Int6/eIF3e Silencing Promotes Functional Blood Vessel Outgrowth and Enhances Wound Healing by Upregulating Hypoxia-Induced Factor 2α Expression
Background— We previously identified INT6/eIF3e as a novel regulator of hypoxia-inducible factor 2α (HIF2α) activity. Small interfering RNA (siRNA)–Int6 adequately stabilized HIF2α, even under normoxic conditions, and thereby enhanced the expression of several angiogenic factors in vitro, suggesting that siRNA-Int6 may induce angiogenesis in vivo.
Methods and Results— We demonstrated a 6- to 8-fold enhanced formation of normal arteries and veins in the subcutaneous regions of adult mice 5 days after a single siRNA-Int6 application. Subcutaneous fibroblasts were identified as the major source of secreted angiogenic factors that led to the formation of functional vessels during Int6 silencing. Fibroblasts transfected ex vivo with siRNA-Int6 induced potent neoangiogenesis when transplanted into a subcutaneous region of nude mice. Application of siRNA-Int6 promoted neoangiogenesis in the area surrounding the injury in wound healing models, including genetically diabetic mice, thereby accelerating the closure of the injury. HIF2α accumulation caused by siRNA-Int6 was confirmed as the unequivocal cause of the angiogenesis by an in vivo angiogenesis assay. Further analysis of the Int6 silencing–induced neoangiogenesis revealed that a negative feedback regulation of HIF2α stability was caused by HIF2α-induced transcription of Int6 via hypoxia-response elements in its promoter. Thus, siRNA-Int6 temporarily facilitates an accumulation of HIF2α protein, leading to hypoxia-independent transcription of angiogenic factors and concomitant neoangiogenesis.
Conclusions— We suggest that the pathway involving INT6/HIF2α acts as a hypoxia-independent master switch of functional angiogenesis; therefore, siRNA-Int6 application might be of clinical value in treating ischemic diseases such as heart and brain ischemia, skin injury, and diseases involving obstructed vessels.
Received December 19, 2009; accepted June 18, 2010.
The process of vessel formation is complex but well coordinated, involving the combined action of numerous growth factors and related signaling pathways.1 Nevertheless, a single angiogenic factor such as vascular endothelial growth factor (VEGF),2 fibroblast growth factor (FGF),3,4 or platelet-derived growth factor5 can induce neoangiogenesis,6–8 albeit with incomplete and leaky vessels.6,9 The transgenic expression of angiopoietin-1 (ANG-1) and VEGF significantly increases both the size and number of blood vessels.8 These results suggest that several angiogenic factors are essential for the formation of functional vessels and that they must be expressed in a complementary and coordinated manner10 to strike a balance among many stimulatory and inhibitory signals.11
Clinical Perspective on p 919
The expression of various angiogenic factors such as VEGF, ANG-1, and pleiotrophin12 is triggered by hypoxia through the action of hypoxia-inducible factors (HIFs); these angiogenic factors play important roles in blood vessel formation and oncogenesis.13 During hypoxia, HIF1α and HIF2α regulate the expression of at least 150 genes involved in metabolism, cell survival, erythropoiesis, and vascular remodeling13,14 by binding to cis-acting hypoxia-response elements (HREs) in the promoters of these genes.15–18 Under normoxic conditions, the subunits of HIFα are rapidly ubiquitinated and degraded through their direct interaction with an E3 ubiquitin ligase complex containing the von Hippel-Lindau tumor suppressor protein (pVHL).19,20 The different phenotypes of Hif1α-18,21 and Hif2α-knockout mice,22–24 and the nonoverlapping expression patterns of these proteins,25 which may contribute to mutant lethality even if the proteins regulate similar genes, have raised questions about the individual roles of each of these HIFαs.
We previously reported that the tumor suppressor INT6/eIF3e binds in a subtype-specific manner to HIF2α and is involved in HIF2α regulation.26 The Int6 gene was first identified as a frequent integration site of the mouse mammary tumor virus in preneoplastic and neoplastic mammary lesions.27 It was later characterized as the translation initiation factor subunit e (eIF3e) in rabbits,28 Schizosaccharomyces pombe,29–31 and Arabidopsis thaliana32,33; however, little is known about the functional role of eIF3e. Unlike the mechanism of HIF1α degradation, HIF2α degradation depends not only on an intact pVHL but also on the Int6/proteasome pathway.26 When an siRNA against Int6 (siRNA-Int6) was used, HIF2α activity was stabilized even under normoxic conditions, and the expression of several angiogenic factors such as ANG-1, basic FGF (bFGF), and VEGF in HeLa and MCF-7 cells subsequently increased.26
In this article, we demonstrate that siRNA-Int6 promotes neoangiogenesis in subcutaneous regions and in the tissues surrounding a wound in adult mice. In addition, we describe the physiological mechanism by which Int6 silencing accelerates the formation of functional vessels.
Cell Cultures, Vector Constructs, and Transfection
HeLa, MCF7, and NIH3T3 cells were cultured in Dulbecco modified Eagle medium, normal human dermal fibroblast (NHDF) in 106S medium. siRNA-Int6, HIF2α plasmids, point mutants of HIF2α, and reporter assay vectors were constructed as described in the online-only Data Supplement. Transfection was performed with Lipofectamine 2000 reagent, calcium/phosphate, and Nucleofection. For details, see the Materials and Methods section in the online-only Data Supplement.
Study of In Vivo Angiogenesis in Mouse Skin and Immunohistochemistry
The HVJ envelope suspension mixed with the indicated amount of siRNA plasmid was injected into 6- to 8-week-old female BALB/c mice and/or 33-week-old female C57BL/KsJ-db/db mice after they were anesthetized with isoflurane. For the wound healing model, the excisional wounds were made into the dorsal skin of the mice with a 6-mm biopsy punch.
Staining was performed in 4-μm paraffin tissue sections to assess the expression of CD31, α-smooth muscle actin (αSMA), S100A4, green fluorescent protein (GFP), and collagen. For details, see the Materials and Methods section in the online-only Data Supplement.
Quantitative Real-Time Polymerase Chain Reaction and Western Blotting
Target messenger RNAs (mRNAs) were measured as previously described.26,34 The respective primer pairs and probes were supplied with the TaqMan Gene Expression Reagent Kit. Protein samples were prepared from cells or tissues transfected with control (siRNA-C) or siRNA-Int6 plasmids. Standard blotting protocols were performed by using specific polyclonal antibodies against INT6, HIF2α, and Myc and monoclonal antibodies against HIF1α and tubulin. For details, see the Materials and Methods section in the online-only Data Supplement.
Immuno–Polymerase Chain Reaction
Immuno–polymerase chain reaction (MUSTag®) assays were performed according to the manufacturer’s instructions. The protein levels of bFGF, hepatocyte growth factor (HGF), VEGF, and ANG-1 in the supernatant samples of NHDF-transfected cells were analyzed. For details, see the Materials and Methods section in the online-only Data Supplement.
Reporter Assay and HaloCHIP Assay
Luciferase activity was determined with the dual luciferase reagent assay system according to the manufacturer’s instructions. Identification of HRE-promoter binding of HIF2α and INT6 was performed with HaloCHIP assays. For details, see the Materials and Methods section in the online-only Data Supplement.
The results are expressed as means with SD or as individual data and subjected to 2-sided statistical analyses with Prism version 5 (GraphPad Software Inc, San Diego, Calif). The Kolmogorov-Smirnov test was used to asses normal distributions for each of the 2 groups compared in any t test; in case of small P values (P<0.05), the data were considered to fail the normality test, and the Mann-Whitney test was performed to compare the angiogenetic effects between siRNA-C and siRNA-Int6 treatment. A 2-sample (unpaired) Student t test was done for protein levels of NHDF cells and area of vessels after ex vivo transplantation, as well as relative HIF2α promoter activity of HeLa cells treated with siRNA-Int6 and comparison of mRNAs of each target factors in 2 groups that were treated with siRNA-C and siRNA-Int6 after 1 day. Multiplicity adjustment methods were applied to study differences in continuous outcome variables among more than 2 groups in time (1-way analysis of variance [ANOVA]), followed by a Bonferroni’s multiple-comparisons posttest or Dun’s posttest, whereas an ordinary 2-way ANOVA with Bonferroni’s multiple-comparisons posttest was performed to determine how the siRNA-Int6 effect response was affected by a different time course without repeated measurement. A repeated-measures 2-way ANOVA with Bonferroni’s multiple-comparison posttest was performed to analyze how the siRNA-Int6 effect response in the mouse wound healing model was affected by time course with repeated measurements. Values of P<0.05=(*), <0.01=(**), and <0.001=(***), according to the experiments were considered significant.
Int6 Silencing Induced the Formation of Physiologically Normal Subcutaneous Vessels in Adult Mice
On the basis of our previous findings, we assumed that the in vivo expression of siRNA-Int6 might induce angiogenesis by upregulating HIF2α. We evaluated the effect of siRNA-Int6 in an in vivo assay by subcutaneously injecting an siRNA-Int6 plasmid-GenomONE mixture into the dorsal skin of mice (dotted circle in Figure 1A).35,36 Strong induction of neoangiogenesis was observed 5 days after injection of siRNA-Int6–219 and -358 (Figure 1A). The total length of blood vessels induced by siRNA-Int6–219 (lane 2; 1603±331 pixels) and siRNA-Int6–358 (lane 3; 1172±141) was 6.2- and 4.5-fold greater, respectively, than the control (lane 1; 259±71). The total surface area of the new vessels induced by siRNA-Int6–219 (lane 5; 7573±2326) and siRNA-Int6–358 (lane 6; 5361±944) was 8- and 6-fold higher, respectively, than the negative control (lane 4; 687±308; Figure 1B).
To investigate whether the efficacy of angiogenesis was dose dependent on the silencing of Int6, we subcutaneously injected the negative control or different amounts of siRNA-Int6 plasmids into mice. Imaging analysis of the mice subcutis harvested 5 days after injection (Figure II in the online-only Data Supplement) indicated that neoangiogenesis started at 10 μg siRNA-Int6 plasmid and followed a dose-dependent enhancement up to 40 μg (lane 9; 69 860±1151 pixels) relative to the negative control (lane 8; 16 530±598 pixels) (Figure 1C). Tissue mRNA levels of HIF2α, VEGF-A, and HGF in Figure 1C at day 1 started to increase after injection of 10 to 20 μg siRNA-Int6 and reached a maximum at 40 μg. mRNA of bFGF showed a similar pattern but with a significant enhancement only at 40 and 50 μg siRNA-Int6. Although Int6 mRNA showed no significant decrease after 10 and 20 μg siRNA-Int6 injection, the level correlated inversely with the increasing HIF2α mRNA at the higher doses of siRNA-Int6 (Figure III in the online-only Data Supplement). A time-course analysis of vessel formation after a single injection of 40 μg siRNA-Int6 plasmid (Figure 1D) revealed visible neoangiogenesis on day 3 after injection (lane 2; 25 900±5456 pixels) and reached a maximum on day 5 (62 823±6790 pixels).
The New Vessels Tested Positive for CD31 and αSMA
Histopathological examination of the subcutaneous skin sections around the injection site of 40 μg siRNA-C (Figure 2A) or siRNA-Int6 after 5 days (Figure 2B) revealed an increase in newly formed vessels that make up normal arteries and veins (red arrows) with an internal membrane as shown by hematoxylin-eosin and elastica–van Gieson staining at the siRNA-Int6 injection site.
We also stained the tissue with antibodies against CD31 and αSMA. Compared with the negative control (Figure 2C), a significantly greater number of cells stained positive for both antigens in the skin injected with siRNA-Int6 (Figure 2D). We also measured the number of CD31- or αSMA-positive vessels per field (× 400) in each section of 3 different mice for 3 experiments; 2.4-fold more CD31-positive vessels were in the siRNA-Int6 than in the control group (siRNA-Int6, 11.25±2.66; siRNA-C, 4.75±1.98; Figure 2E) and αSMA-positive vessels were 5.5-fold higher (siRNA-Int6, 5.63±1.99; siRNA-C, 1.25±1.04; Figure 2F). These findings indicated that these vessels were newly formed and caused by the injection of siRNA-Int6.
Enhanced Angiogenic Factor Expression Within the Injected Skin Region in Adult Mice
The protein levels of HIF1α, HIF2α, and INT6 in mice injected with 40 μg siRNA-C or siRNA-Int6 plasmids showed a decrease of Int6 as early as day 1 that remained low for 5 days. In contrast, HIF2α expression rose significantly on day 1 and remained high until day 5 (with no significant difference of HIF2α expression between days 3 and 5; P>0.05). HIF1α levels did not differ significantly between the siRNA-C and siRNA-Int6 groups (Figure 3A and 3B).
Further examination of mRNA levels revealed that after injection of siRNA-Int6, the amount of Hif2α mRNA rose on day 1 and was enhanced until day 5, which correlated inversely with the change of Int6 mRNA levels, whereas Hif1α mRNA did not vary significantly compared with the negative control (Figure 3C). mRNA levels of VEGF-A, HGF, bFGF, Tyrosine kinase with immunoglobulin-like and EGF-like domains 2 (TIE2), ANG-1, ANG-2, monocyte chemoattractant protein-1 (MCP-1), and pleiotrophin were 2- to 6-fold higher than in the control tissues expressing siRNA-C on day 1 (Figure 3D). These results are in accordance with previous reports that bFGF, VEGF-A, and pleiotrophin are transcribed by HIF2α through HREs in their promoters.12
Subcutaneous Fibroblasts Are the Major Target for siRNA-Based Int6 Silencing in Mice
To determine the cells affected by siRNA-Int6 injection in the subcutaneous regions of mice, an siRNA-Int6-GFP vector was used as a probe. Hematoxylin and eosin staining (Figure 4A) and immunohistochemical analysis with anti-GFP antibodies (Figure 4B) indicated that GFP expression was localized primarily in spindle-shaped cells and, to a certain extent, in macrophage-like cells. To confirm the identity of the spindle-shaped cells, we stained the tissue sections with an anti-GFP antibody (brown or blue) in combination with specific fibroblast markers: anti-collagen type Ι (blue; Figure 4C) or anti-S100A4 (brown; Figure 4D) antibodies. The results clearly indicated that the spindle-shaped cells were fibroblasts and the main target cells of siRNA-Int6 injection.
Transplantation of Primary Fibroblasts Treated Ex Vivo With siRNA-Int6 Induces Functional Blood Vessel Formation in Adult Nude Mice
To confirm that the fibroblasts in the subcutaneous region play a major role in angiogenesis caused by Int6 silencing, we examined in vitro mRNA expression of several angiogenic factors in NHDFs after INT6 silencing with quantitative real-time polymerase chain reaction. INT6 mRNA significantly decreased to 34% of the control cells 18 hours after transfection, whereas HIF2α mRNA increased to 140% of the control at 18 hours and continued to increase to 185% at 72 hours. In contrast, HIF1α mRNA remained constant throughout the 72-hour time course. VEGF-A, bFGF, HGF, ANG-2, and ANG-1 mRNA significantly increased 48 hours after transfection (Figure 5A). We also measured secreted protein concentrations in the culture supernatant of NHDFs transfected with siRNA-Int6 or siRNA-C over a 72-hour period with a supersensitive multiplex assay (MUSTag). Compared with negative controls, the levels of VEGF increased from 0.2 to 1.5 pg/mL; HGF, from 700 to 900 pg/mL; and ANG-1, from 120 to 180 pg/mL. bFGF levels did not change significantly, probably because 15 ng/mL bFGF in the culture medium masked the changes (Figure 5B). Western blot analysis showed enhanced expression of endogenous HIF2α in fibroblasts transfected with siRNA-Int6 for 48 hours compared with the control, whereas the expression of endogenous INT6 was lower (Figure 5C).
Next, we transfected NHDFs ex vivo with siRNA-Int6 or siRNA-C plasmids and transplanted them into subcutaneous regions of nude mice after 24 hours of incubation. The transplanted siRNA-Int6–treated NHDFs strongly induced neoangiogenesis 5-fold higher than siRNA-C–treated NHDFs (Figure 5D and Figure IV in the online-only Data Supplement), similar to the levels observed in Figure 1A. These results confirm that Int6-silenced subcutaneous fibroblasts release adequate amounts of angiogenic factors to induce the formation of functional and stable blood vessels.
siRNA-Int6 Accelerates Wound Healing by Promoting Neoangiogenesis in Mice
Extending experiments to wound healing, we injected a total of 40 μg siRNA-Int6 in the area surrounding a wound. siRNA-Int6 significantly enhanced wound closure as early as 24 hours after injury with an ∼38% decrease in wound size, followed by 42% from days 2 to 4; the most significant the wound closure difference was 54% on day 5 compared with the control (Figure 6A). Figure 6B shows pictures of representative wounds on days 0 and 5 after injury in both groups.
The number of blood vessels (red arrows) in the subcutaneous tissue surrounding the wound of the siRNA-C group was significantly lower than in the siRNA-Int6 group as evaluated from photos (Figure 6C and 6D) and by elastica–van Gieson (Figure V in the online-only Data Supplement) and CD31 staining (Figure 6D). siRNA-Int6–treated wounds showed a significant increase of ∼4-fold in vessel formation with concomitantly accelerated wound closure compared with those receiving siRNA-C (Figure 6E). We also applied the siRNA-Int6 to genetically diabetic C57BL/KsJ-db/db mice37 and observed a significant acceleration of wound closure even over an extended time course compared with control mice (Figure VI in the online-only Data Supplement). These results prefigure an encouraging therapeutic value of siRNA-Int6 for treating delayed wound healing, especially in diabetic patients with impaired microcirculation.
siRNA-Int6 Induced Neoangiogenesis Directly Involves the Accumulation of HIF2α
We previously reported that Int6 binds to the Int6 binding site (IBS) of HIF2α (571 to 700 aa), which leads to the proteasomal degradation of HIF2α. Our in vivo results prompted us to examine whether overexpression of HIF2α or HIF2α mutants that lacked pVHL binding or the IBS showed angiogenic effects similar to Int6 silencing. We generated a mutant (P403A/P530A) in which the proline (P) residues 405 and 531 were substituted with alanine (A) to inhibit the degradation of HIF2α by the binding of proline hydroxylase.16,17 We also generated a deletion mutant (ΔIBS) that lacked the IBS to disrupt Int6-induced degradation. Finally, we constructed a double mutant (P403A/P530A+ΔIBS) lacking both pVHL and IBS (Figure 7A).
As a preliminary step, we evaluated the expression of each HIF2α mutant and wild-type HIF2α cotreated with siRNA-Int6 or siRNA-C in NIH3T3 cells under normoxic or hypoxic conditions (Figure VII in the online-only Data Supplement). The Western blot results suggest that under normoxic condition, the protein stability of HIF2α is regulated mainly by INT6, and P403A/P530A+ΔIBS was expressed at approximately the same level as ΔIBS.
Next, we optimized the injection dose of the plasmids and found that 40 μg plasmids triggered the maximum effect in an in vivo angiogenesis assay (data not shown). Then, we injected the optimized amount of 40 μg each pcDNA3-HIF1α, HIF2α, or the above mutants into the subcutaneous region of mice. Overexpression of wild-type HIF2α induced angiogenesis only 1.6-fold higher than the control but 1.9-fold higher with HIF2α-P403A/P530A, 2.8-fold higher with HIF2α-ΔIBS, and 3.0-fold higher with HIF2α-P403A/P530A+ΔIBS overexpression. Angiogenesis induced by HIF2α-P403A/P530A+ΔIBS was strongest but not significantly different (P>0.05) compared with siRNA-Int6. HIF1α overexpression showed no significant effect on neoangiogenesis (Figure 7B).
To clarify whether Int6 silencing–induced angiogenesis is unequivocally caused by concomitant HIF2α accumulation, we first examined the efficiency of 2 siRNAs against Hif2α (HIF2α-R1 and HIF2α-R2) in a Western blot analysis. The results indicated that the transfection of siRNA-HIF2α-R1 or -R2 strongly inhibited endogenous HIF2α protein expression in NIH3T3 cells (Figure VIII in the online-only Data Supplement). We next subcutaneously coinjected 30 μg siRNA-Int6 and different amounts of 2 siRNA-HIF2αs into the skin of the mice. The neoangiogenesis resulting from Int6 silencing was completely abolished by cotransfection of 10 μg HIF2α-R1 or HIF2α-R2 (Figure 7C and 7D and Figure IX in the online-only Data Supplement). These results unambiguously indicate that the effects of siRNA-Int6 can be attributed to HIF2α activation.
Negative Feedback of HIF2α Stability by HIF2α-Dependent INT6 Induction Through HREs in the Int6 Promoter
The expression of endogenous INT6 protein was upregulated in a dose-dependent manner by transfection of HIF2α in MCF7 cells, implying an HIF2α-induced transcription of Int6 (Figure 8A). To address this possibility, we used a luciferase reporter plasmid containing a 2089-bp upstream sequence of the human INT6 promoter. After cotransfection of the pGL3–INT6 promoter with pcDNA3-HIF2α in HeLa cells, the transcriptional activity of the INT6 promoter was increased in an HIF2α expression–dependent manner under both normoxic and hypoxic conditions (Figure 8B). We also discovered an HIF2α-mediated transcription of HIF2α by using a luciferase reporter plasmid pGL3-HIF2α promoter; HIF2α activity also was influenced by siRNA-Int6 (Figure 8C).
HIF2α-related transcription of Int6 and Hif2α38 was also supported by the in silico detection of putative HREs (CACG or CGTG) in the promoters of Int6 and Hif2α (Figure X [left] in the online-only Data Supplement). We performed a HaloCHIP assay that clearly demonstrated the binding of HIF2α to HREs in the murine Hif2α (−367 to −1233 nt) and Int6 (−26 to −598 nt) promoters (Figure X [right] in the online-only Data Supplement). Thus, in hypoxic conditions or after HIF2α overexpression, HIF2α protein activity triggers its self-transcription. In parallel, the induced and stabilized HIF2α also transcribes endogenous INT6, which binds and degrades HIF2α. Therefore, we suggest that the self-activation of HIF2α and the negative feedback of HIF2α through HIF2α-dependent INT6 induction form a delicate balance between the degradation and stability of HIF2α. These mechanisms take place on both the transcriptional and protein levels.
Our previous results suggested that INT6 is a key factor suppressing the induction of angiogenic factors by destabilizing HIF2α via proteasomal degradation.26 Here, we extend our research on Int6 silencing to in vivo angiogenesis. We injected siRNA-Int6 into the subcutaneous tissues of mice and analyzed the angiogenic effects of Int6 silencing in vivo. Significant neoangiogenesis in an siRNA-Int6 expression–dependent manner was observed within 5 days (Figure 1). A small amount of siRNA-Int6 (20 μg) application (Figure III in the online-only Data Supplement) did not lead to a significant reduction of Int6 mRNA, whereas HIF2α mRNA and the angiogenic factors VEGF-A and HGF mRNA were slightly enhanced compared with siRNA-C. Measurements of changes in in vivo tissues are susceptible to high background disturbances, especially because the basic level of Int6 is high (data not shown) and small changes are difficult to detect, whereas small changes in VEGF-A and HGF become more obvious because of the low basic mRNA level (data not shown). We also consider another reason (Figure 8D): The enhanced transcription of Int6 mRNA by stabilized HIF2α after endogenous Int6 silencing may mask the decrease in the basic Int6 mRNA, whereas higher doses of Int6-siRNA (>30 μg) showed the significance because of efficient silencing of both the basic and enhanced Int6 mRNA.
We could determine subcutaneous fibroblasts and, to a lesser extent, macrophages as the source of Int6 silencing effects (Figure 4). Fibroblasts stimulate angiogenesis, stabilize the neovascular endothelium,39 and play a pivotal role in angiogenesis as progenitors of endothelial cells in newly formed blood vessels.40 Our results are in agreement with this, as demonstrated by the induction of potent neoangiogenesis after subcutaneous ex vivo transplantation with siRNA-Int6–transfected fibroblasts into nude mice (Figure 5D). Ito et al41 reported that an increase in endogenous MCP-1 can enhance inflammatory processes by attracting monocytes, the accumulation of which leads to increased collateral vessel formation and blood flow in a femoral artery occlusion model. We also observed that MCP-1 mRNA (Figure 3D) increased 24 hours after siRNA-Int6 injection. Thus, fibroblasts are the principal effectors of Int6 silencing in the mouse skin releasing angiogenic factors necessary for neoangiogenesis, although macrophages might be partially involved.
To investigate why Int6 silencing induces neoangiogenesis stronger than the hypoxic/ischemic stabilization of HIF2α in vivo (Figure 7B), we injected plasmids encoding an HIF2α mutant (P403A/P531A) and an HIF2α mutant lacking ΔIBS into the skin of mice. Higher levels of neoangiogenesis were observed with the expression of HIF2α-ΔIBS than with HIF2α-P403A/P531A. A triple mutation (HIF2α-P403A/P531A-ΔIBS) triggered significantly higher levels of neoangiogenesis than HIF2α-P403A/P531A, whereas Int6 silencing showed an effect similar to or a trend toward even a higher effect than HIF2α-P403A/P531A-ΔIBS (Figure 7B). Moreover, the neoangiogenesis induced by siRNA-Int6 was abolished in mouse skin after coinjection with siRNA-HIF2α (Figure 7C and 7D and Figure IX in the online-only Data Supplement); therefore, we confirmed that siRNA-Int6 unambiguously induces neoangiogenesis and enhances wound healing by upregulating HIF2α.
We further discovered an HIF2α-driven transcription of Int6 via the involvement of HREs in the Int6 promoter. Promoter activities of both Int6 and Hif2α increased in an HIF2α expression–dependent manner, even under normoxic conditions. Furthermore, we proved functional HREs in the Int6 (−26 to −598 nt) and Hif2α promoter (−367 to −1233 nt; Figure X in the online-only Data Supplement). Overall, our findings revealed that HIF2α transcribes Int6 and itself and that INT6, as a negative regulator of HIF2α stability, instantly degrades and reduces the HIF2α protein levels (Figure 8D). Int6 silencing at a certain level leads to the inactivation of the existing and de novo–transcribed INT6, resulting in the reduced degradation of HIF2α; the accumulating levels of HIF2α then further activate the self-transcription of Hif2α,42 and the resulting enhanced accumulation of HIF2α leads to a potent neoangiogenesis (Figure 8E). Thus, Int6 silencing causes a stronger effect on angiogenesis than the overexpression of HIF2α.
In this way, INT6 acts as one of several neoangiogenesis master switches by controlling the protein levels of HIF2α in an oxygen-independent manner. Recent reports have focused on micro-RNAs that positively or negatively affect the transcriptional levels of their specific mRNAs.43 In our scheme of Int6 transcriptional regulation, we cannot discount the micro-RNA regulation of Int6; therefore, its possible involvement needs to be investigated further.
Int6 silencing is an effective way to facilitate HIF2α activity without hypoxia, leading to physiological and functional neoangiogenesis in mice.
We are grateful to F. McKeon, T. Hashimoto, Y. Suzuki, and I. Nozawa for suggestions and critical reading of the manuscript; T. Suzuki and T. Kietzmann for providing the HIF2α and HIF3α plasmids; T. Yamashita for providing full-length mouse HIF2α ; N. Makisaka for technical assistance; T. Nagaoka for taking photographs; and all the members of our laboratory for their input during the course of this work.
Sources of Funding
This work was supported in part by a grant-in-aid from the Ministry of Education, Culture, Sports, Science, and Technology of Japan (Drs Shibasaki and Chen).
Forsythe JA, Jiang BH, Iyer NV, Agani F, Leung SW, Koos RD, Semenza GL. Activation of vascular endothelial growth factor gene transcription by hypoxia-inducible factor 1. Mol Cell Biol. 1996; 16: 4604–4613.
Shing Y, Folkman J, Sullivan R, Butterfield C, Murray J, Klagsbrun M. Heparin affinity: purification of a tumor-derived capillary endothelial cell growth factor. Science. 1984; 223: 1296–1299.
Folkman J. Can mosaic tumor vessels facilitate molecular diagnosis of cancer? Proc Natl Acad Sci U S A. 2001; 98: 398–400.
Li A, Dubey S, Varney ML, Dave BJ, Singh RK. IL-8 directly enhanced endothelial cell survival, proliferation, and matrix metalloproteinases production and regulated angiogenesis. J Immunol. 2003; 170: 3369–3376.
Thurston G, Suri C, Smith K, McClain J, Sato TN, Yancopoulos GD, McDonald DM. Leakage-resistant blood vessels in mice transgenically overexpressing angiopoietin-1. Science. 1999; 286: 2511–2514.
Antoine M, Tag CG, Wirz W, Borkham-Kamphorst E, Sawitza I, Gressner AM, Kiefer P. Upregulation of pleiotrophin expression in rat hepatic stellate cells by PDGF and hypoxia: implications for its role in experimental biliary liver fibrogenesis. Biochem Biophys Res Commun. 2005; 337: 1153–1164.
Semenza GL. HIF-1 and human disease: one highly involved factor. Genes Dev. 2000; 14: 1983–1991.
Semenza GL, Wang GL. A nuclear factor induced by hypoxia via de novo protein synthesis binds to the human erythropoietin gene enhancer at a site required for transcriptional activation. Mol Cell Biol. 1992; 12: 5447–5454.
Wang GL, Jiang BH, Rue EA, Semenza GL. Hypoxia-inducible factor 1 is a basic-helix-loop-helix-PAS heterodimer regulated by cellular O2 tension. Proc Natl Acad Sci U S A. 1995; 92: 5510–5514.
Iyer NV, Kotch LE, Agani F, Leung SW, Laughner E, Wenger RH, Gassmann M, Gearhart JD, Lawler AM, Yu AY, Semenza GL. Cellular and developmental control of O2 homeostasis by hypoxia-inducible factor 1 alpha. Genes Dev. 1998; 12: 149–162.
Cockman ME, Masson N, Mole DR, Jaakkola P, Chang GW, Clifford SC, Maher ER, Pugh CW, Ratcliffe PJ, Maxwell PH. Hypoxia inducible factor-alpha binding and ubiquitylation by the von Hippel-Lindau tumor suppressor protein. J Biol Chem. 2000; 275: 25733–25741.
Ryan HE, Lo J, Johnson RS. HIF-1 alpha is required for solid tumor formation and embryonic vascularization. EMBO J. 1998; 17: 3005–3015.
Tian H, Hammer RE, Matsumoto AM, Russell DW, McKnight SL. The hypoxia-responsive transcription factor EPAS1 is essential for catecholamine homeostasis and protection against heart failure during embryonic development. Genes Dev. 1998; 12: 3320–3324.
Peng J, Zhang L, Drysdale L, Fong GH. The transcription factor EPAS-1/hypoxia-inducible factor 2alpha plays an important role in vascular remodeling. Proc Natl Acad Sci U S A. 2000; 97: 8386–8391.
Compernolle V, Brusselmans K, Acker T, Hoet P, Tjwa M, Beck H, Plaisance S, Dor Y, Keshet E, Lupu F, Nemery B, Dewerchin M, Van Veldhoven P, Plate K, Moons L, Collen D, Carmeliet P. Loss of HIF-2alpha and inhibition of VEGF impair fetal lung maturation, whereas treatment with VEGF prevents fatal respiratory distress in premature mice. Nat Med. 2002; 8: 702–710.
Tian H, McKnight SL, Russell DW. Endothelial PAS domain protein 1 (EPAS1), a transcription factor selectively expressed in endothelial cells. Genes Dev. 1997; 11: 72–82.
Chen L, Uchida K, Endler A, Shibasaki F. Mammalian tumor suppressor Int6 specifically targets hypoxia inducible factor 2 alpha for degradation by hypoxia- and pVHL-independent regulation. J Biol Chem. 2007; 282: 12707–12716.
Marchetti A, Buttitta F, Miyazaki S, Gallahan D, Smith GH, Callahan R. Int-6, a highly conserved, widely expressed gene, is mutated by mouse mammary tumor virus in mammary preneoplasia. J Virol. 1995; 69: 1932–1938.
Asano K, Merrick WC, Hershey JW. The translation initiation factor eIF3–p48 subunit is encoded by int-6, a site of frequent integration by the mouse mammary tumor virus genome. J Biol Chem. 1997; 272: 23477–23480.
Crane R, Craig R, Murray R, Dunand-Sauthier I, Humphrey T, Norbury C. A fission yeast homolog of Int-6, the mammalian oncoprotein and eIF3 subunit, induces drug resistance when overexpressed. Mol Biol Cell. 2000; 11: 3993–4003.
Akiyoshi Y, Clayton J, Phan L, Yamamoto M, Hinnebusch AG, Watanabe Y, Asano K. Fission yeast homolog of murine Int-6 protein, encoded by mouse mammary tumor virus integration site, is associated with the conserved core subunits of eukaryotic translation initiation factor 3. J Biol Chem. 2001; 276: 10056–10062.
Bandyopadhyay A, Lakshmanan V, Matsumoto T, Chang EC, Maitra U. Moe1 and spInt6, the fission yeast homologues of mammalian translation initiation factor 3 subunits p66 (eIF3d) and p48 (eIF3e), respectively, are required for stable association of eIF3 subunits. J Biol Chem. 2002; 277: 2360–2367.
Burks EA, Bezerra PP, Le H, Gallie DR, Browning KS. Plant initiation factor 3 subunit composition resembles mammalian initiation factor 3 and has a novel subunit. J Biol Chem. 2001; 276: 2122–2131.
Yahalom A, Kim TH, Winter E, Karniol B, von Arnim AG, Chamovitz DA. Arabidopsis eIF3e (INT-6) associates with both eIF3c and the COP9 signalosome subunit CSN7. J Biol Chem. 2001; 276: 334–340.
Kato H, Tamamizu-Kato S, Shibasaki F. Histone deacetylase 7 associates with hypoxia-inducible factor 1alpha and increases transcriptional activity. J Biol Chem. 2004; 279: 41966–41974.
Matsumoto T, Bohman S, Dixelius J, Berge T, Dimberg A, Magnusson P, Wang L, Wikner C, Qi JH, Wernstedt C, Wu J, Bruheim S, Mugishima H, Mukhopadhyay D, Spurkland A, Claesson-Welsh L. VEGF receptor-2 Y951 signaling and a role for the adapter molecule TSAd in tumor angiogenesis. EMBO J. 2005; 24: 2342–2353.
Sato M, Tanaka T, Maeno T, Sando Y, Suga T, Maeno Y, Sato H, Nagai R, Kurabayashi M. Inducible expression of endothelial PAS domain protein-1 by hypoxia in human lung adenocarcinoma A549 cells: role of Src family kinases-dependent pathway. Am J Respir Cell Mol Biol. 2002; 26: 127–134.
Ito WD, Arras M, Winkler B, Scholz D, Schaper J, Schaper W. Monocyte chemotactic protein-1 increases collateral and peripheral conductance after femoral artery occlusion. Circ Res. 1997; 80: 829–837.
Hu CJ, Iyer S, Sataur A, Covello KL, Chodosh LA, Simon MC. Differential regulation of the transcriptional activities of hypoxia-inducible factor 1 alpha (HIF-1alpha) and HIF-2alpha in stem cells. Mol Cell Biol. 2006; 26: 3514–3526.
During the past decades, several studies have tried to trigger revascularization by exogenously applying angiogenic factors into the tissues of interest. A common obstacle was that the application of the specific angiogenic factor alone or an unbalanced combination with others led to the development of unphysiological and incomplete leaky blood vessels. In our research, we first found that hypoxia-inducible factor 2α is under subtype-specific and negative regulation of INT6, which directly binds to hypoxia-inducible factor 2α and triggers its degradation through the proteasome pathway. Further in vivo experiments revealed that Int6 silencing led to enhancement of hypoxia-inducible factor 2α activity even under normoxic condition, temporarily inducing a physiological and potent neovascularization. In mouse skin, we determined subcutaneous fibroblasts as the major source of angiogenic factors. The fibroblasts treated ex vivo with siRNA-Int6 demonstrated that transplantation into the skin also led to the same strong induction of physiological neovascularization. Further application of siRNA-Int6 in a wound healing model (normal and db/db mice) showed a significantly enhanced wound repair with concomitant formation of new vessels. These results prefigure an encouraging therapeutic value of siRNA-Int6 for treating delayed wound healing, especially in diabetic patients with impaired microcirculation. Therefore, siRNA-Int6–transfected fibroblast cell treatment or direct application of siRNA-Int6 might be of clinical value in treating ischemic diseases such as heart and brain ischemia, skin injury, and diseases involving obstructed vessels.
The online-only Data Supplement is available with this article at http://circ.ahajournals.org/cgi/content/full/CIRCULATIONAHA.109.931931/DC1.