Vascularized device for blood filtration and methods thereof

Abstract

Described herein are biomimetic microfluidic devices comprising electrospun silk fibroin membranes. In some embodiments, the biomimetic microfluidic device is a kidney glomerulus organ-on-chip device. Also described herein are methods of filtering a sample, methods of screening an agent for treatment of a kidney or glomerular disorder, and methods of generating fenestrated endothelial cells using the biomimetic microfluidic devices described herein.

Description

[0001]VASCULARIZED DEVICE FOR BLOOD FILTRATION AND METHODS THEREOF CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to U.S. Provisional Patent Application No.63 / 559,532 filed on February 29, 2024, which is incorporated by reference herein in its entirety. FEDERALLY SPONSORED RESEARCH This invention was made with government support under DP2DK139544 awarded by the National Institutes of Health. The government has certain rights in the invention. REFERENCE TO SEQUENCE LISTING This application was filed with a Sequence Listing XML in ST.26 XML format in accordance with 37 C.F.R. § 1.831 and PCT Rule 13ter. The Sequence Listing XML file submitted in the USPTO Patent Center, “028193-0050-WO01_sequence_listing_xml_25-FEB-2025.xml,” was created on February 252025, contains 36 sequences, has a file size of 36.0 kilobytes (36,864 bytes), and is incorporated by reference in its entirety into the specification. BACKGROUND Organ-on-chip (OOC) technology integrates techniques from materials science, tissue engineering, and cell and developmental biology to produce in vitro platforms that mimic the physiological environment of human organs. OOCs are compact—about the size of a USB flash drive—and contain fluidic microchannels that support cell culture and tissue formation. Dynamic fluid flow is used to apply shear stress to cells to stimulate mechanosensitive gene expression and regulate cell fate decisions, proliferation, maturation, and signaling. The orientation of the fluidic microchannels can be designed to match the in vivo tissue organization of the organ of interest, with neighboring channels forming tissue-tissue interfaces that mimic those in the physiological environment. For example, the blood-brain-barrier, lung alveoli, and the glomerular filtration barrier were modeled by stacking two microchannels separated by a porous membrane to facilitate cell adhesion. OOC systems have been applied to drug screening, disease modeling, and developmental biology research. Since a report in 2010, the field of OOC systems has grown rapidly as OOCs provide in vitro models that avoid interspecies differences that limit the utility of animal models for understanding human biological responses in disease research and therapeutic discovery. For example, OOCs can employ human stem cell differentiation techniques to generate human- specific organ models for preclinical studies, while over 90% of drug candidates that pass animal model-based preclinical studies fail in human clinical trials, OOCs developed with human cells have demonstrated improved prediction of human responses and increased success rates in drug development. Recognizing the groundbreaking advantages of OOCs, the U.S. Food and Drug Administration (FDA) approved OOCs for use as preclinical study models as part of the FDA Modernization Act 2.0 passed in 2022, eliminating the mandate for animal testing before in-human clinical trials. However, nonbiological materials in OOCs often limit their performance. For example, the membranes separating fluidic channels in OOCs are typically composed of polymers such as polydimethylsiloxane (PDMS) or polycarbonate (PC) which are employed due to their mechanical integrity under fluid flow and biocompatibility, but these polymers result in permanent nondegradable membranes, which are orders of magnitude thicker than basement membranes in human organs. PDMS and PC membranes used in OOCs have been ~30–50 m thick, while basement membranes in human organs are often less than 1 μm thick. Creating ultrathin PDMS membranes for OOCs requires specialized and cost-prohibitive equipment and techniques (e.g., reactive-ion etching with specialized gases), which presents a notable manufacturing barrier. Additionally, the inert nature of PDMS, PC, and other synthetic polymers results in a permanent barrier between cells and tissues that prevents functional remodeling of extracellular matrix (ECM). Such non-physiological membranes hinder transmembrane crosstalk between cells and the study of basement membranes in health and disease. Moreover, PDMS / PC membranes possess flat surfaces that lack the topographical features and microstructures found in native tissues that are important for cell fate determination and tissue function. An alternative membrane that enables in vivo-like transmembrane intercellular crosstalk and basement membrane remodeling by surrounding cells is needed to enhance the physiological relevance of OOCs. What is needed is an organ-on-chip device capable of recapitulating kidney glomerulus structure and morphology. SUMMARY One embodiment described herein is a biomimetic microfluidic device, comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane. In one aspect, the polymeric base material comprises polydimethylsiloxane (PDMS), polycarbonate (PC), poly-methyl-meta-acrylate (PMMA), cyclic olefin copolymer (COC), polyimide, or combinations thereof. In another aspect, the electrospun silk fibroin membrane further comprises polyethylene oxide (PEO). In another aspect, the biomimetic microfluidic device further comprises a pump fluidly connected to the first channel and the second channel and configured to pump fluids through the first channel and the second channel. In another aspect, the electrospun silk fibroin membrane further comprises a laminin coating on the first surface, the second surface, or a combination thereof. In another aspect, the electrospun silk fibroin membrane comprises a thickness of at least about 0.1 μm. In another aspect, the electrospun silk fibroin membrane comprises a thickness of about 0.1 μm to about 100 μm. In another aspect, the electrospun silk fibroin membrane comprises a thickness of about 0.5 μm to about 7.5 μm. In another aspect, the electrospun silk fibroin membrane is porous and comprises a porosity of about 1% to about 70%. In another aspect, the electrospun silk fibroin membrane is porous and comprises a porosity of about 40% to about 70%. In another aspect, the electrospun silk fibroin membrane comprises a Young’s modulus of about 25 kPa to about 100 kPa. In another aspect, the population of cells is cultured on the first surface of the electrospun silk fibroin membrane, the second surface of the electrospun silk fibroin membrane, or a combination thereof. In another aspect, the population of cells comprises induced pluripotent stem cells (iPSCs), embryonic stem cells (ESCs), kidney glomerular cells, podocytes, intermediate mesoderm (IM) cells, mesoderm cells, endothelial cells, embryoid bodies, organoids, spheroids, or combinations thereof. In another aspect, podocytes are cultured on the first surface of the electrospun silk fibroin membrane and endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane. In another aspect, the endothelial cells comprise fenestrated endothelial cells. In another aspect, the electrospun silk fibroin membrane mimics a kidney glomerular basement membrane. In another aspect, the first channel mimics a urinary compartment of a kidney glomerulus and the second channel mimics a vascular compartment of a kidney glomerulus. In another aspect, the biomimetic microfluidic device is a kidney glomerulus organ-on-chip device. In another aspect, the biomimetic microfluidic device is part of an extracorporeal dialysis system. Another embodiment described herein is a method of filtering a sample, the method comprising: applying a sample to a biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; wherein podocytes are cultured on the first surface of the electrospun silk fibroin membrane and endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane; and perfusing one or more fluids through the biomimetic microfluidic device. In one aspect, the method further comprises collecting a filtered fraction of the sample. In another aspect, the podocytes are induced pluripotent stem cell (iPSC)-derived podocytes. In another aspect, the sample is a blood sample, a urine sample, or a combination thereof. In another aspect, the method filters one or more uremic toxins from the sample. In another aspect, the method filters one or more biological metabolites from the sample. In another aspect, the one or more biological metabolites comprise inulin, albumin, urea, -2-microglobumin (B2M), visfatin, creatinine, or combinations thereof. In another aspect, the method has an albumin clearance rate of less than about 6%. In another aspect, the method has a urea reduction ratio (URR) of at least about 75%. Another embodiment described herein is a method of screening an agent for treatment of a kidney or glomerular disorder, the method comprising: applying an agent to a biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; wherein podocytes are cultured on the first surface of the electrospun silk fibroin membrane and endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane, and wherein the agent contacts the podocytes; perfusing one or more fluids through the biomimetic microfluidic device; and detecting a response of the podocytes to the agent, the response being associated with a kidney or glomerular disorder. In one aspect, the podocytes are induced pluripotent stem cell (iPSC)-derived podocytes. In another aspect, the response associated with the kidney or glomerular disorder comprises altered podocyte gene expression, altered podocyte protein expression, altered podocyte albumin uptake, altered formation or extension of podocyte foot processes, or combinations thereof. In another aspect, the agent comprises a protein, a peptide, a nucleic acid, a small molecule, or combinations thereof. In another aspect, the kidney or glomerular disorder comprises podocyte injury, proteinuria, glomerulosclerosis, diabetic nephropathy, chemotherapy-related nephrotoxicity, a podocytopathy, or combinations thereof. Another embodiment described herein is a method of generating fenestrated endothelial cells, the method comprising: seeding a population of podocytes or intermediate mesoderm (IM) cells and a population of endothelial cells in a biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; wherein the podocytes or IM cells are cultured on the first surface of the electrospun silk fibroin membrane and the endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane; and perfusing one or more fluids through the biomimetic microfluidic device to generate the fenestrated endothelial cells. Another embodiment described herein is a kit for filtering a sample, comprising: a biomimetic microfluidic device, comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; optionally, pumps, tubing, syringes, and containers for applying fluids to the biomimetic microfluidic device; optionally, cell culture buffers and reagents; and optionally, packaging, labels, or instructions for use. Another embodiment described herein is the use of a biomimetic microfluidic device for filtering a sample, the biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane. Another embodiment described herein is the use of a biomimetic microfluidic device for screening an agent for treatment of a kidney or glomerular disorder, the biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane. DESCRIPTION OF THE DRAWINGS The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee. FIG. 1A–F show the glomerulus biomimetic microfluidic (bioMF) device engineering and characterization. FIG.1A shows a schematic showing SF membrane synthesis and bioMF OOC device manufacturing process. The order of cell seeding, and differentiation of hiPS cell-derived podocytes and vascular endothelial cells are illustrated. SF, silk fibroin; PDMS, polydimethylsiloxane; ECs, endothelial cells; IM, intermediate mesoderm. FIG. 1B shows a representative photograph of the chips (left) and a schematic demonstrating the cross-section of the chip (right). Scale bar 1.5 mm. FIG.1C shows electron micrographs showing cross-sectional views of the bioMF OOC device with ultrathin SF membrane (left) and a conventional chip with a substantially thicker PDMS membrane (right). Scale bars 250 μm. FIG. 1D shows electron micrograph quantification of membrane thickness in the bioMF OOC device compared to conventional PDMS chips, and the estimated in vivo glomerular basement membrane thickness, N = 2 samples for conventional PDMS chips, N = 3 samples for bioMF OOC devises, n = 25 measurements for each sample. Purple squares (small): individual chip data points from the first replicate; purple squares (large): mean of all chip data points from the first replicate; orange circles (small): individual chip data points from the second replicate; orange circles (large): mean of all chip data points from the second replicate; cyan triangles (small): individual chip data points from the third replicate; cyan triangle (large): mean of all chip data points from the third replicate. FIG. 1E shows a zoomed-in view of SF membrane cross-section. Scale bar 5 μm. FIG. 1F shows electron micrographs showing the morphology, fiber network, and porosity of the SF membrane and a PDMS membrane. Scale bars 250 μm (the first and the third panel), 5 μm (the second and fourth panels). FIG. 2A–H show enhanced cell differentiation and glomerular tissue formation in the bioMF OOC device. FIG. 2A shows confocal microscopy of hiPS cell-derived podocytes differentiated in the device and immunostained for cell lineage identification markers nephrin (cyan) and podocin (yellow), and counterstained with DAPI (blue). Red arrows indicate strong expression of nephrin at the cell boundaries, white arrows indicate filamentous pattern of podocin in podocytes. Scale bars: 100 μm (top row), 50 μm (middle row), 25 μm (bottom row). FIG.2B shows confocal microscopy of the hiPS cell-derived endothelial cells cultured in the device and immunostained for the endothelial marker VE-cadherin (magenta) and counterstained with DAPI (blue). Scale bar, 100 μm. FIG.2C shows 3D reconstructed images of the epithelial (podocyte) and vascular (endothelial) cell layers separated by the electrospun SF membrane in the device. Cells were immunostained for nephrin (green) and VE-cadherin (red). Scale bar, 25 μm. FIG. 2D shows characterization of cell-secreted basement membrane proteins in the chips. Cultured glomerulus bioMF devices were immunostained for the ECM proteins collagen IV (first column, green), nidogen-1 (second column, red), and agrin (third column, yellow). Scale bar, 100 μm. FIG. 2E shows urea, inulin, and albumin filtration function of the cellularized devices with podocytes and endothelial cells; N = 3 independent replicates for inulin and albumin, N = 2 independent replicates for urea. FIG.2F shows estimated inulin and albumin filtration rate in vivo. FIG. 2G shows inulin and albumin filtration function of acellular devices; N = 3 independent replicates. Cyan triangles (small): chip data points from the first replicate; cyan triangle (large): mean of all first replicate chip data points; orange circles (small): chip data points from the second replicate; orange circles (large): mean of all second replicate chip data points; purple squares (small): chip data points from the third replicate; purple squares (large): mean of all third replicate chip data points. Statistical analysis was performed using unpaired Student’s t test. FIG.3A–E show modeling drug-induced nephrotoxicity in the glomerulus-on-a-chip. FIG. 3A shows fluorescent images of podocytes in the devices with and without Adriamycin treatment, and immunostained for nephrin (cyan) and podocin (yellow), and counterstained with DAPI (blue). Red arrows indicate strong expression of nephrin on the cell boundaries, white arrows indicate filamentous pattern of podocin in podocytes. Scale bars: 100 μm (top two rows), 25 μm (bottom two rows). FIG. 3B shows digital zoom-in of the podocyte fluorescent images from FIG. 3A, immunostained for nephrin (cyan) and podocin (yellow), and counterstained with DAPI (blue). Red arrows indicate strong expression of nephrin at the cell boundaries, white arrows indicate filamentous pattern of podocin in podocytes. Scale bar 12.5 μm. FIG. 3C shows fluorescent images of the endothelial cells in the devices with and without Adriamycin treatment, immunostained for VE-cadherin (magenta) and counterstained with DAPI (blue). Scale bar, 100 μm. FIG. 3D shows albumin filtration function of non-treated (vehicle control) and Adriamycin (ADR)-treated OOC devices; N = 3 independent replicates. FIG.3E shows inulin filtration function of vehicle control and ADR-treated devices; N = 3 independent replicates. Cyan triangles (small): individual chip data points from the first replicate; cyan triangle (large): mean of all chip data points from the first replicate; orange circles (small): individual chip data points from the second replicate; orange circles (large): mean of all chip data points from the second replicate; purple squares (small): individual chip data points from the third replicate; purple squares (large): mean of all chip data points from the third replicate. Statistical analysis was performed using unpaired Student’s t test. FIG. 4A–H show high resolution electron microscopy analysis of tissue-specific phenotypes in the engineered glomerulus bioMF device. FIG.4A shows a schematic highlighting the cross-sectional view of the engineered tissue model and structure. FIG. 4B shows a transmission electron micrograph showing an actual cross-sectional view of the engineered glomerular filtration barrier in the chips. Scale bar, 2 μm. FIG.4C shows transmission electron microscope images of podocytes and endothelial cells in the chips. Arrows indicate podocyte foot processes and formation of interdigitation-like organization, formation of secondary and tertiary foot processes, and formation of short protrusions around the SF nanofibers. Scale bars: 2 μm (left and middle), 1 μm (right). EC, endothelial cells; SF, SF membrane; POD, podocytes; FP, foot processes. FIG. 4D shows TEM images of endothelial cells phenotype in the chips when interfaced with podocytes. Arrows indicate fenestration-like structures. Scale bars: 1000 nm (left), 500 nm (middle), 1 μm (right). FIG.4E shows TEM images of endothelial cells in the chips when podocytes were absent. Scale bars: 2 μm (left and middle panels), and 1000 nm (right panel). FIG.4F shows quantification of endothelial cell fenestration numbers in regions of interest from TEM images in coculture / monoculture chips. N = 10 field of views on different cells. FIG. 4G shows pseudo-colored electron micrograph of podocytes (green) and fenestrated endothelial cells (purple) in the chips showing each cell layer on opposing sides of the SF membrane (blue), demonstrating the engineered model of the glomerular filtration barrier. FIG.4H shows pseudo- colored electron micrograph of monolayer-cultured endothelial cells (purple) in the chips on the SF membrane (blue), underscoring the lack of fenestrae when podocyte interactions are absent. Statistical analysis was performed using unpaired Student’s t test. FIG. 5A–C show uncovering intercellular crosstalk in the engineered glomerulus-on-a- chip. FIG. 5A shows a schematic showing cell culture conditions in the chips. Left panel: podocytes and endothelial cells are cocultured in the chip; middle panel: only podocytes are cultured in the chip; right panel: only endothelial cells are cultured in the chip. FIG. 5B shows confocal microscopy images evaluating VEGF-A expression in the podocyte and endothelial cell layers under coculture and monoculture conditions in the devices. Scale bar, 100 μm. FIG.5C shows ELISA characterization of secreted VEGF-A concentration in the urinary and vascular channels in the cocultured chips; N = 3 independent samples. Statistical analysis was performed using unpaired Student’s t test. FIG. 6A–B show scanning electron micrographs of SF membrane and porous PDMS membrane with and without laminin functionalization. FIG. 6A shows bare / unfunctionalized SF membrane (left) and SF membrane coated / functionalized with laminin-511 (right). Red arrows indicate topographical features of the electrospun SF nanofibers appearing after laminin-511 coating; scale bars 2.5 m (top row) and 0.5 m (bottom row). FIG. 6B shows bare / unfunctionalized porous PDMS membrane (left) and porous PDMS membrane coated / functionalized with laminin-511 (right). Blue arrows indicate ridges on bare PDMS membranes; scale bars, 25 m (top row), 2.5 m (middle row), and 1.5 m (bottom row). FIG.7A–B show material property characterization of electrospun SF membranes. FIG. 7A shows Young’s modulus of SF membranes measured by uniaxial tensile testing compared to human kidneys, N = 3 independent experiments and n = 10 total number of data points for the electrospun SF membranes. FIG. 7B shows Fourier-Transform Infrared Spectroscopy characterization of electrospun silk fibroin membranes from various batches across independent experiments showing the same characteristic absorbance peaks, highlighting the reproducibility of the material synthesis method. FIG.8A–B show hiPS cell-derived endothelial cell adhesion in the vascular channel on SF membrane on PDMS sides. FIG.8A shows a schematic illustration of endothelial cell adhesion in the vascular channel. FIG. 8B shows DAPI staining of hiPS cell-derived endothelial cells demonstrating adhesion on all interior sides of the chip vascular channel, including the SF membrane side (left), PDMS wall (channel wall) side (middle), and PDMS bottom side (right), scale bar 150 m. FIG. 9A–B show characterization of podocyte and endothelial cell markers in the bioMF OOC devices. FIG.9A shows 3D-reconstruction image of the bioMF OOC device demonstrating the podocyte-endothelial cell interface on the SF membrane, scale bar 25 m. FIG.9B shows western blot results of cells collected from the bioMF OOC devices showing bands for synaptopodin and VE-Cadherin. FIG. 10 shows 3D-reconstructed confocal images of glomerular basement membrane proteins / proteoglycans produced by the cells (podocyte layer) differentiated in the bioMF OOC device. Scale bar, 100 m. FIG.11A–B show measurement of the porosity of SF membranes using TEM micrographs. FIG.11A shows percentage area of nanofibers versus open pores in the SF membranes. N = 2 independent experiments and a total of 10 field of view for the measurement (n = 10), all data were presented as min to max with all data points shown. FIG. 11B shows an illustration of sections selected for quantification of SF membrane fiber density and porosity; the box indicates the SF membrane area (within the engineered tissue) used for quantification. FIG. 12A–B show quantification of transmission electron micrographs for measuring endothelial cell fenestration. FIG.12A shows endothelial cell body void spaces (fenestration-like structures) after co-culture by interfacing with differentiated podocytes or monoculture (without podocytes). N = 2 independent experiments and a total of 11 fields of view for the measurement for co-culture measurement (n = 11), and a total of 16 field of view for monoculture measurement (n = 16). FIG.12B shows the number of endothelial fenestrations normalized by pixel size after co-culture with or without differentiated podocytes. N = 2 independent experiments and a total of 10 fields of view for the measurement for both co-culture and monoculture measurement (n = 10). Statistical analysis was performed using unpaired Student’s t test, all data were presented as min to max with all data points shown. FIG. 13 shows quantification of VEGF-A immunofluorescence intensity expressed by podocytes and endothelial cells in the bioMF OOC device under coculture vs. monoculture conditions. 100 measurements were performed for each condition, including 25 measurements per field of view, and 4 fields of view were measured from an independent experiment (yielding n = 100). Statistical analysis was performed using unpaired Student’s t test. FIG. 14 shows bright-field images of endothelial cells and intermediate mesoderm cells showing examples of suboptimal adhesion in the silk fibroin membrane-based organ-on-a-chip (SF-OOC) devices. The left image shows endothelial cells, where the black arrows denote dark strips of detaching cells along the channel walls. The right image shows intermediate mesoderm cells, where the white arrow denotes cell clumps from detached intermediate mesoderm cells. Scale bar, 100 μm. FIG.15 shows nanofiber diameter distribution of SF membranes with and without laminin- 511 functionalization. FIG. 16 shows FTIR spectra of SF membranes immediately after electrospinning, methanol treatment, PEO removal, and laminin-511 functionalization. Vertical lines: shift of peaks for Amide-I, -II, and -III, and diminishment of peak for C-O-C bond. FIG.17 shows a brightfield image of the SF membrane in the device. Scale bar, 250 m. FIG.18 shows a photograph of an exemplary final setup of the SF-OOC device perfusion system. FIG.19 shows application of the SF-OOC devices in modeling human kidney glomerulus filtration function of urea, inulin, B2M, visfatin, and albumin. N = 2 independent replicates for urea, B2M, and visfatin. N = 3 independent replicates for inulin and albumin. FIG. 20 shows a top PDMS mold design of a 3D-printed PDMS soft-lithography mold in Fusion 360. Unit of measurements: mm. FIG.21 shows a bottom PDMS mold design of a 3D-printed PDMS soft-lithography mold in Fusion 360. Unit of measurements: mm. FIG.22 shows a design of a 3D-printed chip holder in Fusion 360. Unit of measurements: mm. FIG.23A–G. FIG. 23A show the family pedigree of the twin Northern European kindred (DUK40130) with biopsy-proven Focal Segmental Glomerulosclerosis (FSGS). FIG.23B shows the urinary protein creatine ratio (UPCR) in Duke 5 and Duke 7. FIG.23C shows the sequence of the TRPC6 C terminus. Both patients had a novel compound, C-terminal TRPC6 mutation (p.L899H and p.P924T). FIG.23D shows the computationally generated superimposed images of the WT reference and computationally predicted mutant single-chain TRPC6 proteins with a focus on the C terminal region of the protein. FIG.23E shows the pore architecture of the WT and the mutant TRPC6 protein (WT TRPC6 diameter = 9.71 Å, predicted WT TRPC6 diameter = 5.67 Å, and TRPC6899H,924Tdiameter = 5.85 Å). FIG.23F shows a hematoxylin and eosin staining of Duke 5’s kidney biopsy with focal glomerular sclerosis and hyalinosis. FIG. 23G shows an electron micrograph of Duke 5’s biopsy with dilated ER with electron-dense ER congestion. FIG.24A–G. FIG.24A shows the immunofluorescence of patient-specific human induced pluripotent stem (iPS) cell-derived podocytes’ cell bodies labeled with nephrin, podocin, and TRPC6. FIG.24B shows the immunofluorescence of patient-specific iPS cell-derived podocytes’ foot processes labeled for nephrin, podocin, and TRPC6 expression. FIG. 24C shows the distribution of nephrin, podocin, and TRPC6 in the podocytes' cell bodies and foot processes. FIG. 24D shows healthy patient podocytes immunostained for nephrin, calnexin, and ionositol 1,4,5 triphosphate receptors. FIG.24E shows the quantification of the ER aspect ratio based on calnexin expression in podocytes. FIG. 24F shows the immunofluorescence of treated and untreated patient-specific podocytes immunostained for nephrin, podocin, and TRPC6. FIG.24G shows the Mander’s overlap coefficient for nephrin and podocin in the podocytes' cell bodies and foot processes before and after treatment with sildenafil and losartan. FIG. 24E and 24G represent three biologically independent experiments, and the data is defined as ± standard error of the mean. Scale bars: FIG.24A, 20 μm; FIG.24B, 20 μm; FIG.24C, column 1 images 15 μm, column 2 and 3 images 5 μm; FIG.24D column 1, 4 μm; column 2, 2 μm; FIG.24F, 5 μm. FIG.25A–G. FIG.25A shows the schematic representation of an engineered glomerular capillary wall-on-a-chip with patient-specific iPS cell-derived podocytes and endothelial cells. The protein aggresomes are shown in the mutant organ chip. Podocyte lineage-specific marker expression in the glomerular capillary wall-on-a-chip was measured for FIG.25B, podocin; FIG. 25C, nephrin; FIG. 25D, TRPC6; FIG. 25E, CD2AP. FIG. 25F shows the urinary clearance % quantified from the urinary channel after perfusing the chips with fluorescently labeled albumin for 6 hours. FIG. 25G shows protein aggresome size in podocytes before and after sildenafil and losartan combinatorial therapy, as represented in FIG. 31B. FIG. 25B–G represent three biologically independent experiments, and the data is defined as ± standard error of the mean. Each data point represents signal intensity emanating from the immunostained chips / pixel2as represented in FIG.30 across an area of ~75 μm. FIG. 26A–E. FIG. 26A shows the pore diameter of the WT, WT predicted, and mutant TRPC6 channel core diameters. FIG.26B shows the immunofluorescence of protein aggresomes observed in patient-specific iPS cell-derived podocytes’ cell bodies immunolabeled for Nephrin, Podocin, and TRPC6. The bottom panel represents the zoomed-in view. FIG.26C shows the z- stacks of the podocytes showing Nephrin-Podocin complex formation in the cell body of the podocytes. The mutant podocytes demonstrated massive podocin-enriched aggresomes. FIG. 26D shows Mander’s overlap coefficient for Nephrin and Podocin in the cell body of the untreated podocytes. FIG. 26E shows Mander’s overlap coefficient for Nephrin and Podocin in the foot processes of the untreated podocytes. FIG. 26C–D represent data from three biologically independent experiments, and the data is defined as ± standard error of the mean. One-way ANOVA with Sidak’s multiple comparison post hoc analyses was performed with a family-wise alpha threshold of 0.05. Scale bars: FIG.26A, top row 20 μm; bottom row 5 μm; FIG.26B, column 1, 20 μm, column 2, 5 μm. FIG.27A–E. FIG.27A shows representative immunoblots of podocyte markers including Nephrin, Synaptopodin, TRPC6, Podocin, CD2AP, GLEPP1, and WT1, the housekeeping protein GAPDH. FIG. 27B shows the quantification of immunoblots for the podocyte markers. Data represents three biologically independent experiments and is defined as ± standard error of the mean. One-way ANOVA with Dunnett’s multiple comparison post hoc analyses was performed with a family-wise alpha threshold of 0.05. FIG.27C shows electron micrograph of differentiated podocytes with protein aggresomes in the ER of Duke 5 and Duke 7 podocytes. FIG.27D shows LAMP1 vesicles, Podocin, and TRPC6 expression in podocytes. The LAMP1 vesicle number remains unchanged in the podocytes. FIG.27E shows lysosomal biogenesis-associated RAB7 expression in the podocytes. RAB7 expression was reduced in the mutants. Data is representative of three biologically independent experiments and the data is defined as ± standard error of the mean. One-way ANOVA with Sidak’s multiple comparison post hoc analyses was performed with a family-wise alpha threshold of 0.05. Scale bar: FIG. 27C, top row 1000 μm, bottom row 500 nm; FIG.27D, top row 20 μm, bottom row 5 μm; FIG.27E, 100 μm. FIG. 28A–B show patient podocytes immunostained for Nephrin, Podocin, and TRPC6 expression after treatment with BI749327 (FIG. 28A) and Sparsentan (FIG. 28B). Scale bars within each panel: top row 20 μm, bottom row 5 μm. FIG. 29A–C show z stacks of podocytes on the glomerular capillary wall-on-a-chip immunostained for Podocin, TRPC6, Nephrin, and CD2AP in healthy WT (FIG.29A), Duke 5 (FIG. 29B), and Duke 7 (FIG. 29C) samples. The X-, Y-, and Z-axes have been highlighted in the figures. FIG.30A–B show immunofluorescence of podocytes on the glomerular capillary wall-on- a-chip immunostained for Podocin and TRPC6 (FIG. 30A) and Nephrin and CD2AP (FIG. 30B) for WT, Duke 5, and Duke 7 samples. Corresponding quantifications are presented in FIG.25B– E. FIG.31A–B. FIG.31A shows albumin sequestration by the podocytes in the glomerular capillary wall-on-a-chip. Albumin uptake was reduced in the mutant podocytes. One-way ANOVA with Dunnett’s multiple comparison post hoc analyses was performed with a family-wise alpha threshold of 0.05. FIG.31B shows an electron micrograph of untreated and treated podocytes differentiated in the glomerular capillary wall-on-a-chip system. FIG. 31A represents three biologically independent experiments, and the data is defined as ± standard error of the mean. Scale bars: FIG.31A, 20 μm, FIG.31B, 1000 nm. FIG.32A–G show validation of lentiviral-delivered shRNA-mediated knockdown of CTGF, Cyr61, and YAP. FIG. 32A shows a schematic representation of lentiviral preparation and experiment workflow. FIG. 32B shows RT-qPCR of shRNAs targeting CTGF (shCTGF), Cyr61(shCyr61), and YAP (shYAP). N = 3 biological replicates, unpaired t-test. FIG.32C shows ELISA quantification of the secreted levels of CTGF and Cyr61. N = 3 biological replicates. One- Way ANOVA with multiple comparisons to non-transduced (NT) negative control. FIG.32D shows western blot analysis of cell lysates collected from non-transduced (NT), vehicle control (shGFP) and shYAP-transduced cells. FIG. 32E shows densitometric quantification of FIG. 32D. N = 3 biological replicates, One-way ANOVA with multiple comparisons and post-hoc Dunnett’s test for significance between means. FIG. 32F shows immunostain of YAP (green, fire) in shGFP and shYAP podocytes (DAPI, white). FIG.32G shows quantification of YAP nuclear staining intensity N = 3 biological replicates, unpaired t-test. ns, not significant; *p<0.05; **p<0.01; *** p< 0.001; ****p<0.0001. FIG. 33A–G show CTGF, Cyr61, and YAP knockdown significantly alter podocyte morphology and susceptibility to drug-induced injury via Vinculin delocalization. FIG.33A shows counterstain for F-actin (grey) in control and knockdown podocytes (DAPI, cyan). FIG.33B shows quantification of cell area in the knockdowns compared to negative control (non-transduced, NT). N = 3 biological replicates, One-way ANOVA with multiple comparisons and post-hoc Dunnett’s test. FIG. 33C shows quantification of actin fibers / cell. N = 3 biological replicates, One-way ANOVA with multiple comparisons and post-hoc Dunnett’s test. FIG. 33D shows CCK-8 assay showing percent viability of shCyr61, shCTGF, shCTGF+shCyr61(double knockdown) and shYAP relative to control group (shGFP). Data are presented as mean and error bars as SEM. One- way ANOVA with multiple comparisons and post-hoc Dunnett’s test. For shGFP, shCTGF+shCyr61, and shYAP, N = 4 biological replicates; otherwise, N = 3 biological replicates. FIG.33E shows CCK-8 assay showing percent viability of shCTGF+shCyr61 and shYAP relative to control (shGFP) before and after ADR treatment. N = 3 biological replicates, Two-Way ANOVA with Tukey multiple comparisons test. Dotted line denotes basal level of viability after ADR treatment. FIG.33F shows the percent difference of the DMSO-normalized cell viability of ADR- treated podocytes with shCTGF+shCyr61 and shYAP compared to the cell viability of ADR-treated non-targeting control (shGFP) in FIG. 33E. N = 3 biological replicates, One-way ANOVA with multiple comparisons and post-hoc Dunnett’s. FIG. 33G shows immunostaining showing presence of vinculin-positive puncta in shGFP negative control that are largely absent in the shCTGF / shCyr61 double knockdown. Dotted white lines denote cell borders, yellow arrows indicate vinculin-positive puncta. Ns, not significant; *p<0.05; **p<0.01; *** p< 0.001; ****p<0.0001. FIG.34A–J show development and validation of lentiviral mediated TetOn-inducible YAP, CTGF, and Cyr61 overexpression. FIG. 34A shows a schematic representation of lentiviral preparation and experiment workflow to illustrate induction of YAP overexpression prior to ADR- injury. FIG.34B shows brightfield and fluorescence microscopy images of TetOn-MCS-eGFP, - YAP-2A-mcherry, -CTGF-2A-mcherry, and -Cyr61-2A-mcherry. FIG.34C shows RT-qPCR of YAP gene expression TetOn-YAP podocytes relative to control group. N = 4 biological replicates. Unpaired t-test. FIG.34D shows western blot of pYAP / pTAZ (Ser127 / Ser89) and Total YAP / TAZ in TetOn-YAP podocytes relative to control. FIG.34E shows densitometric quantification of YAP and pYAP in FIG. 34D. N = 3 independent replicates. Unpaired t-test. FIG. 34F shows immunostain of YAP (green) in NT and TetOn-YAP podocytes, counterstained to DAPI (blue). FIG. 34G shows RT-qPCR of CTGF and Cyr61 gene expression in TetOn-CTGF and -Cyr61 podocytes, respectively. N = 4 biological replicates. Unpaired t-test. FIG.34H shows RT-qPCR of CTGF and Cyr61 in TetOn-YAP podocytes relative to control group. FIG. 34I shows immunostain of CTGF (top panel, yellow) and Cyr61 (bottom panel, cyan) in control and TetOn- YAP podocytes (grey), counterstained with DAPI (magenta). FIG.34J shows ELISA quantification of the secreted levels of CTGF and Cyr61 in NT control group, and TetOn-YAP, -CTGF, and - Cyr61 podocytes. N = 3 biological replicates. *p<0.05; **p<0.01. FIG. 35A–G show the effects of CTGF / Cyr61 and YAP overexpression on podocyte resistance to ADR-induced injury. FIG.35A shows CCK-8 viability assay showing percent viability of human iPS cell-derived podocytes transduced with TetOn-CTGF, TetOn-Cyr61, TetOn-YAP, andnegative control (non-transduced, NT) following 24 hours treatment with 0.5 μg mL 1 ADR, relativeto untreated control. N = 3 biological replicates. A two-way ANOVA with multiple comparisons was used to determine significance. FIG. 35B shows immunostaining of YAP (green) and EdU incorporation (red) in non-transduced and TetOn-YAP podocytes compared to intermediate mesoderm (podocyte progenitor cell), counterstained to DAPI (blue). FIG. 35C shows flow cytometry analysis of the cell membrane integrity (7-AAD) in healthy and ADR-injured TetOn-YAP podocytes, gated with respect to untreated non-transduced (NT). FIG. 35D shows the percent difference of cells with compromised membranes in FIG.35C with respect to internal untreated groups. N = 3 biological replicates. A two-way ANOVA with multiple comparisons was used to determine significance. FIG.35E shows RT-qPCR of Nphs1 corresponding to the Nephrin gene in TetOn-YAP podocytes (with and without ADR) relative to untreated, non-transduced control. N = 4 biological replicates. Two-Way ANOVA with Tukey post-hoc test. FIG. 35F shows immunostain of Nephrin in NT and TetOn-YAP podocytes (with and without ADR). FIG. 35G shows quantification of mean Nephrin intensity in FIG.35F. N = 6 images per condition *p<0.05; **p<0.01; *** p< 0.001. FIG. 36A–N show that YAP overexpression affects genes and proteins of the podocyte contractile apparatus and enhances myosin activity to promote re-extension of primary processes. FIG. 36A shows a schematic representation of lentiviral generation and experiment design to illustrate induction of YAP overexpression following ADR-injury. FIG.36B shows western blot of pYAP / pTAZ (Ser127 / Ser89) and YAP / TAZ in NT and TetOn-YAP podocytes, with and without ADR. FIG.36C shows densitometric quantification of pYAP and YAP in FIG.36B. N = 3 biological replicates. Two-Way ANOVA with Tukey’s post hoc test. FIG.36D shows CCK-8 viability assay when YAP overexpression is induced following 24 hours ADR treatment relative to untreated, NT podocytes. N = 3 biological replicates. FIG.36E shows immunostain of YAP (green) in NT and TetOn-YAP podocytes, with and without ADR. FIG.36F shows immunostain of YAP (green) and Beta-Tubulin III (Tuj1, orange) in NT and TetOn-YAP podocytes, with and without ADR, counterstained with DAPI (blue). FIG.36G shows YAP vs. Beta-Tubulin III (Tuj1) intensity in NT and TetOn-YAP podocytes, with and without ADR. N = 2 independent replicates, n = 5 cells per replicate. FIG. 36H shows F-actin (grey, top panel; fire, bottom panel) immunostain in NT and TetOn-YAP podocytes, with and without ADR. FIG.36I shows mean actin intensity of FIG.36H. N = 6 images per condition. FIG. 36J shows RT-qPCR of actin-associating genes (SYNPO, MYL12A, MYO5A) and Nphs2, corresponding gene to podocin, in NT and TetOn-YAP podocytes, relative to negative control (untreated, NT). N = 4 biological replicates. Two-Way ANOVA with Tukey’s post-hoc test. FIG.36K shows western blot of Myosin Va and Synaptopodin in NT and TetOn-YAP podocytes, with and without ADR. FIG. 36L shows densitometric quantification of FIG.36K. Myosin Va, N = 4 biological replicates; Synaptopodin N = 3 biological replicates. Two- Way ANOVA with Tukey post-hoc test. FIG. 36M shows co-immunostain of synaptopodin (magenta) and Myosin Va (green) in NT and TetOn-YAP podocytes, with and without ADR, counterstained with DAPI (blue). FIG. 36N shows immunostain of Podocin (yellow) in NT and TetOn-YAP podocytes, with and without ADR, counterstained with DAPI (blue). *p<0.05; **p<0.01; *** p< 0.001; ****p<0.0001. FIG. 37A–G show translation and validation of TetOn-system to higher order model (Glomerulus Chip). FIG.37A shows a schematic representation of TetOn-system in chip timeline: Human iPSC differentiation, chip fabrication, lentiviral transduction, drug-treatment, and induction of YAP overexpression. FIG.37B shows an epifluorescent microscopy image of LV-TetOn-YAP transduction in Intermediate Mesoderm cells at 20MOI compared to non-transduced; Epifluorescent and fluorescent microscopy images of NT and TetOn-YAP podocytes from their respective IM cells. FIG.37C shows albumin urinary clearance (% total Albumin introduced into vascular channel) in NT and lentiviral-transduced glomerulus chips following 5 days of podocyte induction. N = 5 chips per condition. FIG. 37D shows immunostain of YAP (green) in NT and TetOn-YAP glomerular chips, with and without ADR. FIG. 37E shows immunostain of Nephrin (green) and Podocin (yellow) in NT and TetOn-YAP podocytes, with and without ADR. FIG.37F shows albumin urinary clearance (% total Albumin introduced into vascular channel) in NT and TetOn-YAP glomerular chips with ADR treatment, followed by YAP overexpression and 2-day recovery period in Culture Boost (Extended Culture). NT, N = 5 chips per condition; TetOn-YAP N = 7 chips per condition. Two-Way ANOVA with Tukey’s post hoc test. FIG. 37G shows the percent difference of albumin urinary clearance in extended culture compared to initial ADR albumin clearance in FIG.37F. Unpaired t-test. *p<0.05. FIG.38 shows plasmid maps for shRNA inhibition studies. Map of pLV-H1TetO-GFP-puro (10,281 bp, Biosettia, SORT-C01) vector backbone for cloning shRNA oligos, which places expression of shRNA under control of a doxycycline inducible promoter. FIG. 39A–B show that target gene knockdown did not affect overall expression of focal adhesion and actin-associated proteins. FIG.39A shows a representative western blot of vinculin, alpha-actinin 4, synaptopodin, and podocin with GAPDH as housekeeping protein in vehicle control as well as shCyr61, -CTGF, and double knockdown conditions. FIG. 39B shows densitometric quantification of FIG.39A. N = 3 biological replicates. One-Way ANOVA. FIG.40A–E show plasmid maps for TetOn-constructs. FIG.40A shows FUW-TetO-MCS (Addgene Plasmid #84008); Purpose: Doxycycline inducible lentiviral vector (empty backbone vector, serves as negative control). FIG.40B shows FUdeltaGW-rtTA (Addgene Plasmid #19780); Lentiviral vector expresses reverse tetracycline transactivator (rtTA) for use with Tet-ON constructs. FIG. 40C shows FUW-TetO-wtYAP (Addgene Plasmid #84009); Lentiviral vector expresses FLAG-tagged wild-type YAP under control of doxycycline inducible promoter. FIG.40D shows TetO-CTGF-2A-mCherry (10,319 bp); Lentiviral vector expresses wildtype CTGF and mCherry reporter under control of doxycycline inducible promoter. FIG.40E shows FUW-TetO- Cyr61-2A-mCherry (10,415 bp); Lentiviral vector expresses wildtype Cyr61 and mCherry reporter under control of doxycycline inducible promoter. FIG. 41 shows the mean intensity quantification in FIG. 34F where YAP significantly elevated compared to negative control. FIG.42A–B show that YAP expression in podocytes can regulate Nephrin in healthy and disease states. FIG.42A shows flow cytometric analysis of NT and TetOn-YAP podocytes, with and without ADR (detached and adherent), showcasing cell area vs cell permeability (7-AAD). FIG. 42B shows RT-qPCR analysis of Nphs1 in wild type and TetON-YAP podocytes (with and without ADR); N = 4 independent replicates. FIG. 43A–D show that actomyosin activity, a key player in podocyte development and homeostasis, can be enhanced by YAP overexpression following ADR injury. FIG.43A shows an all-condition compilation of FIG.36G. FIG.43B shows RT-qPCR analysis of upstream (ROCK2, RAC1) genes and representative western blot of downstream proteins (beta-tubulin III and vinculin) that were unaffected neither by YAP overexpression nor ADR exposure; N = 3 biological replicates. FIG.43C shows RT-qPCR analysis of MYO5A activity over podocyte differentiation; N = 3 biological replicates. FIG. 43D shows co-immunostain of Myosin V (green) and Nephrin (yellow) over podocyte differentiation, counterstained to DAPI (blue). DETAILED DESCRIPTION Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art. For example, any nomenclatures used in connection with, and techniques of biochemistry, molecular biology, immunology, microbiology, genetics, cell and tissue culture, and protein and nucleic acid chemistry described herein are well known and commonly used in the art. In case of conflict, the present disclosure, including definitions, will control. Exemplary methods and materials are described below, although methods and materials similar or equivalent to those described herein can be used in practice or testing of the embodiments and aspects described herein. As used herein, the terms “amino acid,” “nucleotide,” “polynucleotide,” “vector,” “polypeptide,” and “protein” have their common meanings as would be understood by a biochemist of ordinary skill in the art. Standard single letter nucleotides (A, C, G, T, U) and standard single letter amino acids (A, C, D, E, F, G, H, I, K, L, M, N, P, Q, R, S, T, V, W, or Y) are used herein. As used herein, terms such as “include,” “including,” “contain,” “containing,” “having,” and the like mean “comprising.” The present disclosure also contemplates other embodiments “comprising,” “consisting essentially of,” and “consisting of” the embodiments or elements presented herein, whether explicitly set forth or not. As used herein, “comprising,” is an “open- ended” term that does not exclude additional, unrecited elements or method steps. As used herein, “consisting essentially of” limits the scope of a claim to the specified materials or steps and those that do not materially affect the basic and novel characteristics of the claimed invention. As used herein, “consisting of” excludes any element, step, or ingredient not specified in the claim. As used herein, the term “a,” “an,” “the” and similar terms used in the context of the disclosure (especially in the context of the claims) are to be construed to cover both the singular and plural unless otherwise indicated herein or clearly contradicted by the context. In addition, “a,” “an,” or “the” means “one or more” unless otherwise specified. As used herein, the term “or” can be conjunctive or disjunctive. As used herein, the term “and / or” refers to both the conjunctive and disjunctive. As used herein, the term “substantially” means to a great or significant extent, but not completely. As used herein, the term “about” or “approximately” as applied to one or more values of interest, refers to a value that is similar to a stated reference value, or within an acceptable error range for the particular value as determined by one of ordinary skill in the art, which will depend in part on how the value is measured or determined, such as the limitations of the measurement system. In one aspect, the term “about” refers to any values, including both integers and fractional components that are within a variation of up to ± 10% of the value modified by the term “about.” Alternatively, “about” can mean within 3 or more standard deviations, per the practice in the art. Alternatively, such as with respect to biological systems or processes, the term “about” can mean within an order of magnitude, in some embodiments within 5-fold, and in some embodiments within 2-fold, of a value. As used herein, the symbol “~” means “about” or “approximately.” All ranges disclosed herein include both end points as discrete values as well as all integers and fractions specified within the range. For example, a range of 0.1–2.0 includes 0.1, 0.2, 0.3, 0.4. . . 2.0. If the end points are modified by the term “about,” the range specified is expanded by a variation of up to ±10% of any value within the range or within 3 or more standard deviations, including the end points, or as described above in the definition of “about.” As used herein, the terms “room temperature,” “RT,” or “ambient temperature” refer to the typical temperature in an indoor laboratory setting. In one aspect, the laboratory setting is climate controlled to maintain the temperature at a substantially uniform temperature or with a specific range of temperatures. In one aspect, “room temperature” refers a temperature of about 15–30 °C, including all integers and endpoints within the specified range. In another aspect, “room temperature” refers a temperature of about 15–30 °C; about 20–30 °C; about 22–30 °C; about 25–30 °C; about 27–30 °C; about 15–22 °C; about 15–25 °C; about 15–27 °C; about 20–22 °C; about 20–25 °C; about 20–27 °C; about 22–25 °C; about 22–27 °C; about 25–27 °C; about 15 °C ± 10%; about 20 °C ± 10%; about 22 °C ± 10%; about 25 °C ± 10%; about 27 °C ± 10%; ~20 °C, ~22 °C, ~25 °C, or ~27 °C, at standard atmospheric pressure. As used herein, the terms “active ingredient” or “active pharmaceutical ingredient” refer to a pharmaceutical agent, active ingredient, compound, or substance, compositions, or mixtures thereof, that provide a pharmacological, often beneficial, effect. As used herein, the terms “control,” or “reference” are used herein interchangeably. A “reference” or “control” level may be a predetermined value or range, which is employed as a baseline or benchmark against which to assess a measured result. “Control” also refers to control experiments or control cells. As used herein, the term “dose” denotes any form of an active ingredient formulation or composition, including cells, that contains an amount sufficient to initiate or produce a therapeutic effect with at least one or more administrations. “Formulation” and “composition” are used interchangeably herein. As used herein, the term “prophylaxis” refers to preventing or reducing the progression of a disorder, either to a statistically significant degree or to a degree detectable by a person of ordinary skill in the art. As used herein, the terms “effective amount” or “therapeutically effective amount,” refers to a substantially non-toxic, but sufficient amount of an action, agent, composition, or cell(s) being administered to a subject that will prevent, treat, or ameliorate to some extent one or more of the symptoms of the disease or condition being experienced or that the subject is susceptible to contracting. The result can be the reduction or alleviation of the signs, symptoms, or causes of a disease, or any other desired alteration of a biological system. An effective amount may be based on factors individual to each subject, including, but not limited to, the subject’s age, size, type or extent of disease, stage of the disease, route of administration, the type or extent of supplemental therapy used, ongoing disease process, and type of treatment desired. As used herein, the term “subject” refers to an animal. Typically, the subject is a mammal. A subject also refers to primates (e.g., humans, male or female; infant, adolescent, or adult), non- human primates, rats, mice, rabbits, pigs, cows, sheep, goats, horses, dogs, cats, fish, birds, and the like. In one embodiment, the subject is a primate. In one embodiment, the subject is a human. As used herein, a subject is “in need of treatment” if such subject would benefit biologically, medically, or in quality of life from such treatment. A subject in need of treatment does not necessarily present symptoms, particular in the case of preventative or prophylaxis treatments. As used herein, the terms “inhibit,” “inhibition,” or “inhibiting” refer to the reduction or suppression of a given biological process, condition, symptom, disorder, or disease, or a significant decrease in the baseline activity of a biological activity or process. As used herein, “treatment” or “treating” refers to prophylaxis of, preventing, suppressing, repressing, reversing, alleviating, ameliorating, or inhibiting the progress of biological process including a disorder or disease, or completely eliminating a disease. A treatment may be either performed in an acute or chronic way. The term “treatment” also refers to reducing the severity of a disease or symptoms associated with such disease prior to affliction with the disease. “Repressing” or “ameliorating” a disease, disorder, or the symptoms thereof involves administering a cell, composition, or compound described herein to a subject after clinical appearance of such disease, disorder, or its symptoms. “Prophylaxis of” or “preventing” a disease, disorder, or the symptoms thereof involves administering a cell, composition, or compound described herein to a subject prior to onset of the disease, disorder, or the symptoms thereof. “Suppressing” a disease or disorder involves administering a cell, composition, or compound described herein to a subject after induction of the disease or disorder thereof but before its clinical appearance or symptoms thereof have manifest. Organ-on-chip (OOC) systems are revolutionizing tissue engineering by providing dynamic models of tissue structure, organ-level function, and disease phenotypes using human cells. However, nonbiological components of OOC devices often limit the recapitulation of in- vivo-like tissue-tissue crosstalk and morphogenesis. Described herein is a kidney glomerulus- on-a-chip that recapitulates glomerular morphogenesis and barrier function using a biomimetic ultrathin membrane and human induced pluripotent stem cells. The resulting chip comprised a proximate epithelial-endothelial tissue interface, which reconstituted the selective molecular filtration function of healthy and diseased kidneys. Additionally, fenestrated endothelium was successfully induced from human pluripotent stem cells in an OOC device, through in vivo-like paracrine signaling across the ultrathin membrane. Thus, this device provides a dynamic tissue engineering platform for modeling human kidney-specific morphogenesis and function, enabling mechanistic studies of stem cell differentiation, organ physiology and pathophysiology. An alternative membrane that enables in vivo-like transmembrane intercellular crosstalk and basement membrane remodeling by surrounding cells is needed to enhance the physiological relevance of OOCs. To address this unmet need, silk fibroin (SF) was employed to engineer an alternative class of basement membranes for OOC devices. SF is a protein generated by Bombyx mori silkworms that possesses superior mechanical properties, biocompatibility, and versatility, and has been used to create various synthetic materials for drug delivery, biosensor and bioelectronics, and tissue engineering. When SF is exposed to methanol, it undergoes a conformational transition from random coil to tightly packed -sheet that enhances the mechanical properties of SF-based materials and makes them insoluble in water. SF material properties surpass those of other protein-based biomaterials such as collagen and fibril materials, which tend to be more susceptible to rupture under fluid shear stress and handling, making SF a promising candidate for tissue engineering applications. SF scaffolds can support the propagation and differentiation of various cell types including fibroblasts, neurons, cardiomyocytes, lung epithelial stem cells, and kidney cells. For example, electrospun SF membranes were previously found to enhance the adhesion and differentiation of human induced pluripotent stem (hiPS) cell-derived cells when compared to tissue culture plates. As described herein, a kidney glomerulus-on-a-chip system was engineered by using electrospun SF to create an ultrathin membrane that resembles in vivo basement membranes with ECM-like topography and molecular properties, and that robustly supports kidney cell propagation, transmembrane intercellular crosstalk, basement membrane remodeling, molecular size-specific glomerular filtration, and an OOC platform capable of inducing the development of fenestrated endothelium from unspecialized hiPS cell derivatives with specificity. In some embodiments, the devices, methods, and cells described herein may be used in various applications, including, for example, but not limited to, as an in vitro model for a kidney / glomerular disorder, therapeutic applications (e.g., tissue regeneration and / or repair or transplantation), drug discovery and / or developments, and / or tissue engineering. In one aspect, a method of modeling a kidney-specific condition in vitro is provided herein. In some embodiments, podocytes (e.g., immature or mature; post-mitotic) can be used in different applications where podocytes are required, including, but not limited to, as an in vitro model for a kidney / glomerular disorder, therapeutic applications (e.g., tissue regeneration and / or repair or transplantation), drug discovery and / or developments, and / or tissue engineering. The podocytes can be pre-formed and then transferred to the biomimetic microfluidic device described herein, or they can be differentiated in the device from pluripotent stem cells, mesodermal cells, and / or intermediate mesodermal cells using various differentiation methods prior to the culturing. One embodiment described herein is a biomimetic microfluidic device, comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane. In some aspects described herein, the biomimetic microfluidic devices described herein may also be used in air-liquid communication applications. The first channel and the second channel can be of substantially equal (e.g., within 10% or within 5% or less) heights or of different heights. The channel dimensions (e.g., heights and / or widths) can be varied to achieve different levels of fluid shear stress, which can influence cell differentiation and / or cell function. The electrospun silk fibroin membrane can be porous (e.g., permeable, or selectively permeable), non-porous (e.g., non-permeable), rigid, flexible, elastic, or any combination thereof. In some embodiments, the membrane can be porous, e.g., allowing exchange / transport of fluids (e.g., gas and / or liquids), passage of molecules such as nutrients, cytokines and / or chemokines, cell transmigration, or any combinations thereof. In some embodiments, the membrane can be non-porous. In one aspect, the electrospun silk fibroin membrane is porous and comprises a porosity of about 1% to about 70%. In another aspect, the electrospun silk fibroin membrane is porous and comprises a porosity of about 40% to about 70%. In one aspect, the polymeric base material comprises polydimethylsiloxane (PDMS), polycarbonate (PC), poly-methyl-meta-acrylate (PMMA), cyclic olefin copolymer (COC), polyimide, or combinations thereof. In another aspect, the electrospun silk fibroin membrane further comprises polyethylene oxide (PEO). In another aspect, the biomimetic microfluidic device further comprises a pump fluidly connected to the first channel and the second channel and configured to pump fluids through the first channel and the second channel. Pneumatic pressures or vacuum can be applied to the channels to cause the electrospun silk fibroin membrane to flex or stretch. In another aspect, the electrospun silk fibroin membrane further comprises a laminin coating on the first surface, the second surface, or a combination thereof. The membrane is amenable to functionalization with appropriate extracellular matrix (ECM) proteins. In another aspect, the electrospun silk fibroin membrane comprises a thickness of at least about 0.1 μm. In another aspect, the electrospun silk fibroin membrane comprises a thickness of at least about 0.5 μm, at least about 1 μm, at least about 5 μm, at least about 10 μm, at least about 25 μm, or at least about 50 μm. In another aspect, the electrospun silk fibroin membrane comprises a thickness of about 0.1 μm to about 100 μm. In another aspect, the electrospun silk fibroin membrane comprises a thickness of about 0.5 μm to about 100 μm, about 1 μm to about 100 μm, about 5 μm to about 100 μm, about 10 μm to about 100 μm, about 25 μm to about 100 μm, about 50 μm to about 100 μm, or about 75 μm to about 100 μm. In another aspect, the electrospun silk fibroin membrane comprises a thickness of about 0.1 μm to about 75 μm, about 0.1 μm to about 50 μm, about 0.1 μm to about 25 μm, about 0.1 μm to about 10 μm, about 0.1 μm to about 5 μm, about 0.1 μm to about 1 μm, or about 0.1 μm to about 0.5 μm. In another aspect, the electrospun silk fibroin membrane comprises a thickness of about 0.5 μm to about 7.5 μm. In another aspect, the electrospun silk fibroin membrane comprises a Young’s modulus of about 25 kPa to about 100 kPa. In another aspect, the electrospun silk fibroin membrane comprises a Young’s modulus of about 50 kPa to about 100 kPa, or about 50 kPa to about 75 kPa. In another aspect, the population of cells is cultured on the first surface of the electrospun silk fibroin membrane, the second surface of the electrospun silk fibroin membrane, or a combination thereof. In another aspect, the population of cells comprises induced pluripotent stem cells (iPSCs), embryonic stem cells (ESCs), kidney glomerular cells, podocytes, intermediate mesoderm (IM) cells, mesoderm cells, endothelial cells, embryoid bodies, organoids, spheroids, or combinations thereof. In another aspect, podocytes are cultured on the first surface of the electrospun silk fibroin membrane and endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane. In another aspect, the endothelial cells comprise fenestrated endothelial cells. In some embodiments, the biomimetic microfluidic devices described herein may comprise one or more kidney-associated cells distributed in the first channel or second channel. Non- limiting examples of kidney-associated cells include, but are not limited to, endothelial cells, mesangial cells, epithelial cells, smooth muscle cells or myocytes, granular cells (Juxtaglomerular cells), parietal cells, proximal tubular cells, loop of Henle thin segment cells, duct cells, connective tissue fibroblasts, pericytes, insulin-producing cells, and combinations thereof. In another aspect, the electrospun silk fibroin membrane mimics a kidney glomerular basement membrane. In another aspect, the first channel mimics a urinary compartment of a kidney glomerulus and the second channel mimics a vascular compartment of a kidney glomerulus. In another aspect, the biomimetic microfluidic device is a kidney glomerulus organ- on-chip device. In another aspect, the biomimetic microfluidic device is part of an extracorporeal dialysis system. Another embodiment described herein is a method of filtering a sample, the method comprising: applying a sample to a biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; wherein podocytes are cultured on the first surface of the electrospun silk fibroin membrane and endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane; and perfusing one or more fluids through the biomimetic microfluidic device. In one aspect, the method further comprises collecting a filtered fraction of the sample. In another aspect, the podocytes are induced pluripotent stem cell (iPSC)-derived podocytes. In another aspect, the sample is a blood sample, a urine sample, or a combination thereof. In another aspect, the method filters one or more uremic toxins from the sample. In another aspect, the method filters one or more biological metabolites from the sample. In another aspect, the one or more biological metabolites comprise inulin, albumin, urea, -2-microglobumin (B2M), visfatin, creatinine, or combinations thereof. In another aspect, the method has an albumin clearance rate of less than about 6%. In another aspect, the method has a urea reduction ratio (URR) of at least about 75%. Another embodiment described herein is a method of screening an agent for treatment of a kidney or glomerular disorder, the method comprising: applying an agent to a biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; wherein podocytes are cultured on the first surface of the electrospun silk fibroin membrane and endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane, and wherein the agent contacts the podocytes; perfusing one or more fluids through the biomimetic microfluidic device; and detecting a response of the podocytes to the agent, the response being associated with a kidney or glomerular disorder. In one aspect, the podocytes are induced pluripotent stem cell (iPSC)-derived podocytes. In another aspect, the response associated with the kidney or glomerular disorder comprises altered podocyte gene expression, altered podocyte protein expression, altered podocyte albumin uptake, altered formation or extension of podocyte foot processes, or combinations thereof. In another aspect, the agent comprises a protein, a peptide, a nucleic acid, a small molecule, or combinations thereof. In another aspect, the kidney or glomerular disorder comprises podocyte injury, proteinuria, glomerulosclerosis, diabetic nephropathy, chemotherapy-related nephrotoxicity, a podocytopathy, or combinations thereof. Another embodiment described herein is a method of generating fenestrated endothelial cells, the method comprising: seeding a population of podocytes or intermediate mesoderm (IM) cells and a population of endothelial cells in a biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; wherein the podocytes or IM cells are cultured on the first surface of the electrospun silk fibroin membrane and the endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane; and perfusing one or more fluids through the biomimetic microfluidic device to generate the fenestrated endothelial cells. Another embodiment described herein is a kit for filtering a sample, comprising: a biomimetic microfluidic device, comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; optionally, pumps, tubing, syringes, and containers for applying fluids to the biomimetic microfluidic device; optionally, cell culture buffers and reagents; and optionally, packaging, labels, or instructions for use. Another embodiment described herein is the use of a biomimetic microfluidic device for filtering a sample, the biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane. Another embodiment described herein is the use of a biomimetic microfluidic device for screening an agent for treatment of a kidney or glomerular disorder, the biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane. It will be apparent to one of ordinary skill in the relevant art that suitable modifications and adaptations to the compositions, formulations, methods, processes, and applications described herein can be made without departing from the scope of any embodiments or aspects thereof. The compositions and methods provided are exemplary and are not intended to limit the scope of any of the specified embodiments. All of the various embodiments, aspects, and options disclosed herein can be combined in any variations or iterations. The scope of the compositions, formulations, methods, and processes described herein include all actual or potential combinations of embodiments, aspects, options, examples, and preferences herein described. The exemplary compositions and formulations described herein may omit any component, substitute any component disclosed herein, or include any component disclosed elsewhere herein. The ratios of the mass of any component of any of the compositions or formulations disclosed herein to the mass of any other component in the formulation or to the total mass of the other components in the formulation are hereby disclosed as if they were expressly disclosed. Should the meaning of any terms in any of the patents or publications incorporated by reference conflict with the meaning of the terms used in this disclosure, the meanings of the terms or phrases in this disclosure are controlling. Furthermore, the foregoing discussion discloses and describes merely exemplary embodiments. All patents and publications cited herein are incorporated by reference herein for the specific teachings thereof. Various embodiments and aspects of the inventions described herein are summarized by the following clauses: Clause 1. A biomimetic microfluidic device, comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane. Clause 2. The biomimetic microfluidic device of clause 1, wherein the polymeric base material comprises polydimethylsiloxane (PDMS), polycarbonate (PC), poly-methyl-meta- acrylate (PMMA), cyclic olefin copolymer (COC), polyimide, or combinations thereof. Clause 3. The biomimetic microfluidic device of clause 1 or 2, wherein the electrospun silk fibroin membrane further comprises polyethylene oxide (PEO). Clause 4. The biomimetic microfluidic device of any one of clauses 1–3, further comprising a pump fluidly connected to the first channel and the second channel and configured to pump fluids through the first channel and the second channel. Clause 5. The biomimetic microfluidic device of any one of clauses 1–4, wherein the electrospun silk fibroin membrane further comprises a laminin coating on the first surface, the second surface, or a combination thereof. Clause 6. The biomimetic microfluidic device of any one of clauses 1–5, wherein the electrospun silk fibroin membrane comprises a thickness of at least about 0.1 μm. Clause 7. The biomimetic microfluidic device of any one of clauses 1–6, wherein the electrospun silk fibroin membrane comprises a thickness of about 0.1 μm to about 100 μm. Clause 8. The biomimetic microfluidic device of any one of clauses 1–7, wherein the electrospun silk fibroin membrane comprises a thickness of about 0.5 μm to about 7.5 μm. Clause 9. The biomimetic microfluidic device of any one of clauses 1–8, wherein the electrospun silk fibroin membrane is porous and comprises a porosity of about 1% to about 70%. Clause 10. The biomimetic microfluidic device of any one of clauses 1–9, wherein the electrospun silk fibroin membrane is porous and comprises a porosity of about 40% to about 70%. Clause 11. The biomimetic microfluidic device of any one of clauses 1–10, wherein the electrospun silk fibroin membrane comprises a Young’s modulus of about 25 kPa to about 100 kPa. Clause 12. The biomimetic microfluidic device of any one of clauses 1–11, wherein the population of cells is cultured on the first surface of the electrospun silk fibroin membrane, the second surface of the electrospun silk fibroin membrane, or a combination thereof. Clause 13. The biomimetic microfluidic device of any one of clauses 1–12, wherein the population of cells comprises induced pluripotent stem cells (iPSCs), embryonic stem cells (ESCs), kidney glomerular cells, podocytes, intermediate mesoderm (IM) cells, mesoderm cells, endothelial cells, embryoid bodies, organoids, spheroids, or combinations thereof. Clause 14. The biomimetic microfluidic device of any one of clauses 1–13, wherein podocytes are cultured on the first surface of the electrospun silk fibroin membrane and endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane. Clause 15. The biomimetic microfluidic device of any one of clauses 1–14, wherein the endothelial cells comprise fenestrated endothelial cells. Clause 16. The biomimetic microfluidic device of any one of clauses 1–15, wherein the electrospun silk fibroin membrane mimics a kidney glomerular basement membrane. Clause 17. The biomimetic microfluidic device of any one of clauses 1–16, wherein the first channel mimics a urinary compartment of a kidney glomerulus and the second channel mimics a vascular compartment of a kidney glomerulus. Clause 18. The biomimetic microfluidic device of any one of clauses 1–17, wherein the biomimetic microfluidic device is a kidney glomerulus organ-on-chip device. Clause 19. The biomimetic microfluidic device of any one of clauses 1–18, wherein the biomimetic microfluidic device is part of an extracorporeal dialysis system. Clause 20. A method of filtering a sample, the method comprising: applying a sample to a biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; wherein podocytes are cultured on the first surface of the electrospun silk fibroin membrane and endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane; and perfusing one or more fluids through the biomimetic microfluidic device. Clause 21. The method of clause 20, further comprising collecting a filtered fraction of the sample. Clause 22. The method of clause 20 or 21, wherein the podocytes are induced pluripotent stem cell (iPSC)-derived podocytes. Clause 23. The method of any one of clauses 20–22, wherein the sample is a blood sample, a urine sample, or a combination thereof. Clause 24. The method of any one of clauses 20–23, wherein the method filters one or more uremic toxins from the sample. Clause 25. The method of any one of clauses 20–24, wherein the method filters one or more biological metabolites from the sample. Clause 26. The method of any one of clauses 20–25, wherein the one or more biological metabolites comprise inulin, albumin, urea, -2-microglobumin (B2M), visfatin, creatinine, or combinations thereof. Clause 27. The method of any one of clauses 20–26, wherein the method has an albumin clearance rate of less than about 6%. Clause 28. The method of any one of clauses 20–27, wherein the method has a urea reduction ratio (URR) of at least about 75%. Clause 29. A method of screening an agent for treatment of a kidney or glomerular disorder, the method comprising: applying an agent to a biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; wherein podocytes are cultured on the first surface of the electrospun silk fibroin membrane and endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane, and wherein the agent contacts the podocytes; perfusing one or more fluids through the biomimetic microfluidic device; and detecting a response of the podocytes to the agent, the response being associated with a kidney or glomerular disorder. Clause 30. The method of clause 29, wherein the podocytes are induced pluripotent stem cell (iPSC)-derived podocytes. Clause 31. The method of clause 29 or 30, wherein the response associated with the kidney or glomerular disorder comprises altered podocyte gene expression, altered podocyte protein expression, altered podocyte albumin uptake, altered formation or extension of podocyte foot processes, or combinations thereof. Clause 32. The method of any one of clauses 29–31, wherein the agent comprises a protein, a peptide, a nucleic acid, a small molecule, or combinations thereof. Clause 33. The method of any one of clauses 29–32, wherein the kidney or glomerular disorder comprises podocyte injury, proteinuria, glomerulosclerosis, diabetic nephropathy, chemotherapy-related nephrotoxicity, a podocytopathy, or combinations thereof. Clause 34. A method of generating fenestrated endothelial cells, the method comprising: seeding a population of podocytes or intermediate mesoderm (IM) cells and a population of endothelial cells in a biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; wherein the podocytes or IM cells are cultured on the first surface of the electrospun silk fibroin membrane and the endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane; and perfusing one or more fluids through the biomimetic microfluidic device to generate the fenestrated endothelial cells. Clause 35. A kit for filtering a sample, comprising: a biomimetic microfluidic device, comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; optionally, pumps, tubing, syringes, and containers for applying fluids to the biomimetic microfluidic device; optionally, cell culture buffers and reagents; and optionally, packaging, labels, or instructions for use. Clause 36. Use of a biomimetic microfluidic device for filtering a sample, the biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane. Clause 37. Use of a biomimetic microfluidic device for screening an agent for treatment of a kidney or glomerular disorder, the biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane. EXAMPLES Example 1 Generation of Electrospun SF Membranes 5% SF aqueous solution (Advanced Biomatrix, 5154) was dialyzed against 90% polyethylene glycol (Sigma, P6667) aqueous solution in a Slide-A-Lyzer Dialysis Cassette (Thermo Fisher Scientific, 66380) for 22 h at room temperature, followed by reconstitution in ultrapure water to achieve an 8% SF solution. The solution was mixed gently with 10% polyethylene oxide (PEO; Sigma, 189456) aqueous solution at a ratio of 3.9:1 SF:PEO, and was loaded into a 10-mL syringe (BD, 14-823-2A) (slowly to prevent air bubble formation). A 16-gauge blunt-tip needle (McMASTER-CARR, 75165A753) was attached to the syringe, and the syringe was loaded onto a syringe pump (Chemyx, F10071) to provide a flow rate of 0.015 mL / min. For electrospinning, power was provided by a high-voltage power supply (Gamma High Voltage Research, ES100P-20W) at 11 kV, and the SF:PEO mixture solution was electrospun onto flat non-stick aluminum foil at a distance of 20 cm for 10 min. The resulting electrospun mats were used immediately for device engineering (described below) or stored at 4 °C. BioMF OOC Device Engineering PDMS chips were generated by casting a prepolymer mixture containing Sylgard 184 elastomer and curing agent (Ellsworth, 184 SIL ELAST KIT 0.5KG) at a ratio of 10:1 in custom- designed 3D-printed molds, including a top channel mold with a 18 mm × 1 mm × 1 mm (L × W × H) dimension and a bottom channel mold with a 18 mm × 1 mm × 0.2 mm dimension (Protolabs) for 4 h at 65 °C. The resulting PDMS chips were peeled from the molds and stored at room temperature. To assemble the chips, electrospun SF / PEO membranes were cut into 5 cm × 0.5 cm rectangles and immersed in 90% methanol solution for 20 min in a fume hood. Afterwards, the membranes were air-dried overnight at room temperature in the fume hood, followed by immersion in DI water for 48 h to remove the polyethylene oxide. Subsequently, the membranes were air-dried at room temperature and peeled from the foil. The resulting SF membranes were transferred onto a bottom PDMS chip pre-stamped with spin-coated PDMS prepolymer mixture glue (3500 rpm for 10 min) that contained Sylgard 184 elastomer and curing agent at a ratio of 10:3. The SF / PDMS bottom chip was stored at room temperature overnight to allow the PDMS glue to cure. A top PDMS chip was then stamped with the spin-coated PDMS glue and stamped against the SF / PDMS bottom chip to coat the area of the SF membrane positioned outside the fluidic microchannel with PDMS glue, followed by curing of the PDMS glue for overnight at room temperature. The SF / PDMS bottom chip was then coated with PDMS glue again using the same method, followed by curing of the PDMS glue for 48 h at room temperature. Subsequently, a new top PDMS chip and the SF / PDMS bottom chip were treated with oxygen plasma at 50 W and 0.8 mbar for 60 s using a plasma asher (Emitech K-1050X). The top PDMS chip was pressed against the SF / PDMS bottom chip while carefully aligning the fluidic microchannels. Immediately after bonding the chips, 25 L of 25 g / mL laminin-511 solution (Biolamina, LN511) were added to the bottom channel of the chip and 30 L to the top channel of the chip, followed by incubation at 37 °C overnight. HiPS Cell Culture The hiPS cell line PGP-1 (Personal Genome Project) was propagated on 6-well tissue- culture-treated polystyrene plates (VWR, 10062-892) coated with pluripotent stem cell-qualified Matrigel (BD Biosciences, 354277). The cells were cultured in mTeSR1 medium (Stem Cell Technologies, 85857) and incubated at 37 °C and 5% CO2, and the medium was replenished daily. The cells were passaged every 4 to 5 days by treatment with StemPro Accutase (Thermo Fisher Scientific, 07920). The cells were tested quarterly for and were found to be free of mycoplasma contamination (Mycoplasma PCR Detection Kit, ABM, G238). Chromosomal analysis confirmed that these cells were karyotypically normal. Differentiation of Intermediate Mesoderm Cells The hiPS cells were dissociated from Matrigel-coated plates by treatment with enzyme- free cell dissociation buffer (Thermo Fisher Scientific, 13150016) and were centrifuged for 5 min at 200 × g in DMEM / F12. The cells were resuspended in a mesoderm differentiation medium containing DMEM / F12 with GlutaMax (GIBCO, 10565042) supplemented with 100 ng / mL activin A (Thermo Fisher Scientific, PHC9561), 3 M CHIR99021 (Stemgent, 04-0004), 10 M Y27632 (TOCRIS Bioscience, 1254), and 1× B27 serum-free supplement (GIBCO, 17504044). The cells were seeded at a density of 100,000 cells per well of a 12-well tissue-culture-treated polystyrene plates (VWR, 10062-894) coated with 5 g / mL laminin-511-E8 (iMatrix-511, T304) solution in sterile distilled water (GIBCO, 15230204). The cells were then cultured for 2 days with mesoderm differentiation media at 37 °C and 5% CO2, and the culture medium was replaced with an intermediate mesoderm induction medium (IM medium) containing DMEM / F12 with GlutaMax supplemented with 100 ng / mL BMP7 (Thermo Fisher Scientific, PHC9541), 3 M CHIR99021, and 1× B27 serum-free supplement. The cells were incubated at 37 °C and 5% CO2for a minimum of 14 days with daily medium replacement. Differentiation of Endothelial Cells A previously established protocol was adapted for the differentiation of endothelial cells. Briefly, hiPS cells were propagated on 6-well tissue-culture-treated plates coated with Matrigel, and were cultured with mTeSR1 supplemented with 10 μM ROCK inhibitor Y-27632 medium for 1 day at 37 °C and 5% CO2, followed by culturing with N2B27 media (neurobasal media; Invitrogen, 21103049) and DMEM / F12 glutamax (Invitrogen) at a 1:1 ratio with 100× N2 (Invitrogen,17502048) and B27 ( ) Vitamin A (Invitrogen, 12587010) supplemented with 8 M CHIR99021(Stemgent) and 25 ng / mL Hbmp4 (PeproTech, 120-05ET). The cells were cultured in N2B27 media for 3 days (without media exchange) at 37 °C and 5% CO2to induce lateral mesoderm differentiation. Afterwards, N2B27 media was replaced with endothelial induction media containing StemPro-34 SFM media (Invitrogen, 10639011) supplemented with Glutamax and Pen / Strep (Gibco Cell Culture, 15140122) at a 100:1:1 ratio, 2 M forskolin (Abcam, ab120058), and 200 ng / mL VEGF (Invitrogen, PHC9391). The endothelial induction media was exchanged daily for 2 days, and conditioned media was collected 1 day after each media exchange. The cells were then dissociated with Accutase and MACS-sorted using QuadroMACS Separator (Miltenyibiotec, 130-042-302) to harvest CD144+ and CD31+ cells using human CD144 microbeads (Miltenyibiotec, 130-097-857) and human CD31 microbeads (Miltenyibiotec, 130- 091-935). The sorted cells were propagated on a tissue-culture-treated plate (6-well polystyrene plates, VWR) coated with 25 g / mL human fibronectin (Advanced Biomatrix, 5050) and cultured with the conditioned media / StemPro-34 SFM (1:1) supplemented with 2 g / mL heparin (STEMCELL Technologies, 7980). Media was exchanged every 2 days until the conditioned media was used up. The cells were further propagated with CultureBoost medium (Cell Systems, 4Z0-500-R) supplemented with 100 U / mL Pen / Strep. Cell Seeding, Podocyte Differentiation, and Cell Culture in the bioMF OOC Device Before cell seeding, the chip bottom channels were rinsed with 200 L pre-warmed CultureBoost medium supplemented with 100 U / mL Pen / Strep, and the chip top channels were rinsed with 200 L pre-warmed DMEM / F12. The chips were then kept at 37 °C until later use. Vascular endothelial cells derived from hiPS cells were dissociated from the 6-well plate by incubating them with 2 mL / well Accutase (Invitrogen) for 5 min at 37 °C, and then transferred to a 15-mL conical tube (VWR, 10026-076). The cell suspension was centrifuged for 5 min at 200 g, and the supernatant was aspirated. Then, the cell palate was resuspended to 6 × 106cells / mL suspension using pre-warmed CultureBoost medium supplemented with 100 U / mL Pen / Strep.For endothelial cell seeding, the chip bottom channel media were removed by using a 200- L pipette, followed by adding 25 L of the endothelial cell suspension into the same bottom channel. After adding the cell suspension, the chip was immediately flipped over to allow cell adhesion to the SF membrane. The chips were incubated at 37 °C and 5% CO2for 4 h to ensure cell attachment. Then, the chip bottom channels were rinsed again with pre-warmed CultureBoost supplemented with 100 U / mL Pen / Strep, and the top channels were rinsed with pre-warmed DMEM / F12. The chips were incubated in the upside-down orientation for overnight at 37 °C and 5% CO2. To prepare for intermediate mesoderm cell seeding, the chip’s fluidic channels were gently rinsed with CultureBoost medium supplemented with 100 U / mL Pen / Strep (bottom channel) and IM media supplemented with 100 U / mL Pen / Strep (top channel) and incubated at 37 °C and 5% CO2while preparing the cell suspension. Intermediate mesoderm cells in a 12-well plate were dissociated with 500 L / well 0.05% trypsin / EDTA (Gibco) by incubating for 5 min at 37 °C. Then, the cells were further dissociated using a cell scraper (Corning) and collected into 2 mL Trypsin Neutralization Buffer (10% HI-FBS, Sigma, in DMEM / F12) for each 500 L trypsin / EDTA. The cell suspension was centrifuged for 5 min at 200 g, followed and the supernatant was aspirated. The cells were then resuspended to 200,000 cells / mL concentration using pre-warmed IM media supplemented with 100 U / mL Pen / Strep. For seeding the intermediate mesoderm cells, the chiptop channel medium was removed gently using a 200- L pipette, followed by adding 30 L of theIM cell suspension into the same top channel. The chips were then incubated at 37 °C and 5% CO2for 4 h before gently rinsing the chip bottom and top channels again with pre-warmed CultureBoost and IM media, respectively. Subsequently, both the IM cells and the endothelial cells in the chips were incubated overnight at 37 °C and 5% CO2. Afterwards, the chip top channel was rinsed gently with pre-warmed podocyte induction media consisting of DMEM / F12 with GlutaMAX supplemented with 50 ng / mL VEGF, 1× B27, 100 ng / mL Activin A, 100 ng / mL BMP-7, 0.1 M all-trans retinoic acid (Stem Cell Technologies, 72262), and 3 M CHIR99021) supplemented with 100 U / mL Pen / Strep, and the chip bottom channel was rinsed gently with pre- warmed media consisting of CultureBoost supplemented with 100 U / mL Pen / Strep, followed by further incubation overnight at 37 °C and 5% CO2. Subsequently, the chips were connected to the Human Emulation System (Emulate, Inc) to provide the top and bottom channels with pulsatile flow of the podocyte induction media and the CultureBoost (for endothelial cells) at a flow rate of 1 L / min. The chips were perfused at the flow rate of 1 L / min in a 37 °C and 5% CO2incubator for 4 days, and the media reservoirs were replenished every 2 days. Results demonstrating Nidogen-1 and Agrin production, as well as VEGF-A immunostaining and secretion, were obtained by perfusing the chips using a different peristaltic pump (Ismatec Ip Digital peristaltic pump, Cole-Parmer, EW-78001-32) at a flow rate of 4.09 L / min (based on equipment capability), where the podocyte induction media and the CultureBoost were recirculated in the system and replenished every 2 days. The chips were successfully propagated with cells using either the Emulate system or the more affordable Ismatec peristaltic pump. Disease Modeling Adriamycin (ADR; LC Laboratories, D-4000) was reconstituted in DMSO (Sigma, D2438) to generate a stock solution of 1 mg / mL concentration. ADR medium was prepared by adding 0.5 L stock solution and 0.5 L DMSO into every 1 mL of CultureBoost to achieve a final Adriamycin concentration of 0.5 μg / mL. DMSO medium (vehicle control) was made by adding 1 L DMSO into every 1 mL of CultureBoost to achieve a final DMSO concentration of 0.1% v / v. ADR medium or DMSO medium were added into the chip’s bottom channel inlet media reservoirs, and CultureBoost was added into each chip’s top channel inlet media reservoirs. The chips were further perfused for 3 days at 37 °C and 5% CO2. Molecular Filtration Assay Albumin-Texas Red (Thermo Fisher Scientific, A23017) was reconstituted in 1 mL CultureBoost to prepare a 5 mg / mL albumin stock solution; inulin-FITC (Sigma, F3272) was reconstituted in CultureBoost to prepare a 7.6 mg / mL inulin stock solution, and urea (Sigma, U5378) was reconstituted in sterile water to prepare a 104.24 mg / mL urea stock solution. The albumin, inulin, and urea stock solutions were added to the CultureBoost medium to achieve a final concentrations of 100 g / mL for albumin, 10 g / mL for inulin, concentration of, and 300 g / mL for urea. CultureBoost supplemented with albumin, inulin, and urea was added into all chip bottom channel inlet media reservoirs, and CultureBoost was added into all chip top channel inlet media reservoirs. Media from the outlet media reservoirs were aspirated. The chips were then perfused continuously for 6 h at 37 °C and 5% CO2. Then, 100 L media from the top channel outlet reservoirs and bottom channel inlet reservoirs were collected into a flat clear bottom black 96-well plate (Corning, CLS3614) in technical replicates with albumin and inulin standard made by serial dilutions of 100 g / mL albumin and 10 g / mL inulin. The fluorescence intensities of all wells were detected using a plate reader (Molecular Devices SpectraMax iD3). Albumin was detected using excitation at 550 nm and emission at 615 nm; inulin was detected using excitation at 513 nm and emission at 557 nm. A urea assay kit (Abcam, ab83362) was used to determine urea concentration. Data in FIG. 2E were generated using the Emulate system, and data in FIG.2H were generated using the Ismatec peristaltic pump. The data were processed and analyzed using Microsoft Excel 2023 and GraphPad Prism (version 9.5.1.). Clearance (%) was calculated by the equation shown below, where [U] is urinary outflow concentration and [V] is the total vascularstock concentration. Urea Assay Urea concentrations in channel outflow media were determined by using a urea assay kit (Abcam, ab83362). Briefly, cell culture supernatant was collected from the peristaltic pump outlet media reservoirs and was centrifuged at 1,500 rpm for 10 min at 4 °C, the resulting supernatant was further diluted 2-fold and 50 L of each sample was added along with the standard dilution (50 L) into a flat bottom 96-well plate (Thermo Fisher Scientific, 168055) in technical duplicates. 50 L CultureBoost without supplements (background control) was included along with the samples and standard dilution and mixed with 50 L background reaction mix, and the samples and standards were mixed with 50 L reaction mix. The samples, background controls, and standards were then incubated for 1 h at 37 °C. The plate was then immediately analyzed by fluorescence plate reading (Molecular Devices SpectraMax iD3) at 570 nm. The data were processed and analyzed using Microsoft Excel 2023 and GraphPad Prism (version 9.5.1.). Immunofluorescence Microscopy Analyses Cells in the glomerulus bioMF OOC devices were rinsed once with 1× DPBS (Gibco, 14190144), followed by fixation with 4% paraformaldehyde (Thermo Fisher Scientific, 28908) for 20 min at room temperature. Subsequently, the cells were washed once with 0.125% Triton- X / DPBS (Triton-X, Thermo Fisher Scientific, HFH10) for 5 min at room temperature to permeabilize the cells, followed by incubation with 1% BSA in 0.125% Triton-X / DPBS (BSA, Sigma) for 1 h at room temperature to block unspecific binding. Afterwards, the cells were rinsed 3 times with 0.125% Triton-X / DPBS, 5 min at room temperature for each rinse, then the cells were incubated overnight at 4 °C with primary antibodies of interest: anti-nephrin, 1:200 dilution (American Research Products, GPN02), anti-VE-Cadherin 1:200 dilution (Santa Cruz, sc-9989), anti-podocin, 1:200 dilution (Abcam, ab50339), anti-collagen-IV, 1:200 dilution (Thermo Fisher Scientific, 14-9871-82), anti-nidogen-1, 1:50 dilution (Thermo Fisher Scientific, MA5-23911), anti- agrin, 1:100 dilution (Thermo Fisher Scientific, PA5-103585), and anti-VEGF-165, 1:20 dilution (Bio-Techne, AF-293-SP). The cells were rinsed three times with 0.125% Triton-X / DPBS, 5 min at room temperature for each rinse, and were incubated with the respective secondary antibodies (at 1:1000 dilution) for 1 h at room temperature, followed by rinsing three times with 1× DPBS and once with distilled water (Gibco, 15230001). The secondary antibodies used in this study were AF488 goat anti-guinea pig (Thermo Fisher Scientific, A11073), AF594 donkey anti-mouse (Thermo Fisher Scientific, A21203), AF488 donkey anti-mouse (Thermo Fisher Scientific, A21202), AF594 donkey anti-rabbit (Thermo Fisher Scientific, A21207), AF488 donkey anti-rabbit (Thermo Fisher Scientific, A21206), and AF488 donkey anti-goat (Thermo Fisher Scientific, A11055). The cells were then incubated with DAPI (1:1000 in distilled water) for 5 min at room temperature, followed by incubation with 0.03% Sudan Black B (Sigma, 199664) in 70% ethanol (VWR, 89125-172) for 5 min at room temperature. Then, the cells were rinsed 3 times with 1× DPBS and stored in 1× DPBS for immunofluorescence visualization. The cells in the chips were visualized by Leica SP8 upright confocal system (Leica), the z-stack images were processed and analyzed by Fiji ImageJ (version 1.53f51). The 3D-reconstructed image shown in FIG.2C was captured by flipping the device over to visualize each side separately at the same location, due to the opacity of the SF membrane. Another 3D-reconstructed image obtained without flipping the device over (FIG.7) shows the native proximity of the podocytes and endothelial cells in the bioMF OOC device. Enzyme-Linked Immunosorbent Assay An enzyme-linked immunosorbent assay for VEGF-A was performed using a Human VEGF-A ELISA kit (RayBiotech). Briefly, cell culture supernatant was collected from the Human Emulation System® outlet media reservoirs and was centrifuged at 1,500 rpm for 10 min at 4 °C, the resulting supernatant was further diluted 3-fold and 100 L of each sample was added along with the standard dilution (100 L) into 8-well strips from the assay kit in technical duplicates. The samples and standards were then incubated for 2.5 h followed by rinsing four times with 1× Wash Solution, incubation with biotinylated antibody for 1 h at room temperature, incubation with streptavidin solution for 45 min at room temperature, incubation with TEM One-Step Substrate Reagent for 30 min at room temperature in dark, and then adding Stop Solution. The plate was immediately analyzed by fluorescence plate reading (Molecular Devices SpectraMax iD3) at 450 nm. The results were processed and analyzed using Microsoft Excel 2023 and GraphPad Prism (version 9.5.1.). Scanning Electron Microscopy For imaging the glomerulus bioMF OOC devices, air-dried chips were sliced using a scalpel (Thermo Fisher Scientific, 08-927-5B) for cross-sectional imaging, and the resulting chip slices (~5 mm thick) and SF / PDMS bottom chips were sputter-coated with gold at 11 mA for 200 s. For imaging PDMS chips, air-dried PDMS chips were sliced and sputter-coated following the same method used for the glomerulus bioMF OOC devices. For PDMS membrane morphological characterization, the PDMS chip top compartment was peeled off, and the resulting exposed PDMS membrane was sputter coated with gold at 11 mA for 200 s. The resulting chip and PDMS chip samples were visualized using an Apreo S scanning electron microscopy system (Thermo Fisher Scientific) at 2kV and 25 pA. The images were processed and analyzed by Fiji Image J (version 1.53f51). Transmission Electron Microscopy Cells in the chips were rinsed once with 1× DPBS, followed by fixation with a fixative solution containing 10 mL of 20% formaldehyde (Sigma, 8.18708), 4 mL of 25% glutaraldehyde (Sigma, G7651), 5 mL of 10X phosphate buffered saline (Gibco, 14200075), and 31 mL of distilled water (Gibco, 15230001), overnight at 4 °C or 3 h at room temperature (~25 °C). The samples were post-fixed with 1% osmium tetroxide (Electron Microscopy Sciences, 19180) for 1 h at room temperature in the fume hood, followed by fixation with 0.5% uranyl acetate (Electron Microscopy Sciences, 22400) and dehydration with ethanol (VWR) gradient. The chip’s top compartment was then removed, followed by immersion of the resulting chips in the Spurr’s resin containing 4.1 g ERL 4221 (Electron Microscopy Sciences, 15004), 5.9 g NSA (Electron Microscopy Sciences, 19050), 1.43 g DER 736 (Electron Microscopy Sciences, 13000), and 0.1 mL DMAE (Electron Microscopy Sciences, 13300), overnight at room temperature. Afterwards, the chips were immersed in a freshly prepared Spurr’s resin and cured at 50-60 °C for 24 to 48 h. The resulting chip samples were microtome-sliced after peeling off the bottom PDMS compartment, loaded on TEM grids, counterstained, and visualized using TEM (FEI Tecnai G2Twin). Micrographs were organized using Adobe Illustrator 2023. SF membrane porosity was quantified using Fiji Image J (version 1.53f51) by specifically focusing on the center part of the SF membrane (to avoid background signals from the cell layers), as depicted in FIG. 9. Endothelial fenestration was quantified using Fiji Image J by quantifying the cells in the field of view. Micro-strain analyzer Young’s modulus of the electrospun SF membranes was measured by a micro-strain analyzer (TA Instruments) using uniaxial tensile testing. The SF membranes were cut into 5 cm × 0.5 cm rectangles and loaded onto a tension immersion fixture to immerse the SF membrane sample in water at room temperature (25 °C) throughout the measurement. The stress-strain curves were obtained at a 100 s testing duration and a 0.47 mm / s uniaxial stretching rate, using n = 10 individual samples from two independent SF membrane synthesis sessions. Each sample was measured once. The Young’s modulus values were calculated by measuring the slope of the linear region of the stress-strain curves using GraphPad Prism software (version 9.5.1.). Attenuated total reflectance accessory – Fourier transformed infrared spectroscopy Fourier transformed infrared spectroscopy (FTIR) analysis of the electrospun SF membrane was performed using a Thermo Electron Nicolet 8700 (Thermo Fisher Scientific) with an attenuated total reflectance (ATR) accessory. All spectra were obtained in absorbance modeon a Ge crystal at a resolution of 4 cm 1 with 32 scans and a spectral range of 500 to 2000 cm 1.The background spectrum was acquired by using a piece of bare non-stick Aluminum foil prior to sampling. For each sample, three different locations were measured, and a total of three independent samples were measured for quantification and to assess reproducibility. For each sample, the average values of 3 measured locations were used for the absorption peak analysis using GraphPad Prism software (version 9.5.1.). Statistical Analysis All data were presented as mean ± standard deviation. To compare two groups with one variable, the data were analyzed using unpaired two-tailed Student’s t-test analysis with a 95% confidence interval. Differences were considered significant at * p 0.05, ** p 0.01, and *** p 0.001. All statistical analyses were performed in GraphPad Prism software (version 9.5.1.). Engineering and Characterization of the Kidney Glomerulus-On-A-Chip Device The biomimetic microfluidic (bioMF) OOC device manufacturing process is illustrated in FIG. 1A. SF membranes were generated by electrospinning as described in the Experimental Methods and were sectioned into 5 cm × 0.5 cm rectangles, followed by immersion in 90% methanol for 20 min to induce the conformational change of SF chains from random coils to - sheets. Polyethylene oxide (PEO) was added during electrospinning to increase the solution’s viscosity, and was removed by immersing the SF membranes in DI water for 48 h. The PEO-free SF membrane was air-dried and attached to a PDMS chip that contained a microfluidic channel (1 mm × 0.2 mm × 18 mm; bottom channel in FIG. 1B) representing the dedicated vascular compartment using spin-coated PDMS glue. The resulting hybrid SF / PDMS chip was then stamped against a second PDMS chip that contained a larger channel (1 mm × 1 mm × 18 mm; top channel in FIG. 1B) representing the urinary filtrate compartment. PDMS glue was used to secure the portions of SF membranes outside the parallel fluidic channels and between the chip walls. The stamping step was performed twice to ensure sufficient spreading and infiltration of the PDMS glue into the SF membranes. The glue was cured at room temperature for 48 h, and the SF / PDMS bottom chip and a new PDMS top chip were treated with oxygen plasma to facilitate PDMS-PDMS bonding through Si-O-Si covalent bonds. The oxygen plasma treatment also introduced hydroxyl groups on the SF membrane, which increased its hydrophilicity and enhanced adsorption of laminin-511, an ECM protein important for cell adhesion, propagation, and differentiation. To demonstrate the biological function of the chips, hiPS cell-derived vascular endothelial cells and podocytes were utilized to reconstitute the structure and function of the human kidney’s blood filtration barrier—the glomerular capillary wall—which has a unique three- layered tissue-tissue interface that provides size-selective blood filtration. The resulting chips contained two fluidic channels separated by a SF membrane (FIG. 1B). Notably, the membrane is an order of magnitude thinner than the PDMS membranes used previously in other OOC devices) (3.5 μm versus 45 μm) (FIG.1C–D). The SF membranes were highly porous (FIG.1E–F), an important feature for molecular transport and intercellular crosstalk between cell layers on opposite sides of the membrane. Despite the desirably high porosity of the SF membrane, the bioMF OOC device did not leak even after prolonged continuous perfusion. The SF membranes also exhibited a microstructure characterized by intertwining layers of fibers resembling the ultrastructure of native ECM (FIG. 1F and FIG. 6A). When functionalized with laminin to enhance cell adhesion, the SF membranes retain their characteristic fibrous morphological features, while the individual nanofibers appeared to exhibit a slightly rougher surface topology compared to uncoated membranes, indicating robust incorporation of laminin- 511 on the electrospun SF nanofibers (FIG.6A). By contrast, laminin functionalization on porous PDMS membranes resulted in a smoother surface than bare PDMS membranes, where ridges were observed on the surface, however, these ridges can no longer be observed after laminin functionalization (FIG.6B). Additionally, the SF membrane had a bulk Young’s modulus of ~67 kPa (FIG. 7A) that is closely related to the mechanical properties of the human kidneys; this in vivo-like topography and mechanical property of electrospun SF can enhance cell propagation and differentiation. Naturally derived matrices have batch-to-batch variations in their molecular content which limits reproducibility between experiments. To examine whether this engineered platform can avoid this issue, Fourier transform infrared (FTIR) spectroscopy analysis was performed to examine the chemical and functional group characteristics of independently synthesized SF membranes. It was found that membranes from different batches exhibited similar infrared absorbance peak patterns, indicating consistent molecular and conformational composition (FIG. 7B). Engineering the Glomerular Filtration Barrier from Patient-Specific hiPS Cells The glomerulus chip consists of two parallel microfluidic channels, making it amenable to modeling multicomponent tissue systems such as the three-layered podocyte–glomerular basement membrane–endothelial cell interface at the kidney’s blood filtration barrier. In the kidney glomerulus-on-a-chip device, the top channel (1 × 1 mm cross-section) represented the urinary compartment while the bottom channel (1 × 0.2 mm cross-section) represented the vascular compartment (schematic in FIG.1B). To mimic the development of human glomerulus tissue in situ (in the engineered device), podocytes were differentiated from hiPS cell-derived intermediate mesoderm cells (IM cells; precursor of podocytes) and co-cultured them with vascular endothelial cells derived from the same (isogenic) hiPS cells. The IM cells seeded in the top channel were differentiated into mature podocytes by using media supplemented with defined inductive factors while the cells were interfaced with endothelial cells layered on the opposite side of the SF membrane in the vascular channel. After differentiation, the podocytes formed a uniform cell layer and expressed mature podocyte lineage markers, including nephrin and podocin (FIG. 2A), which are important for regulating the formation of podocyte foot processes and blood filtration function. Nephrin at the interface between two or more podocyte foot processes maintains the structural integrity of the slit diaphragm, and podocin regulates foot process formation. Interestingly, podocytes differentiated in the chips expressed high levels of nephrin in the resulting tissue areas or boundaries where multiple podocytes interact to form the zipper-like structures or interdigitating foot process-like patterns observed in native tissues. These cellular phenotypes suggest that the podocytes differentiated in situ were mature and formed cell-cell interactions necessary for tissue function (FIG. 2A, top left, middle left, bottom left, and bottom middle images). Additionally, podocin formed filamentous distribution patterns in podocytes differentiated in the chips, indicating maturation of the cells and formation of foot processes (FIG. 2A, top right, middle right, and bottom right images). HiPS cell-derived endothelial cells maintained a confluent layer on the opposite side of the SF membrane and expressed the endothelial cell marker VE-cadherin (FIG.2B), an adhesion molecule important for regulating cell- cell contact, cell proliferation, and vascular permeability. Additionally, it was found that the endothelial cells lined all sides of the vascular channel (FIG. 8A–B). Given this study’s goal of modeling the glomerular filtration barrier, the endothelial cells lining the SF membrane and at the podocyte-endothelium interface were specifically focused on. These analyses revealed the formation of continuous podocyte and endothelial cell layers on the dedicated apical and basal sides of the SF membrane embedded in the microfluidic chip, as confirmed by 3D confocal imaging of the device (FIG.2C and FIG.9A). Additionally, western blot analysis further confirmed the presence of mature podocytes and endothelial cells in the bioMF OOC device through the presence of strong bands corresponding to synaptopodin and VE-Cadherin (FIG. 9B). Synaptopodin is one of the key podocyte markers that binds to the podocyte actin-cytoskeleton, regulating podocyte actin integrity and exerting a protective effect against podocyte injury caused by dramatic actin cytoskeletal rearrangement or disruption. Podocytes are the major contributors of basement membrane proteins in the native glomerulus. ECM protein expression in the glomerular tissue formed in the chips was examined and it was found that the podocyte layer produced notably more collagen IV than the endothelial cell layer (FIG. 2D and FIG. 10), consistent with the mature kidney phenotype. Other key glomerular basement membrane molecules including nidogen-1 and agrin were also secreted by the podocytes and deposited onto the underlying SF membrane, with low to no expression in the endothelial cells (FIG. 2D and FIG. 10). Nidogen-1 binds laminin and collagen to facilitate basement membrane formation, and agrin is a negatively charged heparan sulfate proteoglycan contributing to the net negative charge of the glomerular basement membrane in vivo. This net negative charge may help restrict filtration and loss of plasma albumin. These cell-secreted ECM molecules were deposited onto the underlying SF membrane, suggesting that the SF membrane can be remodeled by podocytes through secretion of glomerular basement membrane molecules to generate a native ECM-like interface between podocytes and endothelial cells. The kidney’s glomerular filtration barrier removes waste from blood by filtering based on molecular size and charge. Large, negatively charged plasma proteins are retained while smaller waste molecules pass through and enter the urine if they are not reabsorbed. For example, inulin is a small polysaccharide (3–5 kDa) that efficiently crosses the glomerular filtration barrier and is widely used for clinical measurement of glomerular filtration rate in patients. In contrast, albumin is a large (66.5 kDa) negatively charged protein that cannot cross the filtration barrier in a healthy kidney and is retained in the vasculature. To determine whether the podocyte- and endothelial cell-lined chip could selectively retain albumin in the vasculature channel while removing inulin by filtration into the urinary filtrate channel, a permselective filtration assay using inulin conjugated to FITC (inulin-FITC) and albumin conjugated to Texas Red (albumin-Texas Red) was performed. It was found that chips lined with podocytes and endothelial cells showed physiologically relevant levels of selective molecular filtration, removing more than 7% of inulin within 6 hours while retaining more than 99.5% of albumin (FIG. 2E). This selective filtration performance is comparable to estimated in vivo values of 6.6% inulin removal and 99.9% albumin retention in 6 hours (FIG.2F). The in vivo-like three-component system was important for achieving selective filtration rates similar to in vivo glomerular filtration rates as chips lacking the specialized glomerular epithelial and endothelial cells failed to prevent the loss of albumin into the urinary outflow (FIG.2G). These results demonstrate the functionality of the engineered device, enabled by the concerted actions of the two cell layers—podocytes and endothelium—separated by a thin, porous, cell-modified SF membrane. Urea is another important molecule in glomerular filtration but is often neglected in glomerulus-on-a-chip models. Urea is a small molecule (60 Da) formed by catabolism of the toxic cell metabolic waste molecule ammonia in the liver. In the kidneys, urea crosses the glomerular filtration barrier and is excreted with urine. Urea filtration in the bioMF OOC device was evaluated, and it was found that urea exhibited higher clearance than inulin and albumin: urea (60 Da), 11% clearance; inulin (3–5 kDa), 7% clearance; albumin (66.5 kDa), 0.5% clearance (FIG.2H). These results demonstrated the size-selective filtration function of the device. In patients with chronic kidney disease, filtration of urea is compromised leading to uremia, a fatal disease if left untreated. Dialysis is used to facilitate urea removal for late-stage chronic kidney disease patients. Two metrics have been used to evaluate the efficiency of urea removal in dialysis, Kt / V (K: dialyzer clearance; t: dialysis time; V: urea distribution volume) and URR (urea reduction ratio), with the National Kidney Foundation’s dialysis guidelines recommending a Kt / V of 1.2, and URR 65%. In the device, Kt / V reached ~1.8 and URR was ~83%, surpassing the recommended values, indicating that the bioMF OOC device can remove urea and meet dialysis standards. As a result, it was anticipated that the device can be further explored as a potential blood filtration device for removal of urea and other uremic toxins in future work. Together, these selective molecular filtration results demonstrate that the bioMF OOC device exhibits size-selective molecular filtration resembling the human kidney glomerulus. Glomerular Disease Modeling Glomerular diseases are associated with various risk factors including drug toxicity, genetics, bacterial and viral infections, and patient lifestyle. Recapitulation of glomerular injury phenotypes in a biomimetic microfluidic chip would provide a valuable platform for drug discovery and mechanistic studies. Whether the glomerulus chips could recapitulate nephrotoxicity was tested by administering the chemotherapy drug Adriamycin, which causes acute glomerular injury in patients, to the vascular channel of the device to mimic intravenous drug administration. After three days of exposure to a clinically relevant dose of Adriamycin (0.5 μg / mL), the glomerulus bioMF OOC devices developed compromised cellular phenotypes and aberrant filtration function. Podocytes in the Adriamycin-treated chips lost foot processes, became less arborized than the vehicle control (DMSO-treated chips), and exhibited obscure cell-cell boundaries (FIG. 3A), characteristics of injured kidney glomeruli. These observations were supported by the altered expression pattern of podocin; filamentous podocin distribution patterns were no longer observed in the Adriamycin-treated chip’s podocyte layer, where podocyte loss was apparent (FIG.3A–B). These observations are consistent with previous studies that showed podocyte foot process effacement due to Adriamycin-induced podocyte injury. In contrast, the endothelial cells maintained their structure and VE-cadherin expression after Adriamycin treatment (FIG. 3C), successfully modeling cell-type-specific sensitivity to Adriamycin observed in the kidney glomerulus. Permselective molecular filtration was also compromised upon Adriamycin treatment of the chip. Both inulin and albumin clearance rates were significantly higher in the Adriamycin- treated group than in the DMSO-treated control (FIG.3D–E), indicating the onset of albuminuria and loss of size-selective filtration upon exposure to the drug. These results mimic clinical indicators of microalbuminuria in acute kidney injury, in which elevated albumin in patient urine was observed. The demonstration of drug-induced nephrotoxicity in the chips indicates their ability to recapitulate pathophysiological outcomes in response to injury stimuli. Thus, the glomerulus chip platform can be useful for screening drugs for nephrotoxicity, studying disease mechanisms and discovering new therapeutic candidates. Induction of Fenestrated Glomerular Endothelium from hiPS Cells OOCs are often challenging to section for high resolution electron microscopy due to material mechanical property mismatch between the elastic PDMS and rigid cured resins used for tissue embedding and processing. The ultrathin SF membrane-containing chips developed in this study allow easy microtome sectioning for applications in high-power tissue characterization and analysis of cell morphological features (FIG.4A). High resolution transmission electron microscopy (TEM) analysis confirmed that the podocytes and endothelial cells propagated in the glomerulus chips formed respective tissue layers on opposite sides of the SF membrane, creating a sandwich-like three-layered tissue structure that mimics the in vivo glomerular filtration barrier (FIG. 4B). In the glomerulus chips, podocytes formed long protrusions (foot processes; indicated by arrows in the left image in FIG. 4C). Intriguingly, some of the podocyte foot processes encased the engineered SF membrane and wrapped around individual electrospun SF nanofibers, forming secondary and tertiary foot processes (arrows in the middle image in FIG. 4C), indicating robust cell attachment and interaction with the SF membrane (arrows in the right image in FIG.4C). The formation of primary, secondary, and tertiary foot processes indicates a mature podocyte phenotype capable of forming interdigitations with neighboring cell’s foot processes to facilitate glomerular filtration. The SF membranes exhibit high porosity (55.9% void / pore area in the SF membrane cross-section, FIG. 11) with interconnected pores amenable to diffusion and transport of nutrients and signaling molecules across the engineered glomerular filtration barrier. Despite the high porosity, both podocytes and endothelial cells remained adherent to their respective sides of the SF membrane surfaces, without undesirable infiltration into the adjacent tissue. This observation accurately recapitulates the structural organization and features of the healthy glomerular filtration barrier in vivo, where podocytes and endothelial cells are proximal to each other but remain on either side or surface of the ultrathin glomerular basement membrane. The thinness of the SF membrane positions podocytes and endothelial cells proximally, enabling intercellular crosstalk and formation of in vivo-like tissue-tissue interfaces. Indeed, it was found that hiPS cell-derived endothelial cells developed fenestrations only when interfaced with podocytes in the chip (FIG. 4B–D); ~40% of the endothelial cell body constituted the characteristic void-like fenestration morphology (FIG.12). Endothelial cells propagated without podocytes in the chip failed to form the highly specialized fenestration-like structures observed in vivo (FIG.4E and FIG.12). By quantifying the number of fenestrations per pixel in the differentiated endothelial cells, it was found that the number of fenestration-like structures was significantly higher when endothelial cells were cocultured with podocytes than when podocytes were absent (FIG.4F). To visually highlight the distinct cell layers and resulting tissues formed on the SF membrane, pseudo-colored versions of the TEM micrographs were generated as shown in FIG. 4B and FIG. 4E (middle) (FIG. 4G–H). These pseudo-colored images reveal the formation of fenestrations by endothelial cells interfaced with podocytes in the chip. Fenestrations are a key morphological feature of the glomerular endothelium and are induced in vivo by endothelial cell interactions with podocyte-secreted VEGF-A via the receptor VEGFR2. Subsequent activation of PI3-kinase signaling, and rearrangement of the endothelial cell cytoskeleton enable fenestrae development. Intriguingly, methods to induce this tissue- specific endothelial phenotype in vitro from differentiated hiPS cells have yet to be explored. The ability to generate this specialized cell type in this study suggests that the proximity of the podocytes and endothelial cells in the chip supported robust intercellular communication that allowed formation of a tissue-tissue interface and tissue-specific morphogenesis. To explore whether VEGF-A signaling between podocytes and endothelial cells occurs in the chips, the expression of VEGF-A in both podocytes and endothelial cells in the devices was examined (FIG. 5A–B). VEGF-A expression was observed in the podocytes when cocultured with endothelial cells (FIG. 5B and FIG.13), but when only podocytes or only endothelial cells were cultured in the device, the cells failed to form a confluent monolayer and podocyte expression of VEGF-A was low or near background levels (FIG.5B and FIG.13). These results indicate that coculturing podocytes and endothelial cells in the chips leads to the development of a specialized cell phenotype, and that interactions between the podocyte and endothelial tissue layers across the ultrathin SF membrane facilitates podocyte maturation and production of VEGF-A to promote glomerular endothelial cell specialization. Intracellular expression of VEGF-A and its secretion by cells into the soluble medium in the chips was quantified. A VEGF-A concentration of ~14 ng / mL was found in the top channel (FIG. 5C), indicating secretion of VEGF-A from podocytes. Meanwhile, despite the drastically low levels of intracellular VEGF-A expression in the endothelial cell layer (FIG. 5B), a VEGF-A concentration of approximately 2.5 ng / mL was detected in the vascular channel fluid outflow (FIG.5C). In contrast, there were undetectable levels of VEGF-A in the cell culture medium alone (background or medium unexposed to the cells) (FIG. 5C). Together, these data underscore the functional characteristics of the bioMF OOC device, which include podocyte production and secretion of VEGF-A. The presence of soluble VEGF-A in the vascular channel along with the formation of fenestration in the endothelial cells indicated that the chips supported VEGF-A production by the podocytes and crosstalk between podocytes and endothelial cells to induce a specialized glomerular endothelial phenotype from a generic vascular endothelial cell population derived from undifferentiated hiPS cells. Thus, this bioMF OOC system could be used to study intercellular signaling across basement membranes in the kidney glomerular filtration barrier and in other organ models. In this study, a type of OOC device was engineered by integrating an ultrathin electrospun SF biomimetic basement membrane to interface two tissue layers (glomerular epithelium and vascular endothelium) in a dynamic microfluidic system. The resulting device supports the differentiation and propagation of hiPS cell-derived kidney glomerular cells, podocytes and vascular endothelial cells, to generate an isogenic human glomerulus-on-a-chip that exhibits in vivo-like cell structure, tissue organization, ECM remodeling, and molecular filtration functions. Intriguingly, it was observed that the SF membranes support cyclic mechanical stretching, suggesting the SF membranes would be amenable to physiological levels of mechanical strain, which could be a topic for follow-up studies, especially when applying this new organ-on-a-chip system to model lung function or inter-organ communication and crosstalk in the future. The utility of the device for modeling glomerular injury and proteinuria disease was also demonstrated. The porous structure of the electrospun SF membrane supported the integration of basement membrane proteins secreted by the surrounding podocytes and endothelial cells, yielding a physiologically relevant scaffold for prolonged tissue culture, differentiation, and maturation. The ultrathin (3.5 μm thick) membrane was amenable to microtome sectioning of the bioMF OOC devices for high resolution TEM analysis, which has been challenging to perform using PDMS- membrane-based OOC devices, as microtome blades cannot easily cut through the thick, elastic PDMS membranes embedded in resins. TEM analysis revealed that the glomerulus bioMF device supports formation of a tissue-tissue interface when podocytes and endothelial cells are cocultured on opposing side of the SF membrane, in which unspecialized vascular endothelial cells develop fenestrations that mimic the unique morphological and functional feature of glomerular endothelial cells in vivo. Future studies could employ this platform to gain insight into the mechanisms of fenestrae development and cell fate determination, including the roles of intracellular vesicles and their aggregation and fusion to form openings in the cell cytoplasm by examining endothelial morphology at different time points in the bioMF OOC device. The capability to recapitulate in vivo-like podocyte-endothelial crosstalk and fenestrae development will help elucidate the mechanism by which disruption of specific signaling pathways alters the function of tissue-specific endothelium and influences disease progression (e.g., drug-induced nephrotoxicity, viral infection, and pathogenic genetic mutations). Because the electrospinning methodology used here can produce membranes of different desired thicknesses, future work could involve modulation of SF membrane thickness to investigate its impact on cell physiology and tissue-tissue crosstalk, and to study the role of basement membrane thickness in kidney disease and organ failure. For example, there might be an optimum SF membrane thickness to achieve in vivo-like cellular responses in different tissues and organs. In addition, since glomerular basement membrane thickening is a hallmark of several glomerular disorders (e.g., membranous nephropathy, diabetic nephropathy, and focal segmental glomerulosclerosis), understanding how membrane thickness affects podocyte-endothelial cell crosstalk could improve the understanding of kidney disease mechanisms and help identify novel therapeutic targets. To demonstrate the functional capability of the bioMF OOC device, recapitulating the glomerular filtration barrier was focused on because of its characteristic three-layered structure and blood filtration function, which benefit from the structural organization of the chip and the ultrathin, porous SF membrane. The selective molecular filtration of albumin, inulin, and urea only after incorporating podocytes and endothelial cells in the device (not in acellular chips) indicates that the cell layers regulate the filtration rates of these molecules. This strategy for engineering the bioMF OOC device is broadly applicable to other tissues and organs, including brain, heart, and lung tissues via the propagation, differentiation, and maturation of neurons, cardiomyocytes, and lung epithelial stem cells, respectively. Using the layered chip design and SF membrane, various tissue types can be modeled in vitro by incorporating their corresponding cell types and by modulating the thickness of the SF membrane and its functionalization with tissue-specific or cell-adhesive ECM proteins. For example, the chips could be functionalized with user-defined matrices that support the co-culture of astrocytes, pericytes, and endothelial cells to recapitulate the structure and function of the blood-brain-barrier to model neurological disorders such as stroke. Building on this strategy, it may be possible to create networks of organ-specific chips toward the development of “body-on-a-chip” systems for systemic mechanistic studies of human physiology and pathophysiology and for the discovery and testing of therapeutic candidates. As demonstrated in this study, electrospun SF membranes provide an ultrathin, porous, biocompatible scaffold alternative to synthetic polymers such as PDMS and PC for OOC engineering. Still, SF membranes possess limitations that could be addressed in future studies. For example, the membranes have limited translucence, which prevents visualization of non- fluorescently labeled cells during live culture with standard microscopy. The use of reporter cell lines may be required for experiments that involve extensive live-cell imaging. As a result, the chips are better suited for end-point morphological characterization such as by immunofluorescence, electron microscopy, qPCR, western blot, and ELISA. However, the SF membranes were still sufficiently translucent to allow tracking of cell spreading, and it was noticed that when intermediate mesoderm cells were not properly attached and spread, they formed clumps, observed as large dark cell clusters under a phase contrast or dissecting microscope, and when the endothelial cells were not properly attached, they formed thick strings along the chip channel walls (FIG.14). The absence of these cell clusters and strings indicated well-spread cell cultures and formation of intact epithelial and endothelial tissue layers in the bioMF OOC device. Example 2 Organ-on-a-Chip Devices with Silk Fibroin Membranes for In Vivo-Like Tissue Modeling Organ-on-a-chip (OOC) platforms are promising for use in disease modeling and drug discovery as they can recapitulate human organ physiology by organizing cultured cells in fluidic channels to mimic in vivo-like tissue structure, patterning, and functions. Previously established OOC devices rely heavily on non-biological plastic materials to provide structural support between adjacent cell layers in the fluidic channels. Such materials yield a thick physical barrier that can limit transport of signaling molecules between cell layers, leading to suboptimal cell morphology and function. Here, a protocol for the microfabrication, processing, and application of a next- generation OOC system is described, which employs an ultra-thin, porous, naturally derived biomaterial membrane composed of silk fibroin to replace the conventionally used plastic materials. Electrospinning was applied to synthesize thin and porous silk fibroin membranes, which are subsequently integrated with polydimethylsiloxane-based microfluidic channels. The resulting silk fibroin membrane-based organ-on-a-chip (SF-OOC) device differentiates itself from existing devices by providing an in vivo-like tissue-tissue interface that supports specialized tissue ultrastructure development through intercellular crosstalk. As an example of application, the system was evaluated in a kidney glomerulus-on-a-chip system consisting of human vascular endothelial cells and podocytes—both derived from human induced pluripotent stem cells—on opposite sides of the silk fibroin membrane. This SF-OOC was used to model tissue formation, glomerular filtration, drug-induced disease, and development of tissue-specific endothelial fenestration. SF-OOC devices will enable modeling of various tissue-tissue interfaces across organs with in vivo-like performance and can be interconnected to model multi-organ interactions for use in “body-on-a-chip” applications. OOC devices are small microfluidic systems (about the size of a USB flash drive) commonly made of thermoplastics (e.g., polydimethylsiloxane (PDMS)), polycarbonate, and glass, which contain membranes for cell culture and synthetic fluidic channels for introduction of dynamic fluid flow. OOCs are designed to allow incorporation of cells along with manipulation of their microenvironment to provide in vivo-like tissue organization, fluid shear stress, and matrix mechanical properties. OOCs have been used to model human physiology and pathophysiology in various organs including the lungs, heart, liver, small intestine, and kidneys. Compared to conventional in vivo animal models whose developmental and molecular pathways have key differences from humans’, human cell-laden OOC devices allow for more accurate prediction of human biological responses. Incorporation of patient-derived cells into OOCs also paves the way for personalized medicine applications. Current OOCs typically employ one of the following designs: (1) a single channel, (2) horizontal / parallel channels, or (3) vertically stacked channels separated by a porous membrane. Single-channel devices have been widely used to study how matrix mechanical properties affect cell behavior under in vivo-like fluid shear stress, building on previous studies of cell mechanotransduction in response to fluid shear stress. Devices with horizontal / parallel channels commonly contain a hydrogel layer between the channels, which is useful for modeling cell transmigration (e.g., of immune cells) as well as molecular transport across the interstitium (e.g., kidney tubules). Devices with vertically stacked channels are the most widely used type of OOC due to the capability of the porous membrane to support cell adhesion on both sides, mimicking basement membranes in various tissues, including the blood brain barrier, the lung alveoli, and the glomerular filtration barrier in the kidney. The vertically stacked channel design recapitulates the spatial organization of these tissues while also allowing modeling of molecular exchange across the cell-laden barrier. The OOC device described in this protocol is a vertically stacked channel device. In vertically stacked channel OOCs, the porous membranes are commonly manufactured with PDMS, which provides an elastic and mechanically robust structural support for cell adhesion and propagation. The elastic nature of PDMS allows introduction of cyclic mechanical strain, for example, in capillaries modeled in glomerulus-on-a-chip devices, and to model breathing motion in lung-on-a-chip devices. The robust mechanical properties of PDMS provide resistance to fluid shear stress without rupturing, and the porosity of PDMS membranes supports transmembrane transport of molecules between fluid channels, facilitating studies of drug metabolism in intestine- on-a-chip devices and molecular filtration in glomerulus-on-a-chip devices. However, PDMS membranes cannot be made thinner than around 50 m, which is over 150-fold thicker than native basement membranes in vivo (~300 nm to 1 m). In addition, due to the non-degradable nature of PDMS, the membrane represents a permanent physical barrier between cell layers, restricting intercellular crosstalk between the cells and limiting development of advanced tissue patterns and morphogenesis within OOC systems. Additionally, validation of tissue ultrastructure and morphogenesis has been elusive due to the inability of PDMS membranes to support microtome sectioning necessary for high power tissue analysis. Attempts to reduce the thickness of PDMS membranes often result in drastically compromised mechanical properties, leading to rupture under fluid shear stress. Current techniques to manufacture porous PDMS membranes include soft molding combined with spin coating and reactive ion etching. Although soft molding on spin-coated PDMS membranes provides reliable replication of thin, porous membranes using pre-made stamping molds, the manufacturing process is limited to a small membrane area due to pre-polymer viscosity that tends to resist the applied pressure required for formation of through-pores. Moreover, reactive ion etching requires highly specialized gases, which are usually not readily available in research labs, and controlling the depth of etching has not been fully optimized compared to dry etching on silicon, resulting in low etch rates and incomplete pores. In addition, the flat surfaces of PDMS membranes lack extracellular matrix-like topological features that have been shown to modulate cell differentiation through mechanosensitive pathways. PDMS membranes also possess a much larger elastic modulus than native human organs, and the hydrophobic property of PDMS may lead to adsorption of hydrophobic drugs and altered pharmacokinetics / pharmacodynamics during drug screening and toxicity testing. As a result, an alternative membrane is desired to overcome these limitations of PDMS membranes for OOC applications. Silk fibroin (SF), a protein polymer naturally derived from the fibrin of silk produced by the silkworm Bombyx mori, is a promising biomaterial scaffold for tissue engineering. SF is biocompatible, non-immunogenic, and possesses desirable mechanical properties that promote maintenance of scaffold structural integrity. Additionally, SF supports versatile material fabrication and processing techniques, allowing for manufacturing into various material formats including films, hydrogels, sponges, microbeads, and electrospun fibers and mats. Among existing material fabrication techniques, electrospinning has gained interest for cell culture because it produces a highly porous and fibrous structure, mimicking the molecular and structural properties of native extracellular matrices. In addition, membranes generated by electrospinning can be as thin as basement membranes found in vivo, significantly thinner than achievable with PDMS membranes. In this work, a protocol to fabricate OOCs with ultra-thin, porous silk fibroin membranes was developed. Development of the Protocol This protocol describes the fabrication of a silk fibroin membrane-based organ-on-a-chip (SF-OOC) device. The first step in developing this protocol was to engineer a biomimetic electrospun SF membrane (FIG. 1A) that exhibits an extracellular matrix-like structure with pronounced fibrous and porous architecture (FIG. 6A). The resulting membrane exhibited an average nanofiber diameter of around 0.45 μm (FIG.15) and an average thickness of around 5 m, which is 10-fold thinner than widely used PDMS membranes (FIG.1D). The average Young’s modulus of the SF membranes was 67 kPa (FIG.7A), which is within the range for human kidneys (95.64 ± 9.39 kPa to 180.32 ± 11.11 kPa)49. Treatment of the resulting SF membranes with 90% methanol induced -sheet formation, rendering the membranes insoluble. The conformational change from random coil to -sheet was confirmed via Fourier-transform infrared spectrometry (FTIR) by the observation of shifts in peaks for amide-I, amide-II, and amide-III (FIG. 16). Following methanol treatment and drying overnight, the membranes were immersed in deionized water to remove polyethylene oxide (PEO), which was initially mixed into the SF solution to facilitate electrospinning. Successful removal of PEO was confirmed in FTIR by a diminished peak for the -C-O-C- bond (FIG.16). Two pieces of PDMS with engrafted fluid channels (bottom channel: 1 mm × 1 mm × 18 mm, top channel: 1 mm × 0.2 mm × 18 mm) were prepared by soft lithography (with an elastomer- to-curing agent ratio of 10:1), using 3D-printed molds designed with Autodesk Fusion 360. A lower elastomer-to-curing agent ratio (10:3) was subsequently used to prepare the prepolymer mixture, followed by spin coating to generate a thin layer of uncured PDMS which was used as glue to bond the membrane to the bottom of the PDMS piece (FIG.1A–B). The glue prevents leakage around the edges of the hydrophilic SF membrane and facilitates bonding to the top PDMS portion. The fluid channel was covered by the membrane (FIG. 1A). The bottom PDMS piece with the membrane was coated twice with PDMS glue using the top PDMS piece to transfer the glue, ensuring that the middle fluid channel area was not covered or blocked by the glue, which can disturb the SF-cell interface. The resulting assembly was treated with oxygen plasma to activate the silicon surface of PDMS and form covalent Si-O-Si bonds to form a seal (FIG.1A–B). The oxygen plasma treatment also served to sterilize the device; therefore, no additional sterilization steps were required. The water-tight seal was validated by introducing water into the channels to perfuse the device overnight, with no leakage observed. Although oxygen plasma treatment is widely used in other fields (e.g., electronic device engineering) to remove organic debris from silicon wafers, the treatment conditions in this protocol are significantly milder (50 W,60 s) than those used in electronics engineering applications ( 100 W, 60 s). The treatment alsointroduces -OH groups on the membrane surface, rendering the membranes hydrophobic, which facilitates adsorption of exogenously added laminin, as indicated by the presence of a pronounced -NH2 absorbance peak in the FTIR results (FIG.16). The incorporation of laminin was confirmed by the “hairy” morphology observed on electrospun nanofibers without a notable increase in nanofiber diameters (FIG. 6A and FIG. 15), as well as cell-membrane interactions (discussed below). The resulting device can be visualized under regular brightfield microscopy to examine membrane integrity after chip assembly (FIG. 17). To confirm that the device manufacturing process does not alter the morphology of the SF membranes, scanning electron microscopy (SEM) was performed to visualize the top and side of the membrane (FIG. 1B–F). The results indicated that the membrane remained fibrous, porous, and intact after device assembly. Demonstration of a Biomedical Application The SF-OOC device was applied in a kidney glomerulus-on-a-chip in which human vascular endothelial cells and podocytes derived from human induced pluripotent stem (iPS) cells were cultured on opposite sides of the silk fibroin membrane. This SF-OOC was used to model tissue formation, glomerular filtration, drug-induced disease, and tissue-specific endothelial fenestration. Human iPS cell differentiation techniques were utilized to recapitulate the kidney glomerulus, which in humans consists of a layer of highly specialized epithelial cells known as podocytes and a layer of vascular endothelial cells separated by a thin glomerular basement membrane. To recapitulate this tissue-tissue interface, human vascular endothelial cells and intermediate mesoderm cells (podocyte precursor cells) were differentiated from human iPS cells. The SF-OOC devices were assembled one day prior to cell seeding. Immediately after oxygen plasma treatment and assembly, the device was coated with laminin-511 to functionalize the membrane and promote cell adhesion, propagation, and differentiation (FIG. 1A). The human iPS cell-derived endothelial cells were propagated on the bottom of the membrane, while the intermediate mesoderm cells were propagated on the top side of the same membrane. The intermediate mesoderm cells were induced to differentiate into mature podocytes while interfaced with endothelial cells located on the opposite side of the membrane. Podocyte induction medium was delivered through the top channel of the device and endothelial cell culture medium was delivered through the bottom channel (FIG. 1A and FIG. 18). The resulting device showed confluent podocyte and endothelial cell coverage on the top and bottom of the membrane, respectively. Podocytes showed pronounced expression of the lineage markers nephrin and podocin, and endothelial cells expressed VE-cadherin (FIG.2A–B and FIG.9A). The device also demonstrated in vivo-like glomerular filtration function as indicated by a size-selective barrier which permitted transport of small molecules (inulin and urea) while blocking transport of a large molecule (albumin). A significantly larger amount of inulin and urea was filtered across the tissue barrier into the top (urinary) channel, while over 99.5% of albumin was retained in the bottom (vascular) channel (FIG. 2E). A closer examination of the size-selectivefiltration function using uremic toxin molecules [ -2-microglobumin (B2M) and visfatin] revealed agradually decreasing filtration rate consistent with increasing molecular weight (FIG.19). The cells propagated in the SF-OOC secreted and deposited native glomerular basement membrane proteins, including collagen IV, agrin, and nidogen-1 onto the SF membrane, indicating the potential of the membrane to be gradually remodeled by the cells. This paves the way for the use of the SF-OOC in modeling kidney diseases associated with pathological changes in the basement membrane (e.g., Alport syndrome and diabetic nephropathy). The glomerulus SF-OOC device also demonstrated the ability to model drug-induced nephrotoxicity. Perfusing a nephrotoxic chemotherapy drug, Adriamycin (doxorubicin), at clinically relevant dose for 3 days through the bottom (vascular) channel disrupted the cell morphology and compromised the size-selective filtration function (FIG. 3A–E). A closer examination of the cell morphology in the devices indicated that podocytes damaged by the drug treatment exhibited retracted foot processes, loss of podocyte-specific marker patterns, and loss of cells (FIG. 3A– C). In contrast, endothelial cells, which are in direct contact with the administered drug, did not seem to alter their morphology and cell-specific marker expression (FIG.3A–C). The results are in agreement with the previous observation that podocytes are significantly more sensitive to Adriamycin than endothelial cells in the in vivo. These results suggest that the SF-OOC device may be useful for drug testing and screening for nephrotoxicity during drug development. Comparison with Other Methods The SF membrane is made of a nature protein polymer and is significantly thinner than most membranes in existing OOC devices. PDMS membranes commonly used in OOC devices are at least 50 m thick, which is over 150-fold thicker than basement membranes in vivo. Such drastically larger membrane thickness can limit crosstalk between cells located on each side of the membrane. In the glomerulus SF-OOC described here, endothelial fenestrations were successfully induced in situ by podocytes located on the opposite side of the SF membrane, whereas mono-culturing the endothelial cells in the device results in non-fenestrated endothelial cells (FIG. 4G–H). Whether endothelial cells can develop such tissue-specific morphology in PDMS membrane-based OOC devices is yet to be determined as these devices are not amenable to microtome sectioning, which is required for ultrastructural analysis by transmission electron microscopy (TEM) and related histological analyses. The SF-OOC provides a robust platform for modeling in vivo-like tissue-specific morphologies and phenotypes, and the thinness of the membrane enhances compatibility for ultrastructural imaging. A recently published method describes the engineering of a thin membrane derived from collagen gels for use in OOCs. However, this study was focused mainly on illustrating how the membrane was engineered and incorporated into an OOC device, but did not explore in details regarding tissue-tissue interfaces, or integrity of the membrane for biomedical applications. In addition, the membrane exhibited a Young’s modulus of around 3–5 MPa, which is 30--50 times higher than that of in vivo basement membranes. While the device was able to support cell adhesion and culture of various cell types (e.g., Caco-2, A549, HUVEC, and NIH3T3 cells), this study could have investigated the device’s capacity to facilitate tissue-tissue interfaces, model disease phenotypes under injury stimuli, and assess cellular ultrastructure and intercellular crosstalk. In contrast, the protocol described herein involves comprehensive analysis of the SF- OOC barrier function under healthy and diseased conditions (FIG.2E and FIG.3D–E), while cell morphology is examined by confocal microscopy and TEM (FIG. 2A–B, 3A–C, and 4G–H). Additionally, this protocol validates the presence of advanced tissue-specific endothelial morphogenesis, leading to the induction of fenestrations (FIG. 4F–H), with VEGF signaling between podocytes and endothelial cells in the device (FIG.5B–C). Therefore, this protocol may be better suited for modeling the kidney and other organs under physiological or pathophysiological conditions. Future Applications SF has been shown to support the propagation and differentiation of numerous cell types from various organs, including neurons, cardiomyocytes, lung epithelial cells, and gastrointestinal epithelial cells. SF-OOC devices can therefore be used to model various organs by incorporating the corresponding cell types. For example, human iPS cell-derived endothelial cells can be cocultured with human iPS cell-derived astrocytes and pericytes to model the patient-specific blood-brain-barrier, which can be utilized to model pathological stimuli that result in a leaky barrier and to engineer drugs that can be delivered across the barrier for brain-targeted therapies. Another potential SF-OOC application is modeling the microenvironment of the small intestine, by incorporating gastrointestinal epithelial cells and endothelial cells into the device to model intestinal absorption, glucose transport, and how the microbial microenvironment contributes to homeostasis and disease in the small intestine. Liver sinusoids could also be modeled using the SF-OOC devices by incorporating hepatocytes, stellate cells, and endothelial cells. In this case, the SF membrane could be modulated by reducing the electrospinning duration to decrease or obtain the desired nanofiber density, allowing infiltration of stellate cells into the membrane and subsequent seeding of endothelial cells and hepatocytes on each side of the membrane. The resulting liver SF-OOCs could be used to model drug / metabolite metabolism, albumin secretion, and liver diseases such as metabolic dysfunction-associated steatohepatitis. Different OOCs could be connected to form multi-organ platforms. For example, blood-brain-barrier SF-OOCs could be connected with glomerulus SF-OOCs to model the inter-organ communication between the brain and the kidneys, which could shed light on the mechanisms of Alzheimer’s disease development and progression from a multiorgan perspective, as chronic kidney disease has been shown to be associated with a higher risk of Alzheimer’s disease in clinical studies. Also, liver SF-OOCs can be connected with small intestine SF-OOCs to model nutrient uptake and metabolism, as well as how diet and exposure to environmental factors contribute to metabolic diseases such as type 2 diabetes. Moreover, establishing various organ-specific SF-OOC devices could ultimately allow interconnection of these devices to model the human body-on-a- chip, for studying whole body drug metabolism, modeling cancer metastasis and diseases that involve multiple organs (e.g., metabolic syndromes and viral tropism), and for developing therapeutics to target these diseases. The kidney glomerulus SF-OOC described here can be expanded to study the role of intercellular crosstalk pathways between podocytes and glomerular endothelial cells in the development of glomerular endothelial fenestrations and glycocalyx. This device can also be used to investigate how glomerular endothelial fenestrations and podocyte foot processes change under pathophysiological conditions. As nephrotoxicity has been a major obstacle to new pharmaceutical development, the ability of the glomerulus SF-OOC to model nephrotoxic drug- induced injury indicates that this platform may be useful for drug screening. Improved drug screening using a more relevant human-cell kidney model could reduce the high failure rate of drug clinical trials (currently ~90%) resulting from predominant use of animal models or biologically and genetically unmatched models for pre-clinical studies aimed at treating human diseases. The glomerulus SF-OOC device could thus help overcome limitations of animal models, including ethical concerns and their insufficiency to accurately model and predict human physiological responses. Using this device, disease models could be built by introducing injury stimuli to the system (e.g., nephrotoxic drug treatment, hyperglycemia, hypertension, viral infection), and the disease models could subsequently be used for testing therapeutic candidates. The glomerulus SF-OOC device can also support incorporation of patient-specific human iPS cells to develop personalized devices that can be used to optimize treatment at different stages of a disease. Current therapeutic approaches for end-stage renal disease patients are limited to dialysis or organ transplantation, but neither of these approaches can fully and reliably compensate for the loss of kidney function. Dialysis is limited by insufficient removal of medium- to-large molecular weight uremic toxins and protein-bound toxins, leading to toxin buildup in patients and associated cardiovascular complications. Kidney transplantation has an extremely long waitlist and patient ineligibility due to risk-benefit determination and race-based calculations. Even if a patient does receive a kidney transplant, they must take immunosuppressive drugs for the rest of their life, and they still have elevated risk of infections, hypertension, and organ failure or rejection. A more effective treatment approach is needed. A SF-OOC kidney device for blood filtration could provide insights for the future development of an artificial kidney that removes uremic toxins more efficiently and reliably than traditional dialysis machines. MATERIALS Human iPS Cells PGP1 (The Personal Genome Project) human iPS cells were used. Growth factors and media supplements for human iPS cell-podocyte induction mTeSR™1 medium (Stem Cell Technologies, cat. no.85850) Advanced DMEM / F-12 (Thermo Fisher Scientific, cat. no.12634028) DMEM / F-12, GlutaMAX™ supplement (Thermo Fisher Scientific, cat. no.10565018) Complete Medium Kit with Serum and CultureBoost-R™ (Cell Systems, cat. no.4Z0-500- R) Human Activin A Recombinant Protein (Thermo Fisher Scientific, cat. no. PHC9564) Gibco™ Human BMP-7 Recombinant Protein (Thermo Fisher Scientific, cat. no. PHC9541) B-27™ Supplement, serum free (Thermo Fisher Scientific, cat. no.17504044) Stemolecule CHIR99021 (Stemgent, cat. no.04-0004) Gibco™ Human VEGF-165 Recombinant Protein (Thermo Fisher Scientific, cat. no. PHC9391) All-Trans Retinoic Acid (Stem Cell Technologies, cat. no.72264) Y-27632 dihydrochloride (Tocris Bioscience, cat. no.1254) Penicillin-Streptomycin, liquid (Thermo Fisher Scientific, cat. no.15140122) Growth factors and media supplements for human iPS cell-derived endothelial cell induction Neurobasal Medium (Thermo Fisher Scientific, cat. no.21103049) B27 Supplement, minus vitamin A (Thermo Fisher Scientific, cat. no.12587010) N-2 Supplement (Thermo Fisher Scientific, cat. no.17502048) Recombinant Human BMP-4, E. coli derived (Peprotech, cat. no.120-05ET-10UG) StemPro™-34 SFM (Thermo Fisher Scientific, cat. no.10639011) GlutaMAX™ Supplement (Thermo Fisher Scientific, cat. no.35050061) Forskolin, Adenylyl cyclase activator (Abcam, cat. no. ab120058) Heparin Solution (Stem Cell Technology, cat. no.7980) Penicillin-Streptomycin, liquid (Thermo Fisher Scientific, cat. no.15140122) ECM molecules Corning® Matrigel® hESC-Qualified Matrix (VWR, cat. no. BD354277) This reagent shows lot-to-lot variation. Lot testing for successful human iPS cell culture is recommended. Keep Matrigel cold and use a manufacturer-recommended dilution factor for human iPS cell culture. iMatrix-511, recombinant laminin-511 E8 (Takara Bio, cat. no. T304) Human recombinant Laminin 511 protein (Biolamina, cat. no. LN511-0502) Human Fibronectin (Advanced Biomatrix, cat. no.5050) Antibodies for immunocytochemistry and western blot Nephrin (American Research Products, cat. no.03-GP-N2) VE-Cadherin (Santa Cruz Biotech, cat. no. sc-9989) Podocin (Abcam, cat. no. ab50339) Collagen-IV (Thermo Fisher Scientific, cat. no.14-9871-82) Nidogen-1 (Thermo Fisher Scientific, cat. no. MA5-23911) Agrin (Thermo Fisher Scientific, cat. no. PA5-103585) VEGF (Bio-techne, cat. no. AF-293-SP) AF488 Goat anti-guinea pig (Thermo Fisher Scientific, cat. no. A11073) AF594 Donkey anti-mouse (Thermo Fisher Scientific, cat. no. A21203) AF488 Donkey anti-mouse (Thermo Fisher Scientific, cat. no. A21202) AF594 Donkey anti-rabbit (Thermo Fisher Scientific, cat. no. A21207) AF488 Donkey anti-rabbit (Thermo Fisher Scientific, cat. no. A21206) AF488 Donkey anti-goat (Thermo Fisher Scientific, cat. no. A11055) DAPI Nucleic Acid Stains (Thermo Fisher Scientific, cat. no. D1306) Anti-mouse IgG, HRP-linked Antibody (Cell Signaling Technology, cat. no.7076) Anti-rabbit IgG, HRP-linked Antibody (Cell Signaling Technology, cat. no.7074) Imaging / Sample Preparation Reagents Formaldehyde (Sigma-Aldrich, cat. no.8.18708) Glutaraldehyde (Sigma-Aldrich, cat. no. G7651) Phosphate buffered saline (Thermo Fisher Scientific, cat. no.14200075) Osmium tetroxide (Electron Microscopy Science, cat. no.19180) Uranyl acetate (Electron Microscopy Science, cat. no.22400) ERL 4221 (Electron Microscopy Science, cat. no.15004) NSA (Electron Microscopy Science, cat. no.19050) DER 736 (Electron Microscopy Science, cat. no.13000) DMAE (Electron Microscopy Science, cat. no.13300) Molecular Filtration Assay and ELISA Albumin from Bovine Serum (BSA), Texas Red™ conjugate (Thermo Fisher Scientific, cat. no. A23017) Inulin-FITC (Sigma-Aldrich, cat. no. F3272) Urea (Sigma-Aldrich, cat. no. U5378) Urea Assay Kit (Abcam, cat. no. ab83362) -Microglobulin, Human Urine (Sigma-Aldrich, cat. no.475823- M) Human beta 2 Microglobulin ELISA Kit (Abcam, cat. no. ab99977) Recombinant Human Visfatin (PeproTech, cat. no.139-09) Human Visfatin ELISA Kit (Abcam, cat. no. ab264623) Human VEGF-A ELISA Kit (RayBiotech, cat. no. ELH-VEGF-2) Western Blot Analysis Pierce™ BCA Protein Assay Kits (Thermo Fisher Scientific, cat. no.23225) 4x Laemmli Sample Buffer (Bio-Rad, cat. no.1610747) 10x Tris / Glycine / SDS (Bio-Rad, cat. no.1610732) TWEEN® 20, PROTEIN GRADE® Detergent, 10% Solution, Sterile Filtered (Sigma- Aldrich, cat. no.655206) Mini-PROTEAN® Cassette Opening Lever (Bio-Rad, cat. no.4560000) cOmplete™, Mini, EDTA-free Protease Inhibitor Cocktail (Sigma-Aldrich, cat. no. 4693159001) PhosSTOP™ (Sigma-Aldrich, cat. no.4906845001) Restore™ Western Blot Stripping Buffer (Thermo Fisher Scientific, cat. no.21059) SuperSignal™ West Femto Maximum Sensitivity Substrate (Thermo Fisher Scientific, cat. no.34094) Enzymes and Other Reagents Mycoplasma PCR Detection Kit (Applied Biological Materials, cat. no. G238) Cell Dissociation Buffer, enzyme-free, Hanks' Balanced Salt Solution (Thermo Fisher Scientific, cat. no.13150016) Accutase (Stem Cell Technology, cat. no.07920) Trypsin-EDTA (Thermo Fisher Scientific, cat. no.25200056) Fetal Bovine Serum (Corning, cat. no.35-011-CV) Paraformaldehyde Solution (Thermo Fisher Scientific, cat. no.28908) DPBS, no calcium, no magnesium (Thermo Fisher Scientific, cat. no.14190144) DPBS, calcium, magnesium (Thermo Fisher Scientific, cat. no.14040133) Triton X-100, 1% (Thermo Fisher Scientific, cat. no. HFH10) Bovine Serum Albumin (Sigma-Aldrich, cat. no.9048) Dimethyl sulfoxide (Sigma-Aldrich, cat. no. D2438) Cell Counting Kit - 8 (Sigma-Aldrich, cat. no.96992) Sudan Black B (Sigma-Aldrich, cat. no.199664) Distilled Water (Thermo Fisher Scientific, cat. no.15230204) Adriamycin (LC Laboratories, cat. no. D-4000) CD144 MicroBeads, human (Miltenyi Biotec, cat. no.130-097-857) CD31 MicroBead Kit, human (Miltenyi Biotec, cat. no.130-091-935) OCC Device Equipment Custom-designed 3D PDMS mold (FIG.20–21) Sylgard 184 silicon encapsulant clear kit (Ellsworth, cat. no.4019862) Sterile blunt needles, 18-G (SAI infusion technology, cat. no. B18-100) Miltex Premium no.3 Scalpel (Integra life science, cat. no.2970018589) 205TC Balance (VWR, cat. no.76446-796) Corning™ 10015mm Polystyrene Petri Dishes (Thermo Fisher Scientific, cat. no.07-202- 011) Scienceware® vacuum desiccator (Sigma-Aldrich, cat. no. Z119016) Nalgene™ Non-Phthalate PVC Vacuum Tubing (Thermo Fisher Scientific, cat. no.76403- 596) Heratherm™ General Protocol Ovens (Thermo Fisher Scientific, cat. no.51028112) Fisherbrand™ Premium Plain Glass Microscope Slides (Thermo Fisher Scientific, cat. no. 12-544-4) Carbon Steel Sterile Surgical Blades, no.11 (Integra life science, cat. no.4-111) Feather™ Single-Use Scalpels (Thermo Fisher Scientific, cat. no.08-927-5B) Disposable Spatulas (VWR, cat. no.80081-188) Silk fibroin (Advanced Biomatrix, cat. no.5154) Poly (ethylene glycol) (Sigma-Aldrich, cat. no. P6667) Slide-A-Lyzer™ Dialysis Cassettes, 3.5K MWCO (Thermo Fisher Scientific, cat. no. 66380) BD Luer-Lok™ Disposable Syringes without Needles, 10 mL (BD, cat. no.14-823-2A) BD PrecisionGlide™ needles, 11 / 2 in, 18G (BD, cat. no.305196) Slide-A-Lyzer™ Accessory Float Buoys and Syringes (Thermo Fisher Scientific, cat. no. 66430) Poly (ethylene oxide) (Sigma-Aldrich, cat. no.189456) KIMBLE® KIMAX® Griffin Beakers (100 mL; cat. no.14000-100) Aldrich® stirring rods (Sigma, cat. no. Z549768) 16-13-G blunt-tip needle (McMASTER-CARR, cat. no.75165A753) Fusion 101A Syringe Pump (Chemyx, cat. no. F10071) High voltage line operated power source (Gamma High Voltage Research, cat. no. E8100P-20W) Premium Aluminum Foil (VWR, cat. no.89107-724) Methanol (Thermo Fisher Scientific, cat. no.176840010) Petri Dishes, Glass (VWR, cat. no.75845-544) Bemis™ Parafilm™ M Laboratory Wrapping Film (Thermo Fisher Scientific, cat. no. PM996) Spin Coat Hood (Air control) Spin Coater (Headway) P type test silicon wafers (University Wafer) Emitech K-1050X Plasma Asher (Emitech) Ismatec IPC Digital Peristaltic Pump, 0.4 to 45 rpm, 16-Channel, Click-'N-Go Cartridge; 115 VAC (Cole-Parmer, cat. no. EW-78001-32) BPT 2-stop tubing (VWR, cat. no. MFLX95723-12) Sterile Blunt Needles, 19-G (SAI technology, cat. no. B19-50) Disposable Beakers (Sigma-Aldrich, cat. no. Z245399) 3D printed chip racks (FIG.22) Human Emulation System Chip Rack (Emulate, Inc) Self-Sealing Sterilization Pouch (VWR, cat. no.89140-802) Tissue Culture Supplies Cell Lifter (Corning, cat. no.3008) 6-wells Cell Culture Plates (VWR, cat. no.10062-892) 12-wells Cell Culture Plates (VWR, cat. no.10062-894) Corning® 75 cm² U-shaped cell culture flask, canted neck (Sigma-Aldrich, cat. no. CLS430641U) Corning® bottle-top vacuum filter system (Sigma-Aldrich, cat. no. CLS430758) Corning® Vacuum Filter / Storage Systems, Sterile, Corning (VWR, cat. no.28199-764) Millipore® Steriflip® Vacuum Tube Top Filter (Sigma-Aldrich, cat. no. SCGP00525) VWR® High-Performance Centrifuge Tubes with Flat or Plug Caps, 15 mL (VWR, cat. no. 89039-666) VWR® High-Performance Centrifuge Tubes with Flat or Plug Caps, 50 mL (VWR, cat. no. 89039-656) Falcon™ Polystyrene Aspirating Pipets, Without Graduations (Thermo Fisher Scientific, cat. no.13-675-10CC) Nunc™ Biobanking and Cell Culture Cryogenic Tubes (Thermo Fisher Scientific, cat. no. 14-666-318) Fisherbrand™ Microcentrifuge Tubes with Locking Snap Cap (Thermo Fisher Scientific, cat. no.377267) Mr. Frosty™ Freezing Container (Thermo Fisher Scientific, cat. no.5100-0001) P10 barrier pipette tips (VWR, cat. no.76322-528) P20 barrier pipette tips (VWR, cat. no.76322-134) P100 barrier pipette tips (VWR, cat. no.76322-136) P200 barrier pipette tips (VWR, cat. no.76322-150) P1000 barrier pipette tips (VWR, cat. no.76322-154) Disposable Serological Pipettes, 5 mL (VWR, cat. no.76201-710) Disposable Serological Pipettes, 10 mL (VWR, cat. no.75816-100) Disposable Serological Pipettes, 25 mL (VWR, cat. no.75816-090) Disposable Serological Pipettes, 50 mL (VWR, cat. no.75816-088) KIMWIPES™ delicate task wipers, small (VWR, cat. no.21905-011) KIMWIPES™ delicate task wipers, big (VWR, cat. no.21903-005) Basic Plus Lab Coats, Kimberly-Clark (VWR, cat. no.37000-928) Ultraform® Powder-Free Nitrile Exam Gloves, Microflex®, small (VWR, cat. no. 89235- 578) Ultraform® Powder-Free Nitrile Exam Gloves, Microflex®, medium (VWR, cat. no.89235- 580) Ultraform® Powder-Free Nitrile Exam Gloves, Microflex®, large (VWR, cat. no. 89235- 582) Ethanol absolute, KOPTEC, meets analytical specification of BP, Ph. Eur., USP (VWR, cat. no.89125-172) Tissue Culture Equipment Ethanol absolute, KOPTEC, meets analytical specification of BP, Ph. Eur., USP (VWR, cat. no.89125-172) Biosafety cabinet, Series 1300, Type A2 (Thermo Fisher Scientific, cat. no.1333) CO2 incubator, Forma Steri-Cycle CO2 Incubator (Thermo Fisher Scientific, cat. no. 201370) Light microscope (Nikon, model no. EcliPSeTS100-F, equipped with a Zeiss, model no. AxioCam MRc 5) Hemacytometer chamber with cover glass (VWR, cat. no.15170-172) Avanti J-15 benchtop centrifuges (Beckman Coulter, cat. no. C01994) Micro centrifuge 20R (Beckman Coulter, cat. no. B31608) Countess 3 Automated Cell Counter (Thermo Fisher Scientific) Imaging Equipment Leica SP8 upright confocal system (Leica) Apreo™ 2 SEM for Materials Science (Thermo Fisher Scientific) FEI Tecnai G2 Spirit Twin TEM (FEI) EVOS™ M7000 Imaging System (Thermo Fisher Scientific, cat. no. AMF7000) ChemiDoc MP Imaging System (Biorad, cat. no.12003154) Fiji ImageJ (version 1.53f51) GraphPad Prism software (version 9.5.1.) Other Equipment Micro-Strain analyzer (TA instruments) Nicolet™ 8700 FT-IR spectrometers (Thermo Fisher Scientific) Corning® 96 Well Special Optics Microplate (Corning, cat. no. CLS3614) SpectraMax iD3 and iD5 Multi-Mode Microplate Readers (Molecular device) Nunc™ MicroWell™ 96-Well, Nunclon Delta-Treated, Flat-Bottom Microplate (Thermo Fisher Scientific, cat. no.168055) QuadroMACS™ Separator (Miltenyi Biotec, cat. no.130-042-302) Desk V sputter coater (Denton) Precision Tweezers (VWR, cat. no.89259-986) Mini-PROTEAN® Tetra Vertical Electrophoresis Cell for Mini Precast Gels (Bio-Rad, cat. no.1658004) Bolt™ Bis-Tris Plus Mini Protein Gels, 4-12%, 1.0 mm, WedgeWell™ format (Thermo Fisher Scientific, cat. no. NW04120BOX) PowerPac Universal Power Supply (Bio-Rad, cat. no.1645070) Standard Dry Block Heaters (VWR, cat. no.75838-318) Trans-Blot Turbo Midi 0.2 μm PVDF Transfer Packs (Bio-Rad, cat. no.1704157) Trans-Blot® Turbo™ Transfer System (Bio-Rad, cat. no.1704150) Reagent Preparation Human Pluripotent Stem Cells To achieve optimum differentiation results, human iPS cells should be expanded on feeder-free culture system with mTeSR1 medium in Matrigel-Coated plate for 4 days after passing or around 70% confluence. Matrigel-Coated Plates Matrigel is to be stored at –20 °C to avoid crosslinking before use. Dilute the appropriate amount of Matrigel in 25 mL of cold DMEM / F12 without directly mixing the concentrated Matrigel solution. All cultures for human iPS cells were carried out on 6-well plates. For each well, add 1 mL of diluted Matrigel solution and incubate for 2 h at 37 °C or 24 h at 4 °C. The coated plate can be used immediately or stored at 4 °C for up to 2 weeks. For the Matrigel plate stored at 4 °C, 30 min of incubation at 37 °C is recommended before seeding human iPS cells. Laminin 511-E8-Coated Plates All 2D differentiation steps were carried out in 12-well plates. Laminin 511-E8 coating solution was prepared by diluting Laminin 511-E8 protein solution with ultrapure water in a 1:100 ratio to achieve a final concentration of 5ug / mL. For each well, add 700 μL of the coating solution and incubate at room temperature for 2 h or 4 °C overnight. The Laminin 511-E8-coated plate can be stored for up to 1 week at 4 °C. For the functionalized plate stored at 4 °C, 30 min of incubation at room temperature is recommended before seeding cells. Fibronectin-Coated Plates Plate-based differentiation steps for endothelial cells were carried out in 6-well plates. Fibronectin solution was prepared by diluting human fibronectin recombinant protein solution with ultrapure water in 1:20 ratio to achieve a final concentration of 0.025 g / mL. For each well, add 1000 μL of the coating solution and incubate at room temperature for 2 h or 4 °C overnight. The fibronectin-coated plate can be stored at 4 °C for up to 1 week. The plate should be brought to room temperature (25 °C) 30 min before use to equilibrium the temperature. Activin A Human recombinant activin A was reconstituted in sterile PBS with 0.1% (wt / vol) BSA to a final concentration of 100 g / mL. Aliquot in 100 μL and store in –20 °C for up to 6 months before use. BMP7 Human recombinant BMP7 was reconstituted in sterile PBS with 0.1% (wt / vol) BSA to a final concentration of 100 g / mL. Aliquot in 100 μL and store in –20 °C for up to 6 months before use. VEGF Human recombinant VEGF was reconstituted in sterile PBS with 0.1% (wt / vol) BSA to a final concentration of 100 g / mL. Aliquot in 100 μL and store in –20 °C for up to 6 months before use. CHIR99021 CHIR99021 was reconstituted by dissolving 2 mg of CHIR99021 in 143.4 μL of sterile DMSO to a final concentration of 30 mM. Aliquot in 5 μL and store in –20 °C for up to 1 month before use. Y27632 Y-27632 dihydrochloride was reconstituted by dissolving 10 mg of Y27632 in 3.079 mL of sterile distilled water. Aliquot in 100 μL and store in –20 °C for up to 6 months before use. All-trans Retinoic Acid All-trans retinoic acid was reconstituted by dissolving 10mg of all-trans retinoic acid in 3.33 mL of sterile DMSO. Aliquot in 500 μL and store in –20 °C for up to 6 months before use. BMP4 Human recombinant BMP7 was reconstituted in 10 mM acetic acid to a final concentration of 2.5 g / mL. Aliquot in 100 μL and store in –20 °C for up to 6 months before use. Forskolin Forskolin was reconstituted by dissolving 10 g of forskolin in 2.436 mL of sterile warm DMSO to a final concentration of 4.105 g / mL. Aliquot in 100 μL and store in –20 °C for up to 6 months before use. Defined mTesR1 medium Mix 400 mL mTeSR1 Basal Medium with 100 mL of thawed mTeSR15x Supplement to make 500 mL stock. Aliquot in 50 mL and store in –20 °C for up to 6 months. Thaw the aliquot at 4 °C overnight before use, and thawed aliquot can be stored for up to 2 weeks. The plate should be brought to room temperature (25 °C) 30 min before use to equilibrium the temperature. Complete Medium with CultureBoost-R Add CultureBoost-R supplement to 500 mL basal medium per manufacturer’s guidelines. Store at 4 °C up to two weeks before use. Mesoderm Differentiation Medium Before initiating the differentiation protocol for Mesoderm from human iPS cells, prepare fresh mesoderm differentiation medium by mixing DMEM / 12 with GlutaMax, 100 ng / mL activin A, 3 M CHIR99021, 10 M Y27632, 1× B27 serum-free supplement, and 1% (vol / vol) penicillin– streptomycin. Prepare the medium according to the scale of the experiment. Intermediate Mesoderm Differentiation Medium Prepare fresh intermediate mesoderm differentiation medium by mixing DMEM / 12 with GlutaMax, 100 ng / mL BMP7, 3 M CHIR99021, 1× B27 serum-free supplement, and 1% (vol / vol) penicillin–streptomycin. Prepare the medium according to the scale of the experiment. Newly prepared intermediate mesoderm media can be stored at –20 °C for up to 3 months. Frozen aliquot can be thawed overnight at 4 °C before use. Podocyte Induction Medium Prepare fresh podocyte induction medium by mixing DMEM / 12 with GlutaMax 100 ng / mL BMP7, 100 ng / mL activin A, 50 ng / mL VEGF, 3 M CHIR99021, 1× B27 serum-free supplement, 0.1 M all-trans retinoic acid, and 1% (vol / vol) penicillin–streptomycin. Prepare the medium according to the scale of the experiment. Newly made intermediate mesoderm media can be stored at –20 °C for up to 3 months. Frozen aliquot can be thawed overnight at 4 °C before use. N2B27 Medium Before initiating the differentiation protocol for lateral mesoderm from human iPS cells, prepare 1031 mL of fresh N2B27 medium by mixing 500 mL of neurobasal media, 500 mL DMEM / 12 with GlutaMax, 20 mL of B27 (-) Vitamin A, 10 mL N2 (100x) and 1 mL of - mercaptoethanol. Prepare the medium according to the scale of the experiment and adjust the amount for each ingredient based on the same ratio. Lateral Mesoderm Differentiation Medium Before initiating the differentiation protocol for lateral mesoderm from human iPS cells, prepare fresh lateral mesoderm differentiation medium by mixing N2B27 medium, 8 M CHIR99021, 25 ng / mL hBMP4, and 1% (vol / vol) penicillin–streptomycin. Prepare the medium according to the scale of the experiment. StemPro-34 SFM Medium Prepare fresh StemPro-34 SFM medium by mixing StemPro-34 basal medium, StemPro- 34 supplement, 1% (vol / vol) penicillin–streptomycin, and 1% (vol / vol) Glutamax. Prepare the medium according to the scale of the experiment. Freshly made StemPro-34 media can be stored at –20 °C for up to 3 months. Frozen aliquot can be thawed overnight at 4 °C before use. Endothelial Cell Induction Medium Before initiating the endothelial cell induction protocol, prepare fresh endothelial cell induction medium by mixing StemPro-34 SFM medium, 200 ng / mL VEGF-165, 2 M Forskolin and 1% (vol / vol) penicillin–streptomycin. Prepare the medium according to the scale of the experiment. Newly made endothelial cell induction media can be stored at –20 °C for up to 3 months. Frozen aliquot can be thawed overnight at 4 °C before use. Trypsin-Neutralizing Solution Fresh Trypsin-neutralizing solutions are to be made every time trypsin EDTA is used with DMEM / F12 and 10% (vol / vol) FBS. Permeabilization Buffer Permeabilization buffer is made with 0.125% (vol / vol) Triton X-100 in PBS. This solution can be prepared and stored at room temperature for up to 1 month. Spurr Resin Prepare 40 mL of fresh Spurr resin for TEM sample embedding by mixing 16.4 g of ERL4221, 23.6 g of NSA, 5.72 g of DER 736, and 0.4 mL of DMAE in a plastic cup. Prepare the resin quantity to the scale of the experiment and adjust accordingly using the same ratio. RIPA Buffer RIPA cell lysis buffers are prepared by supplementing RIPA buffer with phosphatase inhibitor cocktail tablets and protease inhibitor cocktail tablets. The RIPA buffer should be made and used on ice, prepared RIPA cell lysis buffer can be stored in 4 °C for up to 6 months. 90% Methanol Prepare 90% Methanol by adding 90 mL of 100% methanol in a clean glass beaker in the fume hood with 10 mL of ultrapure water. Mix well using a 50 mL serological pipette before use. 8% SF Solution 10% PEG solution can be prepared by adding 100 g of Poly (ethylene glycol) to 900 mL of Milli-Q water. Allow the PEG to fully dissolve by sterling the solution on a stir plate at 250 rpm for 30 min. Soak the 3 mL dialysis cassettes in Milli-Q water for 1 min / side before use. 5% SF solution from Advance Biomatrix were first thawed at 4 °C for 20 h or until fully thawed. On the second day prepare the 10% PEG solution by diluting 100 g of Poly (ethylene glycol) in 900 mL of ultra-pure water. Use a 10 mL syringe to very slowly transfer 10 mL of the 5% SF solution into the syringe. Invert the syringe and attach a 27-G 1’’ thin needle on to the syringe and insert the needle into one corner of the cassette until the tip of the needle is inside the transparent membrane area, avoid inserting too deep or breaking the cassette membrane. Inject the solution very slowly into the cassette until it’s 70% filled. Remove the air bubble inside the cassette by inserting another 27-G thin needle without a syringe attached into the opposite corner to allow venting. Continue pushing the syringe until no visible air bubble in the cassette before removing the top empty needle. Inject all remaining SF solution into the cassette before removing the needle and the syringe. Attach the cassette to a buoy and transfer it into the 10% PEG solution for 20 h dialysis. On the second day, take out the cassette with the concentrated SF solution in it. A 10 mL syringe with a 27-G, 1’’ thin needle is used to transfer all the concentrated SF solution into a 15 mL conical tube. Re-dilute the SF solution to 8% with Milli-Q water to a total volume of 6.25 mL. Prepare 10% PEO solution by adding 1 g of Poly (ethylene oxide) to 9 mL of ultrapure water. Use a glass stir bar to thoroughly mix the solution until the PEO powder is completely dissolved and no clumps are visible. Add 10% PEO solution to 8% SF solution in 3.9:1 ratio, mix gently using a glass rod until no PEO clumps are visible in the mixture. Aliquot the SF / PEO solution into centrifuge tubes using 16-G needle and store in –80 °C for future use. Procedure Generation of Electrospun SF Membranes (~4 d) Frozen SF / PEO solution should be thawed overnight in 4 °C before use. Using a 10 mL syringe without a needle, carefully transfer 1 mL of SF / PEO solution into the syringe. Too much shear stress by transferring the liquid too fast will result in SF beta- sheet forming. Remove all air bubbles in the syringe by inverting the syringe and pushing out all air. The presence of air bubbles in the SF solution will result in droplets forming during electrospinning; this will cause defects on the spun membrane. Once all air is removed, attach an 18-G needle to the syringe and push until a tiny droplet of SF / PEO solution is visible at the needle tip. Load the syringe onto the syringe pump. Troubleshooting (Table 1). Cut out a 10 × 10 cm square nonstick foil, flatting the foil by pressing the foil with a piece of KimWipe against a flat bench surface. Use a double-sided tape to tape the flattened foil onto the metal plate. Set the distance between the metal plate and the tip of the needle to 15 cm. Connect the needle to the 20 W power source using a wire. Ground the metal plate using a wire. Turn on the pump, select the correct type of syringe, set the total volume to 1 mL, and set the speed to 0.015 mL / min. Wait until a tiny droplet is visible on the tip of the syringe. Turn on the power source and increase the voltage to 9.5 kV. Note: The voltage can be slightly adjusted based on the fiber condition, but the recommended range is between 9 kV and 10 kV. Continue to observe the fiber formation and membrane thickness during electrospinning. After 10 min, turn off the power source and pause the pump, remove the attached foil from the metal plate for further processing of the membrane. Troubleshooting (Table 1). Cut the membrane area into 0.5 cm by 3 cm strips using sharp scissors; avoid any area with SF / PEO solution droplets and avoid areas toward the edge of the circular e-spun membrane since the thickness of the membrane will drastically decrease toward the edge. Transfer all membrane strips to a clean glass petri dish, submerge all membrane in 90% methanol (see reagent preparation) in a chemical hood for 20 min to crosslink the SF. Avoid overlaying of the membrane. Since 90% methanol will evaporate rapidly, use a lid to cover the petri dish to avoid membrane drying. Remove the methanol from the petri-dish, let the membrane air dry overnight. Pause Point: The dried methanol-treated SF membranes can be stored at room temperature (25 °C) for up to 1 year. On the second day, submerge all membrane under DI water to remove all the PEO, parafilm the petri dish to avoid evaporation. Remove DI water after 2 d and let the membrane air dry for later use. Generation of PDMS Chips (~5 h) Add Sylgard 184 PDMS elastomer and curing agent to a clean plastic dish at 10:1 ratio. 4 g of PDMS and 0.4 g of curing agent is required for each SF-OOC device. Mix the elastomer / curing agent mixture thoroughly using a plastic spatula. Note: uneven mixture will result in inadequate curing. Remove all air bubbles in the PDMS mixture using a vacuum desiccator for at least 30 min. Clean the PDMS molds (FIG.20–21) by washing with DI water and spray with 70% ethanol. Air dries the mold before use. Pour the PDMS mixture onto the top and the bottom PDMS piece mold slowly, making sure the mold is fully covered by PDMS. Use a glass slide to remove all the excess PDMS outside of the mold area by sweeping the glass slide along the surface of the mold. Remove all visible air bubbles by poking the air bubble with a surgical blade. Note: Any air bubble would result in an empty air pocket inside the PDMS chip and compromise its structural integrity. Transfer the mold with PDMS to a general protocol oven preheated to 65 °C. Bake for 4 h for the PDMS to fully cure. Remove the mold from the oven and demold the PDMS chip. Use a surgical blade to cut along the edges of the PDMS mold. For the top PDMS piece, use a tweezer to dig out one end of the PDMS with the handle, pull the handle vertically from the mold while pushing down the other end of the PDMS chip to avoid breaking the pole in the mold. Demold the final portion of the chip by slightly wiggling the chip left and right until it can be pulled out vertically. For the bottom PDMS piece, use a tweezer to dig out one corner of the chip and peel it off from the mold. Trim off the excess PDMS on Chip and remove the handle. Use an 18-G needle, punching it through the SF-OOC channel pores on the top PDMS piece to make sure there are no PDMS left in the channel pores. Note: any PDMS remaining in the channel could block media flow into the SF-OOC device Fabrication of Bottom SF / PDMS Pieces (~5 d) Make the PDMS glue following the procedures below: Add Sylgard 184 PDMS with its curing agent to a clean plastic dish in a 10:3 ratio. 1 g of PDMS and 0.3 g of curing agent are required for coating ~8 bottom PDMS pieces. Mix the PDMS and crosslinker mixture thoroughly using a plastic spatula. Load the plastic dish on the Spin Coater, and make sure the spin coater is aligned and locked in. Set the speed to 3500 rpm, time to 600 s, and ramp to 500 on the Spin coater and start the program. Peel the SF membrane from the foil using a tweezer. Troubleshooting (Table 1). Dip the bottom PDMS piece channel side facing down into the PDMS glue, make sure the whole surface of the bottom PDMS piece is covered with the glue. Place the bottom PDMS piece channel side facing up on the bench, and carefully transfer the peeled-off membrane onto the bottom PDMS piece channel. Note: Make sure the full channel area is covered completely by SF membrane, make sure there are no wrinkles or cracks in the SF membrane during transfer. Cut the excess membrane from the bottom PDMS piece by trimming the membrane at 1 / 2 at the inlet / outlet channel using a blade. Allow the PDMS glue to cure overnight. Troubleshooting (Table 1). On day 2, make a new batch of PDMS glue following Step 11. Dip the top PDMS piece in the glue channel side facing down, making sure the whole chip surface is covered with PDMS glue. Flip the top PDMS piece to channel side facing down, align the top and bottom channel carefully and bind the top channel to bottom channel. Note: any misalignment will cause PDMS glue to leak into the channel area and block the channel to prevent normal liquid flow. Flip the chip, lightly press on the bottom PDMS piece at area outside of the channel to make sure that PDMS glue infiltrates into all areas outside of the channel itself. Peel off the bottom PDMS piece from the top PDMS piece by lifting it up on one end and let the top PDMS piece naturally drop off by gravity. Put the coated bottom PDMS piece face up, let the PDMS glue cure overnight. On day 3, repeat the bottom PDMS piece coating process in Step 14 to double coat the bottom PDMS piece. Wait for all PDMS glue to cure for another 2 days at room temperature before the chips are ready to assemble. Quality Control Check Point: Place the bottom chip under a microscope to observe the microscopic structure and identify any potential defects. If cracks are observed in the membrane, it has been damaged and should be discarded. If visible tiny circular areas on the membrane with less silk fibroin can be observed, this is likely due to improper electrospinning, and the membrane should be used as a substitute. Pause Point: The resulting SF / PDMS chips can be stored at room temperature (25 °C) for up to 1 year before assembling into the final SF-OOC devices. Troubleshooting (Table 1). Assembling the SF-OOC Device / Laminin Coating (~2.5 h) Transfer 10 mL of sterile 1×PBS (+Ca2+, +Mg2+) in a 15 mL conical tube and put on ice. Transfer the frozen human recombinant Laminin 511 protein on ice. Take an additional empty 15 mL conical tube for mixing laminin solution on ice. Bring the Ice bucket with all three items into a biosafety cabinet. Wash all top PDMS pieces with 70% ethanol and allow it to air dry before transferring all to a sterile petri dish. Spray the entire surface of a large square sterile tissue culture plate before bringing it into the biosafety cabinet. Spray a p-200 pipette and a new box of p200 barrier tips into the biosafety cabinet, along with a tweezer to help handling the chips. Set the power of the plasma asher to 50 W and set the timer to 1 min. Adjust the vacuum blob to make sure the vacuum purge level sits at 0. Transfer 2 top and 2 bottom PDMS pieces into the plasma asher with channel side facing up. Make sure the door for the plasma Asher chamber is firmly closed. Turn on the vacuum and wait for the pressure to go down to 0.5 before starting the program by pressing start. Calibrate the plasma Asher by adjusting the Gas1 blob to 50. Troubleshooting (Table 1). When the program is finished, turn off the vacuum and adjust the purge blob to 10 to vent the plasma asher chamber. Plasma treatment sterilizes the top and bottom PDMS piece. Therefore, after plasma treatment, the chip should be kept in sterile environment, spray the bench surface and hands to avoid contamination while handling the chips. Troubleshooting (Table 1). Transfer one bottom PDMS piece from the plasma asher to a clean bench area, take out a top PDMS piece from the plasma asher and carefully align the bottom PDMS piece with the top PDMS piece before binding the two. Note: Chips that underwent plasma treatment will become very sticky; any misalignment will not be able to be adjusted once the surface of top and bottom PDMS pieces get contact. Gently press the combined top and bottom PDMS piece for 10~20 s to better facilitate the binding process. Test whether the two parts are bonded closely by lightly attempting to peel off the bottom chip with a tweezer. If the pieces can be separated easily, the bonding is insufficient. In that case, apply more pressure to the chips to ensure sufficient bonding. Troubleshooting (Table 1). Remove the assembled chip from the bench surface using a tweezer, transfer both chips into the biosafety cabinet. Mix 50 μL of human recombinant Laminin 511 protein solution with 50 μL of sterile 1× PBS (+Ca2+) (+Mg2+) in an empty conical tube using a p200 pipette. Transfer 50 μL of the Laminin 511 / PBS mixture into both channels following the procedures below: (i) Use a p-200 pipette barrier tip to take 50 μL Laminin 511 / PBS mixture on ice. (ii) Transfer the tip close to the SF-OOC device bottom channel inlet on the top right corner. (iii) Gently push the pipette to remove the air bubble near the tip of the pipette tip, Insert the pipette tip gently into the SF-OOC device bottom channel inlet, push enough laminin solution so that the whole bottom channel is filled. (iv) While holding the pipette push button in place, use a tweezer to stabilize the chip while pulling out the pipette tip from the bottom channel inlet. Note: Do not release the pipette push bottom while pulling out the pipette tip, it will cause air bubbles to enter the SF-OOC channels. (v) While holding the pipette push button in place, insert the pipette tip gently into the SF- OOC device top channel inlet (bottom right corner), slowly push all the remaining laminin solution into the top channel all at once. (vi) without releasing the push bottom, use a tweezer to stabilize the chip while pulling out the pipette tip from the top channel inlet. (vii) Check to make sure there is no visible air bubble remains in both top and bottom channel. In case of visible air bubble in the channel, repeat step (i) to step (vii) until no air bubble is visible and both channels are filled with laminin solution. Troubleshooting (Table 1). Transfer the SF-OOC device to the ethanol sprayed large square sterile tissue culture plate, make sure the interior of the plate is air dried before transferring the SF-OOC device. Take 3~4 sterile 15 mL conical tube caps and place them on the plate, add 2.5 mL of sterile PBS into each cap. The PBS would maintain the moisture in the plate so that the channels will not dry up. Close the plate lid and parafilm, transfer the parafilm plate into a 37 °C in a 5% CO2 incubator for overnight laminin binding. Troubleshooting (Table 1). Maintenance of Human iPS Cells and Differentiation into Intermediate Mesoderm Cells, (~21 d) Following an established protocol10, briefly, human iPS cells are cultured in Matrigel-coated six- well plates, the cells are fed with mTeSR1 medium, and the medium is replaced daily until ~70% to 80% confluency for passaging or differentiation. For intermediate mesoderm cell differentiation, the human iPS cells are dissociated and seeded into a laminin-coated twelve-well plate (Reagent preparation) in mesoderm induction medium (Reagent preparation). The resulting cells are cultured in mesoderm induction medium for 2 d, the medium is replaced daily. Upon differentiation into mesoderm cells, the cells are cultured with intermediate mesoderm differentiation medium (Reagent preparation), and the medium is replaced daily. The cells are cultured for 14 d to fully differentiate into intermediate mesoderm cells. Differentiation of Human iPS Cells into Endothelial Cells (~13 d) Following an established protocol98, briefly, at ~70% to 80% confluency, human iPS cells are dissociated and seeded into a Matrigel-coated T-75 flask (Reagent preparation) in mTeSR1 supplemented with ROCK-inhibitor, followed by 1 d culture. Subsequently, the cells were cultured with lateral mesoderm induction medium (Reagent preparation) for 3 d without media change, followed by culturing with endothelial cell induction medium (Reagent preparation) for 3 d, the medium is replaced daily, and conditioned medium is collected for later use. Upon endothelial cell differentiation, the cells are sorted against CD31 and CD144 to obtain a pure endothelial cell population. Following cell sorting, the cells can be cryopreserved as P0, or the cells can be seeded in fibronectin-coated T-75 flasks (Reagent preparation) with complete conditioned medium for expansion. For cell expansion, the medium is replaced every 2 d until depletion of the complete conditioned medium. Then, the cells are cultured with CultureBoost-R, and the medium is replaced every 2 d. Expansion of Endothelial Cells (~4 d) Freshly differentiated and sorted endothelial cells at P0 (immediately after sorting) are recommended to expand into at least p2 for application in SF-OOC devices. Before expanding P0 endothelial cells, prepare a fibronectin-coated six-well plate (Reagent preparation). Aspirate fibronectin from the plates and rinse three times with prewarmed DMEM / F12, followed by adding 2 mL of CultureBoost-R into each well. Add the sorted endothelial cells into the six-well plate at a density of around 50,000 cells / well and shake the plates gently in figure eight to evenly distribute the cells. Maintain the endothelial cells at 37 °C in a 5% CO2 incubator. Replace the medium after overnight culture and then replace every 2 days. Passage the cells every 4 days or when the cells reach around 80% confluency. Seeding of Endothelial Cells into SF-OOC Devices (~6 h) Before seeding endothelial cells, visually inspect the SF-OOC devices to ensure both top and bottom fluid channels are filled with laminin solution without the presence of any air bubbles. Also, visually inspect endothelial cell culture plates to ensure they have reached around ~100% confluency. Rinse the SF-OOC devices with prewarmed CultureBoost-R for the bottom channel and DMEM / F12 for the top channel, using gravity wash. Gravity wash is performed using the following procedures. (i) Use a p-200 pipette barrier tip to take 200 μL prewarmed CultureBoost-R (ii) Transfer the tip close to the SF-OOC device bottom channel inlet on the top right corner. (iii) Gently push the pipette to remove the air bubble near the tip of the pipette tip, then hold the pipette in place without releasing. (iv) Insert the pipette tip gently into the SF-OOC device bottom channel inlet without releasing the pipette push button. (v) While holding the pipette push button in place, use another hand to gently remove the pipette tip by using the tip releasing button. The medium should start to flow through the pipette tip and the bottom channel slowly. (vi) Repeat step (i) to step (v) using DMEM / F12, adding the medium into the SF-OOC device top channel through the top channel inlet (bottom right). (vii) Aspirate the media from the channel outlets and remove the pipette tips once the medium inside stops flowing. Troubleshooting (Table 1). Keep the SF-OOC devices at 37 °C in a 5% CO2 incubator until later use. Aspirate the CultureBoost-R medium from endothelial cell culture plates. Add 1 mL of prewarmed Accutase to each well and incubate at 37 °C in a 5% CO2 incubator for 4 min or until most of the cells begin to partially lift. Perform a visual check under a brightfield microscope. Use a p-1000 pipette barrier tip to pipette up and down 4~5 times gently against the wells to dissociate the cells. Transfer the Accutase (mixed with suspending cells) into a 15 mL or 50 mL conical tube, this depends on the number of wells that are dissociated. Add the same volume of prewarmed DMEM / F12 as the Accutase into the same conical tube. Add 1 mL of prewarmed DMEM / F12 to each well and use a cell scraper to scrap the cells gently in an “X” motion and then rotate the scrap clockwise. Transfer the DMEM / F12 (mixed with remaining suspending cells) into the same conical tube. Centrifuge the cell suspension at 200 g for 5 min. Aspirate the supernatant and resuspend the cells in prewarmed CultureBoost-R, the volume depends on the number of SF-OOC devices to be used, the equation is shown below. (L) = 25 L ×Count the number of cells and calculate the desired volume of CultureBoost-R using the following equation. 1000Add prewarmed CultureBoost-R to the cell suspension to achieve the Desired Volume. Take the SF-OOC devices out from the incubator, use a p-200 pipette barrier tip to gently remove the medium from the top channel and the bottom channel, through drawing medium from the channel outlets (top channel outlet: top left; bottom channel outlet: bottom left). Laminin can become deactivated if left dried, proceed with this step timely to avoid drying of the channels. Using a p-200 pipette barrier tip, add 25 μL endothelial cell suspension into the SF-OOC device bottom channel through the channel inlet (top right). Note: The suspension needs to be added all at once to minimize air bubbles forming in the channel. Using a new p-200 pipette barrier tip, add 30 L prewarmed DMEM / F12 to the top channel through the channel inlet (bottom right). Flip the SF-OOC devices and perform inverted culture at 37 °C in a 5% CO2 incubator for 4 h, with a few caps of sterile 1×DPBS. Rinse the SF-OOC devices with prewarmed DMEM / F12 and CultureBoost-R in the top and bottom channel, respectively, using gravity wash. Small air bubbles can build up in the channel ports after 4 h incubation, gently push the pipette to expel the air bubble before releasing the tip to gravity wash. Aspirate the media that flows out from the channel outlets and remove the pipette tips once the medium inside stops flowing. Avoid aspirating directly on the channel ports, the suction can damage the cells and the SF membrane. Flip the SF-OOC devices and perform inverted culture at 37 °C in a 5% CO2 incubator overnight, with a few caps of sterile 1×DPBS. Troubleshooting (Table 1). Seeding of Intermediate Mesoderm Cells into SF-OOC Devices (~5 h) Before dissociating the intermediate mesoderm cells, visually inspect the cells under a 10× brightfield microscope to ensure the cells have reached 100% confluency. Troubleshooting (Table 1). Take the SF-OOC devices out from the incubator, flip the devices to return to their normal orientation, with channel ports facing up. Rinse the SF-OOC devices with prewarmed DMEM / F12 and CultureBoost-R in the top and bottom channel, respectively, using gravity wash. Keep the SF-OOC devices at 37 °C in a 5% CO2 incubator until later use. Prepare trypsin-neutralizing solution at a volume suitable for the scale of the experiment. A well of 100% confluent intermediate mesoderm cells is sufficient for an experiment using 16 SF-OOC devices. Therefore, 2 mL of trypsin-neutralizing solution is recommended. In a twelve-well plate of intermediate mesoderm cells, aspirate the medium in one well. Add 500 L of prewarmed 0.05% Trypsin-EDTA into the well, incubate for 1 min at 37 °C in a 5% CO2 incubator. Visually inspect the cells to ensure the cell edges begin to roll up. Use a cell scarper to scrap the cells in an “X” motion and then rotate the scrap clockwise. Pipette the cells several times using a p-1000 barrier tip. Visually inspect the cells again to ensure they are properly dissociated into individual cells. Avoid any noticeable cell clumps as they cannot spread out in the SF-OOC devices, leading to suboptimal barrier function. Transfer the cell suspension into the trypsin-neutralizing solution. Centrifuge the cells at 200 g for 5 min. Aspirate the supernatant and resuspend the cells with the intermediate mesoderm medium, the volume of the medium is calculated based on the following equation. (L) = 30 L ×Count the cells and calculate the desired volume of the Intermediate Mesoderm medium using the following equation.2 × 10 1000Add prewarmed Intermediate Mesoderm medium to the cell suspension to achieve the Desired Volume. Take the SF-OOC devices out from the incubator, use a p-200 pipette barrier tip to gently remove the medium from the top channel, through drawing medium from the channel outlet (top channel outlet: top left). Using a p-200 pipette barrier tip, add 30 L intermediate mesoderm cell suspension into the SF-OOC device top channel through the channel inlet (bottom right). Note: The suspension needs to be added all at once to minimize air bubbles forming in the channel. Culture the devices at 37 °C in a 5% CO2 incubator for 4 h, with a few caps of sterile 1×DPBS. Gently insert an empty p-200 pipette barrier tip into both top and bottom channel outlet port. Rinse the SF-OOC devices with prewarmed intermediate mesoderm medium and CultureBoost- R in the top and bottom channel, respectively, using gravity wash. Once the media stops flowing, gently press the pipette tips into the channel ports. Avoid pressing the tips too deep into the channel ports, over-pressing can lead to air bubble formation. Culture the devices at 37 °C in a 5% CO2 incubator for overnight. Remove all pipette tips on the devices, gently insert a new empty p-200 pipette barrier tip into both top and bottom channel outlet port. Rinse the SF-OOC devices with prewarmed podocyte induction medium and CultureBoost-R in the top and bottom channel, respectively, using gravity wash. Culture the devices at 37 °C in a 5% CO2 incubator for overnight. Troubleshooting (Table 1). Connection of SF-OOC Devices to a Peristaltic Pump (~2 h) Transfer the SF-OOC devices on to a presterilized customized 3D-printed holder (Reagent preparation). Bring the devices back into the incubator until later use. Arrange desired numbers of presterilized plastic beakers in a sterile six-well plate, the number depends on the number of SF-OOC devices to be connected. One device needs two beakers. Add 10 mL of the podocyte induction medium and 10 mL of CultureBoost-R into each beaker. Cover the beakers using a presterilized p-1000 tip box lid (with ~2 mm holes drilled on top for tubing connection) and label on top of the beakers as “top” and “bottom”, respectively. Work under a biosafety cabinet under sterile conditions, on an incubator tray, load desired numbers of presterilized 2-stop tubing onto a peristaltic pump and insert 16-G blunt needle pins to the outlet side of the 2-stop tubing. One device needs two 2-stop tubing. Insert the inlet side of the 2-stop tubing into the beakers through the drilled holes on the tip box lid. Connect the peristaltic pump to power and start the pump to maximize the flow rate using the “MAX” button. Hold the “MAX” button until media comes out from the outlet side (media will be visible with pink color). Pause the peristaltic pump by pressing the “Stop” button. Insert sterile 10-mL syringes (attached with 16-G blunt tip needles) into the SF-OOC device channel outlets through holes on the 3D-printed holder. Start the pump by pressing the “Start” button. Insert the blunt needle pins into their corresponding SF-OOC device channel inlets. Move the incubator tray with the entire SF-OOC device setup back into an incubator and perfuse at 4.09 L / min at 37 °C in 5% CO2. Maintenance of SF-OOC Devices on the Peristaltic Pump (~4 d) After 24 h, bring the incubator tray with the entire SF-OOC device setup back into a biosafety cabinet. Use a 10-mL serological pipette tip to transfer the media from the 10-mL syringes (the outlet side) back to their corresponding SF-OOC device medium reservoir beakers (the inlet side). Bring the incubator tray back into the incubator and perfuse at 4.09 L / min at 37 °C in 5% CO2. Troubleshooting (Table 1). After 24 h, bring the incubator tray with the entire SF-OOC device setup back into a biosafety cabinet. Aspirate the media from the 10-mL syringes (the outlet side). Add 10 mL of prewarmed podocyte induction medium and CultureBoost-R into the top channel beaker and the bottom channel beaker, respectively. Bring the incubator tray back into the incubator and perfuse at 4.09 L / min at 37 °C in 5% CO2. After 24 h, repeat step 49. After 24 h, podocytes are fully differentiated in the SF-OOC devices, and the devices are ready for downstream applications. Troubleshooting (Table 1). Filtration Assay of SF-OOC Devices (~7 h) Upon podocyte differentiation, prepare a solution consisting of 10 g / mL inulin-FITC and 100 g / mL albumin-Texas Red in CultureBoost-R. Keep the solution in dark to avoid light exposure. Bring the incubator tray with the entire SF-OOC device setup back into a biosafety cabinet. Aspirate the media from the 10-mL syringes (the outlet side) and from the inlet reservoir beakers (the inlet side). Add 3 mL of the inulin / albumin CultureBoost-R into the bottom channel beakers, add 3 mL of regular CultureBoost-R to the top channel beakers. Move the incubator tray with the entire SF-OOC device setup back into an incubator and perfuse at 4.09 L / min at 37 °C in 5% CO2 for 6 h. Transfer 100 L of outflow from the top channel syringe (the outlet side) into a clear flat bottom 96-well plate and measure the fluorescence intensity of FITC and Texas Red using a plate reader. Take technical duplicates for each condition measured and a minimum of three independent replicates per experimental condition. Calculate the amount of inulin-FITC and albumin-Texas Red filtered from the bottom channel into the top channel using the following equation for clearance (%). Perform data analysis and visualization using GraphPad Prism or Microsoft Office Excel. GraphPad Prism v9 and Microsoft Office Excel 2024 were used for data analysis. Immunocytochemistry of SF-OOC Devices (~2 d) Aspirate all media in 10-mL syringes (the outlet side) and in beakers (the inlet side). Disconnect the SF-OOC devices from the peristaltic pump. Transfer the SF-OOC devices to a petri dish for handling. Label the petri dishes appropriately with the order of the SF-OOC devices. Rinse the SF-OOC devices by adding 200 L of 1×DPBS into both top and bottom channel through gravity wash. Aspirate outflow from channel outlet ports. Remove the pipette tips once DPBS inside the tips stop flowing. Add 200 L of 4% (v / v) paraformaldehyde in 1×DPBS into both top and bottom channel through gravity wash. Remove the pipette tips once the buffer inside the tips stops flowing. Incubate the devices with 4% (v / v) paraformaldehyde for 20 min at room temperature. Remove paraformaldehyde using p-200 pipette barrier tips. Add 200 L of 0.125% (v / v) Triton X- 100 in PBS to the devices through gravity wash and incubate for 5 min at room temperature. Add 200 L of 1% (wt / v) BSA and 0.125% (v / v) Triton X-100 in PBS to the devices through gravity wash and incubate for 1 h at room temperature. Add 200 L of 0.125% (v / v) Triton X-100 in PBS to the devices through gravity wash to rinse the cells. Incubate for 10 min at room temperature for each wash. Repeat three times. Add 200 L of primary antibodies in permeabilization buffer and incubate overnight at 4 °C. For example, if aiming to stain the human iPS cell-derived podocytes with nephrin and stain the human iPS cell-derived endothelial cells with VE-Cadherin, prepare the dilutions of a mixture of nephrin and VE-Cadherin primary antibodies in 0.125% (v / v) Triton X-100 in PBS. Then, add the mixture antibody solution to the SF-OOC devices. Add 200 L of 0.125% (v / v) Triton X-100 in PBS to the devices through gravity wash to rinse the cells. Incubate for 10 min at room temperature for each wash. Repeat three times. Add 200 L of the first secondary antibody in permeabilization buffer and incubate for 1 h at room temperature. For example, if aiming to stain the human iPS cell-derived podocytes with nephrin secondary antibody, prepare the dilutions (1:1000) of nephrin secondary antibody in 0.125% (v / v) Triton X-100 in PBS. Then, add the antibody solution to the SF-OOC devices. Add 200 L of 0.125% (v / v) Triton X-100 in PBS to the devices through gravity wash to rinse the cells. Incubate for 10 min at room temperature for each wash. Repeat three times. Add 200 L of the second secondary antibody in permeabilization buffer and incubate for 1 h at room temperature. For example, if aiming to stain the human iPS cell-derived endothelial cells with VE-Cadherin secondary antibody, prepare the dilutions (1:1000) of VE-Cadherin secondary antibody in 0.125% (v / v) Triton X-100 in PBS. Then, add the antibody solution to the SF-OOC devices. Add 200 L of 0.125% (v / v) Triton X-100 in PBS to the devices through gravity wash to rinse the cells. Incubate for 10 min at room temperature for each wash. Repeat three times. Add 200 L of distilled water to the devices through gravity wash to rinse the cells. Counterstain the cells with DAPI (1:1000) in distilled water for 5 min. Quench the autofluorescence with 0.03% (v / v) Sudan Black B in 70% ethanol for 5 min. Add 200 L of 1× DPBS to the devices through gravity wash to rinse the cells. Repeat three times. Gently peel off the top PDMS portion of the SF-OOC device, visualize the cells by using an upright confocal microscope. Images can be analyzed, and 3D reconstructed by Fiji ImageJ. TEM of SF-OOC Devices (~5 d) Aspirate all media in 10-mL syringes (the outlet side) and in beakers (the inlet side). Disconnect the SF-OOC devices from the peristaltic pump. Transfer the SF-OOC devices to a petri dish for handling. Label the petri dishes appropriately with the order of the SF-OOC devices. Prepare fixation buffer by adding 10 mL of 20% formaldehyde, 4 mL of 25% glutaraldehyde, 5 mL of 10×DPBS, and 31 mL of distilled water. This formula makes 50 mL of fixation buffer, and the volume can be adjusted based on the number of SF-OOC devices to be fixed. Rinse the SF-OOC devices by adding 200 L of 1×DPBS into both top and bottom channel through gravity wash. Aspirate outflow from channel outlet ports. Remove the pipette tips once DPBS inside the tips stop flowing. Add 200 L of the fixation buffer into both top and bottom channel through gravity wash. Remove the pipette tips once the buffer inside the tips stops flowing. Incubate the devices with the fixation buffer for 2 h at room temperature or overnight at 4 °C. Remove fixation buffer using p-200 pipette barrier tips. Add 200 L of 1×DPBS to the devices through gravity wash and incubate for 10 min at room temperature. Repeat twice. Remove PBS using p-200 pipette barrier tips. Add 200 L of 1% OsO4 to the devices through gravity wash and incubate for 1 h at room temperature, avoiding light exposure. Remove OsO4 and add 200 L of 1×DPBS to the devices through gravity wash and incubate for 10 min at room temperature. Repeat twice. Add 200 L of 0.1 N acetate buffer to the devices through gravity wash and incubate for 10 min at room temperature. Remove acetate buffer. Add 200 L of 0.5% uranyl acetate to the devices through gravity wash and incubate for 1 h at room temperature. Remove uranyl acetate. Add 200 L of 0.1 N acetate buffer to the devices through gravity wash and incubate for 10 min at room temperature. Repeat twice. Remove acetate buffer. Add 200 L of 30% ethanol to the devices through gravity wash and incubate for 10 min at room temperature. Repeat twice. Add 200 L of 50% ethanol to the devices through gravity wash and incubate for 10 min at room temperature. Repeat twice. Add 200 L of 70% ethanol to the devices through gravity wash and incubate for 10 min at room temperature. Repeat twice. Add 200 L of 90% ethanol to the devices through gravity wash and incubate for 10 min at room temperature. Repeat twice. Add 200 L of 100% ethanol to the devices through gravity wash and incubate for 10 min at room temperature. Repeat three times. Gently peel off the top PDMS portion of the devices. In 100% ethanol, cut away the bottom PDMS portion around the biomimetic membrane using a scalpel blade. Prepare Spurr’s resin (Reagent preparation) in a fume hood. In the fume hood, transfer the devices into a glass vial and add around 2 mL of Spurr’s resin to ensure complete immersion of the devices. Incubate the devices with Spurr’s resin for overnight at room temperature, without the lid. Make new Spurr’s resin (Reagent preparation) in a fume hood. Incubate the devices with the old Spurr’s resin for 10 min at 55 °C. In the fume hood, remove the old Spurr’s resin, and add 2 mL of new Spurr’s resin to the devices. Incubate the devices with the old Spurr’s resin for 10 min at 55 °C. In the fume hood, remove the old Spurr’s resin. Transfer the devices to a plastic mold and ensure the bottom PDMS portion faces up. Add ~1 mL of new Spurr’s resin to cover the devices. Incubate the devices with the new Spurr’s resin for 48 h at 55 °C. Demold the resin from the mold. Microtome section the device in the resin, followed by negative staining following the procedures below: (a) Coat TEM grids with formvar followed by a thin layer of carbon. (b) Place the microtome sectioned samples on the coated grids and incubate for 1-15 min. (c) Dry the samples on the grids using filter paper. (d) Wash the grids with ultrapure water once. (e) Dry the samples on the grids using filter paper. (f) Add a drop of 0.5% uranyl acetate on the grids and incubate for 30 s to 15 min. (g) Dry the samples on the grids using filter paper and allow the grids to air dry. Visualize the samples with TEM. Protein Extraction of SF-OOC Devices (~2 h) Aspirate all media in 10-mL syringes (the outlet side) and in beakers (the inlet side). Disconnect the SF-OOC devices from the peristaltic pump. Transfer the SF-OOC devices to a petri dish for handling. Label the petri dishes appropriately with the order of the SF-OOC devices. Prepare cell lysing buffer RIPA on ice. (Reagent preparation). Rinse the SF-OOC devices by adding 200 L of ice-cold 1×DPBS into both top and bottom channel through gravity wash. Aspirate outflow from channel outlet ports. Remove the pipette tips once DPBS inside the tips stop flowing. Remove 1×DPBS from the devices using p-200 pipette barrier tips. Add 25 L of RIPA to the bottom channel of the devices and add 30 L of RIPA to the top channel of the devices. Incubate the device with RIPA on ice for 5 min. Pipette RIPA in both top and bottom channel several times to sufficiently lyse all the cells. On ice, transfer the RIPA into a 1.5-mL microtube. Shake the microtube in a cold room for 1 h. Centrifuge the microtube at 10,000 rpm for 1 min. Transfer the supernatant into a new microtube on ice. Aliquot and store at -80 °C. The resulting protein lysate can be subsequently used for western blot to detect both podocyte and endothelial cell-specific markers, such as SYNPO and VE-Cadherin, respectively. Cyclic Stretching of SF-OOC Devices The highly elastic feature of the biomimetic membrane shows promises in application of cyclic stretching in the SF-OOC devices. While both podocytes and endothelial cells were shown to form competent layers to demonstrate proper size-selective filtration function without cyclic stretching, other applications where cyclic stretching is preferred could include a periodic vacuum generation system to the device. The vacuum system can be connected to the SF-OOC devices through pre-designed ports for vacuum channels that line next to the fluid channel in the middle. Alternatively, as the SF-OOC devices demonstrated compatibility with the Emulate® system, which has an embedded vacuum regulation function. In this case, cyclic stretching can be performed by simply turning on the cyclic stretching function using the Emulate® system control panel. Timing Step 1 – 6, SF electrospinning and post-treatment: ~4 d Step 7 – 10, PDMS soft lithography: ~5 h Step 11 – 23, SF-OOC device fabrication and laminin functionalization: ~5 d Step 24, Intermediate mesoderm cell differentiation from human iPS cells: ~21 d Step 25 – 27, Endothelial cell differentiation from human iPS cells: ~17 d Step 28 – 43, Cell seeding in SF chips, including endothelial cell seeding, intermediate mesoderm cell seeding, and static culture: ~3 d Step 44 – 52, SF-OOC device maintenance under fluid flow for podocyte differentiation and maturation: ~4 d Option A, Filtration function assay with inulin / albumin: ~7 h Option B, step i – vii, Immunocytochemistry, primary antibody staining: ~2.5 h Option B, step viii – xvii, Immunocytochemistry, secondary antibody staining: ~3.5 h Option C, step i – xiv, TEM, sample fixation and dehydration: ~8 h Option C, step xv, TEM, sample resin embedding: ~1 h Option C, step xvi – xvii, TEM, sample negative staining and visualization: ~5 h Option D, Cell lysing and protein extraction: ~2 h Troubleshooting Troubleshooting solutions can be found in Table 1. This protocol provides a detailed method for manufacturing of an SF-OOC device with an ultrathin biomimetic membrane, and its application in mimicking the tissue morphology and function of the human glomerular filtration barrier. With this protocol, the biomimetic membrane can be fabricated and integrated into an SF-OOC device within 9 days, while several pause points can be included due to capability to store the biomimetic membrane long term (up to 1 year) at room temperature. At the same time, following established podocyte and vascular endothelial cell differentiation protocols using human iPS cells, the podocyte differentiation protocol requires an approximate 16 days of to generate ready-to-use intermediate mesoderm cells for SF-OOC device seeding, while the endothelial cell differentiation protocol requires an approximate 17 days to generate ready-to-use endothelial cells. Once all components (SF-OOC devices, endothelial cells, intermediate mesoderm cells) are ready, this protocol requires around 7 additional days to generate the final glomerulus SF-OOC device that is ready for subsequent applications. For example, immunocytochemistry can be performed to analyze the morphology and cell-specific marker expression patterns of podocytes and endothelial cells (FIG.2A–B). Both podocytes and endothelial cells show a confluent cell coverage on each side of the biomimetic membrane, contributing to its size-selective filtration function, where small molecules inulin and urea are freely filtered from the bottom (vascular) channel into the top (urinary) channel, and large molecules such as albumin, B2M, and Visfatin are restricted in the bottom (vascular) channel (FIG.2E and FIG.19). Infusion of a chemotherapy drug, Adriamycin, at clinical dosage for 3 days through the bottom (vascular) channel results in compromised podocyte morphology and molecular characteristics, as nephrin and podocin expression in the podocytes decreased substantially, and the cells appeared to exhibit disrupted subcellular nephrin and podocin expression patterns (FIG. 3A–C). On the other hand, endothelial cells seem to be intact in their morphology even though the drug was administered through the bottom (vascular) channel where the endothelial cells were lined (FIG. 3A–C). This result demonstrates the kidney-specific toxicity of Adriamycin as podocytes appear to be more sensitive to Adriamycin treatment than endothelial cells, consistent with in vivo and clinical reports. In addition to the morphological changes observed in podocytes after Adriamycin treatment, the SF-OOC devices also exhibit compromised size-selective filtration function, as the diseased devices became leakier to albumin, where Adriamycin-treated devices show significantly increased albumin clearance rate compared to vehicle control (FIG.3D). This result demonstrates the capability of the SF-OOC devices to mimic microalbuminuria as observed in patients affected by Adriamycin-induced acute kidney injury. The application of the ultrathin biomimetic membrane in the SF-OOC devices in this protocol also supports a facile procedure to prepare the samples for high-resolution imaging, especially TEM. Such imaging techniques allow for visualizing ultra-structures of the cells cultured in the SF-OOC devices. For example, the glomerulus SF-OOC devices developed in this protocol demonstrate formation of endothelial fenestrations when co-cultured with podocytes, whereas mono-cultured endothelial cells in the SF-OOC devices did not form any of the similar fenestration structures (FIG. 4F–H). Endothelial fenestration is a hallmark of glomerular endothelial cells, and these fenestrations play crucial roles in regulating the filtration function of the glomerular filtration barrier. Visualization of cell ultra-structures at high resolution with TEM has been challenging with other pre-existing SF-OOC systems, where a highly elastic PDMS porous membrane is used as a barrier to separate different fluid channels. Due to the high elasticity of these membranes, mechanical mismatch between the PDMS membrane, formed tissues, and the assay resin leads to difficulties in microtome sectioning, therefore limiting its compatibility with TEM. Moreover, compared to PDMS membranes which have an around 50 m thickness (~100-fold thicker than native glomerular basement membrane), the biomimetic membrane in this protocol has a thickness of around 3–5 m (10-fold thinner than porous PDMS membranes). Such ultra-thinness enhances intercellular crosstalk between podocytes and endothelial cells located on opposite sides of the membrane, as demonstrated by the formation of endothelial fenestrations in the SF-OOC devices. The formation of endothelial fenestration is further validated by the presence of VEGF signaling between podocytes and endothelial cells in the devices (FIG. 5B–C). The capability of the SF-OOC devices to recapitulate intercellular crosstalk between podocytes and endothelial cells and resulting tissue-specific morphogenesis (endothelial fenestrations) paves the way for more in-depth investigation of how endothelial fenestrations are developed in a human-specific model. Additionally, how endothelial fenestrations can be affected under pathophysiological conditions can also be explored using this device, thus shedding light on potential therapeutic targets on glomerular endothelial fenestrations to improve patient clinical outcomes. Together, this protocol provides a facile and robust approach to engineer a transformative SF-OOC device with an ultrathin biomimetic membrane for modeling the tissue morphogenesis, function, and disease of the human kidney glomerulus. Given the superior biocompatibility of SF to support culture of various other cell types, this SF-OOC device can also be used to model other tissue / organ types, such as small intestines, lung alveoli, and the blood brain barrier. This approach allows for interconnecting various SF-OOC devices to study inter-organ crosstalk, as well as potential development of human body-on-a-chip for pharmacodynamics / pharmacokinetics modeling in drug development and for patient-specific modeling of disease and testing of novel therapeutics when existing standard-of-care treatment does not achieve desirable clinical outcomes. Example 3 Glomerular Capillary Wall-on-A-Chip Device Engineering The organ-on-a-chip device was built with two channels separated by a biomimetic silk- based membrane. Human iPS-derived intermediate mesoderm cells were seeded in the top channel and differentiated into podocytes in the presence of iPS-derived endothelial cells seeded in the bottom channel. The vascularized chips were maintained for at least nine days in culture under constant fluid flow and then analyzed using confocal and electron microscopy. Generation of Patient-Specific Induced Pluripotent (iPS) Cells Protocols for the characterization of the patients and the creation of iPS cells were developed in accordance with the requirements of the institutional review board at Duke University Hospital. The kindred’s parents provided written informed consent. Patient-Specific Cell Isolation Whole blood was collected for Duke 5 and Duke 7 using EDTA blood collection tubes. Peripheral blood mononuclear cells (PBMCs) were isolated by density gradient centrifugation with SepMate-50 PBMC Isolation Tubes (StemCell Technologies, 85450) used in conjunction with the density gradient medium Lymphoprep (StemCell Technologies, 07811) per manufacturer’s protocol. PBMCs were then washed with DPBS with 2% FBS (StemCell Technologies, 07905) and cryopreserved in CryoStor CS10 (Stemcell Technologies, 07930) for long-term storage in liquid nitrogen. Human iPS Cell Generation To generate iPSCs derived from Duke 5 and Duke 7, PBMCs were thawed gently at 37°C and diluted 1:10 into DPBS with 2% FBS to remove cryopreservation solution. PBMCs were resuspended in StemSpan SFEM II (StemCell Technologies, 09655), StemSpan Erythroid Expansion Supplement (1X) (Stemcell Technologies, 02692), Primocin (0.2%) (InvivoGen, ant- pm-1), and Penicillin-Streptomycin (1%) (Gibco, 15140122) and plated on a 6-well plate. PBMC media was changed every other day. After seven days PBMCs were collected and counted. 30,000 PBMCs were transduced (Day 0) using CytoTune-iPS 2.0 Sendai Reprogramming Kit (ThermoFisher Scientific, A16517) prepared in PBMC Media (without Primocin) following manufacturer’s protocol. Transduced cells were seeded into one well each of a 48-well plate and incubated for 24 hours in a 37°C incubator at 5% CO2. The media was changed into PBMC Media to remove Sendai virus, and cells were incubated for an additional 48 hours. A feeder layer of mouse embryonic fibroblasts (MEFs) (Gibco, A24903) was prepared on Geltrex LDEV-Free, hESC-Qualified, Reduced Growth Factor Basement Membrane Matrix (Gibco, A1569601)-coated 12-well plates in MEF Media: DMEM (Gibco, 11965092), human Embryonic Stem Cell-qualified FBS (10%) (Gibco, 16141061), MEM Non-Essential Amino Acids Solution (1mM) (Gibco, 11140050), Beta-mercaptoethanol (55uM), Penicillin-Streptomycin (1%) (Gibco, 15140122). On Day 3, the transduced PBMCs were collected, resuspended in iPSC Media: DMEM F12 + GLutamax (Gibco, 10565018), Knockout Serum Replacement (20%) (Gibco, 10828010), MEM Non-Essential Amino Acids Solution (1X) (Gibco, 11140050), Penicillin-Streptomycin (1%) (Gibco, 15140122), Beta-mercaptoethanol (55uM) (Gibco, 21985023), Human FGF-basic (10ng / mL) (Peprotech, 100-18B), and seeded onto the MEF-coated plates. The freshly seeded plate was centrifuged at 210-300 RCF for five minutes in order to encourage cell-to-cell adhesion between the PBMCs and MEFs. Media was changed with fresh iPSC Media every other day or as needed. Between Days 19-21 robust iPSC colonies were dissociated from the MEFs with Versene Solution (Gibco, 15040066), incubated for two minutes at 37°C, picked via suction with a P200 pipette tip, and resuspended in StemFlex Basal Medium plus StemFlex Supplement (1X) (Gibco, A3349401), Penicillin-Streptomycin (1%) (Gibco, 15140122 ), and rho-kinase inhibitor Y-27632 dihydrochloride (0.1%) (Selleck Chem, S1049), and seeded into individual wells of a 12-well, Vitronectin (VTN-N) Recombinant Human Protein, Truncated (ThermoFisher Scientific, A14700) coated plate. Up to five clones per sample were selected. After 24 hours cells had adhered to the plate and media was changed to remove Y-27632. Cells were passaged such that confluence was maintained between 30 - 70% and media was changed every other day or as needed. Y- 27632 was used in StemFlex media to passage cells and removed 24 hours after each passage. iPS cells morphology was monitored and assessed daily to ensure cells maintained compact structures with distinct borders. Ultimately, two clones per sample were selected for further maturation. After no fewer than 10-15 passages, iPS cells were considered to be stable and ready for quality assessments. Two iPSC clones from each sample (Duke 5 and Duke 7) were column-purified with anti-TRA-1-60 microbeads (Miltenyi Biotec, Inc., 130-100-832), passaged and plated on a 6-well VTN-N coated plate for expansion, along with two wells of a 96-well coated plate for assessment of iPS cells via immunocytochemistry. The Pluripotent Stem Cell 4-Marker Immunocytochemistry Kit (Invitrogen, A24881) was used to detect SOX2, TRA-1-60, OCT4, and SSEA4, known expression factors of mature iPSC cells. A cell pellet was collected and KaryoStat assay (Thermo Fisher Scientific, 905403) was performed to ensure no chromosomal abnormalities were introduced through the iPSC derivation process. Human iPS cells that passed quality control assessments were frozen in CryoStor CS10 and stored in liquid nitrogen until ready for downstream use. iPS Cell Culture Patient-specific iPS cells (WT, Duke 5, and Duke 7) were propagated on 6-well plates coated with Matrigel Matrigel (VWR; 75796-278) by using mTeSR1 (StemCell Technologies; 85870) medium without antibiotics. The cells were split every 5 to 6 days (or until 75% confluency is reached) by treatment with Accutase (Thermo / Life Technologies; A1110501). The cell lines were propagated in a 37 °C incubator with 5% CO2.The cell lines were routinely tested for mycoplasma and the cells were found to be free of mycoplasma contamination (Mycoplasma PCR Detection Kit from abm. G238). Differentiation of Mesoderm, IM, and Podocytes For directed differentiation of Duke 5 and Duke 7 iPS cells towards mesoderm, iPS cells were carefully dislodged with enzyme-free dissociation buffer (Gibco, 13150-016) and centrifuged once at 200 g for 5 mins in advanced DMEM / F12 (Gibco; 12634010) to remove residual Matrigel or stem cell culture media components. The supernatant was aspirated, and the cells were resuspended in mesoderm differentiation media. Cells were seeded in a laminin 511-E8 (Takara T304)-coated 12 well plate. The mesoderm differentiation media comprised of 100 ng / ml activin A (Thermo Fisher Scientific, PHC9561), 3 M CHIR99021 (Stemgent, 04-0004), 10 M Y27632 (TOCRIS, 1254), and 1× B27 serum-free supplement (GIBCO, 17504044) dissolved in DMEM / F12 with GlutaMax (GIBCO, 10565018). After 2 days of differentiation, the media was replaced with Intermediate Mesoderm Differentiation Media. The Intermediate Mesoderm Differentiation Media comprised of 100 ng / ml BMP7 (Thermo Fisher Scientific, PHC9541), 3 M CHIR99021 (Stemgent, 04-0004), and 1× B27 serum-free supplement (GIBCO, 17504044) dissolved in DMEM / F12 with GlutaMax (GIBCO, 10565018). After 14 days of differentiation, Intermediate Mesoderm cells were trypsinized with 0.25% trypsin-EDTA (Gibco; 25200-056) and cells were replated on freshly prepared Laminin 511-E8 coated 12 well plates at a density of 100,000 cells / well with Podocyte Differentiation Medium. The Podocyte Differentiation Medium comprised of 100 ng / ml BMP7, 100 ng / ml activin A, 50 ng / ml VEGF (Thermo Fisher Scientific, PHC9391), 3 M CHIR99021, 1× B27 serum-free supplement, and 0.1 M all-trans retinoic acid (Stem Cell Technologies, 72262) dissolved in DMEM / F12 with GlutaMax (GIBCO, 10565018)16. The cells were cultured for 5 days with regular media change. Drug Treatment iPS-derived podocytes were maintained in CultureBoost-R for one day after the completion of the podocyte differentiation. For drug treatment, drugs (dissolved in DMSO) were added to CultureBoostR, mixed well, and introduced to the cells. DMSO served as a control. The drugs and their concentrations used in this study were as follows: Tacrolimus (MedchemExpress, HY-13756A, 1 μM), BI749327 (MedchemExpress, HY-111925, 500 nM), Sparsentan (MedchemExpress, HY-17621, 10 μM), Sildenafil (MedchemExpress, HY-15025,1 μM), and Losartan (MedChemExpress #HY-17512 / CS-2116, 10 μM). Differentiation of Endothelial Cells WT, Duke 5, and Duke 7 iPS cells were harvested with Accutase (Thermo / Life Technologies; A1110501) and reseeded on Matrigel-coated 6-well plates at a cell density of 420,000 cells / well. After 1 day of incubation, Lateral Mesoderm Differentiation Media was added to the wells. The Lateral Mesoderm Differentiation Media comprised of N2B27 media, which contains neurobasal media (Invitrogen, 21103049) and DMEM / F12 glutamax (Invitrogen) in 1:1 ratio with N2 (100×) (GIBCO, 21103049) and B27 without Vitamin A (GIBCO, 12587010), 8 μM CHIR99021 and 25 ng / mL hBMP4 (VWR International LLC, 10273-372) for 3 days. On Day 4, media was replaced with Endothelial cell Differentiation Media for an additional 3 days. The Endothelial cell Differentiation Media comprised of StemPro-34 SFM media (GIBCO, 10639011) supplemented with Glutamax (GIBCO, 35050061) in 100:1 ratio, 2 μM forskolin (Abcam Inc., ab120058), and 200 ng / mL VEGF165. Cells were fed daily, and conditioned media was collected for endothelial expansion media preparation. On Day 7, differentiated cells were dislodged with cold Accutase treatment for 5 mins and MACS-sorted (MidiMACS™ Separator) to harvest CD144+and CD31+cell populations. Purified cell populations were expanded in Endothelium Maintenance Media comprised of conditioned media diluted at 1:1 ratio with StemPro-34 SFM supplemented with 2 μg / mL heparin (STEMCELL Technologies, 07980). Media was replenished every other day or until the conditioned media was depleted. For continued expansion beyond the first passage, the cells were fed with viEC maintenance media, which comprised of StemPro- 34 supplemented with 10% HI-FBS (Invitrogen, 10082147), 2 μg / mL heparin, and 50 ng / mL VEGF165. Western Blot Analysis WT, Duke 5, and Duke 7 podocytes were washed with cold PBS (Gibco; 14190144) on Day 6 and lysed on ice in RIPA buffer (Millipore / Sigma; R0278-500ML) supplemented with PhosSTOP phosphatase inhibitors (Millipore / Sigma; 04906837001) and complete EDTA-free protease inhibitor cocktail (Millipore / Sigma; 4693132001). One tablet of each PhosSTOP phosphatase inhibitor and complete EDTA-free protease inhibitor was added in 10 mL RIPA Buffer. Protein samples were separated by SDS-PAGE using Mini-PROTEAN TGX Stain-Free Precast 4-15% Gels (Bio-rad; 4568083) and transferred onto a PVDF membrane (Bio-rad; 1704157) using a Trans-blot Turbo semi-dry transfer system (Biorad; 1704150). Membranes were blocked with 5% Blotto (ChemCruz; sc-2324) in Tris-buffered saline with Tween 20 (TBST), and immunoblotting was carried out according to standard procedure. Primary antibodies were always diluted in TBST supplemented with 5% Blotto and incubated for 1 h at room temperature on a platform rocker. Primary antibodies used for Western blot were guinea pig Anti-Nephrin antibody (ARP; #GPN02, 1:500), rabbit Anti-Podocin (Abcam, mAb #ab50339, 1:1000), mouse Anti-Synaptopodin (D-9) (SantaCruz, mAb #sc-515842, 1:1000), mouse Anti-CD2AP (B-4) (SantaCruz, mAb #sc-25273, 1:1000), mouse Anti-WT1 (6F-H2) (Millipore Sigma #05-753, 1:250), mouse Anti-GLEPP1 (B-6) (SantaCruz, mAb #sc-365354, 1:1000), and rabbit Anti- GAPDH (Millipore Sigma; #ABS16, 1:10000). Secondary antibodies used in this study included HRP-conjugated goat anti-mouse (Cell Signaling; #7076, 1:5000), HRP-conjugated goat anti- rabbit (Cell Signaling; #7074, 1:10000), HRP-conjugated goat anti-guinea pig polyclonal (ARP; #90001, 1:1000), and HRP-conjugated donkey anti-goat polyclonal (R&D Systems, #HAF109, 1:1000). Secondary antibodies were also diluted in TBST supplemented with 5% Blotto and incubated for 1 h at room temperature on a platform angle rocker. Chemiluminescence was detected using the Super Signal West Femto kit (Thermo; 34094). Signal intensities were analyzed using a ChemiDoc imager (Bio-rad). Immunostaining and Microscopy WT, Duke 5, and Duke 7 podocytes were washed with cold PBS (Gibco; 14190144) on Day 6 and were fixed with 4% paraformaldehyde (Sigma-Aldrich, #16005) in dPBS followed by permeabilization with 0.125% triton X-100 in dPBS for 5 min. Cells were blocked by incubation with a solution of 1% BSA and 0.125% triton X-100 in dPBS (blocking buffer) for 30 min on ice. The cells were incubated with primary antibodies in a permeabilization buffer overnight at 4 °C. The cells were washed with permeabilization buffer 3-4 times and incubated with secondary antibodies conjugated to either Alexafluor-488 (Life Technologies; #A212021:1000), Alexafluor- 594 (Life Technologies; #A212031:1000) or Alexafluor-700 (Life Technologies; #A210381:1000) in permeabilization buffer for 1 h at room temperature. Following incubation, the cells werewashed three times with permeabilization buffer and counterstained with 4 ,6-diamidino-2-phenylindole (DAPI) (Invitrogen #D1306). The primary antibodies used included guinea pig Anti- Nephrin (ARP; #GPN02, 1:500), rabbit Anti-Podocin (Abcam, mAb #ab50339, 1:1000), rabbit Anti- TRPC6 (Abcam, pAb #ab228771) and mouse Anti-CD2AP (B-4) (Santa Cruz, #sc-25272), Calnexin (C5C9) Anti-Rabbit (Cell Signaling, mAb #2679), LAMP1 (D2D11) XP Anti-Rabbit (Cell Signaling, mAb #9091), RAB7 (D95F2) XP and Anti-Rabbit mAb (Cell Signaling, mAb #9367). Immunofluorescence images were captured using the Zeiss 780 Upright confocal microscope equipped with 63x / 1.4 Oil Zeiss Plan-Apochromat 440762 (02) WD 0.19 mm objective. Data was analyzed using the Zen 2.3 Black software. Generation of Glomerular-Capillary-on-a-Chip Biomimetic silk fibroin chips were treated with oxygen plasma at 50 W and 0.8 mbar for 60 s using a plasma asher (Emitech, K-1050X) to activate the membrane for surface adsorption of laminin-511 solution (25 g / ml; Biolamina, LN511), followed by incubation at 37°C overnight. The next day, the channels were washed with advanced DMEM / F12 (Gibco; 12634010), and endothelial cells were seeded on the bottom channel. After 4 h incubation, isogenic Intermediate mesoderm cells were seeded on the top channel and incubated overnight at 37 C. The next day, the top channel was flushed with Podocyte Differentiation Medium, and the bottom was flushed with viEC medium. The cells were then continuously perfused with their respective cell-culture media using an Ismatec IPC-N digital peristaltic pump (Cole-Parmer) at a volumetric flow rate of246 l h 1 (shear stress of 4.09 × 10 3 dyn cm 2 for the top channel and 0.07 dyn cm 2 for thebottom channel). Cell culture media was recirculated, and recirculated media was replaced every alternate day from the falcon tubes. After 4 days of podocyte differentiation, the podocyte induction media was replaced with CultureBoost-R, and the resulting glomerulus chip was maintained for an additional day before treatment with the drugs. The chips were perfused for an additional 2 days with the drug solutions, followed by immunofluorescence analyses of electron microscopy. For barrier function analyses, CultureBoost-R media was supplemented with 100 g / mL albumin conjugated to Alexa Fluor 555 (Thermo Fisher Scientific; A34786) and perfused through the bottom channel. Outflow media was collected from the apical channel outlet, and fluorescence intensity was measured using a SpectraMax Fluorescent Plate reader (SpectraMax i3x, Molecular Devices). The amount of Albumin filtered from the basal to the apical channel was analyzed using an equation for renal clearance: ([U] × UV) / [P]), where [U] is the urinary concentration of albumin, UV = volume of media collected from the apical channel outlet reservoir, and [P] = dosing concentration in the basal channel. For immunofluorescence studies, chips were fixed with 4% paraformaldehyde (Sigma- Aldrich, #16005) in dPBS, followed by permeabilization with 0.125% triton X-100 in dPBS for 5 min. Cells were blocked by incubation with a solution of 1% BSA and 0.125% triton X-100 in dPBS (blocking buffer) for 30 min on ice. The chips were stained with guinea pig Anti-Nephrin (ARP; #GPN02, 1:500), rabbit Anti-Podocin (Abcam, mAb #ab50339, 1:1000), rabbit Anti-TRPC6 (Abcam, pAb #ab228771) and mouse Anti-CD2AP (B-4) (Santa Cruz, #sc-25272) overnight and imaged using the Zeiss 780 Upright confocal microscope equipped with 63x / 1.4 Oil Zeiss Plan- Apochromat 440762 (02) WD 0.19 mm objective. Data was analyzed using the Zen 2.3 Black software. Transmission Electron Microscopy After drug treatment, cells were washed with cold PBS (Gibco; 14190144) followed by fixation with 10 ml of 20% formaldehyde (Sigma-Aldrich, 8.18708), 4 ml of 25% glutaraldehyde (Sigma-Aldrich, G7651), 5 ml of 10× phosphate-buffered saline (Gibco, 14200075), and 31 ml of distilled water (Gibco, 15230001), overnight at 4°C. The chips were postfixed with 1% osmium tetroxide (Electron Microscopy Sciences, 19180) for 1 hour at room temperature in the fume hood, followed by fixation with 0.5% uranyl acetate (Electron Microscopy Sciences, 22400) and dehydration with ethanol (VWR) gradient. The chip’s top compartment was then removed, followed by immersion of the resulting chips in the Spurr’s resin containing 4.1 g of 3,4- Epoxycyclohexanemethyl 3,4-epoxycyclohexanecarboxylate (ERL 4221) (Electron Microscopy Sciences, 15004), 5.9 g of Nonenyl Succinic Anhydride (NSA) (Electron Microscopy Sciences, 19050), 1.43 g of Dow epoxy resins (DER) 736 (Electron Microscopy Sciences, 13000), and 0.1 ml of 2-Dimethylaminoethanol (DMAE) (Electron Microscopy Sciences, 13300), overnight at room temperature. Afterward, the chips were immersed in a freshly prepared Spurr’s resin and cured at 50° to 60°C for 24 to 48 hours. The resulting chip samples were microtome-slice, loaded on TEM grids, counterstained, and visualized using TEM (FEI Tecnai G2Twin). Computational Modeling of TRPC6 Protein Structure ColabFold was utilized to make AlphaFold predictions of the structure of a single TRPC6 chain containing variants L899H and P924T. The resulting structure was then comparatively analyzed against the reference PDB contained in the AlphaFold database utilizing the molecular graphics software UCSF ChimeraX. The protein structure was analyzed for both localized and global changes. UCSF ChimeraX was also utilized to predict the structure of the TRPC6 four- chain multimer by superimposing Alpha Fold’s single-chain predictions onto the reference multimer contained in the UniProt database based on the matchmaker function. The multimer was then analyzed for changes in pore diameter utilizing the tape measure tool in UCSF ChimeraX. These predictions and analyses were repeated with the wildtype sequence as a negative control and six known mutations (P112Q, L395A, G757D, L780P, P924R, P924S) as positive controls (data not shown). Statistical Analyses A Two-way Analysis of Variance (ANOVA) with Dunnett’s multiple comparison post hoc analyses was performed with a family-wise alpha threshold of 0.05 and with individual mean difference computed for each comparison for FIG. 24E. One-way ANOVA with Sidak’s multiple comparison post hoc analyses was performed with a family-wise alpha threshold of 0.05 for FIG. 24G and Tukey’s multiple comparison post hoc analyses for FIG. 25F. Two-way ANOVA with Tukey’s multiple comparison post hoc analyses test was performed for FIG.25B–E and unpaired t-test for FIG.25G. Molecular Genetic Analyses The DUK40130 proband (Duke 5, Female) and her affected twin sibling (Duke 7, Female) (FIG.23A) were diagnosed with proteinuria at the age of 22 with dipstick proteinuria. Quantitation of urine protein-to-creatinine ratio (UPCR) at age 24 revealed 1.1 and 2.1 mg / mg, respectively. Both individuals started therapy with Losartan at age 24 and tacrolimus was added at age 25. The UPCR has remained at ~ 0.8 - 1 mg / mg with normal eGFR (FIG.23B). Exome analyses of Duke 5 and Duke 7 revealed two single-nucleotide substitutions (c.2696 T > A and c.2770 C > A) in TRPC6, which result in the pathogenic L899H and P924T variants of the C-terminal helical region of the protein (FIG.23C). The variants are predicted to cause distortion of the C-terminal alpha helix, which was not observed in the reference or predicted WT structures (FIG.23D). The variants have not been documented together in the ~200,000 individuals present in gnomAD and ClinVar databases. Computational modeling of the novel compound variants revealed greater tertiary structural changes (RMSD ~19.2 Å) than other known variants (RMSD ~17.2-18.9 Å), without a significant alteration of the channel pore diameter (WT TRPC6 diameter = 9.71 Å, predicted WT TRPC6 diameter = 5.67 Å, and diameter = 5.85 Å) (FIG. 23E and FIG.26A). This distortion of the C-terminal alpha helix can possibly be explained by the removal of the helix-breaking proline at position 924 (FIG. 23D). Biopsy of Duke 5 revealed focal glomerular sclerosis and hyalinosis by light microscopy (FIG. 23F). Electron microscopy demonstrated podocyte foot process (FP) effacement and marked disruption of podocyte ER architecture associated with numerous electron-dense ER inclusions (FIG. 23G). TRPC6 is a major component of the podocyte SD complex, where it associates with other important SD proteins, including nephrin and podocin. Podocin and TRPC6 interact at their respective C- termini. This region of protein-protein contact operates as a molecular switch to negatively modulate TRPC6 activity24. Given the functional importance of this known intermolecular interaction, the presence of the variants within the region, and the unique presence of podocyte ER inclusions associated with the variants, it was hypothesized that the L899H / P924T variants may physically disrupt Nephrin-Podocin-TRPC6 complex formation, leading to trafficking disturbances and ER congestion. Slit Diaphragm Protein Localization Is Altered in Patients’ Podocytes Podocyte differentiation from the Duke 5 and Duke 7 iPS cells revealed large podocin- enriched ER inclusions associated with significantly reduced TRPC6 and nephrin expression (FIG.24A and FIG.26B–C). There is significant disruption of nephrin and podocin colocalization in Duke 5 and Duke 7 podocytes relative to healthy controls as confirmed by Mander’s correlation coefficient (FIG. 26D). Nephrin and podocin are important components of the filtration SD that also facilitate the assembly of pro-survival signaling assemblies and cytoskeletal regulatory networks at the glomerular filtration barrier (GFB). Given the established role of TRPC6 in regulating podocyte actin-cytoskeleton, it was questioned how the compound L899H and P924T variants might impact nephrin-podocin-TRPC6 subcellular localization to terminal podocyte arborizations. Immunostaining of healthy control podocytes revealed colocalized punctate expression of nephrin, podocin, and TRPC6 in terminal podocyte arborizations. Conversely, the colocalized expression of these markers was significantly reduced in Duke 5 and Duke 7 podocytes (FIG.24B and FIG.26E). Next, it was investigated how SD proteins interact spatially within the cell body and terminal podocyte arborizations. A 3D rendering of high-resolution confocal images revealed a uniform distribution of the nephrin-podocin-TRPC6 complexes throughout the cytoplasm and in the cell body of WT podocytes, highlighting the importance of these protein-protein interactions in podocyte morphological phenotype and molecular physiology. Punctate localization of nephrin- podocin-TRPC6 in podocyte arborizations was also observed. By contrast, this spatial patterning was disrupted in Duke 5 and Duke 7 podocytes, consistent with an impairment in SD complex formation in the diagnostic biopsy (FIG.23G and FIG.24C). Immunoblots from patient-specific differentiated podocytes also revealed significant downregulation of nephrin, TRPC6, and synaptopodin (FIG.27A–B) with massive podocin-enriched aggresomes in the ER (FIG.27C). To further explore potential disruption in trafficking or endosomal processing of the SD proteins, whole cell lysates from Duke 5 and Duke 7 podocytes were evaluated by immunoblot. It was demonstrated that the presence of high molecular weight podocin aggresomes were almost nonexistent in the WT. Evaluation of the ER morphology by high-resolution confocal microscopy demonstrated a regular perinuclear distribution of ER tubules in healthy controls, while the ER tubules in mutant podocytes exhibited a fragmented and attenuated appearance with mislocalized nephrin expression and high inositol 1,4,5 triphosphate receptor indicating calcium efflux from the ER (FIG.24D–E). The expression of late endosome and lysosomal biogenesis markers LAMP1 and RAB7, respectively, were then examined. While LAMP1 vesicle number remained similar in mutant and healthy podocytes (FIG.27D), RAB7 expression was significantly downregulated in the mutant podocytes, suggesting that Duke 5 and Duke 7 podocytes may exhibit reduced lysosomal biogenesis, leading to uncontrolled protein accumulation and ER congestion. (FIG. 27E). Sildenafil and Losartan Cotreatment Reduce Podocin-Rich Aggresomes and Recruit Nephrin Next, non-immunosuppressive candidate therapies that could improve podocyte SD marker localization and resolve podocin-enriched aggresomes were evaluated. To recapitulate patient clinical management, podocytes were first treated with tacrolimus and losartan (TL). Combination therapy with these agents did not restore nephrin-podocin colocalization (FIG.24F). The effects of the selective TRPC6 blocker, BI749327, was also tested, which has been shown to suppress TRPC6-mediated cation conductance and inhibit activation of Nuclear Factor of Activated T-cell (NFAT) in HEK cells overexpressing disease-causing gain-of-function TRPC6 mutations (i.e., P112Q, M132T, R157Q, R895C, and R895L). BI749327 failed to reduce podocin aggresomes and improve nephrin-podocin-TRPC6 colocalization in Duke 5 and Duke 7 podocytes, suggesting that the formation of the ER aggresomes was unrelated to aberrations in TRPC6-mediated cation conductance (FIG. 28A). Patient-derived podocytes with sparsentan were also tested, a dual endothelin-angiotensin II receptor antagonist that can suppress TRPC6- mediated calcium influx with preservation of podocyte nephrin and podocin expression27. Sparsentan also failed to reduce podocin-rich aggresomes or to reestablish nephrin-podocin colocalization (FIG.28B), highlighting again that the formation of ER protein aggresomes in the mutant podocytes may be independent of dysregulated TRPC6-mediated cation conductance28. When Duke 5 and Duke 7 podocytes were treated with the Phosphodiesterase 5 (PDE5) inhibitor, Sildenafil in combination with Losartan (SL), a significant reduction in podocin-rich aggresomes and complete restoration of nephrin-podocin colocalization was observed (FIG. 24F–G) which has important implications in bridging podocyte-podocyte junctions. Although PDE5 inhibitors have been shown to downregulate TRPC6 activity, it is likely that the beneficial effect of restoring podocyte proteostasis are pleiotropic and independent of the ability to suppress aberrant TRPC6 activity. Sildenafil and Losartan Ameliorate Albuminuria in an Engineered Glomerular Capillary Wall-On- A-Chip To explore the effects of therapy on podocyte barrier functions, a vascularized glomerular capillary wall-on-a-chip platform was engineered to recapitulate the GFB using patient-derived isogenic iPS cells (FIG. 25A). The tripartite glomerular filtration barrier is comprised of endothelium on the luminal surface of the glomerular basement membrane and terminally differentiated podocytes on the opposing urinary surface. Podocytes, with their interdigitating foot processes, provide the primary support for the filtration slit diaphragm, a heteroporous, zipper- like assembly of cadherin and immunoglobulin-like proteins that possess molecular charge and molecular size selectivity properties for ultrafiltration. After differentiation on the microfluidic chip, immunostaining of mutant podocytes revealed significantly elevated levels of podocin and TRPC6 and reduced levels of nephrin and CD2AP compared to WT (FIG.29). Treatment with TL or SL, significantly reduced podocin and TRPC6 aggregation in Duke 5 and Duke 7 podocytes (FIG. 25B–E and FIG.30). However, Duke 5 and Duke 7 podocytes treated with SL exhibited higher Nephrin and CD2AP levels as compared with their TL-treated counterparts (FIG.25C and 25E, and FIG. 30). These findings demonstrate that SL more effectively restores the pattern of SD protein localization than TL in mutant podocytes. Next, albumin excretion was examined by quantifying albumin filtration into the “urinary” channel effluent of the microfluidic chips. It was demonstrated that the effluent from Duke 5 chips had significantly higher albumin levels relative to controls, consistent with the clinical manifestation of the disease in the affected siblings (FIG. 25F). The mutant podocytes also demonstrated reduced albumin sequestration from the filtrate (FIG. 31A). Treatment with both SL and TL reduced albumin leakage (FIG. 25F) in the Duke 5 chips; however, SL was superior to TL in reducing albumin excretion. Albumin leakage in Duke 7 chips remained similar to WT chips after SL treatment, while TL-treated Duke 7 chips demonstrated increased albumin excretion (FIG. 25F). Subsequent TEM imaging of untreated Duke 5 and Duke 7 podocytes on the microfluidic chips revealed fragmented ER tubules with ER inclusions. In SL-treated patients’ podocytes, ER aggresomes were significantly reduced (FIG. 25G), and ER morphology was markedly improved (FIG.31B), consistent with a beneficial effect of improved proteostasis on podocyte barrier function. In this report, novel compound C-terminal TRPC6 variants that induce podocyte injury via alteration of SD protein expression and targeting were described, disruption of SD complex assembly, and ER congestion / fragmentation. These novel features of TRPC6-associated podocyte injury and dysfunction were amenable to therapeutic correction using combined ARB and PDE5 inhibition but were not ameliorated by treatment with tacrolimus or other agents targeting TRPC6-mediated calcium conductance. Although the precise mechanisms of the therapeutic benefit are yet to be determined, it is well-known that PDE5 inhibitors can ameliorate dysregulated TRPC6 activity in various cell types. Inhibition of PDE5 catalytic activity prevents the hydrolytic conversion of cyclic guanosine monophosphate (cGMP) to the inactive metabolite,5 -GMP. cGMP is an important activator of Protein Kinase G, a member of the AGC kinase familythat has been shown to inhibit TRPC6 activity and expression via direct phosphorylation of the threonine residue 69 in podocytes. PKG has also been shown to positively regulate the proteasome-mediate degradation of misfolded proteins, to promote proteostasis, to enhance autophagic flux and to reduce protein aggregation. PDE5 inhibitors have also been shown toupregulate the expression of peroxisome proliferator-activated receptor (PPAR ). PPAR is anuclear transcription factor and positive regulator of autophagy and lysosomal biogenesis via suppression of mTOR activity. Previous work showed that Sildenafil suppressed podocyteTRPC6 expression through the recruitment of PPAR to the TRPC6 promoter, demonstrating arole for the PPAR as a transcriptional repressor. Although these studies do not delineate thecontributions of PKG and PPAR signaling to the amelioration of podocyte injury caused by thecompound L899H / P924T variants, it is possible that the known cytoprotective actions of these molecules on proteostasis could underlie the beneficial effects observed on podocyte morphology and function. Although no direct influence of ARBs on proteostasis have been defined, the potential for additive or synergistic benefit from combined therapy on podocyte function cannot be excluded by these studies and warrant further investigation. Dysregulated TRPC6 activity can significantly contribute to many non-genetic forms of podocytopathy, such as diabetic nephropathy, membranous nephropathy, and minimal change disease. Consequently, deciphering the molecular mechanisms of TRPC6-mediated podocytopathy is a subject of intense scientific and clinical / translational interest. While the precise mechanisms of TRPC6-mediated injury across the spectrum of glomerular disease are largely undefined, aberrant calcium transport has been identified as a key feature of the disease- relevant physiologic disturbances in podocytes. In this study, novel compound C-terminal TPRC6 variants (L899H / P924T) were reported that induce the accumulation of podocin-enriched ER aggresomes, disrupts ER morphology, and impairs nephrin-podocin-TRPC6 trafficking to actin- based, terminal podocyte arborations. This is the first study to identify these features of TRPC6- mediated podocyte injury and explore the therapeutic utility of targeting TRPC6-induced alterations in podocyte proteostasis and suggest various pathogenic mechanisms for the effects of TRPC6 mutations. These findings expand the spectrum of TRPC6-mediated podocyte cellular injury and provide further evidence of the pleiotropic benefits of PDE5 inhibitors. What remains unknown are the specific pathomechanisms of L899H / P924T on TRPC6 interactions with nephrin and podocin, the role of impaired proteostasis in other forms of TRPC6-mediated podocytopathy, and the molecular underpinnings of PDE5 inhibitor-induced podocyte cytoprotection. Example 4 Derivation of Mature Podocytes from Human iPS Cells Human iPS cell-derived podocytes were induced as previously described. The use of all human iPS cell lines used in these studies was approved by the institutional review board and stem cell research oversight committee and obtained under approved material transfer agreements. The cell lines were subject to routine mycoplasma testing and found to be mycoplasma free at the time of study (Mycoplasma PCR Detection Kit, abm, G238). The PGP-1 human iPS cell line was obtained from the Personal Genome Project at Harvard University. The DU-11 human iPS cell line was generated at the Duke University iPSC Core Facility. Tosummarize, iPS cells were cultured to 70% con uence on Matrigel-coated plates (BDBiosciences, 354277) and were then dissociated with warm enzyme-free dissociation bu er(Gibco, 13150 016). Colonies were scraped using a cell-lifter (Fischer Scientific; 08100240) andthen pelleted by centrifuging the cell suspension twice at 200g for 5 min each in Advanced DMEM / F12 (Gibco; 12634010). The second centrifugation is an important step that helps to remove residual Matrigel, which can affect differentiation. Following centrifugation, iPS cells were resuspended in mesoderm induction media, consisting of DMEM / F12 with GlutaMax (Gibco;10565042) supplemented with 100ng mL 1 Activin A (Invitrogen; PHC9564), 3 M CHIR99021(Stemgent; 04-0004), 10 M Y27632 (TOCRIS; 1254), and 1x B27 serum-free supplement (Gibco;17504044). A cell count was performed, and the cell suspension (consisting of mostly singelized iPS cells and small colonies) was plated at a seeding density of 100,000 cells per well of a tissue culture treated 12-well plate (VWR; 10062-894) pre-coated with Laminin 511-E8 (Takara; T304). Cells were then cultured in mesoderm induction induction media for 2 days with daily media change. At the end of this 2-day induction period, cells were transitioned to intermediatemesoderm media, (containing DMEM / F12 with GlutaMax supplemented with 100 ng mL 1 BMP7(Invitrogen; Phc9543), 3 M CHIR99021, and 1x B27 serum-free supplement) for a minimum of14 days, with daily media change. After 14 days in intermediate mesoderm induction, the cellscan either be cryopreserved or di erentiated to podocytes. To initiate podocyte induction, theintermediate mesoderm cells were dissociated using 0.0505% trypsin-EDTA (Gibco; 53425300- 054), and then plated at a seeding density of 100,000 cells per well of a laminin-511-E8-coated 12-well plate. The resultant cultures were fed daily for the next five days with a podocyte inductionmedium consisting of Advanced DMEM / F12 with GlutaMax supplemented with 100 ng mL 1 ofBMP7, 100 ng mL 1 of Activin A, 50 ng mL 1 of VEGF (Gibco; PHC9394), 3 M CHIR99021, 1×B27 serum-free supplement, and 0.1 M all-trans retinoic acid (Stem Cell Technologies; 72262). It is recommended to keep this medium protected from light using aluminum foil, given the photo- instability of retinoic acid. Following the 5-day podocyte induction period, mature podocytes weretransitioned to a maintenance media (Culture Boost-R, Cell Systems; 4Z0 500) for a minimumof 11 day prior to experimentation and drug treatments. Derivation of Endothelial Cells from Human iPS Cells Human iPS-derived endothelial cells were induced according to previously established methods and optimized for cell lines. Briefly, human iPS cells were dissociated and pelleted, and then resuspended in mTeSR1 medium and seeded at 45,000 cells / cm2onto Matrigel-coated 6- well plates. The next day, the medium was changed to N2B27 medium to induce lateral mesoderm cells and cultured without changing the medium for 3 days. On day 4, N2B27 medium was replaced with endothelial induction medium that was replenished every day for 3 days. On day 7, vascular endothelial cells were isolated via magnet-activated cell sorting (MACS, Militenyi Biotec). Briefly, cells were prepared incubated with Accutase, and the cell suspension diluted 1:1 with StemPro-34 (Gibco, 10639011) and pelleted via centrifugation. The cell pellet was washed with MACS buffer (Dulbecco’s phosphate buffered saline (DPBS, Gibco 14190144), 0.5% bovine serum albumin (BSA, Millipore Sigma A9418), 2 mM Ethylenediaminetetraacetic acid (EDTA, Thermo Scientific R1021)), and resuspended in 80 μL / 10 million cells MACS buffer and 20 μL / 10 million cells each of FcR blocking reagent (Militenyi Biotec, 130-059-901), CD31 Microbeads (Militenyi Biotec, 130-091-935), and CD144 Microbeads (Militenyi Biotec,130-097-857). Following 15 min incubation, cells were washed with MACS buffer and sorted on a QuadroMACS Separator (Militenyi Biotec, 130-091-051). CD31+ / CD144+ cells were expanded in conditioned medium diluted at a 1:1 ratio with StemPro-34 supplemented with 2 μg / mL heparin (StemCell Technologies, 07980) that was replaced every other day. At 90% confluence, the cells were passaged up to a 1:4 split in maintenance media (Culture Boost-R) or used for downstream experiments. Plasmid Construction The Biosettia pLV-RNAi vector system was used for shRNA mediated gene silencing studies. Briefly, plasmids targeting CTGF, Cyr61, and YAP were obtained by cloning the respective shRNA oligo (Table 2) into the pLV-TetO-GFP-puro cloning vector (Biosettia; SORTC01) according to manufacturer’s instruction. Plasmids for inducible YAP overexpression studies were purchased from Addgene (YAP: FUW-TetO-wtYAP, plasmid #84009; reverse tetracycline transactivator (rtTA), plasmid #19780; vector backbone: FUW-TetO-MCS, plasmid #84008). A YAP reporter plasmid, FUW-TetO-wtYAP-2A-mCherry, was generated using the Gibson cloning method. Briefly, the FUW-TetO-MCS backbone vector was digested using BamHI (NEB; R3136) and EcoRI (NEB; R3101S) YAP insert was amplified from FUW-TetO-wtYAP with the primers shown below. The 2A-mCherry insert (from Nanog-2A-mCherry, Addgene #59995) was amplified using the primers shown below. Table 2. Design of Inducible shRNA Targeting YAP, CTGF, Cyr61 Variant All Gene RefS Targe Pos GC s var SEQ ID Targ iants shRNA Sequence ID X7 AAAAGGAGTTGACGA NM_ 794 shCyr61Cyr61 0015 CDS (exo 42.1 11 Yes (1)GAAACAATTTGGATC CAAATTGTTTCTCGT 2 54 n 3) CAACTCC text text sequence. Abbreviations: CDS, protein coding sequence; UTR, untranslated region; pos, position. GC, percent GC content of the indicated sequence. Table 3. FUW-TetO-wtYAP and 2A-mCherry Primer Name DNA Sequence FUW-TetO-wtYAP FwdCAGGCTAGCTATCAGGATCCGCCACCATGGACTACAAGGACG5FUW-TetO-wtYAP RevCCCTTGCTCACCATTAACCATGTAAGAAAGCT62A-mCherry FwdGCTTTCTTACATGGTTAATGGTGAGCAAGGGCGAG72A-mCherry RevCCTTGAGGCCCAAGAGAATTCTTACTTGTACAGCTCGTCCATGCC8The amplified fragments were then combined with the digested FUW-TetO-MCS vector at a 2:1 molar ratio using the NEB Gibson Assembly Master Mix (NEB; E2611S), according to manufacturer’s instructions. Positive Gibson colonies were confirmed via PCR screen, and plasmids were extracted and purified using the Macherey-Nagel Nucleospin kit (740588.50) and sent for sequencing. The resulting construct, FUW-TetO-YAP-2A-mCherry, was then used as the backbone to generate inducible FUW-TetO-CTGF-2A-mCherry and FUW-TetO-Cyr61-2A- mCherry. The CTGF and Cyr61 inserts were amplified from iPS cell-derived podocyte cDNA (PGP-1 cell lines) with BamHI / PacI overhangs using the primers shown below. Table 4. CTGF and Primer Name SequenceCTGF BamHI Fwd CTGF PacI RevGTTAATTAATCATGCCATGTCTCCGTA10Cyr61 BamHI FwdCyr61 PacI Rev The resulting amplicons and backbone vector, FUW-TetO-YAP-2A-mCherry, were subject to digestion with BamHI (NEB; R0136S) and PacI (NEB; R0547S). The digested fragments containing compatible overhangs were then gel purified using the Takara PCR clean up and gel extraction kit (cat.no. 740609.250. Ligation was performed using the T4 Ligation kit from Promega (cat.no.M1801). The ligated constructs were then gel purified, extracted, and sent for sequencing. Plasmid maps for the constructs used in this study can be found in Supplementary Information. Lentiviral Preparation For lentiviral packaging, the HEK293T cell line was used, which was maintained in HG- DMEM (Thermo; 11965092) supplemented with 10% Tetracycline-approved FBS (Thermo; A4736401) and 1% Pen / Strep. Lentiviral particles were prepared by transiently transfecting HEK293T cells (80–90% confluent in T-75 flasks) with lentiviral transfer vectors (9.24 μg) together with packaging vectors pMD2-VSVG (Addgene plasmid 12259, 2.5 μg) psPAX2 (Addgene plasmid 12260, 6.16 μg) and pREV (Addgene plasmid 12253, 1.54 μg) using the Lipofectamine 3000 kit (Thermo; L3000015) according to manufacturer instructions for T-75 flasks, and in the absence of antibiotics. After 8 hours, HG-DMEMDMEM medium was changed. Lentiviral supernatants were collected at 24- and 48-hours post-transfection, after which the supernatants were filtered through 0.22-micron filters (Millipore; SCGP00525) and stored directly at 80°C; lentiviral supernatants were not concentrated for these experiments. Titer of the unconcentrated supernatants was estimated by PCR of integrated lentiviral DNA of HEK293T cells transduced with an empty vector (psPAX2, VSV-G, pRev) or HIV-1 p24 ELISA (Origene, TR30038).). Lentiviral-Mediated Inducible Gene Knockdown and Overexpression Studies For gene silencing studies, podocytes were transduced with lentiviral supernatants encoding inducible shRNAs at 50MOI for 48 hours. For overexpression studies, podocytes were transduced with a 1:1 vol / vol mix of rtTA and TetOn-inducible YAP, CTGF, or Cyr61 lentiviruses inthe presence of 5 μg mL 1 polybrene (Sigma; TR-1003-G) at 50MOI for 48 hours. For glomeruluschip studies, intermediate mesoderm cells were transduced at 20MOI with YAP lentivirus and polybrene for 48 hours. It was found that the addition of polybrene to the lentiviral media was important for obtaining high transduction efficiencies. For gene silencing studies and CTGF / Cyr61 overexpression studies, transgene expression was induced by supplementingCultureBoost-R media (Cell Systems; 4CB-500-R) with 2 μg mL 1 doxycycline hyclate (Sigma; 24390-14-5) for 4 days and media was refreshed every other day. For YAP overexpression, transgene expression was induced by culturing the cells in doxycycline-containing media for 11 Podocyte Injury Models Podocyte injury experiments were performed. To model drug-induced cytotoxicity and foot-process effacement, podocytes were treated with Adriamycin (ADR, LC Laboratories, D4000)at a concentration of 0.5 μg mL 1 for 24 hours in the plate or up to 3 days in the glomerulus chip.To model disrupted mechanobiology, podocytes were treated with Cytochalasin-D at 0.5 g mL 1to inhibit F-actin polymerization (Sigma-Aldrich; C8273) for 24 hours.Immuno uorescent Imaging and AnalysesCell fixation and immunolabeling were performed. Briefly, all primary antibody incubations were performed overnight at 4°C in a solution containing 0.125% Tween / PBS. The primary antibodies used in this study were mouse anti-YAP (Santa-Cruz; SC-101199, 1:100 dilution), rabbit anti-Vinculin (Thermo; PA5-29688, 1:100 dilution) rabbit anti-CTGF (CST; 86641S, 1:100 dilution), and rabbit anti-Cyr61 (CST; 14479S, 1:100 dilution), rabbit, anti-podocin (Abcam, ab50339), and rabbit anti-beta-tubulin III (Sigma-Aldrich, T2200). Secondary antibody incubations were carried out for 1 hour in the dark, at a dilution of 1:1000 in a solution containing 0.125% Tween / PBS. Secondary antibodies used in this study included donkey anti-rabbit Alexa fluor 594 (Life Technologies; A21207), donkey anti-rabbit Alexa fluor 488 (Life Technologies; A21206), and donkey anti-mouse Alexa fluor 488 (Life Technologies; A21202). All immunofluorescent images were captured at 4×, 10×, or 20× objectives using an EVOS-M7000 imaging system (Life Technologies; AMF7000). Podocyte F-actin was labeled with Phalloidin-594 (Thermo; A12381) following manufacturer’s instruction. For quantification of cell area and actin fiber density, F-actin labeled cells were imaged at 10× magnification using an EVOS-M7000 epifluorescent microscope (Life Technologies). Calculation of cell-spread was performed manually in FIJI. Assessment of F-actin fiber density was performed using the FIJI plugin FiloQuant. Briefly, FiloQuant threshold parameters was set to 25 to detect cell edges, and 50 to detect intracellular actin fibers. Repair cycles was set to 5. Cell Viability Assay Measurements of cell viability were preformed using a CCK-8 assay and per manufacturer’s direction. Briefly, the CCK-8 stock reagent was diluted 1:10 in CultureBoost-R media (Cell Systems; 4CB-500-R) and 0.8mLs of this solution was applied per well of 12 well plate and incubated at 37 °C for 2 hours. Following the incubation period, 100 μL of the supernatants were collected into 96 well plates (Nunc; 243656) and absorbance was read at 450 nm using a Tecan Spark multi-plate reader. Percent viability was obtained by normalizing ODs to their respective controls. Western Blot Analysis Western blot was performed. All primary antibodies were diluted in TBS-Tween (TBS-T) supplemented with 5% BSA and incubated overnight in a 4 °C cold room on a platform angle rocker. Primary antibodies used for western blot in this study were mouse anti-YAP / TAZ (Santa Cruz Biotech; Sc-101199, 1:500), rabbit anti-Vinculin (Thermo; PA5-29688; 1:2000 dilution), rabbit anti-alpha-actinin-4 (Abcam; ab108198, 1:1000), mouse anti-Synaptopodin (Santa Cruz; sc- 515842, 1:1000), rabbit anti-Myosin Va (Sigma-Aldrich, M4812), and mouse anti-GAPDH (Invitrogen; ma515738, 1:5000). Following overnight incubation, blots were washed 3× 5 minutes in TBS-T to remove any unbound antibody. Secondary antibodies were then incubated for 1 hour at room temperature in 5% Blotto / TBST on a platform angle rocker. Secondary antibodies were HRP-conjugated goat anti-mouse (Cell Signaling Tech; 7076, 1:5000) and HRP conjugated goat anti-rabbit (Cell Signaling Tech; 7074, 1:5000). Following incubation with secondary antibody, blots were washed 3× 5 minutes in TBST. Chemiluminescence was detected using the Super Signal West Femto kit (Thermo; 34094). Signal intensities were captured using a ChemiDoc imager and quantified using ImageLab software (Biorad). ELISA ELISAs were performed. To assess CTGF and Cyr61 expression in knockdown and overexpression constructs, cell conditioned supernatants were collected after 4 days of doxycycline treatment and stored at 80°C. Both the Human CTGF ELISA kit (Abcam; ab261851) and Cyr61 ELISA kit (Abcam; ab238267) were used according to the manufacturer’s protocol. The total concentration of secreted protein in the supernatants was interpolated from a standard curve using a 4PL fit in Graphpad Prism. If necessary (i.e. to control for loss of viability), the amount of secreted protein was then normalized to total protein collected from whole cell lysates using a BCA assay (Pierce; 23225). Gene Expression Analysis Gene expression studies were conducted. Reverse transcription and qPCR were performed using the Luna Universal One-Step RT-qPCR kit (NEB; E3005L). Gene expression was then quantified using a Quantstudio 396-well, 0.2 mL block instrument (Applied Biosystems) set to detect SYBR Green. The Thermo iCloud Relative Quantification Suite was used to compile and digitally analyze the resultant CT values. Fold change was calculated using the double delta CT method, in which samples were first internally normalized to either -actin or GAPDH housekeeping genes and then normalized to their respective controls either non-transduced or non-targeting). The sequences of the qPCR primers used in this study are shown in Table 5. Table 5. DNA Primers Glomerulus Chip Fabrication and Molecular Filtration Assay The glomerulus chip with an electrospun, silk fibroin membrane was fabricated. Briefly, a 5% silk fibroin aqueous solution (Advanced Biomatrix, 5154) was dialyzed against 90% polyethylene glycol (Sigma-Aldrich, P6667) aqueous solution in a Slide-A-Lyzer Dialysis Cassette (Thermo Fisher, 66380) for 22 hours at room temperature, and then reconstituted in ultrapure water to 8%. The solution was mixed with 10% poly(ethylene oxide) (Sigma-Aldrich, 189456) at a ratio of 3.9 silk fibroin : 1 poly(ethylene oxide) and loaded into a syringe (BD Biosciences, 14- 823-2A), followed by attachment with a 16-gauge blunt-tip needle (McMASTER-CARR, 7516A753) and loading onto a syringe pump (Chemyx, F10071) with a high-voltage power supply (Gamma High Voltage Research, ES100P-20W). Silk fibroin sheets were electrospun at 11kV onto nonstick aluminum foil at a distance of 20cm for 10 minutes. PDMS chips were manufactured by mixing Sylgard 184 elastomer and curing agent (Ellsworth, 184 SIL ELAST KIT 0.5KG) at a 10:1 ratio in custom-designed 3D-printed molds for 4 hours at 65 °C. To assemble the chips, 5 cm × 0.5 cm membranes were first immersed in 90% methanol for 20 minutes and then air-dried overnight at room temperature, followed by water immersion for 48 hours. The silk fibroin membranes were transferred onto a bottom PDMS channel, prestamped with spin-coated PDMS prepolymer mixture and cured at room temperature overnight. The cured silk fibroin + PDMS bottom channel is then stamped against a spin-coated (with PDMS glue) PDMS top channel and cured overnight at room temperature. This step was then repeated with a 48-hour, room temperature cure. A new top PDMS channel and the silk fibroin + PDMS bottom channel were treated with oxygen plasma at 50W and 0.8mbar for 60s (Emitech, K-1050X). The two channels were pressed together to bond, followed by the addition of laminin-511 solution at 25ug / mL (Biolamina, LN511) into both top and bottom channels of the newly formed chip and incubated overnight at 37 °C. Cells were prepared according to the methods mentioned above. Endothelial cells were seeded into the vascular channel at a density of 147,000 cells / chip in Culture Boost-R, allowed to adhere to the bottom of the membrane (upside-down) for 4 hours, before rinsing and incubating overnight at 37 °C. The next day, transduced and non-transduced IM cells were seeded at a density of 60,000 cells / chip, respectively, in IM induction media, allowed to adhere to the top of the membrane for 4 hours, before rinsing and incubating overnight at 37 °C. The following day, all channels were rinsed, and IM media was replaced by podocyte induction media then incubated overnight. Subsequently, chips were connected to the peristaltic pump (Ismatec Ip Digital peristaltic pump, Cole-Palmer, EW-28001-32) at a flow rate of 4.09 μL / min and replenished every 2 days for up to 5 days. Chips were used for subsequent disease modelling and transgene activation as mentioned above. The molecular filtration assay (Albumin clearance) was carried out according to previously established methods. Briefly, vascular channel media was supplemented with 100 g / mL albumin conjugated to Alexa Fluor 555 (Thermo, A34786) and perfused for 6 hours. Spent medium was collected form the urinary channel outlet and fluorescence intensity was measures using a SpectraMax Fluorescent Plate reader (SpectraMax i3x, Molecular Devices. Albumin urinary clearance was calculated using the following equation: The molecular filtration assay (Albumin clearance) was carried out according to established methods. Briefly, vascular channel media was supplemented with 100 g / mL albumin conjugated to Alexa Fluor 555 (Thermo, A34786) and perfused for 6 hours. Spent medium was collected form the urinary channel outlet and fluorescence intensity was measures using a SpectraMax Fluorescent Plate reader (SpectraMax i3x, Molecular Devices. Albumin urinary clearance was calculated using the following equation: ([U] × UV) / [P]), where [U] is urinary concentration of albumin, UV = volume of media collected from the urinary channel outlet reservoir, and [P] = dosing concentration in the microvascular channel (100 g / mL).Quanti cation and Statistical AnalysisUnless otherwise indicated, all experiments were performed as three independent replicates. For F-actin fiber density analysis and analysis of shRNA effects on cell viability, a one- way ANOVA with post-hoc Dunnett’s test compared to control was used to test for statisticalsigni cance. For testing effects of ADR treatment on YAP, CTGF, and Cyr61 overexpressing cells,a two-way ANOVA with multiple comparisons and post-hoc Sidak’s test was performed to determine significance. For shRNA qPCR validation studies, student’s unpaired t-tests were performed assuming normality. For Western blot, a One-way ANOVA with multiple comparisons to control and post-hoc Dunnett’s test was performed. For all studies, only p values of 0.05 orlower were considered statistically signi cant (p > 0.05 [ns, not signi cant], p < 0.05 [*], p < 0.01[**], p < 0.001 [***], p < 0.0001 [****]). Statistical testing for all studies was performed using the GraphPad Prism 9 software package for Windows. Development of a Lentivirus-Mediated Inducible shRNA Knockdown Assay It was hypothesized that CTGF, Cyr61, and YAP play an indispensable role in podocyte homeostasis and may serve as viable therapeutic targets to prevent injury or repair the cell following acute injury. To probe the effect of target gene inhibition on podocyte biology, treating the cells with Verteporfin was initially considered, a small molecule inhibitor of YAP. Verteporfin inhibits signaling of YAP (and consequently, expression of its target genes) by either upregulating 14-3-3 protein, which sequesters YAP in the cytoplasm and targets it for ubiquitin-mediated degradation, or by directly interfering with the YAP / TEAD binding. However, some studies have also demonstrated that Verteporfin is not a selective inhibitor of YAP and can have several off- target, cytotoxic effects. To overcome this issue, a series of tetracycline (tet)-inducible gene constructs was developed that would allow the interrogation of the expression of CTGF, Cyr61, and YAP reliably and specifically without compromising cell viability. Since post-mitotic cells are often resistant to DNA-transfection methods, these constructs were packaged and delivered in lentivirus, pseudo-typed with the protein G of vesicular stomatitis virus, VSV-G, to stably and efficiently transduce post-mitotic cells. A series of tet-inducible short hairpin RNAs (shRNA) were designed to conditionally knockdown expression of CTGF, Cyr61, or YAP, respectively. The RefSeq identifiers, target position, variant specificity, sequences, and plasmid map for all the shRNA constructs tested in this study are provided in Table 2 and FIG.38, and the summary of the experimental workflow is schematized in FIG. 32A as follows: Human iPS cell-derived podocytes were transduced with lentiviruses (LVs) encoding tetracycline-inducible constructs for 48 hours in the presence of polybrene. To activate transgene expression, 2μg mL-1 doxycycline was administered for 4 days. All shRNAs gave approximately 90% transcriptional knockdown of their intended target compared to negative (non-transduced) and positive / vehicle (non-targeting) controls – now termed shCTGF, shCyr61, and shYAP (FIG. 32B). To determine the efficiency and specificity of the knockdown at the protein level, CTGF and Cyr61 were first probed via ELISA because their matricellular secretion is most correlated to their function and they share 40-60% of their primary structure. ELISA analysis of cell-supernatants revealed that there is a significant reduction of the proteins via their respective shRNA but knockdown of CTGF does not affect secreted levels of Cyr61, and vice versa (FIG.32C). Since decreased levels of both CTGF and Cyr61 were routinely observed in the injured podocytes, it was anticipated that this assay system would allow the probing of possible synergistic effects of CTGF and Cyr61 double knockdown which has not been systematically studied. It was confirmed that LVs can be co-delivered to yield a CTGF+Cyr61 double knockdown at the protein level (FIG. 32C). Furthermore, the shYAP construct yields potent protein-level knockdown of total YAP, approximately 90% (FIG. 32D–E). This effect also translated to a significant reduction in nuclear staining intensity (FIG. 32F–G), where the protein is established to be transcriptionally-active. Taken together, the requisite toolset to interrogate the lack of biological functions for CTGF, Cyr61, and YAP in human iPS cell-derived podocytes was designed and validated. Next, this toolset was applied to ascertain how inhibited expression of the genes of interest – CTGF, Cyr61, or YAP – could affect podocyte health. Inhibition of CTGF, Cyr61, or YAP Disrupts Podocyte Morphology and Viability, and Increases Susceptibility to Adriamycin Injury Which aspects of podocyte biology are directed by YAP, CTGF, and Cyr61 were sought to be determined. Given that nuclear YAP is suggested to be important for podocyte cell survival, and CTGF / Cyr61 influence cell-matrix adhesion, it was hypothesized that depletion of these proteins would disrupt podocyte cell morphology via cytoskeletal disassembly and subsequently reduce cell viability. In fact, F-actin is an important determinant of podocyte function and is often disrupted in glomerular disease. Indeed, a strikingly similar phenotype was observed across all knockdowns: F-actin labeling revealed that shCTGF, shCyr61, shCTGF+shCyr61 double knockdown, and shYAP resulted in a significant reduction in cell spread and area compared to control conditions (FIG.33A–B). Given these drastic changes in cell morphology, it was evaluated whether any of the knockdown conditions affected the assembly of F-actin. Surprisingly, it was found that only the double knockdown of CTGF / Cyr61 induced a statistically significant reduction in actin fiber density (FIG.33C). As individual proteins, YAP and its target genes can significantly perturb the structural integrity of the actin cytoskeleton, but only the double knockdown of CTGF and Cyr61 disrupts the molecular composition. Additionally, while it is generally accepted that expression of CTGF / Cyr61 can be mechanically regulated37, whether CTGF / Cyr61 regulate cytoskeletal assembly in turn is less understood. Thus, these data raise the intriguing possibility of a bi-directional feedback loop between CTGF / Cyr61 signaling and podocyte cytoskeletal dynamics. Whether the knockdown targets would affect general cell viability or sensitivity to Adriamycin (ADR) was the studied. Interestingly, it was found that only YAP knockdown induced a significant reduction in cell viability, as measured by CCK-8 assay (FIG. 33D). Next, it was examined whether CTGF+Cyr61 double knockdown or YAP knockdown further sensitized podocytes to drug-induced injury compared to the ADR-injured, non-targeting control. As expected, all knockdown conditions had significant reductions in cell viability following ADR injury (FIG. 33E). Moreover, it was found that by knocking down either CTGF and Cyr61 or YAP, podocytes exhibited increased sensitivity to ADR treatment compared to the ADR-injured control, 11% and 14%, respectively (FIG.33F). It was reasoned that even though CTGF+Cyr61 double knockdown did not inherently affect cell viability by itself, their depletion may render cells more sensitive to ADR injury through impaired cell-matrix adhesion. In podocytes, proper focal adhesion assembly is thought to be important to glomerular function, since podocyte detachment is a leading factor in the progression of glomerular disease. Thus, the expression of the focal adhesion marker, vinculin was examined, which is a force sensitive, cytoskeleton-associated protein that regulates cell-matrix junctions by recruiting actin remodeling proteins to the junctional complex and stimulating actin polymerization. While no disruption to total Vinculin protein expression was found via western blot (FIG.39A–B), immunostaining revealed that CTGF+Cyr61 double knockdown appear to express fewer large punctate vinculin positive complexes compared to the shGFP control (FIG.33F), indicating a potential disruption to the localization of this protein to mature focal adhesion complexes. However, intracellular expression of lineage restricted or actin-bound markers such as podocin, synaptopodin, and alpha-actinin 4 were unaffected (FIG. 39A–B). Therefore, these data suggest that CTGF+Cyr61 double knockdown may interfere with localization of focal adhesion proteins such as vinculin, but not necessarily expression. Taken together, these results highlight that individually, each target (CTGF, Cyr61, and YAP) is an important determinant of podocyte cytoskeletal integrity. Podocyte homeostasis is significantly perturbed through CTGF and Cyr61 depletion, either directly via double knockdown or indirectly through disruption of upstream YAP signaling. The early detection of disrupted CTGF and Cyr61 secretion may serve as a robust surrogate to detect declining podocyte health and susceptibility to further injury, potentially serving as an indicator to progressive proteinuria. Development and Characterization of the Lentiviral-Mediated, Tetracycline-Inducible System to Conditionally Overexpress CTGF, Cyr61, and YAP Given these findings that YAP, CTGF, and Cyr61 expression appear to heavily influence podocyte morphology and susceptibility to Adriamycin-induced injury, it was studied whether overexpressing these proteins could serve as endogenous therapeutic targets. LV-packaged, tet- inducible constructs were first developed that would enable the conditional overexpression of YAP, CTGF, and Cyr61 modified with a 2A-mcherry reporter (FIG. 40). In the same manner as the shRNA assays, podocytes were transduced with LVs for 48 hours in the presence of polybrenefollowed by 1-4 days of 2 μg mL 1 doxycycline administration, depending on the transgene (FIG.34A). A series of assays was then performed to validate the efficacy of the TetOn-system. It was confirmed that all the TetOn-constructs expressed their respective reporter protein, eGFP for the vehicle control and mcherry for the target genes (FIG. 34B). It was found that YAP was transcriptionally overexpressed by the TetOn-YAP system, approximately 17-fold respective to the control (FIG. 34C). This transcriptional activation translated to overexpression of total YAP protein, particularly through phosphorylated YAP, approximately 6-fold, respectively (FIG. 34D– E). A significant increase in YAP intensity through immunostain was also confirmed (FIG. 34F and FIG.41). The YAP immunostain supported the western blot, suggesting a significant increase in overall YAP immunostaining intensity per field of view. The TetOn-CTGF- and TetOn-Cyr61- constructs enabled transcriptional overexpression of their respective transgenes at approximately 27- and 13-fold increase, respective to the control (FIG.34G). Interestingly, a significant increase in transcriptional expression of CTGF and overall higher expression of Cyr61 (though not statistically significant) in YAP-overexpression podocytes was detected (FIG.34H). It was then found that both CTGF and Cyr61 had higher nuclear intensity in YAP-overexpressing podocytes compared to the control, (FIG.34I), but there was not a significant increase in secreted levels of these proteins outside of their respective transgene group, TetOn-CTGF and -Cyr61 (FIG.34J). This result is likely due to the increase in phosphorylated YAP which is not transcriptionally active and therefore may not confer an increase in YAP target genes. In addition to the gene knockdown analyses via LV-packaged shRNA, an appropriate system has now been developed and validated to conditionally overexpress CTGF, Cyr61, and YAP using a LV-packaged TetOn-system in the human iPS cell-derived podocytes. Given their roles in maintaining podocyte structure and function, this system was applied to explore the therapeutic potential of CTGF, Cyr61, and YAP modulation for the prevention of injury or restoration of homeostasis following injury. YAP Overexpression (not CTGF or Cyr61) Regulates Cell Size, Lineage-Restricted Marker Expression to Prevent Insult from Adriamycin The therapeutic potential of the target genes was explored by assessing whether CTGF-, Cyr61-, or YAP-overexpressing podocytes were protected from 24 hours treatment with 0.5 μgmL 1 ADR. It was found that neither overexpression of CTGF nor Cyr61 could protect againstADR-induced injury (FIG.35A). It was speculated that this could be a case of diminishing returns, in which basal level of CTGF / Cyr61 already confers maximal protection in this model. This would be in line with the findings that their (individual or joint) depletion disrupts podocyte morphology and the joint depletion exacerbates podocyte sensitivity cells to Adriamycin. In contrast, it was found that TetOn-YAP-podocytes did confer statistically significant protection from ADR-induced injury (FIG.35A), suggesting that modulating YAP expression may have therapeutic value that is independent of CTGF / Cyr61 signaling. Therefore, subsequent studies aimed to unveil the mechanisms to support his observation. It was observed that some TetOn-YAP-podocytes grew much larger compared to control cells. As such, it was initially anticipated that YAP overexpression in mature podocytes might have induce...

Claims

CLAIMS What is claimed:

1. A biomimetic microfluidic device, comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane.

2. The biomimetic microfluidic device of claim 1, wherein the polymeric base material comprises polydimethylsiloxane (PDMS), polycarbonate (PC), poly-methyl-meta-acrylate (PMMA), cyclic olefin copolymer (COC), polyimide, or combinations thereof.

3. The biomimetic microfluidic device of claim 1, wherein the electrospun silk fibroin membrane further comprises polyethylene oxide (PEO).

4. The biomimetic microfluidic device of claim 1, further comprising a pump fluidly connected to the first channel and the second channel and configured to pump fluids through the first channel and the second channel.

5. The biomimetic microfluidic device of claim 1, wherein the electrospun silk fibroin membrane further comprises a laminin coating on the first surface, the second surface, or a combination thereof.

6. The biomimetic microfluidic device of claim 1, wherein the electrospun silk fibroin membrane comprises a thickness of at least about 0.1 μm.

7. The biomimetic microfluidic device of claim 1, wherein the electrospun silk fibroin membrane comprises a thickness of about 0.1 μm to about 100 μm.

8. The biomimetic microfluidic device of claim 7, wherein the electrospun silk fibroin membrane comprises a thickness of about 0.5 μm to about 7.5 μm.

9. The biomimetic microfluidic device of claim 1, wherein the electrospun silk fibroin membrane is porous and comprises a porosity of about 1% to about 70%.

10. The biomimetic microfluidic device of claim 9, wherein the electrospun silk fibroin membrane is porous and comprises a porosity of about 40% to about 70%.

11. The biomimetic microfluidic device of claim 1, wherein the electrospun silk fibroin membrane comprises a Young’s modulus of about 25 kPa to about 100 kPa.

12. The biomimetic microfluidic device of claim 1, wherein the population of cells is cultured on the first surface of the electrospun silk fibroin membrane, the second surface of the electrospun silk fibroin membrane, or a combination thereof.

13. The biomimetic microfluidic device of claim 1, wherein the population of cells comprises induced pluripotent stem cells (iPSCs), embryonic stem cells (ESCs), kidney glomerular cells, podocytes, intermediate mesoderm (IM) cells, mesoderm cells, endothelial cells, embryoid bodies, organoids, spheroids, or combinations thereof.

14. The biomimetic microfluidic device of claim 13, wherein podocytes are cultured on the first surface of the electrospun silk fibroin membrane and endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane.

15. The biomimetic microfluidic device of claim 14, wherein the endothelial cells comprise fenestrated endothelial cells.

16. The biomimetic microfluidic device of claim 1, wherein the electrospun silk fibroin membrane mimics a kidney glomerular basement membrane.

17. The biomimetic microfluidic device of claim 1, wherein the first channel mimics a urinary compartment of a kidney glomerulus and the second channel mimics a vascular compartment of a kidney glomerulus.

18. The biomimetic microfluidic device of claim 1, wherein the biomimetic microfluidic device is a kidney glomerulus organ-on-chip device.

19. The biomimetic microfluidic device of claim 1, wherein the biomimetic microfluidic device is part of an extracorporeal dialysis system.

20. A method of filtering a sample, the method comprising: applying a sample to a biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; wherein podocytes are cultured on the first surface of the electrospun silk fibroin membrane and endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane; and perfusing one or more fluids through the biomimetic microfluidic device.

21. The method of claim 20, further comprising collecting a filtered fraction of the sample.

22. The method of claim 20, wherein the podocytes are induced pluripotent stem cell (iPSC)- derived podocytes.

23. The method of claim 20, wherein the sample is a blood sample, a urine sample, or a combination thereof.

24. The method of claim 20, wherein the method filters one or more uremic toxins from the sample.

25. The method of claim 20, wherein the method filters one or more biological metabolites from the sample.

26. The method of claim 25, wherein the one or more biological metabolites comprise inulin, albumin, urea, -2-microglobumin (B2M), visfatin, creatinine, or combinations thereof.

27. The method of claim 20, wherein the method has an albumin clearance rate of less than about 6%.

28. The method of claim 20, wherein the method has a urea reduction ratio (URR) of at least about 75%.

29. A method of screening an agent for treatment of a kidney or glomerular disorder, the method comprising: applying an agent to a biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; wherein podocytes are cultured on the first surface of the electrospun silk fibroin membrane and endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane, and wherein the agent contacts the podocytes; perfusing one or more fluids through the biomimetic microfluidic device; anddetecting a response of the podocytes to the agent, the response being associated with a kidney or glomerular disorder.

30. The method of claim 29, wherein the podocytes are induced pluripotent stem cell (iPSC)- derived podocytes.

31. The method of claim 29, wherein the response associated with the kidney or glomerular disorder comprises altered podocyte gene expression, altered podocyte protein expression, altered podocyte albumin uptake, altered formation or extension of podocyte foot processes, or combinations thereof.

32. The method of claim 29, wherein the agent comprises a protein, a peptide, a nucleic acid, a small molecule, or combinations thereof.

33. The method of claim 29, wherein the kidney or glomerular disorder comprises podocyte injury, proteinuria, glomerulosclerosis, diabetic nephropathy, chemotherapy-related nephrotoxicity, a podocytopathy, or combinations thereof.

34. A method of generating fenestrated endothelial cells, the method comprising: seeding a population of podocytes or intermediate mesoderm (IM) cells and a population of endothelial cells in a biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; wherein the podocytes or IM cells are cultured on the first surface of the electrospun silk fibroin membrane and the endothelial cells are cultured on the second surface of the electrospun silk fibroin membrane; andperfusing one or more fluids through the biomimetic microfluidic device to generate the fenestrated endothelial cells.

35. A kit for filtering a sample, comprising: a biomimetic microfluidic device, comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane; optionally, pumps, tubing, syringes, and containers for applying fluids to the biomimetic microfluidic device; optionally, cell culture buffers and reagents; and optionally, packaging, labels, or instructions for use.

36. Use of a biomimetic microfluidic device for filtering a sample, the biomimetic microfluidic device comprising: a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane.

37. Use of a biomimetic microfluidic device for screening an agent for treatment of a kidney or glomerular disorder, the biomimetic microfluidic device comprising:a polymeric base material; an electrospun silk fibroin membrane embedded in the polymeric base material, the electrospun silk fibroin membrane comprising a first surface and a second surface, the first surface and the second surface being configured for culturing a population of cells; and a first channel and a second channel separated by the electrospun silk fibroin membrane, the first channel being in fluid communication with the first surface of the electrospun silk fibroin membrane, and the second channel being in fluid communication with the second surface of the electrospun silk fibroin membrane.

Images