The 5-HT7 receptor system as a treatment target for mood and anxiety disorders

Search methods

The systematic search was conducted using Medline, Cochrane Library, EMBASE, PsycINFO, the National Institute of Health website Clinicaltrials.gov, controlled-trials.com, and relevant grey literature) from 1993 inception to March 2021. The following search string was used:

[(5-HT7 OR serotonin receptor 7 OR 5-hydroxytryptamine 7) AND (depress* OR bipolar disorder OR anxiety disorder)] OR [(5-HT7 OR serotonin receptor 7 OR 5-hydroxytryptamine 7) AND (animals OR humans OR preclinical study OR clinical trial OR experimental medicine)] OR [(5-HT7 OR serotonin receptor 7 OR 5-hydroxytryptamine 7) AND (lurasidone OR vortioxetine)] OR (5-HT7 antagonists OR 5-HT7 agonists)

Reference lists of included articles were further searched for eligible studies. If papers were not written in English, we attempted to obtain a translated version.

Systematic searches of the preselected databases were carried out independently by two researchers (TYL and NG), using the predetermined search string. Results were compiled using Rayyan QCRI software (Ouzzani et al., 2016), and titles and abstracts were independently screened by both researchers. Any studies that appeared eligible, or if there was any uncertainty about eligibility, underwent a full-text review. Final inclusion lists were compared, and any disagreements were discussed until a consensus was reached. An additional (PRS) reviewer was consulted as needed.

Study and participant type

Only full-text original research articles using an appropriate control group (sham or placebo) were included. Animal studies that used a mood or anxiety disorder model and a relevant genetic or pharmacological manipulation to the 5-HT7 system were included.

For human experimental medicine studies or clinical trials, randomized control trials (RCTs) using males and females over the age of 18 fulfilling International Classification of Diseases (ICD) or Diagnostic and Statistical Manual of Mental Disorders (DSM) criteria diagnosis for an MDD, major depressive episode (MDE), bipolar affective disorder (BD) or an anxiety disorder were included. All subtypes of MDD or MDE (mild, moderate, severe, with/without psychotic features) and bipolar disorder (rapid cycling, type I, type II and other) were included. Participants who only met criteria for dysthymia or cyclothymia were excluded.

Intervention type

Any studies that used an appropriate genetic or pharmacological manipulation to the 5-HT7 system were included in the present review. Pharmacological agents, in both preclinical and human studies, that have substantial selectivity for 5-HT7 receptors were included, such as selective agonists (AS-19, LP-44, LP-12, LP-211, E55888) (Brenchat et al., 2009^; Godínez-Chaparro et al., 2011^; Leopoldo et al., 2004^, 2008^; Sanin et al., 2004) and antagonists (SB-238719, SB-269970, SB-656104, DR-4004, DR-4446, PZ-766, JNJ-18038683, asenapine, amisulpride, imipramine, and desipramine) (Azima and Vispo, 1958^; Bonaventure et al., 2012^; Brodie et al., 1961^; Canale et al., 2016a^; Forbes et al., 2002^; Frånberg et al., 2008^; Hagan et al., 2000^; Kikuchi et al., 1999^; Puech et al., 1998^; Thomas et al., 1999^; Zhang et al., 2002^; Zajdel et al., 2012a). Lurasidone and vortioxetine, antipsychotic and antidepressant medications, respectively, were also included due to their partial 5-HT7 receptor antagonism (Adell, 2010^; Ohno et al., 1997). Studies that did not use pharmacological agents with reasonable 5-HT7 affinity were excluded.

Outcome measures

The primary outcome measure of interest was changes in mood and anxiety behaviors after pharmacological interventions with agents that bind to, or alter, 5-HT7 receptor function. In preclinical animal studies, these behavioral changes were evaluated by use of validated functional tests that may measure depressive or anxiety symptomatology (e.g. light dark transfer test, elevated plus-maze tests, forced swim tests, and tail suspension tests). Secondary outcome measures of interests were changes in sleep or cognition measures.

For human studies, changes from baseline to endpoint in mood status, assessed by change in mood-related symptoms measured by validated rating scales such as the Hamilton Depression Rating Scale (HAM-D) or Quick Inventory of Depressive Symptomology (QIDS-CR) was the primary outcome measure (Hamilton, 1960^; Rush et al., 2003). Any included data on change in sleep, cognition, or other experimental medicine markers (e.g. neuroimaging) were also collected.

Quality measures

Study quality was measured by using the “Quality Assessment Tool for Quantitative Studies,” which was designed by Effective Public Health Practice Project (Armijo-Olivo et al., 2012^; Thomas et al., 2004). Included studies were assessed in the following domains: selection bias, study design, confounders, blinding, data collection methods, withdrawal, and dropout. A global quality rating was scored independently, and any disagreements discussed by reviewers.

Systematic search results

The initial search identified 8097 papers. After removal of duplicates, 4989 studies underwent initial title and abstract screening leaving 109 studies for full-text review. Review of full-text articles excluded 49 articles for the following reasons: article type (7 review or meta-analysis papers and 18 conference abstracts or posters), wrong disorder or outcome measure (20), data reported elsewhere (2), not a randomized control trials study (1), and article language (1). Two further studies were added from hand searching after the original search. In total, 64 studies were included (52 animal studies and 12 human studies), and details are outlined in Figure 1.

Open multimedia modal

Study characteristics

Fifty preclinical animal studies were included in the review. Most of the studies (48/52) examined 5-HT7 drug or knockout (KO) effects on preclinical models of depression, with a further 23 examining models of anxiety. Study details and summarized findings are listed in Table 1.

Open multimedia modal

A comprehensive listing of all relevant study results, with corresponding mood-related behavioral paradigms and mood models, comparators, individual results and p-values can be found in Supplemental Information—Preclinical Behavioral Results.

While all included studies used compounds or genetic manipulations that reasonably targeted 5-HT7 receptors, some compounds had higher selectivity than others.

Thirty used nonspecific 5-HT7 antagonist agents, which also have moderate to high affinity to other receptors, including other 5-HT and dopaminergic receptors, while 25 studies used compounds selective to 5-HT7 receptors.

For the purpose of this review, selective compounds are considered those which had preferential binding defined as a 5-HT7 inhibition constant (K_i) at least fourfold less than all other receptors of interest. A complete list of compounds considered in this review, with their chemical (IUPAC) name and receptor binding profiles, can be found in Supplemental Information—Receptor Activity. Receptor bindings are described by inhibition constants (K_i in nanomoles) unless otherwise noted (e.g. IC50 or pK_i values). The Psychoactive Drug Screening Program K_i database was used to reference all compounds (Helmuth, 2000^; Roth et al., 2000). If compounds were not listed in the database, this is noted in the Supplemental Information along with source data for receptor bindings.

Antidepressant effects

The forced swim test (FST) and the tail suspension test (TST) are often used to evaluate antidepressant agents in preclinical rodent studies, and these tests measure immobility time as a correlate of negative mood, or hopelessness (Can et al., 2011^; Yankelevitch-Yahav et al., 2015). Even though the FST and TST are not recognized as effective models of depression, they are still used for screening potential antidepressants in preclinical studies (Bourin et al., 2001^; Sewell et al., 2021). Within our systematic review, 43 studies used the FST, and 15 studies used the TST.

Twenty-five studies that used compounds nonspecific to the 5-HT7 receptor identified potential antidepressant effects depression such as immobility time, in both the FST or TST (Abbas et al., 2009^; Cates et al., 2013^; Chlon-Rzepa et al., 2013^; Gu et al., 2019^, 2018^, 2017^; Guilloux et al., 2013^; Jankowska et al., 2020^; Kim et al., 2014^; Kołaczkowski et al., 2014^; Latacz et al., 2018^; Li et al., 2013^; Mørk et al., 2012^; Partyka et al., 2017^, 2019^; Pytka et al., 2015^, 2017a^, 2017b^, 2018^; Wang et al., 2019^; Waszkielewicz et al., 2015^; Wróbel et al., 2019^; Zagórska et al., 2015^, 2016^; Zajdel et al., 2013).

Furthermore, 19 studies using agents with selective 5-HT7 antagonist properties found significant improvements in potential markers of depression; for example, SB-269970 was found to decrease animals’ immobility time both in water (FST) and land (TST) (Bonaventure et al., 2012^, 2007^; Canale et al., 2015^, 2016b^, 2016a^, 2017^; Guscott et al., 2005^; Hedlund et al., 2005^; Kim et al., 2016^; Kucwaj-Brysz et al., 2018^; Lax et al., 2018^; Medina et al., 2014^; Partyka et al., 2019^; Stroth and Svenningsson, 2015^; Wesolowska et al., 2006a^, 2006b^, 2007^; Zajdel et al., 2011^, 2015). An additional study found decreased immobility time in 5-HT7 gene KO mice, compared to wild type (WT) mice, in the FST (Balcer et al., 2019). Lastly, one study using a highly selective 5-HT7 agonist, AS-19, reported increased immobility (i.e. more depressed-like behavior) in the FST (Li et al., 2013).

Of note, the antipsychotic medication amisulpride and antidepressants with 5-HT7 antagonist properties, such as imipramine and desipramine, were found to have potential antidepressant effects in these tests (Abbas et al., 2009^; Wesolowska et al., 2007). Abbas and colleagues found that amisulpride, an antipsychotic that acts primarily as a dopaminergic receptor antagonist but also has potent 5-HT7 receptor antagonism (Abbas et al., 2009), showed potential antidepressant effects in both the TST and FST in WT mice. These effects were not seen in 5-HT7 KO mice, suggesting that activity at the 5-HT7 receptor may be specifically associated with potential antidepressant properties. Another study using citalopram, imipramine, desipramine, and moclobemide found that while low doses of these agents had no significant effect, antidepressant properties were evident once combined with SB-269970, a selective 5-HT7 receptor antagonist, suggesting that it was enhanced 5HT7 antagonism that produced potential antidepressant-like effects (Wesolowska et al., 2007).

Two studies found mixed results, each with a 5-HT7 antagonist compound showing potential antidepressant effects in the FST, but not TST (Lax et al., 2018^; Wang et al., 2019). A further three studies showed no significant antidepressant effects of their tested compounds in either the TST or FST (Delcourte et al., 2017^; Maxwell et al., 2019^; Volk et al., 2011).

Anxiolytic effects

In addition to modulation of depressive symptoms, there is also evidence from preclinical animal studies that 5-HT7 receptor modulation may mediate changes in measures of anxiety. A total of 23 studies measured anxiety behaviors using a variety of paradigms.

Similarly to preclinical depression studies, 11 studies found potential anxiolytic effects in animal models associated with the use of 5-HT7 antagonist agents, which also have strong to moderate affinities at other serotonergic and dopaminergic receptors (Chlon-Rzepa et al., 2013^; Guilloux et al., 2013^; Latacz et al., 2018^; Pytka et al., 2015^, 2017b^, 2018^; Volk et al., 2008^, 2011^; Zagórska et al., 2015^, 2016^; Zajdel et al., 2012b). Eight studies found potential anxiolytic effects when investigating selective 5-HT7 receptor antagonists (Canale et al., 2016a^; Hedlund and Sutcliffe, 2007^; Kucwaj-Brysz et al., 2018^; Lax et al., 2018^; Mnie-Filali et al., 2011^; Wesolowska et al., 2006a^, 2006b^; Zajdel et al., 2015).

These studies used animal models of anxiety where anxiolytic effects were measured by a decrease in rodents’ motion or increase their searching behaviors in anxious environments (e.g. maintaining motionless to escape from electric shock in Four-Plate Test, increasing attempts to get food under electrical shocks in Vogel Conflict Drinking Test, or burying more marbles in their bedding in Marble Burying Test). The selective 5-HT7 antagonist, SB-269970, increased antianxiety behaviors in the Vogel Conflict Drinking Test and in the elevated plus-maze test (Wesolowska et al., 2006a). A similar finding was identified in the shock threshold test and open-field test (Wesolowska et al., 2006b). In addition, mice showed less anxious behavior in the marble burying test, which may measure obsessive-compulsive behaviors (Hedlund and Sutcliffe, 2007).

Balcer and colleagues found mixed results, with 5-HT7 gene KO mice being significantly less anxious during a novelty suppressed feeding task, but not during an elevated maze plus task (Balcer et al., 2019). Two studies found no significant anxiolytic effects associated with any compounds with 5-HT7 antagonistic properties (Cates et al., 2013^; Maxwell et al., 2019). One study found that the 5-HT7 agonists, LP-211 and LP-378, improved potential anxiety-related behaviors in the light/dark test and increased exploration of black and white boxes (Adriani et al., 2012).

Other effects

In addition to antidepressant and anxiolytic effects, 5-HT7 antagonists have been investigated in animal models of mania, sleep, and cognition. One study found that hyperlocomotion in rodents induced by sleep deprivation, a behavioral model of mania, was reduced after treatment with asenapine (Delcourte et al., 2017). Rodents treated with JNJ-18038683, a selective 5-HT7 antagonist, and 5-HT7 gene KO mice both displayed increased latency to REM sleep and decreased REM duration (Bonaventure et al., 2012^; Hedlund et al., 2005). Additionally, the selective 5-HT7 antagonist SB-269970 increased REM latency, decreased REM sleep duration, and reversed increases in sleep fragmentation induced by citalopram (Bonaventure et al., 2007).

Several studies have also reported that compounds with 5-HT7 antagonist properties were associated potential pro-cognitive effects, such as reversing natural forgetting impairment in rats (Canale et al., 2017^; Pytka et al., 2017a) and reversing drug-induced memory impairments (Canale et al., 2016b^; Jankowska et al., 2020^; Zajdel et al., 2015).

Study characteristics

Twelve RCTs were included in the systematic review. Most of the studies (11) were between 6 and 8 weeks long (mean study duration 7.4 weeks); however, one study lasted for 32 weeks. Eight studies explored effects in participants with MDD, two in participants with bipolar disorder, and two in participants with generalized anxiety disorder (GAD). Only one study used a highly selective 5-HT7 antagonist (JNJ-18038683), while the others used medications that had higher affinity to other receptors, in addition to 5-HT7 receptor activity. All clinical trial details and findings are summarized in Table 2.

Open multimedia modal

Major depressive disorder

Six studies investigated the effect of vortioxetine (LU AA21004), a medication with 5-HT1A, 5-HT1B, 5-HT3, and 5-HT7 receptors and 5-HT transporter affinities, on MDD symptoms. All studies administering vortioxetine at 20 mg found significant improvements in mood as measured by Montgomery–Åsberg Depression Rating Scale (MADRS) (Montgomery and Asberg, 1979) between baseline and week 8 (Jacobsen et al., 2015^; McIntyre et al., 2014). Three studies found significant differences in mood at a 10-mg dose (Alvarez et al., 2012^; Henigsberg et al., 2012^; McIntyre et al., 2014); however, two only found a trend toward significance at this dosage (Jacobsen et al., 2015^; Mahableshwarkar et al., 2015). Lower doses of vortioxetine (5 mg or less) found no significant changes associated with treatment (Alvarez et al., 2012^; Henigsberg et al., 2012^; Mahableshwarkar et al., 2013). It is important to emphasize that vortioxetine is not a 5-HT7-specific agent, and vortioxetine also blocks 5-HT1D and 5HT-3 receptors and stimulates 5-HT1A and 5HT1B receptors.

Only one study examined change in depressive symptoms after 7 weeks of 20-mg JNJ-18038683, a highly selective 5-HT7 receptor antagonist, compared to placebo and escitalopram 20 mg (Bonaventure et al., 2012). Although JNJ-18038683 did not significantly decrease depression symptoms compared to placebo, escitalopram (20 mg) also did not significantly change mood scores in this study.

Anxiety disorders

Two studies examined the effects of medications with 5-HT7 antagonistic properties in participants with GAD. Rothschild and colleagues measured anxiety symptoms using the Hamilton Anxiety Scale (HAM-A) (Hamilton, 1959) before and after 8 weeks of treatment with 5 mg vortioxetine, but found no differences compared to placebo (Rothschild et al., 2012). Baldwin and colleagues investigated the effect of either 5 or 10 mg of vortioxetine on time to relapse in participants with GAD. At the end of a 20-week open-label treatment period, participants who responded were then randomized to 24–56 weeks of a double-blind treatment of vortioxetine (n = 229) or placebo (n = 230). The study found a statistically significant effect of vortioxetine relative to the placebo in time to relapse (Baldwin et al., 2012).

Bipolar disorders

Lurasidone and asenapine are both atypical antipsychotics often used to treat schizophrenia and bipolar disorders, and they both have moderate affinity to 5-HT7 receptors. Lurasidone is a dopaminergic D2 and D3 receptor, 5-HT2A, 5-HT7, and α2C-adrenergic receptor antagonist, and a partial 5-HT1A agonist (Bawa and Scarff, 2015). In one study investigating bipolar depression, participants treated with lurasidone (at both 20–60 mg and 80–120 mg doses) experienced a significant improvement in MADRS scores after 6 weeks compared to placebo (Loebel et al., 2014). One study, with a sample size of nine participants, found significant improvements in both depressive and anxiety symptoms in those with bipolar depression after 8-week treatment with asenapine (El-Mallakh et al., 2020). In participants with MDD with mixed features, who experienced least 2–3 manic or hypomanic episodes, depressive symptoms measured by MADRS and manic symptoms measured by the Young Mania Rating Scale (YMRS) improved after 6 weeks of treatment of 20–60 mg lurasidone (Suppes et al., 2016^; Young et al., 1978).

Other effects

As with rodent studies, human clinical trials have observed effects of 5-HT7 receptor modulation outside of mood symptoms. McIntyre and colleagues found that both 10- and 20-mg vortioxetine significantly improved participants’ composite cognition scores compared to placebo using a predefined efficacy analysis. Additionally, they also found significant improvements in most secondary objectives and subjective patient-reported cognitive measures (McIntyre et al., 2014). Another study measuring cognitive impairment using the Cognitive and Physical Functioning Questionnaire (CPFQ) found while both 10- and 15-mg vortioxetine numerically improved scores, these improvements were not statistically significant versus placebo (Fava et al., 2009^; Mahableshwarkar et al., 2015).

Only one study in the systematic review examined changes in sleep in human participants. JNJ-18038683, a specific 5HT-7 antagonist, was found to prolong REM latency and reduced REM sleep duration in healthy participants, and enhanced REM sleep suppression induced by citalopram (Bonaventure et al., 2012).

We would like to thank Dr. Anthony Vernon for his help and advice on interpretation of preclinical compound binding.