Synthesis, Pharmacophore Modeling, and Biological Evaluation of Novel 5H-Thiazolo[3,2-a]pyrimidin-5-one Derivatives as 5-HT2A Receptor Antagonists

Novel 5H-thiazolo[3,2-a]pyrimidin-5-one derivatives linked through an ethylene bridge to various phenylpiperazine groups were prepared for evaluation as 5-HT2A receptor antagonists. The target compounds 11a–p were prepared through the initial synthesis of the 2-chloroethyl intermediates 10a–d which were then reacted with the appropriate phenylpiperazines. All compounds were tested for their antagonistic activity on 5-HT2Areceptors using inhibition of 5-hydroxytryptophan(5-HTP)-induced head twitches in mice. Pharmacophore modeling study, based on a hypothetical pharmacophore template generated from a set of diverse known active ligands, revealed good fitting of the designed compounds to the generated hypothetical pharmacophore.


Introduction
Serotonin (5-hydroxytryptamine, 5-HT) has been recognized for more than 50 years as an effector on various smooth muscles and, subsequently, as an important neurotransmitter in the brain [1].The serotonin receptors represent a diverse family of receptors that has been initially classified pharmacologically into multiple subtypes [2].The advent of molecular cloning provided additional discrimination within the serotonin receptor family revealing the existence of seven subfamilies containing a total of fourteen distinct receptors based on primary sequence, pharmacology and signal transduction pathways [1].Serotonin receptor dysfunction has been implicated in many neuropsychiatric disorders including anxiety, depression and schizophrenia [1].Clinical use of therapeutics affecting serotonergic function is exemplified by selective serotonin reuptake inhibitors for depression [2], 5-HT 1A partial agonists for anxiety and depression, 5-HT 1D agonists for acute migraine [3], 5-HT 2A and 5-HT 2C antagonists for migraine prophylaxis, depression, schizophrenia [4,5], 5-HT 3 antagonists for chemotherapy-induced emesis [6], and 5-HT 4 agonists for gastrointestinal disorders [7].
Inspired by this knowledge, the newly synthesized compounds were designed as structural mimics of ketanserin 1 and its analogs 2 and 3.The new compounds possessed an additional phenyl group on the thiazolopyrimidine nucleus, hoping to enhance the 5-HT 2A antagonistic activity.Moreover, the piperidine ring in the lead compounds was replaced by the isosteric piperazine ring with the aromatic substitution directly attached to it as in risperidone.The two carbon spacer (the ethylene bridge) was retained to keep the distance between the heterocyclic nucleus and the protonatable nitrogen (N 1 of the piperazine ring) as supported by the modeling study.

Chemistry
The target compounds 11a-p were prepared according to Scheme 1.The 2aminothiazole derivatives 9a-d were obtained through a cyclocondensation reaction of the corresponding α-bromoacetophenones 8a-d [14] to thiourea according the reported procedures [15,16].Reaction of the appropriate 2aminothiazole 9a-d, and 2-acetylbutyrolactone in phosphorous oxychloride furnished the 2-chloroethyl intermediates 10a-d.This was followed by alkylation of various phenylpiperazines with these intermediates in dry dimethylformamide in the presence of triethylamine under nitrogen atmosphere to give the final compounds Sch. 1.

Pharmacology
The new compounds were evaluated, in vivo, for their 5-HT 2A antagonistic activity by measuring their ability to inhibit the 5-HTP induced head twitches in mice [17,18].Risperidone was chosen as the standard drug.Risperidone was preferred over ritanserin as the standard drug, although the latter is more closely related to the tested compounds, because the former, unlike ritanserin, is widely marketed for clinical treatment of psychosis.The observed data were summarized in Tab. 1 which showed the response percent calculated as the percent of mice showing no head twitches to the total number of mice in each group.ED 50 for all compounds are shown in Tab. 2. The doses were calculated in mmol/kg and were administered intraperitoneally (ip).From the obtained data (Tabs.1,2 and Fig. 7), all compounds were found to exhibit 5-HT 2A receptor antagonistic activity with no significant difference from the activity of risperidone, except for compounds 11n, 11o, and 11p.Compounds 11i and 11k possess an activity profile identical to that of risperidone.The impact of the phenylpiperazine substituent (R') viz.2-OCH 3 , 4-Cl, 3-CF 3 , and H, upon activity is not absolute.However, the substituent (R) on the phenyl ring at position 3 of the heterocyclic nucleus seems to play role in switching the activity; so that compounds with R=CH

Pharmacophore modeling
A molecular modeling experiment was carried out to develop a hypothetical pharmacophore model for the 5-HT 2A receptor aiming to study the fitting of the designed compounds to this pharmacophore.A ligand-based pharmacophore model was developed using a training set of ten 5-HT 2A antagonists of diverse chemical structures including compounds 1-6 (Chart 1), in addition to, setoperone, chlorpromazine, cyproheptadine, and tefludazine [11].Fig. 1 shows the alignment of the ten selected structures.The generated hypothetical pharmacophore (Fig. 2) shows seven overlapping points with similar chemical properties in the training set, F1: a hydrogen bond acceptor center; F2: an aromatic or hydrophobic center with two parallel places F3 and F5 for Pi orbital accommodation; F4: an aromatic center with a H-bond donor group; F6: a hydrophobic center; and F7: a second H-bond acceptor center.
Fitting of the designed compounds 11a-d to the pharmacophore revealed the presence of the appropriate chemical groups superimposed on the pharmacophoric elements (Figs.[3][4][5][6].The phenylpiperazine tail represents the aromatic ring with its Pi orbitals properly oriented to fit to F2, F3 and F5.Also, N 1 of the piperazine ring is a H-bond acceptor corresponding to F1.The ethylene spacer is the hydrophobic group fitting to F6; it is worth mentioned that this 2 carbon spacer was perfect in adjusting the distance between other groups in the molecule.The thiazolopyrimidine ring is a suitable aromatic group for F4 with the nitrogen atom, at position 8, as a H-bond acceptor group for F7.Finally, it is interesting to note that, according to the created model, the phenyl group at position 3 is not part of the pharmacophore.Therefore, it will not be required to study the fitting of the rest of the compounds 11e-p since variations in the substituents on the 3-phenyl group will not affect the fitting results.
In conclusion, it could be assumed that the presence of the phenyl ring on the heterocyclic nucleus, although seeming not essential for activity as suggested by the molecular modeling study, yet the activity of the new compounds is in great part dependent on it.The role of the 3-phenyl group is unclear but it can be assumed that it fits into an accessory site of the receptor.Pharmacological data, supported by the modeling study, showed that the phenylpiperazine moiety is a suitable bioisostere of the arylpiperidine group present in ketanserin 1 and its analogs.In addition, the ethylene bridge is the proper spacer that keeps other groups in the molecule in the perfect position.given in ppm.Mass spectra were performed on Fennigan MAT, SSQ 7000 mass spectrophotometer at 70eV.IR spectra were recorded on Bruker FT-IR spectrophotometer as potassium bromide disc.Compounds 8a-d [14], and 9a [15], and 9b-d [16] were prepared according to the published procedures.

Pharmacological screening
Inhibition of 5-HTP induced head twitches in mice.
The antagonism of 5-HT 2A receptors was determined in vivo by inhibition of 5-HTP induced head twitches in mice [18].The animals were divided into seventeen groups each of six male mice (18-22g) to test compounds 11a-p and risperidone as the standard drug (Tab.1).The animals received 75mg/kg pargyline HCl as a solution in water intraperitoneally (ip).Thirty minutes later, they were treated with the test drug ip in a dose of 24.4 mmol/kg as a suspension in saline and Tween 80. Thirty minutes after the test drug, the animals were injected ip with 10mg/kg DL-5-HTP as a suspension in saline and Tween 80.An additional group of six mice was used as a control group which was injected with water-Tween 80 as a vehicle.An animal was considered to give positive response if it does not show head twitches 20 min.after 5-HTP injection.According to the observed response%, the experiment was repeated similarly using a lower and a higher dose of the drug

1 .
Experimental structure activity data for 5-HT 2A antagonists suggest that two planar aromatic or heterocyclic ring systems separated by an aliphatic or alicyclic chain, containing basic protonatable nitrogen, seem to constitute a potent 5-HT 2A ligand.Additional hydrophobic substituent in the heterocyclic ring, as demonstrated by the 4-fluorophenyl group in sertindole 6 and the chlorophenyl ring in clozapine 5,