The 5q-syndrome.

Little is known about the relationship between the 5q- syndrome and the other subtypes of MDS, primary ANLL, or therapy-related MDS and ANLL with 5q- . The natural history of these disorders is very different to the 5q- syndrome, suggesting the presence of different underlying abnormalities. In particular, the good prognosis of the 5q- syndrome and rare transformation to ANLL contrasts with the poor prognosis and clinically aggressive course of RAEBt and therapy-related MDS and ANLL. It has previously been postulated by others that the disease genes causing the 5q- syndrome and therapy-related MDS and ANLL may be distinct. The recent molecular delineation of the different critical regions of the 5q- chromosome in malignant myeloid disorders supports this view. However, it remains possible that different mutations of a single gene localized to 5q31 or additional mutations elsewhere in the genome are responsible for some of the different phenotypic manifestations of the 5q- in myeloid disorders. Whether there are different disease genes implicated in the development of the 5q- syndrome and the other malignant myeloid disorders with a 5q- will only be finally resolved by the identification and cloning of these genes. In the absence of a fortuitous discovery it now seems probable that the identification of the 5q- syndrome gene will be achieved by an approach analogous to that recently used in the cloning of the APC gene in colorectal cancer. In these studies, a yeast artificial chromosome (YAC) contig was constructed encompassing the entire 5- Mb critical region and novel coding sequences mapping to the deleted region identified by screening the YAC clones against cDNA libraries. These candidate genes were then examined for rearrangements and/or mutations in patient material. The identification of small nested deletions in two patients with familial APC considerably reduced the complexity of the task, and the APC gene was identified as a result of its frequent mutation. Similarly, any candidate genes mapping to the critical region of the 5q- syndrome will need to be closely examined for rearrangements or mutations on the normal chromosome 5 homolog. The number of genes contained within the approximately 10-Mb region encompassing the proposed critical regions is unknown; some regions are relatively rich in genes and others are relatively depleted. A recent study that gives some insight into this question examined the number of genes found in one megabase of the HLA class I region. cDNA selection was applied to three overlapping YACs spanning 1 Mb of this region. In addition to the recognized class I genes, 20 additional anonymous cDNA clones were identified. Similarly, YAC coverage of 880 kb of the Huntington disease region (2.2 mb) has allowed for the isolation of 13 cDNA clones. Therefore, based on these analyses of gene-rich regions, the approximately 10 Mb encompassing the putative critical regions in 5q31 may possess up to 200 genes. It is probable that most of the genes in this region will be identified over the next 5 years or so, given the progress of the Human Genome Mapping project. Such comprehensive information on chromosome 5 should enable the resolution of the genetic basis of the 5q- syndrome