Interferon alpha induces nucleus-independent apoptosis by activating extracellular signal-regulated kinase 1/2 and c-Jun NH2-terminal kinase downstream of phosphatidylinositol 3-kinase and mammalian target of rapamycin

Abstract

Interferon (IFN)alpha induces apoptosis via Bak and Bax and the mitochondrial pathway. Here, we investigated the role of known IFNalpha-induced signaling cascades upstream of Bak activation. By pharmacological and genetic inhibition of the kinases protein kinase C (PKC)delta, extracellular signal-regulated kinase (ERK), and c-Jun NH(2)-terminal kinase (JNK) in U266-1984 and RHEK-1 cells, we could demonstrate that all three enzymes are critical for the apoptosis-associated mitochondrial events and apoptotic cell death induced by IFNalpha, at a step downstream of phosphatidylinositol 3-kinase (PI3K) and mammalian target of rapamycin (mTOR). Furthermore, the activation of JNK was found to occur in a PKCdelta/ERK-dependent manner. Inhibition of these kinases did not affect the canonical IFNalpha-stimulated Janus tyrosine kinase-signal transducer and activator of transcription signaling or expression of IFN-responsive genes. Therefore, enucleated cells (cytoplasts) were examined for IFNalpha-induced apoptosis, to test directly whether this process depends on gene transcription. Cytoplasts were found to undergo apoptosis after IFNalpha treatment, as analyzed by several apoptosis markers by using flow cytometry, live cell imaging, and biochemical analysis of flow-sorted cytoplasts. Furthermore, inhibition of mTOR, ERK, and JNK blocked IFNalpha-induced apoptosis in cytoplasts. In conclusion, IFNalpha-induced apoptosis requires activation of ERK1/2, PKCdelta, and JNK downstream of PI3K and mTOR, and it can occur in a nucleus-independent manner, thus demonstrating for the first time that IFNalpha induces apoptosis in the absence of de novo transcription.

Figure 1.

PKCδ, ERK, and JNK are involved in IFNα-induced apoptosis. U266-1984 cells were treated with or without 300 U/ml IFNα for 48 h in the presence or absence of rottlerin, U0126, SP600125, or a combination of inhibitors. To assess apoptosis, cells were stained for annexin V (a and c) and for Bak activation (b) and analyzed by flow cytometry. The percentages of annexin V- and active Bak-positive cells were analyzed by flow cytometry. The bars represent the mean value of three independent experiments.

Figure 2.

Dominant-negative mutants of PKCδ and JNK inhibit IFNα-induced apoptosis. (a) RHEK-1 cells were treated for 24 and 48 h with the indicated amounts of IFNα, and the percentage of annexin V-positive cells was quantified by using the CellQuest software (BD Biosciences). (b) RHEK-1 cells were transiently transfected with GST (control) or JNK1-APF and the amount of FLICA-positive cells in response to IFNα (5000 U/ml; 48 h) was analyzed by flow cytometry. The results shown are representative of two independent experiments. (c and d) U266-1984 cells were coinfected with adGFP plus adPKCδ-WT or adPKCδ-KD or adMXM, and the PKCδ protein expression and the effect on interferon-induced mitochondrial depolarization in GFP-positive cells (d) was analyzed. PKCδ-FL, PKCδ full length. The bars represent the mean value of two independent experiments.

Figure 3.

PKCδ, ERK, and JNK are sequentially activated downstream of PI3K and mTOR. (a) U266-1984 cells were treated for the indicated time points with 5000 U/ml IFNα in the presence or absence of Ly294002 (10 μM) or rapamycin (1 μM), respectively. The cleavage of PKCδ was detected with immunoblotting. PKCδ-CL: PKCδ cleaved. Tubulin was used as loading control. (b) Immunoblot analysis of IFNα-stimulated phosphorylation of ERK1/2 and JNK1/2 after treatment of U266-1984 cells with Ly294002 (10 μM), rapamycin (1 μM), rottlerin (2 μM), U0126 (10 μM), or SP600125 (10 μM) in the presence or absence of IFNα 5000 U/ml for 16 h. β-Actin was used as a loading control. (c) To further confirm JNK activation as being most proximal to the apoptotic machinery, cells were cultured in the absence or presence of IFNα, with or without pre-incubation with the indicated kinase inhibitors and stained for the phosphorylated form of JNK (Thr183/Tyr185) and analyzed by flow cytometry. The percentages of pJNK-positive cells were quantified using CellQuest software (BD Biosciences) and expressed as relative induction compared with the control cells. (d) Immunoblot analysis of cJun phosphorylation on the serine (S63) residue after treatment of U266-1984 cells with SP600125 (10 μM) in the presence or absence of 5000 U/ml IFNα for 16 h. Tubulin was used as loading control. All results shown are representative of two independent experiments.

Figure 4.

Rottlerin, U0126, and SP600125 do not affect IFNα-induced JAK/STAT signaling. (a) Immunoblot analysis of phosphorylation on residue tyrosine (Y701) and serine (S727) of STAT1 and residue tyrosine (Y689) of STAT2 after 30 min of 300 U/ml IFNα treatment in the presence or absence of inhibitors: Ly294002 (10 μM), rapamycin (1 μM), rottlerin (2 μM), SP600125 (10 μM), or U0126 (10 μM). (b) Flow cytometric analysis of the PML protein levels after 300 U/ml IFNα treatment for 24 h. Gray thin line, control cells; black bold line, IFNα-treated cells. i, black dash and dot, IFNα + rottlerin-treated cells; black dashed line, rottlerin-treated cells. ii, black dash and dot, IFNα + U0126-treated cells; black dashed line, U0126-treated cells. iii, black dash and dot, IFNα + SP600125-treated cells; black dashed line, SP600125-treated cells. The results shown are representative of two independent experiments. (c) IFN-induced transcription of a GAS-regulated luciferase construct in the presence and absence of kinase inhibitors rapamycin (1 μM), SP600125 (10 μM), and U0126 (10 μM). (d) IFN-induced transcription of an ISRE regulated luciferase construct in the presence and absence of kinase inhibitors rapamycin (1 μM), SP600125 (10 μM), and U0126 (10 μM).

Figure 5.

IFNα activated signaling is intact in enucleated cells. (a) IFNα-receptor signaling is intact in enucleated cells. Immunoblot analysis of the FACS-sorted RHEK-1 cytoplasts and nucleated cells treated with 5000 U/ml IFN for 45 min for the tyrosine 701 phosphorylated STAT1, total STAT1, and lamin A/C. These data are representative of two independent experiments. (b) Immunostaining of RHEK-1 cytoplasts and nucleated cells treated with 5000 U/ml IFN for 20 min and stained for total IRS-1. Cytoplasts are indicated by arrows. (c) Flow cytometric analysis of JNK phosphorylation in enucleated cells after 8 h of IFNα. Hoechst staining was used to separate enucleated from nucleated. The histogram shows phosphorylated JNK in the cytoplast population only. C, control; U, U0126 (10 μM); I, IFN (5000 U/ml); and I/U, IFN (5000 U/ml) + U0126 (10 μM).

Figure 6.

IFNα induces cell death in enucleated cells. Live cell imaging by time-lapse microscopy. RHEK-1 cytoplasts and nucleated cells were either treated with 5000 U/ml IFN for 45 h (a) or left untreated (b). Images were taken every 15 min, and the indicated time points were selected to demonstrate changes in the apoptotic markers. RHEK-1 cytoplasts and nucleated cells were stained for nuclear morphology (Hoechst, blue) in combination with TMRE (red) and annexin V (green). Cytoplasts are indicated by arrows. (c) Immunocytochemical analysis of caspase-3 activation (green) and cyt c (red). Release of cyt c manifests as absence of mitochondrial staining. Cytoplasts are indicated by arrows.

Figure 7.

IFNα induces caspase activation in cytoplasts. (a) The cytoplast preparation from RHEK-1 cells was either left untreated or treated with 5000 U/ml IFN for 40 h. The caspase-3 activity in relation to DNA content was assessed by flow cytometry by using an antibody that specifically recognizes active caspase-3, and PI, respectively. i, nucleated RHEK-1 cells (with ∼10-fold higher PI fluorescence, in the top part of the dot plot) represent a contamination of the cytoplast preparation. Numbers represent percentage of the total population staining positive for active caspase-3. Left, untreated cytoplasts; right, cytoplasts treated with 5000 U/ml IFNα for 40 h. The shift to the right in the dot plot represents an increase in active caspase-3–specific immunofluorescence. ii, RHEK-1 cytoplasts and nucleated cells were treated with IFNα for 24 and 40 h. The levels of active caspase-3 were analyzed by flow cytometry. One representative experiment, shown here as -fold increase of active caspase-3 relative to control. (b) After treatment with IFNα for 40 h, the RHEK-1 cytoplasts were separated from nucleated cells by using Hoechst staining and then sorted by using FACSDiVa. The isolated cytoplasts were analyzed by Western blotting, and the cleavage of cytokeratin-18 was detected using M30 antibody. Calreticulin was used as loading control and cytoplasmic marker. (c) Rapamycin, SP600125, and U0126 inhibit IFNα induced caspase cleavage in cytoplasts. RHEK-1 cytoplasts and nucleated cells were pretreated with the indicated inhibitors for 1 h, followed by IFNα for another 40 h. The levels of active caspase-3 were analyzed by flow cytometry, and they are presented as -fold relative to control (i). SP600125 and U0126 had a moderate toxicity in control cells. For clarity, ii shows the same experiment as in i, presented as -fold increase of active caspase-3 by IFNα relative to treatment with respective inhibitor in the absence of IFNα.

Figure 8.

Proposed model of IFNα-induced activation of noncanonical signaling, shown to be essential for the proapoptotic effects of IFNα.