1045-1050 27/2/2009 12:25 ÌÌ ™ÂÏ›‰·1045 INTERNATIONAL JOURNAL OF ONCOLOGY 34: 1045-1050, 2009 Involvement of p16 and PTCH in pathogenesis of melanoma and basal cell carcinoma MAJA CRETNIK1, GORAZD POJE2, VESNA MUSANI1, BOZO KRUSLIN3, PETAR OZRETIC1, DAVOR TOMAS3, MIRNA SITUM4 and SONJA LEVANAT1 1Laboratory for Hereditary Cancer, Division of Molecular Medicine, Rudjer Boskovic Institute; 2University Department of Otorhinolaryngology and Head and Neck Surgery, Clinical Hospital Center Rebro, University of Zagreb; 3Ljudevit Jurak Department of Pathology; 4University Department of Dermatovenerology, Sestre milosrdnice University Hospital, Zagreb, Croatia Received November 12, 2008; Accepted January 20, 2009 DOI: 10.3892/ijo_00000230 Abstract. The involvement of two tumor suppressors p16 and The pathway is aberrantly activated in many human Ptch in pathogenesis of cutaneous melanomas and basal cell cancers and in basal cell carcinomas (BCC) mutations in carcinomas (BCCs) was studied through expression of Ptch Patched, Hedgehog and even Smoothened are frequent (4). and p16 and genetic alterations in 9p21 region (p16) and in In melanoma pathology, indications of involvement of the 9q22.3 region (PTCH) of chromosome 9. Immunohisto- Hh-Gli pathway started to show promise in therapeutic chemical analyses of paraffin-embedded tissues with Ptch and strategies very recently. Although involvement of p16 in p16 antibodies, typing for 9q22-q31 and 9p21 region with melanoma development has been shown either through loss polymorphic markers and p16 and Ptch mutation detection was of heterozygosity or mutation screenings of p16, connections done. Higher expression of p16 and Ptch in melanoma and of melanoma and Hh-Gli pathway are very recent (5). The BCC of the skin was frequently detected in studied cases. CDKN2A locus encodes a cyclin-dependent kinase inhibitor However, allelic loss of PTCH region occurs more frequently p16 that acts to inhibit cell cycle progression in the G1 phase in BCCs than loss of heterozygosity of p16 region. Both types by binding and inhibiting CDK4/6 kinases. Loss of p16 leads of tumors, BCCs and melanomas, suggest involvement of to deregulated CDK4/6 action and promotion of cell divisions. Hh-Gli signaling pathway, but using different mechanisms. We show a combined role of p16 and Hh-Gli signaling pathway in the pathogenesis of skin neoplasia, basal cell Introduction carcinoma and melanoma as the two most frequent types of skin neoplasia. The Hh-Gli signaling pathway is receiving increasing attention as a crucial regulator of not only embryonic organogenesis but Materials and methods also as an oncogenic pathway implicated in diverse human tumors. The pathway begins with binding of the ligand protein BCC and melanoma cases. Twenty sporadic malignant Hedgehog to the transmembrane receptor Patched (Ptch), melanomas (Table I) (Breslow thickness 3-5 mm, Clark's which then releases its inhibition of Smoothened (Smo) and a level 5 or less, classified as superficial spreading melanoma cytoplasmatic cascade of phosphorylation and dephosphory- (SSM), nodular melanoma (NM), acral lentigous melanoma lation events leads to activation of transcription factors Gli1-3 (ALM) or lentigo maligna melanoma (LMM) and twenty BCC and transcription of target genes, which include Cyclin D, samples (Table II) classified as carcinoma baseocellulare Cyclin E, members of Wnt and TGFß signaling pathways and solidum (CBS), adenocysticum (AC), morpheiphorme (M) or Ptch itself (1-3). superfitiale multicentricum (SM) were collected from Sestre milosrdnice University Hospital in Zagreb. The main clinical and histopathological features are summarized in Tables I _________________________________________ and II. The samples were stripped of identifiers and were collected with consideration to all necessary ethical and legal requirements. Ethics Committee of the hospital approved Correspondence to: Dr Sonja Levanat, Laboratory for Hereditary these studies. The experiments were conducted in accordance Cancer, Division of Molecular Medicine, Rudjer Boskovic Institute, Bijenicka 54, 10002 Zagreb, Croatia with the Declaration of Helsinki. E-mail: [email protected] All cases have been routinely processed and paraffin- embedded. Two 5-μm sections of tumor tissue were cut from Key words: p16, PTCH, melanoma, basal cell carcinoma, signaling paraffin block for each immunohistochemical analysis and pathways 25-μm sections were microdissected for loss of heterozygosity (LOH) analysis (Table III). The tissues were histologically evaluated by a trained pathologist. 1045-1050 27/2/2009 12:25 ÌÌ ™ÂÏ›‰·1046 1046 CRETNIK et al: p16 AND PTCH IN MELANOMA AND BASAL CELL CARCINOMA Table I. Melanoma samples and IH analyses. ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– No. Age Gender Typea Clark Breslow Size No. of mitoses Localization Immunohistochemistry Ptch p16 ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– 1 55 M LMM 4 4 13 6 Face ++ ++ 2 33 F SSM 3 3 19 3 Abdomen - - 3 60 M SSM 3 3 15 2 Neck - - 4 35 M SSM 4 4 7 8 Back ++ +/- 6 70 M SSM 4 5 15 5 Face - ++ 8 46 M SSM 3 5 10 20 Back ++ ++ 9 47 M SSM 3 5 11 10 Neck + + 18 62 F SSM 3 4 16 10 Upper arm + ++ 10 74 M NM 5 5 39 11 Back ++ ++ 11 62 M NM 5 5 45 11 Abdomen ++ ++ 12 73 F NM 4 5 25 20 Back + +/- 13 77 F NM 4 5 12 3 Leg ++ - 15 48 M NM 4 5 11 8 Back ++ +/- 16 20 M NM 3 4 10 13 Neck ++ ++ 7 56 F ALM 4 4 15 7 Face ++ ++ ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– aSSM, superficial spreading melanoma; NM, nodular melanoma; ALM, acral lentigous melanoma and LMM, lentigo maligna melanoma. ––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Immunohistochemical method. Ptch and p16 proteins were MgCl2, 10 pmol/l each primer, 1.25 U Taq polymerase analyzed as follows: slides were deparaffinized, prewarmed (AmpliTag Gold® DNA polymerase, Applied Biosystems, in Epitope Retrieval Solution (Dako, Glostrup, Denmark) and Foster City, CA) in 10 mmol/l Tris-HCl buffer. PCR system kept at 95-99˚C for 40 min and then washed in PBS 3x5 min. (Gene Amp PCR system 2400, Applied Biosystems) was set Endogenous peroxidase was blocked with 3% H2O2/methanol for 25 to 35 cycles, with the parameters published previously for 10 min, washed 3x5 min in PBS. The sample was (6). For LOH analysis 5-10 μl of PCR product was loaded on surrounded by PAP-PEN (Kiyota, Baltimore, MD). Protein 0.1x40x32 cm native 8-12% polyacrylamide gel, at 10-15 V/ block serum-free (Dako) was added for 10 min. cm in 1X Tris/borate/EDTA (TBE) buffer and for SSCP 3-5 μl For Ptch protein: primary anti-PTCH-CTP rabbit antibody of PCR product was loaded with denaturing buffer, denatured in 2X BSA/PBS (a kind gift from Dr Allen Bale) was 10 min at 55˚C and then loaded on 0.1x16x18 cm native 6-9% incubated overnight at 4˚C. For p16 protein: primary anti-p16 polyacrylamide gel at 250 V in 1X TBE buffer. mouse antibody was used according to manufacturer's DNA was visualized by the silver staining method, briefly: instructions (Dako). After washing in PBS 3x5 min secondary gels were fixed by submersion in ethanol, oxygenated with anti-rabbit or anti-mouse antibody (Dako) was added for HNO3, followed by treatment with AgNO3 and visualized with 60 min at room temperature. Slides were again washed in Na2CO3/formalin. The reaction was stopped with acetic PBS 3x5 min and PAP (Dako) was added for 60 min. After acid. washing in PBS 3x5 min DAB was added for 7 min. Slides Only the PCR products that showed aberrant SSCP pattern were then counterstained with hematoxylin and embedded in were sequenced. Before sequencing, amplified PCR products Canada balsam. Negative controls were stained by the same were purified with QIAquick purification kit (Qiagen, Hilden, protocol, omitting the primary antibody step. Instead, slides Germany), according to the manufacturer's instructions. PCR were incubated with 2% BSA/PBS for the same period of products were then sequenced in both directions by using the time as the samples. same primers and the big dye terminator cycle sequencing kit (Applied Biosystems). Sequencing analysis was performed LOH, single-strand conformation polymorphism (SSCP) and on an automatic sequencer (ABI PRISM 310 Genetic Analyzer, sequencing. DNA samples from blood leukocytes and tumor Applied Biosystems). fresh or paraffin tissues were extracted by standard methods. DNA samples were typed for eight short tandem repeat Statistical analysis. Non-parametric Spearman rank correlation polymorphisms: D9S196, PTCH intra, D9S287, D9S180, and coefficient was used to assess the degree of correlation D9S127 spanning chromosome region 9q22.3-q31, and between pairs of variables. Mann-Whitney test was used to D9S104, D9S126 and IFNA spanning 9p21 region. PCR check if there were differences in protein expression between reactions were performed in 25 μl reaction mixture containing different types and subtypes of skin tumors. Correlations and 100 ng of template DNA or 1-5 μl of crude extract from differences were considered significant if two-tailed p-value paraffin-embedded tissue prepared after microdissection, was <0.05. All statistical calculations were done using 200 μmol/l deoxynucleoside triphosphate (dNTP), 1.5 mmol/l Statistica 7.0 software (Statsoft, Inc., Tulsa, OK, USA). 1045-1050 27/2/2009 12:25 ÌÌ ™ÂÏ›‰·1047 INTERNATIONAL JOURNAL OF ONCOLOGY 34: 1045-1050, 2009 1047