Login processing...

Trial ends in Request Full Access Tell Your Colleague About Jove


A High-Throughput Platform for Culture and 3D Imaging of Organoids

Published: October 14, 2022 doi: 10.3791/64405


This paper presents a fabrication protocol for a new type of culture substrate with hundreds of microcontainers per mm2, in which organoids can be cultured and observed using high-resolution microscopy. The cell seeding and immunostaining protocols are also detailed.


The characterization of a large number of three-dimensional (3D) organotypic cultures (organoids) at different resolution scales is currently limited by standard imaging approaches. This protocol describes a way to prepare microfabricated organoid culture chips, which enable multiscale, 3D live imaging on a user-friendly instrument requiring minimal manipulations and capable of up to 300 organoids/h imaging throughput. These culture chips are compatible with both air and immersion objectives (air, water, oil, and silicone) and a wide range of common microscopes (e.g., spinning disk, point scanner confocal, wide field, and brightfield). Moreover, they can be used with light-sheet modalities such as the single-objective, single-plane illumination microscopy (SPIM) technology (soSPIM).

The protocol described here gives detailed steps for the preparation of the microfabricated culture chips and the culture and staining of organoids. Only a short length of time is required to become familiar with, and consumables and equipment can be easily found in normal biolabs. Here, the 3D imaging capabilities will be demonstrated only with commercial standard microscopes (e.g., spinning disk for 3D reconstruction and wide field microscopy for routine monitoring).


In organotypic 3D cell cultures, hereafter referred to as organoids, stem cells differentiate and self-organize into spatial structures that share strong morphological and functional similarities with real organs. Organoids offer valuable models to study human biology and development outside the body1,2,3. A growing number of models are being developed that mimic the liver, brain, kidney, lung, and many other organs2,4,5. Differentiation in organoids is directed by the addition of soluble growth factors and an extracellular matrix in a precise time sequence. However, in marked contrast to organs, the development of organoids is quite heterogeneous.

Beyond numerous biological challenges6,7, organoid cultures also pose technological challenges in terms of cell culture methods, characterization of transcriptomics, and imaging. In vivo organ development occurs in a biological environment that results in a highly stereotypical self-organization of cell arrangements. Any phenotypic alteration can be used as a proxy to diagnose a diseased state. In contrast, organoids develop in vitro in minimally controlled microenvironments compatible with cell culture conditions, resulting in large variability in the development path and shape formation for each individual organoid.

A recent study8 demonstrated that quantitative imaging of organoid shape (phenotype descriptors) coupled to the assessment of a few genetic markers allow the definition of phenotypic development landscapes. Arguably, the ability to relate the diversity of genomic expression in organoids with their phenotypic behavior is a major step toward unleashing the full potential of organotypic cultures. Thus, it begs for the development of dedicated, high-content imaging approaches allowing the characterization of organoid features at subcellular, multicellular, and whole-organoid scales in 3D9,10.

We developed a versatile high-content screening (HCS) platform allowing streamlined organoid culture (from isolated human embryonic stem cells [hESCs], human induced pluripotent stem cells [hIPSCs], or primary cells to 3D, multicellular, differentiated organoids) and fast, non-invasive 3D imaging. It integrates a next-generation, miniaturized, 3D cell culture device, called the JeWells chip (the chip hereafter), that contains thousands of well-arrayed microwells flanked with 45° mirrors that allow fast, 3D, high-resolution imaging by single-objective light-sheet microscopy11. Compatible with any standard, commercial, inverted microscope, this system enables the imaging of 300 organoids in 3D with subcellular resolution in <1 h.

The microfabrication of the cell culture device starts from an existing micro structured mold, which contains hundreds of micropyramids (Figure 1A) with a square base and sidewalls at 45° with respect to the base. Figure 1C shows electron microscope (EM) images of such structures. The mold itself is made of poly(dimethylsiloxane) (PDMS) and can be made as a replica cast of a primary mold (not shown here) with corresponding features (as cavities) using standard soft-lithographic procedures. The primary mold can be produced by different procedures. The one used for this protocol was made using silicon wet etching as reported in Galland et al.11; the procedure for the fabrication of the primary mold is not critical for this protocol. The pyramids are arranged in a squared array, with the same pitch for the X and Y directions (in this case the pitch is 350 µm).

As an illustration, proof-of-concept experiments12 were published to demonstrate that the chip allows long-term culture (months) and differentiation protocols while precisely defining the number of initial cells in the wells. Individual development of a large number of organoids can be automatically monitored live using standard brightfield and 3D light-sheet fluorescence microscopes. Moreover, organoids can be retrieved to perform further biological investigations (e.g., transcriptomic analysis). This paper outlines detailed protocols for the fabrication of the cell culture coverslips, the seeding and staining procedure for fluorescence microscopy, as well as the retrieval of the organoids.

Subscription Required. Please recommend JoVE to your librarian.


NOTE: The first part of this protocol details the microfabrication of the cell culture device. An original primary mold with pyramidal cavities can be produced in-house-if micro-fabrication facilities are available-or outsourced to external companies. The primary mold used in this work is produced in-house, with fabrication process steps described elsewhere11,13. A basic protocol for the microfabrication of the mold is available in Supplementary File 1. CRITICAL: The operations in steps 1 to 6 need to be conducted in a dust-free environment. A laminar flow hood or a clean room, if available, are preferred. All through these steps, personal protective equipment (PPE) needs to be used, such as gloves, a lab coat, and safety glasses.

1. Dicing of PDMS mold

  1. Cut small portions of the PDMS mold to the dimension required for the final device (e.g., 1 cm x 1 cm [Figure 1B]). Cut them parallel to the XY directions of the array.
    ​CRITICAL: When cutting the PDMS mold in 1 cm x 1 cm dices, execute the cuts in single steps; using a one-sided razorblade, apply pressure and cut through the PDMS in one single step. This is to prevent the formation of small particles of PDMS, which can deposit on the surface of the mold and affect the quality of the replica steps (step 3).

2. Preparation of flat PDMS substrates

  1. Weigh ~15-20 g of PDMS base resin and 1.5-2 g of its reticulation agent (i.e., a ratio of 10:1), mix them carefully, and degas in a vacuum jar for ~20 min.
  2. Slowly pour the degassed PDMS onto a plastic, 15 cm wide (diameter) Petri dish.
  3. Cure the PDMS by keeping the Petri dish in an oven set at 65° C for 2 h. After waiting for the curing to complete, a flat PDMS film of ~1.5 mm thickness (contained in the Petri dish) is ready for use (Figure 2A).
  4. Using a sharp blade, cut 2 cm x 2 cm pieces out of the PDMS (Figure 2B-D).
    ​NOTE: The exact thickness of this PDMS sheet is not critical; ~1-2 mm is a suitable range. The dimension for the cut should be bigger than the textured mold but not much larger, which would result in wasting of material.

3. Production of the textured layer made of UV-curable adhesive

  1. Place one PDMS mold dice (as prepared in step 1) face-down on top of the flat PDMS cut (as prepared in step 2) (Figure 3A,B).
    NOTE: Make sure the pyramidal protrusions in the mold are oriented toward the flat PDMS substrate and that only the top surface of the truncated pyramids is in contact. Assess the correct placement with an optical microscope (Figure 3C,D).
  2. To one side of the mold, using a pipette, drop a small amount (two drops, approximately 0.1-0.2 mL) of UV-curable adhesive (Figure 4A).
    NOTE: As the liquid gets in contact with the edge of the mold, capillarity will drive the liquid to fill the cavity between the mold itself and the flat cut of PDMS used as substrate.
    1. Follow the progression of the liquid inside the cavity, for example, using an inverted optical microscope with a 10x magnification objective (Figure 4B,C).
  3. Expose the UV-curable adhesive to UV light to cure it. Adjust the time of exposure depending on the power density of the UV source used (e.g., a UV-LED box with a power density of 35 mW/cm2; 2 min at 50% power in this protocol to cure the adhesive completely).
  4. Using the excess amount of adhesive at one edge, hold the cured adhesive on the flat PDMS substrate by gently pressing with a finger (Figure 5A,B). Meanwhile, use tweezers to pinch one corner of the mold next to the same edge being held down and slowly peel it off while making sure that the textured film is not lifted up as well.
    NOTE: Figure 5C shows the result of a proper mold removal procedure; Figure 5E-G shows the wrong procedure.
  5. Trim the excess adhesive and the excess PDMS substrate using a razor blade to leave the cured, textured, adhesive layer flat on the PDMS with excess PDMS on one edge only (Figure 5D), which will be needed in step 5.

4. Coverslip substrate preparation

NOTE: As a substrate for the final device, standard rounded 1.5H coverslips with 25 mm diameter are used. Before they can be used, they need to be cleaned to remove dust and/or organic residual from their surface.

  1. Coverslip cleaning
    1. Immerse the coverslips in soapy water for 5 min while the ultrasound is applied (40 kHz, 110 W sonic power).
    2. Wash the coverslips in clean deionized (DI) water, first by immersion and finally with running DI water from a tap. Dry the coverslips with an N2 gas blow-gun.
    3. Immerse the coverslips in an acetone bath for 5 min; immediately move to a 2-propanol (IPA) bath for an additional 5 min.
    4. Rinse the coverslips with clean IPA using a squeeze bottle. Dry the coverslips with the N2 gas blow-gun.
      NOTE: Coverslips cleaned using this procedure can be stored in a closed container until needed. Keep them in a dry cabinet to avoid humidity depositing on their surface.
  2. When ready to be used, treat a clean coverslip with a short O2 plasma process to improve its hydrophilicity: O2 20 sccm, 3 mbar pressure, 60 W at the RF power generator, and 60 s duration.
  3. Immediately after the plasma activation, proceed with spin-coating the thin layer of UV-curable adhesive by placing the coverslip on the vacuum chuck of a standard spin-coater and pouring a small drop of adhesive at the center of the coverslip (Figure 6). Run the spin-coating process: spreading for 5 s at 500 rpm, coating at 3,000 rpm for 45 s (with the acceleration and deceleration set at 100 rpm/s).
    1. If spin-coating is not available, use the following alternative way to produce a thin film of UV adhesive on the coverslips:
      1. On a clean coverslip, drop ~0.1 mL of UV-curable adhesive using a pipette (Figure 7A).
      2. Take a second coverslip and place it on top of the first one to make the liquid adhesive spread evenly between the two coverslips (Figure 7B-D).
      3. Once the spreading adhesive has reached the edges of the coverslips, gently separate them by sliding one over the other. Once separated, both coverslips are fully coated with a thin layer of liquid adhesive (Figure 7E).
        NOTE: The coating might not be uniform and smooth only if the separation is not done with a smooth and continuous movement.
  4. Precuring of UV-curable adhesive
    1. After spin-coating, precure the adhesive by exposure to UV. Adjust the time of exposure depending on the power density of the UV source used (here, a UV-LED box with a power density of 35 mW/cm2 was used for 1 min at 50% power).
      CRITICAL: The adhesive used here is an optical glue. See the discussion for key points related to the energy doses for its curing.

5. Transfer of the textured film to the final substrate

  1. Take one of the textured films (prepared in step 3) and place it in contact with the adhesive-coated coverslip (prepared in step 4). Make sure the contact between the partially cured adhesive on the coverslip and the textured film is as uniform as possible (Figure 8A-C).
    CRITICAL: At this stage, the adhesive on the coverslip should be solid enough to avoid reflowing, which would fill the pyramidal cavities of the textured film when placed in contact but also be plastic and adhesive enough that contact can be optimized by gently pressing on the textured film.
  2. Expose the coverslips to UV light until the adhesive layer coated on the coverslip is fully cured. This will seal the textured film on the coverslip and provide leak-proof isolation between the pyramidal cavities.
  3. Finally, peel off the PDMS flat substrate (Figure 8D-F). Using tweezers, pinch the PDMS on one corner at the edge where excess material was left after trimming (step 3.5). This way, the textured film layer is left adhesive to the coverslip with open access at the top for cell seeding.CRITICAL: When peeling off the flat PDMS, the textured film should remain well attached to the coverslip. Adhesion failures are easily confirmed if the textured film can be peeled off from the coverslip after final exposure to UV without any effort.

6. Long-term passivation of the cell culture coverslip for cell culture

NOTE: Passivation is achieved by generating a conformal and continuous coating of a biomimetic copolymer with a structure similar to the polar group of phospholipids in the cell membrane. This conformal coating prevents cells adhesion to the cell culture device

  1. Prepare a solution with 0.5% (w/v) of the biomimetic copolymer dissolved in pure ethanol. Store the solution at 4 °C for future use.
  2. Place the cell culture coverslip in a 35 mm Petri dish and fully cover it with the biomimetic copolymer solution.
  3. After 5 min, remove the cell culture coverslip from the container with the biomimetic copolymer solution and leave it to dry at room temperature inside the final dish in a biosafety hood (>1 h).
    NOTE: A thicker coating can be produced by increasing the concentration of the biomimetic copolymer in the coating solution; the results of a thicker coating are visible under a brightfield microscope (Figure 9A).

7. Cell seeding

  1. Degassing and sterilization
    1. Just before cell seeding, dispense sterile phosphate-buffered saline (PBS) into the cell culture dishes (typically 1 mL for a 35 mm Petri dish). Degas the dish with the sterile PBS using an ultrasonic device for ~10 min, followed by several rounds of pipetting to remove all the bubbles.
      CRITICAL: If air is trapped inside the pyramidal cavities, it will prevent the cells from entering them. To make sure there is no air trapped in the pyramid before cell seeding, it is recommended to visually ensure (under a benchtop brightfield microscope at 10x or 20x magnification) the absence of air in these cavities (Figure 10).
    2. Replace the PBS with sterile culture medium and sterilize the plate with UV light for 30 min under a cell culture hood.
      NOTE: From this step onward, the dish should be considered as sterile and manipulated using sterile techniques. An alternative way to remove trapped air from the cell culture device is to use a vacuum jar with a vacuum pump.
  2. Cell suspension preparation
    NOTE: Cells can be seeded as single or small cell aggregates and enter the sample wells through the top aperture. Over time, the inserted cells aggregate and grow inside the sample wells into spheroids of a size greater than the size of the aperture. As validated cell line models, use HCT116 (CCL-247 ATCC) or MCF7 (HTB-22 ATCC) cancer cells maintained in recommended culture medium (ATCC guidelines).
    1. Prepare a cell suspension (e.g., using a trypsinization process following ATCC guidelines). Follow trypsinization/cell suspension preparation recommendations for the cells of interests.
    2. Count and adjust the cell concentration to 0.5 × 106 cells/mL in the recommended culture medium.
  3. Cell dispensing
    1. Remove the PBS buffer from the 35 mm cell culture dish and then dispense 1 mL of the adjusted cell suspension. See Figure 11A for an optical microscope image of a cell seeding procedure with adequate cell density and homogeneity.
    2. Place the cell culture dish back into the cell incubator (37 °C, 5% CO2, and 100% humidity) for 10 min. Approximately 80-100 cells will enter each pyramidal cavity.
      NOTE: It is possible to increase the number of cells per cell culture dish by increasing either the cell concentration or the time spent before cell suspension removal. Typically, spheroid formation takes several hours (depending on cell type) after cell seeding and can be followed with a brightfield microscope (4x to 40x objectives; Figure 11). From here, culture medium, extracellular matrices, and differentiation growth factors can be changed or added to the cell culture dish containing the spheroids in accordance with typical differentiation protocols that could last a few days, weeks, or months.
    3. After the 10 min incubation, recover the cell culture dish from the incubator and gently aspirate the cellular suspension to remove untrapped cells. Add 1 mL of culture medium to a 35 mm dish and place it back into the cell incubator.
      CRITICAL: At this stage, because spheroids have not formed yet, it is very important to avoid strong aspiration or dispensing that will result in loss of the cells. Visual control using a benchtop brightfield microscope is highly recommended at this step.

8. Immunostaining and imaging

  1. Fixation and staining
    NOTE: Different classical procedures of fixation and immunostaining are completely compatible with the cell culture dish. One typical protocol is described here.
    1. Fix the organoids/spheroids in the cell culture dish for 20 min in 4% paraformaldehyde at room temperature.
    2. Permeabilize the organoids for 24 h in 1% surfactant solution in sterile PBS at 4 °C on an orbital shaker and incubate for 24 h in blocking buffer (2% bovine serum albumin [BSA] and 1% surfactant in sterile PBS) at 4 °C on an orbital shaker.
    3. Incubate the samples with primary antibodies of interest at a dilution between 1/50 and 1/200 (or according to the manufacturer's recommendations) in antibody dilution buffer (2% BSA and 0.2% surfactant in sterile PBS) at 4 °C for 48 h.
    4. Rinse the samples 3x with washing buffer on an orbital shaker (3% NaCl and 0.2% surfactant in sterile PBS) and incubate with corresponding secondary antibodies in antibody dilution buffer (dilution between 1/100 and 1/300 or according to the manufacturer's recommendations), 0.5 µg/mL 4',6-diamidino-2-phenylindole (DAPI), and 0.2 µg/mL Alexa Fluor 647 or 488 Phalloidin at 4 °C for 24 h on an orbital shaker, followed by five rinsing steps with PBS. Optionally, mount the samples using a water-soluble clearing agent prewarmed to 37 ˚C.
  2. Imaging
    NOTE: At this stage, the organoids in the microwell plate can be considered as a normal culture dish containing fixed and stained samples for imaging: any standard imaging procedure can be used with no adaptation or modification required. Figure 12 illustrates a representative result of images and 3D reconstruction obtained using a spinning disk confocal microscope, with a 40x air objective (numerical aperture 0.75).
    1. Use constructor software for automatic image acquisition process with the following settings: exposure time = 50 ms, with a z motorized stage to acquire a z-stack (1 µm Z-step, for a total height of 70 µm).
    2. Perform 3D reconstruction using image analysis software.

9. Release and collection of the organoids

NOTE: The textured adhesive layer of the cell culture dish can be detached from the coverslip to release the living spheroids/organoids (before fixation) contained inside the pyramidal cavities for analysis of the cells with other procedures such as RNA sequencing, -omic approaches, in vitro experiments, and in vivo transplantation.

  1. With the sample still in the Petri dish and in a biosafety hood, use a blade such as a scalpel to cut a corner of the textured adhesive layer.
  2. With tweezers, pinch the textured adhesive layer on the cut edge and gently peel it off from the glass coverslip but keep it immersed in the medium (some organoids might be remaining with the adhesive layer, Figure 13).
  3. Rinse 3x with culture medium and collect the organoids by pipetting.

Subscription Required. Please recommend JoVE to your librarian.

Representative Results

Figure 8F shows the typical aspect of a cell culture coverslip after successful fabrication. The UV-curable adhesive layer appears flat and adheres well to the coverslip. The transfer of the adhesive layer on the coverslip might fail if the layer on the coverslip is overcured, or if the removal of the flat PDMS substrate is done incorrectly (as shown in Figure 8G,H). In both cases, the failure is evident as no textured film is transferred to the coverslip.

For the microwells to work, the textured film produced (as described in protocol step 3) needs to have both the top and the bottom sides of the pyramids open to ensure that cells during the seeding in step 7 can enter the cavities and remain contained once the organoids start forming. Figure 9 shows the increase in thickness of the coating by increasing the concentration of the biomimetic copolymer in the coating solution. Figure 10 shows the results of degassing the cell culture dishes to ensure no air is trapped in the pyramidal cavities. Figure 10A,B shows complete degassing, while Figure 10C,D shows air bubbles in the pyramidal cavities.

In Figure 11, cells are trapped inside the pyramids, and the corresponding organoids are formed after days 2 and 15 of culturing. If the top opening of the microwells is instead closed (as a consequence of incorrect step 3), no cells will be left after rinsing the sample after cell seeding. Figure 12 shows representative images of the organoids and a 3D reconstruction obtained using a spinning disk confocal microscope with a 40x air objective (numerical aperture 0.75). After release and collection, the organoids can be stored in a small vial and used for further analysis (e.g., RNA sequencing). Figure 13B shows an example of a sample of organoids collected as described.

Figure 1
Figure 1: Poly(dimethylsiloxane) mold. (A) The starting PDMS mold obtained as a cast replica of a 4" textured wafer (see Supplementary File 1). (B) 1 cm x 1 cm wide cut of the starting PDMS mold. (C) Scanning electron microscope images of the trapezoidal pillars arranged in a squared array, which populate the surface of the mold. Scale bars = 1 cm (A), 100 µm, and 200 µm (C top and bottom, respectively). Abbreviation: PDMS = poly(dimethysiloxane). Please click here to view a larger version of this figure.

Figure 2
Figure 2: Flat PDMS substrate. (A) A layer of PDMS, ~1 mm thick, is prepared in a standard 15 cm wide Petri dish. (B) Peeling out of a fresh cut of PDMS. (C) The remaining PDMS on the Petri dish can be cut into more pieces for further usage. (D) Flat PDMS cut and PDMS mold side by side; the flat PDMS is bigger than the mold. Abbreviation: PDMS = poly(dimethysiloxane). Please click here to view a larger version of this figure.

Figure 3
Figure 3: Placing the mold on the substrate. (A) The mold is placed on the flat substrate with the pyramids facing down. (B) Different appearance of poor contact (top) and good contact (bottom) between the mold and PDMS substrate. (C) Optical microscope image of an area with poor contact; the red circles highlight the pillars not in contact with the substrate-they appear of a brighter color than the ones in contact in the same image. (D) Optical image of an area with all pillars in good contact. Scale bars = 200 µm (C,D). Abbreviation: PDMS = poly(dimethysiloxane). Please click here to view a larger version of this figure.

Figure 4
Figure 4: Capillary filling of UV-curable adhesive. (A) This sequence of images shows (from left to right) the pouring of a small quantity of adhesive on one edge and the progression of the liquid adhesive in the cavity between the mold and substrate. The yellow arrows point at the direction of the liquid flow. (B) Sequence of optical images (40x magnification) of the liquid front moving; when in good contact, the front of the liquid moves around the edges of the pyramids without leaks. The yellow arrows point at the flow direction and the red arrows point at the leading edge of the liquid front moving around the contact edges. (C) Sequence of optical images (40x) of the liquid front moving when the contact is poor; the red arrows point at the leading edge of the liquid infiltrating between the mold and the substrate. Please click here to view a larger version of this figure.

Figure 5
Figure 5: Removal of the mold after adhesive curing. (A-C) Correct procedure for peeling off the mold: while gently pressing on the excess of adhesive at the left edges with tweezers, the mold is pinched near the same edge and slowly bent to peel off. (D) The result of the correct procedure; the excess is trimmed out on three edges, while a small excess of PDMS is left on the fourth side (yellow arrow). (E-G) Example of an incorrect procedure for removing the mold: the mold is pinched with tweezers from the opposite edge, resulting in peeling off from the flat substrate of the adhesive film together with the mold. Please click here to view a larger version of this figure.

Figure 6
Figure 6: Coating of the glass coverslip. (A,B) Example of a good coating procedure: the glass coverslip is placed on the vacuum chuck of a spin-coater, and enough adhesive is dispensed at the center, giving a homogeneous coating of the coverslip. (C,D) Example of a bad coating procedure: not enough adhesive is dispensed, resulting in non-uniform coating of the coverslip. (E) From left to right, examples of good, acceptable, and bad coating. Please click here to view a larger version of this figure.

Figure 7
Figure 7: Alternative coating process. (A) A small drop of liquid adhesive is placed at the center of the coverslip. (B) A second coverslip is gently placed on top of the first one. (C,D) The liquid spread between the two coverslips and evenly distributed. (E) Examples of results after correct separation of the coverslips (left) or incorrect separation (right). Please click here to view a larger version of this figure.

Figure 8
Figure 8: Transfer to the coverslip. (A) The cured and trimmed adhesive film (Figure 5D) is placed on the coverslip with partially cured adhesive. (B) Example of good contact between the textured film and the coverslip. The inset is an optical image showing uniform contact of the areas between the trapezoidal cavities (dark uniform color). (C) Example of poor contact between the textured film and the coverslip. The brighter areas are not in contact, the red arrows point at those areas. The inset is an optical image showing the different appearance of good contact (two on the left) and poor contact (two on the right). Scale bars (yellow, B,C) = 200 µm. (D,E) Correct procedure for removing the PDMS flat substrate: tweezers are used to pinch the PDMS at the edge with excess of PDMS and bend to gently peel off. (F) The result with the UV adhesive-textured film successfully transferred to the coverslip. (G,H) Example of incorrect peeling procedure: tweezers are used to pinch the flat PDMS on one of the edges where the excess material was trimmed-off; thus, the film is removed together with the flat PDMS. Abbreviation: PDMS = poly(dimethysiloxane). Please click here to view a larger version of this figure.

Figure 9
Figure 9: Coating with biomimetic copolymer. (A) Cell culture device coated with biomimetic copolymer using a solution of 0.5% (left) or (B) 1% (right) (w/v) in pure ethanol. Scale bar = 100 µm. Please click here to view a larger version of this figure.

Figure 10
Figure 10: Degassing of cell culture device. (A) Cell culture device before complete degassing. (B) Cell culture device after complete degassing; (C,D) cell culture device with only partial degassing where air bubbles are trapped in the pyramids. Scale bars = 200 µm (A-D). Please click here to view a larger version of this figure.

Figure 11
Figure 11: Representative brightfield images of cell seeding and growth of organoids. (A) Cells are visible floating on the top. (B) Cells trapped in the pyramidal cavities after rinsing away the cell suspension. Only trapped cells will be retained as shown. (C) Cells aggregate to form spheroids in each cavity (day 2 after cell seeding). (D) Image at day 15 after cell seeding of an organoid that has grown in a pyramidal cavity. Scale bars = 200 µm (A B), 120 µm (C), and 150 µm (D). Please click here to view a larger version of this figure.

Figure 12
Figure 12: 3D imaging of organoids. (A) Three different Z plans acquired using a spinning disk confocal (lens 40x, air, numerical aperture: 0.75) of an organoid under neuroectoderm differentiation (day 7after cell seeding). Actin staining using phalloidin (Alexa Fluor 647) in orange and N-cadherin immunostaining (Alexa Fluor 561) in blue. (B) A 3D reconstruction using image analysis software of the corresponding organoids (80 Z plans, total height of 100 µm). White arrows: clusters of cells around a streak with high level of actin expression. Blue arrows: cell clusters positive for N-cadherin protein expression. Please click here to view a larger version of this figure.

Figure 13
Figure 13: Release of organoids. (A) Removing the textured adhesive layer with tweezers; (B) brightfield image of an organoid suspension obtained after the removal. Scale bar = 200 µm (B). Please click here to view a larger version of this figure.

Supplementary File 1: A step-by-step protocol for the microfabrication of a Si mold with pyramidal microcavities is presented. Please click here to download this File.

Subscription Required. Please recommend JoVE to your librarian.


The procedure for the fabrication of the microwell culture dish, which allows high-density organoid culture and differentiation, has been described in this paper. Owing to the geometry and arrangement of the microcavities, thousands of spheroids can be cultured and stained in a single plate for long periods of time (several weeks or more) with nearly no loss of material. As a comparison, an area of 4 mm x 2 mm on the cell culture plate can contain as many spheroids as a single 384-well plate with an area of 12 cm x 8 cm.

The regular distribution of the microcavities and the flat standard coverslip used as the substrate enable the monitoring of thousands of live or fixed organoids in 3D, without the need for complex and time-consuming sample manipulation between culturing vessels and imaging supports. Due to the pyramid shape with a smaller top than the base, there is no loss of organoids due to pipetting steps during the medium exchange, dispensing of extracellular matrices, or staining procedures. All these steps could be performed directly, as for a 2D cell monolayer without any extra precautions, unlike classical organoid generation techniques (e.g., hanging droplets, low attachment plates, Aggrewell).

Moreover, compared to these existing techniques, the microwells have been designed for high-resolution 3D imaging compatible with all types of immersion lenses (air, water, oil, and silicon). Long-term maintenance of organoids closed to a flat glass coverslip makes the cell culture coverslip compatible with high-resolution light microscopes, which is impossible with the existing methods without complex handling and fastidious organoid transfer. Further, an optimized passivation procedure avoids any adhesion of cells/organoids for long-term culture, thus controlling the differentiation of the 3D culture Finally, since the textured adhesive layer is removable, living organoids can be retrieved to perform other types of experiments such as -omics assays or in vivo transplantation.

It should be noted that the pyramidal shape and array arrangement of the microcavities used for this protocol are derived from the primary mold, which has been fabricated by means of silicon wet-etching to provide surfaces with mirror-like finishing and 45° inclination to allow for single-objective light-sheet microscopy (soSPIM11). While a discussion of these requirements and a full description on how to produce such molds can be found elsewhere12,13, a simple protocol is also given in Supplementary File 1. Different ways of producing a primary mold are available, which are fully compatible with the fabrication of the chips in this protocol. Laser etching of glass and high-resolution 3D printing, for example, could be used to produce a primary mold with suitable cavities for the containment of the organoids, but are not suitable for soSPIM imaging.

The fabrication protocol disclosed here can be adapted for use in any standard biological laboratory, as no dedicated, advanced microfabrication tools are required. Some critical steps of the fabrication of the microwells are the production through capillary filling of the textured adhesive film and its transfer to the glass substrate. However, in our experience, they can be mastered with high reproducibility in a short period of time. The distance between the pyramids should be as small as possible to increase the density of cavities for the growth of organoids, but it also needs to be large enough so that the cavities can be sealed on the substrate without any leak or lack of sealing between them. We found that a distance of 50 µm is a good compromise for this case.

During cell seeding, a critical aspect is the removal of any air bubbles trapped inside the cavities to ensure cell entry. We describe here a validated solution, which uses only classical laboratory equipment (e.g., an ultrasonic homogenizer or a vacuum jar). To allow better homogenization of the number of cells per cell culture dish, it is recommended to dispense the cell suspension from the side of the dish and not directly above the coverslip. Gentle manual agitation could also help to homogenize the cellular solution.

After the seeding of cells, the cell culture dish with organoids can be treated like any other classical culture support; no specific manipulation or critical steps are required. All the protocols that have been used with the microwell devices are practically the same as those used in low-attachment dishes such as U-bottom plates. Hence, organoid culture in these microwells does not require any change in culture conditions compared to other standards.

Another critical factor is that the adhesive used here is an optical glue. In its suggested best usage (also refer to application notes from the vendor), two optical elements to be glued are aligned and UV-curable adhesive is applied at the interface. A first exposure to UV of an energy insufficient to cause full polymerization is used to solidify the glue while maintaining adhesive properties. The optical components can be moved to achieve optimal alignment, and then the glue is fully cured to its final state with a second exposure to UV to provide permanent adhesion. The first and second exposure energy doses (i.e., exposure times if a source of known and constant energy density is used) must be optimized depending on the energy density of the UV source used, the thickness of the layer of glue, the transmission of UV through the optical elements, and the surface texturing (or lack of it) of the surface to be glued.

While the adhesion between the textured film and the adhesive-coated coverslip is required to be strong enough to provide for leak-proof sealing, it also needs to be possible to remove the textured film to retrieve the organoids after imaging, if any further analysis is required, such as genotypic profiling. A correct compromise between leak-proof sealing and the capability to peel off the textured film has been achieved with the protocol described here, but further optimization of the precuring and final curing times for the adhesive thin layer on the coverslip might be required as it depends on the UV-exposure system and film thickness used.

One limitation in the current microwells version is in the seeding procedure; cells must be seeded before the ECM components to allow them to enter the vicinity of the pyramid. Improvements are ongoing to coat the microwells with extracellular matrix components, as a first step. We have not yet performed any electron microscopy (EM) on the fixed organoids within the cavities. Some modification to these fabrication and cell culture protocols will be required before imaging of organoids in the microwells with EM becomes a viable option.

A natural future extension of this method is to provide high-content screening capacities to allow multicondition testing in a single workflow (multiwell plate). This cell culture device provides a unique alternative to existing organoid techniques, offering unsurpassed culturing and imaging throughput and opening new perspectives in the field of organoid research for biomedical application and drug discovery.

Subscription Required. Please recommend JoVE to your librarian.


An international patent application has been published with the Publication Number WO 2021/167535 A1.


The research is supported by the CALIPSO project supported by the National Research Foundation, Prime Minister's Office, Singapore, under its Campus for Research Excellence and Technological Enterprise (CREATE) programme. V.V. acknowledges the support of NRF investigator NRF-NRFI2018-07, MOE Tier 3 MOE2016-T3-1-005, MBI seed funding, and ANR ADGastrulo. A.B. and G.G. acknowledge the support from MBI core funding. A.B. acknowledges Andor Technologies for the loan of the BC43 microscope.


Name Company Catalog Number Comments
2-Propanol Thermofisher scientific AA19397K7
Acetone Thermofisher scientific AA19392K7
BC43 Benchtop Confocal Microscope Andor Technology spinning disk confocal microscope
bovine serum albumin  Thermofisher scientific 37525
Buffered oxide etching solution Merck 901621-1L
CEE Spin Coater Brewer Science 200X
DAPI Thermofisher scientific 62248
Developer AZ400K Merck 18441223164
DI Milliq water Millipore
Fetal Bovine Serum (FBS) Invitrogen 10082147
Glass coverslips Marienfled 117650 1.5H, round 25 mm diameter
Hepes Invitrogen 15630080
Imaris software BitPlane image analysis software
Inverted Transmission optical microscope Nikon TSF100-F
Labsonic M Sartorius Stedium Biotech Ultrasonic homogenizer
Lipidure NOF America CM5206 bio-mimetic copolymer
NOA73 Norland Products 17-345 UV curable adhesive
Penicillin-Streptomycin Invitrogen 15070063
Phalloidin Thermofisher scientific  A12379 Alexa Fluor
Phosphate Buffer Solution Thermofisher scientific 10010023
Photo Resist AZ5214E Merck 14744719710
Pico Plasma tool Diener Electronic GmbH + Co. KG Pico Plasma For O2 plasma treatment
RapiClear 1.52 Sunjin lab RC 152001 water-soluble clearing agent
RCT Hot Plate/Stirrer IKA (MY)
Reactive Ion Etching tool Samco Inc. (JPN) RIE-10NR
RPMI 1640 Invitrogen 11875093 culture medium for HCT116 cells
Sylgard 184 Silicone Elastomer Kit Dow Corning 4019862 Polydimethylsiloxane or in short, PDMS
Trichloro(1H,1H,2H,2H-perfluorooctyl)silane Sigma Aldrich 448931-10G
Triton X-100 Sigma Aldrich T9284 surfactant
Trypsin EDTA Thermofisher scientific 15400054
Ultrasonic Cleaner Bransonic CPX2800
UV-KUB 2 KLOE UV-LED light source, 365 nm wavelength, 35 mW/cm2 power density
UV mask aligner SUSS Microtec Semiconductor (DE) MJB4



  1. Kim, J., Koo, B. -K., Knoblich, J. A. Human organoids: model systems for human biology and medicine. Nature Reviews Molecular Cell Biology. 21 (10), 571-584 (2020).
  2. Takebe, T., Wells James, M. Organoids by design. Science. 364 (6444), 956-959 (2019).
  3. Kratochvil, M. J., et al. Engineered materials for organoid systems. Nature Reviews Materials. 4 (9), 606-622 (2019).
  4. Rossi, G., Manfrin, A., Lutolf, M. P. Progress and potential in organoid research. Nature Reviews Genetics. 19 (11), 671-687 (2018).
  5. O'Connell, L., Winter, D. C. Organoids: past learning and future directions. Stem Cells and Development. 29 (5), 281-289 (2020).
  6. Vives, J., Batlle-Morera, L. The challenge of developing human 3D organoids into medicines. Stem Cell Research & Therapy. 11 (1), 72 (2020).
  7. Busslinger, G. A., et al. The potential and challenges of patient-derived organoids in guiding the multimodality treatment of upper gastrointestinal malignancies. Open Biology. 10 (4), 190274 (2020).
  8. Lukonin, I., et al. Phenotypic landscape of intestinal organoid regeneration. Nature. 586 (7828), 275-280 (2020).
  9. Rios, A. C., Clevers, H. Imaging organoids: a bright future ahead. Nature Methods. 15 (1), 24-26 (2018).
  10. Dekkers, J. F., et al. High-resolution 3D imaging of fixed and cleared organoids. Nature Protocols. 14 (6), 1756-1771 (2019).
  11. Galland, R., et al. 3D high- and super-resolution imaging using single-objective SPIM. Nature Methods. 12 (7), 641-644 (2015).
  12. Beghin, A., et al. High content 3D imaging method for quantitative characterization of organoid development and phenotype. bioRxiv. , (2021).
  13. Beghin, A., et al. Automated high-speed 3D imaging of organoid cultures with multi-scale phenotypic quantification. Nature Methods. 19 (7), 881-892 (2022).
This article has been published
Video Coming Soon

Cite this Article

Grenci, G., Dilasser, F., Mohamad Raffi, S. B., Marchand, M., Suryana, M., Sahni, G., Viasnoff, V., Beghin, A. A High-Throughput Platform for Culture and 3D Imaging of Organoids. J. Vis. Exp. (188), e64405, doi:10.3791/64405 (2022).More

Grenci, G., Dilasser, F., Mohamad Raffi, S. B., Marchand, M., Suryana, M., Sahni, G., Viasnoff, V., Beghin, A. A High-Throughput Platform for Culture and 3D Imaging of Organoids. J. Vis. Exp. (188), e64405, doi:10.3791/64405 (2022).

Copy Citation Download Citation Reprints and Permissions
View Video

Get cutting-edge science videos from JoVE sent straight to your inbox every month.

Waiting X
Simple Hit Counter