Cancer Research Targets Advances in Breast Cancer Research
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Cancer Research 67, 5134, June 1, 2007. doi: 10.1158/0008-5472.CAN-07-0465
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Vincent-Salomon, A.
Right arrow Articles by Heard, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vincent-Salomon, A.
Right arrow Articles by Heard, E.

Molecular Biology, Pathobiology, and Genetics

X Inactive–Specific Transcript RNA Coating and Genetic Instability of the X Chromosome in BRCA1 Breast Tumors

Anne Vincent-Salomon1,3,4, Carine Ganem-Elbaz1,2, Elodie Manié1,3, Virginie Raynal1,3, Xavier Sastre-Garau4, Dominique Stoppa-Lyonnet1,3,4, Marc-Henri Stern1,3,4 and Edith Heard1,2

1 Institut Curie, Centre de Recherche; 2 Centre National de la Recherche Scientifique Unité Mixte de Recherche 218, Institut Curie; 3 Institut National de la Santé et de la Recherche Médicale U830; and 4 Department of Tumor Biology, Institut Curie, Paris, France

Requests for reprints: Edith Heard, Centre National de la Recherche Scientifique Unité Mixte de Recherche 218, Institut Curie, Centre de Recherche, 26 rue d'Ulm, Paris F-75248, France. Phone: 33-1-42-34-66-91; Fax: 33-1-46-33-30-16; E-mail: edith.heard{at}curie.fr or Marc-Henri Stern, Institut National de la Santé et de la Recherche Médicale U830 Institut Curie, Centre de Recherche, 26 rue d'Ulm, Paris F-75248, France. Phone: 33-1-42-34-66-46; Fax: 33-1-42-34-66-30; E-mail: marc-henri.stern{at}curie.fr.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Identification among breast tumors of those arising in a hereditary BRCA1 context remains a medical challenge. Abnormalities in X chromosome copy number and in the epigenetic stability of the inactive X chromosome (Xi) have been proposed to characterize BRCA1 breast tumors. In particular, it has been proposed that loss of BRCA1 function can lead to loss of X inactive–specific transcript (XIST) RNA association with the Xi. However, few studies have addressed this issue in a sufficiently large series of BRCA1 primary tumors. Here we assess X-chromosome status using single-cell (RNA and DNA fluorescence in situ hybridization) and global genomic (array-comparative genomic hybridization and allelotyping) approaches on a series of 11 well-defined BRCA1 tumors. We show that many or most cells of the tumors contain one or more XIST RNA domains. Furthermore, the number of XIST RNA domains per cell varied considerably even within a single tumor. Frequent X-chromosome allelic and copy number aberrations were found, in agreement with aberrant XIST RNA domain numbers. In summary, by combining multiple approaches to assess the genetics and epigenetics of a large series of BRCA1 primary tumors, we can conclude definitively that BRCA1 is not required for XIST RNA coating of the X chromosome. The intratumoral and intertumoral variability in XIST RNA domain number in BRCA1 tumors correlates with chromosomal genetic abnormalities, including gains, losses, reduplications, and rearrangements of the X-chromosome. Finally, we also show the necessity for combined global and single-cell approaches in the assessment of tumors with such a high degree of heterogeneity. [Cancer Res 2007;67(11):5134–40]


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Women carrying germ-line deleterious mutations in the BRCA1 gene have an ~80% lifetime risk of developing breast cancer, which represents a >8-fold relative risk compared with the general population (1). Thus, BRCA1 behaves as a classic tumor suppressor gene because the systematic inactivation of the remaining wild-type allele is a key event of the malignant process in BRCA1 tumors (2). However, the exact reasons underlying this predisposition have remained unclear. As BRCA1 is implicated in a wide variety of processes that can impinge on genomic integrity, including DNA repair, recombination, transcription, chromatin remodeling, and G2-M checkpoints, it is thought that BRCA1 inactivation is likely to induce genetic instability and thus lead to, or participate in, carcinogenesis (for review, see ref. 3). Furthermore, a lack of BRCA1 has been proposed to induce epigenetic instability, which could in turn affect genomic integrity. Several potential links between BRCA1 function and the stable maintenance of heterochromatic structures have been reported, both in the context of constitutive heterochromatin (4) and the facultative heterochromatin of the inactive X chromosome (Xi; ref. 5). Indeed, it has been proposed that loss of BRCA1 function can lead to the loss of association of the X inactive–specific transcript (XIST) transcript from the Xi in female cells (5). This, in turn, may result in X chromosome reactivation through the destabilization of epigenetic marks such as histone modifications, which are known to contribute to the stable maintenance of the inactive state (for review, see ref. 6). The XIST gene and the noncoding functional RNA it produces are known to be essential for the initiation of X inactivation during early development (7, 8). However, it is not thought to be essential for the maintenance of X inactivation in somatic cells, although it can recruit some of the epigenetic marks involved in the stability of the inactive state (for review, see ref. 9), and loss of XIST combined with a loss of DNA methylation can lead to slightly increased levels of X-linked gene reactivation in somatic cells (10).

Although there is evidence linking functional BRCA1 with XIST RNA coating of the X chromosome in cultured cells (5), a recent study has shown that the BRCA1 protein does not show a major association with the Xi (11). The degree to which BRCA1 loss of function generates epigenetic, rather than genetic, instability of the X chromosome in primary tumor samples has remained unclear, however. This is an important issue for the interpretation of longstanding reports about the lack of a cytologically heteropyknotic Barr body in human cancer cells (for review, see ref. 12). The cytologic loss of the Barr body was widely assumed to be due to genetic loss of the inactive X chromosome rather than to its reactivation and decondensation. However, the study of Ganesan et al. (5) suggested that the latter might indeed be a frequent event in a BRCA1 mutant context. Further studies, mainly in cultured breast tumor cell lines, have shown that loss of the Xi and reduplication of the active X (Xa) chromosome (X chromosome isodisomy) are a frequent event in breast cancer (13), particularly in basal-like tumors (14). Very few primary tumors from BRCA1 patients have been analyzed, however, particularly in the context of the proposed role for BRCA1 in XIST RNA coating and epigenetic stability of the inactive X chromosome (5, 13, 14). We set out to address this issue using a panel of carefully characterized tumor samples from patients carrying deleterious germ-line BRCA1 mutations. XIST RNA coating and X chromosome integrity were analyzed using single-cell fluorescence in situ hybridization (FISH) approaches and genomic integrity was assessed using array-comparative genomic hybridization (array-CGH) and allelotyping.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Tumor sample selection and histology. Eleven tumors from patients with germ-line deleterious BRCA1 mutations were selected based on the availability of frozen tumor and normal samples. The BRCA1 mutations, listed in Table 1 , have previously been determined from blood samples according to published methods (15). Truncating BRCA1 mutations were selected, given their unquestionable deleterious properties. Deletion of the second allele was verified in nine of these tumors by array-CGH and allelotype analyses. In addition, frozen tumor samples from four randomly chosen sporadic luminal ductal breast carcinomas were also studied (Table 2 ). A hematein-eosin-safran–stained tissue section was also made in each case to evaluate tumoral cellularity. Normal lobules and ducts were assessable and interpretable in only two of the BRCA1 normal tissue samples (B1-N28 and B1-N07). Characterization of the tumor samples was completed by the determination of estrogen receptor, progesterone receptor, ERBB2, cytokeratin 5/6, and epidermal growth factor receptor (EGFR) status determined by immunohistochemistry done according to previously published protocols (16).


View this table:
[in this window]
[in a new window]

  		Table 1. Patient and tumor characteristics

 

View this table:
[in this window]
[in a new window]

  		Table 2. XIST RNA FISH analysis of sporadic breast tumors

 
RNA and DNA FISH and immunofluorescence. Procedures and materials used for combined RNA FISH and immunofluorescence, as well as DNA FISH, are detailed on the Epigenome Network of Excellence web site.5 For immunostaining and RNA FISH, 8-µm cryosections were generated from frozen tumor samples and immediately fixed in 3% paraformaldehyde/PBS for 10 min, then permeabilized in 1x PBS/0.5% Triton X-100/2 mmol/L vanadyl ribonucleoside complex (New England Biolabs) on ice for 4 min. After three washes in PBS, immunostaining was done in the presence of RNA-Guard (Amersham, GE Healthcare) at 40 units/mL for all antibody incubations, using previously described protocols (17). Anti-KL1 antibody (Dako) was used at a dilution of 1:300 and anti-H3K27me3 antibody (Upstate) was used at a dilution of 1:300. Secondary antibodies were goat anti-rabbit Alexa 488 and goat anti-mouse Alexa 680 (Molecular Probes, Invitrogen). Following immunostaining, the sample was fixed for 10 min in 3% paraformaldehyde and then dehydrated through an ethanol series of washes. RNA FISH was then done by hybridization with a XIST probe overnight at 37°C in a dark and humid chamber. After three washes in 50% formamide/2x SSC and three washes in 2x SSC at 42°C, and 4',6-diamidino-2-phenylindole (DAPI) counterstaining, coverslips were mounted on slides and visualized with a fluorescence microscope. The DNA probe detecting XIST RNA was a 10-kb fragment corresponding to XIST exon 1 (gift from Dr. Carolyn Brown, Department of Medical Genetics, University of British Columbia, Vancouver, British Columbia, Canada), labeled by nick translation (Vysis, Abbott) with SpectrumRed (Vysis) fluorescent dUTP.

For DNA FISH experiments, frozen section were prepared as above, but after fixation and permeabilization were kept in 70% ethanol at –20°C. Before denaturation, the samples were dehydrated through progressive ethanol washes. They were then denatured in 70% formamide/2x SSC at 75°C for 10 min and rinsed several times in ice-cold 70% ethanol followed by dehydration, as above, and then overnight hybridization with fluorescently labeled probes at 42°C in a dark, humid chamber. DNA FISH probes were denatured, competed with Cot-1 DNA (Invitrogen; 2 µg in 10-µL hybridization volume) for 1 h at 37°C before hybridization. DNA FISH hybridization and washes were done using similar conditions to the RNA FISH described above, except wash temperature was 44°C. For X inactivation center (XIC) detection, a BAC clone covering the XIST locus (RP11-13M9) labeled with SpectrumRed was used; for the centromere of the X chromosome, a FITC-labeled centromeric probe CEN-X (Cambio) was used.

Microscopy and image analysis. Three-dimensional images were captured with a Zeiss Axiovert microscope equipped with an Apotome (Zeiss) for three-dimensional acquisitions, an ORCA CCD camera (Hamamatsu), and piloted with Axiovision software (Zeiss). Using a 63x objective, images of ~50 optical sections separated by 0.2 µm were acquired for DAPI (360/40, 470/40), SpectrumGreen (470/40, 525/50), SpectrumRed (545/30, 610/75), and Alexa 680 (620/60, 700/75). On average, 200 nuclei were scored for the presence of XIST RNA or for XIC and CEN-X DNA signals. Tumor cells were identified morphologically (at the DAPI level) and through KL1 staining patterns. Adjacent tumor sections stained with H&E were also used to help identify regions of malignant cells.

Array-CGH. The pan-genomic array used contains 3,922 bacterial artificial chromosome (BAC) and P1-derived artificial chromosome (PAC) DNAs, among which 126 are located on the X chromosome (18). Positions of BAC and PAC on the National Center for Biotechnology Information (NCBI) Build 36.1 reference sequence were determined by end sequencing. Briefly, tumor and normal male DNAs were labeled with Cy5-dCTP or Cy3-dCTP, respectively. Labeled tumor and reference DNAs were purified on Microcon column (Millipore), coprecipitated with 120 µg of human Cot-1 DNA, resuspended in 70 µL of hybridization buffer (50% formamide, 40 mmol/L NaH2PO4, 0.1% SDS, 10% dextran sulfate, 2x SSC), denatured, prehybridized for 90 min at 37°C, and hybridized on treated microarray slides. After 24 h of hybridization at 37°C in a humidity chamber and appropriate washing, slides were scanned using a GenePix 4000B scanner (Axon) and analyzed with GenePix Pro 5.1 image analysis software (Axon), which determined the median intensities for Cy3 and Cy5 signals of each BAC clone. Ratios of Cy5/Cy3 signals observed for each BAC clone were normalized according to the MANOR routine (19). The GLAD method was applied to determine the status of the X chromosome (20).

Allelotyping. Twelve microsatellite markers on the X chromosome (DXS1060, DXS1226, DXS1214, DXS1068, DXS993, DXS991, DXS990, DXS8055, DXS1001, DXS1047, DXS1227, and DXS8043) were selected from the Linkage Mapping set MD10 (Applied Biosystems). Microsatellites were mapped on the NCBI Build 36.1 reference sequence. Microsatellite typing was achieved using PCR and a 3130xl Genetic Analyzer. An allelic ratio was determined using Genemapper V4.0 (Applied Biosystems). Allelic status was defined as retention of heterozygosity (ROH) if |log 2(allelic ratio)| < 0.262; allelic imbalance (AI) if 0.262 < |log 2(allelic ratio)| < 1; and loss of heterozygosity (LOH) if |log 2(allelic ratio)| ≥ 1.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
BRCA1 mutation and tumor characterization. A panel of 11 tumors from 10 patients carrying defined BRCA1 mutations were analyzed for XIST RNA status and X chromosome integrity. In all cases, the BRCA1 mutations, which were scattered across the locus, led to putative truncated proteins. The mutations were all different except for two. The exact BRCA1 mutations and the immunohistochemical and morphologic characteristics of these tumors are shown in Table 1. All BRCA1 tumors were of ductal type and were high grade in 10 of 11 (90%) cases. To determine whether these tumors have basal-like characteristics as has been reported for the majority of BRCA1 tumors (21), we determined their immunophenotype. All tumors but one were found to be basal-like as they were indeed estrogen receptor, progesterone receptor, and ERBB2 negative and cytokeratin 5/6 positive. Eight of the tumors were also strongly EGFR positive (Table 1).

Assessment of XIST RNA domain status in BRCA1 mutant tumors. To assess the association of XIST RNA with the X chromosome in BRCA1-deficient tumors, we analyzed frozen sections of these tumors, as well as of nearby normal tissue, using XIST RNA FISH. We also combined the RNA FISH analysis with immunofluorescence detection of the pan-cytokeratin KL1 marker. This enables positive detection of epithelial tumoral cells, which facilitates their distinction from normal stromal cells that were KL1 negative. This was particularly important to assess whether, within the tumor sample, XIST RNA–positive cells were, in fact, tumoral or normal cells. In normal tissue samples, XIST RNA domains, corresponding to the Xi, were detected as expected (e.g., B1_N07; Fig. 1A and data not shown). The efficiency of detection in normal tissues was in the order of 50% and was rather variable owing to the small number of glandular cells present within what was usually predominantly fat tissue. When we examined the BRCA1 tumor samples, we were surprised to find that XIST RNA domains could be readily detected in >35% of the KL1-positive cells in the majority (6 of 11) of tumors. A complete lack of XIST RNA domains in tumor cells could be found in only one case, B1_T07B (Fig. 1A). In this tumor, the only XIST RNA–positive cells were nonmalignant KL1-negative cells (Fig. 2 ). On the other hand, in 10 of the 11 tumors, XIST RNA domains could be detected in some or most of the tumor cells, often with higher efficiency than in the corresponding normal tissue (Fig. 1A and data not shown). In four tumors, although the number of KL1-positive cells with XIST RNA domains was <10%, these were very clearly tumor cells based on KL1 staining and nuclear morphology (Fig. 1). We were rather surprised at this result given the previously reported findings of XIST RNA status in BRCA1 tumors (5, 14), although it should be noted that in these studies only a limited number (four) of BRCA1 tumors was examined. Furthermore, in 6 of the 11 BRCA1 tumors we analyzed, more than one XIST RNA domain per nucleus could be detected in some cells (Fig. 1A), unlike in normal tissues and four sporadic tumors analyzed, where only one XIST RNA domain per nucleus could be detected (Table 2). The number of cells showing zero, one, or more XIST RNA domains varied considerably, both between and within tumors (Fig. 1C). In fact, even between two asynchronous independent tumors from the same patient, the number of XIST RNA domains per cell differed dramatically (B1_T13A showing 38% cells with one XIST RNA domain and 41% with more than one domain, whereas B1_T13B showed only 5% of cells with one XIST RNA domain per cell; see Fig. 1C). Furthermore, patients with the same BRCA1 mutation showed completely different patterns of XIST RNA–positive cells (compare B1_T17 and B1_T13A and B). This striking degree of heterogeneity suggested genetic instability of the Xi in BRCA1 tumors independent of the primary BRCA1 mutation itself.


Figure 1
View larger version (46K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. XIST RNA FISH and X-chromosome DNA FISH analysis in BRCA1 tumors. A, representative examples of XIST RNA FISH (red) combined with anti-KL1 immunofluorescence (IF; yellow) are shown for part of a normal gland (B1_N07) and three tumors (B1_TO7, B1_T03, and B1_T06). DNA is counterstained with DAPI (blue). One XIST RNA domain per nucleus is seen in normal tissue; no XIST RNA domains could be detected in the B1-T07 tumor; whereas in B1-T03 and B1_T07, various numbers of XIST RNA domains/nucleus could be detected in KL1-positive tumor cells and the XIST RNA patterns varied considerably between cells within the same sample. B, representative examples of dual DNA FISH using an XIC probe (red) and a CEN-X probe (green) on the same tumors as shown in (A). DNA is counterstained with DAPI (blue). A panel of cells is shown above and two representative nuclei are enlarged below in each case. C, a summary of the XIST RNA FISH data as well as of the XIC and CEN-X DNA FISH data is shown for the 11 BRCA1 tumors examined and for the adjacent normal tissue from two of the patients concerned. An average of 200 nuclei were scored in each case. The status of the XIC region based on array-CGH and allelotyping is also shown for comparative purposes. In the DNA FISH analysis, the numbers of signals that were closely linked were scored and in cases where only one of the two signals was detected, whatever the total number, this was categorized as "Other." This category can either be explained by the fact that FISH detection efficiency is not 100% or by genetic aberrations leading to separation/absence of XIC and centromeric regions. Note that XIST RNA FISH detection was in the order of 58% in normal cells (B1_N07), which correlates well with the percentage of cells (55%) showing two linked XIC and CEN-X signals per nucleus detected by DNA FISH.

 

Figure 2
View larger version (40K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Detection of XIST RNA–positive/KL1-negative stromal cells. A representative example of a tumor sample (B1_T07) stained with anti-KL1 to reveal tumoral epithelial (KL1+) versus normal stromal (KL1–) cells, analyzed for the presence of XIST RNA domains. Top, XIST RNA FISH (red) combined with DAPI (blue). Bottom, XIST RNA FISH (red) combined with anti-KL1 immunofluorescence (yellow). No XIST RNA domains could be detected in tumor cells and one XIST RNA domain per nucleus could be detected in normal cells (arrowheads).

 
Finally, we were not able to define any obvious epigenetic deficiencies in these tumors, using trimethylation of Lys27 of histone H3 (H3K27me3) as a marker, which is a well-known histone modification characterizing the inactive X chromosome (22). We found that a correlation between XIST RNA and H3K27me3 accumulation on the X chromosome was quite variable among cells, both in normal and tumor (BRCA1 and sporadic) samples (data not shown). Such cell-to-cell variability in H3K27me3 staining of the Xi has also been noted in normal somatic tissues of adult mice.6

Array-CGH and allelotyping analyses of X chromosome status in BRCA1 tumors. Given the above findings suggesting that the X chromosome is more genetically than epigenetically unstable, we set out to address X-chromosome stability in these 11 tumors using array-CGH. The copy number status of the X chromosome was defined as normal in six tumors, deleted in one, gained in one, and rearranged in three. Examples of array-CGH patterns are shown in Fig. 3A . To assess the allelic status of the X chromosome, we also allelotyped 12 well-defined polymorphic poly-CA microsatellites across the X chromosome in ten of these tumors. Examples of allelotype patterns are shown Fig. 3B. Status of the markers was defined (Fig. 3C) as (a) balanced in two tumors, indicating a ROH; (b) LOH in two tumors, indicating either the loss of one X chromosome or a reduplication of one X chromosome (isodisomy); and (c) AI in one sample, indicating either X-chromosome gain or loss of one X chromosome in the tumoral cells masked by contaminating normal cells in the sample. In two cases, the Xp arm had a ROH status whereas the Xq arm was either LOH or AI. The three other cases displayed complex allelic patterns. Combining these array-CGH and allelotyping analyses, we can thus define four classes of X-chromosome status (Fig. 3C). The "a" group (including tumors B1_T06 and B1_T17; allelotyping was not available for tumor B1_T28) had a normal status both at the allelic and array-CGH levels. The "b" group (tumors B1_T02, B1_T13A, and B1_T07B) had a rearranged X-chromosome status, with concordant allelic and array-CGH patterns. The "c" group (tumors B1_T15 and B1_T13B) had a normal array-CGH status and an AI or LOH allelic status indicating a putative isodisomic X chromosome. The "d" group (tumors B1_T03, B1_T05, and B1_T09) showed dissociated allelic and array-CGH status, indicative of complex aberrations of the X chromosome. Given our assessment of XIST RNA status, we also evaluated the XIC region by combining the array-CGH data and the allelotyping of the nearest distal DXS986 polymorphic marker (Fig. 1C). This analysis was in complete agreement with the XIST RNA status in 9 of 11 cases. In particular, tumors B1_T02 and B1_T07B were found to have lost the XIC region (Fig. 3C), in agreement with their total lack of XIST RNA domains. Moreover, the two tumors with a putative isodisomic XIC region were almost completely devoid of XIST RNA domains, which is in agreement with a duplication of the Xa and loss of the Xi. It is noteworthy that in the same patient, one tumor (B1_T13A) showed a normal XIC status whereas the other (B1_T13B) showed X chromosome isodisomy. However, for two tumors (B1_T09 and B1_T13A), the deduced XIC genomic status was inconsistent with XIST RNA status. As array-CGH and allelotyping approaches can only provide an average read-out from a cell population, we decided to assess the status of the X chromosome at the single-cell level by DNA FISH.


Figure 3
View larger version (26K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Array-CGH and allelotyping of the X chromosome in BRCA1 tumors. A, array-CGH profiles of three tumors. X axis, normalized fluorescence tumoral/normal ratios of individual clones represented by colored dots. Y axis, position on X chromosome. The normal status is defined by a tumoral/normal ratio of 1.4. B, allelotyping of three tumors. X axis, allelic value (AV = |log 2(allelic ratio)|). Y axis, position on X chromosome. Noninformative markers are not represented. C, combined representation of X chromosome status in the BRCA1 tumors. Colored bars, array-CGH status. Squares and diamonds, allelic status. Tumors are split into four groups: (a) normal array-CGH and allelic status; (b) concordant abnormal array-CGH and allelic status; (c) dissociated array-CGH and LOH results indicative of putative isodisomy; and (d) alternation of concordant and dissociated array-CGH and LOH results indicative of complex X chromosome rearrangements. A to C, left, pictogram of the X chromosome; arrow, XIST locus. Array-CGH—green, blue, and red dots or bars, loss, normal, and gain status, respectively. Allelic status—black squares, gray diamonds, and black diamonds, ROH (AV < 0.262), AI (0.262 < AV < 1), and LOH (AV ≥ 1) status, respectively.

 
DNA FISH assessment of the X chromosome in BRCA1-deficient tumors. Given the striking heterogeneity in the number of XIST RNA FISH signals per nucleus detected in BRCA1-deficient tumors, an X-chromosome paint probe was used to try and assess X chromosome status. However, the complexity of the signals that we obtained and the fact that tumor cell nuclei are often irregular in shape and size meant that a systematic analysis of X chromosome ploidy by X chromosome painting was deemed to be inadequate (data not shown). We therefore used a dual DNA FISH approach using an X chromosome centromere (CEN-X) and an XIC-specific BAC probe (XIC) on frozen tumor sections equivalent to those used for the XIST RNA FISH analysis. In this way, we were able to assess the copy number of X chromosome centromeric and XIC regions simultaneously and at the single-cell level with high efficiency. This analysis (Fig. 1B and C) revealed that X chromosome status in these tumor samples is indeed highly variable and often complex. In most of the tumors examined, we could define nuclei containing one, two, three, or four linked XIC/CEN-X signals. In B1_T07B, the majority of nuclei presented only one XIC and CEN-X signal, as expected, for loss of the inactive X chromosome and absence of XIST RNA coating. In the two tumors B1_T09 and B1_T13A, in which the deduced XIC genomic status was inconsistent with the XIST RNA status, we detected cells with one, two, or three sets of signals and a large proportion of cells with more complex combinations of fragmented (unlinked) XIC/CEN signals. Such complex patterns were, in fact, present in the majority of tumors analyzed including those with more than one XIST RNA domain per cell (Fig. 1C). In summary, this DNA FISH analysis reinforces our conclusion, based on XIST RNA FISH analysis and array CGH and allelotyping, that the X chromosome is subject to substantial genetic rearrangements (loss, fragmentation, and/or gain) in these tumors. Furthermore, this analysis highlights the dramatic degree of cellular heterogeneity that can be found within a tumor sample.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Our results lead us to conclude definitively that BRCA1-deficient breast tumors do not show a systematic deficiency in XIST RNA coating of the X chromosome. However, our data reveal that the inactive X chromosome is prone to genetic instability in BRCA1 tumors and that this probably accounts for the lack of XIST RNA signal detected in the small sample of tumors analyzed in the studies by Ganesan et al. (5) and Richardson et al. (14). Our work is in agreement with the study of Sirchia et al. (13) based on breast cancer cell lines and a small number of primary tumors, which showed that there is a frequent loss of the Xi and reduplication of the Xa in breast tumorigenesis. However, our work in a more extensive panel of BRCA1 primary tumors also reveals that genetic instability can lead not only to the loss but also to an increase in the number of Xi. Our study highlights the necessity for combined global and single-cell approaches in the assessment of tumors with a high degree of genomic instability such as that found in a BRCA1-deficient background.

The nature or origin of the BRCA1 mutations examined cannot explain the different XIST RNA patterns observed because a spectrum of different mutations (including one Ashkenazi-type, B1_T28) were included in our study. Furthermore, independent tumors from the same patient showed differing XIST RNA status.

BRCA1-deficient and basal-like breast tumors have been proposed to be predisposed to genetic instability of the X chromosome and of some autosomes (14). However, in other series of BRCA1 tumors and basal-like tumors analyzed by array-CGH, and compared with sporadic breast tumors, X aberrations were not found to be differentially represented (23, 24). Thus, to assess definitively whether the X chromosome is preferentially rearranged in BRCA1 tumorigenesis compared with sporadic tumors, and compared with autosomal genetic instabilities in different types of breast cancer, complete global and single-cell approaches on large series of different subtypes of breast tumors would be required. Nevertheless, data suggesting that the Barr body is often lost in aging individuals or in cancer (for review, ref. 12) make it tempting to speculate that the inactive X chromosome could be prone to genetic instability in multiple tumor types, and that this may be linked to its unique expression status. However, we observed that BRCA1 tumors are often associated with cells showing multiple XIST RNA domains per nucleus. This contrasted sharply with four randomly chosen sporadic breast tumors, which never showed more than one XIST RNA domain per cell (Table 2). These observations would be consistent with a role for BRCA1 in chromosome segregation (4, 25). Indeed, the recent finding that BRCA1 is associated with constitutive (pericentric) heterochromatin rather than facultative heterochromatin (4) is consistent with our evidence that loss (or gain) of the Xi involves increased errors in chromosome segregation, rather than a generalized failure of heterochromatin maintenance as originally proposed by Ganesan et al. (5). Although we have shown that XIST RNA coating of the inactive X chromosome is unaffected by the lack of BRCA1, it will be interesting in the future to determine the extent to which genetic or epigenetic instability of the X chromosome, or chromosomal segregation errors, might participate in breast oncogenesis.


   Acknowledgments
 
Grant support: Canceropole Centre National de la Recherche Scientifique, Institut National de la Santé et de la Recherche Médicale (INSERM), and Institut Curie (Programme Incitatif et Cooperatif); the Carte d'Identité des Tumeurs (CIT) program of the Ligue Nationale Contre le Cancer, for the construction of the BAC array; "Interface INSERM" (A. Vincent-Salomon); and the Schlumberger Foundation (E. Heard).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank M. Guggiari, I. Lebigot, S. Berhouet, and V. Moncoutier for their excellent technical assistance; C. Brown for the XIST probe; and O. Delattre for his support and for access to unpublished data. This work is part of the "Canceropole Ile-de-France Breast Cancer and Epigenetics" program coordinated by G. Almouzni.


   Footnotes
 
Note: A. Vincent-Salomon, C. Ganem-Elbaz, and E. Manié contributed equally to this work.

5 http://www.epigenome-noe.net/researchtools/protocol.php?protid=3 Back

6 E. Heard and J. Chaumeil, unpublished data. Back

Received 2/ 5/07. Revised 3/18/07. Accepted 4/ 2/07.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Antoniou A, Pharoah PD, Narod S, et al. Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detected in case Series unselected for family history: a combined analysis of 22 studies. Am J Hum Genet 2003;72:1117–30.[CrossRef][Medline]
  2. Smith SA, Easton DF, Evans DG, Ponder BA. Allele losses in the region 17q12-21 in familial breast and ovarian cancer involve the wild-type chromosome. Nat Genet 1992;2:128–31.[CrossRef][Medline]
  3. Gudmundsdottir K, Ashworth A. The roles of BRCA1 and BRCA2 and associated proteins in the maintenance of genomic stability. Oncogene 2006;25:5864–74.[CrossRef][Medline]
  4. Pageau GJ, Lawrence JB. BRCA1 foci in normal S-phase nuclei are linked to interphase centromeres and replication of pericentric heterochromatin. J Cell Biol 2006;175:693–701.[Abstract/Free Full Text]
  5. Ganesan S, Silver DP, Greenberg RA, et al. BRCA1 supports XIST RNA concentration on the inactive X chromosome. Cell 2002;111:393–405.[CrossRef][Medline]
  6. Heard E, Disteche CM. Dosage compensation in mammals: fine-tuning the expression of the X chromosome. Genes Dev 2006;20:1848–67.[Abstract/Free Full Text]
  7. Penny GD, Kay GF, Sheardown SA, Rastan S, Brockdorff N. Requirement for Xist in X chromosome inactivation. Nature 1996;379:131–7.[CrossRef][Medline]
  8. Marahrens Y, Panning B, Dausman J, Strauss W, Jaenisch R. Xist-deficient mice are defective in dosage compensation but not spermatogenesis. Genes Dev 1997;11:156–66.[Abstract/Free Full Text]
  9. Heard E. Delving into the diversity of facultative heterochromatin: the epigenetics of the inactive X chromosome. Curr Opin Genet Dev 2005;15:482–9.[CrossRef][Medline]
  10. Csankovszki G, Nagy A, Jaenisch R. Synergism of Xist RNA, DNA methylation, and histone hypoacetylation in maintaining X chromosome inactivation. J Cell Biol 2001;153:773–84.[Abstract/Free Full Text]
  11. Pageau GJ, Hall LL, Lawrence JB. BRCA1 does not paint the inactive X to localize XIST RNA but may contribute to broad changes in cancer that impact XIST and Xi heterochromatin. J Cell Biochem 2006;175:693–701.
  12. Spatz A, Borg C, Feunteun J. X-chromosome genetics and human cancer. Nat Rev Cancer 2004;4:617–29.[CrossRef][Medline]
  13. Sirchia SM, Ramoscelli L, Grati FR, et al. Loss of the inactive X chromosome and replication of the active X in BRCA1-defective and wild-type breast cancer cells. Cancer Res 2005;65:2139–46.[Abstract/Free Full Text]
  14. Richardson AL, Wang ZC, De Nicolo A, et al. X chromosomal abnormalities in basal-like human breast cancer. Cancer Cell 2006;9:121–32.[CrossRef][Medline]
  15. Wagner T, Stoppa-Lyonnet D, Fleischmann E, et al. Denaturing high-performance liquid chromatography detects reliably BRCA1 and BRCA2 mutations. Genomics 1999;62:369–76.[CrossRef][Medline]
  16. Azoulay S, Lae M, Freneaux P, et al. KIT is highly expressed in adenoid cystic carcinoma of the breast, a basal-like carcinoma associated with a favorable outcome. Mod Pathol 2005;18:1623–31.[Medline]
  17. Chaumeil J, Le Baccon P, Wutz A, Heard E. A novel role for Xist RNA in the formation of a repressive nuclear compartment into which genes are recruited when silenced. Genes Dev 2006;20:2223–37.[Abstract/Free Full Text]
  18. Idbaih A, Marie Y, Pierron G, et al. Two types of chromosome 1p losses with opposite significance in gliomas. Ann Neurol 2005;58:483–7.[CrossRef][Medline]
  19. Neuvial P, Hupe P, Brito I, et al. Spatial normalization of array-CGH data. BMC Bioinformatics 2006;7:264.[CrossRef][Medline]
  20. Hupe P, Stransky N, Thiery JP, Radvanyi F, Barillot E. Analysis of array CGH data: from signal ratio to gain and loss of DNA regions. Bioinformatics 2004;20:3413–22.[Abstract/Free Full Text]
  21. Foulkes WD, Stefansson IM, Chappuis PO, et al. Germline BRCA1 mutations and a basal epithelial phenotype in breast cancer. J Natl Cancer Inst 2003;95:1482–5.[Abstract/Free Full Text]
  22. Plath K, Fang J, Mlynarczyk-Evans SK, et al. Role of histone H3 lysine 27 methylation in X inactivation. Science 2003;300:131–5.[Abstract/Free Full Text]
  23. Chin K, DeVries S, Fridlyand J, et al. Genomic and transcriptional aberrations linked to breast cancer pathophysiologies. Cancer Cell 2006;10:529–41.[CrossRef][Medline]
  24. Jonsson G, Naylor TL, Vallon-Christersson J, et al. Distinct genomic profiles in hereditary breast tumors identified by array-based comparative genomic hybridization. Cancer Res 2005;65:7612–21.[Abstract/Free Full Text]
  25. Joukov V, Groen AC, Prokhorova T, et al. The BRCA1/BARD1 heterodimer modulates ran-dependent mitotic spindle assembly. Cell 2006;127:539–52.[CrossRef][Medline]
  26. Elston CW, Ellis IO. Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology 1991;19:403–10.[Medline]



This article has been cited by other articles:


Home page
 JNCI J Natl Cancer InstHome page
F. Lose, D. L. Duffy, G. F. Kay, Kathleen Cuningham Foundation Consortium for Resea, M. A. Kedda, and A. B. Spurdle
Skewed X Chromosome Inactivation and Breast and Ovarian Cancer Status: Evidence for X-Linked Modifiers of BRCA1
J Natl Cancer Inst, November 5, 2008; 100(21): 1519 - 1529.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
I. O-Sullivan, A. Chopra, J. Carr, T. S. Kim, and E. P. Cohen
Immunity to Growth Factor Receptor-Bound Protein 10, a Signal Transduction Molecule, Inhibits the Growth of Breast Cancer in Mice
Cancer Res., April 1, 2008; 68(7): 2463 - 2470.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Vincent-Salomon, A.
Right arrow Articles by Heard, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vincent-Salomon, A.
Right arrow Articles by Heard, E.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online