Optimized protocols for the microdissection of specific areas from archival tissues and the subsequent RNA analysis are needed but challenging due to RNA degradation and chemical modifications. The aim of this study was to present the most appropriate protocol for utilizing mouse FFPE kidney for laser capture microdissection and Nanostring gene expression analysis. We evaluated different section thicknesses (3, 5, 10 μm), 2 RNA extraction kits (Qiagen and Roche) and different H&E staining methods to optimize microdissection and RNA extraction from glomeruli and cortical tubules samples from FFPE mouse kidney. RNA quality and quantity were assessed via Nanodrop and Qubit. The protocol providing the best results consisted of 5 μm sections, a shorter protocol for H&E staining, and RNA extracted with the Roche kit. Higher Nanostring gene counts and lower qPCR cT significantly correlated with RNA concentrations measured with the Qubit, but not with measures obtained with the Nanodrop. The Nanostring data showed that none of the genes included in the panel was differentially expressed in the cortical tubule compartment compared to the whole kidney. However, 25 genes were differentially expressed in the glomerular compartment compared to the whole kidney. Our data showed that sufficient RNA can be extracted from small compartments like mouse renal glomeruli from archival FFPE tissue, and that whole kidney analysis does not accurately represent the transcriptome state of the glomeruli, which comprise only a small proportion of the overall kidney volume.