TY - JOUR
AU1 - Flegeau,, Killian
AU2 - Rubin,, Sébastien
AU3 - Mucha,, Simon
AU4 - Bur,, Pauline
AU5 - Préterre,, Julie
AU6 - Siadous,, Robin
AU7 - L’Azou,, Béatrice
AU8 - Fricain,, Jean-Christophe
AU9 - Combe,, Christian
AU1 - Devillard,, Raphaël
AU1 - Kalisky,, Jérôme
AU1 - Rigothier,, Claire
AB - Abstract Background The development of an artificial glomerular unit may be pivotal for renal pathophysiology studies at a multicellular scale. Using a tissue engineering approach, we aimed to reproduce in part the specific glomerular barrier architecture by manufacturing a glomerular microfibre (Mf). Methods Immortalized human glomerular cell lines of endothelial cells (GEnCs) and podocytes were used. Cells and a three-dimensional (3D) matrix were characterized by immunofluorescence with confocal analysis, Western blot and polymerase chain reaction. Optical and electron microscopy were used to study Mf and cell shapes. We also analysed cell viability and cell metabolism within the 3D construct at 14 days. Results Using the Mf manufacturing method, we repeatedly obtained a cellularized Mf sorting human glomerular cells in 3D. Around a central structure made of collagen I, we obtained an internal layer composed of GEnC, a newly formed glomerular basement membrane rich in α5 collagen IV and an external layer of podocytes. The cell concentration, optimal seeding time and role of physical stresses were modulated to obtain the Mf. Cell viability and expression of specific proteins (nephrin, synaptopodin, vascular endothelial growth factor receptor 2 (VEGFR2) and von Willebrandt factor (vWF)) were maintained for 19 days in the Mf system. Mf ultrastructure, observed with EM, had similarities with the human glomerular barrier. Conclusion In summary, with our 3D bio-engineered glomerular fibre, GEnC and podocytes produced a glomerular basement membrane. In the future, this glomerular Mf will allow us to study cell interactions in a 3D system and increase our knowledge of glomerular pathophysiology. bioassembly, glomerular barrier, glomerular endothelial cells, podocytes, tissue engineering INTRODUCTION Although organ transplantation saves several thousand lives per year, a growing number of patients are on the organ waiting list. Kidney tissue engineering (TE), one of the most promising approaches to constructing an engineered kidney, needs to be developed to potentially overcome the problem of organ shortage [1]. Organ manufacture may be based on various methods: biofabrication, in vitro or in vivo TE, regenerative medicine, bioassembly and bioprinting [2]. The objective of TE is to reproduce a functional organ using an appropriate combination of cells and scaffolds. TE constructs would closely reproduce the architecture, physiology and morphology of the native organ at the cellular level. The final step might be the replacement of deficient organs by improving TE organ compatibility and functionality [3]. Focusing on kidney TE, techniques and methods may be separated if the focus is on tubule or glomerular structure construction. For the last decade, tubule structure has been obtained by TE or self-assemblage [4]. Glomerular manufacture, in contrast, remains an ambitious goal. Different approaches have been developed to obtain glomerular structures using inducible pluripotent stem cells, decellularized kidney scaffold and a combination of cells and scaffolds. Lanza et al. [5] used somatic cell nuclear transfer to demonstrate the ability of a clone of bovine renal cells seeded on a biodegradable scaffold and transplanted in vivo to secrete urinary fluid. A functional glomerulus was also created in vivo using murine embryonic cells. After 5 days of in vitro culture, embryonic cells were injected into the subcapsular renal space with VEGF. After 3 weeks, the formation of glomeruli was observed with reabsorption of albumin at the newly formed convoluted tubule [6]. Rudimentary urine was also obtained by decellularizing rat kidneys and reseeding them with epithelial and endothelial cells [7, 8]. The structure and ultrastructure of these constructions reproduced the glomeruli ultrastructure incompletely. Studies of cell–cell or cell–matrix interactions were then difficult and partial. Vascularization also modulates cell phenotype [9] or crosstalk between different cell types through physical and shear stresses [6, 10]. Fluid shear stresses induce cortical actin cytoskeleton reorganization and a relocalization of adherent complexes in podocytes [9]. Podocyte hypertrophy, apoptosis and detachment have also been reported [11]. Endothelial cells exhibit no adaptive mechanism under shear stress except proliferation [12]. However, TE is today limited by the lack of a microvasculature. Vascular TE is effective and successful, but a lot of work needs to be done to assemble a microvascularized tissue [13]. The development of a three-dimensional (3D) in vitro glomerular structure by bioassembly (building blocks corresponding to hybrid cells and materials construct) may comply with these requirements. For instance, no in vitro renal 3D cylindrical culture model has been reported in the literature to our knowledge. Some studies have developed co-culture models in 2D [14, 15]. Musah et al. [16] and Zhou et al. [17] reproduced the glomerular capillary wall in a chip and succeeded in producing pathophysiological conditions. However, these are 2D models with an artificial or synthetic matrix. Endothelial cells and podocytes were seeded on each side of a matrigel or a polydimethylsiloxane membrane in a chip—a flat structure very different from a capillary under fluid shear stress. We believe that the challenge will be to obtain a perfused 3D cylindrical glomerular unit. Onoe et al. [18] succeeded in producing microvessels containing different cell types such as fibroblasts, myocytes, endothelial cells, nerve cells and epithelial cells using a microfluidic device with double-coaxial laminar flow. The constructs were monocellular fibres without specific cell organization or lumen formation. In this study we developed a plain glomerular microfibre (Mf) that reproduces the 3D cylindrical structure of a glomerular barrier, that is, with, from inside out, an endothelial layer, a glomerular basement membrane (GBM) and podocytes. This project is a step in the development of a tissue-engineered glomerular filtration unit that could serve as an in vitro experimental model for a glomerular pathophysiology study. MATERIALS AND METHODS Cell culture Conditionally immortalized human glomerular cell lines of endothelial cells (GEnCs) and podocytes were used [19, 20]. The culture conditions have been previously described [21]. GEnCs were cultured in endothelial growth media-2 (EGM-2) MV SingleQuots supplemented with the EGM-2 MV BulleKit (Lonza, Walksville, MD, USA). Podocytes were cultured in Roswell Park Memorial Institute (RPMI) 1640 (Life Technologies, Carlsbad, CA, USA) supplemented with 10% foetal calf serum (GE Healthcare, Little Chalfont, UK), 1% insulin transferrin selenium (Gibco, Life Technologies, Carlsbad, CA, USA), 100 μg/mL streptomycin and 100 U/mL penicillin (Life Technologies). Construction of endothelial Mf As a first step, a shell of 2% reticulated alginate (Protanal Sodium Alginate LF 10/60 LS, FMC, Philadelphia, PA, USA) was formed around a polycarbonate capillary tube (internal diameter 350 μm) (Paradigm Optics, Vancouver, WA, USA) by successively soaking the capillary tube in the calcium chloride and sodium alginate solutions [21]. GEnCs were suspended in the collagen I solution (Life Technologies) and injected inside the capillary tube. Appropriate cell number and collagen I concentration were determinated for this study in the ‘Results’ section. Endothelial Mfs were cultured and deposited in six-well plates. Each well was occupied by one Mf. Endothelial Mfs were then seeded with human podocytes. The detailed method is described in the ‘Results’ section. Cell viability assay Cell viability was assessed using a LIVE/DEAD Viability/Cytotoxicity Kit (Life Technologies) following the manufacturer’s instructions. Results were analysed by fluorescence using a DMI3000 B microscope and Leica Application Suite software (Leica, Wetzlar, Germany). Immunofluorescence Mfs were washed with 1 phosphate-buffered saline (PBS) 1× and fixed using 4% paraformaldehyde. Mfs were then washed three times with PBS 1× and stored at 4°C for further experiments. The Mf was deposited onto the paraffin film and permeabilized with 0.1% Triton 100× for 5 min at room temperature (RT). The Mf was washed three times with Tris-buffered saline (TBS) 1× and then blocked for 2 h at RT in blocking buffer [1% bovine serum albumin (BSA), 1% foetal calf serum diluted in TBS]. Primary antibodies (listed in Table 1) were diluted in blocking buffer, deposited onto the Mf and stored overnight at 4°C in the dark. After several washes, the Mf was incubated with the appropriate secondary antibody (listed in Table 1) at a 1/200 dilution, for 2 h at RT in the dark. After several washes, nuclei were labelled with 4′,6-diamidino-2-phenylindole solution (dilution 1/1000) for 10 min at RT. Finally, the Mf was deposited on a glass support dedicated to the immunofluorescence (IF) analysis and mounted using Fluoromount mounting medium (Southern Biotech, Birmingham, AL, USA). Samples were analysed using an SPE confocal microscope (Leica). Table 1 List of the different antibodies Antibodies . Source . Applications . Dilution . Origin . Primary  Nephrin, H-300 Rabbit pAb IF, WB IF: 1/100 Santa Cruz WB: 1/1000  Nephrin (1243–1256) Guinea Pig pAb WB WB: 1/500 Acris Antibodies  Synaptopodin Mouse mAb IF, WB IF: 1/150 Progen WB: 1/100  Podocin Rabbit pAb IF IF: 1/100 Santa Cruz  CD31 Mouse mAb IF, WB IF: 1/100 R&D Systems  vWF Mouse mAb IF IF: 1/100 R&D Systems  VEGFR2 Mouse mAb IF, WB IF: 1/100 R&D Systems WB: 1/500  Collagen IV Mouse mAb IF, WB IF: 1/200 Abcam WB: 1/100  Laminin Rabbit mAb WB WB: 1/500 Novotec  β-actin Mouse mAb WB WB: 1/1000 BD Transduction Secondary  Alexa Fluor 488 goat anti-rabbit IF 1/200 Molecular Probes  Alexa Fluor 594 goat anti-rabbit IF 1/200 Molecular Probes  Alexa Fluor 488 goat anti-mouse IF 1/200 Molecular Probes  Alexa Fluor 594 goat anti-mouse IF 1/200 Molecular Probes  Goat anti-rabbit HRP conjugated WB 1/10000 GE Healthcare  Goat anti-guinea pig HRP conjugated WB 1/5000 GE Healthcare  Goat anti-mouse HRP conjugated WB 1/5000 GE Healthcare Antibodies . Source . Applications . Dilution . Origin . Primary  Nephrin, H-300 Rabbit pAb IF, WB IF: 1/100 Santa Cruz WB: 1/1000  Nephrin (1243–1256) Guinea Pig pAb WB WB: 1/500 Acris Antibodies  Synaptopodin Mouse mAb IF, WB IF: 1/150 Progen WB: 1/100  Podocin Rabbit pAb IF IF: 1/100 Santa Cruz  CD31 Mouse mAb IF, WB IF: 1/100 R&D Systems  vWF Mouse mAb IF IF: 1/100 R&D Systems  VEGFR2 Mouse mAb IF, WB IF: 1/100 R&D Systems WB: 1/500  Collagen IV Mouse mAb IF, WB IF: 1/200 Abcam WB: 1/100  Laminin Rabbit mAb WB WB: 1/500 Novotec  β-actin Mouse mAb WB WB: 1/1000 BD Transduction Secondary  Alexa Fluor 488 goat anti-rabbit IF 1/200 Molecular Probes  Alexa Fluor 594 goat anti-rabbit IF 1/200 Molecular Probes  Alexa Fluor 488 goat anti-mouse IF 1/200 Molecular Probes  Alexa Fluor 594 goat anti-mouse IF 1/200 Molecular Probes  Goat anti-rabbit HRP conjugated WB 1/10000 GE Healthcare  Goat anti-guinea pig HRP conjugated WB 1/5000 GE Healthcare  Goat anti-mouse HRP conjugated WB 1/5000 GE Healthcare HRP: horseradish peroxidase; mAb: monoclonal antibody; pAb: polyclonal antibody. Open in new tab Table 1 List of the different antibodies Antibodies . Source . Applications . Dilution . Origin . Primary  Nephrin, H-300 Rabbit pAb IF, WB IF: 1/100 Santa Cruz WB: 1/1000  Nephrin (1243–1256) Guinea Pig pAb WB WB: 1/500 Acris Antibodies  Synaptopodin Mouse mAb IF, WB IF: 1/150 Progen WB: 1/100  Podocin Rabbit pAb IF IF: 1/100 Santa Cruz  CD31 Mouse mAb IF, WB IF: 1/100 R&D Systems  vWF Mouse mAb IF IF: 1/100 R&D Systems  VEGFR2 Mouse mAb IF, WB IF: 1/100 R&D Systems WB: 1/500  Collagen IV Mouse mAb IF, WB IF: 1/200 Abcam WB: 1/100  Laminin Rabbit mAb WB WB: 1/500 Novotec  β-actin Mouse mAb WB WB: 1/1000 BD Transduction Secondary  Alexa Fluor 488 goat anti-rabbit IF 1/200 Molecular Probes  Alexa Fluor 594 goat anti-rabbit IF 1/200 Molecular Probes  Alexa Fluor 488 goat anti-mouse IF 1/200 Molecular Probes  Alexa Fluor 594 goat anti-mouse IF 1/200 Molecular Probes  Goat anti-rabbit HRP conjugated WB 1/10000 GE Healthcare  Goat anti-guinea pig HRP conjugated WB 1/5000 GE Healthcare  Goat anti-mouse HRP conjugated WB 1/5000 GE Healthcare Antibodies . Source . Applications . Dilution . Origin . Primary  Nephrin, H-300 Rabbit pAb IF, WB IF: 1/100 Santa Cruz WB: 1/1000  Nephrin (1243–1256) Guinea Pig pAb WB WB: 1/500 Acris Antibodies  Synaptopodin Mouse mAb IF, WB IF: 1/150 Progen WB: 1/100  Podocin Rabbit pAb IF IF: 1/100 Santa Cruz  CD31 Mouse mAb IF, WB IF: 1/100 R&D Systems  vWF Mouse mAb IF IF: 1/100 R&D Systems  VEGFR2 Mouse mAb IF, WB IF: 1/100 R&D Systems WB: 1/500  Collagen IV Mouse mAb IF, WB IF: 1/200 Abcam WB: 1/100  Laminin Rabbit mAb WB WB: 1/500 Novotec  β-actin Mouse mAb WB WB: 1/1000 BD Transduction Secondary  Alexa Fluor 488 goat anti-rabbit IF 1/200 Molecular Probes  Alexa Fluor 594 goat anti-rabbit IF 1/200 Molecular Probes  Alexa Fluor 488 goat anti-mouse IF 1/200 Molecular Probes  Alexa Fluor 594 goat anti-mouse IF 1/200 Molecular Probes  Goat anti-rabbit HRP conjugated WB 1/10000 GE Healthcare  Goat anti-guinea pig HRP conjugated WB 1/5000 GE Healthcare  Goat anti-mouse HRP conjugated WB 1/5000 GE Healthcare HRP: horseradish peroxidase; mAb: monoclonal antibody; pAb: polyclonal antibody. Open in new tab Protein extraction Mf extraction required NP40 lysis buffer (Tris 50 mM pH 7.5, NaCl 120 mM, NP40 1%, β-glycerophosphate 40 mM, benzamidine 1 mM, ethylenediaminetetraacetic acid 1 mM) with 10 µL/mL protease and phosphatase inhibitors (Sigma, Lyon, France). Protein quantification was performed on all samples using bicinchoninic acid Protein Assay Reagent (Pierce, ThermoScientific, Brebières, France) and the Victor X3 Multilabel Plate Reader (PerkinElmer, Waltham, MA, USA). Western blot (WB) Equal amounts of proteins were prepared and loaded in each lane of sodium dodecyl sulphate–polyacrylamide gel electrophoresis (SDS-PAGE) gel. Extracted proteins were denatured and loaded in acrylamide SDS-PAGE gel. After protein transfer on polyvinylidene fluoride, membranes were blocked with TBS and Tween-20 (TBST; 10 mM Tris HCl pH 7.5, 140 mM NaCl, 0.1% Tween-20) and 5% BSA and then incubated with primary antibodies overnight at 4°C (listed in Table 1). After several washes with TBST, the membrane was incubated for 1 h at RT with the appropriate secondary antibodies conjugated to horseradish peroxidase (listed in Table 1). The detection was performed using an enhanced chemiluminescent substrate, ECL Plus (GE Healthcare, Little Chalfont, UK) and the ImageQuant LAS 4000 mini camera. Different membrane exposure times were used according to the type of sample: 2D differentiated podocyte extracts (3 min) or Mf extracts (6 min). Cell density was reduced in Mfs with a huge amount of collagen I matrix. Quantitative polymerase chain reaction (PCR) and PCR Total RNA was extracted using the NucleoSpin RNA plus purification kit (Macherey-Nagel, Düren, Germany) following the manufacturer’s instructions. Complementary DNA was synthesized from 1 μg of total RNA for controls and 300 ng for fibres using MaximaRT (Thermo Scientific, Villebon sur Yvette, France) according to the manufacturer’s instructions. PCR primers are listed in Table 2. The cycling parameters for PCR included 40 cycles of denaturation at 95°C for 15 s, annealing at 60–63°C for 30 s and elongation at 70°C for 5 s. The quantitative PCR was performed in triplicate on a CFX Connect Real Time PCR Detection System (Bio-Rad, Marnes-la-Coquette, France) using Invitrogen Ambion Glycoblue Coprecipitant (Invitrogen, Villebon sur Yvette, France). Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as a housekeeping gene standard. Table 2 PCR sequences Type . . Sequences (5′ → 3′) . SYNPO fwd CACGCTCACCACACCAACTT rv TGCTGCATCTCACCTCCTCA COL4A5 fwd AAAGGAGAGCCTGGTGGAAT rv CCGGCTGGGTTATAGTCTGA NID1 fwd ACATTGAGCCCTACACGGAGCT rv GCCACTGGTAAGTGTAGATGCG LAMA1 fwd GAAGGTGACTGGCTCAGCAAGT rv AGGCGTCACAACGGAAATCGTG VWF fwd TTGACGGGGAGGTGAATGTG rv ATGTCTGCTTCAGGACCACG CD31 fwd AACAGAAACCCGTGGAGATG rv GTCTCTGTGGCTCTCGTTCC GAPDH fwd GAGAAGGCTGGGGCTCATTTG rv CCACGATACCAAAGTTGTCAT Type . . Sequences (5′ → 3′) . SYNPO fwd CACGCTCACCACACCAACTT rv TGCTGCATCTCACCTCCTCA COL4A5 fwd AAAGGAGAGCCTGGTGGAAT rv CCGGCTGGGTTATAGTCTGA NID1 fwd ACATTGAGCCCTACACGGAGCT rv GCCACTGGTAAGTGTAGATGCG LAMA1 fwd GAAGGTGACTGGCTCAGCAAGT rv AGGCGTCACAACGGAAATCGTG VWF fwd TTGACGGGGAGGTGAATGTG rv ATGTCTGCTTCAGGACCACG CD31 fwd AACAGAAACCCGTGGAGATG rv GTCTCTGTGGCTCTCGTTCC GAPDH fwd GAGAAGGCTGGGGCTCATTTG rv CCACGATACCAAAGTTGTCAT fwd: forward; rv: reverse. Open in new tab Table 2 PCR sequences Type . . Sequences (5′ → 3′) . SYNPO fwd CACGCTCACCACACCAACTT rv TGCTGCATCTCACCTCCTCA COL4A5 fwd AAAGGAGAGCCTGGTGGAAT rv CCGGCTGGGTTATAGTCTGA NID1 fwd ACATTGAGCCCTACACGGAGCT rv GCCACTGGTAAGTGTAGATGCG LAMA1 fwd GAAGGTGACTGGCTCAGCAAGT rv AGGCGTCACAACGGAAATCGTG VWF fwd TTGACGGGGAGGTGAATGTG rv ATGTCTGCTTCAGGACCACG CD31 fwd AACAGAAACCCGTGGAGATG rv GTCTCTGTGGCTCTCGTTCC GAPDH fwd GAGAAGGCTGGGGCTCATTTG rv CCACGATACCAAAGTTGTCAT Type . . Sequences (5′ → 3′) . SYNPO fwd CACGCTCACCACACCAACTT rv TGCTGCATCTCACCTCCTCA COL4A5 fwd AAAGGAGAGCCTGGTGGAAT rv CCGGCTGGGTTATAGTCTGA NID1 fwd ACATTGAGCCCTACACGGAGCT rv GCCACTGGTAAGTGTAGATGCG LAMA1 fwd GAAGGTGACTGGCTCAGCAAGT rv AGGCGTCACAACGGAAATCGTG VWF fwd TTGACGGGGAGGTGAATGTG rv ATGTCTGCTTCAGGACCACG CD31 fwd AACAGAAACCCGTGGAGATG rv GTCTCTGTGGCTCTCGTTCC GAPDH fwd GAGAAGGCTGGGGCTCATTTG rv CCACGATACCAAAGTTGTCAT fwd: forward; rv: reverse. Open in new tab PCR products were migrated on agarose gel 1.5% with GelRed Nucleic Acid Gel Stain (Biotium, Cabestany, France). The fluorescence analysis was performed with Gene Genius BioImaging System (Syngene, Illkirch, France). Transmission electron microscopy Mfs were fixed with 2.5% glutaraldehyde in phosphate buffer (PB) 0.1 M pH 7.4 for 2 h at RT. Mfs were rinsed in PB and post-fixed in 1% osmium tetroxide with 0.15% potassium ferrocyanide in PB 0.1 M pH 7.4 for 1 h in the dark at RT. After several washes in water, Mfs were dehydrated in graded ethanol and propylene oxide and then embedded in epoxy resin (Epon 812). Mfs were mounted on silicon moulds and stored 60°C for 48 h. Ultrathin sections (65 nm) were picked up on copper grids and then stained with UranyLess (Delta Microscopies, Toulouse, France) and lead citrate. Grids were examined with a transmission electron microscope (H7650, Hitachi, Tokyo, Japan) at 80 kV at the Bordeaux Imaging Center, Bordeaux, France. Statistical analysis Statistical analyses were performed using GraphPad Prism 6 software (GraphPad Software, La Jolla, CA, USA). Data were first tested for normality using a Kolmogorov–Smirnov test. Depending on the result, a parametric or non-parametric test was used to compute a P-value. RESULTS Construction of endothelial Mfs Endothelial Mfs were obtained through an adaptation of the Mf biofabrication method developed by our team [22]. Mf conformation was sustained throughout the whole process with homogeneous cell distribution and cell viability (Figure 1A). The length and diameter of the Mfs were 0.5–1.5 cm and 110–180 μm, respectively. FIGURE 1 Open in new tabDownload slide (A) Microscopic observation of a cellularized microfibre. (a) The cellularized microfibre (cell concentration 35.106/mL, collagen I concentration 5 mg/mL) was contained in the shell formed of successive layers of reticulated alginate. Optical microscopy, magnification ×2.5. (b) Eighteen hours after alginate shell removal, the Mf kept a similar conformation. Optical microscopy, magnification ×2.5. (c) Five days after alginate removal, the structure and conformation of the Mf remained unchanged. Optical microscopy, magnification ×2.5. (d) Endothelial cell viability was maintained at Day 5. Optical microscopy, magnification ×2.5. (B) Compaction index. (a) Method of compaction index calculation. Green line and red line were the initial diameter and final diameter, respectively, of a fibre conserved for 18 h at 33°C. The compaction index was obtained by dividing the final diameter by the initial diameter. (b) Compaction analysis associated to cell concentration. The collagen concentration (5 mg/mL) was similar in all conditions. Data are representative of three independent experiments (*P < 0.05, unpaired t-test). (c) Compaction analysis associated to collagen concentration. The cell concentration (35.106 cells/mL) was similar in all conditions. Data are representative of three independent experiments (*P < 0.05, unpaired t-test). FIGURE 1 Open in new tabDownload slide (A) Microscopic observation of a cellularized microfibre. (a) The cellularized microfibre (cell concentration 35.106/mL, collagen I concentration 5 mg/mL) was contained in the shell formed of successive layers of reticulated alginate. Optical microscopy, magnification ×2.5. (b) Eighteen hours after alginate shell removal, the Mf kept a similar conformation. Optical microscopy, magnification ×2.5. (c) Five days after alginate removal, the structure and conformation of the Mf remained unchanged. Optical microscopy, magnification ×2.5. (d) Endothelial cell viability was maintained at Day 5. Optical microscopy, magnification ×2.5. (B) Compaction index. (a) Method of compaction index calculation. Green line and red line were the initial diameter and final diameter, respectively, of a fibre conserved for 18 h at 33°C. The compaction index was obtained by dividing the final diameter by the initial diameter. (b) Compaction analysis associated to cell concentration. The collagen concentration (5 mg/mL) was similar in all conditions. Data are representative of three independent experiments (*P < 0.05, unpaired t-test). (c) Compaction analysis associated to collagen concentration. The cell concentration (35.106 cells/mL) was similar in all conditions. Data are representative of three independent experiments (*P < 0.05, unpaired t-test). The appropriate cell number and collagen I concentration were determinated by the fibre compaction. Fibre compaction was a pivotal phenomenon during Mf fabrication and crucial for the next step, alginate removal. The formula was: Compaction index = final transversal diameter/initial transversal diameter The compaction index was measured at three different points along the Mf following the method illustrated in Figure 1B(a). We observed an optimal fibre compaction for a cell concentration of 35.106/mL and a collagen concentration of 5 mg/mL [Figure 1B(b) and 1B(c)]. Lower cell or collagen concentrations induced a strong fibre compaction (<0.7) with massive cell death. Low compaction was observed for high cell concentrations (50.106/mL), with difficulties removing alginate. The optimal compaction index was between 0.7 and 0.9. After 18 h at 33°C and optimal index compaction achieved, the reticulated alginate shell of Mfs was manually removed, layer by layer, under a magnifying glass. Alginate was discarded with an improvement in cell viability. Endothelial Mfs showed a specific organization with a collagen I core covered by a GEnC layer (Supplementary data, Endothelial cell orientation in 3D, and Supplementary data, Figures; Figure 1. Cellularized Mfs were then incubated at 37°C for 5 days in six-well plates. Construction of a glomerular barrier In order to obtain an Mf reproducing a glomerular barrier, Mfs cellularized with GEnC were seeded with podocytes at different concentrations and times. A solution containing podocytes was dropped onto the fibre and left for 5 min. In a second step, a GEnC medium was added to the well before incubation at 37°C for 14 days. Cell viability of Mfs covered by podocytes was confirmed at different time points after podocyte seeding (Supplementary data, Figures; Figure 2, left panel), even if some dead cells were observed. FIGURE 2 Open in new tabDownload slide (A) Microscopic observation of Mfs seeded with podocytes at different times. Mfs were seeded with podocytes before 4 days (upper panel) or after 5 days (lower panel) after alginate shell removal. Mfs were observed at Day 0 of podocyte seeding (S0) and 6 days later (S6). The enlarged box shows the magnification (×10). Podocytes were clearly distinguishable and located at the edge of the fibre. (B) Effect of the seeding time on Mf length. Bar charts represent fibres >5 days (left) and <4 days (right). The ratio between the initial and final length is approximately two times higher in fibres less than Day 4 compared with fibres more than Day 5. Bars show mean ± SD, n = 11 (P < 0.05) by unpaired t-test with Welch’s correction. FIGURE 2 Open in new tabDownload slide (A) Microscopic observation of Mfs seeded with podocytes at different times. Mfs were seeded with podocytes before 4 days (upper panel) or after 5 days (lower panel) after alginate shell removal. Mfs were observed at Day 0 of podocyte seeding (S0) and 6 days later (S6). The enlarged box shows the magnification (×10). Podocytes were clearly distinguishable and located at the edge of the fibre. (B) Effect of the seeding time on Mf length. Bar charts represent fibres >5 days (left) and <4 days (right). The ratio between the initial and final length is approximately two times higher in fibres less than Day 4 compared with fibres more than Day 5. Bars show mean ± SD, n = 11 (P < 0.05) by unpaired t-test with Welch’s correction. Cellularized Mfs were seeded with podocytes at various concentrations: 1.106, 5.105, 2.5.105, 105, 5.104 and 104 podocytes/mL (Supplementary data, Figures; Figure 2, right panel). An initial concentration of 5.105–106 podocytes/mL resulted in the formation of podocyte clusters around the Mf and subsequent cell death (Supplementary data, Figures; Figure 2A and B, right panel). Conversely, concentrations of 105 and 5.104 podocytes/mL showed optimal results. Podocytes were clearly adherent to the surface of the Mf without cluster formation (Supplementary data, Figures; Figure 2C and D, right panel). Finally, a concentration of 104 cells/mL resulted in the absence of podocytes attached to the Mf (Supplementary data, Figures; Figure 2E, right panel). Influence of podocyte seeding time on Mf contraction Several Mfs, ranging from Days 1 to 7 after alginate shell removal, were seeded with podocytes at a concentration of 5.104 cells/mL (Figure 2A). When Mfs were seeded with podocytes at Day 1 and 2, a drastic change in the morphology of the fibres was observed after 6 days of incubation at 37°C (Figure 2A, upper panel). Mfs adopted a spherical shape or showed a strong decrease in their initial length the second phenomenon was called contraction. Mfs seeded with podocytes at Day 7 kept their original morphology and length after 6 days of incubation at 37°C (Figure 2A, lower panel). Podocytes were clearly visible and located at the edge of the Mf (Figure 2A, enlarged box). To confirm these observations, a study of the contraction index in Mfs < 4 days was conducted and compared with Mfs >5 days. For each fibre, the contraction index was calculated as the ratio between the initial and final length (position under microscope was similar throughout the experiments). This contraction index was measured at Days 0 and 7 (Figure 2B). The length of long-standing Mfs (>5 days) did not vary after 8 days (ratio = 1.1 ± 0.1). Conversely, for early Mfs (<4 days), their initial length was divided by 2.0 ± 0.9 over the same period. A significant reduction in length was observed between conditions (P < 0.05, unpaired t-test). Expression of cell type–specific proteins in Mfs In differentiated Mfs, specific proteins were expressed at Day 14 with a specific pattern, depending on cell types: endothelial-specific proteins were located at the internal monolayer, whereas the expression of podocyte proteins was located at the periphery. The expression of specific endothelial markers in cellularized Mfs was detected by IF experiments for proteins platelet endothelial cell adhesion molecule-1 (PECAM-1; Figure 3A and E) and von Willebrand factor (vWF; Figure 3B). Expression of podocyte proteins nephrin (Figure 3C), synaptopodin (Figure 3D) and podocin (Figure 3E) was also prominent on cellularized Mfs. Protein expression was also observed by cross-section at a cellular scale (Supplementary data, Figures; Figure 3). As previously reported, podocytes formed a monolayer of cells around the collagen type I core and GEnC monolayer. FIGURE 3 Open in new tabDownload slide Expression of cell type–specific proteins by IF with confocal microscopy. (A) Immunostaining of PECAM-1 on an endothelial Mf. PECAM-1 is labelled in green and nuclear in blue. (B) Immunostaining of vWF. vWF is labelled in green and nuclear in blue. Arrows show the Weibel–Palade bodies. (C) Immunostaining of nephrin. Nephrin (red) and nuclei (blue) were localized at the periphery. Nephrin is expressed inside the membrane and the cytoplasm of some cells. (D) Immunostaining of synaptopodin. Synaptopodin (red) and nuclei (blue) were localized at the periphery. Synaptopodin is expressed inside the membrane and the cytoplasm of some cells. (E) Immunostaining of PECAM-1 and podocin. Podocin (red) and PECAM-1 (green) are expressed. Podocin expression is mostly situated at the edge of the Mf, whereas PECAM-1 is locally expressed within the internal layer. FIGURE 3 Open in new tabDownload slide Expression of cell type–specific proteins by IF with confocal microscopy. (A) Immunostaining of PECAM-1 on an endothelial Mf. PECAM-1 is labelled in green and nuclear in blue. (B) Immunostaining of vWF. vWF is labelled in green and nuclear in blue. Arrows show the Weibel–Palade bodies. (C) Immunostaining of nephrin. Nephrin (red) and nuclei (blue) were localized at the periphery. Nephrin is expressed inside the membrane and the cytoplasm of some cells. (D) Immunostaining of synaptopodin. Synaptopodin (red) and nuclei (blue) were localized at the periphery. Synaptopodin is expressed inside the membrane and the cytoplasm of some cells. (E) Immunostaining of PECAM-1 and podocin. Podocin (red) and PECAM-1 (green) are expressed. Podocin expression is mostly situated at the edge of the Mf, whereas PECAM-1 is locally expressed within the internal layer. The expression of nephrin, synaptopodin, CD31, vWF and vascular endothelial growth factor receptor 2 (VEGFR2) was confirmed by WB (Figure 4A and Supplementary data, Figures; Figure 4) and by PCR analysis on the total extract of Mfs after 14 days of differentiation (Figure 4B). FIGURE 4 Open in new tabDownload slide (A) Expression of specific proteins of each cell type by WB. Nephrin, synaptopodin and VEGFR2 expression were detected in Mf samples in the denaturating condition after 14 days of differentiation. We used 2D differentiated podocyte lysate (Pd) as a control for the expression of podocyte-specific proteins. The blot is representative of five independent experiments. With the same protein load, important variations of β-actin expression were observed in our 3D system compared with 2D cell culture because of a complex mixture of collagen core, podocytes and endothelial cells. MW: molecular weight (kDa). (B) Expression of specific proteins of each cell type by PCR. Synaptopodin, CD31 and vWF expression were detected in Mf samples and reported to GAPDH expression. The result is representative of two independent experiments. The control condition was co-culture of endothelial cells and podocytes in 2D. FIGURE 4 Open in new tabDownload slide (A) Expression of specific proteins of each cell type by WB. Nephrin, synaptopodin and VEGFR2 expression were detected in Mf samples in the denaturating condition after 14 days of differentiation. We used 2D differentiated podocyte lysate (Pd) as a control for the expression of podocyte-specific proteins. The blot is representative of five independent experiments. With the same protein load, important variations of β-actin expression were observed in our 3D system compared with 2D cell culture because of a complex mixture of collagen core, podocytes and endothelial cells. MW: molecular weight (kDa). (B) Expression of specific proteins of each cell type by PCR. Synaptopodin, CD31 and vWF expression were detected in Mf samples and reported to GAPDH expression. The result is representative of two independent experiments. The control condition was co-culture of endothelial cells and podocytes in 2D. GBM formation Neoformation of collagen IV GBM was observed at the cell interface by IF at 14 days (Figure 5A and Supplementary data, Figures; Figure 5A). This expression of collagen IV was detected around the cells between the outer and inner nuclei layers. Collagen IV was not detected in 2D GEnC culture [Supplementary data, Figures; Figure 5B(a)] or non-cellularized Mfs [Supplementary data, Figures; Figure 5B(b)]. A faint signal, mostly cytoplasmic or at the plasma membrane, was detected in 2D podocyte cell culture (Supplementary data, Figures; Figure 5B(c)]. Cell interaction and cooperation between GEnCs and podocytes induced collagen IV expression at the podocytes–GEnC interface. This condition is probably required to obtain a functional GBM. WB, PCR and quantitative PCR analysis confirmed collagen IV, nidogen and laminin secretion at the cell interface in the Mf at 14 days (Figure 5B and C;Supplementary data, Figure; Figures 4B and 5C). Glomerular basement membrane (GBM) protein expression increased during cell differentiation and GBM maturation (Figure 5B). 3D organization induced a strong expression of the constitutive protein of GBM compared with expression in the 2D model (Figure 5C). GBM was enriched in α5 collagen IV, as observed in a fully differentiated GBM in human kidney. FIGURE 5 Open in new tabDownload slide Collagen IV expression. (A) IF of collagen IV on a cellularized Mf. Nuclei (blue) and collagen IV (red). (B) Expression of GBM proteins during Mf maturation analysed by WB on Mf. Collagen IV and laminin expression were detected in denaturated conditions at 75–90 kDa and 200–220 kDa, respectively. Control conditions were co-culture lysate of endothelial cells and podocytes (Control +) and lysate of human mesenchymal cells (Control −). The blot was representative of two independent experiments. MW: molecular weight (kDa). (C) PCR and quantitative PCR of α5 collagen IV, laminin and nidogen on Mf. The result is representative of two independent experiments. The control condition (C) was co-culture of endothelial cells and podocytes in 2D compared with Mf (F) where cells were grown in 3D. FIGURE 5 Open in new tabDownload slide Collagen IV expression. (A) IF of collagen IV on a cellularized Mf. Nuclei (blue) and collagen IV (red). (B) Expression of GBM proteins during Mf maturation analysed by WB on Mf. Collagen IV and laminin expression were detected in denaturated conditions at 75–90 kDa and 200–220 kDa, respectively. Control conditions were co-culture lysate of endothelial cells and podocytes (Control +) and lysate of human mesenchymal cells (Control −). The blot was representative of two independent experiments. MW: molecular weight (kDa). (C) PCR and quantitative PCR of α5 collagen IV, laminin and nidogen on Mf. The result is representative of two independent experiments. The control condition (C) was co-culture of endothelial cells and podocytes in 2D compared with Mf (F) where cells were grown in 3D. Electron microscopy (EM) demonstrated de novo formation of a GBM (Figure 6). A few hours after podocyte seeding, the cell interface between podocytes and GEnC was clearly observed with some cytoplasmic extensions (Figure 6A and Supplementary data, Figures; Figure 6). At 14 days of culture, Mfs exhibited a 3D ultrastructure similar to a glomerular barrier with GEnC, GBM and podocytes from inside to outside. We observed foot processes, that is, podocyte extensions, at the podocyte interface, fenestrations on the endothelial layer and a newly formed matrix between the two cell types. In some places, a densification of the matrix was detected that may correspond to the formation of a lamina rara (Figure 6B). The width of pedicels or foot processes, endothelial fenestrations and neoformed GBM were evaluated at different points in the Mfs (Figure 6C). FIGURE 6 Open in new tabDownload slide Mf ultrastructure. (A) Mf ultrastructure several hours after podocyte seeding, observed by transmission EM. Glomerular endothelial cells (filled diamond) were interfaced with podocytes (asterisks) with cytoplasmic extensions or processes (left arrow). (B) Mf ultrastructure after 14 days of culture at 37°C, observed by transmission EM. The interface between the glomerular endothelial cells (filled diamond) and podocytes (asterisks) was conserved at 14 days of culture. Ultrastructural analysis exhibited some modifications since podocyte seeding. Podocyte extensions (P) and foot processes were observed. Between the two cell layers, neo-GBM was formed (left arrow). In some parts, a densification of the matrix (left arrow) was detected as lamina densa (GBM is formed by three layers, two lamina rara, which appear pale, and a lamina densa). (C) Width (nm) of pedicels, endothelial fenestrations and neo-GBM. Analysis was performed on transmission EM data at different levels of Mf. We observed a significant difference of pedicel width between J0 and J14 (unpaired t-test). FIGURE 6 Open in new tabDownload slide Mf ultrastructure. (A) Mf ultrastructure several hours after podocyte seeding, observed by transmission EM. Glomerular endothelial cells (filled diamond) were interfaced with podocytes (asterisks) with cytoplasmic extensions or processes (left arrow). (B) Mf ultrastructure after 14 days of culture at 37°C, observed by transmission EM. The interface between the glomerular endothelial cells (filled diamond) and podocytes (asterisks) was conserved at 14 days of culture. Ultrastructural analysis exhibited some modifications since podocyte seeding. Podocyte extensions (P) and foot processes were observed. Between the two cell layers, neo-GBM was formed (left arrow). In some parts, a densification of the matrix (left arrow) was detected as lamina densa (GBM is formed by three layers, two lamina rara, which appear pale, and a lamina densa). (C) Width (nm) of pedicels, endothelial fenestrations and neo-GBM. Analysis was performed on transmission EM data at different levels of Mf. We observed a significant difference of pedicel width between J0 and J14 (unpaired t-test). DISCUSSION In this study we reproduced the structure of a glomerular barrier by developing a glomerular Mf. In this 3D structure, human GEnC and podocytes synthetized de novo a GBM at the interface between these two cell types. This glomerular Mf was obtained through our innovative method without any synthetic matrix. In the literature, TE products have mostly assembled cell types in 2D using a synthetic matrix at the cell interface [18]. Here, we organized two different cell types in 3D with a specific pattern leading to improved podocyte and GEnC differentiation, specific matrix secretion and synthesis of a GBM. Initial characteristics of our constructs were related to the contraction/compaction phenomenon directly linked to cell and collagen concentrations. Optimal concentrations allowed us to obtain Mfs. Chrobak et al. [23] reported similar compaction and contraction phenomena. We also observed a straight compaction and contraction of the Mfs due to a cytotoxic phenomenon probably induced by the reticulated alginate shell. Removal of the alginate shell drastically changed cell survival in the Mf. Cell death may be linked to insufficient nutrient diffusion but also to other factors leading to apoptosis or necrosis (Supplementary data, Figures; Figure 7). Following alginate removal, podocytes were seeded onto the cellularized Mfs. We initially thought that an earlier seeding would improve cellular communication and interaction and then potentiate cell differentiation and GBM neoformation [24]. On the contrary, we observed a drastic compaction of the fibres that adopted a spherical shape. We hypothesized that interactions and communications between the two undifferentiated cell types induced a modification of cellular contractility and physical characteristics of the matrix, hence inducing contraction of the Mf. In ‘older’ Mfs, GEnCs were already differentiated with morphological and cell behaviour already established. Their characteristics were unchanged after undifferentiated podocyte seeding. Cell viability experiments conducted a considerable time after podocyte seeding revealed optimal viability. GEnC in cellularized Mfs expressed specific endothelial markers such as vWF, VEGFR2 and PECAM-1. Podocyte-specific proteins such as nephrin, synaptopodin and podocin were also expressed by podocytes at the external layer of the Mf. Hence the expression of these markers clearly demonstrated the optimal differentiation of cells on the Mf. Cell characteristics, that is, protein expressions, in 3D were sustained over time as described in the 2D cell culture [21]. Finally, organization of the cells in 3D with a direct interface induced a strong cell interaction and communication between the two cell types with collagen IV production. We observed a noticeable expression and secretion of collagen IV at the periphery of the Mf with a specific pattern: a thin layer between the outer and inner parts of the Mf. Specific expression of α5 collagen IV was detected by PCR in our glomerular Mfs. Interestingly, this expression was faint to undetectable in 2D co-cultures. Laminin and nidogen expression were also detected. Collagen IV is the main component of the GBM, along with laminin, nidogen and heparan sulphate proteoglycans [25]. Hence the secretion of these specific proteins provided evidence of the formation of a new GBM in our Mf system. Production of neo-GBM confirmed the differentiation of GEnC and podocytes. We demonstrated that the creation of GBM was directly linked to podocyte–endothelial cell communication or interaction. Crosstalk between the two cell types was previously reported as a crucial parameter to induce GBM component secretion [15, 24]. EM confirmed a 3D barrier organization with endothelial and podocyte layers separated by neo-GBM. We have observed some formation of podocyte foot processes at the external layer of the neo-GBM. The structure of the neo-GBM was not identical to the GBM of a normal kidney with three lamina layers. However, matrix densification, as in the lamina densa, was observed in some parts of the neo-GBM. Longer maturation and fluid shear stress exposure may be necessary to obtain a better organized neo-GBM structure. This is, to our knowledge, the first report of de novo synthesis of a GBM, since most previous publications that have reported de novo synthesis of basement membrane have used a synthetic matrix to support cell growth [16, 24]. CONCLUSION Our innovative model allowed us to produce a glomerular Mf with human glomerular cells. Crosstalking between the different cell types induced cell differentiation and formation of a GBM not observed in 2D cell culture. The ultrastructure of the Mf, with GEnC, a GBM and podocytes, reproduced many aspects of a human glomerular barrier ultrastructure. Our 3D system will be useful to study cell phenotypes and interactions in pathophysiological conditions. Our model will be complicated by the association of mesangial cells and the development of a perfused system. ACKNOWLEDGEMENTS The authors would like to thank Melina Petrel for his help. Moin Saleem and Gavin Welsh kindly gifted glomerular endothelial cells and podocytes. FUNDING This work was supported by the Agence de Biomédecine and the Société Francophone de Néphrologie, Dialyse et Transplantation. EM studies were performed at the Bordeaux Imaging Center, Bordeaux University, a Core facility of the national infrastructure ‘France BioImaging’ (ANR-10-INBS-04 France-BioImaging). AUTHORS’ CONTRIBUTIONS K.F., S.R., S.M. and P.B. performed the experiments and analysed and interpreted data. J.P., R.S., B.L. and J.K. performed the experiments. J.-C.F. contributed to the data analysis. C.C. contributed to the data analysis and wrote the article. R.D. reviewed the article. C.R. conceived the experiments, analysed and interpreted the data and wrote the article. All authors read and approved the final article. CONFLICT OF INTEREST STATEMENT None declared. The results presented in this article have not been published previously in whole or part, except in abstract format. REFERENCES 1 Perin L , Da Sacco S , De Filippo RE. Regenerative medicine of the kidney . Adv Drug Deliv Rev 2011 ; 63 : 379 – 387 Google Scholar Crossref Search ADS PubMed WorldCat 2 Groll J , Boland T , Blunk T et al. Biofabrication: reappraising the definition of an evolving field . Biofabrication 2016 ; 8 : 013001 Google Scholar Crossref Search ADS PubMed WorldCat 3 Yoo SS. 3D-printed biological organs: medical potential and patenting opportunity . Expert Opin Ther Pat 2015 ; 25 : 507 – 511 Google Scholar Crossref Search ADS PubMed WorldCat 4 Guimaraes-Souza NK , Yamaleyeva LM , AbouShwareb T et al. In vitro reconstitution of human kidney structures for renal cell therapy . Nephrol Dial Transplant 2012 ; 27 : 3082 – 3090 Google Scholar Crossref Search ADS PubMed WorldCat 5 Lanza RP , Chung HY , Yoo JJ et al. Generation of histocompatible tissues using nuclear transplantation . Nat Biotechnol 2002 ; 20 : 689 – 696 Google Scholar Crossref Search ADS PubMed WorldCat 6 Xinaris C , Benedetti V , Rizzo P et al. In vivo maturation of functional renal organoids formed from embryonic cell suspensions . J Am Soc Nephrol 2012 ; 23 : 1857 – 1868 Google Scholar Crossref Search ADS PubMed WorldCat 7 Song JJ , Guyette JP , Gilpin SE et al. Regeneration and experimental orthotopic transplantation of a bioengineered kidney . Nat Med 2013 ; 19 : 646 – 651 Google Scholar Crossref Search ADS PubMed WorldCat 8 Liu W , Xu S , Woda C et al. Effect of flow and stretch on the [Ca2+]i response of principal and intercalated cells in cortical collecting duct . Am J Physiol Renal Physiol 2003 ; 285 : F998 – F1012 Google Scholar Crossref Search ADS PubMed WorldCat 9 Friedrich C , Endlich N , Kriz W et al. Podocytes are sensitive to fluid shear stress in vitro . Am J Physiol Renal Physiol 2006 ; 291 : F856 – F865 Google Scholar Crossref Search ADS PubMed WorldCat 10 Eremina V , Baelde HJ , Quaggin SE. Role of the VEGF—a signaling pathway in the glomerulus: evidence for crosstalk between components of the glomerular filtration barrier . Nephron Physiol 2007 ; 106 : p32 – p37 Google Scholar Crossref Search ADS PubMed WorldCat 11 Petermann AT , Pippin J , Durvasula R et al. Mechanical stretch induces podocyte hypertrophy in vitro . Kidney Int 2005 ; 67 : 157 – 166 Google Scholar Crossref Search ADS PubMed WorldCat 12 Elger M , Sakai T , Kriz W. The vascular pole of the renal glomerulus of rat . Adv Anat Embryol Cell Biol 1998 ; 139 : 1 – 98 Google Scholar Crossref Search ADS PubMed WorldCat 13 Becquart P , Cambon-Binder A , Monfoulet LE et al. Ischemia is the prime but not the only cause of human multipotent stromal cell death in tissue-engineered constructs in vivo . Tissue Eng Part A 2012 ; 18 : 2084 – 2094 Google Scholar Crossref Search ADS PubMed WorldCat 14 Wang PC , Takezawa T. Reconstruction of renal glomerular tissue using collagen vitrigel scaffold . J Biosci Bioeng 2005 ; 99 : 529 – 540 Google Scholar Crossref Search ADS PubMed WorldCat 15 Slater SC , Beachley V , Hayes T et al. An in vitro model of the glomerular capillary wall using electrospun collagen nanofibres in a bioartificial composite basement membrane . PLoS One 2011 ; 6 : e20802 Google Scholar Crossref Search ADS PubMed WorldCat 16 Musah S , Mammoto A , Ferrante TC et al. Mature induced-pluripotent-stem-cell-derived human podocytes reconstitute kidney glomerular-capillary-wall function on a chip . Nat Biomed Eng 2017 ; 1 : 0069 Google Scholar Crossref Search ADS PubMed WorldCat 17 Zhou M , Zhang X , Wen X et al. Development of a functional glomerulus at the organ level on a chip to mimic hypertensive nephropathy . Sci Rep 2016 ; 6 : 31771 Google Scholar Crossref Search ADS PubMed WorldCat 18 Onoe H , Okitsu T , Itou A et al. Metre-long cell-laden microfibres exhibit tissue morphologies and functions . Nat Mater 2013 ; 12 : 584 – 590 Google Scholar Crossref Search ADS PubMed WorldCat 19 Saleem MA , O'Hare MJ , Reiser J et al. A conditionally immortalized human podocyte cell line demonstrating nephrin and podocin expression . J Am Soc Nephrol 2002 ; 13 : 630 – 638 Google Scholar PubMed OpenURL Placeholder Text WorldCat 20 Satchell SC , Tasman CH , Singh A et al. Conditionally immortalized human glomerular endothelial cells expressing fenestrations in response to VEGF . Kidney Int 2006 ; 69 : 1633 – 1640 Google Scholar Crossref Search ADS PubMed WorldCat 21 Rigothier C , Auguste P , Welsh GI et al. IQGAP1 interacts with components of the slit diaphragm complex in podocytes and is involved in podocyte migration and permeability in vitro . PLoS One 2012 ; 7 : e37695 Google Scholar Crossref Search ADS PubMed WorldCat 22 Kalisky J , Raso J , Rigothier C et al. An easy-to-use and versatile method for building cell-laden microfibres . Sci Rep 2016 ; 6 : 33328 Google Scholar Crossref Search ADS PubMed WorldCat 23 Chrobak KM , Potter DR , Tien J. Formation of perfused, functional microvascular tubes in vitro . Microvasc Res 2006 ; 71 : 185 – 196 Google Scholar Crossref Search ADS PubMed WorldCat 24 Byron A , Randles MJ , Humphries JD et al. Glomerular cell cross-talk influences composition and assembly of extracellular matrix . J Am Soc Nephrol 2014 ; 25 : 953 – 966 Google Scholar Crossref Search ADS PubMed WorldCat 25 Miner JH. Glomerular basement membrane composition and the filtration barrier . Pediatr Nephrol 2011 ; 26 : 1413 – 1417 Google Scholar Crossref Search ADS PubMed WorldCat Author notes Killian Flegeau and Sébastien Rubin contributed equally to this work. © The Author(s) 2019. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/open_access/funder_policies/chorus/standard_publication_model)
TI - Towards an in vitro model of the glomerular barrier unit with an innovative bioassembly method
JF - Nephrology Dialysis Transplantation
DO - 10.1093/ndt/gfz094
DA - 2020-02-01
UR - https://www.deepdyve.com/lp/oxford-university-press/towards-an-in-vitro-model-of-the-glomerular-barrier-unit-with-an-XNk3n5XJWf
SP - 240
VL - 35
IS - 2
DP - DeepDyve
ER -