Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jul 1.
Published in final edited form as: J Invest Dermatol. 2021 Dec 8;142(7):1956–1965.e2. doi: 10.1016/j.jid.2021.09.040

miR-181a promotes multiple pro-tumorigenic functions through targeting TGFβR3

Vida Chitsazzadeh 1,*, Tran N Nguyen 2,*, Alvaro de Mingo Pulido 2, Bruna B Bittencourt 2, Lili Du 1, Charles H Adelmann 1, Ivannie Ortiz Rivera 2,+, Kimberly A Nguyen 2, Leah D Guerra 1, Andrew Davis 3, Marco Napoli 3, Wencai Ma, R Eric Davis, Kimal Rajapakshe 4, Cristian Coarfa 4, Elsa R Flores 3, Kenneth Y Tsai 2,5
PMCID: PMC9174350  NIHMSID: NIHMS1762867  PMID: 34890627

Abstract

Cutaneous squamous cell carcinoma (cuSCC) comprises 15–20% of all skin cancers and has a well-defined progression sequence from precancerous actinic keratosis (AK), to invasive cuSCC. In order to identify targets for chemoprevention, we previously reported a cross-species analysis to identify transcriptional drivers of cuSCC development and identified miR-181a as a potential oncomiR.

We show that upregulation of miR-181a promotes multiple pro-tumorigenic properties by targeting an understudied component of TGFβ signaling, TGFβR3. miR-181a and TGFβR3 are upregulated and downregulated, respectively, in cuSCC. miR-181a overexpression (OE) and TGFβR3 knockdown (KD) significantly suppresses UV-induced apoptosis in HaCaT cells and in primary normal human epidermal keratinocytes (NHEK). In addition, OE of miR-181a or KD of TGFβR3 by shRNA enhances anchorage-independent survival. miR-181a OE or TGFβR3 KD enhances cellular migration and invasion and upregulation of EMT markers. Luciferase reporter assays demonstrate that miR-181a directly targets the 3’UTR of TGFβR3. miR-181a upregulates pSMAD3 levels following TGFβ2 administration and results in elevated SNAIL and SLUG expression. Finally, we confirm in-vivo, that miR-181a inhibition compromises tumor growth. Importantly, these phenotypes can be reversed with TGFβR3 OE or KD in the context of miR-181a OE or KD, respectively, further highlighting the physiologic relevance of this regulation in cuSCC.

Keywords: TGFβR3, miR-181a, epithelial-mesenchymal transition, cutaneous squamous cell carcinoma

INTRODUCTION

In the United States, over 700,000 cases of cutaneous squamous cell carcinoma (cuSCC) are diagnosed annually and account for about 15–20% of all skin cancers (Parikh et al., 2014). Our understanding of the molecular and genetic events that lead to sequential progression of normal skin (NS) to precancerous actinic keratosis (AK) to cuSCC is limited. This represents a fundamental gap in our knowledge and understanding of this progression sequence is of relevance to understanding cancer development and developing more effective chemoprevention strategies. In an effort to identify transcriptional drivers of cuSCC development, we previously genomically profiled the development sequence from NS to AK to invasive cuSCC. This effort identified major microRNA drivers, as identified through functional pair analysis (Chitsazzadeh et al., 2016).

Our initial analysis suggested that miR-181a is upregulated during cuSCC development. The miR-181 family is highly evolutionarily conserved across vertebrates with roles in differentiation of hematopoietic cells, including lymphocytes, natural killer cells, and megakaryocytes (Weng et al., 2015). Pathway analysis reveals that miR-181 family target genes play important roles in cancer, axon guidance, actin cytoskeleton, MAPK signaling, and T cell receptor signaling pathways (Yang et al., 2014).

Several studies show upregulation of miR-181 expression in colorectal carcinoma (Nishida et al., 2012), ovarian cancer (Parikh et al., 2014) and hepatocarcinoma (Brockhausen et al., 2015) with roles in cell cycle, apoptosis, proliferation, migration and invasion (Zhang et al., 2012) through targets such as BCL2L11 (BIM) (Taylor et al., 2013), BCL-2 (Chen et al., 2010), ATM (Wang et al., 2011), and KRAS (Shin et al., 2011). In a recent meta-analysis of 21 studies involving 1685 patients, elevated expression of miR-181a was identified as a negative prognostic molecular marker in human head and neck squamous cell carcinoma (HNSCC) (Jamali et al., 2015), a disease that closely resembles cuSCC genomically (Chitsazzadeh et al., 2016, Li et al., 2015). Here, we define a mechanistic link between miR-181a, susceptibility to apoptosis, cellular motility and EMT, and TGFβR3. We show that miR-181a is able to confer several pro-tumorigenic properties upon epithelial cells, that these phenotypes are due to the direct regulation of TGFβR3 expression, and that inhibition of miR-181a compromises tumor growth in-vivo.

RESULTS

miR-181a overexpression is observed in cutaneous squamous cell carcinoma

First, we examined miR-181a expression in two different non-melanoma skin cancer cohorts (Figure 1a). A tissue microarray containing 37 evaluable cases of cuSCC and 9 normal sun-exposed skin controls (US Biomax) was hybridized with a locked nucleic acid (LNA, Exiqon) anti-miR-181a or nonspecific LNA anti-miR control and expression quantified within tumor cells and normal epidermis, respectively, using the Aperio Image system. We found that miR-181a was upregulated in the majority of cases, by a mean of 3.0 ± 0.5-fold (P < 0.0001, unpaired t-test) (Figure 1a, b). Furthermore, Taqman qPCR showed that miR-181a-5p was increased in 8 out of 11 SCC cell lines when compared to normal human epithelial keratinocyte (NHEK) (Supplementary Figure S1).

Figure 1. miR-181a is overexpressed in cuSCC and enhances anchorage-independent growth.

Figure 1.

Open in a new tab

a, ISH detection using LNA detection probe (blue) or scramble-miR as negative control were performed on FFPE tissue microarray. Error bars represent the mean (+) S.E.M.; **** P<0.0001 (unpaired t-test). b, Representative images show that cuSCC has higher miR-181a expression compared to normal sun-exposed skin. Scale bar = 50 μm. c, Transcriptome of HaCaT stable cell lines overexpressing lenti-miR-181a was analyzed by IPA for their associated molecular pathways. d, HaCaT stable cell lines overexpressing lenti-miR-181a or lenti-control were tested for their survival in liquid culture and enhance colony formation in soft agar. Error bars represent the mean (+) S.E.M.; **** P<0.0001 (unpaired t-test).

miR-181a overexpression increases colony formation efficiency and suppresses UV-induced apoptosis.

Next, we sought to determine the potential mechanism of action of miR-181a by assessing the pathways regulated by it. We performed microarray gene expression analysis (Illumina BeadArray) on lenti-miR-181a HaCat (miR-181a overexpressed) and lenti-miR-00 HaCat (negative control). Ingenuity Pathway Analysis (IPA, Qiagen) revealed enrichment of cell proliferation and cellular movement pathways (Figure 1c, Supplementary Table S1).

HaCaT cells overexpressing with miR-181a were tested for their survival in liquid culture and enhanced colony formation in soft agar. As a positive control we used Ha-RasV12-transformed HaCaT cells, which readily form colonies and tumors in-vivo (Vin et al., 2013). Our results showed that HaCaT miR-181a overexpressing cell lines have an increased capability of colony formation in soft agar. miR-181a-overexpressing HaCaT cells produced 6.3 ± 0.6 fold more colonies (P < 0.0001, unpaired t-test) (Figure 1d, Supplementary Figure S2a), reflective of a significant increase in anchorage-independent proliferation.

Exposure to UV radiation (UVR) can induce apoptosis of mammalian cells. Since UV exposure is the primary environmental cause of skin cancer, UV-induced apoptosis represents an important tumor suppressive mechanism. To identify a potential link between miR-181a expression and UV-induced apoptosis, we used NHEK and HaCaT overexpressing miR-181a and the negative lenti-miR-control. Cells were subjected to a flow cytometry-based apoptosis assay (Annexin V) 24 hours following 750 J/m2 UV-irradiation (Vin et al., 2013). Overexpression of miR-181a strongly suppressed UV-induced apoptosis in both NHEK and HaCaT cells by 86.8% ± 3.5% (P < 0.001, unpaired t-test) and 45.4% ± 2.4 % (P < 0.01, unpaired t-test), respectively (Figure 2a and b). Ataxia telangiectasia (ATM) is a key regulator of DNA damage signaling pathway that activates p53 (Canman and Lim, 1998, Khanna and Chenevix-Trench, 2004) and a validated target of miR-181a (Bisso et al., 2013). ATM regulates cell cycle progression through phosphorylation of cell-cycle checkpoint kinase 1 (CHEK1) and p53, and DNA damage repair through phosphorylation histone H2AX (Smith et al., 2010). Here, we showed that ATM and p53 expression correlate conversely with miR-181a level. We also noticed that phosphorylation of DNA damage response proteins such as CHEK1, p53 and H2AX increased to a lesser extent when cells with higher level of miR-181a were exposed to UV (Figure 2c).

Figure 2. miR-181a suppresses UV-induced apoptosis and increases invasiveness of keratinocytes.

Figure 2.

Open in a new tab

a-b, NHEK and HaCat cells were stably transfected with lenti-miR-181a or lenti-miR-control. Cells were sham irradiated or irradiated with 750 J/m2 of UVB. Apoptotic cells were determined by flow cytometry. Error bars represent the mean (+) S.E.M.; *** P < 0.001, ** P < 0.01 (unpaired t-test). c, Western blots showed that overexpression impairs the proper induction of ATM, p-p53, p-H2AX, and p-CHK1 in response to UV exposure. d-e, Trans-well invasion assay was performed at 48 h after NHEK were transfected with control-mimic, or miR-181a-5p mimic and RDEB2 were transfected with control-inhibitor, or miR-181a-5p inhibitor. Scale bar = 50 μm. Error bars represent the mean (+) S.E.M.; * P < 0.05, ** P< 0.01 (unpaired t-test).

miR-181a overexpression increases keratinocyte invasiveness

Because the cellular movement pathway was found to be highly enriched in miR-181a overexpressing HaCaT cells, we asked whether miR-181a affected keratinocyte morphology and motility and invasiveness. Interestingly, during the course of apoptosis assays, we observed that miR-181a overexpression induced morphological changes in NHEKs, with cells adopting a more spindled appearance (Supplementary Figure S2b). These observations suggested to us that miR-181a might also regulate keratinocyte epithelial-mesenchymal transition (EMT).

In order to address this, we performed Boyden chamber Matrigel invasion assays using NHEK and RDEB2 cuSCC cells (Figure 2d and e). First, we overexpressed miR-181a in NHEK. We observed that miR-181a-5p increased NHEK invasiveness significantly by 1.7 ± 0.2-fold compared to non-targeting control-expressing NHEK (P < 0.05, unpaired t-test). We then depleted miR-181a in RDEB2 cuSCC cells. Depletion of miR-181a decreased RDEB2 invasiveness by 1.7 ± 0.1-fold compared to non-targeting control-expressing cells (P < 0.01, unpaired t-test). Finally, we found that miR-181a expression is anticorrelated, for some cuSCC cell lines, with wound closure times in a scratch assay consistent with the notion that baseline levels of miR-181a may regulate motility and invasiveness.

TGFβR3 is a direct target of miR-181a

Since the biological significance of miRNA-driven regulation relies on the effect upon their cognate mRNA targets, we then analyzed the predicted targets of miR-181a using Targetscan (Agarwal et al., 2015), PicTar (Krek et al., 2005) and miRanda (Enright et al., 2003, Griffiths-Jones et al., 2008). The predicted target genes of miR-181a included transforming growth factor beta receptor III, TGFβR3. By using RNA-seq data from human cuSCC samples (Chitsazzadeh et al., 2016), we confirmed that TGFβR3 expression is significantly decreased in cuSCC as compared to normal skin by 1.8 ± 0.2-fold (P<0.0001, unpaired t-test) (Figure 3a). Western blot analysis demonstrated downregulation of TGFβR3 in addition to downregulation of the established target ATM (Figure 3b).

Figure 3. TGFβR3 is downregulated in cuSCC and is a direct target of miR-181a.

Figure 3.

Open in a new tab

a, Data obtained by RNA-sequencing shows TGFβR3 is downregulated in cuSCC relative to normal skin. Error bars represent the mean (+) S.E.M.; **** P < 0.05 (unpaired t-test). Available data in GSE/GEO. b, Western blot analysis of NHEK transiently transfected with control-mimic, or miR-181a mimic showed that miR-181a regulates TGFβR3 at the post-transcriptional level and downregulates the established target ATM. c, Diagram of miR-181a predicted seed regions in the WT and MUT 3’UTR of TGFβR3. d, Reporter assay in SRB1 with co-transfection of WT reporter plasmid and control-mimic, or miR-181a-5p mimic. e, Reporter assay in RDEB2 with co-transfection of WT reporter plasmid and control-inhibitor or miR-181a-5p inhibitor. f, Reporter assay in HaCat, with co-transfection of WT-or MUT reporter and control-mimic, or miR-181a-5p mimic. Data represent the S.E.M. from three replicates. *P < 0.05; **P < 0.01; ****P < 0.0001 (unpaired t-test).

We then sought to address the hypothesis that TGFβR3 is a direct the target of miR-181a in keratinocytes and we asked whether the 3’-UTR of TGFβR3 could confer direct regulation by miR-181a. We used plasmids harboring the 3’-UTR of TGFβR3 upstream of a luciferase reporter gene (Figure 3c). Because of the size of the 3’-UTR, we split it into two segments labeled “A” and “B”. Only segment “B” contained the two predicted miR-181a binding sites. We use SCC cell line that has low miR-181a (SRB1) and high miR-181a (RDEB2) for the reporter assay. In SRB1, the wild-type (WT) reporter plasmids were co-transfected with a miR-181a mimic or, control-mimic. Reporter assays were performed 48hr post-transfection. Compared with the control-mimic, presence of the miR-181a mimic significantly decreased the relative luciferase activity by 3.5±0.4-fold (P<0.005, unpaired t-test) when co-transfected with the WT reporter plasmid for segment “B” only (Figure 3d). On the other hand, when we depleted miR-181a in RDEB2 and perform the same assay as with SRB1, luciferase activity increased by 1.6 ± 0.2-fold (P<0.05, unpaired t-test) (Figure 3e).

To demonstrate the specificity of control conferred by the predicted miR-181 binding sites, we generated the mutant luciferase reporter plasmids for segment “B”. The WT or mutant reporter plasmid was transfected into miR-181a constitutively overexpressing HaCat cells. While the WT reporter showed downregulation of luciferase activity these cells, the mutant reporters with site 1 and site 2 individually mutated did not show any significant decrease in luciferase activity (Figure 3f). These results show that miR-181a suppresses TGFβR3 by targeting the two predicted sites within the 3’-UTR of TGFβR3.

TGFβR3 depletion phenocopies miR-181a effect in suppressing UV-induced apoptosis

Having demonstrated that TGFβR3 is a direct target of miR-181a, we then wanted to assess the degree to which phenotypes conferred by miR-181a-5p could be accounted for by downregulation of TGFβR3. To this end, we determined the effects of suppressing TGFβR3 expression using shRNA in NHEK and HaCat (Figure 4a, b; Supplementary Figure S3). Following UV-irradiation, we performed apoptosis assays and observed that TGFβR3 depletion suppressed UV-induced apoptosis in both NHEK and HaCaT cells by 63.6% ± 2.1% (P < 0.005, unpaired t-test) and 73.9% ± 1.8% (P<0.01, unpaired t-test), respectively. Taken together, our data show that both overexpression of miR-181a and depletion of TGFβR3 individually can suppress UV-induced apoptosis in keratinocytes (Figure 4a, b). To further functionally corroborate the direct link between mir-181 and TGFβR3, we overexpressed miR-181a and TGFβR3 in the same cells, and we were able to observe loss of the protective anti-apoptotic effect of miR-181a overexpression, clearly indicating that TGFβR3 overexpression can bypass miR-181-mediated activities. In immortalized, non-transformed Ker-CT cells, miR-181a overexpression suppressed UV-induced apoptosis by 2.3-fold ± 0.1 (P<0.01, unpaired t-test) and simultaneous TGFβR3 depletion by shRNA restored apoptosis to 87.5% ± 5.4% of control levels (P<0.01, unpaired t-test) (Figure 4c).

Figure 4. Downregulation of TGFβR3 phenocopies effect of miR-181a upregulation.

Figure 4.

Open in a new tab

a-b, Depletion of TGFβR3 phenocopies the suppression of UV-induced apoptosis by miR-181a in keratinocytes. NHEK and HaCat were transfected with sh-control, or sh-TGFβR3 After 24 h, cells were sham irradiated or irradiated with 750 J/m2 of UVB. Apoptotic cells were determined by flow cytometry. Error bars represent the mean (+) S.E.M.; *** P<0.001, ** P<0.01 (Student’s t-test). c, TGFβR3 overexpression negates effects of miR-181a overexpression in suppressing UV-induced apoptosis in keratinocytes. Using Ker-CT immortalized keratinocytes (due to greater transduction efficiency) miR-181a and TGFβR3 were overexpressed singly and in combination. miR-181a overexpression along suppressed UV-induced apoptosis by 2.3-fold ± 0.1 (red), and simultaneous TGFβR3 depletion by shRNA restored apoptosis to 87.5% ± 5.4% of control levels (blue). Error bars represent the mean (+) S.E.M.; ** P<0.01 (Student’s t-test). d-e, TGFβR3 overexpression negates effects of miR-181a overexpression on promoting migration and invasion in keratinocytes. miR-181a stably expressed HaCat (lenti-miR-181a) and non-targeting control stably expressed HaCat (lenti-miR-control) were transfected with TGFβR3 expressed plasmid or vector control. Assays conducted 48 h after transfection. Scale bar = 50 μm. Error bars show SEMs. Asterisks represent a statistically significant difference from the control (**P< 0.01; ****P < 0.0001).

TGFβR3 overexpression reverses cell invasion promoted by miR-181a

Next, we determined whether TGFβR3 overexpression would negate miR-181a-mediated effects on keratinocyte invasiveness. HaCaT cells were stably transfected with lenti-miR-181a (OE) or lenti-miR-00 (control) and subjected to Boyden chamber assays, using Matrigel-coated membranes. Our results showed that miR-181a overexpression significantly enhanced HaCaT invasion (2.6 ± 0.7 - fold, P<0.01, unpaired t-test). We then assessed whether miR-181a-induced enhancement of cell invasiveness could be blunted by TGFβR3 overexpression. In these rescue experiments, we showed that overexpression of the TGFβR3 can essentially rescue the increased invasion nearly completely (1.3 ± 0.4-fold, P=0.40, unpaired t-test) (Figure 4d, e). Note that the individual effect of TGFβR3 is minimal as the graph is normalized to the same control (Figure 4d, first and third columns). These data show that for the key phenotypes of miR-181a upregulation, including suppression of apoptosis and increased invasiveness can be attributed in large part to the downregulation of TGFβR3.

miR-181a promotes upregulation of EMT markers through TGFβ signaling

By transfecting immortalized keratinocytes (Ker-CT) with miR-181a, we were able to test short term effects of TGFβ2, the most relevant agonist for examining TGFβR3 binding and signaling (Henen et al., 2019, Villarreal et al., 2016), with and without miR-181a OE. Multiple groups have reported that TGFβ2 signaling proceeds through SMAD3 most prominently (You et al., 2007) and that TGFβR3 is a negative regulator of this signaling (Tazat et al., 2015).

Accordingly, we found robust induction of phospho-SMAD3 within 1 hour following TGFβ2 administration, which persists beyond 6 hours only in miR-181a-overexpressing keratinocytes, however this is lost after 24 induction (Figure 5a). In this context, it is clear that the down regulation of TGFβR3 enables signaling from TGFβ2 to be extended. By 24 hours, strong induction of EMT markers SNAIL (SNAI1) and SLUG (SNAI2), and upregulation of CDH1 (E-Cadherin) (Figure 5b). SNAIL and SLUG are highly relevant for EMT in keratinocytes and are upregulated in UV-driven and chemical skin carcinogenesis (De Craene et al., 2014, Hudson et al., 2007, von Maltzan et al., 2016). The increased CDH1 (E-cadherin) is likely to be due to a partial EMT phenotype previously reported in several contexts (Hahn et al., 2016, Yan et al., 2015) and because it has also been described as a SLUG target gene (Hudson et al., 2009). Based on these results, we then tested the direct contribution of TGFβR3 to this augmentation of TGFβ signaling. To this end, we additionally tested the effect of TGFβ2 on miR-181a overexpressing cells In contrast, we have been able to observe a reversion of SMAD3 phosphorylation and some EMT markers (SNAIL and CDH1) when the keratinocytes overexpressed TGFβR3 together when mir-181 overexpression (Figure 3c).

Figure 5. miR-181a augments TGFβ2 signaling and strongly enhances EMT protein expression.

Figure 5.

Open in a new tab

a, TGFβ2 signaling is enhanced by miR-181a expression. Phospho-SMAD3 is strongly induced by TGFβ2 exposure and persists beyond 6 hours (up to 24 hours) only in the context of miR-181a expression. Phospho-SMAD2 is not upregulated. b, The transcription factors SNAIL and SLUG, and E-cadherin (CDH1), highly relevant for keratinocyte EMT, are upregulated significantly only in the miR-181a-expressing keratinocytes upon stimulation with TGFβ2. Neither TGFβ2 alone nor miR-181a overexpression alone changes the expression of these factors. c, TGFβ2 signaling is enhanced by miR-181a expression and TFGβR3 co-expression reverses this. Phospho-SMAD3 is strongly induced by TGFβ2 exposure in the context of miR-181a over-expression; however this is reversed with TFGβR3 co-expression in Ker-CT cells. The transcription factors SNAIL and E-cadherin (CDH1), highly relevant for keratinocyte EMT, are upregulated significantly in the miR-181a-expressing keratinocytes upon stimulation with TGFβ2, but these effects were reversed when TFGβR3 was co-expressed. Interestingly, these effects were not observed for SLUG or SMAD2.

miR-181a inhibition suppresses tumor growth in a xenograft SCC mouse model

To assess whether suppressing miR-181a expression would affect tumor growth in-vivo, we transiently transfected COLO16 cells with miR-181a inhibitor or control inhibitor, and then injected them into flanks of NOD CRISPR Prkdc Il2r gamma (NCG) mice (Figure 6a). We found that tumors derived from COLO16 cells transfected with miR-181a inhibitor grew substantially more slowly, as compared to the negative control (Figure 6a–b). By day 11, the average volume of tumors (n=16; 8 mice each condition) derived from cells transfected with miR-181a inhibitor was 35.2% ± 13.3% smaller (P<0.05, unpaired t-test) smaller than those derived from the cells transfected with the control inhibitor (Figure 6b–d). Most importantly, we observed that silencing TGFβR3 in COLO16 cell lines reverts the effect of miR-181a inhibition, as simultaneous suppression of miR-181a and TGFβR3 expression resulted in reversion of tumor growth to control levels (P=0.78, unpaired t-test; Figure 6b–d). To ensure TGFβR3 levels were modulated as expected in the tumors, TGFβR3 expression was measured by qRT-PCR in tumors at day 11, showing increased expression by 2.1 ± 0.4-fold (P < 0.05, unpaired t-test) in tumors derived from cells transfected with miR-181a inhibitor, as compared to control inhibitors (Figure 6e). Efficacy of the shRNA-mediated suppression of TGFβR3 expression was confirmed (Figure 6e) in both cells treated alone with shRNA-TGFβR3 (decreased 4.3 ± 0.1-fold; P<0.001, unpaired t-test) and in combination with miR-181a inhibitor (decreased 2.4 ± 0.1-fold; P<0.001, unpaired t-test). These results collectively show that miR-181a plays an important role in cuSCC progression in-vivo through the regulation of TGFβR3.

Figure 6. Inhibition of miR-181a function suppresses tumor progression in-vivo.

Figure 6.

Open in a new tab

a, Timeline of COLO16 SCC cell line transfection and xenograft inoculation. b, Comparison of representative gross tumor xenografts at sacrifice shows substantial decrease in. c-d, Xenograft tumor growth was significantly reduced (P < 0.05, paired Student’s t-test, n=16 (8 mice) each condition) in tumors derived from cells transfected with miR-181a inhibitor than the control inhibitor. At day 11, the average volume of tumors derived from cells transfected with miR-181a inhibitor was 35.2% ± 13.3% smaller (P<0.05, unpaired t-test) than those derived from the cells transfected with the control inhibitor (Figure 6b–d). Simultaneous silencing of TGFβR3 in COLO16 cell lines resulted in reversion of tumor growth to control levels (P=0.78, unpaired t-test; Figure 6a–b). e, TGFβR3 expression was measured by qRT-PCR in tumors at day 11, showing increased expression by 2.1 ± 0.4-fold (P < 0.05, unpaired t-test) in tumors derived from cells transfected with miR-181a inhibitor, as compared to control inhibitors. These results show that miR-181a plays an important role in cuSCC progression in-vivo. TGFBR3 expression increased significantly (P < 0.05, two-way ANOVA) in tumors derived from cells transfected with miR-181a inhibitor, compared to control inhibitor.

Discussion

Our data show that miR-181a is significantly overexpressed in a subset of cuSCC, as compared to normal skin. The overexpression of miR-181a significantly suppressed UV-induced apoptosis (Figure 2a–c), enhanced anchorage independent survival in keratinocytes (Figure 1d), and significantly increased invasion (Figure 2), activating TGFβ2 signaling through SMAD3 and upregulating several EMT markers (Figure 5). Importantly, we showed that these phenotypes are due to the direct regulation of TGFβR3 by miR-181a (Figure 4), which is conferred by two sites within the 3’ UTR (Figure 3c–f).

TGFβR3 is downregulated in various human cancers, including in HNSCC(Meng et al., 2011). In early-stage tumors, it has a suppressive role, serving as a homeostatic regulator of the TGF-β pathway. On the other hand, in late-stage tumors it increases TGF-β expression, promoting tumor progression (Bernabeu et al., 2009).

Our data show that TGFβR3 expression is decreased in cuSCC as compared to normal skin (Figure 3a) and that TGFβR3 is a direct and functional target of miR-181. Furthermore, we show that TGFβR3, as directly regulated by miR-181a (Figure 3c–f), opposes TGF- β signaling (Figure 3b), thus enhancing stress-induced apoptosis and suppressing invasiveness in cuSCC with increased miR-181a expression. Importantly, sustained pSMAD3 levels are observed following miR-181a expression in response to TGFβ2 (Figure 5a). This is accompanied by upregulation of key EMT transcription factors SNAIL and SLUG (Figure 5b). Interestingly, pSMAD3, CDH1, and SNAIL upregulation is reversed by TGFβR3 co-expression (Figure 5c). Finally, inhibition of miR-181a expression suppresses the progression of tumors in-vivo, indicating the relevance of these findings in cancer. In this context, the simultaneous suppression of TGFβR3 expression results in reversal of this effect and growth of tumor cells back to control levels (Figure 6b–d).

TGFβ signaling in cancer is enormously complex and while miR-181a has been implicated in regulating TGFβR1 in cancer (Ge et al., 2019), there has not been a previous accounting for how miR-181a regulates TGFβR3 in driving multiple tumorigenic phenotypes in cells and in-vivo. Collectively, our data definitively demonstrate that miR-181a drives several oncogenic functions through direct targeting of TGFβR3, integrating multiple phenotypes such as resistance to apoptosis, motility, and induction of EMT markers.

MATERIALS AND METHODS

Cell Culture

Cell lines were maintained in DMEM/F12 with 10% FBS, RM+ supplement. SRB1, SRB12, and COLO16 were a gift from Jeffrey Myers, MD, PhD (MDACC). SCC IC1, SCCT1, SCCT2, SCCT3, SCCT8, SCRDEB2, SCRDEB3 and SCRDEB4 were provided by Andrew South, PhD (Thomas Jefferson University). HaCaT cells, an immortalized human keratinocyte line cell was obtained from Norbert Fusenig (German Cancer Research Center). Normal Human Epidermal Keratinocytes (NHEK) were purchased (Lonza, Cat. # 192906 ) and maintained in KGM-Gold Keratinocyte Growth Medium (Lonza, Cat. # 1922151). The Ker-CT line was obtained from TACC (CRL-4048) also maintained in KGM-Gold. All cell lines were STR profiled to confirm distinct identities and were maintained mycoplasma-free.

Quantitative RT-PCR

miRNAs and mRNAs were detected by Taqman quantitative reverse transcription PCR (qRT-PCR) method. Expression of miRNAs were normalized by RNU6B. Expression of mRNAs were normalized to GAPDH. All primers were purchased from Thermo Fisher, USA (Cat. # 4427975, Assay ID 000480).

Flow cytometry

TMRE (Invitrogen) was used as a measure of mitochondrial membrane potential, Annexin V-FITC or Annexin V-APC (Invitrogen) as a probe for apoptosis, and Cytox Blue (Invitrogen) as an indicator for dead cells. At 6, 24, or 48 hours post-irradiation, both floating and adherent cells were collected and stained with TMRE, Annexin V and Cytox Blue. Data were collected and analyzed using a flow cytometer (FACScalibur, Becton Dickinson) and FlowJo Software (Tree Star). Data were analyzed and charts were plotted using GraphPad Prism software.

Overexpression and inhibition of miR-181a in cuSCC cells and keratinocytes

Hsa-miR-181–5p mimic (3nM), inhibitor (40nM) and their corresponding mirVana mimic and inhibitor negative control (Ambion) were transfected into the cells using Lipofectamine RNAiMAX in OptiMEM I (Invitrogen). Transient overexpression in the Ker-CT line was performed using jetPRIME 50 nM (Polyplus). Total RNAs were extracted by Trizol and PureLink RNA Mini Kit (Thermo Fisher). Expression of miR-181a was detected to confirm the transfection. Lentivirus-based GFP-tagged vectors for hsa-miR-181a (lenti-miR-181a OE) or scramble control (lenti-miR-00 control) were purchased from System Biosciences (SBI) lenti-miRNA vector bank. Transduction efficiency was assessed by GFP intensity and Taqman qRT-PCR.

Overexpression and depletion of TGFβR3

To overexpress TGFβR3, we purchased pDONR223-TGFΒR3 (Addgene, plasmid # 23478) donor vector which contained TGFβR3 ORF. Then, the TGFβR3 ORF was cloned into pcDNA3.1/V5 expressing vector using Gateway technique. All vector sequences were validated by Sanger sequencing. TGFβR3 expression was validated by western blot.

To deplete TGFβR3, we use GIPZ Human TGFβR3 shRNA (GE Dharmacon, Clone ID V3LHS_352448 and V3LHS_352450). GIPZ non-silencing lentiviral shRNA (GE Open Biosystems, Cat. # RHS4348) was used as control. The shRNA lentivirus was prepared with 293T cells and psPAX2/pVSV.G packaging plasmids, and then was transduced to target cells. Following transduction, cells were puromycin-selected and sorted to obtain cells with high-level suppression TGFβR3 suppression were validated by Taqman qRT-PCR.

Invasion Assay

Cells were starved 24 hours prior to the assay, then 3.5 × 105 cells were suspended in serum-free media and seeded on a 8μM pore size control insert (Corning, Cat. # 08–774-162) or the insert coated with matrigel (Corning, Cat # 08–774-122). Complete media contain 30% FBS was added to the lower compartments as a chemo-attractant for cells. Thereafter, cells were allowed to move for 48 hours. Cells remaining on the upper side of the membrane were removed. Those that invaded to the bottom side of the membrane were fixed and stained with Diff-Quik Stain Set (Siemens B4132–1A). The membranes were air-dried and mounted for photography. Cells from ten random fields were counted.

miRNA in situ hybridization (ISH)

In situ hybridization was conducted by the MD Anderson Center for RNA interference and non-coding RNAs. Skin squamous cell cancer tissue arrays (Cat # SK 801b and # SK 802a) were purchased from US Biomax, Inc. (Rockville, MD).

Plasmid construct and luciferase reporter assay

The pEZX-MT05 expression vector harboring the human TGFβR3 3’ UTR cDNA (GeneCopoeia, Cat. # HmiT066530) was used as DNA template. Site-directed mutagenesis was carried out using the QuickChange II XL Site-Directed Mutagenesis Kit (Catalog #200521, Agilent Technologies), following the Manufacturer’s instruction. The mutagenic primers were synthesized by IDT and the mutations verified by DNA Sanger sequencing.

HaCaT cells were transfected with wild-type and mutant reporter vectors along with scrambled control. 72 hours after transfection, the cell culture medium was collected for the assay. The luciferase reporter assays were performed according to the manufacturer’s guidelines for the Secrete Pair Dual Luminescence Assay Kit. Luciferase activity was normalized to secreted alkaline phosphatase.

Soft agar and colony formation assays

Following plating of bottom agar (0.6% Bacto Agar) with media we plated 2500 to 10,000 cells per well and they were embedded in top agar (0.3%) and plated in 24-well plates. Control or lenti-miR-overexpression media was replaced every 48 hr for 4 to 6 weeks. Cell colonies were stained with 1% crystal violet, imaged and quantified.

Protein extraction and Western blot analysis

Protein lysates were prepared in RIPA buffer (Sigma, Cat # R0278) supplemented with Halt Protease & Phosphatase Inhibitor Cocktail (Thermo Scientific, Cat #1861281). Primary antibodies were obtained from Cell Signaling and Thermo Scientific. Westerns were imaged using the Licor Odyssey CLx system.

Animals and in vivo model of human cutaneous squamous cell carcinoma

NOD CRISPR Prkdc Il2r gamma (NCG) mice (6 to 8 weeks old; Charles River) were housed and monitored in our Animal Research Facility. All experimental procedures and protocols were approved under IACUC 6437 at Moffitt Cancer Center and University of South Florida. RDEB2 cells were transfected with control inhibitor and miR-181a inhibitor. 24 hours post-transfection, cells were trypsinized for xenografting. Mice were subcutaneously inoculated with 2 × 106 SCC COLO16 cells. Once tumors become palpable, measurement was taken every other day. Mice were sacrificed 11 days after inoculation.

Statistical analyses

All biological experiments were repeated at least three times. Numerical data were analyzed using a one-way analysis of variation. The statistical significance between treatments was assessed by one-tailed Student’s t-tests or Paired Wilcoxon tests.

Data Availability

No datasets were generated or analyzed during the current study.

Supplementary Material

1
2
3
4
5

ACKNOWLEDGEMENTS

This work was supported by AACR Landon Foundation INNOVATOR Award in Cancer Prevention Research, T. Boone Pickens Endowment for Early Prevention of Cancer, DXB Biosciences Research Fund, and institutional funds (MDACC), NCI R01CA194062, NCI R01CA194617CA (KYT), and NCI 7R35CA197452 (ERF). VC acknowledges support from the University of Texas CCTS TL1 award. This work has been supported in part by the Analytic Microscopy Core at the H. Lee Moffitt Cancer Center & Research Institute, a comprehensive cancer center designated by the National Cancer Institute NIH P30CA076292.

Footnotes

CONFLICTS OF INTEREST

The authors declare no conflict of interest.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

  1. Agarwal V, Bell GW, Nam JW, Bartel DP. Predicting effective microRNA target sites in mammalian mRNAs. Elife 2015;4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bernabeu C, Lopez-Novoa JM, Quintanilla M. The emerging role of TGF-beta superfamily coreceptors in cancer. Biochim Biophys Acta 2009;1792(10):954–73. [DOI] [PubMed] [Google Scholar]
  3. Bisso A, Faleschini M, Zampa F, Capaci V, De Santa J, Santarpia L, et al. Oncogenic miR-181a/b affect the DNA damage response in aggressive breast cancer. Cell Cycle 2013;12(11):1679–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Brockhausen J, Tay SS, Grzelak CA, Bertolino P, Bowen DG, d’Avigdor WM, et al. miR-181a mediates TGF-beta-induced hepatocyte EMT and is dysregulated in cirrhosis and hepatocellular cancer. Liver international : official journal of the International Association for the Study of the Liver 2015;35(1):240–53. [DOI] [PubMed] [Google Scholar]
  5. Canman CE, Lim DS. The role of ATM in DNA damage responses and cancer. Oncogene 1998;17(25):3301–8. [DOI] [PubMed] [Google Scholar]
  6. Chen G, Zhu W, Shi D, Lv L, Zhang C, Liu P, et al. MicroRNA-181a sensitizes human malignant glioma U87MG cells to radiation by targeting Bcl-2. Oncol Rep 2010;23(4):997–1003. [DOI] [PubMed] [Google Scholar]
  7. Chitsazzadeh V, Coarfa C, Drummond JA, Nguyen T, Joseph A, Chilukuri S, et al. Cross-species identification of genomic drivers of squamous cell carcinoma development across preneoplastic intermediates. Nat Commun 2016;7:12601. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. De Craene B, Denecker G, Vermassen P, Taminau J, Mauch C, Derore A, et al. Epidermal Snail expression drives skin cancer initiation and progression through enhanced cytoprotection, epidermal stem/progenitor cell expansion and enhanced metastatic potential. Cell Death Differ 2014;21(2):310–20. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Enright AJ, John B, Gaul U, Tuschl T, Sander C, Marks DS. MicroRNA targets in Drosophila. Genome Biol 2003;5(1):R1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Ge S, Zhang H, Deng T, Sun W, Ning T, Fan Q, et al. MiR-181a, a new regulator of TGF-beta signaling, can promote cell migration and proliferation in gastric cancer. Invest New Drugs 2019;37(5):923–34. [DOI] [PubMed] [Google Scholar]
  11. Griffiths-Jones S, Saini HK, van Dongen S, Enright AJ. miRBase: tools for microRNA genomics. Nucleic Acids Res 2008;36(Database issue):D154–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hahn JM, McFarland KL, Combs KA, Supp DM. Partial epithelial-mesenchymal transition in keloid scars: regulation of keloid keratinocyte gene expression by transforming growth factor-beta1. Burns Trauma 2016;4(1):30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Henen MA, Mahlawat P, Zwieb C, Kodali RB, Hinck CS, Hanna RD, et al. TGF-beta2 uses the concave surface of its extended finger region to bind betaglycan’s ZP domain via three residues specific to TGF-beta and inhibin-alpha. J Biol Chem 2019;294(9):3065–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hudson LG, Choi C, Newkirk KM, Parkhani J, Cooper KL, Lu P, et al. Ultraviolet radiation stimulates expression of Snail family transcription factors in keratinocytes. Mol Carcinog 2007;46(4):257–68. [DOI] [PubMed] [Google Scholar]
  15. Hudson LG, Newkirk KM, Chandler HL, Choi C, Fossey SL, Parent AE, et al. Cutaneous wound reepithelialization is compromised in mice lacking functional Slug (Snai2). J Dermatol Sci 2009;56(1):19–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Jamali Z, Asl Aminabadi N, Attaran R, Pournagiazar F, Ghertasi Oskouei S, Ahmadpour F. MicroRNAs as prognostic molecular signatures in human head and neck squamous cell carcinoma: a systematic review and meta-analysis. Oral Oncol 2015;51(4):321–31. [DOI] [PubMed] [Google Scholar]
  17. Khanna KK, Chenevix-Trench G. ATM and genome maintenance: defining its role in breast cancer susceptibility. J Mammary Gland Biol Neoplasia 2004;9(3):247–62. [DOI] [PubMed] [Google Scholar]
  18. Krek A, Grun D, Poy MN, Wolf R, Rosenberg L, Epstein EJ, et al. Combinatorial microRNA target predictions. Nat Genet 2005;37(5):495–500. [DOI] [PubMed] [Google Scholar]
  19. Li YY, Hanna GJ, Laga AC, Haddad RI, Lorch JH, Hammerman PS. Genomic analysis of metastatic cutaneous squamous cell carcinoma. Clin Cancer Res 2015;21(6):1447–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Meng W, Xia Q, Wu L, Chen S, He X, Zhang L, et al. Downregulation of TGF-beta receptor types II and III in oral squamous cell carcinoma and oral carcinoma-associated fibroblasts. BMC Cancer 2011;11:88. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Nishida N, Nagahara M, Sato T, Mimori K, Sudo T, Tanaka F, et al. Microarray analysis of colorectal cancer stromal tissue reveals upregulation of two oncogenic miRNA clusters. Clin Cancer Res 2012;18(11):3054–70. [DOI] [PubMed] [Google Scholar]
  22. Parikh SA, Patel VA, Ratner D. Advances in the management of cutaneous squamous cell carcinoma. F1000prime reports 2014;6:70. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Shin KH, Bae SD, Hong HS, Kim RH, Kang MK, Park NH. miR-181a shows tumor suppressive effect against oral squamous cell carcinoma cells by downregulating K-ras. Biochem Biophys Res Commun 2011;404(4):896–902. [DOI] [PubMed] [Google Scholar]
  24. Smith J, Tho LM, Xu N, Gillespie DA. The ATM-Chk2 and ATR-Chk1 pathways in DNA damage signaling and cancer. Adv Cancer Res 2010;108:73–112. [DOI] [PubMed] [Google Scholar]
  25. Taylor MA, Sossey-Alaoui K, Thompson CL, Danielpour D, Schiemann WP. TGF-beta upregulates miR-181a expression to promote breast cancer metastasis. J Clin Invest 2013;123(1):150–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Tazat K, Hector-Greene M, Blobe GC, Henis YI. TbetaRIII independently binds type I and type II TGF-beta receptors to inhibit TGF-beta signaling. Mol Biol Cell 2015;26(19):3535–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Villarreal MM, Kim SK, Barron L, Kodali R, Baardsnes J, Hinck CS, et al. Binding Properties of the Transforming Growth Factor-beta Coreceptor Betaglycan: Proposed Mechanism for Potentiation of Receptor Complex Assembly and Signaling. Biochemistry 2016;55(49):6880–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Vin H, Ojeda SS, Ching G, Leung ML, Chitsazzadeh V, Dwyer DW, et al. BRAF inhibitors suppress apoptosis through off-target inhibition of JNK signaling. Elife 2013;2:e00969. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. von Maltzan K, Li Y, Rundhaug JE, Hudson LG, Fischer SM, Kusewitt DF. Role of the Slug Transcription Factor in Chemically-Induced Skin Cancer. J Clin Med 2016;5(2). [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Wang Y, Yu Y, Tsuyada A, Ren X, Wu X, Stubblefield K, et al. Transforming growth factor-beta regulates the sphere-initiating stem cell-like feature in breast cancer through miRNA-181 and ATM. Oncogene 2011;30(12):1470–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Weng H, Lal K, Yang FF, Chen J. The pathological role and prognostic impact of miR-181 in acute myeloid leukemia. Cancer Genet 2015;208(5):225–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Yan L, Cao R, Wang L, Liu Y, Pan B, Yin Y, et al. Epithelial-mesenchymal transition in keloid tissues and TGF-beta1-induced hair follicle outer root sheath keratinocytes. Wound Repair Regen 2015;23(4):601–10. [DOI] [PubMed] [Google Scholar]
  33. Yang Z, Wan X, Gu Z, Zhang H, Yang X, He L, et al. Evolution of the mir-181 microRNA family. Comput Biol Med 2014;52:82–7. [DOI] [PubMed] [Google Scholar]
  34. You HJ, Bruinsma MW, How T, Ostrander JH, Blobe GC. The type III TGF-beta receptor signals through both Smad3 and the p38 MAP kinase pathways to contribute to inhibition of cell proliferation. Carcinogenesis 2007;28(12):2491–500. [DOI] [PubMed] [Google Scholar]
  35. Zhang X, Nie Y, Du Y, Cao J, Shen B, Li Y. MicroRNA-181a promotes gastric cancer by negatively regulating tumor suppressor KLF6. Tumour Biol 2012;33(5):1589–97. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1762867-supplement-1.tiff (3.2MB, tiff)
2
NIHMS1762867-supplement-2.tiff (3.2MB, tiff)
3
NIHMS1762867-supplement-3.tiff (3.2MB, tiff)
4
NIHMS1762867-supplement-4.docx (18.9KB, docx)
5
NIHMS1762867-supplement-5.xlsx (2.5MB, xlsx)

Data Availability Statement

No datasets were generated or analyzed during the current study.

RESOURCES