Property-Based Design and Synthesis of New Chloroquine Hybrids via Simple Incorporation of 2-Imino-thiazolidin-4-one or 1H-Pyrrol-2,5-dione Fragments on the 4-Amino-7-chloroquinoline Side Chain

No presente trabalho realizou-se a síntese de novos N-derivados da 4-amino-7-cloroquinolina modificando seletivamente o grupo amino terminal das N-(7-cloroquinolin-4-il)-alquildiaminas, a base do fármaco cloroquina (CQ) mediante a incorporação de sistemas heterocíclicos da 2-imino-tiazolidin-4-ona e 1H-pirrol-2,5-diona. Esses derivados foram selecionados graças às suas propriedades características, e avaliados mediante crivado virtual empregando as plataformas OSIRIS e Molinspiration. Os derivados quinolínicos assim desenhados e sintetizados poderiam incrementar a atividade antimalárica dos análogos da CQ sem afetar a lipofilia como tem se descrito na literatura, postulando-se como candidatos em posteriores testes biológicos.


Introduction
Drug-like properties are an integral element of drug discovery projects.This term became commonly used following the pivotal work of Lipinski and co-workers. 1,2rug-like property optimization is an important area of antiparasitic drug discovery, in general, and antiplasmodial agents development, in particular.][5] The development of antiplasmodial agents directed towards a single therapeutic target has shown not to be the best strategy within the progress against this devastating disease.4][5][6] The tertiary amino group present in most 4-aminoquinoline drugs is considered responsible for the drug accumulation within the parasite digestive vacuole, and to provide lipophilicity because of its basic character. 7Nevertheless, the introduction of structural features that highly decrease this group basicity has proved to increase the antimalarial activity of several CQ analogues. 3,5These discoveries established the amino side chain modifications as an alternative strategy for the new and selective antimalarials development.These modifications consist mainly in the introduction of small N-heterocycles whose ring structure would increase the antimalarial activity or potentiate other bioeffects.
Based on this evidence, in the present work, the design and synthesis of potential new 4-aminoquinoline druglike molecules were developed, through the systematic analysis of the desired product structure and the selective modification of the amino group, incorporating the biologically privileged 2-imino-thiazolidin-4-ones and cyclic N-imides scaffolds using a new strategy of "propertybased design". 22These systems are present in diverse bioactive molecules, and additionally could prevent the enzymatic dealkylation without affecting lipophilicity, and increase antimalarial activity of CQ, as described in literature.

Results and Discussion
The present work describes the efficient synthesis of new 4-amino-7-chloroquinoline derivatives functionalized with 2-imino-tihazolidin-4-one or 1H-pyrrol-2,5-dione systems separated by a short aminoalkyl chain.4][25][26][27] The obtained results evidence the potential in further biological studies of these derivatives thanks to their considerable drug-score and excellent reaction yields.

Rational molecular design
The principal idea of molecular hybrids formation is the lead antimalarial agent development with a potential and selective activity against multidrug resistant P. falciparum strains.The 2-imino-thiazolidin-4-one and 1H-pyrrol-2,5dione systems possess the adequate characteristics to be considered as reversal agent candidates within the new quinoline derivatives construction.
Using the Lipinski's rule 25 and drug-score criterion, 24 the structures of a new series of CQ analogues based on the 4-amino-7-chloroquinoline skeleton and these N-heterocyclic moieties were evaluated, assembling these compounds from synthetic and commercially available precursors (Figure 2).
Employing the Molinspiration software, 28 the planned compounds 13-22 were subjected to the Lipinski's rule of five analysis (drug-likeness), which indicates if a chemical substance can be orally active in humans. 1,24Currently, there are many approaches that assess a compound drug-likeness based on topological descriptors, fingerprints of molecular drug-likeness structure keys or other properties such as TPSA.In the Molinspiration program, the occurrence frequency of each fragment is determined within the collection created by shreddering 3300 traded drugs, as well as 15000 commercially available chemicals yielding a complete list of all available fragments. 23In this work, the Molinspiration program was used for the fragment based drug-likeness calculation of all desired compounds   , and nOHNH = 0-4) 1,27 (Table 1).TPSA has been shown to be a very good descriptor characterizing drug absorption, including intestinal absorption, bioavailability, Caco-2 permeability and blood-brain barrier penetration. 25,26rediction results for the compounds 13-22 molecular properties (Table 1), show TPSA values between 63 and 69 Å 2 confirming their drug-relevant properties.In order to assess the possible pharmacological properties of hybrids 13-22, a toxicity profile evaluation was performed employing the OSIRIS software, 29 as it may point to the presence of some fragments generally responsible for the irritant, mutagenic, tumorigenic, or reproductive effects in these molecules. 24As shown in Figure 3 (left panel), with the exception of compounds 21 and 22, all desired products represented low or moderate biological risks.
Moreover, we used the OSIRIS program to predict the compounds drug-score (Figure 3, right panel).Products 17-20 revealed the effect of the anhydride structure over this parameter, and can be attributed to the presence of substitutions on positions 2 or 5 of the pyrrolidone ring.This theoretical data is a clear evidence of the biological potential of the designed 4-aminoquinoline series, and support further experimental bioactivity assessments, pointing these hybrids as lead compounds with a low toxicity risk profile.

Synthesis of 4-aminoquinoline derivatives bearing N-heterocyclic systems on the side chain amino group
Final compounds 13-22 and their 4-amino-7-chloroquinoline precursors 8 were synthesized by direct and effective protolcols. 6,30The side chain amino group  (compound 8) was bound to the 4,7-dichloroquinoline core by means of nucleophilic aromatic substitution S N Ar, using 4,7-dichloroquinoline 6 and the corresponding commercial diaminoalkanes 7.This reaction was performed under reflux, in the presence of excess of diamine as solvent.9][20] In this work, the quinoline derivatives bearing the 2-imino-tiazolidin-4-one framework 13 and 14 were obtained in good yields as white solids from the quinoline diamines 8, potassium thiocyanate, and chloroacetyl chloride in anhydrous acetone at room temperature, without isolation of thiourea derivatives 13a and 14a (Scheme 1, route A).
On the other hand, formation of two isomers iminothiazolidinone and N-alkylimino-thiazolidinone has been described by Ottaná et al., 14 and explained by the possible condensation of chloroacetyl chloride with the two different nitrogen atoms on the thiocarbonyl group, causing the two different thiols intermediates generation.However, according to gas chromatography and nuclear magnetic resonance (NMR) analysis, compounds 13 and 14 were obtained exclusively as 2-imino-thiazolidin-4-ones in a 75-78% yield (Table 2), and N-alkylimino-thiazolidin-4-one formation was not detected.Single isomer synthesis

Not isolated
Not isolated 7a, 8a n = 1 7b, 8b n = 2 is probably favoured by the higher electron availability of the NH 2 group, compared with the highly hindered NH group bearing the N-alkylaminoquinoline framework on intermediates 13a and 14a.In order to extend the molecular library of hybrids between CQ and N-heterocyclic systems, new 4-aminoquinolines functionalized with 1H-pyrrol-2,5dione framework were prepared.Synthesis of compounds 15-22 was performed according to the procedure established by Deng et al., 20 starting from quinoline diamines 8 and cyclic anhydrides 9-12 (Scheme 1, route B).Intramolecular cyclization of the amic acid intermediates 15a-22a was accomplished using acetic anhydride in presence of sodium acetate (5% mol).Most of the anhydride precursors employed were commercially available.Nevertheless, for preparation of hybrids 19 and 20, synthesis of anhydride 11 through intermolecular cycloaddition of furan and maleic anhydride was developed according to Deng et al. 16 Commonly employed conditions within the N-alkylimide synthesis require high temperatures and long reaction times, reaching moderate yields. 31In this work, 4-aminoquinoline derivatives functionalized with N-alkylimide system were obtained in high yield (85-98%), using lower reaction temperature and shorter reaction times (Table 2).
The structures of all new CQ hybrids 13-22 were strongly confirmed by NMR, IR and GC-MS analysis (Table 2 and Experimental part).
In summary, in this paper, we have reported a versatile and convenient route for the synthesis of novel hybrids between 4-aminoquinoline and N-heterocyclic systems previously designed.Obtained theoretical data awakes further experimental assays and establish these molecules as lead compounds with a low toxicity risk profile.The preliminary antimalarial evaluation of some of these hybrid analogues, prepared here against P. falciparum, has been studied. 32Based on these results further synthesis of new molecular hybrids by changing the substituents on the small heterocyclic ring and quinoline nucleus in order to enhance the antimalarial activity is being investigated and will be reported in due course.

Materials and methods
Melting points (mp, uncorrected) were determined on a Fisher-Johns melting point apparatus.The IR spectra were recorded on a Lumex Infralum FT-02 spectrophotometer in KBr. 1 H NMR spectra were recorded on Bruker AM-400 spectrometers, using DMSO-d 6 .Chemical shifts are reported in ppm.Signals are designated as follows: s, singlet; d, doublet; dd, doublet of doublets; ddd, doublet of doublets of doublets; t, triplet; dt, doublet of triplets; td, triplet of doublets; q, quartet; quint., quintet; m, multiplet; br, broad.A Hewlett Packard 5890a series II Gas Chromatograph interfaced to an HP 5972 Mass Selective Detector (MSD) with an MP MS Chemstation Data system was used for MS identification at 70 ev using a 60 m capillary column coated with hp-5 [5%-phenyl-poly(dimethyl-siloxane)].Elemental analyses were performed on a Perkin Elmer 2400 Series II Analyzer and were within ± 0.4 of theoretical values.The reaction progress was monitored using thin layer chromatography on a silufol UV254 TLC aluminium sheet.
Molecular design was performed based on structure activity relationship studies, and virtual screening analysis reported in literature. 23,24Lipinki's rule of five parameters were calculated using the Molinspiration virtual platform (http://www.molinspiration.com/services/).The toxicity risk profile evaluation was accomplished employing the OSIRIS program available free at http://www.organicchemistry.org/prog/peo.The OSIRIS and Molinspiration Property Explorers shown in these pages are an integral part of some pharmaceutical companies' in-house substance registration system.They allow drawing chemical structures and calculating on-the-fly various drug-relevant properties whenever a structure is valid.Prediction results are valued and colour coded.Properties with high risks of undesired effects like mutagenicity or a poor intestinal absorption are shown in red, whereas a green colour indicates drug conform behaviour (Figure 3; see colors online).

Chemistry
General procedure for synthesis of 4-amino-7-chloroquinolines (8) 6,30 4.00 g (20.2 mmol) of 6 and a,w-diaminoalkane (7a, b) (10.1 mmol) was heated at 80 °C for 1 h with stirring and subsequently at 140-150 °C for 6-7 h with continued stirring.The reaction mixture was cooled to room temp and basified with 10% NaOH (70 mL).The resultant mixture was extracted with chloroform-methanol (20:1, 4 × 50 mL).The combined extracts were washed with brine, dried over anhydrous Na 2 SO 4 , concentrated under reduced pressure and the residue (10 mL) was precipitated by the addition of n-heptane (70 mL).The solid was purified by washing with 40 mL ethyl ether.Physicochemical characteristics of the obtained compounds 8 are given in the Supplementary Information.
General procedure for synthesis of 4-amino-7-chloroquinoline and 2-imino-thiazolidin-4-one hybrids (13, 14)  Potassium thiocyanate (0.50 g, 4.50 mmol) was dissolved in 5.0 mL of anhydrous acetone under vigorous agitation.Subsequently, 1.00 g (4.52 mmol) of quinoline diamines (8a or 8b) in acetone were added dropwise to the above solution and the reaction mass was stirred for 7 h according to TLC analysis.Without any further purification process, over the resulting mass, triethylamine (3.00 g, 29.60 mmol) and then a solution of chloroacethyl chloride in anhydrous acetone (1.10 g, 9.30 mmol) was added during 30 min at 0 ºC.The reaction mass was heated to room temperature and stirred over a 24 h period.Resulting yellowish mass was neutralized with 30 mL of sodium bicarbonate 10%, and extracted with ethyl acetate (2 × 30 mL).The organic layer was dried over Na 2 SO 4 and concentrated under vacuum.Finally, hybrids 13, 14 were purified by flash column chromatography using (20:1) ethyl acetate and methanol mixtures as eluents.Physicochemical characteristics of the obtained compounds are given in Table 2. General procedure for synthesis of chloroquine and 1H-pyrrol-2,5-dione hybrids (15-22)  1.0 g (4.24 mmol) of quinoline diamines (8a or 8b) was dissolved in 5.0 mL of anhydrous dichloromethane under vigorous agitation.Then, a solution of anhydride (9-11) (0.7 g, 4.24 mmol) in dichloromethane was added dropwise to the above solution for 30 min at 0 °C, stirring the white precipitate formed for 2 h.Solvent excess was distilled under reduced pressure, and the remaining solid was filtered and washed with diethyl ether.Without any further purification process, the obtained amic acid 15a-22a was mixed with 0.05 g (0.61 mmol) of sodium acetate and dissolved in 1.0 mL acetic anhydride.Reaction temperature was adjusted to 60 ºC and kept constant over a 5 h period.Finally, the reaction mass was allowed to cool and neutralized with sodium bicarbonate (1 mol L -1 ).The coloured precipitate was filtered, washed with cold water, and air dried.Hybrids 15-22 were finally purified by flash column chromatography using (20:1) ethyl acetatemethanol mixtures as mobile phase and TLC control (R f ).Physicochemical characteristics of the obtained compounds are given in Table 2.

+
also comparing them with CQ structure 5. Conversely to the CQ as reference, the obtained calculations demonstrate that all analyzed compounds contain high bioavailability properties, and fulfil all parameters established by this rule (molecular weight = 269.73-408.91,log P = 2.66-4.92,nON = 2-6

Table 1 .
Calculated Lipinski's rule of five parameters for the 4-aminoquinoline bearing N-heterocyclic frameworks Polar surface area; b number of hydrogen bond acceptors; c number of hydrogen bond donors; d number of rotatable bonds. a