1,2,3,4-Tetrahydro-1,5-naphthyridine motif is present in a wide range of important bioactive compounds (
Scheme 1a) [
1–
6]. Importantly, the stereogenic configuration shows considerable influence on their biological activities. For instance, the
in vitro human plasma CETP inhibitory activity of (2
R, 4
S)-
B exceeds that of its enantiomer by >320 times (IC
50 = 21 nmol/L and 7456 nmol/L, respectively) [
3]. In addition, 1,2,3,4-tetrahydro-1,5-naphthyridine framework is amenable to transform into other important compounds, for example 1,5-diaza-
cis-decalin, further enhancing its versatility and utility in synthetic chemistry [
7–
12]. Therefore, the development of efficient methods for the synthesis of 1,2,3,4-tetrahydro-1,5-naphthyridines has attracted considerable attention in the past several decades. Snyder
et al. disclosed the construction of 1,2,3,4-tetrahydro-1,5-naphthyridines by an intramolecular inverse-electron-demand Diels-Alder reaction between imidazoles and 1,2,4-triazines (
Scheme 1b, route 1) [
13,
14]. Aza-vinylogous Povarov reaction of aromatic amines,
α-ketoaldehydes and
α,β-unsaturated dimethylhydrazones under mechanochemical conditions was reported by the group of Menéndez, a number of tetrahydro-1,5-naphthyridines were produced in moderate to high yield (
Scheme 1b, route 2) [
15]. Shao, Zhang, Chen and coworkers reported a rare-earth metal-catalyzed synthesis of 1,2,3,4-tetrahydro-1,5-naphthyridines
via C–H cyclization of functionalized pyridines (
Scheme 1b, route 3) [
16]. The direct hydrogenation of 1,5-naphthyridine to afford such architecture has been extensively studied by Fan, Khusnutdinova, and others (
Scheme 1b, route 4) [
17,
18]. Nevertheless, the asymmetric construction of enantioenriched tetrahydro-1,5-naphthyridines was relatively rare. In 2016, Fan and coworkers described an elegant catalytic synthesis of optically active 1,2,3,4-tetrahydro-1,5-naphthyridines
via enantioselective hydrogenation of 1,5-naphthyridine (
Scheme 1b, route 4), up to 95% yield with 99%
ee was achieved [
18]. Nevertheless, the development of new routes and catalyst systems to access chiral 1,2,3,4-tetrahydro-1,5-naphthyridines with different product distribution is still in high demand.
Organo rare-earth (RE) catalysis has been emerging as one of powerful platforms for a number of chemical transformations [
19–
25]. In the context of rare-earth-mediated C–H functionalizations, significant advances have been achieved with cationic rare-earth alkyl complexes as the catalysts probably due to their relatively stronger Lewis acidity and less steric hindrances around rare-earth metal center than their neutral analogues [
26–
31]. In contrast, neutral rare-earth complexes are much less popular during the past decade [
32–
37]. As part of our continuous efforts [
38–
44] in organo rare-earth catalysis and motivated by previous work [
16] by Shao, Zhang, Chen
et al., we envisioned that chiral bis(oxazolinato) rare-earth amido complex catalyst [
45–
47] was a potential promoter for the titled reaction. The reaction was proposed to be initiated by site-selective deprotonation [
48–
51] with the assistance of bis(oxazolinato) rare-earth amido complex, followed by an enantioselective alkene insertion into the resulting RE-aryl bond to generate the chiral rare-earth alkyl species. The
σ-bond metathesis of rare-earth alkyl intermediate with another substrate afforded the RE-aryl complex, closing the catalytic cycle. Herein, we would like to disclose our effort along this line. La[N(SiMe
3)
2]
3/
PyBox was eventually identified to be the competent catalyst for the enantio-selective intramolecular C–H alkylation of pyridine derivatives with alkene. A number of chiral 1,2,3,4-tetrahydro-1,5-naphthyridines were obtained in an atom-economical and efficient manner with moderate to good yields and high enantioselectivity (34 examples, up to 99% yield, 93%
ee).
To begin with,
N-(but-3-en-1-yl)-
N-phenylpyridin-3-amine (
1a) was chosen as the model substrate for the optimization of the reaction conditions. As depicted in
Table 1, different types of oxazoline ligands and various rare-earth metals were investigated.
When the reaction was carried out with La[N(SiMe
3)
2]
3 in toluene at 100 ℃, no reaction occurred (
Table 1, entry 1). The addition of Bn
2NH as the additive led to the generation of the desired 1,2,3,4-tetrahydro-1,5-naphthyridine product
2a in 99% yield
via 6-
exo-trig cyclization (entry 2). Based on our previous works [
40–
41], La[N(SiMe
3)
2]
3/
L1 was employed as the chiral catalyst, the corresponding product
2a was afforded in 67% yield with 23%
ee (entry 3). The subsequent examination of chiral oxazoline ligands suggested that pyridine-based bisoxazoline ligand
L2 was a better choice, product
2a was isolated in 27% yield with 86%
ee (entry 4, PS8 in Supporting information for more details). The variation of pyridine-based bisoxazoline ligands indicated that decreasing or increasing the steric hindrance of substitution at stereogenic center led to inferior results (entries 5 and 6, methyl substituted
L3: 31% yield, 26%
ee, tert-butyl substituted
L4: 8% yield, 20%
ee). Chiral ligand
L5 with cyclohexyl group provided comparable enantioselectivity (entry 7, 84%
ee), while lower yield and
ee value were afforded by phenyl substituted
L6 (entry 8, 8% yield, 38%
ee). The use of pyridine-based mono-oxazoline ligand
L7 resulted into poor results (entry 9, 48% yield, 6%
ee). Next, the influence of rare-earth ion radius was tested (PS7 in Supporting information for more details). It was found that the enantioselectivity was closely associated with rare-earth ion radius [
52–
58]. La[N(SiMe
3)
2]
3 and Ce[N(SiMe
3)
2]
3, with larger ionic radii (six-coordinate La
3+, 1.032 Å; Ce
3+, 1.010 Å) providing superior
ee values (86%
ee and 84%
ee, respectively) compared to Sm[N(SiMe
3)
2]
3 (0.958 Å, 50%
ee) and Y[N(SiMe
3)
2]
3 (0.900 Å, 10%
ee) (entries 4 and 10–12). When the reaction was run with 20 mol% catalyst loading, the yield of
2a was increased to 68% yield with maintained
ee value (entry 13). Switching the solvent to mesitylene and changing the additive to dihexylamine led to
2a in 78% yield and 90%
ee (entry 14). Performing the reaction at higher concentration provided a better yield with maintained enantioselectivity after 18 h (entry 15, 89% yield, 90%
ee).
Therefore, the optimal reaction conditions were established as 1a (0.1 mmol), La[N(SiMe3)2]3/L2 (20 mol%, 1:1), dihexylamine (20 mol%) in mesitylene (0.8 mL) at 100 ℃ for 18 h.
With the optimized reaction conditions in hand, the substrate scope was investigated (
Scheme 2). Firstly, the protecting group on the nitrogen atom was examined. Regardless of the substituent pattern and the electronic property of the aryl moiety, all of the reactions underwent smoothly, the corresponding products
2a–
2t were obtained in moderate to good yield with high enantioselectivity (61%–99% yield, 75%–90%
ee). The substrate
1a–
1k with
para-substituted aryl group transformed into the products with relatively high
ee value (86%–90%
ee) except 4-F substituted one (75%
ee). The position of substitutions on the phenyl group displayed a limited effect on the outcomes (
2l–
2t, 61%–86% yield, 86%–90%
ee). The reaction of substrate
1u bearing 3,5-dimethylphenyl group afforded the related product
2u in 91% yield, 90%
ee. In contrast, 48% yield with 90%
ee was obtained for 2,6-diethyl-4-methylphenyl substituted
1v with
L5 as the ligand. Switching
N-aryl group to
N-Bn and
N–CH
2Bn were feasible, the desired tetrahydro-1,5-naphthyridines were generated in good yield with high
ee value (
2w: 69% yield, 88%
ee,
2x: 88% yield, 86%
ee, respectively). Next, the substitutions on the pyridine moiety were tested. All of the reaction took place well, delivering the desired adducts
2y–
2ac in 51%–90% yield with 80%–92%
ee. Notably, the reaction was compatible with quinoline-derived substrates
1ad–
1af, moderate yield with high enantioselectivity were obtained (43%–61% yield, 87%–93%
ee). Moderate yield and lower stereoselectivity were obtained for the reaction of styrene-derived substrate
1ag (47% yield, 31%
ee). The attempts to synthesize
2ah, 2aj, 2ak and
2al were unsuccessful. In addition,
2ai was generated in 90% yield and 38%
ee in the presence of Y[N(SiMe
3)
2]
3/
L1. The absolute configuration of adduct
2i was determined to be (
S) by X-ray diffraction analysis.
To demonstrate the potential utility of the current methodology, a scale-up synthesis of product
2a was conducted (
Scheme 3a). Under the optimal reaction conditions, 5 mmol of
1a (1.12g) converted into the corresponding chiral product
2a in 85% yield (0.95g) with 87%
ee. then, the further transformations of
2a were carried out. Utilizing the
[Sc] catalyst developed by our research group, we successfully achieved C–H alkylation at C6 position of the pyridine ring with styrene, the desired product was obtained in 80% yield with 85%
ee (
Scheme 3b). Olefin derived from estradiol were also amenable to the reaction, delivering the related adduct in 71% yield with 92.5/7.5
dr.
Motivated by the diverse bioactivity of 1,2,3,4-tetrahydro-1,5-naphthyridines, we tested the bioactivity of products
2. The
in vitro cytotoxicity against the human hepatocellular carcinoma (HCCLM3) cell and lung cancer cell (A549) were investigated (see Supporting information for details). The outcomes implied that compound
2 g displayed an obvious inhibitory effect on the viability of HCCLM3 and A549 (
Scheme 3c). IC
50 values of
2 g toward HCCLM3 and A549 were 10.23 and 3.12 µmol/L, respectively.
To gain a deeper insight into the reaction mechanism, a series of mechanistic experiments were conducted (see Supporting information for more details). As illustrated in
Scheme 4a, the relationship between the enantiopurity of chiral ligand and the
ee value of product was a linear effect, suggesting that one ligand molecule is contained in each reactive La(Ⅲ) complex. Furthermore, parallel kinetic experiments were performed (
Scheme 4b), a kinetic isotope effect value of 5.5 was observed, implying that C(sp
2)−
H bond cleavage of pyridine might be involved in the turnover limiting step. The dark blue colour of La[N(SiMe
3)
2]
3/
L2 tempered us to probe its structure. As indicated in
Scheme 4c, the chemical shift of H
a, H
b and H
c on pyridine ring in chiral ligand moved to from 8.12 ppm (H
a, H
b) and 6.95 ppm (H
c) to 5.44 ppm (H
a), 5.40 ppm (H
b) and 5.31 ppm (H
c). Moreover, after coordination with La[N(SiMe
3)
2]
3, the chemical shift of C
c changed from 136.1 ppm to 50.8 ppm. Three different [N(SiMe
3)
2] groups were detected as well at 0.39, 0.35 and 0.32 ppm. These observations along with 2D NMR spectra rendered us to proposed that N(SiMe
3)
2 group underwent nucleophilic addition to C4-position of pyridine ring during the coordination process. The related transformations have been reported by other groups [
59–
62].
Based on previous work, we proposed a possible catalytic cycle (
Fig. 1). In order to elucidate the origin of the enantioselective control, DFT calculations were conducted on the transition states involved in the olefin migratory insertion step. Considering the relative position of the two N(SiMe
3)
2 groups, four distinct transition states including both
cis and
trans configurations were identified. As depicted in
Fig. 2, the results indicated the
S-configuration transition state in the
trans form was energetically more favourable (0 kcal/mol
vs. 2.8 kcal/mol), aligning well with experimental data. Further structural analysis revealed that the
R-configuration transition state experienced considerable steric repulsion between the phenyl group on nitrogen in the substrate and
iPr group of the ligand as well the hydrogen of terminal alkene with oxazoline ring, elucidating the preference for the
S-configuration. Similar situations were observed for the
cis type of transition states (0.9 kcal/mol
vs. 3.1 kcal/mol, for more details, see Supporting information).
In conclusion, we have accomplished a highly enantioselective intramolecular C–H alkylation of pyridine derivatives with alkene by La[N(SiMe3)2]3/PyBox complex. This protocol offers a straightforward and atom-economical route to a new family of chiral 1,2,3,4-tetrahydro-1,5-naphthyridines with high stereoselectivity (34 examples, up to 93% yield, 93% ee). The potential utility of the current method in organic synthesis was demonstrated by scale-up synthesis of chiral product and its further transformations. This work provides a new catalyst system for rare-earth-mediated asymmetric C–H functionalization. The further applications of chiral oxazoline-based rare-earth metal complexes in other related asymmetric transformations are in progress.
PDF
Full
110
60
关闭全屏
No. 86 Xueyuan South Road, Haidian District, Beijing
010-62662699(Operation and Technical Support)
qkjq@cast.org.cn
