Introduction
Osteosarcoma is the predominant type of malignant bone tumor, affecting mostly children and adolescents [
1]. The advancement of surgical resection with adequate surgical margins and neoadjuvant chemotherapy has increased the survival rate from less than 20% in 1970s to 60–70% nowadays [
2]. However, clinically detectable metastases are found at diagnosis among approximately 15–20% of patients with osteosarcoma [
3]. The most common metastatic disease occurs in lungs, whereas the second most common site of metastasis is bone. The survival of patients with unresectable, metastatic or relapsed osteosarcoma are still unsatisfactory, with an overall 5-year survival rate of less than 20% [
4]. Current standard treatment seems to be ineffective against these recurrent or metastatic osteosarcomas [
5]. Therefore, there is an urgent demand in identification of therapeutic targets for advanced osteosarcoma patients.
Recently, high-throughput screening projects such as DepMap based on RNA interference silencing and CRISPR-Cas9 knockout techniques have emerged to be a powerful approach to identify potential dependency genes that are crucial to tumor survival, metastasis or recurrence [
6‐
8]. These identified tumor vulnerabilities could be potential therapeutic targets. CERES algorithm was developed to calculate gene-knockout effects. CERES scores represent the median effects of vital genes and nonessential genes for per cell line [
9]. Genes that are only essential in a few cell lines are recognized as better drug targets because it is unlikely to cause toxicity in normal tissues after inhibiting their function. Furthermore, investigation and validation of prognosis-predictive value of these cancer essential genes may allow better clinical decision making for orthopedic surgeons.
In this study, we aimed to identify candidate essential genes that are essential to the survival of osteosarcoma, by screening DepMap database, functional enrichment and LASSO analyses. Then, a predictive model based on these essential genes for overall survival was constructed. Finally, cell proliferation was further evaluated in response to knocking down the candidate gene to determine its oncogenic potential.
Methods
Identification of genes essential for survival of osteosarcoma cells through DepMap database
To identify potential cancer therapeutic targets, the Broad Institute constructed the DepMap (Cancer Dependency Map) database, which contains gene dependency information of more than 700 human tumor cell lines of different tissue origins and gene expression, gene copy number and gene mutation information of more than 1,000 tumor cells (
https://depmap.org/portal/). Through the DepMap database, we can know the gene dependence of different cell lines. Genes with amplified gene copy number can cause severe DNA damage during CRISPR-Cas9 cleavage and cause cell growth arrest or apoptosis, which can lead to false positives. Therefore, in the process of using CRISPR-Cas9 technology to evaluate the dependence of cells on genes, the DepMap database comprehensively considered the two factors of gene copy number and sgRNA loss, and set a new parameter CERES as a parameter to measure the degree of gene necessity. The principle of CERES is that the fewer cells that carry the gene's sgRNA survive, the fewer copies of the gene are in the cell, and the more the cells are dependent on the gene. A negative CERES score indicates that knocking out the gene inhibits the survival and proliferation of the cell lines. In the study, we defined genes as an essential gene with CERES scores < − 1 in more than 75% of osteosarcoma cell lines [
10].
Identification of differentially expressed genes (DEGs)
Limma is a differential expression screening method based on generalized linear models. In this study, the R software package limma (version 3.40.6) was used for differential analysis to obtain the differential genes between the osteosarcoma group and the normal control group. Genes with a fold-change (FC) > 1.5 or < 0.67 (corresponds to |log2FC| ≥ 0.585) and p-value < 0.05 were considered significantly DEGs.
Clinical data collection and extraction
MRNA expression profiles of GSE19276, GSE21257 and GSE39055 were downloaded from the GEO database (
http://www.ncbi.nlm.nih.gov/geo/). The dataset for GSE19276 includes 44 osteosarcoma samples and 5 normal tissue samples. GSE21257 and GSE39055 incorporate 53 patients and 37 patients with follow-up information, respectively.
Patients and specimens
A total of 18 osteosarcoma specimens and 18 paired normal specimens from Fujian Medical University Union Hospital between August 2019 and August 2021 were collected. The project was approved by the Research Ethics Committee of Fujian Medical University Union Hospital. All patients provided written informed consent under an institutionally approved protocol.
GSEA analysis
We obtained the GSEA software (version 3.0) from the GSEA (
http://software.broadinstitute.org/gsea/index.jsp) website, and divided the samples into high expression groups (50%) and low expression groups according to the expression level of LARS group (50%) to assess relevant pathways and molecular mechanisms.
P value of < 0.05 was considered statistically significant.
ROC analysis
We performed ROC analysis using the R software package pROC (version 1.17.0.1) to obtain the AUC. Specifically, we obtained the patient's follow-up time and risk score, and used the roc function of pROC to perform the ROC analysis of 1-year, 3-year, and 5-year.
Gene enrichment analysis
The DAVID (Database for Annotation, Visualization and Integrated Discovery) database can provide biological meaning for genes (
https://david.abcc.Ncifcrf.gov/).In the research, KEGG-pathway and GO-BP enrichment analysis of identified DEGs was performed using the DAVID database.
Protein–protein interaction (PPI) network analysis of DEGs
PPI analysis of DEGs was analyzed using Metascape and the node score was calculated and obtained in Metascape (
http://metascape.org). MCODE (Molecular Complex Detection) algorithm was performed to identify the densely connected network neighborhoods. Each MCODE component is marked with a different color and characterizes their biological significance.
Least absolute shrinkage and selection operator (LASSO) analysis
In this study, we used the R software package glmnet to integrate survival time, survival status, and gene expression data for regression analysis using the lasso-cox method.
Drug sensitivity evaluation
GSCALite is a website used for the analysi of drug sensitivity and genomic cancer (
http://bioinfo.life.hust.edu.cn/web/GSCALite/). In the research, GSCALite was used to evaluate the drug sensitivity of LARS and DNAJC17 to identify potential compounds for treatment.
Cells culture and transfection
The cell lines Saos-2 and U-2 OS were obtained from ATCC (American Type Culture Collection). Saos-2 and U-2 OS cells were cultured in DMEM medium (Gibco) containing 10% fetal bovine serum (FBS, BI, Kibbutz Beit Haemek, Israel), 100 U/mL penicillin and 0.1 mg/mL streptomycin (BBI life sciences, shanghai, China) andat 37 °C in a humidified incubator with 5% CO2. The sequence of shRNAs targeting LARS was cloned into pLVX vector. The sequence of LARS shRNA1 was 5′-CTGGACATCACTTGTTTCT-3′; The sequence of LARS shRNA2 was 5′-AAATGAAGGCGTCCATTCA-3′. The transfection was performed using lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’ s guidelines.
RNA isolation and RT-qPCR
Total RNA was extracted from Saos-2 and U-2 OS cells using TRIzol (Invitrogen, CA, USA), and was reverse transcripted with mRNA reverse transcription kit (Takara, Japan). Specific primers for RT-qPCR were performed to detect the mRNA expression of LARS. All primers were synthesized by Sangon Biotech (Shanghai, China). LARS-forward primer 5′-ATGGCGGAAAGAAAAGGAACAG-3′ and LARS-reverse primer 5′-CAGGCCAAAGGGAAACAGACAAC-3′. GAPDH primers were as follows; GAPDH-forward primer 5′-GCGGGGCTCTCCAGAACATCAT-3′ and GAPDH-reverse primer 5′-CCAGCCCCAGCGTCAAAGGTG-3′. Relative quantification was performed by the 2-ΔΔCt method.
Western blot
Cells were lysed in RIPA lysis buffer (Roche Ltd, Basel, Switzerland) containing protease inhibitors. Protein concentration in the lysates was measured by Micro BCA Protein Assay Kit (Pierce Biotechnology, IL, US). Samples were separated on 10% SDS-PAGE and subsequently transferred to Amersham Protran nitrocellulose membranes (GE Healthcare Life Sciences, Fairfield, USA). The nitrocellulose membranes were then incubated with primary antibodies for the target proteins LARS and GAPDH (Proteintech, Wuhan, China) at a dilution of 1:1000 and 1:5000 for 2 h, respectively. The proteins were detected and quantified using the Odyssey® CLx Infrared Imaging System (LI-COR Biosciences, NE, USA).
Immunohistochemistry (IHC) staining analysis
We conducted IHC staining analysis to measure the protein expression of LARS in OS tissues and adjacent normal tissues according to the standard immunoperoxidase staining procedure. Slides were incubated with anti-LARS (21146-1-AP, Proteintech, Wuhan, China, diluted 1:400). The IHC staining scores of LARS were evaluated by two independent pathologists. The percentage of stained positive cells was calculated from 1 to 4: 1, 0–25%; 2, 26–50%; 3, 51–75%; and 4, 75–100%. The staining intensity score was ranged from 0 to 3: 0, no staining; 1, weak staining; 2, moderate staining; and 3, strong staining. The percentage of positive tumor cells and the staining intensity were multiplied to produce a weighted score for each case.
CCK-8 assay
Saos-2 and U-2 OS cells were seeded into 96-well plates (2 × 104 cells/well) and cultured for 24 h, 48 h, and 72 h. Four hours before absorbance measuring, 10 μL CCK-8 solution was added. The absorbance was measured at 450 nm with a microplate reader after incubating at 37 °C for 2 h.
Statistical analysis
In the study, a statistical correlation was calculated by t-test. Overall survival (OS) was evaluated by the Kaplan–Meier method, and survival curves were compared by log-rank test. Two-way ANOVA was used to analyze CCK8 assay. All p values < 0.05 were considered statistically significant.
Discussion
Osteosarcoma is the most common primary malignant bone tumor, and high morbidity rate and high recurrence rate are the clinical characteristics of osteosarcoma [
11]. Osteosarcoma seriously affects the health of adolescents and children [
12]. Although great progress has been made in the clinical diagnosis and surgical treatment of osteosarcoma, the overall prognosis of patients with osteosarcoma is still poor. Therefore, it is urgent to find effective molecular therapeutic targets.
Cancer therapeutic target identification has focused on oncogenes and tumor suppressor genes that are mutated in cancer. Mutated oncogenes and tumor suppressor genes that cause cancer also confer properties on cancer cells that differ from normal cells, making them selectively dependent on certain genes for growth or survival. Such genes, which we call selective essential genes, have good specificity as therapeutic targets and are ideal potential therapeutic targets. Therefore, the systematic identification of these selectively essential genes is a novel strategy to identify cancer therapeutic targets. A well-established method for identifying essential genes is to use CRISPER-Cas9 technology to perform genome-wide gene function inactivation screening in cell lines. In recent years, Broad and Anger have performed genome-wide screening of CRISPER-Cas9 in more than 700 cancer cells of different tissue origins and identified essential genes in these cells. Using these data, they constructed the DepMap database, which aims to systematically identify the genetic dependencies of cancer cells.
In the study, the dependence score was calculated by CERES and identified 785 genes essential to the proliferation and survival of osteosarcoma cells from the DepMap website. Of the 785 genes, 59 DEGs were identified in osteosarcoma tissues compared with normal tissues through the GSE19276 database. The results showed that these 59 essential genes were mainly associated with the cell cycle-related signaling pathway in the osteosarcoma tissues. Furthermore, we established a gene signature (LARS and DNAJC17) screened from these 59 genes, and this signature could divide osteosarcoma patients into the low-risk and high-risk groups.
LARS and DNAJC17 are highly expressed in osteosarcoma. High expression of LARS is an unfavorable prognostic factor in patients with osteosarcoma. However, high expression of DNAJC17 is a favorable prognostic factor in patients with osteosarcoma. Consequently, we were interested in the influence of LARS on osteosarcoma.
LARS encodes a cytosolic leucine-tRNA synthetase, a member of the class ARSs (aminoacyl-tRNA synthetase) family. The ARSs family is an extremely ancient enzyme in evolution. Its classic function is to catalyze the esterification reaction between amino acids and their corresponding tRNAs to generate aminoacyl-tRNA, which provides raw materials for protein synthesis in organisms and participates in the production of genetic coding process. The 20 ARSs synthetases correspond to 20 amino acids, and their precise identification of the corresponding tRNA and amino acid substrates ensures the precise transmission of genetic information from mRNA to protein. In the past decades, the identification of cancer-related factors has been an important issue in the field of oncology, not only to understand the basic mechanisms of tumor formation, but also to discover related therapeutic targets. However, ARSs have been neglected, mainly because many people think that ARSs are merely housekeeping genes involved in protein synthesis. Mammalian ARSs have evolved over the course of evolution to develop many additional domains that are not necessarily associated with their catalytic domains. Thanks to these domains, ARSs are able to interact with regulatory factors. ARSs participate in signaling pathways by forming complexes with other regulatory factors [
13]. Abnormal expression and mutation of ARSs induce abnormal cellular regulation and protein synthesis. The abnormal function of ARSs is associated with a variety of human diseases, such as autoimmune disorders, metabolic disorders, neuronal diseases, and cancer [
14]. Some ARSs are abnormally up-regulated or down-regulated in a variety of tumors [
15‐
24].
Recently, more and more evidences have shown that a variety of ARSs are responsible for non-canonical functions other than enzymatic catalytic activity, such as translation regulation, transcription regulation, synthesis, signal transduction, angiogenesis, inflammation, apoptosis, and the development of cancer [
25‐
29]. The non-classical function of LARS was first reported in 2012. Downregulation of cellular endogenous LARS inhibits mTORC1 activity, thereby affecting cell proliferation [
30]. Shin et al. found that the LARS gene is overexpressed in lung cancer cells and tissues, and knockdown of LARS expression can inhibit the growth and metastasis of lung cancer cells. Which suggest that LARS may have carcinogenic potential [
31]. However, the expression of LARS was repressed during mammary cell transformation and in human breast cancer. Monoallelic genetic deletion of LARS in mouse mammary glands enhanced breast cancer tumor formation and proliferation. Repression of LARS lead to impaired leucine codon-dependent translation of growth suppressive genes, including gamma-glutamyltransferase 5 (GGT5) and epithelial membrane protein 3 (EMP3). which in turn enhances breast tumor formation and growth [
32]. These results indicated that the function of LARS is complex, as it appears to have different functions in different tumors.
In the study, we knocked down LARS gene expression by shRNA. Knockdown of LARS expression inhibited the proliferative ability of osteosarcoma cells, suggesting that LARS has oncogenic potential in osteosarcoma. The correlation between LARS expression and immune cells suggests that LARS regulates osteosarcoma immunity through multiple immune cell populations, such as Monocytic lineage, T cells, and Fibroblasts. The analysis of drug sensitivity revealed that high LARS expression was associated with higher drug sensitivity to BRD-K30748066, Tozasertib, BI-2536, GSK461364, BRD-K70511574, Triazolothiadiazine, Rigosertib, FQI-1, KX2-391 and SB-225002.
Our findings may contribute to further understanding the molecular mechanism and improving the clinical diagnosis and treatment for osteosarcoma. However, limitations of this study still exist including lack of a large cohort of samples from patients with osteosarcoma and short of survival analysis. In addition, further experimental studies are essential for further verifying the mechanisms of LARS in osteosarcoma.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.