HOXB9 Promotes Laryngeal Squamous Cell Carcinoma Progression by Upregulating MMP12

Background ： The HOX family transcription factor HOXB9 is a crucial element in the progression of various cancers. In the previous study conducted by the investigators, a drastically higher HOXB9 expression was reported in laryngeal squamous cell cancer (LSCC), when compared to adjacent normal laryngeal squamous tissues. Furthermore, a high level of HOXB9 was closely correlated with histological grade and overall survival in LSCC patients. However, the underlying molecular mechanisms have not been fully elucidated. Results: The present study explored the molecular mechanisms of HOXB9 in LSCC progression. Furthermore, the in vitro and in vivo studies revealed that the gene knockout of HOXB9 using the CRISPR/CAS9 system inhibited cell proliferation, migration and invasion, and promoted cell apoptosis. Mechanistic studies in LSCC cell lines and human LSCC specimens demonstrated that HOXB9 promotes LSCC progression by directly upregulating the MMP12 expression at the level of its transcription. Conclusions: Collectively, the present study is the ﬁ rst to demonstrate the role of HOXB9 in the regulation of LSCC progression by enhancing the upregulation of MMP12.


INTRODUCTION
Among all head and neck cancers, laryngeal squamous cell carcinoma (LSCC) remains as the second most common malignant squamous cell carcinoma. 1 At present, the treatment of LSCC is mainly traditional surgical treatment supplemented by radiotherapy and chemotherapy. 2 However, there are many side effects and limited therapeutic effects with its long-term use. At present, a deep understanding of the underlying molecular mechanisms that promote LSCC remains elusive, and there is a lack of effective markers and related targets for the diagnosis and treatment of LSCC.
Therefore, it is of great significance to actively explore the key molecules involved in the occurrence and development of LSCC, in order to improve the treatment effect and survival rate of patients.
The HOX gene has been reported to play an important role in human embryonic development and cell differentiation. 3 According to the similarity of the human HOX gene sequence and chromosome location, the HOX gene can be divided into four clusters (HOXA-D), 4,5 and each cluster has 13 paralogous groups. 6,7 Aberrations in HOX gene expression often leads to abnormal development and malignant tumors. At present, a large number of literatures have reported that the HOX gene is involved in the occurrence and development of different tumors, such as breast cancer, leukemia, lung cancer and gastric cancer. [8][9][10][11][12][13] HOXB9 is one of the important members of the HOX family. As a transcription factor, this plays an important role in embryonic development and cancer progression. [14][15][16] There are few reports on the HOX gene and LSCC. At present, there is no large sample microarray and histological detection of the HOX gene in LSCC. In the previous study conducted by the investigators, it was found that the expression of HOXB9 is higher in LSCC than in paracancerous tissues. In addition, a high level of HOXB9 is notably correlated with the high histological grade and poor prognosis of LSCC patients. 17 However, the mechanism underlying HOXB9 in LSCC progression remains unknown. The present study explored the role of HOXB9 in the proliferation, migration, invasion and apoptosis of LSCC cells. In addition, the present data revealed that MMP12 is an important target gene regulated in the downstream of HOXB9. Importantly, HOXB9 promoted LSCC progression by upregulating the MMP12 protein expression in LSCC cells via direct binding to the promoter of MMP12.

Cell culture
Human LSCC cell lines (Hep-2 and AMC-HN-8) and a normal human keratinocyte cell line (HaCaT) were obtained from the Shanghai Institute of Biochemistry and Cell Biology (Shanghai, China). All cells were cultured in Dulbecco's modified Eagle's medium (DMEM) supplemented by 10% fetal bovine serum (FBS; Gibco, Carlsbad, CA, USA), and cultured at 37°C with 5% CO2.

CRISPR/cas9-mediated knockout of HOXB9 in LSCC cells
The CRISPR/Cas9 system was essentially used to target HOXB9. LSCC cell lines, including AMC-HN-8 and Hep-2 cells, were transfected with single-guide RNA (sgRNA) for three independent specific sequences of HOXB9 or nonspecific sgRNA.
These were designed using the online CRISPR design (http://crispr.mit.edu/). After 48 hours of infection, AMC-HN-8 and Hep-2 cells were screened with 1 μg/ml of puromycin for seven days to generate stably transfected cells. Then, these selected cells were expanded in regular culture medium (DMEM medium supplemented with 10% FBS). The HOXB9 protein expression levels in sgRNA-transfected or control cells were confirmed by western blot.

The reverse transcription-quantitative polymerase chain reaction (RT-qPCR) assay
The total RNA were extracted from different cell lines using Trizol reagent (Invitrogen, Carlsbad, CA, USA), according to manufacturer's instructions.
Approximately 2 μg of total RNA sample was reverse transcribed into cDNA using a high capacity cDNA kit (Thermo Fisher Scientific, Waltham, MA, USA). The relative mRNA levels were normalized against GAPDH, and calculated using the 2 -ΔΔCt method, all measurements were performed in triplicate.

Western blot analysis
The total protein obtained from the different cell lines (20 µg) were separated on 10% sodium dodecyl-sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a polyvinylidene fluoride (PVDF) membrane. After blocking with 5% bovine serum albumin (BSA) at 25°C for one hour, the membrane was incubated overnight at 4°C with different primary target protein antibodies or GAPDH. After incubating with the HRP-conjugated IgG antibody (1:8,000 dilution; Zhongshan Jinqiao Technology, Beijing, China) at 37°C for one hour, the proteins were visualized using an enhanced ECL Kit (Thermo Fisher Scientific, Waltham, MA, USA). GAPDH was used as a loading control, and the bands were quantified using the Quantity One Software v4.62 (Bio-Rad, Hercules, CA, USA). All measurements were performed in triplicate.

Cell proliferation assay
The role of HOXB9 on LSCC cell proliferation was evaluated using Cell Counting Kit-8 (CCK-8) assay (Beyotime Biotechnology; Shanghai, China), according to manufacturer's instructions. Briefly, Hep-2 and AMC-HN-8 cells were added to 96-well plates at a concentration of 3×10 3 cells per well. A total of 10 μl of CCK-8 solution was added for one hour into each well at different time points (24, 48, 72 and 96 hours), and the absorbance at 450 nm was measured using an enzyme-linked immunosorbent assay (ELISA) reader (Bio-Rad Laboratories, Inc., Hercules, CA, USA). The results were representative of three individual experiments performed in triplicate.

Cell cycle analysis
The different cell cycles were measured according to manufacturer's instructions (BD Biosciences, CA, USA). These were performed using flow cytometry, as previously described. 18 All experiments were repeated for at least three times.

Cell migration assay
In order to evaluate the changes in migration ability after the knockout HOXB9, wound healing and Transwell assay were separately performed. Briefly, Hep-2 and AMC-HN-8 cells were seeded on six-well plates, and grown up to 80-90% confluence.
Then, the cell monolayer was wounded using a 200-μl plastic tip, and the wounds were observed at different time points (0, 12 and 24 hours post-wounding) within the scrape line. The images were captured using a microscope, and the migration ability was analyzed by measuring the width of the wounds. In the Transwell assay, the cell suspension that contained 5×10 4 cells were seeded into the upper chamber (8 μm in size) of a 24-well plate in serum-free media. The lower wells contained 10% FBS.
After incubation for 24 hours, cells that passed through the inserts were fixed in 4% methanol for 20 minutes, and stained by 0.1% crystal violet. Then, the migrated cells were counted under a microscope, and images of three random fields were scanned.
These results were representative of three individual experiments.

Cell invasion assay
The Transwell champers were coated with 80 μl of Matrigel (BD Biosciences, CA, USA) on a 24-well plate. Different cells (5×10 4 ) were added to the upper Transwell chambers in serum-free media, while the lower wells contained 10% FBS.
After incubation for 24 hours at 37°C, cells that invaded on the lower membrane were fixed with methanol and stained with 0.1% crystal violet for 20 minutes. The number of invading cells were counted under a microscope, and the images of three random fields were captured. All measurements were repeated in triplicate.

TUNEL assay
The apoptosis of cells was detected using a TUNEL detection kit (Roche), according to manufacturer's instructions. Hep-2 and AMC-HN-8 cells (10 4 /ml) were inoculated on the cover glass of the 24-well plate, and cultured for 24 hours. Then, cells were fixed with 4% paraformaldehyde at room temperature for 20 minutes, and incubated with 0.1% Triton X-100 on ice for two minutes. Afterwards, the TUNEL reagent was added and incubated at 37℃ for 60 minutes, and cells were dyed with DAPI for five minutes, and observed under a microscope.

In vivo tumorigenicity assay
Ten BALB/c nude mice (5 weeks of age), which were obtained from Vital River Laboratories (Beijing, China), were housed in aseptic conditions. This animal experiment was approved by the Ethics Committee of Harbin Medical University. All mice were subcutaneously injected into the flank with 1×10 6 suspension of Hep-2 cells (150 μl of DMEM/Matrigel at 1:1 mixture). Then, the tumor sizes were measured every three days after becoming palpable under the skin. These tumor-bearing mice were sacrificed at four weeks when the tumor reached approximately 1.5 cm in diameter. These tumors were removed for weighing and further study.

Affymetrix microarray analysis
Six RNA samples (three from HOXB9-knockout cells and three from control sgRNA-transfected Hep-2 cells) were used for the microarray test. The total RNA (100 mg) was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA, USA), according to manufacturer's instructions. These microarray experiments were carried out, as we previously described. 17 The microarray data were analyzed using normalized transcript signals. Hierarchical clustering was applied to visualize the expression of all differential genes, the differences in gene expression between HOXB9-knockout and control sgRNA-transfected Hep-2 cell samples were calculated, and those that were greater than 2-fold upregulated/downregulated were considered significant.

Immunohistochemistry analysis
The samples were fixed in 4% formalin and paraffin-embedded.
Immunohistochemistry was performed, as previously described. 17 The slides were incubated with the primary antibody against HOXB9 (1:200; Abcam, Cambridge, MA, USA) and MMP12 (1:400; Abcam, Cambridge, MA, USA) overnight at 4°C . The sections were developed using diaminobenzidine (DAB), and all histological assessments were conducted by two experienced pathologists. The score of the immunohistochemical staining was calculated by multiplying the staining intensity with the stained area of positive cells. Staining intensity score: 0 for negative, 1 for low staining intensity, 2 for medium staining intensity, and 3 for high staining intensity. The staining area of positive cells was scored, as follows: 0, none; 1, <10%; 2, 10-50%; 3, >50%. A total score of 4 was used to distinguish between low (<4) and high (≥4) levels of HOXB9 and MMP12 gene expression.

Luciferase assay
The target cells (Hep-2 cells with or without HOXB9-knockout and HEK-293T cells transfected with pcDNA3.1 or pcDNA3.1-overexpressed HOXB9) were separately transfected with wild-type or mutant MMP12 luciferase reporter plasmids, and the inner control was the luciferase reporter gene. All cells in the logarithmic growth stage were digested with trypsin to make the cell suspension. After 48 hours of transfection, 100 μl of lysate was added into each pore and centrifuged at 12,000 × g for five minutes, and the supernatant was collected for determination. The activities of the firefly and Renilla luciferases were determined using the Dual-Glo Luciferase Assay System (E1960, Promega), according to manufacturer's instructions. The normalized data were calculated as the ratio of Renilla/firefly luciferase activities.

Chromatin immunoprecipitation (ChIP) assays
The HEK-293T and Hep-2 cell lines (1×10 7 ) were crosslinked with 1% formaldehyde for 10 minutes on ice, and subjected to ChIP assays using a specific antibody, according to manufacturer's protocol (Millipore). The ChIP DNA complex was precipitated and subjected to RT-qPCR using MMP12 sense primer F Furthermore, the number of cells in the G1 phase was significantly increased, the S phase was shortened, and the G2 phase had no significant changes ( Figure 3B). These results show that the cell cycle of LSCC cells was blocked in the G1 phase after HOXB9 knockout. Furthermore, in order to investigate the effect of HOXB9 on the migration ability of LSCC cells, wound-healing assays were used to detect the effect of HOXB9 on the migration ability of LSCC cells. By comparing the images taken at 12 and 24 hours after cell migration, it was found that the migration ability of Hep-2 and AMC-HN-8 cells decreased in the HOXB9-KO group, when compared to the control group ( Figure 3C). This suggests that HOXB9 can promote the migration of LSCC cells. Then, it the effect of HOXB9 on the invasiveness of LSCC cells was tested by Transwell assays. It was revealed that cell invasiveness decreased, when compared with the control group ( Figure 3D). In addition, TUNEL assays were performed to determine whether HOXB9 affects LSCC cell apoptosis. The results revealed that after HOXB9 was knocked out in LSCC cells, the apoptosis of cells increased in the HOXB9-KO group, when compared to the control group ( Figure 3E).
These results suggest that HOXB9 could inhibit the apoptosis of LSCC cells.
Furthermore, the mouse xenograft assays performed using HOXB9 knockout Hep-2 cells revealed that the tumor growth rate and tumor volumes significantly decreased in the HOXB9-KO group, when compared to the control group ( Figure 3F). This indicates that HOXB9 can influence the proliferation ability of LSCC cells. Thus, these data suggests that HOXB9 can affect the cellular functional changes of LSCC cells.

Screening of differentially expressed genes after the knockout of HOXB9 in LSCC cells
In order to investigate the downstream target gene regulated by HOXB9 in LSCC, the whole gene expression microarray was carried out after the CRISPR/Cas9-mediated deletion of HOXB9 in the Hep-2 cell line, and the significant differential expression genes were systematically screened out. The results revealed that there were 137 upregulated genes, 148 downregulated genes, and 21,614 genes without significant difference. The differences of these upregulated and downregulated genes are presented in three experimental groups and three control groups ( Figure 4A). Next, a series of genes correlated to tumor cell proliferation, invasion and apoptosis was obtained by Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis, as shown in the typical gene thermogram ( Figure 4B). A total of 10 representative tumor related genes were selected from these differentially expressed genes, and these were verified by RT-qPCR in Hep-2 cells with knocked out HOXB9 (three different biological repeats).
The results revealed that the mRNA levels of these genes were consistent with the results of the ChIP assay ( Figure 4C). According to the differential gene screening experiment, MMP12 was noted to have the most significant difference, and this was also reported to be overexpressed in multiple tumors, and involved in the process of tumor growth, invasion and metastasis. In order to verify the correlation between the expression of HOXB9 and MMP12, the protein expression level of these two genes was detected in Hep-2 cells. The results revealed that the expression of MMP12 was downregulated when HOXB9 was knocked out (Figure 4D), and upregulated when HOXB9 was overexpressed ( Figure 4E). These results show that the expression level of MMP12 changed with the HOXB9.

The expression between HOXB9 and MMP12 is positively correlated
In order to further clarify the correlation between the expression of HOXB9 and MMP12, the expression of HOXB9 and MMP12 was detected in 106 cases of LSCC and its adjacent tissues by immunohistochemistry. The results revealed that MMP12 was highly expressed in the cytoplasm of LSCC tissues, but lowly expressed in adjacent tissues ( Figure 5A).  (Figures 6D and 6E, P<0.001). These results show that there was physical binding between the HOXB9 and MMP12 promoter.

DISCUSSION
LSCC is one of the most common head and neck cancers in the world. Despite the improvements in treatment over the past 10 years, the mortality rate remains high. 19 The mechanism of LSCC remains unclear, and there is no effective treatment target.
Furthermore, the occurrence and development of LSCC is a complex process, which involves the interaction of many factors. In the previous study conducted by the investigators, a group of HOX genes were screened out by microarray, which were upregulated in LSCC tissues. Further large sample microarray studies have shown that the high HOXB9 expression level is highly correlated with LSCC histological grade and poor prognosis. 17 These results suggest that as a member of the HOX gene family, HOXB9 may be an effective biomarker for the development and prognosis of LSCC.
However, it remains unclear how HOXB9 participates in LSCC.
In the present study, an attempt was made to elucidate the relationship between HOXB9 and LSCC progression in vivo and in vitro, and screen the downstream target gene of HOXB9, which is a transcription factor. Previous studies have revealed that HOXB9 is abnormally expressed in many tumors. However, this has different methods of action in different tumors. 20 Matrix metalloproteinase (MMP) is a zinc-dependent endopeptidase family. MMP is considered to be an important inflammatory mediator, which regulate tissue remodeling in physiological and pathological processes. 33 MMPs have been proposed as a potential therapeutic target for various cardiovascular and musculoskeletal diseases and cancer. 34 MMP12 belongs to the elastase family of MMP, and its function is mainly involved in the process of inflammation, tissue repair, respiratory diseases, and tumor growth, invasion and metastasis. 35 Previous studies have reported that MMP12 is involved in the invasion and metastasis of non-small cell lung cancer (NSCLC). 36 In addition, it was found that the expression of MMP12 is significantly elevated in lung adenocarcinoma tissues and cell lines, and that this is highly correlated with the pathological stage and lymph node metastasis. Furthermore,

MMP12 knockout inhibits the proliferation and invasion of lung adenocarcinoma cells,
suggesting that MMP12 may be a target for the treatment of lung adenocarcinoma. 37 Some researchers have reported that the high level of MMP12 in serum is correlated to the poor prognosis of patients, suggesting that MMP12 may promote the invasion and metastasis of colon cancer cells. 38 In the present study, it was speculated that MMP12 may be the downstream target gene of HOXB9, which is regulated by HOXB9 transcription, and has enhanced activity in LSCC, thereby affecting the biological behavior of LSCC cells.
In order to confirm the interaction between HOXB9 and MMP12, the chromatin immunoprecipitation and luciferase reporter gene assay were respectively performed.
This confirmed the physical binding between the HOXB9 and MMP12 promoter, and •Consent for publication Not applicable.

•Availability of data and materials
The datasets during and/or analysed during the current study available from the corresponding author on reasonable request.

•Competing interests
The authors declare that they have no competing interests.

•Funding
This study was supported by the funds of the National Natural Science Foundation of China (Grant no. 82060495 to Chuanhui Sun and Grant no. 81272965, 81772874 to Yanan Sun ).

•Authors' contributions
Chuanhui Sun and Peng Wang performed the cytological experiment. Yujiang Chen and Changsong Han were major contributors in animal and pathological experiment.
Hua Deng and Qiuying Li performed the Immunohistochemical and western test.
Yanan Sun was a major contributor in analyzing the data and writing the manuscript.         The screening of differentially expressed genes after the knockout of HOXB9 in LSCC cells. (A) The microarray analysis of differentially expressed genes in Hep-2 cells after HOXB9 knockout. The red dots represent the upregulated genes, the green dots represent downregulated genes, and the gray dots represent the non-differentially expressed genes. Columns 1-6 represent the HOXB9 knockout (three biological repeats) and control group (three biological repeats), respectively. The change from green to red color indicates that the fold change is from low to high. (B) A typical gene thermogram is shown. The thermogram shows the hierarchical clustering, in which columns 1-6 represent the HOXB9 knockout group and control group, respectively. The change from green to red color indicates that the fold change is from low to high. (C) The RT-qPCR validation of the representative gene mRNA level is presented. The black columnar represents the result of the DNA microarray, while the white columnar represents the result of the RT-qPCR. It can be observed that these two results are basically the same.