Identification of recurrent somatic genomic duplications at the C-MYB locus in T-ALL. (A) Global representation of large-scale analysis of genomic copy number using a 4K array-CGH in MOLT4 cell line (upper panel) and TL63 primary T-ALL case (lower panel). DNA copy gain at the C-MYB locus at 6q23 is indicated, as well as the DNA copy loss at the CDKN2A/p16/ARF locus at 9p21 as an example. (B) Compared analysis of array hybridization using as normal DNA either a healthy subject DNA (control) or germ-line DNA (GL) of the same patient as DNA reference. This analysis allowed us to distinguish copy-number variations (CNVs) and somatic genomic imbalances. As an example, a CNV at 5p15 was resolved in the TL26 and TL29 cases (a gain was observed when the leukemic DNA was cohybridized with an unrelated control DNA, but it disappeared using paired leukemic and GL controls). A somatic loss of CDKN2A/p16/ARF was evidenced in both the TL26 and TL29 cases, because the imbalance persisted after cohybridization with paired control; similarly the somatic gain at the 6q23/C-MYB locus was confirmed in case TL29. For each BAC/PAC of the array, gains were represented in red; losses, in green; and balanced signals, in yellow. (C) Copy-number analysis using a very high-density oligonucleotide array (Agilent), focused on the C-MYB region, enabled the minimal region of genomic gain to be mapped to approximately 230 kb. (C, right panel) A 6-Mb–sized copy gain region was found in case TL01; (left panel) a short minimal genomic gain including the C-MYB gene was evidenced in case TL29 (array-CGH performed with paired leukemic and GL DNAs). The genomic region of gain of chromosome 6 is magnified, with the horizontal cursor (blue) pointing out the C-MYB gene (shown in red in the right panel). The so-called “moving average” ratio between leukemic and GL DNA appears as a blue line in case TL01, and a brown line in case TL29. Two copies (alleles) of the locus appear as a moving average close to 0, whereas a DNA copy gain of an allele shifts the line close to ratio + 0.5 along the corresponding genomic region. (D) Interphasic C-MYB FISH using the RP1–32B1 (green) and RP3–388E23 (red) probes in the MYB^dup cases showed no extra signal (except in the complex MOLT4 pseudotetraploid cell line, not shown), which suggests local duplication. (E) Molecular combing analysis using C-MYB locus probes RP11–55H4 (red) and RP11–166A21 (green) demonstrated a local duplication (bottom); the normal allele is shown on the top. See “Patients, materials, and methods; Cytogenetic and molecular analyses” for details about FISH image acquisition and manipulation.