Spermine Synthase : A Potential Prognostic Marker for Lower-Grade Gliomas
Article information
Abstract
Objective
The objective of this study was to assess the relationship between spermine synthase (SMS) expression, tumor occurrence, and prognosis in lower-grade gliomas (LGGs).
Methods
A total of 523 LGG patients and 1152 normal brain tissues were included as controls. Mann-Whitney U test was performed to evaluate SMS expression in the LGG group. Functional annotation analysis was conducted to explore the biological processes associated with high SMS expression. Immune cell infiltration analysis was performed to examine the correlation between SMS expression and immune cell types. The association between SMS expression and clinical and pathological features was assessed using Spearman correlation analysis. In vitro experiments were conducted to investigate the effects of overexpressing or downregulating SMS on cell proliferation, apoptosis, migration, invasion, and key proteins in the protein kinase B (AKT)/epithelialmesenchymal transition signaling pathway.
Results
The study revealed a significant upregulation of SMS expression in LGGs compared to normal brain tissues. High SMS expression was associated with certain clinical and pathological features, including older age, astrocytoma, higher World Health Organization grade, poor disease-specific survival, disease progression, non-1p/19q codeletion, and wild-type isocitrate dehydrogenase. Cox regression analysis identified SMS as a risk factor for overall survival. Bioinformatics analysis showed enrichment of eosinophils, T cells, and macrophages in LGG samples, while proportions of dendritic (DC) cells, plasmacytoid DC (pDC) cells, and CD8+ T cells were decreased.
Conclusion
High SMS expression in LGGs may promote tumor occurrence through cellular proliferation and modulation of immune cell infiltration. These findings suggest the prognostic value of SMS in predicting clinical outcomes for LGG patients.
INTRODUCTION
Lower-grade glioma (LGG) is a benign tumor originating from brain neuroglial cells. It is characterized by a relatively slow growth rate and a low level of malignancy [6,12,33]. The incidence rate of LGG among all intracranial gliomas ranges from approximately 5% to 18% [10,32,54]. Compared to other intracranial gliomas, LGGs typically exhibit a more favorable prognosis and longer overall survival (OS) [21,26,29]. Notwithstanding the generally high quality of life experienced by most LGG patients, the diagnosis and treatment of LGG still pose considerable challenges in the domains of neurosurgery and neurooncology [9,27,34,53].
Although LGG is less malignant, its treatment strategy remains complex [21,39,40]. At present, surgical resection serves as the primary treatment for LGG. However, achieving complete removal is frequently difficult because of the tumor’s location and proximity to healthy brain tissue [5,49,62]. Furthermore, despite the favorable prognosis observed in most patients with LGG, approximately 15–20% of patients experience recurrence or malignant transformation following treatment. In comparison to other types of tumors, LGGs progress slowly and generally remain inactive. However, the prognosis for LGG patients varies widely, as some may progress to high-grade gliomas (HGG) within months while others remain stable over the long term. Additionally, there are reports indicating that untreated LGG patients may eventually transition to HGG. Therefore, there is an urgent need to identify early and effective treatment targets and methods for LGG, highlighting the critical importance of exploring new therapeutic approaches and prognostic markers [4,35,36,51].
Spermine synthase (SMS) is an essential polyamine biosynthetic enzyme responsible for converting spermidine into spermine. This enzyme plays a pivotal role in various biological processes such as cell proliferation, differentiation, and apoptosis [17,18,20,60]. Under normal physiological conditions, the activity of SMS is tightly regulated to ensure the constant level of polyamines within cells [15,52,63], which is crucial for maintaining normal cellular physiological functions. However, under certain pathological conditions, especially in certain cancers, abnormal expression and activity of SMS may occur [58,60]. The accumulation of spermidine can have deleterious effects on cell viability, and tumor cells may upregulate SMS to maintain a relatively high level of spermidine below the toxic threshold to promote tumor initiation and progression. SMS serves as a protective mechanism for tumors to sustain their proliferation, thus high expression of SMS in cancer patients may indicate an unfavorable clinical prognosis [17].
In recent years, with the advancement of molecular biology technologies, researchers have discovered that SMS is upregulated in various types of cancers, significantly correlating with the initiation, progression, and prognosis of cancer [13,58]. High expression of SMS in certain tumors is closely associated with adverse clinical outcomes, high recurrence rates, and low survival rates [11,43,61]. For instance, in colorectal cancer, elevated SMS levels are negatively correlated with patient prognosis, as SMS enhances forkhead box O3 (FOXO3) acetylation modifications, thereby inhibiting Bim protein transcription, suppressing tumor cell apoptosis, and promoting tumor progression [17]. These findings underscore the close association of SMS with the pathological and physiological processes of cancer occurrence and development in patients, positioning SMS as a potential therapeutic target for cancer treatment.
Given this context, we have developed a significant interest in the role of SMS in LGG. We believe that clarifying the expression pattern and function of SMS in LGG can offer new insights into the treatment and prognosis assessment of LGG. Hence, the primary objective of this study is to examine the expression of SMS in LGG and its correlation with clinical and pathological characteristics. This research aims to provide new data for the fundamental understanding of LGG and as a valuable reference for clinical practice. The implications of this may result in improved treatment outcomes and quality of life for LGG patients.
MATERIALS AND METHODS
This study was approved by the Ethical Committee of the Air Force Medical Center, The Fourth Military Medical University (No. 2023-09-PJ01).
Data processing
We retrieved three cases of high-throughput RNA sequencing data and associated clinical and pathological information from the LGG queue in The Cancer Genome Atlas (TCGA) database. Additionally, we obtained normal tissue data from the Genotype-Tissue Expression (GTEx) database. The study included 523 patients with LGG and 1152 cases of normal brain tissue. The TCGA and GTEx RNA-seq data are processed by UCSC Xena using the TOIL pipeline and are normalized in TPM reads format (UCSC Xena is available at https://xenabrowser.net/datapages/). Tumor data from LGG in the TCGA and corresponding normal tissue data from GTEx should be extracted. The high-throughput RNA-seq data was transformed from the FPKM format, which stands for fragments per kilobase of transcript per million mapped reads, to TPM, an abbreviation for transcripts per million. Before data collection, written informed consent was acquired per the provided guidelines since the TCGA and GTEx databases are publicly accessible [45,48].
Differential gene expression analysis of genes in LGG
Based on the median expression value of SMS, 523 patients with LGG were stratified into a high-expression group and a low-expression group. For identifying differentially expressed genes (DEGs) between two groups, we utilized the R package “DESeq2” and performed a two-tailed test based on the negative binomial generalized linear model. The fold change threshold was greater than 1.5, and the adjusted p-value was less than 0.05. The results can be visualized as a volcano plot and a heatmap showing the correlation of single genes using the R packages ‘Enhanced Volcano’ and ‘pheatmap’ [28,31,37].
Enrichment analysis of SMS-related markers in LGG
Next, utilize the MetaScape online database to conduct functional enrichment analysis on the identified DEGs. DEGs are considered statistically significant when the minimum count is greater than 3, the fold change exceeds 1.5, and the p-value is less than 0.01. Furthermore, the R package “clusterProfiler” was employed to conduct gene set enrichment analysis (GSEA) on the DEGs between the two groups. In GSEA, the reference gene set selected is the C2 gene set, which consists of curated gene sets from MsigDB. A total of 64 clusters were significantly enriched, meeting the criteria of a cluster false discovery rate of less than 0.25 and a p-value of less than 0.05. Protein-protein interaction networks, generated from the string database and considered statistically significant, were visualized using the Cytoscape software (v3.7.1; Cytoscape, San Diego, CA, USA) [7,23,47,68].
The association between somatic mutations in subependymal giant cell astrocytomas and the infiltration of immune cells
Initially, we employed the ssGSEA algorithm from the R package “GSVA” to assess the infiltration levels of 24 common immune cell types, including plasma dendritic cells (pDCs), CD8+ T cells, DCs, gamma delta (γδ) T cells, central memory T cells, regulatory T cells, effector memory T cells, natural killer (NK) CD56 bright cells, NK CD56 dim cells, T follicular helper (Tfh) cells, macrophage 1 cells, Th17 cells, immature DCs, B cells, cytotoxic cells, Th2 cells, NK cells, activated DCs, neutrophils, T cells, macrophages, and eosinophils. Subsequently, we employ Spearman’s correlation analysis to assess the relationship between SMS expression and immune cell infiltration. Additionally, we performed a Mann-Whitney U test to compare immune cell infiltration among different SMS expression groups [8,59,67].
Correlation analysis assessed the relationship between SMS expression and clinical and pathological features in LGG
This study compares the clinical and pathological characteristics of patients with LGG among different groups based on SMS expression levels. The Mann-Whitney U test will be used for continuous variables, while the Pearson chi-square test will be used for ordinal variables. This study assesses the correlation between SMS expression and various clinical and pathological features through logistic analysis [25,30,38].
Flow cytometry
Cell cycle analysis was conducted using the Cell Cycle Assay Kit (E-CK-A351; Elabscience Biotechnology, Wuhan, China) to confirm the cell cycle of each group [22]. Following the treatment, cells from each group were collected using cold phosphate-buffered saline (PBS). Subsequently, they were fixed in 75% ethanol overnight at 4°C. The cell concentration was then adjusted to 1×106 cells/mL, and RNase (50 mg/mL, EN0531; ThermoFisher, Wilmington, MA, USA) and 1 mg/mL PI (25535-16-4; Sangon Biotech, Shanghai, China) were added. After incubating the samples at 37°C for 30 minutes, they were transferred to flow cytometry tubes to detect and analyze cell cycle phases, including G0/G1, S, G2, and M phases.
Analysis of apoptosis
Cell apoptosis was assessed using the Annexin V-FITC/PI apoptosis detection kit (E-CK-A211; Elabscience Biotechnology) [22]. The cells were resuspended in PBS and then incubated with Annexin V-APC and propidium iodide (PI) in a dark chamber at room temperature for 15 minutes. Apoptotic cell detection was conducted using the Annexin V-FITC/PI Apoptosis Detection Kit (E-CK-A211; Elabscience Biotechnology) and a FACScan flow cytometer [22]. Cells were resuspended in PBS and incubated with Annexin V-APC and PI in a dark chamber at room temperature for 15 minutes. Analysis was performed using the FACScan flow cytometry system (Becton Dickinson, San Diego, CA, USA). The top left quadrant represents necrotic cells, the top right quadrant corresponds to late apoptotic cells, the bottom right quadrant indicates early apoptotic cells, and the bottom left quadrant represents cells that did not undergo apoptosis. The cell cycle of each group was determined using the Cell Cycle Kit (E-CK-A351; Elabscience Biotechnology) [22]. After treatment, cells from each experimental group were collected in cold PBS and then fixed overnight at 4°C in 75% ethanol. The cell concentration was adjusted to 1×106 cells/mL, followed by the addition of RNase (50 mg/mL, EN0531; ThermoFisher) and PI (1 mg/mL, 25535- 16-4; Sangon Biotech). The cells were incubated at 37°C for 30 minutes and transferred to flow cytometry tubes to detect and analyze the cellular distribution in the cell cycle’s G0/G1, S, G2, and M phases. A cell flow cytometry system manufactured by Becton Dickinson in San Diego, CA, USA, was utilized for the analysis. The upper left quadrant represents necrotic cells, the upper right quadrant represents late apoptotic cells, the lower right quadrant represents early apoptotic cells, and the lower left quadrant represents non-apoptotic cells [22,38].
Cell culture and treatment
The HS683 (HTB-138) human oligodendroglioma cell line and the U251 (HTB-17) glioblastoma cell line were obtained from ATCC (Manassas, VA, USA). CP-H121, CP-H122, and CP-H123, human neuroglia, astroglia, and microglia cell lines, were purchased from Pricella (https://www.procell.com.cn; China). HS683 and U251 cell lines were cultured in Dulbecco’s modified eagle medium supplemented with 10% fetal bovine serum (FBS) sourced from Australia (10100147; Gibco, Glendale, CA, USA) and 1% penicillin-streptomycin (15140163; Gibco). CP-H121, CP-H122, and CP-H123 cell lines were cultured using complete medium CM-H121, CM-H122, and CMH123 respectively (Pricella, Wuhan, China). The cells were cultured under the conditions of 37°C and 5% CO2. Furthermore, the cells were treated with 1 µM spermine and spermidine for 24 hours before conducting subsequent detection [56].
Detection of SMS expression level
For Western blot analysis, the protein was extracted from cells using RIPA lysis buffer supplemented with PMSF (P0100; Solarbio, Beijing, China). The protein concentration was determined using the BCA Protein Assay Kit (Pierce, Rockford, IL, USA). Load 30 µg of protein lysate onto a 10% or 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gel. The protein was then transferred onto a polyvinylidene fluoride membrane (IPFL00010; Millipore, Billerica, MA, USA) using a semi-dry transfer apparatus manufactured by Bio-Rad Laboratories. The membrane was sealed in 5% skim milk powder for at least 1 hour. The primary antibodies, including anti-SMS antibody (ab156879; Abcam, Cambridge, UK; diluted 1:1000), anti-β-actin antibody (A5441; Sigma-Aldrich, Darmstadt, German; diluted 1:2000), anti-AKT antibody (9272; CST, Denver, MA, USA), anti-p-AKT(S473) antibody (4060; CST), anti-p-AKT(T308) antibody (13038; CST), anti-ERK antibody (4695; CST), anti-p-ERK antibody (4376; CST), anti-mTOR antibody (2983; CST), anti-p-mTOR antibody (2976; CST), anti-E-cadherin antibody (3195; CST), anti-snail antibody (3895; CST), and anti-Vimentin antibody (5741; CST), were incubated overnight at 4°C. Following the incubation period, the membranes were rinsed three times for 10 minutes each in TBST. Subsequently, the membranes were probed with either mouse anti-HRP (115-035-003; Jackson ImmunoResearch, West Chester, PA, USA) or rabbit anti-HRP (111-035-003; Jackson ImmunoResearch, Manassas, VA, USA) antibodies. The chemiluminescent substrate Plus (34580, Sigma-Aldrich) was utilized for protein detection on the Alpha Imager HP imaging system. The relative grayscale value of the target strip and the internal reference strip were analyzed using Image software to determine the relative protein expression level, with β-actin used as an internal reference [1,55].
Quantitative real-time quantitative polymerase chain reaction (qRT-PCR)
Total RNA was extracted from cells using TRIzol reagent (15596026; Invitrogen, Carlsbad, CA, USA). cDNA synthesis was performed using the ABScript II kit (RK20404; Abclone, Seoul, Korea) with 1 µg of total RNA. Subsequently, qRT-PCR was conducted on the Roche LightCycler 480 high-throughput fluorescence quantitative PCR instrument (Roche, Basel, Switzerland) using the SYBR® Green JumpStart™ Taq ReadyMix™ kit (S5193; Roche) with three replicates for each well. Specific primer sequences can be found in Supplementary Table 1. We analyzed the relative expression level of mRNA using the 2-ΔΔCt method. In this method, △△Ct is calculated as the difference between the mean Ct value of the target gene in the experimental group and the mean Ct value of the reference gene in the experimental group, subtracted from the difference between the mean Ct value of the target gene in the control group and the mean Ct value of the reference gene in the control group [65].
SMS gene silencing and overexpression experiments
The SMS gene was silenced using shRNA technology, and two shRNA molecules (shSMS-1 and shSMS-2) were designed specifically to target the SMS gene. The specific sequences for the shRNA molecules can be found in Supplementary Table 2. The shRNA was inserted into the lentiviral vector pLVX-CMV300 (110720; Addgene, Cambridge, MA, USA), while SMS was overexpressed by inserting it into the pCMV-SMS plasmid (20783; Addgene). To package lentiviruses, 1.5 µg of knockdown/overexpression plasmid was transfected into 293T cells along with the corresponding non-targeting control (shNC/con), packaging vector (0.5 µg of VSVG and 2 µg of Δ8.2), and Lipofectamine 2000 (11668030; Thermo Fisher Scientific, Wilmington, MA, USA; https://www.thermofisher.cn/cn/zh/home.html). Collect lentiviral supernatant with a high titer (>108 TU/mL) 48 and 72 hours after transfection by passing it through a 0.45 μm filter. IN THE LOGARITHMIC GROWTH PHASE, the HS683 and U251 cells were infected with lentivirus (MOI=10). After approximately 1 week, the cells were screened with 1 mg/mL puromycin (540222; Sigma-Aldrich) for subsequent experiments [44].
Cell proliferation experiment
The CCK-8 Assay Kit (CK40; Dojindo, Yonezawa, Japan) should be used in accordance with the instructions provided by the manufacturer. Transfer 1000 cultured cells to each well of a 96-well plate containing culture medium and incubate at 37°C for 0, 1, 2, 3, and 4 days. Following the incubation period, add 10 µL of CCK8 reagent to each well and measure the absorbance at 450 nm using a microplate reader (BioTek, Winooski, VT, USA; www.biotek.com/) after 2 hours [18].
Cloning for experimental purposes
Logarithmic phase cells were inoculated into a 6-well plate at 500 cells/well density. The cells were cultured for 10–14 days, replacing fresh media every 48 hours. The clones were fixed in 4% PFA for 1 hour and then stained with a 0.1% crystal violet solution (Merck KGaA; Sigma-Aldrich) at room temperature for 1 hour. The number of clones with more than 50 cells in each treatment group was counted, and this process was independently repeated three times [18].
Scratch experiment
The cells were seeded at a density of 1 million cells per well in a 6-well plate that had been pre-coated with an extracellular matrix substrate consisting of 50 µg/mL poly-L-lysine (C0313-5mg; Beyotime, Shanghai, China). A straight line was drawn using the tip of a pipette to distribute the cells. The remaining cells were cultured in fresh medium supplemented with aphidicolin C, a DNA replication inhibitor, at a concentration of 10 µg/mL (HY-13316; MedChemExpress, Montclair, NJ, USA) for an additional 12 hours. After incubation, the samples were fixed with 3.7% PFA for 15 minutes. Photographic images were captured at 0, 24, and 48 hours. The cell-free Area in each image was manually tracked using ImageJ software (National Institutes of Health, Bethesda, MD, USA). The wound area was then calculated using the following formula : wound closure percentage = (At=0 h – At=48 h) / At=0 h × 100, where At=0 h represents the initial area at time zero, and At=48 h represents the measured wound area after 48 hours of scratching [50].
Cell migration and invasion experiments
We utilized Transwell chambers (24-well plate, #3422; Corning, New York, NY, USA; https://www.corning.com/cn) that were coated with Matrigel for invasion assays and without for migration assays. The cell suspension, consisting of 1×105 cells, was prepared in a serum-free medium. It was then added to the upper chamber of a Transwell system. The lower chamber, which served as a chemotactic agent, contained 600 µL of 10% FBS medium. After incubating for 24 hours, any remaining cells in the upper chamber were removed using a cotton swab. Next, cells were fixed with 4% formaldehyde for 1 hour. Cells on the lower surface of the membrane were then stained with 0.1% crystal violet (Merck KGaA; Sigma-Aldrich) at room temperature for 1 hour. Excess dye was washed off with PBS, and the cells were air-dried before examination under a microscope at 200× magnification. Finally, the cells in five randomly selected fields of view on the lower surface of the chamber were counted [16,57]. Each sample is replicated three times, and the cells are quantified.
Statistical analysis
To assess the diagnostic significance of SMS expression in LGG, we compared its expression levels in LGG tumors and normal brain tissue. Receiver operating characteristic (ROC) analysis was conducted using the R package “pROC”. The clinical outcomes data for LGG patients are obtained from published studies, including OS, progression-free interval (PFI), and disease-specific survival (DSS). Prognostic analysis involves Kaplan-Meier and univariate and multivariate Cox regression analyses. Conduct random forest regression analysis using the R package “randomForest” [23]. The R package “rms” generates scatterplots and calibration plots. The “Qian Gan Tu” R package is used to study subgroups in clinical pathology. A significance level of 0.1 was used for the univariate and multivariate Cox regression analyses.
Furthermore, the multicollinearity test was conducted in parallel [19,66]. All statistical analyses in bioinformatics were conducted using R (v3.6.3). A p-value of less than 0.05 was deemed statistically significant for detecting differences. Data analysis for the experiment was conducted using SPSS ver. 22.0 software (IBM, Armonk, NY, USA). A comparison was made between the experimental and control groups, and a p-value of less than 0.05 was deemed statistically significant.
RESULTS
Expression of somatic mutations in multiple cancers and analysis of DEGs in LGG
We conducted a thorough analysis of 33 types of tumors in the TCGA database and found significant differences in SMS mRNA expression among 28 cancer types. This discovery highlights the potential key role of SMS in various cancers (Fig. 1A). Compared to other tissues, normal brain tissue showed lower levels of SMS expression. Furthermore, particularly in LGG, there was a significant increase in SMS expression compared to normal brain tissue, underscoring the importance of SMS in the pathogenesis of LGG (p<0.005; Fig. 1B). We conducted detailed comparisons of gene expression in LGG patients and identified 512 DEGs associated with high SMS expression. Out of these, 491 genes were upregulated, while 21 genes were downregulated (Fig. 1C) (log2 fold change >1.5, p<0.05). The differential expression of these DEGs can be visualized using a gene co-expression heatmap (Fig. 1D).
mRNA expression profile of LGG patients based on SMS expression. A : Comparison of SMS expression in different cancer types from the TCGA database with their corresponding normal tissues. ns, p≥0.05; **p<0.01; ***p<0.001. B : SMS expression is higher in LGG tumors compared to normal brain tissue. LGG patients were divided into high and low SMS expression groups based on the median expression level in LGG. C : Volcano plot showing the mRNA expression profile of the two groups. D : Heatmap depicting the results of the correlation analysis of individual genes in the two groups. SMS : spermine synthase, TPM : transcripts per million, ACC : adrenocortical carcinoma, BLCA : bladder urothelial carcinoma, BRCA : breast invasive carcinoma, CESC : cervical squamous cell carcinoma and endocervical adenocarcinoma, CHOL : cholangiocarcinoma, COAD : colon adenocarcinoma, DLBC : diffuse large B-cell lymphoma, ESCA : esophageal carcinoma, GBM : glioblastoma multiforme, HNSC : head and neck squamous cell carcinoma, KICH : kidney chromophobe, KIRC : kidney renal clear cell carcinoma, KIRP : kidney renal papillary cell carcinoma, LAML : acute myeloid leukemia, LGG : lowergrade glioma, LIHC : liver hepatocellular carcinoma, LUAD : lung adenocarcinoma, LUSC : lung squamous cell carcinoma, MESO : mesothelioma, OV : ovarian serous cystadenocarcinoma, PAAD : pancreatic adenocarcinoma, PCPG : pheochromocytoma and paraganglioma, PRAD : prostate adenocarcinoma, READ : rectum adenocarcinoma, SARC : sarcoma, SKCM : skin cutaneous melanoma, STAD : stomach adenocarcinoma, TGCT : testicular germ cell tumors, THCA : thyroid carcinoma, THYM : thymoma, UCEC : uterine corpus endometrial carcinoma, UCS : uterine carcinosarcoma, UVM : uveal melanoma, mRNA : messenger RNA, NTAN1 : N-terminal asparagine amidase 1, SPATS2L : spermatogenesis associated serine rich 2 like, TMSB10 : thymosin beta 10, MRPS15 : mitochondrial ribosomal protein S15, OAZ1 : ornithine decarboxylase antizyme 1, DSCAML1 : down syndrome cell adhesion molecule like 1, FERMT1 : fermitin family homolog 1, NOG : noggin, BMP2 : bone morphogenetic protein 2, PHYHIPL : phytanoyl-coa 2-hydroxylase interacting protein-like, TCGA : the cancer genome atlas, ns : not specified.
Functional enrichment analysis of SMS-related DEGs in LGG
To enhance our comprehension of the functions of these DEGs associated with somatic mutation signature in LGG, we conducted a functional enrichment analysis using the “Metascape” software. The results disclosed many enriched functional terms associated with short message service, encompassing cellular development, signal transduction, metabolic processes, and other significant biological processes (Fig. 2A-C, Supplementary Fig. 1). Significantly, GSEA analysis revealed notable enrichment in critical biological processes associated with cell proliferation, cell adhesion, and cell signaling receptors (Fig. 2D-J). This information gives us valuable insights into the possible biological function of SMS in LGGs.
Functional enrichment analysis of differentially expressed genes (DEGs) associated with spermine synthase (SMS) in lower-grade glioma (LGG). Using the Metascape database, functional annotation was performed on 512 DEGs between high and low SMS expression groups. All statistically enriched terms were identified and hierarchically clustered into a tree based on a kappa score threshold of 0.3 (A). The network layout shows the featured terms within each cluster (B). Node size and color represent the proportion of input genes and cluster identity, respectively. Enriched terms with a similarity score above 0.3 are connected by edges (the thickness of the edges represents the similarity score) (C). Enrichment groups connected by edges correspond to similarity scores above 0.3. Nodes are colored based on p-values within the same enrichment network (D). Gene set enrichment analysis was performed on differentially expressed messenger RNAs between high and low SMS expression groups in LGG (D-J). KEGG : Kyoto Encyclopedia of Genes and Genomes.
The correlation between immune cell infiltration in spinal muscular atrophy and LGG research studies
Infiltration of immune cells plays a vital role in the tumor microenvironment. The ssGSEA method was employed to thoroughly analyze immune cell infiltration in LGG. This analysis revealed a significant correlation between SMS and immune cell infiltrations (Fig. 3A). Specifically, it shows a positive correlation with the infiltration of eosinophils, T cells, macrophages, and neutrophils. Conversely, it exhibits a negative correlation with the infiltration of pDCs, CD8 T cells, DCs, and γδ T cells (Fig. 3A). Subsequently, we assessed the infiltration levels of six pertinent immune cell types (eosinophils, T cells, macrophages, DCs, pDCs, and CD8 T cells) in various SMS groups (Fig. 3B-G). These findings offer significant insights into the potential involvement of SMS in modulating the immune microenvironment of LGG.
Correlation of spermine synthase (SMS) expression with immune cell infiltration in lower-grade glioma patients. A : Spearman analysis displaying the correlation between SMS expression profile and levels of 24 immune cell infiltrates. Comparison of the infiltration levels of most correlated immune cells between high and low SMS expression groups, including eosinophils (B), T cells (C), macrophages (D), pDCs (E), CD8 T cells (F), and DCs (G). pDC : plasmacytoid dendritic, DC : dendritic, Tgd : transgenic disease, Tcm : central memory T, Treg : regulatory T, Tem : effector memory T, NK : natural killer, Tfh : T follicular helper, IDC : interdigitating dendritic cell, aDC : activated dendritic. **p<0.01. ***p<0.001.
The clinical significance of SMS in patients with LGG
We conducted additional investigations to examine the relationship between SMS expression and the clinical pathological features of patients with LGG (Tables 1-2). The results suggest elevated SMS expression is linked to several unfavorable clinicopathological characteristics, including age, histological type, grade, and isocitrate dehydrogenase (IDH) status.
Clinical and pathological characteristics of LGG patients with differential SMS expression
Logistic regression analysis of the relationship between clinicopathological features and SMS expression in LGG patients
We utilized Welch’s t-test to examine the disparities in clinical and pathological characteristics among LGG patients with varying levels of SMS expression, as suggested by Welch. In the SMS cohort, we identified patients who were over 40 years old (Fig. 4B), had astrocytoma (Fig. 4C), were classified as grade G3 (Fig. 4D), had a poor prognosis, including patients who died from the disease (Fig. 4E), and patients who developed progressive disease (Fig. 4F). Furthermore, the lack of co-deletions of 1p/19q (Fig. 4G) and wild type IDH (Fig. 4H) exhibited a significant association with elevated expression of SMS. However, there were no significant gender differences observed (Fig. 4A). These findings impart significant information regarding the potential value of SMS in predicting the prognosis of LGG patients.
Evaluation of the clinical significance of spermine synthase (SMS) in patients with lower-grade glioma (LGG). Welch’s t-test was used to analyze the relationship between SMS expression and gender (A), age (B), histological type (C), World Health Organization (WHO) grade (D), disease-specific survival (DSS) (E), initial treatment outcome (F), 1p/19q codeletion (G), and isocitrate dehydrogenase (IDH) status (H). Receiver operating characteristic curve analysis was performed to evaluate the diagnostic utility of SMS in LGG (I-K). Kaplan-Meier analysis of overall survival (OS) (L), progression-free survival (PFI) (M), and DSS (N) between high and low SMS expression groups. TPM : transcripts per million, CR : complete response, PR : partial response, SD : stable disease, PD : progressive disease, WT : wild type, Mut : mutation, AUC : area uner curve, FPR : false positive rate, HR : hazard ratio.
A series of statistical analyses was conducted to evaluate the predictive value of SMS in the prognosis of LGG. The results of the Kaplan-Meier analysis indicated that the group with high SMS expression had a significantly worse prognosis (Fig. 4I-K). Furthermore, time-dependent analysis of the ROC curve further validated the considerable sensitivity and specificity of SMS expression in forecasting the prognosis of patients with LGG (Fig. 4L-N).
Furthermore, we conducted univariate and multivariate Cox regression analyses to further evaluate the predictive value of SMS expression on clinical outcomes. Table 3 demonstrates that in both univariate (hazard ratio [HR], 4.284; p<0.001) and multivariate (HR, 2.587; p<0.001) Cox regression analyses, only SMS expression was found to be an independent risk factor for OS. Furthermore, the univariate Cox regression analysis identified age, histological type, World Health Organization (WHO) grade, 1p/19q co-deletion, and IDH status as predictive factors for clinical outcomes. Age, WHO grade, and IDH status emerged as significant factors in the multivariable Cox regression analysis.
Cox regression analysis of clinical outcomes in LGG patients
Subsequently, we performed a multivariable Cox regression analysis using prognostic factors determined to be statistically significant. Furthermore, a prognostic nomogram was developed, and a calibration curve was plotted to assess the efficacy of the nomogram. The prognostic nomogram included age, WHO grade, IDH status, and MGMT expression as predictors of OS, yielding a C-index of 0.828 (Supplementary Fig. 2A). Similarly, we utilized all four of these features for the column plots predicting PFI and DSS, yielding C-indices of 0.715 (Supplementary Fig. 2B) and 0.830 (Supplementary Fig. 2C). All calibration curves exhibit accurate predictions of clinical outcomes at 1-year, 3-year, and 5-year follow-ups (Supplementary Fig. 2D-F). The forecast data for OS, PFI, and DSS over 5 years exhibit high volatility and a tendency to overestimate. However, the forecast data for the 1–3-year period is stable and accurate.
The prognostic performance of SMS in the clinical-pathological subgroup of LGG patients
Furthermore, we investigated the prognostic performance of SMS in the clinical pathological subgroup of LGG. Thus, we conducted Cox regression analysis on specific subgroups, as shown in Table 4. The data results in the forest plot, as shown in Supplementary Fig. 3. The forest plots demonstrate that stress-induced telomere shortening (SMS) significantly affects clinical outcomes in various scenarios. These include individuals aged ≤40 years (HR, 3.21; p<0.001), individuals aged >40 years (HR, 5.85; p<0.001), males (HR, 7.96; p<0.001), patients with stage III disease (HR, 3.42; p<0.001), patients with co-deletion (HR, 2.94; p<0.001), patients with IDH mutation (HR, 0.001), patients with astrocytoma (HR, 5.14; p<0.001), patients with oligodendroglioma (HR, 3.39; p<0.001), and patients with stable disease as the primary treatment outcome (HR, 4.07; p<0.001). Similar findings also apply to PFI and DSS, as shown in Supplementary Fig. 3B and C. The results of the Cox regression analysis demonstrate that SMS is a significant risk factor for clinical outcomes across multiple subgroups. These findings highlight the crucial significance of SMS in predicting the prognosis of patients with LGG.
Cox regression analysis of the prognostic performance of SMS in predicting clinical outcomes in LGG patient subgroups
In addition, we assessed the clinical outcomes (OS, PFI, and DSS) of different subgroups using Kaplan-Meier analysis. The subgroups were categorized based on age, gender, WHO grade, 1p/19q co-deletion presence, IDH status, histological type, and primary treatment outcomes (Supplementary Figs. 4-7). The findings demonstrate that the clinical outcomes of the high SMS expression group are markedly poorer.
The somatic mutation score expression is strongly associated with the malignant behavior of LGG cells
The malignant behaviors of tumor cells encompass cell proliferation, apoptosis, migration, and invasion. We first detected the expression of SMS in the commonly used normal glial cell lines CP-H121/122/123 and glioma cell lines HS683/U251. Western blot results revealed a significantly higher protein level of SMS in HS683/U251 compared to CP-H121/122/123. Additionally, there was little difference in the expression levels between HS683 and U251 (Fig. 5A). Therefore, we conducted knockdown experiments to decrease the expression of SMS in both HS683 and U251 cells using shSMS-1 and shSMS-2. The knockdown efficiency was evaluated through Western blot and qRT-PCR. The results demonstrated a significant reduction in the expression levels of SMS in shSMS-1 and shSMS-2 compared to the control cell line (shNC) (Fig. 5B and C). Cell proliferation was assessed using CCK-8 assay and colony formation assay. The results demonstrate a significant decrease in the proliferative ability of cells transfected with shSMS-1 and shSMS-2 compared to shNC (Fig. 5D and F).
Knockdown of spermine synthase (SMS) inhibits the malignant behavior of lower-grade glioma cells. A : Western blot analysis of SMS protein expression levels in human normal glial cell lines CP-H121/122/123 and glioma cell line HS683/U251, with statistical analysis on the right. B : The efficiency of SMS knockdown in HS683/U251 glioma cell line detected by Western blot, with protein expression analysis on the right. C : The efficiency of SMS knockdown in HS683/U251 glioma cell line detected by quantitative reverse transcription polymerase chain reaction. D : Cell proliferation before and after SMS knockdown was measured by CCK-8 assay. E : The apoptosis ratio of cells before and after SMS knockdown was measured by flow cytometry, with statistical analysis on the right. F : Several cell clones formed before and after SMS knockdown, with statistical results. G : Migration of cells before and after SMS knockdown measured by scratch wound healing assay. H : Migration and invasive ability of cells before and after SMS knockdown measured by transwell assay. NC : normal control. ***p<0.001.
Furthermore, the flow cytometry analysis revealed a significant increase in the percentage of late-stage apoptotic cells in the shSMS-1 and shSMS-2 groups compared to the shNC group (Fig. 5E). The results of the scratch experiment indicated that, at the same time point, the relative healing area of shSMS-1 and shSMS-2 was significantly lower compared to shNC (Fig. 5G). Additionally, the Transwell experiment results demonstrated a significant reduction in the number of migrated and invaded cells in shSMS-1 and shSMS-2, compared to shNC (Fig. 5H). The results above indicate that inhibiting SMS reduces cell proliferation, migration, and invasion abilities while promoting cell apoptosis.
Simultaneously, we established an SMS cell line that overexpresses oe-SMS and its corresponding control cell line (con) to evaluate the previously mentioned indicators. Initially, the overexpression efficiency was assessed through Western blot and qRT-PCR experiments (Supplementary Fig. 8A and B). Subsequently, the proliferation ability was assessed through CCK-8 experiments and colony formation assays. The results demonstrate that the proliferation ability of oe-SMS was significantly enhanced compared to the control condition (Supplementary Fig. 8C and D). The proportion of late apoptotic cells in oe-SMS is significantly lower than con, as determined by flow cytometry (Supplementary Fig. 8E). The Scratch assay and Transwell assay results demonstrated that the migratory and invasive abilities of oe-SMS were significantly increased compared to the control group (Supplementary Fig. 8F and G). The results above indicate that overexpressing SMS enhances cell proliferation, migration, and invasion abilities while inhibiting cell apoptosis. The malignant behavior of SMS is connected to LGG cells.
The AKT/epithelial-mesenchymal transition (EMT) signaling pathway is responsible for tumor proliferation and SMS-mediated metastasis
Treatment with exogenous agmatine inhibits AKT phosphorylation, thereby suppressing cell growth. Alterations in SMS expression in LGG can potentially influence the polyamine levels, thereby affecting the activation and inhibition of AKT signaling. Phosphorylation at threonine 308 (pAKT (T308)) and serine 473 (pAKT (S473)) sites represent the phosphorylation status of the AKT protein at specific positions, which is essential for its activation. Initially, the key protein in this signal was detected using Western blot analysis. The results demonstrated that downregulation of SMS decreased the level of p-AKT, while overexpression of SMS increased the level of p-AKT, in comparison to the control (shNC and con group) (Fig. 6A).
Spermine synthase (SMS) mediates the proliferation and metastasis of lower-grade glioma (LGG) through the AKT/EMT signaling pathway. A : Western blot analysis of AKT phosphorylation levels in cells before and after SMS knockdown or overexpression, with statistical analysis on the right. B : Western blot analysis of key proteins in the EMT pathway in cells before and after SMS knockdown or overexpression, with statistical analysis on the right. C : Western blot analysis of ERK and mTOR phosphorylation levels in cells before and after SMS knockdown or overexpression, with statistical analysis on the right. D : Western blot analysis of AKT phosphorylation levels in cells treated with spermine (SPM) and spermidine (SPD) with statistical analysis on the right. E : Migration of cells treated with SPM and SPD measured by transwell assay, with statistical results. AKT : protein kinase B, pAKT : phosphorylated AKT, NC : normal control, con : control, oe : overexpression, ERK : extracellular signal-regulated kinase, p-ERK : phosphorylated ERK, mTOR : mechanistic target of rapamycin, p-mTOR : phosphorylated mtor, EMT : epithelial to mesenchymal transition. **p<0.01. ***p<0.001.
Moreover, analysis using Western blot revealed that the knockdown of SMS significantly suppressed the phosphorylation levels of the EMT signaling pathway and the ERK/mTOR, as demonstrated by comparing them to the control (shNC and con). Interestingly, the knockdown of SMS led to an increase in the levels of E-cadherin protein, a crucial protein in EMT, while causing a decrease in the levels of snail and vimentin proteins. Conversely, overexpression of SMS showed the opposite trend in the madding (Fig. 6B and C). Furthermore, our findings indicate that the addition of arginine or ornithine can effectively reverse the expression of p-AKT and modulate the migratory ability of SMS via knockdown or overexpression pathways (Fig. 6D and E). Therefore, SMS promotes the phosphorylation of AKT and impacts EMT signaling by converting spermidine into spermine, consequently boosting progression of LGG.
DISCUSSION
In reexpression pattern cent years, somatic mutation signatures expression pattern in various cancers has garnered significant attention [13,58,60]. Numerous studies have already shown that the expression of SMS is significantly elevated in specific types of cancer while notably reduced in others [17,46]. Current research indicates that elevated SMS expression is associated with poorer prognosis in various cancers, including colorectal and breast cancer. In colorectal cancer, high SMS expression is negatively correlated with patient prognosis. SMS enhances FOXO3 acetylation modification, leading to the inhibition of Bim protein transcription, which suppresses tumor cell apoptosis and promotes tumor progression [17]. In breast cancer, SMS is reported to potentially promote tumor immune escape by inhibiting immune infiltration and MHC-II antigen presentation, thereby facilitating tumor progression [11]. Our study found that the SMS expression patterns in LGG are distinct from those observed in other types of cancer, which contrasts with existing literature. This specific expression pattern might be linked to the biological characteristics and developmental mechanisms of LGG [3,14,43,64].
This study aims to further investigate the role of somatic mutation status in LGG and its association with DEGs. The results demonstrated a significant correlation between SMS and multiple DEGs. These DEGs are vital for the progression of LGG and may be linked to various traits, including tumor invasiveness, growth rate, and drug resistance. Functional enrichment analysis provides a means of gaining a deeper understanding of the biological processes associated with SMS. It has been observed that biological processes associated with SMS primarily consist of crucial processes, including cell proliferation, migration, and apoptosis. These processes are closely linked to the development and immune-related additional insights into its pathogenesis.
Immune cell infiltration in numerous cancers is widely regarded as a crucial prognostic factor [2,42,69]. Our research has discovered a significant correlation between the expression of surface marker signaling and the infiltrated cells in LGG. This correlation could be attributed to the role of SMS in modulating immune responses, which in turn impacts the prognosis of patients with LGG [58,60].
By analyzing samples from patients with LGG, we found a significant correlation between SMS expression and the clinical and pathological characteristics of the patients. Patients who exhibit high SMS expression frequently present with higher tumor grades, poorer prognosis, and increased recurrence rates. These findings offer clues regarding the potential role of SMS in LGG. Our research further validates the potential significance of the compared to other prognostic factor in predicting the prognosis of LGG. Compared to other types of patients of cancer, the expression of s strongly correlated with patients’ survival and recurrence rates. This correlation opens up new possibilities for future treatment strategies.
The main discovery of this study is the close relationship between SMS expression in LGG and patient prognosis, immune cell infiltration, and clinical pathological features. SMS exhibits specificity in tumor expression in LGG, significantly influencing its occurrence and development. This provides a crucial lead for SMS to serve as a potential target for LGG treatment, suggesting further exploration of precision targeting strategies for LGG based on the SMS target. From a clinical perspective, SMS may emerge as a key component of future LGG treatment strategies (Supplementary Fig. 9).
Based on the bioinformatics above and database analyses, we discovered high expression of SMS in LGG, which was positively correlated with eosinophils, T cells, and macrophages in immune infiltration. Conversely, there was a negative correlation with DC, pDC, and CD8+ T cells. It may be linked to the functionality of these particular immune cell types. Currently, studies have demonstrated a negative correlation between eosinophils and glioblastoma grade. It has been observed that LGG possess a higher number of eosinophils compared to HGG. This higher eosinophil count can potentially affect immune responses and the tumor microenvironment through cytokine secretion [24]. In the case of other immune cell categories, additional detailed subtyping is required for validation because they contain multiple types. Macrophages can be classified into two subtypes, M1 and M2, which have contrasting effects and show polarization towards different directions at various stages of disease progression [41].
Furthermore, previous research has demonstrated a strong correlation between elevated SMS expression and promoting cellular proliferation and migration, two processes intricately linked to AKT signaling and EMT [18]. Tests were conducted, and it was found that the phosphorylation of AKT and the expression of snail and vimentin in EMT were significantly reduced when SMS was knocked down, while E-Cadherin was significantly upregulated. Conversely, overexpressing SMS led to a significant increase in the phosphorylation of AKT and the expression of snail and Vimentin in EMT, while E-Cadherin was significantly downregulated, which aligns with previously reported findings. Moreover, we performed further analyses to detect and measure the phosphorylation levels of ERK and mTOR. The observed changes were found to be consistent with the previously mentioned findings. Our research elucidates that in LGG, SMS enhances tumor proliferation, invasion, and migration capabilities by activating the AKT/EMT pathway, thus promoting tumor occurrence and progression, which in turn adversely affects the pathological mechanisms underlying patient prognosis. However, detecting these pathways still has limitations because we have not yet explored other aspects of the signal pathways that have not been previously reported. Therefore, additional research may be required in the future.
CONCLUSION
While our study has yielded meaningful findings, it is important to acknowledge its limitations, including its small sample size and experimental methodology. Further research is needed to explore the specific role of SMS in LGG and its potential as a therapeutic target. Simultaneously, we anticipate additional clinical trials to validate our findings and enhance treatment outcomes for patients with LGG.
Notes
Conflicts of interest
No potential conflict of interest relevant to this article was reported.
Informed consent
Informed consent was obtained from all individual participants included in this study.
Author contributions
Conceptualization : ND; Data curation : CL, HZ; Formal analysis : CL; Funding acquisition : CL; Methodology : HL, XH; Project administration : MY, ZF; Visualization : CL; Writing - original draft : CL; Writing - review & editing : YW, ND
Data sharing
None
Preprint
None
Acknowledgements
The author would like to thank the Air Force Medical Center staff and Xiantao Academic Tools (www.xiantao.love) for their valuable contributions.
Supplementary materials
The online-only data supplement is available with this article at https://doi.org/10.3340/jkns.2024.0080.
qRT-PCR primer sequence
SMS knockdown sequence
Functional annotation of differentially expressed genes (DEGs) with varying levels of spermine synthase (SMS) in lower-grade glioma patients. Using the Metascape database, functional annotation was performed on 512 differentially expressed messenger RNAs between high and low SMS expression groups. Subsequently, all statistically enriched terms were further analyzed and hierarchically clustered into a tree based on a kappa score threshold of 0.3.
Construction and validation of a scoring model based on spermine synthase (SMS) expression. Nomogram risk scoring model for 1-year, 3-year, and 5-year overall survival (OS) (A), progression-free survival (PFI) (B), and disease-specific survival (DSS) (C) based on SMS expression. Calibration plots were generated to validate the utility of the scoring model in predicting OS (D), PFI (E), and DSS (F). HR : hazard ratio, CI : confidence interval, WHO : World Health Organization, IDH : isocitrate dehydrogenase, WT : wild type, Mut : mutation, PD : progressive disease, SD : stable disease, PR : partial response, CR : complete response.
Prognostic performance of spermine synthase (SMS) in predicting clinical outcomes in different subgroups of lower-grade glioma patients. Patients were divided into different subgroups based on age, gender, WHO grade, 1p/19q codeletion, IDH status, histological type, and initial treatment outcome. Cox regression was performed to evaluate the ability of SMS to predict OS (A), PFI (B), and DSS (C) in each subgroup, and the results were presented as hazard ratios. Bar graphs represent the 95% confidence intervals of the hazard ratios. WHO : World Health Organization, IDH : isocitrate dehydrogenase, Mut : mutation, WT : wild type, OS : overall survival, PFI : progression-free survival, DSS : disease-specific survival.
Kaplan-Meier analysis of OS, PFI, and DSS in six representative subgroups. Age, >40 years or ≤40 years; WHO grade G2 or G3; and IDH status, mutated or wild-type for OS (A-F), PFI (G-L), and DSS (M-R). OS : overall survival, SMS : spermine synthase, HR : hazard ratio, CI : confidence interval, WHO : World Health Organization, IDH : isocitrate dehydrogenase, Mut : mutation, WT : wild type, PFI : progression-free survival, DSS : disease-specific survival.
Kaplan-Meier analysis of overall survival for the remaining ten representative subgroups based on gender (A and B), 1p/19q co-deletion (C and D), histological type (E-G), and initial treatment outcome (H-J). SMS : spermine synthase, HR : hazard ratio, PD : progressive disease, PR : partial response.
Kaplan-Meier analysis of progression-free survival for the remaining eleven representative subgroups based on gender (A and B), 1p/19q co-deletion (C and D), histological type (E-G), and initial treatment outcome (H-K). SMS : spermine synthase, HR : hazard ratio, CR : complete response, PD : progressive disease, PR : partial response, SD : stable disease.
Kaplan-Meier analysis of disease-specific survival (DSS) for the remaining ten representative subgroups based on gender (A and B), 1p/19q co-deletion (C and D), histological type (E-G), and initial treatment outcome (H-J). SMS : spermine synthase, HR : hazard ratio, PD : progressive disease, PR : partial response, SD : stable disease.
Overexpression of spermine synthase (SMS) enhances the malignant properties of lower-grade glioma cells. A : Efficiency of SMS overexpression in CP-H121 cell line detected by qRT-PCR. B : Efficiency of SMS overexpression in CP-H121 cell line detected by Western blot. C : Cell proliferation before and after SMS overexpression was measured by CCK-8 assay. D : Several cell clones formed before and after SMS overexpression, with statistical results. E : The apoptosis ratio of cells before and after SMS overexpression was measured by flow cytometry, with statistical analysis on the right. F : Migration of cells before and after SMS overexpression measured by scratch wound healing assay. G : Migration and invasive ability of cells before and after SMS overexpression measured by transwell assay. con : control, oe-SMS : overexpression of spermine synthase, qRT-PCR : reverse transcription quantitative polymerase chain reaction. **p<0.01. ***p<0.001.
Schematic representation of the molecular mechanisms involved in the occurrence and development of lower-grade glioma (LGG) involving SMS. DC : dendritic cells, pDC : plasmacytoid dendritic cells, ERK : extracellular signal-regulated kinase, AKT : protein kinase B, mTOR : mechanistic target of rapamycin, TF : transcription factor, EMT : epithelial to mesenchymal transition.