Common mechanism of pathogenesis in various types of metastatic osteosarcoma
- Authors:
- Published online on: September 15, 2017 https://doi.org/10.3892/ol.2017.6955
- Pages: 6307-6313
Abstract
Introduction
Osteosarcoma (OS), which is one of the common primary bone tumors in children and adolescents is highly aggressive and metastatic (1); it comprises ~5% of all pediatric tumors and patients with pulmonary metastasis present a poor prognosis (2). There are definite differences in the mechanism between primary and metastatic OS, and numerous previous studies have been performed to expound them (3–5).
Multiple signaling pathways, including the Notch signaling pathway, are involved in cancer migration initiation through the extracellular matrix (ECM) towards the bloodstream (6). Survival of cancer cells in the bloodstream is mediated by the integrin signaling pathway to activate the phosphatidylinositol 3-kinase/protein kinase B survival pathway (7). Signaling pathways, including those of Src and Wnt, are involved in apoptosis resistance throughout metastasis (8,9). In addition, the immune system is also invaded and modulated for survival advantage (10). Subsequently, cell adhesion is arrested, and the cells extravasate to their new environment and re-establish cell-cell adhesions (11). There are numerous various viewpoints regarding the mechanism underlying metastatic tumor cell arrest to distant organ sites (12,13). Angiogenesis is necessary for cancer cell proliferation and metastasis (12); however, there remain a number of unknown mechanisms, including the precise mechanism underlying how cells alternate between exhibiting primary and metastatic properties.
Therefore, in the present study, gene expression profiles were used to further investigate the metastatic mechanism of OS. The GSE49003 dataset considered in the present study was previously analyzed by Diao et al (14) to identify differentially expressed genes (DEGs) between metastatic and non-metastatic patients with OS, and crucial microRNAs associated with OS metastasis, by merging data from different metastatic or non-metastatic OS cell lines. However, different types of metastatic OS may be regulated by different molecular mechanisms. In addition, transcription factors (TFs) may also serve a vital role in this pathomechanism. Therefore, the dataset was reanalyzed in the present study to emphasize the different mechanisms of different metastatic OS cell lines.
Materials and methods
Gene expression profile data
The raw expression data (dataset GSE49003; http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE49003), as provided by Endo-Munoz et al on July 18, 2013, were used in the present study. The microarray expression profile was obtained from two metastatic OS cell lines and two non-metastatic OS cell lines. The metastatic KHOS and KRIB cell lines and the non-metastatic HOS cell line were used with three-duplicated samples, and gene expression data from each of these cell lines were used. The platform of this dataset was GPL6847 Illumina HumanHT-12 V3.0 expression beadchip (Illumina Inc., San Diego, CA, USA).
Data preprocessing and DEG screening
The Limma package (15) in Bioconductor was used to add probe annotation files for each Illumina chip in order to preprocess the expression profile. Background correction, quantile normalization and probe summarization were performed to generate the gene expression data matrix.
DEGs between KHOS and HOS and between KRIB and HOS were determined using the Limma package. The differential expression of genes were analyzed by Student's t-test, and those with a false discovery rate adjusted P-value of <0.01 and |fold change| ≥2 were screened.
Functional enrichment analysis of DEGs
The Gene Ontology (GO) (16) project was established for gene classifications by molecular function, biological process (BP) and cellular component. DEGs of KHOS vs. HOS and KRIB vs. HOS were functionally enriched by GO. In addition, the Kyoto Encyclopedia of Genes and Genomes (KEGG) (17), which has a become a major database resource for understanding high-level functions of genes, was utilized. The default threshold of P<0.01 was selected for the hypergeometric enrichment test.
Protein-protein interaction (PPI) network construction and subnetwork mining
The Search Tool for the Retrieval of Interacting Genes/Proteins is a biological database that provides known and predicted PPIs (13). The tool was applied in the present study to identify interacting protein pairs between DEGs with a PPI score of 0.9. Subsequently, Cytoscape software version 2.8.0 (18) was used to visualize the constructed PPI network. Subnetworks (modules) with a hypergeometric P-value <0.05 were identified by the ClusterONE plugin (19) from Cytoscape. Furthermore, the Database for Annotation, Visualization and Integrated Discovery (20,21) was utilized to perform the KEGG pathway cluster analyses of DEGs in modules with P<0.05.
Overlapping gene analysis
Overlapping DEGs that were upregulated in KHOS and KRIB cells compared with HOS cells were identified, and downregulated DEGs that were common in KHOS and KRIB cells were identified. Thereafter, KEGG signaling pathways of these two types of overlapping genes were enriched.
Furthermore, based on the regulatory association between TFs and target genes recorded in the University of California Santa Cruz (UCSC) (22) database, regulatory associations between TFs and their target DEGs were identified.
Results
DEGs of various groups
A total of 1,552 (711 downregulated and 841 upregulated) and 1,330 DEGs (570 downregulated and 760 upregulated) were obtained from the KHOS vs. HOS and KRIB vs. HOS comparisons, respectively.
Functional enrichment analyses of DEGs
Significant enriched terms of GO and KEGG pathway enrichment analyses in KHOS and KRIB groups are presented in Tables I and II, respectively. Upregulated genes of KHOS were associated with GO-BP terms of anatomical structure morphogenesis and cellular response to extracellular stimulus, and the downregulated genes were enriched in BP terms of multicellular organismal development and anatomical structure morphogenesis. Using KEGG, focal adhesion, axon guidance and peroxisome proliferator-activated receptor signaling pathways were significantly enriched by upregulated genes of KHOS, and pathways in cancer, tight junctions and cell adhesion molecules were enriched by downregulated DEGs in KHOS cells (Table I). GO and pathway enrichment analyses of KRIB were similar to those of KHOS, and the significantly enriched GO term of KRIB overexpressed genes was anatomical structure morphogenesis, and the significantly enriched pathway was focal adhesion (Table II).
PPI network and subnetwork
A total of 856 pairs of interacting proteins and 495 nodes were noted in the PPI network of KHOS (Fig. 1A). The top 6 nodes with a greater degree of connectivity compared with the others were jun proto-oncogene (JUN; degree=33), vascular endothelial growth factor A (VEGFA; degree=29), epidermal growth factor receptor (EGFR; degree=23), cyclin-dependent kinase inhibitor 1A (degree=19), interleukin 6 (IL-6; degree=19) and signal transducer and activator of transcription 3 (STAT3; degree=17). Furthermore, two modules with P<0.05 were obtained (Fig. 1B and C). A total of 9 genes, including 7 collagen molecules and 2 integrin family members involved in module 1 were significantly enriched in terms of ECM-receptor interaction and focal adhesion.
There were 719 interacting pairs and 397 nodes of the PPI network of KRIB (Fig. 2). VEGFA (degree=29), Fyn oncogene related to Src, FGR, Yes (degree=25), JUN (degree=23), EGFR (degree=22) and IL-6 (degree=19), were the top 5 hub nodes. In addition, one module was screened (Fig. 2), and the significant pathways enriched by its genes were ECM-receptor interaction and focal adhesion.
KEGG pathway enrichment and transcriptional regulatory network analysis of overlapping DEGs
A total of 421 upregulated and 595 downregulated overlapping genes were obtained. Pathway of focal adhesion, glutathione metabolism and endocytosis were enriched by their upregulated genes, whereas pathways in cancer, focal adhesion and cell adhesion molecules were enriched by downregulated genes. In addition, the mitogen-activated protein kinase (MAPK) and transforming growth factor-β (TGF-β) signaling pathways were enriched by overlapping genes with decreased expression levels (Table III).
Fig. 3 demonstrates the transactional regulatory network of overlapping DEGs. A total of 7 TFs, including activating transcription factor 3, forkhead box (FOX)A1, STAT3, FOXA2, hes-related family bHLH transcription factor with YRPW motif 1 (HEY1), SUZ12 polycomb repressive complex 2 subunit and zinc finger ZZ-type containing 3 were identified. The first three of these were downregulated and the others were overexpressed.
Discussion
Due to the high risk for recurrence and the poor survival of metastatic OS patients (23), two different metastatic OS cell lines and one non-metastatic cell lines were used to investigate the mechanism underlying metastasis. The GO and pathway enrichment analysis results were similar of these two metastatic types of OS. Besides, a large number of overlapping genes, including 7 TFs, were obtained. These genes and biological functions may be important in OS tumor metastasis.
The most prominent node in the PPI networks of KHOS and KRIB was VEGFA, which is a growth factor that is activated in vasculogenesis, angiogenesis and endothelial cell growth (24). It is not only a commonly upregulated angiogenic factor during tumor growth, but also the most frequent lymphangiogenic factor contributing to the lymphatic metastasis (25). A previous study suggested that VEGFA promoted tumor growth and metastasis via the VEGF-VEGFR1 signaling pathway (26). In addition, EGFR, which is a member of the EGF family, was another common hub node with a high degree (27). It was revealed that EGFR participated in the promotion of cell division, migration and angiogenesis, and inhibited cell apoptosis (28). Furthermore, these two genes were enriched in the development and differentiation process, including anatomical structure morphogenesis and vasculature development. Yang et al (29) demonstrated that Twist, which is a master regulator of embryonic morphogenesis, contributed to metastasis by promoting an epithelial-mesenchymal transition (29). Therefore, tumor metastasis is the recapitulation of morphogenetic processes (30). Morphogenetic factors, including hepatocyte growth factor/scatter factor have metastasis-promoting effects on breast carcinoma cells (31). Based on these studies, VEGFA and EGFR may be factors involved in the morphogenetic processes to promote tumor metastasis.
The KEGG pathway enrichment analysis results of module 1 of the KHOS group was similar to that of module 1 of the KRIB group. ECM-receptor interaction and focal adhesion were the observed pathways enriched by multiple collagen genes and integrin genes. These results were in accordance with previous studies (32,33) and demonstrated the accuracy and practicability of them. Additionally, functional enrichment analysis of overlapping genes of these two group would be more direct to exhibit commonality of metastatic OS.
Focal adhesion was the significant pathway enriched by upregulated and downregulated overlapping genes. Furthermore, these downregulated genes remained enriched in signaling pathways that included the MAPK and TGF-β signaling pathways. The former pathway serves an important role in the early metastasis of the tumor (34,35), whereas the latter has positive and negative effects on carcinogenesis (36), and further investigation for is required to fully elucidate the molecular mechanism.
Finally, a total of 7 TFs, which regulated multiple genes, were screened. Among them, FOXA1, STAT3 and HEY1 were more important in the regulatory network. Candy et al (37) indicated that overexpressed HEY1 was a negative prognostic factor for colorectal cancer, and it is closely associated with vascular and perineural invasion and lymph node metastasis. Therefore, the upregulated HEY1 may be associated with OS metastasis. In addition, FOXA1 was revealed to be a participant in embryonic development, mediating tissue-specific gene expression in differentiated tissues (38). This gene encoded protein may be another important factor involved in the similar process of morphogenesis.
In conclusion, recapitulation of morphogenetic processes, a disordered signaling system, and activated cell adhesion and migration were involved in the facilitated tumor metastasis. Furthermore, morphogenesis and vasculature development induced angiogenesis, and tumor growth and adhesion, or even the lymph node metastasis induced the metastatic OS. The present study further illustrated the OS tumorigenesis mechanism and displayed the crucial roles of genes such as VEGFA, EGFR and HEY1. The practical clinical significance has not yet been clearly described and requires further investigation.
References
Wu J, Hsieh MY, Wang Y, Roberts E, Vallejo D, Lee A and Ermel R: Antitumor properties of ouabain in lipid double emulsion in orthotopic canine osteosarcoma xenografted mouse model. Cancer Res. 74:54122014. View Article : Google Scholar | |
Roos A, Yustein JT and Donehower L: Oncogenic role of Runx2 in the development of osteosarcoma. Cancer Res. 73:5036. 2013. View Article : Google Scholar | |
Pompetti F, Rizzo P, Simon RM, Freidlin B, Mew DJ, Pass HI, Picci P, Levine AS and Carbone M: Oncogene alterations in primary, recurrent, and metastatic human bone tumors. J Cell Biochem. 63:37–50. 1996. View Article : Google Scholar : PubMed/NCBI | |
Oda Y, Naka T, Takeshita M, Iwamoto Y and Tsuneyoshi M: Comparison of histological changes and changes in nm23 and c-MET expression between primary and metastatic sites in osteosarcoma: A clinicopathologic and immunohistochemical study. Human Pathol. 31:709–716. 2000. View Article : Google Scholar | |
Liang S, Ren Z, Han X, Yang J, Shan L, Li L, Wang B, Zhang Q, Mu T, Chen K, et al: PLA2G16 expression in human osteosarcoma is associated with pulmonary metastasis and poor prognosis. PLoS One. 10:e01272362015. View Article : Google Scholar : PubMed/NCBI | |
Hughes DP: How the NOTCH pathway contributes to the ability of osteosarcoma cells to metastasizePediatric and Adolescent Osteosarcoma. Jaffe N, Bruland OS and Bielack S: 152. Springer; New York, NY: pp. 479–496. 2010, View Article : Google Scholar | |
Casar B, Rimann I, Kato H, Shattil SJ, Quigley JP and Deryugina EI: In vivo cleaved CDCP1 promotes early tumor dissemination via complexing with activated β1 integrin and induction of FAK/PI3K/Akt motility signaling. Oncogene. 33:255–268. 2012. View Article : Google Scholar : PubMed/NCBI | |
Kim LC, Song L and Haura EB: Src kinases as therapeutic targets for cancer. Nat Rev Clin Oncol. 6:587–595. 2009. View Article : Google Scholar : PubMed/NCBI | |
Rubin EM, Guo Y, Tu K, Xie J, Zi X and Hoang BH: Wnt inhibitory factor 1 decreases tumorigenesis and metastasis in osteosarcoma. Mol Cancer Ther. 9:731–741. 2010. View Article : Google Scholar : PubMed/NCBI | |
Krishnan K, Khanna C and Helman LJ: The biology of metastases in pediatric sarcomas. Cancer J. 11:306–313. 2005. View Article : Google Scholar : PubMed/NCBI | |
Eccles SA and Welch DR: Metastasis: Recent discoveries and novel treatment strategies. Lancet. 369:1742–1757. 2007. View Article : Google Scholar : PubMed/NCBI | |
Fidler IJ: The pathogenesis of cancer metastasis: The ‘seed and soil’ hypothesis revisited. Nat Rev Cancer. 3:453–458. 2003. View Article : Google Scholar : PubMed/NCBI | |
Steeg PS: Tumor metastasis: Mechanistic insights and clinical challenges. Nat Med. 12:895–904. 2006. View Article : Google Scholar : PubMed/NCBI | |
Diao CY, Guo HB, Ouyang YR, Zhang HC, Liu LH, Bu J, Wang ZH and Xiao T: Screening for metastatic osteosarcoma biomarkers with a DNA microarray. Asian Pac J Cancer Prev. 15:1817–1822. 2013. View Article : Google Scholar | |
Smyth GK: Limma: Linear models for microarray dataBioinformatics and computational biology solutions using R and Bioconductor. Gentleman R, Carey VJ, Huber W, Irizarry RA and Dudoit S: Springer; New York: pp. 397–420. 2005, View Article : Google Scholar | |
Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, et al: Gene Ontology: Tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 25:25–29. 2000. View Article : Google Scholar : PubMed/NCBI | |
Kanehisa M and Goto S: KEGG: Kyoto encyclopedia of genes and genomes. Nucleic Acids Res. 28:27–30. 2000. View Article : Google Scholar : PubMed/NCBI | |
Shannon P, Markiel A, Ozier O, Baliga NS, Wang JT, Ramage D, Amin N, Schwikowski B and Ideker T: Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res. 13:2498–2504. 2003. View Article : Google Scholar : PubMed/NCBI | |
Nepusz T, Yu H and Paccanaro A: Detecting overlapping protein complexes in protein-protein interaction networks. Nat Methods. 9:471–472. 2012. View Article : Google Scholar : PubMed/NCBI | |
da Huang W, Sherman BT and Lempicki RA: Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources. Nat Protoc. 4:44–57. 2008. View Article : Google Scholar | |
da Huang W, Sherman BT and Lempicki RA: Bioinformatics enrichment tools: Paths toward the comprehensive functional analysis of large gene lists. Nucleic Acids Res. 37:1–13. 2009. View Article : Google Scholar : PubMed/NCBI | |
Meyer LR, Zweig AS, Hinrichs AS, Karolchik D, Kuhn RM, Wong M, Sloan CA, Rosenbloom KR, Roe G, Rhead B, et al: The UCSC Genome Browser database: Extensions and updates 2013. Nucleic Acids Res. 41:(Database issue). D64–D69. 2013. View Article : Google Scholar : PubMed/NCBI | |
PosthumaDeBoer J, Witlox MA, Kaspers GJ and Van Royen BJ: Molecular alterations as target for therapy in metastatic osteosarcoma: A review of literature. Clin Exp Metastasis. 28:493–503. 2011. View Article : Google Scholar : PubMed/NCBI | |
Liang D, Chang JR, Chin AJ, Smith A, Kelly C, Weinberg ES and Ge R: The role of vascular endothelial growth factor (VEGF) in vasculogenesis, angiogenesis, and hematopoiesis in zebrafish development. Mech Dev. 108:29–43. 2001. View Article : Google Scholar : PubMed/NCBI | |
Björndahl MA, Cao R, Burton JB, Brakenhielm E, Religa P, Galter D, Wu L and Cao Y: Vascular endothelial growth factor-a promotes peritumoral lymphangiogenesis and lymphatic metastasis. Cancer Res. 65:9261–9268. 2005. View Article : Google Scholar : PubMed/NCBI | |
Liu W, Xu J, Wang M, Wang Q, Bi Y and Han M: Tumor-derived vascular endothelial growth factor (VEGF)-A facilitates tumor metastasis through the VEGF-VEGFR1 signaling pathway. Int J Oncol. 39:1213–1220. 2011.PubMed/NCBI | |
Zimmermann M, Zouhair A, Azria D and Ozsahin M: The epidermal growth factor receptor (EGFR) in head and neck cancer: Its role and treatment implications. Radiation Oncol. 1:112006. View Article : Google Scholar | |
Italiano A, Saint-Paul MC, Caroli-Bosc FX, François E, Bourgeon A, Benchimol D, Gugenheim J and Michiels JF: Epidermal growth factor receptor (EGFR) status in primary colorectal tumors correlates with EGFR expression in related metastatic sites: Biological and clinical implications. Ann Oncol. 16:1503–1507. 2005. View Article : Google Scholar : PubMed/NCBI | |
Yang J, Mani SA, Donaher JL, Ramaswamy S, Itzykson RA, Come C, Savagner P, Gitelman I, Richardson A and Weinberg RA: Twist, a master regulator of morphogenesis, plays an essential role in tumor metastasis. Cell. 117:927–939. 2004. View Article : Google Scholar : PubMed/NCBI | |
Berx G, Raspé E, Christofori G, Thiery JP and Sleeman JP: Pre-EMTing metastasis? Recapitulation of morphogenetic processes in cancer. Clin Exp Metastasis. 24:587–597. 2007. View Article : Google Scholar : PubMed/NCBI | |
Meiners S, Brinkmann V, Naundorf H and Birchmeier W: Role of morphogenetic factors in metastasis of mammary carcinoma cells. Oncogene. 16:9–20. 1998. View Article : Google Scholar : PubMed/NCBI | |
Lu P, Weaver VM and Werb Z: The extracellular matrix: A dynamic niche in cancer progression. J Cell Biol. 196:395–406. 2012. View Article : Google Scholar : PubMed/NCBI | |
Felding-Habermann B: Integrin adhesion receptors in tumor metastasis. Clin Exp Metastasis. 20:203–213. 2003. View Article : Google Scholar : PubMed/NCBI | |
Khanna C, Wan X, Bose S, Cassaday R, Olomu O, Mendoza A, Yeung C, Gorlick R, Hewitt SM and Helman LJ: The membrane-cytoskeleton linker ezrin is necessary for osteosarcoma metastasis. Nat Med. 10:182–186. 2004. View Article : Google Scholar : PubMed/NCBI | |
Wan X, Mendoza A, Khanna C and Helman LJ: Rapamycin inhibits ezrin-mediated metastatic behavior in a murine model of osteosarcoma. Cancer Res. 65:2406–2411. 2005. View Article : Google Scholar : PubMed/NCBI | |
Derynck R, Akhurst RJ and Balmain A: TGF-beta signaling in tumor suppression and cancer progression. Nat Genet. 29:117–129. 2001. View Article : Google Scholar : PubMed/NCBI | |
Candy P, Phillips MR, Redfern AD, Colley SM, Davidson JA, Stuart LM, Wood BA, Zeps N and Leedman PJ: Notch-induced transcription factors are predictive of survival and 5-fluorouracil response in colorectal cancer patients. Br J Cancer. 109:1023–1030. 2013. View Article : Google Scholar : PubMed/NCBI | |
Bernardo GM and Keri RA: FOXA1: A transcription factor with parallel functions in development and cancer. Biosci Rep. 32:113–130. 2012. View Article : Google Scholar : PubMed/NCBI |