OLFML2A Downregulation Inhibits Glioma Proliferation Through Suppression of Wnt/β-Catenin Signaling

Glioma is a highly heterogeneous and lethal tumor with an extremely poor prognosis. Through analysis of TCGA data, we identified that OLFML2A is a key promotor of gliomagenesis. However, the molecular function of OLFML2A and its underlying mechanism of action in glioma remain unclear. In this study, we found that OLFML2A expression was significantly upregulated in glioma specimens and positively correlated with pathological grades in glioma patients. Moreover, Kaplan–Meier survival analysis of TCGA data revealed that glioma patients with higher OLFML2A expression had shorter overall survival. Importantly, OLFML2A knockdown in glioma cells inhibited cell proliferation and promoted apoptosis. Mechanistically, OLFML2A downregulation inhibits Wnt/β-catenin signaling by upregulating amyloid precursor protein (APP) expression and reducing stabilized β-catenin levels, leading to the repression of MYC, CD44, and CSKN2A2 expression. Furthermore, OLFML2A downregulation suppressed the growth of transplanted glioma subcutaneously and intracranially by inhibiting Wnt/β-catenin pathway-dependent cell proliferation. By uncovering the oncogenic effects in human and rodent gliomas, our data support OLFML2A as a potential therapeutic target for glioma.


INTRODUCTION
Glioma is characterized by extremely poor outcomes (1). In particular, glioblastoma (GBM), a World Health Organization (WHO) grade IV tumor, is the most common and aggressive malignant brain tumor in humans, and even with radical surgery, radiation therapy, and chemotherapy, the median survival time is less than 12 months (2,3). Studies have partly characterized the genetic basis of tumorigenesis in glioma over the past two decades (4,5). Based on these findings, the molecular classification standard of central nervous system tumors was established by the WHO in 2016 (6). However, glioma is a highly heterogeneous tumor (7), and there are many specific molecules that can be used for its classification. Thus, identifying the molecules that regulate glioma progression may contribute to the development of reliable therapeutic targets for glioma.
Olfactomedin-like 2A (OLFML2A) belongs to subfamily IV of olfactomedin domain-containing (OLFM) proteins (8), also known as photomedins-1, which were first identified and characterized in the mouse retina (9). In normal human tissue, OLFML2A is predominantly detected in the photoreceptor layer of the retina but is not expressed in neuronal tissues (10). In general, the function of OLFML2A remains poorly understood at present. Recent studies have shown that OLFML2A is highly expressed in triple-negative breast cancer (TNBC) and positively related to cell growth, proliferation, and migration (11). In acute myeloid leukemia, high OLFML2A expression is associated with extramedullary infiltration and predicts a poor prognosis (12). Consistent with these findings, Dai et al. showed that OLFML2A is also overexpressed in liver hepatocellular carcinoma (LIHC) and many other cancers, such as breast cancer and leukemia, by comparing the Oncomine dataset to normal samples (13). Ultimately, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway and process enrichment analysis suggested that OLFML2A regulates the tumorigenesis of LIHC through angiogenesis, negative regulation of DNA-binding transcription factor activity, regulation of neuronal differentiation, positive regulation of apoptosis, and autophagy via the Wnt, Notch, and hypoxia signaling pathways (13). The above observations underscore that OLFML2A may be involved in the development of many cancers. However, the detailed regulatory mechanisms have not yet been thoroughly validated, and the underlying molecular mechanisms need to be elucidated. Moreover, the role of OLFML2A in glioma has not been reported.
Therefore, to better understand the function of OLFML2A and identify new markers for targeted glioma therapy, in vitro and in vivo assays and bioinformatics analyses were performed to elucidate the function and mechanisms of OLFML2A in glioma progression.

Cell Culture
Human glioblastoma cell lines (U87MG and U251) were purchased from GeneChem (Shanghai, China) and cultured in DMEM (HyClone, USA) supplemented with 10% fetal bovine serum (Sigma, USA) and 1% penicillin and streptomycin (Beyotime). All cells were cultured in a humidified incubator at 37°C under an atmosphere with 5% CO 2 .

Lentiviral Transduction
Cells were seeded in 6-well plates at a density of 5×10 4 cells/well and incubated at 37°C until reaching 30% confluence. According to the determined multiplicity of infection, an appropriate amount of lentivirus was added to the culture medium of glioma cells for transduction, and the cells were cultured for another 14 h. Subsequently, the medium containing the lentivirus was removed, and the cells were cultured in normal medium for 58 h. GFP expression was observed under a fluorescence microscope 3 days after infection, and glioma cells with an infection efficiency of >80% were selected for subsequent analysis.
All lentiviruses were purchased from GeneChem (Shanghai, China). The forward sequences of the shRNAs targeting OLFML2A in U251 and U87MG cells were as follows: sh-OLFML2A-1 (TCTATGTCACCAACTACTA) and sh-OLFML2A-2 (GCCAAACAAACATTCACTA). The forward sequence of the shRNA targeting OLFML2A in C6 cells was as follows: AGGCCGGTGGAGTAATATGTA. To increase the efficiency of RNA interference, four shRNAs targeting amyloid precursor protein (APP) were designed, and the four plasmids carrying the different shRNAs were mixed in equal proportion for lentiviral packaging. The forward sequences of the shRNAs targeting APP in U87MG cells were as follows: psc10391 (ccCTGTTCATTGTAAGCACTT), psc10392 (gcAGACACA GACTATGCAGAT), psc11434 (ccCAAAGTTTACTCA AGACTA), and psc11435 (gcCATCTTTGACCGAAACGAA).

Cell Proliferation Analysis by Celigo Assay
Cell proliferation was analyzed using a Celigo assay as previously described (14). Briefly, cells were cultured in 96-well plates at a density of 1,000 cells/well and treated with the indicated shRNAs or control shRNA (shCtrl). The medium was changed every other day, and the cells were imaged with a Celigo Imaging Cytometer Abbreviations: OLFML2A, olfactomedin-like 2A; APP, amyloid precursor protein; TCGA, The Cancer Genome Atlas; LGG, low-grade glioma; TNBC, triple-negative breast cancer; LIHC, liver hepatocellular carcinoma; IHC, immunohistochemistry; DEGs, differentially expressed genes; IPA, ingenuity pathway analysis. for 5 consecutive days. For each experimental well of a 96-well plate, Celigo scans of four visual fields at each time point (40 × magnification) were performed, and the images were analyzed with the corresponding software. Each assay was conducted in triplicate. After 5 days of consecutive measurements, a cell growth curve was plotted to assess cell proliferation.

Quantitation of Apoptosis by Flow Cytometry
An Annexin-V-APC apoptosis detection kit (eBioscience, Cat. No. 88-8007) was used to evaluate cell apoptosis according to the manufacturer's instructions. First, glioma cells were infected with a lentivirus expressing OLFML2A or a scrambled shRNA sequence (Scr-shRNA). Subsequently, after culturing for another 4 days, the cells were harvested, washed with PBS, and then resuspended in staining buffer at a final density of 1×10 6 -1×10 7 /ml. Then, 5 ml of Annexin-V-APC was added to 100-ml aliquots of cell suspensions and incubated at room temperature for 10-15 min. Signals were detected with a FACSCalibur instrument (Becton-Dickinson, USA).

Animal Studies
All animals were purchased from the Animal Experiment Center of Gansu College of Traditional Chinese Medicine (Lanzhou, China) and housed in a standardized specific pathogen-free animal facility. All experimental procedures were approved by the Institutional Animal Care and Use Committee of Lanzhou University Second Hospital. All mice were maintained and were used in accordance with the guidelines approved by the Institutional Animal Care and Use Committee.
For the tumor xenograft implantation model, 4-week-old mice were randomly divided into two groups (n=10 per group) and implanted with 1×10 6 U87MG cells that were previously transfected with OLFML2A shRNA lentivirus or the negative control. Then, the cells were mixed with Matrigel (50% volume) and subcutaneously implanted into the right flanks of the nude mice (BALB/c). Tumor volume was determined using an external caliper every 3-4 days and calculated using the formula (Length×Width 2 )/2. Mice were sacrificed 50 days after transplantation, at which time tumors were excised and used in subsequent analyses.
For the orthotopic model, 1×10 6 C6-OLFML2A-shRNA or control cells were injected into the right corpus striatum of the brains of 6-week-old Wistar rats using a stereotactic frame (n=5 per group). Spectral computed tomography (HD750 CT scanner; GE Healthcare, Little Chalfont, UK) was performed to record tumor growth at 14 and 19 days after injection using the same scanning parameters described in a previous study (15). Rats were monitored and sacrificed when neurological signs appeared or after 35 days.

Gene Chip Microarray Analysis
Total RNA was extracted using an RNAiso Plus kit (TaKaRa Biotechnology Co., Ltd., Dalian, China) and assessed using a NanoDrop 2000 Spectrophotometer (Thermo Scientific, MA, USA) and an Agilent Bioanalyzer 2100 (Palo Alto, CA, USA) (16). Only samples of sufficient quality were used in subsequent gene chip experiments. The RNA quality control standards were as follows: 1.7<A260/A280<2.2 on the NanoDrop 2000 and an RNA integrity number (RIN)> =7.0 and a 28S/18S ratio > 0.7 on the Agilent 2100 Bioanalyzer. The microarray was processed as described in a previous study (17). Briefly, amplified RNA (aRNA) was prepared from total RNA using a GeneChip 3' IVT Express kit (Affymetrix, Santa Clara, CA, USA). Then, DNA was synthesized, after which double-stranded template DNA was synthesized. Biotin-labeled aRNA was obtained by in vitro transcription. Then, aRNA was purified, fragmented, labeled, and hybridized with an Affymetrix GeneChip ® Human Transcriptome Array 2.0 (Affymetrix, Santa Clara, CA, USA). After hybridization, the chips were washed and stained. Finally, raw intensity data were analyzed using Affymetrix Expression Console software after the images and raw data were scanned. Genes that were up-or downregulated with fold changes of > 2.0 or < 2.0 were analyzed using Affymetrix Transcriptome Analysis Console Software.

Ingenuity Pathway Analysis
Ingenuity pathway analysis (IPA) of differentially expressed genes (DEGs) was conducted with Qiagen's Ingenuity Pathway Analysis algorithm (www.qiagen.com/ingenuity, Qiagen, Redwood City, CA, USA). Canonical pathway analysis, functional analysis, regulatory effect analysis, and interaction network analysis were performed, with activation z-scores and overlapping p-values calculated as described in a previous study (18).

Quantitative Reverse Transcription PCR
Cells in the exponential growth phase were collected and lysed with an RNAiso Plus kit (TaKaRa Biotechnology Co., Ltd., Dalian, China) to extract total RNA. A NanoDrop-2000 spectrophotometer (Thermo Scientific, MA, USA) was used to measure the purity and concentration of RNA. According to the manufacturer's protocol, total RNA was reverse-transcribed using the Prime Script RT kit (TaKaRa Biotechnology Co., Ltd.) for 15 min at 37°C before being heat inactivated for 20 s at 85°C. Subsequently, quantitative reverse transcription PCR (qRT-PCR) was performed using SYBR ® Premix Ex Ta Master Mix (TaKaRa Biotechnology Co., Ltd.) and a Bio-Rad CFX96 Real-time PCR system (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The primers used for qRT-PCR are shown in Table 1. qRT-PCR was performed in a final volume of 10 ml with the following thermocycling program: an initial denaturation at 95°C for 10 min followed by 40 cycles of 95°C for 10 s, 60°C for 30 s, and 72°C for 30 s. As the reaction progressed, fluorescence signals were obtained. After amplification, a melting curve analysis was performed, the results of which were used to determine the dissociation characteristics of the PCR products. The 2 -DDCq method was used to calculate the mRNA expression level relative to that of the internal reference gene GAPDH. Each sample was assayed in triplicate.

Statistics and Data Analysis
All data are presented as means ± standard deviation (SD). Statistical analyses were performed using Prism 8 (GraphPad Software, San Diego, CA, USA). Two-tailed Student's t-test was used to assess the significance of the differences between two groups, and one-way ANOVA was used to assess the significance of the differences among multiple groups. Survival curves were analyzed using the Kaplan-Meier method and assessed by the log-rank test. P-values<0.05 were considered to indicate a significant difference.

OLFML2A Overexpression Is Correlated With Glioma Progression and Poor Prognosis
By analyzing datasets from The Cancer Genome Atlas (TCGA), we observed that OLFML2A expression was significantly upregulated in glioma tissues compared to normal brain tissue and that OLFML2A expression was positively correlated with glioma grades (Figures 1A, B). Kaplan-Meier survival analysis of TCGA data revealed that glioma patients with higher OLFML2A expression had shorter overall survival ( Figure 1C). Real-time PCR results showed that OLFML2A expression was upregulated in four different glioma cell lines ( Figure 1D). Taken together, these findings suggested that OLFML2A is overexpressed in human glioma.

OLFML2A Expression Is Efficiently Inhibited by Lentiviral-Based shRNA in Human Glioma Cell Lines
A positive correlation between OLFML2A expression and glioma pathological grades suggested that OLFML2A may be involved in the development and progression of glioma. To assess the biological role of OLFML2A in glioma, we used a lentivirus-based shRNA strategy to knockdown OLFML2A expression in glioma cell lines. U251 and U87MG cells were infected with a lentivirus specifically targeting human OLFML2A (OLFML2A-shrNA) or a negative control (Scr-shRNA) to evaluate the knockdown efficiency.
Cells infected with the lentivirus were cultured and harvested, and total RNA/protein samples were collected. Then, the OLFML2A mRNA and protein levels were detected by qRT-PCR and Western blotting. The results demonstrated that OLFML2A mRNA expression was significantly reduced (Figure 2A), and OLFML2A protein was barely detectable in cells treated with OLFML2A-shRNA ( Figures 2B, C). These results indicated that lentivirus-based shRNA strategy could inhibit OLFML2A expression at the protein and mRNA levels.
OLFML2A Knockdown Leads to Reduced Glioma Cell Proliferation In Vitro U251 and U87MG cells express high levels of OLFML2A protein.
To determine whether OLFML2A is required for the proliferation of these cell lines, we generated U251 and U87MG cells with lentivirus-delivered OLFML2A shRNA knockdown as previously described. Then, Celigo assays were performed to monitor cell growth for 5 days. As shown in Figures 3A-F, silencing OLFML2A decreased the total cell numbers and slowed the growth rate of U251 and U87MG cells. Subsequently, the MTT assay was used to further evaluate the effect of OLFML2A knockdown on the proliferation of U251 and U87MG cells. As indicated in Figures 3G-J, the cell proliferation rate of the OLFML2A-shRNA group was markedly reduced compared to that of the control group.

Suppression of OLFML2A Expression Leads to Increased Glioma Cell Apoptosis
The ability to resist cell death, another hallmark of cancers, is essential for tumor development (19). The reduction in cell numbers induced by OLFML2A knockdown may be attributable to cell proliferation inhibition and increased cell apoptosis. To further investigate whether OLFML2A promotes glioma cell proliferation by regulating cell apoptosis, we used Annexin-V staining to assess apoptosis in OLFML2Aknockdown and control U251 cells. OLFML2A knockdown significantly induced apoptosis in U251 cells ( Figures 4A, B), and similar results were observed in U87MG cells following OLFML2A silencing by shRNA ( Figures 4C, D).

Disruption of Multiple Crucial Cell Cycle Regulation Pathways Involved in Cancer Development by OLFML2A Knockdown
The results described above showed that OLFML2A is crucial for tumor development in human glioma cell lines. However, the mechanisms underlying OLFML2A-mediated glioma development and its downstream pathways have not been systematically explored. Therefore, global gene expression profiling of U87MG cells infected with lentiviruses expressing either Scr-shRNA or OLFML2A-shRNA was performed using a microarray platform, and 1,911 genes showing significant differential expression were identified (based on |fold change|≥2.0 and FDR<0.05), of which 658 were upregulated and 1,253 were downregulated ( Figures 5A, B and Supplementary Table 1, Supplementary Data Sheet 1). Then, functional analysis was performed via IPA, and the DEGs were observed to be enriched in cancer, cell cycle, cell death and cell survival, cellular growth, and proliferation, suggesting that OLFML2A is significantly (based on a P<0.001 threshold) involved in proliferation and cell cycle regulation processes ( Figure 5C). Canonical pathway analysis using IPA was conducted to further evaluate potential signaling pathways, resulting in 136 pathways being identified (Supplementary Table 2 cycle regulation, were also inhibited ( Figure 4D). Furthermore, Apoptosis Signaling (Z-score=1.508), Cell Cycle: G1/S Checkpoint Regulation (Z-score=1.291), and Cell Cycle: G2/M DNA Damage Checkpoint Regulation (Z-score=0.943) were activated ( Figure 5D). Collectively, these findings provided a strong rationale for the role of OLFML2A as a crucial molecule that regulates the proliferation of glioma cells.

OLFML2A Knockdown Suppresses the Wnt/b-Catenin Pathway in Glioma Cells
The results of previous studies have shown that the Wnt/bcatenin pathway is activated in glioma and is essential for glioma growth, proliferation, cell cycle regulation, and apoptosis (20). Importantly, the IPA data showed that Wnt/b-catenin signaling was one of the inhibited pathways enriched in the OLFML2Aassociated gene signatures ( Figure 5D). The expression of 21 molecules involved in the Wnt/b-catenin pathway was altered ( Figures 6A, B). Subsequently, we validated the expression levels of the 21 molecules by real-time PCR and observed that the expression levels of Wnt ligands (Wnt3, Wnt5a, and Wnt5b), the transcription factor LEF1, and genes downstream of Wnt (MYC, CD44, and CSKN2A2) were decreased when OLFML2A was downregulated ( Figure 6C). The changes in the remaining molecules were generally consistent with the gene chip results (G-J) U251 and U87MG cell growth was measured by the MTT assay for 5 days. The data are presented as mean ± SD. *P < 0.05, **P < 0.01, ***P < 0.001.
( Figure 6C). These data preliminarily confirmed that Wnt activity is suppressed upon OLFML2A downregulation. Furthermore, we observed that the mRNA levels of the Wnt target genes MYC, CD44, and CSKN2A2 were significantly and positively correlated with the OLFML2A mRNA levels in GBM patients by analyzing TCGA data ( Figure 6D). Accumulating evidence has demonstrated that the regulation of b-catenin and GSK-3b phosphorylation controls Wnt/bcatenin pathway activation (20). Therefore, we subsequently assessed the phosphorylation of b-catenin and GSK-3b in U87MG cells. The results showed that the Thr41/Ser37/Ser33 phosphorylation levels of b-catenin were significantly increased upon OLFML2A downregulation ( Figure 6E). Conversely, stabilized b-catenin levels were decreased when OLFML2A was downregulated ( Figure 6E). Additionally, we observed that phosphorylated GSK-3b levels were significantly decreased while total GSK-3b levels were elevated upon OLFML2A downregulation ( Figure 6E). Thus, our results indicated that high levels of OLFML2A expression may activate the Wnt/bcatenin pathway to promote cell proliferation and inhibit cell apoptosis.

APP Acts as an Intermediate Molecule to Mediate the Downstream Regulatory Effects of OLFML2A in Glioma
To further elucidate the regulatory mechanism of OLFML2A on the Wnt/b-catenin pathway and cell proliferation, an IPA-based interaction network analysis was performed, the results of which confirmed that APP is an important intermediate molecule that mediates the downstream regulatory effects of OLFML2A ( Figure 7A). APP has been reported to be a negative regulator of Wnt/b-catenin signaling pathways (21). Our results showed that OLFML2A knockdown increased APP expression (Figures 7B, C). Furthermore, we observed that APP expression was significantly downregulated in GBM tissues compared to normal brain tissue by analyzing TCGA data ( Figure 7D). Additionally, we demonstrated that APP downregulation partly increases the proliferation of OLFML2A-knockdown U87MG cells (Figures 7E, F). These data further suggested that APP is a functional regulator that mediates the downstream regulatory effects of OLFML2A through the Wnt/ b-catenin pathway in glioma.

OLFML2A Downregulation Leads to the Repression of Glioma Cell Proliferation In Vivo
We used a xenograft model to verify whether OLFML2A knockdown reduces tumor growth in vivo. U87MG cells infected with Scr-shRNA or OLFML2A-shRNA were transplanted into the right flanks of immunocompromised nude mice, and the results showed that OLFML2A knockdown led to significant decreases in tumor volume and weight ( Figures 8A-C) (n = 10). Moreover, the IHC results showed that OLFML2A knockdown could significantly reduce the positivity rate of Ki-67 in subcutaneous tissues ( Figure 8D). In addition, the level of APP expression was consistent with the in vitro results ( Figure 8E). Furthermore, the IHC results also indicated that b-catenin phosphorylation levels were significantly elevated while stabilized b-catenin levels were markedly decreased following OLFML2A downregulation ( Figure 8E). These results confirmed that OLFML2A functions as an oncogene in glioma by regulating the Wnt/b-catenin pathway via APP in vivo.
To further verify the effects of OLFML2A on glioma growth in immunocompetent animals, we transplanted C6 cells transfected with Scr-shRNA or OLFML2A-shRNA into the right striatum of Wistar rats (n = 5) and observed that OLFML2A downregulation strongly suppressed the growth of intracranial glioma by CT scan on days 14 and 19 after transplantation ( Figure 9A). Similarly, gross observations also showed that OLFML2A downregulation could significantly suppress orthotopic glioma growth ( Figure 9B), and the positive rate of Ki-67 was also decreased after OLFML2A knockdown ( Figure 9C).

DISCUSSION
Glioma is a very aggressive and heterogeneous tumor that lacks effective targeted therapies and has a relatively poor prognosis (22).
Therefore, there is an unmet need to identify new targets to develop potential diagnostic and therapeutic strategies. In the present study, we first analyzed an available genomic database (TCGA) and identified OLFML2A as a putative molecule involved in glioma tumorigenesis. Our results indicated that OLFML2A is upregulated in glioma, positively correlated with tumor grade, and negatively correlated with the prognosis of glioma patients. In addition, we observed that the level of OLFML2A is correlated with glioma Our results suggested that OLFML2A promotes the progression of glioblastoma and may be a potential novel prognostic factor for glioblastoma. Importantly, OLFML2A may be a promising target gene for glioma treatment.
OLFML2A is a member of the OLF family (8). Interestingly, data obtained over the past several years demonstrated that OLF family proteins play important roles in neurogenesis, neural crest formation, dorsal ventral patterning, cell-cell adhesion, cell cycle regulation, and tumorigenesis and may act as modulators of critical signaling pathways (Wnt, BMP) (8). The human OLFML2A gene is located on chromosome 9q33.3 and encodes a secreted glycoprotein known as photomedin-1 (10). An understanding of the biological functions of proteins possessing the OLF domain has remained elusive (8,10,23). In the field of cancer research, the tumorigenic function of OLFM4 has been verified in many tumors (24,25). Intriguingly, the tumorigenic function of OLFML2A is also gradually emerging. OLFML2A expression was shown to be negatively correlated with the biological progression and clinical features of TNBC, LIHC, and leukemia (11)(12)(13). Moreover, as previously reported, the Wnt/b-catenin pathway is a potential regulatory target of OLFML2A, through which OLFML2A regulates several malignant biological processes in LIHC (13). Consistent with previous studies, our results indicate that OLFML2A acts as an oncogene that exerts an important effect on glioma progression. Moreover, an IPA-based pathway enrichment analysis identified the Wnt/b-catenin pathway as being altered after OLFML2A downregulation.
Extensive studies have shown that abnormal activation of the Wnt signaling pathway leads to the malignant biological process of glioma, including cell proliferation, cell apoptosis, and cell invasion (26). In the classic Wnt/b-catenin pathway, Wnt ligands combine with the receptor protein frizzled (Fzd) and low-density lipoprotein receptor related protein 5/6 (LRP5/6) to form a protein complex, which then activates Dishevelled (Dvl), leading to the phosphorylation and inactivation of glycogen synthase kinase 3b (GSK-3b) and preventing stabilized b-catenin from being phosphorylated and degraded (26,27). Stable b-catenin in the cytoplasm further enters the nucleus and binds to the T-cytokine/ lymph enhancer binding factor (TCF/LEF) transcription factor, resulting in the transcription of Wnt target genes such as MYC, CD44, cyclin D1, and PPAR-d (26,27). In our present study, we demonstrated that the expression levels of Wnt ligands (Wnt3, Wnt5a, and Wnt5b), the transcription factor LEF1, and Wntdownstream genes (MYC and CD44) were decreased when OLFML2A was downregulated. Furthermore, total GSK-3b levels were significantly elevated, and GSK-3b phosphorylation was decreased upon OLFML2A downregulation. Additionally, the levels of stabilized b-catenin were decreased while the Thr41/ Ser37/Ser33 phosphorylation levels of b-catenin were increased when OLFML2A was downregulated. Overall, the results of our present study indicate that OLFML2A downregulation can inactivate the Wnt/b-catenin pathway to inhibit cell proliferation and promote cell apoptosis.
The results of an IPA-based interaction network analysis confirmed that APP is an important intermediate molecule that mediates the downstream regulatory effects of OLFML2A in glioma. Accumulating evidence has shown that abnormal cleavage of APP is a central process in the pathological mechanisms of Alzheimer's disease (AD) (27)(28)(29), and the Wnt/ b-catenin pathway is dysfunctional in AD brains (30)(31)(32). Furthermore, APP has been reported to downregulate b-catenin expression by increasing its degradation (32,33), and overexpressed APP prevents b-catenin translocation into the nucleus through physical binding and precludes the transcription of Wnt target genes in AD disease models (21). These findings indicated a reciprocal regulation between APP and the Wnt/b-catenin signaling pathway. The results of the interaction network analysis were consistent with the function of APP acting as a suppressor of the Wnt/b-catenin pathway. In our present study, we observed that APP was expressed at low levels in glioma tissue. APP expression was observed to be negatively correlated with OLFML2A, and OLFML2A knockdown led to increased APP levels both in vitro and in vivo. Moreover, APP downregulation partially attenuated inhibitory effect of OLFML2A knockdown on cell proliferation. Therefore, we concluded that OLFML2A downregulation inhibits Wnt/b-catenin signaling by upregulating APP expression and reducing stabilized b-catenin levels.

CONCLUSION
In summary, our results suggest that OLFML2A plays a crucial role in the proliferation of glioma and acts as an oncogene to promote the progression of glioma in humans. Mechanistically, OLFML2A knockdown inhibits the Wnt/b-catenin signaling pathway by upregulating APP expression and promoting b-catenin phosphorylation, leading to repressed MYC, CD44, and CSKN2A2 expression. Our data indicate that the oncogenic effect of OLFML2A in glioma occurs through regulation of Wnt/b-catenin signaling, which may provide a new potential therapeutic target for glioma.

DATA AVAILABILITY STATEMENT
The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/Supplementary Material.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the ethics committee at the Lanzhou University Second Hospital (No. 2019A-077). The patients/participants provided their written informed consent to participate in this study. The animal study was reviewed and approved by the ethics committee at the Lanzhou University Second Hospital (D2019-117). Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.