Expression and clinical significance of TYRP1,ABCB5,and MMP17 in sinonasal mucosal melanoma

Abstract

BACKGROUND:

Sinonasal mucosal melanoma (SNMM) is a lethal malignancy with poor prognosis. Treatment outcomes of SNMM are poor. Novel prognostic or progression markers are needed to help adjust therapy.

METHODS:

RNA-seq was used to analyze the mRNA expression of tumor tissues and normal nasal mucosa from primary SNMM patients ( $n=$ 3). Real-time fluorescent quantitative PCR (qRT-PCR) was used to validate the results of RNA-seq ( $n=$ 3), while protein expression was analyzed by immunohistochemistry (IHC, $n=$ 31) and western blotting ( $n=$ 3). Retrospective studies were designed to determine the clinical parameters and the total survival rate, and correlation between the protein expression levels of the most significant key genes and prognosis was analyzed.

RESULTS:

In total, 668 genes were upregulated and 869 genes were downregulated in SNMM (fold change $\geqslant$ 2, adjusted $p$ value $<$ 0.01). Both mRNA and protein expression levels of the key genes in SNMM tumor tissues were higher than those in the normal control nasal mucosal tissues. The expression rates of TYRP1, ABCB5, and MMP17 in 31 primary SNMM cases were 90.32%, 80.65%, and 64.52%, respectively. In addition, age, typical symptoms, and AJCC stage were related to overall survival rate of patients with SNMM ( $p<$ 0.05). Furthermore, the expression of ABCB5 was age-related ( $p=$ 0.002). Compared with individuals with negative ABCB5 expression, those with positive expression exhibited significantly poor overall survival ( $p=$ 0.02).

CONCLUSION:

The expression levels of TYRP1, ABCB5, and MMP17 were significantly upregulated in SNMM tissues, and the expression of ABCB5 was related to poor prognosis in SNMM. Thus, ABCB5 may serve as a progression marker and can predict unfavorable prognosis in patients with SNMM.

Keywords

RNA-seq sinonasal mucosal melanoma mucosal melanoma prognosis TYRP1 ABCB5 MMP17

1. Introduction

Sinonasal mucosal melanoma (SNMM) is a rare but highly aggressive malignancy, usually occurring in patients older than 50 years and running a lethal clinical course [1]. Patients are usually diagnosed at advanced stages, which are characterized by epistaxis, nasal congestion, and polypoid masses, with or without pigment [2]. In addition, treatment outcomes of SNMM are poor, with a 5-year overall survival rate less than 30%, which is nearly 2.5 times lower than that of cutaneous malignant melanoma (MM) [3, 4]. Therefore, we need useful prognostic biomarkers to more accurately diagnose patients who have failed treatment and to support the establishment of new therapeutic approaches.

The uncertain molecular mechanisms underlying disease pathogenesis and complex pathological processes hinder the process of finding a useful prognostic biomarker. Some studies have found different gene mutations in SNMM patients, such as KIT mutation, NRAS mutation, and BRAF mutation [5, 6, 7, 8, 9]. In recent years, targeted therapies and immunotherapy for SNMM have been used in clinical practice; the current treatment options vary greatly among different individuals and have an important influence on treatment outcomes. It is usually necessary to balance tumor treatment and treatment-related side effects [10, 11, 12, 13]. Therefore, it is very important to accurately assess the prognosis of individual patients. A retrospective study based on 27 SNMM patients reported that tumor-infiltrating lymphocytes’ (TILs) density can be used as an important prognostic indicator of SNMM, but clinical studies with larger sample sizes and more in-depth basic studies are still needed to confirm this finding [14]. Previous studies provide possible factors related to the onset of SNMM and its prognosis, but there is a lack of new technologies and methods to screen the molecular biomarkers of SNMM and further verify their relationship with prognosis.

Gene expression plays a vital role in cellular phenotype; thus, a comprehensive classification of gene transcripts can help to elucidate how gene expression determines phenotypic expression. High-throughput sequencing techniques have been applied since 2005 [15], helping us to understand life activities at the molecular level and to conduct detailed research to elucidate the genome and transcriptome [16]. As an essential part of high-throughput sequencing, RNA sequencing (RNA-seq) has been applied to many aspects of tumor research and treatment, including the discovery and characterization of biomarkers of tumor heterogeneity and evolution, drug resistance, tumor immune microenvironment and immunotherapy, and tumor neoantigens [17, 18]. The pathophysiological and molecular characteristics of SNMM are not yet established. Furthermore, there has been no RNA-seq study on SNMM. Moreover, previous studies have never used RNA-seq to accurately analyze differentially expressed genes (DEGs) in SNMM. RNA-seq may provide valuable insights for the discovery and identification of potential new transcriptional biomarkers in SNMM.

The objectives of this study were to assess clinical and histologic characteristics, validate gene expression, and identify possible prognostic factors in SNMM.

2. Materials and methods

2.1 Patients and tissues

The tumor tissue and normal control tissue samples for RNA-seq, qRT-PCR, and western blotting were obtained from three primary SNMM patients. Healthy control nasal mucosal tissues were obtained from the contralateral nasal cavity. SNMM was diagnosed based on the intraoperative freezing or postoperative pathology report signed out by two certified pathologists in the institution. The study protocols were approved by the local ethics committee of the First Affiliated Hospital of Nanchang University (Nanchang, China) [(2020) Medical Research Review no. (013)] and written informed consent was obtained during each preoperative clinic visit.

A retrospective review was performed to analyze data from 31 primary SNMM patients who had undergone endoscopic sinus surgery (ESS) treatments at the First Affiliated Hospital of Nanchang University between June 1, 2010 and June 30, 2020. All patients had negative surgical margins. Overall, 31 patients diagnosed with primary mucosal MM of the nasal cavities or paranasal sinuses and meeting the clinical and histopathological criteria of SNMM were included in the retrospective review. The stage of SNMM was determined in accordance with the AJCC 8th edition guidelines. Clinical data and paraffin blocks were collected for all the patients. Basic clinical demographic data of SNMM patients are presented in Supplementary Table S1.

2.2 Total RNA isolation and rRNA depletion

Total RNA was isolated from the tumor tissues of SNMM patients by using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s instructions, and then, 4–8 $\mu$ g of each RNA sample was depleted of rRNA by using a Ribo-Zero ${}^{\text{TM}}$ Gold kit (Epicentre Biotechnologies, Madison, WI, USA). Quantification of RNA was performed using NanoDrop1000 spectrophotometer (Thermo Fisher Scientific, Wilmington, DE, USA), while the integrity and purity of RNA were evaluated through capillary electrophoresis by using 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Only total RNA samples with RNA Integrity Number values higher than 8 were used for library construction and the next-generation sequencing analysis.

2.3 Library preparation and next-generation sequencing

The cDNA library was constructed in accordance with the manufacturer’s instructions of NEBNext Ultra RNA Library Prep Kit for Illumina (NEB, E7530) and NEBNext Multiplex Oligos for Illumina (NEB, E7500). Briefly, the enriched mRNA was fragmented into approximately 200-nt-long RNA inserts, which were used to synthesize the first-strand cDNA and the second cDNA. Then, end-repair/dA-tail and adaptor ligation were performed on the double-stranded cDNA. The suitable fragments were isolated by using Agencourt AMPure XPbeads (Beckman Coulter Inc.) and enriched by PCR amplification. Finally, the constructed cDNA libraries were sequenced using an Illumina HiSeq 4000 sequencer (Biomarker Technologies, China).

2.4 Quantitative real-time PCR (qRT-PCR)

For quantitative real-time PCR (qRT-PCR), total RNA was extracted using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s instructions. PCR was performed at 37 ${}^{\circ}$ C for 15 min and at 85 ${}^{\circ}$ C for 5 s, and cDNA was then stored at $-$ 20 ${}^{\circ}$ C. Quantitative RT-PCR was carried out on a CFX96 real-time PCR system (Bio-Rad Laboratories, Hercules) using the SYBR Green Premix Ex Taq TM Kit (TaKaRa, Ohtsu, Japan) with 35 cycles at 95 ${}^{\circ}$ C for 5 s and at 60 ${}^{\circ}$ C for 30 s. We selected the nine most significantly upregulated genes for validation, and the primer and probe set for TYRP1, ABCB5, MMP17, POU3F3, MLIP, CYP19A1, EDNRB, TSPAN10,and ADAMTS3 was purchased from Sangong Biotech. Gene expression data analysis included normalization of TYRP1, ABCB5, MMP17, POU3F3, MLIP, CYP19A1, EDNRB, TSPAN10, and ADAMTS3 CT values to the housekeeping gene ( $\beta$ -actin). Relative transcriptional fold changes were calculated as 2 ${}^{-\Delta\Delta Ct}$ . (See Supplementary Table S2 for the primer sequences.)

2.5 Western blotting

Western blotting was performed in a manner consistent with the method described by Wang et al. [19]. Collected tissue samples from various groups were lysed with RIPA buffer with protease inhibitors. Protein quantification of tissues was performed using a BCA assay kit. Samples containing 20 $\mu$ g of protein were loaded and heated to 95 ${}^{\circ}$ C for 5 min, subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis in 10% or 12% gels, and transferred to a polyvinylidene fluoride membrane. The membrane was blocked using 5% nonfat dried milk and incubated at 4 ${}^{\circ}$ C overnight with primary antibodies TYRP1 (1:1000), ABCB5 (1:1000), MMP17 (1:1000), and $\beta$ -actin (1:5000) diluted in Tris-Buffered Saline-0.1% Tween (TBST). After incubation with secondary antibodies, chemiluminescence signals were imaged by using a Tanon 5200 (Tanon Science & Technology Co. Ltd., Shanghai, China); the target protein band signals were quantified and analyzed by ImageJ 1.5.2 software.

2.6 Immunohistochemical staining

Immunohistochemical staining of SNMM tissues was conducted in accordance with the manufacturer’s instructions. Heat-induced antigen retrieval process was used, and deparaffinized sections were treated with 10 mmol/L citrate buffer (pH 6.0) at 95 ${}^{\circ}$ C for 15 min. Immunohistochemistry was performed at 4 ${}^{\circ}$ C overnight with an anti-TYRP1 antibody (clone; Bioss, Beijing, China, dilution 1:200), an anti-ABCB5 antibody (clone; Bioss, Beijing, China, dilution 1:200), and an anti-MMP17 antibody (Clone Bioss, Beijing, China, dilution 1:200); next, the samples were incubated with anti-rabbit antibody (Zhongshan Golden Bridge, Beijing, China) at 37 ${}^{\circ}$ C for 30 min. Subsequently, 3, 3’-diaminobenzidine (DAB, Zhongshan Golden Bridge, Beijing, China) was applied at 20 ${}^{\circ}$ C for 5 min, and the sections were counterstained with hematoxylin. The sections were then dehydrated in gradient alcohol and sealed in neutral resins.

To further evaluate protein expression levels and pathological characteristics in SNMM, we estimated the extent and intensity of immunoreactivity [20, 21]. The extent of immunoreactivity ranged from 0 to 4 according to the positive cells in each microscopic field of view (0, 0%; 1, 1%–25%; 2, 25%–50%; 3, 50%–75%; 4, 75%–100%), and the scores of intensity of staining were as follows: 0, negative staining; 1, weak staining; 2, moderate staining; 3, strong staining.

2.7 Statistical analysis

Statistical analysis was performed with the GraphPad Prism 8 statistical package (GraphPad Software, La Jolla, CA, USA). All data are reported as means $\pm$ standard deviation, unless otherwise specified. A between-group comparison was performed using the $\chi^{2}$ -test or the precision probability method. Kaplan–Meier curves were plotted to show the survival time differences. The resulting $p$ values were adjusted using Benjamini and Hochberg’s approach to control the false discovery rate. A $p$ value or an adjusted $p$ value (‘adj. $p$ ’) $<$ 0.05 was considered statistically significant.

2.8 Data analysis of RNA-seq

To obtain clean data from raw sequencing data in FASTQ format, the following reads were removed: 1) reads with adapter, 2) reads with more than 10% N bases, and 3) reads with less than 50% bases above Q10. After that, the remaining clean data were mapped to reference genome GRCh38 (release 95) with HISAT2 (version 2.0.4) [22]. Transcript abundances were quantified with StringTie (version 1.3.4d) [23]. Gene expression level was normalized using fragments per kilobase of transcript per million fragments (FPKM). For differential analysis, Limma and DESeq2 were employed [24, 25]. In this study, significantly differentially expressed genes (DEGs) were defined as genes with a fold change of $\geqslant$ 2 and an adjusted $p$ value of $<$ 0.01. Principal component analysis (PCA) was performed to validate biological replicates. The gene interaction network was inferred using Cytoscape with STRING database.

2.9 Functional annotation of DEGs using KEGG and GO pathway enrichment analyses

The Gene Ontology (GO) is the most widely used database for defining the properties of gene products, and it covers three domains, namely, biological processes, cellular components, and molecular functions. Kyoto Encyclopedia of Genes and Genomes (KEGG; https://www.kegg.jp/) is one of the most widely used biological databases in the world. To analyze GO and signaling pathway enrichment, differentially expressed genes (DEGs) were uploaded to an online tool, Metascape, a website for gene annotation, visualization, and attributes (https://metascape.org/) [26].

2.10 The cancer genome atlas and genotype-tissue expression databases

The gene expression data of pan-cancer cases were downloaded from the TCGA (https://portal.gdc.cancer .gov/) and GTEx (https://gtexportal.org/home/datasets) databases and were analyzed to determine TYRP1, ABCB5, and MMP17 expression levels in skin cutaneous melanoma (SKCM). In total, 15,776 tissue profiles were downloaded from the XENA-TCGAGTEx database. In total, 1282 cases of TCGA SKCM RNASeq-TPM, including 813 cases of normal tissues and 469 cases of SKCM tissues, were obtained. TYRP1, ABCB5, and MMP17 expression in SKCM tissues was analyzed for TCGA-derived data.

3. Results

Figure 1.

Differentially expressed genes in the control and SNMM groups, as detected by the RNA-seq analysis. (A) Volcano plot of DEGs (adjusted $p<$ 0.01) identified from the RNA-seq libraries of the control and SNMM groups; each gene is represented by one point on the graph; and 668 upregulated genes (red) and 869 downregulated genes (green) were found (up $=$ total upregulated genes, down $=$ total downregulated genes). (B) Hierarchical clustering analysis of top 30 upregulated DEGs (adjusted $p<$ 0.01), with color intensity signifying the magnitude of change. (C) Principal component analysis (PCA) of obtained transcriptomes. (D) Topological gene–gene interaction network analysis of 50 significantly upregulated genes.

3.1 The differences in gene expression between two groups and analysis

We compared the gene expression levels in the control and SNMM groups by using LIMMA software (moderate t-statistics test, screening condition: $|\text{Fold change}|\geqslant$ 2, adjusted $p$ value $<$ 0.01). Specifically, 1537 DEGs were identified, compared with the control group, including 668 upregulated genes and 869 downregulated genes in the SNMM group (Fig. 1A). Hierarchical clustering analysis showed the expression profiles of the top 30 upregulated DEGs (Fig. 1B). The obtained transcriptomes were subjected to PCA (Fig. 1C). In addition, a topological gene–gene interaction network analysis of 50 significantly upregulated genes is shown in Fig. 1D.

3.2 Functional annotation of DEGs using KEGG and GO analyses

The data of DEGs were uploaded to the Metascape website to identify GO terms and KEGG pathways. In this study, the terms were considered significant and grouped into clusters only when the $p$ -value was lower than 0.01 and the number of enriched genes was at least 3, respectively, based on similarities of membership. We found that the upregulated DEGs were mainly enriched in GO:0051129 (negative regulation of cellular component organization), GO:0048812 (neuron projection morphogenesis), and GO:1903047 (mitotic cell cycle process). Downregulated DEGs were mainly enriched in GO:0030855 (epithelial cell differentiation), GO:0035082 (axoneme assembly), and GO:0003341 (cilium movement) (Fig. 2A). Each node represented an enriched term and was colored by its cluster ID (Fig. 2B) and then by its $p$ -value (Fig. 2C).

3.3 Relative expression of the key genes in SNMM

Figure 2.

Enrichment analysis of DEGs using Metascape. (A) Heat map of enriched terms across input DEG lists, colored by $p$ -values. (B) Network of enriched terms colored by cluster identity. (C) Network of enriched terms colored by $p$ -value, where terms containing more genes tend to have a more significant $p$ -value.

We selected the nine most significantly upregulated genes for validation, namely, TYRP1, ABCB5, MMP17, POU3F3, MLIP, CYP19A1, EDNRB, TSPAN10 and ADAMTS3, which might be associated with the development of SNMM, through the above-mentioned DEG analysis and based on previous studies. Then, qRT-PCR was performed to validate the RNA-seq results of these genes. The results showed that all of the above key genes were expressed at significantly higher levels in the tumor tissues than in that of normal control nasal mucosal tissues, which was consistent with the RNA-seq results (Fig. 3A). In addition, TYRP1, ABCB5 and MMP17 showed nearly 10-fold upregulation in the SNMM group ( $p<$ 0.01), thereby notably surpassing the rest of the genes. We further determined the translation difference of these key genes using western blotting.

The results showed only baseline expression of TYRP1, ABCB5, and MMP17 in the extracts of health tissues (Fig. 3B). However, tumor tissue extracts from SNMM patients showed higher protein expression levels of TYRP1, ABCB5, and MMP17 ( $p<$ 0.01, $p<$ 0.05, and $p<$ 0.01, respectively) (Fig. 3C). Overall, nine key genes that might be associated with the development of SNMM were screened, and their upregulation of mRNA was validated with qRT-PCR. Three genes with larger difference between the control and SNMM groups were selected for validation at the protein level and for further study. Moreover, Pan-cancer data from multiple databases showed that TYRP1, ABCB5, and MMP17 were abnormally expressed in pan-cancer tissues (Supplementary Fig. S1A–C). The expression levels of TYRP1, ABCB5, and MMP17 were obviously increased in melanoma tissues from SKCM when compared with control group (Supplementary Fig. S1D–E).

3.4 The IHC analysis of TYRP1, ABCB5 and MMP17 as well as their relationship with clinicopathological features

Table 1
Immunohistochemical analysis of TYRP1, ABCB5, and MMP17 expression in tumor tissues of SNMM patients

Project ( $n$ )	( $-$ )	( $+$ )	( $++$ )	( $+++$ )	Positive rate (%)
TYRP1	3	1	4	23	90.32
ABCB5	6	8	6	11	80.65
MMP17	11	5	9	6	64.52

Figure 3.

Validation analysis by using qRT-PCR and western blotting. (A) qRT-PCR analysis of TYRP1, ABCB5, MMP17, POU3F3, MLIP, CYP19A1, EDNRB, TSPAN10, and ADAMTS3 in normal control nasal mucosal tissues and SNMM tumor tissues. qRT-PCR was performed by using gene-specific primers; the fold change in gene expression was calculated by using the $\Delta\Delta Ct$ method. (B) Blots of TYRP1, ABCB5, MMP17, and $\beta$ -actin in normal control nasal mucosal tissues and SNMM tumor tissues. (C) Semiquantitative analysis of proteins; the relative expression of each protein normalized to $\beta$ -actin expression. NC: normal control nasal mucosal tissues extracted from the contralateral nasal cavity of SNMM patients. SNMM: tumor tissues extracted from SNMM patients. ${}^{*}p<$ 0.05, ${}^{**}p<$ 0.01, ${}^{***}p<$ 0.001.

Table 2

Relationship between the TYRP1, ABCB5, and MMP17 expression levels and the clinicopathological features of SNMM patients

Parameters	$n$	TYRP1		$p$	ABCB5		$p$	MMP17		$p$
		Positive	Negative		Positive	Negative		Positive	Negative
Total	31	28	3		25	6		20	11
Age (years)				0.56			0.002			1.00
$<$ 60		11	2		7	6		8	5
$\geqslant$ 60		17	1		18	0		12	6
Gender
Male		17	0	0.08	15	2	0.37	11	6	1.00
Female		11	3		10	4		9	5
Pigment				1.00			1.00			0.02
Yes		17	2		15	4		9	10
No		11	1		10	2		11	1
Tumor location				1.00			0.31			1.00
Nasal cavity		22	3		19	6		16	9
Paranasal sinus		5	0		5	0		4	2
AJCC stage				1.00			1.00			0.45
III		17	2		15	4		11	8
IVa–IVb		11	1		10	2		9	3

Figure 4.

Immunohistochemical staining of TYRP1, ABCB5, and MMP17 in SNMM tumor tissues. (A) No pigment-type HE; (B) pigment-type HE; (C) negative control (IgG is the primary antibody); (D) TYRP1 expression mainly in the nuclear membrane and cytoplasm; (E) ABCB5 expression mainly in the cell membrane and cytoplasm; (F) MMP17 expression mainly in the extracellular matrix ( $\times$ 400). HE, hematoxylin and eosin.

According to the results of IHC staining, the positive expression rate was 90.32% (28/31) for TYRP1, 80.65% (25/31) for ABCB5, and 64.52% (20/31) for MMP17 (Table 1). We found that TYRP1 was mainly expressed in the cytoplasm and nuclear membrane; the ABCB5 was expressed in the cell membrane and cytoplasm; and the MMP17 was expressed in extracellular matrix (Fig. 4D–F). In addition, we found that the positive ABCB5 expression was associated with age, and there was a higher positive expression rate in patients older than 60 years than in those younger than 60 years ( $p=$ 0.02). MMP17 expression was associated with pigment, as there was a higher expression rate in patients without pigment than in those with pigment ( $p=$ 0.02). The positive TYRP1 expression had no obvious correlation with age, sex, pigment, tumor location, and stage ( $p>$ 0.05) (Table 2).

3.5 Relationship between ABCB5 and MMP17 protein expression levels and survival rate of patients with SNMM

We analyzed the relationship between ABCB5 and MMP17 protein expression levels and survival rate in patients with SNMM. The median survival time in individuals with ABCB5 positive expression was 18 months, and the 1-, 3-, and 5-year survival rates of these patients were 82.29%, 19.66%, and 13.11%, respectively. In addition, the median survival time in individuals without ABCB5 expression was 37 months, and the 1-, 3-, and 5-year survival rates were 77.19%, 22.51%, and 15.01%, respectively. The overall survival time of the two groups was statistically significant ( $p=$ 0.002) (Fig. 5A).

Figure 5.

The relationship between the expression levels of ABCB5 and MMP17 and survival rate of SNMM patients. (A) ABCB5 and the overall survival of SNMM patients. $+$ , Censored. $p=$ 0.02 (compared between ABCB5-positive and ABCB5-negative expression). (B) MMP17 and the overall survival of SNMM patients. $+$ , Censored. $p=$ 0.09 (compared between MMP17-positive and MMP17-negative expression).

The median survival time in individuals with MMP17 positive expression was 14.5 months; the 1-, 3-, and 5-year survival rates of these patients were 77.19%, 22.51%, and 15.01%, respectively. The median survival time in individuals without MMP17 expression was 34 months; the 1-, 3-, and 5-year survival rates were 100%, 43.75%, and 29.17%, respectively. The difference in overall survival time between the two groups was not statistically significant ( $p=$ 0.09) (Fig. 5B).

4. Discussion

SNMM is a rare and highly aggressive malignancy with poor prognosis. There is currently a lack of evidence-based therapy guidelines that can improve outcomes in SNMM. Thus, diagnosing SNMM early and applying appropriate treatments are critical to reduce mortality and improve the prognosis of patients with SNMM.

This is the first study using RNA-seq to identify DEGs in SNMM tumor tissues and normal nasal mucosal tissues. A total of 1537 DEGs were detected, of which 688 were upregulated and 869 genes were downregulated in the SNMM group. In our study, we screened nine key genes associated with the development of SNMM by aggregating published literature and the abovementioned DEG analysis; we further confirmed the three most potential differential genes through qRT-PCR and western blotting analysis. We found that TYRP1, ABCB5, and MMP17 were extremely highly expressed at both mRNA levels and protein levels in the SNMM group compared with the control group. In addition, the expression levels of TYRP1, ABCB5, and MMP17 were abnormally expressed in pan-cancer tissues and melanoma tissues from SKCM. These obtained results were consistent. IHC analysis was used to identify the relationship between the expression of the key genes and prognosis of SNMM patients.

TYRP1 is a tyrosinase that is involved in pigmentation; its expression is stimulated by melanocyte inducing transcription factor (MITF) [27]. MITF plays an important role in melanocytes in activating pigmentation, cell proliferation, and differentiation [28]. A previous study has also shown that increased TYRP1 activity in metastatic melanoma leads to adverse clinical outcomes and is related to melanoma development [29]. In this study, we examined TYRP1 protein expression in paraffin sections of SNMM tissues by IHC and showed that the positive expression rate of TYRP1 in SNMM was as high as 90.32% (28/31). However, we were not able to analyze the relationship between the expression rate of TYRP1 and prognosis owing to the sample limitation. In SNMM, TYRP1 was observed to be located mainly at the nuclear membrane and cytoplasm, suggesting that it is a functional protein. Bolander et al. [30] have found that most melanoma tissues and normal melanocytes express TYRP1, and its expression negatively correlates with disease stage, but there is no significant correlation between TYRP1 protein expression and overall survival. In this study, there were no significant associations between the expression rate of TYRP1 and gender, age, clinical stage, clinical tumor site, and pigment content ( $p>$ 0.05). We did not find the prognostic value of TYRP1 in SNMM, and this result of our study is inconsistent with findings reported in the previous studies [31, 32]. Therefore, the specific mechanism of TYRP1 in the development of SNMM is unclear, and further research is needed.

ABCB5 is mainly localized on the cell membrane; it acts as a drug efflux pump and mediates the cell membrane transportation of specific substances [33]. The expression of ABC transport family has been confirmed in many cancer types, and the positive expression of ABC transport proteins is associated with poor prognosis [34]. We found that the positive expression rate of ABCB5 was related to age ( $p<$ 0.01), given that patients with SNMM older than 60 years had a higher positive expression rate of ABCB5 than patients younger than 60 years; however, no significant relationship was found between ABCB5 positivity and other variables, such as gender, clinical stage, clinical tumor site, and pigment content ( $p>$ 0.05). In addition, we found that age was related to the overall survival rate of SNMM ( $p<$ 0.01), and patients with SNMM older than 60 years had a poor prognosis. This finding is similar to the age distribution of ABCB5 positivity. A recent study has shown that ABCB5 is a key factor in promoting the metastasis of SKCM; compared with the ABCB5 ${}^{-}$ melanoma subgroup, ABCB5 ${}^{+}$ malignant melanoma initial cells (MMICs) had a higher metastasis potential, and inhibition of the ABCB5 expression helped to reduce the migration and invasion of melanoma cells in vitro [35]. In this study, the positive expression rate of ABCB5 was 80.65% (25/31). The difference in overall survival time between the two groups was statistically significant ( $p=$ 0.02). It indicates that ABCB5 ${}^{+}$ SNMM patients have poorer prognosis than ABCB5 ${}^{-}$ SNMM patients. These findings suggest that the accumulation of ABCB5 plays a vital role in the development of SNMM. Indeed, our findings are similar to previous studies on other types of malignant tumors [36, 37, 38, 39]. However, the regulatory mechanisms underlying the expression of ABCB5 are still at an early research stage, and further studies are needed.

Matrix metalloproteinases (MMPs) are the main enzymes participating in extracellular matrix remodeling, degradation, and angiogenesis during tumor invasion and metastasis. The overexpression of MMPs is related to poor clinical prognosis in patients with malignant tumors [40]. A previous study has explored the relationship between the expression levels of MMP17 and MMP25 and the clinical and pathological characteristics in stomach cancer. It showed that the expression levels of MMP17 and MMP25 were significantly associated with the depth of tumor invasion and lymph node metastasis, regardless of age, gender, tumor wall invasion, clinical tumor site, and clinical stage [41]. Similar result was found in breast cancer; specifically, IHC analysis showed that MMP17 was expressed at a high level in breast cancer tissues, whereas normal breast tissues exhibited negative or weak MMP17 staining [42]. In this study, we found that MMP17 had a high positive expression rate (64.52%) in SNMM. In addition, we found that MMP17 expression was associated with pigmentation, that is, SNMM without pigmentation had a higher positive expression rate of MMP17 than SNMM with pigmentation ( $p=$ 0.02). However, based on the existing sample sizes, we did not find a significant relationship between the expression of MMP17 and overall survival time in patients with SNMM. To date, tissue inhibitors of metalloproteinases are considered potential therapeutic targets for cancer treatment. However, from our data, we did not confirm that patients with SNMM would benefit from such a treatment.

Despite rigorous bioinformatics analysis and validation in our study, there are still some limitations. First, the limitation of sample size might have led to some deviations in the results. Second, this study only conducted experimental verification with a small number of clinical samples; a larger clinical sample size and in vivo animal models will be needed to validate and possibly extend these findings. Third, we extended our analysis to clinical data sets, but we only found that ABCB5 was related to poor prognosis. Therefore, the in-depth mechanism of upregulation of these genes in SNMM remains to be elucidated, and future research is needed to clarify the mechanisms by which increased ABCB5 expression is related to the development of SNMM.

5. Conclusion

SNMM is a highly specific form of MM with poor prognosis. Low incidence and complex molecular mechanism hinder sufficient pathophysiological knowledge to allow effective treatment. Our study further validated the expression of key genes based on the RNA-seq analysis. Based on our results, ABCB5 might be associated with poor prognosis in SNMM, especially in patients over 60 years.

Ethical statement

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards. The trial protocols were approved by the Ethics Committee of the First Affiliated Hospital of Nanchang University (IIT 2020 No.013).

Funding

This study was supported by the National Natural Science Foundation of China (grant no. 81860182) and the Innovation Fund Designated for Graduate Students of Jiangxi Province (grant no. YC2020-S128).

Author contributions

Conception: Jing Ye and Jun Wang.

Interpretation or analysis of data: Jun Wang, Junhao Tu, Binxiang Tang, Zhiqiang Zhang, Mei Han, Mengyue Li, Jieqing Yu, Li Shen, and Meiping Zhang.

Preparation of the manuscript: Junhao Tu and Jun Wang.

Revision of important intellectual content: Junhao Tu and Meiping Zhang.

Supervision: Jing Ye.

Supplementary data

The supplementary files are available to download from http://dx.doi.org/10.3233/CBM-220093.

Footnotes

Acknowledgments

First, we acknowledge the Medical Innovation Center, the First Affiliated Hospital of Nanchang University. Second, we are grateful to Dr. Jingming Zhang and Dr. Meng Zhang for their helpful advice for this article and to pathologists Xianhe Yang and Xiaojun Zhu for their help with the evaluation of pathological characteristics.

Conflict of interest

All authors declare no conflict of interest.

References

Gal

Silver

and Huang

, Demographics and treatment trends in sinonasal mucosal melanoma, The Laryngoscope 121 (2011), 2026–2033.

Gilain

et al., Mucosal melanoma of the nasal cavity and paranasal sinuses, Eur Ann Otorhinolaryngol Head Neck Dis 131 (2014), 365–369.

Amit

Na’ara

and Hanna

E.Y.

, Contemporary Treatment Approaches to Sinonasal Mucosal Melanoma, Curr Oncol Rep 20 (2018), 10.

Moya-Plana

et al., Oncologic outcomes, prognostic factor analysis and therapeutic algorithm evaluation of head and neck mucosal melanomas in France, European Journal of Cancer (Oxford, England: 1990) 123 (2019), 1–10.

Ablain

et al., SPRED1 Human tumor genomics and zebrafish modeling identify loss as a driver of mucosal melanoma, Science (New York, NY) 362 (2018), 1055–1060.

Carvajal

et al., KIT as a therapeutic target in metastatic melanoma, JAMA 305 (2011), 2327–2334.

Freiberger

et al., Who’s Driving? Switch of Drivers in Immunotherapy-Treated Progressing Sinonasal Melanoma, Cancers 13 (2021).

Mikkelsen

et al., The molecular profile of mucosal melanoma, Melanoma Res 30 (2020), 533–542.

Hayward

et al., Whole-genome landscapes of major melanoma subtypes, Nature 545 (2017), 175–180.

10.

Shoushtari

et al., Clinical features and response to systemic therapy in a historical cohort of advanced or unresectable mucosal melanoma, Melanoma Res 27 (2017), 57–64.

11.

Teterycz

et al., Multimodal Treatment of Advanced Mucosal Melanoma in the Era of Modern Immunotherapy, Cancers 12 (2020).

12.

Wang

Yan

and Jiang

, Evaluation of the prognostic impact of postoperative adjuvant radiotherapy on head and neck mucosal melanoma: a meta-analysis, BMC Cancer 15 (2015), 758.

13.

Skiba-Tatarska

Kusa-Podkańska

Surtel

and Wysokińska-Miszczuk

, The side-effects of head and neck tumors radiotherapy, Polski Merkuriusz Lekarski: Organ Polskiego Towarzystwa Lekarskiego 41 (2016), 47–49.

14.

Ledderose

Penkava

and Ledderose

, Prognostic Value of Tumor-Infiltrating Lymphocytes in Sinonasal Mucosal Melanoma, The Laryngoscope 132 (2021). doi: 10.1002/lary.29820.

15.

Margulies

et al., Genome sequencing in microfabricated high-density picolitre reactors, Nature 437 (2005), 376–380.

16.

Hong

et al., RNA sequencing: new technologies and applications in cancer research, J Hematol Oncol 13 (2020), 166.

17.

Cieślik

and Chinnaiyan

, Cancer transcriptome profiling at the juncture of clinical translation, Nature Reviews Genetics 19 (2018), 93–109.

18.

Suvà

and Tirosh

, Single-Cell RNA Sequencing in Cancer: Lessons Learned and Emerging Challenges, Mol Cell 75 (2019), 7–12.

19.

Wang

et al., Protease-Activated Receptor-2 Decreased Zonula Occlidens-1 and Claudin-1 Expression and Induced Epithelial Barrier Dysfunction in Allergic Rhinitis, Am J Rhinol Allergy 35 (2021), 26–35.

20.

et al., Expression of human chorionic gonadotropin, CD44v6 and CD44v4/5 in esophageal squamous cell carcinoma, World J Gastroenterol 11 (2005), 7401–7404.

21.

Remmele

and Stegner

, Recommendation for uniform definition of an immunoreactive score (IRS) for immunohistochemical estrogen receptor detection (ER-ICA) in breast cancer tissue, Der Pathologe 8 (1987), 138–140.

22.

Kim

Langmead

and Salzberg

, HISAT: a fast spliced aligner with low memory requirements, Nat Methods 12 (2015), 357–360.

23.

Pertea

et al., StringTie enables improved reconstruction of a transcriptome from RNA-seq reads, Nat Biotechnol 33 (2015), 290–295.

24.

Ritchie

et al., Limma powers differential expression analyses for RNA-sequencing and microarray studies, Nucleic Acids Res 43 (2015), e47.

25.

Love

Huber

and Anders

, Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2, Genome Biol 15 (2014), 550.

26.

Zhou

et al., Metascape provides a biologist-oriented resource for the analysis of systems-level datasets, Nat Commun 10 (2019), 1523.

27.

Seo

et al., UCHL1 Regulates Melanogenesis through Controlling MITF Stability in Human Melanocytes, The Journal of Investigative Dermatology 137 (2017), 1757–1765.

28.

Bertolotto

et al., A SUMOylation-defective MITF germline mutation predisposes to melanoma and renal carcinoma, Nature 480 (2011), 94–98.

29.

Gilot

et al., A non-coding function of TYRP1 mRNA promotes melanoma growth, Nat Cell Biol 19 (2017), 1348–1357.

30.

Bolander

et al., The protein expression of TRP-1 and galectin-1 in cutaneous malignant melanomas, Cancer Genomics Proteomics 5 (2008), 293–300.

31.

El Hajj

et al., SNPs at miR-155 binding sites of TYRP1 explain discrepancy between mRNA and protein and refine TYRP1 prognostic value in melanoma, Br J Cancer 113 (2015), 91–98.

32.

El Hajj

et al., Tyrosinase-related protein 1 mRNA expression in lymph node metastases predicts overall survival in high-risk melanoma patients, Br J Cancer 108 (2013), 1641–1647.

33.

Guo

et al., ATP-binding cassette member B5 (ABCB5) promotes tumor cell invasiveness in human colorectal cancer, J Biol Chem 293 (2018), 11166–11178.

34.

Kadioglu

et al., Effect of ABC transporter expression and mutational status on survival rates of cancer patients, Biomed Pharmacother 131 (2020), 110718.

35.

Wang

et al., ABCB5 promotes melanoma metastasis through enhancing NF-kappaB p65 protein stability, Biochem Biophys Res Commun 492 (2017), 18–26.

36.

Wilson

B.J.

et al., ABCB5 identifies a therapy-refractory tumor cell population in colorectal cancer patients, Cancer Res 71 (2011), 5307–5316.

37.

Cheung

S.T.

Cheung

P.F.

Cheng

C.K.

Wong

N.C.

and Fan

S.T.

, Granulin-epithelin precursor and ATP-dependent binding cassette (ABC)B5 regulate liver cancer cell chemoresistance, Gastroenterology 140 (2011), 344–355.

38.

Yao

et al., ABCB5-ZEB1 Axis Promotes Invasion and Metastasis in Breast Cancer Cells, Oncol Res 25 (2017), 305–316.

39.

Lee

et al., Targeting the ABC transporter ABCB5 sensitizes glioblastoma to temozolomide-induced apoptosis through a cell-cycle checkpoint regulation mechanism, The Journal of Biological Chemistry 295 (2020), 7774–7788.

40.

Yip

Foidart

Noel

and Sounni

N.E.

, MT4-MMP: The GPI-Anchored Membrane-Type Matrix Metalloprotease with Multiple Functions in Diseases, Int J Mol Sci 20 (2019).

41.

Wang

S.J.

Y.X.

and Luo

H.S.

, Expression and clinical significance of matrix metalloproteinase-17 and -25 in gastric cancer, Oncol Lett 9 (2015), 671–676.

42.

Chabottaux

et al., Membrane-type 4 matrix metalloproteinase promotes breast cancer growth and metastases, Cancer Res 66 (2006), 5165–5172.

Expression and clinical significance of TYRP1,ABCB5,and MMP17 in sinonasal mucosal melanoma

Abstract

BACKGROUND:

METHODS:

RESULTS:

CONCLUSION:

Keywords

1. Introduction

2. Materials and methods

2.1 Patients and tissues

2.2 Total RNA isolation and rRNA depletion

2.3 Library preparation and next-generation sequencing

2.4 Quantitative real-time PCR (qRT-PCR)

2.5 Western blotting

2.6 Immunohistochemical staining

2.7 Statistical analysis

2.8 Data analysis of RNA-seq

2.9 Functional annotation of DEGs using KEGG and GO pathway enrichment analyses

2.10 The cancer genome atlas and genotype-tissue expression databases

3. Results

3.2 Functional annotation of DEGs using KEGG and GO analyses

3.3 Relative expression of the key genes in SNMM

Table 1 Immunohistochemical analysis of TYRP1, ABCB5, and MMP17 expression in tumor tissues of SNMM patients

5. Conclusion

Ethical statement

Funding

Author contributions

Supplementary data

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Immunohistochemical analysis of TYRP1, ABCB5, and MMP17 expression in tumor tissues of SNMM patients