Skip to main content
Download PDF

Comprehensive analysis of ESCRT transcriptome-associated signatures and identification of the regulatory role of LMO7-AS1 in osteosarcoma

Cancer Cell International volume 25, Article number: 29 (2025) Cite this article

This article has been updated

Abstract

Osteosarcoma (OS) is a commonly observed malignant tumor in orthopedics that has a very poor prognosis. The endosomal sorting complex required for transport (ESCRT) is important for the development and progression of cancer and may be a significant target for cancer therapy. First, we built a prognostic signature using 7 ESCRT-related genes (ERGs) to predict OS patient prognosis. Analysis of internal and external datasets revealed that the ERG signature has good predictive ability and reproducibility. Immune analysis demonstrated a significant correlation between OS patient immune status and ERG signature score. Moreover, ERG signature score was found to be associated with the response of OS patients to immunotherapy and anticancer drugs. Additionally, we constructed a prognostic signature consisting of 10 ESCRT-related long noncoding RNAs (ERLs) that effectively predicted the prognosis of OS patients. Furthermore, two subgroups of OS patients with distinct prognoses (clusters 1 and 2) were identified. Finally, LMO7-AS1 was chosen for functional experimental validation. The knockdown of LMO7-AS1 suppressed the malignant progression of OS cells. Furthermore, transcriptome sequencing was performed on OS cells and revealed a correlation between LMO7-AS1 and the PI3K-Akt signaling pathway. In conclusion, our ESCRT transcriptome-associated signatures can act as prognostic biomarkers for OS, and LMO7-AS1 is a novel therapeutic target for the treatment of OS.

Introduction

Osteosarcoma (OS) is the most common malignant bone tumor and is most common in children and adolescents [1], with a slightly higher incidence in males than in females [2]. Since the 2000s, the incidence of OS has increased threefold in patients aged 0 to 24 years [3]. OS exhibits a high metastatic potential, is highly malignant and has a high mortality rate [4]. Although chemotherapy has significantly improved the survival rate of OS, the prognosis for OS has not improved significantly in recent decades [5, 6]. The main factors contributing to the poor prognosis of OS are lung metastasis and chemotherapy resistance [7, 8]. Thus, the discovery of novel biomarkers for prognosis prediction and improvement in OS patients is urgently needed.

The ESCRT complex is involved in important biological processes such as viral outgrowth [9], vesicle outgrowth, cytoplasmic division, autophagy, and membrane repair [10, 11]. Recent studies have shown that ESCRT-mediated membrane repair can protect tumor cells from T-cell attack, and inhibition of ESCRT can promote tumor cell death and may contribute to improving cancer immunotherapy [12,13,14]. In addition, the ESCRT mechanism has been found to be related to the malignant pathological process of many tumors, including breast cancer [15], thyroid cancer [16], colorectal cancer [17], hepatocellular carcinoma [18], melanoma [19], etc. Therefore, the ESCRT machinery is significantly related to tumorigenesis and development, and may be a crucial component of tumor therapy.

Yang et al. developed a risk model that utilizes ESCRT-III pathway genes to predict the prognosis of endometrial cancer [20]. However, the entire ESCRT transcriptome has not been studied in tumors. In this study, we established an ERG signature for predicting OS patient prognosis. We then validated the ERG signature in an external dataset. Subsequently, we conducted immune analysis and predicted the response of OS patients to immunotherapy and anticancer drugs. To obtain ERLs, we evaluated the relationship between ERG and lncRNA expression levels by calculating Pearson correlation coefficient, and screened 187 ERLs based on selection criteria of |R|>0.4 and P < 0.05. Furthermore, we also identified a signature consisting of 10 ERLs and identified two subgroups of OS patients based on these 10 ERLs. Subsequent cell and animal experiments demonstrated that the inhibition of LMO7-AS1 significantly suppressed the malignant progression of OS cells. Therefore, our findings indicated that the ESCRT transcriptome-associated signatures can predict the prognosis, guide personalized treatment and offer a viable treatment target for OS patients.

Materials and methods

Data collection

We obtained gene expression profiles and clinical data for 88 OS samples from the TARGET database (https://ocg.cancer.gov/programs/target) and gene expression profiles for 396 normal samples from the GTEx database (https://gtexportal.org). The two gene expression profiles were then batch corrected and normalized using the R“sva” package, and samples with incomplete survival data were removed. Next, we downloaded expression and clinical data for 90 OS samples from the GSE21257 (n = 53) and GSE39055 (n = 37) cohorts from the GEO dataset (https://www.ncbi.nlm.nih.gov/geo/). GSE21257 and GSE39055 were combined to form an external validation cohort. A total of 251 ERGs were collected from previously published literatures (Appendix T1). Pearson correlation coefficient was calculated to assess the relationship between ERG and lncRNA expression levels. Then, 187 ERLs were screened based on the selection criteria of |R| > 0.4 and P < 0.05 (Appendix T2).

Construction, evaluation and verification of prognostic signatures

First, we used the “limma” R package to identify the differentially expressed ERGs and ERLs between OS and normal tissues. Next, we used univariate Cox analysis to screen ERGs and ERLs that were associated with patient prognosis in the TARGET database. Then, we constructed the ERG and ERL signatures using LASSO Cox regression. The risk score was calculated according to the formula:

\({\rm{Risk}}\,{\rm{ score = }}\sum\nolimits_{i = 1}^n {} \,{\rm{(Coe}}{{\rm{f}}_{\rm{i}}}{\rm{ \times }}{{\rm{x}}_{\rm{i}}}{\rm{)}}\)

where Coefi is the risk coefficient and xi is the ERG or ERL expression value.

Then, we classified patients into high-risk and low-risk groups using the median risk score. The Kaplan‒Meier (K‒M) curve was utilized to compare the overall survival of the two groups. A receiver operating characteristic (ROC) curve was used to evaluate the predictive effect. Moreover, to assess whether the risk score was an independent prognostic factor, univariate and multivariate Cox regression analyses were carried out. Subsequently, subgroup analyses were performed based on different clinical characteristics or individual genes. Finally, the GEO datasets were used as an external dataset to verify the reliability of the ERG signature.

Immune-related analyses

The ESTIMATE algorithm [21] was employed to estimate the proportion of stromal and immune cells in each patient. It mainly analyzes the gene expression profile of tumor samples, uses specific gene sets to infer the proportion of stromal cells and immune cells in tumor tissues, and generates a comprehensive score. The resulting scores were classified into four categories, namely stromal, immune and ESTIMATE score, and tumor purity. The stromal score and immune score mainly reflect the relative content of stromal cells and immune cells in tumor tissue. ESTIMATE score is a comprehensive evaluation that considers both stromal cells and immune cells. The higher the ESTIMATE score, the higher the overall proportion of stromal cells and immune cells in the tumor microenvironment. Tumor purity refers to the proportion of tumor cells in a tumor sample. The ssGSEA algorithm [22] was used to assess immune cell infiltration in each patient, while immune checkpoint molecules were analyzed via the “ggpubr” R package and presented as a box plot using R language.

Predicting immunotherapy and chemotherapy responses

We employed the TIDE algorithm [23] and subclass mapping [24] to assess patients’ treatment responses to two widely utilized clinical immune checkpoint inhibitors. Furthermore, the “pRRophetic” R package was utilized to estimate the sensitivity of patients to anticancer drugs, with the results presented in terms of the IC50.

Consensus clustering

Consensus clustering analysis was carried out using the “ConsensusClusterPlus” R package. By using the cumulative distribution function (CDF) and consensus matrix, the ideal cluster number was determined.

Functional analysis

To explore the biological processes linked to the ERL signature, we first obtained the differentially expressed genes (DEGs) between the two groups of patients using the “limma” R package. Then, we conducted Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses using the “clusterProfiler” R package. Moreover, gene set enrichment analysis (GSEA) was implemented to reveal the functions associated with the ERL signature via GSEA software.

Cell culture and transfection

The OS cell lines MG-63, SAOS-2, 143B, R-1059-D and HOS were obtained from the American Type Culture Collection (USA). The cells were cultured using MEM (Gibco, USA) supplemented with 10% fetal bovine serum (FBS; Gibco, USA). Then, the cells were placed in a 37 °C incubator containing 5% carbon dioxide. To knock down LMO7-AS1, we purchased siRNA from GenePharma Technology (Shanghai, China). LMO7-AS1 siRNA included three sequences (siLMO7-AS1-1-forward CCAGGCAUAUUCUACCAGUTT, siLMO7-AS1-1-reverse ACUGGUAGAAUAUGCCUGGTT, siLMO7-AS1-2-forward GGAUAAAGAAGUAUCCAAATT, siLMO7-AS1-2-reverse UUUGGAUACUUCUUUAUCCTT, siLMO7-AS1-3-forward GAUGGGCCAACAAGUGUUATT, siLMO7-AS1-3-reverse UAACACUUGUUGGCCCAUCTT. The negative control (NC) sequences were as follows: NC-forward UUCUCCGAACGUGUCACGUTT, NC-reverse ACGUGACACGUUCGGAGAATT).

A total of 2 × 105 cells were inoculated into each well of a 6-well plate. When the cell density reached 60–70%, the cells were transfected using the transfection reagent jetPRIME® (Polyplus, France) according to the instructions.

RNA extraction and RT‒qPCR

TRIzol reagent (Invitrogen, USA) and RNA extraction auxiliary reagents (ECOTOP, China) were employed to extract the total RNA. Following quantification, PrimeScript™ RT Master Mix (Takara, China) was used for the reverse transcription process. Next, TB Green® Premix Ex Taq™ II (Takara, China) was used to perform RT‒qPCR. Ultimately, the gene expression was calculated using the 2−ΔΔCt method with GAPDH as a reference. We used the PCR primers indicated below: LMO7-AS1 forward: 5’-CTTTCCCGGCCAAGAAACCT-3’; reverse: 5’-TGGTAGAATATGCCTGGCGG-3’. LINC01549 forward: 5’- TTGAGACTCCAGGGGTGTTC’; reverse: 5’-GCTGGAGGATGGAGCAGATT-3’; GAPDH forward: 5’-GTCGGAGTCAACGGATT-3’; reverse: 5’-AAGCTTCCCGTTCTCAG-3’.

Cell count kit-8 (CCK-8) assay

The CCK-8 experiment was utilized to evaluate cell proliferation status. Cells were seeded into 96-well plates at 2,000 cells per well. After culturing for 24 h, 48 h, and 72 h, 10 µl of CCK-8 reagent (Biosharp, China) was added per well. Subsequently, the optical density (OD) value of each well was measured at 450 nm after 2 h of incubation. Finally, cell viability was calculated using the formula provided in the manufacturer’s instructions. Cell viability = [OD (Transfected Group) – OD (Blank)] / [OD (Untransfected Group) – OD (Blank)] ×100%.

Migration and invasion assays

These experiments were performed using Transwell inserts (Corning, USA). For the migration experiment, we suspended 2 × 104 cells in 100 µl of serum-free MEM and added them to the upper chamber. For the invasion experiment, we added 1 × 105 cells suspended in 200 µl of serum-free MEM to the Matrigel-coated (Corning, USA) upper chamber. Then, 600 µl of MEM containing 30% FBS was added to the lower chamber. For the migration and invasion experiments, cells were cultured for 24 h and 48 h, respectively. Next, the cells were fixed with 4% paraformaldehyde (Biosharp, China) for 30 min, after which they were stained using crystal violet (Beyotime, China) for 15 min. Finally, each chamber was cleaned, the cells on the upper chamber were gently wiped away, and the cells were observed and imaged using a microscope (Olympus, Japan).

Establishment of stable LMO7-AS1 knockdown OS cells

Lentivirus (LV) containing packaged short hairpin RNA plasmids was procured from General Biological Corporation (Anhui, China). When the cell density in the six-well plate reaches 20–30%, the cells will be infected with 1 × 106 units of recombinant lentivirus with the assistance of 6 µg/ml polybrene. Finally, 1.5 µg/ml puromycin (Biosharp, China) was used to select cells with stable knockdown.

Xenograft tumor formation assay

Four-week-old BALB/c nude mice were acquired from Hangzhou Ziyuan Laboratory Animal Technology Co., Ltd (Zhejiang, China). The animal experiment strictly followed the ARRIVE guidelines and approved by the Laboratory Animal Ethics Committee of Anhui Medical University. OS cells were suspended at 5 × 107 cells/ml density in a mixture of PBS and Matrigel (1:1). Subsequently, 100 µl of LV-control 143B cells and LV-LMO7-AS1 143B cells were separately injected subcutaneously on the ventral side of nude mice. The volume of tumors was assessed every 7 days using the formula V = 0.5 ×length × width2. All nude mice were euthanized 28 days after cell inoculation, and tumor tissues were subsequently collected for further study.

Immunohistochemistry (IHC)

The tumor tissues from the control group (LV- control) and experimental group (LV-LMO7-AS1) were fixed using 4% paraformaldehyde. Subsequently, the tissues were fixed in paraffin and cut into 4 μm slices for hematoxylin and eosin (HE) staining and IHC staining. IHC staining was conducted employing horseradish peroxidase-conjugated IgG and primary antibodies targeting specific antigens, namely proliferating cell nuclear antigen (PCNA) and Ki67 (Affinity, China). Protein staining was performed utilizing DAB substrate liquid (Biosharp, China). Finally, positive brown staining was observed and photographed by light microscopy.

Transcriptome sequencing analysis

Total RNA was extracted from the interference group of si-LMO7-AS1 and the control group of 143B cells, and technical services were provided by Beijing Novogene Bioinformatics Technology Co., Ltd (Beijing, China). RNA samples meeting the quality criteria were selected for transcriptome sequencing. The “DESeq” R package was utilized to obtain DEGs according to the criteria of |log2FC| > 0.7 and p < 0.05. Finally, GO and KEGG enrichment analyses were performed on these DEGs.

Statistical analysis

R software 4.2.1, GraphPad Prism 9.0.0, and Cytoscape 3.8 were used for statistical analysis and data visualization. Student’s t test or one-way ANOVA was employed to compare groups of continuous variables. All experiments were performed at least three times. P < 0.05 was considered to indicate a statistically significant difference. *p < 0.05; **p < 0.01; ***p < 0.001; ns p > 0.05.

Results

Construction and evaluation of the ERG prognostic signature

Firstly, we conducted differential expression analysis on 251 ERGs in the TARGET and GTEx databases and identified 60 differentially expressed ERGs, as shown in the heatmap (Fig. 1).We then screened 9 prognostic genes by univariate Cox regression analysis (Fig. 2A). Subsequently, through LASSO Cox regression analysis (Fig. 2B), we constructed a risk signature bases on 7 ERGs: CDK1, CBL, CHMP4C, SH3KBP1, MYD88, APOE, and RILP. Then, the OS patients were split into high-risk and low-risk groups. The K‒M curve demonstrated a significantly lower overall survival rate in the high-risk group (Fig. 2C). Furthermore, the prediction accuracy of the ERG signature was assessed using the ROC curve. The larger the area under the curve (AUC), the better the accuracy. As shown in Fig. 2D, the ERG signature had great predictive power, with AUC values of 0.679, 0.747 and 0.733 for 1, 3 and 5 years, respectively. We also plotted the risk score distribution map, the survival status distribution map, and 7 ERG expression profiles (Supplementary Fig. 1A). Next, we compared the expression levels of the 7 ERGs between the two risk groups. The findings revealed that the expression levels of CDK1, CBL, and CHMP4C were higher in the high-risk group, while the expression levels of SH3KBP1, MYD88, APOE, and RILP were lower (Supplementary Fig. 1B). Finally, the univariate and multivariate Cox analyses revealed that metastasis and risk score were independent predictors of patient outcome (Fig. 2E, F).

Fig. 1
figure 1

Heatmap of 60 differentially expressed ERGs

Full size image
Fig. 2
figure 2

Construction and evaluation of the ERG prognostic signature. (A) Univariate Cox regression analysis identified 9 ERGs that were correlated with OS patients’ overall survival. (B) Lasso Cox regression analysis ultimately identified 7 ERGs to construct a prognostic signature. (C) K‒M curve. (D) ROC curve analyses for 1, 3 and 5 years. (E-F) Univariate and multivariate Cox regression analyses

Full size image

Validation of the ERG prognostic signature

Based on the ERG prognostic signature in the TARGET cohort, 90 samples in the GEO cohort were also divided into two risk groups. The K‒M curve revealed that high-risk patients had lower survival rates (Supplementary Fig. Appendix T2A). The AUCs of the ROC curves were 0.849, 0.676 and 0.637 for 1, 3 and 5 years, respectively (Supplementary Fig. Appendix T2B). Supplementary Fig. Appendix T2C-E show the risk score, survival status, and levels of the 7 ERGs. These results suggested that the ERG prognostic signature has favorable reproducibility.

Immune analysis in the ERG prognostic signature

We used the ESTIMATE algorithm to score each sample. Fig. 3A–D demonstrate that the high-risk group had higher tumor purity and lower stromal, immune, and ESTIMATE scores. We further analyzed the difference in immune cell abundance between the two groups using the ssGSEA algorithm. The findings revealed higher infiltration of B cells, T helper cells, and Tcm cells in the high-risk group but lower infiltration of CD8 + T cells, cytotoxic cells, iDCs, macrophages, Mast cells, neutrophils, NK cells, T cells, Th1 cells, and Th17 cells (Fig. 3E, F). Moreover, we analyzed differences in immune checkpoint molecule levels. The results indicated higher expression levels of TNFRSF9, CD86, CD274, TNFRSF18, B2M, CD40, PDCD1LG2, ICOSLG, CD8A, LGALS9, CD80, HAVCR2, PTPRC, CD28, CTLA4, CD40LG, and LAG3 in the low-risk group, while IL12A exhibited lower expression (Fig. 3G).

Fig. 3
figure 3

Immune analysis in the ERG prognostic signature. (A-D) Differences in stromal score, immune score, ESTIMATE score, and tumor purity between the two groups. (E) The heatmap displays the infiltration levels of 24 immune cell types. (F) The boxplots show the difference in immune cells between the two groups. (D) The boxplots show the difference in immune checkpoint molecules between the two groups

Full size image

Prediction of immunotherapy and chemotherapeutic responses in the ERG prognostic signature

Given the difference in immune status between the two risk groups, we speculated that the two groups of patients might have different responses to immunotherapy. Therefore, we investigated the different responses of the two groups of patients to immune checkpoint blockade therapy (Fig. 4A). We found that anti-PD1 immunotherapy was more effective in low-risk patients (Fig. 4B). Chemotherapy is an imperative component in OS treatment. Hence, we also assessed the response of patients to commonly utilized chemotherapy medications. Our findings indicated that the low-risk group demonstrated potentially greater sensitivity to the 6 tested medications (dasatinib, MG-132, midostaurin, PF-02341066, roscovitine, and SB-216763) (Fig. 4C-H). Moreover, the high-risk group exhibited potentially greater sensitivity to OSI-906, thapsigargin, and vorinostat (Fig. 4I-K).

Fig. 4
figure 4

Prediction of immunotherapy and chemotherapeutic responses in the ERG prognostic signature. (A) TIDE values and predictions of response to immunotherapy. (B) Submap analysis suggested that low-risk groups were more likely to benefit from anti-PD-1 therapy. (C-K) Estimated IC50 values showed differences in sensitivity between the two groups for 9 chemotherapy drugs

Full size image

Screening of ERLs and construction of the ERL prognostic signature

By calculating Pearson correlation coefficients of the expression levels of ERGs and lncRNAs, we screened 187 ERLs using |R| > 0.4 and P < 0.05 as criteria (Supplementary Fig. 3). Next, we identified 95 differentially expressed ERLs in normal and OS tissues (Fig. 5). Then, 18 prognosis-related ERLs were screened by univariate Cox regression analysis (Fig. 6A). Finally, the ERL prognostic signature was established using LASSO Cox regression analysis (Fig. 6B).

Fig. 5
figure 5

Heatmap of 95 differentially expressed ERLs

Full size image
Fig. 6
figure 6

Construction and evaluation of the ERL prognostic signature. (A) Univariate Cox regression analysis identified 18 ERLs that were correlated with OS patients’ overall survival. (B) Lasso Cox regression analysis ultimately identified 10 ERLs to construct prognostic signature. (C) K‒M curve. (D) ROC curve analyses for 1, 3 and 5 years. (E-F) Univariate and multivariate Cox regression analyses

Full size image

Evaluation of the ERL prognostic signature

The K‒M curve demonstrated a significantly lower overall survival rate in the high-risk group (Fig. 6C). The AUCs of the ROC curves were 0.821, 0.873 and 0.893 for 1, 3 and 5 years, respectively (Fig. 6D). The risk score distribution map and the survival status distribution map were also plotted (Supplementary Fig. 4A-C). The univariate and multivariate Cox analyses revealed that metastasis and risk scores were independent predictors of patient outcome (Fig. 6E, F). Finally, we conducted a stratified analysis to examine the prognostic value of the ERL risk signature in different clinical subgroups. The K‒M curve revealed that high-risk patients in different clinical subgroups had worse outcomes (Supplementary Fig. 1).

Consensus clustering

Based on the 10 ERLs in the ERL signature, we performed a consensus cluster analysis and selected K = 2 for further analysis (Fig. 7A-C). Fig. 7E shows that the OS patients were divided into cluster 1 (n = 45) and cluster 2 (n = 40). The K‒M curve indicated a significantly improved prognosis in cluster 1 compared to cluster 2 (Fig. 7D). Although cluster 1 had a higher stromal score and ESTIMATE score and lower tumor purity than cluster 2, the immune score and immune cell infiltration levels did not differ significantly (Fig. 7F).

Fig. 7
figure 7

Consensus clustering based on the 10 ERLs. (A) The CDF of consensus clustering. (B) The CDF curve’s length and slope when the index ranged from 2 to 9. (C) The consensus score matrix of 85 OS samples when k = 2. (D) K‒M curves of two subclusters. (E) Heatmap of the expression of 10 ERLs in two subclusters. (F) Enrichment levels of 24 immune cell types in two subclusters

Full size image

Functional analysis of the ERL prognostic signature

We performed GO, KEGG, and GSEA analyses to explore potential functional characteristics associated with the ERL signature. The GO enrichment analysis revealed a predominant enrichment of DEGs in biological processes associated with immune system activation (Supplementary Fig. 2A). Furthermore, cell-substrate junctions and focal adhesions associated with the malignant progression of tumors were also enriched (Supplementary Fig. 2A). The KEGG analysis demonstrated that these DEGs were linked to the MAPK and Rap1 signaling pathways (Supplementary Fig. 2B). These signaling pathways are associated with malignant phenotypes of tumors. Finally, GSEA revealed a strong relationship between the ERL signature and PI3K-Akt-mTOR signaling, epithelial-mesenchymal transition, inflammatory response, and apoptosis (Supplementary Fig. 2C-F).

Silencing LMO7-AS1 suppressed the malignant progression of OS cells in vitro

To learn more about the roles of these 10 ERLs in OS, we analyzed these 10 lncRNAs and found that all 10 lncRNAs were expressed at higher levels in OS tissue (Figs. 5 and 8A). Research evidence has confirmed that ELFN1-AS1, UNC5B-AS1, and HIF1A-AS2 play a malignant role in OS [25,26,27]. The malignant effects of FOXN3-AS1, MIRLET7BHG, and SENCR have also been reported in other tumors [28,29,30]. Moreover, we found that upregulation of LINC01549 and LMO7-AS1 was associated with worse prognosis in OS, while OLMALINC and LBX2-AS1 did not show this trend (Supplementary Fig. 3A-J). Therefore, we selected LOM7-AS1 and LINC01549 for RT‒qPCR experiments.

Fig. 8
figure 8

Silencing LMO7-AS1 inhibits OS cells proliferation, migration and invasion in vitro. (A) Relative expression of LMO7-AS1 in normal and OS tissues. (B) Relative expression of LMO7-AS1 in OS cell lines. (C, D) The relative silencing levels of LMO7-AS1 in HOS (C) and 143B (D) cells. (E–F) Effects of LMO7-AS1 silencing on the proliferation of HOS (E) and 143B (F) cells. (G, H) Images (G) and counts (H) of HOS and 143B cell migration experiments after LMO7-AS1 silencing. (I–J) Images (I) and counts (J) of HOS and 143B cell invasion experiments after LMO7-AS1 silencing

Full size image

The expression of LMO7-AS1 and LINC01549 in OS cells was examined using RT‒qPCR. The RT‒qPCR outcomes demonstrated that 143B and HOS cells expressed LMO7-AS1 at higher levels (Fig. 8B), while LINC01549 was only higher in Saos-2 cells (Supplementary Fig. 4). Consequently, we selected 143B and HOS cells to study the function of LMO7-AS1 in OS. After transfection of 143B and HOS cells for 24 h, LMO7-AS1 knockdown effectiveness was assessed by RT‒qPCR. The results revealed that si-LMO7-AS1 sequences 1 and 2 were more effective at knocking down LMO7-AS1 (Fig. 8C-D). When LMO7-AS1 was silenced, the CCK-8 assay revealed that compared with the NC group, the proliferation rates of 143B and HOS cells were significantly reduced statistically (Fig. 8E-F). Furthermore, silencing LMO7-AS1 inhibited the migration and invasion of OS cells (Fig. 8G-J).

Silencing LMO7-AS1 inhibited tumor growth in vivo

We subsequently examined the effect of LMO7-AS1 silencing on in vivo OS growth. We constructed 143B cells with stable LMO7-AS1 knockdown using lentivirus. These cells, along with control cells, were later inoculated into nude mice. The findings revealed that the LMO7-AS1 knockdown group had significantly lower tumor volume and weight than the control group (Fig. 9A-C). Furthermore, IHC analysis revealed a significant reduction in the levels of the proliferation-related proteins Ki67 and PCNA in the low LMO7-AS1 expression group (Fig. 9D).

Fig. 9
figure 9

Silencing LMO7-AS1 inhibits OS growth in vivo. (A) Subcutaneous tumors were dissected and photographed after 4 weeks. (B) Subcutaneous tumor volume was recorded every week. (C) Subcutaneous tumor weight after 4 weeks. (D) Immunohistochemical analysis of HE, Ki67, and PCNA

Full size image

LMO7-AS1 is involved in the PI3K-Akt signaling pathway in OS cells

To further investigate the mechanism of LMO7-AS1 in OS, transcriptome sequencing was performed to examine the altered gene expression in 143B cells with LMO7-AS1 knockdown. Compared to the control group, the group with LMO7-AS1 silencing had 201 upregulated genes and 373 downregulated genes (Fig. 10A). GO enrichment analysis demonstrated that the DEGs were mainly enriched in epithelial cell proliferation (Fig. 10B). KEGG pathway analysis demonstrated that the DEGs were primarily enriched in the PI3K-Akt signaling pathway (Fig. 10C). Fig. 10D shows the 18 DEGs enriched in the PI3K-Akt pathway.

Fig. 10
figure 10

Transcriptome sequencing analysis. (A) Volcano plot of DEGs. (B) GO and (C) KEGG enrichment analysis. (D) 18 DEGs enriched in the PI3K-Akt pathway

Full size image

Discussion

OS is the most common primary malignant bone tumor [1]. Despite the use of surgical treatment and adjuvant chemotherapy, the outcomes are not satisfactory [31]. Over the past 40 years, OS treatment has shown limited progress [32]. Recently, there have been significant advancements in studying prognostic signatures for OS. Wu and colleagues established a prognostic signature for OS based on the tumor microenvironment, which not only enables accurate prediction of patient prognosis but also provides guidance for immunotherapy and chemotherapy treatments [33]. Moreover, cuproptosis functions as an emerging biomarker for assessing the prognosis and immunotherapy response of OS [34]. Additionally, the metastasis-related gene signature constructed by Qin et al. is capable of predicting the prognosis of OS patients and presents five potential therapeutic targets [35]. Although many prognostic signatures have been constructed for prognosis prediction and treatment guidance for OS, the association between ESCRT-related signatures in OS and overall survival has not been thoroughly investigated. Therefore, this study focused on the construction and evaluation of ESCRT-related prognostic signatures of OS for the first time.

In our research, we constructed ESCRT transcriptome-associated signatures, which were superior to traditional classification systems in predicting prognosis in patients with OS. Additionally, the ESCRT transcriptome-associated signatures displayed independence as prognostic markers in OS. Moreover, the ESCRT transcriptome-associated signatures were correlated with the immune status of OS patients and helped to predict the outcomes of immunotherapy and chemotherapy treatment in OS patients. Then, through consensus clustering, we identified two subgroups of OS patients with significantly different prognoses. Finally, we identified LMO7-AS1 as a promising therapeutic target for OS and found a correlation between LMO7-AS1 and the PI3K-Akt signaling pathway.

Immune analysis revealed that high-risk patients with a worse prognosis had less immune cell infiltration and lower immune, stromal, and ESTIMATE scores but higher tumor purity. The tumor immune microenvironment (TIME) comprises tumor cells, immune cells, and cytokines, and it plays a pivotal role in tumor development and metastasis [36, 37]. The infiltration of immune cells in OS primarily consists of T lymphocytes and macrophages [38]. T cells, especially CD8+ T cells, play a vital role in the antitumor response in OS [38]. CD8+ T cells are the most critical antitumor immune cells and are effective at clearing cancer cells [39]. In the TIME, CD8+ T cells are continuously stimulated due to prolonged exposure to the antigen, and exhausted CD8+ T cells gradually lose their effector function [40, 41]. Exhaustion of CD8+ T cells is one of the leading causes of immune dysfunction in cancer patients [39]. Fritzsching et al. found that patients with OS had a better prognosis with a higher CD8+ T-cell/FOXP3 regulatory T-cell ratio [42]. In addition, recent studies have shown that infiltration of CD8+ T cells in tumors is positively correlated with patient response to PD-1 therapy [43,44,45]. Therefore, CD8+ T cells will be the central focus of future targeted therapy and immunotherapy for the treatment of OS. Furthermore, tumor-associated macrophages (TAMs) are intimately connected to tumor occurrence, metastasis, and drug resistance [46,47,48,49]. TAMs include M1 and M2 subtypes, representing the two extremes of TAMs. Multiple studies have revealed that M1-type TAMs may prevent OS development and metastasis [50, 51]. Conversely, M2-type TAMs have the ability to enhance the migration and invasion of OS cells, subsequently resulting in unfavorable patient outcomes [52, 53]. Several studies have employed all-trans retinoic acid to suppress the differentiation of M2-type TAMs as a therapeutic approach for OS and metastasis prevention [52, 54]. Thus, strategies for harnessing the antitumor function of TAMs, inhibiting the protumor activity of TAMs, and thereby improving the prognosis of OS patients will be very important.

In recent years, immunotherapy has developed rapidly and achieved remarkable performance in the treatment of tumors [55,56,57]. Immune checkpoint inhibitors (ICIs) are commonly used in oncology immunotherapy. Although numerous patients would benefit, most cancer patients do not respond to the therapy [58]. Therefore, it is critical to distinguish patients who are sensitive to ICI therapy. Patients in the low-risk group exhibited higher sensitivity to PD-1 therapy, in line with previous studies [59, 60]. Despite the potential benefits of immunotherapy for patients with OS, adjuvant chemotherapy continues to be recognized as a crucial treatment approach. According to the predicted IC50 results, the two groups of patients exhibited different sensitivities to the 9 chemotherapy drugs. Previous studies have reported the positive effects of dasatinib, MG-132, midostaurin, PF-02341066, roscovitine, SB-216,763, OSI-906, and vorinostat in OS treatment [61,62,63,64,65,66,67,68]. However, there are few reports on the use of thapsigargin in OS. Thapsigargin is an inhibitor of the endoplasmic reticulum Ca2 + pump that exhibits anti-inflammatory and antitumor activities against multiple types of tumors [69]. Thus, thapsigargin potentially serves as an efficacious chemotherapeutic agent for OS treatment. In conclusion, our research findings may have the potential to provide novel approaches for personalized treatment of OS patients.

LMO7-AS1 is an antisense RNA that belongs to the lncRNA class. Through bioinformatics analysis, several studies have found that the elevated expression of LMO7-AS1 is correlated with reduced survival rates in patients with Wilms tumor [70] and colorectal cancer [71]. However, the role of LMO7-AS1 in OS remains unclear. Functional experiments showed that LMO7-AS1 knockdown can significantly inhibit the proliferation, migration and invasion of OS cells in vitro. Moreover, the xenograft mouse model demonstrated that silencing LMO7-AS1 can inhibit tumor growth in vivo and reduce the expression of Ki67 and PCNA. These results indicated that the silencing of LMO7-AS1 inhibited the malignant progression of OS cells. Hence, LMO7-AS1 shows potential as a viable therapeutic target for treating OS.

Finally, transcriptome sequencing of OS cells revealed a close association between LMO7-AS1 and the PI3K-Akt signaling pathway. The PI3K pathway is primarily activated by AKT, and most OS patients have elevated AKT and p-AKT levels [72]. Multiple investigations have indicated that the activation of the PI3K-Akt pathway is related to the proliferation, metastasis, and drug resistance of OS [73,74,75,76]. In summary, LMO7-AS1 is closely associated with the PI3K-Akt pathways, and targeting this pathway may provide hope for the treatment of OS. However, the mechanism by which LOM7-AS1 regulates the PI3K-Akt pathway in OS remains unclear, so additional research is needed.

Conclusions

In summary, we identified ESCRT transcriptome-associated signatures that exhibited good predictive ability for OS patient survival. Furthermore, we identified the crucial role of LMO7-AS1 in OS. LMO7-AS1 may affect the malignant progression of OS cells. Additionally, LMO7-AS1 is related to the PI3K-Akt signaling pathway and further investigation is necessary to examine the effect of LMO7-AS1 on the PI3K-Akt pathway. Overall, our study provides valuable insights into prognosis prediction and personalized treatment for OS patients and reveals a novel target for OS treatment.

Data availability

The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding authors.

Change history

  • 14 February 2025

    The incorrect ESM figures and labels was mapped correctly.

Abbreviations

OS:

Osteosarcoma

ESCRT:

Endosomal sorting complex required for transport

ERGs:

ESCRT-related genes

ERLs:

ESCRT- related long noncoding RNAs

K‒M:

Kaplan‒Meier

ROC:

Receiver operating characteristic

CDF:

Cumulative distribution function

DEGs:

Differentially expressed genes

GO:

Gene ontology

KEGG:

Kyoto encyclopedia of genes and genomes

GSEA:

Gene set enrichment analysis

FBS:

Fetal bovine serum

NC:

Negative control

CCK8:

Cell count kit-8

LV:

Lentivirus

IHC:

Immunohistochemical

HE:

Hematoxylin and eosin

PCNA:

Proliferating cell nuclear antigen

AUC:

Area under the curve

TIME:

Tumor immune microenvironment

TAMs:

Tumor-associated macrophages

ICIs:

Immune checkpoint inhibitors

References

  1. Mirabello L, Troisi RJ, Savage SA. Osteosarcoma incidence and survival rates from 1973 to 2004: data from the Surveillance, Epidemiology, and end results program. Cancer. 2009;115(7):1531–43.

    Article  PubMed  Google Scholar 

  2. Mirabello L, Troisi RJ, Savage SA. International osteosarcoma incidence patterns in children and adolescents, middle ages and elderly persons. Int J Cancer. 2009;125(1):229–34.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  3. Cole S, Gianferante DM, Zhu B, Mirabello L. Osteosarcoma: a Surveillance, Epidemiology, and end results program-based analysis from 1975 to 2017. Cancer. 2022;128(11):2107–18.

    Article  PubMed  Google Scholar 

  4. Liu W, Zhao Y, Wang G, Feng S, Ge X, Ye W, Wang Z, Zhu Y, Cai W, Bai J, et al. TRIM22 inhibits osteosarcoma progression through destabilizing NRF2 and thus activation of ROS/AMPK/mTOR/autophagy signaling. Redox Biol. 2022;53:102344.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  5. Berner K, Johannesen TB, Berner A, Haugland HK, Bjerkehagen B, Bøhler PJ, Bruland ØS. Time-trends on incidence and survival in a nationwide and unselected cohort of patients with skeletal osteosarcoma. Acta Oncol (Stockholm Sweden). 2015;54(1):25–33.

    Article  Google Scholar 

  6. Smeland S, Bielack SS, Whelan J, Bernstein M, Hogendoorn P, Krailo MD, Gorlick R, Janeway KA, Ingleby FC, Anninga J et al. Survival and prognosis with osteosarcoma: outcomes in more than 2000 patients in the EURAMOS-1 (European and American Osteosarcoma Study) cohort. European journal of cancer (Oxford, England: 1990) 2019, 109:36–50.

  7. Isakoff MS, Bielack SS, Meltzer P, Gorlick R. Osteosarcoma: current treatment and a collaborative pathway to Success. J Clin Oncology: Official J Am Soc Clin Oncol. 2015;33(27):3029–35.

    Article  CAS  Google Scholar 

  8. Khanna C, Fan TM, Gorlick R, Helman LJ, Kleinerman ES, Adamson PC, Houghton PJ, Tap WD, Welch DR, Steeg PS, et al. Toward a drug development path that targets metastatic progression in osteosarcoma. Clin cancer Research: Official J Am Association Cancer Res. 2014;20(16):4200–9.

    Article  Google Scholar 

  9. Russell T, Samolej J, Hollinshead M, Smith GL, Kite J, Elliott G. Novel Role for ESCRT-III Component CHMP4C in the Integrity of the Endocytic Network Utilized for Herpes Simplex Virus Envelopment. mBio 2021, 12(3).

  10. Willan J, Cleasby AJ, Flores-Rodriguez N, Stefani F, Rinaldo C, Pisciottani A, Grant E, Woodman P, Bryant HE, Ciani B. ESCRT-III is necessary for the integrity of the nuclear envelope in micronuclei but is aberrant at ruptured micronuclear envelopes generating damage. Oncogenesis. 2019;8(5):29.

    Article  PubMed Central  PubMed  Google Scholar 

  11. Vietri M, Radulovic M, Stenmark H. The many functions of ESCRTs. Nat Rev Mol Cell Biol. 2020;21(1):25–42.

    Article  CAS  PubMed  Google Scholar 

  12. Ritter AT, Shtengel G, Xu CS, Weigel A, Hoffman DP, Freeman M, Iyer N, Alivodej N, Ackerman D, Voskoboinik I, et al. ESCRT-mediated membrane repair protects tumor-derived cells against T cell attack. Sci (New York NY). 2022;376(6591):377–82.

    Article  CAS  Google Scholar 

  13. Jia Y, Peng Z, Li J, Qin Y, Wang G. Unilateral percutaneous kyphoplasty with O-Arm Navigation for the Treatment of Kümmell’s Disease. J pain Res. 2022;15:257–66.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  14. Monypenny J, Milewicz H, Flores-Borja F, Weitsman G, Cheung A, Chowdhury R, Burgoyne T, Arulappu A, Lawler K, Barber PR, et al. ALIX regulates tumor-mediated immunosuppression by Controlling EGFR Activity and PD-L1 Presentation. Cell Rep. 2018;24(3):630–41.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  15. Guo Y, Cui Y, Li Y, Jin X, Wang D, Lei M, Chen F, Liu Y, Xu J, Yao G, et al. Cytoplasmic YAP1-mediated ESCRT-III assembly promotes autophagic cell death and is ubiquitinated by NEDD4L in breast cancer. Cancer Commun (London England). 2023;43(5):582–612.

    Article  PubMed Central  Google Scholar 

  16. Liu RT, Huang CC, You HL, Chou FF, Hu CC, Chao FP, Chen CM, Cheng JT. Overexpression of tumor susceptibility gene TSG101 in human papillary thyroid carcinomas. Oncogene. 2002;21(31):4830–7.

    Article  CAS  PubMed  Google Scholar 

  17. Chen F, Zhang L, Wu J, Huo F, Ren X, Zheng J, Pei D. HCRP-1 regulates EGFR-AKT-BIM-mediated anoikis resistance and serves as a prognostic marker in human colon cancer. Cell Death Dis. 2018;9(12):1176.

    Article  PubMed Central  PubMed  Google Scholar 

  18. Zheng Y, Yang S, Dai W, Wang J, Bi S, Zhang X, Zheng Z, Sun Y, Wu S, Kong J. CHMP3 promotes the progression of hepatocellular carcinoma by inhibiting caspase–1–dependent pyroptosis. Int J Oncol 2024, 64(1).

  19. Zhang W, Yang J, Wang B, Lu Y, Yang J, Zhong W, Yu Z, Qin Z, Xiao B, Wang K, et al. HRS mediates tumor immune evasion by regulating proteostasis-associated interferon pathway activation. Cell Rep. 2023;42(11):113352.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  20. Yang Y, Wang M. Genomic analysis of the endosomal sorting required for transport complex III pathway genes as therapeutic and prognostic biomarkers for endometrial carcinoma. Translational cancer Res. 2022;11(9):3108–27.

    Article  CAS  Google Scholar 

  21. Yoshihara K, Shahmoradgoli M, Martínez E, Vegesna R, Kim H, Torres-Garcia W, Treviño V, Shen H, Laird PW, Levine DA, et al. Inferring tumour purity and stromal and immune cell admixture from expression data. Nat Commun. 2013;4:2612.

    Article  PubMed  Google Scholar 

  22. Barbie DA, Tamayo P, Boehm JS, Kim SY, Moody SE, Dunn IF, Schinzel AC, Sandy P, Meylan E, Scholl C, et al. Systematic RNA interference reveals that oncogenic KRAS-driven cancers require TBK1. Nature. 2009;462(7269):108–12.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  23. Jiang P, Gu S, Pan D, Fu J, Sahu A, Hu X, Li Z, Traugh N, Bu X, Li B, et al. Signatures of T cell dysfunction and exclusion predict cancer immunotherapy response. Nat Med. 2018;24(10):1550–8.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  24. Hoshida Y, Brunet JP, Tamayo P, Golub TR, Mesirov JP. Subclass mapping: identifying common subtypes in independent disease data sets. PLoS ONE. 2007;2(11):e1195.

    Article  PubMed Central  PubMed  Google Scholar 

  25. Wang B, Wang X, Li P, Niu X, Liang X, Liu G, Liu Z, Ge H. Osteosarcoma Cell-Derived Exosomal ELFN1-AS1 mediates macrophage M2 polarization via sponging mir-138-5p and miR-1291 to promote the tumorgenesis of Osteosarcoma. Front Oncol. 2022;12:881022.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  26. Yan Y, Liu X, Li Y, Yan J, Zhao P, Yang L. EPB41L4A-AS1 and UNC5B-AS1 have diagnostic and prognostic significance in osteosarcoma. J Orthop Surg Res. 2023;18(1):261.

    Article  PubMed Central  PubMed  Google Scholar 

  27. Wang X, Peng L, Gong X, Zhang X, Sun R. LncRNA HIF1A-AS2 promotes osteosarcoma progression by acting as a sponge of miR-129-5p. Aging. 2019;11(24):11803–13.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  28. Molaei Ramshe S, Ghaedi H, Omrani MD, Geranpayeh L, Alipour B, Ghafouri-Fard S. Up-regulation of FOXN3-AS1 in invasive ductal carcinoma of breast cancer patients. Heliyon. 2021;7(10):e08179.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  29. Xia Y, Zhen L, Li H, Wang S, Chen S, Wang C, Yang X. MIRLET7BHG promotes hepatocellular carcinoma progression by activating hepatic stellate cells through exosomal SMO to trigger hedgehog pathway. Cell Death Dis. 2021;12(4):326.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  30. Han C, Qi Y, She Y, Zhang M, Xie H, Zhang J, Zhao Z, Peng C, Liu Y, Lin Y, et al. Long noncoding RNA SENCR facilitates the progression of acute myeloid leukemia through the miR-4731-5p/IRF2 pathway. Pathol Res Pract. 2023;245:154483.

    Article  CAS  PubMed  Google Scholar 

  31. Pilavaki P, Gahanbani Ardakani A, Gikas P, Constantinidou A. Osteosarcoma: current concepts and evolutions in Management principles. J Clin Med 2023, 12(8).

  32. Hu Z, Wen S, Huo Z, Wang Q, Zhao J, Wang Z, Chen Y, Zhang L, Zhou F, Guo Z et al. Current Status and Prospects of Targeted Therapy for Osteosarcoma. Cells: 2022, 11(21).

  33. Wu C, Gong S, Duan Y, Deng C, Kallendrusch S, Berninghausen L, Osterhoff G, Schopow N. A tumor microenvironment-based prognostic index for osteosarcoma. J Biomed Sci. 2023;30(1):23.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  34. Ni S, Hong J, Li W, Ye M, Li J. Construction of a cuproptosis-related lncRNA signature for predicting prognosis and immune landscape in osteosarcoma patients. Cancer Med. 2023;12(4):5009–24.

    Article  CAS  PubMed  Google Scholar 

  35. Qin S, Li L, Liu D. Metastasis-related gene signature associates with immunity and predicts prognosis accurately in patients with osteosarcoma. Aging. 2023;15(14):7219–36.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  36. Lv B, Wang Y, Ma D, Cheng W, Liu J, Yong T, Chen H, Wang C. Immunotherapy: reshape the Tumor Immune Microenvironment. Front Immunol. 2022;13:844142.

    Article  PubMed Central  PubMed  Google Scholar 

  37. Xiao Y, Yu D. Tumor microenvironment as a therapeutic target in cancer. Pharmacol Ther. 2021;221:107753.

    Article  CAS  PubMed  Google Scholar 

  38. Heymann MF, Lézot F, Heymann D. The contribution of immune infiltrates and the local microenvironment in the pathogenesis of osteosarcoma. Cell Immunol. 2019;343:103711.

    Article  PubMed  Google Scholar 

  39. Philip M, Schietinger A. CD8(+) T cell differentiation and dysfunction in cancer. Nat Rev Immunol. 2022;22(4):209–23.

    Article  CAS  PubMed  Google Scholar 

  40. Kurachi M. CD8(+) T cell exhaustion. Semin Immunopathol. 2019;41(3):327–37.

    Article  PubMed  Google Scholar 

  41. Hashimoto M, Kamphorst AO, Im SJ, Kissick HT, Pillai RN, Ramalingam SS, Araki K, Ahmed R. CD8 T cell exhaustion in chronic infection and Cancer: opportunities for interventions. Annu Rev Med. 2018;69:301–18.

    Article  CAS  PubMed  Google Scholar 

  42. Fritzsching B, Fellenberg J, Moskovszky L, Sápi Z, Krenacs T, Machado I, Poeschl J, Lehner B, Szendrõi M, Bosch AL, et al. CD8(+)/FOXP3(+)-ratio in osteosarcoma microenvironment separates survivors from non-survivors: a multicenter validated retrospective study. Oncoimmunology. 2015;4(3):e990800.

    Article  PubMed Central  PubMed  Google Scholar 

  43. Ribas A, Dummer R, Puzanov I, VanderWalde A, Andtbacka RHI, Michielin O, Olszanski AJ, Malvehy J, Cebon J, Fernandez E, et al. Oncolytic Virotherapy promotes Intratumoral T Cell Infiltration and improves Anti-PD-1 immunotherapy. Cell. 2017;170(6):1109–e11191110.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  44. Herbst RS, Soria JC, Kowanetz M, Fine GD, Hamid O, Gordon MS, Sosman JA, McDermott DF, Powderly JD, Gettinger SN, et al. Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients. Nature. 2014;515(7528):563–7.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  45. Guan L, Wu B, Li T, Beer LA, Sharma G, Li M, Lee CN, Liu S, Yang C, Huang L, et al. HRS phosphorylation drives immunosuppressive exosome secretion and restricts CD8(+) T-cell infiltration into tumors. Nat Commun. 2022;13(1):4078.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  46. Punzo F, Bellini G, Tortora C, Pinto DD, Argenziano M, Pota E, Paola AD, Martino MD, Rossi F. Mifamurtide and TAM-like macrophages: effect on proliferation, migration and differentiation of osteosarcoma cells. Oncotarget. 2020;11(7):687–98.

    Article  PubMed Central  PubMed  Google Scholar 

  47. Zhao CC, Han QJ, Ying HY, Gu XX, Yang N, Li LY, Zhang QZ. TNFSF15 facilitates differentiation and polarization of macrophages toward M1 phenotype to inhibit tumor growth. Oncoimmunology. 2022;11(1):2032918.

    Article  PubMed Central  PubMed  Google Scholar 

  48. Binnewies M, Pollack JL, Rudolph J, Dash S, Abushawish M, Lee T, Jahchan NS, Canaday P, Lu E, Norng M, et al. Targeting TREM2 on tumor-associated macrophages enhances immunotherapy. Cell Rep. 2021;37(3):109844.

    Article  CAS  PubMed  Google Scholar 

  49. Li H, Yang P, Wang J, Zhang J, Ma Q, Jiang Y, Wu Y, Han T, Xiang D. HLF regulates ferroptosis, development and chemoresistance of triple-negative breast cancer by activating tumor cell-macrophage crosstalk. J Hematol Oncol. 2022;15(1):2.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  50. Zhao Y, Zhang B, Zhang Q, Ma X, Feng H. Tumor-associated macrophages in osteosarcoma. J Zhejiang Univ Sci B. 2021;22(11):885–92.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  51. Dumars C, Ngyuen JM, Gaultier A, Lanel R, Corradini N, Gouin F, Heymann D, Heymann MF. Dysregulation of macrophage polarization is associated with the metastatic process in osteosarcoma. Oncotarget. 2016;7(48):78343–54.

    Article  PubMed Central  PubMed  Google Scholar 

  52. Shao XJ, Xiang SF, Chen YQ, Zhang N, Cao J, Zhu H, Yang B, Zhou Q, Ying MD, He QJ. Inhibition of M2-like macrophages by all-trans retinoic acid prevents cancer initiation and stemness in osteosarcoma cells. Acta Pharmacol Sin. 2019;40(10):1343–50.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  53. Yang D, Liu K, Fan L, Liang W, Xu T, Jiang W, Lu H, Jiang J, Wang C, Li G, et al. LncRNA RP11-361F15.2 promotes osteosarcoma tumorigenesis by inhibiting M2-Like polarization of tumor-associated macrophages of CPEB4. Cancer Lett. 2020;473:33–49.

    Article  CAS  PubMed  Google Scholar 

  54. Zhou Q, Xian M, Xiang S, Xiang D, Shao X, Wang J, Cao J, Yang X, Yang B, Ying M, et al. All-trans retinoic acid prevents Osteosarcoma Metastasis by inhibiting M2 polarization of Tumor-Associated macrophages. Cancer Immunol Res. 2017;5(7):547–59.

    Article  CAS  PubMed  Google Scholar 

  55. Mailankody S, Devlin SM, Landa J, Nath K, Diamonte C, Carstens EJ, Russo D, Auclair R, Fitzgerald L, Cadzin B, et al. GPRC5D-Targeted CAR T cells for Myeloma. N Engl J Med. 2022;387(13):1196–206.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  56. Duerinck J, Schwarze JK, Awada G, Tijtgat J, Vaeyens F, Bertels C, Geens W, Klein S, Seynaeve L, Cras L et al. Intracerebral administration of CTLA-4 and PD-1 immune checkpoint blocking monoclonal antibodies in patients with recurrent glioblastoma: a phase I clinical trial. J Immunother Cancer 2021, 9(6).

  57. Grimm MO, Schmitz-Dräger BJ, Zimmermann U, Grün CB, Baretton GB, Schmitz M, Foller S, Leucht K, Schostak M, Zengerling F, et al. Tailored Immunotherapy Approach with Nivolumab in Advanced Transitional Cell Carcinoma. J Clin Oncology: Official J Am Soc Clin Oncol. 2022;40(19):2128–37.

    Article  CAS  Google Scholar 

  58. Bagchi S, Yuan R, Engleman EG. Immune Checkpoint inhibitors for the treatment of Cancer: clinical impact and mechanisms of response and resistance. Annu Rev Pathol. 2021;16:223–49.

    Article  CAS  PubMed  Google Scholar 

  59. Li LQ, Zhang LH, Yuan YB, Lu XC, Zhang Y, Liu YK, Wen J, Khader MA, Liu T, Li JZ, et al. Signature based on metabolic-related gene pairs can predict overall survival of osteosarcoma patients. Cancer Med. 2021;10(13):4493–509.

    Article  PubMed Central  PubMed  Google Scholar 

  60. Yi C, Li Z, Zhao Q, Gong D, Zhao S, Chen Z, Cheng C, Bian E, Tian D. Single-cell RNA sequencing pro-angiogenic macrophage profiles reveal novel prognostic biomarkers and therapeutic targets for Osteosarcoma. Biochemical genetics 2023.

  61. Beck O, Paret C, Russo A, Burhenne J, Fresnais M, Steimel K, Seidmann L, Wagner DC, Vewinger N, Lehmann N et al. Safety and activity of the combination of Ceritinib and Dasatinib in Osteosarcoma. Cancers 2020, 12(4).

  62. Li X, Huang T, Jiang G, Gong W, Qian H, Zou C. Proteasome inhibitor MG132 enhances TRAIL-induced apoptosis and inhibits invasion of human osteosarcoma OS732 cells. Biochem Biophys Res Commun. 2013;439(2):179–86.

    Article  CAS  PubMed  Google Scholar 

  63. Brounais B, Chipoy C, Mori K, Charrier C, Battaglia S, Pilet P, Richards CD, Heymann D, Rédini F, Blanchard F. Oncostatin M induces bone loss and sensitizes rat osteosarcoma to the antitumor effect of midostaurin in vivo. Clin cancer Research: Official J Am Association Cancer Res. 2008;14(17):5400–9.

    Article  CAS  Google Scholar 

  64. Sampson ER, Martin BA, Morris AE, Xie C, Schwarz EM, O’Keefe RJ, Rosier RN. The orally bioavailable met inhibitor PF-2341066 inhibits osteosarcoma growth and osteolysis/matrix production in a xenograft model. J bone Mineral Research: Official J Am Soc Bone Mineral Res. 2011;26(6):1283–94.

    Article  CAS  Google Scholar 

  65. Vella S, Tavanti E, Hattinger CM, Fanelli M, Versteeg R, Koster J, Picci P, Serra M. Targeting CDKs with Roscovitine increases sensitivity to DNA damaging drugs of human osteosarcoma cells. PLoS ONE. 2016;11(11):e0166233.

    Article  PubMed Central  PubMed  Google Scholar 

  66. Nishimura H, Nakamura O, Yamagami Y, Mori M, Horie R, Fukuoka N, Yamamoto T. GSK-3 inhibitor inhibits cell proliferation and induces apoptosis in human osteosarcoma cells. Oncol Rep. 2016;35(4):2348–54.

    Article  CAS  PubMed  Google Scholar 

  67. Pettke A, Hotfilder M, Clemens D, Klco-Brosius S, Schaefer C, Potratz J, Dirksen U. Suberanilohydroxamic acid (vorinostat) synergistically enhances the cytotoxicity of doxorubicin and cisplatin in osteosarcoma cell lines. Anticancer Drugs. 2016;27(10):1001–10.

    Article  CAS  PubMed  Google Scholar 

  68. Kuijjer ML, Peterse EF, van den Akker BE, Briaire-de Bruijn IH, Serra M, Meza-Zepeda LA, Myklebost O, Hassan AB, Hogendoorn PC. Cleton-Jansen AM: IR/IGF1R signaling as potential target for treatment of high-grade osteosarcoma. BMC Cancer. 2013;13:245.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  69. Pagliaro L, Marchesini M, Roti G. Targeting oncogenic notch signaling with SERCA inhibitors. J Hematol Oncol. 2021;14(1):8.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  70. Zheng H, Li BH, Liu C, Jia L, Liu FT. Comprehensive Analysis of lncRNA-Mediated ceRNA Crosstalk and Identification of Prognostic Biomarkers in Wilms’ Tumor. BioMed research international 2020, 2020:4951692.

  71. Qin F, Xu H, Wei G, Ji Y, Yu J, Hu C, Yuan C, Ma Y, Qian J, Li L, et al. A prognostic model based on the Immune-related lncRNAs in Colorectal Cancer. Front Genet. 2021;12:658736.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  72. Ji Z, Shen J, Lan Y, Yi Q, Liu H. Targeting signaling pathways in osteosarcoma: mechanisms and clinical studies. MedComm. 2023;4(4):e308.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  73. Keremu A, Maimaiti X, Aimaiti A, Yushan M, Alike Y, Yilihamu Y, Yusufu A. NRSN2 promotes osteosarcoma cell proliferation and growth through PI3K/Akt/MTOR and Wnt/β-catenin signaling. Am J cancer Res. 2017;7(3):565–73.

    CAS  PubMed Central  PubMed  Google Scholar 

  74. Qiu C, Su W, Shen N, Qi X, Wu X, Wang K, Li L, Guo Z, Tao H, Wang G, et al. MNAT1 promotes proliferation and the chemo-resistance of osteosarcoma cell to cisplatin through regulating PI3K/Akt/mTOR pathway. BMC Cancer. 2020;20(1):1187.

    Article  CAS  PubMed Central  PubMed  Google Scholar 

  75. Li H, Shen X, Ma M, Liu W, Yang W, Wang P, Cai Z, Mi R, Lu Y, Zhuang J, et al. ZIP10 drives osteosarcoma proliferation and chemoresistance through ITGA10-mediated activation of the PI3K/AKT pathway. J Experimental Clin cancer Research: CR. 2021;40(1):340.

    Article  PubMed Central  Google Scholar 

  76. Yang B, Li L, Tong G, Zeng Z, Tan J, Su Z, Liu Z, Lin J, Gao W, Chen J, et al. Circular RNA circ_001422 promotes the progression and metastasis of osteosarcoma via the miR-195-5p/FGF2/PI3K/Akt axis. J Experimental Clin cancer Research: CR. 2021;40(1):235.

    Article  CAS  PubMed Central  Google Scholar 

Download references

Funding

This work was supported by the Research fund of Anhui Institute of Translation Medicine (Grant No. 2021zhyx-C49), Foundation of Anhui Medical University (Grant No. 2021xkj030 and 2021xkj041), and Nature Science Foundation of Anhui Province (Grant No. 2208085MH255).

Author information

Author notes

  1. Shibing Zhao, Dasheng Tian, Fei Huang contributed equally to this work.

Authors and Affiliations

  1. Department of Orthopaedics, The Second Affiliated Hospital of Anhui Medical University, Hefei, 230601, China

    Shibing Zhao, Dasheng Tian, Fei Huang, Jinhao Cheng, Zhitao He, Qitian Shen, Shuai Liang, Deliang Gong, Jun Liu, Chengfeng Yi, Chun Zhang, Erbao Bian, Juehua Jing & Tao Wang

  2. Institute of Orthopaedics, Research Center for Translational Medicine, The Second Affiliated Hospital of Anhui Medical University, Hefei, 230601, China

    Shibing Zhao, Dasheng Tian, Fei Huang, Jinhao Cheng, Zhitao He, Qitian Shen, Shuai Liang, Deliang Gong, Jun Liu, Chengfeng Yi, Chun Zhang, Erbao Bian, Juehua Jing & Tao Wang

  3. Department of Orthopaedics, Division of Life Sciences and Medicine, The First Affiliated Hospital of USTC, University of Science and Technology of China, Hefei, 230601, China

    Lei Wang

  4. Department of Bone and Soft Tissue Surgery, Anhui Provincial Cancer Hospital, Hefei, 230601, China

    Lei Wang

Authors
  1. Shibing Zhao
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Dasheng Tian
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Fei Huang
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Lei Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Jinhao Cheng
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Zhitao He
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Qitian Shen
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Shuai Liang
    View author publications

    You can also search for this author inPubMed Google Scholar

  9. Deliang Gong
    View author publications

    You can also search for this author inPubMed Google Scholar

  10. Jun Liu
    View author publications

    You can also search for this author inPubMed Google Scholar

  11. Chengfeng Yi
    View author publications

    You can also search for this author inPubMed Google Scholar

  12. Chun Zhang
    View author publications

    You can also search for this author inPubMed Google Scholar

  13. Erbao Bian
    View author publications

    You can also search for this author inPubMed Google Scholar

  14. Juehua Jing
    View author publications

    You can also search for this author inPubMed Google Scholar

  15. Tao Wang
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

TW, JJ, EB, SZ and DT conceived and designed the experiments; SZ, DT, FH, JC and CY analyzed the data and performed the experiments; QS, DG, JL, ZH and CZ screened and checked the data; SZ, TW, LW and SL wrote the paper. All authors reviewed the manuscript.

Corresponding authors

Correspondence to Erbao Bian, Juehua Jing or Tao Wang.

Ethics declarations

Ethical approval

The animal experiment strictly followed the ARRIVE guidelines and approved by the Laboratory Animal Ethics Committee of Anhui Medical University. The ethical approval number is LLSC20241416.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1: Appendix T1 251 ESCRT-related genes.

Supplementary Material 2: Appendix T2 187 ESCRT- related long noncoding RNAs.

Supplementary Material 3: Appendix T3 88 clinical patient data samples from the TARGET dataset.

Supplementary Material 4: Appendix T4 90 clinical patient data samples from GEO dataset.

12935_2025_3659_MOESM5_ESM.jpg

Supplementary Material 5: Fig. 1 (A) The distribution of risk scores and survival statuses and 7 ERG expression profiles. (B) Heatmap of differences in the expression of 7 ERGs in the two risk groups.

12935_2025_3659_MOESM6_ESM.jpg

Supplementary Material 6: Fig. 2 Validation of the ERG prognostic signature in the GEO cohort. (A) K‒M curves based on the ERG signature. (B) ROC curve analyses for 1, 3 and 5 years. (C-D) Risk score distribution and survival status of OS patients. (E) ​Heatmap of the expression of 7 ERGs in the two risk groups.

12935_2025_3659_MOESM7_ESM.jpg

Supplementary Material 7: Fig. 3 ERL and ERG interaction network. Squares represent ERLs and circles represent ERGs. Blue denotes a negative correlation, whereas red denotes a positive correlation.

12935_2025_3659_MOESM8_ESM.jpg

Supplementary Material 8: Fig. 4 (A-B) Risk score distribution and survival status of OS patients based on the ERL signature. (C) Heatmap of the expression of the 10 ERLs in the two risk groups.

12935_2025_3659_MOESM9_ESM.jpg

Supplementary Material 9: Fig. 5(A-H) Stratified analysis to explore the prognostic value of the ERL signature in different clinical features (age, sex, metastatic potential, and tumor primary site).

12935_2025_3659_MOESM10_ESM.jpg

Supplementary Material 10: Fig. 6 Functional enrichment analyses. (A) The GO enrichment analysis. (B) ​The KEGG pathway analysis. (C-F) The GSEA analysis.

12935_2025_3659_MOESM11_ESM.jpg

Supplementary Material 11: Fig. 7 (A-J) The K‒M curves for patients grouped based on the expression level of each of the 10 ERLs.

Supplementary Material 12: Fig. 8 Relative expression of LINC01549 in different OS cell lines.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Zhao, S., Tian, D., Huang, F. et al. Comprehensive analysis of ESCRT transcriptome-associated signatures and identification of the regulatory role of LMO7-AS1 in osteosarcoma. Cancer Cell Int 25, 29 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12935-025-03659-4

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12935-025-03659-4

Keywords

Cancer Cell International

ISSN: 1475-2867

Contact us