BASIC RESEARCH www.jasn.org

A Rare Autosomal Dominant Variant in Regulator of Calcineurin Type 1 (RCAN1) Gene Confers Enhanced Calcineurin Activity and May Cause FSGS

Brandon M. Lane ,1 Susan Murray,2 Katherine Benson,3 Agnieszka Bierzynska,4 Megan Chryst-Stangl,1 Liming Wang ,5 Guanghong Wu,1 Gianpiero Cavalleri ,3 Brendan Doyle,6 Neil Fennelly,6 Anthony Dorman,6 Shane Conlon,2 Virginia Vega-Warner,7 Damian Fermin ,7 Poornima Vijayan ,8 Mohammad Azfar Qureshi,8 Shirlee Shril,9 Moumita Barua,8 Friedhelm Hildebrandt,9 Martin Pollak,10 David Howell,11 Matthew G. Sampson,9,12 Moin Saleem ,4 Peter J. Conlon ,2,13 Robert Spurney,5 and Rasheed Gbadegesin 1,5

Due to the number of contributing authors, the affiliations are listed at the end of this article.

ABSTRACT Background Podocyte dysfunction is the main pathologic mechanism driving the development of FSGS and other morphologic types of steroid-resistant nephrotic syndrome (SRNS). Despite significant prog- ress, the genetic causes of most cases of SRNS have yet to be identified. Methods Whole-genome sequencing was performed on 320 individuals from 201 families with familial and sporadic NS/FSGS with no pathogenic mutations in any known NS/FSGS genes. Results Two variants in the gene encoding regulator of calcineurin type 1 (RCAN1) segregate with disease in two families with autosomal dominant FSGS/SRNS. In vitro,lossofRCAN1 reduced human podocyte viability due to increased calcineurin activity. Cells expressing mutant RCAN1 displayed increased calci- neurin activity and NFAT activation that resulted in increased susceptibility to apoptosis compared with wild-type RCAN1. Treatment with GSK-3 inhibitors ameliorated this elevated calcineurin activity, suggest- ing the mutation alters the balance of RCAN1 regulation by GSK-3b, resulting in dysregulated calcineurin activity and apoptosis. Conclusions These data suggest mutations in RCAN1 can cause autosomal dominant FSGS. Despite the widespread use of calcineurin inhibitors in the treatment of NS, genetic mutations in a direct regulator of calcineurin have not been implicated in the etiology of NS/FSGS before this report. The findings highlight the therapeutic potential of targeting RCAN1 regulatory molecules, such as GSK-3b, in the treatment of FSGS.

JASN 32: ccc–ccc, 2021. doi: https://doi.org/10.1681/ASN.2020081234

Glomerular diseases, including diabetic nephropathy, aretheprimaryknowncauseofCKDintheUnited States and the rest of the world.1,2 Most glomerular Received August 27, 2020. Accepted February 25, 2021. diseases are due to primary dysfunction or secondary Published online ahead of print. Publication date available at injury to the podocyte, the visceral epithelial cell of the www.jasn.org. trilayer glomerular filtration barrier. Primary podocyte Correspondence: Dr. Rasheed Gbadegesin, Division of Ne- dysfunction, referred to as podocytopathy, typically phrology, Duke Molecular Physiology Institute, Carmichael manifests as steroid-resistant nephrotic syndrome building, RM 51-104, 300 N Duke St., Durham, NC 27701-2047. (SRNS) with morphologic changes of FSGS or minimal Email: [email protected] change disease apparent on kidney biopsy specimens.3,4 Copyright © 2021 by the American Society of Nephrology

JASN 32: ccc–ccc, 2021 ISSN : 1046-6673/3207-ccc 1 BASIC RESEARCH www.jasn.org

It is estimated that 5%–30% of all podocytopathies are due Significance Statement to mutation in single genes, especially in children and young adults.5–7 More than 60 genes have been identified as causes of Whole-genome sequencing of 320 individuals with nephrotic syn- monogenic SRNS; however, these genes are responsible for drome (NS) of unclear genetic etiology and data from several only 20% of all genetic SRNS, suggesting there are other, un- independent patient cohorts provided insight into the genetic ar- chitecture of the condition. The strategy identified a disease- fi 6,8–11 fi identi ed, single-gene causes of SRNS. Identi cation of causing autosomal dominant mutation in regulator of calcineurin these causal genes has the potential to improve our under- type 1 (RCAN1) that increased cellular calcineurin (CN) activity, standing of disease pathogenesis, the identification of disease NFAT (NF of activated T cells) activation, and susceptibility to ap- biomarkers, the identification of new therapeutic agents, and optosis of podocytes in vitro. Inhibition of an RCAN regulator, GSK- b RCAN1 the repurposing of existing agents to treat nephrotic 3 , rescued the increased CN activation. Mutations in are a novel cause of NS and reveal a potential target for developing syndrome (NS). personalized therapy. To identify new, single-gene causes of SRNS, we carried out whole-genome sequencing (WGS) on 320 individuals from 201 families with familial and sporadic NS, and reviewed METHODS whole-exome sequencing data from patients with NS of un- clear genetic etiology. We identified two segregating, hetero- WGS fi zygous mutations in the regulator of calcineurin (CN) type 1 WGS was performed at GENEWIZ (South Plain eld, NJ). fl (RCAN1) in two large Northern European families. There are Brie y, genomic DNA samples were assessed for purity, quan- three genes in the RCAN family RCAN1–3, all of which encode tity, and quality by using the NanoDrop 2000 Spectrophotom- proteins capable of interacting with CN and inhibiting CN- eter (Thermo Fisher), Qubit 2.0 Fluorometer, Qubit dsDNA dependent signaling pathways.12–22 Therefore, we screened HS Assay Kit (Thermo Fisher), and agarose gel electrophore- families with hereditary and sporadic NS in other independent sis. Library construction was then performed using Illumina’s cohorts for rare variants in RCAN1–3 genes. We identified TruSeq DNA PCR-Free library preparation kit following the four possible disease-causing variants: three in RCAN2,and manufacturer’s protocol. Genomic DNA was fragmented by one in RCAN3. acoustic shearing with a Covaris S220 instrument. Sheared The RCAN family of proteins form a complex with the DNA was then end repaired and A-tailed, followed by adaptor catalytic subunit of CN, and regulate both CN phosphatase ligation. Final libraries were analyzed on the Agilent TapeSta- activity and its ability to bind key substrates like NF of acti- tion, for library sizing, and quantified using the Qubit dsDNA vated T cells (NFAT).12–14,16–18,20,23–27 Unregulated CN acti- HS Assay Kit along with the KAPA Library Quantification Kit vation is central to the pathogenesis of multiple glomerular for quantitative PCR. DNA libraries were sequenced using disease processes, and CN inhibitors (CNIs) are often used to Illumina platforms to generate $120 Gb of raw data per sam- treat glomerular diseases.28–36 The rationale for treating ac- ple, with a 23150-bp, paired-end sequencing configuration. quired forms of NS with CNIs has historically been that the immune system was thought to play a significant role in ac- Variant Calling and Annotation quired forms of NS, such as minimal change disease DNA-sequencing data were processed using fastp1 to trim low- and FSGS.28,29 However, CN has nonimmunologic actions quality bases and Illumina sequencing adapters from the 39 end that are important in the pathogenesis of kidney diseases. of reads.39 Reads were then aligned to the GRCh37 version of the For example, CN causes cytoskeletal instability by dephosphor- human genome with the BWA2 algorithm.40 PCR duplicates ylating synaptopodin and promoting its degradation.28–31,33 were flagged using the PICARD Tools3 software suite.41 Align- Moreover, podocyte loss plays a key role in pathogenesis ment processing and variant calling were performed using the of FSGS, and CN promotes a decrease in the number of GATK4 toolkit following the Broad Institute’s Best Practices glomerular podocytes by both genetic and nongenetic Workflow.42,43 Functional consequences and genotype prove- mechanisms.28–31,34,37,38 nances of variants were annotated using Ensembl Variant In this study, we discovered that a disease-causing RCAN1 Predictor.44 After annotation, variants meeting the following variant in individuals with FSGS had a reduced ability to in- criteria were selected for further analysis: having a “pass” status hibit activated CN compared with wild-type (WT) RCAN1. after GATK’s Variant Quality Score Recalibration, found to re- The increase in CN activation induced by the RCAN1 variant side in a coding region, and had an allele frequency of ,5% in at was inhibited by treatment with antagonists of glycogen syn- least one population of the Genome Aggregation Database (gno- thase kinase 3 (GSK-3). In addition, cells expressing this mAD).45 Second-level filtering to identify disease-causing vari- RCAN1 variant were more sensitive to apoptotic stimuli, ants is as shown in Supplemental Figure 1. Variants of interest which could be rescued by CNI treatment. Collectively, our were confirmed by Sanger sequencing. findings suggest mutations in RCAN1 are a novel genetic cause of NS, and use of CNIs and GSK antagonists may represent RCAN1 Knockdown Podocytes targeted or personalized therapy for individuals with NS/FSGS Multiple, conditionally immortalized, human podocyte lines due to RCAN1 mutations. (courtesy of Dr. Jeffrey Kopp) with reduced RCAN1 expression

2 JASN JASN 32: ccc–ccc,2021 www.jasn.org BASIC RESEARCH were created using lentiviral transduction of short hairpin RNA total phosphatase levels. EGTA-supplemented buffer was (shRNA) against RCAN1 (TRCN0000019848; Millipore added to additional wells for each sample to measure the Sigma). Lentiviral control lines were created using shRNA non-CN phosphatase activity. All sample wells received water with no known target (SHC016V; Millipore Sigma). Podocyte and CN substrate (RII phosphopeptide), except for back- lentiviral transduction was performed as described previ- ground wells in which water was substituted for substrate. A ously.46 RCAN1 KD was confirmed through immunoblotting positive control (CN enzyme supplied by the kit) was used to (LS-C162511; LifeSpan Biosciences). ensure assay effectiveness. The plate was incubated at 30°C for 10 minutes, lysate was added to all sample wells, and then it Immunoprecipitation Studies was incubated at 30°C for 30 minutes. BioMol Green reagent Immunoprecipitation was performed using a protocol modi- (100 ml) was then added to all wells and incubated for 25 min- 13 fied from Fuentes et al. For the studies, human embryonic utes at room temperature before reading the OD620nm using a kidney cells (HEK293) cells were grown in DMEM supple- Tecan (Männedorf, Switzerland) Infinite 200 microplate mented with 10% FCS, penicillin (100 U/ml), and streptomy- reader, with two reads per well. Background well readings cin (100 mg/ml) (all from Gibco, Gaithersburg, MD), as were subtracted from all experimental well readings, and mo- previously described.47 For transfection, HEK293 cells were les of phosphate were calculated using the phosphate standard plated in six-well Costar tissue culture plates (Corning, Corn- curve. The CN-specific phosphatase activity was calculated by ing, NY) and grown to approximately 80% confluency. Cells subtracting the phosphate present in EGTA-treated cells were then cotransfected with the FLAG-tagged CN construct (non–CN-related phosphatase activity) from calmodulin- (GenScript, Piscataway, NJ) and the Myc-tagged RCAN1 treated cells (total phosphatase activity). All experiments construct (WT or mutant as indicated; GenScript) using Lip- were repeated in triplicate. ofectamine 2000, according to the manufacturer’s recommen- dations (ThermoFisher Scientific, Waltham, MA). Cells were GSK-3 Inhibition harvested 48 hours after transfection, and cell pellets were We diluted LY2090314 (Selleck Chemicals, Houston, TX) and lysed in ice-cold 50 mM Tris-hydrochloride (pH 7.5), tideglusib (Selleck Chemicals) in water (1:2000) from DMSO 150 mM sodium chloride, and 1% nonidet P-40 in the pres- stock (10 mM and 2 mM, respectively) and replaced 2 mlof ence of protease inhibitors (Protease Inhibitor Cocktail; water in the CN activity assay described above with an equal Sigma-Aldrich, St. Louis, MO). For immunoprecipitation of volume of diluted GSK inhibitors to reach a final concentra- RCAN1 constructs, protein lysates were incubated with anti- tions of 200 nM LY2090314 and 1 mM tideglusib. Untreated Myc antibodies (ThermoFisher Scientific) at 4°C for 1 hour, samples wells received 2 ml of 1:2000 diluted DMSO. To ac- and then Protein A Plus Protein G (Millipore, Bedford, MA) commodate the additional sample conditions for these inhi- was added to the lysate and rocked for approximately 4 hours bition studies, we used cell lysate(SPACE)te diluted 1:1 in lysis at 4°C. For immunoprecipitation of CN constructs, protein buffer. CN activity was calculated by subtracting the phos- lysates were incubated with anti-FLAG antibodies linked to phate activity in EGTA-treated wells from the activity in sepharose beads (Cell Signaling Technology, Danvers, MA) treated or untreated wells. This experiment was repeated and rocked for approximately 4 hours at 4°C. After three in triplicate. washes with ice-cold lysis buffer, Laemmli sample buffer was added to the pellet and boiled for approximately 10 minutes, NFAT Luciferase Assay and immunoblotting was then performed. This experiment The NFAT luciferase assay was performed using the Dual Fire- was repeated in triplicate. fly and Renilla Luciferase Assay Kit (Biotium, Freemont, CA), according to the manufacturer’s protocol. Briefly, HEK293 cells CN Activity Assay were grown in 24-well plates and transfected with Lipofecta- CN activity was examined using the Cellular Calcineurin Ac- mine 2000 according to the manufacturer’s protocol. Cells were tivity assay kit according to the manufacturer’s protocol (Enzo transfected with equal parts of an NFAT-luciferase reporter Life Sciences, Farmingdale, NY). Conditionally immortalized construct (Promega), loading control construct (pRL-TSK; podocytes were grown on collagen-coated, six-well dishes un- Promega), PPP3CA,andoneoftheRCAN1 constructs til confluent. HEK293 cells were grown in six-well dishes and (0.6 ng DNA per construct per well). Lysates were harvested transfected with PPP3CA and RCAN1 constructs using Lip- using the supplied lysis buffer after 48 hours and plated in ofectamine 2000, as described above, for 48 hours. Immuno- duplicate on 96-well plates. Measurements were taken using blotting was performed to ensure equal levels of PPP3CA and an Infinite Pro 200 microplate reader with automated injection RCAN1 transfection between HEK293 cell samples. All exper- (Tecan), which injected 100 ml of luciferase assay reagent into imental cells were washed with Tris-buffered saline before ly- each well, recorded the fluorescence, added 100 ml of Renilla sates were harvested and cleared of free nucleotides using a assay reagent, and performed another fluorescence reading. desalting column. Phosphate standards were loaded in dupli- The relative luminescence units for each well were then calcu- cate on a 96-well plate. Calmodulin-supplemented buffer was lated by dividing the luciferase reagent readings by the Renilla added to appropriate wells for measuring the background and reagent readings. The experiment was repeated in triplicate.

JASN 32: ccc–ccc, 2021 RCAN1 Variants Can Cause NS 3 BASIC RESEARCH www.jasn.org

Automated Cell Apoptosis Imaging readings. The experiment was repeated in quadruplicate with a total Toboth visualize and quantify the apoptosis and total cell death, we Nof at least 16 for each cell type, and full videos of the representative used a Lionheart FX automated microscope from BioTek along HEK293 cell images are available in Supplemental Videos 1–3. with fluorescent apoptosis reagents. Podocytes were plated and grown to confluency before beginning the assay. HEK293 cells Three-Dimensional In Silico Protein Modeling were grown in 96-well plates, as described above, and transfected Molecular graphics and analyses of PDB files created in the with PPP3CA and RCAN1 constructs using Lipofectamine LTX, I-TASSER software48,49 was performed with UCSF ChimeraX, according to the manufacturer’s protocol, for 48 hours before be- which was developed by the Resource for Biocomputing, Vi- ginning the assay. Cells were exposed to serum-free media contain- sualization, and Informatics at the University of California, ing a 1:500 dilution of NucView Caspase-3 Alexa 488 (Biotium) San Francisco, with support from National Institutes of Health and a 1:2000 dilution of propidium iodide (Sigma-Aldrich). The (NIH) (R01-GM129325) and the Office of Cyber Infrastruc- NucView reagent consists of a substrate of caspase-3 that emits ture and Computational Biology, National Institute of Allergy green fluorescence when cleaved, whereas propidium iodide fluo- and Infectious Diseases.50 resces in late apoptotic and necrotic cells. These media also con- tained either 1 mM FK506 or an equal concentration of vehicle Illustrations (ethanol). Bright-fieldimages,alongwithgreenandredfluorescent The summary graphic was created using Biorender.com. images, were collected every 2 hours for 48 hours. Using automated GEN5 software from BioTek, the images were processed to remove Immunoblotting background, and the number of fluorescent cells was quantified Immunoblotting was performed using standard methods for each well using label-free cell counting. Wells containing full and visualized by enhanced chemiluminescence, as previously de- serum were used as a control to test the validity of the apoptosis scribed.51 Antibodies were used at the following concentrations:

A 40030 PEDIGREE

? * ? ?? ? 10017 10015 10126 10559 10124 11041 10557 10016 10585 10586 28141 L I/T S L I/T S L I/T S L I/T S LIS L I/T S

CCTTTA /G CCC CCTTTA /G CCC CCTTTA /G CCC CCTTTA /G CCCCCCTTTG A C CCTTTA /G CCC

?? ? ? ? ? ? 10027 10568 10231 10230 10056 11067 10228 10229 10052 10486 LIS L I/T S L I/T S

CCCTTTG A C CCTTTA /G CCC CCTTTA /G CCC

? 11220 BCDE

Figure 1. The RCAN1 p.I162T mutation segegrates with disesase in a family with FSGS. (A) Pedigree of European family 40030 with FSGS and the RCAN1 I162T variant segregating with the disease in the family. Family members that are currently unnaffected but may develop disease later in life are depicted with a question mark. Sequenced individuals are shown with a chromatogram and associated amino acid sequence (L5Leucine, I5 Isoleucine, T5 Threonine, S5Serine). Asterisk indicates obligate carrier. (B–E) Kidney histology from individual 10557 in family 40030. (B and C) FSGS on hematoxylin and eosin staining at (red arrow). (D) Mild foot process ef- facement (red arrows) and thinned glomerular basement membrane. (E) Capillary loop double contour formation (red arrows) on silver staining. Original magnification, 320 in (B), 340 in (C) and (D).

4 JASN JASN 32: ccc–ccc,2021 www.jasn.org BASIC RESEARCH

Table 1. In silico prediction of disease-causing RCAN1 heterozygous variant Variant gnomAD (EUR) MAF (EUR) MAF (all population) CADD PolyPhen SIFT MutationTaster RCAN1 c. T485C, p. I162T 2 of 128,738 0.00001 0.000007 26.7 Probably damaging Damaging Disease causing GRCH37: RCAN1-001, transcript ENST00000313806.4. EUR, European; MAF, minor allele frequency; CADD, combined annotation dependent depletion; SIFT, sorting intolerant from tolerant.

1:500 for caspase-3 (Cell Signaling Technology), 1:1000 for MYC harvesting, RNA extraction (RNAeasy kit; Qiagen), and gener- tag (Cell Signaling Technology), 1:1000 for DYKDDDDK tag (Cell ation of cDNA (Promega), as previously described.52 TaqMan Signaling Technologies), and 1:3000 for b-actin (Sigma-Aldrich). probes (Invitrogen) were used in to analyze gene expression for For apoptosis experiments, HEK293 cells were grown in six-well CD2AP (Hs00961451_m1), RCAN1 (Hs01120954_m1), plates and transfected with PPP3CA and RCAN1 constructs using RCAN2 (Hs00195165_m1), RCAN3 (Hs00203728_m1), and Lipofectamine LTX,according to the manufacturer’s protocol. After PTPRO (Hs00958177_m1). The analysis was repeated in trip- 48 hours of serum starvation, the cells were washed with PBS, licate, with multiple wells per sample in each replicate. harvested, and the lysates were analyzed with immunoblotting. The experiments were repeated in triplicate and quantified using Electron Microscopy of Renal Biopsy Specimens ImageJ software. UnmodifiedWesternblotimagesareshownin The harmonic mean of the glomerular basement membrane the Supplemental Materials. thickness was calculated using multiple measurements, using reference ranges for men and women as reported by Das et al.53 Quantitative Real-Time PCR Conditionally immortalized, human podocytes (generously Statistical Analyses provided by Dr. Jeffrey Kopp) were grown until confluent in The two-tailed t test was used for the comparison of RCAN1- T75 and differentiated at 37° for 14 days before lysate KD podocyte immunoblotting against RCAN1 (t59.802;

A FLISPPxSPP PK||Q TxxP RCAN1 RRM LxxP ExxP Px|x|T 252

K128E I162T*

FLISPPxSPP PK||Q TxxP RCAN2 RRM LxxP ExxP Px|x|T 243

P149T R234H N243H

FLISPPxSPP TxxP RCAN3 RRM LxxP ExxP Px|x|T 241 C173G

B RCAN1 WT RCAN1 I162T

Figure 2. Potentially pathogenic variants disrupt conserved RCAN protein domains. (A) Depictions of RCAN1, RCAN2, and RCAN3 peptides showing the conserved protein domains within the RCAN1 family of proteins and locations (arrows) of the newly identified variants. The variants identified in these patients are all located in or near conserved domains, including the RNA recognition motif (RRM; RNA binding–like) domain (blue-green); the carboxy-terminal CN binding motifs (orange and red); and the SP motif (green-yellow), which contains the LxxP, FLISPPxSPP, and ExxxP sequences. (B) Three-dimensional modeling of WT RCAN1 and the p.I162T variant revealed disruption of the amino- and carboxy- terminal regions (blue and red, respectively), and the region around the SP motif (green), when compared with WT RCAN1.

JASN 32: ccc–ccc, 2021 RCAN1 Variants Can Cause NS 5 BASIC RESEARCH www.jasn.org

Table 2. Phenotype of individuals with RCAN1 mutations Family Number Individual Number Age at Onset (yr) Proteinuria (quantity) Biopsy (histology) CKD Stage Transplant Recurrence 40030 10056 29 UNK N 5 Y N 40030a 10126 NA N N 0 N NA 40030 10559 48 Y Y (UNK) 5 Y UNK 40030 10552 35 Y (825 mg) Y (FSGS) 5 N NA 40030 10557 65 Y (5000 mg) Y (FSGS) 5 N NA 40030a 10586 NA N N 0 N NA 40030 10017 UNK UNK UNK UNK UNK UNK 40030 10229 45 Y(.3000 mg) N 1 N NA 6559 1 5 Y (1860 mg) Y (FSGS) 5 Y N 6559 101 40 Y (2410 mg) UNK 1 N NA 6559a 0106 NA N N 0 N NA UNK, unknown; N, no; Y, yes; NA, not available. aAsymptomatic. degrees of freedom [df]56) and cleaved caspase-3 (t54.764; variants using previously published algorithms for the identi- df56). Two-way ANOVA, followed by a Dunnett multiple fication of causal variants in families with Mendelian kidney comparison analysis, was used to analyze automated live im- disease (Supplemental Figure 1). We identified four variants in aging of RCAN1-KD podocytes (df52; F57.698). One-way four genes that are present in all affected individuals ANOVA, followed by a Tukey multiple comparisons test, was (Supplemental Table 1) in a large Irish family recruited as used to determine the differences between means for the anal- part of our collaboration with the Irish Kidney Gene Project ysis of RCAN1-variant apoptosis immunoblotting results (Figure 1).54,55 One of these variants was in RCAN1 (c. T485C, (df53; F556.94). A two-tailed t test analysis was used for p.I162T; transcript, ENST00000313806.4; GRCh37). A search 5 5 the variant NFAT luciferase assay (t 11.84; df 10) and the for pathogenic variants in these four genes in existing whole- 5 5 CN activity assay (t 6.579; df 22). Two-way ANOVA, fol- exome sequencing data from 191 families with NS of unclear lowed by a Dunnett multiple comparisons analysis, was used etiology identified a second family with another segregating to compare groups for RCAN1-variant automated live-cell variant in RCAN1 (Supplemental Figure 2), but did not find 5 5 apoptosis imaging (df 3; F 10.92) and GSK-3 inhibition pathogenic variants in the other three candidate genes. The 5 5 experiments (df 2; F 27.03). second variant, c.A382G, p.K128E, is rare, with a minor allele frequency of 0.0004172 in the gnomAD database, and both RCAN1 variants are conserved in evolution (Supplemental RESULTS Table 2).

Clinical Ascertainment and WGS In Silico Modeling We identified 320 individuals from 201 families with familial In silico modeling revealed the p.I162T RCAN1 variant is and sporadic NS/FSGS with no pathogenic mutations in any predicted to be damaging by at least three prediction tools, known NS/FSGS genes. We performed WGS and filtered with a Combined Annotation Dependent Deletion score

Table 3. Rare heterozygous RCAN variants in NS cohorts Study Number Phenotype Variant Allele Count gnomAD MAF (All) PolyPhen SIFT MutationTaster Conservation 159 SRNS, MCD RCAN2 0 0.000000 Damaging Probably Disease causing Zebrafish c.C445A damaging p.P149T 260 SRNS, FSGS RCAN2 2 of 248,944 0.000008 Damaging Probably Disease causing Frog c.A728C damaging p.N243H 3NSRCAN2 3 of 249,306 0.00001 Damaging Probably Disease causing Zebrafish c.C700T damaging p.R234H 459 SSNS, FR/SD RCAN3 0 0.000000 Tolerated Probably Disease causing Zebrafish c.T517G damaging p.C173G GRCH37: RCAN1-001, transcript ENST00000313806.4; RCAN2-002, transcript ENST00000371374.1; RCAN3-001, transcript ENST00000374395.4. MAF, minor allele frequency; MCD, minimal change disease; SSNS, steroid-sensitive NS; FR/SD, frequent relapsing/steroid-dependent; SIFT, sorting intolerant from tolerant.

6 JASN JASN 32: ccc–ccc,2021 www.jasn.org BASIC RESEARCH

ABPodocyte RCAN1 KD Quantification CRCAN1 KD CN activity Control RCAN1 KD 1.5 15 Podocyte podocyte *p<0.0001 *p=0.0003 10 RCAN1 1.0

5 -actin 0.5 0 Relative RCAN1 /  -Actin Relative Calcineurin Activity 0.0 -5 Control RCAN1 KD Control RCAN1 KD Podocyte RCAN1 KD Apoptosis podocyte podocyte podocyte podocyte D 45 RCAN1 KD E F 40 Control podocyte RCAN1 KD Apoptosis Quantification FK506 treated RCAN1 KD 2.0 35 Control RCAN1 KD *p=0.0031 30 podocyte podocyte 1.5 25 Cleaved Caspase 3 20 1.0

15 3/  -Actin -actin

Podocyte Apoptosis % 0.5 10

5 Relative Cleaved Caspase 0.0 0 Control RCAN1 KD 0 8 16 24 32 40 48 56 64 72 podocyte podocyte Hours with Serum Starvation

Figure 3. Loss of RCAN1 increases CN activity and reduces viability in human podocytes. (A) Immunoblotting and (B) subsequent quantification revealed a significant reduction in RCAN1 expression in conditionally immortalized podocyte cell lines transduced with shRNA against RCAN1 (RCAN1 KD) when compared with lentiviral control podocytes (P,0.001, n54, t test). (C) RCAN1 KD podocytes had increased levels of CN activity compared with controls (P50.0003, n512, t test). (D) RCAN1 KD podocytes exposed to serum starvation displayed an increased susceptibility to apoptosis, as measured by automated live-cell imaging of a fluorescent reporter of cleaved caspase-3 activity (P,0.01 for all time points after 30 hours, n.16 for each group, two-way ANOVA) and (E and F) immu- noblotting (P50.003, n54, t test). This apoptosis could be rescued by CNI treatment (1 mM FK506), demonstrating the contributions of altered CN regulation to podocyte viability (P.0.35 for all time points).

.20 (Table 1). Modeling the mutation in the three- Two individuals from the index family carry an RCAN1 vari- dimensional structure of RCAN1, using I-TASSER and Chi- ant but did not present with kidney disease at the time of meraX software, revealed that this mutation is predicted to ascertainment and follow-up, whereas one individual in the disrupt both the amino and carboxy termini of the protein second family also carries the second RCAN1 variant without involved in CN binding, and the highly conserved serine- any evidence of kidney disease (Table 2). Our data suggest an proline (SP) motif, which is required for regulation of CN autosomal dominant pattern of inheritance with incomplete activity (Figure 2, Supplemental Videos 1 and 2).48,49,56 The penetrance. SP motif also contains GSK-3 and other kinase phosphoryla- tion sites that can alter the activity of RCAN1.19,26,57,58 Sequencing RCAN1, RCAN2, and RCAN3 in Independent Cohorts Phenotype of Individuals with RCAN1 Mutations There are three genes in the RCAN family of genes, RCAN1 The phenotypes of the eight affected individuals from the two (chromosome 21q 22.11), RCAN2 (chromosome 6p12.3), and families with RCAN1 variants are displayed in Table 2. Af- RCAN3 (chromosome 1p 36.11).13,21,22 Because all of the pro- fected individuals presented between the ages of 5 and 65 years. teins encoded by RCAN genes interact with CN and modulate Six individuals had proteinuria at presentation, four of which CN-dependent signaling pathways, and to provide supportive presented with nephrotic-range proteinuria. However, one of evidence for the pathogenicity of RCAN1 variants, we the two individuals with unknown proteinuria at presenta- screened for variants in RCAN1–3 in additional cohorts of tion ultimately progressed to ESKD. Biopsy specimen reports patients with NS/FSGS. We screened patients for rare variants were available for three individuals, and all had FSGS on light- in the three genes in five other independent cohorts (cohorts microscopy results and podocyte foot process effacement on from the Boston Children’s Hospital and Beth Israel Hospital electron-microscopy results (Figure 1). Three patients had un- Boston, the Nephrotic Syndrome Study Network Consortium dergone a kidney transplant, follow-up data are available in [NEPTUNE] cohort, UK NephroS cohort, and a cohort from two of these individuals, and none of these patients developed the Toronto General Hospital) (Supplemental Table 3). We recurrence of primary disease after the kidney transplant. identified four potentially pathogenic variants in RCAN2

JASN 32: ccc–ccc, 2021 RCAN1 Variants Can Cause NS 7 BASIC RESEARCH www.jasn.org and RCAN3 (Table 3). The RCAN3 variant and one of the cell lysates using a cellular CN activity assay revealed the three RCAN2 variants are novel and they are not found in p.I162T variants disrupted the regulatory function of the gnomAD (approximately 250,000 chromosomes ana- RCAN1, resulting in increased CN activity compared with lyzed). The other two variants in RCAN2 have minor allele WT RCAN1–expressing cells (P,0.001). An NFAT luciferase frequency of #0.00001 in gnomAD. All of the variants are assay confirmed this increased CN activity resulted in elevated predicted to be damaging by three in silico prediction tools, NFAT activation compared with WT RCAN1–expressing cells and they are all conserved in evolution. Other missense vari- (P,0.001) (Figure 4). To verify the effectiveness of our con- ants found in the three genes are listed in Supplemental structs, we also examined CN activity in cells transfected with Table 4. either PPP3CA or RCAN1 WT alone to ensure CN activation increased and decreased, respectively, in these assays, as com- Loss of RCAN1 Disrupts Podocyte CN Regulation and pared with untransfected controls (Supplemental Figure 5). Decreases Podocyte Viability To determine the relevance of RCAN1 in the maintenance of RCAN1 p.I162T Induces Increased Apoptosis that Can podocyte functional integrity, we first confirmed expression in Be Rescued by CN Inhibition podocytes through quantitative real-time PCR of condition- Transfected HEK293 cell lines expressing PPP3CA and WTor ally immortalized, human podocyte cell lines. RCAN1, mutant RCAN1 were exposed to serum starvation–induced RCAN2,andRCAN3 are all expressed in podocytes at compa- apoptosis and evaluated using both automated live-cell imag- rable levels with key podocyte genes, such as CD2AP and ing and Western blot quantification of caspase-3 activity. PTPRO (GLEPP1) (Supplemental Figure 3). With the known Overexpression of RCAN1 p.I162T induced a significant in- role of RCAN proteins in CN regulation, we examined the crease in apoptosis and total cell death relative to the WT effects of loss-of-function RCAN1 mutations on podocyte RCAN1–expressing cells (Figure 5, Supplemental Figures 6 CN activity using shRNA-mediated RCAN1 KD in condition- and 7, and Supplemental Videos 3 and 4). Pretreatment with ally immortalized podocytes. As expected, podocytes with re- FK506 rescued the increased apoptosis phenotype in the mu- ducedfunctionalRCAN1displayedincreasedCNactivity tant cell lines, confirming the increased apoptosis in the compared with WT controls (Figure 3). RCAN1 p.I162T–expressing cell lines is due to increased CN Increased CN activity is known to induce podocyte apo- activity (Figure 5, Supplemental Video 5). ptosis both in vitro and in vivo, a key phenotype associated with FSGS.38 To examine the effects of decreased RCAN1- mediated CN regulation on podocyte viability, we examined the susceptibility of podocytes to serum starvation using au- Calcineurin Activity Assay fi A tomated live-cell imaging and quanti cation of cleaved 200 p<0.0001 caspase-3 activity. RCAN1-KD podocytes displayed increased 150 apoptosis compared with WT controls that could be rescued by treatment with CN inhibition (FK506) (Figure 3). These 100 PPP3CA + RCAN1 WT data suggest loss-of-function variants in RCAN1 have the 50 PPP3CA + RCAN1 |162T potential to alter podocyte CN regulation and reduce 0 cell viability. Relative Calcineurin Activity -50

RCAN1 Variants and CN Binding B NFAT-Luciferase Assay Regulation of CN by RCAN1 is mediated by direct protein- 1.3 protein interactions between RCAN1 and the catalytic subunit p<0.0001 of the phosphatase.12,13,16–18 To determine if the RCAN1 mu- 1.2 PPP3CA + RCAN1 WT tations identified in the affected families altered docking of 1.1 RCAN1 and CN, we performed immunoprecipitation studies PPP3CA + RCAN1 |162T in HEK293 cells expressing PPP3CA (CN catalytic subunit) 1.0 and WT RCAN1 or the p.I162T variant, as described in the Relative NFAT expression 0.9 Methods . As shown in Supplemental Figure 4, this RCAN1 mutation did not disrupt the protein-protein interaction be- Figure 4. CN and NFAT activity is increased in RCAN1 mutant tween RCAN1 and CN. cell lines. (A) HEK293 cells were transfected with constructs containing PPP3CA (CN) and either WT RCAN1 or the p.I162T RCAN1 RCAN1 variant. At 48 hours after transfection, the RCAN1 p.I162T Enhances CN Activation p.I162T variant–expressing cells displayed increased CN activity ’ To investigate the effects of the p.I162T variant on RCAN1 s (P,0.001, n512, t test) and (B) NFAT expression (P,0.001, n56, function as a regulator of CN activity, we examined the CN t test) compared with WT RCAN1–expressing cells. For the NFAT activity of HEK293 cells that were transfected with PPP3CA luciferase assay, cells were also transfected with constructs to and RCAN1 constructs (WT or p.I162T). Evaluation of these allow for both Firefly and Renilla dual-luminescence detection.

8 JASN JASN 32: ccc–ccc,2021 www.jasn.org BASIC RESEARCH

A B Cleaved Propidium RCAN1 Apoptosis Imaging Brightfield Caspase-3 Iodide CC3 + BF PI + BF Merge 40 35 PPP3CA + PPP3CA + RCAN1 |162T 30 RCAN1 WT PPP3CA + RCAN1 WT 25 FK506 treated 20 PPP3CA + RCAN1 |162T PPP3CA + 15 FK506 treated RCAN1 I162T PPP3CA + RCAN1 WT Apoptosis cell % 10 FK506 treated 5 PPP3CA + 0 RCAN1 I162T 0 4 8 12162024 Hours with Serum Starvation C D RCAN1 Immunoblot Apoptosis Cleaved 1.5 p=0.0378 Caspase 3 PPP3CA + RCAN1 WT Myc 1.0 PPP3CA + RCAN1 |162T FK506 treated -actin PPP3CA+ RCAN1 |162T Flag PPP3CA + + + + 0.5 FK506 treated Myc RCAN1 WT + - - + PPP3CA+ RCAN1 WT

Myc RCAN1 I162T - + + - Relative CC3 /  -Actin 0.0 FK506 600nm - - + +

Figure 5. Mutant RCAN1 causes increased apoptosis that can be rescue by CNI FK506. (A) HEK293 cells were transfected with constructs containing PPP3CA (CN) and either WT RCAN1 or the p.I162T variant, and the cells were exposed to serum deprivation. We analyzed the susceptibility to apoptosis using a fluorescent reporter of caspase-3 activity over 24 hours. RCAN1 I162T–expressing cells (red) displayed increased apoptosis compared with WT RCAN1 cells (black) (P,0.02 for all time points between 18 and 24 hours, two- way ANOVA). This increased apoptosis in the RCAN1 mutants was rescued by treatment with 1 mMFK506(P.0.3 for all time points), demonstrating this aberrant apoptosis in mutant cells is due to the increased CN activity (n.16 for all samples). (B) This increased apoptosis could be seen in still images taken 24 hours after serum starvation, which showed increased apoptosis (green, cleaved caspase-3 [CC3]) and necrosis (red, propidium iodide [PI]) in RCAN1 I162T–expressing cells compared with WT. (C and D) The in- creased apoptosis and rescue was confirmed through Western blot analysis of cleaved caspase-3 expression after 48 hours of serum starvation (P50.02, n53, one-way ANOVA). BF, bright-field imaging.

RCAN1 Mutations Can Cause Disease through Altered dysregulated CN activity, which promotes apoptosis in podo- GSK-3 Signaling cytes. Furthermore, the deficiencies in CN regulation caused Having ruled out deficiencies in CN binding as a driving force by RCAN1 p.I162T are likely due to structural changes that of increased CN activity in mutant RCAN1–expressing cells, we affect critical interactions with GSK-3b. These particular examined the regulation of RCAN1 activity. A key feature of RCAN1 mutations disrupt the balance of the feedback cycle proteins in the RCAN family is the presence of an SP motif with of CN regulation by promoting phosphorylation of RCAN1 its “signature” amino acid sequence FLISPPxSPP that begins at proteins by GSK-3, ultimately leading to increased CN activa- amino acid 160 of RCAN1 isoform 1. This highly conserved tion and apoptosis (Figure 7).26 region of the protein is known to be phosphorylated by GSK-3 kinases, although the consequences of these modifications on RCAN1 protein function and CN activity may be context de- DISCUSSION pendent and have yet to be fully elucidated. The p.I162T variant disrupts this motif directly and is predicted to alter the struc- In this study, we carried out WGS in a cohort of families with ture of this region to make the GSK-3b site at serine 163 more hereditary FSGS/NS that is not due to pathogenic variants in accessible for modification (Figure 6, A and B). To determine if .60 known FSGS/NS genes, and identified a loss-of-function GSK-3 activity is a component of the pathogenic CN activation, variant in RCAN1 in a family with autosomal dominant FSGS. we examined CN activity of HEK293 cells overexpressing CN Despite the widely known deleterious effects of uncontrolled and RCAN1 constructs in the presence of a dual GSK-3a/b activation of CN in the glomerulus and other compartments inhibitor, LY2090314, and GSK-3b–specific inhibitor, tideglu- of the kidney, this is the first time that mutations in genes sib (Figure 6C). Both of these potent GSK-3 inhibitors were encoding a direct regulator of CN will be implicated in the able to correct the aberrant CN activity of RCAN1 p.I162T, etiology of NS/FSGS.34,35,38 Although RCAN1 is the most suggesting GSK-3b activity likely plays a pivotal role in the plausible candidate based on the genetic data and the function pathogenesis of RCAN1-mediated kidney disease. of RCAN1, the role of the other variants that segregated with On the basis of the combined data, we propose that genetic the FSGS phenotype observed in the leading family still re- variants in RCAN1 can induce glomerular disease due to mains unknown.

JASN 32: ccc–ccc, 2021 RCAN1 Variants Can Cause NS 9 BASIC RESEARCH www.jasn.org

with Alzheimer disease. RCAN2 and RCAN3, located on chro- A RCAN1 WT RCAN1 I162T mosome 6p12.3 and 1p33.11, respectively, are also highly ex- pressed in the developing brain and the heart.13,21,22,61,62 Previous studies have reported that RCAN1, RCAN2,and RCAN3 are expressed in the podocyte and other kidney cellu- lar components; however, their role in human kidney disease is unknown.63–65 In mice with doxorubicin-induced nephrop- athy, a murine model of human FSGS, knockout of RCAN1 increased susceptibility to podocyte injury and albuminuria.63 B RCAN1 WT RCAN1 I162T The regulation of RCAN activity by molecules such as GSK- 3b and its subsequent regulation of CN activation are com- plex, with most studies to date limited to RCAN1. Numerous potential regulators of RCAN1 activity have been identified, including a potential priming phosphorylation by Big MAP kinase1atserine167inRCAN1-1(serine112inisoform RCAN1-4) that precedes phosphorylation of serine 163 by GSK-3b (serine 108 in isoform RCAN1-4).19,57,58 RCAN1 has also been shown to be a potential facilitator of CN activity C when phosphorylated by TGF-b–activated kinase 1 and phos- Calcineurin Assay with GSK3 Inhibition phorylation by NF-kB–inducing kinase can increase RCAN1 100 66,67 p=0.005 stability. Whereas RCAN1 phosphorylation may activate or repress CN activity, depending on the context, phosphor- 50 ylated RCAN1 is also a target of CN (Figure 7).68 Furthermore, 0 RCAN1 expression can be altered by the NFAT transcriptional network, providing an additional feedback regulatory -50 mechanism.69,70 Regulation of CN by RCAN1 can either inhibit or activate -100 CN, depending on the context.20,24–26 The highly conserved Relative Calcineurin Activity SP motif with its signature amino acid sequence FLISPPxSPP -150 is able to inhibit CN activation in vitro, although RCAN1 mu- Control LY2090314 Tideglusib tants truncated after the SP motif display a lower affinity than PPP3CA + RCAN1 WT the full-length RCAN1 protein.12,16,17,19,26 In vivo, however, PPP3CA + RCAN1 |162T the SP motif is not sufficient for inhibition of CN.18,19 In this Figure 6. Inhibition of GSK-3b can rescue the aberrant CN ac- regard, inhibition of CN by RCAN1 requires: (1)theLxxP tivation caused by mutant RCAN1. (A) Three-dimensional mod- motif within the SP domain; and (2) the PxIxIT-like domain, eling of the RCAN1 p.I162T variant revealed alterations to the which is also used for docking of many CN substrates.18 More- positioning of the serine 163 (S163) residue (gold sphere marker) over, stimulation of CN signaling requires both the LxxP and that is a target of GSK-3 kinase. (B) Analysis of the surface Exx(x)P domains within the SP motif, and the highly con- structure of the protein shows that the S163 residue appears to served GSK-3 phosphorylation site within the FLISPPxSPP be more accessible for modification in RCAN1 p.I162T compared 18,26 with WT. (C) HEK293 cells were transfected with constructs sequence. In addition to multiple RCAN family members, containing PPP3CA and either WT RCAN1 or the p.I162T variant, each RCAN gene produces multiple splicing variants. Isoform and the lysates were analyzed for CN activity after 48 hours using RCAN1-4, in particular, has been highly studied in cardiovas- a CN cellular activity assay. The CN activity is increased in un- cular disease and may also contribute to CN regulation in – treated I162T samples (red) compared with WT RCAN1 (P50.005). the kidney.24,71 73 The CN activity was restored to WT levels when the lysates were We have shown that rare variants in RCAN1 can cause FSGS treated with 0.2 mM of the dual GSK-3a/b inhibitor LY2090314 through uncontrolled activation of CN. Due to the similar (P50.67) and 1 mM of the GSK-3b–specific inhibitor tideglusib protein functions between RCAN family members, it is rea- 5 5 (P 0.09, n 6 for all samples, one-way ANOVA). sonable to expect that variants in RCAN2 and RCAN3 are also capable of disrupting CN regulation. With the known roles of The gene encoding RCAN1 is located in chromosome CN activation and NFAT signaling in the regulation of im- 21q22.12, the region classically referred to as the minimal mune responses and podocyte cytoskeletal dynamics, CN candidateregionfortheDownsyndromephenotype. regulatory molecules make attractive therapeutic targets for kid- RCAN1 has been reported to be overexpressed in the brain ney disease. CNIs are widely used in the treatment of FSGS and of babies with Down syndrome during development and it other morphologic forms of NS. However, CNIs are not uni- has been associated with neurofibrillary tangles in patients formly effective, for example, only about 30%–50% of patients

10 JASN JASN 32: ccc–ccc,2021 www.jasn.org BASIC RESEARCH

Activated P CN activity

CN RCAN1 GSK-3

WT podocyte Inhibited CN activity

Activated P CN activity

CN Mutant Apoptosis RCAN1 GSK-3 Inhibited Mutant RCAN1 podocyte CN activity

Figure 7. RCAN1 mutants decrease cell viability by altering the GSK-3b–mediated regulation of CN activity. On the basis of the data acquired in this study, we posit that in WT cells (top), RCAN1 regulates CN activity through a regulatory feedback loop that requires GSK-3b kinase activity. When RCAN1 is phosphorylated (P) by GSK-3b, it either activates CN or dampens RCAN1’s ability to fully inhibit CN, both of which promote cellular apoptosis. The increased phosphatase activity of CN also decreases the levels of phosphorylated RCAN1, which then allows RCAN1 to resume inhibition of CN activity. In cells with mutant RCAN1 (bottom), the structural changes in RCAN1 promote increased phosphorylation by GSK-3b, which shifts the balance of the feedback loop in favor of increased CN activity and, ultimately, apoptosis. with SRNS will achieve partial or full remission after CNI treat- In summary, we identified, for the first time, contributions to ment.74 There are currently no biomarkers to predict response causality from mutations in RCAN1 in families with autosomal to CNI therapy, despite major renal and nonrenal toxicities. dominant FSGS. We showed that the RCAN1 variant, p.I162T, Our in vitro studies suggest that certain functional variants in disrupts the ability of RCAN1 to regulate CN activation, result- RCAN genes may be able to identify patients with SRNS who ing in reduced cell survival that can be rescued by CNIs. In are likely to respond to CNIs. Unfortunately, none of the pa- addition, GSK-3 inhibitors can rescue the increased CN activa- tients in the two index RCAN1 families were treated with CNIs; tion caused by the RCAN1 mutation. Therefore, the use of CNI therefore, we do not have human data to corroborate our cell and GSK antagonists may represent targeted or personalized culture findings. therapy for individuals with NS due to RCAN1 mutations. With limited treatment options available for patients with glomerular disease, the identification of new and repurposed pharmaceutical therapies is critical to increasing therapeutic DISCLOSURES options for these conditions. In this study, we found that GSK- 3 inhibitors could reverse the increased CN activity induced by M. Barua reports having ownership interest in AstraZeneca; serving on the RCAN1 mutations. The GSK-3 inhibitors used in this study, editorial board of Glomerular Diseases; and receiving research funding from tideglusib and LY2093014, are potent and highly selective Otsuka, Regulus, and Sanofi. K. Benson reports serving as chair of the ClinGen small-molecule inhibitors that have both been examined in Kidney Cystic and Ciliopathy Disorders Variant Curation Expert Panel. R. Gbadegesin reports receiving research funding from AstraZeneca, Bristol human phase 2 clinical trials for a variety of diseases. Although Myers Squibb, and Goldfinch Biotech; and having consultancy agreements GSK-3 activity is an attractive therapeutic target, any potential with Keryx Pharmaceutical. F. Hildebrandt reports having consultancy agree- inhibition in patients would need to be carefully titrated due to ments with, ownership interest in, and serving as a scientific advisor for or the importance of GSK-3 activity for maintaining podocyte member of Goldfinch Bio as cofounder; and receiving honoraria from Sanofi. viability and kidney function.75 With numerous molecules S. Murray reports receiving research funding from Amgen. M. Pollak reports having ownership interest in Apolo1Bio; having patents and inventions with implicated in the regulation of RCAN1 activity, additional Athena Diagnostics; serving on the NephCure Foundation scientific advisory therapeutic targets will likely emerge as the RCAN protein board; having consultancy agreements with, and receiving research funding interactome becomes more defined. from, Vertex; and receiving honoraria from various academic talks. Because M.

JASN 32: ccc–ccc, 2021 RCAN1 Variants Can Cause NS 11 BASIC RESEARCH www.jasn.org

Pollak is an editor of the JASN, he was not involved in the peer review process for SUPPLEMENTAL MATERIAL this manuscript. A guest editor oversaw the peer review and decision-making process for this manuscript. M. Saleem reports receiving research funding from This article contains the following supplemental material online at http:// Evotec, Retrophin, and UCB; having consultancy agreements with Mission Ther- jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2020101525/-/ fi apeutics, P zer, and Retrophin; having ownership interest in Purespring Thera- DCSupplemental. peutics; and receiving honoraria from Purespring Therapeutics as director and Supplemental Figure 1. Filtering of rare variants in WGS data from fi fi chief scienti cof cer. M. G. Sampson reports having consultancy agreements family 40030. with Janssen Pharmaceutical, Kohlberg Kravis Roberts & Co.; and serving as a Supplemental Figure 2. Pedigree of second FSGS family with segregating fi scienti c advisor for, or member of, Natera. R. Spurney reports serving as scien- RCAN1 variant. fi ti c advisor for, or member of, the American Journal of Physiology; and having Supplemental Figure 3. RCAN gene expression in cultured human consultancy agreements with Amgen and Tectonic. Additional funding and/or podocytes. programmatic support for NEPTUNE has also been provided by the University Supplemental Figure 4. RCAN1 and CN binding. of Michigan, NephCure Kidney International, and the Halpin Foundation.Ad- Supplemental Figure 5. RCAN1 single transfection calcineurin activity. ditional funding and/or programmatic support for NEPTUNE has also been Supplemental Figure 6. Late apoptosis/necrosis quantification. provided by the University of Michigan, NephCure Kidney International, Supplemental Figure 7. Unmodified Western blots. and the Halpin Foundation. All remaining authors have nothing to disclose. Supplemental Table 1. Segregating heterozygous variants found in family 40030. Supplemental Table 2. Evolutionary conservation of RCAN1 variant FUNDING residues. Supplemental Table 3. Description of patient cohorts. R. Gbadegesin is supported by National Institute of Diabetes and Digestive Supplemental Table 4. Heterozygous missense variants in RCAN1-3 genes. and Kidney Diseases (NIDDK) grants 5R01DK098135 and 5R01DK094987, Supplemental Summary 1. Members of the Nephrotic Syndrome Study Doris Duke Charitable Foundation Clinical Scientist Development Award (NEPTUNE). 2009033, Borden Scholars Award, and the Duke Health Scholars Award. B. Supplemental Video 1. RCAN1 WT protein modeling. M. Lane is supported by NIDDK Duke Nephrology Award, grant T32- Supplemental Video 2. RCAN1 p.I162T protein modeling. DK007731. A. Bierzynska is funded by Kidney Research UK (personal non- Supplemental Video 3. RCAN1 WT apoptosis. clinical fellowship). M. Barua has received Canadian Institutes of Health Supplemental Video 4. RCAN1 p.I162T apoptosis. Research grant 432980, McLaughlin Accelerator Award (2019), NephCure Kid- Supplemental Video 5. RCAN1 p.I162T apoptosis rescue with FK506. ney International–NEPTUNE Ancillary Studies Grant (2016), and Physicians Services Incorporated Health Research Grant 14-04 (2015); and support from the Can-SOLVE CKD Network (https://www.cansolveckd.ca/) and Toronto Gen- eral Hospital Foundation. M.G. Sampson is supported by National Institutes of REFERENCES Health (NIH) grants R01-DK108805 and R01-DK119380. NEPTUNE is a part of the NIH Rare Disease Clinical Research Network, supported through a collabo- 1. Bikbov B, Purcell CA, Levey AS, Smith M, Abdoli A, Abebe M, et al.; fi ration between the Of ce of Rare Diseases Research, National Center for Advanc- GBD Chronic Kidney Disease Collaboration: Global, regional, and na- ‐ ‐ ing Translational Sciences, and NIDDK, under grant U54 DK 083912. tional burden of chronic kidney disease, 1990-2017: A systematic analysis for the Global Burden of Disease Study 2017. Lancet 395: 709–733, 2020 ACKNOWLEDGMENTS 2. USRDS annual data report: Epidemiology of kidney disease in the United States. Available at: https://www.usrds.org/annual-data-report. The authors acknowledge all of the participants in the study. We appreciate the Accessed May 20, 2020 technical support provided by the Duke Molecular Physiology Institute Ge- 3. Barisoni L, Kriz W, Mundel P, D’Agati V: The dysregulated podocyte nomics Core and the personnel of Duke Center for Genomic and Computa- phenotype: A novel concept in the pathogenesis of collapsing idio- tional Biology. The authors acknowledge Goldfinch Bio for their support with pathic focal segmental glomerulosclerosis and HIV-associated ne- WGS of patient samples. The authors acknowledge the Irish Kidney Gene phropathy. JAmSocNephrol10: 51–61, 1999 Project and the Genome England consortium for their collaboration. 4. Wiggins R-C: The spectrum of podocytopathies: A unifying view of The full list of members of NEPTUNE are listed in Supplemental glomerular diseases. Kidney Int 71: 1205–1214, 2007 Summary 1. 5. Lovric S, Fang H, Vega-Warner V, Sadowski CE, Gee HY, Halbritter J, B.M. Lane, R. Spurney, and R. Gbadegesin designed the experiments and et al.; Nephrotic Syndrome Study Group: Rapid detection of mono- wrote the manuscript; B.M. Lane, R. Gbadegesin, G. Wu, L. Wang, M. Chryst- genic causes of childhood-onset steroid-resistant nephrotic syndrome. Stangl, S. Murray, S. Conlon, K. Benson, and R. Spurney performed the ex- Clin J Am Soc Nephrol 9: 1109–1116, 2014 periments. Subject enrollment, sequencing, and analysis of sequencing data 6. Sadowski CE, Lovric S, Ashraf S, Pabst WL, Gee HY, Kohl S, et al.; SRNS were carried out by M. Chryst-Stangl, S. Murray, K. Benson, A. Bierzynska, Study Group: A single-gene cause in 29.5% of cases of steroid-resistant G. Cavalleri, B. Doyle, N. Fennelly, S. Conlon, V. Vega-Warner, D. Fermin, nephrotic syndrome. JAmSocNephrol26: 1279–1289, 2015 P. Vijayan, M. A. Qureshi, S. Shril, M. Barua, F. Hildebrandt, M. Pollak, 7. Yao T, Udwan K, John R, Rana A, Haghighi A, Xu L, et al.: Integration of M.G. Sampson, M. Saleem, P.J. Conlon, and R. Gbadegesin. D. Howell and genetic testing and pathology for the diagnosis of adults with FSGS. A. Dorman evaluated kidney biopsy samples. All of the authors read, edited, Clin J Am Soc Nephrol 14: 213–223, 2019 and approved the manuscript for submission. 8. Vivante A, Hildebrandt F: Exploring the genetic basis of early-onset chronic kidney disease. Nat Rev Nephrol 12: 133–146, 2016 9. Preston R, Stuart HM, Lennon R: Genetic testing in steroid-resistant nephrotic syndrome: Why, who, when and how? Pediatr Nephrol 34: 195–210, 2019 DATA SHARING STATEMENT 10. Trautmann A, Bodria M, Ozaltin F, Gheisari A, Melk A, Azocar M, et al.; PodoNet Consortium: Spectrum of steroid-resistant and congenital Material requests and all correspondence should be sent to nephrotic syndrome in children: The PodoNet registry cohort. Clin [email protected]. JAmSocNephrol10: 592–600, 2015

12 JASN JASN 32: ccc–ccc,2021 www.jasn.org BASIC RESEARCH

11. Warejko JK, Tan W, Daga A, Schapiro D, Lawson JA, Shril S, et al.: 31. Bensman A, Niaudet P: Non-immunologic mechanisms of calcineurin Whole exome sequencing of patients with steroid-resistant nephrotic inhibitors explain its antiproteinuric effects in genetic glomer- syndrome. Clin J Am Soc Nephrol 13: 53–62, 2018 ulopathies. Pediatr Nephrol 25: 1197–1199, 2010 12. Rothermel B, Vega RB, Yang J, Wu H, Bassel-Duby R, Williams RS: A 32. Yoo T-H, Fornoni A: Nonimmunologic targets of immunosuppressive protein encoded within the Down syndrome critical region is enriched agents in podocytes. Kidney Res Clin Pract 34: 69–75, 2015 in striated muscles and inhibits calcineurin signaling. JBiolChem275: 33. Faul C, Donnelly M, Merscher-Gomez S, Chang YH, Franz S, Delfgaauw 8719–8725, 2000 J, et al.: The actin cytoskeleton of kidney podocytes is a direct target of 13. Fuentes JJ, Genescà L, Kingsbury TJ, Cunningham KW, Pérez-Riba M, the antiproteinuric effect of cyclosporine A. Nat Med 14: 931–938, Estivill X, et al.: DSCR1, overexpressed in Down syndrome, is an in- 2008 hibitor of calcineurin-mediated signaling pathways. Hum Mol Genet 9: 34. Wang Y, Jarad G, Tripathi P, Pan M, Cunningham J, Martin DR, et al.: 1681–1690, 2000 Activation of NFAT signaling in podocytes causes glomerulosclerosis. 14. Mulero MC, Aubareda A, Schlüter A, Pérez-Riba M: RCAN3, a novel JAmSocNephrol21: 1657–1666, 2010 calcineurin inhibitor that down-regulates NFAT-dependent cytokine 35. Gooch JL, Barnes JL, Garcia S, Abboud HE: Calcineurin is activated in gene expression. Biochim Biophys Acta 1773: 330–341, 2007 diabetes and is required for glomerular hypertrophy and ECM accu- 15. Cao X, Kambe F, Miyazaki T, Sarkar D, Ohmori S, Seo H: Novel human mulation. Am J Physiol Renal Physiol 284: F144–F154, 2003 ZAKI-4 isoforms: Hormonal and tissue-specific regulation and function 36. D’Agati VD, Kaskel FJ, Falk RJ: Focal segmental glomerulosclerosis. as calcineurin inhibitors. Biochem J 367: 459–466, 2002 NEnglJMed365: 2398–2411, 2011 16. Görlach J, Fox DS, Cutler NS, Cox GM, Perfect JR, Heitman J: Identi- 37. Pedigo CE, Ducasa GM, Leclercq F, Sloan A, Mitrofanova A, Hashmi T, fication and characterization of a highly conserved calcineurin binding et al.: Local TNF causes NFATc1-dependent cholesterol-mediated protein, CBP1/calcipressin, in Cryptococcus neoformans. EMBO J 19: podocyte injury. JClinInvest126: 3336–3350, 2016 3618–3629, 2000 38. Wang L, Chang J-H, Paik S-Y, Tang Y, Eisner W, Spurney RF: Calcineurin 17. Kingsbury TJ, Cunningham KW: A conserved family of calcineurin (CN) activation promotes apoptosis of glomerular podocytes both regulators. Genes Dev 14: 1595–1604, 2000 in vitro and in vivo. Mol Endocrinol 25: 1376–1386, 2011 18. Mehta S, Li H, Hogan PG, Cunningham KW: Domain architecture of the 39. Chen S, Zhou Y, Chen Y, Gu J: fastp: an ultra-fast all-in-one FASTQ regulators of calcineurin (RCANs) and identification of a divergent preprocessor. Bioinformatics 34: i884–i890, 2018 RCAN in yeast. Mol Cell Biol 29: 2777–2793, 2009 40. Li H, Durbin R: Fast and accurate short read alignment with Burrows- 19. Vega RB, Yang J, Rothermel BA, Bassel-Duby R, Williams RS: Multiple Wheeler transform. Bioinformatics 25: 1754–1760, 2009 domains of MCIP1 contribute to inhibition of calcineurin activity. JBiol 41. PicardToolkit 2019. Broad Institute, GitHub Repository. Available at: Chem 277: 30401–30407, 2002 http://broadinstitute.github.io/picard. Accessed February 17, 2020 20. Sanna B, Brandt EB, Kaiser RA, Pfluger P, Witt SA, Kimball TR, et al.: 42. DePristo MA, Banks E, Poplin R, Garimella KV, Maguire JR, Hartl C, Modulatory calcineurin-interacting proteins 1 and 2 function as calci- et al.: A framework for variation discovery and genotyping using next- neurin facilitators in vivo. Proc Natl Acad Sci U S A 103: 7327–7332, generation DNA sequencing data. Nat Genet 43: 491–498, 2011 2006 43. Van der Auwera GA, Carneiro MO, Hartl C, Poplin R, Del Angel G, Levy- 21. Strippoli P, Lenzi L, Petrini M, Carinci P, Zannotti M: A new gene family Moonshine A, et al.: From FastQ data to high confidence variant calls: including DSCR1 (Down Syndrome Candidate Region 1) and ZAKI-4: The genome analysis toolkit best practices pipeline. Curr Protoc Bio- Characterization from yeast to human and identification of DSCR1-like informatics 43: 11.10.1–11.10.33, 2013 2, a novel human member (DSCR1L2). Genomics 64: 252–263, 2000 44. McLaren W, Gil L, Hunt SE, Riat HS, Ritchie GRS, Thormann A, et al.: The 22. Miyazaki T, Kanou Y, Murata Y, Ohmori S, Niwa T, Maeda K, et al.: Ensembl variant effect predictor. Genome Biol 17: 122, 2016 Molecular cloning of a novel thyroid hormone-responsive gene, ZAKI- 45. Karczewski KJ, Francioli LC, Tiao G, Cummings BB, Alföldi J, Wang Q, 4, in human skin fibroblasts. JBiolChem271: 14567–14571, 1996 et al.: The mutational constraint spectrum quantified from variation in 23. Chang KT, Shi Y-J, Min K-T: The Drosophila homolog of Down’s syn- 141,456 humans [published correction appears in Nature 590: E53, drome critical region 1 gene regulates learning: Implications for mental 2021 10.1038/s41586-020-03174-8]. Nature 581: 434–443, 2020 retardation. Proc Natl Acad Sci U S A 100: 15794–15799, 2003 46. Hall G, Lane BM, Khan K, Pediaditakis I, Xiao J, Wu G, et al.: The human 24. Vega RB, Rothermel BA, Weinheimer CJ, Kovacs A, Naseem RH, FSGS-causing ANLN R431C mutation induces dysregulated PI3K/AKT/ Bassel-Duby R, et al.: Dual roles of modulatory calcineurin-interacting mTOR/Rac1 signaling in podocytes. JAmSocNephrol29: 2110–2122, protein 1 in cardiac hypertrophy. Proc Natl Acad Sci U S A 100: 2018 669–674, 2003 47. Spurney RF: Role of C-terminal serines in desensitization and phos- 25. Ryeom S, Greenwald RJ, Sharpe AH, McKeon F: The threshold pattern phorylation of the mouse thromboxane receptor. JBiolChem273: of calcineurin-dependent gene expression is altered by loss of the 28496–28503, 1998 endogenous inhibitor calcipressin. Nat Immunol 4: 874–881, 2003 48. Zhang Y: I-TASSER: Fully automated protein structure prediction in 26. Hilioti Z, Gallagher DA, Low-Nam ST, Ramaswamy P, Gajer P, CASP8. Proteins 77[Suppl 9]: 100–113, 2009 Kingsbury TJ, et al.: GSK-3 kinases enhance calcineurin signaling by 49. Yang J, Zhang Y: Protein structure and function prediction using phosphorylation of RCNs. Genes Dev 18: 35–47, 2004 I-TASSER. Curr Protoc Bioinformatics 52: 5.8.1–5.8.15, 2015 27. Martínez-Martínez S, Genescà L, Rodríguez A, Raya A, Salichs E, Were 50. Goddard TD, Huang CC, Meng EC, Pettersen EF, Couch GS, Morris JH, F, et al.: The RCAN carboxyl end mediates calcineurin docking- et al.: UCSF ChimeraX: Meeting modern challenges in visualization and dependent inhibition via a site that dictates binding to substrates and analysis. Protein Sci 27: 14–25, 2018 regulators. Proc Natl Acad Sci U S A 106: 6117–6122, 2009 51. Wang L, Chang J-H, Buckley AF, Spurney RF: Knockout of TRPC6 28. Spurney RF: Non-immunologic actions of calcineurin inhibitors in pro- promotes insulin resistance and exacerbates glomerular injury in Akita teinuric kidney diseases. Front Endocrinol (Lausanne) 5: 181, 2014 mice. Kidney Int 95: 321–332, 2019 29. Schönenberger E, Ehrich JH, Haller H, Schiffer M: The podocyte as a 52. Hall G, Lane B, Chryst-Ladd M, Wu G, Lin J-J, Qin X, et al.: Dysregu- direct target of immunosuppressive agents. Nephrol Dial Transplant lation of WTI (-KTS) is associated with the kidney-specific effects of the 26: 18–24, 2011 LMX1B R246Q mutation. Sci Rep 7: 39933, 2017 30. Hall G, Wang L, Spurney RF: TRPC channels in proteinuric kidney dis- 53. Das AK, Pickett TM, Tungekar MF: Glomerular basement membrane eases. Cells 9: 44, 2019 thickness—a comparison of two methods of measurement in patients

JASN 32: ccc–ccc, 2021 RCAN1 Variants Can Cause NS 13 BASIC RESEARCH www.jasn.org

with unexplained haematuria. Nephrol Dial Transplant 11: 1256–1260, 65. Grgic I, Hofmeister AF, Genovese G, Bernhardy AJ, Sun H, Maarouf 1996 OH, et al.: Discovery of new glomerular disease-relevant genes by 54. Connaughton DM, Bukhari S, Conlon P, Cassidy E, O’Toole M, Mohamad translational profiling of podocytes in vivo. Kidney Int 86: 1116–1129, M, et al.: The Irish kidney gene project--prevalence of family history in 2014 patients with kidney disease in Ireland. Nephron 130: 293–301, 2015 66. Liu Q, Busby JC, Molkentin JD: Interaction between TAK1-TAB1-TAB2 55. Connaughton DM, Kennedy C, Shril S, Mann N, Murray SL, Williams PA, and RCAN1-calcineurin defines a signalling nodal control point. Nat et al.: Monogenic causes of chronic kidney disease in adults. Kidney Int Cell Biol 11: 154–161, 2009 95: 914–928, 2019 67. Lee EJ, Seo SR, Um JW, Park J, Oh Y, Chung KC: NF-kappaB-inducing 56. Roy A, Yang J, Zhang Y: COFACTOR: An accurate comparative algo- kinase phosphorylates and blocks the degradation of Down syndrome rithm for structure-based protein function annotation. Nucleic Acids candidate region 1. JBiolChem283: 3392–3400, 2008 Res 40: W471–W477, 2012 68. Porta S, Serra SA, Huch M, Valverde MA, Llorens F, Estivill X, et al.: 57. Abbasi S, Su B, Kellems RE, Yang J, Xia Y: The essential role of MEKK3 RCAN1 (DSCR1) increases neuronal susceptibility to oxidative stress: A signaling in angiotensin II-induced calcineurin/nuclear factor of acti- potential pathogenic process in neurodegeneration. Hum Mol Genet vated T-cells activation. JBiolChem280: 36737–36746, 2005 16: 1039–1050, 2007 58. Abbasi S, Lee J-D, Su B, Chen X, Alcon JL, Yang J, et al.: Protein kinase- 69. Minami T, Horiuchi K, Miura M, Abid MR, Takabe W, Noguchi N, mediated regulation of calcineurin through the phosphorylation of et al.: Vascular endothelial growth factor- and thrombin-induced modulatory calcineurin-interacting protein 1. J Biol Chem 281: termination factor, Down syndrome critical region-1, attenuates 7717–7726, 2006 endothelial cell proliferation and angiogenesis. JBiolChem279: 59. Turro E, Astle WJ, Megy K, Gräf S, Greene D, Shamardina O, et al.; 50537–50554, 2004 NIHR BioResource for the 100,000 Genomes Project: Whole-genome 70. Stathatos N, Bourdeau I, Espinosa AV, Saji M, Vasko VV, Burman KD, sequencing of patients with rare diseases in a national health system. et al.: KiSS-1/G protein-coupled receptor 54 metastasis suppressor Nature 583: 96–102, 2020 pathway increases myocyte-enriched calcineurin interacting protein 1 60. Bierzynska A, McCarthy HJ, Soderquest K, Sen ES, Colby E, Ding WY, expression and chronically inhibits calcineurin activity. JClinEndo- et al.: Genomic and clinical profiling of a national nephrotic syndrome crinol Metab 90: 5432–5440, 2005 cohort advocates a precision medicine approach to disease manage- 71. Cheng L, Li P, Wang H, Yang X, Zhou H, Tao W, et al.: Decreased activity ment. Kidney Int 91: 937–947, 2017 of RCAN1.4 is a potential risk factor for congenital heart disease in a 61. Porta S, Martí E, de la Luna S, Arbonés ML: Differential expression of Han Chinese population. Protein Cell 9: 1039–1044, 2018 members of the RCAN family of calcineurin regulators suggests se- 72. Qin L, Zhao D, Liu X, Nagy JA, Hoang MV, Brown LF, et al.: Down lective functions for these proteins in the brain. Eur J Neurosci 26: syndrome candidate region 1 isoform 1 mediates angiogenesis 1213–1226, 2007 through the calcineurin-NFAT pathway. Mol Cancer Res 4: 811–820, 62. Canaider S, Facchin F, Griffoni C, Casadei R, Vitale L, Lenzi L, et al.: 2006 Proteins encoded by human Down syndrome critical region gene 1-like 73. Lange AW, Yutzey KE: NFATc1 expression in the developing heart 2 (DSCR1L2) mRNA and by a novel DSCR1L2 mRNA isoform interact valves is responsive to the RANKL pathway and is required for endo- with cardiac troponin I (TNNI3). Gene 372: 128–136, 2006 cardial expression of cathepsin K. Dev Biol 292: 407–417, 2006 63. Li H, Zhang W, Zhong F, Das GC, Xie Y, Li Z, et al.: Epigenetic regulation 74. Gipson DS, Trachtman H, Kaskel FJ, Greene TH, Radeva MK, Gassman of RCAN1 expression in kidney disease and its role in podocyte injury. JJ, et al.: Clinical trial of focal segmental glomerulosclerosis in children Kidney Int 94: 1160–1176, 2018 and young adults. Kidney Int 80: 868–878, 2011 64. Jang C, Lim JH, Park CW, Cho Y-J: Regulator of calcineurin 1 isoform 4 75. Hurcombe JA, Hartley P, Lay AC, Ni L, Bedford JJ, Leader JP, et al.: (RCAN1.4) is overexpressed in the glomeruli of diabetic mice. Korean Podocyte GSK3 is an evolutionarily conserved critical regulator of J Physiol Pharmacol 15: 299–305, 2011 kidney function. Nat Commun 10: 403, 2019

AFFILIATIONS

1Division of Nephrology, Department of Pediatrics, Duke Molecular Physiology Institute, Duke University School of Medicine, Durham, North Carolina 2Irish Kidney Gene Project, Department of Genetics, Royal College of Surgeons of Ireland, Dublin, Republic of Ireland 3School of Pharmacy and Biomolecular Sciences, Royal College of Surgeons of Ireland, Dublin, Republic of Ireland 4Department of Pediatrics, Bristol Royal Hospital for Children and University of Bristol, Bristol, United Kingdom 5Division of Nephrology, Department of Medicine, Duke University School of Medicine, Durham, North Carolina 6Department of Pathology, Beaumont General Hospital, Dublin, Republic of Ireland 7Department of Pediatrics, University of Michigan, Ann Arbor, Michigan 8Division of Nephrology, Department of Medicine, University of Toronto and Toronto General Hospital, Toronto, Ontario, Canada 9Division of Nephrology, Department of Pediatrics, Boston Children’s Hospital and Harvard University Medical School, Boston, Massachusetts 10Division of Nephrology, Department of Medicine, Beth Israel Hospital and Harvard University Medical School, Boston, Massachusetts 11Department of Pathology, Duke University School of Medicine, Durham, North Carolina 12Broad Institute of Harvard and Massachusetts Institute of Technology, Cambridge, Massachusetts 13Division of Nephrology, Department of Medicine, Beaumont General Hospital, Dublin, Republic of Ireland

14 JASN JASN 32: ccc–ccc,2021 SUPPLEMENTARY MATERIALS

A rare autosomal dominant variant in regulator of calcineurin type1 (RCAN1) gene confers enhanced calcineurin activity and may cause FSGS

Brandon M Lane1, Susan Murray2, Katherine Benson3, Agnieszka Bierzynska4, Megan Chryst-Stangl1, Liming Wang13, Guanghong Wu1, Gianpiero Cavalleri3, Brendan Doyle5, Neil Fennelly5, Anthony Dorman5, Shane Conlon2, Virginia Vega-Warner6, Damian Fermin6, Poornima Vijayan7, Mohammad Azfar Qureshi7, Shirlee Shril8, Moumita Barua7, Friedhelm Hildebrandt8, Martin Pollak9, David Howell10, Matthew G. Sampson8,11, Moin Saleem4, Peter J Conlon2, 12, Robert Spurney13, Rasheed Gbadegesin1,13 1 Department of Pediatrics, Division of Nephrology and Duke Molecular Physiology Institute, Duke University School of Medicine, Durham, NC, USA 2 Irish Kidney Gene Project, Department of genetics, Royal College of Surgeons of Ireland, Republic of Ireland 3 School of Pharmacy and Biomolecular Sciences, Royal College of Surgeons of Ireland, Dublin, Republic of Ireland 4 Department of Pediatrics, Bristol Royal Hospital for Children and University of Bristol, Bristol BS2 8BJ, UK 5 Department of Pathology, Beaumont General Hospital, Dublin, Republic of Ireland 6 Department of Pediatrics, University of Michigan, Ann Arbor, MI, USA 7 Department of Medicine, Division of Nephrology, University of Toronto and Toronto General Hospital, Toronto, ON, Canada

8 Department of Pediatrics, Division of Nephrology, Boston Children’s Hospital and Harvard University Medical School, Boston MA, USA 9 Department of Medicine, Division of Nephrology, Beth Israel Hospital and Harvard University Medical School, Boston, MA, USA

10 Department of Pathology, Duke University School of Medicine, Durham, NC, USA 11 Broad Institute of Harvard and MIT, Cambridge, MA, USA 12 Department of Medicine, Division of Nephrology, Beaumont General Hospital, Dublin, Republic of Ireland 13 Department of Medicine, Division of Nephrology, Duke University School of Medicine, Durham, NC, USA Supplemental Material Table of Contents

Supplementary Figure 1: Filtering of rare variants in WGS data from Family 40030 Supplementary Figure 2: Pedigree of second FSGS family with segregating RCAN1 variant Supplementary Figure 3: RCAN gene expression in cultured Human podocytes Supplementary Figure 4: RCAN1 and CN binding Supplementary Figure 5: RCAN1 single transfection calcineurin activity. Supplementary Figure 6: Late Apoptosis/Necrosis quantification Supplementary Figure 7: Unmodified western blots Supplementary Table 1: Segregating heterozygous variants found in Family 40030 Supplementary Table 2: Evolutionary conservation of RCAN1 variant residues Supplementary Table 3: Description of patient cohorts Supplementary Table 4: Heterozygous missense variants in RCAN1-3 genes Supplementary Table 5: Members of the Nephrotic Syndrome Study Network (NEPTUNE) Supplementary Movie 1: RCAN1 WT protein modeling Supplementary Movie 2: RCAN1 p.I162T protein modeling Supplementary Movie 3: RCAN1 WT apoptosis Supplementary Movie 4: RCAN1 p.I162T apoptosis Supplementary Movie 5: RCAN1 p.I162T apoptosis rescue with FK506

Supplementary Figure 1: Filtering of rare variants in WGS data from Family 40030

Supplementary Figure 2: Pedigree of second FSGS family with segregating RCAN1 variant

Supplementary Figure 3: RCAN gene expression in cultured Human podocytes Quantitative Real-Time PCR analysis of three independent samples of differentiated conditionally immortalized human podocytes revealed that RCAN1, RCAN2, and RCAN3 are expressed in the podocyte. The mean Ct values for each gene are listed to provide a relative comparison of RCAN genes expression to known podocyte genes CD2AP and PTPRO (GLEPP1). These values suggest that RCAN1-3 are expressed at a similar level to CD2AP and PTPRO. This analysis was repeated in triplicate with multiple wells per sample for each replicate.

Supplementary Figure 4: RCAN1 and CN binding To determine if the RCAN1 mutations affected docking of RCAN1 and calcineurin, we transfected human embryonic kidney cells (HEK293) cells with a flag-tagged calcineurin construct and myc-tagged RCAN1 constructs as indicated. Immunoprecipitation studies were then performed as described in the Methods Section. (a) Immunoprecipitation of the calcineurin construct co-immunoprecipitated equal amounts of wild type and mutant RCAN1 proteins. (b) Conversely, immunoprecipitation of either wild type or mutant RCAN1 constructs co-immunoprecipitated equal amounts of the calcineurin protein.

Supplementary Figure 5: RCAN1 single transfection calcineurin activity. To ensure that the calcineurin activity and NFAT luciferase activity assays were working appropriately, we examined the activity of unmodified HEK293 cells as well as the single transfections of RCAN1 WT or PPP3CA. (a&b) Both the calcineurin activity (a) and NFAT expression (b) decreased in RCAN1 WT expressing cells and increased in PPP3CA expressing cells compared to unmodified cells, demonstrating the efficacy of the assays.

Supplementary Figure 6: Late Apoptosis/Necrosis In addition to apoptosis readings using cleaved caspase 3 reporters, live cell automated imaging of propidium iodide staining was simultaneously used to measure late apoptosis/necrosis levels in HEK293 cells transfected with PPP3CA and either WT RCAN1 (black circle) or I162T (red triangle) mutants. The I162T expressing cells displayed increased late apoptotic/necrotic cells compared to WT RCAN1 expressing cells (p=0.0152 at 24 hours respectively, two-way ANOVA). This increased cell death was eliminated when the cells were treated with 1uM of FK506 (blue triangle) (p>0.4 for all time points in both mutants).

Supplementary Figure 7: Unmodified western blots for RCAN1 (a), β-actin (b), Caspase 3 (c), and β-actin (d) in conditionally immortalized podocytes. Unmodified western blots for Caspase-3 (e), Myc-tag (f), and β-actin (g) in 293 cells.

Supplementary Table 1: Segregating variants found in Family 40030

Gene HGVSc HGVSp gnomAD MAF CADD Polyphen SIFT Mut (Eur) Taster RCAN1 c.485T>C p.I162T 2/128738 0.00001 26.7 Probably Damaging Disease Damaging causing MAST3 c.1160G>A p.R387H 49/122646 0.0003 26.0 Possibly Damaging Disease Damaging causing POLD1 c.653G>A p.R218H 41/124252 0.0003 32.0 Possibly Damaging Disease Damaging causing CBR3 c.605 C>A p.T202K 28/129144 0.0002 34.0 Probably Damaging Disease Damaging causing

GRCH37: RCAN1-001: ENST0000031380

Supplementary Table 2: Evolutionary conservation of RCAN1 variant residues

Supplementary Table 3: Description of cohorts

Cohort Individuals/ Race* Histology* Therapy* response Family White/Black/ FSGS/MCD/ SRNS/SSNS/ Asian/Others Others /Unknown Unknown Duke 547/392 182/66/119/25 118/55/29/190 110/195/87

Boston Children’s 138/114 114/0/0/0 7/0/107/0 7/0/107 Hospital NEPTUNE cohort 627/627 332/149/67/79 183/168/129/147 NA++

Beth Israel Hospital, 524/337 216/31/21/69 337/0/0 337/0/0 Boston

Toronto General 193/193 NA NA 42/45/106 Hospital , Canada

UK NephroS cohort NA NA NA NA

* Findings in proband NA: Not available ++: Partial or complete remission ever (Yes/No/Unknown) 451/98/78

Supplementary Table 4: Heterozygous missense variants in RCAN1-3 genes in patients with nephrotic syndrome

Gene rsID Exon Nucleotide Protein

RCAN1 1 c.A107C p.L36R RCAN1 1 c.A109C p.S37A RCAN1 rs749675544 1 c.C115T p. A39T RCAN1 rs377673728 2 c.C368T p.R123K RCAN1 rs140515920 2 c.A382G p.K128E RCAN1 rs146806035 3 c.C448T p.H150Y RCAN1 rs145120179 3 c.C458T p.P153L RCAN1 rs1178734954 3 c.T485C p.I162T RCAN2 4 c.C445A p.P149T RCAN2 5 c.G596C p.G199A RCAN2 rs201948840 5 c.C700T p.R234H RCAN2 rs138769310 5 c.G721T p.V241L RCAN2 5 c.A728C p.N243H RCAN3 rs114568126 4 c.C391T p.R131W RCAN3 4 c.A482T p.E161V RCAN3 4 c.T517G p.C173G RCAN3 rs779074675 5 c.G566C p.G189A RCAN3 5 c.A605C p.E202A RCAN3 rs201998230 5 c.G668T p.R223L RCAN3 rs201388228 5 c.G673A p.D225N

GRCH37: RCAN1-001: ENST00000313806.4, RCAN2-002: ENST00000371374.1, RCAN3-001: ENST00000374395.4

Supplementary Table 5: Members of the Nephrotic Syndrome Study Network (NEPTUNE) NEPTUNE Enrolling Centers Cleveland Clinic, Cleveland, OH: K Dell*, J Sedor**, M Schachere#, J Negrey# Children’s Hospital, Los Angeles, CA: K Lemley*, E Lim# Children’s Mercy Hospital, Kansas City, MO: T Srivastava*, A Garrett# Cohen Children’s Hospital, New Hyde Park, NY: C Sethna*, K Laurent # Columbia University, New York, NY: G Appel*, A Pradhan# Emory University, Atlanta, GA: L Greenbaum*, C Wang**, C Kang# Harbor-University of California Los Angeles Medical Center: S Adler*, J LaPage# John H. Stroger Jr. Hospital of Cook County, Chicago, IL: A Athavale*, M Itteera Johns Hopkins Medicine, Baltimore, MD: M Atkinson*, S Boynton# Mayo Clinic, Rochester, MN: F Fervenza*, M Hogan**, J Lieske*, V Chernitskiy# Montefiore Medical Center, Bronx, NY: F Kaskel*, M Ross*, P Flynn# NIDDK Intramural, Bethesda MD: J Kopp*, J Blake# New York University Medical Center, New York, NY: H Trachtman*, O Zhdanova**, F Modersitzki#, S Vento# Stanford University, Stanford, CA: R Lafayette*, K Mehta# Temple University, Philadelphia, PA: C Gadegbeku*, S Quinn-Boyle# University Health Network Toronto: M Hladunewich**, H Reich**, P Ling#, M Romano# University of Miami, Miami, FL: A Fornoni*, C Bidot# University of Michigan, Ann Arbor, MI: M Kretzler*, D Gipson*, A Williams#, J LaVigne# University of North Carolina, Chapel Hill, NC: V Derebail*, K Gibson*, E Cole#, J Ormond-Foster# University of Pennsylvania, Philadelphia, PA: L Holzman*, K Meyers**, K Kallem#, A Swenson# University of Texas Southwestern, Dallas, TX: K Sambandam*, Z Wang#, M Rogers# University of Washington, Seattle, WA: A Jefferson*, S Hingorani**, K Tuttle**§, M Bray #, M Kelton#, A Cooper#§ Wake Forest University Baptist Health, Winston-Salem, NC: JJ Lin*, Stefanie Baker#

Data Analysis and Coordinating Center: M Kretzler, L Barisoni, J Bixler, H Desmond, S Eddy, C Gadegbeku, B Gillespie, D Gipson, L Holzman, V Kurtz, M Larkina, J Lavigne, S Li, CC Lienczewski, J Liu, T Mainieri, L Mariani, M Sampson, M Wladkowski, A Williams, J Zee

Digital Pathology Committee: Carmen Avila-Casado (UHN-Toronto), Serena Bagnasco (Johns Hopkins), Joseph Gaut (Washington U), Stephen Hewitt (National Cancer Institute), Jeff Hodgin (University of Michigan), Kevin Lemley (Children’s Hospital LA), Laura Mariani (University of Michigan), Matthew Palmer (U Pennsylvania), Avi Rosenberg (NIDDK), Virginie Royal (Montreal), David Thomas (University of Miami), Jarcy Zee (Arbor Research) Co-Chairs: Laura Barisoni (Duke University) and Cynthia Nast (Cedar Sinai)

*Principal Investigator; **Co-investigator; #Study Coordinator §Providence Medical Research Center, Spokane, WA

Supplementary Protein Modeling (Movies 1-2): Movie files showing the predicted 3D structures of wildtype RCAN1 (Supplementary Movie 1), and RCAN1 I162T (Supplementary Movie 2).

Supplementary Apoptosis (Movies 2-5): Movie files of HEK 293 cells transfected with PPP3CA and either RCAN1 WT (Supplementary Movie 3), and RCAN1 I162T (Supplementary Movie 4) undergoing serum starvation (4hr-24hr of starvation). Cell death can be observed and quantified with green fluorescence indicating cleaved caspase 3 activity (apoptosis) which is then followed by red propidium idiode staining (late apoptosis/necrosis). Reduced apoptosis can be observed in RCAN1 I162T (Supplementary Movie 5) cells treated with 1µM FK506.