Identification of small ubiquitin-related modifier (SUMO)-related genes-based biomarkers in Alzheimer's disease based on bioinformatics analysis

Identification of small ubiquitin-related modifier (SUMO)-related genes-based biomarkers in Alzheimer's disease based on bioinformatics analysis | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Identification of small ubiquitin-related modifier (SUMO)-related genes-based biomarkers in Alzheimer's disease based on bioinformatics analysis Peng Chen, Mengting Fan, Hailiang Lin, Zhong Chen, Qiuyang Zhang, and 2 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-6983921/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 20 Apr, 2026 Read the published version in Scientific Reports → Version 1 posted 11 You are reading this latest preprint version Abstract Background To explore the role of small ubiquitin-related modifier (SUMO) in Alzheimer's disease (AD) and its pathogenesis, bioinformatics was used to search for SUMO-related genes (SRGs)-based biomarkers. Methods Datasets related to AD (GSE140831, GSE63060), a dementia dataset (GSE140830), and 189 SRGs were retrieved from public databases. Candidate genes were identified by intersecting differentially expressed genes (DEGs) with SRGs. A protein-protein interaction (PPI) network was constructed to select the top 15 core genes, and the support vector machine-recursive feature elimination (SVM-RFE) model identified feature genes. Validation was done using the GSE140831 and GSE63060 datasets, and the nomogram model was assessed by receiver operating characteristic (ROC) curve analysis. Gene set enrichment analysis (GSEA) and other analyses were performed. Reverse transcription quantitative polymerase chain reaction (RT-qPCR) was used for further validation. Results Overlapping 189 SRGs and 12,853 DEGs identified 107 candidate genes. Five feature genes were selected using the SVM-RFE algorithm. CREBBP , PIAS1 , and TRIM28 were confirmed as AD biomarkers due to their increased expression in AD and strong ROC performance. GSEA highlighted their involvement in pathways such as olfactory transduction, lysosome, and spliceosome. Immune infiltration analysis suggested immune cell involvement in AD progression. Additionally, 21 potential drugs for AD therapy were predicted. RT-qPCR confirmed the over-expression of CREBBP and TRIM28 in AD samples, consistent with dataset trends. Conclusion CREBBP , PIAS1 , and TRIM28 were identified as SRG-based biomarkers for AD diagnosis, providing new insights into AD pathogenesis. Health sciences/Biomarkers Biological sciences/Computational biology and bioinformatics Health sciences/Diseases Biological sciences/Neuroscience Alzheimer’s disease Small ubiquitin-related modifier-related genes Bioinformatics analysis Biomarkers Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 1. Introduction Alzheimer's disease (AD) is the most common neurodegenerative disease worldwide, with progressive cognitive decline as the main clinical feature [ 1 ]. Alzheimer's Association International estimated there are around 50 million people with dementia worldwide, which is expected to triple by 2050 [ 2 ]. The main manifestations of AD are progressive cognitive impairment and personality disorder [ 3 ]. The APOE gene is the most common risk gene for AD [ 4 ]. Deposition of amyloid beta (Aβ) in the brain parenchyma and cerebrovascular system, as well as the presence of neurofibrillary tangles within neurons and progressive loss of synapses, are central neuropathological markers of AD [ 5 ]. Current drug treatments of AD only have symptomatic effects, and there is no treatment to alleviate the disease [ 6 ], so new advances in the diagnosis and treatment of AD remain an urgent task. The search for biomarkers of AD is of great significance in further promoting AD treatment. Small ubiquitin-related modifier (SUMO), a 97-residue protein, mediates post-translational modification through covalent attachment to a specific lysine residue on the target protein [ 7 ]. Ubiquitination plays a important role in the biochemical pathways of eukaryotes by influencing the location, function, and stability of modified proteins [ 8 ]. In histopathology, AD is characterized by the insoluble aggregation of Aβ and microtubule-associated protein tau in the brain, both of these proteins are associated with SUMO. It has been demonstrated that tau SUMOylation reciprocally promotes its hyperphosphorylation, which results in aggregation of tau to form neurofibrillary tangles, and eventually interferes with the normal function of neurons and leads to nerve cell death [ 9 ]. And early studies have indicated that the SUMO system may impact Aβ levels and tau aggregation by altering with AD-type pathology [ 10 ]. Telomere length is related to biological aging, a recent study has shown that telomere attrition is associated with the occurrence of AD [ 11 ]. Accumulating evidence indicates that the ubiquitin and SUMO pathways are crucial for telomere structure integrity because of they are significant regulators of the shelterin complex and other chromatin modifiers [ 12 ]. However, the role of SUMO-related genes (SRGs) in AD is still unclear, so identifying SRGs-based biomarkers for AD’s diagnosis and investigating the mechanisms by which these factors influence AD may be of great significance to improving the intervention of AD patients and offering a novel insight for the treatment of AD. In this study, candidate genes were acquired based on AD-related datasets (GSE140831) and 189 SRGs, then SUMO-related genes associated with AD were determined by using the Support Vector Machine-Recursive Feature Elimination (SVM-RFE) algorithm, differential expression, and functional enrichment analyses. Our study could lead to earlier and more accurate AD diagnosis by identifying SRG-based diagnostic biomarkers. In addition, understanding the role of these biomarkers may be of great significance to discovering new therapeutic targets and providing new insights into the development of drugs or other treatment modalities to slow or stop AD progression. 2 Materials and methods 2.1 Data collection The expression profiles of GSE140831 (GPL15988), GSE63060 (GPL6947), and GSE140830 (GPL15988) datasets were searched from the Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo/) [13]. Specifically, 734 peripheral blood samples from 204 patients with AD and 530 healthy persons (controls) in the GSE140831 dataset were selected as a training set. The GSE63060 dataset offered peripheral blood samples from 145 AD and 104 controls. Besides, the GSE140830 dataset was composed of peripheral blood samples from 261 dementia patients (80 of behavioral variant frontotemporal dementia (bvFTD), 44 in semantic variant primary progressive aphasia (svPPA), 47 from non-fluent variant primary progressive aphasia (nfvPPA), 54 of progressive supranuclear palsy (PSP) and 36 of corticobasal syndrome (CBS) and 281 controls. A sum of 189 small ubiquitin-related modifiers (SUMO) related genes (SRGs) was retrieved from the Molecular Signatures Database (MSigDB, https://www.gsea-msigdb.org/) as 'SUMOylation' being a keyword. 2.2 Differential expression and functional enrichment analyses The differentially expressed genes (DEGs) of AD between AD and control groups in the GSE140831 dataset were obtained employing the R package 'limma' (Ver. 3.54.1) [14] (|Log 2 fold-change (FC)|>0.1, adjusted p<0.05). Subsequently, utilizing the R package 'ggVenndiagram' (Ver. 1.2.2) [15], the candidate genes associated with AD and SRGs were acquired through intersecting observably up-regulated and down-regulated DEGs with SRGs, respectively. Subsequently, the relevant functions and signaling pathways in which candidate genes were involved were investigated via Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genome (KEGG), implemented with the R package 'clusterProfiler' (Ver. 4.6.2) [16] (adj p<0.05). Additionally, a protein-protein interaction (PPI) network revealing interactions among proteins encoded by candidate genes was built and visualized by the Search Tool for Recurring Instances of Neighboring Genes database (STRING, https://string-db.org) (Interaction score=0.4) and the plug-in CytoHubba of Cytoscape software severally (Ver. 3.9.0) [17]. The TOP 15 candidate genes in the PPI network were singled out as core genes for screening biomarkers. 2.3 Acquisition of the biomarkers and construction of a nomogram model For the sake of searching biomarkers recognizing and diagnosing AD, the SVM-RF model-selected feature genes were obtained firstly from core genes when the model attained the highest accuracy and the lowest error rate after 10-fold cross-validation by R package 'e1071' (Ver. 1.7-13) [18]. Afterwards, the expression verification of feature genes in the GSE140831 and GSE63060 datasets proceeded to ascertain genes with observably discrepant expression levels in AD and control groups and consistent expression trends in the two datasets as the biomarkers for AD. Wilcoxon's test estimated the significantly different expression levels of inter-group feature genes (p<0.05). Furthermore, based on the GSE140831 dataset, a nomogram model predicting the risk of AD was established through the R package 'rms' (Ver. 6.5-0) [19]. The nomogram model could visualize a total point for each sample in the GSE140831 dataset, and the risk of AD grew along with the increased total points. The predictive performance of the nomogram model was appraised by plotting the receiver operating characteristic (ROC) curve through the R package 'pROC' (Ver. 1.18.0) [20] and computing the area under the ROC curve (AUC). 2.4 Gene set enrichment analysis (GSEA) To explore whether biomarkers were involved in regulating the progression of AD, the underlying pathways enriched by them were defined via GSEA. The Spearman's correlation coefficients between biomarker and other genes in the GSE140831 dataset were calculated adopting the R package 'psych' (Ver. 2.2.9) [21] and queued in order, and the sorted genes were input into the 'clusterProfiler' package in R to match relevant pathways of biomarkers (adj.p1). 2.5 Construction of molecular regulatory network The prediction of upstream molecules targeting biomarkers was beneficial in understanding the biological functions of biomarkers. So, the transcription factors (TFs) regulating biomarkers were predicted via the NetworkAnalyst (http://www.networkanalyst.ca) database. Moreover, the microRNAs (miRNAs) retrieved from the miRWALK (http://mirwalk.umm.uni-heidelberg.de/) and miRDB (http://www.mirdb.org) databases were overlapped as miRNAs corresponding to biomarkers. The feat of the Cytoscape software synthesized the TF-mRNA and miRNA-mRNA networks. 2.6 Immunoinfiltration characterization and drug prediction To characterize the discrepancies in immune cell infiltration between AD and control groups, the abundance of 64 kinds of immune cells in each sample of the GSE140831 dataset was estimated through the xCell algorithm [22]. The discrepant immune cells in the AD and control groups were found through Wilcoxon's test (p0.4, p<0.05). Moreover, the Drug-Gene Interaction Database (DGIdb, www.dgidb.org) identified feasible drugs targeting biomarkers for AD therapy, and a gene-drug network was created through Cytoscape software. 2.7 Assessment of expression levels of biomarkers in the five dementia-related diseases To understand the expression of biomarkers in five dementia-related diseases (bvFTD, svPPA, nfvPPA, PSP, and CBS), the expression patterns of the biomarkers in AD and control groups were first displayed in a heatmap plotted by the R package 'ggplot2' (Ver. 3.3.6) [23]. Besides, the differences in the expression of biomarkers in five dementia-related diseases were compared through Wilcoxon's test (p<0.05). 2.8 Real-time reverse transcriptase-polymerase chain reaction (RT-qPCR) The RT-qPCR was the assay of choice for detecting relative biomarker expression levels in five pairs of AD and control whole blood samples gathered from Fuzhou Second General Hospital. All sample donors signed the informed consent forms. This study was conducted in accordance with the Declaration of Helsinki and obtained ethical approval from the Ethics Committee of Fuzhou Second People's Hospital (Approval No.: 2024005).The total RNA was extracted from the ten samples using the TRIzol reagent (Ambion, USA) per the manufacturer's protocol. The concentrations of total RNA were detected in rapid sequence via the NanoPhotometer N50. Ulteriorly, the cDNA synthesis was reverse-transcribed using the SureScript-First-strand-cDNA-synthesis-kit (Servicebio, China). Additionally, the CFX Connect Thermal Cycler (Bio-Rad, USA) was operated to detect the relative expression level of mRNA using the GAPDH as an internal reference gene for RT-qPCR through the 2 -ΔΔCT method. The details of all primers are attached in Online Resource 1 which can be found in the supplementary file. 2.9 Statistical analysis Bioinformatics analysis was implemented through this study's R software (Ver. 4.3.2). The Wilcoxon's test was used to distinguish the inter-group discrepancies, and a p-value or adj. p-value of less than 0.05 was deemed statistically significant. 3 Results 3.1 The functions of candidate genes might be related to nucleic acid and cytoplasmic transport and cancer regulation 12,853 DEGs were identified between the AD and control groups in the GSE140831 dataset ( Fig. 1a ). 107 candidate genes related to AD and SRGs, with 85 up-regulated and 22 down-regulated, were obtained by overlapping 189 SRGs and 6,640 up-regulated and 6,213 down-regulated DEGs ( Fig. 1b-c ). They were enriched in 624 GO entries, comprising 498 biological processes (BPs), 44 cellular components (CCs), and 82 molecular functions (MFs). Moreover, the functions of candidate genes were associated with nucleic acid transport, nucleocytoplasmic transport, protein sumoylation, and so on ( Fig. 1d ). Concerning the 34 KEGG pathways involved by candidate genes, they included not only cytoplasmic transport but also nucleocytoplasmic transport, transcriptional misregulation in cancer, chronic myeloid leukemia, etc ( Fig. 1e ). Generally, 107 candidate genes might participate in nucleic acid and cytoplasmic transport and cancer regulation. Additionally, the PPI-based screening manifested that the interactions among 15 core genes, PARP1 , TRIM28 , NPM1 , RANGAP1 , SUMO3 , MDM2 , NCOR2 , PPARG , TP53 , SUMO2 , CREBBP , HDAC1 , EP300 , SUMO1 and PIAS1 were close at protein level, especially among TP53, EP300 and SUMO1 ( Fig. 1f ). 3.2 The performance of the biomarker-based nomogram model in predicting the risk of AD was reliable There were five feature genes ( CREBBP , PIAS1 , TRIM28 , MAD2 , and NCOR2 ) in the SVM-RFE model when the model had the highest accuracy ( Fig. 2a ), of which CREBBP, PIAS1 and TRIM28 were served as biomarkers for AD as a consequence of their observably increased expression levels in AD group and consistent expression trends in GSE140831 and GSE63060 datasets (p<0.05) ( Fig. 2b-c ). The nomogram model created by absorbing three biomarkers demonstrated that the risk of AD increased with the total points corresponding to each sample in the GSE140831 dataset ( Fig. 2d ). The precise predictive performance of the nomogram model must be further confirmed by the AUC value of 0.809 ( Fig. 2e ). 3.3 Three biomarkers were involved in identical pathways, such as olfactory transduction, lysosome, and spliceosome Exploring three biomarkers by GSEA was essential for understanding their effects on AD. The findings of GSEA unveiled that CREBBP , PIAS1 , and TRIM28 gathered in olfactory transduction, lysosome, and spliceosome were significantly and synchronously. Besides, Huntington's disease and neuroactive ligand receptor interaction were involved by PIAS1 and TRIM28 . In conclusion, GSEA certifies that three biomarkers might retain similar functions in regulating AD's progression ( Fig. 3a-c ). 3.4 The expression of three biomarkers was regulated by multiple TFs and miRNAs How upstream molecules regulated three biomarkers was further explored; 58 TFs and 284 miRNAs targeted three. The TF-mRNA displayed that three biomarkers could be regulated by EGR1 , DMRT1 , E2F1 , CREM , and ZFX simultaneously ( Fig. 3d ). In addition, as presented in Fig. 3e , the transcription of TRIM28 and CREBBP could be regulated by hsa-miR-6812-5p, hsa-miR-1227-5p, and hsa-miR-7113-5p synchronously, as well as CREBBP and PIAS1 by hsa-miR-4452, hsa-miR-1284, hsa-miR-6883-3p, and so forth. 3.5 Most of the immune cells might influence the progression of AD Among the 64 immune cells, a total of 55 immune cells showed markedly discrepant abundance between the AD and control groups (p<0.05). For instance, AD's percentages of basophils, class-switched memory B cells, myocytes, and mesenchymal stem cells (MSCs) were higher. In contrast, the proportions of natural killer T (NKT) cells, B cells, neutrophils, smooth muscle, and naive CD8(+) T cells were inverse in AD ( Fig. 4a ). Furthermore, Spearman's correlation analysis among discrepant immune cells uncovered that memory B cells and B cells maintained the highest positive correlations of 0.76 (p<0.05), yet T helper type 1 (Th1) cells and neurons kept the highest negative correlations of -0.72 (p<0.05) ( Fig. 4b ). Basophils were positively correlated with PIAS1 and CREBBP , and myocytes with TRIM28 . Besides neutrophils and CREBBP , PIAS1 maintained significantly negative correlations, as well as smooth muscle and TRIM28 (|cor|>0.4, p<0.05) ( Fig. 4c-e ). 3.6 A sum of 21 alternative drugs might be developed for AD therapy What is more, a total of 21 drugs targeting CREBBP were retrieved in DGIdb, such as triazolam, colchicine, alprazolam, nocodazole, etc. ( Fig. 4f ). Unfortunately, no drugs targeting TRIM28 and PIAS1 were found in this database. Among these, triazolam and alprazolam were proven to be therapeutic for AD. However, applying them in large doses could produce side effects such as decreased blood pressure and respiratory circulation suppression, so drug prediction targeting biomarkers might provide data for developing available drugs for AD. 3.7 The expression level of TRIM28 was significantly increased in svPPA The expression patterns of three biomarkers in five dementia-related diseases were exhibited in the heat map ( Fig. 5a ). Expression analysis of biomarkers confirmed that CREBBP and PIAS1 showed no significant discrepancies between bvFTD, svPPA, nfvPPA, PSP, and CBS and controls, except for TRIM28 with upregulated expression levels in svPPA ( Fig. 5b-f ). 3.8 CREBBP and TRIM28 were all over-expressed in AD samples Consistent with the expression trend in the GSE140831 and GSE63060 datasets, the relative expression levels of CREBBP (p = 0.0043) and TRIM28 (p = 0.0032) were increased in AD samples. While the expression level of PIAS1 was not significant (p > 0.05) ( Fig. 6a-c ). 4 Discussion Alzheimer's disease (AD) is a degenerative disease of the nervous system. whose primary genetic risk factor is the APOE gene and is characterized by insoluble aggregation of proteins such as amyloid-beta (Aβ) [ 24 ]. As A post-translational modification, SUMOylation has been shown to regulate multiple cellular pathways. SUMO is believed to be closely related to key proteins in AD, such as Aβ and the microtubule-associated protein tau, and is involved in regulating various cellular functions associated with AD pathology [ 25 ]. In addition, the SUMO pathway is also associated with AD's occurrence and development by affecting the telomere structure's integrity, which can provide a new research perspective for diagnosing and treating AD [ 26 , 27 ]. However, bioinformatic analysis of SRGs in AD has not yet been performed, and crucial SRG-based biomarkers involved in the pathogenesis of AD have not been confirmed. In the present study, we identified the SRGs associated biomarkers in AD based on the GSE140831, GSE63060, and GSE140830 datasets, specifically CREBBP , PIAS1 , and TRIM28 . These biomarkers are potential diagnostic targets for AD because of their increased expression levels in the AD group and reliable predictive performance. GSEA and the prediction results of TFs suggest that these biomarkers may have similar biological functions in AD’s progression, and the expression of three biomarkers could be regulated simultaneously by some TFs. Immune infiltration analysis suggested that, between the AD and control groups, the proportion of specific immune cells correlated significantly with these biomarkers. Finally, we speculate a total of 21 drugs as potential therapeutic agents for AD on DGIdb, in addition to triazolam and alprazolam, which have been used in the treatment of AD. CREBBP is a transcriptional coactivator with histone acetyltrans-ferase (HAT) activity and is involved in post-translational modification by acetylation of histones [ 28 ], there are studies showed that dysregulation of histone acetylation pathways is thought to play may play a key role in the disease progression of AD. Hence, dysfunction of CREBBP is likely to contribute to neurodegenerative diseases [ 29 , 30 ]. In addition, recent studies have shown that CREBBP may be a hub gene that provides new insight into treating AD by mediating the metabolism of niacin and lysosome [ 31 , 32 ]. PIAS1 , a protein inhibitor of the activated STAT1 , is one of E3 the E3-type SUMO ligases. It plays important roles in the STAT pathway, which is essential to inflammatory activation in AD [ 33 ]. Besides that, PIAS1 is thought to play a neuroprotective role against Aβ toxicity in AD [ 34 , 35 ]. TRIM28 has also been demonstrated to be an E3 SUMOylation ligase [ 36 ]. Maxime et al found that TRIM28 regulates the steady state levels of a-Synuclein (a-Syn) and tau, which are supposed to be neurodegeneration-driving proteins in AD by mediating the SUMOylation [ 37 , 38 ]. GSEA showed that lysosome, olfactory transduction, and spliceosome are the pathways involved in these three biomarkers. Studies have shown that TRIM28 inhibits autophagy degradation by inducing SUMOylation of BCR-ABL [ 39 ]. Lysosomes are a significant cell signaling center, influencing fundamental processes such as cell membrane repair, energy metabolism, and inflammatory pathways [ 40 ]. The autophagy-lysosome pathway is closely related to protein homeostasis and other processes in AD [ 41 ]. Therefore, it is reasonable to speculate that the biomarker TRIM28 may affect AD by participating in the regulation of the lysosomal pathway. Olfactory dysfunction often occurs in the early stages of AD, studies have found that the olfactory mucosal cells from AD patients secrete an increased amount of Aβ1–42, and the ratio of secretion of Aβ1–42 / Aβ1–40 also increases [ 42 ]. Splicing contributes to the diversity and function of neuronal transcription; the disruption of the splicing mechanism leads to neurological diseases, including AD [ 43 ]. Yi-Chen et al supposed that the interaction of tau spliceosomes disrupts the function of snRNP, resulting in loss of transcriptome fidelity and splicing errors, and leading to neurodegenerative changes in AD ultimately [ 44 ]. The TF-mRNA indicated that EGR1 , DMRT1 , CREM , ZFX , and E2F1 could simultaneously regulate three biomarkers. As a member of immediate early genes (IEGs), EGR1 is a mediator and regulator of synaptic plasticity and neuronal activity under physiological conditions and in pathological conditions [ 45 ], and has recently been considered a candidate gene for schizophrenia. CREM and CREB are members of the same family of transcription factors, and CREB may have a significant effect on forming and consolidating of memory by interacting with CREBBP [ 46 ]. ZFX is a zinc finger protein of the Zfy family. David found that ZFX is a repressor of core and linker histones [ 47 ]. E2F1 is a critical transcription factor regulating cell cycle, which is primarily considered to be a key factor in cell proliferation and cancer. Still, recent studies have suggested that it may also act as an important part in neurodegenerative diseases [ 48 ]. The expression levels of E2F-1 in the brain tissue of AD model rats had changed, suggesting that E2F-1 may be involved in neuronal death and loss of function by regulating cell cycle and apoptosis in AD [ 49 ]. These transcription factors associated with CREBBP , PIAS1 , and TRIM28 may mediate the development of AD by regulating histone acetylation, synaptic plasticity, and the cell cycle. However, the mechanisms of these effects still need to be further explored. Neuroimmune interactions are a significant focus of research into neurodegenerative diseases, including AD, recently [ 50 ]. Our study found that in AD, the proportion of specific immune cells increased, such as basophils, class-switching memory B cells, muscle cells. In contrast, the proportion of specific immune cells decreased, such as natural killer T cells (NKT), B cells, neutrophils, smooth muscle, and naive CD8(+) T cells. The results are similar to previous studies [ 51 , 52 ]. In addition, among these immune cells, basophils were positively correlated with PIAS1 and CREBBP , myocytes were positively associated with TRIM28 , and there was a significant negative correlation between neutrophils with CREBBP and PIAS1 , as well as smooth muscle and TRIM28 . Basophils are a subset of granulocytes involved in various immunomodulatory and inflammatory processes. A study found that the number of basophils increased significantly in AD samples [ 53 ]. This is consistent with our findings, so we could speculate that PIAS1 and CREBB may regulate the inflammatory response in AD by affecting neutrophils. However, Shad found blood levels of basophils were constantly low in AD [ 54 ], so the role of basophils in AD is still under more research. Similar to basophils, other types of cells have a complex relationship with this neurodegenerative disease. For example, myocytes also play an essential role in multiple physiological processes closely related to AD. Myocytes are mainly involved in muscle function, motor control, and energy metabolism. Cerebral small vessel disease (CSVD) is characterize with depletion of vascular myocytes, so vascular myocytes are considered as central to brain aging [ 55 ], and there is evidence that myocytes are the source of amyloid deposits in the meninges and cortical blood vessels [ 56 ], which may affect the progression of AD. SO TRIM28 may participate in the pathological process of AD by affecting myocyte function and further increasing amyloid deposition in the brain. It is worth to noting that recent studies have suggested that blood-brain barrier (BBB) dysfunction is correlated with human cognitive impairment including the early stages of AD [ 57 ]. BBB is made up of various cell including smooth muscle cells, besides that, it has been suggested that neutrophils may be involved in the early development of AD by mediating BBB damage, intravascular adhesion, and invasion of the central nervous system [ 58 ]. Therefore, we speculated that these three biomarkers can mediates peripheral or central inflammatory processes and affect the integrity of BBB by affecting the function of immune cells, thereby contribute to the occurrence and development of AD. Finally, we identified the potential therapeutic drugs targeting CREBBP through the DGIdb database, totally 21 drugs were screened including colchicine, glucocorticoid, benzodiazepines an so on, among which triazolam and alprazolam have been used to improve behavioral and psychological symptoms of AD [ 59 ]. As we know, although cholinesterase inhibitors, which designed to increase the levels of acetylcholine (ACh) in the brain, are the main drugs for AD, there are no therapeutic drugs with definite efficacy for AD [ 60 ]. Interestingly, Keisuke et al recently founded that triazolam could inhibit rhAChE activity by ≥ 20% [ 61 ], which may provide new evidence for the use of benzodiazepines in the treatment of AD. In a word, our drug predictions based on these biomarkers, particularly those related to CREBBP , can provide valuable data for the exploit of more effective and safer AD drugs, but further research involving animal models and clinical trials is required to assess the safety and efficacy of these drugs. In conclusion, we identified CREBBP , PIAS1 and TRIM28 as SUMO-SRGs-based biomarkers diagnosing AD through bioinformatics analysis firstly in this study, furthermore, we revealed the regulatory networks and identified potential therapeutic agents aimed at these biomarkers. These genes may be involved in the onset and progression of AD, they have potential to be considered as novel biomarkers for AD's diagnosing and progression monitoring. Our study contributes to clarifying the the relationship between SUMO and the pathological process of AD and also offers insights into exploring novel treatment strategies for AD. Of course, there were some limitations in our study. Firstly, the sample size for qRT-PCR analysis is relatively insufficient, expanding the sample size is needed to verify our findings. Secondly, the diagnostic efficacy and mechanism by which these potential AD-related SUMO-SRGs-based biomarkers must be substantiated by further research and laboratory validation. Declarations Funding This study was financially supported by the the Intramural Research Project of Fuzhou Second General Hospital, China (Award Number: 2024ZY04) and the Natural Science Foundation of Fujian Province, China (Award Number: 2023J011519) . Competing interests The authors declare no competing interests. Author Contributions All authors contributed to the study conception and design. Peng Chen and Mengting Fan contributed equally to this work. Conceptualization:Peng Chen, Mengting Fan, Jing Xu; Methodology: Peng Chen; Formal analysis and investigation: Peng Chen; Writing - original draft preparation: Mengting Fan; Writing - review and editing: Peng Chen, Mengting Fan, Hailiang Lin, Zhong Chen, Qiuyang Zhang, Yunfan Cheng; Funding acquisition: Qiuyang Zhang and Jing Xu. Data availability statements The datasets analysed during the current study are available in the [GEO] repository, [https://www.ncbi.nlm.nih.gov/geo/, reference number GSE140831 (GPL15988), GSE63060 (GPL6947), GSE140830 (GPL15988)]. and [MSigDB] repository, [ https://www.gsea-msigdb.org/]. Ethics approval and consent to participate This study was conducted in accordance with the Declaration of Helsinki. Ethical approval was obtained from the Ethics Committee of Fuzhou Second General Hospital, with the approval number 2024005 and the approval date February 29, 2024. All patients provided written informed consent when their clinical samples were used for RT-qPCR experiments, ensuring that the research process complied with ethical standards and fully respected the rights and wishes of the patients. Consent to Publish N/A References Arvanitakis Z, Shah RC, Bennett DA (2019) Diagnosis and Management of Dementia: Review. Jama. 322:1589-1599. https://doi.org/10.1001/jama.2019.4782. Scheltens P, De Strooper B, Kivipelto M, et al. (2021) Alzheimer's disease. Lancet. 397:1577-1590. https://doi.org/10.1016/s0140-6736(20)32205-4. Barbe C, Jolly D, Morrone I, et al. (2018) Factors associated with quality of life in patients with Alzheimer's disease. BMC Geriatr. 18:159. https://doi.org/10.1186/s12877-018-0855-7. van der Lee SJ, Wolters FJ, Ikram MK, et al. (2018) The effect of APOE and other common genetic variants on the onset of Alzheimer's disease and dementia: a community-based cohort study. Lancet Neurol. 17:434-444. https://doi.org/10.1016/s1474-4422(18)30053-x. Rostagno AA (2022) Pathogenesis of Alzheimer's Disease. Int J Mol Sci. 24:107. https://doi.org/10.3390/ijms24010107. Passeri E, Elkhoury K, Morsink M, et al. (2022) Alzheimer's Disease: Treatment Strategies and Their Limitations. Int J Mol Sci. 23:13954. https://doi.org/10.3390/ijms232213954. Martins WC, Tasca CI, Cimarosti H (2016) Battling Alzheimer's Disease: Targeting SUMOylation-Mediated Pathways. Neurochem Res. 41:568-578. https://doi.org/10.1007/s11064-015-1681-3. Grabbe C, Husnjak K, Dikic I (2011) The spatial and temporal organization of ubiquitin networks. Nat Rev Mol Cell Biol. 12:295-307. https://doi.org/10.1038/nrm3099. Nortley R, Korte N, Izquierdo P, et al. (2019) Amyloid β oligomers constrict human capillaries in Alzheimer's disease via signaling to pericytes. Science. 365:eaav9518. https://doi.org/10.1126/science.aav9518. Lee L, Sakurai M, Matsuzaki S, et al. (2013) SUMO and Alzheimer's disease. Neuromolecular Med. 15:720-736. https://doi.org/10.1007/s12017-013-8257-7. Kuan XY, Fauzi NSA, Ng KY, et al. (2023) Exploring the Causal Relationship Between Telomere Biology and Alzheimer's Disease. Mol Neurobiol. 60:4169-4183. https://doi.org/10.1007/s12035-023-03337-4. Zalzman M, Meltzer WA, Portney BA, et al. (2020) The Role of Ubiquitination and SUMOylation in Telomere Biology. Curr Issues Mol Biol. 35:85-98. https://doi.org/10.21775/cimb.035.085. Chang B, Hu Z, Chen L, et al. (2023) Development and validation of cuproptosis-related genes in synovitis during osteoarthritis progress. Front Immunol. 14:1090596. https://doi.org/10.3389/fimmu.2023.1090596. Ritchie ME, Phipson B, Wu D, et al. (2015) limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43:e47. https://doi.org/10.1093/nar/gkv007. Pelikan A, Herzel H, Kramer A, et al. (2022) Venn diagram analysis overestimates the extent of circadian rhythm reprogramming. Febs j. 289:6605-6621. https://doi.org/10.1111/febs.16095. Wu T, Hu E, Xu S, et al. (2021) clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb). 2:100141. https://doi.org/10.1016/j.xinn.2021.100141. Shannon P, Markiel A, Ozier O, et al. (2003) Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 13:2498-2504. https://doi.org/10.1101/gr.1239303. Hu X, Wang J, Ju Y, et al. (2022) Combining metabolome and clinical indicators with machine learning provides some promising diagnostic markers to precisely detect smear-positive/negative pulmonary tuberculosis. BMC Infect Dis. 22:707. https://doi.org/10.1186/s12879-022-07694-8. Liu TT, Li R, Huo C, et al. (2021) Identification of CDK2-Related Immune Forecast Model and ceRNA in Lung Adenocarcinoma, a Pan-Cancer Analysis. Front Cell Dev Biol. 9:682002. https://doi.org/10.3389/fcell.2021.682002. Liu C, Fang J, Kang W, et al. (2023) Identification of novel potential homologous repair deficiency-associated genes in pancreatic adenocarcinoma via WGCNA coexpression network analysis and machine learning. Cell Cycle. 22:2392-2408. https://doi.org/10.1080/15384101.2023.2293594. Goncharova IA, Nazarenko MS, Babushkina NP, et al. (2020) [Genetic Predisposition to Early Myocardial Infarction]. Mol Biol (Mosk). 54:224-232. https://doi.org/10.31857/s0026898420020044. Zhuang X, Schlunck G, Wolf J, et al. (2022) Time- and Stimulus-Dependent Characteristics of Innate Immune Cells in Organ-Cultured Human Corneal Tissue. J Innate Immun. 14:98-111. https://doi.org/10.1159/000516669. Liu S, Xie X, Lei H, et al. (2019) Identification of Key circRNAs/lncRNAs/miRNAs/mRNAs and Pathways in Preeclampsia Using Bioinformatics Analysis. Med Sci Monit. 25:1679-1693. https://doi.org/10.12659/msm.912801. Mangialasche F, Solomon A, Winblad B, et al. (2010) Alzheimer's disease: clinical trials and drug development. Lancet Neurol. 9:702-716. https://doi.org/10.1016/s1474-4422(10)70119-8. Hoppe JB, Salbego CG, Cimarosti H (2015) SUMOylation: Novel Neuroprotective Approach for Alzheimer's Disease? Aging Dis. 6:322-330. https://doi.org/10.14336/ad.2014.1205. Se Thoe E, Fauzi A, Tang YQ, et al. (2021) A review on advances of treatment modalities for Alzheimer's disease. Life Sci. 276:119129. https://doi.org/10.1016/j.lfs.2021.119129. Cummings JL, Tong G, Ballard C (2019) Treatment Combinations for Alzheimer's Disease: Current and Future Pharmacotherapy Options. J Alzheimers Dis. 67:779-794. https://doi.org/10.3233/jad-180766. Goodman RH, Smolik S (2000) CBP/p300 in cell growth, transformation, and development. Genes Dev. 14:1553-1577. Schueller E, Paiva I, Blanc F, et al. (2020) Dysregulation of histone acetylation pathways in hippocampus and frontal cortex of Alzheimer's disease patients. Eur Neuropsychopharmacol. 33:101-116. https://doi.org/10.1016/j.euroneuro.2020.01.015. Jeong H, Then F, Melia TJ, Jr., et al. (2009) Acetylation targets mutant huntingtin to autophagosomes for degradation. Cell. 137:60-72. https://doi.org/10.1016/j.cell.2009.03.018. Li T, Yin L, Kang X, et al. (2022) TFEB acetylation promotes lysosome biogenesis and ameliorates Alzheimer's disease-relevant phenotypes in mice. J Biol Chem. 298:102649. https://doi.org/10.1016/j.jbc.2022.102649. Wang Z, Zou Z, Li Q (2022) Nicotinic acid supplementation contributes to the amelioration of Alzheimer's disease in mouse models. Ann Transl Med. 10:1049. https://doi.org/10.21037/atm-22-1148. He K, Zhang J, Liu J, et al. (2021) Functional genomics study of protein inhibitor of activated STAT1 in mouse hippocampal neuronal cells revealed by RNA sequencing. Aging (Albany NY). 13:9011-9027. https://doi.org/10.18632/aging.202749. Liu SY, Ma YL, Hsu WL, et al. (2019) Protein inhibitor of activated STAT1 Ser(503) phosphorylation-mediated Elk-1 SUMOylation promotes neuronal survival in APP/PS1 mice. Br J Pharmacol. 176:1793-1810. https://doi.org/10.1111/bph.14656. Tao CC, Hsu WL, Ma YL, et al. (2017) Epigenetic regulation of HDAC1 SUMOylation as an endogenous neuroprotection against Aβ toxicity in a mouse model of Alzheimer's disease. Cell Death Differ. 24:597-614. https://doi.org/10.1038/cdd.2016.161. Neo SH, Itahana Y, Alagu J, et al. (2015) TRIM28 Is an E3 Ligase for ARF-Mediated NPM1/B23 SUMOylation That Represses Centrosome Amplification. Mol Cell Biol. 35:2851-2863. https://doi.org/10.1128/mcb.01064-14. Rousseaux MW, Revelli JP, Vázquez-Vélez GE, et al. (2018) Depleting Trim28 in adult mice is well tolerated and reduces levels of α-synuclein and tau. Elife. 7:e36768. https://doi.org/10.7554/eLife.36768. Rousseaux MW, de Haro M, Lasagna-Reeves CA, et al. (2016) TRIM28 regulates the nuclear accumulation and toxicity of both alpha-synuclein and tau. Elife. 5:e19809. https://doi.org/10.7554/eLife.19809. Liu W, Zhu Y, Ye W, et al. (2025) Redox regulation of TRIM28 facilitates neuronal ferroptosis by promoting SUMOylation and inhibiting OPTN-selective autophagic degradation of ACSL4. Cell Death Differ. https://doi.org/10.1038/s41418-025-01452-4. Lawrence RE, Zoncu R (2019) The lysosome as a cellular centre for signalling, metabolism and quality control. Nat Cell Biol. 21:133-142. https://doi.org/10.1038/s41556-018-0244-7. Udayar V, Chen Y, Sidransky E, et al. (2022) Lysosomal dysfunction in neurodegeneration: emerging concepts and methods. Trends Neurosci. 45:184-199. https://doi.org/10.1016/j.tins.2021.12.004. Lampinen R, Fazaludeen MF, Avesani S, et al. (2022) Single-Cell RNA-Seq Analysis of Olfactory Mucosal Cells of Alzheimer's Disease Patients. Cells. 11:676. https://doi.org/10.3390/cells11040676. Tollervey JR, Wang Z, Hortobágyi T, et al. (2011) Analysis of alternative splicing associated with aging and neurodegeneration in the human brain. Genome Res. 21:1572-1582. https://doi.org/10.1101/gr.122226.111. Hsieh YC, Guo C, Yalamanchili HK, et al. (2019) Tau-Mediated Disruption of the Spliceosome Triggers Cryptic RNA Splicing and Neurodegeneration in Alzheimer's Disease. Cell Rep. 29:301-316.e310. https://doi.org/10.1016/j.celrep.2019.08.104. Duclot F, Kabbaj M (2017) The Role of Early Growth Response 1 (EGR1) in Brain Plasticity and Neuropsychiatric Disorders. Front Behav Neurosci. 11:35. https://doi.org/10.3389/fnbeh.2017.00035. Bartolotti N, Lazarov O (2019) CREB signals as PBMC-based biomarkers of cognitive dysfunction: A novel perspective of the brain-immune axis. Brain Behav Immun. 78:9-20. https://doi.org/10.1016/j.bbi.2019.01.004. Gokhman D, Livyatan I, Sailaja BS, et al. (2013) Multilayered chromatin analysis reveals E2f, Smad and Zfx as transcriptional regulators of histones. Nat Struct Mol Biol. 20:119-126. https://doi.org/10.1038/nsmb.2448. Verdaguer E, Susana Gde A, Clemens A, et al. (2007) Implication of the transcription factor E2F-1 in the modulation of neuronal apoptosis. Biomed Pharmacother. 61:390-399. https://doi.org/10.1016/j.biopha.2006.11.001. Tuerxun M, Muhda A, Yin L (2021) The molecular mechanisms of signal pathway activating effect of E2F-1/NF-κB/GSK-3β on cognitive dysfunction of Alzheimer rats. Bioengineered. 12:10000-10008. https://doi.org/10.1080/21655979.2021.1989261. Jorfi M, Maaser-Hecker A, Tanzi RE (2023) The neuroimmune axis of Alzheimer's disease. Genome Med. 15:6. https://doi.org/10.1186/s13073-023-01155-w. Zhuang X, Zhang G, Bao M, et al. (2023) Development of a novel immune infiltration-related diagnostic model for Alzheimer's disease using bioinformatic strategies. Front Immunol. 14:1147501. https://doi.org/10.3389/fimmu.2023.1147501. Lu Y, Li K, Hu Y, et al. (2021) Expression of Immune Related Genes and Possible Regulatory Mechanisms in Alzheimer's Disease. Front Immunol. 12:768966. https://doi.org/10.3389/fimmu.2021.768966. Johnson MD, Blitman J (2020) Basophilic invasion in the neurohypophysis is increased in patients with Alzheimer's disease. Clin Neuropathol. 39:271-274. https://doi.org/10.5414/np301255. Shad KF, Aghazadeh Y, Ahmad S, et al. (2013) Peripheral markers of Alzheimer's disease: surveillance of white blood cells. Synapse. 67:541-543. https://doi.org/10.1002/syn.21651. Ahmed-Jushuf F, Jiwa NS, Arwani AS, et al. (2016) Age-dependent expression of VEGFR2 in deep brain arteries in small vessel disease, CADASIL, and healthy brains. Neurobiol Aging. 42:110-115. https://doi.org/10.1016/j.neurobiolaging.2016.03.002. Wisniewski HM, Wegiel J, Kotula L (1996) Review. David Oppenheimer Memorial Lecture 1995: Some neuropathological aspects of Alzheimer's disease and its relevance to other disciplines. Neuropathol Appl Neurobiol. 22:3-11. https://doi.org/10.1111/j.1365-2990.1996.tb00839.x. van de Haar HJ, Burgmans S, Jansen JF, et al. (2017) Blood-Brain Barrier Leakage in Patients with Early Alzheimer Disease. Radiology. 282:615. https://doi.org/10.1148/radiol.2017164043. Zenaro E, Pietronigro E, Della Bianca V, et al. (2015) Neutrophils promote Alzheimer's disease-like pathology and cognitive decline via LFA-1 integrin. Nat Med. 21:880-886. https://doi.org/10.1038/nm.3913. Wang F, Feng TY, Yang S, et al. (2016) Drug Therapy for Behavioral and Psychological Symptoms of Dementia. Curr Neuropharmacol. 14:307-313. https://doi.org/10.2174/1570159x14666151208114232. Lane CA, Hardy J, Schott JM (2018) Alzheimer's disease. Eur J Neurol. 25:59-70. https://doi.org/10.1111/ene.13439. Obara K, Mori H, Ihara S, et al. (2024) Inhibitory Actions of Antidepressants, Hypnotics, and Anxiolytics on Recombinant Human Acetylcholinesterase Activity. Biol Pharm Bull. 47:328-333. https://doi.org/10.1248/bpb.b23-00719. Additional Declarations No competing interests reported. Supplementary Files SupplementaryInformation.pdf Cite Share Download PDF Status: Published Journal Publication published 20 Apr, 2026 Read the published version in Scientific Reports → Version 1 posted Editorial decision: Revision requested 15 Dec, 2025 Reviews received at journal 01 Dec, 2025 Reviewers agreed at journal 23 Nov, 2025 Reviews received at journal 06 Nov, 2025 Reviewers agreed at journal 16 Oct, 2025 Reviewers agreed at journal 16 Oct, 2025 Reviewers invited by journal 14 Oct, 2025 Editor assigned by journal 08 Oct, 2025 Editor invited by journal 08 Jul, 2025 Submission checks completed at journal 02 Jul, 2025 First submitted to journal 02 Jul, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-6983921","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":534369865,"identity":"e2a6357c-8ae2-4166-ad5e-d506b5ebaba4","order_by":0,"name":"Peng Chen","email":"","orcid":"","institution":"Fuzhou Second General Hospital","correspondingAuthor":false,"prefix":"","firstName":"Peng","middleName":"","lastName":"Chen","suffix":""},{"id":534369866,"identity":"8bb7ffc3-cfa1-4eb6-9245-b73648286672","order_by":1,"name":"Mengting Fan","email":"","orcid":"","institution":"Fuzhou Second General Hospital","correspondingAuthor":false,"prefix":"","firstName":"Mengting","middleName":"","lastName":"Fan","suffix":""},{"id":534369867,"identity":"93ea7e41-97bb-4ac6-8468-f94af0cf3700","order_by":2,"name":"Hailiang Lin","email":"","orcid":"","institution":"Fuzhou Second General Hospital","correspondingAuthor":false,"prefix":"","firstName":"Hailiang","middleName":"","lastName":"Lin","suffix":""},{"id":534369868,"identity":"4898f8c3-0796-4bbb-a11d-6997b45a3726","order_by":3,"name":"Zhong Chen","email":"","orcid":"","institution":"Fuzhou Second General Hospital","correspondingAuthor":false,"prefix":"","firstName":"Zhong","middleName":"","lastName":"Chen","suffix":""},{"id":534369869,"identity":"fc86ec6f-1fb9-4ab3-bd65-0cfea5c3055c","order_by":4,"name":"Qiuyang Zhang","email":"","orcid":"","institution":"Fuzhou Second General Hospital","correspondingAuthor":false,"prefix":"","firstName":"Qiuyang","middleName":"","lastName":"Zhang","suffix":""},{"id":534369870,"identity":"94594008-cfdd-477c-be2d-ab1558952512","order_by":5,"name":"Yunfan Cheng","email":"","orcid":"","institution":"Fuzhou Second General Hospital","correspondingAuthor":false,"prefix":"","firstName":"Yunfan","middleName":"","lastName":"Cheng","suffix":""},{"id":534369871,"identity":"a2c30275-ca6c-4aad-a4de-d436d61c8fba","order_by":6,"name":"Jing Xu","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA9UlEQVRIiWNgGAWjYHCChAMMDEDE3gBkV0jIyROvhQeIGc5YGBs2EGcTULmEAwMDY1tFIoiNFxjcSHh4uODXHTkGCd6HjwvnSSQwNjA/fHQDjxbJGQkJh2f2PTNmkG43Np65TSKPnYHN2DgHjxZ+CaAW3p7DifvvHGOT5t0mUczYwMMmjU8LG1RLfYNEGvtv3jkSiQ0HCGgB28Lz43ACg0QaGzNvAxFaJHseAG1pOGzYwHOMWZrnmISxYTMBvxgcz0n+zPPnsDwDexvjZ56aOjl59uaHj/FpAUZhAjA6kAWY8SoHAfYDDAx/CKoaBaNgFIyCkQwASeZMSxK0GeUAAAAASUVORK5CYII=","orcid":"","institution":"Fuzhou Second General Hospital","correspondingAuthor":true,"prefix":"","firstName":"Jing","middleName":"","lastName":"Xu","suffix":""}],"badges":[],"createdAt":"2025-06-26 13:38:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-6983921/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-6983921/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1038/s41598-026-49884-3","type":"published","date":"2026-04-20T15:59:03+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":94587773,"identity":"bb5c34d8-0398-4b9a-ae3e-6add22e09b23","added_by":"auto","created_at":"2025-10-28 18:18:41","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":26475845,"visible":true,"origin":"","legend":"","description":"","filename":"Manuscript.docx","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/f250ebafb168523f6e782d3e.docx"},{"id":94588034,"identity":"e08fc2c5-a258-467e-865f-552f08dc53ea","added_by":"auto","created_at":"2025-10-28 18:18:57","extension":"json","order_by":1,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":8819,"visible":true,"origin":"","legend":"","description":"","filename":"8eba0228606343a1a180d6d9b2bbd8ad.json","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/4219bd35ab6cf848d49c1226.json"},{"id":94587787,"identity":"bc213336-5a58-4b92-ac06-829b0978ecc6","added_by":"auto","created_at":"2025-10-28 18:18:46","extension":"pdf","order_by":2,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":79500,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/6e16b0ba0163e0b6610c6178.pdf"},{"id":94587755,"identity":"41d6d698-650f-41c8-bea2-e5f9d4d0f814","added_by":"auto","created_at":"2025-10-28 18:18:39","extension":"xml","order_by":3,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":131898,"visible":true,"origin":"","legend":"","description":"","filename":"8eba0228606343a1a180d6d9b2bbd8ad1enriched.xml","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/9d21e917b647ecda8ef9bd61.xml"},{"id":94588488,"identity":"377d0679-1094-40c2-a85a-cffa24818c70","added_by":"auto","created_at":"2025-10-28 18:19:26","extension":"jpeg","order_by":4,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5518714,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/b3ebb208bc3334f91cb6ce25.jpeg"},{"id":94587607,"identity":"58bacc84-7c6a-47bb-9047-f40d34da1019","added_by":"auto","created_at":"2025-10-28 18:18:26","extension":"jpeg","order_by":5,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":3760224,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/dcbc36ef05adbe5d109eea85.jpeg"},{"id":94588498,"identity":"0e04c83c-b4f7-4ffb-ae4d-e4a86ed57ae2","added_by":"auto","created_at":"2025-10-28 18:19:26","extension":"jpeg","order_by":6,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5180692,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/2cb3933a54f5417f244648a7.jpeg"},{"id":94587750,"identity":"843236fb-0562-435e-aaf9-28fd9d865dfe","added_by":"auto","created_at":"2025-10-28 18:18:38","extension":"jpeg","order_by":7,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5180692,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/bf5f0ab641135916600cade3.jpeg"},{"id":94588321,"identity":"2d618ca0-b6b5-4a47-b13c-d4a715e9043f","added_by":"auto","created_at":"2025-10-28 18:19:11","extension":"jpeg","order_by":8,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":5180692,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage5.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/ce77693a7523bf29ac409943.jpeg"},{"id":94588163,"identity":"e213119e-25fc-4c76-b6ae-3aa20eb59d6e","added_by":"auto","created_at":"2025-10-28 18:19:04","extension":"jpeg","order_by":9,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":1523202,"visible":true,"origin":"","legend":"","description":"","filename":"floatimage6.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/35107628428879809569a049.jpeg"},{"id":94587758,"identity":"96dfe0bc-b723-4857-91be-06377a1bc167","added_by":"auto","created_at":"2025-10-28 18:18:40","extension":"png","order_by":10,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":103568,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/da6535894f2df129a56b248a.png"},{"id":94588077,"identity":"0a135721-d8c0-4535-ac3f-7d09e21b81c2","added_by":"auto","created_at":"2025-10-28 18:18:59","extension":"png","order_by":11,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":39556,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/d12eff240d0f607ab3a1af93.png"},{"id":94588385,"identity":"6e03d7e6-ac92-47b5-af7e-5f9ebf46140f","added_by":"auto","created_at":"2025-10-28 18:19:12","extension":"png","order_by":12,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":118656,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/48c3c3cedccfd9bebc7fa9fc.png"},{"id":94587354,"identity":"6ca779e4-403d-45f0-b455-bab3c0d045e0","added_by":"auto","created_at":"2025-10-28 18:18:09","extension":"png","order_by":13,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":150256,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/a5a735bbf35201cd291e4f46.png"},{"id":94588490,"identity":"4e6c9d1a-eeb7-427f-8fb5-32a7815c677a","added_by":"auto","created_at":"2025-10-28 18:19:26","extension":"png","order_by":14,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":29730,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/63969911e71eae5bfa2b512a.png"},{"id":94587760,"identity":"6faa01ee-f180-4262-b8fa-fefc97a7c1bb","added_by":"auto","created_at":"2025-10-28 18:18:40","extension":"png","order_by":15,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":22280,"visible":true,"origin":"","legend":"","description":"","filename":"Onlinefloatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/82859431ca536eeae091c32d.png"},{"id":94588560,"identity":"23598900-74d8-4e19-ba17-ee6814fa717a","added_by":"auto","created_at":"2025-10-28 18:19:30","extension":"xml","order_by":16,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":130198,"visible":true,"origin":"","legend":"","description":"","filename":"8eba0228606343a1a180d6d9b2bbd8ad1structuring.xml","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/733916e8b57132c755cb5822.xml"},{"id":94588080,"identity":"33515838-b49d-42cf-a2c6-dc00366b719a","added_by":"auto","created_at":"2025-10-28 18:19:00","extension":"html","order_by":17,"title":"","display":"","copyAsset":false,"role":"acdc-reference","size":148521,"visible":true,"origin":"","legend":"","description":"","filename":"earlyproof.html","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/4d01c5fcfbca75b1c83fffe6.html"},{"id":94588033,"identity":"b1b95afd-4bac-4bfa-be4b-8cef463e91af","added_by":"auto","created_at":"2025-10-28 18:18:57","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":383379,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eDifferential Expression of Genes \u003c/strong\u003e(\u003cstrong\u003ea\u003c/strong\u003e) Volcano plot of differentially expressed genes between Alzheimer's disease (AD) and normal groups. Red dots represent upregulated genes, and blue dots represent downregulated genes. (\u003cstrong\u003eb\u003c/strong\u003e) Intersection of significantly upregulated genes and SUMOylation-related genes. (\u003cstrong\u003ec\u003c/strong\u003e) Intersection of significantly downregulated genes and SUMOylation-related genes. (\u003cstrong\u003ed\u003c/strong\u003e) Gene ontology (GO) enrichment analysis of candidate genes. Blue dots represent downregulated genes, and red dots represent upregulated genes. The red box in the middle indicates the degree of enrichment, with higher levels indicating greater enrichment. (\u003cstrong\u003ee\u003c/strong\u003e) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of candidate genes. Blue dots represent downregulated genes, and red dots represent upregulated genes. The red box in the middle indicates the degree of enrichment, with higher levels indicating greater enrichment. (\u003cstrong\u003ef\u003c/strong\u003e) Protein-protein interaction (PPI) network of candidate genes.\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/90adbc3e56070124ba232c09.png"},{"id":94596461,"identity":"320c90d7-b457-4ca1-b961-c021aca4d8ac","added_by":"auto","created_at":"2025-10-28 18:42:10","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":178504,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eBiomarker Screening and Nomogram Construction \u003c/strong\u003e(\u003cstrong\u003ea\u003c/strong\u003e) Results of feature selection using the SVM-RFE algorithm. The error rate of the support vector machine model is shown. (\u003cstrong\u003eb-c\u003c/strong\u003e) Expression validation analysis of feature genes in the training set and validation set. Red represents the AD group, and green represents the normal group. ****, p \u0026lt; 0.0001; **, p \u0026lt; 0.01; ns, not significant. (\u003cstrong\u003eb\u003c/strong\u003e) Training set GSE140831. (\u003cstrong\u003ec\u003c/strong\u003e) Validation set GSE63060. (\u003cstrong\u003ed\u003c/strong\u003e) Nomogram based on biomarkers. (\u003cstrong\u003ee\u003c/strong\u003e) ROC validation of the nomogram.\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/8ff07b18b21e6a78fdeb5246.png"},{"id":94588295,"identity":"368b623b-29f8-42f0-9416-91a8181f89d5","added_by":"auto","created_at":"2025-10-28 18:19:07","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":442586,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eEnrichment Analysis of Biomarkers and Construction of Related Networks \u003c/strong\u003e(\u003cstrong\u003ea-c\u003c/strong\u003e) GSEA analysis of biomarkers. The y-axis represents the enrichment score, and the x-axis represents genes, with each vertical line representing a gene. (\u003cstrong\u003ea\u003c/strong\u003e) CREBBP. (\u003cstrong\u003eb\u003c/strong\u003e) PIAS1. (\u003cstrong\u003ec\u003c/strong\u003e) TRIM28. (\u003cstrong\u003ed\u003c/strong\u003e) Biomarker-transcription factor network. Red nodes represent mRNAs (biomarkers), and green nodes represent transcription factors. (\u003cstrong\u003ee\u003c/strong\u003e) Biomarker-MicroRNA network. Purple nodes represent mRNAs (biomarkers), and orange nodes represent MicroRNAs.\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/50ab04624e922961a5341eee.png"},{"id":94588564,"identity":"7081e3d6-4c31-4a80-b572-4a394e9c99c4","added_by":"auto","created_at":"2025-10-28 18:19:30","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":609331,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eImmune Microenvironment Analysis and Potential Drug Prediction for Biomarkers \u003c/strong\u003e(\u003cstrong\u003ea\u003c/strong\u003e) Immune infiltration analysis, showing the expression differences of 55 immune cells in the AD and control groups of the training set. (\u003cstrong\u003eb\u003c/strong\u003e) Correlation heatmap of 55 differentially expressed immune cells. (\u003cstrong\u003ec-e\u003c/strong\u003e) Correlation analysis between biomarkers and differentially expressed immune cells. (\u003cstrong\u003ec\u003c/strong\u003e) CREBBP. (\u003cstrong\u003ed\u003c/strong\u003e) PIAS1. (\u003cstrong\u003ee\u003c/strong\u003e) TRIM28. (\u003cstrong\u003ef\u003c/strong\u003e) Biomarker-targeted drug network. Orange nodes represent mRNAs (biomarkers), and purple nodes represent targeted drugs.\u003c/p\u003e","description":"","filename":"floatimage4.png","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/0722fd28800b8b995ed12b9f.png"},{"id":94588078,"identity":"fcb14180-41bd-41d0-b3f8-675153cbaba7","added_by":"auto","created_at":"2025-10-28 18:18:59","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":246980,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExpression Differences of Biomarkers in Different Dementia-Related Diseases \u003c/strong\u003e(\u003cstrong\u003ea\u003c/strong\u003e) Heatmap of biomarker expression in different dementia-related diseases. (\u003cstrong\u003eb\u003c/strong\u003e) Expression differences of biomarkers in bvFTD and control groups in GSE140830. (\u003cstrong\u003ec\u003c/strong\u003e) Expression differences of biomarkers in svPPA and control groups in GSE140830. (\u003cstrong\u003ed\u003c/strong\u003e) Expression differences of biomarkers in nfvPPA and control groups in GSE140830. (\u003cstrong\u003ee\u003c/strong\u003e) Expression differences of biomarkers in PSP and control groups in GSE140830. (\u003cstrong\u003ef\u003c/strong\u003e) Expression differences of biomarkers in CBS and control groups in GSE140830.\u003c/p\u003e","description":"","filename":"floatimage5.png","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/e6fda11a0d3759a8c2446eb0.png"},{"id":94587353,"identity":"1a07736a-acfb-46be-a014-c267a4cd4bf2","added_by":"auto","created_at":"2025-10-28 18:18:08","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":75141,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eExpression Validation of Key Genes in Clinical Tissue Samples, p \u0026lt; 0.05, ns, no significance \u003c/strong\u003e(\u003cstrong\u003ea\u003c/strong\u003e) CREBBP. (\u003cstrong\u003eb\u003c/strong\u003e) PIAS1. (\u003cstrong\u003ec\u003c/strong\u003e) TRIM28.\u003c/p\u003e","description":"","filename":"floatimage6.png","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/1a21f1a127b1f73397236d2d.png"},{"id":107928082,"identity":"510dbc4a-93fd-4794-9950-aec528417427","added_by":"auto","created_at":"2026-04-27 16:07:30","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2265932,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/45a6a9e2-a1d4-43c2-82d4-2ef44ac36e74.pdf"},{"id":94588112,"identity":"b1f665c2-1ed1-4d64-99d8-b0ef4df223f5","added_by":"auto","created_at":"2025-10-28 18:19:02","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":79500,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryInformation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-6983921/v1/d1ee87ce2d714f41450c3772.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"Identification of small ubiquitin-related modifier (SUMO)-related genes-based biomarkers in Alzheimer's disease based on bioinformatics analysis","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003eAlzheimer's disease (AD) is the most common neurodegenerative disease worldwide, with progressive cognitive decline as the main clinical feature [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. Alzheimer's Association International estimated there are around 50\u0026nbsp;million people with dementia worldwide, which is expected to triple by 2050 [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. The main manifestations of AD are progressive cognitive impairment and personality disorder [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. The APOE gene is the most common risk gene for AD [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e]. Deposition of amyloid beta (Aβ) in the brain parenchyma and cerebrovascular system, as well as the presence of neurofibrillary tangles within neurons and progressive loss of synapses, are central neuropathological markers of AD [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e]. Current drug treatments of AD only have symptomatic effects, and there is no treatment to alleviate the disease [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e], so new advances in the diagnosis and treatment of AD remain an urgent task. The search for biomarkers of AD is of great significance in further promoting AD treatment.\u003c/p\u003e\u003cp\u003eSmall ubiquitin-related modifier (SUMO), a 97-residue protein, mediates post-translational modification through covalent attachment to a specific lysine residue on the target protein [\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. Ubiquitination plays a important role in the biochemical pathways of eukaryotes by influencing the location, function, and stability of modified proteins [\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. In histopathology, AD is characterized by the insoluble aggregation of Aβ and microtubule-associated protein tau in the brain, both of these proteins are associated with SUMO. It has been demonstrated that tau SUMOylation reciprocally promotes its hyperphosphorylation, which results in aggregation of tau to form neurofibrillary tangles, and eventually interferes with the normal function of neurons and leads to nerve cell death [\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. And early studies have indicated that the SUMO system may impact Aβ levels and tau aggregation by altering with AD-type pathology [\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Telomere length is related to biological aging, a recent study has shown that telomere attrition is associated with the occurrence of AD [\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. Accumulating evidence indicates that the ubiquitin and SUMO pathways are crucial for telomere structure integrity because of they are significant regulators of the shelterin complex and other chromatin modifiers [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. However, the role of SUMO-related genes (SRGs) in AD is still unclear, so identifying SRGs-based biomarkers for AD\u0026rsquo;s diagnosis and investigating the mechanisms by which these factors influence AD may be of great significance to improving the intervention of AD patients and offering a novel insight for the treatment of AD.\u003c/p\u003e\u003cp\u003eIn this study, candidate genes were acquired based on AD-related datasets (GSE140831) and 189 SRGs, then SUMO-related genes associated with AD were determined by using the Support Vector Machine-Recursive Feature Elimination (SVM-RFE) algorithm, differential expression, and functional enrichment analyses. Our study could lead to earlier and more accurate AD diagnosis by identifying SRG-based diagnostic biomarkers. In addition, understanding the role of these biomarkers may be of great significance to discovering new therapeutic targets and providing new insights into the development of drugs or other treatment modalities to slow or stop AD progression.\u003c/p\u003e"},{"header":"2 Materials and methods","content":"\u003cp\u003e\u003cstrong\u003e2.1 Data collection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe expression profiles of GSE140831 (GPL15988), GSE63060 (GPL6947), and GSE140830 (GPL15988) datasets were searched from the Gene Expression Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo/) [13]. Specifically, 734 peripheral blood samples from 204 patients with AD and 530 healthy persons (controls) in the GSE140831 dataset were selected as a training set. The GSE63060 dataset offered peripheral blood samples from 145 AD and 104 controls. Besides, the GSE140830 dataset was composed of peripheral blood samples from 261 dementia patients (80 of behavioral variant frontotemporal dementia (bvFTD), 44 in semantic variant primary progressive aphasia (svPPA), 47 from non-fluent variant primary progressive aphasia (nfvPPA), 54 of progressive supranuclear palsy (PSP) and 36 of corticobasal syndrome (CBS) and 281 controls. A sum of 189 small ubiquitin-related modifiers (SUMO) related genes (SRGs) was retrieved from the Molecular Signatures Database (MSigDB, https://www.gsea-msigdb.org/) as \u0026apos;SUMOylation\u0026apos; being a keyword.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.2 Differential expression and functional enrichment analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe differentially expressed genes (DEGs) of AD between AD and control groups in the GSE140831 dataset were obtained employing the R package \u0026apos;limma\u0026apos; (Ver. 3.54.1) [14] (|Log\u003csub\u003e2\u003c/sub\u003efold-change (FC)|\u0026gt;0.1, adjusted p\u0026lt;0.05). Subsequently, utilizing the R package \u0026apos;ggVenndiagram\u0026apos; (Ver. 1.2.2) [15], the candidate genes associated with AD and SRGs were acquired through intersecting observably up-regulated and down-regulated DEGs with SRGs, respectively. Subsequently, the relevant functions and signaling pathways in which candidate genes were involved were investigated via Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genome (KEGG), implemented with the R package \u0026apos;clusterProfiler\u0026apos; (Ver. 4.6.2) [16] (adj p\u0026lt;0.05). Additionally, a protein-protein interaction (PPI) network revealing interactions among proteins encoded by candidate genes was built and visualized by the Search Tool for Recurring Instances of Neighboring Genes database (STRING, https://string-db.org) (Interaction score=0.4) and the plug-in CytoHubba of Cytoscape software severally (Ver. 3.9.0) [17]. The TOP 15 candidate genes in the PPI network were singled out as core genes for screening biomarkers.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.3 Acquisition of the biomarkers and construction of a nomogram model\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor the sake of searching biomarkers recognizing and diagnosing AD, the SVM-RF model-selected feature genes were obtained firstly from core genes when the model attained the highest accuracy and the lowest error rate after 10-fold cross-validation by R package \u0026apos;e1071\u0026apos; (Ver. 1.7-13) [18]. Afterwards, the expression verification of feature genes in the GSE140831 and GSE63060 datasets proceeded to ascertain genes with observably discrepant expression levels in AD and control groups and consistent expression trends in the two datasets as the biomarkers for AD. Wilcoxon\u0026apos;s test estimated the significantly different expression levels of inter-group feature genes (p\u0026lt;0.05). Furthermore, based on the GSE140831 dataset, a nomogram model predicting the risk of AD was established through the R package \u0026apos;rms\u0026apos; (Ver. 6.5-0) [19]. The nomogram model could visualize a total point for each sample in the GSE140831 dataset, and the risk of AD grew along with the increased total points. The predictive performance of the nomogram model was appraised by plotting the receiver operating characteristic (ROC) curve through the R package \u0026apos;pROC\u0026apos; (Ver. 1.18.0) [20] and computing the area under the ROC curve (AUC).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.4 Gene set enrichment analysis (GSEA)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo explore whether biomarkers were involved in regulating the progression of AD, the underlying pathways enriched by them were defined via GSEA. The Spearman\u0026apos;s correlation coefficients between biomarker and other genes in the GSE140831 dataset were calculated adopting the R package \u0026apos;psych\u0026apos; (Ver. 2.2.9) [21] and queued in order, and the sorted genes were input into the \u0026apos;clusterProfiler\u0026apos; package in R to match relevant pathways of biomarkers (adj.p\u0026lt;0.05, |NES|\u0026gt;1).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.5 Construction of molecular regulatory network\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe prediction of upstream molecules targeting biomarkers was beneficial in understanding the biological functions of biomarkers. So, the transcription factors (TFs) regulating biomarkers were predicted via the NetworkAnalyst (http://www.networkanalyst.ca) database. Moreover, the microRNAs (miRNAs) retrieved from the miRWALK (http://mirwalk.umm.uni-heidelberg.de/) and miRDB (http://www.mirdb.org) databases were overlapped as miRNAs corresponding to biomarkers. The feat of the Cytoscape software synthesized the TF-mRNA and miRNA-mRNA networks.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.6 Immunoinfiltration characterization and drug prediction\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo characterize the discrepancies in immune cell infiltration between AD and control groups, the abundance of 64 kinds of immune cells in each sample of the GSE140831 dataset was estimated through the xCell algorithm [22]. The discrepant immune cells in the AD and control groups were found through Wilcoxon\u0026apos;s test (p\u0026lt;0.05). Ulteriorly, Spearman\u0026apos;s correlation analysis uncovered associations among discrepant immune cells and between discrepant immune cells and each biomarker (|cor|\u0026gt;0.4, p\u0026lt;0.05).\u003c/p\u003e\n\u003cp\u003eMoreover, the Drug-Gene Interaction Database (DGIdb, www.dgidb.org) identified feasible drugs targeting biomarkers for AD therapy, and a gene-drug network was created through Cytoscape software.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.7 Assessment of expression levels of biomarkers in the five dementia-related diseases\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo understand the expression of biomarkers in five dementia-related diseases (bvFTD, svPPA, nfvPPA, PSP, and CBS), the expression patterns of the biomarkers in AD and control groups were first displayed in a heatmap plotted by the R package \u0026apos;ggplot2\u0026apos; (Ver. 3.3.6) [23]. Besides, the differences in the expression of biomarkers in five dementia-related diseases were compared through Wilcoxon\u0026apos;s test (p\u0026lt;0.05).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.8 Real-time reverse transcriptase-polymerase chain reaction (RT-qPCR)\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe RT-qPCR was the assay of choice for detecting relative biomarker expression levels in five pairs of AD and control whole blood samples gathered from Fuzhou Second General Hospital. All sample donors signed the informed consent forms. This study was conducted in accordance with the Declaration of Helsinki and obtained ethical approval from the Ethics Committee of Fuzhou Second People\u0026apos;s Hospital (Approval No.: 2024005).The total RNA was extracted from the ten samples using the TRIzol reagent (Ambion, USA) per the manufacturer\u0026apos;s protocol. The concentrations of total RNA were detected in rapid sequence via the NanoPhotometer N50. Ulteriorly, the cDNA synthesis was reverse-transcribed using the SureScript-First-strand-cDNA-synthesis-kit (Servicebio, China). Additionally, the CFX Connect Thermal Cycler (Bio-Rad, USA) was operated to detect the relative expression level of mRNA using the GAPDH as an internal reference gene for RT-qPCR through the 2\u003csup\u003e-\u0026Delta;\u0026Delta;CT\u003c/sup\u003e method. The details of all primers are attached in\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003eOnline Resource 1 which can be found in the supplementary file.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e2.9 Statistical analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBioinformatics analysis was implemented through this study\u0026apos;s R software (Ver. 4.3.2). The Wilcoxon\u0026apos;s test was used to distinguish the inter-group discrepancies, and a p-value or adj. p-value of less than 0.05 was deemed statistically significant.\u003c/p\u003e"},{"header":"3 Results","content":"\u003cp\u003e\u003cstrong\u003e3.1 The functions of candidate genes might be related to nucleic acid and cytoplasmic transport and cancer regulation\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e12,853 DEGs were identified between the AD and control groups in the GSE140831 dataset (\u003cstrong\u003eFig. 1a\u003c/strong\u003e). 107 candidate genes related to AD and SRGs, with 85 up-regulated and 22 down-regulated, were obtained by overlapping 189 SRGs and 6,640 up-regulated and 6,213 down-regulated DEGs (\u003cstrong\u003eFig. 1b-c\u003c/strong\u003e). They were enriched in 624 GO entries, comprising 498 biological processes (BPs), 44 cellular components (CCs), and 82 molecular functions (MFs). Moreover, the functions of candidate genes were associated with nucleic acid transport, nucleocytoplasmic transport, protein sumoylation, and so on (\u003cstrong\u003eFig. 1d\u003c/strong\u003e). Concerning the 34 KEGG pathways involved by candidate genes, they included not only cytoplasmic transport but also nucleocytoplasmic transport, transcriptional misregulation in cancer, chronic myeloid leukemia, etc (\u003cstrong\u003eFig. 1e\u003c/strong\u003e). Generally, 107 candidate genes might participate in nucleic acid and cytoplasmic transport and cancer regulation. Additionally, the PPI-based screening manifested that the interactions among 15 core genes, \u003cem\u003ePARP1\u003c/em\u003e, \u003cem\u003eTRIM28\u003c/em\u003e,\u003cem\u003e\u0026nbsp;NPM1\u003c/em\u003e, \u003cem\u003eRANGAP1\u003c/em\u003e, \u003cem\u003eSUMO3\u003c/em\u003e, \u003cem\u003eMDM2\u003c/em\u003e, \u003cem\u003eNCOR2\u003c/em\u003e, \u003cem\u003ePPARG\u003c/em\u003e, \u003cem\u003eTP53\u003c/em\u003e, \u003cem\u003eSUMO2\u003c/em\u003e,\u003cem\u003e\u0026nbsp;CREBBP\u003c/em\u003e, \u003cem\u003eHDAC1\u003c/em\u003e, \u003cem\u003eEP300\u003c/em\u003e, \u003cem\u003eSUMO1\u003c/em\u003e and\u003cem\u003e\u0026nbsp;PIAS1\u003c/em\u003ewere close at protein level, especially among TP53, EP300 and SUMO1 (\u003cstrong\u003eFig. 1f\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.2 The performance of the biomarker-based nomogram model in predicting the risk of AD was reliable\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThere were five feature genes (\u003cem\u003eCREBBP\u003c/em\u003e, \u003cem\u003ePIAS1\u003c/em\u003e, \u003cem\u003eTRIM28\u003c/em\u003e, \u003cem\u003eMAD2\u003c/em\u003e, and \u003cem\u003eNCOR2\u003c/em\u003e) in the SVM-RFE model when the model had the highest accuracy (\u003cstrong\u003eFig. 2a\u003c/strong\u003e), of which CREBBP, PIAS1 and TRIM28 were served as biomarkers for AD as a consequence of their observably increased expression levels in AD group and consistent expression trends in GSE140831 and GSE63060 datasets (p\u0026lt;0.05) (\u003cstrong\u003eFig. 2b-c\u003c/strong\u003e). The nomogram model created by absorbing three biomarkers demonstrated that the risk of AD increased with the total points corresponding to each sample in the GSE140831 dataset (\u003cstrong\u003eFig. 2d\u003c/strong\u003e). The precise predictive performance of the nomogram model must be further confirmed by the AUC value of 0.809 (\u003cstrong\u003eFig. 2e\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.3 Three biomarkers were involved in identical pathways, such as olfactory transduction, lysosome, and spliceosome\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eExploring three biomarkers by GSEA was essential for understanding their effects on AD. The findings of GSEA unveiled that \u003cem\u003eCREBBP\u003c/em\u003e, \u003cem\u003ePIAS1\u003c/em\u003e, and \u003cem\u003eTRIM28\u0026nbsp;\u003c/em\u003egathered in olfactory transduction, lysosome, and spliceosome were significantly and synchronously. Besides, Huntington\u0026apos;s disease and neuroactive ligand receptor interaction were involved by \u003cem\u003ePIAS1\u003c/em\u003e and \u003cem\u003eTRIM28\u003c/em\u003e. In conclusion, GSEA certifies that three biomarkers might retain similar functions in regulating AD\u0026apos;s progression (\u003cstrong\u003eFig. 3a-c\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.4 The expression of three biomarkers was regulated by multiple TFs and miRNAs\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eHow upstream molecules regulated three biomarkers was further explored; 58 TFs and 284 miRNAs targeted three. The TF-mRNA displayed that three biomarkers could be regulated by \u003cem\u003eEGR1\u003c/em\u003e, \u003cem\u003eDMRT1\u003c/em\u003e, \u003cem\u003eE2F1\u003c/em\u003e, \u003cem\u003eCREM\u003c/em\u003e, and \u003cem\u003eZFX\u003c/em\u003e simultaneously (\u003cstrong\u003eFig. 3d\u003c/strong\u003e). In addition, as presented in \u003cstrong\u003eFig. 3e\u003c/strong\u003e, the transcription of \u003cem\u003eTRIM28\u003c/em\u003e and \u003cem\u003eCREBBP\u003c/em\u003e could be regulated by hsa-miR-6812-5p, hsa-miR-1227-5p, and hsa-miR-7113-5p synchronously, as well as \u003cem\u003eCREBBP\u003c/em\u003e and \u003cem\u003ePIAS1\u003c/em\u003e by hsa-miR-4452, hsa-miR-1284, hsa-miR-6883-3p, and so forth.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.5 Most of the immune cells might influence the progression of AD\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAmong the 64 immune cells, a total of 55 immune cells showed markedly discrepant abundance between the AD and control groups (p\u0026lt;0.05). For instance, AD\u0026apos;s percentages of basophils, class-switched memory B cells, myocytes, and mesenchymal stem cells (MSCs) were higher. In contrast, the proportions of natural killer T (NKT) cells, B cells, neutrophils, smooth muscle, and naive CD8(+) T cells were inverse in AD (\u003cstrong\u003eFig. 4a\u003c/strong\u003e). Furthermore, Spearman\u0026apos;s correlation analysis among discrepant immune cells uncovered that memory B cells and B cells maintained the highest positive correlations of 0.76 (p\u0026lt;0.05), yet T helper type 1 (Th1) cells and neurons kept the highest negative correlations of -0.72 (p\u0026lt;0.05) (\u003cstrong\u003eFig. 4b\u003c/strong\u003e). Basophils were positively correlated with \u003cem\u003ePIAS1\u003c/em\u003e and \u003cem\u003eCREBBP\u003c/em\u003e, and myocytes with \u003cem\u003eTRIM28\u003c/em\u003e. Besides neutrophils and \u003cem\u003eCREBBP\u003c/em\u003e, \u003cem\u003ePIAS1\u003c/em\u003e maintained significantly negative correlations, as well as smooth muscle and \u003cem\u003eTRIM28\u003c/em\u003e (|cor|\u0026gt;0.4, p\u0026lt;0.05) (\u003cstrong\u003eFig. 4c-e\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.6 A sum of 21 alternative drugs might be developed for AD therapy\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWhat is more, a total of 21 drugs targeting \u003cem\u003eCREBBP\u003c/em\u003e were retrieved in DGIdb, such as triazolam, colchicine, alprazolam, nocodazole, etc. (\u003cstrong\u003eFig. 4f\u003c/strong\u003e). Unfortunately, no drugs targeting\u003cem\u003e\u0026nbsp;TRIM28\u003c/em\u003e and \u003cem\u003ePIAS1\u003c/em\u003e were found in this database. Among these, triazolam and alprazolam were proven to be therapeutic for AD. However, applying them in large doses could produce side effects such as decreased blood pressure and respiratory circulation suppression, so drug prediction targeting biomarkers might provide data for developing available drugs for AD.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.7 The expression level of TRIM28 was significantly increased in svPPA\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe expression patterns of three biomarkers in five dementia-related diseases were exhibited in the heat map (\u003cstrong\u003eFig. 5a\u003c/strong\u003e). Expression analysis of biomarkers confirmed that \u003cem\u003eCREBBP\u003c/em\u003e and\u003cem\u003e\u0026nbsp;PIAS1\u003c/em\u003e showed no significant discrepancies between bvFTD, svPPA, nfvPPA, PSP, and CBS and controls, except for \u003cem\u003eTRIM28\u003c/em\u003e with upregulated expression levels in svPPA (\u003cstrong\u003eFig. 5b-f\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e3.8 CREBBP and TRIM28 were all over-expressed in AD samples\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eConsistent with the expression trend in the GSE140831 and GSE63060 datasets, the relative expression levels of \u003cem\u003eCREBBP\u003c/em\u003e (p = 0.0043) and \u003cem\u003eTRIM28\u003c/em\u003e (p = 0.0032) were increased in AD samples. While the expression level of \u003cem\u003ePIAS1\u003c/em\u003e was not significant (p \u0026gt; 0.05) (\u003cstrong\u003eFig. 6a-c\u003c/strong\u003e).\u003c/p\u003e"},{"header":"4 Discussion","content":"\u003cp\u003eAlzheimer's disease (AD) is a degenerative disease of the nervous system. whose primary genetic risk factor is the APOE gene and is characterized by insoluble aggregation of proteins such as amyloid-beta (Aβ) [\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e]. As A post-translational modification, SUMOylation has been shown to regulate multiple cellular pathways. SUMO is believed to be closely related to key proteins in AD, such as Aβ and the microtubule-associated protein tau, and is involved in regulating various cellular functions associated with AD pathology [\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. In addition, the SUMO pathway is also associated with AD's occurrence and development by affecting the telomere structure's integrity, which can provide a new research perspective for diagnosing and treating AD [\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. However, bioinformatic analysis of SRGs in AD has not yet been performed, and crucial SRG-based biomarkers involved in the pathogenesis of AD have not been confirmed.\u003c/p\u003e\u003cp\u003eIn the present study, we identified the SRGs associated biomarkers in AD based on the GSE140831, GSE63060, and GSE140830 datasets, specifically \u003cem\u003eCREBBP\u003c/em\u003e, \u003cem\u003ePIAS1\u003c/em\u003e, and \u003cem\u003eTRIM28\u003c/em\u003e. These biomarkers are potential diagnostic targets for AD because of their increased expression levels in the AD group and reliable predictive performance. GSEA and the prediction results of TFs suggest that these biomarkers may have similar biological functions in AD\u0026rsquo;s progression, and the expression of three biomarkers could be regulated simultaneously by some TFs. Immune infiltration analysis suggested that, between the AD and control groups, the proportion of specific immune cells correlated significantly with these biomarkers. Finally, we speculate a total of 21 drugs as potential therapeutic agents for AD on DGIdb, in addition to triazolam and alprazolam, which have been used in the treatment of AD.\u003c/p\u003e\u003cp\u003e\u003cem\u003eCREBBP\u003c/em\u003e is a transcriptional coactivator with histone acetyltrans-ferase (HAT) activity and is involved in post-translational modification by acetylation of histones [\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e], there are studies showed that dysregulation of histone acetylation pathways is thought to play may play a key role in the disease progression of AD. Hence, dysfunction of \u003cem\u003eCREBBP\u003c/em\u003e is likely to contribute to neurodegenerative diseases [\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. In addition, recent studies have shown that \u003cem\u003eCREBBP\u003c/em\u003e may be a hub gene that provides new insight into treating AD by mediating the metabolism of niacin and lysosome [\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e, \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. \u003cem\u003ePIAS1\u003c/em\u003e, a protein inhibitor of the activated \u003cem\u003eSTAT1\u003c/em\u003e, is one of E3 the E3-type SUMO ligases. It plays important roles in the STAT pathway, which is essential to inflammatory activation in AD [\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Besides that, \u003cem\u003ePIAS1\u003c/em\u003e is thought to play a neuroprotective role against Aβ toxicity in AD [\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e, \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. \u003cem\u003eTRIM28\u003c/em\u003e has also been demonstrated to be an E3 SUMOylation ligase [\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. Maxime et al found that \u003cem\u003eTRIM28\u003c/em\u003e regulates the steady state levels of a-Synuclein (a-Syn) and tau, which are supposed to be neurodegeneration-driving proteins in AD by mediating the SUMOylation [\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eGSEA showed that lysosome, olfactory transduction, and spliceosome are the pathways involved in these three biomarkers. Studies have shown that TRIM28 inhibits autophagy degradation by inducing SUMOylation of BCR-ABL [\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e]. Lysosomes are a significant cell signaling center, influencing fundamental processes such as cell membrane repair, energy metabolism, and inflammatory pathways [\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. The autophagy-lysosome pathway is closely related to protein homeostasis and other processes in AD [\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. Therefore, it is reasonable to speculate that the biomarker \u003cem\u003eTRIM28\u003c/em\u003e may affect AD by participating in the regulation of the lysosomal pathway. Olfactory dysfunction often occurs in the early stages of AD, studies have found that the olfactory mucosal cells from AD patients secrete an increased amount of Aβ1\u0026ndash;42, and the ratio of secretion of Aβ1\u0026ndash;42 / Aβ1\u0026ndash;40 also increases [\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e]. Splicing contributes to the diversity and function of neuronal transcription; the disruption of the splicing mechanism leads to neurological diseases, including AD [\u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e43\u003c/span\u003e]. Yi-Chen et al supposed that the interaction of tau spliceosomes disrupts the function of snRNP, resulting in loss of transcriptome fidelity and splicing errors, and leading to neurodegenerative changes in AD ultimately [\u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e44\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe TF-mRNA indicated that \u003cem\u003eEGR1\u003c/em\u003e, \u003cem\u003eDMRT1\u003c/em\u003e, \u003cem\u003eCREM\u003c/em\u003e, \u003cem\u003eZFX\u003c/em\u003e, and \u003cem\u003eE2F1\u003c/em\u003e could simultaneously regulate three biomarkers. As a member of immediate early genes (IEGs), \u003cem\u003eEGR1\u003c/em\u003e is a mediator and regulator of synaptic plasticity and neuronal activity under physiological conditions and in pathological conditions [\u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e45\u003c/span\u003e], and has recently been considered a candidate gene for schizophrenia. \u003cem\u003eCREM\u003c/em\u003e and \u003cem\u003eCREB\u003c/em\u003e are members of the same family of transcription factors, and \u003cem\u003eCREB\u003c/em\u003e may have a significant effect on forming and consolidating of memory by interacting with \u003cem\u003eCREBBP\u003c/em\u003e [\u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e46\u003c/span\u003e]. \u003cem\u003eZFX\u003c/em\u003e is a zinc finger protein of the Zfy family. David found that \u003cem\u003eZFX\u003c/em\u003e is a repressor of core and linker histones [\u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e47\u003c/span\u003e]. \u003cem\u003eE2F1\u003c/em\u003e is a critical transcription factor regulating cell cycle, which is primarily considered to be a key factor in cell proliferation and cancer. Still, recent studies have suggested that it may also act as an important part in neurodegenerative diseases [\u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e48\u003c/span\u003e]. The expression levels of \u003cem\u003eE2F-1\u003c/em\u003e in the brain tissue of AD model rats had changed, suggesting that \u003cem\u003eE2F-1\u003c/em\u003e may be involved in neuronal death and loss of function by regulating cell cycle and apoptosis in AD [\u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e49\u003c/span\u003e]. These transcription factors associated with \u003cem\u003eCREBBP\u003c/em\u003e, \u003cem\u003ePIAS1\u003c/em\u003e, and \u003cem\u003eTRIM28\u003c/em\u003e may mediate the development of AD by regulating histone acetylation, synaptic plasticity, and the cell cycle. However, the mechanisms of these effects still need to be further explored.\u003c/p\u003e\u003cp\u003eNeuroimmune interactions are a significant focus of research into neurodegenerative diseases, including AD, recently [\u003cspan citationid=\"CR50\" class=\"CitationRef\"\u003e50\u003c/span\u003e]. Our study found that in AD, the proportion of specific immune cells increased, such as basophils, class-switching memory B cells, muscle cells. In contrast, the proportion of specific immune cells decreased, such as natural killer T cells (NKT), B cells, neutrophils, smooth muscle, and naive CD8(+) T cells. The results are similar to previous studies [\u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e51\u003c/span\u003e, \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e52\u003c/span\u003e]. In addition, among these immune cells, basophils were positively correlated with \u003cem\u003ePIAS1\u003c/em\u003e and \u003cem\u003eCREBBP\u003c/em\u003e, myocytes were positively associated with \u003cem\u003eTRIM28\u003c/em\u003e, and there was a significant negative correlation between neutrophils with \u003cem\u003eCREBBP\u003c/em\u003e and \u003cem\u003ePIAS1\u003c/em\u003e, as well as smooth muscle and \u003cem\u003eTRIM28\u003c/em\u003e. Basophils are a subset of granulocytes involved in various immunomodulatory and inflammatory processes. A study found that the number of basophils increased significantly in AD samples [\u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e53\u003c/span\u003e]. This is consistent with our findings, so we could speculate that \u003cem\u003ePIAS1\u003c/em\u003e and \u003cem\u003eCREBB\u003c/em\u003e may regulate the inflammatory response in AD by affecting neutrophils. However, Shad found blood levels of basophils were constantly low in AD [\u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e54\u003c/span\u003e], so the role of basophils in AD is still under more research. Similar to basophils, other types of cells have a complex relationship with this neurodegenerative disease. For example, myocytes also play an essential role in multiple physiological processes closely related to AD. Myocytes are mainly involved in muscle function, motor control, and energy metabolism. Cerebral small vessel disease (CSVD) is characterize with depletion of vascular myocytes, so vascular myocytes are considered as central to brain aging [\u003cspan citationid=\"CR55\" class=\"CitationRef\"\u003e55\u003c/span\u003e], and there is evidence that myocytes are the source of amyloid deposits in the meninges and cortical blood vessels [\u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e56\u003c/span\u003e], which may affect the progression of AD. SO \u003cem\u003eTRIM28\u003c/em\u003e may participate in the pathological process of AD by affecting myocyte function and further increasing amyloid deposition in the brain. It is worth to noting that recent studies have suggested that blood-brain barrier (BBB) dysfunction is correlated with human cognitive impairment including the early stages of AD [\u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e57\u003c/span\u003e]. BBB is made up of various cell including smooth muscle cells, besides that, it has been suggested that neutrophils may be involved in the early development of AD by mediating BBB damage, intravascular adhesion, and invasion of the central nervous system [\u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e58\u003c/span\u003e]. Therefore, we speculated that these three biomarkers can mediates peripheral or central inflammatory processes and affect the integrity of BBB by affecting the function of immune cells, thereby contribute to the occurrence and development of AD.\u003c/p\u003e\u003cp\u003eFinally, we identified the potential therapeutic drugs targeting \u003cem\u003eCREBBP\u003c/em\u003e through the DGIdb database, totally 21 drugs were screened including colchicine, glucocorticoid, benzodiazepines an so on, among which triazolam and alprazolam have been used to improve behavioral and psychological symptoms of AD [\u003cspan citationid=\"CR59\" class=\"CitationRef\"\u003e59\u003c/span\u003e]. As we know, although cholinesterase inhibitors, which designed to increase the levels of acetylcholine (ACh) in the brain, are the main drugs for AD, there are no therapeutic drugs with definite efficacy for AD [\u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e60\u003c/span\u003e]. Interestingly, Keisuke et al recently founded that triazolam could inhibit rhAChE activity by \u0026ge;\u0026thinsp;20% [\u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e61\u003c/span\u003e], which may provide new evidence for the use of benzodiazepines in the treatment of AD. In a word, our drug predictions based on these biomarkers, particularly those related to \u003cem\u003eCREBBP\u003c/em\u003e, can provide valuable data for the exploit of more effective and safer AD drugs, but further research involving animal models and clinical trials is required to assess the safety and efficacy of these drugs.\u003c/p\u003e\u003cp\u003eIn conclusion, we identified \u003cem\u003eCREBBP\u003c/em\u003e, \u003cem\u003ePIAS1\u003c/em\u003e and \u003cem\u003eTRIM28\u003c/em\u003e as SUMO-SRGs-based biomarkers diagnosing AD through bioinformatics analysis firstly in this study, furthermore, we revealed the regulatory networks and identified potential therapeutic agents aimed at these biomarkers. These genes may be involved in the onset and progression of AD, they have potential to be considered as novel biomarkers for AD's diagnosing and progression monitoring. Our study contributes to clarifying the the relationship between SUMO and the pathological process of AD and also offers insights into exploring novel treatment strategies for AD. Of course, there were some limitations in our study. Firstly, the sample size for qRT-PCR analysis is relatively insufficient, expanding the sample size is needed to verify our findings. Secondly, the diagnostic efficacy and mechanism by which these potential AD-related SUMO-SRGs-based biomarkers must be substantiated by further research and laboratory validation.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e This study was financially supported by the the Intramural Research Project of Fuzhou Second General Hospital, China (Award Number: 2024ZY04) and the Natural Science Foundation of Fujian Province, China (Award Number: 2023J011519) .\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e The authors declare no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor Contributions \u003c/strong\u003eAll authors contributed to the study conception and design. Peng Chen and Mengting Fan contributed equally to this work. Conceptualization:Peng Chen, Mengting Fan, Jing Xu; Methodology: Peng Chen; Formal analysis and investigation: Peng Chen; Writing - original draft preparation: Mengting Fan; Writing - review and editing: Peng Chen, Mengting Fan, Hailiang Lin, Zhong Chen, Qiuyang Zhang, Yunfan Cheng; Funding acquisition: Qiuyang Zhang and Jing Xu.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability statements \u003c/strong\u003eThe datasets analysed during the current study are available in the [GEO] repository, [https://www.ncbi.nlm.nih.gov/geo/, reference number GSE140831 (GPL15988), GSE63060 (GPL6947), GSE140830 (GPL15988)]. and [MSigDB] repository, [ https://www.gsea-msigdb.org/].\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003eThis study was conducted in accordance with the Declaration of Helsinki. Ethical approval was obtained from the Ethics Committee of Fuzhou Second General Hospital, with the approval number 2024005 and the approval date February 29, 2024. All patients provided written informed consent when their clinical samples were used for RT-qPCR experiments, ensuring that the research process complied with ethical standards and fully respected the rights and wishes of the patients.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to Publish\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eN/A\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eArvanitakis Z, Shah RC, Bennett DA (2019) Diagnosis and Management of Dementia: Review. Jama. 322:1589-1599. https://doi.org/10.1001/jama.2019.4782.\u003c/li\u003e\n\u003cli\u003eScheltens P, De Strooper B, Kivipelto M, et al. (2021) Alzheimer\u0026apos;s disease. Lancet. 397:1577-1590. https://doi.org/10.1016/s0140-6736(20)32205-4.\u003c/li\u003e\n\u003cli\u003eBarbe C, Jolly D, Morrone I, et al. (2018) Factors associated with quality of life in patients with Alzheimer\u0026apos;s disease. BMC Geriatr. 18:159. https://doi.org/10.1186/s12877-018-0855-7.\u003c/li\u003e\n\u003cli\u003evan der Lee SJ, Wolters FJ, Ikram MK, et al. (2018) The effect of APOE and other common genetic variants on the onset of Alzheimer\u0026apos;s disease and dementia: a community-based cohort study. Lancet Neurol. 17:434-444. https://doi.org/10.1016/s1474-4422(18)30053-x.\u003c/li\u003e\n\u003cli\u003eRostagno AA (2022) Pathogenesis of Alzheimer\u0026apos;s Disease. Int J Mol Sci. 24:107. https://doi.org/10.3390/ijms24010107.\u003c/li\u003e\n\u003cli\u003ePasseri E, Elkhoury K, Morsink M, et al. (2022) Alzheimer\u0026apos;s Disease: Treatment Strategies and Their Limitations. Int J Mol Sci. 23:13954. https://doi.org/10.3390/ijms232213954.\u003c/li\u003e\n\u003cli\u003eMartins WC, Tasca CI, Cimarosti H (2016) Battling Alzheimer\u0026apos;s Disease: Targeting SUMOylation-Mediated Pathways. Neurochem Res. 41:568-578. https://doi.org/10.1007/s11064-015-1681-3.\u003c/li\u003e\n\u003cli\u003eGrabbe C, Husnjak K, Dikic I (2011) The spatial and temporal organization of ubiquitin networks. Nat Rev Mol Cell Biol. 12:295-307. https://doi.org/10.1038/nrm3099.\u003c/li\u003e\n\u003cli\u003eNortley R, Korte N, Izquierdo P, et al. (2019) Amyloid \u0026beta; oligomers constrict human capillaries in Alzheimer\u0026apos;s disease via signaling to pericytes. Science. 365:eaav9518. https://doi.org/10.1126/science.aav9518.\u003c/li\u003e\n\u003cli\u003eLee L, Sakurai M, Matsuzaki S, et al. (2013) SUMO and Alzheimer\u0026apos;s disease. Neuromolecular Med. 15:720-736. https://doi.org/10.1007/s12017-013-8257-7.\u003c/li\u003e\n\u003cli\u003eKuan XY, Fauzi NSA, Ng KY, et al. (2023) Exploring the Causal Relationship Between Telomere Biology and Alzheimer\u0026apos;s Disease. Mol Neurobiol. 60:4169-4183. https://doi.org/10.1007/s12035-023-03337-4.\u003c/li\u003e\n\u003cli\u003eZalzman M, Meltzer WA, Portney BA, et al. (2020) The Role of Ubiquitination and SUMOylation in Telomere Biology. Curr Issues Mol Biol. 35:85-98. https://doi.org/10.21775/cimb.035.085.\u003c/li\u003e\n\u003cli\u003eChang B, Hu Z, Chen L, et al. (2023) Development and validation of cuproptosis-related genes in synovitis during osteoarthritis progress. Front Immunol. 14:1090596. https://doi.org/10.3389/fimmu.2023.1090596.\u003c/li\u003e\n\u003cli\u003eRitchie ME, Phipson B, Wu D, et al. (2015) limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43:e47. https://doi.org/10.1093/nar/gkv007.\u003c/li\u003e\n\u003cli\u003ePelikan A, Herzel H, Kramer A, et al. (2022) Venn diagram analysis overestimates the extent of circadian rhythm reprogramming. Febs j. 289:6605-6621. https://doi.org/10.1111/febs.16095.\u003c/li\u003e\n\u003cli\u003eWu T, Hu E, Xu S, et al. (2021) clusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innovation (Camb). 2:100141. https://doi.org/10.1016/j.xinn.2021.100141.\u003c/li\u003e\n\u003cli\u003eShannon P, Markiel A, Ozier O, et al. (2003) Cytoscape: a software environment for integrated models of biomolecular interaction networks. Genome Res. 13:2498-2504. https://doi.org/10.1101/gr.1239303.\u003c/li\u003e\n\u003cli\u003eHu X, Wang J, Ju Y, et al. (2022) Combining metabolome and clinical indicators with machine learning provides some promising diagnostic markers to precisely detect smear-positive/negative pulmonary tuberculosis. BMC Infect Dis. 22:707. https://doi.org/10.1186/s12879-022-07694-8.\u003c/li\u003e\n\u003cli\u003eLiu TT, Li R, Huo C, et al. (2021) Identification of CDK2-Related Immune Forecast Model and ceRNA in Lung Adenocarcinoma, a Pan-Cancer Analysis. Front Cell Dev Biol. 9:682002. https://doi.org/10.3389/fcell.2021.682002.\u003c/li\u003e\n\u003cli\u003eLiu C, Fang J, Kang W, et al. (2023) Identification of novel potential homologous repair deficiency-associated genes in pancreatic adenocarcinoma via WGCNA coexpression network analysis and machine learning. Cell Cycle. 22:2392-2408. https://doi.org/10.1080/15384101.2023.2293594.\u003c/li\u003e\n\u003cli\u003eGoncharova IA, Nazarenko MS, Babushkina NP, et al. (2020) [Genetic Predisposition to Early Myocardial Infarction]. Mol Biol (Mosk). 54:224-232. https://doi.org/10.31857/s0026898420020044.\u003c/li\u003e\n\u003cli\u003eZhuang X, Schlunck G, Wolf J, et al. (2022) Time- and Stimulus-Dependent Characteristics of Innate Immune Cells in Organ-Cultured Human Corneal Tissue. J Innate Immun. 14:98-111. https://doi.org/10.1159/000516669.\u003c/li\u003e\n\u003cli\u003eLiu S, Xie X, Lei H, et al. (2019) Identification of Key circRNAs/lncRNAs/miRNAs/mRNAs and Pathways in Preeclampsia Using Bioinformatics Analysis. Med Sci Monit. 25:1679-1693. https://doi.org/10.12659/msm.912801.\u003c/li\u003e\n\u003cli\u003eMangialasche F, Solomon A, Winblad B, et al. (2010) Alzheimer\u0026apos;s disease: clinical trials and drug development. Lancet Neurol. 9:702-716. https://doi.org/10.1016/s1474-4422(10)70119-8.\u003c/li\u003e\n\u003cli\u003eHoppe JB, Salbego CG, Cimarosti H (2015) SUMOylation: Novel Neuroprotective Approach for Alzheimer\u0026apos;s Disease? Aging Dis. 6:322-330. https://doi.org/10.14336/ad.2014.1205.\u003c/li\u003e\n\u003cli\u003eSe Thoe E, Fauzi A, Tang YQ, et al. (2021) A review on advances of treatment modalities for Alzheimer\u0026apos;s disease. Life Sci. 276:119129. https://doi.org/10.1016/j.lfs.2021.119129.\u003c/li\u003e\n\u003cli\u003eCummings JL, Tong G, Ballard C (2019) Treatment Combinations for Alzheimer\u0026apos;s Disease: Current and Future Pharmacotherapy Options. J Alzheimers Dis. 67:779-794. https://doi.org/10.3233/jad-180766.\u003c/li\u003e\n\u003cli\u003eGoodman RH, Smolik S (2000) CBP/p300 in cell growth, transformation, and development. Genes Dev. 14:1553-1577. \u003c/li\u003e\n\u003cli\u003eSchueller E, Paiva I, Blanc F, et al. (2020) Dysregulation of histone acetylation pathways in hippocampus and frontal cortex of Alzheimer\u0026apos;s disease patients. Eur Neuropsychopharmacol. 33:101-116. https://doi.org/10.1016/j.euroneuro.2020.01.015.\u003c/li\u003e\n\u003cli\u003eJeong H, Then F, Melia TJ, Jr., et al. (2009) Acetylation targets mutant huntingtin to autophagosomes for degradation. Cell. 137:60-72. https://doi.org/10.1016/j.cell.2009.03.018.\u003c/li\u003e\n\u003cli\u003eLi T, Yin L, Kang X, et al. (2022) TFEB acetylation promotes lysosome biogenesis and ameliorates Alzheimer\u0026apos;s disease-relevant phenotypes in mice. J Biol Chem. 298:102649. https://doi.org/10.1016/j.jbc.2022.102649.\u003c/li\u003e\n\u003cli\u003eWang Z, Zou Z, Li Q (2022) Nicotinic acid supplementation contributes to the amelioration of Alzheimer\u0026apos;s disease in mouse models. Ann Transl Med. 10:1049. https://doi.org/10.21037/atm-22-1148.\u003c/li\u003e\n\u003cli\u003eHe K, Zhang J, Liu J, et al. (2021) Functional genomics study of protein inhibitor of activated STAT1 in mouse hippocampal neuronal cells revealed by RNA sequencing. Aging (Albany NY). 13:9011-9027. https://doi.org/10.18632/aging.202749.\u003c/li\u003e\n\u003cli\u003eLiu SY, Ma YL, Hsu WL, et al. (2019) Protein inhibitor of activated STAT1 Ser(503) phosphorylation-mediated Elk-1 SUMOylation promotes neuronal survival in APP/PS1 mice. Br J Pharmacol. 176:1793-1810. https://doi.org/10.1111/bph.14656.\u003c/li\u003e\n\u003cli\u003eTao CC, Hsu WL, Ma YL, et al. (2017) Epigenetic regulation of HDAC1 SUMOylation as an endogenous neuroprotection against A\u0026beta; toxicity in a mouse model of Alzheimer\u0026apos;s disease. Cell Death Differ. 24:597-614. https://doi.org/10.1038/cdd.2016.161.\u003c/li\u003e\n\u003cli\u003eNeo SH, Itahana Y, Alagu J, et al. (2015) TRIM28 Is an E3 Ligase for ARF-Mediated NPM1/B23 SUMOylation That Represses Centrosome Amplification. Mol Cell Biol. 35:2851-2863. https://doi.org/10.1128/mcb.01064-14.\u003c/li\u003e\n\u003cli\u003eRousseaux MW, Revelli JP, V\u0026aacute;zquez-V\u0026eacute;lez GE, et al. (2018) Depleting Trim28 in adult mice is well tolerated and reduces levels of \u0026alpha;-synuclein and tau. Elife. 7:e36768. https://doi.org/10.7554/eLife.36768.\u003c/li\u003e\n\u003cli\u003eRousseaux MW, de Haro M, Lasagna-Reeves CA, et al. (2016) TRIM28 regulates the nuclear accumulation and toxicity of both alpha-synuclein and tau. Elife. 5:e19809. https://doi.org/10.7554/eLife.19809.\u003c/li\u003e\n\u003cli\u003eLiu W, Zhu Y, Ye W, et al. (2025) Redox regulation of TRIM28 facilitates neuronal ferroptosis by promoting SUMOylation and inhibiting OPTN-selective autophagic degradation of ACSL4. Cell Death Differ. https://doi.org/10.1038/s41418-025-01452-4.\u003c/li\u003e\n\u003cli\u003eLawrence RE, Zoncu R (2019) The lysosome as a cellular centre for signalling, metabolism and quality control. Nat Cell Biol. 21:133-142. https://doi.org/10.1038/s41556-018-0244-7.\u003c/li\u003e\n\u003cli\u003eUdayar V, Chen Y, Sidransky E, et al. (2022) Lysosomal dysfunction in neurodegeneration: emerging concepts and methods. Trends Neurosci. 45:184-199. https://doi.org/10.1016/j.tins.2021.12.004.\u003c/li\u003e\n\u003cli\u003eLampinen R, Fazaludeen MF, Avesani S, et al. (2022) Single-Cell RNA-Seq Analysis of Olfactory Mucosal Cells of Alzheimer\u0026apos;s Disease Patients. Cells. 11:676. https://doi.org/10.3390/cells11040676.\u003c/li\u003e\n\u003cli\u003eTollervey JR, Wang Z, Hortob\u0026aacute;gyi T, et al. (2011) Analysis of alternative splicing associated with aging and neurodegeneration in the human brain. Genome Res. 21:1572-1582. https://doi.org/10.1101/gr.122226.111.\u003c/li\u003e\n\u003cli\u003eHsieh YC, Guo C, Yalamanchili HK, et al. (2019) Tau-Mediated Disruption of the Spliceosome Triggers Cryptic RNA Splicing and Neurodegeneration in Alzheimer\u0026apos;s Disease. Cell Rep. 29:301-316.e310. https://doi.org/10.1016/j.celrep.2019.08.104.\u003c/li\u003e\n\u003cli\u003eDuclot F, Kabbaj M (2017) The Role of Early Growth Response 1 (EGR1) in Brain Plasticity and Neuropsychiatric Disorders. Front Behav Neurosci. 11:35. https://doi.org/10.3389/fnbeh.2017.00035.\u003c/li\u003e\n\u003cli\u003eBartolotti N, Lazarov O (2019) CREB signals as PBMC-based biomarkers of cognitive dysfunction: A novel perspective of the brain-immune axis. Brain Behav Immun. 78:9-20. https://doi.org/10.1016/j.bbi.2019.01.004.\u003c/li\u003e\n\u003cli\u003eGokhman D, Livyatan I, Sailaja BS, et al. (2013) Multilayered chromatin analysis reveals E2f, Smad and Zfx as transcriptional regulators of histones. Nat Struct Mol Biol. 20:119-126. https://doi.org/10.1038/nsmb.2448.\u003c/li\u003e\n\u003cli\u003eVerdaguer E, Susana Gde A, Clemens A, et al. (2007) Implication of the transcription factor E2F-1 in the modulation of neuronal apoptosis. Biomed Pharmacother. 61:390-399. https://doi.org/10.1016/j.biopha.2006.11.001.\u003c/li\u003e\n\u003cli\u003eTuerxun M, Muhda A, Yin L (2021) The molecular mechanisms of signal pathway activating effect of E2F-1/NF-\u0026kappa;B/GSK-3\u0026beta; on cognitive dysfunction of Alzheimer rats. Bioengineered. 12:10000-10008. https://doi.org/10.1080/21655979.2021.1989261.\u003c/li\u003e\n\u003cli\u003eJorfi M, Maaser-Hecker A, Tanzi RE (2023) The neuroimmune axis of Alzheimer\u0026apos;s disease. Genome Med. 15:6. https://doi.org/10.1186/s13073-023-01155-w.\u003c/li\u003e\n\u003cli\u003eZhuang X, Zhang G, Bao M, et al. (2023) Development of a novel immune infiltration-related diagnostic model for Alzheimer\u0026apos;s disease using bioinformatic strategies. Front Immunol. 14:1147501. https://doi.org/10.3389/fimmu.2023.1147501.\u003c/li\u003e\n\u003cli\u003eLu Y, Li K, Hu Y, et al. (2021) Expression of Immune Related Genes and Possible Regulatory Mechanisms in Alzheimer\u0026apos;s Disease. Front Immunol. 12:768966. https://doi.org/10.3389/fimmu.2021.768966.\u003c/li\u003e\n\u003cli\u003eJohnson MD, Blitman J (2020) Basophilic invasion in the neurohypophysis is increased in patients with Alzheimer\u0026apos;s disease. Clin Neuropathol. 39:271-274. https://doi.org/10.5414/np301255.\u003c/li\u003e\n\u003cli\u003eShad KF, Aghazadeh Y, Ahmad S, et al. (2013) Peripheral markers of Alzheimer\u0026apos;s disease: surveillance of white blood cells. Synapse. 67:541-543. https://doi.org/10.1002/syn.21651.\u003c/li\u003e\n\u003cli\u003eAhmed-Jushuf F, Jiwa NS, Arwani AS, et al. (2016) Age-dependent expression of VEGFR2 in deep brain arteries in small vessel disease, CADASIL, and healthy brains. Neurobiol Aging. 42:110-115. https://doi.org/10.1016/j.neurobiolaging.2016.03.002.\u003c/li\u003e\n\u003cli\u003eWisniewski HM, Wegiel J, Kotula L (1996) Review. David Oppenheimer Memorial Lecture 1995: Some neuropathological aspects of Alzheimer\u0026apos;s disease and its relevance to other disciplines. Neuropathol Appl Neurobiol. 22:3-11. https://doi.org/10.1111/j.1365-2990.1996.tb00839.x.\u003c/li\u003e\n\u003cli\u003evan de Haar HJ, Burgmans S, Jansen JF, et al. (2017) Blood-Brain Barrier Leakage in Patients with Early Alzheimer Disease. Radiology. 282:615. https://doi.org/10.1148/radiol.2017164043.\u003c/li\u003e\n\u003cli\u003eZenaro E, Pietronigro E, Della Bianca V, et al. (2015) Neutrophils promote Alzheimer\u0026apos;s disease-like pathology and cognitive decline via LFA-1 integrin. Nat Med. 21:880-886. https://doi.org/10.1038/nm.3913.\u003c/li\u003e\n\u003cli\u003eWang F, Feng TY, Yang S, et al. (2016) Drug Therapy for Behavioral and Psychological Symptoms of Dementia. Curr Neuropharmacol. 14:307-313. https://doi.org/10.2174/1570159x14666151208114232.\u003c/li\u003e\n\u003cli\u003eLane CA, Hardy J, Schott JM (2018) Alzheimer\u0026apos;s disease. Eur J Neurol. 25:59-70. https://doi.org/10.1111/ene.13439.\u003c/li\u003e\n\u003cli\u003eObara K, Mori H, Ihara S, et al. (2024) Inhibitory Actions of Antidepressants, Hypnotics, and Anxiolytics on Recombinant Human Acetylcholinesterase Activity. Biol Pharm Bull. 47:328-333. https://doi.org/10.1248/bpb.b23-00719. \u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Alzheimer’s disease, Small ubiquitin-related modifier-related genes, Bioinformatics analysis, Biomarkers","lastPublishedDoi":"10.21203/rs.3.rs-6983921/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-6983921/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e\u003cp\u003eTo explore the role of small ubiquitin-related modifier (SUMO) in Alzheimer's disease (AD) and its pathogenesis, bioinformatics was used to search for SUMO-related genes (SRGs)-based biomarkers.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e\u003cp\u003eDatasets related to AD (GSE140831, GSE63060), a dementia dataset (GSE140830), and 189 SRGs were retrieved from public databases. Candidate genes were identified by intersecting differentially expressed genes (DEGs) with SRGs. A protein-protein interaction (PPI) network was constructed to select the top 15 core genes, and the support vector machine-recursive feature elimination (SVM-RFE) model identified feature genes. Validation was done using the GSE140831 and GSE63060 datasets, and the nomogram model was assessed by receiver operating characteristic (ROC) curve analysis. Gene set enrichment analysis (GSEA) and other analyses were performed. Reverse transcription quantitative polymerase chain reaction (RT-qPCR) was used for further validation.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e\u003cp\u003eOverlapping 189 SRGs and 12,853 DEGs identified 107 candidate genes. Five feature genes were selected using the SVM-RFE algorithm. \u003cem\u003eCREBBP\u003c/em\u003e, \u003cem\u003ePIAS1\u003c/em\u003e, and \u003cem\u003eTRIM28\u003c/em\u003e were confirmed as AD biomarkers due to their increased expression in AD and strong ROC performance. GSEA highlighted their involvement in pathways such as olfactory transduction, lysosome, and spliceosome. Immune infiltration analysis suggested immune cell involvement in AD progression. Additionally, 21 potential drugs for AD therapy were predicted. RT-qPCR confirmed the over-expression of CREBBP and TRIM28 in AD samples, consistent with dataset trends.\u003c/p\u003e\u003ch2\u003eConclusion\u003c/h2\u003e\u003cp\u003e\u003cem\u003eCREBBP\u003c/em\u003e, \u003cem\u003ePIAS1\u003c/em\u003e, and \u003cem\u003eTRIM28\u003c/em\u003e were identified as SRG-based biomarkers for AD diagnosis, providing new insights into AD pathogenesis.\u003c/p\u003e","manuscriptTitle":"Identification of small ubiquitin-related modifier (SUMO)-related genes-based biomarkers in Alzheimer's disease based on bioinformatics analysis","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-10-28 16:39:33","doi":"10.21203/rs.3.rs-6983921/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2025-12-15T14:34:57+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-12-01T13:20:52+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"240339921284852244312706395111682745031","date":"2025-11-23T10:09:09+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-11-06T20:27:31+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"329693784423128883334197189224826194430","date":"2025-10-16T18:26:07+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"226208849757938470877965069197273915243","date":"2025-10-16T08:32:05+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-10-14T08:01:49+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2025-10-08T07:27:48+00:00","index":"","fulltext":""},{"type":"editorInvited","content":"","date":"2025-07-08T07:25:05+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-07-02T13:54:03+00:00","index":"","fulltext":""},{"type":"submitted","content":"Scientific Reports","date":"2025-07-02T13:50:26+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"scientific-reports","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"scirep","sideBox":"Learn more about [Scientific Reports](http://www.nature.com/srep/)","snPcode":"","submissionUrl":"","title":"Scientific Reports","twitterHandle":"","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"stoa","reportingPortfolio":"Scientific Reports","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"5879c4a5-c367-4b6a-ab5c-579b0b52159b","owner":[],"postedDate":"October 28th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[{"id":56817433,"name":"Health sciences/Biomarkers"},{"id":56817434,"name":"Biological sciences/Computational biology and bioinformatics"},{"id":56817435,"name":"Health sciences/Diseases"},{"id":56817436,"name":"Biological sciences/Neuroscience"}],"tags":[],"updatedAt":"2026-04-27T16:05:04+00:00","versionOfRecord":{"articleIdentity":"rs-6983921","link":"https://doi.org/10.1038/s41598-026-49884-3","journal":{"identity":"scientific-reports","isVorOnly":false,"title":"Scientific Reports"},"publishedOn":"2026-04-20 15:59:03","publishedOnDateReadable":"April 20th, 2026"},"versionCreatedAt":"2025-10-28 16:39:33","video":"","vorDoi":"10.1038/s41598-026-49884-3","vorDoiUrl":"https://doi.org/10.1038/s41598-026-49884-3","workflowStages":[]},"version":"v1","identity":"rs-6983921","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-6983921","identity":"rs-6983921","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}