TALENs Facilitate Single-step Seamless SDF Correction of F508del CFTR in Airway Epithelial Submucosal Gland Cell-derived CF-iPSCs.

Cystic fibrosis (CF) is a recessive inherited disease associated with multiorgan damage that compromises epithelial and inflammatory cell function. Induced pluripotent stem cells (iPSCs) have significantly advanced the potential of developing a personalized cell-based therapy for diseases like CF by generating patient-specific stem cells that can be differentiated into cells that repair tissues damaged by disease pathology. The F508del mutation in airway epithelial cell-derived CF-iPSCs was corrected with small/short DNA fragments (SDFs) and sequence-specific TALENs. An allele-specific PCR, cyclic enrichment strategy gave ~100-fold enrichment of the corrected CF-iPSCs after six enrichment cycles that facilitated isolation of corrected clones. The seamless SDF-based gene modification strategy used to correct the CF-iPSCs resulted in pluripotent cells that, when differentiated into endoderm/airway-like epithelial cells showed wild-type (wt) airway epithelial cell cAMP-dependent Cl ion transport or showed the appropriate cell-type characteristics when differentiated along mesoderm/hematopoietic inflammatory cell lineage pathways.

Introduction

Cystic fibrosis (CF) is caused by mutations in the CF transmembrane conductance regulator (CFTR) gene.[1] Of the >2,000 disease-causing mutations detected in the CFTR gene (http://www.genet.sickkids.on.ca/cftr/), a trinucleotide (CTT) deletion that spans codons 507 and 508 and results in loss of a phenylalanine at amino acid 508 (F508del) of the CFTR protein is found in ~70% of all CF alleles.[2] CF patients typically exhibit a variety of pathologies that include abnormal mucus accumulation in airways and lungs, accompanied by opportunistic bacterial infections that appear to be associated with both airway epithelial cell (AEC) and immune cell dysfunction. Recent studies suggest that CFTR is a component of the monocyte and macrophage response to infection in CF patients.[3,4] Since CF-associated pathologies result in extensive tissue damage, treatment of CF will require a comprehensive strategy that both corrects the underlying genetic defect and repairs/regenerates damaged tissues. In this context, the ability to reprogram mature somatic cells into induced pluripotent stem cells (iPSCs)[5,6] has opened the door for development of a comprehensive, personalized cellular therapy for CF.[7] These patient-specific iPSCs have the potential of generating transplantable, autologous cells/tissues that circumvent rejection by the host immune response, enhancing the potential for successful engraftment and tissue repair and avoiding the need for immunosuppressive drugs.[8,9,10] Several studies have already indicated that embryonic stem cells and fibroblast-derived CF-iPSCs can be differentiated into cells that have properties of endoderm[11,12,13] and airway epithelium.[14,15,16,17,18] Ultimately, further refinement of such cell differentiation protocols should be able to produce cells that will successfully rebuild damaged airways.

An important component of a comprehensive therapy for CF is the repair of the disease-causing CF mutation(s). Restoration of wild-type (wt) CFTR function in the repaired tissues will be critical in ameliorating the dysfunction associated with the CFTR mutation. The sequence-specific gene-editing approach, small/short fragment homologous replacement (SFHR), has been applied to numerous genomic targets, including CFTR, in multiple cell systems both in vitro and in vivo.[19,20,21,22] SFHR is mediated by polynucleotide small/short DNA fragments (SDFs), but until now, has not been applied to correct CFTR mutations in human CF-iPSCs. While SFHR-driven homologous exchange (HE) efficiencies as high as ~10% have been observed with microinjection,[22,23] the efficiency of HE can range between 0.05 to ~5%, depending on the cells, the method of nucleic acid delivery or other transfection parameters.[19,24] Since transcription activator-like effector nucleases (TALENs)[25,26,27,28] and clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 nuclease[29,30,31] mediate DNA double strand breaks (DSBs) by enhancing the efficiency of homologous recombination between donor plasmid DNA and a genomic target, we reasoned that this induction of DSBs could facilitate SDF-mediated HE as well. In this study, TALENs were used to minimize off-target effects associated with the CRISPR/Cas9 system[32,33] and enhance SDF-mediated correction of the F508del CFTR in CF-iPSCs.

Results

Generation of CF-iPSCs

Primary airway submucosal gland AECs (CFSME101) from a CF patient homozygous for the F508del mutation were reprogrammed by transduction with four individual retroviruses, each containing one canonical transcription factor (OCT4, SOX2, KLF4, or c-MYC) and grown for 3–4 weeks on mitomycin C inactivated mouse embryo fibroblast feeders until candidate iPSC colonies appeared. Initially, five candidate iPSC colonies were generated (CF1-iPS1, CF1-iPS2, CF1-iPS3, CF1-iPS4, and CF1-iPS5) giving a reprogramming efficiency of ~2 × 10−4. Of the five clones generated, four were selected for further analysis and experimentation.

The F508del/F508del CFTR genotype of the parental CFSME101 primary cells and the CF1-iPSC lines was confirmed by allele-specific PCR (AS-PCR; Supplementary Figure S1a) and DNA sequence analysis of PCR products generated by non-AS-PCR (Supplementary Figure S1b). Immunocytochemical analysis showed that the CF1-iPSC clones expressed pluripotent markers SSEA3, SSEA4, TRA-1–60, TRA-1–81, and NANOG (Supplementary Figure S1c, Supplementary Table S1). Pluripotence was further demonstrated in vitro by expression of α-fetoprotein (endoderm), TUJ1 (ectoderm), and α-smooth muscle actin (mesoderm) in embryoid body cells (Supplementary Figure S1d, Supplementary Table S1) and in vivo by tissues derived from teratomas generated in immunodeficient NGS mice representing the three primordial germ layers (Supplementary Figure S1e). Cytogenetic analysis of cell lines CF1-iPS1, -iPS4, and -iPS5 between P5.6-P5.8 (where passage number PX.Y.etc = X passages before transduction/reprogramming, Y passages since candidate colony isolation) showed a normal diploid female karyotype (46,XX; Supplementary Figure S1f).

TALEN enhanced correction of F508del CFTR

A 491 bp wtCFTR donor SDF (491z-SDF) was used for correction, since this SDF alone has been previously shown to mediate HE at the genomic F508del CFTR locus in AECs.[19,20,22] Sequence-specific CFTR-TALEN pairs (CFTAL-1B and CFTAL-2B; Supplementary Figure S2a) were designed to enhance 491z-SDFs-mediated HE and also cut only in the mutant DNA. CF1-iPS4 cells were transfected with either 107 or 2 × 107 SDFs/cell in absence or presence of CFTAL-1B and CFTAL-2B expression vectors. The advantage of this TALEN pair is that one of the pair is allele specific for the mutant allele and, as such, has enhanced binding to and cutting of the mutant allele. Cells were harvested after 3 days with Accutase or Dispase, subcultured, and analyzed by AS-PCR for wtCFTR DNA. Subcultured cells were harvested again on days 7 and 9 for analysis. CF1-iPS4 cells cotransfected with SDFs and TALENs appeared to have significantly more wtCFTR DNA than those transfected with SDFs alone ( and Supplementary Figure S2b,c), indicating enhancement of SDF-mediated HE at the CFTR locus by TALENs.

Isolation and characterization of corrected CF-iPSCs

Corrected CF1-iPS4 cell clones were isolated by a cyclic dissociation/PCR enrichment protocol involving dissociation of the cells with Dispase, subculturing as cell clumps into individual wells of a 12-well plate and screening by AS-PCR (see Supplementary Materials and Methods).[34] Enhancement of SDF-mediated HE was observed in the corrected cell subpopulation at day 9 after cotransfection with 2 × 107 SDFs/cell and TALENs CFTAL-1B and CFTAL-2B (, lane 21).

Approximately 100-fold enrichment in the corrected CF1-iPS4 cell population (~10% corrected cells) was achieved after five to six enrichment cycles (ECs) ( and Supplementary Figure S2d). At the sixth enrichment cycle (EC6), cells in well EC6 C2.1 (P5.55.15), derived from well C2 (P5.55.14), had the same apparent ratio of F508delCFTR:wtCFTR as in well EC6 C2 (P5.55.14) (). The EC6C2.1 cells were plated at single cell densities on a 60-mm dish. Of the 20 clones isolated, one clone (c1) (P5.55.17) indicated both wt and F508delCFTR by AS-PCR ( and Supplementary Figure S3). A subsequent isolation from EC8, EC8 B3.1 (P5.55.19), yielded 6 heterozygote clones of the 30 isolated (20%). DNA sequence analysis of CF1-iPS4 clone 1 (CF1-iPS4c1) confirmed heterozygosity (). Allele-specific RT-PCR analysis (Supplementary Table S2) of CFTR mRNA expression at P5.55.21 showed expression of both wt and F508delCFTR mRNA, whereas untransfected CF1-iPS4 p5.14 showed only F508delCFTR mRNA expression (). While iPSCs express low levels of CFTR mRNA (S Suzuki, DC Gruenert, unpublished observations), it was still possible to differentiate between wt and F508delCFTR mRNA expression in cultures also expressing pluripotence markers. The pluripotence of the corrected cells (, Supplementary Table S1) was further indicated by their ability to form three primordial germ layers in embryoid bodies in vitro (, Supplementary Table S1).

Off-target events due to TALEN/SDF-mediated genome editing were evaluated by Southern blot hybridization for detection of random integration and by screening the most likely off-target TALEN-binding sites that differ from the intended target site by up to seven mismatches per TALEN half-site. Southern blot analysis showed no spurious donor DNA random integration bands in two distinct corrected clones (CF1-iPS4c1 and CF1-iPS4D2) when compared to untransfected parental CF1-iPS4 cells (PC) (). All three samples gave a single band at 8746- or 7530-bp, following gDNA digestion with EcoRV and XbaI, respectively, indicating that there was no random integration of the SDF with TALEN-enhanced correction.

Candidate off-target sequences were selected based on sequence homology and repeat-variable diresidue-binding preferences using the PROGNOS algorithms. The 24 top-ranked sites were assessed by Sanger sequence analysis of genomic DNA from parental, CF1-iPS4 P5.50, and corrected CF1-iPS4c1 P5.55.29 cells (Supplementary Table S3). No mutations could be attributed to TALEN-induced off-target events (). Based on the reference hg19 genome, the several single-nucleotide polymorphisms present in both corrected and uncorrected CF1-iPS4 were identical. Sequence analysis at three sites failed and was not further considered. While this is not a comprehensive off-target investigation of the entire genome, it is, nonetheless, encouraging that no off-target events were observed at any of the 21 independent sites evaluated.

Differentiation and phenotypic characterization of corrected CF1-iPSc1 cells

Corrected CF1-iPS4c1 cells were also evaluated for CFTR function and the ability to differentiate into cells showing airway epithelial-like features to determine their potential usefulness in cell therapy for CF. Transepithelial Cl ion transport, a hallmark of CF, was assessed in CF1-iPS4c1 cells grown to confluence on cell culture inserts with an air–liquid interface. After 19 and 20 days in culture, respective transepithelial resistances (Rt) in Ussing chambers were 45 and 63 Ωcm2 and forskolin-activated, CFTRinh172-blocked, CFTR-specific Cl currents were −0.5 and −1.0 μA/cm2 (mean = −0.75 μA/cm2) ( i top). Expression of zonula occludens-1 (ZO1) protein, a marker of epithelial cell differentiation, was found widely expressed in the cultures, although it was not uniform throughout the monolayer ( i bottom, Supplementary Table S1), suggesting incomplete epithelial differentiation. A previous study with human mesenchymal stem cells showed transdifferentiation of mesenchymal stem cells co-cultured with human AECs under air–liquid interface conditions that led to AECs with functional CFTR expression.[35] In that context, CF1-iPS4c1 cells were co-cultured with CFBE41o-AECs and assayed for differentiation and CFTR function. CF1-iPS4c1/CFBE41o- co-cultures were evaluated up to 20 days for transepithelial Cl secretion and ZO1 expression (, Supplementary Table S1). After 7 days, cAMP-dependent Cl currents, activated by forskolin and blocked by CFTRinh172, were similar to those observed for CFBE41o- alone ( i.v., Rt > 1,000 Ω.cm2, ICl = −0.80 μA/cm2). However, from days 12 to 20, the average CFTRinh172 blocked CFTR current was −4.70 ± 2.0 μA/cm2 ( iii) (significantly larger than Cl currents in CF1-iPS4 alone). All CF1-iPS4c1 co-culture islands showed prominent ZO1 expression, indicating epithelial differentiation, tight junction formation, and cell polarity. Characteristic membrane localized peripheral ZO1 staining of CF1-iPS4c1 islands became more prominent with time in co-culture; however, the CF1-iPS4c1 cells still showed tight junction discontinuities when compared to pure CFBE41o- cultures (, Supplementary Table S1). Overall, these observations indicate that CF1-iPS4c1 cells begin to adopt features of the CFBE41o- AECs and show CFTR expression and function. HE-mediated correction of CFTR-associated Cl current was further demonstrated by comparing three iPSCs with distinct CFTR genotypes: (i) non-CF, FHTF-iPSCs - wt/wt CFTR, (ii) corrected CF1-iPS4c1 cells - wt/F508del, and (iii) uncorrected CF1-iPS4A1 cells - F508del/F508del (Supplementary Figure S4a). When co-cultured with CFBE41o- cells (Supplementary Figure S4b), assessed for their CFTR mRNA expression (Supplementary Figure S4c) and assayed for transepithelial Cl ion transport (), CFTR (wt and/or F508del) mRNA expression in all three cell lines was much higher on day 9 than on day 6 and was maintained (FHTF-iPSCs) or increased (CF1-iPS4c1 and A1) with additional days in co-culture (Supplementary Figure S4c). While there were no changes in short circuit current (▵ISC≈0) on day 6 in response to Cl transport activators, significant changes in ISC were observed on days 9, 13 (), and 15. At day 9, uncorrected cells showed 6.1 ± 1 µA/cm2 and CFTR-corrected cells showed 8.2 ± 1.6 µA/cm2 of amiloride-blocked currents. Treatment of the co-cultures with CFTR-specific VX-770 that stimulated or CFTRinh172 that blocked CFTR-dependent Cl current, in FHTF-iPS and CF1-iPS4c1 cells, but not in CF1-iPS4 A1 cells after 13 and 15 days in co-culture (,), indicating that the TALEN/SDF-mediated genetic correction in CF1-iPS4c1 cells led to wtCFTR functional activity.

Using a more controlled, defined airway differentiation protocol (, Supplementary Table S1), lineage specific proximal airway-like epithelial cell progression of CF1-iPS4c1 cells was achieved. With the defined airway differentiation protocol, cultures showed enrichment for cells expressing NKX2.1 and FOXJ1, markers of ciliated AECs (days 43–61) (, Supplementary Table S1) as well as concomitant expression of CFTR protein with wtCFTR function (day 62) (). These observations are consistent with proximal airway-like differentiation into ciliated cells and functional correction of the CF1-iPS4c1.

Because CF pathology has a significant inflammatory component associated with macrophages,[36,37,38,39] CF1-iPS4c1 cells were directed to differentiate along a hematopoietic lineage-specific pathway into monocyte- and macrophage-like cells (Supplementary Figure S5a). After day 7, CF1-iPS4c1 cells expressing CD34 and KDR, but not CD45, were observed by flow cytometry, indicating the early stages of hematopoietic/endothelial cell differentiation (Supplementary Figure S5b, Supplementary Table S1). By day 16, cells with endothelial markers KDR, CD34, CD31, and CD144, as well as a small population of cells expressing hematopoietic markers, CD45 and CD34 (including myeloid progenitors expressing CD45 and CD33) (Supplementary Figure S5c, Supplementary Table S1) was detected. On day 49, flow analysis, indicated the presence of some CD34low expressing cells, and a distinct CD45+ CD33- population that included a small proportion of CD14 expressing monocytes (Supplementary Figure S5d, Supplementary Table S1). The flow data was corroborated by qRT-PCR analysis. On day 49, CD14 mRNA levels were significantly increased, consistent with the presence of cells in the monocyte/macrophage lineage pathway (Supplementary Figure S5e, Supplementary Table S2).

Discussion

The differentiation of CF-iPSCs into cells with an inflammatory cell features provides the potential for assessing and comparing the parental CF1-iPS4–derived inflammatory cells to the corrected CF1-iPS4c1 cells. As with airway-like epithelial cell differentiation, differentiation into inflammatory cells will aid in development of a cellular therapy for regenerating an intact hematopoietic system in CF patients, as well as in development of a disease model system to assess potential pharmacological therapeutic agents. While further studies are required to identify the appropriate engraftable progenitor cells to be used in repopulating the airways and hematopoietic system, these studies lay the foundation for developing a seamless gene correction strategy in iPSCs generated with a nonintegrating reprogramming vehicle. In addition, the airway and hematopoietic differentiation studies are a critical part of developing and refining xeno-free cellular therapies for CF.

One recent study in CF adult intestinal stem cell organoids showed apparent functional correction of F508del CFTR using CRISPR/Cas9 to facilitate homologous recombination.[32] Genetic correction of these multipotent adult stem cells resulted in swelling properties similar to those observed in organoids derived from healthy individuals. However, due to off-target events and the selection marker remaining in the genomic DNA, the approach described is not footprint free and has the risk of unintended adverse mutagenesis. In addition, the cells that were corrected were not pluripotent and had a limited potential in terms of their ability to differentiate into other cell types within the body. The TALEN/SDF strategy described here shows no apparent off-target events (,) and appears to be seamless in the sense that there are no other alterations in the targeted region. Moreover, cells/tissues associated with CF pathology, will need to have pluripotence for a multi-tissue therapeutic value, a requirement that intestinal stem cell organoids will not satisfy. Two other recent papers demonstrate functional correction of CFTR in iPSCs using ZFNs[16] or CRISPR/Cas9s.[18] However, unlike the present study, the correction of the CFTR mutation utilizes a donor plasmid that contains a selection marker gene and relies on a protocol that is not footprint free and a seamless single-step strategy for isolating cells with corrected genomic DNA. Therefore, the overall strategy presented here of generating patient-derived iPSCs combined with seamless, single-step, mutation correction as well as co-culture-induced and/or directed multilineage differentiation will provide an effective starting platform for building a personalized cell and gene-based CF therapy.

In summary, the studies described are distinct from previous studies, showing that human airway submucosal gland epithelial cells can be reprogrammed into CF-iPSCs and corrected by a seamless DNA fragment (polynucleotide)-based SFHR strategy that combines wtCFTR-SDFs (not a plasmid donor DNA) with CFTR-specific TALENs. Through a straightforward cyclic enrichment protocol[34] corrected clones that retain their pluripotence were isolated without selection. Another unique aspect of this study is the demonstration that a corrected CF-iPSCs can be “differentiated” into airway epithelial-like cells through co-culture with immortalized AECs. These corrected CF-iPSCs showed wild-type cAMP-dependent Cl ion transport properties after co-culture and directed differentiation along an endodermal lineage pathway into cells that had proximal airway epithelial-like properties. In addition, the corrected CF1-iPS4c1 cells were also able to differentiate into inflammatory cells expressing macrophage and monocyte markers.

Materials and Methods

Cells and culture conditions. All human tissue studies were approved by the UCSF Committee on Human Research (CHR) and California Pacific Medical Center (CPMC) Institutional Review Board (IRB). Primary CF tracheobronchial submucosal gland epithelial cells (CFSME101), homozygous for the F508del CFTR mutation, were obtained from the UCSF CF Cell Culture Core Laboratory (Director, Dr WE Finkbeiner) and grown in MLHC8e medium.[40] Immortalized CFBE41o- cells,[41,42,43,44] used for the co-culture studies, are homozygous for the F508del CFTR mutation were routinely grown in supplemented Eagle's Minimal Essential Medium and subcultured with polyvinylpirrolidone/ethylene glycol tetraacetic acid/trypsin. Co-culture with iPSC-derived cells was carried out on Transwell inserts in mTeSR1 medium.

Human iPSCs were generated by retroviral reprogramming[5] according to guidelines developed by the CPMC Research Institute Stem Cell Research Oversight Committee and the UCSF Gamete and Embryonic Stem Cell Research Committee. The iPSCs were grown and subsequently subcultured by mechanical isolation and enzymatic dissociation of individual iPSC colonies with collagenase type IV (Invitrogen, Carlsbad, CA)[45] or Dispase (StemCells, Inc, Vancouver, British Columbia, Canada).

Cell passage number is denoted as Pn1.n2.n3. - nz, where n1 = number of passages as primary cells before reprogramming, n2 = number of passages since reprogramming, n3 = number of passages after transfection with SDF, etc., where each period delineates the onset of a specific protocol or treatment that alters the character of the cells.[46]

Generation of sequence-specific TALENs. TALEN targeting sequences were designed using Web-based software, TALE-NT 2.0 (https://boglab.plp.iastate.edu/).[47] The following sequences were selected: TALEN pairs CFTAL1B: 5′-T CTCAGTTTTCCTGGATTAT, spacer: gcctggcaccattaaagaa, CFTAL2B: AATATCATTGGTGTTTCCT A-3′. CFTAL2B is allele specific and preferentially binds to the mutant genomic sequence, and will therefore not bind and contribute to cleavage of donor and repaired DNA (Supplementary Figure S2a). We were unable to generate a comparable CRISPR/Cas9 nickase (Cas9n) pair for this locus.[48] CFTR-B TALEN plasmids were assembled by following Golden Gate TALEN assembly method[25] with the Golden Gate TALEN plasmid kit (Kit # 1000000024, Addgene, Cambridge, MA). In the second Golden Gate reaction, a novel backbone plasmid (MR015; MH Porteus and M Rahdar, unpublished data) was used for optimal mammalian cell expression.

SDF-mediated correction with TALENs. CF1-iPS4 cells, P5.55, were nucleofected with the 491z-SDF at 107 SDFs/cell or 2 × 107 SDFs/cell in absence or presence of CFTAL-B TALENs. Genomic DNA was isolated on days 3, 7, and 9 posttransfection and amplified with AS-PCR primers CF1B/CF7C or CF1B/CF8C (Supplementary Table S2). The resulting PCR products were analyzed on a 2% agarose gel. The well containing the highest relative amount of corrected cells as indicated by band intensity was then subjected to a multiple enrichment cycle protocol similar to those described previously.[34,49] This protocol involved (i) dissociation with dispase and approximately equal distribution of cells into each well of a multiwell plate (Falcon-Beckton Dickenson Labware, Franklin Lakes, NJ), (ii) further growth, (iii) wtCFTR AS-PCR analysis (Supplementary Table S2) of an aliquot from each well, and (iv) subculture/distribution of the cells in the well with the apparent highest proportion of corrected cells, into another multiwell plate.

DNA from untransfected CF1-iPS4 and non-CF-iPSCs was used as controls for the F508del and wtCFTR AS-PCR amplification, respectively. The sensitivity of the AS-PCR analysis and the proportion of corrected cells within the population of transfected cells were determined by mixing genomic DNA from a non-CF-iPSC line (SC2-iPSrs3) with varying percentages of CF1-iPS4 DNA.[50]

Allele-specific reverse transcription PCR. Total RNA was extracted using ReliaPrep RNA Cell Miniprep System (Promega, Madison, WI) and then synthesized into cDNA using ProtoScript M-MuLV First Strand cDNA Synthesis Kit (New England BioLabs, Ipswich, MA). Allele-specific reverse transcription PCR for CFTR expression was performed with the CF7C (wt) or CF8C (F508del) paired with a non–allele-specific primer, CF17 (Supplementary Table S2).

Corrected CF1-iPS4c1 and CF airway epithelial cell co-cultures. Immortalized CF bronchial epithelial cells, CFBE41o-, have been shown to promote trans-differentiation of marrow-derived mesenchymal stem cells to acquire an AEC phenotype.[35] In a similar context, CFBE41o- cells were co-cultured with CFTR corrected CF1-iPS4c1, CFTR wild-type FHTF-iPS1 and CFTR uncorrected CF1-iPS4A1 cells to promote the differentiation of the iPSCs into AECs. CFBE41o- cells are homozygous for the F508del mutation, express very low levels of F508del CFTR mRNA, and readily form tight epithelial cell monolayers.[44]

Mixed CFBE41o-/iPSC co-cultures were generated by adding ~5 × 104 CFBE41o- cells/cm2 to iPSCs plated on 12-mm clear polyester Snapwell inserts (Corning Life Sciences, Acton, MD). The CF1-iPS4c1/CFBE41o- co-culture was grown in mTeSR1 medium and transepithelial ion transport measurements were made up to 20 days post-plating. An air–liquid interface[51] was established at day 12 and maintained throughout the rest of the experiment.

Measurement of CFTR-dependent Cl transport. Short circuit current (ISC) of CFTR-dependent Cl currents were measured in Easy Mount Ussing chambers (Physiologic Instruments, San Diego, CA) across cell monolayers grown on Snapwell inserts as previously described.[52] Transepithelial voltage was clamped to 0 mV with a standard four-electrode voltage clamp. ISC was continuously recorded. At 60-second intervals, transepithelial voltage was clamped to 1 mV for 1 second to monitor transepithelial resistance (Rt). A serosal-to-mucosal Cl gradient was established to increase the electrochemical driving force for Cl secretion. Transepithelial ISC measured under these conditions was termed ICl. The serosal solution comprised: (in mmol/l) 120 NaCl, 25 NaHCO3, 5 KCl, 1.2 NaH2PO4, 5.6 glucose, 1.0 CaCl2, and 1.2 MgCl2, while the apical Cl-free solution comprised: (in mmol/l) 120 Na-gluconate, 20 NaHCO3, 5 KHCO3, 1.2 NaH2PO4, 5.6 glucose, 2.5 Ca(gluconate)2, and 1.2 MgSO4. Experiments were carried out at 37 °C, and solutions were gassed with 5% CO2 in air for a solution pH of 7.4. Positive currents were defined as the movement of anions from the serosa to mucosa. Forskolin (Calbiochem, La Jolla, CA) (5 or 20 μmol/l) was serosally added to stimulate CFTR activity. VX-770 (5 μmol/l), a CFTR channel opener was added to the apical side. CFTRinh172 (CF Foundation Therapeutics; 50 μmol/l), a CFTR-specific chloride channel inhibitor, was added to the apical side quantifying CFTR involvement in transepithelial Cl currents. In addition, glibenclamide (500 μmol/l), a general Cl transport inhibitor that blocks both CFTR and non-CFTR Cl transport, was used to determine the amount of non-CFTR current contributed to the overall Cl current. Recordings on matched cultures were performed side-by-side.

Figure S1. Characterization of CF1-iPSC lines. Figure S2. Clonal enrichment of CF-iPSCs corrected with SDFs and TALENs. Figure S3. Corrected clone isolation. Figure S4. Assessment of control iPSCs. Figure S5. Lineage differentiation of CF1-iPS4c1 cells into inflammatory hematopoietic cells. Table S1. Antibodies. Table S2. PCR primers and products. Table S3. Potential CFTAL-B off-target binding sites and PCR analytical primers. Materials and Methods