MNX1 Promotes Malignant Progression of Cervical Cancer via Repressing the Transcription of p21cip1

Motor neuron and pancreas homeobox 1 (MNX1) is a development-related genes and has been found to be highly expressed in several cancers. However, its biological function in cervical cancer remains largely unexplored. QRT-PCR, western blot, and IHC showed that MNX1 was abnormally overexpressed in cervical cancer tissues and cell lines. The high expression level of MNX1 correlated with poorer clinicopathologic characteristics in cervical cancer patients. Evaluated by RTCA (Real Time Cellular Analysis) proliferation assay, colony formation assay, EdU assay, transwell assay, and matrigel assay, we found that knockdown of MNX1 inhibited proliferation, migration and invasion of cervical cancer in vitro, while overexpression of MNX1 promoted malignant phenotype of cervical cancer. And subcutaneous xenograft model confirmed the malignant phenotype of MNX1 in vivo. Furthermore, flow cytometry, chromatin immunoprecipitation, and luciferase reporter assay indicated that MNX1 accelerated cell cycle transition by transcriptionally downregulating cyclin-dependent kinases p21cip1. In summary, our study revealed that MNX1 exerted an oncogenic role in cervical cancer via repressing the transcription of p21cip1 and thus accelerating cell cycle progression. Our results suggested that MNX1 was a potential diagnostic marker and therapeutic target for cervical cancer patients.


INTRODUCTION
As one of the most common gynecological malignant tumors, cervical cancer is the fourth leading cause of cancer-related death among women worldwide (1). Although efforts (including periodic cancer screening, prompt surgical treatment, chemotherapy, and radiotherapy) have been made to decrease the mortality of cervical cancer, the prognosis of patients is still poor and cervical cancer remains an important public health issue (2). The pathogenesis of cervical cancer has not been clearly illustrated, but it is confirmed that the activation of tumor-promoting genes and the inactivation of tumor suppressor genes participate in the progression of cervical cancer (3). To screen for novel abnormally expressed genes functioning in cervical cancer may provide potential prognostic markers and therapeutic targets for treatment. MNX1 (Motor neuron and pancreas homeobox 1, also known as HB9, HLXB9) is a member of homeobox gene family and encodes a nuclear protein (4). The homeobox genes are a group of genes containing homeobox (a 180 base pairs long DNA sequence) and encode homeodomain proteins that act as transcription factors (5). Many homeobox genes have been proved to be implicated in various human cancers, acting as oncogenes, or tumor suppressors (6)(7)(8). MNX1 was firstly found to be expressed in lymphoid and pancreatic tissues and defined as a novel human homeobox gene in 1994 (9). Early studies showed that MNX1 was involved invertebrate and pancreatic development (10,11) and motor neuronal differentiation (12). Defects in this gene result in Currarino syndrome, an autosomic dominant congenital malformation (13). In follow-up study, MNX1 was found to be abnormally expressed in several cancer types, including prostate cancer, hepatocellular carcinoma and acute myeloid leukemia (14)(15)(16). Furthermore, recent studies confirmed that MNX1 played oncogenic roles in colorectal cancer, breast cancer, and bladder cancer (17)(18)(19).
The aim of this study is to identify the expression and function of MNX1 in cervical cancer. Our results revealed that MNX1 was significantly upregulated in cervical cancer and correlated with poorer prognosis. The knockdown or overexpressed MNX1 inhibited or promoted aggressiveness of cervical cancer, including proliferation, migration, and invasion capacities, by enhancing or repressing the transcription of p21 cip1 thus regulating the G2/M cell cycle transition. These findings suggested that MNX1 might be a potential diagnostic marker and therapeutic target for cervical cancer.

Human Cervical Cancer Tissues
The 40 pairs of cervical cancer tissues and adjacent tissues were selected from the Affiliated Cancer Hospital of Nanjing Medical University and informed consent was obtained from all subjects. All tumors and paired non-tumor tissues were confirmed by experienced pathologists and no patients received chemotherapy or radiotherapy before surgery. The mRNA expression of MNX1 and p21 cip1 in cervical cancer tissues was detected by qRT-PCR. Collection of human tissue samples was conducted in accordance with the International Ethical Guidelines for Biomedical Research Involving Human Subjects (CIOMS). This study was approved by the Ethics Committee of the Nanjing Medical University Affiliated Cancer Hospital.

Tissue Microarrays
Paired cervical cancer tissue microarrays were obtained from Shanghai Outdo Biotech Co., Ltd. (Cat. No. OD-CT-RpUtr03-004 and OD-CT-RpUtr03-006). Totally 62 pairs of paraffinembedded human cervical cancer sections were analyzed for MNX1 expression. All tissues were examined by two experienced pathologists and the TNM stage was confirmed in each patient with blinded methods. The sections were incubated with an anti-MNX1 primary antibody (1:100, Abcam, ab79541). The IHC scores were calculated according to intensity and percentage of positive cells. The staining intensity was evaluated as the basis of four grades: 0 (negative staining), 1(weak staining), 2 (moderate staining), or 3 (strong staining). The product (percentage of positive cells and respective intensity scores) was used as the final staining scores (a minimum value of 0 and a maximum value of 300).

RNA Preparation, Reverse Transcription, and qRT-PCR
TRIzol reagent (Invitrogen, Carlsbad, CA, USA) was used to extract total RNA from tissue samples or cultured cells according to the manufacturer's protocol. A Reverse Transcription Kit (Takara, Cat: RR036A, KeyGEN) was utilized to generate cDNA. QRT-PCR was performed with SYBR Select Master Mix (Applied Biosystems,Cat: 4472908. KeyGEN, Nanjing, China) and primers are shown in Table S2.

siRNA and Plasmid Transfection
The siRNAs targeting MNX1 and p21 cip1 were conducted and purchased from RiboBio, Guangzhou, China. All siRNA sequences are shown in Table S3. The full-length cDNA of human MNX1 were PCR-amplified and cloned into the expression vector Pgpu6/gfp/neo (Vigene Biosciences, Shandong, China). The siRNAs and overexpression plasmids were transfected into cervical cancer cells according to the Lipofectamine 3000 Reagent (Invitrogen, Carlsbad, CA, USA) protocol. Non-sense RNAi (si-NC) and empty plasmids (oe-NC) was used as negative controls.

Cell Proliferation Assay
The cell proliferation assays were performed 24 h after transfection. For Real TimexCELLigence analysis system (RTCA), 8,000 cells/100 µL were seeded in E-plates, and the plates were locked into the RTCA DP device in the incubator to calculate the "cell index" value. In colony formation assay, a total of 200 cells were placed in afresh 6-well-plate and the cells were stained with 0.1% crystal violet solution after 10-14 days. For 5-ethynyl-2 ′ -deoxyuridine (EdU) assay (keyFluor488 Click-IT EDU Kit, RiboBio, Guangzhou, China), the transfected cells were placed in 96-well-plates (8,000 cells/well) overnight in a CO 2 incubator. Then, cells were incubated with 100 µL/well of 10 µM EdU for 2 h at 37 • C and fixed with 50 µL 4% paraformaldehydecontaining PBS for 30 min at room temperature. Subsequently, the cells were cultured for 5 min with 50 µL of 2 mg/mL glycine and then washed with 100 µL 3% BSA in PBS. After permeabilization with 0.5% Triton X-100 for 20 min, the cells were cultured with 1 × Click-iT reaction solution for 30 min at room temperature in dark conditions. After that, cells were incubated with 100 µL/well of 1× Hoechst 33,342 solutions for 30 min at room temperature in the dark after washing with 100 µL of PBS. The cells were then imaged using fluorescence microscopy and proliferation cell ratios were counted from five random fields in every well. Each experiment was repeated three times. A total of 400 cells in a fresh six-well-plates were maintained in medium containing 10% FBS; the medium was replaced every 3 or 4 days. After 2 weeks, the cells were fixed with 4% paraformaldehyde and stained with 0.1% crystal violet. Each experiment was repeated three times.

Migration and Invasion Assay
For wound healing assay, cells were growing on the 6-wellplate, then artificial scratch on a confluent monolayer of cells was created with a 200 µL pipette tip. The medium was replaced with the serum-free and cells imaged 48 h later. For transwell and matrigel assay, totally 40,000 transfected cells were added to the upper chamber of Transwell assay inserts (8 µM PET, 24-well Millicell) or a Matrigel coated membrane (BD Biosciences) containing 200 µL serum-free DMEM media. The lower chambers were filled with 800 µl DMEM media containing 10% FBS. After a 24-h (migration assay) or 48-h (invasion assay) incubation, the cells were fixed with 4% polyformaldehyde, stained with crystal violet, and imaged. Migration and invasion were assessed by counting cell nuclei from five random fields on every filter. Each experiment was repeated three times. RTCA was also used to evaluated the ability of migration and invasion. CIM-plates installation and baseline measurement was carried out according to the instructions. Add 100 µl of mixed, serumfree cell suspension (4 × 10 4 cells) to the upper chamber in CIMplates, and the plates were locked into the RTCA DP device in the incubator to calculate the "cell index" value.

Cell Cycle Analysis
Cells were digested with 0.25% trypsin-EDTA and fixed with 70% ethanol for 12 h at 4 • C. The ethanol-suspended cells were centrifuged and stained with PI staining solution for 10 min in the dark at 37 • C. A FACSCalibur flow cytometer was used to detect cell cycle distribution. The percentage of the cells in G0-G1, S, and G2-M were counted and compared.

Chromatin Immunoprecipitation (ChIP)
Cells were cross-linked in 4% paraformaldehyde and the reaction was quenched with glycine. After two washes with cold PBS, cells were added with pre-cooling PBS containing cocktail (Halt TM Protease Inhibitor Cocktail, Thermo Scientific, #78430) and scraped into a centrifuge tube. The cells were centrifuged for 10 min at 800 g at 4 • C, then added with 500 µL cell lysis buffer (containing 2.5 µL cocktail) and incubated on ice for 15 min. Cells were then centrifuged for 5 min at 800 × g, 4 • C and cell precipitates were resuspended in 500 µL nucleus lysis buffer (containing 2.5 µL cocktail). The cells were sonicated (amplitude 30%) on ice for 10 min and soluble chromatin was obtained by centrifuging for 10 min at 12,000 g at 4 • C. Five micrograms of anti-MNX1 antibody (Sigma-Aldrich; SAB2101494) coupled to magnetic beads (Resin M2, Sigma, Shanghai, China) was used to immunoprecipitate the DNAprotein complex, and the IgG antibody was used as a negative control. The immunoprecipitation products were washed with 500 µL low salt buffer, high salt buffer, LiCi buffer, and TE buffer successively, all for 5 min at 4 • C. The ChIP elution buffer (containing proteinase K) was used for DNA purification. The beads were wiped out on a magnetic frame and the DNA was eluted with elution buffer C from adsorption column. ChIP DNA samples were subjected to PCR amplification with primers specific to p21 cip1 promoter region. PCR products were then used for agarose gel electrophoresis. The sequence of primers used are shown in Table S4 and GAPDH was used as a control.

Luciferase Reporter Assay
The p21 cip1 (CDKN1A) promoter region (−2,000 bp) was amplified and cloned into luciferase reporter plasmid (pGL3basic). The p21 cip1 promoter wild-type plasmids or mutanttype plasmids were co-transfected with CMV-MNX1 expression plasmids in HEK293T cells, and CMV-empty vectors were used as a negative control. Relative luciferase activity was corrected for Renilla luciferase activity of pGL3-basic, and normalized to the activity of the control.

Xenograft Model
All animal studies were conducted in accordance with NIH animal use guidelines and protocols were approved by Nanjing Medical University Animal Care Committee. Sixteen female nude mice (4-6 weeks old) were purchased from Nanjing Medical University School of Medicine's accredited animal facility. The mice were randomly divided into two groups using random number generator. In each group, 1.0 × 10 6 exponentially growing cervical cancer cells were injected in axilla subcutaneously. Before tumor transplantation, cells were transfected with shRNAs or overexpression plasmids. The transfection was performed by transient transfection according to the specification of Lipofectamine 3000 (Invitrogen, Carlsbad, CA, USA). The sh-NC and empty vector pcDNA3.1 were used as controls and totally 5 µg plasmid vectors were transfected into cells for each group. The sequences of shRNAs are shown in Table S5. Tumors were harvested at 6 weeks after injection. The weight of tumor was measured on the scale and tumor volume was estimated using calipers ([length × width 2 ]/2). And tissues were immunohistochemically stained with MNX1 (Abcam, ab79541, 1:100), Ki67 (Abcam, ab79541, 1:100), and p21 cip1 (Abcam, ab109520, 1:100). Western blotting was performed as previously described. No blinding was done in the animal studies.

Statistical Analysis
Results are presented as the mean ± standard deviation (SD). Statistical analyses were performed using SPSS Statistics (version 20.0, Chicago, Ill) and GraphPad Prism 6 software (GraphPad Software, Inc., La Jolla, CA, USA). P < 0.05 was considered statistically significant.

Overexpression of MNX1 Correlates With Poorer Prognosis and More Aggressive Clinical Features
Analysis of TCGA dataset revealed that the mRNA expression of MNX1 was remarkably upregulated in cervical cancer tissues compared with para-tumor tissues (p = 0.0003, Figure 1A). In GEPIA (Gene Expression Profiling Interactive Analysis) website, patients with higher expression of MNX1 bore a worse disease free survival (n high = 73, n low = 72, p = 0.019, Figure 1B). The expression of MNX1 in cervical cancer tissues were significantly higher than adjacent tissues in 85% (34/40) of 40 cervical cancer patients (p < 0.001, Figures 1C,D). IHC assays based on tissue microarrays (TMAs) were performed to detect the protein expression of MNX1 in 62 paired human cervical cancer tissues and para-tumor tissues, and results showed that staining scores of MNX1 were higher in cancer tissues (p < 0.0001, Figure 1E).

Knockdown of MNX1 Inhibited Progression of Cervical Cancer in vitro
To evaluate the expression of MNX1 in cell lines, qRT-PCR and western blotting were performed and results showed that MNX1 was generally upregulated in cervical cancer cell lines compared with normal human cervical cell lines (Hacat) (Figures 2A,B). To further investigate the biological function of MNX1 in cervical cancer, two specific siRNAs targeting MNX1 were transfected into HeLa and Siha cells. Both two siRNAs showed favorable suppression efficiency in HeLa (Figures 2C,D) and Siha cells (Figures 2E,F). The RTCA proliferation assay (Figure 2G), EDU assay (Figure 2H), and colony formation assay (Figure 2I) showed that knockdown of MNX1 inhibited the proliferation ability of cervical cancer in HeLa and Siha cells. Moreover, RTCA migration assay (Figure 2J), transwell assay, and matrigel assay (Figure 2K), and wound healing assay ( Figure 2L) revealed that silencing MNX1 inhibited the ability of cervical cancer cells to migrate and invade. These results suggest that MNX1 plays a vital role in the malignant phenotype of cervical cancer.

Ectopic Expression of MNX1 Enhanced Aggressiveness of Cervical Cancer in vitro
To further verify the biological role of MNX1 in cervical cancer, a pcDNA3.1 plasmid to overexpress MNX1 was constructed and transfected into C33a and HeLa cells. The plasmid effectively upregulated the expression of MNX1, confirmed by qRT-PCR and western blotting (Figures 3A,B). Consistently, our results showed that ectopic expression of MNX1 promotes proliferation, migration, and invasion (Figures 3C-G) of cervical cancer cells.
si-MNX1 Induced G2/M Cell Cycle Arrest and Upregulated the Expression of p21 cip1 Two hundred and eight genes highly correlated with MNX1 were used for GO, KEGG, and Reactome pathway analysis. Results showed that MNX1 may participate in "transcription" and "metabolism" pathway ( Figure 4A). Cell cycle detection showed that knockdown of MNX1 induced G2/M cell cycle arrest in HeLa and Siha cells ( Figure 4B). Furthermore, we examined the effect of MNX1 on the expression of cell cycle key-related genes, including p15 ink4b , p16 ink4a , p21 cip1 , p27 kip1 , CDK1, CDK2, CDK4, cyclinB1, cyclinD1, and cyclinE1. Both in HeLa and Siha cells, knockdown of MNX1 upregulated the expression of p21 cip1 , which has been confirmed as a tumor suppressor gene in multiple cancers (Figure 4C). And western blotting results suggested that knockdown of MNX1 increased the expression of p21 cip1 while decreased the expression of phosphorylated CDK1 (p Thr161 -CDK1), a downstream effector of p21 cip1 (Figure 4D). Consistently with these results, ectopic expression of MNX1 decreased the expression of p21 cip1 while increased the expression of p Thr161 -CDK1 in C33a and HeLa cells (Figures 4E,F). It suggested that MNX1 might exerted its biological function via modulating the expression of p21 cip1 .

MNX1 Suppressed the Expression of p21 cip1 via Binding to Its Promoter Region
Our previous results showed that knockdown or ectopic expression of MNX1 altered the expression of p21 cip1 . To further verify the mechanism, we analyzed the correlation between MNX1 and p21 cip1 in 40 cases of CC samples, and the results were shown that MNX1 and p21 cip1 had a negative correlation (n = 40, p < 0.001) ( Figure 5A). As transcription factors usually bind to sequence-specific DNA to regulate transcription, we utilized Jaspar Database to predict the binding site between MNX1 and the promoter region (upstream 2,000 bp of coding region) of CDKN1A (the gene symbol of p21 cip1 ). It turned out that MNX1 was predicted to have four binding sites with the promoter region of CDKN1A, of which 1,371-1,380 bp (AACAATAAAT) and 226-235 bp (GCCCATTAAT) showed higher combination scores ( Figure 5B). Accordingly, the wild CDKN1A promoter region and mutant types (226-MT and 1371-MT) were generated and cloned into luciferase reporter vector (pGL3-basic; Figure 5C). And in luciferase reporter assay, overexpression of MNX1 inhibited the transcriptional activity of the wild CDKN1A promoter but not mutant type ( Figure 5D). Moreover, ChIP assay also revealed that MNX1 bound to the p21 cip1 promoter region in HeLa and Siha cells (Figures 5E,F).

Silencing p21 cip1 Rescued the Function of si-MNX1
To determine whether the function of MNX1 was relied on p21 cip1 , we designed three siRNAs (Table S3) to knockdown the expression of p21 cip1 in HeLa cells. The si1-p21 cip1 showed the best transfection efficiency ( Figure 6A) and it was used for the following experiment. RTCA proliferation assay, colony formation assay, EDU assay, transwell assay, matrigel assay, and would healing assay revealed that silencing p21 cip1 partially rescued the decreased proliferation, migration, and invasion ability of HeLa cells caused by knockdown of MNX1 (Figures 6B-F). And western blotting showed that the protein level of p21 cip1 and p Thr161 -CDK1 were partially reversed by silencing p21 cip1 (Figure 6G).

MNX1 Promoted Tumor Growth of Cervical Cancer in vivo
The xenograft models were used to explore the function of MNX1 in vivo. The shRNA-MNX1 (shRNA-NC as control) was transfected into HeLa cells and the knockdown efficiency was confirmed by qRT-PCR and western blotting (Figures 7A,B).
Results showed that knockdown of MNX1 inhibited tumor growth (measured by tumor weight and volume) in vivo (Figures 7C-E). IHC staining and western blotting of harvested tumors revealed that knockdown of MNX1 upregulated the protein level of p21 cip1 and downregulated ki-67 and p Thr161 -CDK1 in vivo (Figures 7F,G). Moreover, ectopic expression of MNX1 promoted tumor growth and altered the expression of p21 cip1 and ki-67 in vivo (Figures 7H-K).

DISCUSSION
In this study, we identified MNX1, a transcription factor of homeobox family, was significantly upregulated and involved in the progression of cervical cancer. The overexpression of MNX1 correlated with advanced clinical stages and poorer prognosis of cervical cancer patients. Furthermore, MNX1 exerted its oncogenic role via modulating the expression of p21 cip1 , especially by targeting the promoter region of p21 cip1 thus to repress its transcription. In accordance with our findings, a recent article showed that MNX1 had a role in the progression of cervical cancer, partially through upregulating cell cycle regulators CCNE1 and CCNE2 (24). And MNX-AS1, the antisense lncRNA of MNX1, was also reported to promote the invasion and metastasis of gastric cancer through repression of CDKN1A (25). All this results indicated that MNX1 played a critical role in cancer growth and cell cycle progression, and MNX1 might serve as a useful diagnostic and treatment target for cervical cancer. MNX1is a member of homeobox gene family, which all contain a homeobox (a DNA sequence, around 180 base pairs long) and encode homeodomain protein products as transcription factors (26). This cluster of genes has been identified to participate in the regulation of development and morphogenesis in animals, fungi, and plants (27). For example, CDX1, which is stably expressed in the human intestine, plays an important role in embryonic epicardial development (28,29). And the protagonist of our study, MNX1, participates in motor neuron development and neurodegenerative processes of zebrafish (30) and moreover controls cell fate choice in the developing endocrine pancreas (31). In recent years, more and more researches uncovered the role of development-related homeobox genes in carcinogenesis and these genes show great application prospect in tumor diagnosis and prevention, as the role of carcino-embryonic antigen (CEA) in gastroenteric tumors and alpha fetal protein (AFP) in liver cancer (32)(33)(34). For instance, PDX1 is a key regulator in pancreatic development and β-cell function (35) and meanwhile dynamically regulates pancreatic ductal adenocarcinoma initiation and maintenance (36). HOXC13, a highly conserved transcription factor involved in morphogenesis of all multicellular organisms, is aberrantly expressed and associated with cancer progression in esophageal cancer (37), lung adenocarcinoma (38), and liposarcomas (39). Likewise, MNX1 has been reported to promote sustained proliferation in bladder cancer by upregulating CCNE1/2 (40) and to act as a novel oncogene in prostate cancer (41). And in our study, MNX1 was also confirmed to be upregulated in cervical cancer and enhance the progression of cervical cancer.
In terms of mechanism, we found that MNX1 promoted tumor growth of cervical cancer via accelerating the progression of the cell cycle, especially by modulating the expression of p21 cip1 . Cell cycle is a vital process by which a cell leads to duplication and disorders of the cell cycle regulation may lead to tumor formation (42). The cell cycle progress is determined by two types of regulatory factors, cyclins and cyclindependent kinases (CDKs) (43). Active cyclin-CDK complexes phosphorylate proteins to elevate the expression levels of cyclins and enzymes required for DNA replication (44). Conversely, the cell cycle progression can be prevented by inhibitors by binding to and thus inactivating cyclin-CDK complexes, such as p21 cip1 , p27 kip1 , p16 ink4a , and so on (45). The p21 cip1 , also known as cyclin-dependent kinase inhibitor 1 (CDKN1A), has been identified as a regulator of cell cycle and a tumor suppressor in multiple kinds of cancers (46). Our results proved that MNX1 repressed the transcription of p21 cip1 by directly targeting its promoter region and furthermore promoted the phosphorylation of downstream CDK1. The MNX1-p21 cip1p Thr161 CDK1 axis played crucial roles in the progression of cervical cancer and meanwhile provided new evidence for the pathogenesis of cervical cancer. Moreover, the association between cervical cancer and HPV has long been identified. As a sexually transmitted agent, HPV are involved in transformation and maintaining of transformed status (47). Many studies have reported that HPV can also alter the expression of p21 (48)(49)(50). Thus, we searched the GEO dataset to seek for information about the relationship between MNX1 and HPV viral infection. We analyzed the GSE dataset 103,546 and found that there were no significant changes in the expression of MNX1 (NM_005515) in HaCat cells infected with HPV11E6 or HPV18E6. In GSE3292 (GDS1667), HPV positive or negative head and neck squamous cell carcinoma (HNSCC) showed no expression differences of MNX1 ( Figure S1). This information suggests that MNX1 might not be directly involved in HPV carcinogenesis and further investigations are needed in the future. In addition, cervical cancer is almost invariably associated with p53 loss (either mutation of HPV infection) and p53 is a very well know activator of p21. In this study, we proved that MNX1 exerted an oncogenic role in cervical cancer via suppressing the expression of p21 with binding to its promoter region. And we performed qRT-PCR and western blot in cervical cancer cells and results showed that knockdown of MNX1 did not affect the expression of p53. Transcription factor could directly regulate the expression of target genes by binding to the gene promoter. We speculated that MNX1 mediated downregulation of p21 was independent on p53. And in this study, three cervical cancer cells (HeLA, Siha, and C33A) were used to study the function of MNX1. We noted that p53 was lowly expressed in Hela cells, positively expressed in Siha cells and mutant of codon 273 in C33A cells. It indicated that MNX1 could function in cervical cancer cells with different p53 status. Moreover, it has been reported that in p53 mutation acute myeloid leukemia, MNX1 was identified as one of the hub genes from the protein-protein interaction network (51). And in hematopoietic stem and progenitor cells, the oncogenic potential of MNX1 was mediated via p53-p21 signaling pathway (52). In breast cancer, enrichment analysis suggested that MNX1 is probably involved in biological processes and pathways related to nuclear division, cell cycle, and p53 signaling (18). And a recent study showed that MNX1 had a role in the progression of cervical cancer, partially through upregulating cell cycle regulators CCNE1 and CCNE2 (24). These studies suggested that the function of MNX1 might be translatable to other tumor types.
In summary, we identified the homeobox member MNX1 as a tumor-promoting gene in cervical cancer. The upregulated MNX1 correlated with more advanced clinical pathological characteristics and poorer prognosis of cervical cancer patients. And MNX1 exerted its oncogenic role by repressing the transcription of p21 cip1 thus to promote the progression of cell cycle. We believe that there are other genes beside p21 regulated by MNX1 in cervical cancer. RNA-seq and ChIP-seq experiments may need to confirm this cluster of genes. And as the downregulation of MNX1 inhibited tumor growth of cervical cancer, MNX1 may represent promising targets for the diagnosis and anti-tumor therapy in cervical cancer patients. Through knockdown of MNX1, it might combine the function of MNX1 with chemotherapy.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary files, further inquiries can be directed to the corresponding author/s.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Ethics Committee of the Nanjing Medical University Affiliated Cancer Hospital. The patients/participants provided their written informed consent to participate in this study. The animal study was reviewed and approved by Nanjing Medical University Animal Care Committee.

AUTHOR CONTRIBUTIONS
LX, EL, and BR designed and supervised the study. BZ, YW, and JL were responsible for acquisition of data, interpretation of data, and article drafting. QZ contributed to experiments in vitro. JH and QL helped to analyze the data and revise the article. All authors contributed to the article and approved the submitted version.