Microbioreactor-Based Production of Anchorage-Dependent Mesenchymal Stromal Cells Primed for Acute Respiratory Distress Syndrome

Summary

A protocol is described for microcarrier-microbioreactor production of anchorage-dependent cells as potential cell-based therapies. Targeting a use case of acute respiratory distress syndrome, comparable cell yield and favorable gene expression relative to flask-based process conditions are illustrated. This protocol is extensible to other anchorage-dependent cells and target indications.

Abstract

Cell-based therapy is an emerging therapeutic approach for a range of health conditions. However, producing sufficient quantities of quality cells remains an active scientific and technical challenge. The final cell product cannot be sterilized; therefore, sterility and quality-by-design considerations are paramount. For cells that are anchorage-dependent, flask-based expansion is still utilized at the development and industrial scales, requiring large physical footprints, limiting control of the biochemical environment (pH, CO₂ and O₂, diffusional gradients of nutrients and metabolites), and providing physical cues that can affect final cell quality. These challenges have motivated the adoption of bioreactor-based manufacturing approaches, but such approaches typically proceed at volumetric scales that confer challenges and can be cost-prohibitive for exploratory benchtop research or process development. Microbioreactor-microcarrier cell therapy production has been reported to provide biochemical control during culture for a test case of cell source type - mesenchymal stromal cells. For initial assessment of cell quality in vitro, acute respiratory distress syndrome was considered as a potential use case. The approach illustrated here enables rapid process development at benchtop scale due to the small (2 mL) volume and tight biochemical control in the microbioreactor. To achieve this for an anchorage-dependent cell type such as mesenchymal stromal cells required substantial modification of the microbioreactor process and utilized a gelatin microcarrier platform produced in-house through a microfluidic process. Here, additional detail is illustrated for the method and process parameters used to culture such anchorage-dependent cells in the microbioreactor, with the intention to enable other researchers to adapt these practices for their own use cases and cell source types.

Introduction

The use of living cells as a therapeutic product has recently emerged as a promising new treatment option for several health challenges. For example, chimeric antigen receptor (CAR) T cells have brought durable remission to those with so-called liquid cancers — leukemia, lymphoma, and multiple myeloma — that were refractory to all previously available treatments¹. More recently, a mesenchymal stem/stromal cell (MSC) product (Ryoncil) was approved to treat children suffering from steroid-refractory acute graft-versus-host disease (SR-aGvHD), a condition with > 50% mortality². But extension to other unmet clinical needs requires intensive development of both the therapy and the means to develop and manufacture that product.

One such unmet clinical need can be found in individuals suffering from acute respiratory distress syndrome (ARDS). ARDS is discussed here to provide a specific use case to illustrate the method, results, and discussion of possible extensions to other disease targets. A critical (progression in hours to days), prevalent (> 200,000 annual cases in the US), and deadly indication (27%-46% mortality, depending on the severity of presentation), ARDS drew renewed and widespread attention during the COVID-19 Pandemic as the typical mechanism of death for patients infected with SARS-CoV-2^3,4. ARDS is characterized by increased pulmonary vascular permeability leading to interstitial edema, alveolar flooding, and infiltration of cellular components, including neutrophils and red blood cells, into the alveolar compartment⁵. In the initial exudative phase of ARDS, innate immune-cell mediated inflammation drives these pathogenic results by damaging endothelial and epithelial barrier integrity and promoting immune cell recruitment and activation through secretion of pro-inflammatory cytokines; successful resolution of the inflammatory response, repair of barrier integrity, and avoidance of excessive fibrosis are critical for survival³. While anti-viral therapies such as Remdesivir have shown promise in combating SARS-CoV-2-mediated ARDS, for other etiologies, there are currently no approved pharmacological interventions available that reduce mortality⁵.

MSCs may offer a promising approach to treating ARDS, given their lung-specific tropism upon systemic injection and their ability to secrete immunomodulatory paracrine signals into their surrounding microenvironment^6,7. Mechanistic studies have implicated expression of such paracrine factors as interleukin-10 (IL-10)⁸, indoleamine 2,3-dioxygenase 1 (IDO1)⁹, prostaglandin E2 (PGE2)¹⁰, and others in driving putative MSC therapeutic potency for multiple indications, including acute lung injury (ALI)/ARDS¹¹; therefore, expression levels of these gene products represent potential critical quality attributes (pCQAs) that may be predictive of clinical efficacy, providing tractable release criteria for the final cell product during manufacture. However, producing sufficient quantities of cells to reach an effective therapeutic dose is challenging; Ryoncil, for example, is indicated for dosing at 2 x 10⁶ cells/kg of patient mass, administered twice weekly for 4 weeks¹². For an 80 kg patient, this equates to > 1.2 billion MSCs for a single course of treatment, necessitating a large physical footprint to house sufficient surface area for this 2D growth condition. Given that this living cell product cannot be sterilized prior to patient administration without harming the cells' therapeutic potential, utilization of current Good Manufacturing Practice guidelines for production in a sterile fashion is paramount, adding to the logistic complexity of cell manufacture and thus the cost of goods¹³.

Further, prior studies have shown evidence that typical MSC production in tissue culture polystyrene (TCPS) flasks contributes to the onset of intra-population phenotypic heterogeneity. For some end uses, this renders quality control efforts between batches challenging in research and in clinical translation^14,15,16,17. It is speculated that the limited biochemical control available in TCPS flasks, including diffusional gradients of gases, nutrients, and metabolic waste products, poor pH regulation, etc., contributes to these challenges.

These issues motivate the adoption of bioreactor-based culture to leverage the superior biochemical regulation and cell density available in a controlled three-dimensional growth format with dynamic mixing¹⁸. However, bioreactor-based processing or manufacturing comes with its own set of challenges: as an anchorage-dependent cell type, MSCs require some form of substrate upon which to grow in a bioreactor, necessitating the addition of microcarriers. Additionally, many commercially available bioreactors operate at volume scales that are prohibitively resource-intensive (> 3 L of growth media) for the kind of exploratory benchtop research needed to identify and refine unknown process parameters that affect expression of measurable indicators of cell quality. These features are described as potential critical quality attributes or pCQAs¹⁹.

Krupczak et al. previously reported a microcarrier-microbioreactor approach that produced MSCs with improved pCQA expression for ARDS, limited to in vitro assessments²⁰. This approach combined the use of a dissolvable gelatin microcarrier platform produced in-house with a commercially available microbioreactor platform, the Mobius Breez microbioreactor²¹. This microbioreactor utilizes 2 mL single-use disposable reactor chambers and incorporates many desirable features for improved biochemical control, including regulation of temperature, pH, CO₂, and O₂ in a dynamically mixed, closed-loop perfusion-fed system, but was not designed with anchorage-dependent cells like MSCs in mind. Indeed, the standard microfluidic channels used to inject reactor contents are too small to allow passage of microcarriers, necessitating substantial adaptation to enable successful production of MSCs in this platform.

Here, a detailed protocol is described and illustrated for the purpose of adapting this microbioreactor for the anchorage-dependent culture of MSCs. The process parameters reported herein were identified with the intention to inform process development and manufacturing of MSCs for pre-clinical, in vitro assessment of therapeutic efficacy for ARDS. However, in principle, this protocol can be adapted for other end-user applications, other inflammatory or immunomodulatory-related conditions, and other anchorage-dependent cell types that could benefit from enhanced biochemical control in a benchtop-scale setting. Figure 1 schematizes the protocol described in the accompanying video, and Figure 2, Figure 3, and Figure 4 illustrate key results.

Protocol

1. Microbioreactor setup and preparation

1. Preparation of the required number of microbioreactor consumables
   1. Ensure the biosafety cabinet (BSC) is on and sterile. Remove the foil pouch covering the new consumable by puncturing the top of the bag and retaining the foil bag until the conclusion of this experiment (to record details of lot# in case of defect). Open the exterior blister packaging and discard the hydration pack contained within. Discard the exterior blister packaging.
   2. Spray the interior blister packaging with 70% ethanol and bring it into the BSC. Open the interior blister packaging and discard.
   3. Using a sterile syringe, remove the plunger and connect the nozzle to a 0.22 µm sterile filter. Pour 50 mL of unsterilized ultrapure water into the barrel of the syringe. While holding the syringe over a waiting, sterile 50 mL tube, re-insert the plunger into the syringe and inject the sterile filtered water into the waiting collection tube.
   4. Remove the clear plastic cap connecting to Bottle LL in the bottle rack. Using a new sterile 50 mL syringe, remove the plunger and connect the nozzle to Bottle LL. Pour the sterile filtered water into the syringe barrel and inject it into bottle LL.
      ​NOTE: The needleless connection ports can become stuck during the manufacturer's sterilization procedure, so you may need to connect and disconnect the syringe several times in order to break the seal to the bottle and successfully inject the contents. Alternatively, you can use a tube welder to connect a source bottle to the weldable tubing in a sterile fashion for the appropriate bottle.
   5. Using this same procedure, fill bottle L with 25 mL of non-FBS-supplemented low glucose DMEM, bottle R with 25 mL of sterile PBS, and bottle RR with 8 mL of anti-adherence rinse solution. Use a new sterile syringe for each liquid.
   6. Repeat this procedure for as many consumables as required in the experiment.
2. Perform a POD self-test in each POD being used for this experiment.
   NOTE: The self-test should complete with no issues; if a component of the self-test fails, address the failure before continuing.
3. Remove the self-test cassettes to their light-protected storage bags.
4. Mount the consumable to the bioreactor as described below.
   1. Remove the green tape from the valves of the consumable and mount it into position in the POD. Reinstall the clamps and ensure no gas/air can be heard hissing as it escapes.
   2. Connect the white bottle rack pressure sensor line, the red air pressure line, and the blue vacuum line between the consumables and the base station/POD. Ensure the side injection port is clamped shut with a white teardrop clamp.
   3. Follow the software steps to initiate a new experiment. These culminate with the autoprime and calibrate feature, which takes ~4-5 h and can run overnight.
5. Label each bottle rack with its POD location so it is easy to distinguish which POD to return the consumable to after it has been removed from the bioreactor. Use colored tape to indicate the position: Pod0 = red, Pod1 = orange, Pod2 = yellow, Pod3 = green. Write the name of the experiment and the date on the tape.
6. Leave the consumables to calibrate overnight. Ensure the media can enter the reactor and exit into the waste line.
7. If planning to use cells stored in liquid nitrogen, thaw and plate them so they have > 24 h to recover from cryo storage prior to injection.

2. Recalibration and microcarrier preparation

1. Ensure the autopriming and calibrating steps have been completed successfully.
   1. If calibration has not been completed successfully, identify and fix any root cause issues and then attempt a recalibration. Common issues include neglecting to remove the white bottle rack clamp, neglecting to remove the green tape from the consumable before placing it in the POD, and insufficient media volume in the media bottle.
2. Ensure the dissolved oxygen sensors are stable.
   1. Nearly 10% of consumables will exhibit sensor drift in the DO sensor after calibration. Perform the preferred corrective action, which is to re-run the full calibration. If required, perform one of the following.
      1. Ensure there is no air in the cassette, then uncheck the Display plots checkbox to show controls, then click Autocalibrate DO and pH.
         1. If air was present in the reactor when a recalibration is needed, return to the first page of the MBS dashboard by clicking on the Blue left arrow in the top left corner of the window for the specific POD in question. Click on the Button to engage manual ops. Check the Waste tube ready box and enter W in the text entry box to select the waste bottle. Click the Eject-wash button, then click Yes on the pop-up window. This will empty the reactor chambers and refill them with media from the media source bottle.
3. Prepare the correct number of microcarriers in 15 mL conical tubes, one for each condition.
   NOTE: Microcarrier number, aggregate surface area, and cell seeding density will depend on your application. For this protocol, it is recommended to inject an absolute number of microcarriers that results in 75 cm² of available microcarrier growth area for comparison to a T75 flask.
4. Allow microcarriers to settle in the tube; aspirate out as much supernatant as possible without removing microcarriers in order to make room for cells and fresh media.

3. Cell and microcarrier injection

1. Ensure the autoprime and calibrate features have completed successfully.
2. Click through the software window as if the reactor is being inoculated according to the standard suspension cell workflow.
3. Enter a note in the pop-up window to indicate that no injection of cells has occurred at this stage.
4. Engage manual ops to pause reactor mixing and add a white C-clamp to the bottle rack lines.
5. Remove each consumable and transfer to the BSC. Remove the spent media from waste and perfusion bottles and discard.
   1. Be sure to remove all waste from the large perfusion bottle as well as the smaller 1.5 mL tube contained within the perfusion bottle if you plan to harvest conditioned media at any point in the study.
6. Remove the non-serum-supplemented growth media from bottle L and discard. Rinse out bottle L with ~20 mL of sterile PBS and discard. Add 25 mL of serum-supplemented growth media to bottle L.
7. Return the consumables to their respective PODs. Put media source bottles on ice for the remaining duration of the experiment. Refresh ice daily.
   NOTE: Wet ice can be supplemented with freezer packs to prolong the duration of temperature control.
8. Wash out the reactors with PBS and flush the lines.
   1. Click the Flush bottles button, enable the Radio button for Bottle L, select Output to Waste, then click Start Flush MSCserum.
      NOTE: This will flush the newly added hMSC media through the line to ensure no more non-FBS-supplemented media enters the reactor.
   2. Click the Flush bottles button again, select the Radio button for Bottle RR, and start flushing. This will flush anti-adherence rinse solution through the line and ensure it is primed for entry into the reactor.
   3. In the Bottle Contents window, indicate in the software that MSCserum is located in Bottle R, which actually contains PBS.
   4. Go to manual ops and perform three eject-wash cycles on the waste bottle.
9. Add anti-adherence rinse solution to the reactor.
   1. Go back to the Bottle Contents window and indicate that MSCserum is located in Bottle RR, which actually contains anti-adherence rinse solution.
   2. Go to manual ops and perform one eject-wash cycle on the waste bottle.
      NOTE: The former PBS wash was to prevent any components in the DMEM from interfering with the anti-adherence wash; now, the reactor should be full of anti-adherence rinse solution with trace PBS.
10. After 30 min have elapsed with anti-adherence solution present, rinse out the consumable with PBS.
    1. In the Bottle Contents window, indicate in the software that MSCserum is located in Bottle R, which actually contains PBS.
    2. Go to manual ops and perform three eject-wash cycles on the waste bottle.
    3. Go back to the Bottle Contents window and indicate in the software that MSCserum is located in Bottle L, which contains hMSC media.
11. Empty the reactor to await cell and microcarrier injection. In the manual ops panel, click Reinoculate and leave the system at the stage where the dashboard indicates to inoculate the reactor.
    NOTE: Here, the term inoculate simply means injection of the cells intended for expansion in the microbioreactor.
12. Add the additional resistors to modulate mixing speed in line with the default R0 resistor.
13. Lift cells from their flask and count them on a hemocytometer. Perform all open operations in a BSC to ensure sterility.
    1. Aspirate the media, then wash with 5-10 mL of pre-warmed PBS. Aspirate the PBS and add 5 mL of trypsin + EDTA to the flask.
    2. Incubate at 37 °C for 5 min or until all cells have visibly lifted from the flask surface.
    3. Add 10 mL of pre-warmed hMSC media to neutralize the trypsin activity. Withdraw the liquid contents and deposit them in a 15 mL centrifuge tube.
    4. Centrifuge the cells at 300 x g for 5 min. Discard the supernatant and resuspend the cell pellet in an appropriate volume of hMSC media for counting. For a fully confluent T75 flask, a suitable volume is typically 2-4 mL, which will yield a cell concentration of ~1-2E6 cells per mL.
    5. Mix a sample of the resuspended cells with equal part Trypan Blue viability dye and load the mixture into a hemocytometer for counting. Count the cells according to the manufacturer's instructions.
14. Calculate the volume of cell solution needed to reach 150,000 cells and pipette that amount into the waiting 15 mL conical tube containing microcarriers for each condition. Top up the volume in the tube to be between 2.2 - 2.5 mL with fresh, warm hMSC media.
    NOTE: The volume in each 15 mL conical should be slightly greater than 2 mL such that the injection will result in a slightly overfilled reactor condition to avoid bubbles, which will be equalized in a later step.
15. Add the appropriate volume of cells to the T75 flask control.
16. Working with one to two consumables at a time, remove the consumables from the POD and bring them into the BSC.
17. Withdraw cells and microcarriers in a 5 mL syringe using green 4 inch blunt needles. Inject the cells and microcarriers into the reactor.
18. Return the consumable to the base station and repeat until all consumables have been injected.
19. Click through the Software Windows to indicate that the inoculation has taken place. When the software returns to the normal home page, remove the white clamp from the consumable output lines.
20. Enter manual ops and perform an Eject Excess function with output port set to P for perfusion. Engage static mixing mode, using 1680 s on and 120 s off. Keep the mixing frequency at 5 Hz for static mixing.
21. Turn off the optical density reference measurement.
    1. Open the MBS Client and click Connect. Enter the following command contained within the quotations: "dmem_write 0,1,2,3 F_measure_ODref_skip 1"
    2. Click Send, await confirmation that the command has been sent successfully, and close the MBS Client.
22. After 24 h, disable static mixing mode and re-enable normal mixing. Reduce the mixing frequency to 1.5 Hz.

4. Microbioreactor operation

1. Monitor the microbioreactor daily to ensure no issues disrupt the experiment. Ensure the media source bottles maintain >5 mL of media to successfully perfuse the reactor without issue.
2. Change out the ice buckets as needed to maintain temperature control of the media source bottle. Keep the light-protective drape over the front of the microbioreactor to prevent light-based degradation of volatile media components.
3. Ensure no bubbles are introduced into the reactor. Do not forget to periodically change the media in the TCPS flask control.

5. Microbioreactor harvest

1. Ensure the BSC is sterile and ready for operation.
   1. Warm the following required reagents in a bead or water bath to 37 °C: ~25 mL of warm hMSC media, 10 mL of warm PBS, and 1.5 mL of 10 mg/mL Pronase needed for harvesting.
2. Prepare the bottle rack for harvest.
   1. Engage manual ops mode for each POD in the software.
   2. Clamp the consumable input lines using a white C-clamp along the clear portion of the line.
   3. Remove the consumables to the BSC. Remove the contents of the waste bottle and discard.
   4. Flush the waste bottle with ~20 mL of PBS and discard. Assess the remaining volume in the media bottle and top up with fresh (cold) hMSC media if needed.
      NOTE: Here, > 4 mL of media is required to complete the three necessary eject-wash cycles for harvest.
   5. Line the waste bottle with ~5 mL of cold hMSC media to cushion the harvested fluid.
3. Harvest the cells from the microbioreactor.
   1. Using a 1 mL syringe, draw up 200 µL of 10 mg/mL Pronase into the syringe and ~300-400 µL of air.
   2. Unscrew the side port injection line cap and spray the needleless valve connector with 70% ethanol. Remove the tape securing the knotted side port injection line and untie the knot. Unclamp the white teardrop clamp.
   3. Connect the 1 mL syringe and gently inject the Pronase into the reactor. Chase the liquid down the injection line with air to ensure all liquid contents are delivered to the reactor.
   4. Re-engage the teardrop clamp and re-cap the side port injection line. Return the consumable to its POD, reconnect all lines, remove the white C-clamp, and resume mixing.
   5. Ensure the waste bottle and media source bottle are held on ice. Start a 5 min timer and allow the Pronase to digest the microcarriers within the reactor.
   6. After 5 min, check to see if the microcarriers have degraded by verifying they are no longer visibly present in the reactor. If microcarriers are still visible, allow one additional minute for digestion. If no microcarriers are visible, proceed.
   7. Harvest the cells. Engage manual ops mode, set the # of cycles to 3, and output port to W for waste, and click Eject Wash.
   8. After the eject wash cycles are complete, remove the consumable back to the BSC and withdraw the contents of the waste bottle using a 10- or 20-mL syringe.
   9. Dispense the contents of the waste bottle into a 15 mL conical centrifuge tube. Centrifuge the cells at 300 x g for 5 min, aspirate the supernatant, and evaluate the size of the cell pellet.
   10. Resuspend the cell pellet in 1-5 mL of growth media or PBS and count the cells on a hemocytometer. Repeat this step for all consumables.
4. Harvest the flask condition(s) using your normal cell culture protocols. A typical protocol for MSCs is described in 3.13.

6. Analyze the cells from the microbioreactor conditions and flask conditions for downstream application

NOTE: The analytical methods used to produce the featured results are provided in brief below and described in further detail in Krupczak, Farruggio, and Van Vliet (2024)²⁰.

1. Environmental control: Use graphical user interface provided by the microbioreactor to access and display data from each of a consumable's integrated sensors during the duration of a culture period. These data are provided in the Breez Explorer software that comes installed in the laptop provided with the microbioreactor.
2. Cell Yield: Harvest MSCs as described above and pellet the cells at 300 x g for 5 min. Carefully remove the supernatant and thoroughly resuspend the cell pellet in an appropriate volume of PBS or hMSC media (1-2 mL). Take a sample of the resuspended cell mixture and mix with an equal volume of Trypan Blue dye. Micropipette the stained sample mixture into a hemocytometer chip and measure cell number and viability on an automated or manual hemocytometer.
3. Check gene expression by two-step reverse transcription quantitative polymerase chain reaction as described below.
   1. RNA purification: Use an RNeasy Mini kit to isolate and purify RNA from each sample following the manufacturer's instructions. In brief, lyse the cells, transfer the cellular contents into a purification spin tube, and wash the samples with each buffer to remove unneeded cellular components, nucleic acids, and proteins that could interfere with subsequent PCR.
   2. cDNA conversion: Use a High-Capacity Reverse Transcription kit and thermal cycler to convert the purified RNA to complementary DNA (cDNA) according to the manufacturer's instructions. In brief, thaw the kit components and create a master mix by combining components in the appropriate volumes for your number of samples. Load equal parts purified RNA and prepared master mix (10 µL each) into each well of a PCR plate, maintaining care to track the locations of each sample identity. Follow the manufacturer's protocol in the thermal cycler to complete the reverse transcription reaction.
   3. qPCR: Using exon-spanning, sequence-specific primers validated to avoid primer dimer, load a 384-well qPCR plate with forward primer, reverse primer, template cDNA, nuclease-free water, and master mix in appropriate volumes (typically, 1 µL, 1 µL, 1 µL, 7 µL, 10 µL, respectively) using multiple technical replicate wells for each gene and sample. Split the contents across multiple plates if necessary to capture all gene targets. Ensure that all samples for a particular gene target are loaded onto the same plate to avoid plate-to-plate variability in results, and that each plate has an appropriate housekeeping gene and no-template controls. Measure fluorescence intensity on an appropriate qPCR machine and use the ΔΔCt method for relative quantitation to calculate gene expression.

Results

The microbioreactor enabled monitoring and regulation of environmental factors over approximately 10 days in culture. Environmental factors, including pH (Figure 2A), temperature (Figure 2B), dissolved oxygen (Figure 2C), and carbon dioxide (Figure 2D), were precisely measured and maintained in the microbioreactor conditions, but not in flask conditions, during the cultu...

Discussion

In this protocol, a method to expand, prime, and harvest cells using a microcarrier-microbioreactor approach is described. This process enables anchorage-dependent cells such as MSCs to increase cell number under the superior biochemical control of a bioreactor platform (relative to conventional TCPS flasks) at benchtop scale, suitable for rapid process development. Further, it is demonstrated that microcarrier-microbioreactor production of MSCs results in a primed cell phenotype that may exhibit enhanced therapeutic pot...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
Anti-adherence rinse solution	Stemcell Technologies	7010
Biosafety cabinet
Breez Microbioreactor	MilliporeSigma
Centrifuge
Conical tube, 15 mL
Conical tube, 50 mL
Dulbecco's Modified Eagle Medium, low glucose	Gibco	11885084
Ethanol
Fetal Bovine Serum	Gibco	A5670701
Hemocytometer
Laboratory Tape
Microcarriers
Microcentrifuge tube, 1.5 mL
Microscope
Needle, blunt, 4", 18 gauge	Amazon	B0D663ZR6D
Phosphate Buffered Saline	Corning	21-040-CM
Pronase, from Streptomyces griseus	MilliporeSigma	10165921001
Serological Pipette, 10 mL	VWR	75816-100
Sterile deionized water
Syringe filter, 0.22 um, sterile	Celltreat Scientific Products	229762
Syringe, 1 mL	BD	309628
Syringe, 20 mL	BD	302830
Syringe, 5 mL	BD	309646
Syringe, 50 mL	BD	309653
Tissue Culture Polystyrene Flask, T75	Thermo Fisher Scientific	130190
Trypan Blue	SigmaAldrich	T8154-100ML