Graduate student University at Buffalo Amherst, New York, United States
Objectives: There is currently no quantitative systems pharmacology (QSP) model for diabetes-induced glomerular fibrosis, an important marker of progression in diabetic kidney disease. In this project, our aim was to develop a QSP model that describes the mechanism of how high glucose in the blood causes glomerular fibrosis in the diabetic mice.
Methods: A previous QSP model of inflammation and fibrosis in lupus was used as a starting point to construct our QSP model of glomerular fibrosis in the diabetic condition [1]. A priori our questions were: • Are each mechanistic assumptions in the lupus model relevant to glomerular fibrosis in diabetes? • What additional mechanisms need to be added to adapt model to the diabetic case? • Which parameters, if any, are translatable to our model? Mechanisms were only incorporated into the model if there were in vitro studies showing that a specific process could be occurring in the diabetic condition in glomerular cells or within glomeruli of diabetic mice. In the end, we added high glucose induced formation of inflammatory advanced glycation end products (AGEs) inducing the recruitment of macrophages. Similar to the lupus model, the recruitment of macrophages induces the activation of the resident fibroblasts, in our case mesangial cells, to a phenotype that is known to be the source of excess collagen, the accumulation of which results in glomerular fibrosis. Additionally, recruitment and activation of cell types were mediated through growth factors and cytokines such as monocyte chemoattractant protein (MCP) and transforming growth factor – β (TGF-β). Data from in vitro and in vivo diabetic model studies were then used to calibrate 19 model parameters alongside 13 parameters that were retained from the previous model [1]. Model equations followed mass action kinetics to describe production, death and degradation of cells and biomolecules, and Michaelis Menten and Hill function kinetics for biomolecule mediated recruitment and activation of cells.
Results: Our model predicted that immediate regulation of blood glucose control does not lead to immediate amelioration of glomerular fibrosis. Immediate regulation of blood glucose down to healthy levels results in 0% reduction in collagen content over the span of 24 weeks and only ~20% reduction over a span of 3.5 years. This same trend is observed with inhibition of AGEs. However, enhanced degradation of AGEs results in the slowing down of collagen accumulation and complete return to baseline of collagen content in the span of 24 weeks.
Conclusions: Our model suggests that degradation of AGEs would accelerate the recovery for glomerular fibrosis and posits a mechanism for why certain approaches to treating glomerular fibrosis may not be effective. Additionally, our QSP model provides an initial model from which further modifications and analysis can be made to better understand the complexity of diabetes-induced glomerular fibrosis.
Citations: [1] Hao et al. Mathematical model of renal fibrosis. Proc Natl Acad Sci. 2014;111(39)
“The results in this abstract have been previously presented in part at the AIChE Conference in Phoenix, Arizona in November 2022”
