It has been reported in a huge number of studies that TME plays critical roles in multi cancer types and promotes the progression of molecular classification systems and treatment strategies. Our study identified LRRN4CL, HS3ST3A1, PCOLCE2, and CAPN8 as TME-related genes associated with prognosis in TC samples collected from the TCGA database, and verified two of them, HS3ST3A1 and CAPN8, through clinical specimen and cellular functional experiments.
LRRN4CL lacks in-depth research, while only one study reported that the upregulation of the cell surface protein LRRN4CL promoted metastases of pulmonary in mice. HS3ST3A1 encodes the enzyme 3-Osulfotransferase, which catalyzes the biosynthesis of 3-O-sulfated heparan sulfate, a specific subtype of heparan sulfate (HS). It has been reported that HS3ST3A1 is involved in respiratory papillomatosis, human immunodeficiency virus (HIV) infection, and lung cancer. PCOLCE2 is reported to encode a functional collagen-binding protein procollagen C-proteinase enhancer (PCPE2) , and evidence has proved its role in colorectal cancer patients. CAPN8 belongs to the Calpains family and exhibits restricted expression patterns. CAPN8 has been proposed to be involved in vesicle trafficking and gastric cancer. Synthesize previous studies, there are hardly any reports on their functions in the tumor microenvironment, but their roles as prognosis indicators are certainly clarified in several types of cancer. It’s our study that collects LRRN4CL, HS3ST3A1, PCOLCE2, and CAPN8 all together for the first time and sets them as biomarkers related to TME in thyroid cancer, which might become potential candidate targets for TC immunotherapy.
It’s commonly recognized that inflammation and autoimmunity are risk factors for TC. Evidence also indicated that TC patients might benefit from targeting cancer-related inflammation, which provided new strategies for diagnosis and treatment . Studies of immune infiltration in TC have made many major advances, particularly in the study of primary immune cells, such as NK cells, TAMs, MCs, DCs, CD8 + T cells, B cells, neutrophils, and Tregs. [22, 24, 25]. In our study, the proportion of TIICs was estimated using the CIBERSORT algorithm, which indicated that TME-related DEGs had a notably association with specific immune cells, including monocytes, neutrophils, T cells, and so on. Previous studies revealed that TAMs, MCs, DCs, Tregs, and neutrophils were positively related to TC progression[26–30]. While NK cells, iDCs, CD8 + T cells, and B cells are negatively associated with TC progression[31–35]. However, some studies clarified that immune cells are related to various outcomes in cancer. It’s for sure that further studies will be conducted to explore the role of different immune cells plays in cancer for better prognostic evaluation.
As for TME-related genes in thyroid cancer, a previous study has identified 30 hub genes by constructing the PPI network and set CXCL10 as the top hub gene. Our study has different logical methods and conducted further data mining. First, we obtained more gene expression and clinical profiles so that our study might have more credibility. CXCL10 was a specific differential expressed TME-related gene in the PPI network with the most nodes, but it tended to lose its significance when performing univariate Cox regression analysis. Furthermore, we analyzed the immune infiltrating profiles of TC and concluded the correlation between immune cells and DEGs using the CIBERSORT algorithm. These make sense for better identifying TME-related biomarkers and prognostic indicators in patients with TC.
HS3ST3A1 and CAPN8 were considered to serve as prognostic predictors in our study. Clinical specimens were collected and subjected to RT-qPCR and IHC stain, which showed the higher HS3ST3A1 and CAPN8 expression levels in tumor tissues, especially in tumor with LNM. In vitro, the downregulation of HS3ST3A1 and CAPN8 presented the inhibition of proliferation and invasion in papillary thyroid cancer cells. These findings indicated that TME-related genes HS3ST3A1 and CAPN8 function in immune process and contribute to tumor development.
However, it is necessary to acknowledge the limitations of this study. The biases were unavoidable because the data were primarily obtained from the TCGA database. Besides, we chose two of the genes for experimental verification; the other two genes, LRRN4CL and PCOLCE2 were excluded due to their adverse expression both in paired and un-paired tissue. As for clinical specimens, our data lacks survival information due to the good outcome of TC patients. Furthermore, our experiments mainly focused on papillary thyroid cancer cells, which is the most common subtype of TC, but the verification in other thyroid cancer subtypes is lacking. Animal experiments and the depth of molecular mechanisms still need further investigation.
Overall, we used the ESTIMATE algorithm to identify DEGs related to the TME in TC specimens obtained from the TCGA dataset. LRRN4CL, HS3ST3A1, PCOLCE2, and CAPN8 were observed as potential prognostic indicators for patients with TC. Furthermore, HS3ST3A1 and CAPN8 were highly expressed in thyroid tumor tissue, especially in tumor with LNM, and their downregulation can inhibit the proliferation and invasion of thyroid cancer cells. However, the underlying molecular mechanisms of tumor micro-metastasis and the potential clinical value for early diagnosis still require further experimental study.