Fig 1.
Generation of PIGAc.1234C>T mutation using the PiggyBac transposon system.
(A). Map of PiggyBac (PB) constructs for PIGAwt (top), PIGAc.1234C>T (middle) and PIGAtr411 (bottom). c.1234C>T (p.Arg412*) is a nonsense point mutation in the 6th exon of the PIGA gene that predicts a truncated protein missing the final C-terminal 73 amino acids. PIGAtr411 is the truncated form of the PIGA cDNA, lacking the coding sequence of the C-terminal 73 amino acids. (B). Representative FACS analysis CD59 expression in TF1PIGAnull cells transfected with PB-PIGAwt, PB-PIGAc.1234C>T or PB-PIGAtr411. Transfected TF1PIGAnull cells were stained with an APC-conjugated CD59 antibody to assess PIGA gene expression. Non-transfected TF1PIGAnull cells were used as a control. MFI represents mean fluorescence intensity.
Fig 2.
The PIGAc.1234C>T mutation increases PIGA function compared to PIGAnull hiPSCs.
(A). Representative example of FACS analysis CD59 expression in the three hiPSC lines. Overlay histogram shows that CD59 expression was significantly higher in PIGAc.1234 C>T hiPSCs compared to PIGAnull hiPSCs. MFI was 445.4 in PIGAwt hiPSCs and 332.6 in PIGAc.1234C>T hiPSCs (p>0.05, NS). However, MFI in PIGAc.1234C>T hiPSCs was significantly higher than 17.9 in PIGAnull hiPSCs (*p<0.05). The results indicated PIGA gene function was partially restored in PIGAc.1234C>T hiPSCs. PIGAnull hiPSCs (purple), PIGAc.1234C>T hiPSCs (orange) and PIGAwt hiPSCs (blue). (B). Representative example of Alkaline Phosphatase (AP) activity in the three hiPSC lines was detected by an Alkaline Phosphatase Detection Kit. AP activity was examined under light microscopy (20X magnification, scale bar 50μm). AP activity was increased in PIGAc.1234C>T hiPSCs compared to PIGAnull hiPSCs. Further confirming PIGAc.1234C>T hiPSCs possess partial PIGA gene function. (C). Representative Western blot: BMP4 induction in three hiPSC lines. PIGAwt, PIGAnull and PIGAc.1234C>T, were treated with or without 50ng/mL BMP4 for 4 hours and levels of phosphoSmad1 (Ser463/465)/Smad5 (Ser463/465)/Smad8 (Ser426/428) were detected by immunoblotting. Intensity of Smad 1/5/8 phosphorylation was comparably increased following BMP4 induction in both of PIGAwt hiPSCs and PIGAc.1234C>T hiPSCs; decreased responsiveness to BMP4 induction was observed in PIGAnull hiPSCs as previously described. β -actin was used as an internal control.
Fig 3.
The PIGAc.1234C>T mutation does not impair terminal hematopoietic differentiation during mesoderm induction.
(A). Quantitation of EB-derived blood-like cells (EB-BLCs). A total of 100 hiPSC-derived EBs were assessed for each hiPSC derived cell line and the percentage of EB-BLCs present was determined. The percentage of EB-BLCs derived from PIGAwt hiPSCs was significantly higher than EB-BLCs from PIGAc.1234C>T hiPSCs (**P<0.01, Mann-Whitney test). Results shown are average and standard deviation (mean ± SD) based on three independent experiments. (B). Expression of hematopoietic markers CD33, CD34 &CD45 in EB-BLCs from PIGAwt and PIGAc.1234C>T hiPSCs. The cells were stained with anti-human CD33, anti-human CD34 and anti-human CD45 for flow cytometry analysis. Antibodies and analyzed by flow cytometry. The values shown are from three independent experiments. All values represent average and standard error (mean ± SE). (C). Representative example of FACS analysis of hematopoietic phenotypes in the EB-BLCs from PIGAwt and PIGAc.1234C>T. The zebra plot shows expression of CD59 (X-axis) and CD45 (Y-axis) after three and eight days of hematopoietic differentiation. Unstained PIGAwt cells were used as a control. (D). Enumeration of colony forming units (CFU) from the BLCs derived from PIGAwt and PIGAc.1234C>T. There was no significant difference in CFU colony formation between PIGAwt and PIGAc.1234C>T hematopoietic cells (p>0.05, NS, one-tailed, Unpaired T test). All values represent mean ± SE. Abbreviations: CFU-Macrophage (M); CFU-Granulocyte-Macrophage (GM); committed erythroid BFU-E (BFU) and CFU-E (CFU) progenitors; multipotent progenitor cells CFU-GEMM (GEMM).
Fig 4.
GPI anchored proteins are required for neural differentiation.
(A). Representative example of images of hiPSC-derived EBs and EB-derived rosettes during neural differentiation. Neural induction and rosette formation upon neural induction was assessed in three cell lines using a serum-free EB generation method. On day 2 (left) of hiPSC-derived EBs from PIGAwt, PIGAc.1234C>T, and PIGAnull after forced aggregation (20X magnification, scale bar is 50μm). On day 4 (middle), single homogeneous hiPSC-EBs collected were pooled in a 10 cm plate (4X magnification, scale bar is 100μm). On day 11 (right), neuroepithelial cells appeared and neural tube-like rosettes formed (EB-derived rosettes) and scale bar is 50μm. (B). Neural induction rates from EB-derived rosettes. The percentage of EB derived rosettes was 88.8% ± 4.6, 75.5% ± 9.8 and 68.4% ± 6.9 for PIGAwt, PIGAc.1234C>T, and PIGAnull, respectively. PIGAwt versus PIGAc.1234C>T (p>0.05, NS) and PIGAwt versus PIGAnull (*p<0.05, one way ANOVA and Multiple comparisons). Neural induction from PIGAnull hiPSCs was less than 70%. All values were mean ±SD. (C). Representative confocal images showing expression of neuron stem cell marker SOX1 (in red) combined proliferation by EdU labeling in hNPCs derived from isolated neural rosettes. Nuclei were visualized with DAPI (blue) and scale bar 100μm. (D). Representative confocal images showing expression of neuron progenitor marker PAX6 (in red) and combined proliferation by EdU (in green) in hNPCs derived from isolated neural rosettes. Nuclei were visualized with DAPI (blue) and scale bar 200μm. hNPCs from PIGAnull cell lines showed reduced expression of SOX1 and PAX6. (E). Proliferation rate in hNPCs was assessed and plotted in all three cell lines. EdU positive cells were counted and normalized by total number of nuclei staining with DAPI (blue). Proliferation was significantly decreased in PIGAnull and PIGAc.1234C>T compared to PIGAwt. (F). Graphs depict the percentage of positive cells for SOX-1 (left) and Pax6 (right) in hNPCs derived from PIGAwt, PIGAc.1234C>T, and PIGAnull hiPSC lines. The hNPCs derived from the PIGAnull hiPSCs showed significantly decreased expression of SOX1 and PAX6. Similar levels of SOX1 and PAX6 were expressed in hNPCs from PIGAwt and PIGAc.1234C>T. All values represent mean ± SD.
Fig 5.
Characterization of human neural progenitor cell (hNPC) derived neurons during neuronal differentiation.
(A). Representative confocal scan images of MAP2 and GABA neurons. Four weeks after neuronal differentiation, hNPC-derived neurons from PIGAwt and PIGAc.1234C>T hNPCs were stained with anti-human MAP2 (microtubule-associated protein 2, red) and anti-human GABA (γ-amino butyric acid, green); nuclei were stained with DAPI (blue). Scale bars represent 50μm. Comparison of hNPC-derived neurons from PIGAwt (top) and PIGAc.1234C>T (bottom) showed dramatically reduced growth in the neurons from the PIGAc.1234C>T cell line (S2 Fig). (B). Representative confocal images of immunofluorescence staining for Synapsin (red) and VGLuT1 (vesicular glutamate transporter-1, green) in hNPC-derived neurons from PIGAwt and PIGAc.1234C>T cell lines. After 4 weeks of neuronal differentiation, neuron proliferation was markedly decreased in PIGAc.1234C>T. The density of Synapsin and VGLuT1 was substantially decreased in hNPC-derived neurons from PIGAc.1234C>T compared to the PIGAwt cell line. Scale bars represent 100 μm. (C). Quantification of VGLuT1 density in hNPC-derived neurons from PIGAwt and PIGAc.1234C>T. The number of VGLuT1 positive neurons was counted and plotted. Values represent the mean ± SD. (D). Representative confocal images showing the density of VGAT (vesicular GABA transporter, green) and Synapsin (red) in hNPC-derived neurons from PIGAwt and PIGAc.1234C>T cell lines. There was decreased density of VGAT and synapse formation in the hNPC-derived neurons from PIGAc.1234C>T cells compared to the control cell line (PIGAwt) (S3 Fig). Scale bars represent 20 μm. (E). Density of VGAT per 100 μm in PIGAwt and PIGAc.1234C>T derived neurons. Values represent the mean ± SD.
Fig 6.
Electrophysiological characterization of neurons derived from PIGAc.1234C>T hNPCs revealed a severe defect in neuronal activity.
Representative example of (A). Light microscopy images of hNPC derived neurons from PIGAwt (top), and PIGAc.1234C>T (bottom) hiPSC lines after 2 weeks of neuronal maturation (20X magnification, scale bar, 50μm). (B). Measurement of spontaneous neuronal activity by electrophysiology. Whole cell patch-clamp recording was performed on 2-week-old hiPSC derived neurons. PIGAwt neurons showed normal transit of inward sodium (Na+) currents and sustained outward potassium (K+) currents in response to membrane depolarization. Resting membrane potential was approximately -40 mV. PIGAc.1234C>T neurons showed normal Na+/K+ currents, though with reduced amplitude compared PIGAwt neurons. (C). Induction of action potential in hiPSC derived neurons by current injection. Multiple action potential peaks were observed in PIGAwt neurons. A single action potential peak with smaller amplitude was observed in PIGAc.1234C>T neurons. (D). Sample trace of excitatory spontaneous synaptic currents in PIGAwt neurons.
Fig 7.
Susceptibility of hNPCs to complement-mediated cytotoxicity.
The complement-mediated cytotoxicity assay was performed in PIGAwt, PIGAc.1234C>T, and PIGAnull hNPCs. The cells were incubated with normal serum, cobra venom factor or atypical HUS serum for 30 minutes. The percentage of non-viable cells was measured using WST-1 cell proliferation reagent. PIGAnull hNPCs were the most susceptible to complement-mediated cell killing, followed by PIGAc.1234C>T cells, which were more susceptible to complement-mediated killing compared to PIGAwt cells (p<0.05). Values were mean ± SD. NS (normal human serum, black), CVF (cobra venom factor, patterned), aHUS (serum from a patient with atypical hemolytic-uremic syndrome, white).