Critical role of miR-10b in B-RafV600E dependent anchorage independent growth and invasion of melanoma cells

Recent high-throughput-sequencing of cancer genomes has identified oncogenic mutations in the B-Raf genetic locus as one of the critical events in melanomagenesis. B-Raf encodes a serine/threonine kinase that regulates the MAPK/ERK kinase (MEK) and extracellular signal-regulated kinase (ERK) protein kinase cascade. In normal cells, the activity of B-Raf is tightly regulated and is required for cell growth and survival. B-Raf gain-of-function mutations in melanoma frequently lead to unrestrained growth, enhanced cell invasion and increased viability of cancer cells. Although it is clear that the invasive phenotypes of B-Raf mutated melanoma cells are stringently dependent on B-Raf-MEK-ERK activation, the downstream effector targets that are required for oncogenic B-Raf-mediated melanomagenesis are not well defined. miRNAs have regulatory functions towards the expression of genes that are important in carcinogenesis. We observed that miR-10b expression correlates with the presence of the oncogenic B-Raf (B-RafV600E) mutation in melanoma cells. While expression of miR-10b enhances anchorage-independent growth of B-Raf wild-type melanoma cells, miR-10b silencing decreases B-RafV600E cancer cell invasion in vitro. Importantly, the expression of miR-10b is required for B-RafV600E-mediated anchorage independent growth and invasion of melanoma cells in vitro. Taken together our results suggest that miR-10b is an important mediator of oncogenic B-RafV600E activity in melanoma.


Introduction
Melanoma is the most aggressive of all the skin cancers. B-Raf is a serine/threonine protein kinase that activates the MEK/ERK-signaling pathway. About 25-70% of malignant melanomas harbor gain-of-function mutations in the oncogene B-Raf [1]. Among several B-Raf gain-a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 of-function mutations, B-Raf V600E is the most common mutation and accounts for nearly 80% of them [1]. Not only does B-Raf V600E cause a sustained activation of ERK signaling pathway in melanoma, it is also critical for the malignant process and is one of the few identified driver mutations essential for melanoma proliferation and survival. The transformation of melanocytes to melanoma by B-Raf V600E requires activation of the MEK-ERK kinases cascade with multiple downstream components [2][3][4]. The mechanism that integrates the diverse components into a coordinated response to the B-Raf V600E mutation remains undefined.
MicroRNAs (miRNAs) are small, non-protein coding RNA molecules and they regulate gene expression through a combination of translational repression and mRNA destabilization [5]. Each miRNA targets~200 mRNA molecules [6]. Because of their pleiotropic potentials, miRNAs are attractive candidates as master regulators of the B-Raf V600E oncogenic transformation program. In this study, we identified for the first time, significant positive correlation between B-Raf V600E mutation and microRNA-10b (miR-10b) expression. Furthermore, we show that miR-10b is a novel downstream effector of B-Raf V600E and that B-Raf V600E plays a causal role in the induction of miR-10b in melanoma cell lines. Our results suggested that B-Raf V600E increased miR-10b expression by increasing the expression levels of helix-loophelix transcription factor Twist1. We also show that miR-10b induced by B-Raf V600E is able to increase invasive capacity and anchorage independent growth of melanoma cells.

In vitro Matrigel invasion assay
The polycarbonate membrane (8μ pore size) of FluoroBlok cell culture inserts (BD Biosciences) was coated with 60 uL of Matrigel (1:26 in serum-free medium) (BD Biosciences) and incubated at 37˚C for 2-3 hours. 1x10 4 of sk-mel-28 or M249 cells were plated on these inserts. 700mL of chemo-attractive medium (Dulbecco's Modified Eagle's medium, 1% P/S and 10% FBS) was added to the lower chambers (24-well BD Falcon TC companion plate). After 24 hours of incubation, the insert bottoms were dipped in 1X PBS and stained in Calcein AM reconstituted in DMSO to 1mg/mL-1ml in 700 ml 1X PBS (BD Biosciences). Digital images were captured on EVOS inverted microscope along with manual cell count and fluorescence reading (485-538nM).

ChIP assay
Chromatin Immunoprecipitations (ChIPs) were performed as described previously [15]. Briefly, cells were crosslinked with 1% formaldehyde for 10 minutes at room temperature. Chromatin was sheared by sonication and immunoprecipitated overnight with control rabbit IgG or an antibody to Twist (Rabbit polyclonal Anti-Twist, Santa Cruz (H-81), sc15393). Following reversal of cross-links, the DNA was purified by proteinase K digestion followed by phenol-chloroform extraction and ethanol precipitation. The purified DNA was subjected to quantitative real-time PCR using SYBR Green Master Mix (Qiagen) with an Applied Biosystems Prism 7500 PCR system and analyzed with the SDS software. Primers for Ebox1 or Ebox2 were described previously [13].

RNA extraction and TaqMan miRNA quantitative real-time PCR assay
RNA extraction was carried out using Qiazol reagent (Qiagen). TaqMan miRNA qRT-PCR was performed according to manufacturer's instructions (TaqMan). RNU6 or RNU44 were used as the internal control. Briefly, in the reverse transcription (RT) step, cDNA was reverse transcribed from total RNA samples using specific looped miRNA primers from the TaqMan miRNA assays and reagents from the TaqMan. In the PCR step, PCR products were amplified from cDNA samples using the TaqMan miRNA assay together with the TaqMan universal PCR master mix.

Soft agar assay for anchorage independence
1mL layer of 1:1 mixture of 1.2% agar and 2X medium (to give a final 0.6% agar concentration) was evenly spread on 35mm plates. It was allowed to set for 30 minutes at RT. Cells were counted and plated in a mixture of 1.2% agar and 2X medium (1:4) to give a final agar concentration of 0.3%. 3x10 3 of sk-mel-28 cells or 1x10 4 of Mel 505 cells or 1x10 4 of sk-mel-197 cells were plated. Number of colonies formed were stained with MTT and counted under bright field microscope at 4X objective. Cells were plated in triplicates and four fields were counted per 35 mm plate.

TCGA data analysis
The publicly available TCGA data were directly downloaded from the TCGA Data Portal at https://tcga-data.nci.nih.gov/. TCGA melanoma cases (TCGA-SKCM) annotated for B-Raf WT or V600E mutation were extracted from the GDC portal (https://portal.gdc.cancer.gov). Among those, 407 samples were profiled for both mRNA expression and miRNA expression. Fragments per Kilobase per Million mapped reads (FPKM) from the RNA-sequencing data and reads per million miRNA mapped from the miRNA data were used as normalized read counts. Pearson correlation coefficients and P-values for miR-10b expression with Twist1 and MYC expression were calculated, respectively.

Western blot analysis
Cell extracts were prepared and western blotting was carried out. Briefly, cells were lysed with 20 mM Tris, pH 7.4, 150mM NaCl, 2mM EDTA, and 1% Triton X-100. Samples (10-50 μg protein) were separated by sodium dodecyl sulfate-polyacrylamide electrophoresis and then electrophoretically transferred from the gel to polyvinylidene fluoride membranes (Millipore). All the primary antibodies were diluted in phosphate buffered saline, pH 7.4, 0.2% Tween-20, 5% bovine serum albumin and 0.002% sodium azide. Following three washes, blots were incubated with the appropriate horseradish peroxidase-conjugated secondary antibody for 1 h at room temperature. Proteins were visualized with enhanced chemiluminescence with the BioRad ChemiDoc EQ system.

Statistical analysis
All experiments were performed in triplicates. All the statistical analyses were performed using the two-tailed student's t-test (GraphPad Prism software).

Results and discussion
We previously reported that the expression of several miRNAs was altered upon introduction of B-Raf V600E in primary melanocytes [16]. Among the miRNAs whose expression levels were affected by B-Raf V600E , we chose miR-10b for further study because its expression positively correlates with V600E mutation in B-Raf in established melanoma cell lines (Fig 1), clinical melanoma samples (S1 Fig), and because it is involved in several other malignancies [17].
To investigate if B-Raf V600E plays a causal role in regulating miR-10b expression in melanoma cell lines we undertook a loss-of-function approach. We stably knocked down B-Raf V600E expression by siRNA in two different B-Raf V600E melanoma cell lines. As expected, B-Raf protein expression significantly decreased in B-Raf V600E knockdown cells with a concomitant decrease in phosphorylation of ERK1/2. In both studied cell lines, we found that miR-10b expression was significantly down-regulated upon B-Raf V600E knock down (Fig 2A). A decrease in miR-10b expression was also observed in M249, which are B-Raf V600E positive melanoma cells when B-Raf V600E activity was down-regulated by vemurafenib (PLX4032), a pharmacological inhibitor of B-Raf V600E (Fig 2B).
We reasoned that if the effect of knocking down of B-Raf V600E expression on miR-10b is physiological we should observe an opposite effect with over-expression of the oncogenic B-Raf V600E in melanoma cell lines carrying wild-type B-Raf. Indeed, we found that the expression of miR-10b was significantly upregulated in cells ectopically expressing B-Raf V600E as compared to empty vector control (Fig 2C). The drawback of over-expression of an oncogene is that the cells might get adapted to its sustained expression and this can lead to non-specific effects that may not be attributable to the expression of the oncogene alone. As a complementary approach, we used an estrogen receptor fusion system where B-Raf V600E was rendered functionally hormone-dependent by fusion with synthetic steroid 4-hydroxytamoxifen (4-OHT)-binding domain of mutated estrogen receptor (ER T2 ) [19]. As expected, upon treating the cells with 4-OHT for 24 hours, we observed potent induction of ERK dual phosphorylation at Thr 202 and Tyr 204 indicating that the kinase activity of B-Raf V600E was temporally turned on by 4-OHT treatment (Fig 2D, lower right panel). Interestingly, transient activation of B-Raf V600E is better at activating miR-10b expression than sustained expression of the constitutively active kinase (Compare Fig 2C and 2D). Taken together, our results suggest that the expression of miR-10b in melanoma is B-Raf V600E dependent.
In humans, the miR-10b gene resides upstream from the developmental gene Hoxd4. It has been shown that the basic helix-loop-helix transcription factor Twist positively regulates miR-10b expression in breast cancer cells by directly binding to the proximal E-box present in the miR-10b gene promoter. The expression levels of Twist are often deregulated in a vast majority of melanoma tumors and cell lines [20,21]. Furthermore, the expression of Twist was shown to be regulated at the transcription level by B-Raf V600E mutation in human melanoma cells [22], raising the possibility that B-Raf V600E may increase miR-10b expression by activating Twist. Indeed, hyper activation of B-Raf signaling pathway increases Twist1 transcript in established melanoma cell lines (Fig 3A). In addition, we also observed significant correlation of miR-10b and Twist1 expression in B-Raf V600E mutated but not B-Raf wild-type melanoma clinical samples (Fig 3B and Table 1). The observed correlation is specific as no significant correlation was observed between another E-box transcription factor c-myc and miR-10b (Fig 3C  and Table 2). Importantly, knocking down of B-Raf V600E diminished Twist1 expression while the expression of Twist was important for miR-10b synthesis (Fig 3D-3F, S2 Fig). It is of interest to note that knocking down of wild type B-Raf in Mel505 cells had no noticeable effect on the expression levels of Twist1 (Fig 3E, right panel). Finally, Twist1 may have a direct effect on miR-10b expression as the promoter occupancy of miR-10b by Twist1 was elevated upon exogenous expression of B-Raf V600E in B-Raf wild-type Mel505 melanoma cell line (Fig 3G). Our results therefore suggested that the activation miR-10b expression by B-Raf V600E required the presence of Twist1 in melanoma. It is possible that B-Raf gain-of-function mutation is necessary but not sufficient for miR-10b expression in melanoma. Indeed, we did not observe significant correlation of B-Raf mutation status and miR-10b expression level in melanoma TCGA datasets before the clinical samples were stratified into Twist1 positive and negative groups (Fig 3D).
Studies with cancer cell transplantation and autochthonous cancer mouse models have demonstrated the causal role of miR-10b in breast cancer initiation, growth, progression and metastasis [23][24][25]. However, the causality of loss-or gain-of-function of miR-10b in other cancers including melanoma is not known. Anchorage-independent growth and increased invasive capacity are two well-characterized hallmarks of cancer. Since Twist-miR-10b axis has a demonstrated role in anchorage-independent growth [26], miR-10b may confer these important cancerous phenotypes on melanoma cells. Indeed, in low-miR-10b-expressing wild-type B-Raf melanoma cells Mel505, ectopic expression of miR-10b was sufficient to increase anchorage-independent growth as measured in soft agar (Fig 4A). This effect was not cell type specific as a similar effect was observed with other wild-type B-Raf melanoma cells sk-mel-197. Similarly, depletion of miR-10b in high-miR-10b-expressing B-Raf V600E melanoma cells skmel-28 and M249 decreased their invasiveness (Fig 4B). B-Raf V600E is an oncoprotein and its presence is essential for anchorage-independent growth and invasive capacity of B-Raf V600E melanoma cells. B-Raf V600E induces miR-10b expression in melanoma cell lines. Therefore, B-Raf V600E melanoma cells may acquire cancer phenotypes partly by increasing the expression levels of miR-10b through increasing the expression of the transcription factor Twist1. Consistent with this line of thinking, as shown in Fig 5A and 5B, ectopic miR-10b expression significantly rescued the decrease in anchorage independent growth and invasiveness due to the loss of B-Raf V600E in sk-mel-28 cells. The mean value for the wild-type cells was 1254.7, and for the mutant cells 1627.4. The p-value for a one-tailed t-test was 0.0422. The data are from GSE89438 [18]. The expression values for miR-10b were downloaded as part of the GSE89438_series_matrix file from NCBI GEO. The miR-10b values were imported into the R programming environment, and the boxplot function was applied to the values for wild-type and B-Raf V600E mutant samples. https://doi.org/10.1371/journal.pone.0204387.g001 Critical role of miR-10b in B-Raf V600E dependent anchorage independent growth and invasion of melanoma cells The mechanisms of how augmented miR-10b expression in B-Raf V600E melanoma cells enhances their anchorage independent growth and invasiveness is currently not known. miR-NAs function mainly by silencing expression of target genes by binding to specific sites of the targeted gene mRNAs. Studies of genetic deletion of miR-10b in MMTV-PyMT mice identified Tbx5, Pten, and HoxD10 as key miR-10b targets [27]. While Tbx5 and Pten are well-characterized tumor suppressor genes, HoxD10 is a proven metastasis suppressor gene. To grow continuously after being detached from extracellular matrix, cancer cells develop resistance to a specific form of caspase-mediated cell death known as anoikis by upregulating the survival pathways. PI3K pathway is a major survival pathway cancer cells employ to overcome anoikis to achieve anchorage-independent growth [28]. Of the three target genes of miR-10b, Pten has been shown to play a pivotal role in dampening the PI3K pathway. It is therefore possible that miR10b enhances anchorage independent growth in melanoma by decreasing the expression of Pten. Both Tbx2 and HoxD10 are transcription factors that can potentially regulate a multitude of genes. It has been shown that HoxD10 represses the expression of metastasis gene RhoC in breast cancer cells and that miR-10b increases cancer cells invasion by repressing the repressor of RhoC expression. RhoC is a well-studied metastasis gene in melanoma and it is possible that B-Raf V600E utilizes the same miR-10b-HoxD10-RhoC axis to drive cancer cells invasion. Likewise, Tbx2 can also play an equally important role in mediating the different effects on melanomagenesis resulting from B-Raf mutation.
In 2010, an orally available inhibitor of B-Raf V600E increased the median progression-free survival of patients with malignant melanoma harboring this mutation, to more than 7 months [29]. However, patients develop resistance to this drug due to several mechanisms including Critical role of miR-10b in B-Raf V600E dependent anchorage independent growth and invasion of melanoma cells receptor tyrosine kinase (RTK) or N-Ras upregulation among many others [30]. Recently, combinatorial therapy utilizing B-Raf and MEK inhibition delayed the emergence of resistance while progression-free survival remained more or less similar to B-Raf inhibition alone [31]. However, relapse due to drug-resistance is commonplace and warrants a treatment strategy based on new downstream molecular targets of the mutant B-Raf protein. The expression of miR-10b is often deregulated in cancers including melanoma and its expression was positively associated with worse patient survival [24]. Here we identify miR-10b as a novel molecular target downstream of mutant B-Raf protein raising the possibility that targeting miR-10b and/or its effectors may represent a rational alternative treatment for relapsed patient with B-Raf V600E melanoma.