Characterization at nucleotide resolution of the homogeneously staining region sites of insertion in two cancer cell lines

The mechanisms of formation of intrachromosomal amplifications in tumours are still poorly understood. By using quantitative polymerase chain reaction, DNA sequencing, chromosome walking, in situ hybridization on metaphase chromosomes and whole-genome analysis, we studied two cancer cell lines containing an MYC oncogene amplification with acquired copies ectopically inserted in rearranged chromosomes 17. These intrachromosomal amplifications result from the integration of extrachromosomal DNA molecules. Replication stress could explain the formation of the double-strand breaks involved in their insertion and in the rearrangements of the targeted chromosomes. The sequences of the junctions indicate that homologous recombination was not involved in their formation and support a non-homologous end-joining process. The replication stress-inducible common fragile sites present in the amplicons may have driven the intrachromosomal amplifications. Mechanisms associating break-fusion-bridge cycles and/or chromosome fragmentation may have led to the formation of the uncovered complex structures. To our knowledge, this is the first characterization of an intrachromosomal amplification site at nucleotide resolution.


Amplicon analysis.
The level of amplification was measured by real-time quantitative PCR using the 7500 Real-Time PCR System and SYBR Green PCR kits (Applied-Biosystems).
Amplification levels were calculated using a standard curve constructed with serial dilutions of control DNA amplified in a parallel experiment (DNA from normal lymphocytes). DNAs were extracted from lymphocytes and tumor using Qiagen kits.
All measurements were performed at least in duplicate. Primers were selected by using PrimerExpress program (Applied Biosystems). Only primer pairs with efficiency higher than 90% were retain for further experiments. The absence of sequences identical to the PCR target in the rest of the genome was verified by using BLAT.
PCR fragments were directly sequenced using Big Dye Terminator Sequencing kits (Applied Biosystems).

Whole-genome DNA Copy number determination.
The Affymetrix Genome-Wide Human SNP Array 6.0 was used to detect copy number alterations. Samples were processed at the Genomic Platform of the Institut Curie following the instructions provided by the manufacturer. Data were normalized, analyzed and visualized using Partek Genomic Suite version 6.6 (Partek, St Louis, MO). Single nucleotide polymorphisms with smoothing values lower and greater than 2 ± 0.28 were considered as gain and loss, respectively. Data S1 GBM11, localisation of the hsr insertion site in 17p11.
The BAC CTD-2277H24 covers the site of insertion of the hsr ( Figure 1D in publication). In order to have to precisely localize the insert, FISH were performed using fosmids overlapping the BAC. When a fosmid overlapped the site of insertion, hybridization spots could be observed both side of the hsr (Figure 1). We assumed that the percentage of spots centromeric or telomeric of the hsr was proportional to the lenght of the fosmid sequence in the corresponding side of the insertion site Figure 2A). For the series of fosmids, the percentage of centromeric spots was plotted as a fonction of their 3' or 5' ends positions. The extrapolation of the curves to a percentage of 0 (3' side) or 100 (5' side) determine the location of the insertion.
Based on these data, we defined a region of about 30 kb (17,714,000 -17,743,000) where the insertion site was expected ( Figure 2B). Chromosome walking was performed using a series of primers along this chomosome sequence and it was possible to find the junction between chromosomes 17 and the amplicon (see main  FISH of chromosome 17 paintings show that in the chromosomes 17 bearing the hsr a fragment of the long arm was fused at the end of the short arm ( Figure 1). In addition, three small marker chromosomes contained segments from the two arms of chromosome 17. The structure of these chromosomes was studied using SNP analysis. The copy number profile of chromosome 17 sequences deduced from SNP data shows a complex situation (Figure 2A). The allele ratio profile also indicated complex rearrangements ( Figure 2B). Eleven regions (A -K) were defined corresponding to the main copy number and allele composition changes ( Figure 2C). The boundaries between these regions were not localized at the nucleotide level, but FISH established their distribution in the rearranged chromosomes. We hybridized BACs (Table 1) localized in the segments A -K ( Figure 3). The results are summarized in Table 2. Figure 2D shows the chromosome structure.     Table 1 and in Figure 2C. Table 2 summarizes the data. A -K: chromosome segments defined in Figure 2C. Photos are identified by the letter(s) corresponding to the hybridized BAC. BAC were labelled in green and the hsr was visualized in red using the BAC RP11-1136L8 containing MYC in the A-K photos. Arrowheads indicate the BAC hybridization sites.

Supporting figure S5
Deleted region in the chromosome 17. Locaization of copy number variants, blue and red correspond to gain and loss in size relative to the reference. From the UCSC genome browser on human Feb 2009 (GRCh37/hg9) assembly.

Supporting figure S6
SW613-Tu1. Position of the junctions between segments from chromosome 17, 21 and 8 at the site of insertion of the hsr in relationship to the position of the Alu repeats. The table gives the positions of the Alu repeats and of the junctions.