The V617F mutation in the JAK2 tyrosine kinase, recently described in a majority of patients with myeloproliferative disorders (MPDs), confers growth factor independence in vitro and oncogenicity in mice. Therefore, targeted inhibition of mutant JAK2 kinase activity may be an effective strategy for treatment of MPD patients with this mutation. The ability to measure the activation status of JAK2 in patient samples will thus be of substantial value for monitoring therapeutic efficacy. We have developed quantitative flow cytometry-based assays for rapid and reproducible measurement of intracellular total and phosphorylated proteins of the canonical JAK/STAT pathway, as well as heat shock proteins (HSPs). In this study we examined the ability of these assays to detect altered levels of total and phosphorylated JAK/STAT signaling pathway components and HSP in a cell line (HEL) that is homozygous for the V617F JAK2 mutant. HEL cell cultures were incubated with 17AAG, a geldanamycin analog with clinical utility in a broad range of diseases. 17AAG exerts its inhibitory effect by binding to heat shock protein 90 (HSP90), preventing its chaperone association with client oncoproteins. AKT is among these client proteins and a component of the JAK/STAT pathway, representing a potential therapeutic target. 17AAG exposure reduced total AKT protein levels by 42%. 17AAG also inhibited mutant JAK2 activity by 66% and had a smaller effect (17%) on total JAK2 levels, suggesting that mutant JAK2 activation may rely on HSP90, either directly or through dependence on other client proteins. Exposure to 17AAG also reduced levels of P-STAT5 (50%) and, to a lesser extent, total STAT5 (27%). 17AAG-treated cells showed a 55% reduction in HSP90 levels and a 14% increase in HSP70 protein levels. JAK Inhibitor I (Calbiochem), a potent pan-JAK Inhibitor that blocks JAK1, JAK2, and JAK3 activity, caused reductions in P-JAK2 and P-STAT5 levels (29% and 26% decreases, respectively). However, the combining of JAK Inhibitor I with 17AAG did not result in an enhanced effect beyond what was observed with 17AAG treatment alone. Similar results were seen with AG490, a potent and selective JAK2 inhibitor. 17AAG caused a 40% decrease in viable cells after 18 hrs of treatment, compared with a 35% reduction for the pan-JAK inhibitor and a 20% decrease for AG490. Combining 17AAG with the pan-JAK inhibitor or AG490 caused only minor enhancement of these cytotoxic effects (46% and 41% reduction in cell viability, respectively). Our data support the potential utility of HSP90 inhibitors such as 17AAG in the development of small-molecule therapy for mutant JAK2 kinase-positive MPD. These results also show that flow cytometry-based assays for JAK/STAT signaling components and HSPs can be used to quantitatively monitor drug efficacy at the protein level in intact cells. These tests are likely to have broad clinical utility given the spectrum of diseases in which a pathogenic role for mutant JAK2 kinase is implicated.

Author notes

Corresponding author