Free
Biochemistry and Molecular Biology  |   May 2015
Patient-Specific Induced Pluripotent Stem Cell–Derived RPE Cells: Understanding the Pathogenesis of Retinopathy in Long-Chain 3-Hydroxyacyl-CoA Dehydrogenase Deficiency
Author Affiliations & Notes
  • Padmini P. Polinati
    Research Program of Molecular Neurology, Biomedicum 1, University of Helsinki, Helsinki, Finland
  • Tanja Ilmarinen
    Institute of Biomedical Technology, University of Tampere, Tampere, Finland
    BioMediTech, Tampere, Finland
  • Ras Trokovic
    Research Program of Molecular Neurology, Biomedicum 1, University of Helsinki, Helsinki, Finland
  • Tuulia Hyotylainen
    Valtion Teknillinen Tutkimuskeskus (VTT) Technical Research Centre of Finland, Tietotie 2, Espoo, Finland
  • Timo Otonkoski
    Research Program of Molecular Neurology, Biomedicum 1, University of Helsinki, Helsinki, Finland
    Children's Hospital, Helsinki University Central Hospital, Helsinki, Finland
  • Anu Suomalainen
    Research Program of Molecular Neurology, Biomedicum 1, University of Helsinki, Helsinki, Finland
    Children's Hospital, Helsinki University Central Hospital, Helsinki, Finland
  • Heli Skottman
    Institute of Biomedical Technology, University of Tampere, Tampere, Finland
    BioMediTech, Tampere, Finland
  • Tiina Tyni
    Research Program of Molecular Neurology, Biomedicum 1, University of Helsinki, Helsinki, Finland
    Children's Hospital, Helsinki University Central Hospital, Helsinki, Finland
  • Correspondence: Tiina Tyni, Research Program of Molecular Neurology, Biomedicum 1, University of Helsinki, FI-00290, Helsinki, Finland; tiina.tyni@hus.fi
Investigative Ophthalmology & Visual Science May 2015, Vol.56, 3371-3382. doi:10.1167/iovs.14-14007
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to authenticated users only.
      Sign In or Create an Account ×
    • Get Citation

      Padmini P. Polinati, Tanja Ilmarinen, Ras Trokovic, Tuulia Hyotylainen, Timo Otonkoski, Anu Suomalainen, Heli Skottman, Tiina Tyni; Patient-Specific Induced Pluripotent Stem Cell–Derived RPE Cells: Understanding the Pathogenesis of Retinopathy in Long-Chain 3-Hydroxyacyl-CoA Dehydrogenase Deficiency. Invest. Ophthalmol. Vis. Sci. 2015;56(5):3371-3382. doi: 10.1167/iovs.14-14007.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose.: Retinopathy is an important manifestation of trifunctional protein (TFP) deficiencies but not of other defects of fatty acid oxidation. The common homozygous mutation in the TFP α-subunit gene HADHA (hydroxyacyl-CoA dehydrogenase), c.1528G>C, affects the long-chain 3-hydroxyacyl-CoA dehydrogenase (LCHAD) activity of TFP and blindness in infancy. The pathogenesis of the retinopathy is unknown. This study aimed to utilize human induced pluripotent stem cell (hiPSC) technology to create a disease model for the disorder, and to derive clues for retinopathy pathogenesis.

Methods.: We implemented hiPSC technology to generate LCHAD deficiency (LCHADD) patient-specific retinal pigment epithelial (RPE) monolayers. These patient and control RPEs were extensively characterized for function and structure, as well as for lipid composition by mass spectrometry.

Results.: The hiPSC-derived RPE monolayers of patients and controls were functional, as they both were able to phagocytose the photoreceptor outer segments in vitro. Interestingly, the patient RPEs had intense cytoplasmic neutral lipid accumulation, and lipidomic analysis revealed an increased triglyceride accumulation. Further, patient RPEs were small and irregular in shape, and their tight junctions were disorganized. Their ultratructure showed decreased pigmentation, few melanosomes, and more melanolysosomes.

Conclusions.: We demonstrate that the RPE cell model reveals novel early pathogenic changes in LCHADD retinopathy, with robust lipid accumulation, inefficient pigmentation that is evident soon after differentiation, and a defect in forming tight junctions inducing apoptosis. We propose that LCHADD-RPEs are an important model for mitochondrial TFP retinopathy, and that their early pathogenic changes contribute to infantile blindness of LCHADD.

Mitochondrial fatty acid β-oxidation (FAO) is a major adenosine triphosphate (ATP)-providing pathway for tissues such as skeletal muscle, heart, and liver, especially at the time of increased energy expenditure. In mitochondrial FAO, the trifunctional protein (TFP) is a multienzyme complex composed of four molecules of the α-subunit containing the enoyl-CoA hydratase and 3-LCHAD domains and four molecules of the β-subunit containing the 3-ketoacyl-CoA thiolase domain (Fig. 1)1 with two genes, HADHA and HADHB, encoding the subunits.1 The homozygous mutation c.1528G>C, in HADHA (p.Glu510Gln), leads mostly to deficiency of long-chain 3-hydroxyl-CoA dehydrogenase (LCHAD) activity of TFP, and is the most common cause of LCHAD deficiency (LCHADD; Online Mendelian Inheritance in Man [OMIM] 609016). The carrier frequency of this mutation in Finland is 1:240.1 Disorders affecting this pathway manifest as hypoketotic hypoglycemia, hepatic steatosis, cardiomyopathy, and rhabdomyolysis.1 Most of the β-oxidation disorders, including LCHADD, can be treated with a low-fat diet. However, progressive pigment retinopathy is a typical manifestation of LCHAD and TFP deficiencies, but not common in other FAO defects. In several countries, β-oxidation disorders are included in the neonatal screening programs.2 Without early diagnosis and low-fat diet treatment, retinopathy leads to chorioretinal atrophy of the central fundus, high myopia, posterior staphyloma, and poor vision3 (Fig. 2A). In our recent study,4 patient fundus photographs from stage 2 pigment retinopathy were classified based on the severity of the fundus changes, that is, pigment deposits (P1–P3) and retinal pigment epithelium (RPE) atrophy (A1–A3). Figure 2A shows a 5-year-old patient in whom the stage of retinopathy was P3 and A3 (Fig. 2A; P3 and A3). 
Figure 1
 
Schematic illustration of mitochondrial β-oxidation pathway.
Figure 1
 
Schematic illustration of mitochondrial β-oxidation pathway.
Figure 2
 
Characterization of LCHADD patient samples. (A) Fundus photographs of LCHADD patients showing different stages of pigment retinopathy upper left, patient 1 [early stage], upper right, patient 2 [late stage], lower left and right, patient 3 [stage 2 - grades of pigment deposits (P3) and RPE atrophy (A3) of LCHADD]. Reprinted with permission from Tyni T, Immonen T, Lindahl P, Majander A, Kivelä T. Refined staging for chorioretinopathy in long-chain 3-hydroxyacyl coenzyme A dehydrogenase deficiency. Ophthalmic Res. 2012;48:75–81. Copyright 2012 S. Karger AG, Basel. (B) Sequencing HADHA homozygous c.1528G>C: Control fibroblasts (FiPSC 5–7), homozygous c.G>C substitution in LCHADD fibroblasts, hiPSC from LCHADD fibroblasts (clone 12.22), and hiPSC-differentiated RPE cells (clone 12.22) showed the same substitution. Ctrl, control; Pat, patient.
Figure 2
 
Characterization of LCHADD patient samples. (A) Fundus photographs of LCHADD patients showing different stages of pigment retinopathy upper left, patient 1 [early stage], upper right, patient 2 [late stage], lower left and right, patient 3 [stage 2 - grades of pigment deposits (P3) and RPE atrophy (A3) of LCHADD]. Reprinted with permission from Tyni T, Immonen T, Lindahl P, Majander A, Kivelä T. Refined staging for chorioretinopathy in long-chain 3-hydroxyacyl coenzyme A dehydrogenase deficiency. Ophthalmic Res. 2012;48:75–81. Copyright 2012 S. Karger AG, Basel. (B) Sequencing HADHA homozygous c.1528G>C: Control fibroblasts (FiPSC 5–7), homozygous c.G>C substitution in LCHADD fibroblasts, hiPSC from LCHADD fibroblasts (clone 12.22), and hiPSC-differentiated RPE cells (clone 12.22) showed the same substitution. Ctrl, control; Pat, patient.
The molecular pathogenesis of pigment retinopathy remains unclear; however, the hypothesized mechanism leading to this pigment retinopathy appears to be degeneration of the RPE, which secondarily disturbs either the function or the maintenance of the neural retina and the underlying choriocapillaris.5 The question whether locally malfunctioning LCHAD or peripheric production and retinal accumulation of toxic intermediates underlie the pathogenesis has remained open. Recently, LCHADD pigment retinopathy was hypothesized to be caused by toxicity of accumulating fatty acid intermediates.6 Long-chain acyl-CoA esters accumulating in LCHADD were shown to inhibit oxidative phosphorylation7 by shifting pyruvate metabolism to formation of lactate, thereby leading to decreased mitochondrial ATP/adenosine diphosphate (ADP) ratio and function of the dicarboxylic carrier protein.8 The expression of β-oxidation proteins in the retina suggests their importance for retinal metabolism.9,10 Furthermore, we have shown that FAO enzymes, including carnitine palmitoyl transferase (CPT)1A, CPT1C, and TFP, are expressed in human RPE and in several layers of the neural retina.10 Cultured porcine RPE cells are able to metabolize palmitate and express LCHAD and long-chain 3-ketoacyl-CoA thiolase (LCKT) activities.10 This evidence suggests that the molecular pathogenesis of LCHADD pigment retinopathy may involve local mechanisms. 
Retinal pigment epithelium is a pigmented monolayer of cells responsible for the metabolic support of retina. Melanosomes are the most prominent pigment granules in RPE that tend to act against reactive oxygen species and protect the neural retina.11 During aging, RPE pigmentation decreases, and a new type of pigmented granule, melanolysosome, increases in number, as a sign of melanin degradation.11 
We generated several human induced pluripotent stem cell (hiPSC) clones from skin fibroblasts of an LCHADD patient and differentiated those into RPE cells in vitro, thus generating a human RPE cell layer containing the homozygous common HADHA mutation p.Glu510Gln. We report here LCHADD-specific early pathogenic changes in hiPSC-derived RPEs, involving robust changes in lipid composition and pigmentation as well as cell-to-cell contacts. Furthermore, we suggest that this novel in vitro tool is valuable in elucidating the pathogenesis of LCHADD retinopathy and potentially can be used to evaluate new strategies for preventing the disease progression. 
Materials and Methods
Reprogramming and Maintenance of hiPSC Lines
The Ethics Committee of Helsinki University and Helsinki University Central Hospital District approved the generation of hiPSC lines from control and LCHADD patient fibroblasts (Nro 423/13/03/00/08) and the research adhered to the tenets of the Declaration of Helsinki. Induced PSCs were generated by retroviruses (Oct4, Sox2, Klf4, cMyc) and further propagated as described previously.12 Controls used were previously established hiPSC lines (A116,13 FiPS5-714) generated from fibroblasts without LCHADD. 
Differentiation, Enrichment, and Culture of RPE Cells
Three selected patient hiPSC clones carrying the homozygous HADHA p.Glu510Gln change (12.1, 12.22, 12.4), as well as control hiPSC lines (FiPS5-7, A116) confirmed not to have HADHA changes, were adapted on human foreskin feeder cell layers for at least eight passages prior to the differentiation toward RPE cells. Retinal pigment epithelium differentiation was performed as previously described.15 Briefly, pluripotent stem cell colonies were picked manually and plated onto Nunc Low Cell binding six-well plates (Nalgene NUNC, Tokyo, Japan), in which the colonies formed floating aggregates. At this point, knockout serum replacement (KO-SR) concentration was reduced from 20% to 15%, and bFGF (basic fibroblast growth factor) was removed from the hiPSC culture medium to initiate spontaneous RPE differentiation. This modified culture medium (RPE medium) was changed thrice a week. After 65 to 173 days of differentiation, the pigmented areas were manually isolated from the floating aggregates. Thereafter the cells were dissociated with 1× trypsin-EDTA (Lonza, Walkersville, MD, USA) and seeded on collagen IV (5 μg/cm2) (Sigma-Aldrich Corp., St. Louis, MO, USA)-coated 12-well plates (Nalge NUNC), permeable collagen IV-coated BD Biocoat culture inserts (BD Bioscience, Franklin Lakes, NJ, USA), or ThinCert culture inserts (Greiner Bio-One North America, Inc., Monroe, NC, USA) for RPE enrichment and further analyses. 
Genomic DNA Extraction and Sequencing
Genomic DNA was isolated from control and patient fibroblasts, patient hiPSC clones, and patient hiPSC-RPE using the Flexi gene DNA kit from Qiagen (Hilden, Germany). The DNA concentration was determined by the absorbance at 260 nm. Total genomic DNA (100 ng) was used for amplifying exon 15 of the human HADHA gene. Polymerase chain reaction products were purified using a PCR purification kit (ExoSAP-IT; Affymetrix, Santa Clara, CA, USA) and sequenced using the ABI BigDye terminator sequencing kit (ABI, Foster City, CA, USA). 
Analysis of RPE Cell Surface Area
The perimeter of each individual cell (10 cells per image) with a clearly defined border was outlined, and cell surface area was calculated using ImageJ software (National Institutes of Health, Bethesda, MD, USA). 
Immunocytochemistry
Cell fixation and antibody hybridization were performed as described earlier.15 The primary antibodies used and their dilutions are listed in the Table. Donkey anti-mouse IgG, goat anti-rabbit IgG, and chicken anti-goat IgG (all from Alexa Fluor 488), as well as goat anti-mouse IgG, goat anti-rabbit IgG, and donkey anti-sheep IgG (all from Alexa Fluor 568) (Molecular Probes, Life Technologies, Paisley, UK), diluted 1:1500, were used as secondary antibodies. Images were taken with a Zeiss Axioplan 2 microscope (Zeiss, Jena, Germany). 
Table
 
List of Primary Antibodies
Table
 
List of Primary Antibodies
Reverse Transcription–Polymerase Chain Reaction
The RT-PCR analysis was performed from control and patient hiPSC-RPE as described earlier.15 Gene-specific primer sequences have been published previously.13,15 Polymerase chain reaction products were analyzed on 2% agarose gels with a 100-bp DNA ladder and visualized with 4.5.2 Basic Program (Bio-Rad Laboratories, Inc., Hercules, CA, USA). 
In Vitro Phagocytosis Assay
Photoreceptor outer segments (POS) were isolated from freshly slaughtered porcine eyes using a continuous sucrose gradient as previously described.15 Photoreceptor outer segments were labeled with fluorescein isothiocyanate (0.04 μg/μL; Sigma-Aldrich) in 0.1 M NaHCO3 (pH 9) for 1 hour at room temperature (RT). Labeled POS were washed, resuspended in RPE culture medium, and seeded onto collagen IV–coated 0.3-cm2 BD Biocoat culture plate inserts (BD Bioscience). Cells were incubated at +37°C in 5% CO2 for 16 hours. External fluorescence was removed by trypan blue treatment for 10 minutes. Cells were then washed with PBS and used for immunocytochemistry as described earlier.15 Filamentous actin was stained with 1:40 dilution with PBS (Invitrogen, Carlsbad, CA, USA) by incubating for 10 minutes at RT following several washes with PBS. The images were captured using a confocal microscope (TCS SP2; Leica, Wetzlar, Germany). 
Transmission Electron Microscopy
Retinal pigment epithelium cell medium was removed from the collagen-coated inserts and the cell layer was cut into pieces, rinsed with PBS, and fixed in 2.5% glutaraldehyde in 0.1 M HEPES for 2 hours at RT and then transferred to 4% paraformaldehyde for 3 days at 4°C. Then the cells were postfixed with 1% osmium tetroxide, dehydrated, and embedded in Epon. Images were captured with a Jeol 1200-EXII and Jeol 1400 electron microscope (Tokyo, Japan). 
Neutral and Phospholipid Staining
Neutral lipid and phospholipid (PL) stainings were performed using HCS LipidTOX (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions. Pictures were taken using a Zeiss Axioplan 2 microscope. 
Lipidomics
Lipidomics was performed as described earlier,16 and the internal standard mixture comprised labeled species. For sample preparation, cell pellets from control and patient hiPSC-RPE were mixed with 170 μL 0.90% NaCl solution and sonicated for 3 minutes at 7°C, 40 kHz (Finnsonic m03; FinnSonic Oy, Lahti, Finland). For the ultraperformance liquid chromatography–quadrupol time-of-flight mass spectrometry analyses, a standard mixture (10 μL) containing phosphatidyl choline (PC) (17:0/0:0), PC (17:0/17:0), phosphatidylethanolamine (PE) (17:0/17:0), and ceramide (d18:1/17:0) (Avanti Polar Lipids, Inc., Alabaster, AL, USA) and triglyceride (TG) (17:0/17:0/17:0) (Larodan Fine Chemicals, Limhamn, Sweden) was added to 15 μL cell homogenates. High-performance liquid chromatogaphy–grade chloroform and methanol (2:1; 100 μL) were added to the samples, vortexed for 2 minutes, and allowed to stand for 30 minutes. Subsequently, samples were centrifuged, the lower phase (60 μL) was collected, and 20 μL internal standard mixture 2 was added. 
With respect to instrumental conditions, the extracts were analyzed on a Waters Q-Tof Premier mass spectrometer (Milford, MA, USA) combined with Acquity Ultra Performance LC (Chicago, IL, USA) as described previously.16 Data processing using MZmine2 software (Espoo, Finland) was performed as previously described.16 Finally, the data were further normalized by dividing the normalized lipid concentration into the protein content of the sample. 
Statistical Analysis
The differences between samples were compared using Student's t-test. Data were expressed as mean ± standard error. Values of P < 0.05 were considered significant. 
Results
Reprogramming of hiPSC Lines From LCHADD Fibroblasts and Differentiation Into RPE Cells
Human skin fibroblasts were cultured from an LCHADD patient with the common homozygous c.1528G>C (p.Glu510Gln) HADHA mutation, and reprogrammed to generate patient-specific iPSCs. Three different hiPSC clones (12.1, 12.22, and 12.4) were selected for further studies. All three clones were cultured as individual hiPSC lines, and all stained positive for pluripotent cell surface markers typical of hiPSCs (TRA1-60 and SSEA3 and OCT4 with nuclear localization) (Supplementary Fig. S1A). All hiPSC clones were cultured for at least 10 passages and exhibited growth properties and morphology similar to those of human embryonic stem cell (hESC) and control hiPSC (A116,13 FiPS5-714) lines. The homozygous common HADHA mutation c.1528G>C/p.Glu510Gln was confirmed to be present in the LCHADD patient fibroblasts, the patient hiPSC clones, and differentiated RPEs (Fig. 2B), but it was not present in controls. 
Retinal pigment epithelial cells differentiated from both patient and control hiPSCs showed the typical cobblestone-like structure reported previously.13,15,17 However, the patient RPEs were smaller in size, with a more irregular shape and less pigmentation than in the controls (Figs. 3A, 3B). The area of a single RPE cell from a patient was 2.5-fold smaller (P = 0.0365) than those of controls (Fig. 3C). 
Figure 3
 
Morphology of the hiPSC clones. (A) Differentiation of RPE: morphology of control (FiPSC5-7) and patient (clone 12.22) hiPSC-RPE. Magnification: ×20. Scale bars: 20 μm. (B) Control (A116) and patient hiPSC-RPE (clone 12.22) lines showing differences in the pigmentation after 54 days of differentiation. (C) Bar graph showing quantification of cell surface area. Data from control hiPSC-RPE (FiPSC 5–7, A116) and patient hiPSC-RPE (clones 12.1, 12.4, and 12.22) are the average of 100 RPE cells each. Bars represent average with SEM as error bars (*P < 0.05) [P = 0.0365]. Ctrl, control; Pat, patient.
Figure 3
 
Morphology of the hiPSC clones. (A) Differentiation of RPE: morphology of control (FiPSC5-7) and patient (clone 12.22) hiPSC-RPE. Magnification: ×20. Scale bars: 20 μm. (B) Control (A116) and patient hiPSC-RPE (clone 12.22) lines showing differences in the pigmentation after 54 days of differentiation. (C) Bar graph showing quantification of cell surface area. Data from control hiPSC-RPE (FiPSC 5–7, A116) and patient hiPSC-RPE (clones 12.1, 12.4, and 12.22) are the average of 100 RPE cells each. Bars represent average with SEM as error bars (*P < 0.05) [P = 0.0365]. Ctrl, control; Pat, patient.
Characterization of the Gene and Protein Expression of hiPSC-RPE Cells
After 170 days of RPE differentiation, the cells originating from patient hiPSC and controls expressed genes typical and specific for functional RPEs (Figs. 4A–E). The expressed genes includes BEST1 (bestrophin 1), MITF (microphthalmia-associated transcription factor), RPE65 (RPE-specific protein 65 kDa), OTX2v1 (orthodenticle homeobox 2 variant 1), TYR (tyrosinase), PMEL (premelanosome protein), and PEDF (pigment epithelium–derived factor). As the patient RPE cell morphology was disorganized, the cells did not adhere to each other and the cell junctions appeared irregular. We analyzed the expression of cellular retinaldehyde-binding protein (CRALBP), Na+/K+ ATPase, RPE65, and zona occludens (ZO-1). Expression of CRALBP was detected in the cytoplasm and cell membranes in both the control and the patient. In controls, Na+/K+ ATPase staining localized to the cell membranes, as did that of the patient cells, but the staining was weak in the patient hiPSC-RPEs. RPE65 labeling was detected throughout the cytoplasm and in the cell membranes of both control and patient hiPSC-RPEs. The ZO-1 localization in the patient hiPSC-RPEs was disorganized and showed an irregular staining pattern and further revealed a discontinuous staining pattern of the tight junctions, whereas the control cells showed a regular staining pattern and revealed a clearly discerned labeling in the cell membrane tight junctions. 
Figure 4
 
Immunocytochemistry of RPE markers. (AD) Immunocytochemistry demonstrating the expression of RPE markers in control (FiPSC 5-7) and patient (clones 12.1, 12.22) hiPSC-RPE cells. (A) CRALBP in the cytoplasm. (B) Na+/K+ ATPase localized to cell membrane. (C) RPE65 in the cytoplasm. (D) ZO-1, tight junction protein expression in the membrane. Magnification: ×63. Scale bars: 10 μm and 5 μm in inset (AD). (E) Agarose gel electrophoresis of RT-PCR products for RPE differentiation markers, control (FiPSC 5–7) and patient hiPSC-RPE (clone 12.1).
Figure 4
 
Immunocytochemistry of RPE markers. (AD) Immunocytochemistry demonstrating the expression of RPE markers in control (FiPSC 5-7) and patient (clones 12.1, 12.22) hiPSC-RPE cells. (A) CRALBP in the cytoplasm. (B) Na+/K+ ATPase localized to cell membrane. (C) RPE65 in the cytoplasm. (D) ZO-1, tight junction protein expression in the membrane. Magnification: ×63. Scale bars: 10 μm and 5 μm in inset (AD). (E) Agarose gel electrophoresis of RT-PCR products for RPE differentiation markers, control (FiPSC 5–7) and patient hiPSC-RPE (clone 12.1).
In Vitro Functional Characterization of hiPSC-RPE Cells
Phagocytosis activity is a crucial function for RPE cells; hence a phagocytosis assay was performed to confirm the in vitro functionality of the patient and control hiPSC-RPEs. Confocal microscopy showed that both the patient and control RPEs were able to internalize POS isolated from porcine eyes (Fig. 5), indicating that they work as functional RPEs. 
Figure 5
 
Phagocytosis assay performed in control (FiPSC 5–7) and patient (clone 12.4). Human-induced pluripotent stem cell-RPE internalizes porcine POS (green). For cell morphology, F-actins were stained using phalloidin (red). View through the depth of image stacks at the position representing red and green lines. Images were taken with a Leica TCS SP2 confocal microscope. Magnification: ×63.
Figure 5
 
Phagocytosis assay performed in control (FiPSC 5–7) and patient (clone 12.4). Human-induced pluripotent stem cell-RPE internalizes porcine POS (green). For cell morphology, F-actins were stained using phalloidin (red). View through the depth of image stacks at the position representing red and green lines. Images were taken with a Leica TCS SP2 confocal microscope. Magnification: ×63.
Expression of Mitochondrial FAO Enzymes
Immunocytochemical analysis revealed that mitochondrial FAO enzymes, medium-chain acyl-CoA dehydrogenase (ACADM), acyl-CoA dehydrogenase 9 (ACAD9), very long chain acyl-CoA dehydrogenase (ACADVL), TFP, and CPT1A were expressed in both control and patient hiPSC-RPEs (Figs. 6A–E). The patient cells showed more intense staining, which was likely due to a blocking of the fluorescence by pigment granules in control RPEs and less pigmentation in the patient cells. 
Figure 6
 
(AE) Immunocytochemistry staining of mitochondrial β-oxidation enzymes, control (FiPSC 5–7) and patient (clones 12.4 and 12.22) hiPSC-RPE cells. (A) ACADM in the cytoplasm. (B) ACAD9 in both membranes and cytoplasm. (C) ACADVL in both membranes and cytoplasm. (D) TFP in both membrane and cytoplasm. (E) CPT1A in the cytoplasm. Scale bars: 10 μm and 5 μm in inset (AE).
Figure 6
 
(AE) Immunocytochemistry staining of mitochondrial β-oxidation enzymes, control (FiPSC 5–7) and patient (clones 12.4 and 12.22) hiPSC-RPE cells. (A) ACADM in the cytoplasm. (B) ACAD9 in both membranes and cytoplasm. (C) ACADVL in both membranes and cytoplasm. (D) TFP in both membrane and cytoplasm. (E) CPT1A in the cytoplasm. Scale bars: 10 μm and 5 μm in inset (AE).
Transmission Electron Microscopy
Ultrastructural study revealed that both the control and patient hiPSC-RPEs had basal nuclei and apical microvilli, typical of RPE morphology (Fig. 7E), further supporting that LCHADD did not compromise differentiation. The control RPEs revealed plenty of melanosomes with a uniform structure and small lipid droplets, whereas the patient cells had several larger lipid droplets, a few melanosomes, and several melanolysosomes (Figs. 7A–E). These results suggest degradation of melanosomes, a finding typically associated with aging-related retinal changes. 
Figure 7
 
Representative electron microscope images of control (FiPSC 5–7, A116) and patient (clone 12.4, 12.22) hiPSC-RPE cells. (A) Several melanolysosomes (arrow) were detected in patient compared to control. (B) Fewer melanosomes were detected in patient cells (asterisk) than in control. (C) Lipid accumulation observed in patient (arrow) compared to control. (D) Tight junctions in control and patient (arrowheads). (E) Basal nuclei (arrow) and apical microvilli (Mv) in control and patient.
Figure 7
 
Representative electron microscope images of control (FiPSC 5–7, A116) and patient (clone 12.4, 12.22) hiPSC-RPE cells. (A) Several melanolysosomes (arrow) were detected in patient compared to control. (B) Fewer melanosomes were detected in patient cells (asterisk) than in control. (C) Lipid accumulation observed in patient (arrow) compared to control. (D) Tight junctions in control and patient (arrowheads). (E) Basal nuclei (arrow) and apical microvilli (Mv) in control and patient.
Neutral Lipid and Phospholipid Staining and Lipidomics
As toxic accumulation of lipids has been suggested as a pathogenic mechanism for LCHADD retinopathy, we analyzed the RPE lipid content. Immunofluorescence labeling with LipidTOX revealed intense accumulation of cytoplasmic neutral lipids in the patient hiPSC-RPEs, as well as increased PL signals (Fig. 8A). Further lipidomic profiling by liquid chromatography–mass spectometry allowed identification of 276 lipid species out of a total of 1000, including PL, ceramides, sphingomyelins, diacylglycerols, and TG. A significant (Fig. 8B) increase in TG levels was observed in the patient hiPSC-RPEs, in particular the long-chain TGs [TG(14:0/18:1/18:1)+TG(16:0/16:1/18:1), TG(16:0/18:0/18:0), TG(16:0/18:1/18:1), TG(16:0/18:1/20:1)+TG(18:0/18:1/18:1), TG(18:1/18:1/18:1), and TG(18:1/18:2/18:1)] (Fig. 8B). Other lipid classes, such as PC, PE, and ceramides, showed no remarkable differences. These results indicate that patient hiPSC-RPEs replicate the clinical findings in LCHADD pigment retinopathy, showing excessive accumulation of neutral lipids, especially long-chain TGs. 
Figure 8
 
Lipid analysis control (FiPSC 5–7, A116) and patient (clone 12.4, 12.22) hiPSC-RPE cells. (A) Neutral lipid staining (×20) shows a highly dense accumulation of lipids in patient but not in control hiPSC-RPE. Scale bars: 10 μm. (B) Lipidomics data revealed upregulation of TGs in patient hiPSC-RPE. Significant upregulation of polyunsaturated TGs in patient. Controls and patient data are average of two independent experiments using two different cell batches with triplicates in each. Bars represent average with SEM as error bars. *P < 0.05, **P < 0.01 when compared to controls.
Figure 8
 
Lipid analysis control (FiPSC 5–7, A116) and patient (clone 12.4, 12.22) hiPSC-RPE cells. (A) Neutral lipid staining (×20) shows a highly dense accumulation of lipids in patient but not in control hiPSC-RPE. Scale bars: 10 μm. (B) Lipidomics data revealed upregulation of TGs in patient hiPSC-RPE. Significant upregulation of polyunsaturated TGs in patient. Controls and patient data are average of two independent experiments using two different cell batches with triplicates in each. Bars represent average with SEM as error bars. *P < 0.05, **P < 0.01 when compared to controls.
Discussion
We report here a hiPSC-derived RPE cell model for LCHADD pigment retinopathy. To our knowledge, this is the first in vitro cell model for LCHADD pigment retinopathy. The patient hiPSC clones expressed pluripotent stem cell markers and differentiated toward RPE with efficiency similar to that of the controls, indicating that LCHADD did not affect reprogramming of pluripotent state or RPE differentiation per se. However, the patient hiPSC-RPEs revealed several interesting changes, including small cell size, decreased pigmentation, few melanosomes, increased number of melanolysosomes, accumulation of TGs, and disorganized cell-to-cell contacts (ZO-1), all of which have potential to significantly contribute to disease progression and eventually to blindness. 
The decreased cell pigmentation phenotype in LCHADD-RPEs very well mimics the clinical outcome of LCHADD pigment retinopathy. These patients develop diffuse macular hypopigmentation, which may develop within 2 years, leading to pigment clumping at the level of RPE and resulting in the progressive loss of RPE.2,18 Melanin in RPE cells exhibits several potentially antioxidant properties, including sequestration of redox-active metal ions and scavenging of free radicals.19 Melanosome biogenesis is completed prior to birth, and these pigment granules may be retained throughout life. They are the most prominent pigment granules in the young RPE, decreasing during aging.19 Depigmentation of RPE cells is known to occur in normal aging people and in a number of degenerative retinopathies,20 associating itself with degradation of melanosomes that fuse within lysosomes.21 Our control hiPSC-RPE cells had regular melanosomes, whereas the patient hiPSC-RPE cells revealed few melanosomes and an increased number of melanolysosomes. These findings suggest that depigmentation and melanosome degradation are early changes in the pathologic development of LCHADD pigment retinopathy. 
The robust expression of mitochondrial FAO enzymes in both hiPSC-RPE cell lines correlated well with the previous immunohistochemical evidence11,22 suggesting that mitochondrial FAO enzymes are important for RPE and neural retinal metabolism. Furthermore, excessive lipid load, especially long-chain TGs, indicates that LCHAD activity is necessary for RPE lipid processing. Nonadipose cells such as RPEs have a limited capacity for lipid storage. Excess free fatty acids can impair normal cell signaling, are prone to peroxidation, and can cause cellular dysfunction and apoptotic cell death.23 
In addition to lipid loading, the patient hiPSC-RPEs showed a rounded shape, instead of the hexagons of the control RPEs, and had deformed cell-to-cell adhesion and tight junctions. Despite these changes, these RPEs were able to internalize POS by phagocytosis, indicating their functionality. As the RPE monolayer forms the outer blood–retina barrier, the integrity of the tight junctions is vital.10 These findings, as well as our previous histopathological findings on LCHADD pigment retinopathy, suggest that disruption of barrier functions may lead to degeneration of the RPE layer in LCHADD.5 
In conclusion, reprogramming of somatic patient cells to hiPSC and further differentiation to RPEs provides a powerful model to study early pathogenic changes of LCHADD pigment retinopathy and retinopathies with RPE dysfunction such as macular degeneration, causing blindness in developed countries and affecting more than 14 million people.23 Our LCHADD pigment retinopathy model revealed a gross disruption of the RPE cell morphology with less pigmentation, few melanosomes, several melanolysosomes, loose cell-to-cell attachments, and an excessive accumulation of TGs early upon differentiation, leading to reduced oxidant scavenging potential, lipid toxicity, and loose blood–retina barrier. This model aids in clarifying the molecular pathology of LCHADD pigment retinopathy. 
Acknowledgments
We thank the patients and their families for donating fibroblasts. Thanks to Biomedicum Molecular Imaging Unit and Electron Microscopy Unit Viikki, Helsinki, for microscopy services. The authors are grateful to Outi Heikkilä, Hanna Koskenaho, and Outi Melin for technical assistance. Eva Roomets is thanked for helping with clinical discussions. Ophthalamic Research is thanked for reusing eye images. 
Supported by Jane and Aatos Erkko Foundation, Helsinki, Finland; Pediatric Research Foundation, Helsinki, Finland; The Mary and Georg Ehrnooth Foundation, Helsinki, Finland; and Academy of Finland (Grants 130497 [HS and TI] and 218050 [AS]) and Sigrid Juselius Foundation and European Research Council (AS). 
Disclosure: P.P. Polinati, None; T. Ilmarinen, None; R. Trokovic, None; T. Hyotylainen, None; T. Otonkoski, None; A. Suomalainen, None; H. Skottman, None; T. Tyni, None 
References
Tyni T, Pihko H. Long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency. Acta Paediatr. 1999; 88: 237–245.
Gillingham M, Van Calcar S, Ney D, Wolff J, Harding C. Dietary management of long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency (LCHADD). A case report and survey. J Inherit Metab Dis. 1999; 22: 123–131.
Tyni T, Kivela T, Lappi M, Summanen P, Nikoskelainen E, Pihko H. Ophthalmologic findings in long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency caused by the G1528C mutation: a new type of hereditary metabolic chorioretinopathy. Ophthalmology. 1998; 105: 810–824.
Tyni T, Immonen T, Lindahl P, Majander A, Kivela T. Refined staging for chorioretinopathy in long-chain 3-hydroxyacyl coenzyme A dehydrogenase deficiency. Ophthalmic Res. 2012; 48: 75–81.
Tyni T, Pihko H, Kivela T. Ophthalmic pathology in long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency caused by the G1528C mutation. Curr Eye Res. 1998; 17: 551–559.
Fletcher AL, Pennesi ME, Harding CO, Weleber RG, Gillingham MB. Observations regarding retinopathy in mitochondrial trifunctional protein deficiencies. Mol Genet Metab. 2012; 106: 18–24.
Ventura FV, Ruiter J, Ijlst L, de Almeida IT, Wanders RJ. Differential inhibitory effect of long-chain acyl-CoA esters on succinate and glutamate transport into rat liver mitochondria and its possible implications for long-chain fatty acid oxidation defects. Mol Genet Metab. 2005; 86: 344–352.
Ventura FV, Tavares de Almeida I, Wanders RJ. Inhibition of adenine nucleotide transport in rat liver mitochondria by long-chain acyl-coenzyme A beta-oxidation intermediates. Biochem Biophys Res Commun. 2007; 352: 873–878.
Tyni T, Johnson M, Eaton S, Pourfarzam M, Andrews R, Turnbull DM. Mitochondrial fatty acid beta-oxidation in the retinal pigment epithelium. Pediatr Res. 2002; 52: 595–600.
Tyni T, Paetau A, Strauss AW, Middleton B, Kivela T. Mitochondrial fatty acid beta-oxidation in the human eye and brain: implications for the retinopathy of long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency. Pediatr Res. 2004; 56: 744–750.
Strauss O. The retinal pigment epithelium in visual function. Physiol Rev. 2005; 85: 845–881.
Trokovic R, Weltner J, Manninen T et al. Small molecule inhibitors promote efficient generation of induced pluripotent stem cells from human skeletal myoblasts. Stem Cells Dev. 2013; 22: 114–123.
Toivonen S, Ojala M, Hyysalo A, et al. Comparative analysis of targeted differentiation of human induced pluripotent stem cells (hiPSCs) and human embryonic stem cells reveals variability associated with incomplete transgene silencing in retrovirally derived hiPSC lines. Stem Cells Transl Med. 2013; 2: 83–93.
Hussein SM, Batada NN, Vuoristo S et al. Copy number variation and selection during reprogramming to pluripotency. Nature. 2011; 471: 58–62.
Vaajasaari H, Ilmarinen T, Juuti-Uusitalo K, et al. Toward the defined and xeno-free differentiation of functional human pluripotent stem cell-derived retinal pigment epithelial cells. Mol Vis. 2011; 17: 558–575.
Robciuc A, Hyotylainen T, Jauhiainen M, Holopainen JM. Hyperosmolarity-induced lipid droplet formation depends on ceramide production by neutral sphingomyelinase 2. J Lipid Res. 2012; 53: 2286–2295.
Juuti-Uusitalo K, Delporte C, Gregoire F et al. Aquaporin expression and function in human pluripotent stem cell-derived retinal pigmented epithelial cells. Invest Ophthalmol Vis Sci. 2013; 54: 3510–3519.
Fahnehjelm KT, Holmstrom G, Ying L, et al. Ocular characteristics in 10 children with long-chain 3-hydroxyacyl-CoA dehydrogenase deficiency: a cross-sectional study with long-term follow-up. Acta Ophthalmol. 2008; 86: 329–337.
Rozanowski B, Burke JM, Boulton ME, Sarna T, Rozanowska M. Human RPE melanosomes protect from photosensitized and iron-mediated oxidation but become pro-oxidant in the presence of iron upon photodegradation. Invest Ophthalmol Vis Sci. 2008; 49: 2838–2847.
Bonilha VL. Age and disease-related structural changes in the retinal pigment epithelium. Clin Ophthalmol. 2008; 2: 413–424.
Burke JM, Kaczara P, Skumatz CM, Zareba M, Raciti MW, Sarna T. Dynamic analyses reveal cytoprotection by RPE melanosomes against non-photic stress. Mol Vis. 2011; 17: 2864–2877.
Roomets E, Kivela T, Tyni T. Carnitine palmitoyltransferase I and acyl-CoA dehydrogenase 9 in retina: insights of retinopathy in mitochondrial trifunctional protein defects. Invest Ophthalmol Vis Sci. 2008; 49: 1660–1664.
Meyer JS, Howden SE, Wallace KA, et al. Optic vesicle-like structures derived from human pluripotent stem cells facilitate a customized approach to retinal disease treatment. Stem Cells. 2011; 29: 1206–1218.
Figure 1
 
Schematic illustration of mitochondrial β-oxidation pathway.
Figure 1
 
Schematic illustration of mitochondrial β-oxidation pathway.
Figure 2
 
Characterization of LCHADD patient samples. (A) Fundus photographs of LCHADD patients showing different stages of pigment retinopathy upper left, patient 1 [early stage], upper right, patient 2 [late stage], lower left and right, patient 3 [stage 2 - grades of pigment deposits (P3) and RPE atrophy (A3) of LCHADD]. Reprinted with permission from Tyni T, Immonen T, Lindahl P, Majander A, Kivelä T. Refined staging for chorioretinopathy in long-chain 3-hydroxyacyl coenzyme A dehydrogenase deficiency. Ophthalmic Res. 2012;48:75–81. Copyright 2012 S. Karger AG, Basel. (B) Sequencing HADHA homozygous c.1528G>C: Control fibroblasts (FiPSC 5–7), homozygous c.G>C substitution in LCHADD fibroblasts, hiPSC from LCHADD fibroblasts (clone 12.22), and hiPSC-differentiated RPE cells (clone 12.22) showed the same substitution. Ctrl, control; Pat, patient.
Figure 2
 
Characterization of LCHADD patient samples. (A) Fundus photographs of LCHADD patients showing different stages of pigment retinopathy upper left, patient 1 [early stage], upper right, patient 2 [late stage], lower left and right, patient 3 [stage 2 - grades of pigment deposits (P3) and RPE atrophy (A3) of LCHADD]. Reprinted with permission from Tyni T, Immonen T, Lindahl P, Majander A, Kivelä T. Refined staging for chorioretinopathy in long-chain 3-hydroxyacyl coenzyme A dehydrogenase deficiency. Ophthalmic Res. 2012;48:75–81. Copyright 2012 S. Karger AG, Basel. (B) Sequencing HADHA homozygous c.1528G>C: Control fibroblasts (FiPSC 5–7), homozygous c.G>C substitution in LCHADD fibroblasts, hiPSC from LCHADD fibroblasts (clone 12.22), and hiPSC-differentiated RPE cells (clone 12.22) showed the same substitution. Ctrl, control; Pat, patient.
Figure 3
 
Morphology of the hiPSC clones. (A) Differentiation of RPE: morphology of control (FiPSC5-7) and patient (clone 12.22) hiPSC-RPE. Magnification: ×20. Scale bars: 20 μm. (B) Control (A116) and patient hiPSC-RPE (clone 12.22) lines showing differences in the pigmentation after 54 days of differentiation. (C) Bar graph showing quantification of cell surface area. Data from control hiPSC-RPE (FiPSC 5–7, A116) and patient hiPSC-RPE (clones 12.1, 12.4, and 12.22) are the average of 100 RPE cells each. Bars represent average with SEM as error bars (*P < 0.05) [P = 0.0365]. Ctrl, control; Pat, patient.
Figure 3
 
Morphology of the hiPSC clones. (A) Differentiation of RPE: morphology of control (FiPSC5-7) and patient (clone 12.22) hiPSC-RPE. Magnification: ×20. Scale bars: 20 μm. (B) Control (A116) and patient hiPSC-RPE (clone 12.22) lines showing differences in the pigmentation after 54 days of differentiation. (C) Bar graph showing quantification of cell surface area. Data from control hiPSC-RPE (FiPSC 5–7, A116) and patient hiPSC-RPE (clones 12.1, 12.4, and 12.22) are the average of 100 RPE cells each. Bars represent average with SEM as error bars (*P < 0.05) [P = 0.0365]. Ctrl, control; Pat, patient.
Figure 4
 
Immunocytochemistry of RPE markers. (AD) Immunocytochemistry demonstrating the expression of RPE markers in control (FiPSC 5-7) and patient (clones 12.1, 12.22) hiPSC-RPE cells. (A) CRALBP in the cytoplasm. (B) Na+/K+ ATPase localized to cell membrane. (C) RPE65 in the cytoplasm. (D) ZO-1, tight junction protein expression in the membrane. Magnification: ×63. Scale bars: 10 μm and 5 μm in inset (AD). (E) Agarose gel electrophoresis of RT-PCR products for RPE differentiation markers, control (FiPSC 5–7) and patient hiPSC-RPE (clone 12.1).
Figure 4
 
Immunocytochemistry of RPE markers. (AD) Immunocytochemistry demonstrating the expression of RPE markers in control (FiPSC 5-7) and patient (clones 12.1, 12.22) hiPSC-RPE cells. (A) CRALBP in the cytoplasm. (B) Na+/K+ ATPase localized to cell membrane. (C) RPE65 in the cytoplasm. (D) ZO-1, tight junction protein expression in the membrane. Magnification: ×63. Scale bars: 10 μm and 5 μm in inset (AD). (E) Agarose gel electrophoresis of RT-PCR products for RPE differentiation markers, control (FiPSC 5–7) and patient hiPSC-RPE (clone 12.1).
Figure 5
 
Phagocytosis assay performed in control (FiPSC 5–7) and patient (clone 12.4). Human-induced pluripotent stem cell-RPE internalizes porcine POS (green). For cell morphology, F-actins were stained using phalloidin (red). View through the depth of image stacks at the position representing red and green lines. Images were taken with a Leica TCS SP2 confocal microscope. Magnification: ×63.
Figure 5
 
Phagocytosis assay performed in control (FiPSC 5–7) and patient (clone 12.4). Human-induced pluripotent stem cell-RPE internalizes porcine POS (green). For cell morphology, F-actins were stained using phalloidin (red). View through the depth of image stacks at the position representing red and green lines. Images were taken with a Leica TCS SP2 confocal microscope. Magnification: ×63.
Figure 6
 
(AE) Immunocytochemistry staining of mitochondrial β-oxidation enzymes, control (FiPSC 5–7) and patient (clones 12.4 and 12.22) hiPSC-RPE cells. (A) ACADM in the cytoplasm. (B) ACAD9 in both membranes and cytoplasm. (C) ACADVL in both membranes and cytoplasm. (D) TFP in both membrane and cytoplasm. (E) CPT1A in the cytoplasm. Scale bars: 10 μm and 5 μm in inset (AE).
Figure 6
 
(AE) Immunocytochemistry staining of mitochondrial β-oxidation enzymes, control (FiPSC 5–7) and patient (clones 12.4 and 12.22) hiPSC-RPE cells. (A) ACADM in the cytoplasm. (B) ACAD9 in both membranes and cytoplasm. (C) ACADVL in both membranes and cytoplasm. (D) TFP in both membrane and cytoplasm. (E) CPT1A in the cytoplasm. Scale bars: 10 μm and 5 μm in inset (AE).
Figure 7
 
Representative electron microscope images of control (FiPSC 5–7, A116) and patient (clone 12.4, 12.22) hiPSC-RPE cells. (A) Several melanolysosomes (arrow) were detected in patient compared to control. (B) Fewer melanosomes were detected in patient cells (asterisk) than in control. (C) Lipid accumulation observed in patient (arrow) compared to control. (D) Tight junctions in control and patient (arrowheads). (E) Basal nuclei (arrow) and apical microvilli (Mv) in control and patient.
Figure 7
 
Representative electron microscope images of control (FiPSC 5–7, A116) and patient (clone 12.4, 12.22) hiPSC-RPE cells. (A) Several melanolysosomes (arrow) were detected in patient compared to control. (B) Fewer melanosomes were detected in patient cells (asterisk) than in control. (C) Lipid accumulation observed in patient (arrow) compared to control. (D) Tight junctions in control and patient (arrowheads). (E) Basal nuclei (arrow) and apical microvilli (Mv) in control and patient.
Figure 8
 
Lipid analysis control (FiPSC 5–7, A116) and patient (clone 12.4, 12.22) hiPSC-RPE cells. (A) Neutral lipid staining (×20) shows a highly dense accumulation of lipids in patient but not in control hiPSC-RPE. Scale bars: 10 μm. (B) Lipidomics data revealed upregulation of TGs in patient hiPSC-RPE. Significant upregulation of polyunsaturated TGs in patient. Controls and patient data are average of two independent experiments using two different cell batches with triplicates in each. Bars represent average with SEM as error bars. *P < 0.05, **P < 0.01 when compared to controls.
Figure 8
 
Lipid analysis control (FiPSC 5–7, A116) and patient (clone 12.4, 12.22) hiPSC-RPE cells. (A) Neutral lipid staining (×20) shows a highly dense accumulation of lipids in patient but not in control hiPSC-RPE. Scale bars: 10 μm. (B) Lipidomics data revealed upregulation of TGs in patient hiPSC-RPE. Significant upregulation of polyunsaturated TGs in patient. Controls and patient data are average of two independent experiments using two different cell batches with triplicates in each. Bars represent average with SEM as error bars. *P < 0.05, **P < 0.01 when compared to controls.
Table
 
List of Primary Antibodies
Table
 
List of Primary Antibodies
Supplement 1
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×