Synthesis of novel series of 3,5-disubstituted imidazo[1,2-d] [1,2,4]thiadiazoles involving SNAr and Suzuki–Miyaura cross-coupling reactions

The first access to 3,5-disubstituted imidazo[1,2-d][1,2,4]thiadiazole derivatives is reported. The series were generated from 2-mercaptoimidazole, which afforded the key intermediate bearing two functional positions. The SNAr reactivity toward tosyl release at the C-3 position was investigated and a regioselective electrophilic iodination in C-5 position was performed to allow a novel C–C bond using Suzuki–Miyaura reaction. Palladium-catalyzed cross-coupling conditions were optimized. A representative library of various boronic acids was employed to establish the scope and limitations of the method. To complete this methodological study, the influence of the nature of the C-3 imidazo[1,2-d][1,2,4]thiadiazole substitutions on the arylation in C-5 was investigated.


Introduction
For the past few decades, sulfur-containing [5,5] fused ring systems with a bridgehead nitrogen have received considerable attention in the drug discovery eld due to their interesting biological activities. [1][2][3][4][5][6] For example, structures containing them have been reported in various therapeutic anticancer 7-9 and antitubercular 10 agents, and for cardiovascular treatments. 11 Moreover, other representative molecules have demonstrated their potential in the treatment of neurodegenerative disorders. 12 For these reasons, this heterocyclic family plays an increasingly important role in exploring uncovered regions of chemical space for the discovery of new biologically active drugs.
However, this exploration remains underdeveloped when we consider the sub-family bearing a sulfur-nitrogen bond. [13][14][15][16][17][18][19] The main reasons are the lack of knowledge about their formation, reactivity or how to successfully position the desired substituents step by step. There has therefore been tremendous interest in overcoming this major hindrance in order to increase the molecular diversity around these series and develop highly original cores for the design of future original bioactive molecules.

Results and discussion
First, we focused our attention on the tosyl platform 4, which can be prepared by using three short steps from commercially available 2-mercaptoimidazole. 27 The condensation of 1 and nbutyl isocyanate led to amide 2 in a near quantitative yield. The oxidative ring closure in presence of bromine and triethylamine afforded the [5,5] fused bicyclic heterocycle 3 in an excellent 93% yield. To nish, a ring-opening/ring-closure sequence in presence of tosyl cyanide led to 4 in 88% yield (Scheme 1).
To tackle the usefulness of 4 as a building block and taking advantage of the tosyl as leaving group, we began by the C-3 functionalization using a S N Ar reaction. Using the Leung-Toum conditions, 27 derivative 5 was synthesized with n-propylamine as nucleophile and Et 3 N as base, in toluene at r.t. aer 4 hours in a 91% yield (Table 1, entry 1). Aer the successfully accomplished condensation of a primary aliphatic amine with 4, we next explored the scope of this method by treating the versatile platform 4 with other types of amines. The use of cycloalkylamines such as cycloproyl-or cyclohexyl-amine decreased the yield to 65% (entries 3 and 4). With N-methylpropylamine, the efficiency of the reaction was maintained (entry 2 vs. 1) while with other secondary cyclic amines such as piperidine or N-methylpiperazine, the efficiency signicantly diminished (entries 5, 6). Fortunately, when morpholine was condensed with 4, the S N Ar reaction led to compound 11 in an excellent 94% yield (entry 7). Interestingly, we obtained a yield of 74% with the less nucleophilic benzylamine (entry 8) but no reaction was observed with aniline (entry 9). To achieve this investigation, we switched to alkoxides as nucleophiles (entries 10, 11) and all the attempted compounds were isolated in near quantitative yields.
Selective halogenation in C-5 position with N-bromo or Niodosuccinimide in DMF at r.t. was performed and showed a better reactivity for the introduction of iodine atom ( Table 2 entries 22, 24 versus 26 and 27). The scope of the reaction was studied with compounds 5-11 to afford derivatives 16-27 (Table  2) without any signicant problems as iodo derivatives were mainly isolated in satisfying yields, except in the case of 18 and 21 for which the purication was more problematic.
With these compounds in hand, we then achieved the iodine displacement by Suzuki-Miyaura cross coupling as no C-C bond formation in C-5 position appears to be currently described on this skeleton. This prompted us to propose to the community a general and efficient catalytic system by optimizing the main reaction parameters (Table 3). First, we used 22 as starting material, Pd(PPh 3 ) 4 as the palladium source, Cs 2 CO 3 as base, and dioxane as solvent under microwave irradiation during 1 h. With these conditions, the desired product 28 was isolated in a low but encouraging yield (33%, Table 3, entry 1). When the catalyst was switched for PdCl 2 (dppf).DCM, the reactivity was improved and the desired compound 28 was obtained in 47% yield. In the following experiment, we catalyzed the reaction with a bidentate palladium complex, which was formed by using a mixture of Pd(OAc) 2 (10 mol%) and Xantphos (20 mol%). The reaction was achieved in only 1 h and  product 28 was isolated in a satisfying yield of 60% (Table 3, entry 3). Changing the nature of the base indicated that the use of K 2 CO 3 did not affect the reaction yield whereas K 3 PO 4 partially reduced the reactivity. Next, the scope and potential limitations of the Pd-coupling step were investigated by modulation of the boron derivatives ( Table 4). The use of electron-rich or neutral phenyl boronic acids was well tolerated and furnished the derivatives 29 and 30 in good yields (entries 2 and 3). In contrast, the presence of electron-withdrawing substituents such as nitro or uorine slightly decreased the efficiency of the reaction and compounds 34 and 35 were isolated in 40% and 38% yields, respectively. Next, we investigated the inuence of steric hindrance using the methoxy position switch on the phenyl ring. While the ortho orientation induced a dramatic decrease in yield (34% versus 65% for 28), the meta orientation led to the desired compound in a 50% yield (entries 3-5). The only identied limit concerned the presence of labile hydrogens such as OH or NH, which totally inhibited the reaction (entries 6, 10 and 11). This constraint was easily removed by the use of a protective group such as THP for the phenol derivative (entry 6) and an aryl entity was successfully introduced in a good overall yield of 59% aer a tandem sequence including the cross coupling reaction and the in situ deprotection. Finally, the introduction of electronrich heterocycle was studied with thiophene-3-boronic acid, and the desired product 36 was isolated in a good 71% yield.
To complete this Suzuki-Miyaura study, we then evaluated the inuence of the nature of the substituent in C-3 position  PdCl 2 (dppf)$DCM (0.1 eq.) Cs 2 CO 3 , (2.0 eq.) 47% 3 Pd(OAc) 2 (0.1 eq.), xantphos (0.2 eq.) Cs 2 CO 3 , (2.0 eq.) 60% 4 Pd(OAc) 2 (0.1 eq.), xantphos (0.2 eq.) K 2 CO 3 , (2.0 eq.) 60% 5 Pd(OAc) 2 (0.1 eq.), xantphos (0.2 eq.) K 3 PO 4 , (2.0 eq.) 36% a Yield is indicated as isolated product.  (Table 5) and selected p-tolylboronic acid as the sole arylation partner. As previously described, the presence of a hydrogen on the C-5 nitrogen atom totally inhibited the catalytic cycle (entries 1, 3, 4 and 8) whereas its substitution restored the efficiency of the reaction (entry 2 vs. 1). In fact, whatever the nature of the tertiary amine (i.e. aliphatic or cyclic) in C-3 position, the C-N bond was efficiently generated and products were isolated in fairly good yields ranging from 56% to 71% (Table 5, entries 2, 5-7). To nish, we performed the cross coupling reaction in the presence of an alkyloxy in the C-3 position and proved that the nal compound could be obtained in moderate yield (entries 9, 10) suggesting that electronic enrichment of the heterocycle played a major role in the result. Heck (with methyl acrylate) or Buchwald (with aniline) cross-coupling reactions are performed with this catalytic system and only 10% of conversion were obtained. This limitation prompted us to identify, in the future, a new catalytic system able to remove this limitation.