We examined the involvement of sphingosine kinase-1 (SphK1), which governs the ceramide/sphingosine-1-phosphate balance, in susceptibility to imatinib of either sensitive or resistant chronic myeloid leukemia cells. Imatinib-sensitive LAMA84-s displayed marked SphK1 inhibition coupled with increased content of ceramide and decreased pro-survival sphingosine-1-phosphate. Conversely, no changes in the sphingolipid metabolism were observed in LAMA84-r treated with imatinib. Overcoming imatinib resistance in LAMA84-r with farnesyltransferase or MEK/ERK inhibitors as well as with cytosine arabinoside led to SphK1 inhibition. Overexpression of SphK1 in LAMA84-s cells impaired apoptosis and inhibited the effects of imatinib on caspase-3 activation, cytochrome c and Smac release from mitochondria through modulation of Bim, Bcl-xL and Mcl-1 expression. Pharmacological inhibition of SphK1 with F-12509a or its silencing by siRNA induced apoptosis of both imatinib-sensitive and -resistant cells, suggesting that SphK1 inhibition was critical for apoptosis signaling. We also show that imatinib-sensitive and -resistant primary cells from chronic myeloid leukemia patients can be successfully killed in vitro by the F-12509a inhibitor. These results uncover the involvement of SphK1 in regulating imatinib-induced apoptosis and establish that SphK1 is a downstream effector of the BcrAbl/Ras/ERK pathway inhibited by imatinib but upstream regulator of Bcl-2 family members.

Background:Head and neck squamous cell carcinoma (HNSCC) are resistant to standard treatments, partly due to cancer stem cells (CSCs) localised in hypoxic niches. Compared to X-rays, carbon ion irradiation relies on better ballistic properties, higher relative biological effectiveness and the absence of oxygen effect. Hypoxia-inducible factor-1α (HIF-1α) is involved in the resistance to photons, whereas its role in response to carbon ions remains unclear.Methods:Two HNSCC cell lines and their CSC sub-population were studied in response to photons or carbon ion irradiation, in normoxia or hypoxia, after inhibition or not of HIF-1α.Results:Under hypoxia, compared to non-CSCs, HIF-1α is expressed earlier in CSCs. A combined effect photons/hypoxia, less observed with carbon ions, results in a synergic and earlier HIF-1α expression in both subpopulations. The diffuse ROS production by photons is concomitant with HIF-1α expression and essential to its activation. There is no oxygen effect in response to carbon ions and the ROS localised in the track might be insufficient to stabilise HIF-1α. Finally, in hypoxia, cells were sensitised to both types of radiations after HIF-1α inhibition.Conclusions:Hypoxia-inducible factor-1α plays a main role in the response of CSCs and non-CSCs to carbon ion and photon irradiations, which makes the HIF-1α targeting an attractive therapeutic challenge.

Although conventional radiotherapy promotes the migration/invasion of cancer stem cells (CSCs) under normoxia, carbon ion (C-ion) irradiation actually decreases these processes. Unraveling the mechanisms of this discrepancy, particularly under the hypoxic conditions that pertain in niches where CSCs are preferentially localized, would provide a better understanding of the origins of metastases. Invasion/migration, proteins involved in epithelial-to-mesenchymal transition (EMT), and expression of MMP-2 and HIF-1α were quantified in the CSC subpopulations of two head-and-neck squamous cell carcinoma (HNSCC) cell lines irradiated with X-rays or C-ions. X-rays triggered HNSCC-CSC migration/invasion under normoxia, however this effect was significantly attenuated under hypoxia. C-ions induced fewer of these processes in both oxygenation conditions. The differential response to C-ions was associated with a lack of HIF-1α stabilization, MMP-2 expression, or activation of kinases of the main EMT signaling pathways. Furthermore,we demonstrated a major role of reactive oxygen species (ROS) in the triggering of invasion/migration in response to X-rays. Monte-Carlo simulations demonstrated that HO● radicals are quantitatively higher after C-ions than after X-rays, however they are very differently distributed within cells. We postulate that the uniform distribution of ROS after X-rays induces the mechanisms leading to invasion/migration, which ROS concentrated in C-ion tracks are unable to trigger.