- Email: [email protected]
MicroRNA-dependent regulation of cKit in cutaneous melanoma
Biochemical and Biophysical Research Communications 379 (2009) 790–794
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
MicroRNA-dependent regulation of cKit in cutaneous melanoma O. Igoucheva, V. Alexeev * Department of Dermatology and Cutaneous Biology, Jefferson Medical College, Thomas Jefferson University, 233 S. 10th Street, BLSB, Suite 326, Philadelphia, PA 19107, USA
a r t i c l e
i n f o
Article history: Received 18 December 2008 Available online 4 January 2009
Keywords: cKit MicroRNA Cutaneous melanoma Melanocytes mir-221
a b s t r a c t Loss of cKit receptor in cutaneous melanomas was attributed to the down-regulation of AP2 transcription factor. Our analysis of 27 melanoma cell lines showed no correlation between AP2 and c-kit expression. Suggesting a post-transcriptional mechanism of cKit down-modulation, we performed genome-wide microRNA (miRNA) expression profiling and found that several miRNA species are commonly up-regulated in melanomas. Among them was mir-221, which can directly interact with c-kit 30 UTR and inhibit cKit protein translation. Observed inverse correlation of the c-kit and mir-221 expression in various melanocytic cells pointed to its involvement in regulation of cKit in melanoma. Moreover, a series of functional assays demonstrated that mir-221 could directly inhibit cKit, p27Kip1 and, possibly, other pivotal proteins in melanoma. Collectively, the studies presented here indicate that mir-221 could be a novel therapeutic target for the treatment of cutaneous melanoma. They also suggest that regulation of expression and functional activity of identified up-regulated miRNAs should be further studied in the context of malignant melanoma. Ó 2009 Elsevier Inc. All rights reserved.
cKit tyrosine kinase receptor for the stem cell factor (SCF), encoded by the proto-oncogene c-kit, belongs to the PDGF family of the receptor tyrosine kinases (RTK). The intracellular signaling from cKit RTK plays a critical role in the development of a variety of mammalian cell types including hematopoietic progenitor cells, mast cells, primordial germ cells, intestinal cells of Cajal, and melanocytes. It is well established that cKit RTK transduce intracellular signals that lead to various, sometimes conflicting cellular responses. It has been demonstrated that this RTK can transmit pro-proliferation [1], anti-proliferation [2], survival, pro-migration [3], and pro-apoptotic [4] signals. In normal melanocytes, SCFdependent cKit-mediated signaling supports proliferation, migration and differentiation of the cells. Constitutive activation of cKit RTK alone does not induce tumorigenic transformation of the melanocytes in vitro and in vivo [5]. However, in certain subtypes of melanomas characterized by activating mutations and/or increased copy number of c-kit gene [6] with rare occurrence of BRAF mutations [7], cKit RTK activity enhances tumorigenic potential of transformed melanocytes. Studies on pharmacological inhibition of cKit RTK in certain subpopulations of cKit-positive melanomas with Imatinib showed increased apoptosis and G1 phase cell cycle arrest associated with the inhibition of phosphoERK and elevated expression of p27Kip1 tumor suppressor [8]. In contrast to normal melanocytes and cKit-expressing melanomas, lesions of cutaneous melanoma occurring on intermittent sun exposed skin are often characterized by the loss of cKit receptor * Corresponding author. Fax: +1 215 503 5788. E-mail address: [email protected] (V. Alexeev). 0006-291X/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2008.12.152
expression [6] and high frequency of activating mutations in the BRAF [9]. Earlier studies on cKit-negative BRAF mutant, A375 melanoma cells demonstrated that re-expression and restoration of cKit-mediated signaling in these cells induced cKit-dependent apoptosis and resulted in a loss of tumorigenic potential of melanoma [10]. Despite extensive studies of the cKit in melanocytes and melanomas, the molecular mechanisms that govern the expression of this RTK remain incompletely understood. Several recent studies demonstrated that a family of 19–25nucleotide-long RNA molecules termed as microRNA (miRNA) represents a new class of highly conserved non-coding RNAs that function as negative regulators of protein expression. In multiple cases, it was shown that deregulation of miRNA expression contributes to cancer development and facilitates the progression of multiple myeloma, prostate cancers, Burkitt lymphoma and other types of tumors. However, very limited and discrete information regarding miRNA expression and function in melanoma is currently available. As cKit RTK plays a critical role in a variety of mammalian cell types, several studies examined whether cKit protein expression is controlled by miRNA. Based on bioinformatics predictions, ckit mRNA can be targeted by at least 12 different miRNAs. Studies on erythropoietic CD34+ cord blood progenitor cell showed that supplementation of cells with highly homologous mir-221 and mir-222 caused impaired proliferation of the cells directly coupled with down-modulation of cKit protein [11]. Recent miRNA expression profiling in thyroid papillary carcinoma (TPC) also showed a distinguished up-regulation of these two miRNAs in TPC characterized by the loss of cKit [12]. Other studies showed that mir-221
O. Igoucheva, V. Alexeev / Biochemical and Biophysical Research Communications 379 (2009) 790–794
could also down-modulate the expression of the p27Kip1 tumor suppressor and promote proliferation of various tumor cells [13]. Very recently, Felicetti and co-workers identified the promyelocytic leukemia zinc finger (PLZF) transcription factor as a repressor of miR-221 and miR-222. Authors demonstrated that targeted PLZF silencing in cKitlow melanoma cell line unblocks miR-221 [14]. Collectively, these studies demonstrated that cKit protein expression can be regulated by miRNA, mir-221 in particular. Therefore, the goals of this study were to identify deregulated miRNAs in cutaneous melanomas and determine whether mir-221 contributes to the progression of this malignancy via inhibition of cKit.
Materials and methods Cell lines. WM and Lu melanoma cell lines were provided by Dr. M. Herlyn (Wistar Institute, Philadelphia, PA). Primary human melanocytes were provided by Drs. R.E. Boissy and Z. Abdel-Malek (University of Cincinnati, Cincinnati, OH). Melanoma cells were cultured as described at www.wistar.org/herlyn/resources_intro.htm. Primary melanocytes were cultured in melanocyte-specific media with TPA-free supplements (Invitrogen, Carlsbad, CA). DNA microarray. Analysis of AP2 a and c isoform expression in various melanocytic cells was done using Affymetrix gene expression HG-U95Av2 chips utilizing cervices provided by the Center for Translational Medicine at Thomas Jefferson University. Semi-quantitative RT-PCR analysis of mir-221 expression. Total RNA from various melanocytic cells was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA). Total RNA was treated with DNase-1 and reverse transcribed using SuperScript II system (Invitrogen, Carlsbad, CA). AP2 a and c, c-kit and mir-221 were amplified using gene-specific primers: AP2 a: 50 -GCCCCGTGTCCCTGTCCAA-30 and 50 -TGAGGAGCGAGAGGCGACC-30 AP2 c: 50 -GGCCCAGCAACT GTGTAAAGA-30 and 50 -GCAGTTCTGTATGTTCGTCTCCAA-30 c-kit: 50 -GATCATGCAGAAGCTGCACT-30 and 50 -AAAATCCCATAGGACCAGA C-30 mir-221: 50 -TGAATGCAGTAGGCAGTTGTG-30 and 50 -GGGGTAG CATTGGTGAGACA-30 GAPDH: 50 -AACGTGTCAGTGGTGGACCT-30 and 50 -TGCTGTAGCCAAATTCGTTG-30 . Amplification of GAPDH was used as a control for equal RT-PCR loading. Nucleofection of melanoma cells. Optimization of anti-miRTM RNA oligonucleotides delivery into cultured melanoma cells was done using Cy3-labeled anti-miRTM RNA oligonucleotides (Ambion Inc., Applied Biosystems, Austin, TX). Using Amaxa nucleofection system II (Lonza, Gaithersburg, MD) and 50 nM of Cy3-labeled oligonucleotides, several nucleofection conditions were tested. The highest efficacy of oligonucleotide delivery (up to 90%) and viability of the cells (up to 80%) was achieved when oligonucleotides were nucleofected using NHEM Neo nucleofection kit (Lonza) and U-024 program (data not shown). These conditions were also optimal for the delivery of plasmids DNA into cultured melanoma cells. Generation of miRNA and cKit expression vectors. A genomic sequence containing mir-221 was amplified by PCR from the total human genomic DNA using direct: 50 -CCCAGCATTTCTGACTG-30 and reverse: 50 -ATTTCACCCAAATGGTCT-30 primers. Amplified fragments were ligated into pEF6-TOPO expression vector (Invitrogen, Carlsbad, CA) containing human EF1 promoter and blasticidin resistance gene (Bsd) for rapid selection of the transduced cells. The integrity of the promoter and gene sequences were verified by direct DNA sequencing. The resultant plasmid was designated as pEF1-mir-221. Human c-kit cDNA with or without 30 UTR was obtained by RTPCR reaction using total RNA isolated from primary human melanocytes and gene-specific primers: cDNA: 50 -ATGAGAGGCGCTCGCGG CGCC-30 and 50 -TCAGACATCGTCGTGCACAAGCA-30 cDNAUTR: 50 -AT GAGAGGCGCTCGCGGCGCC-30 and 50 -TCCAATTGTTGACAGTTCTGAA G-30 . Resultant sequences were ligated into pEF6-TOPO vector.
791
Western blot, immunofluorescent and TUNEL assays: For the Western blot analysis, cells were lysed in a buffer containing 20 mM Tris–HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton X100, 0.5% NP-40, 1 mM PMSF, and protease inhibitors (Pierce, Rockford, IL). For the analysis of cKit phosphorylation, a cocktail of phosphatase inhibitors (Pierce, Rockford, IL) was added to lysis buffer. cKit was detected on blots by protein-specific antibodies (Ab81). Phospho-cKit was detected by monoclonal antibodies (Tyr823) specific to phosphorylated cKit receptor. PARP was detected by monoclonal antibodies (19F4) specific to the cleaved form of the protein. All antibodies were purchased from Cell Signaling Technologies (Danvers, MA) and used in dilution 1:1000. For the direct immunofluorescent detection of cKit in melanoma cells, AlexaFluor488-lebeled monoclonal (Ab81) antibodies were used in dilution 1:50 (Cell Signaling Technologies, Danvers, MA). Ki-67 and p27Kip1 were detected by indirect immunofluorescence using protein-specific antibodies (Millipore, Bedford, MA) as devised by the manufacturer. Apoptosis in 1205Lu melanoma cells was detected using TUNEL assay (In Situ Cell Death detection kit, Roche Diagnostics, Mannheim, Germany) as devised by the manufacturer.
Results and discussion AP2 expression in melanocytic cells In human melanocytic cells, AP2 transcription factor was shown to be directly involved in regulation of c-kit gene transcription. As only limited number of melanomas were previously tested for the expression of the AP2 a, here we examined the expression of structurally and functionally similar a and c isoforms of the AP2 in six lines of normal, two lines of vitiligo melanocytes, and 27 lines of melanoma cell isolated from tumors on different stages of tumor progression (Fig. 1A). Based on intensity of hybridization signals, our analysis confirmed earlier findings showing general down-regulation of AP2 a expression in primary melanoma cells. However, in contrast to previous findings, we did not observe any direct correlation between down-regulation of AP2 a and progression of the malignancy from radial growth phase (RGP) through vertical growth phase (VGP) to metastatic (MM) form. Both high and low levels of the AP2 a were found in melanomas, isolated from GRP, VGP and MM tumors. In almost all melanoma cells, lower expression of AP2 a was accompanied by the elevated expression AP2 c. Considering structural and functional similarities between these two isoforms, it is likely that in melanomas, similarly to germ cell tumors and gliomas [15], up-regulated AP2 c compensates for the down-regulated a-isoform. Based on our findings, average combined expression of both AP2 isoforms in malignant melanoma was only 1.4 times lower than that in normal adult melanocytes (Fig. 1B). Therefore, down-regulation or complete lack of cKit in the majority of cutaneous melanomas cannot be attributed exclusively to the down-regulation of the AP2 a. These observations directly pointed to the existence of an alternative mechanism(s) of the cKit regulation in melanoma. Analysis of miRNA expression in melanoma Suggesting post-transcriptional mechanism of cKit regulation, we hypothesized that microRNA may inhibit cKit protein expression. Having an original setup for genome-wide miRNA gene expression profiling at Thomas Jefferson University [16], we analyzed the expression of more than 300 miRNAs in primary normal melanocytes and melanoma cell lines as described previously [16]. In contrast to previously published data showing down-regulation of many miRNA species in various cancers, we did not find any
792
O. Igoucheva, V. Alexeev / Biochemical and Biophysical Research Communications 379 (2009) 790–794
Fig. 1. AP2 a and c expression in melanocytic cell. Expression of AP2 a and c was examined in melanocytes (F—fetal, A—adult, V—vitiligo) and melanoma cells. Cell lines are positioned based on the expression of the AP2 a (from the highest to the lowest). Cell lines and other available information (stage of tumor progression) are indicated above the columns. (A) Variations of AP2 a (open columns) and AP2 c (solid columns) expression in melanocytic cells. (B) Combined expression of both a and c isoforms.
Table 1 miRNA expression in melanocytes and melanoma. miRNA
Location
Expression in melanocyte/melanoma
Some predicted or confirmed targets
Normal
RGP
VGP
MM
mir-23a mir-24-2 mir-25 mir-92
19p13.3 19p13.8 7q22 13q31
+ + ±
++ + ++ ++++
++++ +++ ++ ++++
++++ ++++ ++ ++++
mir-124 mir-146b mir-149
8p23.1 5q34 2q37.3
+++
+ +++ +++++
++++ +++ ++++
mir-213 mir-221 mir-324 mir-338
1q32.1 Xp11.3 17p13.1 17q25.3
+ ++ ++
++ +++ +++ +
++ ++++ +++ +
MAP4K4 (2), metastasis suppressor 1 (1) BCL2-like 11 (apoptosis facilitator) (2),integrin, alpha V (2), MAP2K4 (2), PI3K regulatory subunit 3 (1), heart and neural crest derivatives (HAND1) mRNA MITF (2), RAB27A (2). neuro-oncological ventral antigen 1 (2), periaxin (2) AP2 associated kinase 1 (1), cKit (1)? BCL2-like apoptosis facilitator (1) ??? cKit (1), p27Kip1 (2), Tyrosinase (1) ??? MAGE10 (2), cKit (1)
Predicted number of miRNA binding sites on 30 UTRs indicated in parenthesis. Not yet predicted targets are indicated with question marks (?).
significantly down-regulated miRNAs in melanomas, as compared to melanocytes. However, several miRNA species were up-regulated in RGP, VGP and MM tumors (Table 1). As many miRNA genes are located at regions of amplification or translocation, we compared miRNA loci to the known chromosomal abnormalities reported in melanoma. Our analysis showed that the loci of miRNA species highly expressed in melanoma (Table 1) did not map to the known amplification or translocation sites associated with this tumor. Out of all deregulated miRNAs, elevated expression of mir-221 (and homologous mir-222) was of particular interest. This miRNA was expressed in melanoma cells showing decreased levels of cKit transcripts (WM35, WM983A, WM164, and 1205Lu). Similar results were obtained when RNA samples isolated from additional melanoma cell lines were analyzed for the genome-wide miRNA expression using GenoSensor services (GenoSensor, Tempe, AZ). In addition to the already established profiles, this analysis demonstrated that mir-149 and mir-338 were up-regulated in cKit-negative melanoma cells isolated from tumor cells on earlier stages of tumor progression (RGP and RGP/VGP). Based on bioinformatics predictions, mir-221, mir-149, and mir-338 can form favorable duplexes with the 30 UTR of the c-kit mRNA with energies 20.9 kcal/mol, 23.4 kcal/mol, and 16.7 kcal/mol, respectively (see: http://microrna.sanger.ac.uk/). To examine mir-221 expression in larger number of samples, we established the conditions for miRNA detection using RT-PCR (see Materials and methods). For this analysis, we used additional samples of primary normal and vitiligo melanocytes, 17 cutaneous
melanoma cell lines, 1 BRAF normal melanoma line, and two samples of uveal melanomas. Same RNA samples were used for the analysis of AP2 a, AP2 c and c-kit expression. As shown in Fig. 2, AP2 a and AP2 c were equally expressed in all analyzed cells (with few exceptions). However, level of c-kit transcription varied from the highest in primary melanocytes to the lowest or absent in cutaneous melanomas. The level of mir-221 transcription varied as well (Fig. 2). Overall, this RT-PCR-based analysis clearly demonstrated
Fig. 2. Analysis of mir-221, AP2 and cKit expression in melanocytic cells. Cell line names shown above the photographs; analyzed genes shown on the right. Abbreviations: Cells were isolated from N—normal skin; Vt—vitiligo skin; P—primary melanoma lesion; LN—lymph node metastasis; LM—lung metastasis; R—RGP melanoma; V—VGP melanoma; R/V—RGP/VGP melanoma; LvM—liver metastasis of primary uveal melanoma; Br—BRAF normal melanoma.
O. Igoucheva, V. Alexeev / Biochemical and Biophysical Research Communications 379 (2009) 790–794
that levels of c-kit transcription do not correlate with the AP2 expression, but inversely correlate with the expression of mir-221. miRNA inhibition and cKit expression Further, we tested whether enforced, nucleic acid-based inhibition of mir-221 could lead to the restoration of cKit expression and cKit-mediated signaling. Therefore, cKit-negative, mir-221-positive 1205Lu melanoma cells were nucleofected with control or mir-221specific anti-miRTM inhibiting oligonucleotides (Ambion) under optimal conditions (see Materials and methods). Nucleofected cells were cultured for 48 h with or without stem cell factor (SCF). Cells were then examined for the expression of cKit and phosphorylation. Oligonucleotide-based inhibition of mir-221 resulted in a rapid restoration of the cKit protein expression and autophosphorylation of the receptor when cells were exposed to SCF (Fig. 3A). Moreover, cells with restored cKit expression and signaling showed an increase of SCF-dependent apoptosis as compared with cells cultured in the absence of SCF or control cells as deduced from the presence of cleaved PARP protein, the hallmark of apoptotic cell death (Fig. 3B). Induction of apoptosis was confirmed by TUNEL assay, which demonstrated increased apoptosis of cells treated with antimir-221 oligonucleotides and cultured in the presence of SCF (Fig. 3C). Collectively, these studies demonstrated that targeted inhibition of mir-221 led to the restoration of cKit expression and liganddependent signaling, which resulted in the induction of apoptosis of BRAF-mutant melanoma. These findings directly indicate that targeted inhibition of mir-221 could be a novel therapeutic approach for the treatment of cutaneous melanoma. miRNA-dependent inhibition of cKit To confirm cKit-inhibiting activity of mir-221, we generated mammalian expression vector coding for human mir-221 (see
Fig. 3. Western blot analysis of the cKit re-expression and induction of SCFdependent apoptosis. (A) cKit receptor expression and phosphorylation. Lanes: 1— 1205Lu cells; 2—1205Lu cells treated with control oligonucleotide (+SCF); 3— 1205Lu cells treated with anti-mir-221 RNA oligonucleotides (+SCF). (B) Detection of cleaved PARP in cells treated with anti-mir-221 RNA oligonucleotides; Lanes: 1— 1205Lu cells; 2—1205Lu cells treated with control anti-mir oligonucleotides (+SCF); 3—1205Lu cells treated with anti-mir-221 RNA oligonucleotides (SCF); 4—1205Lu cells treated with anti-mir-221 RNA oligonucleotides (+SCF). Arrows on the left point to the position of the molecular weight markers. Arrows on the right point to cKit and cleaved PARP. Antibodies indicated below the panels. (C) Detection of apoptosis by TUNEL assay. Treatment and culture conditions indicated on the panels. Apoptotic cells—green. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this paper.)
793
Materials and methods). Expression vector was nucleofected into cKithigh 451Lu melanoma cells under optimized conditions (see Materials and methods). Ninety-six hours later, cells were analyzed for the miRNA and cKit expression. As shown in Fig. 4, mir221 was expressed in transduced cells. In a given timeframe, expression of miRNA did not lead to c-kit mRNA cleavage, as seen from the presence of the c-kit gene transcript, but resulted in the inhibition of cKit protein translation (Fig. 4A). Direct immunofluorescent detection of cKit in parental 451Lu and cells expressing mir-221 confirmed these observations (Fig. 4B). Both parental and mir-221-expressing cells were actively proliferating, as judged by the positive staining for Ki-67 proliferation marker (Fig. 4B). As p27Kip1 is one of the targets for mir-221, we also examined the expression of this protein. Indirect immunofluorescent staining for p27Kip1 demonstrated a characteristic cytoplasmic relocation of the protein in parental 451Lu cells, consistent with the data obtained on other melanomas [17]. However, p27Kip1 was absent in mir-221-expressing cells (Fig. 4B). Collectively, these studies confirmed that mir-221 could simultaneously inhibit cKit and p27Kip1 protein expression in melanoma cells. To determine whether mir-221 directly acts on c-kit mRNA 30 UTR, we established a stable cell line of CHO cells expressing this miRNA and nucleofected these cells with mammalian expression vectors containing human c-kit cDNA with or without 30 UTR (see Materials and methods). Forty-eight hours after nucleofection, cells were lysed and expression of the cKit was examined by Western blot analysis. In mir-221-positive cells, expression of the cKit from 30 UTR-containing plasmid was decreased, whereas in cells transduced with cDNA it was unaffected (Fig. 4C). These studies confirmed that mir-221 could directly act on 30 UTR of the c-kit mRNA and inhibit protein translation. We further investigated how constitutive expression of mir-221 affects cKit-positive 451Lu melanoma cells. Cells were stably nucleofected with mir-221 encoding plasmid and seeded in a clonal density. Parental 451Lu cells nucleofected with mock vector were used as a control. As shown in Fig. 4, after two weeks of selection with Bsd, control cells migrated throughout the culture plate and gave rise to many small sparse and flat colonies. On contrary, mir-221 expressing cells gave rise to a smaller number of colonies, which were on average five times bigger in diameter and contained up to 20-times more cells per individual clone (Fig. 4C). This data demonstrated that mir-221-mediated inhibition of cKit and p27Kip1 resulted in elevated proliferation of the cells, and, possibly, impaired motility. Overall, our current study demonstrated that in cutaneous melanomas, expression of cKit RTK does not depend on AP2 transcription factor but is regulated by miRNAs, mir-221 in particular. Since more than 70% of cutaneous melanomas are characterized by the mutations in BRAF, it is possible that miRNAdependent inhibition of cKit in these melanomas serves as a mechanism to escape cKit-dependent apoptosis. Considering the fact that p27Kip1, a member of cyclin dependent kinase family that negatively controls cell cycle progression, is also targeted by mir-221, it is likely that simultaneous inhibition of cKit and p27Kip1 in melanoma permits uncontrolled proliferation of transformed cells leading to rapid progression of the tumors. It is also plausible that besides cKit and p27Kip1, mir-221 acts on other yet unidentified targets inhibition of which leads to the impaired motility of melanoma cells. Collectively, we demonstrated that mir-221 is an important regulator of pivotal proteins, inhibition of which may significantly contribute to the overall tumorigenicity and progression of cutaneous melanoma. Our data also suggest that this and other miRNAs found to be up-regulated in melanoma could be novel therapeutic targets, expression and biological activities of which should be further investigated in the context of cutaneous melanoma.
794
O. Igoucheva, V. Alexeev / Biochemical and Biophysical Research Communications 379 (2009) 790–794
Fig. 4. Plasmid-derived expression and functional activity of mir-221 in 451Lu melanoma cells. (A) RT-PCR and Western blot (WB) analyses of miRNA and cKit expression. Lanes: 1—parental 451Lu; 2—mir-221 expressing 451Lu. (B) Immunofluorescent detection of cKit and p27Kip1 protein expression and proliferation of 451Lu melanoma cells. Green—cKit, red—Ki-67, magenta—p27Kip1, blue—DAPI nuclear staining. Cell types and detected proteins are indicated next to the photographs. Scale bar—40 lm. (C) Western blot analysis of cKit derived from expression vectors with or without 30 c-kit UTR in mir-221-expressing CHO cells. (D) Proliferation and colony forming potentials of 451Lu and mir-221-expressing 451Lu cells. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this paper.)
Acknowledgments We would like to thank Drs. M. Herlyn, R.E. Boissy and Z. AbdelMalek for providing us with various melanocytic cells and for helpful discussions. This research was supported by the Jefferson Medical College to V.A. References [1] J. Lennartsson, P. Blume-Jensen, M. Hermanson, E. Ponten, M. Carlberg, L. Ronnstrand, Phosphorylation of Shc by Src family kinases is necessary for stem cell factor receptor/c-kit mediated activation of the Ras/MAP kinase pathway and c-fos induction, Oncogene 18 (1999) 5546–5553. [2] N. Carpino, D. Wisniewski, A. Strife, D. Marshak, R. Kobayashi, B. Stillman, B. Clarkson, p62(dok): a constitutively tyrosine-phosphorylated, GAP-associated protein in chronic myelogenous leukemia progenitor cells, Cell 88 (1997) 197–204. [3] I. Timokhina, H. Kissel, G. Stella, P. Besmer, Kit signaling through PI 3-kinase and Src kinase pathways: an essential role for Rac1 and JNK activation in mast cell proliferation, EMBO J. 17 (1998) 6250–6262. [4] W.W. An, M.W. Wang, S. Tashiro, S. Onodera, T. Ikejima, Mitogen-activated protein kinase-dependent apoptosis in norcan-tharidin-treated A375-S2 cells is proceeded by the activation of protein kinase C, Chin. Med. J. (Engl.) 118 (2005) 198–203. [5] V. Alexeev, K. Yoon, Distinctive role of the cKit receptor tyrosine kinase signaling in mammalian melanocytes, J. Invest. Dermatol. 126 (2006) 1102– 1110. [6] J.A. Curtin, K. Busam, D. Pinkel, B.C. Bastian, Somatic activation of KIT in distinct subtypes of melanoma, J. Clin. Oncol. 24 (2006) 4340–4346. [7] J.L. Maldonado, J. Fridlyand, H. Patel, A.N. Jain, K. Busam, T. Kageshita, T. Ono, D.G. Albertson, D. Pinkel, B.C. Bastian, Determinants of BRAF mutations in primary melanomas, J. Natl. Cancer Inst. 95 (2003) 1878–1890. [8] K.S. Smalley, R. Contractor, T.K. Nguyen, M. Xiao, R. Edwards, V. Muthusamy, A.J. King, K.T. Flaherty, M. Bosenberg, M. Herlyn, K.L. Nathanson, Identification of a novel subgroup of melanomas with KIT/cyclin-dependent kinase-4 overexpression, Cancer Res. 68 (2008) 5743–5752.
[9] M.S. Brose, P. Volpe, M. Feldman, M. Kumar, I. Rishi, R. Gerrero, E. Einhorn, M. Herlyn, J. Minna, A. Nicholson, J.A. Roth, S.M. Albelda, H. Davies, C. Cox, G. Brignell, P. Stephens, P.A. Futreal, R. Wooster, M.R. Stratton, B.L. Weber, BRAF and RAS mutations in human lung cancer and melanoma, Cancer Res. 62 (2002) 6997–7000. [10] S. Huang, M. Luca, M. Gutman, D.J. McConkey, K.E. Langley, S.D. Lyman, M. BarEli, Enforced c-KIT expression renders highly metastatic human melanoma cells susceptible to stem cell factor-induced apoptosis and inhibits their tumorigenic and metastatic potential, Oncogene 13 (1996) 2339–2347. [11] N. Felli, L. Fontana, E. Pelosi, R. Botta, D. Bonci, F. Facchiano, F. Liuzzi, V. Lulli, O. Morsilli, S. Santoro, M. Valtieri, G.A. Calin, C.G. Liu, A. Sorrentino, C.M. Croce, C. Peschle, MicroRNAs 221 and 222 inhibit normal erythropoiesis and erythroleukemic cell growth via kit receptor down-modulation, Proc. Natl. Acad. Sci. USA 102 (2005) 18081–18086. [12] H. He, K. Jazdzewski, W. Li, S. Liyanarachchi, R. Nagy, S. Volinia, G.A. Calin, C.G. Liu, K. Franssila, S. Suster, R.T. Kloos, C.M. Croce, A. de la Chapelle, The role of microRNA genes in papillary thyroid carcinoma, Proc. Natl. Acad. Sci. USA 102 (2005) 19075–19080. [13] S. Galardi, N. Mercatelli, E. Giorda, S. Massalini, G.V. Frajese, S.A. Ciafre, M.G. Farace, miR-221 and miR-222 expression affects the proliferation potential of human prostate carcinoma cell lines by targeting p27Kip1, J. Biol. Chem. 282 (2007) 23716–23724. [14] F. Felicetti, M.C. Errico, L. Bottero, P. Segnalini, A. Stoppacciaro, M. Biffoni, N. Felli, G. Mattia, M. Petrini, M.P. Colombo, C. Peschle, A. Care, The promyelocytic leukemia zinc finger-microRNA-221/-222 pathway controls melanoma progression through multiple oncogenic mechanisms, Cancer Res. 68 (2008) 2745–2754. [15] S.M. Hong, H.F. Frierson Jr., C.A. Moskaluk, AP-2gamma protein expression in intratubular germ cell neoplasia of testis, Am. J. Clin. Pathol. 124 (2005) 873– 877. [16] C.G. Liu, G.A. Calin, B. Meloon, N. Gamliel, C. Sevignani, M. Ferracin, C.D. Dumitru, M. Shimizu, S. Zupo, M. Dono, H. Alder, F. Bullrich, M. Negrini, C.M. Croce, An oligonucleotide microchip for genome-wide microRNA profiling in human and mouse tissues, Proc. Natl. Acad. Sci. USA 101 (2004) 9740– 9744. [17] C. Denicourt, C.C. Saenz, B. Datnow, X.S. Cui, S.F. Dowdy, Relocalized p27Kip1 tumor suppressor functions as a cytoplasmic metastatic oncogene in melanoma, Cancer Res. 67 (2007) 9238–9243.
Contents lists available at ScienceDirect
Biochemical and Biophysical Research Communications journal homepage: www.elsevier.com/locate/ybbrc
MicroRNA-dependent regulation of cKit in cutaneous melanoma O. Igoucheva, V. Alexeev * Department of Dermatology and Cutaneous Biology, Jefferson Medical College, Thomas Jefferson University, 233 S. 10th Street, BLSB, Suite 326, Philadelphia, PA 19107, USA
a r t i c l e
i n f o
Article history: Received 18 December 2008 Available online 4 January 2009
Keywords: cKit MicroRNA Cutaneous melanoma Melanocytes mir-221
a b s t r a c t Loss of cKit receptor in cutaneous melanomas was attributed to the down-regulation of AP2 transcription factor. Our analysis of 27 melanoma cell lines showed no correlation between AP2 and c-kit expression. Suggesting a post-transcriptional mechanism of cKit down-modulation, we performed genome-wide microRNA (miRNA) expression profiling and found that several miRNA species are commonly up-regulated in melanomas. Among them was mir-221, which can directly interact with c-kit 30 UTR and inhibit cKit protein translation. Observed inverse correlation of the c-kit and mir-221 expression in various melanocytic cells pointed to its involvement in regulation of cKit in melanoma. Moreover, a series of functional assays demonstrated that mir-221 could directly inhibit cKit, p27Kip1 and, possibly, other pivotal proteins in melanoma. Collectively, the studies presented here indicate that mir-221 could be a novel therapeutic target for the treatment of cutaneous melanoma. They also suggest that regulation of expression and functional activity of identified up-regulated miRNAs should be further studied in the context of malignant melanoma. Ó 2009 Elsevier Inc. All rights reserved.
cKit tyrosine kinase receptor for the stem cell factor (SCF), encoded by the proto-oncogene c-kit, belongs to the PDGF family of the receptor tyrosine kinases (RTK). The intracellular signaling from cKit RTK plays a critical role in the development of a variety of mammalian cell types including hematopoietic progenitor cells, mast cells, primordial germ cells, intestinal cells of Cajal, and melanocytes. It is well established that cKit RTK transduce intracellular signals that lead to various, sometimes conflicting cellular responses. It has been demonstrated that this RTK can transmit pro-proliferation [1], anti-proliferation [2], survival, pro-migration [3], and pro-apoptotic [4] signals. In normal melanocytes, SCFdependent cKit-mediated signaling supports proliferation, migration and differentiation of the cells. Constitutive activation of cKit RTK alone does not induce tumorigenic transformation of the melanocytes in vitro and in vivo [5]. However, in certain subtypes of melanomas characterized by activating mutations and/or increased copy number of c-kit gene [6] with rare occurrence of BRAF mutations [7], cKit RTK activity enhances tumorigenic potential of transformed melanocytes. Studies on pharmacological inhibition of cKit RTK in certain subpopulations of cKit-positive melanomas with Imatinib showed increased apoptosis and G1 phase cell cycle arrest associated with the inhibition of phosphoERK and elevated expression of p27Kip1 tumor suppressor [8]. In contrast to normal melanocytes and cKit-expressing melanomas, lesions of cutaneous melanoma occurring on intermittent sun exposed skin are often characterized by the loss of cKit receptor * Corresponding author. Fax: +1 215 503 5788. E-mail address: [email protected] (V. Alexeev). 0006-291X/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.bbrc.2008.12.152
expression [6] and high frequency of activating mutations in the BRAF [9]. Earlier studies on cKit-negative BRAF mutant, A375 melanoma cells demonstrated that re-expression and restoration of cKit-mediated signaling in these cells induced cKit-dependent apoptosis and resulted in a loss of tumorigenic potential of melanoma [10]. Despite extensive studies of the cKit in melanocytes and melanomas, the molecular mechanisms that govern the expression of this RTK remain incompletely understood. Several recent studies demonstrated that a family of 19–25nucleotide-long RNA molecules termed as microRNA (miRNA) represents a new class of highly conserved non-coding RNAs that function as negative regulators of protein expression. In multiple cases, it was shown that deregulation of miRNA expression contributes to cancer development and facilitates the progression of multiple myeloma, prostate cancers, Burkitt lymphoma and other types of tumors. However, very limited and discrete information regarding miRNA expression and function in melanoma is currently available. As cKit RTK plays a critical role in a variety of mammalian cell types, several studies examined whether cKit protein expression is controlled by miRNA. Based on bioinformatics predictions, ckit mRNA can be targeted by at least 12 different miRNAs. Studies on erythropoietic CD34+ cord blood progenitor cell showed that supplementation of cells with highly homologous mir-221 and mir-222 caused impaired proliferation of the cells directly coupled with down-modulation of cKit protein [11]. Recent miRNA expression profiling in thyroid papillary carcinoma (TPC) also showed a distinguished up-regulation of these two miRNAs in TPC characterized by the loss of cKit [12]. Other studies showed that mir-221
O. Igoucheva, V. Alexeev / Biochemical and Biophysical Research Communications 379 (2009) 790–794
could also down-modulate the expression of the p27Kip1 tumor suppressor and promote proliferation of various tumor cells [13]. Very recently, Felicetti and co-workers identified the promyelocytic leukemia zinc finger (PLZF) transcription factor as a repressor of miR-221 and miR-222. Authors demonstrated that targeted PLZF silencing in cKitlow melanoma cell line unblocks miR-221 [14]. Collectively, these studies demonstrated that cKit protein expression can be regulated by miRNA, mir-221 in particular. Therefore, the goals of this study were to identify deregulated miRNAs in cutaneous melanomas and determine whether mir-221 contributes to the progression of this malignancy via inhibition of cKit.
Materials and methods Cell lines. WM and Lu melanoma cell lines were provided by Dr. M. Herlyn (Wistar Institute, Philadelphia, PA). Primary human melanocytes were provided by Drs. R.E. Boissy and Z. Abdel-Malek (University of Cincinnati, Cincinnati, OH). Melanoma cells were cultured as described at www.wistar.org/herlyn/resources_intro.htm. Primary melanocytes were cultured in melanocyte-specific media with TPA-free supplements (Invitrogen, Carlsbad, CA). DNA microarray. Analysis of AP2 a and c isoform expression in various melanocytic cells was done using Affymetrix gene expression HG-U95Av2 chips utilizing cervices provided by the Center for Translational Medicine at Thomas Jefferson University. Semi-quantitative RT-PCR analysis of mir-221 expression. Total RNA from various melanocytic cells was isolated using TRIzol reagent (Invitrogen, Carlsbad, CA). Total RNA was treated with DNase-1 and reverse transcribed using SuperScript II system (Invitrogen, Carlsbad, CA). AP2 a and c, c-kit and mir-221 were amplified using gene-specific primers: AP2 a: 50 -GCCCCGTGTCCCTGTCCAA-30 and 50 -TGAGGAGCGAGAGGCGACC-30 AP2 c: 50 -GGCCCAGCAACT GTGTAAAGA-30 and 50 -GCAGTTCTGTATGTTCGTCTCCAA-30 c-kit: 50 -GATCATGCAGAAGCTGCACT-30 and 50 -AAAATCCCATAGGACCAGA C-30 mir-221: 50 -TGAATGCAGTAGGCAGTTGTG-30 and 50 -GGGGTAG CATTGGTGAGACA-30 GAPDH: 50 -AACGTGTCAGTGGTGGACCT-30 and 50 -TGCTGTAGCCAAATTCGTTG-30 . Amplification of GAPDH was used as a control for equal RT-PCR loading. Nucleofection of melanoma cells. Optimization of anti-miRTM RNA oligonucleotides delivery into cultured melanoma cells was done using Cy3-labeled anti-miRTM RNA oligonucleotides (Ambion Inc., Applied Biosystems, Austin, TX). Using Amaxa nucleofection system II (Lonza, Gaithersburg, MD) and 50 nM of Cy3-labeled oligonucleotides, several nucleofection conditions were tested. The highest efficacy of oligonucleotide delivery (up to 90%) and viability of the cells (up to 80%) was achieved when oligonucleotides were nucleofected using NHEM Neo nucleofection kit (Lonza) and U-024 program (data not shown). These conditions were also optimal for the delivery of plasmids DNA into cultured melanoma cells. Generation of miRNA and cKit expression vectors. A genomic sequence containing mir-221 was amplified by PCR from the total human genomic DNA using direct: 50 -CCCAGCATTTCTGACTG-30 and reverse: 50 -ATTTCACCCAAATGGTCT-30 primers. Amplified fragments were ligated into pEF6-TOPO expression vector (Invitrogen, Carlsbad, CA) containing human EF1 promoter and blasticidin resistance gene (Bsd) for rapid selection of the transduced cells. The integrity of the promoter and gene sequences were verified by direct DNA sequencing. The resultant plasmid was designated as pEF1-mir-221. Human c-kit cDNA with or without 30 UTR was obtained by RTPCR reaction using total RNA isolated from primary human melanocytes and gene-specific primers: cDNA: 50 -ATGAGAGGCGCTCGCGG CGCC-30 and 50 -TCAGACATCGTCGTGCACAAGCA-30 cDNAUTR: 50 -AT GAGAGGCGCTCGCGGCGCC-30 and 50 -TCCAATTGTTGACAGTTCTGAA G-30 . Resultant sequences were ligated into pEF6-TOPO vector.
791
Western blot, immunofluorescent and TUNEL assays: For the Western blot analysis, cells were lysed in a buffer containing 20 mM Tris–HCl, pH 7.5, 150 mM NaCl, 1 mM EDTA, 1% Triton X100, 0.5% NP-40, 1 mM PMSF, and protease inhibitors (Pierce, Rockford, IL). For the analysis of cKit phosphorylation, a cocktail of phosphatase inhibitors (Pierce, Rockford, IL) was added to lysis buffer. cKit was detected on blots by protein-specific antibodies (Ab81). Phospho-cKit was detected by monoclonal antibodies (Tyr823) specific to phosphorylated cKit receptor. PARP was detected by monoclonal antibodies (19F4) specific to the cleaved form of the protein. All antibodies were purchased from Cell Signaling Technologies (Danvers, MA) and used in dilution 1:1000. For the direct immunofluorescent detection of cKit in melanoma cells, AlexaFluor488-lebeled monoclonal (Ab81) antibodies were used in dilution 1:50 (Cell Signaling Technologies, Danvers, MA). Ki-67 and p27Kip1 were detected by indirect immunofluorescence using protein-specific antibodies (Millipore, Bedford, MA) as devised by the manufacturer. Apoptosis in 1205Lu melanoma cells was detected using TUNEL assay (In Situ Cell Death detection kit, Roche Diagnostics, Mannheim, Germany) as devised by the manufacturer.
Results and discussion AP2 expression in melanocytic cells In human melanocytic cells, AP2 transcription factor was shown to be directly involved in regulation of c-kit gene transcription. As only limited number of melanomas were previously tested for the expression of the AP2 a, here we examined the expression of structurally and functionally similar a and c isoforms of the AP2 in six lines of normal, two lines of vitiligo melanocytes, and 27 lines of melanoma cell isolated from tumors on different stages of tumor progression (Fig. 1A). Based on intensity of hybridization signals, our analysis confirmed earlier findings showing general down-regulation of AP2 a expression in primary melanoma cells. However, in contrast to previous findings, we did not observe any direct correlation between down-regulation of AP2 a and progression of the malignancy from radial growth phase (RGP) through vertical growth phase (VGP) to metastatic (MM) form. Both high and low levels of the AP2 a were found in melanomas, isolated from GRP, VGP and MM tumors. In almost all melanoma cells, lower expression of AP2 a was accompanied by the elevated expression AP2 c. Considering structural and functional similarities between these two isoforms, it is likely that in melanomas, similarly to germ cell tumors and gliomas [15], up-regulated AP2 c compensates for the down-regulated a-isoform. Based on our findings, average combined expression of both AP2 isoforms in malignant melanoma was only 1.4 times lower than that in normal adult melanocytes (Fig. 1B). Therefore, down-regulation or complete lack of cKit in the majority of cutaneous melanomas cannot be attributed exclusively to the down-regulation of the AP2 a. These observations directly pointed to the existence of an alternative mechanism(s) of the cKit regulation in melanoma. Analysis of miRNA expression in melanoma Suggesting post-transcriptional mechanism of cKit regulation, we hypothesized that microRNA may inhibit cKit protein expression. Having an original setup for genome-wide miRNA gene expression profiling at Thomas Jefferson University [16], we analyzed the expression of more than 300 miRNAs in primary normal melanocytes and melanoma cell lines as described previously [16]. In contrast to previously published data showing down-regulation of many miRNA species in various cancers, we did not find any
792
O. Igoucheva, V. Alexeev / Biochemical and Biophysical Research Communications 379 (2009) 790–794
Fig. 1. AP2 a and c expression in melanocytic cell. Expression of AP2 a and c was examined in melanocytes (F—fetal, A—adult, V—vitiligo) and melanoma cells. Cell lines are positioned based on the expression of the AP2 a (from the highest to the lowest). Cell lines and other available information (stage of tumor progression) are indicated above the columns. (A) Variations of AP2 a (open columns) and AP2 c (solid columns) expression in melanocytic cells. (B) Combined expression of both a and c isoforms.
Table 1 miRNA expression in melanocytes and melanoma. miRNA
Location
Expression in melanocyte/melanoma
Some predicted or confirmed targets
Normal
RGP
VGP
MM
mir-23a mir-24-2 mir-25 mir-92
19p13.3 19p13.8 7q22 13q31
+ + ±
++ + ++ ++++
++++ +++ ++ ++++
++++ ++++ ++ ++++
mir-124 mir-146b mir-149
8p23.1 5q34 2q37.3
+++
+ +++ +++++
++++ +++ ++++
mir-213 mir-221 mir-324 mir-338
1q32.1 Xp11.3 17p13.1 17q25.3
+ ++ ++
++ +++ +++ +
++ ++++ +++ +
MAP4K4 (2), metastasis suppressor 1 (1) BCL2-like 11 (apoptosis facilitator) (2),integrin, alpha V (2), MAP2K4 (2), PI3K regulatory subunit 3 (1), heart and neural crest derivatives (HAND1) mRNA MITF (2), RAB27A (2). neuro-oncological ventral antigen 1 (2), periaxin (2) AP2 associated kinase 1 (1), cKit (1)? BCL2-like apoptosis facilitator (1) ??? cKit (1), p27Kip1 (2), Tyrosinase (1) ??? MAGE10 (2), cKit (1)
Predicted number of miRNA binding sites on 30 UTRs indicated in parenthesis. Not yet predicted targets are indicated with question marks (?).
significantly down-regulated miRNAs in melanomas, as compared to melanocytes. However, several miRNA species were up-regulated in RGP, VGP and MM tumors (Table 1). As many miRNA genes are located at regions of amplification or translocation, we compared miRNA loci to the known chromosomal abnormalities reported in melanoma. Our analysis showed that the loci of miRNA species highly expressed in melanoma (Table 1) did not map to the known amplification or translocation sites associated with this tumor. Out of all deregulated miRNAs, elevated expression of mir-221 (and homologous mir-222) was of particular interest. This miRNA was expressed in melanoma cells showing decreased levels of cKit transcripts (WM35, WM983A, WM164, and 1205Lu). Similar results were obtained when RNA samples isolated from additional melanoma cell lines were analyzed for the genome-wide miRNA expression using GenoSensor services (GenoSensor, Tempe, AZ). In addition to the already established profiles, this analysis demonstrated that mir-149 and mir-338 were up-regulated in cKit-negative melanoma cells isolated from tumor cells on earlier stages of tumor progression (RGP and RGP/VGP). Based on bioinformatics predictions, mir-221, mir-149, and mir-338 can form favorable duplexes with the 30 UTR of the c-kit mRNA with energies 20.9 kcal/mol, 23.4 kcal/mol, and 16.7 kcal/mol, respectively (see: http://microrna.sanger.ac.uk/). To examine mir-221 expression in larger number of samples, we established the conditions for miRNA detection using RT-PCR (see Materials and methods). For this analysis, we used additional samples of primary normal and vitiligo melanocytes, 17 cutaneous
melanoma cell lines, 1 BRAF normal melanoma line, and two samples of uveal melanomas. Same RNA samples were used for the analysis of AP2 a, AP2 c and c-kit expression. As shown in Fig. 2, AP2 a and AP2 c were equally expressed in all analyzed cells (with few exceptions). However, level of c-kit transcription varied from the highest in primary melanocytes to the lowest or absent in cutaneous melanomas. The level of mir-221 transcription varied as well (Fig. 2). Overall, this RT-PCR-based analysis clearly demonstrated
Fig. 2. Analysis of mir-221, AP2 and cKit expression in melanocytic cells. Cell line names shown above the photographs; analyzed genes shown on the right. Abbreviations: Cells were isolated from N—normal skin; Vt—vitiligo skin; P—primary melanoma lesion; LN—lymph node metastasis; LM—lung metastasis; R—RGP melanoma; V—VGP melanoma; R/V—RGP/VGP melanoma; LvM—liver metastasis of primary uveal melanoma; Br—BRAF normal melanoma.
O. Igoucheva, V. Alexeev / Biochemical and Biophysical Research Communications 379 (2009) 790–794
that levels of c-kit transcription do not correlate with the AP2 expression, but inversely correlate with the expression of mir-221. miRNA inhibition and cKit expression Further, we tested whether enforced, nucleic acid-based inhibition of mir-221 could lead to the restoration of cKit expression and cKit-mediated signaling. Therefore, cKit-negative, mir-221-positive 1205Lu melanoma cells were nucleofected with control or mir-221specific anti-miRTM inhibiting oligonucleotides (Ambion) under optimal conditions (see Materials and methods). Nucleofected cells were cultured for 48 h with or without stem cell factor (SCF). Cells were then examined for the expression of cKit and phosphorylation. Oligonucleotide-based inhibition of mir-221 resulted in a rapid restoration of the cKit protein expression and autophosphorylation of the receptor when cells were exposed to SCF (Fig. 3A). Moreover, cells with restored cKit expression and signaling showed an increase of SCF-dependent apoptosis as compared with cells cultured in the absence of SCF or control cells as deduced from the presence of cleaved PARP protein, the hallmark of apoptotic cell death (Fig. 3B). Induction of apoptosis was confirmed by TUNEL assay, which demonstrated increased apoptosis of cells treated with antimir-221 oligonucleotides and cultured in the presence of SCF (Fig. 3C). Collectively, these studies demonstrated that targeted inhibition of mir-221 led to the restoration of cKit expression and liganddependent signaling, which resulted in the induction of apoptosis of BRAF-mutant melanoma. These findings directly indicate that targeted inhibition of mir-221 could be a novel therapeutic approach for the treatment of cutaneous melanoma. miRNA-dependent inhibition of cKit To confirm cKit-inhibiting activity of mir-221, we generated mammalian expression vector coding for human mir-221 (see
Fig. 3. Western blot analysis of the cKit re-expression and induction of SCFdependent apoptosis. (A) cKit receptor expression and phosphorylation. Lanes: 1— 1205Lu cells; 2—1205Lu cells treated with control oligonucleotide (+SCF); 3— 1205Lu cells treated with anti-mir-221 RNA oligonucleotides (+SCF). (B) Detection of cleaved PARP in cells treated with anti-mir-221 RNA oligonucleotides; Lanes: 1— 1205Lu cells; 2—1205Lu cells treated with control anti-mir oligonucleotides (+SCF); 3—1205Lu cells treated with anti-mir-221 RNA oligonucleotides (SCF); 4—1205Lu cells treated with anti-mir-221 RNA oligonucleotides (+SCF). Arrows on the left point to the position of the molecular weight markers. Arrows on the right point to cKit and cleaved PARP. Antibodies indicated below the panels. (C) Detection of apoptosis by TUNEL assay. Treatment and culture conditions indicated on the panels. Apoptotic cells—green. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this paper.)
793
Materials and methods). Expression vector was nucleofected into cKithigh 451Lu melanoma cells under optimized conditions (see Materials and methods). Ninety-six hours later, cells were analyzed for the miRNA and cKit expression. As shown in Fig. 4, mir221 was expressed in transduced cells. In a given timeframe, expression of miRNA did not lead to c-kit mRNA cleavage, as seen from the presence of the c-kit gene transcript, but resulted in the inhibition of cKit protein translation (Fig. 4A). Direct immunofluorescent detection of cKit in parental 451Lu and cells expressing mir-221 confirmed these observations (Fig. 4B). Both parental and mir-221-expressing cells were actively proliferating, as judged by the positive staining for Ki-67 proliferation marker (Fig. 4B). As p27Kip1 is one of the targets for mir-221, we also examined the expression of this protein. Indirect immunofluorescent staining for p27Kip1 demonstrated a characteristic cytoplasmic relocation of the protein in parental 451Lu cells, consistent with the data obtained on other melanomas [17]. However, p27Kip1 was absent in mir-221-expressing cells (Fig. 4B). Collectively, these studies confirmed that mir-221 could simultaneously inhibit cKit and p27Kip1 protein expression in melanoma cells. To determine whether mir-221 directly acts on c-kit mRNA 30 UTR, we established a stable cell line of CHO cells expressing this miRNA and nucleofected these cells with mammalian expression vectors containing human c-kit cDNA with or without 30 UTR (see Materials and methods). Forty-eight hours after nucleofection, cells were lysed and expression of the cKit was examined by Western blot analysis. In mir-221-positive cells, expression of the cKit from 30 UTR-containing plasmid was decreased, whereas in cells transduced with cDNA it was unaffected (Fig. 4C). These studies confirmed that mir-221 could directly act on 30 UTR of the c-kit mRNA and inhibit protein translation. We further investigated how constitutive expression of mir-221 affects cKit-positive 451Lu melanoma cells. Cells were stably nucleofected with mir-221 encoding plasmid and seeded in a clonal density. Parental 451Lu cells nucleofected with mock vector were used as a control. As shown in Fig. 4, after two weeks of selection with Bsd, control cells migrated throughout the culture plate and gave rise to many small sparse and flat colonies. On contrary, mir-221 expressing cells gave rise to a smaller number of colonies, which were on average five times bigger in diameter and contained up to 20-times more cells per individual clone (Fig. 4C). This data demonstrated that mir-221-mediated inhibition of cKit and p27Kip1 resulted in elevated proliferation of the cells, and, possibly, impaired motility. Overall, our current study demonstrated that in cutaneous melanomas, expression of cKit RTK does not depend on AP2 transcription factor but is regulated by miRNAs, mir-221 in particular. Since more than 70% of cutaneous melanomas are characterized by the mutations in BRAF, it is possible that miRNAdependent inhibition of cKit in these melanomas serves as a mechanism to escape cKit-dependent apoptosis. Considering the fact that p27Kip1, a member of cyclin dependent kinase family that negatively controls cell cycle progression, is also targeted by mir-221, it is likely that simultaneous inhibition of cKit and p27Kip1 in melanoma permits uncontrolled proliferation of transformed cells leading to rapid progression of the tumors. It is also plausible that besides cKit and p27Kip1, mir-221 acts on other yet unidentified targets inhibition of which leads to the impaired motility of melanoma cells. Collectively, we demonstrated that mir-221 is an important regulator of pivotal proteins, inhibition of which may significantly contribute to the overall tumorigenicity and progression of cutaneous melanoma. Our data also suggest that this and other miRNAs found to be up-regulated in melanoma could be novel therapeutic targets, expression and biological activities of which should be further investigated in the context of cutaneous melanoma.
794
O. Igoucheva, V. Alexeev / Biochemical and Biophysical Research Communications 379 (2009) 790–794
Fig. 4. Plasmid-derived expression and functional activity of mir-221 in 451Lu melanoma cells. (A) RT-PCR and Western blot (WB) analyses of miRNA and cKit expression. Lanes: 1—parental 451Lu; 2—mir-221 expressing 451Lu. (B) Immunofluorescent detection of cKit and p27Kip1 protein expression and proliferation of 451Lu melanoma cells. Green—cKit, red—Ki-67, magenta—p27Kip1, blue—DAPI nuclear staining. Cell types and detected proteins are indicated next to the photographs. Scale bar—40 lm. (C) Western blot analysis of cKit derived from expression vectors with or without 30 c-kit UTR in mir-221-expressing CHO cells. (D) Proliferation and colony forming potentials of 451Lu and mir-221-expressing 451Lu cells. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this paper.)
Acknowledgments We would like to thank Drs. M. Herlyn, R.E. Boissy and Z. AbdelMalek for providing us with various melanocytic cells and for helpful discussions. This research was supported by the Jefferson Medical College to V.A. References [1] J. Lennartsson, P. Blume-Jensen, M. Hermanson, E. Ponten, M. Carlberg, L. Ronnstrand, Phosphorylation of Shc by Src family kinases is necessary for stem cell factor receptor/c-kit mediated activation of the Ras/MAP kinase pathway and c-fos induction, Oncogene 18 (1999) 5546–5553. [2] N. Carpino, D. Wisniewski, A. Strife, D. Marshak, R. Kobayashi, B. Stillman, B. Clarkson, p62(dok): a constitutively tyrosine-phosphorylated, GAP-associated protein in chronic myelogenous leukemia progenitor cells, Cell 88 (1997) 197–204. [3] I. Timokhina, H. Kissel, G. Stella, P. Besmer, Kit signaling through PI 3-kinase and Src kinase pathways: an essential role for Rac1 and JNK activation in mast cell proliferation, EMBO J. 17 (1998) 6250–6262. [4] W.W. An, M.W. Wang, S. Tashiro, S. Onodera, T. Ikejima, Mitogen-activated protein kinase-dependent apoptosis in norcan-tharidin-treated A375-S2 cells is proceeded by the activation of protein kinase C, Chin. Med. J. (Engl.) 118 (2005) 198–203. [5] V. Alexeev, K. Yoon, Distinctive role of the cKit receptor tyrosine kinase signaling in mammalian melanocytes, J. Invest. Dermatol. 126 (2006) 1102– 1110. [6] J.A. Curtin, K. Busam, D. Pinkel, B.C. Bastian, Somatic activation of KIT in distinct subtypes of melanoma, J. Clin. Oncol. 24 (2006) 4340–4346. [7] J.L. Maldonado, J. Fridlyand, H. Patel, A.N. Jain, K. Busam, T. Kageshita, T. Ono, D.G. Albertson, D. Pinkel, B.C. Bastian, Determinants of BRAF mutations in primary melanomas, J. Natl. Cancer Inst. 95 (2003) 1878–1890. [8] K.S. Smalley, R. Contractor, T.K. Nguyen, M. Xiao, R. Edwards, V. Muthusamy, A.J. King, K.T. Flaherty, M. Bosenberg, M. Herlyn, K.L. Nathanson, Identification of a novel subgroup of melanomas with KIT/cyclin-dependent kinase-4 overexpression, Cancer Res. 68 (2008) 5743–5752.
[9] M.S. Brose, P. Volpe, M. Feldman, M. Kumar, I. Rishi, R. Gerrero, E. Einhorn, M. Herlyn, J. Minna, A. Nicholson, J.A. Roth, S.M. Albelda, H. Davies, C. Cox, G. Brignell, P. Stephens, P.A. Futreal, R. Wooster, M.R. Stratton, B.L. Weber, BRAF and RAS mutations in human lung cancer and melanoma, Cancer Res. 62 (2002) 6997–7000. [10] S. Huang, M. Luca, M. Gutman, D.J. McConkey, K.E. Langley, S.D. Lyman, M. BarEli, Enforced c-KIT expression renders highly metastatic human melanoma cells susceptible to stem cell factor-induced apoptosis and inhibits their tumorigenic and metastatic potential, Oncogene 13 (1996) 2339–2347. [11] N. Felli, L. Fontana, E. Pelosi, R. Botta, D. Bonci, F. Facchiano, F. Liuzzi, V. Lulli, O. Morsilli, S. Santoro, M. Valtieri, G.A. Calin, C.G. Liu, A. Sorrentino, C.M. Croce, C. Peschle, MicroRNAs 221 and 222 inhibit normal erythropoiesis and erythroleukemic cell growth via kit receptor down-modulation, Proc. Natl. Acad. Sci. USA 102 (2005) 18081–18086. [12] H. He, K. Jazdzewski, W. Li, S. Liyanarachchi, R. Nagy, S. Volinia, G.A. Calin, C.G. Liu, K. Franssila, S. Suster, R.T. Kloos, C.M. Croce, A. de la Chapelle, The role of microRNA genes in papillary thyroid carcinoma, Proc. Natl. Acad. Sci. USA 102 (2005) 19075–19080. [13] S. Galardi, N. Mercatelli, E. Giorda, S. Massalini, G.V. Frajese, S.A. Ciafre, M.G. Farace, miR-221 and miR-222 expression affects the proliferation potential of human prostate carcinoma cell lines by targeting p27Kip1, J. Biol. Chem. 282 (2007) 23716–23724. [14] F. Felicetti, M.C. Errico, L. Bottero, P. Segnalini, A. Stoppacciaro, M. Biffoni, N. Felli, G. Mattia, M. Petrini, M.P. Colombo, C. Peschle, A. Care, The promyelocytic leukemia zinc finger-microRNA-221/-222 pathway controls melanoma progression through multiple oncogenic mechanisms, Cancer Res. 68 (2008) 2745–2754. [15] S.M. Hong, H.F. Frierson Jr., C.A. Moskaluk, AP-2gamma protein expression in intratubular germ cell neoplasia of testis, Am. J. Clin. Pathol. 124 (2005) 873– 877. [16] C.G. Liu, G.A. Calin, B. Meloon, N. Gamliel, C. Sevignani, M. Ferracin, C.D. Dumitru, M. Shimizu, S. Zupo, M. Dono, H. Alder, F. Bullrich, M. Negrini, C.M. Croce, An oligonucleotide microchip for genome-wide microRNA profiling in human and mouse tissues, Proc. Natl. Acad. Sci. USA 101 (2004) 9740– 9744. [17] C. Denicourt, C.C. Saenz, B. Datnow, X.S. Cui, S.F. Dowdy, Relocalized p27Kip1 tumor suppressor functions as a cytoplasmic metastatic oncogene in melanoma, Cancer Res. 67 (2007) 9238–9243.