Chronic lymphocytic leukaemia (CLL) is the most frequently diagnosed type of leukaemia in adults in many countries. Patients diagnosed with this condition vary considerably in both initial symptoms and disease prognosis. The average five-year survival for CLL is about 50% (slightly lower for men than for women) but this masks extreme variability. Genetically, the disease can be divided into two basic molecular subtypes according to the number of somatic mutations observed in the variable regions of the immunoglobulin genes. However, the detailed molecular mechanisms that lead to the development of either subtype of DLL are still largely unknown.

A large group of mainly Spanish researchers led by Carlos López-Otin from Universidad de Oviedo, Oviedo and Elias Campo from Universidad de Barcelona, Barcelona, has now elucidated a number of these molecular changes using whole-genome sequencing. They sequenced the complete genomes and the exomes, or protein-coding DNA regions, from four patients with CLL, two with no mutations in immunoglobulin genes (referred to as IGHV-unmutated) and two with mutations in these genes (IGHV-mutated). About 1,000 somatic mutations were detected in each tumour, which was lower than the numbers observed in some other tumour types but in agreement with previous estimates of mutation rates in leukaemia. The most commonly occurring mutation type was, as expected, the G>A/C>T transition, which usually occurred in a CpG context. The two IGHV-mutated cases also had a high rate of A>C/T>G mutations, and the researchers suggested that this might arise due to the involvement of the error-prone polymerase h, which is known to introduce somatic mutations into immunoglobulin genes. This single finding illustrates the genetic and molecular differences between the subclasses.

The researchers identified forty-six somatic mutations in these four cases that altered the protein-coding sequences of genes and that, therefore, might affect gene function. None of these had been previously linked to CLL and only one, in the NOTCH1 gene, had been found in any lymphoid malignancy. The sequences of these genes in samples from a further 363 CLL patients were then analysed in order to discover which were recurrently mutated in this disease. Mutations were found to occur in more than one case overall in four of these genes: NOTCH1, MYD88, XPO1 and KLHL6. The most commonly mutated gene was NOTCH1, with mutations found in approximately 12% of all CLL patients tested, more often in patients with IGHV-unmutated disease. The mutations observed in this gene caused a truncation of the NOTCH1 protein prior to its C-terminal domain; this form is more stable and active than the wild type and accumulates in leukaemia cells. Twenty-three of 46 genes assigned to the NOTCH1 signalling pathway were found to be differentially expressed in leukaemia cases with mutations in this gene. Four patients in the complete cohort, all with IGHV-umnutated disease, had mutations in the same codon of the XPO1 gene that encodes the protein exportin 1; two of these patients also carried NOTCH1 mutations. These mutations occurred within a highly conserved region in the gene, which further suggests that they will affect protein function.

Mutations in the gene MYD88 were more commonly observed in patients with the IGHV-mutated form of leukaemia, with one specific missense mutation in MYD88 being found in nine patients. This gene encodes a protein that is involved in interleukin-1 and Toll-like receptor signalling pathways during the immune response; further analysis of proteins in this pathway suggested that this is an activating mutation in a novel proto-oncogene. On average, patients with MYD88-mutated leukaemia presented at a younger age and with more advanced disease than patients in which this gene was wild type. Mutations in the gene KLHL6, which encodes the kelch-like protein 6 that is involved in the maturation of B cells, were also observed more commonly in patients with IGHV-unmutated CLL.

López-Otin, Campo and their co-workers conclude by suggesting that these genes are all likely to be involved in the molecular processes driving the development of CLL. Furthermore, they identify mutations in NOTCH1 and MYD88, in particular, as likely activating events in leukaemia development, and the protein products of these genes as potential therapeutic targets.

Reference

Puente, X.S., Pinyol M., Quesada, V. and 62 others, Whole-genome sequencing identifies recurrent mutations in chronic lymphocytic leukaemia Nature (2011), published online ahead of print 5 June 2011