Control of 6-Exo and 7-Endo Cyclizations of Alkynylamides using Platinum and Bismuth Catalysts
Control of 6-Exo and 7-Endo Cyclizations of Alkynylamides using Platinum and Bismuth Catalysts
复制标题
DOI：
10.1002/asia.201100209
复制
发表时间：
2011-06-06
影响因子：
4.1
通讯作者：
Takemoto, Yoshiji
中科院分区：
文献类型：
--
作者：
Girard, Anne-Lise;Enomoto, Taro;Takemoto, Yoshiji
哌嗪-2-酮和1，4-二氮杂环庚烷-2-酮支架在生物活性天然产物和药物化合物中具有特色，[1]例如肽模拟物，[2]具有逆转癌症治疗中的多药耐药性潜力的化合物（Ardeemins），[3]抗生素（caprazamicins），[4]镇静剂和安定剂（地西泮）[5]。因此，这些结构作为药物发现领域的重要药效团吸引了相当大的关注。通过内酰胺化、Mitsunobu反应、还原胺化和Ugi反应，已经开发了多种合成方法用于构建这些N-杂环结构。[6，7]虽然哌嗪-2-酮与1，4-二氮杂环庚烷-2-酮是同源的，但是迄今为止还没有关于从同一化合物合成两种基序的报道。为了建立一个简单而统一的方法来制备这些结构，我们研究了刘易斯酸催化的2-（丙-2-炔基氨基）-乙酰胺的杂环化反应。虽然有许多关于这种前体的5-exo-dig/6-endo-dig环化的报道[8]，但相对较少报道竞争性6-exo-dig/7-endo-dig环化的控制。[9]据我们所知，在控制非对映选择性的情况下，末端炔的氢酰胺化从未被描述用于合成1，4-二氮杂环庚烷-2-酮和哌嗪-2-酮。[10]钯和金催化取代炔酰胺的环化反应已被本课题组和其他几个研究组报道。[11然而，这些条件以前没有应用于末端炔。本文报道了2-（丙-2-炔基氨基）乙酰胺分子内氢酰胺化反应高选择性合成1，4-二氮杂环庚烷-2-酮或哌嗪-2-酮的研究进展，其中区域选择性取决于所用的刘易斯酸催化剂。首先，我们研究了N-Ns炔酰胺（Ns=邻硝基苯磺酰基（nosyl））1a在不同催化体系中在1，2-二氯乙烷（DCE）中在708 ℃和室温下的环化反应（表1）。催化剂筛选表明，根据所用的刘易斯酸，6-外-二氢N-环化产物2a和7-内-二氢N-环化产物3a以不同的比例产生。[13]令人惊讶的是，当使用PtCl 2作为催化剂时，[14]以67%的产率获得3a，仅产生痕量的2a（表1，条目1）。相反，使用PtCl 4的反应得到2a作为主要产物（条目2）。测试的其他铂催化体系在促进环化方面无效（条目3和4）。当使用阳离子金催化剂[15]时，3a也是主要产物，但与PtCl 2相比，获得较低的区域选择性（条目5）。通过使用AgNTf 2作为（Tf= CF 3SO 2）催化剂导致低反应性和选择性，并且观察到副产物（条目6）。[PdACHTUNGTRENNUNG（PhCN）2Cl 2][16]在室温下不促进向环化产物的转化，并且在加热时仅观察到降解产物（条目7）。当使用根据硬和软酸碱（HSAB）原理[17]被认为是边界的金属时，例如CuI、CuACHTUNGTRENNUNG（OTf）2或InACHTUNGTRENNUNG（OTf）3，起始材料的回收率超过80%（表1，条目8-10）。与此形成鲜明对比的是，与BiACHTUNGTRENNUNG（OTf）3 [18]的反应在室温下缓慢进行，并以75%的产率得到作为单一产物的6-外环化产物2a [a] Dr.A. - L.吉拉德，T. Enomoto，S.横内博士Tsukano，Prof. Dr. Y.竹本
Piperazin-2-one and 1, 4-diazepan-2-one scaffolds are featured in biologically active natural products and pharmaceutical compounds,[1] such as peptidomimetics,[2] compounds with potential to reverse multi-drug resistance in cancer treatments (ardeemins),[3] antibiotics (caprazamicins),[4] sedatives, and tranquilizers (diazepam)[5]. These structures have therefore attracted considerable attention as important pharmacophores in the field of drug discovery. Diverse synthetic methods have been developed for the construction of these N-heterocyclic structures, by means of lactamization, the Mitsunobu reaction, reductive amination, and Ugi reactions.[6, 7] While piperazin-2-ones are homologous with 1, 4-diazepan-2-ones, however, to date no reports on the synthesis of both motifs from the same compound have been reported. To establish a simple and unified method to generate these constructs, we investigated the Lewis acid catalyzed heteroannulation of 2-(prop-2-ynylamino)-acetamide. While there are a number of reports on the 5-exo-dig/6-endo-dig cyclization of this precursor,[8] there are relatively few reporting the control of the competitive 6-exo-dig/7-endo-dig cyclization.[9] To our knowledge, the hydroamidation of terminal alkynes has never been described for the synthesis of 1, 4-diazepan-2-one and piperazin-2-one with control of diastereoselectivity.[10] The palladium-and gold-catalyzed cyclization of substituted alkynylamides has been reported by our research group and several other groups.[11, 12] However, these conditions have not previously been applied to terminal alkynes. Herein, we report the development of highly selective syntheses of 1, 4-diazepan-2-ones or piperazin-2-ones by an intramolecular hydroamidation of 2-(prop-2-ynylamino) acetamides, in which regioselectivity is dependent on the Lewis acid catalyst employed. Initially, we investigated the cyclization of N-Ns alkynylamide (Ns= o-nitrobenzenesulfonyl (nosyl)) 1a with different catalytic systems in 1, 2-dichloroethane (DCE) at 708C and at room temperature (Table1). A catalyst screening revealed that the 6-exo-dig N-cyclized product 2a and the 7-endo-dig N-cyclized product 3a were produced in different ratios depending on the Lewis acids used.[13] Surprisingly, when PtCl2 was used as a catalyst,[14] 3a was obtained in 67% yield with only trace quantities of 2a generated (Table 1, entry 1). In contrast, the reaction using PtCl4 gave 2a as the major product (entry 2). Other platinum catalytic systems tested were not effective at promoting cyclization (entries 3 and 4). When a cationic gold catalyst [15] was employed, 3a was also the major product but lower regioselectivity was obtained compared to that of PtCl2 (entry 5). By using AgNTf2 as a (Tf= CF3SO2) catalyst resulted in low reactivity and selectivity, and by-products were observed (entry 6).[PdACHTUNGTRENNUNG (PhCN) 2Cl2][16] did not promote conversion to the cyclized products at room temperature and only degradation products were observed upon heating (entry 7). When metals considered to be borderline according to the hard and soft acids and bases (HSAB) principle,[17] such as CuI, CuACHTUNGTRENNUNG (OTf) 2, or InACHTUNGTRENNUNG (OTf) 3 were employed, the starting material was recovered in over 80% yield (Table 1, entries 8–10). In sharp contrast, the reaction with BiACHTUNGTRENNUNG (OTf) 3 [18] proceeded slowly at room temperature, and afforded the 6-exocyclized product 2a as a single product in 75% yield [a] Dr. A.-L. Girard, T. Enomoto, S. Yokouchi, Dr. C. Tsukano, Prof. Dr. Y. Takemoto