The spatially resolved transcriptome signatures of glomeruli in chronic kidney disease
Geremy Clair, Hasmik Soloyan, Paolo Cravedi, Andrea Angeletti, Fadi Salem, Laith Al-Rabadi, Roger E. De Filippo, Stefano Da Sacco, Kevin V. Lemley, Sargis Sedrakyan, Laura Perin
Geremy Clair, Hasmik Soloyan, Paolo Cravedi, Andrea Angeletti, Fadi Salem, Laith Al-Rabadi, Roger E. De Filippo, Stefano Da Sacco, Kevin V. Lemley, Sargis Sedrakyan, Laura Perin
View: Text | PDF
Research Article Nephrology

The spatially resolved transcriptome signatures of glomeruli in chronic kidney disease

Abstract

Here, we used digital spatial profiling (DSP) to describe the glomerular transcriptomic signatures that may characterize the complex molecular mechanisms underlying progressive kidney disease in Alport syndrome, focal segmental glomerulosclerosis, and membranous nephropathy. Our results revealed significant transcriptional heterogeneity among diseased glomeruli, and this analysis showed that histologically similar glomeruli manifested different transcriptional profiles. Using glomerular pathology scores to establish an axis of progression, we identified molecular pathways with progressively decreased expression in response to increasing pathology scores, including signal recognition particle–dependent cotranslational protein targeting to membrane and selenocysteine synthesis pathways. We also identified a distinct signature of upregulated and downregulated genes common to all the diseases investigated when compared with nondiseased tissue from nephrectomies. These analyses using DSP at the single-glomerulus level could help to increase insight into the pathophysiology of kidney disease and possibly the identification of biomarkers of disease progression in glomerulopathies.

Authors

Geremy Clair, Hasmik Soloyan, Paolo Cravedi, Andrea Angeletti, Fadi Salem, Laith Al-Rabadi, Roger E. De Filippo, Stefano Da Sacco, Kevin V. Lemley, Sargis Sedrakyan, Laura Perin

×

Figure 3

Gene expression and cell deconvolution analysis of glomerular and tubular ROIs.

Options: View larger image (or click on image) Download as PowerPoint
Gene expression and cell deconvolution analysis of glomerular and tubula...
(A) Box plots of normalized intensities for PODXL, SYNPO, and EDH3 (glomeruli-specific genes) as well as HPN, CDH16, and TMEM37 (tubule-specific genes). Each dot represents either a glomerular (orange) or tubular (blue) ROI. Binomial GLM adjusted P values are indicated in red. (B) Dendrogram showing hierarchical clustering and transcriptional link between glomeruli (all samples) and tubules (all samples). A, adult; Y, young. (C). Unsupervised principal component analysis based on label-free quantification of the transcripts expressed in glomerular versus tubular ROIs in all kidney tissue segments analyzed based on principal components (PC1, PC2, PC3) constructed to capture the most variation in the samples. Percentage of total variance is indicated after each principal component. (D) Heatmap depicting the transcripts significantly modulated between glomerular and tubular ROIs in all kidney tissue segments analyzed (Student’s t test; binomial GLM test, Benjamini-Hochberg-adjusted [BH-adjusted] P < 0.05). Transcripts less than LOQ in value are shown in white. (E) Volcano plot representing the results of the Student t test (BH-adjusted P < 0.05) comparison of differential gene expression between glomerular and tubular ROIs in all the biopsies. A select list of GO terms and KEGG pathways significantly enriched (EASE-modified Fisher’s exact, P < 0.05) for genes upregulated in both glomeruli and tubules (Student’s t test and binomial test) is depicted on each side of the volcano plot. (F and G) Cell deconvolution analysis showing the abundance of cell types in glomerular ROIs (n = 73, F) and tubular ROIs (n = 29, G) from AS (no. 1–3), FSGS (no. 4 and 5), and MN (no. 7, 6) biopsies and their corresponding nondiseased controls, based on publicly available single-cell RNA-Seq experiments compiled into profile matrices by aggregating gene-wise counts of all annotated cell types (61).