Formal total syntheses of classic natural product target molecules via palladium-catalyzed enantioselective alkylation

Summary Pd-catalyzed enantioselective alkylation in conjunction with further synthetic elaboration enables the formal total syntheses of a number of “classic” natural product target molecules. This publication highlights recent methods for setting quaternary and tetrasubstituted tertiary carbon stereocenters to address the synthetic hurdles encountered over many decades across multiple compound classes spanning carbohydrate derivatives, terpenes, and alkaloids. These enantioselective methods will impact both academic and industrial settings, where the synthesis of stereogenic quaternary carbons is a continuing challenge.


Introduction
Catalytic enantioselective allylic alkylation has emerged as a powerful method for the construction of building blocks bearing quaternary carbon and fully substituted tertiary centers [1,2]. A recent addition developed by our laboratory is the allylic alkylation of nonstabilized enolate precursors to form α-quaternary carbonyl compounds (Scheme 1) [3]. Once the key stereocenter is set by this chemistry, further elaboration allows access to many bioactive small molecules. In our lab alone, this palladium-catalyzed alkylation has enabled the enantioselective total syntheses of dichroanone [4], elatol [5], cyanthiwigins [6][7][8], carissone [9], cassiol [10], chamigrenes [11], and liphagal [12]. Other labs have also utilized our method in natural product total synthesis [13,14]. Often, it is the case that a new technology that allows the synthesis of building blocks will open up new avenues to complex structures of long standing interest [15,16]. Herein we detail the application of this asymmetric chemistry in formal total syntheses of "classic" natural product targets across a range of compound families by strategic selection of allylic alkylation substrates and subsequent product transformations.

D) Aspidospermine
The aspidosperma alkaloids have garnered much attention as beautiful targets for the synthetic chemist. Most of the 250-plus Scheme 7: Formal total synthesis of (−)-dysidiolide. compounds in this class share a pentacyclic core, from the clinical anticancer therapeutics vincristine and vinblastine to the simpler aspidospermidine [60]. To address the challenging synthetic features of the aspidosperma alkaloids, many clever synthetic approaches have been reported [61,62]. One popular target in this family is aspidospermine (36, Scheme 8). Although its medicinal potency is inferior to other members of the class, this alkaloid has served as a proving ground for many synthetic chemists.
In 1989, Meyers reported an enantioselective synthesis of the (4aS,8aR,8S)-hydrolilolidone core 37 [63,64] present in aspidospermine (36), and thus a formal total synthesis of the alkaloid itself, intercepting Stork's classic route [61]. One precursor described in the core synthesis is enone 38, which bears the quaternary stereocenter of the natural product. Contrasting Meyers' approach, which employed a chiral auxiliary as part of 39, we thought a catalytic enantioselective alkylation strategy would be ideal for a formal total synthesis of natural (−)-aspidospermine (36) via the antipode of 38.
In 2001, Magnus reported a total synthesis of rhazinilam in racemic form (Scheme 10) [83]. In their approach, the first retrosynthetic disconnection of the amide C-N bond in the ninemembered ring led to tricyclic compound 45. The pyrrole ring of 45 was formed by intramolecular condensation of cinnamyl amide 46, which is prepared via union of quaternary piperidinone 47 and cinnamyl electrophile 48. We envisioned that our allylic alkylation of lactam enolates would furnish enantioenriched piperidinone 47, and thus a single enantiomer of rhazinilam may be prepared.
Magnus' route. This formal synthesis demonstrates the utility of our recently developed asymmetric lactam alkylation chemistry.

F) Quebrachamine
Quebrachamine (51) is an indole alkaloid isolated from the Aspidosperma quebracho tree bark [60]. It has been found to possess adrenergic blocking activities for a variety of urogenital tissues [85]. Structurally, it features a tetracycle including an indole nucleus, a 9-membered macrocycle, and an all-carbon quaternary stereocenter. Due to its structural complexity and biological activities, quebrachamine has received considerable attention from the chemistry community. A number of total syntheses have been reported [86][87][88], with several examples of asymmetric syntheses [89][90][91].
tion at the macrocycle led to amide 52, which was prepared from 3,3-disubstituted piperidine 53. The all-carbon quaternary stereocenter in 53 was installed by double alkylation of lactam 55, using an auxiliary to control the stereoselectivity. We envisioned that an alternative way of constructing this motif would again make use of our recently developed palladium-catalyzed asymmetric alkylation of lactam enolates. The formal synthesis of (+)-quebrachamine commenced with benzoyl lactam 50 (Scheme 13), which was prepared in excellent yield and ee by alkylation of carboxy-lactam 49 (see Scheme 11) [84]. Oxidative cleavage of the terminal double bond and subsequent reduction with LiAlH 4 afforded N-benzylpiperidine-alcohol 56 [84]. Hydrogenolysis of the N-benzyl group and re-protection with di-tert-butyl dicarbonate Scheme 13: Formal total synthesis of (+)-quebrachamine.
Recently, Pandey reported a highly efficient synthesis of (+)vincadifformine (Scheme 14) [106]. The key step in the synthesis was an iminium ion cascade reaction that formed the fused ring systems by coupling 3,3-disubstituted tetrahydropyridine 57 with indole derivative 58. The former coupling partner was derived from chiral α-quaternary lactam 60, which was constructed using a chiral auxiliary strategy. We envisioned that chiral lactam 60 could again be readily accessed by our palladium-catalyzed enantioselective alkylation chemistry.
The formal synthesis of (−)-vincadifformine commenced with ruthenium-catalyzed isomerization of the terminal olefin moiety in unprotected piperidinone (+)-47 (made previously in the formal synthesis of (+)-rhazinilam shown in Scheme 11) to produce internal olefin 63 (Scheme 15) [109]. Ozonolysis of the double bond furnished aldehyde 64, which was reduced under Luche conditions to alcohol 65, a compound identical in structure and enantiomeric to the intermediate employed by Pandey in the synthesis of (+)-vincadifformine.

Conclusion
The development of a series of Pd-catalyzed methods for constructing stereogenic quaternary carbons has provided two generations of building blocks (Scheme 16). The described derivatization enabled the formal total syntheses of an array of classic natural products including sugar derivatives, terpenes, and alkaloids, adding significantly to the growing list of uses for this powerful C-C bond construction. An efficient route to the sesquiterpenoid (−)-thujopsene (10) has been delineated, allowing access to the compound's natural antipode. Our lab's novel approach to (−)-quinic acid (21) allowed access to either Scheme 15: Formal total synthesis of (−)-vincadifformine. enantiomer of this important substance. We have also intercepted a key intermediate in Danishefsky's synthesis of (±)dysidiolide (29), rendering the former racemic route enantioselective. Additionally, a rapid approach to a compound in Meyers' formal synthesis of (+)-aspidospermine (36) granted access to the natural product without the use of a chiral auxil- iary. Finally, we have demonstrated the application of lactam alkylation products in the catalytic asymmetric syntheses of (+)rhazinilam (44), (+)-quebrachamine (51), and (−)-vincadifformine (59). The powerful catalytic enantioselective allylic alkylation will undoubtedly enable new synthetic endeavors in the context of both academic and industrial research.

Supporting Information
Supporting information features experimental procedures, characterization data of synthesized compounds, copies of 1 H and 13 C NMR spectra, and single crystal structure data.

Supporting Information File 1
Experimental data, NMR spectra and X-ray data.