Activation and Role of MAP Kinase-Dependent Pathways in Retinal Pigment Epithelium Cells: JNK1, P38 Kinase, and Cell Death

Abstract

purpose. Retinal pigment epithelial (RPE) cell death is an important step in the pathogenesis of ocular diseases. JNK1 and P38 kinase, two stress-activated kinases, play key roles relaying stress signals leading to cell death through cyclin D1 and c-Myc. Recently, stress-activated kinases have been shown to regulate cell proliferation. In the current study, the involvement of the JNK1 and P38 kinase signaling pathways in RPE cell proliferation and death was investigated.

methods. RPE cell proliferation was stimulated with 10% fetal calf serum (FCS). Activation of the JNK1 and P38 kinase cascades and their potential targets was detected by Western blot analysis. Pharmacologic inhibitors and activators, and antisense oligodeoxynucleotides (ODN) directed against the stress kinases were used to analyze the signaling involved in RPE cell death.

results. P38 and JNK1 and their respective upstream activating kinases, MKK3/6 and -4, were all transiently activated in FCS-stimulated RPE cell cultures. Ras controlled only the activation of JNK1, whereas Rho transmitted the activation of both JNK1 and P38, suggesting parallel signaling pathways and cross talk between the two kinases. Pharmacologic inhibition of JNK1 did not affect cell proliferation in FCS-stimulated cells. Inactivation of P38 kinase and antisense ODN-induced downregulation of P38 kinase also had no affect on cell proliferation. Long-term, high-level activation of JNK1 and P38 kinase occurred during serum depletion-induced RPE cell death. Overactivation of JNK1 and P38 kinase was also observed during pharmacologically induced cell death, suggesting that this process is common to RPE cell-death-signaling pathways induced by various stress stimuli. Cell death mediated by the overactivation of JNK1 and P38 kinase was cyclin D1- and c-Myc-independent.

conclusions. The inhibition of JNK1 or P38 kinase had no effect on FCS-stimulated proliferation of RPE cells, whereas the overactivation of these two enzymes was involved in RPE cell death in FCS-depleted cultures. Parallel upstream signaling pathways and cross talk between the two kinases suggest that the regulation of signaling in RPE cell death is complex.

In many situations, cells must adapt to environmental changes by monitoring and reacting quickly to extracellular stimuli. The mitogen-activated protein kinase (MAPK) cascade is one of the most ubiquitous signal transduction systems and is rapidly activated in response to a variety of cellular stimuli, including cellular stress and cell death. ¹ Stress-activated c-Jun NH₂-terminal kinase 1 (JNK1)/SAPK and P38 kinase, two members of the MAPK family, are known to control gene transcription in response to cellular stress. ² They are arranged in modules. ³ MKK3/6 activates P38 kinase and MKK4 activates JNK1. P38 kinase has been shown to regulate the activity of various transcription factors, such as activating transcription factor (ATF)-1 and -2, and cAMP response element-binding protein (CREB), whereas JNK1 regulates activation of the transcription factor, c-Jun. ⁴ The Rho GTPases activate the JNK1 and P38 kinase signaling cascades, but Ras may also be indirectly involved in this activation, ⁵ showing diversity in the mode of control of these two MAPKs. This suggests that the generation of multiple signal inputs through intentional cross talk between different signaling molecules of the stress-signaling protein kinase pathways is a mechanism by which cells ensure that the required complement of signal is present before committing themselves to responses critical in cellular stress. 

The JNK1/P38 kinase and extracellular signal-regulated kinase (ERK)1/2 signaling cascades have been shown to have opposite effects on apoptosis regulation, ^{6 7} suggesting that the JNK1 and P38 kinase are cell death mediators. However, JNK1 and P38 kinase have recently been demonstrated to be essential for the induction of cell proliferation by growth factors, ⁸ showing new functions for these stress-activated kinases. Conversely, it has been shown that inhibition of the activity of P38 kinase stimulates cell proliferation, ⁹ demonstrating an opposite biological function for P38 kinase. Treatment with the pharmacologic inhibitor of P38 kinase, SB203580, induces mitotic conversion, suggesting that inhibition of P38 kinase is necessary for progression through mitosis. ¹⁰ The activation of P38 kinase also has a negative role in cell proliferation on treatment with the protein synthesis inhibitor anisomycin and with nocodazole. ¹¹ Oxidative stress, genotoxic agents, or gamma irradiation can also induce mitotoxic arrest dependent on the selective activation of P38 kinase. ¹² Moreover, depletion of P38 kinase, by antisense ODNs, has been shown to abolish the G2 delay induced by UV irradiation. ¹³ These data suggest that JNK1 and P38 kinase may have different (cell death-cell proliferation), and sometimes opposite (cell proliferation-cell proliferation arrest) biological activities depending on the cell type and stimulus. The molecular mechanisms by which P38 kinase regulates progression through G1 during the cell cycle are not completely understood. The inhibitory role of P38 kinase at the G₁-S transition correlates with the downregulation of cyclin D1 levels in vitro and in vivo. ^{14 15} The transcription factor c-Myc is involved in the control of both cell proliferation and cell death. It has also been shown that activation of P38 kinase induces the downregulation of c-Myc in nonapoptotic cells, ¹⁶ whereas the initiation of translation of c-Myc is mediated by P38 kinase during apoptosis. ¹⁷ These apparently conflicting data demonstrate the opposite roles of P38 kinase in the control of c-Myc production depending on the status of the cell. In contrast, numerous data have clearly shown that c-Myc is involved in the JNK1-regulated apoptosis triggered by various stimuli, such as anticancer drug treatment and oxidative stress in normal and transformed cells. ^{18 19 20}  

In pathologic conditions, RPE cells are exposed to a variety of stimuli that may induce cell death. It would therefore be useful to determine precisely which signaling pathways mediate RPE cell death, with a view toward inhibiting this process specifically in RPE diseases such as age-related macular degeneration (AMD). The molecular mechanisms mediating RPE cell death are mostly poorly understood and the role of stress-activated signaling kinases in this process has not been investigated. We therefore investigated the activation status of the JNK1 and P38 kinase signaling pathways and determined their respective upstream intracellular activators, possible cross talk between these pathways, and effects of inhibiting their activation in RPE cells. We also investigated whether the overactivation of JNK1 and P38 kinase occurs during RPE cell death, whether drug-induced overactivation of both JNK1 and P38 kinase induces RPE cell death, and whether cyclin D1 and c-Myc mediates the effects of JNK1 and P38 kinase. 

Materials and Methods

RPE Culture and Treatment of Cells

Human RPE cells were isolated as previously described. ²¹ Primary human RPE cell cultures were seeded on 0.1% gelatin-coated dishes and cultured in F99RPE growth medium. Subcultures were grown in F99 basal medium (medium 199/Ham’s F12; Gibco/BRL, Grand Island, NY), supplemented with 10% FCS, 1 mM sodium pyruvate, 1 μg/mL insulin, 50 μg/mL gentamicin, and 2.5 μg/mL amphotericin B. The proliferation of RPE cells was assessed daily by counting the number of cells and by determining incorporation of [3H]-thymidine (0.92 TBq/mmol; Amersham, Orsay, France) as previously described. The number of cells was determined by two methods: trypan blue exclusion and MTT (3 (4,5-dimethylthiazol-,yl) 2,5 diphenyltetrazolium bromide) staining. ²²  

In some experiments, specific pharmacologic inhibitors of Ras (FTS), Rho guanosine triphosphatases (GTPases; toxin B, Clostridium difficile), MEK1/2 (U0126; Calbiochem, Meudon, France), JNK1 (D-JNKI1) and P38 kinase (SB203580), and an activator of both P38 kinase and JNK1 (anisomycin) were used to analyze the involvement of specific signaling pathways in RPE cell death. U0126 had no effect on the activities of JNK1 and P38 kinase, D-JNKI1 did not affect ERK1/2 and P38 kinase activities, and SB203580 did not affect ERK1/2 and JNK1 activities, demonstrating that the signaling inhibition achieved with U0126, D-JNKI1, and SB203580 was highly specific. In contrast, SB202190 was found to inhibit P38 kinase and to activate JNK1 and P38 kinase, when added to RPE cell cultures at low (below 25 μM) and high (40 μM) concentrations, respectively. SB202190 had no effect on JNK1 activation at concentrations below 25 μM. Anisomycin, which is known to inhibitor protein synthesis at concentrations of 0.1 to 100 μg/mL had no effect on protein synthesis in RPE cell cultures, when added at concentrations of 3 to 0.03 ng/mL. Pharmacologic agents were added 2 hours before cell stimulation with 10% FCS on the day of serum stimulation and on day 3 of the culture period. Cells were again stimulated with 10% FCS on day 3 of the culture period. Stock solutions of each inhibitor were made in dimethyl sulfoxide (DMSO) and diluted in DMEM, so that the final concentration of DMSO in test solutions did not exceed 0.1% (a concentration that has no effect on RPE cell proliferation or cell death). 

Western Blot Analysis

RPE cells were washed twice in PBS, lysed in ice-cold lysis buffer (50 mM Tris-HCl [pH 7.5], 100 mM NaCl, 50 mM NaF, 5 mM EDTA, 40 mM β-glycerophosphate, 0.2 mM sodium orthovanadate, 1 μg/mL leupeptin, and 1 μM pepstatin) and centrifuged at 4°C for 10 minutes at 10,000g. A monoclonal antibody directed against β-actin was used as an internal standard for checking protein loading. Cell lysates were mixed with 3× Laemmli buffer and heated for 5 minutes at 95°C. They were then subjected to SDS-PAGE (12%–15% polyacrylamide gel), and the proteins were transferred to nitrocellulose filters by electroblot and probed with polyclonal antibodies directed against Raf-1, ERK2, P90^RSK, c-Myc (dilution 1:100, Santa Cruz Biotechnology, Santa Cruz, CA), c-Jun (dilution 1:1000; Cell Signaling Technology, Beverly, MA), and cyclin D1 (NeoMarkers, Fremont, CA) to determine the amounts of these kinases present during the period of cell proliferation and death. Polyclonal antibodies directed against phospho-JNK1 (thr¹⁸³ and tyr¹⁸⁵), phospho-P38 kinase (thr¹⁸⁰ and tyr¹⁸²), and phospho-ERK1/2 (thr¹⁹² and tyr¹⁹⁴; dilution 1:5000; Promega, Madison, WI,) and against phospho-c-Jun (ser⁶³), phospho-MKK4 (thr²⁶¹), and phospho-MKK3/6 (ser¹⁸⁹ and ser²⁰⁷; dilution 1:1000; Cell Signaling) were used to analyze the activation of intracellular signaling during RPE cell proliferation. The primary antibodies were detected with a horseradish peroxidase-conjugated goat anti-rabbit secondary antibody. Enhanced chemiluminescence (ECL) substrate was used to detect the secondary antibody according to the manufacturer’s instructions (Amersham), and the membrane was placed against autoradiograph film (Hyperfilm; Amersham). The bands on the fluorograph were quantified with a laser densitometer (LKB Ultrascan XL; Pharmacia, Saclay, France). 

ODNs and ODN Treatment of the Cells

Sense and antisense phosphorothioate ODNs directed against P38 kinase were designed based on the published sequence of the human P38 kinase gene. The sense ODN sequence was identical with those of the P38 kinase genes conserved in human, mouse and rat. ²³ The antisense P38 kinase ODN was 5′-GTC-TTG-TTC-AGC-TCC-TGC-3′ (referred to as AS P38) and the corresponding sense ODN was 5′-GCA-GGA-GCT-GAA-CAA-GAC-3′ (referred to as S P38). Sense and antisense phosphorothioate ODNs directed against c-Myc were designed based on the published sequence of the human c-Myc gene. The antisense c-Myc ODN was 5′-AAC-GTT-GAG-GGG-CAT-3′ (referred to as AS c-Myc) and the sense c-Myc ODN was 5′-AGC-TGG-GGT-AGC-AAT-3′ (referred as to S c-Myc). This antisense ODN is the most potent antisense inhibitor yet. It reduces tumor growth and metastatic potential in a human melanoma model. ²⁴ We used lipofectin (GibcoBRL, Cergy Pontoise, France), a cationic lipid, to deliver the ODNs, because this method results in high levels of ODN uptake and stability in the intracellular compartment and does not affect the final nuclear location of the ODNs after endocytosis and release from the endocytic compartment. All ODNs were synthesized commercially (Invitrogen, Cergy-Pontoise, France) and purified by HPLC. Lipofectin-ODN complexes were produced according to the manufacturer’s instructions. The results presented were obtained with 1 μM P38 kinase and 0.5 μM c-Myc ODNs and with 7 μg/mL lipofectin. Cells were incubated for 3 days in the presence of 10% heat-treated calf serum. RPE cells were then treated with ODNs and lipofectin for 4 hours, washed twice with serum-free basal F99 medium and fresh complete culture F99 medium to which the appropriate concentration of ODNs was added. The cells were cultured for 7 days. For 3 days, beginning on day 4 of incubation, cells were treated with the appropriate concentration of ODNs without lipofectin. 

Statistics

Each figure shows the results of experiments repeated at least three times. Data are expressed as the mean ± SEM. Mann-Whitney tests (nonparametric tests) were used for statistical analysis. 

Results

Activation of the MKK3/6-P38 Kinase and MKK4-JNK1 Modules during RPE Cell Proliferation

We investigated the role of the P38 kinase signaling pathway in RPE cell proliferation. In this pathway, MKK3 and -6, two closely related kinases (thus, MKK3/6), are activated through phosphorylation and, in turn, stimulate P38 kinase and activate nuclear protein substrates, such as the transcription factors ATF-2 and CREB. The states of phosphorylation of MKK3/6 and P38 kinase were studied by Western blot analysis. MKK3/6 phosphorylation was barely detectable before serum stimulation, and the stimulation of RPE cell proliferation by the addition of serum caused an increase in MKK3/6 phosphorylation by a factor of 8- to 11-fold after 10 minutes (Fig. 1A) . MKK3/6 activation was constant during the first 30 minutes and then returned to levels similar to those detected before serum stimulation (Fig. 1A) . Phosphorylated P38 kinase was detected before serum stimulation (Fig. 1B) . Treatment with 10% serum resulted in an increase in P38 kinase phosphorylation by a factor of three to four within the first 10 minutes of treatment (Fig. 1B) . Low levels of P38 kinase phosphorylation were observed during the first 30 minutes, followed by a decline to levels of phosphorylation observed before serum stimulation. The production of MKK3 and P38 kinase was constant during the 6-day period of culture (data not shown). We investigated whether P38 kinase was involved in serum-mediated signaling for RPE proliferation by treating cell cultures with the highly specific inhibitor of P38 kinase, SB203580 (0–20 μM). Serum-induced RPE cell proliferation was not significantly affected by the inhibition of P38 kinase (Fig. 1C) . The inhibition of P38 kinase activity with 20 μM SB202190, detected by monitoring c-Jun production on Western blots, did not affect RPE cell proliferation (data not shown). To confirm that P38 kinase was not involved in the proliferation of serum-stimulated RPE cells, we downregulated P38 kinase production by an antisense ODN approach. P38 kinase depletion cells by treatment with AS P38 ODNs in serum-stimulated did not significantly affect cell proliferation or induce cell death (Fig. 1D) , demonstrating that although serum greatly stimulates P38 kinase activation, this activation is not necessary for cell proliferation or cell survival. 

Although the effect of JNK1 on cell survival is unclear and cell specific, it has been shown that the duration of JNK1 activation may determine cell death and proliferation. ²⁵ As with other MAPK pathways, the core signaling unit is composed of an MAPK, typically MKK4, which, once activated, phosphorylates and activates JNK1, which in turn activates transcription factors such as ATF-2 and c-Jun. We studied the state of activation of MKK4 and JNK1 during a 24-hour period by Western blot analysis. MKK4 phosphorylation was barely detectable before stimulation with serum (Fig. 2A) . Stimulation of RPE cell proliferation resulted in the activation of MKK4 phosphorylation after 10 minutes of culture and during a 6-hour period of culture. MKK4 activation then returned to levels similar to those detected before stimulation with serum (Fig. 2A) . The production of MKK4 was constant during the 24-hour period of culture (data not shown). Levels of JNK1 phosphorylation were very low before stimulation with serum (Fig. 2B) . Serum stimulated JNK1 phosphorylation by a factor of 6 to 8 after 10 minutes. At 30 minutes, levels of JNK1 phosphorylation had decreased slightly (four to five times higher than before stimulation with serum). At 6 hours, JNK1 phosphorylation had decreased more strongly, and at 24 hours, it was identical with that in the absence of serum. The production of JNK1 was constant during the 24-hour period of culture (data not shown). To investigate whether JNK1 was involved in serum-mediated signaling for RPE proliferation, we treated cell cultures with the highly specific inhibitor of JNK1, D-JNKI1 (0–20 μM). Inhibition of JNK1 during a 6-day period of culture did not significantly affect serum-stimulated cell proliferation or induce cell death, suggesting that JNK1 activation is not involved in RPE cell proliferation and survival (Fig. 2C) . 

Analysis of Upstream Cross-regulation of the JNK1 and P38 Signaling Pathways

Rho GTPases have been reported to regulate JNK1 and P38 kinase pathways, whereas Ras has been shown to be a direct activator of the ERK1/2 signaling pathway. ⁵ To determine whether the activation of JNK1 and P38 kinase was controlled by Ras or Rho proteins, we treated serum-stimulated RPE cells with the Ras inhibitor, FPT III inhibitor, and the Rho protein inhibitor toxin B from C. difficile. We investigated the levels of activation of JNK1 and P38 kinase by Western blot analysis with antibodies that recognized phosphorylated forms of JNK1 and P38 kinase, respectively, and investigated the levels of production of JNK1 and P38 kinase by Western blot with antibodies that recognized total JNK1 and P38 kinase, respectively. The ratio of the production levels to the phosphorylation levels of the kinases shows the specificity of the inhibitor-induced change of the phosphorylation. Ras inhibition reduced JNK1 phosphorylation after 30 minutes and inhibited JNK1 phosphorylation during the entire 24-hour period of culture (Fig. 3A) . In contrast, Ras inhibition did not significantly affect P38 kinase phosphorylation (Fig. 3B) . In control experiments, the effects of Ras inhibition on the phosphorylation of ERK1/2 were also investigated by Western blot analysis. As expected, ERK1/2 phosphorylation was strongly affected by treatment with the Ras inhibitor. No phosphorylation of ERK2 was detected from 30 minutes until the end of the 24-hour period of culture. JNK1 phosphorylation decreased after 10 minutes of treatment with the Rho GTPase inhibitor (Fig. 4A) . Rho GTPase inhibition reduced P38 kinase phosphorylation to levels similar to those detected before stimulation with serum (Fig. 4B) . In contrast, ERK2 phosphorylation was not affected by the Rho GTPase inhibitor (Fig. 4C) . These data strongly suggest that the activation of JNK1 and P38 kinase is differentially controlled by Ras and Rho GTPases, and suggest potential cross-regulation between JNK1 and P38 kinase by Rho GTPases and between JNK1 and ERK1/2 by Ras. 

Effect of Overactivation of the JNK1-P38 Kinase Signaling Pathways

The activation of JNK1 and P38 kinase was detected during cell death induced by various stimuli. We therefore analyzed the levels of JNK1 and P38 kinase phosphorylation during RPE cell death induced by serum depletion. Serum depletion induced RPE cell death after 3 days of culture (Fig. 5) . Serum depletion reduced the number of RPE cells on days 3 and 6 by 54% and 77%, respectively, compared with the number of cells present on day 1 of the culture period (Fig. 5C) . We then performed Western blot analysis to analyze the levels of phosphorylation of JNK1 and P38 kinase after serum depletion. Serum depletion altered the kinetics of activation of both JNK1 and P38 kinase by increasing the duration of phosphorylation for both JNK1 and P38 kinase (Fig. 5A) . High levels of phosphorylation of JNK1 and P38 kinase were still detectable after 2 hours and during the 24-hour culture period (Fig. 5A) . These data suggest that the long-term activation of both JNK1 and P38 kinase is involved in cell death signaling in serum-depleted RPE cell cultures. To confirm the specific role of the overactivation of JNK1 and P38 kinase in RPE cell death, we inhibited JNK1 and P38 kinase by adding 20 μM D-JNKI1 and 20 μM SB202190, respectively, to serum-depleted RPE cell cultures. Inhibition of long-term activation of JNK1 or P38 kinase by adding D-JNKI1 and SB202190 was checked in serum-depleted RPE cell cultures (Fig. 5B) . The inhibition of JNK1 or P38 kinase reversed by 73% and 54%, respectively, the level of cell death induced by serum depletion on day 3, and by 67% and 55%, respectively, that observed on day 6 of the culture period (Fig. 5C) . In addition, activation of JNK1 and P38 kinase was not inhibited by 20 μM SB202190 and 20 μM D-JNKI1, respectively (data not shown), demonstrating the specificity of the JNK1 and P38 kinase inhibitors. These data show that inhibiting overactivation of the JNK1 or P38 kinase signaling pathways was sufficient to reduce RPE cell death significantly, strongly suggesting that serum-depletion-induced cell death is caused by the overactivation of JNK1 and P38 kinase. 

It has been shown that the persistent activation of SAPK correlates with cell apoptosis. ²⁵ We therefore thought that overactivation of both the JNK1 and P38 kinase signaling pathways might induce RPE cell death, even in the absence of cellular stress such as serum depletion. We therefore investigated the effect of persistent activation of JNK1 and P38 kinase on cell death in RPE cells cultured in the presence of serum. RPE cell treatment with 40 μM SB202190 resulted in the strong and sustained activation of JNK1 during the 24-hour culture period, whereas only transient JNK1 activation was observed with serum alone (Fig. 6A) . Cell treatment with 40 μM SB202190 also rapidly increased the activation of P38 kinase (Fig. 6A) but did not affect the production of JNK1 and P38 kinase (data not shown). We investigated whether the overactivation of both JNK1 and P38 kinase also results in stimulation of the downstream pathways for JNK1 and P38 kinase by analyzing the effects of 40 μM SB202190 on the activation of c-Jun, CREB, and ATF2 by Western blot analysis. After 2 hours, serum induced the weak and transient phosphorylation of c-Jun in untreated RPE cells, whereas it induced a sustained production of c-Jun (Fig. 6B) . Cell treatment with 40 μM SB202190 resulted in stronger, sustained phosphorylation of c-Jun during the 24-hour culture period (Fig. 6B) . In contrast, cell treatment with SB202190 did not affect production of c-Jun during the 24-hour period of culture (Fig. 6B) . These data suggest that activation of the downstream cascade of JNK1 with 40 μM SB202190 involves the stimulation of c-Jun transcriptional activity. Activation of CREB and ATF-2 also increased after cell treatment with 40 μM SB202190 (data not shown). Overactivation of both JNK1 and P38 kinase with 40 μM SB202190 rapidly induced RPE cell death. Cell treatment with 40 μM SB202190 led to a 71% and 90% decrease in the total number of cells after 3 and 6 days of culture, respectively, from the total number of cells present before treatment (1.2 × 10⁴ and 0.4 × 10⁴ cells on days 3 and 6 of the treatment with 40 μM SB202190 versus 4.2 × 10⁴ cells before cell treatment; Fig. 6C ). SB202190 at a concentration of 20 μM, a concentration effective for P38 kinase inhibition but not for JNK1 inhibition (data not shown), no difference in the total number of cells was observed between untreated cell cultures and 20 μM SB202190-treated cultures (Fig. 6C) , confirming that the inhibition of P38 kinase, but not of JNK1, had no effect on cell proliferation in serum-stimulated RPE cell cultures. These data suggest that overactivation of both the JNK1 and P38 kinase pathways induces RPE cell death, even in culture conditions (presence of serum) permissive of cell proliferation. To confirm that the overactivation of JNK1 and P38 kinase was involved in RPE cell death, we treated cells with another pharmacologic agent that induces persistent activation of these two kinases. Treatment of serum-stimulated RPE cells with 0.03 ng/mL anisomycin, a very low concentration that does not inhibit protein synthesis in RPE cell cultures (data not shown), persistently stimulated the phosphorylation of both JNK1 and P38 kinase and also rapidly resulted in the sustained phosphorylation of c-Jun and ATF2 (data not shown). The treatment of serum-stimulated cells with 0.03 ng/mL anisomycin also induced cell death. Anisomycin induced a 52% and 43% decrease in the total number of cells after 3 and 6 days of culture, respectively, from that observed before treatment with 40 μM SB202190 (Fig. 6C) . 

In conclusion, all these data show that despite the transient serum-mediated activation of both P38 kinase and JNK1, these two SAPKs were not involved in serum-induced RPE cell growth signaling. In contrast, they suggest that an increase in the duration of activation of JNK1 and P38 kinase is involved in the control of signal transduction in RPE cell death. 

Although it has been demonstrated that production of cyclin D1 is regulated positively by ERK1/2 and negatively by the P38 kinase pathway ¹⁴ and that the downregulation of cyclin D1 induces cell apoptosis, ²⁶ recent data have demonstrated that the induction of cyclin D1 involves the activation of both JNK1 and P38 kinase. ²⁷ We therefore analyzed the effects of overactivating both JNK1 and P38 kinase on the levels of cyclin D1 in serum-treated RPE cells. The stimulation of RPE cell cultures with serum induced the production of cyclin D1 after a 6-hour period of treatment and more than a 24-hour period of culture (Fig. 7A) . The overactivation of both JNK1 and P38 kinase did not significantly affect the kinetics of cyclin D1 production, suggesting that cell death induction by overactivation of both JNK1 and P38 kinase was not mediated by changes in cyclin D1 levels (Fig 7A) . The inactivation of ERK1/2 by the MEK1/2 inhibitor, U0126, resulted in a marked decrease in serum-induced cyclin D1 production (Fig. 7B) , confirming that RPE cell proliferation was controlled by the ERK1/2-mediated stimulation of cyclin D1 production. 

The mitogenic and proapoptotic activities of c-Myc are functionally inseparable and JNK/P38 kinases have been shown to regulate c-Myc-mediated apoptosis. ^{28 29} We therefore also investigated, by Western blot analysis, the production of c-Myc after ERK inhibition and JNK1-P38 kinase overactivation. The stimulation of RPE cells by serum increased the production of c-Myc after 2 hours of treatment and during the 24-hour period of culture (Fig. 7C) . Overactivation of both JNK1 and P38 kinase did not alter c-Myc production during the 24-hour period of culture (Fig. 7C) , suggesting that c-Myc was not involved in the JNK1/P38 kinase overactivation-induced pathway relaying RPE cell death signals. The inhibition of ERK signaling reduced c-Myc production late, after 24 hours of culture (Fig. 7D) , suggesting that ERK signaling pathway did not directly control c-Myc production. These data suggesting that ERK, JNK1, and P38 kinase does not control c-Myc production do not exclude a role for c-Myc in RPE cell proliferation and death. To investigate the role of c-Myc in RPE cell proliferation and death, we downregulated c-Myc production by means of an antisense ODN approach. The depletion of c-Myc by treatment with AS c-Myc ODNs resulted in a 44% decrease in cell proliferation after 3 days of culture and a 37% decrease after a 7-day period of culture (Fig. 8) . In contrast, cell treatment with sense c-Myc ODNs did not significantly affect serum-stimulated cell proliferation. These data strongly suggest that c-Myc is involved in signaling for cell proliferation and not cell death in serum-stimulated RPE cell culture. 

Discussion

Regulation of Equilibrium between Cell Proliferation and Cell Death by the Levels of Activation of JNK1 and P38 Kinase in RPE Cells

The purpose of this study was to determine whether SAPK signaling pathways were involved in the proliferation and death of RPE cells. We demonstrated that although both JNK1 and P38 kinase were activated during RPE cell proliferation, the inhibition of either JNK1 or P38 kinase had no effect on cell death in serum-stimulated RPE cells. In contrast, overactivation of both the JNK1 and P38 kinase pathways was involved in RPE cell death induced by various stimuli. A proposed scheme for activation of the JNK1 and P38 kinase signaling pathways in RPE cell death is presented in Figure 9 . 

It has been shown in previous studies that the major MAPKs involved in proliferation of many cell types are ERK1/2. ³⁰ However, recent studies have also demonstrated that JNK1 and P38 kinases may act together with ERK1/2 and may be involved in the mediation of cell proliferation signaling. ^{31 32} It has also been shown that these stress-activated kinases may transmit proliferative signals in the absence of ERK1/2 activation. Thus, the functions of JNK1 and P38 kinase in cell proliferation remain unclear and seem to be stimulus and cell specific. We demonstrated that although all three MAPKs were activated during RPE cell proliferation, only the ERK1/2 pathway controlled FCS-induced RPE cell proliferation, because inhibition of ERK, but not of JNK1 or P38 kinase, reduced cell proliferation. In our study, JNK1 and P38 kinase had no positive effect on the activation of this process. It has recently been demonstrated that the activation of JNK1 and P38 kinase is correlated with the inhibition of cell cycle progression. ⁹ If JNK1 or P38 kinase had antiproliferative activity, then the inhibition of JNK1 or P38 kinase activation would induce an increase in RPE cell proliferation. We found that, in serum-stimulated cells, the pharmacologic inhibition of P38 kinase or JNK1 did not reduce serum-induced cell proliferation. This strongly suggests that JNK1 and P38 kinase are not independently involved in antiproliferative activities in serum cultured RPE cells. 

Another widely accepted dogma was that the SAPKs, JNK1 and P38 kinase, are activated when cell death is induced. Our data are in agreement with this dogma, but also show that JNK1 and P38 kinase can be activated without occurrence of cell death in serum-stimulated RPE cells. Moreover, although the activation of both JNK1 and P38 kinase was observed in FCS-stimulated RPE cells, the inhibition of cell proliferation but not of cell death was detected if the activities of MEK1/2-ERK1/2 were inhibited in PD98059-treated RPE cell cultures (Mascarelli F, unpublished data, 2002). This strongly suggests that, in the absence of ERK stimulation, the activation of JNK1 and P38 kinase was not involved in cell death signaling in serum-stimulated RPE cells. 

In contrast, we showed that JNK1 and P38 kinase were overactivated in serum-depleted cells and that this overactivation correlated with cell death. This suggests that the duration of JNK1 and P38 kinase activation may play a role in the specificity of the transmitted signal for RPE cell proliferation and death. The key role of activation kinetics in the function of the SAPK was confirmed by showing that pharmacologic inhibition of JNK1 or P38 kinase overactivation greatly reduced serum-depletion-induced RPE cell death. This also suggests that the inhibition of only one of the signaling pathways is sufficient to inhibit cell death to a large extent. It would therefore be of value to determine the specific intracellular targets of the overactivated JNK1 and P38 kinase signaling pathways in serum-depleted RPE cells. We have also previously demonstrated that ERK1/2 was also overactivated during cell death in serum-depleted RPE cell cultures. ²² This suggests that the overactivation of ERK1/2 signaling cannot overcome the overactivation of JNK1 and P38 kinase to rescue RPE cells from cell death. It would be of interest to determine whether there is cross talk between the JNK1/P38 kinase and ERK1/2 signaling pathways in negative control of the survival proteins induced by ERK1/2 in serum-depleted RPE cell cultures. 

RPE cells were treated with 0.03 ng/mL anisomycin to induce the overactivation of both JNK1 and P38 kinase, leading to RPE cell death. At this low concentration, anisomycin did not block protein synthesis in RPE cell cultures, suggesting that the overactivation of both JNK1 and P38 kinase and subsequent cell death occurs through a pathway other than the inhibition of protein synthesis. Anisomycin induced the overproduction of c-Jun in RPE cells, suggesting that c-Jun is involved in anisomycin-induced cell death. This hypothesis was confirmed by recent data showing that low concentrations of anisomycin induce c-Jun activation by Ras-dependent and Ras-independent signaling pathways. ³³ High concentrations of SB202190 also induced the overactivation of both JNK1 and P38 kinase, whereas low concentrations of SB202190 inhibited only P38 kinase. The molecular mechanism by which opposite effects were obtained with the same chemical agents remains to be precisely defined. It is interesting to note that anisomycin also triggered opposite biological effects (pro- and antiapoptotic processes) depending on the concentration. Is this due to the common activation of the JNK1/P38 kinase pathway? This remains to be elucidated. Whatever the molecular mechanism involved in the overactivation of both JNK1 and P38 kinase by high concentrations of SB202190, it also leads to RPE cell death, suggesting that the strong and sustained activation of both the JNK1 and P38 kinase pathways is a signaling mechanism common to different cell death inducers. This confirms recent data showing that the duration of JNK1 activation may determine cell death. Transient JNK1 activation (minutes) leads to cell proliferation, whereas sustained JNK1 activation (hours) causes cell apoptosis. ²⁵ It is therefore possible that the transient activation of JNK1/P38 kinase was sufficient to trigger the cell survival signaling necessary for the induction of RPE cell proliferation by serum, whereas the sustained activation of JNK1/P38 kinase induced by pharmacologic agents contributed to cell death in serum-stimulated RPE cell cultures. It would be of interest to study the specific intracellular targets induced by the persistent and the transient activation of JNK1 and P38 kinase for cell proliferation and cell death equilibrium. The transient activation of ERK1/2 induced by serum could not counterbalance the sustained activation of both JNK1 and P38 kinase induced by anisomycin and SB202190. It would therefore useful to investigate, in the presence of serum, the effects of sustained ERK1/2 activation in the presence of JNK1 and P38 kinase overactivation and to determine whether the levels of activation of JNK1 and P38 kinase are abnormally high in the RPE cells of patients with AMD, a pathologic condition characterized by a loss of RPE cell viability. 

Potential and Specific Targets of the JNK1 and P38 Kinase Signaling Pathways for the Inhibition of RPE Cell Death

The small GTPases Ras and Rho have been shown to play a crucial role in transmitting signals in normal and pathologic situations. Indeed, we showed that the inhibition of Ras completely blocked serum-induced cell proliferation. These data support a key role for Ras in transmitting a serum signal to ERK1/2 for cell proliferation in RPE cell cultures. This seems to exclude Ras as a target in therapeutic strategy to combat RPE cell death, a hallmark of AMD, although other Ras-controlled signaling pathways, such as the PI3 kinase pathway, have been demonstrated to be antiapoptotic. In contrast, Rho GTPases, which control both JNK1 and P38 kinase in RPE cells, are potential targets for selective strategies to inhibit RPE cell death. 

Cyclin D1 production is stimulated by serum in RPE cell culture. ERK1/2 activation is associated with accumulation of cyclin D1, leading to progression to the sense phase, ¹⁴ whereas antisense cyclin D1 induces apoptosis, ²⁶ suggesting a role for cyclin D1 in both cell proliferation and cell survival. In addition, cyclin D1 production is regulated positively by ERK1/2 and negatively by the P38 kinase pathway in normal fibroblasts. ¹⁴ It is also regulated positively by collaborative interactions between ERK1/2, JNK1, and P38 kinase in src-transformed cells, ²⁷ suggesting that the control of cyclin D1 production is under complex regulation. The inhibition of ERK1/2 activation, which reduced RPE cell proliferation, also affected the kinetics of cyclin D1 production in RPE cell cultures. Inhibition of P38 kinase alone had no effect on RPE cell proliferation, suggesting that ERK1/2, but not the P38 kinase pathway, controls the levels of production of cyclin D1 for the induction of RPE cell proliferation. In contrast, overactivation of both JNK1 and P38 kinase induced cell death, but did not affect the production of cyclin D1, suggesting that the JNK1 and P38 kinase pathways do not control cell survival by modifying levels of cyclin D1 production. Our data rule out the negative regulation of cyclin D1 by JNK1 and P38 kinase ¹⁴ through an ERK-independent pathway. ^{34 35} Thus, cyclin D1 does not seem to be a good candidate target for use in a therapeutic strategy to inhibit RPE cell death. However, cyclin D1 production has also been shown to be regulated by an mTOR-P70^S6K-independent PI3 kinase pathway. ³⁶ Thus, if apoptosis is the cell death process involved in AMD, PI3 kinase is a potential target for strategies against RPE cell death and the role of the PI3 kinase pathway in the control of cyclin D1 and in RPE cell death should also be investigated. 

The transcription factor CREB has been demonstrated to be a target in ERK1/2 signaling for the mediation of cell proliferation in many cell types. ³⁷ It is also a potential target in the search for an RPE cell death inhibitor in the JNK1 and P38 kinase signaling. CREB phosphorylation is not affected by the inhibition of ERK1/2 activation in RPE cells, suggesting that the ERK1/2 signaling pathway does not control CREB activation and that CREB activation is not required for RPE cell proliferation (Mascarelli F, unpublished data, 2002). This result may be accounted for by recent data showing that activation of CREB is also regulated by P38 kinase and that the transactivation activities of the CREB transcription factor may require the activation of P38 kinase. ³⁸ It would therefore be of interest to investigate whether the inhibition of CREB may affect RPE cell death. 

c-Myc is also a key regulator of cell proliferation but has recently been demonstrated to be involved in the control of MAPK-regulated cell death. ³⁹ We therefore investigated whether this factor was involved in RPE cell death. We showed that c-Myc was not a target of the overactivated JNK1/P38 kinase signaling pathway, ruling out the use of an anti-c-Myc approach as a strategy against RPE cell death. 

In conclusion, the Rho GTPases/JNK1-P38 kinase pathways are involved in cell death only when overactivated. The cell death signaling pathway is cyclin D1- and c-Myc-independent in RPE cells. Further studies are therefore needed to identify the specific intracellular targets of the JNK1/P38 kinase signaling pathways. Identification of the complete signaling network and potential cross-control regulatory loops may help us to develop selective strategies for the treatment of degenerative diseases involving RPE cells. 

 Supported by Grants 9697 and 4474 from the Association for Cancer Research (FM), Grant 75/01-RS/63 from the League against Cancer (FM), and the Ministry of Education and Research (GL).

 Submitted for publication May 30, 2002; revised August 30, 2002; accepted September 20, 2002.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Analysis of activation of the MKK3/6-P38 kinase module during RPE cell proliferation and of the effects of inhibition and downregulation of P38 kinase on RPE cell proliferation. (A, B) RPE cells were cultured for 3 days and then stimulated with 10% FCS. Cells were lysed at the indicated times, and equal amounts of protein were reduced and subjected to SDS-PAGE and Western blot analysis with anti-active phosphorylated MKK3/6 (A) and P 38 (B) kinase antibodies. (C) RPE cells were treated with the indicated amount of the P 38 kinase inhibitor SB203580, and cell proliferation was analyzed on days 1, 3, and 6. (D) The role of P38 kinase in cell proliferation was analyzed by means of an antisense P38 kinase ODN strategy, and the production of P38 kinase in sense and antisense P38 kinase ODN-treated cells was analyzed by Western blot analysis. Similar results were obtained in three independent experiments.

Figure 2.

 

Analysis of activation of the MKK4-JNK1 module during RPE cell proliferation and effects of JNK1 inhibition on RPE cell proliferation. (A, B) RPE cells were cultured for 3 days and then stimulated with 10% FCS. Cells were lysed at the indicated times, and equal amounts of protein were reduced and subjected to SDS-PAGE and Western blot analysis with anti-active phosphorylated MKK4 (A) and JNK1 (B) antibodies. (C) RPE cells were treated with the indicated amount of the JNK1 inhibitor D-JNKI1, and cell proliferation was analyzed on days 1, 3, and 6. Similar results were obtained in three independent experiments.

Figure 3.

 

Effects of the inhibition of Ras on the activation of ERK2, JNK1, and P38 kinase in FCS-stimulated RPE cell cultures. RPE cells were cultured for 3 days and then stimulated with 10% FCS in the presence or absence of 10 μM FPT III inhibitor. Cells were lysed at the times indicated, and proteins were reduced and subjected to SDS-PAGE and Western blot analysis. Phosphorylated JNK1 (A), P38 kinase (B), and ERK2 (C) were detected with an anti-active JNK1, an anti-active P38 kinase, and an anti-active ERK2 antibody, respectively. The levels of production of JNK1, P38 kinase, and ERK1/2 were analyzed with an anti-JNK1, an anti-P38 kinase, and an anti-ERK1/2 antibody, respectively. Similar results were obtained in three independent experiments.

Figure 4.

 

Effects of the inhibition of Rho GTPases on the activation of ERK2, JNK1, and P38 kinase in FCS-stimulated RPE cell cultures. RPE cells were cultured for 3 days and then stimulated with 10% FCS in the presence or absence of 500 nM toxin B from C. difficile. Cells were lysed at the times indicated, and proteins were reduced and subjected to SDS-PAGE and Western blot analysis. Phosphorylated JNK1 (A), P38 kinase (B), and ERK2 (C) were detected with an anti-active JNK1, an anti-active P38 kinase, and an anti-active ERK2 antibody, respectively. The levels of production of JNK1, P38 kinase, and ERK1/2 were analyzed with an anti-JNK1, an anti-P38 kinase, and an anti-ERK1/2 antibody, respectively. Similar results were obtained in three independent experiments.

Figure 5.

 

Effect of serum depletion on RPE cell death and the activation of both JNK1 and P38 kinase signaling pathways. RPE cells were initially cultured for 3 days and then, for a further period, in the presence or absence of serum (A). The effects of the inhibition of JNK1 and P38 kinase on cell death were analyzed with the JNK1 inhibitor D-JNKI1and the P38 kinase inhibitor SB202190, respectively (B). Activation of JNK1 and P38 kinase (A, B) and cell death (C) were analyzed at the times indicated by Western blot analysis and cell counting, respectively. Phosphorylated JNK1 and P38 kinase were detected with an anti-active JNK1 and an anti-active P38 kinase antibody, respectively. Similar results were obtained in three independent experiments. Data are the mean ± SD and differences between means were analyzed by Mann-Whitney test (*P < 0.05).

Figure 6.

 

Effect of the activation of both JNK1 and P38 kinase on c-Jun production and cell death. Cultures of FCS-stimulated cells were treated with the JNK1 and P38 kinase activators SB202190 and anisomycin and the phosphorylation of JNK1 and P38 kinase (A) and c-Jun (B) and the production of c-Jun (B) were analyzed by Western blot analysis. The effects of the activation of JNK1 and P38 kinase on cell death were analyzed with the JNK1 and P38 kinase activators, SB202190 and anisomycin (C). Similar results were obtained in three independent experiments. Data are the mean ± SD; differences between means were analyzed by the Mann-Whitney test (**P < 0.01 and ***P < 0.005).

Figure 7.

 

Effects of overactivation of JNK1 and P38 kinase and of inhibition of ERK1/2 on production of cyclin D1 and c-Myc in FCS-stimulated RPE cells. RPE cells were cultured in DMEM with 10% FCS for 3 days and then stimulated with 10% FCS in the presence or absence of 10 μM U0126 (B, D) and 0.03 ng/mL anisomycin (A, C). Cells were lysed at the times indicated, and proteins were reduced and subjected to SDS-PAGE and Western blot analysis with specific anti-cyclin D1 (A, B) and c-Myc antibodies (C, D). Similar results were obtained in three independent experiments.

Figure 8.

 

Effect of downregulation of c-Myc on serum-stimulated RPE cell proliferation. The role of c-Myc in cell proliferation was analyzed with the use of an antisense ODN c-Myc strategy, and the production of c-Myc in sense and antisense c-Myc ODN-treated cells was analyzed by Western blot analysis (top) and cell counts obtained at 0, 1, 3, and 7 days of culture (bottom). Similar results were obtained in three independent experiments. Data are the mean ± SD; differences between means were analyzed by the Mann-Whitney test (*P < 0.05).

Figure 9.

 

Schematic representation of the activation of the JNK1 and P38 kinase pathways in FCS-stimulated RPE cells. Serum engages a major pathway involving ERK1/2, which induces RPE cell proliferation (filled arrow). Serum transiently stimulates the activation of JNK1 and P38 kinase, which are not involved in cell proliferation. In contrast, serum depletion is a stress stimulus that engages two major parallel pathways involving JNK1 and P38 kinase (open arrows). Serum depletion or pharmacologically induced overactivation of JNK1 and P38 kinase induce RPE cell death in a cyclin D1- and c-Myc-independent manner. Ras controls activation of NK1 and ERK2, whereas Rho GTPases control the activation of both JNK1 and P38 kinase. The exact role of the serum-induced activation of the JNK1 and P38 kinase pathways remains unknown.