Skip to main content
Download PDF

Peptides based on the interface of hnRNPA2B1-transthyretin complex repress retinal angiogenesis in diabetic retinopathy

Journal of Translational Medicine volume 23, Article number: 458 (2025) Cite this article

Abstract

Background

Heterogeneous nuclear ribonucleoprotein A2B1 (hnRNPA2B1) plays a vital role in angiogenesis, when its nucleic acid-binding domain is occupied by transthyretin (TTR), the neovascularization of human retinal microvascular endothelial cells (hRECs) is repressed under hyperglycemic conditions.

Methods

HnRNPA2B1-targeting peptides (THIPs) were designed based on the core fragments at the TTR-hnRNPA2B1 interface. Biacore, Langmuir equilibrium adsorption, and co-immunoprecipitation (co-IP) assays were performed to determine the association between the THIPs and hnRNPA2B1. Proliferation and DNA synthesis in hRECs were detected using CCK-8 and EdU assays. Transwell, wound healing, and tube formation assays were used to evaluate migratory and the angiogenic capacity of hRECs. Related RNA and protein expression levels were tested by quantitative PCR and western blot assays, respectively. Streptozotocin (STZ)-induced diabetic retinopathy (DR) model rats were intravitreally injected with 5 μL of AAV9 virus (1 × 1012 vg/mL) every 8 weeks, with sterile saline used as control. After 16 weeks, the retinas were extracted and subjected to Evans blue leakage and retinal trypsin digestion assays. Retinal paraffin sections were prepared and stained with hematoxylin and eosin (H&E) or subjected to immunohistochemical or immunofluorescence assays.

Results

Biacore, Langmuir equilibrium adsorption, and co-IP analyses demonstrated that the four designed THIPs specifically recognized hnRNPA2B1. CCK-8 and EdU labeling assays showed that the THIPs inhibited proliferation and DNA synthesis in hRECs under hyperglycemia. Transwell, wound healing and tube formation assays demonstrated that the THIPs inhibited the migratory and angiogenic capacity of hRECs. Quantitative PCR and western blot assays suggested that the THIPs exerted their effects via the STAT4/miR-223-3p/FBXW7 and the downstream Notch1/Akt/mTOR axes. In vivo studies using DR model rat revealed that the intravitreal administration of THIP-4 significantly mitigated retinal leakage, capillary decellularization, pericyte loss, fibrosis, and gliosis during DR progression.

Conclusion

Our findings demonstrated that under hyperglycemia, THIP-4 suppressed DR progression via the STAT4/miR-223-3p/FBXW7 and Notch1/Akt/mTOR axes both in vitro and in vivo. These results indicated that THIP-4 has strong potential for clinical application in DR and other angiogenesis associated diseases.

Introduction

With the rapid increase in the global prevalence of diabetes [1,2,3], diabetic retinopathy (DR) has emerged as a major health concern, affecting individuals of all ages [4,5,6]. DR is a severe microvascular complication resulting from a sustained hyperglycemic and hypoxic microenvironment in the retina. Typical pathological features of DR include retinal neovascularization, vascular leakage, retinal fibrosis, and gliosis, which can result in vision loss, visual impairment, and even blindness [7,8,9,10]. Understanding the development and progression of DR, as well as the associated signaling network and molecular mechanisms, is crucial for establishing novel diagnostic and therapeutic strategies for this condition.

Preventing retinal neovascularization is a primary goal in clinical DR therapy. Anti-VEGF compounds or antibodies, such as pegaptanib, ranibizumab, bevacizumab, and aflibercept are commonly applied via intraocular injection to suppress retinal vascular leakage and angiogenesis [11,12,13]; however, resistance to these antibodies and potential side effects in normal vascular remain problematic. Other molecules, including long noncoding RNAs [14, 15], microRNAs [16,17,18], vitamins [19, 20], and peptides [21, 22], are also being explored for their diagnostic and therapeutic value in DR.

Heterogeneous nuclear ribonucleoproteins (hnRNPs) comprise a group of nuclear proteins involved in transcriptional and posttranscriptional regulation [23]. HnRNPA2B1, a core member of this protein family, has been identified as a “reader” or “switcher” of N6-methyladenosine (m6A) modification on RNA [24,25,26]. Studies have shown that hnRNPA2B1 plays a vital role in various biological processes, e.g., hnRNPA2B1 mediates the level or stability of long noncoding RNAs and microRNAs via reading their m6A modification [25], promotes the progression of multiple myeloma via m6A‑dependent stabilization of TLR4 or ILF3 mRNAs [27, 28], facilitates the development of colorectal cancer via m6A-dependent stabilization of TCF7L2 mRNA [29], initiates and amplifies the innate immune response the immune response to DNA viruses via recognizing pathogenic DNA [30], and accelerates angiogenesis via eliminating the tumor suppressor miR-204-3p or promoting Stop codon decoding and synthesis of VEGF-Ax [31, 32]. However, the role of hnRNPA2B1 in the progression of DR has not been systematically investigated, nor have the underlying mechanisms. We have previously shown that transthyretin (TTR), a ~ 55-kDa homotetrameric transporter protein, inhibits the neovascularization of hRECs in a high-glucose environment by directly binding to the nucleic acid binding domain of hnRNPA2B1 in the nucleus, which could affect the m6A reading function of hnRNPA2B1 and its downstream STAT4/miR-223-3p/FBXW7 signaling axis; in addition, TTR could not affect the level of hnRNPA2B1 [33, 34]. The TTR-hnRNPA2B1 interface has also been characterized via structural simulation, and has been further optimized using energy minimization and site-directed substitution [35].

In this work, focusing on the TTR-hnRNPA2B1 interface, TTR-hnRNPA2B1 interface peptides (THIPs) were designed based on the interaction complex, further optimized, and then synthesized. In vitro, we found that the THIPs recognized hnRNPA2B1 and suppressed proliferation, DNA synthesis, migration, wound healing, and tube formation in human retinal endothelial cells (hRECs) under hyperglycemic conditions, effects that were mediated via the STAT4/hsa-miR-223-3p/FBXW7 pathway. In vivo, meanwhile, THIP 4 significantly suppressed the development of DR in a streptozotocin (STZ)-induced DR rat model. This research represents the first report of peptides designed to target hnRNPA2B1 for the treatment of DR-related neovascularization. Our findings offer promising avenues for the development of novel clinical therapies for this condition.

Methods

Cell culture

HRECs were obtained from the Cell Bank of the Chinese Academy of Science, Shanghai, China, and were cultured in phenol red-free Dulbecco’s modified Eagle’s medium (DMEM; Life Technologies, Carlsbad, CA, USA). HRECs (passages 5–6) were plated in six-well plates (Corning, Acton, MA, USA) at a density of 3000 cells/mL and then cultured in low-glucose (5.5 mM glucose; LG), high-glucose (25 mM glucose; HG), or low-glucose + mannitol (20 mM; LM, osmotic control) medium for 120 h. For biofunction assessment, 1 μM TTR homotetramer (4 μM monomer) or 1 μM THIPs was added to the HG medium.

Antibodies and reagents

The details of the antibodies used in this work are provided in Table S1. Protease Inhibitor mixture (50 × ; P1046-1) and Phosphatase Inhibitor mixture (50 × ; P1046-2) were obtained from Beyotime (Shanghai, China). CNBr-activated Sepharose 4B was purchased from GE Healthcare (Uppsala, Sweden). The Pierce BCA Protein Assay Kit and the In-Gel Tryptic Digestion Kit were acquired from Thermo Scientific (Hudson, NH, USA). Other chemicals were of analytical reagent grade and were purchased locally.

Preparation of TTR and hnRNPA2B1

Human TTR (NCBI Reference Sequence: NP_000362.1, amino acids 21–147) cDNA was synthesized by GenePharma (Suzhou, China), inserted into a pRham N-HIS plasmid (Lucigen, Middleton, WI, USA), expressed in Escherichia coli BL21 (DE3) cells, and purified using a Ni-Sepharose column; a ToxinEraser Endotoxin Removal Kit (GenScript, Nanjing, China) was employed for endotoxin depletion [33]. Additionally, human hnRNPA2B1 (UniProtKB/Swiss-Prot: P22626.2, amino acids 12–195) cDNA was synthesized by GenePharma and processed as described above.

Peptide map fingerprinting (PMF) analysis was used to identify the prepared proteins. Following a previously reported protocol [36, 37], after separation by 12% SDS–PAGE, the single bands of TTR (~ 15 kDa) and hnRNPA2B1 (~ 25 kDa) were extracted, purified, trypsinized, and further subjected to matrix-assisted laser desorption ionization time-of-flight (MALDI–TOF) analysis. The peaks were analyzed by Mascot Search using the nr and SwissProt 57.15 databases.

Biacore assay

For chip preparation, 400 mM EDC and 100 mM NHS (GE Healthcare) were mixed immediately before injection as the activator, and the mixture was injected for 800 s over flow cell 1 (Fc1) and Fc2 of the sample channel at a flow rate of 10 μL/min. Purified hnPNA2B1 (50 μg/mL) in 10 mM NaAc (pH 4.5) was then injected for 1650 s over Fc2 at a flow rate of 10 μL/min, with an immobilization density of approximately 6954 RU. The chip was deactivated by the injection of 1 M ethanolamine hydrochloride (GE Healthcare) at a flow rate of 10 μL/min for 800 s into Fc1 and Fc2.

TTR was diluted with running buffer (1 × EP Buffer, GE Healthcare) to 10,000, 5000, 2500, 1250, 625, 312.5, 156.3, and 0 nM, while the THIPs were diluted to 35,000, 17,500, 8750, 4375, 2187.5, 1093.75, 546.88, 273.44, and 0 nM. These samples were then injected onto Fc1 and Fc2 of the sample channel at a flow rate of 30 μL/min for an association phase of 90 s, followed by 180 s of dissociation. The association and dissociation processes were performed in the running buffer. A Surface Plasmon Resonance (SPR) assay was used to investigate the affinity between hnRNPA2B1 and TTR and between hnRNPA2B1 and each of the THIPs.

Langmuir equilibrium adsorption

To immobilize hnRNPA2B1 on sepharose beads, 5 g of CNBr-activated Sepharose 4B wet gel was mixed with 75 mg of purified hnRNPA2B1 (dissolved in 0.1 M NaHCO3/0.5 M NaCl, pH 8.3), and the mixture was gently rotated at 4 ℃ for 12 h. The gel was subsequently blocked in 0.1 M Tris–HCl (pH 8.0) for 2 h and then washed sequentially (three cycles) with 0.1 M acetate/0.5 M NaCl (pH 4.0), 0.1 M Tris–HCl/0.5 M NaCl (pH 8.0), and 20 mM PBS (pH 7.2). The hnRNPA2B1 immobilized on the sepharose beads was detected using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific Inc.).

As previously reported [38], TTR and the THIPs (5, 10, 15, 20, and 25 μM) were dialyzed against 20 mM PBS (pH 7.2), mixed with 0.1 g of hnRNPA2B1 immobilized on sepharose beads, and gently rotated at 4 °C for 4 h to equilibrium. After centrifugation at 1000 × g for 5 min, the supernatants were assayed for concentrations.

Localization of TTR and the THIPs in hRECs

HRECs (passage 6) were incubated in HG medium at 37 °C for 24 h in 24-well plates, after which 1 μM TTR homotetramer or 1 μM THIPs was added to the cells. After a 2-h incubation, the cells were washed three times with PBS, fixed in 4% paraformaldehyde for 20 min, permeabilized in 0.1% Triton X-100, and exposed to anti-hnRNPA2B1, anti-TTR, or anti-His-tag antibodies overnight at 4 °C. After exposure to secondary antibodies and counterstaining with Hoechst 33342 (4082, Cell Signaling Technology), the cells were visualized using confocal microscopy (Zeiss, Oberkochen, Germany).

Co-immunoprecipitation (co-IP) assay

As described in our previous work [33, 35], hRECs (passage 6) were cultured in HG medium at 37 °C for 24 h in 6-well plates and then lysed in RIPA buffer. A 4-μL volume of hnRNPA2B1 antibody was then mixed with 800 μL of cell extract, followed by incubation at 4 °C for 12 h. Then, 40 μL of sepharose beads coupled with Protein A and THIPs (400 μL, 1 μM) was added to the mixture for a further 4 h of incubation at 4 °C.

THIPs (400 μL, 1 μM) were mixed with the above cell extracts together with 4 μL of anti-His-tag antibody, followed by incubation at 4 °C for 12 h. Then, 40 μL of sepharose beads coupled with Protein A was added to the mixture for a further 4 h of incubation at 4 °C.

After washing twice with PBS and three times with washing buffer (1% NP-40, 0.1% SDS, 0.5% sodium deoxycholate, 1 mM Na3VO4, 5 mM NaF, 10 mM sodium pyrophosphate, 10 mg/mL aprotinin, 5 mg/mL leupeptin, 1 mM PMSF), the sepharose beads were boiled in 50 μL of 1 × SDS loading buffer for 10 min, and then subjected to western blot or dot blot assays.

To identify the proteins captured by TTR and THIPs in IP, the beads were incubated in SDT buffer (100 mM Tris, 4% SDS and 1 mM DTT, pH 8.0) for 10 min at 100 ℃, after a 5 min centrifugation (8000 × g) at 4 ℃, the supernatant was collected and the DTT concentration was adjusted to 100 mM. 30 μL of the collected sample was incubated at 95 ℃ for 8 min, and then subjected to ultra-filtration (10 kDa MWCO) with UA buffer (8 M Urea, 150 mM Tris–HCl pH8.5) and IAA buffer (100 mM iodoacetamide in UA buffer) subsequently. The proteins were then digested in Trypsin buffer (4 μg Trypsin in 40μL 25 mM NH4HCO3) at 37 ℃ for 18 h. In the nanoElute system (Bruker), the peptides were subjected to an EASY column™ C18 column (Solvent A, 0.1% formic acid and 99.9% water; Solvent B, 0.1% formic acid and 99.9% acetonitrile), using a gradient time of 30 min at a flow rate of 300 nL/min as follows: 5% to 35% B in 18 min; 35% to 80% B in 2 min; and 80% B in 10 min. LC–MS/MS analysis was then performed on a timsTOF Pro mass spectrometry (Bruker) for 30 min. MS.d folder data was analyzed with maxquant1.6.14. Proteins identified with unique peptides ≥ 2 were believable.

Quantitative real-time PCR (qPCR)

Total RNA was extracted from hRECs as previously described [34, 35] and the mRNA levels of STAT4 and FBXW7 (GAPDH as control) together with hsa-miR-223-3p (cel-miR-39-3p as spike-in control) were detected by qPCR. The primers are listed in Table S2.

Cell viability assays

Cell Counting Kit-8 (CCK-8) (Sigma-Aldrich, St. Louis, MO, USA) assays were used to monitor the proliferation of hRECs after 5 days of culture as previously described and per the manufacturer’s instructions [33, 39]. DNA synthesis in hRECs was tested using an EdU labeling kit (RiboBio, Guangzhou, China) and confocal microscopy (Zeiss). The migratory properties of hRECs were investigated using a Transwell assay (Corning Transwell chambers; Corning, NY, USA) and were assessed under EVOS FL Auto Imaging System (Life Technologies, Carlsbad, CA, USA). HREC migration over 24 h was monitored using a wound healing assay kit (ab242285, Abcam, Cambridge, MA, USA). In addition, the tube formation (angiogenic) potential of hRECs was tested on a basement membrane matrix (BD Biosciences, San Jose, CA, USA). The apoptosis of hRECs was investigated by Cytoflex S Flow Cytometry (Beckman, Indianapolis, IN, USA) using a FITC Annexin V Apoptosis Detection Kit I (BD Biosciences, San Jose, CA, USA).

Western blot and dot blot assays

As previously described [33, 39], for western blot, total protein was extracted from cells or tissues using RIPA lysis buffer supplemented with a protease inhibitor mixture and a phosphatase inhibitor, separated by 10% SDS–PAGE, and transferred to PVDF membranes. For dot blotting, approximately 2 μL of sample was dotted on a membrane and dried at room temperature. In both cases, the membrane was blocked with 5% skimmed milk, and subsequently incubated first with primary and then secondary antibodies. After exposure to an enhanced chemiluminescent (ECL) reagent, the target proteins were visualized with the Odyssey Imaging System (LI-COR Biosciences, Lincoln, NE, USA).

STZ-induced DR rat model

All animal experiments were conducted according to the institutional guidelines for the humane treatment of animals, the Principles of Laboratory Animal Care (National Institutes of Health, Bethesda, MD, USA) (https://www.ncbi.nlm.nih.gov/books/NBK54050), and the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research (https://www.arvo.org/About/policies/statementfor-the-use-of-animals-in-ophthalmic-and-visionresearch/). All animal-related experiments were approved by the Ethics Committee of The Affiliated Wuxi People’s Hospital of Nanjing Medical University (2022-033, 18-Oct-2022).

Male Sprague–Dawley (SD) rats (Changzhou Cavens Laboratory Animals Co., Ltd, Changzhou, China), aged 6 to 8 weeks, were used in this work. Five days after the intraperitoneal injection of STZ (60 mg/kg), blood glucose levels > 16.7 mmol/L indicated that the rat model of diabetes had been established.

Rat TTR (GenBank: AAH86946.1; PDB ID: 1IE4) and THIP4 were cloned into the pAAV-MCS vector and separately packaged in adeno-associated viruses (AAVs; GenePharma). After the intraperitoneal injection of ketamine (80 μg/g) and xylazine (4 μg/g), the animals were intravitreally injected with 5 μL of AAV9 (1 × 1012 vg/mL) every 8 weeks, with sterile saline used as control. The following groups were set up: Control, STZ, STZ + NC, STZ + TTR, and STZ + THIP 4 (n = 8 rats per group).

As per our previously published protocols [39, 40], after 16 weeks, the retinas were extracted and subjected to Evans blue leakage and retinal trypsin digestion assays. Retinal paraffin sections (5 μm thick) were prepared and stained with hematoxylin and eosin (H&E) or subjected to immunohistochemical or immunofluorescence assays.

For the determination of the pathological indexes of the liver and kidney, the concentrations of blood urea nitrogen (BUN) and serum creatinine (SCr) and, the serum activities of alanine aminotransferase (ALT), aspartate aminotransferase (AST), and alkaline phosphatase (ALP) activities were measured using an automatic biochemical instrument analyzer (BS-480, Shenzhen Mindray Bio-Medical Electronics Co., Ltd, Shenzhen, China).

Statistical analysis

All data were analyzed with SPSS 13.0 software (SPSS Inc., Chicago, IL, USA) and are presented as means ± standard deviation. Statistical significance was assessed using one-way ANOVA followed by Fischer’s least significant difference (LSD) post hoc test. P-values < 0.05 were considered significant.

Results

Design of the peptides

We previously demonstrated that TTR specifically recognizes and targets the mRNA binding domain of hRNPA2B1 (Fig. 1A), thereby inhibiting hREC neovascularization in a high-glucose environment [33]. Additionally, employing an energy minimization strategy, we showed that two amino acid substitutions (I68R and D99R) at the TTR-hnRNPA2B1 interface (Fig. 1A) helped stabilize the complex and enhanced the suppression of angiogenesis [35]. In this study, THIPs were designed based on the three core fragments on the interface of the TTR homotetramer; as the interface was not a continuous or linear sequence, and the core fragments were located in two separate monomers, softer linkers (–GGG–) were designed between these fragments, and the I68R and D99R substitutions were also generated; for localization analysis, a His-tag was inserted in the C-terminus (Fig. 1B). The designed peptides, along with the scramble control, were synthesized by GL Biochem (Shanghai) Ltd (Shanghai, China).

Fig. 1
figure 1

The design of TTR-hnRNPA2B1 interface peptides (THIPs). A The TTR homotetramer (marked in green)/hnRNPA2B1 (marked in purple) complex; the interface is marked in yellow. Two amino acid substitutions—I68R and D99R—are marked in red. B The sequences of the THIPs and the scramble peptide. C Following the lysis of human retinal microvascular epithelial cells (hRECs), anti-hnRNPA2B1 and anti-His-tag antibodies were used to pull down the hnRNPA2B1-THIPs complexes, and these were then subjected to western blot or dot blot analysis. D After 2 h of incubation with 1 μM THIPs, TTR, or scrambled peptide, the localization of these molecules was detected using anti-hnRNPA2B1, anti-human TTR, anti-His-tag, Alexa Fluor 488-labeled anti-IgG, and Alexa Fluor 594-labeled anti-IgG antibodies; nuclei were counterstained with Hoechst 33342. Scale bar: 25 μm

Full size image

The THIPs specifically recognized hnRNPA2B1

Human TTR and hnRNPA2B1 were expressed and purified, yielding single bands at approximately 15 and 25 kDa, respectively, on SDS–polyacrylamide gels (Fig. S1); in addition, the single bands were extracted and their identities were confirmed via PMF analysis (Fig. S2).

In the Biacore assay, the association rate constant (Ka), dissociation rate constant (Kd), and dissociation constant (KD) between hnRNPA2B1 and TTR or between hnRNPA2B1 and the THIPs could be calculated using SPR; in contrast, no interaction was detected between the scrambled peptide and hnRNPA2B1 (Fig. S3) (Table 1). Furthermore, the low KD detected indicated that the THIPs had a high affinity for hnRNPA2B1.

Table 1 The interaction constants between hnRNPA2B1 and THIPs
Full size table

For the Langmuir equilibrium adsorption analysis, ~ 5 mg of recombinant hnRNPA2B1 (approximately 0.2 μmol) was immobilized on 1 g of wet sepharose gel. The KD values between the THIPs and hnRNPA2B1 were found to be lower than that between TTR and hnRNPA2B1, suggesting that the THIPs had a higher affinity for hnRNPA2B1; however, for THIP 1–3, the theoretical maximum adsorption values (Qmax) were also lower (Fig. S4) (Table 1). In addition, the data for the scrambled peptide did not fit a typical equilibrium adsorption model. The interaction between each of the THIPs and hnRNPA2B1 was further confirmed by co-IP assays (Fig. 1C).

Via LC–MS/MS assay, the proteins captured by THIPs and TTR were identified, and those with > fivefold enhancement of LFQ intensity, and > 200,000 ΔLFQ intensity were compared (Fig. S5 and Table S3). HnRNPA2B1 was overlapped in all the groups with the most significant Δ LFQ intensity and enhanced fold. These results suggested that the same as TTR, THIPs could specifically target hnRNPA2B1.

The THIPs entered the nucleus and interacted with hnRNPA2B1

Next, we investigated whether the THIPs could enter hRECs and sought to determine their putative cellular localization. For this, we first incubated hRECs with the THIPs for 2 h (with TTR and the scrambled peptide serving as controls), and then sequentially exposed the cells to anti-hnRNPA2B1, anti-human TTR, anti-His-tag, Alexa Fluor 488-labeled anti-IgG, and Alexa Fluor 594-labeled anti-IgG antibodies; nuclei were counterstained with Hoechst 33342. As shown in Fig. 1D, like TTR, both the THIPs and the scrambled peptide could enter the hRECs. Additionally, they all mainly accumulated in the nucleus and their signals overlapped with that of hnRNPA2B1.

The THIPs inhibited proliferation and DNA synthesis in hRECs in a high-glucose environment

An EdU assay was used to detect DNA synthesis in hRECs (Fig. 2A, B). The results showed that DNA synthesis was significantly higher in hRECs cultured in a high-glucose environment than in cells cultured in a low-glucose one; however, both TTR and the THIPs significantly reversed this trend. In addition, among the assessed peptides, THIP-4 showed the strongest repressive effects.

Fig. 2
figure 2

The THIPs inhibited DNA synthesis and proliferation of hRECs. (A, B) EdU assays (1, vs. HG; 2, vs. HG + NC; 3, vs. HG + THIP-4; n = 5, *P < 0.05, **P < 0.01, ***P < 0.001); nuclei were counterstained with Hoechst 33342, scale bar: 100 μm. (C) CCK-8 assays after 5 days of culture (1, vs. HG; 2, vs. HG + NC; 3, vs. HG + THIP-4; n = 5, *P < 0.05, **P < 0.01, ***P < 0.001). HG high glucose, NC negative control

Full size image

Meanwhile, the results of the CCK-8 assay (Fig. 2C) showed that, after 5 days of culture, the proliferative ability of hRECs was markedly increased under high-glucose conditions compared with that in a low-glucose environment; like in the EdU assay, these effects were significantly abrogated by the addition of either TTR or the THIPs, with THIP-4 again displaying the most significant repressive effect. In addition, in a low-glucose environment, neither TTR nor the THIPs inhibited hREC proliferation (Fig. S6). Furthermore, the apoptosis of hRECs was not significantly influenced by either the THIPs or TTR (Fig. S7).

The THIPs suppressed migration, wound healing, and tube formation in hRECs under hyperglycemic conditions

The results of the Transwell assay revealed that hREC migration was enhanced in a high-glucose environment, but not in the presence of TTR or the designed THIPs (Fig. 3A). In addition, the wound healing assay (Fig. 3B, C) demonstrated that both TTR and the THIPs significantly inhibited the accelerated hREC healing observed under hyperglycemic conditions, with THIP-4 showing the most significant suppressive effect. Furthermore, in the tube formation analysis (Fig. 3D, E), hyperglycemia significantly increased tube length, a trend that was reversed by TTR and the THIPs; again, THIP-4 displayed the strongest inhibitory action.

Fig. 3
figure 3

The THIPs inhibited migration, wound healing, and tube formation of hRECs. A Transwell assay (32 h; scale bar: 500 μm). B, C Wound healing assay (24 h; 1, vs. HG; 2, vs. HG + NC; 3, vs. HG + THIP-4; n = 5, *P < 0.05, **P < 0.01, ***P < 0.001); scale bar: 200 μm. D, E Tube formation assay (4 h; 1, vs. HG; 2, vs. HG + NC; 3, vs. HG + THIP-4; n = 5, *P < 0.05, **P < 0.01, ***P < 0.001); scale bar: 200 μm. HG high glucose, NC negative control

Full size image

The THIPs influenced hREC viability via the STAT4/hsa-miR-223-3p/FBXW7 and Notch1/Akt/mTOR axes

In the gene expression analysis, qPCR results showed that STAT4 and hsa-miR-223-3p expression levels were significantly increased in a high-glucose environment, effects that were reversed in the presence of TTR or the THIPs, with THIP-4 displaying the strongest suppressive effect (Fig. 4A, B). In contrast, the expression level of FBXW7 was markedly decreased under hyperglycemia, but this trend was reversed with the addition of TTR or THIP1–4; among these, THIP-4 exhibited the most significant promotive effect (Fig. 4C).

Fig. 4
figure 4

The THIPs regulated the STAT4/hsa-miR-223-3p/FBXW7 and Notch1/Akt/mTOR pathways in hRECs. A–C The expression levels of STAT4 mRNA, hsa-miR-223-3p, and FBXW7 mRNA (1, vs. HG; 2, vs. HG + NC; 3, vs. HG + THIP-4; n = 5, *P < 0.05, **P < 0.01, ***P < 0.001). D Western blot assay. E–I The relative protein levels of STAT4 and p-STAT4 (1, vs. HG STAT4; 2, vs. HG p-STAT4; 3, vs. HG + NC STAT4; 4, vs. HG + NC p-STAT4; n = 3, *P < 0.05, **P < 0.01, ***P < 0.001), FBXW7, Notch1, p-AKT, and p-mTOR (1, vs. HG; 2, vs. HG + NC; n = 3, *P < 0.05, **P < 0.01, ***P < 0.001). HG high glucose, NC negative control

Full size image

Meanwhile, western blotting analysis demonstrated that the protein levels of STAT4, p-STAT4 (Fig. 4D, E), Notch 1 (Fig. 4D, G), p-AKT (Fig. 4D, H), and p-mTOR (Fig. 4D, I) were increased in a high-glucose environment, whereas those of FBXW7 were decreased (Fig. 4D, F); notably, these alterations were significantly reversed after the addition of TTR or THIP1–4.

Meanwhile, the HIF-1α expression level remained stable in all the groups (Fig. S8A, B). Additionally, although the level of VEGFA was significantly enhanced under hyperglycemia, neither the THIPs nor TTR could reverse this effect (Fig. S8A, C).

THIP-4 mitigated DR progression in rats

The high conservation of amino acid sequences between human (UniProtKB/Swiss-Prot: P22626.2, amino acid) and rat (NCBI Reference Sequence: NP_001098083.2) hnRNPA2B1 (100% coverage and 100% sequence similarity) (Fig. S9A) suggested that the THIPs could also target rat hnRNPA2B1 in vivo. In addition, human (NCBI Reference Sequence: NP_000362.1; PDB ID: 1ICT) and rat (GenBank: AAH86946.1; PDB ID: 1IE4) TTRs display over 95% sequence similarity (Fig. S9B and C). Meanwhile, we also noted that the amino acid sequence of STAT4 is highly conserved between human and rat (Fig. S10A), as is the sequence of miR-223-3p and its interaction with the 3′-UTR of FBXW7 mRNA (Fig. S10B). These observations suggested that the STAT4/miR-223-3p/FBXW7 signaling pathway is also conserved between the two species. As THIP-4 showed the most significant anti-neovascularization effect in vitro, we inserted its sequence into the pAAV-MCS vector, which was subsequently packaged in AAV9 particles and administered to rats by intravitreal injection. Saline and TTR serving served as controls.

After 16 weeks, the rats were euthanized and the retinas were extracted. The retinas were subsequently subjected to an Evans blue assay using a reverse model, in which the retinal vascular leakage areas were highlighted as dark shadows (Fig. 5A). Compared with the control group (10.1 ± 1.8%), the leakage area was significantly increased in both the STZ (30.8 ± 2.9%) and STZ + NC (29.6 ± 2.2%) groups; however, these trends could be partially rescued by THIP-4 administration (13.3 ± 2.5%) or TTR overexpression (11.5 ± 1.4%) (Fig. 5B). Notably, serum albumin leakage showed a similar trend, as determined by western blotting (Fig. 5C).

Fig. 5
figure 5

THIP-4 mitigated the progression of DR in a rat model. A Evans blue (EB) stained retinas were visualized with the EVOS FL Auto Imaging System using a reverse-phase setting, where vascular and leakage appeared in dark shadow (scale bar: 100 μm). B Retinal vascular leakage (n = 5 visual fields in three retinas; 1, vs. normal control; 2, vs. STZ + NC; *P < 0.05, **P < 0.01, ***P < 0.001). C Retinal vascular leakage quantified by serum albumin western blotting (n = 3; 1, vs. normal control; 2, vs. STZ + NC; *P < 0.05, **P < 0.01, ***P < 0.001). D Retinas were digested with trypsin, following which the retinal vasculature was subjected to periodic acid-Schiff staining (scale bar: 25 μm; red arrows: acellular capillaries; green arrows: retained pericytes; blue arrows: lost pericytes). E Acellular capillary number and (F) human retinal microvascular epithelial cell (hREC)/pericyte ratio (n = 5, 1, vs. normal control; 2, vs. STZ + NC; *P < 0.05, **P < 0.01, ***P < 0.001). STZ streptozotocin, NC negative control

Full size image

The Evans blue-stained retinas were washed in sterile saline, digested with trypsin, and then subjected to periodic acid-Schiff staining (Fig. 5D). Compared with the control group (4.4 ± 2.1), the number of acellular capillaries was significantly increased in both the STZ (18.2 ± 3.0) and STZ + NC (15.6 ± 1.9) groups; however, this trend was reversed by THIP-4 administration (4.6 ± 3.1) or TTR overexpression (6.2 ± 2.2) (Fig. 5E). Additionally, as shown in Fig. 5F, the hREC/pericyte ratio was significantly increased in the STZ (8.90 ± 1.29) and STZ + NC (9.08 ± 0.79) groups compared with that in the control group (3.75 ± 0.48); however, the administration of THIP-4 (4.18 ± 0.53) or TTR overexpression (4.41 ± 0.57) partially offset this increase.

The eyeballs of the rats were paraffin-embedded, sliced into 5-μm-thick sections, and then subjected to H&E (Fig. 6A) or immunohistochemical staining. ImageJ was used for the analysis of the relative positivity ratios of target proteins. The results showed that positivity for fibronectin (a biomarker of fibrosis) (Fig. 6B) and PCNA (a biomarker of proliferation) in the STZ and STZ + NC groups was significantly increased compared with that in the normal control group (Fig. 6C); however, the positivity ratios for these two proteins were lower in the STZ + THIP-4 and STZ + TTR groups than in the STZ and STZ + NC groups. Additionally, although the levels of HIF-1α (induced by a hypoxic microenvironment) (Fig. 6D) and VEGFA (Fig. 6E) were upregulated in the STZ-induced DR microenvironment, neither THIP-4 administration nor TTR overexpression affected the levels of these two proteins under the same conditions.

Fig. 6
figure 6

Immunohistochemical assays of the rat retina. A Retinal Sects. (5 μm thick) were subjected to hematoxylin and eosin (H&E) staining (scale bar: 200 μm. GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; RPE, retinal pigment epithelium) or immunohistochemistry targeting fibronectin, PCNA, HIF-1α, or VEGFA (scale bar: 100 μm). B–E The relative positivity ratio for fibronectin, PCNA, HIF-1α, and VEGFA (n = 3; 1, vs. normal control; 2, vs. STZ + NC; *P < 0.05, **P < 0.01, ***P < 0.001). STZ streptozotocin, NC negative control

Full size image

The eyeball sections were then subjected to immunoreactivity assays (Fig. 7A, B). Compared with the normal control group, the levels of glial fibrillary acidic protein (GFAP, a biomarker of retinal gliosis) were markedly increased (Fig. 7A, C) while those of glutamate synthase (GS) were decreased (Fig. 7A, D) in the STZ and STZ + NC groups, suggesting that retinal gliosis was significantly enhanced in the DR rat model; however, this trend was efficiently reversed by the administration of THIP-4 or the overexpression of TTR. Furthermore, the immunoreactivity assay indicated that both THIP-4 (Fig. 7B, E) and TTR (Fig. 7B, F) were highly expressed in the eye. Similar results were obtained using western blot and dot blot assays (Fig. S11).

Fig. 7
figure 7

Immunofluorescence assays of the rat retina. Immunostaining for (A) GFAP/GS and (B) TTR/His-tagged THIP-4 on retinal paraffin sections (scale bar: 100 μm). GCL, ganglion cell layer; IPL, inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; RPE, retinal pigment epithelium. C–F The relative positivity ratio for GFAP, GS, TTR, and His-tagged THIP-4 in the retina (n = 3; 1, vs. normal control; 2, vs. STZ + NC; *P < 0.05, **P < 0.01, ***P < 0.001). STZ streptozotocin, NC negative control

Full size image

Meanwhile, in dot blot assay, the signal of THIP-4 could only be detected in retina, but not in brain and serum (Fig. S12A). Compared with the normal group, the levels of SCr (Fig. S12B) and BUN (Fig. S12C), and the activities of ALT (Fig. S12D), AST (Fig. S12E), and ALP (Fig. S12F), were all significantly promoted, which could not be affected by TTR or THIP-4. These results demonstrated that using such an intravitreal AAV administration strategy, the THIP-4 was only located in eye.

THIP-4 inhibited DR progression in rats via the STAT4/rno-miR-223-3p/FBXW7 and Notch1/Akt/mTOR axes

QPCR analysis showed that the expression of rno-miR-223-3p was significantly increased in STZ-treated rats, a trend that was reversed by the administration of THIP-4 or the overexpression of TTR (Fig. 8A). Meanwhile, western blot assay results (Fig. 8B) demonstrated that following STZ treatment, the protein levels of STAT4, p-STAT4 (Fig. 8C), Notch 1 (Fig. 8E), p-AKT (Fig. 8F), and p-mTOR (Fig. 8G) were increased, whereas those of FBXW7 were decreased in retinal tissue of DR rats (Fig. 8D). Nevertheless, these alterations were all significantly reversed following THIP-4 administration or TTR overexpression.

Fig. 8
figure 8

THIP-4 regulated the STAT4/rno-miR-223-3p/FBXW7 and Notch1/Akt/mTOR pathways in a DR rat model. A The expression level of rno-miR-223-3p in the rat retina (1, vs. normal control; 2, vs. STZ + NC; n = 3, *P < 0.05, **P < 0.01, ***P < 0.001). B Western blot assay of the proteins extracted from retinal tissue. C–G The relative protein levels of STAT4 and p-STAT4 (1, vs. normal control STAT4; 2, vs. normal control p-STAT4; 3, vs. STZ + NC STAT4; 4, vs. STZ + NC p-STAT4; n = 3, *P < 0.05, **P < 0.01, ***P < 0.001), FBXW7, Notch1, p-AKT, and p-mTOR (1, vs. normal control; 2, vs. STZ + NC; n = 3, *P < 0.05, **P < 0.01, ***P < 0.001). STZ streptozotocin; NC negative control

Full size image

Additionally, the levels of HIF-1α and VEGFA in the eyes of rats from all the STZ treatment groups were upregulated following long-term exposure to a hypoxic and hyperglycemic microenvironment (Fig. S13A and B). However, neither THIP-4 administration nor elevated TTR expression affected the levels of these two proteins.

Discussion

The increasing prevalence of diabetes has emerged as a major global health challenge [1,2,3]. DR is an important microvascular complication of diabetes [4,5,6], often leading to vision loss, visual impairment, and even blindness [7,8,9,10]. One of the most effective current clinical treatments for DR is the intraocular injection of anti-VEGF antibodies such as pegaptanib, ranibizumab, and bevacizumab [11,12,13]. Despite their efficacy, these therapies pose challenges, including the development of resistance and potential adverse effects on healthy vasculature. Consequently, identifying novel biomolecules with anti-neovascularization properties that can complement anti-VEGF antibody therapy has become a focal point in DR research.

HnRNPA2B1 is a nuclear protein involved in transcriptional and posttranscriptional regulation, acting as a “reader” of the m6A modification of mRNA [23,24,25,26]. This protein has been implicated in different biological processes, including RNA regulation [25] and angiogenesis [31, 32], as well as in a variety of diseases, such as multiple myeloma [27, 28] colorectal cancer [29], breast cancer [41, 42], and gastric cancer [43]. Pathological angiogenesis is associated with the development of various tumors [44, 45], and, like DR, anti-VEGF antibodies, which inhibit angiogenesis, have been used in clinical settings for the treatment of these cancers. Given its role as a regulator of angiogenesis, hnRNPA2B1 represents a potential therapeutic target for both DR and tumor neovascularization. In our previous work, we demonstrated that TTR directly interacts with hnRNPA2B1 and inhibits neovascularization in hRECs in vitro via the STAT4/miR-223-3p/FBXW7 signaling axis, however, TTR did not regulate the level of hnRNPA2B1 [33,34,35].

In the present study, we designed a set of THIPs and systematically investigated their hnRNPA2B1 recognition potential and biofunctions in vitro and in vivo. As shown in Fig. 1A, B, in our previous study [35], three sites of TTR on the TTR-hnRNPA2B1 interface were linked via a soft linker, and a His-tag was added to the C-terminus for localization analysis (THIP-1); additionally, employing energy minimization, we identified two Arg substitutions that stabilized the TTR-hnRNPA2B1 complex [35]. In this study, based on the THIP-1 sequence, we designed three additional THIPs through single or double substitutions.

To investigate the affinities of the THIPs for hnRNPA2B1, several protocols were employed. In the Biacore assay, the Kd values for TTR, THIP-1, THIP-2, THIP-3, and THIP-4 in hnRNPA2B1 targeting were determined as 3.678, 2.568, 1.989, 1.934, and 2.285 μM, respectively (Fig. S3) (Table 1). Additionally, using Langmuir equilibrium adsorption analysis, the KD value for the interaction between TTR and hnRNPA2B1 was calculated as 7.13 ± 0.37, while those for the interaction between THIP 1–4 and hnRNPA2B1were found to be 5.34 ± 0.37, 4.58 ± 0.21, 4.53 ± 0.36, and 6.49 ± 0.21 μM, respectively (Fig. S4) (Table 1). These KD values suggested that both TTR and the THIPs could specifically recognize hnRNPA2B1. Although the determined affinities were at a level similar to that observed with polyclonal antibodies and lower than those of well-designed monoclonal antibodies [46,47,48], the sequences have scope for optimization for further reducing the dissociation constant. The interaction between each of the THIPs and hnRNPA2B1 was also confirmed via a co-IP assay (Fig. 1C).

Like TTR, the THIPs rapidly entered the nucleus of hRECs (Fig. 1D), and reduced DNA synthesis (Fig. 2A, B), proliferation (Fig. 2C), migration (Fig. 3A), wound healing (Fig. 3B, C), and tube formation (Fig. 3D, E) in the cells under a high-glucose environment. In addition, a CCK-8 assay showed that neither TTR nor the designed THIPs could inhibit hREC proliferation in low glucose environment (Fig. S6). These results highlighted the anti-neovascularization activity of THIPs in hyperglycemia in vitro. Western blot and qPCR analyses indicated that the THIPs and TTR suppressed angiogenesis by reducing the levels of STAT4, p-STAT4 (Fig. 4A, D, E), hsa-miR-223-3p (Fig. 4B), Notch 1 (Fig. 4D, G), p-AKT (Fig. 4D, H), and p-mTOR (Fig. 4D, I) and increasing that of FBXW7 (Fig. 4C, D, F); notably, all of which these factors act in the STAT4/miR-223-3p/FBXW7 pathway and its downstream Notch1/Akt/mTOR signaling axis (Fig. 9). Because neither the THIPs nor TTR affected hREC apoptosis under conditions of hyperglycemia (Fig. S7), apoptosis-associated pathways were not investigated in this work.

Fig. 9
figure 9

The mechanism underlying the anti-angiogenic function of the TTR-hnRNPA2B1 interface peptides (THIPs)

Full size image

Alignment analysis showed that the hnRNPA2B1 sequence was 100% conserved between human and rat (Fig. S9A). Given that THIP-4 exhibited the most significant repressive effect on neovascularization in vitro, its sequence was inserted into the pAAV-MCS vector, packaged in AAV9 particles, and expressed in the eye of DR rats via intravitreal injection. Although the sequences of human and rat TTR do not completely overlap, to avoid potential antigenicity, rat TTR was overexpressed in the rat eye as a control. In the DR rat model, THIP-4 administration significantly mitigated the typical DR-associated retinal phenotypes (Fig. 5), including retinal vascular leakage [49], capillary decellularization [50], and pericyte loss [51]. Additionally, THIP-4 inhibited two other typical features of DR, namely, retinal fibrosis (quantified using fibronectin) and gliosis (quantified using GFAP) (Figs. 6 and 7). Retinal gliosis is predominantly associated with Müller cells [52, 53]. While this study focused only on hRECs for in vitro investigations, understanding how THIP-4 affects Müller cell gliosis in DR rats remains an intriguing question. Notably, recent studies have highlighted cellular senescence as a critical factor in DR development [54]. Klotho, an anti-senescence gene implicated in aging, neurodegeneration, and kidney disease, is downregulated by miR-223-3p [55, 56]. In future work, we plan to systematically investigate whether there exists a hnRNPA2B1/STAT4/hsa-miR-223-3p/Klotho pathway in Müller cells.

As a m6A reader, in theory, hnRNPA2B1 could recognize m6A modification on various RNAs, and it has been reported to facilitate tumor development via regulating TLR4, ILF3, and TCF7L2 mRNAs [27,28,29], to accelerate angiogenesis via eliminating miR-204-3p or promoting Stop codon decoding and synthesis of VEGF-Ax [31, 32]. TTR and THIPs could specifically target hnRNPA2B1, and then regulate STAT4/miR-223-3p/FBXW7 pathway and its downstream Notch1/Akt/mTOR signaling axis. However, after the nucleic acid binding domain of hnRNPA2B1 was blocked, STAT4/miR-223-3p/FBXW7 pathway should not be the only affected signaling axis. In future work, we hope to investigate the effects of TTR and THIPs on other reported hnRNPA2B1 associated pathways, and to find out the potential crosstalk system between these signal axes.

As previously reported, high glucose environment could induce the significantly increased VEGFA levels in retina cells, including hRECs [57, 58], ARPE-19 cells [59] and RF/6A cells [60]. VEGF has been wildly accepted as the clinical target for anti-angiogenesis therapy, not only for DR but also for a number of cancers, and anti-VEGF antibodies have been commonly applied for clinical therapy; but the resistance and potential side effects in normal tissues are still the important challenges. And in this work, we tested the effects of THIPs and TTR on VEGF and HIF-1α. As in the in vitro experiment (Fig. S8), the administration of either THIP-4 or TTR did not affect HIF-1α and VEGFA protein levels in DR rats (Fig. S13), despite the known importance of the HIF-1α/VEGF signaling pathway in DR and tumor associated angiogenesis [61, 62]. These results suggested that the TTR or THIPs dominated hnRNPA2B1/STAT4/hsa-miR-223-3p/FBXW7 axis and its downstream Notch1/Akt/mTOR pathway should be parallel to the classic HIF-1α/VEGF pathway. Therefore, the designed THIPs exhibit substantial promise as co-therapeutics alongside anti-VEGF antibodies in the clinical treatment of DR or tumor. In the future work, in vivo and invitro experiments would be designed for the co-administration of anti-VEGF antibodies and THIPs.

In this work, the safety assessment of THIP-4 was not systematically investigated. However, regarding Fig. S12A, THIP-4 was only detected in retina, and it did not pass the blood-eye and blood–brain barriers. Additionally, all the STZ induced rats showed similar kidney (Fig. S12B and C) and liver (Fig. S12D-F) injuries. Via intravitreal AAV injection, THIP-4 was located in the micro-environment in eye, it could not spread to the whole body and could not affect other organs. In future work, the administration strategies for THIPs would be systematically investigated, e.g., intravitreal peptide or AAV injection, intravenous peptide or AAV injection, and then the long-term safety of THIPs would be carefully studied.

To investigate the specific of THIPs, the captured proteins in IP assay were trypsinized and identified via LC–MS/MS assay. HnRNPA2B1 was identified with the highest Δ LFQ intensity in each group (Fig. S5 and Table S3), which suggested that the peptides designed based on the interface of TTR-hnRNPA2B1 complex could specifically recognize hnRNPA2B1. Additionally, several differential THIPs-captured proteins were identified with much lower LFQ intensities; and in our further work, the potential relationship between these proteins and angiogenesis would be studied.

Conclusions

In conclusion, we designed THIPs based on the TTR-hnRNPA2B1 interface. THIP-4 and TTR exhibited anti-angiogenic properties in vitro and suppressed DR progression in vivo via the STAT4/miR-223-3p/FBXW7 pathway and the downstream Notch1/Akt/mTOR axis. These observations suggested that the THIPs can be used in conjunction with anti-VEGF antibodies in the clinical treatment of DR or some tumors. To maximize their therapeutic potential, future research efforts should focus on enhancing their affinity for hnRNPA2B1 and extending their half-life.

Availability of data and materials

All data associated with this study are available within the article itself and its supplemental files. Raw data are available upon request from the corresponding authors.

Abbreviations

HnRNPA2B1:

Heterogeneous nuclear ribonucleoprotein A2B1

TTR:

Transthyretin

THIPs:

The TTR-hnRNPA2B1 interface peptides

AAV:

Adeno-associated viruses

DMEM:

Dulbecco’s modified Eagle’s medium

DR:

Diabetic retinopathy

EB:

Evans blue

GFAP:

Glial fibrillary acidic protein

GS:

Glutamine synthetase

H&E:

Hematoxylin and eosin

LSD:

Least significant difference

PMF:

Peptide map fingerprinting

STZ:

Streptozotocin

References

  1. Chen Y, Lundeen EA, Koyama AK, Kompaniyets L, Andes LJ, Benoit SR, Imperatore G, Rolka DB. Prevalence of testing for diabetes among US adults with overweight or obesity, 2016–2019. Prev Chronic Dis. 2023;20:E116.

    Article  PubMed  PubMed Central  Google Scholar 

  2. Kautzky-Willer A, Leutner M, Harreiter J. Sex differences in type 2 diabetes. Diabetologia. 2023;66(6):986–1002.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Krastev E, Abanos S, Kovachev P, Tcharaktchiev D. Diabetes prevalence and duration data extracted from outpatient records representative for the Bulgarian population. Stud Health Technol Inform. 2023;305:230–3.

    PubMed  Google Scholar 

  4. Lundeen EA, Burke-Conte Z, Rein DB, Wittenborn JS, Saaddine J, Lee AY, Flaxman AD. Prevalence of diabetic retinopathy in the US in 2021. JAMA Ophthalmol. 2023;141(8):747–54.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Hou XH, Wang LM, Zhu DL, Guo LX, Weng Ji, Zhang M, Zhou ZG, Zou DJ, Ji QH, Guo XH, et al. Prevalence of diabetic retinopathy and vision-threatening diabetic retinopathy in adults with diabetes in China. Nat Commun. 2023; 14(1).

  6. Jensen ET, Rigdon J, Rezaei KA, Saaddine J, Lundeen EA, Dabelea D, Dolan LM, D’Agostino R, Klein B, Meuer S, et al. Prevalence, progression, and modifiable risk factors for diabetic retinopathy in youth and young adults with youth-onset type 1 and type 2 diabetes: the SEARCH for diabetes in youth study. Diabetes Care. 2023;46(6):1252–60.

    Article  PubMed  PubMed Central  Google Scholar 

  7. Li HB, Liu XY, Zhong H, Fang JN, Li XA, Shi R, Yu Q. Research progress on the pathogenesis of diabetic retinopathy. BMC Ophthalmol. 2023; 23(1).

  8. Haider S, Adderley N, Tallouzi MO, Sadiq SN, Steel DH, Chavan R, Sheikh I, Nirantharakumar K, Snell KIE. Diabetic retinopathy progression in patients under monitoring for treatment or vision loss: external validation and update of a multivariable prediction model. BMJ Open. 2023;13(4): e073015.

    Article  PubMed  PubMed Central  Google Scholar 

  9. Kong XR. Diabetic retinopathy in the US—where we are now and what is next. JAMA Ophthalmol. 2023;141(8):754.

    Article  PubMed  Google Scholar 

  10. Tomkins-Netzer O, Niederer R, Lightman S. The role of statins in diabetic retinopathy. Trends Cardiovasc Med. 2024;34(2):128–35.

    Article  CAS  PubMed  Google Scholar 

  11. Martinez-Zapata MJ, Salvador I, Martí-Carvajal AJ, Pijoan JI, Cordero JA, Ponomarev D, Kernohan A, Solà I, Virgili G. Anti-vascular endothelial growth factor for proliferative diabetic retinopathy. Cochrane Database Syst Rev. 2023; 2023(3).

  12. Muns SM, Villegas VM, Flynn HW, Schwartz SG. Update on current pharmacologic therapies for diabetic retinopathy. Expert Opin Pharmacother. 2023;24(14):1577–93.

    Article  CAS  PubMed  Google Scholar 

  13. Sharma D, Zachary I, Jia HY. Mechanisms of acquired resistance to anti-VEGF therapy for neovascular eye diseases. Invest Ophthalmol Vis Sci. 2023;64(5):28.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  14. Chen CM, Ding P, Yan WM, Wang ZM, Lan YY, Yan XL, Li T, Han J. Pharmacological roles of lncRNAs in diabetic retinopathy with a focus on oxidative stress and inflammation. Biochem Pharmacol. 2023;214: 115643.

    Article  CAS  PubMed  Google Scholar 

  15. Geng MR, Liu W, Li JJ, Yang G, Tian Y, Jiang X, Xin Y. LncRNA as a regulator in the development of diabetic complications. Front Endocrinol. 2024;15:1324393.

    Article  Google Scholar 

  16. Anuradha U, Mehra NK, Khatri DK. Understanding molecular mechanisms and miRNA-based targets in diabetes foot ulcers. Mol Biol Rep. 2024; 51(1).

  17. Luo Y, Li C. Advances in research related to MicroRNA for diabetic retinopathy. J Diabetes Res. 2024;2024:1–21.

    Article  Google Scholar 

  18. Wu JL, Chen JQ, Hu J, Yao MD, Zhang M, Wan XL, Jia HX, Wang FH, Sun XD. CircRNA Uxs1/miR-335-5p/PGF axis regulates choroidal neovascularization via the mTOR/p70 S6k pathway. Transl Res. 2023;256:41–55.

    Article  CAS  PubMed  Google Scholar 

  19. Simó R, Hernández C. What else can we do to prevent diabetic retinopathy? Diabetologia. 2023;66(9):1614–21.

    Article  PubMed  PubMed Central  Google Scholar 

  20. Srivastava SP, Upadhyay P, Das S, Tiwari N, Mishra S, Tripathi SM. Managing diabetic complications with alternative therapeutic strategies. Clin Exp Optom. 2024;9:1–10.

    Google Scholar 

  21. Chen JT, Huang YJ, Liu CF, Chi JW, Wang YG, Xu LL. The role of C-peptide in diabetes and its complications: an updated review. Front Endocrinol. 2023; 14.

  22. Lin TY, Kang EYC, Shao SC, Lai ECC, Hwang YS. Risk of diabetic retinopathy between sodium-glucose cotransporter-2 inhibitors and glucagon-like peptide-1 receptor agonists. Diabetes Metab J. 2023;47(4):573–4.

    Article  PubMed  PubMed Central  Google Scholar 

  23. Keene JD. RNA regulons: coordination of post-transcriptional events. Nat Rev Genet. 2007;8(7):533–43.

    Article  CAS  PubMed  Google Scholar 

  24. Alarcón Claudio R, Goodarzi H, Lee H, Liu XH, Tavazoie S, Tavazoie SF. HNRNPA2B1 is a mediator of m6a-dependent nuclear RNA processing events. Cell. 2015;162(6):1299–308.

    Article  PubMed  PubMed Central  Google Scholar 

  25. Jiang XL, Liu BY, Nie Z, Duan LC, Xiong QX, Jin ZX, Yang CP, Chen YB. The role of m6A modification in the biological functions and diseases. Transduct Target Ther. 2021; 6(1).

  26. Wu BX, Su SC, Patil DP, Liu HH, Gan JH, Jaffrey SR, Ma JB. Molecular basis for the specific and multivariant recognitions of RNA substrates by human hnRNP A2/B1. Nat Commun. 2018;9(1):420.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Jia CM, Guo YW, Chen Y, Wang XY, Xu QT, Zhang Y, Quan LN. HNRNPA2B1-mediated m6A modification of TLR4 mRNA promotes progression of multiple myeloma. J Transl Med. 2022; 20(1).

  28. Jiang FJ, Tang XZ, Tang C, Hua Z, Ke MY, Wang C, Zhao JM, Gao SY, Jurczyszyn A, Janz S, et al. HNRNPA2B1 promotes multiple myeloma progression by increasing AKT3 expression via m6A-dependent stabilization of ILF3 mRNA. J Hematol Oncol. 2021; 14(1).

  29. Liu H, Li DX, Sun LN, Qin HQ, Fan AH, Meng LN, Graves-Deal R, Glass SE, Franklin JL, Liu Q, et al. Interaction of lncRNA MIR100HG with hnRNPA2B1 facilitates m6A-dependent stabilization of TCF7L2 mRNA and colorectal cancer progression. Mol Cancer. 2022; 21(1).

  30. Wang L, Wen MY, Cao XT. Nuclear hnRNPA2B1 initiates and amplifies the innate immune response to DNA viruses. Science. 2019; 365(6454).

  31. Guo QD, Fan Y, Wang QT, Li B, Qiu W, Qi YH, Pan ZW, Zhang SJ, Zhao SL, Yang K, et al. Glioblastoma upregulates SUMOylation of hnRNP A2/B1 to eliminate the tumor suppressor miR-204-3p, accelerating angiogenesis under hypoxia. Cell Death Dis. 2023; 14(2).

  32. Smith MR, Costa G. RNA-binding proteins and translation control in angiogenesis. FEBS J. 2021;289(24):7788–809.

    Article  PubMed  Google Scholar 

  33. Gu Y, Hu D, Xin Y, Shao J. Transthyretin affects the proliferation and migration of human retinal microvascular endothelial cells in hyperglycemia via hnRNPA2B1. Biochem Biophys Res Commun. 2021;557:280–7.

    Article  CAS  PubMed  Google Scholar 

  34. Shao J, Fan GM, Yin XW, Gu Y, Wang XL, Xin Y, Yao Y. A novel transthyretin/STAT4/miR-223-3p/FBXW7 signaling pathway affects neovascularization in diabetic retinopathy. Mol Cell Endocrinol. 2019;498: 110541.

    Article  CAS  PubMed  Google Scholar 

  35. Tian YK, Ye L, Shao J, Xin Y. The interface targeting hnRNPA2B1 regulates the repression of transthyretin against human retinal microvascular endothelial cells in high-glucose environment. Diabet Med. 2023;40(9): e15125.

    Article  CAS  PubMed  Google Scholar 

  36. Fang YK, Zhou YJ, Xin Y, Shi Y, Guo ZT, Li YR, Gu ZH, Ding ZY, Shi GY, Zhang L. Preparation and characterization of a novel thermostable lipase from Thermomicrobium roseum. Catal Sci Technol. 2021;11(22):7386–97.

    Article  CAS  Google Scholar 

  37. Xin Y, Hao MY, Fan GM, Zhang Y, Zhang L. Soluble expression of Thermomicrobium roseum sarcosine oxidase and characterization of N-demethylation activity. Mol Catal. 2019;464:48–56.

    Article  CAS  Google Scholar 

  38. Wang JJ, Huang ZX, Ji L, Chen C, Wan Q, Xin Y, Pu ZN, Li KK, Jiao JT, Yin Y, et al. SHF acts as a novel tumor suppressor in glioblastoma multiforme by disrupting STAT3 dimerization. Adv Sci. 2022;9(26):2200169.

    Article  CAS  Google Scholar 

  39. Shao J, Gu WD, Ye L, Xin Y. The hsa_circ_0004805/hsa_miR-149-5p/TGFB2 axis plays critical roles in the pathophysiology of diabetic retinopathy in vitro and in vivo. Mol Cell Endocrinol. 2023;576: 112042.

    Article  CAS  PubMed  Google Scholar 

  40. Ye L, Chen YX, Gu WD, Shao J, Xin Y. Hsa_circ_0004776 regulates the retina neovascularization in progression of diabetic retinopathy hsa-miR-382-5p/BDNFaxis. Arch Physiol Biochem. 2024;130(6):921–33.

    Article  CAS  PubMed  Google Scholar 

  41. Wang JY, Zhang JY, Liu H, Meng LN, Gao XC, Zhao YH, Wang C, Gao XL, Fan AH, Cao TY, et al. N6-methyladenosine reader hnRNPA2B1 recognizes and stabilizes NEAT1 to confer chemoresistance in gastric cancer. Cancer Commun. 2024;44(4):469–90.

    Article  Google Scholar 

  42. Zhou XQ, Hong YL, Liu YP, Wang L, Liu X, Li Y, Yuan H, Hu FQ. Intervening in hnRNPA2B1-mediated exosomal transfer of tumor-suppressive miR-184-3p for tumor microenvironment regulation and cancer therapy. J Nanobiotechnol 2023; 21(1).

  43. Jin T, Yang LP, Chang C, Luo HJ, Wang R, Gan YB, Sun Y, Guo YT, Tang R, Chen SC, et al. HnRNPA2B1 ISGylation regulates m6A‐tagged mRNA selective export via ALYREF/NXF1 complex to foster breast cancer development. Adv Sci. 2024; 11(24).

  44. Liu ZL, Chen HH, Zheng LL, Sun LP, Shi L. Angiogenic signaling pathways and anti-angiogenic therapy for cancer. Signal Transduct Target Ther. 2023; 8(1).

  45. Lugano R, Ramachandran M, Dimberg A. Tumor angiogenesis: causes, consequences, challenges and opportunities. Cell Mol Life Sci. 2019;77(9):1745–70.

    Article  PubMed  PubMed Central  Google Scholar 

  46. Hao JQ, Jin R, Yi YM, Jiang XS, Yu JY, Xu Z, Schnicker NJ, Chimenti MS, Sugg SL, Li B. Development of a humanized anti-FABP4 monoclonal antibody for treatment of breast cancer. Breast Cancer Res. 2024;26:119.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  47. Song J, Jeong BS, Kim SW, Im SB, Kim S, Lai CJ, Cho W, Jung JU, Ahn MJ, Oh BH. Noncovalent antibody catenation on a target surface greatly increases the antigen-binding avidity. Elife. 2023;12: e81646.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  48. Sun ZH, Chu XJ, Adams C, Ilina TV, Guerrero M, Lin GW, Chen C, Jelev D, Ishima R, Li W, et al. Preclinical assessment of a novel human antibody VH domain targeting mesothelin as an antibody-drug conjugate. Mol Ther Oncolytics. 2023;31: 100726.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  49. Hammes HP. Diabetic retinopathy: hyperglycaemia, oxidative stress and beyond. Diabetologia. 2017;61(1):29–38.

    Article  PubMed  Google Scholar 

  50. Xu GT, Zhang JF, Tang L. Inflammation in diabetic retinopathy: possible roles in pathogenesis and potential implications for therapy. Neural Regen Res. 2023;18(5):976.

    Article  PubMed  Google Scholar 

  51. Tomita Y, Lee D, Tsubota K, Negishi K, Kurihara T. Updates on the current treatments for diabetic retinopathy and possibility of future oral therapy. J Clin Med. 2021;10(20):4666.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  52. Hachana S, Pouliot M, Couture R, Vaucher E. Diabetes-induced inflammation and vascular alterations in the Goto-KAKIZAKI rat Retina. Curr Eye Res. 2020;45(8):965–74.

    Article  PubMed  Google Scholar 

  53. Rosales MAB, Silva KC, Lopes de Faria JB, Lopes de Faria JM. Exogenous SOD mimetic tempol ameliorates the early retinal changes reestablishing the redox status in diabetic hypertensive rats. Invest Ophth Vis Sci. 2010;51(8):4327.

    Article  Google Scholar 

  54. Crespo-Garcia S, Fournier F, Diaz-Marin R, Klier S, Ragusa D, Masaki L, Cagnone G, Blot G, Hafiane I, Dejda A, et al. Therapeutic targeting of cellular senescence in diabetic macular edema: preclinical and phase 1 trial results. Nat Med. 2024;30(2):443–54.

    Article  CAS  PubMed  Google Scholar 

  55. Zhang XY, Li L, Tan HS, Hong X, Yuan Q, Hou FF, Zhou LL, Liu YH. Klotho-derived peptide 1 inhibits cellular senescence in the fibrotic kidney by restoring Klotho expression via posttranscriptional regulation. Theranostics. 2024;14(1):420–35.

    Article  PubMed  PubMed Central  Google Scholar 

  56. Zhu ZH, Quadri Z, Crivelli SM, Elsherbini A, Zhang LP, Tripathi P, Qin HY, Roush E, Spassieva SD, Nikolova-Karakashian M, et al. Neutral sphingomyelinase 2 mediates oxidative stress effects on astrocyte senescence and synaptic plasticity transcripts. Mol Neurobiol. 2022;59(5):3233–53.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  57. Wu S, Zhang Y, Hou Y, Zhu J, Yang H, Cui Y. Research on the role of exosomes secreted by immortalized adipose-derived mesenchymal stem cells differentiated into pericytes in the repair of high glucose-induced retinal vascular endothelial cell damage. Exp Eye Res. 2024;247: 110046.

    Article  CAS  PubMed  Google Scholar 

  58. Li J, Lu X, Wei L, Ye D, Lin J, Tang X, Cui K, Yu S, Xu Y, Liang X. PHD2 attenuates high-glucose-induced blood retinal barrier breakdown in human retinal microvascular endothelial cells by regulating the Hif-1alpha/VEGF pathway. Inflamm Res. 2022;71(1):69–79.

    Article  CAS  PubMed  Google Scholar 

  59. Sanguineti R, Puddu A, Nicolò M, Traverso CE, Cordera R, Viviani GL, Maggi D. miR-126 mimic counteracts the increased secretion of VEGF-A induced by high glucose in ARPE-19 cells. J Diabetes Res. 2021;2021:6649222.

    Article  PubMed  PubMed Central  Google Scholar 

  60. Xie WJ, Zhou P, Qu MW, Dai ZR, Zhang XL, Zhang CY, Dong X, Sun GB, Sun XB. Ginsenoside Re attenuates high glucose-induced RF/6A injury via regulating PI3K/AKT inhibited HIF-1α/VEGF signaling pathway. Front Pharmacol. 2020;11:695.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  61. Palazon A, Tyrakis PA, Macias D, Veliça P, Rundqvist H, Fitzpatrick S, Vojnovic N, Phan AT, Loman N, Hedenfalk I, et al. An HIF-1α/VEGF-A axis in cytotoxic T cells regulates tumor progression. Cancer Cell. 2017;32(5):669–83.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  62. Song S, Zhang GH, Chen XT, Zheng J, Liu XD, Wang YQ, Chen ZJ, Wang YX, Song YL, Zhou Q. HIF-1α increases the osteogenic capacity of ADSCs by coupling angiogenesis and osteogenesis via the HIF-1α/VEGF/AKT/mTOR signaling pathway. J Nanobiotechnol. 2023;21(1):257.

    Article  CAS  Google Scholar 

Download references

Funding

This work was supported by the China Postdoctoral Science Foundation (2020M671541) and the Top Talent Support Program for young and middle-aged people of Wuxi Health Committee (BJ2023017).

Author information

Authors and Affiliations

  1. The Key Laboratory of Industrial Biotechnology, Ministry of Education, School of Biotechnology, National Engineering Research Center for Cereal Fermentation and Food Bio Manufacturing, Jiangnan University, Wuxi, 214122, Jiangsu, China

    Yixiu Chen, Lu Ye, Shujing Cui & Yu Xin

  2. Department of Ophthalmology, The Affiliated Wuxi People’s Hospital of Nanjing Medical University, Wuxi People’s Hospital, Wuxi Medical Center, Nanjing Medical University, Wuxi, 214023, Jiangsu, China

    Jun Shao

Authors
  1. Yixiu Chen
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Lu Ye
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Shujing Cui
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Jun Shao
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Yu Xin
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

YC involved in formal analysis, investigation, methodology, visualization and writing—original draft. LY and SC involved in formal analysis, investigation and visualization. JS involved in methodology, resources, supervision, funding acquisition, writing—review and editing. YX involved in methodology, supervision, resources, project administration, writing—original draft, writing—review and editing. All authors approved the final manuscript.

Corresponding authors

Correspondence to Jun Shao or Yu Xin.

Ethics declarations

Ethics approval and consent to participate

The study was approved by the Ethics Committee of the Affiliated Wuxi People’s Hospital of Nanjing Medical University (2022-033, 18-Oct-2022) in China.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Chen, Y., Ye, L., Cui, S. et al. Peptides based on the interface of hnRNPA2B1-transthyretin complex repress retinal angiogenesis in diabetic retinopathy. J Transl Med 23, 458 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12967-025-06437-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12967-025-06437-y

Keywords

Journal of Translational Medicine

ISSN: 1479-5876

Contact us