Kidney organoids reveal redundancy in viral entry pathways during ACE2-dependent SARS-CoV-2 infection

Abstract

With a high incidence of acute kidney injury among hospitalized COVID-19 patients, considerable attention has been focussed on whether SARS-CoV-2 specifically targets kidney cells to directly impact renal function, or whether renal damage is primarily an indirect outcome. To date, several studies have utilized kidney organoids to understand the pathogenesis of COVID-19, revealing the ability for SARS-CoV-2 to predominantly infect cells of the proximal tubule (PT), with reduced infectivity following administration of soluble ACE2. However, the immaturity of standard human kidney organoids represents a significant hurdle, leaving the preferred SARS-CoV-2 processing pathway, existence of alternate viral receptors, and the effect of common hypertensive medications on the expression of ACE2 in the context of SARS-CoV-2 exposure incompletely understood. Utilizing a novel kidney organoid model with enhanced PT maturity, genetic- and drug-mediated inhibition of viral entry and processing factors confirmed the requirement for ACE2 for SARS-CoV-2 entry but showed that the virus can utilize dual viral spike protein processing pathways downstream of ACE2 receptor binding. These include TMPRSS- and CTSL/CTSB-mediated non-endosomal and endocytic pathways, with TMPRSS10 likely playing a more significant role in the non-endosomal pathway in renal cells than TMPRSS2. Finally, treatment with the antihypertensive ACE inhibitor, lisinopril, showed negligible impact on receptor expression or susceptibility of renal cells to infection. This study represents the first in-depth characterization of viral entry in stem cell-derived human kidney organoids with enhanced PTs, providing deeper insight into the renal implications of the ongoing COVID-19 pandemic.

Importance: Utilizing a human iPSC-derived kidney organoid model with improved proximal tubule (PT) maturity, we identified the mechanism of SARS-CoV-2 entry in renal cells, confirming ACE2 as the sole receptor and revealing redundancy in downstream cell surface TMPRSS- and endocytic Cathepsin-mediated pathways. In addition, these data address the implications of SARS-CoV-2 exposure in the setting of the commonly prescribed ACE-inhibitor, lisinopril, confirming its negligible impact on infection of kidney cells. Taken together, these results provide valuable insight into the mechanism of viral infection in the human kidney.

Keywords: COVID-19; SARS-CoV-2; kidney; kidney organoids; stem cells.

Fig 1

PT-enhanced organoids show extensive viral entry factor expression. (A) Line plot depicting viral titre as determined by Vero cell assays (Median Tissue Culture Infectious Dose; log₁₀ TCID₅₀) of culture media sampled from WA1 SARS-CoV-2 (icSARS-CoV-2-GFP)-infected PT-enhanced organoids across three independent experiments replicated identically (red, green, and blue lines), as well as mock-infected organoids (gray line). Mock-infected line is representative of all mock results across each experiment. LOD and dotted line represent lower limit of detection. Error bars represent standard error of the mean (SEM) from n = 3 individual wells of organoids (three organoids per well) at each timepoint (note these data represent the “no drug controls” from experiments depicted in Fig. 3A). (B) Confocal immunofluorescence of a representative PT-enhanced organoid 6 days post-infection, demonstrating SARS-CoV-2 double stranded RNA (dsRNA; red) localization, co-stained for markers of proximal tubules (LTL; blue), loop of Henle (SLC12A1; green, apical membrane staining), late distal tubule/connecting segment (GATA3; green, apical), and podocytes of the glomeruli (NPHS1; gray). Arrows indicate examples of a LTL-positive tubule (white arrow), SLC12A1-positive tubule (yellow arrow), and GATA3-positive tubule (cyan arrow). Scale bar represents 50 µm. (C) scRNAseq DotPlot of PT-enhanced organoids (day 14 of organoid culture) depicting the expression of entry factors (receptors and proteases) and pro-viral factors reported in literature to be involved in SARS-CoV-2 infectivity. Identities of each cluster are labeled (left axis), with numbers in brackets differentiating kidney cell populations for which multiple clusters exist. (D) Confocal immunofluorescence of PT-enhanced organoid examples depicting ACE2 expression (green) within LTL + proximal tubules (blue) and TMPRSS2 (red) in LTL structures. Epithelium is marked with EPCAM (gray). Arrows depict examples of ACE2 and TMPRSS2 staining. Yellow squares highlight fields of view shown at higher magnification in the images on far right. Scale bars represent 50 µm.

Fig 2

ACE2 is the sole receptor for SARS-CoV-2 in renal cells. (A) Confocal immunofluorescence of a representative PT-enhanced organoid 6 days post-infection demonstrating SARS-CoV-2 double-stranded RNA (dsRNA; red) localization within ACE2-positive (gray) proximal tubule cells. Organoid is co-stained for markers of proximal tubule brush-border membrane (LTL; blue) and nephron epithelium (EPCAM; green). Arrows indicate examples of dsRNA staining, with yellow arrow indicating the region shown at higher magnification in the inset images. Scale bar represents 50 µm. (B) Confocal immunofluorescence of PT-enhanced organoids (day 14 of organoid culture) generated from ACE2 knockout and wild-type iPSCs, depicting nephron epithelium (EPCAM; green), podocytes of the glomeruli (NPHS1; gray), and proximal tubules (LTL; blue). Scale bars represent 200 µm. Each confocal image depicts 3 × 3 stitched tiles, generated using the standard rectangular grid tile scan mode with automated stitching during image acquisition using ZEISS ZEN Black software (Zeiss Microscopy, Thornwood, NY) installed on a ZEISS LSM 780 confocal microscope (Carl Zeiss, Oberkochen, Germany). (C) Bar graphs from two independent experiments (top and bottom) depicting the viral titres (log₁₀ TCID₅₀/mL) of culture media sampled from ACE2 knockout (KO) and wild-type (WT) PT-enhanced organoids, both infected with VIC01 SARS-CoV-2 (dark and light red bars) or remaining uninfected (controls; light and dark blue bars). Error bars represent SEM from three biological replicates per timepoint. LOD, lower limit of detection.

Fig 3

Drug inhibition assays revealing dual viral entry pathway usage by SARS-CoV-2. (A) Bar graph depicting the viral titres (log₁₀ TCID₅₀/mL) of culture media sampled from PT-enhanced organoids 6 days post-infection, treated with either protease inhibitors (Camostat and E64d, alone or in combination; dark/light red bars), drug reconstitution reagent (DMSO controls; dark/light blue bars), or remaining untreated (no drug controls; light/dark green bars). PT-enhanced organoids were infected with WA1 SARS-CoV-2. (icSARS-CoV-2-GFP). LOD and dotted line represent lower limit of detection. Error bars represent SEM from three independent experiment, with three (drug inhibition tests and no drug controls) or two (DMSO controls) biological replicates per timepoint. Significance was calculated using a one-way ANOVA with Tukey’s multiple comparisons test. Asterisks represent two-tailed P values (*P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001). NS, not significant. (B) scRNAseq DotPlot of PT-enhanced organoids (day 14 of organoid culture) depicting the expression of TMPRSS genes within the highest ACE2-expressing proximal tubule clusters and distal (connecting segment) nephron cluster. The proximal tubule cluster most enriched with ACE2 expression is outlined in gray. Dot size represents the percentage of cells expressing a gene within each cluster, while shade intensity correlates with gene expression level. (C and D) Confocal immunofluorescence of TMPRSS10 (C, red) and corresponding rat isotype control (D, red), co-stained with markers of proximal tubule (LTL, blue) and nephron epithelium (EPCAM, green). Arrows indicate region shown at higher magnification in inset images. Scale bars represent 50 µm. (E) Confocal immunofluorescence of SARS-CoV-2-infected and uninfected (control) PT-enhanced organoids, depicting virus (dsRNA, red) within TMPRSS10-expressing (green) proximal tubules (marked with LTL, blue) within infected samples and lack of dsRNA detection in uninfected control. Arrows depict examples of dsRNA staining. Scale bars represent 50 µm.

Fig 4

PT-enhanced organoids do not show increased susceptibility to SARS-CoV-2 infection following lisinopril exposure. (A) Confocal immunofluorescence of a representative PT-enhanced organoid following 8 days of lisinopril treatment, depicting nephron epithelium (EPCAM; green), podocytes of the glomeruli (NPHS1; gray), proximal tubules (LTL; blue), and loop of Henle (SLC12A1; red). Inset depicts a region of the same merge image shown at higher magnification. Scale bar represents 200 µm. (B) qRT-PCR analyses of untreated (gray bars) and lisinopril-treated (orange bars) PT-enhanced organoids (14 days of organoid culture) from three independent experiments for expression of the lisinopril target, ACE, and relevant SARS-CoV-2 entry factors. Error bars represent SEM from three biological replicates. Statistical significance was assessed using an unpaired t test. NS, not significant. (C) Bar graph depicting the viral titres (log₁₀ TCID₅₀/mL) of culture media sampled from untreated and lisinopril-treated PT-enhanced organoids, either infected with WA1 SARS-CoV-2 (icSARS-CoV-2-GFP) 48 h post-treatment (light/dark red bars) or remaining uninfected (light/dark blue bars). LOD and dotted line represent lower limit of detection. Significance was assessed using unpaired t tests adjusted for multiple comparisons using the Holm-Sidak method. NS, not significant.