Isoform‐selective phosphoinositide 3‐kinase inhibitors induce apoptosis in chronic lymphocytic leukaemia cells

Phosphoinositide 3-kinase (PI3K) has been reported to be constitutively active in chronic lymphocytic leukaemia (CLL), and to contribute to increased cell survival and resistance to apoptosis (Barragan et al, 2002; Pleyer et al, 2009). PI3Ks generate phosphoinositide lipids in response to extracellular stimuli, regulating survival, proliferation, differentiation and migration (Manning & Cantley, 2007). PI3K class I are heterodimers that consist of a catalytic subunit and a regulatory subunit, and are further divided into class IA (p110α, β and δ), activated downstream tyrosine kinase receptors and class IB (p110γ), activated downstream G protein-coupled receptors (Engelman et al, 2006). p110α and β are widely distributed whereas p110δ and γ are mainly, but not exclusively, expressed in leucocytes. LY294002 and wortmannin have been largely used as PI3K inhibitors. However, neither of them can discriminate between the different isoforms of PI3K. Recently, isoform-selective PI3K inhibitors have been described (Jackson et al, 2005; Marone et al, 2008; Draghetti et al, 2009). Here, we have examined the effect of isoform-selective PI3K inhibitors (Table I) in the viability of CLL cells and lymphocytes from healthy donors. Chronic lymphocytic leukaemia was diagnosed according to standard clinical and laboratory criteria. Blood samples were obtained from the Hospital de Bellvitge, L'Hospitalet de Llobregat, Spain. Mononuclear cells from peripheral blood samples were isolated and cultured as previously described (Barragan et al, 2002). First, the ability of isoform-selective PI3K inhibitors to induce apoptosis in CLL cells was examined. Apoptosis was determined by annexin V-fluorescein isothiocyanate (Bender MedSystems, Vienna, Austria) and propidium iodide (PI; Bender MedSystems) double staining, as previously described (Barragan et al, 2002), and cell viability was measured as the percentage of annexin V and PI double-negative cells. Samples were acquired and data was analysed by using the FACSCalibur and CellQuest software (Becton Dickinson, Mountain View, CA, USA). Thus, CLL cells from 5 different patients were incubated with a range of concentrations (5–50 μmol/l) of isoform-selective PI3K inhibitors for 24 and 48 h and the cell viability was measured. PI3K inhibitors induced apoptosis in a dose- and time-dependent manner (Fig 1A, n = 5), except MSC1822169B (p110γ inhibitor), which did not induce apoptosis in CLL cells. The inhibitors that showed a stronger effect on cell viability were MSC1902994A (p110α inhibitor), MSC1829899A (p110δ inhibitor) and MSC1831419A (p110β and δ inhibitor). LY294002 was used as a control for broad PI3K inhibition. Incubation of CLL cells with 20 μmol/l LY294002 for 24 and 48 h reduced cell viability to 58 ± 6% and 50 ± 10%, respectively (n = 5). Cytotoxic effect of isoform-specific PI3K inhibitors on CLL cells and on cells from healthy donors. (A) Cells from CLL patients were incubated for 24 and 48 h with or without a range of doses (5, 10, 20, 50 μmol/l) of MSC1902994A, MSC1829899A, MSC1822169B, MSC1831419A (n = 5). (B) Cells from CLL patients were incubated for 48 h with 10 μmol/l MSC1902994A, 10 μmol/l MSC1829899A, 10 μmol/l MSC1822169B, 10 μmol/l MSC1831419A or 20 μmol/l LY294002 (n = 22, five samples were resistant and 17 samples were sensitive). (C) Cells from CLL patients were incubated for 48 h with 10 μmol/l MSC1902994A, 10 μmol/l MSC1829899A, 10 μmol/l MSC1831419A, or with combinations of two or three inhibitors (n = 8). (D) Cells from CLL patients (n = 9) and normal cells from healthy donors (n = 4) were incubated for 48 h with 20 μmol/l LY294002 or 10 μmol/l MSC1902994A, 10 μmol/l MSC1829899A or 10 μmol/l MSC1831419A. Viability was measured by analysis of phosphatidylserine exposure and propidium iodide uptake (A, B, C) or as nonapoptotic CD3+/CD19− T cells or CD3−/CD19+ B cells (D), and is expressed as the percentage of the viability of untreated cells. Data are shown as the mean value ± SEM. (B, C) *P < 0·001 treated versus untreated cells. (D) **P < 0·005, *P < 0·01 treated versus untreated cells, †P < 0·005, ‡P < 0·01 B cells versus T cells. The isoform-selective PI3K inhibitors were then tested in samples from 22 CLL patients. Surprisingly, CLL cells from five of the 22 patients showed resistance to PI3K inhibitor-induced apoptosis when incubated with 10 μmol/l isoform-selective PI3K inhibitors. The samples were grouped into two classes, designated as sensitive and resistant, according to whether the apoptosis induced by all the isoform-selective PI3K inhibitors at 48 h was higher or lower than 20% when compared to the controls. The patients resistant to all PI3K inhibitors were also resistant to LY294002-induced apoptosis (Fig 1B). Cells from the sensitive patients were examined to determine whether the combination of two or more PI3K inhibitors could increase the apoptosis induced by a single PI3K inhibitor treatment. CLL cells were incubated with one, two or three inhibitors at the same time, and cell viability was measured at 48 h. Isoform-selective PI3K inhibitors showed additive effects for inducing apoptosis of CLL cells, but no combination induced synergistic effects on cell viability (Fig 1C). The sensitivity of normal B and T cells to apoptosis induced by isoform-selective PI3K inhibitors was analysed by co-incubating cells with anti-CD3-allophycocyanin and anti-CD19-phycoerythrin antibodies (Becton Dickinson), and the percentage of non-apoptotic cells was determined for CD19+/CD3− cells (B cells) and CD19−/CD3+ (T cells). The number of apoptotic B and T cells was measured in CLL samples and blood samples from healthy donors exposed to PI3K inhibitors for 48 h. Incubation with 10 μmol/l isoform-selective PI3K inhibitors reduced the percentage of viable CLL cells (n = 9) (Fig 1D). In contrast, T cells from CLL samples were resistant to the isoform-selective PI3K inhibitors (n = 9). Incubation with 10 μmol/l isoform-selective PI3K inhibitors also reduced the percentage of viable normal B cells derived from healthy donors, whereas T cells derived from healthy donors were resistant to isoform-selective PI3K inhibitor-induced apoptosis (n = 4) (Fig 1D). These results indicate that normal and leukaemic B cells are more sensitive than T cells to isoform-selective PI3K inhibitor-induced apoptosis. Here, we have analysed the contribution of PI3K isoforms to CLL cell survival, using a set of recently developed isoform selective class I PI3K inhibitors. Our results show that inhibition of p110α, β and δ induced apoptosis whereas inhibition of p110γ did not affect the viability of CLL cells, suggesting that p110α, β and δ are contributing to maintain basal CLL cell viability. Similar results using two other p110α inhibitors (PIK90 and PI103) and the same inhibitors for p110β and p110δ, have been recently reported (Niedermeier et al, 2009). It has been described that p110δ contributes to 60% of total PI3K activity in normal B cells whereas p110α and β only represent the 40% of the remaining activity (Bilancio et al, 2006). p110δ is the isoform mostly involved in signalling through the interleukin 4 and B-cell receptors in normal B cells (Bilancio et al, 2006). However our results suggest that the most important PI3K isoform in CLL cell survival is p110α. Interestingly, some patients were resistant to the apoptosis induced by PI3K inhibitors, suggesting that, in these patients, pathways other than PI3K could be activated to maintain cell viability. Our results indicate that B cells are more sensitive than T cells to the specific PI3K-isoform inhibition. Chemotherapeutic drugs induce apoptosis equally in both B and T cells, leading to immunosuppression (Hamblin & Hamblin, 2008). Thus, the differential effect of PI3K inhibitors on B and T cells is of interest. In conclusion, we propose that isoform-selective PI3K inhibitors may provide a novel therapeutic option for CLL patients. This study was supported by grants from the Ministerio de Ciencia e Innovación and FEDER (SAF2007-60964 and ISCIII-RTICC RD06/0020), and the AGAUR-Generalitat de Catalunya (2005SGR-00549). MdeF is a recipient of a fellowship from the AGAUR-Generalitat de Catalunya, AMC, DMGG, APP and CRP are recipients of research fellowships from the Ministerio de Ciencia e Innovación, and DIS is recipient of fellowship from the José Carreras International Leukemia Foundation (FIJC-07/ESP-FCAJAMADRID). We also thank the Unitat de Biologia and the Unitat de Genòmica from the Serveis Cientificotècnics at the Universitat de Barcelona for their technical support. Thomas Rückle and Montserrat Camps are employed by Merck Serono SA.