Abstract
Objective
This study aims to investigate the correlation between the levels of hypoxia-inducible factor-1α (HIF-1α) in the peripheral blood of patients with type 2 diabetes (T2D) and early-stage diabetic retinopathy (DR) and to evaluate its potential as a predictor of early DR.
Methods
From November 2021 to December 2023, 70 patients who underwent fundus photography at Zhangzhou Second Hospital were recruited. Based on the results, patients were categorized into a T2D group comprising 25 patients and a DR group comprising 45 patients. In addition, 18 healthy individuals who underwent routine physical examinations during the same period were included as a control group. Serum levels of the HIF-1α protein were measured in all three groups.
Results
The results indicated that patients in the early-stage DR group had significantly higher levels of HIF-1α, albumin (Alb), Ca and blood urea nitrogen (BUN) compared to the T2D group (P < 0.05). In addition, patients in the DR group exhibited higher levels of total cholesterol, high-density lipoprotein cholesterol (HDL-C), fasting plasma glucose (FPG) and glycated hemoglobin (HbA1C) than the control group (P < 0.05). The incidence rates of DR in groups A, B, C and D were 18.18, 31.82, 59.09 and 95.45%, respectively, with statistically significant differences (P < 0.05). Spearman’s correlation analysis revealed significant positive correlations between HIF-1α and age, disease duration (years), systolic blood pressure, Cr, FPG and HbA1c, with a strong positive correlation between HIF-1α and HbA1c (P < 0.05) optimal cutoff value of 2.3855 ng/mL.
Conclusion
Increased levels of HIF-1α in peripheral blood are closely linked to the development of DR, suggesting that HIF-1α may act as a potential biomarker for the early detection and prediction of DR risk.
Introduction
There are approximately 19.5 million patients with diabetic retinopathy (DR) in China, of whom about 3 to 4 million experience vision loss or even blindness due to retinal complications (1). Simultaneously, DR is one of the leading causes of blindness in adults over the age of fifty worldwide (2). DR is classified as a specific microvascular complication of diabetes (3), where patients with DR often remain in a prolonged state of hyperglycemia, leading to abnormal energy metabolism and tissue hypoxia (4), which can easily trigger microcirculatory disorders, resulting in microvascular lesions and subsequent pathological changes such as nerve and vascular damage, hemorrhage and exudation (5). In the early stages, retinal changes in DR patients are non-proliferative (NDR) and asymptomatic, making it difficult for patients to notice (6). Often, due to a lack of timely medical consultation, it develops into proliferative diabetic retinopathy, which has a higher risk of vision-threatening complications. The occurrence and progression of DR are closely related to various factors, with hypoxia being a central element in its development. Hypoxia-inducible factor-1α (HIF-1α) is currently the only known transcriptional regulatory factor that remains active in hypoxic environments (7). By regulating the expression of a series of downstream functional genes such as vascular endothelial growth factor (VEGF) (8), it mediates the response of organs and tissues to microcirculatory hypoxia, thereby influencing the pathological progression of diabetes and its complications. Given the central role of HIF-1α in the hypoxic response, this study aims to analyze the relationship between DR and peripheral blood levels of HIF-1α and evaluate its potential as a biomarker for the early prediction of DR.
Materials and methods
Study design and population
From November 2021 to December 2023, a total of 88 patients who underwent fundus photography at Zhangzhou Second Hospital were recruited as study subjects, comprising 42 men (47.7%) and 46 women (52.3%). According to the diagnostic criteria for diabetes, they were divided into 25 participants in the T2D group (no retinopathy) and 45 participants in the DR group (with early-stage DR). The control group consisted of 18 healthy individuals who underwent physical examinations during the same period. Based on the quartiles of peripheral blood HIF-1α levels among the 88 subjects, they were categorized into four groups (A, B, C and D). This study obtained approval from the Institutional Review Board of the hospital, and all enrolled patients signed informed consent forms.
Participants were included in the study if they met the following criteria: i) individuals who fulfill the diagnostic criteria for T2D as outlined in the national T2D prevention and treatment guidelines (9); ii) patients aged 30 years or older with comprehensive medical records; iii) patients who consent to undergo fundus photography and have been diagnosed by an ophthalmologist as being in the early-stage DR; iv) patients who have signed an informed consent form. Participants were excluded from the study if they met the following criteria: i) patients with other forms of diabetes; ii) patients with severe liver or kidney impairment; iii) patients with infections, hematological disorders (particularly thalassemia), autoimmune diseases or malignant tumors; iv) patients with glaucoma, hemangiomas or other ocular conditions; v) patients with psychological disorders that may hinder cooperation with examinations and follow-up studies; vi) participants with other microvascular complications, including diabetic nephropathy; vii) patients determined by an ophthalmologist to be in the proliferative retinopathy stage of DR; and viii) patients who refuse to sign the informed consent form or whose information is incomplete.
Diagnostic criteria
Type 2 diabetes diagnostic criteria
According to the ‘Clinical guidelines for prevention and treatment of type 2 diabetes mellitus in the elderly in China (2022 edition)’ (9), the diagnosis of T2D is established when a patient exhibits typical symptoms of diabetes such as increased thirst, polydipsia, polyuria, polyphagia and unexplained weight loss alongside one or more of the following criteria: random blood glucose level ≥200 mg/dL; fasting plasma glucose (FPG) ≥126 mg/dL; 2 h glucose tolerance test (OGTT) blood glucose ≥200 mg/dL; glycated hemoglobin (hemoglobin A1C, HbA1C) ≥6.5%.
DR diagnostic criteria
According to diabetic retinopathy guidelines (10), patients with diabetes can be diagnosed with DR if fundus photography reveals any of the following findings, contingent upon a certain duration of the disease: microaneurysms; small hemorrhages; hard exudates; cotton wool spots (soft exudates); neovascularization; fibrovascular proliferation; vitreous hemorrhage; vitreous organization; tractional retinal detachment or blindness.
Clinical data collection
Basic clinical data of participants: age, gender, body mass index (BMI), duration of illness, smoking history and blood pressure will be consistently collected by specialized nurses. Biochemical indicators: laboratory physicians will measure the blood biochemical indicators of participants using a fully automated biochemical analyzer (Hitachi HITACHI7600, Japan). The following parameters will be assessed: low-density lipoprotein cholesterol (LDL-C), albumin (Alb), high-density lipoprotein cholesterol (HDL-C), calcium (Ca), phosphorus (P), creatinine (Cr), blood urea nitrogen (BUN), HbA1C, triglycerides (TG), FPG and total cholesterol (TC). Fundus photography results: fundus photography will be performed by a specialized ophthalmologist from our hospital using the Nikon AFC330 fundus camera. Both the quality control personnel and the examiner conducting the fundus photography will be part of the same team.
Serum sample collection and methods
Participants in each group will fast for 12 h before venipuncture to collect 5–6 mL of venous blood. This blood will then be centrifuged at 3,000 rpm (approximately 1,600 g ) for 15 min, and the supernatant will be collected and stored at −80°C for later analysis. The levels of HIF-1α in the serum samples from each group will be measured using an enzyme-linked immunosorbent assay (ELISA) kit (Jianglai Biological JL11823, 96T, China). The instruments employed will include a full-wavelength microplate reader (MDSpectraMax Plus 384, USA) and a high-precision spectrophotometer (Merinton SMA4000, USA).
Statistical analyses
Data were organized using Excel 2021. The normality of continuous variables was assessed with the Kolmogorov–Smirnov test. Variables conforming to a normal distribution were reported as such, while those not conforming were summarized using the median and interquartile range. Pairwise comparisons were conducted using the independent samples t-test, and multiple groups were analyzed using one-way ANOVA or the Mann–Whitney U test as appropriate. Categorical variables were expressed as percentages. Inter-group comparisons were performed with the chi-square test, and the correlation between HIF-1α and other indicators was evaluated using Spearman’s rank correlation. The diagnostic value of serum HIF-1α for early-stage DR was assessed using the receiver operating characteristic (ROC) curve. All statistical analyses were performed using SPSS 26.0 software, with a two-tailed P < 0.05 indicating statistical significance. All figures were visualized using GraphPad Prism 9.6.0.
Quality control measures
Study subjects were meticulously selected in strict accordance with the established inclusion and exclusion criteria to prevent selection bias; a consistent team of experienced ophthalmologists conducted and evaluated the fundus photography results to ensure accuracy and consistency; the collection of clinical data was personally overseen by the principal investigator, with rigorous checks and verifications performed to guarantee the authenticity and reliability of the data samples; serum samples were stored at −80°C upon collection; all pipette tips and centrifuge tubes used in the experiments were subjected to sterilization processes to maintain aseptic conditions.
Results
General demographic characteristics
The age of all participants ranged from 35 to 82 years, with a mean age of 62.98 ± 11.33 years. This study received approval from the Institutional Review Board (IRB) of the hospital, and all enrolled participants provided informed consent. In the study, there were 18 participants in the control group, 25 participants in the T2DM group and 45 participants in the DR group. A comparison of the baseline data of the three groups of subjects revealed that the DR group had significantly higher age (65.40 ± 10.71), disease duration (7.21 ± 6.59) and systolic blood pressure (SBP) (141.42 ± 24.16 mmHg) than the control group. Compared to the T2D group (3.37 ± 4.73), only the disease duration in the DR group was significantly higher, with a statistically significant difference (P < 0.05). In addition, there were no statistically significant differences in gender, smoking status, diastolic blood pressure (DBP) and BMI among the three groups (P < 0.05) (Table 1).
Baseline characteristics of the participants.
| Variables | Control (n = 18) | T2D (n = 25) | DR (n = 45) | F/χ2 | P |
|---|---|---|---|---|---|
| Age (years) | 58.56 ± 9.64 | 61.80 ± 12.72 | 65.40 ± 10.71* | 2.631 | 0.078 |
| Smoking | 1.804 | 0.406 | |||
| Yes n (%) | 6.00 (6.80%) | 4.00 (4.50%) | 10.00 (11.40%) | ||
| No n (%) | 12.00 (17.60%) | 21.00 (23.90%) | 35.00 (39.80%) | ||
| Sex | 1.172 | 0.557 | |||
| Woman n (%) | 8.00 (9.10%) | 12.00 (13.60%) | 26.00 (29.50%) | ||
| Man n (%) | 10.00 (11.40%) | 13.00 (14.80%) | 19.00 (21.60%) | ||
| Disease duration (years) | 0.00 (0.00, 0.00) | 1.50 (0.24, 5.00) | 6.00 (2.00, 10.00)* , # | 25.130 | <0.001 |
| SBP (mmHg) | 126.94 ± 5.35 | 136.20 ± 19.75 | 141.42 ± 24.16* | 3.249 | 0.044 |
| DBP (mmHg) | 81.39 ± 4.74 | 85.00 ± 13.20 | 82.76 ± 13.80 | 0.487 | 0.616 |
| BMI (Kg/m2) | 22.49 ± 2.73 | 25.53 ± 4.87 | 24.17 ± 4.13 | 2.847 | 0.064 |
and
represent the DR group comparisons with the control and T2D group, respectively. P < 0.05.
Biochemical indicators
One-way ANOVA and non-parametric tests were employed to assess the biochemical indicators of the three groups of patients. The results indicated that patients in the DR group had significantly higher levels of HIF-1α, Alb, Ca and BUN when compared to the T2D group (P < 0.05). Furthermore, patients in the DR group had higher levels of TC, high-density lipoprotein HDL-C, FPG and HbA1c than those in the control group. The LDL-C levels were higher in the DR group when compared to the T2D group, with statistically significant differences (P < 0.05). However, levels of TG, LDL-C, Cr and BUN did not display statistically significant differences when compared to the T2D group (P < 0.05) (Table 2).
Comparison of clinical biochemical indicators among control, T2D and DR groups.
| Variables | Control (n = 18) | T2D (n = 25) | DR (n = 45) | F/U | P |
|---|---|---|---|---|---|
| HIF-α (ng/mL) | 1.02 (0.72, 1.31) | 1.48 (1.08, 2.20) | 2.95 (2.12, 3.90)* , # | 25.884 | <0.001 |
| Alb (g/L) | 47.91 ± 2.18 | 40.54 ± 3.77 | 39.13 ± 6.02* , # | 21.336 | <0.001 |
| Ca (mmol/L) | 2.45 ± 0.17 | 2.29 ± 0.22 | 2.23 ± 0.14* , # | 10.535 | <0.001 |
| P (mmol/L) | 1.11 (0.95, 1.23) | 1.14 (1.01, 1.31) | 1.22 (1.13, 1.35) | 2.493 | 0.287 |
| Cr (mmol/L) | 76.60 (69.05, 89.03) | 69.00 (56.00, 79.00) | 81.00 (59.00, 118.60) | 2.738 | 0.254 |
| TC (mmol/L) | 5.30 (4.56, 5.66) | 4.96 (4.15, 6.04) | 4.58 (4.01, 5.09)* | 4.535 | 0.104 |
| TG (mmol/L) | 1.61 (1.00, 2.32) | 1.52 (1.15, 1.93) | 1.78 (1.05, 2.42) | 1.138 | 0.566 |
| LDL-C (mmol/L) | 3.18 (2.58, 3.91) | 3.34 (2.55, 4.39) | 2.50 (2.11, 3.28) # | 7.884 | 0.019 |
| HDL-C (mmol/L) | 1.25 (1.17, 1.37) | 1.36 (1.02, 1.58) | 1.45 (1.16, 1.72)* | 4.684 | 0.096 |
| FPG (mmol/L) | 5.40 (5.30, 5.80) | 9.20 (6.49, 11.59) | 9.31 (7.50, 12.15)* | 23.040 | <0.001 |
| BUN (mmol/L) | 5.33 (4.22, 5.86) | 4.71 (3.82, 5.23) | 6.50 (4.82, 8.56)* , # | 14.000 | <0.001 |
| HbA1C (%) | 5.55 (5.30, 6.43) | 7.80 (7.00, 8.80) | 8.60 (8.00, 10.60)* | 17.671 | <0.001 |
and
represent the DR group comparisons with the control and T2D group, respectively. P < 0.05.
Peripheral blood HIF-1α expression levels across different groups
The levels of HIF-1α in the DR group, T2D group and control group were (3.35 ± 0.02), (1.76 ± 0.00) and (1.35 ± 0.01) ng/mL, respectively. Moreover, the HIF-1α levels in the DR group were significantly higher than those in the T2D group and the control group (P < 0.05) (Fig. 1).
The level of serum HIF-1α in three groups.
Citation: Endocrine Connections 14, 5; 10.1530/EC-25-0004
Comparison of examination results in subjects with varying levels of HIF-1α
Statistical analysis revealed significant differences in disease duration (years), Alb, Ca, TC, FPG, BUN and HbA1C among the four patient groups (P < 0.05). In addition, disease duration (years), FPG, BUN and HbA1C significantly increased from group A to group D (P < 0.05), suggesting that these parameters significantly increase with elevated levels of HIF-1α. Furthermore, other clinical indicators among the four groups did not demonstrate statistically significant differences (P ≥ 0.05) (Table 3).
Comparison of clinical biochemical indicators among A, B, C and D groups.
| Variables | A (n = 22) | B (n = 22) | C (n = 22) | D (n = 22) | F/U/χ 2 | P |
|---|---|---|---|---|---|---|
| Age (years) | 60.00 ± 11.65 | 60.55 ± 9.43 | 66.86 ± 11.90 | 66.50 ± 11.55 | 1.806 | 0.152 |
| Smoking | 5.284 | 0.152 | ||||
| Yes (%) | 16.00 (18.20%) | 18.00 (20.50%) | 16.00 (18.20%) | 18.00 (20.50%) | ||
| No (%) | 6.00 (6.80%) | 4.00 (4.50%) | 6.00 (6.80%) | 4.00 (4.50%) | ||
| Sex | 1.035 | 0.793 | ||||
| Woman (%) | 11.00 (12.50%) | 9.00 (13.60%) | 10.00 (11.40%) | 16.00 (18.20%) | ||
| Man (%) | 11.00 (12.50%) | 13.00 (14.80%) | 12.00 (14.81%) | 6.00 (6.80%) | ||
| Disease duration (years) | 0.00 (0.00, 2.00) | 0.38 (0.04, 3.50) | 0.875 (6.00, 10.00) | 4.25 (9.00, 10.00) | 18.769 | <0.001* |
| SBP (mmHg) | 131.45 ± 15.05 | 131.55 ± 16.85 | 138.57 ± 23.33 | 144.91 ± 24.68 | 2.302 | 0.083 |
| DBP (mmHg) | 82.64 ± 7.54 | 85.82 ± 12.33 | 80.73 ± 14.49 | 83.27 ± 13.82 | 0.638 | 0.593 |
| BMI (Kg/m2) | 23.20 ± 4.02 | 24.95 ± 4.16 | 24.87 ± 5.03 | 23.83 ± 3.55 | 0.885 | 0.452 |
| Alb (g/L) | 47.91 ± 2.18 | 41.56 ± 4.61 | 40.56 ± 5.50 | 37.72 ± 6.59 | 17.091 | <0.001* |
| Ca (mmol/L) | 2.41 ± 0.18 | 2.31 ± 0.17 | 2.26 ± 0.17 | 2.20 ± 0.14 | 5.515 | 0.002* |
| P (mmol/L) | 1.17 ± 0.21 | 1.17 ± 0.13 | 1.14 ± 0.18 | 1.28 ± 0.30 | 1.912 | 0.134 |
| Cr (mmol/L) | 72.85 (52.80, 91.80) | 71.55 (60.25, 82.00) | 75.00 (57.50, 103.73) | 90.50 (59.05, 175.25) | 2.545 | 0.467 |
| TC (mmol/L) | 4.83 (4.55, 5.43) | 5.09 (4.43, 5.92) | 4.58 (3.88, 5.54) | 4.46 (3.91, 5.12) | 8.141 | 0.043* |
| TG (mmol/L) | 1.60 (1.21, 1.93) | 1.85 (1.09, 2.44) | 1.40 (1.01, 2.25) | 1.55 (1.00, 2.95) | 2.545 | 0.467 |
| LDL-C (mmol/L) | 3.13 (2.61, 3.62) | 3.27 (2.45, 4.35) | 2.88 (2.12, 4.47) | 2.50 (1.94, 3.35) | 3.273 | 0.351 |
| HDL-C (mmol/L) | 1.29 (1.22, 1.58) | 1.29 (1.06, 1.63) | 1.36 (1.05, 1.57) | 1.47 (1.23, 1.84) | 2.545 | 0.467 |
| FPG (mmol/L) | 5.75 (5.38, 9.06) | 7.73 (5.91, 9.33) | 8.51 (6.65, 9.73) | 10.55 (9.26, 13.42) | 14.182 | 0.003* |
| BUN (mmol/L) | 5.24 (4.23, 5.86) | 4.41 (4.02, 5.29) | 4.65 (4.65, 7.24) | 7.53 (5.02, 10.42) | 12.000 | 0.007* |
| HbA1C (%) | 6.28 ± 1.21 | 7.53 ± 0.98 | 8.82 ± 1.60 | 10.51 ± 2.14 | 30.177 | <0.001* |
Indicates that there are statistical differences in variables among the four groups. P < 0.05.
The correlation of HIF-1α with other indicators
Spearman’s correlation analysis indicated that HIF-1α had significant positive correlations with age, disease duration (years), SBP, Cr, FPG and HbA1C. Notably, HIF-1α showed a strong positive correlation with HbA1c (ρ = 0.748) (|ρ| ≥ 0.70, P < 0.05). Interestingly, HIF-1α exhibited weak yet significant negative correlations with Alb) (ρ = −0.480) and Ca (ρ = −0.446) (0.3 ≥ |ρ| > 0.5, P < 0.05). The remaining variables, such as BMI, P and HDL-C, exhibited positive correlations with HIF-1α; however, no statistically significant differences were observed for BMI, P, HDL-C, sex, smoking, DBP, TC, TG and LDL-C (P ≥ 0.05) (Fig. 2).
Correlation between HIF-1α and other indexes.
Citation: Endocrine Connections 14, 5; 10.1530/EC-25-0004
Incidence of DR at varying levels of HIF-1α
The incidence of DR from group A to group D increased significantly by 18.18, 31.82, 59.09 and 95.45%, respectively (P < 0.05) (Fig. 3).
Diagnostic value of predicting HIF-1α level by ROC curve in DR.
Citation: Endocrine Connections 14, 5; 10.1530/EC-25-0004
Diagnostic value of HIF-1α levels in DR
According to the ROC curve analysis, the area under the curve (AUC) for diagnosing DR based on HIF-1α levels is 0.833 (95% CI (0.743–0.923)), with an optimal cutoff value of 2.3855 ng/mL (Fig. 4).
Diagnostic value of predicting HIF-1α level by ROC curve in DR.
Citation: Endocrine Connections 14, 5; 10.1530/EC-25-0004
Discussion
Diabetes is a chronic metabolic disorder characterized by persistent hyperglycemia, which significantly increases the risk of damage to the heart, brain and major blood vessels in the limbs. This condition is also associated with a higher likelihood of developing various cardiovascular diseases (8). In addition, diabetes insidiously affects the microvasculature of the retina, kidneys and nervous system. DR is one of the most common microvascular complications of diabetes and is recognized as one of the leading causes of blindness worldwide. In DR, abnormalities in retinal microcirculation result in microvascular damage and the pathological formation of new blood vessels. These changes can also be associated with optic neuropathy and cataracts, significantly threatening patients’ quality of life (11). Many patients present in the proliferative stage of DR at their first visit due to the absence of specific clinical manifestations in earlier stages, leading to irreversible vision impairment. Therefore, the early identification and treatment of DR are essential for enhancing patients’ visual and overall quality of life.
In patients with DR, the ocular microcirculation, particularly the microvessels in the retinal fundus, frequently experiences pathological damage due to oxidative stress, hemodynamic changes and inflammatory responses, with hypoxia-induced microcirculatory disorders playing a central role (12). HIF-1 is a transcriptional activator that responds to fluctuations in oxygen levels and regulates the expression of genes critical for processes such as cell proliferation and apoptosis (13). HIF-1 consists of two subunits, HIF-1α and HIF-1β, with HIF-1α as the primary active subunit. Under hypoxic conditions (14), HIF-1α acts as a key transcriptional regulator (15), promoting microvascular endothelial cell proliferation (16) and facilitating new blood vessel formation (17). This process attempts to mitigate the hypoxic state of the retina (18) and is linked to the progression of diabetic complications (19). Research has demonstrated that as oxygen levels in tissue cells decline, the concentration of HIF-1α increases correspondingly (20). In various tumor studies, overexpression of HIF-1α is often associated with abnormal cell growth and tissue hypoxia (21).
HIF-1α serves as a central transcriptional regulator in neovascularization and plays a crucial role in the pathogenesis of DR (22). Recent studies have established a strong association between HIF-1α and DR. Notably, one investigation demonstrated a significant increase in HIF-1α expression in the retinas of DR model mice compared to healthy control mice (23). In our study, the levels of peripheral serum HIF-1α in patients with DR were significantly higher than those in both the T2D group and the healthy control group. Moreover, HIF-1α levels in the T2D group were also significantly elevated compared to the control group, suggesting that diabetic patients may exist in a state of chronic hypoxia. These findings further substantiate the close relationship between elevated HIF-1α levels and the onset and progression of DR, aligning with previous research findings.
To further investigate the association between HIF-1α levels and DR, we categorized 88 patients into four groups based on the quartiles of HIF-1α levels and statistically analyzed their clinical baseline data and biochemical indicators. The analysis revealed significant differences among the four groups in several parameters, including disease duration (years), Alb, HbA1c, TC, FPG, BUN and Ca (P < 0.05). Notably, group D exhibited higher levels of disease duration, FPG, BUN and HbA1c compared to the other groups. Spearman’s correlation analysis indicated significant positive correlations between HIF-1α and age, disease duration, SBP, Cr, FPG and HbA1c, with a particularly strong correlation between HIF-1α and HbA1c (ρ = 0.748) (|ρ| ≥ 0.70, P < 0.05). Elevated HbA1c levels reflect poor long-term glycemic control, and traditional risk factors for DR include diabetes duration, HbA1c and FPG. HIF-1α levels tend to increase with worsening retinal ischemia and hypoxia, which may influence the development of retinal neovascularization alongside these risk factors. Interestingly, our study also identified weak negative correlations between HIF-1α and Alb (ρ = −0.480) and Ca (ρ = −0.446) (0.3 ≥ |ρ| >0.5, P < 0.05). These findings suggest additional factors that may affect HIF-1α levels and retinal health, highlighting the need for further research to clarify the clinical significance of these relationships.
Recent studies have investigated changes in HIF-1α levels across various stages of DR and identified its predictive value for disease progression (24). In addition, Xu et al. indicate that HIF-1α can upregulate the expression of VEGF, thereby promoting the formation of new blood vessels in the retina (25). Other studies corroborate these findings, demonstrating that levels of HIF-1α and VEGF in the serum of DR patients were significantly higher compared to diabetic patients without DR (26). In clinical practice, anti-VEGF therapy has shown efficacy for patients with proliferative DR (27). VEGF is a crucial mediator in the formation of new blood vessels in the retina and its expression in DR is finely regulated by HIF-1α (24).
Through fundus examination, a significant increase in the incidence of DR was observed corresponding to elevated HIF-1α levels. Subsequent analysis utilizing the ROC curve assessed the diagnostic value of HIF-1α levels for DR. The results indicated that when HIF-1α levels reached ≥2.3855 ng/mL, the incidence of DR significantly increased compared to those in patients with NDR, indicating high predictive specificity and sensitivity for the progression from T2D to DR. These findings underscore the potential of HIF-1α as a biomarker for DR and suggest novel perspectives for future treatment strategies (Fig. 5). Therefore, we recommend using HIF-1α as a biomarker for evaluating the risk of DR in diabetic patients, which could enable early diagnosis and intervention.
The role of hypoxia and HIF-1α in DR development, including key mechanisms such as oxidative stress, HIF-1α activation, VEGF upregulation and subsequent microvascular damage.
Citation: Endocrine Connections 14, 5; 10.1530/EC-25-0004
Nonetheless, first, although we observed significant associations between HIF-1α levels and DR, a limitation of this study is the small sample size; therefore, future research should aim to validate the correlation between HIF-1α and DR in larger cohorts and explore the therapeutic potential of HIF-1α in the management of DR. Second, longitudinal studies are needed to determine whether HIF-1α levels dynamically change over time during DR progression. Such studies could track HIF-1α expression at non-proliferative and proliferative stages of DR and correlate these changes with clinical outcomes. Third, while we controlled for several confounding factors, unmeasured variables such as dietary habits, physical activity or genetic predispositions may have influenced our results. Future studies should incorporate comprehensive lifestyle and genetic data to better account for these factors.
Declaration of interest
The authors confirmed that no potential conflicts of interest exist that could influence or bias the research reported in this work.
Funding
This study was supported by the Fujian Provincial Department of Education Middle and Young Teachers’ educational research project (No. JAT20866) and Zhangzhou Health Vocational College Institutional Research Project (No. ZWYZ202103).
Author contribution statement
ZQS and LJC wrote the main manuscript. MJL prepared the data collection. JYL and LPQ prepared figure and tables. LTS, ZQS and LJC analyzed and interpreted the results. All authors reviewed the results and approved the final version of the manuscript. All authors would be informed about each step of manuscript processing including submission, revision and revision reminder.
Data availability
The experimental data used to support the findings of this study are available from the corresponding author upon request.
Ethics approval and consent to participate
This study was approved by the Ethics Committee of the Second Hospital of Zhangzhou. Informed consent was obtained from all the participants. All methods were carried out in accordance with the Declaration of Helsinki.
Consent for publication
All the authors confirm that written informed consent was obtained from all subjects and/or their legal guardian(s).
References
- 1↑
Hou X , Wang L , Zhu D , et al. China National Diabetic Chronic Complications (DiaChronic) Study Group. Prevalence of diabetic retinopathy and vision-threatening diabetic retinopathy in adults with diabetes in China. Nat Commun 2023 14 4296. (https://doi.org/10.1038/s41467-023-39864-w)
- 2↑
Li Y , Yu Y & VanderBeek BL . Anaemia and the risk of progression from non-proliferative diabeticretinopathy to vision threatening diabetic retinopathy. Nat Eye 2020 34 934–941. (https://doi.org/10.1038/s41433-019-0617-6)
- 3↑
Chen M , Zhang Q , Wang S , et al. Inhibition of diabetes-induced Drp1 deSUMOylation prevents retinal vascular lesions associated with diabetic retinopathy. Exp Eye Res 2023 226 109334. (https://doi.org/10.1016/j.exer.2022.109334)
- 4↑
Rorex C , Cardona SM , Church KA , et al. CX3CR1-Fractalkine dysregulation affects retinal GFAP expression, inflammatory gene induction, and LPS response in a mouse model of hypoxic retinopathy. Int J Mol Sci 2025 26 1131. (https://doi.org/10.3390/ijms26031131)
- 5↑
Chen X , Wang Y , Wang JN , et al. Lactylation-driven FTO targets CDK2 to aggravate microvascular anomalies in diabetic retinopathy. EMBO Mol Med 2024 16 294–318. (https://doi.org/10.1038/s44321-024-00025-1)
- 6↑
Dan AO , Ștefănescu-Dima A , Bălășoiu AT , et al. Early retinal microvascular alterations in young type 1 diabetic patients without clinical retinopathy. Diagnostics 2023 13 1648. (https://doi.org/10.3390/diagnostics13091648)
- 7↑
Fallah J & Rini BI . HIF inhibitors: status of current clinical development. Curr Oncol Rep 2019 21 6. (https://doi.org/10.1007/s11912-019-0752-z)
- 8↑
Jin J , Wang X , Zhi X , et al. Epigenetic regulation in diabetic vascular complications. J Mol Endocrinol 2019 63 R103–R115. (https://doi.org/10.1530/jme-19-0170).
- 9↑
Chinese Elderly Type 2 Diabetes Prevention and Treatment of Clinical Guidelines Writing Group . Clinical guidelines for prevention and treatment of type 2 diabetes mellitus in the elderly in China (2022 edition). Zhonghua Nei Ke Za Zhi 2022 61 12–50. (https://doi.org/10.3760/cma.j.cn112138-20211027-00751)
- 10↑
Fong DS , Aiello L , Gardner TW , et al. Diabetic retinopathy. Diabetes Care 2003 26(Supp. Suppl 1) 99–102. (https://doi.org/10.2337/diacare.26.2007.s99)
- 11↑
Balaratnasingam C , An D , Hein M , et al. Studies of the retinal microcirculation using human donor eyes and high-resolution clinical imaging: insights gained to guide future research in diabetic retinopathy. Prog Retin Eye Res 2023 94 101134. (https://doi.org/10.1016/j.preteyeres.2022.101134)
- 12↑
Min J , Zeng T , Roux M , et al. The role of HIF1α-PFKFB3 pathway in diabetic retinopathy. J Clin Endocrinol Metab 2021 106 2505–2519. (https://doi.org/10.1210/clinem/dgab362)
- 13↑
Vadlapatla RK , Vadlapudi AD & Mitra AK . Hypoxia-inducible factor-1 (HIF-1): a potential target for intervention in ocular neovascular diseases. Curr Drug Targets 2013 14 919–935. (https://doi.org/10.2174/13894501113149990015)
- 14↑
Kiani AA , Kazemi A , Halabian R , et al. HIF-1α confers resistance to induced stress in bone marrow-derived mesenchymal stem cells. Arch Med Res 2013 44 185–193. (https://doi.org/10.1016/j.arcmed.2013.03.006)
- 15↑
Zhao M , Wang S , Zuo A , et al. HIF-1α/JMJD1A signaling regulates inflammation and oxidative stress following hyperglycemia and hypoxia-induced vascular cell injury. Cell Mol Biol Lett 2021 26 40. (https://doi.org/10.1186/s11658-021-00283-8)
- 16↑
Sun K , Chen Y , Zheng S , et al. Genipin ameliorates diabetic retinopathy via the HIF-1α and AGEs-RAGE pathways. Phytomedicine 2024 129 155596. (https://doi.org/10.1016/j.phymed.2024.155596)
- 17↑
Zhang J , Qin Y , Martinez M , et al. HIF-1α and HIF-2α redundantly promote retinal neovascularization in patients with ischemic retinal disease. J Clin Invest 2021 131 e139202. (https://doi.org/10.1172/jci139202)
- 18↑
D’Amico AG , Maugeri G , Reitano R , et al. PACAP modulates expression of hypoxia-inducible factors in streptozotocin-induced diabetic rat retina. J Mol Neurosci 2015 57 501–509. (https://doi.org/10.1007/s12031-015-0621-7)
- 19↑
Ai X , Yu P , Luo L , et al. Berberis dictyophylla F. inhibits angiogenesis and apoptosis of diabetic retinopathy via suppressing HIF-1α/VEGF/DLL-4/Notch-1 pathway. J Ethnopharmacol 2022 296 115453. (https://doi.org/10.1016/j.jep.2022.115453)
- 20↑
Kim SY , Yoon NG , Im JY , et al. Targeting the mitochondrial chaperone TRAP1 alleviates vascular pathologies in ischemic retinopathy. Adv Sci 2024 11. (https://doi.org/10.1002/advs.202302776)
- 21↑
Lu QL , Liu J , Zhu XL , et al. Expression of nerve growth factor and hypoxia inducible factor-1α and its correlation with angiogenesis in non-small cell lung cancer. J Huazhong Univ Sci Technolog Med Sci 2014 34 359–362. (https://doi.org/10.1007/s11596-014-1283-3)
- 22↑
Bilgin B , Bilak S & Özay Y . Comparison of HIF-1α and survivin levels in patients withdiabetes and retinopathy of varying severity. Arq Bras Oftalmol 2024 87 e2023. (https://doi.org/10.5935/0004-2749.2023-0112)
- 23↑
Song S , Zhang G , Chen X , et al. HIF-1α increases the osteogenic capacity of ADSCs by coupling angiogenesis and osteogenesis via the HIF-1α/VEGF/AKT/mTOR signaling pathway. J Nanobiotechnology 2023 21 257. (https://doi.org/10.1186/s12951-023-02020-z)
- 24↑
Chen B , Zou J , Xie L , et al. WNT-inhibitory factor 1-mediated glycolysis protects photoreceptor cells in diabetic retinopathy. J Transl Med 2024 22 245. (https://doi.org/10.1186/s12967-024-05046-5)
- 25↑
Xu Y , Zou H , Ding Q , et al. tiRNA-Val promotes angiogenesis via Sirt1-Hif-1α axis in mice with diabetic retinopathy. Biol Res 2022 55 14. (https://doi.org/10.1186/s40659-022-00381-7)
- 26↑
Ai X , Yu P , Luo L , et al. Berberis dictyophylla F. inhibits angiogenesis and apoptosis of diabetic retinopathy via suppressing HIF-1α/VEGF/DLL-4/Notch-1 pathway. J Ethnopharmacol 2022 296 115–453. (https://doi.org/10.1016/j.jep.2022.115453)
- 27↑
Uludag G , Hassan M , Matsumiya W , et al. Efficacy and safety of intravitreal anti-VEGF therapy in diabetic retinopathy: what we have learned and what should we learn further. Expert Opin Biol Ther 2022 22 1275–1291. (https://doi.org/10.1080/14712598.2022.2100694)
