Cancer cells are capable of sphere formation (transformation) through reactive oxygen species (ROS) and glycolysis shift. Transformation is linked to tumorigenesis and therapy resistance, hence targeting regulators of ROS and glycolysis is important for cancer therapeutic candidates. Here, we demonstrate that Smac mimetic AZ58 in combination with tumour necrosis factor-α (TNF-α) was able to inhibit the production of ROS, inhibit glycolysis through Pim-1 kinase-mediated Ser-112 phosphorylation of BAD, and increase depolarization of mitochondria. We also identified mitochondrial isoforms of Pim-1 kinase that were targeted for degradation by AZ58 in combination with TNF-α or AZ58 in combination with Fas ligand (FasL) plus cycloheximide (CHX) through caspase-3 to block transformation. Our study demonstrates that Smac mimetic in combination with TNF-α is an ideal candidate to target Pim-1 expression, inhibit ROS production and to block transformation from blebbishields.

INTRODUCTION

Sphere formation from cells is an important hallmark of cellular transformation [1]. Cellular transformation is an important characteristic feature of tumorigenic cancer stem cells [2]. Recent studies indicated that production of reactive oxygen species (ROS) is essential for K-Ras-mediated cellular transformation [3]. K-Ras-mediated cellular transformation also depends on Pim-1 kinase to fuel transformation through glycolysis by phosphorylating BAD at Ser-112 (pBADS112) [47].

ROS are known to destabilize the mitochondrial outer membrane potential, often initiating apoptosis [8]. Smac is one among the pro-apoptotic molecules that is released from mitochondria upon destabilization of mitochondrial outer membrane potential, which targets cellular inhibitor of apoptosis protein (c-IAP)- and X-linked inhibitor of apoptosis protein (XIAP)-mediated inhibition of active caspases [9,10]. Smac mimetics mimic this function to enable the activation of caspases [11]. However, tumour necrosis factor-α (TNF-α) is known to increase the action of Smac mimetic tremendously to induce apoptosis [11,12]. Although apoptosis was classically considered the end of cells, we recently identified that the blebbishield emergency programme helps cellular transformation after induction of apoptosis [13,14]. It is not known whether ROS-induced apoptosis is part of the transformation process or not. However, Pim-1 kinase has been implicated in the protection of mitochondria against ROS [15], has been implicated in the protection of cells from apoptosis [16] and, importantly, mediates K-Ras-induced cellular transformation [4].

We previously found that Smac mimetic in combination with death ligands targets internal ribosome entry site (IRES) translational targets such as XIAP and c-IAP2 [12,17]. Interestingly, Pim-1 is an IRES translational target [18]. Hence, we investigated whether the Smac mimetic AZ58 can target Pim-1 expression and ROS production under conditions of blebbishield-mediated cellular transformation or its inhibition.

Here, we demonstrate that Pim-1 isoforms are selectively expressed in sphere-forming blebbishields but not in non-sphere-forming blebbishields. Three Pim-1 isoforms (corresponding to AUG, CUG and UUG translation initiation codons) localize at mitochondria; the UUG isoform localizes to the inner mitochondrial membrane, but the other two isoforms localize to the outer mitochondrial membrane. Inhibition of protein translation by cycloheximide (CHX) selectively blocks mitochondrial localization of the Pim-1 AUG isoform but not the CUG and UUG isoforms and reduces pBADS112 accumulation at the outer mitochondrial membrane. Inhibition of protein translation in combination with AZ58 and death ligand TNF-α or Fas ligand (FasL) enhances the degradation of all three isoforms of Pim-1, leading to increased mitochondrial depolarization. AZ58 with TNF-α, but not AZ58 with FasL, effectively targets ROS production to abrogate transformation. Our study demonstrates that Smac mimetic AZ58 with TNF-α or FasL plus CHX is efficient in targeting Pim-1 isoforms to inhibit transformation from blebbishields whereas Smac mimetic in combination with TNF-α but not FasL effectively targets ROS production.

MATERIALS AND METHODS

Cells and maintenance

Human bladder cancer cells RT4v6 (described previously [12,13]), 1A6, T24, ScaBer, HT-1376, HT-1197, RT-112 and RT4 (parent cells of RT4v6, denoted RT4P) were maintained in minimal essential medium (MEM) with 10% fetal bovine serum, L-glutamine, pyruvate, non-essential amino acids, vitamins, penicillin and streptomycin.

Reagents and antibodies

AZ58 is a proprietary Smac mimetic provided by AstraZeneca used at a concentration of 30 nM throughout the study [19,20]. Antibodies against Pim-1 (Sc-13513; used at 1:500 for Western blotting and 1:50 dilution for immunofluorescence, overnight) and cyclo-oxygenase 2 (Cox-2) (Sc-1745) were purchased from Santa Cruz Biotechnology. Antibodies against phospho-BAD-Ser-112 (9296) and phospho-BAD-Ser-136 (9295) were purchased from Cell Signaling Technology. BAD antibody (B31420) was purchased from BD Transduction Laboratories. CHX (239763) was purchased from Calbiochem and used at a concentration of 10 μg/ml throughout the study. Caspase-3 inhibitor benzyloxycarbonyl-Asp-Glu-Val-Asp-fluoromethylketone (z-DEVD-fmk) (FMK004) (used at 20 μM throughout) and recombinant human cytokines TNF-α (210-TA) (used at 17 ng/ml throughout) and FasL (126FL/CF) (used at 17 ng/ml throughout), and antibody against Smac (AF7891) were purchased from R&D Systems. Tigecycline (S1403) (used at 20 μM throughout) was purchased from Selleckchem. Alexa Fluor 488-conjugated anti-mouse antibody (A10680; used at 1:600 dilution for immunofluorescence) and MitoTracker-Deep Red (M22426) were purchased from Molecular Probes/Thermo Fisher Scientific. JC-1 (15003) was purchased from Cayman Chemicals.

Pim-1 isoform genetics

The coding sequence of Pim-1 reference mRNA (NCBI: NM_001243186.1) was subjected to BLAT analysis in UCSC Genome Browser to identify exons. Previously known start codons AUG and CUG were mapped within exon-1 [21]. Exon-1 was scanned for in-frame start codons, including alternative start codons, to predict the first amino acid of isoform-3. In-frame start codons between AUG and CUG were sought because of the reactivity of isoform-3 with anti-Pim-1 antibody [22], which suggested that the isoform was ‘in frame’ similar to AUG and CUG isoforms. An UUG alternative start codon was identified that code for leucine, similar to CUG in the CUG isoform. On the basis of differences in start codon positions, the total molecular mass was calculated for each isoform by multiplying the total number of amino acids by the average amino acid molecular mass of 110 kDa. CUG isoform has 404 amino acids with a calculated molecular mass of 44.44 kDa, UUG isoform has 371 amino acids with a calculated molecular mass of 40.81 kDa, and AUG isoform has 313 amino acids with a calculated molecular mass of 34.43 kDa.

Isolation of sphere-forming and non-sphere-forming blebbishields

Sphere-forming and non-sphere-forming blebbishields were isolated from RT4v6 cells. Briefly, cells were plated at a density of 200 000 cells/ml (13 ml/T-75 flask × four flasks), and at 24 h treated with 10 μg/ml of CHX for 24 h. The floating pyknotic cell populations were then gently collected, pelleted down at 1200 rev./min for 3 min at room temperature and re-plated in 150-mm plates with complete MEM for a further 4 h. The floating cells (non-sphere-forming blebbishields) were washed with complete MEM and collected separately, and the attached spheres (sphere-forming blebbishields) were scraped in fresh ice-cold complete MEM. Both types of blebbishields were pelleted at 3500 rev./min and washed with ice-cold PBS before lysis and Western blotting.

SDS/PAGE and western immunoblotting

The cells were washed with PBS and resuspended in lysis buffer [50 mM Tris/HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA, 25 mM NaF, 1% Triton X-100, 1% NP-40, 0.1 mM Na3VO4, 12.5 mM β-glycerophosphate, 1 mM PMSF and Complete protease inhibitor cocktail (Roche)] and incubated on ice for 30–40 min with vortex-mixing for 20 s every 10 min (total 40 min) (increase the extraction of the Pim-1 UUG isoform from mitochondria). The lysates were clarified at 13 000 rev./min for 10 min, and the supernatants were quantified using BCA assay reagent (Pierce). The lysates were subjected to SDS/PAGE and Western blotting on nitrocellulose membranes. The membranes were developed using ECL reagent (GE Healthcare).

Isolation of mitochondria

The cells were treated as indicated in the figures and were scraped in medium, pelleted and washed with ice-cold PBS at 1000 g × for 5 min each time. Mitochondria were isolated as previously described [23]. Isolated mitochondria were lysed in lysis buffer (described in the Western blotting section) for 40 min, quantified and subjected to Western blotting.

Immunofluorescence and mitochondrial distribution analysis

The cells were fixed overnight using ice-cold methanol, blocked with 1% BSA and 0.3% Triton X-100 in PBS, incubated with anti-Pim-1 antibody overnight at 4°C in blocking buffer, washed with PBS and incubated in Alexa Fluor 488-conjugated secondary antibody for 1 h at room temperature. The cells were then washed with PBS, stained with propidium iodide before imaging with a fluorescence microscope (Olympus). Independently MitoTracker Deep Red was used as per the manufacturer's instructions to validate mitochondrial distribution in RT4v6 cells.

Biochemical localization of proteins at sub-mitochondrial locations

For biochemical localization of proteins at sub-mitochondrial locations, mitochondria were isolated as described above and fresh mitochondria were equally distributed into three parts and pelleted again. The pellets were subjected to protein digestion by proteinase K (50 ng/μl) in M-buffer (210 mM mannitol, 70 mM sucrose, 10 mM HEPES and 1 mM EDTA, pH 7.5) with or without 1% Triton X-100 (final concentration) for 30 min at room temperature. Control mitochondria were resuspended in parallel with M-buffer and incubated for 30 min at room temperature. The reactions were stopped by adding 5 mM PMSF (final concentration) as described in [24]. The mitochondria from these conditions were lysed and subjected to Western blotting as described above.

Quantification of mitochondrial outer membrane permeabilization (MOMP)/depolarization

RT4v6 cells were plated at a density of 50 000 cells/ml and 4 ml/well on six-well plates. After 24 h, cells were treated as indicated in the figures for 23.5 h. At this point, JC-1 dye was added and incubated for a further 30 min before trypsinization and FACS analysis. Cells with intact mitochondria were detected using FL-3 channels, and cells with permeabilized/depolarized mitochondria were detected using FL-1 channels.

ROS measurement

RT4v6 cells were plated at a density of 50 000 cells/ml and 4 ml/well on six-well plates. After 24 h, cells were treated as indicated in the figures for 23.5 h. At this point, 10 μM DCF-DA (2′,7′-dichlorofluorescein diacetate) was added to all wells, and cells were incubated for a further 30 min before analysis of ROS-activated DCF-DA fluorescence (FL-1/525 nm) by FACS (Beckman Coulter, FC500) [25,26]. The histograms were merged using FlowJo software (http:://www.flowjo.com).

Statistical analyses

Statistical analyses were performed using Microsoft Excel 2010. The statistical significance was determined based on Student's t-test with two-tailed distribution and two-sample unequal variance, and P values below 0.05 were considered significant. Error bars represent S.E.M.

RESULTS

Pim-1 AUG, CUG and UUG isoforms are widely expressed in cancer cell lines and are required for transformation from blebbishields

Pim-1 has been reported to have isoforms that are generated by alternative translation initiation [21]. AUG is the classical start codon for Pim-1 isoform-1 which is utilized by ribosomes for translation [21]. Pim-1 isoform-1 has a calculated molecular mass of 34.43 kDa (Figure 1A). Pim-1 isoform-2 is translated from the CUG start codon, which is an alternative start codon usually utilized by mitochondria of lower organisms [21,27]. Pim-1 isoform-2 has a calculated molecular mass of 44.44 kDa (Figure 1A). Another Pim-1 isoform with an intermediate size between that of isoform-1 and isoform-2 has been widely reported [16,22,28]. We have identified an alternative start codon, UUG, which corresponds to this third isoform (Pim-1 isoform-3) (Figure 1A). The UUG alternative start codon falls between the AUG and CUG start codons in frame and can generate a product with calculated molecular mass of 40.81 kDa (Figure 1A). CUG and UUG are established alternative start codons (Figure 1A) [21,27,2932]. In the present paper, for clarity, we designate Pim-1 isoforms based on their start codon: isoform-1 is referred to as ‘the AUG isoform,’ isoform-2 as ‘the CUG isoform’ and isoform-3 as ‘the UUG isoform.’

Pim-1 isoforms AUG, UUG and CUG are selectively expressed in sphere-forming blebbishields but not in non-sphere-forming blebbishields

Figure 1
Pim-1 isoforms AUG, UUG and CUG are selectively expressed in sphere-forming blebbishields but not in non-sphere-forming blebbishields

(A) Pim-1 isoform genetics. Pim-1 mRNA was compared with known protein Pim-1 isoforms (initiated from start codons CUG and AUG) to predict the intermediate in-frame start codon that can generate a Pim-1 isoform intermediate in size between the CUG and AUG isoforms. The bottom panel shows the authentic and alternate start codons (shaded codon boxes) in addition to two additional CUG in-frame alternative start codons (unshaded codon boxes) that did not yield the expected product size by molecular mass calculation. (B) Protein levels of all three Pim-1 isoforms in bladder cancer cell lines. Note the 40-kDa Pim-1 isoform corresponding to the UUG start codon (isoform-3). (C) Selective expression of Pim-1 isoforms and BAD Ser-112 phosphorylation in sphere-forming (SF*) blebbishields but not in non-sphere-forming (NSF*) blebbishields. Blebbishields were generated from RT4v6 cells by exposure to 10 μg/ml CHX for 24 h, and isolated blebbishields were subjected to sphere formation for 4 h in complete MEM.

Figure 1
Pim-1 isoforms AUG, UUG and CUG are selectively expressed in sphere-forming blebbishields but not in non-sphere-forming blebbishields

(A) Pim-1 isoform genetics. Pim-1 mRNA was compared with known protein Pim-1 isoforms (initiated from start codons CUG and AUG) to predict the intermediate in-frame start codon that can generate a Pim-1 isoform intermediate in size between the CUG and AUG isoforms. The bottom panel shows the authentic and alternate start codons (shaded codon boxes) in addition to two additional CUG in-frame alternative start codons (unshaded codon boxes) that did not yield the expected product size by molecular mass calculation. (B) Protein levels of all three Pim-1 isoforms in bladder cancer cell lines. Note the 40-kDa Pim-1 isoform corresponding to the UUG start codon (isoform-3). (C) Selective expression of Pim-1 isoforms and BAD Ser-112 phosphorylation in sphere-forming (SF*) blebbishields but not in non-sphere-forming (NSF*) blebbishields. Blebbishields were generated from RT4v6 cells by exposure to 10 μg/ml CHX for 24 h, and isolated blebbishields were subjected to sphere formation for 4 h in complete MEM.

Examining all three Pim-1 isoforms is necessary to understand the role of Pim-1 during transformation from blebbishields. We first evaluated a panel of bladder cancer cell lines for expression of all three isoforms at protein levels. We found that all three isoforms were expressed at the protein level in RT4v6 cells, including the UUG isoform, at a molecular mass close to 40 kDa (Figure 1B). Expression of AUG isoform was higher in RT4v6 cells than in any other cell lines tested (Figure 1B). In RT4v6 total lysates, expression of the UUG isoform was lower in expression than the other two isoforms (Figure 1B) (we discovered later in the present study that this was due to mitochondrial localization: see Figures 2A and 3A). We focused further studies on RT4v6 cells because we previously demonstrated blebbishield-mediated cellular transformation in this cell line [13].

We next examined whether expression of Pim-1 isoforms is required for blebbishield-mediated transformation. For this purpose, we induced apoptotic blebbishield formation from RT4v6 cells by exposing them to 10 μg/ml CHX for 24 h and isolated pyknotic blebbishields. We allowed a further 4 h of sphere formation from these blebbishields to isolate sphere-forming blebbishields (attached) and non-sphere-forming (floating) blebbishields. Western blotting analysis of these blebbishields revealed Pim-1 expression in sphere-forming blebbishields but not in non-sphere-forming blebbishields, demonstrating that Pim-1 is essential for transformation from blebbishields (Figure 1C).

Cellular transformation requires glycolysis and is regulated by Pim-1 by phosphorylating BAD at Ser-112 (pBADS112) [4]. Analysis of sphere-forming and non-sphere-forming blebbishields demonstrated pBADS112 phosphorylation only in the sphere-forming blebbishields, but not in non-sphere-forming blebbishields, further confirming the role of Pim-1 in blebbishield-mediated transformation (Figure 1C). Taken together, these results demonstrate that Pim-1 isoforms are widely expressed by bladder cancer cell lines and that Pim-1 isoform expression and pBADS112 are essential for transformation from blebbishields.

The AUG isoform mediates BAD Ser-112 phosphorylation despite all three Pim-1 isoforms being differentially localized at mitochondria

Next, we examined the localization of Pim-1 isoforms to precisely understand their connection to transformation. Cell fractionation and Western blotting revealed that the UUG isoform localized specifically to mitochondria, whereas the AUG and CUG isoforms localized to both nuclei and mitochondria-depleted cytoplasm in addition to mitochondria (Figure 2A). We confirmed the identity of the mitochondrial fraction by detecting Smac (Figure 2A). To confirm that BAD Ser-112 phosphorylation coincided with Pim-1 localization, we evaluated the distribution of pBADS112 and BAD phosphorylated at Ser-136 (pBADS136) and found that only pBADS112 specifically localized to mitochondria (Figure 2A). To further confirm the mitochondrial localization of Pim-1, we carried out immunofluorescence analysis and found that Pim-1 predominantly localized to mitochondria distributed around nucleus (Figure 2A1). To validate the mitochondrial distribution around nucleus in RT4v6 cells, we used MitoTracker Deep Red and found similar distribution pattern of mitochondria around nuclei (Figure 2A2).

Pim-1 isoforms AUG and CUG localize on the outer mitochondrial membrane with pBADS112, whereas Pim-1 isoform UUG localizes inside mitochondria

Figure 2
Pim-1 isoforms AUG and CUG localize on the outer mitochondrial membrane with pBADS112, whereas Pim-1 isoform UUG localizes inside mitochondria

(A) Subcellular localization of Pim-1 isoforms and pBADS112 in RT4v6 cells. The Pim-1 UUG isoform specifically localized to mitochondria, where pBADS112 but not pBADS136 co-localized with Pim-1. Cytoplasm includes mitochondria. MD*cytoplasm: mitochondria-depleted cytoplasm. (A1) Immunofluorescence localization of Pim-1 in RT4v6 cells where Pim-1 localized predominantly in mitochondria (arrows) around nuclei (stained with propidium iodide, PI). (A2) Mitochondria distribution around nuclei in RT4v6 cells was visualized using MitoTracker Deep Red. (B) Sub-mitochondrial localization of Pim-1 isoforms and pBADS112 in RT4v6 mitochondria. The Pim-1 UUG isoform localized inside mitochondria (lane 2), but AUG and CUG isoforms co-localized with pBADS112 at the outer mitochondrial membrane. (C) Serum withdrawal increased expression of the Pim-1 AUG isoform and subsequently pBADS112 in RT4v6 cells. *SFM: serum-free medium. (C1) Densitometric quantification of the Pim-1 AUG isoform and pBADS112, both were normalized to actin of corresponding lanes. Black asterisk indicates the fall in Pim-1 AUG isoform quantity and red asterisk indicates subsequent fall in pBADS112 quantity. (D) Schematic representation suggesting that Pim-1 AUG isoform mediates BAD Ser-112 phosphorylation (also see Figure 3).

Figure 2
Pim-1 isoforms AUG and CUG localize on the outer mitochondrial membrane with pBADS112, whereas Pim-1 isoform UUG localizes inside mitochondria

(A) Subcellular localization of Pim-1 isoforms and pBADS112 in RT4v6 cells. The Pim-1 UUG isoform specifically localized to mitochondria, where pBADS112 but not pBADS136 co-localized with Pim-1. Cytoplasm includes mitochondria. MD*cytoplasm: mitochondria-depleted cytoplasm. (A1) Immunofluorescence localization of Pim-1 in RT4v6 cells where Pim-1 localized predominantly in mitochondria (arrows) around nuclei (stained with propidium iodide, PI). (A2) Mitochondria distribution around nuclei in RT4v6 cells was visualized using MitoTracker Deep Red. (B) Sub-mitochondrial localization of Pim-1 isoforms and pBADS112 in RT4v6 mitochondria. The Pim-1 UUG isoform localized inside mitochondria (lane 2), but AUG and CUG isoforms co-localized with pBADS112 at the outer mitochondrial membrane. (C) Serum withdrawal increased expression of the Pim-1 AUG isoform and subsequently pBADS112 in RT4v6 cells. *SFM: serum-free medium. (C1) Densitometric quantification of the Pim-1 AUG isoform and pBADS112, both were normalized to actin of corresponding lanes. Black asterisk indicates the fall in Pim-1 AUG isoform quantity and red asterisk indicates subsequent fall in pBADS112 quantity. (D) Schematic representation suggesting that Pim-1 AUG isoform mediates BAD Ser-112 phosphorylation (also see Figure 3).

We further dissected the sub-mitochondrial localization of Pim-1 and pBADS112 in mitochondria by isolating mitochondria and subjecting them to proteinase K digestion in the presence or absence of Triton X-100. In the presence of Triton X-100, proteinase K can enter mitochondria and digest proteins both inside and outside mitochondria, whereas in the absence of Triton X-100, proteinase K can only digest proteins on the mitochondrial surface (mitochondrial outer membrane) [24]. The present study revealed that the Pim-1 AUG and CUG isoforms localized at the outer surface of mitochondria, whereas the UUG isoform localized exclusively inside mitochondria (Figure 2B). Interestingly, pBADS112 also localized at the outer surface of mitochondria, and there was only a very small amount of total BAD (non-pBADS112) localized inside mitochondria (Figure 2B). These results suggested that the Pim-1 UUG isoform may not have a role in phosphorylation of BAD at Ser-112.

To identify whether AUG or CUG isoform phosphorylates BAD at Ser-112, we performed a pulse–chase analysis by withdrawing serum from culture medium and detected the changes in expression of pBADS112 in relation to the Pim-1 AUG and CUG isoforms in total lysates over time. Serum withdrawal markedly induced the expression of the AUG isoform, which was followed by an increase in the expression of pBADS112 (Figure 2C). Interestingly, the expression of pBADS112 fell between 8 and 16 h, and this reduction in pBADS112 expression was preceded by a reduction in AUG isoform expression, indicating that the Pim-1 AUG isoform mediates BAD Ser-112 phosphorylation (Figures 2C, 2C1 and 2D).

Inhibiting Pim-1 AUG isoform mitochondrial localization reduces BAD Ser-112 phosphorylation

To further investigate which Pim-1 isoform mediates BAD Ser-112 phosphorylation, we took advantage of the fact that different Pim-1 isoforms use different start codons and the fact that UUG can initiate translation using mitochondrial ribosomes [33]. We used two different translation inhibitors: CHX, which predominantly blocks cytoplasmic protein translation, and tigecycline, which predominantly blocks mitochondrial protein translation [3436]. Interestingly, inhibition of cytoplasmic and endoplasmic reticulum protein translation using CHX blocked Pim-1 AUG isoform localization to mitochondria (although the mitochondria-depleted cytoplasm had a significant amount of AUG isoform) and reduced the levels of pBADS112 on mitochondria (Figure 3A). In contrast, CUG and UUG isoforms had no or little effect on mitochondrial localization (Figure 3A). Interestingly, CHX enhanced mitochondrial Cox-2 (Figure 3A), a protein product translated from mRNA by mitochondrial translation [34]. Inhibition of mitochondrial protein translation using tigecycline did not reduce mitochondrial Pim-1 AUG, CUG or UUG isoforms or mitochondrial pBADS112 (Figure 3A). Lack of Pim-1 isoform reduction in response to tigecycline indicates that the Pim-1 isoforms might have long protein stability even though mitochondrial Cox-2 was reduced (Figure 3A). Taken together with the results shown in Figure 2, these results indicate that Pim-1 AUG and CUG isoforms localize with pBADS112 at the outer mitochondrial membrane, where only the AUG isoform regulates BAD Ser-112 phosphorylation, and demonstrate that the UUG isoform localizes inside mitochondria without affecting BAD Ser-112 phosphorylation (Figure 3B).

The Pim-1 AUG isoform but not the CUG or UUG isoform mediates phosphorylation of BAD at Ser-112

Figure 3
The Pim-1 AUG isoform but not the CUG or UUG isoform mediates phosphorylation of BAD at Ser-112

(A) CHX blocked Pim-1 AUG isoform translocation from the cytosol to mitochondria and led to reduced pBADS112. Cox-2 was examined to assess the effect of tigecycline on mitochondrial translation. Cy-Cox-2: cytoplasmic Cox-2; Mt-Cox-2: mitochondrial Cox-2. (B) Schematic diagram showing that CHX inhibits Pim-1 AUG isoform translocation to the outer mitochondrial membrane and reduces mitochondrial pBADS112.

Figure 3
The Pim-1 AUG isoform but not the CUG or UUG isoform mediates phosphorylation of BAD at Ser-112

(A) CHX blocked Pim-1 AUG isoform translocation from the cytosol to mitochondria and led to reduced pBADS112. Cox-2 was examined to assess the effect of tigecycline on mitochondrial translation. Cy-Cox-2: cytoplasmic Cox-2; Mt-Cox-2: mitochondrial Cox-2. (B) Schematic diagram showing that CHX inhibits Pim-1 AUG isoform translocation to the outer mitochondrial membrane and reduces mitochondrial pBADS112.

Smac mimetic AZ58 targets Pim-1 isoforms for degradation through caspase-3 to inhibit transformation from blebbishields

Although CHX reduced Pim-1 AUG isoform localization to the outer mitochondrial membrane and reduced BAD Ser-112 phosphorylation, CHX was not sufficient to block transformation from blebbishields as CHX-generated blebbishields could still transform (sphere formation) and express pBADS112 (Figure 1C). Thus, we investigated whether degradation of Pim-1 isoforms is necessary to block transformation from blebbishields. For this purpose, we chose two treatment conditions that differentially influence transformation from blebbishields: TNF-α plus AZ58 (Smac mimetic), which inhibits transformation from blebbishields, and FasL plus AZ58, which permits transformation from blebbishields (our unpublished work). As expected, TNF-α plus AZ58 abolished expression of Pim-1 isoforms, whereas FasL plus AZ58 did not (Figure 4A). In addition, TNF-α plus AZ58 abolished expression of pBADS112, whereas FasL plus AZ58 did not (Figure 4A). These results demonstrate that Pim-1 isoform degradation is necessary to block transformation from blebbishields.

Smac mimetic AZ58 plus TNF-α targets Pim-1 isoforms through caspase-3, whereas FasL plus AZ58 requires CHX supplement to target Pim-1 isoforms

Figure 4
Smac mimetic AZ58 plus TNF-α targets Pim-1 isoforms through caspase-3, whereas FasL plus AZ58 requires CHX supplement to target Pim-1 isoforms

(A) TNF-α plus AZ58 efficiently targeted Pim-1 isoforms and abolished sphere formation from blebbishields, whereas FasL plus AZ58 could not target Pim-1 isoforms and permits sphere formation (transformation) from blebbishields. Note the loss of pBADS112 and monomeric BAD following treatment with TNF-α plus AZ58. (B) CHX in combination with TNF-α and CHX in combination with FasL plus AZ58 targeted all three Pim-1 isoforms for degradation at mitochondria and in total blebbishields to abolish sphere formation (transformation). (C) AZ58 with TNF-α targeted Pim-1 isoforms through caspase-3. Bottom panel, schematic representation of Pim-1 isoform degradation. (D) JC-1 dye-stained cells were subjected to FACS to show that inhibition of transformation is caused by a significantly reduced number of cells with polarized mitochondria (red), and an increased number of cells with depolarized mitochondria (green).

Figure 4
Smac mimetic AZ58 plus TNF-α targets Pim-1 isoforms through caspase-3, whereas FasL plus AZ58 requires CHX supplement to target Pim-1 isoforms

(A) TNF-α plus AZ58 efficiently targeted Pim-1 isoforms and abolished sphere formation from blebbishields, whereas FasL plus AZ58 could not target Pim-1 isoforms and permits sphere formation (transformation) from blebbishields. Note the loss of pBADS112 and monomeric BAD following treatment with TNF-α plus AZ58. (B) CHX in combination with TNF-α and CHX in combination with FasL plus AZ58 targeted all three Pim-1 isoforms for degradation at mitochondria and in total blebbishields to abolish sphere formation (transformation). (C) AZ58 with TNF-α targeted Pim-1 isoforms through caspase-3. Bottom panel, schematic representation of Pim-1 isoform degradation. (D) JC-1 dye-stained cells were subjected to FACS to show that inhibition of transformation is caused by a significantly reduced number of cells with polarized mitochondria (red), and an increased number of cells with depolarized mitochondria (green).

Although Pim-1 AUG and CUG isoform degradation were clear from these results, the status of UUG isoform in total lysates remained obscure. We therefore examined UUG isoform status in isolated mitochondria. For this purpose we chose additional conditions which are known to permit or inhibit transformation from blebbishields: blebbishields generated with CHX alone or FasL plus AZ58 can transform, whereas blebbishields generated with CHX plus TNF-α and blebbishields generated with CHX plus FasL plus AZ58 are unable to transform (our unpublished work). As expected, all three Pim-1 isoforms were lost in blebbishields that are not capable of transformation, whereas these isoforms were retained in blebbishields that are capable of transformation (Figure 4B).

We found that the caspase-3 inhibitor z-DEVD-fmk could completely inhibit Pim-1 degradation in response to TNF-α plus AZ58 (Figure 4C), indicating that Pim-1 isoforms were degraded by caspase-3. These results demonstrated that caspase-3-mediated degradation of Pim-1 isoforms is necessary to block transformation from blebbishields.

Since Pim-1 loss is associated with loss of transformation from blebbishields (Figure 4B), and since Pim-1 is known to protect mitochondria [37], we examined the mitochondrial polarization status by using JC-1 staining which emits red fluorescence when the mitochondria are polarized (healthy) and emits green fluorescence upon mitochondrial depolarization (damaged/dysfunctional) [38]. We found that Pim-1 loss is associated with significant increase in cells with depolarized mitochondria at the expense of cells with polarized mitochondria (Figure 4D).

Smac mimetic targets ROS production effectively with TNF-α but not with FasL

Pim-1 is linked to the protection of cells from ROS [15], and ROS production is essential for cellular transformation [3]. Hence, we examined whether Pim-1 isoform targeting by AZ58 with TNF-α or FasL could target ROS production in blebbishields. TNF-α efficiently reduced the number of ROS-positive cells with AZ58; however, with FasL plus AZ58, it was insignificant (Figure 5A). We also examined the intensity of ROS and found that high-ROS-positive cells were targeted more efficiently by TNF-α with AZ58 than by FasL with AZ58 (Figure 5B, rightmost panel). Of note, AZ58, TNF-α and FasL as single agents could not alter ROS intensity, indicating that their combination is required to target ROS production (Figure 5B).

AZ58 plus TNF-α but not AZ58 plus FasL efficiently blocks ROS production in blebbishields

Figure 5
AZ58 plus TNF-α but not AZ58 plus FasL efficiently blocks ROS production in blebbishields

(A) AZ58 plus TNF-α targeted ROS-positive cells more efficiently than AZ58 plus FasL did. RT4v6 cells were treated as indicated and ROS-positive cells were identified by DCF-DA fluorescence. (B) Quantitatively, AZ58 plus TNF-α targeted high-ROS-positive cells more efficiently than AZ58 plus FasL did. Arrows indicate high-ROS-positive cells.

Figure 5
AZ58 plus TNF-α but not AZ58 plus FasL efficiently blocks ROS production in blebbishields

(A) AZ58 plus TNF-α targeted ROS-positive cells more efficiently than AZ58 plus FasL did. RT4v6 cells were treated as indicated and ROS-positive cells were identified by DCF-DA fluorescence. (B) Quantitatively, AZ58 plus TNF-α targeted high-ROS-positive cells more efficiently than AZ58 plus FasL did. Arrows indicate high-ROS-positive cells.

Since FasL could not inhibit ROS production significantly (Figure 5), we examined the impact of adding CHX to TNF-α or FasL plus AZ58 on ROS-positive cells and ROS production (Figure 6). CHX plus TNF-α completely abrogated ROS-positive blebbishields, whereas CHX plus FasL plus AZ58 did not (Figure 6A). Of note, CHX and its vehicle, methanol, increased ROS-positive cells (methanol as a single agent does not produce blebbishields or pyknosis, whereas CHX does) (Figure 6A), but blebbishields generated by CHX were capable of transformation (Figure 1C). Analysis of ROS intensity revealed that FasL plus AZ58 reduced high-ROS-positive cells, and addition of CHX further reduced high-ROS-positive blebbishields (Figure 6B). However, addition of CHX to FasL plus AZ58 could not abrogate ROS-positive cells as addition of CHX to TNF-α did (Figure 6B). Although CHX (with methanol vehicle) increased high-ROS-positive cells, the combination of CHX with TNF-α targeted the high-ROS-positive population (Figure 6B). Together, these data demonstrate that high ROS positivity is not detrimental to blebbishield-mediated transformation and that ROS production is efficiently targeted by AZ58 plus TNF-α but not AZ58 plus FasL.

ROS production is abrogated in CHX plus TNF-α generated blebbishields but not in FasL plus AZ58 plus CHX-generated blebbishields

Figure 6
ROS production is abrogated in CHX plus TNF-α generated blebbishields but not in FasL plus AZ58 plus CHX-generated blebbishields

(A) CHX plus TNF-α targeted ROS production in blebbishields more efficiently than CHX plus FasL plus AZ58 did. RT4v6 cells were treated as indicated, and ROS-positive cells were identified by DCF-DA fluorescence. (B) Quantitatively, CHX plus TNF-α targeted high-ROS-positive cells more efficiently than CHX plus FasL plus AZ58 did. Arrows indicate high-ROS-positive cells. Note that CHX alone induced ROS because of the methanol solvent. However, CHX plus TNF-α completely abrogated ROS production.

Figure 6
ROS production is abrogated in CHX plus TNF-α generated blebbishields but not in FasL plus AZ58 plus CHX-generated blebbishields

(A) CHX plus TNF-α targeted ROS production in blebbishields more efficiently than CHX plus FasL plus AZ58 did. RT4v6 cells were treated as indicated, and ROS-positive cells were identified by DCF-DA fluorescence. (B) Quantitatively, CHX plus TNF-α targeted high-ROS-positive cells more efficiently than CHX plus FasL plus AZ58 did. Arrows indicate high-ROS-positive cells. Note that CHX alone induced ROS because of the methanol solvent. However, CHX plus TNF-α completely abrogated ROS production.

DISCUSSION

Pim-1 is an important oncogene that mediates K-Ras-induced cellular transformation, regulates glycolysis by phosphorylating BAD, and mediates protection of mitochondria from ROS and other forms of mitochondria-damaging signals such as calcium overload [47,15,37]. Targeting Pim-1 is thus crucial for a cancer therapeutic to be efficient, either alone or in combination with common cytokines. Here, we demonstrated the sub-mitochondrial localization of Pim-1 isoforms and identified the distinct role of the Pim-1 AUG isoform but not the UUG or CUG isoforms in phosphorylation of BAD at Ser-112. Furthermore, we demonstrated that Pim-1 isoforms were efficiently targeted for degradation when AZ58 is delivered with TNF-α or when AZ58 is delivered with FasL supplemented with CHX.

Our findings from the present study show that Pim-1 is an important IRES target with survival roles in blebbishield-mediated cellular transformation. Complete down-regulation of Pim-1 by TNF-α plus AZ58 is an important aspect in blocking transformation from blebbishields (Figures 4A and 4B), because CHX alone does not degrade Pim-1 isoforms (Figure 4B) yet supports transformation from blebbishields. FasL plus AZ58 (Figure 1C), on the other hand is unable to down-regulate Pim-1 isoforms as efficiently as TNF-α with AZ58 but supports transformation (Figures 4A and 4B). Concurrent with this, increased depolarization of mitochondria is associated with loss of transformation from blebbishields (Figure 4D).

In summary, our study identified Pim-1 isoform localization in sub-mitochondrial locations, demonstrated the importance of Pim-1 expression and localization at the outer membrane of mitochondria in blebbishield-mediated transformation and identified AZ58 as an effective therapeutic agent to target Pim-1 isoforms and ROS production in combination with TNF-α.

AUTHOR CONTRIBUTION

Goodwin Jinesh conceived the hypothesis, designed the study, performed all experiments, interpreted the data and wrote the paper. Naomi Laing contributed AZ58. Ashish Kamat generated the RT4v6 cell line, interpreted the data, and provided scientific and editorial oversight. All authors read and agree to the contents of this paper.

We sincerely thank Dr David J. McConkey for providing various bladder cancer cell lines, providing various reagents used in the present study, and discussing data; Ms I-Ling Lee, Ms Christina Medina, Ms Maria Ramirez, Ms Melissa Zenil, Dr Chinedu Mmeje, Dr Rikiya Taoka and Dr Hisashi Takeuchi for technical help; and Ms Stephanie Deming for editorial assistance. Naomi Laing is employed by AstraZeneca, which provided the Smac mimetic AZ58.

FUNDING

This work was supported by the Cancer Center Support Grant : P30CA016672.

Abbreviations

     
  • CHX

    cycloheximide

  •  
  • c-IAP

    cellular inhibitor of apoptosis protein

  •  
  • Cox-2

    cyclo-oxygenase 2

  •  
  • DCF-DA

    2′,7′-dichlorofluorescein diacetate

  •  
  • FasL

    Fas ligand

  •  
  • IRES

    internal ribosome entry site

  •  
  • MEM

    minimal essential medium

  •  
  • pBADS112

    BAD phosphorylated at Ser-112

  •  
  • pBADS136

    BAD phosphorylated at serine-136

  •  
  • ROS

    reactive oxygen species

  •  
  • TNF-α

    tumour necrosis factor-α

  •  
  • XIAP

    X-linked inhibitor of apoptosis protein, z-DEVD-fmk, benzyloxycarbonyl-Asp-Glu-Val-Asp-fluoromethylketone

References

References
1
Bapat
S.A.
Mali
A.M.
Koppikar
C.B.
Kurrey
N.K.
Stem and progenitor-like cells contribute to the aggressive behavior of human epithelial ovarian cancer
Cancer Res.
2005
, vol. 
65
 (pg. 
3025
-
3029
)
[PubMed]
2
Lobo
N.A.
Shimono
Y.
Qian
D.
Clarke
M.F.
The biology of cancer stem cells
Annu. Rev. Cell Dev. Biol.
2007
, vol. 
23
 (pg. 
675
-
699
)
[PubMed]
3
Park
M.T.
Kim
M.J.
Suh
Y.
Kim
R.K.
Kim
H.
Lim
E.J.
Yoo
K.C.
Lee
G.H.
Kim
Y.H.
Hwang
S.G.
, et al. 
Novel signaling axis for ROS generation during K-Ras-induced cellular transformation
Cell Death Differ.
2014
, vol. 
21
 (pg. 
1185
-
1197
)
[PubMed]
4
Xu
D.
Allsop
S.A.
Witherspoon
S.M.
Snider
J.L.
Yeh
J.J.
Fiordalisi
J.J.
White
C.D.
Williams
D.
Cox
A.D.
Baines
A.T.
The oncogenic kinase Pim-1 is modulated by K-Ras signaling and mediates transformed growth and radioresistance in human pancreatic ductal adenocarcinoma cells
Carcinogenesis
2011
, vol. 
32
 (pg. 
488
-
495
)
[PubMed]
5
Leung
C.O.
Wong
C.C.
Fan
D.N.
Kai
A.K.
Tung
E.K.
Xu
I.M.
Ng
I.O.
Lo
R.C.
PIM1 regulates glycolysis and promotes tumor progression in hepatocellular carcinoma
Oncotarget
2015
, vol. 
6
 (pg. 
10880
-
10892
)
[PubMed]
6
Aho
T.L.
Sandholm
J.
Peltola
K.J.
Mankonen
H.P.
Lilly
M.
Koskinen
P.J.
Pim-1 kinase promotes inactivation of the pro-apoptotic Bad protein by phosphorylating it on the Ser112 gatekeeper site
FEBS Lett.
2004
, vol. 
571
 (pg. 
43
-
49
)
[PubMed]
7
Fang
X.
Yu
S.
Eder
A.
Mao
M.
Bast
R.C.
Jr
Boyd
D.
Mills
G.B.
Regulation of BAD phosphorylation at serine 112 by the Ras-mitogen-activated protein kinase pathway
Oncogene
1999
, vol. 
18
 (pg. 
6635
-
6640
)
[PubMed]
8
Suski
J.M.
Lebiedzinska
M.
Bonora
M.
Pinton
P.
Duszynski
J.
Wieckowski
M.R.
Relation between mitochondrial membrane potential and ROS formation
Methods Mol. Biol.
2014
, vol. 
810
 (pg. 
183
-
205
)
9
Adrain
C.
Creagh
E.M.
Martin
S.J.
Apoptosis-associated release of Smac/DIABLO from mitochondria requires active caspases and is blocked by Bcl-2
EMBO J.
2001
, vol. 
20
 (pg. 
6627
-
6636
)
[PubMed]
10
Srinivasula
S.M.
Hegde
R.
Saleh
A.
Datta
P.
Shiozaki
E.
Chai
J.
Lee
R.A.
Robbins
P.D.
Fernandes-Alnemri
T.
Shi
Y.
Alnemri
E.S.
A conserved XIAP-interaction motif in caspase-9 and Smac/DIABLO regulates caspase activity and apoptosis
Nature
2001
, vol. 
410
 (pg. 
112
-
116
)
[PubMed]
11
Varfolomeev
E.
Blankenship
J.W.
Wayson
S.M.
Fedorova
A.V.
Kayagaki
N.
Garg
P.
Zobel
K.
Dynek
J.N.
Elliott
L.O.
Wallweber
H.J.
, et al. 
IAP antagonists induce autoubiquitination of c-IAPs, NF-kappaB activation, and TNFalpha-dependent apoptosis
Cell
2007
, vol. 
131
 (pg. 
669
-
681
)
[PubMed]
12
Jinesh
G.G.
Chunduru
S.
Kamat
A.M.
Smac mimetic enables the anticancer action of BCG-stimulated neutrophils through TNF-alpha but not through TRAIL and FasL
J. Leukoc. Biol.
2012
, vol. 
92
 (pg. 
233
-
244
)
[PubMed]
13
Jinesh
G.G.
Choi
W.
Shah
J.B.
Lee
E.K.
Willis
D.L.
Kamat
A.M.
Blebbishields, the emergency program for cancer stem cells: sphere formation and tumorigenesis after apoptosis
Cell Death Differ.
2013
, vol. 
20
 (pg. 
382
-
395
)
[PubMed]
14
Goodwin Jinesh
G.
Willis
D.L.
Kamat
A.M.
Bladder cancer stem cells: biological and therapeutic perspectives
Curr. Stem Cell Res. Ther.
2014
, vol. 
9
 (pg. 
89
-
101
)
[PubMed]
15
Lilly
M.
Sandholm
J.
Cooper
J.J.
Koskinen
P.J.
Kraft
A.
The PIM-1 serine kinase prolongs survival and inhibits apoptosis-related mitochondrial dysfunction in part through a bcl-2-dependent pathway
Oncogene
1999
, vol. 
18
 (pg. 
4022
-
4031
)
[PubMed]
16
Xie
Y.
Xu
K.
Dai
B.
Guo
Z.
Jiang
T.
Chen
H.
Qiu
Y.
The 44 kDa Pim-1 kinase directly interacts with tyrosine kinase Etk/BMX and protects human prostate cancer cells from apoptosis induced by chemotherapeutic drugs
Oncogene
2006
, vol. 
25
 (pg. 
70
-
78
)
[PubMed]
17
Jinesh
G.G.
Kamat
A.M.
Redirecting neutrophils against bladder cancer cells by BCG and Smac mimetic combination
Oncoimmunology
2012
, vol. 
1
 (pg. 
1161
-
1162
)
[PubMed]
18
Johannes
G.
Carter
M.S.
Eisen
M.B.
Brown
P.O.
Sarnow
P.
Identification of eukaryotic mRNAs that are translated at reduced cap binding complex eIF4F concentrations using a cDNA microarray
Proc. Natl. Acad. Sci. U.S.A.
1999
, vol. 
96
 (pg. 
13118
-
13123
)
[PubMed]
19
Hennessy
E.J.
Adam
A.
Aquila
B.M.
Castriotta
L.M.
Cook
D.
Hattersley
M.
Hird
A.W.
Huntington
C.
Kamhi
V.M.
Laing
N.M.
, et al. 
Discovery of a novel class of dimeric Smac mimetics as potent IAP antagonists resulting in a clinical candidate for the treatment of cancer (AZD5582)
J. Med. Chem.
2013
, vol. 
56
 (pg. 
9897
-
9919
)
[PubMed]
20
Lee
E.K.
Jinesh
G.G.
Laing
N.M.
Choi
W.
McConkey
D.J.
Kamat
A.M.
A Smac mimetic augments the response of urothelial cancer cells to gemcitabine and cisplatin
Cancer Biol. Ther.
2013
, vol. 
14
 (pg. 
812
-
822
)
[PubMed]
21
Saris
C.J.
Domen
J.
Berns
A.
The pim-1 oncogene encodes two related protein-serine/threonine kinases by alternative initiation at AUG and CUG
EMBO J.
1991
, vol. 
10
 (pg. 
655
-
664
)
[PubMed]
22
Hu
X.F.
Li
J.
Vandervalk
S.
Wang
Z.
Magnuson
N.S.
Xing
P.X.
PIM-1-specific mAb suppresses human and mouse tumor growth by decreasing PIM-1 levels, reducing Akt phosphorylation, and activating apoptosis
J. Clin. Invest.
2009
, vol. 
119
 (pg. 
362
-
375
)
[PubMed]
23
Graham
J.M.
Isolation of mitochondria from tissues and cells by differential centrifugation
Curr. Protoc. Cell Biol.
2001
, vol. 
Chapter 3
 pg. 
Unit 3.3
 
[PubMed]
24
Zhao
J.
Liu
T.
Jin
S.B.
Tomilin
N.
Castro
J.
Shupliakov
O.
Lendahl
U.
Nister
M.
The novel conserved mitochondrial inner-membrane protein MTGM regulates mitochondrial morphology and cell proliferation
J. Cell Sci.
2009
, vol. 
122
 (pg. 
2252
-
2262
)
[PubMed]
25
Negre-Salvayre
A.
Auge
N.
Duval
C.
Robbesyn
F.
Thiers
J.C.
Nazzal
D.
Benoist
H.
Salvayre
R.
Detection of intracellular reactive oxygen species in cultured cells using fluorescent probes
Methods Enzymol.
2002
, vol. 
352
 (pg. 
62
-
71
)
[PubMed]
26
Evgeniy
E.
Sergei
K.
Identification of ROS using oxidized DCFDA and flow-cytometry. Advanced protocols in oxidative stress II
Methods Mol. Biol.
2010
, vol. 
594
 (pg. 
57
-
72
)
[PubMed]
27
Touriol
C.
Bornes
S.
Bonnal
S.
Audigier
S.
Prats
H.
Prats
A.C.
Vagner
S.
Generation of protein isoform diversity by alternative initiation of translation at non-AUG codons
Biol. Cell
2003
, vol. 
95
 (pg. 
169
-
178
)
[PubMed]
28
Guo
S.
Mao
X.
Chen
J.
Huang
B.
Jin
C.
Xu
Z.
Qiu
S.
Overexpression of Pim-1 in bladder cancer
J. Exp. Clin. Cancer Res.
2010
, vol. 
29
 pg. 
161
 
[PubMed]
29
Chang
K.J.
Wang
C.C.
Translation initiation from a naturally occurring non-AUG codon in Saccharomyces cerevisiae
J. Biol. Chem.
2004
, vol. 
279
 (pg. 
13778
-
13785
)
[PubMed]
30
Chen
S.J.
Lin
G.
Chang
K.J.
Yeh
L.S.
Wang
C.C.
Translational efficiency of a non-AUG initiation codon is significantly affected by its sequence context in yeast
J. Biol. Chem.
2008
, vol. 
283
 (pg. 
3173
-
3180
)
[PubMed]
31
Starck
S.R.
Jiang
V.
Pavon-Eternod
M.
Prasad
S.
McCarthy
B.
Pan
T.
Shastri
N.
Leucine-tRNA initiates at CUG start codons for protein synthesis and presentation by MHC class I
Science
2012
, vol. 
336
 (pg. 
1719
-
1723
)
[PubMed]
32
Gerashchenko
M.V.
Su
D.
Gladyshev
V.N.
CUG start codon generates thioredoxin/glutathione reductase isoforms in mouse testes
J. Biol. Chem.
2010
, vol. 
285
 (pg. 
4595
-
4602
)
[PubMed]
33
Rusconi
C.P.
Cech
T.R.
Mitochondrial import of only one of three nuclear-encoded glutamine tRNAs in Tetrahymena thermophila
EMBO J.
1996
, vol. 
15
 (pg. 
3286
-
3295
)
[PubMed]
34
Skrtic
M.
Sriskanthadevan
S.
Jhas
B.
Gebbia
M.
Wang
X.
Wang
Z.
Hurren
R.
Jitkova
Y.
Gronda
M.
Maclean
N.
, et al. 
Inhibition of mitochondrial translation as a therapeutic strategy for human acute myeloid leukemia
Cancer Cell
2011
, vol. 
20
 (pg. 
674
-
688
)
[PubMed]
35
Hanas
J.
Linden
G.
Stuart
K.
Mitochondrial and cytoplasmic ribosomes and their activity in blood and culture form Trypanosoma brucei
J. Cell Biol.
1975
, vol. 
65
 (pg. 
103
-
111
)
[PubMed]
36
Sutton
C.A.
Ares
M.
Jr
Hallberg
R.L.
Cycloheximide resistance can be mediated through either ribosomal subunit
Proc. Natl. Acad. Sci. U.S.A.
1978
, vol. 
75
 (pg. 
3158
-
3162
)
[PubMed]
37
Borillo
G.A.
Mason
M.
Quijada
P.
Volkers
M.
Cottage
C.
McGregor
M.
Din
S.
Fischer
K.
Gude
N.
Avitabile
D.
, et al. 
Pim-1 kinase protects mitochondrial integrity in cardiomyocytes
Circ. Res.
2010
, vol. 
106
 (pg. 
1265
-
1274
)
[PubMed]
38
Perelman
A.
Wachtel
C.
Cohen
M.
Haupt
S.
Shapiro
H.
Tzur
A.
JC-1: alternative excitation wavelengths facilitate mitochondrial membrane potential cytometry
Cell Death Dis.
2012
, vol. 
3
 pg. 
e430
 
[PubMed]