Roles of fluid shear stress and retinoic acid in the differentiation of primary cultured human podocytes

Roles of fluid shear stress and retinoic acid in the differentiation of primary cultured human podocytes

Experimental Cell Research xxx (xxxx) xxx–xxx Contents lists available at ScienceDirect Experimental Cell Research journal homepage:

2MB Sizes 1 Downloads 40 Views

Experimental Cell Research xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

Experimental Cell Research journal homepage:

Roles of fluid shear stress and retinoic acid in the differentiation of primary cultured human podocytes Seung Hee Yanga,b,1, Jin Woo Choic,1, Dongeun Huhd, Hyung Ah Joa, Sejoong Kima,e, Chun Soo Lima,b,f, Jung Chan Leeg,h,i, Hee Chan Kimg,h,i, Hyug Moo Kwonj, Chang Wook Jeongk, ⁎ Cheol Kwakk, Kwon Wook Jooa,b, Yon Su Kima,b, Dong Ki Kima,b, a

Department of Internal Medicine, Seoul National University College of Medicine, Seoul, Republic of Korea Kidney Research Institute, Seoul National University, Seoul, Republic of Korea c Interdisciplinary Program in Bioengineering Major, Graduate School, Seoul National University, Seoul, Republic of Korea d Department of Bioengineering, University of Pennsylvania, Philadelphia, USA e Department of Internal Medicine, Seoul National University Bundang Hospital, Seongnam, Republic of Korea f Department of Internal Medicine, Seoul National University Boramae Medical Center, Seoul, Republic of Korea g Department of Biomedical Engineering, Seoul National University College of Medicine, Seoul, Republic of Korea h Department of Biomedical Engineering, Seoul National University Hospital, Seoul, Republic of Korea i Institute of Medical and Biological Engineering, Medical Research Center, Seoul National University, Seoul, Republic of Korea j School of Life Sciences, Ulsan National Institute of Science and Technology, Ulsan, Republic of Korea k Department of Urology, Seoul National University Hospital, Seoul, Republic of Korea b



Keywords: Podocyte Differentiation Microfluidics Retinoic acid

Due to the distinct features that distinguish immortalized podocyte cell lines from their in vivo counterparts, primary cultured human podocytes might be a superior cell model for glomerular disease studies. However, the podocyte de-differentiation that occurs in culture remains an unresolved problem. Here, we present a method to differentiate primary cultured podocytes using retinoic acid (RA) and fluid shear stress (FSS), which mimic the in vivo environment of the glomerulus. RA treatment induced changes in the cell shape of podocytes from a cobblestone-like morphology to an arborized configuration with enhanced mobility. Moreover, the expression of synaptopodin and zonula occludens (ZO)−1 in RA-treated podocytes increased along with Krüppel-like factor 15 (KLF15) expression. Confocal microscopy revealed that RA increased the expression of cytoplasmic synaptopodin, which adopted a filamentous arrangement, and junctional ZO-1 expression, which showed a zipper-like pattern. To elucidate the effect of FSS in addition to RA, the podocytes were cultured in microfluidic devices and assigned to the static, static+RA, FSS, and FSS+RA groups. The FSS+RA group showed increased synaptopodin and ZO-1 expression with prominent spikes on the cell-cell interface. Furthermore, interdigitating processes were only observed in the FSS+RA group. Consistent with these data, the mRNA expression levels of synaptopodin, podocin, WT-1 and ZO-1 were synergistically increased by FSS and RA treatment. Additionally, the heights of the cells were greater in the FSS and FSS+RA groups than in the static groups, suggesting a restoration of the 3D cellular shape. Meanwhile, the expression of KLF15 increased in the RA-treated cells regardless of fluidic condition. Taken together, FSS and RA may contribute through different but additive mechanisms to the differentiation of podocytes. These cells may serve as a useful tool for mechanistic studies and the application of regenerative medicine to the treatment of kidney diseases.

1. Introduction Podocytes are terminally differentiated cells that cover the surface of the glomerular basement membrane and play a crucial role in regulating the glomerular filtration barrier [1]. Thus, genetic abnorm-


alities in podocyte proteins and podocyte injury are major causes of hereditary and acquired kidney diseases, respectively [2,3]. In this regard, many laboratories have focused their efforts on understanding podocyte pathobiology in various models of glomerular disease. Traditional animal models are considered to be the gold standard

Correspondence to: Department of Internal Medicine, Seoul National University Hospital, 101 Daehak-ro, Jongro-gu, Seoul 110-744, Republic of Korea. E-mail address: [email protected] (D.K. Kim). These authors contributed equally to this work. Received 30 September 2016; Received in revised form 10 March 2017; Accepted 13 March 2017 0014-4827/ © 2017 Elsevier Inc. All rights reserved.

Please cite this article as: Yang, S.H., Experimental Cell Research (2017),

Experimental Cell Research xxx (xxxx) xxx–xxx

S.H. Yang et al.

for research related to podocyte injuries or diseases. However, animal models do not always completely replicate human disease phenotypes [4–6]. In vitro cell culture studies offer an important alternative to animal experiments or at least provide data that supplement the data from animal experiments and allow for the precise manipulation of cellular environments, thereby permitting a more detailed understanding of cellular biology. Podocyte cell lines have been widely used for more than two decades due to their high availability and differentiation potential, which can be achieved through the conditional immortalization through the use of oncogenes [7,8]. Even though these podocyte cell lines share biochemical characteristics with podocytes in vivo [7], cell lines might differ from their in vivo counterparts in their expression levels of genes and proteins involved in important cellular functions [9–12]. Moreover, podocyte cell markers, including synaptopodin and nephrin, are expressed at significantly lower levels in podocyte cell lines than in podocytes in vivo [13]. Therefore, primary cultured human podocytes that has been freshly isolated from human kidneys are increasingly being recognized as a superior cell source for mechanistic research. Primary cultured podocytes are of particular interest in studies evaluating organ-on-a-chip technologies and, more importantly, evaluating primary cultured podocytes as an important source of cells for kidney regenerative medicine [14,15]. A widely accepted method of podocyte primary culture is to isolate the outgrowing podocytes from the isolated glomeruli [16]. Unfortunately, however, outgrowing podocytes rapidly de-differentiate, losing their cell-specific markers and specialized architecture [17]. With this in mind, it is important to reconfigure the functions and morphologies of podocytes by inducing differentiation using a technique other than the ectopic expression of oncogenes. Recently, fluid shear stress (FSS), which normally occurs through the nephron, was found to induce cytoskeletal reorganization and functional polarization of renal tubular cells [18,19]. This finding indicates that the in vitro reconstitution of nephrons via microfluidic techniques can lead to the functional and morphological differentiation of the cells in the kidneys. However, whether FSS can induce the differentiation of podocytes has remained elusive. Here, we introduce a novel in vitro cell culture protocol to induce differentiation of primary cultured human podocytes using a simple microfluidic device fabricated with polydimethyl siloxane (PDMS) in combination with all-trans retinoic acid (RA) stimulation, which may induce Krüppel-like factor 15 (KLF15), a novel transcriptional regulator of podocyte differentiation [20].

Fig. 1. Flow chart of the experimental procedure for the human primary podocytes differentiation with a microfluidics device.

5% CO2 atmosphere. The media consisted of DMEM/F12 (Lonza, Basel, Switzerland) supplemented with 15% fetal bovine serum (FBS) (Gibco, MA, USA), 1× insulin-transferrin-selenium (Gibco), 10 mmol HEPES buffer (Sigma-Aldrich, MO, USA), 200 µmol L-glutamine (Gibco), 50 nmol hydrocortisone (Sigma-Aldrich), 100 U/ml penicillin (Gibco) and 100 pg/ml streptomycin (Gibco) and was changed every 3 days. To determine the optimal media, the media was supplemented in some experiments with 2% or 15% FBS. Plating was routinely done on plastic dishes coated with 10 µg/ml fibronectin (Sigma-Aldrich). 2.2. Human primary podocyte differentiation with a microfluidics device 2.2.1. Microfluidic device fabrication A single channel microfluidic device was designed to culture the human primary podocytes inside the channel. The microdevice was fabricated using soft lithography, as reported previously [21]. Briefly, an SU-8 master mold was fabricated through photolithography. A mixture of PDMS (Sygald, MI, USA) and curing agent (10:1 w/w ratio of PDMS to curing agent) was poured onto the master mold and baked at 75 °C for 4 h. The dimensions of the microdevice were 750 µm (width) ×250 µm (height) ×1.5 µm (length). For bonding, the PDMS mold containing the desired channel and an empty PDMS slab were treated with plasma in a plasma cleaner (Harrick Plasma, NY, USA). The microdevice was reversibly bonded by exposing the PDMS parts to plasma for one second to ease the post-analysis processes, such as quantitative PCR and immunostaining, and to reduce cell loss.

2. Materials and methods 2.1. Human primary podocytes preparation The schematic experimental procedure is presented in Fig. 1. Human primary podocytes were harvested as previously reported [16]. Surgically resected kidney specimens were obtained from patients diagnosed with renal cell carcinoma, and the kidney cortices were dissected mechanically. The glomeruli were isolated by sieving techniques, and the isolated glomeruli were cultured for 8 days. The outgrowing cells were trypsinized and passed through sieves with a 25-µm pore size to remove the remaining glomerular cores, which primarily consisted of mesangial cells [7]. To quantitatively analyze the podocyte from the isolated glomeruli, flow cytometry analysis was conducted. On day 8, the cultured cells were enumerated, and 1×106 cells were incubated with Fc receptor blocking reagent (1 μg/ml, BD Bioscience, CA, USA). Podocyte were identified using rabbit antihuman purified anti-Nephrin (Abcam, Cambridge, UK) and FITClabeled anti-rabbit IgG (BD PharMingen, CA, USA). Human primary glomerular cells were determined with PE-conjugated anti-CD31 for endothelial cells (BD PharMingen) and PE-conjugated anti-CD90 for mesangial cell (BD PharMingen). Stained cells were sorted and analyzed using fluorescence-activated cell sorting (FACS) Calibur instrument (BD Biosciences). The cells were kept at 37 °C in a humid

2.3. Schematics of system and experimental timeline The microfluidic device was treated with UV light for 30 min to sanitize the inside of the channel, which was coated with fibronectin (10 µg/ml) and incubated for 1 h. The podocytes were plated on the device (3×105 cells/device) and allowed to settle for 4 h in the incubator at 37 °C, and 5% CO2 for attachment. Before the shear stress was applied, the podocytes were exposed to a low shear stress and cultured for 1 day in the incubator to allow for adjustment to the microenvironment. Then, different conditions were employed to test the podocytes: static with medium containing vehicle (DMSO, Sigma2

Experimental Cell Research xxx (xxxx) xxx–xxx

S.H. Yang et al.

Aldrich); static with medium containing RA (Sigma-Aldrich); fluidic with medium containing vehicle; and fluidic with medium containing RA. Various magnitudes of FSS 0.5–2 dyne/cm 2) and various concentrations of RA (1 and 2 µM) were applied to the de-differentiated human podocytes for 4 days.

Real-time qPCR was performed using Assay-on-Demand TaqMan probes and primers for synaptopodin, zonula occludens (ZO)−1, podocin, P-cadherin, WT-1, KLF15 and GAPDH (Applied Biosystems, CA, USA) and an ABI PRISM 7500 sequence detection system. Relative quantification was performed using the 2-ΔΔCT method. GAPDH was used as a loading control. All experiments were completed in triplicate.

2.4. Assessment of podocyte differentiation 2.5. Western blot analysis 2.4.1. Real-time quantitative PCR analysis Human primary podocytes were harvested from the microfluidics device after 4 days of induction. Total RNA was extracted from human primary podocytes, and the mRNA levels of target genes were assayed by real-time quantitative PCR. Briefly, total RNA was isolated from the primary cultured podocytes using the RNeasy kit (Qiagen GmBH, Germany), and 500 ng of total RNA was reverse-transcribed using oligo-d(T) primers and AMV-RT Taq polymerase (Promega, WI, USA).

The podocytes were harvested from culture plates and proteins were extracted using RIPA buffer containing Halt protease inhibitor (Pierce, IL, USA). Western immunoblotting was performed using primary antibodies against synaptopodin (Progen, Heidelberg, Germany), ZO-1 (Invitrogen, NY, USA), KLF15 (Abcam) and β-actin (Sigma-Aldrich). Briefly, equal amounts (30 μg) of extracted protein were separated by 10% SDS-polyacrylamide gels and transferred onto

Fig. 2. Characterization of human primary podocytes. (A) Representative inverted phase-contrast images of outgrowing cells from glomeruli on days 0, 6, 8 and 10 (Original magnification, ×200). (B) Representative FACS profiles of nephrin and CD90 in cells at day 8. (C) Representative FACS profile showing the sorting strategy used to isolate nephrinpositive and CD31-negative cells. (D) Representative confocal image of FACS-sorted nephrin-positive and CD31-negative cells (Original magnification, ×400).


Experimental Cell Research xxx (xxxx) xxx–xxx

S.H. Yang et al.

Immobilon-FL 0.4 μM polyvinylidene difluoride membranes (Millipore, MA, USA). Anti-rabbit IgG (Cell Signaling Technology, MA, USA) and anti-mouse IgG (Cell Signaling Technology) were used as HRP-conjugated secondary antibodies. Labeled proteins were detected using an enhanced chemiluminescence system (ECLTM PRN 2106; Amersham Pharmacia Biotech, Buckinghamshire, UK), and the intensities of the bands were analyzed using a gel documentation system (Bio-Rad Gel Doc 1000 and Multi-Analyst®version 1.1).

High power images of ZO-1 (Invitrogen) were acquired using a Leica TCS Super-resolution stimulated emission depletion (STED)-CW laser scanning microscope. Briefly, podocytes on the PDMS slabs were stained with the appropriate antibodies, mounted in ProLong Gold (Molecular Probes) and covered with a 22×22-mm cover glass (Fisher Scientific). STED images (1024×1024 pixels) were acquired through a ×100/1.4 numerical aperture oil immersion objective lens at a scan speed of 1000 lines/sec. The acquired images were deconvoluted after background subtraction using Leica LAS AF software Version 2.6.07266 (Leica). The data sets were then visualized using BioImageXD 1.0 software. For live images, the podocytes in the culture plates were stained with GFP-cell Tracker™ (Dharmacon, CO, USA) according to the manufacturer's instructions and imaged using a Leica TCS SP8 inverted confocal microscope (Leica) for 48 h.

2.6. Confocal microscopic examination The microfluidics devices were disassembled for immunofluorescence staining, and the cells on the PDMS slabs were washed with PBS and fixed in 4% paraformaldehyde for 20 min. Following fixation, the cells were permeabilized with 0.3% Triton X and stained with antibodies against nephrin (Abcam), synaptopodin (Progen), ZO-1 (Invitrogen) and KLF15 (Abcam) in a blocking agent overnight at 4 °C. Alexa 488/555-conjugated probes (Molecular Probes, OR, USA) were used as secondary antibodies and 4′,6-diamidino-2-phenylindole (DAPI; Molecular Probes) was used to counterstain the nuclei. The primary antibodies were omitted in the negative controls. After stimulation, Z-sectioned fluorescent images were used to determine the localization of synaptopodin in the podocytes. Immunofluorescence images were acquired with a confocal microscope (Leica TCS SP8, Leica Microsystem GmbH Wetzlar, Germany) and were analyzed with Leica IMARIS 7.6.

2.7. Scanning electron microscopic examination The cells were fixed with a mixture of cold 2.5% glutaraldehyde in 0.1 M phosphate buffer (pH 7.2) (Sigma) and 2% paraformaldehyde in 0.1 M phosphate buffer (pH 7.2) (Sigma) overnight. After, the sample was post-fixed in 2% osmium tetroxide in 0.1 M phosphate buffer at 37 °C for 1.5 h. The sample was briefly washed with phosphate buffer solution and rapidly dehydrated with ethanol. Scanning electron microscope (SEM) images were acquired using JSM 7410F at an accelerating voltage of 5 kV (JEOL Ltd.) with veleta camera at x6000

Fig. 3. Effects of RA on podocyte morphology. (A) Representative inverted phase-contrast images of human primary podocytes treated with vehicle, 1 µM or 2 µM RA for 0, 12 and 24 h (Original magnification, ×400). (B) Representative still images from live images of green fluorescent protein-transfected podocytes treated with vehicle (supplementary video 1.mp4), 1 µM (supplementary video 2.mp4) or 2 µM RA (supplementary video 3.mp4) for 48 h (Original magnification, ×1000).


Experimental Cell Research xxx (xxxx) xxx–xxx

S.H. Yang et al.

under an inverted phase-contrast microscope. The cells showed a cobblestone-like morphology with a high rate of proliferative activity until day 10, which indicated that the cells, including podocytes, were morphologically de-differentiated (Fig. 2A). On day 8 of culture, the cells reached approximately 80% confluence and were analyzed by flow cytometry, as these cells may have had heterogeneous origins from three major glomerular resident cells (podocytes, mesangial cells and endothelial cells). Flow cytometry revealed that less than 1% of the outgrowing cells from the glomeruli expressed the mesangial cell marker CD90 (Fig. 2B). Therefore, the glomerular cells were stained with CD31 and nephrin to distinguish endothelial cells and podocytes, respectively, before sorting with FACS. Among the heterogeneous population, the CD31-negative and nephrin-positive portion was considered a homogenous collection of podocyte-derived cells. The results of flow cytometry showed that 68.9 ± 10.1% of cells were podocytes; around 7×105 of these cells were collected among 1×106 input cells for further analysis (Fig. 2C). The purity of FACS-sorted cells was verified by confocal microscopy, which revealed a prevalent localization of nephrin in the cytoplasm without the expression of CD31, and the purity of sorted podocytes was > 98% (Fig. 2D). These results indicated that the FACS-sorted cells probably originated from podocytes and were de-differentiated based on their morphological features.

magnification. 2.8. Statistical analysis The results were expressed as the means ± SD or means ± SEM where indicated. Statistical analysis was performed using GraphPad Prism 5.0 (Graph Pad Software, CA, USA). A P-value < 0.05 was considered statistically significant. 2.9. Ethics statement The study protocol complies with the Declaration of Helsinki and received full approval from the institutional review board at the Seoul National University Hospital (no. H-1306-108-500). All the samples were immediately recruited, stored and monitored by the Seoul National University Hospital Human Biobank. 3. Results 3.1. Purification of human primary de-differentiated podocytes Around 200 to 300 glomeruli were harvested from the kidney tissue. Cells outgrowing from isolated glomeruli were observed daily

Fig. 4. Effects of RA on podocyte expression of KLF15, ZO-1 and synaptopodin. (A) Representative western blot and relative densitometric quantification of KLF15, ZO-1 and synaptopodin (representative of four blots) in podocytes after 48-h exposure to vehicle, 1 µM or 2 µM RA. Data are expressed as the fold change in protein expression relative to vehicletreated cells (*p < 0.05 vs. vehicle group; ** p < 0.001 vs. vehicle group). (B) Representative confocal microscopy images of podocytes after 48-h exposure to vehicle (left) and 1 µM RA (right). DAPI (nucleus, blue), synaptopodin (green), and ZO-1 (red) (Original magnification, ×200; ×800 [white box]).


Experimental Cell Research xxx (xxxx) xxx–xxx

S.H. Yang et al.

3.2. Effects of RA on podocyte differentiation

3.3. Effects of FSS with RA on podocyte differentiation

To test whether the differentiation of primary cultured podocytes was induced by RA, the morphological changes of the podocytes were observed for 24 h after exposure to RA or vehicle. Podocytes treated with RA changed in shape to an arborized configuration and increased in size, whereas podocytes in the absence of RA retained their cobblestone-like morphology. However, RA did not appear to have a significant dose-dependent effect on podocyte morphology (Fig. 3A). Cell mobility was also analyzed via live imaging of GFP-transfected podocytes; these cells showed increased mobility in the presence of RA (Fig. 3B and supplementary video 1–3). In addition to the changes in morphology and mobility, the expression of KLF15, synaptopodin and ZO-1 proteins were evaluated by western blotting to verify the effects of RA on molecular differentiation (Fig. 4A). As expected, the protein levels of KLF15, a key transcriptional regulator of podocyte differentiation, were significantly increased in podocytes treated with RA compared to those treated with vehicle only. Similarly, the protein levels of synaptopodin and ZO-1 were significantly increased in cells treated with RA. However, none of these effects exhibited a dosedependent pattern. Therefore, the subsequent experiments were performed using 1 µM RA. Confocal microscopy was then performed to further determine the subcellular localization and expression patterns of synaptopodin and ZO-1 (Fig. 4B). The RA-treated podocytes showed increased cytoplasmic expression of synaptopodin compared with the vehicle-treated podocytes. Moreover, in RA-treated podocytes, synaptopodin staining showed a punctate expression pattern with a filamentous arrangement, which is typical of human podocytes (Fig. 4B [right upper], white box). The expression of ZO-1 also increased in RAtreated podocytes compared to cells treated with vehicle only. Moreover, dense ZO-1 staining in a zipper-like pattern was observed in the cell periphery of only RA-treated podocytes (Fig. 4B [right lower], white box). Supplementary material related to this article can be found online at

To test the strength of FSS required to induce differentiation, podocytes were cultured in the microfluidics channels under static and fluidic FSS conditions at 0.2, 0.5, 1.0, and 2.0 dyne/cm 2 for 4 days. The mRNA expression of synaptopodin and ZO-1 significantly increased in the podocytes subjected to FSS to 0.5 and 0.2 dyne/cm 2, respectively, compared to the cells under the static condition. Thus, the subsequent experiments were performed with FSS of 0.5 dyne/cm2 (Supplementary Fig 1). To evaluate the effect of FSS with or without concomitant RA stimulation on podocyte differentiation, the cells were grouped into 4 conditions: static, static+RA, FSS and FSS+RA. As shown in Fig. 5A and B, confocal microscopy analysis of the podocytes after 4 days of culture revealed that the signal for synaptopodin significantly increased in the FSS and FSS+RA groups compared with the static and static+RA groups. Although the expression of synaptopodin in the FSS+RA group was not significantly higher than that in the FSS group, the FSS+RA group showed enhanced peripheral expression of synaptopodin in a filamentous arrangement (Fig. 5A, white arrowhead). This result indicates that FSS and RA might synergistically induce podocyte differentiation. Meanwhile, KLF15 showed increased expression only in the RA-treated groups regardless of FSS, indicating that FSS might act independently of the transcriptional regulator KLF15 in mediating podocyte differentiation. In addition, superresolution fluorescence microscopy for ZO-1 was used to image the morphology of epithelial tight junctions (Fig. 5C). The overall intensity of ZO-1 increased in the FSS and FSS+RA groups. Moreover, spikes on the cell-cell interfaces were observed in the static+RA, FSS and FSS +RA groups; these spikes were most prominent in the FSS+RA group. Furthermore, SEM revealed the formation of interdigitating cell processes only in the FSS+RA group (Fig. 5D). Consistent with these fluorescence microscopy data, the mRNA expression of synaptopodin and ZO-1 simultaneously increased in the FSS+RA group. Moreover, the specific podocyte markers podocin and WT-1 showed mRNA expression patterns that were similar to that of synaptopodin.

Fig. 5. Effects of fluid shear stress and RA on podocyte differentiation. (A) Representative confocal microscopy images of DAPI (blue), synaptopodin (green) and KLF15 (red) in podocytes cultured under different conditions: static, static+RA, FSS, and FSS+RA (Original magnification, ×400; ×800 [white box]; arrowhead, interdigitating cell processes). (B) Relative fluorescence signal intensity of synaptopodin and KLF15 by the podocytes. Data are expressed as the fold change in fluorescence signal intensity relative to the static group (*P < 0.05; **P < 0.01; ***P < 0.005; NS, no significance). (C) Representative super-resolution microscopy images of podocytes stained with ZO-1 (Original magnification, ×3000; scale bar, 5 µm). (D) Representative scanning electron microscope images of podocytes (Original magnification, ×6000; scale bar, 1 µm).


Experimental Cell Research xxx (xxxx) xxx–xxx

S.H. Yang et al.

Fig. 6. mRNA expression levels of synaptopodin, ZO-1, podocin, WT-1, ZO-1, P-cadherin and KLF15 in podocytes cultured under different conditions: static, static+RA, FSS, and FSS +RA. Data are expressed as the fold change in mRNA expression relative to the static group (*P < 0.05; **P < 0.01; ***P < 0.005; NS, no significance).

filtration barrier, the structural and molecular abnormalities of podocytes may be important pathophysiological features of various glomerular diseases [22,23]. Thus, proper in vitro models are required for the analysis of podocyte differentiation. While human-relevant in vitro podocyte models may be crucial to studying glomerular diseases, traditional in vitro models primarily consist of immortalized cell lines cultured under static conditions; these models are limited by their inability to reconstitute the dynamic aspects of their in vivo counterparts as well as the podocyte cell lines themselves [13,24]. In this respect, the continuous exposure of podocytes to mechanical shear stress induced by urinary flow may be a dynamic aspect that must be recreated in in vitro podocytes models. In the present study, we successfully induced the molecular and structural differentiation of primary cultured human podocytes via the application of FSS using a simple single-layer microfluidic device and chemical stimulation with

Meanwhile, the mRNA expression of P-cadherin increased in groups treated with FSS independent of RA treatment. The KLF15 mRNA expression showed a similar pattern in the fluorescence microscopy data (Fig. 6). To indirectly assess the effects of FSS and/or RA on the shape of the podocytes, the heights of the podocytes were measured using confocal immunofluorescence microscopy. Regardless of RA treatment, the height of cells in the FSS and FSS+RA groups showed significantly greater heights compared to cells under the static condition, indicating that FSS might partly restore the three-dimensional cellular shape from the flattened morphology that is commonly observed in cells cultured under the usual static conditions (Fig. 7).

4. Discussion Due to their contribution to the formation of the glomerular 7

Experimental Cell Research xxx (xxxx) xxx–xxx

S.H. Yang et al.

Fig. 7. Mean heights of podocytes under different conditions: static, static+RA, FSS, and FSS+RA. Confocal microscope images shown with X-Y and Y-Z reconstruction (Original magnification, ×400) (*P < 0.05; **P < 0.01; ***P < 0.005; NS, no significance).

lity; 2) the increased expression of synaptopodin, a crucial podocyte differentiation marker, in a filamentous arrangement; and 3) the increased junctional expression of ZO-1 in a zipper-like pattern, which is a characteristic finding in differentiated podocytes [33]. The expression of the nuclear transcription factor KLF15, which mediates podocyte differentiation [20], increased in RA-treated cells, indicating that the RA-induced differentiation of podocytes may have been mediated by KLF15. In addition to the biochemical stimulation using RA, the in vitro podocytes acquired in vivo-like structures and functions when cultured under continuous FSS. Compared to the static and static+RA groups, the cytoskeletal and slit diaphragm complex proteins of the podocytes cultured under FSS were significantly increased. Notably, the expression of the actin-binding cytoskeletal protein synaptopodin, which is highly expressed in the dynamic part of podocytes (such as the foot process) in vivo [34], was increased in the peripheral cytoplasm of podocytes under FSS and RA stimulation. These treatment also resulted in the formation of interdigitating processes. Similarly, the tight junctional protein ZO-1 also increased in the cell-cell junctions of cells in the FSS+RA group; in these cells, the formation of spikes, which could be the in vitro equivalents of podocyte foot processes, was also observed [35]. Interestingly, the heights of the podocytes under FSS increased compared to those under static conditions, indicating that the structural differentiation of the cytoskeleton and junctions induced by FSS enabled the cells to maintain their natural three dimensional shape. Several studies support the effects of FSS on cellular heights and cytoskeletal reorganization of kidney cells, including in the inner medullary collecting duct cells and proximal tubule cells [18,27]. Despite the fact that this system successfully induced differentiation of human primary cultured podocytes, there are some limitations to the present study. First, further studies are needed to validate the use of this microfluidic culture system to generate an in vitro model of podocyte diseases. Second, the present single-layer microfluidic culture system cannot be used to assess cell-to-cell interactions; thus, the structure of the microdevice system should be modified for greater complexity such that it mimics the physiology of the glomerulus. For example, a system with two or more layers containing other glomerular resident cells may reveal mechanisms of podocyte-endothelial crosstalk, enable functional analyses of the glomerular filtration barrier, and

RA. This culture system effectively induced cell differentiation without the use of oncogenes, which can potentially change the differentiated phenotype [25]. Previous studies have demonstrated that proximal tubular cells exposed to microfluidic conditions exhibit both structural and functional differentiation regarding cytoskeleton organization, cilia formation, ion transporter polarization, and increased expression of adhesion and junctional complexes [18,26–28]. As a consequence, renal cells cultured under microfluidic conditions are able to more precisely recapitulate the in vivo response to toxic stimuli [18,29,30], suggesting that culture conditions that mimic the in vivo fluidic conditions of the kidney might provide a more relevant in vitro model of kidney cells. However, although several previous studies have achieved microfluidic culture systems using tubule cells of the nephron [18,26–29], those authors did not clarify whether the microfluidic system could be used to differentiate primary cultured podocytes to achieve in vivo-like structures and functions. Unlike tubular epithelial cells, podocytes are difficult to purify due to the heterogeneity of the glomerular cells and terminally differentiated cells in vivo which, once de-differentiated, are difficult to redifferentiate [31,32]. In this study, we used FACS to purify podocytederived cells to reduce the likelihood of contamination by other glomerular resident cells. Because the outgrowing cells were passed through sieves with a 25-µm pore size, the remaining glomerular cores, which primarily consisted of mesangial cells, were removed. Thus, the proportion of CD90-positive mesangial origin cells comprised less than 1% of the outgrowing cells. While podocin and synaptopodin rapidly lose their expression after de-differentiation, nephrin has been reported to maintain its expression in the cytoplasm of podocytes after de-differentiation [8]. Therefore, in this study, podocyte origin cells were positively selected using FACS with an anti-nephrin antibody. After purification of the podocyte-derived cells, we used simultaneous biophysical and biochemical stimulation to effectively induce and maintain the differentiated status of podocytes. The podocytes were then cultured under FSS in media containing RA, which previously was reported to induce the differentiation of primary cultured murine podocytes [20]. As expected, RA successfully triggered the differentiation of human podocytes, showing 1) the induction of a morphological change into an arborized configuration with enhanced cellular mobi8

Experimental Cell Research xxx (xxxx) xxx–xxx

S.H. Yang et al.

further improve podocyte differentiation. Nevertheless, this system is relatively faster than existing methods and increases the induction of human primary podocyte differentiation, which has posed a challenge to ongoing investigations of podocyte biology. Moreover, this system integrates human-derived primary podocytes with a microfluidic environment that is similar to the environment experienced by cells in vivo and that can provide more biologically significant data pertaining to human kidney diseases. Furthermore, the differentiation of podocytes without the use of ectopic oncogene expression might serve as a useful cell source for regenerative medicine in kidney diseases in terms of avoiding the risk of infection, cancer and nonimmune toxic effects of immunosuppression [36].






Authors’ contributions [15]

SHY and JWC performed the experiments, analyzed the results, and drafted the manuscript. DH, HAJ, SK, CSL, JCL, HCK, HMK, CWJ, CK, KWJ and YSK conceived the experiment and interpreted the data. DKK designed the study, analyzed the results, interpreted the data, and reviewed the manuscript. All of the authors read and approved the final manuscript.

[16] [17] [18] [19]

Conflicts of interests


The authors have no conflicts of interests to declare in relation to this article.





This work was supported by a grant from the Korea Healthcare Technology R & D Project, Ministry of Health and Welfare, Republic of Korea (HI13C1921). The Biospecimens were provided by the Seoul National University Hospital Human Biobank, a member of the National Biobank of Korea, which is supported by the Ministry of Health and Welfare, Republic of Korea.

[24] [25] [26]


Appendix A. Supporting information


Supplementary data associated with this article can be found in the online version at doi: 10.1016/j.yexcr.2017.03.026.




[1] H. Pavenstadt, W. Kriz, M. Kretzler, Cell biology of the glomerular podocyte, Physiol. Rev. 83 (1) (2003) 253–307. [2] J.J. Li, S.J. Kwak, D.S. Jung, et al., Podocyte biology in diabetic nephropathy, Kidney Int. Suppl. 106 (2007) S36–S42. [3] H. Cheng, R.C. Harris, The glomerulus–a view from the outside – the podocyte, Int. J. Biochem. Cell Biol. 42 (9) (2010) 1380–1387. [4] L.G. Griffith, M.A. Swartz, Capturing complex 3D tissue physiology in vitro, Nat. Rev. Mol. Cell Biol. 7 (3) (2006) 211–224. [5] A. Rangarajan, S.J. Hong, A. Gifford, R.A. Weinberg, Species- and cell type-specific requirements for cellular transformation, Cancer Cell. 6 (2) (2004) 171–183. [6] G.J. Becker, T.D. Hewitson, Animal models of chronic kidney disease: useful but not perfect, Nephrol. Dial. Transplant. 28 (10) (2013) 2432–2438. [7] S.J. Shankland, J.W. Pippin, J. Reiser, P. Mundel, Podocytes in culture: past present, future, Kidney Int. 72 (1) (2007) 26–36. [8] M.A. Saleem, M.J. O'Hare, J. Reiser, et al., A conditionally immortalized human podocyte cell line demonstrating nephrin and podocin expression, J. Am. Soc. Nephrol. 13 (3) (2002) 630–638. [9] C.S. Alge, S.M. Hauck, S.G. Priglinger, A. Kampik, M. Ueffing, Differential protein



[33] [34]




profiling of primary versus immortalized human RPE cells identifies expression patterns associated with cytoskeletal remodeling and cell survival, J. Proteome Res. 5 (4) (2006) 862–878. C. Pan, C. Kumar, S. Bohl, U. Klingmueller, M. Mann, Comparative proteomic phenotyping of cell lines and primary cells to assess preservation of cell typespecific functions, Mol. Cell Proteom. 8 (3) (2009) 443–450. A.D. van der Meer, V.V. Orlova, P. Ten Dijke, A. van den Berg, C.L. Mummery, Three-dimensional co-cultures of human endothelial cells and embryonic stem cellderived pericytes inside a microfluidic device, Lab Chip 13 (18) (2013) 3562–3568. A. Birgersdotter, R. Sandberg, I. Ernberg, Gene expression perturbation in vitro–a growing case for three-dimensional (3D) culture systems, Semin. Cancer Biol. 15 (5) (2005) 405–412. S. Chittiprol, P. Chen, D. Petrovic-Djergovic, T. Eichler, R.F. Ransom, Marker expression, behaviors, and responses vary in different lines of conditionally immortalized cultured podocytes, Am. J. Physiol. Ren. Physiol. 301 (3) (2011) F660–F671. J.J. Song, J.P. Guyette, S.E. Gilpin, G. Gonzalez, J.P. Vacanti, H.C. Ott, Regeneration and experimental orthotopic transplantation of a bioengineered kidney, Nat. Med. 19 (5) (2013) 646–651. F. An, Y. Qu, X. Liu, R. Zhong, Y. Luo, Organ-on-a-chip: new platform for biological analysis, Anal. Chem. Insights 10 (2015) 39–45. P. Mundel, J. Reiser, W. Kriz, Induction of differentiation in cultured rat and human podocytes, J. Am. Soc. Nephrol. 8 (5) (1997) 697–705. L. Ni, M. Saleem, P.W. Mathieson, Podocyte culture: tricks of the trade, Nephrol. 17 (6) (2012) 525–531. K.J. Jang, A.P. Mehr, G.A. Hamilton, et al., Human kidney proximal tubule-on-achip for drug transport and nephrotoxicity assessment, Integr. Biol. (2013). Y. Duan, N. Gotoh, Q. Yan, et al., Shear-induced reorganization of renal proximal tubule cell actin cytoskeleton and apical junctional complexes, Proc. Natl. Acad. Sci. USA 105 (32) (2008) 11418–11423. S.K. Mallipattu, R. Liu, F. Zheng, et al., Kruppel-like factor 15 (KLF15) is a key regulator of podocyte differentiation, J. Biol. Chem. 287 (23) (2012) 19122–19135. D. Huh, B.D. Matthews, A. Mammoto, M. Montoya-Zavala, H.Y. Hsin, D.E. Ingber, Reconstituting organ-level lung functions on a chip, Science 328 (5986) (2010) 1662–1668. K. Tryggvason, J. Wartiovaara, Molecular basis of glomerular permselectivity, Curr. Opin. Nephrol. Hypertens. 10 (4) (2001) 543–549. S.J. Shankland, The podocyte's response to injury: role in proteinuria and glomerulosclerosis, Kidney Int. 69 (12) (2006) 2131–2147. D. Kim, X. Wu, A.T. Young, C.L. Haynes, Microfluidics-based in vivo mimetic systems for the study of cellular biology, Acc. Chem. Res. (2014). F. Castro-Munozledo, Corneal epithelial cell cultures as a tool for research, drug screening and testing, Exp. Eye Res. 86 (3) (2008) 459–469. Z. Du, Q. Yan, Y. Duan, S. Weinbaum, A.M. Weinstein, T. Wang, Axial flow modulates proximal tubule NHE3 and H-ATPase activities by changing microvillus bending moments, Am. J. Physiol. Ren. Physiol. 290 (2) (2006) F289–F296. K.J. Jang, K.Y. Suh, A multi-layer microfluidic device for efficient culture and analysis of renal tubular cells, Lab Chip 10 (1) (2010) 36–42. V. Raghavan, Y. Rbaibi, N.M. Pastor-Soler, M.D. Carattino, O.A. Weisz, Shear stress-dependent regulation of apical endocytosis in renal proximal tubule cells mediated by primary cilia, Proc. Natl. Acad. Sci. USA 111 (23) (2014) 8506–8511. L.C. Snouber, F. Letourneur, P. Chafey, et al., Analysis of transcriptomic and proteomic profiles demonstrates improved Madin-Darby canine kidney cell function in a renal microfluidic biochip, Biotechnol. Prog. 28 (2) (2012) 474–484. S. Kim, S.C. LesherPerez, B.C. Kim, et al., Pharmacokinetic profile that reduces nephrotoxicity of gentamicin in a perfused kidney-on-a-chip, Biofabrication 8 (1) (2016) 015021. A. Murakami, H. Oshiro, S. Kanzaki, et al., A novel method for isolating podocytes using magnetic activated cell sorting, Nephrol. Dial. Transplant. 25 (12) (2010) 3884–3890. M.J. Wilmer, C.P. Ng, H.L. Lanz, P. Vulto, L. Suter-Dick, R. Masereeuw, Kidney-ona-chip technology for drug-induced nephrotoxicity screening, Trends Biotechnol. 34 (2) (2016) 156–170. J. Reiser, W. Kriz, M. Kretzler, P. Mundel, The glomerular slit diaphragm is a modified adherens junction, J. Am. Soc. Nephrol. 11 (1) (2000) 1–8. P. Mundel, H.W. Heid, T.M. Mundel, M. Kruger, J. Reiser, W. Kriz, Synaptopodin: an actin-associated protein in telencephalic dendrites and renal podocytes, J. Cell Biol. 139 (1) (1997) 193–204. Y. Li, Y.S. Kang, C. Dai, L.P. Kiss, X. Wen, Y. Liu, Epithelial-to-mesenchymal transition is a potential pathway leading to podocyte dysfunction and proteinuria, Am. J. Pathol. 172 (2) (2008) 299–308. S.F. Badylak, D. Taylor, K. Uygun, Whole-organ tissue engineering: decellularization and recellularization of three-dimensional matrix scaffolds, Annu. Rev. Biomed. Eng. 13 (2011) 27–53.