Classics in Chemical Neuroscience: Aripiprazole

Abstract

Aripiprazole was the first antipsychotic developed to possess agonist properties at dopamine D₂ autoreceptors, a groundbreaking strategy that presented a new vista for schizophrenia drug discovery. The dopamine D₂ receptor is the crucial target of all extant antipsychotics, and all developed prior to aripiprazole were D₂ receptor antagonists. Extensive blockade of these receptors, however, typically produces extrapyramidal (movement) side effects which plagued first-generation antipsychotics, such as haloperidol. Second-generation antipsychotics, such as clozapine, with unique polypharmacology and D₂ receptor binding kinetics, have significantly lower risk of movement side effects, but can cause myriad additional ones, such as severe weight gain and metabolic dysfunction. Aripiprazole’s polypharmacology—characterized by its unique agonist activity at dopamine D₂, D₃ and serotonin 5-HT_1A receptors as well as antagonist activity at serotonin 5-HT_2A receptors—translates to successful reduction of positive, negative, and cognitive symptoms of schizophrenia, while also mitigating risk of weight gain and movement side effects. New observations, however, link aripiprazole to compulsive behaviors in a small group of patients, an unusual side effect for antipsychotics. In this review, we discuss the chemical synthesis, pharmacology, pharmacogenomics, drug metabolism, and adverse events of aripiprazole, and we present a current understanding of aripiprazole’s neurotherapeutic mechanisms, as well as the history and importance of aripiprazole to neuroscience.

Graphical abstract

INTRODUCTION

Schizophrenia—characterized by positive symptoms (delusions, hallucinations, disorganized speech and behavior), negative symptoms (catatonia, blunted affect, apathy, and anhedonia), and cognitive symptoms (deficits in executive function and working memory)—is a devastating psychiatric disorder that affects approximately 1% of the population.¹ A key early finding was an association between positive symptoms of psychosis and an overactive striatal dopamine system,^2–5 which lead to the enduring “dopamine hypothesis” of schizophrenia. Specifically, persons with schizophrenia show elevated baseline and psychostimulant-induced striatal dopamine release^{2, 6–7} and are hypersensitive to dopaminergic psychostimulants.⁸ Also, acutely psychotic patients, as well as patients prodromal for schizophrenia, display increased presynaptic striatal dopamine synthesis (measured by increased uptake of radiolabeled L-dihydroxyphenylalanine (L-DOPA), dopamine’s precursor).^{3, 9} Despite these associations, most studies do not observe a correlation between schizophrenia and the availability of the dopamine transporter (DAT).^10–12 Also, though increases in dopamine D₂ receptor density in schizophrenia are observed,¹³ interpreting the findings is often confounded by prior antipsychotic treatment, which itself increases D₂ receptor density.¹⁴ Other alterations in the dopamine system are reported, such as increases in D₂ receptor homodimers and D₂ receptors existing in high-affinity (active) conformations,³ but evidence regarding these observations, presently, is not definitive.

Most antipsychotic medications that are effective at reducing positive symptoms block D₂ receptors, which are highly expressed in the striatum, to recalibrate dopamine signaling and remodel dopaminergic circuits.^15–17 Neuroimaging studies demonstrate that clinical improvement in positive symptoms using first-generation antipsychotics (FGAs), such as chlorpromazine and haloperidol, requires approximately 65% striatal D₂ receptor occupancy.^18–20 However, grave drawbacks of FGAs are extrapyramidal side effects (EPS)—motor abnormalities such as rigidity, muscle spasms, tremors, restlessness, and involuntary movements (e.g. tardive dyskinesia)—^20–24 and increased serum prolactin levels (hyperprolactinemia), which can lead to lactation, decreased bone density, and disturbances in sex hormones.²⁵ EPS are caused by a blockade of nigrostriatal D₂ receptors, and hyperprolactinemia is caused by blockade of tuberoinfundibular D₂ receptors, which normally function to suppress prolactin secretion from the anterior pituitary gland.^26–27 Closely titrating the dose of FGAs is necessary to reduce the likelihood or severity of EPS,²⁸ whereas reducing hyperprolactinemia may require switching to a second-generation antipsychotic (SGA).²⁹

SGAs, including clozapine and quetiapine, are clinically as effective as FGAs,³⁰ and they are associated with a lower incidence of EPS and hyperprolactinemia.^{26, 31} Many hypotheses have been put forward to explain the neuropharmacology that underlies these characteristics. In contrast to FGAs, positron emission tomography indicates SGAs can be effective when occupying less than 60% of striatal D₂ receptors—a major finding that overturned the seemingly ineluctable paradigm that antipsychotic efficacy is directly proportional to striatal D₂ receptor occupancy.^{20, 32–36} Though, differences in ligand-receptor on/off binding kinetics may also be a causal factor, i.e. effective doses of SGAs and FGAs may occupy similar numbers of striatal D₂ receptors, but SGAs may dissociate more quickly than FGAs from D₂ receptor binding sites, as is documented with quetiapine and clozapine.^37–38 A rapid off-rate would likely permit endogenous dopamine to maintain an adequate level of D₂ receptor signaling to prevent hyperprolactinemia and EPS. Whereas tight binding of FGAs to D₂ receptors may decrease endogenous D₂ receptor signaling to a level that causes these side effects.

Additionally, it has been argued that antagonism of serotonin 5-HT_2A receptors by SGAs contributes to their efficacy, reducing the need for relatively extensive D₂ receptor occupancy, consequentially reducing side effects.^39–40 In support of this, studies show that selective 5-HT_2A receptor antagonism attenuates amphetamine-elicited release of dopamine into the striatum while also blocking amphetamine’s psychomotor effects in non-human primates.⁴¹ Also, the selective 5-HT_2A antagonist, pimavanserin, effectively treats psychosis in Parkinson’s disease without causing EPS.^42–43 However, FGAs such as chlorpromazine and haloperidol have appreciable potencies at 5-HT_2A that are comparable to some SGAs, challenging this hypothesis.^44–45 Other targets, such as muscarinic and 5-HT_1A receptors may contribute to lower EPS risk of newer antipsychotics.^{40, 46–48} The pharmacology of amisulpiride, however, casts doubt on this hypothesis—it has very low affinity at 5-HT_1A, and at each of the muscarinic receptors (and at 5-HT_2A receptors),⁴⁹ yet has low EPS liability.⁵⁰

From a neural systems perspective, there is evidence that SGAs, like olanzapine and clozapine, target the mesolimbic (ventral striatum) dopamine system, sparing the nigrostriatal (dorsal striatum) dopamine system that is intimately involved in motor processing, which may underlie clozapine’s low EPS characteristics.^51–52 Nevertheless, some SGAs also have added risk of causing obesity and metabolic dysfunction (e.g., diabetes, high cholesterol).^{50, 53–54} Exploration of the precise neurobiological and neuropharmacological mechanisms underlying distinct effects of individual antipsychotics remains vigorous.

Due in part to the aforementioned side effects and others caused by both FGAs and SGAs, including sedation and emotional dampening,⁵⁵ approximately two-thirds of patients with a psychotic disorder are noncompliant with their antipsychotic medications.^56–57 Finally, antipsychotic drugs have limited efficacy in approximately one-third of patients,⁵⁸ and treatment of negative and cognitive symptoms in schizophrenia remains a challenge—likely because they involve different neural systems and mechanisms than positive symptoms. For example, in contrast to a hyperactive striatal dopamine system underlying positive symptoms, a hypoactive mesocortical dopamine system is proposed to underlie negative and cognitive symptoms.^{5, 59–61}

This review discusses the prototypical third-generation antipsychotic (TGA), aripiprazole, distinguished by its agonist pharmacology at certain receptor targets, most notably D₂ autoreceptors.^62–65 This pharmacology was the first of its kind for an approved antipsychotic medication. Furthermore, aripiprazole can be effective at attenuating negative and cognitive symptoms, in addition to positive symptoms, of schizophrenia in some patients and has lower EPS liability than FGAs (e.g., haloperidol), lower weight gain and metabolic liabilities than SGAs (e.g., clozapine), and does not cause hyperprolactinemia.^{50, 66–69} Though, aripiprazole is not a panacea, having limited effects in some patients, and new reports suggest a link between aripiprazole and impulse control deficits, a unique side effect for an antipsychotic, which we briefly discuss. Aripiprazole opened new vistas for exploration of the neurobiological underpinnings of psychotic symptoms and of side-effects caused by antipsychotics, and it inspired a paradigm shift from an antagonist-based to an agonist-based approach for antipsychotic drug discovery.

CHEMICAL PROPERTIES AND CHEMICAL SYNTHESIS

Aripiprazole, 7-[4-[4-(2,3-dichlorophenyl)piperazin-1-yl]butoxy]-3,4-dihydro-1H-quinolin-2-one, is an achiral quinolinone derivative (CAS No: [129722-12-9]; Figure 1). It possesses a single hydrogen bond donor and five acceptors, has a molecular weight of 448.4 g/mol, and a Log P value of 4.55. These physicochemical properties comply with Lipinski’s rule of five and provide the compound with high bioavailability, protein binding, and an acceptable metabolic profile.^70–71 Otsuka Pharmaceutical patented aripiprazole in 1988 [US patent 4,734,416 filed in 1979], along with many other carbostyril derivatives and salts thereof, as potential antihistamine and central nervous system controlling agents.^72–73 However, the patent did not explicitly mention aripiprazole by name, nor did it describe a synthetic procedure used for it.

Figure 1.

Chemical structures of aripiprazole (OPC-14597, left) and its predecessor OPC-4392 (right).

The first synthetic procedure and reference to its antipsychotic activity was described by Oshiro et al. in Otsuka’s 1991 patent [US patent 5,006,528 filed in 1989].⁷⁴ In 1998, Otsuka scientists described a similar synthesis for the free base, but with slightly different conditions (Scheme 1).⁷⁵ The synthesis begins with the alkylation of 7-hydroxy-3,4-dihydro-2(1H)-quinolinone by stirring it with 1,4-dibromobutane (3 molar equivalents (mol. equiv.)) in the presence of potassium carbonate (1 mol. equiv.) in dimethylformamide at 60 °C for four hours to give 7-(4-bromobutoxy)-3,4-dihydro-2(1H)-quinolinone. The reaction mixture is then diluted with an equal volume of water, and the organic phase is extracted with ethyl acetate. After rotary evaporation, the resulting product is recrystallized in ethyl alcohol.^74–75 The product is subsequently combined with sodium iodide (2 mol. equiv.) in acetonitrile and refluxed for 30 minutes before cooling to room temperature. Next, 1-(2,3-dichlorophenyl)piperazine (1.5 mol. equiv., prepared based on⁷⁶), and triethylamine (2 mol. equiv.) are added to the reaction mixture, and refluxed for another four hours. The resulting precipitate is filtered and discarded. The filtrate is evaporated in a low-pressure environment, and dissolved in ethyl acetate. Then, it is washed, dried, and subjected to rotary evaporation to yield a resin. The resin is recrystallized in ethyl alcohol to provide the free base of aripiprazole as a white powdery substance. The powder may then be dissolved in ethyl alcohol with acid to yield a variety of salts. Other compounds, such as OPC-4392 (Figure 1), aripiprazole’s predecessor, were prepared using similar procedures by condensing (4-bromobutoxy)-2(1H)-quinolinone (or a structural analog) with various phenylpiperizines;^74–75 the procedure has since been optimized.⁷⁷

Scheme 1.

Synthesis of aripiprazole reported in the primary literature.

MANUFACTURING INFORMATION

Aripiprazole was approved by the FDA for schizophrenia on November 15^th, 2002, as an oral tablet formulation in the dose range of 2 to 30 mg. PubMed-indexed publications with “aripiprazole” in their abstract rapidly increased around this time, peaking at ~300/year in 2008, then plateauing to present day (Figure 2). Aripiprazole was originally manufactured by Otsuka Pharmaceutical and was co-marketed with Bristol Meyers-Squibb under the brand name Abilify®. The following formulations have since been approved: orally disintegrating tablets, oral solution, and an aqueous solution for intramuscular injection. The original patent expired April, 2015, and generic 2 to 30 mg oral tablets have subsequently been produced by numerous manufacturers (Teva, Torrent, Hetero Labs, Alembic, and Ajanta, among others). Consistent annual sales figures for Abilify® are unavailable, however, global sales in 2013 and 2014 were approximately $7.824 and $9.285 billion, respectively,⁷⁸ illustrating the drug’s enormous financial success.

Figure 2.

PubMed-indexed citations containing “aripiprazole” in their Abstract (1995–2016).

APPROVED INDICATIONS AND DOSING

Aripiprazole, oral formulation, is approved for the following indications: 1) schizophrenia in adults and adolescents (13–17 years); 2) bipolar I disorder (manic and mixed episodes) in adults and pediatric patients (10–17 years); 3) major depressive disorder in adults, as an adjunctive therapy; 4) irritability associated with autism spectrum disorder in pediatric patients (6–17 years); and 5) vocal and motor tics associated with Tourette’s syndrome in pediatric patients (6–18 years).⁷⁰ Additionally, the intramuscular formulation is approved to treat agitation associated with schizophrenia or bipolar mania in adults.⁷⁰ Generally, children are started on a 2 mg/day dose of aripiprazole, and may be titrated up to 10 mg/day, depending on the disorder and severity.⁷⁰ Adults on the other hand are usually started on 10 mg/day, and may be titrated up to a maximum of 30 mg/day depending on their response.⁷⁰

PHARMACOKINETICS AND PHARMACOGENOMICS

Aripiprazole displays linear kinetics, with a bioavailability of 87% (independent of low-fat food intake).^{70, 79} The maximum plasma concentration occurs 3–5 hours following administration, and a steady-state is achieved after approximately 14 days of daily dosing.^{70, 79} At steady-state, the volume of distribution is 404 L due to extensive binding to plasma proteins (>99%).^{70, 79} In vitro studies using human liver microsomes and recombinant cytochrome P450 enzymes indicate that aripiprazole is metabolized predominantly via phase I mechanisms by both CYP3A4 and CYP2D6 to yield dehydrogenation and hydroxylation products, while CYP3A4 also mediates N-dealkylation (Figure 3).⁷⁰ Phase II metabolism is also present, albeit to a lesser extent (FDA document NDA No. 21-436).⁸⁰ Interestingly, the major circulating metabolite is dependent upon the frequency of administration. With acute dosing, the phase II product BMS-337041 is the most abundant circulating metabolite, whereas the phase I product dehydroaripiprazole is more prevalent after chronic dosing. In addition to compounds provided in Figure 3, McEnvoy et al. report a motley of aripiprazole metabolites from LC-MS/MS analysis of human urine, including iminium ion, epoxide, and glutathione species.⁸⁰

Figure 3.

Structures of phase I metabolites of aripiprazole (adapted from FDA document NDA No. 21-436). Dehydroaripiprazole is the major metabolite after chronic dosing, and is pharmacologically active.

Aripiprazole is metabolized almost equally along both CYP3A4 and CYP2D6 pathways (1:1) in extensive metabolizers, however, this ratio approaches 3:1 in intermediate metabolizers.^81–82 At steady-state the average elimination half-life of aripiprazole is approximately 75 hours, and the primary metabolite, dehydroaripiprazole, is 95 hours.⁷⁰ Considering that dehydroaripiprazole is pharmacologically active at D₂ receptors and exists at a concentration that is ~40% of the plasma concentration of aripiprazole, it likely contributes to the sustained pharmacologic effects of aripiprazole.^{70, 83}

Serum concentrations of aripiprazole at therapeutic doses range between 150–300 ng/mL,⁸³ but they can vary substantially between patients administered equivalent doses. This is likely due to polymorphisms in the CYP2D6 gene. For example, in individuals with little to no CYP2D6 activity, the elimination half-life of aripiprazole extends from ~95 to 146 hours.^{70, 81–83} Also, psychiatric patients are often poly-medicated, which likely impacts aripiprazole concentrations. In particular, aripiprazole is often prescribed as an adjunctive antidepressant; since many antidepressants are also metabolized by CYP2D6 and CYP3A4 (e.g., paroxetine, fluvoxamine, fluoxetine), the potential for increased serum concentrations of these drugs is inherent.^{70, 83} Patients also taking medications that are inhibitors or inducers of CYP3A4 and/or CYP2D6⁸⁴ should undergo close therapeutic drug monitoring.

Genetic association studies show that polymorphisms in several genes impact antipsychotic efficacy.^85–87 For example, the Taq1A (rs1800497) polymorphism, a transition mutation (C→T) that generates the A1 allele of D2DR, is associated with an increased therapeutic response to aripiprazole in Asian patients.⁸⁸ This association, however, is not observed in other ethnicities.^{85, 89} Multiple neuroimaging studies show that healthy volunteers, of European descent, who are either homozygous or heterozygous for the A1 allele exhibit less striatal D₂ receptor availability.^90–92 Two other D2DR polymorphisms, Ser311Cys (rs1801028) and –141C Ins/Del (rs1799732), also appear to affect antipsychotic drug response. The Ser311Cys polymorphism associates with a favorable response to risperidone in Han Chinese populations.⁹³ Conversely, some studies show that-141 Del carriers tend to have a poorer response to several antipsychotics, but not aripiprazole.^{85–86, 94} Also, the C-1019G polymorphism in HTR1A, and A-1438G-T102C (rs6311/rs6313) in HTR2A (genes encoding the 5-HT_1A and 5-HT_2A receptors, respectively) associate with decreased efficacy of aripiprazole to treat negative and cognitive symptoms in schizophrenic patients of Han Chinese descent.^95–96 Furthermore, other polymorphisms in HTR2A and DRD2 may interact to facilitate aripiprazole’s antipsychotic efficacy.⁹⁷ Unfortunately, many of these experiments have small effect sizes, and results depend upon the ethnic population studied and how “clinical response” is defined. Thus, pharmacogenomics currently provides insufficient data to guide personalized medicine for schizophrenia.

PHARMACOLOGY AND ADVERSE EVENTS

Like other antipsychotics, aripiprazole’s pharmacology is complex (Table 1),^{62, 98} and is representative of the polypharmacology—many targets—approach to treat psychotic disorders. Aripiprazole has high affinity (defined here as K_i < 30 nM) at serotonin 5-HT_1A, 5-HT_2A, 5-HT_2B, 5-HT_2C, and 5-HT₇, dopamine D₂ and D₃, adrenergic α_1a, and histamine H₁ receptors. Aripiprazole is distinguished from earlier antipsychotics by its partial agonist activity at D₂, D₃, 5-HT_1A, and 5-HT_2C receptor targets^{62, 64, 98–101}. In vitro, aripiprazole is a neutral antagonist or very weak partial agonist at 5-HT_2A and 5-HT₇, and is an inverse agonist at 5-HT_2B receptors.⁶² According to the prescribing information for Abilify®, aripiprazole is an “α1 antagonist.” Its functional effects at H₁ receptor have not been reported in scientific literature, to our knowledge. Though, since histamine activation effects are not typically reported, and since aripiprazole was part of a series of compounds designed to be antihistamines, it is likely an H₁ antagonist.

Table 1.

Aripiprazole: Receptor Affinity Profile

Receptor Target	Ki, nM	Radioligand and Source
“High” Affinity (Ki < 30 nM)
Serotonin 5-HT2B	0.4	[3H]LSD*
Dopamine D2	0.95	[3H]NMSP, PDSP Certified
Dopamine D3	5.4	[3H]NMSP*
Serotonin 5-HT1A	5.6	[3H]8-OH-DPAT, PDSP Certified
Serotonin 5-HT2A	8.7	[3H]Ketanserin*
Serotonin 5-HT7	10	[3H]LSD, PDSP Certified
Serotonin 5-HT2C-INI	22	[125I]DOI*
Adrenergic α1a	25	[125I]HEAT, PDSP Certified
Histamine H1	29	[3H]Pyrilamine, PDSP Certified
“Moderate” Affinity (Ki = 30–300 nM)
Adrenergic α1b	34	[125I]HEAT, PDSP Certified
Adrenergic α2c	38	[3H]Clonidine, PDSP Certified
Serotonin 5-HT1D	63	[3H]GR-125743, PDSP Certified
Adrenergic α2a	74	[3H]Clonidine, PDSP Certified
Adrenergic α2b	102	[3H]Clonidine, PDSP Certified
Adrenergic β1	141	[125I]Pindolol, PDSP Certified
Adrenergic β2	163	[125I]Pindolol, PDSP Certified
“Low” Affinity (Ki > 300 nM)
Dopamine transporter, D1, D4, D5, serotonin transporter, 5-HT1B, 5-HT1E, 5-HT3, 5-HT5a, 5-HT6, muscarinic M1, M2, M3, M4, M5, nicotinic α7, α1β2, α2β2, α2β4, α3β2, histamine H2, H4, norepinephrine transporter– all PDSP Certified. Δ-opioid*, β-opioid*, κ-opioid*

K_i values obtained from the NIMH Psychoactive Drug Screening Program database, http://pdsp.med.unc.edu/pdsp.php; search conducted 02/17/2017. Affinities were determined by displacement of radioligand from human cloned receptors. ^*Shapiro et al., 2003

At effective doses, aripiprazole occupies up to 90% of D₂ receptors, and also occupies 5-HT_1A and 5-HT_2A receptors but to a lesser extent.^{22, 102} As a D₂ receptor partial agonist with moderate intrinsic activity (as characterized in some in vitro assays), aripiprazole may block postsynaptic D₂ receptors in neural systems with high dopaminergic tone, i.e. the striatal dopamine system of schizophrenic patients, which may account for its effects on positive symptoms. Conversely, it may activate postsynaptic D₂ receptors in neural systems with low dopaminergic tone, i.e. the mesocortical system in schizophrenic patients, which may account for its effects on negative and cognitive symptoms.^{65, 99, 103–104} Other proposed pharmacological mechanisms focus on aripiprazole’s full agonism at presynaptic D₂ autoreceptors in vivo in animal models.^{62, 64, 98–99, 103, 105} D₂ autoreceptors modulate dopamine neurotransmission via a negative feedback mechanism. For example, at relatively high dopamine concentrations (or during phasic dopamine release), presynaptic D₂ autoreceptors are activated to decrease dopamine synthesis and release, and somatodendritic D₂ autoreceptors decrease neuronal firing rate.^106–109 Striatal dopamine neurons express high levels of D₂ autoreceptors, whereas they are scantly expressed in the mesocortical dopamine pathway.^110–111 Thus, aripiprazole, by acting as a D₂ autoreceptor agonist, may decrease dopaminergic tone selectively in the striatum. Combined with antagonist activity at postsynaptic D₂ receptors in vivo,¹⁰⁵ this pharmacology may contribute to its unique clinical effects. Finally, aripiprazole’s functional effects at the D₂ receptor vary depending on signaling pathway, cell type, and cellular context^{62, 64, 98, 112–113} providing evidence that it may be a biased agonist at the D₂ receptor in vivo^{62, 64, 98, 112–113}. How its unique intracellular signaling effects impact clinical symptoms is not known.

Despite extensive D₂ receptor occupancy, aripiprazole exhibits a relatively low risk for EPS and no risk of hyperprolactinemia—both caused by chronic D₂ receptor blockade.^{22, 114–115} Aripiprazole may even reduce prolactin secretion in certain patients, potentially via its D₂ receptor partial agonist effects.⁵⁰ Besides partial agonist activity at D₂ receptors, aripiprazole’s agonist activity at 5-HT_1A receptors, from partial to full agonist depending on cellular system, may also contribute to its efficacy and reduced side effects, relative to FGAs.^{46–47, 62, 101, 116–118} Aripiprazole activates somatodendritic 5-HT_1A receptors, reducing serotonin release and subsequently increasing dopamine release in the cortex, which may translate to treat negative and cognitive symptoms of schizophrenia.^119–120 Similarly, SGAs, such as clozapine and ziprasidone, that can attenuate negative and cognitive symptoms are also 5-HT_1A partial agonists^121–123 and enhance central dopamine release via activation of 5-HT_1A receptors.^124–125 More recently, it was shown that the ability of clozapine to reverse phencyclidine-elicited cortical desynchronization—a model of neural activity underlying negative symptoms— requires activation of 5-HT_1A receptors.¹²⁶

Relative to SGAs, such as clozapine and olanzapine, aripiprazole has a lower propensity to induce weight gain.^{50, 127} SGAs are potent 5-HT_2C and H₁ antagonists or inverse agonists—likely contributing to their obesity side effects.¹²⁷ Aripiprazole, however, is a partial agonist at the 5-HT_2C receptor, and the selective 5-HT_2C agonist, lorcaserin (Belviq®), is effective at decreasing weight in humans.^127–129 Nevertheless, some patients taking aripiprazole gain weight, and its putative antagonist activity at histamine H₁ receptors (K_i ~29 nM, Table 1), which is associated with weight gain,^{45, 62} may offset beneficial effects of 5-HT_2C partial agonism. Much less is known about the contribution of aripiprazole’s other targets—including 5-HT_2B (which it binds with highest affinity), 5-HT₇, D₃, and adrenergic α_1a—to its mechanism of action, though each of these targets warrants further investigation. For example, new evidence links inactivation or knockout of 5-HT_2B receptors to impulsivity both preclinically and clinically.^130–131 This is intriguing in light of recent reports of serious impulse control deficits and compulsive behaviors in a small group of patients taking aripiprazole.¹³² Although immediate focus moved to aripiprazole’s dopamine agonist activity, aripiprazole’s highly potent inverse agonist activity at 5-HT_2B receptors may also contribute to impulsive and compulsive behavior.

In preclinical studies of rats and rabbits, teratogenic and developmental toxicity effects of aripiprazole are observed at doses 2–11 times the maximum recommended human dose (based on area under the curve comparisons) (FDA document NDA No. 21-436). These findings guided designation of aripiprazole to pregnancy risk category C. However, the evidence for serious adverse fetal events in humans remains weak due to a lack of well-designed, clinical trials. A recent, prospective, cohort study assessed the risk of major malformations in infancy following exposure to aripiprazole in the first trimester of pregnancy. Results show an absolute risk of 3.13% (N = 96), not different from the risk associated with exposure to any SGA antipsychotic, 1.3% (N = 312), or unexposed infants, 0.6% (N = 177).¹³³ Though, small sample sizes and wide confidence intervals may have obfuscated actual differences. These data corroborate an earlier study in pregnant women taking aripiprazole (N = 86 vs 172 unexposed).¹³⁴ Although this latter study reports a significantly increased rate of pre-term delivery (odds ratio = 2.30; 95% CI 0.32–16.7) and restricted fetal growth (odds ratio = 2.97; 95% CI 1.23–7.16) in the aripiprazole group. The aripiprazole dosing regimen, however, was not controlled. Furthermore, approximately one third of patients taking aripiprazole reported also smoking tobacco and drinking alcohol while pregnant, known risk factors for prematurity and restricted fetal growth.^134–135 Aripiprazole is detected in human breast milk, and women taking aripiprazole are advised not to breastfeed.¹³⁶ Considering the increased risk of a psychiatric relapse and of adverse pregnancy outcomes in schizophrenic patients, (e.g. stillbirths)¹³⁷ a reduction in dose or a postponement of treatment should be assessed on a case-by-case basis.

The most commonly reported side effects of aripiprazole are akathisia in adults and tremors in adolescents with schizophrenia.^{70, 83, 138–140} Also, adjunctive treatment with antidepressants carries a black box warning of an increased risk of suicidal thoughts and behavior in patients 24 years of age and younger.^{70, 141} Elderly patients treated off-label for dementia-related psychosis are at an increased risk of death, and a significant dose-response relationship exists between aripiprazole and cerebrovascular events in the elderly.^{70, 142–144} Other side effects include dizziness, drowsiness, sedation, insomnia, somnolence, weight gain, drooling, restlessness, anxiety, and headaches.^{70, 138, 145}

HISTORY AND IMPORTANCE TO NEUROSCIENCE

Pharmacotherapy for schizophrenia began in the early 1950s with the discovery that the FGA chlorpromazine produced a powerful, non-narcotic, calming effect, reducing agitation in hospital patients undergoing surgery. Not long afterwards it was tested in psychotic patients, and proved tremendously effective, so much so that it freed many from institutionalization, and spawned the “psychopharmacological revolution”.¹⁴⁶ Chlorpromazine and later SGAs, such as clozapine, were found to be dopamine receptor antagonists, inspiring the dopamine hypothesis of schizophrenia. This hypothesis has withstood the test of time. Despite intensive research efforts to discover new drug targets for schizophrenia, all approved antipsychotics share activity at dopamine receptors as an essential part of their pharmacology. FGAs and SGAs block dopamine D₂ receptors, and vanquish positive symptoms of schizophrenia that result from hyperactivation of the striatal dopamine system. They are less effective at treating negative and cognitive symptoms, which are believed to result from hypoactivation of the mesocortical dopamine system. Though, a careful clozapine treatment regimen can alleviate some negative and cognitive symptoms, which may be due to its unique pharmacology at serotonin receptors. [Pointedly, clozapine remains the most effective antipsychotic, yet because of the risk of agranulocytosis, a potentially fatal condition caused by suppression of white blood cells,¹⁴⁷ it is only approved for treatment-resistant schizophrenia and for reducing suicidal behavior in patients with schizophrenia or schizoaffective disorder.^{50, 148}] FGAs and SGAs carry inherent serious side effect risks. Namely, FGAs are fraught with EPS and hyperprolactinemia issues, and many SGAs cause metabolic syndrome, obesity, and type II diabetes.^{69, 149–150}

Side effect issues of FGAs and SGAs and their limited efficacy in treating negative and cognitive symptoms in schizophrenia provided the impetus and the opportunity for drug discovery targeting novel biological targets or pharmacological mechanisms, which seeded the discovery of aripiprazole. In the late 1970s, Otsuka scientists were exploring 2(1H)-quinolinone derivatives for anti-histamine functionality devoid of central nervous system side effects when they fortuitously discovered compounds with antipsychotic activity in preclinical tests.¹⁵¹ Compounds were derivatized into a series of (4-phenyl-1-piperazinyl)alkoxy-2(1H)-quinolinone molecules. Lead candidates, such as OPC-4392, were selected based on their agonist activity at presynaptic D₂ autoreceptors, a then novel, alternative approach for reducing dopaminergic activity compared to direct blockade of postsynaptic D₂ receptors.^151–152 This rationale was based, in part, on observations that D₂ autoreceptors are predominantly expressed in the striatal dopamine system. Activation would reduce dopamine synthesis there, without largely impacting cortical dopamine activity.

OPC-4392’s D₂ autoreceptor agonist activity was assessed ex vivo and in vivo. Ex vivo, in rat striatal slices, OPC-4392 dose-dependently inhibited L-DOPA formation—an indirect measure of inhibition of tyrosine hydroxylase, which is the key enzyme for dopamine synthesis. This effect did not involve direct suppression of tyrosine hydroxylase activity, and it was reversed by the D₂ receptor antagonist, sulpiride, suggesting a presynaptic D₂ receptor mechanism.¹⁵³ In vivo, in anesthetized rats, OPC-4392 inhibited ventral tegmental area (VTA) dopamine spikes elicited by nucleus accumbens stimulation (antidromic effects), an effect blocked by simultaneous application of a D₂ receptor antagonist, providing evidence of D₂ autoreceptor agonist effects in the ventral striatum.¹⁵⁴ In addition to dopamine autoreceptor agonist effects, OPC-4392 was also purported to block postsynaptic D₂ receptors, based on observations that it inhibited behaviors in mice elicited by the D₂ receptor agonist apomorphine and reversed the inhibitory effect of apomorphine on acetylcholine release in rat striatal slices.¹⁵⁵

OPC-4392 showed efficacy in preclinical behavioral models of psychosis at doses that did not induce catalepsy, a model of EPS.¹⁵¹ For example, it attenuated jumping behavior in mice treated with the dopamine precursor L-DOPA and dopamine releaser, methamphetamine. OPC-4392 also prevented lethality caused by high dose methamphetamine. Despite the promising preclinical data, clinical development for schizophrenia was halted. “Unpublished observations” from clinical studies included an improvement in negative symptoms without EPS, but an aggravation of positive symptoms due to an “activation” effect.¹⁰⁵ Nevertheless, these clinical studies corroborated a link between dopamine autoreceptor agonism and efficacy to treat negative symptoms with low EPS liability, found earlier using terguride.¹⁵⁶

Beginning with OPC-4392, structure activity relationships were used to increase antagonist potency at postsynaptic dopamine receptors to treat positive symptoms, while maintaining autoreceptor agonism to treat negative symptoms.¹⁰⁵ By replacing the –propoxy linker in OPC-4392 with –butoxy, the 2,3-dimethyl moiety with 2,3-dicholoro, and by changing the carbostyril moiety to 3,4-dihydrocarbostyril, aripiprazole (OPC-14597) was discovered. The approach for developing aripiprazole closely paralleled OPC-4392, which was used as a prototype to guide the preclinical development of aripiprazole. Relative to OPC-4392, aripiprazole has nearly equipotent agonist activity at presynaptic dopamine autoreceptors in vivo, but has increased antagonist potency at postsynaptic dopamine receptors.¹⁰⁵

Its dopamine autoreceptor agonist activity was determined by its efficacy to block increases in DOPA caused by reserpine and by gamma-butyrolactone in several brain regions; effects that were reversed by haloperidol. Moreover, like OPC-4392, direct application of aripiprazole inhibited VTA dopamine neurons, an effect blocked by a D₂-selective receptor antagonist, domperidone, but not a D₁-selective receptor antagonist, SCH-23390, suggesting agonist activity at D₂ autoreceptors.¹⁵⁷ A recent clinical neuroimaging study, however, shows that aripiprazole does not affect dopamine synthesis, even at doses that occupy up to 79% of striatal D₂ receptors, challenging the conclusion that activation of D₂ autoreceptors is a critical mechanism for its antipsychotic effects.¹⁵⁸ The doses, 3–9 mg oral, used in this study are lower than the typical prescribed, starting dose for adults with schizophrenia, 10 mg. Also, this study involved a single, acute treatment in healthy subjects, so conclusions should be taken with caution.

Aripiprazole’s antagonist activity at postsynaptic receptors was inferred in preclinical studies by its failure to induce both locomotion in mice treated with reserpine, and contralateral rotations in rats with striatal 6-hydroxydopamine lesions—both behaviors are observed after postsynaptic dopamine receptor activation. Additionally, aripiprazole inhibited apomorphine induced stereotypy, locomotion, and ipsilateral rotations in a kainic acid striatal lesion model at potencies >10-fold relative to OPC-4392, suggesting more potent antagonist activity at postsynaptic D₂ receptors. Aripiprazole’s ED₅₀ doses in these models were substantially less than its ED₅₀ for inducing catalepsy, together suggesting antipsychotic activity without EPS.¹⁰⁵

In the same year that these preclinical results were published, Otsuka reported aripiprazole’s efficacy in both positive and negative symptoms of schizophrenia with a low risk of EPS, minor weight gain, and without significant prolactin elevation, results later corroborated.^{66, 159} As clinical data on aripiprazole’s efficacy accumulated in the following decades, it was found to be effective in treating numerous neuropsychiatric disorders (described under Approved Indications).

The discovery of aripiprazole broke ground, because of its novel agonist activities at D₂ autoreceptors that translated to efficacy for positive, negative, and cognitive symptoms with reduced side effects. Remarkably, or as if aligned with the history of antipsychotic drug discovery in general, it wasn’t until after clinical trials for schizophrenia that aripiprazole’s pharmacological activity at 5-HT_1A receptors was discovered. Clozapine, a partial agonist at 5-HT_1A, shares a similar history. These discoveries were important for neuroscience, as they helped unveil the interaction between 5-HT_1A receptors and cortical dopamine neurotransmission, and they guided rational antipsychotic drug discovery targeting 5-HT_1A receptors to reduce negative and cognitive symptoms. Since aripiprazole, other antipsychotics designed to possess 5-HT_1A agonist activity have been approved, including cariprazine, asenapine, and brexpiprazole.

Schizophrenia pathophysiology is more than appreciably complex, involving genetic and environmental factors interacting with neurodevelopmental factors, and basic science remains crucial for uncovering new biological targets for schizophrenia drug discovery. There has been much research into other neural circuits impacted in the disease, such as the glutamate and acetylcholine systems.^160–163 Moreover, basic science has led to the discovery of new biological mechanisms for existing antipsychotics. For example, new research shows that aripiprazole inhibits microglia activation, an anti-inflammatory effect that may have clinical utility for schizophrenia.^164–166 Aripiprazole has inspired and continues to inspire new questions regarding the neurobiological mechanisms of schizophrenia.

ABBREVIATIONS

EPS

extrapyramidal side effects

FGA

first generation antipsychotic

SGA

second generation antipsychotic

TGA

third generation antipsychotic