NIPBL+/− haploinsufficiency reveals a constellation of transcriptome disruptions in the pluripotent and cardiac states

Cornelia de Lange syndrome (CdLS) is a complex disorder with multiple structural and developmental defects caused by mutations in structural and regulatory proteins involved in the cohesin complex. NIPBL, a cohesin regulatory protein, has been identified as a critical protein responsible for the orchestration of transcriptomic regulatory networks necessary for embryonic development. Mutations in NIPBL are responsible for the majority of cases of CdLS. Through RNA-sequencing of human induced pluripotent stem cells and in vitro-derived cardiomyocytes, we identified hundreds of mRNAs, pseudogenes, and non-coding RNAs with altered expression in NIPBL+/− patient-derived cells. We demonstrate that NIPBL haploinsufficiency leads to upregulation of gene sets identified in functions related to nucleosome, chromatin assembly, RNA modification and downregulation of Wnt signaling, cholesterol biosynthesis and vesicular transport in iPSC and cardiomyocytes. Mutations in NIPBL result in the dysregulation of many genes responsible for normal heart development likely resulting in the variety of structural cardiac defects observed in the CdLS population.


Single Polymorphism (SNP) Microarray Analysis
DNA was isolated from each fibroblast and iPSC line using DNeasy Blood and Tissue Kit (Qiagen, Inc), and genotyped on the Illumina BeadChip at the Genomics Diagnostic laboratory, Children's Hospital of Philadelphia. The NIPBL1, NIPBL2, NIPBL3, and NIPBL4 fibroblast and iPSC lines were genotyped on HumanOmni1-Quad arrays with 1,140,419 markers. CNV analysis was performed using the PennCNV software5 based on the signal intensity (Log R Ratio and B Allele Frequency) data exported from Illumina GenomeStudio. The default parameters were used, and only large CNVs (>200kb) were retained for subsequent analysis. The signal intensity data around all CNV calls were plotted and visually inspected to ensure accuracy. We used the scan_region.pl program within PennCNV to determine if a CNV call is unique to iPSC lines, in comparison to the parental fibroblast as a reference.

Immunofluorescence
Glass coverslips are cleaned, autoclaved, and placed in the bottom of a 12-well tissue culture dish. Coverslips are then covered with 1:100 matrigel and placed in 37 o C incubator overnight. The next day media is removed and dishes with matrigel + coverslips are placed at RT for 1h. Cells were plated onto coverslip dishes by dissociating to single cells with Collagenase II (Worthington Biochemical Corporation), centrifuge cell suspension plus RPMI (Invitrogen) for 3 minutes. Resuspend pellet in cardiac media with rock inhibitor and plate on 12-well dish with coverslips. Maintain cells for 2-3 days or until a confluency of 30-40% has been reached and then prepare for fixation. Wash cells once with 1X dPBS. Fix cells by adding 4% paraformaldehyde in dPBS to each well, and let sit for 15 minutes at room temperature. Wash twice with cold dPBS for 5 minutes each and begin immunostaining of cardiomyocyte samples with cardiac specific antibody markers.

RNA isolation, library preparation and sequencing
Total RNA purification was carried out on iPSC and CMs as directed in PureLink TM RNA Micro Kit protocol (Invitrogen) using QIAzol Lysis Reagent (Qiagen) including the optional on-column DNAse step. All samples were submitted for bioanalyzer analysis for quality control. Only samples receiving a RNA integrity number (RIN) score ≥9.4 were provided further RNA sequencing analysis. RNA-seq libraries were prepared with Illumina TruSeq RiboZero Sample Prep kits according to the manufacturer's protocol and sequenced on a HiSeq 2500 (Illumina), using the paired end protocol with 100-bp read length to obtain three technical sequencing replicates per sample, each individual was prepared in triplicate.

First Strand cDNA Synthesis
RNA was isolated from iPSC or CMs and first strand synthesis was performed using TaqMan® Reverse Transcription Reagents. Protocol was followed as instructed by manufacturer, with each sample yielding a 20µL reaction from 1µg of RNA by using random hexamers as the primer type. cDNA was stored at -20C until needed for downstream analysis.

qRT-PCR analysis
Gene expression was quantified using the QX100 TM Droplet Digital TM PCR system.
Expression levels from all samples were measured using primer/probes to NIPBL, and normalized to TBP (ABI: 4325803). The ddPCR assay was performed as previously published 5 . Briefly, a 25uL reaction contained 20uL of ddPCR mastermix, 2.5uL cDNA (25ng cDNA) and 1.25uL of Target   Error bars are representative of ±SEM.