#5904 COMPARATIVE TRANSCRIPTOMICS OF THREE CANONICAL MOUSE MODELS OF CHRONIC KIDNEY DISEASE

Abstract

Abstract Background and Aims A range of mouse models are available in preclinical chronic kidney disease (CKD) research but differ in their disease etiology and pathological hallmarks. To enable better selection of the mouse CKD model optimal for preclinical drug discovery studies, we characterized the kidney transcriptome signature of three well-established models of CKD induced by unilateral ureter obstruction (UUO), unilateral ischemic reperfusion injury (uIRI) and adenine-supplemented diet feeding (ADI). Method Male C57BL/6JRj mice were used in all studies. UUO or sham surgery was performed in 9 weeks old mice, which were terminated two weeks post-surgery. uIRI or sham surgery was performed in 10 weeks old mice terminated 6 weeks post-surgery. 12 weeks old mice received an adenine-supplemented diet or a control diet for 6 weeks. Endpoints included plasma biochemistry, kidney histology and RNA sequencing on kidney samples. Results Compared to corresponding controls, plasma urea was only increased in ADI mice. All three models presented with increased inflammation (F4/80, UUO>ADI>uIRI) and fibrosis (COl1A1, UUO = uIRI>ADI) in the histological analysis. Compared to corresponding controls, all three models demonstrated substantial increases in differential expressed genes (DEGs, UUO, n = 12,046; uIRI, n = 12,236; ADI, n = 11,468), with significant overlap in transcriptome signatures between the models. Gene expression markers of inflammation (e.g., Ccl2, Cd68 and Il1b), fibrogenesis (e.g., Tgfb1, Serpine1 and Col1a1) and kidney injury (e.g., Havcr1, Lcn2 and Spp1) supported histological findings in the three models. Interestingly, several molecular targets pursued in CKD drug discovery were significantly regulated, however with different directionality, in UUO (upregulated Atg7, Bax, Ednra/b, Gli2, Ifng, Shh; downregulated Glp1r, Nr3c2, Ptch1, Slc5a2), uIRI (upregulated Ednra, Gli2, Glp1r, Ifng, Nr3c2; downregulated Agt, Egfr, Slc5a2), and ADI mice (upregulated Agt, Atg7, Bax, Ednra; downregulated Ednrb, Egfr, Flt1, Nr3c2, Ptch1, Slc5a2). Conclusion UUO, uIRI and ADI mouse models demonstrate histopathological hallmarks of CKD, characterized by increased macrophage infiltration and fibrosis, which was corroborated by their individual transcriptome signatures. Despite similarities in histological phenotype, the models were distinguished based on kidney transcriptome changes, with several current drug targets regulated in a model-specific fashion. In conclusion, the current work will enable researchers to select the mouse model optimal for profiling individual preclinical test drugs with therapeutic potential in CKD. Also, our data may serve to enable further development of CKD mouse models with improved clinical translatability.