CAREER: Multiscale Modeling of a Virtual Kidney during the Onset and Progression of Diabetic Kidney Disease

Diabetic kidney disease (DKD) is a serious complication of both type 1 and type 2 diabetes and is the leading cause of kidney failure. Yet, it is still not clear how the many underlying chemical, physical, and biological processes interact to damage the kidneys during diabetes. It is challenging to monitor the damage to the kidneys inside a patient.

The regions of the kidney that are damaged are very small and are deep within the body. It takes a long time for irreversible damage to accumulate to the point where non-invasive urine samples contain detectable quantities of proteins that leaked through the kidneys. This Faculty Early Career Development Program (CAREER) project will connect several processes that have been shown individually to contribute to injury in kidneys due to diabetes into a sophisticated computer simulation.

This computational tool will aid in understanding how the processes interact and how diabetic kidney damage begins and changes over time. In the long-term, results from this project will help to predict the impacts of many competing factors on kidney health during diabetes management. The computer simulation produced in the project can also be used to test and optimize treatments to slow the progression of DKD.

The project also will involve a set of educational activities related to the scientific work. Several undergraduate and graduate students including women and underrepresented minorities will work with the PI to conduct the research and educational activities. Educational modules related to the research will be delivered to a variety of groups including K-12 students, college students, and grandparents.

Physical models of kidney tissues will be 3D-printed and shared with students and educators. The activities will expose many students and members of the public to biomedical engineering and computational science through engaging scientific demonstrations and interactive experiences.

The principal investigator's long-term career goal is to develop multiscale computational models to enhance understanding of the mechanisms governing tissue remodeling and damage as a result of diseases and infections and to simulate the treatment of those conditions to improve human health. Toward this goal, this project will develop a novel computational approach for studying diabetic kidney disease (DKD) through a virtual kidney that can be used like a powerful, non-invasive microscope to look into the body to detect and monitor damage to the glomeruli (where most damage occurs) during the onset and progression of diabetic complications in the kidney.

The virtual kidney platform will use multiscale computational modeling to connect effects at different length scales from smaller to larger: inside cells, between adjacent cells, across a single glomerulus, and among collections of glomeruli interacting with other tissues in a kidney. A hybrid computational approach will take advantage of the benefits of stochastic differential equations to describe chemical species that react and interact in large quantities and of agent-based models to describe cells and chemical species that interact in small quantities or in qualitative up- or down-regulation fashions.

The Research Plan is organized under two aims. The FIRST AIM is to formulate cellular level mathematical models of biochemical cell signaling networks responsible for damage in each of the three main glomerular zones: a) podocytes, b) mesangial cells and mesangial matrix and c) endothelial cells and the glomerular basement membrane. Computational tools will be developed for importing biochemical networks, conducting uncertainty and sensitivity analyses, and validating model results for the three zones.

To assess validity, computational models will be compared to experimental evidence gleaned from the literature. The SECOND AIM is to build a virtual kidney computational model focusing on crosstalk, structural, and hemodynamic effects on the primary tissue compartments within multiple nephrons. Steps include: a) constructing a hybrid computational model for tissue level simulation of glomerular cells that combines information from reactions in single cells, connects multiple interacting cells and transports glucose and other molecules through the mesangium and the glomerular filtration barrier; b) connecting the glomerular models to a tubule and vessel centric-model for blood flow and filtration regulation and c) validating the virtual kidney model and simulating responses to stimuli.

A user-friendly interface will be created for specifying model input and stimuli, running the simulations, and visualizing results. Simulations of the virtual kidney model will be rigorously compared to pharmacological, physiological, and histological data, primarily from humans. After model validation to a subset of the data cases, in silico experiments will be conducted to predict the magnitudes and rates of change for glomerular injury outcomes including proteinuria and glomerular filtration rate (GFR) under DKD pathophysiological conditions.

The metric for success will be the accurate simulation of the classic trajectory of onset and progression of DKD subject to clinically relevant input and stimuli. Computational tools and code developed will be distributed openly via GitHub repositories, which may be of broad interest to the systems biology research community using SBML and/or CompuCell3D.

In summary, the project addresses the critical need to compile the multiple mechanistic processes that contribute to DKD onset and progression into a user-friendly systematic framework capable of taking the interconnected chemical, physical, and biological factors into account in a coupled fashion and in the appropriate magnitudes and sequences to make testable predictions.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.