1. Introduction
Pathological neovascularization is a common feature of various retinal diseases, including retinopathy of prematurity (ROP), proliferative diabetic retinopathy (PDR), retinal vein occlusion (RVO), wet age-related macular degeneration (wet AMD), etc. [
1]. It causes retinal edema, hemorrhage, inflammation, and fibrotic scarring, leading to severe visual impairment and even blindness [
2,
3]. Despite variations in the underlying causes and risk factors, hypoxia-induced production of angiogenic mediators such as vascular endothelial growth factor (VEGF) is widely recognized to play a critical role in triggering the formation of new blood vessels [
4]. In the clinic, several anti-VEGF drugs and retinal laser photocoagulation techniques have been applied to reduce retinal hypoxia, consequently leading to the inhibition of retinal angiogenesis. However, there remains limitations such as resistance to anti-VEGF drugs and irreversible peripheral visual function damage caused by laser photocoagulation [
5,
6]. Therefore, alternative approaches that target the hypoxia-induced angiogenesis are actively being explored.
Hypoxia-inducible factor-1α (HIF-1α), a crucial transcription factor that forms a heterodimer with HIF-1β and binds to hypoxia-responsive elements (HREs) within target gene promoters, plays a pivotal role in cellular responses to hypoxia. HIF-1α has been recognized as inducing the expression of angiogenic factors such as VEGF, promoting the formation of new blood vessels. In addition, HIF-1α activation in response to hypoxia insult drives metabolic adaptations, to ensure immune cell activation and effective responses, orchestrating the inflammatory environment for retinal angiogenesis. Targeting HIF-1α signaling seems to be a promising approach for regulating hypoxia-induced angiogenesis.
Oxygen-induced retinopathy (OIR) is the most widely used model for retinal neovascular diseases, such as ROP, PDR, and RVO [
4,
7,
8]. In the application of this hypoxia-induced angiogenesis model, it is common to perform intravitreal injection of various agents, to elucidate the molecular mechanisms involved in retinal neovascularization or to test the efficacy of new potential anti-angiogenic agents. When performing intravitreal injection, a Hamilton microliter syringe is usually used to inject 0.5–2 μL reagent in an OIR model. In some cases, the needle of the microliter syringe is very soft and may become dull after multiple injections, making it difficult to penetrate the eyeball effectively. Therefore, some researches chose sharp and disposable needles to pre-create a puncture channel, followed by using the microliter syringe to inject the corresponding volume of reagents [
9,
10]. In this study, we serendipitously found that a 0.5 mm puncture (using a 25 G needle) without delivering any reagents could almost completely inhibit the formation of pathological angiogenesis in the OIR model, while a 0.3 mm puncture (using a 30 G needle) or 0.21 mm direct injection procedure (using 33 G needle) did not influence the development of neovascularization.
Compared with the 0.3 mm or 0.21 mm puncture group, the 0.5 mm puncture apparently caused the leakage of vitreous fluid, concomitant with a collapsed eyeball, indicating a sudden drop in intraocular pressure (IOP) induced by the 0.5 mm puncture. As we know, ocular perfusion pressure, a vital factor influencing ocular blood flow, is obviously affected by changes in IOP [
11]. Elevated IOP can hinder the blood flow and diminish the perfusion pressure within the retinal vasculature, leading to insufficient oxygen delivery and subsequent retinal hypoxia [
12,
13]. Remarkably, 0.5 mm puncture induced a substantial decrease in IOP, subsequently elevating ocular perfusion pressure and retinal oxygen partial pressure (pO
2). These alterations hold the potential to reduce HIF-1α expression and mitigate the pathological angiogenesis. In addition, the strikingly contrary outcomes observed with varying needle sizes suggested that large size needles (≥0.5 mm) for intravitreal injection should be avoided in the OIR model, due to the potential role of IOP reduction in inhibiting HIF-1α-mediated retinal neovascularization.
2. Materials and Methods
2.1. Mice for OIR Model and Needle Puncture
C57BL/6J mice were obtained from the Animal Center of Guangzhou University of Chinese Medicine and were housed in a specific pathogen-free facility in the Animal Laboratories of Zhongshan Ophthalmic Center. All animal experiments were approved by Institutional Animal Care and Use Committee of Zhongshan Ophthalmic Center, Sun Yat-sen University (Ethic number: Z2022042, approved on 14 June 2022). The OIR model was established as described previously [
7]. Newborn mice and their nursing mothers were exposed to 75% O
2 in an oxygen chamber from postnatal day 7 (P7) to P12. These mice were randomized into groups and received intravitreal puncture using needles of different calibers immediately upon returning to room air environment on P12. A disposable sharp needle (25 G with a 0.5 mm needle outer diameter, BD/30 G with a 0.3 mm needle outer diameter, BD) was inserted vertically into the eye wall 1 mm posterior to the limbus. The resulting scleral tunnel could be easily accessed for the 33 G microliter syringe injection (a 0.21 mm needle outer diameter, Hamilton). Alternatively, the 33 G microliter syringe needle could be directly inserted vertically into the eye wall. Pups weighing more than 6 g were euthanized using carbon dioxide asphyxiation on P17 for further experimental analysis.
2.2. Measurement of pO2
The measurement of particle oxygen pressure was performed as previously described [
14]. Pups were anesthetized through intraperitoneal injection of 50 mg/kg pentobarbital sodium. An incision was made in the lateral and temporal eyelid to expose the eye ball, and then mice were positioned in a robotic stereotaxic apparatus. To measure oxygen levels, a custom-made oxygen sensitive microelectrode for mice was inserted into the vitreous through the limbus. The electrode’s position inside the eye was visualized using a contact lens and an operating microscope. The oxygen pressure was recorded until the values stabilized.
2.3. Measurement and Management of IOP
Pups were anesthetized using an intraperitoneal injection of 50 mg/kg pentobarbital sodium. IOP was measured daily at 10 AM using a rebound tonometer (iCare TONOLAB, Vantaa, Finland) from P12 to P17. At P12, as the mice received intravitreal puncture using needles immediately upon returning to the room air at 6 AM, two additional IOP measurements were taken, both before and after the puncture. Six consecutive probe-to-cornea contact measurements were averaged as one record. For lowering IOP, Travatan (0.004% travoprost, Novartis) and Azarga (10 mg/mL Brinzolamide + 5 mg/mL Timolol, Novartis) eye drops were applied to one eye at a dose of 5 µL from P12 to P17 once a day, while 5 µL of PBS eye drops were applied to the other eye as a control in the same pup mouse. IOP was measured before and 2 h after drug administration at P12. Subsequently, IOP was measured daily at 10 AM from P13 to P17.
2.4. H&E and Immunohistochemistry Staining
Eyes were fixed with 4% formalin overnight, embedded in paraffin, and cut into 5 μm vertical slices. Sections were then washed and treated with hematoxylin buffer for 10 min at room temperature. After rinsing in deionized water and dipping in 1% eosin solution for 15 s, sections were rehydrated in an alcohol gradient, rewashed, and mounted. The neovascular cell nuclei above the internal limiting membrane (ILM) nuclei were counted, and the mean of the 10 counted sections yielded the average neovascular cell nuclei per eye. At least six eyes were used for analysis. For immunohistochemistry staining, sections were subjected to antigen retrieval, then blocked with normal serum, and incubated overnight with diluted primary antibodies at 4 °C. The following primary antibodies were used: polyclonal rabbit anti-TGFb (Abcam, Cambridge, UK) and monoclonal rabbit anti-collagen III (Abcam). Normal IgG was used as a negative control. The secondary antibody Goat anti-Rabbit IgG (H+L) Cross-Adsorbed Secondary Antibody, HRP (Invitrogen, Carlsbad, CA, USA) was incubated for 2 h. Retinal histological changes were observed under a microscope (Leica DM4000, Wetzlar, Germany).
2.5. Immunofluorescence Staining of Retinal Wholemounts
Eyes were enucleated and fixed in 4% paraformaldehyde (PFA) fixative for 1 h and the retinae were removed carefully. The retinal whole-mounts were stained with Isolectin B4 (IB4, 1:50, Invitrogen) for 2 h at room temperature for quantification of neovascularization. For other immunofluorescence stainings, the retinae were incubated with 5% Bovine Serum Albumin (BSA) at 4 °C overnight for blocking and incubated with primary and secondary antibodies. The primary antibodies include anti-CD31 antibody (Merck Millipore, Darmstadt, Germany), anti-NG2 antibody (Servicebio, Wuhan, Hubei, China), anti-α-SMA antibody (Invitrogen), and anti-IBA1 antibody (Wako, Tokyo, Japan). The secondary antibodies included donkey anti-rabbit IgG H&L (Alexa Fluor 555) secondary antibody, goat anti-Armenian hamster IgG H&L (Alexa Fluor 488), and goat anti-mouse IgG H&L (Alexa Fluor 647) (Abcam). The retinae were washed extensively, flat-mounted, and observed using a confocal microscope (Carl Zeiss, Jena, Germany). Areas of retinal neovascularization and the avascular region were analyzed using Image-Pro Plus 6.0 (Media Cybernetic, Rockville, MD, USA).
2.6. Visualization of Retinal Vascular Permeability
Retinal vascular permeability was visualized through the leakage of Evans blue (EB). EB powder (Sigma, Darmstadt, Germany) was dissolved in PBS to make 2% EB solution and then intravenously injected at a dosage of 150 μg/kg. After 2 h, the mice were sacrificed, and their eyes were perfused with 4% PFA. Retinae were dissected, flat-mounted, and examined through fluorescence microscopy (Carl Zeiss, Jena, Germany).
2.7. Hypoxyprobe Staining
The hypoxic state in the retina was evaluated using a Hypoxyprobe RedAPC Kit, as previously described [
15]. OIR mice at P12 or P17 were injected with Hypoxyprobe intraperitoneally at 2.5 mg per pup 1 h before they were sacrificed. Retinas were isolated after fixation of eyeballs in 4% PFA for 1 h, washed 3 times with PBST, blocked with 5% BSA for 1 h, and incubated with RED APC dye -MAb1 diluted with PBST at 4 °C overnight. Retinae were washed extensively and cut into four pieces before being flat-mounted on slides. All retinae were photographed using a confocal microscope (Carl Zeiss, Jena, Germany).
2.8. Real-Time Quantitative PCR Analysis
Total RNA was extracted from the retinae on ice using TRIzol (Invitrogen), and cDNA synthesis was performed using a Reverse Transcriptase Superscript II Kit (Invitrogen) according to the manufacturer’s instructions. Briefly, qPCR was performed in a total volume of 20 µL mixture containing 2 µL of cDNA, 10 µL of 2 × SYBR Premix Ex Taq, and 10 µmol/L of the primer pairs. GAPDH was used as a reference gene. Samples were incubated at 95 °C for 30 s and up to 40 cycles of 95 °C for 5 s and 60 °C for 34 s according to the protocol. The real-time PCR reaction was performed using a StepOnePlus™ real-time PCR instrument. The experiments were repeated independently three or more times.
2.9. Statistics
Representative images were displayed in figures, and the results were repeated at least 3 times. All data were statistically analyzed and presented as the mean ± standard error of the mean (SEM). Comparisons of three or four groups were analyzed using one-way ANOVA, and comparisons between two groups were performed using a two-tailed Student t-test. A value of p < 0.05 was considered statistically significant (* p < 0.05; ** p < 0.01; *** p < 0.001).
4. Discussion
In this study, we surprisingly found that large needle puncture (0.5 mm) for intravitreal injection caused an unexpected “anti-angiogenic” effect in OIR mice, indicating that an acute and significant reduction in IOP had the potential to counteract the hypoxia-induced angiogenesis. The 0.5 mm puncture induced a reduction in IOP and an improvement in pO2 and enhancement in oxygen delivery to the retina. Consequently, this ameliorated the hypoxic stimulus, leading to a decrease in the expression of Hif1a and angiogenic factors, thereby inhibiting pathological angiogenesis in the retina. This is the first study to report that IOP reduction achieved through intravitreal puncture can elicit such a significant anti-angiogenesis effect.
As we know, elevated IOP in glaucoma is associated with impaired blood flow to the retina and optic nerves, resulting in ischemic and hypoxic damage to the neuroretinal tissue [
22,
23]. Lowering IOP has proven beneficial in alleviating the mechanical compression and vascular compromise that occur in glaucoma. Clinically, it has been observed that patients with neovascular glaucoma experience significant regression of new blood vessels after the glaucoma surgery for lowering IOP. In addition, studies have indicated that lowering IOP can increase choroidal blood perfusion and alleviate scleral hypoxia, suggesting its potential role in managing high myopia progression. In patients with wet AMD receiving anti-VEGF therapy, simultaneous IOP reduction has been reported to contribute to improved visual outcomes after 2 years [
24]. However, direct evidence demonstrating the notion that lowering IOP can prevent neovascularization is limited and the underlying mechanisms are not fully understood. In this study, use of a 25 G needle (0.5 mm) for intravitreal injection induced a sharp decline of IOP and probably enhanced the retinal perfusion and improved the blood supply to the retina, leading to alleviation of retinal hypoxia. Hypoxia, a recognized potent trigger, promotes the activation of Hif1a transcription factor, which subsequently instigates the expression of pro-angiogenic genes, including
Vegfa and other angiogenic factors. Such processes lead to the development of pathological neovascularization [
25]. Here, the IOP reduction achieved through intravitreal puncture downregulated the expression of
Hif1a and
Vegfa. This regulatory effect ultimately suppressed hypoxia-induced angiogenesis in the retina.
We have further proposed the therapeutic concept that IOP-lowering drugs may serve as a potential therapeutic strategy for managing retinal neovascular diseases. We detected the anti-angiogenesis effects of Azarga and Travatan, two commonly prescribed IOP-lowering medications in clinical practice. Despite Travatan lowering IOP by promoting aqueous humor outflow, and Azarga reducing IOP by decreasing aqueous humor production, both drugs achieve a similar level of IOP reduction and alleviation of neovascularization in OIR mice, indicating these non-invasive interventions aimed at lowering IOP could have potential for the treatment of retinal neovascular diseases. However, the direct vascular impact of these drugs could not be ruled out. Previous studies have shown that prostaglandin analogue, the main component of Travatan, could induce vessel vasodilation and increase capillary permeability [
26]. Blockers of carbonic anhydrase, similarly to Azarga, may relax pericytes and improve the retinal blood supply by increasing the pH difference between the intracellular and extracellular space [
27]. Therefore, the responses to IOP-lowering drugs are variable and individual, and further research and clinical studies are needed to fully understand the extent of their efficacy and applicability in managing retinal neovascular diseases.
The OIR mouse, the most commonly used and classical retinal neovascularization model, is applied to investigate the pathogenesis of retinal neovascularization and evaluate the efficacy of various agents in inhibiting neovascularization. In many cases, the agents need to be injected into the vitreous cavity and standardization of the intravitreal injection procedure directly affects the research results. Therefore, there is an urgent need to standardize the intravitreal injection procedure, to optimize the use of the OIR model. Our study found that using a 0.5 mm 25 G needle for intravitreal injection caused unexpected “anti-angiogenic” effects, while a 0.3 mm 30 G needle had no corresponding effect on angiogenesis and nor did a typical 0.21 mm Hamilton 33 G needle. This suggested that attention should be paid to the needle size during the intravitreal injection procedure in mice, to avoid confusing results caused by IOP decrease. It is advisable to avoid using needle diameters larger than 0.3 mm, as this may cause fluctuations in IOP, potentially affecting the outcome of neovascularization in the OIR model. In addition, it is worth noting that the anatomy of the mouse and human eyes is very different. It is indeed possible that a 0.5 mm intravitreal puncture on human eyes may not lead to a significant IOP reduction as observed in mice.
In summary, our findings revealed that a large puncture (≥0.5 mm) remarkably suppressed the formation of pathological angiogenesis during an intravitreal injection procedure in the OIR model. This observation identified a crucial risk factor of IOP fluctuation that has been ignored in the past and affecting the neovascularization outcome in the OIR model. Moreover, our results provide a novel therapeutic perspective regarding anti-angiogenesis therapy based on the management of IOP. IOP-lowering drugs, most commonly used for reducing IOP in the clinic, are potentially a new therapeutic approach for preventing the development of retinal neovascular diseases.