Research Projects of the Biological Psychiatry Group (Elias Eriksson, Agneta Ekman, Christer Nilsson, Charlotta Sundblad, Mikael Landén, Monika Hellstrand, Lars Westberg, Hoi-Por Ho, Marie Olsson, Jonas Melke, and Olle Bergman)
PROJECT A: The role of serotonin in sex steroid-related psychiatric disorders & the role of sex steroids in serotonin-related disordersBackground:
Enhanced irritability/aggression and enhanced sexual activity are the two most prominent characteristics of a serotonin-depleted rat. Both aggression and sexual behaviour are regulated also by sex steroids; a major role for serotonin in animals hence seems to be to modulate sex steroid-driven behaviour. With respect to the influence of serotonin on sexual function, a relationship similar to that observed in experimental animals seems to exist also in man, reduced libido and anorgasmia being the side-effects most frequently reported by non-depressed patients treated with serotonin reuptake inhibitors (SRIs). To what extent serotonin influence irritability and aggression in humans is however somewhat less obvious; to investigate this has been one purpose of this project (see below). A more general aim has been to explore the possibility that serotonin may be of importance for sex steroid-related psychiatric disorders, such as premenstrual dysphoria (PMD), and, conversely, that sex steroids (in particular testosterone) may be involved in serotonin-related psychiatric disorders (such as depression, obsessive compulsive disorder, panic disorder, and depression). In addition, we have recently become interested in the influence of both serotonin and sex steroids on personality traits, as well as on personality disorders. The influence of sex steroid-related genes for the development of transsexualism, criminality, and cognitive function also has been the subject of our studies.
Key findings:
1) We were the first to show that SRIs are very effective for the treatment of PMD. This observation (first presented at a congress in 1989) has now been replicated in approximately 20 placebo-controlled trials.
2) We were the first to show that SRIs are superior to noradrenaline reuptake inhibitors for the treatment of PMD. This observation has now been replicated in three placebo-controlled trials.
3) We were the first to show that the effect of SRIs when used for PMD is characterized by a very short onset of action (1-2 days), hence enabling intermittent drug administration during the luteal phase only. This observation has now been replicated in five placebo-controlled trials.
4) We were the first to show that intermittent treatment may in fact be more effective than continuous drug administration, suggesting that the effect of SRIs when used for PMD is characterized by a certain development of tolerance. This observation has recently been replicated in one US study.
5) We have reported that both PMD and bulimia nervosa may be
associated with a relative hyperandrogenicity.6) We were the first to suggest (in a case report) that an androgen antagonist (flutamide) may be effective for the treatment of bulimia nervosa. In a recent yet unpublished placebo-controlled trial, we could confirm this effect.
7) In a study comparing PMD women and controls, we have shown that cerebrospinal levels of monoamine metabolites seem to be dependent on cycle phase, and related to serum levels of sex steroids; also, we showed that the HVA/5HIAA ratio is lower in PMD patients than in controls.
8) In animal experiments, we have shown that the irritability displayed by female Wistar rats in the resident-intruder paradigm is cycle-dependent, and that it is effectively reduced by treatment with an SRI; tentatively, this may represent the first animal model of PMD.
9) In recent and ongoing studies, a series of candidate genes coding either for proteins involved in the formation or action of sex steroid, or for proteins involved in serotonergic neurotransmission, have been investigated in patients with various psychiatric conditions (PMD, bulimia nervosa, anorexia nervosa, transsexualism) and in controls; also, the relationship between gene variants on the one hand, and serum levels of sex steroids on the other, has been studied. So far, these studies have resulted in several very intriguing findings, regarding, e.g., sex steroid related genes in patients with transsexualism, serotonin related genes in patients with eating disorders, and the relationship between sex steroids related genes and serum levels of sex steroids in normal controls.
10) Prompted by the fact that serotonin and androgens may play opposite roles for the pathophysiology of PMD and bulimia nervosa (see above), and also by the fact that androgens and serotonin are known to exert opposite effects on various aspects of behaviour, such as aggression, sexual drive, and impulsivity, we have studied to what extent androgenization influences serotonin activity in the female rat brain. These experiments suggest that androgenization of female rats, either neonatally or at adult age, leads to a reduction in serotonin release in the amygdala. In addition, we have shown that androgenization of female rats neonatally leads to insulin resistance at adult age. In ongoing experiments, the effect of androgens on serotonin receptor density, and on the expression of mRNA for genes related to serotonin transmission, is being studied.
11) Prompted by the fact that sexual side-effects (reduced libido and anorgasmia) are common in both men and women treated with SRIs, we have investigated the effect of SRIs on sexual function in male and female rats. These studies suggest that SRIs exert a similar influence on sexual activity in animals as in man, and that animal experiments hence could be used in order to elucidate the mechanisms (in terms of receptor involvement etc) underlying SRI-induced sexual dysfunction.
12) In several clinical studies, we have investigated the influence of SRIs on sexual function in humans. Also, we have shown that adding buspirone (a 5HT1A agonist with an alpha-2-blocking metabolite) to an SRI reduces the sexual side-effects; this finding had previously been reported by others; our study however was the first double-blind trial showing a difference between buspirone and placebo in this respect.
13) We were the first to show that the strong SRI clomipramine is superior to imipramine for the treatment of panic disorder, suggesting that serotonin reuptake inhibition is more important than noradrenaline reuptake inhibition when treating panic disorder with tricyclic antidepressants. This finding has been replicated in one controlled trial.
14) We have published a number of reports regarding putative biological markers in patients with panic disorder and controls, including one study showing elevated CSF levels of beta-endorphine immunoreactivity in panic disorder patients.
15) Prompted by the finding (by others) that panic attacks in panic disorder patients may be elicited by an infusion of sodium lactate, we have investigated the effect of lactate infusion on cardiovascular and respiratory parameters in Wistar rats, and suggested that the response to lactate in this rat strain may serve as an animal model of panic anxiety.
Significance (theoretical & clinical)
1) Many studies have shown that serotonin depletion induces irritability or aggression in experimental animals. Our discovery that SRIs are effective in reducing the severe irritability characterizing PMD was one of the first pieces of evidence suggesting that serotonin may counteract irritability also in humans. It is hence of some theoretical importance.
2) Our observation that SRIs are effective for PMD means that women with this very stressful condition (afflicting at least 5% of all women of fertile age) may now be effectively treated. PMD recently became an approved indication for one SRI in the UK and the US.
3) Our observation that SRIs reduce premenstrual complaints with a very short onset of action (1-2 days) is (in our opinion) of considerable theoretical importance, since it is the first time ever that such a prompt symptom-reducing effect of serotonin reuptake inhibition has been reported.
4) The short onset of action of SRIs when used for PMD is also of clinical significance, since it makes it possible for women with PMD to restrict the medication (and hence also the side-effects) to two weeks per cycle.
5) Our finding that intermittent administration is more effective than continuous administration also is important, since it suggests that SRIs, when used for PMD, are associated with a certain development of tolerance, than can be effectively avoided by repeated drug holidays. Previously, it has generally been assumed that the use of antidepressant drugs is not marred by development of tolerance.
6) Our observation that an SRI is more effective than a noradrenaline reuptake inhibitor for the treatment of PMD suggests that PMD should not be regarded as a variant of depression, but as a distinct diagnostic entity. Also, it underlines the importance of serotonin for the regulation of premenstrual complaints. Needless to say, the conclusion that the effect of SRIs in PMD is not equivalent to the antidepressant effect also lends support from the short onset of action in PMD (see above), and from our finding that much lower doses of clomipramine (25 mg) are required when treating PMD as compared to when treating depression.
7) Our clinical and preclinical studies on the relation between sex steroids (particularly testosterone) and serotonin may increase our understanding of the pathophysiology of a number of serotonin-related psychiatric disorders, including PMD, bulimia nervosa, obsessive compulsive disorder, and depression; also, it may increase our understanding on how androgens (and other sex steroids) influence various forms of normal behaviour, such as sexual activity.
8) Our recent observation that androgen antagonism may be effective for the treatment of bulimia nervosa may lead to an enhanced understanding of the pathophysiology of eating disorders, and may lead to the development of new strategies for the management of these conditions.
9) Our collaboration (during the 80s) with the manufacturer of clomipramine, Ciba-Geigy (now Novartis), led to the registration of clomipramine for panic disorder, first in Sweden, then in many other countries. For many years, clomipramine was regarded as the drug of choice for the treatment of panic disorder; now it is gradually being replaced by the selective SRIs. Partly due to our discovery that clomipramine is superior to imipramine for the treatment of panic disorder, clomipramine rather than imipramine is now generally used as reference compound in panic disorder drug trials.
10) Our attempts to establish animal models reflecting PMD and panic disorder may improve the possibilities to investigate how SRIs influence cycle-related irritability and panic anxiety, respectively (in which brain areas are these effects exerted? via which postsynaptic serotonin receptors are they mediated? etc). Likewise, our experiments regarding the effect of SRIs on sexual activity in rat may serve to shed light on the mechanisms underlying SRI-induced sexual dysfunction.
11) Our recent observation regarding the involvement of sex steroid-related genes in transsexualism may lead to an enhanced understanding of the aetiology of this enigmatic condition.
PROJECT B: Studies on partial and inverse agonism vis-à-vis G-protein-coupled receptors
Background:
It has for long been known that the intrinsic activity displayed by a partial agonist vis-à-vis a G-protein-coupled receptor may vary considerably, depending on the responsiveness of the receptor population upon which it acts. According to classical receptor theory, the receptor number (=receptor density) is the most important factor in this context; thus, when the receptor population studied is characterized by a high receptor density, the intrinsic activity of a partial agonist is relatively high. On the other hand, when some of the receptors are being inactivated, e.g. by means of the administration of receptor alkylators, the efficacy of partial agonists is reduced.
Receptor antagonists are defined as compounds that do not influence receptor activity per se, but reduce the effect of an agonist. In the 70s, it was shown that certain agents could influence the benzodiazepine receptor, which is a part of the GABA A receptor/chloride ion channel complex, in a way opposite to that of an agonist, also in the absence of agonists. This paradoxical effect was referred to as inverse agonism. Until recently, G-protein-coupled receptors were not assumed to be the subject of inverse agonism.The dopaminergic D2 receptor, which is a member of the family of G-protein-coupled receptors, is the major target for compounds with antipsychotic (antagonists) or anti-parkinsonian (agonists) effects, respectively. In addition, D2 receptors situated on the prolactin-producing cells in the pituitary inhibit the release of prolactin.
Key findings:
1) We were the first to show that a number of partial D2 agonists that act as antagonists vis-à-vis postsynaptic D2 receptors in the striatum act as full agonist vis-à-vis the prolactin regulating D2 receptors in the pituitary of male rats. This observation has subsequently been confirmed both in experimental animals and in humans.
2) We were the first to show that female and male rats differ markedly with respect to the responsiveness of prolactin regulating D2 receptors; thus, partial D2 agonists that stimulate prolactin regulating D2 receptors in males act as antagonists on these receptors in females. However, after chronic dopamine depletion, the responsiveness of prolactin regulating D2 receptors in females is upregulated, and hence similar to that in males.
3) We have shown that a receptor alkylating compound (EEDQ) may switch the effect of partial agonists vis-à-vis prolactin regulating D2 receptors from agonism to antagonism, without influencing the density of pituitary D2 receptors. We hence suggested that factors other than receptor number (probably G-protein-related) are of importance for the intrinsic activity displayed by a partial agonist, and that receptor alkylators may influence these factors.
4) By means of in vitro experiments (using D2-transfected clonal cells), we have obtained further support for our assumption that factors other than the receptor number are crucial for the intrinsic activity displayed by partial D2 agonists; thus, in an experimental setting in which the receptor number is constant, the efficacy of a partial D2 agonist may switch from agonism to antagonism depending on which intracellular transduction systems that are being simultaneously activated (so-called protean agonism).
5) In in vitro experiments, using D2-transfected clonal cells, we have shown that a number of compounds previously regarded as neutral D2 antagonists, such as the antipsychotic compound haloperidol, exert an effect on D2 receptors opposite to that of agonists, also in the absence of agonism. Our first report on this phenomenon was in fact the first paper ever in which inverse agonism vis-à-vis a G protein coupled receptor could be demonstrated using a functional response in intact cells. Subsequently, the fact that G-protein-coupled receptors may be the subject of inverse agonism has been confirmed in a very large number of studies. The inverse agonism of haloperidol and other antipsychotic drugs vis-à-vis dopamine D2 receptors also has been replicated by several groups.
6) We have shown that several compounds, generally regarded as dopamine D3 receptor antagonists, in fact may act as D3 receptor agonists.
7) By administrating an antisense oligonucleotide in vivo, we have obtained evidence for an influence of dopamine D3 receptors on dopamine turnover and locomotion in rat.
Significance:
1) The findings that G-protein-coupled receptors (such as the D2 receptor) may be the subject of inverse agonism, and that the intrinsic activity displayed by partial agonists is dependent on G-protein-related factors, are both observations of some general pharmacological importance. Also, they may have direct implications for the difference in clinical profile between typical and atypical antipsychotic drugs.
2) The observation that partial D2 agonists reduce rather than increase prolactin release is of some clinical importance, since this class of drugs are presently being evaluated for the treatment of schizophrenia and Parkinson's disease, and also for hyperprolactinemia.
3) Studies on the functional role of the D3 receptor may, tentatively, be of importance for the future development of antipsychotic (and anti-parkinsonian) drugs.