Still Unconquered: Enantioselective Passerini and Ugi

Article Cite This: Acc. Chem. Res. 2018, 51, 1290−1300

pubs.acs.org/accounts

Still Unconquered: Enantioselective Passerini and Ugi Multicomponent Reactions Qian Wang,† De-Xian Wang,‡ Mei-Xiang Wang,*,§ and Jieping Zhu*,† †

Laboratory of Synthesis and Natural Products, Institute of Chemical Sciences and Engineering, Ecole Polytechnique Fédérale de Lausanne, EPFL-SB-ISIC-LSPN, BCH5304, CH-1015 Lausanne, Switzerland ‡ Beijing National Laboratory for Molecular Sciences, CAS Key Laboratory of Molecular Recognition and Function, Institute of Chemistry, Chinese Academy of Sciences, Beijing 100190, China § Laboratory of Bioorganic Phosphorous Chemistry and Chemical Biology (Ministry of Education), Department of Chemistry, Tsinghua University, Beijing 100084, China CONSPECTUS: The Passerini three-component (P-3CR) and the Ugi four-component (U-4CR) are two of the most prominent isocyanide-based multicomponent reactions (IMCRs). The P-3CR transforms isocyanides, aldehydes (ketones), and carboxylic acids to α-acyloxy carboxamides, while the U-4CR converts isocyanides, aldehydes (ketones), amines, and carboxylic acids to α-acetamido carboxamides. Conversion of the high energy formal divalent isocyano carbon into a tetravalent amide carbonyl carbon provides the driving force for these reactions. While the prototypical P-3CR and U-4CR provide linear adducts, many heterocycles and macrocycles are now readily synthesized by modifying these truly versatile reactions. As one stereocenter is generated by the nucleophilic addition of the isocyanide to the carbonyl and imine functions, the search for enantioselective versions of these reactions has become a much sought after goal among synthetic chemists. This seemingly trivial endeavor turns out to be extremely difficult to achieve, in sharp contrast to the remarkable progress documented in the field of asymmetric synthesis in general and catalytic enantioselective nucleophilic addition to CX bond in particular. Since Denmark’s first report in 2003 on the catalytic enantioselective Passerini two-component reaction of isocyanides with aldehydes, several Lewis acid (LA) and Brønsted acid-catalyzed enantioselective protocols have been developed. However, it is fair to say that truly catalytic enantioselective P-3CR and U-4CR with wide application scope remain elusive. In this Account, we summarize the progress recorded in this field over the past 15 years. We entered the field by investigating the enantioselective reaction of α-isocyanoacetamides with aldehydes and imines, which was previously developed in our lab for the synthesis of functionalized 5-aminooxazoles. Our initial experimental results, in conjunction with Dömling’s and Schreiber’s earlier findings, prompted us to assume that the low turnover number in LA-catalyzed asymmetric IMCRs is a main hurdle for enantioselectivity. We speculated that the LA incapable of forming chelates would be the catalyst of choice for enantioselectivity, the rational being that the P-3CR and the U-4CR afforded bidentate intermediates (α-hydroxy imidates, α-amino imidates) and products (α-acyloxy carboxamides, α-acetamido carboxamides) from nonchelating inputs. Therefore, the transfer of catalyst from these chelating intermediates or products to the monocoordinating starting materials would be difficult, hence the problem with catalyst turnover. This working hypothesis turned out to be a valuable guide that allowed us to develop Al-salen and Al-phosphate-catalyzed enantioselective P-3CR and enantioselective construction of chiral heterocycles such as oxazoles and tetrazoles. Nevertheless, all our attempts to apply these LA catalysts to the Ugi reaction failed. Indeed, to date, no reports on the successful LA-catalyzed asymmetric Ugi-type reactions exist in the literature. However, significant progress has been made in recent years employing organocatalysts. We developed a chiral phosphoric acid (CPA)-catalyzed enantioselective three-component synthesis of 2-(1-aminoalkyl)-5-aminooxazoles, a four-component synthesis of epoxy-tetrahydropyrrolo[3,4-b]pyridin-5-ones and a Ugi four-center, three-component reaction of isocyanides, anilines, and 2-formylbenzoic acids for the synthesis of isoindolinones. Other groups have found that chiral dicarboxylic acid and BOROX are effective catalysts for truncated Ugi three-component reactions. MCRs have long remained an underexploited domain.2 In this context, the development of enantioselective isocyanide-based multicomponent reactions (IMCRs) is especially challenging.3

1. INTRODUCTION Multicomponent reactions (MCRs) combine three or more reactants in a single operation to afford an adduct with minimum loss of atoms.1 After half a century of intensive research in the field of asymmetric synthesis, many classic reactions now have enantioselective variants. In sharp contrast, asymmetric © 2018 American Chemical Society

Received: March 9, 2018 Published: April 30, 2018 1290

DOI: 10.1021/acs.accounts.8b00105 Acc. Chem. Res. 2018, 51, 1290−1300

Article

Accounts of Chemical Research The divalent isocyano carbon undergoes α-addition with both nucleophiles and electrophiles (Scheme 1a).4 The Passerini

and the U-4CR enantioselective. However, this seemingly trivial task turned out to be extremely challenging and even after many years of intensive research, we are still at the beginning of this adventure. The difficulties associated with the development of catalytic enantioselective P-3CR and U-4CR are multiple and are summarized by the following. (a) Competition from background reactions, as both P-3CR and U-4CR occur spontaneously in an appropriate solvent at room temperature. (b) Difficulty in chemoselective activation of carbonyl or imine function by a Lewis acid (LA) due to the Lewis basicity of the other components. (c) Coordination of isocyanide to metal11 is known to divert the reaction pathway of isocyanides.12 (d) Low catalyst turnover due to product inhibition. Indeed, the transfer of chiral catalyst from the bidentate P-3CR and U-4CR products to the monodentate reactant could be a kinetically unfavorable process. (e) The complexity of the reaction mechanism. The addition of isocyanide to carbonyl or imine leading to nitrilium intermediate has long been considered as a reversible process. However, a recent theoretical study indicated that the formation of nitrilium was a rate-limiting step in these MCRs and that conversion to the α-adduct 10 was highly exothermic with a low activation energy.13 It was concluded that the key C−C bond forming step that creates the stereocenter was irreversible, contradicting common wisdom. This mechanistic insight is critically important for the development of enantioselective process. Nevertheless, the existence of other reaction manifolds rendered the task more difficult than it might appear. For example, Saegusa et al. have shown that BF3·Et2O-catalyzed P-3CR can proceed, in addition to the classic pathway leading to 4 (Scheme 3a), through

Scheme 1. Isocyanides, the P-3CR, and the U-4CR

three-component (P-3CR)5,6 and the Ugi four-component reactions (U-4CR),7,8 two of the most prominent IMCRs, each feature this key elementary step. The P-3CR converts isocyanides 1, aldehydes (ketones) 2, and carboxylic acids 3 to α-acyloxy carboxamides 4 (Scheme 1b), while the U-4CR transforms isocyanides 1, aldehydes (ketones) 2, amines 5, and carboxylic acids 3 to α-acetamido carboxamides 6 (Scheme 1c). Apolar solvents (e.g., CH2Cl2) are generally preferred for the P-3CR whereas polar protic solvents (e.g., MeOH) are the conventional reaction media for the U-4CR. Being extremely powerful in generating the molecular complexity and diversity, both the P-3CR and the U-4CR have been extensively applied to the synthesis of heterocycles, natural products,4 and pharmaceuticals.9 Since one stereocenter is generated, controlling the stereochemical outcome of these reactions would significantly expand their synthetic utility. Whereas diastereoselective P-3CR and U-4CR using chiral substrates3,10 are known, progress on the development of truly catalytic asymmetric P-3CR and U-4CR reactions has been slow, despite dedicated effort. The widely accepted reaction pathway describing the mechanism for P-3CR and U-4CR is highlighted in Scheme 2.

Scheme 3. P-3CR and U-4CR: Alternative Reaction Pathways

Scheme 2. Ugi’s Mechanistic Proposal for P-3CR and U-4CR

an imino oxirane intermediate 11 resulting from the intramolecular trapping of the nitrilium by the adjacent alkoxide. Ring opening of the epoxide by carboxylate would then afford ent-4 (Scheme 3b).14 In the Ugi-type reaction, formation of hemiaminal 12 followed by an SN2 displacement of acetate group by isocyanide has also been proposed to account for the formation of nitrilium ent-9, which would then be converted to the U-4CR adduct ent-6 following the usual pathway.15 The concurrent presence of these reaction pathways, inconsequential to the achiral transformations, would be detrimental to the reaction’s enantioselectivity as the Ugi’s reaction manifold (Scheme 3a) and those competitive processes (Scheme 3b,c) afforded two enantiomers. The aforementioned factors rendered the development of catalytic enantioselective P-3CR and U-4CR highly challenging. Since the pioneering report of Denmark in 2003 on the asymmetric

Nucleophilic addition of the divalent isocyano carbon to the activated carbonyl/imine 8 would afford nitrilium 9 which, upon trapping by carboxylate, would furnish imidate 10. The Mumm rearrangement of 10 would then provide P-3CR adduct 4 (X = O) or U-4CR adduct 6 (X = NR4). Despite the reversibility of the intermediate steps, the last irreversible Mumm rearrangement pulls the reaction toward formation of the multicomponent adducts. Catalytic enantioselective addition of nucleophiles to carbonyls or imines has been extensively investigated and many important organic reactions involving this stereocenter-generating step can now be performed in a highly enantioselective manner. Therefore, one might expect it to be easy to render the P-3CR 1291


Article

Accounts of Chemical Research two-component Passerini reaction, several catalytic enantioselective protocols have been developed. In this Account, we summarize the progress made in this field in the hope of stimulating other researchers to tackle this challenging task and, ultimately, to provide a long-awaited solution to this interesting problem.

Scheme 5. Mechanism of Two-Component Passerini Reaction

2. ENANTIOSELECTIVE PASSERINI REACTIONS 2.1. Enantioselective Two-Component Passerini Reaction of Aldehydes with Isocyanides

Denmark reported the first examples of enantioselective α-addition of isocyanides to aldehydes that afforded, after hydrolysis, the enantioenriched α-hydroxy carboxamides 13.16 Different from other asymmetric transformations based on the concept of Lewis base activation of a Lewis acid,17 a control experiment showed that SiCl4, without being activated by an external Lewis base, effectively promotes the addition of tBuNC to benzaldehyde, even at −78 °C. Reasoning that isocyanide could increase the Lewis acidity of SiCl4 via coordination, they proposed reducing the effective concentration of isocyanide in the reaction mixture to minimize the background reaction. Indeed, slow addition of tBuNC (1.2 equiv) to a solution of SiCl4 (1.0 equiv), benzaldehyde, phosphoramide (R,R)-14 (0.05 equiv), and Hünig’s base (0.1 equiv) in CH2Cl2 at −74 °C afforded 2-hydroxy carboxamide in 93% yield with an er of 96.4/3.6. Excellent enantioselectivities were observed with aromatic, heteroaromatic, and α,β-unsaturated aldehydes, but the ee dropped notably with aliphatic aldehydes (Scheme 4).

leads to 16. Nucleophilic addition of the isocyanide to the activated aldehyde would produce the nitrilium intermediate 17, which is then trapped by the chloride to provide imidoyl chloride 18. The catalyst turnover was proposed to be the transfer of the chiral ligand from 18 to SiCl4, rather than cleavage of the O−Si bond as is generally observed in conventional LA catalysts. The attenuated Lewis acidity of the silicon center in the imidoyl chloride 18 enables a facile transfer of the Lewis base catalyst to SiCl4. Finally, aqueous workup converts 19 to the product 13. 2.2. Enantioselective Passerini Three-Center, Two-Component Reaction of Aldehydes with α-Isocyanoacetamides

Scheme 4. SiCl4/Chiral Phosphoramide-Catalyzed Two-Component Passerini Reaction

The reaction of aldehydes with α-isocyanoacetamides 20 affords 2-(1-hydroxyalkyl)-5-aminooxazoles (Scheme 6).18 In our initial Scheme 6. Enantioselective Reaction of α-Isocyanoacetamides with 2-(Benzyloxy)acetaldehyde

survey, we found that a combination of Sn(OTf)2 and PyBOX indeed catalyzes the reaction of 20a with 2-(benzyloxy)acetaldehyde to afford 5-aminooxazole 21a in a 63% yield with an 80% ee (Scheme 6).19 Control experiments showed that the chirality of 20a had no influence on the enantioselectivity, since both the (S)- and the (±)-20a provided similar results. Unfortunately, applying the above catalytic system to nonchelating aldehydes provided racemic 5-aminooxazoles. This led us to hypothesize that there was a catalyst turnover problem in this reaction. Indeed, the 2-(1-hydroxyalkyl)-5-aminooxazoles, such as 21a, are bidentate ligands that might form chelates with Lewis acids. The transfer of a catalyst from these chelates to monocoordinating aldehydes, an essential step for catalyst turnover, would therefore be inefficient. On the basis of these considerations, we suspected that any Lewis acid capable of chelating the bidentate substrates/products would not be a good candidate for the enantioselective process. Guided by this working hypothesis, we re-investigated the reaction of α-isocyanoacetamide with simple aldehydes and found that the (salen)Al(III)Cl complex

A mechanistic proposal is shown in Scheme 5. Coordination of a siliconium ion/hexachlorosilicate ion pair 15 to an aldehyde 1292


Article

Accounts of Chemical Research

the same activity as that generated in situ. We assumed that the Al−CPA complex existed mainly as a trans metal complex of type 24, which is essential for enantioselectivity. Matsunaga, Shibasaki, and co-workers demonstrated that a heterobimetallic Schiff base complex 26, formed in situ from Ga(OiPr)3/Yb(OTf)3 and (S,S)-25, was highly effective in catalyzing the reaction of α-isocyanoacetamides 20 with aldehydes 2 (Scheme 9).22 Various aromatic, heteroaromatic, and alkenyl

22 (0.1 equiv) can catalyze the reaction to give 5-aminooxazoles 21 in good yields, albeit with moderate enantioselectivities (Scheme 7).20 Scheme 7. Chiral (Salen)Al(III)Cl Complex-Catalyzed Reaction of α-Isocyanoacetamides with Aldehydes

Scheme 9. Bimetallic Schiff Base Complex-Catalyzed Enantioselective Reaction of α-Isocyanoacetamides with Aldehydes

We subsequently developed a chiral aluminum-organophosphate-catalyzed enantioselective reaction of 20 with aldehydes. The catalyst [23]2Al(III)Cl was prepared by mixing chiral phosphoric acid (CPA) 23 and Et2AlCl in a 2:1 mol ratio.21 Under optimized conditions (10 mol % of 23, 5 mol % of Et2AlCl, toluene, −70 °C, c 0.1 M), the reaction between 2 and 20 afforded 5-aminooxazoles 21 in good yields and enantioselectivities (Scheme 8). The preformed [23]2Al(III)Cl displayed Scheme 8. Al-Organophosphate-Catalyzed Reaction of α-Isocyanoacetamides with Aldehydes aldehydes participated in the reaction to afford 5-aminooxazoles 21 in excellent yields and enantioselectivities (95−98% ee). 3-Phenylpropanal is also involved in the reaction and provides the corresponding aminooxazole with slightly reduced enantioselectivity (88% ee). Catalyst 26 was assumed to not only activate the aldehyde but also interact with the α-isocyanoacetamide to effectively control the orientation of the two substrates in the enantio-discriminating step. This is by far the most efficient catalyst for the enantioselective synthesis of 5-aminooxazoles. Zhong and co-workers found that CPA (R)-27 efficiently catalyzed the reaction of α-isocyanoacetamides 20 with aldehydes 2 (Scheme 10).23 Under their optimized conditions, the reaction of aliphatic aldehydes 2 (2.0 equiv) with 20 in toluene at −40 °C in the presence of a catalytic amount of (R)-27 and molecular sieves (5 Å) afforded 5-aminooxazoles 21 in excellent yields (85−98%) with good to excellent enantioselectivities (68−99% ee). Aromatic aldehydes were unreactive under these conditions. 2.3. Catalytic Enantioselective Passerini Three-Component Reaction

Dömling and co-workers found that a stoichiometric amount of Ti-TADDOL complex 28 was capable of promoting P-3CR.24 1293


Article

Accounts of Chemical Research Scheme 10. Chiral Phosphoric Acid-Catalyzed Enantioselective Reaction of α-Isocyanoacetamides with Aldehydes

Scheme 12. Cu(II) Indan PyBOX-Catalyzed P-3CR

(S)-Enriched α-acyloxy carboxamides 4 were isolated in low to moderate yields and ee’s (Scheme 11). While the reaction was We applied the (salen)Al(III)Cl complex 22 to the threecomponent Passerini reaction (Scheme 13).26 By slowly adding

Scheme 11. Ti-TADDOL Complex-Promoted P-3CR

Scheme 13. (Salen)Al(III)Cl-Catalyzed Enantioselective P-3CR

not catalytic and the enantioselectivity remained low, this nonetheless represented the first examples of Lewis acid promoted asymmetric P-3CR. Schreiber and co-workers demonstrated that tridentate Cu(II)-PyBOX complex 29 was capable of catalyzing the Passerini reaction (Scheme 12).25 The reaction must be performed under anhydrous conditions to ensure high enantioselectivity as water can deactivate the catalyst and render the background reaction more competitive. As an acidic component, benzoic acid gave superior results to phenylacetic acid. Both aliphatic and aromatic isocyanides took part in the reaction, with tBuNC and 4-methoxyphenyl isocyanide giving the best results. Aldehydes were limited to the chelating species [2-(benzyloxy)acetaldehyde, furan-2-carbaldehyde, thiophene-2-carbaldehyde] as the reaction involving benzaldehyde afforded the racemic P-3CR adduct.

carboxylic acid 3 to a solution of isocyanide 1, aldehyde 2, and catalyst 22, a variety of nonchelating aliphatic aldehydes including α-substituted, as well as aryl, alkyl, and alkenyl carboxylic acids, and aryl and benzyl isocyanides participated in this catalytic asymmetric transformation to afford 4 with good to excellent enantiomeric excess. Chloroacetic acid gave the highest ee of the α-acyloxy carboxamides 4. The less-reactive aromatic isocyanides afforded the adducts with higher ee than their aliphatic counterpart, probably due to the diminution of the uncatalyzed background reaction. The salen-Al complexes can adopt trans, cis-β, and cis-α conformers (Figure 1).27 The former is unable to form chelates 1294


Article

Accounts of Chemical Research

Scheme 15. [(Salen)Al(III)Me]-Catalyzed Enantioselective Three-Component Reaction of Isocyanides, Aldehydes, and Hydrazoic Acid to 5-(1-Hydroxyalkyl)tetrazoles

Figure 1. Conformations of salen−metal complexes.

with bidentate ligands, while the two others can. The salen-AlCl complex 22 is known to adopt a trans conformer, which we assumed to be beneficial for catalyst turnover. As a control experiment, we synthesized the Al-complex 30, a known cis-β complex.28 As shown in Scheme 14, the reaction of isobutyraldehyde, Scheme 14. Importance of the Catalyst Conformation: A Control Experiment

2-chloroacetic acid, and 4-methoxyphenylisocyanide in the presence of 22 afforded the three-component adduct in 64% yield with 99% ee, while the same reaction under identical conditions using 30 as catalyst furnished the racemic adduct in 58% yield.29 These results provide indirect evidence that support our initial working hypothesis. In 1961, Ugi and Meyr reported that hydrazoic acid can replace carboxylic acid in P-3CR to afford tetrazole 31 in moderate to good yields.30 Having succeeded in the development of enantioselective P-3CR, we subsequently investigated a catalytic enantioselective three-component Passerini reaction of isocyanides 1, aldehydes 2, and hydrazoic acid (32).31 Our initial study using (salen)Al(III)Cl complex (22) afforded the desired tetrazole 31a (R1 = iBu, R2 = 4-MeOC6H4) together with a significant amount of 2-hydroxy carboxamide 33 with similar ee’s (Scheme 15a). We hypothesized that the formation of 31 and 33 proceed through the same nitrilium intermediate 34. Addition of HN3 to 34 would provide tetrazole 31, while trapping of 34 by in situ generated HCl would afford chloroimidate 35, which, in the presence of adventitious water, would be converted to hydroxy carboxamide 33 (Scheme 15b). Mechanistic studies also indicate that the azide group is directly transferred from hydrazoic acid rather than from the Al-bound azide and that the tetrazole itself does not significantly modify the catalytic properties of the aluminum complex. According to this mechanistic insight, several other Al(salen) complexes were examined. Optimized conditions consisted of performing the reaction in toluene (c = 0.2 M) at −40 °C and using [(salen)Al(III)Me] (36) as a catalyst. A wide range of linear and α-branched aliphatic aldehydes and aromatic and

aliphatic isocyanides including α-isocyanoacetates were employed in the reaction to afford the corresponding tetrazoles in good to excellent yields and ee’s. Using acrolein as a carbonyl input, a sequence of Michael addition/enantioselective P-3CR occurred to afford highly functionalized tetrazoles in good yields and ee’s (Scheme 15). Liu, Tan, and co-workers reported a CPA 37-catalyzed enantioselective P-3CR (Scheme 16).32 Under their optimized conditions, a wide range of aldehydes, carboxylic acids, and hindered isocyanides took part in the reaction to furnish the three-component adducts in excellent yields and enantioselectivities. Based on precedents from the literature,33 the authors proposed that the formation of a H-bonded heterodimer 38 between CPA and carboxylic acid was important for the success of the reaction, as it could increase the acidity of the CPA and enhance the nucleophilicity of the carboxylic acid. This protocol represents, to date, the most efficient catalytic enantioselective P-3CR. 1295


Article

Accounts of Chemical Research Scheme 16. CPA-Catalyzed Enantioselective P-3CR

Scheme 17. CPA-Catalyzed Enantioselective Reaction of Aldehydes, Amines, and α-Isocyanoacetamides

3. ENANTIOSELECTIVE UGI REACTION 3.1. Enantioselective Ugi Three-Component Reaction of α-Isocyanoacetamides, Aldehydes, and Amines

We have developed a three-component synthesis of 2-(1-aminoalkyl)-5-aminooxazoles 39 from α-isocyanoacetamides, aldehydes, and anilines (Scheme 17).34 Many Brønsted and Lewis acids were known to catalyze or promote this reaction; however, our extensive effort in developing an enantioselective version using chiral LA was met with failure. Finally, we found that CPA was a competent catalyst.35 Stirring a toluene solution of aldehydes 2, anilines 5, and α-isocyanoacetamides 20 (c = 0.05 M, −20 °C) in the presence of CPA 40 (20 mol %) afforded oxazoles 39 in excellent yields with moderate to good enantioselectivities. 4-Substituted anilines with different electronic properties took part in the reaction. Aliphatic aldehydes, including linear and α-branched, were acceptable substrates with the hindered pivalaldehyde being the best in terms of enantioselectivity. Aromatic aldehydes were converted to the corresponding adducts, albeit with reduced ee (results not shown). The α-phenyl, benzyl, or methyl substituted αisocyanoacetamides, as well as a dipeptidic isocyanide participated in this reaction to afford the corresponding 5-aminooxazoles in excellent yields with good to high ee’s. This represented the first examples of catalytic enantioselective Ugi-type MCR.36 This demonstrates, also for the first time, the utility of chiral Brønsted acid in the development of catalytic enantioselective IMCRs. Scheme 18 depicts a possible reaction pathway. Protonation of imine 41 by CPA 40 would lead to iminium salt 42. Addition of isocyanide 20 onto the Si face of the iminium would afford nitrilium 43, which then cyclizes to form 44. Deprotonation of 44 would furnish 5-aminooxazole 39 with concurrent regeneration of CPA. Aminooxazole 39 contains three basic nitrogen atoms, which can compete with imine 41 to form a hydrogen bond or to form a salt with CPA 40, which would be detrimental to the

Scheme 18. Reaction Pathway of CPA-Catalyzed Enantioselective Three-Component Synthesis of 5-Aminooxazoles

catalytic cycle. However, the results of control and NMR titration experiments demonstrated that CPA selectively activates the imine, making product inhibition minimal. As expected, aliphatic amines afforded the three-component adducts with low ee’s. α-Isocyanoacetates bearing an electron-withdrawing group (45, EWG = 4-NO2) at the α-position display a reactivity akin to 1296


Article

Accounts of Chemical Research α-isocyanoacetamides.37 We subsequently developed a CPAcatalyzed enantioselective four-component synthesis of epoxytetrahydropyrrolo[3,4-b]pyridin-5-ones 46 (Scheme 19).38

Scheme 20. Chiral Dicarboxylic Acid-Catalyzed Enantioselective Ugi Four-Center, Three-Component Reaction of Isocyanides, Aldehydes, and Benzhydrazides

Scheme 19. CPA-Catalyzed Enantioselective Four-Component Reaction

Stirring a CH2Cl2 solution of aldehydes 2, anilines 5, α-isocyanoacetate 45, and a catalytic amount of CPA 47 at −35 °C for 24 h provided 5-ethoxyoxazoles 48. Addition of α,β-unsaturated acyl chloride 49 and refluxing the resulting solution for 5 h afforded, via a sequence of acylation and intramolecular Diels−Alder cycloaddition, the polycycle 46 in high yields and ee’s. While only one diastereomer was detected with pivalaldehyde, a minor diastereomer was formed with cyclohexanecarboxaldehyde. Various substituted anilines worked well to afford the cycloadducts 46 in high yields with excellent ee’s. In this one-pot four-component reaction, five contiguous stereocenters including two quaternary ones were created. The absolute configuration of 2-(1-aminoalkyl)-5-alkoxyoxazoles 48 dictated those of the remaining stereocenters in 46. The relative stereochemistry can be accounted for assuming that the cycloaddition proceeded through an amide-exo/R3-endo mode and that the bulky R2 group adopted a pseudoequatorial position to avoid a steric clash with other substituents.

by trapping the transient nitrilium intermediates 54 by the internal carboxamide furnished heterocycles 52 with the concurrent regeneration of catalyst 51. Hydrolysis of 52a (R1 = 2-BzOC6H4, R2 = Ph, R3 = 2-MeOBn) under acidic conditions gave the corresponding α-hydrazinocarboxamide 55 without loss of ee. On the other hand, treatment of a trifluoroethanol solution of 52a with K2CO3 afforded the benzoxazole derivative 56 with a slight erosion of enantiopurity. Finally, the N−N bond in 56 can be reduced with SmI2 to afford the secondary amine 57 in 75% yield (Scheme 20). 3.3. Enantioselective Ugi Three-Component Reaction of Isocyanides, Aldehydes, and Amines

The BOROX catalyst, developed by Wulff and co-workers, is an ion pair containing a chiral boroxinate anion and a protonated substrate. After screening 13 different chiral ligands, 12 amines, and 47 alcohols/phenols in combination with BH3·SMe2, they identified the optimal chiral BOROX catalyst 59 for the enantioselective Ugi three-component reaction of tBuNC, dibenzylamine, and aromatic aldehydes.40 Under optimized conditions, the α-amino carboxamides 58 were isolated in high yields and ee’s (Scheme 21). The reaction was, nevertheless, limited to aromatic aldehydes.

3.2. Enantioselective Ugi Three-Component Reaction of Isocyanides, Aldehydes, and Benzhydrazides

Maruoka and co-workers developed a chiral dicarboxylic acid-catalyzed reaction of aldehydes 2, benzhydrazides 50, and 2-benzoyloxyphenyl isocyanides (Scheme 20).39 Under optimum conditions [(R)-51 (5 mol %), 4 Å MS, m-xylene, −30 °C, 40 h], heterocycles 52 were isolated in excellent yields and ee’s from aromatic aldehydes. Cyclohexanecarboxaldehyde and hydrocinnamaldehyde participated in the reaction to provide the products with reduced ee’s. The reaction was proposed to proceed through acyclic azomethine imine intermediates 53, generated in situ by the acid-catalyzed condensation of aldehydes 2 and benzhydrazides 50. Subsequent nucleophilic addition of isocyanides 1 to 53 followed

3.4. CPA-Catalyzed Enantioselective Ugi Four-Center, Three-Component Reaction of 2-Formylbenzoic Acids, Amines, and Isocyanides

The carboxylic acid has been missing in all of the aforementioned enantioselective Ugi-type reactions. With the objective of developing a true Ugi reaction involving a carboxylic acid input, we began by investigating the catalytic enantioselective two-component 1297


Article

Accounts of Chemical Research Scheme 21. BOROX-Catalyzed Enantioselective Three-Component Ugi Reaction

Scheme 23. CPA-Catalyzed Enantioselective U-4C-3CR

reaction of isocyanides 1 with 3-(phenylamino)isobenzofuran1(3H)-one 60.41 We found that the reaction of 1 with (±)-60 in dichloroethane (c = 0.1 M, molar ratio 1/60 = 1.2:1) in the presence of CPA 62 (10 mol %) and 3 Å MS at 0 °C afforded isoindolinones 61 in good to excellent yields and enantioselectivities (Scheme 22).

relative to the two-component version. This work represents the first example of an asymmetric Ugi reaction involving carboxylic acids as reaction partners. Scheme 24 depicts the hypothetical reaction manifolds for the formation of 61 from 1, 5, and 63. Condensation of 5 and

Scheme 22. Catalytic Enantioselective Two-Component Synthesis of Isoindolinones

Scheme 24. Possible Reaction Pathway of the U-4C-3CR

63 in the presence of CPA 62 would afford iminium salt 64. Nucleophilic addition of isocyanide 1 to 64 would provide nitrilium 65 (step a), which could, in turn, be captured intramolecularly by the carboxylic acid to afford 66. Transannular addition of the aniline nitrogen to the carbonyl group would afford the bridged intermediate 67, which, upon fragmentation, would provide isoindolinone 61 (step b). However, 66 could also undergo tautomerization to isocoumarin 68, which lacks chirality.42 Two scenarios could be proposed that account for the observed enantioselectivity: (a) the Mumm rearrangement (step b) proceeds rapidly at the expense of the imine−enamine tautomerization (kb ≫ kc, kd). In this case, the absolute configuration of the final adduct would be determined by the nucleophilic addition of isocyanide to the iminium salt (step a); (b) imine−enamine tautomerization (step c/d) prevails over the Mumm rearrangement (kc, kd ≫ kb). In this event, the enantioselective protonation of 68 would be the step that determines the stereochemistry. In other word, the enantioselectivity could result from the dynamic kinetic resolution (DKR) of racemic 66.

We subsequently investigated a catalytic enantioselective Ugi four-center, three-component reaction of isocyanides 1 with anilines 5 and 2-formylbenzoic acids 63. As illustrated in Scheme 23, the reaction under the previously optimized conditions provided adducts 61 in excellent yields. Higher catalyst loading (30 mol %) was needed, and the enantioselectivity was slightly decreased 1298


Article

Accounts of Chemical Research A series of control experiments have been performed the results of which clearly indicate that the imine−enamine isomerization was a kinetically competent process (Scheme 24, steps c, d) and that asymmetric protonation of enamine 68 determines the absolute configuration of the isoindolinones 61.

Mei-Xiang Wang received his B.Sc. degree in chemistry from Fudan University, Shanghai. After spending three years at the General Research Institute of Nonferrous Metals (GRINM, Beijing) as a research associate, he joined the Institute of Chemistry, Chinese Academy of Sciences (ICCAS), at Beijing as a research student. He obtained his M.Sc. degree in 1989 and Ph.D. degree in 1992 under the supervision of Professor Z.-T. Huang. In the next 17 years, he worked at ICCAS ranking from assistant professor to associate professor to professor. During 2000−2004, he served as the Director of ICCAS and Center for Molecular Science, Chinese Academy of Sciences. Since May 2009, he has been a professor of chemistry at Tsinghua University in Beijing. His research interests include enantioselective biotransformations using whole cell catalysts, macrocyclic and supramolecular chemistry, and selective organic reactions for the synthesis of natural products and bioactive compounds.

■

CONCLUSIONS AND PERSPECTIVES By capitalizing on the catalyst turnover problem inherent to the LA-catalyzed nucleophilic addition of isocyanide to polarized double bonds, we have advanced a working hypothesis that a chiral LA incapable of forming chelates with bidentate ligands (intermediate or product of the reaction) would be ideal for developing catalytic enantioselective P-3CR and U-4CR. While this design principle turned out to be rewarding in the development of enantioselective P-3CR and related reactions, it failed for the U-4CR. Organocatalysts, including chiral phosphoric acids, dicarboxylic acids, and BOROX, were found to be more efficient in catalyzing the enantioselective addition of isocyanides to imines (hydrazones) and few catalytic enantioselective truncated Ugi-type reactions have been developed. A Brønsted acidcatalyzed Ugi four-center, three-component reaction of isocyanides, anilines, and 2-formyl benzoic acid has also recently been reported. However, the enantioselective Ugi four-component reaction of isocyanides, aldehydes, amines, and carboxylic acids remains unknown. With the experimental results accumulated during the past 15 years, it is reasonable to expect that the solution to this long-awaited challenge will be found in the very near future with organocatalysts representing the most probably way to uncover this odyssey. In view of the tremendous synthetic power of Ugi reaction and its numerous variants in the generation of molecular complexity and diversity, the availability of such an enantioselective process would be an important addition to the toolkit of synthetic and medicinal chemists.

■

Jieping Zhu received his B.Sc. from the Hangzhou Normal University and his M.Sc. from the Lanzhou University (P. R. China) under the guidance of Professor Yulin Li. He obtained his Ph.D. degree from the University Paris XI (France) under the supervision of Professor H.-P. Husson and Professor J.-C. Quirion. After a postdoctoral stay with Professor Sir D. H. R. Barton at the Texas A & M University in USA, he was nominated as Chargé de Recherche in the “Institut de Chimie des Substances Naturelles”, CNRS (France), and was promoted to Director of Research in 2000. He moved to EPFL, Switzerland, in September 2010 as Professor of Chemistry.

■

ACKNOWLEDGMENTS We thank all our co-workers for their invaluable intellectual and experimental contributions. This international collaborative project has been initiated by National Natural Science Foundation of China (NSFC). We thank NSFC, CNRS (France), EPFL (Switzerland), and the Swiss National Science Foundation (SNSF) for generous financial support.

AUTHOR INFORMATION

■

Corresponding Authors

*E-mail: [email protected]. *E-mail: jieping.zhu@epfl.ch.

REFERENCES

(1) Multicomponent Reaction; Zhu, J., Bienaymé, H., Eds.; Wiley-VCH, Weinheim, Germany, 2005. (2) De Graaff, C.; Ruijter, E.; Orru, R. V. A. Recent Developments in Asymmetric Multicomoponent Reactions. Chem. Soc. Rev. 2012, 41, 3969−4009. (3) Van Berkel, S. S.; Bögels, B. G. M.; Wijdeven, M. A.; Westermann, B.; Rutjes, F. J. T. Recent Advances in Asymmetric Isocyanide-Based Multicomponent Reactions. Eur. J. Org. Chem. 2012, 2012, 3543−3559. (4) (a) Zhu, J. Recent Developments in the Isonitrile-Based Multicomponent Synthesis of Heterocycles. Eur. J. Org. Chem. 2003, 2003, 1133−1144. (b) Dömling, A. Recent Developments in Isocyanide Based Multicomponent Reactions in Applied Chemistry. Chem. Rev. 2006, 106, 17−89. (c) El Kaim, L.; Grimaud, L. Beyond the Ugi Reaction: Less Conventional Interactions between Isocyanides and Iminium Species. Tetrahedron 2009, 65, 2153−2171. (d) Wessjohann, L. A.; Rivera, D. G.; Vercillo, O. E. Multiple Multicomponent Macrocyclizations (MiBs): A Strategic Development Toward Macrocycle Diversity. Chem. Rev. 2009, 109, 796−814. (e) Gulevich, A. V.; Zhdanko, A. G.; Orru, R. V. A.; Nenajdenko, V. G. Isocyanoacetate Derivatives: Synthesis, Reactivity, and Application. Chem. Rev. 2010, 110, 5235−5331. (f) Qiu, G.; Ding, Q.; Wu, J. Recent Advances in Isocyanide Insertion Chemistry. Chem. Soc. Rev. 2013, 42, 5257−5269. (g) Tron, G. C. Off the Beaten Track: The Use of Secondary Amines in the Ugi Reaction. Eur. J. Org. Chem. 2013, 2013, 1849−1859. (h) Giustiniano, M.; Basso, A.; Mercalli, V.; Massarotti, A.; Novellino, E.; Tron, G. C.; Zhu, J. To Each His Own: Isonitriles for All Flavors. Functionalized Isocyanides as Valuable Tools in Organic Synthesis. Chem. Soc. Rev. 2017, 46, 1295−1357.

ORCID

De-Xian Wang: 0000-0002-9059-5022 Mei-Xiang Wang: 0000-0001-7112-0657 Jieping Zhu: 0000-0002-8390-6689 Notes

The authors declare no competing financial interest. Biographies Qian Wang received her B.Sc. and M.Sc. degrees from the Lanzhou University (P. R. China) under the guidance of Professor Yulin Li. She got her Ph.D. degree from the Chinese University of Hong Kong under the supervision of Professor Henry N. C. Wong. After several postdoctoral stays in Switzerland and in France, she joined CNRS (France) as a research engineer. In 2010, she moved to the Swiss Federal Institute of Technology Lausanne (EPFL) as a senior research scientist. De-Xian Wang received her B.Sc. degree from the Lanzhou University (P. R. China) and her M.Sc. (2000) and Ph.D. degrees from Hebei University (2003). During 2002 to 2004, she spent 2 years in Inha University (South Korea) as a visiting researcher and then postdoctoral fellow. In 2004, she joined Institute of Chemistry, Chinese Academy of Sciences (ICCAS), at Beijing as an associate professor and was promoted to full professor in 2014. Her research interests cover supramolecular chemistry, molecular recognition, and self-assembly. 1299


Article

Accounts of Chemical Research (5) Passerini, M.; Simone, L.; Sopra, G. Isonitrili (I). Composto Del pIsonitril-Azobenzole Con Acetone Ed Acido Acetica. Gazz. Chim. Ital. 1921, 51, 126−129. (6) Banfi, L.; Riva, R.; Charette, A. B. The Passerini Reaction. Org. React. 2005, 65, 1−140. (7) Ugi, I.; Meyr, R.; Fetzer, U.; Steinbrückner, C. Versuche mit Isonitrilen. Angew. Chem. 1959, 71, 386−386. (8) Dömling, A.; Ugi, I. Multicomponent Reactions with Isocyanides. Angew. Chem., Int. Ed. 2000, 39, 3168−3210. (9) Akritopoulou-Zanze, I. Isocyanide-based Multicomponent Reactions in Drug Discovery. Curr. Opin. Chem. Biol. 2008, 12, 324−331. (10) Banfi, L.; Basso, A.; Moni, L.; Riva, R. The Alternative Route to Enantiopure Multicomponent Reaction Products: Biocatalytic or Organocatalytic Enantioselective Production of Inputs for Multicomponent Reactions. Eur. J. Org. Chem. 2014, 2014, 2005−2015. (11) Michelin, R. A.; Pombeiro, A. J. L.; Guedes da Silva, A. F. Coord. Chem. Rev. 2001, 218, 75−112. (12) For recent examples from our group, see: (a) Odabachian, Y.; Tong, S.; Wang, Q.; Wang, M.-X.; Zhu, J. Zinc Bromide Promoted Coupling of Isonitriles with Carboxylic Acids to 2,4,5-Trisubstituted Oxazoles. Angew. Chem., Int. Ed. 2013, 52, 10878−10882. (b) Tong, S.; Zhao, S.; He, Q.; Wang, M.-X.; Zhu, J.; Wang, Q. Development of New ESIPT Fluorophores Enabled by Y(OTf)3-Catalyzed Reaction of Isocyanides with Thiocarboxylic Acids. Angew. Chem., Int. Ed. 2017, 56, 6599−6603. (13) (a) Maeda, S.; Komagawa, S.; Uchiyama, M.; Morokuma, K. Finding Reaction Pathways for Multicomponent Reactions: The Passerini Reaction is a Four-Component Reaction. Angew. Chem., Int. Ed. 2011, 50, 644−649. (b) Chéron, N.; Ramozzi, R.; El Kaim, L.; Grimaud, L.; Fleurat-Lessard, P. Challenging 50 years of Established Views on Ugi Reaction: A Theoretical Approach. J. Org. Chem. 2012, 77, 1361−1366. (14) Saegusa, T.; Taka-Ishi, N.; Fujii, H. Reaction of Carbonyl Compound with Isocyanide. Tetrahedron 1968, 24, 3795−3798. (15) Katsuyama, A.; Matsuda, A.; Ichikawa, S. Revisited Mechansitic Implications of the Joullié-Ugi Three-Component Reaction. Org. Lett. 2016, 18, 2552−2555. (16) (a) Denmark, S. E.; Fan, Y. The First Catalytic, Asymmetric αAdditions of Isocyanides. Lewis-Base-Catalyzed, Enantioselective Passerini-Type Reactions. J. Am. Chem. Soc. 2003, 125, 7825−7827. (b) Denmark, S. E.; Fan, Y. Catalytic, Enantioselective α-Additions of Isocyanides: Lewis Base Catalyzed Passerini-Type Reactions. J. Org. Chem. 2005, 70, 9667−9676. (17) Denmark, S. E.; Stavenger, R. A. Asymmetric Catalysis of Aldol Reactions with Chiral Lewis Bases. Acc. Chem. Res. 2000, 33, 432−440. (18) Cuny, G.; Gamez-Montaño, R.; Zhu, J. Truncated Diastereoselective Passerini Reaction, a Rapid Construction of Polysubstituted Oxazole and Peptides Having an α-Hydroxy-β-amino acid Component. Tetrahedron 2004, 60, 4879−4885. (19) Wang, S.-X.; Wang, M.-X.; Wang, D.-X.; Zhu, J. Asymmetric Lewis Acid Catalyzed Addition of Isocyanides to Aldehydes - Synthesis of 5-Amino-2-(1-hydroxyalkyl)oxazoles. Eur. J. Org. Chem. 2007, 2007, 4076−4080. (20) Wang, S.-X.; Wang, M.-X.; Wang, D.-X.; Zhu, J. Chiral SalenAluminum Complex as a Catalyst for Enantioselective α-Addition of Isocyanides to Aldehydes: Asymmetric Synthesis of 2-(1-Hydroxyalkyl)-5-aminooxazoles. Org. Lett. 2007, 9, 3615−3618. (21) Yue, T.; Wang, M.-X.; Wang, D.-X.; Masson, G.; Zhu, J. Catalytic Asymmetric Passerini-Type Reaction: Chiral Aluminum-Organophosphate-Catalyzed Enantioselective α-Addition of Isocyanides to Aldehydes. J. Org. Chem. 2009, 74, 8396−8399. (22) Mihara, H.; Xu, Y.; Shepherd, N. E.; Matsunaga, S.; Shibasaki, M. A. Heterobimetallic Ga/Yb-Schiff Base Complex for Catalytic Asymmetric α-Addition of Isocyanides to Aldehydes. J. Am. Chem. Soc. 2009, 131, 8384−8385. (23) Zeng, X.; Ye, K.; Lu, M.; Chua, P. J.; Tan, B.; Zhong, G. Chiral Brønsted Acid Catalyzed Enantioselective Addition of α-Isocyanoacetamides to Aldehydes. Org. Lett. 2010, 12, 2414−2417.

(24) Kusebauch, U.; Beck, B.; Messer, K.; Herdtweck, E.; Dömling, A. Massive Parallel Catalyst Screening: Toward Asymmetric MCRs. Org. Lett. 2003, 5, 4021−4024. (25) Andreana, P. R.; Liu, C. C.; Schreiber, S. L. Stereochemical Control of the Passerini Reaction. Org. Lett. 2004, 6, 4231−4234. (26) Wang, S.-X.; Wang, M.-X.; Wang, D.-X.; Zhu, J. Catalytic Enantioselective Passerini Three-Component Reaction. Angew. Chem., Int. Ed. 2008, 47, 388−391. (27) Katsuki, T. Unique Asymmetric Catalysis of cis-β Metal Complexes of Salen and Its Related Schiff-Base Ligands. Chem. Soc. Rev. 2004, 33, 437−444. (28) Evans, D. A.; Janey, J. M.; Magomedov, N.; Tedrow, J. S. Chiral Salen-Aluminum Complexes as Catalysts for Enantioselective Aldol Reactions of Aldehydes and 5-Alkoxyoxazoles: An Efficient Approach to the Asymmetric Synthesis of syn and anti β-Hydroxy-α-amino Acid Derivatives. Angew. Chem., Int. Ed. 2001, 40, 1884−1888. (29) Wang, S. Ph.D. dissertation (no. 200518003208093), Institute of chemistry, Chinese Academy of Sciences, Beijing, 2008. (30) Ugi, I.; Meyr, R. Isonitrile, V. Erweiterter Anwendungsbereich der Passerini-Reaktion. Chem. Ber. 1961, 94, 2229−2233. (31) Yue, T.; Wang, M.-X.; Wang, D.-X.; Zhu, J. Asymmetric Synthesis of 5-(1-Hydroxyalkyl)tetrazoles by Catalytic Enantioselective PasseriniType Reactions. Angew. Chem., Int. Ed. 2008, 47, 9454−9457. (32) Zhang, J.; Lin, S.-X.; Cheng, D.-J.; Liu, X.-Y.; Tan, B. Phosphoric Acid-Catalyzed Asymmetric Classic Passerini Reaction. J. Am. Chem. Soc. 2015, 137, 14039−14042. (33) Monaco, M. R.; Poladura, B.; de Los Bernardos, M. D.; Leutzsch, M.; Goddard, R.; List, B. Activation of Carboxylic Acids in Asymmetric Organocatalysis. Angew. Chem., Int. Ed. 2014, 53, 7063−7067. (34) Janvier, P.; Sun, X.; Bienaymé, H.; Zhu, J. Ammonium ChloridePromoted Four-Component Synthesis of Pyrrolo[3,4-b]pyridin-5-one. J. Am. Chem. Soc. 2002, 124, 2560−2567. (35) Yue, T.; Wang, M.-X.; Wang, D.-X.; Masson, G.; Zhu, J. Brønsted Acid Catalyzed Enantioselective Three-Component Reaction Involving the α Addition of Isocyanides to Imines. Angew. Chem., Int. Ed. 2009, 48, 6717−6721. (36) List showed that chiral phosphoric acid (TRIP) can catalyze the reaction of benzaldehyde, 4-methoxyaniline, and tBuNC to afford the desired α-amino amide, albeit with low yield (15%) and enantioselectivity (er 59:41). See footnote 12 of: Pan, S. C.; List, B. Catalytic Three-Component Ugi Reaction. Angew. Chem., Int. Ed. 2008, 47, 3622−3625. (37) (a) Bonne, D.; Dekhane, M.; Zhu, J. Modulating the Reactivity of α-Isocyanoacetates: Multicomponent Synthesis of 5-Methoxyoxazoles and Furopyrrolones. Angew. Chem., Int. Ed. 2007, 46, 2485−2488. (b) Lalli, C.; Bouma, M. J.; Bonne, D.; Masson, G.; Zhu, J. Exploiting the Divergent Reactivity of α-Isocyanoacetate: Multicomponent Synthesis of 5-Alkoxyoxazoles and Related Heterocycles. Chem. - Eur. J. 2011, 17, 880−889. (38) Su, Y.; Bouma, M. J.; Alcaraz, L.; Stocks, M.; Furber, M.; Masson, G.; Zhu, J. Organocatalytic Enantioselective One-Pot Four-Component Ugi-Type Multicomponent Reaction for the Synthesis of Epoxytetrahydropyrrolo[3,4-b]pyridin-5-ones. Chem. - Eur. J. 2012, 18, 12624−12627. (39) Hashimoto, T.; Kimura, H.; Kawamata, Y.; Maruoka, K. A Catalytic Asymmetric Ugi-type Reaction with Acyclic Azomethine Imines. Angew. Chem., Int. Ed. 2012, 51, 7279−7281. (40) Zhao, W.; Huang, L.; Guan, Y.; Wulff, W. D. Three-Component Asymmetric Catalytic Ugi Reaction-Concinnity from Diversity by Substrate-Mediated Catalyst Assembly. Angew. Chem., Int. Ed. 2014, 53, 3436−3441. (41) Zhang, Y.; Ao, Y.-F.; Huang, Z.-T.; Wang, D.-X.; Wang, M.-X.; Zhu, J. Chiral Phosphoric Acid Catalyzed Asymmetric Ugi Reaction by Dynamic Kinetic Resolution of the Primary Multicomponent Adduct. Angew. Chem., Int. Ed. 2016, 55, 5282−5285. (42) Faggi, C.; García-Valverde, M.; Marcaccini, S.; Menchi, G. Isolation of Ugi Four-Component Condensation Primary Adducts: A Straightforward Route to Isocoumarins. Org. Lett. 2010, 12, 788−791.

1300


Still Unconquered: Enantioselective Passerini and Ugi

Recommend Documents