- Email: [email protected]
Anti-Angiogenic Effects of Ribonucleic Acid Interference Targeting Vascular Endothelial Growth Factor and Hypoxia-Inducible Factor-1α
Anti-Angiogenic Effects of Ribonucleic Acid Interference Targeting Vascular Endothelial Growth Factor and Hypoxia-Inducible Factor-1␣ FARZIN FOROOGHIAN AND BIKUL DAS ● PURPOSE:
To compare the in vitro anti-angiogenic effects of inhibiting vascular endothelial growth factor (VEGF) and hypoxia-inducible factor 1-␣ (HIF-1␣) using ribonucleic acid (RNA) interference (RNAi). ● DESIGN: Laboratory investigation. ● METHODS: VEGF or HIF-1␣ was antagonized in human retinal pigment epithelial (RPE) cells using RNAi, and then cells were cultured under hypoxia. Angiogenic proteins secreted into the media were measured using enzyme-linked immunosorbent assay. Media from hypoxic RPE cells was used to grow human umbilical vein endothelial cells (HUVECs). Capillary tube formation by HUVECs was quantified and compared to assess the effectiveness of angiogenesis. ● RESULTS: RNAi targeting VEGF caused a significant decrease in VEGF in addition to several other clinically important angiogenic factors, including angiogenin, interleukin-6 (IL-6), interleukin-8 (IL-8), monocyte chemoattractant protein-1 (MCP-1), and tumor growth factor 1 (TGF-1). Although HIF-1␣ RNAi reduced the production of VEGF, angiogenin, and TGF-1, we observed an increase in the levels of several other angiogenic factors like IL-6, IL-8, and MCP-1. RNAi of VEGF and HIF-1␣ was effective in inhibiting angiogenesis, although the effect was more pronounced for VEGF RNAi. ● CONCLUSIONS: RNAi of VEGF and HIF-1␣ may have therapeutic potential in ischemic retinal diseases like diabetic retinopathy. Targeting VEGF seems to have the advantage of decreasing the production of several clinically important angiogenic factors, thereby effectively inhibiting angiogenesis. Antagonism of HIF-1␣ may lead to the overactivation of alternate transcription factors and their respective gene products, leading to less effective inhibition of angiogenesis. (Am J Ophthalmol 2007;144:761–768. © 2007 by Elsevier Inc. All rights reserved.) Accepted for publication Jul 16, 2007. From the Department of Ophthalmology and Vision Sciences (F.F.), the Division of Oncology (B.D.), and the Research Institute (F.F., B.D.), The Hospital for Sick Children, Toronto, Ontario, Canada; and the Institute of Medical Science, University of Toronto, Toronto, Ontario, Canada (F.F., B.D.). Inquiries to Farzin Forooghian, Hospital for Sick Children, 555 University Avenue, Room 10128, Elm Wing, Toronto, Ontario, Canada M5G 1X8; e-mail: [email protected] 0002-9394/07/$32.00 doi:10.1016/j.ajo.2007.07.022
©
2007 BY
I
T IS WELL-KNOWN THAT ANGIOGENIC GROWTH FAC-
tors contribute to the pathologic neovascularization that occurs in ischemic retinal diseases like diabetic retinopathy. Vascular endothelial growth factor (VEGF) has been studied extensively in this respect, and monoclonal antibodies and aptamers targeting VEGF currently are being investigated as potential therapies for diabetic retinopathy.1,2 A potential drawback of newly emerging anti-VEGF therapies is that only one of multiple potentially important angiogenic factors is targeted. Laser photocoagulation and intravitreal triamcinolone are two currently used treatment methods for diabetic retinopathy. These treatments are believed to exert their effects partly through the downregulation of multiple angiogenic factors.3,4 These therapeutic methods, however, are not without significant side effects.5,6 Recently, the pathogenic role of erythropoietin (EPO) in diabetic retinopathy has been recognized.7,8 Both VEGF and EPO are regulated by the transcription factor hypoxiainducible factor-1␣ (HIF-1␣), which has been proposed as a novel therapeutic target for ischemic retinal disease.9 With respect to diabetic retinopathy, HIF-1␣ is an attractive target because elevated glucose concentrations have been shown to upregulate HIF-1␣ in human retinal pigment epithelial (RPE) cells.10 Ribonucleic acid (RNA) interference (RNAi) recently emerged as a potentially exciting therapeutic method for various disease states. A key therapeutic advantage to using RNAi lies in its ability to knock down the expression of disease-causing genes of known sequence specifically and potently. Interestingly, the first clinical application involving RNAi has been directed at the treatment of age-related macular degeneration. Clinical trials involving RNAi targeting VEGF or its receptor through intravitreal injection delivery currently are underway.11 The eye is a unique organ in terms of allowing direct drug delivery, thereby overcoming many of the difficulties associated with systemic RNAi delivery.11 Investigations into the therapeutic potential of RNAi for other blinding ocular diseases such as diabetic retinopathy thus are not only interesting, but also are highly warranted. In addition to VEGF and EPO, numerous other angiogenic factors are clinically important in diabetic retinopathy. Angiogenin, interleukin-6 (IL-6), interleukin-8 (IL-8), monocyte chemoattractant protein-1 (MCP-1), leptin, placental
ELSEVIER INC. ALL
RIGHTS RESERVED.
761
FIGURE 1. Bar graphs demonstrating decreased expression of angiogenic factors by retinal pigment epithelial (RPE) cells after ribonucleic acid (RNA) interference. After transfection with either random short interfering RNA (siRNA) or siRNA targeting vascular endothelial growth factor (VEGF) or hypoxia-inducible factor-1␣ (HIF-1␣), human RPE (ARPE-19) cells were incubated under hypoxia (3% 02) for 24 hours and then conditioned media were assayed using enzyme-linked immunosorbent assay. Cells incubated without any siRNA were used as a negative control. Error bars indicate standard deviation (SD). All pairwise comparisons between histogram bars are statistically significant (P < .01), except for those with an asterisk (*). IL ⴝ interleukin; MCP-1 ⴝ monocyte chemoattractant protein 1; TGF-1 ⴝ transforming growth factor 1.
growth factor (PIGF), transforming growth factor-1 (TGF-1), and basic fibroblast growth factor (bFGF) all have been found to be elevated in the vitreous of patients with diabetic retinopathy.12–18 Furthermore, the severity of diabetic retinopathy has been correlated to vitreous levels of IL-6, IL-8, MCP-1, and bFGF.17,19,20 Any study that attempts to address the anti-angiogenic properties of a novel therapeutic method therefore must examine these angiogenic factors as well. The RPE is an abundant source of angiogenic factors in the retina. In addition to VEGF,21 these cells have been shown to express IL-6, IL-8, MCP-1, bFGF, TGF-1, and PIGF under various conditions.22–26 RPE cells thus are an attractive therapeutic target. RNAi in the form of short interfering RNA (siRNA) targeting VEGF previously has been shown to inhibit VEGF production by human RPE cells,27 demonstrating that these cells are amenable to siRNA transfection. Although VEGF is an attractive target for RNAi experiments on ocular angiogenesis, other targets need to be considered as well. The effects of HIF-1␣ inhibition on ischemia-induced retinal angiogenesis are not yet known. The purpose of this study was to compare the antiangiogenic properties of RNAi targeting HIF-1␣ with that of RNAi targeting VEGF in an in vitro model of ischemiainduced retinal angiogenesis. 762
AMERICAN JOURNAL
METHODS ● RETINAL PIGMENT EPITHELIAL CELL CULTURE:
Human RPE cells (ARPE-19) were purchased commercially (ATCC, Manassas, Virginia, USA). Cells were cultured (37 C, 5% CO2) to confluence in 56.7 cm2 Nunclon culture dishes (Nalge Nunc International, Rochester, New York, USA) with D-MEM/F-12 growth medium (15 ml; ATCC) supplemented with 10% heat-inactivated fetal bovine serum (FBS) and penicillin (100 units/ml) and streptomycin (100 g/ml). For hypoxia experiments, the medium was replaced with D-MEM/F-12 growth medium (15 ml) supplemented with penicillin (100 units/ml) and streptomycin (100 g/ml) only. After 24 hours of growth in a hypoxic incubator (37 C, 3% O2), culture medium was removed and centrifuged to remove any cellular debris, and the supernatant was frozen at ⫺80 C until further use. Conditioned medium from two separate experiments was pooled together before freezing at ⫺80 C. All experiments were carried out using cells within the first five passages. Observation of the cells by light microscopy after over 36 hours of hypoxia demonstrated that they were viable and healthy.
● siRNA TRANSFECTION:
Cells were grown to 90% confluence for transfection. For siRNA experiments, ONTARGETplus SMARTpool HIF-1␣ siRNA or VEGF
OF
OPHTHALMOLOGY
NOVEMBER 2007
siRNA was used, whereas fluorescent siGLO RISC-free siRNA was used as a negative control (Dharmacon, Lafayette, Colorado, USA). Each siRNA was a mixture of four individual siRNAs sequences that had been validated previously by the manufacturer. Before transfection, cells were washed with Opti-MEM I reduced-serum medium (Invitrogen, Burlington, Ontario, Canada). After this, cells were incubated (37 C, overnight) in reduced-serum medium (15 ml) containing 600 pmol siRNA and 30 l Lipofectamine 2000 (Invitrogen). The transfection solution then was aspirated and the cells were washed with culture medium. Cells were given 36 hours to recover from transfection, and then hypoxic experiments were performed. The amount of siRNA and transfection agent used was optimized, and under the conditions used, we observed 100% transfection under fluorescent microscopy and little to no cell toxicity under light microscopy. Using Western blot techniques, we confirmed that HIF-1␣ and VEGF levels were barely detectable up to 72 hours after siRNA transfection.
FIGURE 2. Bar graph demonstrating differential anti-angiogenic effects of retinal pigment epithelium conditioned media after RNA interference on human umbilical vein endothelial cell (HUVEC) tube formation. After transfection with either random siRNA or siRNA targeting VEGF or HIF-1␣, ARPE-19 cells were incubated under hypoxia (3% 02) for 24 hours, and then the conditioned media was used in a HUVEC tube formation assay. The average number of tubes per microscopic field (ⴛ10 magnification) is shown for the various siRNA treatments. Cells incubated without any siRNA were used as a negative control. Error bars indicate SD.
● ENZYME-LINKED IMMUNOSORBENT ASSAY (ELISA):
Conditioned medium was assayed for angiogenic factors (VEGF, angiogenin, PIGF, EPO, leptin, TGF-1, bFGF, IL-6, IL-8, and MCP-1) using Quantikine human immunoassay kits (R&D Systems, Minneapolis, Minnesota, USA) according to the manufacturers’ instructions. All assays were carried out in triplicate. We also assayed for pigment epitheliumderived factor (PEDF), an important anti-angiogenic factor secreted by RPE cells,4 using Chemikine pigment epithelium-derived factor sandwich enzyme-linked immunosorbent assay (ELISA) kit (Millipore, Billerica, Massachusetts, USA).
graphed using a phase-contrast microscope. Confluent cells of passages two through six were used for the tube formation assay. ● STATISTICAL ANALYSIS:
ELISA and HUVEC assay data were analyzed using a one-way analysis of variance with the Newman-Keuls post hoc test. Post hoc analysis was performed only if the groups were significantly different. Statistical significance was set at .05.
● HUMAN UMBILICAL VEIN ENDOTHELIAL CELL TUBE FORMATION ASSAYS: Human umbilical vein endothe-
RESULTS
lial cells (HUVEC-2) were purchased commercially (BD Biosciences, San Jose, California, USA) and grown in 56.7 cm2 Nunclon culture dishes (Nalge Nunc International) with Medium 200 (Cascade Biologics, Portland, Oregon, USA) supplemented with low serum growth supplement (Cascade Biologics) and penicillin (100 units/ml) and streptomycin (100 g/ml). One day before tube formation assays, medium was replaced with Medium 199 (Sigma, St Louis, Missouri, USA) supplemented with 10% FBS, heparin (100 g/ml), and bFGF (3 ng/ml). Matrigel (8 mg/ml; BD Biosciences) was added to 48-well plates and allowed to gel for two hours at 37 C. HUVECs (5 ⫻ 104) were preincubated with conditioned media from RPE cell cultures for 30 minutes at 37 C before plating. Endothelial cell rearrangement and tube formation was visualized after 16 hours, and the number were counted per low-power (⫻10 magnification) microscopic field as described.28 Each experiment was performed as six replicates (six wells). Representative fields were photoVol. 144, No. 5
RNA INTERFERENCE
AFTER TRANSFECTION OF ARPE-19 CELLS WITH HIF-1␣ RNA,
we observed that several angiogenic factors were decreased significantly compared with the cells treated with random siRNA or no siRNA (Figure 1). VEGF, angiogenin, and TGF-1 levels all were significantly lower (P ⬍ .01) in the conditioned media from cells that were transfected with HIF-1␣ siRNA. The concentrations of these factors, however, were significantly (P ⬍ .001) higher than those achieved with VEGF siRNA. Surprisingly, we observed significantly (P ⬍ .001) elevated levels of IL-6, IL-8, and MCP-1 after HIF-1␣ transfection when compared with cells transfected with random siRNA or no siRNA (Figure 1). After transfection of ARPE-19 cells with VEGF siRNA, we observed that several angiogenic factors were decreased significantly compared with the cells treated with random siRNA (Figure 1). VEGF, angiogenin, IL-6, IL-8, TGF-1, and MCP-1 levels all were significantly lower (P ⬍ .01) in the conditioned media from cells that were transfected OF
VEGF
AND
HIF-1␣
763
FIGURE 3. Representative photographs (ⴛ10 magnification) of HUVEC tube formation using conditioned medium from RPE cells after RNA interference. After transfection with either (Second row) random siRNA, (Third row) siRNA targeting VEGF, or (Fourth row) HIF-1␣, ARPE-19 cells were incubated under hypoxia (3% 02) for 24 hours, and then the conditioned media was used in a HUVEC tube formation assay. Cells incubated without any siRNA were used as a control (First row). The number of tubes per microscopic field was counted and averaged for the various siRNA treatments. Representative photographs of such fields are shown.
with VEGF siRNA. When we compared the concentrations of angiogenic factors after VEGF siRNA transfection with those of cells without any siRNA treatment, the same angiogenic factors were found to be decreased significantly (P ⬍ .01). After transfection of ARPE-19 cells with random siRNA, we observed statistically significant differences in several angiogenic factors (Figure 1). Levels of VEGF, angiogenin, and MCP-1 in the conditioned media were significantly lower (P ⬍ .001) after transfection with random siRNA. Levels of IL-6 and TGF-1 were significantly higher (P ⬍ .01) in the conditioned media after transfection with random siRNA. There was no significant difference in IL-8 in the conditioned media of cells treated with random siRNA and cells without any pretreatment. We did not detect any leptin, EPO, PEDF, bFGF, or PIDF in any of the conditioned media (data not shown). Furthermore, none of the angiogenic factors that we assayed for in this study could be detected in the unconditioned medium (data not shown). 764
AMERICAN JOURNAL
To examine the anti-angiogenic properties of RNAi targeting HIF-1␣ and VEGF, we used the conditioned media from treated and untreated RPE cells in a HUVEC tube formation assay. Conditioned media from RPE cells grown in hypoxia after HIF-1␣ siRNA transfection resulted in significantly less HUVEC tube formation than conditioned medium from hypoxic RPE cells without any siRNA pretreatment (P ⬍ .05; Figure 2). Interestingly, conditioned medium from RPE cells grown in hypoxia after VEGF siRNA transfection seemed to result in even less tube formation when compared with hypoxic RPE conditioned medium (P ⬍ .01; Figure 2). There was no statistically significant difference between HUVEC tube formation between the nontreated hypoxic conditioned medium and the random siRNA conditioned medium. Furthermore, we did not observe a statistically significant difference in tube formation between VEGF siRNA– and HIF-1␣ siRNA–treated conditioned medium. Representative photographs (Figure 3) show impaired HUVEC tube formation after growth in conditioned media from OF
OPHTHALMOLOGY
NOVEMBER 2007
RPE cells pretreated with VEGF siRNA and HIF-1␣ siRNA.
DISCUSSION WE OBSERVED THAT siRNA TARGETING VEGF ATTENUATED
the production of several clinically important angiogenic factors from ARPE-19 cells, including VEGF, angiogenin, IL-6, IL-8, TGF-1, and MCP-1. When the conditioned medium from these cells was used in a HUVEC tube forming assay, we observed markedly diminished angiogenesis compared with conditioned medium from RPE cells without siRNA treatment. These data not only highlight the important angiogenic properties of VEGF, but they also demonstrate the autocrine ability of VEGF to regulate the synthesis of multiple angiogenic proteins. The autocrine effect of VEGF on retinal elements previously has been reported, both in retinal glial cells29 and on RPE cells.30,31 Furthermore, RPE cells have been shown to have VEGF receptors.30,31 A review of the literature reveals interesting insights into how VEGF may be regulating synthesis of multiple angiogenic factors in RPE cells. Nuclear factor– B (NF-B) is an ubiquitous transcription factor that, by regulating the expression of multiple inflammatory and immune genes, plays a critical role in host defense and in chronic inflammatory diseases.32 NF-B is not the only transcription factor involved in regulating these genes, however, and it frequently functions together with other transcription factors, such as activator protein 1 (AP-1), that also are involved in the regulation of inflammatory and immune genes. NF-B and AP-1 both are activated under hypoxic conditions.33 NF-B regulates the transcription of many genes including IL-6, IL-8, and MCP-1.34 –36 Interestingly, it has been demonstrated that VEGF is able to activate NF-B in endothelial cells.37,38 If similar autocrine functions of VEGF are present in RPE cells, this would explain our observation of decreased IL-6, IL-8, and MCP-1 synthesis after VEGF silencing. VEGF also has been demonstrated to interact with HIF-1␣ via an autocrine loop,39 and HIF-1␣ has been shown to be associated with angiogenin and TGF-1 expression.40,41 Diminished signaling via this autocrine loop may explain our observations of decreased angiogenin and TGF-1 after VEGF siRNA. Based on our results, future investigations into the autocrine effects of VEGF on HIF-1␣ and NK-B expression are warranted. We observed that inhibition of HIF-1␣ attenuated the production of some angiogenic factors, while simultaneously causing increased production of other angiogenic factors. Although we observed diminished HUVEC tube formation following HIF-1␣ silencing, this effect was not as great as that observed with VEGF silencing. Presumably, the upregulation of various angiogenic factors (IL-6, IL-8, and MCP-1) after HIF-1␣ silencing may have negated some of the anti-angiogenic effect of HIF-1␣ siRNA. Vol. 144, No. 5
RNA INTERFERENCE
FIGURE 4. Schematic diagrams representing proposed crosstalk between angiogenic factors and transcription factors in human RPE cells. Based on our observations and findings in the literature, the diagrams outline the proposed interaction between angiogenic factors and their transcription factors in human RPE cells under hypoxia (3% 02) with (Top) no siRNA, (Middle) VEGF siRNA, and (Bottom) HIF-1␣ siRNA. White indicates the intracellular space, and grey indicates extracellular space. (Top) Under hypoxic conditions, VEGF normally has autocrine effects that lead to activation of both HIF-1␣ and nuclear factor- B (NF-B). (Middle) When VEGF is diminished using siRNA, these transcription factors are downregulated along with their downstream products (angiogenin [Ang], TGF-1, interleukin [IL]-6, IL-8, and monocyte chemoattractant protein 1 [MCP-1]). (Bottom) When HIF-1␣ is diminished using siRNA, NF-B is upregulated along with its downstream products (IL-6, IL-8, MCP-1). VEGFR ⴝ VEGF receptor.
OF
VEGF
AND
HIF-1␣
765
nonspecific effects of random siRNA did not significantly affect HUVEC tube formation, indicating that the total net result of these nonspecific effects was not deleterious to cell growth. Our in vitro angiogenesis assay, however, did not assess the metabolic changes or long-term effects secondary to siRNA nonspecificity. RNAi may have therapeutic potential in the treatment of ischemic retinal diseases like diabetic retinopathy. There is considerable cross-talk and autocrine effects among angiogenic factors and their transcription factors. In particular, VEGF seems to affect the production of multiple angiogenic factors. A schematic diagram illustrating the proposed cross-talk between angiogenic factors and their transcription factors in human RPE cells, based on our observations and evidence from the literature, is shown in Figure 4. Targeting VEGF seems to have the advantage of decreasing the production of several clinically important angiogenic factors, likely via the interruption of important VEGF autocrine loops. Although HIF-1␣ is important in the induction of several hypoxia-responsive elements, it is only one of many transcription factors that are activated by hypoxia. Inhibition of HIF-1␣ thus may lead to the overactivation of other hypoxia-responsive transcription factors and their respective gene products, leading to unwanted effects and diminished anti-angiogenic activity. For this reason, VEGF may be a superior RNAi target compared with HIF-1␣. An important caveat of RNAi is the nonspecific and unpredictable effects on the production of many angiogenic factors. Future in vivo work on RNAi as a therapy for ocular neovascularization should take into consideration the effects mentioned above.
Although HIF-1␣ plays a major role in controlling the ubiquitous transcriptional response to hypoxia, it is clear that a number of other transcription factors also are activated either directly or indirectly.42 As previously mentioned, NF-B and AP-1 are two important hypoxiadriven transcription factors that regulate the production of many inflammatory mediators.33 The important pathogenic role of both the HIF-1␣ and NF-B/AP-1 systems in diabetic retinopathy is evidenced by their induction with insulin growth factor 1, a factor that is increased in diabetic vitreous and correlates with severity of diabetic retinopathy.43,44 After silencing of HIF-1␣, it is plausible that hypoxic cells are able detect the reduction in downstream hypoxia-adaptive proteins and attempt to compensate for this by upregulating alternate transcription factors (e.g., NF-B and AP-1). Interaction between HIF-1␣ and NF-B has been demonstrated previously.45,46 An upregulation of NF-B may explain our observed increase in IL-6, IL-8, and MCP-1 after HIF-1␣ siRNA. Future studies examining the effects of HIF-1␣ siRNA on NF-B expression likely will shed light on this mechanism. We observed nonspecific effects on the production of several angiogenic factors with random siRNA. Nonspecific effects both at the messenger RNA and protein level are well known to occur with siRNA methods and may represent one of the limitations of this technology.47 Induction of an antiviral cellular response is another known unwanted effect of siRNA.42 Despite this, we still observed significant effects on angiogenic protein production with VEGF- and HIF-1␣-specific siRNA when compared with the random siRNA controls. Furthermore, the
THIS STUDY WAS SUPPORTED BY CIHR GRANT NO. 11535 FROM THE CANADIAN INSTITUTES OF HEALTH RESEARCH, OTTAWA, Ontario, Canada. The authors indicate no financial conflict of interest. Both authors were involved in design and conduct of study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.
REFERENCES 7.
1. Adamis AP, Altaweel M, Bressler NM, et al. Changes in retinal neovascularization after pegaptanib (Macugen) therapy in diabetic individuals. Ophthalmology 2006;113:23–28. 2. Avery RL, Pearlman J, Pieramici DJ, et al. Intravitreal bevacizumab (Avastin) in the treatment of proliferative diabetic retinopathy. Ophthalmology 2006;113:1695 e1– e15. 3. Wilson AS, Hobbs BG, Shen WY, et al. Argon laser photocoagulation-induced modification of gene expression in the retina. Invest Ophthalmol Vis Sci 2003;44: 1426 –34. 4. Obata R, Iriyama A, Inoue Y, Takahashi H, Tamaki Y, Yanagi Y. Triamcinolone acetonide suppresses early proangiogenic response in retinal pigment epithelial cells after photodynamic therapy in vitro. Br J Ophthalmol 2007;91:100 –104. 5. Jonas JB. Intravitreal triamcinolone acetonide for diabetic retinopathy. Dev Ophthalmol 2007;39:96 –110. 6. Bandello F, Brancato R, Menchini U, et al. Light panretinal photocoagulation (LPRP) versus classic panretinal photoco-
766
AMERICAN JOURNAL
8.
9.
10.
11. 12.
OF
agulation (CPRP) in proliferative diabetic retinopathy. Semin Ophthalmol 2001;16:12–18. Watanabe D, Suzuma K, Matsui S, et al. Erythropoietin as a retinal angiogenic factor in proliferative diabetic retinopathy. N Engl J Med 2005;353:782–792. Hernandez C, Fonollosa A, Garcia-Ramirez M, et al. Erythropoietin is expressed in the human retina and it is highly elevated in the vitreous fluid of patients with diabetic macular edema. Diabetes Care 2006;29:2028 –2033. Arjamaa O, Nikinmaa M. Oxygen-dependent diseases in the retina: role of hypoxia-inducible factors. Exp Eye Res 2006; 83:473– 483. Xiao Q, Zeng S, Ling S, Lv M. Up-regulation of HIF-1alpha and VEGF expression by elevated glucose concentration and hypoxia in cultured human retinal pigment epithelial cells. J Huazhong Univ Sci Technolog Med Sci 2006;26:463– 465. Check E. A crucial test. Nat Med 2005;11:243–244. Ozaki H, Hayashi H, Oshima K. Angiogenin levels in the vitreous from patients with proliferative diabetic retinopathy. Ophthalmic Res 1996;28:356 –360.
OPHTHALMOLOGY
NOVEMBER 2007
13. Canataroglu H, Varinli I, Ozcan AA, Canataroglu A, Doran F, Varinli S. Interleukin (IL)-6, interleukin (IL)-8 levels and cellular composition of the vitreous humor in proliferative diabetic retinopathy, proliferative vitreoretinopathy, and traumatic proliferative vitreoretinopathy. Ocul Immunol Inflamm 2005;13:375–381. 14. Hernandez C, Segura RM, Fonollosa A, Carrasco E, Francisco G, Simo R. Interleukin-8, monocyte chemoattractant protein-1 and IL-10 in the vitreous fluid of patients with proliferative diabetic retinopathy. Diabet Med 2005;22:719 – 722. 15. Gariano RF, Nath AK, D’Amico DJ, Lee T, Sierra-Honigmann MR. Elevation of vitreous leptin in diabetic retinopathy and retinal detachment. Invest Ophthalmol Vis Sci 2000;41:3576 –3581. 16. Khaliq A, Foreman D, Ahmed A, et al. Increased expression of placenta growth factor in proliferative diabetic retinopathy. Lab Invest 1998;78:109 –116. 17. Boulton M, Gregor Z, McLeod D, et al. Intravitreal growth factors in proliferative diabetic retinopathy: correlation with neovascular activity and glycemic management. Br J Ophthalmol 1997;81:228 –233. 18. Yu XB, Sun XH, Dahan E, et al. Increased levels of transforming growth factor-betal and -beta2 in the aqueous humor of patients with neovascular glaucoma. Ophthalmic Surg Lasers Imaging 2007;38:6 –14. 19. Funatsu H, Yamashita H, Noma H, Mimura T, Yamashita T, Hori S. Increased levels of vascular endothelial growth factor and interleukin-6 in the aqueous humor of diabetics with macular edema. Am J Ophthalmol 2002;133:70 –77. 20. Funatsu H, Yamashita H, Noma H, et al. Aqueous humor levels of cytokines are related to vitreous levels and progression of diabetic retinopathy in diabetic patients. Graefes Arch Clin Exp Ophthalmol 2005;243:3– 8. 21. Young TA, Wang H, Munk S, et al. Vascular endothelial growth factor expression and secretion by retinal pigment epithelial cells in high glucose and hypoxia is protein kinase C-dependent. Exp Eye Res 2005;80:651– 662. 22. Holtkamp GM, van Rossem M, de Vos AF, Willekens B, Peek R, Kijlstra A. Polarized secretion of IL-6 and IL-8 by human retinal pigment epithelial cells. Clin Exp Immunol 1998;112:34 – 43. 23. Uetama T, Ohno-Matsui K, Nakahama K, Morita I, Mochizuki M. Phenotypic change regulates monocyte chemoattractant protein-1 (MCP-1) gene expression in human retinal pigment epithelial cells. J Cell Physiol 2003;197:77– 85. 24. Bost LM, Aotaki-Keen AE, Hjelmeland LM. Coexpression of FGF-5 and bFGF by the retinal pigment epithelium in vitro. Exp Eye Res 1992;55:727–734. 25. Miyamoto N, de Kozak Y, Jeanny JC, et al. Placental growth factor-1 and epithelial hemato-retinal barrier breakdown: potential implication in the pathogenesis of diabetic retinopathy. Diabetologia 2007;50:461– 470. 26. Nagineni CN, Cherukuri KS, Kutty V, Detrick B, Hooks JJ. Interferon-gamma differentially regulates TGF-beta1 and TGF-beta2 expression in human retinal pigment epithelial cells through JAK-STAT pathway. J Cell Physiol 2007;210: 192–200. 27. Murata M, Takanami T, Shimizu S, et al. Inhibition of ocular angiogenesis by diced small interfering RNAs (siRNAs)
Vol. 144, No. 5
RNA INTERFERENCE
28.
29.
30.
31.
32. 33. 34.
35.
36. 37.
38.
39.
40.
41.
42.
43.
OF
specific to vascular endothelial growth factor (VEGF). Curr Eye Res 2006;31:171–180. Maffucci T, Piccolo E, Cumashi A, et al. Inhibition of the phosphatidylinositol 3-kinase/Akt pathway by inositol pentakisphosphate results in anti-angiogenic and antitumor effects. Cancer Res 2005;65:8339 – 8349. Noda K, Ishida S, Shinoda H, et al. Hypoxia induces the expression of membrane-type 1 matrix metalloproteinase in retinal glial cells. Invest Ophthalmol Vis Sci 2005;46:3817– 3824. Ohno-Matsui K, Yoshida T, Uetama T, Mochizuki M, Morita I. Vascular endothelial growth factor upregulates pigment epithelium-derived factor expression via VEGFR-1 in human retinal pigment epithelial cells. Biochem Biophys Res Commun 2003;303:962–967. Guerrin M, Moukadiri H, Chollet P, et al. Vasculotropin/ vascular endothelial growth factor is an autocrine growth factor for human retinal pigment epithelial cells cultured in vitro. J Cell Physiol 1995;164:385–394. Barnes PJ. Nuclear factor-kappa B. Int J Biochem Cell Biol 1997;29:867– 870. Cummins EP, Taylor CT. Hypoxia-responsive transcription factors. Pflugers Arch 2005;450:363–371. Yoshida A, Yoshida S, Khalil AK, Ishibashi T, Inomata H. Role of NF-kappaB-mediated interleukin-8 expression in intraocular neovascularization. Invest Ophthalmol Vis Sci 1998;39:1097–1106. Barnes PJ, Karin M. Nuclear factor-kappaB: a pivotal transcription factor in chronic inflammatory diseases. N Engl J Med 1997;336:1066 –1071. Baeuerle PA, Baltimore D. NF-kappaB: 10 years after. Cell 1996;87:13–20. Grosjean J, Kiriakidis S, Reilly K, Feldmann M, Paleolog E. Vascular endothelial growth factor signalling in endothelial cell survival: a role for NF-kappaB. Biochem Biophys Res Commun 2006;340:984 –994. Marumo T, Schini-Kerth VB, Busse R. Vascular endothelial growth factor activates nuclear factor-kappaB and induces monocyte chemoattractant protein-1 in bovine retinal endothelial cells. Diabetes 1999;48:1131–1137. Das B, Yeger H, Tsuchida R, et al. A hypoxia-driven vascular endothelial growth factor/Flt1 autocrine loop interacts with hypoxia-inducible factor-1alpha through mitogen-activated protein kinase/extracellular signal-regulated kinase 1/2 pathway in neuroblastoma. Cancer Res 2005;65:7267–7275. Jiang Y, Dai A, Li Q, Hu R. Hypoxia induces transforming growth factor-beta1 gene expression in the pulmonary artery of rats via hypoxia-inducible factor-1alpha. Acta Biochim Biophys Sin (Shanghai) 2007;39:73– 80. Nakamura M, Yamabe H, Osawa H, et al. Hypoxic conditions stimulate the production of angiogenin and vascular endothelial growth factor by human renal proximal tubular epithelial cells in culture. Nephrol Dial Transplant 2006;21: 1489 –1495. Karpala AJ, Doran TJ, Bean AG. Immune responses to dsRNA: implications for gene silencing technologies. Immunol Cell Biol 2005;83:211–216. Meyer-Schwickerath R, Pfeiffer A, Blum WF, et al. Vitreous levels of the insulin-like growth factors I and II, and the
VEGF
AND
HIF-1␣
767
NFkappaB site. Arterioscler Thromb Vasc Biol 2007;27: 755–761. 46. Tacchini L, De Ponti C, Matteucci E, Follis R, Desiderio MA. Hepatocyte growth factor-activated NF-kappaB regulates HIF-1 activity and ODC expression, implicated in survival, differently in different carcinoma cell lines. Carcinogenesis 2004;25:2089 –2100. 47. Jackson AL, Linsley PS. Noise amidst the silence: off-target effects of siRNAs? Trends Genet 2004;20:521–524.
insulin-like growth factor binding proteins 2 and 3, increase in neovascular eye disease. Studies in nondiabetic and diabetic subjects. J Clin Invest 1993;92:2620 –2625. 44. Poulaki V, Joussen AM, Mitsiades N, Mitsiades CS, Iliaki EF, Adamis AP. Insulin-like growth factor-I plays a pathogenetic role in diabetic retinopathy. Am J Pathol 2004;165: 457– 469. 45. Bonello S, Zahringer C, BelAiba RS, et al. Reactive oxygen species activate the HIF-1alpha promoter via a functional
AJO History of Ophthalmology Series Early Stereoscopy
S
ir Charles Wheatstone in June 1838 described “. . . a Stereoscope, to indicate its property of representing solid figures.” In 1849, Sir David Brewster described a binocular camera and the first stereoscopic photographs began to be produced. The stereoscope took off in a big way when Queen Victoria and Prince Albert observed one at the exhibition at the Crystal Palace, and Brewster presented her with a stereoscope made by Jules Duboscq. This signaled the
768
AMERICAN JOURNAL
beginning of a huge trade in stereoscopes and images; it is estimated that by the mid-1850s over a million homes owned one. One of the most successful salesmen of stereoscopic cards was George Nottage, later Lord Mayor of London, with his catalogues listing over one hundred thousand views. Provided by Richard W. Hertle, MD and Richard H. Hertle, MD, of the Cogan Ophthalmic History Society.
OF
OPHTHALMOLOGY
NOVEMBER 2007
To compare the in vitro anti-angiogenic effects of inhibiting vascular endothelial growth factor (VEGF) and hypoxia-inducible factor 1-␣ (HIF-1␣) using ribonucleic acid (RNA) interference (RNAi). ● DESIGN: Laboratory investigation. ● METHODS: VEGF or HIF-1␣ was antagonized in human retinal pigment epithelial (RPE) cells using RNAi, and then cells were cultured under hypoxia. Angiogenic proteins secreted into the media were measured using enzyme-linked immunosorbent assay. Media from hypoxic RPE cells was used to grow human umbilical vein endothelial cells (HUVECs). Capillary tube formation by HUVECs was quantified and compared to assess the effectiveness of angiogenesis. ● RESULTS: RNAi targeting VEGF caused a significant decrease in VEGF in addition to several other clinically important angiogenic factors, including angiogenin, interleukin-6 (IL-6), interleukin-8 (IL-8), monocyte chemoattractant protein-1 (MCP-1), and tumor growth factor 1 (TGF-1). Although HIF-1␣ RNAi reduced the production of VEGF, angiogenin, and TGF-1, we observed an increase in the levels of several other angiogenic factors like IL-6, IL-8, and MCP-1. RNAi of VEGF and HIF-1␣ was effective in inhibiting angiogenesis, although the effect was more pronounced for VEGF RNAi. ● CONCLUSIONS: RNAi of VEGF and HIF-1␣ may have therapeutic potential in ischemic retinal diseases like diabetic retinopathy. Targeting VEGF seems to have the advantage of decreasing the production of several clinically important angiogenic factors, thereby effectively inhibiting angiogenesis. Antagonism of HIF-1␣ may lead to the overactivation of alternate transcription factors and their respective gene products, leading to less effective inhibition of angiogenesis. (Am J Ophthalmol 2007;144:761–768. © 2007 by Elsevier Inc. All rights reserved.) Accepted for publication Jul 16, 2007. From the Department of Ophthalmology and Vision Sciences (F.F.), the Division of Oncology (B.D.), and the Research Institute (F.F., B.D.), The Hospital for Sick Children, Toronto, Ontario, Canada; and the Institute of Medical Science, University of Toronto, Toronto, Ontario, Canada (F.F., B.D.). Inquiries to Farzin Forooghian, Hospital for Sick Children, 555 University Avenue, Room 10128, Elm Wing, Toronto, Ontario, Canada M5G 1X8; e-mail: [email protected] 0002-9394/07/$32.00 doi:10.1016/j.ajo.2007.07.022
©
2007 BY
I
T IS WELL-KNOWN THAT ANGIOGENIC GROWTH FAC-
tors contribute to the pathologic neovascularization that occurs in ischemic retinal diseases like diabetic retinopathy. Vascular endothelial growth factor (VEGF) has been studied extensively in this respect, and monoclonal antibodies and aptamers targeting VEGF currently are being investigated as potential therapies for diabetic retinopathy.1,2 A potential drawback of newly emerging anti-VEGF therapies is that only one of multiple potentially important angiogenic factors is targeted. Laser photocoagulation and intravitreal triamcinolone are two currently used treatment methods for diabetic retinopathy. These treatments are believed to exert their effects partly through the downregulation of multiple angiogenic factors.3,4 These therapeutic methods, however, are not without significant side effects.5,6 Recently, the pathogenic role of erythropoietin (EPO) in diabetic retinopathy has been recognized.7,8 Both VEGF and EPO are regulated by the transcription factor hypoxiainducible factor-1␣ (HIF-1␣), which has been proposed as a novel therapeutic target for ischemic retinal disease.9 With respect to diabetic retinopathy, HIF-1␣ is an attractive target because elevated glucose concentrations have been shown to upregulate HIF-1␣ in human retinal pigment epithelial (RPE) cells.10 Ribonucleic acid (RNA) interference (RNAi) recently emerged as a potentially exciting therapeutic method for various disease states. A key therapeutic advantage to using RNAi lies in its ability to knock down the expression of disease-causing genes of known sequence specifically and potently. Interestingly, the first clinical application involving RNAi has been directed at the treatment of age-related macular degeneration. Clinical trials involving RNAi targeting VEGF or its receptor through intravitreal injection delivery currently are underway.11 The eye is a unique organ in terms of allowing direct drug delivery, thereby overcoming many of the difficulties associated with systemic RNAi delivery.11 Investigations into the therapeutic potential of RNAi for other blinding ocular diseases such as diabetic retinopathy thus are not only interesting, but also are highly warranted. In addition to VEGF and EPO, numerous other angiogenic factors are clinically important in diabetic retinopathy. Angiogenin, interleukin-6 (IL-6), interleukin-8 (IL-8), monocyte chemoattractant protein-1 (MCP-1), leptin, placental
ELSEVIER INC. ALL
RIGHTS RESERVED.
761
FIGURE 1. Bar graphs demonstrating decreased expression of angiogenic factors by retinal pigment epithelial (RPE) cells after ribonucleic acid (RNA) interference. After transfection with either random short interfering RNA (siRNA) or siRNA targeting vascular endothelial growth factor (VEGF) or hypoxia-inducible factor-1␣ (HIF-1␣), human RPE (ARPE-19) cells were incubated under hypoxia (3% 02) for 24 hours and then conditioned media were assayed using enzyme-linked immunosorbent assay. Cells incubated without any siRNA were used as a negative control. Error bars indicate standard deviation (SD). All pairwise comparisons between histogram bars are statistically significant (P < .01), except for those with an asterisk (*). IL ⴝ interleukin; MCP-1 ⴝ monocyte chemoattractant protein 1; TGF-1 ⴝ transforming growth factor 1.
growth factor (PIGF), transforming growth factor-1 (TGF-1), and basic fibroblast growth factor (bFGF) all have been found to be elevated in the vitreous of patients with diabetic retinopathy.12–18 Furthermore, the severity of diabetic retinopathy has been correlated to vitreous levels of IL-6, IL-8, MCP-1, and bFGF.17,19,20 Any study that attempts to address the anti-angiogenic properties of a novel therapeutic method therefore must examine these angiogenic factors as well. The RPE is an abundant source of angiogenic factors in the retina. In addition to VEGF,21 these cells have been shown to express IL-6, IL-8, MCP-1, bFGF, TGF-1, and PIGF under various conditions.22–26 RPE cells thus are an attractive therapeutic target. RNAi in the form of short interfering RNA (siRNA) targeting VEGF previously has been shown to inhibit VEGF production by human RPE cells,27 demonstrating that these cells are amenable to siRNA transfection. Although VEGF is an attractive target for RNAi experiments on ocular angiogenesis, other targets need to be considered as well. The effects of HIF-1␣ inhibition on ischemia-induced retinal angiogenesis are not yet known. The purpose of this study was to compare the antiangiogenic properties of RNAi targeting HIF-1␣ with that of RNAi targeting VEGF in an in vitro model of ischemiainduced retinal angiogenesis. 762
AMERICAN JOURNAL
METHODS ● RETINAL PIGMENT EPITHELIAL CELL CULTURE:
Human RPE cells (ARPE-19) were purchased commercially (ATCC, Manassas, Virginia, USA). Cells were cultured (37 C, 5% CO2) to confluence in 56.7 cm2 Nunclon culture dishes (Nalge Nunc International, Rochester, New York, USA) with D-MEM/F-12 growth medium (15 ml; ATCC) supplemented with 10% heat-inactivated fetal bovine serum (FBS) and penicillin (100 units/ml) and streptomycin (100 g/ml). For hypoxia experiments, the medium was replaced with D-MEM/F-12 growth medium (15 ml) supplemented with penicillin (100 units/ml) and streptomycin (100 g/ml) only. After 24 hours of growth in a hypoxic incubator (37 C, 3% O2), culture medium was removed and centrifuged to remove any cellular debris, and the supernatant was frozen at ⫺80 C until further use. Conditioned medium from two separate experiments was pooled together before freezing at ⫺80 C. All experiments were carried out using cells within the first five passages. Observation of the cells by light microscopy after over 36 hours of hypoxia demonstrated that they were viable and healthy.
● siRNA TRANSFECTION:
Cells were grown to 90% confluence for transfection. For siRNA experiments, ONTARGETplus SMARTpool HIF-1␣ siRNA or VEGF
OF
OPHTHALMOLOGY
NOVEMBER 2007
siRNA was used, whereas fluorescent siGLO RISC-free siRNA was used as a negative control (Dharmacon, Lafayette, Colorado, USA). Each siRNA was a mixture of four individual siRNAs sequences that had been validated previously by the manufacturer. Before transfection, cells were washed with Opti-MEM I reduced-serum medium (Invitrogen, Burlington, Ontario, Canada). After this, cells were incubated (37 C, overnight) in reduced-serum medium (15 ml) containing 600 pmol siRNA and 30 l Lipofectamine 2000 (Invitrogen). The transfection solution then was aspirated and the cells were washed with culture medium. Cells were given 36 hours to recover from transfection, and then hypoxic experiments were performed. The amount of siRNA and transfection agent used was optimized, and under the conditions used, we observed 100% transfection under fluorescent microscopy and little to no cell toxicity under light microscopy. Using Western blot techniques, we confirmed that HIF-1␣ and VEGF levels were barely detectable up to 72 hours after siRNA transfection.
FIGURE 2. Bar graph demonstrating differential anti-angiogenic effects of retinal pigment epithelium conditioned media after RNA interference on human umbilical vein endothelial cell (HUVEC) tube formation. After transfection with either random siRNA or siRNA targeting VEGF or HIF-1␣, ARPE-19 cells were incubated under hypoxia (3% 02) for 24 hours, and then the conditioned media was used in a HUVEC tube formation assay. The average number of tubes per microscopic field (ⴛ10 magnification) is shown for the various siRNA treatments. Cells incubated without any siRNA were used as a negative control. Error bars indicate SD.
● ENZYME-LINKED IMMUNOSORBENT ASSAY (ELISA):
Conditioned medium was assayed for angiogenic factors (VEGF, angiogenin, PIGF, EPO, leptin, TGF-1, bFGF, IL-6, IL-8, and MCP-1) using Quantikine human immunoassay kits (R&D Systems, Minneapolis, Minnesota, USA) according to the manufacturers’ instructions. All assays were carried out in triplicate. We also assayed for pigment epitheliumderived factor (PEDF), an important anti-angiogenic factor secreted by RPE cells,4 using Chemikine pigment epithelium-derived factor sandwich enzyme-linked immunosorbent assay (ELISA) kit (Millipore, Billerica, Massachusetts, USA).
graphed using a phase-contrast microscope. Confluent cells of passages two through six were used for the tube formation assay. ● STATISTICAL ANALYSIS:
ELISA and HUVEC assay data were analyzed using a one-way analysis of variance with the Newman-Keuls post hoc test. Post hoc analysis was performed only if the groups were significantly different. Statistical significance was set at .05.
● HUMAN UMBILICAL VEIN ENDOTHELIAL CELL TUBE FORMATION ASSAYS: Human umbilical vein endothe-
RESULTS
lial cells (HUVEC-2) were purchased commercially (BD Biosciences, San Jose, California, USA) and grown in 56.7 cm2 Nunclon culture dishes (Nalge Nunc International) with Medium 200 (Cascade Biologics, Portland, Oregon, USA) supplemented with low serum growth supplement (Cascade Biologics) and penicillin (100 units/ml) and streptomycin (100 g/ml). One day before tube formation assays, medium was replaced with Medium 199 (Sigma, St Louis, Missouri, USA) supplemented with 10% FBS, heparin (100 g/ml), and bFGF (3 ng/ml). Matrigel (8 mg/ml; BD Biosciences) was added to 48-well plates and allowed to gel for two hours at 37 C. HUVECs (5 ⫻ 104) were preincubated with conditioned media from RPE cell cultures for 30 minutes at 37 C before plating. Endothelial cell rearrangement and tube formation was visualized after 16 hours, and the number were counted per low-power (⫻10 magnification) microscopic field as described.28 Each experiment was performed as six replicates (six wells). Representative fields were photoVol. 144, No. 5
RNA INTERFERENCE
AFTER TRANSFECTION OF ARPE-19 CELLS WITH HIF-1␣ RNA,
we observed that several angiogenic factors were decreased significantly compared with the cells treated with random siRNA or no siRNA (Figure 1). VEGF, angiogenin, and TGF-1 levels all were significantly lower (P ⬍ .01) in the conditioned media from cells that were transfected with HIF-1␣ siRNA. The concentrations of these factors, however, were significantly (P ⬍ .001) higher than those achieved with VEGF siRNA. Surprisingly, we observed significantly (P ⬍ .001) elevated levels of IL-6, IL-8, and MCP-1 after HIF-1␣ transfection when compared with cells transfected with random siRNA or no siRNA (Figure 1). After transfection of ARPE-19 cells with VEGF siRNA, we observed that several angiogenic factors were decreased significantly compared with the cells treated with random siRNA (Figure 1). VEGF, angiogenin, IL-6, IL-8, TGF-1, and MCP-1 levels all were significantly lower (P ⬍ .01) in the conditioned media from cells that were transfected OF
VEGF
AND
HIF-1␣
763
FIGURE 3. Representative photographs (ⴛ10 magnification) of HUVEC tube formation using conditioned medium from RPE cells after RNA interference. After transfection with either (Second row) random siRNA, (Third row) siRNA targeting VEGF, or (Fourth row) HIF-1␣, ARPE-19 cells were incubated under hypoxia (3% 02) for 24 hours, and then the conditioned media was used in a HUVEC tube formation assay. Cells incubated without any siRNA were used as a control (First row). The number of tubes per microscopic field was counted and averaged for the various siRNA treatments. Representative photographs of such fields are shown.
with VEGF siRNA. When we compared the concentrations of angiogenic factors after VEGF siRNA transfection with those of cells without any siRNA treatment, the same angiogenic factors were found to be decreased significantly (P ⬍ .01). After transfection of ARPE-19 cells with random siRNA, we observed statistically significant differences in several angiogenic factors (Figure 1). Levels of VEGF, angiogenin, and MCP-1 in the conditioned media were significantly lower (P ⬍ .001) after transfection with random siRNA. Levels of IL-6 and TGF-1 were significantly higher (P ⬍ .01) in the conditioned media after transfection with random siRNA. There was no significant difference in IL-8 in the conditioned media of cells treated with random siRNA and cells without any pretreatment. We did not detect any leptin, EPO, PEDF, bFGF, or PIDF in any of the conditioned media (data not shown). Furthermore, none of the angiogenic factors that we assayed for in this study could be detected in the unconditioned medium (data not shown). 764
AMERICAN JOURNAL
To examine the anti-angiogenic properties of RNAi targeting HIF-1␣ and VEGF, we used the conditioned media from treated and untreated RPE cells in a HUVEC tube formation assay. Conditioned media from RPE cells grown in hypoxia after HIF-1␣ siRNA transfection resulted in significantly less HUVEC tube formation than conditioned medium from hypoxic RPE cells without any siRNA pretreatment (P ⬍ .05; Figure 2). Interestingly, conditioned medium from RPE cells grown in hypoxia after VEGF siRNA transfection seemed to result in even less tube formation when compared with hypoxic RPE conditioned medium (P ⬍ .01; Figure 2). There was no statistically significant difference between HUVEC tube formation between the nontreated hypoxic conditioned medium and the random siRNA conditioned medium. Furthermore, we did not observe a statistically significant difference in tube formation between VEGF siRNA– and HIF-1␣ siRNA–treated conditioned medium. Representative photographs (Figure 3) show impaired HUVEC tube formation after growth in conditioned media from OF
OPHTHALMOLOGY
NOVEMBER 2007
RPE cells pretreated with VEGF siRNA and HIF-1␣ siRNA.
DISCUSSION WE OBSERVED THAT siRNA TARGETING VEGF ATTENUATED
the production of several clinically important angiogenic factors from ARPE-19 cells, including VEGF, angiogenin, IL-6, IL-8, TGF-1, and MCP-1. When the conditioned medium from these cells was used in a HUVEC tube forming assay, we observed markedly diminished angiogenesis compared with conditioned medium from RPE cells without siRNA treatment. These data not only highlight the important angiogenic properties of VEGF, but they also demonstrate the autocrine ability of VEGF to regulate the synthesis of multiple angiogenic proteins. The autocrine effect of VEGF on retinal elements previously has been reported, both in retinal glial cells29 and on RPE cells.30,31 Furthermore, RPE cells have been shown to have VEGF receptors.30,31 A review of the literature reveals interesting insights into how VEGF may be regulating synthesis of multiple angiogenic factors in RPE cells. Nuclear factor– B (NF-B) is an ubiquitous transcription factor that, by regulating the expression of multiple inflammatory and immune genes, plays a critical role in host defense and in chronic inflammatory diseases.32 NF-B is not the only transcription factor involved in regulating these genes, however, and it frequently functions together with other transcription factors, such as activator protein 1 (AP-1), that also are involved in the regulation of inflammatory and immune genes. NF-B and AP-1 both are activated under hypoxic conditions.33 NF-B regulates the transcription of many genes including IL-6, IL-8, and MCP-1.34 –36 Interestingly, it has been demonstrated that VEGF is able to activate NF-B in endothelial cells.37,38 If similar autocrine functions of VEGF are present in RPE cells, this would explain our observation of decreased IL-6, IL-8, and MCP-1 synthesis after VEGF silencing. VEGF also has been demonstrated to interact with HIF-1␣ via an autocrine loop,39 and HIF-1␣ has been shown to be associated with angiogenin and TGF-1 expression.40,41 Diminished signaling via this autocrine loop may explain our observations of decreased angiogenin and TGF-1 after VEGF siRNA. Based on our results, future investigations into the autocrine effects of VEGF on HIF-1␣ and NK-B expression are warranted. We observed that inhibition of HIF-1␣ attenuated the production of some angiogenic factors, while simultaneously causing increased production of other angiogenic factors. Although we observed diminished HUVEC tube formation following HIF-1␣ silencing, this effect was not as great as that observed with VEGF silencing. Presumably, the upregulation of various angiogenic factors (IL-6, IL-8, and MCP-1) after HIF-1␣ silencing may have negated some of the anti-angiogenic effect of HIF-1␣ siRNA. Vol. 144, No. 5
RNA INTERFERENCE
FIGURE 4. Schematic diagrams representing proposed crosstalk between angiogenic factors and transcription factors in human RPE cells. Based on our observations and findings in the literature, the diagrams outline the proposed interaction between angiogenic factors and their transcription factors in human RPE cells under hypoxia (3% 02) with (Top) no siRNA, (Middle) VEGF siRNA, and (Bottom) HIF-1␣ siRNA. White indicates the intracellular space, and grey indicates extracellular space. (Top) Under hypoxic conditions, VEGF normally has autocrine effects that lead to activation of both HIF-1␣ and nuclear factor- B (NF-B). (Middle) When VEGF is diminished using siRNA, these transcription factors are downregulated along with their downstream products (angiogenin [Ang], TGF-1, interleukin [IL]-6, IL-8, and monocyte chemoattractant protein 1 [MCP-1]). (Bottom) When HIF-1␣ is diminished using siRNA, NF-B is upregulated along with its downstream products (IL-6, IL-8, MCP-1). VEGFR ⴝ VEGF receptor.
OF
VEGF
AND
HIF-1␣
765
nonspecific effects of random siRNA did not significantly affect HUVEC tube formation, indicating that the total net result of these nonspecific effects was not deleterious to cell growth. Our in vitro angiogenesis assay, however, did not assess the metabolic changes or long-term effects secondary to siRNA nonspecificity. RNAi may have therapeutic potential in the treatment of ischemic retinal diseases like diabetic retinopathy. There is considerable cross-talk and autocrine effects among angiogenic factors and their transcription factors. In particular, VEGF seems to affect the production of multiple angiogenic factors. A schematic diagram illustrating the proposed cross-talk between angiogenic factors and their transcription factors in human RPE cells, based on our observations and evidence from the literature, is shown in Figure 4. Targeting VEGF seems to have the advantage of decreasing the production of several clinically important angiogenic factors, likely via the interruption of important VEGF autocrine loops. Although HIF-1␣ is important in the induction of several hypoxia-responsive elements, it is only one of many transcription factors that are activated by hypoxia. Inhibition of HIF-1␣ thus may lead to the overactivation of other hypoxia-responsive transcription factors and their respective gene products, leading to unwanted effects and diminished anti-angiogenic activity. For this reason, VEGF may be a superior RNAi target compared with HIF-1␣. An important caveat of RNAi is the nonspecific and unpredictable effects on the production of many angiogenic factors. Future in vivo work on RNAi as a therapy for ocular neovascularization should take into consideration the effects mentioned above.
Although HIF-1␣ plays a major role in controlling the ubiquitous transcriptional response to hypoxia, it is clear that a number of other transcription factors also are activated either directly or indirectly.42 As previously mentioned, NF-B and AP-1 are two important hypoxiadriven transcription factors that regulate the production of many inflammatory mediators.33 The important pathogenic role of both the HIF-1␣ and NF-B/AP-1 systems in diabetic retinopathy is evidenced by their induction with insulin growth factor 1, a factor that is increased in diabetic vitreous and correlates with severity of diabetic retinopathy.43,44 After silencing of HIF-1␣, it is plausible that hypoxic cells are able detect the reduction in downstream hypoxia-adaptive proteins and attempt to compensate for this by upregulating alternate transcription factors (e.g., NF-B and AP-1). Interaction between HIF-1␣ and NF-B has been demonstrated previously.45,46 An upregulation of NF-B may explain our observed increase in IL-6, IL-8, and MCP-1 after HIF-1␣ siRNA. Future studies examining the effects of HIF-1␣ siRNA on NF-B expression likely will shed light on this mechanism. We observed nonspecific effects on the production of several angiogenic factors with random siRNA. Nonspecific effects both at the messenger RNA and protein level are well known to occur with siRNA methods and may represent one of the limitations of this technology.47 Induction of an antiviral cellular response is another known unwanted effect of siRNA.42 Despite this, we still observed significant effects on angiogenic protein production with VEGF- and HIF-1␣-specific siRNA when compared with the random siRNA controls. Furthermore, the
THIS STUDY WAS SUPPORTED BY CIHR GRANT NO. 11535 FROM THE CANADIAN INSTITUTES OF HEALTH RESEARCH, OTTAWA, Ontario, Canada. The authors indicate no financial conflict of interest. Both authors were involved in design and conduct of study; collection, management, analysis, and interpretation of the data; and preparation, review, or approval of the manuscript.
REFERENCES 7.
1. Adamis AP, Altaweel M, Bressler NM, et al. Changes in retinal neovascularization after pegaptanib (Macugen) therapy in diabetic individuals. Ophthalmology 2006;113:23–28. 2. Avery RL, Pearlman J, Pieramici DJ, et al. Intravitreal bevacizumab (Avastin) in the treatment of proliferative diabetic retinopathy. Ophthalmology 2006;113:1695 e1– e15. 3. Wilson AS, Hobbs BG, Shen WY, et al. Argon laser photocoagulation-induced modification of gene expression in the retina. Invest Ophthalmol Vis Sci 2003;44: 1426 –34. 4. Obata R, Iriyama A, Inoue Y, Takahashi H, Tamaki Y, Yanagi Y. Triamcinolone acetonide suppresses early proangiogenic response in retinal pigment epithelial cells after photodynamic therapy in vitro. Br J Ophthalmol 2007;91:100 –104. 5. Jonas JB. Intravitreal triamcinolone acetonide for diabetic retinopathy. Dev Ophthalmol 2007;39:96 –110. 6. Bandello F, Brancato R, Menchini U, et al. Light panretinal photocoagulation (LPRP) versus classic panretinal photoco-
766
AMERICAN JOURNAL
8.
9.
10.
11. 12.
OF
agulation (CPRP) in proliferative diabetic retinopathy. Semin Ophthalmol 2001;16:12–18. Watanabe D, Suzuma K, Matsui S, et al. Erythropoietin as a retinal angiogenic factor in proliferative diabetic retinopathy. N Engl J Med 2005;353:782–792. Hernandez C, Fonollosa A, Garcia-Ramirez M, et al. Erythropoietin is expressed in the human retina and it is highly elevated in the vitreous fluid of patients with diabetic macular edema. Diabetes Care 2006;29:2028 –2033. Arjamaa O, Nikinmaa M. Oxygen-dependent diseases in the retina: role of hypoxia-inducible factors. Exp Eye Res 2006; 83:473– 483. Xiao Q, Zeng S, Ling S, Lv M. Up-regulation of HIF-1alpha and VEGF expression by elevated glucose concentration and hypoxia in cultured human retinal pigment epithelial cells. J Huazhong Univ Sci Technolog Med Sci 2006;26:463– 465. Check E. A crucial test. Nat Med 2005;11:243–244. Ozaki H, Hayashi H, Oshima K. Angiogenin levels in the vitreous from patients with proliferative diabetic retinopathy. Ophthalmic Res 1996;28:356 –360.
OPHTHALMOLOGY
NOVEMBER 2007
13. Canataroglu H, Varinli I, Ozcan AA, Canataroglu A, Doran F, Varinli S. Interleukin (IL)-6, interleukin (IL)-8 levels and cellular composition of the vitreous humor in proliferative diabetic retinopathy, proliferative vitreoretinopathy, and traumatic proliferative vitreoretinopathy. Ocul Immunol Inflamm 2005;13:375–381. 14. Hernandez C, Segura RM, Fonollosa A, Carrasco E, Francisco G, Simo R. Interleukin-8, monocyte chemoattractant protein-1 and IL-10 in the vitreous fluid of patients with proliferative diabetic retinopathy. Diabet Med 2005;22:719 – 722. 15. Gariano RF, Nath AK, D’Amico DJ, Lee T, Sierra-Honigmann MR. Elevation of vitreous leptin in diabetic retinopathy and retinal detachment. Invest Ophthalmol Vis Sci 2000;41:3576 –3581. 16. Khaliq A, Foreman D, Ahmed A, et al. Increased expression of placenta growth factor in proliferative diabetic retinopathy. Lab Invest 1998;78:109 –116. 17. Boulton M, Gregor Z, McLeod D, et al. Intravitreal growth factors in proliferative diabetic retinopathy: correlation with neovascular activity and glycemic management. Br J Ophthalmol 1997;81:228 –233. 18. Yu XB, Sun XH, Dahan E, et al. Increased levels of transforming growth factor-betal and -beta2 in the aqueous humor of patients with neovascular glaucoma. Ophthalmic Surg Lasers Imaging 2007;38:6 –14. 19. Funatsu H, Yamashita H, Noma H, Mimura T, Yamashita T, Hori S. Increased levels of vascular endothelial growth factor and interleukin-6 in the aqueous humor of diabetics with macular edema. Am J Ophthalmol 2002;133:70 –77. 20. Funatsu H, Yamashita H, Noma H, et al. Aqueous humor levels of cytokines are related to vitreous levels and progression of diabetic retinopathy in diabetic patients. Graefes Arch Clin Exp Ophthalmol 2005;243:3– 8. 21. Young TA, Wang H, Munk S, et al. Vascular endothelial growth factor expression and secretion by retinal pigment epithelial cells in high glucose and hypoxia is protein kinase C-dependent. Exp Eye Res 2005;80:651– 662. 22. Holtkamp GM, van Rossem M, de Vos AF, Willekens B, Peek R, Kijlstra A. Polarized secretion of IL-6 and IL-8 by human retinal pigment epithelial cells. Clin Exp Immunol 1998;112:34 – 43. 23. Uetama T, Ohno-Matsui K, Nakahama K, Morita I, Mochizuki M. Phenotypic change regulates monocyte chemoattractant protein-1 (MCP-1) gene expression in human retinal pigment epithelial cells. J Cell Physiol 2003;197:77– 85. 24. Bost LM, Aotaki-Keen AE, Hjelmeland LM. Coexpression of FGF-5 and bFGF by the retinal pigment epithelium in vitro. Exp Eye Res 1992;55:727–734. 25. Miyamoto N, de Kozak Y, Jeanny JC, et al. Placental growth factor-1 and epithelial hemato-retinal barrier breakdown: potential implication in the pathogenesis of diabetic retinopathy. Diabetologia 2007;50:461– 470. 26. Nagineni CN, Cherukuri KS, Kutty V, Detrick B, Hooks JJ. Interferon-gamma differentially regulates TGF-beta1 and TGF-beta2 expression in human retinal pigment epithelial cells through JAK-STAT pathway. J Cell Physiol 2007;210: 192–200. 27. Murata M, Takanami T, Shimizu S, et al. Inhibition of ocular angiogenesis by diced small interfering RNAs (siRNAs)
Vol. 144, No. 5
RNA INTERFERENCE
28.
29.
30.
31.
32. 33. 34.
35.
36. 37.
38.
39.
40.
41.
42.
43.
OF
specific to vascular endothelial growth factor (VEGF). Curr Eye Res 2006;31:171–180. Maffucci T, Piccolo E, Cumashi A, et al. Inhibition of the phosphatidylinositol 3-kinase/Akt pathway by inositol pentakisphosphate results in anti-angiogenic and antitumor effects. Cancer Res 2005;65:8339 – 8349. Noda K, Ishida S, Shinoda H, et al. Hypoxia induces the expression of membrane-type 1 matrix metalloproteinase in retinal glial cells. Invest Ophthalmol Vis Sci 2005;46:3817– 3824. Ohno-Matsui K, Yoshida T, Uetama T, Mochizuki M, Morita I. Vascular endothelial growth factor upregulates pigment epithelium-derived factor expression via VEGFR-1 in human retinal pigment epithelial cells. Biochem Biophys Res Commun 2003;303:962–967. Guerrin M, Moukadiri H, Chollet P, et al. Vasculotropin/ vascular endothelial growth factor is an autocrine growth factor for human retinal pigment epithelial cells cultured in vitro. J Cell Physiol 1995;164:385–394. Barnes PJ. Nuclear factor-kappa B. Int J Biochem Cell Biol 1997;29:867– 870. Cummins EP, Taylor CT. Hypoxia-responsive transcription factors. Pflugers Arch 2005;450:363–371. Yoshida A, Yoshida S, Khalil AK, Ishibashi T, Inomata H. Role of NF-kappaB-mediated interleukin-8 expression in intraocular neovascularization. Invest Ophthalmol Vis Sci 1998;39:1097–1106. Barnes PJ, Karin M. Nuclear factor-kappaB: a pivotal transcription factor in chronic inflammatory diseases. N Engl J Med 1997;336:1066 –1071. Baeuerle PA, Baltimore D. NF-kappaB: 10 years after. Cell 1996;87:13–20. Grosjean J, Kiriakidis S, Reilly K, Feldmann M, Paleolog E. Vascular endothelial growth factor signalling in endothelial cell survival: a role for NF-kappaB. Biochem Biophys Res Commun 2006;340:984 –994. Marumo T, Schini-Kerth VB, Busse R. Vascular endothelial growth factor activates nuclear factor-kappaB and induces monocyte chemoattractant protein-1 in bovine retinal endothelial cells. Diabetes 1999;48:1131–1137. Das B, Yeger H, Tsuchida R, et al. A hypoxia-driven vascular endothelial growth factor/Flt1 autocrine loop interacts with hypoxia-inducible factor-1alpha through mitogen-activated protein kinase/extracellular signal-regulated kinase 1/2 pathway in neuroblastoma. Cancer Res 2005;65:7267–7275. Jiang Y, Dai A, Li Q, Hu R. Hypoxia induces transforming growth factor-beta1 gene expression in the pulmonary artery of rats via hypoxia-inducible factor-1alpha. Acta Biochim Biophys Sin (Shanghai) 2007;39:73– 80. Nakamura M, Yamabe H, Osawa H, et al. Hypoxic conditions stimulate the production of angiogenin and vascular endothelial growth factor by human renal proximal tubular epithelial cells in culture. Nephrol Dial Transplant 2006;21: 1489 –1495. Karpala AJ, Doran TJ, Bean AG. Immune responses to dsRNA: implications for gene silencing technologies. Immunol Cell Biol 2005;83:211–216. Meyer-Schwickerath R, Pfeiffer A, Blum WF, et al. Vitreous levels of the insulin-like growth factors I and II, and the
VEGF
AND
HIF-1␣
767
NFkappaB site. Arterioscler Thromb Vasc Biol 2007;27: 755–761. 46. Tacchini L, De Ponti C, Matteucci E, Follis R, Desiderio MA. Hepatocyte growth factor-activated NF-kappaB regulates HIF-1 activity and ODC expression, implicated in survival, differently in different carcinoma cell lines. Carcinogenesis 2004;25:2089 –2100. 47. Jackson AL, Linsley PS. Noise amidst the silence: off-target effects of siRNAs? Trends Genet 2004;20:521–524.
insulin-like growth factor binding proteins 2 and 3, increase in neovascular eye disease. Studies in nondiabetic and diabetic subjects. J Clin Invest 1993;92:2620 –2625. 44. Poulaki V, Joussen AM, Mitsiades N, Mitsiades CS, Iliaki EF, Adamis AP. Insulin-like growth factor-I plays a pathogenetic role in diabetic retinopathy. Am J Pathol 2004;165: 457– 469. 45. Bonello S, Zahringer C, BelAiba RS, et al. Reactive oxygen species activate the HIF-1alpha promoter via a functional
AJO History of Ophthalmology Series Early Stereoscopy
S
ir Charles Wheatstone in June 1838 described “. . . a Stereoscope, to indicate its property of representing solid figures.” In 1849, Sir David Brewster described a binocular camera and the first stereoscopic photographs began to be produced. The stereoscope took off in a big way when Queen Victoria and Prince Albert observed one at the exhibition at the Crystal Palace, and Brewster presented her with a stereoscope made by Jules Duboscq. This signaled the
768
AMERICAN JOURNAL
beginning of a huge trade in stereoscopes and images; it is estimated that by the mid-1850s over a million homes owned one. One of the most successful salesmen of stereoscopic cards was George Nottage, later Lord Mayor of London, with his catalogues listing over one hundred thousand views. Provided by Richard W. Hertle, MD and Richard H. Hertle, MD, of the Cogan Ophthalmic History Society.
OF
OPHTHALMOLOGY
NOVEMBER 2007