The computer-aided discovery of novel family of the 5-HT6 serotonin receptor ligands among derivatives of 4-benzyl-1,3,5-triazine

Article abstract of DOI:10.1016/j.ejmech.2017.04.033

The work describes a discovery of new chemical family of potent ligands for the 5-HT₆ serotonin receptors. During the search for new histamine H₄ receptor antagonists among 1,3,5-triazine derivatives, compound 2 (4-benzyl-6-(4-methylpiperazin-1-yl)-1,3,5-triazin-2-amine) was found. Compound 2, weakly active for the H₄ receptor but fitted in 3/4 of pharmacophore features of the 5-HT₆R ligand, occurred to be a moderate 5-HT₆R agent, useful as a lead structure for further modifications. A series of new derivatives (3–19) of the lead 2 was synthesized, evaluated in the radioligand binding assay (RBA) and explored in comprehensive molecular modelling, including both pharmacophore- and structure-based approaches with docking to the homology model of 5-HT₆R. The most active compounds displayed a potent affinity for the 5-HT₆R in the nanomolar range (K_i?=?20–30?nM), some of them (4, 11 and 19) were tested in the rat forced swim test that revealed their antidepressant-like effect. SAR-analysis on the basis of both, RBA and docking results, indicated that action on the receptor is related to the hydrophobicity and the size of aromatic moiety substituted by a methylene linker at the position 4 of 1,3,5-triazine.

Full text of DOI:10.1016/j.ejmech.2017.04.033

Accepted Manuscript
The computer-aided discovery of novel family of the 5-HT serotonin receptor ligands
6
among derivatives of 4-benzyl-1,3,5-triazine
Dorota Łażewska, Rafał Kurczab, Małgorzata Więcek, Katarzyna Kamińska, Grzegorz
Satała, Magdalena Jastrzębska-Więsek, Anna Partyka, Andrzej J. Bojarski, Anna
Wesołowska, Katarzyna Kieć-Kononowicz, Jadwiga Handzlik
PII:
S0223-5234(17)30290-8
10.1016/j.ejmech.2017.04.033
EJMECH 9380
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 21 January 2017
Revised Date: 16 March 2017
Accepted Date: 12 April 2017
Please cite this article as: D. Łażewska, Rafał. Kurczab, Mał. Więcek, K. Kamińska, G. Satała, M.
Jastrzębska-Więsek, A. Partyka, A.J. Bojarski, A. Wesołowska, K. Kieć-Kononowicz, J. Handzlik,
The computer-aided discovery of novel family of the 5-HT serotonin receptor ligands among
6
derivatives of 4-benzyl-1,3,5-triazine, European Journal of Medicinal Chemistry (2017), doi: 10.1016/
j.ejmech.2017.04.033.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
Graphical abstract

ACCEPTED MANUSCRIPT
The computer-aided discovery of novel family of the 5-HT serotonin receptor ligands
6
among derivatives of 4-benzyl-1,3,5-triazine
a
b
a
a
Dorota Łażewska *†, Rafał Kurczab †, Małgorzata Więcek , Katarzyna Kamińska , Grzegorz
b
c
c
b
Satała , Magdalena Jastrzębska-Więsek , Anna Partyka , Andrzej J. Bojarski , Anna
c
a
a
Wesołowska , Katarzyna Kieć-Kononowicz , Jadwiga Handzlik *
a
Department of Technology and Biotechnology of Drugs, Jagiellonian University,
Medical College, Medyczna 9, PL 30-688 Kraków, Poland
b
Department of Medicinal Chemistry Institute of Pharmacology, Polish Academy of
Sciences, Smętna 12, PL 31-343 Kraków, Poland
c
Department of Clinical Pharmacy, Jagiellonian University, Medical College,
Medyczna 9, PL 30-688 Kraków, Poland
*
Corresponding authors:
D. Łażewska) E-mail address: dlazewska@cm-uj.krakow.pl and (J. Handzlik) E-mail
address: j.handzlik@uj.edu.pl
These authors contributed equally to this work.
(
†
1

ACCEPTED MANUSCRIPT
Abstract
The work describes a discovery of new chemical family of potent ligands for the 5-HT6
serotonin receptors. During the search for new histamine H receptor antagonists among
4
1
,3,5-triazine derivatives, compound 2 (4-benzyl-6-(4-methylpiperazin-1-yl)-1,3,5-triazin-2-
amine) was found. Compound 2, weakly active for the H receptor but fitted in 3/4 of
4
pharmacophore features of the 5-HT R ligand, occurred to be a moderate 5-HT R agent,
6
6
useful as a lead structure for further modifications. A series of new derivatives (3-19) of the
lead 2 was synthesized, evaluated in the radioligand binding assay (RBA) and explored in
comprehensive molecular modelling, including both pharmacophore- and structure-based
approaches with docking to the homology model of 5-HT R. The most active compounds
6
displayed a potent affinity for the 5-HT R in the nanomolar range (K =20-30 nM),some of
6
i
them (4, 11 and 19) were tested in the rat forced swim test that revealed their antidepressant-
like effect . SAR-analysis on the basis of both, RBA and docking results, indicated that action
on the receptor is related to the hydrophobicity and the size of aromatic moiety substituted by
a methylene linker at the position 4 of 1,3,5-triazine.
Keywords:
serotonin receptors; 5-HT ligands; 1,3,5-triazine; docking; CNS drugability
6
2

ACCEPTED MANUSCRIPT
1
. Introduction
The 5-HT receptor is the most recently identified member of the serotonin (5-HT)
6
receptor superfamily. The human 5-HT receptor (5-HT R) was discovered by Kohen et al in
6
6
1
996 [1], three years later after the first isolation of this receptor from rat striatum by two
scientific groups independently [2,3]. The 5-HT R is distributed in the central nervous system
6
(CNS), especially, is located in brain areas involved in learning and memory processes.
Intensive preclinical studies have shown that 5-HT R antagonists could be a promising drug
6
with cognitive improvement in psychiatric (e.g. schizophrenia, depression) or
neurodegenerative diseases (e.g. Alzheimer’s disease), and for obesity treatment [4]. Several
studies have shown that not only antagonists but also agonists have potency for the treatment
of obesity or cognitive disfunctions [5]. Since 1999 when the first selective 5-HT R
6
antagonists were described, various compounds have been synthesized as potential agents for
this protein target. Some of them were successful in the primary pharmacological screening or
even have reached clinical studies, e.g. dimebon (discontinued after phase III) or
LUAE58054 (idalopirdine), where their utility has been or still is verified [6]. However till
now, no 5-HT ligand has been accepted as a CNS-drug that could reach pharmaceutical
6
market. Thus, the further search for new chemical families of the 5-HT R ligands, including
6
compounds with high affinity and good CNS-drugability properties, are a challenge for
medicinal chemistry.
More than ten years ago, López-Rodríguez suggested a pharmacophore model for the 5-
HT R antagonists. This model includes a triangle topology with tops of a bulky hydrophobic
6
area (HYD), a positive ionizable nitrogen (PI) and a hydrogen bond acceptor (HBA) as well
as a central aromatic fragment (AR, Fig. 1) [7].
Figure 1
For the most potent 5-HT R ligands, the pharmacophore feature HYD is frequently
6
represented by fused aromatic rings of a naphthalene or a halogen substituted benzene, an
indole or a benzothiophene. A piperazine (un)substituted with methyl seems to be a most
profitable moiety containing the feature PI. The feature HBA, occurring in a number of
sulfonyl ligands, can also fit in the carbonyl moieties whereas the aromatic heterocyclic
groups, e.g. a pyridine, a benzimidazole or a quinolone, form a benefit aromatic feature (AR,
Fig. 1) that is desirable for interactions with the 5-HT R [8].
6
3

ACCEPTED MANUSCRIPT
On the other hand, our previous studies provided a series of 2-amino-4-(4-
methylpiperazin-1-yl)-1,3,5-triazine derivatives [9–11] that had been designed as potential
histamine H receptor ligands. That aim was achieved in the case of derivatives with aryl
4
rings directly substituted at the triazine, e.g. TR7 (1) [9], whereas the compounds with an
aromatic moiety separated by the methylene group, e.g. TR20 (2) [11], displayed a weak
action on the histamine H receptor with K values only in micromolar range (Fig. 2).
4
i
Figure 2
During an analysis of structural properties of the 2,4,6-trisubstituted 1,3,5-triazines (1 and
2
2
2
) some slight similarity to known selective 5-HT R ligands, namely Ro 04-6790 [12], SB-
6
71046 [13] or SB-399885 [14] (Fig. 3) can be noted as well as the triazine compounds (Fig.
) imposed on 3 out of 4 features of the 5-HT R pharmacophore model of López-Rodríguez.
6
Figure 3
Furthermore, precise search for structural similarities to the triazine compounds (1 and
2
) among compounds deposited in ChEMBL v20 database confirmed that no
triazine 5-HT R ligand had been identified so far. The most similar 5-HT R ligands contained
6
6
the benzimidazole or the naphthalene core (see Supplementary Table 1S). Taking this into
account, we decided to perform modelling studies and evaluate the group of 1,3,5-triazine
compounds for their binding ability to this important protein target.
The preliminary in vitro studies have shown a very weak micromolar affinity for 5-HT R
6
in the case of a series of 19 aryl-triazine compounds with the aromatic ring directly linked to
the triazine core (Modification A, Fig. 4). In contrary, a good 5-HT R affinity (K = 96 nM)
6
i
was observed for the methylene-spacer compound 2 (Modification B, Fig. 4).
Figure 4
Hence, compound 2 has been selected as a lead structure for further modifications to give a
new chemical group of 5-HT R ligands with therapeutic perspectives. Here, we present the
6
computer-aided design, synthesis and pharmacological evaluation (in vitro and in vivo) as
4

ACCEPTED MANUSCRIPT
well as structure-activity relationship (SAR) analysis supported by molecular modelling for
the new generation of triazine 5-HT R agents, including lead 2 and its 17 derivatives (2-19,
6
Table 1).
2
. Results and discussion
2
.1. Chemical synthesis
Compounds 2-19 (Table 1) have been obtained according to the synthesis route shown in
Scheme 1, on the basis of methods described previously [9,10].
Table 1
In the first step, commercial 1-methylpiperazine dihydrochloride 20 and 1-cyanoguanidine
2
1 were heated under gradually increasing temperature in butanol to give the 4-
methylpiperazin-1-yl biguanide dihydrochloride 22 (Scheme 1a). Carboxylic acid esters 3e-5e
and 7e-15e were prepared involving two different methods (Scheme 1a).
Scheme 1
Methyl esters 3e and 7e-15e were obtained by acid-catalyzed esterification of
corresponding commercial phenylacetic acids (3a, 7a-15a). In the case of esters 4e and 5e, the
reaction of suitable carboxylic acids (4a and 5a) with iodomethane in the presence of DBU
(1,8-diazabicyclo[4.5.0]undec-7-ene) was carried out [15]. In the last step, suitable esters,
including commercially available ones (2e, 6e, 16e-19e) and those synthesized by us (3e-5e
and 7e-15e) were used for cyclic condensations in basic conditions [9,10] to give final
products 2-19 (Scheme 1b).
2
2
.2. Molecular Modelling
.2.1. Docking studies
Since 3-dimensional pharmacophore model for 5-HT R ligands was developed and
6
successfully used in several studies [7,16], it was reconstructed using the same dataset and
parameters as in the original paper [7]. The lead 2 and compounds 3-19 matched partially on
the pharmacophore model (Fig. 5A) with a missed HBA feature. It should be noted that
compounds with no HBA element were studied in order to assess their influence on the ligand
binding [16] showing that the HBA is not essential, however, the replacement of the sulfonyl
5

ACCEPTED MANUSCRIPT
3
or sulfonamide moiety by sp methylene, in general, caused a slight decrease of the antagonist
potency [8].
Figure 5
To study the molecular mechanism of action for synthesized library of compounds with the
5
-HT R, a homology model, built on the β adrenergic receptor crystal structure [17–19] as a
6 2
template, was used (see Experimental Section). In this model, the protonated piperazine
moiety (PI group) of the ligand interacts with Asp3.32, the AR feature, mapped on the 1,3,5-
triazine ring forms CH-π interaction with Phe6.52 and partially with Phe6.51. Additionally,
the NH group of 1,3,5-triazine can form hydrogen bonds with carbonyl oxygen of Ala5.42
2
and Val3.33 (Fig. 5B). The HYD feature expands into a hydrophobic cavity formed by
transmembrane domains 3, 4, 5 and extracellular loop 2. The hydrophobic nature of the
residues forming this crevice tolerates a different substitution at this aromatic site (-Cl, -CH3).
It should be emphasized that the 1,3,5-triazine rings and HYD features form an angle of
approximately 90°, what is resulted of the tetrahedral conformation of the methylene linker.
In order to get a better understanding of the changes in the activity of a lead structure by
application of different substituents, a binding free energy (∆G) using MM/GBSA algorithm
[
20] implemented in Schrödinger Suit was calculated and used to obtain difference (∆∆G)
between binding free energy of the lead and its derivatives (Table 1).
2
.2.2. CNS-drugability
The Multiparameter Optimization (MPO) based on six factors (molecular weight, number
of hydrogen bond donors, the basic pK , topolgical polar surface area, logP and logD) was
a
calculated for all compounds in the series using in-house script according to rules defined by
Wager et al. for CNS drugs [21]. The obtained CNS MPO scores (Table 1), ranging from 4.33
(
compound 6) to 5.58 (compound 16), suggest a high preference of synthesized compounds
2-19) to become potential CNS drugs.
(
2
.3. Affinities for 5-HT serotonin receptors
6
The benzyl-triazine derivatives 2-19 were examined on their affinity for the human 5-HT6
receptors (Table 1) in the radioligand binding assay, in the same conditions as those for the
aforementioned compound 1 and its aryl derivatives (Modification A, Fig. 4). Displacement
6

ACCEPTED MANUSCRIPT
3
curves of [ H]-LSD binding were measured in HEK293 cells stably transfected with full-
length sequence of the 5-HT R [22]. All tested compounds, excluding 16, exhibited
6
significant activities with K values lower than 1 µM. Most of them (2-12, 18 and 19) had the
i
K values in the range of 100 nM or lower. The most active compounds (18 and 19) displayed
i
affinities in the range of 20 nM, which is more potent than that of known agent Ro 04-6790
[
12].
2
.4. Behavioural tests in vivo
Preclinical data provide support for use of 5-HT R ligands as promising drugs for treating
6
mood disorders [23-26]. Thus, compounds 4, 11, and 19 were tested in the forced swim test
(FST). All tested triazines significantly decreased immobility time (by 29%, 30% and 37 %
compared to respective control group) in this test at the dose of 0.3 mg/kg (Fig. 6). The U-
shaped dose-response effects in the FST test observed after administration of compounds 4,
1
1, and 19 are difficult to explain especially at this stage of behavioral experiments. However
noteworthy fact is that some antidepressants with different mechanisms of action also produce
U-shaped dose-response effects in some animal models and tests for antidepressant-like
activity [27, 28]. Moreover this effect was also observed for other 5-HT R ligands (i.e.
6
SB258585) [29].
Figure 6
2
.5. Structure-activity relationship
The structure-activity relationship (SAR) analysis indicates that the 5-HT R affinity is
6
distinctly determined by both, a kind of aromatic moieties linked by the methylene group to
the triazine ring and substituents at the aromatic moieties. Thus, more extended aromatic areas
of the fused rings of 1-naphthyl (18) or 3-indolyl (19) seem to be most profitable.
Regarding the benzyl fragment, it is observed that the type and the position of substituents
had significant influence on the affinity. Hence, the mono- meta substitution is undeniably
more favorable for all investigated substituents (Cl, F, CH , CH O). Both, the para and the
3
3
ortho, positions provided some decrease of the desirable activity, which was substituent-
dependent. This position preference seems to be corresponding to electron properties within
the phenyl ring, since both the ortho and the para positions are equally susceptible for
electron withdrawing/donating action, whereas the meta position is significantly less liable.
7

ACCEPTED MANUSCRIPT
The substituents represent the following chemical class: the weakly electron withdrawing
agents, Cl and F (3-9), the weakly electron donating substituent, methyl (10-12), and the
stronger electron donating substituent, methoxyl (13-15). Taking into account the obtained
results of the 5-HT R affinities, it is hard to find a regular preference in the electron properties
6
of substituents. The chlorine substituent was the most profitable (4), followed by methyl (11),
then fluorine (8), whereas the methoxyl substituent (13-15) caused a significant decrease of
activity in case of all compared positions, also in comparison to the substituent-free benzyl
compound 2 (Table 1).
In contrary, hydrophobic properties of the substituents affect the receptor affinity in
distinctly regular way. The least favorable methoxyl group is clearly less hydrophobic than
the rest of considered substituents, and even than hydrogen. Apart from the position of the
substituent at the phenyl ring, only the methoxyl group represents hydrophilic properties [30].
The hydrophobicity of the substituents decreases in the following order:
Cl>CH >F>H>OCH , and thus is in total accordance to the receptor affinity results for both
3
3
the meta or the ortho substituted compounds (Table 1). Nevertheless, the SAR analysis
indicates that the strongest decrease of the 5-HT R affinity was induced by an exchange of the
6
lead phenyl fragment into the basic and much more hydrophilic 4-pyridyl moiety (16). The
similar replacement with the 2-thiophenyl group was also unprofitable but the affinity
decrease was not so keen (17, Table 1).
Results of molecular modelling endorse the importance of hydrophobic properties of the
benzyl fragment for the binding with the 5-HT R. They suggest that the hydrophobic cavity
6
formed by transmembrane domains (TM) 3, 4, 5 and extracellular loop 2 (ECL2) is
particularly receptive for more hydrophobic moieties, and the use of more bulky rings in the
case 18 or 19 (1-naphthyl or 3-indolyl, respectively) may increase the CH- π/π-π interaction
giving stronger binding effect. In the case of methoxy-substituted benzyl moiety (13-15), not
only the decrease of hydrophobicity but also the presence of electron-donating oxygen, able to
interact with various amino acid residues, may hinder a desirable fitting in the hydrophobic
cave. For the similar reasons, the weakest activity is observed for the 4-pyridyl compound
(16) that possesses definitely the least hydrophobic properties within the whole investigated
series (2-19). The receptor affinity of the compound with the heterocyclic thiophenyl HYD
feature (17) was not spectacular, as well.
The aforementioned role of both, the kind of aromatic moieties and the kind of substituents
at phenyl rings, has been confirmed by docking studies (Fig. 7).
8

ACCEPTED MANUSCRIPT
Figure 7
Figure 7A shows comparison of the binding poses of the lead 2 (green) and 18 (cyan), the
most active modification of the lead in the series. The results showed that an increase of an
activity may be resulted by an enlargement of hydrophobic surface of a substituent that
interacts stronger with the hydrophobic crevice formed by TM 3, 4, 5 and ECL2. This
observation is supported by the increase of the binding free energy of about 11.1 kcal/mol
compared to the lead molecule 2.
For contrast, the worst modification (16) showed very similar binding mode to the lead 2
(Fig. 7B), however, replacement of a phenyl by a 4-pyridyl ring leads to decrease of the
binding free energy of about 7.1 kcal/mol. Such destabilization can be caused by repulsion
between ECL2 backbone carbonyl and 4-pyridyl units. Results of docking studies for
compounds with the meta-substitution (Fig. 7C), allow to analyze the role of substituents at
the phenyl ring for the interaction with the receptor. All meta-substituted analogs of 2
demonstrated high coherence of the binding mode. The change of binding free energy for this
substitution revealed that the best substituent, in concordance with the experimental binding
data, is chlorine (∆∆G increase about 10.2 kcal/mol). The next order is as follows: methoxyl,
methyl and fluorine, what compared to the experimental affinity data showed that binding free
energy was overestimated for methoxyl group, which was indicated as the least profitable in
the radioligand binding assay. On the basis of four variants of substituents, the docking
studies allowed for an insight into the role of the type and position of the substituent in the
molecular mechanism of receptor binding. The example influence of the position of the
substitution in the phenyl ring on the conformation change in the binding site is demonstrated
for the methyl substituent (Fig. 7D). Due to the steric hindrance of ECL2, the para-methyl
derivative is bent much more than the rest ones, what leads to deeper penetration of these
derivatives into the hydrophobic crevice.
Results of both, the radioligand binding assays and the docking studies, in respect to the
pharmacophore model of López-Rodríguez, demonstrate that the HBA, usually represented by
the sulfonyl group, is not required for potent affinity to the 5-HT R. Instead, a one-atom
6
spacer that can be represented by the sulfonyl as well as by the methylene group seems to be
crucial, and lack of this motif, observed in the case of compound 1, caused immense decrease
of the affinity to the 5-HT R
6
.
9

ACCEPTED MANUSCRIPT
The performed MPO calculations provided particular view on the ability of each
investigated compound (2-19) to be a CNS drug. The SAR in this case is almost inversely
proportional to that observed during the studies on the affinity for the 5-HT R (Table 1).
6
However, all compounds (2-19) showed significant values of the CNS MPO score, close to 5
even in the case of the worst compound 6. These results promote the tested series of 2,4,6-
substituted 1,3,5-triazines to be considered as potential CNS drug candidates with the
pharmacological action via serotonin receptors 5-HT R. This result of computational
6
simulation has been confirmed by the primary in vivo assays that were performed for most
active compounds (4, 11 and 19).
3
. Conclusion
The performed investigations allowed to identify a totally new group of potent 5-HT6
receptor ligands among arylmethyl derivatives of 1,3,5-triazine on the basis of previously
obtained group of aryl 1,3,5-triazine derivatives acted on the other GPCR receptor. Similarity
searching of ChEMBL database proved that no triazine compound has been annotated as a 5-
HT R agent so far. The series of new compounds designed as chemical modifications of the
6
lead structure 2, (4-benzyl-6-(4-methylpiperazin-1-yl)-1,3,5-triazin-2-amine), was synthesized
by the use of the guanylation and the cyclic condensation processes with suitable
(hetero)aromatic esters. The radioligand binding studies indicated the significant sub-
micromolar affinity for the 5-HT R for most of the tested benzyltriazine derivatives (2-19),
6
whereas the corresponding aryl-substituted triazines were weakly active. Results of
radioligand binding assays together with profound molecular modelling studies, including
docking to the homology model of 5-HT R, allowed to carry out SAR analysis to recognize
6
physicochemical properties that affect the affinity of the novel group of 5-HT R agents. The
6
SAR analysis underlines that the hydrophobic moiety substituted at the position 4 of the 1,3,5-
triazine is the crucial activity factor. Thus, (i) more extended and more hydrophobic aromatic
rings linked to the 1,3,5-triazine via methylene spacer or (ii) a benzyl with hydrophobic
substituents at the meta position, were identified as beneficial moieties required by the
hydrophobic cavity of the 5-HT R binding pocket. These studies allowed to reconsider the
6
previous pharmacophore model eroding the HBA pharmacophore feature. Results of primary
screening in vivo using the Porsolt assay confirmed an ability of this series to be CNS drugs.
Compounds 4, 11 and 19 with both, potent action on the 5-HT R and favorable CNS-
6
drugability properties, are especially interesting candidates for wider studies in a search for
1
0

ACCEPTED MANUSCRIPT
new drugs in the treatment of the civilization CNS diseases related to the serotonin 5-HT R
6
signal transduction.
4
. Experimental
4
.1. Chemistry
Reagents were manufactured by Alfa Aesar (Karlsruhe, Germany) or Sigma Aldrich
Darmstadt, Germany). Solvents were dried over calcium hydrochloride (toluene) or calcium
(
oxide (methanol). Reaction progress was verified using thin layer chromatography (TLC),
which was carried out on 0.2 mm Merck silica gel 60 F254 plates. Spots were visualized by
UV light or treatment with Dragendorff reagent. Melting points (mp) were determined using
1
13
MEL-TEMP II apparatus and are uncorrected. The H NMR and C NMR spectra were
obtained on a Varian Mercury-VX 300 Mz spectrometer in DMSO-d or CDCl . Chemical
6
3
1
shifts in H NMR spectra were reported in parts per million (ppm) on the δ scale using the
solvent signal as an internal standard. Data are reported as follows: chemical shift,
multiplicity (s, singlet; br.s., broad singlet; d, doublet; t, triplet; m, multiplet), coupling
constant J in Hertz (Hz), number of protons, proton’s position (Ph-phenyl, Pp-piperazine, Ind-
indole, Pyr-pyridine, Thio-thiophene). LC-MS were carried out on a system consisting of a
Waters Acquity UPLC, coupled to a Waters TQD mass spectrometer. Retention times (t ) are
R
given in minutes. The UPLC/MS purity of all final compounds was determined (%). Esters
3
e-15e were obtained according the procedure described previously [9,10], and some of them
(3e-5e, 7e-15e) are known in Chemical Abstract Database (see Supplementary Information).
4
.1.2. General procedure of synthesis of 1,3,5-triazines (2-19)
Sodium (10 mmol) was dissolved in 10 mL of absolute methanol, then 4-methylpiperazin-
1
-yl biguanidine x 2HCl (5 mmol) and the appropriate carboxylic acid ester 2e-19e (5 mmol)
were added. The reaction mixture was refluxed for 15-30h. After cooling to room
temperature, water (10 mL) was added and the mixture was stirred for 0.5 h. The precipitated
triazine product (2-19) was separated and crystallized from methanol.
4
-(2-Chlorobenzyl)-6-(4-methylpiperazin-1-yl)-1,3,5-triazin-2-amine (3)
Ester 3e (5 mmol) was used. White solid, mp 86-88 ˚C, yield: 25%, C H N Cl (MW
1
5
19
6
1
3
18.85). H NMR [DMSO-d ]: δ: 7.34-7.42 (m, 2H, Ph-3,5-H), 7.21-7.30 (m, 2H, Ph-4,6-H),
6
1
1

ACCEPTED MANUSCRIPT
6
.75 (br.s, 2H, NH ), 3.85 (s, 2H, Ph-CH ), 3.63 (t, J=4,92 Hz, 4H, Pp-2,6-H), 2.26 (t, J=4,89
2
2
1
3
Hz, 4H, Pp-3,5-H), 2.17 (s, 3H, Pp-CH ); C NMR [300 MHz, DMSO-d ] δ: 175.59, 167.29,
3
6
1
64.89, 136.15, 134.05, 132.23, 129.39, 128.57, 127.35, 54.73, 46.20, 42.77, 42.48.
+
/
+
LC/MS : purity: 99.76%, t =2.65,(ESI) m/z [M+H] 319.13.
R
Detailed synthetic procedures and analytical data for compounds 2, 4-19 are described in
Supplementary Data.
4
4
.2. Molecular modelling
.2.1. Homology modelling
Homology models of 5-HT receptor were built on a GPCR crystal structures, i.e. 5-HT
6
1B
(PDB ID: 4IAR, 4IAQ), 5-HT2B (PDB ID: 4IB4), A2A (PDB ID: 3QAK, 2YDV, 4UHR), β2
(
PDB ID: 3P0G, 4LDE, 3PDS), M (PDB ID: 4MQS), CXC chemokine receptor type 4 (PDB
2
ID: 3OE0), dopamine 3 receptor (PDB ID:3PBL) and histamine 1 receptor (PDB ID: 3RZE)
receptors, retrieved from the Protein Data Bank [31]. Sequences of the modelled receptor and
selected templates were obtained from the UniProtKB/Swiss-Prot database and next aligned
manually using Discovery Studio [32], making sure that the most conserved amino acid in
each helix, and motifs characteristic for class A GPCRs are on equivalent positions. The
prediction of the ranges of helices was supported using the web services: PONGO [33],
ExPASY PROSITE [34], and Psipred [35]. For each template, 20 models were generated
using Modeller v9.9 software [36], and tested by docking of active and inactive compounds
fetched from ChEMBL database [37]. For every model, a ROC curve was calculated based on
the Glide Score values of docked compounds (the undocked actives and inactives were
considered as false negatives and true negatives, respectively). The quality of models was
determined by the Area Under ROC curve (AUROC). Three models per template with the
highest AUROC value were selected for further studies.
The final validation of the selected receptor models was performed by docking of the all
set of studied molecules. Only those models were kept for which coherent for the whole set of
compounds, and explaining the main structure-activity relationships binding mode were
observed.
4
.2.2. CNS MPO Calculation
1
2

ACCEPTED MANUSCRIPT
The Central Nervous System Multiparameter Optimization (CNS MPO) score was
calculated based on the six molecular descriptors, i.e.: logP, logD, molecular weight (MW),
topological polar surface area (TPSA), the count of hydrogen bond donor (HBD) and the
strongest basic pK using Calculator Plugins [38] implemented in Chemaxon. For logP, logD,
a
MW, HBD, and pK monotonic decreasing function, whereas a hump function for TPSA was
a
used to transform the values of the descriptors into dimensionless scale. The most desirable
and the least desirable ranges for each descriptor were adopted from original paper [21].
Transformed values (desirability score) of the descriptors were determined and summed up
for each compound, to obtained the final CNS MPO, which can range from 0 (the worst CNS
drug) to six 6 (the best CNS drug).
4
.2.3. Molecular docking
The 3-dimensional structures of the ligands were prepared using LigPrep v3.6 [39], and the
appropriate ionization states at pH = 7.4 were assigned using Epik v3.4 [40]. The Protein
Preparation Wizard was used to assign the bond orders, appropriate amino acid ionization
states and to check for steric clashes. The receptor grid was generated (OPLS_2005 force
field) by centring the grid box with a size of 12 Å on Asp3.32. Automated docking was
performed using Glide v6.9 at the SP level with the flexible docking option turned on [41].
4
.2.4. Binding free energy calculation
MM-GBSA (Generalized-Born/Surface Area) was used to calculate the binding free
energy for all ligand-receptor complexes using Prime software from Schrödinger. To assess
the influence of a given substituent on the binding, the ∆∆G was calculated as a difference
between binding free energy (∆G) of unsubstituted 2 (TR20) and investigated analogues (3-
1
9).
4
.3. Radioligand binding assay
Radioligand binding assays were used to determine the affinity of the synthesized
compounds for human serotonin 5-HT R, which were stably expressed in HEK293 cells. This
6
3
-1
procedure was accomplished via the displacement of [ H]-LSD (85.2 Ci mmol ) for 5-HT R.
6
-11
-4
Each compound was tested in triplicate at 7 to 8 different concentrations (10 to 10 M). The
inhibition constants (K ) were calculated using the Cheng–Prusoff equation [42], and the
i
results are expressed as the mean of at least two independent experiments.
1
3

ACCEPTED MANUSCRIPT
4
4
.4. Behavioral studies in vivo
.4.1. Animals
The experiments were performed on male Wistar rats (170-200 g) obtained from an accredited
animal facility at Jagiellonian University Medical College, Poland. The animals were housed
0
in group of four in controlled environment (ambient temperature 21±2 C; relative humidity
5
0-60%; 12-h light/dark cycles (lights on at 8:00). Standard laboratory food (LSM-B) and
filtered water were freely available. Animals were assigned randomly to treatment groups. All
the experiments were performed by two observers unaware of the treatment applied between
9
:00 and 14:00 on separate groups of animals. All animals were used only once. Procedures
involving animals and their care were conducted in accordance with current European
Community and Polish legislation on animal experimentation. Additionally, all efforts were
made to minimize animal suffering and to use only the number of animals necessary to
produce reliable scientific data. The experimental protocols and procedures described in this
manuscript were approved by the I Local Ethics Commission in Cracow (no 293/2015) and
complied with the European Communities Council Directive of 24 November 1986
(86/609/EEC) and were in accordance with the 1996 NIH Guide for the Care and Use of
Laboratory Animals.
4
.4.2. Drugs
All compounds were suspended in 1% Tween 80 immediately before administration in a
volume of 2 mL/kg. Compounds were administered intraperitoneally (i.p.) 60 minutes before
testing. Control animals received vehicle (1% Tween 80) according to the same schedule.
4
.4.3. Forced swim test in rats procedures
The experiment was carried out according to the method of Porsolt et al. [43]. On the
first day of an experiment, the animals were gently individually placed in Plexiglas cylinders
40 cm high, 18 cm in diameter) containing 15 cm of water maintained at 23-25° C for 15
(
min. On removal from water, the rats were placed for 30 min in a Plexiglas box under 60-W
bulb to dry. On the following day (24 h later), the rats were re-placed in the cylinder and the
total duration of immobility was recorded during the whole 5-min test period. The immobility
was assigned when no additional activity was observed other than that necessary to keep the
rat’s head above the water. Fresh water was used for each animal.
1
4

ACCEPTED MANUSCRIPT
4
.4.4 Statistical analysis
The data of behavioral studies were evaluated by an analysis of variance one-way ANOVA
followed by Bonferroni’s post hoc test (statistical significance set at p<0.05).
Acknowledgements
This study was financially supported by Polish National Science Centre (NCN) grants: No
UMO-2015/17/B/NZ7/02973 and DEC-2011/02/A/NZ4/00031 (synthesis of 1, 2, 16-19).
1
5

ACCEPTED MANUSCRIPT
References
[
1] R. Kohen, M.A. Metcalf, N. Khan, T. Druck, K. Huebner, J.E. Lachowicz, H.Y.
Meltzer, D.R. Sibley, B.L. Roth, M.W. Hamblin, Cloning, Characterization, and
Chromosomal Localization of a Human 5-HT6 Serotonin Receptor, J. Neurochem. 66
(2002) 47–56.
[
[
2] M.J. FJ, Y. Shen, W. RP, H. MW, S. DR, Cloning and expression of a novel serotonin
receptor with high affinity for tricyclic psychotropic drugs, Mol Pharmacol. 43 (1993)
3
20–327.
3] M. Ruat, E. Traiffort, J.M. Arrang, J. Tardivellacombe, J. Diaz, R. Leurs, J.C.
Schwartz, A Novel Rat Serotonin (5-HT6) Receptor: Molecular Cloning, Localization
and Stimulation of cAMP Accumulation, Biochem. Biophys. Res. Commun. 193
(1993) 268–276.
[
4] D. Marazziti, S. Baroni, F. Borsini, M. Picchetti, E. Vatteroni, V. Falaschi, M. Catena-
Dell’Osso, Serotonin Receptors of Type 6 (5-HT6): From Neuroscience to Clinical
Pharmacology, Curr. Med. Chem. 20 (2013) 371–377.
[
[
5] D. Karila, T. Freret, V. Bouet, M. Boulouard, P. Dallemagne, C. Rochais, Therapeutic
Potential of 5-HT 6 Receptor Agonists, J. Med. Chem. 58 (2015) 7901–7912.
6] A. Quiedeville, M. Boulouard, V. Da Silva Costa-Aze, F. Dauphin, V. Bouet, T. Freret,
5
-HT6 receptor antagonists as treatment for age-related cognitive decline, Rev.
Neurosci. 25 (2014) 417–427.
[
[
[
7] M.L. López-Rodríguez, B. Benhamú, T. De La Fuente, A. Sanz, L. Pardo, M.
Campillo, A three-dimensional pharmacophore model for 5-hydroxytryptamine6 (5-
HT6) receptor antagonists, J. Med. Chem. 48 (2005) 4216–4219.
8] B. Benhamú, M. Martín-Fontecha, H. Vázquez-Villa, L. Pardo, M.L. López-Rodríguez,
Serotonin 5-HT 6 Receptor Antagonists for the Treatment of Cognitive Deficiency in
Alzheimer’s Disease, J. Med. Chem. 57 (2014) 7160–7181.
9] D. Łażewska, M. Więcek, J. Ner, K. Kamińska, T. Kottke, J.S. Schwed, M. Zygmunt,
T. Karcz, A. Olejarz, K. Kuder, G. Latacz, M. Grosicki, J. Sapa, J. Karolak-
Wojciechowska, H. Stark, K. Kieć-Kononowicz, Aryl-1,3,5-triazine derivatives as
1
6

ACCEPTED MANUSCRIPT
histamine H4 receptor ligands, Eur. J. Med. Chem. 83 (2014) 534–546.
[
[
10] K. Kamińska, J. Ziemba, J. Ner, J.S. Schwed, D. Łażewska, M. Więcek, T. Karcz, A.
Olejarz, G. Latacz, K. Kuder, T. Kottke, M. Zygmunt, J. Sapa, J. Karolak-
Wojciechowska, H. Stark, K. Kieć-Kononowicz, (2-Arylethenyl)-1,3,5-triazin-2-
amines as a novel histamine H4 receptor ligands, Eur. J. Med. Chem. 103 (2015) 238–
2
51.
11] G. Latacz, P. Kechagioglou, R. Papi, D. Łażewska, M. Więcek, K. Kamińska, P.
Wencel, T. Karcz, J.S. Schwed, H. Stark, D.A. Kyriakidis, K. Kieć-Kononowicz, The
Synthesis of 1,3,5-triazine Derivatives and JNJ7777120 Analogues with Histamine H 4
Receptor Affinity and Their Interaction with PTEN Promoter, Chem. Biol. Drug Des.
8
8 (2016) 254–263.
[
[
12] A.J. Sleight, F.G. Boess, M. Bös, B. Levet-Trafit, C. Riemer, A. Bourson,
Characterization of Ro 04-6790 and Ro 63-0563: potent and selective antagonists at
human and rat 5-HT6 receptors., Br. J. Pharmacol. 124 (1998) 556–62.
13] S.M. Bromidge, A.M. Brown, S.E. Clarke, K. Dodgson, T. Gager, H.L. Grassam, P.M.
Jeffrey, G.F. Joiner, F.D. King, D.N. Middlemiss, S.F. Moss, H. Newman, G. Riley, C.
Routledge, P. Wyman, 5-Chloro-N-(4-methoxy-3-piperazin-1-yl- phenyl)-3-methyl-2-
benzothiophenesulfon- amide (SB-271046): a potent, selective, and orally bioavailable
5
-HT6 receptor antagonist., J. Med. Chem. 42 (1999) 202–5.
[
14] W.D. Hirst, T.O. Stean, D.C. Rogers, D. Sunter, P. Pugh, S.F. Moss, S.M. Bromidge,
G. Riley, D.R. Smith, S. Bartlett, C.A. Heidbreder, A.R. Atkins, L.P. Lacroix, L.A.
Dawson, A.G. Foley, C.M. Regan, N. Upton, SB-399885 is a potent, selective 5-HT6
receptor antagonist with cognitive enhancing properties in aged rat water maze and
novel object recognition models, Eur. J. Pharmacol. 553 (2006) 109–119.
[
[
15] K. Kulig, M. Ignasik, B. Malawska, Synthesis of 3,3- and 4,4-Alkyl-phenyl-substituted
Pyrrolidin-2-one Derivatives, Pol. J. Chem. 83 (2009) 1629–1636.
16] T. De La Fuente, M. Martín-Fontecha, J. Sallander, B. Benhamú, M. Campillo, R.A.
Medina, L.P. Pellissier, S. Claeysen, A. Dumuis, L. Pardo, M.L. López-Rodríguez,
Benzimidazole derivatives as new serotonin 5-HT6 receptor antagonists. Molecular
mechanisms of receptor inactivation, J. Med. Chem. 53 (2010) 1357–1369.
1
7

ACCEPTED MANUSCRIPT
[
17] S.G.F. Rasmussen, Choi, Fung, Pardon, Casarosa, Chae, Devree, Rosenbaum, Thian,
Kobilka, A. Schnapp, Konetzki, R.K. Sunahara, Gellman, Pautsch, J. Steyaert, Weis,
Kobilka, Structure of a nanobody-stabilized active state of the β(2) adrenoceptor,
Nature. 469 (2011) 175–180.
[
[
18] A.M. Ring, A. Manglik, A.C. Kruse, M.D. Enos, W.I. Weis, K.C. Garcia, B.K.
Kobilka, Adrenaline-activated structure of β2-adrenoceptor stabilized by an engineered
nanobody., Nature. 502 (2013) 575–9.
19] D.M. Rosenbaum, C. Zhang, J. a Lyons, R. Holl, D. Aragao, D.H. Arlow, S.G.F.
Rasmussen, H.-J. Choi, B.T. Devree, R.K. Sunahara, P.S. Chae, S.H. Gellman, R.O.
Dror, D.E. Shaw, W.I. Weis, M. Caffrey, P. Gmeiner, B.K. Kobilka, Structure and
function of an irreversible agonist-β(2) adrenoceptor complex., Nature. 469 (2011)
2
36–40.
[
[
20] Schrödinger Release 2016-3: Prime, Schrödinger, LLC, New York, NY, 2016.
21] T.T. Wager, X. Hou, P.R. Verhoest, A. Villalobos, Moving beyond rules: The
development of a central nervous system multiparameter optimization (CNS MPO)
approach to enable alignment of druglike properties, ACS Chem. Neurosci. 1 (2010)
4
35–449.
[
22] P. Zajdel, K. Marciniec, A. Maślankiewicz, K. Grychowska, G. Satała, B. Duszyńska,
T. Lenda, A. Siwek, G. Nowak, A. Partyka, D. Wróbel, M. Jastrzębska-Więsek, A.J.
Bojarski, A. Wesołowska, M. Pawłowski, Antidepressant and antipsychotic activity of
new quinoline- and isoquinoline-sulfonamide analogs of aripiprazole targeting
serotonin 5-HT1A/5-HT2A/5-HT7 and dopamine D2/D3 receptors, Eur. J. Med. Chem.
6
0 (2013) 42–50.
[
[
23] K. Hirano, T.M. Piers, K.L. Searle, N.D. Miller, A.R. Rutter, P.F. Chapman,
Procognitive 5-HT6 antagonists in the rat forced swimming test: potential therapeutic
utility in mood disorders associated with Alzheimer’s disease, Life Sci. 84 (2009) 558–
6
2.
24] A. Wesołowska, The anxiolytic-like effect of the selective 5-HT6 receptor antagonist
SB-399885: the impact of benzodiazepine receptors, Eur. J. Pharmacol. 580 (2008)
3
55–60.
1
8

ACCEPTED MANUSCRIPT
[
[
25] G.V. Carr, L.E. Schechter, I. Lucki, Antidepressant and anxiolytic effects of selective
5
-HT receptor agonists in rats, Psychopharmacology (Berl) 213 (2011) 499–507.
6
26] M. Jastrzębska-Więsek, A. Siwek, A. Partyka, P. Mierzejewski, P. Bieńkowski, M.
Kołaczkowski M, Wesołowska A, Potential antidepressant-like effect of the potent
serotonin 5-HT6 receptor agonist EMD 386088 in the forced swin test in rats,
Pharmacol. Rep. 62 (S1) (2010) 45.
[
[
[
27] F. Borsini, A. Meli, Is the forced swimming test a suitable model for revealing
antidepressant activity?, Psychopharmacology (Berl), 94 (1988) 147–60.
28] F. Borsini, Role of the serotonergic system in the forced swimming test, Neurosci.
Biobehav. Rev. 19 (1995) 377–95.
29] A. Wesołowska, A. Nikiforuk, K. Stachowicz, Anxiolytic-like and antidepressant-like
effects produced by the selective 5-HT6 receptor antagonist SB-258585 after
intrahippocampal administration to rats, Behav. Pharmacol. 18 (2007) 439–46.
[
30] H. Kubinyi, QSAR: Hansch Analysis and Related Approaches, Wiley-VCH Verlag
GmbH, Weinheim, Germany, 1993.
[
[
31] H.M. Berman, The Protein Data Bank, Nucleic Acids Res. 28 (2000) 235–242.
32] Dassault Systèmes BIOVIA, Discovery Studio Modeling Environment, Release 2016,
San Diego: Dassault Systèmes.
[
33] M. Amico, M. Finelli, I. Rossi, A. Zauli, A. Elofsson, H. Viklund, G. von Heijne, D.
Jones, A. Krogh, P. Fariselli, P. Luigi Martelli, R. Casadio, PONGO: a web server for
multiple predictions of all-alpha transmembrane proteins, Nucleic Acids Res. 34 (2006)
W169–W172.
[
34] C.J.A. Sigrist, L. Cerutti, N. Hulo, A. Gattiker, L. Falquet, M. Pagni, A. Bairoch, P.
Bucher, PROSITE: a documented database using patterns and profiles as motif
descriptors., Brief. Bioinform. 3 (2002) 265–274.
[
[
35] L.J. McGuffin, K. Bryson, D.T. Jones, The PSIPRED protein structure prediction
server., Bioinformatics. 16 (2000) 404–405.
36] A. Šali, T.L. Blundell, Comparative Protein Modelling by Satisfaction of Spatial
1
9

ACCEPTED MANUSCRIPT
Restraints, J. Mol. Biol. 234 (1993) 779–815.
[
[
37] A. Gaulton, L.J. Bellis, a P. Bento, J. Chambers, M. Davies, A. Hersey, Y. Light, S.
McGlinchey, D. Michalovich, B. Al-Lazikani, J.P. Overington, ChEMBL: a large-scale
bioactivity database for drug discovery., Nucleic Acids Res. 40 (2012) D1100-7.
38] Calculator Plugins were used for structure property prediction and calculation, Marvin
1
5.4.13.0, 2015, ChemAxon (http://www.chemaxon.com).
[
[
[
[
39] Schrödinger Release 2016-3: LigPrep, Schrödinger, LLC, New York, NY, 2016.
40] Schrödinger Release 2016-3: Epik, Schrödinger, LLC, New York, NY, 2016.
41] Schrödinger Release 2016-3: Glide, Schrödinger, LLC, New York, NY, 2016.
42] Y. Cheng, W.H. Prusoff, Relationship between the inhibition constant (K1) and the
concentration of inhibitor which causes 50 per cent inhibition (I50) of an enzymatic
reaction., Biochem. Pharmacol. 22 (1973) 3099–108.
[
43] R.D. Porsolt, A. Bertin A, M. Jalfre, Behavioural despair” in rats and mice: strain
differences and the effects of imipramine, Eur. J. Pharmacol. 51 (1978) 291–4.
2
0

ACCEPTED MANUSCRIPT
List of Tables, Figures and Schemes
st
1
2
. Table 1 Structures and pharmacological properties of the 1 generation of 5-HT R triazine agents
6
. Fig. 1. Pharmacophore model for the 5-HT R ligands elaborated by López-Rodríguez et al.[7]
6
(
(
distances in Å); PI (red) – positive ionizable atom; HBA (green) – hydrogen bond acceptor, HYD
blue) – hydrophobic site, AR (yellow) – aromatic-hydrophobic ring.
3
. Fig. 2. Structures and the human histamine H R (hH R) affinity of TR7(1) [9] and TR20(2) [11].
4 4
4
. Fig. 3. Structures and the receptor affinity of selected, first identified 5-HT R ligands. The CNS
6
MPO values were calculated using in-house script (see Experimental section).
5
. Fig. 4. Identification of a lead structure 2 (TR20) for 1,3,5-triazine 5-HT ligands.
6
6
. Fig. 5. Reconstructed 3-dimensional pharmacophore model for 5-HT R ligands proposed by
6
López-Rodríguez [7] including four key pharmacophore features: a positive ionizable group (PI,
red), an aromatic ring (AR, orange), a hydrogen bond acceptor group (HBA, green) and
hydrophobic group (HYD, blue). The lead compound 2 mapped onto the pharmacophore model (A)
and binding site of 5-HT R (B).
6
7
. Fig. 6. Effects of compounds 4, 11, and 19 on the immobility time of rats in the forced swim test.
Compounds 4, 11, and 19 were administered i.p. 60 min before the test. The animals were observed
for 5 min. The data are presented as the mean ± SEM of 6-8 rats. The data were statistically
evaluated by one-way ANOVA followed by Bonferroni’s post-hoc test, * p<0.01, **p<0.0001 vs
respective vehicle group.
8
. Fig. 7. Detailed view of the binding modes of a lead compound 2 (green) with the best 18 (cyan)
and worst 16 (grey) derivatives from the series, (A) and (B) respectively. Binding mode (C) of
meta-substituted derivatives, i.e.: -Cl 4 (magenta), -F 8 (brown), -CH 11 (pink) and -OCH 14
3
3
(
yellow). Comparison of the binding mode of methyl derivatives (10-12) substituted in ortho (red),
meta (pink) and para (dark green) positions in the phenyl ring (D).
9
. Scheme 1. Synthesis route for intermediates 22, 3e-5e, 7e-15e (a), and final products 2-19 (b).
Substituents R for acids (3a-5a, 7a-15a), esters (2e-19e) and triazines (2-19) are number
compatible, according to Table 1. Reagents and conditions: (i) BuOH, temperature gradually
increased from 50 to 90 ˚C during 1 h, 5 h reflux; (ii) CH I, DBU, toluene, 48 h stirring, r.t. (for 4a
3
+
and 5a); (iii) MeOH, H , reflux, 7-9 h (for 3a, 7a-15a); (iv) MeONa, reflux for 15-30 h.
2
1

ACCEPTED MANUSCRIPT
Table 1
st
1
. Table 1 Structures and pharmacological properties of the 1 generation of 5-HT R triazine agents.
6
b
c
Cmpd R
5-HT K ±SEM
∆∆G
CNS MPO
6
i
a
[nM]
[kcal/mol]
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
Phenyl
95 ± 7
56 ± 4
28 ± 3
52 ± 6
32 ± 5
88 ± 11
64 ± 3
100 ± 16
80 ± 7
30 ± 4
64 ± 9
193 ± 18
158 ± 11
440 ± 59
1828 ± 313
305 ± 27
20 ± 3
-
5.54
5.07
5.07
5.07
4.33
5.45
5.45
5.45
5.23
5.23
5.23
5.56
5.56
5.56
5.58
5.33
4.55
4.85
2-Chlorophenyl
3-Chlorophenyl
4-Chlorophenyl
3,4-dichlorophenyl
2-Fluorophenyl
3-Fluorophenyl
4-Fluorophenyl
2-Methylphenyl
3-Methylphenyl
4-Methylphenyl
2-Methoxylphenyl
3-Methoxyphenyl
4-Methoxyphenyl
4-Pyridyl
-4.9
-10.2
-7.7
-14.1
-5.0
-1.7
0.4
-4.0
-5.1
-4.1
2.3
-7.2
-4.9
7.1
-1.9
-11.1
-5.7
0
1
2
3
4
5
6
7
8
9
2-Thiophenyl
1-Naphthyl
3-Indolyl
22 ± 2
a
Tested experimentally in the radioligand binding assay, binding affinity, K , expressed as the average of at least
i
b
two independent experiments; calculated as a difference of binding free energy of 2 and a given compound, the
negative values indicate an increase of L-R complex stability; calculated using Multiparameter Optimization
c
(
MPO) procedure based on 6 factors.

ACCEPTED MANUSCRIPT
Fig. 1. Pharmacophore model for the 5-HT R ligands elaborated by López-Rodríguez et al.[7]
6
(
distances in Å); PI (red) – positive ionizable atom; HBA (green) – hydrogen bond acceptor, HYD
(
blue) – hydrophobic site, AR (yellow) – aromatic-hydrophobic ring.

ACCEPTED MANUSCRIPT
Scheme 1. Synthesis route for intermediates 22, 3e-5e, 7e-15e (a), and final products 2-19 (b).
Substituents R for acids (3a-5a, 7a-15a), esters (2e-19e) and triazines (2-19) are number compatible,
according to Table 1. Reagents and conditions: (i) BuOH, temperature gradually increased from 50 to
9
0 ˚C during 1 h, 5 h reflux; (ii) CH I, DBU, toluene, 48 h stirring, r.t. (for 4a and 5a); (iii) MeOH,
3
+
H , reflux, 7-9 h (for 3a, 7a-15a); (iv) MeONa, reflux for 15-30 h.

ACCEPTED MANUSCRIPT
Fig. 2. Structures and the human histamine H R (hH R) affinity of TR7(1) [9] and TR20(2)
4
4
[
11].

ACCEPTED MANUSCRIPT
H CO
3
H C
3
Cl
Cl
NH
N
Cl
H CO
S
O
3
NH2
O
N
S
N
NH
H CO
O
S
O
N
H
HN
N
H C
O
CH3
3
S
N
H
N
H
HN
3
O
SB-399885
h5-HT pK = 9.14 [13]
CNS MPO = 4.36
Ro 04-6790
h5-HT pK = 7.26 [11]
CNS MPO = 3.96
SB-271046
h5-HT pK = 9.2 [12]
CNS MPO = 4.17
6
i
6
i
6
i
Fig. 3. Structures and the receptor affinity of selected, first identified 5-HT R ligands. The CNS MPO
6
values were calculated using in-house script (see Experimental section).

ACCEPTED MANUSCRIPT
Fig. 4. Identification of a lead structure 2 (TR20) for 1,3,5-triazine 5-HT ligands.
6

ACCEPTED MANUSCRIPT
Fig. 5. Reconstructed 3-dimensional pharmacophore model for 5-HT R ligands proposed by López-
6
Rodríguez [7] including four key pharmacophore features: a positive ionizable group (PI, red), an
aromatic ring (AR, orange), a hydrogen bond acceptor group (HBA, green) and hydrophobic group
(
HYD, blue). The lead compound 2 mapped onto the pharmacophore model (A) and binding site of 5-
HT R (B).
6