Catalytic asymmetric desymmetrization reactions are powerful and efficient tools for the synthesis of chiral molecules.[i] These transformations convert simple achiral substrates into complex enantioenriched products through the differentiation of two enantiotopic groups, and can generate complex structures bearing multiple stereocenters in a highly controlled fashion. As such, the development of asymmetric desymmetrization reactions that allow for the construction of important structural motifs is of considerable utility. Tricyclic guanidines are an interesting class of compounds that could potentially be accessed via catalytic asymmetric desymmetrization reactions (Figure 1). These scaffolds are displayed in a wide variety of biologically active natural products,[ii] including the batzelladine alkaloids[iii] (e.g. batzelladine K, 1),[iiic] the merobatzelladine alkaloids (e.g., merobatzelladine B, 2),[iv] and the crambescidin alkaloids[v] (e.g., crambescidin 359, 3).[vb] Many synthetic routes to these compounds involve the generation of a fused-bicyclic urea or guanidine derivative (e.g., 4), which is then transformed to the tricyclic guanidine in subsequent steps.[vi,vii,viii] As such, development of a concise asymmetric synthesis of 4 could provide access to a broad array of interesting alkaloids. Figure 1 Bioactive guanidine alkaloids prepared from bicyclic urea precursors We recently reported an asymmetric synthesis of the tricyclic guanidine natural product (+)-merobatzelladine B (2), which featured a new strategy for the construction of bicyclic ureas and polycyclic guanidines via Pd-catalyzed carboamination reactions of enantiomerically enriched 2-allylpyrrolidine-1-carboxamide derivatives 5 (Scheme 1).[viii] These reactions provided bicyclic urea products 6 in good yield with excellent diastereoselectivity, but control of absolute stereochemistry required the chiral-auxiliary mediated introduction of the C2 stereocenter during the fairly lengthy asymmetric synthesis of 5 (7–9 steps).[ix] Scheme 1 Synthesis of bicyclic ureas via Pd-catalyzed asymmetric desymmetrization A potentially more attractive route to enantiomerically enriched bicyclic ureas and related bi- and tricyclic guanidines would involve the asymmetric Pd-catalyzed desymmetrization of meso-2,5-diallylpyrrolidnyl urea 7. This approach would allow for facile introduction of different R1-substituents, and the alkene present in product 8 provides a convenient handle for further elaboration to tricyclic guanidine products or more highly substituted urea derivatives. In addition, the meso substrate 7 can be prepared in only four steps. Our preliminary studies in this area are described in this communication. These transformations represent the first examples of asymmetric desymmetrizations of bis-alkene substrates in intermolecular Pd-catalyzed alkene carboamination reactions, and also the first examples of six-membered ring formation in an asymmetric Pd-catalyzed alkene carboamination.[x,xi,xii] In initial experiments we elected to employ a catalyst composed of Pd2(dba)3/(S)-Siphos-PE[xiii] for desymmetrization reactions of 7, as we previously illustrated this complex provides good results in related asymmetric carboamination reactions of simple N-allyl urea derivatives.[xiv,xv] We decided to first optimize the structure of the urea N-aryl group, as prior studies in our lab suggested this group may have a significant influence on the level of asymmetric induction.[xiv] Thus, we explored the coupling of Z-1-bromobutene[xvi] with ureas 7 bearing different N-aryl substituents. As shown in Table 1, the use of electron-poor p-cyanophenyl or p-nitrophenyl N-aryl groups resulted in the formation of products 8 with the highest levels of both diastereoselectivity and enantioselectivity. However, these electron-poor substrates were transformed in modest chemical yield due to competing cleavage of the urea moiety (entries 5–6). Use of the electron-rich p-methoxyphenyl group led to improved yields but with lower levels of stereocontrol. After some exploration we found that a substrate bearing a p-chlorophenyl group was transformed to the desired product with both good chemical yield and stereoselectivity (entry 3).[xvii] Table 1 N-Aryl group effects.[a] As shown in Table 2, the asymmetric desymmetrization reactions of 7c are effective with a number of different alkenyl and aryl bromide electrophiles. The main side products generated in these reactions were cis-2,5-diallylpyrrolidine (resulting from competing urea cleavage) and an unsaturated bicyclic urea that is generated by competing β-hydride elimination of an intermediate alkylpalladium complex.[xviii] In the reaction of 7c with E-1-bromohexene a regioisomeric side product bearing a 2-hex-1-enyl group was also generated.[xix] Table 2 Reaction Scope.[a] The best enantioselectivities were obtained when either alkenyl bromides, electron-rich aryl bromides, or electron-neutral aryl bromides were employed as substrates. Diastereoselectivities were generally higher with the alkenyl electrophiles than with aryl electrophiles. Use of sterically hindered aryl bromides (entries 14–15) or electron-poor aryl bromides (entries 9, 11, and 13) led to lower diastereo- and enantioselectivities. Selectivities improved when NaOMe was used in place of NaOtBu in reactions of electron-poor aryl bromides (entries 10 and 12), although yields decreased in these cases. To further demonstrate the utility of the asymmetric desymmetrization reactions, we examined the deprotection of 8c and the conversion of 8c to tricyclic guanidine derivatives. As shown in equation 1, cleavage of the N-p-chlorophenyl group can be accomplished via Pd-catalyzed N-arylation with acetamide[xx, xxi] followed by oxidation of the resulting N-aryl amide with ceric ammonium nitrate. This sequence afforded 9 in 65% yield over two steps. (1) The conversion of 8c to tricyclic guanidine 12 was carried out as shown in Scheme 2. Treatment of 8c with POCl3 followed by NH3 provided bicyclic guanidine 10 in 78% yield. Wacker oxidation of 10 afforded hemiaminal 11, which was then transformed to tricyclic product 12 in 70% yield with 5:1 dr via reductive amination with NaBH3CN.[xxii] Overall, the synthesis of 12, which is structurally related to the batzelladine and merobatzelladine alkaloids, was accomplished in 5 steps and 41% yield from meso-2,5-diallylpyrrolidinyl urea 7c. In addition, this is the first example of a Wacker oxidation/ring-closure sequence to generate a tricyclic guanidine.[xxiii,xxiv] Scheme 2 Conversion of 8c to a tricyclic guanidine derivative Finally, 8c was converted to tricyclic guanidine 16, which is an unnatural stereoisomer of batzelladine k,[xxv,xxvi] as shown in Scheme 3. To avoid problems with base-mediated epimerization of the C4 stereocenter, the Pd-catalyzed N-arylation with acetamide was carried out prior to Wacker oxidation of the alkene. This two-step sequence provided 13 in 65% yield. Reduction of the alkene followed by CAN deprotection generated urea 14, which was converted to guanidine aminal 15 by O-methylation and treatment with ammonia.[xxvii] The reduction of 15 proceeded with modest diastereoselectivity (3:1 dr), but upon purification 9-epi-batzelladine k was isolated as a single stereoisomer in 48% yield over three steps from 14. Scheme 3 Conversion of 8c to 9-epi-batzelladine K. In conclusion we have developed a concise route to enantiomerically enriched bicyclic ureas via Pd-catalyzed desymmetrizing carboamination reactions of meso-diallylpyrroldinyl ureas. These transformations effect formation of both a C–N and a C–C bond, and provide products bearing three stereocenters with good levels of diasteroselectivity and enantioselectivity. These reactions illustrate the potential utility of asymmetric Pd-catalyzed alkene carboamination for desymmetrization processes and provide synthetically valuable products in a straightforward manner. Further exploration of enantioselective Pd-catalyzed alkene difunctionalization reactions are currently underway.