|Home | About | Journals | Submit | Contact Us | Français|
miR-199a-5p plays a critical role in controlling cardiomyocyte survival. However, its significance in endothelial cell biology remains ambiguous. Here, we report the first evidence that miR-199a-5p negatively regulates angiogenic responses by directly targeting v-ets erythroblastosis virus E26 oncogene homolog 1 (Ets-1). Induction of miR-199a-5p in human dermal microvascular endothelial cells (HMECs) blocked angiogenic response in Matrigel® culture, whereas miR-199a-5p-deprived cells exhibited enhanced angiogenesis in vitro. Bioinformatics prediction and miR target reporter assay recognized Ets-1 as a novel direct target of miR-199a-5p. Delivery of miR-199a-5p blocked Ets-1 expression in HMECs, whereas knockdown endogenous miR-199a-5p induced Ets-1 expression. Matrix metalloproteinase 1 (MMP-1), one of the Ets-1 downstream mediators, was negatively regulated by miR-199a-5p. Overexpression of Ets-1 not only rescued miR-199a-5p-dependent anti-angiogenic effects but also reversed miR-199a-5p-induced loss of MMP-1 expression. Similarly, Ets-1 knockdown blunted angiogenic response and induction of MMP-1 in miR-199a-5p-deprived HMECs. Examination of cutaneous wound dermal tissue revealed a significant down-regulation of miR-199a-5p expression, which was associated with induction of Ets-1 and MMP-1. Mice carrying homozygous deletions in the Ets-1 gene exhibited blunted wound blood flow and reduced abundance of endothelial cells. Impaired wound angiogenesis was associated with compromised wound closure, insufficient granulation tissue formation, and blunted induction of MMP-1. Thus, down-regulation of miR-199a-5p is involved in the induction of wound angiogenesis through derepressing of the Ets-1-MMP1 pathway.
Angiogenesis, sprouting of pre-existing blood vessels, is one of the major biological responses supporting both cutaneous wound healing and tumorogenesis. Emerging evidence has revealed that microRNAs (miRs)2 play a critical role in regulation of angiogenesis (1, 2). These small RNA exert substantial gene regulatory effects by a number of ways including physical interaction with the 3′-UTR of mRNA, hindering translation or leading to transcript degradation. Our group and others have characterized the significance of specific miRs in modulating the angiogenic response. miR-210, for example, supports angiogenesis by targeting angiostatic proteins ephrin-A3 and protein-tyrosine phosphatase 1B (1). miR-200b, a hypoxia-repressible miR (3), negatively regulates angiogenic response by silencing vascular endothelial growth factor (4), GATA-binding protein 2 (5), and vascular endothelial growth factor receptor 2 (5).
miR-199a-5p is a multifaceted miR that regulates cell survival during ischemia-reperfusion injury (6–8). It is expressed in a variety of organs including the brain (6), liver (9), stomach (10), ovarian tissue (11, 12), testicular tissue (13), and the cardiovascular system, such as in cardiomyocytes (7, 8) and endothelial cells (14, 15). Recent evidence revealed that miR-199a-5p plays a critical role in attenuating cell migration and in limiting cancer cell metastasis by targeting receptor tyrosine kinase Discoidin domain receptor-1 (9), Axl (16), and anti-adhesive transmembrane glycoprotein podocalyxin-like protein 1 (13). We hypothesized that miR-199a-5p exerts angiostatic effects on endothelial cells. In this report, we provide the first evidence that inhibition of endogenous miR-199a-5p in endothelial cells supports angiogenesis, and first characterized v-ets erythroblastosis virus E26 oncogene homolog 1 (Ets-1) as a bona fide miR-199a-5p target. We found that matrix metalloproteinase 1 (MMP-1) is the key downstream mediator of the miR-199a-5p-Ets-1 pathway. Down-regulation of miR-199a-5p enables cutaneous wound angiogenesis via derepression of Ets-1 and subsequent induction of MMP-1.
Human dermal microvascular endothelial cells (HMECs) were grown in MCDB-131 medium supplemented with 10% FBS, 10 mm l-glutamine, 100 IU/ml penicillin, and 0.1 mg/ml streptomycin as described previously (2, 3, 5). Primary adult human dermal microvascular endothelial cells were cultured in EBM-2 medium (Lonza, Basel, Switzerland) supplemented with EGM-2MV single quotes as described by the manufacturer. HEK-293 cells were maintained with high glucose DMEM supplemented with 10% FBS and 100 IU/ml penicillin, 0.1 mg/ml streptomycin. All of the cells were cultured in standard culture conditions (at 37 °C with 20% O2 and 5% CO2). The cells were seeded in a 12-well plate at a density of 0.12 × 106 cells/well 24 h before treatment. Delivery of small RNAs was achieved using DharmaFECTTM 1 transfection reagent (Dharmacon RNA Technologies, Denver, CO). miR-199a-5p mimic (50 nm), miR-199a-5p inhibitor (100 nm), or siRNA smart pool for human Ets-1 or MMP-1 (100 nm) were obtained from Dharmacon RNA Technologies. Unless specific, the cells were lysed after 72 h (HMECs and HEK-293) or 48 h (primary endothelial cells) of transfection and RNA/protein were collected for gene expression study. For Ets-1 overexpression studies, expressing plasmid encoding Ets-1 (Ets-1 pcDNA), a generous gift from Dr. Michael C. Ostrowski (The Ohio State University, Columbus, OH) (17) (1 μg/well), was delivered to the endothelial cells using Lipofectamine TM LTX/plus reagent as described previously (3). Two days after the plasmid delivery, the cells were subjected to miR-199a-5p mimic (or control mimic) delivery as described above for another 48 h for Matrigel® analysis, and RNA/protein extraction. For miR-199a-5p inhibitor-Ets-1 siRNA co-transfection study, both miR-199a-5p inhibitor and Ets-1 siRNA (or control siRNA) were mixed together and incubated with DharmaFECTTM 1 transfection reagent. For 72 h, the gene-liposome complex was delivered to endothelial cells that were subjected to in vitro angiogenesis studies or for RNA/protein extraction.
Male C57BL/6 mice (age, 8–10 weeks old) were obtained from Harlan Laboratory. The Ets-1-deficient mice (129/Sv × C57BL/6 × FVB/N) were kindly provided by Dr. Michael C. Ostrowski. Genotyping was performed with DNA from tail biopsies by PCR using specific primers (supplemental Table S1) with the following protocol: 94 °C for 45 s, 58 °C for 45 s, and 72 °C for 1 min (40 cycles). Two or four 6-mm diameter full thickness excisional wounds were developed on the dorsal skin of mice with a 6-mm disposable biopsy punch. A surgical splinted wound model was employed for the wound closure study, histological analysis, and collagen deposition assessment as described previously (18). Briefly, a donut-shaped splint with a 8-mm inner diameter was created from a 0.5-mm-thick silicone sheet (Grace Bio-Laboratories, Bend, OR) and placed such that the wound was centered within the splint. To fix the splint to the skin, an immediate bonding adhesive was used, followed by interrupted 5–0 nylon sutures (Ethicon, Inc., Somerville, NJ) to maintain position. All of the animal studies were performed in accordance with protocols approved by the Laboratory Animal Care and Use Committee of the Ohio State University. During the wounding procedure, the mice were anesthetized by low dose isoflurane inhalation for 5–10 min/standard recommendation. Each wound was digitally photographed at the time point indicated. Wound size was calculated by the software Image J. The animals were sacrificed at the indicated postwounding time point, and wound tissues were harvested (as frozen, in optimal cutting temperature compound or fixed in buffered formalin) for further analysis.
The miRVana miRNA isolation kit was employed to extract total RNA from tissue and cells according to the manufacturer's protocol (Ambion). miR expression was determined by miR TaqMan assays and TaqMan microRNA reverse transcription kit, followed by quantitative real time PCR using Universal PCR Master Mix (Applied Biosystems) (3, 5). For mRNA expression studies, total cDNA synthesis was achieved by a SuperScriptTM III first strand synthesis system (Applied Biosystems). The transcription levels of Ets-1, MMP-1, MMP-3, and MMP-9 were assessed by real time PCR using SYBR green-I (Invitrogen) as described previously (3, 5). The sequences of primers are shown in supplemental Table S1.
pLu-wild type Ets1–3′-UTR plasmid or construct carrying the mutation of the predicted miR-199a-5p binding site in Ets1–3′-UTR (100 ng) (Signosis, Sunnyvale, CA) was delivered to HEK-293 cells using LipofectamineTM LTX/plus reagent (Invitrogen) according to the manufacturer's protocol. The constructs were designed based on the sequence of miR-199a-5p binding sites and a total of 300 bp (starting from positions 2551–2850 of Ets-1 3′-UTR) was cloned in the 3′-UTR of constitutively active firefly luciferase construct. Mutation was made in the predicted miR-199a-5p binding site (from ACACUGG to UGUGACC, position 2708–2714 of Ets-1 3′-UTR). Constitutively active Renilla luciferase plasmid (10 ng) was delivered to cells for normalization. The cell lysates were assayed with dual luciferase reporter assay kit (Promega, Madison, WI). The data are presented as ratio of firefly to Renilla luciferase activity as described previously (3, 5).
In vitro angiogenesis assay were assessed by tube formation ability on Matrigel® culture as described previously (3, 5). Endothelial cells were seeded on Matrigel® precoated plate at a density of 5 × 104 cells/well. The angiogenic ability was assessed 8 h (HMECs) or 24 h (primary endothelial cells) after plating on Matrigel®, and the tube length was measured using the software AxioVision Rel 4.6 (Zeiss) (3, 5).
Cell migration was determined using culture insert (Ibidi, Verona, WI) according to the manufacturer's instruction. Briefly, the cells were seeded at the confluent monolayer on the chambers of the culture insert. After seeding for 1 h, the insert was removed, and the cells were allowed to migrate. Cell migratory distance was measured 4 h after removal of insert using AxioVision Rel 4.6 software (Zeiss).
Transfected cells were seeded in 96-well plate, and cell proliferation was determined by the CyQUANT® cell proliferation assays (Invitrogen) as described previously (19).
Western blot was performed using antibodies against Ets-1 (Santa Cruz Biotechnology, Santa Cruz, CA) (3) and mouse MMP-1 (Abbiotec, San Diego, CA) (20). Briefly, proteins were extracted from cells or mouse tissues with lysis buffer consisting of 10 mm Tris, pH 7.4, 150 mm NaCl, 1% Triton X-100, 1% deoxycholic acid, 0.1% SDS, and 5 mm EDTA. Cell lysate or tissue homogenate was resolved in SDS-PAGE and transblotted to PVDF membrane (GE Healthcare), followed by blocking with 10% skim milk and incubation with primary antibody against Ets-1 (1:3,000), MMP-1 (1:1,000) or β-actin (Sigma; 1: 10,000) (as loading control) overnight at 4 °C. The signal was visualized using corresponding HRP-conjugated secondary antibody (GE Healthcare; 1:3,000) and ECL PlusTM Western blotting detection reagents (GE Healthcare).
Immunohistochemistry of Ets-1 and CD31 was performed on cryosections of wound sample using specific antibodies as described previously (5). The wound tissue isolated was embedded in optimal cutting temperature compound and cryosectioned at 10 μm thick. Tissue sections were fixed with acetone, blocked with 10% normal goat serum, followed by incubation with antibodies against endothelial marker CD31 (BD Bioscience, San Jose, CA; 1:400) and Ets-1 (1:500) overnight at 4 °C. Signal was visualized by reaction with fluorescence-tagged secondary antibodies (FITC-tagged α-rat, 1:200; Alexa 568-tagged α-rabbit, 1:200) (Invitrogen; 1:200), counterstained with DAPI (Invitrogen; 1:10,000). For immunocytochemistry, the cells were fixed with 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, blocked, and incubated with primary antibody against Ets-1 (1: 500) or MMP-1 (Abcam, Cambridge, MA; 1:50) overnight at 4 °C. Signal was visualized using Alexa Fluor® 488 dye-conjugated antibody against rabbit, counterstained with rhodamine phalloidin (Invitrogen; 1:250), and DAPI. The images were captured by microscope, and quantification of fluorescent intensity of image was achieved by the software AxioVision Rel 4.6 (Carl Zeiss Microimaging).
Hematoxylin-eosin and Masson's trichrome staining were performed as described by our group previously (21). Briefly, tissue was collected and fixed in 10% buffered formalin solution for 2 weeks, after which they were processed for dehydration and paraffin embedding. Paraffin-embedded tissue was sectioned at 5 μm and processed for hematoxylin-eosin and Masson's trichrome staining.
LCM was performed using the laser microdissection system from PALM Technologies (Bernreid, Germany) containing a PALM MicroBeam and RoboStage for high throughput sample collection and a PALM RoboMover (PALM Robo software, version 2.2) as described previously (5). For immunofluorescence-directed LCM, blood vessels were stained with CD31 antibody (1:25), FITC-conjugated secondary antibody (1:200), and subsequently visualized using fluorescent lamp. For dermal LCM, dermal fraction was identified based on the histology after hematoxylin staining. Endothelial or dermal tissue elements were isolated and captured under a 20× ocular lens, using cut elements. Approximately 150,000 μm2 of tissue area was captured into lysis extraction buffer, and the extract was then held at −80 °C for extraction process (5).
Dermal blood flow was analyzed by laser Doppler imager as described previously (5). The MoorLDI-Mark 2 laser Doppler blood perfusion imager (resolution, 256 × 256 pixels; visible red laser beam at 633 nm) was used for mapping tissue blood flow under isofuran anesthetization.
The data reported represent the means ± S.E. Difference between two means was tested by Student t test, whereas one-way analysis of variance analysis was employed to compare three groups or more. p < 0.05 was considered statistically significant.
To determine the role of miR-199a-5p specifically on the angiogenic response of endothelial cells in vitro, miR-199a-5p mimic was transiently delivered to HMECs, and Matrigel® tube formation was analyzed. Delivery of miR-199a-5p to the endothelial cells not only led to significant accumulation of miR-199a-5p (Fig. 1A) but also exerted potent angiostatic effects (80% down-regulation compared with control) (Fig. 1C). Treatment of miR-199a-5p hairpin inhibitor blocked the endogenous expression of miR-199a-5p (Fig. 1B). Down-regulation of miR-199a-5p was accompanied by significant induction of tube formation ability by 40% in Matrigel® culture (Fig. 1D). To investigate whether miR-199a-5p exerts general angiostatic effect on endothelial cells, we examine the effect of miR-199a-5p mimic or inhibitor on primary adult human dermal microvascular endothelial cells. Delivery of miR-199a-5p mimic to primary endothelial cells significantly attenuated the tube formation ability on Matrigel® (supplemental Fig. S1, A and B), which was comparable with that on HMECs. Similarly, depletion of miR-199a-5p in primary endothelial cells significantly resulted in enhancement of angiogenic response. (supplemental Fig. S1, C and D). Consistent with the Matrigel® study, delivery of miR-199a-5p mitigates endothelial cell migration (supplemental Fig. S2A). miR-199a-5p does not exert any effect on endothelial cell proliferation (supplemental Fig. S2B).
To delineate the mechanisms underlying the angiostatic effect, bioinformatics analyses were performed to investigate the possible direct target using algorithms including Targetscan (22), Pictar (23), MiRanda (22), and miRDB (24). In silico analysis revealed that Ets-1 3′-UTR contains a highly conserved binding site at positions 2708–2714 for miR-199a-5p (Fig. 2A and supplemental Fig. S3A). To experimentally validate the possible interaction between miR-199a-5p and Ets-1 3′-UTR, miR target reporter luciferase assay was performed using construct pLuc-Ets1–3′-UTR plasmid (supplemental Fig. S3B). Delivery of miR-199a-5p mimic resulted in 35% reduction of luciferase activity compared with control mimic delivered cells (Fig. 2B). To test whether the binding was specific to the predicted binding site, similar reporter assay was performed using the construct carrying the mutation of the predicted site (supplemental Fig. S3B). Mutation of the predicted binding site significantly abolished miR-199a-5p-associated inhibition of luciferase activity, suggesting that miR-199a-5p specifically binds to the predicted binding site in Ets-1 3′-UTR exerting translational repression (Fig. 2B). Western blot analysis suggested that miR-199a-5p negatively regulates Ets-1 expression in endothelial cells. Delivery of miR-199a-5p blunted Ets-1 expression in HMECs by 65% (Fig. 2C). On the other hand, repression of endogenous miR-199a-5p significantly induced Ets-1 expression by 2.5-folds (Fig. 2D). These observations were consistent with the findings from immunocytochemistry. Ets-1 protein expression in both cytosolic (signal co-localized with Phalloidin) and nuclear (signal co-localized with DAPI) compartment of endothelial cells is down-regulated by exogenous miR-199a-5p (Fig. 2E). Similarly, primary endothelial cells with miR-199a-5p mimic delivered exhibited attenuated Ets-1 expression (supplemental Fig. S1E). It was thus identified that Ets-1 is a bona fide target of miR-199a-5p in human dermal microvascular endothelial cells.
Ets-1 is a multifaceted transcription factor controlling the expression of genes essential for cell migration and motility. In particular, angiogenesis requires modulation of extracellular matrix facilitating endothelial migration. Previous investigations revealed that MMP-1 is one of the major Ets-1 downstream mediators enabling endothelial migration (25). Delivery of siRNA against Ets-1 or MMP-1 alone recapitulated the angiostatic effect of miR-199a-5p (supplemental Fig. S4). Knocking down Ets-1 in HMECs significantly inhibited MMP-1 mRNA expression, exerting 75% repression (Fig. 3A). Intriguingly, delivery of synthetic miR-199a-5p suppressed MMP-1 expression mimicking the effect of Ets-1 knockdown (Fig. 3B). Inhibition of endogenous miR-199a-5p expression induced MMP-1 expression by 45% in endothelial cells (Fig. 3C). These results were in good agreement with the MMP-1 protein analysis using immunocytochemistry (supplemental Fig. S5). Primary dermal microvascular endothelial cells with miR-199a-5p mimic delivery exhibited reduced MMP-1 expression (supplemental Fig. S1F). Other Ets-1-associated MMPs such as MMP-3 and MMP-9 were not involved in miR-199a-5p-dependent angiostatic effects (supplemental Fig. S6).
To test the hypothesis that Ets-1 serves as a downstream mediator of miR-199a-5p signaling in endothelial cells, HMECs were subjected to Ets-1 overexpression followed by delivery of miR-199a-5p mimic. Overexpression of Ets-1 not only significantly induced Ets-1 expression (supplemental Fig. S7A) but also rescued the angiostatic effects of miR-199a-5p mimic (Fig. 4A). On the other hand, treatment of Ets-1 siRNA inhibited the expression of Ets-1 in the presence of miR-199a-5p inhibitor (supplemental Fig. S7B) and blunted the pro-angiogenic response induced by anti-miR-199a-5p treatment (Fig. 4B). Similarly, miR-199a-5p-associated down-regulation of MMP-1 was reversed by Ets-1 overexpression (Fig. 4C). Delivery of Ets-1 siRNA to HMECs blunted miR-199a-5p inhibitor-dependent induction of MMP-1 (Fig. 4D). These data suggest that Ets-1 serves as a downstream mediator of miR-199a-5p-dependent angiostatic control.
To connect the aforementioned observations to in vivo angiogenesis, we asked whether the miR-199a-5p-Ets-1-MMP-1 pathway takes place during postnatal angiogenesis such as during cutaneous wound healing. Our group previously reported that the miR-dependent angiogenic signals emerge during inflammatory phase (5). We thus studied the expression level of miR-199a-5p day 3 postwounding in the excisional cutanous wound model. Data from dermal miR expression studies revealed a down-regulation of miR-199a-5p in response to wounding compared with intact skin in day 3 (Fig. 5A). These data are further supported by the real time PCR analysis of LCM endothelial tissue element from both skin and day 3 wound edge tissue, indicating that the expression of endothelial miR-199a-5p is down-regulated in response to wounding (Fig. 5B). The repression of miR-199a-5p was negatively associated with the expression of Ets-1 and MMP-1. Wounding significantly induced endothelial Ets-1 expression in day 3 as evidenced from double-immunostaining (Fig. 5, C and D). The induction of Ets-1 was sustained until day 7 postwounding. Western blot analyses demonstrated that MMP-1 protein expression was significantly up-regulated in day 7 postwounding (Fig. 5E).
To test the significance of Ets-1 in wound angiogenesis, we studied wound closure and wound angiogenesis in mice carrying homozygous deletions in the Ets-1 gene (Ets-1−/−) (Fig. 6A). Genetic deletion of Ets-1 revealed significant reduction of Ets-1 protein expression in the skin tissue compared with their heterozygous (Ets-1+/−) and wild type (Ets-1+/+) littermates (Fig. 6B). Ets-1−/− mice exhibited compromised wound closure response (Fig. 6C). Consistent with this finding, histological analysis using hematoxylin-eosin (Fig. 7A) and Masson trichrome staining (Fig. 7B) showed significant impairment in re-epithelialization and collagen deposition in the wound edge tissue from Ets-1−/− mice. Interestingly, Ets-1−/− mice exhibited lower wound site blood flow as evidenced by laser Doppler analysis (Fig. 8A). This observation is in good agreement with the analysis of endothelial cell abundance using immunohistochemistry (Fig. 8B). Wound edge tissue from Ets-1−/− mice exhibited significant loss (by 65%) of endothelial cells compared with that in Ets-1+/− and Ets-1+/+ mice. Such compromised angiogenic response was associated with blunting of wound-induced MMP-1 expression (Fig. 8C).
miR-199a-5p, first characterized in 2003, is derived from two genetic loci in human genome (chromosome 19 for miR-199a-1; chromosome 1 for miR-199a-2). It is well documented that inducible miR-199a-5p arrests cell proliferation and contributes to cell death (7, 8, 26, 27). Our group and others have previously reported that miR-199a-5p regulates cellular detoxifying systems by targeting multidrug resistance-associated protein 1 (6) and CD44 (12). In cardiovascular biology, although miR-199a-5p is known to play a critical role in regulating the function of cardiomyocytes (7, 8, 28–31), information on its significance in vascular biology is scanty. This study provides the first evidence demonstrating that endogenous miR-199a-5p blocks angiogenic response by targeting master transcription factor Ets-1 in endothelial cells.
Ample evidence supports the notion that miRs controlling cell migration serve as endogenous regulators of angiogenesis in endothelial cells. miR-125b, for example, blocks epithelial cell migration (32, 33) and inhibits angiogenic response by targeting VE-Cadherin (34). miR-27b, a miR facilitating breast cancer cell invasiveness (35), promotes angiogenesis in endothelial cells by targeting angiostatic protein semaphorin 6A (36). miR-107, a miR down-regulated by hypoxia, attenuates cell migration (37, 38) and blocks VEGF-dependent angiogenesis by targeting hypoxia-inducing factor (HIF)-1β (39). The current study, together with findings from previous reports, reinforces the concept that miR-199a-5p not only inhibits cell migration (9, 13, 16) but also exerts potent angiostatic effect on endothelial cells by targeting a discrete subset of gene(s).
Ets-1 serves as a master transcription factor regulating angiogenic gene expression in endothelial cells including MMPs, urokinase type plasminogen activator, and vascular endothelial growth factor receptor 2 (3, 25). Induction of Ets-1 expression has been reported during tissue repair such as skin burn wounds (40), gastric ulcer (41), and aortic injury (42) aiding angiogenesis. Pro-angiogenic stimuli induce the expression of Ets-1 via transcriptional control. Moderate hypoxia, a potent angiogenic stimulus, induces Ets-1 promoter activity and Ets-1 expression via activation of HIF (43). Hydrogen peroxide, a well known pro-angiogenic stimulus (44, 45), activates nuclear factor (erythroid-derived 2)-like 2, leading to binding to antioxidant response elements and subsequent enhancement of Ets-1 promoter activity (46). Retinoic acid induces angiogenesis (47), with concomitant induction of Ets-1 transactivation (48). The findings from the current investigation, which agree with our previously published work (3), suggest an additional mechanism: Ets-1 expression is regulated through post-transcriptional modification in response to angiogenic stimulation. Indeed, miRs targeting Ets-1, namely miR-125b (49) and miR-222 (50), was also reported to serve as a potent angiostatic signal in endothelial cells (34, 51, 52), suggesting a sophisticated regulatory circuit in fine-tuning the behavior of endothelial cells in response to angiogenic signal.
In this study, we connect our in vitro observation to in vivo murine wound healing outcomes and report for the first time the involvement of miR-199a-5p-Ets-1 pathway in regulating MMP-1 expression in the context of cutaneous wound angiogenic response. Using the genetic approach, we demonstrate that the loss of Ets-1 compromises wound angiogenesis, compromises granulation tissue formation, and impairs wound closure. Intriguingly, mice with genetic deletion of Ets-1 show normal keratinocyte proliferation and differentiation (53), suggesting that the nonhealing phenotype of Ets-1 knock-out mice is likely independent of keratinocyte malfunction. We further characterized that MMP-1, which is reported to be vital in wound re-epithelialization (54), is specific to the miR-199a-5p-Ets-1 pathway. Consistent with findings from previous reports (25), Ets-1 regulates other MMP genes such as MMP-3 and MMP-9 in HMECs. However, the expression of these genes is not solely controlled by Ets-1. Instead, these genes are modulated by other transcription factors or transrepressors that might be targeted by miR-199a-5p, resulting in an off-set effect of the expression level.
A number of studies have reported the molecular mechanism of miR-199a-5p down-regulation in response to extracellular stimuli. Hypoxia induces the loss of miR-199a-5p in cardiomyocytes (7) and endothelial cells (14). Inhibition of HIF-1α activation resulted in miR-199a-5p induction in human colon cancer cells (55). Interestingly, down-regulation of miR-199a-5p derepressed HIF-1α (7), resulting in a positive feed forward loop under modest low oxygen environment. Apart from the HIF-dependent pathway, Akt is another intracellular signal that is implicated in miR-199a-5p down-regulation. Insulin stimulation inhibits the expression of miR-199a-5p with the concomitance of Akt activation (8). Constitutively active Akt repressed miR-199a-5p expression and was associated with accumulation of miR-199a-5p target HIF-1α (8). Given the fact that activation of both HIF (56, 57) and Akt (58, 59) takes place in the early stage of skin wound healing, it is tempting to speculate that these two pathways might act simultaneously to suppress miR-199a-5p expression, resulting in derepression of Ets-1 and enabling wound angiogenesis in a MMP-1-dependent mechanism. Further investigation is required to characterize the upstream stimuli in regulating the wound-associated down-regulation of miR-199a-5p.
In summary, our results demonstrate a novel post-transcriptional control of Ets-1 expression by miR-199a-5p, and the regulation of its associated downstream mediator MMP-1. These findings reinforce the notion that miRs serve as common endogenous signals in regulating the motility of both epithelial and endothelial cells by targeting discrete set of genes. This work also provides the first evidence that wound-associated down-regulation of miR-199a-5p facilitates angiogenesis via desilencing Ets-1. Taken together, this investigation provides novel mechanistic insight explaining miR-dependent regulation of wound angiogenesis and provides the foundation of developing therapeutic intervention in treating complications of vasculopathy such as chronic nonhealing wounds.
We thank Dr. Michael Ostrowski for providing Ets-1 cDNA clone and Ets-1 knock-out mice as generous gifts.
*This work was supported, in whole or in part, by National Institutes of Health Grants GM 077185 and GM069589 (to C. K. S.) and DK076566 (to S. R.).
This article contains supplemental Table S1 and Figs. S1–S7.
2The abbreviations used are: