Stereocontrolled total synthesis of (+/-)-gelsemine

Stereocontrolled total synthesis of (+/-)-gelsemine
复制标题

DOI:
10.1021/ja961701s
复制
发表时间:
1996-08-07
影响因子:
15
通讯作者:
Liu, G
Liu, G
中科院分区:
化学1区
文献类型:
--
作者:
Fukuyama, T;Liu, G

文献摘要

被引文献

相似文献

钩吻碱(1)一直被认为是钩吻(卡罗莱纳茉莉)的主要生物碱成分。2自从1959年确定钩吻碱的结构以来,3由于其独特的六环笼状结构,吸引了众多的合成努力。[4]虽然1994年有三个研究小组报道了通过其次要同系物21-氧代钩吻碱(2)全合成(()-钩吻碱(1)),但他们都没有成功地控制关键的螺吲哚啉酮系统的立体化学。5本文报道了(()-钩吻素(1)的立体控制全合成,其特征在于双环[3.2. 1]通过二乙烯基环丙烷-环庚二烯重排的方式得到骨架。6我们的合成开始于根据Kondo的方案制备必需的中间体3。因此,将衍生自乙酰乙酸甲酯的二价阴离子加入山梨醛中,然后立即保护不稳定的醇,得到乙氧基乙基醚4(方案1)。β-酮酯4在标准条件下的重氮转移反应得到重氮化合物5,将其进行铜介导的环丙烷化得到双环酮6。用硼氢化钠还原酮6,所得醇的乙酰化,乙氧基乙基醚的水解,和随后的烯烃的臭氧分解提供醛3。醛3和羟吲哚的Knoevenagel缩合得到E-和Z-异构体7和8的4:1混合物(方案2)。尝试将E-异构体光化学异构化为所需的Z-异构体,最多得到1:1的混合物。为了进一步使产物分布偏向,我们决定在羟吲哚的4-位上引入一个大体积的取代基。如预期的,4-碘羟吲哚9与醛3的8缩合以89%的产率提供(Z)-亚烷基吲哚酮9作为唯一产物。醇9的Pfitzner-Moffatt氧化10,然后消除乙酸,得到不稳定的烯酮10。当在90 ℃加热时,化合物10经历异常平稳的重排,得到所需的双环[3.2]。1]系统11 a以98%产率作为高度结晶固体。螺环中心的立体化学通过从相同合成途径获得的相应溴化物11b的单晶X-射线分析来证实。随后的自由基脱碘提供了关键中间体12。与临界双环[3.2。1]框架在手,然后我们把注意力转向剩余的吡咯烷和四氢吡喃环的结构。由于酮和12的R,β-不饱和酯对亲核试剂具有相似的反应性,因此酮的选择性延伸被证明是相当困难的。幸运的是,用(EtO)2 POCH-
Gelsemine (1) has long been known as the major alkaloid component of Gelsemium semperVirens (Carolina jasmine). 2 Since the structure of gelsemine was determined in 1959, 3 it has attracted numerous synthetic efforts due to its unique hexacyclic cage structure. 4 While three groups reported total syntheses of (()-gelsemine (1) in 1994 via its minor congener 21-oxogelsemine (2), none of them have succeeded in controlling the stereochemistry of the critical spiroindolinone system. 5 Herein we report a stereocontrolled total synthesis of (()-gelsemine (1), which features a stereoselective construction of the bicyclo [3.2. 1] framework by means of a divinylcyclopropane-cycloheptadiene rearrangement. 6 Our synthesis started with the preparation of the requisite intermediate 3 according to the protocol of Kondo. 7 Thus, addition of the dianion derived from methyl acetoacetate to sorbic aldehyde followed by immediate protection of the unstable alcohol gave ethoxyethyl ether 4 (Scheme 1). Diazo transfer reaction of the β-keto ester 4 under standard conditions furnished diazo compound 5, which was subjected to coppermediated cyclopropanation to give the bicyclic ketone 6. Reduction of ketone 6 with sodium borohydride, acetylation of the resultant alcohol, hydrolysis of the ethoxyethyl ether, and subsequent ozonolysis of the olefin furnished the aldehyde 3. Knoevenagel condensation of aldehyde 3 and oxindole gave a 4: 1 mixture of E-and Z-isomers 7 and 8 (Scheme 2). Attempted photochemical isomerization of the E-isomer to the desired Z-isomer gave a 1: 1 mixture at best. In an effort to further bias the product distribution, we decided to introduce a bulky substituent to the 4-position of the oxindole. As expected, 8 condensation of 4-iodooxindole9 with aldehyde 3 furnished (Z)-alkylidene indolinone 9 in 89% yield as the exclusive product. Pfitzner-Moffatt oxidation10 of alcohol 9 followed by elimination of acetic acid furnished the unstable enone 10. 11 When heated at 90 C, compound 10 underwent an exceptionally smooth rearrangement to give the desired bicyclo [3.2. 1] system 11a in 98% yield as a highly crystalline solid. The stereochemistry of the spiro center was confirmed by a single-crystal X-ray analysis of the corresponding bromide 11b obtained from the same synthetic pathway. The subsequent radical deiodination provided the key intermediate 12. With the critical bicyclo [3.2. 1] framework in hand, we then turned our attention to the construction of the remaining pyrrolidine and tetrahydropyran rings. Since the ketone and the R, β-unsaturated ester of 12 have similar reactivities toward nucleophiles, selective elongation of the ketone proved to be quite difficult. Fortunately, treatment of 12 with (EtO) 2POCH-