Simple and Efficient Enantioselective α-Deuteration Method of α-Amino Acids without External Chiral Sources

Abstract

Deuterium-labeled α-amino acids are useful in research related to drug discovery and biomedical science. However, a high degree of site selectivity and stereoselectivity in the deuterium incorporation process is still difficult to achieve. Herein, we report a new enantioselective deuteration method at the α-position of several amino acids without external chiral sources. The proposed deuteration methods (NaOEt and EtOD) are highly selective and simple. Additionally, we provide a mechanistic study for this enantioretentive deuteration.

Introduction

Owing to the kinetic isotope effect, deuterium-labeled α-amino acids have found numerous important applications in research, especially related to drug discovery and biomedical science.¹⁻⁴ For example, they are used in elucidating secondary and tertiary structures of proteins.² They are also widely used to investigate biochemical processes such as the metabolism of peptides and the mechanism of enzymes.³ Further, compounds with an α-deuterated amino acid showed improved metabolic stability compared to those with an α-amino acid moiety.⁴ Therefore, various methods have been developed to obtain chiral deuterated amino acids.^5,6 Notable recent methods include the use of chiral metal complexes and organocatalysts.⁵ However, these methods are associated with poor site- and stereoselectivity depending on the amino acid type.^5a,5c,5e,5f Other notable methods are based on the use of enzymes.⁶ They are often limited in substrate scope or provide deuterated amino acids with moderate enantioselectivity and deuterium incorporation.

While most developed methods for asymmetrical deuterium attachment require external chiral sources, a few methods can access deuterated amino acids without using external chiral intervention.⁷ The principles of Seebach’s “self-regeneration of stereocenters (SRS)”⁸ have been mainly employed.^7c,7e Although the SRS-based methods could effectively and asymmetrically incorporate deuterium, they required a strong base (e.g., LDA) under anhydrous conditions and additional steps to transfer carbon-centered chirality to adjacent carbons and remove accessory groups (Scheme 1a). Kawabata group reported a different method^7b that used the concept of “memory of chirality (MOC)⁹”, a phenomenon in which the sp³ chirality of the starting material is preserved in a reactive intermediate as an axial chirality.^9c Unlike SRS-based methods, it does not need preinstallation of carbon-centered chirality. However, this method still required a strong base under anhydrous conditions at a low temperature (−78 °C). Therefore, the development of a convenient and practical production approach for α-deuterated amino acids is still in great demand.

Scheme 1. Enantioretentive α-Deuteration of Amino Acids.

In the course of our previous study on asymmetric total synthesis,¹⁰ we obtained a valuable clue for asymmetrically incorporating a deuterium atom into the α-position of amino acids without external chiral intervention under conventional reaction conditions. When the proline derivative 1 was treated with NaOEt in EtOD at room temperature, the aldol product d-2 was formed with retention of the C-α configuration (Scheme 1b). Interestingly, the analysis of the incomplete reaction mixtures showed that the hydrogen at the α-position of the remaining aldol substrate was replaced with deuterium to give d-1 while preserving its original chirality. Unlike Seebach’s SRS-based and Kawabata’s MOC-based methods, this stereoselective deuteration did not require the use of an anhydrous strong base. We envisioned that this phenomenon could be utilized to rapidly and efficiently access α-deuterated chiral prolines and other amino acids from readily available amino acids under conventional reaction conditions.

Herein, we report a new deuteration method at the α-position of amino acids, including proline, serine, cysteine, phenylalanine, and lysine (Scheme 1c). The developed deuteration method was simple and did not require expensive equipment, strong bases, or extra chiral sources. A mechanistic study for this enantio-retentive deuteration is also provided.

Result and Discussion

Initially, N-isobutyryl l-proline ester 4a was designed as a model substrate (Scheme 2). Unlike proline derivative 1, substrate 4a does not possess a carbonyl group in the amide side chain and therefore cannot undergo intramolecular aldol reaction. When 4a was treated with NaOEt (5.0 equiv) in EtOD (0.2 M) at room temperature, deuteration proceeded gradually to afford d-4a in 94% yield after 8 h. The deuteration level (DL) of thus obtained d-4a was excellent (98%) and its enantiomeric purity was greater than 97%. The prolonged reaction time (16 h) did not affect the reaction outcomes. Interestingly, negligible deuterium incorporation was observed at the α-position of the amide carbonyl group.

Scheme 2. Enantioretentive α-Deuteration of Proline Derivatives 4a^,,,

Reagents and conditions: 4a (0.2 mmol), NaOEt (5.0 equiv), EtOD (0.2 M), room temperature.

The indicated yields are isolated.

The DL is estimated based on ¹H NMR analysis.

The enantiomeric excess (ee) value is estimated by using chiral HPLC.

We modified the reaction conditions to understand the reaction. The reaction concentration did not considerably affect the reaction efficiency (Table 1, entries 2 and 3). The change in the cation of the ethoxide base did not alter the reaction outcome (entries 4 and 5). The influence of the solvent composition on the reaction outcome was noteworthy. The use of THF as a major cosolvent resulted in a lower yield and the enantiomeric purity of d-4a compared to the reaction in EtOD only (entry 1 vs 6). However, at −10 °C, the chemical yield and enantiomeric purity were similar, although the reaction time was prolonged (48 h; entry 7). The use of polar DMF substantially decreased the reaction efficiency and selectivity (entry 8). Further, the addition of a nonpolar solvent, i.e., toluene, led to a considerable decrease in the yield and enantioselectivity (entry 9). When EtOH was used as a cosolvent, the yield and ee values were the same as those obtained from the reaction in EtOD only; however, the deuterium level decreased at the given reaction time (entry 1 vs 10). The deuterium level continuously increased with the reaction time while maintaining the yield and enantioselectivity (entries 10 and 11). The deuteration reaction could be enlarged without significantly losing the reaction efficiency and selectivity. For instance, at 1.0 M and 1.0 equiv of NaOEt, a 40 mmol (9.6 g) scale reaction of 4a afforded d-4a in 87% yield, with a DL and enantiomeric purity of 94 and 96%, respectively (entry 12).

Table 1. Modification of the Reaction Conditions^a.

entry	deviation	yieldb	% Dc	eed
1	none	94%	98%	97%
2	EtOD (0.1 M)	96%	98%	98%
3	EtOD (1.0 M)	98%	96%	98%
4	LiOEt (5.0 equiv)	95%	95%	97%
5	KOEt (5.0 equiv)	97%	98%	98%
6	THF/EtOD = 10:1	75%	89%	75%
7	THF/EtOD = 10:1, – 10 °C, 48 h	91%	92%	94%
8	DMF/EtOD = 10:1	42%	84%	56%
9	toluene/EtOD = 10:1	46%	96%	4%
10	EtOH/EtOD = 1:1	95%	43%	97%
11	EtOH/EtOD = 1:1, 60 h	94%	60%	97%
12e	EtOD (1.0 M), NaOEt (1.0 equiv)	87%	94%	96%

^aReaction conditions: 4a (0.2 mmol), NaOEt (5.0 equiv), EtOD (0.2 M), room temperature, 16 h.

^bIsolated yields.

^cDetermined based on ¹H NMR analysis.

^dDetermined using chiral HPLC.

^e40 mmol scale.

Based on the aforementioned results, the enantioretentive deuteration mechanism was investigated. Inter- or intramolecular metal coordination mechanisms were not considered because no substantial differences were observed in the reaction outcomes between countercations (Table 1, entries 1, 4, and 5). Deuteration transfer from the amide moiety to the proline moiety was not considered because no deuterium incorporation was observed at the α-position of the amide carbonyl group in 4a. Thus, a conceivable deuteration pathway is associated with the deuteration of chiral enolates directly from the EtOD solvent.

We conducted a density functional theory (DFT)-based computational investigation using the M06-2X functional with the 6-311++G(d,p) basis set to understand the deuteration mechanism. The calculated free energy profile of EtOD at room temperature is shown in Figure 1. Based on the computational conformation analysis of 4a, 10 distinct conformers were found (see SI for details). The most stable conformers were 4a-I, which was used as a starting point for our computational studies. Deprotonation at the proline α-position on conformers 4a-I generated an axially chiral enolate 5-I with a free energy barrier (ΔG^‡) of 6.4 kcal/mol (TS1). The transient enolate could undergo several reaction pathways, including the epimerization of axial chirality, reprotonation, and deuteration by EtOD. The ΔG^‡ of deuteration from the Re face (TS2_Re), which lead to d-4a-I, is only 5.6 kcal/mol.¹¹ On the other hand, deuteration from the Si face resulted in less stable conformer ent-d-4a-II with the free energy barrier (ΔG^‡) of 16.0 kcal/mol (TS2_Si). These results suggested the preferential deuteration of the Re face, providing d-4a as a deuteration product.¹² The ΔG^‡ value of epimerization from enolate 5-I to ent-5-I was 11.7 kcal/mol (TS_rac).¹³ This implies that deuteration from the Re face can occur faster than epimerization in the deuterated solvent.

Figure 1.

Plausible reaction pathways and free energy profiles of enantioretentive α-deuteration of 4a.

We also considered the entropic penalties stemming from coordinating the substrate and EtOD in 5-I by calculating their free energies separately. The alignment of substrate and EtOD for deuteration results in the free energy penalty of 5.6 kcal/mol, leading to the overall ΔG^‡ of 11.2, 17.3, 21.6 kcal/mol for TS2_Re, TS2_rac, and TS2_Si, respectively (Figure 1b). Accounting for translational entropies of substrates and EtOD did not change the trends of the free energy barriers of these three TSs. The kinetic isotope effect (KIE) was also investigated for TS2_Re and TS2_Si; the calculated reaction rate of deuterium transfer was 4.9–5.0 times slower than that of proton transfer for both cases.¹⁴ Overall, the computational calculation results agreed well with the experimental outcomes of enantioretentive deuteration.

The scope of the enantioretentive deuteration process was examined by varying the amide group (Table 2). The employed conditions were 5 equiv of NaOEt in 0.2 M EtOD at room temperature. For practical reasons, the reaction time was fixed at 16 h, although a longer reaction time might lead to higher levels of deuterium exchange. The simple unbranched aliphatic-acid-conjugated proline ester underwent a high degree of enantioretentive deuteration. The reaction efficiency was almost the same as that of substrate 4a with an isobutyryl group. For instance, the substrates 4b and 4c with propionyl or pentenoyl group obtained the high yields and ee values of the corresponding deuterated compounds. Further, substrate 4d with the acetyl group successfully provided d-4d in high DL and high ee, albeit in a lower yield (57%) due to the instability of the acetylamide group under the aforementioned conditions.¹⁵ Moreover, when the phenylacetyl group was employed, the obtained ee value and DL of d-4e were high. In this case, almost complete deuterium incorporation was observed at the α-position of the amide carbonyl group. When substrate 4f with a bulky pivaloyl group was subjected to the above reaction conditions, the DL was only 60% at the given reaction time. After 48 h, the deuterium level reached 94%, and the enantiomeric purity of the obtained d-4f was 96%. We attributed the slow deuteration rate of 4f to the bulky pivaloyl group because it might cause a low kinetic acidity for the C-2 hydrogen atom. Further, substrate 4g with an N-Boc group produced the high yield and ee values of the corresponding deuterated compounds but with a low DL (24%). The prolonged reaction time (48 h) led to an increased DL (56%) without compromising the yield and enantioselectivity.

Table 2. Substrate Scope^a.

^aStandard reaction conditions: 4 (0.2 mmol), NaOEt (5.0 equiv), EtOD (0.2 M), room temperature, 16 h. The indicated yields are isolated. The DL and diastereomeric ratio (d.r.) were estimated by ¹H NMR analysis. The ee value was estimated using chiral HPLC.

^bThe reaction was processed for 48 h.

^cThe reaction was processed at 0 °C for 24 h.

^dThe reaction was processed at −10 °C for 48 h.

When the t-butyl ester moiety of 4a was changed to ethyl ester (4h), deuteration proceeded effectively while maintaining almost the same ee value and DL. On the other hand, When the t-butyl ester of 4a was changed to the dimethylamide group (4i), deuterium incorporation was not realized due to the lower acidity for the C-2 hydrogen atom of 4i. Based on this result, we envisioned that site-selective deuteration at the C-terminal of polypeptides would be possible. As a feasibility study, dipeptides 4j (Boc-l-Val-l-Pro-O^tBu) and 4k (Boc-l-Ala-l-Pro-O^tBu) were subjected to NaOEt/EtOD conditions. Both dipeptides showed site-selective deuteration at the C-2 position with high DL and stereoselectivity. This result suggests our method could be applied to longer peptides.

To further examine the substrate scope and limitations, we tested several substrates 4l–n with a noncarbonyl N-protecting group. When N-tosyl substrate 4l was subjected to the above reaction conditions, ee was only 12% at the given reaction time and temperature. However, when the reaction was performed at −10 °C for 48 h, the enantiomeric purity of d-4l was increased to 92% ee with 84% DL. When a benzyl (4m) or trityl group (4n) was employed as an N-protecting group, no deuteration products were observed. These results suggested the importance of the carbonyl group of the N-protecting group in the enantio-retentive deuteration.

The scope of the deuteration process was expanded to other amino acid derivatives. The pseudoprolines of serine, threonine, and cysteine were first examined (Table 3). They underwent deuteration more rapidly than proline derivatives under the given reaction conditions at room temperature; however, the chemical yields were lower due to the instability. Thus, the reactions were tested at 0 °C for 2 h. The oxazolidine-protected serine (6a and 6b) gave results similar to those obtained with the corresponding proline derivatives (4a and 4h). Moreover, the oxazolidine-protected threonine also afforded the deuterated product d-6c in >50:1 d.r. and 88% yield. In contrast, the cysteine-derived thiazolidine derivate 6d gave the corresponding deuterated compound in high yield but with almost complete racemization. This racemization could be overcome by performing the deuteration at lower reaction temperatures. At −50 °C, 6d produced d-6d (97% DL) in 90% yield with perfect enantioselectivity (For computational details, see SI and ref (16)).

Table 3. α-Deuteration of Pseudo-proline Derivatives^a.

^aStandard reaction conditions: 6 (0.2 mmol), NaOEt (5.0 equiv), EtOD (0.2 M), 0 °C, 2 h. The indicated yields are isolated. The DL and diastereomeric ratio (d.r.) were estimated by ¹H NMR analysis. The enantiomeric excess (ee) value was estimated using chiral HPLC.

^bThe reaction was processed at −50 °C for 12 h.

While the enantioretentive deuteration of proline was readily extended to the aforementioned pseudoprolines, the reaction protocols could not be directly applied to the acyclic amino acid derivatives. For instance, the alanine derivative 7a with an isobutyryl group as the only amine substituent produced a high yield of the α-deuterated compound at 0 °C (Table 4). However, almost complete racemization was observed. Meanwhile, the alanine derivative 7b possessing both isobutyryl and benzyl groups at the nitrogen atom exhibited a low degree of chirality preservation (23% ee) at 0 °C. The enantiomeric purity of d-7b was increased to 98% ee with 94% DL when the reaction was performed at −40 °C for 3 d. Under these conditions, N^ε-Cbz-protected lysine derivative 7c showed enantioretentive (99% ee) α-deuteration in excellent yield (95%) and DL (99%). Moreover, phenylalanine derivative 7d produced the corresponding deuterated compound d-7d in high yield and ee values with 70% DL. Similarly, the valine derivative 7e yielded good results in terms of both yield and DL, achieving 93% ee when the reaction was conducted at −30 °C for 7 d. The isoleucine-derived substrate 7f also showed good results in yield and selectivity at −10 °C for 3 d. However, its deuteration was rather slow compared to less sterically bulky derivatives 7b and 7d. The O-protected serine derivative 7g failed to produce the corresponding deuterated compounds because of the rapid β-elimination under the conditions.

Table 4. α-Deuteration of Acyclic Amino Acid Derivatives^a.

^aStandard reaction conditions: 7 (0.2 mmol), NaOEt (5.0 equiv), EtOD (0.2 M). The indicated yields are isolated. The DL was estimated based on ¹H NMR analysis. The enantiomeric excess (ee) value was estimated using chiral HPLC.

^bReaction runs by using EtOD/THF = 10:1 (0.2 M) as the solvent.

To demonstrate the synthetic utility of the current methodology in preparing deuterated amino acids, the transformations of deuterated products were explored. The proline ester d-4a with an isobutyryl group was converted to the deuterated proline 8 in high yield by acid-catalyzed hydrolysis without loss of the DL (Scheme 3). The pentenoyl group in d-4c was selectively removed upon treatment with iodine in aqueous THF to give a high yield of the deuterated proline t-butyl ester 9. The deuterated serine 10 was obtained in good yield from the oxazolidine-protected serine d-6a by acid-catalyzed hydrolysis. Further, acyclic amino ester derivative d-7b was successfully hydrolyzed in acidic conditions to afford N-benzyl amino acid 11 without lowering DL.

Scheme 3. Transformation of α-Deuterated α-Amino Acid Derivatives^a.

Conclusions

In summary, we established an efficient method for the site-selective and stereoselective synthesis of α-deuterated chiral amino acids from readily available amino acids themselves. The substrate proline and pseudoproline derivatives, possessing the N-isobutyryl group, were readily prepared. The reaction conditions were simple (NaOEt in EtOD solvent), and the developed method did not require any expensive equipment, strong bases, or extra chiral sources. The deuterated product was afforded with a high level of enantiopurity and deuteration. The EtOD amount can be reduced while maintaining enantioselectivity by using THF as a major cosolvent. Our mechanistic study suggested that the enantioretentive H/D exchange involves the enantioselective generation of axially chiral enolates and the facial-selective deuteration of enolates. This study demonstrated the first case of the retention of chiral information on optically active α-amino acid derivatives during α-deuteration in a protic solvent. Moreover, the scope of the enantioretentive deuteration process was extended to acyclic amino acids. The several acyclic amino ester derivatives possessing isobutyryl and benzyl groups at the nitrogen atom exhibited an excellent degree of chirality preservation at a low temperature of −40 °C. Further studies are being conducted to extend this base-promoted enantioretentive hydrogen isotope exchange reaction for asymmetrically assessing various isotope compounds.

Method

General Information

All chemicals were of reagent grade and were used as purchased. All reactions were performed under an inert atmosphere of dry nitrogen by using distilled dry solvents. The reactions were monitored with TLC analysis using silica gel 60 F-254 thin-layer chromatography plates. Compounds on the TLC plates were visualized under UV light and by spraying with either potassium permanganate or anisaldehyde solutions. Flash column chromatography was conducted on silica gel 60 (230–400 mesh). Melting points were measured using a Buchi B-540 melting point apparatus without correction. ¹H and ¹³C NMR spectra were recorded on a JEOL JNM-ECZ400S/L1 (400 MHz) and JEOL JNM-ECA-600 (600 MHz) at 298 K if not noted otherwise. The IR spectra were measured by an Agilent Technologies 5500 Series FT-IR spectrometer. High-resolution mass spectra (HRMS) were recorded using fast atom bombardment (FAB), or electrospray ionization (ESI) mass spectrometry.

General Procedure for α-Deuteration of Amino Acids

To a solution of amino acid derivatives (0.2 mmol, 1.0 equiv) in EtOD (1.0 mL) was added NaOEt (1.0 mmol, 5.0 equiv) at a given temperature. After stirring for the corresponding reaction time, the reaction mixture was quenched with saturated NH₄Cl aqueous solution (3.0 mL). Then, EtOD was removed under reduced pressure. The resulting mixture was poured into water (5.0 mL) and extracted with EtOAc three times (3 × 10 mL). The combined organic fraction was dried in MgSO₄ and concentrated under reduced pressure. The residue was purified by flash chromatography on silica gel to give α-deuterated amino acid derivatives.

Computational Mechanistic Studies

We describe in detail in the Supporting Information.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/jacsau.4c00185.

• Experimental procedures of all new compounds; spectroscopic data analysis; copies of NMR spectra; computational studies (PDF)

Author Contributions

CRediT: Soojun Park data curation, formal analysis, investigation, methodology, resources, visualization, writing-original draft; Jae Hyun Kim conceptualization, data curation, formal analysis, methodology, resources, validation, visualization; Dongjun Kim data curation, methodology, validation, visualization; Sanghee Kim conceptualization, formal analysis, funding acquisition, investigation, project administration, supervision, writing-review & editing.

The authors declare no competing financial interest.