Modulation of c-Jun N-Terminal Kinase Signaling and Specific Glucocorticoid Receptor Phosphorylation in the Treatment of Major Depression

Abstract

Glucocorticoid resistance is a common finding in major depressive disorder. Increased glucocorticoid receptor (GR) phosphorylation at serine 226 is associated with increased glucocorticoid resistance. Previously we have demonstrated that depressed patients exhibit higher levels of GR phosphorylated at serine 226 compared to healthy controls. The enzyme that is involved in this specific GR phosphorylation is c-Jun N-Terminal Kinase (JNK). We propose that modulation of glucocorticoid phosphorylation at serine 226, by targeting JNK signaling pathway, could be a potential strategy for antidepressant treatment. We base this assumption on the results of previous research that examined GR phosphorylation and JNK signaling in animal models and human studies. We also discuss the potential challenges in targeting JNK signaling pathway in depression.

Introduction

The etiology of depression is still poorly understood, and present treatment strategies fail to alleviate the symptoms of 10–40% of depressed individuals [1]. Current trends in mood disorders research highlight the need of distinguishing between the specific syndromes that fall under the wide “umbrella” of major depression, in order to personalize treatment and achieve better therapeutic response. This notion found support in basic research, which demonstrated significantly different gene expression profiles between stress-related depression models and endogenous depression models [2].

Dysregulation of the hypothalamic-pituitary-adrenal (HPA) axis has long been associated with stress-related depression [3]. Namely, a significant percentage of depressed patients exhibit increased blood cortisol levels, and this effect is higher in those with comorbid anxiety [4, 5]. Effects of cortisol are mediated through the glucocorticoid receptor (GR), including the feedback regulation of the HPA axis. Therefore, the impairment of GR signaling is suggested to play a pivotal role in the HPA axis dysregulation in stress-related depression [6, 7].

The GR regulates the expression of a variety of genes, including ones that are involved in metabolism and immunity, neuronal survival, neurogenesis and regulation of HPA axis [8]. Therefore, in addition to its role in the neuroendocrine system, GR signaling is strongly involved in immune system response. Cortisol is one of the most potent anti-inflammatory hormones in the body, and pathways coupled with glucocorticoid resistance may conspire during chronic stress to contribute to chronic activation of inflammatory responses [9]. Namely, a large body of research supports the notion of inflammatory disturbance in depression [10, 11]. Chronic stress, through glucocorticoid resistance (i.e. decreased sensitivity to anti-inflammatory effects of glucocorticoids), may contribute to chronic activation of immune system in major depressive disorder [12].

Considering the complex interplay between glucocorticoid and inflammatory signaling (in both physiological and pathological conditions), as well as evidence of dysfunction of both these systems in major depression, targeting one or both systems could be a potential therapeutic strategy for this disorder [13].

1.1. The glucocorticoid receptor signaling

GR is a ligand-dependent transcriptional factor that resides in the cell cytoplasm in its inactive form. After ligand binding it translocates to the nucleus where it interacts with specific DNA sequences and modulates gene expression [8]. For instance, GR stimulates the transcription of genes related to anti-inflammatory and metabolic actions, but also interacts with other transcrptional factors involved in expression of specific genes. For example GR inhibits the activity of NF-κB, a transcriptional factor that promotes inflammation and cell proliferation [14].

The activity of the GR can be modulated by numerous cellular factors, which act either to stimulate or inhibit the GR-mediated transcription of specific genes. For instance, FKBP51, a co-chaperone of heat shock protein 90, modulates GR sensitivity by decreasing GR’s affinity for cortisol. Increased expression of FKBP51 is related to increased glucocorticoid resistance, increased recurrence of depression and better treatment response [15]. The significance of FKBP51 in the vulnerability to depression is highlighted by considerations of targeting this protein in the treatment of major depressive disorder [16].

Specific kinases which phosphorylate the GR can also alter the receptor conformation and thereby change its transcriptional activity [17]. There are several amino-acid residues on the human GR that can be phosphorylated, with serines 203, 211, 226, and 404 being the most characterised [8, 17]. The focus of this paper is the phosphorylation of GR at serine 226 (pGR-226), due to its potential contribution to glucocorticoid resistance of the cell. Namely, phosphorylation of human GR at serine 226 was found to blunt hormone signalling by enhancing nuclear export of the GR [18]. The enzyme involved in this specific GR phosphorylation is the c-Jun N-terminal kinase (JNK) [19], a member of the mitogen-activated protein kinase (MAPK) family.

1.2. The JNK signalling cascade

JNK proteins (also known as stress-activated protein kinases - SAPKs) include at least 10 isoforms that are transcribed from 3 genes, JNK1, JNK2 and JNK3, and which are differently expressed in specific tissues. The JNK signalling pathway is complex, and implicated in diverse cellular functions, such as proinflammatory cytokine signalling, cellular death, proliferation and survival [20]. The JNK signalling cascade consists of a three-level module of upstream kinases that transfer the signals from the cell surface to intracellular effectors. When the first one in line, the MAP kinase kinase kinase (MAP3K) is activated, it phosphorylates the MAP kinase kinase (MAP2K), which then phosphorylates and activates the JNK isoforms. Once activated, the JNK can phosphorylate the serine or threonine residues on specific substrates (such as different oncogenes and transcription factors) [21]. The JNK signalling pathway has been the focus of extensive research in the fields of various inflammatory diseases and cancer, where glucocorticoids represent a valuable treatment option. Long-term glucocorticoid treatment may lead to glucocorticoid resistance, resulting in reduced anti-inflammatory and pro-apoptotic response and presenting a severe obstacle to recovery [22], and increased JNK activation has been demonstrated in steroid unresponsiveness [23]. However, JNK pathway has been associated with both pro-apoptotic and proliferative activity (depending of the context, cell type and JNK isoforms) [24]. Therefore, finding a way to decrease glucocorticoid resistance or modulate JNK activity has been an important goal in these fields of medicine.

Concerning the research of depressive disorders, JNK signalling has received significantly less attention, despite the fact that JNKs modulate GR function [25] and therefore could contribute to glucocorticoid resistance and increased inflammatory response, commonly found in a substantial subgroup of depressed patients.

1.3. GR phosphorylation in depression

GR phosphorylation at serine 226 inhibits GR’s transcriptional activity and enhances the nuclear export of the receptor [18], thus possibly contributing to glucocorticoid resistance. We have recently demonstrated that chronic stress induced by social isolation in rats increased the levels of hippocampal GR phosphorylated at serine 246 (analogous to 226 in humans) and that antidepressant treatment reduced the levels of this parameter [26]. Moreover, we demonstrated that depressed patients have increased levels of pGR-226 compared to healthy controls [27]. Particularly interesting are the findings we observed in healthy women – levels of pGR-226 were positively correlated to neuroticism, personality trait associated with negative bias in attention, interpretation and recall of information, increased reactivity and ineffective coping, which is closely related to vulnerability to depression [28].

In the following hypothesis we propose a mechanism of antidepressant action that concentrates on the modulation of GR phosphorylation and JNK signaling. We base our hypothesis on the current knowledge on JNK activity and on our recent findings on GR phosphorylation in depression.

The Hypothesis/Theory

Inhibition of glucocorticoid receptor phosphorylation at serine 226 may lead to symptomatic improvement in stress-related depression by stimulating GR transcriptional activity and decreasing glucocorticoid resistance.

This inhibition can be achieved by targeting the JNK signalling pathway: by acting on the JNK itself, or further up, by modulating the activity of upstream kinases.

Evaluation of the hypothesis

In the attempt to evaluate the hypotheses we will:

• discuss the basic and clinical studies that explored the role of JNK pathway in in stress response and inflammation, considering that both systems are relevant for major depression

• review possible targets for the inhibition of JNK signalling (Figure 1)

• underline the limitations and challenges of JNK signalling inhibition

Figure 1.

Simplified representation of the JNK signaling pathway in relation to the GR phosphorylation. Two kinases in the three-level module (MAP2K, JNK) are considered as potential targets for inhibition.

2.1. Studies that explored the role of JNK in stress-related depression

Only a few studies have investigated the association of JNK signalling with stress-related psychiatric disorders, and most of them were based on animal models. Moreover, most of the studies focused on the effects of acute stress, even though chronic stress is also an important risk factor for depression [29].

Shen et al. demonstrated that acute stress in rats significantly and strongly increased the levels of MAP2K4 (the immediate upstream activator of JNK) in several brain areas, while the increase of active JNK was present to a lesser extent [30]. This finding is particularly interesting since it points out to a potential target of the JNK pathway other than the JNK itself. Clarke et al. found that a JNK antagonist reversed certain behavioural and neurochemical effects of acute stress in mice. However, in basal conditions JNK antagonist expressed stressor-like monoamine and behavioural changes [31], suggesting that JNK activity has different effects depending on the context (basal/stress). Galeotti and Ghelardini demonstrated acutely stressed mice had a significant JNK activation in the cortex, and JNK inhibition resulted in antidepressant-like behaviour [32].

Concerning chronic stress, Li et al. found that traditional herbal medicine with antidepressant properties improved depressive-like symptoms of stressed rats, which was accompanied by inhibited expression of JNK in the hippocampus [33]. In our studies we observed significant gender differences in JNK activation in rat hippocampus [26, 34]. Basal JNK levels were higher in females compared to males, and chronic stress decreased JNK levels in females while having the opposite effect in males. Moreover, in males, the increase of activated (phosphorylated) JNK was related to the increase of GR phosphorylation at serine 246 in response to chronic stress, which was not the case in females. This suggests that in females the nuclear pGR-246 could be targeted by some other kinases other than JNK, implying for a consideration of gender-specific antidepressant treatment.

To our knowledge, there are no human studies of JNK signalling in stress-related depression. A study by Spiliotaki et al. demonstrated that patients with chronic bipolar depression had significantly lower levels of lymphocyte JNK1 compared to healthy controls, while in the remitted bipolar patients JNK1 levels were in between the two groups [35]. Also, our preliminary unpublished data showed that patients with acute episode of major depression had lower levels of JNK1 in nuclei of lymphocytes from peripheral blood. The major limitation of these studies is the absence of drug free patients, preventing the conclusion if the changes of JNK levels are due to the disease itself or treatment.

Besides altering the stress response, modulation of JNK pathway could be beneficial in the inflammation-related phenomena involved in the vulnerability to depression, either by reducing GR resistance (and stimulating glucocorticoid-medicated anti-inflammatory actions) or by other pathways. For instance, JNK inhibition was demonstrated to inhibit inflammatory response in vitro, e.g. through inactivation of NF-κB [36], or by reducing oxidative stress through COX-2 inhibition [37]. JNK inhibitors were also beneficial in reducing inflammation in animal models of colitis [38], ischemic brain damage [39] and rheumatoid arthritis [40].

2.2. Potential targets for JNK signalling pathway inhibition

The most extensively studied JNK targets are the ATP-competitive JNK inhibitors. They are aimed at the highly conserved ATP-binding site and act by blocking all JNK isoforms. Also, at higher doses they may block p38 signalling. Therefore, the lack of specificity and the potential to triger off-target effects could hinder their use in the clinical setting [21].

Another option is targeting the docking sites of JNK substrates, or the regulatory protein site of the JNK. For example, efforts are focused on the JNK-interacting proteins (JIPs). JIPs are scaffold proteins that mediate signaling in the JNK pathway and they are more specific for particular isoforms of JNKs [24]. However, further research and development is needed before they can be considered for clinical application.

Inhibiton of upstream kinases could potentialy overcome the toxic side effects of direct JNK inhibition. Namely, MAP2K is proposed as a possible target for inhibition, especially since certain studies demonstrated the effect of stress on the expression of these kinases [30]. Several organic compunds found in soy and plant pigment Cyanidin have been demostrated to inhibit MAP2K4 (a member of MAP2K family), but their acitivity is yet to be explored in models of human disorders [24].

Another possible pharmacological target in the GR signaling pathway are the phosphatases that revert the action of kinases by dephosphorylating the GR. Althoug the topic is beyond the scope of this paper, it is noteforwthy of mentioning since phosphatases have been discussed as potential therapeutic targets in inflamatory diseases. However, the general conclusion remains that further research and a better understanding is needed before the translation to the clinical setting can be made [41].

2.3. Limitations and challenges of JNK signalling inhibition

Although JNKs are implicated in a plethora of diseases and disorders, they are crucial for a variety of cellular functions. Hence, the main challenges in targeting the JNK pathway are the potential side effects that could arise from its inhibition, one of the most serious being the induction of tumour formation [42]. Current inhibitors of JNK signalling demonstrated potential benefits for a variety of disorders in in vitro and animal studies [e.g. 40, 43, 44], although their safety profile needs to be explored and modified in order to progress through clinical trials. New inhibitors that would aim at different components of the JNK pathway and obtain higher specificity could potentially overcome these limitations. On the other hand, further research is needed to find out if and which one of different JNK isoforms has predominant role in contributing to depressive symptoms. For example, selective activation of JNK2/3 during neuronal stress was demonstrated in vitro, further adding to the evidence of cell and tissue specificity of JNK singling [45]. Therefore, it should be investigated which JNK isoform and upstream cascade is involved in increased phosphorylation of GR at serine 226 in depression, with special focus on brain structures relevant to depression.

Finally, JNK signalling inhibition cannot be beneficial for the whole anxiety-depression spectrum disorders, since the stress-related conditions might be associated with decreased levels of blood and urinary cortisol as well. For subjects who manifest affective dysregulation associated with hypocortisolemia or greater cortisol suppression after dexamethasone (e.g. some cases of post-traumatic stress disorder [46]), approach to HPA regulation could be putatively different.

Implications of the hypothesis

It is unlikely that the drugs targeting the JNK pathway would be effective in all cases of depressive disorders. Since the reviewed research suggests that JNK activity is dependent of the context and gender, we can hypothesize that a certain population of depressed patients, for example those with a more pronounced glucocorticoid resistance could be more suitable for the JNK related treatment. Specifically, male gender, the presence of psychological stressor(s), higher levels of anxiety associated with specific biological parameters (increased cortisol levels, higher levels of GR phosphorylated at serine 226, or higher JNK expression) could potentially predict a better response to JNK inhibition treatment, if one should arise from current scientific efforts.

It would be of great interest to further explore the changes in the JNK pathway in relation to symptomatic improvement and treatment response in depressed patients. Although there is optimism regarding the JNK-related treatment strategies, the possibility of their use as biomarkers of depression subtype - specific biology could be more feasible in the foreseeable future.