LY96 Upregulation Promotes Kidney Renal Clear Cell Carcinoma (KIRC)

Background: LY96 has been reported to be relevant with kidney inammatory injury but the function of this gene in kidney renal clear cell carcinoma (KIRC) remains unknown. Methods: Various online tools were applied to analyze the roles of LY96 in KIRC using data from the Cancer Genome Atlas. Differential LY96 expression and overall survival (OS) based on different expression levels were analyzed through Oncomine and GEPIA tools. The alterations, related genes, Gene Ontology, and Kyoto Encyclopedia of Genes and Genomes pathways of LY96 were explored via cBioPortal and STRING database. LinkedOmics and Cistrome DB Toolkit were utilized to identify targets of kinase, miRNAs, and transcription factors. The relationship between LY96 and some associated genes or regulatory factors was displayed via GeneMANIA and TIMER tool. TISIDB revealed correlations between LY96 expression and immune-associated factors in the tumor microenvironment. Results: High LY96 expression level was observed in KIRC and associated with poor prognosis and diverse clinical characteristics. LY96 often amplied in KIRC and was mostly linked to the inammatory response. Several highly correlated genes, kinase targets, transcription factors, and DNA methyltransferase that may interact with LY96 were all identied. Our study also demonstrated that various immune-related factors were relevant to LY96 in KIRC. Conclusions: Our study has shown the complex relationships between LY96 and KIRC from diverse angles. High LY96 expression had an adverse effect on the prognosis of KIRC. To nd effective demethylation agents and transcription factors inhibitors targeting LY96 may have benecial effects on the survival of KIRC patients.


Introduction
Renal cell carcinoma (RCC) is categorized into three subtypes: clear cell renal cell carcinoma (ccRCC), papillary renal cell carcinoma (pRCC), and chromophobe renal cell carcinoma (chRCC). Kidney renal clear cell carcinoma (KIRC) is the most common type of renal malignancy, accounting for 75% of RCC [1]. The incidence and cancer-related mortality of KIRC is increasing year by year and have become a global health problem. Early stage RCC is often treated by surgery and the overall survival (OS) can reach 60%-70% [2]. However, metastasis RCC remains incurable and is mainly treated with targeted therapies [3]. Clinically, tyrosine kinase inhibitors (TKIs) sunitinib and sorafenib are used to improve the OS of mRCC greatly [4][5][6]. Thus, discovery of novel therapeutic drugs or targets is an urgent need.
MD2/LY96 (Myeloid Differentiation Protein-2/Lymphocyte Antigen 96) is an accessory receptor lacking a transmembrane domain and the co-receptor for Toll-like receptor 4 (TLR4) [7]. MD2/LY96 contributes to lipopolysaccyaride (LPS) recognition and the subsequent TLR4 activation [8]. It can also cooperate with TLR4 in the innate immune response to bacterial LPS, respond to cell wall components from G+ and Gbacteria with TLR-2 and enhance TLR4-dependent activation of NF-κB [9][10]. Previous studies indicated that stimulation of TLR-4/MD2 complex by LPS increased the risk of liver metastasis from primary colorectal cancer (CRC) [11]. Inhibition of MD2 was also con rmed to be an important target for colon cancer therapies [12]. Beyond this, TLR-4 and MD-2 were reported to be associated with gastric cell proliferation and invasion [13].
In this study, we comprehensively analyzed the relationship between LY96 expression levels and KIRC using the Oncomine, GEPIA, UALCAN, c-BioPortal, STRING, LinkedOmics, GeneMANIA, TIMER, TISIDB, CancerSEA, MEXPRESS, Cistrome DB Toolkit, ENCORI and TRRUST public database. We analyzed genomic alterations, functional network, and survival characteristics of LY96 in KIRC. And the propose of this study aims to determine whether LY96 can be a novel potential biomarker of KIRC.

Oncomine analysis
Oncomine is the largest oncogene chip database and data mining platform including 715 gene expression data sets and 86733 tumor and normal tissue samples (http://www.oncomine.org) [14]. We  [15]. Correlations between LY96 expression levels and prognosis of KIRC were displayed by Kaplan-Meier (KM) survival curves. Besides, the relationships between LY96 expression and the outcomes of KICH (Kidney Chromophobe) and KIRP (kidney renal papillary cell carcinoma) were also showed via GEPIA.

UALCAN analysis
UALCAN is a comprehensive, user-friendly, and interactive web resource for analyzing cancer OMICS data (http://ualcan.path.uab.edu) [16]. The ualcan database provides convenient access to public cancer transcriptome data, identi cation of biomarkers and patient survival information based on different gene expression, cancer stages, tumor grade and clinical characteristics.

c-BioPortal analysis
The cBio Cancer Genomics Portal (http://cbioportal.org) is an open-access resource for interactive exploration of multidimensional cancer genomics data sets [17]. The OncoPrint showed an overview of LY96 alterations in KIRC, mainly including mutation, CNVs, and mRNA expression.

STRING analysis
The STRING database (https://string-db.org/) aims to collect, score and integrate all publicly available sources of protein-protein interaction information [18]. Gene Ontology (GO) function, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses, protein-to-protein interaction (PPI) networks of LY96 were performed using the STRING and related results were obtained.

LinkedOmics analysis
LinkedOmics is publicly available portal that includes multi-omics and clinical data from all 32 TCGA Cancer types [19]. It also includes mass spectrometry-based proteomics data generated by the Clinical Proteomics Tumor Analysis Consortium (CPTAC) on selected TCGA tumor samples. The differentially expressed genes (DEGs) associated with LY96 were screened from the TCGA KIRC cohort via the LinkFinder module. Pearson correlation coe cient was utilized for correlation analysis. Outcomes were visualized graphically using volcano plots, heat maps or scatter plots. Data from the results of LinkFinder module were ranked and analysis of kinase target, miRNA-target, transcription factor-target were explored via LinkInterpreter module. The rank criteria was FDR<.05 and 500 stimulations were performed.

GeneMANIA analysis
GeneMANIA (http://www.genemania.org) is a web site for constructing protein-protein interaction (PPI) network, generating hypotheses about gene function and analyzing gene lists [20]. Given a query list, GeneMANIA can nd functionally similar genes using a number of genomics and proteomics data.
Another important role of GeneMANIA is gene function prediction. GeneMANIA can nd genes likely to share function with LY96 according to their interactions. In addition, many genes have been proved to be related to KIRC. Thus, we used GeneMANIA to explore the relationship between LY96 and several signi cant genes.

TIMER analysis
TIMER is a public resource dedicated to tumor immune in ltrates across 32 cancer types incorporating 10897 samples from The Cancer Genome Atlas (TCGA) (http://timer.comp-genomics.org/) [21]. The relevance of LY96 expression with several related genes and regulatory factors were analyzed by TIMER.
TISIDB analysis TISIDB (http://cis.hku.hk/TISIDB) is a web portal that integrates multiple heterogeneous data types for analysis of tumor and immune system interactions [22].
Spearman correlations between LY96 and diverse immune factors were analyzed from the TISIDB database including immune-inhibitory and stimulatory factors, receptors, and chemokines.

CancerSEA analysis
CancerSEA (http://biocc.hrbmu.edu.cn/CancerSEA/) is the rst single-cell sequencing database aimed to comprehensively analyze the 14 functional states of cancer cells at single-cell level [23]. In this study, we used it to explore the potential roles of LY96 in KIRC.

MEXPRESS analysis
MEXPRESS (https://mexpress.be/) is a data visualization tool for the relationships between DNA methylation status and clinical data [24]. And the methylation status of LY96 was performed by MEXPRESS web tool.

Cistrome Data Brower (DB) Toolkit analysis
The Cistrome DB Toolkit (http://cistrome.org/db) is a user-friendly, up to date, and well maintained resource, which was utilized to depict the genome-wide locations of transcription factor binding sites, histone post-translation modi cations and regions of chromatin accessible to endonuclease activity [25].
We used this tool to infer which transcription factors (TFs) were most likely to increase the expression of LY96 in KIRC.

ENCORI analysis
ENCORI is an open-source platform for studying the miRNA-NcRNA, miRNA-mRNA, ncRNA-RNA, RNA-RNA, RBP-ncRNA, and RBP-mRNA interactions from CLIP-seq, degradome-seq and RNA-RNA interactome data [26]. This tool was performed to verify the difference of LY96 expression in KIRC and normal tissues.

LY96 expression in KIRC
We analyzed LY96 transcription levels in KIRC tumors by Oncomine database and observed that the mRNA expression of LY96 in KIRC tissues was signi cantly higher than normal tissues ( Figure 1).
Besides, the mRNA expression of LY96 was among the top 8% and the differences were all more than twofold between KIRC patients and normal tissues. To further determine the role of LY96 in KIRC, we utilized multiple clinical factors of KIRC samples in the TCGA database to compare the LY96 expression levels in each group, including histological stages, race, gender, age, tumor grade, KIRC subtype, and nodal metastasis status (Figure 2a-h). The results of all groups showed that KIRC patients always displayed a higher transcription level of LY96 than normal people. And this conclusion was also validated by ENCORI (P<.01) (Figure 2i). Thus, LY96 may serve as a novel biomarker in KIRC.
The relationship between LY96 expression levels and prognosis of KIRC patients Survival curves were utilized to assess the effect of LY96 on KIRC, KICH, and KIRP prognosis. The results showed a signi cant association between LY96 expression levels and survival of KIRC patients. And the high expression of LY96 was prognostic for worse survival in KIRC. However, no signi cant relevance was observed between LY96 expression and the OS of KICH or KIRP patients ( Figure 3).

Genomic alterations of LY96 in KIRC
We used the cBioPortal to nd out the frequency of LY96 alterations in KIRC according to the data from the TCGA database. LY96 was 0.9% (16/1598) altered in KIRC patients. The alterations included ampli cation in 12 cases (0.75%), mutation in 3 case (0.19%), and deep deletion in 1 case (0.06%).
Therefore, ampli cation was the most frequent type of LY96 CNV in KIRC (Figure 4).

Enrichment analysis of LY96 functional network in KIRC
To further explore the interactions of LY96, GO functional enrichment and KEGG pathway analysis were carried out ( Table 1). The GO enrichment analysis is composed of three main parts: GO biological process (GO-BP), GO molecular function (GO-MF), and GO cellular component (GO-CC). GO-BP included 243 GO-terms, the most important of which were toll-like receptor signaling pathway, innate immune response-activating signal transduction and MyD88-dependent toll-like receptor signaling pathway. GO-MF contained 28 items and LY96 was mainly involved in signaling receptor activity, lipopolysaccharide receptor activity and lipopolysaccharide binding. Besides, LY96 was enriched for 41 GO-CC terms and endolysosome membrane, phagocytic vesicle, and lipopolysaccharide receptor complex were the top three categories. A total of 21 items were included in the KEGG pathways analysis. LY96 was predominantly related to Toll-like receptor signaling pathway, Tuberculosis, and Chagas disease (American trypanosomiasis). Based on the correlation score, the most related genes were TLR4, LY86, TLR2, TLR1, and CD180 ( Table 2). The PPI network displayed intricate relationship between and neighboring genes in Figure 5. LinkedOmics was also used to analyze the mRNA sequencing data from 533 KIRC patients. The positive and negative correlations between 50 signi cant genes and LY96 were exhibited in the heatmap (Figure 6a

Relevance of LY96 across 14 functional states in RCC and the relationships among LY96 and transcription factors, DNA methyltransferase, clinical information
To further investigate the functions of LY96 in RCC, single-cell analysis was performed using CancerSEA. The results indicated that no biological process was signi cantly correlated with LY96 expression in RCC (Figure 8a). The ndings suggested that LY96 was not associated with the occurrence and development of renal cell carcinoma (Table 4). LY96 was probably speci c for KIRC. In addition, in order to identify members that regulate LY96 expression, transcription factors (TFs) that possibly in uenced transcription of the LY96 gene were explored (Figure 8b). Associations between the 20 top TFs and LY96 expression levels in KIRC were investigated using the TIMER database. The results showed that CEBPB, E2F7, SNAI2, MTA2, STAT3, and RAD21 were all signi cantly correlated with LY96 in KIRC (P<.001) (Figure 8c). A previous study pointed out that the heterogeneity of DNA methylation was strongly associated with a high risk of RCC [31]. So, the relationships between LY96 and DNA methyltransferase (DNMT) expression levels were examined by TIMER. The result demonstrated that DNMT1 positively correlated with LY96 in KIRC (cor=0.257, P<.001) (Figure 8c). It seems that TFs and DNA methylation modi cations regulate LY 96 expression in KIRC. Finally, we used MEXPRESS to study the relationships between LY96 expression and multiple clinical variables. The analysis indicated that history of neoadjuvant treatment (no, yes), lactate dehydrogenase result (elevated, normal), neoplasm histologic grade (g1, g2, g3, g4, gx), pathologic T (T1-T4), pathologic (m0, m1, mx), performance status scale timing (at recurrence/progression of disease, post adjuvant therapy, post secondary therapy, other), person neoplasm cancer status (tumor free, with tumor), platelet qualitative result (elevated, low, normal), white cell count result (elevated, low, normal), tumor stage (I-IV) and sample type (primary tumor, solid tissue normal) were closely related to the expression levels of LY96 (P<.05) (Figure 8d). Besides, the results also showed LY96 expression was negatively associated with the OS of KIRC patients (r=-0.162, P<.001).

Discussion
LY96, also known as MD2, is a co-receptor of TLR4 and is necessary for bacterial LPS binding. A previous study has reported that targeting TLR4/MD2 is an important therapeutic strategy against immunosuppressive disease [32]. Another literature proposed that MD2 was a signi cant contributor in the Ang II-induced kidney in ammatory injury in chronic renal diseases [33]. Our study has revealed that LY96 expression levels between tumor and normal tissues differed in various cancer types, especially in KIRC (supplementary Figure 3). Besides, high expression of LY96 was signi cantly correlated with advanced tumor stage, high histological grade and poor prognosis.
To date, molecular markers for RCC diagnosis and prognosis prediction remain lacking, so renal mass biopsy is still the most common and reliable diagnostic means. We sought to identify the relationships between developmental and progressive mechanisms of RCC and LY96 expression, however, no speci c correlation was found. Furthermore, LY96 expression was not associated with KICH and KIRP prognosis. Thus, LY96 might be an indicator of good speci city for KIRC and deserves further validation as a potential diagnostic, therapeutic, and prognostic biomarker.
Analysis of whole-genome alterations in genitourinary malignancies has reported that higher CNV was closely associated with inferior survival for KIRC [34]. Increased LY96 gene CNV was observed in KIRC and ampli cation was the major type of alteration. The ampli cation of LY96 played an important role in tumor invasion associated with poor prognosis. Abnormal expression and dysfunction of the LY96 gene may result from the alterations in chromosomal structure.
In ammation, particularly chronic in ammation, has been proved to be closely associated with the occurrence and progression of cancers [35]. LY96 has a close relation with in ammation and in ammation-related genes, such as TLR4, LY86, TLR2, TLR1, and CD180, etc. In particular, the correlation coe cient between LY96 and TLR4 was up to 0.999. TLR4 involved in various in ammatory pathways and several studies have revealed that the activation of TLR4 was implicated in the initiation and progression of many cancer types, e.g., cancers of the liver, lung, ovary, and stomach, etc [36][37][38][39]. Other Toll-like receptors had similar functions as TLR4. Moreover, overexpression of TLR3 was observed in both primary and metastatic ccRCC [40] and MD2 was critical for generating the in ammatory response. Thus, we speculated that the high expression of LY96 might trigger in ammatory response and eventually lead to cancer.
GSEA of LinkInterpreter module in Linkedomics helped nd signi cant networks of target kinase, miRNAs, and transcription factors. DNA damage causes genomic instability and gradually may contribute to cancer development. Kinase and their related signaling pathways frequently help balance and repair genomic DNA and have become attractive targets for developing new anticancer drugs. Our research discovered that LY96 was most relevant with kinases including LYN, LCK, and SYK. And LYN and LCK are members of the Src family of non-receptor tyrosine kinases. Src is correlated with many oncogenic cellular processes, such as migration, adhesion, invasion, and proliferation, etc. A study has investigated the role of Src family kinases (SFKs) in KIRC [41]. Nuclear accumulations of LYN were observed in solid tumors and associated with inducing cell proliferation and poor prognosis. And SFK inhibitors dasatinib induced apoptosis and inhibited cell proliferation and migration [42]. Thus, the choice of the kinase target is the most vital aspect of drug target identi cation. Development of inhibitors for SFKs may be an appealing novel therapeutic method for KIRC patients. The loss of cell-cycle regulation and disturbance of apoptotic mechanisms may result in the occurrence and progression of cancers. ETS transcription factor (TF) family enhanced tumorigenesis through extensive mechanisms including chromosomal translocations, DNA damage, genome instability, metabolism, and epigenetics [43]. The results showed that LY96 was closely related to ETS family, such as ELF1_Q6, ETS_Q4, ETS1_B, and ETS2_B. Many studies have con rmed that ELF1, ETS1, ETS2 were all recognized as contributors to malignancies [44][45][46]. In addition, other transcription factors like NFKB, PPARA, PEA3, CEBPB, E2F7, SNAI2, MTA2, STAT3, and RAD21 can also be important targets of LY96. LY96 could act through these factors to regulate cell cycle and proliferation ability. Previous studies have reported that TF CREB5 exhibited an inhibition on LY96 [47] while TF STAT1 acted as an activator [48]. The relationship between the two TFs and LY96 in KIRC deserves further investigation.
Signi cant DNA methylation modi cations were observed in LY96 and its expression was positively related to DNMT expression. DNMT inhibitors can induce DNA demethylation and are considered as potential anticancer agents [49]. Demethylation drugs 5-Aza-2'-deoxycytidine (5-Aza) and PBA treatment in KIRC cell lines have been con rmed [50]. And 5-Aza-induced upregulation of miR-200c might inhibit invasion, migration, and EMT in KIRC cells. Therefore, we deduced that the initiation and progression of KIRC may be medicated by DNA modi cation alterations of LY96. To better understand how altered DNA demethylation in uences the development and progression of KIRC will be the focus of future research and more effective demethylating agents are in urgent need. Meanwhile, we found that lower methylation levels of the LY96 genes were more frequently identi ed in Caucasian and advanced stage.
We studied tumor cell in ltration to determine whether various immune cells play a signi cant role in tumor growth. Immunomodulation of the tumor microenvironment is a new cancer therapy and immunotherapy based on immunomodulatory antibodies is becoming a mainstream of oncology. The composition of tumor-in ltrating lymphocytes in the tumor microenvironment can induce host immune responses to tumor cells and impact tumor growth and clinical prognosis [51][52]. Positive associations between various TILs and LY96 gene in KIRC were identi ed through DISIDB and the most relevant cells were macrophage (cor=0.72) and MDSC (cor=0.702) (P<.001). Macrophages are highly related to tumor immunity and often balance modulates cancer and in ammatory disease. Diverse macrophage subsets have been connected with either protective or pathogenic function in cancers. MDSC have potent immune suppressive functions on host immune cells to help cancers escape the immune response. High expression of the LY96 gene may promote the proliferation of macrophages and MDSCs and thereby result in a poor prognosis. Besides, we explored the correlations between lots of immunoinhibitors, Immunostimulators, MHCs, chemokines, receptors, and LY96 gene in KIRC. Among them, LGALS9, TNFSF13B, HLA-DRA, CCL18, and CCR1 displayed tight links withLY96 in KIRC.
LGALS9 encodes Galectin-9 (Gal-9), which has been demonstrated to be relevant to various cancer types. High expression level of Gal-9 was signi cantly associated with poor prognosis of KIRC patients [53]. TNFSF13B was reported to be involved in cell proliferation and apoptosis and correlated with a worse prognosis in cancers [54]. A study proposed that the dysregulation of HLA-DRA (major histocompatibility complex, class II, DR alpha) was relevant to systemic in ammation [55], which may lead to the occurrence of cancers. CCL18 could promote angiogenesis and induce EMT in cancer cells [56]. CCR1 was a crucial facilitator of cancer invasion and metastasis [57]. These immune checkpoints, which have a close relationship with LY96, promoted cell migration, and accelerate cancer progression. This may explain why LY96 has a negative effect on KIRC at the level of the tumor microenvironment. Different tumor-in ltrating lymphocytes varied among diverse KIRC subtypes, as well as in the immune evasion mechanisms. Therefore, we should take these factors into account when developing new strategies for immunotherapy. The results showed that LY96 expression was highest in C6 (TGF-b dominant) while lowest in C5 (immunologically quiet) immune subtypes.
In conclusion, our study showed that LY96 expression levels were strongly correlated with a worse prognosis of KIRC patients and may be a potential marker in KIRC. LY96 was crucial for generating the in ammatory response and signi cantly related to several kinase targets (such as LYN, LCK, SYK, etc) and transcription factors (such as ELF1, NFKB, ETS, etc). To nd demethylation agents and TFs inhibitors that downregulate LY96 is meaningful. Furthermore, LY96 played an important role in the immune microenvironment and may serve as an immunotherapy target in KIRC. Our study used online tools with a large sample size and low cost, which may lead to new directions of research in KIRC. Nevertheless, further experiments are required to validate its diagnostic and therapeutic potential.

Declarations
Ethics approval and consent to participate Not applicable.

Consent for publication
Not applicable.

Availability of data and materials
The datasets generated and/or analysed during the current study are available in the TCGA database (https://tcga-data.nci.nih.gov/tcga).