Burkitt lymphoma (BL) is the most common form of non-Hodgkin lymphoma in children. Although curable in most cases, those patients who relapse have a dismal prognosis with a 4-year survival of 16% (Cairo, Blood 2007). Identification of biomarkers to predict relapse/refractory disease is needed, and an understanding of the underlying mechanisms is vital to develop novel targeted therapies. We previously identified a recurrent gain of chromosome 13q31 that was associated with relapse and increased expression of miR-17-92 in pediatric BL (Schiffman/Miles, Br J Haematol, 2011). In the present study, we sought to determine the role of miR-17-92 expression in vitro using TALENs mediated deletion of the miR-17-92 locus (MIR17HG) in Raji BL cells that express high levels of miR-17-92 (Shiffman/Miles, 2011).
The modified restriction enzyme and ligation assembly method (Lee/Cairo, ASH 2012) was used to construct MIR17HG TALENs. Expression of miR-17 was assayed by TaqMan assay, and expression of the miR-17-92 target PTEN mRNA and protein were assayed by RT-PCR and western blot, respectively. Proliferation and caspase activity were measured at baseline and after treatment with cyclophosphamide. AKT and phospho (p)AKT were assayed by western blotting. To evaluate mTOR and MAPK pathway activation, pS6 and pERK were assayed by phosphoflow cytometry. In addition, caspase 3 activity was assayed by phosphoflow cytometry to assess induction of apoptosis. Cells were treated with chemotherapy (cyclophosphamide or 4-hydroperoxycyclophosphamide, 4HC), rapamycin, the PI3 kinase inhibitor PIK90, and the dual PI3K/mTOR inhibitor BEZ235 (Axon Medchem). To further validate the role of miR-17 in chemoresistance, miR-17 expression levels were measured by RT-PCR TaqMan assay in the previously developed rituximab-chemotherapy resistant Raji cell lines (RCRCL) Raji 2R and Raji 4RH (Barth et al. Br J Haematol, 2012).
MIR17HG TALENs were used to generate two Raji cell lines with a hemizygous deletion of MIR17HG, #31(+/-) and #48(+/-). Relative to vector control Raji cells, cell lines with hemizygous MIR17HG deletion showed lower expression of miR-17: Raji = 1.0, #31= 0.80-fold, and #48= 0.71-fold. PTEN mRNA was higher in #31 and #48 cells than in the control (Raji = 1.0, #31 = 1.38-fold, and #48 = 2.24-fold), but there was no apparent difference in PTEN protein by western blot. Western blotting showed decreased pAKT with no difference in total AKT. Cyclophosphamide (10mM) induced a significant decrease in proliferation at 48 hours (90.5 +/- 2.9%, p=0.007) in #31 cells vs. control cells that was associated with a trend toward increased caspase 3/7 activity (148.3+/-8.9%, p=0.08). By phosphoflow cytometry, Raji #48 cells showed decreased pS6 compared to control cells and a modestly greater induction of caspase 3 in response to 4HC (3.4 uM) or PIK90 after 48 hours. The combination of 4HC and PIK90, however, significantly induced caspase 3 in Raji #48 cells relative to control Raji cells. Raji cells showed high basal levels of pERK and pS6. BEZ235 or PIK90 treatment decreased p-ERK. Rapamycin did not affect p-ERK but did decrease pS6. While BEZ235 and 4HC each modestly induced caspase alone in Raji cells, the combination caused a much greater increase in caspase 3. In RCRCLs Raji 2R and Raji 4RH, which show high pAKT levels by western blot and phosphoflow cytometry, miR-17 levels were increased (1.73- and 1.69-fold, respectively) relative to chemosensitive Raji cells (1.0) that exhibit lower pAKT levels.
Increased miR-17-92 expression in BL may contribute to therapy resistance, which is further supported by finding increased miR-17 expression in chemotherapy resistant Raji cell lines. Hemizygous deletion of the miR-17-92 locus in BL cells led to increased PTEN mRNA, decreased mTOR pathway activation (decreased pAKT and pS6), and increased sensitivity to chemotherapy. Resistance to PIK90 and 4HC in parental Raji cells that overexpresses miR-17-92 may be due in part to a compensatory upregulation of the MAPK and mTOR pathways, as demonstrated by increased pERK and pS6. Inhibiting this survival mechanism with the PI3K/mTOR inhibitor BEZ235 restored sensitivity to cytotoxic chemotherapy. These findings suggest potential mechanisms underlying BL therapy resistance as well as targeted therapies to overcome resistance.
Cairo:Roche/Genentech: advisory board Other.
Asterisk with author names denotes non-ASH members.