Studies on the Synthesis of DMAP Derivatives by Diastereoselective Ugi Reactions

Diastereoselective Ugi reactions of DMAP-based aldehydes with α-amino acids and tert-butyl isocyanide were examined. The reactions of 4-(dimethylamino)-2-pyridine-carboxaldehyde with various α-amino acids afforded 2-substituted DMAP derivatives with low diastereoselectivity. On the contrary, reactions with 4-(dimethylamino)-3-pyridine-carboxaldehyde delivered 3-substituted DMAP derivatives with moderate to high diastereoselectivity. The combination of α-amino acid and DMAP-based aldehyde is thus important to achieve high diastereoselectivity. Kinetic resolution of a secondary alcohol using a chiral DMAP derivative obtained through these reactions was also examined.

In the course of our research, we have been interested in utilizing an asymmetric Ugi reaction for the synthesis of chiral nucleophilic organocatalysts [29]. Among these catalysts, the chiral molecule 4-(dimethylamino)pyridine (DMAP) [30] is known as a versatile catalyst for various asymmetric transformations, such as kinetic resolution of racemic alcohols [31], desymmetrization of anhydrides [32], and inter or intramolecular reactions of oxazolones [33]. However, incorporating a chiral environment in the DMAP structure is still a challenging issue because of long synthetic steps required to obtain optically pure catalysts. We anticipated that by developing an efficient protocol for the diastereoselective Ugi reactions of DMAP-based aldehydes, one-pot synthesis of diverse chiral DMAP structures may be easily carried out simply by changing substrate combinations. Furthermore, highly functionalized and easily tunable chiral DMAP derivatives are attractive as potential highly active, enantioselective catalysts for asymmetric transformation. In this paper, we describe the synthetic studies of diastereoselective Ugi reactions using DMAP-based aldehydes and the use of the Ugi products as chiral nucleophilic catalysts.

Diastereoselective Ugi Reaction of 4-(Dimethylamino)-2-pyridinecarboxaldehyde (1)
We carried out the Ugi reaction of 4-(dimethylamino)-2-pyridinecarboxaldehyde (1), L-valine, and tert-butyl isocyanide in MeOH as the solvent and an external nucleophile (U-5C-4CR). We speculated that the formation of the new stereogenic center could be controlled by L-valine to afford the product as a single diastereomer through simple purification; the product could then be utilized directly as a chiral nucleophilic catalyst (chiral DMAP). Because the Ugi reactions of DMAP-based aldehydes were not reported, we proceeded with the optimization of the Ugi reaction conditions. Because the Ugi reaction is a condensation reaction between organic components, the concentration of substrates may be important for obtaining the desired product in reasonable yield. Thus, we examined different substrate concentrations for the Ugi reaction ( Table 1). The reaction of DMAP-based aldehyde 1, L-valine, and tert-butyl isocyanide in MeOH was carried out at room temperature for 15 h. Lower concentrations (0.1 and 0.2 M) of 1 delivered the desired product 2a in which MeOH was incorporated, even though almost no diastereoselectivity was observed (79% and 84% yields; entries 1 and 2, respectively). At 0.5 M concentration, the reaction was accelerated sufficiently to afford 2a in 98% isolated yield with a 63:37 diastereomeric ratio (d.r.) (entry 3). Higher concentration (1.0 M) afforded a yield slightly inferior to that achieved by 0.5 M concentration (91%, 60:40 d.r.). On the basis of these results, 0.5 M substrate concentration was determined to be optimal for the model reaction.
To improve the diastereoselectivity of the Ugi product, we carried out reactions with various  -amino acids under the aforementioned conditions. The structure of the -amino acid side chain might be important to control the newly formed stereogenic center. As shown in Table 2, we tested various commercially available -amino acids. Reaction with L-t-leucine gave the desired product 2b in 70% isolated yield with a 62:38 d.r., thus suggesting that the sterically congested side chain of the -amino acid did not improve diastereoselectivity (entry 2 vs. 1).

Diastereoselective Ugi Reaction of 4-(Dimethylamino)-3-pyridinecarboxaldehyde (3)
Next, we used 4-(dimethylamino)-3-pyridinecarboxaldehyde (3) in the diastereoselective Ugi reaction. As shown in Table 3, the Ugi reaction of 3 with L-valine and tert-butyl isocyanide in MeOH at 20 °C afforded the Ugi product 4a in 37% isolated yield with an 84:16 d.r. The low yield of 4a is assumed to be due to the presence of an adjacent bulky dimethyl amino group, which encumbers the formation of the imine or iminium species derived from 3 and L-valine. However, because of the reaction of DMAP-based aldehyde 3 and L-valine, large dimethyl amino group might fix imine or iminium configuration (E or Z isomer) to avoid steric repulsion against dimethyl amino group. Accordingly, a high d.r. could be observed when 3 was used instead 1. To enhance the formation of the imine or iminium species, we next carried out the reactions under various reaction temperatures. The reaction at 40 °C proceeded to 79% conversion and afforded 4a in 53% yield with a 90:10 d.r. (entry 2). Higher conversion (>90%) was achieved at 50 and 60 °C delivering 4a in 57% and 48% isolated yields with 89:11 and 85:15 d.r., respectively. According to these results, the reaction at 50 °C is preferred considering the yield and diastereoselectivity of the product. Although most of the aldehyde was consumed (92% conversion; entry 3), the isolated yield of 4a remained moderate. Owing to this, we considered the determination of appropriate substrate ratio of the reaction components (aldehyde,  -amino acid, and tert-butyl isocyanide) to improve the efficiency of the reaction. To address the aforementioned issue, the substrate ratio was screened under the previously mentioned conditions (Table 4). Although all reactions proceeded in >90% conversion with excess L-valine (1.3 and 1.5 equiv.; entries 2 and 3, respectively) and excess tert-butyl isocyanide (1.3 and 1.5 equiv.; entries 4 and 5), the isolated yield was almost same as that of the control reaction (entry 1). The reason for the relatively low isolated yield of the Ugi products with respect to consumption of the starting aldehyde is unclear. Considering a relatively effective diastereoselective Ugi reaction of DMAP-based aldehyde 3, various -amino acids were investigated under the optimized reaction conditions (Table 5). Reactions with -amino acids bearing alkyl side chains led to the corresponding products in moderate yield with good to high diastereoselectivity (45%-63%; 59:41-93:7 d.r; entries 1 and 4-8). Ciufolini reported that Ugi reactions of aromatic aldehydes, -amino acids and t-BuNC in MeOH with TiCl 4 as a catalyst showed improved isolated yields of Ugi product [22]. TiCl 4 was thus added to the reaction mixture, however the result was almost identical to that obtained under the uncatalyzed reaction (entry 2 vs. 1). Furthermore, using MgSO 4 as a dehydrating agent resulted in slightly decreasing in isolated yield and diastereoselectivity (entry 3 vs. 1). Side chains having a heteroatom also showed the same results as those of their alkyl counterparts (entries 9-16). It is surprising that the d.r. of the Ugi products was not dramatically changed by modifying the structure of -amino acid. Furthermore, the reaction proceeding through the cyclic iminium intermediate derived from L-proline and aldehyde 3 did not improve both yield and diastereoselectivity (51% yield; 74:26 d.r.; entry 17). The stereochemistry of major diastereomer 4a was determined by X-ray structure analysis, showing R configuration at newly formed stereogenic center ( Figure 1). The absolute configuration of the major product was different from related reaction reported by Ciufolini [22]. Table 5. Ugi reaction of 3 with various -amino acids.

Kinetic Resolution of a Racemic Alcohol by Chiral DMAP Derivatives
Next, we explored the possibility of using Ugi products as chiral nucleophilic catalysts. The kinetic resolution of racemic alcohol was selected as a model study. A major diastereomer of Ugi product 4a, which can be easily obtained by column chromatography, was utilized as the catalyst in toluene at −60 °C (Scheme 1). The selectivity factor [34], which was estimated by ee of acetylated product 6 and unreacted alcohol 5 [35], was indicated to be 2.33. It was noted that the Ugi product 4a was capable of catalyzing the acylation reaction, although the selectivity was not satisfied at the moment. Further study to develop more efficient catalysts is now under way.

General
All melting points were determined using a Yanaco micro melting point apparatus MP-S3 and are uncorrected. Solvents were generally distilled and dried by standard literature procedures prior to use. The IR spectra were recorded on a JASCO FT/IR-4100 spectrometer. NMR spectra were recorded on a Varian VNMRS-400 spectrometer at SC-NMR Laboratory (Okayama University), operating at 400 MHz for 1 H-NMR and 100 MHz for 13 C-NMR. Chemical shifts in CDCl 3 were reported in the  scale relative to CHCl 3 (7.26 ppm) as an internal reference for 1 H-NMR. For 13 C-NMR, chemical shifts were reported in the scale relative to CHCl 3 (77.0 ppm) as an internal reference. Column chromatography was performed with silica gel 60 N (spherical, neutral, 40-50 m) purchased from Kanto Chemical. Optical rotations were measured on a Horiba Model SEPA-300 High-sensitive polarimeter. FAB mass spectra (for HRMS) were measured on a JEOL JMS-700 MStation at the Mass Spectrometry Facility (Okayama University). The enantiomeric excess (ee) was determined by HPLC analysis. HPLC was performed on Shimadzu HPLC systems consisting of the following: pump, LC-10AD; detector, SPD-10A, 254 nm; column, Daicel Chiracel OD-H; mobile phase, hexane/2-propanol. (1) To a suspension of L-valine (23.1 mg, 0.20 mmol) and aldehyde 1 (36.6 mg, 0.24 mmol) in dry methanol (0.4 mL) in a screw-cap test tube, tert-butyl isocyanide (17.2 mg, 0.21 mmol) was added and the reaction mixture was stirred for 15 h at room temperature. The solvent was evaporated in vacuo and the resulting residue was purified by silica gel column chromatography (EtOAc/Et 3 N = 97:

Procedure for the Synthesis of 2h
To the mixture of L-proline (33.9 mg, 0.29 mmol) and aldehyde 1 (39.8 mg, 0.27 mmol) in dry methanol (0.4 mL) in a screw-cap test tube, tert-butyl isocyanide (30.4 L, 0.27 mmol) was added and the reaction mixture was stirred for 24 h at room temperature. The solvent was evaporated in vacuo. To a solution of the crude product in MeOH (4 mL) was added NaBH 4 (31.6 mg, 0.84 mmol) at 0 °C owing to reduction of unreacted aldehyde 1. The reaction mixture was stirred at the same temperature for one hour before being quenched with saturated aq. NH 4 Cl (4 mL). The resulting solution was warmed to room temperature and extracted with Et 2 O (3 × 4 mL). The combined organic phase was washed with brine (15 mL), dried over MgSO 4 , filtered, and concentrated in vacuo to give (1SR) N-(1-(N-tert-butylcarbamoyl)-1-(4-(dimethylamino)pyridyn-2-yl)methyl)-L-proline methyl ester (2h) as a pale yellow oil (

Kinetic Resolution of a Racemic Alcohol by Chiral DMAP Derivative
To a solution of major diastereomer of 4a (3.6 mg, 0.01 mmol), 1-phenylethanol (5, 25.4 L, 0.21 mmol) and triethylamine (21.0 L, 0.15 mmol) in toluene (0.4 mL) at -60 °C, was added acetic anhydride (14.0 L, 0.15 mmol) and the reaction mixture was stirred at the same temperature for 15 h. The reaction was quenched with methanol and concentrated in vacuo. The resulting residue was filtered through a short plug of silica gel (hexane/Et 2 O = 3:1, v/v), affording a 28% ee of (S) acetate and a 23% ee of unreacted (R) alcohol at 43% conversion determined by 1 (Table 6).

Data Reduction
Of the 31,451 reflections that were collected, 4,665 were unique (R int = 0.0603); equivalent reflections were merged. Data were collected and processed using CrystalClear (Rigaku) [36].

Structure Solution and Refinement
The structure was solved by direct methods [37] and expanded using Fourier techniques. The non-hydrogen atoms were refined anisotropically. Hydrogen atoms were refined using the riding model. The final cycle of full-matrix least-squares refinement [38] on F [39] was based on 4665 observed reflections and 236 variable parameters and converged (largest parameter shift was 0.00 times its esd) with unweighted and weighted agreement factors of: R1 = S ||Fo| − |Fc|| / S |Fo| = 0.0471 wR2 = [S(w (Fo 2 -Fc 2 ) 2 )/ S w(Fo 2 ) 2 ] 1/2 = 0.1109 The standard deviation of an observation of unit weight [40] was 1.50. Unit weights were used. The maximum and minimum peaks on the final difference Fourier map corresponded to 1.48 and −0.80 e − /Å 3 , respectively. The absolute structure was deduced based on Flack parameter, −2(4), using 1983 Friedel pairs [41].
Neutral atom scattering factors were taken from Cromer and Waber [42]. Anomalous dispersion effects were included in Fcalc [43]; the values for Df' and Df" were those of Creagh and McAuley [44]. The values for the mass attenuation coefficients are those of Creagh and Hubbell [45]. All calculations were performed using the Yadokari-XG 2009 [46] crystallographic software package except for refinement, which was performed using SHELXL-97 [47].

Conclusions
We have studied the diastereoselective Ugi reactions of DMAP-based aldehydes with -amino acids and tert-butyl isocyanide. The reactions of 4-(dimethylamino)-2-pyridinecarboxaldehyde (1) with various -amino acids as a chiral source proceeded to afford the desired Ugi products 2a-h in moderate to high yield (19%-98%) with low diastereoselectivity ratio (50:50-63:37; d.r.), even though various -amino acids structures were investigated. On the other hand, the reactions of 4-(dimethylamino)-3-pyridinecarboxaldehyde (3) with various -amino acids delivered the desired Ugi products 4a-o in moderate yield (18%-63%) with high diastereoselectivity (up to 93:7 d.r.). The fact that the combination of α-amino acid and 3-formyl DMAP is critical to achieve high diastereoselectivity is noteworthy. We also demonstrated that the kinetic resolution of racemic alcohol 5 using a major diastereomer of the Ugi product 4a as a catalyst afforded enantioenriched acetylated product 6 and unreacted alcohol 5 with a selectivity factor of 2.33. The result indicated that the Ugi products are potential chiral nucleophilic catalysts. Further optimization of the catalyst structure and the application of these catalysts to other important asymmetric transformations are currently in progress.