Multicolor fluorescence in situ hybridization and comparative genomic hybridization reveal molecular events in lung adenocarcinomas and squamous cell lung carcinomas

Available online at

www.sciencedirect.com Biomedicine & Pharmacotherapy 63 (2009) 396e403

Original article

Multicolor fluorescence in situ hybridization and comparative genomic hybridization reveal molecular events in lung adenocarcinomas and squamous cell lung carcinomas Hua Shen, Wen Gao, Yu-jie Wu, Hai-rong Qiu, Yong-qian Shu* Cancer Biotherapy Center, The First Affiliated Hospital of Nanjing Medical University, Nanjing 210029, PR China Received 16 July 2008; accepted 24 August 2008 Available online 16 September 2008

Abstract We have used the molecular cytogenetic techniques of multicolor fluorescence in situ hybridization (M-FISH) and comparative genomic hybridization (CGH) to analyze two established lung cancer cell lines (A549, H520), 80 primary lung adenocarcinoma samples and 80 squamous cell lung carcinoma samples in order to identify common chromosomal aberrations. M-FISH revealed numerous complex chromosomal rearrangements. Chromosomes 5, 6, 11, 12, and 17 were most frequently involved in interchromosomal translocations. CGH revealed regions on 1q, 2p, 3q, 5p, 5q, 7p, 8q, 11q, 12q, 14q, 16p, 17p, 19q, 20q, 21q and 22q to be commonly over-represented and regions on 2q, 3p, 4p, 5q, 7q, 8p, 9p, 13q, 14q, and 17p to be under-represented. In lung adenocarcinomas the most common gains were found in 16p13 (50%); while in squamous cell lung carcinomas the common gains were found in 17q21 (45%) and these alterations were observed to be associated with their specific pathological subtype. In conclusion, the present study contributes to the molecular biological characterization in lung adenocarcinomas and squamous cell lung carcinomas and through evaluation of molecular events to the recently emergent focus on novel markers for lung cancer treatment. Published by Elsevier Masson SAS. Keywords: NSCLC; M-FISH; CGH

1. Introduction Lung cancer is responsible for the highest cancer-related morbidity and mortality worldwide [1]. Non-small cell lung cancer (NSCLC) comprises approximately 80% of all lung cancers, among which adenocarcinoma (AdC) and squamous cell carcinoma (SqC) are now the two most common histological subtypes. Cigarette smoking continues to be an important aetiological factor, with a clear implication of involvement in over 95% of male cases [2]. A correlation between the incidence of lung cancer in females and smoking habit has also been observed in recent years [3].

* Corresponding author. Tel.: þ86 25 83718836 6428; fax: þ86 51257700891. E-mail address: [email protected] (Y.-q. Shu). 0753-3322/$ - see front matter Published by Elsevier Masson SAS. doi:10.1016/j.biopha.2008.08.010

Due to the high incidence and mortality rates, much effort has been drawn towards investigating the cause and course of NSCLC. Although classical cytogenetic studies by banding analysis can provide an overall view of structural and numerical abnormalities, the frequent presence of karyotypic complexity has precluded more accurate interpretation. Conversely, molecular characterization by comparative genomic hybridization (CGH) and allelotyping has shown common over-representations of 1q, 3q, 5p, 8q and 20, and loss of heterozygosity (LOH) on 3p, 8p, 13q and 17p in NSCLC [4e6]. Despite the reportedly frequent genomic imbalances, the pattern of karyotypic alterations associated with AdC and SqC remains uncertain. Comparative genomic hybridization (CGH) allows screening of the entire genome in order to identify regions of gain or loss of DNA and map them according to chromosomal location [7]. A major advantage of CGH is that tissue culture

H. Shen et al. / Biomedicine & Pharmacotherapy 63 (2009) 396e403

and preparation of metaphase spreads are not required as DNA extracted from a tumor specimen is used. Previous CGH studies of NSCLC cell lines and primary tumors have identified chromosomal regions that are commonly over- or underrepresented, including gains of 1q, 3q, 5p, 7p and 8q and losses of 3p, 6q, 8p, 9p and 17p [8e10]. Multicolor fluorescence in situ hybridization (M-FISH) is a karyotyping technique that uses whole chromosome paints (WCP) to label each chromosome with a unique color [11]. Use of this technique facilitates the karyotyping of tumor cells and allows the identification of marker chromosomes and complex chromosomal rearrangements that would previously have been uninterpretable. Additionally, the use of color has opened up the field to non-cytogeneticists. Despite these technological advances, acquisition of suitable metaphase spreads is still a problem and consequently much of the work on solid tumors with M-FISH has been carried out using established cell lines. We have now conducted a comprehensive molecular cytogenetic characterization of AdC and SqC tumors. By utilizing M-FISH we analysed one established lung adenocarcinoma cell line (A549) and another squamous cell lung carcinoma cell line (H520). Eighty primary lung adenocarcinoma samples and 80 squamous cell lung carcinoma samples are detected by CGH. Our combined M-FISH and CGH analysis shows distinct patterns of genetic aberration in the two specific NSCLC subtypes, which might in turn influence the different pathogenetic pathways adopted by these tumors. It is hoped that the identification of such changes may eventually lead to improvements in diagnosis and identification of prognostic factors for lung adenocarcinomas and squamous cell lung carcinoma. 2. Materials and methods 2.1. Primary tumor samples One hundred and sixty lung cancer samples were collected from patients with histologically confirmed lung AdC (80) and lung SqC (80) at the departments of thoracic surgery at the First Affiliated Hospital of Nanjing Medical University and oncology hospital of Jiangsu province, China, after informed consent was obtained. The clinicopathologic characteristics of the samples are shown in Table 1. The genomic DNA was extracted using the PureGene DNA isolation kit (Qiagen, Hilden, Germany). The reference DNA was pooled from the 10 gender-matched (male), normal, healthy control subjects. 2.2. Lung AdC and SqC cell lines One established lung adenocarcinoma cell line was acquired from American Type Culture Collection (ATCC, Manassas, USA). It was A549 (ATCC no. CCL-185). Another squamous cell carcinoma of lung cell line was obtained from Shanghai Institute of Cell Biology, China. All of the cell lines were fully authenticated prior to distribution. Cell lines were cultured in RPMI 1640 medium supplemented with 10% v v1

397

Table 1 Clinicopathological features of the patients with non-small cell lung cancer (NSCLC) NSCLC Squamous cell carcinoma Age range (years) 42e75 Sex Male 68 Female 12 T values T1 21 T2 45 T3 12 T4 2 Lymph node involvement N0 51 N1 22 N2 7 Metastatic behaviour M0 78 M1 2

Adenocarcinoma

p

35e78

0.583

61 19

0.161

27 42 8 3

0.603

47 23 10

0.699

79 1

1.0

fetal calf serum, glutamine, penicillin, streptomycin and fungizone. 2.3. Preparation of metaphase spreads Cell cultures were treated with Colcemid (Life Technologies) prior to reaching confluence and incubated at 37  C. Cell lines received 200 ml of 10 mg ml1 Colcemid for 1 h. Cells were harvested and treated with hypotonic solution (0.075 M KCl) for 20 min, fixed in three changes of fixative solution (3:1, methanol:acetic acid) and stored at 20  C overnight. Chromosome preparations were dropped onto cleaned, humidified standard microscope slides and stored overnight at room temperature. Slides were assessed using phase contrast microscopy to ensure the presence of metaphase spreads of suitable quality before performing M-FISH. 2.4. M-FISH M-FISH was performed as described previously [12] using the SpectraVysionTM Assay system (Abbott Laboratories, Maidenhead, UK). SpectraVysion consists of a 52-probe mix of WCPs labeled with different combinations of five fluorochromes (SpectrumGold, SpectrumRed, SpectrumGreen, SpectrumAqua and SpectrumFarRed). In total, 24 combinations of no more than three fluorochromes are used, resulting in a unique color for each chromosome (22 autosomes and two sex chromosomes). Before the probe was applied, metaphase preparations were enzymatically treated with RNase and pepsin to remove RNA molecules and cytoplasmic proteins, respectively. Exact conditions varied between preparations depending on the amount of residual cytoplasm. Slides were fixed in a 1% formaldehyde solution, dehydrated, denatured in 70% formamide/2  saline sodium citrate (SSC) and dehydrated for a second time. A measure of 10 ml of SpectraVysion probe was denatured, applied to the slides and incubated at 37  C for 72 h

398

H. Shen et al. / Biomedicine & Pharmacotherapy 63 (2009) 396e403

to allow the probe to hybridize to the chromosomes. Following hybridization, slides were washed in 0.4  SSC/0.3% NP-40 (Abbott Laboratories, Maidenhead, UK) at 72  C for 2 min followed by 2  SSC/0.1% NP-40 at room temperature for 30 s. Slides were air dried and counterstained with 10 ml of 42 ng ml1 4,6-diamidino-2-phenylindole (DAPI) counterstain in antifade (Abbott Laboratories, Maidenhead, UK). Images of metaphase spreads were captured using a Nikon E800 epifluorescent microscope equipped with a Ludl 6position filter wheel and a Photometrics Sensyst cooled charge coupled device (CCD) camera. Six images were captured using filter combinations specific for each of the five fluorochromes and DAPI. All metaphases of suitable quality for analysis were captured. Spreads ideally contained long, well spread out chromosomes and did not contain large amounts of cell debris. Images were analyzed using the Quips SpectraVysion software (Abbott Laboratories, Maidenhead, UK) by examining each individual color plane (Red, Gold, Green, Aqua and Far Red) and identifying the chromosomal material present by the combinations of fluorochromes observed. The analysis of two primary lung carcinomas is shown in Fig. 1. A composite karyotype was compiled for each sample using six to nine metaphase spreads. Following ISCN recommendations (ISCN 1995), rearranged chromosomes were included when two or more cells exhibited the same structural aberration. The modal number of each whole or rearranged chromosome was used in the composite karyotype. Translocations were reported in shortened ISCN format as derivative (der) chromosomes, but the breakpoints could not be defined as identification using DAPI banding is difficult and requires cytogenetic expertise. Inversions and small deletions or amplifications are not identifiable using M-FISH. 2.5. CGH CGH was performed as described previously [13]. Briefly, labeling of tumor DNA with biotin-16-dUTP (Roche,

Mannheim, Germany) and normal reference DNA with digoxigenin-11-dUTP (Roche) was performed by standard Nick translation. The denatured DNA probes containing each 1.5e2 mg of tumor DNA and reference DNA, and 80 mg of COT-1 DNA were hybridized for 3 days to normal metaphase spreads (Vysis, Downers Grove, IL). Subsequently, the slides were washed extensively, blocked with bovine serum albumin solution, and incubated with fluorescein conjugated avidin (Vector Laboratories, Burlingame, CA) and rhodamineconjugated antidigoxigenin (Roche). The slides were then washed once again and finally mounted in an antifade medium (Vectashield, Vector Laboratories) supplemented with actinomycin (12.5 mg/ml) and DAPI (4,6-diamidino-2-phenylindole) (1.25 mg/ml) for counterstain. Image acquisition was performed on a Nikon E800 epifluorescent microscope equipped with three separate bandpass filters for DAPI, green, and red fluorescence spectrum and a Photometrics Sensyst cooled charge coupled device (CCD) camera. For each analysis, the averaged chromosome-specific green-to-red fluorescence ratios from at least 10 well-selected metaphases were plotted using the Quips CGH software (distributed through Applied Imaging, Newcastle, UK). Relative copy-number changes were interpreted as gains when the average greento-red ratio exceeded 1.15 (1.5 for amplifications), and as losses when the corresponding ratio was less than 0.85. In some exceptional cases, where imbalances did not reach the aforementioned thresholds, deviation from normal was classified as a trend gain or loss when the 95% confidence interval (CI) varied beyond the ratio of 1.0. 2.6. Statistical analysis Clinicopathological features of AdC and SqC patients were compared by the ManneWhitney test and Pearson’s c test. The average copy-number aberrations, whether gains or losses, were compared by student’s t-test. All statistical analyses were done with SPSS13.0 for Windows XP.

Fig. 1. M-FISH-A549.

H. Shen et al. / Biomedicine & Pharmacotherapy 63 (2009) 396e403

399

3. Results

Table 2 Composite karyotype for each tumor cell line

3.1. M-FISH

Sample

Composite karyotype

A549

58e63, XX, þ2, þ5, del(6p), þder(6)t(6;17), þ8, þ8, þ9, þ10, þder(11)t(5;11), þ12, þ14, þ14, þ16, þder(16)t(16;21), þder(17)t(5;17), þ19, þ21, þ21[cp15]

H520

80e96, XX, þder(1)t(1;6), þ2, þ2, þ3, þ3, þ4, þ4, þ5, þ5, der(5)t(5;17), þ6, þ6, þder(6)t(6;17), þ7, þder(8)t(8;19), þ9, þ10, þ11, þder(11)t(5;11), þ12, þ12, þder(12)t(12;19), þ13, þ14, þ14, þ15, þ15, þ16, þ17, þ17, der(17)t(17,19), þ18, þ18, þ19, þ20, þ20, þ20, þ21, þ21, þ22, þ22[cp16]

The composite karyotype for each sample analyzed by MFISH (Figs. 1 and 2) is shown in Table 2. In total, eight different interchromosomal translocations were observed in the composite karyotypes from the two samples and two of these were observed in both AdC cell line (A549) and SqC cell line (H520) samples. They were t(5;11) and t(6;17). Thirteen chromosomes were involved in at least one translocation, but five chromosomes (5, 6, 11, 12 and 17) appeared to be involved particularly frequently, and the number of chromosome in SqC cell line (H520) was more than in AdC cell line (A549). 3.2. CGH CGH data for 80 lung adenocarcinoma samples and 80 squamous cell lung carcinoma samples were available. The most common areas of loss and gain of DNA sequences and minimal regions of overlap are shown in Table 3 and Fig. 3. The results show that the individual chromosomal aberration pattern was not random and much of them can be found both in AdC group and SqC group. DNA gains were mainly found in 1q, 2p, 3q, 5p, 5q, 7p, 8q, 11q, 12q, 14q, 16p, 17q, 19q, 20q, 21q, and 22q and DNA losses were mainly found in 2q, 3p, 4p, 5q, 7q, 8p, 9p, 13q, 14q, and 17p. DNA gains, rather than DNA losses, were more frequently observed. In AdC group the most frequent gain detected was at 16p13 (50%) terminal and the most frequent loss detected was at 17p12-13 (41.25%), while in SqC group the most frequent gain detected was at 3q24-26 (56.25%) and the most frequent loss detected was at 3p24-25 (38.75%). The X and Y chromosomes were excluded from CGH analysis as samples and reference DNA were not sex-matched.

3.3. Clinicopathological correlation 16p13 (50%) amplification can be commonly found in AdC group, but seldom in SqC group, while 17q21 (45%) amplification can be frequently found in SqC group but not in AdC group. The results show significant difference between these two pathological subtypes ( p < 0.05), Table 3. These alterations were observed to be associated with their specific pathological subtype. 4. Discussion The karyotypes of the primary tumor samples were highly complex and contained many structural aberrations, including the involvement of some chromosomes in many different rearrangements within a single karyotype. We have analyzed two established lung cancer cell lines (A549, H520), 80 primary lung adenocarcinoma samples and 80 squamous cell lung carcinoma samples using modern molecular cytogenetic techniques that allow genetic aberrations to be elucidated, which would have been uninterpretable using conventional cytogenetic methods.

Fig. 2. M-FISH H520.

H. Shen et al. / Biomedicine & Pharmacotherapy 63 (2009) 396e403

400

Table 3 Genomic imbalances in non-small cell lung cancers Regions DNA gains 1q21 2p14 3q24-26 5p15 5q35 7p15 8q24 11q13 12q13 14q12-13 16p13 17q21 17q25 19q12 20q 21q22 22q13 DNA losses 2q24qter 3p24-25 4p15 5q 7q31qter 8p22 9p23 13q21 14q11 17p12-13

Squamous cell carcinoma (%)

Adenocarcinoma (%)

p

18/80 22/80 45/80 20/80 17/80 32/80 35/80 20/80 33/80 17/80 13/80 36/80 42/80 33/80 22/80 26/80 38/80

(22.5%) (27.5%) (56.25%) (25%) (21.25%) (40%) (38.89%) (25%) (36.67%) (18.89%) (16.25%) (45%) (52.5%) (41.25%) (27.5%) (32.5%) (47.5%)

14/80 16/80 33/80 25/80 15/80 39/80 37/80 25/80 26/80 18/80 40/80 10/80 38/80 23/80 27/80 21/80 34/80

(17.5%) (20%) (41.25%) (31.25%) (18.75%) (48.75%) (46.25%) (31.25%) (32.5%) (22.5%) (50%) (12.5%) (47.5%) (28.75%) (33.75%) (26.25%) (42.5%)

0.429 0.265 0.058 0.379 0.693 0.265 0.332 0.379 0.569 0.561 0.000 0.000 0.572 0.097 0.391 0.385 0.525

16/80 31/80 23/80 13/80 22/80 27/80 17/80 19/80 16/80 29/80

(20%) (38.75) (28.75%) (16.25%) (27.50%) (33.75%) (21.25%) (23.75%) (20.00%) (36.25%)

20/80 27/80 17/80 15/80 18/80 30/80 13/80 22/80 19/80 33/80

(25%) (33.75%) (21.25%) (18.75%) (22.50%) (37.50%) (16.25%) (27.50%) (23.75%) (41.25%)

0.449 0.511 0.273 0.677 0.465 0.620 0.418 0.587 0.566 0.516

M-FISH reveals eight translocations in two established lung NSCLC cell lines. Translocations such as t(5;11) and t(6;17) were observed in both AdC cell line (A549) and SqC cell line (H520) samples. It suggests that combinations of the above chromosomes may occur both in AdC cell line and SqC cell line and have no specificity in these two major subtypes of NSCLC cell lines. Of the translocations observed in three samples, only t(1;6), t(5;11) and t(8;19) have been reported by other authors [10,14]. Those translocations that appear in more than one sample are of particular interest, especially t(5;11), as this rearrangement reported by Gunawan et al. [14] in primary squamous cell carcinomas of the lung was also observed in both SqC cell line and AdC cell line in our study. This means that this translocation occurs not only in SqC cell line but also in AdC cell line and may have some relationship with the development of NSCLC. The remaining frequent translocation observed in our study, t(1;6), has also been previously reported by two authors [4,10]. The translocation (17;19) appearing in SqC cell line has never been reported and needs further investigation to find the relation with SqC cell carcinoma of lung. Despite the small number of samples involved, the fact that these combinations of chromosomes have been reported in more than one study may imply that they could be important in the development of NSCLC. It is not possible to say whether the breakpoints

involved in these common translocations are the same in each case. Further investigation of the chromosomes involved will be necessary to ascertain if this is so, and if specific genes located at the breakpoints are perturbed by the rearrangements. Five chromosomes (5, 6, 11, 12 and 17) were shown to be involved in rearrangements more often than the other chromosomes. Chromosome 5 was involved particularly frequently, with two interchromosomal translocations. In addition, the p-arm of this chromosome was the most frequently gained region as indicated by CGH. Speicher et al. [10] reported frequent involvement of chromosomes 1, 2, 3, 11, 12 and 22 in intrachromosomal and interchromosomal changes. Chromosomes 5, 8 and 19 were observed by these authors to be involved less frequently, and chromosome 4 was involved infrequently in rearrangements. Our data therefore partially support those of Speicher et al. [10], and suggest that in addition to chromosome 5, chromosomes 11 and 12 may be of particular interest. The differences between the two studies may be caused by the different composition of NSCLC sample we analysed or may be a reflection of the small sample sizes used. This would only be resolved by a larger scale study. Although not all of the frequently rearranged chromosomes have been observed to participate in specific fusions with other chromosomes, the breakpoints involved warrant further investigation to discover if the same loci are involved each time. CGH reveals that more profound degree of chromosomal instability was evident in SqC and several genetic changes were nonetheless common in both subtypes. Regional gains on 3q24-26 (42e56%), 7p15 (40e48%), 8q24 (39e47%), 17q25 (48e53%), 22q13 (43e48%), and loss of 3p24-25 (34e39%), 8p22 (34e38%), 17p12-13 (37e42%) were frequent in both SqC and AdC (p > 0.05) (Table 3). The high percentage incidence of þ3q24-q26 (56.25%), þ17q25 (52.5%), þ22q13 (47.5%), 17q21 (45%) and loss of 3p24-25 (38.75%) detected in our present SqC series could represent specific tumor-progression pathways that promote the development of SqC. Amplification of chromosome 3q has been described in SqC of different origins, such as lung, oral and vulvar [15]. Although a 3q gain might have a role in the early stages of SqC of the uterine cervix and head and neck [16], recent studies have also shown a positive correlation between 3q over-representation and the invasiveness and progress of oral, oesophageal and hypopharyngeal cancers [17e19]. Positional mapping by array-based CGH analysis on the 3q amplicon in SqC of the lung has implicated PIK3CA as one putative oncogene [20]. Genomic amplification of other genes within the same region, such as BCHE and SLC2A2 (both on 3q26), has also been confirmed in SqC by FISH analysis [21]. It is well known that varying degrees of keratinisation are a characteristic of SqC. Recent expression profiling has suggested that there is a frequent upregulation of cytokeratin over-expression in SqC. Type I cytokeratins KRT10, 13, 14, 15, 16, 19, 20 and KRT23, located at 17q21, might be involved in the keratinisation of SqC [22]. In our study we found that in SqC group the amplification of 17q21 was obviously more often than in AdC group ( p < 0.05). It

H. Shen et al. / Biomedicine & Pharmacotherapy 63 (2009) 396e403

401

Fig. 3. Summary of comparative genomic hybridisation abnormalities identified in non-small cell lung carcinoma. Genetic changes detected in (a) squamous cell carcinoma and (b) adenocarcinoma. Each vertical line represents a single genetic aberration observed in a single tumor specimen. Losses are shown on left and gains on the right of individual chromosomes. High-level amplifications are shown as thick lines.

may be the cause by their different pathological subtype and this loci may be used as a biomarker to distinguish these two subtypes. The frequent loss of 13q21 and the proximal translocations detected in the present series might influence the perturbation of the putative tumor-suppressor BRCA2 gene (on 13q21). Deletion of the BRCA2 gene results in the loss of

genes that regulate the cell-cycle, enabling proliferation and tumorigenesis [23]. BRCA2 protein is also associated with the activation of doublestrand break (DSB) repair via homologous recombination. Since chromosomal instability could occur as a consequence of the disruption of DNA repair or recombination pathways [24], it is possible that the gross chromosomal

402

H. Shen et al. / Biomedicine & Pharmacotherapy 63 (2009) 396e403

alterations detected in SqC may be related to defective DSB repair mechanisms caused by an abnormality in the BRCA pathway. In AdC group, the most frequently altered loci were gain of 16p13 (50% of samples) region, gain of 7p15 (48.75% of samples) and gain of 17q25 (46% of samples). Chromosome 16p13 contains the tuberous sclerosis (TSC2) gene. Tuberous sclerosis is an autosomal dominant disorder characterized by seizures, mental retardation, and hamartomatous tumors of the brain, heart, kidney, lung, and skin [25]. Chromosome 17p15 contains the HOXA9 gene and HOX genes encode transcription factors that control patterning and cell fates. Previous studies reported that HOXA9 was frequently upregulated in lung cancer cell lines and direct tumors although less pronounced [26]. Chromosome 17q25 contains the v-maf musculoaponeurotic fibrosarcoma oncogene homolog G (MAFG) gene. MAFG, which is a member of the MAF protooncogene and cap’n’collar families of basic-leucine zipper transcription factors, plays important roles in development, differentiation, oncogenesis, and stress signaling [27]. Gain of 22q13 (40%) is important for our study because plateletderived endothelial cell growth factor (ECGF1) located at 22q13 is known as thymidine phosphorylase (TP), expressed at higher levels in a variety of human cancers than in adjacent normal tissue [28]. Activating mutations of the EGFR gene have recently been identified in cases of bronchioloalveolar carcinomas and ADC [29], showing a striking association with response to therapy with the tyrosine kinase inhibitor gefitinib, whereas cases without activating mutations were nonresponders. In this context, Lynch et al. [30] concluded that gliomas with EGFR gene amplification might require higher doses of gefitinib to disrupt EGFR signaling. Whether EGFR amplification, mutational status, or protein expression are all reliable predictors for response to EGFR-targeted therapy in NSCLC requires further examination [31]. We also found gains of many chromosome regions which have clinical value such as 5p15 (32%), 14q12-13 (26%), and 3q24-26 (36%). 5p15 contains the hTERT gene that conducts telomerase activity [32]. 14q12 has been determined as a prognostic marker region in a CGH study [33]. Among others, p63 [34] and the established protooncogenes PIK3CA and TERC (telomerase RNA component), which are localized at 3q26, are associated with cell proliferation and carcinogenesis in several types of malignancies and may also represent candidate genes for NSCLC pathogenesis [35,36]. Furthermore, we found amplified chromosomes containing oncogenes such as MYC (8q24) and CCND1 (11q13) which were described in previous reports [37e39]. Overall, loss of chromosomal regions was less frequent than gains in this CGH analysis. The observation of losses on chromosomes 3p, 9p and 17p is also in agreement with the current body of literature relating to NSCLC [5,8]. Chromosomes 9p and 17p are known to harbor the tumor-suppressor genes p16 and p53. p53 gene is located at 17p13 and we found the loss of 17p12-13 (36e41%) containing this gene, but the location of p16 reported at 9p21 does not correspond to the minimal regions of overlap identified in this study.

Losses of 3p, including 3p25, are known to be frequent in lung cancer. The minimal region of loss observed here (3p24p25) is in agreement with this. Potentially important genes within this region are retinoic acid receptor beta (RARB) at 3p24, which is a nuclear transcription factor that is thought to limit cell growth by regulating gene expression, and the von HippeleLindau (VHL) tumor-suppressor gene at 3p25. Losses of 3p and 6q have previously been associated with the metastatic phenotype in both adenocarcinoma and squamous cell carcinoma of the lung. In conclusion, most of the genomic alterations in SqC and AdC samples did not show obvious differences. We found that 16p13 which contains TSC2 gene was frequently amplified in AdC but not in SqC while 17q21 which contains type I cytokeratin genes was obviously amplified in SqC but not in AdC. We can use these two loci to distinguish AdC and SqC, and make further research to determine whether these genomic aberrations can be used as biomarker to diagnose SqC or AdC. References [1] Jemal A, Thomas A, Murray T, Thun M. Cancer statistics. CA Cancer J Clin 2002;2002(52):23e47. [2] Wynder EL, Goodman MT, Hoffmann D. Lung cancer etiology: challenges of the future. Carcinog Compr Surv 1985;8:39e62. [3] Wang T-J, Zhou B-S. Meta-analysis of the potential relationship between exposure to environmental tobacco smoke and lung cancer in nonsmoking Chinese women. Lung Cancer 1997;16:145e50. [4] Luk C, Tsao M-S, Bayani J, Shepherd F, Squire JA. Molecular cytogenetic analysis of non-small cell lung carcinoma by spectral karyotyping and comparative genomic hybridization. Cancer Genet Cytogenet 2001; 125:87e99. [5] Petersen I, Bujard M, Petersen S, Wolf G, Goeze A, Schwendel A, et al. Patterns of chromosomal imbalances in adenocarcinoma and squamous cell carcinoma of the lung. Cancer Res 1997;57:2331e5. [6] Pei J, Balsara BR, Li W, Litwin S, Gabrielson E, Feder M, et al. Genomic imbalances in human lung adenocarcinomas and squamous cell carcinomas. Genes Chromosomes Cancer 2001;31:282e7. [7] Kallioniemi A, Kallioniemi O-P, Sudar D, Rutovitz D, Gray JW, Waldman F, et al. Comparative genomic hybridization for molecular cytogenetic analysis of solid tumours. Science 1992;258:818e21. [8] Balsara BR, Sonoda G, du Manoir S, Siegfried JM, Gabrielson E, Testa JR. Comparative genomic hybridization analysis detects frequent, often highlevel, overrepresentation of DNA sequences at 3q, 5p, 7p and 8q in human non-small cell lung carcinomas. Cancer Res 1997;57:2116e20. [9] Bjo¨rkqvist AM, Husgafvel-Pursiainen K, Anttila S, Karjalainen A, Tammilehto L, Mattson K, et al. DNA gains in 3q occur frequently in squamous cell carcinoma of the lung, but not in adenocarcinoma. Genes Chromosomes Cancer 1998;22:79e82. [10] Speicher MR, Petersen S, Uhrig S, Jentsch I, Fauth C, Eils R, et al. Analysis of chromosomal alterations in non-small cell lung cancer by multiplex-FISH, comparative genomic hybridization, and multicolor bar coding. Lab Invest 2000;80:1031e41. [11] Speicher MR, Ballard SG, Ward DC. Karyotyping human chromosomes by combinatorial multi-fluor FISH. Nat Genet 1996;12:368e75. [12] Ashman JNE, Brigham J, Cowen ME, Bahia H, Greenman J, Lind M, et al. Chromosomal alterations in small cell lung cancer revealed by multicolour fluorescent in situ hybridization. Int J Cancer 2002;102: 230e6. [13] Gunawan B, Schulten HJ, von Heydebreck A, Schmidt B, Enders C, Ho¨er J, et al. Site-independent prognostic value of chromosome 9q loss in primary gastrointestinal stromal tumours. J Pathol 2004;202:421e9.

H. Shen et al. / Biomedicine & Pharmacotherapy 63 (2009) 396e403 [14] Gunawan B, Mirzaie M, Schulten H-J, Heidrich B, Fu¨zesi L. Molecular cytogenetic analysis of two primary squamous cell carcinomas of the lung using multicolor fluorescence in situ hybridization. Virchows Arch 2001;439:85e9. [15] Allen DG, Hutchins AM, Hammet F, White DJ, Scurry JP, Tabrizi SN, et al. Genetic aberrations detected by comparative genomic hybridisation in vulvar cancers. Br J Cancer 2002;86:924e8. 2828. [16] Umayahara K, Numa F, Suehiro Y, Sakata A, Nawata S, Ogata H, et al. Comparative genomic hybridization detects genetic alterations during early stages of cervical cancer progression. Genes Chromosomes Cancer 2002;33:98e102. [17] Oga A, Kong G, Tae K, Lee Y, Sasaki K. Comparative genomic hybridization analysis reveals 3q gain resulting in genetic alteration in 3q in advanced oral squamous cell carcinoma. Cancer Genet Cytogenet 2001;127:24e9. [18] Noguchi T, Kimura Y, Takeno S, Chujo M, Uchida Y, Mueller W, et al. Chromosomal imbalance in esophageal squamous cell carcinoma: 3q gain correlates with tumor progression but not prognostic significance. Oncol Rep 2003;10:1393e400. [19] Steinhart H, Bohlender JE, Constantinidis J, Urbschat S, Fischer U, Iro H, et al. Genetic imbalances in preinvasive tissue of hypopharynx provide evidence for cytogenetic heterogeneity. Oncol Rep 2001;8: 1229e31. [20] Massion PP, Kuo W-L, Stokoe D, Olshen AB, Treseler PA, Chin K, et al. Genomic copy number analysis of non-small cell lung cancer using array comparative genomic hybridization: implications of the phosphatidylinositol 3-kinase pathway. Cancer Res 2002;62:3636e40. [21] Brass N, Racz A, Heckel D, Remberger K, Sybrecht GW, Meese EU. Amplification of the genes BCHE and SLC2A2 in 40% of squamous cell carcinoma of the lung. Cancer Res 1997;57:2290e4. [22] Virtanen C, Ishikawa Y, Honjoh D, Kimura M, Shimane M, Miyoshi T, et al. Integrated classification of lung tumors and cell lines by expression profiling. Proc Natl Acad Sci USA 2002;99:12357e62. [23] Tutt A, Gabriel A, Bertwistle D, Connor F, Paterson H, Peacock J, et al. Absence of Brca2 causes genome instability by chromosome breakage and loss associated with centrosome amplification. Curr Biol 1999;9: 1107e10. [24] Venkitaraman AR. Cancer susceptibility and the functions of BRCA1 and BRCA2. Cell 2002;108:171e82. [25] Astrinidis A, Khare L, Carsillo T, Smolarek T, Au KS, Northrup H, et al. Mutational analysis of the tuberous sclerosis gene TSC2 in patients with pulmonary lymphangioleiomyomatosis. J Med Genet 2000;37:55e7. [26] Calvo R, West J, Franklin W, Erickson P, Bemis L, Li E, et al. Altered HOX and WNT7A expression in human lung cancer. Proc Natl Acad Sci USA 2000;97:12776e81.

403

[27] Che´nais B, Derjuga A, Massrieh W, Red-Horse K, Bellingard V, Fisher SJ, et al. Functional and placental expression analysis of the human NRF3 transcription factor. Mol Endocrinol 2005;19:125e37. [28] Sato J, Sata M, Nakamura H, Inoue S, Wada T, Takabatake N, et al. Role of thymidine phosphorylase on invasiveness and metastasis in lung adenocarcinoma. Int J Cancer 2003;106:8630e70. [29] Paez JG, Ja¨nne PA, Lee JC, Tracy S, Greulich H, Gabriel S, et al. EGFR mutations in lung cancer: correlation with clinical response to gefitinib therapy. Science 2004;304:1497e500. [30] Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, Brannigan BW, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small cell lung cancer to gefitinib. N Engl J Med 2004;350:2129e39. [31] Hirsch FR, Bunn Jr PA. Epidermal growth factor receptor inhibitors in lung cancer: smaller or larger molecules, selected or unselected populations? J Clin Oncol 2005;23:9044e7. [32] Garnis C, Lockwood WW, Vucic E, Ge Y, Girard L, Minna JD, et al. High resolution analysis of non-small cell lung cancer cell lines by whole genome tiling path array CGH. Int J Cancer 2006;118:1556e64. [33] Petersen S, Aninat-Meyer M, Schluns K, Gellert K, Dietel M, Petersen I. Chromosomal alterations in the clonal evolution to the metastatic stage of squamous cell carcinomas of the lung. Br J Cancer 2000;82:65e73. [34] Massion PP, Taflan PM, Jamshedur Rahman SM, Yildiz P, Shyr Y, Edgerton ME, et al. Significance of p63 amplification and overexpression in lung cancer development and prognosis. Cancer Res 2003; 63:7113e21. [35] Yokoi S, Yasui K, Iizasa T, Imoto I, Fujisawa T, Inazawa J. TERC identified as a probable target within the 3q26 amplicon that is detected frequently in non-small cell lung cancers. Clin Cancer Res 2003;9: 4705e13. [36] Samuels Y, Wang Z, Bardelli A, Silliman N, Ptak J, Szabo S, et al. High frequency of mutations of the PIK3CA gene in human cancers. Science 2004;304:554. [37] Tan D, Deeb G, Wang J, Slocum HK, Winston J, Wiseman S, et al. HER2/neu protein expression and gene alteration in stage IeIIIA non-smallcell lung cancer: a study of 140 cases using a combination of high throughput tissue microarray, immunohistochemistry, and fluorescent in situ hybridization. Diagn Mol Pathol 2003;12:201e11. [38] Wikman H, Nymark P, Va¨yrynen A, Jarmalaite S, Kallioniemi A, Salmenkivi K, et al. CDK4 is a probable target gene in a novel amplicon at 12q13.3-q14.1 in lung cancer. Genes Chromosomes Cancer 2005;42: 193e9. [39] Yakut T, Egeli U, Gebitekin C. Investigation of c-myc and p53 gene alterations in the tumor and surgical borderline tissues of NSCLC and effects on clinicopathologic behavior: by the FISH technique. Lung 2003; 181:245e58.