Manufacture and preparation of human placenta-derived mesenchymal stromal cells for local tissue delivery

ARTICLE IN PRESS Cytotherapy, 2017; ■■: ■■–■■

Manufacture and preparation of human placenta-derived mesenchymal stromal cells for local tissue delivery

LEE LANKFORD, Y. JULIA CHEN, ZOE SAENZ, PRIYADARSINI KUMAR, CONNOR LONG, DIANA FARMER & AIJUN WANG Surgical Bioengineering Laboratory, Department of Surgery, University of California, Davis School of Medicine, Sacramento, California, USA Abstract Background. In this study we describe the development of a Current Good Manufacturing Practice (CGMP)-compliant process to isolate, expand and bank placenta-derived mesenchymal stromal cells (PMSCs) for use as stem cell therapy. We characterize the viability, proliferation and neuroprotective secretory profile of PMSCs seeded on clinical-grade porcine small intestine submucosa extracellular matrix (SIS-ECM; Cook Biotech). Methods. PMSCs were isolated from early gestation placenta chorionic villus tissue via explant culture. Cells were expanded, banked and screened. Purity and expression of markers of pluripotency were determined using flow cytometry. Optimal loading density and viability of PMSCs on SISECM were determined using 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium (MTS) cell proliferation and fluorescent live/dead assays, respectively. Growth factors secretion was analyzed using enzymelinked immunosorbent assays (ELISA). Results. PMSCs were rapidly expanded and banked. Viable Master and Working Cell Banks were stable with minimal decrease in viability at 6 months. All PMSCs were sterile, free from Mycoplasma species, karyotypically normal and had low endotoxin levels. PMSCs were homogeneous by immunophenotyping and expressed little to no pluripotency markers. Optimal loading density on SIS-ECM was 3–5 × 105 cells/cm2, and seeded cells were >95% viable. Neurotrophic factor secretion was detectable from PMSCs seeded on plastic and SIS-ECM with variability between donor lots. Discussion. PMSCs from early gestation placental tissues can be rapidly expanded and banked in stable, viable cell banks that are free from contaminating agents, genetically normal and pure. PMSC delivery can be accomplished by using SIS-ECM, which maintains cell viability and protein secretion. Future work in vivo is necessary to optimize cell seeding and transplantation to maximize therapeutic capabilities. Key Words: cell delivery, cell manufacturing, Current Good Manufacturing Practice, mesenchymal stromal cells, placenta, small intestine submucosa

Introduction Stem cell–based regenerative therapies likely represent the future of healthcare as we move further into an era of personalized medicine. Mesenchymal stromal cells (MSCs) have garnered interest for clinical use due to their differentiation capabilities, wound healing and immunomodulatory properties [1–7]. Our previous work demonstrated that placenta-derived mesenchymal stromal cells (PMSCs) represent a readily available, allogeneic cell source that is compatible with existing delivery vehicles [8,9]. Additionally, compared with bone marrow MSCs, isolation of PMSCs is relatively noninvasive and yields larger cell numbers [10]. Studies have shown that early gestation PMSCs display superior immunomodulatory, neuroprotective and wound-healing capabilities compared with later

gestation PMSCs and MSCs from other sources [5,8,9,11–18]. The therapeutic effect of MSCs is thought to result from paracrine secretion of unique cytokines and growth factors promoting endogenous healing and growth [19,20] and thus localized rather than systemic delivery of these cells can potentiate their effect. Use of a matrix delivery vehicle promotes precise cell delivery and improves cell retention and survival [21]. Porcine small intestine submucosa-derived extracellular matrix (SIS-ECM; Cook Biotech) is a commercially available, US Food and Drug Administration (FDA)-approved vehicle compatible for use with MSCs [22–26]. SIS-ECM is a natural candidate for cell delivery because it is a decellularized biomaterial that retains its inherently porous quality, allowing new cells to repopulate and thrive [25]. Studies

Correspondence: Aijun Wang, PhD, Surgical Bioengineering Laboratory, Department of Surgery, University of California, Davis School of Medicine, Research II, Suite 3005, 4625 2nd Avenue, Sacramento, CA 95817, USA. E-mail: [email protected] (Received 18 November 2016; accepted 3 March 2017) ISSN 1465-3249 Copyright © 2016 International Society for Cellular Therapy. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcyt.2017.03.003

ARTICLE IN PRESS 2

L. Lankford et al.

have shown that MSCs seeded onto SIS-ECM have increased secretion of vascular endothelial growth factor (VEGF) when compared with other culture conditions [25,27]. PMSCs seeded onto SIS-ECM represents an optimal combination of an ideal cell source with a functional delivery vehicle for allogeneic cell therapy; however, there lacks a common manufacturing protocol and stringent quality control measures [28]. In this study we describe the development of a Current Good Manufacturing Process (CGMP)-compliant method to isolate, expand and bank early gestation PMSCs. In addition, we characterize the viability, proliferation and neuroprotective growth factor secretory profile of PMSCs seeded on SIS-ECM. Materials and methods Isolation, expansion and banking of PMSCs from human early gestation placenta Discarded early gestation placental tissue (15–19 weeks) was collected at the University of California, Davis Medical Center. Cell isolation occurred within 24 h of tissue collection from three donor tissues using an explant culture method. Chorionic villus tissue was carefully dissected from placental tissue and washed in sterile 1X phosphate-buffered saline (PBS) containing 100 U/mL penicillin and 100 µg/mL streptomycin (P-S; Thermo Fisher Scientific). Then, 0.5 g of tissue was transferred to a clean 100 mm petri dish and minced into <1 mm3 fragments before being collected in a suspension of complete culture media and transferred into a tissue culture-treated T150 flask. Flasks were rocked gently to ensure even distribution and incubated in a humidified 37°C, 5% CO2 incubator. Four flasks were cultured to initiate each PMSC lot, and media changed every 3–4 days for 2–4 weeks until flasks were confluent. Complete culture media for PMSCs consisted of Dulbecco’s Modified Eagle’s Medium (DMEM) high glucose with 5% fetal bovine serum (FBS; Hyclone), 20 ng/mL recombinant human basic fibroblast growth factor (R&D Systems), 20 ng/mL recombinant human epidermal growth factor (R&D Systems), and P-S. PMSCs were cryopreserved throughout the manufacturing process at passages one (P1), two (P2) and four (P4). When explant cultures became confluent, cells were harvested and filtered through 70 µm nylon filters to remove tissue debris. Pellets were pooled before counting with a hemocytometer using a 1:1 dilution in trypan blue 0.4% solution. Six T150 flasks were reseeded with cells at P1 at a density of 4 × 103 cells/cm2, and remaining cells were cryopreserved generating a small P1 cell bank. PMSCs were then expanded and cryopreserved at P2 in the Master Cell Bank (MCB). After qualification at P2–P4, PMSCs from the MCB

were expanded and cryopreserved in a Working Cell Bank (WCB) at P4. PMSC cryopreservation was accomplished via resuspension in a solution of 10% dimethyl sulfoxide (DMSO) and 90% FBS and controlled freezing at −80°C using Mr. Frosty freeze containers (Thermo Fisher Scientific).Viability of cells from MCB and WCB was assessed at 6 months from their generation using trypan blue exclusion viability counts. Sterility, endotoxin and mycoplasma screening Sterility and endotoxin levels of PMSC culture supernatants collected at P2–P3 were assessed by the UC Davis Institute for Regenerative Cures (IRC) Quality Control Laboratory. 14-day sterility testing was completed according to United States Pharmacopeia Chapter 71 standards. Endotoxin testing was performed on an Endosafe-PTS system (Charles River Laboratories) and concentrations reported as Endotoxin Units (EU)/mL, where <0.5 EU/mL was considered acceptable. Mycoplasma screening was performed at the UC Davis Comparative Pathology Lab. Screened specimens included 3 × 106 cells and 500 µL of supernatant that were delivered on ice to the Comparative Pathology Lab within 2–3 h of collection before being screened using quantitative PCR (qPCR) for the presence of Mycoplasma species. Karyotyping Cells were karyotyped at the UC Davis IRC Karyotyping Core. PMSCs were seeded in T75 flasks at 1 × 104 cells/cm2 24–48 h prior to being treated with Colcemid (Thermo Fisher Scientific) to arrest mitotic cells in metaphase. Cells were detached from flasks using TrypLE Select (Thermo Fisher Scientific) and treated with a 0.067 mol/L KCl hypotonic solution before fixation with 3:1 methanol:acetic acid solution. Fixed cells were then dropped on slides and slides were aged before trypsin-Giemsa stain and microscope analysis. Microscopy included analysis of 20 metaphase spreads and karyotype of two metaphases. Growth kinetics PMSC growth kinetics were calculated using cell counts obtained from P1–P4 during expansion for cell banking.The following formulas were used to calculate kinetics data, where T is equal to the time in hours, n is equal to the final cell number divided by the initial seeded number and “initial” refers to the starting cell number at each passage: 1. Doubling Time ( DT ) = T ∗ log (2) log (n ) 2. Population Doublings ( PD ) = log (n ) log (2) 3. Log Cell Number = Initial ∗ 2PD

ARTICLE IN PRESS Manufacture/preparation of human PMSCs for local delivery Multipotency assays For osteogenesis and adipogenesis, 5 × 104 cells at P3 were seeded per well in a 12-well plate and allowed to adhere overnight. After 24 h, media was changed to lineage-specific StemPro differentiation media (Thermo Fisher Scientific) and cultures subsequently fed with fresh media every 3–4 days for 2 weeks. After 2 weeks, cells were fixed in 10% formalin. Osteogenesis was confirmed by staining with Alizarin Red to assess calcium deposition. Adipogenesis was confirmed by staining with Oil Red O to assess lipid accumulation, and counterstained with hematoxylin to observe cell nuclei. For chondrogenesis, 1 × 106 cells were pelleted in a 15 mL conical tube and grown in pellet culture. Pellets were immediately fed with 1 mL of StemPro differentiation media and subsequently fed with fresh media every 3–4 days for 2 weeks. After 2 weeks, pellets were fixed in 10% formalin before being dehydrated in a solution of 30% sucrose in 1X PBS overnight at 4°C. Dehydrated pellets were then embedded in Optimum Cutting Temperature (O.C.T.) compound (Thermo Fisher Scientific) and frozen at −80°C. Frozen pellets were then cryosectioned at 9-µm thickness and sections were placed on microscope slides and allowed to dry overnight. Chondrogenesis was confirmed by staining pellet sections with Alcian Blue to detect glycosaminoglycans and counterstaining with Nuclear Fast Red. Flow cytometry analysis Cells were analyzed using flow cytometry as previously described [8]. Briefly, cells were expanded from the MCB before being harvested with StemPro Accutase (Thermo Fisher Scientific). Unstained cells were taken from the total fraction and fixed immediately. The remaining cells were stained with fixable LIVE/DEAD Near-Infrared (NIR) dead cell stain before being fractioned into samples containing 1 × 106 cells. Individual samples were stained with either: fluorescein isothiocyanate (FITC)-CD44 (560977) and phycoerythrin (PE)-CD73 (561014), allophycocyanin (APC)-CD45 (560973) and PE-CD31 (560983),APCCD29 (561794) and PE-CD90 (561970), PE-CD34 (550761) and APC-CD105 (562408), or appropriate isotype controls (556650, 550854 and 556655), all from BD Biosciences. BD anti-mouse immunoglobulin (Ig), κ CompBeads were used to generate compensation controls. Flow cytometry was performed using the FACSCanto cytometer (BD Biosciences), and collected data were analyzed using FlowJo software (FlowJo LLC). Analysis of pluripotency marker expression by PMSCs was performed using the Human Pluripotent Stem Cell Sorting and Analysis Kit (560461) and

3

the Human Pluripotent Stem Cell Transcription Factor Analysis Kit (560589), both from BD Biosciences. Assays were performed on PMSCs at P3 according to the manufacturer’s protocols. Transduction of PMSCs with green fluorescent protein PMSCs were transduced with green fluorescent protein (GFP) as previously described [8]. Briefly, cells were seeded in T150 flasks at 4 × 103 cells/cm2 and allowed to adhere overnight. The next morning, media was changed to 15 mL of transduction media consisting of DMEM High Glucose, 5% FBS, 8 µg/mL protamine sulfate (MP Biomedicals, LLC) and GFPcontaining lentiviral vector (pCCLc-MNDU3-LUCPGK-EGFP-WPRE; provided by UC Davis IRC Vector Core) at a multiplicity of infection (MOI) of 5. After 6 h, transduction media was removed and cells were washed twice with 1X PBS and fresh complete media was reintroduced. After 72 h, successful GFP expression was confirmed using fluorescent microscopy using a Carl Zeiss Axio Observer D1 inverted microscope. Seeding GFP-PMSCs onto SIS-ECM SIS-ECM was a gift from Cook Biotech Inc. Using a sterile punch biopsy tool, 12-mm diameter discs of SIS-ECM were created and placed coarse-side up into wells of an ultra-low attachment 24-well plate (Corning Inc.) and soaked overnight in complete culture media. GFP-labeled PMSCs were resuspended in 15 µL of complete media per ECM. Then, 15 µL of cell suspension was carefully pipetted onto the surface of the matrix, distributing the suspension as much as possible. The plate was then transferred to a 37°C, 5% CO2 incubator for 1 h to allow for cell adherence before addition of 0.5 mL of complete media per well. Cell loading density The optimal loading density of PMSCs on SIS-ECM was determined using the 3-(4,5-dimethylthiazol-2yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)2H-tetrazolium (MTS)-based CellTiter 96 AQueous One Solution Cell Proliferation Assay (Promega). PMSCs were seeded on 12-mm SIS-ECM discs as described above at the following densities: 0, 4.2 × 104, 1 × 105, 3 × 105, 5 × 105 and 1 × 106 cells/cm2. At 24 h after seeding, the assay was performed according to the manufacturer’s instructions. Absorbance was measured at 490 nm using a SpectraMax i3 plate reader instrument (Molecular Devices LLC). Fluorescent viability assay PMSC viability on the ECM was assessed using a LIVE/DEAD Viability/Cytotoxicity Kit (Molecular

ARTICLE IN PRESS 4

L. Lankford et al.

Probes) along with Hoechst 33342 counterstain (Thermo Fisher Scientific) to label cell nuclei. PMSCs were seeded onto 12-mm SIS-ECM discs at a density of 4.2 × 104 cells/cm2 and incubated for 24 h. After 24 h, media was removed from wells and ECMs were stained with a viability dye consisting of 4 µmol/L Ethidium Homodimer (EthD1), 2 µmol/L Calcein acetoxymethyl (AM) and 2 µg/mL Hoechst 33342 in PBS at room temperature for 45 min. After staining, SIS-ECMs were flipped face-down using sterile forceps and imaged at 10 × magnification using a Zeiss Observer Z1 microscope. Blinded manual cell counts of 6–12 fields of view per donor were used to determine viability percentage, which was calculated using the following formula:

(Total nuclei − EthD1 positive nuclei ) % Viability = (Total nuclei ) × 100 Enzyme-linked immunosorbent assays Culture supernatants from PMSCs seeded on SISECM and tissue culture (TC)-treated plastic at a density of 1 × 105 cells/cm2 were collected at 24 h and assayed for neuroprotective growth factors using enzyme-linked immunosorbent assays (ELISAs). Human brain-derived neurotrophic factor (BDNF, DY248), hepatocyte growth factor (HGF, DY294) and VEGF (DY293B) levels were detected using Duoset ELISA kits from R&D Systems. All ELISAs were performed according to the manufacturer’s instructions and absorbance was measured using a SpectraMax i3 plate reader instrument (Molecular Devices LLC). Protein levels were normalized to the seeding number (1.9 × 105 for plastic and 1.131 × 105 for SIS-ECM), and negative controls consisted of complete media collected from ECM-only wells.

cell number was 1.81 × 108 (±4.54 × 107) cells with a mean viability of 96.5% (±0.63%) at initial creation, which decreased to 90.73% (±3%) at 6 months (Figure 1). The mean DTs observed at P2, P3 and P4 were 40.94 h (±3.77), 38.09 h (±3) and 35.87 h (±1.34), respectively (Figure 2A). Mean cumulative PDs were 4.17 (±0.37) at P2, 8 (±0.57) at P3 and 12.24 (±0.46) at P4 (Figure 2B).The mean log cell numbers per gram of tissue were 1.16 × 108 (±4.86 × 107), 1.72 × 109 (±6.86 × 108) and 3.14 × 1010 (±1.3 × 1010) for P2, P3 and P4, respectively (Figure 2C).

PMSC qualification screening All PMSCs contained low levels of endotoxin (0.087 EU/mL ± 0.044), and were free of microbial growth in 14-day sterility testing. Mycoplasma species were not detected in cells or culture media using qPCR. PMSCs were multipotent and demonstrated the ability to differentiate into osteogenic, adipogenic and chondrogenic phenotypes (Figure 2D). All PMSCs were positive for well-established MSC markers CD105, CD73, CD90, CD44 and C29 (99.77 ± 0.03%, 99.9 ± 0.06%, 95.83 ± 2.09%, 100.0 and 99.87 ± 0.13%, respectively) and negative for endothelial and hematopoietic lineage markers CD31, CD34 and CD45 (0.76 ± 0.09%, 2.65 ± 0.48% and 1.07 ± 0.38%, respectively) (Figure 2E–2F). Less than 2% of PMSCs on average expressed pluripotent stem cell transcription factors Sox2, Nanog and Oct 3/4 (1.67 ± 1.36%, 1.84 ± 0.55% and 0.09 ± 0.03%, respectively) and PMSCs were uniformly negative for pluripotent stem cell surface markers SSEA-1, SSEA-3 and TRA-1-81 (0.30 ± 0.04%, 0.25 ± 0.05% and 0.00%, respectively) (Figure 2G–2H). PMSCs had 20/20 normal karyotypes (Figure 2I).

PMSC loading and secretion on SIS-ECM Statistics Data reported as mean ± SEM. Statistical analysis was performed using PRISM 7 (GraphPad Software Inc). One-way analysis of variances (ANOVAs) with multiple comparisons were used to determine statistical significance between multiple groups. Results PMSC expansion and banking Large numbers of viable, rapidly expanding PMSCs were generated from 2 g of chorionic villus tissue from each donor. The MCB mean cell number was 6.81 × 107 (±1.59 × 107) cells with a mean viability of 95.58% (±1.29%) at initial creation, which decreased to 92.84% (±2.89%) after 6 months.The WCB mean

Assessment of PMSC loading on SIS-ECM demonstrated that mean absorbance increased with increasing loading densities up to 5 × 105 cells/cm2 before decreasing at 1 × 106 cells/cm2 (Figure 3A). All mean absorbance values were significantly higher than the background ECM, and only mean absorbance from the lowest density (4.2 × 104 cells/cm2) and 5 × 105 cells/cm2 were significantly different (P = 0.0328). Conditioned media collected from PMSCs seeded directly on SIS-ECM or on TC-treated plastic all had detectable levels of BDNF, HGF and VEGF (Figure 3B–D). BDNF levels for each donor were lower and HGF levels were higher when cultured on SIS-ECM as compared with cells cultured on plastic. Secreted VEGF levels showed no obvious trend in differences between SIS-ECM and plastic.

ARTICLE IN PRESS Manufacture/preparation of human PMSCs for local delivery

5

Figure 1. Manufacturing process overview. Summary of the PMSC manufacturing process including generation of MCB and WCB, qualification screening processes, and cell bank stability testing at 6 mo. Abbreviation: GA, gestational age.

PMSC viability on SIS-ECM The mean viability of PMSCs on SIS-ECM for all donors was >95% (Figure 3E–3H). The specific viabilities of Donor 1, Donor 2 and Donor 3 seeded on SIS-ECM were 96.67% (±0.4%), 95.09% (±0.6%) and 98.05% (±0.4%), respectively. Discussion Protocols for pre-clinical development of stem cell therapies must be generated by methods that can be rapidly translated to clinical manufacturing scenarios. Although it is impractical to generate bona fide CGMP stem cells for pre-clinical studies, these standards should guide production processes used in medical research laboratories. Here we describe methods to isolate early gestation human PMSCs based on measured expansion and cell banking along with qualification to ensure product purity and sterility. These methods are intended to demonstrate the feasibility of this tissue as a source of therapeutic cells and to provide a framework that can be easily adapted to a CGMP production within the United States.

The production process described here generated banks that were large and capable of supporting research studies in a single laboratory for years, possibly decades. However, this process was limited by factors of space, time and cost that could be readily managed in a true CGMP cell production that accounts for the immense expansion potential of PMSCs. For example, in comparing the log cell numbers with the real cell numbers banked (6.3 × 1010 cells calculated vs 1.81 × 108 cells actually banked at P4–a nearly 350fold decrease), it is clear that even the relatively large WCBs created here represent only a fraction of what can be achieved with appropriate scaling. The MCBs andWCBs were also stable at 6 months, although mean cell viabilities decreased over time (2.87% for the MCB and 5.98% for the WCB) indicating the need for continuous monitoring. Another important aspect in the process of clinical cell manufacturing is the rate and degree of cell expansion. DTs of PMSCs decreased throughout the manufacturing process, reaching their lowest point prior to WCB generation. Total doublings, however, remained fairly constant between passages regardless

ARTICLE IN PRESS 6

L. Lankford et al.

Figure 2. PMSC growth kinetics and characterization. PMSC expansion kinetics was assessed throughout the manufacturing process, and characterization completed prior to the generation of the WCB for each donor. (A) Mean DTs for all 3 donor PMSC lots decreased over the first 4 passages. (B) Cumulative doublings and (C) log cell number demonstrate the logarithmic expansion of PMSCs in the early passages after isolation. (D) Multipotency of PMSCs was confirmed by successful differentiation into bone (top panel), fat (middle panel) and cartilage (bottom panel). Scale bars = 100 µm. (E–F) All PMSCs qualified for use were positive for established MSC surface markers CD105, CD73, CD90, CD44 and C29, and negative for hematopoietic and endothelial lineage markers CD31, CD34 and CD45. (G–H) Cells displayed very low levels of pluripotent stem cell transcription factors Sox2 and Nanog, and were negative for transcription factor Oct 3/4. PMSCs were negative for pluripotent stem cell surface markers SSEA-1, SSEA-3 and TRA-1-81. (I) Representative image of Donor 2 karyotype analysis. All PMSCs were determined to have normal (20/20) karyotypes prior to the generation of a WCB.

of DT, and despite cells being subjected to a freeze/ thaw cycle at P2 during generation of the MCB. Qualification screening confirmed that PMSCs were sterile, free of Mycoplasma and had low endotoxin levels.The cells met accepted standards of MSCs using multipotency assays, purity using flow cytometry for commonly assessed MSC markers [29] and had few to no cells displaying pluripotency markers. All PMSCs had normal karyotypes prior to the generation of the WCB; however, additional testing is required to determine if karyotype instability develops at later passages. These results indicate the potential for this

process to generate large allogeneic cell banks for future clinical trials. PMSCs from WCBs were transduced with GFP, a step that would be omitted from clinical cell production but allows for easy imaging of cells on SISECM, which is incompatible with phase contrast microscopy. Efforts to determine loading density of PMSCs on ECM demonstrated that mean absorbance peaks at 5 × 105 cells/cm2. Although this density may be considered the maximum loading density, we speculate that the optimal density is between 3 × 105 and 5 × 105 cells/cm2. Seeding slightly below maximum

ARTICLE IN PRESS Manufacture/preparation of human PMSCs for local delivery

7

Figure 3. PMSC loading density, secretion and viability on SIS-ECM. (A) Mean results of MTS loading assay for all 3 donors demonstrates the absorbance values of increasing concentrations of PMSCs seeded on ECM up to 5 × 105 cells/cm2. (B–D) Secretion of BDNF (B), HGF (C) and VEGF (D), in culture media from each donor at 24 h post-seeding on ECM compared with the normal plastic adherence condition. Values were then normalized to seeding density for comparison. Assessment of viability of PMSCs seeded on ECM show (E) all cell nuclei stained in blue by Hoechst 33342 and overlaid with Ethidium Homodimer staining of dead cells in red. (F) Green fluorescence indicated by addition of Calcein AM demonstrates that the majority of cells seeded on ECM are live. (G) Overlay of Hoechst, Ethidium Homodimer and Calcein AM staining in this field of view, scale bar = 100 µm. (H) Percent viability of cells from each donor seeded on ECM was quantified as previously described in the Methods.

ARTICLE IN PRESS 8

L. Lankford et al.

density is likely better suited for transplantation studies because it allows space for cell growth and redistribution. Given that the total degradation time of a single-ply SIS-ECM could be greater than 3 months [30], cells on SIS-ECM may persist after transplantation for at least several weeks. There was no observable cytotoxicity of cells seeded on SIS-ECM, and viability levels from PMSCs on SIS-ECM were comparable with those obtained from viability counts prior to cell banking. Various studies have emphasized the importance of paracrine secretion on the therapeutic potential of MSCs and cell-based therapies. In our study, we chose to focus on three growth factors known to contribute to wound healing and neuroprotection. VEGF is a component of cellular differentiation and angiogenesis pathways [31]. HGF is an angiogenic factor shown to activate endothelial cell migration and proliferation, promoting vascularization and wound healing [32]. BDNF is a neurotrophin that affects neuronal survival, synaptic plasticity, nerve fiber regrowth and inflammation and may play a critical role in the neuroprotective effects of PMSCs [33]. PMSCs showed notable differences in secretion of BDNF and HGF when seeded on SIS-ECM as compared with plastic. Trends were persistent within conditions for each cell lot, but a large degree of donor-to-donor variability was observed. This underscores the importance of screening the secretory profile of candidate cells, given the critical role secretion may play in the efficacy of transplanted cells. Still, little is known about how delivery systems affect cell secretion and their related signaling pathways [25,34]. This study indicates that there may be predictable changes in secretion from cells cultured on SIS-ECM, but additional work is necessary to determine the nature and persistence of this effect. In summary, early gestation chorionic villus tissue is a source of easily obtainable, rapidly expandable PMSCs. These cells can be banked in large quantities using processes that generate cells that are free of contamination and phenotypically homogeneous. SIS-ECM is an excellent delivery vehicle, maintains cell viability and supports large numbers of seeded PMSCs. Secreted protein levels vary between plastic and SIS-ECM culture conditions; however, more work is necessary to understand the mechanism behind changes in secretory profile and its persistence and impact on in vivo transplantation. Acknowledgments We would like to acknowledge Jon Walker at the University of California, Davis, IRC Quality Control Laboratory and Catherine Nacey at the Karyotype Core for their assistance with cell qualification screening

and karyotyping, and Gerhard Bauer, director of the University of California, Davis, IRC GMP-facility for his guidance on development of standard operating procedures that mirror CGMP-compliant processes. Lastly, we would like to acknowledge Catherine Inta at the University of California, Davis, Comparative Pathology Lab for her assistance with qPCR screening. This work was funded by California Institute for Regenerative Medicine (CIRM) pre-clinical development award PC1-08103, Shriners Hospital for Children research grant #8512 and March of Dimes Foundation grant #5FY1682. Disclosure of interests: The authors have no commercial, proprietary, or financial interest in the products or companies described in this article. References [1] Maxson S, Lopez EA, Yoo D, Danilkovitch-Miagkova A, Leroux MA. Concise review: role of mesenchymal stem cells in wound repair. Stem Cells Transl Med 2012;1(2):142–9. [2] Pittenger MF, Martin BJ. Mesenchymal stem cells and their potential as cardiac therapeutics. Circ Res 2004;95(1):9–20. [3] Fleiss B, Guillot PV, Titomanlio L, Baud O, Hagberg H, Gressens P. Stem cell therapy for neonatal brain injury. Clin Perinatol 2014;41(1):133–48. [4] Nuschke A. Activity of mesenchymal stem cells in therapies for chronic skin wound healing. Organogenesis 2014;10(1):29– 37. [5] Hsieh JY, Wang HW, Chang SJ, Liao KH, Lee IH, Lin WS, et al. Mesenchymal stem cells from human umbilical cord express preferentially secreted factors related to neuroprotection, neurogenesis, and angiogenesis. PLoS ONE 2013;8(8):e72604. [6] Wu KJ, Yu SJ, Chiang CW, Cho KH, Lee YW, Yen BL, et al. Transplantation of human placenta-derived multipotent stem cells reduces ischemic brain injury in adult rats. Cell Transplant 2015;24(3):459–70. [7] Nauta AJ, Fibbe WE. Immunomodulatory properties of mesenchymal stromal cells. Blood 2007;110(10):3499–506. [8] Lankford L, Selby T, Becker J, Ryzhuk V, Long C, Farmer D, et al. Early gestation chorionic villi-derived stromal cells for fetal tissue engineering. World J Stem Cells 2015;7(1):195– 207. [9] Wang A, Brown EG, Lankford L, Keller BA, Pivetti CD, Sitkin NA, et al. Placental mesenchymal stromal cells rescue ambulation in ovine myelomeningocele. Stem Cells Transl Med 2015;4(6):659–69. [10] Pelekanos RA, Sardesai VS, Futrega K, Lott WB, Kuhn M, Doran MR. Isolation and expansion of mesenchymal stem/ stromal cells derived from human placenta tissue. J Vis Exp 2016;112. [11] Brown EG, Keller BA, Lankford L, Pivetti CD, Hirose S, Farmer DL, et al. Age does matter: a pilot comparison of placentaderived stromal cells for in utero repair of myelomeningocele using a Lamb model. Fetal Diagn Ther 2016;39(3):179–85. [12] Jones GN, Moschidou D, Puga-Iglesias TI, Kuleszewicz K, Vanleene M, Shefelbine SJ, et al. Ontological differences in first compared to third trimester human fetal placental chorionic stem cells. PLoS ONE 2012;7(9):e43395. [13] Poloni A, Rosini V, Mondini E, Maurizi G, Mancini S, Discepoli G, et al. Characterization and expansion of mesenchymal progenitor cells from first-trimester chorionic villi of human placenta. Cytotherapy 2008;10(7):690–7.

ARTICLE IN PRESS Manufacture/preparation of human PMSCs for local delivery [14] Selim AO, Selim SA, Shalaby SM, Mosaad H, Saber T. Neuroprotective effects of placenta-derived mesenchymal stromal cells in a rat model of experimental autoimmune encephalomyelitis. Cytotherapy 2016;18(9):1100–13. [15] Vellasamy S, Sandrasaigaran P, Vidyadaran S, George E, Ramasamy R. Isolation and characterisation of mesenchymal stem cells derived from human placenta tissue. World J Stem Cells 2012;4(6):53–61. [16] Lee JM, Jung J, Lee HJ, Jeong SJ, Cho KJ, Hwang SG, et al. Comparison of immunomodulatory effects of placenta mesenchymal stem cells with bone marrow and adipose mesenchymal stem cells. Int Immunopharmacol 2012; 13(2):219–24. [17] Calzarossa C, Bossolasco P, Besana A, Manca MP, De Grada L, De Coppi P, et al. Neurorescue effects and stem properties of chorionic villi and amniotic progenitor cells. Neuroscience 2013;234:158–72. [18] Yust-Katz S, Fisher-Shoval Y, Barhum Y, Ben-Zur T, Barzilay R, Lev N, et al. Placental mesenchymal stromal cells induced into neurotrophic factor-producing cells protect neuronal cells from hypoxia and oxidative stress. Cytotherapy 2012; 14(1):45–55. [19] Gnecchi M, Zhang Z, Ni A, Dzau VJ. Paracrine mechanisms in adult stem cell signaling and therapy. Circ Res 2008; 103(11):1204–19. [20] Mirotsou M, Jayawardena TM, Schmeckpeper J, Gnecchi M, Dzau VJ. Paracrine mechanisms of stem cell reparative and regenerative actions in the heart. J Mol Cell Cardiol 2011; 50(2):280–9. [21] Rane AA, Christman KL. Biomaterials for the treatment of myocardial infarction: a 5-year update. J Am Coll Cardiol 2011;58(25):2615–29. [22] Hodonsky C, Mundada L, Wang S, Witt R, Raff G, Kaushal S, et al. Effects of scaffold material used in cardiovascular surgery on mesenchymal stem cells and cardiac progenitor cells. Ann Thorac Surg 2015;99(2):605–11. [23] Nakatsu H, Ueno T, Oga A, Nakao M, Nishimura T, Kobayashi S, et al. Influence of mesenchymal stem cells on stomach tissue engineering using small intestinal submucosa. J Tissue Eng Regen Med 2015;9(3):296–304. [24] Klinger A, Kawata M, Villalobos M, Jones RB, Pike S, Wu N, et al. Living scaffolds: surgical repair using scaffolds

[25]

[26]

[27]

[28]

[29]

[30]

[31]

[32]

[33]

[34]

9

seeded with human adipose-derived stem cells. Hernia 2016;20(1):161–70. Lin X, Robinson M, Petrie T, Spandler V, Boyd WD, Sondergaard CS. Small intestinal submucosa-derived extracellular matrix bioscaffold significantly enhances angiogenic factor secretion from human mesenchymal stromal cells. Stem Cell Res Ther 2015;6:164. Chang CW, Petrie T, Clark A, Lin X, Sondergaard CS, Griffiths LG. Mesenchymal stem cell seeding of porcine small intestinal submucosal extracellular matrix for cardiovascular applications. PLoS ONE 2016;11(4):e0153412. Liu S, Zhang H, Zhang X, Lu W, Huang X, Xie H, et al. Synergistic angiogenesis promoting effects of extracellular matrix scaffolds and adipose-derived stem cells during wound repair. Tissue Eng Part A 2011;17(5–6):725–39. Mendicino M, Bailey AM, Wonnacott K, Puri RK, Bauer SR. MSC-based product characterization for clinical trials: an FDA perspective. Cell Stem Cell 2014;14(2):141–5. Dominici M, Le Blanc K, Mueller I, Slaper-Cortenbach I, Marini F, Krause D, et al. Minimal criteria for defining multipotent mesenchymal stromal cells. The International Society for Cellular Therapy position statement. Cytotherapy 2006;8(4):315–17. Record RD, Hillegonds D, Simmons C, Tullius R, Rickey FA, Elmore D, et al. In vivo degradation of 14C-labeled small intestinal submucosa (SIS) when used for urinary bladder repair. Biomaterials 2001;22(19):2653–9. Zisa D, Shabbir A, Suzuki G, Lee T. Vascular endothelial growth factor (VEGF) as a key therapeutic trophic factor in bone marrow mesenchymal stem cell-mediated cardiac repair. Biochem Biophys Res Commun 2009;390(3):834–8. Bussolino F, Di Renzo MF, Ziche M, Bocchietto E, Olivero M, Naldini L, et al. Hepatocyte growth factor is a potent angiogenic factor which stimulates endothelial cell motility and growth. J Cell Biol 1992;119(3):629–41. Lu B, Nagappan G, Lu Y. BDNF and synaptic plasticity, cognitive function, and dysfunction. Handb Exp Pharmacol 2014;220:223–50. Rustad KC, Wong VW, Sorkin M, Glotzbach JP, Major MR, Rajadas J, et al. Enhancement of mesenchymal stem cell angiogenic capacity and stemness by a biomimetic hydrogel scaffold. Biomaterials 2012;33(1):80–90.