- Email: [email protected]
The Alstoscholarisine Alkaloids: Isolation, Structure Determination, Biogenesis, Biological Evaluation, and Synthesis
ARTICLE IN PRESS
The Alstoscholarisine Alkaloids: Isolation, Structure Determination, Biogenesis, Biological Evaluation, and Synthesis Jeremy D. Mason, Steven M. Weinreb1 Department of Chemistry, The Pennsylvania State University, University Park, PA, United States 1 Corresponding author: e-mail address: [email protected]
Contents 1. Introduction 2. Isolation and Characterization 2.1 Isolation and Purification 2.2 Structure Elucidation 3. Putative Biogenesis 3.1 Biosynthesis of ()-Alstoscholarisines A–E 3.2 Biosynthesis of (+)-Alstoscholarisine G 3.3 Biosynthesis of (+)-Alstoscholarisines H–J 4. Biological Activity 5. Total Synthesis 5.1 Syntheses of Alstoscholarisines A and E 5.2 Weinreb/Mason Synthesis of ()-Alstoscholarisines B, C, and D 5.3 Luo/Xia Synthesis of ()-Alstoscholarisine H 5.4 Liao Synthesis of (–)-Alstoscholarisines A, E, and Their Enantiomers 6. Summary References
2 3 3 4 17 17 19 19 20 21 21 29 31 33 35 35
Abstract The alstoscholarisines are a small family of biologically and structurally interesting polycyclic monoterpenoid indole alkaloids isolated from the leaf extracts of Alstonia scholaris. The alkaloids can be divided into three different subtypes based upon their structures and putative biogenesis: (1) ()-alstoscholarisines A–E, (2) (+)-alstoscholarisine G, and (3) (+)-alstoscholarisines H–J. This review discusses the isolation, structure determination, biological activity, and proposed biosynthesis of these metabolites. In addition, synthetic studies on the alkaloids are described including total syntheses of racemic alstoscholarisines A–E, a total synthesis of ()-alstoscholarisine A, and a synthesis of racemic alstoscholarisine H. The Alkaloids ISSN 1099-4831 https://doi.org/10.1016/bs.alkal.2018.09.001
#
2019 Elsevier Inc. All rights reserved.
1
ARTICLE IN PRESS 2
Jeremy D. Mason and Steven M. Weinreb
1. INTRODUCTION Extracts of the tree Alstonia scholaris have a long history of use in traditional medicine in many parts of the world including China, Thailand, Malaysia, India, the Philippines, Africa, and Australia.1 These extracts have been utilized for treatment of an exceptionally diverse range of ailments including malaria, jaundice, microbial infectious diseases, and cancer. This organism and related species have been the subject of a number of phytochemical investigations, and a variety of structurally and biologically interesting secondary metabolites have been isolated from these plants. In an ongoing search for biologically active natural products, Luo et al. found in 2014 that leaf extracts of A. scholaris increased the proliferation of adult mouse hippocampal neural stem cells in vitro (vide infra).2 Upon biologically guided purification of the crude alkaloid mixture, five novel monoterpenoid indole alkaloids,3 ()-alstoscholarisines A–E (1–5), were initially isolated (Fig. 1). Two additional alkaloids, (+)-alstoscholarisine F (6) and (+)-G (7), were also isolated from the same A. scholaris material in 2015 (Fig. 2).4 However, since compound 6 is not closely related in structure or biosynthesis to the other alstoscholarisines, this metabolite will not be considered in this review. Shortly afterward, three more alstoscholarisines
Figure 1 Structures and absolute stereochemistry of ()-alstoscholarisines A–E.
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
3
Figure 2 Structures and stereochemistry of (+)-alstoscholarisines F–J.
((+)-H (8), (+)-I (9), and (+)-J (10)), which are biogenetically related to alstoscholarisines A–E and G, but incorporate a modified skeleton, were isolated from the leaves of A. scholaris (Fig. 2).5
2. ISOLATION AND CHARACTERIZATION 2.1 Isolation and Purification 2.1.1 (2)-Alstoscholarisines A–E Alstoscholarisines A–E could be obtained by a standard purification procedure involving acid extraction of the crude alkaloid mixture from the plant material, followed by extensive chromatography to provide samples of the pure compounds.2 Thus, leaves of A. scholaris, that were collected in Xishuangbanna, Yunnan Province, China (18 kg), were heated at reflux in EtOH (3 40 L), and the solvent was removed under vacuum. The residue was dissolved in 0.37% HCl, and the solution was basified to pH 9–10 using ammonium hydroxide. The basic solution was extracted with EtOAc and evaporated. The residue (180 g) was then dissolved in MeOH, and the resulting solution was subjected to column chromatography on silica gel eluting with CHCl3/MeOH (100% to 20:80) to afford six fractions (fractions A–F). Fraction A (25 g) was further chromatographed using CHCl3/acetone (gradient 9:1 to 7:3) as eluent to give five fractions (fractions A1–A5). Fraction A3 (1.6 g) was further purified by preparative HPLC (MeCN/ H2O, 1:1) to afford ()-alstoscholarisine A (1) (103 mg, white crystals),
ARTICLE IN PRESS 4
Jeremy D. Mason and Steven M. Weinreb
()-alstoscholarisine B (2) (8 mg, white needles), ()-alstoscholarisine C (3) (35 mg, white crystals), and ()-alstoscholarisine E (5) (26 mg, white crystals). ()-Alstoscholarisine D (4) (15 mg, white powder) was obtained in the same way from fraction C (15 g). 2.1.2 (+)-Alstoscholarisine G This alkaloid was obtained from fraction E (6.3 g) described earlier via initial purification by column chromatography followed by preparative HPLC (MeCN/H2O, 1:4) to afford (+)-alstoscholarisine G (7) (6 mg).4 2.1.3 (+)-Alstoscholarisines H–J Dried, powdered leaves (100 kg) of A. scholaris that were collected in the Simao region, Yunnan Province, China, were extracted with EtOH (3 300 L) at room temperature, and the solvent was evaporated in vacuo.5 The residue was dissolved in 0.3% HCl, and the resulting solution was basified with 5% aqueous ammonia to achieve pH 9–10. This solution was then partitioned using EtOAc, to afford a basic aqueous solution, an EtOAc fraction, and emulsion fractions. The emulsion (150 g) was subjected to column chromatography on silica gel eluting with CHCl3/MeOH (gradient) to give five fractions (fractions I–V). Fraction II (15 g) was chromatographed multiple times on silica gel eluting with CHCl3/MeOH (gradient 25:1 to 20:1) to yield (+)-alstoscholarisine H (8) (115 mg, colorless needles) and (+)alstoscholarisine I (9) (85 mg, colorless oil). Similarly, (+)-alstoscholarisine J (10) (5 mg, amorphous solid) could be obtained from fraction III (10 g).
2.2 Structure Elucidation 2.2.1 (2)-Alstoscholarisines A–E Initial UV–visible spectroscopic analysis of the major component of the alkaloid mixture, ()-alstoscholarisine A (1), indicated the presence of an indole moiety by absorptions at 233 and 287 nm in the spectrum, and HR-EIMS revealed a molecular formula of C19H24N2O (m/z 296.1890).2 Proton and carbon NMR data for these five alstoscholarisines are compiled in Tables 1 and 2, respectively. The complete structure of ()alstoscholarisine A (1) was determined by extensive two-dimensional NMR studies, and some key interactions are shown in Fig. 3. A methyl group directly attached to the indole ring was revealed by 1H–13C HMBC correlations between H6 (CH3, 2.21) and C2 (136.9), C7 (105.0), and C8 (130.2). The inclusion of an aminal moiety was assigned by 1H–13C HMBC correlation of a downfield methine (H21, 5.47) with C2 (136.9), C13 (138.5), and N-CH3 (45.6).
—
3.57 (dd, J ¼ 2.6, 10.2 Hz) 3.75 (d, J ¼ 10.5 Hz)
1.96 (m)
2.03 (m)
H-17
1.93 (m)
1.74 (br d, J ¼ 13.6 Hz)
H-14
3.03 (br s)
7.66 (d, J ¼ 7.9 Hz)
7.50 (d, J ¼ 7.9 Hz)
H-12
H-16
7.10 (t, J ¼ 7.9 Hz)
7.04 (t, J ¼ 7.9 Hz)
H-11
2.41 (m)
7.02 (t, J ¼ 7.9 Hz)
6.98 (t, J ¼ 7.9 Hz)
H-10
2.18 (m)
7.48 (d, J ¼ 7.9 Hz)
7.42 (d, J ¼ 7.9 Hz)
H-9
H-15
2.10 (s)
2.21 (s)
3.48 (d, J ¼ 10.5 Hz)
—
2.57 (m)
1.88 (m)
1.88 (m)
7.52 (d, J ¼ 7.9 Hz)
7.05 (t, J ¼ 7.9 Hz)
6.98 (t, J ¼ 7.9 Hz)
7.44 (d, J ¼ 7.9 Hz)
2.06 (s)
2.34 (m)
Alstoscholarisine E (5)a
3.50 (d, J ¼ 10.5 Hz)
—
2.57 (br s)
2.24 (m)
1.85 (m)
7.48 (d, J ¼ 7.9 Hz)
7.07 (t, J ¼ 7.9 Hz)
7.00 (t, J ¼ 7.9 Hz)
7.44 (d, J ¼ 7.9 Hz)
2.21 (s)
Continued
3.37 (dd, J ¼ 2.6, 10.2 Hz)
3.09 (br s)
2.50 (m)
2.03 (m)
1.81 (m)
7.49 (d, J ¼ 8.0 Hz)
7.04 (t, J ¼ 8.0 Hz)
6.99 (t, J ¼ 8.0 Hz)
7.41 (d, J ¼ 8.0 Hz)
2.21 (s)
2.46 (br d, J ¼ 10.5 Hz) 2.33 (dd, J ¼ 6.8, 11.3 Hz)
2.46 (br d, J ¼ 10.5 Hz) 1.85 (td, J ¼ 4.0, 11.3 Hz)
Alstoscholarisine C (3)b Alstoscholarisine D (4)a
2.22 (td, J ¼ 4.5, 12.0 Hz) 2.18 (m)
2.30 (dd, J ¼ 6.4, 12.0 Hz) 2.38 (m)
1.84 (td, J ¼ 4.0, 12.0 Hz)
Alstoscholarisine B (2)b
H-6
H-3
Proton(s) Alstoscholarisine A (1)a
Table 1 Proton NMR data for alstoscholarisines A–E
ARTICLE IN PRESS
3.80 (qd, J ¼ 3.4, 6.4 Hz)
2.33 (m)
5.57 (d, J ¼ 2.3 Hz)
2.35 (s)
3.87 (s)
3.69 (qd, J ¼ 3.0, 6.8 Hz)
2.12 (br s)
5.47 (d, J ¼ 2.6 Hz)
2.24 (s)
—
H-19
H-20
H-21
NCH3
OCH3
b
Recorded in CD3OD (600 MHz). Recorded in (CD3)2CO (600 MHz).
1.16 (d, J ¼ 6.4 Hz)
1.22 (d, J ¼ 6.8 Hz)
H-18
a
3.85 (d, J ¼ 10.5 Hz)
Alstoscholarisine B (2)
3.60 (br d, J ¼ 10.2 Hz)
Proton(s) Alstoscholarisine A (1)
Table 1 Proton NMR data for alstoscholarisines A–E—cont’d
3.82 (s)
2.31 (s) —
2.38 (s)
5.49 (d, J ¼ 2.3 Hz)
2.27 (br s)
2.28 (t, J ¼ 3.0 Hz) 5.38 (d, J ¼ 3.0 Hz)
3.94 (q, J ¼ 6.8 Hz)
1.33 (d, J ¼ 6.8 Hz)
4.10 (d, J ¼ 10.5 Hz)
Alstoscholarisine D (4)
3.94 (q, J ¼ 6.8 Hz)
1.28 (d, J ¼ 6.8 Hz)
4.04 (d, J ¼ 10.5 Hz)
Alstoscholarisine C (3)
—
2.26 (s)
5.37 (d, J ¼ 2.3 Hz)
2.16 (br s)
4.04 (q, J ¼ 6.8 Hz)
1.32 (d, J ¼ 6.8 Hz)
3.92 (dd, J ¼ 1.9, 10.2 Hz)
Alstoscholarisine E (5)
ARTICLE IN PRESS
136.9
47.3
8.1
105.0
130.2
118.7
119.8
121.3
111.3
138.5
31.1
34.5
C-2
C-3
C-6
C-7
C-8
C-9
C-10
C-11
C-12
C-13
C-14
C-15
39.2
29.0
137.3
110.8
121.2
119.4
118.4
129.6
105.3
9.1
46.3
133.8
33.5
28.7
137.5
111.1
121.1
119.3
118.3
129.5
105.1
9.0
46.0
133.9
33.1
28.5
138.1
110.9
121.7
120.1
118.7
130.8
107.1
9.8
47.1
136.7
28.5
31.0
138.7
111.6
121.3
119.8
118.6
130.2
105.0
8.0
47.1
137.0
Continued
Carbon Alstoscholarisine A (1)a Alstoscholarisine B (2)b Alstoscholarisine C (3)b Alstoscholarisine D (4)a Alstoscholarisine E (5)a
Table 2 Carbon NMR data for alstoscholarisines A–E
ARTICLE IN PRESS
173.3
52.1
43.0
67.7
45.6
—
C-20
C-21
NCH3
–CO–
OCH3 —
b
Recorded in CD3OD (150 MHz). Recorded in (CD3)2CO (150 MHz).
45.1
75.3
C-19
a
66.9
18.7
C-18
42.5
74.5
18.2
74.8
74.3
C-17
48.8
35.8
Alstoscholarisine B (2)
C-16
Carbon Alstoscholarisine A (1)
Table 2 Carbon NMR data for alstoscholarisines A–E—cont’d
52.2
173.4
45.0
71.8
42.9
72.1
17.9
68.9
49.1
Alstoscholarisine C (3)
—
179.3
44.5
71.9
42.9
72.9
18.2
70.6
50.8
Alstoscholarisine D (4)
—
—
45.5
72.4
43.4
73.3
18.3
68.8
36.0
Alstoscholarisine E (5)
ARTICLE IN PRESS
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
9
Figure 3 Key two-dimensional NMR correlations for ()-alstoscholarisine A (1): (left) 1 H–1H COSY ( ); (center) 1H–13C HMBC ( ); (right) 1H–1H ROESY ( ).
The presence of an N-methylpiperidine ring (D-ring) was determined based on a 1H–1H COSY correlation between H21 (5.47) and H20 (2.12), H20 with H15 (2.18), H15 with H14 (1.74, 2.03), and H14 with H3 (1.84, 2.30), as well as an HMBC cross-peak between the N-Me (2.24) and C3 (47.3). An HMBC correlation of the H16 methine (3.03) with C2 (136.9) and C7 (105.0) indicated that C16 was attached at C2 of the indole ring. Incorporation of a six-membered ring (C-ring) was then elucidated by observation of a COSY relationship between H15 (2.18) and H16 (3.03), as well as the previously discussed correlations between H21, H20, and H15. The presence of an α-methyltetrahydropyran ring was established by COSY cross-peaks between H16 (3.03) and H17 (3.57, 3.60), H20 (2.12) and H19 (3.69), and H19 with the upfield methyl group (H18, 1.22), as well as the downfield chemical shifts of C17 and C19, which indicated the presence of an oxygen in the ring. Furthermore, the H18 methyl group was assigned as being in an equatorial position on the tetrahydropyran ring due to an NOE cross-peak between this signal and H21 in the ROESY spectrum. The structure of ()-alstoscholarisine A (1) was also unambiguously confirmed by single-crystal X-ray analysis. ()-Alstoscholarisine B (2) showed an HR-EIMS peak 58amu greater (m/ z 354.1940) than that for alkaloid 1, which indicated a molecular formula of C21H26N2O3. The IR spectrum (1730 cm1) and the 13C NMR spectrum (173.3) of metabolite 2 indicated the presence of an ester moiety. The COSY and HMBC spectra of 2 showed the same correlations as those discussed earlier for alkaloid 1, except for the absence of the C16 methine hydrogen. Furthermore, an NOE correlation between the H18 methyl group and the H21 aminal proton was retained, which indicated an equatorial methyl group, as in compound 1. It was therefore determined that alkaloid 2 is a structural analogue of ()-alstoscholarisine A (1) containing a methyl ester group at C16. ()-Alstoscholarisine C (3) had the same HR-EIMS peak (m/z 354.1946) as compound 2, indicating the possibility that the two alkaloids are stereoisomers. The HMBC and COSY spectra for metabolite 3 were analogous to
ARTICLE IN PRESS 10
Figure 4 Key 1H–1H ROESY (
Jeremy D. Mason and Steven M. Weinreb
) NMR correlations for ()-alstoscholarisine C (3).
those for 2. However, the ROESY spectrum of ()-alstoscholarisine C (3) revealed an NOE correlation of the H18 methyl group with H15 and H17, indicating an axial methyl group on the tetrahydropyran ring (Fig. 4). Furthermore, the H19 methine showed an ROESY cross-peak with the H21 aminal. In addition, the structure of ()-alstoscholarisine C (3) was unambiguously determined by single-crystal X-ray analysis. ()-Alstoscholarisine D (4) showed an HR-EIMS peak at m/z 340.1780, 14 amu less than that for alkaloids 2 and 3, indicating the possibility that this compound was the methyl ester hydrolysis product of 2 or 3. Indeed, the IR spectrum of metabolite 4 supported the presence of a carboxylic acid moiety (3418 cm1). ()-Alstoscholarisine D (4) showed the same HMBC and COSY correlations as seen for alkaloids 2 and 3. Finally, the methyl group at C19 was determined to be axial due to observed NOE interactions of H18 with H15 and H17, as well as those between H19 and H21. ()-Alstoscholarisine E (5) had the same molecular mass as alkaloid 1 (m/z 296.1881), as well as analogous HMBC and COSY spectra. However, compound 5 was assigned as having an axial methyl group at C19, based on the same type of NOE interactions as discussed for metabolites 3 and 4. The absolute configuration of ()-alstoscholarisine A (1) was determined to be (15R,16R,19R,20S,21S) by single-crystal X-ray analysis utilizing the method developed by Hooft and coworkers.6 The absolute configurations of alkaloids 2–5 were assigned based on the similarity of their UV-circular dichroism spectra with that of ()-alstoscholarisine A (1). 2.2.2 (+)-Alstoscholarisine G HR-TOFMS of alstoscholarisine G (7) indicated a formula of C20H24N3O2 ([M + H]+ 340.2022).4 UV–visible spectroscopic analysis of alkaloid 7 revealed the presence of an indole ring system based on absorptions at 232 and 286 nm. The IR spectrum revealed the presence of an amide or lactam moiety (1683, 3437 cm1). Proton and carbon NMR data for alstoscholarisine G (7) are summarized in Table 3. A δ-lactam ring fused to the indole moiety was elucidated based
ARTICLE IN PRESS 11
The Alstoscholarisine Alkaloids
Table 3 Proton and carbon NMR data for alstoscholarisine G
Proton(s)
H-3
H-6
δHa
Carbon
δCb
2.12 (m)
C-2
135.3
2.53 (m)
C-3
52.3
4.80 (d, J ¼ 15.0 Hz)
C-6
40.2
4.90 (d, J ¼ 15.0 Hz)
C-7
107.2
C-8
125.8
H-9
7.63 (d, J ¼ 7.5 Hz)
C-9
118.1
H-10
7.23 (t, J ¼ 7.5 Hz)
C-10
119.5
H-11
7.26 (t, J ¼ 7.5 Hz)
C-11
121.9
H-12
7.57 (d, J ¼ 7.5 Hz)
C-12
112.0
H-14
2.04 (m)
C-13
138.2
2.40 (br d, J ¼ 13.9 Hz)
C-14
28.4
H-15
3.80 (br d, J ¼ 6.4 Hz)
C-15
39.0
H-17
4.65 (d, J ¼ 10.5 Hz)
C-16
54.4
4.98 (d, J ¼ 10.5 Hz)
C-17
66.1
H-18
1.46 (d, J ¼ 6.8 Hz)
C-18
13.8
H-19
5.50 (q, J ¼ 6.8 Hz)
C-19
126.2
H-21
2.91 (d, J ¼ 12.0 Hz)
C-20
135.0
3.10 (d, J ¼ 12.0 Hz)
C-21
62.5
1.96 (s)
C-22
174.7
NCH3 N1 H CONH a
12.10 (s) 9.28 (br s)
Recorded in pyridine-d5 (600 MHz). Recorded in pyridine-d5 (150 MHz).
b
NCH3
45.6
ARTICLE IN PRESS 12
Jeremy D. Mason and Steven M. Weinreb
Figure 5 Key two-dimensional NMR correlations for (+)-alstoscholarisine G (7): (left) 1 H–13C HMBC ( ); (right) 1H–1H ROESY ( ).
on the presence of HMBC correlations between a lactam NH proton (9.28) and C6 (40.2), C7 (107.2), C16 (54.4), and C22 (174.7), as well as those between H6 (4.80–4.90) with C2 (135.3), C7 (107.2), C8 (125.8), and C22 (174.7) (Fig. 5). An N-methyl-4-piperidinyl unit connected to this lactam at C16 was assigned based on HMBC correlations of a deshielded N-methyl group (1.96) with C3 (52.3), C14 (28.4), C20 (135.0), and C21 (62.5), as well as those between H15 (3.80) and C3 (52.3), C14 (28.4), C16 (54.4), C19 (126.2), C20 (135.0), and C21 (62.5). An ethylidene moiety at C20 of the N-methylpiperidine ring was assigned based on HMBC cross-peaks between H19 (5.50) and C15 (39.0) and C21 (62.5), as well as those between an allylic methyl group (H18, 1.36) and C19 (126.2) and C20 (135.0). Finally, a hydroxymethyl group attached at C16 was assigned by HMBC correlations between the downfield protons at C17 (4.65, 4.98) with C2 (135.3), C15 (39.0), C16 (54.4), and C22 (174.7). The double bond geometry of the ethylidene group was assigned as (E) due to NOE correlations between H18 (1.46) and H15 (3.80) as well as those between H19 (5.50) and H21 (2.91, 3.10) in the ROESY spectrum. The relative and absolute stereochemistry of (+)-alstoscholarisine G (7) is assumed to be (15S,16S) due to the putative biosynthetic pathway originating from (+)-(E)-vallesamine (vide infra), although this assumption has not been verified. 2.2.3 (+)-Alstoscholarisines H–J (+)-Alstoscholarisine H (8) showed an HR-EIMS peak at m/z 296.1886, indicating a formula of C19H24N2O.5 Complete 1H and 13C data for alstoscholarisines H–J are summarized in Tables 4 and 5, respectively. As
ARTICLE IN PRESS 13
The Alstoscholarisine Alkaloids
Table 4 Proton NMR data for alstoscholarisines H–J
(+)-Alstoscholarisine Proton(s) H (8)a
(+)-Alstoscholarisine I (9)b
(+)-Alstoscholarisine J (10)c
H-3
5.34 (t, J ¼ 2.8 Hz)
5.45 (t, J ¼ 2.6 Hz)
5.39 (t, J ¼ 3.0 Hz)
H-6
2.38 (s)
2.38 (s)
2.49 (s)
H-9
7.54 (d, J ¼ 7.5 Hz)
7.46 (d, J ¼ 7.5 Hz)
7.52 (d, J ¼ 7.5 Hz)
H-10
7.14 (t, J ¼ 7.5 Hz)
7.02 (t, J ¼ 7.5 Hz)
7.05 (t, J ¼ 7.5 Hz)
H-11
7.19 (t, J ¼ 7.5 Hz)
7.12 (t, J ¼ 7.5 Hz)
7.14 (t, J ¼ 7.5 Hz)
H-12
7.42 (d, J ¼ 7.5 Hz)
7.47 (d, J ¼ 7.5 Hz)
7.50 (d, J ¼ 7.5 Hz)
H-14
2.02 (dt, J ¼ 2.8, 12.8 Hz)
2.16 (m)
2.05 (dt, J ¼ 3.0, 12.9 Hz)
2.31 (dt, J ¼ 2.8, 12.8 Hz)
2.64 (m)
2.44 (dt, J ¼ 3.0, 12.9 Hz)
H-15
3.52 (m)
3.65 (m)
3.26 (t, J ¼ 3.0 Hz)
H-16
3.60 (m)
—
—
H-17
3.70 (t, J ¼ 10.5 Hz)
3.68 (d, J ¼ 10.9 Hz)
4.99 (s)
4.30 (dd, J ¼ 3.5, 10.5 Hz)
4.65 (br d, J ¼ 10.9 Hz)
—
H-18
1.84 (d, J ¼ 6.5 Hz)
1.79 (d, J ¼ 6.8 Hz)
1.75 (d, J ¼ 6.8 Hz)
H-19
5.67 (q, J ¼ 6.5 Hz)
5.60 (q, J ¼ 6.8 Hz)
5.40 (q, J ¼ 6.8 Hz)
H-21
2.90 (d, J ¼ 12.5 Hz)
2.68 (d, J ¼ 12.8 Hz)
2.78 (d, J ¼ 12.5 Hz)
2.99 (d, J ¼ 12.5 Hz)
2.87 (d, J ¼ 12.8 Hz)
3.06 (d, J ¼ 12.5 Hz)
NCH3
2.45 (s)
2.21 (s)
2.32 (s)
OCH3
—
3.66 (s)
3.38 (s)
OCH3
—
—
3.59 (s)
a
Recorded in CDCl3 (400 MHz). Recorded in CD3OD (400 MHz). Recorded in CD3OD (600 MHz).
b c
ARTICLE IN PRESS 14
Jeremy D. Mason and Steven M. Weinreb
Table 5 Carbon NMR data for alstoscholarisines H–J
(+)-Alstoscholarisine Carbon H (8)a
(+)-Alstoscholarisine I (9)b
(+)-Alstoscholarisine J (10)c
C-2
132.0
130.0
135.2
C-3
67.2
57.6
68.3
C-6
10.1
10.7
10.6
C-7
106.6
110.3
110.0
C-8
128.9
130.1
130.4
C-9
117.7
119.1
119.4
C-10
119.2
120.3
120.4
C-11
121.0
122.7
123.0
C-12
109.6
111.4
111.5
C-13
136.9
138.2
138.4
C-14
33.9
32.7
31.7
C-15
28.0
33.4
38.5
C-16
40.2
54.0
76.9
C-17
63.0
65.1
111.5
C-18
12.9
13.5
13.4
C-19
122.1
124.8
122.2
C-20
135.0
134.9
135.8
C-21
56.5
57.6
58.0
NMe
43.9
44.0
44.0
–CO–
—
175.1
OCH3
—
52.8
58.1
OCH3
—
—
59.1
a
Recorded in CDCl3 (100 MHz). Recorded in CD3OD (100 MHz). c Recorded in CD3OD (150 MHz). b
—
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
15
Figure 6 Key two-dimensional NMR correlations for ()-alstoscholarisine H (8): (left) 1 H–1H COSY ( ); (center) 1H–13C HMBC ( ); (right) 1H–1H ROESY ( ).
in alstoscholarisines A–E, a methyl group attached to an indole system was assigned based on HMBC correlations between H6 (10.1) and C2 (132.0), C7 (106.6), and C8 (128.9). An indole aminal moiety was revealed by HMBC cross-peaks between a downfield methine H3 (5.34) and C2 (132.0) and C13 (136.9) (Fig. 6). A six-membered ring (C-ring) was elucidated based on 1H–1H COSY correlation between H3 (5.34) and H14 (2.02, 2.31), H14 with H15 (3.52), and H15 with H16 (3.60), as well as HMBC cross-peaks between H16 (3.60) with C2 (132.0) and C15 (28.0). An N-methyl piperidine moiety (D-ring) was assigned by HMBC correlations between the N-Me (2.45) and C3 (67.2) and C21 (56.5), H15 (3.52) with C14 (33.9) and C20 (135.0), as well as the previously discussed COSY interactions between H15/H14/H3. An ethylidene group on this piperidine ring was revealed by HMBC correlations of H19 (5.67) with C15 (28.0) and C21 (56.5), as well as those between H18 (1.84) and C20 (135.0). A hydroxymethyl unit attached at C16 was assigned based on HMBC cross-peaks of deshielded methylene protons H17 (3.70, 4.30) with C2 (132.0), C15 (28.0), and C16 (40.2), as well as COSY correlation between H17 and H16 (3.60). The relative stereochemistry at C16 was determined by NOE correlation between H16 (3.60) and H14 (2.02, 2.31) in the ROESY spectrum. The stereochemistry of the aforementioned ethylidene group was assigned as (E) based upon NOE relationships between H18 (1.84) and H15 (3.52), as well as those between H19 (5.67) and H21 (2.90, 2.99). Finally, the complete structure of metabolite 8 was unambiguously established by X-ray crystallographic analysis. (+)-Alstoscholarisine I (9) showed an HR-EIMS peak at m/z 354.1940, 58 amu greater than that of (+)-alstoscholarisine H (8), which indicated the possibility that compound 9 was an analogue of 8 containing a methyl ester moiety. Indeed, the IR spectrum (1728 cm1) and 13C NMR spectrum (175.1) of this metabolite showed the presence of an ester carbonyl group.
ARTICLE IN PRESS 16
Jeremy D. Mason and Steven M. Weinreb
Furthermore, the one-dimensional proton NMR spectrum was very similar to that of (+)-alstoscholarisine H (8), except for the absence of the C16 methine proton and the addition of a methoxyl peak (3.66). The location of the methyl ester group attached at C16 was further supported by HMBC correlations between the hydroxymethyl methylene protons (H17, 3.68, 4.65) and C16 (54.0) and the ester carbonyl carbon (175.1) (Fig. 7). The stereochemistry at C16 was assigned based upon NOE interactions between the methoxyl peak (3.66) and H14 (2.64), and the ethylidene group geometry was assigned as (E) based on analogous correlations to those described earlier for alstoscholarisine H (8). (+)-Alstoscholarisine J (10) showed an HR-EIMS peak 60 amu greater than alstoscholarisine H (8) (m/z 356.2094, C21H28N2O3). The onedimensional proton NMR spectrum of the alkaloid was similar to that of 8, except for the absence of the C16 methine and C17 methylene protons and the addition of a downfield methine at 4.99 ppm and two methoxy peaks at 3.38 and 3.59 ppm. In the 13C spectrum, the signals corresponding to C16 and C17 in alstoscholarisine H were absent, and instead four new signals were observed: a quaternary carbon at 76.9 ppm, a tertiary carbon at 111.5 ppm, and two primary carbons at 58.1 and 59.1 ppm. These data led to the proposal of a dimethyl acetal at C17 and a hydroxyl group at C16. The stereochemistry at C16 was determined by the presence of an NOE cross-peak between the allylic methyl group (H18, 1.75) and one of the methoxyl group signals (3.59) (Fig. 8). The ethylidene stereochemistry was again confirmed as (E) by analogous ROESY cross-peaks as described earlier for alstoscholarisines H and I. Based upon the biogenetic hypothesis for these alkaloids as originating from (+)-(E)-vallesamine (vide infra), the absolute configurations of compounds (+)-8, (+)-9, and (+)-10 are assumed to be (3R,15R,16R), (3R,15S,16S), and (3R,15S,16S), respectively, although this supposition has not yet been firmly established.
Figure 7 Key two-dimensional NMR correlations for (+)-alstoscholarisine I (9): (left) 1 H–13C HMBC ( ); (right) 1H–1H ROESY ( ).
ARTICLE IN PRESS 17
The Alstoscholarisine Alkaloids
Figure 8 Key 1H–1H ROESY (
) correlations for alstoscholarisine J (10).
3. PUTATIVE BIOGENESIS 3.1 Biosynthesis of (2)-Alstoscholarisines A–E Although there are no experimental studies to date, ()-alstoscholarisines C, D, and E most likely arise biogenetically from the well-known monoterpenoid indole alkaloid (+)-stemmadenine (13) which is formed via an established pathway from two common monoterpenoid indole alkaloid precursors, tryptamine (11) and secologanin (12) (Scheme 1).2,7 (+)-Stemmadinine is first transformed to (+)-(E)-vallesamine (14), a known alkaloid that has been isolated previously from A. scholaris. It might be noted that the conversion of stemmadenine to vallesamine, which occurs by an established sequence of steps,8 serves to excise one of the two side-chain carbons of the original tryptamine building block. (+)-(E)-Vallesamine (14) could then be N-methylated to generate quaternary ammonium salt 15, which could undergo ring cleavage to form azafulvene 16. This pivotal intermediate (vide infra) could undergo reduction of the azafulvene moiety and oxidation of the piperidine ring to generate (E)-α,β-unsaturated iminium ion (E)-17, which would undergo a stereospecific conjugate addition of the attendant alcohol moiety to form the tetrahydropyran ring with the axial α-methyl group in tetracyclic iminium ion 18. A final nucleophilic addition of the indole nitrogen onto iminium ion 18 would provide ()-alstoscholarisine C (3). Ester hydrolysis would generate the corresponding carboxylic acid, ()-alstoscholarisine D (4), which could undergo decarboxylation to form ()-alstoscholarisine E (5). The biosynthesis of ()-alstoscholarisines A (1) and B (2) is proposed by Luo and coworkers2 to involve (+)-(Z)-vallesamine (19), an alkaloid which is also found in A. scholaris (Scheme 2). Thus, by the same pathway described earlier originating from (+)-(E)-vallesamine (14), (+)-(Z)-vallesamine (19) could be transformed into (Z)-α,β-unsaturated iminium ion (Z)-17.
ARTICLE IN PRESS 18
Jeremy D. Mason and Steven M. Weinreb
Scheme 1 Proposed biosynthesis of ()-alstoscholarisines C (3), D (4), and E (5) via (+)stemmadenine (13) and (+)-(E)-vallesamine (14).
Scheme 2 Putative biosynthesis of ()-alstoscholarisines A (1) and B (2) from (+)-(Z)vallesamine (19).
This intermediate could undergo an analogous stereospecific cyclization sequence as described earlier for (E)-17 to ultimately generate ()alstoscholarisine B (2) having an equatorial α-methyl group in the tetrahydropyran ring. Hydrolysis of the methyl ester and decarboxylation of the intermediate carboxylic acid would produce ()-alstoscholarisine A (1).
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
19
Scheme 3 Biosynthetic hypothesis for (+)-alstoscholarisine G (7).
3.2 Biosynthesis of (+)-Alstoscholarisine G Alstoscholarisines G–J are thought to all arise from the azafulvene 16 derived from (+)-(E)-vallesamine (14) (Scheme 3).4,8 Thus, (+)-alstoscholarisine G (7) is proposed to be formed by conjugate addition of ammonia (or an ammonia equivalent) to this intermediate to produce an aminomethyl indole like 20, which cyclizes onto the attendant methyl ester to form the δ-lactam ring of the alkaloid.
3.3 Biosynthesis of (+)-Alstoscholarisines H–J In an alternative biogenetic pathway leading to (+)-alstoscholarisines H–J, the azafulvene moiety of intermediate 16 could be reduced, and the piperidine ring could then be oxidized to form iminium ion 21 (Scheme 4), a regioisomer of iminium ion 17 (cf. Scheme 2).5 This intermediate could undergo 1,2-addition of the indole nitrogen onto the iminium ion to generate (+)alstoscholarisine I (9). Alkaloid 9 could then undergo methyl ester hydrolysis and subsequent decarboxylation to form (+)-alstoscholarisine H (8). Finally, metabolite 8 could be oxidized to form the α-hydroxyaldehyde 22, which would produce (+)-alstoscholarisine J (10) by formation of the corresponding dimethyl acetal.
ARTICLE IN PRESS 20
Jeremy D. Mason and Steven M. Weinreb
Scheme 4 Proposed biosynthesis of (+)-alstoscholarisines H (8), I (9), and J (10).
4. BIOLOGICAL ACTIVITY The primary biological activity associated with alstoscholarisines A–E (1–5) is their ability to promote neuronal stem cell (NSC) proliferation and differentiation.2 Neural stem cells have shown promise for the discovery of new CNS therapies potentially of use for treatment of neurodegenerative ailments.9 In recent years, there has been a search for small molecules and natural products with the ability to modulate stem cell fate and population, and such compounds may play a key role in the development of treatments for various neurogenic disorders such as Alzheimer’s disease, Parkinson’s disease, and Huntington’s disease.10 For example, several phase I/II clinical trials using stem cell strategies are currently underway for Alzheimer’s disease.11 ()-Alstoscholarisine A (1) is the most active of the five alkaloids, promoting NSC proliferation at a dosage-dependent concentration of 0.1 μg/mL. Compound 1 also enhanced NSC sphere formation and partial neurogenic fate commitment. It was reported that ()-alstoscholarisine G (7) has no significant activity in enhancing hippocampal NSC proliferation in vitro.4 Moreover, ()-alstoscholarisines H (8), I (9), and J (10) were also found to be inactive toward both NSC proliferation and PC12 cell differentiation (neurite outgrowth-promoting activity).5 In addition, these latter three alkaloids did not demonstrate any cytotoxic activity.
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
21
5. TOTAL SYNTHESIS 5.1 Syntheses of Alstoscholarisines A and E 5.1.1 Bihelovic/Ferjancic Approach to ()-Alstoscholarisine A In 2016, Bihelovic and Ferjancic completed a total synthesis of ()alstoscholarisine A (1) involving 12 steps starting from readily available compounds.12 The synthesis began with an aldol reaction between the enolate of indole ester 23 and protected β-aminoaldehyde 24 to provide adduct 25, which was dehydrated with concomitant Boc removal to form condensation product 26 as a single (E)-isomer (Scheme 5). Removal of the allyl carbamate protecting group of 26 with Pd(OAc)2/morpholine then gave secondary amine 27. Condensation of amine 27 and γ-phenylselenylaldehyde 28 formed enamine 29, which subsequently cyclized via an intramolecular Michael addition to the α,β-unsaturated ester moiety to provide iminium ion 30. This intermediate reacted further with the indole nitrogen to form bridged aminal 31/32 as a separable mixture of C16 ester epimers (dr 1:1).
Scheme 5 Synthesis of key bridged tetracyclic aminal 33.
ARTICLE IN PRESS 22
Jeremy D. Mason and Steven M. Weinreb
Equilibration of this mixture of esters with DBU in ethanol improved the diastereomeric ratio to 5:1, with the major epimer being the desired exo-ester 32 needed for the synthesis. Oxidation of selenide 32 to the corresponding selenoxide with m-CPBA and subsequent elimination provided the requisite terminal alkene 33. Reduction of the ester moiety of 33 with DIBALH gave alcohol 34 (Scheme 6). An Upjohn dihydroxylation13 of 34, followed by oxidative diol cleavage with Pb(OAc)4,14 yielded axial aldehyde 35, which was epimerized under basic conditions with DBU to form equatorial aldehyde 36. This intermediate spontaneously cyclized to lactol 37, which was immediately oxidized with Dess–Martin periodinane to produce δ-lactone 38. A key sequence in the total synthesis was construction of the tetrahydropyran moiety with the requisite configuration of the α-methyl substituent using a strategy that has been widely applied in this field.15,16 Thus, addition of one equivalent of methyllithium to this lactone 38 could first be effected to provide hemiketal 39 (single diastereomer, configuration not determined). This intermediate was further transformed to ()-alstoscholarisine A (1) by
Scheme 6 Completion of the Bihelovic/Ferjancic total synthesis of ()-alstoscholarisine A (1).
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
23
initial conversion to an intermediate oxocarbenium ion 40 through ionization with trimethylsilyl trifluoromethanesulfonate (TMSOTf ), followed by a reduction from the least hindered convex face using triethylsilane as the hydride source. It should be noted that this stereochemical outcome is also consistent with a model predicting axial attack of hydride from triethylsilane based on the intermediacy of a favored chair-like transition state.17 5.1.2 Enantioselective Synthesis of (2)-Alstoscholarisine A by Yang et al. In 2016, Yang and coworkers reported an enantioselective total synthesis of ()-alstoscholarisine A (1) which required 12 steps.18 The synthesis commenced with an acylation of 3-methylindole (42) with vinyl-γ-lactone 41 to provide N-acylindole 43 (Scheme 7). This compound underwent a key enantioselective intramolecular Friedel–Crafts alkylation mediated by
Scheme 7 Early stages of the Yang total synthesis of ()-alstoscholarisine A (1).
ARTICLE IN PRESS 24
Jeremy D. Mason and Steven M. Weinreb
Scheme 8 Yang total synthesis of ()-alstoscholarisine A (1).
Carreira’s iridium catalyst system using the chiral ligand 44 and Sc(OTf )3 to afford tricycle 45 with high enantioselectivity (>99:1 er) and in good yield.19 Dihydroxylation of olefin 45 provided diol 46, which was not isolated but was directly protected as the acetonide 47 (5:1 mixture of diastereomers). The lactam moiety of 47 was next deprotonated with lithium hexamethyldisilazide (LiHMDS), and the resulting enolate was trapped with PhSeBr to form selenide 48. This selenide was then oxidized to the corresponding selenoxide, which underwent elimination to afford α,β-unsaturated lactam 49. Conjugate addition of vinyl cuprate reagent 50 to α,β-unsaturated lactam 49 from the re face and subsequent trapping of the intermediate enolate with acetaldehyde provided alcohol 52 with good (8:1) diastereoselectivity (Scheme 8). The diastereoselectivity of this aldol addition is in accord with a Zimmerman–Traxler chair-like transition state 51, which predicts the antiadduct 52 being the major diastereomeric product.20 Ketal removal with methanolic HCl then resulted in triol 53. Oxidative cleavage of this intermediate with NaIO4 formed aldehyde 54, which underwent spontaneous cyclization to give lactol 55.
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
25
Scheme 9 Completion of the total synthesis of ()-alstoscholarisine A (1).
Lactol 55 was then reduced to tetrahydropyran 56 with Et3SiH/ BF3Et2O (Scheme 9). Hydroboration/oxidation of terminal olefin 56 provided alcohol 57, which was oxidized with Dess–Martin periodinane to produce aldehyde 58. The synthesis was completed by condensation of aldehyde 58 with methylamine to generate intermediate imine 59, followed by reduction of both the imine and indole lactam moieties with LiAlH4 to provide hemiaminal 60, which cyclized directly to form ()-alstoscholarisine A (1). 5.1.3 Mason/Weinreb Synthesis of ()-Alstoscholarisines A and E Mason and Weinreb have reported total syntheses of racemic alstoscholarisines A–E.21 A key intermediate in the synthesis of all five of the alkaloids was bridged aminal 68, prepared in about six operations from readily available materials as discussed below: Thus, indole ester 61 was first deprotonated with LiHMDS at low temperature and α,β-unsaturated-Nsulfonyllactam 62 was added to the resulting enolate, leading to the Michael adduct 63 in good yield as a complex mixture of stereoisomers (Scheme 10). Lactam ester 63 subsequently could be C-allylated using allyl iodide/ K2CO3 in acetonitrile to form 64, again as a complex mixture, followed by in situ addition of Pd(PPh3)4/morpholine to promote decarboxylation of the allyl ester, affording the desired C-allylated lactam 65. This product was formed exclusively as the more stable C15, C20 trans isomer shown, but was a mixture of stereoisomers at the C16 ester-bearing center. N-Sulfonyllactam 65 was then partially reduced to the N,O-hemiacetal 66 using DIBALH. Exposure of this material to trifluoroacetic acid at room temperature resulted in the removal of the Boc group on the indole nitrogen and subsequent closure via an N-sulfonyliminium ion22 to afford
ARTICLE IN PRESS 26
Jeremy D. Mason and Steven M. Weinreb
Scheme 10 Construction of bridged aminal ester 68.
the desired bridged aminal 67 in good yield as a mixture of C16 ester stereoisomers. However, this mixture could be readily epimerized with sodium methoxide/methanol to yield exclusively the more stable ester exoisomer 68. In order to continue the synthesis, the terminal alkene double bond of intermediate 68 needed to be isomerized to an internal position. Two strategies were developed for this transformation, leading to intermediates which were subsequently processed to form alstoscholarisines A and E.21 In the first approach, alkene ester 68 was first reduced to afford alcohol 69 which was then protected as silyl ether 70 (Scheme 11). Heating compound 70 with DEAD in refluxing toluene effected an ene reaction, producing the (E)-allylic hydrazine derivative 71 (along with a by-product derived from the reaction of DEAD with ene product 71 at the C3 position of the indole).23 After removal of the silyl protecting group of compound 71 with fluoride, the resulting internal hydroxyalkene 72 could be cleaved by a two-step procedure to yield axial hydroxyaldehyde 73. Exposure of this
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
27
Scheme 11 Initial route to lactol acetate 77 and δ-lactone 78.
Scheme 12 Alternative route to aldehyde intermediate 73.
intermediate to DBU led to epimerization, followed by spontaneous cyclization of the resulting equatorial hydroxyaldehyde to afford the desired lactol 76, along with some of the ring-opened product 74. Without purification, this mixture could be acylated with acetic anhydride to form lactol acetate 77 as well as acetate 75. Alternatively, the crude lactol mixture could be oxidized to produce δ-lactone 78. A second strategy for double bond isomerization involved heating terminal alkene 68 with the Grubbs II ruthenium metathesis catalyst in methanol,21b,24 leading to propenyl derivative 79 in excellent yield (2.8:1 E/Z mixture) (Scheme 12). The ester group of 79 could subsequently be reduced with lithium aluminum hydride to generate alcohol
ARTICLE IN PRESS 28
Jeremy D. Mason and Steven M. Weinreb
80. The same two-step alkene cleavage protocol used earlier then led to the formation of axial hydroxyaldehyde 73, which was processed into lactol acetate 77 and δ-lactone 78 in somewhat better yields compared to those obtained from the ene-based procedure shown in Scheme 11. In addition, this latter sequence was more efficient overall in that alcohol protection and deprotection was not required, and formation of the ene reaction by-product noted above was avoided. Conversion of δ-lactone 78 to alstoscholarisine A made use of the same strategy previously applied by Bihelovic and Ferjancic for elaborating the tetrahydropyran ring of the metabolite.12 Thus, addition of methyllithium to lactone 78 afforded hemiketal 81, which upon treatment with trimethylsilyl triflate/triethylsilane in methylene chloride at low temperature presumably generated intermediate oxocarbenium ion 82, which upon silane reduction from the least hindered face gave the α-methyltetrahydropyran 83 as a single stereoisomer in good yield (Scheme 13).15,16 Finally, reductive removal of the tosyl group of 8325 produced amine 84 and subsequent N-methylation26 with formalin and sodium cyanoborohydride yielded ()-alstoscholarisine A (1). A synthesis of alstoscholarisine E (5) could be executed by simply reversing the order of steps for generating the α-stereogenic center of the tetrahydropyran. Therefore, exposing lactol acetate 77 to trimethylsilyl trifluoromethanesulfonate and trimethylaluminum in methylene chloride at 78°C resulted in transfer of a methyl group from the aluminum reagent to the least hindered face of putative oxocarbenium ion 85 to afford tetrahydropyran 86 in good yield as a single stereoisomer (Scheme 14).15,16 The tosyl group could then be removed from intermediate 86 using Mg/MeOH,25 and the resulting amine 87 was N-methylated26 to afford ()-alstoscholarisine E (5).
Scheme 13 Conversion of δ-lactone 78 to ()-alstoscholarisine A (1).
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
29
Scheme 14 Conversion of lactol acetate 77 to ()-alstoscholarisine E (5).
Scheme 15 Construction of lactol acetate 94 and δ-lactone 95.
5.2 Weinreb/Mason Synthesis of ()-Alstoscholarisines B, C, and D Intermediate bridged aminal ester 67 has also been used by Mason and Weinreb in total syntheses of racemic alstoscholarisines B, C, and D.21 This ester could be deprotonated with potassium hexamethyldisilazide (KHMDS) and the resulting enolate was alkylated from the least hindered face using anhydrous formaldehyde27 to give hydroxymethyl compound 88 in high yield as a single stereoisomer having the configuration shown in Scheme 15. The alcohol group of 88 was first protected as the silyl ether 89, and a subsequent thermal ene reaction with DEAD occurred smoothly
ARTICLE IN PRESS 30
Jeremy D. Mason and Steven M. Weinreb
to afford (E)-allylic hydrazine derivative 90 in good yield. Oxidative cleavage of the alkene double bond of intermediate 90 led to axial aldehyde 91, which upon exposure to HCl underwent removal of the TBS protecting group to produce hydroxyaldehyde 92. The tetrahydropyran moiety of the alkaloids was then constructed from this compound using a strategy similar to that previously applied for alstoscholarisines A and E (vide supra). Thus, epimerization of aldehyde 92 with DBU resulted in the formation of lactol 93, which could be acylated to generate lactol acetate 94. In addition, oxidation of lactol 93 produced δ-lactone 95. An alternative, more efficient, and higher yielding route to the key intermediates 94 and 95 was also developed utilizing the approach previously applied to alstoscholarisines A and E (cf. Scheme 12).21b Therefore, an isomerization of the allyl group of 88 could be effected using the Grubbs II metathesis catalyst in MeOH24 to afford propenyl compound 96 (3.6:1 E/Z) in high yield (Scheme 16). This compound could then be processed via an oxidative cleavage of the double bond to eventually produce both the lactol acetate 94 and the δ-lactone 95 in good yields. Once again, this sequence precluded silyl protection of the alcohol moiety of intermediate 88. The synthesis of alstoscholarisine B (2) made use of the δ-lactone 95 (Scheme 17).21 Thus, methyllithium could be added selectively to the lactone carbonyl group of 95 at 78°C to afford hemiketal 97 as a single
Scheme 16 Alternate route to lactol acetate 94 and δ-lactone 95.
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
31
Scheme 17 Conversion of δ-lactone 95 to ()-alstoscholarisine B.
stereoisomer (configuration not determined). Treatment of this compound with trimethylsilyl trifluoromethanesulfonate and triethylsilane15,16 at 78°C presumably generates an oxocarbenium ion intermediate which is then reduced from the less hindered face to afford the equatorial α-methyltetrahydropyran 98 as a single stereoisomer. Removal of the tosyl group from 98 with Mg/MeOH25 and subsequent N-methylation26 of the resulting secondary amine 99 led to ()-alstoscholarisine B (2). For the synthesis of alstoscholarisine C (3), lactol acetate 94 was first treated with a mixture of trimethylsilyl trifluoromethanesulfonate and trimethylaluminum15,16 to afford α-methyltetrahydropyran 100 as a single stereoisomer having the requisite configuration as shown (Scheme 18). To complete the synthesis of the alkaloid, the N-Ts group of intermediate 100 was removed reductively with magnesium metal in methanol25 providing amine 101, followed by N-methylation,26 to give ()-alstoscholarisine C (3). A basic hydrolysis of the methyl ester with sodium hydroxide in aqueous ethanol at 70°C served to transform alstoscholarisine C (3) into alstoscholarisine D (4) in good yield.
5.3 Luo/Xia Synthesis of ()-Alstoscholarisine H Luo, Xia, and coworkers have described a short five-step total synthesis of racemic alstoscholarisine H (8).5 In this approach, indole ester 102 was first
ARTICLE IN PRESS 32
Jeremy D. Mason and Steven M. Weinreb
Scheme 18 Conversion of lactol acetate 94 to ()-alstoscholarisines C and D.
deprotonated with LDA and the resulting enolate was added to the pyridinium salt 103, followed by treatment of the crude product with triflic acid. This sequence allowed a one-pot synthesis of a separable mixture of epimeric adducts 104 and 105 (5.6:1 ratio) containing the requisite bridged aminal ring system via a formal [3 + 3]-cycloaddition process. The actual isolated yield for this transformation is low, but is more reasonable when calculated based on recovered starting material. This mixture of epimeric esters (or either one individually) could be hydrolyzed and decarboxylated under acidic conditions, followed by in situ reduction of the resulting iminium ion, leading to a single product 106 in moderate yield, where the methyl ester group at C16 had epimerized under the reaction conditions to the more stable exo configuration. In order to establish the required C16 stereochemistry of the alkaloid it was necessary to utilize a three-step sequence: air autoxidation of the potassium enolate derived from ester 106 utilizing KHMDS occurred from the least hindered face to give a good yield of α-hydroxyester 107, which could then be reduced to diol 108. Finally, treatment of diol 108 with triethylsilane under Lewis acidic conditions led to racemic alstoscholarisine H (8) in high yield, presumably via reduction of intermediate azafulvene 109 from the least hindered exo face. It might also be noted that diol 108 could be oxidized to the corresponding aldehyde (i.e., 22, Scheme 4), but all attempts to convert this compound to alstoscholarisine J (10) by acetalization failed (Scheme 19).
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
33
Scheme 19 Synthesis of ()-alstoscholarisine H.
Scheme 20 Enantioselective synthesis of bicyclic iodo lactams 113 and 116.
5.4 Liao Synthesis of (–)-Alstoscholarisines A, E, and Their Enantiomers A 2018 report by Liao and coworkers has described an enantioselective approach to (–)-alstoscholarisines A (1) and E (5), as well as their (+)-enantiomers.28 The approach commenced from the known, enantiomerically pure silylalkynyl lactam 110, which upon exposure to TBAF led to terminal alkyne lactam 111 (Scheme 20). This intermediate was then converted to the corresponding boron enolate, which underwent an aldol condensation with acetaldehyde, followed by rhodium-catalyzed cyclization of the aldol product onto the alkyne to afford a 1:2.4 mixture of bicyclic lactam enol ethers 112 (required for alstoscholarisine A) and 114 in good overall yield.
ARTICLE IN PRESS 34
Jeremy D. Mason and Steven M. Weinreb
trans-Fused compound 114 could be isomerized to the cis-fused ring system 115 (required for alstoscholarisine E) upon treatment with potassium hydride. Both enol ethers 112 and 115 could subsequently be converted to vinyl iodides 113 and 116, respectively, with N-iodosuccinimide/silver nitrate. For the synthesis of (–)-alstocholarisine A (1), it was found that iodide 113 can be coupled in a Suzuki-Miyaura reaction with indole boronic acid 117 to afford tetracycle 118 (Scheme 21). Removal of the Boc protecting groups of 118 was effected with methanesulfonic acid, and subsequent hydrogenation of the double bond of the product from the least hindered face using Pearlman’s catalyst then led to lactam 119. A number of methods were attempted to partially reduce lactam 119, and it was eventually discovered that treating the compound with triflic anhydride and tributyltin hydride led to the desired bridged aminal 121, presumably via the intermediate iminium ion 120. Subsequent N-methylation of 121 with formalin and sodium cyanoborohydride then afforded ()-alstoscholarisine A (1). Thus, an eight-step synthesis of the alkaloid was completed starting from lactam 110. Using the same sequence of reactions, it was possible to prepare (–)alstoscholarisine E (5) starting from the epimeric bicyclic iodo lactam 116 in generally similar yields. Moreover, starting from the enantiomer of lactam 110, (+)-alstoscholarisines A and E were also produced.
Scheme 21 Synthesis of ()-alstoscholarisine A (1) from iodo lactam 113.
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
35
6. SUMMARY The alstoscholarisines are a small family of biologically and structurally interesting polycyclic indole alkaloids recently isolated from the leaf extracts of A. scholaris. The alkaloids can be divided into three structural types: (1) ()-alstoscholarisines A–E, (2) (+)-alstoscholarisine G, and (3) (+)-alstoscholarisines H–J. These compounds appear, however, to have a related biogenesis, all deriving from the well-known indole alkaloid (+)-stemmadenine. ()-Alstoscholarisines A–E show significant in vitro activity in NSC proliferation and differentiation. No biological activity has as yet been reported for the remaining four alkaloids. These alkaloids have received limited attention to date from a synthetic standpoint. Two groups have reported total syntheses of racemic alstoscholarisine A, and one of these groups has also described total syntheses of racemic alstoscholarisines B–E. Moreover, a report has appeared describing a total synthesis of ()-alstoscholarisine A, the natural enantiomer of the alkaloid. In addition, a concise synthesis of racemic alstoscholarisine H has appeared. The unique molecular architectures of these metabolites should make them attractive for future synthetic endeavors.
REFERENCES 1. Khyade, M. S.; Kasote, D. M.; Vaikos, N. P. J. Ethnopharmacol. 2014, 153, 1–18. 2. Yang, X.-W.; Yang, C.-P.; Jiang, L.-P.; Qin, X.-J.; Liu, Y.-P.; Shen, Q.-S.; Chen, Y.-B.; Luo, X.-D. Org. Lett. 2014, 16, 5808–5811. 3. For reviews of monoterpenoid indole alkaloids see: (a) Cordell, G. A. Introduction to Alkaloids: A Biogenetic Approach; Wiley-Interscience: New York, 1981; (b) Kutchan, T. M. Phytochemistry 1993, 32, 493–506; (c) Pelletier, S. W. In: Alkaloids: Chemical and Biological Perspectives; Pelletier, S. W. Ed.; Vol. 1; Wiley: New York, 1983; (d) Bisset, N. G. In: Indoles and Biogenetically Related Alkaloids; Phillipson, J. D., Zenk, M. H., Eds.; Academic Press: London, 1980; (e) Herbert, R. B. In: The Chemistry of Heterocyclic Compounds. Indoles, Part 4, the Monoterpenoid Indole Alkaloids; Saxton, J. E. Ed.; Wiley: New York, 1983; (f ) Rahman, A.; Basha, A. Biosynthesis of Indole Alkaloids. Clarendon Press: Oxford, 1983; (g) Ishikura, M.; Yamada, K.; Abe, T. Nat. Prod. Rep. 2010, 27, 1630–1680; h Ishikura, M.; Abe, T.; Choshi, T.; Hibino, S. Nat. Prod. Rep. 2013, 30, 694–752; (i) Dewick, P. M. Medicinal Natural Products: A Biosynthetic Approach, 3rd ed.; Wiley: Chichester, 2009; p. 369; (j) O’Connor, S. E.; Maresh, J. J. Nat. Prod. Rep. 2006, 23, 532–547. 4. Yang, X.-W.; Song, C.-W.; Zhang, Y.; Khan, A.; Jiang, L.-P.; Chen, Y.-B.; Liu, Y.-P.; Luo, X.-D. Tetrahedron Lett. 2015, 56, 6715–6718. 5. Pan, Z.; Qin, X.-J.; Liu, Y.-P.; Wu, T.; Luo, X.-D.; Xia, C. Org. Lett. 2016, 18, 654–657. 6. Hooft, R. W. W.; Straver, L. H.; Spek, A. L. J. Appl. Crystallogr. 2008, 41, 96–103. 7. (a) Scott, A. I. Acc. Chem. Res. 1970, 3, 151–157; (b) Benayad, S.; Ahamada, K.; Lewin, G.; Evanno, L.; Poupon, E. Eur. J. Org. Chem. 2016, 1494–1499.
ARTICLE IN PRESS 36
Jeremy D. Mason and Steven M. Weinreb
8. (a) Kutney, J. P.; Nelson, V. R.; Wigfield, D. C. J. Am. Chem. Soc. 1969, 91, 4279–4280; (b) Scott, A. I.; Yeh, C.-L.; Greenslade, D. J. Chem. Soc., Chem. Commun. 1978, 947–948. 9. (a) Hyman, S. E. Nat. Neurosci. 2016, 19, 1383–1384; (b) Pankevich, D. E.; Altevogt, B. M.; Dunlop, J.; Gage, F. H.; Hyman, S. E. Neuron 2014, 84, 546–553. 10. (a) Efe, A. J.; Ding, S. Philos. Trans. R. Soc., B 2011, 366, 2208–2221; (b) Lyssiotis, C. A.; Lairson, L. L.; Boitano, A. E.; Wurdak, H.; Zhu, S.; Schultz, P. G. Angew. Chem., Int. Ed. 2011, 50, 200–242; (c) Johnson, T. C.; Siegel, D. Chem. Rev. 2017, 117, 12052–12086. 11. Duncan, T.; Valenzuela, M. Stem Cell Res. Ther. 2017, 8, 111–119. 12. Bihelovic, F.; Ferjancic, Z. Angew. Chem., Int. Ed. 2016, 55, 2569–2572. 13. VanRheenen, V.; Kelley, R. C.; Cha, D. Y. Tetrahedron Lett. 1976, 17, 1973–1976. 14. Criegee, R. Chem. Ber. 1931, 64, 260–266. 15. For a recent review of stereoselective Synthesis of tetrahydropyrans see: Zhang, Z.; Tong, R. Synthesis 2017, 4899–4917. 16. See inter alia: (a) Lewis, M. D.; Cha, J. K.; Kishi, Y. J. Am. Chem. Soc. 1982, 104, 4976–4978.; (b) Crawley, G. C.; Briggs, M. T. J. Org. Chem. 1995, 60, 4264–4267; (c) Yamashita, Y.; Hirano, Y.; Takada, A.; Takikawa, H.; Suzuki, K. Angew. Chem., Int. Ed. 2013, 52, 6658–6661. 17. Kirby, A. J. The Anomeric Effect and Related Stereoelectronic Effects at Oxygen. SpringerVerlag: Berlin, 1983. 18. Liang, X.; Jiang, S.-Z.; Wei, K.; Yang, Y.-R. J. Am. Chem. Soc. 2016, 138, 2560–2562. 19. Schafroth, M. A.; Sarlah, D.; Krautwald, S.; Carreira, E. M. J. Am. Chem. Soc. 2012, 134, 20276–20278. 20. Zimmerman, H. E.; Traxler, M. D. J. Am. Chem. Soc. 1957, 79, 1920–1923. 21. (a) Mason, J. D.; Weinreb, S. M. Angew. Chem., Int. Ed. 2017, 56, 16674–16676; (b) Mason, J. D.; Weinreb, S. M. J. Org. Chem. 2018, 5877–5896. 22. For a review of N-sulfonylimines see: Weinreb, S. M. Top. Curr. Chem. 1997, 190, 131–184. 23. (a) Thaler, W. A.; Franzus, W. B. J. Org. Chem. 1964, 29, 2226–2235; (b) Stephenson, L. M.; Mattern, D. L. J. Org. Chem. 1976, 41, 3614–3619; (c) Desimoni, G.; Faita, G.; Righetti, P. P.; Sfulcini, A.; Tsyganov, D. Tetrahedron 1994, 50, 1821–1832. 24. Hanessian, S.; Giroux, S.; Larsson, A. Org. Lett. 2006, 8, 5481–5484. 25. Nyasse, B.; Grehn, L.; Ragnarsson, U. Chem. Commun. 1997, 1017–1018. 26. Lewin, G.; Bernadat, G.; Aubert, G.; Cresteil, T. Tetrahedron 2013, 69, 1622–1627. 27. Schlosser, M.; Jenny, T.; Guggisberg, Y. Synlett 1990, (11), 704. 28. Hu, L.; Yao, L.; Xu, B.; Wang, X.; Liao, X. Org. Lett. 2018, 20, 6202–6205.
The Alstoscholarisine Alkaloids: Isolation, Structure Determination, Biogenesis, Biological Evaluation, and Synthesis Jeremy D. Mason, Steven M. Weinreb1 Department of Chemistry, The Pennsylvania State University, University Park, PA, United States 1 Corresponding author: e-mail address: [email protected]
Contents 1. Introduction 2. Isolation and Characterization 2.1 Isolation and Purification 2.2 Structure Elucidation 3. Putative Biogenesis 3.1 Biosynthesis of ()-Alstoscholarisines A–E 3.2 Biosynthesis of (+)-Alstoscholarisine G 3.3 Biosynthesis of (+)-Alstoscholarisines H–J 4. Biological Activity 5. Total Synthesis 5.1 Syntheses of Alstoscholarisines A and E 5.2 Weinreb/Mason Synthesis of ()-Alstoscholarisines B, C, and D 5.3 Luo/Xia Synthesis of ()-Alstoscholarisine H 5.4 Liao Synthesis of (–)-Alstoscholarisines A, E, and Their Enantiomers 6. Summary References
2 3 3 4 17 17 19 19 20 21 21 29 31 33 35 35
Abstract The alstoscholarisines are a small family of biologically and structurally interesting polycyclic monoterpenoid indole alkaloids isolated from the leaf extracts of Alstonia scholaris. The alkaloids can be divided into three different subtypes based upon their structures and putative biogenesis: (1) ()-alstoscholarisines A–E, (2) (+)-alstoscholarisine G, and (3) (+)-alstoscholarisines H–J. This review discusses the isolation, structure determination, biological activity, and proposed biosynthesis of these metabolites. In addition, synthetic studies on the alkaloids are described including total syntheses of racemic alstoscholarisines A–E, a total synthesis of ()-alstoscholarisine A, and a synthesis of racemic alstoscholarisine H. The Alkaloids ISSN 1099-4831 https://doi.org/10.1016/bs.alkal.2018.09.001
#
2019 Elsevier Inc. All rights reserved.
1
ARTICLE IN PRESS 2
Jeremy D. Mason and Steven M. Weinreb
1. INTRODUCTION Extracts of the tree Alstonia scholaris have a long history of use in traditional medicine in many parts of the world including China, Thailand, Malaysia, India, the Philippines, Africa, and Australia.1 These extracts have been utilized for treatment of an exceptionally diverse range of ailments including malaria, jaundice, microbial infectious diseases, and cancer. This organism and related species have been the subject of a number of phytochemical investigations, and a variety of structurally and biologically interesting secondary metabolites have been isolated from these plants. In an ongoing search for biologically active natural products, Luo et al. found in 2014 that leaf extracts of A. scholaris increased the proliferation of adult mouse hippocampal neural stem cells in vitro (vide infra).2 Upon biologically guided purification of the crude alkaloid mixture, five novel monoterpenoid indole alkaloids,3 ()-alstoscholarisines A–E (1–5), were initially isolated (Fig. 1). Two additional alkaloids, (+)-alstoscholarisine F (6) and (+)-G (7), were also isolated from the same A. scholaris material in 2015 (Fig. 2).4 However, since compound 6 is not closely related in structure or biosynthesis to the other alstoscholarisines, this metabolite will not be considered in this review. Shortly afterward, three more alstoscholarisines
Figure 1 Structures and absolute stereochemistry of ()-alstoscholarisines A–E.
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
3
Figure 2 Structures and stereochemistry of (+)-alstoscholarisines F–J.
((+)-H (8), (+)-I (9), and (+)-J (10)), which are biogenetically related to alstoscholarisines A–E and G, but incorporate a modified skeleton, were isolated from the leaves of A. scholaris (Fig. 2).5
2. ISOLATION AND CHARACTERIZATION 2.1 Isolation and Purification 2.1.1 (2)-Alstoscholarisines A–E Alstoscholarisines A–E could be obtained by a standard purification procedure involving acid extraction of the crude alkaloid mixture from the plant material, followed by extensive chromatography to provide samples of the pure compounds.2 Thus, leaves of A. scholaris, that were collected in Xishuangbanna, Yunnan Province, China (18 kg), were heated at reflux in EtOH (3 40 L), and the solvent was removed under vacuum. The residue was dissolved in 0.37% HCl, and the solution was basified to pH 9–10 using ammonium hydroxide. The basic solution was extracted with EtOAc and evaporated. The residue (180 g) was then dissolved in MeOH, and the resulting solution was subjected to column chromatography on silica gel eluting with CHCl3/MeOH (100% to 20:80) to afford six fractions (fractions A–F). Fraction A (25 g) was further chromatographed using CHCl3/acetone (gradient 9:1 to 7:3) as eluent to give five fractions (fractions A1–A5). Fraction A3 (1.6 g) was further purified by preparative HPLC (MeCN/ H2O, 1:1) to afford ()-alstoscholarisine A (1) (103 mg, white crystals),
ARTICLE IN PRESS 4
Jeremy D. Mason and Steven M. Weinreb
()-alstoscholarisine B (2) (8 mg, white needles), ()-alstoscholarisine C (3) (35 mg, white crystals), and ()-alstoscholarisine E (5) (26 mg, white crystals). ()-Alstoscholarisine D (4) (15 mg, white powder) was obtained in the same way from fraction C (15 g). 2.1.2 (+)-Alstoscholarisine G This alkaloid was obtained from fraction E (6.3 g) described earlier via initial purification by column chromatography followed by preparative HPLC (MeCN/H2O, 1:4) to afford (+)-alstoscholarisine G (7) (6 mg).4 2.1.3 (+)-Alstoscholarisines H–J Dried, powdered leaves (100 kg) of A. scholaris that were collected in the Simao region, Yunnan Province, China, were extracted with EtOH (3 300 L) at room temperature, and the solvent was evaporated in vacuo.5 The residue was dissolved in 0.3% HCl, and the resulting solution was basified with 5% aqueous ammonia to achieve pH 9–10. This solution was then partitioned using EtOAc, to afford a basic aqueous solution, an EtOAc fraction, and emulsion fractions. The emulsion (150 g) was subjected to column chromatography on silica gel eluting with CHCl3/MeOH (gradient) to give five fractions (fractions I–V). Fraction II (15 g) was chromatographed multiple times on silica gel eluting with CHCl3/MeOH (gradient 25:1 to 20:1) to yield (+)-alstoscholarisine H (8) (115 mg, colorless needles) and (+)alstoscholarisine I (9) (85 mg, colorless oil). Similarly, (+)-alstoscholarisine J (10) (5 mg, amorphous solid) could be obtained from fraction III (10 g).
2.2 Structure Elucidation 2.2.1 (2)-Alstoscholarisines A–E Initial UV–visible spectroscopic analysis of the major component of the alkaloid mixture, ()-alstoscholarisine A (1), indicated the presence of an indole moiety by absorptions at 233 and 287 nm in the spectrum, and HR-EIMS revealed a molecular formula of C19H24N2O (m/z 296.1890).2 Proton and carbon NMR data for these five alstoscholarisines are compiled in Tables 1 and 2, respectively. The complete structure of ()alstoscholarisine A (1) was determined by extensive two-dimensional NMR studies, and some key interactions are shown in Fig. 3. A methyl group directly attached to the indole ring was revealed by 1H–13C HMBC correlations between H6 (CH3, 2.21) and C2 (136.9), C7 (105.0), and C8 (130.2). The inclusion of an aminal moiety was assigned by 1H–13C HMBC correlation of a downfield methine (H21, 5.47) with C2 (136.9), C13 (138.5), and N-CH3 (45.6).
—
3.57 (dd, J ¼ 2.6, 10.2 Hz) 3.75 (d, J ¼ 10.5 Hz)
1.96 (m)
2.03 (m)
H-17
1.93 (m)
1.74 (br d, J ¼ 13.6 Hz)
H-14
3.03 (br s)
7.66 (d, J ¼ 7.9 Hz)
7.50 (d, J ¼ 7.9 Hz)
H-12
H-16
7.10 (t, J ¼ 7.9 Hz)
7.04 (t, J ¼ 7.9 Hz)
H-11
2.41 (m)
7.02 (t, J ¼ 7.9 Hz)
6.98 (t, J ¼ 7.9 Hz)
H-10
2.18 (m)
7.48 (d, J ¼ 7.9 Hz)
7.42 (d, J ¼ 7.9 Hz)
H-9
H-15
2.10 (s)
2.21 (s)
3.48 (d, J ¼ 10.5 Hz)
—
2.57 (m)
1.88 (m)
1.88 (m)
7.52 (d, J ¼ 7.9 Hz)
7.05 (t, J ¼ 7.9 Hz)
6.98 (t, J ¼ 7.9 Hz)
7.44 (d, J ¼ 7.9 Hz)
2.06 (s)
2.34 (m)
Alstoscholarisine E (5)a
3.50 (d, J ¼ 10.5 Hz)
—
2.57 (br s)
2.24 (m)
1.85 (m)
7.48 (d, J ¼ 7.9 Hz)
7.07 (t, J ¼ 7.9 Hz)
7.00 (t, J ¼ 7.9 Hz)
7.44 (d, J ¼ 7.9 Hz)
2.21 (s)
Continued
3.37 (dd, J ¼ 2.6, 10.2 Hz)
3.09 (br s)
2.50 (m)
2.03 (m)
1.81 (m)
7.49 (d, J ¼ 8.0 Hz)
7.04 (t, J ¼ 8.0 Hz)
6.99 (t, J ¼ 8.0 Hz)
7.41 (d, J ¼ 8.0 Hz)
2.21 (s)
2.46 (br d, J ¼ 10.5 Hz) 2.33 (dd, J ¼ 6.8, 11.3 Hz)
2.46 (br d, J ¼ 10.5 Hz) 1.85 (td, J ¼ 4.0, 11.3 Hz)
Alstoscholarisine C (3)b Alstoscholarisine D (4)a
2.22 (td, J ¼ 4.5, 12.0 Hz) 2.18 (m)
2.30 (dd, J ¼ 6.4, 12.0 Hz) 2.38 (m)
1.84 (td, J ¼ 4.0, 12.0 Hz)
Alstoscholarisine B (2)b
H-6
H-3
Proton(s) Alstoscholarisine A (1)a
Table 1 Proton NMR data for alstoscholarisines A–E
ARTICLE IN PRESS
3.80 (qd, J ¼ 3.4, 6.4 Hz)
2.33 (m)
5.57 (d, J ¼ 2.3 Hz)
2.35 (s)
3.87 (s)
3.69 (qd, J ¼ 3.0, 6.8 Hz)
2.12 (br s)
5.47 (d, J ¼ 2.6 Hz)
2.24 (s)
—
H-19
H-20
H-21
NCH3
OCH3
b
Recorded in CD3OD (600 MHz). Recorded in (CD3)2CO (600 MHz).
1.16 (d, J ¼ 6.4 Hz)
1.22 (d, J ¼ 6.8 Hz)
H-18
a
3.85 (d, J ¼ 10.5 Hz)
Alstoscholarisine B (2)
3.60 (br d, J ¼ 10.2 Hz)
Proton(s) Alstoscholarisine A (1)
Table 1 Proton NMR data for alstoscholarisines A–E—cont’d
3.82 (s)
2.31 (s) —
2.38 (s)
5.49 (d, J ¼ 2.3 Hz)
2.27 (br s)
2.28 (t, J ¼ 3.0 Hz) 5.38 (d, J ¼ 3.0 Hz)
3.94 (q, J ¼ 6.8 Hz)
1.33 (d, J ¼ 6.8 Hz)
4.10 (d, J ¼ 10.5 Hz)
Alstoscholarisine D (4)
3.94 (q, J ¼ 6.8 Hz)
1.28 (d, J ¼ 6.8 Hz)
4.04 (d, J ¼ 10.5 Hz)
Alstoscholarisine C (3)
—
2.26 (s)
5.37 (d, J ¼ 2.3 Hz)
2.16 (br s)
4.04 (q, J ¼ 6.8 Hz)
1.32 (d, J ¼ 6.8 Hz)
3.92 (dd, J ¼ 1.9, 10.2 Hz)
Alstoscholarisine E (5)
ARTICLE IN PRESS
136.9
47.3
8.1
105.0
130.2
118.7
119.8
121.3
111.3
138.5
31.1
34.5
C-2
C-3
C-6
C-7
C-8
C-9
C-10
C-11
C-12
C-13
C-14
C-15
39.2
29.0
137.3
110.8
121.2
119.4
118.4
129.6
105.3
9.1
46.3
133.8
33.5
28.7
137.5
111.1
121.1
119.3
118.3
129.5
105.1
9.0
46.0
133.9
33.1
28.5
138.1
110.9
121.7
120.1
118.7
130.8
107.1
9.8
47.1
136.7
28.5
31.0
138.7
111.6
121.3
119.8
118.6
130.2
105.0
8.0
47.1
137.0
Continued
Carbon Alstoscholarisine A (1)a Alstoscholarisine B (2)b Alstoscholarisine C (3)b Alstoscholarisine D (4)a Alstoscholarisine E (5)a
Table 2 Carbon NMR data for alstoscholarisines A–E
ARTICLE IN PRESS
173.3
52.1
43.0
67.7
45.6
—
C-20
C-21
NCH3
–CO–
OCH3 —
b
Recorded in CD3OD (150 MHz). Recorded in (CD3)2CO (150 MHz).
45.1
75.3
C-19
a
66.9
18.7
C-18
42.5
74.5
18.2
74.8
74.3
C-17
48.8
35.8
Alstoscholarisine B (2)
C-16
Carbon Alstoscholarisine A (1)
Table 2 Carbon NMR data for alstoscholarisines A–E—cont’d
52.2
173.4
45.0
71.8
42.9
72.1
17.9
68.9
49.1
Alstoscholarisine C (3)
—
179.3
44.5
71.9
42.9
72.9
18.2
70.6
50.8
Alstoscholarisine D (4)
—
—
45.5
72.4
43.4
73.3
18.3
68.8
36.0
Alstoscholarisine E (5)
ARTICLE IN PRESS
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
9
Figure 3 Key two-dimensional NMR correlations for ()-alstoscholarisine A (1): (left) 1 H–1H COSY ( ); (center) 1H–13C HMBC ( ); (right) 1H–1H ROESY ( ).
The presence of an N-methylpiperidine ring (D-ring) was determined based on a 1H–1H COSY correlation between H21 (5.47) and H20 (2.12), H20 with H15 (2.18), H15 with H14 (1.74, 2.03), and H14 with H3 (1.84, 2.30), as well as an HMBC cross-peak between the N-Me (2.24) and C3 (47.3). An HMBC correlation of the H16 methine (3.03) with C2 (136.9) and C7 (105.0) indicated that C16 was attached at C2 of the indole ring. Incorporation of a six-membered ring (C-ring) was then elucidated by observation of a COSY relationship between H15 (2.18) and H16 (3.03), as well as the previously discussed correlations between H21, H20, and H15. The presence of an α-methyltetrahydropyran ring was established by COSY cross-peaks between H16 (3.03) and H17 (3.57, 3.60), H20 (2.12) and H19 (3.69), and H19 with the upfield methyl group (H18, 1.22), as well as the downfield chemical shifts of C17 and C19, which indicated the presence of an oxygen in the ring. Furthermore, the H18 methyl group was assigned as being in an equatorial position on the tetrahydropyran ring due to an NOE cross-peak between this signal and H21 in the ROESY spectrum. The structure of ()-alstoscholarisine A (1) was also unambiguously confirmed by single-crystal X-ray analysis. ()-Alstoscholarisine B (2) showed an HR-EIMS peak 58amu greater (m/ z 354.1940) than that for alkaloid 1, which indicated a molecular formula of C21H26N2O3. The IR spectrum (1730 cm1) and the 13C NMR spectrum (173.3) of metabolite 2 indicated the presence of an ester moiety. The COSY and HMBC spectra of 2 showed the same correlations as those discussed earlier for alkaloid 1, except for the absence of the C16 methine hydrogen. Furthermore, an NOE correlation between the H18 methyl group and the H21 aminal proton was retained, which indicated an equatorial methyl group, as in compound 1. It was therefore determined that alkaloid 2 is a structural analogue of ()-alstoscholarisine A (1) containing a methyl ester group at C16. ()-Alstoscholarisine C (3) had the same HR-EIMS peak (m/z 354.1946) as compound 2, indicating the possibility that the two alkaloids are stereoisomers. The HMBC and COSY spectra for metabolite 3 were analogous to
ARTICLE IN PRESS 10
Figure 4 Key 1H–1H ROESY (
Jeremy D. Mason and Steven M. Weinreb
) NMR correlations for ()-alstoscholarisine C (3).
those for 2. However, the ROESY spectrum of ()-alstoscholarisine C (3) revealed an NOE correlation of the H18 methyl group with H15 and H17, indicating an axial methyl group on the tetrahydropyran ring (Fig. 4). Furthermore, the H19 methine showed an ROESY cross-peak with the H21 aminal. In addition, the structure of ()-alstoscholarisine C (3) was unambiguously determined by single-crystal X-ray analysis. ()-Alstoscholarisine D (4) showed an HR-EIMS peak at m/z 340.1780, 14 amu less than that for alkaloids 2 and 3, indicating the possibility that this compound was the methyl ester hydrolysis product of 2 or 3. Indeed, the IR spectrum of metabolite 4 supported the presence of a carboxylic acid moiety (3418 cm1). ()-Alstoscholarisine D (4) showed the same HMBC and COSY correlations as seen for alkaloids 2 and 3. Finally, the methyl group at C19 was determined to be axial due to observed NOE interactions of H18 with H15 and H17, as well as those between H19 and H21. ()-Alstoscholarisine E (5) had the same molecular mass as alkaloid 1 (m/z 296.1881), as well as analogous HMBC and COSY spectra. However, compound 5 was assigned as having an axial methyl group at C19, based on the same type of NOE interactions as discussed for metabolites 3 and 4. The absolute configuration of ()-alstoscholarisine A (1) was determined to be (15R,16R,19R,20S,21S) by single-crystal X-ray analysis utilizing the method developed by Hooft and coworkers.6 The absolute configurations of alkaloids 2–5 were assigned based on the similarity of their UV-circular dichroism spectra with that of ()-alstoscholarisine A (1). 2.2.2 (+)-Alstoscholarisine G HR-TOFMS of alstoscholarisine G (7) indicated a formula of C20H24N3O2 ([M + H]+ 340.2022).4 UV–visible spectroscopic analysis of alkaloid 7 revealed the presence of an indole ring system based on absorptions at 232 and 286 nm. The IR spectrum revealed the presence of an amide or lactam moiety (1683, 3437 cm1). Proton and carbon NMR data for alstoscholarisine G (7) are summarized in Table 3. A δ-lactam ring fused to the indole moiety was elucidated based
ARTICLE IN PRESS 11
The Alstoscholarisine Alkaloids
Table 3 Proton and carbon NMR data for alstoscholarisine G
Proton(s)
H-3
H-6
δHa
Carbon
δCb
2.12 (m)
C-2
135.3
2.53 (m)
C-3
52.3
4.80 (d, J ¼ 15.0 Hz)
C-6
40.2
4.90 (d, J ¼ 15.0 Hz)
C-7
107.2
C-8
125.8
H-9
7.63 (d, J ¼ 7.5 Hz)
C-9
118.1
H-10
7.23 (t, J ¼ 7.5 Hz)
C-10
119.5
H-11
7.26 (t, J ¼ 7.5 Hz)
C-11
121.9
H-12
7.57 (d, J ¼ 7.5 Hz)
C-12
112.0
H-14
2.04 (m)
C-13
138.2
2.40 (br d, J ¼ 13.9 Hz)
C-14
28.4
H-15
3.80 (br d, J ¼ 6.4 Hz)
C-15
39.0
H-17
4.65 (d, J ¼ 10.5 Hz)
C-16
54.4
4.98 (d, J ¼ 10.5 Hz)
C-17
66.1
H-18
1.46 (d, J ¼ 6.8 Hz)
C-18
13.8
H-19
5.50 (q, J ¼ 6.8 Hz)
C-19
126.2
H-21
2.91 (d, J ¼ 12.0 Hz)
C-20
135.0
3.10 (d, J ¼ 12.0 Hz)
C-21
62.5
1.96 (s)
C-22
174.7
NCH3 N1 H CONH a
12.10 (s) 9.28 (br s)
Recorded in pyridine-d5 (600 MHz). Recorded in pyridine-d5 (150 MHz).
b
NCH3
45.6
ARTICLE IN PRESS 12
Jeremy D. Mason and Steven M. Weinreb
Figure 5 Key two-dimensional NMR correlations for (+)-alstoscholarisine G (7): (left) 1 H–13C HMBC ( ); (right) 1H–1H ROESY ( ).
on the presence of HMBC correlations between a lactam NH proton (9.28) and C6 (40.2), C7 (107.2), C16 (54.4), and C22 (174.7), as well as those between H6 (4.80–4.90) with C2 (135.3), C7 (107.2), C8 (125.8), and C22 (174.7) (Fig. 5). An N-methyl-4-piperidinyl unit connected to this lactam at C16 was assigned based on HMBC correlations of a deshielded N-methyl group (1.96) with C3 (52.3), C14 (28.4), C20 (135.0), and C21 (62.5), as well as those between H15 (3.80) and C3 (52.3), C14 (28.4), C16 (54.4), C19 (126.2), C20 (135.0), and C21 (62.5). An ethylidene moiety at C20 of the N-methylpiperidine ring was assigned based on HMBC cross-peaks between H19 (5.50) and C15 (39.0) and C21 (62.5), as well as those between an allylic methyl group (H18, 1.36) and C19 (126.2) and C20 (135.0). Finally, a hydroxymethyl group attached at C16 was assigned by HMBC correlations between the downfield protons at C17 (4.65, 4.98) with C2 (135.3), C15 (39.0), C16 (54.4), and C22 (174.7). The double bond geometry of the ethylidene group was assigned as (E) due to NOE correlations between H18 (1.46) and H15 (3.80) as well as those between H19 (5.50) and H21 (2.91, 3.10) in the ROESY spectrum. The relative and absolute stereochemistry of (+)-alstoscholarisine G (7) is assumed to be (15S,16S) due to the putative biosynthetic pathway originating from (+)-(E)-vallesamine (vide infra), although this assumption has not been verified. 2.2.3 (+)-Alstoscholarisines H–J (+)-Alstoscholarisine H (8) showed an HR-EIMS peak at m/z 296.1886, indicating a formula of C19H24N2O.5 Complete 1H and 13C data for alstoscholarisines H–J are summarized in Tables 4 and 5, respectively. As
ARTICLE IN PRESS 13
The Alstoscholarisine Alkaloids
Table 4 Proton NMR data for alstoscholarisines H–J
(+)-Alstoscholarisine Proton(s) H (8)a
(+)-Alstoscholarisine I (9)b
(+)-Alstoscholarisine J (10)c
H-3
5.34 (t, J ¼ 2.8 Hz)
5.45 (t, J ¼ 2.6 Hz)
5.39 (t, J ¼ 3.0 Hz)
H-6
2.38 (s)
2.38 (s)
2.49 (s)
H-9
7.54 (d, J ¼ 7.5 Hz)
7.46 (d, J ¼ 7.5 Hz)
7.52 (d, J ¼ 7.5 Hz)
H-10
7.14 (t, J ¼ 7.5 Hz)
7.02 (t, J ¼ 7.5 Hz)
7.05 (t, J ¼ 7.5 Hz)
H-11
7.19 (t, J ¼ 7.5 Hz)
7.12 (t, J ¼ 7.5 Hz)
7.14 (t, J ¼ 7.5 Hz)
H-12
7.42 (d, J ¼ 7.5 Hz)
7.47 (d, J ¼ 7.5 Hz)
7.50 (d, J ¼ 7.5 Hz)
H-14
2.02 (dt, J ¼ 2.8, 12.8 Hz)
2.16 (m)
2.05 (dt, J ¼ 3.0, 12.9 Hz)
2.31 (dt, J ¼ 2.8, 12.8 Hz)
2.64 (m)
2.44 (dt, J ¼ 3.0, 12.9 Hz)
H-15
3.52 (m)
3.65 (m)
3.26 (t, J ¼ 3.0 Hz)
H-16
3.60 (m)
—
—
H-17
3.70 (t, J ¼ 10.5 Hz)
3.68 (d, J ¼ 10.9 Hz)
4.99 (s)
4.30 (dd, J ¼ 3.5, 10.5 Hz)
4.65 (br d, J ¼ 10.9 Hz)
—
H-18
1.84 (d, J ¼ 6.5 Hz)
1.79 (d, J ¼ 6.8 Hz)
1.75 (d, J ¼ 6.8 Hz)
H-19
5.67 (q, J ¼ 6.5 Hz)
5.60 (q, J ¼ 6.8 Hz)
5.40 (q, J ¼ 6.8 Hz)
H-21
2.90 (d, J ¼ 12.5 Hz)
2.68 (d, J ¼ 12.8 Hz)
2.78 (d, J ¼ 12.5 Hz)
2.99 (d, J ¼ 12.5 Hz)
2.87 (d, J ¼ 12.8 Hz)
3.06 (d, J ¼ 12.5 Hz)
NCH3
2.45 (s)
2.21 (s)
2.32 (s)
OCH3
—
3.66 (s)
3.38 (s)
OCH3
—
—
3.59 (s)
a
Recorded in CDCl3 (400 MHz). Recorded in CD3OD (400 MHz). Recorded in CD3OD (600 MHz).
b c
ARTICLE IN PRESS 14
Jeremy D. Mason and Steven M. Weinreb
Table 5 Carbon NMR data for alstoscholarisines H–J
(+)-Alstoscholarisine Carbon H (8)a
(+)-Alstoscholarisine I (9)b
(+)-Alstoscholarisine J (10)c
C-2
132.0
130.0
135.2
C-3
67.2
57.6
68.3
C-6
10.1
10.7
10.6
C-7
106.6
110.3
110.0
C-8
128.9
130.1
130.4
C-9
117.7
119.1
119.4
C-10
119.2
120.3
120.4
C-11
121.0
122.7
123.0
C-12
109.6
111.4
111.5
C-13
136.9
138.2
138.4
C-14
33.9
32.7
31.7
C-15
28.0
33.4
38.5
C-16
40.2
54.0
76.9
C-17
63.0
65.1
111.5
C-18
12.9
13.5
13.4
C-19
122.1
124.8
122.2
C-20
135.0
134.9
135.8
C-21
56.5
57.6
58.0
NMe
43.9
44.0
44.0
–CO–
—
175.1
OCH3
—
52.8
58.1
OCH3
—
—
59.1
a
Recorded in CDCl3 (100 MHz). Recorded in CD3OD (100 MHz). c Recorded in CD3OD (150 MHz). b
—
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
15
Figure 6 Key two-dimensional NMR correlations for ()-alstoscholarisine H (8): (left) 1 H–1H COSY ( ); (center) 1H–13C HMBC ( ); (right) 1H–1H ROESY ( ).
in alstoscholarisines A–E, a methyl group attached to an indole system was assigned based on HMBC correlations between H6 (10.1) and C2 (132.0), C7 (106.6), and C8 (128.9). An indole aminal moiety was revealed by HMBC cross-peaks between a downfield methine H3 (5.34) and C2 (132.0) and C13 (136.9) (Fig. 6). A six-membered ring (C-ring) was elucidated based on 1H–1H COSY correlation between H3 (5.34) and H14 (2.02, 2.31), H14 with H15 (3.52), and H15 with H16 (3.60), as well as HMBC cross-peaks between H16 (3.60) with C2 (132.0) and C15 (28.0). An N-methyl piperidine moiety (D-ring) was assigned by HMBC correlations between the N-Me (2.45) and C3 (67.2) and C21 (56.5), H15 (3.52) with C14 (33.9) and C20 (135.0), as well as the previously discussed COSY interactions between H15/H14/H3. An ethylidene group on this piperidine ring was revealed by HMBC correlations of H19 (5.67) with C15 (28.0) and C21 (56.5), as well as those between H18 (1.84) and C20 (135.0). A hydroxymethyl unit attached at C16 was assigned based on HMBC cross-peaks of deshielded methylene protons H17 (3.70, 4.30) with C2 (132.0), C15 (28.0), and C16 (40.2), as well as COSY correlation between H17 and H16 (3.60). The relative stereochemistry at C16 was determined by NOE correlation between H16 (3.60) and H14 (2.02, 2.31) in the ROESY spectrum. The stereochemistry of the aforementioned ethylidene group was assigned as (E) based upon NOE relationships between H18 (1.84) and H15 (3.52), as well as those between H19 (5.67) and H21 (2.90, 2.99). Finally, the complete structure of metabolite 8 was unambiguously established by X-ray crystallographic analysis. (+)-Alstoscholarisine I (9) showed an HR-EIMS peak at m/z 354.1940, 58 amu greater than that of (+)-alstoscholarisine H (8), which indicated the possibility that compound 9 was an analogue of 8 containing a methyl ester moiety. Indeed, the IR spectrum (1728 cm1) and 13C NMR spectrum (175.1) of this metabolite showed the presence of an ester carbonyl group.
ARTICLE IN PRESS 16
Jeremy D. Mason and Steven M. Weinreb
Furthermore, the one-dimensional proton NMR spectrum was very similar to that of (+)-alstoscholarisine H (8), except for the absence of the C16 methine proton and the addition of a methoxyl peak (3.66). The location of the methyl ester group attached at C16 was further supported by HMBC correlations between the hydroxymethyl methylene protons (H17, 3.68, 4.65) and C16 (54.0) and the ester carbonyl carbon (175.1) (Fig. 7). The stereochemistry at C16 was assigned based upon NOE interactions between the methoxyl peak (3.66) and H14 (2.64), and the ethylidene group geometry was assigned as (E) based on analogous correlations to those described earlier for alstoscholarisine H (8). (+)-Alstoscholarisine J (10) showed an HR-EIMS peak 60 amu greater than alstoscholarisine H (8) (m/z 356.2094, C21H28N2O3). The onedimensional proton NMR spectrum of the alkaloid was similar to that of 8, except for the absence of the C16 methine and C17 methylene protons and the addition of a downfield methine at 4.99 ppm and two methoxy peaks at 3.38 and 3.59 ppm. In the 13C spectrum, the signals corresponding to C16 and C17 in alstoscholarisine H were absent, and instead four new signals were observed: a quaternary carbon at 76.9 ppm, a tertiary carbon at 111.5 ppm, and two primary carbons at 58.1 and 59.1 ppm. These data led to the proposal of a dimethyl acetal at C17 and a hydroxyl group at C16. The stereochemistry at C16 was determined by the presence of an NOE cross-peak between the allylic methyl group (H18, 1.75) and one of the methoxyl group signals (3.59) (Fig. 8). The ethylidene stereochemistry was again confirmed as (E) by analogous ROESY cross-peaks as described earlier for alstoscholarisines H and I. Based upon the biogenetic hypothesis for these alkaloids as originating from (+)-(E)-vallesamine (vide infra), the absolute configurations of compounds (+)-8, (+)-9, and (+)-10 are assumed to be (3R,15R,16R), (3R,15S,16S), and (3R,15S,16S), respectively, although this supposition has not yet been firmly established.
Figure 7 Key two-dimensional NMR correlations for (+)-alstoscholarisine I (9): (left) 1 H–13C HMBC ( ); (right) 1H–1H ROESY ( ).
ARTICLE IN PRESS 17
The Alstoscholarisine Alkaloids
Figure 8 Key 1H–1H ROESY (
) correlations for alstoscholarisine J (10).
3. PUTATIVE BIOGENESIS 3.1 Biosynthesis of (2)-Alstoscholarisines A–E Although there are no experimental studies to date, ()-alstoscholarisines C, D, and E most likely arise biogenetically from the well-known monoterpenoid indole alkaloid (+)-stemmadenine (13) which is formed via an established pathway from two common monoterpenoid indole alkaloid precursors, tryptamine (11) and secologanin (12) (Scheme 1).2,7 (+)-Stemmadinine is first transformed to (+)-(E)-vallesamine (14), a known alkaloid that has been isolated previously from A. scholaris. It might be noted that the conversion of stemmadenine to vallesamine, which occurs by an established sequence of steps,8 serves to excise one of the two side-chain carbons of the original tryptamine building block. (+)-(E)-Vallesamine (14) could then be N-methylated to generate quaternary ammonium salt 15, which could undergo ring cleavage to form azafulvene 16. This pivotal intermediate (vide infra) could undergo reduction of the azafulvene moiety and oxidation of the piperidine ring to generate (E)-α,β-unsaturated iminium ion (E)-17, which would undergo a stereospecific conjugate addition of the attendant alcohol moiety to form the tetrahydropyran ring with the axial α-methyl group in tetracyclic iminium ion 18. A final nucleophilic addition of the indole nitrogen onto iminium ion 18 would provide ()-alstoscholarisine C (3). Ester hydrolysis would generate the corresponding carboxylic acid, ()-alstoscholarisine D (4), which could undergo decarboxylation to form ()-alstoscholarisine E (5). The biosynthesis of ()-alstoscholarisines A (1) and B (2) is proposed by Luo and coworkers2 to involve (+)-(Z)-vallesamine (19), an alkaloid which is also found in A. scholaris (Scheme 2). Thus, by the same pathway described earlier originating from (+)-(E)-vallesamine (14), (+)-(Z)-vallesamine (19) could be transformed into (Z)-α,β-unsaturated iminium ion (Z)-17.
ARTICLE IN PRESS 18
Jeremy D. Mason and Steven M. Weinreb
Scheme 1 Proposed biosynthesis of ()-alstoscholarisines C (3), D (4), and E (5) via (+)stemmadenine (13) and (+)-(E)-vallesamine (14).
Scheme 2 Putative biosynthesis of ()-alstoscholarisines A (1) and B (2) from (+)-(Z)vallesamine (19).
This intermediate could undergo an analogous stereospecific cyclization sequence as described earlier for (E)-17 to ultimately generate ()alstoscholarisine B (2) having an equatorial α-methyl group in the tetrahydropyran ring. Hydrolysis of the methyl ester and decarboxylation of the intermediate carboxylic acid would produce ()-alstoscholarisine A (1).
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
19
Scheme 3 Biosynthetic hypothesis for (+)-alstoscholarisine G (7).
3.2 Biosynthesis of (+)-Alstoscholarisine G Alstoscholarisines G–J are thought to all arise from the azafulvene 16 derived from (+)-(E)-vallesamine (14) (Scheme 3).4,8 Thus, (+)-alstoscholarisine G (7) is proposed to be formed by conjugate addition of ammonia (or an ammonia equivalent) to this intermediate to produce an aminomethyl indole like 20, which cyclizes onto the attendant methyl ester to form the δ-lactam ring of the alkaloid.
3.3 Biosynthesis of (+)-Alstoscholarisines H–J In an alternative biogenetic pathway leading to (+)-alstoscholarisines H–J, the azafulvene moiety of intermediate 16 could be reduced, and the piperidine ring could then be oxidized to form iminium ion 21 (Scheme 4), a regioisomer of iminium ion 17 (cf. Scheme 2).5 This intermediate could undergo 1,2-addition of the indole nitrogen onto the iminium ion to generate (+)alstoscholarisine I (9). Alkaloid 9 could then undergo methyl ester hydrolysis and subsequent decarboxylation to form (+)-alstoscholarisine H (8). Finally, metabolite 8 could be oxidized to form the α-hydroxyaldehyde 22, which would produce (+)-alstoscholarisine J (10) by formation of the corresponding dimethyl acetal.
ARTICLE IN PRESS 20
Jeremy D. Mason and Steven M. Weinreb
Scheme 4 Proposed biosynthesis of (+)-alstoscholarisines H (8), I (9), and J (10).
4. BIOLOGICAL ACTIVITY The primary biological activity associated with alstoscholarisines A–E (1–5) is their ability to promote neuronal stem cell (NSC) proliferation and differentiation.2 Neural stem cells have shown promise for the discovery of new CNS therapies potentially of use for treatment of neurodegenerative ailments.9 In recent years, there has been a search for small molecules and natural products with the ability to modulate stem cell fate and population, and such compounds may play a key role in the development of treatments for various neurogenic disorders such as Alzheimer’s disease, Parkinson’s disease, and Huntington’s disease.10 For example, several phase I/II clinical trials using stem cell strategies are currently underway for Alzheimer’s disease.11 ()-Alstoscholarisine A (1) is the most active of the five alkaloids, promoting NSC proliferation at a dosage-dependent concentration of 0.1 μg/mL. Compound 1 also enhanced NSC sphere formation and partial neurogenic fate commitment. It was reported that ()-alstoscholarisine G (7) has no significant activity in enhancing hippocampal NSC proliferation in vitro.4 Moreover, ()-alstoscholarisines H (8), I (9), and J (10) were also found to be inactive toward both NSC proliferation and PC12 cell differentiation (neurite outgrowth-promoting activity).5 In addition, these latter three alkaloids did not demonstrate any cytotoxic activity.
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
21
5. TOTAL SYNTHESIS 5.1 Syntheses of Alstoscholarisines A and E 5.1.1 Bihelovic/Ferjancic Approach to ()-Alstoscholarisine A In 2016, Bihelovic and Ferjancic completed a total synthesis of ()alstoscholarisine A (1) involving 12 steps starting from readily available compounds.12 The synthesis began with an aldol reaction between the enolate of indole ester 23 and protected β-aminoaldehyde 24 to provide adduct 25, which was dehydrated with concomitant Boc removal to form condensation product 26 as a single (E)-isomer (Scheme 5). Removal of the allyl carbamate protecting group of 26 with Pd(OAc)2/morpholine then gave secondary amine 27. Condensation of amine 27 and γ-phenylselenylaldehyde 28 formed enamine 29, which subsequently cyclized via an intramolecular Michael addition to the α,β-unsaturated ester moiety to provide iminium ion 30. This intermediate reacted further with the indole nitrogen to form bridged aminal 31/32 as a separable mixture of C16 ester epimers (dr 1:1).
Scheme 5 Synthesis of key bridged tetracyclic aminal 33.
ARTICLE IN PRESS 22
Jeremy D. Mason and Steven M. Weinreb
Equilibration of this mixture of esters with DBU in ethanol improved the diastereomeric ratio to 5:1, with the major epimer being the desired exo-ester 32 needed for the synthesis. Oxidation of selenide 32 to the corresponding selenoxide with m-CPBA and subsequent elimination provided the requisite terminal alkene 33. Reduction of the ester moiety of 33 with DIBALH gave alcohol 34 (Scheme 6). An Upjohn dihydroxylation13 of 34, followed by oxidative diol cleavage with Pb(OAc)4,14 yielded axial aldehyde 35, which was epimerized under basic conditions with DBU to form equatorial aldehyde 36. This intermediate spontaneously cyclized to lactol 37, which was immediately oxidized with Dess–Martin periodinane to produce δ-lactone 38. A key sequence in the total synthesis was construction of the tetrahydropyran moiety with the requisite configuration of the α-methyl substituent using a strategy that has been widely applied in this field.15,16 Thus, addition of one equivalent of methyllithium to this lactone 38 could first be effected to provide hemiketal 39 (single diastereomer, configuration not determined). This intermediate was further transformed to ()-alstoscholarisine A (1) by
Scheme 6 Completion of the Bihelovic/Ferjancic total synthesis of ()-alstoscholarisine A (1).
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
23
initial conversion to an intermediate oxocarbenium ion 40 through ionization with trimethylsilyl trifluoromethanesulfonate (TMSOTf ), followed by a reduction from the least hindered convex face using triethylsilane as the hydride source. It should be noted that this stereochemical outcome is also consistent with a model predicting axial attack of hydride from triethylsilane based on the intermediacy of a favored chair-like transition state.17 5.1.2 Enantioselective Synthesis of (2)-Alstoscholarisine A by Yang et al. In 2016, Yang and coworkers reported an enantioselective total synthesis of ()-alstoscholarisine A (1) which required 12 steps.18 The synthesis commenced with an acylation of 3-methylindole (42) with vinyl-γ-lactone 41 to provide N-acylindole 43 (Scheme 7). This compound underwent a key enantioselective intramolecular Friedel–Crafts alkylation mediated by
Scheme 7 Early stages of the Yang total synthesis of ()-alstoscholarisine A (1).
ARTICLE IN PRESS 24
Jeremy D. Mason and Steven M. Weinreb
Scheme 8 Yang total synthesis of ()-alstoscholarisine A (1).
Carreira’s iridium catalyst system using the chiral ligand 44 and Sc(OTf )3 to afford tricycle 45 with high enantioselectivity (>99:1 er) and in good yield.19 Dihydroxylation of olefin 45 provided diol 46, which was not isolated but was directly protected as the acetonide 47 (5:1 mixture of diastereomers). The lactam moiety of 47 was next deprotonated with lithium hexamethyldisilazide (LiHMDS), and the resulting enolate was trapped with PhSeBr to form selenide 48. This selenide was then oxidized to the corresponding selenoxide, which underwent elimination to afford α,β-unsaturated lactam 49. Conjugate addition of vinyl cuprate reagent 50 to α,β-unsaturated lactam 49 from the re face and subsequent trapping of the intermediate enolate with acetaldehyde provided alcohol 52 with good (8:1) diastereoselectivity (Scheme 8). The diastereoselectivity of this aldol addition is in accord with a Zimmerman–Traxler chair-like transition state 51, which predicts the antiadduct 52 being the major diastereomeric product.20 Ketal removal with methanolic HCl then resulted in triol 53. Oxidative cleavage of this intermediate with NaIO4 formed aldehyde 54, which underwent spontaneous cyclization to give lactol 55.
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
25
Scheme 9 Completion of the total synthesis of ()-alstoscholarisine A (1).
Lactol 55 was then reduced to tetrahydropyran 56 with Et3SiH/ BF3Et2O (Scheme 9). Hydroboration/oxidation of terminal olefin 56 provided alcohol 57, which was oxidized with Dess–Martin periodinane to produce aldehyde 58. The synthesis was completed by condensation of aldehyde 58 with methylamine to generate intermediate imine 59, followed by reduction of both the imine and indole lactam moieties with LiAlH4 to provide hemiaminal 60, which cyclized directly to form ()-alstoscholarisine A (1). 5.1.3 Mason/Weinreb Synthesis of ()-Alstoscholarisines A and E Mason and Weinreb have reported total syntheses of racemic alstoscholarisines A–E.21 A key intermediate in the synthesis of all five of the alkaloids was bridged aminal 68, prepared in about six operations from readily available materials as discussed below: Thus, indole ester 61 was first deprotonated with LiHMDS at low temperature and α,β-unsaturated-Nsulfonyllactam 62 was added to the resulting enolate, leading to the Michael adduct 63 in good yield as a complex mixture of stereoisomers (Scheme 10). Lactam ester 63 subsequently could be C-allylated using allyl iodide/ K2CO3 in acetonitrile to form 64, again as a complex mixture, followed by in situ addition of Pd(PPh3)4/morpholine to promote decarboxylation of the allyl ester, affording the desired C-allylated lactam 65. This product was formed exclusively as the more stable C15, C20 trans isomer shown, but was a mixture of stereoisomers at the C16 ester-bearing center. N-Sulfonyllactam 65 was then partially reduced to the N,O-hemiacetal 66 using DIBALH. Exposure of this material to trifluoroacetic acid at room temperature resulted in the removal of the Boc group on the indole nitrogen and subsequent closure via an N-sulfonyliminium ion22 to afford
ARTICLE IN PRESS 26
Jeremy D. Mason and Steven M. Weinreb
Scheme 10 Construction of bridged aminal ester 68.
the desired bridged aminal 67 in good yield as a mixture of C16 ester stereoisomers. However, this mixture could be readily epimerized with sodium methoxide/methanol to yield exclusively the more stable ester exoisomer 68. In order to continue the synthesis, the terminal alkene double bond of intermediate 68 needed to be isomerized to an internal position. Two strategies were developed for this transformation, leading to intermediates which were subsequently processed to form alstoscholarisines A and E.21 In the first approach, alkene ester 68 was first reduced to afford alcohol 69 which was then protected as silyl ether 70 (Scheme 11). Heating compound 70 with DEAD in refluxing toluene effected an ene reaction, producing the (E)-allylic hydrazine derivative 71 (along with a by-product derived from the reaction of DEAD with ene product 71 at the C3 position of the indole).23 After removal of the silyl protecting group of compound 71 with fluoride, the resulting internal hydroxyalkene 72 could be cleaved by a two-step procedure to yield axial hydroxyaldehyde 73. Exposure of this
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
27
Scheme 11 Initial route to lactol acetate 77 and δ-lactone 78.
Scheme 12 Alternative route to aldehyde intermediate 73.
intermediate to DBU led to epimerization, followed by spontaneous cyclization of the resulting equatorial hydroxyaldehyde to afford the desired lactol 76, along with some of the ring-opened product 74. Without purification, this mixture could be acylated with acetic anhydride to form lactol acetate 77 as well as acetate 75. Alternatively, the crude lactol mixture could be oxidized to produce δ-lactone 78. A second strategy for double bond isomerization involved heating terminal alkene 68 with the Grubbs II ruthenium metathesis catalyst in methanol,21b,24 leading to propenyl derivative 79 in excellent yield (2.8:1 E/Z mixture) (Scheme 12). The ester group of 79 could subsequently be reduced with lithium aluminum hydride to generate alcohol
ARTICLE IN PRESS 28
Jeremy D. Mason and Steven M. Weinreb
80. The same two-step alkene cleavage protocol used earlier then led to the formation of axial hydroxyaldehyde 73, which was processed into lactol acetate 77 and δ-lactone 78 in somewhat better yields compared to those obtained from the ene-based procedure shown in Scheme 11. In addition, this latter sequence was more efficient overall in that alcohol protection and deprotection was not required, and formation of the ene reaction by-product noted above was avoided. Conversion of δ-lactone 78 to alstoscholarisine A made use of the same strategy previously applied by Bihelovic and Ferjancic for elaborating the tetrahydropyran ring of the metabolite.12 Thus, addition of methyllithium to lactone 78 afforded hemiketal 81, which upon treatment with trimethylsilyl triflate/triethylsilane in methylene chloride at low temperature presumably generated intermediate oxocarbenium ion 82, which upon silane reduction from the least hindered face gave the α-methyltetrahydropyran 83 as a single stereoisomer in good yield (Scheme 13).15,16 Finally, reductive removal of the tosyl group of 8325 produced amine 84 and subsequent N-methylation26 with formalin and sodium cyanoborohydride yielded ()-alstoscholarisine A (1). A synthesis of alstoscholarisine E (5) could be executed by simply reversing the order of steps for generating the α-stereogenic center of the tetrahydropyran. Therefore, exposing lactol acetate 77 to trimethylsilyl trifluoromethanesulfonate and trimethylaluminum in methylene chloride at 78°C resulted in transfer of a methyl group from the aluminum reagent to the least hindered face of putative oxocarbenium ion 85 to afford tetrahydropyran 86 in good yield as a single stereoisomer (Scheme 14).15,16 The tosyl group could then be removed from intermediate 86 using Mg/MeOH,25 and the resulting amine 87 was N-methylated26 to afford ()-alstoscholarisine E (5).
Scheme 13 Conversion of δ-lactone 78 to ()-alstoscholarisine A (1).
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
29
Scheme 14 Conversion of lactol acetate 77 to ()-alstoscholarisine E (5).
Scheme 15 Construction of lactol acetate 94 and δ-lactone 95.
5.2 Weinreb/Mason Synthesis of ()-Alstoscholarisines B, C, and D Intermediate bridged aminal ester 67 has also been used by Mason and Weinreb in total syntheses of racemic alstoscholarisines B, C, and D.21 This ester could be deprotonated with potassium hexamethyldisilazide (KHMDS) and the resulting enolate was alkylated from the least hindered face using anhydrous formaldehyde27 to give hydroxymethyl compound 88 in high yield as a single stereoisomer having the configuration shown in Scheme 15. The alcohol group of 88 was first protected as the silyl ether 89, and a subsequent thermal ene reaction with DEAD occurred smoothly
ARTICLE IN PRESS 30
Jeremy D. Mason and Steven M. Weinreb
to afford (E)-allylic hydrazine derivative 90 in good yield. Oxidative cleavage of the alkene double bond of intermediate 90 led to axial aldehyde 91, which upon exposure to HCl underwent removal of the TBS protecting group to produce hydroxyaldehyde 92. The tetrahydropyran moiety of the alkaloids was then constructed from this compound using a strategy similar to that previously applied for alstoscholarisines A and E (vide supra). Thus, epimerization of aldehyde 92 with DBU resulted in the formation of lactol 93, which could be acylated to generate lactol acetate 94. In addition, oxidation of lactol 93 produced δ-lactone 95. An alternative, more efficient, and higher yielding route to the key intermediates 94 and 95 was also developed utilizing the approach previously applied to alstoscholarisines A and E (cf. Scheme 12).21b Therefore, an isomerization of the allyl group of 88 could be effected using the Grubbs II metathesis catalyst in MeOH24 to afford propenyl compound 96 (3.6:1 E/Z) in high yield (Scheme 16). This compound could then be processed via an oxidative cleavage of the double bond to eventually produce both the lactol acetate 94 and the δ-lactone 95 in good yields. Once again, this sequence precluded silyl protection of the alcohol moiety of intermediate 88. The synthesis of alstoscholarisine B (2) made use of the δ-lactone 95 (Scheme 17).21 Thus, methyllithium could be added selectively to the lactone carbonyl group of 95 at 78°C to afford hemiketal 97 as a single
Scheme 16 Alternate route to lactol acetate 94 and δ-lactone 95.
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
31
Scheme 17 Conversion of δ-lactone 95 to ()-alstoscholarisine B.
stereoisomer (configuration not determined). Treatment of this compound with trimethylsilyl trifluoromethanesulfonate and triethylsilane15,16 at 78°C presumably generates an oxocarbenium ion intermediate which is then reduced from the less hindered face to afford the equatorial α-methyltetrahydropyran 98 as a single stereoisomer. Removal of the tosyl group from 98 with Mg/MeOH25 and subsequent N-methylation26 of the resulting secondary amine 99 led to ()-alstoscholarisine B (2). For the synthesis of alstoscholarisine C (3), lactol acetate 94 was first treated with a mixture of trimethylsilyl trifluoromethanesulfonate and trimethylaluminum15,16 to afford α-methyltetrahydropyran 100 as a single stereoisomer having the requisite configuration as shown (Scheme 18). To complete the synthesis of the alkaloid, the N-Ts group of intermediate 100 was removed reductively with magnesium metal in methanol25 providing amine 101, followed by N-methylation,26 to give ()-alstoscholarisine C (3). A basic hydrolysis of the methyl ester with sodium hydroxide in aqueous ethanol at 70°C served to transform alstoscholarisine C (3) into alstoscholarisine D (4) in good yield.
5.3 Luo/Xia Synthesis of ()-Alstoscholarisine H Luo, Xia, and coworkers have described a short five-step total synthesis of racemic alstoscholarisine H (8).5 In this approach, indole ester 102 was first
ARTICLE IN PRESS 32
Jeremy D. Mason and Steven M. Weinreb
Scheme 18 Conversion of lactol acetate 94 to ()-alstoscholarisines C and D.
deprotonated with LDA and the resulting enolate was added to the pyridinium salt 103, followed by treatment of the crude product with triflic acid. This sequence allowed a one-pot synthesis of a separable mixture of epimeric adducts 104 and 105 (5.6:1 ratio) containing the requisite bridged aminal ring system via a formal [3 + 3]-cycloaddition process. The actual isolated yield for this transformation is low, but is more reasonable when calculated based on recovered starting material. This mixture of epimeric esters (or either one individually) could be hydrolyzed and decarboxylated under acidic conditions, followed by in situ reduction of the resulting iminium ion, leading to a single product 106 in moderate yield, where the methyl ester group at C16 had epimerized under the reaction conditions to the more stable exo configuration. In order to establish the required C16 stereochemistry of the alkaloid it was necessary to utilize a three-step sequence: air autoxidation of the potassium enolate derived from ester 106 utilizing KHMDS occurred from the least hindered face to give a good yield of α-hydroxyester 107, which could then be reduced to diol 108. Finally, treatment of diol 108 with triethylsilane under Lewis acidic conditions led to racemic alstoscholarisine H (8) in high yield, presumably via reduction of intermediate azafulvene 109 from the least hindered exo face. It might also be noted that diol 108 could be oxidized to the corresponding aldehyde (i.e., 22, Scheme 4), but all attempts to convert this compound to alstoscholarisine J (10) by acetalization failed (Scheme 19).
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
33
Scheme 19 Synthesis of ()-alstoscholarisine H.
Scheme 20 Enantioselective synthesis of bicyclic iodo lactams 113 and 116.
5.4 Liao Synthesis of (–)-Alstoscholarisines A, E, and Their Enantiomers A 2018 report by Liao and coworkers has described an enantioselective approach to (–)-alstoscholarisines A (1) and E (5), as well as their (+)-enantiomers.28 The approach commenced from the known, enantiomerically pure silylalkynyl lactam 110, which upon exposure to TBAF led to terminal alkyne lactam 111 (Scheme 20). This intermediate was then converted to the corresponding boron enolate, which underwent an aldol condensation with acetaldehyde, followed by rhodium-catalyzed cyclization of the aldol product onto the alkyne to afford a 1:2.4 mixture of bicyclic lactam enol ethers 112 (required for alstoscholarisine A) and 114 in good overall yield.
ARTICLE IN PRESS 34
Jeremy D. Mason and Steven M. Weinreb
trans-Fused compound 114 could be isomerized to the cis-fused ring system 115 (required for alstoscholarisine E) upon treatment with potassium hydride. Both enol ethers 112 and 115 could subsequently be converted to vinyl iodides 113 and 116, respectively, with N-iodosuccinimide/silver nitrate. For the synthesis of (–)-alstocholarisine A (1), it was found that iodide 113 can be coupled in a Suzuki-Miyaura reaction with indole boronic acid 117 to afford tetracycle 118 (Scheme 21). Removal of the Boc protecting groups of 118 was effected with methanesulfonic acid, and subsequent hydrogenation of the double bond of the product from the least hindered face using Pearlman’s catalyst then led to lactam 119. A number of methods were attempted to partially reduce lactam 119, and it was eventually discovered that treating the compound with triflic anhydride and tributyltin hydride led to the desired bridged aminal 121, presumably via the intermediate iminium ion 120. Subsequent N-methylation of 121 with formalin and sodium cyanoborohydride then afforded ()-alstoscholarisine A (1). Thus, an eight-step synthesis of the alkaloid was completed starting from lactam 110. Using the same sequence of reactions, it was possible to prepare (–)alstoscholarisine E (5) starting from the epimeric bicyclic iodo lactam 116 in generally similar yields. Moreover, starting from the enantiomer of lactam 110, (+)-alstoscholarisines A and E were also produced.
Scheme 21 Synthesis of ()-alstoscholarisine A (1) from iodo lactam 113.
ARTICLE IN PRESS The Alstoscholarisine Alkaloids
35
6. SUMMARY The alstoscholarisines are a small family of biologically and structurally interesting polycyclic indole alkaloids recently isolated from the leaf extracts of A. scholaris. The alkaloids can be divided into three structural types: (1) ()-alstoscholarisines A–E, (2) (+)-alstoscholarisine G, and (3) (+)-alstoscholarisines H–J. These compounds appear, however, to have a related biogenesis, all deriving from the well-known indole alkaloid (+)-stemmadenine. ()-Alstoscholarisines A–E show significant in vitro activity in NSC proliferation and differentiation. No biological activity has as yet been reported for the remaining four alkaloids. These alkaloids have received limited attention to date from a synthetic standpoint. Two groups have reported total syntheses of racemic alstoscholarisine A, and one of these groups has also described total syntheses of racemic alstoscholarisines B–E. Moreover, a report has appeared describing a total synthesis of ()-alstoscholarisine A, the natural enantiomer of the alkaloid. In addition, a concise synthesis of racemic alstoscholarisine H has appeared. The unique molecular architectures of these metabolites should make them attractive for future synthetic endeavors.
REFERENCES 1. Khyade, M. S.; Kasote, D. M.; Vaikos, N. P. J. Ethnopharmacol. 2014, 153, 1–18. 2. Yang, X.-W.; Yang, C.-P.; Jiang, L.-P.; Qin, X.-J.; Liu, Y.-P.; Shen, Q.-S.; Chen, Y.-B.; Luo, X.-D. Org. Lett. 2014, 16, 5808–5811. 3. For reviews of monoterpenoid indole alkaloids see: (a) Cordell, G. A. Introduction to Alkaloids: A Biogenetic Approach; Wiley-Interscience: New York, 1981; (b) Kutchan, T. M. Phytochemistry 1993, 32, 493–506; (c) Pelletier, S. W. In: Alkaloids: Chemical and Biological Perspectives; Pelletier, S. W. Ed.; Vol. 1; Wiley: New York, 1983; (d) Bisset, N. G. In: Indoles and Biogenetically Related Alkaloids; Phillipson, J. D., Zenk, M. H., Eds.; Academic Press: London, 1980; (e) Herbert, R. B. In: The Chemistry of Heterocyclic Compounds. Indoles, Part 4, the Monoterpenoid Indole Alkaloids; Saxton, J. E. Ed.; Wiley: New York, 1983; (f ) Rahman, A.; Basha, A. Biosynthesis of Indole Alkaloids. Clarendon Press: Oxford, 1983; (g) Ishikura, M.; Yamada, K.; Abe, T. Nat. Prod. Rep. 2010, 27, 1630–1680; h Ishikura, M.; Abe, T.; Choshi, T.; Hibino, S. Nat. Prod. Rep. 2013, 30, 694–752; (i) Dewick, P. M. Medicinal Natural Products: A Biosynthetic Approach, 3rd ed.; Wiley: Chichester, 2009; p. 369; (j) O’Connor, S. E.; Maresh, J. J. Nat. Prod. Rep. 2006, 23, 532–547. 4. Yang, X.-W.; Song, C.-W.; Zhang, Y.; Khan, A.; Jiang, L.-P.; Chen, Y.-B.; Liu, Y.-P.; Luo, X.-D. Tetrahedron Lett. 2015, 56, 6715–6718. 5. Pan, Z.; Qin, X.-J.; Liu, Y.-P.; Wu, T.; Luo, X.-D.; Xia, C. Org. Lett. 2016, 18, 654–657. 6. Hooft, R. W. W.; Straver, L. H.; Spek, A. L. J. Appl. Crystallogr. 2008, 41, 96–103. 7. (a) Scott, A. I. Acc. Chem. Res. 1970, 3, 151–157; (b) Benayad, S.; Ahamada, K.; Lewin, G.; Evanno, L.; Poupon, E. Eur. J. Org. Chem. 2016, 1494–1499.
ARTICLE IN PRESS 36
Jeremy D. Mason and Steven M. Weinreb
8. (a) Kutney, J. P.; Nelson, V. R.; Wigfield, D. C. J. Am. Chem. Soc. 1969, 91, 4279–4280; (b) Scott, A. I.; Yeh, C.-L.; Greenslade, D. J. Chem. Soc., Chem. Commun. 1978, 947–948. 9. (a) Hyman, S. E. Nat. Neurosci. 2016, 19, 1383–1384; (b) Pankevich, D. E.; Altevogt, B. M.; Dunlop, J.; Gage, F. H.; Hyman, S. E. Neuron 2014, 84, 546–553. 10. (a) Efe, A. J.; Ding, S. Philos. Trans. R. Soc., B 2011, 366, 2208–2221; (b) Lyssiotis, C. A.; Lairson, L. L.; Boitano, A. E.; Wurdak, H.; Zhu, S.; Schultz, P. G. Angew. Chem., Int. Ed. 2011, 50, 200–242; (c) Johnson, T. C.; Siegel, D. Chem. Rev. 2017, 117, 12052–12086. 11. Duncan, T.; Valenzuela, M. Stem Cell Res. Ther. 2017, 8, 111–119. 12. Bihelovic, F.; Ferjancic, Z. Angew. Chem., Int. Ed. 2016, 55, 2569–2572. 13. VanRheenen, V.; Kelley, R. C.; Cha, D. Y. Tetrahedron Lett. 1976, 17, 1973–1976. 14. Criegee, R. Chem. Ber. 1931, 64, 260–266. 15. For a recent review of stereoselective Synthesis of tetrahydropyrans see: Zhang, Z.; Tong, R. Synthesis 2017, 4899–4917. 16. See inter alia: (a) Lewis, M. D.; Cha, J. K.; Kishi, Y. J. Am. Chem. Soc. 1982, 104, 4976–4978.; (b) Crawley, G. C.; Briggs, M. T. J. Org. Chem. 1995, 60, 4264–4267; (c) Yamashita, Y.; Hirano, Y.; Takada, A.; Takikawa, H.; Suzuki, K. Angew. Chem., Int. Ed. 2013, 52, 6658–6661. 17. Kirby, A. J. The Anomeric Effect and Related Stereoelectronic Effects at Oxygen. SpringerVerlag: Berlin, 1983. 18. Liang, X.; Jiang, S.-Z.; Wei, K.; Yang, Y.-R. J. Am. Chem. Soc. 2016, 138, 2560–2562. 19. Schafroth, M. A.; Sarlah, D.; Krautwald, S.; Carreira, E. M. J. Am. Chem. Soc. 2012, 134, 20276–20278. 20. Zimmerman, H. E.; Traxler, M. D. J. Am. Chem. Soc. 1957, 79, 1920–1923. 21. (a) Mason, J. D.; Weinreb, S. M. Angew. Chem., Int. Ed. 2017, 56, 16674–16676; (b) Mason, J. D.; Weinreb, S. M. J. Org. Chem. 2018, 5877–5896. 22. For a review of N-sulfonylimines see: Weinreb, S. M. Top. Curr. Chem. 1997, 190, 131–184. 23. (a) Thaler, W. A.; Franzus, W. B. J. Org. Chem. 1964, 29, 2226–2235; (b) Stephenson, L. M.; Mattern, D. L. J. Org. Chem. 1976, 41, 3614–3619; (c) Desimoni, G.; Faita, G.; Righetti, P. P.; Sfulcini, A.; Tsyganov, D. Tetrahedron 1994, 50, 1821–1832. 24. Hanessian, S.; Giroux, S.; Larsson, A. Org. Lett. 2006, 8, 5481–5484. 25. Nyasse, B.; Grehn, L.; Ragnarsson, U. Chem. Commun. 1997, 1017–1018. 26. Lewin, G.; Bernadat, G.; Aubert, G.; Cresteil, T. Tetrahedron 2013, 69, 1622–1627. 27. Schlosser, M.; Jenny, T.; Guggisberg, Y. Synlett 1990, (11), 704. 28. Hu, L.; Yao, L.; Xu, B.; Wang, X.; Liao, X. Org. Lett. 2018, 20, 6202–6205.