Oligomer‐Aβ42 suppress glioma progression via potentiating phagocytosis of microglia

Abstract Aims Glioma is characterized by an immunosuppressed environment and a poor prognosis. The accumulation of Amyloid β (Aβ) leads to an active environment during the early stages of Alzheimer's disease (AD). Aβ is also present in glioma tissues; however, the biological and translational implications of Aβ in glioma are elusive. Methods Immunohistochemical (IHC) staining, Kaplan–Meier (KM) survival analysis and Cox regression analysis on a cohort of 79 patients from our institution were performed to investigate the association between Aβ and the malignancy of glioma. Subsequently, the potential of oligomer‐Aβ42 (OAβ42) to inhibit glioma growth was investigated in vivo and in vitro. Immunofluorescence staining and phagocytosis assays were performed to evaluate the activation of microglia. Finally, RNA‐seq was utilized to identify the predominant signaling involved in this process and in vitro studies were performed to validate them. Results A positive correlation between Aβ and a favorable prognosis was observed in glioma. Furthermore, OAβ42 suppressed glioma growth by enhancing the phagocytic activity of microglia. Insulin‐like growth factor 1 (IGF‐1) secreted by OAβ42‐activated microglia was essential in the engulfment process. Conclusion Our study proved an anti‐glioma effect of Aβ, and microglia could serve as a cellular target for treating glioma with OAβ42.

Glioma is the most prevalent primary brain tumor, and its treatment has struggled for decades.Despite conventional chemotherapy, radiotherapy, and surgery, glioma's prognoses remain poor.Although the current immunotherapy achieved success in the peripheral system and have also been developed against glioma.Not surprisingly, current clinical trials of immunotherapies observed limited benefits for patients. 1Compared to the peripheral system, the brain's immune system has a very distinct structure, with microglia dominating while T cells are scarce. 2Microglia are the fundamental innate immune cells developmental, originating from the yolk sac within the brain. 3Numerous pieces of evidence have proved that microglia can mediate the immune surveillance of our brain by engulfing potential pathogens or mutant cells.Microglia also directly recognize, uptake, and degrade tumor cells through phagocytosis. 4,5However, the primitive phagocytosis function of microglia was masked in the glioma environment. 6,7Therefore, The efficient clearing of tumor cells by microglia is a promising strategy in glioma treatment.
The early stage of AD has an activated immune environment. 8,9It is asymptomatic and distinct from the late stage with declined cognitive function.Moreover, the time course from the early stage to the late stage can be over 10 years. 102][13][14][15][16] The representative work by Phillip P. Wolf from the heart center showed a 60% decreased cancer rate (including brain tumors) from AD patients. 11The observational evidence, the similar immune organism structure, and the status of microglia support the hypothesis that AD-associated microglia are unfriendly to glioma. 17In this circumstance, the translational study based on the hypothesis can be valuable, and the immune agonist for the microglia activation could be essential. 18, the dominant metabolites of AD, accumulate in distinct phenotypes over the clinical stages of AD-the mono, oligomer, and fibrillar type with 40 or 42 amino acids of Aβ. 19 Oligomer-Aβ 42 (OAβ42) can promote an inflammatory activation of microglia in the early stage of AD, representing the immune feature of the AD brain.20 Besides activating microglia in early AD, the Aβ can be regarded as a systemic peptide with many functions, including antibacterial, BBB protection, and Calcium homeostasis.The Aβ also associated with cancer.Previous study have revealed that rats with naked mole phenotypes never develop tumors and accumulate Aβ at levels similar to AD mice.21,22 Another group proved that OAβ42 could inhibit glioma growth via degenerating vascular capillaries. 23Indeed, Aβ has been demonstrated to exist in human and mouse glioma tissues.24,25 However, the role of Aβ was not assessed in glioma, and whether OAβ42 could be anti-glioma needs to be evaluated thoroughly.In this study, we assessed the clinical role of OAβ42 in clinical samples and public datasets.We also identified its potential application against glioma progression and the underlying mechanism.

| Patients
We retrospectively reviewed the medical documents of 79 patients diagnosed with glioma (except for pediatric subtypes) submitted to the Department of Neurosurgery, the Sun Yat-sen University Cancer Center, from February 2011 to November 2017.The overall survival (OS) was defined as the period from the day of surgical operation to death or the latest follow-up data updated on December 31, 2022.
All clinical and pathological data were recorded, including the age and clinical diagnosis, histological pattern, and OS (Table S1).Univariate and multivariate Cox Regression Analyses were conducted to identify the glioma risk factor based on the OS.Kaplan-Meier (KM) survival analysis was performed to compare the OS outcomes of the Aβ positive and negative groups in the glioma cohort and GBM cohort from our center.

| Immunohistochemistry analysis
10% formalin-fixed glioma tissues were embedded in paraffin and cut into sections.The immunohistochemistry staining was conducted using an immunostaining system (BenchMark ULTRA system, Ventana-Roche, Switzerland) after deparaffinized, rehydrated, inactivated endogenous peroxidase, and antigen retrieval.The slides of tissue sections were treated with Aβ1-42 antibody (at a concentration of 1:1000) and left to incubate overnight at a temperature of 4°C, then staining with 3,3′-diaminobenzidine and counterstaining with hematoxylin, the specimen was dehydrated, xylene-treated, and mounted.3 peripheral glioma tissues from the 79 patients' cohort were applied as the normal control.3 experienced neuropathologists examined and reviewed all slides, and representative pictures were captured with a Leica DM6 microscope (Leica).

| Preparation of oligomeric amyloid β
We purchased human recombinant Aβ1-42 (purity>95%) from Meilunbio, Dalian.Lyophilized peptide was diluted in 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) 1 mg/mL at 4°C in order to decrease the sheet structures' formation and the helical structures' stable.The peptide was placed in a chemical fume hood to air dry for 1 h at room temperature, and further drying in a Speed Vac (Thermo-Savant) for 30 min.A clear film was formed after resuspension in 100% dimethylsulfoxide (DMSO) at 1 mM, aliquoting, and storing at −80°C.To assemble the oligomers, the peptide film was resuspended in DMSO at 5 mM with water bash ultrasonic for 10 min, diluted to 100 mM in F-12 medium (phenol red-free) and incubated for 24 h at 4°C.Each sample was kept on ice and used in experiments the same day.

| Atomic force microscopy (AFM)
To confirm the oligomeric state of the peptide, the AFM micrograph was taken immediately after dissolving Aβ42.The samples were tested with AFM (Dimension Edge) at Puchuan Company, Guangzhou.The mica surface was examined on 4 different regions to ensure similar structures were present.The images presented are subtracted from the top view containing height and error channel data (Figure S1A,B).

| Cell culture
The murine glioma cell lines GL261 and the murine microglia cell line BV2 were obtained from the State Key Laboratory of Oncology in South China and used for this study.Mycoplasma infection was found to be absent in all cell lines tested.The GL261 cells and BV2 cells were maintained in DMEM with high glucose (Gibco BRL), 10% fetal bovine serum (FBS, HyClone Inc, Logan, UT), 100 U/mL penicillin and 100 μg/mL streptomycin, in a humidified atmosphere at 37 °C under 5% CO2.Supernatants were collected after glioma cells were cultured in an FBS-free medium for 24 h, and then filtered with a polyethersulfone 0.22 μm membrane, serve as the glioma cell-derived condition medium (GCM) for the usage of the in vitro experiments.

| Establishment of the allogeneic glioma mouse model
Anesthesia was administered with isoflurane during all surgery, and all operations were made to minimize suffering.Mice aged 6-8 weeks received GL261 cells implanted into their right cerebral hemisphere.The implant was performed according to the description here.Incisions on the scalp were made in the midline after mice were anesthetized with isoflurane.The skull was drilled with a burr hole of 0.5 mm diameter at stereotaxic coordinates of the bregma, 2 mm lateral, 1 mm caudal, and 3 mm ventral.A 2 μL glioma cell suspension (1 × 10 5 cells/μL in phosphate buffer solution (PBS)) was administered at a depth of 3 mm for 2 min.OAβ42 (1 μM) treatment was treated in Glioma-bearing mice by inoculating glioma cells.Moreover, the control group was inoculated with glioma and vehicle control (DMSO) with a corresponding concentration.

| Analysis of mouse survival and brain tumor growth
Survival and glioma incidence of the mice were observed over 2 months after inoculating glioma cells and OAβ42.In addition, mice were injected with D-luciferin (Macklin, Shanghai, China; 100 μL intraperitoneally at 15 mg/mL in PBS) to image JX-594-driven luciferase expression post-anesthetization. Data acquisition and analysis were performed using Indigo Image v2.0.5.0 software (Berthold).

In vivo imaging was conducted with the NightOWL LB 983 in vivo
Imaging System (Berthold) at every 7 days post the implantation.

| Immunofluorescence staining
28 days post implantation, the glioma-bearing mice were anesthetized with pentobarbital (50 mg/kg) and perfused with PBS followed by 4% paraformaldehyde (PFA) transcardially.Post-fixed of the removed brain tissues in 4% PFA for 24 h, dehydrated by a followed sucrose at a concentration of 0.15 M, 0.5 M, and 0.8 M at 4°C.Using a Leica CM1950 cryostat, the brains were embedded in an optimal cutting-temperature compound (Sakura).Staining of sections was performed according to our previous methods. 26Tissue sections were incubated with rabbit Ki-67 primary antibody (CST, CA, 1:100), IBA1 primary antibody (Abcam, CA, 1:100), GFAP primary antibody from mouse(DAKO, CA, 1:100), CD3 primary antibody from rat(Biolegend, CA, dilution 1:100), and rabbit CD68 primary antibody (Biolegend, CA, 1:100) overnight at 4°C.Then, tissue sections were incubated with Alexa 488 or Alexa 555 conjugated anti-mouse secondary antibodies, or Alexa 647 conjugated anti-rabbit secondary antibodies for 60 min at room temperature.We examined tissue sections using a Zeiss LSM880 confocal microscope (Zeiss).Table S4 provides the antibodies used for this study.

| The EdU assay
EdU was assayed using BeyoClick™ EdU Cell Proliferation Kit as directed by the manufacturer.First, the cells were exposed to the vehicle control (DMSO, 0.2%) and various concentrations of OAβ42 for a duration of 24 h.Then, each well was treated with EdU (final concentration: 10 M) for 2 h at 37°C, after which the cells were harvested.Using Click Reaction Solution containing Azide 555 to mark EdU, 50 liters were added to each well, and the cells were incubated for 30 minutes at room temperature in the dark.Finally, cell nuclei were counterstained with Hoechst 33342 at room temperature for 10 min.The images were taken with a Leica DM6 microscope (Leica).

| Live cell phagocytosis assay
Fluorescent beads phagocytosis assay was employed to analyze live microglia phagocytosis in co-cultures with glioma cells.OAβ42 in simple F12 medium was thawed on ice and polymerized in a 24-well plate (Corning) at 37 °C.GL261 was labeled with GFP.BV2 was unstained and replicated by staining with CellTracke Red CMTPX (Yeasen).The plates were imaged for 10 h at the basetime of addition with lower concentration (1 μM), higher concentration (5 μM) OAβ42, or vehicle in a motorized ImageXpress microscope (Molecular Devices).Image sequences were combined with the ImageXpress, and quantification was analyzed with ImageJ.Only when the GFP-labeled cell interacts with the unstained cell a phagocytosis issue can be counted.And the phagocytosis index was identified as the reduced percentage of these GFP-labeled cells from the base-time to the ending time.   as the ratio of phagocytosis in microglia.The images were captured by a DMi8 microscope (Leica) and a LSM880 confocal microscope (Zeiss).

| Quantitative PCR (qPCR)
The steps were performed according to our previous methods. 26Briefly, total RNA was retrieved from BV-2 cells grown after treated by OAβ42 or vehicle with the EZ-press RNA Purification Kit (EZBioscience) according to the manufacturer's instructions.Reverse transcription was conducted with an iScript™ cDNA synthesis kit (Bio-Rad, Hercules).qPCR was conducted with a CFX96 PCR detection system (Bio-Rad) using Bio-Rad SYBRGreen® dye.A standard curve was established to calculate starting quantities by the iCycleriQ optical system software (Bio-Rad).The experiments were performed in triplicate.The primer sequences applied in this study were provided in Table S5.

| RNA extraction and sequencing
BV2 microglia cells pre-trained with glioma condition medium and 1 μM OAβ42 or vehicle were harvested.Total RNA was extracted with a RNAmini kit (Qiagen).Enrichment of mRNA, fragmentation, reverse transcription, and library construction were conducted with the Illumina Novaseq 6000, and data analysis were performed by Genergy Biotechnology Co. Ltd.

| Western blot
Cell lysates were presented to SDS-polyacrylamide gel electrophoresis and then transferred to polyvinylidene difluoride membranes.
Then, primary antibodies were prepared to incubate the blots at 4°C overnight.Next, the membrane was washed and incubated within a corresponded secondary antibody for 1 h at room temperature.At last, the protein signals were detected with an enhanced chemiluminescence.The antibodies prepared to detect the proteins of interest were also listed in Table S4.

| Statistical analysis
Before data statistical analyses, the Shapiro-Wilk test was performed for the normality of data distribution, and Levene's test was performed for the assumption of homogeneity of variance.Only when the data reached a normal distribution and a homogeneity of variance would the unpaired Student's two-tailed t-test, one-way ANOVA, or two-way ANOVA be conducted.When the data did not reach a normal distribution and a homogeneity of variance, the nonparametric equivalent, including the Mann-Whitney U-Test and the Kruskal-Wallis test, was conducted for analysis.To characterize the cohort, mean ± SD, median (interquartile range, IQR), and counts (percentages) were used to present normally distributed, non-normally distributed, and categorical variables, respectively.The Coxproportional Hazards Regression was used to study the relationship between the variables and mortality.Only when a variable reaches p ≤ 0.1 in the univariate analysis will it be used in the multivariate analysis.In Cox regression analysis, KM survival curves were plotted for variables significantly associated with patient prognosis (p ≤ 0.05).
All statistical analyses in this study were calculated using the IBM SPSS 23.0 program and GraphPad Prism 8.

| Aβ 42 deposition is the prognosis biomarker in glioma
To determine whether Aβ was associated with the malignancy of glioma, we performed IHC staining in tumor tissues from a cohort of 79 glioma patients, including LGG and HGG diagnosed in pathology, as well as 3 peripheral tumor tissues as the normal control (Table S1).
Aβ with a higher expression level was examined in the LGG group compared to the HGG group (p = 0.0075; Figure 1A,B).To determine whether the Aβ-positive could be an independent risk factor for glioma, univariate and multivariate Cox regression analysis was performed.Aβpositive was significantly related to long-term OS (p = 0.029, p = 0.033; Tables S2 and S3).Furthermore, the OS of the Aβ-positive group has a significantly improved OS compared to the negative group in the glioma and GBM cohort, respectively (p = 0.0025, p = 0.0391; Figure 1C,D).
To further confirm the clinical value of Aβ, we reviewed the production process for Aβ.Aβ was derived from amyloid precursor protein (APP) and cleaved by β-Site APP-cleaving enzyme 1 (BACE1).We enrolled TCGA database to analyze the correlation between APP or BACE1 and the OS of glioma patients.Though the level of the APP gene was not significantly different within the LGG and HGG patients, the high level of the BACE1 gene was positively correlated to the longer survival of glioma patients (p < 0.01; Figure S2A,B).The expression of the APP gene and the BACE1 gene was positively correlated in the glioma (p < 0.001; Figure S2C).These also supported the protective role of Aβ in glioma.
These results suggest that the expression of Aβ is negatively associated with glioma's malignant progression.Further exploration of how Aβ suppresses glioma growth is worthwhile.

| OAβ 42 suppresses the progression of glioma in vivo
To determine the potential role of Aβ in glioma formation, OAβ42 (the dominant immune modulator) was prepared, and glioma cell

| OAβ 42 activates microglia in phagocytosis in the glioma microenvironment
To detect changes in the microenvironment induced by OAβ42 in glioma, IF staining with CD3, GFAP, IBA1, and CD68 was performed (Figure 3A).Both positive expressions of CD3, GFAP, and IBA1 showed no significant differences in glioma treated by OAβ42 or vehicle (Figure 3B-D).However, the expression of CD68 was upregulated in the OAβ42 group significantly (p = 0.003; Figure 3E).These revealed that the glioma microenvironment was relatively maintained in homeostasis even with the treatment of OAβ42.In the meantime, microglia were activated in the treatment of OAβ42.Therefore, the activation of microglia may be the cause of microenvironment homeostasis.

Meanwhile, we observed more phagocytosis phenomena in the
OAβ42 treated group.GFP-labeled GL261 cells were surrounded by and close to the CD68-positive microglia (Figure 4A).And ameboid phenotype of microglia appeared more frequently in the OAβ42 treated group (Figure 4B).The high level of activated microglia in OAβ42 was assured by the ratio of CD68 to IBA1 (p = 0.0049; Figure 4C).
To confirm the role of OAβ42 in the anti-glioma process, we performed direct cell culture and co-culture experiments in vitro.The viability of glioma cells did not increase or decrease by OAβ42 in a low concentration of OAβ42 addition for 12 h (Figure S3A,B), consistent with the previous study by Paris et al. 23 However, within the live cell observation station, under a suitable condition for cell incubation for 72 h, microglia and glioma cells were co-cultured with vehicle or OAβ42 for 10 h (Figure 5A).Glioma cell growth was inhibited in OAβ42 exposure compared to vehicle exposure (Figure 5B).And the significant difference in glioma decrease emerged in the low or high concentrate OAβ42 exposed groups at the 7th (n.s., p = 0.0219) and 8th (p = 0.0384, p = 0.048) hour, respectively.Phagocytosis issues, including contraction and overlap between cells, increased in the group of OAβ42 addition.The increasing phagocytosis issues were induced under the low concentrate (p = 0.026) or high (p = 0.0007) concentrate OAβ42 exposure (Figure 5C).The live glioma cell labeled with GFP engulfed by CellTracke Red CMTPX stained microglia was captured in the OAβ42 group (Videos S1 and S2).
To study whether the ability to phagocytosis in microglia was enhanced by the single OAβ42, a beads-based phagocytosis assay was performed.Beads can be engulfed by microglia with OAβ42 even at a low concentration (Figure 5D).The phagocytosis index showed a significant increase in the OAβ42 of low (p = 0.0038) or high (p = 0.0015) concentration.Confocal microscopy was used to confirm phagocytosis and adherence to the surface (Figure 5F).The viability assay also ensured no toxicity of OAβ42 to the microglia (Figure S3C,D).

These results indicate that the phagocytosis function in microglia
was enhanced by OAβ42 in vitro, which can explain the phagocytosis phenomenon observed in vivo.However, the potential molecular mechanisms remain unrevealed.

| OAβ 42-induced phagocytosis of microglia was mediated by up-regulation of IGF-1
To find out the molecular mechanism of enhanced phagocytosis of microglia stimulated by OAβ42, we compared expression profiles among BV2 + vehicle, BV2+ OAβ42, TAMs+vehicle, and TAMs+ OAβ42.To obtain a biological understanding of these OAβ42phagocytosis relationships, enrichment analysis was conducted on each module significantly associated with GO terms.Upregulated cell surface receptor signaling pathway, external side of the plasma membrane, and cell adhesion molecule binding were the relevant biological process, molecular function, and cellular component, respectively (Figure 6A and Figure S4B).Consequently, molecular pathways associated with phagocytosis and AD were assessed.
To find out the related biological pathways, gene set enrichment analysis (GSEA) was performed.Pathways related to regulated mononuclear cell migration, signal receptor binding, and wound healing were enriched among positive genes (Figure S4C-E), supporting the changes in phagocytosis partially characterize the function of OAβ42.Furthermore, the AD, phagosome, and endocytosis pathways were enriched among positive gene loadings and glioma condition medium pretreatment and OAβ42 (Figure 6B).Principal component analysis (PCA) revealed that IGF-1 was the common top gene enriched in the OAβ42 stimulated microglia and the glioma condition medium pretreated microglia (Figure 6C and Figure S4A).
Together, these data provide differences in global expression and specific pathways in all samples, indicating that IGF-1 was molecularly expressed in the phagocytosis state of OAβ42 treated microglia.
To ensure the role of IGF-1 in phagocytosis, we obtained the gene expression of IGF-1 by qPCR (Figure 6D).A significant upregulation of IGF-1 was detected (p = 0.0039).To make sure whether the IGF-1 related receptor can be activated in the OAβ42-treated microglia F I G U R E 6 IGF1 is essential for the OAβ42 induced microglia phagocytosis.(A) The Gene Ontology (GO) and functional pathway analysis of coding genes associated with positive correlation coefficients in TAMs treated with OAβ42, p-value ≤0.05.The x-axis is the -log 10 p value; (B) GSEA of TAMs concerning genes upregulated in the exposure of OAβ42.The analysis demonstrates a positive correlation (q < 0.1, Enrichment Score = 0.15) between many genes upregulated in phagocytosis and AD; (C) PCA of the transcriptome of glioma condition medium pre-treated BV2 (TAMs) treated with OAβ42 compared to vehicle.Volcano plot shows gene expression differences between microglia treated with OAβ42 and vehicle.Log 2 gene ratios are plotted against negative log 10 p values; (D) qPCR reveals IGF-1 upregulated in OAβ42 stimulated microglia, the non-parametric equivalent with Mann-Whitney U-Test was performed to data analysis; (E) Western blot of IGF-1 signaling related proteins of microglia treated with vehicle, OAβ42, and OAβ42+ IGFBP3

| DISCUSS ION
The immune microenvironment plays a critical role in the pathological development of AD.Microglia are the most dominant myeloid-derived immune cells in the brain.Microglia activation has been identified as a prominent feature of the AD microenvironment and could drive AD's pathological process.In the tumor ecosystem, tumor cells were surrounded by various immune cells that could act in pro-and anti-tumorigenic functions. 27Increasing evidence showed that higher densities of tumor-associated microglia (TAMs) were found in HGG compared with LGG, suggesting TAMs serve as a villain in the environment. 28,29TAMs can be divided into brain-resident macrophages (microglia) and bone marrowderived macrophages (BMDMs) based on developmental origin. 30[33] Moreover, microglia are also resistant to myeloablative irradiation with genetic stability. 34,35Only in the cases of blood-brain barrier (BBB) disruption, such as glioma progression, could BMDMs be recruited into the brain and become TAMs. 36The primitive functions of microglia are maintaining the homeostasis of the brain through phagocytosis and the scavenging of abnormal cells or debris.In the current study, we attempted to reeducate the TAMs to normal status with the restored engulfing function to delay the glioma progression.We found that OAβ42, the immune stimulator of early AD, can enhance microglia phagocytosis of glioma cells. 37ese results indicated that the intrinsic function of microglia was masked in the tumor environment and can be resettled by niche modulators such as OAβ42.And the results were also consistent with the observations that lower glioma incidence found in the AD population.
Sporadic attempts targeting microglia to treat glioma were reported.Colony stimulating factor 1-receptor (CSF1-R) is predominantly expressed by microglia and necessary for its survival. 38Blocking CSF1-R in glioma-bearing mice led to a reduced infiltration of GAMs and decreased tumor volume significantly. 39wever, the CSF1-R inhibitors failed in clinical trials. 40The fundamental cause of the failure lies in the mechanism that the CSF1-R inhibitor modified the pro-inflammation of microglia but failed to re-educate the microglia into a normal state with healthy functions.Another research found the CD47/SIRPα axis as a "do not eat me signal" of glioma and blocking CD47 effectively restraining the glioma progression. 41,42But severe hemolytic reaction caused by CD47 monoclonal antibody was noted in clinical trials. 43Though the anti-CD47 therapy did not concern the re-education of microglia and failed in the pre-clinical stage, the enhanced phagocytosis was proved an effective target in glioma treatment.While both CSF1-R and CD47/SIRPα are vital for microglia to survive in the physical status, the effectiveness of targeting these targets was limited in a pathological status.Our results provide a new angle in drug development against glioma, based on the pathological character of the target cells and exert its primitive functions.Therefore, re-educating microglia based on their plasticity of phagocytosis, rather than clearance themselves, can serve as a promising strategy in glioma therapy.
Phagocytosis can be modulated by cytokines, adhesion to the matrix, as well as by host colony-stimulating factors and microbial stimuli. 44Limited evidence is known about the role of growth factors in the phagocytosis of innate immune cells.IGF-1, as a growth hormone, is conventionally known to stimulate cancer cell proliferation and tumor growth. 45,46Indeed, IGF-1 possesses pleiotropic properties that directly stimulate monocyte and macrophage to regulate inflammatory responses via enhanced TNF alpha production and chemotactic migration. 47,48Recent studies reported that IGF-1 could regulate the phagocytosis of astrocytes and epithelial cells. 49,50Here, our study demonstrated that IGF-1 is the essential factor in the phagocytosis of microglia enhanced by OAβ42, helping to inhibit glioma growth.
Two possibilities may explain the diverse function of IGF-1: 1.The plasticity or status of macrophages stimulated by OAβ42 in the brain differs from that caused by inflammatory factors of bronchial injury in the lung; 2. Though the lung-resident macrophage and the brain-resident microglia originated from the yolk sac, they live within a unique organ environment.These two resident engulfing cells acquired tissue-specific adaptability within the terminated differentiation organ and reacted in distinct cellular reactions to scavenge abnormal cells or debris.In cellular endocytosis, IGF-1 signaling is essential for mitochondrial biogenesis, mitophagy, and the Golgi apparatus. 51And the clearance of Aβ can be efficient for microglia in a sustained condition.Moreover, lower serum IGF-1 level was observed to increase the risk of AD incidents. 52,53These all indicated that IGF-1 could benefit glioma phagocytosis by microglia.Besides the benefits of microglia phagocytosis, the pro-function of glioma cells of IGF-1 should also be considered.As the glioma cells were not proliferated in vivo and in vitro co-culture systems, the concentration of IGF-1 secreted by microglia might be low enough and affect microglia itself locally.
Other than IGF-1, another concern in our study may lie in the side effect of OAβ42.Although the consensus for the causative role of Aβ is uncertain, as we mentioned in the background, Aβ is a representative biomarker for AD.Indeed, the amyloid hypothesis is one of the speculations for the etiology of AD. 54 The Aβ targeted peptide could not inhibit AD progression. 55The concentration of Aβ in cerebrospinal fluid was changed 10 years before the occurrence of AD. 56 Aβ is more like a systemic metabolic peptide with no toxicity in a proper phenotype and limited concentration.Oligomer and 42 BV2 cells were seeded in 24-well chambers (Corning) and treated with vehicle, lower concentration (1 μM), higher concentration (5 μM) OAβ42 or IGFBP3 + 1 μM OAβ42 for 12 h.Cells were captured after cultured with 1-μm-diameter latex beads conjugated to a red fluorescent protein (RFP) (Sigma) with a 10-20 beads per cell ratio.Alternatively, Green Actin Tracking Stain (Invitrogen) pretreated cells were filmed on coverslips, incubated with beads for 1 h, washed with PBS for three times, and treated with DMEM medium for an extra hour.Finally, cells in the chambers or on the coverslips were collected for immunofluorescence analysis.DNA was stained with DAPI (Life Technologies). 2 or more beads overlapping in the microglia localization can be determined as a phagocytosis issue.The phagocytosis index was identified F I G U R E 1 The clinical significance of Aβ in glioma patients.(A) Human glioma tissues were collected for IHC staining of Aβ. White bars represent 100 μm, and black bar represent 20 μm.Each grade of glioma tissues was repeated fourth, and the results were obtained and analyzed by experienced pathologists; (B) Quantification of DAB signal intensity was performed and represented in graphs, the nonparametric equivalent with the Kruskal-Wallis test was used; (C) Survival analysis of Aβ in 79 glioma patients' cohort, the Kaplan-Meier survival analysis with Log Rank test was applied.(D) Survival analysis of Aβ in 62 GBM patients' cohort, the Kaplan-Meier survival analysis with Log Rank test was conducted.Columns indicate median [95% CI]; n.s., not significant; p ≥ 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

F I G U R E 2
OAβ42 suppresses glioma growth in vivo.(A) schematic diagram of OAβ42 treatment in GL261-inoculated C57 allogeneic model; (B) Representative (In vivo Imaging System) IVIS images of glioma-inoculated mice treated with vehicle or OAβ42 in early (2 weeks) and late stage (4 weeks); (C) Quantification of tumor volumes in two groups of mice recorded in IVIS, data was analyzed by the nonparametric equivalent with Mann-Whitney U-Test; (D) Survival analysis of glioma-inoculated mice along with vehicle or OAβ42 treatment, the Kaplan-Meier survival analysis with Breslow test was applied; (E, F) Representative images of Ki-67 staining and quantification of Ki-67 staining.White bars represent 100 μm, the Student's t-test was conducted for the data analysis.Columns indicate median [95% CI]; n.s., not significant; p ≥ 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

F I G U R E 3
OAβ42 active microglia and re-modulate the tumor microenvironment.(A) Representative images of the immune microenvironment of Glioma-bearing mice treated with vehicle or OAβ42.White bars represent 100 μm; (B-E) Quantification of the immune cells in the two groups, the Student's t-test was conducted for the data analysis.CD3 represented T cells (B), GFAP represented astrocytes (C), IBA1 represented microglia (D), and CD68 represented activated microglia (E).Columns indicate median [95% CI]; n.s., not significant; p ≥ 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.| 7 of 14 LU et al.

F I G U R E 4
OAβ42 potentiates phagocytosis of microglia.(A) Representative confocal images of microglia that engulf glioma cells after OAβ42 stimulation, White bars represent 20 μm; (B) Representative images of microglia morphology with vehicle treatment or OAβ42, White bars represent 100 μm; (C) Quantification of activation ratio by the ratio of CD68 to IBA1 positive microglia, the Student's t-test was conducted for the data analysis.Columns indicate median [95% CI]; n.s., not significant; p ≥ 0.05; *p < 0.05; **p < 0.01; ***p < 0.001.

F I G U R E 5 | 11 of 14 LU
Figure 2D).Ki-67 was stained in the two groups to assess the glioma's viability.The expression of Ki-67 was decreased in the OAβ42 group (p = 0.0421; Figure2E,F).To further investigate the underlying mechanism of glioma inhibition induced by OAβ42, microenvironment cells in the brain were considered.
Figure6F,G).This indicates that the IGF-1 secreted by microglia showed an enhanced phagocytosis function for the microglia itself indeed.