Abstract
Introduction. Nucleophosmin (NPM1) gene mutations occur in 50-60% of adult AML with normal karyotype (Falini et al, NEJM 2005). About 50 NPM1 mutations have been so far identified, all clustering in exon-12 (Falini et al, Blood 2007) but few sporadic cases involving either exon-9 (one) (Mariano et al, Oncogene 2006) or exon-11 (two) (Albiero et al, Leukemia 2007). In spite of molecular heterogeneity, all mutations cause common changes at the C-terminus of NPM1 mutants, i.e. loss of tryptophans 288 and 290 (or 290 alone) and creation of a new nuclear export signal (NES) motif (Falini et al, Blood 2006a).As a consequence, all NPM1 mutants aberrantly accumulates in the cytoplasm of leukemic cells and can be detected by immunohistochemistry, which is fully predictive of NPM1 gene mutations (Falini et al, Blood 2006b).
Methods. From 2005 to 2015, 702 AML patients samples were analyzed at diagnosis by both immunohistochemistry (IHC) for NPM1 subcellular localization and western blot (WB) with anti-NPM1 mutant specific rabbit polyclonal antibodies antibodies, produced in our laboratory (Martelli et al, Leukemia 2008). Discordant cases were further analyzed by NPM1 gene Sanger sequencing. Newly discovered NPM1 mutated genes were subcloned in pEGFP-C1 vector and transiently expressed in NIH-3T3 adherent cells to study the NPM1 mutant subcellular localization by immunofluorescence microscopy. The NESbase version 1.0 program was used to identify putative NES within the new protein sequence, and their efficiency was evaluated by the pREV1.4-based NES efficiency assay, as previously described (Bolli et al, Cancer Res 2007).
Results. At IHC and WB analyses, concordance in diagnosis was obtained in 695/702 samples (291 NPM1-mutated and 404 NPM1-unmutated AML). In 7/702 (1%), IHC detected cytoplasmic NPM1 whilst WB with anti-NPM1 mutant antibodies was negative. Unfortunately, in 3 out of these cases, the original patient sample was not available for further analyses. In the other 4, exon-12 NPM1 gene sequence was wild-type (WT), in keeping with the negative WB results. One of these cases harbored the previously described exon-11 NPM1 mutation, in 1 case no mutation was detected (further studies are ongoing), and in 2 cases new mutations involving exon-6 were discovered. Strikingly, in the latter cases, WB analysis with different anti-NPM1 antibodies revealed a new band at different molecular weight (MW) than NPM1-WT. Indeed, in 1 case an in frame 21 nucleotides insertion at exon-6 lead to a 7 aa longer than WT protein, whilst in the other a 19 nucleotides insertion created a new stop codon leading to a truncated protein. In both cases, a new NES motif was created. Importantly, cell transfection experiments confirmed that the new NPM1 mutants localized at least partly in the cytoplasm, and the pREV1.4-based NES efficiency assay showed the new NES were active.
Conclusions. Here, we report on the identification and functional characterization of two novel NPM1 mutations in AML. Our observations further support the view that cytoplasmic NPM1 dislocation is a critical step in leukemogenesis, and that immunohistochemistry, that detects, through cytoplasmic dislocation on NPM, 'all types' of NPM1 mutations, might be used as first step for directing further molecular studies.
Disclosures
Di Raimondo: Janssen-Cilag: Honoraria.
Large-scale prediction and analysis of protein sub-mitochondrial localization with DeepMito