Yuzhi
Ren
a,
Chen
Lin
a,
Han
Zhang
b,
Zuquan
Liu
a,
Donghui
Wei
*b,
Jie
Feng
a and
Ding
Du
*a
aSchool of Science, China Pharmaceutical University, Nanjing, 210009, P. R. China. E-mail: ddmn9999@cpu.edu.cn
bCollege of Chemistry, Zhengzhou University, Zhengzhou, Henan Province 450001, P. R. China. E-mail: donghuiwei@zzu.edu.cn
First published on 28th March 2025
Atropisomers bearing multiple stereogenic axes are of increasing relevance to materials science, pharmaceuticals, and catalysis. However, the catalytic enantioselective construction of these atropisomers in a single step remains synthetically challenging. We herein report the first NHC-organocatalytic enantioselective synthesis of a new class of monoaxially and 1,4-diaxially chiral fused uracil scaffolds. Preliminary studies on the antitumor activity of selected compounds demonstrated that this new class of axially chiral uracil derivatives may have potential applications in the discovery of new lead compounds in medicinal chemistry.
Over the past few decades, N-heterocyclic carbenes (NHCs) have provided a diverse range of unique reactive intermediates and reaction modes, enabling the acquisition of various functional molecules, especially those with central chirality.12 The application of NHC catalysis as an effective tool for constructing axially chiral molecules has emerged as a growing area in recent years.13 However, the reported strategies typically involve constructing diverse atropisomers with a single axis. The enantioselective construction of atropisomers with multiple stereogenic axes is still challenging and very limited.14 To the best of our knowledge, atropisomers exhibiting C–C and C–N 1,4-diaxial chirality have not been disclosed yet. To address this synthetic challenge, we aimed to design a step-economic and stereoselective protocol that focused on NHC-catalyzed atroposelective de novo construction of novel heterocyclic frameworks featuring C–C and C–N 1,4-diaxes. Herein, we wish to report a single-step de novo construction of a new class of fused uracil scaffolds (pyrido[2,3-d]pyrimidines) 3/4 with C–C and C–N 1,4-diaxes through NHC-catalyzed atroposelective formal (3 + 3) annulation of 2,6-disubstituted alkyne esters 1 and 6-aminouracils 2 (Scheme 1B) via alkynyl acylazolium intermediates.15 This protocol also features a one-stone-two-birds strategy for highly enantioselective synthesis of fused uracil atropisomers 5 with a C–N single axis.
We started our study by using 3-(2-methoxynaphthalene-1-yl)propiolate 1a and 6-(2-tert-butylphenyl)aminouracil 2a as the model substrates for reaction condition optimization, to test the feasibility of constructing a fused uracil framework bearing stereogenic C–C and C–N 1,4-diaxes (Table 1). Initially, an array of NHC precursors (A–F) were used to promote the reaction using Et3N as a base in DMF at room temperature (entries 1–6). As a result, the reaction in the presence of precatalysts A and C–F proceeded smoothly to afford the desired diastereomers 3a/4a with stereogenic 1,4-diaxes. F was found to be the optimal catalyst, in the presence of which diastereomers 3a and 4a were obtained in 76% combined yield with 98:
2 er and 90
:
10 er, respectively (entry 6). Further screening of different organic and inorganic bases in the presence of catalyst F did not give superior results (entries 7–11). Although the results of Cs2CO3 were comparable to those of Et3N, diastereoselectivity of the products and enantioselectivity of diastereomer 4a were not improved (entry 10). Solvent screening showed that the reaction in DCM delivered the desired products with maintained enantioselectivity and better diastereoselectivity (2
:
1) albeit in a diminished yield (48%) (entry 12). Therefore, DMF and DCM were used as a combined solvent in an attempt to improve both diastereoselectivity and enantioselectivity of the reaction (entries 16–17). It was found that when DCM and DMF (2
:
1) were used as the solvent, the reaction gave diastereoisomers 3a and 4a in a 65% combined yield, both of which exhibited excellent enantioselectivity (98
:
2 er), although the diastereoselectivity was not improved. Finally, the conditions shown in entry 17 were established as the optimal ones for further substrate scope exploration.
Entry | NHC | Base | Solvent | Yieldb (%) | drc, erd |
---|---|---|---|---|---|
a 1a (0.10 mmol, 1.0 equiv.), 2a (0.12 mmol, 1.2 equiv.), base (0.20 mmol, 2.0 equiv.), precatalyst (0.01 mmol, 10 mol%), 4 ÅMS (25 mg), solvent (2–3 mL), under a N2 atmosphere. b Combined isolated yields based on 1a. c The dr values were determined by 1H NMR of the crude product. d The er values were determined by chiral HPLC analysis. e The ratio of volume. | |||||
1 | A | Et3N | DMF | 71 | 1.4![]() ![]() ![]() ![]() ![]() ![]() |
2 | B | Et3N | DMF | 0 | — |
3 | C | Et3N | DMF | 77 | 1![]() ![]() ![]() ![]() ![]() ![]() |
4 | D | Et3N | DMF | 78 | 1![]() ![]() ![]() ![]() ![]() ![]() |
5 | E | Et3N | DMF | 73 | 1![]() ![]() ![]() ![]() ![]() ![]() |
6 | F | Et3N | DMF | 76 | 1![]() ![]() ![]() ![]() ![]() ![]() |
7 | F | DIPEA | DMF | 72 | 1![]() ![]() ![]() ![]() ![]() ![]() |
8 | F | DABCO | DMF | 42 | 1.5![]() ![]() ![]() ![]() ![]() ![]() |
9 | F | DBU | DMF | Trace | — |
10 | F | Cs2CO3 | DMF | 68 | 1![]() ![]() ![]() ![]() ![]() ![]() |
11 | F | CsOAc | DMF | 37 | 1![]() ![]() ![]() ![]() ![]() ![]() |
12 | F | Et3N | DCM | 48 | 2![]() ![]() ![]() ![]() ![]() ![]() |
13 | F | Et3N | CHCl3 | 0 | — |
14 | F | Et3N | CH3CN | <10 | — |
15 | F | Et3N | THF | <10 | — |
16 | F | Et3N | DCM![]() ![]() ![]() ![]() |
41 | 1.5![]() ![]() ![]() ![]() ![]() ![]() |
17 | F | Et3N | DCM![]() ![]() ![]() ![]() |
65 | 1.5![]() ![]() ![]() ![]() ![]() ![]() |
With the optimized conditions in hand, the reaction scope for the synthesis of fused uracils 3/4 bearing 1,4-diaxes was then investigated (Scheme 2). Initially, 3-(naphthalene-1-yl)propiolates 1 bearing different substituents at the 6-position were examined. The reactions of alkyne esters 1 bearing 6-aryls, bromo or alkyl groups proceeded smoothly at room temperature to afford the desired products 3b/4b–3d/4d and 3f/4f–3k/4k in moderate to good combined yields. Although the diastereoselectivity was not satisfactory for most substrates, diastereomers 3b–d and 3f–k were obtained with high to excellent enantioselectivity. Diastereomers 4b–d and 4f–k exhibited decreased enantioselectivity compared to model product 4a. There was an exception that the reaction of 6-(3-chlorophenyl)alkyne ester 1e produced the diastereomer 4e in 70% yield with high diastereoselectivity (>20:
1) and enantioselectivity (96
:
4 er). Then, alkyne esters 1 with diverse 7-substituents were evaluated. It was found that 7-phenyl substituted alkyne ester 1l yielded the single diastereomer 4l in 65% yield with 85
:
15 er, while alkyne esters 1m–p delivered the corresponding products 3m/4m–3p/4p in moderate to good yields with high enantioselectivity for diastereomer 3m–p.
It should be noted that the reaction of 7-bromo alkyne ester 1o did not work under the standard conditions. The employment of NHC-precatalyst A could promote the reaction to give diastereomers 3o/4o in 46% combined yield with 1:
1.5 dr, but 4o was obtained in almost racemic form. Two representative 4-phenyl and 3-phenyl-substituted alkyne esters 1q and 1r were also examined. The results were found to be similar to those of the previous substrates with high enantioselectivity for 3q/3r and moderate enantioselectivity for 4q/4r. Further replacement of the 2-OMe group of alkyne ester 1a with 2-OEt (3s/4s), 2-OBn (3u/4u), 2-Me (3v/4v) and 2-SMe (3w/4w) resulted in unchanged or slightly decreased enantioselectivities, while replacement of 2-OMe group of 2a with 2-OiPr (3t/4t) led to significantly improved diastereoselectivity (>20
:
1 dr) while maintaining enantioselectivity. Gratifyingly, the reaction of tetrahydronaphthalene alkyne substrate 1x worked equally well to deliver products with 96
:
4 er for 3x and 83
:
17 er for 4x. The reaction of 2-methoxy-6-phenyl-substituted alkyne ester 1y produced diastereomers 3y and 4y in 78% combined yield, with 4y isolated as the major product with 80
:
20 er. Moreover, a five-membered benzothiophen-3-yl alkyne ester 1z was used, yielding the desired products 3z/4z with high diastereoselectivity (12
:
1 dr) and excellent enantioselectivity (97
:
3 er) for the major diastereomer 3z, albeit with a lower combined yield (38%) due to the low conversion of the substrates. Two representative aminouracils, 6-(2-tert-butyl-4-iodophenyl)aminouracil 2b and 6-(2-tert-butyl-4-phenylphenyl)aminouracil 2c, were also employed to further explore the reaction scope. Interestingly, the reaction with 2b led to a decrease in enantioselectivity for both diastereomers 3aa and 4aa, while the reaction with 2c maintained enantioselectivity (96
:
4 er) for 3ab but decreased enantioselectivity (84.5
:
15.5 er) for 4ab. Last, the effect of the N-phenyl ortho-group of substrates 2 on the stereoselectivity of this reaction was investigated by replacing the t-butyl group with an iodine group. Unfortunately, this variation led to a significant decrease in the enantioselectivity of both diastereomers 3ac and 4ac, which indicated the necessity of an N-phenyl ortho-group with large steric hindrance.
We then turned our attention to testing the feasibility of constructing fused uracils bearing a single C–N axis.16 Initially, the reaction conditions for the synthesis of fused uracil 6a with a single C–N axis were investigated, and the results are included in the ESI (Table S1†). The desired product 6a was obtained in 79% yield with 97:
3 er under the catalysis of precatalyst C using DBU as the base and DCM as the solvent, establishing the optimal conditions for scope exploration (Scheme 3). It was found that the (3 + 3) annulation reaction of aromatic alkyne esters 5 with 6-(2-tert-butylphenyl)aminouracil 2a proceeded smoothly to afford the corresponding C–N monoaxial products 6b–h in good yields and with high enantioselectivity. Gratifyingly, the protocol was also applicable to two representative β-alkyl-substituted aliphatic alkyne esters, and the desired products 6i and 6j were obtained with slightly decreased enantioselectivities. Unfortunately, the reaction of 4-nitrophenyl propiolate gave product 6k in 77% yield with significantly decreased enantioselectivity (60
:
40 er), indicating the significant role of the β-substituents of alkyne esters 5 in the enantio-control process. Then, the influence of substituents on the phenyl ring of 6-phenylaminouracil 2 was studied. For 6-(2-tert-butyl-4-iodophenyl)aminouracil 2b and 6-(2-tert-butyl-4-phenylphenyl)aminouracil 2c, the desired C–N monoaxial products 6l and 6m were both obtained in high yields and enantioselectivities. However, replacement of the 2-tert-butyl group of substrate 2a with smaller groups like iodine (2d), phenyl (2e) and isopropyl (2f) resulted in the formation of product 6n with significantly decreased enantioselectivity and even racemic products 6o and 6p. Moreover, the reaction of 6-phenylaminouracil 2g with 3-(2-methoxynaphthalene-1-yl)propiolate 1a was carried out to examine the possibility of constructing a C–C monoaxial uracil product. Unfortunately, the desired product 7 was obtained in almost racemic form albeit with a high yield. Based on the above results, it can be concluded that the 2-tert-butyl group of 6-aminouracils 2 played a crucial role in the enantioselective construction of both monoaxially and 1,4-diaxially chiral uracil products.
To further explore the utility of this protocol, some synthetic applications were then investigated as shown in Fig. 1. First, a 1 mmol scale-up reaction between 1a and 2a under the standard conditions was carried out to afford products 3a and 4a with almost unchanged yields and stereoselectivity (Fig. 1A). Then, late-stage functionalization of product 3a was conducted (Fig. 1B). As a result, bromination and demethylation of 3a delivered products 8 and 9, respectively, in high yields while maintaining enantioselectivity.
To investigate the possible bioactivity of this new class of axially chiral fused uracil scaffolds, the proliferation inhibitory activity of several selected compounds 3a, 6b, 6h and 6i on MV4-11 human myeloid monocytic leukemia cells was evaluated (Fig. 1C). Among them, enantiopure 3a, 6b and 6h exhibited potent cell proliferation inhibitory activity against MV4-11 cancer cells, with low IC50 values ranging from 4.2 to 7.5 μM. In particular, the enantiopure 1,4-biaxially chiral uracil 3a showed much more potent cell proliferation inhibitory activity (IC50 = 4.2 μM) than racemic 3a (IC50 = >100 μM). Additionally, neither enantiopure nor racemic monoaxially chiral compound 6i with an alkyl group at the 5-position exhibited cell proliferation inhibitory activity against MV4-11 cancer cells. The above results demonstrated the potential application of this new class of axially chiral uracils as lead compounds for drug discovery.
The thermal stability and rotational barriers of the representative products 3a, 6a and 7 were also examined (Fig. 2). The racemization experiments for 3a and 6a were conducted in m-xylene at different temperatures (25–140 °C). The results revealed that both diaxially and monoaxially chiral products 3a and 6a were stable even at high temperatures because no obvious racemization of 3a and 6a was observed at 140 °C for 24 h (Fig. 2A). The experimental results were found to be consistent with the DFT calculated results (Fig. 2B). The rotation barriers for compounds 3a, 6a and 7 are higher than 36 kcal mol−1, indicating the high stability of both diaxially and monoaxially chiral products. In order to explain the origin of the formation of racemic product 7, we further calculated the energy barrier difference between the C2–C8 bond formation transition states, TS7R and TS7S, involved in the stereoselectivity-determining step of the reaction. The results shown in Fig. 2C indicate that the energy difference between the two transition states is only 0.3 kcal mol−1, which well explains the racemization of 7 in kinetic terms.
![]() | ||
Fig. 2 Determination of the rotational barriers of products 3a/6a/7 and explanation for racemization of 7 through DFT calculations. |
DFT calculations have also been performed to explore the origin of stereoselectivity in the synthesis of 1,4-diaxially chiral uracil products, as shown in Fig. 3. The whole reaction is initiated by the nucleophilic addition to 1a by the actual NHC catalyst (denoted as Cat) to form the alkynyl acylazolium intermediate. The subsequent conjugate addition between anion 2a− and the alkynyl acylazolium intermediate leads to the formation of the C2–C8 bond via four possible transition states TS3SR, TS3RR, TS3RS and TS3SS, respectively. The Gibbs free energy barriers of TS3SR, TS3RR, TS3RS and TS3SS are 5.5, 8.8, 9.0 and 9.9 kcal mol−1, respectively, indicating that the SR-configurational pathway is the most energetically favorable pathway. Meanwhile, the smaller energy barrier difference between the above diastereomeric transition states may account for the lower diastereoselectivity of this reaction. Then, only the favorable pathway associated with intermediate M3SR has been discussed in the following part. The intermediate M3SR transforms to intermediate M5SR through a stepwise [1,3]-proton shift through transition states TS4SR (ΔG‡ = 2.2 kcal mol−1) and TS5SR (ΔG‡ = 8.9 kcal mol−1). The intramolecular nucleophilic addition was responsible for the C–N bond formation via transition state TS6SR (ΔG‡ = 6.8 kcal mol−1), forming intermediate M6SR. At the final step, the free energy barrier for the dissociation of NHC from product 3a is 4.1 kcal mol−1via transition state TS7SR. For a detailed description of the reaction process and calculation methods, refer to the ESI.†
Footnote |
† Electronic supplementary information (ESI) available. CCDC 2412375 and 2412377. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d5sc00452g |
This journal is © The Royal Society of Chemistry 2025 |