Screening for cryptic rearrangements by FISH mapping

Possible candidate genes in the 20q CDR have been identified and prioritised according to their expression in CD34-positive cells.³ For this study, PACS were selected to contain sequences from one or more of the potential candidate genes. The selection was therefore biased to gene-rich areas in the chromosome light bands on 20q. In all, 12 PACS, which spanned the MPD CDR on 20q11.2/q13.1, were hybridised to BM chromosomes preparations in pools of three probes labelled in three colours (Figure 1a). Chromosome preparations from the cell line MHB14 with known del(20)(q11.2q13.1) were used as positive control. In all, 10–31 cells were analysed for each hybridisation experiment. Two hybridisation signals were obtained with each of the PAC probes on all 14 patient samples analysed (Figure 1a and b). In each metaphase cell, six hybridisation signals were seen indicating no loss. In all MH1B14 cells, one hybridisation signal was obtained for each PAC probe from the CDR – a result consistent with the presence of a deletion in one of the chromosome 20 homologues (results not shown).

Figure 1.

Molecular cytogenetic analysis of PV. In all, 14 BM samples from patients with PV were studied at the time of clinical presentation. The diagnosis was determined by the criteria proposed by Pearson and Messinezy.⁵ (-) (a) Chromosome 20 ideogram with 12 PACs from the 20q12/q13.1 region used for this study. The PACs were applied in four sets of three probes labelled with three fluorochromes. PACs were obtained from the PRC Human PAC library, The Wellcome Trust Sanger Institute (Hinxton, Cambs) and mapping data is available in Bench et al.³ DNA was extracted using a Qiagen mini plasmid kit (Qiagen, West Sussex, UK). The PAC inserts were 80–180 kb. Probes were labelled by nick translation with SpectrumOrange d-UTPs, SpectrumGreen d-UTPs (Vysis Inc., USA) or biotin d-UTPs (Boerhinger Mannheim, UK). The biotin labelled probes were detected using an avidin Cy5 antibody (Amersham, UK). (b) A representative metaphase cell from patient 13 with a set of three PACs. (c) A mitotic figure at a 300 banding stage from a PV patient hybridised with SpectraVysion 24 colour paint (Vysis Inc., USA). (d) A 800 banding stage metaphase cell from a PV patient as above. (e) A mitotic figure from patient 13 WCP signals (SpectrumGreen, Vysis Inc., USA) and a 20q telomere-specific probe (SpectrumOrange, Vysis Inc., USA). (f) Band-specific paints (Research Genetics, Inc., USA) were labelled and detected as described by Reid et al:⁹ 20q11 probe visualised by Cy3 (in red) and 20q13 by biotin–avidin/Cy5 (in blue). Their simultaneous application in BM preparations of nine PV patients failed to detect intrachromosomal aberrations, such as inv(20)(p11q13). (g) Comparative karyogram of CGH FR profiles obtained for seven PV patient DNA samples (red). CGH was conducted as before.⁸ DNA from a PV patient with a trisomy 8 was used as a control sample (green). A green arrow indicates the trisomy 8 FR profile. Normal vs normal control FR profiles are in black. Threshold values are set at 0.75 and 1.25. Black arrow indicates over-representation of chromosome X in patients with a sex-mismatched reference DNA. FISH images were captured using an Axioplan 2 microscope (Zeiss), CCD camera (Orca, Hamamatsu, Japan), narrow bandpass filters (Chroma, USA) and SmartCapture 2 software (Digital Scientific Ltd, Cambs.). CGH and M-FISH images were analysed using Quips software (Applied Imaging Inc., USA). CGH FR profiles for each case were collated and compared using the Formatter program (Digital Scientific Ltd, Cambs).

Full figure and legend (230K)

M-FISH and bands-specific painting analysis

Colour karyotyping was conducted on BM chromosome suspensions from three PV patients. In each case, 30 cells were captured and analysed to increase the chances of identifying an aberrant clone present at a low level. No cryptic structural chromosome rearrangements were detected in any of the metaphase cells even when high-resolution mitotic figures were analysed (Figure 1c and d). Simultaneous hybridisation of a whole chromosome 20 paint (WCP) and a 20q telomere probe established in all 14 cases that both chromosome 20 homologues were not involved in cryptic translocations (Figure 1e). BSP for 20q11 and q13 regions were applied together to search for intrachromosomal rearrangements, which could have been missed by 24 colour karyotyping and WCP. Unequivocal interpretation of the BSP signals was only possible of mitotic figures with longer chromosomes (400 banding stage or above) obtained in 9/14 patients. Neither inv(20)(p11q13) as described in AML/MDS cases (6) nor any other aberrations were found (Figure 1f).

Search for genetic imbalances by CGH

We have previously shown that CGH can detect imbalances of chromosome regions in the range of 10 Mbp, including 'small' deletions of 20q (7). The fluorescence ratio (FR) profiles for each chromosome in every sample were within the threshold values (1.25 and 0.75) showing balanced diploid genomes (Figure 1g). Only in two instances imbalances were identified: (a) for chromosome X, when test and reference DNAs were of opposite sexes and, (b) for chromosome 8, in the control PV patient BM sample that had trisomy of chromosome 8 by G banding in all dividing cells. The FR profiles also touched or crossed the thresholds at 1p36, 9q34, and 19p/q. FISH analysis using probes specific for a 1p36 locus (p58 gene, G-Biogene, UK), 9q34/qter subband region and chromosome 19 centromere (G-Biogene, UK) did not reveal any aberrations of these regions (results not shown).

In conclusion

The FISH screening of BM chromosome preparations from PV patients at presentation with apparently normal karyotype failed to detect cryptic aberrations within the 20q11.2/q13.1 region. Nevertheless, very small deletions could possibly have been missed by our approach because only approximately half of the 2.7 Mb CDR is represented by our PAC selection. This is unlikely as the remaining 50% of the region is gene poor (containing one gene not expressed in CD34-positive cells). It is also possible that our screening may have failed to detect an aberrant clone present at a low level. However, this also seems unlikely, because the analysis of 30 cells provides a 90% chance (95% confidence interval) of detecting an abnormal cell clone.

Similarly, a genome wide screen in these samples by colour karyotyping showed no cryptic intrachromosomal rearrangements. Since the application of the same M-FISH technique has identified a single chromosome band (on average being 10 Mbp) aberrations in CML cells,⁸ it was concluded that such occult abnormalities are not present in studied samples. Nevertheless, the presence of inversions or very small telomeric translocations cannot be excluded. Screening with FISH probes specific for 20q telomeric region and with 20q BSP failed to detect any aberrations in all PV cases.

CGH was also used to search for aberrant nondiving clonal cell population(s) in the BM samples. No imbalances were detected by CGH. Again, there are limitations associated with CGH. These include firstly, 'small' deletions, that are detectable only if the size of the affected regions are 5 Mbp or more, while extra copies of segments sized less than 3–5 Mb can be identified with ease and secondly, genomic changes can only be reliably detected if present in at least 40-60% of the sample.⁸ New technologies such as genomic microarrays will enable more detailed and speedier genome screens in the future.

The genetic abnormalities in PV remain elusive; however, this study eliminates the possibility of a consistent small 20q deletion, a recurrent cryptic chromosome rearrangement or a genome imbalance, detectable by molecular cytogenetic techniques, as the cause of PV.

References

1. Bench AJ, Nacheva EP, Hood TL, Holden JL, French L, Swanton S et al. Chromosome 20 deletions in myeloid malignancies: reduction of the common deleted region, generation of a PAC/BAC contig and identification of candidate genes. UK Cancer Cytogenetics Group (UKCCG). Oncogene 2000; 19: 3902–3913. | Article | PubMed | ISI | ChemPort |
2. Anderlid BM, Schoumans J, Anneren G, Sahlen S, Kyllerman M, Vujic M et al. Subtelomeric rearrangements detected in patients with idiopathic mental retardation. Am J Med Genet 2002; 107: 275–284. | Article | PubMed |
3. Mathew S, Rao PH, Dalton J, Downing JR, Raimondi SC. Multicolour spectral karyotyping identifies novel translocations in childhood acute lymphoblastic leukemia. Leukemia 2001; 15: 468–472. | Article | PubMed | ISI | ChemPort |
4. Asimakopoulos FA, Gilbert JG, Aldred MA, Pearson TC, Green AR. Interstitial deletion constitutes the major mechanism for loss of heterozygosity on chromosome 20q in polycythemia vera. Blood 1996; 88: 2690–2698. | PubMed |
5. Pearson TC, Messinezy M. The diagnostic criteria of polycythaemia rubra vera. Leukaemia Lymphoma 1996; 22(Suppl 1): 87–93.
6. Dewald GW, Schad CR, Lilla VC, Jalal SM. Frequency and photographs of HGM11 chromosome anomalies in bone marrow samples from 3,996 patients with malignant hematologic neoplasms. Cancer Genet Cytogenet 1993; 68: 60–69. | PubMed |
7. Nacheva EP, Grace CD, Bittner M, Ledbetter DH, Jenkins RB, Green AR. Comparative genomic hybridisation: a comparison with molecular and cytogenetic analysis. Cancer Genet Cytogenet 1997; 97: 1–13.
8. Gribble SM, Sinclair PB, Grace C, Green AR, Nacheva EP. Comparative analysis of G-banding, chromosome painting, locus-specific fluorescence in situ hybridisation and comparative genomic hybridisation in chronic myeloid leukaemia blast crisis. Cancer Genet Cytogenet 1999; 111: 7–17.  | PubMed |
9. Reid A, Gribble SM, Andrews KM, Green AR, Nacheva EP. Chromosome band specific FISH probes allow improved detection of terminal translocations in leukaemic metaphases. Leukemia 2001; 15: 860–861. | Article |

Acknowledgements

This work was funded by Kay Kendall Leukemia Fund and Leukemia Research Fund, UK.