A Radical Cascade Enabling Collective Syntheses of Natural Products

Article abstract of DOI:10.1016/j.chempr.2017.04.007

Natural products have long been important inspirations for the development of chemical methodologies, theories, and technologies, and ultimately, discoveries of new drugs and materials. Chemical syntheses have traditionally yielded individual or small groups of natural products; however, methodology development allowing the synthesis of a large collection of natural products remains scarce. Here, we report an efficient photocatalytic radical cascade method that enables access to libraries of chiral and multiple-ring-fused tetrahydrocarbolinones. The radical cascade can controllably introduce complexity and functionality into products with excellent chemo-, regio-, and diastereoselectivity. The power of this distinct method has been demonstrated by the efficient syntheses of 33 monoterpenoid indole alkaloids belonging to four families.

Full text of DOI:10.1016/j.chempr.2017.04.007

Article
A Radical Cascade Enabling Collective
Syntheses of Natural Products
Xiaobei Wang, Dongliang Xia,
Wenfang Qin, ..., Hao Song,
Xiao-Yu Liu, Yong Qin
yongqin@scu.edu.cn
HIGHLIGHTS
Direct generation of nitrogen-
centered radical from sulfonamide
by photocatalysis
Radical cascade reactions with
excellent selectivities
Efficient syntheses of 33 indole
alkaloids and related analogs
Potential applications for the
syntheses of other nitrogen-
containing molecules
Qin and co-workers use photocatalytic conditions to generate a nitrogen-
centered radical from aniline-type sulfonamide, which reverses the conventional
reactivity between two electron-donating amine and enamine groups and
initiates radical cascade reactions with excellent chemo-, regio-, and
diastereoselectivity. The power of this distinct method has been demonstrated by
the efficient syntheses of 33 monoterpenoid indole alkaloids belonging to four
families.
Wang et al., Chem 2, 803–816
June 8, 2017 ª 2017 Elsevier Inc.
http://dx.doi.org/10.1016/j.chempr.2017.04.007

Article
A Radical Cascade Enabling Collective
Syntheses of Natural Products
Xiaobei Wang,1,3 Dongliang Xia,1,3 Wenfang Qin,2,3 Ruijie Zhou,1,3 Xiaohan Zhou,1,3 Qilong Zhou,2,3
Wentao Liu,2 Xiang Dai,2 Huijing Wang,2 Shuqing Wang,2 Ling Tan,1 Dan Zhang,1 Hao Song,1
Xiao-Yu Liu,1 and Yong Qin1,4,
*
SUMMARY
The Bigger Picture
Today, achieving the chemical
synthesis of a single natural
product is rarely out of reach with
sufficient manpower. However,
considering the urgent demand
for chemical biology studies and
drug discovery, efficient
Natural products have long been important inspirations for the development
of chemical methodologies, theories, and technologies, and ultimately, dis-
coveries of new drugs and materials. Chemical syntheses have traditionally
yielded individual or small groups of natural products; however, methodology
development allowing the synthesis of a large collection of natural products
remains scarce. Here, we report an efficient photocatalytic radical cascade
method that enables access to libraries of chiral and multiple-ring-fused tetra-
hydrocarbolinones. The radical cascade can controllably introduce complexity
and functionality into products with excellent chemo-, regio-, and diastereo-
selectivity. The power of this distinct method has been demonstrated by the
efficient syntheses of 33 monoterpenoid indole alkaloids belonging to four
families.
preparation of a large collection
of complex natural products and
natural-product-like libraries with
stereochemical and architectural
divergence remains highly
desirable. Here, we describe a
visible-light-mediated radical
cascade reaction that generates a
nitrogen-centered radical from
aniline-type sulfonamide and
reverses the conventional
INTRODUCTION
Structurally complex and diverse natural products and their analogs have long been
inspirations for the development of new medicines, which have greatly improved
human health.1 When access to such molecules from natural sources is uncertain
or uneconomical, chemical synthesis is one of the most important alternative ways
to obtain them.2,3 Although syntheses of individual or several natural products
belonging to a certain family have been achieved, a breakthrough in synthetic
methods that yields diverse skeletons of natural products and that benefits the
syntheses of a large collection of natural products for biological activity tests is
highly anticipated. With over 15 families and more than 3,000 known members,
monoterpenoid indole alkaloids represent the most stereochemically and architec-
turally diverse natural products.4,5 Most of them have exhibited important biological
activities, and some of them, such as vinblastine and reserpine,6,7 have been used as
clinical drugs for decades (Figure 1A). Three major types of monoterpenoid indole
alkaloids, namely, corynanthe-, aspidosperma-, and iboga-type alkaloids, have
been categorized on the basis of the carbon skeletons of the monoterpenoid
units (bold carbon chain in Figure 1A).4 All of these alkaloids can be roughly catego-
rized into the non-rearranged families (shown in blue in Figure 1A) and the rear-
ranged families (shown in black in Figure 1A), which are believed to have been
biosynthesized originally from tryptamine and secologanin by strictosidine syn-
thase- (STR1) and strictosidine glucosidase (SG)-catalyzed condensation.8 Since
the milestone syntheses of reserpine and strychnine by Woodward et al. in the
1950s,9,10 the syntheses of complex monoterpenoid indole alkaloids has been
an interesting and challenging topic because of their rich structural diversity and
biological activities.11–14
reactivity between two electron-
donating amine and enamine
groups, allowing efficient
syntheses of 33 complex indole
alkaloids with varied architectures
and divergent stereochemistry.
This approach could be expanded
to intermolecular reaction
patterns and could be expected
to facilitate the preparation of
nitrogen-containing
architectures, which are of great
interest for the syntheses of
natural products and bioactive
substances.
Chem 2, 803–816, June 8, 2017 ª 2017 Elsevier Inc. 803

Radical cascades are among the most powerful reactions for bond formation.15–19
However, their application in constructing complex naturally predefined architec-
tures, and especially in resolving the major problems—namely collective syntheses
of bioactive natural products20—is still underexploited.21–23 This is largely because
of the high reactivity of radicals, which is difficult to control with multiple functional-
ities in the substrate, thus resulting in poor stereoselectivity and chemoselectivity. In
addition, tedious preparation of an initiating group for the classic radical reaction
further impedes its application. A design for precise control of a radical pathway is
required for successful assembly of complex natural products. Recent advances in
photocatalytic radical reactions24–31 have provided great opportunities in the syn-
thesis of natural products.32–35 Here, we report a photocatalytic radical cascade
with excellent stereo-, regio-, and chemoselectivity to prepare libraries of chiral tet-
rahydrocarbolinones under mild conditions. This radical cascade can be scaled up
and can generate two of the three monoterpenoid indole-alkaloid skeletons,
including the corynanthe-type and aspidosperma-type alkaloids, by varying easily
accessible substrates. This cascade provided indoline skeletons with stereochemical
divergence, which is typically difficult to achieve and which requires one-by-one
synthesis using traditional synthetic methods. Rapid assembly of the core struc-
tures enabled efficient syntheses of 33 complex monoterpenoid indole alkaloids
belonging to four families and also four unnatural diastereoisomers.
RESULTS AND DISCUSSION
Radical Cascade Design
Given that efficient preparation of aspidosperma-type and corynanthe-type inter-
mediates is expected to facilitate the syntheses of diverse families of indole
alkaloids,36 we envisioned that an electron-deficient nitrogen-centered radical
(Ts-I; Figure 1B), initiated from chiral aniline 2 (easily synthesized from the chiral com-
pound 1, the preparation of which37 is described in the Supplemental Experimental
Procedures; see also Figures S1–S7), would enable the formation of a C–N bond
between both electron-donating aniline nitrogen and the b carbon of the enamine
(Ts-I). The resultant heteroatom-associated carbon radical might intramolecularly
attack the pre-installed double bond tethered to the enamine nitrogen to afford
aspidosperma-type intermediate 3 (pathway a, via Ts-IIa). Similarly, the intermolec-
ular Michael addition (Ts-IIb, pathway b) via Ts-III after radical quenching would give
tetrahydrocarbolinone 4, an important intermediate used for previous indole-alka-
loid syntheses.38–40 In addition, a further cascade reaction might occur via intramo-
lecular addition of the resultant radical in Ts-III with a tethered double bond or triple
bond, thus leading to the corynanthe-type intermediate 5. Compounds 3–5 possess
different ring systems and functionalities suitable for the syntheses of a variety of
indole alkaloids. The untested aspect of the designed radical cascade was whether
the stereochemistry of the adducts could be controlled by the chiral substrate 2.
Given that non-rearranged monoterpenoid indole alkaloids have been described
with varying configurations at the C15 and C20 positions,41 this concern was partly
diminished.
1Key Laboratory of Drug Targeting and Drug
Delivery Systems of the Ministry of Education,
West China School of Pharmacy, State Key
Laboratory of Biotherapy, Sichuan University,
Collaborative Innovation Center, Chengdu
610041, China
Radical Cascade Results
The first key task in achieving the above radical cascade was the direct initiation of a
nitrogen-centered radical from Ts-protected aniline by cleavage of the N–H bond
in 2. Classic generation of amide-type nitrogen-centered radicals usually requires
cleavage of an N–heteroatom bond.42,43 Direct conversion of an N–H bond of ani-
line-type amide to a nitrogen-centered radical via photocatalysis is promising but
has proved challenging and unprecedented; there are only a few recent examples
on alkyl-type amide substrates.44–47 With the above radical cascade strategy for
2School of Pharmaceutic Science, Chongqing
University, Chongqing 401331, China
3These authors contributed equally
4Lead Contact
*Correspondence: yongqin@scu.edu.cn
http://dx.doi.org/10.1016/j.chempr.2017.04.007
804 Chem 2, 803–816, June 8, 2017

Et
20
17
A
N
21
14
OH
5
7
15
3
N
21
3
7
H
3
3
2
14
N
N
3
20
7
N
16
14
14
19
17
2
21
N
15
20
16
20
H
15
20
21
16
H
15
2
MeO2C
7
2
N
N
16
14
21
HO
MeO2C
Et
17
OH
H
H
15
19
OH
CO2Me
MeO2C
17
Et
17
N
R
MeO
OAc
16
Catharanthine
Vincamine
(Aspidosperma-type
carbon skeleton)
H
Stemmadenine
(Iboga-type carbon skeleton)
CO2Me
Vinblastine (R=Me)
Vincristine (R= CHO)
6
6
7
7
5
5
19
18
21
3 CHO
2
2
N
N
21
H
O
21
MeO
N
N
1) STR1
2) SGD
20
15
14
3
3
OGlc
H
H
H
H
H
H
20
+
15
H
14
22
19
18
20
19
NH2
N
15
16
22
H
H
16
18
MeO2C
16
MeO2C
22
17
17
OR
MeO2C
Tryptamine
OH
17
Reserpine
Corynanthe-type
carbon skeleton
Secologanin
OMe
R = 3,4,5-(OMe)3PhCO
6
7
5
5
22
N
17 CO2Me
2
6
21
20
N
6
5
O
22
16
21
20
N
OH
3
18
3
7
14
15
7
CO2Me
H
H
14
6
3
16
17
2
15
N
21
16
19
18
2
19
3
20
21
15
N
14
2
H
N
20
19
N
16
H
H
15
H
H
5
17
19
N
17
14
Me
O
O
Geissoschizol
H
18
Rhazimal
Strychnine
Vincamajine
OH
B
H
H
O
H
O
H
O
O
R
Ir+ / B–
blue LEDs
H
H
b
N
N
N
R
N
R
N
N
N
N
H
Ts
R3
Ts
H
deprotonation;
then oxidation
Ts
Ts
2
Ts-I
Ts-IIb
EWG
Ts-III
EWG
a
3 steps
RNH2
H
N
H
H
H
O
O
O
O
H
CO2Me
CHO
H
H
H
H
N
N
N
R
N
N
N
N
H
H
H
R1
Ts
Ts
Ts
Ts
NO2
1
R1
EWG
EWG
R2
R2
Ts-IIa
R3
Aspidosperma-type
intermediate 3
Corynanthe-type
intermediate 5
Tetrahydrocarbolinone 4
Figure 1. Background and Synthetic Plan
(A) Proposed biosynthetic pathway and structural categorization of monoterpenoid indole alkaloids.
(B) Envisioned photocatalytic radical cascade to generate aspidosperma-type intermediate 3 (path a), tetrahydrocarbolinone 4, and corynanthe-type
intermediate 5 (path b) for the syntheses of monoterpenoid indole alkaloids. Abbreviations: D-A reaction, Diels-Alder reaction; STR1, strictosidine
synthase; SG, strictosidine glucosidase.
indole-alkaloid synthesis in mind, we used tosyl-protecting aniline amide 2a (Figures
S8–S11) as a substrate for generating the nitrogen-centered radical via photocatal-
ysis in order to test the designed two-step radical cascade (pathway a; Figure 1B).
After extensive screening of photocatalysts, bases, solvents, and temperature,
optimal conditions (0.5 mol % Ir(dtbbpy) (ppy)2PF6, 5 equiv KHCO3, degassed tetra-
hydrofuran [THF], 35ꢀC, 5 W blue LEDs, Ar) were determined. Under these condi-
tions, the radical cascade of 2a proceeded smoothly to produce the desired 3a
Chem 2, 803–816, June 8, 2017 805

Figure 2. Preparation of Aspidosperma-Type Derivatives 3
Optimal conditions: 0.5 mol % Ir(dtbbpy) (ppy)2PF6, 5 equiv KHCO3, degassed THF, 5 W blue LEDs, 35ꢀC, Ar. The dr values were determined by LC-MS analysis.
The relative carbon configurations at the C-ring and D-ring were determined by nuclear Overhauser effect experiments. 3a and 3l: the absolute configuration was
determined by X-ray crystallography of the detosylated intermediates (see the Supplemental Information). 3a–3m: the dr values or E/Z ratios were determined by
LC-MS analysis. A dr > 50:1 indicated that other diastereoisomers were not detected in the LC-MS experiments. 3a: the yield in parentheses was obtained with 50%
aqueous THF as the solvent. 3n: the dr value was determined by 1H NMR analysis. Ir(dtbbpy) (ppy)2PF6, (4,40-di-tert-butyl-2,20-bipyridine)bis[(2-pyridinyl)phenyl]
iridium(III) hexafluorophosphate. See also Figures S8–S156.
(Figures S20–S22) in 70% yield and >50:1 diastereomeric ratio (dr), along with a 2%
yield of by-product 3a0 (Figure 2; see also Figures S23–S29). Notably, the reaction for
production of 3a could be performed in aqueous THF without loss of yield and dia-
stereoselectivity. The radical reaction did not take place under the same conditions
when the Ts protecting group in 2a was replaced with Me, CO2Me, Bz, and Boc (see
the Supplemental Experimental Procedures and Figures S12–S19). Presumably, this
could be due to their tendency for weaker deprotonation, which suppressed the
initiation of a nitrogen-centered radical. In addition, a control experiment using a
similar substrate without tosyl amide functionality was carried out under the same
conditions; the starting material remained unchanged after 12 hr (Scheme S1), indi-
cating that oxidation of the nitrogen anion to initiate a nitrogen-centered radical was
crucial for the cascade reaction.
A variety of substrates (2b–2n; Figures S30–S87) with either an electron-rich double
bond or an electron-deficient Michael acceptor linked to the enamine nitrogen
were then used for the radical cascade. To our delight, all reactions afforded the
806 Chem 2, 803–816, June 8, 2017

corresponding products 3b–3m (Figures S90–S150) with a fused pyrrolidine or
piperidine ring in moderate to high yields and high to excellent dr. An exceptionally
low dr value (2.5:1) was observed in adduct 3n (Figures S151 and S152), which has a
more flexible azepane ring. This substrate-controlled cascade proceeded in a highly
stereoselective manner to completely control the relative configurations of C2 and
C3 as a trans relationship.
In view of the stereoselectivity for the radical cascade, the addition of the nitrogen-
centered radical to the enamine double bond in the first step proceeded from the
upper face (Ts-I; Figure 2) to give transition states Ts-IIa0 and Ts-IIa, both possessing
a cis relationship between H2 and H7. Ts-IIa could be more reactive than Ts-IIa0,
considering that twisting of the amide in Ts-IIa would dramatically increase the
donating ability of the lone electron pair of the amide nitrogen,48 thus rendering
the radical much more nucleophilic. Furthermore, because the upper face of the
piperidinone in Ts-IIa0 should be blocked by the Ts-protected indoline group, addi-
tion of a radical to the double bond (triple bond for 3m) from the upper face of
the piperidinone was unfavorable because of steric repulsion. An equilibrium shift
of Ts-IIa0 to Ts-IIa resulted in a dominant intramolecular attack of the radical to the
double bond (or triple bond) from the bottom face of the piperidinone ring, which
gave adduct 3 with a trans relationship between H2 and H3. Direct quenching of
the radical in Ts-IIa0 or Ts-IIa before the addition to the double bond (or triple
bond) provided the by-product 3a0.
Having successfully achieved the two steps of the intra- and intramolecular radical
cascade for generating aspidosperma-type derivatives 3, we then chose 2o (Fig-
ures S157–S160) with an inert N-benzyl (N-Bn) group as a substrate to explore the
planned intra- and intermolecular radical cascade (Figure 3A). Under the same con-
ditions as above, Michael acceptors 6–9 (1.1–4.0 equiv) were then used in the radical
cascade to afford the corresponding tetrahydrocarbolinones 4a–4n (Figures S161–
S198) in moderate to high yields. Although the E/Z configuration at the side chain
in 4i–4l was not controllable, the newly generated stereocenters at C2 and C3
were formed exclusively in a trans configuration in 4a–4n. Using Michael acceptors
8 and 9, we were able to isolate 4m and 4n with a methylene group a and b to C3,
respectively. The reaction generating 4b was also performed in aqueous THF
without loss of yield and diastereoselectivity.
For the preparation of corynanthe-type derivatives 5, we planned to use substrate 10
(Figures S199–S239) bearing a Michael acceptor or a double bond tethered to
the piperidinone ring in an attempt to expand the radical chain reaction from the
intermolecular radical addition (Ts-IIb) in the second step to the intramolecular
radical addition (Ts-III) in the third step to give 5 (Figure 3B). In this case, the Ts-IIb
conformer would be crucial for enabling intermolecular trapping by increasing the
lifetime of the radical and disfavoring direct intramolecular 4-exo-trig cyclization
with the double bond on the nitrogen side chain. According to this design, we
then used alkynyl-type Michael acceptors 7a–7c to perform the programmed radical
cascade, which provided 5a–5h (Figures S240–S278) in 36%–76% yields with excel-
lent diastereoselectivities, as well as 7%–44% yields of by-product 11 in some cases
(see the Supplemental Information). The minor diastereoisomers were the C20
epimers. The by-product 11 was produced by direct radical quenching of Ts-III0
and Ts-III. With regard to the excellent stereocontrol for the D-ring formation, the
transition-state Ts-III rather than Ts-III0 might be responsible for generating the
D ring with H20 located at the a bond in adduct 5. The strong steric repulsion be-
tween the electron-withdrawing group (EWG) and R2 was thus avoided in Ts-III.
Chem 2, 803–816, June 8, 2017 807

A
B
in (A)
Figure 3. Preparation of Tetrahydrocarbolinones 4 and Corynanthe-Type Derivatives 5
(A) Optimal conditions: 0.5 mol % Ir(dtbbpy) (ppy)2PF6, 5 equiv KHCO3, degassed THF, 5 W blue LEDs, 35ꢀC, Ar. For the syntheses of 4a–4h, 4m, and 4n,
1.1–2.5 equiv Michael acceptors 6, 8, and 9 were used. For the syntheses of 4i–4l, 4 equiv Michael acceptor 7 was used. 4b: the yield in parentheses was
obtained with 50% aqueous THF as the solvent.
(B) 7: 2.5–5 equiv was used. 5a–5i: the dr values of 5 in (B) were determined by LC-MS analysis. The relative configurations of 5a, 5b, and 5d–5i were
determined by comparison of their NMR spectra with that of 5c; the absolute configuration was confirmed by conversion of 5c to 28a and 29a (see
Scheme 2 and the Supplemental Information for details). 5h: dioxane was used as the solvent. See also Figures S157–S283.
808 Chem 2, 803–816, June 8, 2017

A
B
C
Scheme 1. Total Syntheses of the Eburnamine-Vincamine Family Alkaloids
(A) Reagents and conditions: (a) 1 mol % Ir(dtbbpy) (ppy)2PF6, 5 equiv KHCO3, degassed acetonitrile, 30 W blue LEDs, 35ꢀC, Ar; (b) p-TsOH, CH2Cl2,
1,3-propanediol, room temperature (RT); (c) sodium naphthalenide, THF, ꢁ78ꢀC; (d) benzeneseleninic anhydride, THF, 40ꢀC; (e) BH3$SMe2, THF, 0ꢀC to
RT; (f) trifluoroacetic acid (TFA), CH2Cl2, RT; (g) 0.5 M HCl (aq), RT; (h) pyridinium dichromate (PDC), CH2Cl2, RT.
(B) Reagents and conditions: (a) KOtBu, THF, Ph3P+CH3Brꢁ, 0ꢀC to RT; (b) sodium naphthalenide, THF, ꢁ78ꢀC; (c) benzeneseleninic anhydride, THF,
ꢀ
ꢀ
˚
40 C; (d) BH3$SMe2, THF, 0 C to RT, then NaBO3; (e) tetrapropylammonium perruthenate, N-methyl morpholine-N-oxide, CH2Cl2, 4 A molecular sieves,
RT, then PDC, CH2Cl2, RT; (f) sodium bis(trimethylsilyl)amide (NaHMDS), tBuONO, PhMe, 50ꢀC; (g) TsOH, paraformaldehyde, AcOH, 100ꢀC; then KOtBu,
MeOH, RT.
(C) Reagents and conditions: (a) NaH, dimethylformamide, MeI, 0ꢀC to RT; (b) 6 M HCl (aq)/THF (1:1 v/v), RT; (c) hexamethyl phosphoryl triamide, SmI2,
THF, reflux; (d) NaH, CS2, THF; then MeI, 0ꢀC to RT; (e) azodiisobutyronitrile (AIBN), Bu3SnH, PhMe, 80ꢀC; (f) 15 mol % tris(triphenylphosphine)rhodium(I)
carbonyl hydride (Rh(H)(CO)(PPh3)3), PhSiH3, THF, RT. See also Figures S284–S345.
When the R2 group was a methyl group, reaction of 10g with 7b (EWG = Ac) afforded
5i (Figures S279–S283) in a lower yield (17%) and a moderate dr value (7:1). The
results were reasonable given that a simple double bond rather than a Michael
acceptor in 10g was relatively inert toward the intramolecular attack of an olefin-
type radical in the third step, and the steric repulsion between the methyl group
and the acetyl group in Ts-III0 was partially alleviated.
Syntheses of the Eburnamine-Vincamine Family Alkaloids
Application of the established radical cascade to total syntheses of indole alkaloids
was first demonstrated in the syntheses of eburnamine-vincamine family alkaloids.
As shown in Scheme 1A, by conducting a two-step radical cascade of 2p (Figures
S284–S297) on a 14 g scale, we were able to isolate two diastereomers, 12a and
12b (Figures S298–S303), in 81% combined yield and 1:1.5 ratio. After masking of
the aldehyde in 12b as a dioxane, the tosyl (Ts)-protecting group in the resultant
13 was removed with sodium naphthalenide to afford amine 14 in excellent yield.
Facile oxidation of the indoline moiety to indole (14 to 15), followed by amide
Chem 2, 803–816, June 8, 2017 809

reduction and treatment with trifluoroacetic acid, gave (+)-eburnamenine (16; Fig-
ures S312 and S313) in 73% yield in three steps. Hydration of 16 with HCl/H2O
resulted in (ꢁ)-eburnamine (17; Figures S314 and S315) and (+)-isoeburnamine
(18; Figures S316 and S317). Further oxidation of the mixture of 17 and 18 led to
the formation of the alkaloid (+)-eburnamonine (19; Figures S318–S321).49
Intermediate 12b was used for the synthesis of (ꢁ)-vincamine (23) in seven steps
(Scheme 1B). Wittig methylenation of 12b and removal of the Ts group afforded
20 in excellent yield. After oxidation of the indoline moiety, the double bond in 20
was then converted into a hydroxy group by hydroboration oxidation, which resulted
in concomitant reduction of the amide to give alcohol 21. Oxidation of the hydroxy
group in 21 led to the formation of lactam 22. Finally, 23 (Figures S332–S344) was
synthesized from 22 according to a literature procedure.50
The synthesis of (ꢁ)-vallesamidine (27) was achieved with 15 as the starting material
(Scheme 1C). After two steps of methylation and deprotection, aldehyde 24 was
treated with SmI2 to generate a ketyl radical, which was then added to the indole
double bond to afford 25 as a pair of diastereomers. Removal of the hydroxy group
in 25, followed by reduction of the resulting 26 with rhodium(I) hydride,51 provided
27 (Figures S342–S345) in excellent yield.
Syntheses of the Yohimbine Family Alkaloids
In the analysis of the radical cascade in Figure 3B, we noticed that a yohimbine-type
core could be easily constructed if methyl vinyl ketone (6c) was used as a Michael
acceptor in place of the alkynyl-type counterpart (Scheme 2A). In this context, sepa-
rable diastereoisomers 28a, 28b, 29a, and 29b (Figures S346–S355) were isolated in
56% combined yield and a 3:5:73:19 ratio if compounds 10a and 6c (1.2 equiv) were
subjected to the radical cascade conditions on a 12 g scale, followed by treatment
with p-TsOH in toluene. The absolute configurations of 28a, 28b, 29a, and 29b
were determined by X-ray crystallography (see the Supplemental Information for
details). Major isomers 29a and 29b were then used as starting materials for the
syntheses of yohimbane (31) and alloyohimbane (32) via a six-step transformation.
Specifically, four steps of individual hydrogenation, reduction, and the removal of
hydroxy and Ts groups in 29a and 29b provided the corresponding amines 30a
and 30b, respectively. Further indoline oxidation and amide reduction afforded 31
(Figures S376–S379) and 32 (Figures S380–S383), respectively.
For the syntheses of natural 17-epi-yohimbol (36), yohimbone (37), b-yohimbine (41),
yohimbine (43), and the unnatural yohimbinone (42) and 16-epi-b-yohimbine (44), in-
termediate 29a with a trans-fused D/E ring system was used as the starting material
(Scheme 2B). Mukaiyama hydration of enone 29a led to the desired alcohol 33 as
well as an unexpected by-product 34.52 It is clear that 34 was generated by enol re-
arrangement from the unstable 17-epimer of 33 under the chromatographic condi-
tions, which were then used for the syntheses of 36 and 37. Removal of the hydroxy
group and the Ts group in 34 resulted in simultaneous stereoselective reduction of
ketone to give 35 as a single diastereoisomer. The two subsequent steps of oxida-
tion of the indoline moiety and reduction of the amide in 35 afforded 36 (Figures
S394–S399). Further oxidation of 36 afforded 37 (Figures S400–S403) in 60% yield.
However, after protection of the hydroxy group in 33, Corey-Chaykovsky epoxida-
tion of the resultant 38, followed by opening of the epoxide ring,53 generated two
separable diastereoisomers, 39a and 39b. Compounds 39a and 39b were then indi-
vidually converted to the corresponding esters 40a and 40b in approximately 50%
yield. Further functional group transformation of 40a, including reduction of the
810 Chem 2, 803–816, June 8, 2017

A
B
C
Scheme 2. Total Syntheses of the Yohimbine Family Alkaloids
(A) Reagents and conditions: (a) 0.5 mol % Ir(dtbbpy) (ppy)2PF6, 5 equiv KHCO3, THF, 30 W blue LEDs, 25ꢀC; (b) p-TsOH, PhMe, 60ꢀC; (c) Pd(OH)2, H2,
EtOAc, RT; (d) NaBH4, MeOH, RT; (e) CS2, NaH, Mel, THF, 0ꢀC; then AIBN, Bu3SnH, PhMe, 80ꢀC; (f) Mg, MeOH, RT; (g) benzeneseleninic anhydride, THF,
40ꢀC; (h) 15 mol % Rh(H)(CO)(PPh3)3, PhSiH3, THF, RT.
i
(B) Reagents and conditions: (a) tris(2,2,6,6-tetramethyl-3,5-heptanedionato)manganese(III) (Mn(dpm)3), PhSiH3, O2, PrOH/CH2Cl2 (1:4 v/v), RT;
then silica gel; (b) 1,10-thiocarbonyldiimidazole, 4-dimethylaminopyridine, 1,2-dichloroethane (DCE), RT; then AIBN, Bu3SnH, DCE, reflux; (c) Mg,
MeOH, RT; (d) benzeneseleninic anhydride, THF, 40ꢀC; (e) 15 mol % Rh(H)(CO)(PPh3)3, PhSiH3, THF, RT; (f) dicyclohexylcarbodiimide, Cl2CHCOOH,
DMSO, 35ꢀC; (g) BnBr, Ag2O, PhMe, 45ꢀC; (h) Me3S+Iꢁ, KOtBu, DMSO, RT; (i) Zn, Cp2TiCl2, 1,4-cyclohexadiene (1,4-CHD), THF, RT; (j) 2,2,6,6-
t
tetramethyl-1-piperidyloxy (TEMPO), PhI(OAc)2, MeCN/H2O (1:1 v/v); then NaClO2, NaH2PO4, 2-methyl-2-butene, BuOH/H2O (1:1 v/v), RT; then
trimethylsilyldiazomethane (TMSCHN2), PhMe/MeOH (5:1 v/v), RT; (k) Mg, MeOH, RT; (l) benzeneseleninic anhydride, THF, 45ꢀC; (m) BBr3,
CH2Cl2, ꢁ78ꢀC; (n) Ac2O, DMSO, RT; (o) lithium triisobutylhydroborate, THF, ꢁ78ꢀC, 43/41 6:1.
i
(C) Reagents and conditions: (a) Mn(dpm)3, PhSiH3, O2, PrOH/CH2Cl2 (1:4 v/v), RT; (b) tert-butyldimethylsilyl chloride, imidazole, CH2Cl2, RT; (c)
n
Me3S+Iꢁ, BuLi, THF, 0ꢀC; (d) Zn, Cp2TiCl2, 1,4-CHD, THF, 50ꢀC; (e) TEMPO, PhI(OAc)2, MeCN/H2O (1:1 v/v); then NaClO2, NaH2PO4, 2-methyl-
t
2-butene, BuOH/H2O (1:1 v/v), RT; then TMSCHN2, PhMe/MeOH (5:1 v/v), RT; (f) Mg, MeOH, RT; (g) benzeneseleninic anhydride, THF, 40ꢀC;
(h) 15 mol % Rh(H)(CO)(PPh3)3, PhSiH3, THF, RT; (i) 5 M HCl/THF (1:3 v/v), 30ꢀC; (j) pyridine hydrofluoride, THF, 40ꢀC; (k) 1 M MeONa in MeOH, 30ꢀC.
See also Figures S346–S467.
Chem 2, 803–816, June 8, 2017 811

amide and O-debenzylation with BBr3, afforded 41 (Figures S428–S431). Ultimately,
a sequential oxidation-reduction of 41 via 42 (Figures S434 and S435) delivered
(ꢁ)-yohimbine (43; Figures S436–S439).14,54 Similarly, the diastereoisomer 44 (Fig-
ures S432 and S433) was synthesized from 40b in 49% yield according to the same
procedure as that applied for the synthesis of 41.
Having developed a synthetic route to yohimbine (43) and related alkaloids from
29a, we were able to synthesize natural rauwolscine (48), 17-epi-alloyohimbine
(50), and the unnatural 17-epi-rauwolscine (49) from 29b, which possessed a cis-
fused D/E ring system (Scheme 2C). Unlike 29a, Mukaiyama hydration of 29b pro-
vided two inseparable diastereoisomers 45a and 45b in 52% yield and a 1:2 ratio
after tert-butyldimethylsilyl (TBS) protection of the resultant hydroxy group. A
mixture of 46a and 46b was obtained in 49% yield when Corey-Chaykovsky epoxida-
tion, followed by opening of the epoxide ring, was performed. After oxidation of the
hydroxy groups in 46a and 46b to esters, the Ts groups were removed to give two
separable diastereoisomers, 47a and 47b, in 69% combined yield. Finally, three
steps of indoline oxidation, amide reduction, and TBS deprotection from 47a fur-
nished rauwolscine (48; Figures S456–S459). In a similar procedure, 49 (Figures
S464 and S465) was readily synthesized from 47b, and 50 (Figures S466 and S467)
was obtained by epimerization of 49 at C16.
Syntheses of the Corynanthe and Heteroyohimbine Family Alkaloids
In addition to the divergent stereochemistry at C15 and C20, the distinction among
the large members of the corynanthe family alkaloids is the geometric variation of
the double bond at C19. Therefore, for syntheses of alkaloids in this family, we
used 6i (5 equiv) and 10h (Figures S469–S472) with a 2-butynyl side chain to conduct
a three-step radical cascade on a 20 g scale, which afforded two mixtures of the two
diastereoisomers 51a–51b and 51c–51d in excellent combined yield (Scheme 3A;
see also Figures S468 and S473–S480). The E/Z geometric isomers were completely
separated at this stage. After homologation of the ester side chain at C15 and
removal of the Ts group in 51 via a five-step transformation from individual mixtures
of 51a–51d, the four separable diastereomers 52a–52d were obtained. Starting from
each isomer, we were able to synthesize eight corynanthe family alkaloids, 54–61
(Figures S513–S520 and S527–S542), and unnatural 3-epi-isogeissoschizol (53; Fig-
ures S511 and S512) via two to four steps of functionality manipulations (see Supple-
mental Experimental Procedures).
For further syntheses of the corynanthe family alkaloids 67–71, a strategy that intro-
duces a C18–C19 terminal double bond at an early stage of synthesis was designed
(Scheme 3B). This strategy also enabled the formation of a dihydropyran ring by con-
verting the terminal alkene into a hydroxy group at C19, thus facilitating the synthe-
ses of heteroyohimbine family alkaloids 74 and 76–80 with varied stereochemistry at
C15, C19, and C20 (Scheme 3C).41 To form a C18–C19 terminal double bond, we
performed the radical cascade by using 6i (5 equiv) and 10f to provide a mixture
of 62a–62c (Figures S543 and S544) and a single diastereomer 62d (Figures S545
and S546) in excellent yield. Removal of the Ts group in 62d with Mg/MeOH readily
effected epimerization at C15 to give 63a. Coincidentally, there is no corynanthe
family alkaloid that possesses the same stereochemistry at C3, C15, and C20 as
that of 62d. Treatment of the mixture of the three diastereoisomers 62a–62c with
Mg/MeOH afforded a mixture of two diastereoisomers 63a and 63b in 65% yield
and a single diastereoisomer 63c in 17% yield. After oxidation of the indoline moiety
in 63a and 63b, diastereoisomers 64a and 64b were separated. Similarly, 64c
was obtained by oxidation of 63c. After extension of the C15 ester chain by a
812 Chem 2, 803–816, June 8, 2017

A
B
C
Chem 2, 803–816, June 8, 2017 813

Scheme 3. Total Syntheses of the Corynanthe and Heteroyohimbine Family Alkaloids
(A) Reagents and conditions: (a) 1 mol % Ir(dtbbpy) (ppy)2PF6, 5 equiv KHCO3, THF, 30 W blue LEDs, 35ꢀC, Ar; (b) LiBH4, THF, 0ꢀC to RT; (c)
methanesulfonyl chloride (MsCl), Et3N, CH2Cl2, RT; (d) trimethylsilyl cyanide (TMSCN), tetra-n-butylammonium fluoride (TBAF), THF, reflux; (e) HCl (g),
MeOH, 0ꢀC to RT; (f) Mg, MeOH, RT; (g) benzeneseleninic anhydride, THF, 40ꢀC; (h) LiAlH4, PhMe, 0ꢀC to RT; (i) 5 mol % Rh(H)(CO)(PPh3)3, PhSiH3, THF,
RT; (j) lithium diisopropylamide, HCO2Me, THF, ꢁ78ꢀC to RT; (k) AcOH, NaBH4, MeOH, 0ꢀC.
(B) Reagents and conditions: (a) 0.5 mol % Ir(dtbbpy)(ppy)2PF6, 5 equiv KHCO3, acetonitrile, 15 W blue LEDs, Ar, 37ꢀC; (b) Mg, MeOH, RT; (c)
benzeneseleninic anhydride, THF, 40ꢀC; (d) LiAlH4, THF, ꢁ30ꢀC; (e) MsCl, Et3N, CH2Cl2, RT; (f) TMSCN, TBAF, THF, reflux; (g) HCl (g), MeOH, 0ꢀC–60ꢀC;
(h) 7 mol % Rh(H)(CO)(PPh3)3, PhSiH3, THF, RT; (i) LiAlH4, THF, 0ꢀC to RT; (j) H2, 10% PtO2, MeOH, RT.
(C) Reagents and conditions: (a) Pd(OAc)2, HClO4, 1,4-benzoquinone, MeCN/H2O (7:1 v/v), RT; (b) 7 mol % Rh(H)(CO)(PPh3)3, PhSiH3, THF, RT; (c) NaBH4,
MeOH, ꢁ30ꢀC; (d) NaHMDS or LiHMDS, NCCO2Me, THF, ꢁ78ꢀC; (e) diisobutylaluminum hydride or NaBH4; (f) p-TsOH, CH2Cl2, reflux; then K2CO3,
MeOH, RT; (g) Ph3P, diethyl azodicarboxylate, p-nitrobenzoic acid, THF, 0ꢀC–30ꢀC; then Cs2CO3, MeOH, RT. See also Figures S468–S640.
four-step transformation (64a to 66a), stepwise reduction of the amide and ester and
then the double bond in 66a generated 67 (Figures S575 and S576) and then 68 (Fig-
ures S579–S582), respectively. Similarly, 69–71 (Figures S577, S578, and S583–S586)
were synthesized from 66b and 66c according to the above procedure for the syn-
theses of 67 and 68, respectively.
For the syntheses of heteroyohimbine family alkaloids, 66a, 66b, and 65c were used
as starting materials (Scheme 3C). A three-step sequence of Wacker oxidation,
amide reduction, and stereoselective ketone reduction was applied to 66a to pro-
vide the single diastereoisomer 72 in 44% overall yield. Carbonylation of 72 resulted
in the formation of the lactone ring and the introduction of two methoxycarbonyl
groups in one pot to yield 73. Further reduction and hydrolysis of 73 gave rise to
ajmalicine (74; Figures S601–S604). Lactone 75 was obtained in 63% yield from 72
by Mitsunobu inversion of the C19 hydroxy group configuration, followed by hydro-
lysis. Mayumbine (76; Figures S596–S598) was prepared according to a procedure
similar to that applied for the synthesis of ajmalicine (74). In addition, after similar
functional-group manipulations, alkaloids 77–80, possessing different stereochem-
istry at C15 and C19, were synthesized from 66b and 65c, respectively (see Supple-
mental Experimental Procedures and Figures S605–S640).
Conclusion
We have developed an innovative photocatalytic radical cascade that reverses the
conventional reactivity between two electron-donating amine and enamine groups
and allows highly efficient assembly of manifold core structures of indole alkaloids
with divergent stereochemistry and functionalities from simple starting materials.
The power of this radical cascade has been demonstrated by easy preparation of
focused libraries of natural-product-like compounds with high to excellent diaster-
eoselectivity. By using this radical cascade as a key reaction, we synthesized 33
indole alkaloids belonging to four families and four unnatural diastereoisomers
in 6–14 steps, which are difficult to achieve by traditional methods. Promisingly,
the radical cascade is green, water insensitive, and readily scalable (Figure S641
and Table S1). With the current success, we anticipate that generation of a nitro-
gen-centered radical from direct cleavage of an aniline N–H bond can be involved
in many intermolecular reaction protocols. Expansion of such ideas would provide
expedient access to more nitrogen-containing architectures, such as reserpine,
strychnine, and vincamajine, which are of great value. Application of this method
to the syntheses of other natural products and bioactive substances is in progress
and will be reported in due course.
EXPERIMENTAL PROCEDURES
Full experimental procedures are provided in the Supplemental Information.
814 Chem 2, 803–816, June 8, 2017

ACCESSION NUMBERS
The data for the X-ray crystallographic structures of detosylated derivatives of 3a
and 3l, 28a, 28b, 29a, and 29b in this article have been deposited in the Cambridge
Crystallographic Data Center under accession numbers CCDC: 1518109, 1520164,
1519507, 1519506, 1495501, and 1495500, respectively.
SUPPLEMENTAL INFORMATION
Supplemental Information includes Supplemental Experimental Procedures, 641
figures, 1 table, 1 scheme, and 6 data files and can be found with this article online
at http://dx.doi.org/10.1016/j.chempr.2017.04.007.
AUTHOR CONTRIBUTIONS
Y.Q. conceived and directed all aspects of the research and wrote the manuscript
with the assistance of X.-Y.L. The main experiments, which included methodology
development and syntheses of natural products, were performed by X.W., D.X.,
W.Q., R.Z., X.Z., and Q.Z. W.L., X.D., and H.W. carried out the scale-up experiments
of the radical cascade. S.W. and L.T. conducted the large-scale preparation of start-
ing materials. D.Z., H.S., and all other authors analyzed the results and commented
on the manuscript.
ACKNOWLEDGMENTS
We thank the National Natural Science Foundation of China (21132006
and 21572140) and the National Science and Technology Major Projects for ‘‘Major
New Drugs Innovation and Development’’ (2017ZX09101005 and 2017ZX09101003)
for financial support. Daibing Luo is acknowledged for X-ray crystallographic anal-
ysis. We thank Litao Yuan for recording the high-resolution mass spectra. We would
like to thank Prof. Weidong Li (Chongqing University) and Prof. Xiaoming Feng (Si-
chuan University) for helpful discussions.
Received: February 13, 2017
Revised: March 30, 2017
Accepted: April 19, 2017
Published: June 8, 2017
REFERENCES AND NOTES
1. Newman, D.J., and Cragg, G.M. (2016). Natural
products as sources of new drugs from 1981
to 2014. J. Nat. Prod. 79, 629–661.
Catharanthus. In The Alkaloids: Chemistry and
PharmacologyA. Brossi, ed. (Academic),
pp. 229–240.
11. Adams, G.L., and Smith, A.B., III (2016). The
chemistry of the akuammiline alkaloids. In
Alkaloids: Chemistry and Biology, H.J. Knolker,
¨
ed. (Elsevier), pp. 171–257.
2. Nicolaou, K.C. (2016). Catalyst: synthetic
organic chemistry as a force for good. Chem 1,
331–334.
7. Mashour, N.H., Lin, G.I., and Frishman, W.H.
(1998). Herbal medicine for the treatment of
cardiovascular disease: clinical considerations.
Arch. Int. Med. 158, 2225–2234.
12. Smith, J.M., Moreno, J., Boal, B.W., and Garg,
N.K. (2015). Cascade reactions: a driving force
in akuammiline alkaloid total synthesis. Angew.
Chem. Int. Ed. 54, 400–412.
3. Seiple, I.B., Zhang, Z., Jakubec, P., Langlois-
Mercier, A., Wright, P.M., Hog, D.T., Yabu, K.,
Allu, S.R., Fukuzaki, T., Carlsen, P.M., et al.
(2016). A platform for the discovery of new
macrolide antibiotics. Nature 533, 338–345.
8. O’Connor, S.E., and Maresh, J.J. (2006).
Chemistry and biology of monoterpene indole
alkaloid biosynthesis. Nat. Prod. Rep. 23,
532–547.
13. Mizoguchi, H., Oikawa, H., and Oguri, H. (2014).
Biogenetically inspired synthesis and skeletal
diversification of indole alkaloids. Nat. Chem.
6, 57–64.
4. Creasey, W.A. (1983). The monoterpenoid
indole alkaloids. In Chemistry of Heterocyclic
Compounds: IndolesJ.E. Saxton, ed. (Wiley),
pp. 783–829.
14. Lebold, T.P., Wood, J.L., Deitch, J., Lodewyk,
M.W., Tantillo, D.J., and Sarpong, R. (2013). A
divergent approach to the synthesis of the
yohimbinoid alkaloids venenatine and
9. Woodward, R.B., Bader, F.E., Bickel, H., Frey,
A.J., and Kierstead, R.W. (1956). The total
synthesis of reserpine. J. Am. Chem. Soc. 78,
2023–2025.
5. Dewick, P.M. (2009). Medicinal Natural
Products: A Biosynthetic Approach, Third
Edition (Wiley).
alstovenine. Nat. Chem. 5, 126–131.
10. Woodward, R.B., Cava, M.P., Ollis, W.D.,
Hunger, A., Daeniker, H.U., and Schenker, K.
(1954). The total synthesis of strychnine. J. Am.
Chem. Soc. 76, 4749–4751.
15. Lackner, G.L., Quasdorf, K.W., and Overman,
L.E. (2013). Direct construction of quaternary
carbons from tertiary alcohols via photoredox-
catalyzed fragmentation of tert-alkyl
6. Neuss, N., and Neuss, M.N. (1990).
Therapeutic use of bisindole alkaloids from
Chem 2, 803–816, June 8, 2017 815

N phthalimidoyl oxalates. J. Am. Chem. Soc.
135, 15342–15345.
free reductive dehalogenation reaction. J. Am.
Chem. Soc. 131, 8756–8757.
43. Stella, L. (2001). Nitrogen-centered radicals. In
Radicals in Organic Synthesis, P. Renaud and
M.P. Sibi, eds. (Wiley), pp. 407–426.
16. Jamison, C.R., and Overman, L.E. (2016).
Fragment coupling with tertiary radicals
generated by visible-light photocatalysis. Acc.
Chem. Res. 49, 1578–1586.
31. Dai, C., Narayanam, J.M.R., and Stephenson,
C.R.J. (2011). Visible-light-mediated
conversion of alcohols to halides. Nat. Chem. 3,
140–145.
44. Choi, G.J., and Knowles, R.R. (2015). Catalytic
alkene carboaminations enabled by oxidative
proton coupled electron transfer. J. Am. Chem.
Soc. 137, 9226–9229.
17. Lo, J.C., Gui, J., Yabe, Y., Pan, C.-M., and
Baran, P.S. (2014). Functionalized olefin
cross-coupling to construct carbon–carbon
bonds. Nature 516, 343–348.
32. Weiss, M.E., and Carreira, E.M. (2011). Total
synthesis of (+)-daphmanidin E. Angew. Chem.
Int. Ed. 50, 11501–11505.
45. Choi, G.J., Zhu, Q., Miller, D.C., Gu, C.J., and
Knowles, R.R. (2016). Catalytic alkylation of
remote C-H bonds enabled by proton couple
electron transfer. Nature 539, 268–271.
33. Furst, L., Narayanam, J.M.R., and Stephenson,
C.R.J. (2011). Total synthesis of (+)-gliocladin C
enabled by visible-light photoredox catalysis.
Angew. Chem. Int. Ed. 50, 9655–9659.
18. Yan, M., Lo, J.C., Edwards, J.T., and Baran, P.S.
(2016). Radicals: reactive intermediates with
translational potential. J. Am. Chem. Soc. 138,
12692–12714.
46. Chu, J.C.K., and Rovis, T. (2016). Amide-
directed photoredox-catalysed C–C bond
formation at unactivated sp3 C–H bonds.
Nature 539, 272–275.
34. Schnermann, M.J., and Overman, L.E. (2012). A
concise synthesis of (–)-aplyviolene facilitated
by a strategic tertiary radical conjugate
19. Studer, A., and Curran, D.P. (2016). Catalysis
of radical reaction: a radical chemistry
47. Hu, X.-Q., Qi, X., Chen, J.-R., Zhao, Q.-Q.,
Wei, Q., Lan, Y., and Xiao, W.-J. (2016).
Catalytic N-radical cascade reaction of
hydrazones by oxidative deprotonation
electron transfer and TEMPO mediation. Nat.
Commun. 7, 11188.
perspective. Angew. Chem. Int. Ed. 55, 58–102.
addition. Angew. Chem. Int. Ed. 51, 9576–9580.
20. Jones, S.B., Simmons, B., Mastracchio, A., and
MacMillan, D.W.C. (2011). Collective synthesis
of natural products by means of
35. Sun, Y., Li, R., Zhang, W., and Li, A. (2013). Total
synthesis of indotertine A and drimentines A, F,
and G. Angew. Chem. Int. Ed. 52, 9201–9204.
organocascade catalysis. Nature 475, 183–188.
48. Vatsadze, S.Z., Loginova, Y.D., dos Passos
Gomes, G., and Alabugin, I.V. (2017).
Stereoelectronic chameleons: the donor–
acceptor dichotomy of functional groups.
Chem. Eur. J. 23, 3225–3245.
36. Takayama, H., and Sakai, S.I. (1998).
Monoternenoid indole alkaloid syntheses
utilizing biomimetic reaction. In The Alkaloids:
Chemistry and BiologyG.A. Cordell, ed.
(Academic), pp. 415–452.
21. Brill, Z.G., Grover, H.K., and Maimone, T.J.
(2016). Enantioselective synthesis of
anophiobolin sesterterpene via a programmed
radical cascade. Science 352, 1078–1082.
49. Takano, S., Yonaga, M., Morimoto, M., and
Ogasawara, K. (1985). Chiral synthesis of
(+)-eburnamine, (–)-eburnarnenine, and
(–)-eburnamonine. J. Chem. Soc. Perkin Trans.
1, 305–309.
22. Asaba, T., Katoh, Y., Urabe, D., and Inoue, M.
(2015). Total synthesis of crotophorbolone.
Angew. Chem. Int. Ed. 54, 14457–14461.
37. Brandau, S., Landa, A., Franze´ n, J., Marigo, M.,
and Jørgensen, K.A. (2006). Organocatalytic
conjugate addition of malonates to
a,b-unsaturated aldehydes: asymmetric
formal synthesis of (–)-paroxetine, chiral
lactams, and lactones. Angew. Chem. Int. Ed.
45, 4305–4309.
23. Nagatomo, M., Koshimizu, M., Masuda, K.,
Tabuchi, T., Urabe, D., and Inoue, M. (2014).
Total synthesis of ryanodol. J. Am. Chem. Soc.
136, 5916–5919.
50. Desmaele, D., Mekouar, K., and d’Angelo, J.
¨
(1997). Stereocontrolled elaboration of
quaternary carbon centers through the
asymmetric Michael-type alkylation of
chiralimines/secondary enamines:
enantioselective synthesis of (+)-vincamine.
J. Org. Chem. 62, 3890–3901.
24. Narayanam, J.M.R., and Stephenson, C.R.J.
(2011). Visible light photoredox catalysis:
applications in organic synthesis. Chem. Soc.
Rev. 40, 102–113.
38. Stockigt, J., Antonchick, A.P., Wu, F., and
¨
Waldmann, H. (2011). The Pictet-Spengler
reaction in natural and in organic chemistry.
Angew. Chem. Int. Ed. 50, 8538–8564.
51. Das, S., Li, Y., Bornschein, C., Pisiewicz, S.,
Kiersch, K., Michalik, D., Gallou, F., Junge, K.,
and Beller, M. (2015). Selective rhodium-
catalyzed reduction of tertiary amides in amino
acid esters and peptides. Angew. Chem. Int.
Ed. 54, 12389–12393.
25. Prier, C.K., Rankic, D.A., and MacMillan, D.W.C.
(2013). Visible light photoredox catalysis with
transition metal complexes: applications in
organic synthesis. Chem. Rev. 113, 5322–5363.
39. Yu, J., Wang, T., Liu, X., Deschamps, J., Flippen-
Anderson, J., Liao, X., and Cook, J.M. (2003).
General approach for the synthesis of
sarpagine indole alkaloids. Enantiospecific total
synthesis of (+)-vellosimine, (+)-normacusine
B, (–)-alkaloid Q3, (–)-panarine, (+)-Na-
methylvellosimine, and (+)-Na-methyl-16-
epipericyclivine. J. Org. Chem. 68, 7565–7581.
26. Nicewicz, D.A., and MacMillan, D.W.C. (2008).
Merging photoredox catalysis with
52. Magnus, P., Payne, A.H., Waring, M.J., Scott,
D.A., and Lynch, V. (2000). Conversion of a,
b-unsaturated ketones into a-hydroxy ketones
using an MnIII catalyst, phenylsilane and
dioxygen: acceleration of conjugate hydride
reduction by dioxygen. Tetrahedron Lett. 41,
9725–9730.
organocatalysis: the direct asymmetric
alkylation of aldehydes. Science 322, 77–80.
40. Deiters, A., Chen, K., Eary, C.T., and Martin,
S.F. (2003). Biomimetic entry to the
sarpagan family of indole alkaloids: total
synthesis of (+)-geissoschizine and (+)-N-
methylvellosimine. J. Am. Chem. Soc. 125,
4541–4550.
27. Johnson, C.P., Smith, R.T., Allmendinger, S., and
MacMillan, D.W.C. (2016). Metallaphotoredox-
catalysed sp3–sp3 cross-coupling of carboxylic
acids with alkyl halides. Nature 536, 322–325.
53. Barrero, A.F., Quılez del Moral, J.F., Sa´ nchez,
´
28. Ischay, M.A., Anzovino, M.E., Du, J., and Yoon,
T.P. (2008). Efficient visible light photocatalysis
of [2+2] enone cycloadditions. J. Am. Chem.
Soc. 130, 12886–12887.
E.M., and Arteaga, J.F. (2006). Titanocene-
mediated radical cyclization: an emergent
method towards the synthesis of natural
products. Eur. J. Org. Chem. 2006, 1627–
1641.
41. Stavrinides, A., Tatsis, E.C., Caputi, L., Foureau,
E., Stevenson, C.E., Lawson, D.M.,
Courdavault, V., and O’Connor, S.E. (2016).
Structural investigation of heteroyohimbine
alkaloid synthesis reveals active site elements
that control stereoselectivity. Nat. Commun. 7,
12116.
29. Yoon, T.P., Ischay, M.A., and Du, J. (2010).
Visible light photocatalysis as a greener
approach to photochemical synthesis. Nat.
Chem. 2, 527–532.
54. Abue´ , J., Ghosh, S., and Tanol, M. (1994).
Symmetry-driven synthesis of indole alkaloids:
asymmetric total syntheses of (+)-yohimbine,
(–)-yohimbone, (–)-yohimbane, and
(+)-alloyohimbane. J. Am. Chem. Soc. 116,
9009–9018.
30. Narayanam, J.M.R., Tucker, J.W., and
Stephenson, C.R.J. (2009). Electron-transfer
photoredox catalysis: development of a tin-
42. Zard, S.Z. (2008). Recent progress in the
generation and use of nitrogen-centered
radicals. Chem. Soc. Rev. 37, 1603–1618.
816 Chem 2, 803–816, June 8, 2017