Neuropathic pain (NP) refers to pain caused by damage or illness to the somatosensory system, manifested as spontaneous pain, hyperalgesia, abnormal pain, and sensory abnormalities
[1]. NP is prone to recurrence, and the treatment effect is not satisfactory clinically
[2]. It has been reported that about half of the patients’ pain cannot be fully relieved through treatment
[3], which seriously affects their physical and mental health. Therefore, exploring the pathogenesis of NP and developing effective therapeutic measures are of great significance for improving patients’ health and quality of life.
The activation of microglia is an important mechanism in the pathogenesis of NP
[4-5]. Activated microglia can participate in the occurrence and maintenance of NP by secreting pro-inflammatory factors, upregulating cell surface receptors, and activating related signaling pathways
[4-7]. In addition, activated microglia also exhibit characteristics of migration and phagocytosis, which plays an important role in the onset of many diseases like Alzheimer’s disease
[8], multiple sclerosis
[9], and intracerebral hemorrhage
[10]. A recent study
[11] has found that after nerve injury, activated microglia engulf a large number of perineuronal nets, which wrap around neurons projecting from the spinal dorsal horn to the lateral parabrachial nucleus, thus promoting pain. This suggests that activated microglial cells can trigger pain through phagocytosis. However, the regulatory factors for pain caused by microglial phagocytosis are currently unknown.
The early growth response factor (EGR) belongs to the transcription factor family, and its member EGR2 has been shown to regulate various functions of macrophages, including inflammatory response
[12], tissue repair
[13], phagocytic function
[14], etc. Previous study
[15] has reported that EGR2 is upregulated in microglia, a special type of macrophages, in the cerebral cortex of mice after spared nerve injury. In our sequencing results, we also noted that EGR2 was significantly upregulated in the spinal dorsal horn after NP modelling. However, it remains unclear whether EGR2 causes pain by affecting the function of microglia. Recent literature
[16] has reported that targeted EGR2 therapy can effectively rescue the phagocytic and barrier function of human-induced pluripotent stem cells-derived retinal pigment epithelium, suggesting the possibility of EGR2 regulating cell phagocytic ability. In this study, we found in our early experiments that EGR2-mediated pain is not dependent on the regulation of inflammatory response and the overexpression of EGR2 can significantly affect the expression of CD68, C1q and spleen tyrosine kinase (Syk), which are closely related to the phagocytic function of microglia
[17-19]. Therefore, we speculate that EGR2 may affect the maintenance of NP by regulating the phagocytic function of microglia.
1 Materials and methods
1.1 Ethics statement
The Institutional Ethics Committee of Xiangya Hospital approved all procedures that were performed according to the National Institutes of Health Guide for the Care and Use of Laboratory Animals (approval number: 2020sydw0445).
1.2 Animals
Adult Sprague-Dawley (SD) rats were purchased from Hunan SJA Laboratory Animal Co., Ltd., Changsha, China, weighing 220-250 g. During the experiments, rats were housed in a 12-hour light/dark cycle, a temperature-controlled environment and a room free of pathogens (25-28 °C).
1.3 Rat model of chronic constriction injury
Rats were fully anesthetized with 3% sodium pentobarbital (50 mg/kg) and the chronic constriction injury (CCI) model was constructed based on the protocol described by Bennett, et al
[20]. Then, the left sciatic nerve was exposed and 4 snug ligatures (4-0) were tied to the trunk of the sciatic nerve with the same tightness. Following ligation, the nerve was sent to the intermuscular space and the wound was closed. The sham group had the same procedure but without the ligation of the sciatic nerve.
1.4 Behavioral assessments
CCI caused NP in rats and we measured paw withdrawal mechanical threshold (PWMT) and paw withdrawal thermal latency (PWTL) in the ipsilateral paw from 1 to 14 days after CCI. The PWMT was evaluated by von Frey filaments (North Coast Medical, San Jose, CA), which has been described in our published study
[21]. Briefly, the rats were kept isolated from each other in a chamber with a mesh floor and allowed to adapt to their environment for 30 minutes. Von Frey Filaments were used to stimulate the rat left hind limb, ranging from 0.4 to 15 grams. Finally, up-and-down stimulation was applied and mechanical withdrawal threshold (MWT) was calculated.
The PWTL was evaluated by a thermal pain tester (Tes7370, Ugo Basile, Comerio, Italy). In brief, rats were kept in separate chambers on heat-conductive glass plates and subjected to a 30-minute habituation. The cut-off time for thermal stimulation was set at 30 seconds. Thermal stimulation was applied to the rat left hind limb three times at 5-minute intervals. The PWTL was calculated by taking the average of the three latencies.
1.5 Intraspinal injection
To further explore the function of EGR2, we overexpressed EGR2 by microinjecting AAV-EGR2 in the naive rats (AAV-EGR2 group). Adeno-associated virus (AAV) was used to construct an overexpressed viral vector of EGR2 (rAAV-EF1a-EGR2-mCherry-WPRE-hCG polyA). An rAAV-EF1a-mCherry-WPRE-hCG polyA was used as a negative control (AAV-NC group). Following complete anesthesia of rats, the location of the incision on the T13 vertebral body was identified by touching the most caudal rib. The skin was then dissected to isolate the subcutaneous connective tissue and muscle to fully visualize the T12-L1 spine plate. Then, a minor amount of the vertebral plate at the junction of the lower margin of the T13 vertebral body and the upper margin of the L1 vertebral body was carefully and gently removed. With the use of a stereotactic apparatus, AAV was microinjected into the spinal dorsal horn (SDH) of the rats. The glass pipette needle of the microinjection device was transferred to 0.8 mm on the left side of the midline of the rat spinal cord and 0.6 mm depth. After injection, the procedure was suspended for equilibration (about 1-2 minutes) and then the glass was withdrawn slowly. Eventually, the rat recovered awareness after the wound was closed.
1.6 RNA extraction and real-time quantitative PCR (RT-qPCR)
Affected rat spinal cord horn lumbar segments were collected for RT-qPCR to detect changes in gene expression levels. Total RNA was extracted according to the manufacturer’s instructions (No. RC112-01, Vazyme Biotech Co., Ltd., Nanjing, China). RT-qPCR mixes were made up with the TransStart Tip Green qPCR SuperMix Kit (TransGen Biotech, Beijing, China) and detected using an AB7500 real-time PCR instrument (Applied Biosystems, Foster City, CA, USA). The 2
-ΔΔCt method was used to calculate the relative expression of the target genes compared to
β-actin. The primer sequences in
Table 1 were obtained from Shanghai Biotechnology Company.
1.7 Protein extraction and Western blotting
The spinal cord horn tissues on the operated side were taken and weighed. The protein lysate was added proportionally (RIPA: protease inhibitor: phosphatase inhibitor=100꞉1꞉1). Tissues were broken by ultrasonic tissue homogenizer and the samples were placed on ice for 30 minutes. Then, the samples were centrifuged at 4 ℃ and 12 000 g for 30 minutes, and the supernatant was collected as the extracted protein. The protein concentration of each sample was measured by the BCA protein quantification kit, and after leveling, it was heated at 99 ℃ for 10 minutes for subsequent experiments, while the rest of the protein solution was stored at -20 ℃. Proteins were separated by 12% precast gel electrophoretic separation followed by wet transfer to the polyvinylidene difluoride (PVDF) membrane. The blocking process was carried out for two hours with 5% skimmed milk powder and primary antibodies were then incubated as follows (4 ℃, overnight): Rabbit anti-EGR2 (1꞉500, Novus), mouse anti-GAPDH (1꞉5 000, Abcam). The primary antibody was recovered on the next day and the membrane with TBST was washed for 10 minutes, 3 times. The secondary antibodies corresponding to the primary antibody species (1꞉5 000, goat anti-mouse, goat anti-rabbit; Abcam, USA) were added and were incubated for 2 hours at room temperature. Protein bands were revealed with the ECL Plus blot kit (Advansta K-12045-D50, Shanghai) and imaged using the ChemiDoc XRS System and Image Lab software (Bio-Rad, Universal Hood III, USA).
1.8 Immunofluorescence and microscopy
To further investigate the change of EGR2 after CCI surgery, we performed immunofluorescence staining to clarify the spatial localization of EGR2 in the SDH. After anesthetizing the rats with sodium pentobarbital, we began the perfusion of them using phosphate buffered saline (PBS) and 4% paraformaldehyde. Then, the rat spinal cord was instantly taken out and put into a tube that had been filled with paraformaldehyde. After waiting for 12 hours fixing, the samples were dehydrated with 20% and 30% sucrose water. Next, the lumbar spinal cord was processed for embedding and cut into 10 µm frozen sections with a cryostat (Leica CM1860 Cryostat, Germany). The slices were dried at room temperature, fixed in 4% paraformaldehyde solution for 10 minutes and rinsed 3 times with PBS buffer for 5 minutes each time. A circle around the sample was drawn using an immunohistochemical pen, immunostaining blocking solution was added into the circle, and was incubated for 30 minutes at room temperature. The sections were incubated with primary antibodies as follows (4 ℃, overnight): Rabbit anti-EGR2 (1꞉100; Novus, USA), mouse anti-ionized calcium-binding adapter molecule 1 (IBA-1) (1꞉200; Abcam, USA), mouse anti-glial fibrillary acidic protein (GFAP) (1꞉200; Cell Signaling Technology, USA) and mouse anti-NeuN (1꞉200; Proteintech, USA). On the next day, the sections were washed and they were respectively incubated with the corresponding secondary antibodies (1꞉200; goat anti-mouse and goat anti-rabbit, Abcam, USA) for 1 hour at room temperature. Lastly, staining was visualized under a Leica DM5000B microscope (Leica Biosystems, Wetzlar, Germany).
1.9 HAPI microglia cell line and transfection
The rat microglia cell line HAPI (Huiying Biological Technology CO., LTD, Shanghai, China) was cultured in complete Dulbecco’s Modified Eagle Medium (DMEM) with 10% (v/v) fetal bovine serum, 1% (v/v) penicillin and streptomycin (Thermo Fisher Scientific, Waltham, MA) at 37 ℃ with 5% CO2 in an incubator. The cells were seeded in 6-well plates (2×105 cells/well) overnight and the expression of EGR2 was overexpressed by transfection of AAV-EGR2. AAV-NC was also transfected to generate the control cell lines. The transfection time was controlled at about 48 hours, and the effect was observed under a fluorescence microscope.
1.10 Phagocytosis assay
HAPI cells were cultured into 6-well plates at an appropriate density to ensure that cells were conserved to about 1.2×10
5 cells/well for the assay. Phagocytosis experiments were conducted according to the protocol previously described
[22]. In brief, pre-opsonize the fluorescent latex beads (1 μm, L1030) in 50% FBS and PBS. Preconditioned microspheres were added dropwise to 6-well plates at 50 microspheres per HAPI cell and incubated for 2 hours at 37 ℃. At the end of the incubation, the remaining beads on the cells were gently washed away and fixed with 4% paraformaldehyde (PFA) at room temperature. Finally, the cells were stained with 4',6-diamidino-2-phenylindole (DAPI) (Solarbio, Beijing) and capture the images were captured under an N2Ti2-A inverted microscope.
1.11 Bioinformatics analysis
By analyzing previously published sequencing data of the spinal dorsal horn
[23], we searched for upregulated EGR family members after CCI. It has been reported that the upregulation of EGR2 is related to the activation of phagocytosis signals in alveolar macrophages
[24]. To verify this, we analyzed previously published transcriptomic data
[23]. Then, we conducted RT-qPCR and Western blotting to verify the results of bioinformatics analysis. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses of differentially expressed mRNAs after overexpression of
EGR2 was conducted.
1.12 Statistical analysis
Statistical analyses were conducted using Prism 7 software. For comparisons between 2 groups, two-tailed unpaired Student’s t-tests or Mann-Whitney U tests were employed. For comparisons among multiple groups, one-way or two-way ANOVA was performed, followed by Bonferroni’s multiple comparisons test. Differences were considered significant at a P<0.05.
2 Results
2.1 CCI induced NP and upregulated the expression of EGR2 in the SDH
Behavioral assessments showed that there was no significant difference in baseline between rats in the sham group and the CCI group (
P>0.05). PWMT and PWTL in the CCI group started to decrease from postoperative day 3 to day 14 (all
P<0.05, Figure
1A and
1B). The above results demonstrated that the CCI-induced NP model was constructed successfully.
We identified that solely EGR2 within the EGR family exhibited upregulation after CCI (Figure
1C and
1D). The results of RT-qPCR and Western blotting showed that compared with the sham group, the expression of
EGR2 was significantly increased on 7 days and 14 days after CCI both in mRNA and protein levels (all
P<0.05, Figure
1E and
1F).
2.2 CCI mainly induced EGR2 upregulation in microglia
Immunofluorescence staining showed EGR2 was mainly co-localized in microglia and neurons (
Figure 2). Compared with the sham group, the expression of EGR2 in the microglia was significantly increased after CCI (
P=0.005,
Figure 2B), while there were no statistically significant changes in the expression of EGR2 in neurons and astrocytes (
P>0.05, Figure
2A and
2C). These results indicated that EGR2 may play a role in the microglia during the maintenance of NP.
2.3 Overexpression of EGR2 induced nociceptive hypersensitivity in naive rats
Figure 3A shows the flow chart of the mouse before and after the virus injection. Under the fluorescence microscope, the red fluorescence of the virus can be seen in the SDH (
Figure 3B). Transfection efficiency (mCherry
+IBA-1
+ cells/mCherry
+ cells×100%) of intra-spinal injection AAV-EGR2 in microglia was 13.8%. The results indicated that AAV-EGR2 successfully transfected into microglia. Furthermore, RT-qPCR and Western blotting verified the significant increase of
EGR2 mRNA and protein (all
P<0.05, Figure
3C and
3D). Behavioral assessments revealed that AAV-EGR2 caused a dramatic decrease in the PWMT and PWTL on day 3 after the stabilization of intraspinal injection of the virus, and continued to decrease until day 14 (all
P<0.05, Figure
3E and
3F). Altogether, these results showed that overexpressing
EGR2 induced mechanical and thermal pain hypersensitivity in naive rats.
2.4 CCI induced increase of phagocytosis in microglia
We found that many genes associated with microglia phagocytosis were upregulated after CCI (Figure
4A and
4B), including
CD14,
CD33,
CD68,
C1qa,
C1qb,
C1qc,
P2ry6,
Myd88, and
Syk. Results of RT-qPCR showed that the expression of
CD68,
C1q, and
Syk was significantly increased after CCI surgery (all
P<0.05,
Figure 4C).
To further evaluate the changes in microglia phagocytosis, we used rat microglial HAPI cell line. We treated HAPI with LPS to mimic the activation of microglia in rats with NP. Subsequently, green fluorescent particles were introduced, and the ratio of the total area covered by the fluorescent signal to the overall number of HAPI was quantified to evaluate alterations in microglial phagocytic function following LPS treatment. The results showed that phagocytosis of HAPI was enhanced after LPS stimulation (
P<0.05, Figure
4D and
4E). In conclusion, CCI surgery increased the phagocytosis of microglia.
2.5 Overexpression of EGR2 induced increase of phagocytosis in microglia
To investigate the regulatory effect of EGR2 on the phagocytosis function of microglia after peripheral nerve injury, we harvested the spinal dorsal horn sample from AAV-EGR2 injected rats and examined the expression of genes related to microglia phagocytosis. The results showed that
EGR2,
CD68,
C1q, and
Syk were upregulated in the rat spinal dorsal horn after the injection of overexpression virus (all
P<0.05,
Figure 5A), indicating that microglia phagocytosis was enhanced. We then performed transfection experiments using the rat microglia HAPI cell line. Fluorescence inverted microscopy showed that AAV-EGR2 was successfully transfected into HAPI (
Figure 5B). We used fluorescent latex particles to detect phagocytosis of HAPI. The results showed that HAPI transfected with AAV-EGR2 phagocytosed significantly more fluorescent particles (
Figure 5C).
2.6 Transcriptome sequencing screening of differentially expressed mRNAs after overexpression of EGR2
To identify EGR2 targets, we used RNA-seq in the
EGR2 overexpressed naive rats. The GO analysis of the sequencing results (Figure
6A-
6C) showed that the molecular function terms were mainly protein binding, ion binding, and organic cyclic compound binding. The cellular components were enriched in membrane, intrinsic component of membrane and integral component of membrane. The biological processes were focused on cellular process, response to stimulus and regulation of cellular process. KEGG pathway enrichment analysis (
Figure 6D) showed that the differentially expressed genes were mainly involved in pathways related to NP, including cytokine-cytokine receptor interaction, cell adhesion receptor interaction, and cell adhesion receptor interaction.
We combined our RNA-seq with the CHIP results and found that among these downstream molecules, the expression of CD74 and Lag3 were significantly down-regulated (
Figure 6E). The RT-qPCR results showed that the expression of
Lag3 was significantly suppressed after the overexpression of
EGR2 (
P=0.0053,
Figure 6F). The expression of
CD74 also tended to be down-regulated but had not yet reached statistical significance (
P>0.05,
Figure 6F).
3 Discussion
Microglia have received widespread attention as a potential therapeutic target for NP. In the present study, we found that the expression of EGR2 in the spinal dorsal horn of CCI rats was upregulated from postoperative day 3 and continuing until day 14, accompanied by enhanced phagocytosis of microglia and the development of NP. We explored the important role of EGR2 in regulating the phagocytic function of microglia and modulating the maintenance of NP through overexpression of EGR2 intervention. The results showed that upregulation of EGR2 in the dorsal horn of the spinal cord promoted microglia phagocytosis and was associated with the maintenance of NP.
Microglia activation and proliferation are considered to be hallmarks of many central nervous system (CNS) disorders. Microglia activation responses resulting from peripheral nerve injury include morphological changes, proliferation and upregulation of microglia markers
[25-26]. Activated microglia can contribute to central sensitization and disinhibition through the secretion of pro-inflammatory mediators, leading to NP
[27-28]. Inhibitors of microglia have been found to attenuate chronic pain and peripheral nerve injury does not cause mechanical nociceptive hypersensitivity after specific removal of microglia
[29-30]. In addition to these effects, microglia, as the primary phagocytes of the CNS, continuously monitor changes in the environment of the CNS and respond rapidly to “danger”
[31-32]. The phagocytosis of microglia plays an important role in the maintenance of CNS homeostasis. The surveillance and phagocytosis functions of microglia in the brain of elderly individuals tend to decline
[33], and these abnormal microglia may exacerbate neurodegeneration and cause dysfunction of neurons and oligodendrocytes
[34-35]. The role of microglia phagocytosis in NP has also been gradually discovered by scientists
[36-38]. Makoto Tsuda’s research team
[4] found that CD11c-positive microglia were able to phagocytose medullary phospholipids, and their phagocytosis was enhanced after damage to peripheral nerves, and that this change was associated with spontaneous healing and recurrence of pain in mice. Arkady Khoutorsky’s team
[11] found that microglia phagocytosis of the peripheral nerve network mediated pain behaviors, as evidenced by decreased mechanical pain threshold and thermal hyperalgesia in mice. Zhang’s team
[39] revealed dynamic changes in sciatic nerve injury-induced microglia activation through in vivo optical microscopic imaging and found that the morphology of microglia shifted from branching in the resting state to phagocytosis in the activated state from the 2nd day to the 14th day after injury.These studies suggest that the phagocytosis of microglia is enhanced after peripheral nerve injury, which is closely related to the maintenance of NP.
In this study, we extracted tissues from the lumbar expanded segment of the spinal dorsal horn for transcriptome sequencing on day 7 after rat CCI. The results showed that a large number of genes related to microglia phagocytosis were upregulated. After further validation by RT-qPCR, we found that
CD68,
C1q, and
Syk were up-regulated, which indicated that the phagocytosis of microglia in the dorsal horn of the spinal cord of the rat was enhanced after CCI. CD68 is a lysosomal protein marker that is expressed at high levels on the surface of macrophages and activated microglia, and at low levels on the surface of resting microglia
[17, 40]. C1q is a complement system initiator that labels synapses in the brain and acts as a signaling molecule for “Eat Me”, mediating the pruning of labeled synapses by microglia in the brain
[18]. Activation of the Syk signaling pathway promotes phagocytosis of Aβ proteins as well as myelin debris by microglia
[41]. To visualize the phagocytosis function of microglia more intuitively, we stimulated the rat HAPI cell line with LPS to mimic the state of microglia in NP state, and found that LPS-stimulated HAPI was able to phagocytose more fluorescent latex particles, which further proved the change in phagocytosis capacity of microglia. However, the HAPI microglia cell line we used may not accurately reflect the specific changes in microglia in rats, so microglia extracted from the spinal dorsal horn of rats will be validated in the future to ensure the accuracy of the results.
The EGR family, which includes 4 family members, EGR1, EGR2, EGR3 and EGR4, plays an important role in transcriptional regulation. However, the role played by EGR in pain has been less studied in the published literature
[42]. We found that only EGR2 was upregulated on the 7th day after CCI. Subsequently, we performed immunofluorescence co-labeling of EGR2 and neurons, microglia and astrocytes within the dorsal horn of the spinal cord, and the results showed a significant increase in EGR2 in microglia after CCI, whereas no significant change in EGR2 was seen in astrocytes and neurons. Recent study
[16] has found that EGR2 can regulate phagocytic phenotype and that targeting EGR2 treatment induces activation of phagocytosis in phagocytes. Therefore, we hypothesized that after peripheral nerve injury, EGR2 may contribute to NP by enhancing microglia phagocytosis. To verify this idea, we overexpressed
EGR2 in the dorsal horn of the spinal cord by intrathecal injection of AAV-EGR2 and found that microglia phagocytosis-related genes were up-regulated, and at the cellular level, we verified that HAPI was able to phagocytose more fluorescent latex particles after overexpression of EGR2. However, we did not use EGR2 inhibitors or viral knockdown of EGR2 for reverse validation, which is one of the shortcomings of this experiment. In addition, the role of EGR2 in other types of cells in the spinal dorsal horn cannot yet be ruled out. Future studies could clarify the regulatory role of EGR2 on microglia through targeted intervention of EGR2 in microglia in the dorsal horn of the spinal cord, and could also delve into the specific mechanisms of the phagocytosis process, including the regulation of upstream and downstream target genes as well as the types of phagocytosed substances. These further studies will help to better understand the role of EGR2 in pain perception and provide new targets and strategies for pain treatment.
Through gene sequencing in conjunction with the chromatin immunoprecipitation followed by sequencing (Chip-seq) database, we conducted a preliminary exploration of the downstream target genes of
EGR2. Our GO analysis of the sequencing results showed that the differentially expressed genes in the spinal dorsal horn after overexpression of EGR2 were mainly involved in protein binding in terms of molecular function, in cellular components, and in cellular processes. Enrichment analysis of KEGG signaling pathways showed that the differentially expressed genes were mainly involved in pathways related to NP, including cytokine-cytokine receptor interactions, cell adhesion molecules, phagosomes, NOD-like receptor signaling pathways, and cellular adhesion molecules. interactions, cell adhesion molecules, phagosomes, and NOD-like receptor signaling pathways. These results suggest that EGR2-mediated phagocytosis may be associated with a receptor on the microglia membrane. Further combining with the published Chip-seq database, we predicted that Lag3 might be a downstream molecule of EGR2. The protein Lag3 belongs to the immunosuppressive receptor, a type 1 transmembrane protein, and its expression in microglia is associated with depression
[43-44], and
Science has reported that Lag3 has the function of regulating the phagocytosis of α-synaptic nuclear proteins by microglia in Parkinson’s disease
[45]. However, no studies have yet been able to elucidate the relationship between Lag3 and EGR2-regulated phagocytosis and the association between NP, which is what we need to confirm next. Considering the accuracy of the disease model and cellular localization, in the future we will use chip technology to detect the molecules that may bind EGR2 in the spinal dorsal horn microglia extracted from the CCI model and refine the regulatory relationship between EGR2 and Lag3.
Several critical issues warrant further exploration in future research. Firstly, the intricate molecular mechanisms underlying the function of EGR2 and its interactions within various signaling pathways in microglia remain to be elucidated. Additionally, the therapeutic potential of EGR2 warrants validation through comprehensive animal models and advanced preclinical trials. For instance, pharmacological investigations involving EGR2 inhibitors or specific agonists could be pursued to assess their efficacy in treating NP. Moreover, extending these studies to include human samples is crucial for delineating the role of EGR2 in human NP, thereby laying a foundation for novel treatment paradigms. Through these focused endeavors, we aim to develop innovative therapies that offer enhanced relief for patients suffering from NP.
In summary, EGR2 can be involved in the maintenance of NP by enhancing the phagocytosis of spinal dorsal horn microglia, and this study provides a new theoretical basis for the mechanism of NP maintenance.
the National Natural Science Foundation of China(82071249)
the National Natural Science Foundation of China(81771207)
©Journal of Central South University (Medical Science). All rights reserved.
,
京公网安备11010202008535号
PDF
(5643KB)
