<p>Recurrent chromosome breakpoints in tumour cells may point to cancer genes, but not many have been molecularly characterised. We have used multicolour-banding fluorescence in situ hybridisation (mbanding-FISH) on breast tumour cell lines to identify regions of chromosome break created by inversions, duplications, insertions and translocations on chromosomes 1, 5, 8, 12 and 17. We delineate a total of 136 regions of break, some of them occurring with high frequency. We further describe two examples of dual-colour FISH characterisation of breakpoints, which target the 1p36 and 5p11-12 regions. Both breaks involve genes whose function is unknown to date. The mbanding-FISH strategy constitutes an efficient first step in the search for potential cancer genes.</p>

A clonal origin of hematopoiesis was studied by investigation of X-chromosome inactivation patterns (XCIP) in isolated granulocyte, CD14(+) and CD3(+) subpopulations obtained from bone marrow and peripheral blood of 36 female patients with primary myelodysplastic syndrome (MDS). Clonality was assessed by PCR amplification of polymorphic short tandem repeats of the human androgen receptor (HUMARA) gene and by investigation of silent polymorphism of iduronate sulphatase (IDS) or p55 genes. On the basis of results in a control group of 20 healthy age related females, a ratio of at least 9:1 between the two alleles was considered a significant marker of monoclonal hematopoiesis. Ten of the 11 patients with advanced forms of MDS (RAEB, RAEB-T, CMML) had clonal granulocytes and CD14(+) cells in peripheral blood. In patients with early disease, only 2 out of 11 patients (18%) with RA or RARS, according to WHO classification, had clonal granulocytes and CD14(+) cells in peripheral blood and bone marrow and 2 other patients with 5q-syndrome exhibited extremely oligoclonal granulocyte subpopulation in bone marrow. In contrast, we found clonal granulocytes in 12 out of 14 patients (86%) with refractory cytopenia with multilineage dysplasia (RCMD) and 8 of them simultanously exhibited clonal CD14(+) cells. Estimated 3 years survival of patients with early disease and clonal cell subpopulations was 61% as compared with 88% in patients without clonal hematopoiesis. Karyotype abnormalities were detected in 11 of the 25 females with early disease. Clonal patterns were present in 7 out of 8 patients with abberations diagnosed by routine cytogenetics, nevertheless, FISH revealed 5q deletion in 3 patients without signs of clonality in XCIP assay. No correlation was found between the presence of clonal subpopulations and the degree of telomere shortening in early MDS. Despite some limitations, the measurement of XCIP remains a sensitive tool for diagnosis of the first transforming mutation in the clonal development of MDS especially when combined with FISH and when an age related group is used to establish an appropriate allele ratio to exclude constitutional or acquired skewing. The occurrence of clonal cell subpopulations in most of the RCMD patients in contrast to RA may reflect a proposed multistep pathogenesis of MDS with dysplastic changes limited to erythropoiesis in early step and with subsequent development of multilineage dysplasia. The results also support the usefulness of separation of RCMD from 'pure' RA; however, a more complex insight combining different molecular techniques performed in a large number of patients is needed for refined classification of MDS on the basis of new molecular prognostic factors and for indication of more effective targeted therapy.

Small supernumerary marker chromosomes (SMCs) in human can be defined as additional centric chromosome fragments smaller than chromosome 20. For most small or minute SMCs a correlation with clinical symptoms is lacking, mostly due to problems in visualizing their euchromatic content. Recently we described two new molecular cytogenetic approaches for the comprehensive characterization of small SMCs, excluding those few cases with neo-centromeres. Minute SMCs, consisting preferentially of alpha-satellite DNA, are characterizable in one step by the centromere-specific multicolor FISH (cenM-FISH) approach. For further characterization of minute SMCs and eventually present euchromatic content, the recently developed centromere-near-specific multicolor FISH (subcenM-FISH) technique can be applied. These two approaches are highly informative and easy to perform, as demonstrated in the present report on the example of a prenatal case with a minute SMC derived from chromosome 3 cytogenetically described as min(3)(:p12.1→q11.2:).

<p>Genomic fingerprints of mutagenic agents would have wide applications in the field of cancer biology, epidemiology and prevention. The differential spectra of chromosomal aberrations induced by different clastogens suggest that ratios of specific aberrations can be exploited as biomarkers of carcinogen exposure. We have tested this hypothesis using the novel technique of multicolor banding in situ hybridization (mBAND) in human peripheral blood lymphocytes exposed in vitro to X rays, neutrons, heavy ions, or the restriction endonuclease AluI. In the heavy-ion-irradiated cells, we further analyzed aberrations in chromosome 5 using multicolor FISH (mFISH). Contrary to the expectations of biophysical models, our results do not support the use of the ratios of inter-/intrachromosomal exchanges or intra-/interarm intrachanges as fingerprints of exposure to densely ionizing radiation. However, our data point to measurable differences in the ratio of complex/simple interchanges after exposure to different clastogens. These data should be considered in current biophysical models of radiation action in living cells.</p>

Precise breakpoint definition of chromosomal rearrangements using conventional banding techniques often fails, especially when more than two breakpoints are involved. The classic banding procedure results in a pattern of alternating light and dark bands. Hence, in banded chromosomes a specific chromosomal band is rather identified by the surrounding banding pattern than by its own specific morphology. In chromosomal rearrangements the original pattern is altered and therefore the unequivocal determination of breakpoints is not obvious. The multicolor banding technique (mBAND, see Chudoba et al., 1999) is able to identify breakpoints unambiguously, even in highly complex chromosomal aberrations. The mBAND technique is presented and illustrated in a case of intrachromosomal rearrangement with seven breakpoints all having occurred on one chromosome 16, emphasizing the unique analyzing power of mBAND as compared to conventional banding techniques.

We report a 13-year-old female patient with multiple congenital abnormalities (microcephaly, facial dysmorphism, anteverted dysplastic ears and postaxial hexadactyly), mental retardation, and adipose-gigantism. Ultrasonography revealed no signs of a heart defect or renal abnormalities. She showed no speech development and suffered from a behavioural disorder. CNS abnormalities were excluded by cerebral MRI. Initial cytogenetic studies by Giemsa banding revealed an aberrant karyotype involving three chromosomes, t(2;4;11). By high resolution banding and multicolour fluoresence in-situ hybridisation (M-FISH, MCB), chromosome 1 was also found to be involved in the complex chromosomal aberrations, confirming the karyotype 46,XX,t(2;11;4).ish t(1;4;2;11)(q43;q21.1;p12-p13.1;p14.1). To the best of our knowledge no patient has been previously described with such a complex translocation involving 4 chromosomes. This case demonstrates that conventional chromosome banding techniques such as Giemsa banding are not always sufficient to characterise complex chromosomal abnormalities. Only by the additional utilisation of molecular cytogenetic techniques could the complexity of the present chromosomal rearrangements and the origin of the involved chromosomal material be detected. Further molecular genetic studies will be performed to clarify the chromosomal breakpoints potentially responsible for the observed clinical symptoms. CONCLUSION: This report demonstrates that multicolour-fluorescence in-situ hybridisation studies should be performed in patients with congenital abnormalities and suspected aberrant karyotypes in addition to conventional Giemsa banding.

Using comparative genomic hybridization (CGH), we present a genome-wide screening of a mixed mesenchymal-epithelial hepatoblastoma, its recurrence and 2 novel hepatoblastoma cell lines raised from the ascites, 18 (HepU1) and 23 (HepU2) months after diagnosis of a hepatoblastoma in a 35-month-old boy. Both cell lines were also characterized by GTG-banding, multicolor-fluorescence in situ hybridization (M-FISH) and multicolor banding (M-Band). On the basis of CGH, we compared the cytogenetics of histologically different tumor areas of the parental tumor and its recurrence with the hepatoblastoma cell lines. We found different CGH profiles in the parental tumor rev ish enh(1q31-q32,8p,12,17,20,X), dim(4q34-q35,18q23)[cp] and its recurrence rev ish enh(8q24,17,Xq26-q28), dim(7q11.2-q21,13q34)[cp]. Although both epithelial cell lines were obtained at different times and the clonal ancestor of HepU2 had been exposed to a higher cumulative dose of chemotherapy, HepU1 and HepU2 have an identical karyotype: 48-56,XY,+Y,dup(2)(q32-q34),t(3;4)(q21;q34),+8,+12,+13, +17,+t(18;19)(q21;q?),+20[cp] and identical CGH profiles: rev ish enh(2q24-q33,8,12,13q,17,20), dim(4q34-q35,18q22-q23). In common with previously described hepatoblastoma cell lines, HepU1 and HepU2 demonstrate a gain of chromosome 20. The in situ aberrations most closely resembling that of HepU1 and HepU2 were found in areas of fetal-embryonal differentiation of the primary tumor. Interestingly, both cell lines mimic this histology in their three-dimensional growth pattern in vitro. HepU1 and HepU2 are thus cytogenetically and phenotypically highly characteristic of fetal-embryonal hepatoblastoma.

We report on a rare additional numerical chromosomal aberration in a child with Down syndrome due to free trisomy 21. The karyotype showed 48,XY,+21,+mar after GTG banding, with the marker present in 80% of cells. The supernumerary marker chromosome (SMC) was as small as approximately one-third of 18p, and with the recently developed centromere-specific multi-color fluorescence in situ hybridization (cenM-FISH) technique, it was shown that the SMC was a derivative chromosome 4. The SMC was not specifically stained by arm-specific probes for chromosome 4; thus, it has been described as der(4)(:p11 --> q11:). Microsatellite analysis resulted in a partial maternal uniparental isodisomy (UPD) for chromosome 4p15-16 and a maternal origin for two chromosomes 21. Until now only two similar cases have been described in the literature, but without clarifying the origin of the SMC and without looking for an additional UPD. This is the only reported case of a UPD 4p in a liveborn child.

All molecular alterations that lead to breast cancer are not precisely known. We are evaluating the frequency and consequences of reciprocal translocations in breast cancer. We surveyed 15 mammary cell lines by multicolor fluorescence in situ hybridization (M-FISH). We identified nine apparently reciprocal translocations. Using mBanding FISH and FISH with selected YAC clones, we identified the breakpoints for four of them, and cloned the t(3;20)(p14;p11) found in the BrCa-MZ-02 cell line. We found that the breakpoint targets the potential tumor-suppressor gene FHIT (fragile histidine triad) in the FRA3B region; it is accompanied by homozygous deletion of exon 5 of the gene and absence of functional FHIT and fusion transcripts, which leads to the loss of FHIT protein expression. Additional experiments using comparative genomic hybridization provided further information on the genomic context in which the t(3;20)(p14;p11) reciprocal translocation was found.

Rearrangements of 12p, resulting from deletions or translocations, are common findings in hematologic malignancies. In many cases, these rearrangements target the ETV6 gene (previously called TEL) located at 12p13. Various partner genes have been implicated in the formation of fusion genes with ETV6. These include PDGFRB, JAK2, NTRK3, ABL2, and ABL1, each of which encodes for proteins with tyrosine kinase activity. To date, ETV6/ABL1 transcripts have been detected in only four patients with a leukemic disorder. Here, we describe one adult with chronic myeloid leukemia and a child with T-cell acute lymphocytic leukemia with ETV6/ABL1. Molecular cytogenetic analysis confirmed that formation of an ETV6/ABL1 fusion in these patients required at least three chromosomal breaks and showed that each of these translocations is the result of a complex chromosomal rearrangement. Molecular analysis showed the presence of two fusion transcripts in both patients as the result of alternative splicing, questioning the suggested role of these transcripts in the lineage specificity. Clinical findings of these patients were compared to those of previously reported cases, and the possible clinical and biological similarities between ETV6/ABL1 and other fusion genes leading to increased tyrosine kinase activity are discussed.

<p>Characterisation of chromosome rearrangements using conventional banding techniques often fails to determine the localisation of breakpoints precisely. In order to improve the definition of chromosomal breakpoints, the high-resolution multicolour banding (MCB) technique was applied to identify human chromosome 5 breakpoints from 40 clinical cases previously assessed by conventional banding techniques. In 30 cases (75%), at least one breakpoint was redefined, indicating that MCB markedly improves chromosomal breakpoint localisation. The MCB pattern is highly reproducible and, in contrast to conventional banding pattern, is consistent in both short and elongated chromosomes. This might be of fundamental interest for the detection of chromosomal abnormalities, especially in tumour cells. Moreover, MCB even allows the detection of abnormalities that remain cryptic in GTG-banding analysis.</p>