Functional differentiation and scalable production of renal proximal tubular epithelial cells from human pluripotent stem cells in a dynamic culture system

Abstract Objective To provide a standardized protocol for large‐scale production of proximal tubular epithelial cells (PTEC) generated from human pluripotent stem cells (hPSC). Methods The hPSC were expanded and differentiated into PTEC on matrix‐coated alginate beads in an automated levitating fluidic platform bioLevitator. Differentiation efficacy was evaluated by immunofluorescence staining and flow cytometry, ultrastructure visualized by electron microscopy. Active reabsorption by PTEC was investigated by glucose, albumin, organic anions and cations uptake assays. Finally, the response to cisplatin‐treatment was assessed to check the potential use of PTEC to model drug‐induced nephrotoxicity. Results hPSC expansion and PTEC differentiation could be performed directly on matrix‐coated alginate beads in suspension bioreactors. Renal precursors arose 4 days post hPSC differentiation and PTEC after 8 days with 80% efficiency, with a 10‐fold expansion from hPSC in 24 days. PTEC on beads, exhibited microvilli and clear apico‐basal localization of markers. Functionality of PTECs was confirmed by uptake of glucose, albumin, organic anions and cations and expression of KIM‐1 after Cisplatin treatment. Conclusion We demonstrate the efficient expansion of hPSC, controlled differentiation to renal progenitors and further specification to polarized tubular epithelial cells. This is the first report employing biolevitation and matrix‐coated beads in a completely defined medium for the scalable and potentially automatable production of functional human PTEC.

detoxification and secretion of exogenous compounds such as drugs and xenobiotics into the urine.
Due to high susceptibility to toxic and waste compounds, PTEC are extremely vulnerable, and their injury may result in renal failure or total destruction. There is a high demand of PTEC for tissue modeling, large-scale drug-induced nephrotoxicity screening, and potentially for regenerative therapies. Immortalized and primary PTEC cultivated as 3-dimensional microtissue were reported to dedifferentiated within 10 days. 3 Moreover, although exhibiting a variety of functional transporters, primary PTEC are variable depending on donors, and partially dedifferentiate in vitro while immortalized PTEC lines show functional changes related to the immortalization procedures. 4 To improve PTEC-models, several differentiation protocols of human pluripotent stem cells (hPSC) into PTEC have been developed. [4][5][6] The use of human induced pluripotent stem cells (hiPSC) provides an unlimited source of cells that are donor specific and can be genetically modified to present specific kidney disease backgrounds. However, general limitations of these hPSC-derived PTEC are often their immature transporter properties, limited polarization and short lifespan. A lack of technologies for efficient, robust and automatable mass production of high quality PTEC limits their applicability.
Many biomaterials have been developed for expansion, embedding, and differentiation of stem cells. 7 Alginate hydrogel, for example, offers a variety of advantages such as low cost, environmental friendliness, high biocompatibility, low cytotoxicity, easy purification, functionalization, and adjustable gelation. 8 Spherical alginate beads after coating with extracellular matrix allow cultivation of cells and expansion of surfaces by bead supplementation. 9 In addition, the beads are easily applied to fluidic culture systems of diverse designs including stirring, rotating, and agitation bioreactors. These fluidic systems can be adapted to mimic the fluidic environment of the proximal tubule epithelia and may increase polarization, barrier, and transport functions of PTEC. 10 In 2015, Elanzew and colleagues first reported long-term expansion of hPSC as undifferentiated aggregates with low inoculation density using a BioLevitator TM , now CERO from OLS. 11 In 2017, expansion of human stem cells on Matrigel-coated alginate beads using this system was reported. 9 We used a biolevitation-based approach allowing scalable cell culture to expand and differentiate hPSC into human PTEC cultivated on Matrigel-coated alginate beads. The biolevitation together with the floating cell-covered beads models a fluidic environment. This system supported the efficient expansion of hPSC and their immediate differentiation to renal progenitors in a single system. This is the first report of using biolevitation of cell-coated alginate beads for scalable and potentially automated production of functional human PTEC.

| Cell culture and maintenance in static culture
The hiPSC lines WISCi004-A (referred to as IMR90-4-iPS, derived from female lung fibroblast) from passages 35 to 65, BCRTi005 12 derived from urinary cells at passage 25 to 35, and WAe001-A derived from male blastocyst were cultured on 6-well plates (Falcon) coated with Geltrex (Thermo Fisher Scientific) in serum-free, defined Essential 8 (E8) medium (STEMCELL Technologies). Cells were maintained in a humidified 5% CO 2 atmosphere at 37°C. 0.5mM ethylenediaminetetraacetic acid (EDTA Gibco) in calcium/magnesium free phosphatebuffered saline (PBS) was used to passage hPSC as colonies.

| Preparation, seeding and expansion of hPSC on Matrigel-coated alginate beads
Matrigel-coated alginate beads or Matrigel-coated beads were supplied by Fraunhofer Institute for Biomedical Engineering (IBMT), Project Centre for Stem Cell Process Engineering, Würzburg, Germany. Growth factor-reduced Matrigel was used to cover alginate beads. Matrigel-coated beads were stored at 4°C until use.
Before hPSC seeding, Matrigel-coated beads were rinsed twice with E8 medium. Around 40cm 2 Matrigel-coated beads were used for each 50 ml CEROtubes TM . Confluent hPSC were harvested using 0.5mM EDTA and reseeded on Matrigel-coated beads at a density of 2.6 × 10 6 cells in a final volume of 4ml E8 medium for each tube. On the next day, E8 was filled up to 10ml and changed every day. 4 days after seeding, 90% of Matrigel-coated beads were covered by cells.
In static culture of hPSCs, the differentiation protocol for renal progenitors and PTEC was performed with identical media and timelines with the exception that Geltrex was used for coating instead of Matrigel. For matrix comparison, renal progenitor cells were harvested on day 8 using Accutase cell dissociation reagent (Gibco), re-plated on Laminin521 (LN521) (BioLamina) and differentiated into PTEC.

| Paraffin embedding and sectioning
Matrigel-coated beads covered with cells were harvested, fixed, washed and encapsulated in 4% low melting agarose. Half an hour after gelation, agarose blocks containing beads were dehydrated and subsequently embedded into paraffin. The samples were sectioned at 4µm thickness using a microtome (Leica RM2255). After removing the paraffin by Xylene (Sigma-Aldrich), the sections were heated in Target Retrieval solution (Dako) at 96°C in a water bath for 30 minutes for antigen retrieval and stained with antibodies.

| Flow Cytometry
Adherent cells were dissociated to a single-cell suspension using Trypsin/EDTA 0.5% (Thermo Fisher) and discriminated alive cells and

| Transmission electron microscopy
Matrigel-coated beads covered with cells were harvested, rinsed with PBS and fixed with 2.5% glutaraldehyde (Serva, Heidelberg, Germany) After cutting the agarose in smaller blocks, the samples were dehydrated in a graded ethanol series and transferred to Epon resin (Roth, Karlsruhe, Germany). Finally, ultrathin sections of the samples (70nm) were stained with uranyl acetate, and lead citrate. The examination was carried out with a Zeiss EM 906 electron microscope at 80kV acceleration voltage (Carl Zeiss, Oberkochen, Germany).

| Glucose assay
Glucose uptake of cells on Matrigel-coated beads was measured

| Organic anion uptake assay
To investigate organic anion transport by the basolateral organic anion transporters OAT1 and OAT3, an assay was performed using the fluorescent anion 6-Carboxyfluorescein (6-CF) (Thermo Fisher), a tracer dye, as described by Lawrence et al.. 13 Briefly, d16 cells on Matrigel-coated beads were incubated with 50µM 6-CF for 40 minutes in presence or absence of 2.5mM Probenecid inhibitor (Sigma-Aldrich). Beads were washed twice with PBS before cells were harvested from beads using Trypsin/EDTA. Uptake of 6-CF was measured using MACSQuant Analyzer (Miltenyi Biotec) and data analyzed with FlowJo software.

| Organic cation uptake assays
Organic cation uptake by PTEC was investigated using the fluorescent cationic molecule DAPI, transported into live PTEC through the organic cation transporter 2 (OCT2) as described by Lawrence et al.. 13 Shortly, uptake of 1µM DAPI for 90 minutes by OCT2 was The uptake of these fluorescent substrates was measured using MACSQuant Analyzer (Miltenyi Biotec) and data analyzed with FlowJo software.
Calcein retention was measured using MACSQuant Analyzer (Miltenyi Biotec) and data analyzed with FlowJo software.

| Cytotoxicity assay
PTEC were harvested from the beads at d14 using TrypLE TM Express The absorbance or optical density (OD) was measured at 570nm and 650nm using a microplate reader (Spectra max 384). Cisplatin untreated cells were used as negative controls.
Cell viability was calculated by formula:

Attachment and expansion of hPSC in 4 days was immediately
followed by successive differentiation into mesoderm, renal progenitors and PTEC (Figure 1(A)). 6 After an initial increase in cell numbers during expansion of hPSC (d0), a steady number of renal progenitor cells was maintained until d8 (Table 1, Figure 1(C)). After subsequent differentiation and expansion of PTEC (Figure 1(A)), the cell numbers further increased until d20 (Table 1) (Figure 1(D)).
In summary, starting with hPSC, the average number of cells increased reproducibly 9.2 and 10.2 times using hiPSC lines and the ESC line, respectively, after expansion and differentiation into PTEC (Figure 1(C)).

| Differentiation of hPSC to PTEC on matrix-coated alginate beads recapitulates the developmental stages of nephrogenesis
hPSC treatment with Activin A, Retinoic Acid and BMP4 increased SIX2 expression, a key transcription factor for metanephric mesenchyme. 14 About 80% of BCRTi005-A-derived cells and 68% of WISCi004-A and WAe001-A cells showed SIX2 expression on d4 ( Figure 2(A)). These percentages decreased with further specification in the renal lineage, to 50% on d8 for all three pluripotent cell lines. Around 60% of the cells expressed RET, an indicator of ureteric bud formation, 15 in all three cell lines on d4 (Figure 2(A)). Occurrence of a RET positive population already on d4, before inducing ureteric bud by GDNF, could be explained by SIX2 + cells producing GDNF locally for ureteric bud formation and growth. 14 The nephron progenitors emerging by d4 progressed to the renal vesicle stage as visualized by the appearance of WT1 and JAG1 by d8 (Figure 2(B)).
The three hPSC lines showed a higher population of JAG1 than WT1.
The Notch ligand JAG1 is typically expressed in renal vesicle cells closest to the ureteric bud tip as well as in the prospective proximal tubular cells. 16 Its consistent appearance at d8 indicates initiation of PTEC differentiation. The successive appearance of metanephric mesenchyme, ureteric bud and renal vesicle cells during the induced differentiation process is consistent with renal development and the results obtained in static culture in previous work 6 ( Figure S1).

Lotus lectin (LTL) that binds to Fucose on microvilli of PTEC and
is an indicator for mature PTEC 17 was also abundant (around 70%) on d16 (Figure 3(B)). Moreover, Phalloidin allowed visualization of actin bundles, suggesting the presence of microvilli 18 on the hPSCderived PTEC, confirmed by immunofluorescence analysis on d16 ( Figure 3(C)). In contrast, in static culture with the same medium and the same protocol, around 60% of the cells expressed AQP1 and 55% were positive for LTL on d16 ( Figure S3(A)). When cultivated on Laminin 521 during static culture, PTEC expressed LTL (66%) and AQP1 (82%) (( Figure S3(B)), which was localized on the membrane but not homogenously in the culture; however, they showed a cuboidal morphology ( Figure S3 Figure S4). In addition, the basement membrane was explored by analyzing Laminin localization in combination with the typical apical membrane localized cotransporter SGLT2. Both proteins showed the expected apico-basal localization in hPSC-derived PTEC (Figure 4(A)).
The ability of d16 cells on Matrigel-coated beads to take up organic cations was investigated using the fluorescent cationic molecule DAPI, which can be transported into live PTEC through the basolateral organic cation transporter 2 (OCT2). 13 Around 45% of d16 cells were able to uptake the fluorescent OCT substrate 4-Di-2-ASP and about 26% showed DAPI transport. In the presence of OCT2 inhibitors, uptake of both substrates decreased appreciably ( Figure S5(A,B)). The fluorescent anion 6-Carboxyfluorescein (6-CF), a tracer dye, was used to investigate organic anion transport by the basolateral organic anion transporters OAT1 and OAT3. 13 Activity of OAT1/OAT3 was determined as probenecid-sensitive fluorescein uptake since only around 8% of the inhibitor-treated cells took up 6-CF as opposed to untreated cells, where up to 50% cells were capable of 6-CF uptake ( Figure S5(C)). PTEC were incubated with Calcein AM and an increase in fluorescence intensity was seen due to the cell-permeant nature of the dye. To investigate if the efflux of the dye was mediated by P-gp, cells were treated with the inhibitor PSC-833. The fluorescence intensity of Calcein AM was unchanged indicating that the P-gp ( Figure S5(D)), though expressed on d16 cells, was not fully functional.

| DISCUSS ION
To establish an effective, economical, simple and reproducible protocol for expansion and differentiation of hPSC-derived PTEC, a platform was developed using a levitating fluidic bioreactor together with an adjustable surface for cell adhesion, proliferation and directed differentiation based on alginate beads coated with Matrigel.
The system allows expansion and differentiation without further cell processing in a single culture system. Cell yield and differentiation efficacy were compared with conventional cultivation in static culture on Matrigel-coated polystyrene. 6 Both, expansion rates and yield of PTEC was 2 times higher for hiPSC lines and 2,2 times for Alginate can be easily modified in terms of stiffness and its surface chemically modified, to allow protein coating. 33 Figure S2(B), Figure S2(C)). The successful large-scale production of renal cells on matrixcoated alginate beads by biolevitation offers an effective and lowcost production of high quality PTEC derived from hPSC for multiple applications where there is a high demand, such as bioprinting, therapeutic application or as cellular components for tissue engineering. The monolayer coverage of the alginate beads with polarized PTEC and tight junctions furthermore may provide a system for high throughput screening of PTEC function where each bead mimics a tubular element with a basal tubular surface and an outer urinary surface. Additionally, even when harvested from the beads, these cells are capable of forming a tight epithelium within two days and express functional solute transporters, making them superior to existing immortalized cell lines.
In conclusion, we have successfully developed a platform technology for differentiation and expansion of hPSC-derived PTEC in a serum-free xeno-free medium without the need for passaging, in a single cultivation unit, providing high cell numbers in a reproducible manner for multiple applications.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study, are available from the corresponding author upon reasonable request.