Abstract
Abstract 4185
RAF kinase inhibitor protein (RKIP) has been described as a negative regulator of the RAS-mitogen activated protein kinase (MAPK) signaling cascade and as a metastasis suppressor gene in solid tumors. As constitutive activation of this pathway occurs frequently in acute myeloid leukemia (AML), we investigated the role of RKIP in the pathogenesis of this disorder. Analysis of RKIP expression by Western blot revealed complete or partial loss in 21/96 (22%) AML patient samples and 5/19 (26%) AML cell lines, but in none of ten purified normal CD34+ hematopoietic stem and precursor cell specimens. To evaluate the biological consequences of RKIP loss in AML, we examined the effects of introducing an RKIP transgene in 32D murine hematopoietic cells and THP-1 AML cells, the latter characterized by low RKIP expression. Overexpression of RKIP induced a significant decrease in the number of viable cells in both cell lines which was due to a decrease in proliferation as measured by cell cycle analysis after bromodeoxyuridine (BrdU) labeling. However, no differences were shown in the percentage of apoptotic cells - subG1 peak – in these assays, a finding that was confirmed by analysis of PARP- and caspase-3 cleavage. Importantly, colony formation of leukemic THP-1 cells in soft agar was significantly decreased following RKIP reconstitution going along with a tumor suppressor function of RKIP in hematopoietic cells. In a next step, we analyzed RKIP expression in a previously described transcriptomic data set of 285 AML patients (Valk et al. NEJM 2004) and sought to identify cooperating alterations. 28/285 (10%) patients harbored mutations in either NRAS or KRAS, 84/285 (29%) in NPM1, 17/285 (6%) in CEBPA, 31/285 (11%) in the FLT3 tyrosine kinase domain and 68/285 (24%) a FLT3 internal tandem duplication. Out of these, only mutations in RAS were significantly associated with decreased expression of RKIP (P = 0.00005) suggesting co-existence of mutant RAS and RKIP loss in AML. To test for a functional synergism, we employed NIH3T3 cells, a system commonly used to study RAS biology. The oncogenic potential of mutant RAS – as assessed by its ability to transform cells and induce focus formation – could be significantly reduced by simultaneous overexpression of RKIP indicating cooperation between RAS mutations and RKIP loss in malignant transformation. We next analyzed if RKIP loss was associated with specific AML features and observed a highly significant correlation with monocytic phenotypes (FAB M4 and M5) in two independent patient cohorts (cohort one with 96 patients, P = 2.13 × 10-7; cohort two with 285 patients, P = 1.6 × 10-8). To test for a causative role of RKIP in monocytic differentiation, we analyzed immature HL-60 AML cells, a cell line which can be converted into a monocytic phenotype by incubation with 1,25-dihydroxyvitamin D3 (1,25 D3). Whereas siRNA knockdown of RKIP alone failed to induce the monocytic phenotype, 1,25 D3 driven effects could be significantly increased by siRNA knockdown of RKIP and inhibited by RKIP overexpression indicating that RKIP loss is involved in the development of a monocytic AML phenotype. In conclusion, our data demonstrate that loss of RKIP is a frequent event in AML, particularly in subgroups with a monocytic phenotype. RKIP acts as a tumor suppressor in hematopoietic cells and cooperates with mutant RAS in malignant transformation.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.