PTAFR Exaggerates Microglia-Mediated Microenvironment via IL10-STAT3 Signaling: A Novel Perspective on Potential Biomarker and Target for AD Diagnosis and Treatment

Background: Early diagnosis and effective intervention become the key points for delaying Alzheimer’s progression. Therefore, we aim to nd new biomarkers for early diagnosis of Alzheimer (AD) by bioinformatics, and to reveal the possible mechanisms. Methods and results: GSE1297, GSE63063, and GSE110226 in the GEO database were used to screen out the differentially highly expressed genes. We found out a potential biomarker PTAFR differentially highly expressed in the brain tissue, peripheral blood and cerebrospinal uid of AD patients. Furthermore, we found higher PTAFR level in the plasma and brain tissues of APP/PS1 mice. Simultaneously, it was uncovered that PTAFR mediated the inammatory response to exaggerate microenvironment specially mediated by microglia through the IL10-STAT3 pathway. In addition, we also found PTAFR was a possible target for some anti-AD compounds such as EGCG, donepezil, curcumin, memantine, and Huperzine A. Conclusions: PTAFR was a potential biomarker for early AD diagnosis and treatment correlated with microglia-mediated microenvironment, and an important possible target to make novel strategy for clinical treatment and new drug discovery in AD.


Introduction
As a neurodegenerative disease closely related to age, Alzheimer's Disease (AD) seriously endangers the lives of the elderly [1,2]. Cognitive de cits, memory loss, and language dysfunction were the major clinical characteristics of AD [3]. And the hallmark pathophysiological changes include senile plaques formed by Aβ deposition in the brain, and the neuro brillary tangles cause by tau hyperphosphorylation, gliosis and etc [4,5]. Once clinical symptoms occur, it is hard to reverse the procession [6,7]. Therefore, early diagnosis and treatment are overwhelmingly signi cant for delaying the progression of AD.
The clinical methods commonly used for AD diagnosis mainly include MMSE score, 18FDG-PET, CT or MRI scanning, EEG and biomarkers in cerebrospinal uid of AD patients nowadays. Despite MMSE score is a convenient and low-cost method, it is usually used for screening moderate to severe AD. 18 FDG-PET, CT or MRI scanning are often used for excluding other diseases through imaging and improving the reliability of AD diagnosis, which are inconvenience and high-cost. EEG is insensitive for detecting early AD. And the changes of biomarkers in cerebrospinal uid of AD patients seem more believable for AD diagnosis, but the sample collection is invasive, compared with those of other body uids such as blood or urine [7,8]. Therefore, nding out potential biomarkers with high speci city in the blood is an effective way for early diagnosis of AD. Ideal AD biomarkers should be able to predict the incipient pathophysiological changes in AD brains and CSF, and can be simultaneously detected in peripheral body uids such as blood. Also, high sensitivity and convenience for detection are necessary. As the brain tissues of AD patients are hard to obtain, it is limited to get enough information in AD brains. Therefore, in this study, we use GEO database to further analyze the differentially expressed genes (DEGs) in AD brains, CSF and blood. We nally screened out a high-expressed DEG called PTAFR closely related to AD progression. Furthermore, its predictive e cacy and the possible mechanism involved in AD were also validated and investigated in APP/PS1 mice model and in LPS + Aβ-induced BV2 cells.

Conversion and difference analysis of raw data
The GEO2R interactive online tool (https://www.ncbi.nlm.nih.gov/geo/geo2r) was applied to convert the raw data into a recognizable format. The differential expression genes (DEGs) were identi ed with P <0.05 and |logFC|>1 as threshold values. The expression of genes obtained by VENN intersection was corrected by Bonferroni method.

GO and KEGG pathway enrichment analysis and protein interaction analysis
The online enrichment platform David (http://david-d.ncifcrf.gov/) was used to conduct enrichment analysis of KEGG pathway and GO function on DEGs and screen the top-10 signi cant biological pathways with P<0.05. Then, we used the bisoGenet plug-in in Cytoscape3.6.1 software to analyze the protein interaction of DEGs. Simultaneously, the protein interactions with the target gene PTAFR were evaluated through the String online tool (https://string-db.org/).

Brain samples and blood samples collection
C57BL/6 mice are sourced from the Experimental Animal Center of China Medical University, and APP/PS1 transgenic mice are from Jackson laboratory. All animal care and experimental procedures are in compliance with the "Ethical Standards for Animal Laboratory Animals" of China Medical University.
We collected the hippocampus of 12-month-old APP/PS1 mice (n=5) and 12-month-old C57BL/6 mice (n=5). We chose to study only female mice because incidence of AD is biased towards female [9]. The right hemisphere was used for Western Blotting, while the left hemisphere was immersed in paraformaldehyde for immuno uorescence staining. Blood was obtained from the orbit. Then, the blood samples were centrifuged at 3000 rpm for 10 minutes at room temperature. The bottom layer was collected and used for RNA extraction.

Cell Culture
The cells used in our experiment are BV2, SH-SY5Y, SVGp12. Mouse-derived microglial BV2 cells and human-derived neuroblastoma SH-SY5Y cells were purchased from the Institute of Basic Medicine Chinese Academy of Medical Sciences while human-derived astrocytes SVGp12 were purchased from Bei Na Chuang Lian company. Both BV2 and SVGp12 cells were cultured in DMEM medium, and SH-SY5Y cells were cultured in DMEM/F12 medium. All cells need to add 10% fetal bovine serum (FBS) and 100U/ml penicillin-streptomycin, and place them in a 37°C, 5% CO2 incubator for culture. We use cells of passage 5-20 for experiments.

Transfection and treatment of BV2 cells
We transfected the si-PTAFR plasmid (purchased from Shen Gong Bioengineering Co., Ltd.) into BV2 cells as required, and the transfection reagent was Lipofectamine 3000 (purchased from Thermo). In the nucleotide sequence of si-PTAFR, the sense strand is GCUAUGGGUCUUUGCUAACUUTT; the anti-sense strand is AAGUUAGCAAAGACCCAUAGCTT. When the cell density was about 60%-70%, cells were transfected by using Lipofectamine 3000 according to the manufacturer's instructions for 24 hours.
1μg/ml of LPS and 10μM/L of Aβ (Aβ25-35 were placed in a 37°C incubator for 7 days before use) were incubated for another 24h, then collected the corresponding proteins and mRNAs and stored them at -80°C.

Real-time PCR
We used the reverse transcription kit (Evo M-MLV RT Premix for qPCR) to reverse transcribe the extracted total RNA into complementary cDNA, and then conduct the qPCR kit (SYBR Green Premix Pro Taq HS qPCR Kit) according to the instructions required by the system (5µl SYBR, 0.2µl upstream and downstream primers, 0.2µl ROX, 1µl cDNA, 3.4µl DEPC water) for real-time quantitative PCR detection.
The primers listed in this article are PTAFR, IL10, STAT3, and IL6. All primers were obtained from Sangon Biotech. The primer sequences are shown in Table 1. The results were processed by 2 -△△CT method to compare the relative expression of RNA. The BV2 cells and brain tissue were homogenized with protein lysate containing protease inhibitors and the proteins were extracted, then quanti ed with a BCA kit (Bi Yun Tian company). Equal amounts of protein were separated by SDS-polyacrylamide gels and transferred to polyvinylidene uoride membranes. Then, the membrane was incubated in a blocking solution (a mixture of TBST containing 0.1% Tween-20 and 5% BSA) for 1 h at room temperature, and placed in a primary antibody containing PTAFR(Abcam,1:200), IL10(Wanlei,1:1000), STAT3(CST,1:1000), IL6(Wanlei,1:1000), MAP2(CST,1:1000), and Syn (CST,1:1000) at 4°C overnight. The next day, after washing with TBST, the membrane was incubated with the corresponding HRP secondary antibody. The immune response band was observed by ECL with luminescence and quanti ed by measuring the density of each band using Image-J software.

CCK8 detection
The cell survival viability was assessed by CCK8 assay. First, SH-SY5Y cells were spread in a 96-well plate (n=5000 cells/well) and cultured for 24 hours. Then, the medium was changed to the transfected and modeled BV2 cell supernatant for 1h. 1μg/ml LPS and 10μM/L Aβ were given, and the culture was continued for 24h. After that, change the medium to serum-free DMEM medium containing 10% CCK8 reagent, 100µl per well, and incubate at 37°C for 2h. Finally, use a microplate reader to detect the absorbance at 450nm wavelength, and calculate the cell viability.

Flow cytometry to measure apoptosis
For apoptosis assays, SH-SY5Y cells were seeded in a 6-well plate (n=2*10 5 cells/well) and cultured for 24 hours. Then, the medium was changed to the transfected and modeled BV2 cell supernatant for 1h.
1μg/ml LPS and 10μM/L Aβ were given, and the culture was continued for 24h. Cells were washed twice with pre-cooled PBS, incubated with Annexin V-FITC and PI in the dark. Ultimately, cell apoptosis was analyzed by ow cytometer (BD Company).
11. Immuno uorescence BV2 cells and SH-SY5Y cells were inoculated on a sterile cover glass into 12-well plates. After transfection and conditioned culture, the cells were xed with 4% paraformaldehyde at room temperature.
After washing with PBS, the cells were permeabilized with 0.5% TritonX-100 for 20 minutes. The cells and brain tissue were blocked with goat serum and stained with PTAFR or MAP2 in a humid box at 4°C overnight. Afterward, they were incubated with TRITC-conjugated rabbit anti-goat IgG for 1 h at 37°C in a dark and humid box, followed by counterstaining with DAPI. Finally, the immuno uorescence image was obtained by laser scanning of a confocal microscope.

MOE molecular docking
The protein secondary structure of PTAFR were downloaded from the PDB website (http://www1.rcsb.org/) while the three-dimensional structure of EGCG, donepezil, curcumin, memantine, and Huperzine A were acquired from the PubChem website (https://pubchem.ncbi.nlm.nih.gov/). Then, we imported the protein and small molecule drug structure into the MOE (2018 version) software and convert the protein secondary structure into the tertiary structure. At last, we perform molecular docking between protein and drugs after the small molecule drug is optimized.

Statistical Analysis
All data are statistically analyzed using GraphPad Prism 8.0.1 version. The data are expressed as mean ± standard deviation. Differences between groups were evaluated by one-way analysis of variance. All determinations were repeated three times. P<0.05 was considered statistically signi cant.

DEGs was found closely related to AD progression by further analysis of GSE1297
To obtain the DEGs related to AD progression, we found GSE 1297 chip from GEO database. The GSE 1297 included the expression pro ling of brain hippocampus from 22 postmortem samples of AD patients at various stages of severity (incipient, moderate, and severe AD). The DEGs were screened out with the criteria (|logFC|>1, P < 0.05), shown as Fig. 1a. There were 174 up-regulated DEGs and 54 downregulated DEGs in incipient AD; 270 up-regulated DEGs and 157 down-regulated DEGs in moderate AD; 688 up-regulated DEGs and 367 down-regulated DEGs in severe AD (Fig. 1b). Then, VENN graph network tool (https://bioinfogp.cnb.csic.es/tools/venny/index.html) was employed to nd the DEGs signi cantly changed in all the stages of AD, and 32 DEGs were obtained nally (Fig. 1c, Table 2). They were further analyzed by heat map analysis by the Sangerbox program (Fig. 1d). It was shown that 25 genes were signi cantly up-regulated throughout the various stages of incipient, moderate and severe AD, including :  CR1, OGFOD3, AKAP13, MRPS12, ZDHHC17, ERF, UMOD, GRP107, PTGER4P2-CDK2AP2P2, HRP, RAD51B,  NPAT, FGF20, RPL21P28, PTAFR, IL9R, AVPR2, LTB4R, PMS2P9, MYRF, SLC16A5, ATP11A, ITGB3, BGN,  LOC389906. And 6 genes such as TNFRSF25, E2F5, BICD2, RFXAP, TAC1, B4GALT6 were signi cantly down-regulated in each stage of AD. Only ITGB1 gene was up-regulated in incipient and moderate AD, while down-regulated in severe AD (Table.2). Therefore, considering for the characteristics of biomarkers, we focused on the 25 up-regulated DEGs in each stage of AD. In order to identify the dominant pathways of the screening DEGs and their biological roles, we performed enrichment analysis of the KEGG pathway and GO function by David network enrichment tool (http://david-d.ncifcrf.gov/) and obtained top-10 biological pathways with statistical signi cance (Fig. 2a). The biological processions were hematopoietic cell lineage, viral entry into host cells, mesodermal cell differentiation, in ammatory response, cell adhesion mediated by integrin, cell-substrate adhesion, heterotypic cell-cell adhesion, positive regulation of cell proliferation, extracellular matrix organization and leukocyte cell-cell adhesion. Afterwards, we also explored the interactions between the 32 DEGs using the bisoGenet plug-in of Cytoscape3.6.1 software. It was found that 22 genes had the possible interactions with core pathogenic genes APP of AD, such as PTAFR, BGN, E2F5, BICD2, SLC16A5, AVPR2, TNFRSF25, IL9R, RFXAP, AKAP13, CR1, NPAT, RAD51B, GPR107, ITGB1, ITGB3, ERF, MRPS12, TAC1, OGFOD3, LTB4R, and ZDHHC17 (Fig. 2b).
Up-regulated PTAFR in peripheral blood, cerebrospinal uid and hippocampus of AD patients As we obtained 32 DEGs in the hippocampus of AD patients in different stages, it was considered for the possible application as potential biomarkers for AD diagnosis. The potential biomarkers should signi cantly change with the occurrence and progression of AD, it is essential the changes of DEGs in hippocampus synchronize with the changes in blood and cerebrospinal uid (CSF). Therefore, we found out the CSF chip (GSE110226) and the blood chip (GSE63063) for the further investigation. VENN graph network tool was used for screening the target DEGs changed in blood, CSF and hippocampus of AD. We nally got two DEGs, PTAFR and AKAP13 (Fig. 3a). Then, the possible correlations between the mRNA expressions of PTAFR and AKAP13 with MMSE scores were examined. We found that the mRNA expression of PTAFR was correlated with the MMSE score (P = 0.0006), while that of AKAP13 was not signi cantly correlated with MMSE score (P = 0.1308), shown in Fig. 3b. Also, the Braak staging was more strongly correlated with PTAFR than that with AKAP13 (PTAFR, P < 0.0001; AKAP13, P = 0.0274; Fig. 3c). We also found the expression of PTAFR was positively correlated with neuro brillary tangles scores than AKAP13(PTAFR, P = 0.0015; AKAP13, P = 0.1078; Fig. 3d). The mRNA expression of PTAFR was also gradually increased with AD progression from incipient to severe AD (P < 0.05, Fig. 3e), while AKAP13 did not follow this manner (Fig.S1a). Besides, both PTAFR and AKAP3 were all increased in both choroid plexus and peripheral blood (Fig. 3f & Fig.S1a). PTAFR was also highly expressed in entohinal cortex, hippocampus, and temporal cortex in AD brains, while AKAP13 was just highly expressed in hippocampus (Fig. 3h & Fig.S1c). Furthermore, we investigated the possible mechanism of PTAFR involved in AD progression. Using the AlzData online network platform (http://www.alzdata.org) [10], the PTAFR is only highly expressed in microglia which contributes with in ammatory response in the pathogenesis of AD while AKAP13 is expressed in many cell lines (Fig. 3g, S1b). Taken together, our ndings suggested that PTAFR might be the target biomarkers with higher e cacy for AD diagnosis.
Further validation for the highly-expressed PTAFR in 12month-old APP/PS1 mice, and in BV2 cells Our previous results of GEO database indicated PTAFR was a possibly potential biomarker highly expressed in AD patients, so next we validated it both in vivo and in vitro. We performed the validation on a 12-month-old APP/PS1 double transgenic AD mouse model. As we expected, it was showed that the PTAFR in APP/PS1 mice was signi cantly increased in both mRNA and protein levels (P < 0.01, Fig. 4a, b, c), compared with the age-matched C57 BL/6J mice. The peripheral blood levels of PTAFR in APP/PS1 mice was also signi cantly increased (P < 0.01, Fig. 4a).
Also, we validated the results in vitro. As we previously mentioned, the predictive results on AlzData platform showed PTAFR was speci cally highly expressed in microglia compared with other neural cell in central nervous system. A homologous sequence alignment of murine and human PTAFR gene were performed, and it was indicated the PTAFR gene of murine and homo sapiens are in high homology (Fig.  S2c). Therefore, we examined the PTAFR mRNA and protein levels in different neural cell such as BV2 (microglia-like), SH-SY5Y(neuron-like), and SVGp12 (astrocyte-like) cell lines. We found that after LPS + Aβ treatment, the mRNA and protein levels of PTAFR is signi cantly elevated in BV2 cells (P < 0.01, Fig. 4a, b, d), while there were no signi cant changes of SVGp12 or SH-SY5Y cells compared with control group (P > 0.05, Fig. S2a, b), indicating that PTAFR was high-expressed gene in microglia.
By literature reviewing, PTAFR is closely related to the secretion of in ammatory factors in kidney injury and retinal neovascularization [11][12][13], indicating neuroin ammation mediated by microglia might be the possible mechanism of PTAFR involved in AD progression. First, we found that PTAFR is closely associated with the in ammatory factors IL-10 and STAT3 by the STRING online tool (https://stringdb.org/) (Fig. 5a). Extensive studies revealed that the IL10-STAT3 pathway was involved in the occurrence of many diseases [14][15][16][17][18][19]. Then, we hypothesized PTAFR kindled the microglia-mediated neuroin ammation via IL10-STAT3 signaling and exaggerated the microenvironment of neurons in the procession of AD. After LPS and Aβ treatment, the mRNA and protein levels of IL-10 were decreased, and those of STAT3 and IL-6 were elevated, following the PTAFR up-regulation (P < 0.05, Fig. 5b-d). Further, we silenced PTAFR gene in BV2 cells, and found that the mRNA and protein levels of IL-10 were increased, while those of STAT3 and IL-6 were decreased (P < 0.05, Fig. 5b-d).
Then, the conditional medium (CM) of the treated BV2 cells was given to SH-SY5Y cells treated with LPS + Aβ to investigate the subsequent in ammatory e cacy of microglia on neurons. The CM of LPS + Aβinduced BV2 cells signi cantly reduced the cell viability of SH-SY5Y cells (P < 0.0001, Fig. 5e-f), while the CM of LPS + Aβ-induced BV2 cells after silencing PTAFR improved the cell viability of SH-SY5Y cells in both CCK8 experiment and Annexin V/PI ow cytometry (P < 0.001, Fig. 5e-f). Besides, the CM also remarkedly enhanced the expression of neuroplasticity indexes MAP2 and Syn (Fig. 5g, h). Taken together, PTAFR, as a potential biomarker, exaggerated microglia-mediated microenvironment by increasing the in ammatory factors via IL10-STAT3 signaling.

Targeted docking of PTAFR with anti-AD drugs commonly used in clinical practice
As we mentioned previously, PTAFR was veri ed to play an important role in exaggerating the microgliamediated neuronal microenvironment by IL10-STAT3 signaling, and closely correlated with AD progression. That is to say, it could be considered as a potential biomarker or an essential point for R&D of anti-AD drugs. Here, we made targeted molecular docking of PTAFR with several drugs recognized in clinical and scienti c research for AD treatment, such as donepezil, memantine, EGCG, curcumin, and Huperzine A. The molecular docking was performed by MOE software. S value obtained after docking is used for evaluation of the possible binding. If S<-7 it is considered to have the probability of binding with each other. EGCG is considered as a potential compound with anti-AD e cacy, the docking result showed it could possibly be bound with PTAFR (S=-7.7826, Fig. 6a). Donepezil is a medication used to treat mild or moderated AD, and commonly used in clinical practice. We also found the possible bind sites (S=-7.5199, Fig. 6b). Curcumin, another compound found to have neuroprotective e cacy, can bind with PTAFR, which S=-7.5698 (Fig. 6c). However, the S values of memantine and Huperzine A were − 5.3495 and S=-5.3781, respectively (Fig. 6d, e). It was also indicated that those compounds with planar structure with multiple benzene ring such as EGCG, curcumin, and donepezil seemed to be more likely to bind with PTAFR, while those compounds with stereo conformation, such as memantine and huperzine A, have signi cantly decreased binding effects on PTAFR. These results indicated that PTAFR could be combined with some anti-AD drugs, and might be used as a potential target for the treatment of AD in the future.

Discussion
The pathophysiological changes of AD often precede the major clinical symptoms such as cognitive dysfunction, as well as the characteristic pathological changes such as Aβ deposition and Tau hyperphosphorylation [20]. Therefore, early diagnosis of AD and the intervention targeting initial events of AD will bene t the prognosis and effectively improve the quality of life. At present, AD diagnosis mainly relies on MMSE score, 18 FDG-PET, CT or MRI, and the identi cation of T-tau, p-tau, Aβ42 and other biomarkers in cerebrospinal uid [7,21,22]. However, they still have shortcomings such as high cost, lack of speci city, or invasive detection. Therefore, it is of great signi cance to nd candidate biomarkers for early diagnosis of AD and to develop intervention strategies for early diagnosis and treatment of AD.
Here, we discovered a potential biomarker named PTAFR screening out of GEO database and validated its e cacy in vivo and in vitro.
The ideal AD biomarker should be able to warn early AD brain lesions, detect in peripheral body uids, have correlation with brain lesions, and be sensitive and simple in detection. Human brain tissue is extremely di cult to obtain and the related ethics are limited, so we screened the potential candidates in GEO database. 32 DEGs related to the progress of AD were got by GSE1297, they also closely related to in ammation processes, and 22 of them interacted with the core disease-causing gene APP of AD.
Castillo et al found that the expression of genes closely related to in ammatory response were signi cantly increased in APP NL−GF/NL−GF mice [23]. Venegas et al found that the activation of in ammasomes is closely related to the formation and progression of Aβ plaques in AD [24]. Welikovitch's study proved that the neuron-speci c in ammatory response may be earlier than the formation of Aβ plaques [25]. These above studies suggested that there was a close correlation between the in ammatory response and Aβ deposition in the end-stage AD, which also supported our results. To nd ideal biomarkers re ecting the same changes in peripheral uids as in the brain tissues, we further found a peripheral blood chip GSE63063 and the choroid plexus chip GSE110226. PTAFR was nally found to correlate with the severity of AD MMSE score, Braak staging and neuro brillary tangle scores synchronously. However, there is no report related to PTAFR to reveal the predictive role in AD. Therefore, PTAFR gene became the focus of our follow-up research. We performed veri cation on 12-month-old APP/PS1 mice and found that PTAFR was highly expressed in the brain and peripheral blood of APP/PS1 mice, which was consistent with the predicted results, suggesting that PTAFR had the potential to be a possibly candidate biomarker for AD diagnosis.
As a platelet activating factor receptor, PTAFR plays an important role in many diseases. It has been reported in the literature that the expression of PTAFR in breast cancer cells and osteoclasts was signi cantly increased, and when PTAFR expression was down-regulated, the breast cancer cell migration and osteoclast production were prevented in breast cancer bone metastasis models [26]. Besides, PTAFR down-regulation was also corelated to the proliferation of cardiac broblasts and the deposition of collagen, nally inhibiting the brous bers after myocardial infarction in cardiac broblasts treated with angiotensin II [27]. However, PTAFR is rarely involved in the neurological eld, and the role and pathogenic mechanism in AD have not been reported. Specially, PTAFR was highly expressed in microglia, while no increased expression was found in astrocytes, oligodendrocytes, or neurons by AlzData website. Further, we also found PTAFR had the speci c high-expression in BV2 cells. So, we illustrated that the function of PTAFR gene in central nervous system correlated with microglia-mediate biological process. Microglia contributes to the neuroin ammation and mediates the microenvironment of neurons. Previous studies also found that PTAFR distributed on the surface of vascular endothelial cells could increase IL-1β expression, leading to in ammatory-dependent vascular occlusion in the ischemic retinopathy model [13]. In kidney injury model, PTAFR is highly expressed, and further aggravated the kidney injury by inducing the expression of the in ammatory factor TNFα [12]. Besides, in the renal brosis model induced by a large amount of ethanol intake, the expression of the in ammatory factor TGFβ and closely related indicators of renal brosis were signi cantly decreased in the PTAFR knockout mice, and the renal brosis was inhibited [11]. Furthermore, we also found that the expression of PTAFR was signi cantly increased in BV2 cells induced by LPS + Aβ, suggesting that PTAFR might be involved in the microgliamediated in ammation of AD. However, almost no studies were reported about the possible mechanism of PTAFR in AD. We analyzed the possible interacting proteins with PTAFR by STRING online tool, and found IL10-STAT3 might closely related with PTAFR. Our study revealed IL-10 was increased, and STAT3 and IL-6 were increased after silencing PTAFR in BV2 cells. After conditioned culture of SH-SH5Y cells with treated BV2 cells, the survival and plasticity of SH-SH5Y cells were enhanced after PTAFR silence, suggesting PTAFR promoted the microglia-mediated neuroin ammation through IL10-STAT3 signaling and exaggerated the neuronal microenvironment in AD. IL-10 and STAT3 is involved in the occurrence and development of many diseases, such as Alzheimer's disease, breast cancer, glioma, autosomal dominant hereditary high lgE syndrome and so on[28-31]. Kiyota's research found that IL-10 signi cantly reduced neuroin ammation, enhanced neuron generation, and improved spatial cognitive impairment in APP/PS1 mice [32]. Reichenbach's research con rmed that APP/PS1 transgenic mice with STAT3 knockout reduced Aβ plaque deposition in the brain, inhibited astrocyte proliferation and secretion of pro-in ammatory factors, and improved learning and memory impairment [33]. The above studies suggested that IL-10 and STAT3 were involved in the AD in ammatory response, supporting our ndings. Therefore, it is the rst report to reveal the relationship between PTAFR and IL10-STAT3 signaling in AD.
Since PTAFR was con rmed to be used as a potential biomarker for AD diagnosis, and as a possible target closely related to in ammation, can it be used as a potential target for drug therapy, or as a target to develop new anti-AD drugs? For the rst time, we conducted MOE molecular docking between PTAFR and several drugs or compounds that are recognized in clinical and scienti c research for the treatment of AD, and we found that PTAFR and these anti-AD drugs had a certain degree of combination. Among them, PTAFR has the highest binding degree with EGCG, donepezil and curcumin, indicating that the treatment of EGCG, donepezil or curcumin could interact with PTAFR to a certain extent, thereby exerting anti-in ammatory effects in AD. The above results further highlighted the potential signi cance of PTAFR as an AD biomarker and therapeutic target, and provided a possible reference for the molecular design of new anti-AD drugs.
In summary, PTAFR was a potential biomarker for early AD diagnosis and treatment correlated with microglia-mediated microenvironment, and an important possible target to make novel strategy for clinical treatment and new drug discovery in AD.

Conclusions
As a gene that is highly expressed in AD brain tissue, peripheral blood and cerebrospinal uid, PTAFR were found to be a potential candidate biomarker for AD diagnosis. And PTAFR was highly-expressed in microglia, and induced the neuron in ammatory response to exaggerate the microenvironment of AD neurons via IL10-STAT3 signaling in vivo and in vitro. It also provided us a novel intervention target for AD treatment, and a possible reference for the molecular design of new anti-AD drugs.

Declarations Data Availability
The datasets during and/or analyzed during the current study are available from the corresponding author on reasonable request.