Molecular Vision 2026; 32:325-341
<http://www.molvis.org/molvis/v32/325>
Received 11 June 2025 |
Accepted 20 July 2026 |
Published 05 August 2026
Fangfang Ji, Ji Zhang, Ji Jin
Department of Ophthalmology, the Second Affiliated Hospital of Soochow University, Jiangsu Province, China
Correspondence to: Ji Jin, Department of Ophthalmology, the Second Affiliated Hospital of Soochow University, Suzhou 215004, Jiangsu Province, China; Phone: 0512-67783408; email: [email protected]
Purpose: Ischemic retinal diseases, such as retinopathy of prematurity, retinal vein occlusion, and diabetic retinopathy, are leading causes of vision loss. Current treatments focus on late-stage complications and have significant limitations. S100A9, a calcium-binding protein involved in inflammation and angiogenesis, remains understudied in ischemic retinal diseases. This study aimed to investigate S100A9 expression in oxygen-induced retinopathy (OIR) mice and assess the therapeutic effects of the S100A9 inhibitor tasquinimod.
Methods: OIR was induced in C57BL/6J mice, and tasquinimod (10 mg/kg/d) was administered orally from P12 to P17. Retinal nonperfusion areas were analyzed using immunofluorescence, and angiogenesis-related markers were examined by reverse transcription quantitative polymerase chain reaction (RT-qPCR) and western blotting. Human retinal microvascular endothelial cells (HRECs) were exposed to cobalt chloride–induced hypoxia, treated with tasquinimod, and assessed for S100A9 and angiogenesis-related marker expression, as well as migration and tube formation.
Results: Tasquinimod significantly reduced retinal nonperfusion areas in OIR mice. RT-qPCR and western blot analysis showed that tasquinimod suppressed S100A9, VEGF, ICAM-1, and TSP-1 expression. In vitro, hypoxia upregulated S100A9 and HIF-1α in HRECs, while tasquinimod dose-dependently inhibited hypoxia-induced migration, tube formation, and the expression of angiogenesis-related factors, including HIF-1α, VEGF, ICAM-1, and TSP-1.
Conclusions: S100A9 is upregulated in ischemic retinal conditions and promotes pathological angiogenesis. Tasquinimod effectively inhibits retinal neovascularization by targeting S100A9 and downstream angiogenic pathways, highlighting its potential as a therapeutic strategy for ischemic retinal diseases.
Ischemic retinal diseases, such as diabetic retinopathy (DR), retinal vein occlusion (RVO), and retinopathy of prematurity (ROP), represent major causes of irreversible visual impairment across different age groups [1,2]. DR, in particular, is the leading cause of blindness among the working-age population, affecting approximately one-third of individuals with diabetes worldwide [3]. By 2030, the global diabetic population is projected to exceed 550 million, with DR cases surpassing 220 million [4]. RVO is the second most prevalent retinal vascular disorder, with an estimated global prevalence of 0.77%, impacting over 28 million individuals aged 30 to 89 years [5]. In neonates, ROP remains a leading cause of childhood blindness, particularly in low- and middle-income countries, where neonatal survival rates are rising but access to timely screening and treatment remains limited [1]. According to a global burden of disease analysis, ROP accounted for approximately 101,600 cases of vision impairment in 2019, including moderate and severe visual loss as well as blindness. The incidence of new-onset ROP is notably higher in regions such as Africa and East Asia, and without early intervention, its burden is projected to grow [6].
The retina is one of the most metabolically active tissues in the human body. Under ischemic conditions, the imbalance between oxygen demand and supply leads to neovascular complications that underlie vision-threatening stages of these diseases [7]. Currently available treatments, including laser photocoagulation, intravitreal anti–vascular endothelial growth factor (VEGF) injections, and vitrectomy, primarily target late-stage complications. While these interventions mitigate disease progression, they often entail retinal damage, require repeated administrations, or are limited by transient efficacy and cost [8-10]. Notably, anti-VEGF therapy has shown significant clinical efficacy but is limited by transient effects, the need for repeated injections, financial burden, and potential retinal atrophy due to vascular and neuronal degeneration [11,12]. Therefore, identifying upstream regulators of ischemia-induced angiogenesis is critical for developing more effective therapeutic strategies.
The oxygen-induced retinopathy (OIR) mouse model is a well-established platform for studying hypoxia-driven retinal neovascularization [13]. While OIR does not fully mimic the metabolic disturbances seen in DR, it shares crucial characteristics with various ischemic retinal diseases, including early vascular loss, aberrant neovascularization, breakdown of the blood–retinal barrier, plasma leakage, and neuroinflammatory responses [14]. Hence, it provides a practical in vivo system for investigating angiogenic mechanisms and evaluating potential therapeutic interventions.
Emerging evidence implicates the damage-associated molecular pattern (DAMP) protein S100A9 as a mediator of inflammation and vascular pathology [15]. S100A9 is abundantly expressed in myeloid cells and acts via pattern recognition receptors such as Toll-like receptor 4 (TLR4) and the receptor for advanced glycation end products (RAGE) [15-18], triggering downstream MAPK, PI3K, and NF-κB pathways to upregulate proinflammatory and angiogenic mediators including IL-6, CCL2, ICAM-1, and VCAM-1 [19-25]. SERPINE1 has also been implicated in S100A9-induced suppression in MDS [26]. These inflammatory responses contribute to vascular permeability [27], endothelial proliferation, migration, and pathological neovascularization [28]. Moreover, S100A9 has been associated with multiple inflammatory diseases, including arthritis [29], inflammatory bowel disease [30], and diabetes [31].
Members of the S100 protein family have been implicated in ocular neovascularization, particularly under ischemic and inflammatory conditions [32-34]. For example, S100A4 has been reported to promote neovascularization in the OIR model [32,33]. However, the role of S100A9 in ischemic retinal angiogenesis remains uncharacterized. This is particularly relevant given recent proteomic analyses identifying S100A9 upregulation in OIR mice and in the vitreous of patients with proliferative diabetic retinopathy [21,35].
Tasquinimod is a small-molecule oral inhibitor with high affinity for S100A9, preventing its interaction with TLR4 and RAGE [36]. It has demonstrated antiangiogenic activity in prostate cancer models and is currently under clinical evaluation for multiple myeloma (NCT04405167) [37,38]. Mechanistically, tasquinimod modulates the expression of thrombospondin 1 (TSP-1) and suppresses hypoxia-inducible factor 1α (HIF-1α) signaling [39-41], implicating these pathways in its antiangiogenic effects. However, its potential to attenuate ischemia-induced retinal neovascularization has not been investigated.
This study aims to evaluate the expression changes of S100A9 in the OIR mouse model and to investigate the therapeutic potential of tasquinimod in attenuating retinal neovascularization. In addition, we examined the effects of tasquinimod on human retinal capillary endothelial cells (HRECs) under hypoxic conditions, which serve as an in vitro model to mimic the ischemic microenvironment of the retina. Using this approach, we assessed the inhibitory effects of tasquinimod on hypoxia-induced endothelial migration and vascular network formation. By characterizing the expression dynamics of S100A9 and the antiangiogenic responses to tasquinimod, this study provides mechanistic insights and supports the therapeutic potential of S100A9 inhibition in ischemia-associated retinal vascular pathology.
All animal experiments were conducted in compliance with the Guidelines for the Care and Use of Laboratory Animals of the Chinese Medical Academy and adhered to the Association for Research in Vision and Ophthalmology (ARVO) Statement for the Use of Animals in Ophthalmic and Vision Research. The study was approved by the Animal Care Committee of Soochow University, Suzhou, China, and all experimental procedures followed the regulations set forth by the Laboratory Animal Management Committee of Soochow University.
C57BL/6J mice were obtained from Shanghai SLAC Laboratory Animal Co. Ltd. (Shanghai, China) and housed under specific pathogen-free conditions with a 12-h light–dark cycle, receiving a standard laboratory diet and water ad libitum. OIR was induced as previously described [42]. Briefly, postnatal (P) day 7 mouse pups were exposed to 75% oxygen (hyperoxia) in a sealed chamber until P12, then returned to room air (21% oxygen; normoxia) until P17. Mice were randomly assigned to five experimental groups: (1) normal group (n = 7): mice maintained under normoxia throughout the experiment; (2) OIR12 group (n = 14): mice exposed to 75% oxygen from P7 to P12; (3) OIR17 group (n = 14): mice exposed to 75% oxygen from P7 to P12, followed by normoxia until P17; (4) OIR12 + Tas 5d group (n = 14): mice exposed to 75% oxygen from P7 to P12, followed by oral administration of tasquinimod (10 mg/kg/d) from P12 to P17 under normoxia; and (5) OIR12 + PBS 5d group (n = 14): mice exposed to 75% oxygen from P7 to P12, followed by oral administration of phosphate-buffered saline (PBS) (10 mg/kg/d) from P12 to P17 under normoxia. Although formal power analysis was not conducted, group sizes were selected based on prior studies using the OIR model that detect treatment-related differences in retinal nonperfusion areas [13].
For in vitro use, tasquinimod (Absin Bioscience, Shanghai, China) was dissolved in 100% DMSO to prepare a concentrated stock of 10 mM. Working solutions for cell-based assays were freshly prepared by diluting the stock into culture medium to 1 to 5 µM; the final DMSO content was ≤0.05% v/v at 5 µM. Vehicle controls were DMSO-matched to the highest DMSO percentage used in the corresponding experiment.
For in vivo use, tasquinimod powder was dissolved in DMSO to a concentrated stock of 100 mg/ml, then diluted with PBS to a 1-mg/ml working solution (final DMSO ≤1% v/v). Neonatal mice received 10 mg/kg once daily by oral administration (gavage) at 10 ml/kg from P12 to P17. To minimize systemic toxicity in neonatal mice, PBS was used as the vehicle control instead of DMSO alone; although pharmacologic matching was not achieved, ethical considerations informed this decision.
To visualize the retinal vasculature, retinal dissection and vessel staining were performed as described by Tual-Chalot et al. [43]. Postnatal mice were sacrificed by cervical dislocation, and the eyeballs were carefully enucleated. The eyes were fixed in 4% paraformaldehyde at room temperature for 10 to 15 min. After fixation, the cornea, sclera, uveal tissue, and lens were removed, and the isolated retinas were washed in PBS. For reverse transcription quantitative polymerase chain reaction (RT-qPCR) analysis, retinas were immediately transferred to RNA Later and stored at −80 °C.
For whole-mount immunostaining, the retinas were flattened by making four radial cuts and further fixed in −20 °C methanol to enhance permeabilization. Retinas were incubated overnight at 4 °C with isolectin-B4, an endothelial cell-specific marker, followed by 4′,6-diamidino-2-phenylindole nuclear staining. Immunofluorescence images were acquired using a Nikon (Tokyo, Japan) OFN25 fluorescence microscope (4× magnification).
The nonperfusion area of the retina was measured according to the method described by Smith et al. [44]. Images of the entire retina were stitched together using Photoshop CS6 (Adobe, San Jose, CA), and the lasso tool was used to outline the avascular regions and the total retinal area. The ratio of avascular area to total retinal area (in pixels) was calculated to determine the relative size of the nonperfusion zone. To minimize subjective bias, the analysis was performed by an independent investigator blind to the experimental groups. For each group, measurements were obtained from three retinas, with each retina analyzed three times, and the average value was used. Representative images are included in this study.
Retinal tissues were rapidly frozen in liquid nitrogen and ground into a fine powder using a mortar and pestle. The powdered tissue was transferred to a 5-ml centrifuge tube, and four volumes of lysis buffer (8 M urea, 1% protease inhibitor cocktail; Calbiochem, Merck, Darmstad, Germany) were added. The mixture was sonicated three times on ice using a high-intensity ultrasonic processor (Ningbo Scientz Biotechnology Co., Ltd., Zhejiang, China) to ensure complete lysis. The homogenate was centrifuged at 12,000g at 4 °C for 10 min, and the supernatant containing retinal protein extract was collected. The protein concentration of each sample was determined using the bicinchoninic acid (BCA) protein assay kit (Beyotime, Jiangsu, China).
HRECs were obtained from the BeNa Culture Collection (BNCC340909, Beijing, China) and cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum. Cells were maintained at 37 °C in a humidified incubator with 5% CO2. Cell morphology and growth were monitored daily under an inverted microscope, and the culture medium was refreshed every 2 to 3 days. When cells reached approximately 80% confluence, cells were subcultured immediately (within ≤2 h of assessment) at a 1:2 ratio; cultures were not maintained beyond confluence.
For hypoxia modeling, 200 μM cobalt chloride (CoCl2) was applied for 24 h to chemically stabilize HIF-1α, thereby mimicking the ischemic microenvironment [45-47]. Cells were divided into the following groups for analysis: (1) control (normoxia), (2) hypoxia (CoCl2 200 μM, 24 h), (3) hypoxia + 1 μM tasquinimod (cotreatment with CoCl2 for 24 h), and (4) hypoxia + 5 μM tasquinimod (cotreatment with CoCl2 for 24 h).
Cell migration was assessed using the Transwell migration assay as described previously [39]. HRECs in the logarithmic phase were suspended in serum-free DMEM and adjusted to a concentration of 1 × 105 cells/ml. A 24-well Transwell system was used, with 600 μl of medium containing 200 μM CoCl2 and different tasquinimod concentrations added to the lower chamber. A total of 150 μl of the cell suspension was seeded into the upper chamber and incubated at 37 °C with 5% CO2 for 24 h. After incubation, the Transwell inserts were carefully removed, and the nonmigrated cells were wiped off. Migrated cells on the lower membrane surface were fixed with methanol for 30 min at room temperature and stained with crystal violet solution (15–30 min at room temperature). The membranes were washed with PBS, air-dried, and visualized under a 200× microscope, and the number of migrated cells was counted.
A Matrigel-based tube formation assay was performed to evaluate the angiogenic potential of HRECs as described previously [40]. Matrigel, μ-Slide angiogenesis slides, and 200-μl pipette tips were precooled at 4 °C for 20 min. A 10-μl aliquot of Matrigel was pipetted into each well, ensuring the absence of air bubbles, and incubated at 4 °C for 30 min, followed by polymerization at 37 °C for 30 to 60 min. HRECs were trypsinized (0.25% trypsin, 1–2 min), neutralized with complete medium, centrifuged at 200 ×g for 5 min, and resuspended at a density of 4 × 104 cells/well. Medium containing 200 μM CoCl2 with or without tasquinimod was added, and the cells were incubated at 37 °C. Tube formation was captured after 8 h, and representative images were recorded for analysis.
Total RNA was extracted from mice retinal tissues or HREC samples using TRIzol reagent (Thermo Fisher Scientific, Waltham, MA) following the manufacturer’s instructions. RNA was reverse-transcribed into complementary DNA (cDNA) using the PrimeScript RT Reagent Kit (Takara, Tokyo, Japan). RT-qPCR was performed with the SYBR Green PCR Kit (Takara) on a LightCycler 96 System (Roche Diagnostics, Mannheim, Germany). β-Actin was used as an internal control for normalization. Primer sequences for TSP-1, HIF-1α, VEGF, ICAM-1, and S100A9 were synthesized by GenScript (Nanjing, China), with the sequences described in Table 1 (human gene) and Table 2 (mouse gene).
HRECs were lysed in protein lysis buffer (50 mM Tris-HCl, pH 7.5, 2% sodium dodecyl sulfate, and 1 mM phenylmethylsulfonyl fluoride [PMSF]), with PMSF freshly added to 1 mM before use. Protein concentration was determined using the BCA Protein Assay Kit (Beyotime, Shanghai, China). Proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis, transferred to nitrocellulose membranes (MilliporeSigma, Burlington, MA), and blocked with 5% skimmed milk in PBS-T (PBS + Tween-20) at room temperature for 1 h. The membranes were incubated overnight at 4 °C with primary antibodies against HIF-1α, S100A9, TSP-1, VEGF, ICAM-1, and β-actin (catalog no. ab179483, EPR3555, ab267388, A4782, A12303, ab53013, and AC038; Abcam, Cambridge, UK). Horseradish peroxidase–conjugated secondary antibodies (A0208; Beyotime) were applied for 1 h at room temperature. Proteins were visualized using enhanced chemiluminescence (ECL; Millipore), and band intensities were quantified using ImageJ software (National Institutes of Health, Baltimore, MD). β-Actin served as the loading control.
To assess the effects of CoCl2-induced hypoxia and tasquinimod treatment on HREC viability, a cell counting kit-8 (CCK-8) assay was performed. HRECs in the logarithmic growth phase were collected and seeded into 96-well plates at a density of 5,000 cells per well with six replicates per condition, in a total volume of 200 μl per well. The cells were cultured under 37 °C, 5% CO2, and saturated humidity conditions for 12 h or 24 h. At the time points (12 h and 24 h), 20 μl CCK-8 reagent was added to each well and incubated at 37 °C for 2 h. The absorbance was measured at 450 nm using a microplate reader after a color change from red to yellow, and statistical analysis was performed.
All experiments were performed in triplicate, and data are presented as mean ± standard deviation (SD). Statistical analyses were conducted using GraphPad Prism 5 (GraphPad Software, La Jolla, CA). For comparisons between two groups, an unpaired two-tailed Student t test was used. For comparisons among three or more groups, one-way analysis of variance (ANOVA) followed by post hoc test was performed. Additionally, for experiments involving two independent variables (e.g., different concentrations of CoCl2 across multiple time points), a two-way ANOVA was applied to assess the effects of treatment, time, and their interaction. A p value of <0.05 was considered statistically significant (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001).
To assess the effect of tasquinimod on ischemia-induced retinal neovascularization, a well-established OIR model was used [21]. Neonatal mice were exposed to 75% oxygen for 5 days (P7–P12) to induce vascular obliteration, followed by normoxic recovery (room air, 21% oxygen) from P12 to P17 to stimulate ischemia-induced retinal neovascularization (Figure 1A). Whole-mount isolectin B4 staining was performed to visualize retinal vasculature (Figure 1B), and the avascular area (nonperfusion region) was quantified as a percentage of total retinal area. To provide temporal context for disease progression, we included an OIR12 group representing the time point immediately after hyperoxic exposure and before neovascular responses begin. Although not directly used for treatment comparison, this group illustrates the transition between vessel obliteration and subsequent revascularization. Retinal flat-mount imaging revealed that OIR17 and OIR12 + PBS 5d contained prominent preretinal neovascular tufts with irregular, glomeruloid vessel clusters, whereas tasquinimod-treated retinas displayed a more uniform intraretinal capillary network with fewer preretinal tufts (Figure 1B). Consistent with these features, mice treated with tasquinimod from P12 to P17 exhibited significantly reduced avascular areas (9.62% ± 1.13%) compared to the OIR17 group (20.51% ± 0.93%) and the vehicle control group receiving PBS (20.27% ± 0.36%; p < 0.001; Figure 1C).
To further corroborate the revascularization and structural recovery of the retina, hematoxylin and eosin staining of retinal cross sections was performed at P17 (Figure 1D). Compared with the normal group, OIR17 sections exhibited disorganization of the inner retina (ganglion cell layer [GCL], inner nuclear layer [INL], inner plexiform layer (IPL), and inner nuclear layer [INL]) with focal preretinal cellular tufts and blurred laminar boundaries, consistent with hypoxia-driven neovascular changes [48]. The OIR12 + PBS 5d group showed persistent lamination irregularities with blurred INL–IPL boundaries and focal unevenness of the inner retina, compared with the normal group. In contrast, tasquinimod-treated retinas displayed more continuous lamination and fewer ectopic preretinal cellular clusters.
Taken together with the reduction in avascular area on flat-mounts, and given that retinal injury occurs during the P7 to P12 hyperoxic phase, these imaging and histologic findings indicate that tasquinimod primarily promotes intraretinal revascularization. This process, along with structural recovery and repair, occurs from P12 to P17, rather than preservation before injury.
To investigate the molecular mechanism by which tasquinimod suppresses ischemia-induced retinal neovascularization in OIR mice, we evaluated the messenger RNA (mRNA) expression of angiogenesis-associated genes, including S100A9, VEGF, ICAM-1, and TSP-1 (Figure 2). Compared with the normal group, the OIR17 group (P17) exhibited significant upregulation of all four genes: S100A9 (3.00 ± 0.13), VEGF (2.49 ± 0.10), ICAM-1 (2.91 ± 0.38), and TSP-1 (2.60 ± 0.31). Oral administration of tasquinimod significantly reduced the expression of each factor compared with both the untreated OIR17 group and the OIR12 + PBS 5d control group (p < 0.001 for all comparisons), suggesting that the observed downregulation is a specific effect of tasquinimod rather than vehicle exposure. In contrast, the OIR12 + PBS 5d group did not show any significant differences from the untreated OIR17 group, supporting the specificity of tasquinimod’s action. Western blot analysis corroborated the qPCR findings at the protein level (Figure 3). The OIR17 group exhibited marked increases in S100A9, VEGF-A, ICAM-1, and TSP-1 expression compared to the normal group. Tasquinimod treatment significantly attenuated the expression of these proteins, reducing S100A9 by 38.4%, VEGF by 29.8%, ICAM-1 by 43.3%, and TSP-1 by 33.6% compared to the OIR + PBS 5d group (p < 0.001 for all). No significant difference was observed between OIR17 and OIR12 + PBS 5d groups, confirming that the therapeutic effect was not attributable to vehicle administration.
These findings demonstrate that tasquinimod effectively downregulates angiogenesis-related factors in the ischemic retina. This occurs potentially through inhibition of S100A9 signaling, thereby reducing retinal inflammation and pathological neovascularization.
To investigate the impact of hypoxia on S100A9 expression in HRECs, we used CoCl2 to simulate hypoxic conditions. CoCl2 is a well-established hypoxia-mimetic agent that stabilizes HIF-1α, leading to the activation of hypoxia-responsive pathways [45-47]. Before examining S100A9 regulation, we first assessed the effects of CoCl2 exposure on cell viability using a CCK-8 assay. The results demonstrated that CoCl2 reduced HREC viability in a dose- and time-dependent manner. While short-term exposure (0–12 h) had minimal impact, prolonged hypoxia (24 h) significantly reduced cell viability, particularly at 200-μM and 300-μM concentrations (Figure 4A). This decline in cell survival suggests that sustained hypoxia impairs endothelial function, which may reflect the pathological state of ischemic retinal diseases.
Having established that CoCl2-induced hypoxia alters HREC viability, we next examined its effect on S100A9 expression. HRECs were cultured under normoxic conditions or treated with 200 μM CoCl2 for 24 h to induce hypoxia. RT-qPCR analysis revealed that, compared to the normoxic control, CoCl2 treatment resulted in a significant increase in HIF-1α mRNA levels (approximately threefold) and S100A9 mRNA levels (approximately twofold; Figure 4B,C). Western blot analysis further confirmed that the protein levels of HIF-1α and S100A9 were markedly elevated under hypoxic conditions, with HIF-1α increasing by approximately 1.6-fold and S100A9 by approximately 1-fold compared to controls (Figure 4D–F). These findings suggest that hypoxia not only induces S100A9 expression in HRECs but may also contribute to endothelial dysfunction, as reflected by decreased cell viability under prolonged hypoxic conditions.
To further investigate the impact of hypoxia and tasquinimod treatment on angiogenesis-related gene expression in HRECs, we first assessed the effects of tasquinimod treatment on cell viability using a CCK-8 assay. HRECs were cultured under normoxic conditions or exposed to 200 μM CoCl2 for 24 h, with or without tasquinimod cotreatment. At 0 h, there were no significant differences in cell viability across all treatment groups. However, after 24 h of hypoxia, 5 μM tasquinimod significantly improved cell viability compared to the hypoxia-only group, suggesting that tasquinimod may mitigate hypoxia-induced cytotoxicity (Figure 5A). Importantly, tasquinimod alone (5 μM) did not affect cell viability, as there was no significant difference between the untreated control (HREC) and HREC + 5 μM tasquinimod group. This confirms that tasquinimod does not exert cytotoxic effects at the tested concentrations, further supporting its safety profile in retinal endothelial cells. However, while tasquinimod partially rescued cell viability under hypoxia, it did not completely restore viability to normoxic levels, indicating that additional mechanisms may contribute to hypoxia-induced cell damage beyond S100A9 inhibition.
We then performed RT-qPCR to evaluate whether tasquinimod modulates the expression of angiogenesis-related genes under hypoxic conditions. HRECs were divided into four groups: normal (normoxia), hypoxia (200 μM CoCl2), hypoxia + 1 μM tasquinimod, and hypoxia + 5 μM tasquinimod. RT-qPCR analysis revealed that hypoxia significantly upregulated the mRNA expression levels of HIF-1α, VEGF, ICAM-1, and TSP-1, with HIF-1α increasing approximately 3.91-fold, VEGF by 1.50-fold, ICAM-1 by 2.86-fold, and TSP-1 by 1.98-fold compared to the normal control. Treatment with 1 μM tasquinimod markedly reduced the expression of these genes, while 5 μM tasquinimod further suppressed hypoxia-induced upregulation, indicating a dose-dependent inhibitory effect (Figure 5B–E).
To further validate the impact of tasquinimod on hypoxia-induced angiogenic factor expression, we performed western blot analysis to assess the protein levels of HIF-1α, VEGF, ICAM-1, and TSP-1 in HRECs under the same conditions. Consistent with the qPCR findings, hypoxia significantly upregulated the protein expression of all these factors compared to the normoxic group. Treatment with 1 μM tasquinimod resulted in a notable reduction in the expression of these proteins, while 5 μM tasquinimod led to an even more pronounced suppression, suggesting a dose-dependent inhibitory effect (Figure 6).
Endothelial cell migration is a crucial process in angiogenesis, facilitating new vessel sprouting and tissue remodeling [49]. To investigate the effect of tasquinimod on hypoxia-induced endothelial cell migration, we performed a Transwell migration assay. As shown in Figure 7, hypoxia significantly enhanced the migration of HRECs compared to the normal condition. However, treatment with 1 μM tasquinimod notably suppressed cell migration, and this inhibitory effect was further enhanced by 5 μM tasquinimod. These findings suggest that tasquinimod effectively inhibits endothelial migration under hypoxic conditions.
To further evaluate the antiangiogenic effect of tasquinimod, we conducted a Matrigel tube formation assay. Hypoxic conditions markedly enhanced tube formation, as evidenced by a significant increase in the number of junctions (Figure 8B) and total branching length (Figure 8C) compared to the normoxia group. Treatment with 1 μM tasquinimod significantly attenuated these effects, and 5 μM tasquinimod led to further reductions in both junction number and branching length. These results demonstrate that tasquinimod suppresses key functional processes involved in angiogenesis—including endothelial migration and capillary-like network formation—under hypoxic conditions, supporting its potential role as an antiangiogenic agent.
Retinal neovascularization, a hallmark of ischemic retinopathies such as ROP, DR, and RVO, involves both aberrant angiogenesis and chronic inflammation driven by hypoxic stress [50-52]. In this study, we demonstrated that the inflammatory mediator S100A9 is upregulated in the ischemic retina of OIR mice and in hypoxia-stimulated retinal endothelial cells. Pharmacological inhibition of S100A9 using tasquinimod [36] led to a significant reduction in nonperfusion areas and downregulation of key angiogenesis- and inflammation-related factors, including VEGF, ICAM-1, and TSP-1. These findings support the concept that S100A9 contributes to hypoxia-induced vascular dysfunction and that its inhibition offers a novel anti-inflammatory and antiangiogenic strategy for ischemic retinal diseases. Moreover, histologic assessment of P17 retinal sections was consistent with these findings. Relative to the OIR17 and OIR12 + PBS 5d group, tasquinimod-treated retinas exhibited more continuous lamination (clearer GCL, IPL, and INL boundaries) with fewer preretinal nuclei on the vitreous side. Although not quantified, these qualitative differences indicate improved inner-retina organization in the treated group.
In the OIR paradigm, return to room air after hyperoxia leads to two separable vascular outcomes: pathological preretinal neovascularization (tufts) and physiological intraretinal revascularization of avascular zones [13,44]. In our data, tasquinimod reduced the nonperfused area from P12 to P17, and the flat-mounts showed fewer preretinal tufts with a more uniform intraretinal capillary network (Figure 1B,C). Consistently, cross sections revealed fewer preretinal nuclei and more continuous inner-retina lamination after tasquinimod (Figure 1D). Thus, the smaller avascular areas at P17 most likely reflect improved physiological revascularization, while the concomitant suppression of VEGF and ICAM-1 (Figure 2, Figure 3) indicates attenuation of inflammation-driven pathological sprouting [53-56]. This pattern is compatible with a vascular-normalization framework, in which limiting inflammatory, disorganized tuft formation permits or even facilitates orderly intraretinal revascularization and perfusion recovery [57,58]. Finally, because tasquinimod targets the S100A9 axis, which is implicated in retinal inflammation and endothelial activation [27,40,59], its anti-inflammatory action provides a plausible mechanistic basis for reducing pathological tufts while allowing physiological capillary regrowth.
Importantly, the timing of tasquinimod administration differed between our two experimental arms. In the OIR model, the drug was started after the hyperoxia-induced injury window and given throughout P12 to P17, thereby probing postinsult recovery. By contrast, in HRECs, tasquinimod was applied concurrently with hypoxic stimulation, modeling an early or coincident intervention. This difference helps explain why cell assays show more immediate molecular changes, whereas the in vivo readouts emphasize revascularization and structural recovery. To bridge these designs, future OIR studies will vary the initiation window (e.g., P7, P12, P14) and compare effects on avascular area, tuft area, and intraretinal revascularization, together with molecular endpoints (VEGF, ICAM-1, and S100A9).
This study addresses a gap in the current literature by linking S100A9 inhibition to the suppression of pathological angiogenesis in a hypoxic retinal context. Previous clinical and experimental studies have established S100A9 as a DAMP protein involved in leukocyte activation and vascular inflammation [60,61]. For example, elevated S100A9 levels have been observed in the vitreous fluid of patients with proliferative diabetic retinopathy [35], supporting its role in ocular inflammation. In addition, members of the S100 protein family have been implicated in various ocular diseases involving neovascularization or vascular dysfunction, such as age-related macular degeneration and central retinal vein occlusion [62,63], although not all are strictly ischemia-driven. However, no prior study has specifically evaluated the therapeutic effects of an S100A9 inhibitor in ischemia-driven retinal neovascularization. Our findings extend previous observations by demonstrating that tasquinimod not only reduces S100A9 expression under hypoxia but also suppresses angiogenic and inflammatory pathways.
Mechanistically, we observed that tasquinimod downregulated VEGF and ICAM-1 expression in both OIR retinas and hypoxia-treated HRECs. VEGF is a well-characterized target in retinal vascular diseases. Although anti-VEGF agents such as ranibizumab and aflibercept have demonstrated clinical efficacy, their therapeutic effects are transient, necessitating repeated intravitreal injections [64,65]. Additionally, these treatments may be associated with ocular inflammatory responses, such as intraocular inflammation and retinal vasculitis [66,67]. Moreover, their ability to reverse capillary nonperfusion remains uncertain and is subject to ongoing investigation [68-71]. In contrast, S100A9 inhibition may affect upstream inflammatory signaling, thereby offering a complementary therapeutic mechanism. Nevertheless, due to the absence of direct comparison with anti-VEGF agents in our study, the relative efficacy and potential additive benefit of tasquinimod require further investigation.
Tasquinimod also reduced TSP-1 expression under hypoxia, a finding that may appear paradoxical given TSP-1’s known antiangiogenic role. However, prior studies have shown that TSP-1 expression increases during the neovascular phase in the OIR model, potentially as a reactive response to VEGF elevation [72]. The observed TSP-1 downregulation may therefore reflect an overall amelioration of hypoxia rather than a direct anti-TSP effect. These results highlight the need to further delineate the signaling hierarchy between S100A9, VEGF, and TSP-1. While tasquinimod is a well-characterized inhibitor of the S100A9 axis, context-specific causal evidence in our HREC system is still required. To address this, future work will incorporate exogenous S100A8/A9 dose–response assays in HRECs to directly test whether tasquinimod antagonizes S100-driven NF-κB activation and ICAM-1/VEGF induction, thereby establishing a causal link between the S100A9 pathway and the observed protective effects.
Despite these findings, our study has several limitations. First, we used only a single in vivo dosage (10 mg/kg tasquinimod) and a single time point (P17) to assess therapeutic efficacy in the OIR model. A more comprehensive dose–response analysis and longitudinal evaluation would better characterize the pharmacodynamics and potential toxicity. Second, we did not include comparator groups treated with anti-VEGF agents such as ranibizumab or aflibercept, precluding conclusions regarding tasquinimod’s relative efficacy or combinatorial value. Third, while tasquinimod is known to inhibit S100A9, our data do not establish a causal relationship between S100A9 suppression and the observed vascular improvements. Functional validation using S100A9 knockdown, overexpression, or neutralizing antibodies would provide stronger mechanistic evidence. Finally, we assessed a limited number of angiogenesis and inflammation markers. Unbiased transcriptomic profiling (e.g., RNA sequencing) or proteomic approaches may uncover additional tasquinimod-regulated targets and broaden the mechanistic landscape.
This study demonstrates that tasquinimod attenuates ischemia-induced retinal neovascularization and inflammation, potentially via suppression of S100A9 signaling. By downregulating VEGF, ICAM-1, and TSP-1, tasquinimod modulates key pathways involved in endothelial activation and angiogenesis. These results suggest that targeting S100A9 may represent a novel therapeutic strategy for ischemic retinopathies. Future investigations should validate the specificity and durability of this effect in chronic disease models, compare tasquinimod with current standard-of-care therapies, and explore its clinical applicability in ROP, DR, and RVO.
Author Contribution Conceptualization: Ji Jin Data curation: Fangfang Ji Formal Analysis: Fangfang Ji Funding acquisition: Ji Jin Investigation: Ji Jin Methodology: Ji Jin Project administration: Fangfang Ji Resources: Ji Jin Software: Fangfang Ji Supervision: Ji Jin Validation: Ji Zhang Visualization: Ji Zhang Writing – original draft: Fangfang Ji Writing – review & editing: Ji Jin Conflict of interest statement None declared This study was supported by the Suzhou Science and Technology Fund of China (SYS2020137).
This is an open-access article distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits reusers to copy and distribute the material in any medium or format in unadapted form only, for noncommercial purposes only, and only so long as attribution is given to the creator. CC BY-NC-ND includes the following elements: BY: credit must be given to the creator. NC: Only noncommercial uses of the work are permitted. ND: No derivatives or adaptations of the work are permitted