Epigenetic inactivation of HOXD10 is associated with human colon cancer via inhibiting the RHOC/AKT/MAPK signaling pathway
To examine the influence of HOXD10 on the metabolism and growth of colon carcinoma cells by suppressing the RHOC/AKT/MAPK pathway.
Thirty-seven paired colon cancer and its adjacent samples from The Cancer Genome Atlas (TCGA) were analyzed. Chip Analysis Methylation Pipeline (ChAMP) analysis was employed for differential methylated points (DMPs) and the differential methylation regions (DMRs) screening. The HOXD10 mRNA expression and DNA methylation levels were detected by RT-PCR. The Cell proliferation, migration, invasion and apoptosis were respectively measured by MTT assay, transwell assay, wound healing assay and flow cytometry assay in carcinoma cell lines after treated with 5-aza-2′-deoxycytidine (5-Aza-dC) or transfected with HOXD10-expressing plasmid. The expression of HOXD10 and RHOC was revealed by immunohistochemistry in disparate differentiation colon carcinoma tissues, and the dephosphorylation of AKT and MAPK pathways were detected by RT-PCR and western blot.
The bioinformatics analysis demonstrated that HOXD10 was hypermethylated and low-expressed in colorectal cancer tissues. The detection of RT-PCR indicated the similar results in colorectal cancer cell lines and tissues. The induction of demethylation was recovered by treatment with 5-Aza-dC and the HOXD10 in colorectal cancer cell lines was re-expressed by transfection with a HOXD10 expression vector. The demethylation or overexpression of HOXD10 suppressed proliferation, migration, invasion and promoted apoptosis in colorectal cancer cells. HXOD10 suppressed the tumor growth and detected an opposite trend of protein RHOC. AKT and MAPK pathways were notably inactivated after the dephosphorylation due to the overexpression of HOXD10.
HOXD10 was suppressed in colon adenocarcinoma cells, which down-regulated RHOC/AKT/MAPK pathway to enhance colon cancer cells apoptosis and constrain the proliferation, migration and invasion.
KeywordsColon cancer Methylation 5-Aza-dC HOXD10 RHOC/AKT/MAPK pathway
Chip Analysis Methylation Pipeline
Differential methylated points
Differential methylation regions
DNA methyltransferase inhibitors
Extracellular signal-regulated kinase
Mitogen-activated protein kinase
Specific mortality rate
Specific incidence rate
The Cancer Genome Atlas
The American Cancer Society (http://www.cancer.org) showed that colorectal cancer was the third leading cause of morbidity and mortality in the United States (US). In addition, the specific incidence rate (SIR) and specific mortality rate (SMR) of CRC was the highest in the World Health Organization (WHO) European region . The most effective therapeutic method for patients with CRC was complete surgical resection, provided the best possibility of long-term survival . It has been illustrated that CRC was still a major health problem in several studies [3, 4, 5] and the cancer burden of CRC was anticipated to increase over the next several decades in both developed and developing countries . However, the etiology and underlying molecular mechanisms of colorectal cancer are still unclear.
Epigenetic gene silencing of anti-oncogene through promoter DNA hypermethylation is a common feature in human cancers . DNA methylation refers to the enzymatic addition of a methyl group to the 5 position of cytosine ring by DNA methyltransferase (DNMT) to generate 5-methylcytosine , which plays a vital role in the colorectal carcinogenesis . DNMT inhibitor 5-aza-2′-deoxycytidine (5-Aza-dC) is already used in the investigation of many different kinds of cancer, such as gastric cancer , breast cancer  and urothelial carcinoma  on DNA methylation of regulatory elements and gene expression. So far the knowledge of studies which investigate DNA methylation profile in recurrent CRC has been very limited.
HomeoboxD10 (HOXD10) is a member of the homeobox gene family, which serves as important transcription factors targeting carcinogenesis-related proteins [13, 14, 15, 16]. HOXD10 is regarded as the primary effector that negatively regulates neoplasm metastasis . Emerging evidence indicates that the expression of HOXD10 gene is regulated by DNA methylation. In gastric carcinogenesis, HOXD10 gene is downregulated through promoter hypermethylation . In hepatocellular carcinoma, HOXD10 gene is silenced by promoter region hypermethylation, which is associated with ERK signaling . The dysregulation of HOXD10 gene may be involved in diverse pathways that play important roles in tumorigenesis . In malignant breast cancer, loss of HOXD10 expression during the progression leads to the increment of the pro-metastatic gene RHOC . In gastric tumors, microRNA-10b can promote cell invasion and provoke the up-regulation of RHOC and phosphorylation through targeting HOXD10 . However, the methylation status of HOXD10 and mechanism of action in colon cancer with RHOC and AKT pathway are still unclear.
The mitogen-activated protein kinase (MAPK) pathway is a key regulator for apoptosis related to most of the hypermethylated genes while the PI3K/AKT signaling pathway is involved in proliferation process in colorectal cancer . MAPK pathway is over expressed and associated with functional mutation of HOXD10 gene in human cholangiocellular carcinoma  and ovarian cancer . The phosphorylation activation of extracellular signal-regulated kinase (ERK) is a vital regulator for the metastasis and viability of cancer cells . Nevertheless, the underlying molecular mechanisms between the above-mentioned pathways and CRC-associated gene HOXD10 remain unknown.
This study was designed to confirm the mechanisms and the expression level of HOXD10 in CRC. We determined HOXD10 for the follow-up studies, which showed hypermethylation and decreased mRNA expression in CRC. 5-Aza-dC treatment can alter the DNA methylation level of HOXD10. Our results revealed the overexpression of HOXD10 had adverse influence on colorectal cancer.
For RT-PCR analysis, 15 pairs of frozen colon adenocarcinoma and its adjacent normal tissue specimens were collected from patients with CRC that were diagnosed from 2016 to 2017 at the Department of Gastroenterology and Hepatology, Sun Yat-sen Memorial Hospital. No other therapy, including radiotherapy, chemotherapy was performed prior to entry into the research. Samples used in the study were certified by local ethics committees, and all subjects were given informed consent from patient with available follow-up information.
The colon cancer dataset was obtained from The Cancer Genome Atlas (TCGA) data portal (https://gdc.cancer.gov/). Data for 74 patients were available with complete DNA methylation and were evaluated via the Illumina Infinium Human Methylation 450 BeadArray platform. DNA methylation index (MI) was accounted as β-values. The mean methylated (M) and unmethylated (U) signal intensities for each sample and locus were calculated by the formula (β = M/ [M + U]).
Demethylation with 5-Aza-dC
5-Aza-2′-deoxycytidine (5-aza-dC) (Sigma-Aldrich, USA) was dissolved in DMSO at 50 mg/ml. Cell lines were plated in 1 × 106 cells/ml for 24 h and treated with 0.5 μM 5-Aza-dC in 0.5% DMSO for 24 h, before growing for 7 days. Cells were harvested for RNA and DNA extraction.
Sequences of MSP primers for qRT-PCR
5’ GAATTTGGTAGGTCGAAGGAC 3’
5’ ATAAACCGCCCTACGAAAAC 3’
5′ GGAATTTGGTAGGTTGAAGGAT 3’
5’ AATAAACCACCCTACAAAAAC 3”
Quantitative real-time RT-PCR (qRT-PCR)
Sequences of primers for qRT-PCR
Expression vector for HOXD10
Full-length HOXD10 cDNA (GenBank accession number NM_002148.3) was cloned into the pcDNA3.1 expression vector. Transient transfection was performed using Lipofectamine 3000 (Intrivogen, Carlsbad, CA) according to the manufacturer’s instructions. Empty pcDNA3.1 expression vector was transfected as vector control.
Cell pellets were solubilized in electrophoresis sample buffer, sonicated for 10 s, and boiled for 10 min. The protein concentration of cell lysates was measured, and 3 mg protein was separated by SDS-PAGE, and then transferred to a polyvinylidene fluoride microporous membrane (MILLI-PORE, Billerica, MA). Membranes were blocked in PBS with 5% milk powder and probed with anti-AKT (ab81283, ABcam, USA), anti-ERK (ab54230, ABcam, USA), anti-β-actin (ab11003, ABcam, USA), anti-RHOC (ab64659, ABcam, USA), anti-p-ERK (ab201015, ABcam, USA), ani-p-AKT (ab38449, ABcam, USA), at 4 °C overnight. After washing, the membrane was cultured with HRP-linked goat anti-rabbit Ab (Cell Signaling Technology). Blots were visualized by ECL (ECL Plus; Amersham Pharmacia Biotech, Uppsala, Sweden), on the basis of the manufacturer protocol. Results were captured digitally using a LAS1000 Lumino Image Analyzer (Fuji Photo Film, Tokyo, Japan).
SW480 and LoVo cells were incubated at 1 × 104 cells/well in 96-well plates. Cells were harvested for 24 h at 37 °C in the presence of 5% CO2, before the medium took the place of serum-free medium. The cell lines were treated with 5-Aza-dC (1 μM) at 0 h, 24 h, 48 h, 72 h, 96 h and cell proliferation was detected using the MTT assay. Absorbance at 490 nm was surveyed by a microplate reader (Labcompare, San Diego, CA, USA).
Wound healing assay
The tumor cell migration capacity was evaluated by the wound-healing assay. In short, 1 × 106 cells were seeded in six-well plates, cultured overnight, and transfected with HOXD10 expression vector or empty vector. The cell layer was scratched with a sterile plastic tip when the culture had reached approximately 90% confluency, followed by washing with culture medium twice and cultured again for 48 h with serum-reduced medium containing 1% FBS. At different point in time, photographic images of the plates were captured under a microscope and the data were summarized on account of sextuple assays for each experiment.
The ability of cell invasion was appraised by 6.5-mm Transwell chambers with a pore size of 8 μm (Corning, USA). A concentration of 50 μg/chamber of a basement membrane pre-coated the Transwell upper chambers in the form of Matrigel (BD Biosciences, USA). Putting RPMI-1640 with 10% FBS into the bottom chamber, before CRC cells (3 × 104/chamber) were seeded in serum-containing media in the upper well of the Transwell chambers. The cells were harvested for 48 h, and then removed those that not yet invade through the pores with the cotton swab. The invasive cells in chamber under the filter surface were immobilized in 70% ethanol, dyed with 0.1 mg/ml crystal violet solution, and counted below a microscope (× 20 magnification). Specific experiments had inserts in triplicate, and five selected fields were counted per insert at random.
Flow cytometry (FCM) assay
Cell apoptosis was detected by the Annexin V-FITC apoptosis detection kit on the basis of the manufacturer instructions. In brief, a total of 3 × 105 CRC cells and various concentrations (10, 25, 50,100 μmol/L) of emodin were added to the wells after 24 h of incubation at 37 °C. The control group was treated with the equivalent quantity of DMSO (0.2%). Cells of each sample were harvested after an additional 12 or 24 h and suspended in 500 μl of Annexin V binding buffer (1X). Annexin V-FiTC (5 μl) and 5 μl of propidium iodide (PI) were added and incubated for 15 min in dark. The stained cells were evaluated by flow cytometry using a FACS Calibur (BD Biosciences, San Jose, CA, USA).
After collection of the tumor, 5 μm thin sections of colorectal cancer were immobilized in 4% Paraformaldehyde (PFA) and then embeded in paraffin. Sections were placed on poly L-lysine coated glass slides, immunostained with Envision Flex mini kit (Dako) and in a graded alcohol series to dehydrate and then embed the paraffin. The Anti-Ki67 antibody (Abcam, USA, ab15580) was used to the sections (1:250), and then harvested overnight. The tumor sections were cleaned, which applied (1:500) an anti-Rabbit secondary antibody (Abcam, ab98488).
GraphPad Prism 6.0 Software (GraphPad Inc., La Jolla, CA, USA) was used analysis statistical. All results were repeated for three times. One-way analysis of variance (ANOVA) tested differences of expression among different groups. A figure of P < 0.05 was supposed to statistically significant.
Distribution of top 1000 differentially methylated imprinted CpG sites
According to CpG island neighborhood (shores, shelves, islands, and opensea), the distribution of top 1000 CpG sites which were differentially methylated demonstrated that the CpG regions based on island of the notably hyper- or hypo- methylated CpG sites were scattered separately.
Significant difference of CpG sites between colon cancer and its adjacent tissues
HOXD10 hypermethylation was related to lower mRNA expression in CRC
HOXD10 gene was hyper-enriched by CpGs and showed significant difference between normal and tumor group
Eight CpG sites for HOXD10 in DMR_540
Methylation of HOXD10 is associated with gene silencing
HOXD10 inhibited CRC cell proliferation, cell function and induced apoptosis
The expression of HOXD10 displayed contrary tendency with malignant neoplasms gene RHOC which related to AKT/ERK pathway activities
In this research, we had uncovered that HOXD10 played a vital role in the progression of the colorectal cancer. The evidence has been provided HOXD10 showed hypermethylation and decreased mRNA expression in CRC. Also, we focused on analyzing the pro-metastatic gene RHOC and the phosphorylation activation of the AKT and MAPK signaling pathways altered after HOXD10 demethylation. Exploring the role of anti-oncogenes HOXD10 in human colorectal would lead to a deeper understanding of the underlying mechanism of CRC, as well as stimulate the development of new treatment strategies.
DNA methylation could cause transcriptional silencing of anti-oncogene, which was a common epigenetic modification. Existing research indicated that the methylation of cancer suppressor genes was in close relationship with cancer development . Growing evidence showed clearly that the CpG island hypermethylation in oncogenes played a fundamental role in the process of colon cancer development [27, 28]. Previous studies have showed that HOXD10 which was aberrantly hypermethylated in papillary thyroid cancer may act as a tumor suppressor . HOXD10 was upregulated in 82 (89.13%) normal gastric mucosa samples and in 242 of 436 (55.5%) cases of gastric cancer . Lin et al. showed that of promoter hypermethylation HOXD10 combined with ZIC1 and RUNX3 might be a potential early detection of gastric cancer and precancerous lesions . Daugaard et al. demonstrated that homeobox gene family including HOXD10, HOXD3, HOXB3/HOXB4 have potential as biomarkers in lung adenocarcinoma . Thus, HOXD10 was aberrantly hypermethylated in various carcinoma, which showed the potential of it as biomarker or therapeutic target. Similarly, our study illustrated the HOXD10 hypermethylation and low mRNA expression in colorectal carcinoma tissues and cells.
5-aza-2′-deoxycytidine (5-Aza-dC) can reactivate epigenetically silenced genes, which was able to inhibit DNA methylation . Besides, 5-Aza-dC is the clinically approved DNA methyltransferase inhibitors (DNMTi) . It is reported that 5-Aza-dC was used in colorectal carcinoma cell lines for APC gene, which methylation promoted the migration and proliferation abilities . Moreover, after the treatment with a DNA methyltransferase inhibitor 5-Aza-2′-deoxycytidine, the methylation degree of HOXD10, which was correlated with decreased transcript expression, was restored in oral cancer cell lines . Our study has illuminated that HOXD10 might function as a vital effector that negatively regulated the metastasis and development in oncogenesis of the colon after the treatment of 5-Aza-dC.
Fassan M et al. has proved that a weak cytoplasmic HOXD10 IHC cytoplasmic positivity with membranous reinforcement can be seen in normal samples while 80% breast cancer tissue samples indicated moderate to strong positivity cytoplasmic HOXD10 IHC positivity . The result of IHC indicated that positive rate of HOXD10 protein expression dropped gradually with the increment of histological grade in CRC patients, which was closely connected with a higher expression of pro-metastatic gene RHOC, a gene related to the AKT and MAPK pathway. The same result can be seen in human cholangiocellular carcinoma, the HOXD10 expression was pretty high in well-differentiated cancerous tissues (25/28, 89.3%), while the HOXD10 was lowly expressed in well-differentiated cancerous tissues (25/28, 89.3%) . It has been demonstrated that abnormally activated AKT and MAPK pathway is closely associated with growth and metastasis of CRC [37, 38, 39, 40]. Previous research suggested that inactivated MAPK/ERK and PI3K/AKT signaling pathways resulted from decreasing the phosphorylation of ERK1/2 and AKT, which were classical signal transduction pathways and played a significant role in colorectal carcinoma progression .
In conclusion, the high methylation and low relative mRNA expression of HOXD10 in CRC tissue microarray data, cell lines, and tumor tissue samples, combined with the loss of function associated with the AKT/MAPK signaling pathway for the role as anti-oncogenes in the pathogenesis of CRC. Follow-up studies are needed to completely evaluate HOXD10 in colorectal carcinoma, thus make detailed risk analysis and prognosis easier. Our findings gave an in-depth understanding of HOXD10 methylation involved in the development of colorectal cancer cell metastasis, which may serve as a possible target for therapy in times to come.
This work received Grant (2013) 163 from the Key Laboratory of Malignant Tumor Molecular Mechanism and Translational Medicine of Guangzhou Bureau of Science and Information Technology, and Grant No. KLB09001 from the Key Laboratory of Malignant Tumor Gene Regulation and Target Therapy of Guangdong Higher Education Institutes also supported the present study.
Availability of data and materials
The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Substantial contribution to the conception and design of the work: YY, HW and WZ; Analysis and interpretation of the data: WL, FY and YL; Drafting the manuscript: YY, ZY and HW; Revising the work critically for important intellectual content: YL and JC; Collecting of grants: YL. All authors read and approval the final manuscript.
Ethics approval and consent to participate
All procedures performed in studies involving human participants were in accordance with the ethical standards of Sun Yat-sen Memorial Hospital. Written informed consents were obtained from all individual participants included in the study.
Consent for publication
The authors consent for publication.
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.