Inhibition of histone deacetylase overcomes rapamycin-mediated resistance in diffuse large B-cell lymphoma by inhibiting Akt signaling through mTORC2

Abstract

The mammalian target of rapamycin (mTOR) has emerged as an important therapeutic target for diffuse large B-cell lymphoma (DLBCL), as recent studies have demonstrated that 30% of relapsed patients respond to mTOR inhibitors. Why some lymphomas are resistant is incompletely understood. In the present study, we demonstrated that rapamycin inhibits mTORC1 in DLBCL lines and primary tumors but is minimally cytotoxic. Subsequent investigations revealed that rapamycin also activated eIF4E and the mTORC2 target Akt, suggesting a potential mechanism of rapamycin resistance. Furthermore, knockdown of the mTORC2 component rictor, but not the mTORC1 component raptor, inhibited rapamycin-induced Akt phosphorylation in lymphoma cells. Addition of the histone deacetylase inhibitor (HDI) LBH589 (LBH) overcame rapamycin resistance by blocking mTOR, thus preventing Akt activation. Further studies support the involvement of the protein phosphatase PP1 in LBH-mediated Akt dephosphorylation, which could be mimicked by knockdown of HDAC3. This is the first demonstration that a HDI such as LBH can overcome rapamycin resistance through a phosphatase that antagonizes mTORC2 activation. These results provide a mechanistic rationale for a clinical trial of a combination of HDI and mTOR inhibitors for DLBCL.

Figures

Figure 1

Activation of mTOR signaling in DLBCL. (A) Constitutive activation of the mTOR signaling pathway in patient samples of DLBCL. Activation of Akt, mTOR, S6rp, and 4EBP1 was assessed in DLBCL tissue by immunohistochemistry with phospho-specific antibodies for Akt mTOR or S6rp or 4EBP1. (B-E) Effect of rapamycin on DLBCL survival. After DHL-6, Ly7, and Ly3 cell lines (B-D) or primary cells from 3 DLBCL patients (E) were treated with the indicated concentrations of rapamycin for 48 hours, survival was assessed using annexin and propidium iodide followed by flow cytometry. Data from 1 representative experiment are shown for each cell line (n = 3). Bar graph shows mean ± standard deviation (SD) from 3 determinations.

Figure 2

Effect of rapamycin on mTORC1/mTORC2 signaling. DHL-6 and Ly7 cells were treated with 50 nM rapamycin for 2, 4, and 8 hours (A) or for 24 hours (B) with the indicated rapamycin concentration. Phosphorylation of Akt, eIF4E, S6, 4EBP1, and mTOR was assessed by Western blot analysis. (C) Malignant cells from DLBCL patients (n = 4) were treated with 50 nM rapamycin for 6 hours before Akt, eIF4E, and S6 phosphorylation was analyzed by Western blot. (D) Levels of PAkt and p-eIF4E were evaluated in peripheral blood samples from 3 patients (n = 3) with aggressive lymphoma treated in the clinic with single-agent temsirolimus.

Figure 3

Effect of LBH on survival, proliferation, and histone acetylation in diffuse large B-cell lymphoma cells. (A) DHL-6, Ly7, and Ly3 cells were treated with indicated concentrations of LBH589 for 24 hours. Western blot analysis of acetylated histones H3 and H4 were performed using specific antibodies. (B-C) DHL-6 and Ly7 diffuse large B-cell lymphoma (DLBCL) cells were treated with indicated concentrations of LBH589 for 48 hours and survival was assessed by annexin V/PI staining by flow cytometry. Data from 1 representative experiment are shown for each cell line (n = 3). (D) Primary cells from DLBCL patients were treated with various doses of LBH589 and survival was assessed by annexin V/PI staining. (E) LY7 cells were pretreated with 50 μM Z-VAD-fmk alone or with LBH589 (50 nM) for 48 hours, and survival was assessed by annexin V/PI staining. (F-G) DHL-6 and Ly7 cells were treated with indicated doses of LBH589 for 48 hours and cell proliferation was assessed by 3H-thymidine incorporation. Error bars indicate mean ± SD from 3 determinations.

Figure 4

Cotreatment with LBH and rapamycin exerts synergistic cytotoxic effects against DLBCL cells. (A-B) DHL-6 (A) and Ly7 (B) cells were treated with the indicated concentrations of LBH589 and/or rapamycin for 48 hours and survival was assessed as described before. (C-D) Effect of LBH + Rap on DLBCL cell proliferation. DHL-6 and Ly7 cells were treated with the indicated doses of LBH and/or rapamycin for 48 hours and cell proliferation was assessed by 3H-thymidine incorporation. One representative experiment is shown for each line (n = 3). Error bars indicate mean ± SD from 3 determinations.

Figure 5

Effect of LBH on mTORC1 and mTORC2. (A-B) DHL-6 and Ly7 DLBCL lines were treated with 100 nM LBH for the indicated times (A) or with the indicated concentrations (B) of LBH for 24 hours and then subjected to Western blot analysis using p-S6 or p-4EBP1 or p-mTOR antibody. (C) Malignant cells from DLBCL patients (n = 4) were treated with 100 nM LBH for 6 hours and analyzed by Western blot. (D) DHL-6 cells were cotreated with 10 nM LBH and 10 nM rapamycin for 24 hours before S6 and 4EBP1 phosphorylation was analyzed. Effect of LBH on mTORC2 targets. (E-F) DHL-6 and Ly7 DLBCL lines were treated with 100 nM LBH for the indicated times (E) or with the indicated concentrations (F) of LBH for 24 hours and then analyzed for Akt or eIF4E phosphorylation. (G) Malignant cells from DLBCL patients (n = 4) were treated with 100 nM LBH for 6 hours and were analyzed for Akt or eIF4E phosphorylation. (H) DHL-6 and Ly7 cells were treated with 100 nM LBH and 20 nM rapamycin for 24 hours before analysis of Akt phosphorylation.

Figure 6

Effect of LBH on upstream proteins involved in Akt activation. (A) After DHL-6 cells were treated with the indicated doses of LBH for 24 hours, cell lysates were immunoblotted for p-p85 and p-PDK1. (B) Immunoprecipitation of PP1 in lysates of LBH-treated cells, followed by immunoblotting with Akt. Simultaneously LBH-treated whole-cell lysates (WCLs) were immunoblotted with PP1 and PP1A. (C) Effect of protein phosphatase inhibitor calyculin A on LBH-induced Akt dephosphorylation. DHL-6 cells were exposed to LBH and calyculin A (50 nM) for 24 hours and cell lysates were immunoblotted. (D) Repressed expression of HDAC3, but not HDAC4, led to Akt dephosphorylation in a manner similar to that of LBH treatment. The cell lysates were immunoblotted with antibodies against phospho-Akt. (E) DHL-6 cells were transiently transfected with CA-Akt and empty vector, and subsequently treated with the combination for 48 hours, and survival was assessed. Bar graph shows mean ± SD. (F) CA-Akt–transfected cells were analyzed for Akt activation using p-Akt antibody. (G-H) Effect of LBH on rictor/raptor complex. Ly7 cells were treated with rapamycin (10 nM) or LBH (50 nM) or the combination for 24 hours. WCLs and mTOR immunoprecipitates prepared from the lysates were analyzed by Western blot for the levels of mTOR, rictor, and raptor.

Figure 7

LBH589 alters the level of intact mTORC2 and c-Myc expression. (A-B) Effect of siRNA knockdown of raptor or rictor. (A) Rictor siRNA blocks phosphorylation of Akt in the presence of rapamycin. Ly7 cells were transiently transfected with raptor or rictor siRNA and then treated with rapamycin (R) or left untreated for 24 hours and subjected to Western blot analysis using p-Akt antibody. (B) After raptor or rictor knockdown, DHL-6 cells were treated with rapamycin (R) or left untreated (0) for 24 hours. Survival was analyzed by staining with annexin V and PI. (C-D) Effect of LBH and rapamycin on cyclin D1 and c-Myc expression. (C) Ly7 cells were treated with the indicated concentrations of LBH589 or rapamycin for 24 hours and then subjected to Western blot analysis using c-Myc and cyclin D1 antibodies. (D) Ly cells were cotreated with 50 nM LBH and rapamycin for 24 hours, and c-Myc and cyclin D1 expression was analyzed. (E-F) DHL-6 cells were transiently transfected with control siRNA or c-Myc siRNA, and proliferation was done as described earlier. Western blot analysis showed successful knockdown of c-Myc (50 nM or 200 nM). (G) HDAC3-transfected DHL-6 cells were analyzed for c-Myc and cyclin D1 expression. Error bars indicate mean ± SD from 3 determinations.