Generation of Functional Endodermal Hepatic Organoids

This protocol describes the generation of fast and reproducible endodermal hepatic organoids (eHEPOs). With this protocol, eHEPOs can be produced within 2 weeks and expand long-term (more than 1 year) without losing their differentiation and functionality.

Organoid technology has allowed us to generate a variety of human organ-like mini structures, such as for the liver, brain, and intestine, in vitro. The remarkable advances in organoid models have recently opened a new experimental era for various applications in disease modeling, developmental biology, and drug discovery. Adult stem cells or induced pluripotent stem cell (iPSC)-derived liver organoids govern the generation of hepatocytes to use for diverse applications. Here, we present a robust and reproducible protocol for generating hepatic organoids from pluripotent stem cells. This protocol is applicable to healthy and patient-derived cells. To achieve 3D endoderm-derived hepatic organoids (eHEPOs), iPSCs were directly first differentiated into endodermal cells, and then FACS-enriched EpCAM-positive (EpCAM+) cells were used to establish hepatic organoids using the expansion medium. We provide a fast and efficient method to generate hepatic organoids within 2 weeks. The generated organoids mimic the essential properties and functions of hepatocytes, such as albumin secretion, glycogen storage, and cytochrome P450 enzyme activity. Besides the liver-specific gene expression similarities, eHEPOs comprise polarized epithelial cells with bile canaliculi in between. In addition, eHEPOs can be expanded and serial passages long term (1 year) without losing their capacity to differentiate into mature hepatocytes. Thus, eHEPOs provide an alternative source to produce functional hepatocytes.

Organoids are miniaturized organ-like structures grown in 3-dimensional culture conditions that mimic the organ microenvironment and include the intrinsic factors necessary for self-organization and self-renewal in organ development itself. Organoids can be derived either from pluripotent stem cells (PSCs) or adult tissue-derived cells (stem or progenitor cells)¹. Although their accurate organ-like organization and functional similarity to the specific organ make them valuable tools for disease modeling, they still need further improvements in terms of standardization in culture. In particular, several protocols have been published for the generation of liver organoids, and they differ in their complexity and reproducibility². For instance, the liver bud organoids developed by Takebe et al. take the form of dense, multi-cellular structures containing the following induced pluripotent stem cells (iPSCs): hepatic endodermal progenitors, human umbilical vein endothelial cells (HUVECs), and mesenchymal stem cells (MSCs). However, those organoids do not have long-term self-renewal capacity^3,4.

From a historical perspective, Huch et al. first reported the production of human hepatic epithelial organoids derived from adult tissue, in which the cells polarize and specialize to reproduce aspects of the native epithelium⁵. Then, Guan et al. used iPSC-derived hepatic organoids to model Alagille syndrome (ALGS), a rare genetic disorder associated with bile duct reduction within the liver⁶. Both of these organoids have self-renewal capacity and can gain mature hepatocyte functions, such as bile and albumin secretion, glycogen storage, and liver-specific drug detoxification. In a recent study, Ramli et al. introduced a PSC-derived liver organoid model containing functional bile canaliculi networks between polarized hepatocyte-like cells (HLCs) that empty cholestatic drugs into biliary cysts composed of cholangiocyte-like cells (CLCs)⁷.

This study presents a unique culture for generating iPSC-derived endodermal hepatic organoids, called eHEPOs. The iPSC culture and differentiation into endoderm are described step by step, and the generation of eHEPOs from enriched EpCAM+ progenitors is demonstrated. Finally, the characterization of the functionality and structural organization of the eHEPOs, as well as the cryopreservation of the organoids, are described.

Permissions related to experimental steps were obtained from the local Clinical Research Ethics Committee of Dokuz Eylul University Medical Faculty (2013/25-4, May 12, 2013; 2016/30-29, November 24, 2016).

1. Preparing solutions for cell culture

1. Prepare the endoderm differentiation medium by mixing the following components in a laminar flow hood: 1% B27, 100 ng/mL Activin A, Wnt3a or 30% wnt3a-CM, and R-spo1or 10% R-spo1-CM (see the Table of Materials).
2. Prepare the fluorescence-activated cell sorting (FACS) buffer with the following composition: PBS, 1 mM EDTA, 25 mM HEPES, 1% FBS.
3. Prepare the expansion medium (EM): 1% N2, 1% B27, 1.25 mM N-acetyl-L-cysteine, 50 ng/mL EGF, 10% R-spo1 conditioned medium, 100 ng/mL FGF 10, 10 mM nicotinamide, 5 µM A8301, and 10 µM forskolin (FSK) in Advanced DMEM/F-12. Add 10 nM Y-27632 and 0.3 nM WNT3a (see the Table of Materials).
   NOTE: Keep the complete medium for a maximum of 2 weeks at 2-8 °C.
4. Prepare the differentiation medium (DM): 1% N2, 1% B27, 10 nM gastrin, 50 ng/mL EGF, 100 ng/mL FGF 10, 25 ng/mL HGF, 500 nM A8301, 10 µM DAPT, 25 ng/mL BMP7, and 30 µM dexamethasone in Ad-DMEM/F-12 (see the Table of Materials).
   ​NOTE: Keep the complete medium for 2 weeks at 2-8 °C.

2. Thawing the iPSCs on the feeder-free plate

1. Coat the cell culture plate with basement membrane matrix (BMM).
   1. Thaw one vial of an aliquot of BMM overnight on ice or at least 2 h before starting the coating. Add 6 mL of cold DMEM/F-12 into a 15 mL conical tube, immediately add 80 µL of thawed BMM, and mix well.
   2. Use tissue culture-treated well plates, coat the culture plate with BMM solution, and dispense it to cover the well plate surface. Incubate the plate in a 37 °C incubator for at least 1 h before use. After the incubation time, discard the BMM solution, and immediately add the iPSC-specific medium.
2. Remove one vial of iPSCs from liquid nitrogen storage using cryogenic gloves. Immediately place the vial on a floating tube rack, and immerse the rack in a 37 °C water bath until the ice crystals have melted.
   1. Remove the vial from the water bath, and thoroughly wipe the outside of it with 70% ethanol. Then, place it in the laminar hood. It is important to be quick at this step.
   2. Under aseptic conditions, in a laminar hood, transfer the thawed cells into a sterile 15 mL conical tube containing 5 mL of iPSC-specific medium.
      1. Centrifuge the cells at 200 × g (at room temperature) for 3 min. Discard the supernatant slowly, resuspend the cells in 2 mL of iPSC-specific medium by pipetting very gently a few times with a 5 mL serological pipet, and seed the cells onto the coated wells.
   3. Move the plate quickly to disperse the cells throughout the well surface. Incubate the plate into the 37 °C, 5% CO₂ incubator.
   4. Refresh the culture medium every day.

3. Endoderm differentiation

1. Coat a new well plate with BMM 1 h before splitting the iPSCs, as mentioned in section 2.
2. Control the integrity and viability of the iPSCs under light microscopy. Undifferentiated iPSCs are cells with large nuclei, small cytoplasm, and compact cell colonies with distinct edges.
   NOTE: There are two enzymatic solutions for splitting cells. In the case of using dispase (1 U/mL), areas of differentiated cells must be removed. If a specific enzyme is used, there is no need to manually differentiate the regions with scraping or suction. Keep the specific enzyme at room temperature.
3. Warm the iPSC medium to room temperature. Aspirate the medium, and rinse the iPSC plate with DMEM/F-12 or D-PBS. Add 1 mL of the specific enzyme to one well of a 6-well plate, incubate for 1 min at room temperature, and then aspirate the specific enzyme, and incubate the plate at 37 °C for 5-7 min.
   NOTE: The optimal treatment time with enzymes should be optimized based on the iPSC line used.
4. After the incubation time, add 1 mL of pre-warmed iPSC medium, and gently tap the side of the plate to detach undifferentiated colonies aggregates from the surface of the plate. The optimal size of the aggregates is 50-200 µM.
   1. With a 5 mL serological pipette, gently mix the suspension, and then seed the cell suspension at the desired density onto the BMM-coated well. Incubate the cells at 37 °C for 3-5 days to reach a confluency of 70%.
5. Before starting the differentiation, replace the iPSC medium with the endoderm medium (step 1.1). Refresh the medium every day. Control the morphological changes of the cells during the differentiation steps. Cell-cell interactions begin to appear following differentiation, and the cells gain a spiky shape.

4. eHEPO establishment

1. Perform the enrichment of the EpCAM+ endodermal cells with FACS.
   1. Thaw one vial of organoid-specific BMM (see the Table of Materials) on ice before the cell sorting experiment. Pre-warm the tissue culture plates at 37 °C overnight before the cell sorting experiment.
   2. Rinse the endoderm cells with 1x PBS. Add 300 µL of trypsin, and incubate for 5 min in the incubator. Add 3 mL of cold DMEM-F12, and pipet with 1,000 µL filter tips to generate single cells.
   3. Collect the cell suspension in a 15 mL conical tube. Centrifuge the cells at 300 x g (at room temperature) for 5 min. Discard the supernatant, and wash the pellet with 10 mL of FACS buffer (step 1.2).
   4. To make a single-cell suspension, filter the cells firstly with a 100 µm mesh strainer and then a 40 µm mesh strainer.
      NOTE: Before using the strainer, rinse the surface of the strainer with 1 mL of FACS buffer.
   5. Centrifuge the cell suspension at 300 x g (at room temperature) for 5 min. Count the cell number with Trypan Blue⁸.
   6. The ratio for staining is 1:11. For example, resuspend 15 x 10⁶ cells with 105 µL of FACS buffer, and add 45 µL of FcR and 15 µL of EpCAM (see the Table of Materials). Incubate the cells for 10 min at 40 °C, and protect the tube from light. After the incubation, add 500 µL of FACS buffer to wash the cell suspension. Centrifuge the cells at 300 x g (at room temperature) for 5 min.
   7. Resuspend the pellet in 500 µL of FACS buffer. Add DAPI (0.5 µM/1 mL, final concentration) to the cell suspension to quantify the dead cells. Keep the cells on ice until starting the cell sorting step.
2. Embed the EpCAM and sorted cells into BMM. After the cell sorting, centrifuge the cells at 300 x g for 5 min to withdraw the FACS buffer. Remove the supernatant, and resuspend 3,000-5,000 cells in 20 µL of BMM.
   1. Seed the cells by adding a droplet of BMM to the center of each 48-well plate. Incubate the cells for 2 min at 37 °C until the BMM is solidified. Carefully revert the plate, and keep it in the incubator for another 10 min. Overlay the droplet with the expansion medium (250 µL per well for a 48-well plate).
   2. Refresh the organoid medium every 3 days. At days 10-11, the organoid size will be larger than 100 µm.
3. Perform eHEPO mechanical dissociation and differentiation.
   1. Split each organoid droplet in a 1:4 ratio. To split, discard the organoid medium and add 500 µL of cold Ad-DMEM/F12. With 1,000 µL filter tips, pipette the BMM droplets vigorously.
   2. Collect a maximum of 10 BMM droplets into 15 mL conical tubes, and add 5 mL of cold Ad-DMEM/F12. Keep the tube on ice for 3-5 min.
   3. Centrifuge the cells at 200 x g for 5 min at 4 °C. Discard the supernatant until 1 mL remains. Vigorously pipet the cell pellet 30-40 times using 200 µL filter tips. Check the organoid size under light microscopy, carefully avoiding the formation of a single-cell suspension.
   4. After mechanical dissociation, add 5 mL of cold Ad-DMEM/F12, and centrifuge the cells at 200 x g for 5 min at 4 °C.
   5. Discard the supernatant, and keep the suspension on the ice until seeding the cells.
   6. Seed a 50 µL BMM droplet (containing 35 µL of BMM + 15 µL of cells and Ad-DMEM/F-12 per well for a 24-well plate).
   7. Add BMP7 (25 ng/mL, see the Table of Materials) to the EM medium, and culture for 3 days. On day 4, replace the medium with the differentiation medium (step 1.4). Refresh the medium every 3 days, and culture the cells for at least 14 days.

5. Characterization of eHEPOs

1. Perform histological examinations.
   1. Perform paraffin embedding.
      1. Remove the medium, add 1-2 mL of cold 1x PBS, and gently resuspend the BMM in the cold 1x PBS. Carefully transfer the organoids to a 15 mL centrifuge tube. Allow the organoids to settle under gravity, and then carefully aspirate the PBS. Repeat the washing step three times.
      2. Add 10 mL of fresh fixative (10% neutral buffered formalin) to the organoids, and incubate for 30 min. Gently remove the fixative, add 10 mL of cold 1x PBS, and incubate for 10 min. Carefully aspirate the PBS.
      3. Stain the organoids in 0.5% eosin (see Table of Materials) solution dissolved in 96% ethanol for 30 min at room temperature.
         NOTE: If immunostaining is to be done, do not stain with eosin.
      4. Dehydrate the organoids by washing using an alcohol series (70% ethanol, 80% ethanol, and 96% ethanol) for 5 min each at room temperature.Soak the organoids in 100% ethanol (absolute ethanol) for 30 min.
      5. Clear the organoids using xylene (see the Table of Materials) three times for 20 min each time⁹.
         CAUTION: Xylene is a very toxic and flammable compound, so handle it using appropriate safety gear. Avoid working in large amounts.
      6. Gently transfer the organoids to a prewarmed metal base mold (see the Table of Materials) with molten paraffin using a plastic pipette. Hold the organoids in molten paraffin three times for 30 min each in a 65 °C incubator (replace it with new fresh melted paraffin at every change).
      7. Then, embed the organoids into new fresh paraffin. Move the organoids gently into the middle of the metal base mold.
         NOTE: For this step, it is crucial to work quickly because paraffin solidifies at room temperature.
      8. Transfer the metal base mold to the cold plate for 10-20 s to stabilize the organoids. Then, pour molten paraffin to cover the metal base mold, and add a tissue cassette (see the Table of Materials) on the top of the metal base mold for labeling. Ensure enough paraffin is in the mold, and then let them solidify on the cold plate. Remove the metal base mold.
      9. Cut 5-7 µm thick sections using a standard microtome (see the Table of Materials). Place serial paraffin sections on adhesive microscope slides. Let the slides dry overnight.
   2. Perform hematoxylin and eosin (H&E) and periodic acid-Shiff (PAS) staining.
      1. Put the paraffin-embedded slides in a 63 °C incubator overnight for deparaffinization. The next day, place the slides in xylene three times for 20 min each time. Do not allow the slides to dry out.
      2. Rehydrate the slides with a graded alcohol series (100% ethanol, 96% ethanol, 80% ethanol, and 70% ethanol) for 2 min. Then, place the slides into the distilled water for 1 min.
      3. For H&E staining, perform the following steps.
      4. Put the slides in hematoxylin solution for 5 min. Rinse the slides in running tap water for 5 min. Stain the slides with 1% eosin Y solution (see Table of Materials) for 10 s.
      5. Dehydrate the sections with a series of 80% ethanol, 96% ethanol, and two changes of 100% ethanol for 30 s each.
      6. Put the slides into the xylene for clearing three times for 20 min each time.
      7. Carefully mount coverslips onto the sections on the glass slides with a mounting medium (see the Table of Materials). Ensure there are no bubbles left.
      8. For the PAS staining, perform steps 5.1.2.4-5.1.2.6, followed by the procedures for PAS staining as reported by Akbari et al.¹⁰. Then carefully mount the coverslips onto the sections on the glass slides with the mounting medium. Ensure there are no bubbles left.
   3. Perform immunohistochemistry (IHC) staining.
      1. Perform step 5.1.2.1 and step 5.1.2.2 in the H&E staining protocol.
      2. Heat the slides in 0.1 M citrate buffer (pH 6.0) in the microwave for 10 min.
      3. Hold until the slides cool to room temperature (approximately 20 min), and rinse the slides with distilled water.
      4. Put the slides into freshly prepared 3% H₂O₂ in methanol for 10 min at room temperature. Then, rinse the slides in distilled water for 1 min.
      5. Block the non-specific binding of primer antibodies by incubating the sections with a blocking solution (2% serum + 0.1% Triton X-100 + 0.1% BSA) at room temperature for 1 h.
      6. Gently remove the blocking solution, and add the primary antibodies (E-CAD, ALB, EpCAM, A1AT, CK19) (see the Table of Materials) diluted in the blocking solution at 4 °C overnight.
      7. On the next day, rinse the slides carefully two times for 10 min (.
      8. Add the secondary antibodies (Goat Anti-Rabbit, Goat Anti-Mouse) (see the Table of Materials) for 1 h at room temperature, and rinse the slides in distilled water for 5 min (three times).
      9. Prepare 3'3 diaminobenzidine tetrahydrochloride (DAB, see the Table of Materials) solution (this must be prepared 30 min before use).
      10. Apply the DAB solution to the slides for 1-5 min, and rinse the slides in distilled water for 5 min (three times).
      11. Counter-stain the sections with hematoxylin for 1 min, and rinse the slides in running tap water for 1 min.
      12. Dehydrate the sections with a series of 80% ethanol, 96% ethanol, and two changes of 100% ethanol for 30 s each.
      13. Put the slides into xylene for 20 min (three times).
      14. Carefully mount the coverslip onto the section on the glass slide with a mounting medium. Ensure there are no bubbles left.
   4. Perform OCT embedding.
      1. Fix the organoids using freshly prepared cold 4% PFA overnight at 4 °C. Carefully remove the fixative, and wash with 1x PBS for 15 min (three times).
      2. Prepare a 30% sucrose solution (w/v) using distilled water. Remove the 1x PBS, and add the 30% sucrose to dehydrate the organoids.
      3. Incubate the organoids at 4 °C overnight. Initially, most tissues will float in 30% sucrose and sink once the permeation is complete. The cryoprotection may be considered complete at this point.
      4. Prepare dry ice or liquid nitrogen for the embedding process.
      5. Put two to three drops of OCT (see the Table of Materials) into plastic cryomolds. Place the organoids on top in the correct orientation for cutting.
      6. Slowly and carefully pour the OCT on top of the organoids. Be careful to avoid bubbles. Allow the organoids to settle under gravity for 10-15 min at room temperature.
      7. Label the cryomolds before the freezing step. Place the molds on dry ice or liquid nitrogen for rapid freezing. Transfer the cryomolds to −80 °C, and store them there until the sectioning process.
   5. Perform immunofluorescence (IF) staining.
      1. Perform frozen section staining.
         1. Air-dry the cryo-sections for 20-30 min at room temperature.Rehydrate the slides with distilled water for 5 min (three times). Surround the tissue with a hydrophobic barrier using a pap-pen (see the Table of Materials).
         2. Add 200-300 µL of 0.2 % Triton X-100 per section. Carefully wash the slides for 5 min (three times). Ensure that the sections do not fall off the slides during the washing steps.
         3. Add blocking solution (2% serum + 0.1% Triton X-100 + 0.1% BSA) for 1 h at room temperature.
         4. Gently remove the blocking solution, add the primary antibodies (E-CAD, A1AT, HNF4α) diluted in the blocking solution (see the Table of Materials), and incubate the slides overnight at 4 °C.On the next day, rinse the slides carefully for 10 min (two times).
         5. Add suitable secondary antibodies (Alexa Fluor 488 [Mouse], Alexa Fluor 594 [Mouse], Alexa Fluor 594 [Rabbit]) for 1 h at room temperature (see the Table of Materials), and rinse the slides in distilled water for 5 min (three times).
         6. Add 100-200 µL of Hoechst or DAPI nuclear dye per section for 3 min, wash the slides with 1x PBS, and repeat the wash step three times.
         7. Mount the coverslips onto the sections on glass slides with an anti-fade mounting medium.Keep the slides overnight at 4 °C, and protect them from light until the imaging process.
      2. Perform whole-mount staining.
         1. Perform step 5.1.1.1 and step 5.1.1.2 in the paraffin embedding procedure.
         2. Add blocking solution (0.1%-1% Triton X-100, 1% DMSO, 1% BSA, and 1% goat serum in 1x PBS, see the Table of Materials), and incubate the organoids with blocking solution for 2-3 h at room temperature.
         3. Gently remove the blocking solution, and add the primary antibodies (CK19, CK18, ZO-1, ALB) diluted in the blocking solution overnight at 4 °C (see the Table of Materials).
         4. On the next day, gently remove the solution, and add 1x PBS.Let the organoids settle under gravity for up to 10 min.Remove the solution carefully, and repeat the wash step three times.
         5. Incubate the organoids with the secondary antibodies (Alexa Fluor 594 [Mouse], Alexa Fluor 488 [Rabbit], Alexa Fluor 488 [Mouse]) for at least 1 h at room temperature.Remove the solution carefully, and repeat the wash step four times.
         6. For nuclear staining, incubate the organoids with Hoechst or DAPI nuclear dye for 10 min.Wash the organoids by adding 1x PBS, and repeat the wash step three times.Let the organoids settle to the bottom of the tube.
         7. Using a plastic Pasteur pipette, remove the organoids from the tube, and place them on glass slides. Ensure to do it with minimal damage to the organoids.
         8. Remove the excess PBS, and add an amount (one to two drops) of mounting medium.Carefully mount the coverslips onto the organoids on the glass slides with the mounting medium. Keep the slides at 4 °C, and protect them from light until the imaging process.
2. Perform electron microscopy.
   1. Fix the organoids using Karnovsky solution (2.5% buffered glutaraldehyde + 2% paraformaldehyde in 0.1 M sodium phosphate buffer [Sorensen's buffer], pH 7.4, see the Table of Materials) overnight at 4 °C.
   2. Wash the organoids with 0.1 M Sorenson's buffer + 0.1 M sucrose (1:1) for 15 min (three times). Then, postfix the organoids in 2% sodium phosphate-buffered osmium tetroxide (see the Table of Materials) for 90 min.
   3. Embed the organoids in 2% warm agar. Cut off the excess agar around the organoids using a lancet. Wash the organoids in Sorensen's buffer for 15 min (three times).
   4. Dehydrate using 50% acetone for 15 min (two times).
   5. Contrast overnight using 70% acetone + 1% phosphotungstic acid + 0.5% uranyl acetate (see the Table of Materials) at 4 °C.
   6. On the next day, dehydrate the organoids using an acetone series (80% acetone, 90% acetone, 96% acetone, and 100% acetone) for 15 min (two times).
   7. Clear the organoids using propylene oxide for 10 min (two times), and thenput the organoids in an epon:propylene oxide (see the Table of Materials) mixture for 30 min at room temperature.
   8. Add a new epon solution to the organoids, and incubate overnight at 4 °C.
   9. On the next day, put a new fresh epon solution into the base mold, and then embed the organoids.Put in an incubator for 48 h at 65 °C for polymerization.
   10. Cut the organoid epon blocks with an ultramicrotome set to a section thickness of 50-100 nm.Put the sections to formvar covered grids.
3. Measure the CYP3A4 enzyme activity.
   1. Measure the CYP activity using a commercially available kit (see the Table of Materials) following the manufacturer's instructions.
      NOTE: CYP enzymes convert the substrate to a detectable intermediate product^11,12,13. There are two options (cell lytic and nonlytic) for measuring the CYP enzyme activity. The kit provides an application to analyze the basal CYP activities and induced CYP activities after compound treatment.
   2. Run the experiment in triplicate.Read the luminescence with a luminometer, and calculate the signal. For normalizing, trypsinize the organoids, and count the number of cells.
4. Perform human albumin quantification.
   1. Replace the cell culture medium with a fresh medium 24 h before the ELISA test. After 24 h of incubation, collect the cell culture medium (300-500 µL), and centrifuge at 500 x g (at 4 °C) for 2 min to remove the cell debris and particles. Aliquot the supernatant, and store the samples at −80 °C (up to 3 months).
   2. Run the standard and sample in duplicate. Matured organoids secrete albumin into the culture medium; quantify this albumin with an ELISA quantification kit (see the Table of Materials). Measure the absorbance with an ELISA plate reader at 450 nm within 30 min after stopping the solution addition.Trypsinize the organoids, and count the number of cells to normalize.
5. Perform the ammonia elimination assay.
   1. Perform the ammonia elimination assay following the manufacturer's instructions (see the Table of Materials). Briefly, add 1.5 mM NH₄Cl to the culture medium, and incubate for 24 h.
   2. After incubation, determine the ammonia concentration with an ammonia assay kit (see the Table of Materials). Calculate the measurements using a multimode multi-plate reader, and normalize the values to the cell numbers.
6. Perform the cholylglycylamido-fluorescein (CLF) assay.
   1. Briefly, treat the organoids with 10 µmol/L CLF (see the Table of Materials), and incubate for 1 h in a 37 °C incubator. After the incubation, wash the organoids with 1x PBS. Perform imaging using fluorescence microscopy.
7. Perform gene expression analysis.
   1. Follow the RNeasy kit (see the Table of Materials) instructions to isolate total RNA from the eHEPOs.
   2. Set up the PCR reaction using a commercially available master mix (see the Table of Materials), following the manufacturer's instructions. After preparing the PCR mix for each primer set, place 10 µL of each sample into the 96-well qPCR plates (Table 1).
   3. Seal the plates with adhesive sealer, and then centrifuge the plates at 1,000 x g for 2 min at room temperature. Load the plates into a PCR machine. Normalize the relative gene expression to an internal control (RPL41, Table 1), and analyze by using the 2−ΔΔCt method ¹⁴.

6. Cryopreservation of the eHEPOs

1. Label cryovials with a nitrogen-resistant marker. Check the isopropanol level of the freezing container, and keep it at 4 °C for 15-30 min before starting the experiment.
2. Freeze one 50 µL BMM droplet in one cryovial. Proceed with organoid mechanic dissociation as described in step 4.3. After dissociation, resuspend the cell aggregate in 500 µL of cold freezing medium, then gently transfer the suspension to cryovials, and keep them in the freezing container.
3. Transfer the cell freezing container to −80 °C, and after 24 h, transfer the cryovials to liquid nitrogen for long-term storage.

7. Thawing of the eHEPOs

1. Pre-warm a water bath and Ad-DMEM/F12 to 37 °C.
2. Remove one vial of eHEPOs from the liquid nitrogen storage using cryogenic gloves. Place the vial in a 37 °C water bath until the ice crystals have melted.Transfer the cell suspension to a 15 mL conical tube, and centrifuge at 200 x g for 5 min at 4°C.
3. Aspirate the supernatant, and proceed to seeding the eHEPOs as described in section 4. Overlay the BMM droplet with the expansion medium supplemented with 10 µM Y-27632 (see the Table of Materials).Refresh the medium every 3 days, and remove the ROCK inhibitor after 3 days.

Firstly, human fibroblast cells or peripheral blood mononuclear cells (PBMC) cells were cultured and converted to iPSCs via episomal reprogramming. The fresh knockout serum was essential for obtaining healthy iPSCs. Then, the iPSCs were seeded into the BMM-coated culture plates with 50%-60% confluency. Having iPSC colonies of a small/medium size improved the differentiation efficiency. Then, the iPSCs were differentiated into definitive endoderm with medium containing Activin A, Wnt3a, and R-spo1 factors for 5 days. During the differentiation, the number and size of the cells increased significantly; therefore, paying attention to refreshing the medium is necessary. At the end of the endoderm differentiation, the EpCAM+ cells were enriched by fluorescence-activated cell sorting (FACS) method, and the sorted cells were embedded into BMM and grown for 10-14 days in the expansion medium (EM). The organoid formation capacity depends on the fresh medium components and the pure cell population; therefore, successful cell sorting is a critical step. When the size of the eHEPOs was larger than 100 µm and they had 70%-80% confluency, the cells were dissociated. eHEpOs have the capacity for long-term culture (1 year), and generated eHEPOs can be cryopreserved. Four organoid droplets with 70%-80% confluency are ideal for cryopreservation. The organoid splitting step before starting the maturation/differentiation is very critical. The viability of large organoids decreases during differentiation. In this work, the generated organoids were dissociated mechanically, re-embedded into BMM (1:4 ratio), and differentiated toward mature hepatocytes for 10-14 days in the differentiation medium (DM) (Figure 1).

An immunostaining analysis was performed to validate whether the organoids originating from EpCAM+ cells committed toward the hepatic lineage. The presence of CK19+ and HNF4ɑ+ cells indicates that hepatoblast and/or bile duct progenitor cells are in the organoids. Additionally, organoids expressing epithelial tissue markers such as E-CAD+ cells and cubical/polyhedral epithelial cells expressing ZO-1 indicate the presence of tight junctions between the cells. We observed organoids expressing ALB+, A1AT+, and CK18+, indicating the existence of mature hepatocytes^4,5,6,7 (Figure 2).

Histological analysis with hematoxylin-eosin staining showed that the organoids maintained a well-organized epithelial structure. E-CAD⁺ cells displayed a characteristic polygonal, epithelioid morphology resembling hepatocyte-like cells. The detection of CK19⁺ cells indicated the presence of hepatoblasts or cholangiocyte-like populations. In addition, ALB⁺ and A1AT⁺ cells confirmed the presence of more mature hepatocyte-like cells within the organoids^5,6 (Figure 3).

In the ultrastructure analysis, it was observed that intercellular tight junction complexes developed in the eHEPOs, and apical surface differentiations were observed. It was observed that bile canaliculi-like structures were formed between the intercellular junctional complexes. Bile canaliculi (BCs) were observed between the cell junctional complexes (Figure 4). Mitochondria (Mt), vacuolar inclusions (V), and collagen (Col) were abundant within the hepatocytes (Figure 4).

The qPCR results showed that the expression level of mature hepatocyte markers (ALB, A1AT, CYP3A4, and CYPA7)¹⁰ increased following differentiation, whereas the expression level of endoderm marker (EpCAM) and AFP decreased following differentiation. The RPL41 gene was used as a housekeeping gene. In this protocol, there is no need to remove the BMM during the sample collection for PCR experiments. Instead, the lysis buffer can be directly poured onto the eHEPOs. The macroscopic structure of the organoids was examined with H&E staining (Figure 5B, upper panel), which indicated structure-like-hepatocyte cords (remark) and a structure-like bile canaliculus. Glycogen storage analysis was done with a PAS stain. In Figure 5, the pink-stained areas represent glycogen storage in the cell cytoplasm (Figure 5B, lower panel). Moreover, the results showed that the eHEPOs gained mature hepatocyte functions, such as CYP enzyme activity and albumin secretion to the culture medium upon differentiation.

Figure 1: Schematic representation of endoderm-derived hepatic organoids (eHEPOs). This figure defines the timeline and critical steps during the organoid culture generation. Please click here to view a larger version of this figure.

Figure 2: Verification of the eHEPOs via immunofluorescence staining. Confocal images of whole-mount immunostaining of organoids for (A) CK19/DAPI, (C) DAPI/CK18, and (B) DAPI/ZO-1/ALB. (D,E,F) DAPI/E-CAD/A1AT, DAPI/HNF4ɑ/E-CAD, and DAPI/E-CAD were stained from frozen sections. DAPI was used to co-stain the nuclei. (A,C) The blue color in the nucleus marks the DAPI staining, the green color in the cytoskeleton marks the CK19 and CK18 staining, (B,D) the green color in the membrane shows ZO1 and E-CAD, and the green color in the nucleus shows HNF4ɑ staining. Scale bars: (A) upper panel: 50 µm, lower panel: 10 µm; (B) upper panel: 50 µm, lower panel: 20 µm; (C) upper panel: 50 µm, lower panel: 10 µm; (D) upper panel: 20 µm, lower panel: 20 µm; (E) 20 µm; (F) 50 µm. Please click here to view a larger version of this figure.

Figure 3: Characterization of the eHEPOs via immunohistochemical staining. Histological and immunohistochemical analysis. (A) Hematoxylin and eosin (H&E) staining of organoids, immunohistochemical staining of organoids for (B), (C) E-CAD, (D) ALB, (E) A1AT, and (F) CK19. Scale bars: (A) 200 µm; (B) 100 µm; (C,D,F) 50 µm; (E) 200 µm. Please click here to view a larger version of this figure.

Figure 4: Ultrastructural analysis of eHEPOs with electron microscopy. Transmission electron microscopy (TEM) images of eHEPOs. (A-C) The white circles indicate intercellular junctional complexes, the black arrows show apical surface differentiation (microvilli), and (B) the red arrow shows a transverse section of microvilli. Key: Mt = mitochondria; V = vacuolar inclusions; Col = collagen; BC =  Bile canaliculi; black arrow = microvilli; red arrow = transverse section of microvilli; red asterisks = apoptotic cell debris. Scale bars: (A) 2 µm; (B) 1 µm; (C) 200 nm. Please click here to view a larger version of this figure.

Figure 5: Functional analysis of the eHEPOs. (A) qPCR analysis of the eHEPOs. The data represent the average of at least three independent experiments. Student's t-tests were used for statistic analysis; (B) hematoxylin and eosin (H&E) and periodic acid-Shiff (PAS) staining of the eHEPOs in paraffin sections (black circles: like hepatocyte cords; black arrow: like bile canaliculus; red asterisks: glycogen storage area PAS [+]). Scale bars: top panels, 200 µm; bottom panels, 50 µm. (C) CYP3A4 enzyme activity in eHEPOs as shown in RLU/mL/million cells. (D) Albumin secretion levels as measured with ELISA. The data are expressed as ngALB/day/million cells. (E) Albumin secretion of organoids from different passage numbers (p10, p20, p30) cultured in DM conditions, as measured by ELISA. (F) Percentage of ammonia elimination capacity of EM and DM organoids. (G) Visualization of bile transport, which was analyzed with the cholyglylamido-fluorescein assay in control and CLF-treated organoids. Scale bar: 200 µm. The error bars show the standard deviation (SD), and significant differences between groups were considered at p < 0.05, ** p < 0.001, and *** p < 0.0001. Please click here to view a larger version of this figure.

Table 1: PCR primer sequences list.

The present protocol describes a comprehensive method for generating, expanding, and freezing/thawing hepatic organoids starting from iPSCs. This protocol covers all the steps, including culturing the iPSCs on the feeder and feeder-free culture, 2-dimensional endoderm differentiation, enrichment of the progenitor cells with FACS and organoid formation, and generating function-gaining organoids. Moreover, detailed instructions for validating and characterizing the organoids are also provided. The major obstacle to the wide application of mature and functional hepatocytes is the limited sources of primary hepatocytes, the inability to expand the hepatocytes long-term, and the cryopreservation of the cells^15,16. Combining iPSCs and organoid technology provides a powerful tool to generate unlimited functional hepatocytes.

In the first 5 days of the protocol, iPSCs were differentiated into endoderm cells using Activin A, Wnt3A, and R-spo1. R-spo1 acts as a Wnt pathway agonist protein and induces increased numbers of EpCAM+ cells¹⁰. Obtaining EpCAM+ cells of more than 60%-70% confluency indicates proper endoderm differentiation. To this aim, the size and cell confluency of the iPSCs affect the differentiation efficiency. The centers of large organoids are not differentiated very well, and starting with a high iPSC confluency of more than 60% negatively affects the cell viability. During cell differentiation, all published protocols utilize a defined method to direct the cell fates; however, the differentiation efficiency is not very high, and at the end of the path, heterogeneous cell populations emerge¹⁷. To reduce the heterogeneity within the organoid-initiating cell population, in this protocol, the EpCAM+ endodermal progenitor cells are enriched. It was found that only EpCAM+, rather than EPCAM−, cells had the potency to form 3D organoids. In addition, this method allows one to expand and passage organoids for more than 1 year in the culture. During long-term cell culture, the organoids preserve their functional properties. The eHEPO viability depends on the fresh medium components (especially R-spo1 and FGF10) and the cell splitting ratio (ideally 1:4 or 1:6). Additionally, in this work, the gene expression of mature hepatocyte markers was stronger in the DM condition in comparison with EM medium. After day 5 of differentiation, the cells started to express albumin; therefore, the eHEPOs can be used at an earlier step.

Histological and ultrastructural analyses have shown that organoids (eHEPOs) develop like mature liver tissue during embryogenesis. The formation of intercellular connections between the hepatocyte cells, the appearance of apical surface differentiation in the lumen (microvilli), and the development of bile canaliculi structures have been observed, similar to in the liver organogenesis process¹⁸. In terms of immunostaining in this protocol, staining was performed with two different techniques. To determine the eHEPO cell content, whole-mount staining was performed. In this staining protocol, the incubation times are longer than in the section staining. Indeed, longer incubation times with 3-dimensional structures allow one to improve the staining efficiency. For a more specific protein expression analysis, the staining should be done on cryosections or paraffin sections. Ultrastructural analysis is a more specific method for understanding the structure of organoids. In this method, the size of the organoids affects the incubation time. The incubation times can be shortened if the organoids are smaller than 200 µm. If the time is kept longer (especially acetone series or propylene oxide steps), the structure becomes more fragile and unstable. A long incubation time usually affects the size of the organoids and causes them to shrink. However, all the steps in this protocol are critical, and we want to emphasize the steps for the generation and sorting of the EpCAM+ cells.

Starting with pure population directly affects the organoid formation capacity. In addition, culturing the eHEPOs in the proper conditions, such as ensuring the appropriate cell confluency and performing the subculturing at the correct time, affects the long-term stability of the eHEPOs. eHEPOs are superior over other hepatic organoid models derived from iPSCs for the following reasons^19,20: (1) the organoids form in just 14 days, and during differentiation, there is no need for passaging the organoids, contrary to published protocols^6,19; (2) the organoids can be increased in culture in a proper way for over 1 year (Passage 48), whereas this culturing time for other protocols is 2 months. Like other iPSCs-derived hepatic organoid models, one of the disadvantages of this protocol relates to the expression of early liver development markers such as AFP. To improve the organoid formation and differentiation capacity, we are trying to discover the cell composition of eHEPOs using single-cell RNA sequencing resolution. Moreover, adding liver microenvironment clues to the culture medium will be beneficial for generating more mature eHEPOs.

In order to use eHEPOs for cellular therapies applications, one needs to replace the BMM with a human-compatible matrix. In this regard, tissue engineering approaches and scaffolds that bio-mimic the liver's scaffold could be utilized. Taking together, eHEPOs will be rather convenient in terms of personalized drug screening, preclinical hepatotoxicity analyses, and disease modeling.

The authors have nothing to disclose. The authors have no conflicts of interest to declare. Esra Erdal is co-founder of the ORGANO-ID Biotechnology company.

This research was supported by The Scientific and Technological Research Council of Turkey (TÜBİTAK) via projects SBAG-115S465 and SBAG-213S182. Figure 1 was generated using BioRender.