Controlled Growth and the Maintenance of Human Pluripotent Stem Cells by Cultivation with Defined Medium on Extracellular Matrix-Coated Micropatterned Dishes

Here, we introduce a new serum-free defined medium (SPM) that supports the cultivation of human pluripotent stem cells (hPSCs) on recombinant human vitronectin-N (rhVNT-N)-coated dishes after seeding with either cell clumps or single cells. With this system, there was no need for an intervening sequential adaptation process after moving hPSCs from feeder layer-dependent conditions. We also introduce a micropatterned dish that was coated with extracellular matrix by photolithographic technology. This procedure allowed the cultivation of hPSCs on 199 individual rhVNT-N-coated small round spots (1 mm in diameter) on each 35-mm polystyrene dish (termed “patterned culture”), permitting the simultaneous formation of 199 uniform high-density small-sized colonies. This culture system supported controlled cell growth and maintenance of undifferentiated hPSCs better than dishes in which the entire surface was coated with rhVNT-N (termed “non-patterned cultures”). Non-patterned cultures produced variable, unrestricted cell proliferation with non-uniform cell growth and uneven densities in which we observed downregulated expression of some self-renewal-related markers. Comparative flow cytometric studies of the expression of pluripotency-related molecules SSEA-3 and TRA-1-60 in hPSCs from non-patterned cultures and patterned cultures supported this concept. Patterned cultures of hPSCs allowed sequential visual inspection of every hPSC colony, giving an address and number in patterned culture dishes. Several spots could be sampled for quality control tests of production batches, thereby permitting the monitoring of hPSCs in a single culture dish. Our new patterned culture system utilizing photolithography provides a robust, reproducible and controllable cell culture system and demonstrates technological advantages for the mass production of hPSCs with process quality control.


Introduction
Cell therapy using human pluripotent stem cell (hPSC)-derived cells has been used for the treatment of several diseases. Applications include the use of human embryonic stem cell (hESC)-derived oligodendrocyte progenitor cells for the treatment of acute spinal injury [1] or hESC-derived retinal pigment epithelium (RPE) for the treatment of dry type age-related macular degeneration (AMD) [2]. Moreover, human induced pluripotent stem cell (hiPSC)derived RPEs have been used for treatment of wet type AMD [3][4][5][6]. It is apparent that improved methods for culturing hPSCs are needed to meet regulatory demands ensuring the safety and quality of cellular products. For example, hPSCs can be maintained with animal component-free, chemically-defined and xeno-free media [7][8][9][10] under feeder-free conditions, e.g., cultivation in the presence of laminin-521, laminin-511 E8, pronectin, or rhVTN-N [11][12][13][14]. These synthetic peptides can be coated on dishes to permit feeder-free cultivation. Such peptides must be evaluated in terms of their potential for anchoring hPSCs in the culture medium as well as their cost. Among these synthetic peptides, rhVNT-N is competitive in price, costing 1-5% as much per dish as other available peptides. As for culture media, chemically-defined media without animal components would be ideal in terms of regulatory concerns, but several elaborate adaptation procedures are required in many cases (https://tools. lifetechnologies.com/content/sfs/manuals/feeder_free_hPSCs_in_essential8_medium.pdf, http://www.stemcell.com/~/media/Technical%20Resources/2/E/7/7/B/29267MAN.pdf?la = en) to adapt hPSCs on feeder layers to grow in a feeder-free environment (especially on rhVTN-N). In this report, we introduce a new medium and culture technique. The advantages of single cell patterning culture are discussed.

Materials and Methods
All experiments using human cell lines and animals were reviewed and approved by the committee for non-clinical research and the committee for animal experimentation of the Foundation for Biomedical Research and Innovation (FBRI).
Cryopreservation and thawing procedures hPSCs in single cell cultures were harvested with TrypLE Select and resuspended with chemically defined freezing medium (STEM-CELLBANKER; TakaraBio, Shiga, Japan) (1 x 10 6 cells/ mL) followed by transfer to cryovials (Iwaki, Tokyo, Japan) using a standard slow-freezing method. Cryovials were placed into a freezing container (Nalgene Cryo 100B0C Freezing Container, Nalgene, NY, USA) and cooled overnight in a -80°C freezer (MDF-U32V, Panasonic). Cells were then transferred to and stored in a -150°C freezer (MDF-1155AT, Panasonic) for at least 1 week before performing thawing experiments. For thawing, cryovials were warmed in a water bath at 37°C until the icy masses disappeared, and cell suspensions were transferred to 15 mL centrifuge tubes (BD Bioscience, MA, USA,) and diluted by addition of 4 mL SPM medium containing 10 μM Rock inhibitor Y-27632 (Nacalai Tesque, Kyoto, Japan). Cells were pelleted by centrifugation (120 x g, 3 min) and resuspended in 1 mL of fresh SPM medium containing 10 μM Y-27632 and seeded in 1 well of rhVTN-N-coated 6-well dishes as single cells and cultured at 37°C, 5% CO 2 . Media were changed every day.
In vitro differentiation and teratoma formation assay hPSCs were transferred to ultra-low attachment 6-well plates (Corning, NY, USA) to demonstrate their 3 germ layer differentiation potential by forming embryoid bodies (EBs). Cell clumps were incubated in the culture medium without bFGF and the medium was changed every three days. After 21 days of cultivation, EBs were transferred to 0.1% gelatin-coated 6-well plates. Cells were cultured for another 9 days with the same medium for differentiation, changing medium every three days. Expression of lineage-specific genes was examined by quantitative RT-PCR at the end of culture and cells were immunostained with antibodies for differentiation markers. For teratoma formation assays, 1 x 10 6 hPSCs that had been cultured for 20 passages in SPM on rhVNT-N-coated dishes were embedded in 100 μL of Matrigel (BD Bioscience) and were transplanted under the epidermal space of the neck of NOD.Cg-Prkdc Il2rg/SzJ mice (Charles River, Kanagawa, Japan). Eight weeks after transplantation, mice developed teratomas at the injection site. Teratomas were fixed with 4% formalin (Wako, Kyoto, Japan) after extraction, and sliced sections were stained with hematoxylin (Sakura Finetek, Tokyo, Japan) and eosin (Merck Millipore, Darmstadt, Germany) for histological examination for 3 germ layer differentiation.

Quantitative RT-PCR
Total RNA was extracted using an RNeasy micro kit (74004, QIAGEN, Hilden, Germany) according to the manufacturer's instructions. One μg of total RNA was used to synthesize cDNA with the QuantiTect Reverse Transcription Kit (205311, QIAGEN). Quantitative PCR (qPCR) was performed with TaqMan hPSC Scorecard Panel (A15870, Life Technologies) [19], using a qRT-PCR device (QuantStudio 12K Flex, Life Technologies).

Flow cytometric analysis
hPSCs were harvested and dissociated to single cells with TrypLE Select. The cells were washed once in a final concentration of 2% human serum albumin (Tanabe-Mitsubishi Pharma, Osaka Japan) in PBS(-). A total of 5 × 10 5 cells was incubated with the same buffer containing 1/100 volume of Alexa Fluor 647-conjugated anti-SSEA-3 (BD, 561145) or BV421-conjugated anti-TRA-1-60 (562711, BD Bioscience) for 30 min at 4°C. The dead cells were distinguished with 7-amino-actinomycin D (7AAD, 559925, BD Bioscience). The cells were analyzed with a FACS Aria II cell sorter (BD Bioscience) and the data were analyzed using FlowJo software (FlowJo LLC, OR, USA).

Karyotypic analysis
G band karyotyping service was provided by Nihon Gene Research Laboratories, Inc., (Sendai, Japan). Briefly, hiPSC were treated with colcemid (Sigma) and were harvested by treatment with 0.25% trypsin/EDTA. Cells were fixed on slides with Carnoy's solution and soaked in Giemsa stain solution (Merck Millipore). After washing with water, 50 metaphase spreads were screened and 20 of them were evaluated for chromosomal rearrangements by microscopy (Eclipse E600, Nikon, Tokyo, Japan) at 1000x magnification for G-band analysis. For multicolor fluorescein in situ hybridization mFISH analysis, hPSCs fixed on glass slides were hybridized overnight with a 24XCyte mFISH probe kit (000000-0514-056; MetaSystems, MA, USA). Sections on glass slides were hybridized, and DAPI/anti-fade (000000-0542-060; MetaSystems) was applied per the manufacturer's instructions for multi-color fluorescence in situ hybridization (mFISH) analysis. Metaphase cells were identified at 64x magnification using an Axio Imager.

Micropatterning of culture dishes
Micropatterning of culture dishes was achieved by photolithographic technology as follows. The entire bottom of each polystyrene 35-mm dish (BD Bioscience) was coated with a watersoluble PVA-based, UV-sensitive polymer (BIOSURFINE-AWP, Toyo Gosei Co., Ltd., Tokyo Japan) that became insoluble after crosslinking with UV irradiation at 30 mW/cm 2 for 1 sec with an MA6 mask aligner (SUSS MicroTec AG, Garching, Germany). The spots for cell culture were masked whereas the area not used for cell culture was not masked, permitting UV exposure. The masked area remained soluble and could be washed with Milli Q water. rhVTN-N was directly coated on the bottom of the polystyrene dish on the masked area by incubating 0.5 μg/cm 2 of rhVTN-N for 1 h at room temperature just prior to use. rhVTN-N did not coat the unmasked insoluble PVA after UV irradiation. In this way, we were able to readily prepare rhVTN-N-coated micropatterned dishes.

Results
hPSCs cultured on feeder cells were readily adapted to feeder-free culture with SPM hPSCs (hESCs/hiPSCs) cultured on SNL feeder cells were harvested either with CTK (clumps) or CTK followed by TrypLE Select (single cells). They were subsequently seeded on rhVNT-Ncoated dishes. The first passage of cell clumps was split at ratios of 1:1 to 1:2 and cultured for 4 or 5 days. hPSCs must adapt to their new feeder-free environment and recover from the perturbation of cell growth after the first passage. Then, they were passaged at 1:4 or 1:5 with GCDR for subsequent passages (from the second passage onward). Cells (3-5 x 10 5 ) in a single cell suspension were seeded either on 35-mm dishes (the entire surface was coated with rhVNT-N: non-patterned culture) or on 35-mm dishes micropatterned with 199 discrete spots (1 mm in diameter) coated with rhVNT-N (patterned culture) for the first passage and cultured for 4 days. Cells (2 x 10 5 ) in a single cell suspension were seeded on either non-patterned dishes or patterned dishes and cultured for 4 days from the second passage onward as their proliferation potential stabilized during their first cultivation on rhVNT-N-coated dishes. TrypLE Select was used for dissociation. The culture protocol is shown in Fig 1. It is notable that SPM allowed hPSCs to be cultured directly from a feeder layer to feeder-free conditions (both patterned and non-patterned cultures) without a sequential adaptation procedure.

Single cell culture supported the maintenance of hPSCs
Morphological observations and flow cytometric analyses showed that hiPSC cell lines PXF#9 and 201B7 and the hESC cell lines KhES-1 and H9 could be cultured in an undifferentiated state in three ways: as cell clumps or as single cells with SPM on rhVTN-N-coated dishes (nonpatterned) or on rhVTN-N-coated spots in micropatterned dishes (patterned culture) (Fig 2A,  2B, 2C, 2D, 2E and 2F; S1A, S1B and S1C Fig). The SPM/rhVNT-N-coated dish culture system supported cell proliferation of hPSCs (Fig 2G and 2H, S1D, S1E, S1F and S1G Fig). Gene expression profiling of self-renewal-related genes determined by a qRT-PCR panel supported these data (Fig 3A, S3 Fig). Patterned cultures sustained controlled cell growth and the undifferentiated state of hPSCs A greater yield of hPSCs was harvested from non-patterned cultures than from patterned cultures (Fig 2H, S1E, S1F and S1G Fig). In our micropatterned 35-mm dishes, the total area coated with rhVNT-N was 156 mm 2 , whereas that in non-patterned 35-mm dishes was 962 mm 2 . Whereas the coated area of the non-patterned dish was 6.16-fold greater, the number of cells harvested was only 4.23-fold greater than that from patterned cultures (Fig 2H). We harvested an average of 3.9 ± 0.81 X 10 6 PXF#9 cells (n = 18) from non-patterned cultures and 9.2 ± 1.36 x 10 5 PFX#9 cells (n = 18) from patterned cultures 4 days after seeding 2 x 10 5 cells in both types of dishes. The number of cells harvested from a patterned dish could be increased by increasing the number of 1mm diameter spots and/or decresing the pitch between patterned spots. Nevertheless, our micropatterned 35-mm dish with a 156 mm 2 rhVNT-N-coated area produced a comparable cell number with clump cultures in 35-mm dish (Fig 2H, S1E, S1F and Some of the cells that were seeded on non-rhVNT-N-coated areas in patterned dishes could migrate to rhVNT-N coated areas, but the others could not survive because they were not anchored with rhVNT-N. Indeed, patterned culture restricted cell growth to the coated spots, producing uniform cell clumps (1mm in diameter) with small uniform cell sizes similar to cells in "good" clump cultures (Fig 2E, S1 Movie). In contrast, in non-patterned cultures, we observed uneven, unrestricted and amorphous spread of cell growth and disorganized orientation (Fig 2E). The number of harvested cells from non-patterned and patterned culture changed depending on when we harvested cells on "day 4"; thus, the calculated cell density It is possible that uneven cell growth on non-patterned cultures as evidenced by distinctively different cell morphologies in the rim and in the center of the colony may not be favorable for the maintenance of the self-renewal potential of hPSCs as a population in long-term culture (Fig 2E). In contrast, patterned cultures in 1 mm diameter spots could support controlled cell growth and the maintenance of the undifferentiated state of hPSCs. We compared the expression of SSEA-3 and TRA-1-60 from non-patterned and patterned cultures of KhES-1 and PFX#9 by flow cytometry at passage 15 from the same population ( Fig 2B, S1B Fig), and found higher expression of SSEA-3 and TRA-1-60 in patterned cultured than that in non-patterned culture. Further flow cytometric data of PFX#9 cells cultured in non-patterned dishes collected 5 passages after sorting showed downregulation of SSEA-3 and TRA-1-60 expression during passage (S2A Fig). These data supported this idea. The size of the patterned spots was designed to be 1 mm in diameter, as we know that we could maintain the self-renewal potential of hPSCs relatively well in this size in cell clump culture. Furthermore, patterned culture provides a location (address) and number for each individual cell spot and allows visual inspection and image acquisition of each colony with identification. It also introduces the concept of quality control of production units (spots) and a batch control by sampling a few spots for testing (S2B Fig). Differentiation potential and genetic stability of hPSCs cultured with SPM and rhVNT-N hPSCs that were cultured with SPM on VTN-N-coated dishes as single cells and clumps were examined for their differentiation potential by forming embryoid bodies (EB). The three germ layer differentiation potentials of PFX#9 cells (Fig 3A, 3B, and 3C)   EB formation and found differentiation potential was not perturbed in these culture conditions. Karyotypes of PFX#9 or KhES-1 maintained as single cells with SPM on rhVNT-Ncoated dishes were analyzed by mFISH or by G-Banding (Fig 4 and S5A and S5B Fig). Karyotypes of hPSCs in clumps were assessed by G-Band analysis (S5C and S5D Fig). We found normal karyotypes in all hPSCs that have been tested.

Cryopreservation and thawing of hPSCs as single cells
KhES-1 cells that were cultured with SPM in single cell suspensions were cryopreserved with chemically defined freezing medium (STEM-CELLBANKER). After thawing, they grew robustly in single cell suspension with SPM on VNT-N-coated non-patterned dishes. Gene expression profiles of cells 3 passages after thaw are shown in Fig 5A. KhES-1 cells maintained an undifferentiated state after the thaw. A photograph of KhES-1 cells 3 passages after thawing is shown in Fig 5B. Expression of self-renewal-related molecules SSEA-3 and TRA1-60 by KhES-1 5 passages after thawing was determined by flow cytometry (Fig 5C). Post-thaw cell growth of KhES-1 after the second passage was comparable to that seen before cryopreservation ( Fig 5D). These data indicated that KhES-1 could be efficiently cryopreserved with freezing medium STEM-CELLBANKER and thawed and cultured with SPM as single cells on rhVNT-N-coated dishes.

Discussion
In this report, we have shown that hPSCs can be cultured with a defined medium, SPM, on a very cost-effective ECM substrate (rhVNT-N) in the form of cell clumps or single cells (nonpatterned or patterned). One of the features of SPM is that hPSCs cultured on feeders can be easily adapted to feeder-free conditions in the form of cell clumps or single cells without complicated sequential adaptation procedures. Such procedures may become an issue when chemically defined media and Vitronectin-N are used for cultivation and coating materials of dishes. SPM is also optimized to culture hPSCs in ECM-coated patterned culture dishes. SPM contains BSA produced in New Zealand, and it meets safety criteria as stated in the TSE guideline (EMA/410/01 rev.3) [15] and geographical risk assessment of BSE (GRB, European Food Safety Authority) [16], considerations for future clinical application. In September, 2014, consultations with MHRA in the UK for clinical application of BSA-containing medium supported our strategy for clinical application.
Photolithographic technology has been used for manufacturing patterned culture dishes. This technology can generate a patterned dish of any size and any shape with very sharp boundaries at micrometer scales. Such patterned dishes have been used for culturing mesenchymal stem cells [20]. But, coating ECM on patterned dishes is unique. This technology can reproducibly generate patterned dishes coated with ECM of any type including Laminin 521, Laminin 511 and Vitronectin-XF with even thickness and can avoid non-homogeneous coating. The latter sometimes occurs with ECM stamping using a mold or when a drop of ECM is placed directly on a dish. Micropatterned dishes with several shapes using the same photolithographic technology and the same PVA-base material BIOSURFINE-AWP are available commercially (Toyo Gosei Co. http://www.toyogosei.co.jp/eng/product/medical/). A remaining concern for safety issues in clinical application of micropatterned dish is the possibility that PVA might elute into the culture medium and cause toxic effects to cells in culture. However, the growth rates of hPSCs in PVA-coated dish are comparable to those in noncoated dishes per culture area and no genetic abnormality was reported in several cell lines that have been tested for up to 20 passages. We believe that we will be able to prepare a toxicological test package for PVA prior to the application for clinical trial.
The passage of hPSCs as single cells was advantageous compared to passage via cell clumps (classical method). Some hPSC clumps include differentiated cells, and such differentiated clumps can attach to ECM after passage and proliferate. It is necessary to pick up and remove such differentiated clumps daily to maintain the culture's undifferentiated state. We speculated that differentiated cells in single cell suspension could no longer attach to ECM with the same affinity, probably due to changes in their adhesion molecule expression profile and could be washed out at every medium change. In contrast, differentiated cells in clumps (differentiated clump) could remain attached to the ECM due to the strong affinity of the remaining undifferentiated cells for the ECM. To prove this speculation, a novel tracing method to mark cells early in differentiation was needed. Alternatively, we conducted comparison studies of gene expression profiles of cells from several culture conditions. Specifically, we used the qRT-PCR score card panel to examine the following: (a) cells from clumps that included differentiated cells (differentiated clumps), (b) cells from clumps lacking differentiated cells (undifferentiated clumps, after removal of differentiated clumps), (c) cells from single cell culture that were passaged from single cells for several passages, and (d) cells from single cell culture that were passaged from differentiated clumps (a). We found that cells from differentiated clumps most likely differentiated toward ectoderm among the three germ layers (S4A and S4B Fig), just like the first developmental event occurs in embryo after the implantation into uterus is ectodermal differentiation. However, cell populations ceased to demonstrate a potential for ectodermal differentiation after seeding in flat dishes in single cell suspensions, showing a comparable gene expression profile to undifferentiated clumps or undifferentiated single flat cell culture maintained as single flat cultures for extended periods (S4C and S4D Fig). Single cell passage also permitted quality control of cultured cells (examining viability and expression of surface molecules during passage and before/after cryopreservation) at the single cell level, which is an additional advantage of single cell culture over clump culture.
Patterning hPSCs in hundreds of round shaped spots provides several advantages over nonpatterned single cell culture. For example, patterned culture supports controlled cell growth and the maintenance of the undifferentiated state of hPSCs by controlling the area for cell culture at around 1 mm in diameter, while cells cultured in a non-patterned dish grow unevenly and may lose the potential for pluripotency after many passages (Fig 2B, S1B and S2A Figs). It also facilitates the sampling of a few spots for quality control of a batch defined as one culture dish and monitoring the cell growth of every colony having designated address in a time course manner.
In conclusion, we have introduced a new hPSCs culture system, combining the use of a defined culture medium (SPM) and rhVNT-N-coated patterned dishes. This culture system provides a new method that can assure the quality of cultured hPSCs and the technological basis for the design of fully automated hPSCs mass closed culture system in the near future.