CLEC10A is a prognostic biomarker and correlated with clinical pathologic features and immune infiltrates in lung adenocarcinoma

Abstract CLEC10A, (C‐type lectin domain family 10, member A), as the member of C‐type lectin receptors (CLRs), plays a vital role in modulating innate immunity and adaptive immunity and has shown great potential as an immunotherapy target for cancers. However, there is no functional research of CLEC10A in prognostic risk, immunotherapy or any other treatment of lung adenocarcinoma (LUAD). We performed bioinformatics analysis on LUAD data downloaded from TCGA (The Cancer Genome Atlas) and GEO (Gene Expression Omnibus), and jointly analysed with online databases such as HPA, LinkedOmics, TIMER, ESTIMATE and TISIDB. We found that lower expression of CLEC10A was accompanied with worse outcomes of LUAD patients. Moreover, CLEC10A expression was significantly correlated with a variety of the tumour‐infiltrating immune cells (TIICs). As a promising prognosis predictor and potential immunotherapy target, the potential influence and mechanisms of CLEC10A in LUAD deserve further exploring.


| INTRODUC TI ON
The morbidity and mortality of lung cancer were both increasing, 1 among which lung adenocarcinoma (LUAD) was a common histological subtype. 2 Despite the progresses made in cancer-related treatment technology during the past years, the five-year survival rate for lung cancer was still extremely poor, mainly because most patients are diagnosed at a last stage. 3 Generally, the five-year overall survival rate for patients diagnosed at advanced LUAD was just 15%; however, more than 60% of LUAD patients missed targetable gene alterations period, which could improve their survival rate. 4,5 Thus, discovering specific early detection markers and therapeutic targets is the key to improve survival rate of LUAD patients.
The level of immune cell infiltration in the tumour microenvironment (TME) plays a cardinal role in the tumour initiation, progression, metastasis and treatment resistance of cancer. [6][7][8] CLEC10A, (C-type lectin domain family 10, member A), is also named MGL, (macrophage galactose type C-type lectin). As a member of the CLRs, CLEC10A, like other members, has been determined to involve in improving the immune response activity of immune cells. CLEC10A recognizes and acts on tumour-associated Tn antigens and effectively presents the antigens to CD4 T cells. 9 Furthermore, the binding of CLEC10A to tumour-associated antigens carrying α-N-acetylgalactosamine can obviously increase antigen-specific CD8 T cell activation. 10 Effective tumour eradication requires tumour-specific CD8 and CD4 T cells. The CLEC10A' s function in improving the anti-tumour activity of immune cells has clearly received people's attention and proposed it as a target for cancer immunotherapy. 11 In this study, we conducted a comprehensive analysis of CLEC10A expression in the risk of LUAD progress based on more than 1200 patients, and then correlated CLEC10A different expression level and the alteration of tumour immune microenvironment.
The results revealed the significant prognostic value of CLEC10A expression and a potentially promising target for immunotherapeutic strategies in LUAD. nlm.nih.gov/gds). Of the eight microarray data sets, GSE72094 has detailed clinical prognostic information, so it is used as a validation set to participate in the study, and the other seven sets of data sets are used to study the differential expression of genes. The CLEC10A protein expressed level in lung tissue was explored based on the immunohistochemistry data from HPA (Human Protein Atlas) database (https://www.prote inatl as.org/).

| Differential expression analysis of CLEC10A
The Wilcox test and Kruskal test were applied to assess the differential expression of CLEC10A.

| Linked omics database analysis
The LinkedOmics database (http://www.linke domics.org/login.php) is utilized to analyse 32 TCGA cancer-associated multidimensional data sets. The differentially expressed genes related to CLEC10A were screened from the TCGA LUAD cohort through the LinkFinder module in the database, and the correlation of results was tested by the Pearson correlation coefficient and presented respectively in volcano plot and heat maps. Function module analysis of Gene Ontology biological process (GO_BP), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways by the gene set enrichment analysis (GSEA) in the LinkInterpreter module.

| TIMER and ESTIMATE database analysis
Tumor IMmune Estimation Resource (TIMER) database (https://cistr ome.shiny apps.io/timer/) is a comprehensive resource for systematic analysis of immune infiltrates across diverse cancer types, which includes 32 cancer types. TIMER used a deconvolution method to infer the abundance of tumour-infiltrating immune cells (TIICs) from gene expression profiles of the LUAD samples in the TCGA and GEO dataset.
Estimation of STromal and Immune cells in MAlignant Tumor tissues using Expression data (ESTIMATE) database (https://bioin forma tics.mdand erson.org/publi c-softw are/estim ate/) is an algorithm using gene expression data to generate Immunescore, which represents the degree infiltration of immune cells in tumour tissue.

| TISIDB database analysis
The TISIDB database (http://cis.hku.hk/TISIDB) is a web portal for tumour and immune system interaction, which integrates multiple heterogeneous data types. They pertain to 988 reported immunerelated anti-tumour genes, high-throughput screening techniques, molecular profiling and para-cancerous multi-omics data, as well as numerous resources for immunological data gathered from seven public databases. Here, TISIDB provides us associations for CLEC10A with lymphocytes, immunomodulators and chemokines.

| Decreased CLEC10A expression in LUAD
From six LUAD studies of GEO and TCGA database, CLEC10A expressed lowly in LUAD compared to non-cancer tissues ( Figure 1A). This comparison result was still accurate in 192 pairs of LUAD tissues and matched non-cancer tissues in GEO and TCGA database ( Figure 1B-D). And the differential expression of CLEC10A between LUAD and normal tissues was also reflected in the protein expression level (Supplementary Figure 1). These results revealed that CLEC10A may play an inhibitory role in the LUAD process.

| Correlations between CLEC10A expression and clinicopathological parameters in LUAD patients
Since the function of CLEC10A in LUAD remains unclear, the cor- Other clinical parameters, such as advanced T stage, N stage, M stage and TNM stage, also correlated with the worse overall survival time ( Figure 2E). For verifying the prognostic value of CLEC10A in LUAD, we performed the multivariate analysis. The results showed that only CLEC10A expression and TNM stage were independently associated with the overall survival time ( Figure 2F), which shows that CLEC10A expression can not only participate in guiding clinical work like common clinical phenotypes, but also means that its role in evaluating patients' clinical prognosis is superior to T stage, N stage and M stage.

| CLEC10A Co-expression network in LUAD
For gaining the knowledge of CLEC10A biological function in LUAD, the LinkFinder module in the LinkedOmics web portal was deployed to check the co-expression pattern of CLEC10A in TCGA-LUAD. As is plotted in Figure 3A, it showed that 6116 genes (dark red dots) positively correlated with CLEC10A, and 4456 genes (dark green dots) negatively correlated. Figure

| Association between CLEC10A with immune infiltration level
We explored whether CLEC10A expression can affect various immune cell infiltration levels in LUAD from the TIMER database.

| Relation between CLEC10A with immune molecules
Finally, to broaden the cognition of the correlation between CLEC10A and immune infiltration, we investigated the connections between CLEC10A expression and various immune signatures, which included the immune-related signatures of 28 TIL types from Charoentong's study, three kinds of immunomodulators, chemokines and receptors.
Associations between CLEC10A expression and various immune signatures were obtained from the TISIDB database.

| D ISCUSS I ON
Tumour cells exist in a complex tumour microenvironment (TME). 12 The essence of TME is the cellular and non-cellular components present in and around the tumour. Generally, TME is subdivided into extracellular matrix (ECM), stromal cells and immune cells. 13  However, in most cases, the main role of TME is immunosuppression, which blocked anti-tumour immunity and sustain tumour progression. 14 Immunosuppressive effect of TME regulated by all immune cell types with immunomodulatory activities. 12 Within TME, macrophages tend to become tumour-associated macrophages (TAMs) to drive tumour progression, invasion and metastasis 16 ; tumour-infiltrating dendritic cells also incline to promote immunosuppression and tolerance, rather than drive anti-tumour immunity 6 ; neutrophils recruited into tumour, prevalently polarize towards the N2-subtype with pro-tumoral functions. 6 As the main performers of anti-tumour responses, T cell-mediated immune response reduced by most tumours with multiple strategies, including inhibiting T cell transport to tumour, interfering with antigen-presenting cells and effector T cells. 7,14 The most typical is that tumour cells express PD-L1 or PD-L2 ligand matched to the PD-1 protein of T cells, make it cannot find the tumour and send a signal to the immune system to attack the tumour, and directly cause T cells exhaustion. Clinically, by blocking this event, that is, the application of anti-PD-1 and anti-PD-L1 antibodies can not only facilitate the T cells' proliferation, but also restore their cytotoxic responses against tumour cells. 14,17 Therefore, under the immense progress and bright prospects of tumour immunotherapy, all factors that participating in the modulation of immune cells in TME are worthy of our consideration and research.
In the past decade, the C-type lectin receptors (CLRs) have acquired rising attention due to their functions in fine-tuning of innate and adaptive immunity. CLRs are a huge family of receptors, containing more 1,000 members, with diverse functions, including cell adhesion, complement activation, tissue remodelling, platelet activation, endocytosis, phagocytosis and innate immune activation. 18,19 CLRs contain multiple C-type lectin-like domains, which can specifically recognize specific glycosylated antigens, 20 the form of most foreign antigen display derived from tumour cells or viruses. 21 In innate immunity, CLRs are mainly expressed on antigen-presenting cells (macrophages, neutrophils and dendritic cells (DCs)) and play a critical role in identifying diverse pathogens, such as fungi, bacteria, viruses and parasites. 22 Activation of the innate immune system was a crucial basis for constructing an adaptive immune response.
The binding of CLRs to ligands results in various cellular responses, including respiratory burst, production of cytokines and chemokines, and consequently shaping the adaptive immune responses. 23,24 CLRs-mediated innate immune responses can direct the progress of cellular immunity including Th1, Th17 and CD8 cytotoxic T lymphocytes cells immune responses through triggering the production of multiple cytokines. [25][26][27][28] Furthermore, it has been recently recognized that the CLRs activation involved in developing of regulatory T cells to modulate the function of CD4 T cells. 29 The immunomodulatory effect of CLRs on cellular immunity has been treated as a promising solution for future cancer treatment. Several CLRs agonists or antagonists have proved to be potential anticancer drug candidates. The most representative one is β-glucan, agonist of dectin-1 (dendritic cell-associated C-type lectin-1). In murine model, the application of β-glucan showed a significant tumour growth inhibitory effect. 30,31 Moreover, the combined application with anti-tumour monoclonal   Figure 2). This result clearly shows that CLEC10A has a wide range of effects on the infiltration of immune cells in TME of most cancers, and it is worth further investigation in the field of tumour immunotherapy.
We concluded that there is a possible prognostic molecular marker for poor survival in LUAD, called CLEC10A expression.
Decreased CLEC10A expression leads to worsening of clinical features (primary tumour scope, distant metastasis, pathological stage of tumour and prognosis). CLEC10A is the member of CLRs family, which have been confirmed to be involved in the regulation of the migration of various immune cells. In this research, we revealed that the raising expression level of CLEC10A was obviously linked to the increasing infiltration level of various TIICs in tumour tissues. The immune microenvironment composed of these TIICs profoundly affects the prognosis of LUAD. Therefore, in our clinical work, we can try to evaluate the degree of malignancy of the patient by measuring the expression of CLEC10A in the surgical specimens of LUAD patients, predict the prognosis of the patient, and even better assess the status of immune microenvironment, and develop immunotherapeutic drugs targeting CLEC10A. We recommend strongly that researchers in the field of tumour immunology conduct further research on CLEC10A in LUAD to gradually elaborate the biological function of CLEC10A in the immune microenvironment and prognosis of LUAD patients.

ACK N OWLED G EM ENTS
We thank the authors of TCGA, GEO, HPA, LinkedOmics, TIMER, ESTIMATE and other online databases for their selfless givenness and harding work on collecting and collating these detailed databases for us, scientific researchers in the broad field of biomedicine.

CO N FLI C T O F I NTE R E S T
The authors declare that there are no conflicts of interest.