Leukemia-initiating cells reside within the bone marrow (BM) in specialized niches where they undergo complex interactions with their surrounding stromal cells. In order to identify genes being implicated in the interaction of acute myeloid leukemia (AML) cells and stromal cells, we performed co-cultures of primary AML cells with primary endothelial cells and osteoblasts. The gene expression of co-cultured AML blasts was compared to AML cells grown without adherent cells using microarray analysis. Amongst those genes being dysregulated upon co-culture was the actin binding protein plastin-3 (PLS3). Further RT-qPCR analysis revealed an endogenous PLS3 expression in about 50% of BM samples from AML patients (n=25). In contrast, expression of PLS3 was only detected in 2 of 12 analyzed AML cell lines with Kasumi-1 showing strong and THP-1 showing only weak expression.

Therefore, functional analysis of PLS3 in AML was studied using shRNA knockdown and overexpression of PLS3 in Kasumi-1 cells. We could show that PLS3 has an impact on the colony formation capacity of AML cells in vitro as the knockdown resulted in significantly reduced colony numbers while increased colony growth was observed in the Kasumi-1 cells overexpressing PLS3 (p<0.001 and p<0.001, respectively). To investigate the role of PLS3 in vivo, NSG mice were transplanted with the PLS3 knockdown Kasumi-1 cells. Compared to mice transplanted with Kasumi-1 cells transduced with a vector carrying a scrambled shRNA, the PLS3 knockdown mice survived significantly longer (median survival time 64 vs. 110 days, respectively; p<0.001; n=9 mice per group).

Finally, we investigated whether the expression of PLS3 was associated with AML patients' outcome using published microarray-based gene expression data (Verhaak et al, Haematologica 2009;94). Clinical data of 290 AML patients were available. Based on the mean gene expression value, the patient cohort was divided into high vs low PLS3 expressors. The overall survival was analyzed in a multivariate Cox proportional hazards model including PLS3 gene expression and the baseline parameters age, karyotype and FLT3 mutational status. After a stepwise removal of insignificant terms, the patient's age and a high PLS3 expression remained as independent prognostic survival markers (for PLS3: HR 1.58 (CI 1.05 - 2.37) and for age: HR 1.01 (CI 1.00 - 1.03)).

In conclusion, our results identify the actin binding protein PLS3 as potential novel therapeutic target in AML.

Disclosures

Stamm:Astellas: Other: Travel, Accommodation, Expenses. Heuser:BerGenBio: Research Funding; Tetralogic: Research Funding; Novartis: Consultancy, Research Funding; Celgene: Honoraria; Bayer Pharma AG: Research Funding; Pfizer: Research Funding; Karyopharm Therapeutics Inc: Research Funding. Fiedler:Kolltan: Research Funding; Ariad/Incyte: Consultancy; Novartis: Consultancy; Gilead: Other: Travel; Teva: Other: Travel; GSO: Other: Travel; Pfizer: Research Funding; Amgen: Consultancy, Other: Travel, Patents & Royalties, Research Funding. Wellbrock:Astellas: Other: Travel, Accommodation, Expenses.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution