Abstract
α1-Adrenoceptor (AR) antagonists are widely used for the relief of urinary retention secondary to benign prostatic hyperplasia (BPH). While the five Food and Drug Administration–approved α1-AR antagonists (terazosin, doxazosin, alfuzosin, tamsulosin, and silodosin) share similar efficacy, they differ in tolerability, with reports of ejaculatory dysfunction. The aim of the present work was to revisit their α1-AR subtype selectivity as well as of LDT5 (1-(2-methoxyphenyl)-4-[2-(3,4-dimethoxyphenyl) ethyl]piperazine monohydrochloride), a compound previously described as a multitarget antagonist of α1A-/α1D-AR and 5-HT1A receptors, and to estimate their affinity for D2, D3, and 5-HT1A receptors, which are putatively involved in ejaculatory dysfunction. Competition binding assays were performed with native (D2, 5-HT1A) or transfected (human α1A-, α1B-, α1Dt-AR, and D3) receptors for determination of the drug’s affinities. Tamsulosin and silodosin have the highest affinities for α1A-AR, but only silodosin is clearly a selective α1A-AR antagonist, with Ki ratios of 25.3 and 50.2 for the α1D- and α1B-AR, respectively. Tamsulosin, silodosin, and LDT5 (but not terazosin, doxazosin, and alfuzosin) have high affinity for the 5-HT1A receptor (Ki around 5–10 nM), behaving as antagonists. We conclude that the uroselectivity of tamsulosin is not explained by its too-low selectivity for the α1A- versus α1B-AR, and that its affinity for D2 and D3 receptors is probably too low for explaining the ejaculatory dysfunction reported for this drug. Present data also support the design of “better-than-LDT5” new multitarget lead compounds with pharmacokinetic selectivity based on poor brain penetration and that could prevent hyperplastic cell proliferation and BPH progression.
SIGNIFICANCE STATEMENT The present work revisits the uroselectivity of the five Food and Drug Administration–approved α1 adrenoceptor antagonists for the treatment of benign prostatic hyperplasia (BPH). Contrary to what has been claimed by some, our results indicate that the uroselectivity of tamsulosin is probably not fully explained by its too-weak selectivity for the α1A versus α1B adrenoceptors. We also show that tamsulosin affinity for D3 and 5-HT1A receptors is probably too low for explaining the ejaculatory dysfunction reported for this drug. Based on our lead compound LDT5, present data support the search for a multitarget antagonist of α1A–α1D and 5-HT1A receptors with poor brain penetration as an alternative for BPH treatment.
Introduction
Benign prostatic hyperplasia (BPH) is an age-related disease affecting the quality of life of men mainly due to bladder outlet obstruction, among other bothersome lower urinary tract symptoms (LUTS), such as urgency and nocturia (Berry et al., 1984). α1A-Adrenoceptor (AR) and α1D-AR mRNA have been described in normal and hyperplastic stromal human prostates, and the expression of α1A-ARs is upregulated during BPH (Price et al., 1993; Faure et al., 1994; Nasu et al., 1996; Walden et al., 1999; Roehrborn and Schwinn, 2004; Kojima et al., 2006). The stromal α1A-ARs have been considered important for human prostate contraction (Forray et al., 1994) and, consequently, for the dynamic component of BPH, so that their blockade would explain the observed relief of the micturition difficulties observed with antagonists. On the other hand, cellular proliferation in the periurethral region is related to the static component of BPH and is classically treated at advanced stages of the disease (larger prostates) with the association of α1-AR antagonists (AARAs) and 5-α-reductase inhibitors (Alawamlh et al., 2018). However, other receptors are now being considered putative targets for blocking cellular proliferation, such as the α1D-ARs and 5-HT1A receptors (McVary et al., 2011; Oelke et al., 2013). 5-HT1A receptors are considered as an attractive target for antiproliferative drugs since 5-HT acts as a growth factor on several types of nontumoral and tumoral cells (Fiorino et al., 2014). Earlier works already reported that neuroendocrine cells are present in normal and malignant prostate tissue releasing 5-HT (Abrahamsson et al., 1986), and that prostate cells, including those from BPH patients, express 5-HT1A receptors (Dizeyi et al., 2004). Moreover, these authors showed that prostate cell proliferation was reduced by NAN190, a 5-HT1A receptor antagonist (Dizeyi et al., 2004). Finally, we previously showed that LDT5 inhibited the in vitro growth of prostate cells from BPH patients, induced by 5-HT, similarly to that observed for p-MPPF, a classic 5-HT1A receptor antagonist (Nascimento-Viana et al., 2016). Based on these data, we proposed that a multitarget antagonist toward the α1A-AR, α1D-AR, and 5-HT1A receptor, such as LDT5 (Nascimento-Viana et al., 2016), could be a rational nonhormonal alternative in the search of new drugs for the pharmacotherapy of BPH.
Moderate to severe LUTS associated with BPH are mainly treated with AARAs. The five Food and Drug Administration (FDA)–approved AARAs for BPH treatment have similar efficacies, but they differ in tolerability (Schwinn and Roehrborn, 2008; Michel, 2010; Oelke et al., 2013). The so-called uroselective drugs (tamsulosin, silodosin, and alfuzosin) are better tolerated and have a lower incidence of orthostatic hypotension than the first-generation drugs (terazosin and doxazosin) (Michel, 2010; Hennenberg et al., 2014). As commented by Korstanje et al. (2011), uroselectivity has been classically defined in terms of α1-AR subtype selectivity (pharmacological uroselectivity), preferential reduction of urethral pressure versus blood pressure in animals (functional/physiologic uroselectivity), or desired clinical effects on obstruction and LUTS versus unwanted adverse effects (clinical uroselectivity). As differences exist between the Ki values and selectivities for the five FDA-approved AARA among laboratories (Supplemental Material; Table 1), claims such as tamsulosin’s selectivity for α1A-AR should be carefully checked.
Affinity (Ki values) and selectivity (ratios of Ki values) for binding of α1-AR antagonists to the three human α1-AR subtypes
Ki values are expressed as geometric means of [n] individual experiments. One-way ANOVA and post hoc Holm-Sidak test was performed on pKi values.
Here, we compared these five drugs in the exact same experimental conditions with respect to their affinities for the three human α1-AR subtypes, together with our LDT5 compound. Furthermore, we considered their selectivity not only toward the classic off-target α1B-AR but also toward the D3 and 5-HT1A receptors, putatively responsible for sexual disorders, such as abnormal ejaculation, reported for silodosin and tamsulosin (Giuliano et al., 2006; Wolters and Hellstrom, 2006; Andersson and Abdel-Hamid, 2011; Lepor et al., 2012; La Torre et al., 2016).
Different from silodosin and contrary to what has been claimed by some, our results indicate that the uroselectivity of tamsulosin is probably not fully explained by its too-weak α1A- versus α1B-AR selectivity. We also showed that tamsulosin affinity for D3 and 5-HT1A receptors is probably too low for explaining the ejaculatory dysfunction reported for this drug. Finally, we discuss how multitarget antagonists of α1A–α1D and 5-HT1A receptors, such as LDT5, could be planned for avoiding safety problems at the central nervous system.
Materials and Methods
HEK-293 Cells Transfected with Human α1-AR
Human embryonic kidney (HEK-293; CRL-1573; American Type Culture Collection) was transfected with human α1A-AR, α1B-AR, and α1Dt-AR (Pupo et al., 2003). As the recombinant full-length human α1D-AR is poorly expressed in recombinant systems, a truncated mutant in which the first 79 amino acids were deleted (α1Dt-AR) was used to increase the number of binding sites (Pupo et al., 2003; Nojimoto et al., 2010) The cells were cultured in Dulbecco’s modified Eagle’s medium (GIBCO) containing 25 mM glucose, 44 mM sodium bicarbonate, 10% fetal bovine serum from South America, 1% pyruvate, and 1% penicillin (10,000 U/ml)/streptomycin (10,000 μg/ml; Invitrogen) and incubated (37°C, 5% CO2) until confluence when they were washed with 1 ml of phosphate-buffered saline and scraped to obtain the homogenate. Subsequently, the homogenate was centrifuged at 30,000g for 20 minutes at 4°C, the supernatant discarded, and the pellet resuspended in approximately 10 ml of solution (25 mM HEPES, 150 mM NaCl, 3 mM phenylmethylsulfonyl fluoride, and 1 mM protease inhibitor cocktail, pH 7.4). This material was homogenized with an Ultra-Turrax (IKA Labortechnik) apparatus (twice for 15 seconds at a speed of 9500 rpm). The homogenate was then centrifuged at 30,000g for 20 minutes at 4°C, the supernatant was discarded, and the new pellet was resuspended in buffer containing 25 mM HEPES and 150 mM NaCl, pH 7.4 (Akinaga et al., 2013).
Binding Experiments
Binding to the α1-ARs.
The membrane preparation of transfected HEK-293 cells (150 µg of protein) was incubated for 45 minutes at 30°C in 1 ml of medium containing 0.05 nM [3H]-prazosin, 50 mM Tris-HCl 50 mM (pH 7.4), and 1 mM EDTA (Nascimento-Viana et al., 2016). Nonspecific binding was defined in the presence of 1 µM prazosin. The incubation was terminated by filtration, washing, and treatment of the filters as described previously (Nascimento-Viana et al., 2016).
Binding to the 5-HT1A Receptor.
For binding assays to the 5-HT1A receptors, hippocampi of adult male Wistar rats were homogenized and centrifuged as previously described (Noël et al., 2014). Binding to the low-affinity and high-affinity states of the receptor was performed as detailed previously together with the rationale for estimating the intrinsic efficacy of the ligands by using the Ki ratio (Noël et al., 2014). The protein was incubated at 37°C under yellow light for either 45 minutes with 0.5 nM [3H]-p-MPPF (4-Fluoro-N-(2-[4-(2-methoxyphenyl)1-piperazinyl]ethyl)-N-(2-pyridinyl)benzamide), 50 mM Tris-HCl (pH 7.4), and 1 mM GTP (low-affinity state) or 15 minutes in a solution containing 1 nM [3H]-8-OH-DPAT, 1 mM CaCl2, 1 mM MnCl2, 10 mM pargyline, and Tris-HCl 50 mM (pH 7.4; high-affinity state).
Binding to the D2 and D3 Receptors.
For binding assays to the D2-like receptors, the striatum of adult male Wistar rats was homogenized and centrifuged as previously described (Pompeu et al., 2013; protocol number DFBCICB021, Institutional Ethical Committee for Animal Care from the Federal University of Rio de Janeiro). For binding to the D3 receptor, we used commercially available (Chemiscreen; Millipore) crude membrane preparations of recombinant Chem-1 cells that have been transfected with the cDNA encoding the human D3 receptor (accession number NM_000796). Membranes, compounds, and radioligand (0.1 nM [3H]-YM-09151-2) were incubated at 37°C for 60 minutes under yellow light in a solution containing 120 mM NaCl, 5 mM KCl, 5 mM MgCl2, 1.5 mM CaCl2, 1 mM EDTA, and Tris-HCl 50 mM (pH 7.4) as previously described (Betti et al., 2017).
Statistical Analysis
Data were analyzed by nonlinear regression using GraphPad Prism 6.0 (GraphPad Software) using the classic equations for simple concentration-effect curves (saturation experiments) and competition binding assays to estimate affinity (Kd) of the radioligand and potency [median inhibitory concentrations (IC50)] of the unlabeled competitor ligands, respectively. The affinity of the unlabeled competitor ligands (Ki) was calculated using the IC50 values and the Cheng-Prusoff equation (Cheng and Prusoff, 1973). Ki values were expressed as geometric means with their 95% confidence interval.
Drugs
[3H]-Prazosin (85 Ci/mmol), [3H]-8-OH-DPAT (154.2 Ci/mmol), [3H]-p-MPPF (74.2 Ci/mmol), and [3H]-YM-09151-2 (81.1 Ci/mmol) were purchased from PerkinElmer. Alfuzosin hydrochloride, doxazosin mesylate, tamsulosin hydrochloride, and terazosin hydrochloride were purchased from Sigma-Aldrich (St. Louis, MO), and silodosin was from ShangHai Biochempartner (China). LDT5 hydrochloride ((1-(2-methoxyphenyl)-4-[2-(3,4-dimethoxyphenyl) ethyl]piperazine monohydrochloride) was synthesized as previously described for other N-phenylpiperazine derivatives (Romeiro et al., 2011). Stock solutions (1 and 10 mM) were made in sterile deionized water (LDT5) or 100% DMSO (Sigma-Aldrich) and then diluted in water. At the final concentration used (no more than 0.1%), DMSO had no effect in our assays.
Results
Determination of Kd for [3H]-Prazosin Binding to the Three Human α1-ARs.
We first characterized the binding of [3H]-prazosin to the three subtypes of human α1-AR by performing saturation experiments at equilibrium to determine the Kd values in our experimental conditions. The Kd values were 0.475, 0.354, and 0.577 nM for the α1A-, α1B-, and α1Dt-ARs, respectively (geometric means, n = 2), and were similar to values described elsewhere with these cells (Nojimoto et al., 2010).
Determination of Ki Values and Selectivity of Test Compounds for Binding at the Human α1-ARs.
As illustrated in Fig. 1, we performed full competition curves for the three human α1-AR subtypes with our six compounds using the antagonist [3H]-prazosin as the radioligand. Note that the potency sequence is somewhat different for these three subtypes as exemplified by silodosin, one of the most potent for inhibiting [3H]-prazosin binding to the α1A-AR (Fig. 1A) but less potent for the α1B-AR (Fig. 1B) and α1D-AR (Fig. 1C).
Effect of α1-AR antagonists on the binding of [3H]-prazosin to human α1A-AR (A), α1B-AR (B), and α1D-AR (C). The membrane preparation of transfected HEK-293 cells (150 µg of protein) was incubated for 45 minutes at 30°C in 1 ml of medium containing 0.05 nM [3H]-prazosin, Tris-HCl 50 mM (pH 7.4), and 1 mM EDTA in the presence or absence of increasing concentrations of the tested compounds. The data represent the mean ± S.E.M. of three to four independent experiments performed in triplicate.
Tamsulosin and silodosin have the highest affinities for the α1A-AR but differ mainly with respect to their selectivity profile: whereas tamsulosin affinity for α1A-AR is only slightly higher than for α1D- and α1B-AR, with Ki ratios of 2.92 and 5.1, respectively, silodosin is clearly an α1A-AR selective ligand with Ki ratios of 25.3 and 50.2 for the α1D- and α1B-AR, respectively (Supplemental Fig. 1; Table 1).
With respect to the third uroselective drug, alfuzosin has an affinity 2–4 times lower for the α1A-AR than for the other two subtypes, as also observed with terazosin. Alfuzosin and terazosin showed similar affinities for α1A-AR and similar selectivity profiles (Supplemental Fig. 1; Table 1).
Doxazosin has a similar affinity for the three subtypes. Our compound LDT5 has the same affinity for the α1A- and α1D-ARs, being around 2 to 3 times higher than that for α1B-AR, showing a selectivity profile similar to tamsulosin (Supplemental Fig. 1; Table 1).
Determination of Affinity and Intrinsic Efficacy of Test Compounds at the 5-HT1A Receptor.
To determine the affinity of the six compounds to the 5-HT1A receptor, we used a binding assay with the antagonist radioligand [3H]-p-MPPF and rat hippocampal membranes, as previously described (Noël et al., 2014).
Table 2 shows that LDT5, tamsulosin, and silodosin have a high affinity for this receptor, with Ki values around 5–10 nM, whereas alfuzosin, terazosin, and doxazosin have a much lower affinity, with Ki values higher than 1 µM. Tamsulosin had a Ki value close to the one reported previously by others (4.4 nM; Leonardi et al., 1997). Considering the affinity for the main target receptor of BPH involved in contraction (α1A-AR) as a reference, Table 2 indicates that LDT5 has the same affinity for the 5-HT1A (Ki ratio equal to 1.56) and that silodosin affinity for 5-HT1A is relevant (Ki ratio around 10). On the contrary, albeit tamsulosin binds to 5-HT1A at nanomolar concentrations, its affinity is about 33 times lower than for the α1A-AR. With Ki ratios much higher than 100, alfuzosin, terazosin, and doxazosin are to be considered highly selective α1A-AR ligands toward the 5-HT1A receptor.
Affinity (Ki values) for binding to the 5-HT1A receptor (low-affinity state) and selectivity for binding to α1A-AR versus 5-HT1A receptor
Ki values are expressed as geometric means of [n] individual values calculated from competition curves using the antagonist radioligand (low-affinity state, see Materials and Methods). The ratio of these Ki values and the Ki values for α1A-AR is a measure of selectivity. Intrinsic activity at the 5-HT1A receptor was estimated by the ratio of Ki values for the low- and high-affinity state of the receptor (Noël et al., 2014).
Because not only affinity but also intrinsic efficacy is important for pharmacological effect, we then used a previously validated functional binding assay (Noël et al., 2014) for the three compounds with relevant affinity for the 5-HT1A receptor. As described in Fig. 2 for silodosin and tamsulosin, competition curves were performed using either the antagonist radioligand [3H]-p-MPPF in the presence of GTP (low-affinity state of the receptor) or the agonist radioligand [3H]-8-OH-DPAT in the presence of divalent cations that favor the high-affinity state of the receptor. In such an assay, the intrinsic efficacy of a compound is estimated by its ratio of Ki values measured when the receptor is in the low- to high-affinity state. With Ki ratios not different from 1, LDT5 and silodosin are to be considered as neutral antagonists, whereas tamsulosin harbored a Ki ratio of 3.9, significantly different from 1 (Table 2). As this ratio is much smaller than the one reported for the full agonist 5-HT [76.8; see Noël et al. (2014)], tamsulosin is to be considered as a weak partial agonist of this receptor.
Specific binding of [3H]-8-OH-DPAT (squares) and [3H]-p-MPPF (circles) in rat hippocampal membranes in the presence of increasing concentrations of silodosin and tamsulosin. The protein was incubated at 37°C for either 45 minutes with 0.5 nM [3H]-p-MPPF, Tris-HCl 50 mM (pH 7.4), and 1 mM GTP (low-affinity state) or 15 minutes in a solution containing 1 nM [3H]-8-OH-DPAT, 1 mM CaCl2, 1 mM MnCl2, 10 mM pargyline, and Tris-HCl 50 mM (pH 7.4; high-affinity state). Data are expressed as means ± S.E.M. of three to five individual experiments performed in triplicate.
Determination of Affinity of Tamsulosin and LDT5 at D2 and D3 Receptors.
Due to the putative role of D2 and, mainly, D3 receptors as off targets for drugs used in BPH therapy, we determined the affinity of tamsulosin and LDT5 for human D3 receptors and D2-like receptors present in rat striatal preparations [mainly D2 receptors, according to Booze and Wallace (1995)]. Table 3 shows that both compounds have a higher affinity for the D3 than for the D2 receptors. The Ki ratios (D3 vs. α1A-AR) are around 44 and 8 for tamsulosin and LDT5, respectively.
Affinity of tamsulosin and LDT5 for human D3 and rat striatum
D2–like receptors and selectivity for binding to the α1A-AR versus D3 receptor Ki values are expressed as geometric means of [n] individual experiments.
Discussion
Pharmacological Selectivity of the Five FDA-Approved α1A-AR Antagonists for BPH Treatment.
In vitro off-target receptor binding is a well established method of derisking used in drug discovery programs (Bowes et al., 2012) and is also the basis for defining whether the so-called uroselectivity of some α1-AR antagonists used for BPH treatment is due to pharmacological selectivity or to other reasons (see Introduction). Since α1B-ARs are not involved in the pathophysiology of BPH and are expressed in several tissues, including blood vessels, they have been considered for a long time as off target for α1-AR antagonists used in BPH due to the idea that their blockade was responsible for cardiovascular adverse effects, mainly postural hypotension (Michel, 2010). This idea has now been challenged due to the controversial role of the α1B-AR in controlling blood pressure and the good cardiovascular tolerability of alfuzosin, a nonselective antagonist (Michel, 2010; Akinaga et al., 2019). On the other hand, the α1A-AR is considered the main target for BPH treatment due to the prominence of this subtype in the human prostate and its role in prostate contraction (Forray et al., 1994). As a result of the classic view on BPH, the Ki ratio (α1B-AR/α1A-AR) was used for quantifying pharmacological selectivity of such drugs. As indicated in Table 4, our results support the claimed α1A selectivity of silodosin, since the referred Ki ratio is about 50. On the other hand, our data do not support the usual claim that tamsulosin is an α1A-AR (or α1A/1D-AR) selective antagonist. Indeed, our Ki ratio was 5.1, similar to the low values already reported in two other works (Table 4). Note that even the higher Ki ratios reported by two other groups (around 12) are not sufficient to support the claim, as also criticized by Lepor et al. (2012), who considered that no clinical advantage could be attributed to a receptor selectivity of only about 10 times. Based on a 94%–99% binding to plasma proteins (Flomax CR product monograph, tamsulosin, BOEHRINGER INGELHEIM), we can estimate that the free maximal plasma concentration of tamsulosin at steady state after daily administration of a controlled-release tablet containing 0.4 mg of tamsulosin hydrochloride is below or around the Ki we measured for tamsulosin binding to the α1B-AR. As a consequence, a Ki ratio higher than 5 (α1B vs. α1A) would have a clinical relevance since the active (free) plasma concentration would be in the “selective” range. As an alternative, the explanation elegantly proposed by Sato et al. (2012) sounds plausible. These authors reported that the residence time of tamsulosin at the α1A-AR was much higher than that at the α1B-AR subtype, contrary to what occurred with prazosin. Note that a pharmacodynamics selectivity is expected for drugs with a higher residence time at the target than at the off target (Copeland et al., 2006). As an alternative hypothesis to explain tamsulosin’s reported uroselectivity, Korstanje et al. (2011) concluded that tamsulosin would exhibit a greater uptake into human prostate than would be expected from plasma concentrations based on differences in unbound drug fraction in human prostate (59%) and plasma (0.4%). Based on these data, the area under the curve (0, 24 hours) of unbound tamsulosin in prostate tissue was estimated to be 63-fold higher than the area under the curve (0, 24 hours) in plasma. As it is assumed that, under equilibrium conditions, diffusion of unbound drug will lead to equal drug concentrations in these two compartments, we cannot discard an experimental artifact since the unbound concentrations were not measured directly through in situ microdialysis, the gold standard approach for such experiments. Our data also confirm that the uroselectivity of alfuzosin is not due to a pharmacological selectivity between the α1-AR subtypes, since its affinity for the α1A-AR is even slightly lower (higher Ki ratios) than for the two other subtypes, as also observed for the two “old” nonselective α1-AR antagonists terazosin and doxazosin (Table 4).
Selectivity (ratios of Ki values) for binding to α1-AR subtypes: between target subtypes and between the main target (α1A) and off-target (α1B) subtypes
Comparison between present data and data from the literature.
In addition to the α1B-AR, we also considered the affinity of the five FDA-approved drugs toward the 5-HT1A and D2/D3 receptors, which are poorly discussed in the literature for these drugs. Our interest was based on the proposal that these receptors participate in the central control of ejaculation and could be involved in the ejaculation disorders observed clinically in BPH patients, particularly those treated with silodosin or tamsulosin (Giuliano et al., 2006; Wolters and Hellstrom, 2006; Lepor et al., 2012; La Torre et al., 2016). For silodosin, although harboring some relevant affinity for the 5-HT1A receptor, the central effect could probably be discarded due to its apparently poor brain penetration (Okura et al., 2002). For tamsulosin, the situation is less clear since some penetration into the brain has been reported, albeit without quantitative data (Giuliano et al., 2006), whereas a low potential to cross the blood-brain barrier has also been reported by others ( Andersson and Abdel-Hamid, 2011). Our data do not support the participation of the 5-HT1A, D2, and D3 receptors in the ejaculation disorders due to a relatively low affinity of tamsulosin for these receptors. Note that the Ki value for the D3 receptor reported by Kuo et al. (2000) was much lower than ours (0.28 vs. 15.7 nM) and that we did not find any apparent explication for such difference nor other data in the literature.
LDT5 and Insight for Putative New Multitarget Lead Compounds.
The present data extend our previous data indicating that LDT5 could be considered a multitarget drug for the α1A/D-AR and 5-HT1A receptors (Nascimento-Viana et al., 2016). Previous estimates of affinity (KB) for the α1A- and α1D-AR were based on the antagonism of phenylephrine-induced isometric contractions of rat prostate and aorta, respectively, whereas affinity for the α1B-AR was assessed by competition for [3H]-prazosin binding to rat liver synaptosomes (Ki). Present affinity estimates were all obtained in binding experiments with membranes of cells transfected with each of the three human α1-AR subtypes, a priori, a more suitable assay for a translational point of view. Although the affinity for the α1A- and α1D -ARs are lower (8–14 times), the ratio of Ki values for these two receptors is similar, confirming that LDT5 is a high-affinity α1A/D-AR ligand. Based on the present data, the selectivity toward the off-target α1B-AR should be lower than previously estimated (2.9 vs. 55 times). However, in vivo LDT5 showed an ED50 of 0.09 µg⋅kg−1 for the reduction of intraurethral pressure, and a similar dose (0.1 µg⋅kg−1) did not cause any hypotensive effect (Nascimento-Viana et al., 2016), which could suggest a potential uroselective profile in rats.
The present data give support for designing “better-than-LDT5” new multitarget (α1A/D-AR and 5-HT1A receptor) lead compounds. Indeed, as blockade of brain 5-HT1A receptors could result in on-target adverse effects (see Pharmacological Selectivity of the Five FDA-Approved α1A-AR Antagonists for BPH Treatment), a pharmacokinetic selectivity based on poor brain penetration would be a strategy for such compounds, e.g., by designing a drug that would be a substrate of P-glycoprotein. This concern is also strengthened by the relatively high affinity we reported here for LDT5 binding to the D3 receptor. Considering these data, LDT5 is no more considered as the ideal lead compound since the permeability assay with MDCK-MDR1 showed that it is not a substrate of P-glycoprotein (Noël et al., 2016). However, we suggest that the rationale of such a multitarget drug for BPH treatment is maintained mainly based on our previous data with cells from BPH patients, since LDT5 inhibited prostate hyperplastic cell proliferation and reduced intraurethral pressure without hypotensive effects (Nascimento-Viana et al., 2016).
Authorship Contributions
Participated in research design: Quaresma, Silva, Noël.
Conducted experiments: Quaresma, Pimenta, Santos da Silva.
Contributed new reagents or analytic tools: Romeiro, Pupo.
Performed data analysis: Quaresma, Pimenta, Santos da Silva.
Wrote or contributed to the writing of the manuscript: Quaresma, Pupo, Romeiro, Silva, Noël.
Footnotes
- Received May 21, 2019.
- Accepted July 2, 2019.
This study was supported by Fundação de Amparo à Pesquisa do Estado do Rio de Janeiro and by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Brazil. B.M.C.S.Q. was recipient of a CAPES Ph.D. fellowship from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (Capes, Brazil). A.R.P. and A.C.S.d.S. thank CNPq for the scholarships. A.R.P., C.L.M.S., and F.N. are supported by CNPq.
↵
This article has supplemental material available at jpet.aspetjournals.org.
Abbreviations
- AARA
- α1-AR antagonist
- AR
- α1-adrenoceptor
- BPH
- benign prostatic hyperplasia
- FDA
- Food and Drug Administration
- LUTS
- lower urinary tract symptoms
- p-MPPF
- 4-Fluoro-N-(2-[4-(2-methoxyphenyl)1-piperazinyl]ethyl)-N-(2-pyridinyl)benzamide
- Copyright © 2019 by The American Society for Pharmacology and Experimental Therapeutics
