PMC:2948432 / 8738-19575 JSONTXT

{"project":"2_test","denotations":[{"id":"20140953-18602256-43951734","span":{"begin":701,"end":705},"obj":"18602256"},{"id":"20140953-19887478-43951735","span":{"begin":2081,"end":2085},"obj":"19887478"},{"id":"20140953-19887478-43951736","span":{"begin":2196,"end":2200},"obj":"19887478"},{"id":"20140953-12060695-43951737","span":{"begin":2427,"end":2431},"obj":"12060695"},{"id":"20140953-15714058-43951738","span":{"begin":5263,"end":5267},"obj":"15714058"},{"id":"20140953-18370098-43951739","span":{"begin":5276,"end":5280},"obj":"18370098"},{"id":"20140953-19697059-43951740","span":{"begin":5297,"end":5301},"obj":"19697059"}],"text":"MATERIALS AND METHODS\n\nPatients\nEthical approval for this study was obtained from the Multi-Regional Ethical Committee (MREC) and from the patient information advisory group (PIAG) and all living patients gave informed written consent to the use of their stored tissues for research. Melanoma cases had all been diagnosed at least 3 years previously. Recruitment was irrespective of family history. Cases were then eligible as “relapsers” if relapse occurred after 3 years or as “nonrelapsers” if they were free of relapse. Participants had a tumor thickness greater than 0.75 mm and were recruited between May 2000 and January 2005. Full details of the study were reported previously (Beswick et al.,2008). There were 424 patients eligible for the study, of which 66% (278) participated. The median Breslow thickness was 1.6 mm (range 0.8-20). From the 278 participating patients, 116 (74 relapsers and 42 nonrelapsers) had FFPE primary tumor samples that were available for further sampling and DNA extraction. There was no selection of these blocks, other than that we used blocks which could be traced.\n\nHistology\nSections from the primary tumors were examined (blind to relapse status) by one pathologist according to protocol (AB). The following were recorded: Breslow thickness, site of the primary tumor, histological sub-type, presence of ulceration, and mitotic rate in three categories (0, 1 to 6 and more than 6 per mm2).\n\nDNA Extraction\nDNA was extracted from 116 FFPE primary melanoma tumors, sampled horizontally at the advancing edge of the tumor, in the vertical growth phase, using a 0.8 mm × 2 mm core biopsy needle and haematoxylin and eosin stained slides as a guide. The intent was to choose tissue representative of the deepest part of the tumor but which was sufficiently surrounded by tumor that the sample contained minimal normal stroma and inflammatory cell infiltrate. Horizontal sections of cores were taken during development of the methodology to ensure the technique allowed minimal sampling of normal tissue as described previously (Conway et al.,2009). DNA extraction from cores was carried out using the QIAamp DNA Mini kit (Qiagen, Sussex, UK) (Conway et al.,2009).\n\nCopy Number Analysis of Chromosome 9p21\nGene dosage ratios for 12 CDKN2A/CDKN2B locus sites and 11 other 9p gene sites were determined using the 9p21 MLPA kit (P024 MRC-Holland, Amsterdam, the Netherlands) (Schouten et al.,2002). The genes included in this screen were TEK, ELAV2, CDKN2B, CDKN2A, MTAP, KIAA1354, INFW1, INFB1, MLLT3, and DOCK8 (Fig. 1A). The kit was used in accordance with instructions for all experiments but in 1/4 volumes of those recommended by the supplier, based on previous optimization in our laboratory: 30–100 ng of extracted tumor DNA was denatured and target gene probes were hybridized to the target DNA prior to probe ligation in the presence of ligase-65. The ligation products were subject to polymerase chain reaction (PCR) amplification performed on a GeneAmp PCR System 9700 Thermal Cycler (Applied Biosystems, Warrington, UK) with a hot-start PCR program. MLPA fragments were visualized on an ABI 3130XL Automated DNA Sequencer with a 36 cm capillary array, ABI POP-7 polymer, and GeneScan-ROX 500 size standards (Applied Biosystems). Peak detection analysis has been automated using ABI PRISM Genescan® Analysis software version 3.1 (Applied Biosystems) and GeneMarker software (Softgenetics, State College, Pennsylvania, USA). In each set of experiments, one negative control (no DNA) sample and four normal control samples (human genomic DNA with normal gene dosage at 9p from four different individuals) were included.\nFigure 1 MLPA for analysis of gene dosage at chromosome 9p21. A, a diagrammatical representation of the 9p21 region covered by the P024A MLPA kit. Locations of probed genes are shown in green and MLPA probes in orange with probe names outlined in bold. Probe groupings used in data analysis are represented by orange bars with black dots at top of the picture. B, gene dosage ratios for a representative primary melanoma sample showing an almost total loss of gene dosage (representing ≥80% loss) from MTAP exon 6 to CDKN2B intron.\n\nMLPA Data Analysis\nGene dosage analysis was automated using the MLPA analysis program included with GeneMarker software version 1.6 (Softgenetics, Pennsylvania, USA) according to the manufacturer's instructions. Data were normalized using the “population normalization” mode (as recommended by the MLPA kit manufacturers for analysis of tumor DNA). Peak heights were normalized according to the median height of all test and control peak heights of similar fragment size. Normalized peak heights were then compared to a synthetic control sample (average of peak heights from four human genomic DNA samples with normal gene dosage at 9p) with default analysis parameters to determine gene dosage ratios where the median point within the data set is considered to be 1, and a gene dosage ratio for each probe region was calculated with reference to the synthetic control (Fig. 1B). In previous publications using MLPA on FFPE-derived DNA, gene dosage cut-offs of 0.7 for loss and 1.3 for gain have been used according to manufacturer's instructions (van Dijk et al.,2005; Takata,2008; Buffart et al.,2009) (and author correspondence with MRC-Holland). In our analysis, gene dosage was treated as a continuous variable to account for the effect of any possible contaminating normal DNA.\n\nMutation Analyses\nGene fragments of hotspot mutation regions in BRAF (exon 15) and NRAS (exon 2) were amplified by PCR in separate reactions to screen for common mutations found in melanomas. Standard PCR reactions were carried out using Amplitaq Gold DNA polymerase in 1× PCR buffer (Applied Biosystems) according to manufacturer's instructions. Primers: BRAF exon 15 Forward: 5′-TCA TAATGCTTGCTCTGATAGGA and Reverse: 5′-GGCCAAAAATTTAATCAGTGGA (annealing temperature 59°C); and NRAS exon 2 Forward: 5′-GGTGAAACCTGTTTGTTGGA and Reverse: 5′-TTCAGAACACAAAGATCATC (55°C).\nBRAF and NRAS PCR products were sequenced in both directions using an ABI3100 Automated DNA Sequencer with a 36 cm capillary array, ABI POP-7 polymer and ABI Prism BigDye Terminator Cycle Sequencing Kit version 1.1 (Applied Biosystems) according to the manufacturer's instructions. Sequence analysis was carried out using CodonCode Aligner sequencing software (CodonCode Corporation, Dedham, MA, USA) and mutation detection was based on BRAF and NRAS cDNA sequences (Genbank accession nos. NM_004333 and NM_00254, respectively).\n\nCDKN2A Immunohistochemistry\nA representative set of 17 tumors from the MLPA data set were selected for immunohistochemistry using antibodies to CDKN2A to evidence the validity of the MLPA results. The samples chosen consisted of four samples with gene dosage ratios of ≤0.2 (80% gene dosage loss) at the CDKN2A promoter, or coding regions for CDKN2A; seven samples with gene dosage ratios ∼0.5 (50% gene dosage loss) at the CDKN2A promoter, coding regions for CDKN2A or intronic regions upstream of the promoter; and six samples with gene dosage ratios between 0.7 and 1.3 (normal gene dosage). Positive and negative controls for kit and antibody performance were included. The positive control was a paraffin-embedded section of bladder carcinoma with high CDKN2A expression determined by Western blot analysis. The negative control was a section from the same bladder tumor without the addition of primary antibody.\nSections (5 μm) were cut and fixed on Superfrost plus glass slides before dewaxing and rehydration. Expression of CDKN2A was examined using a CDKN2A monoclonal antibody (1:1500 for 1 hr; Ab-7; Labvision, Freemont, CA, USA) and the catalyzed signal amplification system (CSA system; DakoCytomation, Cambridgeshire, UK) according to the manufacturer's instructions. Endogenous biotin or biotin-binding proteins were blocked using the Avidin Biotin blocking kit (Vector Laboratories, Peterborough, UK) according to manufacturer's instructions and endogenous peroxidase activity was blocked using 3% hydrogen peroxide in water for 5 min (CSA system). Sections were counterstained with hematoxylin, dehydrated and mounted in Depex mounting medium (VWR International, Leicestershire, UK). Expression of CDKN2A was examined by light microscopy and scored in tumors as absent (0), expressed (1) or highly expressed (2).\n\nStatistical Methods\nGene dosage was treated as a continuous variable in the analysis (Fig. 1). To investigate the effect of dosage in relation to each of the transcripts at 9p21, five separate probe groups were created. Probe groups contained all probes located within the coding exons and promoter for each transcript. The probe groups were: (1) CDKN2B (probes: CDKN2B promoter and exon 1), (2) regions coding for P14ARF (probes: CDKN2A CpG island, CDKN2A 1β promoter and exons 1β, 2, and 3), (3) CDKN2A (probes: CDKN2A 1α promoter and exons 1α, 2, and 3), (4) MTAP (probes: MTAP exons 1, 6, and 7), and (5) the interferon gene cluster (probes: KIAA1354 [between IFNA5 and IFNA6], INFW1, and INFB1). A “CDKN2A” group containing all CDKN2A probes was created to investigate the effect of overall loss at the CDKN2A locus (probes: CDKN2A CpG island, CDKN2A 1β promoter, exon 1β, CDKN2A introns 1 and 2, CDKN2A 1α promoter plus exons 1α, 2 and 3). To investigate the effect of loss across the whole 9p21.3 region, an overall 9p21.3 group defined by all of the probes in the region was also considered. Median gene dosage ratio was used to represent the overall gene dosage ratio of the region where there was more than one probe within the region.\nA rolling average heatmap was used to represent graphically the overall gene loss. The rolling average of the gene dosage was calculated as the average gene dosage ratio of the probe itself and adjacent probes. The Wilcoxon two-sample rank test using normal approximation was performed to assess the difference between gene dosage ratio by relapse status (relapse versus no relapse), ulceration status (yes versus no), tumor site (head/neck/foot/hand versus others) and BRAF and NRAS mutation status. The Kruskal Wallis test was applied to assess the difference between tumor histological sub-type (superficial spreading versus nodular) and mitotic rate per mm2 (grouped as 0, 1–6, \u003e6). Spearman correlations (r) were used to assess association between gene dosage ratio and Breslow thickness. Fisher's exact test was used to assess the association between BRAF/NRAS mutation and relapse status, mitotic rate and ulceration. These analyses were carried out using the SAS/STAT statistical software version 9.1 for PC (Copyright, SAS Institute Inc. Cary, NC, USA). The rolling average heatmap was illustrated using heatmap.2\u003cgplots\u003e function in R version 2.9.0 (Vienna, Austria)."}

{"project":"0_colil","denotations":[{"id":"20140953-18602256-957386","span":{"begin":701,"end":705},"obj":"18602256"},{"id":"20140953-19887478-957387","span":{"begin":2081,"end":2085},"obj":"19887478"},{"id":"20140953-19887478-957388","span":{"begin":2196,"end":2200},"obj":"19887478"},{"id":"20140953-12060695-957389","span":{"begin":2427,"end":2431},"obj":"12060695"},{"id":"20140953-15714058-957390","span":{"begin":5263,"end":5267},"obj":"15714058"},{"id":"20140953-18370098-957391","span":{"begin":5276,"end":5280},"obj":"18370098"},{"id":"20140953-19697059-957392","span":{"begin":5297,"end":5301},"obj":"19697059"}],"text":"MATERIALS AND METHODS\n\nPatients\nEthical approval for this study was obtained from the Multi-Regional Ethical Committee (MREC) and from the patient information advisory group (PIAG) and all living patients gave informed written consent to the use of their stored tissues for research. Melanoma cases had all been diagnosed at least 3 years previously. Recruitment was irrespective of family history. Cases were then eligible as “relapsers” if relapse occurred after 3 years or as “nonrelapsers” if they were free of relapse. Participants had a tumor thickness greater than 0.75 mm and were recruited between May 2000 and January 2005. Full details of the study were reported previously (Beswick et al.,2008). There were 424 patients eligible for the study, of which 66% (278) participated. The median Breslow thickness was 1.6 mm (range 0.8-20). From the 278 participating patients, 116 (74 relapsers and 42 nonrelapsers) had FFPE primary tumor samples that were available for further sampling and DNA extraction. There was no selection of these blocks, other than that we used blocks which could be traced.\n\nHistology\nSections from the primary tumors were examined (blind to relapse status) by one pathologist according to protocol (AB). The following were recorded: Breslow thickness, site of the primary tumor, histological sub-type, presence of ulceration, and mitotic rate in three categories (0, 1 to 6 and more than 6 per mm2).\n\nDNA Extraction\nDNA was extracted from 116 FFPE primary melanoma tumors, sampled horizontally at the advancing edge of the tumor, in the vertical growth phase, using a 0.8 mm × 2 mm core biopsy needle and haematoxylin and eosin stained slides as a guide. The intent was to choose tissue representative of the deepest part of the tumor but which was sufficiently surrounded by tumor that the sample contained minimal normal stroma and inflammatory cell infiltrate. Horizontal sections of cores were taken during development of the methodology to ensure the technique allowed minimal sampling of normal tissue as described previously (Conway et al.,2009). DNA extraction from cores was carried out using the QIAamp DNA Mini kit (Qiagen, Sussex, UK) (Conway et al.,2009).\n\nCopy Number Analysis of Chromosome 9p21\nGene dosage ratios for 12 CDKN2A/CDKN2B locus sites and 11 other 9p gene sites were determined using the 9p21 MLPA kit (P024 MRC-Holland, Amsterdam, the Netherlands) (Schouten et al.,2002). The genes included in this screen were TEK, ELAV2, CDKN2B, CDKN2A, MTAP, KIAA1354, INFW1, INFB1, MLLT3, and DOCK8 (Fig. 1A). The kit was used in accordance with instructions for all experiments but in 1/4 volumes of those recommended by the supplier, based on previous optimization in our laboratory: 30–100 ng of extracted tumor DNA was denatured and target gene probes were hybridized to the target DNA prior to probe ligation in the presence of ligase-65. The ligation products were subject to polymerase chain reaction (PCR) amplification performed on a GeneAmp PCR System 9700 Thermal Cycler (Applied Biosystems, Warrington, UK) with a hot-start PCR program. MLPA fragments were visualized on an ABI 3130XL Automated DNA Sequencer with a 36 cm capillary array, ABI POP-7 polymer, and GeneScan-ROX 500 size standards (Applied Biosystems). Peak detection analysis has been automated using ABI PRISM Genescan® Analysis software version 3.1 (Applied Biosystems) and GeneMarker software (Softgenetics, State College, Pennsylvania, USA). In each set of experiments, one negative control (no DNA) sample and four normal control samples (human genomic DNA with normal gene dosage at 9p from four different individuals) were included.\nFigure 1 MLPA for analysis of gene dosage at chromosome 9p21. A, a diagrammatical representation of the 9p21 region covered by the P024A MLPA kit. Locations of probed genes are shown in green and MLPA probes in orange with probe names outlined in bold. Probe groupings used in data analysis are represented by orange bars with black dots at top of the picture. B, gene dosage ratios for a representative primary melanoma sample showing an almost total loss of gene dosage (representing ≥80% loss) from MTAP exon 6 to CDKN2B intron.\n\nMLPA Data Analysis\nGene dosage analysis was automated using the MLPA analysis program included with GeneMarker software version 1.6 (Softgenetics, Pennsylvania, USA) according to the manufacturer's instructions. Data were normalized using the “population normalization” mode (as recommended by the MLPA kit manufacturers for analysis of tumor DNA). Peak heights were normalized according to the median height of all test and control peak heights of similar fragment size. Normalized peak heights were then compared to a synthetic control sample (average of peak heights from four human genomic DNA samples with normal gene dosage at 9p) with default analysis parameters to determine gene dosage ratios where the median point within the data set is considered to be 1, and a gene dosage ratio for each probe region was calculated with reference to the synthetic control (Fig. 1B). In previous publications using MLPA on FFPE-derived DNA, gene dosage cut-offs of 0.7 for loss and 1.3 for gain have been used according to manufacturer's instructions (van Dijk et al.,2005; Takata,2008; Buffart et al.,2009) (and author correspondence with MRC-Holland). In our analysis, gene dosage was treated as a continuous variable to account for the effect of any possible contaminating normal DNA.\n\nMutation Analyses\nGene fragments of hotspot mutation regions in BRAF (exon 15) and NRAS (exon 2) were amplified by PCR in separate reactions to screen for common mutations found in melanomas. Standard PCR reactions were carried out using Amplitaq Gold DNA polymerase in 1× PCR buffer (Applied Biosystems) according to manufacturer's instructions. Primers: BRAF exon 15 Forward: 5′-TCA TAATGCTTGCTCTGATAGGA and Reverse: 5′-GGCCAAAAATTTAATCAGTGGA (annealing temperature 59°C); and NRAS exon 2 Forward: 5′-GGTGAAACCTGTTTGTTGGA and Reverse: 5′-TTCAGAACACAAAGATCATC (55°C).\nBRAF and NRAS PCR products were sequenced in both directions using an ABI3100 Automated DNA Sequencer with a 36 cm capillary array, ABI POP-7 polymer and ABI Prism BigDye Terminator Cycle Sequencing Kit version 1.1 (Applied Biosystems) according to the manufacturer's instructions. Sequence analysis was carried out using CodonCode Aligner sequencing software (CodonCode Corporation, Dedham, MA, USA) and mutation detection was based on BRAF and NRAS cDNA sequences (Genbank accession nos. NM_004333 and NM_00254, respectively).\n\nCDKN2A Immunohistochemistry\nA representative set of 17 tumors from the MLPA data set were selected for immunohistochemistry using antibodies to CDKN2A to evidence the validity of the MLPA results. The samples chosen consisted of four samples with gene dosage ratios of ≤0.2 (80% gene dosage loss) at the CDKN2A promoter, or coding regions for CDKN2A; seven samples with gene dosage ratios ∼0.5 (50% gene dosage loss) at the CDKN2A promoter, coding regions for CDKN2A or intronic regions upstream of the promoter; and six samples with gene dosage ratios between 0.7 and 1.3 (normal gene dosage). Positive and negative controls for kit and antibody performance were included. The positive control was a paraffin-embedded section of bladder carcinoma with high CDKN2A expression determined by Western blot analysis. The negative control was a section from the same bladder tumor without the addition of primary antibody.\nSections (5 μm) were cut and fixed on Superfrost plus glass slides before dewaxing and rehydration. Expression of CDKN2A was examined using a CDKN2A monoclonal antibody (1:1500 for 1 hr; Ab-7; Labvision, Freemont, CA, USA) and the catalyzed signal amplification system (CSA system; DakoCytomation, Cambridgeshire, UK) according to the manufacturer's instructions. Endogenous biotin or biotin-binding proteins were blocked using the Avidin Biotin blocking kit (Vector Laboratories, Peterborough, UK) according to manufacturer's instructions and endogenous peroxidase activity was blocked using 3% hydrogen peroxide in water for 5 min (CSA system). Sections were counterstained with hematoxylin, dehydrated and mounted in Depex mounting medium (VWR International, Leicestershire, UK). Expression of CDKN2A was examined by light microscopy and scored in tumors as absent (0), expressed (1) or highly expressed (2).\n\nStatistical Methods\nGene dosage was treated as a continuous variable in the analysis (Fig. 1). To investigate the effect of dosage in relation to each of the transcripts at 9p21, five separate probe groups were created. Probe groups contained all probes located within the coding exons and promoter for each transcript. The probe groups were: (1) CDKN2B (probes: CDKN2B promoter and exon 1), (2) regions coding for P14ARF (probes: CDKN2A CpG island, CDKN2A 1β promoter and exons 1β, 2, and 3), (3) CDKN2A (probes: CDKN2A 1α promoter and exons 1α, 2, and 3), (4) MTAP (probes: MTAP exons 1, 6, and 7), and (5) the interferon gene cluster (probes: KIAA1354 [between IFNA5 and IFNA6], INFW1, and INFB1). A “CDKN2A” group containing all CDKN2A probes was created to investigate the effect of overall loss at the CDKN2A locus (probes: CDKN2A CpG island, CDKN2A 1β promoter, exon 1β, CDKN2A introns 1 and 2, CDKN2A 1α promoter plus exons 1α, 2 and 3). To investigate the effect of loss across the whole 9p21.3 region, an overall 9p21.3 group defined by all of the probes in the region was also considered. Median gene dosage ratio was used to represent the overall gene dosage ratio of the region where there was more than one probe within the region.\nA rolling average heatmap was used to represent graphically the overall gene loss. The rolling average of the gene dosage was calculated as the average gene dosage ratio of the probe itself and adjacent probes. The Wilcoxon two-sample rank test using normal approximation was performed to assess the difference between gene dosage ratio by relapse status (relapse versus no relapse), ulceration status (yes versus no), tumor site (head/neck/foot/hand versus others) and BRAF and NRAS mutation status. The Kruskal Wallis test was applied to assess the difference between tumor histological sub-type (superficial spreading versus nodular) and mitotic rate per mm2 (grouped as 0, 1–6, \u003e6). Spearman correlations (r) were used to assess association between gene dosage ratio and Breslow thickness. Fisher's exact test was used to assess the association between BRAF/NRAS mutation and relapse status, mitotic rate and ulceration. These analyses were carried out using the SAS/STAT statistical software version 9.1 for PC (Copyright, SAS Institute Inc. Cary, NC, USA). The rolling average heatmap was illustrated using heatmap.2\u003cgplots\u003e function in R version 2.9.0 (Vienna, Austria)."}