Corrigendum: Genetic correction of Werner syndrome gene reveals impaired pro‐angiogenic function and HGF insufficiency in mesenchymal stem cells

We acquired the WS fibroblasts from Coriell Institute. We derived WS iPSC (iWS780 clone 2.10) from WS fibroblast AG00780. This patient carries a 1336C>T mutation in exon-9 of WRN, causing premature translational termination (p.R369*) of WRN protein. We designed a guide RNA sequence (gatgtacttggaaataaaag) near the mutation and cloned into the CRISPR/Cas9 plasmid PX458 (Addgene#48138). We constructed the targeting vector by cloning the left homology arm (including the entire exon-9) and right homology arm (intron-9) into an HR vector HR100PA (SBI). We changed the targeting sequence on the left homology arm by synonymous substitution to avoid secondary gene editing by the gRNA. We transfected the CRISPR/Cas9 plasmid and the linearized targeting vector into WS iPSC by nucleofection. Puromycin-resistant colonies were expanded on MEF. We examined successfully corrected clones by detecting expression of WRN protein by Western blot. We confirmed correct homologous recombination by sequencing the genomic DNA region covering the entire exon-9 and part of intron-9. We confirmed correct mRNA splicing by sequencing the cDNA of the entire WRN open reading frame. To generate an ESC model of WS, we transfected human H1 ESC line with the same gRNA shown above by nucleofection. A nonspecific control gRNA was used to derive the control line. After colony formation from single cells (GFP-sorted), we screened individual clones by Western blot using anti-WRN antibody. WRN-null clones were validated by genomic sequencing across exon-9. 4.2 | Cell culture and differentiation


| Gene correction of WS iPSC
We acquired the WS fibroblasts from Coriell Institute. We derived WS iPSC (iWS780 clone 2.10) from WS fibroblast AG00780. This patient carries a 1336C>T mutation in exon-9 of WRN, causing premature translational termination (p.R369*) of WRN protein. We designed a guide RNA sequence (gatgtacttggaaataaaag) near the mutation and cloned into the CRISPR/Cas9 plasmid PX458 (Addgene#48138). We constructed the targeting vector by cloning the left homology arm (including the entire exon-9) and right homology arm (intron-9) into an HR vector HR100PA (SBI). We changed the targeting sequence on the left homology arm by synonymous substitution to avoid secondary gene editing by the gRNA. We transfected the CRISPR/Cas9 plasmid and the linearized targeting vector into WS iPSC by nucleofection. Puromycin-resistant colonies were expanded on MEF. We examined successfully corrected clones by detecting expression of WRN protein by Western blot. We confirmed correct homologous recombination by sequencing the genomic DNA region covering the entire exon-9 and part of intron-9. We confirmed correct mRNA splicing by sequencing the cDNA of the entire WRN open reading frame. To generate an ESC model of WS, we transfected human H1 ESC line with the same gRNA shown above by nucleofection. A nonspecific control gRNA was used to derive the control line. After colony formation from single cells (GFP-sorted), we screened individual clones by Western blot using anti-WRN antibody. WRN-null clones were validated by genomic sequencing across exon-9.

| Cell culture and differentiation
We cultured iPSC on irradiated MEF using iPSC medium (KO-DMEM containing 20% KSR, 1% NEAA, 1% GlutaMAX, 0.1 mM 2-Mercaptoethanol, 1% P/S, 10 ng/mL bFGF). Medium was changed every day until being passed with Collagenase IV. We cultured iPSC on matrigel (Corning) using mTeSR1 medium (Stemcell Technologies) before differentiation. Medium was changed every day until being passed with Dispase. To differentiate iPSC into MSC, iPSC cultured in feeder-free condition was dissociated into single cells and plated on matrigel with mTeSR1 medium containing 10 μM Y-27632 (Selleckchem). We induced early mesodermal progenitor cells using STEMdiff-ACF Mesenchymal Induction Medium

| Colony-forming unit-fibroblast (CFU-F) assay
For CFU-F assay, we seeded 150 single cells of first-passage MSC on 10-cm dish. Cells are allowed to form colonies in MSC medium (Low-glucose DMEM containing 10% FBS) for 3 weeks. At the end of the experiment, colonies were stained with 0.5% Crystal Violet and photographed.

| In vitro angiogenesis assays
We coated matrigel (BD) in cold 24-well plate (240 μl for each well) and congealed the matrigel in 37°C tissue culture incubator for 20 min. 0.5 × 10 5 HUVEC cells suspended in MSC conditioned medium (CM) and HUVEC medium (v/v 1:1) were seeded onto each well and incubated for 16 h. The morphology of tube-like structure were observed and recorded by a bright-field microscope. The total length of the branching tubes was quantified by ImageJ with an Angiogenesis Analyzer plugin (NIH). CM was collected from the same number of MSC (2 x 10 5 cells/12-well) seeded on culture plate for 16 h. In some experiments, recombinant HGF (1 ng/ml) was added to the CM for analyzing the stimulatory effect on tube formation.

| Measurement of wound
Following MSC treatment, we measured the size of wounds on each mouse on day 0, 3, 6 and 9 using a ruler. Wound closure was calculated as change of the size of the wound at each time point relative to the size at day 0. Article the bone sections of the bone defect region or the full skin tissues was performed using standard protocols as recommended by the manufacturer (EnVision System; Dako). Briefly, the paraffin embedded sections were processed for antigen retrieval by digestion with proteinase K (20 µg/ml) for 10 min, and then incubated with primary antibody against Endomucin (Abcam, ab106100; dilution 1:100), CD31 (Abcam, 28364; dilution 1:50), VEGF (Abcam, ab1316; dilution 1:200) and α-SMA (Abcam, ab23575; dilution 1:200) overnight at 4°C. A horseradish peroxidase (HRP) streptavidin detection system was used to detect the immunoactivity followed by counterstaining with hematoxylin (Sigma). The bone sections incubated with 1% nonimmune serum PBS solution served as negative controls. Quantitation of the Endomucin and CD31 positive vessel area in the bone defect region was performed using ImageJ software.

| Masson's trichrome staining
Slides are deparaffinized and rehydrated through a series of alcohol (from 100% to 70%) and washed in water. Slides are first stained in Weigert's iron hematoxylin working solution for 15 minutes and washed in running water for 15 minutes. Then a second stain in Biebrich scarlet-acid fuchsin solution for 15 minutes was performed.
Slides were incubated in phosphomolybdic-phosphotungstic acid solution until collagen was not red. Then slides were transferred to aniline blue solution and stained for 15 minutes. Finally, slides were washed in distilled water and differentiated in 1% acetic acid solution for 5 minutes. Hydrated slides were mounted with a coverslip.
Collagen was stained as blue whereas cytoplasm, muscle and erythrocytes were stained as red.

| RNAI knockdown and pharmacological inhibition of WRN
We knocked down genes using shRNA cloned in pLVTHM lentivector (Addgene No. 12247). Pseudo-lentivirus packaged in 293 T cells was used to infect MSC which were subsequently GFP-sorted by FACS. We used siRNA (GenePharma) to knock down WRN in primary MSC using siIMPORTER (Millipore). To inhibit WRN, we treated cells with NSC19630 (Millipore) at 1-3 μM for 24 h. DMSO was used as solvent control.

| RNA-SEQ and quantitative RT-PCR
Total RNA from MSC was extracted by Trizol Reagent and purified in Direct-zol RNA columns (Zymo Research). RNA was quantified using NanoDrop2000 spectrophotometer (Thermo Fisher). For RNA-seq, whole transcriptome expression was performed by the NGS platform from GROKEN Bioscience (GROKEN), according to GROKEN's suggested procedures for library construction and data analysis.
Bioinformatics was also provided by GROKEN. The raw data was deposited Gene Expression Omnibus (GSE13 7856). For RT-qPCR, mRNA was first converted to cDNA using PrimeScript RT-PCR kit (Takara). Real-time qPCR was performed using SYBR-based method in QuantStudio 7 Flex system (Applied Biosystems). GAPDH was used as normalization control.

| Statistics
Data are presented as means ±SD for three independent experiments. Statistical comparison in different experiments was calculated by GraphPad Prism 5 using two-way ANOVA or Student's t test. p < 0.05 was considered statistically significant.
The author would like to apologize for the inconvenience caused.