Stirred Suspension Bioreactor Culture of Porcine Induced Pluripotent Stem Cells

Induced pluripotent stem cells (iPSCs) are an attractive cell source for regenerative medicine and the development of therapies, as they can proliferate indefinitely under defined conditions and differentiate into any cell type in the body. Large scale expansion of cells is limited in adherent culture, making it difficult to obtain adequate cell numbers for research. It has been previously shown that stirred suspension bioreactors (SSBs) can be used to culture mouse and human stem cells. Pigs are important pre-clinical models for stem cell research. Therefore, this study investigated the use of SSBs as an alternative culture method for the expansion of iPSCs. Using an established porcine iPSC line as well as a new cell line derived and characterized in the current study, we report that porcine iPSCs (piPSCs) can grow in SSB while maintaining characteristics of pluripotency and karyotypic stability similar to cells grown in traditional two-dimensional static culture. This culture method provides a suitable platform for scale up of cell culture to provide adequate cell numbers for future research applications involving porcine induced pluripotent stem cells.

The purpose of this study was to establish an alternative, scalable culture method for the 91 expansion of piPSCs using SSBs, based upon previous work involving human and murine PSCs 92 (8; 9; 10; 11; 12; 13). Here we demonstrate that piPSCs from an established line and a newly 93 generated line grown in static and SSB culture exhibit similar characteristics in terms of apparent The piPSCs (BT3p17) that constitutively express GFP (14) were generously donated by Drs. 99 Telugu, Ezashi, and Roberts (University of Missouri). Briefly, these piPSCs were derived from   Cryopreserved LIF-dependent piPSCs were thawed in a 37ºC water bath and washed in pre-127 warmed medium. piPSCs (10 5 cells/mL) were plated onto 6-well culture plates (Corning, 353846) 128 coated with 50 µg/mL poly-d-lysine (PDL; Sigma, P0899), 20 µg/mL laminin (LN; Sigma, L-LIF-independent iPSCs were cultured on mitomycin-treated feeder cells for several passages, and 136 then individual colonies were transferred onto Matrigel-coated plates (Corning, 354277) and 137 characterized. Cells originating from a single colony were included in this study. The LIF-138 independent piPSC line was maintained in mTeSR™1 medium at 37°C, 5% CO 2 , in saturated 139 humidity, and passaged using Accutase, when cell confluency reached 80%. In all culture 140 conditions, piPSCs were dissociated into single cells, counted and assessed for cell viability using   For static culture, 10 5 cells/mL from the initial pre-piPSC culture of LIF-dependent and LIF-153 independent cells were plated onto 6-well culture plates coated with PDL/LN and Matrigel-coated 154 plates, respectively, with a full medium change at day 2 and passage at day 3 using Accutase. Each 155 subsequent passage used piPSCs generated from previous static culture up to passage 8. Spinner Flask Assembly Complete, 264500-50) were siliconized as per manufacturer's 160 instructions (Sigma, SL-2) and used with a working volume of 50 mL (Table 1, Fig 1). Fifty mL 161 SSBs were inoculated with 1.406 x10 6 piPSCs from the pre-piPSC culture resuspended in 50 mL 162 N2B27-3i medium or mTeSR™1. The piPSCs were agitated at 104 rpm, corresponding to a 163 maximum shear stress of 3.0 dyne/cm 2 . The cells were passaged on day 3 using Accutase.   (1) = where T d represents apparent doubling time, q 1 is the initial viable cell number at t 1 , and q 2 is the 177 final viable cell number at t 2 (72 hrs). t 1 (0 hr) represents the initial inoculation time. Live cell 178 counts were used for this calculation.

179
Cell viability calculation 180 Live and dead cells were counted, and cell viability was determined using the following formula:

181
(2) where C V represents cell viability as a percentage, C L is the total cells counted after 72 hrs that did  The maximum shear stress ( max ) that a single cell is exposed to, on the surface of an aggregate 190 can be estimated using the following equations (17):

316
Cytogenetic analysis showed a stable karyotype (>95%) throughout the duration of the Samples were collected from SSB culture at each passage to visualize aggregate formation 330 in suspension culture. Aggregates formed freely after inoculation of piPSCs as single cells (Fig. 3 331 B I & C I , respectively) while maintaining GFP expression (Fig. 3 B II & C II , respectively). piPSCs 332 cultured statically also maintained GFP expression (Fig. 3 A-A II ). from pre-, static, and SSB culture conditions (Fig. 4A). The similarity of piPSCs from static and 348 SSB cultures was also apparent using flow cytometry analysis, which determined that there were 349 no significant differences in the percentage of cells expressing the pluripotency-associated markers 350 SSEA-4, SOX2, and OCT4, as well as the proliferation marker Ki67 (Fig. 4B). These similarities 351 were qualitatively confirmed by immunocytochemistry for SSB-cultured piPSCs (Fig. 4C) and 352 piPSCs cultured statically (Fig. 4D). As controls, isotype controls and unstained piPSCs were used to validate staining of 362 piPSCs from pre-, static, and SSB culture conditions (Fig. S1).  Supporting the gene expression data, lineage specific protein expression was observed by 413 staining for SMA and NESTIN in piPSCs from pre-, static, and SSB cultures allowed to 414 spontaneously differentiate for 3 days (Fig. 6). static and SSB cultures (Fig. S2C). In contrast, immunocytochemistry did not detect a difference 446 between SSB and static culture in the expression of exogenous hSOX2, OCT4, and SSEA-4, or 447 Ki67 used here to assess cell proliferation (Fig. S2D). These observations further support the 448 dependency of porcine iPSCs on the exogenous reprogramming genes to maintain their 449 pluripotency.

450
To fully explore the potential of this newly generated LIF-independent piPSC line to 451 differentiate into the three germ layers, we performed in vitro and in vivo differentiation assays.

452
When compared to undifferentiated piPSCs, both static and SSB cultures-derived EBs (Fig. S3A) 453 showed an up-regulation of ACTA2, SOX17 and NESTIN expression, while they maintained 454 GATA4 (Fig. S3B). The gene expression of NESTIN differed in EBs between the two culture 455 conditions. There was no noticeable difference in the expression of SMA, GATA4 or NESTIN 456 observed between the static and SSB-derived EBs (Fig. S3C). Surprisingly, EBs derived from 457 static and SSB cultures maintained the expression of pluripotency markers OCT4, cMYC and KLF4 458 during the course of this experiment (Fig. S3D), and further showed a significant increase in 459 porcine SOX2 expression, as compared to undifferentiated cells (Fig. S3D). 460 We then performed teratoma assays by injecting piPSCs cultured under static and SSB In the present work, we aimed to determine whether SSBs offered an efficient alternative 474 culture platform to the two-dimensional static culture system for the expansion of piPSCs.

543
Formation of teratomas in NOD-SCID mice is considered a gold standard of pluripotency.

544
The LIF-dependent iPSC failed to form teratomas. In previous work, LIF-dependent piPSCs were 545 only able to form teratomas upon administration of DOX via drinking water (36) . Dependence on 546 DOX is consistent with the fact that the pluripotent state of piPSCs still requires the persistence of 547 exogenous genes to maintain pluripotency, possibly due to the inadequate activation of 548 endogenous genes to maintain pluripotency (37). However, the newly derived LIF-independent 549 iPSCs differentiated into ectoderm and mesoderm in vivo, and this potential was maintained in 550 SSB culture.

551
Here we demonstrate that piPSCs grown in static and SSB culture are similar in their 552 functional characteristics. The downstream application of using SSBs is the potential for scale up.

553
Considering the experimental parameters set by this study, such as seeding density and culture 554 duration, the theoretical cell yield can be calculated to illustrate the number of piPSCs generated 555 if all piPSCs were seeded in SSBs after each passage (Supplementary information). Using 10 9 cells 556 as a "therapeutic" dose, 1.406 x10 6 piPSCs can be inoculated into a 50 mL SSB and scaled up over 557 the following 3 passages yielding a total cell number of ~5.91 x10 9 piPSCs if scaled up to 2 -10 558 L SSBs (Table S1). If this theoretical experiment was carried out using the static 6-well tissue 559 culture plates, a "therapeutic" cell yield would be achieved by the 3 rd passage, however 143 6-well 560 tissue culture plates would be required. As this study illustrates that the piPSCs cultured statically 561 or in SSBs are functionally equivalent, scale up would be a more feasible option using SSBs rather 562 than static tissue culture plates for future studies.

564
Conclusions 565 We present here proof-of-principle for stirred suspension bioreactor culture of 566 piPSCs,providing a scalable method of propagating piPSCs in vitro.