UGT1A Gene Family Members Serve as Potential Targets and Prognostic Biomarkers for Pancreatic Cancer

Background Pancreatic cancer (PC) is one of the most common cancers worldwide, with high mortality. The UGT1A gene family plays important roles in pharmacology and toxicology, contributing to interindividual differences in drug disposition. However, mRNA expression and prognostic value of the UGT1A gene family in PC have not been identified. Methods Oncomine, GEPIA2, DAVID 6.8, Metascape, Kaplan-Meier plotter, cBioPortal, GeneMANIA, TRRUST v2, TIMER, and R software were used in our study. Results The transcriptional levels of UGT1A1/3/6/8/9/10 in PC tissues were significantly higher than those in normal tissues. These results were further validated using five pairs of PC tumor tissues and adjacent nontumor tissues. A significant correlation was found between the expression of UGT1A1/6/10 and the pathological stage of PC. PC patients with lower transcriptional levels of UGT1A1/4/5/6/10 were associated with a better prognosis. The differentially expressed UGT1A gene family functions were primarily related to the glucuronidation pathway, cytokine-cytokine receptor interactions, and the ILK signaling pathway. Our data suggest that HNF1A, AHR, and CDX2 are key transcription factors for the UGT1A gene family. Furthermore, the expression levels of UGT1A1/3/8/9/10 were positively correlated with the activities of tumor-infiltrating immune cells, especially B cells. The expression levels of UGT1A6/9 were negatively correlated with macrophage infiltration levels. Conclusions These results suggest that the UGT1A gene family could serve as a potential prognostic biomarker and target for PC. However, future studies are required to validate our findings and promote the clinical utility of the UGT1A gene family in PC.


Introduction
Pancreatic cancer (PC) is characterized by poor prognosis, primarily associated with genetic conditions, diabetes, smoking, and obesity [1][2][3]. Pancreatic adenocarcinoma (PAAD) is the most common type of PC. In 2018, PC was the 13th most common cancer globally, with 458,918 new cases, and the 7th most common cause of cancer-related mortality, with 432,242 deaths [4,5]. Early surgical resection is the most effective treatment for PC. Unfortunately, PC is most commonly discovered in the middle and late stages. Recently, numerous researchers have explored the therapeu-tic targets of PC, from gene and mRNA to miRNA [2,6,7]. However, because these are far from sufficient, it is important to explore additional therapeutic targets and prognostic biomarkers for better prognosis of PC.
UDP-glycosyltransferases (UGTs) are a superfamily of enzymes found in diverse species (including animals, fungi, bacteria, and plants). UGTs catalyze the covalent addition of sugars from nucleotide UDP-sugar donors to functional groups in a broad range of lipophilic molecules [8]. In mammals, the superfamily comprises four families: UGT1, UGT2, UGT3, and UGT8. Among them, the UGT1A gene family (including UGT1A1 and UGT1A3-UGT1A10) has important roles in pharmacology and toxicology, contributing to interindividual differences in drug disposition and cancer risk [8]. Additionally, cumulative evidence has revealed that the UGT1A gene family showed increased expression or activation in several human malignancies [9,10] and can be induced by pathways that sense demand for detoxification and modulation of endobiotic signaling molecules [8,11]. However, there are few related studies on the UGT1A gene family as potential therapeutic targets or prognostic biomarkers in PC.
Recent studies have reported the general expression profile and function of some UGT1A gene family members in PC [12] and other cancers [13]; however, screening suitable UGT1A gene family members as therapeutic targets or prognostic biomarkers for PC is a substantial challenge, which urgently needs to be addressed. Fortunately, with the development of secondgeneration gene sequencing technology and expansive database platforms, a comprehensive analysis of the UGT1A gene family members in patients with PC is possible.
In this study, we used several large public databases to perform a comprehensive analysis of the relationships between the UGT1A gene family members and the pathogenesis, tumor-infiltrating immune cells, and progression of PC. Consequently, we hope to aid clinicians in selecting appropriate therapeutic drugs and more accurately predicting long-term outcomes in patients with PC.

Materials and Methods
The main methods in this study followed the methods of previous studies [14,15].
2.1. Oncomine. The Oncomine platform (http://www .oncomine.org) is a publicly accessible online cancer microarray database with a powerful set of analysis functions that compute gene expression signatures, clusters, and gene set modules, automatically extracting biological insights from the data [16,17]. In this study, analyses of mRNA levels in the UGT1A family in PC and a normal control dataset were performed using Oncomine. The parameter settings were restricted as follows: P value = 0.05, fold change ðFCÞ = 2, and data type = mRNA.

Clinical
Samples. Five pairs of fresh PC specimens and adjacent nontumor tissues were collected from the Zhujiang Hospital, Southern Medical University (Guangzhou, China).
The human specimens used for validation in this study were collected between May and June 2021. The local ethics committee approved the use of these specimens. The adjacent samples were taken at a distance of at least 5 cm from the tumor.

BioMed Research International
Genotype-Tissue Expression (GTEx) dataset projects, which provide customizable functions, such as tumor/normal differential expression analysis, profiling according to cancer type or pathological stage, patient survival analysis, similar gene detection, correlation analysis, and dimensionality reduction analysis [18]. In our study, the expression analysis of the UGT1A family in PC was performed using the expression DIY module of GEPIA2, and the parameter settings were restricted as follows: log 2 FC = 1, P value = 0.05, use = log 2 ðTPM + 1Þ for log scale, jitter size = 0:04, and match = TCGA normal and GTEx data.
2.5. DAVID 6.8. DAVID 6.8 (https://david.ncifcrf.gov/home .jsp) is a comprehensive, functional annotation website that helps investigators better clarify the biological function of submitted genes [19]. In this study, the Gene Ontology (GO) enrichment analysis (including biological processes (BP), cellular components (CC), and molecular function (MF)) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of the UGT1A family were performed using DAVID 6.8.
2.6. Metascape. Metascape (http://metascape.org) is a free, well-maintained, user-friendly gene list analysis tool for gene annotation and analysis [20]. It is an automated metaanalysis tool used to understand common and unique pathways within a group of orthogonal target-discovery studies. In this study, Metascape was used to analyze proteinprotein interaction (PPI) and molecular complex detection (MCODE) in the UGT1A family.

Kaplan-Meier
Plotter. The Kaplan-Meier plotter (http:// www.kmplot.com) is an online database containing microarray gene expression data and survival information of cancer patients derived from Gene Expression Omnibus, TCGA, and the Cancer Biomedical Informatics Grid [21]. In this study, the prognostic value of the mRNA expression of UGT1A family members was evaluated using the Kaplan-Meier plotter. The overall survival (OS) and relapse-free     BioMed Research International pipeline [22]. The frequency of UGT1A family alterations (amplification, deep deletion, and missense mutations), copy number variance obtained from Genomic Identification of Significant Targets in Cancer (GISTC), and the survival between the altered and nonaltered groups was obtained according to the online instructions of cBioPortal.
2.9. GeneMANIA and R Software. GeneMANIA (http://www .genemania.org) is a website that provides information on physical interactions, pathways, coexpression, colocalization, and shared protein domain similarity of submitted genes [23]. In this study, we used GeneMANIA to predict the top 20 genes closely related to the UGT1A family for coexpression, physical interactions, pathways, and shared protein domain similarity of the UGT1A family and then visualized using R software.
2.10. TRRUST v2. TRRUST v2 (https://www.grnpedia.org/ trrust/) is a manually curated database of human and mouse transcriptional regulatory networks, which includes 8444 and 6552 transcription factor-(TF-) target regulatory relationships of 800 human TFs and 828 mouse TFs, respec-tively [24]. In this study, we used TRRUST v2 to explore the TF targets of the UGT1A gene family in patients with PC.
2.11. TIMER 2.0. TIMER 2.0 (http://timer.cistrome.org/) is a comprehensive resource to systematically analyze immune infiltrates across diverse cancer types [25,26]. In our study, we assessed the expression levels of UGT1A gene family members in PC and their correlations with tumor purity and infiltrating immune cells, including CD8+ T cells, CD4 + T cells, B cells, macrophages, and neutrophils.

BioMed Research International
Further, analysis of the expression levels of the UGT1A gene family in PC and normal tissues using GEPIA 2 showed that the transcriptional levels of UGT1A1, UGT1A3, UGT1A4, UGT1A7, UGT1A8, and UGT1A9 in PC tissues were elevated when compared with normal tissue, although there were no significant differences. Notably, the transcriptional levels of UGT1A6 (P < 0:05) and UGT1A10 (P < 0:05) in PC tissues were significantly higher than those in normal tissues (Figure 1). We also compared the relative expression levels of the U UGT1A gene family in PC tissues and found that among all UGT1A gene fam-ilies evaluated, the relative expression of UGT1A10 was the highest (Supplementary 2).
We further assessed the correlation between the expression of the UGT1A gene family and the pathological stage of PC in patients and found a significant correlation between the expression of UGT1A1 (P = 0:00386), UGT1A6 (P = 0:00378), UGT1A10 (P = 0:00244), and pathological stage ( Figure 2). The expression of UGT1A1, UGT1A6, and UGT1A10 increased as the tumor progressed, suggesting that the UGT1A gene family plays an important role in the tumorigenesis and progression of PC.

Verification of the UGT1A Gene Family in Clinical
Samples. We used qRT-PCR to assess the expression of the UGT1A gene family in five pairs of fresh tissues from PC patients and adjacent nontumor tissues. The results showed that the transcript levels of UGT1A1/3/6/9/10 were frequently higher (P < 0:05) in PC tissues than in the corresponding nontumor tissues (Figure 3).

Functional Enrichment Analysis of UGT1A Gene Family
Members in Patients with PC. DAVID 6.8 was utilized to analyze the functions of differentially expressed UGT1A gene family members. The UGT1A gene family members were mainly enriched in glucuronidation, metabolic regulation of BP such as xenobiotic glucuronidation, flavonoid glucuronidation, metabolic processes, negative regulation of glucuronosyltransferase activity, negative regulation of cellular glucuronidation, negative regulation of fatty acid metabolic processes, cellular glucuronidation, flavonoid biosynthetic processes, flavone metabolic processes, and retinoic acid metabolic processes. The integral components of the membrane, endoplasmic reticulum, and endoplasmic reticulum membrane were the most highly enriched items in the CC category. In the MF category, glucuronosyltransferase activity, retinoic acid binding, enzyme binding,    protein heterodimerization activity, protein homodimerization activity, transferase activity, transferring hexosyl groups, enzyme inhibitor activity, steroid binding, protein kinase C binding, drug binding, and transferase activity were the 10 most highly enriched items. KEGG pathway analysis showed that among the top 10 KEGG pathways, ascorbate and aldarate metabolism, pentose and glucuronate interconversions, porphyrin and chlorophyll metabolism, drug metabolism-other enzymes, steroid hormone biosynthesis, drug metabolism-cytochrome P450, metabolism of xenobiotics by cytochrome P450, chemical carcinogenesis, and metabolic pathways were significantly associated with the tumorigenesis and progression of PC (Figures 4 and 5 and Supplementary 6).

PPI Network of UGT1A Gene Family Members and MCODE Component Form in the PPI Network.
Furthermore, to better understand the relationship between UGT1A gene family members and PC, we performed a PPI network analysis of differentially expressed UGT1A gene family members using Metascape to explore the potential interactions among them. The results showed that nine nodes and 35 edges were obtained in the PPI network (Figure 6(a)). After pathway and process enrichment analyses were independently applied to each MCODE component, the results showed that biological function was mainly related to glucuronidation and the glucuronate pathways ( Figure 6(b)).

The Prognostic Value of UGT1A Gene Family Members in Patients with PC.
To evaluate the value of differentially expressed UGT1A gene family members in the progression of PC, we assessed the correlation between differentially expressed UGT1A gene family members and clinical outcomes using GEPIA 2. The value of differentially expressed UGT1A gene family members in the OS of patients with PC was evaluated. We found that PC patients with high transcriptional levels of UGT1A1 (P = 0:0043), UGT1A4 (P = 0:0017), UGT1A5 (P = 0:02), UGT1A6 (P = 0:0036), and UGT1A10 (P = 0:00046) were significantly associated with shorter OS (Figure 7). RFS was also assessed, and the results showed that patients with PC with high transcriptional levels of UGT1A1 (P = 0:015), UGT1A4 (P = 0:025), and UGT1A10 (P = 0:0083) were significantly associated with shorter RFS (Figure 8).

TF Targets of UGT1A Gene Family Members in Patients
with PC. Furthermore, we explored possible TF targets of the differentially expressed UGT1A gene family members using the TRRUST database. The results showed that UGT1A1, UGT1A4, UGT1A6, UGT1A7, UGT1A8, UGT1A9, and UGT1A10 were included in TRRUST. Additionally, we found that HNF1A, AHR, and CDX2 are three key TFs associated with the regulation of the UGT1A family. HNF1A is the key TF for UGT1A1, UGT1A7, UGT1A8, UGT1A9, and UGT1A10, AHR for UGT1A1 and UGT1A6, and CDX2 for UGT1A8 and UGT1A10 (Table 2).
3.9. Immune Cell Infiltration of UGT1A Gene Family Members in PC. Finally, because the prognosis of PC may be affected by inflammatory response and infiltrating immune cells, we further evaluated the association between differentially expressed UGT1A gene family members and immune cell infiltration using the TIMER 2.0 database. As shown in Figure 10

Discussion
The UGT1A gene family encodes pivotal enzymes that play an important role in pharmacology and toxicology, contributing to interindividual differences in drug disposition and cancer risk [8]. Accumulating evidence has demonstrated the differential expression of the UGT1A gene family in a total of 28 tissues, including the pancreas [29][30][31][32]. The most important function of the UGT1A gene family is glucuronidation, which provides protection from environmental toxins and contributes to the clearance of a large proportion of commonly used drugs [8]. In this study, we used several large public databases to perform a comprehensive analysis of the relationships between the UGT1A gene family members and the pathogenesis and progression of PC.
First, we explored the mRNA expression of the UGT1A gene family members and their correlation with the pathological stage in PC. We found that the transcriptional levels of UGT1A1/3/6/8/9/10 in PC tissues were significantly higher than those in normal tissues. These results were then validated using five pairs of PC tumor tissues and adjacent nontumor tissues. A significant correlation was found between the expression of UGT1A1/6/10 and the pathological stage of PC. The expression of UGT1A1, UGT1A6, and UGT1A10 increased as the tumor progressed, suggesting that the UGT1A gene family plays an important role in the tumorigenesis and progression of PC.
Second, we explored the function of differentially expressed UGT1A gene family members using GO enrichment analysis and KEGG pathway enrichment analysis. The results showed that the UGT1A gene family members were mainly enriched in glucuronidation and metabolic regulation functions, and the functions of these genes were primarily related to ascorbate and aldarate metabolism, pentose and glucuronate interconversions, porphyrin and chlorophyll metabolism, drug metabolism-other enzymes, steroid hormone biosynthesis, drug metabolism-cytochrome P450, metabolism of xenobiotics by cytochrome P450, chemical carcinogenesis, and metabolic pathways. Glucuronidation of drugs via elevation of UGT1As correlates with clinical resistance [33]. Glucuronidation plays an important role in clearing metabolites and drug detoxification [34,35]. A study showed that "turning on" UGT1A gene family activity could be the basis of a multidrug resistance mechanism [36]. These data suggest that the UGT1A gene family members are potential drug therapeutic targets.
Third, to evaluate the value of differentially expressed UGT1A gene family members in the progression of PC, we assessed the correlation between the differentially expressed UGT1A gene family and clinical outcomes. The results showed that patients with lower transcriptional levels of UGT1A1/4/5/6/10 were associated with a significantly better prognosis, whereas patients with high transcriptional levels of UGT1A1, UGT1A4, and UGT1A10 were significantly associated with shorter RFS. A study showed that the UGT1A gene family plays an important role in the treatment of modified FOLFIRINOX, improving the long-term survival of patients with PC [12]. These data suggest that the UGT1A gene family is a potential prognostic target in patients with PC.
Because multiple UGT1A gene family members were significantly differentially expressed in PC, we explored the genetic alteration and carcinogenic mechanism of the UGT1A gene family members. The genetic alterations in UGT1A gene family members in PC patients varied from 0.4 to 1.1% for individual genes. Studies have shown that the genetic alteration of UGT1A gene family members reduces the ability of UGT to use UDP-GlcUA and CYP3A4-mediated enhancement of catalytic activity [37].

BioMed Research International
Furthermore, evaluation of the OS, DFS, and PFS among the altered and unaltered groups did not reveal any significant differences.
We also sought to characterize the TF targets of the differentially expressed UGT1A gene family members and found that HNF1A, AHR, and CDX2 are the three key TFs associated with the regulation of the UGT1A gene family.
Recently, an increasing number of studies have shown that the tumor immune microenvironment is of great importance in predicting clinical outcomes and developing immunotherapy in pancreatic cancer [38]. Xu et al. combined the analysis of macrophages and immune checkpoints as an enhanced indicator of survival in patients with PC, implicating the value of the combination therapy [39]. Our results showed that the expression of the UGT1A gene family was negatively correlated with the number of infiltrating macrophages. B cells are special features of pancreatic tumors [40], and studies have shown that a high B cell infiltrate is associated with a better prognosis in patients with PC [41,42]. Importantly, our results also showed that the expression levels of UGT1A1/3/8/9/10 were positively correlated with the activities of tumor-infiltrating immune cells, especially B cells. These results indicate that the UGT1A gene family may noticeably correlate with immune cell infiltration, which may help us better understand the immune microenvironment of patients with PC.
Our study has some limitations. Analysis at the transcriptional level can reflect some aspects of immune status, but not global changes. Moreover, another independent cohort and in vitro or in vivo studies should be performed to validate our results.

Conclusions
Overall, our results suggest that the UGT1A gene family may serve as a potential target and prognostic biomarker for patients with PC. However, future studies are required to validate our findings and thus promote the clinical utility of the UGT1A gene family in PC.

Data Availability
Data in this study can be acquired from the corresponding author on reasonable request and from supplementary files.

Conflicts of Interest
None of the authors have any conflicts of interest.