Natural History and Clinicopathological Associations of TRPC6-Associated Podocytopathy

Visual Abstract

Abstract

Key Points

• We conducted a clinical, genetic, and pathological analysis on 64 cases from 39 families with TRPC6-associated podocytopathy (TRPC6-AP).

• Analysis of 37,542 individuals excluded a major contribution of loss-of-function variants to TRPC6-AP, legitimating current drug discovery approaches.

• This study identifies key features of disease that can help intervention studies design and suggests similarities between TRPC6-AP and primary FSGS.

Background

Understanding the genetic basis of human diseases has become integral to drug development and precision medicine. Recent advancements have enabled the identification of molecular pathways driving diseases, leading to targeted treatment strategies. The increasing investment in rare diseases by the biotech industry underscores the importance of genetic evidence in drug discovery and approval processes. Here we studied a monogenic Mendelian kidney disease, TRPC6-associated podocytopathy (TRPC6-AP), to present its natural history, genetic spectrum, and clinicopathological associations in a large cohort of patients with causal variants in TRPC6 to help define the specific features of disease and further facilitate drug development and clinical trials design.

Methods

The study involved 64 individuals from 39 families with TRPC6 causal missense variants. Clinical data, including age of onset, laboratory results, response to treatment, kidney biopsy findings, and genetic information, were collected from multiple centers nationally and internationally. Exome or targeted sequencing was performed, and variant classification was based on strict criteria. Structural and functional analyses of TRPC6 variants were conducted to understand their effect on protein function. In-depth reanalysis of light and electron microscopy specimens for nine available kidney biopsies was conducted to identify pathological features and correlates of TRPC6-AP.

Results

Large-scale sequencing data did not support causality for TRPC6 protein-truncating variants. We identified 21 unique TRPC6 missense variants, clustering in three distinct regions of the protein, and with different effects on TRPC6 3D protein structure. Kidney biopsy analysis revealed FSGS patterns of injury in most cases, along with distinctive podocyte features including diffuse foot process effacement and swollen cell bodies. Most patients presented in adolescence or early adulthood but with ample variation (average 22, SD ±14 years), with frequent progression to kidney failure but with variability in time between presentation and kidney failure.

Conclusions

This study provides insights into the genetic spectrum, clinicopathological associations, and natural history of TRPC6-AP.

Clinical Trial registry name and registration number:

A Study to Test BI 764198 in People With a Type of Kidney Disease Called Focal Segmental Glomerulosclerosis, NCT05213624.

Introduction

Drug discovery and clinical trials have evolved significantly in the past decades. It is increasingly recognized that understanding the genetic underpinning of human diseases greatly facilitates drug development and precision medicine. Thanks to recent progress in genetic sequencing and multiomic technologies, it is now possible to identify the key molecular pathways, protein functions, and interactions that drive various diseases, paving the way for targeted treatment strategies. To inform precise diagnosis and accurate genotype–phenotype correlation, we can now also evaluate structural consequences or chemical disturbances caused by missense variants, establishing their causal role in genetic-driven diseases.¹

Although many monogenic Mendelian diseases are rare, understanding of the genetic mechanisms of these disorders can lead to the identification of druggable targets for common conditions. This has been exemplified in the study of familial hypercholesterolemia, a disease that in its autosomal recessive form occurs in one in 1,000,000 individuals, but the investigation of which led to the development of statins and PCSK9 inhibitors, key and effective drugs used by millions of individuals in the United States to treat common hypercholesterolemia.^2–8 The importance of these observations is demonstrated by the last decade's trends in drug discovery and investment in rare diseases by the biotech industry.^9–11 An analysis from 2015 suggested that selecting targets that are supported genetically could double the success rate in clinical drug development, quantifying for the first time the potential effect of genetic evidence in guiding therapeutic discoveries.¹¹ Retrospective analysis of US Food and Drug Administration–approved drugs from 2013 to 2022 showed that 63% of the new drugs approved were supported by genetic evidence, and medications without genetic support were mostly symptomatic treatments or molecules that do not target gene products. Moreover, expedited approval was twice as likely for drugs with genetic support as compared with the ones without it.¹² We are entering a new era in which previously untreatable conditions are becoming treatable, as demonstrated by the recent development and approval of the first gene editing therapy for the treatment of sickle cell disease, for which, historically, the only curative option was stem cell transplantation.^13–15

This increasing interest in rare genetic forms of common diseases is also occurring in nephrology.^16–24 TRPC6-associated podocytopathy (TRPC6-AP) is a promising example because it is driven primarily by genetic variants that result in an overall protein gain-of-function increasing the channel opening probability.^25–28 Conditions driven by gain-of-function variants are, at least in principle, more druggable as compared with diseases caused by loss-of-function variants because inhibitors are generally easier to develop than drugs that restore the function of a target protein.^29–32 Indeed, antagonists for the TRPC6 ion channel exist and one is currently in a phase 2 clinical trial, enrolling both cases with primary FSGS as well as cases with TRPC6-AP (ClinicalTrials.gov, ID NCT05213624).

TRPC6 encodes a transmembrane ion channel that, when activated, primarily allows calcium influx. Present in multiple cell types, TRPC6 has been demonstrated to coprecipitate with the podocin apparatus, implying its localization to the slit diaphragm in podocytes.²⁷ There are a number of stimuli that increase open probability of the TRPC6 ion channel in podocytes, including angiotensin II and mechanical stretch at the slit diaphragm.³³ In the open state, TRPC6 permits calcium influx, which results in activation of the calcineurin-NFAT pathway.³⁴ In TRPC6-AP gain-of-function variants, increased calcium flux through the channel is believed to mediate cellular damage and, eventually, apoptosis.³⁵

TRPC6-AP is an autosomal dominant Mendelian disease. First described in 2005,^27,28 there have been a number of pathogenic variants identified.^{25,27,28,36–52} TRPC6-AP causes progressive kidney disease with proteinuria and focal segmental glomerulosclerosis on biopsy.^27,28,53 Most of the available data indicate that TRPC6-AP is largely a disease of adult-onset,⁵⁴ although early pediatric-onset cases have been reported.^38,39

Despite the existence of targeted inhibitors of the TRPC6 ion channel, to date, there have been no trials specifically applying TRPC6 inhibitors to patients with TRPC6-AP due in large part to the rarity of the disorder and uncertainty regarding optimal trial enrollment parameters and outcomes to measure. In fact, it is important noting that variations in TRPC6, like for the vast majority of the approximately 4000 Mendelian genes, cause disease with little to no available natural history data, creating a barrier to clinical development.

We present the natural history, genetic spectrum, and clinicopathological associations of TRPC6-AP in a large cohort of patients with TRPC6 variants to fill this knowledge gap, help define the specific features of the disease, and further facilitate drug development and clinical trials design.

Methods

Refer to the Supplemental Material for detailed methods.

TRPC6-AP Case Study Population and Enrollment

We enrolled 64 individuals from 39 families harboring TRPC6 causal variants. Clinical variables included age of onset, sex, response to immunosuppression, kidney failure status, time to development of kidney failure from presentation, receipt of kidney transplant, and kidney biopsy information. Genomic DNA was isolated from blood following informed consent.

Large-Scale Exome and Genome Sequencing Investigations for TRPC6 Variants

While the current knowledge suggests that loss-of-function TRPC6 variants are not causal for FSGS,^55–58 the missense p.G757D has been previously functionally characterized as loss-of-function.⁵⁹ Moreover, loss-of-function variants are preferentially classified as pathogenic/likely pathogenic by American College of Medical Genetics criteria, irrespective of the underlying mechanism. Therefore, to clarify the contribution of loss-of-function protein-truncating variants to the etiology of TRPC6-AP and precisely select our dataset of study, we analyzed whole exome or genome sequencing from 37,542 individuals, including 5267 cases with nephrotic syndrome and 32,273 healthy and disease controls. All genetic data were aggregated, harmonized, and analyzed at the Institute for Genomic Medicine at Columbia University.⁶⁰

Sequencing, Variant Annotation, and Classification

Exome sequencing for the 24 TRPC6-AP cases from our aggregated nephrotic syndrome cohort was performed at the Institute for Genomic Medicine. The remaining cases were subjected to whole genome sequencing (from the CureGN Study⁶¹) or whole exome sequencing/Sanger sequencing as part of discovery studies previously published by our group.^{27,28,34,38,62–65}

TRPC6 variant prioritization was conducted using Varsome (https://varsome.com/), implementing the American College of Medical Genetics guidelines⁶⁶ to define first-tier and second-tier positive findings. We used Sanger sequencing to confirmation and test for segregation.

Protein Structure Analyses

Affected regions of the TRPC6 protein in cases and controls were visualized using the cBioPortal Mutation Mapper.^67,68 The published structure of TRCP6 (protein data bank ID: 7DXF) was used, and the structural consequences of individual missense variants were evaluated in Coot.⁶⁹ Variants in the vicinity of TRCP6's CBS1 Ca²⁺-coordinating site were evaluated for whether they disturbed hydrogen bonding, the Ca²⁺-coordinating shell, van der Waals contacts, or introduced significant steric clashes. Selected variants with potential mechanisms for disturbing the CBS1 site were represented in ChimeraX⁷⁰ to illustrate the likely impact of the mutant residue on the local structure as compared with wild type.

Kidney Biopsy Analysis

Nine kidney biopsies were available for re-review. Cases were evaluated for % segmental and global glomerulosclerosis, histologic variant of FSGS, % tubular atrophy/interstitial fibrosis, arteriosclerosis, and podocyte features by light and electron microscopy.

Statistical Analyses

All analyses were conducted in R v4.2.2 and Prism v9.4.1; pairwise comparisons were conducted using the Fisher exact or Mann Whitney U test; differences in Kaplan–Meier survival curves were compared using log-rank test.

Results

Large-Scale Burden Analysis Supported No Significant Role of TRPC6 Protein-Truncating Variants in the Pathogenesis of Proteinuric Kidney Disease

We compared the burden of protein-truncating variants between nephrotic syndrome cases, healthy controls, and disease controls. In total, we identified 18 TRPC6 protein-truncating variants in 4/5267 nephrotic syndrome cases (0.08%) and 14/32,273 controls (0.04%). There was no statistically significant excess burden of TRPC6 protein-truncating variants between nephrotic syndrome cases and healthy individuals nor compared with disease controls, individually or together (Supplemental Figure 1 and Supplemental Table 1). Examination of clinical data from the four nephrotic syndrome cases carrying TRPC6 protein-truncating variants revealed a steroid sensitive phenotype in 2/4 cases and a mixed FSGS/membranoproliferative kidney biopsy pattern in another case (Supplemental Table 3). This observation supported a coincidental and noncausal occurrence of TRPC6 protein-truncating variants in individuals with nephrotic syndrome, likely representing mutational background. Interrogation of the United Kingdom Biobank AstraZeneca PheWAS portal (https://azphewas.com/) did not show any significant association of TRPC6 protein-truncating variants with kidney traits. Consistent with these results, the lack of TRPC6 genic constraint (gnomAD v4.0 pLoF=0), and prior evidence,^55–58 there is currently no support for heterozygous protein-truncating loss-of-function variants resulting in haploinsufficiency as causal for TRPC6-AP. The explanation of the predicted loss-of-function mechanism of the missense disease-associated p.G757D suggests a dominant-negative effect, which would mimic more closely the effect of biallelic protein-truncating variants, a model that is not testable with the current datasets because the overall prevalence of heterozygous loss-of-function genotypes was very low and no recessive loss-of-function genotypes have been identified in 37,542 individuals. Hence, for this study, we only considered individuals with TRPC6 missense variants.

Baseline Characteristics of the TRPC6-AP Study Cohort

We analyzed 64 TRPC6 affected variant carriers from 39 families (Table 1 and Supplemental Figure 2); 50% were females. Age at time of clinical recognition of disease was variable, ranging from the first year of life to 69 years. Most patients showed no response to immunosuppressive therapy. Only one case, TAP33-1, presented with initial response to corticosteroid therapy but eventually progressed to kidney failure and underwent kidney transplantation with no recurrence of disease on the graft.

Table 1.

Baseline clinical characteristics of TRPC6 variant carriers and family members

Family ID	GA/SRR	Patient ID	Nucleotide Change	Amino Acid Change	Sex	Age of Clinical Recognition	Descriptive Phenotype	Biopsy	Biopsy Diagnosis	Kidney Failure	Age at Kidney Failure	Age at Last Follow-Up (If No Kidney Failure)	Kidney Tx
TAP01	Eur/NA	TAP01-1	c.2683C>T	p.R895C	F	39	Proteinuria, FSGS, kidney failure	Y	FSGSa	Y	57		N
		TAP01-2	c.2683C>T	p.R895C	F	23	Proteinuria  in pregnancy,  FSGS, kidney failure	Y	FSGS	Y	30		U
		TAP01-3	c.2683C>T	p.R895C	F	22	Collapsing FSGS  during pregnancy	Y	FSGS	Y	22		Y
		TAP01-4	c.2683C>T	p.R895C	M	NA	Healthy at last follow-up	N	—	N		38	N
TAP02	NA/Non-Hispanic White	TAP02-1	c.2683C>T	p.R895C	F	6	FSGS	Y	FSGS	Y	33		Y
		TAP02-2	c.2683C>T	p.R895C	M	23	Nephrotic syndrome	Y	FSGS	Y	28		Y
		TAP02-3	c.2683C>T	p.R895C	F	21	Nephrotic syndrome	N	—	Y	27		N
		TAP02-4	c.2683C>T	p.R895C	M	21	Proteinuria	Y	FSGS	Y	33		Y
TAP03b	NA/Hispanic	TAP03-1	c.2683C>T	p.R895C	M	46	FSGS, kidney failure	Y	FSGS	Y	46		Y
		TAP03-2	c.2683C>T	p.R895C	M	18	FSGS	Y	FSGS	N		26	N
		TAP03-3	c.2683C>T	p.R895C	M	41	FSGS, kidney failure	Y	FSGS	Y	41		Y
TAP04	NA/Asian	TAP04-1	c.2683C>T	p.R895C	F	32	FSGS, kidney failure	Y	FSGS	Y	36		U
TAP05b	NA/Non-Hispanic White	TAP05-1	c.2683C>T	p.R895C	F	3	SRNS	N	—	Y	3		Y
TAP06b	NA/Asian	TAP06-1	c.2683C>T	p.R895C	F	5	Proteinuria	N	—	N		23	N
TAP07	Eur/NA	TAP07-1	c.2683C>T	p.R895C	F	31	Proteinuria in pregnancy, FSGS, kidney failure	Y	FSGSa	Y	40		Y
TAP08	Eur/Non-Hispanic White	TAP08-1	c.2683C>T	p.R895C	M	10	Proteinuria, FSGS	Y	FSGSa	N		25	N
TAP09	Eur/NA	TAP09-1	c.2683C>T	p.R895C	M	5	Nephrotic syndrome, FSGS, kidney failure	Y	FSGS	Y	21		Y
TAP10	Eur/NA	TAP10-1	c.2683C>T	p.R895C	M	42	Proteinuria, IgA nephropathy, FSGS, kidney failure	Y	FSGS+IgA nephropathy	Y	47		Y
TAP11	NA/Non-Hispanic White	TAP11-1	c.2683C>T	p.R895C	M	29	Proteinuria, FSGS, kidney failure	Y	FSGSa	Y	38		Y
TAP12	Eur/NA	TAP12-1	c.2689G>A	p.E897K	F	19	Nephrotic syndrome, FSGS	Y	FSGSa	N		27	N
		TAP12-2	c.2689G>A	p.E897K	M	17	Nephrotic  syndrome, FSGS	Y	FSGSa	N		20	N
		TAP12-3	c.2689G>A	p.E897K	F	26	FSGS, kidney failure	Y	FSGS	Y	35		Y
		TAP12-4	c.2689G>A	p.E897K	F	U	U	U	—	Y	46		Y
TAP13	NA/Non-Hispanic White	TAP13-1	c.2689G>A	p.E897K	F	29	Proteinuria in pregnancy, kidney failure	Y	NA	Y	46		Y
		TAP13-2	c.2689G>A	p.E897K	F	25	Proteinuria in pregnancy, kidney failure	N	—	Y	35		Y
		TAP13-3	c.2689G>A	p.E897K	F	33	Proteinuria	N	—	U		33	N
TAP14	NA/Non-Hispanic White	TAP14-1	c.2689G>A	p.E897K	F	17	FSGS, kidney failure	Y	FSGS	Y	17		U
TAP15b	NA/African American	TAP15-1	c.428A>G	p.N143S	M	33	Proteinuria, kidney failure	N	—	Y	39		Y
		TAP15-2	c.428A>G	p.N143S	M	27	FSGS, kidney failure	Y	Chronic GN/FSGS	Y	30		Y
TAP16	NA/Non-Hispanic White	TAP16-1	c.428A>G	p.N143S	F	16	Nephrotic syndrome	N	—	N		47	N
		TAP16-2	c.428A>G	p.N143S	M	15	FSGS	Y	FSGS	N		18	N
TAP17b	NA/African American	TAP17-1	c.428A>G	p.N143S	F	8	SRNS, FSGS	Y	FSGS	U		8	N
TAP18	NA/Middle Eastern	TAP18-1	c.523C>T	p.R175W	M	U	Kidney failure	U	NA	Y	28		Y
		TAP18-2	c.523C>T	p.R175W	M	6	SRNS, FSGS	Y	FSGS	N		15	N
		TAP18-3	c.523C>T	p.R175W	M	5	SRNS	N	—	N		11	N
TAP19b	NA/Middle Eastern	TAP19-1	c.523C>T	p.R175W	F	7	SRNS, FSGS	Y	FSGS	N		17	N
TAP20	NA/Non-Hispanic White	TAP20-1	c.523C>T	p.R175W	M	8	SRNS, FSGS/C3GN	Y	FSGS/C3GN	Y	8		Y
TAP21	NA/Non-Hispanic White	TAP21-1	c.328T>G	p.N110H	M	35	FSGS, kidney failure	Y	FSGS	Y	38		U
		TAP21-2	c.328T>G	p.N110H	F	34	Proteinuria, FSGS,  kidney failure	Y	FSGS	Y	35		U
		TAP21-3	c.328T>G	p.N110H	M	13	Proteinuria,  kidney failure	N	—	Y	18		U
		TAP21-4	c.328T>G	p.N110H	F	13	Proteinuria	N	—	U		13	N
TAP22b	NA/Hispanic	TAP22-1	c.808T>A	p.S270T	F	24	Proteinuria, FSGS	Y	FSGS	N		25	N
		TAP22-2	c.808T>A	p.S270T	M	21	Proteinuria, kidney failure	N	—	Y	21		Y
		TAP22-3	c.808T>A	p.S270T	F	53	Proteinuria	N	—	U		53	N
TAP23	NA/Non-Hispanic White	TAP23-1	c.319G>C	p.E107Q	M	22	Proteinuria, FSGS/Alport, kidney failure	Y	GBM abnormalities  suggesting Alport syndrome	Y	22		Y
		TAP23-2	c.319G>C	p.E107Q	F	11	Proteinuria, microhematuria, FSGS/Alport	Y	GBM abnormalities suggesting Alport syndrome	N		24	N
TAP24	Admixed/NA	TAP24-1	c.434A>G	p.H145R	F	29	Proteinuria in pregnancy, kidney failure	Y	NA	Y	43		Y
		TAP24-2	c.434A>G	p.H145R	F	27	Proteinuria, FSGS	Y	FSGSa	N		50	N
TAP25	Eur/Non-Hispanic White	TAP25-1	c.643C>T	p.R215W	M	37	Proteinuria, FSGS	Y	FSGSa	N		58	N
TAP26	Eur/NA	TAP26-1	c.643C>T	p.R215W	F	18	Proteinuria in pregnancy, FSGS, kidney failure	Y	FSGS; collapsing  features	Y	34		Y
TAP27	NA/Non-Hispanic White	TAP27-1	c.2270G>A	p.G757D	M	25	Proteinuria, FSGS	Y	FSGS	Y	44		U
		TAP27-2	c.2270G>A	p.G757D	M	0	Proteinuria, FSGS	Y	FSGS	N		16	N
TAP29b	NA/Asian	TAP29-1	c.2684G>T	p.R895L	M	1	SRNS, FSGS	Y	FSGS	Y	1		Y
TAP30b	NA/Non-Hispanic White	TAP30-1	c.2684G>T	p.R895L	F	11	FSGS, kidney failure	Y	FSGS	Y	11		Y
TAP32	NA/Non-Hispanic White	TAP32-1	c.304T>A	p.F102I	M	3	SRNS, MCD, FSGS	Y	MCD (first), FSGS  (second)	N		11	N
TAP33	Eur/Non-Hispanic White	TAP33-1	c.335C>G	p.P112R	F	36	Nephrotic syndrome, FSGS, kidney failure	Y	FSGS-tip lesion	Y	41		Y
TAP34b	NA/Non-Hispanic White	TAP34-1	c.395T>C	p.M132T	M	8	SRNS, MCD, FSGS	Y	MCD with FSGS  features	Y	9		Y
TAP35	Eur/NA	TAP35-1	c.518A>G	p.Y173C	F	29	Proteinuria	Y	Not sampled FSGS	N		36	N
TAP36	Admixed/NA	TAP36-1	c.529G>A	p.V177M	F	28	Nephrotic syndrome, MCD	Y	MCDa	N		29	N
TAP37	Admixed/NA	TAP37-1	c.850G>A	p.A284T	M	69	Nephrotic syndrome, FSGS	Y	FSGS	N		71	N
TAP38	Eur/NA	TAP38-1	c.874G>T	p.D292Y	M	4	Nephrotic syndrome	N	—	N		40	N
TAP39	Admixed/NA	TAP39-1	c.2284G>A	p.V762I	M	9	FSGS	Y	FSGS	Y	9		Y
TAP41	Eur/NA	TAP41-1	c.2680C>A	p.L894I	M	50	CKD, FSGS	Y	Proliferative  mesangial GN+IgM/FSGS	Y	62		Y
TAP42	NA/Non-Hispanic White	TAP42-1	c.2696T>A	p.L899H	F	15	FSGS	Y	FSGS	N		33	N

C3GN, C3 glomerulonephritis; Eur, European; F, female; GA, genetic ancestry; GBM, glomerular basement membrane; M, male; MCD, minimal change disease; NA, not available; SRNS, steroid resistant nephrotic syndrome; SRR, self-reported race; Tx, transplant; U, unknown.

^aKidney biopsy available for review.

^bFamilies and cases previously described.^{27,38,62–64}

TRPC6 Causal Variants Were Enriched in Protein Inhibitory Domains

To identify specific domains affected by pathogenic variants in cases as compared with controls and, therefore, to hypothesize specific functional effects to then test in 3D structure models, we plotted all 21 unique missense variants identified in our families and 22 variants identified among 11,818 multiethnic population controls into the linearized TRPC6 protein construct. The most common variants in TRPC6-AP were p.R895C (n=19, 11 families), p.E897K (n=8, three families), p.N143S (n=5, three families), p.R175W (n=5, three families), and p.N110H (n=4, one family), as reported in Figure 1 and Table 2. Three distinct regions of involvement emerged (Figure 2). The first was in the midst of a proximal region of ankyrin repeats from amino acid positions 100 to 180. The second was in the TRP_2 region from AA 270 to 292. The third cluster of variants occurred in the C-terminal region of the protein, from AA 750 to 900. In controls, there was a notable absence of involvement of the proximal ankyrin repeat section (cluster 1), minimal involvement of the C-terminal region (cluster 3), and several variants involving the TRP_2 region. Interestingly, there were several variants in controls affecting the ion transporter region (AA, 497–727), which was depleted in cases. Intriguingly, in the tertiary structure of the folded TRPC6 protein, clusters 1 and 3 are positioned next to each other, forming an interface that constitutes the intracellular domain of the transmembrane protein. The interface between the ankyrin repeats and C-terminal region has been suggested to form an inhibitory domain favoring the closed state of the TRPC6 cation channel, with variants in these regions possibly resulting in a loss-of-function of the inhibitory domains and, hence, a higher open probability of the channel and more cation flux.⁷¹

Figure 1.

TRPC6 variants identified in this study. Individual variants are arranged by frequency on the y axis, with the number of patients with each variant depicted on the x axis. The most common variants were R985C, E897K, N143S, R175W, and N110H.

Table 2.

Characteristics of the TRPC6 missense variants described in this study

Chromosome Position (Hg38)	Nucleotide Change	Amino Acid Change	ClinVar	Varsome ACMG	REVEL	PrimateAI	AlphaMissense	PolyPhen-2	gnomAD V4 Global AF	Affected (Fam)	Cluster	Previously Reported Variants
101453068	c.2683C>T	p.Arg895Cys	P	P	0.91	0.8582	0.986	1	1.86×10−6	19 (11)	3	Refs. 25,28,45,46,48–50, 59,79,85,86
101453062	c.2689G>A	p.Glu897Lys	P	LP	0.904	0.8391	0.981	0.998	NA	8 (3)	3	Refs. 45,48,85,87,88
101504446	c.523C>T	p.Arg175Trp	Conflicting	LP	0.551	0.8595	0.826	1	NA	5 (3)	1	Refs. 37,39,52,63
101504541	c.428A>Ga	p.Asn143Ser	Conflicting	B	0.464	0.4787	0.206	1	3.84×10-5	5 (3)	1	Refs.27,64
101504641	c.328A>C	p.Asn110His	NA	VUS	0.511	0.6372	0.728	1	NA	4 (1)	1	NA
101504161	c.808T>A	p.Ser270Thr	P	LP	0.895	0.7553	0.933	0.989	NA	3 (1)	2	Refs. 27,49
101504535	c.434A>G	p.His145Arg	Uncertain significance	VUS	0.291	0.8139	0.979	0.106	NA	2 (1)	1	NA
101453067	c.2684G>T	p.Arg895Leu	P	P	0.871	0.8283	0.992	0.998	NA	2 (1)	3	Refs. 25
101471322	c.2270G>A	p.Gly757Asp	NA	VUS	0.847	0.8068	0.949	0.993	NA	2 (1)	3	NA
101504326	c.643C>T	p.Arg215Trp	Uncertain significance	LP	0.752	0.8779	0.869	1	3.72×10-6	2 (2)	-	Refs. 77,88
101504650	c.319G>C	p.Glu107Gln	Uncertain significance	VUS	0.324	0.7591	0.841	1	6.22×10-7	2 (1)	1	NA
101453055	c.2696T>A	p.Leu899His	NA	VUS	0.755	0.7976	0.928	1	NA	Singleton	3	NA
101453071	c.2680C>A	p.Leu894Ile	NA	VUS	0.581	0.8655	0.876	0.998	NA	Singleton	3	NA
101471308	c.2284G>A	p.Val762Ile	LP	VUS	0.428	0.6783	0.187	0.905	NA	Singleton	3	Refs. 88
101504095	c.874G>T	p.Asp292Tyr	NA	VUS	0.949	0.8756	0.957	1	NA	Singleton	2	NA
101504119	c.850G>A	p.Ala284Thr	NA	VUS	0.789	0.7829	0.525	1	12.39×10-6	Singleton	2	NA
101504440	c.529G>A	p.Val177Met	NA	VUS	0.726	0.7758	0.689	0.999	12.39×10-6	Singleton	1	NA
101504451	c.518A>G	p.Tyr173Cys	LP	VUS	0.755	0.8707	0.699	1	NA	Singleton	1	NA
101504574	c.395T>C	p.Met132Thr	NA	VUS	0.699	0.8407	0.97	0.997	NA	Singleton	1	Refs. 49
101504634	c.335C>G	p.Pro112Arg	NA	VUS	0.638	0.7243	0.966	1	NA	Singleton	1	Refs. 36
101504665	c.304T>Ab	p.Phe102Ile	Uncertain significance	LB	0.925	0.7773	0.958	1	8.10×10-6	Singleton	1	NA

ACMG, American College of Medical Genetics; AF, allele frequency; Fam, family; LB, likely benign; LP, likely pathogenic; NA, not available; P, pathogenic; REVEL, rare exome variant ensemble learner; VUS, variant of uncertain significance.

^aThis variant is reported here because it was found segregating in two families, including one from the original discovery paper from Reiser, et al.²⁷ and in one singleton reported by Sadowski et al.⁶⁴

^bThis variant is reported here since all variant prediction tools classified it as deleterious/pathogenic, the minor allele frequency satisfied our criteria for a positive finding, and the likely benign adjudication from Varsome was exclusively driven by the gnomAD variant count.

Figure 2.

TRPC6 variants clustering on protein domains and structural mechanisms of TRPC6 gain-of-function variants. Above, linearized TRPC6 protein with known domains and positional representation of variants identified in population controls (A) and cases (B). Below, one protomer of the TRCP6 tetramer (protein data bank ID: 7DXF) is represented in gray cartoon. In each panel, D890 and E144 that coordinate a Ca²⁺ ion at the CBS1 site are labeled. This Ca²⁺-coordinating site acts as a sensor that inhibits TRCP6 Ca²⁺ conductance in the setting of high intracellular Ca²⁺ concentration.⁷² The consequences of selected mutations are here modeled on the TRCP6 structure to demonstrate how these missense variants may lead to gain-of-function by disturbing the CBS1 site. Wild-type residues are in gray. Mutant residues are in magenta. (C) van der Waals contacts between E144 and a tyrosine in an adjacent α helix help maintain E144 in a favorable conformation for Ca²⁺ coordination. The Y173C mutation ablates these contacts, increasing the degree of freedom of E144 and likely weakening affinity for Ca²⁺ at the CBS1 site. (D) The CBS1 site is composed of coordinating residues from both more N-terminal (E144) and C-terminal (D890) regions of TRPC6. Ca²⁺ coordination at CBS1 requires a close approach of α-helices from these two regions. The P112R mutation introduces a steric clash between an N-terminal bundle of α-helices and a C-terminal α helix. This clash will push the α helices apart and likely weaken Ca²⁺ coordination at the nearby CBS1 site. (E) V177M introduces a large amino acid into the hydrophobic core of a bundle of α helices, disturbing the packing of these helices and the nearby Ca²⁺-coordination site. (F) The H145R mutation adjacent to E144 substitutes a large, positively charged residue for a smaller polar but uncharged residue. This will cause both steric clashes and electrostatic repulsions that will likely disturb the CBS1 site.

TRPC6 Protein Structure Analysis Revealed Diverse Effect of Disease-Associated Missense Variants

TRCP6 contains three intracellular sensors of Ca²⁺ concentration, termed calcium-binding sites 1–3 (CBS1–3). The first of these, CBS1, has been shown to play an inhibitory role in TRCP6 function, such that when the site is occupied by a Ca²⁺ ion at high levels of intracellular Ca²⁺ ion concentration, ion permeation of the TRPC6 channel is reduced.⁷² CBS1 consists of Ca²⁺-coordinating residues from α-helices in the N-terminal ankyrin repeat domain (ARD) and a C-terminal helix (CH2) of TRPC6. Both the side chains of D890 from the CH2 domain and of E144 from the ARD domain contribute to coordinating a Ca²⁺ ion located in a cleft between these N-terminal and C-terminal regions. Structural and functional studies revealed that perturbing CBS1 reduced its affinity for Ca²⁺ ion binding and led to gain-of-function in TRPC6.⁷² Among the variants that have been functionally validated, several are reported in this study, including p.R895C, p.R895L, and p.E897K. Structural data suggested that these variants would disturb the geometry of CBS1, and functional studies confirmed that they cause gain-of-function.⁷² In this study, several variants seemed to perturb CBS1 by similar mechanisms. p.Y173C probably reduced the stability of CBS1 by removing van der Waals contacts from the Ca²⁺ -coordinating residue, E144 (Figure 2C), p.P112R likely pushed apart the ARD and CH2 domains thus reducing Ca²⁺-affinity at CBS1 (Figure 2D), and p.V177M and p.H145R introduced steric clashes in the local environment of CBS1, likely perturbing the geometry of the site (Figure 2, E and F). This structural analysis is consistent with the previously published mechanistic interpretations of p.R895C, p.R895L, and p.E897K. Given that perturbing CBS1 has been documented in functional studies to induce gain-of-function in TRPC6,⁷² structural analysis of TRPC6 variants that reveals probable disturbance of CBS1 supported the likely causality of those variants. In sum, several of the variants reported in this study are likely to cause TRPC6 gain-of-function by disturbing CBS1 and impairing its inhibitory effect on TRPC6 function. Thus, instead of Ca²⁺-binding to CBS1 in the context of high intracellular Ca²⁺-concentration leading to reduced ion permeation through TRPC6, ion transport remains high regardless of the intracellular state, consistent with gain-of-function.

Kidney Biopsy Analysis Revealed Possible Distinctive Features for TRPC6-AP

Of the 64 TRPC6 variant carriers, 49 received a kidney biopsy. Descriptive diagnostic reports were available for 38 cases, while we had complete pathology reports and images re-review in nine. The 38 diagnostic limited reports showed FSGS in 30, of which one was further characterized as collapsing and one as tip lesion; the other eight were distributed as follows: Two had GBM abnormalities suggesting Alport Syndrome, one had IgA deposition associated to FSGS features, one had C3 glomerulonephropathy characteristics and FSGS, one had a proliferative mesangial glomerulonephropathy with IgM deposition, one had inadequate sampling with limited number of glomeruli and was classified as not sampled FSGS, one was classified with minimal change disease with FSGS features, and one patient received sequential biopsies, the first showing minimal change disease and the second FSGS.

Nine cases with complete pathology reports and images were subjected to re-review (Figure 3 and Supplemental Table 4). The major finding was FSGS not otherwise specified variant in 7/9 cases, focal global glomerulosclerosis with minimal change disease features in two biopsies. The percentage of FSGS lesions ranged from 0% to 67%, the percentage of global glomerulosclerosis from 0% to 50%, and the percentage of tubular atrophy/interstitial fibrosis from 5% to 85%, likely reflecting variable timing of kidney biopsy in the disease course. Among six biopsies with slide review, five had podocyte cell body enlargement and swelling and two had focal glomerular basement membrane duplication (Supplemental Table 4). The electron microscopy findings were notable for foot process effacement that, although variable and ranging from 30% to 100% (mean 85%), was ≥80% in seven out of eight biopsies. Microvillous transformation of the podocyte cytoplasm was commonly observed. The most distinctive features were swelling of the podocyte cell bodies and primary processes. Swollen primary processes often appeared blunted and disordered with irregular profiles. The podocyte cytosol exhibited a loose, open quality with paucity of organelles and focal dilatation of endoplasmic reticulum. No mitochondrial swelling or fragmentation was appreciated. Electron lucent lipid inclusions were noted within some podocyte cell bodies. The usual condensation of actin cytoskeleton forming dense mats was seldom observed in the areas of foot process effacement. Incipient lesions of FSGS displayed lifting and detachment of podocyte cell membranes from the glomerular basement membranes with intervening cellular debris. Glomerular basement membranes of patent capillaries had normal thickness and texture. Some patent glomerular capillaries exhibited focal endothelial cell swelling with mild widening of the subendothelial zone.

Figure 3.

Representative kidney biopsy images of cases with TRPC6-AP. Light and electron microscopy: The histologic and electron microscopy features of four different kidney biopsies are illustrated. (A) A low-power view shows FSGS lesions of the usual (NOS) type involving two glomeruli, accompanied by tubular atrophy, interstitial fibrosis, and chronic inflammation (periodic acid–Schiff, original magnification ×200). (B) The lesions of FSGS exhibit segmental sclerosis and hyalinosis with loss of overlying podocytes and adhesion to Bowman's capsule involving portions of the glomerulus not located at the tubular or vascular poles, consistent with FSGS (NOS) variant (JMS, original magnification ×400). (C) A glomerulus with normal size and cellularity has prominent enlargement and swelling of the podocyte cell bodies (JMS, original magnification ×600). (D) In this electron microscopy figure, the podocyte primary processes appear swollen, blunted, and disordered with irregular profiles. A podocyte cell body has dilatation of endoplasmic reticulum (original magnification, ×5000). (E) A different biopsy with approximately 80% foot process effacement also has dysmorphic bulbous primary processes inserting at right angles directly onto the outer glomerular basement membrane (original magnification ×6000). JMS, Jones methenamine silver; NOS, not otherwise specified; TRPC6-AP, TRPC6-associated podocytopathy.

Natural History Study and Genetic Correlations Indicated Variable Presentation, Poor Variant–Phenotype Correlation, and Frequent Progression to Kidney Failure

Age at clinical recognition of kidney disease varied widely, ranging from the first year of life to late adulthood (0–69 years). Most patients presented in adolescence or early adulthood (median, 21; interquartile range [IQR], 9.5–30; mean, 22; SD ±14 years), and there was no difference in age at clinical recognition between carriers of variants in any of the three clusters, nor in individuals with the two most common variants, p.R895C and p.E897K (Figure 4A). Even in families in which data were available in multiple individuals, age at recognition varied significantly (Table 1). Outcome data were available in 60/64 and, of those, 38 (63%) developed kidney failure, with no significant differences between male or female and carriers of different TRPC6 variants (Figure 4B). On average, 50% of cases progressed to kidney failure in 10–18 years from time of clinical recognition, with no difference among subgroups (Figure 4C). When qualitatively looking only at individuals who did progress to kidney failure, the median time from initial presentation to kidney failure was 5 years, with ample range (0–27 years, SD ±7), and variability was large among subgroups (Figure 4D) and even between family members of the same family (Table 1). These data indicate that TRPC6-AP, once it presented to clinical attention, frequently progressed to kidney failure but with very variable time course. Similarly to other Mendelian forms of FSGS,⁷³ none of the cases that received a kidney transplant experienced a recurrence of disease on the graft.

Figure 4.

Natural history and clinical correlations. (A) Histogram depicting frequencies of age of onset among the affected patients. The distribution is skewed toward younger age of onset, with most patients presenting in adolescence or early adulthood. There is a significant number of patients with childhood-onset disease, but this is likely overrepresented in our cohort as one of our recruiting centers is a children's hospital. (B) Proportion of patients, stratified by sex, who are known to have developed kidney failure. Both male and female patients had significant progression, with 63% of male and 64% of female patients developing kidney failure. The true number is likely even higher, owing to incomplete follow-up. (C) Kidney survival analysis from time to clinical recognition of disease; on average, 50% of individuals reached kidney failure after 15 years from clinical recognition of disease, without difference among males, females, or carriers of specific variants. (D) The time to kidney failure, depicting the number of years from disease onset (or initial clinical recognition) to development of kidney failure. The time to kidney failure was notably short for most of the cohort, with male patients having a nonstatistically significant faster progression time than female patients (P = 0.17). There were no significant differences between the clusters of pathogenic variants, nor among individuals with the most common variants.

Discussion

A missense TRPC6 variant was first implicated by Michelle Winn and colleagues as a cause of inherited podocytopathy in 2005.²⁸ That same year, five additional FSGS families were reported to segregate TRPC6 variants.²⁷ Since that time, at least 20 TRPC6 variants have been reported associated with FSGS (Supplemental Table 2). The majority of identified TRPC6 variants were demonstrated to cause an overall gain-of-function in the resulting translated protein, with increase flux of calcium across the ion channel.^{25–28,59,74–79} This makes TRPC6-AP attractive as a candidate for drug development since it is easier to antagonize a transmembrane channel than to facilitate proper folding, trafficking, or function of a mutated protein.^71,80 Indeed, targeted small molecule inhibitors of TRPC6 exist and are being tested clinically for the treatment of primary FSGS.^81,82 To date, there are no clinical trials specifically targeting only individuals with TRPC6 gain-of-function variants using TRPC6 antagonist designed to provide a proof of concept, and natural history studies of TRPC6-AP are lacking.

We first undertook analysis of large-scale sequencing data to formally test the possible contribution of loss-of-function protein-truncating variants to TRPC6-AP. We identified four heterozygous TRPC6 protein-truncating variants among 5267 individuals with nephrotic syndrome caused by minimal change disease or FSGS. This burden of protein-truncating variants was low and not significantly higher than the one observed in a total of 32,275 healthy or disease controls. Moreover, interrogation of clinical data supported a noncausal, bystander role for these TRPC6 protein-truncating variants. These data strongly support the overall gain-of-function mechanism of disease causation and sanctions drug development approaches to inhibit TRPC6 activity as a safe and effective strategy to treat TRPC6-AP and, perhaps, other forms of nephrotic syndrome. We next studied 64 cases diagnosed with primary podocytopathy from 39 independent families segregating 21 different TRPC6 missense variants, of which 11 were not previously reported. When compared with healthy controls, our cases tended to carry variants affecting one of three distinct regions of the TRPC6 protein. Two of the three clusters (1 and 3) affect regions that are known to interact with each other in the formation of a regulatory intracellular calcium-sensing domain, which in the wild-type TRPC6 protein functions to inhibit opening of the transmembrane cation channel in the presence of high intracellular calcium concentration. Mechanistically, this implies that missense variants causing loss-of-function of the inhibitory calcium-sensing domain result in a gain-of-function of the ion channel's overall cation flux. 3D structural analyses demonstrated this at atomic level detail for p.Y173C, p.P112R, p.V177M, and p.H145R analogously to prior analyses of p.R895C and p.E897K. Nonetheless, not all variants seemed to have a direct effect on the CBS1 site, opening possibilities that TRPC6 variants may induce gain-of-function through affecting other features of the TRPC6 tetramer or through changing TRPC6's interactions with other members of the podocin complex.

Analysis of kidney biopsies showed a FSGS pattern of injury, mostly of the not otherwise specified category. These biopsies showed a significant degree of foot process effacement (mean 85%), reinforcing the concept that, while many forms of monogenic FSGS have an indolent presentation with subnephrotic proteinuria and low degree of foot process effacement, variants in genes encoding for slit diaphragm proteins can have a similar presentation to primary forms of FSGS at electron microscopy and overt nephrotic syndrome.^83,84 Cell bodies and primary foot processes of the podocytes were distinctively and markedly swollen, suggesting a specific mechanism of podocyte damage. Therefore, a pathologic finding of swollen podocyte cells bodies and primary processes should prompt genetic testing, even in the presence of diffuse foot process effacement and full nephrotic syndrome. Although speculative, this similarity in podocyte damage between primary/immune forms of FSGS and TRPC6-AP could have therapeutic implications. The podocyte damage and diffuse foot process effacement are common converging effects of extrarenal injury in primary FSGS while in TRPC6-AP constitutes an intrinsic podocyte insult. Therefore, while the intrinsic podocyte defect in individuals with TRPC6 pathogenic variants renders them unresponsive to immunosuppression therapy, the podocyte-protective effect of TRPC6 inhibitors might make these agents promising drugs to protect the end-damaging effect on podocyte and reduce proteinuria in idiopathic FSGS or other forms of nephrotic syndrome, even in absence of TRPC6 genetic variants. If true, TRPC6 inhibitors could represent nephron-protective agents for different forms of proteinuric kidney disease and help spare patients from toxic immunosuppression or other invasive treatments. This provides another illustration of the opportunities that rare genetic disease targets represent for more common indications.

Clinical correlations were notable for poor prognosis. Age of clinical recognition was variable from congenital/neonatal to adult onset. Once the disease became clinically evident, progression was common, with most patients reaching kidney failure, often within a few years of diagnosis, but with ample variability across individuals and within families, irrespectively of sex and genetic variant. This common and variable progression of disease will present a challenge for defining primary outcomes and end points for intervention studies, but also an unmet opportunity in the development of targeted therapies, clinical trials design, and enrollment criteria. The high rate of kidney failure, combined with the variable time between clinical recognition and kidney failure, argue for intervention strategies to be put in place as early in the course of disease as possible. It also emphasizes the importance of early genetic screening of family members. Such cascade genetic testing will help in identifying patients who are earlier in their clinical course and might most benefit from timely targeted therapy.

Our study had limitations. The retrospective nature of our data collection over decades resulted in incomplete clinical characterization. We obviated this issue by focusing on hard end points, such as kidney failure and progression to kidney failure. Because our patients were drawn from a variety of centers and over a long period of time, and because the clinical presentation of TRPC6-AP is not always abrupt but can be indolent, the true point estimate for age of clinical recognition is likely overestimated in some cases. On the other hand, individuals with family members known to have a genetic form of FSGS were more likely to be tested at an earlier age, resulting in a younger age of disease recognition as compared with sporadic cases. Although the rate of kidney failure was high, it should be considered that patients with more severe disease may have been more likely to be enrolled for genetic testing and study inclusion.

In summary, TRPC6-AP is a rare cause of autosomal dominant FSGS, which tends to present in adolescence or early adulthood, with frequent progression to kidney failure. Large-scale genetic studies supported overall gain of TRPC6 ion conductivity as the main mechanism of disease causation, thus legitimating drug development strategies that are designed to inhibit or reduce TRPC6 function. This study identified key characteristics of disease presentation and clinical course that can facilitate intervention studies design and suggests a common kidney pathology phenotype between TRPC6-AP and primary forms of FSGS. This phenotypic similarity poses the provoking idea that TRPC6 inhibitors could be used in primary forms of FSGS or other common proteinuric kidney disease as podocyte-protective agents in an effort to spare toxic immunosuppression or other invasive treatments.

Supplementary Material

jasn-36-274-s001.pdf ^{(2MB, pdf)}

jasn-36-274-s002.pdf ^{(608.1KB, pdf)}

Disclosures

Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/JSN/E875.

Funding

S. Sanna-Cherchi: National Institute of Diabetes and Digestive and Kidney Diseases (DK122397), US Department of Defense (W81XWH-16-1-0451 and W81XWH-22-1-0966), National Center for Advancing Translational Sciences, National Institutes of Health (UL1TR001873), and National Institute of Health (RC2-DK122397). M.R. Pollak: National Institute of Health (RC2-DK122397 and R01-DK007092). M.G. Sampson: National Institute of Health (RC2-DK122397). F. Hildebrandt: National Institute of Health (RC2-DK122397 and R01-DK076683). K. Kiryluk: NIH (R01-DK105124, R01-DK136765, R01-LM013061, 2U01-HG008680, U01-AI152960, and 5UL1-TR001873). R. Gbadegesin: NIH (1R01-DK134347-01) and NIH/NICHD (1R21-HD104176-01). G. Hall: Harold Amos Medical Faculty Development Program, American Society of Nephrology, and NIDDK (K08-DK111940). A. Beenken: NIDDK (K08-DK132511). A.G. Gharavi: NIDDK (1U01-DK100876) and DOD (W81XWH2110550). M. Kretzler: National Institute of Health (RC2-DK116690).

Author Contributions

Conceptualization: David B. Goldstein, Martin R. Pollak, Simone Sanna-Cherchi, Benjamin Wooden.

Data curation: Simone Sanna-Cherchi.

Formal analysis: Andrew Beenken, Juntao Ke, Tze Y. Lim, Elena Martinelli, Simone Sanna-Cherchi, Benjamin Wooden.

Funding acquisition: Rasheed Gbadegesin, David B. Goldstein, Friedhelm Hildebrandt, Krzysztof Kiryluk, Matthias Kretzler, Martin R. Pollak, Matthew G. Sampson, Simone Sanna-Cherchi.

Investigation: Dina F. Ahram, Gerald B. Appel, Farid Arman, Andrew Beenken, Andrew S. Bomback, David Breckenridge, Elizabeth Colby, Marco Delsante, Pasquale Esposito, Sandro Feriozzi, Enrico Fiaccadori, Rasheed Gbadegesin, Loreto Gesualdo, Ali G. Gharavi, David B. Goldstein, Francesca Guglielmi, Gentzon Hall, Margaret E. Helmut, Gina Jin, Krzysztof Kiryluk, Andrea L. Knob, Matthias Kretzler, Brandon Lane, Umberto Maggiore, Riccardo Magistroni, Elena Martinelli, Adele Mitrotti, Corrado Murtas, Amy J. Osborne, Dirk S. Paul, Slavé Petrovski, Isabella Pisani, Martin R. Pollak, Teresa Rampino, Sunil Sahdeo, Ken Saida, Moin A. Saleem, Matthew G. Sampson, Simone Sanna-Cherchi, Shirlee Shril, Chen Wang, Benjamin Wooden, Francesca Zanoni.

Methodology: Andrew Beenken, David B. Goldstein, Tze Y. Lim, Martin R. Pollak, Simone Sanna-Cherchi, Benjamin Wooden.

Project administration: David B. Goldstein, Martin R. Pollak, Simone Sanna-Cherchi, Benjamin Wooden.

Resources: David B. Goldstein, Martin R. Pollak, Simone Sanna-Cherchi.

Software: Simone Sanna-Cherchi.

Supervision: David B. Goldstein, Martin R. Pollak, Simone Sanna-Cherchi.

Validation: Vivette D. D'Agati, Gina Jin, Jeremiah Martino, Simone Sanna-Cherchi, Dominick Santoriello.

Visualization: Andrew Beenken, Massimo R.A Ferrari, Elena Martinelli, Ken Saida, Simone Sanna-Cherchi, Benjamin Wooden.

Writing – original draft: Andrew Beenken, Elena Martinelli, Simone Sanna-Cherchi, Benjamin Wooden.

Writing – review & editing: Andrew Beenken, David B. Goldstein, Elena Martinelli, Martin R. Pollak, Simone Sanna-Cherchi, Benjamin Wooden.

Data Sharing Statement

All data are included in the manuscript and/or supporting information.

Supplemental Material

This article contains the following supplemental material online at http://links.lww.com/JSN/E874.

Supplemental Methods

Supplemental Figure 1. Burden of TRPC6 loss-of-function protein-truncating variants across 37,542 individuals with exome or genome sequencing.

Supplemental Figure 2. Pedigrees of multigeneration families reported in this study.

Supplemental Table 1. Qualifying protein-truncating variants across phenotypic categories and pairwise comparisons by the Fisher exact test.

Supplemental Table 2. TRPC6 pathogenic/likely pathogenic variants as reported in ClinVar.

Supplemental Table 3. Nephrotic syndrome cases with TRPC6 protein-truncating variants.

Supplemental Table 4. Characteristics of light and electron microscopy of the nine biopsies reviewed.