Polypeptide N-Acetylgalactosaminyltransferase 14 (GALNT14) as a Chemosensitivity-Related Biomarker for Osteosarcoma

Purpose Osteosarcoma is the most common primary bone tumor. Polypeptide N-acetylgalactosaminyltransferase 14 (GALNT14), a member of the N-acetylgalactosaminyltransferase family, has been considered to be associated with various cancers. However, its role in osteosarcoma remains unknown. Here, we aimed to explore the expression and potential mechanism of GALNT14 in osteosarcoma through bioinformatics analysis and in vitro experiments. Methods We investigated GALNT14 expression in osteosarcoma using GEO, the TIMER database, and clinical samples. Protein-protein interaction (PPI) network analysis on GALNT14 was performed by STRING. TARGET was used to identify differentially expressed genes (DEGs) between high and low GALNT14 expression. The correlation between GALNT14 and cuproptosis-related genes in osteosarcoma was analyzed by R language. The prognostic significance of GALNT14 was examined by Kaplan–Meier survival analysis. Additionally, we inhibited GALNT14 function in an osteosarcoma cell line by transfecting siRNA and subsequently explored the effect on drug sensitivity by CCK-8, clonogenic assay, and flow cytometry. Results GALNT14 was significantly elevated in osteosarcoma tissue, osteosarcoma cell lines, and metastatic osteosarcoma. PPI analysis revealed that GALNT14 was associated with MUC7, MUC13, MUC5AC, C1GALT1, MUC15, MUC16, MUC1, MUC4, MUC21, and MUC17. In the high GALNT14 expression group, we discovered 81 upregulated DEGs and 73 downregulated DEGs. Functional enrichment analysis of DEGs showed significant enrichment in the Wnt, TGF-β, Hippo, PI3K signaling pathways and cell adhesion molecules. Expression of cuproptosis-related genes was closely related in osteosarcoma, and GALNT14 expression was significantly positively correlated with FDX1, a key regulator of cuproptosis. Kaplan–Meier survival showed that GALNT14 was linked to poor overall survival and disease-free survival in osteosarcoma. In vitro experiments suggested that GALNT14 was associated with chemotherapy resistance in osteosarcoma. Conclusion We identified a GALNT family gene, GALNT14, that was highly expressed in osteosarcoma. This gene was closely associated with metastasis, progression, cuproptosis-related genes, and chemosensitivity of osteosarcoma, and showed correlation with poor overall survival and disease-free survival in osteosarcoma.


Introduction
Osteosarcoma, the most prevalent primary bone cancer, mostly occurs in children and adolescents, with the second highest incidence among older adults [1].Multiagent chemotherapy combined with surgical resection is the standard treatment for patients with localized disease, and it results in long-term survival rates of approximately 70% [2].Metastasis contributes to more than 90% of all cancer-related deaths, including osteosarcoma [3].Te lung is the site of more than 75% of osteosarcoma metastases, and most deaths from osteosarcoma are attributed to lung metastases.It is well understood how tumor cell dispersion and metastatic colonization of distant sites physically occur [4].Wholegenome sequencing has elucidated many genes responsible for the metastatic progression of osteosarcoma.However, the underlying mechanisms are not well defned.Understanding the underlying mechanisms of the progression and metastasis of osteosarcoma may help to improve the prognosis.
Te enzymes that make up the Nacetylgalactosaminyltransferase (GALNT) family add an N-acetylgalactosamine (GalNAc) to a threonine or serine residue of a mucin-type protein to start the process of Oglycosylation [5].Tomsen-nouvelle (Tn) antigens, which are well-knowncancer-associated molecules, are synthesized largely as a result of this process [6].Tere are 20 members in the GALNTfamily, ranging from GALNT1 to GALNT14 and from GANLTL1 to GANLTL6 [5].Te modifcation of Oglycosylation by GALNTs may impact a variety of biological functions associated with cancer, including tumor development, proliferation, and migration [7].
Te GALNT14 gene, which is over 228 kb in length, is located on chromosome 2p23.1.Te 552-amino acid type II membrane protein GALNT14 has a catalytic domain, a transmembrane domain, a stem region, and an N-terminal cytoplasmic domain [6].GALNT14 has been found in various human tissues since it was frst discovered in the gastric cancer cell line MKN45 in 2003 [8].Further, the roles of GALNT14 in many malignancies have been identifed, including the modifcation of migration characteristics, transformation of tissue invasiveness, and change in apoptotic signaling [9].Clinically, GALNT14 has been proposed as a biomarker for anticancer therapy and prognosis [9].However, the function of GALNT14 in osteosarcoma is still unclear.
In this study, we analyzed the expression of GALNT14 in osteosarcoma using expression data found in public databases.Ten protein-protein interaction (PPI) analysis, GO/ KEGG enrichment analysis, and correlation analysis of cuproptosis-related genes preliminarily revealed the potential function of GALNT14.Finally, we assessed the signifcance of GALNT14 in chemosensitivity and prognosis in osteosarcoma.

Diferential Expression Analysis.
Te microarray transcriptome data were collected from three osteosarcoma datasets (GSE12865, GSE11414, and GSE21257) in the Gene Expression Omnibus (GEO, https://www.ncbi.nlm.nih.gov/geo) database.Te three datasets contained samples from osteosarcoma tumor samples and normal human osteoblasts, osteosarcoma and human osteoblast cell lines, and osteosarcoma patients who developed metastases or not, respectively.GEO2R was used for GALNT14 expression analysis [10].Te Tumor Immune Estimation Resource (TIMER, https://cistrome.shinyapps.io/timer/) is a trustworthy and practical database that ofers extensive gene expression data for a variety of cancer types.Using the DifExp module of the TIMER database, we assessed the levels of GALNT14 in adjacent normal tissues and cancerous tissues across cancers [11].

Functional Enrichment Analysis.
With online PPI data obtained from the STRING database (https://cn.string-db.org/) [12], we examined the GALNT14 PPI network.
Gene expression data of 98 osteosarcoma patients were obtained from the Terapeutically Applicable Research to Generate Efective Treatments (TARGET, https://ocg.cancer.gov/programs/target)database, an open database for childhood cancers.To examine the mRNA that difered in expression between groups (high GALNT14 expression vs. low GALNT14 expression), we used the limma package in the R programming language.Te criterion for the diferential expression of mRNAs was established as "adjusted p < 0.05 and Log2 (fold change) > 1 or Log2 (fold change) ≤1." Te diferential expression data were evaluated by functional enrichment in order to better support the underlying function of GALNT14.Gene Ontology (GO) is a popular approach for annotating genes with functions, particularly molecular function (MF), biological pathways (BP), and cellular components (CC).Te Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis is a useful tool for learning about gene functions and related high-level genome functional data.Te ClusterProfler package (version: 3.18.0) in R was used to examine the GO function and the enriched KEGG pathway of possible targets in order to better understand the carcinogenesis of GALNT14.Te boxplot was created using the R package ggplot2.Te heatmap was created using the R software's pheatmap package [13].

Te Correlations among Cuproptosis-Related Genes and GALNT14.
Te cuproptosis-related genes (CDKN2A, FDX1, DLD, DLAT, LIAS, GLS, LIPT1, MTF1, PDHA1, and PDHB) [14] and GALNT14 expression data were obtained from the TARGET database.A parametric (Pearson) or nonparametric (Spearman) test method was employed for the correlation analysis based on the data normality test.Next, the correlations between cuproptosis-related genes and GALNT14 were displayed via the R software package heatmap.Te R software package circlize (version 0.4.12) was used to display the chord diagram.Te correlation between quantitative variables without a normal distribution was described using Spearman's correlation analysis.Statistics were considered signifcant for p values < 0.05.

Kaplan-Meier Survival Analysis of GALNT14.
Te gene expression and clinical information of 98 individuals with osteosarcoma were retrieved from the TARGET database.Te diference in overall survival (OS) and disease-free survival (DFS) between the high GALNT14 expression group and the low GALNT14 expression group was compared using the log-rank test.p values and the hazard ratio (HR) with a 95% confdence interval (CI) were calculated for Kaplan-Meier curves using log-rank testing and univariate Cox proportional hazards regression.All analyses were performed in R version 4.0.3.Statistics were considered signifcant if p < 0.05.[15].

Cell Viability Assay and Clonogenic Assay.
Te cell counting kit-8 (CCK-8) assay was used to determine the viability of the cells.96-well plates were seeded with MG-63 or U-2 osteosarcoma cells.Following adhesion, the cells were given the indicated concentrations and times of cisplatin or doxorubicin.Each well was then flled with 10 μl of CCK-8 reagent, and each was cultured at 37 °C for 1 hour.Te optical density of the cell lysates at 450 nm was measured in order to calculate the relative number of surviving cells.
For colony formation assays, MG-63 or U-2 osteosarcoma cells were seeded at 500 cells/well in 6-well plates and treated with various concentrations of cisplatin or doxorubicin for 14 days.After that, cells were fxed with 4% paraformaldehyde, followed by three washes in PBS and two washes in ddH 2 O.

Flow Cytometry. Apoptosis was measured using
Annexin V/PI double-staining (KeyGEN BioTECH, China).Briefy, cells were incubated with the required conditions and then washed with PBS and trypsinized to get a single cell suspension.Next, 10 5 cells/ml of suspension was suspended in 100 μL binding bufer and stained with Annexin V/PI in the dark for 30 min.Te analysis was performed with FACSVerse (BD Biosciences) and FlowJo software (Tree Star, United States).

Statistical Analyses.
Each result was independently verifed by at least three separate experiments, and all data were displayed as mean ± standard deviation (SD).To determine whether diferences were statistically signifcant, two-sided Student's t-tests or analysis of variance tests were utilized.Statistics were considered to be signifcant at p < 0.05.

GALNT14 Expression in Osteosarcoma and Pan-Cancer.
GSE12865 provided the expression profles of osteosarcoma tumor samples and normal human osteoblasts.As shown in Figure 1(a), GALNT14 expression was signifcantly elevated in osteosarcoma (p � 0.002).We also observed upregulation of GALNT14 in the osteosarcoma cell line U2OS compared to GALNT14 expression in the human osteoblast cell line HOB (p < 0.001) (Figure 1(b)).Furthermore, osteosarcoma at the metastatic stage showed an increased GALNT14 level compared to that at the nonmetastatic stage (p � 0.004) (Figure 1(c)).Te levels of GALNT14 across cancers were examined using the TIMER database.Compared with adjacent normal tissues, GALNT14 presented with a signifcantly higher expression in the tumor tissues of bladder urothelial carcinoma, cholangiocarcinoma, head and neck squamous cell carcinoma, kidney renal clear cell carcinoma, kidney renal papillary cell carcinoma, lung adenocarcinoma, lung squamous cell carcinoma, and uterine corpus endometrial carcinoma.However, kidney chromophobe, liver hepatocellular carcinoma, and prostate adenocarcinoma had lower GALNT14 expression than adjacent normal tissues (Figure 1(d)).All the p values of pan-cancer analysis were summarized in Supplementary Table 1.

PPI and Diferentially Expressed Genes Associated with
GALNT14.We used the STRING database to analyze the GALNT14 PPI network and revealed a link between GALNT14 and MUC7, MUC13, MUC5AC, C1GALT1, MUC15, MUC16, MUC1, MUC4, MUC21, and MUC17 (Figure 2(a)).We analyzed diferentially expressed genes associated with GALNT14 levels using the osteosarcoma expression profle from the TARGET database, which contains the largest sample size of osteosarcoma.Diferentially expressed genes were represented by a heatmap and a volcano plot (Figures 2(b) and 2(c)).In the high GALNT14 expression group, we discovered 81 upregulated genes and 73 downregulated genes.Te functional enrichment analysis showed signifcant enrichment in pathways related to tumorigenesis and progression, such as the Wnt, TGF-β, Hippo, and PI3K signaling pathways.In addition, we also observed the enrichment of cell adhesion molecules, which may be closely related to tumor metastasis (Figure 3).

GALNT14 Was Associated with Cuproptosis-Related
Genes in Osteosarcoma.We analyzed the correlation of the expression of GALNT14 and 10 cuproptosis-related genes by using the TARGET osteosarcoma dataset.As shown in Figure 4(a) and Supplementary Table 2, the expression levels of genes associated with cuproptosis were strongly correlated with osteosarcoma.Ten the relationship between GALNT14 and cuproptosis-related genes was displayed separately.GALNT14 expression was signifcantly positively correlated with FDX1 (R � 0.33, p < 0.001), a key regulator of copper ionophore-induced cell death (Figure 4(b)).Additionally, there was some association between DLAT and GALNT14 expression (R � 0.17, p � 0.103).Te gene expression distributions are displayed in a dot plot (Figure 4(c))..

GALNT14 Was Associated with Poor OS and DFS in
Osteosarcoma.We frst performed a detailed analysis of the impact of GALNT14 on OS in osteosarcoma.Te correlation between gene expression and survival time and status is depicted in Figure 5(a); high GALNT14 expression was associated with poor OS in osteosarcoma.Kaplan-Meier survival curves also demonstrated this trend (p � 0.0305, HR � 2.074, 95% CI (1.071, 4.016)) (Figure 5(b)).In addition, GALNT14 was also associated with poor DFS in osteosarcoma (p � 0.00416, HR � 2.41, 95% CI (1.32, 4.398)) (Supplementary Figure 1 and Figure 5(c)).

GALNT14 Was Related to Chemosensitivity in
Osteosarcoma.First, we measured the expression of GALNT14 in normal human osteoblasts, drug-sensitive osteosarcoma, and drug-resistant osteosarcoma by RT-qPCR.As shown in Figure 6(a), the expression of GALNT14 was increased in osteosarcoma compared with normal osteoblasts, especially in drug-resistant osteosarcoma.Ten, we explored the efect of GALNT14 on drug sensitivity in vitro.A siRNA was used to inhibit the expression of GALNT14 in the osteosarcoma cell lines MG-6 and U-2 (Figure 6(b)).We found that inhibition of GALNT14 improved sensitivity to the chemotherapeutics cisplatin and doxorubicin in MG-6 and U-2 cell lines by detecting cell proliferation with CCK-8 (Figure 6(c)), and consistent trends were also observed in the clonogenic assay (Figure 6(d)).In addition, as shown in Figure 7, fow cytometry revealed that inhibition of GALNT14 resulted in an increased rate of apoptosis after cisplatin or doxorubicin treatment.

Discussion
Te association between GALNT14 expression and cancer characteristics has been studied in multiple tumors.For example, GALNT14 was found to be overexpressed in most breast cancer tissues and associated with lung metastasis [16,17].Overexpression of GALNT14 was also found in ovarian cancer [15,18].In summary, previous reports have suggested abnormal expression of GALNT14 in cancer and metastatic tissues.Our study revealed for the frst time the expression of GALNT14 in osteosarcoma.Consistently, increased expression of GALNT14 was identifed in osteosarcoma.GALNT14 expression was also elevated in metastatic osteosarcoma compared to the nonmetastatic stage.Our results support the potential application of GALNT14 as a predictive biomarker for the diagnosis and metastasis of osteosarcoma.
Previous studies have suggested the signifcance of GALNT14 for the metastasis of a wide variety of cancers [19][20][21].In this study, in addition to aberrant expression in metastatic osteosarcoma, GALNT14 was also found to interact with mucins (except for C1GALT1) by PPI analysis.Altered mucin glycosylation patterns during malignant transformation have been considered to promote cancer cell diferentiation, proliferation, invasion, and metastasis [22].Additionally, we observed that multiple cancer-related pathways were enriched through functional enrichment analysis of genes associated with GALNT14 levels, such as the Wnt, TGF-β, Hippo, and PI3K signaling pathways.Te Wnt signaling pathway is one of the key cascades regulating development and has also been tightly associated with osteosarcoma [23].TGF-βs have been considered to exhibit protumoral and promigratory efects on osteosarcoma [24].Te Hippo signaling pathway shows a close relationship with osteosarcoma and is a potential therapeutic target in the future [25].Tese results reveal the underlying mechanism by which GALNT14 promotes tumorigenesis and metastasis in osteosarcoma.
Te elimination of superfuous and damaged cells is achieved by the vital process of cell death.Apoptosis, necroptosis, and ferroptosis are a few examples of programmed and nonprogrammed cell death that have been uncovered in previous research [26].Recently, abnormal copper ion elevations have been linked to a previously  6 Journal of Oncology unidentifed death pathway called cuproptosis [27].Terefore, signifcant interest has developed in the association of cuproptosis-related genes with osteosarcoma.Our results showed that the expression of multiple cuproptosis-related genes was closely correlated with osteosarcoma.In addition, GALNT14 is also closely associated with the expression of FDX1, a key gene in cuproptosis.In summary, our results suggest a possible novel mechanism of GALNT14 involvement in osteosarcoma cell proliferation and death.
Prognostic signifcance is currently a hot topic in oncology research.Multiple genes have been determined to be associated with the prognosis of osteosarcoma, and some prognostic models have been established [28,29].However, the prognostic signifcance of GALNT14 in osteosarcoma remains unknown.Previous studies have reported an association between GALNT14 and the prognosis of a variety of cancers, including gastric cancer, cholangiocarcinoma, ovarian cancer, and breast cancer [18,21,30,31].Our       Trough an extensive repertoire of resistance mechanisms, tumors elude targeted cancer therapy.Tis is also one of the main reasons for the poor prognosis of tumors.Osteosarcoma is a kind of disease that is particularly resistant to chemotherapy.Tis disease, according to clinical trials, only responds to high doses of chemotherapy and rapidly acquires resistance [32,33].Chemotherapy resistance osteosarcoma potentially correlates to drug buildup in the cell, DNA damage repair, intracellular detoxifcation, signal transduction, apoptosis, the tumor microenvironment, and immunity [34].GALNT14 serves as an emerging marker for predicting therapeutic outcomes in multiple tumors [9], and it has also been identifed to be related to paclitaxel resistance in lung adenocarcinoma [35].
Here, we reveal for the frst time the signifcance of GALNT14 in the chemosensitivity of osteosarcoma.
Tere are some limitations to this study.For example, the molecular interaction of GALNT14 and the mechanism of GALNT14 on drug sensitivity need further experimental verifcation.In conclusion, we identifed a GALNT family gene, GALNT14, that was highly expressed in osteosarcoma.GALNT14 was closely associated with metastasis, progression, cuproptosis-related genes, and the chemosensitivity of osteosarcoma.Finally, the poor prognostic signifcance of GALNT14 in osteosarcoma was also elucidated.Tis discovery is anticipated to ofer fresh perspectives on osteosarcoma diagnosis and therapy.

Data Availability
Te data used to support the fndings of this study are included within the article.

Figure 2 :
Figure 2: GALNT14 PPI network and coexpression genes in osteosarcoma.(a) PPI network of GALNT14.Diferentially expressed genes associated with GALNT14 level showing in heatmap (b) and volcano plot (c).

Figure 3 :
Figure 3: GO and KEGG enrichment analysis of genes associated with GALNT14 levels in osteosarcoma.

Figure 5 :
Figure 5: Te OS signifcance of GALNT14 in osteosarcoma.(a) Te gene expression, survival time, and OS survival status of the TARGET dataset.Te top scatterplot represents gene expression from low to high.Te scatter plot distribution represents the gene expression of diferent samples corresponding to the survival time and survival status.Te bottom fgure is the gene expression heatmap.(b) OS Kaplan-Meier survival analysis.(c) DFS Kaplan-Meier survival analysis.OS, overall survival; DFS, disease-free survival.