ARTICLE https://doi.org/10.1038/s41467-020-19348-x OPEN Site-switchable mono-O-allylation of polyols ✉ Hua Tang1, Yu-Biao Tian1, Hongyan Cui1, Ren-Zhe Li1, Xia Zhang 1 & Dawen Niu 1,2 Site-selective modification of complex molecules allows for rapid accesses to their analogues and derivatives, and, therefore, offers highly valuable opportunities to probe their functions. However, to selectively manipulate one out of many repeatedly occurring functional groups within a substrate represents a grand challenge in chemistry. Yet more demanding is to develop methods in which alterations to the reaction conditions lead to switching of the 1234567890():,; specific site of reaction. We report herein the development of a Pd/Lewis acid co-catalytic system that achieves not only site-selective, but site-switchable mono-O-allylation of polyols with readily available reagents and catalysts. Through exchanging the Lewis acid additives that recognize specific hydroxyls in a polyol substrate, our system managed to install a versatile allyl group to the target in a site-switchable manner. Our design demonstrates remarkable scope, and is amenable to the direct derivatization of various complex, bioactive natural products. 1 Department of Emergency, State Key Laboratory of Biotherapy, West China Hospital, and School of Chemical Engineering, Sichuan University, 610041 ✉ Chengdu, China. 2 State Key Laboratory of Natural Medicines, China Pharmaceutical University, 210009 Nanjing, China. email: [email protected] NATURE COMMUNICATIONS | (2020) 11:5681 | https://doi.org/10.1038/s41467-020-19348-x | www.nature.com/naturecommunications 1 ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-19348-x olecules containing multiple copies of the same func- modification of polyols. In this realm, various methods21,22 have tional group are ubiquitous in Nature and in drug been established to selectively modify the intrinsically most reac- M — candidates. Site-selective transformations reactions tive hydroxyl groups within a substrate (substrate control). that can manipulate one of these repeating functional groups Recently, systems that can override the intrinsic reactivity pre- while keeping others unaffected—provide efficient access to ferences and accomplish catalyst-controlled, site-switchable analogues and derivatives of these compounds, thereby facilitat- modification of complex polyols have emerged23–35. As eminent ing the interrogation and exploitation of their properties1,2. The examples, in a series of landmark studies, the Miller group23–26, same kind of functional groups tend to undergo similar trans- has identified several oligopeptide-based catalysts that enabled formations, however, and differentiation typically only occurs due site-switchable modification of complex antibiotics, such as van- to subtle steric and electronic environments. Accordingly, to comycin, teicoplanin, and erythromycin. In these studies, the achieve high site-selectivity following a generalizable strategy oligopeptide catalysts were designed as mimetics of the catalytic remains a significant task in chemistry3–8. Even more demanding domains of enzymes. Kawabata et al.29–31 devised chiral pyridine is the development of methods in which alterations to the reac- derivatives that were used in the site-selective modification of C4- tion conditions result in switching of the specific site of reaction OH of glucopyranosides as well as natural products such as 32 (Fig. 1a), since it necessitates the selective modification at posi- lanatoside C and avermectin B2a. The Tan group invented a pair tions that are inherently less reactive9–16. of pseudoenantiomeric imidazole-based catalysts that allow site- Polyhydroxylated natural products, carbohydrates in particular, divergent modification of polyols containing cis-1,2-diol moieties, play critical roles in virtually all biological processes and they are including anticancer agent digitoxin. Nargony and coworkers33 essential components of many pharmaceuticals (Fig. 1b). Rapid employed chiral phosphoric acid catalysts to accomplish site- access to the derivatives of these compounds holds tremendous switchable glycosylation of 6-deoxy erythronolide. Our group34 opportunities to understand and modulate key biological pro- reported the site-divergent O-propargylation of various mono- cesses17–19. Polyols contain numerous hydroxyl groups that offer saccharides as well as digitoxin employing a pair of chiral Cu- excellent opportunities for derivatization. Manipulation of only catalysts. These achievements notwithstanding, to develop general one or a certain few hydroxyls in the presence of many others in systems that can achieve catalyst-controlled, switchable site- these substrates, however, used to rely heavily on protection/ selectivity still represents a challenge. In particular, methods deprotection sequences20. Regardless, recent studies have shown capable of introducing a metabolically stable ether bond remain great promise and potential for the direct, site-selective rather limited. a Catalyst B Catalyst A = A functional group b HO MeO OH OH OH O O OH H HO O OH O O HO O OH HO HO HO O H H OH OH HO OH HO OH OH H O Sucrose Geniposide 20-Hydroxyecdysone c Pd-L LA C1-OH Pd-L LA modified product OH Me 19 HO O Me HO 1 HO C19-OH HO HO HO O modified product O 3’ O HO Ouabain (8 hydroxyls) Reaction site controlled by LA Readily available reagents/catalysts Pd-L LA C3’-OH Broad scope (13 examples) modified product Natural product modification Fig. 1 Site-switchable modification of molecules containing multiple identical functional groups. a Reagent/catalyst-controlled, site-selective modification of complex molecules: a challenge in synthetic chemistry. b Some representative, naturally occurring polyols. c Pd/Lewis acid co-catalyzed, site-switchable modification of polyols (this work). 2 NATURE COMMUNICATIONS | (2020) 11:5681 | https://doi.org/10.1038/s41467-020-19348-x | www.nature.com/naturecommunications NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-020-19348-x ARTICLE a c Ph O O O O Pd(PPh3)4 0 All HO Pd -(PPh3)2 OMe Δ high G 9b Ph O All Additive = O O HO + Pd(PPh3)4 + LA OH O HO OBoc OMe THF, 25 °C, 12 h + low G Ph O OBoc O O 9 HO 12 3 O All OMe 9a Additive Yield Yield Entry 9b/9a (30 mol%) of 9b of 9a b OH OH 1 None 36% 36% 1:1 HO HO 2 2 Cu(OAc)2 56% 8% 7:1 Pd0-L O LA1 O All 3 Zn(II) (R,R)-prophenol 50% 50% 1:1 5 6 OH 4 PhB(OH)2 36% 8% 4.5:1 HO 5 Taylor’s catalyst 80% 20% 4:1 Activated Site-selectivity control 0 OH by additive (not by Pd -L) 6 Ph2BOH* 92% (86%) 7% 13:1 4 7 MgBu2 (-10 °C) 14% 42% 1:3 8 MgBu2/BINOL (-10 °C) 8% 84% (83%) 1:10 LA2 OH All OH O 2 OH HO Ph O Ph Ph Pd0-L Ph O N OH N OH OH OH OH B 7 8 OH Me Selectivity determining stage Bond-forming stage (R,R)-prophenol Taylor’s catalyst BINOL Fig. 2 Mechanistic underpinning and reaction design. a The Pd-catalyzed O-allylation of aliphatic alcohols is accelerated by addition of Lewis acids. b Our reaction design: the use of different Lewis acids (LA) to control site-selectivity. c Initial realization of site-switchable modification of polyols by exchanging additives. ‡Reactions were performed on a 0.2 mmol scale. Product ratios were determined by 1H-nuclear magnetic resonance (NMR) analysis of crude 0 * reaction mixtures. Yields in parenthesis are isolated yields. Pd -(PPh3)2 used in this study was generated in situ from Pd2dba3•CHCl3 and PPh3. Generated in situ from the commercial Ph2BOCH2CH2NH2. Ac acetyl, Ph phenyl, Boc t-butoxycarbonyl, Bu butyl, THF tetrahydrofuran, BINOL 1,1’-bi-2-naphthol, dba dibenzylideneacetone. Here we report a strategy that enables site-switchable mono-O- the allylation of aliphatic alcohols by the Tsuji–Trost reaction is allylation of polyols by Pd/Lewis acid co-catalysis36. As a distinct often a slow process (1 + 2 to 3, Fig. 2a, orange arrow), but it can feature of our system, the task of activating electrophiles and that be accelerated by using Lewis acid additives (Fig. 2a, green arrow, of controlling site-selectivity were allocated, respectively, to the and Supplementary Fig. 2 in Supplementary Information). The Pd-catalyst and the Lewis acid additive. Resembling the role of a Lewis acid additives presumably function through complexing gRNA in the Cas/gRNA37 system, the Lewis acid additive serves as with hydroxyls, which would enhance their acidity and facilitate a guide in our system, and determines the site of reaction in the their deprotonation, thereby increasing their nucleophilicity39. modification of various polyols (Fig. 1c). Interestingly, such a Our initial supposition was that the unique reactivity of aliphatic strategy is akin to the switchable screwdrivers we use in our daily alcohols in the Tsuji–Trost reaction could be leveraged to achieve lives. The potential of this principle, however, has not been sys- reagent-controlled site-selectivity during the modification of tematically explored and exploited in chemistry to develop site- polyols. Specifically, we reasoned that if a Lewis acid additive (e.g., selective methodologies. The derivatization of cardiac glycoside LA1 or LA2 in Fig. 2b) could be identified to selectively complex ouabain provides an excellent example to illustrate our strategy. with a certain hydroxyl and enhance its reactivity toward the π- As will be discussed in more detail below, using the same Pd- allylpalladium intermediates, site-selective functionalization of catalyst to activate the electrophile but changing the identity of the polyols would result since hydroxyls not interacting with this specific Lewis acid additive, we accomplish highly selective ally- additive remain largely inert under the conditions (4 to 6 via 5, lation of the C1-OH, C3’-OH, or C19-OH of ouabain (Fig. 1c). or, 4 to 8 via 7). More importantly, in this regime, switch of the reaction site could in principle be realized simply by tuning the Results properties (bulkiness, number of available binding sites, Brønsted Reaction design.