A strategy for macrocyclic ring closure and functionalization aimed toward split-pool syntheses
A strategy for macrocyclic ring closure and functionalization aimed toward split-pool syntheses
复制标题
DOI：
10.1021/ja992658m
复制
发表时间：
1999-11-17
影响因子：
15
通讯作者：
Schreiber, SL
中科院分区：
文献类型：
--
作者：
Lee, D;Sello, JK;Schreiber, SL
具有生物活性的天然产物往往含有中环或大环，红霉素就是一个著名的例子。在这类化合物的合成中已经使用了许多类型的闭环反应。通常，通过改变非环前体的保护基团、立体化学和氧化状态以及通过改变反应条件来优化这些反应的效率。1b当裂池策略2应用于同时合成数百万个含有大环结构的天然产物类化合物3、4时，这些特别的解决方案是不实用的。同样，将大环立体控制的原理应用于大量构象不同的化合物是不切实际的。然而，实现这样的合成可能会产生许多新的生物活性化合物，从而促进化学遗传学，即使用小分子取代突变的类似遗传的筛选被用来探索生物过程。5我们已经开始了一项旨在立体化学和结构多样化化合物的分池合成的努力。作为迈向这一目标的第一步，我们报道了设计的无环化合物的溶液和固相反应，以及可能适用于集体进行的合成的所得大环的反应。导致六元环的闭环反应通常以高有效摩尔进行，这是因为低能构象将反应末端放置在非常接近的位置，并且具有适合键形成的取向。通过将结构元素引入到大环环的非环前驱体中，同样可以得到有利于环闭合的构象，也可以获得高有效摩尔分数。这一策略的一个例子如方案1所示。通过将平面的sp2杂化的具有反式几何构型的两原子(环内)元素插入完全sp3杂化(饱和)的六元环的交替环键中，应该可以得到没有跨环应变和扭转应变的六角形12元环产物。4，7为了保持这种理想的低能构象，必须考虑含取代基环碳的立体化学，以避免烯丙基(A1，3)应变。这一概念存在许多排列，并已成功地应用于结构复杂的10元环和14元环的合成(DL和SLS的未发表结果)。使用方案1中的特定策略，我们利用SP3-SP3碳-碳键7和亚胺(DL和SLS的未发表结果)成环反应，以极高的产率合成了具有六角形的12元环。鉴于烯烃歧化反应被广泛用作影响和研究大环化反应的一种手段，我们选择了闭环歧化反应(RCM)进行研究。尽管RCM反应8以不同的产率进行，但对这些反应的系统研究揭示了支持上述基于构象的分析的启发性趋势。这些结果如下所述。通过1，2-氨基醇1a-j与4-戊烯酸或其2-取代衍生物2a-c(EDC，DMAP，CH2Cl2)同时或顺序的酰化反应(方案2)制备了闭环底物3a-q(方案2)，除了由酒石酸亚甲基缩醛和3-丁烯-1-醇制备的前体5l外。3a-S的RCM反应是在格拉布斯的催化剂(4)(5-15摩尔%)的存在下进行的，在0.003-0.008 M的甲烷中回流24-50 h，以中等到极好的产率与不同数量的24元环二聚体一起合成了5a-S(图1)。10%…
Biologically active natural products frequently contain medium or large ringsserythromycin is a famous example. Many types of ring-closing reactions have been used in syntheses of such compounds. 1a Typically, these reactions are optimized for efficiency by varying protecting groups, stereochemistry, and oxidation states of acyclic precursors and by varying reaction conditions. 1b These ad hoc solutions are not practical when the split-pool strategy2 is applied to simultaneous syntheses of millions of natural product-like compounds3, 4 containing macrocyclic structures. Likewise, applying the principles of macrocyclic stereocontrol to large collections of conformationally heterogeneous compounds is impractical. Achieving such syntheses, however, will likely result in many new biologically active compounds and will thus advance chemical genetics, where genetic-like screens using small molecules in place of mutations are used to explore biological processes. 5 We have initiated an effort aimed at split-pool syntheses of stereochemically and structurally diverse compounds. As a first step toward this goal, we report solution-and solid-phase reactions of designed acyclic compounds and of the resulting macrocycles that may be applicable to syntheses performed en masse. Ring-closing reactions leading to six-membered rings often proceed with high effective molarities as a result of low-energy conformations that place the reacting termini in close proximity and with orientations suitable for bond formation. By introducing structural elements into acyclic precursors of macrocyclic rings that similarly result in conformations favorable to ring closure, high effective molarities may also be achieved. 6 One illustration of this strategy leading to 12-membered rings is shown in Scheme 1. By inserting planar, sp2-hybridized two-atom (intra-annular) elements having trans geometry into alternating ring bonds of a fully sp3-hybridized (saturated) six-membered ring, a hexagonally shaped 12-membered ring product free of trans-annular and torsional strain should result. 4, 7 To maintain this idealized lowenergy conformation, consideration must be given to the stereochemistry at ring carbons bearing substituents so as to avoid, among others, allylic (A1, 3) strain. Many permutations of this concept exist, and it has been applied successfully to the synthesis of structurally complex 10-and 14-membered rings (unpublished results of DL and SLS). Using the specific strategy in Scheme1, we have synthesized 12-membered-rings with hexagonal shapes in excellent yields using both sp3-sp3 carbon-carbon bondforming7 and imine (unpublished results of DL and SLS) bondforming ring-closure reactions. In light of the extensive use of the olefin metathesis reaction as a means to effect8 and study9 macrocylization reactions, we elected to use the ring-closing metathesis (RCM) reaction for this study. Although RCM reactions8 proceeded with variable yields, a systematic study of such reactions revealed illuminating trends that support the above conformation-based analysis. These results are described below. The ring-closure substrates 3a-q were prepared by simultaneous or sequential acylation of 1, 2-amino alcohols 1a-j with 4-pentenoic acid or its 2-substituted derivatives 2a-c (EDC, DMAP, CH2Cl2)(Scheme 2), except for the precursor of 5l, which was prepared from tartaric acid benzylidene acetal and 3-buten-1-ol. RCM reactions of 3a-s were performed in the presence of the Grubbs’ catalyst (4)(5-15 mol%) in CH2Cl2 (0.003-0.008 M, reflux, 24-50 h) to produce 5a-s in moderate to excellent yields together with varying amounts of 24-membered cyclic dimers (Figure 1). 10 …