Heterotrimeric G Proteins: Insights into the Neurobiology of Mood Disorders

Abstract

Mood disorders such as major depression and bipolar disorder are common, severe, chronic and often life-threatening illnesses. Suicide is estimated to be the cause of death in up to approximately 10-15% of individuals with mood disorders. Alterations in the signal transduction through G protein-coupled receptor (GPCR) pathways have been reported in the etiopathology of mood disorders and the suicidal behavior. In this regard, the implication of certain GPCR subtypes such as α_2A-adrenoceptor has been repeatedly described using different approaches. However, several discrepancies have been recently reported in density and functional status of the heterotrimeric G proteins both in major depression and bipolar disorder. A compilation of the most relevant research topics about the implication of heterotrimeric G proteins in the etiology of mood disorders (i.e., animal models of mood disorders, studies in peripheral tissue of depressive patients, and studies in postmortem human brain of suicide victims with mood disorders) will provide a broad perspective of this potential therapeutic target field. Proposed causes of the discrepancies reported at the level of G proteins in postmortem human brain of suicide victims will be discussed.

INTRODUCTION

Neither the neurobiology of mood disorders nor the mechanisms of action of antidepressants are completely understood to date, even if knowledge is progressively accumulating [135]. The essential feature of this group of disorders is a disturbance of mood, accompanied by a manic or depressive syndrome. Mood disorders include major depression and bipolar disorder, which are distinguished by whether or not there has been a manic episode [1].

The diagnostic criteria for major depressive episode include a dysphoric mood or loss of interest or pleasure in all or almost all usual activities, insomnia or hypersomnia, psychomotor agitation or retardation, loss of interest or pleasure in usual activities, fatigue, feelings of worthlessness, and suicidal ideation among others. The manic episode is characterized by a predominantly elevated, expansive, or irritable mood. The elevated or irritable mood must be a prominent part of the illness and relatively persistent, although it may alternate with depressive mood.

Alterations in the signal transduction through G protein-coupled receptor (GPCR) pathways [151] [72] have been reported in the etiopathology of mood disorders [117,118, 173]. The implication of certain GPCR subtypes and, among them, α_2A-adrenoceptors [52,62,69,126], has been repeatedly described using different approaches. Therefore, understanding the possible alterations in the signaling pathways modulated by these GPCR subtypes may open a new therapeutic field. Heterotrimeric G proteins have been described as the main signaling pathway downstream receptor activation (see below). In this context, several discrepancies have been recently reported in density and functional status of the heterotrimeric G proteins both in major depression and bipolar disorder [49,57,62,69,149, 186]. The primary focus of this review will be to summarize recent advances in the study of the molecular biology of heterotrimeric G proteins. The most relevant research topics about the implication of heterotrimeric G proteins in the neurobiology of mood disorders (i.e., animal models of mood disorders, studies in peripheral tissue of depressive patients, and studies in postmortem human brain of suicide victims with mood disorders) will be discussed, providing a broad perspective of this potential therapeutic target field.

MAJOR DEPRESSION AND SUICIDE

Suicide is a leading cause of death [92]. According to recent figures from Centers for Disease Control and Prevention (CDC) in the United States in 2002, suicide was the eleventh most common cause of death (in 1998 it was the eight most common cause) [110]. One of the first studies showing a clear relation between suicide and psychiatric diseases reported that 45% of the suicide victims suffered mood disorders and 23% alcoholism [154]. Thus, several researchers have demonstrated that depression, alcoholism, drugs of abuse, and schizophrenia are the psychiatric pathologies more commonly related to suicide [32,87,119,158].

Depression is itself the principal cause of suicide. In this regard, mood disorders seem to be present in at least 50% of all suicide victims [14,78,79]. Reciprocally, it has been estimated that 60-70% of depressed patients experience suicidal ideation, and that 10-15% of depressed patients commit suicide [130]. These data indicate that close monitoring for suicidal ideation is imperative in patients with mood disorders.

MONOAMINERGIC THEORY OF DEPRESSION

The molecular research about mood disorders have screened for a biological substrate [185]. Several hypotheses have linked the pathogenesis of depression to alterations in the neurotransmission systems in the central nervous system. During the last four decades, alterations in the serotonergic and/or noradrenergic neurotransmission have been related to mood disorders. This monoaminergic theory of depression was based on several reports [27,35,162]. Reserpine, an antihypertensive drug, induced depressive disorders related to catecholamine depletion; whereas iproniazide, a monoamine oxidase inhibitor, was associated with hypomanic episodes in some subjects. The first antidepressant drugs developed in the fifties were reported to inhibit either reuptake or degradation of catecholamines. In this regard, several studies using depletion of neurotransmitters have shown a relationship between alterations in serotonin and catecholamines and mood disorders [2,15,16,23,44,170].

The etiopathogenesis of mood disorders is not fully explained by the monoaminergic theory [109]. For instance, the temporal delay between the acute effect of antidepressants on the neurotransmitter levels and the clinical improvement is not understood, whereas several pharmacological approaches have been reported [5,6,20].

GPCRs are involved in cellular responses to the majority of hormones and neurotransmitters, and represent an enormously significant target for drug discovery. Alterations associated with depressive disorders of certain GPCR subtypes, such as α_2A-adrenoceptor [30,69,164], 5-HT_1A serotonin [4,17,82,160], μ-opioid [52,58,73] and CB₁ cannabinoid [85] receptor, both in terms of density and functionality, have been reported using different approaches. Heterotrimeric G proteins are described as the main signaling pathway downstream GPCR. In this regard, several scientific groups have screened for alterations in heterotrimeric G proteins among the possible molecular causes underlying depressive disorders.

HETEROTRIMERIC G PROTEIN SIGNALING

G Protein-Coupled Receptors

The signaling molecules that control and regulate cellular activity include hormones, neurotransmitters, small peptides and proteins, ions, and lipids, as well as sensory stimuli such as odorants, pheromones, and photons. These signaling molecules bind to receptors molecules inducing specific cellular responses. The cellular receptors have been classified according to their general effector mechanisms. This functional classification describes at least three general types of cell surface receptors: ion-channel receptors, enzyme-associated receptors, and G protein-coupled receptors (GPCRs) [89,136].

The name GPCR refers to diverse heptahelical proteins that mediate the cellular signaling via heterotrimeric G proteins [72,100,151,175]. The first GPCRs cloned in the mid 1980s were the visual pigment opsin [134], and the β-adrenergic receptor [45]. Since then, hundreds of pharmacologically distinct GPCRs have been described [56,71,81].

GPCRs have an extracellular amino terminus, seven α-helical transmembrane domains, which form the transmembrane core, three extracellular loops, three intracellular loops, and an intracellular carboxy terminus. Each of the seven transmembrane domains is formed by 20-27 amino acids. In different GPCRs, the amino terminus (7-595 amino acids), loops (5-230 amino acids), and carboxy terminus (12-359 amino acids) vary considerably in length [71,152,174, 175]. The first crystal structure of any GPCR, the structure of bovine rhodopsin, has been recently solved at 2.8 Å resolution [150].

The most widely used classification of neurotransmitter/hormone receptors has been proposed by the International Union of Pharmacology (IUPHAR) [56,84]. The three major subclasses include the rhodopsin-like receptors (subclass I), the glucagon-related receptors (subclass II), and metabotropic glutamate-related receptors (subclass III) [63,88,89, 184]. While many GPCRs have been found to also couple by G protein-independent mechanisms [26,101], the interaction with heterotrimeric G proteins is the major signaling machinery of these proteins [18,25,76].

Heterotrimeric G Proteins

Heterotrimeric G proteins were reported for the first time in the 1970s by Rodbell [111,155] and Gilman [114]. The “nucleotide binding protein necessary to reconstitute the stimulation of adenylate cyclase” was purified by Gilman’s laboratory [139].

Monomeric G proteins (small G proteins) [113,172], and heterotrimeric G proteins [24,172] have been described as the two main families of signal transduction-related proteins that bind and hydrolyze guanine nucleotides. Heterotrimeric G proteins are formed by three subunits (α, β, and γ). The β and γ form a complex that do not dissociate under physiological conditions. Heterotrimeric G proteins are soluble intracellular proteins that are bound to the plasma membrane through covalent links to fatty acids [182]. G protein activation modulates classical downstream effectors, such as adenylate cyclases [77], phospholipases [90], and ionic channels [39] through well-characterized molecular mechanisms [65,138,166].

Heterotrimeric G proteins are classified according to the G_α subunits. Mammals have more than twenty different described G_α subunits, encoded in seventeen different genes [168]. They are classified in four families according to their homology in their primary structure: G_αs, G_αi, G_αq, G_α12, each family with several subtypes. These G protein subtypes are extremely conserved in evolution.

With the exception of heterotrimeric G proteins expressed in sensory organs (G_αt, G_αgust and G_αolf), and certain subtypes found in hematopoietic (G_α16) or nervous (G_αo) cells, most α-subunits are ubiquitous. In this regard, each cellular subtype expresses at least four or five G_α subtypes. In certain tissues, heterotrimeric G proteins are expressed at high concentrations. Thus, G_αo may represent 12% of the membrane protein in brain [80].

Six different G_β subunits have been described in mammals, expressed in six different genes [33]. These G_β subunits present a 53-90% homology. Twelve different G_γ subunits, which are encoded on twelve different genes, show a high heterogeneity [28]. Although all the possible combinations of G_β and G_γ subunits could form up to seventy two G_βγ dimers, not all the combinations are expressed in vivo. The functional significance of the diversity in the G_βγ dimmers is not yet understood. In this regard, most G_βγ dimers (except for β₁γ₁ expressed in the retina) present similar functional properties.

Molecular Mechanism of Receptor-G Protein Coupling

The diversity of ligands activating GPCRs, as different as a photon of light and a 40 kDa protein, is reflected in the variety of mechanisms of ligand binding, such as the simplest mechanism for a ligand to activate a receptor binding the transmembrane core (norepinephrine, serotonin), or the protease ligand thrombin that cleaves the amino terminus of the receptor [21,91].

The most accepted molecular pharmacological model to describe GPCR activation is the ternary complex model (i.e., agonist-receptor-G protein) [40]. The ternary complex model proposes that the receptor exists in an equilibrium between two conformational states: the functionally inactive (R) and the functionally active (R*) state. This model assumes the classic pharmacological models, in which, in the absence of agonist, there are not detectable signal transduction pathways downstream the receptor. However, it has been demonstrated that, under certain conditions, the R* state (and the G protein activation) may be detected in the absence of agonist [178]. Therefore, the pharmacological model has been extended describing the receptor-G protein coupling in the absence of agonist activation, and the basal functional activity of the signal transduction pathways [99,159]. According to this pharmacological model, the so-called inverse agonists, which are able to decrease the basal activity of the receptors, have been identified [133]. Thus, full and partial agonists bind R* with higher affinity than R, shifting the equilibrium to the activated state, whereas inverse agonists bind R with higher affinity, shifting the equilibrium to the inactive state. This equilibrium is not shifted by the antagonists because they present equivalent affinity for the inactive R and the active R* functional states of the receptor. Several reviews of the inverse agonism have appeared [19,31,41,128,176].

HETEROTRIMERIC G PROTEINS AND MOOD DISORDERS

Several groups have suggested that the heterotrimeric G proteins might be altered in mood disorders. This possibility was first raised in a study where GTP binding to brain membranes was performed in rat chronically treated with lithium [7]. Several works have been published showing alterations in heterotrimeric G proteins in both psychiatric [10,51,83,115], and neurodegenerative [22,36–38,55,70,108, 180] disorders. In this regard, the possible role of heterotrimeric G proteins in the pathophysiology of mood disorders, as well as the G protein signaling cascade as a molecular target for antidepressant therapy, has been mainly studied in three research lines: effects of antidepressant treatments in animal models, studies in peripheral tissue of depressive patients, and studies in postmortem human brain of suicide victims with mood disorders.

Effect of Antidepressant Treatment in Animal Models

Drug discovery in depression has been limited by the lack of a universally accepted animal model that can be used to screen for antidepressant effects. Although there are several animal models that reproduce some features of depression, their relevancy to the specifically human disorder major depression has been debated [185].

Several works have reported that the antidepressant drugs modulate the density of different heterotrimeric G proteins [103,105], as well as their functional response [127,143,44,179] (Table 1). Experimental evidences have shown that lithium attenuates the GPCR-induced modulation of the second messengers, even in the absence of changes in the density of the receptor [117]. It has been consistently reported that chronic lithium affects G protein function [7,34,48,104,106,107,124,181]. The Mg²⁺ binding site in the G_α subunits has been proposed as one of the possible molecular targets for lithium [9,93,94].

Table 1.

Summary of Studies of the Effect of Antidepressant Treatment on Heterotrimeric G Proteins in Animal Models

Chronic antidepressant	Brain region	Finding in animal model	Citation
TCA (imi, clo, dsp) and clg	FCx, HT, HC, Str, DR and LC	↓Gαs and Gαi with TCA and clorgyline	[103]
TCA (imi, dsp) and ECS	Limbic areas	↓cAMP synthesis in vitro	[179]
TCA (amtp, dsp, imi, iprl) and ECS	Cx and HT	↑cAMP synthesis activating G proteins	[127]
dsp	Cx	↓β-adrenoceptor-mediated cAMP synthesis	[143]
dsp	Cx	↓cAMP synthesis activating G proteins and AC	[144]
TCA (imi, clo, dsp), flx and clg	FCx, HT, HC, Str, DR, LC and ME	G protein density: ↓Gαs with imi in HC, with TCA and clg in Str and LC. ↓Gαi1/2 with dsp in FCx, with clg in Str, with imi and clg in HT, with imi in HC. ↑Gαo with TCA and clg. G protein mRNA: ↑Gα2 in FCx, Gαq in Str, and ↓Gαs with flx in ME	[105]
amtp, dsp, trn	FCx	G protein density: o Gαs, Gαi2, Gα35, Gα36 G protein mRNA: o Gαs, Gαi2, Gα1, Gα2	[50]
ECS	FCx	↓ muscarinic and β-adrenergic-stimulated [3H]GTP binding	[8]
ECS	FCx	↑cAMP synthesis activating G proteins and AC	[147]
ECS	HC	↓Gαs, Gαi2, and ↑Gαo mRNA	[124]
Lithium	HC	↑Gαs, Gαi2, Gαo mRNA
Lithium	Cx	↓muscarinic and β-adrenergic-stimulated [3H]GTP binding	[7]
Lithium	Cx	↓Gαi1, Gαi2 mRNA o Gαo, Gαs, Gβ mRNA	[34]
Lithium	FCx, HT, HC, Str, DR, LC	↑Gαi in HT and HC o Gαo, Gαs	[104]
Lithium	Cx	↓Gαs, Gαi2, Gαi2 mRNA	[106]
Lithium	Cx	↓Gαs, Gαi2, Gαi2 mRNA o Gβo, Gβ1, Gβ2, Gβ3 mRNA	[107]
Lithium	Cx	↓5-HT-stimulated Gαs, Gαi, Gαo, Gαq o Gαs, Gαi, Gαo, Gαq densities	[181]

Antidepressant abbreviations: amtp, amitriptyline; clg, clorgyline; clo, clomipramine; dsp; desipramine; ECS, electroconvulsive shock; flx; fluoxetine; imi, imipramine; iprl, iprindole; TCA, tricyclic antidepressants; trn; tranylcypromine. Brain region abbreviations: Cx, cortex; DR, dorsal raphe; FCx, frontal cortex; HC, hippocampus, HT, hypothalamus; LC, locus coeruleus; ME, mesencephalus; Str, striatum.

It has been published that the level of expression of mRNA coding for certain G_α subunits is affected by antidepressants, however, this effect of chronic antidepressant treatment has generated some discrepancies [50]. The Electroconvulsive shock attenuates β-adrenoceptor and muscarinic cholinergic receptor coupling to G proteins [8], and augments heterotrimeric G protein-coupling to adenylate cyclase [147]. A modulation of the expression level of different G_α subunits by the Electroconvulsive shock has also been reported [124]. However, although chronic antidepressant treatment alters the expression and functional activity of heterotrimeric G proteins, they are unlikely a direct molecular target of antidepressant drugs.

Studies in Peripheral Tissue of Depressive Patients

The screening for peripheral markers of depression has been performed in order to facilitate the ease and accuracy of the diagnosis, as well as to improve the knowledge of its etiology. Markers reflecting the psychiatric state of the patient (such as the degree of severity or the response to medication) are termed state markers, whereas those present irrespective of the change in the psychological state are termed trait markers. Therefore, the detection of state markers would provide clinical information about the severity and treatment responsiveness, whereas the presence of a trait marker could be an index of the vulnerability to depression. Several reviews of the peripheral markers have appeared [102,132].

Certain discrepancies have been reported in the quantification of the densities and the functional status of the heterotrimeric G proteins in peripheral tissues of patients with mood disorders (Table 2). However, most of these studies (contrary to the studies performed in postmortem human brain, see below) have been performed in patients with a defined psychiatric diagnosis.

Table 2.

Summary of Studies of Heterotrimeric G Proteins in Peripheral Tissue of Patients with Mood Disorders

Psychiatric diagnosis	Patients/Controls	Peripheral tissue	Finding in peripheral tissue	Citation
Major Depression	22 non-treated depressed patients 22 controls	platelet	↑ Gαi1/2 ↓ Gαi3	[61]
Major Depression	37 non-treated depressed patients 31 controls	mononuclear leukocyte	↓ Gαs, Gαi density and functionality	[13]
Major Depression and Bipolar Disorder	14 non-treated depressed patients 8 non-treated bipolar patients 17 controls	mononuclear leukocyte	↔ Gαs, Gαi in MD ↑ Gαs, Gαi in BD	[188]
Bipolar Disorder	14 non-treated bipolar patients 20 euthymic bipolar patients treated with lithium 11 controls	platelet and mononuclear leukocyte	↑ Gαs in BD ↔ Gαi1/2, Gαq/11 in BD Lithium ↓ Gαq/11 Lithium ↑ PTX labeling	[116]
Bipolar Disorder	44 treated euthymic bipolar patients 27 controls	platelet	↑ Gαs ↔ Gαi1/2, Gαq/11	[129]
Bipolar Disorder	10 non-treated manic bipolar patients 10 non-treated euthymic patients 10 controls	mononuclear leukocyte	↑G protein function in BD ↔ in euthymic patients	[163]
Bipolar Disorder	20 manic bipolar patients 11 depressed bipolar patients 30 controls	mononuclear leukocyte	Manic BD: ↑ G protein function, and ↑ Gαi, Gαs Depressive BD: ↔ G protein function, and ↔ Gαi, Gαs	[12]
Mood Disorder	3 bipolar patients 10 depressed patients 18 controls	platelet	↔ Gαi/o	[140]
Mood Disorder	18 depressed patients 8 depressed bipolar patients 10 controls	mononuclear leukocyte	↓ Gαs density and functionality	[11]
Mood Disorder	16-22 depressed patients 11-16 bipolar patients 14-23 controls	granulocyte	↑ Gαs, Gαi2, and ↔ Gαq mRNA in BD ↔ Gαs, Gαi2, Gαq mRNA in MD	[171]

PTX, pertussis toxin

In platelets of patients with major depression, a higher density of G_αi1/2, and a lower density of G_αi3 subunit have been reported [61]. In mononuclear leukocytes, both decreased immunoreactive and functional levels of G_αs and G_αi [13], and not altered G_αs and G_αi densities [188] have been found. In patients with bipolar disorder, elevated levels the two isoforms of G_αs, both in platelets and in mononuclear leukocytes [116,129,188], without alterations in G_αq/11, G_αz, G_β, or in the different subtypes of G_αi [116,129], have been described. On the contrary, increased levels of G_αi have been found in mononuclear leukocytes of patients with bipolar disorder [188]. In platelets of patients with mood disorders (both major depression and bipolar disorder), decreased densities of 45 kDa G_αs [11], and unaltered levels of G_αi/o have been reported [140].

Regarding the identification of state markers of mood disorders, an increased [³H]GppNHp (GTP analog) binding stimulated by β-adrenoceptor and cholinergic muscarinic receptor agonists has been reported in mononuclear leukocytes in the manic episodes in patients with bipolar disorder [163]. In the same context, the β-adrenoceptor and cholinergic muscarinic stimulated binding of [³H]GppNHp was found elevated in the manic episodes, but decreased in depressive episodes [12]. State markers correlating with the severity of the bipolar disorder have been described according to the increased levels obtained in the density of G_αs [188]. On the contrary, a possible trait marker was proposed when increased levels of G_αs were found without a correlation with the pharmacological treatment of the patient [129]. Similar results have been reported in granulocytes for the mRNA levels codifying for G_αs [171].

Unfortunately, little progress has been made in the research of peripheral markers for depression. A number of hypotheses have been proposed, none of which satisfactorily explains all the discordant results that have been reported [132].

Studies in Postmortem Human Brain of Depressed Subjects

According to the indirect evidence for abnormalities at GPCR signaling described above, and to the molecular mechanism of action of certain classes of antidepressants [86,98,120,121,165,173], several independent laboratories have examined heterotrimeric G proteins in brain of patients with mood disorders [46,64,115,117,177]. These studies have been focused on the density and functionality of these proteins in postmortem human brain of suicide victims with or without a diagnosis of mood disorders, both major depression and bipolar disorder (Table 3).

Table 3.

Summary of Studies of Heterotrimeric G Proteins in Postmortem Human Brain in Suicide and Mood Disorders

Psychiatric diagnosis	Suicide/depressed/controls	Brain region	Finding in suicide or depression	Citation
Suicide victims	43 suicides 38 control	PFCx	mRNA and protein: ↓Gαi2, Gαo ↑ 45 kDa Gαs	[49]
Suicide victims with major depression	13 depressed 13 control	PFCx	↓ Gαi2 ↑ 45 kDa Gαs ↔ 52 kDa Gαs, Gαi1, Gαo, Gαq/11	[149]
Suicide victims with major depression	7 depressed 11 unknown diagnosis 22 control	FCx	↑ 45 kDa Gαs in depressed subjects ↔ 45 and 52 kDa Gαs in suicide victims	[36]
Major depression	4-6 depressed 1-4 dysthymic 7 control	PCx, TCx	↔ 45 and 52 kDa Gαs, Gα, Gαq, Gβ ↑ GTP analog binding to Gαi/o in depressed subjects	[148]
Bipolar disorder	7 bipolar 7 control	TCx, OCx, CB	↑ 45 and 52 kDa Gαs ↔ Gαi1/2, Gα, Gβ	[186]
Bipolar disorder	10 bipolar 10 control	PFCx, TCx, OCx, Thal	↑ 45 and 52 kDa Gαs in PCx, TCx, OCx ↔ Gαi1/2, Gα, Gβ	[187]
Bipolar disorder	5 bipolar 5 control	PFCx	↑ 45 and 52 kDa Gαs ↔ Gαi, Gα, Gαz, Gαq/11, Gβ	[57]
Bipolar disorder	10 bipolar 10 control	FCx, TCx, OCx	↑ Gαq/11 in OCx ↔ Gβ1, Gβ2	[122]
Bipolar disorder	10 bipolar 10 control	FCx, TCx, OCx	↔ 45 and 52 kDa Gαs mRNA	[189]
Major depression and bipolar disorder	15 depressed 15 bipolar 15 control	TCx, OCx	↔ 45 and 52 kDa Gαs, Gαι1/2	[47]
Suicide victims with mood disorder	13 depressed 3 bipolar 8 other diagnosis 9 unknown diagnosis 20 control	PFCx	↑ Gαi1/2 in non-treated suicide victims with mood disorder ↔ Gαi1/2 in treated suicide victims with mood disorder ↔ Gαi1/2 in suicide victims without diagnosis ↔ Gαq/11 in treated suicide victims	[62]
Suicide victims with mood disorder	21 depressed 6 bipolar 1 dysthymic 28 control	PFCx	↔ 45 and 52 kDa Gαs, Gαi1/2,Gαi3,Gαo	[69]

Brain region abbreviations: CB, cerebellum; FCx, frontal cortex; OCx, occipital cortex; PCx, parietal cortex; PFCx, prefrontal cortex; TCx, temporal cortex; Thal, thalamus

Several discrepancies have been reported. Thus, in prefrontal cortex of subjects with major depression, both increased immunoreactive levels of G_αi1/2 [62], and decreased G_αi2 without alterations in G_αi1 [149] have been described. The densities of 45 kDa G_αs have been reported to be increased in prefrontal cortex of suicide victims with major depression [36,149]. A significant decrease in both mRNA and protein levels of G_αi2 and G_αo, and a significant increase in levels of 45 kDa G_αs have been observed in prefrontal cortex of suicide subjects. These alterations were reported to be independent of the retrospective psychiatric diagnosis [49]. On the contrary, several publications have not found alterations in the densities of the different subunits of the heterotrimeric G proteins, both G_α [47,69] and G_β [148]. The G proteins can be directly activated in vitro in an experimental receptor-independent manner with the peptide mastoparan [141,142]. In postmortem prefrontal cortex, mastoparan displayed concentration-response curves with similar pharmacological profiles in suicide victims with mood disorders and in control subjects [69], suggesting that the functional availability of the heterotrimeric G proteins to be activated by GPCR is not altered in depression. Recently, certain polymorphisms in the olfactory Golf have been reported in major depression [190].

In suicide victims with bipolar disorder, increased levels of the two isoforms of G_αs [57,186] and of G_αq/11 [122] have been reported. More recently, the immunoreactive levels of these heterotrimeric G proteins have not shown significant changes in suicide victims with bipolar disorder when compared with control subjects [47,69]. No changes in the level of expression of the mRNA sequences encoding the 45 and 52 kDa G_αs [189], or abnormalities in the gene encoding for these α-subunits [153] have been reported.

The possible alterations in the level of expression and in the functional activity of the heterotrimeric G proteins in postmortem human brain of suicide victims with mood disorders have generated numerous publications with several discrepancies (see above, and Table 3), even within the same scientific group. This disappointing result does not come as surprise because clearcut methodological flaws have been repeated in many studies. Such methodological deficits may have influenced the obtained results in several ways. The presence of small sample sizes [57], heterogeneous groups (suicide victims without a medical diagnosis) [36,62,75], or the establishment of retrospective (postmortem) diagnosis of mood disorders in suicide victims [36,57,122,186,187] may be the basis of certain scientific discrepancies. Thus, it has been recently reported that, opposite to schizophrenia, reaching an accurate diagnosis for mood disorders is difficult to determine using retrospective diagnosis [43].

It has been demonstrated that several of the studied pharmacologic and biochemical parameters are affected by the intrinsic variables of postmortem human brain, such as gender, age, postmortem delay, and freezing storage period of the sample [70,123,156,161]. In this regard, experimental designs testing more suicide victims than control subjects [62], or the absence of matched pairs of subjects (suicides and their respective controls) according to these intrinsic variables [36,47,62,148,149] might be, as discussed somewhere else [69,70], some of the causes of the above reported discrepancies. The impact of these methodological shortcomings must be accounted for.

In summary, the evidence for an association between heterotrimeric G proteins and mood disorders is not as strong as believed [47,69].

G PROTEIN-COUPLED RECEPTORS IN DEPRESSED SUICIDE BRAIN: FUNCTIONAL SUPERSENSITIVITY OF α_2A-ADRENOCEPTORS

The possible alterations in the different subtypes of neurotransmitter GPCRs in postmortem human brain of suicide victims have been reviewed [74,96,98,173]. Several reports have described altered levels of the presynaptic α₂-adrenoceptor in postmortem human brain of suicide victims [30,42,52,62,66,125,126]. It has been demonstrated that the chronic treatment with antidepressant drugs, as well as with lithium, induces desensitization and down-regulation (a loss in the total receptor number) of the α₂-adrenoceptors both in rat brain [97,121,131,169], and in platelets of depressed patients [59,60]. In postmortem human brain of suicide victims with major depression, a down-regulation of the α₂-adrenoceptor promoted by chronic treatment with antidepressants has been reported by antagonist radioligand binding techniques [42]. This selective effect of the antidepressants in suicide victims was not described quantifying the immunoreactive densities of the α₂-adrenoceptors [62].

An important aspect of the obtained results valuating the role the α₂-adrenoceptors in postmortem human brain of suicide victims with mood disorders is the pharmacological characteristic of the radioligand utilized in the quantification. The quantification of the receptor density with α₂-adrenoceptor agonists strongly suggests an increased level of this receptor subtype in several brain regions of suicide victims with mood disorders [66,125,126,146]. On the opposite, no significant differences have been reported quantifying the α₂-adrenoceptor density with radioligand antagonists [30,126]. Increments in the density of α₂-adrenoceptors were suggested in a manuscript using a radioligand antagonist in temporal cortex of suicide victims with a retrospective diagnosis of depression [42]. In postmortem human brain of suicide victims without a specific psychiatric diagnosis, both increased [145] and not altered densities [3] of α₂-adrenoceptors have been reported.

The repeatedly reported increased densities of α2-adrenoceptors in postmortem human brain of suicide victims with mood disorders quantified with radioligand agonists suggest alterations in the functional activity of the receptor without alterations in its total density [30]. Antagonists present, by definition [95,183], comparable affinities to the different functional states of the receptor. According to the above discussed papers, the total density of the α2-adrenoceptors is not altered in postmortem human brain of suicide victims with mood disorders. Agonist chemicals are recognized by the active conformation of the receptor with a higher affinity than antagonists (see above) [40]. Thus, the described alterations in α₂-adrenoceptors might be related to modifications in the functional receptor-G protein coupling.

A direct evaluation of the receptor-G protein coupling can be obtained by measuring the agonist stimulation of [³⁵S]GTPγS binding (a radiolabeled GTP analog) [112,167, 178]. This technique has been optimized in postmortem human brain allowing the quantification of receptor-G protein coupling in membrane preparations [67,68], and in tissue sections [157]. The altered functional state of the α₂-adrenoceptors associated with mood disorders, suggested by the increased radioligand agonist binding (see above), has been recently supported quantifying agonist stimulation of [³⁵S]GTPγS binding in postmortem human brain of suicide victims with antemortem medical diagnosis [69]. A higher potency (lower EC₅₀) of the α₂-adrenoceptor agonist UK14304 promoting receptor-G protein coupling has been reported without alterations in the efficacy (E_max). This higher receptor-G protein coupling was demonstrated to be selective for the α₂-adrenoceptors within the tested GPCRs (5-HT_1A serotonin, μ-opioid, GABA_B, and cholinergic muscarinic receptors) [69]. One of the possible explanations for the selective increased α₂-adrenoceptor-G protein coupling without alterations in the above discussed total receptor or heterotrimeric G protein densities maybe the presence of a higher proportion of constitutively active receptors. According to this proposal, the quantification of the α_2A-adrenoceptor density with a selective inverse agonist [29,133] (presenting higher affinity for the inactive conformational state of the receptor, see above) suggests decreased binding sites in postmortem human brain membranes of suicide victims with mood disorders [30].

Recently, a very promising preliminary study has been published suggesting a polymorphism associated with suicide victims in the third intracellular loop of the α₂-Aadrenoceptor [164]. In this regard, detailed mutational analysis of different GPCRs has demonstrated the role of this intracellular loop 3 playing a key role in determining receptor-G protein coupling in conjunction with residues present in the intracellular loop 2 [25,137,184]. In the therapeutic context, the α₂-adrenoceptor antagonist mirtazapine is prescribed for the treatment of patients with major depression with positive results [53,54].

CONCLUSIONS

Decades of research have not allowed us to understand the molecular basis of mood disorders. The cellular mechanisms of action of many antidepressants are still unknown, and many depressed patients do not respond to antidepressant therapy. By understanding the cellular and molecular alterations associated with depressive disorders, the design of more efficacious therapeutic chemicals could be achievable.

Several discrepancies have been reported in the literature according to the possible alterations in G proteins associated with mood disorders. The last published results suggest that there are no detectable alterations either in the density or in the intrinsic functionality of heterotrimeric G proteins in postmortem human brain of suicide victims with mood disorders.