synthesis and secretion

To assess hepatocyte-specific metabolic activity, albumin (ALB) and fibrinogen (FB) secretion levels were quantified after 72 hours in the culture supernatants of CTRL and EXP microtissues using ELISA kits (E-EL-H6212 for ALB; Elabscience, Houston, TX, USA; EH3057 for FB; FineTest, Wuhan, China) according to the manufacturers’ instructions. Absorbance was measured with a microplate reader (51118650; Thermo Fisher Scientific), and concentrations were calculated based on standard curves.
Moreover, to evaluate the cancer-associated phenotype, alpha-fetoprotein (AFP) levels were measured in culture supernatants using an ELISA kit (PT-AFP-96; Pishtazteb, Tehran, Iran). Supernatants were collected at defined time points, centrifuged to remove debris, and analyzed according to the kit protocol.

Glucose metabolism

Glucose metabolic activity in CTRL and EXP microtissues was assessed after 72 hours by quantifying both glucose consumption and lactate production in the culture medium. The remaining glucose content in the conditioned media was measured using a biochemical glucose assay kit (BXC0101; Biorexfars, Shiraz, Iran), and glucose consumption was calculated relative to freshly prepared medium. Lactate levels in the supernatant were quantified using a biochemical lactate assay kit (BXC0622; Biorexfars). All values were normalized to the number of microtissues for a comparative assessment of glycolytic activity and metabolic reprogramming between EXP and CTRL groups.

Ureagenesis

To assess hepatic nitrogen metabolism, urea secretion was quantified in culture supernatants using a colorimetric urea assay kit (MBS2601488; MyBioSource, San Diego, CA, USA), according to the manufacturer’s instructions. Urea concentrations in the supernatant were calculated using standard curves.

Glycogen storage

Glycogen synthesis and storage within microtissues was assessed using the periodic acid-Schiff (PAS) staining method. Microtissues were fixed after 72 hours in 4% paraformaldehyde, embedded in paraffin blocks, sectioned, and stained with PAS. Glycogen-rich regions were visualized as bright magenta areas, and staining intensity was semi-qualitatively compared between CTRL and EXP microtissues by ImageJ software.

Migration assay

To evaluate collective migration and motility potential of the MT cells, fully formed microtissues were transferred onto tissue culture-treated adhesive plates. The microtissues were gently placed on the surface of plates. After 72 hours, MT spreading in the surrounding area, radial cell escape, and morphological changes were examined using phase-contrast microscopy (Olympus). The qualitative differences in area between the CTRL and EXP groups were evaluated using ImageJ software.
Drug treatment and enzymatic activity assays

Drug sensitivity assay

Microtissues obtained from the CTRL and EXP groups were used to assess drug response. For this purpose, microtissues were transferred to low-attachment plates and treated for 48 hours with acetaminophen (APAP, A7085; Sigma-Aldrich) 0, 1, 10, and 100 mM, isoniazid (INH, I3377; Sigma-Aldrich) 0, 100, 1,000, and 5,000 μM, or 5-fluorouracil (5-FU, F6627; Sigma-Aldrich) 0, 10, 100, and 500 μM. For each treatment condition, the same number of microtissues with a comparable size distribution was used. At the end of the exposure period, viability was measured using the MTS assay (MTS Cell Proliferation Assay Kit). Absorbance values were normalized to the untreated control within each group and expressed as percentages. Each condition was tested in at least three independent biological replicates.

CYPs activity assay

For evaluation of CYP450 inducibility, microtissues from the CTRL and EXP groups were pretreated with enzyme inducers for 72 hours. Phenobarbital (P5178; Sigma-Aldrich) 500 μM was used for induction of CYP3A4 and CYP2B6, whereas Rifampicin (R3501; Sigma-Aldrich) 25 μM) was used for CYP2C9, and Omeprazole (O104, Sigma-Aldrich, USA; 10 μM) was used for induction of CYP1A2. CYP450 activity was then assessed using luminogenic P450-Glo CYP450 Assay kits for CYP3A4 (V9002; Promega, Madison, WI, USA), CYP2B6 (V8321; Promega), CYP2C9 (V8791; Promega), and CYP1A2 (V8771; Promega) according to the manufacturer’s instructions. Microtissues were incubated with luciferin-PFBE for CYP3A4 (1:40 dilution), luciferin-2B6 for CYP2B6 (1:1,000 dilution), luciferin-H for CYP2C9 (1:50 dilution), and luciferin-1A2 for CYP1A2 (1:1,000 dilution) for 3 hours. At the end of the incubation period, medium samples were collected and luciferin metabolites were measured using a Synergy HTX Multi-Mode Microplate Reader (Synergy HTX; BioTek Instruments, Winooski, VT, USA). Enzyme activity was expressed as fold change relative to the basal activity of the vehicle control group.
Statistical analysis
All data are presented as mean±standard deviation. Comparisons among multiple groups at each time point were performed using one-way analysis of variance followed by Tukey’s post hoc test. For comparisons between two groups, an unpaired two-tailed Student’s t-test was used. A P<0.05 was considered statistically significant. Data analysis and graph preparations were performed using GraphPad Prism version 10 (GraphPad Software, San Diego, CA, USA). All experiments were conducted with at least three independent biological replicates. The exact sample size for each comparison was 3 (n=3).
Characterization of liver-derived extracellular matrix and micro-particles
The decellularization protocol effectively removed cellular components while preserving the overall ECM architecture (Supplementary Fig. 1A). H&E staining confirmed the complete removal of nuclei and cellular remnants, whereas Masson’s trichrome staining demonstrated preserved collagen in the resultant LEM (Fig. 1A). Biochemical analysis showed a significant reduction in DNA content in the LEM relative to native liver, while collagen and GAG levels were largely retained, indicating successful decellularization without substantial modifications to the ECM content (Fig. 1B).
MPs generated from digested and crosslinked LEM (Supplementary Fig. 1B) exhibited a uniform, predominantly spherical morphology under light microscopy (Fig. 1C). SEM micrographs further confirmed consistent surface topography and micro-scale size distribution. Quantitative ImageJ-based analysis showed that most MPs were within the desired particle size range of 3–7 μm (Fig. 1C–E). MTS cytotoxicity assays indicated that the MPs were not cytotoxic and were biocompatible, as no significant reduction in cell viability was observed in Huh7 or HFF cells incubated with different concentrations of MPs (Fig. 1E).
Validation of AS-lncRNA transfection and HNF4α-overexpressing Huh7 cells
Huh7 cells that were successfully stably transfected with the AS-lncRNA HNF4α construct were used to establish a resistant polyclonal cell pool after 7 days (Supplementary Fig. 2). qPCR analysis confirmed a significant increase in AS-lncRNA HNF4α expression in these transfected cells compared to intact Huh7 cells, verifying the successful establishment of an AS-lncRNA HNF4α-overexpressing Huh7 cell line (Fig. 2A, B).
Formation and assessment of liver microtissues
The co-culture of three different cell types successfully formed liver microtissues by combining Huh7, HUVEC, and WJ-MSCs at a 10:7:2 ratio with LEM-derived MPs at a 2:1 cell-to-MP ratio in AggreWellTM 400 plates (Fig. 2C). Both EXP and CTRL groups formed compact, spherical, and morphologically uniform microtissues following incubation under standard culture conditions (Fig. 2D). Quantitative image analysis showed that MT diameters fell within a relatively narrow and consistent size range, with comparable size distributions between the EXP and CTRL groups (Fig. 2E). Live/Dead staining revealed predominantly viable cells within microtissues from both groups, with minimal dead cells (Fig. 2F). These findings confirmed the overall bio-compatibility of the LEM-derived MPs and the suitability of the culture conditions for MT formation.
Transcriptional reprogramming of hepatic MT following HNF4α upregulation
qPCR analysis revealed distinct and functionally coherent transcriptional profiles between EXP and CTRL microtissues, as well as in comparison with the human non-tumoral liver reference.

Hepatocyte differentiation markers

Evaluating differentiation markers, EXP microtissues displayed an increase in transcripts for HNF4α (P<0.05) and a robust increase in ALB expression (P<0.01) compared to CTRL microtissues, whereas the cancer-associated marker AFP was markedly (P<0.001) reduced (Fig. 3A). The EXP expression levels of HNF4α and ALB approached those of the healthy liver specimen, while AFP levels remained substantially higher in CTRL microtissues, consistent with a more malignant state.

Adhesion and EMT-related gene expression

Adhesion and EMT-related genes showed a similar shift toward an epithelial, hepatocyte-like phenotype in the EXP group. Transcripts for CDH1 (E-cadherin) were elevated in EXP microtissues, whereas CDH2 (N-cadherin) was decreased relative to CTRL (both P<0.01) (Fig. 3B). This “cadherin switch” pattern in EXP microtissues resembled the adhesion profile observed in normal liver tissue and contrasted with the more mesenchymal/EMT-like signature of the CTRL group.

Glycolysis and gluconeogenesis-related genes

Metabolic reprogramming at the transcriptional level was also evident. Compared to the CTRL group, EXP microtissues exhibited significant downregulation of glycolysis genes such as HK2, LDHA (both P<0.01), and GAPDH (P<0.05) (Fig. 3C), along with considerable upregulation of key gluconeogenesis genes including G6PC (both P<0.0001), GCK (both P<0.01), and PCK1 (both P<0.0001) (Fig. 3D). This combination indicates a shift away from Warburg-like glycolytic dependence toward a more oxidative and gluconeogenic hepatocyte-like metabolism in the EXP microtissues. Compared to the healthy liver control, EXP microtissues showed a closer alignment of G6PC, GCK, and PCK1 expression to normal liver tissue.

Lipid metabolism-related gene expression

In the context of lipid metabolism, EXP microtissues demonstrated increased gene expression of APOA1 (P<0.05) and APOB (P<0.05), along with reductions in SOAT1 (P<0.01) and SOAT2 (P<0.05) expression levels compared to CTRL (Fig. 3E). These changes suggest a shift toward more physiological and normal hepatic function and reduced cholesteryl ester accumulation, again more similar to normal liver tissue.

CYP enzyme expression

Notably, transcripts for all examined CYP enzymes, including CYP3A4 (P<0.05), CYP2C9 (P<0.001), CYP2C19 (P<0.001), CYP1A2 (P<0.01), CYP2B6 (P<0.001), and CYP7A1 (P<0.01), were significantly elevated in EXP microtissues compared to CTRL (Fig. 3F). The transcriptional levels of CYP enzymes in the EXP microtissues approached, or in some cases approximated, those observed in the healthy liver sample, whereas CTRL microtissues exhibited markedly lower CYP expression. This broad upregulation of CYP genes indicates a partial restoration of hepatocyte-specific xenobiotic and bile acid metabolic functions in the EXP group (Fig. 3).
Protein detection by immunofluorescence and western blot
Qualitative assessment, using immunofluorescence staining, confirmed increased expression of HNF4α, CYP3A4, E-cadherin, and GLUT2 in the EXP microtissues compared to the CTRL group, whereas N-cadherin and GLUT1 signals were reduced (Fig. 4). These protein expression patterns closely mirrored the corresponding mRNA data and support a shift toward epithelial, hepatocyte-like, and metabolically normalized phenotypes in the EXP microtissues.
Western blot analysis further corroborated the immunofluorescence findings (Fig. 5A). In EXP microtissues, higher levels of HNF4α (P<0.001), E-cadherin (P<0.001), and GLUT2 (P<0.01) were observed, along with lower levels of N-cadherin (P<0.01), GLUT1 (P<0.0001), and GAPDH (P<0.001), compared to CTRL microtissues (Fig. 5B). These data provide additional evidence that AS-lncRNA HNF4α expression promotes hepatocyte-like reprogramming of liver cancer cells at the protein level.
Functional assessment of hepatic microtissues

Secretory function (ALB and AFP)

The results demonstrated that EXP microtissues secreted higher levels of ALB (P<0.05), as hepatic-specific secretory proteins, while there were no changes in FB production compared to CTRL microtissues. On the other hand, AFP secretion was markedly reduced in EXP microtissues compared to the CTRL, indicating attenuation of the cancer-associated secretory phenotype in response to HNF4α upregulation (Fig. 6A).

Metabolic function (glucose, urea)

Functional glucose metabolism assays showed reduced lactate secretion in EXP microtissues compared to CTRL group (P<0.05). These results suggest suppression of the “Warburg effect” and partial restoration of oxidative metabolic function in the EXP group (P<0.05). EXP microtissues produced higher levels of urea than CTRL microtissues, reflecting improved hepatic nitrogen metabolism and urea synthesis (Fig. 6B).

Migration and glycogen storage

The migration assay showed that CTRL microtissues demonstrated greater radial spreading of cells, and structural dissociation over 72 hours, whereas EXP microtissues largely maintained a compact spheroid morphology. This reduced migratory capacity in EXP microtissues is consistent with less aggressive, more differentiated phenotype of the liver cancer cells in the EXP group (Fig. 6C, D). PAS staining revealed more intense magenta staining in EXP microtissues compared to the CTRL group (P<0.001), indicating higher glycogen content and enhanced hepatocyte-like cell metabolic specialization in the EXP group (Fig. 6E, F).
Functional validation of drug sensitivity and CYP activity

Drug-response profiles

We treated microtissues from the CTRL and EXP groups for 48 hours with APAP, INH, and 5-FU. APAP reduced cell viability more in the EXP group than the CTRL group, especially at higher doses (10 and 100 mM). This shows the EXP group was more sensitive to APAP. Similarly, with INH, the EXP group had lower viability at higher doses (1,000 and 5,000 μM). However, when treated with 5-FU, the EXP group had higher cell viability than the CTRL group, indicating HNF4α may protect against its effects. These results show that HNF4α overexpression changes how the cells respond to different drugs. It makes them more sensitive to some drugs like APAP and INH but offers protection against 5-FU (Fig. 7A).

CYPs functional activity

We measured the activity of CYP3A4, CYP2B6, CYP2C9, and CYP1A2 enzymes in the CTRL and EXP groups. The EXP group showed significantly higher activity for all three enzymes compared to the CTRL group. This was most noticeable after Phenobarbital for CYP3A4 and CYP2B6, Rifampicin for CYP2C9, and Omeprazole for CYP1A2. These results suggest that HNF4α overexpression increases the cells’ ability to metabolize drugs. The increased enzyme activity in the EXP group shows that HNF4α overexpression improves the drug-metabolizing function of these cells (Fig.7B).
In drug discovery and toxicology studies, hepatocytes play a crucial role due to their substantial contribution to detoxification and drug metabolism. As mentioned, researchers do not have access to healthy and sufficiently functional primary human hepatocytes. One available alternative could be the use of liver cancer cell lines, which are available, scalable, and their culture is inexpensive. However, their main drawback is their poor metabolic performance compared to normal primary hepatocytes. In this study, we investigated the induction of the HNF4α in liver carcinoma cells in order to enhance their metabolic capacity. We demonstrated that SINEUP-mediated upregulation of HNF4α improved drug-metabolizing capacity, evidenced by enhanced activity of key CYP enzymes, such as CYP3A4, CYP2B6, CYP2C9, and CYP1A2.
In our study, both the EXP and CTRL groups were treated under the same conditions. The microtissues were uniform in size and presented similar viabilities, which were confirmed by live/dead staining. This ensures that any differences in drug response or metabolism are not due to variations in tissue size, quality or viability.
We established a liver cancer biomimetic MT model using Huh7 cells, HUVECs, and WJ-MSCs enriched with liver-derived extracellular matrix microparticles. We demonstrated that SINEUP-mediated upregulation of HNF4α, partially reduced the cancerous features of liver carcinoma cells and enhanced the liver-specific markers and functions of the EXP microtissues. By examining transcriptome, protein-level, and functional readouts, we provided converging evidence that AS-lncRNA HNF4α enhances liver physiologic function while reducing malignant features in cancerous hepatic microtissues.23
Co-culture of three different cell types in a 3D MT model enriched with tissue specific MPs derived from liver ECM is another strength of this study. Previous studies have shown that 3D culture, multicellular co-culture, and liver-derived ECM components can improve hepatocyte-like functions.24 Decellularized sheep liver ECM retained key structural components, including collagen and GAGs, while being effectively decellularized.22,25 The MPs fabricated from LEM were within a controlled microscale size range. They exhibited favorable morphology and showed no significant cytotoxicity after co-culture with Huh7 or HFF cells. These MPs were successfully incorporated into uniform microtissues when combined with Huh7, HUVECs, and WJ-MSCs at physiologically inspired ratios of 10:7:2 and cell: MP ratio of 2:1 assembled on AggreWellTM platforms. These microparticles supported the formation of compact and uniform microtissues with high viability in both EXP and CTRL groups.22,26,27 Moreover, treatment with APAP and INH revealed that the EXP group exhibited increased sensitivity to these drugs. The EXP group also offered protection against the effects of 5-FU. One of the key findings of this study is that using AS-lncRNA specific for HNF4α effectively increases HNF4α expression in Huh7 cells.28 This increase in HNF4α protein levels is accompanied by an upregulation of ALB expression and a downregulation of AFP at the mRNA and protein levels. It also slightly enhances its own expression in a positive feedback loop.28 HNF4α is an essential transcriptional regulator of hepatocyte differentiation and metabolic function. Its loss-of-function or suppression is a hallmark of hepatocellular carcinoma and the initiation of hepatocyte dedifferentiation.18 In our model, induction of HNF4α via specific AS-lncRNA resulted in broad transcriptional alterations aimed at partially restoring the hepatocyte phenotype in the liver carcinoma cell line Huh7. The increase in ALB and decrease in AFP in the EXP group suggest that SINEUP-mediated increased HNF4α expression can counteract key aspects of the malignant phenotype of the liver carcinoma cell line.29 The observed changes in adhesion molecules and EMT-related genes further support this interpretation. Our data showed a “cadherin switch” toward epithelialization, with increased expression of CDH1 (E-cadherin) and decreased expression of CDH2 (N-cadherin) in the EXP group compared to the CTRL group. This molecular pattern is consistent with reduced EMT markers, decreased invasiveness, and restored epithelial polarity. At the functional level, this was also confirmed by migration assays, in which EXP microtissues remained as compact spheroids with limited radial expansion, whereas CTRL microtissues showed significant expansion and migration. Taken together, these data demonstrate that AS-lncRNA HNF4α not only improved the metabolic features of cells but also stabilized epithelial characteristics and suppressed invasiveness of the liver carcinoma cells in 3D EXP microtissues.
Metabolic reprogramming is a key axis in therapeutic differentiation. Cancer cells, including HCC, typically exhibit increased glycolysis (Warburg effect) and impaired oxidative and gluconeogenic functions.30 In this study, EXP microtissues showed reduced transcription of the glycolysis genes such as HK2, LDHA, and GAPDH. This was also accompanied by increased expression of gluconeogenic related genes including G6PC, GCK, and PCK1. These findings are consistent with higher glucose utilization, reduced lactate secretion, and increased urea production in the EXP group compared to the CTRL group. Thus, they indicate a shift toward improved oxidative metabolic activity of hepatocytes. 31,32
In addition to the upregulation of key glycolysis and gluconeogenic genes and enzymes, our data also revealed a “GLUT switch.” In this regard, GLUT1, which is normally upregulated in cancer cells and is associated with aerobic glycolysis, was significantly downregulated. In contrast, GLUT2, the physiological glucose transporter in mature hepatocytes, was significantly upregulated in EXP microtissues. This switch is a well-known indicator of improving metabolic performance in liver carcinoma cells.33,34 Furthermore, glyceraldehyde 3-phosphate dehydrogenase (GAPDH), a central glycolytic enzyme often upregulated in malignant cells, was significantly downregulated in EXP microtissues after HNF4α induction. Overall, these changes suggest a partial shift from Warburg effect to more oxidative metabolic pathways in the manipulated hepatoma cells.
The main role of GAPDH is the glycolytic process. However, it is known to be a multifunctional protein involved in cancer progression. GAPDH can be indirectly involved in the regulation of EMT, migration, and redox balance.35 Therefore, a significant reduction of GAPDH at the gene expression and protein production levels in EXP microtissues not only reflects a decrease in glycolytic flux but may also contribute to the observed decrease in migratory ability and stabilization of epithelial features.
The combined reduction of HK2, LDHA and GAPDH is consistent with functional evidence of reduced lactate production and increased oxidative metabolism. Importantly, the expression levels of these gluconeogenic genes in EXP microtissues were adjusted to their counterparts in normal liver tissue. These findings suggest that restoring cytoplasmic HNF4α levels can partially induce metabolic reprogramming in liver cancer cells, shifting them toward a more physiological metabolic profile.35-37
Lipid metabolism and xenobiotic processing are critical components of liver function. Both were favorably modulated in EXP microtissues. Increased expressions of APOA1 and APOB, along with decreased expressions of SOAT1 and SOAT2, suggest improved regulation of lipoprotein related genes and a potential reduction in aberrant cholesterol ester storage, which is often associated with carcinogenesis and metabolic stress. Furthermore, the widespread and significant increase in the expression of CPY enzymes, including CYP3A4, CYP2C9, CYP2C19, CYP1A2, CYP2B6, and CYP7A1, in EXP microtissues strongly suggests the restoration of xenobiotic metabolism-related genes. The recovery of CYPs has important implications not only for modeling liver physiology and drug metabolism in vitro but also for developing any potential therapeutic strategies aimed at normalizing physiologic functions in cancerous liver cells.23,38-40
The gene expression data collectively demonstrate that the upregulation of AS-lncRNA HNF4α induces a coordinated transcriptional reprogramming of transfected Huh7 cells. This reprogramming enhances hepatocyte differentiation markers, normalizes adhesion and EMT signatures, suppresses glycolytic and cancer-associated programs, and restores lipid and CYP metabolic pathways. Importantly, the overall expression pattern of EXP microtissues converges toward that of human non-tumoral liver, supporting the notion that this strategy partially redirects cancerous hepatocytes toward a more hepatocyte-like metabolic state (Fig. 3).
The results of the CYP bioavailability and activity assays clearly demonstrated metabolic changes. The EXP group showed greater sensitivity to drugs such as APAP and INH, indicating improved metabolic function. On the other hand, the EXP group showed greater resistance to 5-FU, indicating that HNF4α may provide protection against the toxic effects of this drug.41,42
In the CYP activity assays, the EXP group showed increased activity of the enzymes CYP3A4, CYP2B6, CYP2C9, CYP1A2. This also indicated an increase in metabolic capacity in the EXP group. The increased enzyme activity, and the results of the drug sensitivity test, strongly suggest that increased HNF4α expression not only alters the response to the drug, but also improves the ability of the cells to metabolize drugs. These changes can be considered as an indication of improved liver function in the EXP group.22,43,44 Although EXP microtissues did not fully recapitulate all the features of normal liver, their gene expression patterns for differentiation markers, adhesion molecules, and metabolic/CYPs enzymes shifted toward the non-tumor reference. This gradient in the expression profile of cells-from normal liver tissue to EXP group and to CTRL group-supports the main hypothesis of this study: that upregulation of HNF4α via AS-lncRNA/SINEUP can induce liver carcinoma cells to adopt a more normal hepatocyte-like state in a 3D MT co-culture system.23,40,45,46 In addition to increased CYP expression, the concomitant changes in albumin secretion, urea production, glycogen storage, AFP reduction, glucose metabolism, and migratory capacity further support an overall functional improvement toward a hepatocyte-like phenotype. 47,48 Although the 3D MT model mimics several key aspects of liver function, it does not fully recapitulate the complexity of in vivo systems, including blood flow, oxygen gradients, and systemic interactions. Therefore, it should be considered as a simplified in vitro model. Nevertheless, as regulatory trends increasingly encourage the reduction of animal use in drug discovery and toxicology, this system may serve as a complementary platform for early-stage screening. By providing a more biomimetic microenvironment than conventional 2D cultures, it may improve the predictive value of in vitro assays and help bridge the gap between in vitro and in vivo models.7,49
A limitation of this study is the lack of direct comparison with established hepatic models such as primary human hepatocytes or HepaRG cells. While our results demonstrate a shift toward a hepatocyte-like phenotype, the relative performance of this system compared to these gold-standard models remains to be determined. Therefore, this model should currently be considered as a complementary and scalable system, and future studies will include direct benchmarking to further validate its physiological relevance.50
In this study, we induced a metabolic shift in hepatoma-derived cells by modulating HNF4α using SINEUP technology. While the role of HNF4α in hepatocyte differentiation is well established, the novelty of this work lies in the strategy used to regulate its expression. Unlike conventional approaches based on gene overexpression or genome editing, SINEUP technology enhances the translation of endogenous mRNA without altering the genomic sequence, allowing a more controlled modulation of protein levels. In addition, this approach is implemented within a multicellular 3D microtissue system incorporating endothelial and mesenchymal cells along with liver-derived ECM microparticles, providing a more biomimetic environment. Therefore, driving liver cancer cells metabolically toward a normal hepatocyte phenotype could serve as a promising alternative approach for human primary hepatocytes in drug screening and toxicology. Taken together, these findings suggest that this approach can partially reprogram hepatoma cells toward a hepatocyte-like phenotype. However, this reprogramming remains incomplete, and the engineered microtissues do not fully replicate the complexity and functionality of primary human hepatocytes. Therefore, the results should be interpreted as evidence of partial metabolic improvement rather than full restoration of hepatocyte function.
In summary, this study demonstrated that SINEUP-mediated upregulation of HNF4α expression in Huh7-based co-culture of liver microtissues induces a coordinated shift in differentiation, adhesion, metabolic, and functional properties toward a more normal hepatocyte-like state and reduces malignancy. Using liver ECM-derived microparticles for the enrichment of the hepatic microenvironment, we have generated a stable and reliable MT-based platform. In this model, AS-lncRNA HNF4α upregulates hepatocyte-specific markers, normalizes adhesion and EMT signatures, and suppresses glycolytic and cancer-related genes. It also suggesting partial functional improvement rather than full restoration gluconeogenesis, lipid metabolism, and CYP pathways. Additionally, it improves key liver functions such as ALB secretion, urea production, and glycogen storage, while reducing AFP secretion and migratory behavior. These findings suggest that targeted modulation of HNF4α translation via SINEUP technology, in combination with a physiologically inspired 3D MT system, could be a promising strategy for drug screening and toxicology; however, additional functional validation, particularly in drug metabolism and CYP activity assays, is required to confirm its applicability. This model offers a scalable, physiologically relevant alternative for drug screening. But its use in drug discovery should be viewed as a potential platform rather than a fully established tool at this stage.
Data will be made available on request.

The authors express their gratitude to their fellow colleagues at Royan Institute, department of Regenerative Medicine and Royan Cord Blood Bank for their valuable support and collaboration. Special thanks are extended to Ms. Maedeh Alizadeh for her kind assistance to this work.

Conflicts of Interest

The authors declare no conflicts of interest related to this research.

Ethics Statement

This study was approved by Ethics Committee of Royan Institute (Approval No. IR.ACECR.ROYAN.REC.1402.131).

Funding Statement

This research was supported by the Royan Institute (Grant No. 402000234).

Author Contributions

Conceptualization: ZH, ST, MN, MV

Data curation: ZH,

Funding acquisition: MV

Project administration: ST, MN, MV

Supervision: ST, MN, MV

Writing - original draft: ZH

Writing - review & editing: MZ, MN, AP, AMM, MNT, ST, MN, MV

Supplementary data can be found with this article online https://doi.org/10.17998/jlc.2026.05.13.
Figure 1
Characterization of liver-derived ECM and microparticles (MPs). (A) H&E and Masson’s trichrome staining images. (B) DNA, GAG, collagen quantification bar graphs. (C) SEM image of MPs. (D) MPs size distribution histogram. (E) MTS cytotoxicity assay. Additional preparation methods are provided in Supplementary Fig. 1. LEM, liver extracellular matrix; MT, microtissues; GAG, glycosaminoglycan; CTRL, control; ECM, extracellular matrix; H&E, hematoxylin and eosin; SEM, scanning electron microscopy.
jlc-2026-05-13f1.jpg
Figure 2
MT generation and viability assessment. (A) SINEUP transfection and selection workflow. (B) AS-lncRNA HNF4α qPCR expression. (C) AggreWell MT assembly scheme. (D) Light microscopy of CTRL vs. EXP microtissues. (E) Size distribution analysis. (F) Live/dead assay images. lncRNA, long noncoding RNA; HNF4α, nuclear factor 4 alpha; HUVEC, human umbilical vein endothelial cells; WJ-MSC, Wharton's jelly mesenchymal stem cells; MP, microparticle; CTRL, control; EXP, experimental; AS-lncRNA, antisense lncRNA; qPCR, quantitative polymerase chain reaction.
jlc-2026-05-13f2.jpg
Figure 3
Gene expression by qPCR analysis. (A) Differentiation markers. (B) Adhesion/EMT markers. (C) Glycolysis markers. (D) Gluconeogenesis markers. (E) Lipid metabolism markers. (F) CYP enzyme markers. EMT, epithelial-mesenchymal transition; CYP, cytochrome P450; qPCR, quantitative polymerase chain reaction.
jlc-2026-05-13f3.jpg
Figure 4
Immunofluorescence staining of protein markers in CTRL and EXP MTs. CTRL, control; EXP, experimental; MTs, microtissues.
jlc-2026-05-13f4.jpg
Figure 5
Protein expression analysis by Western blot. (A) Representative protein expression of HNF4α, GAPDH, E-cadherin, N-cadherin, GLUT1, GLUT2, and ß-ACT in CTRL and EXP MTs. (B) Densitometric quantification of protein bands normalized to ACTB. HNF4α, nuclear factor 4 alpha; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; CTRL, control; EXP, experimental; MTs, microtissues.
jlc-2026-05-13f5.jpg
Figure 6
Functional Assessment of Hepatic MTs. (A) Secretion of albumin (ALB), alpha-fetoprotein (AFP), and fibrinogen (FG) measured by ELISA. (B) Glucose consumption, lactate production, urea secretion in CTRL and EXP MTs. (C) Migration/escape assay images. (D) Quantification of migration area. (E) Periodic acid-Schiff (PAS) staining of glycogen in MTs. (F) Quantification of PAS staining intensity. CTRL, control; EXP, experimental; MTs, microtissues.
jlc-2026-05-13f6.jpg
Figure 7
Drug response and CYP functional activity. (A) Dose-dependent effects of APAP, INH, and 5-FU on the viability of CTRL and EXP MTs after 48 hours treatment. (B) Functional analysis of CYP3A4, CYP2C9, CYP2B6, and CYP1A2 in CTRL and EXP MTs. CYP3A4 and CYP2B6 activity were induced by phenobarbital, CYP2C9 activity was induced by rifampicin, and CYP1A2 activity was induced by omeprazole. Data are presented as normalized relative enzyme activities (n=3). *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001. CTRL, control; EXP, experimental; APAP, acetaminophen; INH, isoniazid; 5-FU, 5-fluorouracil; CYP, cytochrome P450.
jlc-2026-05-13f7.jpg
jlc-2026-05-13f8.jpg
  • 1. Hammond TG, Birdsall HH. Hepatocyte CYP2B6 can be expressed in cell culture systems by exerting physiological levels of shear: implications for ADME testing. J Toxicol 2017;2017:1907952. ArticlePubMedPMCPDF
  • 2. Dall M, Stocks B, Cervone DT, Deshmukh AS, Treebak JT. Hepatocyte dedifferentiation in 2D culture reveals extensive transcriptomic and proteomic rewiring. Hepatol Commun 2025;9:e0795. ArticlePubMedPMC
  • 3. Kozyra M, Johansson I, Nordling Å, Ullah S, Lauschke VM, Ingelman-Sundberg M. Human hepatic 3D spheroids as a model for steatosis and insulin resistance. Sci Rep 2018;8:14297. ArticlePubMedPMCPDF
  • 4. Wang T, Amorós MP, Porte C. Lipidome and gene expression profiling in zebrafish liver spheroids: a 3D model for environmental toxicology applications. Environ Sci Technol 2025;59:12543−12553.ArticlePubMedPMCPDF
  • 5. Messner S, Agarkova I, Moritz W, Kelm JM. Multi-cell type human liver microtissues for hepatotoxicity testing. Arch Toxicol 2013;87:209−213.ArticlePubMedPMC
  • 6. Leedale JA, Lucendo-Villarin B, Meseguer-Ripolles J, Kasarinaite A, Webb SD, Hay DC. Mathematical modelling of oxygen gradients in stem cell-derived liver tissue. PLoS One 2021;16:e0244070. ArticlePubMedPMC
  • 7. Yang S, Ooka M, Margolis RJ, Xia M. Liver three-dimensional cellular models for high-throughput chemical testing. Cell Rep Methods 2023;3:100432. ArticlePubMedPMC
  • 8. Hashemian Z, Taleahmad S, Shokouhian B, Najimi M, Vosough M. Functional liver cell-based platforms in biomedical research. Pharmacol Res Perspect 2025;13:e70128. ArticlePubMedPMC
  • 9. Niemeijer M, Więcek W, Fu S, Huppelschoten S, Bouwman P, Baze A, et al. Mapping interindividual variability of toxicodynamics using high-throughput transcriptomics and primary human hepatocytes from fifty donors. Environ Health Perspect 2024;132:37005. ArticlePubMedPMC
  • 10. Wang D, Villenave R, Stokar-Regenscheit N, Clevers H. Human organoids as 3D in vitro platforms for drug discovery: opportunities and challenges. Nat Rev Drug Discov 2026;25:204−226.ArticlePubMedPDF
  • 11. Hamdy H, Shen C, Xu J, Fan D, Zhang Y, Li H, et al. Hepatocyte nuclear factor 4-alpha: a key regulator in liver carcinogenesis. Cell Oncol (Dordr) 2025;48:885−897.ArticlePubMedPMCPDF
  • 12. Nelson LJ, Morgan K, Treskes P, Samuel K, Henderson CJ, LeBled C, et al. Human hepatic HepaRG cells maintain an organotypic phenotype with high intrinsic CYP450 activity/metabolism and significantly outperform standard HepG2/C3A cells for pharmaceutical and therapeutic applications. Basic Clin Pharmacol Toxicol 2017;120:30−37.PubMed
  • 13. Huck I, Gunewardena S, Espanol-Suner R, Willenbring H, Apte U. Hepatocyte nuclear factor 4 alpha activation is essential for termination of liver regeneration in mice. Hepatology 2019;70:666−681.ArticlePubMedPMCPDF
  • 14. Yin C, Xu WP, Dong WH, Ding CH, Cai JR, Zeng X, et al. Differentiation therapy with hepatocyte nuclear factor 4α for patients with hepatocellular carcinoma. Cell Res 2025;35:687−690.ArticlePubMedPMCPDF
  • 15. Miri-Lavasani Z, Torabi S, Solhi R, Shokouhian B, Afsharian P, Heydari Z, et al. Conjugated linoleic acid treatment attenuates cancerous features in hepatocellular carcinoma cells. Stem Cells Int 2022;2022:1850305. ArticlePubMedPMCPDF
  • 16. Yang T, Poenisch M, Khanal R, Hu Q, Dai Z, Li R, et al. Therapeutic HNF4A mRNA attenuates liver fibrosis in a preclinical model. J Hepatol 2021;75:1420−1433.ArticlePubMed
  • 17. Espinoza S, Bon C, Valentini P, Pierattini B, Matey AT, Damiani D, et al. SINEUPs: a novel toolbox for RNA therapeutics. Essays Biochem 2021;65:775−789.ArticlePubMedPMCPDF
  • 18. Shokouhian B, Negahdari B, Heydari Z, Totonchi M, Aboulkheyr Es H, Piryaei A, et al. HNF4α is possibly the missing link between epithelial-mesenchymal transition and Warburg effect during hepatocarcinogenesis. Cancer Sci 2023;114:1337−1352.PubMed
  • 19. Liu K, Guo MG, Lou XL, Li XY, Xu Y, Ji WD, et al. Hepatocyte nuclear factor 4α induces a tendency of differentiation and activation of rat hepatic stellate cells. World J Gastroenterol 2015;21:5856−5866.ArticlePubMedPMC
  • 20. Arnoldi M, Zarantonello G, Espinoza S, Gustincich S, Di Leva F, Biagioli M. Design and delivery of SINEUP: a new modular tool to increase protein translation. Methods Mol Biol 2022;2434:63−87.ArticlePubMedPMC
  • 21. Majidi F, Choshali MA, Izadi A, Hajari MA, Saheli M, Ziai SA, et al. Bio-fabrication of fibrotic liver microtissues as a scalable and drug-responsive in vitro model. Int J Pharm 2025;685:126242. ArticlePubMed
  • 22. Heydari Z, Moeinvaziri F, Agarwal T, Pooyan P, Shpichka A, Maiti TK, et al. Organoids: a novel modality in disease modeling. Biodes Manuf 2021;4:689−716.ArticlePubMedPMCPDF
  • 23. Radi SH, Vemuri K, Martinez-Lomeli J, Sladek FM. HNF4α isoforms: the fraternal twin master regulators of liver function. Front Endocrinol (Lausanne) 2023;14:1226173. ArticlePubMedPMC
  • 24. Saheli M, Sepantafar M, Pournasr B, Farzaneh Z, Vosough M, Piryaei A, et al. Three-dimensional liver-derived extracellular matrix hydrogel promotes liver organoids function. J Cell Biochem 2018;119:4320−4333.ArticlePubMedPDF
  • 25. Allu I, Sahi AK, Koppadi M, Gundu S, Sionkowska A. Decellularization techniques for tissue engineering: towards replicating native extracellular matrix architecture in liver regeneration. J Funct Biomater 2023;14:518. ArticlePubMedPMC
  • 26. Mansouri M, Imes WD, Roberts OS, Leipzig ND. Fabrication of oxygen-carrying microparticles functionalized with liver ECM-proteins to improve phenotypic three-dimensional in vitro liver assembly, function, and responses. Biotechnol Bioeng 2023;120:3025−3038.ArticlePubMed
  • 27. Zahmatkesh E, Ghanian MH, Zarkesh I, Farzaneh Z, Halvaei M, Heydari Z, et al. Tissue-specific microparticles improve organoid microenvironment for efficient maturation of pluripotent stem-cell-derived hepatocytes. Cells 2021;10:1274. ArticlePubMedPMC
  • 28. Kotulkar M, Robarts DR, Apte U. HNF4α in hepatocyte health and disease. Semin Liver Dis 2023;43:234−244.ArticlePubMedPMC
  • 29. Shokouhian B, Aboulkheyr Es H, Negahdari B, Tamimi A, Shahdoust M, Shpichka A, et al. Hepatogenesis and hepatocarcinogenesis: alignment of the main signaling pathways. J Cell Physiol 2022;237:3984−4000.ArticlePubMedPDF
  • 30. Yang F, Hilakivi-Clarke L, Shaha A, Wang Y, Wang X, Deng Y, et al. Metabolic reprogramming and its clinical implication for liver cancer. Hepatology 2023;78:1602−1624.ArticlePubMedPMC
  • 31. Xia H, Huang Z, Wang Z, Liu S, Zhao X, You J, et al. Glucometabolic reprogramming: from trigger to therapeutic target in hepatocellular carcinoma. Front Oncol 2022;12:953668. ArticlePubMedPMC
  • 32. Chen H, Wu Q, Peng L, Cao T, Deng ML, Liu YW, et al. Mechanism, clinical significance, and treatment strategy of warburg effect in hepatocellular carcinoma. J Nanomater 2021;2021:5164100. ArticlePDF
  • 33. Ruan Z, Wang Y, Shi L, Yang XJ. Progress of research on glucose transporter proteins in hepatocellular carcinoma. World J Hepatol 2025;17:104715. ArticlePubMedPMC
  • 34. Pliszka M, Szablewski L. Glucose transporters as a target for anticancer therapy. Cancers (Basel) 2021;13:4184. ArticlePubMedPMC
  • 35. Zhang K, Fang T, Shao Y, Wu Y. TGF-β-MTA1-SMAD7-SMAD3-SOX4-EZH2 signaling axis promotes viability, migration, invasion and EMT of hepatocellular carcinoma cells. Cancer Manag Res 2021;13:7087−7099.ArticlePubMedPMCPDF
  • 36. Xu X, Zhang Y, Wang X, Li S, Tang L. Substrate Stiffness drives epithelial to mesenchymal transition and proliferation through the NEAT1-Wnt/β-catenin pathway in liver cancer. Int J Mol Sci 2021;22:12066. ArticlePubMedPMC
  • 37. Wang Y, Fang X, Xie H, Wang X. GCNT3 promotes hepatocellular carcinoma progression and EMT by activating the PI3K/AKT pathway. Biochem Genet 2025;63:2423−2437.ArticlePubMedPDF
  • 38. Li C, Wang F, Cui L, Li S, Zhao J, Liao L. Association between abnormal lipid metabolism and tumor. Front Endocrinol (Lausanne) 2023;14:1134154. ArticlePubMedPMC
  • 39. Liu X, Lv M, Zhang W, Zhan Q. Dysregulation of cholesterol metabolism in cancer progression. Oncogene 2023;42:3289−3302.ArticlePubMedPDF
  • 40. Zhu T, Wang Z, Zou T, Xu L, Zhang S, Chen Y, et al. SOAT1 promotes gastric cancer lymph node metastasis through lipid synthesis. Front Pharmacol 2021;12:769647. ArticlePubMedPMC
  • 41. Zhao M, Ma J, Li M, Zhang Y, Jiang B, Zhao X, et al. Cytochrome p450 enzymes and drug metabolism in humans. Int J Mol Sci 2021;22:12808. ArticlePubMedPMC
  • 42. Beggs KM, McGreal SR, McCarthy A, Gunewardena S, Lampe JN, Lau C, et al. The role of hepatocyte nuclear factor 4-alpha in perfluorooctanoic acid- and perfluorooctanesulfonic acid-induced hepatocellular dysfunction. Toxicol Appl Pharmacol 2016;304:18−29.ArticlePubMedPMC
  • 43. Sahi J, Shord SS, Lindley C, Ferguson S, LeCluyse EL. Regulation of cytochrome P450 2C9 expression in primary cultures of human hepatocytes. J Biochem Mol Toxicol 2009;23:43−58.ArticlePubMed
  • 44. Li T, Chiang JY. Rifampicin induction of CYP3A4 requires pregnane X receptor cross talk with hepatocyte nuclear factor 4 alpha and coactivators, and suppression of small heterodimer partner gene expression. Drug Metab Dispos 2006;34:756−764.PubMedPMC
  • 45. Deans JR, Deol P, Titova N, Radi SH, Vuong LM, Evans JR, et al. HNF4α isoforms regulate the circadian balance between carbohydrate and lipid metabolism in the liver. Front Endocrinol (Lausanne) 2023;14:1266527. ArticlePubMedPMC
  • 46. Khamis ZI, Pang X, Cui Z, Sang QA, Zhang J. Cytochrome P450–2D6: a novel biomarker in liver cancer health disparity. PLoS One 2021;16:e0257072. ArticlePubMedPMC
  • 47. Prasajak P, Leeanansaksiri W. Developing a new two-step protocol to generate functional hepatocytes from Wharton’s jelly-derived mesenchymal stem cells under hypoxic condition. Stem Cells Int 2013;2013:762196. ArticlePubMedPMCPDF
  • 48. Chen C, Soto-Gutierrez A, Baptista PM, Spee B. Biotechnology challenges to in vitro maturation of hepatic stem cells. Gastroenterology 2018;154:1258−1272.ArticlePubMedPMC
  • 49. Serras AS, Rodrigues JS, Cipriano M, Rodrigues AV, Oliveira NG, Miranda JP. A critical perspective on 3D liver models for drug metabolism and toxicology studies. Front Cell Dev Biol 2021;9:626805. ArticlePubMedPMC
  • 50. Ramaiahgari SC, Waidyanatha S, Dixon D, DeVito MJ, Paules RS, Ferguson SS. From the cover: three-dimensional (3D) HepaRG spheroid model with physiologically relevant xenobiotic metabolism competence and hepatocyte functionality for liver toxicity screening. Toxicol Sci 2017;159:124−136.ArticlePubMedPMC

Figure & Data

References

    Citations

    Citations to this article as recorded by  

      • XML DownloadDownload Citation
        Download Citation
        Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

        Format:
        • RIS — For EndNote, ProCite, RefWorks, and most other reference management software
        • BibTeX — For JabRef, BibDesk, and other BibTeX-specific software
        Include:
        • Citation for the content below
        SINEUP-mediated enhancement of HNF4α improves metabolic function in Huh7-based bioengineered hepatic microtissues
        J Liver Cancer. 2026;26(2):298-316.   Published online August 21, 2026
        Close
      • XML DownloadXML Download
      SINEUP-mediated enhancement of HNF4α improves metabolic function in Huh7-based bioengineered hepatic microtissues
      Image Image Image Image Image Image Image Image
      Figure 1 Characterization of liver-derived ECM and microparticles (MPs). (A) H&E and Masson’s trichrome staining images. (B) DNA, GAG, collagen quantification bar graphs. (C) SEM image of MPs. (D) MPs size distribution histogram. (E) MTS cytotoxicity assay. Additional preparation methods are provided in Supplementary Fig. 1. LEM, liver extracellular matrix; MT, microtissues; GAG, glycosaminoglycan; CTRL, control; ECM, extracellular matrix; H&E, hematoxylin and eosin; SEM, scanning electron microscopy.
      Figure 2 MT generation and viability assessment. (A) SINEUP transfection and selection workflow. (B) AS-lncRNA HNF4α qPCR expression. (C) AggreWell MT assembly scheme. (D) Light microscopy of CTRL vs. EXP microtissues. (E) Size distribution analysis. (F) Live/dead assay images. lncRNA, long noncoding RNA; HNF4α, nuclear factor 4 alpha; HUVEC, human umbilical vein endothelial cells; WJ-MSC, Wharton's jelly mesenchymal stem cells; MP, microparticle; CTRL, control; EXP, experimental; AS-lncRNA, antisense lncRNA; qPCR, quantitative polymerase chain reaction.
      Figure 3 Gene expression by qPCR analysis. (A) Differentiation markers. (B) Adhesion/EMT markers. (C) Glycolysis markers. (D) Gluconeogenesis markers. (E) Lipid metabolism markers. (F) CYP enzyme markers. EMT, epithelial-mesenchymal transition; CYP, cytochrome P450; qPCR, quantitative polymerase chain reaction.
      Figure 4 Immunofluorescence staining of protein markers in CTRL and EXP MTs. CTRL, control; EXP, experimental; MTs, microtissues.
      Figure 5 Protein expression analysis by Western blot. (A) Representative protein expression of HNF4α, GAPDH, E-cadherin, N-cadherin, GLUT1, GLUT2, and ß-ACT in CTRL and EXP MTs. (B) Densitometric quantification of protein bands normalized to ACTB. HNF4α, nuclear factor 4 alpha; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; CTRL, control; EXP, experimental; MTs, microtissues.
      Figure 6 Functional Assessment of Hepatic MTs. (A) Secretion of albumin (ALB), alpha-fetoprotein (AFP), and fibrinogen (FG) measured by ELISA. (B) Glucose consumption, lactate production, urea secretion in CTRL and EXP MTs. (C) Migration/escape assay images. (D) Quantification of migration area. (E) Periodic acid-Schiff (PAS) staining of glycogen in MTs. (F) Quantification of PAS staining intensity. CTRL, control; EXP, experimental; MTs, microtissues.
      Figure 7 Drug response and CYP functional activity. (A) Dose-dependent effects of APAP, INH, and 5-FU on the viability of CTRL and EXP MTs after 48 hours treatment. (B) Functional analysis of CYP3A4, CYP2C9, CYP2B6, and CYP1A2 in CTRL and EXP MTs. CYP3A4 and CYP2B6 activity were induced by phenobarbital, CYP2C9 activity was induced by rifampicin, and CYP1A2 activity was induced by omeprazole. Data are presented as normalized relative enzyme activities (n=3). *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001. CTRL, control; EXP, experimental; APAP, acetaminophen; INH, isoniazid; 5-FU, 5-fluorouracil; CYP, cytochrome P450.
      Graphical abstract
      SINEUP-mediated enhancement of HNF4α improves metabolic function in Huh7-based bioengineered hepatic microtissues

      JLC : Journal of Liver Cancer
      TOP