Downregulation of long noncoding RNA SNHG7 protects against inflammation and apoptosis in Parkinson's disease model by targeting the miR‐425‐5p/TRAF5/NF‐κB axis

Accumulated evidence has manifested that long noncoding RNA (lncRNA) is involved in the progress of Parkinson's disease (PD). SNHG7, a novel lncRNA, has been found to be involved in tumorigenesis. However, SNHG7 expression and its functional effects on PD remain uncharted. Rotenone (Rot) was adopted to construct PD models in Sprague‐Dawley (SD) rats and SH‐SY5Y cells, respectively. The expression levels of caspase 3, tyrosine hydroxylase (TH), ionized calcium‐binding adapter molecule 1 (Iba1) in SD rat striatum were measured via immunohistochemistry and western blot. Additionally, the expressions of inflammatory cytokines (interleukin 1β [IL‐1β], IL‐6, tumor necrosis factor α) and oxidative stress factors (malondialdehyde, superoxide dismutase, and glutathione peroxidase) in the brain tissues were examined using real‐time polymerase chain reaction and enzyme‐linked immunosorbent assay, respectively. Moreover, the protein levels of tumor necrosis factor receptor‐associated factor (TRAF5), I‐κB, nuclear factor‐κB (NF‐κB), HO‐1, Nrf2 were detected via western blot. Bioinformatics was applied to predict the targeting relationship between SNHG7, miR‐425‐5p, and TRAF5. Dual‐luciferase activity assay and RNA immunoprecipitation assays were conducted to verify their interactions. In comparison to healthy donors, SNHG7 was found upregulated while miR‐425‐5p expression was downregulated in PD patients. Functional experiments confirmed that SNHG7 downregulation or miR‐425‐5p overexpression attenuated neuronal apoptosis in the Rot‐mediated PD model, TH‐positive cell loss, and microglial activation by mitigating inflammation and oxidative stress. Mechanistically, SNHG7 served as a competitive endogenous RNA by sponging miR‐425‐5p and promoted TRAF5 mediated inflammation and oxidative stress. Inhibition of SNHG7 ameliorated neuronal apoptosis in PD through relieving miR‐425‐5p/TRAF5/NF‐κB signaling pathway modulated inflammation and oxidative stress, and similar results were observed in the Rot‐mediated rat model of PD.


| INTRODUCTION
As a typical neurodegenerative disorder disease, Parkinson's disease (PD) inflicts serious damage to the patients' quality of life. Clinically, PD patients usually present with symptoms like resting tremors and muscle rigidity. Pathologically, this disease is characterized by the degeneration or progressive loss of dopaminergic neurons in the substantia nigra (SN) pars compacta and striatum of the mid-brain, together with the presence of Lewy bodies (mainly composed of α-synuclein). [1] A series of studies have confirmed that genetic susceptibility, oxidative stress, drug toxicity, and environmental triggers are strongly implicated in PD's pathogenesis. [2,3] In recent years, increasing research has focused on the pathogenic mechanism underlying PD. A great many noncoding RNAs (such as long noncoding RNA [lncRNA] and microRNA [miRNA]) that regulate these genes have attracted wider attention. [4,5] Thus, an improved understanding of lncRNAs in PD helps to provide new insights for finding early diagnostic indicators and effective therapeutic targets of PD.
TRAF5 is a member of the tumor necrosis factor receptorassociated factor (TRAF) family, which has been discovered recently.
Like other TRAF family members, TRAF5 also has a ring finger domain, and it displays E3 ubiquitin ligase viability. Existing studies have validated that TRAF5, as a cytoplasmic adapter, can activate the NF-κB signaling pathway through its receptors, and then participate in mediating inflammation, neuronal apoptosis, glial cell activation, and other processes in the nervous system. [18,19] At present, accumulating studies have highlighted that TRAF5 is directly involved in disease progression as a targeted molecule of miRNAs. For example, in gastric cancer cells, miR-135a regulates the NF-κB pathway by targeting TRAF5. MiR-135a overexpression suppresses the NF-κB pathway, thereby inhibiting gastric cancer cell migration. [20] In the present study, we found that SNHG7 was significantly overexpressed in the serum of PD patients and also positively correlated with the levels of proinflammatory cytokines (interleukin 1β [IL-1β], IL-6, tumor necrosis factor α [TNF-α]). Additionally, we conducted bioinformatics analysis through Starbase (http://starbase. sysu.edu.cn) and found that miR-424-5p shared the binding sites with SNHG7 and TRAF5. Interestingly, miR-424-5p was markedly downregulated in the serum of PD patients and had negative correlations with IL-1β, IL-6, and TNF-α levels. Therefore, we supposed there is a regulatory network of SNHG7-miR-424-5p-TRAF5 in PD development.

| Clinical samples
A total of 56 participants were recruited from XiangYang Center Hospital: 36 surgically treated PD patients and 20 healthy donors who had undergone related examinations as required. All the subjects provided written informed consent. The diagnosis of PD was performed by two experienced clinicians, based on the British PD clinical diagnostic criteria for diagnosis. [21] Exclusion criteria: (1) PD induced by chemical drugs, trauma, and cerebrovascular diseases; (2) PD patients with severe heart, kidney, liver, blood system diseases and infectious or inflammatory diseases; (3) patients with other nervous system diseases were excluded. Moreover, this study garnered approval from the Ethics Committee of XiangYang Center Hospital. All the plasma samples (fasting elbow vein blood, 10 ml) were obtained from the recruiters recruited patients and controls in the morning.
After that, the samples were subjected to experimental examination after centrifugal separation. A total of 45 SD rats were randomly divided into three groups, namely, the Sham group (n = 15), the Rot + LV-si-NC group (n = 15), and the Rot + LV-si-SNHG7 group (n = 15). For the Sham group: sunflower oil was subcutaneously injected into the back of SD rats for 42 days (1 ml/kg/day). For the Rot + LV-si-NC group and Rot + LV-si-SNHG7 group: rotenone (MedChemExpress, HY-B1756) was subcutaneously injected at the rats' back for 42 days consecutively (1.5 mg/kg/day). The small inference RNA (siRNA) lentivirus vector targeting SNHG7 (LV-si-SNHG7) or LV-si-NC (GeneChem Co. Ltd.) were inserted into the linearized vector GV115 to construct the siRNA recombinant lentiviral vector. Two days before Rot treatment, the rats were anesthetized via isoflurane (2%) in oxygen and nitrous oxide and then positioned on a stereotactic frame (Stoelting). The skull surface was exposed with a hole drilled to locate the needle. Hamilton syringes (5 μl, 33-gauge needle) were applied here to inject the rats with 1 μl/side of recombinant LV-si-SNHG7 or LV-si-NC into the dorsal hippocampus (anterior, −5.5; lateral, +1.6; dorsoventral, −7.5 from bregma) (0.2 μl/min, 10 min). Behavioral evaluation was carried out on days 0, 14, and 42. Rats were sacrificed by decapitation following the behavioral evaluation on the 42nd day.

| Animals and treatment
After anesthetization and perfusion, the rat brains were collected, washed, and frozen in liquid nitrogen. They were stored at −80°C for further use. On the bevel experiment: To begin with, the rats were placed on the bevel rough surface at an angle of 60°. Then, the duration during which the rats of each group stay on the bevel was recorded respectively. When the rat stayed on the slope for over 3 min, it was recorded as 180 s. Three experiments were performed on each animal.

| Immunohistochemical staining
On the 42nd day, the rats in different groups were sacrificed, then the brains were collected, fixed in 4% paraformaldehyde, embedded in paraffin, serially sectioned (4 μm). Next, the brain slices were put in a methanol solution (0.3% H 2 O 2 ) and got incubated (37°C, 15 min) to inactivate the endogenous peroxidase. After that, the sections were

| Immunofluorescence
On the 42nd day, the rats in different groups were sacrificed, then the brains were collected and were frozen quickly in a −20°C re-

| Western blot
When rat brain (SN area) or cells were collected, we removed the medium and added protein lysate (Roche) to lyse the cells or tissues.
Then the total proteins were isolated. Next, 50 μg total protein was added to 12% polyacrylamide gel for electrophoresis (100 V, 2 h).
Then, the protein samples were electroblotted onto polyvinylidene fluoride membranes. After blocking with 5% skimmed milk powder

| Enzyme-linked immunosorbent assay (ELISA)
To begin with, the test tube was filled with 10 ml of venous blood.
Standing for 1 h at room temperature, the sample was subject to centrifugation (340 g, 5-10 min). Next, the upper serum was collected and put in the refrigerator (−20°C) for overall detection. Following that, the SD rat brains (SN area) were weighed, shredded, and added with 12 volumes of normal saline. The tissues in normal saline were then transported to a homogenizer for thorough mixing and crushing to deliver a tissue homogenate. Furthermore, the brain tissue homogenate was centrifuged (680g, 15 min, 4°C), and then the supernatant was collected for detection. Then, the supernatant of Rot-treated SNHG7 cells was collected based on the group. After a 10 min centrifugation (1000g, 4°C), the supernatant was collected as required.
All experimental steps conformed to the instructions provided by the ELISA kit in determining the content of IL-1β, IL-6, TNF-α, and lactate dehydrogenase (LDH) release. The test kits were obtained from Nanjing Jiancheng Bioengineering Institute (NanJing JianCheng Bioengineering Institute, Nanjing, China).

| Luciferase reporter assay
The DNA sequences of lncRNA SNHG7 and TRAF5 were amplified.
According to the predicted binding sites from the online database

| RNA immunoprecipitation (RIP) assay
To determine the relations between SNHG7 and miR-425-5p, miR-425-5p, and TRAF5, Magna-RNA combined with the binding protein immunoprecipitation kit (Millipore) was used to conduct the RIP test.
The SH-SY5Y cells (transfected with miR-424-5p mimics or miR-NC) at 80% confluence were collected and lysed in a complete RIP lysate buffer. Then, the cell lines were co-immunized with RIP buffer containing magnetic beads with antiargonaute 2 (Ago2) antibody (Millipore) or negative control healthy rat IgG (Millipore). The immunoprecipitated RNA was isolated and the relative enrichment of SNHG7 and TRAF5 in the lysates was examined by qRT-PCR.

| Statistical analysis
Data analysis was carried out via SPSS software (version 20.0). The data were displayed as mean ± SD. Student's t test, along with χ2 was employed to access the statistical differences between the two groups. One-way analysis of variance followed by Tukey's test is used for multiple group comparisons. The correlation relationship was determined by Pearson's correlation test. p < 0.05 was regarded as statistically significant.

| SNHG7 and miR-425-5p expression characteristics in PD patients
Aiming at determining the SNHG7 and miR-425-5p expressions in PD patients, qRT-PCR was employed to detect the expressions in 36 sera of PD patients and 20 healthy donors. The results indicated that compared with the healthy donors, SNHG7 was upregulated in PD patients' serum, while miR-425-5p was downregulated ( Figure 1A,B).
Pearson's correlation analysis exhibited that SNHG7 in PD patient serum was negatively correlated with miR-425-5p expression ( Figure 1C). Furthermore, the IL-1β, IL-6, and TNF-α expressions in the serum were detected via ELISA. The results indicated that the levels of IL-1β, IL-6, and TNF-α in PD patient serum were significantly increased compared with those in the healthy donors' serum ( Figure 1D). By conducting Pearson's correlation analysis, we found that SNHG7 was positively correlated with IL-1β, IL-6, TNF-α in PD patients' serum, while miR-425-5p revealed a negative correlation with IL-1β, IL-6, TNF-α in PD patients' serum ( Figure 1E,F). ZHANG ET AL.

| 5 of 14
Therefore, the results suggested that both SNHG7 and miR-425-5p play a role in PD.

| SNHG7 inhibition attenuated neuronal apoptosis in PD model in vivo and in vitro
To better understand the biofunctions of SNHG7 on PD, we used Rot to construct a PD model in SD rats. Meanwhile, LV-si-SNHG7 was used to establish a rat model with downregulation of SNHG7.
The spontaneous locomotor activity test, carousel experiment, and bevel experiment were conducted to comprehensively assess the motor skills of rats. As Figure 2A shows, the rats in the Rot + LV-si-NC group showed notable behavioral damage in comparison to the rats in the Sham group. However, the Rot + LVsi-SNHG7 group showed markedly attenuated behavioral deficits than those in the Rot + LV-si-NC group (Figure 2A). The TH-positive cells in SN were detected by immunohistochemistry, and it was found that TH-positive cells were prominently reduced in rats of the Rot + Lv-si-NC group (compared with that in the Sham group), while knocking down SNHG7 reversed Rot's effect on TH positive cell number in SD rats (vs. Rot + Lv-si-NC group, Figure 2B). Additionally, a western blot was applied to detect TH relative expression in the SN area. The result indicated that TH relative expression was enhanced in the Rot + LV-si-SNHG7 group compared to that in the Rot + LV-si-NC group ( Figure 2C F I G U R E 2 SNHG7 inhibition attenuated neuronal apoptosis in PD models in vivo and in vitro. Then, the rat PD model was induced by Rotenone and Lv-si-SNHG7 was used for knocking down SNHG7 in the rat brain. (A) The spontaneous locomotor activity test, carousel experiment, and bevel experiment were used to comprehensively evaluate the rat's motor skills (n = 5). (B) Immunohistochemistry was utilized to detect the number of TH positive cells in the SN area of rats (n = 5). (C) Western blot was employed to detect the TH relative expression in the rat SN region (n = 5). (D) Caspase 3 staining was used to detect the apoptotic neuron number in rats (n = 5). SH-SY5Y cells with low SNHG7 expression were constructed and were then treated with rotenone (500 nM) for 24 h. (E) and qRT-PCR was used to detect the SNHG7 relative expression (n = 3). (F) MTT assay was used for detecting neuronal proliferation (n = 3). (G) ELISA assay was used to detect LDH levels of cells in each group (n = 3). (H) The relative expression of apoptotic proteins: Bax, Bcl2, and caspase 3 was detected by western blot (n = 3). && p <0.01, &&& p < 0.001 (vs Sham); *p < 0.05, **p < 0.01, ***p < 0.001. ELISA, enzyme-linked immunosorbent assay; LDH, lactate dehydrogenase; MTT, 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide; PD, Parkinson's disease; qRT-PCR, quantitative real-time polymerase chain reaction; SN, substantia nigra; TH, tyrosine hydroxylase

| SNHG7 inhibition attenuated inflammation and oxidative stress in the PD model
In the PD rat model, we further validated the expressions of inflammation and oxidative stress factors. First, immunofluorescence was performed to test Iba1-labeled microglia. The findings suggested that microglial activation in the PD model was remarkably stimulated, while SNHG7 knockdown notably reduced microglial activation (vs. Rot + Lv-si-NC group, Figure 3A,B). Moreover, the proinflammatory cytokines in brain was measured by ELISA. The results showed that the levels of IL-1β, IL-6, and TNF-α were prominently overexpressed in the Rot+LV-si-NC group (compared with those in the Sham group), while SNHG7 inhibition attenuated the inflammatory responses in the PD model (vs. Rot + Lv-si-NC group, Figure 4A). Next, the expressions of oxidative stress factors including MDA, SOD, and GSH-PX in the brain of rats were also determined. We discovered that the Rot + LV-si-SNHG7 group had attenuated MDA level but enhanced SOD and GSH-PX level in the brain when compared with the Rot + LV-si-NC group, suggesting that SNHG7 downregulation inhibited oxidative stress response ( Figure 4B). Then, we examined the profiles of TRAF5, I-κB, p-NF-κB, Nrf2, HO-1 using western blot. As the results showed, the expressions of TRAF5 and p-NF-κB in the Rot +LV-si-SNHG7 group were remarkably downregulated compared to those in the Rot + LV-si-NC group, while the expressions of I-κB, Nrf2 and HO-1 were significantly elevated in the Rot + LV-si-SNHG7 group (vs. Rot + LV-si-NC group, Figure 4C). Furthermore, we investigated the regulatory role of SNHG7 on the inflammation and oxidative stress reactions on HMC3 cells induced by LPS. ELISA results showed that in the LPS+si-SNHG7 group, the inflammatory cytokines (including IL-1β, IL-6, and TNF-α) were significantly reduced when compared with those in the LPS + si-NC group ( Figure 4D). By detecting the oxidative stress factors' expression, we found that in LPS-treated HMC3 cells, the level of MDA significantly increased, while SOD and GSH-PX were decreased (compared with the control group, Figure 4E). After downregulating SNHG7, the MDA level was inhibited, but the SOD and GSH-PX expressions were enhanced (compared with the LPS + si-NC group, Figure 4E). Additionally, the expressions of TRAF5 and p-NF-κB in the LPS + si-SNHG7 group were significantly downregulated compared with those in the LPS + si-NC group, while the expressions of I-κB, Nrf2, and HO-1 were significantly increased ( Figure 4F). The above results indicated that SNHG7 inhibition could relieve microglia-mediated inflammation and oxidative stress via regulating the TRAF5/I-κB/NF-κB pathway.

| SNHG7 targeted miR-425-5p
To explore the downstream molecular mechanism of SNHG7, we analyzed the SNHG7 candidate targets through the StarBase database (http://starbase.sysu.edu.cn). Interestingly, miR-425-5p had binding sites with SNHG7 ( Figure 5A). The dual-luciferase reporter assay and RIP assay were applied to determine the interactions between SNHG7 and miR-425-5p. As the results show, miR-425-5p mimics reduced the luciferase activity of HMC3 cells transfected with SNHG7 (wild-type) but exerted a minimal effect on the luciferase activity of mutant SNHG7. Moreover, the enrichment of SNHG7 in the anti-Ago2 antibody precipitation complex was remarkably F I G U R E 3 SNHG7 inhibition attenuated Iba1 labeled microglial activation in the rat PD models. (A). Immunofluorescence was used to detect Iba1-labeled microglia in the SN area (n = 5). (B) Western blot was conducted to detect Iba1 expression in the SN area (n = 5). ***p < 0.001. Iba1, ionized calcium-binding adapter molecule 1; PD, Parkinson's disease; SN, substantia nigra increased following miR-424-5p mimics transfection. In addition, the miR-425-5p levels in the PD model (both in vivo and in vitro) were significantly declined compared with that in the Sham or control group, while downregulation of SNHG7 promoted miR-425-5p expression (compared with that in the Rot + si-NC or Rot + Lv-si-NC group, Figure 5D,E). The above results showed that SNHG7 may function as a competitive endogenous RNA (ceRNA) by sponging miR-425-5p.

| SNHG7 overexpression inhibited miR-425-5p mediated neuroprotective effects
To further investigate the SNHG7/miR-425-5p axis in PD, SH-SY5Y cells were transfected with miR-425-5p mimics and/or SNHG7 overexpressing plasmids. The results showed overexpression of miR-425-5p inhibited SNHG7 level (compared with that in the Rot + miR-NC group, Figure 6A). At the same time, SNHG7 overexpression suppressed the miR-425-5p level (compared with that in the Rot + miR-524-5p group, Figure 6B). Next, the cell viability of SH-SY5Y cells was tested by MTT assay, which showed that the cell viability in the Rot + miR-425-5p group was apparently enhanced (compared with that in the Rot + miR-NC group), while supplementation of the SNHG7 overexpressing plasmids resulted in a remarkable decrease in cell viability (compared with that in the Rot + miR-425-5p group, Figure 6C). Additionally, an ELISA assay was performed to ascertain the LDH level. We found that miR-425-5p mimics obviously reduced LDH expression. Followed with the SNHG7 overexpression, the LDH level was significantly promoted (compared with that in the Rot + miR-425-5p group, Figure 6D).
Besides this, the expressions of apoptosis-related proteins (including Bax, Bcl2, and caspase 3) were examined via western blot. It was found that miR-425-5p decreased the Rot-induced Bax and caspase 3 overexpression and Bcl2 downregulation (compared with those in the Rot + miR-NC group). Nevertheless, the compensation of SNHG7 reversed those effects (compared with that in the Rot + miR-425-5p group, Figure 6E). Hence, SNHG7 promoted rotenone-mediated neuronal damage via inhibiting miR-425-5p.

| SNHG7 promoted TRAF5/NF-κB pathway activation by competitively binding miR-425-5p
To further verify the underlying mechanism of the SNHG7/miR-425-5p axis in regulating microglial activation, the relative expression of inflammatory factors and oxidative stress factors were detected by ELISA. It turned out that miR-425-5p overexpression inhibited the inflammatory cytokines and oxidative stress induced by LPS in HMC3 F I G U R E 4 SNHG7 inhibition mitigated inflammation and oxidative stress in PD models. The rat PD model was induced by Rotenone and Lv-si-SNHG7 was used for knocking down SNHG7 in the rat brain. (A) ELISA was utilized to detect the IL-1β, IL-6, and TNF-α expressions in the SN area of SD rats (n = 5). (B) The oxidative stress-related factors (including MDA, SOD, and GSH-PX) in the SN area of SD rats were detected using the related detection kits (n = 5). (C) Western blot was used to detect the relative expression of TRAF5, I-κB, NF-κB, Nrf2, HO-1 in the SN area of SD rats (n = 5). HMC3 cells were treated with LPS (10 μg/ml) and transfected with si-NC or si-SNHG7. (D) The expressions of inflammatory factors (including IL-1β, IL-6, and TNF-α) in the culture medium of HMC3 cells were detected using ELISA (n = 3). (E) The MDA, SOD, and GSH-PX detection kits were used to detect the expression of oxidative stress factors (including MDA, SOD, and GSH-PX) in HMC3 cells (n = 3). (F) Western blot was carried out to detect the relative expression of TRAF5, I-κB, NF-κB, Nrf2, HO-1 in HMC3 cells (n = 3). *p < 0.05, **p < 0.01, ***p < 0.001. ELISA, enzyme-linked immunosorbent assay; GSH-PX, glutathione peroxidase; IL, interleukin; MDA, malondialdehyde; PD, Parkinson's disease; SOD, superoxide dismutase; TNF-α, tumor necrosis factor α cells, which were inhibited by SNHG7 (compared with that in the Rot + miR-425-5p group, Figure 7A,B). Next, the result of western blot showed that the TRAF5 and p-NF-κB expressions in the LPS + miR-425-5p group declined, while the I-κB, Nrf2, and HO-1 expressions increased in comparison to the LPS+miR-NC group.
However, SNHG7 overexpression reversed the effects induced by miR-425-5p mimics (compared with that in the LPS + miR-425-5p group, Figure 7C,D). The above statistics indicated that SNHG7 promoted the activation of the TRAF5/NF-κB pathway in microglia through competitively binding with miR-425-5p.

| DISCUSSION
In the present study, we explored a novel network of the SNHG7-miR-425-5p-TRAF5 axis in PD progression. Our data showed that downregulation of SNHG7 markedly repressed the inflammation and oxidative stress through upregulating miR-425-5p, thus inhibiting the TRAF5/NF-κB pathway.
With the pathological mechanisms of PD being ceaselessly explored, PD patients are found to manifest reduced TH content, activated microglia, and decreased LDH. [22,23] In addition, the activation  [24,25] Thus, inhibiting the inflammation and oxidative stress of PD might bring out direct therapeutic effects for clinical treatment. For instance, niacin, [26] epalrestat, [27] and bruceine D [28] could ameliorate PD progression via modulating inflammation and oxidative stress. In our study, we also detected the levels of proinflammatory cytokines (including IL-1β, TNF-α, and IL-6) in the serum of PD patients. It was found that the expressions of IL-1β, TNF-α, and IL-6 were overexpressed in PD patients compared with those in healthy donors, and the result was consistent with previous studies. [29,30] Multiple evidence has indicated that the aberrantly expressed lncRNAs are involved in the development of PD, a positive sign in regulating the neural development, oxidative stress, apoptosis, and neuroinflammation of PD. [31][32][33] For example, lncRNA-UCA1 is highly expressed in MPP + -induced PD rat brain tissues and SH-SY5Y cells.
LncRNA-UCA1 leads to neuron damage and apoptosis by upregulating ɑ-synuclein (SNCA). [34] Furthermore, other lncRNAs such as MALAT1, [35] and SNHG14 [36] are all overexpressed in PD and promote its progression. In this study, our data suggested that SNHG7 is upregulated in PD patients and PD models (both in vivo and in vitro).
Interestingly, the high level of SNHG7 had a positive relationship with TNF-α and IL-6 in the serum of PD patients. Functionally, knockdown of SNHG7 expression not only alleviated rotenone-mediated neuronal damage and microglial activation but also significantly mitigated the levels of inflammatory cytokines and oxidative stress. Therefore, SNHG7 is considered to be a biomarker of PD and also play a role in PD progression via modulating neuronal damage and microglial activation.
Moreover, upregulation of miR-425-5p showed mighty effects against rotenone-induced neuronal damage and microglial inflammations. Therefore, miR-425-5p serves as a promising mediator and therapeutic target for PD.
The lncRNA-miRNA interaction has attracted accumulated attention, particularly in tumor development and inflammation. [42,43] Interestingly, by conducting bioinformatics analysis, we found that miR-425-5p contains the binding sites with SNHG7. Our gain-and loss-of assays showed that downregulating SNHG7 enhanced miR-425-5p level, while overexpressing SNHG7 led to miR-425-5p inhibition as well significantly reduced the neuroprotective and anti-inflammatory effects mediated by miR-425-5p. Thereby, we confirmed that the overexpressed SNHG7 could promote neuron damage and microglial activation at least through sponging miR-425-5p.
The TRAF family has been proved to participate in the progress of pathology in PD. For example, triggering receptor expressed on myeloid cells-2 (TREM2), a newly identified receptor expressed on microglia, remarkably reduced MPTP-induced dopaminergic neurodegeneration and neuroinflammation in PD by downregulating the TRAF6/TLR4-mediated activation of the MAPK and NF-κB signaling pathways. [44] In another study, telmisartan has neuroprotective effects in the rotenone rat model of PD through Inhibition of endoplasmic reticulum stress-activated IRE1α-TRAF2-caspase-12 apoptotic pathway. [45] TRAF5 is also a vital member of the TRAF family. Interestingly, it has been found that TRAF5 is overexpressed in NeuN-and GFAP-labeled cells in a rat acute spinal cord injury model. [46] Moreover, TRAF5 is a mediator of several CNS diseases such as ischemic brain infarction, [47] Huntington's disease, [48] and HIV-1 gp120-induced apoptosis in human neurons. [49] In our study, we found that TRAF5 was a target of miR-425-5p by bioinformatics analysis.
Downregulating SNHG7 and overexpressing miR-425-5p both inhibited TRAF5 expression in the PD model in vivo and in vitro. Moreover, as an upstream molecule of NF-κB, TRAF5 promotes NF-κB activation, a feature widely recognized by scholars. Accumulated evidence has shown that TRAF5 mediates the activation of the NF-κB signaling pathway, and engages in cell apoptosis, the occurrence of inflammation, and immune regulation. [50][51][52] In this study, we found that miR-425-5p inhibited TRAF5, which, in turn, constrained the activation of the I-κB/NF-κB pathway, thereby reducing inflammation and oxidative stress, and weakening neuronal apoptosis.

| CONCLUSION
In summary, this study indicated that SNHG7 was upregulated in PD patients; SNHG7 functioned as a ceRNA by sponging miR-425-5p and promoted TRAF5 mediated inflammation and oxidative stress. SNHG7 low expression or miR-425-5p overexpression profoundly attenuated neuronal apoptosis, inflammation, and oxidative stress in PD. Taken together, this study established a novel regulatory axis of SNHG7/miR-425-5p/TRAF5/NF-κB axis in PD. However, more studies are needed to verify this network in the progression of PD in the future.

CONFLICT OF INTERESTS
The authors declare that there are no conflict of interests.

ETHICS STATEMENT
Our study was approved by the Ethics Review Board of XiangYang Center Hospital.

AUTHOR CONTRIBUTIONS
Operating the experiment and writing-original draft and supervision: Keqi Hu. Formal analysis and visualization: Handong Liu. All authors read and approved the final manuscript.

DATA AVAILABILITY STATEMENT
The data sets used and analyzed during the current study are available from the corresponding authors on reasonable request.