Abstract BACKGROUND AND AIMS Diabetic nephropathy (DN), chronic loss of kidney function due to diabetes mellitus, is a common complication in diabetes which currently lacks effective long-term treatment. Microalbuminuria is now the earliest marker for DN, thereby disregarding renal changes that occur before this first indication of decreased kidney function. For better diagnosis and treatment strategies, it is important to study diabetes induced early renal changes. Renal-specific changes of the lipid metabolism could serve as an indicator for early kidney damage. Mass spectrometry imaging (MSI) allows to study the lipid metabolism in the context of tissue histology without the need for labeling (i.e. antibody or radioactive). Existing literature applying this technique focuses on later stages of diabetic kidney disease, thereby overlooking its potential to serve in an earlier stage of diagnosis. The aim of this project is to identify renal cell-specific lipid changes using MSI which could serve as damage markers for early diabetes induced renal pathology. METHOD To study diabetic renal changes, apolipoprotein E-knockout mice were treated with streptozotocin (STZ) and put on an enriched cholesterol diet to induce diabetes. After 12 weeks, both control (n = 4) and diabetic (n = 4) mice were sacrificed and kidneys were harvested for immunohistochemical and matrix assisted laser desorption/ionization (MALDI) MSI. Post-MSI immunofluorescent staining in combination with tissue morphology was used to identify different renal cell types. Analysis of MALDI-MSI data allows us to spatially segment the results per cell type for further statistical analysis of the metabolic MSI data (Figure 1A). RESULTS Two weeks after STZ induction, blood glucose levels of the diabetic mice were significantly elevated compared to control. Spatial segmentation analysis of the MSI data revealed that in both groups different renal cell types could clearly be distinguished based on their metabolic profile. Using the immunofluorescence-based cell type annotation, we zoomed in on various morphological regions of the kidney to specifically test for changes in lipid profiles. Here, the proximal tubular cells in the cortex and the outer stripe of outer medulla had the highest number of significantly changed lipids (Figure 1B). Focusing the analysis on these two groups of cells revealed specific molecular signatures to be discriminative between diabetes and control (Figure 1C). CONCLUSION Using MSI, we were able to identify renal cell type-specific differentially expressed lipids in diabetic mice compared to the healthy control group. The proximal tubular cells in the cortex and outer stripe of the outer medulla had the most altered lipid composition. These molecular signatures might serve as an early indicator of diabetes induced renal changes, thereby opening up a potential window for treatment before the kidney is damaged beyond repair.
Read full abstract