Background
According to GLOBOCAN 2020, esophageal cancer (EC) was reported to be accountable for 604,100 (3.1%) new cases and 544,076 (5.5%) new deaths in 185 countries with about 70% of cases occur in males [
1]. Esophageal squamous cell carcinoma (ESCC), the most common histological subtype of EC, is commonly seen in high-incidence areas such as in Central and Southeast Asia [
2]. The major risk factors for ESCC are smoking and alcohol intake which exert synergistic effects on carcinogenesis [
3]. The clinical management of ESCC is still challenging, and there is currently a lack of approved targeted therapy drugs [
4].
Histone deacetylases (HDACs) are capable of inducing the catalyzation of the removal of acetyl groups from the acetyl-lysine residues in histone and non-histone proteins [
5]. HDACs are of crucial functionality in various cancers due to their modulation in the function of proteins engaged in cancer initiation and progression [
6]. Selective inhibitors for HDACs have been pinpointed as epigenetic drug targets for the treatment of numerous malignancies via diverse molecular mechanisms [
7‐
9]. As a member of the HDACs family, histone deacetylase 3 (HDAC3) has been often activated in several cancers and therefore, its selective inhibitors constitute a preventive strategy against these cancers [
10]. HDAC3 has been documented to be highly expressed in ESCC cells, playing a promoting role in tumor progression [
11]. As previously demonstrated, HDAC3 was able to limit miR-17-92 expression during lung sacculation [
12], highly suggestive of the potential of HDAC3 in modulating microRNAs (miRNAs). miRNAs have been found to exert a tumor suppressive or oncogenic role in EC by mediating their downstream specific target genes [
13,
14]. More importantly, overexpression of miR-494 has been elucidated to exert inhibitory effects on malignant features of ESCC cells [
15].
Moreover, TGIF1 has been reported to be involved in the onset of ESCC [
16]. Notably, TGIF1 has been found to interact with Smad2-Smad4 complex to further block the activation of the TGFβ signaling pathway, which share close correlation with poor prognosis of EC patients and epithelial-mesenchymal transition in ESCC [
17‐
19]. TGFβ signaling pathway is often deregulated in several cancers and possess tumor-suppressor functions in normal cells and early-stage cancer cells [
20].
Summarizing the aforementioned available reports, we proposed a hypothesis that aberrantly expressed HDAC3 in EC may participate in the progression and development of EC through regulation of the miR-484/TGIF1/TGFβ signaling. Herein, this study was conducted with purpose of exploring the effect of HDAC3 on EC cellular capacities of proliferation, invasion, migration, and apoptosis by orchestrating miR-484, TGIF1, and the TGFβ signaling pathway.
Methods
Ethical statement
This study was approved by the ethics committee of the First Affiliated Hospital of Zhengzhou University with the written informed consent filled by the participating patients. The animal experiment was implemented by referring to the guidelines for the care and use of laboratory animals issued by the National Institutes of Health.
Microarray dataset GSE16456 containing ESCC-related miRNAs was retrieved from the Gene Expression Omnibus (GEO) database, including 6 normal samples and 6 ESCC samples. Differential analysis was then conducted utilizing “limma” package of R language with |log FoldChange|> 1 and p value < 0.05 as the threshold. The downstream regulatory miRNAs of HDAC3 were predicted with the RNAInter database. Furthermore, downregulated genes from the GSE16456 dataset and candidate miRNAs were intersected using jvenn. StarBase, mirDIP and miRanda databases were searched for prediction of the downstream target genes of the miRNAs, followed by intersection using jvenn. The expression pattern of genes in ESCC and normal samples available from TCGA was retrieved from the UALCAN database. Kyoto encyclopedia of genes and genomes (KEGG) database was used for analysis of the gene-related signaling pathways.
Study subjects
ESCC tissues and corresponding adjacent tissues were harvested from 79 patients with ESCC (43 males, 36 females, aged 45–72 years with a mean age of 58.11 ± 8.64) who underwent thoracic surgery at the First Affiliated Hospital of Zhengzhou University from January 2017 to January 2018. Patients were included if (1) their ESCC tissues were resected from surgery at the First Affiliated Hospital of Zhengzhou University; (2) they received no radiotherapy, chemotherapy or immunotherapy before surgery; and (3) their baseline characteristics were complete. Patients were excluded if (1) ESCC was not confirmed by pathobiology; or (2) they had undergone radiotherapy or chemotherapy before surgery. Follow-up started after surgery and lasted for 36 months. The correlation between HDAC3 expression and OS of patients was analyzed with the help of the Kaplan–Meier method.
Immunohistochemistry
Paraffin-embedded tissue sections were dewaxed, dehydrated and placed in 3% methanol-H
2O
2 for 20 min, followed by antigen retrieval. Then, sections were blocked with normal goat serum (C-0005, Shanghai Haoranbio Co., Ltd., Shanghai, China) at ambient temperature for 20 min and reacted with primary antibodies (all from Abcam Inc., Cambridge, MA): rabbit-anti HDAC3 (ab7030, 1: 500,), mouse-anti SMAD2 (ab119907, 1: 200), rabbit-anti SMAD4 (ab40759, 1: 500), rabbit-anti TGF-βRII (ab186838, 1: 200) and rabbit-anti Ki67 (ab15580, 1: 200) at 4 ºC overnight as well as with secondary antibody (Abcam), namely goat anti-rabbit immunoglobulin G (IgG) (ab6721, 1: 2000) or goat anti-mouse IgG (ab150113, 1: 2000) at 37 ºC for 20 min. Subsequently, horseradish peroxidase (HRP)-conjugated streptavidin/peroxidase working solution (0343-10000U, Imunbio Co., Ltd., Beijing, China) was introduced for 20-min of incubation with sections at 37 ºC. The results were visualized using diaminobezidin (ST033, Whiga Co., Ltd., Guangzhou, Guangdong, China). The sections were then counterstained with hematoxylin (PT001, Bogoo Biotech Co., Ltd., Shanghai, China) for 1 min and immersed in 1% ammonia to return blue in color. The mounted sections were microscopically observed with 5 high-power fields selected on a random basis (100 cells each field). The percentage of positive cells < 10% was considered negative and > 50% was considered strongly positive, while 10% ≤ the percentage of positive cells < 50% was considered positive [
21].
Reverse transcription quantitative polymerase chain reaction (RT-qPCR)
Total RNA was extracted utilizing TRIzol reagents (15596026, Invitrogen, Carlsbad, CA, USA), and 1 µg of which was used for RT. For miRNA detection, total RNA was reverse-transcribed into complementary DNA (cDNA) employing miRNA First Strand cDNA Synthesis (Tailing Reaction) kit (B532451-0020, Sangon Biotechnology Co. Ltd., Shanghai, China). For mRNA, RT was started with reference to the manual of cDNA RT kit (K1622, Reanta Co., Ltd., Beijing, China). RT-qPCR was operated on the synthesized cDNA using Fast SYBR Green PCR kit (Applied Biosystems, Foster City, CA, USA) in ABI 7500 RT-qPCR system (Applied Biosystems). Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) or U6 functioned as normalizer and the relative expression of genes was analyzed by the 2
−ΔΔCt method. Primer sequences are displayed in Additional file
5: Table S1.
Western blot analysis
Total protein was extracted followed by concentration measurement utilizing a bicinchoninic acid kit. Protein (50 μg) was separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis and transferred onto a polyvinylidene fluoride membrane which was then blocked with 5% skimmed milk powder at ambient temperature for 1 h. Next, the membrane was probed with diluted primary rabbit antibodies to Smad2 (ab33875, 1: 1000, Abcam), Smad4 (ab40759, 1: 5000, Abcam), HDAC3 (ab7030, 1: 500, Abcam), TGIF1 (ab52955, 1: 5000, Abcam), cleaved caspase-3 (ab32042, 1: 1000, Abcam), total caspase-3 (ab32150, 1: 1000, Abcam), MMP-2 (ab97779, 1: 2000, Abcam), MMP-9 (ab38898, 1: 1000, Abcam), TGF-βRII (sc-17791, 1: 1000, Santa Cruz Biotechnology, Santa Cruz, CA) and GAPDH (ab9485, 1: 2500, Abcam) overnight at 4ºC. The next day, the membrane was re-probed with HRP-labeled secondary antibody of goat anti-rabbit antibody to IgG H&L (ab97051, 1: 2000, Abcam) for 1 h. The results were visualized utilizing enhanced chemiluminescence reagents (BB-3501, Ameshame, Little Chalfont, UK). Images were acquired through Bio-Rad image analysis system (Bio-Rad Laboratories, Hercules, CA), followed by analysis using Quantity One v4.6.2 software. The relative protein level was described as the ratio of gray value of protein to be tested to that of GAPDH [
22,
23].
Cell treatment
A normal human esophageal epithelial cell (HEEC) line (Tongpai Biotechnology, Shanghai, China) and 4 ESCC cell lines [EC9706 (Fenghui Biotechnology, Changsha, China), Eca109 (Fenghui Biotechnology), TE-1 (Procell, Wuhan, China) and KYSE-150 (Procell)] were cultured in RPMI-1640 medium appended to 10% fetal bovine serum (FBS) and penicillin–streptomycin solution (1: 1) (final concentration of 100 U/mL) at 37ºC with 5% CO2. Cells were detached with 0.25% trypsin, passaged (1: 3) and seeded into a 6-well plate (3 × 105 cells/well). Upon achieving 70–80% confluence, cells were collected and cell with the relatively high HDAC3 expression was selected for further experiments.
Logarithmically growing cells with the relatively high HDAC3 expression was passaged, seeded into a 6-well plate (1 × 105 cells/well), and transfected with reference to the manuals of Lipofectamine® 2000 reagent (11668019, Invitrogen) utilizing shRNA against HDAC3 (sh-HDAC3), miR-494 mimic, miR-494 inhibitor, overexpression (oe)-TGIF1, and their corresponding NCs as well as treated with SRI-011381 (an activator of the TGFβ signaling pathway), and dimethyl sulfoxide (DMSO) (negative control [NC] for SRI-011381).
The sequences were provided by Shanghai Sangon. The miRNA mimic transfection concentration was 50 nM, the miRNA inhibitor transfection concentration was 100 nM, and the plasmid transfection amount was 1 μg. For SRI-011381 (HY-100347A, MCE, Sovizzo Vicenza, Italia) treatment, cells after transfection were further treated with 10 μM SRI-011381 for 12 h [
24].
5-ethynyl-2'-deoxyuridine (EdU) assay
EdU detection kit (Guangzhou RiboBio Co., Ltd., Guangzhou, China) was adopted for testing cell proliferation [
25]. Three visual fields were selected on a random basis under a fluorescence microscope (FM-600, Shanghai Pudan Optical Instrument Co., Ltd., Shanghai, China) to count the number of positive cells in each field.
Flow cytometry
After transfection for 48 h, cells were detached with 0.25% ethylenediaminetetraacetic acid-free trypsin (YB15050057, Yubo Biological Technology Co., Ltd., Shanghai, China) and centrifuged with the supernatant removed. Annexin-V-fluorescein isothiocyanate (FITC)/propidium iodide (PI) dye liquor was formulated using Annexin-V-FITC, PI and 4-(2-hydroxyerhyl)piperazine-1-erhanesulfonic acid (HEPES) at a ratio of 1: 2: 50 with reference to the manual of Annexin-V-FITC cell detection kits (K201-100, Biovision, Milpitas, CA). Each 100 µL dye liquor was adopted to re-suspend 1 × 10
6 cells and incubated at ambient temperature for 15 min, followed by addition of 1 mL HEPES buffer solution (PB180325, Procell). Finally, cell apoptosis was assessed by detecting FITC and PI fluorescence utilizing a flow cytometer (FACS Calibur, BD Biosciences, Franklin Lakes, NJ, USA) [
26,
27].
Transwell assay
The cell migratory and invasive (added with extracellular matrix gel) capabilities were examined using a Transwell assay, as previously described [
28]. Five randomly selected fields in each chamber were observed, photographed, and counted under an inverted microscope.
Chromatin immunoprecipitation (ChIP) assay
Fresh ESCC and adjacent tissues were cut into 1–3 mm
3 small pieces and transferred to a 50 mL test tube. Next, the tissues were fixed with formaldehyde at ambient temperature for 10 min for generating DNA–protein crosslink. The reaction was terminated by 2.5 M Glycine (final concentration of 0.125 M). The samples were lysed using a homogenizer, sub-packed (1 mL per tube), re-suspended, and fragmented by ultrasonic wave, 10 s each at an interval of 10 s, 15 cycles in total. The supernatant was harvested through centrifugation at 13,000 rpm and 4ºC for 10 min and sub-packed into 3 tubes. The tubes were respectively added with normal mouse antibody to IgG (ab109489, 1: 100, Abcam) as NC and mouse antibody to HDAC3 (ab7030, 1: 100, Abcam) for incubation at 4ºC overnight. The endogenous DNA–protein complex was precipitated by Protein Agarose/Sepharose. The non-specific complex was washed and the DNA–protein complex was incubated at 65ºC overnight for de-crosslinking. DNA fragment was purified and retrieved by phenol/chloroform. The primer was designed by selecting 2000 bp from the transcription start site to the upstream as the promoter sequence. The enrichment of HDAC3 in the miR-494 promoter region was tested by RT-qPCR [
29].
Dual luciferase reporter gene assay
The HEK-293T cell line (American Type culture collection, Manassas, VA) was cultured in DMEM. When achieving 80–90% confluence, cells were detached by 0.25% trypsin, passaged and cultured at 37 ºC with 5% CO
2. Logarithmically growing cells were used for further experiments. The artificially synthesized TGIF1 3’untranslated region (3’UTR) was introduced into pGL3-control (Promega Corp., Madison, WI) through endonuclease sites XhoI and BamH I. The mutation site of complementary sequence of seed sequence was designed on TGIF1-wild-type (WT). After restriction enzyme digestion, the target fragment was inserted into pGL3-control vector using T4 DNA ligase. The correctly sequenced luciferase reporter plasmids WT and mutant type (MUT) were co-transfected with miR-494 mimic/mimic-NC into HEK-293T cells, respectively, for 48 h. The luciferase activity was tested utilizing Dual-Luciferase Reporter Assay System Kit (Promega) on a TD-20/20 luminometer (E5311, Promega) [
30].
Xenograft tumor in nude mice
In total, 36 thymic male nude mice aged 5 to 6 weeks (Shanghai SLAC Laboratory Animal Co., Ltd., Shanghai, China) were housed at 25–27 ºC under constant humidity (60–65%) under 12-h light/dark cycles (drink and eat freely). The animal experiment started after one week of acclimation. Under 80–90% confluence, EC9706 and Eca109 cells transfected with sh-NC + oe-NC, sh-HDAC3 + oe-NC or sh-HDAC3 + oe-TGIF1 were collected, detached, centrifuged, re-suspended and counted. The cell concentration was adjusted to 1 × 107 cells/mL. Then, 20 µL cell suspension was subcutaneously inoculated into the armpit of nude mice (n = 6). The tumor formation was observed every week. Tumor volume (V) was calculated: V = (a longest diameter × b shortest diameter 2)/2, followed by construction of a tumor growth curve. All nude mice were euthanized utilizing CO2 after 5 weeks.
Hematoxylin and eosin (HE) staining
HE staining kits (C0105, Beyotime, Shanghai, China) were used for the staining. In brief, the subcutaneously transplanted tumor tissues were fixed with 10% neutral buffered formaldehyde (G2161, Solarbio, Beijing, China) at 4 ºC for 24 h, The tissues were then dehydrated, paraffin-embedded and sectioned, conventionally dewaxed with xylene and hydrated with gradient alcohol. Hematoxylin was used to stain the sections for 5–10 min, followed by eosin staining solution for 30 s-2 min. Finally, the sections were dehydrated with gradient alcohol, cleared with xylene, sealed with neutral gum, photographed and observed under an inverted microscope (Olympus, IX73) [
31].
Statistical analysis
Statistical analyses were started with the help of SPSS 21.0 (IBM Corp., Armonk, NY, USA). All experimental data were tested for normal distribution and homogeneity of variance. Measurement data were expressed as mean ± standard deviation. Differences between ESCC and adjacent tissues were compared by paired t test while data comparison of unpaired design between other two groups was conducted by independent sample t test. Differences among multiple groups were compared by one-way analysis of variance (ANOVA) in combination with Tukey’s post hoc test. Data at various time points were compared by repeated measures ANOVA with post-hoc Bonferroni-corrected comparisons. The survival rate of patients was calculated by the Kaplan–Meier method. A value of p < 0.05 was summarized as statistically significant.
Discussion
More patients with EC can only be diagnosed at an advanced stage yet the potency of traditional radiotherapy or chemotherapy is limited; therefore, a better understanding of underlying molecular mechanism may provide future direction of developing novel therapeutic strategies [
32]. During the present study, we attempted to unravel the involvement of HDAC3 in the context of ESCC. Collectively, the experimental data in our study offered evidence that shRNA-mediated silencing of HDAC3 harbored the tumor suppressive property to curb proliferative, migratory and invasive capabilities of ESCC cells and to induce apoptotic process through activation of the TGFβ signaling pathway via target-inhibition of TGIF1 by miR-494.
HDAC3 has been highlighted to be a crucial validated target for cancer due to its overexpression in a variety of cancers [
10]. Our study underlined the identification of upregulated HDAC3 expression in ESCC tissues and cells. In line with our findings, esophageal tumor tissues showed upregulation of HDAC3 as compared to normal adjacent tissues, and conversely, its inhibition contributes to the arrested malignant properties of EC cells [
11]. When HDAC3 was silenced in our study, malignant behaviors of ESCC cells were restrained yet apoptosis was accelerated accompanied by upregulated cleaved caspase-3/total caspase-3 and downregulated MMP-2 and MMP-9. Activation of caspase-3 has been used as a hallmark of induced apoptosis in EC cells [
33]. Highly expressed MMP-9 has been found in EC tissues, and is associated with poor prognosis, tumor size and clinical staging [
34,
35]. In the context of ESCC, a high positive rate of MMP-2 has been detected in relation to tumor differentiation degree, invasion depth and LNM so that occurrence of distant metastasis is facilitated [
36]. Notably, HDAC3 has been shown to augment expression of MMP-2 in U87MG cells [
37] but suppress caspase-3 expression in K562 cells [
38]. Therefore, strategies aimed at HDAC3 silencing could prove beneficial for the treatment of ESCC.
An inverse correlation has been confirmed between HDAC3 expression and miR-494 expression in the context of acute ischemic stroke [
39]. Consistently, the current results revealed that HDAC3 played an inhibitory role in miR-494 expression by enriching in its promoter. Moreover, miR-494, in the present study was found to be underexpressed while TGIF1 was overexpressed in ESCC tissues and cells. In accordance to this finding, downregulation of miR-494 has been identified in EC by multiple functional reports [
15,
40,
41]. However, overexpression of miR-494 harbors the potential to curb the proliferative and invasive capabilities of EC cells and to trigger apoptosis by targeting CLPTM1L [
15]. Meanwhile, in the current study, the anti-oncogenic property of miR-494 could be abolished by overexpression of its target gene, TGIF1. miRNAs can interact with the 3’UTR of specific target mRNAs and ultimately cause the repression of their expression [
42]. In this study, we also confirmed that miR-494 bound to the 3’UTR of TGIF1 mRNA and could negatively regulate its expression. TGIF1 is able to encourage non-small cell lung cancer cells to grow and migrate while TGIF1 knockdown ameliorates tumorigenic characteristics [
43]. Additionally, the tumor-promoting action of TGIF1 has been discussed in papillary thyroid cancer where downregulation of TGIF1 limits the progression of malignant cells in vitro and suppresses aggressive phenotypes in vivo [
44]. However, the expression and function of TGIF1 in ESCC have been not fully elucidated. Additionally, owing to the scarcity of literature on the relationship between HDAC3 and TGIF1, further investigation is still required so as to validate the established promoting effect of HDAC3 on ESCC cell malignant phenotypes and tumor growth through blockade of miR-494-mediated TGIF1 inhibition.
In the present investigation, we also found that activation of the TGFβ signaling pathway could be inhibited by TGIF1, facilitating the malignant phenotypes of ESCC cells. The prominent function of TGIF1 has been indicated as the suppression of the TGFβ signaling pathway in the progression of pancreatic ductal adenocarcinoma [
45]. TGIF1 can bind to DNA through interplay with Smad2 in the TGFβ signaling pathway and subsequently inhibit the activation of downstream genes of the TGFβ signaling pathway [
17]. TGFβ signaling pathway has been frequently deregulated in tumors and exerts crucial roles in tumor initiation, development and metastasis [
46]. More importantly, blockade of the TGFβ signaling pathway has been elaborated to accelerate the progression of EC [
47], supporting the validation of our results. In addition, TGFβ1 has been found to be a major factor inducing the upregulation of miR-494 in myeloid-derived suppressor cells [
48]. A previous work also indicated that HDAC3 deficiency contributes to the promotion of TGFβ signaling pathway activation in PM2.5-induced mice with lung injury [
49]. Based on the aforementioned information, we are convinced that the use of a TGFβ signaling pathway activator may lead to a new therapeutic strategy in patients with ESCC (Additional file
6).
Open AccessThis article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/4.0/. 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 in a credit line to the data.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.