Exploring the pathogenesis of diabetic encephalopathy based on the ceRNA regulatory network of lncRNA-miRNA-mRNA

doi:10.21203/rs.3.rs-2454975/v1

Download PDF

Article

Exploring the pathogenesis of diabetic encephalopathy based on the ceRNA regulatory network of lncRNA-miRNA-mRNA

https://doi.org/10.21203/rs.3.rs-2454975/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Diabetes is a chronic systemic metabolic disease. Diabetic encephalopathy is one of many complications. Its pathogenesis is related to abnormal transcriptional regulation. Competitive endogenous RNA regulatory networks are of great significance in diabetic complications. However, comprehensive However, few studies have been reported to analyze the ceRNA network of lncRNA-miRNA-mRNA on the pathogenesis of diabetic encephalopathy.Firstly, GK rats and Wistar rats were screened by behavior, electrophysiology and morphology to confirm the occurrence of diabetic encephalopathy, and then high-throughput sequencing of hippocampus was performed to construct gene expression matrix, and compared with diabetic patients and normal rats obtained from GEO database The human pancreatic islet tissue gene expression matrix was differentially analyzed to construct an interactive regulatory network. Finally, the expression of lncRNA-NONRATT007456.2, lncRNA-NONRATT020546.2, miR-5132-5p and MRC1 in the network was verified by real-time quantitative PCR. Western Blot further confirmed MRC1 protein expression.The results of novel object recognition behavioral experiments, electrophysiological experiments and Golgi staining showed that the diabetic encephalopathy model based on GK-/- diabetic rats was successful. Differential analysis of pancreatic islet tissue of people with diabetes mellitus, obtained the intersection of the two, obtained 20 mRNAs and 188 lncRNAs that were significantly differentially expressed, and constructed a regulatory network consisting of 327 nodes and 488 relational pairs, which are located in the core nodes of the regulatory network. Further analysis of one of the mannose receptor C1 (MRC1) revealed that NONRATT007456.2 and NONRATT020546.2 target the regulation of MRC1 expression by competitively binding to miR-5132-5p.lncRNA-NONRATT007456.2,lncRNA-NONRATT020546.2 can target MRC1 by competitively binding to miR-5132-5p and participate in the regulation of DE, which provides a new idea for the search for new clinical diagnosis and treatment biomarkers for DE.

Biological sciences/Computational biology and bioinformatics

Biological sciences/Neuroscience

Diabetes mellitus is a chronic and lifelong metabolic health problem, and hyperglycemia is its hallmark event^1. Diabetes Mellitus (DM) occurs due to lack of insulin or insulin resistance in the body, or both². Long-term exposure to chronic

hyperglycemia can cause damage to multiple organs such as the eyes, kidneys, heart, blood vessels, and nerves, resulting in a series of complications³. According to the10th edition of the "Global Diabetes Map" released by the International Diabetes Federation in 2021, in 2021, adults with diabetes will account for 10.5% of the total population, reaching 537 million people. As a major health problem, the prevention, occurrence and treatment of diabetes and its complications have become the focus of attention⁴. According to relevant statistics, there are more than 80 complications of diabetes, and its neurological complications include peripheral neuropathy, autonomic neuropathy, cognitive dysfunction and depression⁵. In 1965, biologist Nielon first proposed the concept of "diabetic encephalopathy". Diabetic encephalopathy (DE), also known as diabetic brain injury or type 3 diabetes, is generally believed to be a neurodegenerative disease caused by diabetes that leads to brain neuron damage and cognitive impairment⁶.

As we all know, RNA can be divided into two categories: coding RNA (coding RNA) and non-coding RNA (ncRNA)⁷. ncRNA includes rRNA, tRNA, lncRNA and miRNA and other RNAs with known functions, as well as some RNAs with unknown functions⁸. In 2011, Salmena et al. proposed that there is a special regulatory mechanism between coding RNA and non-coding RNA "the competing endogenous RNA hypothesis"⁹. miRNAs can bind to target mRNAs through response elements (miRNA Response Elements, MERs), thereby inhibiting the translation of target mRNAs, thereby achieving post-transcriptional regulation of gene expression. However, lncRNAs also have MERs that bind to miRNAs, so these lnRNAs will compete for binding to miRNAs, resulting in abnormal regulation of miRNAs on target mRNAs, thereby affecting RNA expression. At present, most of the research on ceRNA focuses on tumors, but the reports on the ceRNA regulatory network and DE are very rare.

In our study, we first constructed an animal model of DE, and combined the sequencing results of the hippocampal tissue of the GK rats and Wistar rats induced with DE and the diabetic islet tissue sequencing results downloaded from the GEO database, and analyzed the sequencing data by bioinformatics. The DE-related ceRNA regulatory network was constructed by the screened target genes, and the accuracy of the results was finally verified by molecular biology experiments, and then the pathogenesis of DE was explored.

Laboratory animals. This study selected SPF grade male GK rats and 12 male Wistar rats each, 24 weeks old, weighing 300–400 g. Among them, GK rats were provided by Changzhou Cavens Laboratory Animal Co., Ltd. [SCXK (Su) 2016-0010], and Wistar rats were provided by Changsha Tianqin Biotechnology Co., Ltd. [SCXK (Xiang) 2019-0013], which passed the quarantine inspection. After (GK rat certificate number: NO.202116131; Wistar rat certificate number: NO.430726210100200363), they were raised in the animal room of Jishou University Medical Research Center, the ambient temperature was 22–24°C, and the humidity was 45%-65%. The working illuminance is 150-300lx, the alternating time between day and night is 12h, feed and drinking water are supplemented every 12h, and the bedding is replaced every 24h. The animal experiment protocol has been reviewed and approved by the Experimental Animal Management and Ethics Committee of Jishou University.

Blood sugar monitoring. Blood was collected from the tail end of the rat, and the blood glucose of the rat was detected by a blood glucose meter.

New object recognition experiment. One day before the new object recognition experiment, the experimental rats were transferred from the animal room into the laboratory, and the rats were allowed to adapt to the environment for a day. Three objects of similar size were prepared. Among them, object one was exactly the same as object two, and object three was in shape and color. Different from things one and two, as new things. Put the rat into the novel object recognition room alone, without anything at this time, and adapt to it for 15 minutes; put object one and object two in the opposite position on the same side in the novel object identification room, and fix them. Put the rat with its back to the object and put it at the same distance from the object 1 and 2. Use the new object recognition system of Beijing Tianming Hongyuan Technology Development Co., Ltd. to record the number of times the rat explores the object 1 and object 2. The time is After 5 minutes, artificial correction was carried out. After the end, the rats were put back into the rearing cage, and the object 1, object 2 and the recognition room were cleaned with 75% alcohol to avoid leaving odor; the rats were placed in the rearing cage for 24 hours, and then conduct the test. At this time, change the object 2 to the completely different object 3, keep the position unchanged, put the rat with its back to the object, and put it in the same distance from the object 1 and the object 3, turn on the video equipment, observe and record the exploration The number of objects, and manual correction, the time is also 5 minutes. After finishing, clean the object and recognition chamber with 75% alcohol. Recognition index = number of touches on new things / (number of touches on new things + number of touches on old things) × 100% to calculate the recognition index of new things.

Hippocampal tissue extraction. The rats were fasted for 12 hours, and blood was collected from the tail vein for blood glucose detection. Intraperitoneal injection of 7% chloral hydrate was performed for sedation and anesthesia, and the anesthesia dose was 0.5ml/100g. After 5–10 minutes, the rats were completely anesthetized for decapitation, and the hippocampal tissue was taken out and quickly transferred to a -80°C refrigerator for subsequent experiments.

Electrophysiological experiments. Put the ACSF into a -80°C freezer for quick freezing for 30 minutes, take out and break the ice cubes to form a mixed solution of ice and water, and then introduce mixed oxygen (95% oxygen and 5% carbon dioxide) for 20 minutes until the liquid is saturated. In addition, all instruments that need to come into contact with the hippocampus are washed with ACSF at room temperature. The incubation tank was filled with ACSF and placed in a water bath, warmed to about 34°C and maintained with continuous oxygen mixing. The rat brain was taken out and immediately put into the ACSF ice-water mixed solution for 1–2 minutes. At the same time, the mixed oxygen was continuously filled to ensure the activity of the brain tissue, and then the hippocampus tissue was peeled off. The slices were sliced at a thickness of 400 µm with a vibrating microtome, and the sliced brain slices were transferred to the incubation tank with a small brush for about 1 h, and kept at 34°C during the period and continuously vented with mixed oxygen. The brain slice perfusion controller was turned on, and the ACSF water circulation was established, so that a sufficient amount of ACSF was loaded into the circulation and oxygen was introduced. The stimulating electrode is a tungsten wire electrode, and the recording electrode is a glass tube-shaped microelectrode filled with 3M NaCl. The electrode enters water and eliminates noise interference. Adjust the measurement state of the amplifier to the calibration state, and check the recording electrode resistance. The recording electrode resistance is about 2–4 MQ. After the incubation of the brain slices, the brain slices were placed on the nylon net of the semi-immersion thermostat recording tank, and the temperature in the bath was 32°C. Mixed oxygen was continuously injected above the brain slice, and unoxygenated ACSF was perfused at a constant rate below the brain slice. Adjust the liquid level of the recording tank, and the liquid level should be kept appropriate, that is, it should not be too high to prevent the brain slice from floating and affect the recording, nor too low to affect the activity of the brain slice. Turn on the microscope to synchronize the imaging, insert the stimulating electrode into the CA3 area and the recording electrode into the hippocampal CA1 area, and induce the electrophysiological experiment using the high-frequency stimulation method. The high-frequency stimulation consists of three 1-second, 100 Hz current clusters with 30 interval in seconds. Twenty minutes after a stable baseline response was obtained, a single stimulus at 100 Hz (1 sec) was administered, and the response was recorded for one hour to record field excitatory postsynaptic potentials (fEPSPs).

Golgi staining. Fixation: After the rat hippocampus was extracted, it was immediately rinsed with 4°C normal saline, and then directly put into neutral buffered formaldehyde solution. The neutral buffered formaldehyde solution should exceed the hippocampal tissue, and it was fixed in the dark for 24 hours. Chromating: Take out the fixed hippocampus and transfer it to a 5% potassium dichromate solution. The medium needs to be changed every day (24h) for 5 days. Silver staining: Take out the hippocampal tissue block, wash the residual potassium dichromate solution on the surface of the hippocampal tissue block with distilled water, and then put it in a 1.5% silver nitrate double-distilled aqueous solution, and store it in the dark for 5 days. The liquid needs to be changed every day. Embedding: After staining, remove the hippocampal tissue block, rinse with PBS buffer, and then absorb excess water with filter paper. Slicing: Fix the hippocampal tissue block on a vibrating microtome with glue, embed it with 4% agar, trim the agar block after the agar solidifies, and add the PBS buffer solution that did not cover the hippocampal tissue block to the slicing tank to a thickness of 110 µm. Sectioning was performed, and the excised tissue sections were rinsed twice with double-distilled water for 3 minutes each, placed on clean glass slides and blotted with excess water. Dehydration: Put the slices into 50%, 75%, 95%, and 95% alcohol in sequence for 4 minutes each time, then wash with absolute ethanol three times, and use an alcohol gradient to quickly dehydrate the slices. Rinse with xylene for 10 minutes to complete the transparent treatment, and finally seal with neutral resin and air dry. Filming: The sections were observed with an optical microscope, and the dendritic spines in the hippocampal CA1 area were photographed, and the images were counted and analyzed.

Data collection. Based on the above experimental results, 3 GK rats with induced diabetic encephalopathy and 3 normal Wistar rats were selected as the experimental group and the control group. The gene expression data of the hippocampus of GK rats and Wistar rats were recorded as data set 1; Through the GEO database, the dataset number is GSE25724, the microarray platform is [HG-U133A] Affymetrix Human Genome U133A Array (GPL96), and the sequencing samples are derived from 7 normal human pancreatic islet cells and 6 type 2 diabetes patients, which are recorded as datasets two.

Data normalization and differential expression analysis. Convert the gene expression data into matrix files, use the "DEGeR" R software package for normalization, and use the TMM normalization method to eliminate the influence of sequencing sample preparation or expression data during the library construction process, and at the same time according to the CPM value. Filter to screen out some low-expression genes or genes whose expression is 0. Then use the "Limma" R software package to perform differential analysis according to the standardized matrix file, and finally screen the mRNAs with different folds of gene expression |Log2FC|>1, P value < 0.05 as significantly differentially expressed genes (DEGs), lncRNA records for DELs. The significantly different gene screening results were visualized as volcano plots and cluster heatmaps using the "ggplot2" R package and the "HeatMap" R package. The R software version used was 4.0.2.

Enrichment analysis of significantly differentially expressed genes. Gene enrichment analysis was performed through DAVID database (http://david.abcc.ncifcrf.gov)¹⁰, according to Gene Ontology (GO) and Kyoto En-cyclopedia of Genes and Genomes (KEGG) to annotate DEGs and analyze the enrichment of DEGs in cellular components, molecular functions, biological functions and biological signaling pathways. Enriched clusters with P-value < 0.05 were considered statistically significant.

Protein interaction network analysis. Protein-Protein Interaction (PPI) network analysis was performed using the STRING database (http://string-db.org)¹¹. The data set was uploaded to the STRING database to obtain the PPI network map, and then the PPI network was imported into Cytoscape software. At the same time, the cytoHubba tool and the MCODE tool were used to perform cluster analysis on the PPI network to obtain the core sub-network that regulates the PPI network. A node with a high topology score can serve as a key node in the network. The Cytoscape software version used was 3.6.1.

Dataset 1 DEGs and DELs co-expression analysis. Using the matrix data of DEGs and DELs, the R language package Hmisc was used to calculate the Pearson correlation coefficient of each DEGs and DELs; correlation tests were performed for significant mRNA-lncRNA pairs with Pearson coefficient |r|>0.99 and p < 0.05, and further A list of DEG-DELs and co-expression relationships was obtained.

miRNA target gene prediction and construction of ceRNA network. Using the miRWalk3.0¹² database, according to the existing or predicted miRNA-mRNA regulatory relationship, the possible miRNA-mRNA regulatory relationship was screened with the binding degree P value > 0.95 (P = 1 considered as existing); The RNAhybrid plugin was used to predict the miRNA-lncRNA regulatory relationship, and the possible miRNA-lncRNA regulatory relationship was screened by thresholds such as binding free energy, number of mismatched bases, and P value < 0.05. According to the co-expression relationship between lncRNA-mRNA and the regulatory relationship between miRNA-mRNA and miRNA-lncRNA, the lncRNAs and mRNAs with significant differential expression and regulation of the same miRNA were screened and integrated. Endogenous competing RNA networks for key genes are finally obtained.

Antibodies and Kits. GAPDH rabbit polyclonal antibody and MRC1 rabbit polyclonal antibody were purchased from Cell Signing Technology, and anti-rabbit secondary antibody was purchased from Beijing Zhongshan Golden Bridge Biotechnology Co., Ltd. Reverse transcription kits and qPCR kits were purchased from Nearshore Protein Technology Co., Ltd., and BCA protein quantification kits were purchased from Beijing Soleibo Technology Co., Ltd.

Real-time PCR. Total RNA was extracted from rat hippocampus tissue using TRIzol reagent (Beijing Soleibo Technology Co., Ltd.) and reverse transcribed using a reverse transcription kit. Quantitative reverse transcription polymerase chain reaction was carried out on the CFX Connect™ real-time polymerase chain reaction detection system, using a qPCR kit, and real-time quantitative polymerase chain reaction was carried out on a real-time fluorescence quantitative PCR instrument of Illumina Company in the United States. The sequence is as follows (Table 1). The data expression ratio analysis was performed using the 2-∆∆Ct method, and the data were analyzed and graphed using GraphPad Prism 8.0.2 software.

Table 1

Primer sequences
RNAs	F	R
NONRATT020546.2	TCACGCTCAACACAAATGTGC	TGAAGGCTGGACTGGGTACG
NONRATT007456.2	AAGGTCAAACAGGAAGATACTCGT	CCATCAAACCCAAGAACAAGC
MRC1	TGGAGTGGCAGGTGGTTTATG	GGTTCAGGAGTTGTTGTGGGC
β-actin	CAGCAAGCAGGAGTACGATGA	GCTCAGTAACAGTCCGCCTAGAA

Western blotting. PMSF (Biyuntian Biotechnology Co., Ltd.) was added to RIPA strong lysis solution (Biyuntian Biotechnology Co., Ltd.) to make the final concentration of PMSF 1 mM, and 150–250 µl of lysis solution was added per 20 mg of tissue to extract protein. The protein concentration was determined by the BCA kit method, and the protein concentration of each histone was adjusted to the same. After SDS-PAGE, it was transferred to PVDF membrane, blocked with 5% nonfat milk powder at room temperature for 1-1.5 hours, and incubated overnight at 4°C with primary antibodies in the following proportions: anti-MRC1 (1:1000), anti-GAPDH (1:1000), then washed with 1xTBST, incubated with horseradish peroxidase labeled secondary antibody at room temperature for 1.5 to 2 hours, photographed and stored using a chemiluminescence imager from Shanghai Qinxiang Scientific Instrument Co., Ltd. degree value.

Blood sugar monitoring results. The results showed that, compared with Wistar rats, GK rats had significantly higher blood glucose levels at each stage on an empty stomach and after eating. After 0.5 h after eating, the blood glucose value rose rapidly, reaching a peak in about 1 h. It was about 27mM/L, and then the blood sugar began to decrease, but the blood sugar level was still high after 2 hours, and the blood sugar changes were consistent with the characteristics of type 2 diabetes.

Recognition cognitive decline in GK rats. A new object recognition experiment was carried out on the experimental rats to observe the recognition and cognitive abilities of the experimental rats. The results showed that the two groups of experimental rats had little difference in the recognition ability of objects of the same distance, the same shape, size and color, but when they were re-inserted into the new object recognition ability. After the third thing, the normal Wistar rats showed obvious ability to recognize new things, and the recognition index was significantly higher than that of the GK rat group, which proved that the GK rats had a certain degree of cognitive dysfunction (Fig. 1).

Significantly lower fEPSPs in GK rats. By applying high-frequency stimulation to the hippocampal slices of rats, it was found that the response of the hippocampus of GK rats to external current stimulation was weakened, and the excitatory postsynaptic potential decreased significantly after stimulation (Fig. 2), indicating the function of synaptic signal transmission in the hippocampus of GK rats. been affected to some extent.

Structural changes of hippocampal dendritic spines in GK rats. In order to observe the changes of neurons in the hippocampus of the rat model, the dendritic spines of the hippocampus were stained and photographed by oil lens. The results were analyzed by Image J software. Compared with the Wistar rats, the structure of the hippocampus of the GK rats was obviously changed. Changes, the density of dendritic spines decreased significantly, the diameter of dendritic spines became smaller, and the dendritic spines became shorter. The structure and function of neurons in the hippocampus were severely damaged (Fig. 3).

Identification of differentially expressed genes. Dataset 1 screened and obtained 344 significantly differentially expressed mRNAs (dataset 1, DEGs), including 278 up-regulated mRNAs and 66 down-regulated mRNAs. The cluster heat map shows that these significantly differentially expressed mRNAs have obvious Clustering features and expression differences (Fig. 4). In the expression analysis of lncRNAs, there were 188 significantly differentially expressed lncRNAs (dataset 1, DELs), including 114 up-regulated lncRNAs and 61 down-regulated lncRNAs. The DELs clustering heat map also shows obvious expression differences (Fig. 5). The screening results of dataset 2 yielded 1844 significantly differentially expressed mRNAs (DEGs in dataset 2), including 1704 up-regulated mRNAs and 140 down-regulated mRNAs (Fig. 6). Through Venn diagram analysis, there were 20 significantly differentially expressed mRNAs in both datasets (Fig. 7): IFITM1, MRC1, CD74, RGS4, SERPING1, RPS7, TOX3, MYOF, CDK10, GPX3, PRSS23, KPNA2, SH2B2, YBX1, RBP4, GEM, PLAGL1, SGK1, VIM, MX2.

DEGs enrichment analysis. In order to understand the biological processes and pathways that significantly different genes may be involved in, and the functional roles of these genes, we performed enrichment analysis on DEGs. The GO enrichment includes cellular component enrichment (CC), molecular function (MF) and biological process (BP). The results showed that dataset 1 DEGs were significantly enriched in CC-enriched clusters such as molecular complexes, cell surfaces, and the perinuclear region of the cytoplasm, and were also enriched in insulin-like growth factor binding, calcium ion binding, transmembrane MF-enriched clusters such as signaling receptor activity, and BP-enriched clusters such as vascular remodeling, positive regulation of gene expression. The KEGG enrichment results showed that DEGs were significantly enriched in signaling pathways such as axon guidance, neuroactive ligand-receptor interactions, and PI3K-Akt signaling pathways (Fig. 8). The enrichment analysis results of these DEGs extracted from rat hippocampus tissue confirmed that they play an important role in diabetic encephalopathy.

The results of data set 2 enrichment analysis showed that these DEGs were significantly enriched in CC-enriched clusters such as proteasome complexes and exosomes, and were also enriched in MFs such as RNA binding and translation initiation factor activity. Enriched clusters, and BP-enriched clusters that regulate mRNA stability and regulate cellular amino acid metabolism processes. The KEGG enrichment results showed that DEGs were significantly enriched in signaling pathways such as Alzheimer's disease, Parkinson's disease, and carbon metabolism (Fig. 9). These DEGs extracted and detected from human pancreatic islet tissue were concentrated in some enriched clusters related to metabolism, regulation of gene expression, protein synthesis, neurological diseases, etc., proving that the abnormal expression of these DEGs is closely related to the occurrence of diseases. big relationship.

Construction of PPI network and core module extraction. Protein interaction network analysis was performed using the STRING database to obtain a PPI network map. Cluster analysis obtained the core sub-network that regulates the PPI network. After removing the discrete points in the PPI network, the results show that the dataset-PPI network consists of 282 nodes and 815 interaction pairs, and the PPI network P value is < 1.0e-16. The MCODE tool analyzes the core module of the PPI network and obtains a core module including 17 key nodes and 29 pairs of interactions; and the nodes with high topological scores in the PPI network can be used as the key nodes of the network. Therefore, using cytoHubba The tool selects the genes with the top 50 node scores as the dataset-core sub-network (Fig. 10), including 50 nodes and 318 interaction pairs

Co-expression analysis. In order to identify the molecular mechanism of lncRNA in diabetic encephalopathy, the co-expression relationship between lncRNA and mRNA was calculated according to the change of gene expression, and the correlation coefficient was proportional to the degree of association between the two. The list of DEGs-DELs and co-expression relationships revealed a total of 1147 co-expression pairs, including 121 DELs and 248 DEGs. By aligning with the significantly differentially expressed genes in the two datasets, 11 DEGs and 19 DELs were finally found to have a certain co-expression relationship.

ceRNA network construction. MiRNA target gene prediction was performed on the genes screened by co-expression analysis, among which a total of 286 miRNA target genes were predicted to match 11 DEGs, and a total of 60 miRNA target genes were predicted to match 19 DELs. These DELs and DEGs can share miRNA The lncRNA-miRNA-mRNA regulatory network constructed with this binding site consists of 327 nodes and 488 relational pairs (Fig. 11).

Screen the target gene. By integrating the above analysis, we found that MRC1 was significantly abnormally expressed in both hippocampus and pancreatic islets. MRC1 protein is also in the core sub-network of the top 50 core modules and nodes in the PPI protein interaction network, and plays a core regulatory role. Finally, through the prediction of miRNA target genes of MRC1, the ceRNA regulatory network of MRC1 was constructed (Fig. 12), and the ceRNA sub-network of NONRATT007456.2/NONRATT020546.2-miR-5132-5p-MRC1 was screened.

Gene expression in ceRNA networks. The expression of the target gene in the ceRNA network was further verified by RT-qPCR. The experimental results showed that the expressions of NONRATT007456.2 and MRC1 were up-regulated, and the expressions of NONRATT020546.2 and miR-5132-5p were down-regulated (Fig. 13).

Protein expression. The expression of MRC1 protein was detected by western blotting. When β-actin was used as an internal reference, it was found that the expression of MRC1 protein was increased in the hippocampus of GK rats (Fig. 14).

With the improvement of living standards and the improvement of dietary conditions, the incidence of diabetes has increased greatly, and various complications caused by diabetes have also increased. As one of the complications of diabetes, DE can cause serious damage to the brain, especially the hippocampus, posing a huge threat to people's lives and even their safety. At present, the research on DE has focused on the fields of blood vessels, inflammation, oxidative stress, and neuronal apoptosis. However, the factors that cause DE are very complex, and the research on the pathogenesis and treatment of DE is still difficult.

The rapid development of transcriptomics has brought new directions for the research of DE. More and more studies have obtained a deeper understanding of the mechanism of disease by analyzing gene expression and gene epigenetic regulation. Treatment brings new ideas. The emergence of the ceRNA regulatory mechanism not only provides a more comprehensive explanation of the functions of lncRNAs, but also confirms the existence of common signaling factors between miRNAs and lncRNAs, which can regulate gene expression by acting together.

Therefore, in this study, the blood glucose detection of GK rats and Wistar normal rats proved that the GK rats screened out in the experiment showed symptoms of T2DM; the results of novel object recognition experiments, electrophysiological experiments and Golgi staining showed that diabetes can lead to cognition. Dysfunction occurs; high-throughput sequencing technology was performed on the hippocampus tissue conforming to DE. Through the gene difference analysis of data set 1 and data set 2, we obtained 20 genes that were significantly differentially expressed in common, and these genes are likely to regulate diabetes. A key factor in brain injury; at the same time, the gene enrichment results of DEGs in the two datasets showed that these screened DEGs were significantly enriched in many DE-related biological processes or pathways, including Alzheimer's disease Silence disease, Parkinson's disease, vascular development, axon guidance, PI3K-AKT signal transduction pathway, etc., have proved that most DEGs play a certain role in the occurrence of DE. The results of PPI network analysis of these DEGs showed that among the 20 genes that were screened out with significant differential expression, some genes played an important regulatory role in the PPI network, and these genes occupied a core position in the PPI network. proteins are closely linked. However, we do not know how these genes are involved in the occurrence of DE; therefore, we constructed a ceRNA network and used this ceRNA regulatory network to explore the mechanism of DE. Through co-expression analysis, the co-expression network of DEGs and DELs was constructed. These mRNA-lncRNA pairs, which are closely related in expression, reveal the functions of lncRNAs to a certain extent. At the same time, the miRNA target prediction of mRNAs and lncRNAs in these co-expression networks formed a potential regulatory network of lncRNA-miRNA-mRNA. After unified integration, the ceRNA regulatory network of these core genes in the occurrence of DE was obtained. After analysis, the target gene MRC1 was identified, and we obtained the ceRNA network of NONRATT007456.2/NONRATT020546.2-miR-5132-5p-MRC1.

A study has constructed a ceRNA network related to the pathogenesis of insulin resistance¹³. In 123 T2DM experimental groups and 106 normal control groups, LncRNA-RP11-773H22.4 was significantly up-regulated, while miRNAs (miR-3163 and miR-3163) were significantly up-regulated. -1) and mRNA (GLUT4 and AKT2) expressions showed marked downregulation in diabetic samples. Afterwards, LncRNA-RP11-773H22.4 was successfully knocked out by CRISPR/Cas9. After gene editing, the expression of the ceRNA network at the RNA and protein levels returned to normal, proving that gene expression in DM is regulated by the ceRNA network. In addition, it has also been reported that LncRNA NORAD is upregulated in DM mice, leading to enhanced cellular inflammation and fibrosis¹⁴. In addition, some studies have pointed out¹⁵ that lncRNA H2k2 promotes the proliferation of diabetic nephropathy mesangial cells through the miR-449a/b/Trim11/Mek signaling pathway; in diabetic mice, lncRNA 3632454L22Rik can absorb miR- 181a-5p promotes corneal epithelial wound healing¹⁶. These ceRNA studies related to DM have greatly deepened our understanding of various complications of DM.

MRC1 is a C1-type mannose receptor mainly present in macrophages, which is involved in immune regulation and regulates the transformation of macrophages. Therefore, MRC1 is closely related to the occurrence of inflammation, and can also participate in the occurrence of DE through inflammation. Constructing the ceRNA regulatory network of MRC1 can help us deepen our understanding of the pathogenesis of DE, and miR-5132-5p, as a rare reported miRNA, may bring new insights into the study of DE.

RT-qPCR results showed that the expressions of NONRATT007456.2 and MRC1 were up-regulated, and the expressions of NONRATT020546.2 and miR-5132-5p were down-regulated, which were consistent with the sequencing results. The expression of MRC1, so the function of miR-5132-5p may be the same as that of most miRNAs, which can inhibit the expression of target genes, and the low expression of miR-5132-5p can also induce MRC1 overexpression to a certain extent; WB The experimental results showed that the expression of MRC1 was the same as that of RT-qPCR.

The target gene MRC1 is a mannose C1-type receptor mainly present on macrophages, also known as CD206, which is involved in immune response and is related to the occurrence of inflammation, expressing a certain anti-inflammatory effect¹⁷. Regarding macrophages, they can be divided into two categories, classical activated M1 type and alternatively activated M2 type, and MRC1 is one of the markers of M2 type¹⁸. When some inflammatory factors appear, it will promote the polarization of MI type to M2 type. For example, after administration of interleukin 4 (IL-4) to rats for 4 hours, MRC1, which is an M2 marker, will be expressed in large quantities, which plays an anti-inflammatory effect¹⁹. At the same time, studies have also pointed out that after the combination of IL-4 and IL-13 induces polarization to form M2 macrophages, it exhibits higher mRNA expression levels of MRC1, Il10, Tgfb1, Arg1 and Ccl17²⁰, while these Genes are closely related to inflammation. After injecting M2 macrophages into kidney-injured mice, it was found to effectively alleviate inflammation and kidney injury. In addition, there are some reports²¹ that MRC1 can integrate multiple signals, thus defining a general safeguard mechanism to protect genome integrity during transcription.

In the persistent chronic high glucose state of diabetes, various metabolic pathways in the body are disturbed, resulting in continuous inflammation. Especially in DE, most of the nerve damage in the brain is caused by inflammation and vascular lesions. Therefore, in DE, the expression of MRC1 is up-regulated due to inflammation. Studies have reported that CD206 macrophages in diabetic patients are increased, are rich in scavenger receptors, have low intracellular lipids, and secrete pro-inflammatory cytokines²². In addition, diabetic visceral fat is rich in CD206-type macrophages and inflammatory signatures, scavenger receptors, and MHC II antigen-presenting genes²³. Immunostaining revealed that CD206-type macrophages were concentrated in their adjacent vascularized lymphoid masses, demonstrating that MRC1 expression is upregulated in diabetic complications.

However, in tumors, MRC1 expresses completely opposite regulatory effects. Among tumor-associated macrophages, M1 tumor macrophages (killing tumors) and M2 tumor macrophages (promoting tumor growth)²⁴, while M2-type tumor macrophages mainly express MRC1, which plays a huge role in the polarization of M1-type macrophages to M2-type macrophages²⁵, and promotes tumorigenesis in most tumors. The M2 phenotype of growth prevails²⁶. However, the expression of MRC1 was found to decrease in breast cancer MCF-7 cells after culturing with hydrogen, which proved that hydrogen can promote the M1 polarization of macrophages, inhibit the M2 polarization of macrophages, effectively inhibit tumors and induce tumor apoptosis²⁷. Therefore, it is not difficult to infer that the high expression of MRC1 will promote the development of tumors to a certain extent. In fibrosis and cirrhosis caused by chronic liver inflammation, highly fibrotic livers have a greater density of CD206 macrophages²⁸, that is, high expression of MRC1. These studies suggest that MRC1 has some other functions in addition to its anti-inflammatory properties, but this may also be species-specific or tissue-specific.

In conclusion, we constructed a lncRNA-miRNA-mRNA regulatory network that may be related to DE occurrence, and these genes may interact through the ceRNA network to play their roles. We conclude that NONRATT007456.2 and NONRATT020546.2 affect MRC1 expression by competitively adsorbing miR-5132-5p, and this ceRNA regulatory network may contribute to the occurrence of diabetic encephalopathy by promoting inflammation.

Ethics approval and consent to participate

Animal research is carried out in full compliance with the basic principles of the Basel Declaration and the guidelines of the Committee on Ethics，And the research was conducted according to the ARRIVE guidelines. The experimental protocol was formulated in accordance with the ethical guidelines of the Declaration of Helsinki and approved by the Ethics Committee of the Institute of Medical Research of the Medical Research Center of Jishou University (JSDX-2022-0090).

Consent for publication

Not Applicable.

Availability of data and materials

The original sequencing data was uploaded to the SRA database with the login number PRJNA897949(https://dataview.ncbi.nlm.nih.gov/object/PRJNA897949?reviewer=mu8pcbchfevf084u6d9bd7fo69). Other data generated or analyzed in the course of this study are included in this paper. For further enquiries, please contact the corresponding author.

Competing interests

The authors declare that they have no competing interests.

Funding

This research was supported by the National Natural Science Foundation of China ( (X.Li)).

Authors' contributions

Fu Chuanjun participated in research design, data processing and statistical analysis and drafted the manuscript; Dong Shuai conducted design studies and participated in data analysis; Tao Jiasheng participated in data processing; Liu Lini participated in manuscript writing. All authors have read and approved the final manuscript.

Acknowledgements

Not applicable.

Zhang, X., et al., Anti-diabetic drugs and sarcopenia: emerging links, mechanistic insights, and clinical implications. 2021. 12(6): p. 1368–1379.
Khalid, M., et al., Insulin Signal Transduction Perturbations in Insulin Resistance. 2021. 22(16).
Abdul, Y., et al., Deferoxamine Treatment Prevents Post-Stroke Vasoregression and Neurovascular Unit Remodeling Leading to Improved Functional Outcomes in Type 2 Male Diabetic Rats: Role of Endothelial Ferroptosis. 2021. 12(4): p. 615–630.
Wu, B. and L.J.E.h.j.C.p. Shi, Cost-utility of ticagrelor plus aspirin in diabetic patients with stable coronary artery disease. 2021. 7(6): p. 529–538.
Henson, J., et al., Spinal Cord Stimulation for Painful Diabetic Peripheral Neuropathy: A Systematic Review. 2021. 10(2): p. 895–908
Luo, Z., et al., CAPE-pNO ameliorates diabetic brain injury through modulating Alzheimer's disease key proteins, oxidation, inflammation and autophagy via a Nrf2-dependent pathway. 2021. 287: p. 119929.
Rincón-Riveros, A., et al., Bioinformatic Tools for the Analysis and Prediction of ncRNA Interactions. 2021. 22(21).
Ning, L., et al., MNDR v3.0: mammal ncRNA-disease repository with increased coverage and annotation. 2021. 49: p. D160-D164.
Salmena L., et al., A ceRNA hypothesis: the Rosetta Stone of a hidden RNA language? Cell. 2011 Aug 5;146(3):353-8. doi: 10.1016/j.cell.2011.07.014. Epub 2011 Jul 28. PMID: 21802130; PMCID: PMC3235919.
Dennis G Jr, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC, Lempicki RA. DAVID: Database for Annotation, Visualization, and Integrated Discovery. Genome Biol. 2003;4(5):P3. Epub 2003 Apr 3. PMID: 12734009.
Szklarczyk D, Gable AL, Nastou KC, Lyon D, Kirsch R, Pyysalo S, Doncheva NT, Legeay M, Fang T, Bork P, Jensen LJ, von Mering C. The STRING database in 2021: customizable protein-protein networks, and functional characterization of user-uploaded gene/measurement sets. Nucleic Acids Res. 2021 Jan 8;49(D1):D605-D612. doi: 10.1093/nar/gkaa1074. Erratum in: Nucleic Acids Res. 2021 Oct 11;49(18):10800. PMID: 33237311; PMCID: PMC7779004.
Sticht C, De La Torre C, Parveen A, Gretz N. miRWalk: An online resource for prediction of microRNA binding sites. PLoS One. 2018 Oct 18;13(10):e0206239. doi: 10.1371/journal.pone.0206239. PMID: 30335862; PMCID: PMC6193719.
Matboli, M., et al., Identification of Novel Insulin Resistance Related ceRNA Network in T2DM and Its Potential Editing by CRISPR/Cas9. 2021. 22(15).
Liu, Y., et al., NORAD lentivirus shRNA mitigates fibrosis and inflammatory responses in diabetic cardiomyopathy via the ceRNA network of NORAD/miR-125a-3p/Fyn. 2021. 70: p. 1113–1127.
Chen, W., et al., viaThe topological key lncRNA H2k2 from the ceRNA network promotes mesangial cell proliferation in diabetic nephropathy the miR-449a/b/Trim11/Mek signaling pathway. 2019. 33(10): p. 11492–11506.
Chen, X., J.J.I.o. Hu, and v. science, Long Noncoding RNA 3632454L22Rik Contributes to Corneal Epithelial Wound Healing by Sponging miR-181a-5p in Diabetic Mice. 2021. 62(14): p. 16.
von Ehr, A., et al., MRC1Inhibition of Microglial TGFβ Signaling Increases Expression of. 2020. 14: p. 66.
Zhou, Y., et al., Mannose receptor modulates macrophage polarization and allergic inflammation through miR-511-3p. 2018. 141(1): p. 350–364.e8.
Rahabi, M., et al., Divergent Roles for Macrophage C-type Lectin Receptors, Dectin-1 and Mannose Receptors, in the Intestinal Inflammatory Response. 2020. 30(13): p. 4386–4398.e5.
van der Zande, H., et al., The Mannose Receptor: From Endocytic Receptor and Biomarker to Regulator of (Meta)Inflammation. 2021. 12: p. 765034.
Duch, A., et al., Multiple signaling kinases target MRC1 to prevent genomic instability triggered by transcription-replication conflicts. 2018. 9(1): p. 379.
Xiang, G., et al., The impact of sitagliptin on macrophage polarity and angiogenesis in the osteointegration of titanium implants in type 2 diabetes. 2020. 126: p. 110078.
Fiani, M., et al., Exploiting Manipulated Small Extracellular Vesicles to Subvert Immunosuppression at the Tumor Microenvironment through Mannose Receptor/CD206 Targeting. 2020. 21(17).
Zhao, S., et al., Tumor-derived exosomal miR-934 induces macrophage M2 polarization to promote liver metastasis of colorectal cancer. 2020. 13(1): p. 156.
Gunassekaran, G., et al., M1 macrophage exosomes engineered to foster M1 polarization and target the IL-4 receptor inhibit tumor growth by reprogramming tumor-associated macrophages into M1-like macrophages. 2021. 278: p. 121137.
Debacker, J., et al., The Prognostic Value of CD206 in Solid Malignancies: A Systematic Review and Meta-Analysis. 2021. 13(14).
Li, T., et al., Effects of a high concentration of hydrogen on neurological function after traumatic brain injury in diabetic rats. 2020. 1730: p. 146651.
Tan-Garcia, A., et al., Liver fibrosis and CD206 macrophage accumulation are suppressed by anti-GM-CSF therapy. 2020. 2(1): p. 100062.

No competing interests reported.

Supplementarymaterials.zip

Download PDF

Version 1

posted

You are reading this latest preprint version

Exploring the pathogenesis of diabetic encephalopathy based on the ceRNA regulatory network of lncRNA-miRNA-mRNA

Status:

Version 1

Abstract

Figures

Main text

Materials And Methods

Result

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1