Design and Microwave Synthesis of New (5Z) 5-Arylidene-2-thioxo-1,3-thiazolinidin-4-one and (5Z) 2-Amino-5-arylidene-1,3-thiazol-4(5H)-one as New Inhibitors of Protein Kinase DYRK1A

Here, we report on the synthesis of libraries of new 5-arylidene-2-thioxo-1,3-thiazolidin-4-ones 3 (twenty-two compounds) and new 2-amino-5-arylidene-1,3-thiazol-4(5H)-ones 5 (twenty-four compounds) with stereo controlled Z-geometry under microwave irradiation. The 46 designed final compounds were tested in order to determine their activity against four representative protein kinases (DYR1A, CK1, CDK5/p25, and GSK3α/β). Among these 1,3-thiazolidin-4-ones, the molecules (5Z) 5-(4-hydroxybenzylidene)-2-thioxo-1,3-thiazolidin-4-one 3e (IC50 0.028 μM) and (5Z)-5-benzo[1,3]dioxol-5-ylmethylene-2-(pyridin-2-yl)amino-1,3-thiazol-4(5H)-one 5s (IC50 0.033 μM) were identified as lead compounds and as new nanomolar DYRK1A inhibitors. Some of these compounds in the two libraries have been also evaluated for their in vitro inhibition of cell proliferation (Huh7 D12, Caco2, MDA-MB 231, HCT 116, PC3, and NCI-H2 tumor cell lines). These results will enable us to use the 1,3-thiazolidin-4-one core as pharmacophores to develop potent treatment for neurological or oncological disorders in which DYRK1A is fully involved.


Introduction
Protein kinases represent an important class of enzymes that play an important role in the regulation of various cellular processes. These enzymes catalyze proteinphosphorylation on serine, threonine, and tyrosine residues, which are frequently deregulated in human diseases. Only a small fraction of the 538 human kinases have been Efforts for the synthetic preparation of new chromeno [3,4-b]indoles as Lamellarin D isosteres [21] have been developed, and compound (XI: 67 nM) emerged among the Lamellarin D derivatives. In 2013, the first successful synthesis and separation of the atropoisomeric (aR)-and (aS)-16-methyl lamellarins N and their different kinase inhibitory action against DYRK1A were reported [22]. The isomer (aR) (XII) showed potent but nonselective inhibition of DYRK1A (42 nM). The 2-amino imidazolone core is present in several marine sponges [23], such as polyandrocarpamines [24], dispacamides [25,26], aplysinopsine [27,28], clathridine [29], hymenialdisine [30][31][32][33][34][35][36], and Leucettamine B. This last alkaloid was isolated in 1993 from the marine calcareous sponge Leucetta microraphis Haeckel of the Argulpelu Reef in Palau [37], and no significant biological activity has been reported. From a biological screening, we discovered that the Leucettamine B analogues [38], named leucettines, exerted a selective inhibition toward protein kinases. A detailed structureactivity relationship (SAR) led to the identification of leucettine L 41 (XIII) as good inhibitor of DYRKs and cdc2-like kinases (CLKs), which are involved in Alzheimer's disease/Down syndrome [39]. We also demonstrated that leucettine L41 inhibits DYRK1A and CLKs in a cellular context and displayed neuropropective properties [40].
The 2-amino 5-arylidene-5H-thiazol-4-ones and their 5-arylidene rhodanine precursors are a class of five-membered heterocyclic rings (FMHRs) considered as "privileged scaffolds" in the medicinal chemistry community [41], and considerable work has been published over decades about their chemistry and biology. Some of them exhibited antiinflammatory [42], anti-tumor effects [43], or have been identified as selective inhibitors of the dual specificity phosphatase (DSP) family [44].
Continuing in the effort to identify new DYRK1A inhibitors, we decided to explore the synthesis of new 5-arylidene-5H-thiazol-4-ones and new 5-arylidene-2-thioxo-1,3thiazolidin-4-ones. We also studied their effects on some protein kinases (DYRK1A, CK1, CDK5-p25, and GSK-3α/β) and on six representative tumor cell lines. Herein, we present the building of this thiazolone/thiazolidinone library, mostly carried out under microwave irradiation, and the biological activities of these compounds.

Synthesis of 5-Arylidene-2-thioxo-1,3-thiazolidin-4-one Library
In order to synthesize a library of 5-arylidene-2-thioxo-1,3-thiazolidin-4-ones 3 (Scheme 1) for a systematic structure-activity relationship (SAR) study, we have examined the reactivity of 2-thioxo-1,3-thiazolidin-4-ones 1a-i with a series of various aromatic aldehydes 2a-p. We selected commercial aldehydes substituted by phenol function at various positions (2f, 2g, 2i, and 2n) and/or methoxy groups (2d,e, 2h) and cyclic ether function (2a-c, 2l, 2o, 2p) to evaluate how the protic character and steric nature of the substituents on the phenyl ring will influence the biological activity of compounds 3. We also examined the bio-isosteric replacement of the benzene ring of 3 with pyridine (2k) or benzofuran (2m) moieties. Compounds 3 are designed to evaluate whether the NH function in position N-1 or a functional group with similar H-bond interactions will retain the activity.
For the 2-thioxo-1,3-thiazolidin-4-one platforms 1, rhodanine 1a, 2-(4-oxo-2-thioxothiazolidin-3-yl)acetic acid 1b and 3-amino rhodanine 1d are commercially available. Access to the 3-(4-oxo-2-thioxo-thiazolidin-3-yl)propanoic acid 1c could be accomplished by a recent patented protocol [45]. For introduction of the arylsulfonamide functionality on the rhodanine moiety in compound 1e, we used the Power's method [46]. Starting from commercial 3-amino-rhodanine 1d, compounds 1f-i were synthesized easily, according to a recent published procedure developed in our laboratory [47]. With the aim to develop a more efficient synthetic process for the synthesis of highly functionalized 5-arylidene rhodanines 3a-v from the building blocks 1a-i and 2a-p, microwave irradiation was employed to shorten the reaction times in comparison to conventional heating. For the reported methods, the Knoevenagel condensations have been realized in the presence of catalyst with various solvents: sodium acetate in glacial acetic acid [48] or in ethanol [46], piperidinium benzoate in toluene [49], tetrabutylammonium bromide in water [50], and piperidine in ethanol [51], using conventional heating in an oil bath. These older protocols suffer from one or more limitations, such as requiring harsh reaction conditions, low moderate yields, cumbersome experimental process, and long reaction time. Considering the latter aspect, we reasoned that microwave dielectric heating assisted chemistry might be an interesting approach for the synthesis of 5-arylidene rhodanine derivatives as potential biological molecules. For this study, we investigated the use of three microwave reactors: the Synthewave ® 402 and the Explorer ® 24 apparatus with continuous irradiation power from 0 to 300 W and the Monowave ® 300 with continuous power from 0 to 800 W. These three microwave irradiation reactors comprise a mono-mode microwave cavity that operates at a frequency of 2.45 GHz; the reaction temperature in the microwave cavity was controlled with a calibrated infrared sensor, and the microwave irradiation parameters (power and temperature) were monitored by a software package. For optimization of the experimental reaction conditions (selection of the appropriate reactor, reaction temperature and irradiation power), we have not realized a comparative study between these three reactors because we focused only our work on the microwave apparatus, which lead us to better results.
As can be seen in Table 1, the experiments revealed that the optimal reaction conditions for the 5-arylidene-2-thioxo-1,3-thiazolidin-4-ones 3a-l were obtained in an open vessel after 60 min at 80 • C (with a power of 90 W) in the Synthewave ® 402 apparatus with a stoichiometry of 1/1-4 of aryl aldehyde 2 and the volatile propylamine in order to be able to develop one-pot tandem reactions (aldimine synthesis/Knoevenagel condensation) [52]. According to this microwave method, the compounds 3a-l were synthesized in yields ranging from 62 to 89%. For the other compounds, 3m, 3n, and 3p-v, we screened a range of another reaction condition parameters in a closed vessel (a tube sealed with a snap cap) with the appropriate microwave reactors. With the conditions shown in Table 1, AcONa (0.1 or 1-1.2 eq) and 0.1 eq of piperidine in glacial acetic acid (0.1-6.6 eq) were identified to execute the Knoevenagel reactions in good yields (62-98% excepted for 3p). Unfortunately, we are not able to use this protocol for the preparation of compound 2o. It was obtained in moderate yield (2o: 26%) using classical heating in an oil bath. The structures of the desired 5-arylidene-2-thioxo-1,3-thiazolidin-4-ones 3a-v were substantiated by 1 H, 13 C NMR (Supplementary Materials), and HRMS analyses, and only the thermodynamically more stable Z-isomers were obtained [50]. Table 1. Results for the preparation under microwave irradiation of (5Z) 5-arylidene-2-thioxo-1,3-thiazolidin-4-one derivatives 3a-v from aromatic aldehydes 2a-p and 2-thioxo-1,3-thiazolidin-4-ones 1a-i. 2.1.2. Synthesis of (5Z) 2-Amino-5-arylidene-1,3-thiazolidin-4-one Library

Reactions Conditions Used under Microwave
With the compounds 3a-v in hand, we investigated the preparation of novel 2-amino-5-arylidene-1,3-thiazolidin-4-ones 5, according to Scheme 1. Transformation of 5-arylidenerhodanine derivatives 3 into their 2-amino-5-arylidene-1,3-thiazolidin-4-ones 5 involved a sulfur/nitrogen displacement in the presence of a primary amine [53] or a secondary amine [54]. Commonly, the synthesis of 2-amino-1,3-thiazol-4-one involved activation of the C=S bond of the starting 5-arylidene rhodanine, and reaction with an halogenoalcane (i.e., EtI or MeI) potentially gave the desired S-alkyl-5-arylidene-1,3-thiazol-4(5H)-one, together with the N-alkyl-5-arylidene rhodanine derivative, as a side product [55]. In this context, we preferred to directly investigate the sulfur/nitrogen displacement in a faster and more efficient way by using microwave dielectric heating for the synthesis of targeted compounds 5 from the intermediate 3 and a variety of polar cyclic secondary amines 4. In order to obtain interesting information for the structure-activity relationship (SAR) study on these compounds 5 as inhibitors of the protein kinase DYRK-1A, compounds 5 were designed to evaluate how the electrostatic and steric nature of the substituents on the phenyl ring and/or the presence of various amino moieties in position C-2 will influence the biological activity. For this study, we have employed eight polar cyclic secondary amines 4a-h as elements of the second point of molecular diversity for the construction of compounds 5. After a preliminary screening of the microwave irradiation conditions (open or closed vessel mode, temperature, power of irradiation, time), the experiments revealed the optimal ratio of 3/4 under microwave irradiation was obtained with 1.5 eq of amino reagent 5 using the "open vessel mode". For the other parameters (temperature: 80-100 • C, power: 80-150 W, time: 20-60 min), the results for each product 5 are presented in Table 2. To verify the versatility and efficiency of this open vessel microwave protocol, different starting 5-arylidene rhodanine derivatives 3 and commercially available cyclic secondary amines 4a-h were used as building blocks to generate a collection of seventeen new 2-amino-5-arylidene-1,3-thiazolidin-4-ones 5a-q: in all cases, the products were easily isolated in good to high yields (72-96%) after a simple precipitation from EtOH. The structure identification of all these new compounds 5 were based on the 1 H and 13 C assignments and was performed extensive 1D and 2D NMR spectroscopy. For the exocyclic double bond (CH=C) of all the 2-amino-5-arylidene-1,3-thiazolidin-4-ones 5, 1 H NMR spectra show only one signal for the methylene proton (CH=) in the range of 7.56-7.61 ppm at lower field values than those expected or previously reported in the literature for the E-isomers which, strongly indicates that the compounds 5a-q have the Z-configuration because of the high degree of thermodynamic stability of this isomer [56]. Table 2. Results for the solvent-free synthesis under microwave irradiation of (5Z) 2-amino-5-arylidene-1,3-thiazol-4(5H)-one derivatives 5a-q by sulfur/nitrogen displacement reaction from (5Z) 5-arylidene-2-thioxo-1,3-thiazolidin-4-one derivatives 3 and cyclic secondary amines 4a-h.  To extend the molecular diversity on the C-2 position of the rhodanine platform, we have also examined introduction of aromatic primary amine (i.e., aniline, C-substituted aniline, amino pyridine, etc.) using the microwave dielectric heating protocol for sulfur/nitrogen displacement from 5-arylidene-2-thioxo-1,3-thiazolidin-4-one 3 and appropriate aromatic primary amine. After screening of the microwave irradiation conditions based on the reaction temperature (90-160 • C), the power for irradiation (90-150 W), the reaction time (30-60 min), the use of polar or non-polar solvent (EtOH, hexane), the use of open or closed vessel mode, and the ratio of the reagents (ratio 3/aromatic amine: 1-3.5), it was not possible to obtain the desired 2-N-arylamino-5-arylidene-1,3-thiazolidin-4-one 5 after analysis of the crude reaction mixture by 1 H NMR. Accordingly, we decided to drop this synthetic strategy for access to 2-N-arylamino-5-arylidene-1,3-thiazolidin-4-one and to focus now on the use of 2-N-heteroarylamino-1,3-thiazolidin-4-ones 8a-e in Knoevenagel condensation (Scheme 2). The designed compounds 8a-e were easily prepared from N-aryl thiourea 6a-d and ethyl bromoacetate according to a protocol developed in our laboratory [55]. As shown in Table 3, the "closed vessel" microwave irradiation procedure involved the use of piperidine (0.1 eq) or AcONa (0.1-1 eq) in glacial acetic acid for the synthesis of 2-N-heteroarylamino-5-arylidene-1,3-thiazolidin-4-one 5r-x from the building blocks 8a-e and aromatic aldehydes 2a-c after 20-40 min at 120 or 150 • C with a power of 100-200 W. This protocol is also very simple to execute because the desired products 5r-x were isolated after crystallization from ethanol in moderate to good yields (34-90%). The expected Z-geometry of these compounds 5 was confirmed by the chemical shift of the exocyclic methylene proton (CH=) obtained in 1 H NMR [56]. Scheme 2. General synthetic approach for the (5Z) 2-N-heteroarylamino-5-arylidene-1,3-thiazol-4(5H)-one derivatives 5r-x from the starting 2-N-heteroarylamino-1,3-thiazolidin-4-one 8a-e and aromatic aldehydes 3(a-c). Reagents and reaction conditions: (i) 7 1.3 eq, EtOH, 55 • C, 3.5-22 h; (ii) piperidine 0.1 eq or AcONa 0.1 eq, AcOH 0.1-6.6 eq, 120-150 • C, MWI in closed vessel (Explorer 24 ® CEM µWave reactor), 20-40 min.
Comparison of all the results obtained in the two series showed the difficulties to define any role for the various substituents on the exocyclic phenyl ring (arylidene moiety in C-5 position), as well on the amino substituents in N-2 position, in order to obtain sub-micromolar affinity for DYRK1A with good selectivity. However, it is interesting to note that the presence of hydroxyl group in para-position of the phenyl ring of 5-arylidene-2-thioxo-1,3-thiazolidin-4-one derivative 3c is very important to access of a sub-micromolar inhibitory activity for DYRK1A. However, the presence of the bulky 1,3-benzodioxol-5yl group on the 5-arylidene moiety is not a crippling factor when it is associated with a thiomorpholin-1-yl substituent in N-2 position, with is the case for the compound 5n (IC 50 0.070 µM). Table 5. Effects of the (5Z) 2-amino-5-arylidene-1,3-thiazol-4(5H)-one derivatives 5a-x on the catalytic activity of four purified protein kinases a .
2 130 a Compounds were tested at various concentrations on each kinase as described in Experimental Section. IC 50 values, calculated from the dose-response curves, are reported in µM. -, inactive at the highest concentration tested (10 µM); >10, inhibitory, but IC 50 > 10 µM.

Compound
Cell Lines IC 50 (µM) a The other derivatives 5a-q have not been tested, or some of them are not soluble in the conditions for the preparation of the biological test. b The other derivatives 3d,e and 3h-v have not been tested, or some of them are not soluble in the conditions for the preparation of the biological test.

Chemistry-General Remarks
Melting points were determined on a Kofler melting point apparatus and were uncorrected. 1 H NMR spectra were recorded on BRUKER AC 300 P (300 MHz) spectrometer, 13 C NMR spectra on BRUKER AC 300 P (75 MHz) spectrometer. Chemical shifts are expressed in parts per million downfield from tetramethylsilane as an internal standard. Data are given in the following order: δ value, multiplicity (s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; br, broad), and number of protons, and coupling constants J is given in Hertz. The mass spectra (HRMS) were taken, respectively, on a MS/MS ZABSpec Tof Micromass (EBE TOF geometry) at an ionizing potential of 8 eV and on a VARIAN MAT 311 at an ionizing potential of 70 eV in the "Centre Régional de Mesures Physiques de l'Ouest" (CRMPO, Rennes, France). Reactions under microwave irradiations were realized in the Synthewave ® 402 apparatus (Merck Eurolab, Div. Prolabo, France) or in the Explorer ® 24 CEM microwave reactor (CEM France, Saclay, France), as well as in the Anton Paar Monowave 300 ® microwave reactor (Anton Paar France, Les Ulis, France). Microwave irradiation reactions were realized in open cylindrical quartz reactor (Ø = 1.8 cm) with the Synthewave ® 402 apparatus or in borosilicate glass vials of 10 mL equipped with snap caps (at the end of the irradiation, cooling reaction was realized by compressed air) with the Explorer ® 24 or Monowave ® 300 reactors. The microwave instrument consists of a continuous focused microwave power output from 0 to 300W for the Synthewave ® 402 or the Explorer ® 24 CEM apparatus and from 0 to 800W for the Anton Paar Monowave 300 ® apparatus. All the experiments in each microwave reactor were performed using the stirring option. The target temperature was reached with a ramp of 2-5 min, and the chosen microwave power stays constant to hold the mixture at this temperature. The reaction temperature is monitored using calibrated infrared sensor and the reaction time included the ramp period. The microwave irradiation parameters (power and temperature) were monitored by the ChemDriver software package for the Explorer ® 24 CEM apparatus and by the Monowave software package for the Anton Paar Monowave 300 ® reactor. Solvents were evaporated with a BUCHI rotary evaporator. All reagents and solvents were purchased from Acros (Geel, Belgium), Sigma-Aldrich Chimie (Saint-Quentin-Fallavier, France), and Fluka France and were used without further purification. In a cylindrical quartz reactor (Ø = 1.8 cm), successively commercial rhodanine 1a (1 eq), aromatic aldehyde 2 (1 eq), and n-propylamine (2 eq) were placed. The reactor was then introduced into the Synthewave ® 402 Prolabo microwave cavity (power: 300 W). The stirred mixture was irradiated at 80 • C (after a ramp of 2 min. from 20 to 80 • C) for 60 min (power level: 30%, 90 W). After microwave dielectric heating, the crude reaction mixture was allowed to cooling down at room temperature, and ethanol (20 mL) was added in the cylindrical quartz reactor. The resulting insoluble product 3 was filtered, recrystallized in EtOH, and dried under high vacuum (10 −2 Torr) for 1 h.     7-Bromo-1,3-benzodioxole-5-carbaldehyde (2l). In a 25 mL round-bottomed flask provided with a magnetic stirrer and condenser, a mixture of commercial 3-bromo-4,5-dihydroxybenz aldehyde (217 mg, 1 mmol), potassium fluoride (529 mg, 9.1 mmol, 9.1 eq), and dibromomethane (70 mL, 173 mg, 1 mmol) in 3 mL of dimethylformamide was stirred vigorously under a stream of nitrogen for 4 h at 140 • C in an oil bath. After cooling down to room temperature, 3 mL of deionized water was added in one portion to the crude reaction mixture. The resulting solution was submitted to extraction with Et 2 O (3 × 5 mL), and, after decantation, the organic layer was dried over MgSO 4 . After filtration, the solvent of the filtrate was eliminated in a rotary evaporator under reduced pressure; then, the crude residue was dried under high vacuum (10 −2 Torr) at 25 • C for 2 h. The desired 7-bromo-1,3-benzodioxole-5-carbaldehyde 2l was obtained as a light-yellow powder in 51% yield (117 mg) and was sufficiently pure to be used further without purification; mp = 123-125 • C. 1  In a 10 mL glass tube were placed successively commercial rhodanine 1a (1 eq), aromatic aldehyde 2o,p (1.1-1.2 eq), sodium acetate (26.4 mg, 0.31 mmol, 0.1 eq), and glacial acetic acid (1.2 eq). The glass tube was sealed with a snap cap and placed in the Monowave 300 ® Anton-Paar microwave cavity (power = 850 W). The mixture was irradiated at 140 • C for 20-30 min under vigorous magnetic stirring. After microwave dielectric heating, the crude reaction mixture was allowed to cool down at room temperature. To this crude mixture, 4 mL of deionized water was added, and the resulting suspension was submitted to ultrasound in a Branson 1510 apparatus at 25 • C for 30 min. Then, the desired compound 3 was collected by filtration, and the crude solid was washed with 5 mL of deionized water and dried under high vacuum (10 −2 Torr) at 25 • C for 2 h. In a 10 mL glass tube were placed successively the 3-N-amino rhodanine derivative 1 (1 eq), piperonaldehyde 2a (1 eq), piperidine (0.1 eq) or sodium acetate (0.1 eq), and glacial acetic acid (0.1-6.6 eq). The glass tube was placed in the Explorer ® 24 CEM microwave cavity (power = 300 W). The mixture was irradiated at 120 or 150 • C (with a power of 100 or 200 W) for 20 min under vigorous magnetic stirring. After microwave dielectric heating, the insoluble compound 3 was separated from the crude reaction mixture by filtration. To the collected crude compound 3, absolute ethanol or deionized water was added, and the resulting suspension was stirred vigorously under magnetic stirring for 18 h. The desired compound 3 was collected by filtration and was dried under high vacuum (10 −2 Torr) at 25 • C for 1 h.

Conclusions
All the data presented in this study demonstrate the interest and the potential of the 1,3-thiazolidin-4-one core in inhibition of the protein kinase DYRK1A. In this preliminary project of pharmacology, we have developed two series of new compounds under microwave irradiation. In the first library, the twenty-two 5-arylidene-2-thioxo-1,3thiazolidin-4-ones 3a-v were synthesized in yields ranging from 26 to 98% by microwave irradiation assisted Knoevenagel condensation using an "open or closed vessel" approach using organic bases (propylamine, TEA, and AcONa). Next, most of the compounds 3 were transformed into 2-amino-5-arylidene-1,3-thiazol-4(5H)-ones 5a-q under microwave irradiation by sulphur/nitrogen displacement with appropriate secondary cyclic amines 4a-h. This second family was completed by the preparation of seven new 2-N-heteroarylamino-5arylidene-1,3-thiazol-4(5H)-ones 5r-x from the 2-N-heteroarylamino-1,3-thiazolidin-4-one building blocks 8a-e under microwave irradiation (34-98% yield). In the two libraries, all the 46 compounds 3 and 5 have been built with a Z-geometry and were evaluated against four protein kinases. Among all these compounds, nine of them turned out to be very interesting because they presented sub-micromolar inhibition activity on DYRK1A (0.090 µM > IC 50 > 0.028 µM). The most effective compounds in these two libraries are the molecules 3e and 5s, as new nano-molar DYRK1A inhibitors (3e: IC 50 0.028 µM and 5s: IC 50 0.033 µM). The current results are the starting point of a new larger program within our group to investigate the biological properties of these new inhibitors with potential application in Alzheimer's disease or in cancer.