Environmental Modulation of Chiral Prolinamide Catalysts for Stereodivergent Conjugate Addition

Abstract

Synthetic chiral catalysts generally rely on proximal functional groups or ligands for chiral induction. Enzymes often employ environmental chirality to achieve stereoselectivity. Environmentally controlled catalysis has benefits such as size and shape selectivity but is underexplored by chemists. We here report molecularly imprinted nanoparticles (MINPs) that utilized their environmental chirality to either augment or reverse the intrinsic selectivity of a chiral prolinamide cofactor. The latter ability allowed the catalyst to produce products otherwise disfavored in the conjugate addition of aldehyde to nitroalkene. The catalysis occurred in water at room temperature and afforded γ-nitroaldehydes with excellent yields (up to 94%) and ee (>90% in most cases). Up to 25:1 syn/anti and 1:6 syn/anti ratios were achieved through a combination of catalyst-derived and environmentally enabled selectivity. The high enantioselectivity of the MINP also made it possible for racemic catalysts to perform asymmetric catalysis, with up to 80% ee for the conjugate addition.

Graphical Abstract

INTRODUCTION

Asymmetric catalysis is vital to the preparation of complex organic molecules and functional materials. The vast majority of synthetic catalysts developed for this purpose, whether transition metal-based or organic, rely on chiral elements in close proximity to the reactive center for chiral induction [1, 2]. As the substrate interacts with the chiral ligand(s) or functional groups on the catalyst over the course of the reaction, different reaction paths may be distinguished if these interactions are sufficiently different in energy, leading to enantio- or diastereoselectivity. Proximity of the chiral elements helps efficient transmission of the differential in the interaction energy to stereoselectivity of the reactive center.

Enzymes use their active site formed from folded peptide chains for chiral induction. Since all natural amino acids except glycine are chiral, the active site is by definition a chiral microenvironment. In such environmentally controlled reactions, residues far away from the catalytic center could also contribute strongly to the stereoselectivity [3].

Using environmental chirality for stereochemical control has certain benefits such as size- and shape-selectivity [4] but is a skill far less mastered by chemists [5, 6]. In the literature, environmental chirality is nearly always used to augment the stereoselectivity of chiral catalysts. Examples include metal–organic cages (MOCs) and metal–organic frameworks (MOFs) [7], which rely on chiral catalysts incorporated in structure for chiral induction [8–10] but employ the organic framework around the catalysts to enhance the stereoselectivity [11–13].

A chiral environment can be generated readily in a cross-linked polymer network through the technique of molecular imprinting [14–16], as long as chiral template molecules are used. Gagné and co-workers synthesized molecularly imprinted polymers (MIPs) containing a chiral bisphosphine platinum complex in the imprinted site [17]. A chiral cavity was also generated through imprinting a BINOL ligand near the metal center. The enantioselectivity of the catalytic reaction was found to be controlled by the chiral phosphine ligand on Pt instead of the chiral nanospace nearby. Tada and colleagues used molecular imprinting to create a chiral cavity near chiral ruthenium complexes and successfully exploited the cavity for the enhancement of the enantioselectivity of the Ru catalyst [18–20].

Our group in recent years developed molecularly imprinted nanoparticles (MINPs) by templated polymerization within the nanoconfined space of micelles [21]. MINPs bear great resemblance to enzymes in their water-solubility, hydrophilic surface, and substrate-tailored active sites in the hydrophobic core of the cross-linked micelle [22–25]. In this work, we show that a chiral cavity near the catalytic center of MINP could be used not only to enhance the stereoselectivity of the catalyst but also to reverse it. The latter ability allowed us to perform stereodivergent conjugate addition of aldehyde to nitroalkene. A stereodivergent synthesis of γ-nitroaldehyde can afford different stereoisomers of γ-amino acids that form foldamers with interesting properties [26–28]. Environmental control is often used for site-isolation of incompatible catalysts [29–31]. Through environmental modulation, we were able to “coerce” the same catalysts into producing otherwise disfavored products, without new catalyst development.

RESULTS AND DISCUSSION

Design and Synthesis of MINP to Control Conjugate Addition.

Conjugate addition of aldehyde to nitroolefin is a highly important reaction because the γ-nitroaldehyde products are versatile synthetic building blocks [32–34]. Prolinamide derivatives are frequently employed to catalyze this reaction and generally afford syn-products [35–37]. Preference for syn-addition is derived from the approaching of (E)-nitroalkene and the preferred s-trans (E)-enamine intermediate from their Si-faces (Scheme 1) [38–41]. anti-Addition has been a long standing challenge in this reaction, with very few examples reported, via enforced Z nitroalkene [42], enforced Z enamine intermediate [43], a C2-symmetric secondary amine [38], and, most recently, a C_δ-substituted pyrrolidine-containing peptidic catalyst [39, 40].

Scheme 1.

Formation of the syn- and anti-product from the s-trans and s-cis (E)-enamine intermediate, respectively. The bonds that define the s-trans and s-cis conformation are shown in red.

In principle, an anti-product can be produced if the more stable s-trans conformer of the enamine intermediate is shifted to the less stable s-cis (Scheme 1). In the literature, such a conformational change is achieved through modifying the prolinamide catalyst to destabilize the s-trans conformer [39]. Our hypothesis is that a suitable microenvironment can also shift the conformational equilibrium, without changing the catalyst. After all, conformation of a molecule is highly dependent on its environment, as amply demonstrated by proteins whose secondary and tertiary structures are strongly impacted by solvents, microenvironments, and also binding partners [44–46]. Our previous work has also shown that the microenvironment around a molecule can strongly impact its conformation, even overriding the intrinsic conformational preferences in certain cases [47–49].

Molecular imprinting is a powerful method to create chiral binding sites [50]. To create a microenvironment to modulate the conformation of a prolinamide catalyst (and ultimately to control its stereoselectivity), we designed template molecule (S,S)-1 for the conjugate addition between aldehyde and (E)-1-nitro-2-phenylethene or β-nitrostyrene. This template molecule has two key units, as shown in Scheme 2: the green-colored moiety is used to create a catalyst-binding pocket through molecular imprinting (i.e., the primary chiral site for the prolinamide catalyst); the magenta-colored substructure is the space-holder for a secondary chiral site, which will be used to accommodate the (white-colored) aldehyde and the (magenta-colored) nitroalkene during the catalysis. The prolinamide catalyst is expected to react with the aldehyde to afford an enamine intermediate. Proximity of the enamine and β-nitrostyrene should help their conjugate addition but the resulting ionic intermediate (i.e., iminium cation) is unlikely to prefer the overall hydrophobic active site that has been imprinted against a neutral, nonpolar template. In addition, the aldehyde, being in large excess in our reactions (vide infra), could displace the iminium from the active site. Release of the iminium intermediate into the aqueous solution helps its hydrolysis to free the prolinamide catalyst, which by its nonpolar nature will want to enter the primary chiral site and then can start another round of reaction.

Scheme 2.

Preparation of chiral MINP-(S, S)-1 with a bound aldehyde and (E)-(2-nitrovinyl)benzene for conjugate addition, showing a schematic illustration of the cross-linked structure around the catalytic active site.

Micellar imprinting of (S,S)-1 afforded MINP-(S,S)-1 (Scheme 2). The procedure has been reported [21] and is detailed in the Supplementary Material. Briefly, the template molecule, divinylbenzene (DVB), and 2,2-dimethoxy-2-phenylacetophenone (DMPA, a photoinitiator) are solubilized by the micelles of cross-linkable surfactant 3. The micelles are cross-linked by the click reaction with diazide 4 and decorated with a layer of hydrophilic ligand 5 by a second round of click reaction. UV-triggered free-radical polymerization then cross-links 3 through its methacrylate and DVB around the template. The templates are removed by precipitation of the MINPs from acetone and washing the resulting precipitate with solvents (see Supporting Information for details). As shown previously, MINPs are typically 5 nm in size and soluble in water due to their polycationic nature and hydrophilic surface ligands [21].

Effects of Environmental Chirality on Enantioselectivity.

To understand the effects of environmental chirality of MINP-(S, S)-1, we first examined a benchmark conjugate addition: the reaction between isobutyraldehyde and β-nitrostyrene creates a single chiral center in the Michael product (7) and a characteristic quaternary carbon next to the carbonyl (Table 1).

Table 1.

Conjugate addition of isobutyraldehyde and β-nitrostyrene catalyzed by prolinamide 2 and MINP.^a

entry	catalyst	co-catalyst	yield (%)	ee (%)
1	10% (S)-2	none	0	--
2	10% (S)-2	NINPb	0	--
3	10% (S)-2	5% MINP-(S, S)-1	72	66 (R)
4	10% (S)-2	10% MINP-(S, S)-1	86	71 (R)
5	10% (S)-2	15% MINP-(S, S)-1	91	94 (R)
6	10% (S)-2	15% MINP-(S, R)-1	69	38 (R)
7	10% (S)-2	15% MINP-(R, R)-1	0	--
8	10% (±)-2	15% MINP-(S, S)-1	66	80 (R)
9	10% (R)-2	15% MINP-(R, R)-1	92	93 (S)
10	10% (S)-2	5% MINP-(S, S)-6	37	41 (R)
11	10% (S)-2	10% MINP-(S, S)-6	41	63 (R)
12	10% (S)-2	15% MINP-(S, S)-6	53	86 (R)
13	10% (S)-2	15% MINP-(S, R)-6	31	32 (R)
14	10% (S)-2	15% MINP-(R, R)-6	0	--
15	10% (R)-2	15% MINP-(R, R)-6	57	84 (S)
16	10% (±)-2	15% MINP-(S, S)-6	31	52 (R)

^aThe reactions were performed with 2.4 mM β-nitrostyrene and 20 mM isobutyraldehyde in 1 mL water at room temperature for 24 h. Yields were determined by ¹H NMR spectroscopy using CH₂Br₂ as the internal standard and enantiomeric excess (ee) was determined by chiral HPLC.

^bNonimprinted nanoparticles (NINPs) were prepared without any template.

Prolinamide (S)-2 at 10 mol % was completely inactive for the conjugate addition in water in the absence of MINP or in the presence of nonimprinted nanoparticles (NINPs) prepared without the templates (Table 1, entries 1 and 2). The inactivity was most likely caused by the poor solubility of the hydrophobic guest in water.

At the same catalyst loading (10 mol % of (S)-2 to β-nitrostyrene), an increase in the amount of MINP-(S, S)-1 brought a steady increase to the yield of the reaction (entries 3–5). The catalyst (S)-2 is inactive on its own in water and any reaction thus must have taken place with the help of the co-catalyst. Although the MINP binds (S)-2 strongly (vide infra), it has to overcome a crystallization energy to bring an insoluble catalyst into solution and then to the imprinted site. A slight excess of the MINP co-catalyst thus is reasonable for optimal results. Interestingly, the ee of the reaction also increased steadily with higher concentrations of the MINP modulator (entries 3–5). As shown in Scheme 2, to enjoy the environmental benefit for the conjugate addition, the reaction between the enamine intermediate and β-nitrostyrene must take place within the imprinted site. Even after solubilization of the prolinamide catalyst, binding of the enamine intermediate by the MINP modulator is a dynamic process. A higher concentration of MINP helps to keep the enamine intermediate inside the imprinted site and is expected to help the stereoselectivity as well.

Acid co-catalysts such as benzoic acid or p-nitrophenol are generally needed for the conjugate addition, because both enamine formation and decomposition of the iminium intermediate are catalyzed by acids [35–37]. In our case, the large amounts of water present (as solvent) enabled us to eliminate the acid co-catalyst altogether. With 15% of the MINP co-catalyst, the reaction gave a 91% yield and 94% ee (entry 5).

Whereas the secondary chiral site derived from the (magenta-colored) (S)-1-(1-naphthyl)ethylamide moiety of (S, S)-1 made (S)-2 an excellent chiral catalyst (entry 5), inverting the secondary chiral site through template (S, R)-1 lowered the ee of the product dramatically, from 94 to 38% (entry 6). Clearly, the secondary site was critical to the chiral induction, suggesting it is possible to employ it to enhance or reverse the diastereoselectivity of the conjugate addition (vide infra).

MINP-(R, R)-1 compounds the wrong environmental chirality of MINP-(S, R)-1 with a mismatched primary chiral site for the catalyst binding. Since the prolinamide catalyst is inactive outside the MINP pocket (entries 1 and 2), the lack of reaction in entry 7 is not surprising. Consistent with the picture, MINP-(S, S)-1 enabled racemic (±)-2 to perform asymmetric catalysis, affording a 66% yield and an impressive 80% ee (entry 8). Apparently, the MINP co-catalyst was able to pick the matched (S)-2 from the racemic catalysts while unbound (S)-2 remained inactive. As expected, if the catalyst and both chiral sites in the MINP co-catalyst were all inverted, an excellent yield and ee could be obtained again, for the opposite enantiomer of the product (entry 9).

Molecular imprinting allowed us to vary the size of the secondary chiral site easily, using templates 6 prepared from 1-phenylethylamine (Scheme 1). The different versions of MINP-6 displayed a similar trend in their influence on the conjugate addition as MINP-1, but the yields and ee were consistently lower (Table 1, entries 10–16). Most likely, the smaller secondary chiral site in MINP-6 had difficulty accommodating both the enamine and β-nitrostyrene simultaneously.

Additional support for the environmentally controlled catalysis came from isothermal titration calorimetry (ITC). MINP-(S, S)-1 bound its matched catalyst (S)-2 strongly in aqueous buffer (which contained 5% DMSO to dissolve the prolinamide catalyst), with a binding constant of K_a = 152 × 10⁴ M-¹ (Table 2, entry 1). The enantiomer (S)-2 was bound with 1/68th of the strength (entry 3). The high enantioselectivity in the binding is consistent with the good ee obtained from the racemic catalyst using MINP-(S,S)-1 as the co-catalyst (Table 1, entry 8). Interestingly, inverting the secondary chiral site also reduced the binding of the primary site for the catalyst as well, as shown in entry 3. Possibly, the two sites, being in close proximity, jointly controlled the binding of the catalyst even though the primary site was mainly responsible (also see Table 2, entries 4–6).

Table 2.

ITC binding data for prolinamide (S/R)-2 and chiral ammonium salts by chiral MINPs-1.^a

entry	MINP	guest	Ka (×104 M−1)	ΔG (kcal/mol)	N b
1	MINP-(S, S)-1	(S)-2	152 ± 8.8	−8.43	1.01 ± 0.08
2	MINP-(S, S)-1	(R)-2	2.22 ± 0.17	−5.92	1.30 ± 0.02
3	MINP-(S, R)-1	(S)-2	111 ± 4.15	−8.24	1.11 ± 0.07
4	MINP-(S, R)-1	(R)-2	2.59 ± 0.89	−6.02	0.97 ± 0.07
5	MINP-(R, R)-1	(S)-2	2.15 ± 0.12	−5.91	0.95 ± 0.04
6	MINP-(R, R)-1	(R)-2	164 ± 14.4	−8.47	1.19 ± 0.02
7	NINP	(S)-2	<0.005	--c	--c
8	MINP-(S, S)-1	(S)-NEAC	86.1 ± 4.25	−8.09	1.15 ± 0.01
9	MINP-(S, S)-1	(R)-NEAC	2.42 ± 0.12	−5.98	1.42 ± 0.16
10	MINP-(S, R)-1	(R)-NEAC	84.0 ± 3.01	−8.08	1.12 ± 0.07
11	MINP-(R, R)-1	(R)-NEAC	85.2 ± 5.53	−8.08	1.08 ± 0.01
12	NINP	(S)-NEAC	<0.005	--c	--c
13	MINP-(S, S)-6	(S)-2	58.0 ± 2.68	−7.86	1.01 ± 0.08
14	MINP-(S, S)-6	(R)-2	0.85 ± 0.07	−5.35	1.10 ± 0.03
15	MINP-(S, R)-6	(S)-2	44.8 ± 0.97	−7.70	1.05 ± 0.03
16	MINP-(S, R)-6	(R)-2	0.78 ± 0.06	−5.31	1.22 ± 0.06
17	MINP-(R, R)-6	(S)-2	0.82 ± 0.03	−5.33	1.14 ± 0.03
18	MINP-(R, R)-6	(R)-2	61.8 ± 6.60	−7.89	1.00 ± 0.02
19	MINP-(S, S)-6	(S)-PEAC	42.0 ± 2.25	−7.67	1.15 ± 0.02
20	MINP-(S, S)-6	(R)-PEAC	1.82 ± 0.84	−5.81	1.86 ± 0.12
21	MINP-(S, R)-6	(R)-PEAC	44.6 ± 0.18	−7.70	0.98 ± 0.01
22	MINP-(R, R)-6	(R)-PEAC	44.0 ± 2.91	−7.69	1.26 ± 0.02
23	NINP	(S)-PEAC	<0.005	--c	--c

^aTitrations were performed in duplicates in 10 mM HEPES buffer pH 7.4 at 298 K and the errors between the runs were <10% (Figures S7–S11). 5% (v/v) DMSO was added to the solution for titrations with the prolinamide 2, which is insoluble in aqueous buffer.

^bN is the average number of binding sites per nanoparticle determined by ITC.

^cBinding was extremely weak and thus −ΔG and N are not listed.

The secondary binding site was created from imprinting of the 1-(1-naphthyl)ethylamide moiety of the template. Indeed, MINP-(S,S)-1 was found to bind its matched guest, (S)-1-(1-naphthyl)ethylammonium chloride (NEAC) strongly, with a K_a of 86 × 10⁴ M⁻¹ (Table 2, entry 8). The binding for its enantiomer was reduced by a factor of 36 (entry 9), indicating a strong imprinting effect for the secondary chiral site as well. The primary chiral site seemed to have little effect on the binding of the secondary site, as long as the guest had the matched stereochemistry (entries 10 and 11). As expected, NINP showed little binding for either the catalyst or (S)-NEAC (entries 7 and 12).

We also performed similar ITC studies for MINP-(S, S)-6, and observed similar trends (entries 13–21) as those displayed by MINP-(S, S)-1.

Effects of Environmental Chirality on Diastereoselectivity.

Encouraged by the strong environmental effect in the enantioselective conjugate addition, we began to investigate the diastereoselectivity of the prolinamide@MINP complex, using a number of aldehydes mostly differing in chain length. The reaction affords two chiral centers in the product and, as discussed earlier, favors syn-addition usually (Table 3). For easier comparison, we also illustrate the yields and d.r. (syn/anti selectivity) of the linear aldehydes in Figure 1, using vertical bars and curves, respectively.

Table 3.

syn-Addition of different aldehydes to β-nitrostyrene catalyzed by prolinamide (S)-2 and MINPs.^a

entry	R1	co-catalyst	major product	yield (%)	syn/anti	eeb
1	Me	MINP-(S, S)-1	(2S,3S)-8a	0	--	--
2	Me	MINP-(S, S)-1	(2S,3S)-8a	93	9/1	91
3	Et	MINP-(S, S)-1	(2S,3S)-8b	94	20/1	92
4	Pr	MINP-(S, S)-1	(2S,3S)-8c	94	22/1	94
5	Bu	MINP-(S, S)-1	(2S,3S)-8d	87	10/1	90
6	Pe	MINP-(S, S)-1	(2S,3S)-8e	53	3/1	88
7	iPr	MINP-(S, S)-1	(2S,3S)-8f	69	10/1	98
8	Me	MINP-(S, S)-6	(2S,3S)-8a	79	9/1	87
9	Et	MINP-(S, S)-6	(2S,3S)-8b	71	25/1	84
10	Pr	MINP-(S, S)-6	(2S,3S)-8c	72	25/1	82
11	Bu	MINP-(S, S)-6	(2S,3S)-8d	62	10/1	81
12	Pe	MINP-(S, S)-6	(2S,3S)-8e	36	3/1	70
13	iPr	MINP-(S, S)-6	(2S,3S)-8f	38	12/1	83

^aThe reactions were performed with 2.4 mM β-nitrostyrene and 20 mM aldehyde in 1 mL water at room temperature for 24 h. Yields and syn/anti selectivities were determined by ¹H NMR spectroscopy using CH₂Br₂ as the internal standard and ee was determined by chiral HPLC.

^bThe enantiomeric excess for the major (2R,3S) product.

Figure 1.

Comparison of yield and d.r. of the conjugate addition of different aldehydes to β-nitrostyrene catalyzed by (S)-2 and MINP(S, S)-1 or MINP(S, S)-6.

The catalyst remained inactive in the absence of MINP (Table 3, entry 1). With 10 mol % (S)-2 and 15 mol % MINP-(S, S)-1, the yield of the Michael product was 93–94% for propanal, butanal, and pentanal (entries 2–4). As the chain length of the aldehyde increased further, the yield decreased, to 87% for hexanal and 53% for heptanal (entries 5 and 6). Branching in the aldehyde also reduced the yield (entry 7). This trend could be understood from the size and geometry of the aldehydes. The aldehyde was used in excess in our reactions and was covalently attached to the MINP-bound prolinamide via enamine formation during catalysis. Thus, it should have preference to occupy the secondary chiral site. A long or branched aldehyde, most likely, can crowd out β-nitrostyrene, making it more difficult to access the enamine intermediate.

More interestingly, the secondary chiral site also had a strong impact on the diastereoselectivity of the reaction. The best d.r. was found with an intermediate chain length in the aldehyde, with too short or two long a chain giving lower stereoselectivity (Figure 1a). The 22–25:1 d.r. achieved by (S)-2 in the presence of the MINP co-catalyst compares favorably with that achieved by the very closely related L-prolinamide 9 reported in the literature (91% yield, 8:1 syn/anti ratio, 72% ee, with 10% catalyst + 10% benzoic acid in dichloromethane, 24 h at room temperature) [35].

What could be the possible reason for the chain length-dependency of the environmentally derived chiral induction? MINP-(S, S)-6 has a smaller secondary chiral site than MINP-(S, S)-1 and yet the highest d.r. was also observed with butanal and pentanal (Figure 1a). Thus, “packing efficiency” of the enamine and β-nitrostyrene in the chiral site could not be the dominant factor in the chiral induction. Assuming the enhancement of the d.r. was a result of the secondary chiral site enforcing the correct conformation (s-trans for the current examples) for the enamine intermediate, the best conformational control for the enamine happened with an intermediate chain length.

Interestingly, while the d.r. for MINP-(S, S)-6 was at least as good as MINP-(S, S)-1, the yields of the Michael products decreased consistently as the secondary chiral site became smaller (Figure 1a, blue versus red bars). In the conjugate addition, the aldehyde is covalently attached to the hydrophobically anchored (S)-2 through enamine formation. β-Nitrostyrene, on the other hand, is noncovalently bound by the secondary chiral site. A small secondary chiral pocket would have difficulty accommodating both enamine and the Michael acceptor, thus decreasing the yield, similarly as what happened to the longest aldehyde in the larger MINP-(S, S)-1. It is also quite likely that some differences in intrinsic reactivity exist among the aldehydes.

As discussed earlier, it has been a long-term challenge to have anti-selective conjugate addition between aldehyde and nitroalkene and only one general method is available currently [39]. Since the secondary chiral site strongly enhanced the diastereoselectivity in the syn-addition (from 3/1 for heptanal to 20–25:1 for butanal and pentanal), it might be possible for the chiral site to override the intrinsic stereoselectivity altogether.

Figure 1b shows the yields and d.r. of the conjugate addition, using (S)-2 and MINP-(S,R)-1 or MINP-(S,R)-6, respectively. The secondary chiral site was thus inverted from the examples above while everything else kept the same. To our delight, the reactions now favored the anti-addition products, with up to 6:1 ratio.

The most interesting observation upon comparing Figure 1a and Figure 1b is the consistent peaking of the d.r. curves at butanal and pentanal, across all four catalyst/MINP combinations—note that the d.r. curves for MINP-(S,R)-1 and MINP-(S,R)-6 overlap completely in Figure 1b. Since the syn- and anti-products form from specific conformers of the (E)-enamine intermediate (Scheme 1), the results confirm that the strongest environmental control of conformation happens at these medium-length aldehydes, with each “enantiomer” of the secondary chiral site promoting the s-trans and s-cis conformation, respectively.

As shown in Table 4, good enantioselectivity was also obtained for the anti-products, ranging from 85 to 92% for MINP-(S,R)-1 and from 79 to 92% for MINP-(S,R)-6.

Table 4.

anti-Addition of different aldehydes to β-nitrostyrene catalyzed by prolinamide (S)-2 and MINPs.^a

entry	R1	co-catalyst	major product	yield (%)	syn/anti	eeb
1	Me	MINP-(S, R)-1	(2S,3S)-8a	88	1/4	85
2	Et	MINP-(S, R)-1	(2S,3S)-8b	90	1/6	86
3	Pr	MINP-(S, R)-1	(2S,3S)-8c	91	1/6	90
4	Bu	MINP-(S, R)-1	(2S,3S)-8d	76	1/3	92
5	Pe	MINP-(S, R)-1	(2S,3S)-8e	49	1/3	92
6	Me	MINP-(S, R)-6	(2S,3S)-8a	71	1/4	79
7	Et	MINP-(S, R)-6	(2S,3S)-8b	66	1/6	88
8	Pr	MINP-(S, R)-6	(2S,3S)-8c	70	1/6	90
9	Bu	MINP-(S, R)-6	(2S,3S)-8d	54	1/3	92
10	Pe	MINP-(S, R)-6	(2S,3S)-8e	31	1/3	92

^aThe reactions were performed with 2.4 mM β-nitrostyrene and 20 mM aldehyde in 1 mL water at room temperature for 24 h. Yields and syn/anti selectivities were determined by ¹H NMR spectroscopy using CH₂Br₂ as the internal standard and ee was determined by chiral HPLC.

^bThe enantiomeric excess for the major (2S,3S) product.

Through a combination of catalyst-derived and environmentally enabled stereoselectivity, we could perform stereodivergent conjugate addition, without developing new catalysts for the anti-reaction. Scheme 3 shows that all four diastereomers of the γ-nitroaldehyde product could be obtained from butanal and β-nitrostyrene. Good to excellent yields, enantioselectivity, and diastereoselectivity could be obtained in all cases, by simply mixing the reactants, a readily synthesized “plain” prolinamide catalyst, and the appropriate MINP in water at room temperature.

Scheme 3.

Stereodivergent conjugate addition of butanal to β-nitrostyrene through a combination of chiral prolinamide 2 and MINPs-1 (also see Figure S6).

CONCLUSIONS

Although many examples exist in the literature that use environmental charity to enhance the stereoselectivity of a catalyst, including with molecularly imprinted materials [17, 19, 20, 51–54], overriding the intrinsic selectivity of the chiral catalyst has not been reported to the best of our knowledge. Our method enables green synthesis of all stereoisomers of the Michael product using readily prepared catalytic modulators—MINP preparation is a one-pot reaction, complete in two days; its purification requires simple precipitation and washing [21]. A powerful feature of molecular imprinting is facile tuning of the imprinted sites by different template molecules, tuning substrate selectivity of imprinted catalysts readily [55, 56]. Although the conjugate addition between aldehyde and nitroalkene is being studied, the principle demonstrated should be applicable to other systems as well. Conformations of substrates [39] and also chiral catalysts [57] have been found to strongly influence the stereoselectivity of a reaction. Since molecules often adopt different conformations in a confined environment [47–49], it is possible that environmental chirality could become a general method for conformational control, to enable catalysts or their substrates to access less stable conformations and form otherwise unfavorable products as a result.

Highlights.

• Enhancement or reversal of stereoselectivity of a chiral catalyst via environmental modulation

• anti- or syn-Selectivity in the conjugate addition of aldehyde to nitroalkene using the same catalyst

• Environmental control of the conformation of catalytic intermediate

• Green synthesis of all stereoisomers of γ-nitroaldehyde in water