Preclinical Evaluation of the Effects of Trazpiroben (TAK-906), a Novel, Potent Dopamine D2/D3 Receptor Antagonist for the Management of Gastroparesis

word count: 250/250 words Introduction word count: 689/750 words Discussion word count: 1500/1500 words


SIGNIFICANCE STATEMENT 80/80 words
Trazpiroben is a novel, potent dopamine D 2 /D 3 selective receptor antagonist designed to avoid adverse effects associated with the current pharmacological therapies, metoclopramide and domperidone. Pre-clinical studies have demonstrated low brain penetration and weak affinity for the hERG channel, indicating that trazpiroben is not expected to be associated with central nervous system or cardiovascular safety issues. With these pharmacological properties, trazpiroben may represent a viable new treatment option for gastroparesis due to a potentially improved safety profile relative to existing therapies.

Introduction
Gastroparesis, a chronic gastric motility disorder, is distinguished by delayed gastric emptying and/or gastric dysrhythmia with diminished peristaltic coordination in the absence of mechanical obstruction of the gastric outlet Gharibans et al., 2019;Grover et al., 2019). Patients report chronic symptoms with periods of exacerbation, which commonly include early satiety, upper abdominal discomfort and/or bloating following a meal, postprandial fullness and nausea and vomiting . Symptoms frequently mimic those observed with other conditions such as functional dyspepsia, leading to issues in identifying and diagnosing the disease (Parkman et al., 2004;Camilleri et al., 2018;Tack and Camilleri, 2018). Gastroparesis may stem from a number of different etiologies, including idiopathic, diabetic or post-surgical causes, which in turn frequently determine the reported pattern and severity of symptoms Nassar and Richter, 2018). Rising disease severity is associated with significant morbidity and healthcare utilization, however, despite the burden associated with gastroparesis, substantial gaps remain in describing the underlying mechanisms leading to the disease, and the interplay between upper gastrointestinal symptoms and gastric emptying requires further investigation (Janssen et al., 2013;Grover et al., 2019).
In addition to challenges in identifying and diagnosing gastroparesis, the current therapeutic landscape remains limited, and includes a patient-guided combination of conservative, medical or surgical treatment options, such as dietary control, gastric electrical stimulation, and pharmacological therapies Tack and Camilleri, 2018). These pharmacological therapies include dopamine receptor antagonists, which can reduce gastroparesis symptoms and are effective at establishing normal gastric myoelectric activity 7 and resolving gastric dysrhythmias (Koch et al., 1989;Acosta and Camilleri, 2015;Gharibans et al., 2019). These benefits occur through reversing the actions of dopamine, which exerts a direct relaxant effect on the gut musculature through activation of muscular D2 receptors in the lower stomach and esophageal sphincter, reducing gut motility (Lee and Kuo, 2010). Dopamine receptor antagonists also offer anti-emetic benefits through preventing D 2 receptor activation in the chemoreceptor trigger zone (CTZ) (Lee and Kuo, 2010).
Two agents with dopamine receptor antagonist activity, metoclopramide and domperidone, are available for the symptomatic management of gastroparesis, though only limited use is permitted owing to safety issues. Metoclopramide, a 5-hydroxytryptamine 4 (5-HT 4 ) receptor agonist and D 2 /D 3 receptor antagonist that penetrates the blood-brain barrier (BBB), carries the risk of potentially serious movement disorders known as extrapyramidal side effects (EPS), in particular tardive dyskinesia as a result of central nervous system (CNS) D 2 receptor blockade. Hence, metoclopramide is labelled only for short-term use, and the drug carries a black box warning from the US Food and Drug Administration regarding treatment lasting longer than 3 months (Bateman et al., 1985;Parkman et al., 2012;Enweluzo and Aziz, 2013).
Domperidone is a D 2 /D 3 receptor antagonist with limited brain penetration, meaning it does not elicit the same CNS adverse effects as metoclopramide, and has been approved by the European Medicines Agency for short-term, low-dose use. Domperidone has been associated with QT interval prolongation, a delay in cardiac polarization that can lead to torsades de pointes, a potentially fatal form of ventricular arrhythmia and sudden cardiac death 8 (Hondeghem, 2011;Renoux et al., 2016;Hellstrom and Al-Saffar, 2018). Owing to these potential effects, domperidone has not been approved by the Food and Drug Administration and is only available in the USA via a single-patient Expanded Access Investigational New Drug Application. A likely mechanism for these cardiac effects is domperidone's inhibitory effect on the human ether-à-go-go-related gene (hERG) channel, with a half-maximal inhibitory concentration (IC 50 ) of 57 nM (Claassen and Zunkler, 2005). As domperidone is primarily metabolized by the cytochrome P450 3A4 enzyme, the risk of cardiovascular adverse events (AEs) is increased when it is administered with inhibitors of this enzyme (Boyce et al., 2012;Chen and Hsiao, 2015). Consequently, there is an unmet need for an effective therapy with a favorable safety profile for the management of gastroparesis.
Trazpiroben is a novel, potent D2/D3 receptor antagonist designed to avoid the potential AEs associated with other dopamine receptor antagonists. This article describes a series of preclinical in vitro and in vivo studies used to define the pharmacological properties of trazpiroben. Trazpiroben,3,decan-3yl]methyl}benzoic acid 1:1 maleic acid salt (Fig. 1), was previously known as TAK-906, ATC-1906M, andATI-24380. ATC-1906M was

Affinity at Human Dopamine Receptors.
The affinity of trazpiroben and domperidone was examined at human D 1 , D 2 , D 3 , D 4 , and D 5 receptors using genetically engineered cell lines expressing recombinant human dopamine receptors; experiments were performed by Cerep, Inc. (Poitiers, France) according to Cerep's published protocols as described below. The affinity of trazpiroben for cloned human dopamine receptors was measured using radioligand binding displacement assays in membrane preparations from stably transfected cell lines as follows: D 1 in Chinese hamster ovary (CHO) cells (Zhou et al., 1990); D 2S in human embryonic kidney (HEK) 293 cells (Grandy et al., 1989); D 2L in human recombinant CHO cells (Grandy et al., 1989;Hayes et al., 1992); D 3 in CHO cells (MacKenzie et al., 1994);D 4.4 in CHO cells (Van Tol et al., 1992); and D 5 in GH4 cells (Sunahara et al., 1991).
Specific ligand binding to the receptors was defined as the difference between total binding and nonspecific binding determined in the presence of an excess of unlabeled ligand. Results were expressed as the percentage of control-specific binding ([measured specific binding/control-specific binding] × 100) and as the percentage inhibition of control-specific binding (100 − [(measured specific binding/control-specific binding) × 100]) obtained in the presence of trazpiroben.
IC 50 s (concentrations causing half-maximal inhibition of control specific-binding) and Hill coefficients (nHs) were determined using nonlinear regression analysis of competition curves generated with mean replicate values using Hill equation curve fitting (Y = D + [(A − D)/(1 + [C/C 50 ] nH )], where Y is specific binding, D is minimum specific binding, A is maximum specific binding, C is the compound concentration, C 50 is the IC 50 , and nH is the slope factor). This analysis was performed using software developed at Cerep (Hill software) and validated by comparison with data generated by commercial software SigmaPlot ® 4.0 for Windows ® ( © 1997 SPSS Inc., Chicago, IL).
Inhibition constants (K i s) were calculated using the Cheng Prusoff equation (K i = IC 50 /[1 + (L/K D )], where L is the concentration of radioligand in the assay and K D is the affinity of the radioligand for the receptor). A Scatchard plot was used to determine the K D .

Receptor Specificity Profiling.
Trazpiroben (as the hydrochloride salt) was tested at a single concentration of 1 µM in an affinity profiling panel of 84 receptors and domperidone was tested at 100 nM (Cerep).
Displacement of known ligands (percentage inhibition) was used to predict trazpiroben and domperidone affinity for these receptors. Binding was deemed significant if inhibition exceeded 50%. Significant binding other than that to D 2 , D 3 , and D 4 receptors was observed for human α 1A , α 1B , 5-HT 1A , 5-HT 2A , and 5-HT 7 receptors. Concentration-response curves were constructed with trazpiroben and domperidone at these receptors to determine affinities.
Subsequently the affinities of trazpiroben (as the maleate salt) and domperidone were compared at α 1B and 5-HT 2A receptors.

Concentrations of Prolactin in Serum after Multiple Oral Doses of Trazpiroben to Rat
In the rat, following a 4-day acclimatization period, individual doses of trazpiroben were calculated based on body weights measured on the first day of dosing. The study involved five male and five female Sprague Dawley rats. Rats were dosed by oral gavage once daily at 100 mg/kg. Blood was collected for serum prolactin determination as follows: day 1, predose and at 1, 6, and 24 h post-dose; day 3, pre-dose and at 1, 6, and 24 h post-dose. Daily observations of the health of the animals were made.
Trazpiroben was prepared in 5% (v/v) dimethyl sulfoxide, 40% (v/v) polyethylene glycol 400, and 55% (v/v) 0.9% sodium chloride for injection on study day 1 for all doses. When not in use the formulations were stored at 2-8°C, protected from light. Rats were dosed at 100 mg/kg via a ball-tipped gavage needle for 4 consecutive days. For determination of prolactin concentration, blood (approximately 0.4 mL) was collected from a jugular vein via a syringe and needle, and transferred into serum separator tubes on days 1 and 3 pre-dose and at 1, 6, and 24 h (prior to the next daily dose, as applicable) post-dose. Blood samples This article has not been copyedited and formatted. The final version may differ from this version. intended for prolactin quantification were processed to serum, frozen, and stored at −70°C (± 15°C) until analysis.
All samples were evaluated for prolactin using an assay based on the competition between unlabeled prolactin and a fixed quantity of 125I-labeled prolactin for a limited number of binding sites on the prolactin-specific antibody. With fixed amounts of antibody and radioactive ligand, the amount of ligand bound by the antibody was inversely proportional to the concentration of prolactin in the sample. The antibody bound-prolactin was then reacted with a second antibody bound to a magnetizable particle. Separation of the antibody boundfraction was accomplished via centrifugation and decantation of the supernatant.
Measurement of the radioactivity in the pellet allowed the amount of labeled prolactin in the bound fraction to be calculated (lower limit of quantification for prolactin was 0.8 ng/mL).
Concentration of the unlabeled prolactin was then determined by interpolation from the standard curve, and results entered into the ClinAxys system v2.6.1. (Clinical Systems Ltd., FL, USA). Assays were performed at Antech Diagnostics (Morristown, NJ).

Concentrations of Prolactin in Serum after Multiple Oral Doses of Trazpiroben to Dog.
This study included five male and five female Purebred beagles (non-naïve) from the Covance stock colony. Animals were individually housed in steel cages during treatment and at least 4 hours following each dose administration to allow monitoring of any test articlerelated effects. Certified Canine Diet #5007 (PMI) was provided ad libitum. Diets were supplemented with canine treats and water was provided fresh daily ad libitum. Following a 3-day acclimatization period, individual doses of trazpiroben were calculated based on body weights measured on the first day of dosing (8.3-12.5 kg). Trazpiroben was prepared in This article has not been copyedited and formatted. The final version may differ from this version.
JPET Fast Forward. Published on July 12, 2021 as DOI: 10.1124/jpet.121.000698 at ASPET Journals on August 21, 2021 jpet.aspetjournals.org Downloaded from capsules (size 12, ¼ oz) by Covance on Study Day 1 for all doses. When not in use, the formulations were stored at ambient temperature, protected from light. Dogs were then dosed via oral capsule (five capsules followed by at least 10 mL of water) once daily at 50 mg/kg for four consecutive days. Blood (approximately 1 mL) was collected from a jugular vein for serum prolactin determination as follows: day 1, pre-dose and at 1, 6, and 24 h post-dose; day 3, pre-dose and at 1, 6, and 24 h post-dose. Daily observations of the health of the animals were made. Handling of blood samples was as described for the rat.
All samples were evaluated for prolactin using a solid phase enzyme immunometric assay in the microplate format, designed for the quantitative measurement of canine prolactin (lower limit of quantification for prolactin was 0.2 ng/mL). The standard curve demonstrated a direct relationship between optical density and prolactin concentration, which allowed for determination of prolactin concentration in controls and unknowns using the SoftMax Pro software (Molecular Devices, Sunnyvale, CA). Results were entered into the ClinAxys system v2.6.1 for reporting. Assays were performed at Antech Diagnostics (Morristown, NJ).

Apomorphine-Induced Emesis in Dog.
Groups of male and female beagle dogs (two male and two female per group) weighing 8-15 kg were used. Animals were fasted overnight before the experiment (water only). The dogs were administered 0.5 mL/kg of trazpiroben oral solution (0.03, 0.1, 0.3, or 1 mg/kg as the hydrochloride salt) or vehicle (10% dimethyl sulfoxide, 40% hydroxypropyl-β-cyclodextrin in sterile water), followed 1 h later by a single subcutaneous injection of 0.3 mL/kg apomorphine hydrochloride (0.3 mg/kg), a highly emetogenic compound. Domperidone and metoclopramide (oral solution) were used as comparators. The latency to first retch and This article has not been copyedited and formatted. The final version may differ from this version. vomit and the number and time of vomit episodes were recorded for 60 min following apomorphine injection. Vomiting was defined as oral expulsion of liquid or solid stomach contents. Meaningful differences were evaluated using nonparametric methods (Wilcoxon rank sum test).

Accelerating Rotarod Performance in Rat.
Male Sprague Dawley rats weighing 180 ± 30 g were obtained from BioLasco Taiwan (under Charles River Laboratories License, Wilmington, MA). Trazpiroben was dissolved in 5% dimethyl sulfoxide/40% polyethylene glycol 400/55% saline. The dosing volume was 10 mL/kg. Groups of eight rats were included in the study based on pre-dose rotarod performance. Animals were trained on a turning rod rotating at a continuous accelerating speed from 4 to 36 rpm over 4 min, for at least three times on day 0. Rats were randomly assigned to different groups with similar baseline training values. The vehicle, trazpiroben at 1, 10, and 30 mg/kg, and chlorpromazine at 30 mg/kg, were orally administered on day 1.
Then, 30 and 60 min later, the animals were placed on the accelerating rotarod (increasing from 4 to 36 rpm during the 4-min period). The time (in s) the animal remained on the turning rotarod was recorded. One-way analysis of variance followed by Dunnett's test was applied for comparison between the vehicle control group and test compound-treated groups. P < 0.05 was considered statistically meaningful.

Concentrations of Trazpiroben in Plasma after Single and Multiple Oral Doses to Rat and
Dog.
These two studies examined the effect of initial and multiple oral administration of oncedaily trazpiroben 100 and 50 mg/kg in the rat and dog, respectively, on plasma trazpiroben This article has not been copyedited and formatted. The final version may differ from this version. For analysis, blood was first processed to plasma via protein precipitation, then 50 µL plasma and 50 µL of warfarin internal standard solution (100 ng/mL) were covered with aluminum foil and vortex mixed for at least 30 seconds. Next, 500 µL of 1% formic acid in acetonitrile was added to the sample, which was then vortexed for approximately 3 min followed by centrifugation at approximately 1600 g for 5 min at room temperature. A 50 µL aliquot of the supernatant was transferred to a 96-well plate that contained 350 µL methanol:water (50:50, v:v). The plate was covered with a dimpled sealing-mat and vortex mixed at low speed for approximately 1 min.
The high-performance liquid chromatography (HPLC) column was a Waters xBridge C18 50 x 2.1 mm column (Milford, MA) with 5 µm particle size maintained at 35°C. The mobile phase was as follows: A, 5 mM ammonium formate in 0.1% formic acid in water; B, 0.1% formic acid in methanol. The initial flow was 0.600 mL/min at 50% B, and was increased to 95% B and then back to 50% B over a cycle time of approximately 4.5 minutes (injection start to next injection start).
Mass spectrometer was a triple quadrapole Sciex API 5000/5500 using positive ion 163.1. The calibration curve was obtained by linear regression on trazpiroben/warfarin peak area ratio with a weighting factor of 1/x 2 . Trazpiroben sample concentrations were reported in ng/mL. The method criteria for calibrator precision and accuracy was ≥75% of standards within ±15% (±20% lower limit of quantification), and the criteria for QC precision and accuracy was ≥75% of standards within ±15%. Matrix freeze-thaw stability was 4 cycles, sample collection stability was 2 hours at room temperature and on wet ice and frozen matrix stability was at least 114 days at -10 to -30°C, and at least 114 days at -60 to -80°C.

Plasma Protein Binding.
Plasma protein binding of trazpiroben was determined in rat (Sprague Dawley) and dog (beagle) plasma using equilibrium dialysis with HPLC MS/MS quantification. The concentration of the test compound was 10 µM. A reference panel of other compounds (acebutolol, quinidine, and warfarin) with known plasma protein binding was also run at the same concentration.

Concentrations of Trazpiroben in Cerebrospinal Fluid after Multiple Oral Doses to Rat and
Dog.
These studies examined the effect of multiple oral administration of once-daily trazpiroben 100 mg/kg in rat and 50 mg/kg in dog on cerebrospinal fluid (CSF) trazpiroben levels to estimate brain exposure. These studies were performed at Covance Laboratories as part of the same experiments described earlier for plasma determinations of trazpiroben levels, and details of animals, dosing, and the timing of blood collection were as summarized there.

Effects of Trazpiroben on the hERG Channel Current.
In vitro effects of trazpiroben were assessed on the hERG channel current (a surrogate for the rapidly activating delayed rectifier cardiac potassium current [I Kr ]; (Redfern et al., 2003)). EGTA, 5; ATP, 4; HEPES, 10; pH adjusted to 7.2 with KOH. Recording took place at 33-35ºC using a combination of inline solution pre-heater, chamber heater, and feedback temperature controller. Cells stably expressing hERG were held at −80 mV. Onset and steady-state inhibition of the hERG potassium current due to trazpiroben were measured using a pulse pattern with fixed amplitudes (conditioning prepulse +20 mV for 1 s; repolarizing test ramp to -80 mV [-0.5 V/s] repeated at 5 s intervals). Each recording ended with a final application of a supramaximal concentration of reference substance to assess the contribution of endogenous currents. The remaining uninhibited current was subtracted offline digitally from the data to determine the potency of the test substance for hERG inhibition.
Trazpiroben was applied to each cell (n = 3) at different concentrations. Peak current was measured during the test ramp and a steady state was maintained for at least 20 s before applying test compound or positive control. Peak current was measured until a new steady state was achieved. Test compound concentrations verified by dose solution analysis were used in IC 50 calculations. Domperidone was also tested by ChanTest Corporation in duplicate using the automated

Affinity at Dopamine Receptors.
Trazpiroben exhibited high affinity for D 2 and D 3 receptors (K i = 1.5 nM for D 2S , 3.1 nM for D 2L , and 3.2 nM for D 3 ), moderate affinity for D 4 receptors (K i = 80 nM for D 4.4 ), and very weak affinity for D 1 and D 5 receptors (negligible affinity observed at 1µM) in stably transfected cell membrane preparations. The affinity of trazpiroben for the human D 2S and D 3 receptors (K i = 1.5 nM and 3.2 nM, respectively) was similar to that of domperidone (0.70 nM and 2.3 nM, respectively; Fig. 2).

Receptor Specificity Profiling.
Trazpiroben had affinity for α-adrenergic (α 1A and α 1B ) and serotonergic (5-HT 1A , 5-HT 2A , and 5-HT 7 ) receptors at the screening concentration of 1 µM. It had negligible affinity for the additional molecular targets tested, suggesting a low probability of off-target AEs. This article has not been copyedited and formatted. The final version may differ from this version. Concentration-response curves were constructed with trazpiroben at these receptors to determine affinities. Trazpiroben affinity for each of these receptors was low for α 1A (K i = 150 nM), 5-HT 1A (K i = 230 nM), and 5-HT 7 (K i = 1.5 µM) receptors and moderate for α 1B (K i = 30 nM) and 5-HT 2A (K i = 66 nM) receptors. Domperidone had a similar receptor affinity profile to trazpiroben, with low affinity at α 1A (K i = 110 nM) and 5-HT 7 (62% inhibition at 0.1µM) receptors and moderate affinity at α 1B (K i = 16 nM) and 5-HT 2A (K i = 19 nM) receptors. These results demonstrated a receptor binding profile for trazpiroben that closely resembled that of domperidone, although these compounds have distinctly different molecular structures.

Functional Assay in Membrane Preparations Expressing the D 2L Receptor.
Functional activity of trazpiroben at the D 2L receptor was determined as G-protein activation in comparison with the agonist dopamine by measurement of Eu-GTP binding in fluorescence assays (Fig. 3). Trazpiroben did not act as an agonist or partial agonist at D 2L receptors in this assay (Fig. 3A). Trazpiroben and domperidone are D 2L receptor antagonists with IC 50 s of 1.8 nM and 1.7 nM, respectively (Fig. 3B). Thus, trazpiroben is a neutral antagonist at the D 2L receptor.

Prolactin Levels in Rat Following a Single Dose of Trazpiroben or Domperidone.
Trazpiroben produced increases in serum prolactin concentrations in rat relative to concurrent control values following oral gavage doses of 0.1-10 mg/kg. A maximal increase in serum prolactin concentrations was observed in animals administered 1 mg/kg, and increases of similar magnitude were seen at 3 and 10 mg/kg (Fig. 4). This article has not been copyedited and formatted. The final version may differ from this version. Domperidone produced increases in serum prolactin concentrations in rat relative to concurrent control values following oral gavage doses of 1-10 mg/kg. Increases in serum prolactin levels observed in rat administered 1, 3, and 10 mg/kg were similar in magnitude, and were also similar to those observed following trazpiroben dosing (Fig. 5).

Concentrations of Prolactin in Serum after Multiple Oral Doses of Trazpiroben to Rat and
Dog.
Oral administration of trazpiroben increased serum prolactin concentrations in both male and female rats (Fig. 5A) and male and female dogs (Fig. 5B) relative to pre-dose concentrations from the initial dose onwards. On day 1, mean serum prolactin concentrations increased relative to pre-dose concentrations and remained elevated at all subsequent time points.
Maximal increases in mean serum prolactin concentrations were observed at 1 hour post-dose on Days 1 and 3. Mean serum prolactin concentrations at 1 h post-dose on Day 1 were approximately 10-and 20-fold greater than pre-dose concentrations in male and female rats, respectively, and 3-and 6-fold greater in male and female dogs, respectively. In both rat and dog, prolactin levels post-dose on day 3 were similar to those on day 1.

Apomorphine-Induced Emesis in Dog.
Apomorphine-induced emesis in beagle dogs is a well-established preclinical model for testing the emetic potential of compounds. (Parkinson and Grasso, 1993) Emetic response in the dog and humans is, among other factors, under the control of D 2 receptors in the CTZ, which is located outside the BBB. In this parallel, single-dose administration study, trazpiroben resulted in statistically significant decreases in retching and vomiting compared This article has not been copyedited and formatted. The final version may differ from this version. to vehicle-treated dog (P =0.0078; differences evaluated using non-parametric Wilcoxon rank sum test [ Fig. 6]) at 0.3 and 1 mg/kg doses and was at least equipotent in antiemetic activity to domperidone.

Accelerating Rotarod Performance in Rat.
Trazpiroben dosed at 1, 10, and 30 mg/kg orally did not affect motor coordination 30 and 60 min after its administration in the rat. In comparison, the centrally acting D 2 receptor antagonist chlorpromazine, given orally at 30 mg/kg, substantially reduced the time rats spent on the accelerating rotarod 30 and 60 min post-dose (Fig. 7).

Dog.
Trazpiroben was rapidly absorbed after the initial dose and each repeat dose in rat and dog.
High plasma concentrations of trazpiroben were observed in all animals at 1 and 6 h after administration on days 1 and 3 (Table 1). At 24 h after dosing, plasma concentrations of trazpiroben were close to the lower limit of quantification (measured on days 1 and 3).
Maximal mean plasma trazpiroben concentrations were higher in female than in male rats.
Concentrations in each animal were more variable in dog than in rat and it was not possible to determine differences between the sexes. Concentrations of trazpiroben after multiple doses did not differ greatly from that measured after the initial dose in both rat and dog (Table 1). This article has not been copyedited and formatted. The final version may differ from this version. The distribution of trazpiroben was derived from in vitro studies of plasma protein binding.
Plasma protein binding was 93% in rat plasma and 92% in dog plasma.

Concentrations of Trazpiroben in CSF after Multiple Oral Doses to Rat and Dog.
CSF concentrations of trazpiroben were very low compared with plasma concentrations ( Table 2)

Effects of Trazpiroben on the hERG Channel Current.
In good laboratory practice manual patch clamp electrophysiology studies of mammalian cells, trazpiroben only weakly blocked the hERG potassium channel current ( Table 3; when fitted to a curve, the IC 50 was 15.6 μM with an nH of 1.3, curve not shown) relative to the positive control compound terfenadine (mean ± SD, 82.7 ± 0.3% inhibition at 60 nM), representing a 10,000-fold selectivity compared with the affinity (1.5 nM) at the target D 2 receptor. In a separate study, domperidone inhibited the hERG current more than 100-fold more potently than trazpiroben, with an IC 50 of 120 nM (data not shown). This potency for hERG current inhibition by domperidone is similar to the previously reported value of 57 nM (Claassen and Zunkler, 2005). This article has not been copyedited and formatted. The final version may differ from this version.

Discussion
Gastroparesis is a gastric motility disorder typified by gastric dysrhythmia and/or slow gastric emptying without mechanical obstruction (Koch, 2014;Gharibans et al., 2019;Camilleri et al., 2018). Despite the associated disease burden and healthcare resource utilization, pharmacological treatments remain limited, and those available carry safety concerns. Trazpiroben, a selective D 2 /D 3 receptor antagonist developed for the chronic treatment of gastroparesis, was designed to avoid the safety issues present with current therapies. In these pre-clinical studies, the pharmacological properties of trazpiroben were evaluated to assess its viability as a treatment for gastroparesis.  , 1999;Kashyap et al., 2009;Lee and Kuo, 2010). D 2 receptors antagonists disinhibit acetylcholine release and increase lower esophageal sphincter and gastric tone, promoting gastric emptying. D 3 receptors also play a role in gastric motility, with D 2 /D 3 receptor activation significantly delaying gastric emptying in the rat, which was partially reversed by administration of a D 3 antagonist (Kashyap et al., 2009). In the area postrema, D 2 /D 3 receptor antagonism has been shown to reduce nausea and vomiting (Yoshikawa et al., 1996;Darmani et al., 1999;Welliver, 2014). Importantly, as the area postrema lies outside the BBB, peripherally selective D 2 /D 3 antagonists can still provide antiemetic benefits (Andrews et al., 1990;Yoshikawa et al., 1996). As a peripherally selective D 2 /D 3 antagonist, trazpiroben is expected to normalize gastric myoelectric rhythm, exhibit prokinetic effects and reduce nausea and vomiting without the potential CNS effects observed with other therapies. This multimodal action renders trazpiroben a promising therapeutic agent for treating gastroparesis.
Adverse effects associated with other D 2 /D 3 antagonists have limited their use when treating gastroparesis in the clinic (Meltzer, 2013;Michaud and Turgeon, 2013;Giudicessi et al., 2018). Metoclopramide is a D 2 /D 3 antagonist and 5-HT 4 agonist approved and marketed in the USA for treatment of symptomatic gastroesophageal reflux and diabetic gastroparesis (maximum 12 weeks duration) (United States Food and Drug Administration). Despite its efficacy, the dose and duration of metoclopramide treatment are limited by well-documented toxicities, most notably EPS (Meltzer, 2013). Trazpiroben does not readily penetrate the BBB and is therefore unlikely to elicit such adverse events. Domperidone, an antiemetic and prokinetic agent, is a D 2 /D 3 antagonist that inhibits the hERG channel and is associated with a risk of drug-induced long QT syndrome, torsades de pointes, and sudden cardiac death This article has not been copyedited and formatted. The final version may differ from this version.  (Straus et al., 2005;Michaud and Turgeon, 2013;Giudicessi et al., 2018). Relative to domperidone, trazpiroben very weakly inhibits the hERG channel, reducing the potential for fatal arrhythmias. Owing to its low BBB penetration and weak hERG affinity, trazpiroben is anticipated to have an improved safety profile over metoclopramide and domperidone while preserving the high-affinity D 2 /D 3 receptor antagonism underpinning the effectiveness of these therapies.
Numerous in vitro assays were employed to characterize trazpiroben. Radioligand binding assays demonstrated its high affinity (single-digit nanomolar) at D 2 (both D 2S and D 2L ) and Domperidone has a relatively high degree of plasma protein binding, reportedly between 92 and 93% (Barone, 1999). In vitro studies with trazpiroben showed a similar degree of plasma protein binding, namely 93% and 92% in the plasma of the rat and dog, respectively. For drug candidates, an important determinant of brain penetration is the degree of plasma This article has not been copyedited and formatted. The final version may differ from this version. JPET Fast Forward. Published on July 12, 2021as DOI: 10.1124 at ASPET Journals on August 21, 2021 jpet.aspetjournals.org Downloaded from protein binding, as only the unbound drug fraction can penetrate the BBB (Reichel, 2009).
Additionally, trazpiroben is amphoteric owing to its carboxylic acid functionality. Thus, trazpiroben is expected to demonstrate low lipophilicity, high polarity, and reduced cell permeability due to the overwhelming dominance of the zwitterionic form at physiological pH. High plasma protein binding and the compound's chemical characteristics indicate poor passive permeability, reflected in the minimal brain penetration observed in the rat and dog and lack of CNS effects observed in the rat and humans, and therefore suggesting that trazpiroben may potentially have an improved CNS safety profile versus more centrally penetrating D 2 /D 3 antagonists like metoclopramide, and may be associated with a low risk of EPS, similar to domperidone (Jasper and Whiting, 2020;Whiting et al., 2021).
The CNS safety of trazpiroben was evaluated via accelerating rotarod performance of the rat following oral gavage. Trazpiroben was not associated with meaningful effects on motor coordination in the rat at any dose investigated. In comparison, the centrally acting D 2 /D 3 antagonist chlorpromazine substantially reduced time spent by the rat on the accelerating rotarod following oral administration (Sokoloff et al., 1992). The lack of trazpirobenmediated effect on accelerating rotarod performance, versus effects seen with a centrally acting D 2 /D 3 antagonist, are consistent with the low brain penetration of trazpiroben observed in the present distribution studies in the rat and dog. Additionally, safety pharmacology studies have shown no meaningful effects on functional observation battery assessments and negligible locomotor effects of trazpiroben in the rat (Kreckler et al., 2020).  (Fitzgerald and Dinan, 2008). D 2 receptor activation on lactotroph cells inhibits prolactin gene transcription and synthesis, and therefore, blockade of D 2 receptors counteracts the tonic inhibitory effect of dopamine on prolactin secretion (Freeman et al., 2000;Fitzgerald and Dinan, 2008). Trazpiroben increased prolactin release in the rat following single oral doses Emetic response in the dog and humans is, among other factors, under the control of D 2 receptors in the CTZ of the area postrema (Welliver, 2014). Administration of the D 2 /D 3 agonist apomorphine can be used to stimulate an emetic response in beagle dogs, and inhibition of this response provides an in vivo measure of D 2 /D 3 antagonism in the CTZ (Levant, 1997). Trazpiroben inhibited apomorphine-induced retching/emetic episodes in dog following oral gavage with a potency (0.1-1 mg/kg) similar to that of the positive control, domperidone (Sokoloff et al., 1992). The potency of trazpiroben-mediated inhibition of emetic responses in the dog was similar to that of trazpiroben-mediated increases of prolactin release in rat. Serum prolactin level may therefore be used as a biomarker to indicate target (D 2 receptor) engagement. As both the anterior pituitary gland and the area postrema lie outside the BBB, effects on prolactin secretion in the rat and dog and on apomorphineinduced emesis in dog are therefore suitable models for demonstrating the potent D 2 antagonism of trazpiroben in vivo, despite its limited BBB penetrability.
In good laboratory practice manual patch clamp electrophysiology studies of mammalian cells, trazpiroben only weakly blocked the hERG potassium channel current (IC 50 = 15.6 μM), representing a 10,000-fold selectivity versus affinities (1.5 nM and 3.2 nM) at target D 2 /D 3 receptors. The cardiac potassium channel, hERG, responsible for I Kr in human ventricles, was evaluated because I Kr inhibition is the most common cause of cardiac action potential prolongation by noncardiac drugs (Brown and Rampe, 2000;Rampe and Brown, 2013). Increased action potential duration causes QT interval prolongation and has been associated with the dangerous ventricular arrhythmia torsades de pointes (Brown and Rampe, 2000;Rampe and Brown, 2013). Domperidone, a potent inhibitor of the hERG channel, is associated with a risk of serious cardiac side effects (Michaud and Turgeon, 2013). In our studies, trazpiroben was a weak inhibitor of the hERG channel relative to domperidone, reducing the potential for fatal arrhythmias and potentially offering an improved cardiac safety profile. This is consistent with safety pharmacology studies in instrumented dog and Phase 1 clinical studies that showed no effects on QTc duration (Kreckler et al., 2020;Whiting et al., 2021).
In summary, the in vivo pharmacological profile of the novel, potent dopamine D 2 /D 3 receptor antagonist trazpiroben is similar to that of domperidone. However, unlike domperidone, trazpiroben has a higher therapeutic safety window in terms of the potential for QTc prolongation. Our data suggest trazpiroben may offer potent anti-nauseant and antiemetic activity, and may have a favorable safety profile. Further studies in patients with gastroparesis are ongoing to confirm the safety and efficacy profile of trazpiroben.

Funding and Acknowledgments
This study was sponsored by Altos Therapeutics LLC. Medical writing support was provided by Alexandra Kisbey-Ascott of Oxford PharmaGenesis, Oxford, UK and was funded by Takeda Pharmaceutical Company Ltd.