Antileukemic Activity of the HSP70 Inhibitor Pifithrin-μ

Heat shock protein (HSP) 70 is aberrantly expressed in acute leukemias and other hematologic and solid malignancies, promoting tumor cell survival and therapy resistance. Recently, the small molecule pifithrin-μ (2-phenylethynesulfonamide) has been identified as a direct inhibitor of inducible HSP70, showing antiproliferative activity in different cell lines of solid tumors. Here, we analysed the in vitro antileukemic effect of pifithrin-μ in acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL) cell lines, as well as in primary AML blasts. In addition, incubations of pifithrin-μ with cytarabine, the histone deacetylase inhibitor SAHA, the HSP90 inhibitor 17-AAG, and the multikinase inhibitor sorafenib were performed to evaluate the potential use of combination therapies with pifithrin-μ in acute leukemias.

Leukemic cell lines KG-1a (AML), K562 (CML in blast crisis), K562r (cytarabine-resistant K562), NALM-6 (B-lineage ALL), TOM-1 (B-lineage ALL, BCR-ABL pos.), Jurkat (T-lineage ALL), BE-13 (T-lineage ALL) and 9 bone marrow cell samples from newly diagnosed or relapsed AML patients were exposed to pifithrin-μ. Cell viability of all cell lines listed above was quantified by WST-1 assay. Subsequent functional analyses were performed on KG-1a and NALM-6 cells. Apoptosis was determined by annexin-V/7-AAD staining and subsequent flow cytometric analysis. Activated caspase-3 was detected by flow cytometry. Levels of the cell signaling kinase Akt were measured by intracellular staining and FACS analysis. Coincubations of pifithrin-μ with cytarabine, SAHA, 17-AAG or sorafenib were performed in KG-1a, NALM-6 and TOM-1, using WST-1 assays to analyse cytotoxic effects of combination therapies.

Pifithrin-μ at low micromolar concentrantions significantly inhibited viability of all acute leukemia cell lines tested, with IC50 values ranging from 2.5 to 12.7 μM independent of the differentiation lineage. Importantly, viability of both cytarabine-sensitive and -resistant K562 cells was effectively inhibited by pifithrin-μ. The median IC50 of primary AML blasts was 8.8 μM, ranging from 5.7 to 11.8 μM with no obvious differences regarding patients' clinical or genetic characteristics.

Apoptosis was induced in a time- and dose-dependent fashion with a rate of specific apoptosis of 46% at 4 μM pifithrin-μ for NALM-6 and 36% at 40 μM pifithrin-μ for KG1a. In NALM-6, treatment with 3 μM pifithrin-μ for 24 hours resulted in a significant increase in the cleaved, active form of caspase-3, whereas in KG1a no increase in active caspase-3 was detected. Intracellular concentrations of Akt were markedly reduced after 12 hours incubation of NALM-6 with pifithrin-μ.

In NALM-6, KG-1a, and TOM-1 combination treatment of pifithrin-μ at concentrations below the IC50 with either SAHA, 17-AAG or sorafenib resulted in a significant decrease of cell viability compared to corresponding monotherapy. Thus in NALM-6 combination of 2 μM pifithrin-μ with 0.6 μM SAHA inhibited viability by 73%, compared to 22% and 0% inhibition for either drug alone (p<0.05). Combination of 2 μM pifithrin-μ with 2 μM 17-AAG led to 58% inhibition, in contrast the monotherapy inhibited cell viability only in 31% for either drug alone. In NALM-6 and TOM-1, the combination of pifithrin-μ with cytarabine decreased viability significantly (47% and 55%, respectively), whereas the single agents were less effective (22% for 2 μM pifithrin-μ, 24% for 9 nM cytarabine in NALM-6; 26% for 3.5 μM pifithrin-μ and 41% for 40 nM cytarabine in TOM-1).

This is, to our knowledge, the first report of the antileukemic effects of the HSP70 inhibitor pifithrin-μ. The inhibitor is highly active against all AML and ALL cell lines tested, including cytarabine resistant cell lines as well as primary leukemic cells. Effectivity of pifithrin-μ could even be increased in combination treatment with other antileukemic agents. Targeting HSP70 might be a promising new therapeutic approach for the treatment of acute leukemias to overcome drug resistance. Thus, our data might build a framework for future clinical trials.

No relevant conflicts of interest to declare.