Abstract Pathological neovascularization is a hallmark of many vision-threatening diseases. However, some patients exhibit poor responses to current anti-VEGF therapies due to resistance and limited efficacy. Recent studies have highlighted the roles of noncoding RNAs in various biological processes, paving the way for RNA-based therapeutics. In this study, we report a marked down-regulation of miR-205 under pathological conditions. miR-205 potently inhibits endothelial cell functions critical for pathological neovascularization, including proliferation, migration, and tube formation. Furthermore, miR-205 strengthens the endothelial barrier, thereby reducing vascular leakage. In mouse models of retinal and choroidal neovascularization, miR-205 administration effectively suppresses abnormal blood vessel formation and leakage. Mechanistically, miR-205 directly targets VEGFA and ANGPT2, which are key drivers of pathological neovascularization. To improve delivery, we successfully loaded miR-205 into exosomes derived from mesenchymal stem cells. This innovative approach avoids cytotoxicity while preserving therapeutic efficacy in both cellular and animal models. Collectively, our findings highlight miR-205 as a promising therapeutic for ocular neovascularization, with exosome delivery offering a novel and efficient strategy for treating vision-threatening vascular diseases. Graphical Abstract [40]graphic file with name 12951_2024_3079_Figa_HTML.jpg Supplementary Information The online version contains supplementary material available at 10.1186/s12951-024-03079-y. Keywords: Ocular neovascularization, miR-205, VEGFA, ANGPT2, Exosomes Introduction Neovascularization is a prevalent pathological feature in various eye diseases, such as retinopathy of prematurity (ROP), diabetic retinopathy, and age-related macular degeneration (AMD) [[41]1]. This condition is characterized by increased vascular permeability and fragility, resulting in significant tissue swelling, bleeding, scarring, retinal detachment, and permanent vision impairment [[42]2]. The primary strategy for managing neovascularization involves the application of anti-VEGF drugs, which effectively block the binding of VEGF to its receptors [[43]3]. Although anti-VEGF therapies are beneficial for many patients, their effectiveness is not uniform across all individuals [[44]4]. Moreover, the need for frequent and sustained injections can increase the risk of both ocular and systemic complications [[45]5, [46]6]. Pathological neovascularization is intricately regulated by a delicate balance of pro- and anti-angiogenic factors within physiological processes. However, this equilibrium is disrupted in disease states. This disruption can arise from an overproduction of pro-angiogenic signals or a suppression of anti-angiogenic factors, resulting in pathological neovascularization [[47]7]. VEGF is a crucial mediator of neovascularization. Its overexpression is frequently observed in various pathological conditions. In addition to VEGF, other angiogenic factors contribute significantly to the intricate dance of blood vessel development. For instance, angiopoietin-2 (ANGPT2) is a key player that can shift the balance toward angiogenesis by acting synergistically with VEGF. Fibroblast growth factor (FGF) is another important contributor, with multiple subtypes that stimulate endothelial cell proliferation and migration, thereby promoting angiogenesis [[48]8]. Platelet-derived growth factor (PDGF) is important for recruiting pericytes and facilitating the maturation of new vessels [[49]9, [50]10]. Additionally, a variety of other molecules, including cytokines, chemokines, and extracellular matrix proteins, play a role in regulating neovascularization [[51]11, [52]12]. They interact within a complex network of signaling pathways that can either promote or inhibit the formation of new blood vessels. Understanding the interplay between these factors is essential for developing effective treatments for ocular vascular diseases [[53]13, [54]14]. Elevated levels of ANGPT2 have been observed in the aqueous and vitreous humor of patients with AMD, proliferative diabetic retinopathy (PDR), and retinal vein occlusion (RVO). In the presence of VEGF, ANGPT2 binds to Tie2, inducing endothelial cell migration and proliferation. This interaction disrupts Tie2 clusters formed by ANGPT1, thereby inhibiting its stabilizing effects on vascular integrity. The resulting consequences include abnormal vascular formation, increased vascular permeability, and heightened inflammation [[55]15]. FGF-2 signaling has been recognized to interact with VEGF signaling in endothelial cells. By activating ERK1/2, FGF-2 stimulates the expression of both VEGF and VEGFR2, consequently promoting neovascularization [[56]16]. Additionally, PDGF promotes endothelial cell proliferation, migration, and VEGF expression [[57]17]. Given the involvement of multiple pro-angiogenic factors in ocular diseases, simultaneous targeting of these factors may be essential to achieve optimal therapeutic benefits. Nucleic acid therapies, including small interfering RNAs (siRNAs), microRNAs (miRNAs), and antisense oligonucleotides (ASOs), have gained prominence for their precision in treating diseases at the genetic level. These therapies use sequence-specific binding to target genes, ensuring high specificity for intended genetic sequences and enabling targeted interventions. A significant advantage of nucleic acid drugs is their broad applicability, as they can reach a range of targets that are often inaccessible to traditional small-molecule and protein therapies due to challenges in engaging specific protein structures. This versatility positions nucleic acid drugs as a cutting-edge modality in the therapeutic landscape [[58]18]. Nucleic acid drugs operate at the genetic level, targeting the root causes of diseases. Their extended half-life is an additional benefit, allowing for repeated action on target genes and leading to a prolonged therapeutic impact. This durability has the potential to reduce the frequency of administration, thereby enhancing patient compliance while maintaining continuous efficacy [[59]19]. miRNAs are known as the non-coding RNA molecules, typically 20 to 24 nucleotides in length. They are found across diverse species and influence gene expression by inhibiting translation or causing post-transcriptional degradation of target mRNAs. A key feature of miRNAs is their capacity to target multiple mRNAs simultaneously, often by binding to 3’ untranslated regions (3’-UTRs) of these mRNAs. This multi-targeting capability highlights their potential for broad-spectrum therapeutic intervention, positioning them as a promising tool for treating complex diseases influenced by multiple genetic pathways [[60]20]. However, the delivery of nucleic acids faces challenges due to their polarity, negative charge, and susceptibility to degradation [[61]21]. Existing delivery systems, such as synthetic nanoparticles, have limitations, including immune reactions, poor cellular absorption, and the necessity for modifications to target specific tissues without causing side effects. Exosomes are tiny cellular messengers, ranging from 30 to 150 nanometers in size, enclosed by a single membrane [[62]22]. They play important roles in cell-to-cell communication by shuttling key biomolecules between cells [[63]23]. Compared to other delivery methods, exosomes offer several advantages. They exhibit low immunogenicity, demonstrate exceptional biocompatibility, and possess good biological safety. Furthermore, exosomes can easily traverse challenging barriers, such as blood-brain barrier [[64]24, [65]25]. Their unique properties enhance the stability of encapsulated nucleic acid drugs and facilitate their passage through biological barriers [[66]26]. Exosomes have emerged as promising drug delivery systems, with successful applications in siRNA delivery for pancreatic cancer treatment. Studies in mice have shown promising results, paving the way for ongoing clinical trials. Furthermore, incorporating pigment epithelium-derived factor (PEDF) can enhance its anti-angiogenic, anti-inflammatory, and neuroprotective effects by improving stability and bioavailability. These advancements highlight the immense potential of exosomes to revolutionize drug delivery systems [[67]27]. In this study, we used mouse models of OIR and CNV to investigate the role of miR-205 in ocular vasculopathy. This study reveals that increasing miR-205 levels significantly inhibits pathological neovascularization and vascular leakage by simultaneously targeting VEGFA and ANGPT2. We also examined the therapeutic potential of small extracellular vesicles derived from mesenchymal stem cells and enriched with miR-205, referred to as Exo-miR-205, in the context of neovascular ocular diseases. Exo-miR-205 significantly inhibits both neovascularization and leakage, highlighting its potential as a promising therapeutic candidate for treating ocular vascular diseases. Materials and methods Cell culture and treatment Human retinal vascular endothelial cells (HRVECs, ACBRI-181, Cell Systems, USA) were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM, [68]C11995, Gibco, USA) supplemented with 10% fetal bovine serum (FBS, 11573397, Gibco, USA) and 1% penicillin-streptomycin. They were transfected with negative control (NC) mimic, miR-205 mimic, NC inhibitor, and miR-205 inhibitor using Lipofectamine 3000 (L3000001, Invitrogen, USA), or left untreated (Ctrl) for 6 h. Following transfection, they were treated with or without VEGF (20 ng/mL) for 48 h. Human umbilical cord-derived mesenchymal stem cells (hUMSCs) were cultured in a medium consisting of basal medium (RP020101, Nuwacell Biotechnology, China) and supplement (RP02010-2, Nuwacell Biotechnology, China) without FBS. All cells were maintained at 37 °C in an incubator with a 5% CO[2] atmosphere. Animals Eight-week-old male C57BL/6J mice (weighing 20–25 g) were acquired from Hangzhou Ziyuan Experimental Animal Technology Co., Ltd (Hangzhou, China). They were housed in a controlled environment, featuring a 12 h light/dark cycle, a room temperature of 25 ± 2 °C, and a humidity level of 50 ± 10%. The animal experiments received approval from the Animal Ethics and Experimentation Committee of the author’s institute. All protocols involving the mice were adhered to the ARVO Statement for the Utilization of Animals in Ophthalmic and Vision Research. 5-Ethynyl-2′-deoxyuridine (EdU) assay EdU assay was performed to assess cell proliferation using an EdU kit (C0071S, Beyotime, China) according to the manufacturer’s instruction. Cells were incubated with 10 µM EdU agent for 2 h. Then, these cells were washed with PBS and fixed in 4% paraformaldehyde for 15 min. After permeabilization with 0.5% Triton X-100 for 10 min, these cells were incubated with the Click reaction buffer in the dark at room temperature for 30 min. Subsequently, the nuclei were stained with DAPI for 5 min. Fluorescence microscopy (IX73, Olympus, Japan) was used to capture the images. Cell proliferation was determined by calculating the average ratio of EdU-positive to DAPI-positive cells. Transwell assay Transwell assay was performed to detect cell migration ability. After the required treatment, HRVECs were adjusted to a concentration of 4 × 10^5 cells/mL. Then, 600 µL of complete growth medium was added into the lower compartment of a Transwell plate (MCEP24H48, Millipore, USA). Next, 100 µL of cell suspension was added to the upper chamber and allowed to incubate overnight at 37 °C. The cells were fixed with 4% paraformaldehyde for 30 min, followed by staining with 0.1% crystal violet for 25 min. Excess cells on the chamber’s surface were wiped off with a cotton swab prior to examination. The images were taken by an Olympus microscope (IX51, Japan). Tube formation assay Tube formation assay was performed to detect the tube-formation ability of endothelial cells. The Matrigel (354234, Corning, USA) was placed in 24-well tissue culture plates (50 µL per well) and incubated at 37 °C for at least 30 min to form a gel layer. The cell suspension was then layered on top of Matrigel-coated plates and incubated at 37 °C. Tube formation was imaged at 10 × magnification by a light microscope. Tube formation ability was analyzed and quantified using Image J software. Spheroid-based sprouting assay Spheroid sprouting assay was performed to detect the sprouting ability of endothelial cells. Cell suspension (8 × 10^4) was added to 5 mL of the complete medium (with 10% FBS) containing 20% methocel (Sigma Aldrich, USA). Drops of 25 µL cell solution was placed onto the lid of a cell culture dish (100 × 20 mm) and incubated upside-down for 24 h. Subsequently, the hanging drops were rinsed off using PBS and the spheroids were re-suspended in the solution containing 20% FBS and 80% methocel. The collagen mix was prepared on ice including Collagen type I (1 mg/mL, 3440-100-01, R&D Systems, USA) and PBS in a 2:3 ratio. The collagen solution was combined with spheroid solution in a 1:1 ratio and added to a 24-well plate for 24 h. Spheroid sprouting were observed by an Olympus microscope and the average sprouting length was determined using the ImageJ software. Laser-induced choroidal neovascularization (CNV) model Mice were anesthetized by intraperitoneal injection of 1.25% tribromoethanol (0.2 mL/10 g, Nanjing Aibei Biotechnology, China) and 1% tropicamide was used to dilate the pupils of the eyes. Subsequently, the fundus was viewed with an imaging camera. A 577 nm laser (power of 140 mW, diameter of 50 μm, exposure time of 0.1 s) was employed to induce laser photocoagulation at a distance of 2 papillary diameter (PD) around optic disc. A total of four laser burns were photocoagulated at equal distances from optic nerve head in each eye. Bubble generation was used as a sign that the Bruch’s membrane had been broken. In CNV model, C57BL/6J mice received a single intravitreous injection of negative control (NC) agomir, miR-205 agomir, NC antagomir, miR-205 antagomir, or PBS immediately after laser injury. After 7 days, choroidal tissues were collected for the subsequent experiment. Oxygen-induced retinopathy (OIR) model C57BL/6J pups and their nursing mothers were exposed to a hyperoxic environment (75% oxygen) from postnatal day 7 (P7) to P12. After this exposure, the pups were returned to normoxic conditions on P12. On the same day, P12 pups received a single intravitreal injection of either negative control (NC) agomir, miR-205 agomir, NC antagomir, miR-205 antagomir, or PBS. Retinas were collected on P17 for subsequent analysis. Evans blue assay Retinal vascular permeability was determined by Evans blue assay. After administering intraperitoneal anesthesia with 1.25% tribromoethanol, Evans blue dye (45 mg/kg) was injected into the femoral vein. Following a 1 h circulation period, the mice were euthanized by cervical dislocation under deep anesthesia. The eyes were promptly enucleated and fixed in a 4% PFA for 1 h at room temperature. After excising the cornea and lens, the retinas were gently separated from eye cups. Following these steps, retinal specimens were captured under a fluorescent microscope. The resulting images were then meticulously analyzed using ImageJ software to quantify and interpret the data [[69]28, [70]29]. Details on the quantitative measurement of EB leakage fluorescence intensity using ImageJ are provided in the supplementary material. Fluorescein fundus angiography (FFA) Fluorescein fundus angiography (FFA) was performed to observe retinal vascular permeability. Following intraperitoneal anesthesia with 1.25% tribromoethanol, pupil dilation was achieved using 0.5% phenylephrine and 0.5% tropicamide. Subsequently, an intraperitoneal injection of approximately 0.2 ml of 2% fluorescein sodium was administered. The images of retinal fundus were obtained and quantified using Multi-Modal Ophthalmic Imaging System for Animal (Saris, Robotrak, China). The permeability was analyzed with Image J. Identification the target genes of miR-205 The potential target genes of miR-205 were predicted using TargetScan, miRtarbase, and miRDB. Venn diagram was performed to identify the common target genes. The DAVID database was used to conduct gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analyses on these common target proteins to obtain the related pathways. P < 0.05 was deemed statistically significant. The results of GO analyses were output in the form of bar chart and the results of KEGG pathway analyses were showed by bubble chart. Exosome isolation and identification Exosomes were isolated from the culture medium of hMSCs using differential ultracentrifugation method. Following incubation at 37 °C with 5% CO[2] for 72 h, the culture medium of hMSCs was collected and subjected to sequential centrifugation step at 300 × g for 10 min, 2,000 × g for 10 min, 10,000 × g for 30 min to remove the live cells, cell debris, and micro-vesicles, respectively. Subsequently, the filtered supernatant was ultra-centrifuged at 100,000 × g for 1 h at 4 °C. The extracellular vesicles were then re-suspended in PBS and stored at -80 °C for further experiments. The nanoparticle tracking analysis (NTA, Particle Metrix, Germany) was performed to determine the size distribution of exosomes derived from MSCs. The high-resolution transmission electron microscopy (TEM, FEI, Tecnai G2, USA) was used to observe the morphology of exosomes. To test cellular uptake of exosomes by HRVECs, exosomes were labeled with anti-CD63 antibody (200 µg/mL, 1:50, sc-5275, Santa Cruz, USA). CD63-labeled exosomes were incubated with HRVECs for 24 h in free medium. Cells were then washed with PBS and fixed with 4% paraformaldehyde for 30 min. Subsequently, the cells were incubated overnight at 4 °C with the Alexa Fluor 488-labeled secondary antibodies (1:500, Invitrogen, USA). The cells were stained with phalloidin and the nuclei were stained with DAPI. The photographs were captured using a fluorescence microscope. Preparation of mir-205-loaded exosome (exo-miR-205) The exosomes were loaded with miR-205 mimics (Exo-miR-205) or NC mimic (Exo-NC) by electroporation using the CUY21EDIT II (BEX, Japan) electroporation system. The electroporation mixture was prepared by mixing exosomes and miR-205 in a 1:1 (wt/wt) ratio in PBS, with final concentration of exosomes was 1 mg/mL. The mixture was transferred into ice-cold cuvettes and electroporated for 10 cycles with a perforation voltage of 110 V, a perforation opening time of 6 ms, a perforation interval of 10 ms, a penetration voltage of 25 V, and a capacitance of 940 µF. Following electroporation, the mixture was transferred to a fresh tube and incubated at 37 °C for 30 min to facilitate the uptake of the mimics. Subsequently, any unincorporated mimics or NC mimics were removed through ultracentrifugation. Cellular uptake in vitro The co-culture assay was performed to evaluate the uptake of Exo-miR-205. miR-205 was fluorescently labeled with FAM (Fluorescein Amidite) for tracking. Following electroporation, the Exo-miR-205-FAM was tagged using the Evlink reagent (EL012100210, illuTINGO, China) to enhance the detection. HRVECs were co-cultured with the Evlink-labeled exosomes in a dark environment for 24 h to facilitate internalization. The nuclei were stained with DAPI. A fluorescence microscopy was used to observe exosome internalization by HRVECs. Ocular distribution To evaluate exosome distribution in CNV model, Exo-miR-205 labeled with Evlink was intravitreally injected immediately following laser-induced injury. Three days post-injection, choroids were carefully isolated for further analysis to assess exosome localization and distribution within ocular tissues. Delivery efficiency was examined using bioluminescence imaging in Institute of Cancer Research (ICR) mice with the IVIS Spectrum In Vivo Imaging System (IVIS^® Spectrum, PerkinElmer, USA). Cy5-labeled miR-205 or Exo-miR-205 was intravitreally administered, and at designated intervals (6 h and 24 h post-injection), the mice were euthanized. Their eyeballs were harvested for imaging using the IVIS Spectrum. Electroretinogram (ERG) An ERG assay was performed to assess functional damage in retinal neurons. After 12 h of dark adaptation, scotopic full-field ERG responses were measured under dim red lighting. Mice were anesthetized via intraperitoneal injection and positioned on a platform, with tropicamide drops applied for pupil dilation. Electrodes were carefully placed on the mouse mandible and corneal surface. Stimulation was performed with a flash intensity of 4 cd·s·m^–2. The initial negative peak was defined as the a-wave, followed by the positive b-wave peak. ERG responses were recorded using a Roland Consult Color Ganzfeld Q450C recording system. TUNEL staining TUNEL staining was performed to label apoptotic cells in vivo, assessing the potential toxicity of Exo-miR-205. Retinas were embedded in paraffin and sectioned at a thickness of 5 μm. After extending, mounting, and baking the sections, they were incubated with a TUNEL reaction mixture (C1088, Beyotime, China) for 1 h at 37 °C in the dark. The sections were then counterstained with DAPI for 10 min. Images were captured using a fluorescence microscope, and the apoptotic index was calculated by determining the percentage of TUNEL-positive cells. Hematoxylin-eosin (HE) staining Histological structural changes were examined using hematoxylin and eosin (HE) staining. Mouse eyeballs were enucleated and fixed in 4% paraformaldehyde overnight, then embedded in paraffin and sectioned at a thickness of 5 μm. The paraffin sections were stained with hematoxylin and eosin, and images were captured under a light microscope. Statistical analysis Statistical analyses were performed using GraphPad Prism version 9.0, with data presented as mean ± standard deviation (SD). Differences between groups were evaluated using Student’s t-test for pairwise comparisons and one-way ANOVA for comparisons among three or more groups. P < 0.05 was considered statistically significant. Result miR-205 level was down-regulated during pathological ocular neovascularization To explore the role of miR-205 in pathological ocular neovascularization, we initially examined its expression in both the OIR and laser-induced CNV models. In the OIR model, miR-205 levels were found to be reduced in the retinas from the hyperoxia group at postnatal day 17 (P17) when compared to the normoxic control (Fig. [71]1A and B). Similarly, in the CNV model, we found that miR-205 expression was markedly lower, reinforcing the notion that decreased levels of miR-205 could be associated with the pathological processes driving ocular neovascularization (Fig. [72]1C and D). HRVECs were treated with VEGF to simulate angiogenic condition. qRT-PCR assays were performed to detect miR-205 expression levels across different groups. miR-205 expression level was significantly reduced in HRVECs following VEGF treatment for 48 h (Fig. [73]1E and F). Collectively, these findings suggest that miR-205 plays a critical role in the context of pathological ocular neovascularization. Fig. 1. [74]Fig. 1 [75]Open in a new tab miR-205 level was down-regulated during pathological ocular neovascularization A Schematic diagram of OIR model. B qRT-PCR analysis of miR-205 expression in OIR model. C Schematic diagram of CNV model. D qRT-PCR analysis of miR-205 expression in CNV model. E, F qRT-PCR assays were performed to detect the expression of miR-205 in HRVECs following VEGFA (20 ng/mL) treatment for 48 h. Data are expressed as the mean ± SD. *P < 0.05, n = 5, Student’s t test miR-205 regulates endothelial angiogenic effects in vitro Endothelial cells (ECs) are crucial for vascular system, as they respond to various stimuli to maintain vascular homeostasis. In this study, we used miR-205 mimics and inhibitors to modulate miR-205 levels in HRVECs. The transfection efficiency was assessed using qRT-PCR assays. The results demonstrated that the introduction of miR-205 mimics resulted in a significant increase in miR-205 expression levels, confirming the successful modulation of miR-205 in HRVECs (Fig. [76]S1A). The success of miR-205 inhibitor transfection into HRVECs was confirmed by immunofluorescence staining (Fig. [77]S1B). EdU assay demonstrated that the transfection of miR-205 mimics significantly decreased the proliferation ability of HRVECs. This finding indicates that miR-205 may play a crucial role in inhibiting cellular growth and proliferation. In contrast, when miR-205 inhibitors were introduced, a notable enhancement in cell proliferation was observed, highlighting the opposing effects of miR-205 on HRVECs (Fig. [78]2A). Transwell migration assays revealed that the overexpression of miR-205 was associated with a marked reduction in the migratory capacity of HRVECs. Conversely, the application of miR-205 inhibitors led to a significant increase in cell migration (Fig. [79]2B). Tube formation assays revealed that elevated levels of miR-205 suppressed the formation of tube-like structures, indicating a potential role in inhibiting angiogenic process. In contrast, the introduction of miR-205 inhibitors contributed to tube formation, suggesting that lower levels of miR-205 facilitate the development of vascular structures (Fig. [80]2C). Moreover, spheroid sprouting assays provided additional evidence of miR-205’s role in angiogenesis. Overexpression of miR-205 resulted in decreased sprouting length suggesting a suppressive effect on angiogenic sprouting. In contrast, the use of miR-205 inhibitors stimulated an increase in sprouting length, further demonstrating the regulatory influence of miR-205 on angiogenic behavior (Fig. [81]2D). Collectively, these results underscore the role of miR-205 in regulating angiogenic effects. Fig. 2. [82]Fig. 2 [83]Open in a new tab miR-205 regulates endothelial angiogenic effects in vitro. HRVECs were transfected with negative control (NC) mimic, miR-205 mimic, NC inhibitor, miR-205 inhibitor, or left untreated (Ctrl) for 6 h, and then treated without or with VEGF (20 ng/mL) for 48 h. A EdU assay was conducted to assess the proliferation of HRVECs. Scale bar, 20 μm. B Transwell assay was performed to detect the migration of HRVECs. Scale bar, 20 μm. C Tube formation ability was detected by Matrigel assays. Scale bar, 100 μm. D Spheroid-based sprouting assay was performed to evaluate the sprouting ability of HRVECs. Scale bar, 50 μm. Data are expressed as the mean ± SD. *P < 0.05, n = 4, One-way ANOVA test miR-205 plays an anti-angiogenic role in pathological neovascularization in vivo To determine the role of miR-205 in pathological neovascularization in vivo, we constructed two animal models of pathological neovascularization: CNV and OIR mouse models. We used miR-205 agomir and antagomir to regulate the expression levels of miR-205. qRT-PCR analysis confirmed increased miR-205 levels in both the retina and choroid following intravitreal injections of miR-205 agomir (Fig. [84]S2A, B). In the OIR model, the injection of miR-205 agomir significantly reduced both avascular areas and neovascular areas compared to the controls, while injection of miR-205 antagomir had an opposing effect (Fig. [85]3A-C). In the CNV model, miR-205 agomir administration led to a significant reduction in CNV regions, while miR-205 antagomir contributed to ocular neovascularization (Fig. [86]3D and E). Choroidal explant sprouting assays demonstrated reduced sprouting capacity in the presence of miR-205 agomir (Fig. [87]3F and G). Collectively, these findings indicate that miR-205 has an inhibitory effect on pathological ocular neovascularization. Fig. 3. [88]Fig. 3 [89]Open in a new tab miR-205 plays an anti-angiogenic role in pathological neovascularization in vivo. In the OIR model, pups at postnatal day 12 (P12) received intravitreal injections of negative control (NC) agomir, miR-205 agomir, NC antagomir, miR-205 antagomir, or PBS. The retinas were harvested at postnatal day 17 (P17) and subsequently stained with GS-IB4 to assess vascular changes. In the CNV model, C57BL/6J mice received intravitreal injection of NC agomir, miR-205 agomir, NC antagomir, miR-205 antagomir, or PBS immediately following laser injury. After a period of 7 days, choroidal tissues were collected for GS-IB4 staining to evaluate the extent of neovascularization. A-C Retinal vaso-obliteration (VO) and neovascularization were detected by GS-IB4 staining. The avascular area is indicated by the margin, while the pathological angiogenic area at P17 is highlighted in yellow. Scale bar: 200 μm. D, E Laser-induced CNV model was used to assess the effect of miR-205 expression on ocular neovascularization, with CNV formation observed through GS-IB4 staining. Images of flat-mounted choroid were captured using a fluorescence microscope. Scale bar: 100 μm. F, G Choroid sprouting assay was performed to examine the suppressive effect of miR-205 on choroidal neovascularization. Representative images, along with computerized quantification of choroidal endothelial sprouts, are presented. Scale bar: 200 μm. Data are expressed as the mean ± SD. *P < 0.05, n = 5, One-way ANOVA test miR-205 suppresses vascular leakage in vivo and in vitro Pathological neovascularization is frequently associated with increased vascular permeability, leading to edema, hemorrhage, and eventual visual impairment [[90]30]. To investigate the role of miR-205 in vascular permeability in vivo, we employed both the OIR and CNV models. In the OIR model, Evans blue assays were conducted to assess vascular leakage. The results demonstrated a significant reduction in vascular leakage in mice treated with miR-205 agomir, while the use of antagomir exacerbated leakage (Fig. [91]4A and C). In the CNV model, the leakage of CNV lesions was assessed using FFA. The results showed that intravitreal injection of miR-205 agomir decreased vascular leakage, whereas miR-205 antagomir had the opposite effect (Fig. [92]4B and D). qRT-PCR and western blot assays were peformed to detect the expression levels of intercellular adhesion molecule-1 (ICAM-1), a key regulator of vascular permeability [[93]31]. The results showed that treatment with miR-205 mimics led to downregulation of ICAM-1, while the miR-205 inhibitor resulted in its upregulation, suggesting that miR-205 plays a regulatory role in ICAM-1 expression (Fig. [94]4E and F). Additionally, an Evans Blue-Bovine Serum Albumin (EB-BSA) cell permeability assay was performed to evaluate the barrier integrity of HRVECs. Overexpression of miR-205 improved barrier function and reduced permeability in VEGF-stimulated HRVECs, indicating a protective role for miR-205 in maintaining vascular integrity (Fig. [95]4G). Fig. 4. [96]Fig. 4 [97]Open in a new tab miR-205 suppresses vascular leakage in vivo and in vitro. In the OIR model, pups at postnatal day 12 (P12) received intravitreal injections of negative control (NC) agomir, miR-205 agomir, NC antagomir, miR-205 antagomir, or PBS. The retinas were harvested at postnatal day 17 (P17), and vascular leakage was detected using Evans blue assay. In the CNV model, C57BL/6J mice received intravitreal injections of NC agomir, miR-205 agomir, NC antagomir, miR-205 antagomir, or PBS immediately following laser injury. After a 7-day period, choroidal tissues were collected for FFA examination was conducted to evaluate the leakage of neovascularization. A, C OIR mice were perfused with Evans blue dye for 1 h, and fluorescence signals from flat-mounted retinas were detected by a fluorescence microscope. Scale bar, 200 μm. B, D FFA was performed to detect leakage in CNV lesions. Scale bar, 100 μm. E ICAM-1 expression was detected by qRT-PCRs, with β-actin as the internal control. F ICAM-1 expression was detected by western blots, with GAPDH as the internal control. G EB-transwell assay was performed to evaluate the permeability of HRVECs following miR-205 transfection. *P < 0.05; n = 4; One-way ANOVA miR-205 targets VEGFA and ANGPT2 to regulate pathological neovascularization To clarify the molecular mechanism through which miR-205 regulates pathological neovascularization, we used the bioinformatics tools (TargetScan, miRtarbase, and miRDB) to predict the potential target genes of miR-205. A venn diagram identified 47 overlapping genes (Fig. [98]5A). Enrichment analysis revealed the involvement of these genes in various GO terms related to pathological angiogenesis, negative regulation of cell proliferation, and negative regulation of vascular permeability, as well as in cellular components like bicellular tight junctions (Fig. [99]S3A). KEGG pathway analysis linked these genes to pathways such as the PI3K-Akt signaling pathway and the tight junction pathway (Fig. [100]5B). The enrichment analysis particularly highlighted genes relevant to vascular biology, including VEGFA and ANGPT2. Fig. 5. [101]Fig. 5 [102]Open in a new tab miR-205 targets VEGFA and ANGPT2 to regulate pathological neovascularization. A Venn diagram showing the intersection of predicted targets of miR-205 by TargetScan, miRtarbase, and miRDB databases. B Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis for the targeted genes of miR-205. C Luciferase activity was measured using the luciferase reporter assay in HRVECs transfected with either VEGFA-WT or VEGFA-Mut, along with either miR-205 mimic or negative control (NC) mimic. D Luciferase activity was measured using the luciferase reporter assay in HRVECs transfected with either ANGPT2-WT or ANGPT2-Mut, along with either miR-205 mimic or NC mimic. E Western blot was performed to detect the expression of VEGFA and ANGPT2. Data are expressed as the mean ± SD. *P < 0.05, n = 4, One-way ANOVA test To validate VEGFA and ANGPT2 as direct targets of miR-205, we conducted luciferase reporter assays. A significant reduction in luciferase activity was observed when miR-205 was co-expressed in HRVECs, indicating a direct interaction. This effect was lost when the binding site was mutated, further confirming the specificity of the interaction (Fig. [103]5C and D). Western blot analysis further confirmed that overexpression of miR-205 led to decreased expression levels of VEGFA and ANGPT2 in HRVECs (Fig. [104]5E). Consistent with the results from western blots, qRT-PCR assays demonstrated a similar trend in gene expression regulation (Fig. [105]S3B, [106]S3C). Collectively, these findings suggest that VEGFA and ANGPT2 are the target genes of miR-205, highlighting a potential regulatory network through which miR-205 exerts its inhibitory effects on pathological neovascularization. Evaluation of exosomes as a delivery vehicle for miR-205 in targeted therapy for ocular neovascularization This study indicates that miR-205, with its dual targeting capability on VEGFA and ANGPT2, plays a key role in pathological neovascularization, suggesting that miR-205 has emerged as a potential nucleic acid drug candidate for treating ocular neovascularization. However, due to the inherent instability of miRNA, it is crucial to select an appropriate delivery system. Exosomes, known for their low immunogenicity, high biocompatibility, and efficient delivery, have been gaining significant interest as drug carriers [[107]32]. To explore the potential of exosomes as a delivery vehicle for miR-205 to endothelial cells, we isolated exosomes using differential centrifugation. The integrity and purity of isolated exosomes were confirmed through NTA, TEM, and western blots. TEM assay revealed the characteristic cup-shaped morphology of exosomes (Fig. [108]S4A). NTA showed that their size was predominantly within the range of 100 to 150 nm (Fig. [109]S4B). Western blot analysis was performed to verify the presence of specific exosome markers such as CD63, Hsp70, and TSG101 (Fig. [110]S4C). Furthermore, we fluorescently labeled the exosomes with CD63 and stained the cytoskeleton with phalloidin to visualize their uptake in HRVECs. The effective internalization of exosomes by HRVECs was confirmed by a fluorescence microscopy (Fig. [111]S4D), demonstrating exosome is a promising delivery carrier for miR-205 to target cells within ocular vasculature. We explored three methods to encapsulate miR-205 mimic into exosomes: co-culture, self-assembly, and electroporation. The encapsulation efficiency was evaluated using qRT-PCR to measure miR-205 content within the exosomes. The data showed that electroporation was the most effective method, yielding a significantly higher concentration of miR-205 (Fig. [112]6A-C). The engineered exosome-miR-205 (Exo-miR-205) vesicles were characterized for their morphology, size, and protein markers. TEM images confirmed the characteristic cup-shaped structure of exosomes (Fig. [113]6D). NTA indicated that the particle size was within the typical 100–150 nm range for exosomes (Fig. [114]6E). Western blot analysis confirmed the presence of exosome-specific markers such as CD63, HSP70, and TSG101 (Fig. [115]6F). Fig. 6. [116]Fig. 6 [117]Open in a new tab Evaluation of exosomes as a delivery vehicle for miR-205 in targeted therapy for ocular neovascularization. A-C Three strategies were used to load miR-205 into exosomes: co-incubation, transfection, and electroporation. qRT-PCR assay was performed to detect the loading efficiency of miR-205 in exosomes. D TEM identified the morphology of Exo-miR-205. Scale bar, 100 nm. E NTA identified the size and particle concentration of Exo-miR-205. F Western blots were performed to detect exosome specific markers including HSP70, TSG101, and CD63 of Exo-miR-205. G Evlink-labelled exosomes and FAM-labelled miR-205 was performed to detect the distribution of Exo-miR-205. Representative images show the cellular uptake of Exo-miR-205 in HRVECs after 24 h incubation. Scale bar, 20 μm. H Evlink-labelled Exo-miR-205. CNV model mice received intraocular injections of Exo-miR-205. Representative images show the uptake of Exo-miR-205 in CNV lesion after 3 days. I Exo- miR-205-Cy5 and miR-205-Cy5 were injected into ICR mice to acquire the images with an IVIS imaging system. Fluorescence of the eyes was imaged at 6 h and 24 h after intraocular injections. Ctrl represents administration with PBS group. Data are expressed as the mean ± SD. *P < 0.05, One-way ANOVA test To investigate whether exosomes could effectively deliver miR-205 to HRVECs, we fluorescently labeled miR-205 with FAM and exosomes with Evlink. After co-culturing Exo-miR-205 with HRVECs for 24 h, HRVECs had successfully internalized the Evlink-labeled exosomes, with FAM-labeled miR-205 localized within these vesicles (Fig. [118]6G). We further assessed the delivery efficiency of Exo-miR-205 using fluorescence staining and the IVIS imaging system. In the CNV model, we observed significant enrichment of Evlink-labeled Exo-miR-205 within the lesions, as confirmed by fluorescence microscopy (Fig. [119]6H). Cy5-labeled miR-205 and Exo-miR-205-Cy5 were administered to ICR mice, an albino strain suitable for imaging with the IVIS system. Six hours post-injection, a prominent Cy5 signal was detected in the eyes of mice treated with Exo-miR-205-Cy5, while a weaker signal was observed in the miR-205-Cy5 group. Notably, the Cy5 signal persisted at 24 h, indicating that Exo-miR-205 exhibits superior stability. The results demonstrated that Exo-miR-205-Cy5 achieved more efficient and sustained delivery to the eye compared to free miR-205-Cy5 (Fig. [120]6I). Collectively, these findings demonstrate the potential of exosomes as a drug delivery system for miR-205. Exo-miR-205 has no toxicity in vivo and in vitro We evaluated the toxicity of Exo-miR-205 in vivo using electroretinography (ERG) and histological analyses. ERG results indicated that intravitreal injections of Exo-miR-205 did not produce significant adverse effects on retinal function (Fig. [121]7A). HE staining revealed no structural abnormalities in the retinas of Exo-miR-205-treated mice compared to the controls (Fig. [122]7B). TUNEL staining confirmed the absence of significant cell apoptosis in the retina following the administration of Exo-miR-205 (Fig. [123]7C). To assess potential systemic toxicity, comprehensive analyses, including hematologic assessments and blood biochemistry, were performed one week after intravitreal injections. No significant differences were observed in white blood cell (WBC) and red blood cell (RBC) counts and hemoglobin (Hb) among the different groups (Fig. [124]S5A-C). Liver function indicators, such as alanine aminotransferase (ALT) and aspartate aminotransferase (AST), also showed no significant variations among the groups (Fig. [125]S5D). Furthermore, histological examination of major organs, including the heart, liver, spleen, lung, and kidney, using HE staining revealed no signs of toxicity attributable to Exo-miR-205 (Fig. [126]S5E). Flow cytometry and Calcein-AM/PI double staining assays demonstrated that Exo-miR-205 did not induce significant cytotoxicity in HRVECs (Fig. [127]7D and E). Additionally, Calcein-AM/PI double staining was used to evaluate apoptosis in other retinal cell types, including retinal ganglion cells (RGCs), retinal pigment epithelium (RPE), and Müller cells (Fig. [128]S5F). The results indicated no significant apoptosis was detected among the different groups. Collectively, these findings suggest that Exo-miR-205 is safe for therapeutic applications, exhibiting no significant local or systemic toxicity. Fig. 7. [129]Fig. 7 [130]Open in a new tab Exo-miR-205 has no toxicity in vivo and in vitro. A-C C57BL/6J mice were administered PBS (Ctrl), exosomes, miR-205, or Exo-miR-205 to evaluate ocular toxicity. ERG was used to assess retinal electrophysiological function (A). H&E and TUNEL staining assays were performed to observe the histological changes and apoptotic cells in the retina and choroid. DNase I was used as a positive control in TUNEL staining assays. GCL: ganglion cell layer, IPL: inner plexiform layer, INL: inner nuclear layer, OPL: outer plexiform layer, ONL: outer nuclear layer. Scale bar, 50 μm (B, C, n = 5). D, E HRVECs were incubated with the exosome, miR-205, Exo-miR-205, or left untreated (Ctrl). Cell apoptosis was determined by the Calcein-AM/propidium iodide (PI) staining. Scale bar, 20 μm (D). Annexin V-FITC/PI double staining was performed to quantify the apoptosis of HRVECs following Exo-miR-205 administration by flow cytometry (E) Exo-miR-205 suppressed pathological neovascularization and vascular leakage We evaluated the efficacy of exosome-mediated delivery of miR-205 in reducing neovascularization and vascular leakage in both the OIR and CNV models. In the OIR model, administration of Exo-miR-205 significantly diminished pathological angiogenesis and avascular areas, demonstrating the most pronounced anti-angiogenic effects (Fig. [131]8A-C). Evans blue assays were performed to evaluate the effect of Exo-miR-205 on vascular permeability. Treatment with either exosomes or miR-205 resulted in improved vascular permeability outcomes, with the Exo-miR-205 group showing the most significant reduction (Fig. [132]8D and E). In the CNV model, GS-IB4 staining revealed that the lesion areas in the Exo-miR-205-treated group were considerably smaller compared to other groups (Fig. [133]8F). Additionally, we used OCT and FFA to observe the inhibitory effects of Exo-miR-205 on choroidal neovascularization and vascular leakage. As shown in Fig. [134]8G, Exo-miR-205 effectively reduced the area of choroidal neovascularization. FFA assays further confirmed that Exo-miR-205 alleviated vascular leakage associated with choroidal neovascularization (Fig. [135]8H). Collectively, these findings demonstrate that Exo-miR-205 significantly reduced both neovascularization and vascular leakage, highlighting its therapeutic potential in treating pathological neovascularization. Fig. 8. [136]Fig. 8 [137]Open in a new tab Exo-miR-205 suppresses pathological neovascularization and vascular leakage. In the OIR model, pups at P12 were administered an intravitreal injection of exosomes, miR-205, Exo-miR-205, or PBS, respectively. The retinas were harvested on P17 for GS-IB4 staining and Evans blue staining. In the CNV model, C57BL/6J mice received a single intravitreal injection of exosome, miR-205, Exo-miR-205, or PBS immediately after laser injury. After 7 days, choroidal tissues were collected to observe neovascularization formation. A-C The retinal avascular area and neovascularization was detected by GS-IB4 staining. Avascular area was highlighted by white staining. Pathologic angiogenic area was highlighted by yellow staining. Scale bar, 200 μm. D, E OIR mice were perfused with Evans blue dye for 1 h. The fluorescence signal of flat-mounted retina was detected by a fluorescence microscope. Scale bar, 200 μm. F GS-IB4 staining was used to detect CNV formation. The fluorescence signal of flat-mounted choroid was detected by a fluorescence microscope. Scale bar, 100 μm. G SD-OCT was performed to detect CNV size. Typical images and quantification of CNV size are shown. Scale bar, 100 μm. H FFA was performed to assess the leakage of CNV lesions. Scale bar, 100 μm. Data are expressed as the mean ± SD. *P < 0.05, n = 5, One-way ANOVA test Discussion Pathological neovascularization plays a critical role in the development of various ocular diseases, including PDR, AMD, ROP, and corneal neovascularization. These conditions can result in significant vision loss and potentially lead to blindness. Therapies targeting VEGF have emerged as the primary strategy for ocular vascular diseases [[138]33]. By inhibiting VEGF, these treatments aim to reduce the growth of abnormal blood vessels and associated complications in the eye. Despite the efficacy of anti-VEGF therapy, long-term use of these medications faces certain challenges. Additionally, some individuals may experience retinal atrophy that can further contribute to vision loss [[139]34]. This study demonstrates that miR-205, a promising therapeutic target, is markedly reduced in both CNV and OIR models. Previous study has also identified abnormal miR-205 expression in ocular neovascular diseases. The serum levels of miR-205 in patients with wet AMD is lower than that in non-AMD controls [[140]35]. Additionally, miR-205 expression is down-regulated under high-glucose and oxidative stress conditions [[141]36, [142]37]. Together, these findings indicate that miR-205 expression is down-regulated in response to pathological stimuli, consistent with our findings. miR-205 down-regulation is associated with abnormal neovascularization, whereas its overexpression significantly inhibits pathological angiogenesis. In cancer and other diseases, miR-205 has been identified as an anti-angiogenic miRNA [[143]38, [144]39]. For instance, miR-205 mitigates psoriasis-related proliferation and angiogenesis by interacting with the Wnt/β-catenin pathway. miR-205-5p inhibits angiogenesis in gastric carcinoma by reducing VEGFA and FGF1 expression. In this study, miR-205 can suppress endothelial cell proliferation, migration, tube formation, and sprouting. Furthermore, abnormal neovascularization often presents with high leakage and irregular structure, leading to edema, hemorrhages, and ultimately vision loss. Our findings demonstrate that miR-205 up-regulation reduces vascular leakage in CNV and OIR model mice, indicating that miR-205 is a promising therapeutic target for modulating neovascularization and vascular permeability. Neovascularization is a complex process finely regulated by a balance between pro-angiogenic and anti-angiogenic factors, including VEGF, ANGPT2, and FGFs. These factors interact within an intricate network of signaling pathways, working together to either promote or inhibit the formation and maturation of new blood vessels [[145]40]. Disruption of this balance can lead to excessive blood vessel production. VEGF is a central molecule in this process, exerting its effects by binding to and activating VEGF receptor-2. The biological activities of VEGF encompass enhancing vascular permeability, promoting endothelial cell mitosis, and inducing the expression of matrix metalloproteinases, which degrade the extracellular matrix and facilitate endothelial cell migration, differentiation, and survival [[146]41]. In addition to VEGF pathway, the Ang/Tie signaling pathway is pivotal for vascular formation. This pathway is regulated by the interaction between angiopoietins (Ang) and Tie2 receptor [[147]42, [148]43]. Preclinical studies have indicated that targeting Ang/Tie pathway in conjunction with VEGF inhibition can be an effective strategy for reestablishing vascular stability. This approach can enhance pericyte coverage around endothelial cells, thus reducing vascular leakage, pathological neovascularization [[149]44–[150]46]. Moreover, ANGPT2 inhibition offers additional therapeutic potential by modulating immune response within the retina [[151]47]. Indeed, the simultaneous inhibition of both VEGF and ANGPT2 can offer enhanced therapeutic benefits over targeting either molecule alone [[152]48]. In our study, we reveal the mechanism by which miR-205 influences ocular neovascularization. We identified VEGF and ANGPT2 as the target genes of miR-205 through computational prediction. Through dual inhibitory effects on ANGPT2 and VEGF, miR-205 could enhance the anti-angiogenic response and stabilize blood vessel permeability. This dual-targeting capability of miR-205 presents a promising therapeutic opportunity for treating ocular neovascularization. RNA-based therapies represent a frontier in medicine [[153]49]. miRNAs are small, non-coding RNA molecules that play a crucial role in gene expression regulation at the post-transcriptional levels. Due to regulatory influence on gene expression, miRNAs represent important biomarkers and therapeutic targets in several diseases [[154]50–[155]52]. Given the complexity of pathological angiogenesis, which involves numerous molecular players, miRNAs offer the unique advantage of targeting multiple processes with minimal adverse effects [[156]53, [157]54]. Our findings indicate that miR-205 holds promise as a nucleic acid drug for treating ocular neovascularization. However, due to the inherent instability of miRNAs in vivo, choosing an effective delivery system is essential. In recent years, exosomes have attracted increasing attention as a promising drug delivery system due to low immunogenicity, low toxicity, high biocompatibility and ability to penetrate biological barriers [[158]55]. In this study, we successfully extracted exosomes from hUMSCs and loaded miR-205 into the exosomes by electroporation. The results showed that these exosomes efficiently delivered miR-205 to HRVECs in vitro. Additionally, Exo-miR-205 accumulated in CNV model mouse lesions and demonstrated improved stability. However, a limitation of our study is the lack of in vivo data on Exo-miR-205 uptake by retinal endothelial cells. In future work, we plan to engineer exosomes to specifically target endothelial cells, aiming to enhance delivery efficiency. Furthermore, we confirm the excellent biosafety of Exo-miR-205 and demonstrate its inhibitory effects on neovascularization and vascular leakage. Beyond their role as ideal carriers for drug or gene delivery, exosomes also have anti-inflammatory properties. Previous study shows that MSC-derived exosomes can alleviate inflammatory cardiomyopathy and reduce cardiomyocyte apoptosis by promoting anti-inflammatory macrophage polarization [[159]56]. Thus, using exosomes to deliver miR-205 not only boosts delivery efficiency but also amplifies miR-205’s impact in vivo, strengthening its therapeutic effects on neovascularization and vascular leakage. In future studies, we will examine the anti-inflammatory effects of exosomes and further investigate the mechanisms behind their synergistic effects. This study underscores the potential advantages of Exo-miR-205 as a therapeutic intervention for ocular neovascularization. While the experimental outcomes are encouraging, the translation to clinical practice remains a challenge for the present. A key limitation is the need for more extensive research, particularly with animal models that can endure long-term treatments, to better simulate the chronic nature of human diseases and sustained therapeutic interventions required. Moreover, the current availability of mesenchymal stem cell-derived exosomes (MSC-exos) is significantly constrained, falling short of the demands for clinical applications. Thus, advancements in scalable production and purification methods for MSC-exo are highly required to ensure their feasibility and efficacy in clinical settings. While our study has demonstrated the efficacy and safety of Exo-miR-205 in preclinical models, careful attention to cell-type-specific binding and targeted in vivo delivery is crucial to prevent off-target gene regulation that could result in unwanted side effects. Rigorous clinical trials will be essential to evaluate the safety, optimal dosing, and therapeutic efficacy of Exo-miR-205 in human patients, as well as to identify any potential side effects and the patient populations most likely to benefit from this treatment. Conclusion This study demonstrates that miR-205 effectively suppresses pathological neovascularization and vascular permeability by targeting VEGFA and ANGPT2. To improve miR-205 stability for therapeutic use, we used MSC-derived exosomes (MSC-exos) as a delivery vehicle. By encapsulating miR-205 into MSC-exos through electroporation, we achieved a preparation with high stability and biocompatibility. This study allows us to validate the safety and therapeutic efficacy of Exo-miR-205, presenting it as a promising, innovative therapy for treating ocular neovascularization. Electronic supplementary material Below is the link to the electronic supplementary material. [160]Supplementary Material 1^ (2.8MB, docx) Acknowledgements