MMP1 overexpression promotes cancer progression and associates with poor outcome in head and neck carcinoma

doi:10.21203/rs.3.rs-1625021/v1

Download PDF

Research Article

MMP1 overexpression promotes cancer progression and associates with poor outcome in head and neck carcinoma

https://doi.org/10.21203/rs.3.rs-1625021/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Purpose: Matrix metalloproteinase-1 (MMP1) has been reported to play key roles in a variety of cancers by degrading the extracellular matrix. However, its carcinogenic roles have not been shown yet in head and neck squamous cell carcinoma (HNSCC). This study aimed to elucidate its expression pattern, functional roles as well as clinical significance in HNSCC.

Methods: The Cancer Genome Atlas (TCGA), Gene Expression Omnibus (GEO) and immunohistochemistry (IHC) were utilized to determine the MMP1 expression pattern and the associations between its expression and patients’ outcome in HNSCC. Mice tongue squamous cell carcinoma model induced by 4-nitroquinoline 1-oxide (4NQO), and siRNA mediated cellular assay in vitro were utilized to evaluate the oncogenic role of MMP1. The biological functions and cancer-related pathways involved in MMP1 related genes were found through bioinformatics analysis.

Results: Both mRNA and protein abundance of MMP1 were highly increased in HNSCC as compared to its non-tumor counterparts. MMP1 overexpression positively correlated with advanced tumor size, cervical node metastasis and advanced pathological grade and lower patients’ survival. In the 4NQO-induced animal model, MMP1 expression increased along with the progression of the disease. In HNSCC cells, siRNA-mediated knockdown of MMP1 significantly inhibited cell proliferation, migration, and invasion, and activated apoptosis and epithelia-mesenchymal transition (EMT). GSEA, GO and KEGG analysis showed that MMP1 expression was significantly related to cancer-related pathways and cancer-related functions.

Conclusion: Together, our results demonstrated MMP1 serves as a novel prognostic biomarker and putative oncogene in HNSCC.

Head and neck squamous cell carcinoma

MMP1

migration

invasion

EMT

Head and neck squamous cell carcinoma (HNSCC) is the ninth common cancer worldwide with more than 550000 new cases and 300000 death cases[1]. Treatment options for HNSCC include surgical resection, chemotherapy, and focused radiation therapy. Nevertheless, the 5-year survival rate of HNSCC patients has not been significantly improved due to cervical lymph node metastasis, locoregional recurrence and therapeutic resistance[2, 3]. Despite extensive study into HNSCC, the underlying cellular and molecular mechanism of HNSCC emerge remains incompletely understood. Therefore, the exploration of molecular mechanisms underlying HNSCC will aid in the development of molecular targeted therapies for this dreaded disease.

Molecular biology research and whole genome sequencing technology provide rich resources for developing biomarkers that will be utilized for early diagnosis, patient stratification, personalized treatment and prognosis prediction[4, 5]. Matrix metalloproteinases (MMPs) are a family of endopeptidases, which are composed of more than 28 kinds of human MMPs. In addition to being vital for development, wound healing, and inflammatory diseases, MMPs are also important in the initiation and progression of tumors^{[6, 7]}. By degrading the extracellular matrix (ECM), they regulate the behavior of various types of cells relevant to cancer biology[8]. Previous studies have reported MMPs play essential roles including tumor growth, differentiation, apoptosis, angiogenesis, migration, invasion and immune surveillance in multiple stages of cancer[7, 9]. As a crucial member of the MMP family, the overexpression of Matrix metalloproteinase-1 (MMP1) was associated with cell proliferation, migration, invasion, and poor prognosis of cancer such as oral squamous cell carcinoma (OSCC), cutaneous squamous cell carcinoma (CSCC), lung cancer and cervical squamous cell carcinoma (CESC)[8, 10–12]. Additionally, the latest studies indicated that MMP1 play a vital role in EMT in colorectal cancer and squamous cell carcinoma of the skin [13, 14]. Nevertheless, the precise biological functions of MMP1 in HNSCC are just now beginning to surface in selected contexts and the associations between MMP1 and EMT remain yet unexplored in HNSCC.

The above results suggest that MMP1 may be a new potential oncogene in human cancer. Nevertheless, the role of MMP1 in the development of HNSCC remains unclear. In the present study, we investigated the expression of MMP1 in HNSCC and revealed the correlations between its expression and clinical parameters by collecting data from TCGA, GEO and our samples. Additionally, small interference RNA and bioinformatics analysis were used to explore the biological function of MMP1 in HNSCC. All these findings indicate that MMP1 as a putative oncogene involving in HNSCC initiation and progression and as a novel biomarker utilizing for HNSCC diagnosis.

Patients’ specimens and immunohistochemical staining (IHC)

103 patients diagnosed with HNSCC from Jan. 2010 to Dec. 2017 underwent radical resections at the Department of Oral and Maxillofacial Surgery, Affiliated Huaian No.1 People’s Hospital of Nanjing Medical University. Patient inclusion criteria were described as we previously reported[15]. The normal oral mucosa of 20 patients with non-cancer surgery were also enrolled as the control group. This study protocol was reviewed and approved by the Research Ethic Committee of the Affiliated Huaian No.1 People’s Hospital. All included patients have signed written informed consent. The detailed steps of IHC and intensity score we had reported previously. MMP1 primary antibody (1:200; ab137332; Abcam) was utilized for IHC and then its expression intensity of each slide was divided into two groups due to the final score: low (<4) or high (>4) MMP1 expression group[15].

Cell culture and small interference RNA (siRNA) transfection

Human Oral Keratinocytes (HOK), Fadu, Cal27, HN6 and HN4 were purchased from American Type Culture Collection (ATCC, Manassas, VA, USA). Cultures were carried out at 37°C and 5% CO2 in DMEN/F12 (Invitrogen) supplemented with 10% heat-inactivated fetal bovine serum (Gibco). The siRNA including MMP1 siRNA-1 (5′-ACACAAGAGCAAGATGTGG-3′) , MMP1 siRNA-2 (5′-TAAGTACTGGGCTGTTCAG-3′) and control siRNA were purchased from GenePharma (Shanghai, China). Cells were transfected with MMP1 siRNA with final concentration 100nM according to Lipofectamine 3000 instruction (Invitrogen, USA). Then, these cells were utilized for further experiments for 48h.

RNA extraction and real time RT-PCR

The RNA was extracted from the HNSCC cells and tissues using Trizol reagent (Invitrogen) and then RT-PCR reactions were performed with a PrimeScript RT reagent kit (Takara Bio, Shiga, Japan) as described previously[16]. The primers were listed as follows: MMP1 (forward: AAATGCAGGAATTCTTTGGG, reverse: ATGGTCCACATCTGCTCTTG), N-cadherin (foward: TTTGATGGAGGTCTCCTAA, reverse: ACGTTTAACACGTTGGAA), E-cadherin (foward: CGAGAGCTACACGTTCACGG, reverse: GGGTGTCGAGGGAAAAATAGG), Vimentin (foward: AAAACACCCTGCAATCTTTCAGA, reverse: CACTTTGCGTTCAAGGTCAA), Snail (foward: ACACGGACAGGATTGACAGA, reverse: GCATGAAGAAGCCGCGAAGTGT) and GAPDH (foward: AGGTGAAGGTCGGAGTCAAC, reverse: AGTTGAGGTCAATGAAGGGG).

Western blot analysis

The proteins of each group were collected, and the precipitate was separated by lysis, shock and centrifugation. The precipitate was separated by electrophoresis in 10% SDS-PAGE gel, then transferred to PVDF membrane. After 5% of the skim milk was sealed 2h, the membranes were incubated with related primary antibodies overnight. Then, the membranes were incubated with secondary antibodies for 1h and imaged by ECL chemiluminescence kit (Bio-Rad). The antibodies utilized as follows: MMP1 (1:1000; ab137332; Abcam), Vimentin (1:2000, #5741, CST), Snail (1:1000, #3879, CST), E-cadherin (1:2000, #3195, CST), N-cadherin (1:1000, #13116, CST) and GAPDH (1:2000 dilution, #2118, CST).

CCK‑8 and colony formation assay

The proliferation of HNSCC cells was assessed by absorbance using a CCK-8 cell viability assay (Cell Counting Kit8, Dojindo, Japan) according to manufacturer's instructions. Detailed CCK-8 and colony formation assay was performed as we previously reported[17].

Flow cytometric assay

The cells of each group were collected and made into single cell suspension. The cells were stained according to the instructions of Annexin V: PI Apoptosis Detection Kit (BD Bioscience), and then detected by flow cytometry. The data were analyzed by Flowjo software.

Wound-healing assay and invasion assay

Cal27 cells were seeded in 2× 10⁵ cell/well in six-well plates. Then, we utilized pipette tip to create an artificial wound on the confluent cell monolayer. The suspended cells were washed with PBS for three times. Then cells were cultured in medium with 1% FBS (Gibco). The wounds were photographed at 0, 24h as indicated. For cell invasion assays, 5×10⁴ cells were placed into the upper chamber precoated with Matrigel and culture medium with 8% FBS was added to the lower chamber. After incubation for 24 h, non-migrated cells were removed with a cotton swab. The chambers were fixed with 4% paraformaldehyde for 15 min at room temperature and incubated with 0.1% crystal violet solution for 30 min at 37℃. Cells were determined under a light microscope.

Chemical-induced HNSCC animal model

A well-known chemical-induced mouse model was utilized to further explore the expression of MMP1 in the occurrence and development of HNSCC. 6-week-old C57BL/6 mice were fed with water containing 50μg/mL 4NQO for 16 weeks, and then fed with normal water for another 8-10 weeks. During these 8-10 weeks, the lesions of mouse tongue were observed every week. Mice that had been drinking normal water were used as control. The tongue specimens of mice at 16, 20 and 24 weeks were collected for histopathological analysis. This study protocol was reviewed and approved by the Research Ethic Committee of the Affiliated Huaian No.1 People’s Hospital.

Bioinformatics analysis of MMP1

MMP1 mRNA expression level and correlated clinicopathological data of HNSCC were obtained from 2 publicly datasets: TCGA (The Cancer Genome Atlas) (https://cancergenome.nih.gov/) and GEO (Gene Expression Omnibus) (https://www.ncbi.nlm.nih.gov/gds/). We utilized the GEPIA2 database (http://gepia2.cancer-pku.cn/) to analyze the overall and disease-free survival of TCGA-HNSCC cohort. Additionally, TCGA-HNSCC raw expression/clinical data was automatically sorted and normalized using R packages. We downloaded MMP1 correlated genes’ chart from cBioPortal (https://www.cbioportal.org/). Then, the related genes with Spearman’s correlation > 0.2 were selected for Gene Ontology and KEGG pathway analysis. Median-cutoff was used to divided the patients from TCGA-HNSC cohort into high and low MMP1 expression subgroups, and then the differentially expressed genes (DEGs) between these two groups were further analysis by R packages limma[18].

Statistical analysis

All data were performed using GraphPad Prism 8, R packages ggplot2 or SPSS 21.0 software. All the experiments were repeated three times, then students’ t-test and ANOVA were utilized for statistical analysis. Chi-square or Fisher exact test were used for evaluating the correlations between MMP1 expression and clinical parameters. The survival rate of our patients’ cohort was estimated by Kaplan-Meier method and log-rank test was used for statistical analysis. Univariate and multivariate Cox-regression analysis were performed to evaluate prognostic factors correlated with patients’ overall survival. P values less than 0.05 were considered statistically significant.

MMP1 mRNA was highly expressed in HNSCC via bioinformatics analyses

Accumulating evidence suggested that MMP1 was highly expressed in most cancers as well as pan-cancer analysis (Figure S1) from GEPIA2[19, 20]. In order to investigate the expression level of MMP1 in tumor specimens of HNSCC patients, we preliminarily evaluated the mRNA expression level of MMP1 by interrogating the public gene expression database including TCGA and GEO. As shown in Fig.1A, TCGA-HNSCC cohort showed that MMP1 mRNA expression was significantly upregulated in HNSCC samples compared to their non-tumor counterparts. Additionally, five independent HNSCC patients’ cohorts from GEO database such as GSE13601, GSE25099, GSE30784, GSE9844X and GSE37991 cohorts were utilized to identify MMP1 mRNA expression. As shown in Fig.1B–F, the abundance of MMP1 mRNA in HNSCC samples was also significantly increased in tumor samples. These figures indicated MMP1 might be involved in the occurrence and development of HNSCC.

MMP1 overexpression was associated with clinical parameters in HNSCC patients

In order to further determine the expression pattern of MMP1 in HNSCC, we next performed immunohistochemical staining on 103 samples of primary HNSCC. The detailed demographic and clinicopathological parameters of these patients were listed in Table 1. In short, 54 males and 49 females were enrolled with an average age of 63.8 years. The follow-up time ranged from 4 to 95 months, with an average of 57.1 months. Until the last follow-up, 54 (52.4%) patients survived with disease-free, 9 (8.7%) patients still alive but with recurrences and/or cervical nodal metastases, 40 (38.9%) patients died due to post-surgical recurrence, metastases or other unrelated diseases. As shown in Fig.2A-F, rare MMP1 staining was observed in most healthy oral mucosa samples and a few HNSCC samples, whereas positive MMP1 cytoplasmic staining was detected in a variety of HNSCC samples. The expression patterns of MMP1 in HNSCC and normal oral epithelium were classified according to immunohistochemical scores. Consequently, MMP1 protein abundance could be categorized into low (40) or high expression (63) in HNSCC samples while negative (6), low (8) or high expression (6) in normal oral epithelial samples, which suggested that MMP1 was abnormally overexpressed in HNSCC (P<0.0001, chi-square test). The relationships between MMP1 expression and clinical parameters of HNSCC patients was shown in Table 1. There were no significant associations between MMP1 expression and gender, age, smoking, alcohol drinking and clinical stage. However, high MMP1 expression positively associated with advanced tumor size, advanced pathological grade and cervical node metastasis and with P-value 0.0097, 0.0006 and 0.0280, respectively.

Table 1

The associations between MMP1 expression and multiple clinicopathological parameters in HNSCC samples.

Clinicopathological parameters

Cases

MMP1

P-values

Low

High

Gender

Male

Female

103

0.4269

Age

≤60

>60

0.6829

Smoking

Yes

0.8211

Alcohol use

Yes

0.1526

Tumor size

T1-T2

T3-T4

0.0097^*

Pathological Grade

III

0.0006^*

Cervical node metastasis

N(0)

N(+)

0.0280^*

Clinical stage

III

0.2842

* indicate statistical significance with p-values less than 0.05.

Aberrant overexpression of MMP1 significantly associated with poor prognosis in HNSCC patients

Next, Kaplan-Meier survival analysis was utilized to investigate the correlation between MMP1 expression and patient prognosis. As shown in Fig.3A-B, the outcome of Kaplan-Meier analysis revealed that high expression of MMP1 was related to poor prognosis in overall survival and disease-free survival (log-rank test, P=0.0248, 0.0202). Moreover, the similar conclusion from GSE41613, GSE42743 and TCGA-HNSCC cohort also showed that overexpression of MMP1 had reduced overall survival compared to low MMP1 expression (Log-rank, P=0.031, 0.0526, 0,046, Fig. 3C-D and Figure S2A). However, there was no correlations between MMP1 mRNA expression and disease-free survival in TCGA-HNSCC cohort, when patients were divided into low and high MMP1 expression subgroups using the median of MMP1 mRNA as cutoff (Figure S2B). We further investigated the associations between the abundance of MMP1 mRNA and pathological grade and clinical stage in TCGA-HNSCC. There were also no correlations between MMP1 mRNA expression and clinicopathological parameters (Figure S2C, D). Additionally, we evaluated the prognostic value of MMP1 expression in HNSCC by univariate and multivariate survival analysis. Consistent with Kaplan-Meier survival analysis, univariate regression analysis also showed that MMP1 expression was significantly correlated with patient survival (P=0.023, Table 2). Moreover, multivariate regression analysis demonstrated MMP1 expression was identified as an independent factor of patient survival after adjusting for other demographic and clinicopathological parameters. (P=0.014, Table 2).

Table 2

Univariate and multivariate survival analyses (proportional hazards method) for patients with primary HNSCC.

Variable	Univariate survival analysis			Multivariate survival analysis
	Hazard ratio	95% CI	p-value	Hazard ratio	95% CI	p-value
Gender (male, female)	1.132	0.558-2.299	0.731	1.070	0.334-3.423	0.909
Smoking (No, Yes)	1.095	0.504-2.378	0.819	1.389	0.469-4.108	0.553
Alcohol use (No, Yes)	0.802	0.389-1.654	0.550	0.648	0.205-2.047	0.460
Age (≤60, >60)	1.192	0.561-2.535	0.648	1.588	0.730-3.452	0.243
Tumorsize (T1-T2, T3-T4)	0.675	0.236-1.931	0.463	0.575	0.199-1.660	0.306
Pathological grade (I, II-III)	1.307	0.871-1.961	0.196	1.206	0.779-1.867	0.838
Cervical nodal metastasis (N0, N+)	1.402	0.628-3.133	0.410	1.816	0.483-6.832	0.377
Clinical stage (I-II, III-IV)	1.302	0.548-4.133	0.549	1.273	0.788-2.198	0.641
MMP1 expression (low, high)	3.306	1.163-7.928	0.023^*	3.410	1.281-9.082	0.014^*

* indicate statistical significance with p-values less than 0.05.

Increased MMP1 expression in chemical‑induced HNSCC tumorigenesis

Based on our findings that high expression of MMP1 in human HNSCC samples, as shown in Fig.4A, a well-established chemical-induced mouse model was utilized to further explore the expression of MMP1 in the occurrence and development of HNSCC. 4NQO treatment led to multiple lesions of the tongue including epithelial hyperplasia, dysplasia, carcinoma in situ and squamous cell carcinoma, which largely reproduced the multi-stage tumorigenic biological behavior of HNSCC. As shown in Fig.4B-I, immunohistochemical staining of MMP1 in mouse specimens indicated significant cytoplasmic staining in dysplasia, carcinoma in situ and invasive carcinoma, while negative or low staining in normal tongue epithelial. Positive MMP1 cytoplasmic staining was commonly observed in carcinoma (100%, 10/10), hyperplasia (60%, 6/10), and dysplasia/carcinoma in situ (60%, 6/10), but much less in samples with healthy mucosa (20%, 2/10). Moreover, consistent with IHC findings, qRT-PCR results revealed that the mRNA levels of MMP1 in carcinoma in situ/SCC samples were highly upregulated compared with normal tongue samples (Fig. 4J). Above all, our findings demonstrated that MMP1 might be a putative oncogene driving the occurrence and development of HNSCC.

MMP1 depletion inhibited cell proliferation, migration/invasion and promoted apoptosis in HNSCC

The above experiments have demonstrated the tumor promoting effect of MMP1, we next aimed to elucidate its oncogenic roles by siRNA-mediated loss of function approach during HNSCC initiation and progression. We first detected the expression of MMP1 in a group of HNSCC cell lines and found that MMP1 protein abundance was highly overexpressed in all HNSCC cell lines compared with HOK cells (Fig.5A). We next selected Cal27 and Fadu cells for siRNA-mediated knockdown experiments owing to relatively higher MMP1 protein in them. By introducing two independent siRNAs targeting human MMP1 into Cal27 and Fadu cells, the subsequent changes of MMP1 expression and cell phenotype were monitored. As shown in Fig.5B, after siRNA transfection, MMP1 protein expression was significantly decreased in Cal27 and Fadu cells. Then we utilized these transfected cells for phenotypic experiments. CCK-8 viability assay and colony formation assay indicated significantly lower proliferation rate in Cal27 and Fadu cells following MMP1-siRNA transfection (Fig.5C-E). In addition, cytometric assay showed that the proportions of apoptotic cells in siMMP1-transfected cells were significantly increased from 1.3% to 5.0/8.3% in Cal27, from 2.6% to 20.8/41.3% in Fadu, respectively (Fig.5F, G). Both wound healing and transwell invasion assays demonstrated the migratory and invasive abilities of cells following MMP1 knockdown were significantly reduced (Fig.5F-I). In line with these findings, western blot and qRT-PCR results indicated the protein and mRNA expression of EMT/metastasis-associated marker Vimentin, N-cadherin and Snail were down-regulated concomitant with E-cadherin up-regulated following MMP1 knockdown (Fig. 5K-M). Finally, we generated generic EMT signature to score the EMT status of these samples from TCGA-HNSCC cohort, which was a useful tool for objective, systematic investigation of EMT roles and dynamics in cancer progression, treatment response, and survival [21]. Consistent with above findings, our results from this scoring system revealed that MMP1 expression was significantly associated with EMT score in TCGA-HNSCC (R=0.322, P<0.0001, Fig.5N). These results suggested that MMP1 knockdown inhibited cell proliferation, migration/invasion and activates apoptosis in HNSCC by EMT.

Functional enrichment analysis of the differentially expressed genes (DEGs) correlated to MMP1

Next, we utilized functional enrichment analysis to complement the in vitro loss-of-function experiment in exploring MMP1 pro-tumorigenic functions. Firstly, we screened 26526 negative and 10403 positive genes correlated to MMP1 from TCGA-HNSCC on GEPIA 2. (Figure S3A). Then, we performed GO and KEGG analysis on these genes. The data suggests that MMP1 correlated genes enriched in numerous cancer related biofunction and pathways such as wound healing, cell-substrate adhesion, cell-matrix adhesion, PI3K-Akt signaling pathway, human papillomavirus infection, focal adhesion, MAPK signaling pathway and proteoglycans in cancer, indicating MMP1 played a pro-tumorigenic role by these pathways (Figure S3B, C). We further divided the expression of MMP1 into high-expression group and low-expression group (median as cutoff) by the method of bipartition in TCGA-HNSCC data. Then, we utilized bioinformatics analysis to recognize the nominee genes that might be correlated with high/low group of MMP1 (Figure S3D-S3F). Consistent with above findings, GO and KEGG analysis for the correlated genes associated with high group of MMP1 were significantly enriched in wound healing, cell-substrate adhesion, cell-matrix adhesion, PI3K-Akt signaling pathway, human papillomavirus infection, focal adhesion and proteoglycans in cancer (Fig. 6A, C). Additionally, KEGG and GO analysis showed the correlated genes associated with low group of MMP1 were related to cell cycle and metabolism (Fig. 6B, D). Finally, we analyzed the differentially expressed genes (DEGs) between two groups using GSEA analysis. The results revealed that DEGs between two groups were significantly enriched in EMT pathway (Fig. 6E), epidermal cell differentiation (Fig. 6F), focal adhesion (Fig. 6G) and p53 signaling pathway (Fig. 6H). Above all, combined with cell assay in vitro and bioinformatics results extremely support the idea that MMP1 is an oncogene promoting cell migration and invasion by EMT in HNSCC.

MMPs have been long appreciated as important mediators of tissue remodeling, targeting for proteolysis numerous extracellular matrix components. Previous studies have reported MMPs play essential roles including tumor growth, differentiation, apoptosis, angiogenesis, migration, invasion and immune surveillance in multiple steps of cancer progression[22, 23]. MMP1 as a critical member of the MMP family, a variety of evidence have reported that its oncogenic roles and prognostic significance underlying multiple cancer contexts[8, 10–12]. Herein, we used public databases, HNSCC samples, vitro loss-of-function experiments and bioinformatics analysis to determine the expression form, prognostic significance and carcinogenic effect of MMP1 in HNSCC. The results demonstrate that MMP1 as a putative oncogene promotes HNSCC initiation and development and also be a novel biomarker with clinical translation potential.

Accumulating evidence have revealed that MMP1 is usually overexpressed in multiple cancers including lung cancer, esophageal squamous cell carcinoma (ESCC), nasopharyngeal carcinoma, breast cancer and OSCC[8, 11, 24–26]. Consitent with previous studies, our results revealed that MMP1 was highly expressed in multple independent patient cohorts from publicly available dataset and immunohistochemistry in primary HNSCC samples. In line with this findings, the results of 4NQO-induced HNSCC model showed that the expression of MMP1 increased with the occurrence and progression from hyperplasia to invasive carcinoma. Although MMP1 amplification was identified in mutiple cancers, the mechanisms of its overexpression were still unclear. Therefore, we utilized cBioPortal to explore genetic alteration of MMP1 and found genetic amplication occurred in only 5% HNSCC sample (Supplement Fig. 3). Thus largely ruled out the possibility of of MMP1 genetic amplification responsible for its overexpression in most HNSCC samples. The accurate mechanisms of its amplication in HNSCC still need further investigation. All these findings strongly demonstrate MMP1 promotes HNSCC initiation and development as a bona fide oncogene.

Several previous reports have proposed important clinical relevance of MMP1 overexpression in human cancer. For example, elevated MMP1 was significantly associated with pN status and clinical stage in ESCC[25]. In colorectal cancer, MMP1 overexpression was positively correlated with TNM stage and lymphatic metastasis[13, 27]. Moreover, abnormally overexpression of MMP1 was positively correlated with the worse outcome in breast cancer, ESCC, melanoma, colon cancer and pancreatic carcinoma[24, 25, 28–30]. In line with these findings, our results from primary HNSCC samples revealed that MMP1 overexpression significantly associated with advanced tumor size, cervical node metastasis and advanced pathological grade. In addition, both Kaplan-Meier survival and univariate/multivariate Cox-regression analyses showed that the overexpression of MMP1 positively correlated with reduced survival, and it could be served as an independent prognostic predictor for patients’ survival. Complementary, TCGA-HNSCC and GEO data also demonstrated that overexpression of MMP1 was associated with decreased overall survival which further strengthened our results. However, the mRNA expression of MMP1 was not related to clinical grades, pathological stages and diease-free survival in TCGA-HNSCC cohort. We reasoned that it’s conceviable because of the significant heterogeneity of HNSCC between TCGA cohort and our cohort. Secondly, different classification methods of TCGA-HNSCC cohort and our cohort and the inconsistency between mRNA and protein expression of MMP1 might lead to this discrepancy. Of course, more centers and more patients are needed to establish the clinical and prognostic significance of MMP1 in HNSCC, so as to determine whether MMP1 can be used as novel biomarker for evaluation of patients’ prognosis.

It is undisputed that MMP1 is critically involved in tumorigenesis by promoting cell proliferation, migration, invasion, EMT and inhibiting apoptosis[10, 13, 31]. For example, the downregulation of MMP1 inhibited invasion and migration in gastric cancer and cervical cancer[32, 33]. In addition, in colorectal cancer and ESCC, the overexpression of MMP1 promoted cell proliferation, migration, invasion, EMT in vitro and metastasis in vivo[13, 25]. In line with the above findings, our results demonstrate that MMP1 has a variety of tumorigenic roles by promoting cell proliferation, migration, invasion and suppressing apoptosis in HNSCC cells. Accumulating evidence have shown EMT is a significant approach regulating migration and invasion necessary for primary tumor cells in recent years[34, 35]. Noticeably, our results demonstrated that MMP1 knockdown strongly inhibited invasion and motility by facilitating MET in HNSCC as evidenced by the mRNA and protein expression of EMT related markers were reversed upon MMP1 depletion as well as EMT scoring system. Moreover, bioinformatics analysis from public databases data sources also suggested MMP1 associated differentially expressed genes were significantly rich in EMT pathway and would healing. These results partly explained the correlation between the high expression of MMP1 and cervical lymph node metastasis. In addition, bioinformatics analysis results also indicated MMP1 have involved in significant cancer-related pathways such as PI3K-Akt signaling pathway, focal adhesion, MAPK signaling pathway and cell cycle. For instance, knockdown of MMP1 inhibited the progression of colorectal cancer by suppressing the PI3K-Akt-myc signaling pathway and EMT[13, 36]. Additionally, Ji et al. have demonstrated the downregulation of MMP-1, MMP-2, and MMP-7 could play anti-metastatic effects of via inactivation of MAPK signaling and induction of focal adhesion formation in hepatocellular carcinoma[37]. Complementary, Yu et al. have reported that MMP1 significantly facilitated colon cancer cell proliferation via accelerating cell cycle transition from G0/G1 to S and G2/M phase[36], which in part strengthened our data. Collectively, our findings together with others strongly suggest that MMP1 probably functions as a pro-tumorigenic oncogene via enhancing cancer cell proliferation, migration and invasion in HNSCC by inducing EMT and cancer-related pathways. However, the precise molecular mechanisms regulating EMT await further elucidation.

Our findings revealed MMP1 was highly expressed in HNSCC, and its high expression was significantly correlated with tumor size, pathological grade, cervical lymph node metastasis and poor prognosis. Moreover, MMP1 suppressed apoptosis and promoted proliferation, migration, and invasion by mediating EMT in HNSCC cells. Therefore, MMP1 may be used as a new biomarker and therapeutic target in HNSCC to participate in the diagnosis and treatment of HNSCC patients.

Acknowledgements

Not applicable.

Authors’ contributions

WZ and XH performed the experimental study, data collection and analysis and manuscript writing. RH, HZ and PY carried out the most experiments. XL, SZ and MW performed histological and statistical analyses. FJ conceived and supervised the whole project. All authors read and approved the final manuscript.

Funding

This work is financially supported by grants from the Science and Technology Development Fund of Nanjing Medical University (Grant No. NMUB2020140).

Data availability

Raw data and images can be obtained from the authors.

Declarations

Conflict of interests Statement

The authors declare that they have no conflict of interest.

Ethics Approval

Our study protocol was reviewed and approved by the the Research Ethic Committee of the Affiliated Huaian No.1 People’s Hospital. Written informed consent was obtained from all subjects in our study.

Consent to participate

All authors stated consent for participation.

Consent for Publication

All the authors agree with the publication of this original article.

Siegel RL, Miller KD, Fuchs HE, Jemal A (2021) Cancer Statistics, 2021. CA Cancer J Clin 1:7-33.https://doi.org/10.3322/caac.21654
Kitamura N, Sento S, Yoshizawa Y, Sasabe E, Kudo Y, Yamamoto T (2020) Current Trends and Future Prospects of Molecular Targeted Therapy in Head and Neck Squamous Cell Carcinoma. Int J Mol Sci 1.https://doi.org/10.3390/ijms22010240
Johnson DE, Burtness B, Leemans CR, Lui VWY, Bauman JE, Grandis JR (2020) Head and neck squamous cell carcinoma. Nat Rev Dis Primers 1:92.https://doi.org/10.1038/s41572-020-00224-3
Furuta K, Arao T, Sakai K, Kimura H, Nagai T, Tamura D, et al (2012) Integrated analysis of whole genome exon array and array-comparative genomic hybridization in gastric and colorectal cancer cells. Cancer Sci 2:221-7.https://doi.org/10.1111/j.1349-7006.2011.02132.x
Agrawal N, Frederick MJ, Pickering CR, Bettegowda C, Chang K, Li RJ, et al (2011) Exome sequencing of head and neck squamous cell carcinoma reveals inactivating mutations in NOTCH1. Science 6046:1154-7.https://doi.org/10.1126/science.1206923
Van Doren SR (2015) Matrix metalloproteinase interactions with collagen and elastin. Matrix Biol:224-31.https://doi.org/10.1016/j.matbio.2015.01.005
Egeblad M, Werb Z (2002) New functions for the matrix metalloproteinases in cancer progression. Nature reviews Cancer 3:161-74.https://doi.org/10.1038/nrc745
Wang C, Mao C, Lai Y, Cai Z, Chen W (2021) MMP1 3'UTR facilitates the proliferation and migration of human oral squamous cell carcinoma by sponging miR-188-5p to up-regulate SOX4 and CDK4. Mol Cell Biochem 2:785-96.https://doi.org/10.1007/s11010-020-03944-y
Yadav L, Puri N, Rastogi V, Satpute P, Ahmad R, Kaur G (2014) Matrix metalloproteinases and cancer - roles in threat and therapy. Asian Pac J Cancer Prev 3:1085-91.https://doi.org/10.7314/apjcp.2014.15.3.1085
Jiang S, Liu H, Zhang J, Zhang F, Fan J, Liu Y (2021) MMP1 regulated by NEAT1/miR-361-5p axis facilitates the proliferation and migration of cutaneous squamous cell carcinoma via the activation of Wnt pathway. Cancer Biol Ther 5-6:381-91.https://doi.org/10.1080/15384047.2021.1941583
Gabasa M, Radisky ES, Ikemori R, Bertolini G, Arshakyan M, Hockla A, et al (2021) MMP1 drives tumor progression in large cell carcinoma of the lung through fibroblast senescence. Cancer Lett:1-12.https://doi.org/10.1016/j.canlet.2021.01.028
Zhao S, Yu M (2020) Identification of MMP1 as a Potential Prognostic Biomarker and Correlating with Immune Infiltrates in Cervical Squamous Cell Carcinoma. DNA Cell Biol 2:255-72.https://doi.org/10.1089/dna.2019.5129
Wang K, Zheng J, Yu J, Wu Y, Guo J, Xu Z, et al (2020) Knockdown of MMP1 inhibits the progression of colorectal cancer by suppressing the PI3K/Akt/cmyc signaling pathway and EMT. Oncol Rep 4:1103-12.https://doi.org/10.3892/or.2020.7490
Fernandez-Figueras MT, Carrato C, Saenz-Sarda X, Musulen E, Fuente MJ, Puig L (2021) MicroRNA31 and MMP-1 contribute to the differentiated pathway of invasion -with enhanced epithelial-to-mesenchymal transition- in squamous cell carcinoma of the skin. Arch Dermatol Res.https://doi.org/10.1007/s00403-021-02288-x
Zhang W, Li J, Wu Y, Ge H, Song Y, Wang D, et al (2018) TEAD4 overexpression promotes epithelial-mesenchymal transition and associates with aggressiveness and adverse prognosis in head neck squamous cell carcinoma. Cancer Cell Int:178.https://doi.org/10.1186/s12935-018-0675-z
Wu Y, Wang Y, Diao P, Zhang W, Li J, Ge H, et al (2019) Therapeutic Targeting of BRD4 in Head Neck Squamous Cell Carcinoma. Theranostics 6:1777-93.https://doi.org/10.7150/thno.31581
Zhang W, Ge H, Jiang Y, Huang R, Wu Y, Wang D, et al (2020) Combinational therapeutic targeting of BRD4 and CDK7 synergistically induces anticancer effects in head and neck squamous cell carcinoma. Cancer Lett:510-23.https://doi.org/10.1016/j.canlet.2019.11.027
Yu G, Wang LG, Han Y, He QY (2012) clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS 5:284-7.https://doi.org/10.1089/omi.2011.0118
Tang ML, Bai XJ, Li Y, Dai XJ, Yang F (2018) MMP-1 Over-expression Promotes Malignancy and Stem-Like Properties of Human Osteosarcoma MG-63 Cells In Vitro. Curr Med Sci 5:809-17.https://doi.org/10.1007/s11596-018-1947-5
Wang QM, Lv L, Tang Y, Zhang L, Wang LF (2019) MMP-1 is overexpressed in triple-negative breast cancer tissues and the knockdown of MMP-1 expression inhibits tumor cell malignant behaviors in vitro. Oncol Lett 2:1732-40.https://doi.org/10.3892/ol.2018.9779
Tan TZ, Miow QH, Miki Y, Noda T, Mori S, Huang RY, et al (2014) Epithelial-mesenchymal transition spectrum quantification and its efficacy in deciphering survival and drug responses of cancer patients. EMBO molecular medicine 10:1279-93.https://doi.org/10.15252/emmm.201404208
Kessenbrock K, Plaks V, Werb Z (2010) Matrix metalloproteinases: regulators of the tumor microenvironment. Cell 1:52-67.https://doi.org/10.1016/j.cell.2010.03.015
Stallings-Mann ML, Waldmann J, Zhang Y, Miller E, Gauthier ML, Visscher DW, et al (2012) Matrix metalloproteinase induction of Rac1b, a key effector of lung cancer progression. Sci Transl Med 142:142ra95.https://doi.org/10.1126/scitranslmed.3004062
Shen CJ, Kuo YL, Chen CC, Chen MJ, Cheng YM (2017) MMP1 expression is activated by Slug and enhances multi-drug resistance (MDR) in breast cancer. PLoS One 3:e0174487.https://doi.org/10.1371/journal.pone.0174487
Liu M, Hu Y, Zhang MF, Luo KJ, Xie XY, Wen J, et al (2016) MMP1 promotes tumor growth and metastasis in esophageal squamous cell carcinoma. Cancer Lett 1:97-104.https://doi.org/10.1016/j.canlet.2016.04.034
Song L, Liu H, Liu Q (2019) Matrix metalloproteinase 1 promotes tumorigenesis and inhibits the sensitivity to 5-fluorouracil of nasopharyngeal carcinoma. Biomed Pharmacother:109120.https://doi.org/10.1016/j.biopha.2019.109120
Bendardaf R, Buhmeida A, Ristamaki R, Syrjanen K, Pyrhonen S (2007) MMP-1 (collagenase-1) expression in primary colorectal cancer and its metastases. Scand J Gastroenterol 12:1473-8.https://doi.org/10.1080/00365520701485449
Wang T, Zhang Y, Bai J, Xue Y, Peng Q (2021) MMP1 and MMP9 are potential prognostic biomarkers and targets for uveal melanoma. BMC Cancer 1:1068.https://doi.org/10.1186/s12885-021-08788-3
Xie J, Zhou X, Wang R, Zhao J, Tang J, Zhang Q, et al (2021) Identification of potential diagnostic biomarkers in MMPs for pancreatic carcinoma. Medicine (Baltimore) 23:e26135.https://doi.org/10.1097/MD.0000000000026135
Jonsson A, Falk P, Angenete E, Hjalmarsson C, Ivarsson ML (2021) Plasma MMP-1 Expression as a Prognostic Factor in Colon Cancer. J Surg Res:254-60.https://doi.org/10.1016/j.jss.2021.04.021
Liu H, Kato Y, Erzinger SA, Kiriakova GM, Qian Y, Palmieri D, et al (2012) The role of MMP-1 in breast cancer growth and metastasis to the brain in a xenograft model. BMC Cancer:583.https://doi.org/10.1186/1471-2407-12-583
Chen YJ, Liang L, Li J, Wu H, Dong L, Liu TT, et al (2020) IRF-2 Inhibits Gastric Cancer Invasion and Migration by Down-Regulating MMP-1. Dig Dis Sci 1:168-77.https://doi.org/10.1007/s10620-019-05739-8
Tian R, Li X, Gao Y, Li Y, Yang P, Wang K (2018) Identification and validation of the role of matrix metalloproteinase-1 in cervical cancer. International journal of oncology 4:1198-208.https://doi.org/10.3892/ijo.2018.4267
Bakir B, Chiarella AM, Pitarresi JR, Rustgi AK (2020) EMT, MET, Plasticity, and Tumor Metastasis. Trends Cell Biol 10:764-76.https://doi.org/10.1016/j.tcb.2020.07.003
Pastushenko I, Blanpain C (2019) EMT Transition States during Tumor Progression and Metastasis. Trends Cell Biol 3:212-26.https://doi.org/10.1016/j.tcb.2018.12.001
Yu J, Xu Z, Guo J, Yang K, Zheng J, Sun X (2021) Tumor-associated macrophages (TAMs) depend on MMP1 for their cancer-promoting role. Cell Death Discov 1:343.https://doi.org/10.1038/s41420-021-00730-7
Yoon JH, Choi YJ, Cha SW, Lee SG (2012) Anti-metastatic effects of ginsenoside Rd via inactivation of MAPK signaling and induction of focal adhesion formation. Phytomedicine 3-4:284-92.https://doi.org/10.1016/j.phymed.2011.08.069

No competing interests reported.

FigureS1S4.pdf

Download PDF

Version 1

posted

You are reading this latest preprint version

MMP1 overexpression promotes cancer progression and associates with poor outcome in head and neck carcinoma

Status:

Version 1

Abstract

Figures

Introduction

Materials And Methods

Results

Discussion

Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1