Inhibition of CML Stem Cell Growth By Targeting WNT Signaling Using a Porcupine Inhibitor

BCR-ABL tyrosine kinase inhibitors (TKIs) have revolutionized the treatment of chronic myeloid leukemia (CML). However, TKIs are unable to eradicate the leukemia stem cells (LSCs) from which CML arises, despite effective inhibition of BCR-ABL kinase activity in this population. There is evidence that extrinsic signals from the bone marrow (BM) microenvironment play an important role in the resistance of CML LSC to TKI treatment. Our previous studies have shown a role for Wnt signaling in protecting CML LSC from TKI treatment (Blood. 2013; 121(10):1824-38). Palmitoylation of Wnts by the Porcupine acyltransferase (PORCN) is required for their secretion and activity. LGK974, a potent PORCN inhibitor, has recently been shown to inhibit Wnt signaling and shows in vivo efficacy against several Wnt-dependent tumors (PNAS. 2013; 110(50):20224-9). In this study, we investigated whether inhibition of Wnt secretion using LGK974 could sensitize CML stem/progenitor cells to TKI treatment. CML and normal CD34+ cells and purified CD34+CD38- stem cells and CD34+CD38+ committed progenitors were treated with LGK974 with or without addition of the BCR-ABL TKI (nilotinib), and in the presence and absence of human BM mesenchymal stromal cells (MSC). CML CD34+ cells showed enhanced Wnt signaling following exposure to recombinant Wnt3a (200ng/ml) compared to normal CD34+ cells as evidenced by increased phospho-LRP6 (ser1490) and β-catenin protein levels. Analysis of all 10 frizzled receptors expression by qRT-PCR revealed a significant upregulation of FZD4 (p=0.026, n=10), and FZD5 (p=0.003, n=10), but downregulation of FZD8 (p=0.009, n=10), in CML compared to normal CD34+ cells. Co-culture with MSC was associated with increased Wnt signaling in CML CD34+ cells as detected by increased expression and nuclear translocation of β-catenin protein, and increased expression of Wnt target genes including c-Myc, Cyclin-D1, PPARδ, and Axin2. Treatment with LGK974 reduced Wnt secretion from MSC measured in a TCF/LEF reporter assay (LGK 0.5μM, 0.74 fold reduction compared to control, p=0.016, n=5, LGK 1μM, 0.62 fold reduction, p=0.007, n=5) and potently inhibited Wnt signaling in CML CD34+ cells both in the absence and presence of MSC. Treatment with LGK974 (1μM) in combination with nilotinib (1μM) resulted in significantly increased inhibition of CML CFC growth compared to nilotinib alone, both in the presence and absence of MSC (34+38- cells LGK alone 82.75±6.38% inhibition of control, Nil alone 43.12±15.32%, combination 23.17±6.48%, p<0.05, n=4; 34+38+ cells LGK alone 87.65±6.13% inhibition of control, Nil alone 26.99±4.54%, combination 16.04±0.48%, p<0.05, n=4) Importantly, LGK974 inhibited CML CFC growth to a significantly greater extent than the normal CFC growth ( normal 34+38- cells LGK alone 88.09±6.85% inhibition of control, Nil alone 89.79±10.09%, combination 70.95±9.39%, p=ns, n=4; 34+38+ cells LGK alone 86.42±8.41% inhibition of control, Nil alone 87.43±4.66%, combination 73.39±6.91%, p=ns, n=4). However, LGK974 treatment did not induce apoptosis in CML or normal progenitors. These results suggest that the effects of LGK974 on clonogenic growth could be related to the long-term inhibition of LSC self-renewal. Therefore, studies designed to evaluate the effect of in vivo administration of LGK974 in a transgenic BCR-ABL mouse model of CML are in progress. Finally, our preliminary studies indicate that in vivo treatment of control healthy mice with LGK974 (5mg/kg bid p.o.), nilotinib (50mg/kg qd p.o.), or the combination for 4 weeks (n=7 per cohort) did not result in clinical toxicity or alter the short-term and long-term hematopoietic stem cell populations. In conclusion, our results indicate that treatment with the PORCN inhibitor LGK974 effectively inhibits both autocrine and paracrine WNT signaling in CML stem cells and inhibits their clonogenic capacity in vitro, and support further evaluation of this approach to selectively target CML stem cells.