Time-dependent activation of MAPK/Erk1/2 and Akt/GSK3 cascades: modulation by agomelatine

Background The novel antidepressant agomelatine, a melatonergic MT1/MT2 agonist combined with 5-HT2c serotonin antagonist properties, showed antidepressant action in preclinical and clinical studies. There is a general agreement that the therapeutic action of antidepressants needs the activation of slow-onset adaptations in downstream signalling pathways finally regulating neuroplasticity. In the last several years, particular attention was given to cAMP-responsive element binding protein (CREB)-related pathways, since it was shown that chronic antidepressants increase CREB phosphorylation and transcriptional activity, through the activation of calcium/calmodulin-dependent (CaM) and mitogen activated protein kinase cascades (MAPK/Erk1/2). Aim of this work was to analyse possible effects of chronic agomelatine on time-dependent changes of different intracellular signalling pathways in hippocampus and prefrontal/frontal cortex of male rats. To this end, measurements were performed 1 h or 16 h after the last agomelatine or vehicle injection. Results We have found that in naïve rats chronic agomelatine, contrary to traditional antidepressants, did not increase CREB phosphorylation, but modulates the time-dependent regulation of MAPK/Erk1/2 and Akt/glycogen synthase kinase-3 (GSK-3) pathways. Conclusion Our results suggest that the intracellular molecular mechanisms modulated by chronic agomelatine may be partly different from those of traditional antidepressants and involve the time-dependent regulation of MAPK/Erk1/2 and Akt/GSK-3 signalling pathways. This could exert a role in the antidepressant efficacy of the drug.


Background
Regulation of gene expression represents a major component in the mechanism of action of antidepressants [1][2][3]. Converging evidence shows that a common target of chronic antidepressants is a positive modulation of cAMPresponsive element binding protein (CREB), a transcription factor that regulates the expression of several genes involved in the control of neuroplasticity, circadian rhythms, cell survival and cognition [4][5][6][7][8][9]. CREB transcriptional activity is regulated by phosphorylation at Ser133, which is induced by multiple signalling cascades. In particular, it was shown that the calcium/calmodulin (CaM)-dependent and the mitogen activated protein (MAP) kinase cascades have a crucial role in the activation of CREB consequent to chronic antidepressant treatments [10][11][12].
Moreover, recent preclinical biochemical and behavioural evidence suggested an involvement of the Akt/ glycogen synthase kinase-3 (GSK-3) signalling, regulating gene expression through the activation of CREB and other transcription factors, both in the modulation of behaviour and in the mechanism of action of psychoactive drugs [13][14][15]. GSK-3 is a Ser/Thr kinase regulated predominantly in an inhibitory manner through phosphorylation at N-terminal serine residues (Ser 21 in GSK-3α and Ser 9 in GSK-3β) by several protein kinases, including Akt [16].
The antidepressant agomelatine is an agonist of melatonergic MT 1 /MT 2 receptors and an antagonist of 5-HT 2c receptors. The unique and novel pharmacological profile of agomelatine has been found to be effective in the treatment of depressive symptoms, with a rapid stabilisation of circadian rhythms and a favourable tolerability profile. Efficacy of agomelatine in behavioral tests has been previously reported in different animal models [25][26][27][28][29]. Thus far, little is known about the signalling pathways affected by agomelatine downstream of receptor modulation. This is especially interesting with regard to the short half-life (1-2 h) of agomelatine, its mechanism of action and the signalling pathways responsible for the changes in gene expression and therapeutic effect.
In this study, we verified whether agomelatine modulates the time-dependent activity of some major signalling pathways known to be regulated by traditional antidepressants. In particular, the expression and activation (phosphorylation) of CREB, αCaM kinase II, MAP kinase/Erk1/2, Akt and GSK-3β were investigated at both nuclear and cytoplasmic level of hippocampus (HPC) and prefrontal and frontal cortex (PFC/FC) of rats treated for three weeks with agomelatine or vehicle and sacrificed 1 h (6 p.m.) or 16 h (9 a.m.) after the last drug administration.
We found that agomelatine, rather than potentiating CREB-related signalling as traditional antidepressants, was able to partly modulate the time-dependent activation of the MAP/Erk1/2 and Akt/GSK cascades, suggesting that its antidepressant properties might be related with a fine-tuning of the time-dependent oscillations in forebrain intracellular signalling.

Chronic agomelatine does not activate CREB
In order to dissect the effect of chronic agomelatine on the signalling pathways regulating CREB activity and its putative time-dependent modulation, rats were chronically treated with vehicle or agomelatine (once daily for 21 days) at 5 p.m. and sacrificed after 1 h (6 p.m.) or 16 hours (9 a.m). CREB expression and activation (phosphorylation at Ser133) were measured in PFC/FC and HPC nuclear fractions. The two-way ANOVA highlighted no time-related changes in CREB expression or phosphorylation levels between 1 h or 16 h posttreatment and, interestingly, differently from traditional antidepressants [11], chronic agomelatine did not induce any significant changes in CREB expression or phosphorylation ( Figure 1A,B for PFC/FC and Figure 2A,B for HPC).
Chronic agomelatine modulates time-dependent changes of MAPK/Erk1/2 activation in frontal/prefrontal cortex but not in hippocampus Different intracellular signalling cascades, including CaM kinases and MAP kinases, are known to modulate CREB phosphorylation in the action of antidepressants [10][11][12]. To investigate whether the lack of CREB activation after chronic agomelatine was dependent on a lack of activation or a dysregulation of intracellular signalling cascades regulating CREB phosphorylation, we assessed the expression and phosphorylation levels of αCaM Kinase II and MAPK/Erk1/2, in PFC/FC and HPC nuclear fractions following agomelatine treatment.
Interestingly, in both PFC/FC and HPC, we found that expression and phosphorylation levels of αCaMKII were unmodified by both time and agomelatine ( Figure 1C, D for PFC/FC; Figure 2C, D for HPC). On the contrary, the activation of MAPK/Erk1/2, through phosphorylation at Thr 202 /Tyr 204 residues, was found to be significantly lower in animals sacrificed in the morning (9 a.m.), than in the evening (  To assess whether agomelatine can modulate the Akt/ GSK-3 signalling pathway, we measured total expression and phosphorylation levels of Akt and GSK-3β in nuclear and cytosolic fractions from PFC/FC and HPC of rats chronically treated with agomelatine and sacrificed at 6 p.m. or 9 a.m. As described in details below, although the expression levels of Akt and GSK-3β were unmodified by chronic agomelatine, the phosphorylation levels of the kinases were partly and selectively regulated by the drug, depending on the brain area and subcellular fraction analysed (Figures 3, 4, 5, 6). Interestingly, in some cases, chronic agomelatine was able to modulate the activation of Akt/GSK-3 phosphorylation, as shown below.
In nuclear fraction from PFC/FC, the 2-way ANOVA showed a significant effect of time on Akt phosphorylation at Thr 308 , which is decreased in rats sacrificed in the morning (F(1,28) = 17.43 p < 0.001) ( Figure 3). Interestingly, post-hoc test was significant only for vehicle treated rats (Bonferroni post-hoc test: VEH 16 h vs. 1 h −33.40% p < 0.001) ( Figure 3B), suggesting that agomelatine dampens the time-dependent modulation of phospho-Thr 308 Akt. Instead, 2-way ANOVA reported a significant effect of both time and treatment for phospho-Ser 473 Akt, whose levels were decreased by agomelatine and were significantly lower in both groups of animals at 16 h vs 1 h In cytosol from PFC/FC, a significant effect of time was found on phospho-Ser 473 Akt levels (2-way ANOVA: time F(1,28) = 8.33 p < 0.01), while post-hoc comparisons were significant only in vehicle treated rats (Bonferroni posthoc test: VEH 16 h vs. 1 h −15.44% p < 0.05) ( Figure 4C). Moreover, 2-way ANOVA showed a significant effect of time on phosphorylation of GSK-3β at Tyr216 (F(1,28) = 12.54 p < 0.01) and the post-hoc test highlighted a significant increase at 16 h selectively in rats chronically treated with agomelatine (Bonferroni post-hoc test: AGO 16 h vs. 1 h +27.33% p < 0.05) ( Figure 4E). No significant differences were found in the phosphorylation state of both pTyr 308 of Akt and pSer 9 of Gsk-3β in cytosol from PFC/ FC ( Figure 4B and F, respectively).
A different pattern was found in HPC. Indeed, in the nuclear fraction from HPC, while phospho-Thr308 Akt levels were similarly reduced in the morning in all animals  Figure 5C). No differences were found in the phosphorylation state of both pTyr 216 and Ser 9 of Gsk-3β in HPC nuclear fraction ( Figure 5E and F, respectively).
In the cytosolic fraction from the same area, the 2-way ANOVA highlighted a significant effect of time for phospho-Thr 308 Akt levels (time F(1,28) = 6.19 P < 0.01) ( Figure 6B For clarity, all significant changes in time-dependent regulation of MAPK/Erk1/2 and Akt/GSK-3 pathways, in nuclei and cytosol from PFC/FC and HPC of vehicle or agomelatine treated rats, and relative statistics are resumed in Table 1.

Discussion
The study of the effects of agomelatine on intracellular signalling pathways modulating CREB activation demonstrated that, differently from other antidepressants that activate the transcription factor CREB, mainly through phosphorylation by αCaM kinase II and IV and MAPK/Erk1/2 cascades [11], chronic agomelatine did not act on these molecular mechanisms. Indeed, we did not observe any significant modification in both CREB and αCaM kinase II activation in nuclear fraction from PFC/FC or HPC. Our results are in line with those of Morley-Fletcher et al. [30], who recently showed in the same rat strain, that chronic agomelatine, although able to reverse the reduction of CREB phosphorylation induced by prenatal stress in rats, had no effect on CREB phosphorylation in non-stressed animals.
A number of previous studies have shown that both acute and chronic agomelatine increase the expression of brain- derived neurotrophic factor and of other neuroplastic molecules such as fibroblast growth factor, and activity-regulated cytoskeleton-associated protein [31][32][33][34][35], suggesting that agomelatine could exert neurotrophic and antidepressant effects through the activation of molecular mechanisms partly different from those of other drugs [11,27].
Interestingly, we observed time-dependent changes in the activation of MAPK/Erk1/2 kinases in both PFC/FC and HPC. Indeed, we found significantly lower phosphorylation levels of both kinases in animals sacrificed in the morning as compared to those sacrificed in the evening.
A circadian modulation of MAPK/Erk1/2 activity was recently demonstrated in hippocampus and shown to contribute to memory consolidation [36][37][38]. Thus, according to our present results, time-related fluctuations in nuclear Erk1/2 activation are present also in PFC/FC, where chronic treatment with agomelatine seems to partly modulate the physiological changes. Further studies are needed to better clarify the circadian nature of these modifications and the potential of agomelatine in the modulation of the MAPK/ Erk1/2 signalling pathways.
We also observed a time-dependent modulation of Akt and GSK-3β activation through phosphorylation that was selectively different depending on the brain area and the subcellular fraction analysed. In particular, we found that Akt phosphorylation was lower in PFC/FC and HPC of animals sacrificed in the morning (16 h after the last injection). This down-regulation of the kinase activation is particularly evident in the nuclear fraction, suggesting a possible involvement in the regulation of gene transcription. Moreover, since the Ser 9 of GSK-3β is phosphorylated by Akt, the reduced phosphorylation of  GSK-3β at Ser 9 observed in nuclei from PFC/FC could be linked to the morning decrease of Akt activity observed in the same fraction. Although previous studies have shown a circadian modulation of Akt and GSK-3 activity, the functional role of these fluctuations is poorly understood. Genetic manipulations, leading to increased Akt activity in neurons of the suprachiasmatic nucleus, lengthen the circadian period of locomotor activity, whereas the reduction of Akt signalling shortens it [39]. Moreover, it was recently demonstrated that both genetic and pharmacological reduction of GSK-3 activity in mouse embryonic fibroblasts have a specific effect on the circadian transcriptional oscillation, showing a phase delay in the transcription of the clock gene mPer2 [40].
Our data suggest that chronic treatment with agomelatine reduces phospho-Akt at Ser 473 in nuclear fraction of PFC/FC. More interestingly, chronic agomelatine was also shown to partly modulate the time-dependent regulation of Akt and GSK-3β activity. Indeed, agomelatine treatment dampened the morning reduction of Akt phosphorylation at Thr 308 in nuclear fraction and at Ser 473 in cytosol of PFC/FC. Moreover, agomelatine induced a significant increase in pTyr 216 GSK-3β in cytosol from PFC/FC and a decrease of pSer 473 Akt in nuclei from HPC 16 h after the last drug injection. Finally, in the cytosolic fraction from HPC of rats sacrificed in the morning, agomelatine dampened the phosphorylation levels of Akt at Thr 308 and induced a significant decrease of pSer 473 Akt.
Agomelatine treatment for 21 days was previously described to induce an increase in Erk1/2, Akt and GSK-3β phosphorylation after chronic treatment in whole hippocampal extract from HPC of rats sacrificed 16h after last drug administration [31]. The following factors might contribute to explaining the apparent contrasting findings compared to our present results: 1-different rat strains were used, Sprague Dawley rats in our study vs. Wistar rats in the previous study; 2-different antibodies were used; 3-the cellular fractions analysed in the two studies were different (nuclear and cytosol vs total extract). In this regard, it is well known that the subcellular distribution and compartmentalization of ERK-MAPKs between cytosol and nucleus play a key role in the regulation of activity and specificity of the kinases [41,42].

Conclusion
In summary, we have shown that chronic treatment with the antidepressant agomelatine does not activate CREB and CREB-related signalling in nuclei and cytosol from both HPC and PFC/FC of naïve rats. This is of particular interest because suggests that the antidepressant effect of agomelatine could be mediated by different molecular mechanisms with respect to classical antidepressants. Although further studies are warranted to identify the different pattern of intracellular signalling pathways modulated by agomelatine, this difference might contribute to the therapeutic benefits.
Moreover, although further studies are required to reveal whether MAPK/Erk1/2 and Akt/GSK-3 signalling pathways could play a role in the circadian control in HPC and PFC/FC, the present work has provided clear evidence of a time-dependent modulation of these pathways in both brain areas. Finally, considering the effects of agomelatine on the time-dependent regulation of Erk1/2, Akt and GSK-3β phosphorylation levels, it can be speculated that this could be a target in the antidepressant effect induced by the drug.

Animals
Twenty-four male Sprague-Dawley rats (175-200 g) were purchased from Charles River (Calco, Italy). Animals were kept at constant temperature (22°C) with a regular 12 h light/dark cycle (light-on at 7 a.m.). The rats were housed in stable groups of four compatible individuals, in 800 cm 3 cages, with sawdust bedding, adequate environmental enrichment and ad libitum access to food and water. The wellbeing of all animals was daily monitored.
All animal procedures were conducted according to current regulations for animal experimentation in Italy (Decreto Legislativo 116/1992) and the European Union (European Communities Council Directive 2010/63/EU), were approved by the Italian Ministry of Health (Decreto Legislativo 295/2012-A) and the whole study adheres to the ARRIVE guidelines.

Drug treatments
Rats were randomly divided in 4 experimental groups (8 animals each): 1-Chronic vehicle (hydroxyethylcelullose 1%, 1 ml/kg, i.p.) sacrificed 1 h after the last injection; 2-Chronic vehicle sacrificed 16 h after the last injection; 3-Chronic agomelatine (40 mg/kg i.p.) sacrificed 1 h after the last injection; 4-Chronic agomelatine sacrificed 16 h after the last injection. Treatments were given once a day for 21 days, at 5.00 p.m. (2 h before the start of the dark cycle, 7 p.m.) The HPC and PFC/FC were quickly dissected on ice and processed for subsequent experiments.

Preparation of subcellular fractions for signalling studies
Individual HPC and PFC/FC were homogenized in 10 volumes of 0.28 M sucrose buffered at pH 7.4 with Tris, containing 20 mM NaF, 5 mM Na 2 H 2 P 2 O 7 , 1 mM Na 3 VO 4 (protein phosphatase inhibitors), and 2 μl/ml of protease inhibitor cocktail (Sigma-Aldrich S.r.l., Milan, Italy), using a glass-teflon tissue grinder (clearance 0.25 mm) in order to obtain the total homogenate fraction [11]. The homogenate was further centrifuged (5 min, 1000 g at 4°C) and the resulting pellet (nuclear fraction) was resuspended in lysis buffer (120 mM NaCl, 20 mM HEPES pH 7.4, 0.1 mM EGTA, 0.1 mM DTT, containing 20 mM NaF, 5 mM Na 2 H 2 P 2 O 7 , 1 mM Na 3 VO 4 , and 2 μl/ml of protease inhibitor cocktail). To obtain the cytosolic fraction, the supernatant of the 1000 g centrifugation was ultracentrifuged at 135,000 g for 1 h. Each fraction was immediately stored at −80°C after being obtained. Protein content of subcellular fractions was evaluated by using the BCA Protein Assay (Thermo Fisher Scientific SAS, Illkirch Cedex, France). Purity of subcellular fractions was checked by measuring subcellular distribution of protein markers, as previously shown [43].

Western blot analysis
Western blot analysis was carried out as previously described [11,44], by incubating PVDF membranes, containing electrophoresed proteins from either nuclear or cytosolic fractions, with antibodies for CREB (Cat#   and β-actin 1:5000 (Sigma-Aldrich Cat# A1978 RRID:AB_476692). Following incubation with peroxidase-coupled secondary antibodies, protein bands were detected by using ECL (GE Healthcare Europe GmbH, Milano, Italy). All protein bands used were within linear range of standard curves, and both total expression and phosphorylation levels were normalized for β-actin level in the same membrane. Standardization and quantitation were performed with Quantity One software (BioRad Laboratories S.r.l., Segrate, Italy).

Statistical analysis
Two-way ANOVA was employed for the analysis of the experiments, with treatment and time from the last administration as independent factors. When appropriate, further differences were analyzed by Bonferroni post-hoc test. Significance was assumed at p < 0.05. Statistical analysis of the data was carried out using GraphPad Prism4 (GraphPad Software Inc., USA). For the sake of simplicity in the graphs were reported only the significant Bonferroni post-hoc comparisons' p values; the complete results of the 2-way ANOVA are described in the text. Data are presented as means ± standard error (SEM), with each individual group composed of 8 samples. For graphic clarity, optical densities from experimental groups were expressed and presented as a mean percentage of the vehicle treated group.