Background
Colorectal cancer (CRC) is the third most common cancer and one of the leading causes of cancer-related death worldwide [
1]. What is worse, metastatic cases account for approximately 40% to 50% of newly diagnosed patients, whose overall survival (OS) reaches only up to 30 months [
2]. Despite the fact that personalized medicine dramatically improves the outcome of cancer treatment, the prognosis of metastatic CRC remains poor. There is no doubt that a better molecular understanding of CRC will provide clinicians with more choices to achieve better outcomes for patients. However, our understanding of the underlying molecular mechanisms through which cancer migrates and metastasizes is not sufficient to address this problem. Some important biological behaviors have been discovered to partly illuminate this process.
With the rapid expansion of cancer cells, the central regions of most solid tumors are subjected to hypoxia because of inadequate vascularization, and the presence of a hypoxic microenvironment and the activation of downstream effectors are common features [
3]. It is generally considered that hypoxia causes resistance to chemotherapy and initiates a number of events, such as epithelial-mesenchymal transition (EMT), inflammatory cell infiltration, and autophagy [
4,
5]. The formation of the so-called “pro-metastatic niche” facilitates the progression of CRC to a large extent.
Silencing information regulator 1 (SIRT1) is a highly evolutionarily conserved histone and non-histone deacetylase. Altered SIRT1 expression is observed during ischemia/reperfusion, and SIRT1 exerts protective roles in the heart, liver and brain tissues [
6,
7]. However, its dysregulation is also frequently observed in many tumor types, and it is thought to play multifaceted roles (promotor or inhibitor) in tumor progression. In ovarian cancer, hypoxia induces the transcriptional repression of SIRT1, which contributes to EMT and enhances cancer metastasis [
8]. Conversely, under hypoxic conditions, SIRT1 is a direct downstream target of hypoxia-inducible factor 1α (HIF-1α), which orchestrates with NF-κB pathway to promote cancer stem cell-like properties in ovarian cancer cells [
9]. From this perspective, alterations in SIRT1 expression under hypoxia and its roles in the hypoxic microenvironment need further research. This information may shed light on the mechanisms of CRC migration and invasion.
Early growth response factor 1 (EGR1) is a zinc-finger transcription factor that binds to canonical GC-rich motifs. EGR1 is induced by a variety of stimuli, including hypoxia, ionizing radiation, hyperglycemia and chemotherapy drugs, and it promotes or inhibits tumor proliferation [
10]. EGR1 expression varies under hypoxic conditions in different cellular contexts and regulates diverse biological processes. In addition, there is evidence that EGR1 binds directly to the SIRT1 promoter and interacts with SIRT1 to regulate its function [
11]. Hence, we wondered whether EGR1 regulates SIRT1 expression under hypoxic conditions and whether the hypoxia/EGR1/SIRT1 axis contributes to the progression of CRC.
In this paper, we report that EGR1 binds directly to the GC-rich box 100 bp upstream of the SIRT1 promoter and mediates hypoxia-induced SIRT1 transcriptional suppression in CRC cells. In addition, the hypoxia/EGR1/SIRT1 axis contributes to the migration and metastasis of CRC cells. Moreover, decreased SIRT1 activity promotes the acetylation of NF-κB and the activation of matrix metalloproteinase 2/9 (MMP-2/-9) and boosts the migration and invasion of CRC cells. As a consequence, the restoration of SIRT1 may suppress CRC progression and might be a promising target for hindering the development of CRC.
Methods and materials
Cell culture
HEK-293T, HCT116 and SW480 cell lines were purchased from the American Tissue Culture Collection (ATCC, Manassas, VA, USA). The HEK-293T cell line was cultured in DMEM with 10% FBS, the HCT116 cell line was cultured in McCoy’s 5A medium with 10% FBS, and the SW480 cell line was cultured in Leibovitz’s L-15 medium with 10% FBS. All cell lines were cultured at 37 °C in a humidified atmosphere containing 5% CO2. Hypoxia (1% O2) was maintained by pumping a mixture of gases (95% N2, 5% CO2) into a Billups-Rothenberg chamber (Del Mar, CA, USA) at 37 °C and monitored by a gas flowmeter.
Lentivirus infection and plasmid transfection
Lentiviruses were used to generate cells with SIRT1 overexpression or downregulation. Based on the protein expression levels of SIRT1 in several CRC cell lines (Additional file
1: Figure S1A), HCT116 cells were stably transfected with an empty vector (Con077) or SIRT1-shRNA (GeneChem, Shanghai, China). SW480 cells were stably transfected with an empty vector (Con195) or lenti-SIRT1 (GeneChem, Shanghai, China). The transfected cells were then screened with puromycin (4 μg/mL) for several passages. SIRT1 promoter luciferase reporter plasmids (truncated SIRT1 promoter sequences and mutated 100 bp promoter sequences inserted into the GV354 vector, GeneChem, Shanghai, China) as well as EGR1, Sp1 and USF2 overexpression plasmids (constructed using GV230 and GV141 vectors, GeneChem, Shanghai, China) were transiently transfected into HEK-293T cells with Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA).
RNA extraction and real-time PCR
Total RNA extraction was performed with RNAiso plus (TaKaRa, Shiga, Japan) according to the manufacturer’s instructions. Reverse transcription was then performed with PrimeScript™ RT Master Mix (TaKaRa, Shiga, Japan). Real-time PCR was performed on an Agilent Technologies system (Santa Clara, CA, USA) using TB Green Premix Ex Taq II (TaKaRa, Shiga, Japan) and the following primers:
SIRT1 forward | 5′-GCAGATTAGTAGGCGGCTTG-3′ |
SIRT1 reverse | 5′-TCATCCTCCATGGGTTCTTC-3′ |
β-Actin forward | 5′-GGACTTCGAGCAAGAGATGG-3′ |
β-Actin reverse | 5′-AGCACTGTGTTGGCGTACAG-3′ |
MMP-2 forward | 5′-CTCTCCTGACATTGACCTTGGCAC-3′ |
MMP-2 reverse | 5′-GTATTCATTCCCTGCAAAGAACAC-3′ |
MMP-9 forward | 5′-AATCTCACCGACAGGCAGCT-3′ |
MMP-9 reverse | 5′-CCAAACTGGATGACGATGTC-3′ |
TIMP-2 forward | 5′-GTGGACTCTGGAAACGACAT-3′ |
TIMP-2 reverse | 5′-CCAGGAAGGGATGTCAGAGC-3′ |
TIMP-3 forward | 5′-TACCGAGGCTTCACCAAGATG-3′ |
TIMP-3 reverse | 5′-CAGGACTTGATCTTGCAGTTAC-3′ |
Three duplicate wells were used for each sample, and each group was analyzed three times. Relative quantitative values were then calculated by the 2−ΔΔCt method.
Western blot analysis
Cell extracts were prepared by lysis in 1× SDS loading buffer; then, the samples were boiled and subjected to Western blot analysis. Proteins (30 μg) were transferred from gels onto PVDF membranes (Millipore, Billerica, MA, USA) and blocked in 5% skimmed milk for 2 h. The following antibodies were used in the experiment to detect proteins: rabbit anti-SIRT1 (1:1000, Abcam, Cambridge, UK), EGR1 (1:1000, Abcam), USF2 (1:1000, Abcam), NF-κB p65 (acetyl K310) (1:1000, Abcam), TIMP-2 (1:1000, Abcam), MMP-2 (1:500, ProteinTech, Wuhan, China), MMP-9 (1:500, ProteinTech), mouse anti-β-actin (1:2000, Cell Signaling Technologies, Beverly, MA, USA), and NF-κB p65 (1:100, Santa Cruz, Dallas, Texas, USA). The antibodies were incubated with the membranes at room temperature for 1 h and then at 4 °C overnight. HRP-linked anti-rabbit and anti-mouse secondary antibodies were incubated with the membranes at 4 °C for 1 h. All blotting was detected by the enhanced chemiluminescence (ECL) detection method.
Dual-luciferase reporter assay
SIRT1 promoter region sequences were segmented from distal to proximal and fused with firefly and Renilla luciferase cDNA in a GV354 vector (GeneChem, Shanghai, China). All constructs were verified by sequencing. HEK-293T cells were seeded in 24 wells at a density of 5 × 104 cells per well 24 h before transfection. These constructs were then transfected into cells and maintained at 37 °C for 12 h. Dual-luciferase reporter assays were performed in accordance with the manufacturer’s instructions (Promega, Madison, WI, USA), and firefly and Renilla luciferase activities were assessed using a dual-luciferase reporter assay system (Promega, Madison, WI, USA). Three duplicate wells were used for each sample, and each sample was analyzed three times. The ratio of firefly luciferase activity to Renilla luciferase activity was used to indicate the relative luciferase activity.
Chromatin immunoprecipitation
Chromatin immunoprecipitation (ChIP) assays were performed using a SimpleChIP Plus Kit (Cell Signaling Technologies) according to the manufacturer’s instructions. HEK-293T cells were seeded in 10 cm dishes at a density of 2 × 106 prior to the experiment. Cells were cross-linked with 1% formaldehyde, and the reaction was terminated with glycine. Chromatins were then digested by micrococcal nuclease. Antibodies against IgG (negative control, anti-rabbit monoclonal IgG, Cell Signaling Technologies) and EGR1 (Cell Signaling Technologies) and Protein G magnetic beads were used to immunoprecipitate the protein–nucleic acid complexes. Chromatin elution and purification were performed after that. The DNA sequence of the SIRT1 promoter bound to EGR1 was amplified by PCR, and the following primers were used: 5′-ACCCGTAGTGTTGTGGTCTG-3′ (forward), 5′-GCCATCTTCCAACTGCCTCT-3′ (reverse). The PCR results were analyzed by Southern blotting.
Transwell assay
Transwell assays were employed to estimate the cell migration and metastasis abilities. Matrigel (BD, Franklin Lakes, NJ, USA) was diluted with serum-free medium at a ratio of 1:9 and then added to the bottom of an 8 μm transwell chamber (Corning, Kennebunk, ME, USA) and cultured in an incubator at 37 °C for 2 h. HCT116 and SW480 cells were pretreated with serum-free medium for 24 h and seeded in the upper chambers at a density of 5 × 104. Serum-free medium was added to the upper chambers, and complete medium was added to the bottom chambers. After 24 h of culture under different conditions (normoxia or hypoxia), the cells were fixed with methanol and stained with a 0.1% crystal violet solution. The cells were counted under a microscope after washing with PBS. Four to six high power fields (400×) photos were taken randomly to count the number of cells in each group.
The migration assays were performed without Matrigel. SRT1720 (200 nM, Selleckchem, Houston, TX, USA) and EX527 (2 μM, Selleckchem) were used to pretreat the SW480 and HCT116 cells, respectively, in the experiment.
Statistical analysis
All data are presented as the mean ± standard deviation (SD). Student’s t test was employed to compare two unpaired treatment groups. LDS-t test was employed for multiple comparisons. One-way ANOVA was used to analyze three or more treatment groups. ImageJ (version 1.3.7, NIH, USA) was used to measure densitometry of immunoblotting for each panel. Statistical analyses were performed by SPSS 22.0 software (SPSS, Inc., Chicago, IL, USA) and graphs were created using GraphPad Prism software (version 5.0, San Diego, CA, USA). Results demonstrating p < 0.05 were considered statistically significant.
Discussion
SIRT1 is an extensively explored class III histone deacetylase and has been proven to be a trigger or brake in carcinogenesis depending on the cellular context and tumor type. In recent years, the important connection of SIRT1 to cancer metastasis has been introduced, further confirming the pivotal roles of SIRT1 in cancer progression. For example, in CRC, SIRT1 promotes EMT and metastasis in a Fra-1-dependent manner [
20]. Likewise, SIRT1 is recruited to the promoter of E-cadherin and acts as a co-suppressor with ZEB1 to synergistically inhibit E-cadherin transcription, thus promoting EMT in prostate cancer [
21]. Li et al. also introduced a novel mechanism whereby SIRT1 enhances PGC-1α-mediated mitochondrion biogenesis and increases cell metabolism; as a result, this process facilitates the motility of tumor cells [
22]. However, some studies argue that SIRT1 suppresses cancer cell migration and invasion, such as, by targeting ARHGAP5 in gastric cancer [
23], by inhibiting the TGF-β/Smad4/MMP-7 axis in breast cancer [
24] and by regulating autophagy in CRC [
25]. Though completely different conclusions were reached, these findings strengthen the hypothesis that the SIRT1 promotion or inhibition of cancer migration and invasion is cellular context-dependent.
Recent reports have shed light on the bi-directional interaction between hypoxia and SIRT1. On the one hand, SIRT1 inhibits the transcriptional activity of HIF-1α through the deacetylation of specific lysine sites [
26]. On the other hand, the expression of SIRT1 is induced or suppressed under hypoxic conditions. SIRT1 is reportedly upregulated by HIF-1α and is involved in the promotion of cancer stem cell-like properties in ovarian cancer [
9]. In contrast, SIRT1 is downregulated in non-small cell lung cancer cells exposed to hypoxia, where it modulates the chemotherapeutic resistance to cisplatin and doxorubicin [
27]. Sun et al. reported that in ovarian cancer, hypoxia suppresses SIRT1 transcription via PIASy mediation of the SUMOylation of Sp1, which emphasizes the roles of SIRT1 in mediating hypoxia-induced cancer cell migration and invasion [
8]. Another study from their lab observed a similar phenomenon in lung cancer [
17]. This evidence prompted us to investigate the effects of the hypoxic microenvironment on the development of CRC and the relevant mechanisms.
Our data indicated that hypoxia-induced SIRT1 transcription and expression suppressed and promoted migration and invasion in CRC. Through changing SIRT1 expression and activity, we found that the bioprocess was definitely SIRT1-dependent. To determine the underlying mechanisms, we analyzed the promoter of SIRT1 and constructed a series of truncated promoter plasmids, which hinted at EGR1 as a potent transcription factor regulating SIRT1 transcription. We further verified that EGR1 binds directly to the GC-rich motifs within 100 bp of the upstream promoter of SIRT1 to significantly promote SIRT1 transcription. Under hypoxia, the expression of EGR1 was decreased, and the binding of EGR1 to the promoter of SIRT1 was decreased; as a result, SIRT1 transcription was reduced. A rescue experiment was used to further confirm the roles of EGR1 in regulating SIRT1 transcription.
EGR1 is an early-phase reactive molecule in response to various stimuli that plays a significant role in hypoxia-induced tumor progression, survival and angiogenesis [
10]. Evidences show that EGR1 is a key hypoxia response factor [
28]. In ischemia/reperfusion (I/R) injury, hypoxemic conditions upregulate EGR1, which is thought to be a switch in regulating inflammation, coagulation and vascular hyperpermeability [
29]. In addition, a genomic profiling analysis of MCF-7 suggests that chronic hypoxia exposure increases EGR1 expression [
30]. Nevertheless, after tumor cells underwent hypoxic treatment, we found that EGR1 expression levels in HCT116 cells started to increase to a peak within 3 h, then gradually declined to a lower level compared with normoxic control; whereas in SW480 cells, EGR1 expression levels continued to decline within 24 h. These results may be attributed mainly to different exposure times because almost all studies concerning alterations in EGR1 in hypoxia use acute stress models with exposure times that are relatively short (ranges from 4 to 8 h) and intermittent [
29,
31]. In our study, the tumor cells were treated with a relatively long and sustained hypoxic microenvironment (24 h). Besides, the difference may be partially attributed to different cell lines that possess different responses to hypoxic stress. Hence, after treated by a long and sustained hypoxic exposure, EGR1 acts a potent regulator to influence other bioprocesses.
The NF-κB pathway is a major downstream target of SIRT1 that mediates a series of inflammatory processes in physiological and pathological conditions. Its activation induces the expression of MMPs. Our data indicated that hypoxia increased the acetylation level of the p65 lysine 310 residue, but had a minor influence on its expression. Furthermore, hypoxia promoted MMP-2 and MMP-9 and inhibited their inhibitors TIMP-2 and TIMP-3 in both CRC cell lines. The Western blot results also showed that SIRT1 overexpression eliminated the p65 acetylation induced by hypoxia along with the decreased MMP-2 and MMP-9, suggesting that NF-κB is a direct downstream target that is deacetylated by SIRT1 and might regulate CRC cell migration and invasion.
To summarize, our data emphasize the importance of SIRT1 in CRC migration and invasion under hypoxic conditions, and this regulation is EGR1 dependent. Our data establish for the first time that EGR1 functions as an important upstream regulator of SIRT1 in modulating the bioprocesses of CRC. In addition, we found that NF-κB might be an active participant in the development of CRC under hypoxic conditions. Our study imply that small molecule compounds be developed to specifically target SIRT1 to inhibit CRC migration and invasion and ultimately achieve better outcomes for patients.
Authors’ contributions
Guarantors of integrity of the entire study: ZJ, QQ; study concepts and study design: ZJ, QQ, YST; definition of intellectual content: ZJ, QQ; literature research and cell culture: LS; experimental studies: YST, ZR, YT; data analysis: CZQ; manuscript preparation: YST; manuscript editing: ZR; manuscript review: LS, CZQ. All authors read and approved the final manuscript.