Protocol for differentiation of bovine adipogenic progenitor cells embedded in alginate sheets

Summary

Here, we present a cost-effective protocol to differentiate bovine fibro-adipogenic progenitors in a thin hydrogel sheet adherent to 96-well plates. We describe steps for the embedding and culturing of cells in alginate sheets, culture maintenance, and analysis. Compared to alternative three-dimensional (3D) models such as hydrogel-based microfibers, this approach simplifies automation while retaining efficient maturation of adipocytes. Embedded cells are still subjected to a 3D environment, but the sheets can be handled and analyzed like two-dimensional cultures.

Graphical abstract

Highlights

• •Protocol for adipogenic differentiation of bovine FAPs
• •Cells embedded in alginate are cultured as thin sheets stuck to a well plate
• •Comparable differentiation to 3D fiber culture
• •Automation friendly and allows for high content analysis

Publisher’s note: Undertaking any experimental protocol requires adherence to local institutional guidelines for laboratory safety and ethics.

Here, we present a cost-effective protocol to differentiate bovine fibro-adipogenic progenitors in a thin hydrogel sheet adherent to 96-well plates. We describe steps for the embedding and culturing of cells in alginate sheets, culture maintenance, and analysis. Compared to alternative three-dimensional (3D) models such as hydrogel-based microfibers, this approach simplifies automation while retaining efficient maturation of adipocytes. Embedded cells are still subjected to a 3D environment, but the sheets can be handled and analyzed like two-dimensional cultures.

Before you begin

Efficient adipogenic differentiation can be achieved in a variety of 3D environments, including aggregates, scaffolds and hydrogels. Embedding adipogenic progenitor cells in a hydrogel microfiber creates a three-dimensional environment that enables the cells to efficiently differentiate towards mature adipocytes with unilocular fat droplets within 4–6 weeks.^1,2 While this differentiation method is well suited for production of larger quantities of fat, it is less so for media optimization or larger-scale screening purposes, since media exchanges with free-floating structures need careful handling. 2D differentiation protocols are relatively easy to handle and can be adapted to fully automated workflows, but they produce less mature adipocytes and are often not representative for 3D differentiation.¹ Therefore, we developed a new culture method, where the cells are embedded in the hydrogel solution (hence the three-dimensional environment). However, instead of free-floating fibers this hydrogel is “glued” to the bottom of a well plate, forming a thin sheet. This enables us to treat the culture like a conventional 2D culture, while maintaining the efficient differentiation capacity of the 3D culture. The sheets are about 200 μm thick, thereby enabling good nutrient and gas exchange through the structure. In addition to the automation-friendly set up, relatively small sheets can be produced, reducing the quantity of cells required to obtain reliable results. Depending on the well plate used, sheets from 1–13 μL (with 60,000 to 390,000 cells) can be produced. Compared to 2D differentiation, we saw higher reproducibility of the results from our sheet culture to 3D differentiation: some media components that were essential in 2D were not needed in sheets or fibers, while some compounds with positive effects on fiber and sheet differentiation had little or no effect in the 2D culture.

The protocol describes the specific steps for using bovine fibro-adipogenic progenitor cells (FAPs) using a liquid handler. This protocol can also be applied to FAPs or other adipogenic progenitor cell types from other species; however, media conditions and potentially cell numbers must be adapted accordingly. Details on the differentiation protocols used here can also be found in a recent publication from our group.³ All steps can also be performed manually in the absence of a liquid handler.

Institutional permissions

For these experiments, fresh bovine skeletal muscle samples were obtained from a registered abattoir according to national guidelines on animal tissue handling. Ethical approval was not required for acquisition of samples from commercially slaughtered cattle. Samples were acquired and transported in accordance with Dutch national guidelines on handling of animal material. Mosa Meat B.V. has been granted a license to handle Category 3 animal material.

Any experiments on live vertebrates or higher invertebrates must be performed in accordance with relevant institutional and national guidelines and regulations. Before using this method, it is therefore essential to acquire the relevant permission for the cell type of choice.

Prepare necessary buffers and solutions in advance

Timing: 2 h

The following solutions are stable for a longer time and can be prepared in advance. Prepare all reagents as described in the materials section.

• 1.
  Prepare gelation buffer.

  • a.
    Weigh in CaCl₂ and HEPES and add to a 1 L glass bottle.
  • b.
    Dissolve in 1 L Type 1 water.
  • c.
    Adjust the pH to 7.4 with 10 M NaOH.
  • d.
    Autoclave.
  • e.
    Sterile gelation buffer can be stored at room temperature for up to 6 months.
• 2.
  Prepare 1% sodium alginate.

  • a.
    Dissolve 1 g sodium alginate in 100 mL Type 1 water by gently mixing for 4–6 h on a shaker table.
  • b.
    Sterilize by filtering through a 0.2 μm filter.
  • c.
    Sterile solution can be stored in the fridge for up to 6 months.

PDL (Poly-D-Lysine) coating of 96-well differentiation plates

Timing: 10 min hands-on time, 4 h run-time

In order to make alginate sheets, it is beneficial to first coat the output plates with a PDL coating, wash with gelation buffer, and then allow it to dry completely before use. This precoating enables the attachment of the cell/alginate mixture to the plate bottom and simplifies the handling of the cultures. Sheets can be formed as well on non-coated tissue culture plates, but with lower reproducibility and they tend to swim up during prolonged culture, which makes media exchanges and analysis more difficult. Therefore, we strongly recommend using PDL coated plates when longer cultures (involving media exchanges) are planned.

• 3.
  Prepare relevant solutions.

  • a.
    Coating solution.
  • b.
    Gelation buffer.
• 4.
  Edit Python file for run variables.

  • a.
    Change NUM_PLATES to the number of plates you have.
  • b.
    Edit staring tip.
  • c.
    [Optional] Adapt incubation times.
• 5.
  Load protocol on Opentrons App.

  • a.
    Place labware on Opentrons as directed by the app (Figure S1A).
  • b.
    Fill required reagents in the correct locations.
• 6.
  The liquid handler steps performed are as follows:

  • a.
    Add 50 μL PDL coating solution per well.
  • b.
    Incubate for 60 min.
  • c.
    Wash twice with 100 μL gelation buffer.
  • d.
    Incubate for 60 min.
  • e.
    Remove all liquid.
  • f.
    Let the plates dry for 60 min [this step can be skipped if the plates are not used immediately].

Optional: When plates are stored for longer, they can be sterilized under UV light for about 15 min prior to use.

Note: All steps should be performed in a sterile environment, we recommend using a liquid handler placed under a laminar flow hood. It is important for the plates to be completely dry before use. If the plates are not dry, the moisture causes the dispensed alginate sheets to deform around the edges of the well, creating a problematic sample.

Pause point: Both the coating and the gelation buffer incubation step can be performed overnight instead of the 1 h on the liquid handler, in case needed. Dried plates can be stored up to one month at room temperature.

Cell preparation

Timing: 3–4 days for proliferation, 30 min hands-on time on day of experiment

For adipogenic differentiation, sufficient amounts of appropriate progenitor cells are needed. We only tested bovine FAPs, but other cell types with adipogenic potential (like stromal-vascular cells, mesenchymal stem cells, or 3T3-L1 cells) should behave in a similar manner. However, adaptations regarding the cell density and the optimal proliferation and differentiation medium would be needed. Cells were derived and cultured in the following way:

• 7.
  Isolate cells from the semitendinosus muscle of commercially slaughtered Belgian Blue cattle (both male and female, aged from 1 to 7 years).

  • a.
    Mince and dissociate muscle with collagenase (CLSAFA, Worthington; 1 h, 37°C).
  • b.
    Filter cell slurries through a 100 μm cell strainer and incubate in an ammonium-chloride-potassium (ACK) erythrocyte lysis buffer (1 min, room temperature).
  • c.
    Resuspend cells in Serum-Free Growth medium (SFGM), and filter through a 40 μm strainer prior to culture.
  • d.
    Seed FAPs in T175 flasks with a density of 4,000 cells/cm² in SFGM and passage every 4–5 days.

Optional: Larger cell stocks from individual isolations are cryopreserved and cells are brought up prior to the experiments for at least one passage before making alginate sheets.

Note: Per 5 μL alginate sheet, 150,000 cells are needed. We suggest having at least three replicates per condition and calculate about 5–15 μL of dead volume per used well of the 96-well PCR plate.

In our experience, FAPs differentiate best when they have at least one recovery passage after cryopreservation and are about 90% confluent at the day of harvesting. Again, some optimizations are needed depending on the cell type used and we suggest performing initial testing on optimal cell densities and media used.

For primary cells, we recommend using cells between passage 1 to 3 for optimal differentiation.

• 8.
  Harvesting FAPs.

  • a.
    Check confluency of the cells under the microscope. Ideal confluency for efficient differentiation is about 90%.
  • b.
    Wash cells 1× with PBS.
  • c.
    Add TrypLE and incubate for 3–5 min at room temperature until cells begin to detach.
  • d.
    Detach the cells with 3× volume PBS and transfer to 15–50 mL reaction tubes.
  • e.
    Spin down cells at 250 g for 5 min at room temperature.
  • f.
    Count cells using a live/dead stain (like Trypan Blue). Used cell numbers should be adjusted for the live cell numbers only.
• 9.
  Make cell/alginate solution.

  • a.
    Resuspend cells in Adipogenic Differentiation Medium (ADM) at 60 M/mL (take pellet size into account for accuracy).
  • b.
    Add same volume of 1% sodium alginate and mix for a final v/v ratio of 1:1.
• 10.Transfer mixture to the appropriate tubes for the step “preparing cell/alginate sheets” (e.g., Hard-Shell® 96-Well PCR Plates from Biorad).

CRITICAL: The final cell concentration in alginate is critical and care should be taken when counting and diluting the cells. In addition, pipetting and mixing steps should be performed carefully to avoid the formation of air bubbles in the mixture. See example in Methods video S2.

Note: All cell culture steps depend on the cell type and species used (initial culture, harvesting, media used) and may need to be adjusted accordingly.

Key resources table

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Biological samples
Semitendinosus bovine muscle	Local abattoir	N/A
Chemicals, peptides, and recombinant proteins
DMEM/F12	Gibco	CAT# 21331-020
Poly-D-Lysine hydrobromide	MP Biomedicals	CAT# 215017550
Pierce 20× borate buffer	Thermo Fisher	CAT# 28341
High viscosity non-functionalized alginate solution	Sigma-Aldrich	CAT# W201502
HEPES	Sigma-Aldrich	CAT# H3375-1KG
CaCl2	Sigma-Aldrich	CAT# C3881-1KG
Chemically defined FBS replacement	Kolkmann et al. 2022	N/A
Phosphate buffer solution (PBS)	Fisher Scientific	CAT# 11530546
NaOH	Sigma-Aldrich	CAT# 1.60309
TrypLE Express Enzyme (1×)	Thermo Fisher	CAT# 12604021
Type 1 water	Milipure Purification System	N/A
Hoechst 34580	Sigma-Aldrich	CAT# 63493
Bodipy	Thermo Fisher	CAT# D3922
Nile red	Sigma-Aldrich	CAT# N3013
D-(+)-Galactose	Sigma-Aldrich	CAT# G5388
Sodium pyruvate	Sigma-Aldrich	CAT# P2256
Insulin recombinant human	PAN-Biotech	CAT# P-2701000
Hydrocortisone	Sigma-Aldrich	CAT# H0135-1mg
L-Ascorbic acid	Sigma-Aldrich	CAT# A8960-5g
Putrescine dihydrochloride	Sigma-Aldrich	CAT# P5780
Chemically defined lipid concentrate	Gibco	CAT# 11905-031
Indomethacin	Sigma-Aldrich	CAT# I7378
EDTA disodium salt	Sigma-Aldrich	CAT# 27285-500G-R
Trypan blue solution 0.4%	Thermo Fisher	CAT# 15250061
Collagenase	Worthington	CAT# CLSAFA
Ammonium chloride	Fisher Scientific	CAT# A661-500
Potassium bicarbonate	Fisher Scientific	CAT# P235-500
Paraformaldehyde	Merck	CAT# 47608
Experimental models: Cell lines
Fibro-adipogenic progenitor cells (FAPs) between passage 1 to 3	Dohmen et al.1	N/A
Software and algorithms
Opentrons API v2.1	Opentrons	https://docs.opentrons.com/v2/index.html
Opentrons Simulator python module	Opentrons	https://pypi.org/project/Opentrons/
Opentrons App Version 4.7+ (Windows/Mac/Linux)	Opentrons	https://opentrons.com/ot-app/
Associated Python Protocol/ labware .json files/ csv	This paper	https://doi.org/10.5281/zenodo.7564523
Custom-made script for analysis of 3D adipogenic differentiation images	Dohmen et al.1	Available on request
Python Version 3.9.8	Python	https://www.python.org/
CellReporterXpress	Molecular Devices	N/A
Leica Application Suite X Version 4.4.0.24861	Leica	N/A
Other
Confocal Leica Stellaris 5 Microscope	Leica	N/A
Opentrons OT-2R Liquid Handler	Opentrons	SKU:999-00111
IX Pico IXP-38-570-5078 High Content Analyzer	Molecular Devices	N/A
Multi-Channel Electronic Pipette p20 (Gen2)	Opentrons	SKU: 999-00005
Multi-Channel Electronic Pipette p300 (Gen2)	Opentrons	SKU: 999-00006
2 well SBS polypropylene reservoir	Agilent	CAT# 203852-100
96 pyramid bottom SBS reservoir	NEST	CAT# 360103
Falcon® 96-well Black/Clear Flat Bottom TC-treated Imaging Microplate with Lid	Thermo Fisher	CAT# 353219
MicroAmp™ Optical 96-Well Reaction Plate Applied Biosystems™	Thermo Fisher	CAT# N8010560
96-well aluminum block	Opentrons	CAT# 999-00028
Spetec 56 laminar flow box	Spetec	N/A
100 μm cell strainer	pluriSelect	CAT# 43-50100
40 μm cell strainer	pluriSelect	CAT# 43-50040-51
T175 Flasks	Thermo Fisher	CAT# 159926

Materials and equipment

Reagents

Gelation buffer

Reagent	Final concentration	Amount
CaCl2 (147.01 g/mol)	66 mM	9.70 g
HEPES (238.30 g/mol)	10 mM	2.38 g
Type 1 water	N/A	1 L
NaOH (10 M)	N/A	N/A
Total	N/A	1 L

Store at room temperature for up to 6 months.

Coating solution

Reagent	Final concentration	Amount
Poly-D-Lysine hydrobromide	50 μg/mL	0.4 mL
Borate Buffer (1 M)	50 mM	2 mL
Type 1 water	N/A	37.6 mL
Total	N/A	40 mL

Prepare fresh.

1% Sodium Alginate

Reagent	Final concentration	Amount
Sodium alginate, high viscosity	1%	1 g
Type 1 water	N/A	100 mL
Total	N/A	100 mL

Store at 4°C for up to 4 months.

Adipogenic Differentiation Medium [ADM]

Reagent	Final concentration	Amount
DMEM/F12	N/A	410 mL
D-(+)-Galactose	17 mM	10 mL
Sodium pyruvate	10 mM	5 mL
Calcium Chloride dihydrate	2 mM	1 mL
Insulin recombinant human	10 μg/mL	500 μL
Hydrocortisone	100 nM	365 μL
L-Ascorbic acid 2-phosphate	227 μM	572 μL
Putrescine dihydrochloride	56.7 M	25 μL
Chemically Defined Lipid Concentrate	0.1%	500 μL
Indomethacin	5 nM	1 mL
Water [Type I]	15%	75 mL
Total	N/A	500 mL

Store base at 4°C for up to 2 weeks, add Indomethacin on day of the experiment.

Serum-Free Growth medium [SFGM]

Reagent	Final concentration	Amount
DMEM/F12	N/A	495 mL
Chemically-defined FBS replacement	1%	5 mL
Total	N/A	500 mL

SFGM stability depends on medium used.

Ammonium-Chloride-Potassium [ACK] Erythrocyte lysis buffer

Reagent	Final concentration	Amount
Type 1 water	N/A	800 mL
Ammonium chloride	15 mM	8.02 g
Potassium bicarbonate	10 M	1 g
Disodium EDTA	0.1 M	0.0372 g
NaOH (10 M)	N/A	N/A
Total	N/A	1 L

Adjust the pH to 7.2–7.4, add Type 1 water until the volume is 1 L.

CRITICAL: Paraformaldehyde is moderately toxic by skin contact and inhalation. Protective gloves should be worn when handling the solutions and all steps using it should be performed under a fume hood.

Cell source: For the presented protocol, we primarily tested the use of bovine fibro-adipogenic progenitors (FAPs), but as 3D hydrogel cultures proved to be efficient for other adipogenic progenitor cell types and species^2,4 we do foresee the method being relevant for additional cell types and species. For primary cells, we recommend using cells between passage 1 to 3 for optimal differentiation.

Growth medium: Appropriate growth medium for the cell type and species used should be applied and optimized for growth and purity accordingly.

Differentiation medium: Appropriate medium for the cell type and species used should be applied and optimized by the researcher. Especially the need for serum or specific inducers highly depends on the species, the growth medium, and the origin of the cells and needs to be determined accordingly.

Hydrogel: The best results were found when bovine FAPs were embedded in 0.5% of high viscosity, non-functionalized alginate, but different percentages, viscosities, or functionalizations can be advantageous for specific applications. These may need some adaptations in the liquid handling steps (like different aspiration and dispensing speeds for high viscosity gels), please refer to the troubleshooting section on how to adapt these parameters in an efficient way. Different hydrogel types may also work in this system but will need adaptations regarding the gelation buffer and the plate coating.

Equipment

Imaging equipment

Confocal microscope: TCS SP8, Leica Stellaris 5 using a 10×/2.00 objective lens and 4 μm Z-steps.

High content analyzer: IX Pico IXP-38-570-5078 with HC PL FLUOTAR 10×/0.32 objective, Filters: DAPI 350-390 / 419-482; FITC 445-485 / 509-539; Brightfield.

Alternatives: Any standard confocal microscope and high content analyzer can be used for the shown applications.

Liquid handler and autoclaved reservoirs

• •
  Opentrons OT2-R or OT-2 unit equipped with:

  • ○
    Opentrons p300 8-Channel Electronic Pipette gen2.
  • ○
    Opentrons p20 8-Channel Electronic Pipette gen2.

Additionally, the plexiglass that the OT-2R unit comes pre-installed with was was removed after placing the unit into the laminar flow hood, both for airflow purposes and for ease-of-access by scientists setting up the liquid handler for usage.

• •Spetec 56 laminar flow box (Figure 1).
• •A standard Windows-based laptop with an i5 4th Generation Intel Core Processor is connected to the OT-2 unit via USB or WiFi (almost any modern laptop would be sufficient).
• •Opentrons 20 μL and 200 μL filter tips.
• •
  MicroAmp™ Optical 96-Well Reaction Plate Applied Biosystems™ (N8010560) placed on top of a 96-well aluminum block from Opentrons.

  • ○
    Alternatively: Standard PCR reaction tubes.
  • ○
    Alternatively: 96-Well PCR Plates from Biorad (HSP9601) (no aluminum block required).
• •Agilent 2 well SBS polypropylene reservoir (203852-100).
• •NEST 96 pyramid bottom SBS reservoir (pn 360103).
• •
  96 well imaging plates e.g., Falcon® 96-well Black/Clear Flat Bottom TC-treated Imaging Microplate with Lid (Falcon, 353219).

  • ○
    Alternatively: Any similar microplate can be used.

Alternatives: All of the above materials could be replaced by similar materials, provided proper labware definitions are made using the Opentrons Labware Creator/pre-loaded by the Opentrons Labware Library. Other liquid handlers could also of course be utilized for this method. For example: if liquid level detection was deemed important for the user’s specific workflow or if another liquid handler is already owned; the main concepts of this method would still apply.

Figure 1.

Opentrons OT-2 unit in Spetec 56 laminar flow box

Setup used for sterile cell culture work.

Software

Programming

Python version 3.9.8 was used to code these protocols (via Microsoft Visual Studio Code), utilizing commands from the Opentrons API Version 2.1. Modifications to this protocol can be relatively easily made using any code editing application. Simulation of any Opentrons protocols before use is highly recommended; further instructions on how to do so can be found in the github repository.

Python protocol

Code for the Python protocol can be found on the provided gitLab repository (https://doi.org/10.5281/zenodo.7564523), as well as custom labware definitions as a JSON file made via the Opentrons custom labware creator and a template CSV file that needs to be uploaded to the Raspberry Pi unit on the OT-2 via the Opentrons App through Jupyter Notebook.

Opentrons Application

The official Opentrons Application was used to load protocols, perform calibrations, and upload csv files to the liquid handler via Jupyter Notebook. This protocol was coded using v2.1 of the Opentrons API. While this protocol was coded with Opentrons version 5.0 of the Opentrons Application / Robot Software in mind, other versions of the application would also suffice.

Image analysis software

We used the following software for the acquisition and analysis of HCA and confocal pictures, respectively. In general, image analysis can be performed with standard imaging software suitable for the microscope used.

CellReporterXpress® by Molecular Devices® (Version 2.9.1) was used for the analysis of the HCA acquisition.

Leica Application Suite X Version 4.4.0.24861 was used for the acquisition of confocal images with the Leica Stellaris 5 confocal microscope.

Step-by-step method details

We here describe an automatable model for adipogenic differentiation, suited for high throughput screening. In short, sheets are made by pipetting 5 μL of a cell/alginate mixture in the middle of a PDL coated well of 96-well plate. Polymerization of the alginate is initialized by a short wash with a gelation buffer. Differentiation medium is added, and medium is exchanged weekly over a four-week period. At the end of the differentiation period, cells are fixed, stained with a lipid stain and the differentiation efficiency is assessed using a high content analyzer (Figure 2).

Figure 2.

Schematic overview of the protocol workflow

This scheme gives an overview of the different steps performed to perform alginate sheet differentiation. The protocol starts with coating the plates with poly-D-lysine (PDL) and making the sheets. Differentiation medium is added on day 0 and changed once a week. After 4 weeks, sheets are fixed, stained and analyzed with a high content analysis device.

Preparing cell/alginate sheets

Timing: 5 min hands-on time and 15 min run-time per 96-well plate

Preparing the alginate sheets is the most critical step for the success of the protocol described here, though the general steps are simple: cell/alginate mixture is transferred to the center of the well using a p20 8-channel pipette, then the sheets are washed with a gelation buffer and excess gelation buffer is washed away with cell culture media. For a graphical representation, please refer to Methods video S1 and for a step-by-step manual for the set up please refer to Methods video S2.

• 1.
  Prepare needed reagents.

  • a.
    Gelation buffer.
  • b.
    DMEM/F12 with 1 mM CaCl₂ added.
• 2.
  Edit Python file for run variables.

  • a.
    Change NUM_PLATES to the number of plates you have.
  • b.
    Edit staring tip column.
  • c.
    Edit csv: Insert column-wise the sheets’ volume and sample number per well (in the format: [<volume> - <sample name>] and upload the edited csv to Jupyter Notebook of the Opentrons liquid handler.
• 3.
  Load protocol on the Opentrons App.

  • a.
    Ensure that a Opentrons p300 8-Channel Electronic Pipette (GEN2) and a Opentrons p20 8-Channel Electronic Pipette (GEN2) are both loaded.
  • b.
    Place sterile labware on Opentrons as directed by the app (Figure S1B).
  • c.
    Fill required volumes of the reagents in the correct locations as specified by the “run” screen.
• 4.
  The liquid handler steps performed are as follows:

  • a.
    Mix the cell/alginate mixture 3 times with the p20 8-channel pipette.
  • b.
    Transfer the specified volumes/samples to the middle of the wells.
  • c.
    Slow wash with gelation buffer (a delay of 5 s per well incubation time before removal is added to ensure complete gelation). The solution is dispensed at the side of the well with a speed of 12.5 μL/s (8 times slower than standard speed) to not disturb the nascent sheets.
  • d.
    Wash 2× with DMEM/F12.
• 5.
  Addition of the differentiation medium.

  • a.
    Remove last DMEM/F12 wash and add the prepared differentiation medium to the wells in the appropriate volume (15 × the volume of the alginate sheets, e.g.: a 5 μL alginate sheet = 75 μL of media) (Step can be performed manually or via liquid handler (not described)).
  • b.
    The sheets are cultured in a humidified atmosphere at 37°C with 5% CO₂.

CRITICAL: During this step, pipette dispense locations and heights are of the utmost importance to achieve reproducibly well formed alginate sheets (troubleshooting Section: “Wrong dispense height or offset during sheet making”). Additionally: Alginate dissolves when in contact with Phosphates, so it is the utmost importance to avoid any contact with Phosphate Buffer Solution (PBS). To maintain the stability of the sheets over the culture period, it is strongly advised to add 1 mM CaCl₂ to the wash buffers and the differentiation media.

Alternatives: All steps described can be done as well manually. To ensure that sheets are placed in the middle of the well, we suggest using a single channel pipette for dispensing the cell/alginate mixture, instead of the 8-channel option used on the liquid handler. For gelation and washing, a multichannel pipette can be used and is best placed at the sides of the wells to ensure that the liquids flow slow and constant over the cell/alginate mixture to not disturb the sheet before proper gelation takes place. As soon as the alginate gelated, sheets are stable and no special precautions have to be taken. In our experience, they can be treated like standard 2D cultures from this point onwards.

Methods video S1. Schematic representation of the sheet making procedure, related to step 4

This video shows a schematic view on the cell/alginate mixture dispensation and the gelation process.

Methods video S2. Video description of the liquid handler set up, related to step 3

Step by step description on how to set up the liquid handler for the sheet making protocol.

Culture maintenance

Timing: 10 min/plate, every 7 days over the 4 week culture period

During the culture maintenance, differentiation media is exchanged once a week.

• 6.
  Prepare solutions.

  • a.
    Differentiation media (specific to the cell’s culture requirements).
• 7.
  Edit Python file for run variables.

  • a.
    Change NUM_PLATES to the number of plates you have.
  • b.
    Edit the starting tip variable for the p200 filter tips used (for example: if you have tips in columns 5–10 in your tip rack, you need to set this to A5).
  • c.
    [Optional] Adjust liquid volumes for the media exchange if needed.
• 8.
  Load protocol on Opentrons App.

  • a.
    Place labware on the OT-2 as directed by the app (Figure S1C).
  • b.
    Fill required reagents in the correct locations from the run screen.
• 9.
  The liquid handler steps performed are as follows:

  • a.
    Remove old media from well plates.
  • b.
    Add new media to well plates.

Fixation and staining

Timing: 10 min hands-on time and 2 h run time

Successful adipogenic differentiation can be assessed by staining the lipid droplets and analyzing the percentage of positive and negative cells and the total lipid volume produced.

To get an accurate measurement of size and volume of the lipid droplets, high resolution confocal images are recommended as the acquisition of several z-stacks enables volumetric analysis. When less accuracy but higher throughput is needed (for example when using alginate sheets for screening purposes), alginate sheets can also be analyzed with a standard high content analyzer without taking z-stacks. When taking cell number, percentage of positive cells, total area stained, and cell integrated intensities for the lipid marker as main readouts the results are robust enough to judge the overall differentiation efficiency.

• 10.
  Prepare solutions.

  • a.
    Fixation solution (4% PFA in gelation buffer).
  • b.
    Staining solution (Hoechst (1:1000) and Bodipy (1:1000) in gelation buffer.
• 11.
  Edit Python file for run variables.

  • a.
    Change NUM_PLATES to the number of plates you have.
  • b.
    Edit staring tip column variable.
  • c.
    [Optional] Adjust liquid volumes and incubation times.
• 12.
  Load protocol on Opentrons App.

  • a.
    Place labware on the liquid handler as directed by the app (Figure S1D).
  • b.
    Fill required reagents in the correct locations.
• 13.
  The liquid handler steps performed are as follows:

  • a.
    Add fixation solution to the well plates.
  • b.
    Incubate for 15 min.
  • c.
    Remove fixation solution.
  • d.
    Add staining solution.
  • e.
    Incubate 60 min.
  • f.
    Remove staining solution.
  • g.
    Wash with gelation buffer.
  • h.
    Final addition of gelation buffer to well plates.
• 14.Image acquisition and analysis (refer to the Quantification and statistical analysis).

CRITICAL: Alginate dissolves when in contact with phosphates, so it is of the utmost importance to avoid any contact with Phosphate Buffer Solution (PBS). To maintain the stability of the sheets while staining, dilute both the PFA and staining solution in 66 mM CaCl₂ (referenced above as gelation buffer).

Note: PFA is moderately toxic upon skin contact or inhalation, so all steps containing PFA must be performed under the fume hood and protective measures (gloves, safety goggles) should be taken. For the fixation and staining protocol, we have a liquid handler unit installed under a fume hood. In case this is not possible, we recommend performing fixation manually or using a non-toxic fixative.

Expected outcomes

The expected result of this procedure is uniform alginate sheets in all wells of a 96-well plate (Figure 3 left panel). It should be noted that some variation of the size of these alginate sheets is normal and does not hinder differentiation in general. However, when an alginate sheet blobs out into the outer portion of the well, differentiation, as well as correct quantification, can be hindered.

Figure 3.

Typical sheet forms that can occur

Representative pictures of alginate sheets (after 4 weeks of differentiation). Blue: Hoechst, Green: Bodipy. Scale bar, 500 μm. Left panel: When everything is set up correctly, we expect the alginate sheets to form as a round sheet in the middle of the well, not touching the edge of the well. Some variation regarding the thickness and cell density within this sheet cannot be avoided (here visible by the blue spot in the middle of the alginate sheets) but does not hinder differentiation and analysis. Middle panel: When too little volume is dispensed or cells do not distribute equally, differentiation can be hindered, and the results cannot be reliably compared to each other. Right panel: If the sheet contorts to the side of the well, the increased thickness of the cells/hydrogel mixture can reduce differentiation efficiency and make the analysis less reliable (due to well edge effects and incomplete capturing of the differentiated cells).

After culturing the alginate sheets for four weeks as described above with an initial addition of differentiation media and weekly media exchanges, followed by fixation with PFA / staining with bodipy, mature adipocytes (usually characterized by unilocular lipid droplets) within the hydrogel matrix are expected. This method is intended to help optimize the culture parameters needed to differentiate and mature adipocytes, and to find new compounds influencing adipogenesis.

To validate that the alginate sheet system is representative of the differentiation in hydrogel microfibers, we ran several experiments where cells were differentiated in sheets and fibers in parallel with different media additives known to have positive or negative effects on the adipogenesis in 3D, including some additives that fail to show the same effect in 2D. These experiments showed (1) that differentiation in both systems is comparable regarding the fat volume per cell and the mean fat droplet size, and (2) that compounds affecting differentiation in the fiber culture affected cells in the sheet culture to a similar extent (Figure 4). To further investigate the usefulness of the protocol in regard to a potential high throughput application, we did an initial test with a small number of compounds tested in 96-well format and assessed the differentiation efficiency with a high content analyzer. The results showed three medium additives with a negative effect on the differentiation and one with a statistically significant positive effect (Figure 5). This result demonstrates that alginate sheet differentiation is a versatile tool to screen a large number of media compositions, as this culture system is easy to handle and can be analyzed in a high-throughput manner, reducing the time and effort for data acquisition and analysis.

Figure 4.

Comparison of 3D microfiber to 2.5D alginate sheet differentiation

(A) Confocal pictures of both cell/alginate mixtures after 28 days of differentiation. Both show a similar efficient differentiation in the positive control (CNT), with nearly all cells differentiating, and detrimental reduction of the differentiation in the negative control (1% solvent). Blue: Nuclei stained with Hoechst, Yellow: lipid droplets stained with Nile Red, Scale bar 100 μm.

(B) When zooming in within both structures, one can see the very comparable size and shape of formed lipid droplets. Scale bar 100 μm.

(C) Quantification of the fat volume per nucleus in both systems shows comparable trends when cells were treated with different compounds.

(D) Measuring the lipid droplet size showed similar diameter in both systems. For quantitative analysis, we analyzed 3 biological replicates (cells coming from different FAP isolations) and took 25 pictures per sample (magnification 10×, zoom 2×, z-step size of 4 μm) to capture the full sample. Total fat volume per nucleus and mean fat droplet diameter were determined using a custom python script.

Figure 5.

Small-scale compound screen

We performed a small-scale compound screen as proof of concept that alginate sheets can be used for high-throughput screenings. For this, we made sheets in 96-well microplates (96-wells per FAP isolation) and added different compounds in different concentrations to the differentiation medium for the first week of differentiation (100% = standard concentration used, 50% = half of the standard concentration, 200% = double of the standard concentration). Medium was changed weekly, and plates were fixed with 4% PFA after 4 weeks of differentiation. For analysis, nuclei were stained with Hoechst and lipid droplets with Bodipy. Acquisition was performed with a high content analyzer (10× magnification, no z-stacking), and internal analysis scripts were used to determine the number of cells, the percentage of positive and negative cells, and the total area stained for the lipid marker. Area per cell was calculated by dividing the total area by the number of nuclei detected.

(A) Plate layout with numbers standing for the compound added per well.

(B) Exemplary thumbnail picture of the analyzed lasagna. Blue = Hoechst, Green = Bodipy.

(C) Percentage of positive cells per compound. The gray bar represents the mean of all replicates, each dot stands for one analyzed well and the color code indicates the FAP isolation used). ∗ indicates compounds that differ significantly from the control (1 - CNT). Differences were tested with student t-test and p < 0.5 is considered significant.

(D) Area stained for lipid marker normalized to the number of cells per well.

Quantification and statistical analysis

For these studies CellReporterXpress® by Molecular Devices® (Version 2.9.1) was used for the analysis of the HCA acquisition. We applied the 2-channel assay using a nuclei marker and a marker to identify differentiated cells to determine cell number and percentage of positive cells and applied an adapted single channel assay for counting cells to determine the total area and droplet number based on the lipid stain.

• •The code for the analysis of 3D adipogenic differentiation images acquired with the confocal microscope is available from the authors upon request (for academic use only).
• •For data visualization and statistical analysis, we used a custom-made Python script.

Limitations

This method is not suitable for any procedure with the goal of harvesting large amounts of mature adipocytes, i.e., production, analyses that require large amounts of cells, or fat content analysis, due to the low volume of cells embedded in alginate. Likewise, due to the cells being embedded in alginate with no to very little contact, this method is not suitable for studying cell-cell interactions. Compared to traditional 2D differentiation, more cells are needed, and results obtained from sheet differentiation are not always predictive for the differentiation in 2D. Therefore, this method is intended as a screening system to test different media compositions for 3D differentiation.

As this method is primarily intended to screen larger amounts of compounds that promote adipocyte differentiation in a 3D environment, and for the endpoint to be a static image analysis, we only optimized the procedure for 96- and 384-well formats. For a proof of concept, we tested the making and analysis of a sheet in a 6-well microplate with a cow-shaped mold using commercially available bovine fat embedded in alginate (Figure S3). While the concept is transferable, larger sheets are less stable and tend to not stick to the well plate in a comparable manner to a smaller format. Further optimization would be necessary but is outside the scope of the presented method.

Troubleshooting

Problem 1

Malformed or no alginate sheets dispensed.

The most frequent issue occurring in this protocol is the incorrect formation of an alginate sheet (Figure 5). This can occur in the step “preparing cell/alginate sheets”. This can have multiple causes and we present various potential troubleshooting methods.

Potential solution

Different common mistakes during the protocol set up can lead to this problem.

• •Remaining liquid on the output plates when coating.

Thorough removal of the gelation buffer solution and completely dried plates during the coating step are critical for the success of the production of stable alginate sheets. To achieve this, the aspiration height must be adjusted accordingly: too high will leave liquid in the plates; too low will make the tips touch the bottom of the plate and prevent complete liquid removal. We therefore suggest a height of 0.5 mm–1 mm above the bottom of the well. We recommend a critical check of the specific labware being used as well as robot specific inaccuracies, and to perform some initial calibration runs first to ensure that the heights are fitting the individual set-up.

• •Wrong dispense height or offset during sheet making.

Dispensation height when dispensing the cell/alginate mixture needs to be exact and adapted to the solution’s viscosity (refer to step: preparing cell/alginate sheets). We found 0.6 mm above the bottom of coated 96 well plates (Falcon®, Cat.#. 353219) works best, but depending upon the user’s plates, calibration settings, and the cell/alginate mixture viscosity some tweaking will be necessary. This can be done by applying different dispensation heights within a Python loop as seen in Figure S2. Additionally, the pipet offset must be correct: if the electronic pipette is not calibrated directly in the middle of the 96 well plate, the droplet may be dispensed too close to the side and then form a blob.

• •Dispense speeds.

Having proper dispensing speeds within the protocol is essential to ensure the correct formation and maintenance of the alginate sheets, especially during the gelation step. We have observed that when using the standard dispense speed (≈ 100 μL/s) the not yet gelated cell/alginate mixture can be flushed to the side of the well. But if the dispense speed is too slow the gelation buffer can form large droplets that may disturb the sheets as well due to the impact of the droplets. Therefore, we determined an optimal speed (12.5 μL/s) that leads to a slow but constant flow of the liquid into the well, enabling equal gelation.

• •Bubbles present within FAP/hydrogel mixture in the PCR plate.

If there are bubbles present within the PCR plate’s liquid, the liquid handler may not successfully aspirate the cell/alginate mixture. Proper mixing is needed to prevent this issue, please refer to the instructional setup video for more details (Methods video S2).

• •Pipette Alignment:

If the liquid handler’s 8-channel pipettes are misaligned or tilted, any assays performed using that pipette head will likely not function as intended. Be sure to follow the Opentrons instructions for leveling an 8-channel pipette. A misaligned pipette will impact washing steps and alginate sheet making, so this is an essential step.

Problem 2

Poor differentiation.

Poor differentiation can have a variety of reasons. It goes without saying that high quality of cells and an appropriate differentiation medium for the cell type and species used are a prerequisite for success and independent of the method used. Still, some factors can hinder reliable differentiation with the method described here, mainly referring to the media volumes. Poor differentiation can arise from having inappropriate media volumes on top of the alginate sheets. If there is too little media, the cells can dry out and die, if there is too much media, the cells can have insufficient oxygen exchange and suffocate. We recommend using the media volumes stated above or carefully evaluate adaptation.

But even if correct volumes are used, evaporation can cause issues; see also the “culture maintenance” section.

Potential solution

Some wells in the well plate may evaporate faster than others in the incubator, causing unreliable maturation of the adipocytes. This can be troubleshot by sealing the plates using parafilm, or by ensuring the humidity levels in the incubators are consistent with specifications for the cell types you are culturing.

In case of doubt, we recommend having controls scattered throughout the plate, to ensure comparable differentiation on different locations of the well.

Resource availability

Lead contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, E. M. Mall (eva@mosameat.com).

Materials availability

This study did not generate new unique reagents.

Data and code availability

• •The code for the analysis of 3D adipogenic differentiation images acquired with the confocal microscope is available from the authors upon request (for academic use only).
• •For data visualization and statistical analysis, we used a custom-made Python script (https://doi.org/10.5281/zenodo.7564523).