MicroRNA-204/CREB5 axis regulates vasculogenic mimicry in breast cancer cells

Abstract

BACKGROUND:

Vasculogenic mimicry (VM) is characterized by formation of three-dimensional (3D) channels-like structures by tumor cells, supplying the nutrients needed for tumor growth. VM is stimulated by hypoxic tumor microenvironment, and it has been associated with increased metastasis and clinical poor outcome in cancer patients. cAMP responsive element (CRE)-binding protein 5 (CREB5) is a hypoxia-activated transcription factor involved in tumorigenesis. However, CREB5 functions in VM and if its regulated by microRNAs remains unknown in breast cancer.

OBJECTIVE:

We aim to study the functional relationships between VM, CREB5 and microRNA-204-5p (miR-204) in breast cancer cells.

METHODS:

CREB5 expression was evaluated by mining the public databases, and using RT-qPCR and Western blot assays. CREB5 expression was silenced using short-hairpin RNAs in MDA-MB-231 and MCF-7 breast cancer cells. VM formation was analyzed using matrigel-based cultures in hypoxic conditions. MiR-204 expression was restored in cancer cells by transfection of RNA mimics. Luciferase reporter assays were performed to evaluate the binding of miR-204 to 3 ${}^{\prime}$ UTR of CREB5.

RESULTS:

Our data showed that CREB5 mRNA expression was upregulated in a set of breast cancer cell lines and clinical tumors, and it was positively associated with poor prognosis in lymph nodes positive and grade 3 basal breast cancer patients. Silencing of CREB5 impaired the hypoxia-induced formation of 3D channels-like structures representative of the early stages of VM in MDA-MB-231 cells. In contrast, VM formation was not observed in MCF-7 cells. Interestingly, we found that CREB5 expression was negatively regulated by miR-204 mimics in breast cancer cells. Functional analysis confirmed that miR-204 binds to CREB5 3 ${}^{\prime}$ -UTR indicating that it’s an ulterior effector.

CONCLUSIONS:

Our findings suggested that CREB5 could be a potential biomarker of disease progression in basal subtype of breast cancer, and that perturbations of the miR-204/CREB5 axis plays an important role in VM development in breast cancer cells.

Keywords

Breast cancer microRNA-204-5p CREB5 prognosis vasculogenic mimicry

1. Introduction

Vasculogenic mimicry (VM) was first described in malignant melanoma cells and involves the formation of three-dimensional (3D) channel-like networks by tumor cells [1]. These cellular networks allow the supply of blood and oxygen in which tumors can feed themselves, operating independently or simultaneously with classical endothelial vessels [2]. Remarkably, VM has been described in multiple solid tumors such as glioblastoma [3], ovarian [4], prostate [5], lung [6], hepatocellular [7] and breast cancers [8, 9]. Unfortunately, the presence of VM has been associated with aggressive tumor phenotypes, increased metastasis, poor clinical prognosis, and shorter overall survival in cancer patients. Thereby, the protein factors and modulators participating in VM are considered as potential biomarkers and therapeutic targets [10]. The tumor microenvironment is a key factor in the development of VM. Hypoxia is the most typical feature of the inner microenvironment in tumors; it is also the most important factor inducing the VM. Hypoxia through the action of the hypoxia inducible factor (HIF-1 $\alpha$ ) promotes the activation of multiple cellular pathways and phenotypes allowing cell survival under low oxygen conditions [11]. One of the HIF-1 $\alpha$ targets is the CRE-binding protein 1 (CREB1) transcription factor which participates in the control of hypoxia-regulated gene expression [12]. CREB1 protein specifically binds with high affinity to consensus and variants of cAMP-responsive elements (CRE) in gene promoters, as a homodimer or heterodimer with c-Jun or CRE-BP1, functioning as a CRE-dependent trans-activator [13]. Multiple stimuli including cytokines and growth factors as well as the hypoxic tumor microenvironment can influence the CREB1 activity by different signal transduction pathways [14]. Another member of this family of transcription factors is CREB5 which also belongs to the ATF/CREB family of proteins, whose members are characterized by specific binding to CRE motif [15]. CREB5 has relevant functions in regulating cell growth, cell cycle, cell proliferation and differentiation. More recent studies have revealed a key role of CREB5 in tumor development and progression [15, 16, 17, 18, 19]. For instance, CREB5 showed increased expression levels, and positively correlated with advanced stages of disease and poor survival in colorectal cancer (CRC) patients [15]. In addition, functional assays suggested that CREB5 is involved in invasiveness and metastasis of CRC cells through regulation of the hepatocyte growth factor-MET gene promoter and downstream signaling [15]. Likewise, CREB5 was found upregulated in epithelial ovarian cancer (EOC) cells, and its high expression was correlated with disease stage, shorter overall survival and metastasis to pelvic lymph nodes in EOC patients [16]. Remarkably, it has been reported that overexpression of CREB5 may induce resistance therapy to androgen-receptors antagonist enzalutamide in prostate cancer [17]. In addition, CREB5 was associated with a poor prognosis in hepatocellular carcinoma (HC) patients [18]. Forced overexpression of CREB5 resulted in increased cell proliferation of HC cells, and the CREB5 silencing had the opposite effect [18]. Taken altogether, these studies suggested that CREB5 is a potential oncogene and that its overexpression was associated with poor prognosis and shorter overall survival in cancer patients.

Recent studies indicated that CREB5 expression could be regulated at posttranscriptional level by non-coding RNAs including long non-coding RNAs (lncRNAs), circular RNAs (circRNAs) and microRNAs in cancer cells [19, 20, 21]. MicroRNAs (miRNAs) are non-coding single-stranded small RNAs of 21-25 nucleotides in length that function as negative regulators of gene expression [22]. The post-transcriptional regulatory functions of miRNAs are mediated by degradation of the target mRNAs and inhibition of mRNA translation, through complementarity between the miRNA seed region and the 3 ${}^{\prime}$ -untranslated region (3 ${}^{\prime}$ UTR) of the target mRNAs [23]. Several studies showed that miRNAs have also a prominent role in the regulation of MV and angiogenesis. For instance, we and others previously reported that microRNA-204-5p (miR-204) was frequently downregulated in breast and gastric cancers, and that its ectopic restoration inhibited cell proliferation, tumor invasion, angiogenesis, metastasis and VM by targeting several oncogenes and cell signaling pathways including PI3K/AKT, RAF1/MAPK, VEGFA, and FAK/SRC [23, 24, 25, 26, 27]. However, the functional role of CREB5 in hypoxia-activated VM in breast cancer cells is unknown. Moreover, the potential regulation of CREB5 by miR-204, an hypoxamiRs previously reported as a VM regulator in ovarian and breast cancers [4, 26], remains also unexplored. In the present study, we investigated the CREB5 expression in public databases, breast cancer tissues and cell lines and its potential role in the development of VM. We also investigated whether miR-204 was able to regulate VM by regulating CREB5. Our data suggest that miR-204 exhibits a potential role in the regulation of VM by targeting CREB5 in breast cancer.

2. Materials and methods

2.1 Cell lines

Human MDA-MB-231 and MCF-7 breast cancer cell lines were obtained from the American Type Culture Collection, and routinely grown in Dulbecco’s modification of Eagle’s minimal medium (DMEM-F12) supplemented with 10% fetal bovine serum and penicillin-streptomycin (50 unit/ml) at 37 ${}^{\circ}$ C in 5% CO ${}_{2}$ atmosphere.

2.2 Analysis of CREB5 expression in public databases

The public datasets for CREB5 gene expression from breast tumors samples was obtained from Cancer Genome Atlas (TCGA) using Xena repository data (https://xenabrowser.net/datapages/ TCGA Brest Cancer dataset ID: TCGA.BRCA.sampleMap/HiSeqV2) and compared with Genotype-Tissue Expression (GTEx) information (dataset ID: TcgaTargetGtex_RSEM_isoform_fpkm) using XenaPhyton tools (https:// github.com/ucscXena/xenaPython) and UCSC Xena browser (https://xenabrowser.net/). A total of 1391 samples were matched and Welch’s $t$ -test was used considering a $p<$ 0.05 as statistically significant.

2.3 Kaplan-Meir analysis

Kaplan-Meier curves for CREB5 expression in breast cancer patients were determined as previously described [28]. Briefly, Kaplan-Meir on line tool (https:// kmplot.com/analysis/) use gene expression data and overall survival information of breast cancer patients reported in GEO (Affymetrix microarrays), EGA and TCGA. The database is handled by a PostgreSQL server, which integrates gene expression and clinical data simultaneously. To analyze the prognostic value of genes the patient samples were split into two groups according to various quantile expression of CREB5. A Kaplan-Meier survival plot compared the patient cohorts and the hazard ratio with 95% confidence intervals and logrank $P$ value were calculated.

2.4 MicroRNA-204 mimics transfection

MicroRNA-204 precursor and miR-204 scramble control were both transfected at 30 nM in MDA-MB-231 and MCF-7 cells using siPORT amine transfection agent (Ambion). Briefly, pre-miR-204 was diluted in 25 $\mu$ l Opti-MEM to obtain a concentration at 30 nM and added to wells containing 1 $\times$ 10 ${}^{7}$ cells grown in 450 $\mu$ l DMEM for 48 h. Then, cells were seeded and used for downstream analyses.

2.5 CREB5 gene silencing

Two oligonucleotides pairs (21–23 nt length) corresponding to specific short hairpin RNAs (shRNA) targeting the CREB5 gene were designed. To minimize the possibility of shRNA off targeting effects, a nucleotide BLAST search was carried out. Each oligonucleotide pair was cloned into the pSilencer 5.1 U6 retro plasmid (ThermoFisher) and sequences were confirmed by automatic sequencing. The resulting recombinant plasmids were named sh-CREB5.2 and sh-CREB5.3. The efficacy of the constructs in CREB5 gene silencing was tested by transient transfection of MDA-MB-231 and MCF-7 cells (2 $\times$ 10 ${}^{5}$ ) using lipofectamine 2000 reagent (Invitrogen) followed by CREB5 protein detection by Western blotting at 48 h after transfection.

2.6 Vasculogenic mimicry assays

Vasculogenic mimicry experiments were performed through 3-dimensional cell cultures on matrigel. Firstly, MDA-MB-231 cells (1 $\times$ 10 ${}^{4}$ cells/well) were transfected with sh-CREB5.2, sh-CREB5.3 and with the negative control previously described. The cells were cultured in 96-well plate covered with geltrex matrix (50 $\mu$ l). Afterward, cells were incubated at 37 ${}^{\circ}$ C in 5% CO ${}_{2}$ atmosphere in hypoxia conditions (1% O ${}_{2}$ ) for 48 h. Then, the formation of 3D tunnels was induced by culture of cells on matrigel during 0–9 hours, and then cells were observed under an inverted microscope (Iroscope SI-PH) and imaged during the course of time. Two observers individually counted the number of branch points and tubular structures. Data were expressed as mean $\pm$ S.D. $p<$ 0.05 was considered as statistically significant.

2.7 Western blots

Proteins were separated on 10% polyacrylamide gels and transferred to 0.2 $\mu$ m nitrocellulose membrane (Bio-Rad) in transfer buffer (25 mM Tris, 192 mM glycine and 10% methanol). Membranes were dried and blocked for 60 min at room temperature with TBST-1X (137 mM NaCl, 20 mM Tris, 0.1% Tween-20, at pH 7.6) containing 5% BSA (Sigma-Aldrich) and individually incubated overnight at 4 ${}^{\circ}$ C with rabbit anti-GAPDH (1:2000 Cell Signaling) and rabbit anti-CREB5 (1:1000 Abcam) primary antibodies, respectively. Membranes were washed three times in TBST-1X and incubated with horseradish peroxidase-conjugated goat anti-rabbit IgG (1:2,500, Zymed). Signal was detected and developed using ECL detection kit. Densitometry analysis of immunodetected bands in Western blots assays were performed using the public domain myImage analysis software.

2.8 Immunofluorescence analysis

Briefly, cells were fixed in 4% formaldehyde in PBS 1X for 30 min at room temperature. Coverslips were incubated with 0.1% Triton X-100 for 3 min. Following washing with PBS 1X, cells were blocked for 40 min at room temperature with 0.2% BSA in PBS 1X, and incubated with anti-CREB5 antibodies overnight at 4 ${}^{\circ}$ C. Samples were incubated with Alexa-488-labelled secondary antibodies and then, stained cells were washed with PBS 1X for 15 min. Finally, slides were assembled with vectashield ${}^{\text{\textregistered}}$ mounting media (Vector) containing DAPI and cells were observed under an Olympus FluoView FV1000 Confocal Microscope with an attached MRC1024 LSCM system (Bio-Rad). Cells were imaged from top to bottom in the Zplane; images from the basal plane of the cells were captured and stored as digital images.

2.9 Luciferase reporter assay

The 3 ${}^{\prime}$ UTR of CREB5 gene were amplified using specific primers and cloned downstream of luciferase gene into p-miR-report vector (Ambion). All constructs were verified through automatic sequencing. Then, recombinant pmiR-LUC-CREB5 and pmiR- LUC-CONTROL plasmids were transfected into MDA-MB-231 cells using lipofectamine 2000 (Invitrogen). At 4 h postransfection, the medium was replaced with complete fresh medium. 24 h after transfection, pre-miR-204 (30 nM) or pre-miR-negative control (scramble) were co-transfected with lipofectamine RNAi max (Invitrogen). Then 24 h after transfection, firefly and renilla luciferase activities were both measured by the Dual-Glo luciferase Assay (Promega) using a Flourskan Ascent FL microplate fluorometer (Thermo Scietific). Firefly luciferase activities were normalized with Renilla reniformis luciferase activities. $P$ values for differences between control and pre-miR-204 transfected cells were determined by two-tailed Student’s t test.

2.10 Statistical analysis

Experiments were performed three times by triplicate and results were represented as mean $\pm$ S.D. One-way analysis of variance (ANOVA) followed by Tukey’s test were used to compare the differences between means. A $p<$ 0.05 was considered as statistically significant.

3. Results

3.1 Expression of CREB5 is upregulated in breast tumors

To evaluate the CREB5 expression in breast cancer patients, we first examined its mRNA levels in a cohort of 1,391 breast tumors and matched adjacent normal samples from TGCA datasets. Our results showed that CREB5 mRNA was significantly ( $p<$ 0.0001) overexpressed in breast tumors compared with normal tissues in this large cohort of patients (Fig. 1A). Furthermore, using data obtained from CellExpress database, we found that CREB5 mRNA was also upregulated in MCF-7, T47-D, HCC2218, MDA-MB-453, MDA-MB-231 and Hs578t breast cancer cell lines compared to non-tumorigenic MCF-10A cells used as normal control (Fig. 1B).

Figure 1.

CREB5 is overexpressed in breast tumors and cancer cell lines. (A) CREB5 expression in 1391 breast tumors and matched adjacent normal samples from TCGA datasets (*** $p<$ 0.0001). (B) CREB 5 mRNA expression in MCF7, T47D, HCC2218, MDA-MB-453, MDA-MB-453 and Hs578t breast cancer cell lines, compared to the normal MCF10A cell line. Data were obtained from the CELLEXPRESS database. (C–D) Kaplan-Meier curves for overall survival according the expression of CREB5. A cohort of breast cancer patients with a mean follow-up of 33 months was analyzed. Gene expression data and overall survival information were downloaded from GEO (Affymetrix microarrays), EGA and TCGA. Prognostic values were obtained by splitting patient’s data into two groups according to various quantile expressions of CREB5. The two patient cohorts were compared by a Kaplan-Meier survival plot, and the hazard ratio with 95% confidence intervals and logrank $P$ value were calculated.

Figure 2.

CREB5 silencing in MDA-MB-231 and MCF-7 breast cancer cell lines. (A) Western blot assays for CREB5 knockdown in MDA-MB-231 and MCF-7 cells transfected with the sh-CREB5.2 or sh-CREB5.3 constructs using CREB5 antibodies (1:1000). GAPDH was used as an internal loading control. (B) Densitometric analysis of immunodetected bands in panel A. Data represents the mean of three independent assays $\pm$ SD. (* $p<$ 0.05).

Figure 3.

Silencing of CREB5 expression inhibits vasculogenic mimicry. (A) Images of vasculogenic mimicry of MDA-MB-231 cells transfected with sh-CREB5.3 (right panel), scramble control (middle panel) and non-transfected (left panel). Cells were submitted to hypoxia for 48 h, and then incubated in matrigel during 0–9-h. (B) Graphical representation of quantification of number of capillary tubes and (B) nodes as depicted in panel A. (D) Images of vasculogenic mimicry of MCF-7 cells after 48 h hypoxia and subsequent incubation in matrigel for 9 h. Experiments were performed three times in triplicate and the data were expressed as mean $\pm$ S.D.

3.2 High expression of CREB5 in basal subtype of breast cancer correlates with poor outcome

Then, we wondered whether high expression of CREB5 had clinical implications in the prognosis of breast cancer patients. Therefore, we performed an overall survival analysis using the Start Kaplan-Meier tool as implemented for breast cancer, which utilizes genome-wide transcriptome data and clinical overall survival information from breast cancer patients with a median follow-up of 150 months. Data showed that high expression of CREB5 transcript was associated with poor overall survival of breast cancer patients with basal subtype and lymph node-positive tumors (HR $=$ 1.88, logrank $p=$ 0.014) (Fig. 1C). Notably, this clinical association was more evident in patients with grade 3 lymph node-positive tumors (HR $=$ 2.54, logrank $p=$ 0.0048) (Fig. 1D). No significant associations were found between CREB5 expression and overall survival in luminal and HER2-positive subtypes of breast tumors (data not shown), suggesting that it could be a potential specific biomarker of disease progression in basal subtype of breast cancer.

Figure 4.

MicroRNA-204 targets the CREB5 gene. (A) Western blot of MDA-MB-231 and MCF-7 cells, non-transfected and transfected with pre-miR-204, using the CREB5 polyclonal antibodies (1:1000). GAPDH specific antibodies were used as control. (B) Densitometric analysis of bands in panel A. Data were normalized using GAPDH expression. Images are representative of three independent experiments. (C) Immunofluorescence assays using anti-CREB5 antibodies for CREB5 protein detection and DAPI for nuclear DNA staining (blue channel) in MDA-MB-231 (Alexa-555, red channel) and MCF-7 (Alexa-488, green channel) cells transfected with miR-204 and scramble as control. (D–E) Densitometric analysis of immunofluorescence signals in panel C. (F) Schematic representation of p-miR-Report construct containing the miR-204 binding sites of the 3 ${}^{\prime}$ UTR of CREB5 gene cloned downstream of luciferase gene. Seed sequences and miR-204 binding sites in targets genes are indicated in grey box. (G) Luciferase reporter activities in MDA-MB-231 cells co-transfected with miR-204 mimics and the recombinant plasmid described in F. Cells transfected with p-miR-Report plasmid alone or co-transfected with scramble or 3 ${}^{\prime}$ UTR mutant plasmid used as controls. Data represent the mean $\pm$ S.D. of three independent experiments performed by triplicate.

3.3 CREB5 silencing impairs vasculogenic mimicry in breast cancer cells

CREB5 is a hypoxia-activated gene but its role in VM is unknown. To determine if selective inhibition of CREB5 affects 3D channels-like formation representative of VM, its expression was knocked-down by RNA interference. Two specific short hairpin RNAs (designated as shCREB5.2 and shCREB5.3) targeting the CREB5 mRNA were designed and cloned into the pSilencer vector. Both constructs were individually transfected in triple negative MDA-MB-231 and luminal MCF-7 breast cancer cell lines, and subsequently CREB5 protein expression was analyzed by Western blot assays using specific antibodies 48 h post transfection. The results showed that the shCREB5.3 sequence efficiently down-regulated CREB5 expression in both cell lines (Fig. 2A), whereas no significant effect was observed with the shCREB5.2 interfering sequence. GADPH protein expression used as a loading control, showed no significant changes between both treatments. Densitometric analysis of the immunodetected bands showed that silencing induced by the shCREB5.3 construct was effective, as this sequence suppressed CREB5 expression about 85% in MDA-MB-231 cells and by 62% in MCF-7 cell line (Fig. 2B). Then, the effects of CREB5 silencing using shCREB5.3 were evaluated by hypoxia-induced 3D channels-like network formation assays representative of the early stages of VM in MDA-MB-231 cells. Results showed that, after 48 h hypoxia, VM was effectively induced in control cells showing an increased number of branch points and capillary tubes during the course of time (0–9 h) reaching the highest levels at 9 h (Fig. 3A, left panel). Interestingly, CREB5 silencing using the sh-CREB5.3 resulted in a dramatic inhibition of VM under hypoxia at 3–9 h (Fig. 3A, right panel). Likewise, a reduction in the number of branch points and capillary tubes compared to untreated control cells was found since the initial time points until 9 h of channels-like formation assays (Fig. 3A, right panel). No significant inhibition of VM was observed in scramble-transfected control cells, as expected (Fig. 3A, middle panel). Intriguingly, VM was not detected in poorly invasive MCF-7 breast cancer cells, suggesting that this process could be a specific feature of highly metastatic tumors such as MDA-MB-231 cancer cells (Fig. 3D).

3.4 MicroRNA-204 targets the transcription factor CREB5

In previous studies, we have shown that miR-204 inhibits the hypoxia-induced channels-like formation [4]; therefore, it was reliable to propose that it may regulates VM at least in part by targeting CREB5 as the bioinformatics analysis showed that it contains potential 3 ${}^{\prime}$ UTR binding sites for this miRNA. To corroborate whether miR-204 can bind to and repress the CREB5 expression, we first analyzed its protein expression levels by Western blot assays in miR-204 mimics transfected MDA-MB-231 cells, and compared with scramble-treated control cells. Data showed that CREB5 protein was abundantly expressed in MCF-7 cells and in less extend in MDA-MB-231 cells (Fig. 4A). In addition, data showed a significant decrease in CREB5 levels in both miR-204-transfected MCF-7 and MDA-MB-231 cells relative to controls (Fig. 4A and B). Immunofluorescence and confocal microscopy assays showed that CREB5 protein was mainly localized in the nucleus in both MCF-7 and MDA-MB-231 cell lines (Fig. 4C). Interestingly, a drastic decrease in CREB5 abundance in nuclei was observed in miR-204 mimics-transfected cells compared to scramble-transfected control cells in both cell lines in agreement with Western blot assays (Fig. 4C–E). In consequence, we next investigated if CREB5 mRNA is direct target of miR-204 using luciferase reporter assays. Using targetscan software, we identified a complementary site for miR-204 in the 3 ${}^{\prime}$ UTR sequence of CREB5 gene which was cloned downstream of the luciferase-coding region in the pmiR-Report plasmid (Fig. 4F). In addition, a mutated version of the miR-204 binding site at the CREB5 3 ${}^{\prime}$ UTR was included as a control. Data showed that ectopic expression of miR-204 and co-transfection of recombinant CREB5 3 ${}^{\prime}$ UTR wild type plasmid into MDA-MB-231 cells resulted in a significant reduction of the relative luciferase activity in comparison with controls (Fig. 4G). In addition, when mutated sequence was assayed no significant changes in luciferase activity were found.

4. Discussion

In spite of recent advances in chemotherapeutic treatment of locally advanced and metastatic disease, breast cancer still remains as a lethal disease. The tumor microenvironment is a key factor in the development of aggressive disease, as it activates diverse cell signaling pathways impacting the cancer hallmarks including angiogenesis, therapy resistance, invasion, metastasis and VM. Notably, VM has been associated with more aggressive tumors, poor prognosis, and shorter overall survival in breast cancer patients. Thereby, there is an urgent need for potential biomarkers of disease progression and novel therapeutic agents targeting VM [10]. Hypoxia activates the hypoxia inducible factor (HIF-1 $\alpha$ ) transcription factor which in turns triggers the activation of several genes involved multiple cellular pathways allowing cell survival under low oxygen concentrations.

cAMP response element-binding protein (CREB) is family of proteins that promote tumorigenesis in different types of cancer. CREB proteins have the ability to activate the expression of diverse target genes which are involved in important processes in tumor development including metabolism, angiogenesis, cell cycle, cell survival, proliferation, and DNA repair [29]. Most of our limited knowledge about the functions of CREB5 in tumors comes from colorectal cancer studies. Previous studies have shown that CREB5 is overexpressed in colorectal cancer, and showed that it may plays a key role in cell proliferation and metastasis by regulation of c-Jun, CSF1R, MMP9, PDGFRB, FIGF and IL6 [30]. Also, it was showed that CREB5 expression was upregulated in colorectal cancer and positively correlated with advanced WHO stages and TNM stages and shorter survival in patients. CREB5 silencing reduced the invasiveness and metastatic capacity of colorectal cancer cells in vivo. Furthermore, CREB5 promoted metastasis mainly through activation of hepatocyte growth factor-MET signaling [30]. Altogether, these studies indicate that CREB5 is an important oncogene with relevant roles in carcinogenesis, however, to our best knowledge no studies about the role of CREB5 in breast cancer have been reported yet. Therefore, here describe for the first time the characterization of CREB5 expression in clinical breast tumors and cancer cell lines, and its role in VM. CREB5 mRNA expression was upregulated in three breast cancer cell lines and breast tumors, which is in agreement with previous reports where it was also found as overexpressed in hepatocellular carcinoma, colorectal and prostate tumors suggesting an oncogenic role related to therapy resistance and worse prognosis [31, 32, 33]. In addition, it has been reported that increased CREB5 expression promotes cell proliferation, angiogenesis and metastasis in diverse types of cancer and it has been correlated with poor prognosis in cancer. Remarkably, CREB5 overexpression in breast tumors was associated with clinical parameters of patients suggested a potential role as a novel biomarker of disease progression.

As CREB5 seems to be a novel oncogene, here we focused in the study of its functions in vasculogenic mimicry in breast cancer. Silencing of CREB5 abolished the hypoxia-induced formation of 3D channels-like structures representative of the early stages of vasculogenic mimicry in breast cancer cells. These findings are novel as no previous role for CREB proteins have been reported, and added complexity to the molecular mechanisms operating at transcriptional level through the action of transcription factors in VM. To obtain clues on the mechanisms of CREB5 regulation we evaluated if its expression could be modulated by miR-204 in breast cancer cells, as they have been reported as major regulators of genes involved in angiogenesis and vasculogenic mimicry. CREB5 contains 4 predicted binding sites for miR-204 suggesting a posttranscriptional regulation by miR-204, a non-coding RNA that has been previously reported to be downregulated in several cancer cell lines and solid tumors, including breast cancer. As a negative regulator, miR-204 may inhibits the angiogenic by directly targeting key factors involved in, such as ANGPT1 and TGFBRII [23]. Interestingly, here we found that CREB5 expression was negatively regulated after restoration of miR-204 levels in breast cancer cells, suggesting that posttranscriptional regulation of CREB5 could represent an important control point of gene expression.

All together these finding raises several questions which remain to be solved in future studies: i) What is the transcriptional program activated by CRE5 transcriptional factor involved in vasculogenic mimicry activation? ii) Are there additional microRNAs potentially regulating CREB5 expression and 3D channel-like formation?, and iii) To what extend do CREB5-miR-204 interactions may occur in other types of human cancers?. In conclusion, we showed that tumor suppressor miR-204 exhibit a potential role in the regulation of vasculogenic mimicry through the miR-204/CREB5 axis in breast cancer.

Footnotes

Acknowledgments

We acknowledge to Universidad Autónoma de la Ciudad de México for support. This research was partially funded by Consejo Nacional de Ciencia y Tecnologia, grant number A1-S-13656.

Conflict of interest

The authors declare that they have no competing interests.

Author contributions

Conception: C.L.C. L.A.M and C.P.P.

Interpretation or analysis of data: E.C.S., C.P.F., A.F.P., Y.M.S.V., M.B.S.C., O.N.H.C., R.G.A.B.

Preparation of the manuscript: C.L.C and M.E.A.S. Revision for important intellectual content: Supervision: C.L.C., L.A.M., and M.E.A.S.

All authors have read and agreed to the published version of the manuscript.

References

Maniotis

A.J.

Folberg

Hess

Seftor

E.A.

Gardner

L.M.G.

Pe’er

Trent

J.M.

Meltzer

P.S.

Hendrix

M.J.

, Vascular channel formation by human melanoma cells in vivo and in vitro: Vasculogenic mimicry, The American Journal of Pathology 155 (1999), 739–752. doi: 10.1016/s0002-9440(10)65173-5.

Seftor

R.E.B.

Hess

A.R.

Seftor

E.A.

Kirschmann

D.A.

Hardy

K.M.

Margaryan

N.V.

and Hendrix

M.J.C.

, Tumor cell vasculogenic mimicry: From controversy to therapeutic promise, The American Journal of Pathology 181 (2012), 1115–1125. doi: 10.1016/j.ajpath.2012.07.013.

El Hallani

Boisselier

Peglion

Rousseau

Colin

Idbaih

Marie

Mokhtari

Thomas

J.L.

Eichmann

Delattre

J.Y.

Maniotis

A.J.

and Sanson

, A new alternative mechanism in glioblastoma vascularization: Tubular vasculogenic mimicry, Brain 133 (2010), 973–982. doi: 10.1093/brain/awq044.

Salinas-Vera

Y.M.

Gallardo-Rincón

García-Vázquez

Hernández-De la Cruz

O.N.

Marchat

L.A.

González-Barrios

J.A.

Ruíz-García

Vázquez-Calzada

Contreras-Sanzón

Resendiz-Hernández

Astudillo-de la Vega

Cruz-Colin

J.L.

Campos-Parra

A.D.

and López-Camarillo

, Hypoxamirs profiling identify miR-745 as a regulator of the early stages of vasculogenic mimicry in SKOV3 ovarian cancer cells, Frontiers in Oncology 9 (2019), 1–14. doi: 10.3389/fonc.2020.00889.

Sharma

Seftor

R.E.B.

Seftor

E.A.

Gruman

L.M.

Heidger

P.M.

Cohen

M.B.

Lubaroff

D.M.

and Hendrix

M.J.C.

, Prostatic tumor cell plasticity involves cooperative interactions of distinct phenotypic subpopulations: Role in vasculogenic mimicry, Prostate 50 (2002), 189–201. doi: 10.1002/pros.10048.

Peng

Wang

Shan

Peng

Dong

Shi

Zhao

Duan

Cheng

Zhang

and Duan

, The long noncoding RNA LINC00312 induces lung adenocarcinoma migration and vasculogenic mimicry through directly binding YBX1, Molecular Cancer 17 (2018), 1–6. doi: 10.1186/s12943-018-0920-z.

Zhang

J.G.

Zhou

H.M.

Zhang

J.N.

Liu

G.L.

and Li

, Hypoxic induction of vasculogenic mimicry in hepatocellular carcinoma: Role of HIF-1 α, RhoA/ROCK and Rac1/PAK signaling, BMC Cancer 20 (2020), 32. doi: 10.1186/s12885-019-6501-87.

Nisar

M.A.

Zheng

Saleem

M.Z.

Ahmmed

Ramzan

M.N.

Ud Din

S.R.

Tahir

Liu

and Yan

, IL-1β promotes vasculogenic mimicry of breast cancer cells through p38/MAPK and PI3K/Akt signaling pathways, Frontiers in Oncology 11 (2021), 618839. doi: 10.3389/fonc.2021.618839.

Huang

Bai

Liu

and Hou

, Effects of miR 93 on epithelial to mesenchymal transition and vasculogenic mimicry in triple negative breast cancer cells, Molecular Medicine Reports 23 (2021), 30. doi: 10.3892/mmr.2020.11668.

10.

Cao

Bao

Miele

Sarkar

F.H.

Wang

and Zhou

, Tumour vasculogenic mimicry is associated with poor prognosis of human cancer patients: A systemic review and meta-analysis, European Journal of Cancer 49 (2013), 3914–3923. doi: 10.1016/j.ejca.2013.07.148.

11.

Wei

Chen

Jiang

Peng

Liu

Ren

Hua

Zhou

Liao

Wang

Xiang

Zhou

Xiong

and Zeng

, Mechanisms of vasculogenic mimicry in hypoxic tumor microenvironments, Molecular Cancer 20 (2021), 7. doi: 10.1186/s12943-020-01288-1.

12.

Nakayama

, CAMP-response element-binding protein (CREB) and NF-κB transcription factors are activated during prolonged hypoxia and cooperatively regulate the induction of matrix metalloproteinase MMP1, The Journal of Biological Chemistry 288 (2013), 22584–2259595. doi: 10.1074/jbc.M112.421636.

13.

Craig

J.C.

Schumacher

M.A.

Mansoor

S.E.

Farrens

D.L.

Brennan

R.G.

and Goodman

R.H.

, Consensus and variant cAMP-regulated enhancers have distinct CREB-binding properties, The Journal of Biological Chemistry 276 (2001), 11719–11728. doi: 10.1074/jbc.M010263200.

14.

Steven

and Seliger

, Control of CREB expression in tumors: From molecular mechanisms and signal transduction pathways to therapeutic target, Oncotarget 7 (2016), 35454–35465. doi: 10.18632/oncotarget.7721.

15.

Wang

Qiu

Liu

Huang

Chen

Zhang

Ding

Liang

and Liao

, CREB5 promotes invasiveness and metastasis in colorectal cancer by directly activating MET, Journal of Experimental and Clinical cancer Research: CR 39 (2020), 168. doi: 10.1186/s13046-020-01673-0.

16.

Deng

Liao

Liu

and Yao

, CREB5 promotes tumor cell invasion and correlates with poor prognosis in epithelial ovarian cancer, Oncology Letters 14 (2017), 8156–8161. doi: 10.3892/ol.2017.7234.

17.

Hwang

J.H.

Seo

J.H.

Beshiri

M.L.

Wankowicz

Liu

Cheung

Qiu

Hong

A.L.

Botta

Golumb

Richter

Sandoval

G.J.

Giacomelli

A.O.

S.H.

Han

Dai

Pakula

Sheahan

and Hahn

W.C.

, CREB5 promotes resistance to androgen-receptor antagonists and androgen deprivation in prostate cancer, Cell Reports 29 (2019), 2355–2370.e6. doi: 10.1016/j.celrep.2019.10.068.

18.

Wang

S.T.

Zhang

Z.J.

Zhou

and Peng

B.G.

, CREB5 promotes cell proliferation and correlates with poor prognosis in hepatocellular carcinoma, International Journal of Clinical and Experimental Pathology 11 (2018), 4908–4916.

19.

Bhardwaj

Singh

Rajapakshe

Tachibana

Ganesan

Pan

Gunaratne

P.H.

Coarfa

and Bedrosian

, Regulation of miRNA-29c and its downstream pathways in preneoplastic progression of triple-negative breast cancer, Oncotarget 8 (2017), 19645–19660. doi: 10.18632/oncotarget.14902.

20.

Zhang

Wang

Lin

and Guo

, LncRNA SNHG5 affects cell proliferation, metastasis and migration of colorectal cancer through regulating miR-132-3p/CREB5, Cancer Biology and Therapy 20 (2019), 524–536. doi: 10.1080/15384047.2018.1537579.

21.

Zhang

Yamaguchi

Zhang

Wang

Tian

and Chen

, Circular RNA circVAPA knockdown suppresses colorectal cancer cell growth process by regulating miR-125a/CREB5 axis, Cancer Cell International 20 (2020), 103. doi: 10.1186/s12935-020-01178-y.

22.

Iorio

M.V.

Visone

Di Leva

Donati

Petrocca

Casalini

Taccioli

Volinia

and Chang-Gong

, MicroRNA signatures in human ovarian cancer, Cancer Research 67 (2007), 8699–8707. doi: 10.1158/0008-5472.can-07-1936.

23.

Guo

Ingolia

N.T.

Weissman

J.S.

and Bartel

D.P.

, Mammalian microRNAs predominantly act to decrease target mRNA levels, Nature 466 (2010), 835–840. doi: 10.1038/nature09267.

24.

Yuan

Wang

Liu

Zhan

Wang

and Xu

A.M.

, Histological and pathological assessment of miR-204 and SOX4 levels in gastric cancer patients, BioMed Research International (2017), 6894675. doi: 10.1155/2017/6894675.

25.

Flores-Pérez

Marchat

L.A.

Rodríguez-Cuevas

Bautista-Piña

Hidalgo-Miranda

Ocampo

E.A.

Martínez

M.S.

Palma-Flores

Fonseca-Sánchez

M.A.

Astudillo-de la Vega

Ruíz-García

González-Barrios

J.A.

Pérez-Plasencia

Streber

M.L.

and López-Camarillo

, Dual targeting of ANGPT1 and TGFBR2 genes by miR-204 controls angiogenesis in breast cancer, Scientific Reports 6 (2016), 34504. doi: 10.1038/srep34504.

26.

Salinas-Vera

Y.M.

Marchat

L.A.

García-Vázquez

González de la Rosa

C.H.

Castañeda-Saucedo

Tito

N.N.

Flores

C.P.

Pérez-Plasencia

Cruz-Colin

J.L.

Carlos-Reyes

Á.

López-González

J.S.

Álvarez-Sánchez

M.E.

and López-Camarillo

, Cooperative multi-targeting of signaling networks by angiomiR-204 inhibits vasculogenic mimicry in breast cancer cells, Cancer Letters 432 (2018), 17–27. doi: 10.1016/j.canlet.2018.06.003.

27.

Hong

B.S.

Ryu

H.S.

Kim

Lee

Moon

Kim

K.H.

Jin

M.S.

Kwon

N.H.

Kim

Chung

D.H.

Jeong

Kim

K.Y.

Lee

H.B.

Han

Yun

Kim

J.I.

Noh

D.Y.

and Moon

H.G.

, Tumor suppressor miRNA-204-5p regulates growth, metastasis, and immune microenvironment remodeling in breast cancer, Cancer Research 79 (2019), 1520–1534. doi: 10.1158/0008-5472.CAN-18-0891.

28.

Györffy

Lanczky

Eklund

A.C.

Denkert

Budczies

and Szallasi

, An online survival analysis tool to rapidly assess the effect of 22,277 genes on breast cancer prognosis using microarray data of 1,809 patients, Breast Cancer Research and Treatment 123 (2010), 725–731. doi: 10.1007/s10549-009-0674-9.

29.

Zhang

Odom

D.T.

Koo

S.H.

Conkright

M.D.

Canettieri

Best

Chen

Jenner

Herbolsheimer

Jacobsen

Kadam

Ecker

J.R.

Emerson

Hogenesch

J.B.

Unterman

Young

R.A.

and Montminy

, Genome-wide analysis of cAMP-response element binding protein occupancy, phosphorylation, and target gene activation in human tissues, Proceedings of the National Academy of Sciences of the United States of America 102 (2005), 4459–4464. doi: 10.1073/pnas.0501076102.

30.

and Ding

, Involvement of the CREB5 regulatory network in colorectal cancer metastasis, Yi Chuan = Hereditas 36 (2014), 679–684. doi: 10.3724/SP.J.1005.2014.0679.

31.

Ding

and Qi

, Screening and regulatory network analysis of survival-related genes of patients with colorectal cancer, Science China. Life Sciences 57 (2014), 526–531. doi: 10.1007/s11427-014-4650-1.

32.

Tang

Dong

and Wang

, CircBACH1/let-7a-5p axis enhances the proliferation and metastasis of colorectal cancer by upregulating CREB5 expression, Journal of Gastrointestinal Oncology 11 (2020), 1186–1199. doi: 10.21037/jgo-20-498.

33.

Pan

and Li

, The dual regulatory role of miR-204 in cancer, Tumour Biology: The Journal of the International Society for Oncodevelopmental Biology and Medicine 37 (2016), 11667–11677. doi: 10.1007/s13277-016-5144-5.