Cyclin D2 Overexpression and Lack of p27 Correlate Positively and Cyclin E Inversely with a Poor Prognosis in Gastric Cancer Cases
2000; Elsevier BV; Volume: 156; Issue: 2 Linguagem: Inglês
10.1016/s0002-9440(10)64763-3
ISSN1525-2191
AutoresYasuo Takano, Yo Kato, Paul J. van Diest, Mitsunobu Masuda, Hiroyuki Mitomi, Isao Okayasu,
Tópico(s)Cancer Research and Treatments
ResumoG1 cyclins and cyclin-dependent kinase (CDK) complexes play important roles in G1 cell cycle transition, and their overexpression is implicated for neoplasia. The p27 protein (p27) negatively regulates G1 progression by binding to G1 cyclins/CDK complexes and inhibits their activity, resulting in inhibition of entry to the cell cycle. We investigated overexpression of cyclin D1 (CCND1), cyclin D2 (CCND2), cyclin E (CCNE), CDK2, and CDK4, in addition to p27, in 260 gastric cancer cases on the basis of Western blots, reverse transcriptase-polymerase chain reaction Southern blots, and immunohistochemistry to clarify the roles of these proteins in tumor progression and prognosis. Examination of 20 cases of fresh cancer and matched normal tissues demonstrated a clear tendency for increased mRNA synthesis to be more frequent than expected from protein levels, and a direct correlation between p27 protein and mRNA was not found. Immunohistochemistry demonstrated 21.5%, 34.2%, 30.4%, 44.2%, and 48.0% positivity for CCND1, CCND2, CCNE, CDK2, and CDK4, respectively, in the 260 gastric cancer cases. Overexpression of CCND2 and CDK4 significantly correlated with tumor progression. Moreover, CCND2 cytoplasmic staining (26.2%) appeared to be strictly linked with progression, whereas nuclear staining (7.8%) demonstrated an inverse correlation. Survival curves showed CCND2 (especially cytoplasmic staining) and CDK4 positivity to be associated with a poor prognosis and CCNE positivity with a better prognosis. Tumors with high p27 labeling indices (LIs) were well differentiated, with low levels of invasion and lymph node metastasis. p27-negative cases (37.3%) demonstrated a poor prognosis. Multivariate analysis revealed positivity for CCND2 and negativity for p27 to be independent prognostic factors. There were no direct links among CCND2, CCNE, CDK4, and p27. The results indicate that CCND2 cytoplasmic localization might reflect an important physiological role in tumor progression, whereas CCNE overexpression correlates with differentiation and a good prognosis, possibly because of accumulation of inactive forms of CCNE-CDK2 complexes. Loss of p27 caused by degradation activity may affect tumor cell growth in the presence of an altered extracellular matrix, facilitating metastasis. Cell-cycle-regulatory proteins appear to work independently. G1 cyclins and cyclin-dependent kinase (CDK) complexes play important roles in G1 cell cycle transition, and their overexpression is implicated for neoplasia. The p27 protein (p27) negatively regulates G1 progression by binding to G1 cyclins/CDK complexes and inhibits their activity, resulting in inhibition of entry to the cell cycle. We investigated overexpression of cyclin D1 (CCND1), cyclin D2 (CCND2), cyclin E (CCNE), CDK2, and CDK4, in addition to p27, in 260 gastric cancer cases on the basis of Western blots, reverse transcriptase-polymerase chain reaction Southern blots, and immunohistochemistry to clarify the roles of these proteins in tumor progression and prognosis. Examination of 20 cases of fresh cancer and matched normal tissues demonstrated a clear tendency for increased mRNA synthesis to be more frequent than expected from protein levels, and a direct correlation between p27 protein and mRNA was not found. Immunohistochemistry demonstrated 21.5%, 34.2%, 30.4%, 44.2%, and 48.0% positivity for CCND1, CCND2, CCNE, CDK2, and CDK4, respectively, in the 260 gastric cancer cases. Overexpression of CCND2 and CDK4 significantly correlated with tumor progression. Moreover, CCND2 cytoplasmic staining (26.2%) appeared to be strictly linked with progression, whereas nuclear staining (7.8%) demonstrated an inverse correlation. Survival curves showed CCND2 (especially cytoplasmic staining) and CDK4 positivity to be associated with a poor prognosis and CCNE positivity with a better prognosis. Tumors with high p27 labeling indices (LIs) were well differentiated, with low levels of invasion and lymph node metastasis. p27-negative cases (37.3%) demonstrated a poor prognosis. Multivariate analysis revealed positivity for CCND2 and negativity for p27 to be independent prognostic factors. There were no direct links among CCND2, CCNE, CDK4, and p27. The results indicate that CCND2 cytoplasmic localization might reflect an important physiological role in tumor progression, whereas CCNE overexpression correlates with differentiation and a good prognosis, possibly because of accumulation of inactive forms of CCNE-CDK2 complexes. Loss of p27 caused by degradation activity may affect tumor cell growth in the presence of an altered extracellular matrix, facilitating metastasis. Cell-cycle-regulatory proteins appear to work independently. Cell cycle transitions are regulated by a series of G1 cyclins (cyclin Ds and E) and cyclin-dependent kinase (CDK) complexes. The cyclin Ds (CCNDs)-CDK4 and cyclin E (CCNE)-CDK2 complexes are catalytically active during late G1 and are linked to regulation of G1/S progression in mammalian cells.1Peters G The D-type cyclins and their role in tumorigenesis.J Cell Sci. 1994; 8: 89-96Crossref Google Scholar, 2Sherr CJ Mammalian G1 cyclins.Cell. 1993; 73: 1059-1065Abstract Full Text PDF PubMed Scopus (1996) Google Scholar, 3Luger H Reed SI Translational control of p27 accumulation during the cell cycle.Science. 1996; 271: 1861-1864Crossref PubMed Scopus (823) Google Scholar mRNA transcription of CCNDs and protein synthesis is highest in middle-late G1 and lowest during S phase.2Sherr CJ Mammalian G1 cyclins.Cell. 1993; 73: 1059-1065Abstract Full Text PDF PubMed Scopus (1996) Google Scholar Cyclin D1 (CCND1), D2 (CCND2), and D3 (CCND3. map to different chromosomes but have an identical cyclin box and PEST sequence.1Peters G The D-type cyclins and their role in tumorigenesis.J Cell Sci. 1994; 8: 89-96Crossref Google Scholar, 4Baldin V Lukas J Marcrote MJ Pagano M Draetta G Cyclin D1 is a nuclear protein required for cell cycle progression in G1.Genes Dev. 1993; 7: 812-821Crossref PubMed Scopus (1445) Google Scholar CCND1 has been proposed as a protooncogene because its derangement contributes to the uncontrolled cell growth characteristic of tumors.5Keyomarsi K Pardee AB Redundant cyclin overexpression and gene amplification in breast cancer cells.Proc Natl Acad Sci USA. 1993; 90: 1112-1116Crossref PubMed Scopus (508) Google Scholar Overexpression and rearrangement of the CCND1 have been reported to be associated with tumor progression and/or poor prognosis in many different tumor types, such as the carcinomas of breast,6Gillett C Fantle V Smith R Fisher C Bartek J Dickson C Barnes D Peters G Amplification and overexpression of cyclin D1 in breast cancer detected by immunohistochemical staining.Cancer Res. 1994; 54: 1812-1817PubMed Google Scholar esophagus,7Gramlich TL Fritsch CR Maurer D Eberle M Gansler TS Differential polymerase chain reaction assay of cyclin D1 gene amplification in esophageal carcinoma.Diagn Mol Pathol. 1994; 3: 255-259Crossref PubMed Scopus (52) Google Scholar and pancreas,8Gansauge S Gansauge F Ramadani M Stobbe H Rau B Harada N Beger HG Overexpression of cyclin D1 in human pancreatic carcinoma is associated with poor prognosis.Cancer Res. 1997; 57: 1634-1637PubMed Google Scholar as well as mantle cell lymphomas.9Nakamura S Yatabe Y Seto M Cyclin D1 overexpression in malignant lymphomas.Pathol Int. 1997; 47: 421-429Crossref PubMed Scopus (44) Google Scholar The commonest genetic abnormality affecting CCND1 is considered to be DNA amplification, resulting in increased transcription.1Peters G The D-type cyclins and their role in tumorigenesis.J Cell Sci. 1994; 8: 89-96Crossref Google Scholar, 2Sherr CJ Mammalian G1 cyclins.Cell. 1993; 73: 1059-1065Abstract Full Text PDF PubMed Scopus (1996) Google Scholar There has been no report concerning any correlation between overexpression of CCND2, normally expressed in B lymphocytes, and tumor progression, although colorectal tumors may occasionally demonstrate amplification.10Leach FS Elledge SJ Sherr CJ Willson JKV Markowitz S Kinzler KW Vogelstein B Amplification of cyclin genes in colorectal carcinomas.Cancer Res. 1993; 53: 1986-1989PubMed Google Scholar It is possible that CCND2 resides in a chromosomal region that does not readily undergo amplification.11Sherr CJ Cancer cell cycle.Science. 1996; 274: 1672-1677Crossref PubMed Scopus (5003) Google Scholar However, it is considered to be a protooncogene because the immortalization of primary B lymphocytes by the Epstein-Barr virus is accompanied by transcriptional activation of CCND212Sinclair AJ Palmero I Peters G Farrell PJ EBNA-2 and EBNA-LP cooperate to cause G0 to G1 transition during immortalization of resting human B-lymphocytes by Epstein-Barr virus.EMBO J. 1994; 13: 3321-3328PubMed Google Scholar Overexpression of CCNE has been reported to be a poor prognostic factor in cancers of various organs,13van Diest PJ Michalides RJAM Jannink I van der Valk P Peterse HL de Jomg JS Meijer CJLM Baak JPA Cyclin D1 in invasive breast cancer; correlations, and prognostic value.Am J Pathol. 1997; 150: 705-711PubMed Google Scholar, 14Sutter T Doi S Carnevale KA Arber N Weinstein IB Expression of cyclin D1 and E in human colon adenocarcinomas.J Med. 1997; 28: 285-309PubMed Google Scholar and in this respect it is similar to CCND1. Over 10% of transgenic mice overexpressing human CCNE spontaneously developed mammary carcinoma.15Bortner DM Rosenberg MP Induction of mammary gland hyperplasia and carcinomas in transgenic mice expressing human cyclin E.Mol Cell Biol. 1997; 17: 453-459Crossref PubMed Scopus (203) Google Scholar Akama et al stated that CCNE amplification and overexpression correlate with gastric cancer progression without any linkage to amplification of CCND1.16Akama Y Yasui W Yokozawa H Kuniyasu H Kitahara K Ishikawa T Tahara E Frequent amplification of the cyclin E gene in human gastric carcinoma.Jpn J Cancer Res. 1995; 86: 617-621Crossref PubMed Scopus (146) Google Scholar CDK4, one of the catalytic partners for CCNDs, also shows DNA amplification, presumably leading to increased activity of CCND-dependent kinase complexes.17Khatib ZA Matsushime H Valentine M Shapiro DN Sherr CJ Look AT Coamplification of the CDK4 gene with MDM2 and GLI in human sarcomas.Cancer Res. 1993; 53: 5535-5541PubMed Google Scholar It has been described that an association with CDK4 is required for CCND1 function in G1 progression.18Tam SW Theodoras AM Shay JW Draetta GF Pagano M Differential expression and regulation of cyclin D1 protein in normal and tumor cells: association with cdk4 is required for cyclin D1 function in G1 progression.Oncogene. 1994; 9: 2663-2674PubMed Google Scholar In addition, CDK2 is a catalytic partner of CCNE and is thought to be a poor prognostic factor in combination with CCNE in various organs.19Kitahara K Yasui W Kuniyasu H Yokozaki H Akama Y Yunotani S Hisatsugu T Tahara E Concurrent amplification of cyclin E and CDK2 genes in colorectal carcinomas.Int J Cancer. 1995; 4: 25-28Crossref Scopus (116) Google Scholar, 20Kawana H Tamaru J Tanaka T Hirai A Saito Y Kitagawa M Mikata A Harigaya K Kuriyama T Role of p27Kip1 and cyclin-dependent kinase 2 in the proliferation of non-small cell lung cancer.Am J Pathol. 1998; 153: 505-513Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar The p27 protein (p27), identified from the observation that transforming growth factor-β and cell-cell contact prevent activation of the CCNE-CDK2 complex during the G1 phase,21Koff A Ohtsuki M Polyak K Roberts JM Massague J Negative regulation of G1 in mammalian cells: inhibition of cyclin E-dependent kinase by TGF-β.Science. 1993; 260: 536-539Crossref PubMed Scopus (513) Google Scholar is a 27-kd heat-stable protein present in extracts from cells made quiescent by these signals. It has been implicated in the negative regulation of G1 progression in response to a number of antiproliferative signals by binding tightly to CCNE-CDK2 and CCND1/CCND2-CDK4 complexes and inhibiting their CDK activity in a stoichiometric manner.22Polyak K Kato J Solomon MJ Sherr CJ Massague J Roberts JM Koff A p27kip1, a cyclin-cdk inhibitor, links transforming growth factor-β and contact inhibition to cell cycle arrest.Genes Dev. 1994; 8: 9-22Crossref PubMed Scopus (1840) Google Scholar p27 activity peaks in G1, and it is thought that protein abundance rather than modification is the primary mode of its regulation. The amount of p27 varies, whereas the abundance of p27 mRNA remains unchanged. It is believed that the posttranslational alteration of p27 levels is achieved in part by a mechanism of translational control.3Luger H Reed SI Translational control of p27 accumulation during the cell cycle.Science. 1996; 271: 1861-1864Crossref PubMed Scopus (823) Google Scholar In addition, a large amount of p27 is found in quiescent cells, suggesting that p27 maintains cells in G0.23Nourse J Firpo E Flanagan M Coats S Polyak C Lee M Massague J Crabtree G Roberts J Interleukin-2-mediated elimination of p27Kip1 cyclin-dependent kinase inhibitor prevented by rapamycin.Nature. 1994; 372: 570-573Crossref PubMed Scopus (905) Google Scholar p27 levels decrease when cells reenter the cell cycle, mostly because of ubiquitin-proteasome-dependent degradation.24Pagano M Tam SW Theodoras AM Romero-Beer P Del Sal G Chau V Yew R Draetta G Rolfe M Role of the ubiquitin-proteasome pathway in regulating abundance of the cyclin-dependent kinase inhibitor p27.Science. 1995; 269: 682-685Crossref PubMed Scopus (1739) Google Scholar Recent research using p27 knock-out mice indicated that p27 is a tumor suppressor gene, because −/− homozygotes often develop pituitary tumors spontaneously, a characteristic similar to that of mice with the Rb mutation.25Nakayama K Ishida N Shirane M Inomata A Inoue T Shishido N Horii I Dennis YL Nakayama K Mice lacking p27Kip1 display increased body size, multiple organ hyperplasia, retinal dysplasia, and pituitary tumors.Cell. 1996; 85: 707-720Abstract Full Text Full Text PDF PubMed Scopus (1482) Google Scholar, 26Kiyokawa H Kineman RD Manova-Todorova KO Soares VC Hoffman ES Ono M Khanam D Hayday AC Frohman LA Koff A Enhanced growth of mice lacking the cyclin-dependent kinase inhibitor function of p27Kip1.Cell. 1996; 85: 721-732Abstract Full Text Full Text PDF PubMed Scopus (1153) Google Scholar Moreover, it has been reported that lack or low expression of p27 is associated with aggressive progression and a poor prognosis in human breast,27Catzavelos C Bhattacharya N Ung YC Wilson JA Roncari L Sandhu C Shaw P Yeger H Morava-Protzner I Kapusta L Franssen E Pritchard KI Slingerland JM Decreased levels of the cell-cycle inhibitor p27Kip1 protein: prognostic implications in primary breast cancer.Nature Med. 1997; 3: 227-230Crossref PubMed Scopus (794) Google Scholar non-small lung,28Esposito V Baldi A De Luca A Groger AM Loda M Giordano GG Caputi M Baldi F Pagano M Giordano A Prognostic role of the cyclin-dependent kinase inhibitor p27 in non-small cell lung cancer.Cancer Res. 1997; 57: 3381-3385PubMed Google Scholar colorectum,29Loda M Cukor Tam SW Lavin P Firentino M Draetta GF Jessup M Pagano M Increased proteasome-dependent degradation of the cyclin-dependent kinase inhibitor p27 in aggressive colorectal carcinomas.Nature Med. 1997; 3: 231-234Crossref PubMed Scopus (992) Google Scholar stomach,30Yasui W Kudo Y Demba S Yokozaki H Tahara E Reduced expression of cyclin-dependent kinase inhibitor p27Kip1 is associated with advanced stage and invasiveness of gastric carcinomas.Jpn J Cancer Res. 1997; 88: 625-629Crossref PubMed Scopus (136) Google Scholar, 31Mori M Mimori K Shiraishi T Tanaka S Ueo H Sugimachi K Akiyoshi T p27 expression and gastric carcinoma.Nature Med. 1997; 3: 593Crossref PubMed Scopus (266) Google Scholar and prostate cancers32De Marzo AM Meeker AK Epstein JI Coffy DS Prostate stem cell compartments: expression of the cell cycle inhibitor p27Kip1 in normal, hyperplastic, and neoplastic cells.Am J Pathol. 1998; 153: 911-919Abstract Full Text Full Text PDF PubMed Scopus (195) Google Scholar, 33Tsihlias J Kapsta LR DeBoer G Morava-Protzer I Zbieranowski I Bhattacharya N Catzavelos GC Klotz LH Slingerland JM Loss of cyclin-dependent kinase inhibitor p27Kip1 is a novel prognostic factor in localized human prostate adenocarcinoma.Cancer Res. 1998; 58: 542-548PubMed Google Scholar and lymphoma.34Sanchez-Beato M Saez AI Martinez JC Mateo MS Sanchez LS Villuendas R Troncone G Piris MA Cyclin-dependent kinase inhibitor p27kip1in lymphoid tissue: p27kip1 expression is inversely proportional to the proliferative index.Am J Pathol. 1997; 151: 151-160PubMed Google Scholar In the present study a comparison of these cell-cycle-related factors was made, using a large series of gastric cancers to shed light on their importance for malignancy. Two hundred and sixty cases of surgically resected gastric cancers were selected from the patients charts at Kitasato University Hospital from 1989 to 1991. All patients underwent total or subtotal gastrectomy, and in 256 cases regional lymph nodes were removed. A total of 253 cases could be followed up until November 1997. The patient series comprised 167 males and 93 females, with a mean age of 60.9 (from 28 to 87) years. The median follow-up time for the surviving patients was 65.3 months. Histopathological diagnoses were made, using the modified classification described elsewhere,35Sugano H Nakamura K Kato Y Pathological studies of human gastric cancer.Acta Pathol Jpn. 1982; 32: 329-347PubMed Google Scholar of well (78 cases), moderately (42), and poorly (140) differentiated adenocarcinomas. Each of the three types were classified into four groups, in accordance with the tumor grading, as follows: limited to the mucosa (m.; 50), invasion within the submucosa (s.m.; 58), proper muscle (m.p.; 32), and beyond (s.; 120). They were subclassified as superficial (m.; 50) or truly invasive (s.m., m.p. and s.; 210). Four-micron-thick sections of formalin-fixed and paraffin-embedded tissue were used for the study, with a combination of the standard labeled streptavidin-biotin-peroxidase (LSAB kit; Dako, Glostrup, Denmark) method and microwave oven heating (500 W, 5 minutes for trhee times in 10 mmol/L citric acid at pH 6.0). DCS-6 (×10; YLEM, Rome, Italy), cyclin E (×40; Novocastra Laboratories, Newcastle, UK), Cdk2 (×100; Transduction Laboratories, Lexington, MA), and Kip1 (×1000 dilution; Transduction Laboratories) mouse monoclonal antibodies for the CCND1, CCNE, CDK2, and p27 proteins and cyclin D2 (c-17) and cdk4 (c-22) rabbit polyclonal antibodies (×1000 dilutions; Santa Cruz Biotechnology, Santa Cruz, CA) for CCND2 and CDK4 proteins were applied. Counterstaining was achieved with 0.3. methyl green solution. While cut-off values were basically determined on the basis of our own large scale of studies, those for CCND1 and CCNE positivity were defined as 5% of the cells, in accordance with an earlier report.36Gillet C Smith P Gregory W Richards M Millis R Peters G Cyclin D1 and prognosis in human breast cancer.Int J Cancer (Pred Oncol). 1996; 69: 92-99Crossref PubMed Scopus (298) Google Scholar Cases were defined as positive for CCND2 immunostaining when over 15% of the cells were stained in each section as described previously.37Takano Y Kato Y Masuda M Ohshima Y Okayasu I Cyclin D2 but not cyclin D1 overexpression closely correlates with gastric cancer progression and prognosis.J Pathol. 1999; 189: 194-200Crossref PubMed Scopus (91) Google Scholar The same criterion was used for CDK2 and CDK4. Gastric cancer cases positive for CCND1, CCND2, CCNE, CDK2, and CDK4 proved by Western blotting were included in each run as positive controls. Samples of gastric normal epithelium similarly established to be negative for all of these proteins were used as negative controls. In areas of cancer tissue, percentages of p27-positive cells were calculated after counting randomly at least 1000 nuclei in the mucosal site in a high-power field (×400). The percentage was used as p27 labeling indices. In 197 cases, p27 indices were also calculated from examination of more than five normal crypts adjacent to cancerous lesions. We used a cut-off of 30% for plus or minus, in accordance with an earlier report.38Yang RM Naitoh J Murphy M Wang H Phillipson J deKernion JB Loda M Reiter RE Low p27 expression predicts poor disease-free survival in patients with prostatic cancer.J Urol. 1998; 159: 941-945Abstract Full Text Full Text PDF PubMed Scopus (277) Google Scholar Lymphocytes in the same specimen were used as internal controls. Nonimmune rabbit serum was used for negative control sections. Statistical analyses were made of data of three gastroenteropathologists concerned with inter- and intraobserver reproducibility. A close correlation was noted in the cases of CCND1, CCND2, CCNE, CDK2, CDK4, and p27 LI (P < 0.0001). Twenty cases of fresh gastric cancers, all invasive, and matched normal epithelium samples frozen at −80°C were examined. Tissue samples were homogenized in 0.01 mol/L phosphate-buffered saline (PBS) solution and centrifuged (12,000 rpm, 30 minutes). The supernatants were mixed with 62.5 mmol/L Tris-HCl buffer (pH 6.8. containing 2% sodium dodecyl sulfate (SDS), 5% 2-mercaptoethanol, 7. glycerol, and 0.01% bromophenol blue and boiled for 10 minutes. Proteins (40 μg protein) were electrophoresed on 8. SDS-polyacrylamide gels at 30 mA for 3 hours and transferred onto 0.45-mm polyvinylidene fluoride membranes (Immobilon-P; Millipore, Bedford, MA), using a semidry system (Biocraft, Tokyo, Japan) at 200 mA for 30 minutes. Membranes were blocked with 15% skimmed milk in phosphate-buffered saline and then incubated with antibodies (double the dilutions used for immunohistochemistry) at 4°C overnight, followed by exposure to horseradish peroxidase-conjugated rabbit anti-mouse IgG and swine anti-rabbit IgG (Dako). Specific binding of antibody was determined using enhanced chemiluminescence of X-ray films (Fuji RX-U). All Western blots were repeated in triplicate for confirmation. To evaluate overexpression on X-ray film, data for both cancer and matched normal tissues were read into a computer (Power Macintosh 7600/120) with a scanner (Epson GT-9500), and densities of bands were calculated with NIH Image software to obtain cancer-to-normal ratios. The cut-off ratio was 2.0 for a positive. mRNA was isolated by the phenol and guanidinium thiocyanate method, and 1 μg of aliquots was dissolved in 20 μl of reaction buffer containing reverse transcriptase (RAV-2; Takara, Ohtu, Japan), random primers (Takara), and ribonuclease inhibitor (Takara). After incubation at 37°C for 1 hour, c-DNAs were obtained and amplified in a final volume of 100 μl of reaction mixture containing 2 U of Taq polymerase (Takara), using 1 μmol/L primers. The polymerase chain reaction (PCR) conditions were 94°C for 1 minute, 55°C for 1 minute, and 72°C for 2 minutes for 25 cycles. Primers, oligonucleotide probes, hybrid temperatures, and lengths of PCR products are listed in Table 1. All primers and oligonucleotide probes except the CCND1 primers were originally designed by ourselves. The PCR products were electrophoresed through 2% agarose gels containing ethidium bromide and photographed under UV light. After gels were denatured, neutralized, and Southern blot transferred onto nylon membranes (Hybon N+; Amersham, Buckinghamshire, UK), hybridization was performed using 10 pmol/ml of digoxygenin-labeled oligonucleotide probes against the respective samples at 48°C to 64°C, adjusted for each probe (Table 1), for 16 hours. After the membrane was washed with 2× SSC, 0.1% SDS and 0.1× SSC, 0.1% SDS at the same temperature as the hybrid temperature, luminescence was detected with X-ray films (Fuji RX-U), employing a Dig luminescent detection kit for nucleic acids (Boehringer Mannheim).Table 1Sequences for Primers and Probes UsedGenesSequences for primers and probesPCR cycles Hib. temp. LengthCCND1Primers5′-CTG/GAG/CCC/GTG/AAA/AAG/AGC-3′255′-CTG/GAG/AGG/AAG/CGT/GTG/AGG-3′58°CProbe5′-GCT/GTG/CAT/CTA/CAC/CGA/CGG/CTC/CAT/CCG-3′415 bpsCCND2Primers5′-TGC/TGT/CTG/CAT/GTT/CCT/GGC/CTC-3′255′-ATC/TTA/GCC/AGC/AGC/TCA/GTC/AGG-3′64°CProbe5′-GCT/GTC/TCT/GAT/CCG/CAA/GCA/TGC/TCA/GAC-3′390 bpsCCNEPrimers5′-ATG/GTA/TAC/TTG/CTG/CTT/CGG/CC-3′255′-AGA/ACG/TGG/AGC/AGG/CGC/GCA/ACT-3′48°CProbe5′-GTA/TCA/TTT/CTC/GTC/ATC/TGA/ATT/GAT/GCA-3′397 bpsCDK2Primers5′-TCC/GAG/AGA/TCT/CTC/TGC/TTA/AGG-3′255′-CGA/GTC/ACC/ATC/TCA/GCA/AAG/ATG-3′62°CProbe5′-CAT/GGA/TGC/CTC/TGC/TCT/CAC/TGG/CAT/TCC-3′452 bpsCDK4Primers5′-TCC/GAG/AGA/TCT/CTC/TGC/TTA/AGG-3′255′-AGC/TCG/GTA/CCA/GAG/TGT/AAC/AAC-3′64°CProbe5′-GCT/GAT/GGA/CGT/CTG/TGC/CAC/ATC/CCG/AAC-3′396 bpsp27Primers5′-AAC/CTC/TTC/GGC/CCG/GTG/GAC/CAC-3′255′-GTC/TGC/TCC/ACA/GAA/CCG/GCA/TTT-3′62°CProbe5′-GCG/ACC/TGC/AAC/CGA/CGA/TTC/TTC/TAC/TCA-3′471 bpsb-actinPrimers5′-TGA/TGA/TAT/CGC/CGC/GCT/CGT/CGT-3′25 cycle5′-CAC/AGC/CTG/GAT/AGC/AAC/GTA/CAT-3′60°CProbe5′-CAT/CGA/GCA/CGG/CAT/CGT/CAC/CAA/CTG/GGA-3′412 bpsHib. temp., hybridization temperature. Length, length of PCR products. Open table in a new tab Hib. temp., hybridization temperature. Length, length of PCR products. Statistical analysis was performed using the Mann-Whitney U test and the χ2Sherr CJ Mammalian G1 cyclins.Cell. 1993; 73: 1059-1065Abstract Full Text PDF PubMed Scopus (1996) Google Scholar test for demonstration of an interrelation among CCND1, CCND2, CCNE, CDK2, CDK4, and p27 and other parameters. The survival period was defined as the time from the operation to cancer death. Multivariate survival analysis was performed with the Cox regression model to assess the additional prognostic value of the different variables, using Statview J4.5 and BMDP software. Figure 1 shows the results of Western blotting with CCND2 and CCNE for 20 cases of matched gastric cancer and normal tissues. Single or doublet CCND2 bands (phosphorylated and unphosphorylated) were found around 34 kd. Two major bands of 49 and 43 kd, corresponding to the two alternatively spliced forms, were obtained for CCNE. Figure 2 shows the results of reverse transcriptase-polymerase chain reaction (RT-PCR) Southern blotting for G1 cyclins, CDKs, and p27 in cancer tissues, compared with β-actin at the bottom. CCND1 protein overexpression was noted in 15% of cases and mRNA overexpression in 35%. The respective figures were 30% and 50% for CCND2, 50% and 55% for CCNE, 55% and 75% for CDK2, and 35% and 45% for CDK4. There was a clear tendency for mRNA positivity to be more frequent than protein synthesis. There was no case demonstrating protein overexpression without mRNA overexpression, except for that of CCND2. In normal tissues, slight protein or mRNA expression was sometimes found, but this was generally very weak. However, there were two discrepant cases (2 and 15) between RNA and protein expression for CCND2. Also, even in normal tissue, overexpression was observed in cases 2, 7, and 11. There were no discrepancies between the results of Western blotting and those of immunohistochemistry.Figure 2Clear single bands can be seen for G1 cyclins and CDKs with the expected lengths. RT-PCR of β-actin proves the adequacy of the mRNA used. mRNA overexpression is evident in 35% (protein overexpression, 15%) for CCND1, 50% for CCND2 (35%), 55% for CCNE (50%), 75% for CDK2 (55%), and 45% for CDK4 (35%).View Large Image Figure ViewerDownload Hi-res image Download (PPT) Figure 3 shows Western blotting results for p27 in fresh cancer (T) and matched normal tissues (N). Single or doublet bands around 27 kd were obtained in all cases except for normal tissue No6. T/N density ratios calculated with NIH Image software are listed under the case numbers. There were nine cases with values under 1.0. The results of immunohistochemistry were not always the same as with Western blotting. RT-PCR Southern blotting revealed a single band of weak to strong density, proving intact production of p27 mRNA in all cases. However, there was no significant linkage between protein and mRNA production. Immunohistochemistry of CCND1 and CCNE demonstrated 56 (21.5%. and 79 (30.4%) of 260 gastric cancer cases to show nuclear staining (Figure 4, a and c). CCND2 was positive in nuclei only in 20 (7.8%), in cytoplasm only in 68 (26.2%), and both in one (0.4%), making a total of 89 cases (34.2%). Characteristic cytoplasmic staining at the perimembranes was often recognized (Figure 4b). CDK2 and CDK4 demonstrated nuclear or both nuclear and cytoplasmic staining in 115 (44.2%) and 125 (48.0%) cases (Figure 4, d and e). Table 2 summarizes data for correlations among G1 cyclin and CDK positivity and clinicopathological data. Although the male and female positive ratios did not differ for either CCND2 or CCNE, the mean age of positive cases was significantly higher in both cases (P = 0.0251 and P = 0.0296). Statistically significant differences were noted with CCND1 positive versus negative for less lymph node metastasis (P = 0.0009), with CCND2 for well differentiated (P < 0.0001) and vessel invasion of cancer cells (P = 0.0008), with CCNE for well differentiated (P = 0.0058) and less depth of cancer invasion (P = 0.0256), with CDK2 for well differentiated (0.0234), and with CDK4 for vessel invasion (P = 0.0053). With regard to CCND2, nuclear stained lesions (21 cases) tended to be well differentiated (P = 0.0006), with a lower depth of cancer invasion (P = 0.0274), fewer lymph node metastases (P = 0.002), and less vessel invasion (P = 0.0659). In contrast, cytoplasmic staining was associated with greater depth of invasion (P = 0.0020), more lymph node metastasis (P = 0.0007), and vessel invasion (P < 0.0001),
Referência(s)