Neutrophil differentiation of X-CGD iPSC line iXC9 and normal control human iPSC line iPS(IMR90). Panels A-C show cytospin Giemsa stain, cytospin NBT assay, and DHR flow cytometric analysis, respectively, of neutrophils arising from differentiation of normal control iPSC line iPS(IMR90). Panels D-F are cytospin Giemsa stain, cytospin NBT assay, and DHR flow cytometric analysis, respectively, of neutrophils arising from differentiation of X-CGD iPSC line iXC9. Arrows in the cytospin Giemsa stain images (A,D) indicate some of the cells with mature neutrophil morphology (bar = 10 μm). Cytospin NBT assay shows blue-black formazan precipitates resulting from robust ROS production by PMA-activated neutrophils derived from normal control iPSCs (B), but lack of formazan precipitates indicating no ROS production by neutrophils derived from X-CGD iPSCs (E). The DHR dot plot images (y-axis side scatter; x-axis DHR fluorescence) demonstrate robust ROS production (high fluorescence) by 38.5% of PMA-activated cells representing mature neutrophils derived from normal control iPSCs (C), but no evidence of any significant ROS production by PMA-activated neutrophils derived from X-CGD iPSCs (F).

Analysis of APC-gp91 minigene plasmid ZFN-mediated HR targeting to the AAVS1 locus in X-CGD iPSCs. (A) Upper schematic of APC-gp91 minigene donor plasmid (pAAVS1-puro-CAG-gp91) and lower schematic of the PPP1R12C gene structure that includes exons 1, 2, and 3 flanking the AAVS1 locus that is within intron 1. Dotted lines indicate AAVS1 homology arms within the targeting donor plasmid. The puromycin-polyA promotorless cassette (SA-2A-Puro-pA) relies on a splice accepter plus 2A linker to use the PPP1R12C promoter for expression. The gp91^phox minigene cassette uses CAG (CMV early enhancer/chicken β-actin) promoter to express a codon-optimized gp91^phox cDNA-polyA. Restriction enzyme Sph1 (S) sites and 3′- and 5′-probes used for Southern blot are shown in both schematics. Not shown for the donor plasmid is a Sph1 restriction site 4.3 kb upstream of the 2 closely spaced Sph1 restriction sites within the schematic of pAAVS1-puro-CAG-gp91 plasmid. This upstream Sph1 site would not be present with a correctly targeted insert, but is often included with off-target random insertion of the donor plasmid. (B) Sph1 restriction Southern blot analysis of some of the APC-gp91 minigene targeted iXC9 (iXC9-APC-gp91) clones. 3′-probe (upper blot) detects wild-type allele (WT) 6.5 kb fragment in all clones except no. 20 (in which both alleles are targeted) and the 3′-targeted integration (TI) 6.0 kb fragment. 5′-probe (lower blot) detects WT 6.5 kb fragment in all but clone 20, the 5′ TI 3.8 kb fragment in all clones, and additional fragment(s) generated by random insertion in some clones (note the commonly seen 4.3 kb fragment derived entirely from donor plasmid sequence that is indicative of random insertion, and the 5.8 kb random insertion fragment seen in clone 20). From this particular analysis, clones indicated in red containing no random insert and a single-copy APC-gp91 insert in AAVS1 (nos. 9, 21, 23, 25, 26) were chosen for NHEJ screening (Table 3). (C) Flow cytometry gp91^phox transgene expression by iXC9-APC-gp91 c21 (right panel) and lack of gp91^phox expression by parental X-CGD iPSC clone iXC9 (left panel) (y-axis side scatter; x-axis fluorescent anti-gp91^phox). (D) iXC9-APC-gp91 c21 normal karyotype.

Functional restoration of ROS production in neutrophils differentiated from iXC1-APC-gp91 c3. Panels A and B are cytospin Giemsa stain and DHR flow cytometric analysis, respectively, of neutrophils arising from differentiation of normal control iPSC line iPS(IMR90); and panels C and D are cytospin Giemsa stain and DHR flow cytometric analysis, respectively, of neutrophils arising from differentiation of ZFN-targeted minigene-corrected X-CGD iPSC clone iXC1-APC-gp91 c3. Arrows in the cytospin Giemsa-stain images (A,C) indicate some of the cells with mature neutrophil morphology (bar = 10 μm). The DHR dot plot images (y-axis side scatter; x-axis DHR fluorescence) demonstrate robust ROS production (high fluorescence) by 28.8% and from 26.8% of cells, respectively, representing PMA-activated neutrophils derived from normal control iPSCs (B) and from ZFN-targeted minigene-corrected X-CGD iPSC clone iXC1-APC-gp91 c3 (D).

Characterization of X-CGD iPSC line iXC1. Line iXC1 (A) exhibits characteristic human ESC-like morphology in bright field (BF) microscopy (top left) and expresses pluripotency markers that include alkaline phosphatase (AP) activity (top right) and fluorescence antibody staining for OCT4, NANOG, TRA-1-60, and SSEA-4 (left panels in green and red). Blue staining (right panels) are 4′-6-Diamidino-2-phenylindole (DAPI) staining of nuclei (bar = 100 μm); (B) has the CYBB 458T→G mutation; (C) has normal male karyotype; (D) forms embryoid bodies in vitro shown by BF (top left) microscopic appearance, and by expression of β-tubulin class III neuroectoderm marker (top right), smooth muscle actin (SMA) mesoderm marker (bottom left), and α-fetoprotein (AFP) endoderm marker (bottom right); (E) forms teratomas in vivo in NSG mice that include all 3 germ layers, identified here as neural rosette ectoderm (top), cartilage mesoderm (middle), and respiratory epithelial endoderm (bottom). Not shown is that RT-PCR analysis of cDNA generated from the iXC1 teratoma confirmed presence of mRNA specific for human PAX6 (ectoderm), CD34 (mesoderm), and α-feto-protein (endoderm).

Lentivirus vector gene transfer mediated expression of gp91^phox or GFP in X-CGD iPSCs. (A) Schematic of CL20i4rEF1α-Puro-T2A-gp91 self-inactivating bicistronic lentivirus vector (LV) insert with internal EF1α (short intronless) promoter co-expressing puromycin resistance gene and (via T2A element) codon-optimized gp91^phox. This vector also contains the 400 bp version of chicken insulator cHS4 (CI). Not shown are the series of closely related CL20 lentivectors constructs with the same backbone used for the experiment in panel D of this figure, but expressing GFP instead of gp91^phox, using EF1α short, EF1α long, CAG, or PGK promoters. (B) Flow cytometric DHR analysis of ROS production by neutrophils differentiated from LV-transduced line iXC9 X-CGD iPSC shows only 4% of cells are weakly DHR positive (compare with Figure 2C and F), even though almost all of the transduced and puromycin selected iXC9 cells strongly expressed gp91^phox just before differentiation (as shown in panel E). (C) Flow cytometric DHR analysis of ROS production by neutrophils differentiated from the same LV-transduced iXC9 line, but maintained in puromycin selection during the differentiation, results in 28% of cells DHR-positive, a value similar to that in Figures 2C and 6B for the normal control. (D) Promoter-based differences in GFP lentivirus expression in unselected polyclonal populations of transduced iXC9 iPSCs during neutrophil differentiation. GFP expression during in vitro neutrophil differentiation of iXC9 iPSCs transduced with lentivirus containing GFP expressed from EF1α short, EF1α long, CAG, or PGK promoter. The proportion of cells remaining GFP positive was calculated relative to the initial level (range 73%-90%) observed in the undifferentiated iPSCs at the start of the observation period, which was set to 100% at baseline to allow direct comparison of the proportional change. (E) Shown is the decrease over time in culture of gp91^phox transgene expression measured by flow cytometry in iXC9 or iXC1 transduced with CL20i4rEF1α-Puro-T2A-gp91 lentivirus (blue diamonds and purple asterisks), compared with the very long-term stability of expression of gp91^phox from the AAVS1 locus targeted APC-gp91 minigene in iXC9 and iXC1 (green triangles, red squares, and orange circles). For all of these transduced or gene-targeted X-CGD iPS lines, day 0 on this graph corresponds to the time at which lines completed puromycin selection and were then grown thereafter for the period of time shown in the absence of puromycin.

Analysis of the top 10 putative off-targets for AAVS1 ZFNs in targeted clones to rule out NHEJ mutations at those sites. Ten SELEX-predicted genomic off-targets were PCR-amplified and products sequenced for iXC1-APC-gp91 c3 (top chromatograph), iXC9-APC-gp91 c9 (middle chromatograph), and iXC9-APC-gp91 c21 (bottom chromatograph) for each of the 10 sites as indicated. Because the PCR products are from a homogenous population in each clone, mutations in the off-target sequence will show as single/double peaks/nucleotides that are different from reference sequence. None of these 3 clones showed any mutations in the off-target sequences.