Main

The prototypical classical neuroleptic agent haloperidol is an effective antipsychotic compound (Leucht et al. 1999) which, unfortunately, also induces severe extrapyramidal side effects (EPS) and hyperprolactinemia. The term “atypical antipsychotic” was originally introduced to describe compounds, such as clozapine, which not only suppressed psychotic symptoms, but differed from haloperidol by having a low tendency to induce EPS and increase plasma prolactin levels. It is postulated that dopamine D2 receptor blockade is the mechanism by which antipsychotic activity is achieved (Creese et al. 1976; Seeman et al. 1976). However, since clozapine also interacts with D2 receptors it is likely that there are additional properties embedded in the clozapine molecule that explain its “atypicality.” One influential theory was put forward by Meltzer et al. (1989), who suggested that about a 10-fold higher affinity for serotonin 5-HT2A over D2 receptors could be the key factor in achieving improved clinical response and absence of EPS. This hypothesis has led to the development of new antipsychotics with the desired 5-HT2A/D2 affinity ratio, although each of these compounds also binds to additional receptor sites. Several of these novel antipsychotic compounds indeed display clear benefits in terms of side-effect profile. Interestingly, clozapine is active in schizophrenic patients who are treatment-refractory to other antipsychotics, including the more recent 5-HT2A/D2 receptor antagonists (Wahlbeck et al. 1999; Conley et al. 1999; Taylor and Duncan-McConnell 2000). The elucidation of the multiple receptor interactions of clozapine suggested that other neurotransmitter systems may also play a role in the efficacy and tolerability of that molecule. Particularly, the anti-adrenergic effects of clozapine are now receiving increased attention (Nutt 1994; Litman et al. 1996; Hertel et al. 1999). Iloperidone is a new psychotropic agent currently undergoing Phase III trials for the treatment of psychotic disorders (Figure 1 ). Iloperidone was selected from a large series of piperidinyl-benzisoxazoles because it showed a 300-fold greater potency in a test for limbic activity (inhibition of apomorphine-induced climbing) than in a test for nigrostriatal activity (inhibition of apomorphine-induced stereotypy) (Strupczewski et al. 1995). The large difference of potency of iloperidone in these tests is expected to result in an improved ratio of therapeutic effect to EPS liability compared with standard antipsychotics.

Figure 1
figure 1

Chemical structure of iloperidone.

Full size image

Previous studies have investigated the receptor binding profile of iloperidone with rat receptors (Szewczak et al. 1995) and a limited number of human homologues of dopamine and 5-HT receptor subtypes (Kongsamut et al. 1996). These experiments demonstated that iloperidone displays the desired 5-HT2A/D2 affinity ratio. The aim of the present study was to determine the receptor affinity profile of iloperidone at a wider range of human neurotransmitter receptors. In resemblance to clozapine, it was noted that iloperidone possesses high affinity for norepinephrine α1- and α2C adrenergic receptors.

METHODS

The radioligand receptor binding assays are listed in Table 1.

Table 1 Affinity Profile of Iloperidone for Human Receptors
Full size table

Materials

Radioligands were purchased from NEN Life Science Products, USA, except for 3H-RX821002 and 3H-Mesulergine, which were obtained from Amersham Pharmacia Biotech Ltd, UK, and 125I-GTI which was obtained from ANAWA, Switzerland. Iloperidone was synthesized by Hoechst Marion Roussel. Unless specified otherwise, all other chemicals were of reagent grade and obtained through standard commercial sources.

Membrane Preparation

Total rat brain (minus cerebellum) membranes were purchased from Analytical Biological Services (ABS) and prepared according to the following customer-supplied protocol. Male Wistar albino rats (250–300 g) were decapitated, the brains removed, the cerebral cortices dissected out and the rest of the brains homogenized in 10 volumes of ice-cold 50 mM Tris-HCl buffer, pH 7.7, for 30 s. The homogenate was centrifuged at 1000 g for 10 minutes, the supernatant collected and centrifuged at 35,000 g for 10 minutes. The pellet was resuspended in buffer and washed by four further 10-min centrifugations at 35,000 g. The final pellet was resuspended in buffer (2 mL/brain) and aliquots (2 mL) were frozen and stored at -80°C. Prior to use, the membrane suspension was thawed quickly at 37°C, centrifuged at 35,000 g for 10 minutes, washed once by suspension in the assay buffer and recentrifuged. The final pellet was resuspended and homogenized in the assay buffer to give the desired membrane concentration.

Guinea pig brain (minus cerebellum) membranes were also purchased from ABS and prepared according to the above protocol for rat brain membranes. The guinea pigs used were male Dunkin-Hartley with a 250–300 g body weight.

Calf brains were obtained from the local slaughterhouse and the caudate dissected over ice. Membranes were prepared as described previously (Bruinvels et al. 1992).

Cell membranes from Chinese hamster ovary (CHO) cell lines expressing recombinant α2A, α2B, α2C, and CRF2α, and cell membranes from human embryonal kidney (HEK) 293 expressing recombinant D2A and 5-HT3, were prepared in the Nervous System Department, Novartis Pharma, Basel, Switzerland. They were thawed and homogenized just before the assay. Membranes expressing recombinant 5-HT7 (Sf9 cells with baculovirus expression) were prepared by the Biotechnology Dept, Novartis Pharma, Basel, Switzerland.

The following membrane preparations were purchased from NEN Life Science Products, USA: NIH 3T3 cells expressing recombinant CCKA and CCKB; Sf9 cells expressing recombinant β1 or β2 receptors (all baculovirus expression); HeLa cells expressing recombinant 5-HT6 receptors; and CHO cells expressing recombinant NK1, NK2, or NK3 receptors. Other membrane preparations were purchased from Receptor Biology Inc: MDCK cells expressing recombinant norepinephrine transporter (hrNET); CHO cells expressing adenosine A1, A2A or A3 receptors or muscarinic M1, M2, M3, M4 and M5 receptors; HEK 293 cells expressing cannabinoid1 receptors; CHO-K1 cells expressing recombinant dopamine transporter (hrDAT); and CHO-K1 cells expressing recombinant opiate δ, opiate μ and opiate κ receptors. Membranes from CHO-K1 cells expressing recombinant 5-HT1A receptors were obtained from EuroScreen, SA.

Radioligand Binding Assays

96-well Microtiter Plate Filtration Assay

In general, the membrane preparations were homogenized after thawing and pretreated as required.

Individual radioligand binding assays for different receptors were performed as outlined by Herz et al. (1997) and references therein, with minor modifications if required. The binding studies were performed in 96-well plates (Falcon) in a total volume of 250 μL, consisting of the radioligand, drug (iloperidone or reference compound) and membrane preparation (cells, rat or guinea pig brain membranes) diluted in appropriate buffer. Non-specific binding was determined in the presence of an appropriate drug specific for the receptor under study. The plates were incubated at equilibrium for a specified time, as determined by kinetic experiments, for each receptor assay. Reactions were terminated by flash filtration and inverse transfer to 96-well filter plates (96-well cell harvester, filter plates GFC, coated with PAI as necessary, Canberra Packard). The plates were dried for 30 minutes at 56°C and sealed at the bottom with an adhesive sheet (Topseal, Canberra Packard). Subsequently, 50 μL of scintillation fluid (Microscint-20, Canberra Packard) was added to each well, the plates sealed on top and the radioactivity counted in a 96-well plate counter (Topcount, Canberra Packard). The displacement curves were established with eight concentrations (10-fold dilution steps) of iloperidone. Each concentration was tested in duplicate. The KI values are the mean value of three separate experiments.

Additional Binding Studies

Additional binding studies were performed in accordance with previously described methods: 5-HT1B/1D and 5-HT7 (Bruinvels et al. 1992; Hoyer et al. 1997). Non-specific binding was determined in the presence of 10 μM 5-HT (5-HT1 and 5-HT7 sites).

Data Analysis

A standard data reduction algorithm was used to calculate percent specific binding in the presence of the test compound as follows: ([B - NSP]/[T - NSP]) × 100where: B = binding in the presence of test compound, NSP = non-specific binding in the presence of excess inhibitor, and T = total binding.

IC50 values were derived (where feasible) from a 4-parameter logistic fit and were converted to KI values using the Cheng-Prusoff equation (Cheng and Prusoff 1973).

The entire data analysis was performed by a dedicated program linking the raw data to a custom driven Excel 7.0 macro and Graphpad Prism V 2.1. All affinities are expressed as KI values (mol/L).

RESULTS

The derived in vitro receptor binding profile of iloperidone is included in Table 1 and 2. Iloperidone displayed moderate affinity (KI 10–100 nM) at human α2C adrenoceptors (16.2 nM), human D2A (21.4 nM), human 5-HT1A (93.1 nM), bovine 5-HT1B/1D (89.1 nM) and human 5-HT6 receptors (63.1 nM). Low affinity (KI 100–1000 nM) was observed at the human α2A adrenoceptor (162 nM), human α2B adrenoceptor (162 nM), guinea pig histamine H1 (437 nM) and human 5-HT7 receptors (112 nM). In addition, iloperidone had very low affinity (KI 1000–10,000 nM) at the human CCKB receptor, human dopamine and norepinephrine transporters, and human muscarinic M1, M2, M4 receptors. There was no significant cross-reactivity (KI > 10 μM) at all other receptors tested.

Table 2 Affinity Profile of Iloperidone for Receptors for Which Human Homologue Was Unavailable
Full size table

DISCUSSION

The affinity of iloperidone for a number of receptor sites considered important for antipsychotic activity has been reported previously (Szewczak et al. 1995; Kongsamut et al. 1996). The highest affinity was observed for the rat α1-adrenoceptor (KI 0.4 nM; Szewczak et al. 1995), while the compound also displayed a high affinity (KI values below 10 nM) at the human recombinant 5-HT2A receptor (KI 5.6 nM), the short splice form (2B) of the human recombinant D2 receptor (KI 6.3 nM), and the D3 receptor (KI 7.1 nM; Kongsamut et al. 1996). These authors reported a KI value of iloperidone for the long splice form of the human D2 receptor of 13.3 nM, which is in good agreement with data obtained in the present study (21.4 nM). In addition, Kongsamut et al. (1996) reported a low affinity of iloperidone to receptors of the D1 family: D1 (KI 216 nM) and D5 (KI 319 nM). The affinity for the 5-HT2C receptor was reported as 42.8 nM.

The present study extends the radioligand binding profile of iloperidone with a large series of human receptors. A moderate affinity (KI values between 10 and 100 nM) was observed for α2C (KI 16.2 nM), 5-HT1A (93.1 nM) and 5-HT6 (63.1 nM) receptors and bovine 5-HT1B/1D receptors (89.1 nM). Since antipsychotic compounds are dosed until a significant occupation (50–80%) of D2 receptors is obtained (Nordström et al. 1993; Kapur et al. 2000), one can assume that therapeutic doses of iloperidone will result in relevant occupancy of D3, 5-HT2A, 5-HT2C, α1 and α2C receptors. Whether 5-HT1A, 5-HT1B and 5-HT6 receptors will be occupied is less certain. At therapeutic doses, in vivo binding of iloperidone at receptors with lesser affinity (e.g. the adenosine family of receptors) would be negligible. The binding profile is summarized in graphical form in Figure 2 . On the basis of this receptor binding profile one may predict some of iloperidone's clinical properties.

Figure 2
figure 2

Bargraph of selected receptor affinities (pKI in nM) of iloperidone. These affinities were measured at recombinant human receptors, with the exception of the α1-adrenoceptor (rat), the 5-HT1B receptor (bovine) and histamine H1 receptor (guinea pig). The displacement curves were established with eight concentrations of iloperidone. Each concentration was tested three times. Variation was within 5% of the mean. Data for the α1, the dopamine receptors, 5-HT2A and 5-HT2C receptor were reported previously (Kongsamut et al. 1996).

Full size image

Overall Receptor Binding Affinity

Considering the present results in conjunction with those reported by Kongsamut et al. (1996), it is evident that iloperidone binds to three classes of monoamine receptor subtypes: dopamine, serotonin and norepinephrine receptors, but not to any other neurotransmitter receptor class. The moderate to high affinity of iloperidone for human D2, 5-HT2A and α1 receptors suggest that iloperidone will have antipsychotic activity (Baldessarini et al. 1992; Willner 1997). Iloperidone's binding to 5-HT2C and α2C receptors will modify the therapeutic response and perhaps add additional therapeutic activity. On the other hand, iloperidone is devoid of affinity for receptor sites which are related to side effects (e.g. muscarine and histamine H1 receptors). The potential relevance of each individual receptor site is discussed, roughly in order of declining affinity.

Affinity for α1 Receptors

Three different subtypes of the human α1-adrenoceptor (α1A, α1B, α1D) have been identified by molecular cloning (Bylund 1992). The affinity of iloperidone to those α1-adrenoceptor subtypes is not yet known. But since iloperidone displayed a very high affinity for rat brain α1 adrenoceptors, it probably will have significant affinity for the human α1A, α1B and α1D subtypes. Animal experiments provide circumstantial evidence that α1 blockade might contribute to antipsychotic activity. For instance, prazosin administration to rats dose-dependently decreased burst firing and regularized the firing pattern of ventral tegmental dopamine neurons (Grenhoff and Svensson 1993). Disruption of prepulse inhibition by the psychotomimetic drug, phencyclidine, in rats (a putative model for the sensory motor gating deficit of schizophrenic patients), is normalized by α1-adrenoceptor blocking antipsychotic compounds (Bakshi and Geyer 1997). These two preclinical experiments suggest that inappropriate α1-adrenoceptor stimulation could be involved in the pathogenesis of schizophrenia and α1-adrenoceptor blockade could thus be a useful pharmacologic attribute for an antipsychotic compound. There are however, side effects such as postural hypotension that can occur through α1-adrenergic blockade. Clinical studies have shown that clozapine frequently causes postural hypotension early in the course of treatment (Baldessarini and Frankenburg 1991). Iloperidone has shown similar effects early in treatment but tolerance is seen to develop as treatment continues (data on file).

Affinity for 5-HT2A Receptors

As indicated in the introductory section, iloperidone adheres to the theory put forward by Meltzer et al. (1989), in that it has a higher affinity for the 5-HT2A receptor than for the D2 receptor. The theory predicts a better tolerability than classical antipsychotic compounds. Interestingly, the hallucinogen psilocybin induced schizophrenia-like psychosis in humans which was blocked by the selective 5-HT2A receptor antagonist, ketanserin (Vollenweider et al. 1998). Effective doses of psilocybin significantly decreased [11C]raclopride binding in the striatum, which is indicative for an increase in endogenous dopamine levels in this brain structure (Vollenweider et al. 1999). Unexpected though, selective 5-HT2A receptor antagonists were not very effective in the treatment of psychoses in schizophrenia (Truffinet et al. 1999; Hoechst Marion Roussel, Company Press Release, 1999). Nevertheless, if combined with D2 receptor blockade, 5-HT2A blockade might still represent a useful pharmacologic principle.

Morisset et al. (1999) have proposed that central 5-HT2A receptor blockade could represent a mechanism for improved cognition. The authors showed that antipsychotics with high 5-HT2A receptor antagonist affinity, including iloperidone, stimulated histamine neuron activity, which enhances alertness via H1 receptor activation.

Affinity for D2A, D2B, D3 and D4 Receptors

According to their different pharmacologic profile and intracellular signaling pathways, dopamine receptors have traditionally been classified into two major populations, designated D1 and D2. Molecular cloning techniques have identified additional subtypes of dopamine receptor whose profiles suggest them to be members of either the D1 or D2 families (Sokoloff and Schwartz 1995). Thus, the cloned D5 receptor resembles the classical D1 receptor in being positively coupled to cAMP production and also in terms of sequence homology and the absence of introns. On the other hand, the D3 and D4 subtypes most closely resemble the D2 receptor in being inhibitory on cAMP production and having a similar intron distribution (Sokoloff and Schwartz 1995). Alternative splicing of the D2 receptor produces a long (D2A) and a short (D2B) variant in humans (reviewed by Sokoloff and Schwartz 1995). The D2 family of receptors has been strongly linked to both the beneficial and side effects associated with antipsychotic agents. Attempts to divorce therapeutic benefit from adverse effects were given great impetus by reports that clozapine bound with high affinity and some selectivity to the dopamine D4 over D2 receptor subtypes (Van Tol et al. 1991). High affinity antagonists with marked selectivity have now been synthesized and, in some cases, examined in schizophrenic patients. No therapeutic efficacy was apparent (Kramer et al. 1997; Truffinet et al. 1999). These results indicate that D4 receptor blockade is unlikely to be a major contributor to antipsychotic activity. Although iloperidone displayed relevant affinity for the human D4 receptor (KI 25 nM; Kongsamut et al. 1996), this probably has no therapeutic consequence.

Iloperidone, like most commonly used antipsychotics, has significant affinity for the dopamine D3 receptor and is likely to interact at these sites at therapeutically relevant doses. Selective D3 receptor antagonists have been described, but these are not yet tested in schizophrenic patients. Based on the pattern of expression of the D3 receptor, this subtype could, however, be relevant for antipsychotic activity. The D3 receptor is mainly expressed in mesocorticolimbic projection areas such as the medial forebrain bundle, the shell of the nucleus accumbens, olfactory tubercle, amygdala and cortical structures, but less in the nigrostriatal and tuberoinfundibular dopamine systems (Bouthenet et al. 1991).

Finally, differences in affinity for the two splice forms of the human D2 receptor have also attracted attention (Malmberg et al. 1993; Usiello et al. 2000). Whereas most antipsychotic compounds display equal affinity for both variants, clozapine and remoxipride, two compounds with a low propensity to cause EPS, bound with higher affinity to the D2B than to the D2A form of the receptor. Usiello et al. (2000) found that the cataleptic effect of the typical antipsychotic haloperidol was absent in D2A receptor deficient mice. These authors suggested that therapeutic activity could be related to blockade of presynaptic D2B receptors, whereas extrapyramidal side effects could be avoided if the compound would fail to bind to D2A receptors. As published by Kongsamut et al. (1996), iloperidone displays higher affinity for the D2B than for the D2A form which, according to the foregoing, would indicate a reduced propensity to cause EPS.

Affinity for α2A and α2C Adrenoceptors

Iloperidone's next highest affinity is to the norepinephrine α2C binding site. Clozapine displays nanomolar affinity for the α2C adrenoceptor (KI 9.1 nM) and some selectivity relative to the α2A subtype (KI 50 nM) (Schotte et al. 1996). It has been speculated that clozapine's superior therapeutic activity is at least partly explained by α2-adrenoceptor blockade (Nutt 1994; Litman et al. 1996; Hertel et al. 1999). Other therapeutic effects of clozapine may also be related to α2-adrenoceptor blockade. In monkeys with MPTP-induced Parkinsonism, dyskinetic movements induced by L-dopa were diminished by idazoxan, an α2 antagonist (Henry et al. 1999; Grondin et al. 2000). Thus, blockade of α2-adrenergic receptors could explain the potent antidyskinetic effect of clozapine (Bennett et al. 1994; Pierelli et al. 1998). Experiments in genetically altered mice show that overexpression of α2C receptors contributes to behavioral despair and accordingly, blockade of α2C receptors could be antidepressive (Sallinen et al. 1999). Other preclinical studies indicate that overexpression of α2C receptors worsens spatial recognition and induces anxiety-like behavior (Björklund et al. 1998, 1999). These effect were reverted by an α2 antagonist, suggesting that blockade of α2C receptors might result in improved cognition and anxiolytic activity. On the other hand, also undesired properties may be related to blockade of α2-adrenoceptors. Idazoxan and other α2 antagonists are proconvulsant in mice (Jackson et al. 1991). Such an effect was not seen in a mutant mice strain that lacked functional α2A receptors (Janumpalli et al. 1998), suggesting that proconvulsant activity is related to blockade of α2A adrenergic receptors.

Iloperidone has a low affinity for α2A receptors indicating a low propensity to induce convulsions. In contrast, the affinity of iloperidone for α2C-adrenoceptors could be of clinical relevance as it might result in antidepressant and anxiolytic activity and in improved cognition.

Affinity for 5-HT2C Receptors

Recent research has shown that the 5-HT2C antagonist, SB206,553 dose-dependently increased the firing rate of VTA and locus coeruleus (LC) adrenergic neurons in rats (Gobert et al. 2000). These authors also reported that the 5-HT2C antagonist dose-dependently increase levels of dopamine (DA) and noradrenaline (NA) but not serotonin in the frontal cortex. Clozapine has been shown to preferentially increase dopamine release in the medial prefrontal cortex (Moghaddam and Bunney 1990). Since clozapine is a potent 5-HT2C antagonist (Schotte et al. 1996), the effects of clozapine on dopamine release in the prefrontal cortex might be due, at least partly, to blockade of 5-HT2C. Dopamine hypofunction in cortical dopamine projection has been suggested to be responsible for negative symptomatology (Davis et al. 1991). Iloperidone shows moderate affinity to the 5-HT2C receptors (KI 42.8 nM; Kongsamut et al. 1996) which might lead to disinhibition of the VTA and LC neurons, enhanced cortical DA and NA in the frontal cortex and thus an effect against negative symptoms of schizophrenia.

Activation of 5-HT2C receptors by fenfluramine or mCPP suppresses food intake in laboratory animals. For this reason the blockade of 5-HT2C receptors is suspected to contribute to the hyperphagia and weight gain observed with antipsychotic treatment. Chronic treatment of rats with selective 5-HT2C receptor antagonists did not lead to weight gain (Kennett et al. 1997). Also the antipsychotic drug ziprasidone displays relevant 5-HT2C blockade (Schotte et al. 1996), but did not induce profound weight gain in humans (Allison et al. 1999). Thus, although 5-HT2C receptor agonists induce hypophagia, the opposite is not necessarily observed with antagonists. Also iloperidone, despite its affinity for 5-HT2C receptors, produced, if compared to placebo, minimal weight gain in schizophrenic patients (data on file).

Selective 5-HT2C receptor antagonists displayed anxiolytic activity (Kennett et al. 1996, 1997) and suppressed haloperidol-induced catalepsy (Reavill et al. 1999). On theoretical grounds, 5-HT2C receptor antagonists might also be useful for the treatment of Parkinson's disease (Fox and Brotchie 1999). Iloperidone's affinity for 5-HT2C receptors could, therefore help to explain the low propensity to induce catalepsy and the anxiolytic effects seen in animal studies (Corbett et al. 1993).

Affinity for 5-HT6 Receptors

The relevance of 5-HT6 receptor blockade in the pharmacologic profile of antipsychotic compounds remains speculative. Roth et al. (1994) reported that clozapine and several atypical antipsychotic agents (rilapine, olanzapine, tiospirone, fluperlapine, clorotepine and zotepine) had high affinities for the 5-HT6 receptor. It is interesting that selective 5-HT6 receptor antagonist, Ro 04-6790 in rats increased cholinergic neurotransmission (Bentley et al. 1999). Similarly, clozapine, which has relevant 5-HT6 receptor affinity (Glatt et al. 1995; Schotte et al. 1996), increased extracellular levels of acetylcholine in rat prefrontal cortex (Parada et al. 1997). It is therefore conceivable that clozapine increases extracellular levels of acetylcholine via blockade of 5-HT6 receptors. Blockade of 5-HT6 receptors could thus, via increased cholinergic neurotransmission, contribute to improved cognitive function. The present results show that iloperidone has moderate affinity for 5-HT6 receptors, while the affinity for muscarine receptors is low. These two features in combination suggest that iloperidone could display efficacy against neurocognitive deficits in patients with schizophrenia. However, it is not certain whether therapeutic doses of iloperidone will be high enough to obtain significant occupancy of the 5-HT6 receptor, as its affinity measured in the present experiments is rather moderate (KI 63.1 nM).

Affinity for 5-HT1A Receptors

The same argument holds for the 5-HT1A receptor since the affinity of iloperidone for human 5-HT1A receptors amounted to 93.1 nM, only. In a cellular assay, iloperidone produced a concentration-dependent surmountable antagonism against the 5-HT1A receptor agonist, 8-OH-DPAT (mean [S.E.M.] pKB 7.69 [0.18]; data on file).

Post-mortem studies of patients with schizophrenia have revealed increased numbers of 5-HT1A receptors in the prefrontal cortex (Hashimoto et al. 1993; Burnet et al. 1997). Intrinsic activation of these receptors hyperpolarizes the neurons and reduces the output of their neurotransmitter glutamate. Loss and/or hypoactivity of cortical glutamatergic neurons has been postulated to underlie the cognitive impairment in Alzheimer's disease (Francis et al. 1993). Therefore, the normalization of glutamate output by a 5-HT1A antagonist such as iloperidone might help to ameliorate cognitive impairment in patients with schizophrenia.

Affinity for Muscarinic Receptors

The low or negligible affinity of iloperidone for muscarinic receptors indicates that it may have a low propensity to cause side effects such as dry mouth, blurred vision, increased frequency of micturition or other anticholinergic effects.

Affinity for Histamine H1 Receptors

It is remarkable that antipsychotic drugs with high affinity for histamine H1 receptors, like clozapine, olanzapine or thioridazine cause profound increases in body weight, whereas compounds with smaller H1 affinity are less active in this respect (Allison et al. 1999). Also other compounds with high H1 receptor affinity like the antidepressant mirtazapine (Fawcett and Barkin 1998), the antihypertensive ketanserin (Brogden and Sorkin 1990), or the migraine prophylactic pizotifen (Cleland et al. 1997) induce significant weight gain. These clinical observations are supplemented by preclinical studies showing that application of H1 receptor antagonists or depletion of histamine elicits a feeding response (Doi et al. 1994). Iloperidone displays low affinity for the histamine H1 receptor and indeed has shown little effect on body weight of schizophrenic patients (data on file).

Affinity for Other Receptors

The affinity of iloperidone for other receptors such as the 5HT3 and nicotine cholinergic receptors is less than 10 μM (pKI < 5.0), indicating that iloperidone is inactive at these sites.

CONCLUSION

Iloperidone is characterized by a broad spectrum of dopamine, norepinephrine and serotonin antagonism. Thus, as with other new antipsychotics, iloperidone has a high affinity for 5-HT2A receptors and α1 adrenergic receptors and moderate affinity for D2 receptors, indicating antipsychotic efficacy, with a reduced propensity to induce EPS. Favourable properties are also suggested by its additional receptor profile. The moderate D2 receptor affinity is balanced by comparable affinity for α2C adrenoceptors, and 5-HT2C, suggesting potential improvements in cognition and negative symptoms. Moreover, blockade of α2C adrenoceptors might translate into antidepressant and anxiolytic activity. Low affinity for histamine H1 receptors suggests that iloperidone has limited propensity to induce weight gain. Extremely low activity at cholinergic receptors suggests that side effects associated with anticholinergic agents such as dry mouth, blurred vision and increased frequency of micturition will be avoided. Due to the low affinity of iloperidone for α2A receptors proconvulsive activity is not expected.

This broad receptor binding profile indicates that iloperidone has the potential to be an effective and well-tolerated agent in the treatment of psychotic disorders. Indeed, preliminary studies have confirmed the favourable efficacy and tolerability of iloperidone in patients with schizophrenia (Davidson et al. 1994; Borison et al. 1996; Cutler et al. 1996).