Genomic instability and increased expression of BUB1B and MAD2L1 genes in ductal breast carcinoma

Cancer Letters 254 (2007) 298–307 www.elsevier.com/locate/canlet

Genomic instability and increased expression of BUB1B and MAD2L1 genes in ductal breast carcinoma Marina Scintu a,1, Rita Vitale a,b,1, Maria Prencipe a, Antonietta Pia Gallo a, Loriana Bonghi a, Vanna Maria Valori c, Evaristo Maiello c, Monica Rinaldi d, Emanuela Signori d,e, Carla Rabitti f, Massimo Carella g, Bruno Dallapiccola g, Vittorio Altomare h, Vito M. Fazio a,e, Paola Parrella a,* a

g

Laboratory of Oncology, Research Department, IRCCS Casa Sollievo della Sofferenza, Viale Padre Pio, San Giovanni Rotondo (FG) 71013, Italy b Department of Medical Biochemistry, Biology and Physics, University of Bari, 70124 Bari, Italy c Oncology Department, IRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo (FG) 71013, Italy d Institute of Neurobiology and Molecular Medicine, Consiglio Nazionale delle Ricerche, Rome, Italy e Molecular Medicine and Biotechnology Laboratory, University Campus BioMedico, Rome 00155, Italy f Department of HystoPathology, University Campus BioMedico, Rome, Italy Laboratory of Medical Genetics, IRCCS Casa Sollievo della Sofferenza, San Giovanni Rotondo (FG) 71013, Italy h Breast Unit, Department of Surgery, University Campus BioMedico, Rome 00168, Italy Received 16 January 2007; received in revised form 14 March 2007; accepted 15 March 2007

Abstract In a series of invasive ductal breast carcinoma, we investigated the status of chromosomal and intrachromosomal instability by fluorescence in situ hybridisation and determined the level of mRNA expression for two genes involved in the mitotic spindle checkpoint pathway, BUB1B and MAD2L1. All breast cancers demonstrated higher chromosomal instability rates in tumor samples (average: 56.86%, range: 36.24–76.78%) than in controls (average: 11.54%, range: 9.91– 14.84%) (P < 0.0001). As well as intrachromosomal instability rates were elevated in tumor (average: 18.45% range: 8.34–35.8%) as compared with controls (average: 4.18% range: 3.47–4.81%) (P < 0.0001). An increase in BUB1B and MAD2L1 transcripts was demonstrated in the majority of the tumor tested. BUB1B mRNA levels but not MAD2L1 levels correlated with intrachromosomal instability (r = 0.722, P = 0.018).  2007 Elsevier Ireland Ltd. All rights reserved. Keywords: Breast cancer; Genomic instability; MAD2L1; BUB1B; Gene expression

Abrreviations: MIN, microsatellite instability; NER, nucleotide excision repair; CIN, chromosomal instability; FISH, fluorescence in situ hybridisation; ICI, intrachromosomal instability; QRT-PCR, quantitative reverse transcription polymerase chain reaction. * Corresponding author. Tel.: +39 0882 416262; fax: +39 0882 416261. E-mail address: [email protected] (P. Parrella). 1 Both authors contribute equally to this work. 0304-3835/$ - see front matter  2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.canlet.2007.03.021

M. Scintu et al. / Cancer Letters 254 (2007) 298–307

1. Introduction The number of genetic abnormalities accumulated in cancer cells cannot be simply explained by the stepwise accumulation of changes in cells with a normal mutation rate, suggesting that tumors have a substantial unstable genome. In tumors with inactivation of DNA mismatch repair (MMR) genes, cancer cells are susceptible to the acquisition of somatic mutations throughout the genome. In this circumstances, repetitive regions are particular susceptible to mutations, and instability is observed especially at short repeated sequences scattered through the genome named microsatellite (microsatellite instability, MIN) [1]. Cancer cells showing MIN have a mutation rate at nucleotide level that is two or three orders of magnitude greater than those observed in normal cells [2]. Another mechanism that is involved in maintaining genomic stability is the nucleotide excision repair (NER) pathway. In the hereditary cancer syndrome Xeroderma pigmentosum, loss of NER genes induces skin cancer whose genome is characterized by high mutation rates at pyrimidine dimers [3]. However, most of the tumors do not show nucleotide instability, but rather they are characterized by the presence of gross chromosomal numerical and structural changes, which is referred as aneuploidy [2]. It is of note that genomic instability at nucleotide levels and aneuploidy are virtually mutually exclusive, thus aneuploidy may reflect a different form of genetic instability that accelerates mutation rates by different mechanisms [2]. One attempt to understand the mechanisms responsible for aneuploidy was to measure the rate at which cancer cell lines gain and lose chromosomes [4]. In colorectal cancer an accelerated rate of chromosomal gain and losses was demonstrated in cell lines with aneuploidy as compared with tumors showing MIN [2]. In similar experiments, frequent loss or gains of whole chromosomes were also demonstrated in breast cancer and head and neck cancer cell lines [5,6]. These accelerated rates of chromosomal aberrations leading to aneuploidy were termed as chromosomal instability (CIN), and has been suggested as an alternative form of genomic instability [4]. Thus aneuploidy describes the condition of a non-diploid karyotype, while CIN is a measure of the flux in karyotype [4]. In primary tumors CIN has been determined by FISH analysis as the percentage of clones deviating from the modal chromosome number [6,7]. Using a similar approach Minhas et al. [6] have

299

investigated the level of chromosomal breakage in head and neck cancer cell line and primary tumors introducing the concept of intrachromosomal instability (ICI) as the rate at which chromosomal rupture occur in cancer cells. Several mechanisms can be involved in the determination of aneuploidy, it can occur as results of aberrant mitotic divisions or by defects in duplications, maturation or segregation of centrosome [7–9]. Another mechanism could be due to chromosome cohesion defects [10] or aneuploidy could arise by improper attachment of chromosome to spindle microtubules [11,12]. In cancer cell lines, several evidences implicate the mitotic spindle checkpoint as the point of failure in chromosomal instability [13]. In normal cells, the spindle checkpoint pathway ensures that all the chromosomes are correctly aligned in metaphase and properly attached to the mitotic spindle before separation [14]. Cells in tissue culture showing chromosomal instability do not arrest in mitosis after the treatment with microtubule disrupting agents as normal cells do, suggesting a failure in mitotic spindle checkpoint [5,6,13,15]. Karyotypic analysis of breast cancer cell lines and primary culture demonstrated the same structural genomic abnormalities found in other human cancer [16–18]. Moreover, a cell-to-cell variability of chromosome numbers within the same tumor was also described, suggesting that chromosomal instability is a feature of breast carcinomas [5,19,20]. A study on breast cancer cell lines has shown a relationship between higher level of chromosomal instability and mitotic spindle checkpoint defects [5]. However, expression analysis of mitotic spindle checkpoint genes have shown conflicting results with either reduced, increased or normal levels of expression [21–23]. 2. Materials and methods 2.1. Patients and samples We evaluated 19 patients affected by invasive ductal carcinoma treated at the Department of Surgery, Breast Unit University Campus BioMedico, Rome, Italy. All patients were female and average age ± SD at the diagnosis was 56 ± 15 years (range: 30–77 years). Tumors were less than 2 cm in size (T1b or T1c), 12 cases were lymph node negative and 7 positive, 3 tumors were grade I, 9 grade II and 7 grade III. The average ± SD percentage of positively stained cells for estrogen receptor, progesterone receptor and ki67/Mib1 determined by routine

300

M. Scintu et al. / Cancer Letters 254 (2007) 298–307

immunohistochemical analysis were 71.16 ± 30.69%, 59.84 ± 31.07% and 17.67 ± 12.58%, respectively. Seven tumors were positive for TP53 expression, and four tumors displayed Her2neu protein overexpression. After surgical resection, tissue from the bulk of the tumor, and normal breast distant from cancer were immediately frozen in liquid nitrogen and stored at 80 C. At the time to be processed, each normal and tumor sample was cut in half. One half of the samples was embedded in OCT compound and a 5 lm eosin/ ematoxylin stained section was prepared to ensure that tumor sample contained more than 70% of tumor cells as previously described [24]. Fresh frozen sections from breast tissues distant from tumors were also examined, four samples showed some grade of fibrocystic disease whereas the remaining 15 samples did not show pathological alterations. Control samples for FISH analysis were chosen among breast tissues without pathological abnormalities. Total RNA was extracted from the other half of the frozen normal or tumor specimen, using the Invitrogen (Carlsbad, CA) TRIzol reagent. The tissues were kept frozen in liquid nitrogen, whereas TRIzol was added, than the samples were carefully and mechanically homogenized, and the mixture was transferred into 1.5-ml tube using a sterile cell scraper. The RNA samples were then further purified using an RNeasy kit (Qiagen, Valencia, CA) following manufacture’s instruction. 2.2. Fluorescence in situ hybridization (FISH) analysis Tumor cells were isolated from 50-lm fresh frozen OCT embedded blocks as previously described [25]. Cellular disaggregation was performed in a 0.5% pepsin (Sigma, St. Louis, MO) solution in 0.9% NaCl adjusted to pH 1.5 with HCl at 37 C for 20 min. After disaggregation the cell suspension was filtered through 40 lm nylon mesh (Sefar Italia S.r.l., Italy), pelletted by centrifugation at 1500g for 1 min. Then 200 ll of supernatant were removed with a micropipette, and the remaining cell suspension was diluted in 150 ll phosphate buffered saline (PBS). A 40 ll aliquot was applied onto poly-L-lysine-coated slides (Superfrost Plus Fisher Scientific, Nepean, Ont., Canada) that were dried at RT for 60 min. Cells were then fixed in PBS-buffered 1% formaldehyde for 5 min, washed in PBS and dehydrated by sequential immersion in Ethanol 70%, 95% and 100%. Dual probe FISH analysis was performed with two probes localized on the p and q arm, respectively, for chromosome 8 (TelVysion 8p Spectrum Green; LSI-c-myc Spectrum Orange; Vysis Inc Downer Grove, IL), and chromosome 17 (TelVysion 17p Spectrum Green; TelVysion 17q Spectrum Orange), and on chromosomal bands 13q14.3 (LSI D13S25 Spectrum Orange) and 13q34 (LSI Spectrum Green) for chromosome 13 (Fig. 1). Target DNA and probes were codenaturated at 72 C and hybridised overnight at 37 C. Nuclei preparations were

Fig. 1. Dual probe FISH analysis using probes for chromosome 8, 13 and 17, representative isolated nuclei from breast tissues are 0 0 shown. Nuclei are counterstained with 4 -6 -diamino phenylindole (DAPI, blue). (a) Normal breast tissue (BN5) hybridised with probes for chromosome 8p (TelVysion 8p Spectrum, green) and 8q (LSI-c-myc, orange). Modal chromosomal number is 2/2 (89% of the nuclei population), CIN and ICI rates are 11.1% and 4.0%, respectively. (b) Tumor BT1 hybridised with probes located at chromosome band 13q14.3 (LSI D13S25, orange) and 13q34 (LSI, green) for chromosome 13. Modal chromosomal number is 1/1 (38% of the nuclei population), CIN and ICI rates are 62.1% and 6.6%, respectively. White arrows nuclei showing one signal for each probe, red arrows nuclei showing two signals for each probe. (c) Tumor BT9 hybridised with probes for chromosomes 17p (TelVysion 17p, green) and 17q (TelVysion 17q, orange). Modal chromosomal number is 2/2 (32% of the nuclei population), CIN and ICI rates are 69.2% and 26.8%, respectively. White arrow nucleus showing three signals for each of the probes (11% of the nuclei population); red arrow nucleus displaying four signals for each of the probes (18% of the nuclei population). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

washed with 0.1· SSC (pH 7.2) + 0.1% NP40 for 5 min at 60 C and 2 · SSC (pH 7.2) at room temperature for 1 min. Slides were counterstained with 4 0 -6 0 -diaminophenylindole (DAPI) and the antifade compound p-phenylenediamine (Vector Laboratories Inc, Burlingam, CA, USA). Nuclear signals were detected with a fluorescent microscope Eclipse 600 (Nikon Instrument S.p.A, Florence, Italy) equipped with a triple pass filter. Images were captured digitally using Genikon system (Nikon Instrument S.p.A, Florence, Italy). At least 200 nuclei were counted for each case and cell clones were identified by evaluating the percentage of cells having the same probe signal pattern [26]. For

M. Scintu et al. / Cancer Letters 254 (2007) 298–307

each normal or tumor specimen the modal chromosome number (M) was determined, and chromosomal instability (CIN) was calculated as the fraction of cells with a chromosomal number different from that of the mode in percent (F%) (Table 1). We also evaluated the rate of chromosomal structural abnormalities (Intrachromosomal Instability, ICI): for each cell nucleus, a value n was assigned to indicate the number of p arm signals for chromosomes 8 and 17 and for the chromosomal band 13q14.3 signal for chromosome 13. The q or 13q34 signals were determined relative to n and counted. The modal number was calculated as the most common number of q signals relative to n (n  2, n  1, n, n + 1, n + 2, etc.) [6]. Nuclei were then counted individually and categorized as the difference of q signals relative to the p signals for chromosomes 8 and 17, and to the 13q14.3 signal for chromosome 13. The modal number (M) of signal differences was determined and the deviation from that mode in percent (F%) was calculated (Table 2) [6]. This method minimizes the impact of loss

301

or gain of entire chromosomes in the determination of chromosomal breakage, for which compensation is made in this calculation. 2.3. Quantitative reverse transcription polymerase chain reaction (QRT-PCR) Total RNA was reverse transcribed with Multiscribe reverse transcriptase (Applied Biosystems, Foster City, CA, USA) and fluorescence based real time RT-PCR was performed with MGB TaqMan chemistry on ABI 7700 sequence detection system (Applied Biosystems, Foster City, CA, USA). Reactions were performed in singleplex using Taqman Gene Expression Assays (Applied Biosystems Foster City, CA, USA) for BUB1B (Hs00176169_m1) and MAD2L1 (Hs00829154_m1) and the human large ribosomal protein (Human RPLP09, Applied Biosystems, Foster City, CA, USA) as endogenous control. Expression of the target gene is normalized by the expression of the endogenous control

Table 1 Chromosomal Instability (CIN) in normal breast tissues and primary breast cancer as determined by FISH analysis Chromosome 8 a

Normal BN1 BN2 BN3 BN4 BN5

Average ± SD a

a

M

(F%)

M

(F%)

(F%)a

2/2 2/2 2/2 2/2 2/2

11.0 11.1 8.1 15.8 11.7

2/2 2/2 2/2 2/2 2/2

9.3 24.9 21.1 20.4 19.1

2/2 2/2 2/2 2/2 2/2

9.4 8.3 9.8 8.3 10.2

9.9 14.8 13.0 14.8 13.6

2/2 2/2 4/4 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 3/3 2/2 2/2 2/2 2/2 2/2 2/2 2/2

43.6 74.4 71.4 79.0 41.3 40.1 65.4 64.6 69.2 50.7 84.9 31.0 51.5 43.5 50.2 71.9 68.4 53.8 55.3 58.43 ± 15.02

a

18.96 ± 5.81

1/1 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2

62.1 76.1 50.7 53.1 49.7 41.4 49.2 40.7 72.8 55.5 81.7 59.8 55.4 65.7 58.8 39.1 70.7 69.0 66.7 58.85 ± 12.25

a

Average

(F%)

11.54 ± 2.76

a

Chromosome 17

M

Average ± SD Tumor BT1 BT2 BT3 BT4 BT5 BT6 BT7 BT8 BT9 BT10 BT11 BT12 BT13 BT14 BT15 BT16 BT17 BT18 BT19

Chromosome 13 a

9.20 ± 0.87

2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2 3/3 2/2 2/2 2/2 2/2 2/2 2/2 2/2 2/2

13.24 ± 2.02

37.9 67.6 53.9 50.5 38.3 27.2 13.7 69.9 80.2 54.4 63.7 70.7 13.1 57.8 67.6 54.2 64.7 70.9 55.9

47.9 72.7 58.7 60.9 43.1 36.2 42.8 58.4 74.1 53.5 76.8 53.8 40.0 55.7 58.9 55.1 67.9 64.6 59.3

53.27 ± 19.18

56.86 ± 11.46

M, Modal number; F(%) is determined as the fraction of cells with chromosomal number of for deviating from the mode. Bold values indicate samples showing a modal chromosomal number other than ‘‘2/2’’.

302

M. Scintu et al. / Cancer Letters 254 (2007) 298–307

Table 2 Intrachromosomal instability (ICI) in normal breast tissues and primary breast cancer as determined by FISH analysis

Normal BN1 BN2 BN3 BN4 BN5

Chromosome 8

Chromosome 13

Chromosome 17

Average

Ma

Ma

Ma

(F%)a

(F%)a

6.2 3.6 4.3 4.6 3.9

3.9 4.3 3.5 4.8 4.4

4.48 ± 1.06

4.18 ± 0.50

2 2 2 2 2

Average ± SD Tumor BT1 BT2 BT3 BT4 BT5 BT6 BT7 BT8 BT9 BT10 BT11 BT12 BT13 BT14 BT15 BT16 BT17 BT18 BT19

(F%)a 4.0 4.0 3.4 5.5 4.2

2 2 2 2 2

4.22 ± 0.78

2 2 4 2 2 2 2 2 2 2 2 3 2 2 2 2 2 2 2

Average ± SD

5.9 33.6 10.7 63.0 14.7 12.9 14.7 22.8 12.1 16.6 35.0 31.0 15.1 12.5 13.9 13.3 14.8 22.9 15.4

(F%)a 1.4 5.4 2.7 4.3 5.2

2 2 2 2 2

3.80 ± 1.71

1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

20.05 ± 13.06

6.6 24.9 20.6 13.7 7.5 4.7 15.8 22.1 28.7 10.5 17.8 15.2 12.2 7.7 13.7 2.0 24.6 16.5 18.1 14.89 ± 7.33

2 2 2 2 2 2 2 2 2 2 3 2 2 2 2 2 2 2 2

13.1 48.8 16.8 19.5 16.9 7.5 13.7 28.9 39.5 19.9 21.9 21.0 13.1 12.1 22.8 18.9 11.2 32.7 9.3

8.5 35.8 16.0 32.1 13.0 8.4 14.7 24.6 26.8 15.7 24.9 22.4 13.5 10.8 16.8 11.4 16.9 24.0 14.2

20.40 ± 10.61

18.45 ± 7.80

a

M, Modal number; F(%) is determined as the fraction of cells with q signals relative to p or pericentromeric signal deviating from the mode. Bold values indicate samples showing a modal chromosomal number other than ‘‘2’’.

ðDC t ¼ C tT  C tEC Þ. To ensure that efficiency of PCR reaction was similar for both target genes and endogenous control, for each determination three dilutions of the cDNA sample were analyzed in triplicates (1:50, 1:25 and 1:10) (Fig. 2). Relative expression (RE) in tumor samples as compared with matched normal breast tissue was calculated according to the following formula: 2DDCt where DDCt is the difference between DCts calculated for normal and tumor, respectively (see ABI PRISM 7700 Sequence Detection System User Bulletin #2). According to this formula mRNA expression in normal breast tissue is 1, thus RE value <1 indicate reduced expression and RE > 1 indicate increased expression. 2.4. Statistical analysis Statistical analyses were performed using SPSS 10.0 software (SPSS Inc., Chicago IL, USA). Mean values were compared using the one-way ANOVA or twotailed t-test when appropriate. Linear regression analysis

was used to look for correlations between continuous variable. All tests are two sided with significance at P 6 0.05. 3. Results 3.1. Determination of chromosomal and intrachromosomal instability by FISH analysis Using two probes for each of the three chromosomes, we determined chromosomal instability (CIN) and intrachromosomal instability (ICI) in nuclei preparations obtained from five pathologically normal breast tissues and 19 tumor specimens. Optimal FISH preparations with at least 200 scorable nuclei were obtained for all samples (average ± SD: 220.14 ± 23.63, range: 200–256). Representative results are shown in Fig. 1. Normal breast tissues were diploid with an average ± SD percentage of cells with two probe signals of 86.7% ± 2.0 (range: 85.2–90.1%), whereas all cancer specimens showed variable degree of aneuploidy with an aver-

M. Scintu et al. / Cancer Letters 254 (2007) 298–307

303

Fig. 2. Real time quantitative PCR for representative case BT5, normal breast tissue (N) and tumor (T) samples were tested in triplicates at three total mRNA dilutions 1:10, 1:25 and 1:50 for the target genes BUB1B and MAD2L1 and for the endogenous control, human large ribosomal protein (Human RPLP09). (a) Amplification plots for BUB1B gene labelled with FAM (left panel) and Human RPLP09 endogenous control labelled with VIC (right panel). (b) Amplification plots for MAD2L1 gene labelled with FAM (left panel) and Human RPLP09 endogenous control labelled with VIC (right panel). To ensure that amplification was specific, for each reaction plate a negative control (water) was also tested, Ct value were 45 for both target genes and endogenous control (not shown on the plots).

age ± SD percentage of cells with two probe signals of 40.6% ± 10.1 (range: 25.7–63.8) (P < 0.0001). All normal tissues and the majority of the tumors displayed a modal chromosomal number for each of the chromosome of 2/2. For tumor BT12 and BT8 chromosome 8 modal numbers were 3/3 and 4/4, respectively. Tumor BT1 chromosome 13 modal number was 1/1, and tumor BT11 chromosome 17 modal number was 3/3 (Table 1). The average chromosomal number deviation from the mode (F%) in controls was 13.24% (range: 9.9–14.8%), whereas for breast cancer was markedly elevated at 56.86% (range: 36.2–76.8) (P < 0.0001) (Table 1). In control the average modal deviation (F%) for chromosomes 8, 13 and 17 was 11.54% (range: 8.1–15.8), 18.96% (range: 9.3–24.9%) and 9.20% (8.3–10.2), respectively, whereas tumor was much greater at 58.43% (range: 31.0–84.9%), 58.95% (range: 39.1– 81.7%) and 53.27% (range: 13.1–80.2%) (P < 0.0001) (Table 1). To determine intrachromosomal instability, individual cells were characterized by the number of q signals relative to p signals for chromosome 8 and 17, and to the 13q14.3 signal for chromosome 13. For both normal and tumor samples the modal number of signals was the same than the modal copy number (Table 2). The average deviation from the modal number of q signals relative to p or

13q14.3 signal (F%) in controls was 4.18% (range: 3.5– 4.8%) whereas it was higher in primary breast cancer 18.45% (range: 8.4–35.8%) (P < 0.0001). For chromosomes 8, 13 and 17 the average deviation from the modal number of q signals relative to p or 13q14.3 signal was in control tissues of 4.22% (range: 3.4–5.5%), 3.80 (range: 1.4–5.4%) and 4.48% (3.6–6.2%), respectively, whereas in tumor samples was much higher 20.05% (range: 5.9– 35.0%), 14.89% (range: 2–28.7%) and 20.40% (range: 7.5–48.8%), respectively (P < 0.0001) (Table 2). This is consistent with primary breast cancer displaying a phenotype of intrachromosomal alterations and instability, either due to strand breakage, translocation or telomeric loss. Results from the FISH analysis were correlated with clinical parameters. No association was found for both CIN and ICI rates with tumor grade, estrogen receptor, progesterone receptor, ki67/mib1 and lymph node status. Tumors positively stained for TP53 (n = 7) displayed Mean ± SD ICI rates of 23.22 ± 8.60 as compared with rates of 15.66 ± 6.01 in tumors TP53 negative (n = 12) (P < 0.04). An association was also found between average ICI rates, chromosome 17 CIN and ICI rates, and Her2neu overexpression. Average mean ± SD ICI rates were 26.70 ± 6.12 in Her2neu positive tumors (n = 4)

304

M. Scintu et al. / Cancer Letters 254 (2007) 298–307

and 16.24 ± 6.74 in negative samples (n = 15) (P < 0.01). Chromosome 17 mean ± SD CIN rates were 69.78 ± 1.54 in Her2neu overexpressing samples and 48.87 ± 19.34 in Her2neu negative tumors (P < 0.001). Chromosome 17 mean ± SD ICI rates were 32.86 ± 11.69 in Her2neu positive tumors, and 17.08 ± 7.71 in negative cases (P < 0.004). 3.2. Increased BUB1B and MAD2L1 mRNA levels and correlation with genomic instability detected by FISH To determine whether changes in the expression of mitotic spindle checkpoint genes could be associated to the chromosomal alterations detected in primary breast cancer we quantitatively measured the transcripts of two of the major mitotic checkpoint genes BUB1B and MAD2L1. In a preliminary study we determined an high variability in the level of expression of those two genes in a series of 5 fresh frozen pathologically normal breast tissues (data not shown), thus the analysis was performed by determining the amount of the transcripts in tumor specimen as relative expression to its matched normal tissue (see Section 2). Representative results for BUB1B and MAD2L1 transcripts quantitative analysis are shown in Fig. 2. The analysis could be performed on 10 of the 19 primary breast cancer included in the study, for other five cases not enough RNA was obtained from the normal tissues, and for the remaining four cases normal tissue was not available. Table 3 shows results for the BUB1B and Table 3 Quantitative determination of BUB1B and MAD2L1 transcripts in primary breast cancer BUB1B RE BT2 BT3 BT5 BT8 BT9 BT11 BT12 BT17 BT18 BT19

a

128.29 1.89 3.69 1.99 2.07 28.73 7.63 7.48 1.45 5.11

MAD2L1 b

Range

REa

Rangeb

115.36–142.67 1.53–2.29 3.40–4.00 1.78–2.23 1.81–2.34 26.92–30.67 6.34–9.16 6.91–8.10 1.27–1.66 3.70–7.06

91.54 25.51 2.67 0.82 4.51 0.41 3.86 34.89 1.67 0.69

84.97–98.617 20.85–31.22 2.57–2.78 0.69–0.99 3.79–5.39 0.37–0.45 3.55–4.19 29.12–41.80 1.54–1.81 0.59–0.80

a RE, BUB1B or MAD2L1 expression in the tumor sample relative to the expression in the matched normal breast tissue. b The range given for BUB1B or MAD2L1 in the tumor sample relative to expression in the matched normal breast tissue is determined by evaluating the expression 2DDCt with DDCt + S and DDCt  S, where S is the standard deviation of DDCt value. The standard deviation of the DDCt value is calculated from the standard deviation of the BUB1B or MAD2L1 (S1) and Human RPLP09 (S2) standard deviation of the Ct value according to the qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi following formula S 21 þ S 22 (User Bullettin #2 ABI PRISM 7700

Sequence Detection System pag 15).

MAD2L1 genes analysis. An overall increase in expression is seen for BUB1B as compared to its normal counterpart with relative expression values ranging from 1.45 to 128.29. For the MAD2L1 gene tumor BT11 showed a reduced expression in the gene as compared with matched normal (RE = 0.41), BT19 and BT8 displayed a moderate decrease of the transcript in the tumor (RE = 0.69 and RE = 0.82, respectively), and seven tumors showed increased mRNA levels in the tumor as compared with matched normal sample (range: 1.67–91.54). A statistically significant correlation was found between BUB1B and MAD2L1 mRNA levels by linear regression analysis (r = 0.879, P = 0.001). Interestingly BUB1B mRNA levels but not MAD2L1 levels correlated with intrachromosomal instability (r = 0.722, P = 0.018).

4. Discussion The level of genomic instability was analyzed in a series of ductal breast carcinomas. Chromosomal and intrachromosomal instability were evaluated on isolated nuclei using dual colour FISH analysis for chromosome 8, 13 and 17. Although chromosome-wide investigation indicated that chromosome numerical abnormality in common cancers are not random [27,28], several studies have shown that the analysis of three chromosomes is sufficient to segregate diploid from aneuploidy tumors and to identify clonal populations with a great level of certainty [7,20]. High clonal heterogeneity is the likely result of aneuploidy originating from chromosomal instability that generates multiple unrelated clones [20,29]. In the present study all cancer specimens showed some degree of aneuploidy although the modal chromosome number for the majority of them was diploid, with only four tumors with modal chromosomal number different from two signals for each of the probes. These results are in agreement with karyotypic studies on cell lines and primary culture that also show variable levels of chromosome numerical changes in breast cancer [17–19,30,31]. The average levels of chromosomal instability ranged from 36.2 for tumor BT6 to 76.8 for tumor BT11. Whereas intrachromosomal instability average rate ranged from 8.5 for tumor BT1 to 35.8 for tumor BT2 (Tables 1 and 2). Since, changes in karyotype have been associated with clinicopathological feature such as tumour types, stage and mitotic index [17], we chose to analyze only tumors of the ductal type less than 2 cm in diameter (pT1b and pT1c). Thus the variations detected in chromosomal and intrachromosomal instability rates are likely to reflect a true different

M. Scintu et al. / Cancer Letters 254 (2007) 298–307

biological behaviour rather than differences related to tumor stages and histopathological subtypes. It is uncertain whether the level of variability measured in normal breast tissues represents the background error in our measure, likely due to different efficiency in probe hybridisation, or a modest level of instability in normal breast from cancer patients. However, in both cases, the difference between chromosomal and intrachromosomal instability in tumor samples is much greater than in normal breast tissues. It is still an ongoing debate whether aneuploidy is an essential contributor to tumorigenesis or merely a remnant of oncogenic transformation [14]. Indeed, most oncogenes and tumor suppressor genes have an effect on tumor proliferation and cell survival. On the other side the ability of a cell to redistribute whole chromosome could facilitate tumorigenesis by increasing the chances of loss of heterozygosity of a tumor suppressor genes or in amplifying an oncogene by duplicating the chromosome. We found a significant association between HER2neu protein expression and average and chromosome 17 alterations. Similar results were reported recently by Igarashi et al. [32] that found an association between Her2neu membrane surface expression and chromosome 17 copy number. Thus, our results indicate a direct relationship between chromosomal instability rates leading to aneuploidy status, and HER2neu oncogenic activation. In a subset of tumor, results from genomic instability analysis were correlated with the level of mRNA expression of BUB1B and MAD2L1 mitotic spindle checkpoint genes. In a preliminary study we found a significant variability in DCt values in normal tissue that we considered as a major issue in determining cut-off values for gene expression analysis. This variability is most likely due to differences in normal breast tissue related to age or to the hormonal status that may affect significantly gene expression. Thus for each gene we performed the analysis on paired normal and tumor tissues, an approach that eliminates the bias due to variability between individuals. An increase in BUB1B and MAD2L1 transcript was detected in the majority of the tumors. These results are surprising because previous studies in colon, ovarian and also some breast cancer cell lines indicated a reduced expression of mitotic checkpoint genes that could have functional significance [13,21,33,34]. However, our results are consistent with a recent report by Yuan et al. [22] that also shows an increase in transcripts

305

of mitotic spindle checkpoint genes including BUB1B and MAD2L1 in a series of breast cancer cell line and primary tumors, and from microarray analysis data that also support our results [35,36]. BUB1B and MAD2L1 transcripts levels were strongly associated (r = 0.879 P = 0.001), thus the same mechanism seems to be responsible for the increased expression. When genomic instability data were compared with expression levels we found that BUB1B levels of mRNA expression correlated with intrachromosomal instability (r = 0.722, P = 0.018) but not with chromosomal instability. A possible explanation of this observation could be that the increased expression of the BUB1B gene represents a cellular response to the alterations of mechanisms involved in the maintenance of chromosome integrity, such as a failure in the pathway devoted to DNA repair. Alternatively, the intrachromosomal instability status may cause a genetic imbalance at the chromosome 15 BUB1B gene locus that will be associated to the increased expression of the gene. Our results suggest that aneuploidy in primary breast cancer is a consequence of an overall genomic instability related to high rates of chromosomal numerical and structural changes. Moreover, the analysis of two major mitotic spindle checkpoint genes supports findings from recent reports in which this pathway is described as up regulated rather than down regulated [22,35,36]. Although, it cannot be ruled out that the increased expression of mitotic checkpoint gene itself may contribute to genome instability as was demonstrated in yeast for the BUB1 gene [37], it is more likely that it represents a compensation for alterations in other mechanisms responsible for the control of genomic stability. First, the total loss of the mitotic spindle checkpoint control is a catastrophic event even for cancer cells that leads to cell death [38]. Thus, the increased expression of the BUB1B and MAD2L1 genes could just represent a partial compensation for other defect in the mitotic spindle checkpoint. However, other mechanisms such as DNA repair and G1/S checkpoint are also implicated in chromosome stability and cytokinesis integrity and many genes involved in these pathways (e.g. BRCA1, BRCA2, CHK2, etc.) play a role in breast cancer development [11,39–41]. Therefore, a deeper analysis of all these control mechanisms and their relationship with instability at chromosomal level will be needed to better clarify the events leading to the aneuploid state of breast cancer cells.

306

M. Scintu et al. / Cancer Letters 254 (2007) 298–307

Acknowledgements This work is supported by a grant from the Ministero della Salute (Italian Ministry of Health) IRCCS 2005–2006, and by a 2005–2007 grant from the ‘‘Associazione Italiana Ricerca sul Cancro, AIRC’’ (Italian Association for Cancer Research).

[16]

[17]

References [18] [1] H. Yamamoto, K. Imai, M. Perucho, Gastrointestinal cancer of the microsatellite mutator phenotype pathway, J. Gastroenterol. 37 (2002) 153–163. [2] C. Lengauer, K.W. Kinzler, B. Vogelstein, Genetic instability in colorectal cancers, Nature 386 (1997) 623–627. [3] J.E. Cleaver, Cancer in xeroderma pigmentosum and related disorders of DNA repair, Nat. Rev. Cancer 5 (2005) 564– 573. [4] C. Lengauer, K.W. Kinzler, B. Vogelstein, Genetic instabilities in human cancers, Nature 396 (1998) 643–649. [5] D.S. Yoon, R.P. Wersto, W. Zhou, F.J. Chrest, E.S. Garrett, T.K. Kwon, E. Gabrielson, Variable levels of chromosomal instability and mitotic spindle checkpoint defects in breast cancer, Am. J. Pathol. 161 (2002) 391–397. [6] K.M. Minhas, B. Singh, W.W. Jiang, D. Sidransky, J.A. Califano, Spindle assembly checkpoint defects and chromosomal instability in head and neck squamous cell carcinoma, Int. J. Cancer 107 (2003) 46–52. [7] W.L. Lingle, S.L. Barrett, V.C. Negron, A.B. D’Assoro, K. Boeneman, W. Liu, C.M. Whitehead, C. Reynolds, J.L. Salisbury, Centrosome amplification drives chromosomal instability in breast tumor development, Proc. Natl. Acad. Sci. USA 99 (2002) 1978–1983. [8] Z. Storchova, D. Pellman, From polyploidy to aneuploidy, genome instability and cancer, Nat. Rev. Mol. Cell Biol. 5 (2004) 45–54. [9] E.A. Nigg, Centrosome aberrations: cause or consequence of cancer progression?, Nat Rev. Cancer 2 (2002) 815–825. [10] P.V. Jallepalli, I.C. Waizenegger, F. Bunz, S. Langer, M.R. Speicher, J.M. Peters, K.W. Kinzler, B. Vogelstein, C. Lengauer, Securin is required for chromosomal stability in human cells, Cell 105 (2001) 445–457. [11] K.B. Kaplan, A.A. Burds, J.R. Swedlow, S.S. Bekir, P.K. Sorger, I.S. Nathke, A role for the adenomatous polyposis coli protein in chromosome segregation, Nat. Cell. Biol. 3 (2001) 429–432. [12] R.A. Green, K.B. Kaplan, Chromosome instability in colorectal tumor cells is associated with defects in microtubule plus-end attachments caused by a dominant mutation in APC, J. Cell Biol. 163 (2003) 949–961. [13] D.P. Cahill, C. Lengauer, J. Yu, G.J. Riggins, J.K. Willson, S.D. Markowitz, K.W. Kinzler, B. Vogelstein, Mutations of mitotic checkpoint genes in human cancers, Nature 392 (1998) 300–303. [14] G.J. Kops, B.A. Weaver, D.W. Cleveland, On the road to cancer: aneuploidy and the mitotic checkpoint, Nat. Rev. Cancer 5 (2005) 773–785. [15] B. Ouyang, J.A. Knauf, K. Ain, B. Nacev, J.A. Fagin, Mechanisms of aneuploidy in thyroid cancer cell lines and tissues: evidence for mitotic checkpoint dysfunction without

[19]

[20]

[21]

[22]

[23]

[24]

[25]

[26]

[27]

[28]

mutations in BUB1 and BUBR1, Clin. Endocrinol. (Oxf.) 56 (2002) 341–350. P.J. Hainsworth, K.L. Raphael, R.G. Stillwell, R.C. Bennett, O.M. Garson, Cytogenetic features of twenty-six primary breast cancers, Cancer Genet. Cytogenet. 53 (1991) 205–218. N. Pandis, I. Idvall, G. Bardi, Y. Jin, L. Gorunova, F. Mertens, H. Olsson, C. Ingvar, K. Beroukas, F. Mitelman, S. Heim, Correlation between karyotypic pattern and clincopathologic features in 125 breast cancer cases, Int. J. Cancer 66 (1996) 191–196. M.R. Teixeira, N. Pandis, G. Bardi, J.A. Andersen, S. Heim, Karyotypic comparisons of multiple tumorous and macroscopically normal surrounding tissue samples from patients with breast cancer, Cancer Res. 56 (1996) 855– 859. S. Takami, C. Kawasome, M. Kinoshita, H. Koyama, S. Noguchi, Chromosomal instability detected by fluorescence in situ hybridization in Japanese breast cancer patients, Clin. Chim. Acta 308 (2001) 127–131. F. Farabegoli, D. Santini, C. Ceccarelli, M. Taffurelli, D. Marrano, N. Baldini, Clone heterogeneity in diploid and aneuploid breast carcinomas as detected by FISH, Cytometry 46 (2001) 50–56. M.J. Percy, K.A. Myrie, C.K. Neeley, J.N. Azim, S.P. Ethier, E.M. Petty, Expression and mutational analyses of the human MAD2L1 gene in breast cancer cells, Genes Chromosomes Cancer 29 (2000) 356–362. B. Yuan, Y. Xu, J.H. Woo, Y. Wang, Y.K. Bae, D.S. Yoon, R.P. Wersto, E. Tully, K. Wilsbach, E. Gabrielson, Increased expression of mitotic checkpoint genes in breast cancer cells with chromosomal instability, Clin. Cancer Res. 12 (2006) 405–410. K.A. Myrie, M.J. Percy, J.N. Azim, C.K. Neeley, E.M. Petty, Mutation and expression analysis of human BUB1 and BUB1B in aneuploid breast cancer cell lines, Cancer Lett. 152 (2000) 193–199. P. Parrella, Y. Xiao, M. Fliss, M. Sanchez-Cespedes, P. Mazzarelli, M. Rinaldi, T. Nicol, E. Gabrielson, C. Cuomo, D. Cohen, S. Pandit, M. Spencer, C. Rabitti, V.M. Fazio, D. Sidransky, Detection of mitochondrial DNA mutations in primary breast cancer and fine-needle aspirates, Cancer Res. 61 (2001) 7623–7626. S. Xiao, A. Renshaw, E.S. Cibas, T.J. Hudson, J.A. Fletcher, Novel fluorescence in situ hybridization approaches in solid tumors. Characterization of frozen specimens, touch preparations, and cytological preparations, Am. J. Pathol. 147 (1995) 896–904. P. Parrella, O.L. Caballero, D. Sidransky, S.L. Merbs, Detection of c-myc amplification in uveal melanoma by fluorescent in situ hybridization, Invest. Ophthalmol. Vis. Sci. 42 (2001) 1679–1684. R. Nakamura, J.P. Song, J. Isogaki, Y. Kitayama, H. Sugimura, Multiple (multicentric and multifocal) cancers in the ipsilateral breast with different histologies: profiles of chromosomal numerical abnormality, Jpn. J. Clin. Oncol. 33 (2003) 463–469. Y. Kitayama, H. Igarashi, H. Sugimura, Different vulnerability among chromosomes to numerical instability in gastric carcinogenesis: stage-dependent analysis by FISH with the use of microwave irradiation, Clin. Cancer Res. 6 (2000) 3139–3146.

M. Scintu et al. / Cancer Letters 254 (2007) 298–307 [29] T. Kuukasjarvi, M. Tanner, S. Pennanen, R. Karhu, O.P. Kallioniemi, J. Isola, Genetic changes in intraductal breast cancer detected by comparative genomic hybridization, Am. J. Pathol. 150 (1997) 1465–1471. [30] P. Devilee, R.F. Thierry, T. Kievits, R. Kolluri, A.H. Hopman, H.F. Willard, P.L. Pearson, C.J. Cornelisse, Detection of chromosome aneuploidy in interphase nuclei from human primary breast tumors using chromosome-specific repetitive DNA probes, Cancer Res. 48 (1988) 5825–5830. [31] B. Dutrillaux, M. Gerbault-Seureau, Y. Remvikos, B. Zafrani, M. Prieur, Breast cancer genetic evolution: I. Data from cytogenetics and DNA content, Breast Cancer Res. Treat. 19 (1991) 245–255. [32] H. Igarashi, K. Yamashita, M. Suzuki, Y. Kitayama, J. Isogaki, K. Maruyama, K. Sunayama, H. Tsuda, T. Ozawa, S. Kiyose, H. Sugimura, Simultaneous imaging of membrane antigen and the corresponding chromosomal locus in pathology archives, Pathol. Int. 55 (2005) 753–756. [33] M. Shichiri, K. Yoshinaga, H. Hisatomi, K. Sugihara, Y. Hirata, Genetic and epigenetic inactivation of mitotic checkpoint genes hBUB1 and hBUBR1 and their relationship to survival, Cancer Res. 62 (2002) 13–17. [34] X. Wang, D.Y. Jin, R.W. Ng, H. Feng, Y.C. Wong, A.L. Cheung, S.W. Tsao, Significance of MAD2 expression to mitotic checkpoint control in ovarian cancer cells, Cancer Res. 62 (2002) 1662–1668. [35] H. Dai, L. van’t Veer, J. Lamb, Y.D. He, M. Mao, B.M. Fine, R. Bernards, M. van de Vijver, P. Deutsch, A. Sachs, R. Stoughton, S. Friend, A cell proliferation signature is a marker of extremely poor outcome in a

[36]

[37]

[38]

[39]

[40]

[41]

307

subpopulation of breast cancer patients, Cancer Res. 65 (2005) 4059–4066. R. Rouzier, C.M. Perou, W.F. Symmans, N. Ibrahim, M. Cristofanilli, K. Anderson, K.R. Hess, J. Stec, M. Ayers, P. Wagner, P. Morandi, C. Fan, I. Rabiul, J.S. Ross, G.N. Hortobagyi, L. Pusztai, Breast cancer molecular subtypes respond differently to preoperative chemotherapy, Clin. Cancer Res. 11 (2005) 5678–5685. C.D. Warren, D.M. Brady, R.C. Johnston, J.S. Hanna, K.G. Hardwick, F.A. Spencer, Distinct chromosome segregation roles for spindle checkpoint proteins, Mol. Biol. Cell 13 (2002) 3029–3041. G.J. Kops, D.R. Foltz, D.W. Cleveland, Lethality to human cancer cells through massive chromosome loss by inhibition of the mitotic checkpoint, Proc. Natl. Acad. Sci. USA 101 (2004) 8699–8704. M.J. Daniels, Y. Wang, M. Lee, A.R. Venkitaraman, Abnormal cytokinesis in cells deficient in the breast cancer susceptibility protein BRCA2, Science 306 (2004) 876–879. K. Shen, Y. Wang, S.C. Brooks, A. Raz, Y.A. Wang, ATM is activated by mitotic stress and suppresses centrosome amplification in primary but not in tumor cells, J. Cell Biochem. 99 (2006) 1267–1274. M. Grigorova, J.M. Staines, H. Ozdag, C. Caldas, P.A. Edwards, Possible causes of chromosome instability: comparison of chromosomal abnormalities in cancer cell lines with mutations in BRCA1, BRCA2, CHK2 and BUB1, Cytogenet. Genome Res. 104 (2004) 333–340.