For the past century, the classification of tumors has been based largely on the concept that tumors can be classified according to their histological features under a light microscope. Over the past 10 years, studies have clarified the genetic basis of tumorigenesis in various tumors, and such understanding raises the possibility that it may contribute to the further classification of tumors (Louis, 2012). Also, a better understanding and identification of predictive and prognostic biomarkers are needed for individualized treatment and improved survival.

Secreted modular calcium-binding protein 1 (SMOC1) was first isolated in 2002, which encodes a secreted modular glycoprotein (Vannahme et al., 2002). SMOC1 contains a follistatin-like domain, an EF-hand calcium-biding domain, two thyroglobulin-like domains, and a unique domain and is usually localized in the basement membrane of different tissues and also can present in other extracellular matrices (Vannahme et al., 2002). It belongs to a family of matricellular proteins that also include basement membrane-40 (also known as secreted protein acidic and rich in cysteine), as well as SMOC2 (Bornstein and Sage, 2002). SMOC2 was involved in multiple biological processes, including cell cycle progression (Liu et al., 2008), cell attachment (Maier et al., 2008), and tumor development (Su et al., 2016; Huang et al., 2017). Also, in previous studies, the expression of SMOC2 was decreased in gallbladder cancer (Gu et al., 2015) and advanced breast cancer (Fidalgo et al., 2015) but increased in metastatic head and neck squamous cell carcinoma (Hyakusoku et al., 2016). In endometrial carcinoma, SMOC2 was able to inhibit cell proliferation and overcome chemoresistance (Lu et al., 2019). SMOC1 has similar domains to SMOC2, except for its own unique domain (Vannahme et al., 2003). We assume that similar to SMOC2, SMOC1 might also be involved in tumor development. The expression of SMOC1 was increased in some brain tumors, such as oligodendroglioma (Brellier et al., 2011) and astrocytic tumors (Boon et al., 2004). Moreover, SMOC1 was identified as a new cancer-related protein by interacting with tenascin-c, which was an ECM protein and was highly expressed in many human cancers (Brellier et al., 2011). In U87 glioma cells, SMOC1 inhibits the tenascin-c-induced chemo-attractive effect (Brellier et al., 2011). Previous studies indicated that SMOC1 was involved in embryogenesis (Gersdorff et al., 2006), endothelial cell proliferation, angiogenesis (Awwad et al., 2015), integrin–matrix interactions, cell adhesion (Klemencic et al., 2013), and glucose metabolism (Montgomery et al., 2020). However, the role of SMOC1 in tumor genesis and progression is still unclear.

In this study, we aimed to systematically explore the gene expression and evaluate the prognostic values of SMOC1 in different cancers and uncover the potential functions of SMOC1 in specific cancers. The findings explored the prognostic value of SMOC1 in low-grade glioma (LGG) and provided new ideas for exploring the potential mechanism of SMOC1 in glioma development and progression.

Gliomas are the most common malignant tumor of the CNS, with a character of aggressive growth, poor outcome of the patient, and high rates of recurrence (Wen and Kesari, 2008). According to the CBTRUS report, glioma represents approximately 25.1% of all primary and 80.8% of malignant tumors in the brain and other CNS tumors diagnosed in the United States in 2013–2017 (Ostrom et al., 2020). Also, the WHO classifies glioma into four grades: WHO grades I, II, III, and IV (Louis et al., 2016). LGG usually refers to gliomas other than WHO grade IV glioma (GBM) and accounts for approximately 10% of all primary brain tumors (Ostrom et al., 2020). The prognosis of gliomas varies widely by WHO classification. Compared with GBM patients, LGG patients usually have an indolent course and longer survival (Forst et al., 2014). However, despite initially growing slow, LGG can transform to GBM with time (Morshed et al., 2019). According to the previous study (Afra et al., 1999), nearly half of the patients with LGG will experience malignant transformation, usually within 5 years. However, the risk factors for the development of LGG are poorly understood at present, and the current understanding of the mechanism underlying the growth and invasion of glioma is limited. The application of IDH mutate-status in 2016 WHO Classification of CNS tumors showed a great significance of molecular diagnosis in glioma (Louis et al., 2016). However, more molecular biomarkers are still needed to be discovered.

SMOC1 was first reported in 2002; we know not much about the biological function of SMOC1. However, its potential importance cannot be ignored because it has been linked with embryogenesis (Gersdorff et al., 2006), osteoblast differentiation (Choi et al., 2010), and some forms of cancer. One in silico study analyzed the expression and prognostic data of long noncoding RNAs, microRNAs, and mRNAs in the colon cancer dataset from the TCGA database and identified five mRNAs, including SMOC1, as potential prognostic biomarkers for colon cancer (Huang and Pan, 2019). Besides, another study suggests that the aberrant expression of SMOC1 can inhibit the proliferation and colony formation in colorectal cancer cells, as well as tumor formation in vivo (Aoki et al., 2018). This suggests that SMOC1 may act as a tumor suppressor. A recent study has analyzed the methylation data of the glioma project from the TCGA database and identified 10 glioma grade-associated cytosine-phosphate guanine sites, which targeted four genes, including SMOC1 (Weng and Salazar, 2021). Another in silico study identified a seven-gene signature, including SMOC1, which was positively correlated with 5-years OS of glioma patients (Zhang G.-H. et al., 2019). However, the specific relationship between the expression level of SMOC1 and the prognosis of glioma patients has not been evaluated yet.

In this study, we first systematically analyzed the expression level of SMOC1 and its prognostic value in glioma. Compared with levels in normal tissue, the expression of SMOC1 was aberrantly in many cancers; it was reduced in bladder urothelial carcinoma, BRCA, cholangiocarcinoma, kidney renal clear cell carcinoma, kidney renal papillary cell carcinoma, liver hepatocellular carcinoma, prostate adenocarcinoma, thyroid carcinoma, and uterine corpus endometrial carcinoma but increased in kidney chromophobe, LUAD, and LGG. These data suggest that the alterations in SMOC1 expression depend on the type of cancer. Survival analyzes from PrognoScan and GEPIA2 showed that the expression level of SMOC1 was correlated with the prognosis of brain glioma (LGG) and LUAD patients. Therefore, we further analyzed the expression of SMOC1 in glioma and LUAD via the GEPIA2 and CGGA databases. The results showed that the expression of SMOC1 was remarkably increased in tumor samples of LGG, whereas there were no significant changes in GBM and LUAD samples, which was consistent with previous studies (Boon et al., 2004; Brellier et al., 2011). In addition, we found that the expression of SMOC1 was significantly increased in subtypes of glioma through the CGGA dataset. Compared with IDH-wild-type, 1p19q non-codeletion, and recurrent subtypes, the expression of SMOC1 was increased in IDH mutant, 1p19q co-deletion, and primary gliomas. In the age < 42 years group, a higher expression level of SMOC1 was also observed. As we know, IDH wild-type, 1p19q non-codeletion, recurrent gliomas, and age > 40 years were poor prognostic factors in LGG (Nabors et al., 2017). These results confirmed the upregulation of SMOC1 expression in LGG and suggested that the expression of SMOC1 increases with the decrease of tumor malignancy. Consistent with the analysis of the TCGA datasets in GEPIA2, the survival analysis of CGGA datasets also showed a significant positive correlation between SMOC1 expression and better prognosis in all WHO grade I, WHO grade II, and WHO grade III gliomas but not in WHO grade IV glioma. These findings further confirmed the prognostic value of SMOC1 in specific types of cancer. Therefore, based on the consistent results of the association between SMOC1 expression and survival of LGG patients, we have reason to believe that SMOC1 can serve as a good prognostic biomarker in LGG. As mentioned before, the expression of SMOC1 was increased in oligodendroglioma (Brellier et al., 2011) and astrocytic tumors (Boon et al., 2004). Also, SMOC1 has even been identified as a new cancer-related protein by interacting with tenascin-c, and it inhibits the tenascin-c induced chemo-attractive effect (Brellier et al., 2011) in U87 glioma cells. However, the function of SMOC1 in glioma is still unclear. Also, it is necessary to focus on its precise effects in this cancer type, as well as the underlying mechanisms.

According to previous studies, the gene SMOC1 encodes a secreted modular glycoprotein (Vannahme et al., 2002). This protein belongs to a family of matricellular proteins, and it was generally expressed on the basement membrane of different tissues in adult animals and also can be present in other extracellular matrices (Vannahme et al., 2002). During mouse development, SMOC1 was expressed in the basement membrane zones of the brain, blood vessels, lung, heart, and many other tissues. This broad distribution suggests that SMOC1 might have multifunctional roles during mouse embryogenesis (Gersdorff et al., 2006). Besides, previous studies have shown that SMOC1 might also involve in angiogenesis. SMOC1 was highly expressed in proliferation endothelial cells, and the expression of SMOC1 was regulated by inflammatory cytokines and nitric oxide (Dreieicher et al., 2009). Another study indicated that SMOC1 acts as a negative feedback regulator of the activin-like kinase 5 signal pathway by binging to endothelin, leading to activation of transforming growth factor-beta signal pathway and activin-like kinase 1, thus promoting endothelial cell proliferation and angiogenesis (Awwad et al., 2015). Moreover, one study reported that SMOC1 might be involved in integrin–matrix interactions and cell adhesion (Klemencic et al., 2013). In addition to its role in matrix remodeling, SMOC1, as a circulating glycoprotein, was identified as a regulator of glucose homeostasis (Montgomery et al., 2020). A recent study identified SMOC1 as a novel thrombin-activating protein, which enhances the action of thrombin in vivo and in vitro (Delgado Lagos et al., 2021).

In this study, with the GSEA tool of LinkedOmics database, we found that the SMOC1 coexpressed genes were mainly enriched for protein localization and in pathways such as ribosome in LGG, whereas the genes involved in processing such as response to lymphocyte- and neutrophil-mediated immunity, leukocyte migration, adaptive immune response, T cell activation and in pathways such as ECM–receptor interaction, Th17 cell differentiation, and leukocyte trans-endothelial migration were inhibited in LGG. In the further analysis in CancerSEA, we found that the expression of SMOC1 was correlated with several important functional states in glioma cells, especially stemness, hypoxia, EMT, and metastasis. These results strongly suggested that SMOC1 might influence various processes in the glioma tumor microenvironment. It is well known that interactions between malignant cells and the extracellular environment are critical for cancer development and progression (Rangarajan and Weinberg, 2003). Additionally, we observed in our analysis that there was a positive correlation between SMOC1 expression and tumor purity in LGG, which suggested that the expression of SMOC1 was more likely from tumor cells, which was consistent with a previous study (Brellier et al., 2011). The expression of SMOC1 was negatively correlated with levels of infiltrating B cells, CD8⁺ T cells, CD4⁺ T cells, macrophages, neutrophils, and dendritic cells, as well as gene markers of most immune cells in the LGG tumor microenvironment. Interestingly, there was no significant correlation between SMOC1 expression and tumor purity or most of the infiltrating immune cells in GBM. These results indicate that the expression of SMOC1 was significantly related to immune infiltrating cells in LGG. Together with the functional analysis, our findings strongly suggest that SMOC1 might play an important role in the tumor microenvironment of glioma, thereby influencing glioma development and progression. Further experimental studies are needed to validate our conclusions and explore the specific function of SMCO1 in glioma.

In conclusion, our findings in this study suggest that SMOC1 was highly expressed in LGG, and the expression of SMOC1 was positively correlated with the survival of LGG patients. The functional analyses indicate that SMCO1 might play an important role in the glioma microenvironment, thereby influencing the development and progression of glioma. However, the prognostic value of SMOC1 in glioma needs to be validated in clinic samples, and as little was known about the biological function of SMOC1 in tumor genesis and progression, further in vitro and in vivo investigations are required to clarify the role of SMOC1 in glioma.