Cross-validation of SARS-CoV-2 responses in kidney organoids and clinical populations

Abstract

Kidneys are critical target organs of COVID-19, but susceptibility and responses to infection remain poorly understood. Here, we combine SARS-CoV-2 variants with genome-edited kidney organoids and clinical data to investigate tropism, mechanism, and therapeutics. SARS-CoV-2 specifically infects organoid proximal tubules among diverse cell types. Infections produce replicating virus, apoptosis, and disrupted cell morphology, features of which are revealed in the context of polycystic kidney disease. Cross-validation of gene expression patterns in organoids reflects proteomic signatures of COVID-19 in the urine of critically ill patients indicating interferon pathway upregulation. SARS-CoV-2 viral variants alpha, beta, gamma, kappa, and delta exhibit comparable levels of infection in organoids. Infection is ameliorated in ACE2-/- organoids and blocked via treatment with de novo-designed spike binder peptides. Collectively, these studies clarify the impact of kidney infection in COVID-19 as reflected in organoids and clinical populations, enabling assessment of viral fitness and emerging therapies.

Keywords: COVID-19; Genetic diseases; Molecular pathology; Nephrology; iPS cells.

Figure 1. SARS-CoV-2 efficiently infects human kidney organoids with tropism for proximal tubules.

(A) Schematic of kidney organoid infection protocol. (B) Whole-well wide-field immunofluorescence images of iPS cell–derived organoids infected with SARS-CoV-2-mNG. Arrowheads point to infected cells. (C) qRT-PCR of SARS-CoV-2 envelope RNA in organoids infected with SARS-CoV-2/WA1 or mock-infected (MOCK). Dots represent a well of organoids. Mean ± SEM, n ≥ 1 well of organoids per infection from 4 independent experiments. Mann-Whitney test, **P < 0.01. (D) Plaque assays of SARS-CoV-2–infected human kidney organoids derived from iPS cells or ES cells. Dots represent a well of organoids. Mean ± SEM, n ≥ 1 well of organoids per infection from 3 independent experiments, respectively. Mann-Whitney test, NS P > 0.05. (E) Representative confocal immunofluorescence images of organoids infected with SARS-CoV-2-mNG. (F) Quantification of infected organoid cellular tropism. Dots represent a single organoid. Mean ± SEM, n ≥ 4 organoids per infection from 3 independent experiments. Two-way ANOVA, multiple comparisons, MOCK vs. infected for each respective region; **P < 0.01, ****P < 0.0001, NS P > 0.05. (G) Representative confocal immunofluorescence images of organoid infected with SARS-CoV-2-GFP, with zoomed images of white boxed areas showing infected (top) versus uninfected (bottom) proximal tubules. Arrowheads indicate areas of disrupted LTL pattern.

Figure 2. SARS-CoV-2 infects PKD organoid cystic epithelium.

(A) Schematic of cystic PKD organoid infection protocol. (B) Representative confocal immunofluorescence images showing cystic PKD organoids infected with SARS-CoV-2-mNG. Outlines denote independent organoids. (C) Quantification of infected organoid area (percent total) of PKD and control (isogenic non-PKD) SARS-CoV-2-mNG–infected cultures. Dots represent a single organoid. Mean ± SEM, n ≥ 4 organoids per infection from 3 independent experiments each. Unpaired t test, NS P > 0.05. (D) Representative immunofluorescence images of cystic PKD organoids infected with SARS-CoV-2-mNG, with zoom of cleaved caspase-3 staining and pyknotic nuclei. (E) Quantification of pyknotic nuclei and elevated cleaved caspase-3 levels of infected and noninfected cells of infected organoids. Dots represent a biological replicate. Mean ± SEM, n ≥ 5 organoids per biological replicate from 3 independent experiments each. Unpaired t test, **P < 0.01, ***P < 0.001.

Figure 3. COVID-19⁺ patient urine expresses signatures found in organoids.

(A) Schematic of patient cohort sample selection and analysis. (B) Scatterplots of protein size versus relative fluorescence units (RFU) detected in patient urine and patient blood. (C) Gene Ontology pathway analysis of urine proteome reads. Red circles represent pathways that hit Bonferroni significance, orange circles represent pathways that hit an FDR less than 0.05, and blue circles represent nonsignificant pathways. (D) Volcano plot of increased and decreased proteins in COVID-19⁺ patient urine compared with COVID-19^– patient urine. Dotted lines represent FDR 0.1 and Bonferroni significance cutoffs. (E) Upregulated proteomic hits between COVID-19⁺ and COVID-19^– patients. (F) qRT-PCR of upregulated proteomic hits in organoids infected with SARS-CoV-2/WA1 or mock-infected. Dots represent a well of organoids. Mean ± SEM, n ≥ 1 well of organoids per infection from 4 independent experiments. Unpaired t test.

Figure 4. SARS-CoV-2 variants show similar rates of infection in kidney organoids.

(A) qRT-PCR of SARS-CoV-2 envelope RNA in infected kidney organoid cultures. Dots represent a well of organoids. Mean ± SEM, n ≥ 1 well of organoids per infection from 3 independent experiments. One-way ANOVA, Kruskal-Wallis post hoc test, ****P < 0.0001, NS P > 0.05. (B) Plaque assays of SARS-CoV-2–infected kidney organoids. Dots represent a well of organoids. Mean ± SEM, n ≥ 1 well of organoids per infection from 3 independent experiments. One-way ANOVA, Kruskal-Wallis post hoc test, *P < 0.05, **P < 0.01, NS P > 0.05. (C) Focus-forming assay of SARS-CoV-2–infected kidney organoids. Dots represent a well of organoids. Mean ± SEM, n ≥ 1 biological replicates per condition from 3 independent experiments. Mann-Whitney test, **P < 0.01. (D) Prevalence of admission AKI, dialysis, and death in COVID⁺ patients over time.

Figure 5. ACE2 is an essential viral entry pathway for SARS-CoV-2 infection of kidney organoids.

(A) Schematic of ACE2 knockout and infection protocol. (B) Representative confocal immunofluorescence images of ACE2^–/– SARS-CoV-2-mNG–infected organoids, compared with isogenic controls. (C) Quantification of GFP⁺ area in ACE2^–/– organoids infected with SARS-CoV-2-mNG, compared with mock-treated control. Dots represent a single organoid. n ≥ 4 organoids per experiment from 2 experiments. (D) Plaque assay of ACE2^–/– and control organoids infected with SARS-CoV-2 or mock-treated. Non-log scale is shown for this figure to emphasize low levels of infection in ACE2^–/– organoids. Dots represent a well of organoids. Mean ± SEM, n ≥ 1 well of organoids per infection from 3 independent experiments, using 2 distinct mutant cell lines. One-way ANOVA with Tukey’s post hoc tests, *P < 0.05, **P < 0.01, NS P > 0.05.

Figure 6. Therapeutic interventions reduce SARS-CoV-2 infection and replication in human kidney organoids.

(A) Schematic of protocol for SARS-CoV-2 kidney organoid infection with remdesivir treatment. (B) Plaque assays of SARS-CoV-2– and SARS-CoV-2-mNG–infected kidney organoids treated with or without remdesivir. Mean ± SEM of 3 independent experiments. Wilcoxon matched-pairs signed rank test, *P < 0.05. (C) Schematic of LCB1 binding to spike glycoprotein receptor-binding domain (RBD). (D) Schematic of LCB1 viral pretreatment and infection of kidney organoids. (E) qRT-PCR expression levels of SARS-CoV-2 envelope RNA in infected kidney organoid cultures, with increasing levels of LCB1 protein preincubated with virus. Dots represent a well of organoids. Mean ± SEM, n ≥ 1 well of organoids per infection from 4 independent experiments, 2 iPS and 2 ES, normalized to β-actin. One-way ANOVA, Kruskal-Wallis post hoc test, *P < 0.05, **P < 0.01, ***P < 0.001, NS P > 0.05. (F) Plaque assays of SARS-CoV-2–infected kidney organoids with increasing levels of LCB1 protein preincubated with virus. Dots represent a well of organoids. Mean ± SEM, n ≥ 1 well of organoids per infection from 4 independent experiments, 2 iPS and 2 ES. One-way ANOVA, Kruskal-Wallis post hoc test, *P < 0.05, ***P < 0.001, ****P < 0.0001, NS P > 0.05.