Abnormal morphology and function in retinal ganglion cells derived from patients-specific iPSCs generated from individuals with Leber's hereditary optic neuropathy

Abstract

Leber's hereditary optic neuropathy (LHON) is a maternally inherited eye disease that results from degeneration of retinal ganglion cells (RGC). Mitochondrial ND4 11778G > A mutation, which affects structural components of complex I, is the most prevalent LHON-associated mitochondrial DNA (mtDNA) mutation worldwide. The m.11778G > A mutation is the primary contributor underlying the development of LHON and X-linked PRICKLE3 allele (c.157C > T, p.Arg53Trp) linked to biogenesis of ATPase interacts with m.11778G > A mutation to cause LHON. However, the lack of appropriate cell and animal models of LHON has been significant obstacles for deep elucidation of disease pathophysiology, specifically the tissue-specific effects. Using RGC-like cells differentiated from induced pluripotent stem cells (iPSCs) from members of one Chinese family (asymptomatic subjects carrying only m.11778G > A mutation or PRICKLE3 p.Arg53Trp mutation, symptomatic individuals bearing both m.11778G > A and PRICKLE3 p.Arg53Trp mutations and control lacking these mutations), we demonstrated the deleterious effects of mitochondrial dysfunctions on the morphology and functions of RGCs. Notably, iPSCs bearing only m.11778G > A or p.Arg53Trp mutation exhibited mild defects in differentiation to RGC-like cells. The RGC-like cells carrying only m.11778G > A or p.Arg53Trp mutation displayed mild defects in RGC morphology, including the area of soma and numbers of neurites, electrophysiological properties, ATP contents and apoptosis. Strikingly, those RGC-like cells derived from symptomatic individuals harboring both m.11778G > A and p.Arg53Trp mutations displayed greater defects in the development, morphology and functions than those in cells bearing single mutation. These findings provide new insights into pathophysiology of LHON arising from RGC deficiencies caused by synergy between m.11778G > A and PRICKLE3 p.Arg53Trp mutation.

Graphical Abstract

Figure 1

Generation of patient-derived iPSCs from human dermal fibroblasts. (A) Portion of one Chinese pedigree family (2. Individuals harboring hemizygous (−/0), heterozygous (+/−) PRICKLE3 (c.157C > T, p.Arg53Trp) mutation and wild-type (+/+ or +/0) were indicated. Individuals related to this study were marked with^*. (B) Fundus photograph of five members in this family. Images were taken by the fundus photography (Cannon CR6-45NM fundus camera). (C) Schematic diagram illustrating that patient-derived iPSCs with various genotypes were generated from human dermal fibroblasts and then differentiated to neural progenitor cells or retinal ganglion cells. (D) Microscopy and alkaline phosphatase staining of iPSCs. The images above were the morphology of iPSCs clones on day 24 from the reprogramming start, surrounded by undifferentiated dermal fibroblasts, and the images below were the alkaline phosphatase staining of iPSCs clones cultured in a feeder-free condition. Scale bars = 100 μm.

Figure 2

Deficient differentiation of iPSCs into RGC-like cells. (A) iPSCs from different genotypes were induced toward optic neuronal progenitors, and then staining with neural progenitor markers PAX6 (green) and RAX (red). Nuclei were stained with DAPI. Scale bars = 50 μm. (B) The proportion of PAX6 positive cells among the DAPI-positive cells on day 14, n≧3. (C) The proportion of RAX positive cells among the DAPI-positive cells on day 14, n≧3. (D) Generations of retinal ganglion were confirmed by immunostaining with retinal ganglion cells markers: BRN3a (red) and β III TUBULIN (green). Nuclei were stained with DAPI. Scale bars = 50 μm. (E) The proportion of BRN3a-positive cells among the DAPI-positive cells. n≧3. Data are presented as mean ± standard error of mean (SEM), P indicates the significance, ^*P < 0.05, ^**P < 0.01, ^***P < 0.001; ns, no statistically significant by one-way ANOVA followed by Bonferroni’s post hoc test.

Figure 3

Abnormal morphology of RGC-like cells and axon elongation. (A) Classical morphology of early retinal organoid at day 21. Scale bars = 50 μm. (B) The axons and neural network formed by purified RGC-like cells in white phage. Scale bars = 100 μm. (C) The axons and neural network formed by purified RGC-like cells. Purified RGC-like cells were stained with β III TUBULIN (red). Scale bars = 100 μm. (D) Morphology of RGC-like cells. Cells were stained with β III TUBULIN on day 28. Scale bars = 20 μm. (E) Representative tracings of RGC-like cells by IMAGE J. (F) The soma size of RGC-like cells. n≧120. (G) The number of neurites per RGC-like cell. n≧120. (H) The total length of neurites per RGC-like cell. n≧120. (I) The average length of neurites per RGC-like cell. n≧120. Data are presented as mean ± standard error of mean (SEM), P indicates the significance, ^*P < 0.05, ^**P < 0.01, ^***P < 0.001; ns, no statistically significant by one-way ANOVA followed by Bonferroni’s post hoc test.

Figure 4

The electrophysiological properties of RGC-like cells. (A) Inward sodium ionic recording of solitary RGC-like cells in voltage-clamp mode. (B) RGC with recording microelectrode attached. (C) I-V curves displayed decreased inward current of RGC-like cells bearing mutations. n≧5. (D) Inward sodium current amplitudes of RGC-like cells. n≧5. (E) Patch-clamp recordings of evoke APs of RGC-like cells. (F) Evoke APs fired numbers of RGC-like cells. n≧5. (G) The amplitudes of evoke APs of RGC-like cells. n≧5. (H) The spontaneous APs fired by RGC-like cells. Data are presented as mean ± standard error of mean (SEM), P indicates the significance, ^*P < 0.05, ^**P < 0.01, ^***P < 0.001; ns, no statistically significant by one-way ANOVA followed by Bonferroni’s post hoc test.

Figure 5

The analysis of ATP contents and apoptosis. (A) Measurement of cellular and mitochondrial ATP levels using bioluminescence assay. RGC-like cells were incubated with 10 mM glucose or 5 mM 2-deoxy-D-glucose plus 5 mM pyruvate to determine ATP generation under mitochondrial ATP synthesis. Average rates of ATP level per cell line in mitochondria are shown. (B) Analysis of apoptosis. The distributions of Cytochrome c from RGC-like cells were visualized by immunofluorescent labeling with TOM20 antibody conjugated to Alex Fluor 488 (green) and Cytochrome c antibody conjugated to Alex Fluor 594 (red) analyzed by confocal microscopy. DAPI-stained nuclei were identified by their blue fluorescence. Scale bars = 20 μm. (C) Measurement of Caspase 3 activity in the presence and absence of toxic TNF-α and SM-164 stimuli. TNF-α and SM-164 are effective toxic stimulus for apoptosis (46,47). RGC-like cells from different genotypes were treated with TNF-α + SM-164 using Apoptosis Inducer Kit (Beyotime) to induce apoptosis. Caspase-3 activity was measured using the Caspase-3 Activity Assay Kit (Beyotime). Three independent experiments were made for each cell line. Data are presented as mean ± standard error of mean (SEM), P indicates the significance, ^*P < 0.05, ^**P < 0.01, ^***P < 0.001; ns, no statistically significant by one-way ANOVA followed by Bonferroni’s post hoc test.