Neuroprotective effects of etanercept on diabetic retinopathy via regulation of the TNF-α/NF-κB signaling pathway

Purpose: To study the influence of etanercept on diabetic retinopathy in rats via tumor necrosis factor alpha (TNF-α)/nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) signaling pathway. Methods: Thirty-six Sprague-Dawley (SD) rats were randomly divided into normal, model and etanercept groups. The expression of Caspase-3 was determined by immunohistochemistry, while the relative protein and mRNA expression levels of TNF-α and NF-κB were determined by Western blotting and quantitative polymerase chain reaction, respectively. Besides, the contents of TNF-α and interleukin-1 beta (IL-1β) were evaluated using enzyme-linked immunosorbent assay (ELISA), while cell apoptosis was assessed by terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling (TUNEL). Results: Immunohistochemical studies showed that the mean optical density of tissues positive for caspase-3 in both the model and etanercept groups were significantly higher than in the normal group (p < 0.05), while the mean optical density in the etanercept group was significantly lower than that in the model group (p < 0.05). The protein expression levels of TNF-α and NF-κB in the etanercept group were significantly lower than those in the model group (p < 0.05). Furthermore, mRNA expressions of TNF-α and NF-κB declined in the etanercept group (p < 0.05); in addition, TNF-α, and IL-1β levels in the etanercept group were lower than in the model group (p < 0.05). Cell apoptosis in the etanercept group was also lower than in the model group. Conclusion: Etanercept suppresses TNF-α/NF-κB signaling pathway thereby repressing inflammation and cell apoptosis in diabetic retinopathy rats. Therefore, etenercept’s neuroprotective effect may potentially be useful in developing a suitable therapy for diabetic neuropathy.


INTRODUCTION
The increase in the morbidity rate of diabetes makes diabetic retinopathy increasingly common in clinic, so that it has gradually become a severe microvascular complication of diabetes [1,2]. According to the statistics, diabetic retinopathy has been one of the leading causes of blindness in patients in the United States, seriously affecting the life, health, labor capacity and quality of life of patients. Therefore, it is vital to further explore the pathogenesis of diabetic retinopathy and the relevant treatment methods for alleviating diabetic retinopathy-induced damage to patients, and improving the life, health and quality of life of patients with this disease [3][4][5][6].
In the pathogenesis of diabetic retinopathy, massive optic nerve cell apoptosis (one of the crucial pathological reactions) can cause optic nerve damage, blindness and other critical conditions. Studies have demonstrated that the pivotal tumor necrosis factor alpha (TNFα)/nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) signaling pathway in organisms can regulate multiple pathological processes, such as apoptosis and inflammation [7,8]. The TNF-α/NF-κB signaling pathwaymediated inflammation and cell apoptosis are now considered to play crucial roles in diabetic retinopathy.
Etanercept, efficacious inhibitor of TNF-α in clinic, binds to the receptor of TNF-α to repress the pro-inflammatory effect of TNF-α, thereby effectively inhibiting inflammation, but the specific mechanism of action therein has not yet been elucidated [9,10]. Therefore, the present study aims to explore the influence of etanercept on diabetic retinopathy in rats through the TNFα/NF-κB signaling pathway.

EXPERIMENTAL Animals
Thirty-six specific pathogen-free laboratory Sprague-Dawley (SD) rats aged 1 month (Shanghai SLAC Laboratory Animal Co. Ltd., certificate No. SCXK (Shanghai, China) 2013-0006) were bred in the Experimental Animal Center of Venus Eye Hospital and given normal feed and sterile filtered water daily under a 12/12 h light-dark cycle, in a temperature-controlled room (21 ± 2 °C). This study was approved by the Animal Ethics Committee of Venus Eye Hospital Animal Center (approval no. 2019-07-12). All procedures were conducted in accordance with the 'Animal Research: Reporting in vivo experiments guidelines 2.0' [11].

Animal grouping
A total of 36 SD rats were randomly divided into normal group (n = 12), model group (n = 12) and etanercept group (n = 12) using a random number table, and they were adaptively fed in the Experimental Animal Center for 7 days before experiments.

Establishment of diabetic retinopathy model
First, 1 % streptozotocin solution was prepared, and intraperitoneally injected into rats at a dose of 60 mg/kg, and 3 days later, venous blood was withdrawn from the tails of the rats for determination of blood glucose, with the blood glucose > 16.7 mmol/L as the sign of successful diabetic modeling.
Following successful modeling, the rats in normal group were given normal food, but not treated, while the rats in model group and etanercept group were intraperitoneally injected with an equal amount of normal saline and etanercept at 5 mg/kg, respectively. After intervention for 7 days, all the rats were sacrificed by cervical dislocation (after being anesthetized using peritoneal administration of pentobarbital sodium at a dose of 40 mg/kg).

Sampling
Upon the completion of anesthesia, the retinal tissues were directly collected from 6 rats in each group, rinsed with normal saline, and preserved in Eppendorf (EP) tubes at -80 °C for subsequent Western blotting and qPCR. The remaining 6 rats in each group were fixed through perfusion for sampling as follows: The thoracic cavity of the rats was cut open to expose the heart, and 400 mL of 4 % paraformaldehyde was perfused through the left atrial appendage. The retinal tissues were taken out, and fixed in 4 % paraformaldehyde for immunohistochemistry and TUNEL assays.

Immunohistochemistry
The pre-paraffin-embedded tissues were sliced and placed in warm water at 42 °C to expand, they are then mounted, baked and prepared into paraffin-embedded tissue sections. The tissue sections were soaked in xylene solution and gradient ethanol successively for routine deparaffinization and hydration. Subsequently, the resulting sections were immersed in citrate buffer and subjected to complete antigen retrieval by heating repeatedly using a microwave oven 3 times (heating for 3 min and braising for 5 min per time). After rinsing, the tissue sections were added dropwise with endogenous peroxidase blocker, reacted for 10 min, rinsed and sealed with goat serum which was added dropwise for 20 min. With the goat serum blocking buffer discarded, the tissue sections were incubated with the anti-Caspase-3 primary antibody (1:200) in a refrigerator at 4 °C overnight. The next day, the sections were rinsed, added dropwise with the secondary antibody solution, and reacted for 10 min. After they were fully rinsed, the sections were reacted with streptavidin-peroxidase solution for 10 min, and added dropwise with DAB for color development, and cell nucleic was counterstained using hematoxylin. Finally, the sections were sealed and observed under a microscope.

Western blotting (WB)
The cryopreserved retinal tissues were added with lysis buffer, bathed on ice for 1 h and centrifuged at 14,000 g in a centrifuge for 10 min. Then, the proteins were quantified using the bicinchoninic acid (BCA) method (Pierce, Rockford, IL, USA), and the concentration of proteins in the tissues was calculated based on the absorbance and standard of proteins determined using a microplate reader.
Subsequently, the extracted proteins were separated using a 10 % sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) gel and subsequently transferred to a polyvinylidene difluoride membrane (Millipore, Billerica, MA, USA), followed by being sealed with blocking buffer for 1.5 h, and incubated successively with the anti-TNF-α primary antibody (1:1,000), anti-NF-κB primary antibody (1:1,000) and secondary antibodies (1:1,000). After being rinsed, the proteins were reacted with chemiluminescent reagent for 1 min in the dark for complete image development.

Quantitative reverse transcriptionpolymerase chain reaction (qRT-PCR)
First, the total ribonucleic acids (RNAs) were extracted using TRIzol® reagent (Invitrogen; Thermo Fisher Scientific, Inc.) from tissue specimens and reversely transcribed into complementary deoxyribose nucleic acid (cDNA) using the reverse transcription kit. Then, qPCR was performed in a 20 μL reaction system through 40 cycles of reaction at 51 °C for 2 min, pre-degeneration at 96 °C for 10 min, degeneration at 96 °C for 10 sec, and annealing at 60 °C for 30 s. Finally, the relative expression levels of the related messenger RNAs (mRNAs) were calculated by 2 -ΔΔCq method, with glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as an internal reference. The primer sequences are detailed in Table 1.

Enzyme-linked immuno sorbent assay (ELISA)
The preserved tissues were ground and subjected to ELISA according to the instructions of the manufacturer. First, the sample and standard were separately loaded into a plate, added with biotinylated antibody working solution and then with enzyme-conjugated substance working solution. Subsequently, the plate was washed. Finally, the resulting standard and samples were detected at 450 nm in a microplate reader.

TUNEL assay
The tissues embedded in paraffin earlier were first made into 5 μm-thick sections, extended in warm water at 42 °C, mounted, baked and prepared into paraffin-embedded tissue sections. Then, these sections were routinely deparaffinized and hydrated by immersing them successively in xylene solution and gradient ethanol. Subsequently, the sections were added dropwise with TdT reaction solution, reacted in the dark for 1 h, and incubated with deionized water added in drops for 15 min to terminate the reaction.

Statistical analysis
In the present study, statistical analysis was conducted using Statistical Package for Social Sciences (SPSS) 20.0 software (IBM, Armonk, NY, USA), and measurement data are presented as mean ± standard deviation (SD). The t-test were performed for data meeting normal distribution and homogeneity of variance, corrected t-test for those fulfilling normal distribution and heterogeneity of variance, and nonparametric test for those not conforming to normal distribution and homogeneity of variance. Ranked data and enumeration data were subjected to rank sum test and chi-square test, respectively. P < 0.05 indicated statistical significance.

Immunohistochemistry results
Tan colored samples represents positive expression, and there were fewer tissues positive for Caspase-3 in the normal group, but more in the other groups. Based on the results in Figure  1, the mean optical density of Caspase-3-positive tissues in both the model group and the etanercept group was significantly higher than that in the normal group, showing a statistically significant difference (p < 0.05). While the mean optical density in the etanercept group was significantly lower than that in the model group (p < 0.05).

Expression of TNF-α and NF-κB proteins
The normal group showed fewer expressed TNFα and NF-κB proteins, while there were more expressed TNF-α and NF-κB proteins in the other groups. The statistical results showed that the relative protein expression levels of TNF-α and NF-κB in the model and etanercept groups were considerably higher than those in the normal group, showing statistically significant differences (p < 0.05), whereas the etanercept group had significantly lower relative protein expression levels of TNF-α and NF-κB than the model group, (Figure 2) (p < 0.05).

Figure 2:
Expressions of proteins detected via WB. Relative protein expression levels in each group. *p < 0.05 vs. normal group, and # p < 0.05 vs. model group

QPCR results
As shown in Figure 3, both the model group and the etanercept group had significantly higher relative mRNA expression levels of TNF-α and NF-κB than the normal group, and the differences were statistically significant (p < 0.05). Moreover, the relative mRNA expression levels of TNF-α and NF-κB in the etanercept group were significantly lower than those in the model group (p < 0.05).

TNF-α and IL-1β levels
The model and etanercept groups had significantly higher TNF-α and IL-1β content than the normal group (p < 0.05), whereas their content in the etanercept group was significantly lower than that in the model group (p < 0.05) ( Figure 4).

DISCUSSION
As one of the common severe complications of diabetes in clinics, diabetic retinopathy mainly damages retinal microvessels, thereby impairing retinal optic nerves and the relevant tissues. Current studies have unraveled that diabetic retinopathy has an increased morbidity rate annually with the increase in that of diabetes, which is one of the leading causes of visual disturbance and even blindness in diabetic patients [12][13][14]. As the pathogenesis of diabetic retinopathy continues to be researched, it has been gradually realized by researchers that the development of diabetic retinopathy is a longterm complex cascade reaction involving multiple physiological and pathological reactions. Cell apoptosis and inflammation play an important role in the pathogenesis of diabetic retinopathy. In hyperglycemia, pathological changes occurred in retinal microvessels and cause ischemia and hypoxia in the retinal tissues, resulting in tissue damage. After tissue injury, numerous signaling pathways are further activated by ischemiahypoxia injury and release of various inflammatory factors and cytokines, thereby triggering a series of complex pathological reactions in which the TNF-α/NF-κB signaling pathway serves as a crucial player [15,16].
Moreover, ischemia-hypoxia injury and the release of various inflammatory factors and cytokines can further aggravate the expression and release of TNF-α in retinal tissue, thereby activating the TNF-α/NF-κB signaling pathway. As such, a vicious circle is formed, which further exacerbate retinal nerve cell apoptosis and retinal tissue injury, leading to visual disturbance, even blindness and other severe conditions in patients [17,18]. The results of this study further corroborated that the expression of apoptosis effector molecule Caspase-3 rose significantly and aberrantly, and was highly expressed in the retinal tissues of rats with diabetic retinopathy, thereby significantly elevating the cell apoptosis rate. The protein and mRNA expressions of the crucial molecules, TNF-α and NF-κB in the TNFα/NF-κB signaling pathway were abnormally raised, suggesting that the TNF-α/NF-κB signaling pathway is activated and involved in the pathological reaction of diabetic retinopathy. The contents of pro-inflammatory factors TNF-α and IL-1β was significantly increased, implying that severe inflammation occurs in the retinal tissues of rats with diabetic retinopathy.
As a commonly used anti-inflammatory drug, etanercept, is a good inhibitor of TNF-α receptor, which suppresses the pro-inflammatory effect of TNF-α through the competitive binding to the receptor of TNF-α [19]. In the present study, it was found that after treatment with etanercept, the diabetic retinopathy model rats had prominently decreased Caspase-3 expression and cell apoptosis rate as well as improvement in the aberrantly high protein and mRNA expressions of TNF-α and NF-κB, suggesting the repressed TNF-α/NF-κB signaling pathway.
Besides, compared in model group, the contents of TNF-α and IL-1β declined significantly, indicating that etanercept exert favorable antiinflammatory and anti-apoptotic effects.

CONCLUSION
Etanercept inhibits NF-α/NF-κB signaling pathway thereby repressing inflammation and cell apoptosis in diabetic retinopathy rats and exerting a neuroprotective effect. Its potential role in the clinical management of diabetic retinopathy should be therefore be further investigated.