Multi Kidney Disease Modelling using hiPSC-Derived Biomimetic Glomerulus-on-a-Chips

Introduction: : Chronic kidney disease (CKD) is characterized by the progressive decline of glomerular filtration function, primarily driven by podocyte injury, including foot process effacement and detachment [1]. Podocytes are highly specialized, terminally differentiated visceral epithelial cells with complex arborized structures that are essential for maintaining the glomerular filtration barrier [2]. Current therapeutic strategies for CKD primarily focus on managing systemic risk factors such as hypertension and hyperglycemia [3,4]. However, due to the asymptomatic nature of early-stage CKD and the limited regenerative capacity of post-mitotic podocytes, patients frequently present with irreversible kidney damage [3]. These limitations highlight the need for therapeutic approaches that directly target podocyte injury and promote renal repair. Our laboratory has developed a biomimetic glomerulus-on-a-chip platform that integrates patient-derived induced pluripotent stem cell (iPSC) podocytes and endothelial cells within a two-channel microfluidic device, separated by a silk fibroin electrospun membrane that mimics the glomerular basement membrane [5]. This system has enabled us to model Adriamycin-induced glomerular injury and quantitatively assess filtration barrier damage by measuring the differential clearance of albumin and inulin [5]. In the present study, we aim to expand this platform to perform parallel multi-kidney disease modeling using human glomerulus-on-a-chip devices in order to identify potential shared pathways of podocyte injury and early biomarkers of CKD. Additionally, we investigate a novel gene therapy approach for Adriamycin-induced injury by overexpressing Yes-associated protein (YAP) post-treatment to evaluate its potential regenerative and cytoprotective effects, paving the way for novel molecular therapies that would target podocytes injury in early stage CKD.

Materials and

Methods: : Silk fibroin glomerulus chips were manufactured based on a previously established protocol (Mou, Science Advances 2024). Human iPSC-derived endothelial cells and intermediate mesoderm cells were seeded into the vascular and urinary channels, followed by a five-day podocyte induction protocol to obtain the final glomerulus-on-chip device (Figure 1).
The following treatments were used for injury modeling: (1) Adriamycin (0.5 μg/mL), (2) lipopolysaccharide (1 μg/mL), (3) pamidronate (100 nM), (4) high glucose (25 mM), and (5) high glucose + inflammatory cytokines (TNFα (1 ng/mL) + IL-6 (1 ng/mL)). Media containing each treatment was added to the vascular channel of the organ chip, and damage to the glomerular filtration barrier was assessed by filtration assay. For the filtration assay, fluorophore-conjugated albumin (100 μg/mL) and inulin (10 μg/mL) were added to the vascular channel for 6 hours. Outflow from the urinary channel was collected and analyzed using a microplate reader at various timepoints. Percent clearance in the outflow was calculated by generating a standard curve from serial dilutions of the original inlet concentration and fitting each urinary outlet readout to the standard curve.
For YAP overexpression, intermediate mesoderm cells were transduced at an MOI of 20 with YAP lentivirus and polybrene prior to seeding into the chip. After Adriamycin treatment, the Tet-On system was induced with a one-day Doxycycline treatment. Podocyte markers (Nephrin and Podocin) as well as YAP expression levels were visualized using immunofluorescence.

Results, Conclusions, and Discussions:: Multi-kidney disease modeling experiments reflected damage to the filtration barrier through increased albumin and inulin clearance percentages, consistent with previous 2D and in vivo studies [6]. In the Adriamycin group, albumin clearance increased from 5% to 12.5% between 2 and 5 days of treatment. In pamidronate-induced injury, a 10% increase was observed as early as day 2, rising above 20% by day 5. For the hyperglycemia model, high glucose alone did not induce significant damage until day 14 (data not shown). However, with inflammatory cytokines, a significant 20% increase in albumin clearance was observed by day 2, with continued damage through day 5, indicating that a diabetic-like condition with inflammation is necessary for hyperglycemia-induced glomerular injury. No significant increase in albumin clearance was observed in the LPS treatment group up to 14 days (Figure 2b). Inulin filtration levels followed similar trends (Figure 2c).
We first validated the YAP overexpression system in the chip by demonstrating a significant increase in YAP fluorescence levels compared to untreated controls, with no accompanying filtration damage (Figure 3a,c). When targeting Adriamycin-induced podocyte injury using YAP overexpression, we found a similar 10% increase in albumin filtration after 3 days of Adriamycin treatment, and prolonged culture of these chips in normal media showed extended damage to the filtration barrier even after Adriamycin removal. Doxycycline-induced YAP overexpression post-Adriamycin treatment ameliorated the exacerbation of filtration barrier damage by maintaining albumin and inulin filtration levels, which was reflected in an overall preservation of cell confluency compared to the Adriamycin group without YAP overexpression (Figure 3b,c). Our study highlights the potential therapeutic value of modulating YAP as a podocyte injury target, which has been shown to be involved in key podocyte injury processes in multiple previous studies from our lab [6,7]. The capability of ameliorating ongoing damage after Adriamycin may result from the enlarged podocyte cell bodies observed after YAP overexpression in 2D culture, allowing the remaining podocytes after Adriamycin treatment to cover a larger area of the filtration barrier and prevent further damage. Next, we will perform transcriptomic analysis for disease models to identify additional therapeutic targets and early biomarkers.

Acknowledgements and/or References (Optional):: Reference:
1. Hostetter, T. H. (1995). Progression of renal disease and renal hypertrophy. Annual Review of Physiology, 57(1), 263–278. https://doi.org/10.1146/annurev.ph.57.030195.001403
2. Musah, S., Bhattacharya, R., & Himmelfarb, J. (2024). Kidney disease modeling with organoids and organs-on-chips. Annual Review of Biomedical Engineering, 26, Article 26. https://doi.org/10.1146/annurev-bioeng-112123-015459
3. Lv, M., Zhang, Y., Zhang, X., Wang, H., & Chen, Y. (2015). Therapeutic strategies of diabetic nephropathy: Recent progress and future perspectives. Drug Discovery Today, 20(3), 332–346. https://doi.org/10.1016/j.drudis.2014.11.017
4. Hart, P. D., & Bakris, G. L. (2010). Hypertensive nephropathy: Prevention and treatment recommendations. Expert Opinion on Pharmacotherapy, 11(16), 2675–2686. https://doi.org/10.1517/14656566.2010.522146
5. Mou, X., Zhang, Y., Nguyen, M., Chen, Y., Tewari, R., & Musah, S. (2024). An ultrathin membrane mediates tissue-specific morphogenesis and barrier function in a human kidney chip. Science Advances, 10(23), eadn2689. https://doi.org/10.1126/sciadv.adn2689
6. Burt, M. A., Nelson, C. E., Chihab, M., Robertson, H., Song, L., & Musah, S. (2021). Adriamycin-induced podocyte injury disrupts the YAP-TEAD1 axis and downregulates Cyr61 and CTGF expression. ACS Chemical Biology, 17(12), 3341–3351. https://doi.org/10.1021/acschembio.1c00485
7. Zhang, Y., & Musah, S. (2024). Mechanosensitive differentiation of human iPS cell-derived podocytes. Bioengineering, 11(10), 1038. https://doi.org/10.3390/bioengineering11101038
Acknowledgement: This research was funded by the NIH Director’s New Innovator Grant (Award Number DP2DK139544) awarded to Samira Musah.