HHS Public Access Author manuscript Author Manuscript
Stem Cell Res. Author manuscript; available in PMC 2016 September 01. Published in final edited form as: Stem Cell Res. 2015 September ; 15(2): 365–375. doi:10.1016/j.scr.2015.08.002.
Development of a scalable suspension culture for cardiac differentiation from human pluripotent stem cells Vincent C. Chena, Jingjing Yea, Praveen Shuklac, Giau Huaa, Danlin Chena, Ziguang Lina, Jian-chang Liua, Jing Chaia, Joseph Goldc, Joseph Wuc, David Hsua, and Larry A. Couturea,b aCenter
Author Manuscript
for Biomedicine and Genetics, Beckman Research Institute of City of Hope, 1500 E. Duarte Road, Duarte, California 91010, USA bCenter
for Applied Technology Development, Beckman Research Institute of City of Hope, 1500 E. Duarte Road, Duarte, California 91010, USA
cStanford
Cardiovascular Institute, Stanford University School of Medicine, Stanford, California,
USA
Abstract
Author Manuscript
To meet the need of a large quantity of hPSC-derived cardiomyocytes (CM) for pre-clinical and clinical studies, a robust and scalable differentiation system for CM production is essential. With an hPSC aggregate suspension culture system we established previously, we developed a matrixfree, scalable, and GMP-compliant process for directing hPSC differentiation to CM in suspension culture by modulating Wnt pathways with small molecules. By optimizing critical process parameters including: cell aggregate size, small molecule concentrations, induction timing, and agitation rate, we were able to consistently differentiate hPSCs to >90% CM purity with an average yield of 1.5 to 2×109 CM/L at scales up to 1L spinner flasks. CM generated from the suspension culture displayed typical genetic, morphological, and electrophysiological cardiac cell characteristics. This suspension culture system allows seamless transition from hPSC expansion to CM differentiation in a continuous suspension culture. It not only provides a cost and labor effective scalable process for large scale CM production, but also provides a bioreactor prototype for automation of cell manufacturing, which will accelerate the advance of hPSCs research towards therapeutic applications.
Author Manuscript
Keywords human pluripotent stem cells; cardiomyocyte differentiation; suspension cell cultures; GMP
Correspondence: Larry A. Couture, Ph.D., Center for Biomedicine and Genetics, Beckman Research Institute of City of Hope, 1500 E. Duarte Road, Duarte, California 91010, Tel: (626)256-8728, Fax: (626)256-8730. Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Chen et al.
Page 2
Author Manuscript
Introduction Myocardial infarction and heart failure are leading causes of death worldwide. As the myocardium has a very limited regenerative capacity, endogenous cell regeneration cannot adequately compensate for heart damage caused by myocardial infarction. The concept of cell replacement therapy is an appealing approach to the treatment of these cardiac diseases. Human pluripotent stem cells (hPSCs) are an attractive cell source for cell replacement therapies because they can be expanded indefinitely in culture and efficiently differentiated into a variety of cell lineages, including cardiac cells. However, current hPSC expansion and differentiation methods rely on adherent cell culture systems that are challenging to scale up to the levels required to support pre-clinical and clinical studies.
Author Manuscript Author Manuscript
Activin/Nodal/TGF-β, BMP, and Wnt signaling play pivotal roles in regulating mesoderm and cardiac specification during embryo development1–7. Significant progress has been made in cardiac differentiation process by modulating Activin, BMP, and Wnt pathways, which can efficiently drive differentiation to over 80% purity of CM8–13. Using an adherent cell culture platform, one study revealed that using 2 small Wnt pathway modulators to sequentially activate and then inhibit Wnt signaling at different differentiation stages of the culture is sufficient to drive cardiac differentiation and generate CM with high purity10. In spite of this, adherent culture systems have limited scalability and are not practical to support the anticipated CM requirements of clinical trials. Alternatively, using an embryoid body (EB) differentiation method, a complex cardiac induction procedure involving stagespecific treatments with growth factors and small molecules to modulate Activin/Nodal, BMP, and Wnt pathways has been reported to be effective in cardiac differentiation in a suspension culture system9, 11. However, the process of generating EBs is inefficient, rendering this method impractical for large scale CM production. An additional limitation of these approaches for scale-up application is that both methods are based on the expansion of the hPSCs in adherent culture and the subsequent CM differentiation process in either adherent culture or as EBs. The labor intensiveness and limited scalability of current processes have been the primary bottle necks to the large scale production of CM for clinical applications of hPSC-derived CM.
Author Manuscript
Pre-clinical studies suggest that doses of up to one billion CM will be required to achieve therapeutic benefit after transplantation14, 15. In order to meet the current CM demand for pre-clinical studies and the anticipated demand for foreseeable clinical studies, development of a robust, scalable and cGMP-compliant differentiation process for the production of both hPSCs and hPSC-derived CM is essential. Suspension cell culture is an attractive platform for large scale manufacture of cell products for its scale-up capacity. Application of a suspension culture platform to support hPSC growth in matrix-free cell aggregates has been well established16–21. We previously also reported the development of a defined, scalable and cGMP-compliant suspension system to culture hPSCs in the form of cell aggregates21. With this suspension culture system, hPSC cultures can be serially passaged and consistently expanded. In the present study we adapted our suspension culture system to establish a robust, scalable and cGMP-compliant process for manufacturing CM. We were able to use hPSC aggregates generated in the suspension culture system directly to produce CM with high efficiency and yield in suspension with various scales of spinner flasks. We optimized
Stem Cell Res. Author manuscript; available in PMC 2016 September 01.
Chen et al.
Page 3
Author Manuscript
various critical process parameters including: small molecule concentration, induction timing and agitation rates for differentiation cultures in spinner flasks with scales up to 1 L. In this study, we integrated undifferentiated hPSC expansion and small molecule-induced cardiac differentiation into a scalable suspension culture system using spinner flasks, providing a streamlined and cGMP-compliant process for scale-up CM differentiation and production.
Materials and methods hPSC suspension cultures
Author Manuscript
We routinely maintained the hPSCs lines H7 (WA07, WiCell), ESI-017 (BioTime), and a hiPSC line (a gift from Dr. Joseph Wu, Stanford) in the form of cell aggregates in suspension culture as previously described21. Briefly, suspension-adapted hPSCs were seeded as single cells at a density of 2.5–3×105 cells/mL in 125, 500, or 1000 mL spinner flasks (Corning) containing culture medium (StemPro hESC SFM, Thermo Fisher Scientific, Life Technologies) with 40 ng/mL bFGF (Life Technologies) and 10 μM Y27632 (EMD Millipore). Stirring rates were adjusted to between 50–70 rpm depending on the vessel size and hPSC line. Medium was changed every day by demi-depletion with fresh culture medium without Y27632. Cells were dissociated with Accutase (Millipore) into single cells and passaged every 3–4 days when the aggregate size reached approximately 300 μm. Cell suspension cultures were maintained in 5% CO2 with 95% relative humidity at 37°C. Calculation of sizes of cell aggregates
Author Manuscript
Aliquots of aggregates from the suspension cultures were evenly spread in 24- or 6-well plates as necessary to allow adequate separation of aggregates. A minimum of 3 pictures were taken of different areas from the edge to the middle of a well under a microscope. The pictures were analyzed with ImagePro software (Media Cybernetics). At least 200 cell aggregates in the pictures were randomly circled and the sizes of individual cell aggregates and average sizes of selected aggregates were calculated using the software. Differentiation of hPSC to CM in suspension
Author Manuscript
Undifferentiated hESC aggregates generated from suspension cultures were directly used for differentiation in suspension without cell dissociation. Low attachment 6-well plates and spinner flasks in sizes of 125, 500, and 1000 mL were used for differentiation. The differentiation cultures were maintained in 5% CO2 with 95% relative humidity at 37°C. RPMI 1640 medium (Life Technologies) with 1× B27® Supplement minus insulin (Life Technologies) was used as the basal medium from CHIR and IWP-4 induction through 2 days post IWP-4 induction. Two days after IWP-4 induction, RPMI 1640 with 1× B27® Supplement (Life Technologies) was used as basal medium. Thereafter, 60–80% of media was change every 2–3 days until cell harvest. Briefly, induction would be initiated on the day hPSC aggregates reached an average size between 160–280 μm. On day 0 of induction, the aggregates were induced with CHIR (Stemgent) at various concentrations. On day 1 the medium was changed to remove the CHIR. On day 2 or 3 IWP-4 (Stemgent) was added at various concentrations for 2 days. On day 4 or 5, the medium was changed to remove
Stem Cell Res. Author manuscript; available in PMC 2016 September 01.
Chen et al.
Page 4
Author Manuscript
IWP-4. After 2 days of IWP-4 induction, basal medium alone were used for further differentiation as described previously. Cryopreservation At harvest, CM aggregates were dissociated with Liberase TH (Roche) at 37°C for 20–30 min. After washing with PBS, the CM aggregates were further dissociated into single cells with TrypLE (Life Technologies) at 37°C for 5–10 min. Dissociated single CM at 1–3×107 cells/mL were cryopreserved with CryoStor CS10 (Biolife Solutions, Inc.) supplemented with 10 μM Y27632 in liquid nitrogen. To carry cells in adherent culture, cryopreserved cells were thawed and plated in 6-well plates coated with Synthemax (Corning) at cell seeding density 1–3×106 cells per well with 3 mL culture medium RPMI supplemented with B27.
Author Manuscript
Flow Cytometry Analysis of the cell surface markers Tra-1-60, SSEA-4, ROR2, PDGFR-α, and CD90: Cells were enzymatically dissociated to single cells, washed in PBS, and counted using a hemocytometer. 2–3×105 cells were resuspended in FACS buffer composed of 0.5% BSA in PBS and incubated with directly conjugated antibodies for 30 min on ice. After incubation, cells were washed 3 times in FACS buffer and analyzed with an Accuri C6 flow cytometer (BD Biosciences). The antibodies used were: Anti-SSEA-4-PE conjugated (R&D Sysstems), Anti-Tra-1-60-FITC conjugated (BD Biosciences), APC Mouse Anti-Human CD90 (BD Biosciences), mROR2-PE conjugated (FAB20641P, R&D Systems), mPDGFr-alpha-APC conjugated (FAB1264A, R&D Systems).
Author Manuscript
Analysis of the intracellular marker Oct-4 and cTnT: Cells were enzymatically dissociated to single cells, washed in PBS, and counted using a hemocytometer. Cells were then fixed in 4% PFA in PBS for 10 min at room temperature. Subsequently, cells were washed in PBS, permeabilized in PBS supplemented with 0.1% BSA and 0.1% saponin (permeabilization buffer) for 10 min at room temperature. 2–3×105 cells were resuspended in permeabilization buffer and incubation with conjugated Oct-4 antibody or cTnT primary antibody for 30 min on ice. Cells stained with cTnT primary antibody were then washed 3 times in permeabilization buffer and incubated with secondary antibody (diluted in permeabilization buffer) for 30 min on ice. After incubation, cells were washed 3 times in FACS buffer and analyzed by Accuri C6. The antibodies used were Anti-Oct3/4-PE conjugated (BD Biosciences) and Anti-Cardiac Troponin T antibody [1C11] (Abcam), and secondary antibody for anti-cardiac Troponin T antibody was R-PE goat anti-mouse IgG (Southern Biotech).
Author Manuscript
RNA sequencing Transcriptome sequencing libraries were constructed with TruSeq RNA Sample Preparation Kit V2 (Illumia) with minor modifications. In brief, 500 ng of total RNA from each sample was used for polyadenylated RNA enrichment with oligo dT magnetic beads, and the poly(A) RNA was fragmented with divalent cations at 94°C for 8 min. First strand cDNA was synthesized using random oligonucleotides and SuperScript II (Life Technologies). Second strand cDNA synthesis was subsequently performed using DNA Polymerase I.
Stem Cell Res. Author manuscript; available in PMC 2016 September 01.
Chen et al.
Page 5
Author Manuscript
Double stranded cDNA was further subjected to end repair, A-tailing, and adapter ligation in accordance with the manufacturer supplied protocols. Purified double strand cDNA templates with ligated adaptor molecules on both ends were selectively enriched by 10 cycles of PCR for 10 s at 98°C, 30 s at 60°C, and 30 s at 72°C using Illumina PCR Primer Cocktail and Phusion DNA polymerase (Illumina). PCR products were cleaned using 1.0 × AmpureXP beads (Beckman Coulter). Purified libraries were validated using Bioanalyzer 2100 system with DNA High Sensitivity Chip (Agilent) and quantified with Qubit (Life Technologies). All libraries were sequenced on the Illumina Hiseq 2500 with single 40 bp reads following the manufacturer’s recommendations. Immunocytochemistry
Author Manuscript
Single cell-dissociated CM were seeded on Synthemax II-coated glass bottom Petri dishes for 2–3 days. Cells were then fixed in 4% paraformaldehyde for 10 min and permeabilized using 0.1% triton X-100 for 30 min followed by blocking in 5% normal goat serum for 60 min at room temperature. After permeabilization and blocking, the cells were incubated overnight at 4°C with primary antibodies, mouse anti-α-Actinin (Sigma), mouse anti-cTNT (Abcam), rabbit anti-cTNT (Abcam), or rabbit anti-cTNI (Abcam), at 1:100. The samples were rinsed 3× with PBST for 30 min and incubated with secondary antibodies (Cy3 antirabbit or FITC anti-mouse, Millipore; 1:500) overnight at 4 C, rinsed, applied a drop of Prolong Gold Anti-fade reagent containing DAPI (Life Technology), and covered with a coverslip. The images were acquired with multiphoton laser scanning confocal microscope (Zeiss LSM 510). Electrophysiology
Author Manuscript Author Manuscript
Whole cell action potentials were recorded with the use of standard patch-clamp technique, as previously described8, 22, 23. Cultured hESC-derived cardiomyocytes (H7 CMs) were dissociated using TrypLE for 10 min at 37 °C, centrifuged at 200×g, suspended in the RPMI media supplemented with B27 (RPMI+B27) and filtered through a 100 μM cell strainer (BD Biosciences), and plated as single cells (1 × 105 cells per well of a 24-well plate) on No. 1 8 mm glass cover slips (Warner Instruments, Hamden, CT, USA) coated with Matrigel (1:50 ratio) in RPMI+B27 media supplemented with 2 μM thiazovivin and allowed to attach for 48–72 hrs, changing the media every other day. Cells were then placed in a RC-26C recording chamber (Warner) and mounted onto the stage of an inverted microscope (Nikon, Tokyo, Japan). The chamber was continuously perfused with the perfused with warm (35– 37 °C) extracellular solution of following composition: (mM) 150 NaCl, 5.4 KCl, 1.8 CaCl2, 1.0 MgCl2, 1.0 Na pyruvate, 15 HEPES, and 15 glucose; pH was adjusted to 7.4 with NaOH. Glass micropipettes (2–3 MΩ tip resistance) were fabricated from standard wall borosilicate glass capillary tubes (Sutter BF 100-50−10, Sutter Instruments, Novato, CA) using a programmable puller (P-97; Sutter Instruments) and filled with the following intracellular solution: 120 KCl, 1.0 MgCl2, 10 HEPES, 10 EGTA and 3 mg ATP; pH was adjusted to 7.2 with KOH. A single beating cardiomyocyte were selected and action potentials (APs) were recorded in whole cell current clamp mode using an EPC-10 patchclamp amplifier (HEKA, Lambrecht, Germany). Data were acquired using PatchMaster software (HEKA, Germany) and digitized at 1.0 kHz. The single cells were paced at constant pacing rate of 1 Hz using a 5–20 ms depolarizing current injections of 150–550 pA. Stem Cell Res. Author manuscript; available in PMC 2016 September 01.
Chen et al.
Page 6
Author Manuscript
The following are the criteria used for classifying observed APs into ventricular-, atrial- and nodal-like cardiomyocytes. For ventricular-like, the criteria were a negative maximum diastolic membrane potential ( 90 mV and AP duration at 90% repolarization/AP duration at 50% repolarization (APD)90/APD50 < 1.4. For atrial-like, the criteria were an absence of a prominent plateau phase, a negative diastolic membrane potential ( 1.7. For Nodal-like, the criteria were a more positive MDP, a slower AP upstroke, a prominent phase 4 depolarization and APD90/APD50 between 1.4 and 1.78. GMP compliance To meet GMP compliance, the processes for cardiomyocyte production from hPSC expansion and cardiac differentiation to cell cryopreservation are defined and standardized. Reagents, materials, and procedures are validated and verified by documentation.
Author Manuscript
Statistics Data are presented as mean±s.e.m. The statistical significance was determined using twotailed Student’s t test. Statistical significance was indicated as *p < 0.05, **p < 0.01, and ***p < 0.001.
Results Optimization of cardiomyocyte differentiation in small scale of static suspension
Author Manuscript Author Manuscript
Manipulating Wnt signaling with small molecules at different differentiation stages has been reported to efficiently induce hPSC to CM in adherent cultures10. Using a similar approach, we developed a cardiac differentiation process for hESCs in a suspension culture system by induction with the Wnt activator, CHIR99021 (CHIR), followed with the Wnt inhibitor, IWP-4. With our previously described hESC suspension culture system21, undifferentiated H7 cells were adapted to suspension culture in the form of cell aggregates. The suspension culture can be serially passaged and expanded as aggregates in suspension culture (Fig. 1a) while maintaining expression of pluripotency markers in over 90% of cells. The undifferentiated cell aggregates were directly used for cardiac differentiation in suspension culture. To optimize cardiac differentiation, we first titrated CHIR and IWP-4 in 6-well plate static suspension culture. As cell aggregate size may affect induction efficiency, we evaluated the relationship between aggregate size and CHIR concentration. Cell aggregates on days 1, 2, and 3 of suspension cultures representing different size ranges had an average diameter of 159±28 μm, 211±40 μm, and 270±49 μm, respectively (Fig. 1b–d). The cell aggregates from each day maintained >95% positive for pluripotency markers (Fig 1d). CHIR was then titrated at 6, 12, 18, and 24 μM in basal medium on day 0 of induction for 24 hrs. IWP4 at 5 μM was added on day 3 of induction for 48 hrs. It has been previously reported that activation of the Wnt pathway by CHIR can induce early mesodermal differentiation10. To evaluate CHIR induction efficiency, the expression of ROR2 and PDGFRα, which were previously identified as mesoderm and cardiac mesoderm markers9, 24, 25, were used to track cardiac mesoderm differentiation by flow cytometry. One set of representative results for analysis on day 2 of differentiation (Fig 2a) showed that without CHIR, cell aggregates seeded from different size ranges of day 1 (first row), 2
Stem Cell Res. Author manuscript; available in PMC 2016 September 01.
Chen et al.
Page 7
Author Manuscript Author Manuscript
(second row), and 3 (third row) suspension cultures did not differentiate to ROR2 and PDGFRα positive cells. However, CHIR concentrations starting from 6 μM were able to induce aggregates seeded from the day 1 suspension culture to generate approximately 42– 83% ROR2+PDGFRα+ cells. In contrast, the cell aggregates of the day 2 suspension culture induced by 6 μM CHIR only gave rise to 14% ROR2+PDGFRα+ cells, suggesting an insufficient induction of cardiac mesodermal cells. CHIR concentration of 12 μM could efficiently drive the differentiation to generate approximately 65–72% ROR2+PDGFRα+ cells. However, when cell aggregates in the size range of the day 3 suspension culture were used for CHIR titration, only 5–8% ROR2+PDGFRα+ cells were detected at induction with CHIR >12 μM. Notably, when 24 μM CHIR was used for induction, a significant amount of dead cells was observed, indicating increased cytotoxicity at higher CHIR concentration. Overall, the results showed that day 1 cell aggregates with a smaller size differentiated into ROR2+PDGFRα+ mesodermal cells more efficiently at CHIR induction, while the differentiation efficiency declined when larger cell aggregates were used for induction (Fig. 2b) The results of cardiac differentiation efficiency assessed for cardiac Troponin T (cTnT)positive cells on day 18 (Fig. 2c) showed that use of day 1 cell aggregates for differentiation generated a broad variability. Aggregates seeded from the day 2 culture exhibited a better cardiac differentiation with generation of 66±10% and 50±15% cTnT+ cells when induced with 12 and 18 μM CHIR. For cell aggregates harvested from day 3 culture, cardiac differentiation efficiency increased when higher CHIR concentrations were used for induction. These results suggest that cardiac differentiation efficiency is dependent on both the CHIR concentration and aggregate size.
Author Manuscript
Following the CHIR titration study, we titrated IWP-4 to further improve cardiac differentiation efficiency. Cell aggregates harvested from day 2 suspension culture with an average size of 200±20 μm were used for IWP-4 titration as aggregates in this size range could be efficiently induced by a broader concentration range of CHIR. Following the differentiation procedure described above, cell aggregates were first induced with 12 and 18 μM CHIR on day 0 followed by 1, 5, and 15 μM IWP-4 on day 3 of differentiation. The results showed that without IWP-4 addition the cells could still differentiate to CM with 12 and 18 μM CHIR induction alone, suggesting an endogenous mechanism inhibited Wnt signaling (Fig. 2d). However, addition of 1 and 5 μM IWP-4 enhanced cardiac differentiation. Optimization of stirring rates for 125mL spinner flasks
Author Manuscript
With a baseline set of differentiation conditions optimized in 6-well plates, we optimized cardiac differentiation in 125mL spinner flasks. Previous reports showed that shear stress could affect CM differentiation26–29 so we first tested the effects of stirring rate on cardiac differentiation and CM yield. As previous results showed that the ROR2+PDGFRα+ cardiac mesoderm population emerged as early as day 2 of differentiation (Fig. 2a), we also evaluated the effect of IWP-4 induction timing (day 2 vs. day 3) on cardiac differentiation efficiency. Differentiation was performed with IWP-4 added on day 2 or 3 at stirring rates of 35, 45, and 55 rpm. At a stirring rate of 35 rpm, the cardiac mesodermal population on day 2–4 of differentiation, marked by ROR2 and PDGFRα, were relatively small as compared with those from cultures stirred at 45 or 55 rpm in both conditions with IWP-4 added on day
Stem Cell Res. Author manuscript; available in PMC 2016 September 01.
Chen et al.
Page 8
Author Manuscript
2 and 3 (Fig. 3a, 3b). When analysis of cTnT+ cells on day 8 and 18, IWP-4 induction on day 2 showed a trend of better cardiac differentiation compared to day 3 induction, and at 45 rpm the cardiac differentiation exhibited better efficiency and consistency with 86±9% cTnT+ cells on day 18 (Fig. 3c,3d). Additionally, we observed that the differentiating aggregates developed at 35 rpm tended to fuse and formed larger aggregates. In contrast, the size of differentiating aggregates was smaller at 55 rpm (Fig. 3e). These results highlights the sensitivity of cardiac differentiation culture to shear and that effective cardiac differentiation only occurs within a certain range of agitation, or shear. Titration of CHIR99021 and IWP-4 induction timing for stirred suspension cultures in spinner flasks
Author Manuscript Author Manuscript
Considering that agitation and shear of a dynamic suspension culture might change the optimal condition of CM differentiation, we chose to re-optimized parameters in the 125mL spinner flask. Based on the previous results from static suspension cultures in 6-well plates and the stirring rate evaluation in 125mL spinner flasks, 3, 6, 12, and 18 μM CHIR were used to induce cell aggregates with an average size range of 200±20 μm in 125mL spinner flasks with an agitation rate of 45 rpm. IWP-4 induction time on day 2 and day 3 were also compared in the setup of CHIR titration. Analyses of the day 2 and day 3 differentiation culture showed that cell aggregates induced with 3 μM CHIR generated
Stem Cell Res. Author manuscript; available in PMC 2016 September 01. Published in final edited form as: Stem Cell Res. 2015 September ; 15(2): 365–375. doi:10.1016/j.scr.2015.08.002.
Development of a scalable suspension culture for cardiac differentiation from human pluripotent stem cells Vincent C. Chena, Jingjing Yea, Praveen Shuklac, Giau Huaa, Danlin Chena, Ziguang Lina, Jian-chang Liua, Jing Chaia, Joseph Goldc, Joseph Wuc, David Hsua, and Larry A. Couturea,b aCenter
Author Manuscript
for Biomedicine and Genetics, Beckman Research Institute of City of Hope, 1500 E. Duarte Road, Duarte, California 91010, USA bCenter
for Applied Technology Development, Beckman Research Institute of City of Hope, 1500 E. Duarte Road, Duarte, California 91010, USA
cStanford
Cardiovascular Institute, Stanford University School of Medicine, Stanford, California,
USA
Abstract
Author Manuscript
To meet the need of a large quantity of hPSC-derived cardiomyocytes (CM) for pre-clinical and clinical studies, a robust and scalable differentiation system for CM production is essential. With an hPSC aggregate suspension culture system we established previously, we developed a matrixfree, scalable, and GMP-compliant process for directing hPSC differentiation to CM in suspension culture by modulating Wnt pathways with small molecules. By optimizing critical process parameters including: cell aggregate size, small molecule concentrations, induction timing, and agitation rate, we were able to consistently differentiate hPSCs to >90% CM purity with an average yield of 1.5 to 2×109 CM/L at scales up to 1L spinner flasks. CM generated from the suspension culture displayed typical genetic, morphological, and electrophysiological cardiac cell characteristics. This suspension culture system allows seamless transition from hPSC expansion to CM differentiation in a continuous suspension culture. It not only provides a cost and labor effective scalable process for large scale CM production, but also provides a bioreactor prototype for automation of cell manufacturing, which will accelerate the advance of hPSCs research towards therapeutic applications.
Author Manuscript
Keywords human pluripotent stem cells; cardiomyocyte differentiation; suspension cell cultures; GMP
Correspondence: Larry A. Couture, Ph.D., Center for Biomedicine and Genetics, Beckman Research Institute of City of Hope, 1500 E. Duarte Road, Duarte, California 91010, Tel: (626)256-8728, Fax: (626)256-8730. Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Chen et al.
Page 2
Author Manuscript
Introduction Myocardial infarction and heart failure are leading causes of death worldwide. As the myocardium has a very limited regenerative capacity, endogenous cell regeneration cannot adequately compensate for heart damage caused by myocardial infarction. The concept of cell replacement therapy is an appealing approach to the treatment of these cardiac diseases. Human pluripotent stem cells (hPSCs) are an attractive cell source for cell replacement therapies because they can be expanded indefinitely in culture and efficiently differentiated into a variety of cell lineages, including cardiac cells. However, current hPSC expansion and differentiation methods rely on adherent cell culture systems that are challenging to scale up to the levels required to support pre-clinical and clinical studies.
Author Manuscript Author Manuscript
Activin/Nodal/TGF-β, BMP, and Wnt signaling play pivotal roles in regulating mesoderm and cardiac specification during embryo development1–7. Significant progress has been made in cardiac differentiation process by modulating Activin, BMP, and Wnt pathways, which can efficiently drive differentiation to over 80% purity of CM8–13. Using an adherent cell culture platform, one study revealed that using 2 small Wnt pathway modulators to sequentially activate and then inhibit Wnt signaling at different differentiation stages of the culture is sufficient to drive cardiac differentiation and generate CM with high purity10. In spite of this, adherent culture systems have limited scalability and are not practical to support the anticipated CM requirements of clinical trials. Alternatively, using an embryoid body (EB) differentiation method, a complex cardiac induction procedure involving stagespecific treatments with growth factors and small molecules to modulate Activin/Nodal, BMP, and Wnt pathways has been reported to be effective in cardiac differentiation in a suspension culture system9, 11. However, the process of generating EBs is inefficient, rendering this method impractical for large scale CM production. An additional limitation of these approaches for scale-up application is that both methods are based on the expansion of the hPSCs in adherent culture and the subsequent CM differentiation process in either adherent culture or as EBs. The labor intensiveness and limited scalability of current processes have been the primary bottle necks to the large scale production of CM for clinical applications of hPSC-derived CM.
Author Manuscript
Pre-clinical studies suggest that doses of up to one billion CM will be required to achieve therapeutic benefit after transplantation14, 15. In order to meet the current CM demand for pre-clinical studies and the anticipated demand for foreseeable clinical studies, development of a robust, scalable and cGMP-compliant differentiation process for the production of both hPSCs and hPSC-derived CM is essential. Suspension cell culture is an attractive platform for large scale manufacture of cell products for its scale-up capacity. Application of a suspension culture platform to support hPSC growth in matrix-free cell aggregates has been well established16–21. We previously also reported the development of a defined, scalable and cGMP-compliant suspension system to culture hPSCs in the form of cell aggregates21. With this suspension culture system, hPSC cultures can be serially passaged and consistently expanded. In the present study we adapted our suspension culture system to establish a robust, scalable and cGMP-compliant process for manufacturing CM. We were able to use hPSC aggregates generated in the suspension culture system directly to produce CM with high efficiency and yield in suspension with various scales of spinner flasks. We optimized
Stem Cell Res. Author manuscript; available in PMC 2016 September 01.
Chen et al.
Page 3
Author Manuscript
various critical process parameters including: small molecule concentration, induction timing and agitation rates for differentiation cultures in spinner flasks with scales up to 1 L. In this study, we integrated undifferentiated hPSC expansion and small molecule-induced cardiac differentiation into a scalable suspension culture system using spinner flasks, providing a streamlined and cGMP-compliant process for scale-up CM differentiation and production.
Materials and methods hPSC suspension cultures
Author Manuscript
We routinely maintained the hPSCs lines H7 (WA07, WiCell), ESI-017 (BioTime), and a hiPSC line (a gift from Dr. Joseph Wu, Stanford) in the form of cell aggregates in suspension culture as previously described21. Briefly, suspension-adapted hPSCs were seeded as single cells at a density of 2.5–3×105 cells/mL in 125, 500, or 1000 mL spinner flasks (Corning) containing culture medium (StemPro hESC SFM, Thermo Fisher Scientific, Life Technologies) with 40 ng/mL bFGF (Life Technologies) and 10 μM Y27632 (EMD Millipore). Stirring rates were adjusted to between 50–70 rpm depending on the vessel size and hPSC line. Medium was changed every day by demi-depletion with fresh culture medium without Y27632. Cells were dissociated with Accutase (Millipore) into single cells and passaged every 3–4 days when the aggregate size reached approximately 300 μm. Cell suspension cultures were maintained in 5% CO2 with 95% relative humidity at 37°C. Calculation of sizes of cell aggregates
Author Manuscript
Aliquots of aggregates from the suspension cultures were evenly spread in 24- or 6-well plates as necessary to allow adequate separation of aggregates. A minimum of 3 pictures were taken of different areas from the edge to the middle of a well under a microscope. The pictures were analyzed with ImagePro software (Media Cybernetics). At least 200 cell aggregates in the pictures were randomly circled and the sizes of individual cell aggregates and average sizes of selected aggregates were calculated using the software. Differentiation of hPSC to CM in suspension
Author Manuscript
Undifferentiated hESC aggregates generated from suspension cultures were directly used for differentiation in suspension without cell dissociation. Low attachment 6-well plates and spinner flasks in sizes of 125, 500, and 1000 mL were used for differentiation. The differentiation cultures were maintained in 5% CO2 with 95% relative humidity at 37°C. RPMI 1640 medium (Life Technologies) with 1× B27® Supplement minus insulin (Life Technologies) was used as the basal medium from CHIR and IWP-4 induction through 2 days post IWP-4 induction. Two days after IWP-4 induction, RPMI 1640 with 1× B27® Supplement (Life Technologies) was used as basal medium. Thereafter, 60–80% of media was change every 2–3 days until cell harvest. Briefly, induction would be initiated on the day hPSC aggregates reached an average size between 160–280 μm. On day 0 of induction, the aggregates were induced with CHIR (Stemgent) at various concentrations. On day 1 the medium was changed to remove the CHIR. On day 2 or 3 IWP-4 (Stemgent) was added at various concentrations for 2 days. On day 4 or 5, the medium was changed to remove
Stem Cell Res. Author manuscript; available in PMC 2016 September 01.
Chen et al.
Page 4
Author Manuscript
IWP-4. After 2 days of IWP-4 induction, basal medium alone were used for further differentiation as described previously. Cryopreservation At harvest, CM aggregates were dissociated with Liberase TH (Roche) at 37°C for 20–30 min. After washing with PBS, the CM aggregates were further dissociated into single cells with TrypLE (Life Technologies) at 37°C for 5–10 min. Dissociated single CM at 1–3×107 cells/mL were cryopreserved with CryoStor CS10 (Biolife Solutions, Inc.) supplemented with 10 μM Y27632 in liquid nitrogen. To carry cells in adherent culture, cryopreserved cells were thawed and plated in 6-well plates coated with Synthemax (Corning) at cell seeding density 1–3×106 cells per well with 3 mL culture medium RPMI supplemented with B27.
Author Manuscript
Flow Cytometry Analysis of the cell surface markers Tra-1-60, SSEA-4, ROR2, PDGFR-α, and CD90: Cells were enzymatically dissociated to single cells, washed in PBS, and counted using a hemocytometer. 2–3×105 cells were resuspended in FACS buffer composed of 0.5% BSA in PBS and incubated with directly conjugated antibodies for 30 min on ice. After incubation, cells were washed 3 times in FACS buffer and analyzed with an Accuri C6 flow cytometer (BD Biosciences). The antibodies used were: Anti-SSEA-4-PE conjugated (R&D Sysstems), Anti-Tra-1-60-FITC conjugated (BD Biosciences), APC Mouse Anti-Human CD90 (BD Biosciences), mROR2-PE conjugated (FAB20641P, R&D Systems), mPDGFr-alpha-APC conjugated (FAB1264A, R&D Systems).
Author Manuscript
Analysis of the intracellular marker Oct-4 and cTnT: Cells were enzymatically dissociated to single cells, washed in PBS, and counted using a hemocytometer. Cells were then fixed in 4% PFA in PBS for 10 min at room temperature. Subsequently, cells were washed in PBS, permeabilized in PBS supplemented with 0.1% BSA and 0.1% saponin (permeabilization buffer) for 10 min at room temperature. 2–3×105 cells were resuspended in permeabilization buffer and incubation with conjugated Oct-4 antibody or cTnT primary antibody for 30 min on ice. Cells stained with cTnT primary antibody were then washed 3 times in permeabilization buffer and incubated with secondary antibody (diluted in permeabilization buffer) for 30 min on ice. After incubation, cells were washed 3 times in FACS buffer and analyzed by Accuri C6. The antibodies used were Anti-Oct3/4-PE conjugated (BD Biosciences) and Anti-Cardiac Troponin T antibody [1C11] (Abcam), and secondary antibody for anti-cardiac Troponin T antibody was R-PE goat anti-mouse IgG (Southern Biotech).
Author Manuscript
RNA sequencing Transcriptome sequencing libraries were constructed with TruSeq RNA Sample Preparation Kit V2 (Illumia) with minor modifications. In brief, 500 ng of total RNA from each sample was used for polyadenylated RNA enrichment with oligo dT magnetic beads, and the poly(A) RNA was fragmented with divalent cations at 94°C for 8 min. First strand cDNA was synthesized using random oligonucleotides and SuperScript II (Life Technologies). Second strand cDNA synthesis was subsequently performed using DNA Polymerase I.
Stem Cell Res. Author manuscript; available in PMC 2016 September 01.
Chen et al.
Page 5
Author Manuscript
Double stranded cDNA was further subjected to end repair, A-tailing, and adapter ligation in accordance with the manufacturer supplied protocols. Purified double strand cDNA templates with ligated adaptor molecules on both ends were selectively enriched by 10 cycles of PCR for 10 s at 98°C, 30 s at 60°C, and 30 s at 72°C using Illumina PCR Primer Cocktail and Phusion DNA polymerase (Illumina). PCR products were cleaned using 1.0 × AmpureXP beads (Beckman Coulter). Purified libraries were validated using Bioanalyzer 2100 system with DNA High Sensitivity Chip (Agilent) and quantified with Qubit (Life Technologies). All libraries were sequenced on the Illumina Hiseq 2500 with single 40 bp reads following the manufacturer’s recommendations. Immunocytochemistry
Author Manuscript
Single cell-dissociated CM were seeded on Synthemax II-coated glass bottom Petri dishes for 2–3 days. Cells were then fixed in 4% paraformaldehyde for 10 min and permeabilized using 0.1% triton X-100 for 30 min followed by blocking in 5% normal goat serum for 60 min at room temperature. After permeabilization and blocking, the cells were incubated overnight at 4°C with primary antibodies, mouse anti-α-Actinin (Sigma), mouse anti-cTNT (Abcam), rabbit anti-cTNT (Abcam), or rabbit anti-cTNI (Abcam), at 1:100. The samples were rinsed 3× with PBST for 30 min and incubated with secondary antibodies (Cy3 antirabbit or FITC anti-mouse, Millipore; 1:500) overnight at 4 C, rinsed, applied a drop of Prolong Gold Anti-fade reagent containing DAPI (Life Technology), and covered with a coverslip. The images were acquired with multiphoton laser scanning confocal microscope (Zeiss LSM 510). Electrophysiology
Author Manuscript Author Manuscript
Whole cell action potentials were recorded with the use of standard patch-clamp technique, as previously described8, 22, 23. Cultured hESC-derived cardiomyocytes (H7 CMs) were dissociated using TrypLE for 10 min at 37 °C, centrifuged at 200×g, suspended in the RPMI media supplemented with B27 (RPMI+B27) and filtered through a 100 μM cell strainer (BD Biosciences), and plated as single cells (1 × 105 cells per well of a 24-well plate) on No. 1 8 mm glass cover slips (Warner Instruments, Hamden, CT, USA) coated with Matrigel (1:50 ratio) in RPMI+B27 media supplemented with 2 μM thiazovivin and allowed to attach for 48–72 hrs, changing the media every other day. Cells were then placed in a RC-26C recording chamber (Warner) and mounted onto the stage of an inverted microscope (Nikon, Tokyo, Japan). The chamber was continuously perfused with the perfused with warm (35– 37 °C) extracellular solution of following composition: (mM) 150 NaCl, 5.4 KCl, 1.8 CaCl2, 1.0 MgCl2, 1.0 Na pyruvate, 15 HEPES, and 15 glucose; pH was adjusted to 7.4 with NaOH. Glass micropipettes (2–3 MΩ tip resistance) were fabricated from standard wall borosilicate glass capillary tubes (Sutter BF 100-50−10, Sutter Instruments, Novato, CA) using a programmable puller (P-97; Sutter Instruments) and filled with the following intracellular solution: 120 KCl, 1.0 MgCl2, 10 HEPES, 10 EGTA and 3 mg ATP; pH was adjusted to 7.2 with KOH. A single beating cardiomyocyte were selected and action potentials (APs) were recorded in whole cell current clamp mode using an EPC-10 patchclamp amplifier (HEKA, Lambrecht, Germany). Data were acquired using PatchMaster software (HEKA, Germany) and digitized at 1.0 kHz. The single cells were paced at constant pacing rate of 1 Hz using a 5–20 ms depolarizing current injections of 150–550 pA. Stem Cell Res. Author manuscript; available in PMC 2016 September 01.
Chen et al.
Page 6
Author Manuscript
The following are the criteria used for classifying observed APs into ventricular-, atrial- and nodal-like cardiomyocytes. For ventricular-like, the criteria were a negative maximum diastolic membrane potential ( 90 mV and AP duration at 90% repolarization/AP duration at 50% repolarization (APD)90/APD50 < 1.4. For atrial-like, the criteria were an absence of a prominent plateau phase, a negative diastolic membrane potential ( 1.7. For Nodal-like, the criteria were a more positive MDP, a slower AP upstroke, a prominent phase 4 depolarization and APD90/APD50 between 1.4 and 1.78. GMP compliance To meet GMP compliance, the processes for cardiomyocyte production from hPSC expansion and cardiac differentiation to cell cryopreservation are defined and standardized. Reagents, materials, and procedures are validated and verified by documentation.
Author Manuscript
Statistics Data are presented as mean±s.e.m. The statistical significance was determined using twotailed Student’s t test. Statistical significance was indicated as *p < 0.05, **p < 0.01, and ***p < 0.001.
Results Optimization of cardiomyocyte differentiation in small scale of static suspension
Author Manuscript Author Manuscript
Manipulating Wnt signaling with small molecules at different differentiation stages has been reported to efficiently induce hPSC to CM in adherent cultures10. Using a similar approach, we developed a cardiac differentiation process for hESCs in a suspension culture system by induction with the Wnt activator, CHIR99021 (CHIR), followed with the Wnt inhibitor, IWP-4. With our previously described hESC suspension culture system21, undifferentiated H7 cells were adapted to suspension culture in the form of cell aggregates. The suspension culture can be serially passaged and expanded as aggregates in suspension culture (Fig. 1a) while maintaining expression of pluripotency markers in over 90% of cells. The undifferentiated cell aggregates were directly used for cardiac differentiation in suspension culture. To optimize cardiac differentiation, we first titrated CHIR and IWP-4 in 6-well plate static suspension culture. As cell aggregate size may affect induction efficiency, we evaluated the relationship between aggregate size and CHIR concentration. Cell aggregates on days 1, 2, and 3 of suspension cultures representing different size ranges had an average diameter of 159±28 μm, 211±40 μm, and 270±49 μm, respectively (Fig. 1b–d). The cell aggregates from each day maintained >95% positive for pluripotency markers (Fig 1d). CHIR was then titrated at 6, 12, 18, and 24 μM in basal medium on day 0 of induction for 24 hrs. IWP4 at 5 μM was added on day 3 of induction for 48 hrs. It has been previously reported that activation of the Wnt pathway by CHIR can induce early mesodermal differentiation10. To evaluate CHIR induction efficiency, the expression of ROR2 and PDGFRα, which were previously identified as mesoderm and cardiac mesoderm markers9, 24, 25, were used to track cardiac mesoderm differentiation by flow cytometry. One set of representative results for analysis on day 2 of differentiation (Fig 2a) showed that without CHIR, cell aggregates seeded from different size ranges of day 1 (first row), 2
Stem Cell Res. Author manuscript; available in PMC 2016 September 01.
Chen et al.
Page 7
Author Manuscript Author Manuscript
(second row), and 3 (third row) suspension cultures did not differentiate to ROR2 and PDGFRα positive cells. However, CHIR concentrations starting from 6 μM were able to induce aggregates seeded from the day 1 suspension culture to generate approximately 42– 83% ROR2+PDGFRα+ cells. In contrast, the cell aggregates of the day 2 suspension culture induced by 6 μM CHIR only gave rise to 14% ROR2+PDGFRα+ cells, suggesting an insufficient induction of cardiac mesodermal cells. CHIR concentration of 12 μM could efficiently drive the differentiation to generate approximately 65–72% ROR2+PDGFRα+ cells. However, when cell aggregates in the size range of the day 3 suspension culture were used for CHIR titration, only 5–8% ROR2+PDGFRα+ cells were detected at induction with CHIR >12 μM. Notably, when 24 μM CHIR was used for induction, a significant amount of dead cells was observed, indicating increased cytotoxicity at higher CHIR concentration. Overall, the results showed that day 1 cell aggregates with a smaller size differentiated into ROR2+PDGFRα+ mesodermal cells more efficiently at CHIR induction, while the differentiation efficiency declined when larger cell aggregates were used for induction (Fig. 2b) The results of cardiac differentiation efficiency assessed for cardiac Troponin T (cTnT)positive cells on day 18 (Fig. 2c) showed that use of day 1 cell aggregates for differentiation generated a broad variability. Aggregates seeded from the day 2 culture exhibited a better cardiac differentiation with generation of 66±10% and 50±15% cTnT+ cells when induced with 12 and 18 μM CHIR. For cell aggregates harvested from day 3 culture, cardiac differentiation efficiency increased when higher CHIR concentrations were used for induction. These results suggest that cardiac differentiation efficiency is dependent on both the CHIR concentration and aggregate size.
Author Manuscript
Following the CHIR titration study, we titrated IWP-4 to further improve cardiac differentiation efficiency. Cell aggregates harvested from day 2 suspension culture with an average size of 200±20 μm were used for IWP-4 titration as aggregates in this size range could be efficiently induced by a broader concentration range of CHIR. Following the differentiation procedure described above, cell aggregates were first induced with 12 and 18 μM CHIR on day 0 followed by 1, 5, and 15 μM IWP-4 on day 3 of differentiation. The results showed that without IWP-4 addition the cells could still differentiate to CM with 12 and 18 μM CHIR induction alone, suggesting an endogenous mechanism inhibited Wnt signaling (Fig. 2d). However, addition of 1 and 5 μM IWP-4 enhanced cardiac differentiation. Optimization of stirring rates for 125mL spinner flasks
Author Manuscript
With a baseline set of differentiation conditions optimized in 6-well plates, we optimized cardiac differentiation in 125mL spinner flasks. Previous reports showed that shear stress could affect CM differentiation26–29 so we first tested the effects of stirring rate on cardiac differentiation and CM yield. As previous results showed that the ROR2+PDGFRα+ cardiac mesoderm population emerged as early as day 2 of differentiation (Fig. 2a), we also evaluated the effect of IWP-4 induction timing (day 2 vs. day 3) on cardiac differentiation efficiency. Differentiation was performed with IWP-4 added on day 2 or 3 at stirring rates of 35, 45, and 55 rpm. At a stirring rate of 35 rpm, the cardiac mesodermal population on day 2–4 of differentiation, marked by ROR2 and PDGFRα, were relatively small as compared with those from cultures stirred at 45 or 55 rpm in both conditions with IWP-4 added on day
Stem Cell Res. Author manuscript; available in PMC 2016 September 01.
Chen et al.
Page 8
Author Manuscript
2 and 3 (Fig. 3a, 3b). When analysis of cTnT+ cells on day 8 and 18, IWP-4 induction on day 2 showed a trend of better cardiac differentiation compared to day 3 induction, and at 45 rpm the cardiac differentiation exhibited better efficiency and consistency with 86±9% cTnT+ cells on day 18 (Fig. 3c,3d). Additionally, we observed that the differentiating aggregates developed at 35 rpm tended to fuse and formed larger aggregates. In contrast, the size of differentiating aggregates was smaller at 55 rpm (Fig. 3e). These results highlights the sensitivity of cardiac differentiation culture to shear and that effective cardiac differentiation only occurs within a certain range of agitation, or shear. Titration of CHIR99021 and IWP-4 induction timing for stirred suspension cultures in spinner flasks
Author Manuscript Author Manuscript
Considering that agitation and shear of a dynamic suspension culture might change the optimal condition of CM differentiation, we chose to re-optimized parameters in the 125mL spinner flask. Based on the previous results from static suspension cultures in 6-well plates and the stirring rate evaluation in 125mL spinner flasks, 3, 6, 12, and 18 μM CHIR were used to induce cell aggregates with an average size range of 200±20 μm in 125mL spinner flasks with an agitation rate of 45 rpm. IWP-4 induction time on day 2 and day 3 were also compared in the setup of CHIR titration. Analyses of the day 2 and day 3 differentiation culture showed that cell aggregates induced with 3 μM CHIR generated
