The Nephrotoxin Puromycin Aminonucleoside Induces Injury in Kidney Organoids Differentiated from Induced Pluripotent Stem Cells

Abstract

Kidney diseases, including acute kidney injury (AKI) and chronic kidney disease (CKD), which can progress to end stage renal disease (ESRD), are a worldwide health burden. Organ transplantation or kidney dialysis are the only effective available therapeutic tools. Therefore, in vitro models of kidney diseases and the development of prospective therapeutic options are urgently needed. Within the kidney, the glomeruli are involved in blood filtration and waste excretion and are easily affected by changing cellular conditions. Puromycin aminonucleoside (PAN) is a nephrotoxin, which can be employed to induce acute glomerular damage and to model glomerular disease. For this reason, we generated kidney organoids from three iPSC lines and treated these with PAN in order to induce kidney injury. Morphological observations revealed the disruption of glomerular and tubular structures within the kidney organoids upon PAN treatment, which were confirmed by transcriptome analyses. Subsequent analyses revealed an upregulation of immune response as well as inflammatory and cell-death-related processes. We conclude that the treatment of iPSC-derived kidney organoids with PAN induces kidney injury mediated by an intertwined network of inflammation, cytoskeletal re-arrangement, DNA damage, apoptosis and cell death. Furthermore, urine-stem-cell-derived kidney organoids can be used to model kidney-associated diseases and drug discovery.

Keywords: AKI; DNA damage; RAAS; apoptosis; iPSCs; inflammation; organoids; puromycin aminonucleoside; urine cells.

Figure 1

Lobular kidney organoids contain distinct kidney structures. (a) Schematic depiction of the protocol for generating kidney organoids. (b) Overview of iPSC spheroids at D8 after generation (a.,b.) and kidney organoids UMK2 at D21 (c.,d.) with binocular. (a.) 4× magnification under light microscope. Scale bar depicts 500 µm. (b.) 10× magnification under light microscope. Scale bar depicts 200 µm. (c.) 1× magnification. Scale bar depicts 2000 µm. (d.) 4× magnification. Scale bar depicts 2000 µm. (c) Morphology of organoid section via histological H&E staining. A typical glomerulus-like structure is depicted by a dashed rectangle. Tubule-like structures are marked with an asterisk. Scale bar depicts 50 µm. (d) Confocal pictures of glomerular (ACTN4, yellow) and tubular (LTL, green) structures in UMK1 sections. Nuclear OCT2 (POU2F2) (red) is expressed in UMK1 sections. A glomerulus-like structure was marked with an arrow. Scale bar depicts 50 µm. (e) Monitoring of iPSC spheroids and kidney organoids in a dextran uptake assay. (a.) iPSC spheroids treated with dextran after 4 h pulse. (b.) kidney organoids treated with dextran after 4 h pulse. (c.) kidney organoids treated with dextran after 24 h chase. Scale bar depicts 200 µm.

Figure 2

Comparative analysis of gene expression in iPSC spheroids and kidney organoids. (a) Similarities between spheroids and non-treated and PAN-treated organoids are shown in the cluster dendrogram. Control and PAN-treated UMK1 cluster together, and iPSC spheroids cluster separately. (b) The common gene-sets between spheroids and untreated kidney organoids consists of 15,332 genes. In total, 500 genes are exclusively expressed in spheroids and 310 genes in control organoids. (c) Expression of pluripotency-associated genes in UMK1_con and SPH depicted in a Pearson heatmap. (d) Expression of kidney-associated genes in UMK1_con and SPH depicted in a Pearson heatmap.

Figure 3

PAN leads to less defined glomerular and tubular structures. (a) PAN induction leads to a decrease in the number of highly proliferative KI67+ cells (red). Nuclei were stained with Hoechst33342 (blue). Scale bar depicts 50 µm. (b) Expression of the podocyte marker SYNPO is upregulated (UMK1, FFK1, FFK2, UFK2) after PAN treatment. Error bars depict standard error. (c) PODXL+ glomeruli (yellow) are less defined after PAN treatment. Glomeruli are marked with a white arrow. Nuclei were stained with Hoechst33342 (blue). Scale bar depicts 50 µm. (d) Confocal microscopy pictures of PODXL+ glomeruli with and without PAN. Scale bar depicts 20 µm. (e) LTL+ proximal tubules (green) and ACTN4+ glomeruli (yellow) are less defined after PAN treatment. Glomeruli are marked with a white arrow. Nuclei were stained with Hoechst33342 (blue). Scale bar depicts 50 µm. (f) Comparative confocal pictures of ACTN4 (yellow) and LTL-stained (green) organoid sections treated with and without PAN. Scale bar depicts 20 µm. Glomeruli are marked with a white arrow. (g) Expression of the tubular markers ABCC4, CLDN10 and NR3C2 are downregulated by PAN. Error bars depict standard error.

Figure 4

Transcriptome analysis of UMK1 with and without PAN treatment. (a) The common gene-sets between UMK1_con and UMK1_PAN consisted of 15344 genes. Exclusively expressed in UMK1_con and UMK1_PAN are 298 and 215 genes, respectively. (b) Expression of kidney-associated genes in UMK1_con and UMK1_PAN is displayed in a Pearson heatmap.

Figure 5

PAN treatment induces DNA damage in kidney organoids. (a) Elevated expression of γH2A.X (red) in PAN-treated kidney organoids UMK1. Nuclei were stained with Hoechst33342 (blue). Scale bar depicts 50 µm. (b) Elevated expression of t-P53 and cleaved Caspase 3 in PAN-treated kidney organoids UMK1. (c) Upregulated enrichment clusters include cell damage. (d) Cell-cycle-related enrichment clusters were downregulated by PAN treatment.

Figure 6

Inflammation-associated gene expression and cytokine secretion in untreated and PAN-treated kidney organoids. (a) Pearson heatmap of the expression of immune-related genes in control and PAN treatment. (b) Expression of IL-6 and IL-8 is elevated in PAN-induced kidney organoids as measured by qRT-PCR. Error bars depict standard error. (c) Cluster dendrogram with the technical duplicates of untreated kidney organoids clustering together and the PAN-treated kidney organoids cluster separately. (d) Cytokine array data comparing expression between untreated control and PAN treatment. (e) Metascape-generated heatmap comparing UMK1_PAN and UMK1_con included inflammation- and immune-response-related GOs. Subjected gene-sets are based on the Venn analysis in Figure 4. (f) Bar graph of non-redundant enrichment clusters after PAN treatment.

Figure 7

PAN induction affects the RAAS in kidney organoids. (a) PAN induces downregulation of genes of the KEGG pathway renin secretion. Downregulated genes are marked in blue. (b) Renin concentration (ng/µL) in conditioned media of untreated and PAN-treated kidney organoids. Significance is determined by α-value ≤ 0.05. (c) Expression of RAAS-associated AGT and AGTR1 is upregulated in PAN-induced kidney organoids. Error bars depict the standard error. (d) Pearson’s heatmap depicting the expression of RAAS-associated genes in untreated and PAN-treated kidney organoids.