FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2075738060> ?p ?o ?g. }

• W2075738060 endingPage "e297" @default.
• W2075738060 startingPage "e289" @default.
• W2075738060 abstract "Purpose: Hyperglycaemia has been identified as major risk factor for diabetic retinopathy (DR). It is widely accepted that the progression of DR is mainly due to a local imbalance of pro- versus anti-angiogenic factors in the retina. In this study, we investigated whether retinal pigment epithelial (RPE) cells produced pro-angiogenic factors under high glucose (HG) conditions in vitro. Methods: Cultured human retinal endothelial (RE) cells were exposed to conditioned medium from retinal pigment epithelium cells (ARPE-19) grown in HG medium and assessed for tube formation. Based on the expression profiles of ARPE-19, we investigated whether ANGPTL4 was a major angiogenic factor released from ARPE-19 under HG conditions using cultured human RE cells as the test system for experiments with recombinant protein, conditioned medium from ARPE-19 and RNA interference (RNAi). Results: The conditioned medium from ARPE-19 cultured under HG conditions promoted tube formation of cultured human RE cells. GeneChip analysis showed that ANGPTL4 was one of the highest upregulated genes under HG conditions. In addition, recombinant ANGPTL4 promoted all of the elements of angiogenesis in human RE cells in vitro. The results of experiments using conditioned medium from ARPE-19 combined with RNAi demonstrated that ANGPTL4 was a major angiogenic factor released from ARPE-19 under HG conditions. Conclusions: ANGPTL4 was induced by high glucose in RPE cells and exhibited potent angiogenic activity on RE cells. Our results are unique and may potentially add a new candidate to the long list of molecules involved in diabetic retinopathy. Hyperglycaemia has been established as a major risk factor for diabetic retinopathy, but the precise mechanisms underlying the pathophysiology of diabetic retinopathy remain unknown. It is widely accepted that the progression of this disease is mainly due to a local imbalance of pro- versus anti-angiogenic factors in the retina. A number of secreted factors, such as vascular endothelial growth factor (VEGF) (Aiello et al. 1994), angiopoietin2 (Watanabe et al. 2005a), IGF-1(insulin-like growth factor 1) (Miller et al. 2007), erythropoietin (Watanabe et al. 2005b) and pigment epithelium-derived factor (PEDF) (Funatsu et al. 2006), have been reported to be involved in the development of diabetic retinopathy, although the precise involvement of each of these factors in the pathogenesis of the disease remains unknown. Moreover, the origin of these factors is controversial, and it is possible that several types of cells in the retina contribute to the production of these molecules. Retinal pigment epithelial (RPE) cells are mono-layered cubical cells that lie in close proximity to the metabolically highly active photoreceptor cells. They have many important functions, including phagocytosis of the outer segment discs of rods and cones; the uptake, processing and release of vitamin A; and mediation of the vectorial transport of nutrients from the choroidal blood to the photoreceptor cells (Hewitt & Adler 1997). Retinal pigment epithelial cells have also been shown to secrete growth factors such as PEDF (Steele et al. 1992), fibroblast growth factor 5 (Dunn et al. 1996) and platelet-derived growth factor (Campochiaro et al. 1989). These factors have been reported to stimulate the proliferation of fibroblasts, glial cells and the RPE cells themselves (Bryan & Campochiaro 1986). Thus, RPE cells may be a good candidate as the putative origin of diabetic retinopathy-related secretory factors in the retina; however, few studies have investigated this possibility (Sone et al. 1996; Young et al. 2005; Hernández et al. 2006; Yokoyama et al. 2006). The gene for ANGPTL4 has been identified by several groups, and the protein is also known as fasting-induced adipose factor (FIAF) (Kersten et al. 2000), hepatic fibrinogen/angiopoietin-related protein (HFARP) (Kim et al. 2000) and PPARγ angiopoietin-related protein (PGAR) (Yoon et al. 2000). Several previous reports have suggested that the ANGPTL4 protein may be involved in angiogenesis (Ito et al. 2003; Le Jan et al. 2003; Hermann et al. 2005). Recently, it has been reported that ANGPTL4 protein may be involved in the development and pathogenesis of retinal angiogenesis in a model of oxygen-induced retinopathy, but little is known about the expression or the function of this protein in the eye (Perdiguero et al. 2011). In this study, we investigated whether RPE cells play a role in retinal angiogenesis under high glucose conditions. Using human retinal endothelial (RE) cells as a model system, we demonstrated that RPE cells produced tube-forming factors when cultured in medium containing a high glucose concentration and that this activity was mainly mediated by ANGPTL4. The human RPE-derived cell line, ARPE-19, was purchased from the American Type Culture Collection (Rockville, MA, USA). Human RE cells were purchased from Cell Systems Corporation (Kirkland, WA, USA). Recombinant human ANGPTL4 and VEGF were purchased from AdipoGen (Seoul, Korea; AG-40A-0033, formerly designated as RP-AL4102) and R and D Systems (Minneapolis, MN, USA; 293-VE), respectively. CS-C medium and CS-C complete medium [containing 10% foetal bovine serum (FBS) and growth factors] were purchased from Cell Systems Corporation. ARPE-19 cells and RE cells were cultured at 37°C in an atmosphere containing 5% CO2. ARPE-19 cells were used at passages 23–25 for all the experiments. Cells were maintained in Dulbecco’s modified Eagle’s medium with nutrient mixture F12 (Gibco BRL, Grand Island, NY, USA) (DMEM/F12) and 15 mm HEPES buffer containing 7.5 mm glucose supplemented with 10% FBS (FBS; Gibco BRL), 100 U/ml of penicillin and 100 μg/ml of streptomycin (Gibco BRL). For the glucose challenge, after the cells were grown to 90% confluence, they were cultured for another 48 hr in the presence of glucose at a concentration of 5 mm, 7.5 mm, 12.5 mm or 17.5 mm in DMEM/F12 supplemented with 10% FBS, which we referred to as very low glucose (VLG), low glucose (LG), medium glucose (MG) or high glucose (HG) medium, respectively. Media from ARPE-19 cells conditioned in the presence of a high or low glucose (17.5 or 7.5 mm) were abbreviated as high glucose-conditioned media (HG-CM) or low glucose-conditioned media (LG-CM), respectively. ARPE-19 cells were cultured for 72 hr, except for the knockdown experiments in which they were incubated for 48 hr. RE cells were used at passages 3–6 for all the experiments. They were grown in type-I collagen-coated dishes (Iwaki, Tokyo, Japan) and maintained in CS-C complete medium. The tube formation assay was conducted using BD BioCoat Angiogenesis System Endothelial Cell Tube Formation plates (BD Biosciences, Bedford, MA, USA) in accordance with the protocol supplied by the manufacturer. Images of the tubules formed were obtained under fluorescence microscopy at 4× magnification. The tube length in each well was measured from one representative image per well using the WinRoof software (MITANI Corporation, Tokyo, Japan) and presented as a percentage of the control. Total RNA was prepared from the ARPE-19 cells cultured in each type of medium using TRIzol Reagent (Invitrogen, Carlsbad, CA, USA). For transcriptome analysis, expression profiles of ARPE-19 cells cultured in LG or HG medium containing 10% FBS were compared by the human HG-U133 Set Array (Affymetrix, Santa Clara, CA, USA). Samples for the analysis were prepared in accordance with the manufacturer’s protocol. Analysis was performed using the Affymetrix GeneChip® Operating Software (GCOS). Probe sets were re-annotated using eldorado software (Genomatix, Munich, Germany), and the cellular components of the genes were defined by Gene Ontology (GO) annotation. Real-time RT-PCR analysis was performed using the TaqMan® probe with an ABI Prism 7900 sequence-detection system (Applied Biosystems, Foster City, CA, USA). The levels of the target genes were normalized to the levels of human ACTB (β-actin) using the comparative threshold method. The identification numbers of the primers for human ANGPTL4 and human ACTB were Hs00211522_ml and Hs99999903_ml, respectively (Applied Biosystems). The effect of glucose on ANGPTL4 expression was further analysed in ARPE-19 cells. After the cells reached 80% confluence in LG medium, they were further incubated in HG medium for 0, 30, 60, 90 or 120 min for a time-course analysis. For concentration–response analysis, they were incubated for 48 hr in the presence of various concentrations of glucose (VLG, LG, MG or HG medium). The expression level of ANGPTL4 mRNA in the ARPE-19 cells was determined by real-time RT-PCR as described previously. ARPE-19 cells were collected and homogenized in 25 mm Tris/HCl pH 7.5, 150 mm NaCl and 1% (w/v) Triton-X, followed by centrifugation at 4°C. The supernatant or culture medium was incubated with 10 μl of anti-ANGPTL4 antibody (Abnova, Taipei City, Taiwan; H0051129-A01) fixed on protein G Sepharose beads (Amersham Pharmacia Biotech, Piscataway, NJ, USA). The immunoprecipitates were boiled for 10 min in the presence of β-mercaptoethanol and resolved by SDS-PAGE. Western blots were performed using anti-ANGPTL4 antibody as described previously (Nishimura et al. 2002). For migration and invasion assays, 1 × 105 RE cells were seeded on BD BioCoat Angiogenesis System Endothelial Cell Migration and Invasion plates (BD Bioscience) in CS-C Medium containing 0.4% FBS and either VEGF (5 ng/ml) or ANGPTL4 (100 ng/ml), in accordance with the protocols supplied by the manufacturer. Cell migration was assayed by motility through an unoccluded pore in the presence of a thin layer of human fibronectin, and cell invasion was measured by the ability for endothelial cells to degrade extracellular matrix and migrate through the pores. The proliferative activity of RE cells was assessed based on cellular mitochondrial dehydrogenase activity. An aliquot of 2 × 104 RE cells was added to each well of a 96-well tray containing CS-C complete medium. After starvation, the cells were exposed to CS-C medium containing 0.4% FBS with either VEGF (5 ng/ml) or ANGPTL4 (100 ng/ml) for 48 hr, and then Cell Count Reagent SF (Nacalai Tesque, Kyoto, Japan) was applied for 4 hr. For RNAi experiments, ARPE-19 cells were seeded at a density of 2 × 105 cells/well into six-well culture plates containing LG medium and incubated for 24 hr. The transfection was performed using Dharmafect® reagent (Dharmacon Inc., Lafayette, CO, USA) in LG medium and siRNA at a final concentration of 20 nm. The siRNAs used were for PPARγ, HIF-1α, EGR1, ANGPTL4, VEGF or Lamin A/C as the control (siGENOME SMART pool M_003436-01, M_004018-02, M_006526-01, M_003550-01-0010, M_007807-01-0010, and D-001050-01-05; Dharmacon Inc.). After 48-hr transfection, the medium was replaced by HG medium, and the incubation was continued for another 48 hr. The number of independent experiments is noted in the text and figure legends. The results are expressed as mean values ± SD. StatView-5.0 software (SAS Institute Inc., Cary, NC, USA) was used for all statistical analyses. The Student’s t-test was carried out to determine whether the differences between groups were statistically significant. Either p < 0.05 or p < 0.01 was considered significant. To evaluate whether the factors secreted by ARPE-19 cells promoted tube formation in RE cells in vitro, conditioned media from ARPE-19 cells were overlaid onto tube formation plates containing RE cells. At first, we examined the effect of glucose by using LG or HG medium. The HG medium (Fig. 1C) increased tube formation in RE cells by 89.7 ± 53.0% (p < 0.05, n = 5), as compared with LG medium (Fig. 1A). Next, we examined the effect of HG-CM. The length of the tubules formed in HG-CM (Fig. 1D) was 413.4 ± 78.2% (p < 0.01, n = 5) higher than that in LG-CM (Fig. 1B). A significant increase (602.0 ± 132.7%; p < 0.01, n = 5) in the length of the tubules formed by RE cells was also observed when the culture system was overlaid with HG-CM (Fig. 1D), as compared with that in HG medium (Fig. 1C). Thus, these results suggest that HG-CM contained high glucose-induced angiogenic factors that promote the tube formation of RE cells (Fig. 1E). Tube formation activity of retinal endothelial (RE) cells in high glucose-conditioned media (HG-CM) in vitro. The tube formation activity of the RE cells in low glucose-conditioned media (LG-CM) or HG-CM was assessed by incubation for 16 hr on BD BioCoat plates. (A) Control: low glucose (LG) medium. (B) LG-CM. (C) HG medium. (D) HG-CM. (E) Quantitative analysis of network structures. The total length of the network structures was measured and the total length per field was calculated as a ratio of that in the control. n = 5. Values are expressed as means ± SD. **p < 0.01, *p < 0.05 as compared with the result in the control. Scale bars, 200 μm. We next attempted to identify the angiogenic factors secreted by the ARPE-19 cells cultured in the presence of a high glucose concentration. Among the upregulated genes revealed by the gene expression analysis of ARPE-19 cells (Table 1), we focused on ANGPTL4 as a strong candidate for the high glucose concentration-induced angiogenic activity, as the expression of this gene showed the highest level of upregulation; its expression ranged from 2.46-fold [with probe set 223333_s_at in experiment 1] to 13.9-fold [with probe set 221009_s_at in experiment 2]. We next investigated the induction of ANGPTL4 expression by glucose in further detail. The temporal profile of ANGPTL4 mRNA expression in the HG medium is shown in Fig. 2A. Significant induction of ANGPTL4 mRNA expression in the presence of a high glucose concentration was observed at 60 min (1.37 ± 0.14-fold, p < 0.05, n = 3); the expression level was further increased at 90 min (2.35 ± 0.13-fold, p < 0.01, n = 3) and at 120 min (4.52 ± 0.54-fold, p < 0.01, n = 3), as compared with the control (0 min). In previous studies, both the full-length and the cleaved forms (fibrinogen-like domain) of ANGPTL4 were reported, which resolved 55 and 38 kDa, respectively (Cazes et al. 2006). While the predicted size of the full-length ANGPTL4 is approximately 40 kDa, the post-transcriptional modification may explain the apparent discrepancy. Immunoblot analyses using anti-ANGPTL4 antibody demonstrated a approximately 55-kDa protein both in cell lysates and conditioned medium from ARPE-19 cells cultured under the high glucose concentrations (Fig. 2B), suggesting the majority of ANGPTL4 in HG-CM was the full-length protein. We also investigated the expression levels of ANGPTL4 in ARPE-19 cells following exposure to different concentrations of glucose. Figure 2C shows that the ANGPTL4 mRNA expression was progressively upregulated by increasing glucose concentrations in the medium. Cells cultured in MG and HG media showed an increase in ANGPTL4 mRNA expression by 2.44 ± 0.37-fold (p < 0.01, n = 3) and 5.78 ± 0.87-fold (p < 0.01, n = 3), respectively, as compared with cells cultured in LG medium, which suggests that the expression of ANGPTL4 mRNA can also be significantly induced by moderately elevated glucose levels, corresponding to the condition that can be expected in patients with fairly well-controlled diabetes. While the induction of ANGPTL4 in the liver and adipose tissue was demonstrated under fasting conditions (Kersten et al. 2000), ARPE-19 cells exposed to VLG medium showed a decrease in its expression (0.38 ± 0.18-fold, p < 0.05, n = 3). These results suggest the tissue specificity of the regulation of ANGPTL4 gene expression. The effects of HG medium on ANGPTL4 mRNA and protein expression. (A) Time-course of the effects in HG medium (0–120 min). n = 3. *p < 0.05, **p < 0.01 as compared with that at 0 min. (B) Immunoblot analyses using anti-ANGPTL4 antibody. Immunoprecipitates from cell lysates (left, Lysate) and culture medium (right, CM) of ARPE-19 cells cultured in the high glucose medium were analysed by SDS-PAGE followed by Western blotting. Expression of both the full-length form (bold arrowhead) and the short form of ANGPTL4 was observed in the lysates, but the former is the predominant form in CM. The image is representative of three experiments. (C) Effects of elevated glucose levels for 48 hrs’ stimulation. n = 3. The mRNA level of ANGPTL4 was normalized to that of ACTB, and the relative values were compared with those at 0 min (A) and low glucose (LG) (C), respectively, and expressed as means ± SD. *p < 0.05, **p < 0.01 as compared with the result in LG medium. To investigate the transcriptional machinery of ANGPTL4 expression in the ARPE-19 cells more directly, we focused on three transcription factors and performed RNAi experiments. ANGPTL4 has been reported to be a downstream target of PPARγ (Yoon et al. 2000). It has been reported that the expression of ANGPTL4 mRNA was highly upregulated in cardiomyocytes under hypoxic conditions by HIF-1α (Belanger et al. 2002). We also focused on EGR1, which was one of the genes upregulated by high glucose in our GeneChip analysis (about two to threefold, Table 1). EGR1 is an 80–82-kDa zinc finger transcription factor known to be involved in cell proliferation and differentiation (Sukhatme et al. 1988) and has been shown to activate the transcription of many genes in the presence of high glucose, including PDGF A and B, VEGF and insulin-like growth factor-2 (Silverman & Collins 1999). The effects of siRNA inhibition of PPARγ, HIF-1α and EGR1 on the ANGPTL4 expression are shown in Fig. 3. Expression of the ANGPTL4 gene in ARPE-19 cells was significantly reduced by 89.8 ± 2.2% (p < 0.01, n = 3) and 94.4 ± 3.5% (p < 0.01, n = 3) by the siRNAs for HIF-1α and PPARγ, respectively, as compared with cells transfected with the control, but no significant change was noted after treatment with the siRNA for EGR1. These results suggest that both PPARγ and HIF-1α are essential for the induction of ANGPTL4 by glucose in ARPE-19 cells. ANGPTL4 mRNA expression in ARPE-19 cells treated with siRNAs. ANGPTL4 mRNA expression in ARPE-19 cells treated with 20 nm of siRNA for either HIF-1α, PPARγ or EGR1 and cultured in HG medium for 48 hr was measured by real-time RT-PCR and expressed as a ratio of that in low glucose medium. n = 3. **p < 0.01 as compared with the result in control, Lamin A/C. NS, no significance. Angiogenesis by endothelial cells involves multiple steps, namely cell invasion, cell migration, cell proliferation and tube formation (Folkman 1986). Therefore, we studied the induction of each of these steps in RE cells by ANGPTL4 in vitro.Figure 4A shows that treatment with recombinant full-length ANGPTL4 (100 ng/ml) or VEGF (5 ng/ml) increased invasion by 83.0 ± 20.7% (p < 0.01, n = 5) and 139.5 ± 35.2% (p < 0.01, n = 5), respectively, as compared to control-treated cells. Figure 4B shows that treatment with ANGPTL4 or VEGF increased migration by 78.1 ± 27.2% (p < 0.01, n = 4) and 160.6 ± 35.7% (p < 0.01, n = 4), respectively, as compared to the control. Figure 4C shows that treatment with ANGPTL4 or VEGF increased the proliferative activity by 73.3 ± 10.0% (p < 0.01, n = 4) and 133.3 ± 36.6% (p < 0.01, n = 4), respectively, as compared with the control. We also investigated the effects of recombinant ANGPTL4 and VEGF on the tube formation by RE cells (Fig. 5A–D). In this experiment, we used 15 ng/ml rather than 5 ng/ml VEGF, as 5 ng/ml VEGF failed to significantly induce tube formation. Figure 5D shows that treatment with ANGPTL4 (100 ng/ml) or VEGF (15 ng/ml) increased the tube length by 154.1 ± 43.3% (p < 0.01, n = 5) and 126.2 ± 34.4% (p < 0.01, n = 5), respectively, as compared with that in the control. These results indicate that ANGPTL4 promotes each step of angiogenesis in RE cells almost as potently as VEGF in vitro. Retinal endothelial (RE) cell migration, invasion and proliferation assays. Either vascular endothelial growth factor (5 ng/ml) or ANGPTL4 (100 ng/ml) human recombinant protein was added to plates containing RE cells. (A) The invasion plate was incubated for 21 hr. n = 5. (B) The migration plate was incubated for 27 hr. n = 4. (C) The proliferation assay was conducted for 48 hr using Count Reagent SF. n = 4. Values are expressed as means ± SD and are calculated as a ratio of that in the control, CS-C Medium. **p < 0.01 compared with control. In vitro tube formation assays. Either recombinant human ANGPTL4 or vascular endothelial growth factor (VEGF) was added to plates containing retinal endothelial (RE) cells for 16 hr. (A) Control. CS-C Medium. (B) VEGF (15 ng/ml). (C) ANGPTL4 (100 ng/ml). (D) Quantitative analysis of the tube formation activity of RE cells treated with either VEGF (15 ng/ml) or ANGPTL4 (100 ng/ml). n = 5. Values are expressed as means ± SD. **p < 0.01 as compared with that in the control. Scale bars, 200 μm. Next, we treated the ARPE-19 cells with siRNA for ANGPTL4, VEGF or HIF-1α (Fig. 6) to determine whether these factors were involved in the tube-forming activity induced by ARPE-19 cells cultured in HG medium. ANGPTL4, VEGF or HIF-1α mRNA were reduced by more than 80% by the corresponding siRNA treatment (data not shown). Effect of treatment of high glucose-conditioned media (HG-CM) treated with siRNAs on the tube formation activity of retinal endothelial (RE) cells in vitro. The HG-CM treated with 20 nm of siRNA for the control, Lamin A/C (A), vascular endothelial growth factor (VEGF) (B), ANGPTL4 (C) or HIF-1α (D) for 48 hr was added to the plates containing RE cells for 16 hr. (E) Quantitative analysis of the tube formation activity of the RE cells overlaid with HG-CM treated with 20 nm of either the control siRNA or the siRNA for VEGF, ANGPTL4 or HIF-1α and calculated as a ratio of that in HG medium. n = 5. Values are expressed as means ± SD. *p < 0.05 **p < 0.01 as compared with the result in the control. Scale bars, 200 μm. While the addition of HG-CM treated with the siRNA for VEGF (Fig. 6B) shows reduction of the tube formation activity by 42.1 ± 5.9% (p < 0.05, n = 5) in this analysis, HG-CM treated with 20 nm of the siRNA for ANGPTL4 (Fig. 6C) showed reduction of the tube formation activity by 68.6 ± 8.3% (p < 0.01) and those treated with the siRNA for HIF-1α (Fig. 6D) showed reduction of the tube formation by 93.6 ± 1.6% (p < 0.01, n = 5) (Fig. 6E) as compared with the cells treated with control siRNA for Lamin A/C (Fig. 6A). In contrast, knockdown of the EGR1 gene did not reduce tube formation by RE cells in HG-CM (data not shown), suggesting that the induction of ANGPTL4 expression was not dependent on EGR1. In this study, we demonstrated that ARPE-19 cells produced factors with potent angiogenic activity on human RE cells in the presence of a high glucose concentration, and that this angiogenic activity could be primarily attributed to ANGPTL4. Contributions from other angiogenic factors such as VEGF cannot be excluded, however; and the identification of these factors is under way. ANGPTL4 belongs to the angiopoietin-like protein family (ANGPTLs), members of which are structurally similar to the angiopoietins, with N-terminal coiled-coiled domains and C-terminal fibrinogen-like domains, but do not bind to either of the angiopoietin receptors, Tie1 or Tie2 (Kim et al. 2000). Although some ANGPTLs, including ANGPTL3 and ANGPTL6, exhibit angiogenic activity (Camenisch et al. 2002; Oike et al. 2004), the potential angiogenic activity of other family members, including ANGPTL2 (Dhanabal et al. 2002; Kubota et al. 2005) and ANGPTL4, is rather controversial. Indeed, previous studies have reported both pro-angiogenic (Le Jan et al. 2003; Hermann et al. 2005; Perdiguero et al. 2011) and anti-angiogenic effects (Ito et al. 2003) of ANGPTL4. In this study, we demonstrated that recombinant human ANGPTL4 protein significantly promoted each of the four steps of angiogenesis on RE cells, namely invasion, migration, proliferation and tube formation (Auerbach et al. 2003) and that its angiogenic activity was as potent as that of VEGF, whose involvement in the development of diabetic retinopathy has been well established (Aiello et al. 1994). Although specific receptors for ANGPTL4 have not been identified, its binding to integrins and other extracellular matrix proteins has been reported in several cell types (Goh et al. 2010; Huang et al. 2011). In this study, MatrigelTM and fibronectin were included in the invasion and the tube formation assay systems and in the migration assay system, respectively. Thus, the effects of this protein may be dependent on the assay system or the origin of endothelial cells. However, as the angiogenic activity of this protein on RE cells was evident in our system at least in vitro, it may be interesting to note that the additive effects of VEGF and ANGPTL4 on angiogenesis have been reported in white adipose tissue (Gealekman et al. 2008). One unexpected finding in this study was the induction of ANGPTL4 expression by high glucose in the ARPE-19 cells. Expression of ANGPTL4 has been reported to be strongly upregulated in liver and adipose tissue under fasting conditions (Kersten et al. 2000; Yoon et al. 2000). Increased and decreased plasma levels of ANGPTL4 in fasting conditions and under chronic high-fat feeding have been reported, indicating that ANGPTL4 may function as an endocrine signal involved in the regulation of metabolism (Kersten et al. 2000; Yoon et al. 2000). The glucose-induced, rather than glucose-suppressed, expression of ANGPTL4 in the ARPE-19 cells suggests that its expression may be regulated in a tissue-specific manner (Sango et al. 2006; Yamada et al. 2006; Dutton & Trayhurn 2008). The mechanisms by which high glucose induces ANGPTL4 expression in RPE cells are unknown, and multiple pathways may be involved. Among the three transcription factors we examined, HIF-1α appeared to be one of the essential factors for the glucose-induced ANGPTL4 expression in ARPE-19 cells. The transcriptional regulation of HIF-1α may be complicated, but our data may have a potential therapeutic implication. Tissue or cellular hypoxia is often reported to be involved in diabetes-related lesions, including atherosclerosis (Williamson et al. 1993). It has reported that pathological neovascularization, which results from tissue hypoxia, was also strongly inhibited in angptl4-deficient mice in a model of oxygen-induced retinopathy, and ANGPTL4 controls hypoxia-driven angiogenesis in the retina (Perdiguero et al. 2011). Also, it has been reported that ANGPTL4 gene expression is upregulated by HIF-1α in cardiomyocytes (Belanger et al. 2002). Although we did not directly demonstrate HIF-1α activation in ARPE-19 cells, the dependence of ANGPTL4 on HIF-1α may have novel implications, as diabetic retinopathy is almost always associated with tissue ischaemia in the retina (Antonetti et al. 2006). ANGPTL4 was reported to be induced by PPARγ activation in adipose tissues (Yoon et al. 2000), and we also showed that PPARγ is essential for glucose-induced ANGPTL4 expression in ARPE-19 cells, although we did not demonstrate direct PPARγ activation by glucose. Insulin has also been reported to suppress ANGPTL4 expression in adipocytes (Yamada et al. 2006). While some patients of type 2 diabetes mellitus exhibit hyperinsulinemia which compensates for insulin resistance, others are insulin-deficient especially after a long clinical course of the disease. It may interesting to speculate that ANGPTL4 induction in the diabetic eye may be context-dependent and affected by the ambient insulin concentrations, although we did not test this hypothesis in our system. Several recent lines of evidence indicate that ANGPTLs may play important roles in metabolic homoeostasis as well as angiogenesis (Oike et al. 2005). Indeed, it has been shown that hepatocyte-derived ANGPTL3 and ANGPTL4 can regulate lipid metabolism (Köster et al. 2005), and variants in the ANGPTL4 gene have been associated with lipid abnormalities in humans (Maxwell et al. 2010). Our study provides further evidence, suggesting that ANGPTL4 may be an important link between metabolism and angiogenesis in the retina. Of course, this does not reduce the importance of known factors such as VEGF in diabetic retinopathy, and although it remains to be determined in future studies whether human RPE cells in vivo would also produce ANGPTL4 under hyperglycaemic conditions, it would be interesting to see whether ANGPTL4 may be involved in some subtypes of diabetic retinopathy. In summary, we demonstrated that the induction of angiogenic activity by glucose in ARPE-19 cells is mediated, at least in part, by the upregulation of ANGPTL4. Although our model, which is based on acute challenge by high glucose, does not precisely mimic the long clinical course of diabetic retinopathy, and thus in vivo validation in diabetic animal models is necessary, our data suggest the novel possibility that RPE cells might contribute to the pathogenesis of diabetic retinopathy via upregulated expression of ANGPTL4. We would like to thank Kazuo Okubo, Dai Suzuki and Kazuko Nagase for their technical Assistance, and Toshiyuki Oshitari for discussions and comments on this manuscript. This work was supported by a Grant from the Ministry of Health, Labour and Welfare of Japan (K.Y.) and a grant from National Center for Global Health and Medicine (K.Y.)." @default.
• W2075738060 created "2016-06-24" @default.
• W2075738060 creator A5021261649 @default.
• W2075738060 creator A5027211308 @default.
• W2075738060 creator A5046304131 @default.
• W2075738060 creator A5060761026 @default.
• W2075738060 creator A5063065419 @default.
• W2075738060 creator A5083477042 @default.
• W2075738060 creator A5089470172 @default.
• W2075738060 date "2013-02-07" @default.
• W2075738060 modified "2023-09-25" @default.
• W2075738060 title "Angiopoietin-like protein 4 (ANGPTL4) is induced by high glucose in retinal pigment epithelial cells and exhibits potent angiogenic activity on retinal endothelial cells" @default.
• W2075738060 cites W1600727360 @default.
• W2075738060 cites W1967466134 @default.
• W2075738060 cites W1971826305 @default.
• W2075738060 cites W1982265214 @default.
• W2075738060 cites W1984024821 @default.
• W2075738060 cites W2001047183 @default.
• W2075738060 cites W2010155782 @default.
• W2075738060 cites W2010903558 @default.
• W2075738060 cites W2014590172 @default.
• W2075738060 cites W2018643656 @default.
• W2075738060 cites W2024008844 @default.
• W2075738060 cites W2026490523 @default.
• W2075738060 cites W2031824514 @default.
• W2075738060 cites W2035216298 @default.
• W2075738060 cites W2035562370 @default.
• W2075738060 cites W2037326491 @default.
• W2075738060 cites W2050866162 @default.
• W2075738060 cites W2057253189 @default.
• W2075738060 cites W2066650638 @default.
• W2075738060 cites W2071269124 @default.
• W2075738060 cites W2086308510 @default.
• W2075738060 cites W2091767953 @default.
• W2075738060 cites W2094719911 @default.
• W2075738060 cites W2100453456 @default.
• W2075738060 cites W2110646133 @default.
• W2075738060 cites W2114234975 @default.
• W2075738060 cites W2116671187 @default.
• W2075738060 cites W2117617326 @default.
• W2075738060 cites W2120895512 @default.
• W2075738060 cites W2126047917 @default.
• W2075738060 cites W2132949972 @default.
• W2075738060 cites W2142853396 @default.
• W2075738060 cites W2145969545 @default.
• W2075738060 cites W2157340523 @default.
• W2075738060 cites W2167604659 @default.
• W2075738060 cites W2324588265 @default.
• W2075738060 cites W4236050547 @default.
• W2075738060 cites W4255733162 @default.
• W2075738060 doi "https://doi.org/10.1111/aos.12097" @default.
• W2075738060 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/23387337" @default.
• W2075738060 hasPublicationYear "2013" @default.
• W2075738060 type Work @default.
• W2075738060 sameAs 2075738060 @default.
• W2075738060 citedByCount "24" @default.
• W2075738060 countsByYear W20757380602013 @default.
• W2075738060 countsByYear W20757380602014 @default.
• W2075738060 countsByYear W20757380602015 @default.
• W2075738060 countsByYear W20757380602016 @default.
• W2075738060 countsByYear W20757380602017 @default.
• W2075738060 countsByYear W20757380602018 @default.
• W2075738060 countsByYear W20757380602019 @default.
• W2075738060 crossrefType "journal-article" @default.
• W2075738060 hasAuthorship W2075738060A5021261649 @default.
• W2075738060 hasAuthorship W2075738060A5027211308 @default.
• W2075738060 hasAuthorship W2075738060A5046304131 @default.
• W2075738060 hasAuthorship W2075738060A5060761026 @default.
• W2075738060 hasAuthorship W2075738060A5063065419 @default.
• W2075738060 hasAuthorship W2075738060A5083477042 @default.
• W2075738060 hasAuthorship W2075738060A5089470172 @default.
• W2075738060 hasConcept C104317684 @default.
• W2075738060 hasConcept C118487528 @default.
• W2075738060 hasConcept C167734588 @default.
• W2075738060 hasConcept C178790620 @default.
• W2075738060 hasConcept C185592680 @default.
• W2075738060 hasConcept C2777025900 @default.
• W2075738060 hasConcept C2779094623 @default.
• W2075738060 hasConcept C2780827179 @default.
• W2075738060 hasConcept C502942594 @default.
• W2075738060 hasConcept C55493867 @default.
• W2075738060 hasConcept C64584667 @default.
• W2075738060 hasConcept C71924100 @default.
• W2075738060 hasConcept C77305985 @default.
• W2075738060 hasConcept C86803240 @default.
• W2075738060 hasConcept C95444343 @default.
• W2075738060 hasConceptScore W2075738060C104317684 @default.
• W2075738060 hasConceptScore W2075738060C118487528 @default.
• W2075738060 hasConceptScore W2075738060C167734588 @default.
• W2075738060 hasConceptScore W2075738060C178790620 @default.
• W2075738060 hasConceptScore W2075738060C185592680 @default.
• W2075738060 hasConceptScore W2075738060C2777025900 @default.
• W2075738060 hasConceptScore W2075738060C2779094623 @default.
• W2075738060 hasConceptScore W2075738060C2780827179 @default.
• W2075738060 hasConceptScore W2075738060C502942594 @default.
• W2075738060 hasConceptScore W2075738060C55493867 @default.
• W2075738060 hasConceptScore W2075738060C64584667 @default.
• W2075738060 hasConceptScore W2075738060C71924100 @default.

• next