Molecular Cancer Research
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Molecular Cancer Research 5, 47-60, January 1, 2007. Published Online First January 8, 2007;
doi: 10.1158/1541-7786.MCR-06-0233
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Supplementary Data
Right arrow All Versions of this Article:
1541-7786.MCR-06-0233v1
5/1/47    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by López-Sánchez, N.
Right arrow Articles by Frade, J. M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by López-Sánchez, N.
Right arrow Articles by Frade, J. M.

DNA Damage and Cellular Stress Responses

Mitochondrial c-Jun NH2-Terminal Kinase Prevents the Accumulation of Reactive Oxygen Species and Reduces Necrotic Damage in Neural Tumor Cells that Lack Trophic Support

Noelia López-Sánchez1, José-Rodrigo Rodríguez2 and José María Frade1

Departments of 1 Developmental Neurobiology and 2 Neuroanatomy and Cellular Biology, Cajal Institute, Consejo Superior de Investigaciones Cientificas, Madrid, Spain

Requests for reprints: José M. Frade, Cajal Institute, Consejo Superior de Investigaciones Cientificas, Avda Doctor Arce 37, E-28002 Madrid, Spain. Phone: 34-91-5854740; Fax: 34-91-5854754. E-mail: frade{at}cajal.csic.es


   Abstract
Top
 Notes
 Abstract
Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
In response to different stress signals, the c-Jun NH2-terminal kinase (JNK) can trigger cell death. However, JNK also facilitates the survival and cell cycle progression of tumor cells by mechanisms that are poorly defined. Here, we show that schwannoma RN22 cells can survive and proliferate under serum-free conditions although serum withdrawal rapidly induces mitochondrial fission and swelling. Although the morphologic changes observed in the mitochondria did not trigger cytochrome c release, they were accompanied by an increase in the mitochondrial membrane potential ({Delta}{Psi}M) and of immunoreactivity for active JNK in these organelles. Pharmacologic inhibition of JNK provoked a further increase of the {Delta}{Psi}M, an increase in reactive oxygen species (ROS) production, and a sustained decrease in cell viability due to necrosis. This increase in necrosis was prevented by the presence of ROS scavengers. Immunoreactivity for active JNK was also observed in the mitochondria of neuroblastoma 1E-115 and neuroblastoma 2a neuroblastoma cell lines on serum withdrawal, whereas active JNK was barely detected in serum-deprived fibroblasts. Accordingly, the reduction in neural tumor cell viability induced by JNK inhibition was largely attenuated in serum-deprived fibroblasts. These data indicate that local activation of JNK in the mitochondria can protect against necrotic cell death associated with ROS production, facilitating the growth of neural tumor cells subjected to serum deprivation. (Mol Cancer Res 2007;5(1):47–60)


   Introduction
Top
 Notes
 Abstract
 Introduction
Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
Tumor progression is normally associated with high rates of proliferation even in the absence of trophic support. Nevertheless, because eukaryotic cells have developed mechanisms to induce cell death when cell proliferation occurs in an uncontrolled manner, high rates of proliferation alone are insufficient for tumors to develop. Indeed, simultaneous induction of proliferation together with a blockage of the apoptosis associated with uncontrolled cell growth is thought to be a crucial step in tumor cell transformation (1). The blockage of apoptosis in cells destined to die can activate alternative pathways leading to cell death, normally involving necrosis (2). Therefore, tumor cells could also set in motion additional molecular mechanisms to prevent the secondary necrosis derived from abortive apoptosis.

The c-Jun NH2-terminal kinase (JNK) was initially described as a stress-activated kinase able to phosphorylate the NH2-terminal transactivation domain of the transcription factor c-Jun in response to UV light (3). JNK belongs to a family of mitogen-activated protein kinases composed of three members (JNK1, JNK2, and JNK3; ref. 4). Apart from c-Jun, JNK has been reported to phosphorylate other transcription factors, such as activating transcription factor 2, E-26–like protein-1, p53, and c-Myc (4), as well as members of the B-cell leukemia/lymphoma 2 (Bcl-2) family of apoptosis regulators (5). JNK has mainly been studied in the context of cell stress and apoptotic cell death after heat shock, the formation of free radicals, and DNA damage (6-8). Meanwhile, expanding on the role of JNKs as effectors of harmful stimuli, it has also been shown that JNK can favor cell viability. In this context, JNK activity is related to tumor cell expansion (9-11), although it is presently unclear whether this is due to stimulation of tumor cell survival or to a direct effect on the cell cycle machinery.

Although JNK activation normally occurs in the nucleus, it can also be activated in the mitochondria, usually in association with the increased production of reactive oxygen species (ROS). Active JNK has been shown to promote ROS detoxification and to confer tolerance to oxidative stress (12, 13). Hence, this kinase may be involved in preventing ROS-derived secondary necrosis in tumor cells, facilitating tumor cell progression.

ROS are mainly generated by the mitochondrial respiratory chain in eukaryotic cells, this being composed of four protein complexes that mediate the transfer of electrons from NADH-flavin adenine dinucleotide to molecular oxygen. Electron transport along the respiratory chain generates a mitochondrial membrane potential ({Delta}{Psi}M), as protons are pumped out of the matrix across the inner membrane. Indeed, this chemical gradient is the basis for oxidative phosphorylation through the F0F1-ATP synthase complex (respiratory complex V). Electrons may escape from the transport chain, mainly through complexes I (NADH-ubiquinone oxidoreductase) and III (ubiquinol-cytochrome c oxidoreductase), and these electrons can react with molecular oxygen to produce oxygen radicals (superoxide anion). Normally, these oxygen radicals are converted into other ROS, such as hydrogen peroxide or hydroxyl radicals before they are eliminated from the cell.

Mitochondria are dynamic organelles whose number and morphology can change within a cell during development, the cell cycle, or when challenged by toxins or stress (14, 15). In healthy cells, these organelles appear as a network of interconnected tubular mitochondria. However, under conditions that compromise mitochondrial function, a decrease in connectivity and the formation of short, round mitochondria occurs due to changes in the rates of fission or fusion. Mitochondrial fission and fusion is regulated by proteins such as dynamin-related protein 1/dynamin 1, a GTPase from the dynamin family crucial for mitochondrial fission, or by mitofusin 1/fuzzy onions 1p, a large GTPase containing COOH-terminal coiled-coil domain required for mitochondrial fusion (16). These proteins are recruited dynamically to the outer mitochondrial membrane as the mitochondria modify their morphology. Time-lapse experiments in yeast mitochondria have revealed that the frequency of fusion and fission events is equivalent, thereby maintaining mitochondrial number and morphology (17). Genetic inactivation of the fusion step leads to fragmentation and genetic inactivation of fission results in an increase in mitochondrial connectivity (16). Fission or fusion rates may change under different growth conditions, leading to an increase or decrease in mitochondrial numbers. Moreover, mitochondrial fission has also been linked to the apoptotic process, which seems to require the fission of these organelles before the release of cytochrome c and the loss of {Delta}{Psi}M (18).

In this study, we present evidence that the activation of JNK in the mitochondria can prevent the necrotic cell death associated with ROS production, and facilitate the survival and growth of RN22 schwannoma and neuroblastoma cells subject to the withdrawal of trophic factors.


   Results
Top
 Notes
 Abstract
 Introduction
 Results
Discussion
 Materials and Methods
 Acknowledgements
 References
 
RN22 Schwannoma Cells Are Viable and They Proliferate in the Absence of Trophic Support
To study the survival and growth of neural tumor cells deprived of trophic support, we focused our attention on the rat RN22 schwannoma cell line, a cell line derived from the rat nerve-root tumor A204 (19) that grows without any evident loss of viability in serum-free conditions (20-22). To verify that RN22 schwannoma cells proliferate in the absence of serum, we followed their growth in the presence or absence of serum. Accordingly, we found that the duplication time of serum-deprived RN22 schwannoma cells (88 h) was nearly 4-fold that in the presence of serum (24 h; see Supplementary Fig. S1). Thus, while RN22 schwannoma cells are capable of proliferating in the absence of mitogens, their rate of division is lower that that of cells maintained in medium containing serum.

Serum Withdrawal Triggers Morphologic Changes in the Mitochondria of RN22 Schwannoma Cells
Mitochondria are dynamic organelles whose morphology changes in response to stressful situations (14, 15). To analyze whether mitochondrial morphology is modified in RN22 schwannoma cells in the absence of serum, these cells were labeled with the mitochondrial-specific dye Mitotracker Red CMXRos (Invitrogen, Carlsbad, CA). As previously described in healthy cells, the mitochondria from RN22 schwannoma cells cultured in the presence of serum displayed a distinctive reticular arrangement (Fig. 1A ). However, serum withdrawal resulted in a high degree of mitochondrial fragmentation, which was evident after 18 h (Fig. 1A). At this time point, different degrees of mitochondrial fragmentation were visible in most RN22 schwannoma cells (Fig. 1A) and after 40 h, the fragmented mitochondria showed signs of swelling (Fig. 1A). Interestingly, despite the existence of mitochondrial swelling, no signs of cytochrome c release could be detected in the majority of cells (see below and Supplementary Fig. S2). These morphologic changes in the mitochondria observed with Mitotracker were confirmed by electron microscopy (Fig. 1B). The morphology of mitochondria from RN22 schwannoma cells cultured in the presence of serum was that of a sectioned reticular arrangement, indicative of the presence of long mitochondria without evidence of swelling. In contrast, mitochondrial fragmentation and swelling was evident after 32 h in the absence serum whereas other organelles displayed no morphologic changes (e.g., the endoplasmic reticulum or nucleus). In accordance with the viability of RN22 schwannoma cells grown in the absence of serum, the morphologic changes observed in the mitochondria were fully reversible upon serum addition (Fig. 1C).


Figure 1
View larger version (136K):
[in this window]
[in a new window]

  		Figure 1. RN22 schwannoma cells display reversible fragmentation and swelling of their mitochondria in response to serum withdrawal. A. RN22 schwannoma cells were maintained in medium containing serum (+FCS) or serum-depleted medium (–FCS) for 18 or 40 h before labeling the cells with the mitochondria-specific dye Mitotracker (Mit.). Serum-depleted cells showed fragmented mitochondria after 18 h, which subsequently underwent swelling (40 h). Bar, 5 µm. B. Electron microscopy analysis confirmed that serum-deprived RN22 schwannoma cells grown in the absence of serum for 32 h contained fragmented and swollen mitochondria (arrows), as well as less healthy mitochondria (arrowheads). Observe how the endoplasmic reticulum (r) and nucleus (n) were unaltered in the absence of serum. C. The addition of serum to RN22 schwannoma cells previously grown for 48 h in the absence of serum was able to induce recovery of the mitochondrial network, as seen using the Mitotracker staining. Representative images. Bar, 10 µm.

 
Active JNK in the Mitochondria of Serum-Deprived RN22 Schwannoma Cells
Because serum withdrawal appears to trigger physiologic alterations in the mitochondria of RN22 schwannoma cells, it is possible that these organelles receive stress signals under such conditions. JNK is a kinase known to be activated in response to stress and its activation classically results in the translocation of JNK to the nucleus. However, JNK has also been seen to translocate to the mitochondria in response to stress signals (23-25). To analyze whether mitochondrial fission correlates with JNK translocation to these organelles, we investigated the distribution of JNK by subcellular fractionation. RN22 schwannoma cells cultured in the presence or absence of serum were lysed, and the fractions enriched in mitochondria were obtained by sequential centrifugation (heavy membrane fractions). When these fractions were Western blotted and probed with a JNK-specific antibody, it was clear that JNK accumulated in the mitochondrial extracts derived from serum-deprived RN22 schwannoma cells (see Supplementary Fig. S3A). Because mitochondria appear to be subject to stress and to undergo morphologic changes when RN22 schwannoma cells are deprived of serum, we assessed whether the mitochondrial pool of JNK had been activated under these conditions. RN22 schwannoma cells cultured in the absence of serum were immunostained with antisera specific for either JNK or for the active form of JNK (phosphorylated on Thr183/Tyr185). Whereas JNK was distributed throughout the nucleus and cytoplasm in RN22 schwannoma cells (Fig. 2, top ), the active form of JNK was restricted to the cytoplasmic compartment. Indeed, double labeling with the nuclear marker bisbenzimide confirmed that active JNK was excluded from the nuclei of serum-deprived RN22 schwannoma cells (Fig. 2, top). Double labeling with FITC-streptavidin (Fig. 2, top), a mitochondrial marker (26), or with the anti-ATP synthase subunit ß (not shown), showed that active JNK was present in the mitochondria. To confirm biochemically that activated JNK was present in the mitochondria of serum-deprived RN22 schwannoma cells, heavy membrane fractions enriched in mitochondria and soluble fractions were obtained from these cells. These fractions were probed with the antisera specific for either JNK or the active form of JNK in Western blots. This analysis showed that the active form of JNK was present in the mitochondria-enriched fraction, while it was barely detectable in the soluble fraction (see Supplementary Fig. S3B).


Figure 2
View larger version (54K):
[in this window]
[in a new window]

  		Figure 2. Active JNK can be detected in the mitochondria of RN22 schwannoma cells and N1E-115 neuroblastoma cells, but its presence is strongly reduced in the mitochondria of mouse embryonic fibroblasts grown in the absence of serum. Immunoreactivity for JNK (JNK) was detected in the cytoplasm and nucleus of RN22 schwannoma cells cultured in the absence of serum for 18 h (RN22). In contrast, immunostaining for the active form of JNK (P-JNK) was restricted to the cytoplasm and colocalized with streptavidin labeling (Strep), a mitochondrial marker. In N1E-115 neuroblastoma cells (N1E-115) deprived of serum for 18 h, JNK was mostly detected in the cytoplasm, and the active form of JNK was mainly cytoplasmic, colocalizing with the mitochondrial marker streptavidin. Note that, in these cells, streptavidin also labeled the nuclear compartment as previously shown in other cellular systems (see text). JNK immunolabeling was weak in mouse embryonic fibroblasts cultured in the absence of serum for 18 h, and the active form of JNK was barely detected in these cells. Large panels, merged pictures, including nuclei labeled with bisbenzimide (blue). Representative images. Bar, 5 µm.

 
JNK Activity Is Required for the Survival of RN22 Schwannoma Cells Cultured in the Absence of Serum
In different cell systems, JNK activity has been shown to be necessary for survival (27-29). Therefore, we determined whether JNK activity was required for RN22 schwannoma cells to survive in the absence of serum. We inhibited the activity of JNK using the pharmacologic inhibitor of JNK SP600125. This molecule interacts with the ATP-binding pocket of JNK, impeding its interaction with ATP and inhibiting its function (30). SP600125 (9 µmol/L) was able to prevent the apoptosis of RN22 schwannoma cells in response to UV radiation (see Supplementary Fig. S4), indicating that this drug was indeed able to prevent a known function of JNK, that of a stress kinase that triggers apoptosis in response to UV light. The viability of RN22 schwannoma cells grown in the absence of serum was analyzed using the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay in the presence of different concentrations of SP600125. A dose-dependent reduction in cell viability of these cells was observed in the presence of SP600125 (Fig. 3A ). When the time course of this effect was followed, a steady decrease in viability was observed from 2 h, which reached a maximum after a 12-h treatment (Fig. 3B). The participation of JNK in the survival of serum-deprived RN22 schwannoma cells was confirmed by exposing the cells to AS601245, a novel JNK inhibitor that, like SP600125, also interacts with the ATP-binding pocket of JNK (31). This compound reduced the viability of RN22 schwannoma cells to a similar extent as that observed with SP600125 (Supplementary Fig. S5A). Moreover, the viability of serum-deprived RN22 schwannoma cells was significantly reduced when they were exposed to JNK inhibitor III, a cell-permeable peptide capable of preventing the interaction of JNK with its substrates (32). The elevated molecular weight (4,413.3 g/mol) and low solubility (1 µg/µL) of this peptide impedes its use at high concentrations without compromising cell viability due to intrinsic toxicity of the vehicle (dimethyl formamide). However, we detected a reduction in survival of ~25% with the MTT assay when this peptide was used at 10 µmol/L (see Supplementary Fig. S5B).


Figure 3
View larger version (9K):
[in this window]
[in a new window]

  		Figure 3. Inhibition of JNK activity specifically reduces cell viability in RN22 schwannoma and N1E-115 neuroblastoma cells deprived of serum. A. The presence of SP600125 (JNK Inh.) reduced the viability of serum-deprived RN22 schwannoma cells in a dose-dependent manner, as quantified by the MTT reduction assay. At 9 µmol/L, the effect of SP600125 reached a maximum. B. Time course of the effect of 9 µmol/L SP600125 on the viability of serum-deprived RN22 schwannoma cells. Note how cell viability gradually decreases. C. SP600125 reduces the viability of N1E-115 neuroblastoma cells in a dose-dependent manner. D. SP600125-dependent reduction of viability was largely attenuated in serum-deprived fibroblasts treated with JNK inhibitors, when compared with RN22 schwannoma and N1E-115 neuroblastoma cells. Points, mean (n = 4); bars, SE.

 
Effect of JNK on Cell Viability Can Be Extended to Other Neural Tumor Cell Lines
To study whether JNK is also activated in other neural tumor cells in response to serum withdrawal and whether this inhibition results in a similar loss of viability, we examined two neuroblastoma cell lines, neuroblastoma 1E-115 (N1E-115; Fig. 2B) and neuroblastoma 2a (N2a; not shown) cells. JNK was distributed throughout the nucleus and cytoplasm in both cell lines (Fig. 2, middle) and like RN22 schwannoma cells, active JNK colocalized with the mitochondrial marker streptavidin (Fig. 2, middle). However, in N1E-115 cells, streptavidin labeling also resulted in additional nuclear background, as previously described in normal rat kidney cells (26). Unlike the neural tumor cell lines, mouse embryonic fibroblasts contained less active JNK (Fig. 2, bottom) despite the fact that they displayed JNK immunoreactivity throughout their cytoplasm (Fig. 2, bottom). As expected, exposure of N1E-115 (Fig. 3C) and N2a (not shown) neuroblastoma cells to SP600125 resulted in a loss of viability similar to that observed in RN22 schwannoma cells. Similarly, exposure of mouse embryonic fibroblasts deprived of serum to the JNK inhibitor SP600125 produce a loss of viability, although to a lower extent than that observed in neural tumor cells (Fig. 3D). These data, together with the decrease in active JNK in their mitochondria, suggests that serum-deprived fibroblasts are less dependent on JNK for their survival than neural tumor cells.

JNK Activity Does Not Regulate Mitochondrial Fragmentation in Serum-Deprived RN22 Schwannoma Cells
Because JNK appears to be active in the mitochondria of serum-deprived RN22 schwannoma cells, it is possible that this molecule could regulate the morphologic changes observed in the mitochondria of these cells. To test this hypothesis, RN22 schwannoma cells were exposed to SP600125 at the time of serum withdrawal and the cells were maintained in the presence of this inhibitor for 18 or 24 h before evaluating their mitochondrial morphology. This analysis showed that the inhibition of JNK activity did not affect the changes in mitochondrial morphology that occur upon serum withdrawal (see Supplementary Fig. S6). Thus, JNK does not seem to participate in mitochondrial fragmentation.

The Loss of Viability of Serum-Deprived RN22 Schwannoma Cells in Response to JNK Inhibitors Involves Necrosis
To decipher the mechanism underlying the loss of viability triggered by JNK inhibitors in serum-deprived RN22 schwannoma cells, we focused on the possible participation of apoptosis in this effect. Because the mitochondrial network undergoes fission and the individual mitochondria show signs of swelling in the absence of serum, we hypothesized that prevention of JNK activity could facilitate the release of cytochrome c from the mitochondria, thus inducing apoptosis. To test this possibility, we examined the distribution of cytochrome c in RN22 schwannoma cells serum-deprived for 18 h and exposed to SP600125 or vehicle alone for an additional period of 24 h. Using a specific antibody, we found cytochrome c in the mitochondria of most control and SP600125-treated RN22 schwannoma cells (Fig. 4A ), in accordance with the results shown in Supplementary Fig. S2. Diffuse cytoplasmic staining of cytochrome c was only observed in 14.89 ± 3.49% (n = 3) of serum-deprived RN22 schwannoma cells and this percentage did not significantly increase upon exposure to SP600125 (12.88 ± 2.11%; n = 3; not significant). The lack of cytochrome c release on exposure to JNK inhibitors suggested that apoptosis is not involved in the loss of viability observed under these conditions. To test this hypothesis, SP600125 was used to block JNK activity in serum-deprived RN22 schwannoma cells treated with the general caspase inhibitor z-VAD. The effects of these compounds were compared with staurosporine-dependent cell death in these cells. Whereas the presence of z-VAD rescued serum-deprived RN22 schwannoma cells from the effects of staurosporine on cell viability, it was not able to prevent the loss of viability in response to SP600125, indicating that caspases were not involved in this SP600125-induced cell death (Fig. 4B). Accordingly, we found that the proportion of terminal deoxyribonucleotide transferase–mediated nick-end labeling (TUNEL)–labeled nuclei in RN22 schwannoma cells cultured in the absence of serum did not significantly change upon treatment with the JNK inhibitor SP600125 (Fig. 4C).


Figure 4
View larger version (29K):
[in this window]
[in a new window]

  		Figure 4. The decrease in cell viability in serum-deprived RN22 schwannoma treated with the JNK inhibitor SP600125 is independent of apoptosis. A. Most serum-deprived RN22 schwannoma cells showed nondiffuse cytochrome c immunolabeling after 24 h in either the presence or absence (Vehicle) of the JNK inhibitor SP600125. Representative images. Bar, 5 µm. B. The presence of the caspase inhibitor z-VAD did not prevent the reduction in the viability of serum-deprived RN22 schwannoma cells in response to the SP600125, as measured by the MTT assay. z-VAD was fully active in our hands as it fully protected these cells from the apoptotic effects of staurosporine (0.5 µmol/L). Columns, mean (n = 4); bars, SE. ***, P < 0.005. C. The proportion of serum-deprived RN22 schwannoma cells positive for TUNEL staining was not significantly affected by treatment with SP600125 at the time points indicated. Columns, mean (n = 3); bars, SE. N.S., nonsignificant.

 
There are other components of the mitochondrial intermembrane space that may be released when caspase-independent apoptotic cell death is triggered, such as apoptosis-inducing factor (AIF; ref. 33). To verify whether this protein is released from the mitochondria and translocated to the nuclear compartment upon inhibition of JNK activity, we did immunostaining with an AIF-specific antibody on serum-deprived RN22 schwannoma cells treated with or without SP600125 (Fig. 5A ). AIF was not detected in the nucleus of cells exposed to the JNK inhibitor, indicating that the loss of viability in RN22 schwannoma cells lacking JNK activity was not induced by caspase-independent apoptosis.


Figure 5
View larger version (63K):
[in this window]
[in a new window]

  		Figure 5. The decrease in cell viability in serum-deprived RN22 schwannoma treated with the JNK inhibitor SP600125 is not due to caspase-independent apoptosis or autophagy. A. Serum-deprived RN22 schwannoma cells were treated for 24 h with vehicle or SP600125, fixed, and subjected to immunocytochemistry for AIF. The cell nuclei were visualized by counterstaining with bisbenzimide (Bisbenz.). Note that AIF is not translocated to the nucleus in the presence of SP600125. Bar, 20 µm. B. RN22 schwannoma cells were transfected with a vector expressing LC3-GFP and then maintained in the presence or absence of serum. Treatment of RN22 schwannoma cells grown in the presence of serum with hydroxychloroquine (+FCS/H-ClQ) resulted in the appearance of cytoplasmic condensations, representing LC3-enriched autophagosomes. Serum starvation resulted in an increase of small LC3-labeled autophagosomes that were unaffected by the presence of SP600125 (–FCS/JNK Inh). Representative images. Bar, 5 µm.

 
An alternative mechanism of programmed cell death that is triggered by cell starvation is based on autophagy (34, 35). A hallmark of autophagy is the existence of vacuoles that include the marker protein, microtubule-associated protein light chain 3 (LC3; ref. 36). To determine whether serum-deprived RN22 schwannoma cells treated with the JNK inhibitor SP600125 suffered autophagy, we transfected these cells with a plasmid expressing a LC3-green fluorescent protein (GFP) fusion protein. Although RN22 schwannoma cells cultured in the presence of serum showed only a diffuse pattern for LC3-GFP (Fig. 5B), the induction of autophagy through exposure to hydroxychloroquine was reflected in the appearance of LC3-positive vesicles in these cells (Fig. 5B; ref. 37). Serum starvation resulted in the accumulation of fewer LC3-GFP aggregates (Fig. 5B), indicating that the absence of serum produced little autophagy. Moreover, the presence of these vesicles did not increase upon exposure to SP600125 (Fig. 5B).

Because regulated cell death does not seem to be involved in the loss of viability observed in serum-deprived RN22 schwannoma cells treated with the JNK inhibitor SP600125, we tested whether necrosis was involved in this process. Serum-deprived RN22 schwannoma cells were labeled with propidium iodide and because an increase in apoptosis does not occur in response to SP600125, nuclei labeled with propidium iodide are not derived from late apoptotic cells whose membrane disintegrates. A steady increase of propidium iodide–positive nuclei was observed in the cultures of serum-deprived RN22 schwannoma cells treated with SP600125 (Fig. 6A ). This increase in propidium iodide–labeled nuclei was coupled with a net decrease in the absolute number of cells in the SP600125 cultures (Fig. 6B). These results indicate that necrosis occurs in serum-deprived RN22 schwannoma cells treated with SP600125. To further confirm this conclusion, electron microscopy was done in the SP600125-treated RN22 schwannoma cells. Accordingly, features of necrosis were readily observed in these cells such as the absence of nuclear condensation and release of the cytoplasmic content (Fig. 6C).


Figure 6
View larger version (72K):
[in this window]
[in a new window]

  		Figure 6. JNK inhibition in serum-deprived RN22 cells results in necrosis. A. Serum-deprived RN22 schwannoma cells were cultured in the presence (JNK Inh) or absence (Vehicle) of the JNK inhibitor SP600125, and necrosis was monitored by nuclear labeling of intact cells with propidium iodide. Note that the presence of SP600125 significantly increased necrotic cell death after 24 h of culture. Points, mean (n = 3); bars, SE. *, P < 0.05; **, P < 0.01; ***, P < 0.005. B. In accordance with its capacity to induce necrosis, SP600125 triggered a significant decrease in the number of serum-deprived RN22 schwannoma cells attached to the culture dishes when compared with the controls (Vehicle). Points, mean (n = 3); bars, SE. *, P < 0.05; **, P < 0.01; ***, P < 0.005. C. Electron microscopy confirmed that the diminished viability of serum-deprived RN22 schwannoma cells treated with SB600125 was due to necrosis. Note that most cytoplasmic components are absent and the mitochondria (arrowheads) are swollen. mb, cell membrane; n, nucleus. Bottom (C), a high-power image from the top panel. Representative images.

 
JNK Prevents ROS Accumulation and Necrosis in RN22 Schwannoma Cells
To analyze the molecular events that trigger necrosis in serum-deprived RN22 schwannoma cells exposed to SP600125, we initially assessed the changes in the {Delta}{Psi}M in these cells. We hypothesized that JNK inhibition might compromise {Delta}{Psi}M, thereby triggering an energy deficit that would cause necrosis in these cells. The {Delta}{Psi}M can be measured with membrane-permeable lipophilic cationic fluorochromes such as rhodamine 123 (38-40). Living cells accumulate this probe in their mitochondria, and the extent of uptake, measured by the intensity of cellular fluorescence, reflects the {Delta}{Psi}M. Contrary to our expectations, serum withdrawal resulted in a 55% increase in the intensity of rhodamine 123 fluorescence that further incremented by 92% in response to JNK inhibition (Fig. 7A and B ).


Figure 7
View larger version (85K):
[in this window]
[in a new window]

  		Figure 7. JNK inhibition of serum-deprived RN22 cells increases the {Delta}{Psi}M and ROS accumulation. A. RN22 schwannoma cells were cultured in the presence or the absence of serum for 18 h, and then incubated with rhodamine 123 (Rhod-123), a membrane-permeable lipophilic cationic fluorochrome that accumulates in the mitochondria and reflects the {Delta}{Psi}M value. Note how relatively higher levels of fluorescence can be observed in RN22 schwannoma cells deprived of serum. Bar, 10 µm. Representative images. B. When rhodamine 123 fluorescence was quantified per cell (in arbitrary units), serum withdrawal produced a significant increment in the intensity of fluorescence with respect to RN22 schwannoma cells maintained in the presence of serum. The rhodamine 123 fluorescence further increased in response to JNK inhibition with SP600125. Columns, mean (n = 5); bars, SE. ***, P < 0.001. C. RN22 schwannoma cells were cultured in the presence or the absence of serum for 10 h and they were then incubated in the presence of carboxy-H2DCF-DA (H2DCFDA), a cell-permeable ROS indicator that fluoresces when oxidation occurs within the cell. Serum withdrawal did not appear to produce an increase in the intensity of carboxy-H2DCF-DA, but treatment with SP600125 induced ROS production that could be detected with carboxy-H2DCF-D. Bottom, phase contrast pictures of the same fields in the top panels. Bar, 10 µm. Representative images.

 
ROS are by-products of the respiratory electron transport chain and, above a defined threshold, their production has been shown to be an exponential function of {Delta}{Psi}M (41, 42). Therefore, we hypothesized that the substantial increase in {Delta}{Psi}M observed in serum-deprived RN22 schwannoma cells in response to JNK inhibition might produce elevated levels of ROS. Carboxy-2',7'-dichlorofluorescein diacetate (carboxy-H2DCF-DA) is a cell-permeable indicator of ROS that fluoresces when oxidation occurs within the cell. We used this compound to measure the accumulation of ROS in serum-deprived RN22 schwannoma cells treated with SP600125. Despite the increase in {Delta}{Psi}M, ROS could not be detected in RN22 schwannoma cells grown in serum-free medium (Fig. 7C). Nevertheless, a fluorescent signal was readily detected in these cells upon inhibition of JNK, indicating that ROS accumulated in RN22 schwannoma cells treated with the JNK inhibitor SP600125 (Fig. 7C).

Because the accumulation of ROS can trigger necrosis (43), ROS accumulation in RN22 schwannoma cells could be the cause of the necrosis observed in response to SP600125 treatment. We attempted to reduce or even block necrosis in serum-deprived RN22 schwannoma cells treated with SP600125 by adding antioxidant compounds that would impair ROS accumulation. Metalloporphyrins are catalytic antioxidants that scavenge a wide range of ROS such as superoxide, peroxide, peroxynitrite, and lipid peroxyl radicals (44). By using the metalloporphyrin Fe(III) 5,10,15,20-tetrakis-4-carboxyphenyl porphyrin (45), we were able to impair the reduction in viability of RN22 schwannoma cells in a dose-dependent manner (Fig. 8 ). This indicates that necrotic cell death caused by JNK inhibition in serum-deprived RN22 schwannoma cells is largely based on oxidative stress.


Figure 8
View larger version (11K):
[in this window]
[in a new window]

  		Figure 8. The ROS scavenger Fe(III) 5,10,15,20-tetrakis-4-carboxyphenyl porphyrin (FeTCP) reverses the loss of viability triggered by JNK inhibition in serum-deprived RN22 cells. Serum-deprived RN22 schwannoma cells were treated for 24 h with vehicle or 9 µmol/L SP600125, together with different concentrations of the ROS scavenger Fe(III) 5,10,15,20-tetrakis-4-carboxyphenyl porphyrin. This treatment completely blocked the reduction in viability of RN22 schwannoma cells measured with MTT, in a dose-dependent manner, indicating that the cause of cell death is oxidative stress. Columns, mean (n = 4); bars, SE.

 

   Discussion
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
Materials and Methods
 Acknowledgements
 References
 
We have shown that serum withdrawal does not compromise the proliferative capacity of RN22 schwannoma cells. Nevertheless, in the absence of serum, their mitochondria undergo morphologic and physiologic changes that include fission and swelling as well as a moderate increase in the {Delta}{Psi}M. In most cells, these changes were not translated into the release of cytochrome c from the mitochondria or the activation of the classic apoptotic machinery. The structural and functional changes in the mitochondria correlated with the presence of active JNK in these organelles, both in RN22 schwannoma and in other tumor cell lines of neural origin such as N2a and N1E-115 neuroblastoma cells. The inhibition of JNK in the presence of specific inhibitors resulted in reduced viability of these neural tumor cells without affecting the morphologic changes in the mitochondria. This effect was independent of the disruption of the outer mitochondrial membrane and the release of cytochrome c or AIF, indicating that it was not linked to caspase-dependent or caspase-independent apoptotic cell death. Autophagy was also ruled out as the cause of the impaired viability because the density of autophagic vacuoles enriched with LC3 did not increase in response to JNK inhibition. Despite the apparent absence of morphologic effects in mitochondria upon JNK inhibition, SP600125 altered the function of these organelles. Indeed, this inhibitor induced a further increase of {Delta}{Psi}M, which was translated into the accumulation of ROS and necrosis in RN22 schwannoma cells. Finally, we have shown that ROS scavenging prevents necrosis, indicating that the loss of viability is directly dependent on the excess accumulation of ROS upon JNK inhibition. Overall, these results indicate that the presence of active JNK in the mitochondria can prevent necrotic cell death associated with the production of ROS and facilitate the survival and growth of neural tumor cells deprived of trophic support.

Trophic Factors, Mitochondrial Morphology, and Cell Death
In most eukaryotic cells, deprivation of trophic factors is normally associated with cell cycle withdrawal. However, some cells still maintain their proliferative capacity in the absence of trophic factors and this, together with a reduction in apoptosis, is a platform both necessary and sufficient for tumor transformation (1). We have shown that serum-deprived RN22 schwannoma cells can proliferate and most of them were viable in the absence of trophic support. Accordingly, the proportion of TUNEL-labeled RN22 schwannoma cells grown in the absence of serum remained at a minimum and cytochrome c release was only detected in a minority of the cells.

Despite the proliferative capacity observed in serum-deprived RN22 schwannoma cells, they exhibited extensive changes in mitochondrial morphology. These morphologic alterations of the mitochondria are reminiscent of the response to different stress signals observed in other cellular systems. As such, in hyperglycemic conditions, mitochondrial fragmentation is triggered (46) and heat shock can affect mitochondrial morphology and trigger swelling without affecting the ultrastructure of other organelles (47).

Mitochondrial swelling normally results in rupture of the outer mitochondrial membrane, causing the release of cytochrome c and other components of the mitochondrial intermembrane space that initiate an apoptotic response. In our cell system, mitochondrial swelling does not seem to produce the rupture of the outer mitochondrial membrane, thus explaining the maintenance of {Delta}{Psi}M and the absence of cytochrome c release in most cells. Accordingly, no rupture of the outer mitochondrial membrane was detected by electron microscopy in heavily swollen mitochondria. Mitochondrial swelling in the absence of cytochrome c release has been described in a number of cases, including CD47-dependent cell death of T cells (48), rat hepatocytes after partial hepatectomy (49), and human osteosarcoma cells treated with the protonophore m-chlorophenylhydrazone (50). Apart from these examples, it is known that the release of cytochrome c from the mitochondrial intermembrane space can be regulated biochemically. For instance, active mitochondrial calcium-independent phospholipase A2 can reduce the amount of cytochrome c released in response to mitochondrial matrix swelling (51), and estrogens prevent calcium-induced release of cytochrome c from heart mitochondria but not mitochondrial swelling (52). These studies indicate that cytochrome c release is a highly dynamic process that can be modulated by molecular mechanisms intrinsic to particular cell types. It is also worth noting that mitochondrial swelling in serum-deprived RN22 schwannoma cells is a delayed effect that is only observed 24 h after the withdrawal of trophic support. Over this period, it is conceivable that new lipids can be added to the original outer mitochondrial membrane, facilitating its swelling without rupture.

Mitochondrial Localization of Phosphorylated JNK
JNK is activated in a large proportion of primary brain tumors (53), but its function in these transformed cells remains unclear. In fact, the activation of JNK in neural tumor cells has been implicated in promoting both apoptosis and cell proliferation in vitro, depending on the cell line. Activation of JNK in the nuclear compartment is normally associated to c-Jun phosphorylation and apoptosis. However, JNK has also been seen to translocate to the mitochondria in different cell systems (23-25). In this work, we present evidence that JNK activity promotes the survival of tumor cells under conditions of trophic factor withdrawal.

In our cell system, active JNK was observed in the mitochondria but not in the nucleus, indicating that JNK acts selectively in the mitochondria. In most examples where active JNK is located in the mitochondria, this enzyme colocalizes with and inactivates Bcl-2 or Bcl-XL (23-25), facilitating the release of cytochrome c and apoptosis. We have shown that activation of JNK in the mitochondria is not related to the induction of apoptosis in our cell system, suggesting that the mitochondrial substrates of JNK in RN22 schwannoma cells are probably not Bcl-2 and Bcl-XL. Apart from these members of the Bcl-2 family, JNK may interact with sarcoma gene homology domain 3 (SH3)-binding protein 5, also known as Sab; a novel JNK-interacting protein able to bind to the SH3 domain of Bruton's tyrosine kinase. The Sab protein is located in the mitochondria of chick embryonic fibroblasts and, in response to anisomycin, it can interact with the active form of JNK in these organelles (54, 55). Sab is thought to play a role in anchoring JNK to the mitochondria although, alternatively, phosphorylation of Sab by JNK could also affect mitochondrial function.

JNK Inhibition, Oxidative Stress, and Necrotic Cell Death
We showed that while RN22 schwannoma cells and other neuroblastoma cell lines are viable and proliferate in the absence of trophic support, they die by necrosis when JNK activity is inhibited. The results obtained here were based on the use of SP600125, a well-known inhibitor of JNK. This compound is a reversible competitive ATP inhibitor with >20-fold selectivity for JNK when compared with a large range of enzymes, including the highly related kinases from the mitogen-activated protein kinase superfamily extracellular signal-regulated kinase and p38 (30). In our hands, SP600125 was able to prevent apoptosis in RN22 schwannoma cells after UV light exposure, a well-characterized effect triggered by JNK (3). The use of JNK inhibitor III, a cell-permeable peptide mimicking the JNK binding domain of c-Jun, has also been shown previously to prevent JNK activity (31). This inhibitor also significantly reduced the viability of serum-deprived RN22 schwannoma cells as defined by the MTT assay, further evidence that JNK seems to be involved in promoting the survival of these cells in the absence of serum. Finally, the ATP-competitive JNK inhibitor AS601245 (31) also compromises the viability of serum-deprived RN22 schwannoma cells.

A number of studies have shown a role for JNK in the prevention of cell death. Thus, inhibition of JNK results in S-phase inhibition and cell death of glioblastoma T98G cells (27), and activation of JNK in cardiac myocytes can promote survival in response to oxidative stress (28). JNK has been shown to be required for IL-3–mediated cell survival of hematopoietic pro-B-factor dependence of lymphoid 5.12 (FL5.12) cells, through Thr201 phosphorylation and the ensuing inactivation of the proapoptotic protein Bcl-2–associated death promoter homologue (29). Additional evidence of the antiapoptotic activity of JNK comes from the analysis of the double-knockout mice for Jnk1 and Jnk2 genes. Null mutations for both genes increase the apoptosis observed in the developing forebrain (56, 57).

Rhodamine 123 labeling was used as a specific marker for {Delta}{Psi}M to study alterations in this variable in response to serum deprivation and inhibition of JNK activity. The level of rhodamine 123–dependent fluorescence increased in RN22 schwannoma cells grown in the absence of serum. This probably reflects the increase of {Delta}{Psi}M in response to trophic deprivation, as observed in FL5.12A cells when deprived of Interleukin-3 (58). Alternatively, the apparent increase of rhodamine 123 uptake in serum-deprived RN22 schwannoma cells may reflect the changes in mitochondrial size and complexity observed in the absence of serum. Indeed, the accumulation of {Delta}{Psi}M-dependent dyes in mitochondria generally depends on their size and complexity (59). Apart from this initial rise of {Delta}{Psi}M in the absence of serum, SP600125 induced an additional increase in rhodamine 123 incorporation. Although this phenomenon may be due to further changes in mitochondrial size or complexity in serum-deprived RN22 schwannoma cells treated with SP600125, no obvious change in the mitochondrial morphology was detected under these conditions (Supplementary Fig. S6; Fig. 4A). Therefore, we conclude that JNK activity appears to be crucial for maintaining relative low levels of {Delta}{Psi}M in RN22 schwannoma cells cultured in the absence of serum.

The {Delta}{Psi}M is directly linked to ROS production and above a particular threshold, any small increase of {Delta}{Psi}M stimulates the production of ROS by mitochondria (41, 42). Thus, it is plausible that the mitochondria from serum-deprived RN22 schwannoma would dramatically increase their {Delta}{Psi}M value in response to JNK inhibition, increasing ROS production to a level that the cellular detoxification machinery could not eliminate. Although we were unable to detect ROS in serum-deprived RN22 schwannoma cells by labeling with carboxy-H2DCF-DA, ROS accumulation was clearly observed in these cells upon exposure to SP600125. This situation is somewhat reminiscent of that previously described, where overexpression of JNK in SK-OV-3 human ovarian adenocarcinoma cells reduces ROS production, an effect that was reversed by JNK inactivation with SP600125 (60).

ROS are potent activators of JNK through mechanisms that include the activation of apoptosis signal-regulating kinase 1 (ASK1; ref. 61); the activation of the Src pathway (62); the activation of the calcium-dependent, proline-rich tyrosine kinase 2, as well as the small GTP binding factors Ras-related C3 botulinum toxin substrate 1, and cell division cycle 42 (63); and the oxidative inactivation of the endogenous JNK inhibitor dual specificity phosphatase M3/6 (64). In RN22 schwannoma cells, JNK appears to be activated before ROS can be detected with carboxy-H2DCF-DA, suggesting that its function is to prevent ROS production and/or detoxification. Nevertheless, we cannot rule out that low levels of ROS that are undetectable with the carboxy-H2DCF-DA probe could be the cause of JNK activation in serum-deprived RN22 schwannoma cells.

At present, the mechanism used by JNK to prevent ROS accumulation in serum-deprived RN22 schwannoma cells is unclear. JNK could phosphorylate mitochondrial substrates leading to a reduction of the {Delta}{Psi}M. Accordingly, a number of phosphoproteins have been described in the mitochondria, several of which are components of the electron transfer chain (65, 66). Moreover, it has recently been suggested that different protein kinases could regulate mitochondrial physiology (67). Indeed, protein kinase A phosphorylation of two mitochondrial membrane proteins of 6 and 18 kDa reduces the formation of superoxide in mitochondrial complex I (68), the major site for ROS production. It is therefore possible that Ser/Thr protein kinases such as protein kinase A or JNK could regulate the phosphorylation of mitochondrial substrates that influence mitochondrial physiology, thereby preventing ROS formation.

JNK can prevent ROS production through an indirect mechanism based on the regulation of carbohydrate metabolism. In HeLa cells, acceleration of mitochondrial respiration in response to pyruvate increases the production of ROS and the subsequent activation of cytosolic JNK. In turn, activation of cytosolic JNK activates glycogen synthase, diminishing the levels of glucose available for mitochondrial respiration and leading to a reduction in ROS production (69). Hyperglycemic conditions can also induce mitochondrial fragmentation and ROS production in liver cells, myoblasts, and endothelial cells (46, 70). Interestingly, ROS production, but not mitochondrial fragmentation, could be prevented in liver cells and myoblasts after inhibition of mitochondrial pyruvate uptake (46). This suggests that enhanced respiration in fragmented mitochondria may be a source of oxidative components, as observed in serum-deprived RN22 schwannoma cells treated with JNK inhibitors.

Finally, a further nonexclusive possibility is that JNK could also participate in the elimination of ROS. JNK is a key element of a signal transduction network that coordinates the induction of protective genes in response to an oxidative challenge, thereby conferring tolerance to oxidative stress and extending the life span of Drosophila (13). Furthermore, Jun proteins are activated by JNK in response to oxidative stress, and they associate with nuclear factor (erythroid-derived 2)–related factor-1/2 to induce the expression of detoxifying and antioxidant genes (12).

Excess accumulation of ROS could provoke necrotic cell death in RN22 schwannoma cells due to the massive cellular damage associated with lipid peroxidation and alterations of proteins and nucleic acids (43). Although there is no evidence that apoptosis or autophagy reduce the viability of serum-deprived RN22 schwannoma cells treated with JNK inhibitors, there is evidence of necrosis in response to these inhibitors. On the one hand, we observed an increase in the capacity of serum-deprived RN22 schwannoma cells to rapidly incorporate propidium iodide upon treatment with SP600125, a classic indicator of cell membrane damage linked to necrotic cell death. This effect was accompanied by a sustained decrease in cell density in our cultures, indicative of the disintegration of the necrotic cells. On the other hand, necrotic morphology was observed by electron microscopy in serum-deprived RN22 schwannoma cells treated with the JNK inhibitor SP600125. High levels of ROS production in response to JNK inhibition are therefore likely to cause necrosis of RN22 serum-deprived schwannoma cells. Accordingly, a metalloporphyrin-derived compound that mimics the prosthetic group of superoxide dismutase and that acts as a superoxide scavenger (45) could prevent the loss of viability in a dose-dependent manner.

JNK Activation and Cancer
The resistance of tumor cells to cell death is one of the major obstacles to overcome in our struggle to combat cancer. We have shown that deprivation of trophic support activates JNK in the mitochondria of RN22 schwannoma cells and in N1E-115 and N2a neuroblastoma cells. Moreover, JNK inhibitors trigger toxic effects mostly in these cells. Tumor cells may acquire the capacity to activate JNK in their mitochondria to prevent ROS accumulation when active proliferation occurs in environments depleted of trophic support. This may be particularly relevant for proliferation in the core of solid tumors where there is little blood flow, or when metastatic cells pass through regions devoid of trophic factors. This survival mechanism may also be useful to tumor cells that are resistant to therapies that provoke ROS production through DNA damage. In some instances, JNK activity has been at least partially shown to be involved in such resistance due to its specific effects on the expression of antioxidant enzymes (71) or in mediating enhanced DNA repair (72). Indeed, inhibition of JNK with SP600125 has proved to be an efficient manner to reduce viability in these tumor cell lines (72). The data we present here suggest that specific inhibition of JNK activity in apoptosis-resistant neural tumor cells may be a suitable therapeutic approach to treat neural-derived tumors.


   Materials and Methods
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
Acknowledgements
 References
 
Antibodies
Both JNK and active JNK (phosphorylated in Thr183/Tyr185) were detected using the PhosphoPlus stress-activated protein kinase/JNK (Thr183/Tyr185) Antibody kit (Cell Signaling Technology, Danvers, MA) following the manufacturers' instructions. The antiserum against JNK was used at a dilution of 1/15,000 for Western blot and at a dilution of 1/200 for immunocytochemistry. Antiserum against active JNK was used at a dilution of 1/2,000 for immunocytochemistry and at a 1/15,000 for Western blots. Detection of AIF was done with the goat polyclonal D-20 antiserum (Santa Cruz Biotechnology, Santa Cruz, CA) diluted 1/100 for immunocytochemistry. The monoclonal antibody 3D5 that recognizes the ß-subunit of ATP synthase (Molecular Probes) was used at a dilution of 1/300 for immunocytochemistry and 1/10,000 for Western blotting. The anti–cytochrome c polyclonal Ab-2 antiserum (Lab Vision NeoMarkers, Fremont, CA) was used at a dilution of 1/300 for immunocytochemistry.

Other Reagents
SP600125, JNK inhibitor III, and AS601245 were purchased from Calbiochem and used at the concentrations indicated. The z-Val-Ala-Asp-fluoromethylketone (Calbiochem, Darmstadt, Germany) was used at 80 µmol/L and hydroxychloroquine (Sanofi-Synthelabo, Paris, France) was used at 30 µg/mL (6-h treatment). Staurosporine was purchased from ALEXIS (Lausanne, Switzerland) and used at 0.5 µg/mL (6-h treatment). Propidium iodide (Sigma, St. Louis, MO) was used at 4.6 µg/mL and bisbenzimide (Sigma) at 1 µg/mL. The porphyrin analogue Fe(III) 5,10,15,20-tetrakis-4-carboxyphenyl porphyrin (Frontier Scientific, Logan, UT) was dissolved in 0.1 mol/L NaOH (Merck, Darmstadt, Germany), neutralized with 10x 136.9 mmol/L NaCl (Merck), 2.7 mmol/L KCl (Merck), 7.9 mmol/L sodium phosphate (Merck; PBS) to prepare a stock solution at 50 mmol/L, and this was used at the concentrations specified. Carboxy-H2DCF-DA (Molecular Probes) was used at 10 µmol/L, rhodamine 123 (Molecular Probes) was used at 1 µg/mL, and Mitotracker (Molecular Probes) was used at 90 µg/mL.

Cell Culture Conditions
RN22 schwannoma cells and N2a and N1E-115 neuroblastoma cells were all maintained at 37°C in DMEM containing 10% FCS (Life Technologies, Carlsbad, CA). For serum deprivation, cells were plated in DMEM alone and cultured for 18 h before treating them with the chemicals described. To induce apoptosis, serum-deprived RN22 schwannoma cells were maintained either in the presence or absence of SP600125 for 25 min and the cells were then UV irradiated (270 mJ/cm2) using a Stratalinker UV cross-linker. UV-treated RN22 schwannoma cells were then maintained for an additional period of 30 min at 37°C, and fixed for 15 min in 4% paraformaldehyde (Merck). Pyknotic nuclei were identified after staining with 1 µg/mL bisbenzimide (Sigma). Primary mouse embryonic fibroblast cultures were established from E14 mouse embryos, which were first eviscerated and then their head, limbs, and tails were removed. The remaining tissue was minced and trypsinized with trypsin-EDTA (Life Technologies) for 30 to 40 min at 37°C. The trypsin was then inactivated by adding 1 volume DMEM/10% FCS, and the tissue was dissociated by gentle trituration, centrifuged at 170 x g for 5 min, and plated in the same medium. The cultures were established at a density of 12,000 cells/cm2 on plates coated with 0.5 mg/mL polyornithine (Sigma) in 150 mmol/L sodium borate (Merck; pH 8.4). Early passages of these cells (second to third) were used for this study.

Mitochondrial Labeling
Mammalian cells contain three biotin-dependent carboxylases in the mitochondrial matrix. This makes it easy to specifically stain mitochondria with streptavidin-coupled fluorophores (31). Therefore, streptavidin labeling of mitochondria was done in cells fixed for 15 min with 4% paraformaldehyde and permeabilized for 30 min in PBS/0.1% Triton X-100 (PBT). The cells were then incubated for 20 min at room temperature with 2 to 5 µg/mL FITC-streptavidin (Jackson ImmunoResearch, West Grove, PA) and washed thrice with PBT. Mitochondria were also labeled in live cells with 90 µg/mL Mitotracker or by immunocytochemistry with the 3D5 monoclonal antibody.

Mitochondrial Membrane Potential
The relative variations in {Delta}{Psi}M were determined with rhodamine 123 (Molecular Probes) using a variation of the protocol described previously (73). Briefly, serum-deprived RN22 schwannoma cells (8,000-20,000 per well in 24-well plates) were cultured either in the presence or absence of SP600125, and they were then incubated with 1 µg/mL rhodamine 123 for 10 min at 37°C. The cells were fixed with 4% paraformaldehyde for 15 min and visualized under a Nikon Eclipse 80i fluorescence microscope. Alternatively, rhodamine 123 was extracted with 400 µL butanol (Merck) in cells previously washed twice with PBS. The fluorescence was determined in a FLUOstar OPTIMA spectrofluorometer (excitation 485 nm; emission 532 nm; BMG Labtechnologies, Offenburg, Germany). The cell density was estimated in parallel cultures (n = 3) and the amount of fluorescence per cell (in arbitrary units) was determined by dividing the total fluorescence by estimated number of cells.

ROS Detection
RN22 schwannoma cells were cultured either in the presence or absence of serum for 18 h. The presence of ROS was then monitored after exchanging the medium with PBS containing 10 µmol/L carboxy-H2DCF-DA (Molecular Probes) and incubating the cells for 1 h at 37°C. The appropriate cell culture media were then replaced, and the cells were incubated either in the presence or absence of SP600125 for a further 4 h. The presence of ROS was then detected under a Nikon Eclipse 80i fluorescence microscope as a green signal (excitation wavelength 492-495 nm, emission wavelength 517-527 nm).

Electron Microscopy
RN22 schwannoma cells were cultured either in the presence or absence of serum for 32 h, and they were then isolated by trypsinization with trypsin-EDTA (Life Technologies) and centrifuged for 5 min at 700 x g. The cell pellets were washed with PBS and fixed with 2% paraformaldehyde/2% glutaraldehyde (TAAB Laboratories Equipment Ltd., Berkshire, United Kingdom) in 100 mmol/L sodium cacodylate (pH 7.4) for 48 to 72 h at 4°C. The cells were postfixed with 1% OsO4 (Sigma), dehydrated in ethanol, and embedded in Epon/Araldite (TAAB Laboratories Equipment). After thin sectioning and poststaining in lead citrate, the specimens were observed by transmission electron microscopy on a JEOL 1200-EXII electron microscope.

Fractionation of Cell Extracts
Heavy membrane extracts enriched in mitochondria and the soluble fraction were obtained as described previously (74). Briefly, RN22 schwannoma cells were harvested in isotonic buffer [210 mmol/L mannitol, 70 mmol/L sucrose, 1 mmol/L EDTA, and 10 mmol/L HEPES (pH 7.5)] supplemented with a protease inhibitor cocktail Complete (Roche, Basel, Switzerland) as well as Phosphatase Inhibitor Cocktail 2 (Sigma), and homogenized with a Dounce homogenizer. The samples were transferred to Eppendorf centrifuge tubes, centrifuged at 900 x g for 5 min to remove the nuclei, and then centrifuged at 10,000 x g for 30 min at 4°C to obtain the HM pellet enriched in mitochondria and the soluble fraction (supernatant). The HM material was resuspended in PBS/0.2% Triton X-100 and the protein concentration from both HM and soluble fractions was determined by the Bio-Rad Protein Assay (Bio-Rad, Hercules, CA).

Western Blot
HM extracts (2-5 µg protein) or soluble fraction extracts (5-8 µg protein) were separated by SDS-PAGE on 11% to 13% acrylamide gels under reducing conditions, and transferred to Immun-Blot polyvinylidene difluoride membranes (Bio-Rad). Membranes were blocked in 2% blocking reagent (ECL Advance; GE Healthcare, Bucks, United Kingdom) in PBS with 0.1% Tween 20 for 1 h, and incubated overnight at 4°C with the antisera in blocking buffer. After washing five times in PBS with 0.1% Tween 20, the membranes were incubated with goat anti-rabbit horseradish peroxidase–conjugated antibody (Jackson Immunoresearch) diluted 1/1,660,000 or anti-mouse horseradish peroxidase–conjugated antibody (Bio-Rad) diluted 1/1,350,000 in blocking buffer for 1 h, and washed as above. Protein bands were visualized using ECL Advance.

Immunocytochemistry
Cells were cultured on coverslips coated with 0.5 mg/mL polyornithine/10 µg/mL laminin (Invitrogen) as described above, fixed with 4% paraformaldehyde for 15 min, and then permeabilized for 30 min at room temperature with PBT. Nonspecific binding was then blocked for 30 min with 10% FCS in PBT and the cells were subsequently incubated for 1 h at room temperature with the primary antibody in PBT/1% FCS. After four washes with PBT, cells were incubated for 40 min at room temperature in a 1,000-fold dilution of an anti-mouse IgG antibody coupled to Alexa 594 (Molecular Probes), or Cy2-conjugated anti-rabbit IgG antibody (Jackson Immunoresearch). For JNK and active JNK immunostaining, we followed the protocol described by the manufacturers [PhosphoPlus stress-activated protein kinase/JNK (Thr183/Tyr185) Antibody kit; Cell Signaling Technology]. Cultures were finally washed four times in PBT and once in PBS alone, and they were mounted in 50% glycerol in PBS. Nuclei were counterstained with 1 µg/mL propidium iodide or bisbenzimide. Images were recorded using a DXM 1200 digital camera (Nikon).

Analysis of Cell Death
Serum-deprived RN22 schwannoma cells (20,000 per well) were plated in 24-well tissue culture dishes, and treated as described. Cell viability was studied by determining the amount of yellow MTT (Sigma) that was reduced to insoluble purple formazan. After removing the medium, the water-insoluble formazan was solubilized with 400 µL DMSO (Sigma), and the dissolved material was measured spectrophotometrically at a wavelength of 570 nm, subtracting the background at 650 nm. Apoptosis in these cells was determined using TUNEL staining done with the In situ Cell Death Detection kit (Roche). The formation of autophagic vacuoles was determined by means of an LC3-GFP plasmid (36), kindly provided by Tamotsu Yoshimori (National Institute of Genetics, Mishima, Japan), which was previously transfected using LipofectAMINE 2000 (Invitrogen), following the manufacturer's instructions.

Cell Counting and Statistics
Cell counting was done using a Nikon E80i microscope with phase contrast and epifluorescence illumination. On average, 300 to 500 cells were analyzed per coverslip. The quantitative data are shown as the mean ± SE from n independent experiments. Statistical differences were analyzed using the Student's t test.


   Acknowledgements
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
References
 
We thank R. Diez del Corral, M.J. Latasa, and M. Sefton for useful scientific comments, and Tamotsu Yoshimori (Osaka University, Osaka, Japan) for providing us with the LC3-GFP plasmid.


   Notes
Top
 Notes
Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 
Grant support: "La Caixa" Foundation BM05-71-0, CIEN Foundation, and Spanish Ministry of Education and Science grant BMC2003-03441 (J.M. Frade); and a Research Personnel Training Fellowship (N. López-Sánchez).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: Supplementary data for this article are available at Molecular Cancer Research Online (http://mcr.aacrjournals.org/).

Received 7/31/06; revised 11/27/06; accepted 11/30/06.


   References
Top
 Notes
 Abstract
 Introduction
 Results
 Discussion
 Materials and Methods
 Acknowledgements
 References
 

  1. Green DR, Evan GI. A matter of life and death. Cancer Cell 2002;1:19–30.[CrossRef][Medline]
  2. Leist M, Jäättelä M. Four deaths and a funeral: from caspases to alternative mechanisms. Nature Rev Mol Cell Biol 2001;2:1–10.[Medline]
  3. Hibi M, Lin A, Smeal T, Minden A, Karin M. Identification of an oncoprotein- and UV-responsive protein kinase that binds and potentiates the c-Jun activation domain. Genes Dev 1993;7:2135–48.[Abstract/Free Full Text]
  4. Davis RJ. Signal transduction by the JNK group of MAP kinases. Cell 2000;103:239–52.[CrossRef][Medline]
  5. Maundrell K, Antonsson B, Magnenat E, et al. Bcl-2 undergoes phosphorylation by c-Jun N-terminal kinase/stress-activated protein kinases in the presence of the constitutively active GTP-binding protein Rac1. J Biol Chem 1997;272:25238–42.[Abstract/Free Full Text]
  6. Morrison DK, Davis RJ. Regulation of MAP kinase signaling modules by scaffold proteins in mammals. Annu Rev Cell Dev Biol 2003;19:91–118.[CrossRef][Medline]
  7. Kyriakis JM, Avruch J. Mammalian mitogen-activated protein kinase signal transduction pathways activated by stress and inflammation. Physiol Rev 2001;81:807–69.[Abstract/Free Full Text]
  8. Chang L, Karin M. Mammalian MAP kinase signalling cascades. Nature 2001;410:37–40.[CrossRef][Medline]
  9. Bost F, McKay R, Dean N, Mercola D. The JUN kinase/stress-activated protein kinase pathway is required for epidermal growth factor stimulation of growth of human A549 lung carcinoma cells. J Biol Chem 1997;272:33422–9.[Abstract/Free Full Text]
  10. Bost F, McKay R, Bost M, Potapova O, Dean NM, Mercola D. The Jun kinase 2 isoform is preferentially required for epidermal growth factor-induced transformation of human A549 lung carcinoma cells. Mol Cell Biol 1999;19:1938–49.[Abstract/Free Full Text]
  11. Yang YM, Bost F, Charbono W, et al. C-Jun NH(2)-terminal kinase mediates proliferation and tumor growth of human prostate carcinoma. Clin Cancer Res 2003;9:391–401.[Abstract/Free Full Text]
  12. Venugopal R, Jaiswal AK. Nrf2 and Nrf1 in association with Jun proteins regulate antioxidant response element-mediated expression and coordinated induction of genes encoding detoxifying enzymes. Oncogene 1998;17:3145–56.[CrossRef][Medline]
  13. Wang MC, Bohmann D, Jasper H. JNK signaling confers tolerance to oxidative stress and extends lifespan in Drosophila. Dev Cell 2003;5:811–6.[CrossRef][Medline]
  14. Karbowski M, Youle RJ. Dynamics of mitochondrial morphology in healthy cells and during apoptosis. Cell Death Differ 2003;10:870–80.[CrossRef][Medline]
  15. Bossy-Wetzel E, Barsoum MJ, Godzik A, Schwarzenbacher R, Lipton SA. Mitochondrial fission in apoptosis, neurodegeneration and aging. Curr Opin Cell Biol 2003;15:706–16.[CrossRef][Medline]
  16. Sesaki H, Jensen RE. Division versus fusion: Dnm1p and Fzo1p antagonistically regulate mitochondrial shape. J Cell Biol 1999;147:699–706.[Abstract/Free Full Text]
  17. Nunnari J, Marshall W, Straight A, Murray A, Sedat J, Walter P. Mitochondrial transmission during mating in Saccharomyces cerevisiae is determined by mitochondrial fusion and fission and the intramitochondrial segregation of mitochondrial DNA. Mol Biol Cell 1997;8:1233–42.[Abstract]
  18. Frank S, Gaume B, Bergmann-Leitner ES, et al. The role of dynamin-related protein 1, a mediator of mitochondrial fission, in apoptosis. Dev Cell 2001;1:515–25.[CrossRef][Medline]
  19. Pfeiffer SE, Wechsler W. Biochemically differentiated neoplastic clone of Schwann cells. Proc Natl Acad Sci U S A 1972;69:2885–9.[Abstract/Free Full Text]
  20. Gentry JJ, Casaccia-Bonnefil P, Carter BD. Nerve growth factor activation of nuclear factor {kappa}B through its p75 receptor is an anti-apoptotic signal in RN22 schwannoma cells. J Biol Chem 2000;275:7558–65.[Abstract/Free Full Text]
  21. Frade JM. Nuclear translocation of the p75 neurotrophin receptor cytoplasmic domain in response to neurotrophin binding. J Neurosci 2005;25:1407–11.[Abstract/Free Full Text]
  22. Susen K, Blochl A. Low concentrations of aggregated ß-amyloid induce neurite formation via the neurotrophin receptor p75. J Mol Med 2005;83:720–35.[Medline]
  23. Ito Y, Mishra NC, Yoshida K, Kharbanda S, Saxena S, Kufe D. Mitochondrial targeting of JNK/SAPK in the phorbol ester response of myeloid leukemia cells. Cell Death Differ 2001;8:794–800.[CrossRef][Medline]
  24. Kharbanda S, Saxena S, Yoshida K, et al. Translocation of SAPK/JNK to mitochondria and interaction with Bcl-x(L) in response to DNA damage. J Biol Chem 2000;275:322–7.[Abstract/Free Full Text]
  25. Brichese L, Cazettes G, Valette A. JNK is