A three enzyme system to generate the Strychnos alkaloid scaffold from a central biosynthetic intermediate

Article abstract of DOI:10.1038/s41467-017-00154-x

Monoterpene indole alkaloids comprise a diverse family of over 2000 plant-produced natural products. This pathway provides an outstanding example of how nature creates chemical diversity from a single precursor, in this case from the intermediate strictosidine. The enzymes that elicit these seemingly disparate products from strictosidine have hitherto been elusive. Here we show that the concerted action of two enzymes commonly involved in natural product metabolism - A n alcohol dehydrogenase and a cytochrome P450 - produces unexpected rearrangements in strictosidine when assayed simultaneously. The tetrahydro-β-carboline of strictosidine aglycone is converted into akuammicine, a Strychnos alkaloid, an elusive biosynthetic transformation that has been investigated for decades. Importantly, akuammicine arises from deformylation of preakuammicine, which is the central biosynthetic precursor for the anti-cancer agents vinblastine and vincristine, as well as other biologically active compounds. This discovery of how these enzymes can function in combination opens a gateway into a rich family of natural products.

Full text of DOI:10.1038/s41467-017-00154-x

A R T I C L E
DOI: 10.1038/s41467-017-00154-x
OPEN
A t h re e e n zy m e s y st e m t o g e n e r a t e t he Strychnos
a lk al o i d s c a ff o l d f r o m a c e n t r a l b i o s y n th e t i c
i nt e rm e d i at e
1
2
2
1
E v a n g e l o s C . T a t s i s , I n ê s C a r qu e i j e i r o
, T h o m a s D u g é d e B e rn o n vi l l e
, J a k o b F r a n k e ,
1
2
2
2
1
2
T
h
u
-
T
h
u
y
T
.
D
a
n
g
,
A
u
d
r
e
y
O
u
d
i
n
,
A
r
n
a
u
d
L
a
n
o
u
e
,
F
l
o
r
e
n
t
L
a
f
o
n
t
a
i
n
e
,
A
n
n
a
K
.
S
t
a
v
r
i
n
i
d
e
s
,
M
a
r
c
C
l
a
s
t
r
e
,
2
1
V
i
n
c
e
n
t
C
o
u
r
d
a
v
a
u
l
t
&
S
a
r
a
h
E
.
O
’
C
o
n
n
o
r
M o no t e r pe n e i nd o l e a l k a l o i d s c o m p r i s e a d i v e r s e f a m i ly o f o ve r 2 0 00 p l a n t- p r od u c e d n a t ur a l
p r od u c t s . T h i s p at h w a y p r o v i d e s a n o u ts t a n d i n g e x a m pl e o f h ow n a t ur e c r ea te s c h e m i c a l
d i v e r si t y f r o m a s in g l e p r e cu r s o r , i n t h i s c a s e f r o m t h e i n t e r m e di a t e s t r i c t o s i d i n e . T h e
e n z y m e s t h a t e l ic i t t h e se s e e m i n g l y d i s p a r a t e p r o d u c ts f ro m s t r i c t o s i d i n e h a v e h i t he r to b ee n
e l u s i ve . H e r e w e s h o w t h a t t h e c o n c e r t e d a c ti o n o f t wo e nz y m es c o m m o nl y i n vo l v e d i n
n a t u r a l p ro d u c t m e t ab o l i s m —a n a l co h o l d e h y d r o g e n a s e a n d a c y t o c h r om e P 4 5 0 —p ro d u c e s
u n ex p e c t e d r e ar ra n g e m e n t s i n s tr i c to s i d i n e w h e n a s s a y ed s i m u l t a n e o u s l y . T he t e t r a h y dr o - β-
c a rb o l i ne o f s t r ic t o s i d i n e a g l y co n e i s c o n v e r t e d i n to a k u a m m i c i ne , a Strychnos a l k a l o i d , a n
e l u s i ve b io s y nt he t ic t r a n sf o r m a t i o n t h a t h a s b e e n i n ve s t i g a t e d f or d e c a de s . I m p or t a nt l y ,
a k ua m mi ci ne a ri s e s f r o m d e f o r m y l a t i o n o f p r e a k u a m m i c i ne , w hi c h i s t h e c e n tr a l b i o s y n t he t i c
p r e c u rs or f o r t h e a nt i - c a n c e r a g e n ts v i n b l a s t i n e a n d v i n c ri s t i n e, a s w e l l a s o t he r b i o l og i c a l l y
a c t i v e c o m po un d s . T h i s d i s co v e r y o f h o w t h e s e e n z y m e s c a n f un c t i o n i n c o m b i n a t i o n o pe n s a
g a t e wa y i nt o a r ic h f a m i l y o f n a tu r a l p r o d u ct s .
1 John Innes Centre, Department of Biological Chemistry, Norwich Research Park, Norwich NR4 7UH, UK. 2 Université François-Rabelais de Tours,
EA2106 Biomolécules et Biotechnologies Végétales, Parc de Grandmont 37200 Tours, France. Evangelos C. Tatsis and Inês Carqueijeiro contributed equally
to this work. Correspondence and requests for materials should be addressed to V.C. (email: vincent.courdavault@univ-tours.fr) or to
S.E.O. (email: sarah.oconnor@jic.ac.uk)
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
8
:
3
1
6
D
O
I
:
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
7
-
0
0
1
5
4
-
x
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
re
c
o
m
m
u
n
ic
a
t
i
o
n
s
1
|
|
|

A R T I C L E
N A T U R E C O M MU N ICA T IO N S | D O I: 1 0 . 1 0 3 8 / s 4 1 4 67 - 0 17 - 0 0 15 4 -x
he medicinal plant Catharanthus roseus synthesizes a on 4,21-dehydrogeissoschizine (strictosidine aglycone) to yield
variety of biologically active monoterpene indole alkaloids, akuammicine, a Strychnos alkaloid. These enzymes are char-
most notably the anti-cancer agents vinblastine and acterized by both in vitro and in planta silencing approaches.
T
vincristine. Vinblastine and vincristine, along with many other Akuammicine is known to arise from deformylation of
10–12
monoterpene indole alkaloids, are derived from the biosynthetic preakuammicine8,
, which is the central biosynthetic
intermediate, preakuammicine, as evidenced by extensive feeding precursor for the anti-cancer agents vinblastine and vincristine, as
studies and isolation of biosynthetic intermediates from plant well as hundreds of other biologically active compounds. There-
material1, 2. However, discovery of the enzymes that are fore, these enzymes open a gateway into a rich family of natural
responsible for biosynthesis of the preakuammicine skeleton has products, and will facilitate metabolic engineering efforts to make
proven elusive.
high-value monoterpene indole alkaloids.
Nearly all monoterpene indole alkaloid pathways begin with
deglycosylation of strictosidine by a dedicated glucosidase, stric-
tosidine glucosidase (SGD)3. Strictosidine aglycone, which can
exist in numerous isomers such as 4,21-dehydrogeissoschizine, is
then enzymatically transformed into the various monoterpene
indole alkaloid skeletons. However, the enzymes responsible for
the synthesis of preakuammicine, or any preakuammicine-
derived product from strictosidine aglycone remain unknown.
While the enzymatic reactions remain cryptic, extensive isolation
of reaction intermediates as well as model chemical studies
suggest a sequence of reactions: 4,21-dehydrogeissoschizine
(strictosidine aglycone isomer) is reduced to geissoschizine, which
is followed by oxidative rearrangement and reduction to pre-
Results
Identification of gene candidates. To identify genes responsible
for generating Strychnos scaffolds such as preakuammicine and
akuammicine (Fig. 1a), we mined a transcriptomic database of
C. roseus tissues13, 14, which also included data from plants with
increased alkaloid levels due to folivory15. Gene candidates iden-
tified were initially screened in C. roseus by virus-induced gene
silencing (VIGS), a transient silencing system in which qualitative
perturbations in the metabolic profile between silenced and control
tissue provide clues to the metabolic function of the gene of
interest. Transformation into preakuammicine requires both
akuammicine4–10
.
Preakuammicine can non-enzymatically
reduction and oxidation steps (Fig. 1b, Supplementary Fig. 1)4–10
.
10–12
deformylate to produce the alkaloid akuammicine8,
serve as a precursor for additional alkaloid scaffolds8.
or
Therefore, we focused on alcohol dehydrogenases (reductase) and
cytochromes P450 (oxidase); alcohol dehydrogenases have been
implicated in other monoterpene indole alkaloid biosynthetic
steps16–18, and cytochromes P450 are well known to play impor-
Here we use a transcriptomic database for the medicinal plant
Catharanthus roseus to identify two enzymes, an alcohol dehy-
drogenase (GS) and a cytochrome P450 (GO), that act in tandem
tant roles in oxidative transformations in plant metabolism16, 19
.
a
N
H
N
H
H
N
O
N
O
H
N
H
NH
CH3O2C
OH
O-β-Glc
H
N
H
Vincamine
H
Strychnine
H
OH
O
N
H3CO2C
CO2CH3
N
N
Preakuammicine
N
Strictosidine
OH
H
OAc
CO2CH3
N
H
H
H
CO2CH3
MeO
N
H
CO2CH3
N
OH
H
R
R: CH3 vinblastine
R: CHO vincristine
Akuammicine
b
N
N
N
SGD
H
H
N
N
H
H
Strictosidine
Akuammicine
H
H
OH
N
H3CO2C
H3CO2C
CO2CH3
OH
OH
4,21-dehydrogeissoschizine
(strictosidine aglycone)
Geissoschizine
Preakuammicine
Fig. 1
M
o
n
m
o
i
t
c
e
r
p
e
e
t
n
h
e
r
i
o
n
u
d
g
o
h
l
e
a
a
n
l
u
k
n
a
k
l
o
n
i
o
d
w
b
n
i
o
s
s
e
y
r
n
t
h
s
e
o
s
f
i
s
e
. a
A
l
l ca
.
2
0
a
0
c
0
m
o
.
n
P
o
r
t
e
e
a
r
k
p
u
e
a
n
m
e
i
m
n
i
d
c
o
l
e
e
a
l
k
h
a
y
l
o
p
i
o
d
t
s
h
a
e
r
s
e
d
e
r
t
i
v
o
e
b
d
e
f
r
t
o
h
m
e
s
t
e
r
c
i
c
u
t
r
o
s
s
o
i
r
d
i
f
n
o
e
r
,
w
h
n
i
c
y
h
s
i
t
s
r
c
o
t
n
u
v
e
l
r
t
e
d
d
i
t
o
r
p
re
a
k
u
a
m
i
n
i
e
n
z
y
m
a
t
i
c
r
e
t
i
o
n
s
i
n
i
s
i
z
e
d
p
r
m
a
u
c
r
a
l
y
v
e
g
e
n
t
a
e
l
n
k
z
a
y
l
o
m
i
d
e
s
,
. b
A
s
c
s
h
e
i
m
a
t
g
i
c
o
c
v
o
e
s
r
i
v
d
i
a
e
s
w
e
o
(
f
S
t
G
h
D
e
p
.
r
A
o
p
n
o
i
s
s
e
o
d
m
p
e
a
r
t
o
h
f
w
s
a
t
y
t
c
h
a
s
t
g
e
n
e
e
r
a
a
g
t
l
e
y
s
c
t
o
h
n
e Strychnos
a
y
l
d
k
r
a
o
l
o
g
i
e
d
a
s
k
o
u
s
a
c
m
m
i
c
e
i
,
n
e
c
.
a
S
n
t
r
b
i
e
c
t
r
o
e
s
d
i
d
u
i
c
n
e
e
d
i
s
t
d
o
e
f
g
o
l
r
y
m
c
o
g
s
e
y
i
l
s
a
s
t
e
o
d
b
h
y
t
i
h
n
e
k
w
n
o
h
w
n
h
s
t
r
i
c
t
o
i
d
n
e
l
u
)
r
i
t
o
i
d
i
n
e
,
4
,
2
1
-
d
e
h
i
s
h
i
z
i
n
s
c
i
z
e
,
ic
f
h
e
y
e
p
d
o
i
n
g
e
s
s
t
u
e
d
d
i
e
t
s
o
s
b
u
e
g
g
t
e
h
s
e
t
r
e
e
a
c
r
u
r
a
r
n
g
r
e
s
t
r
o
f
m
o
a
r
n
m
y
p
d
r
o
e
w
a
n
k
s
u
t
a
r
m
m
m
i
c
a
i
n
e
a
.
P
r
d
e
s
a
.
k
S
u
e
a
e
m
S
m
u
i
p
c
p
i
n
e
m
c
a
e
n
n
d
e
y
f
o
F
r
i
m
y
1
l
a
f
t
o
e
r
t
a
o
f
m
o
o
r
m
t
d
h
e
e
S
t
r
y
c
h
n
o
s
a
l
k
a
o
l
f
o
m
i
d
o
a
n
k
o
u
t
a
e
m
m
i
c
e
i
n
e
d
,
a
o
n
d
i
s
k
a
l
o
s
i
o
d
t
h
i
z
p
r
s
o
f
o
e
a
l
k
l
o
i
l
e
t
a
r
g
.
r
e
t
a
i
l
e
d
p
i
c
t
u
r
e
r
p
e
n
i
n
l
e
a
l
a
l
b
i
o
s
y
n
t
h
e
s
i
s
2
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
8
:
3
1
6
D
O
I
:
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
7
-
0
0
1
5
4
-
x
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
|
|
|

N
A
T
U
R
E
C
O
M
M
U
N
I
CA
T
I
O
NS
|
D
OI
:
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
7
-
0
0
1
5
4
-
x
A
R
T
I
C
L
E
In planta and in vitro assay of enzyme candidates. Prioritization decreases in the production of the alkaloids catharanthine,
of genes through Pearson’s correlation coefficient (PCC) deter- vindoline and vindorosine (Supplementary Fig. 2a), iboga and
mination revealed a limited number of candidates displaying aspidosperma type alkaloids that are derived from pre-
correlation with SGD (549 with PCC > 0.6), with only two can- akuammicine (Supplementary Fig. 1)23.
didates predicted to encode cytochromes P450 (Supplementary
All literature reports suggest geissoschizine (m/z 353), a
Data 1). One was secologanin synthase, which is involved in reduction product of the 4,21-dehydrogeissoschizine isomer of
upstream monoterpene indole alkaloid biosynthesis20, 21 and the strictosidine aglycone, is the precursor for preakuammicine
second was CYP71D1V1, which was previously deposited as (Fig. 1b)4, 5, 10. As part of an effort to characterize alcohol
JN613015 in a previous report, though not functionally char- dehydrogenases of C. roseus, we noted that silencing of one
acterized22. Silencing CYP71D1V1 led to a statistically significant medium chain alcohol dehydrogenase gene (named GS1 for
a
Strictosidine
4,21-dehydrogeissoschizine
(E )-geissoschizine
Preakuammicine
21
6
5
N
5
7
20
N
N
NH
3
2
14
3
6
20
O-β-Glc
H
N
H
SGD
GS
GO
H
N
N
15
H
17 OH
H
H
H
7
H
16
15
O
16
2
N
17
H3CO2C
H3CO2C
H3CO2C
CO2CH3
OH
OH
GS
GS
HCHO
spont.
N
N
N
H
H
N
H
N
H
H
H
N
H
CO2CH3
H3CO2C
H3CO2C
H
H
OH
OH
(16R)-E-isositsirikine
(16R)-Z-isositsirikine
Akuammicine
b
c
Akuammicine
(16R)-Z-isositsirikine
(16R)-E-isositsirikine
GS1 + CYP71D1V1 (GO)
GS1
GS1
GS1 + CYP71D1V1 (GO)
GS2 + CYP71D1V1 (GO)
GS2
THAS1 + CYP71D1V1 (GO)
GS2
GS2 + CYP71D1V1 (GO)
GS1 + microsomes EV
1.00
1.00
2.00
3.00
4.00
2.00
3.00
4.00
Time
Time
Fig. 2
E
S
e
a
n
s
y
z
h
y
o
a
m
w
a
i
(
o
t
n
i
c
g
6R
u
a
c
c
o
)
s
t
m
- E
i
v
p
-
o
i
t
o
s
y
u
o
G
o
n
f
d
t
1
G
s
S
a
1
a
n
d
G
3
n
O
S
5
d
.
2
5
(
a
i
n
n
d
e
)
G
O
y
(
m
C
a
s
Y
P
7
a
1
s
i
i
D
1
y
n
V
s
e
1
)
o
S
h
.
a
R
S
e
1
l
a
a
/
e
c
2
m
o
t
i
a
e
h
o
n
n
o
n
d
t
l
o
G
f
s
O
t
r
.
i
i
I
g
c
s
s
r
t
o
o
l
5–
g
s
a
i
t
d
i
n
n
)
s
e
a
. c
e
w
i
t
h
N
R
o
S
M
M
G
R
D
c
( m/z
d
,
G
S
r
a
n
t
2
d
G
O
.
b
M
u
f
l
t
t
h
a
b
i
p
e
m
l
e
t
m
r
w
i
e
o
c
m
a
c
m
e
n
t
i
a
o
n
r
m
o
m
a
n
i
t
p
o
o
c
r
u
i
n
n
m
o
g
d
(
s
p
M
R
d
M
)
t
o
e
i t
d
n
d
h
a
L
C
M
t m/z
k
a
n
z
t
i
c
s
a
f
G
i
o
n
d
h
a
a
c
e
3
h
r
i
z
a
t
i
o
n
o
j
o
c
o
i
n
ic
a
t
h
a
t
t
h
r
e
1
i
f
s
i
s
i
r
i
i
n
e
d
a
G
1
6
R
-
Z
-
i
s
o
i
t
s
e
i
r
k
i
(
c
u
p
p
a
r
y
e
F
y
1
6
M
3
)
o
f
a
n
a
k
u
i
n
s
t
t
a
n
d
r
e
d
o
d
a
r
e
d
w
e
n
z
y
m
t
i
c
p
r
d
c
t
S
/
2
n
T
H
A
S
1
i
s
a
m
d
u
m
a
i
n
l
c
d
h
d
o
e
n
a
i
n
v
l
v
e
i
n
t
h
e
e
t
e
r
o
y
o
h
i
m
i
n
e
o
o
e
r
p
e
n
i
n
l
e
a
l
k
a
l
o
i
p
a
t
h
w
a
y
(
S
u
p
p
l
e
m
e
n
t
a
r
y
F
i
g
.
1
)
.
E
V
e
m
p
t
y
v
e
c
t
o
r
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
8
:
3
1
6
D
O
I
:
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
7
-
0
0
1
5
4
-
x
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
3
|
|
|

A R T I C L E
N A T U R E C O M MU N ICA T IO N S | D O I: 1 0 . 1 0 3 8 / s 4 1 4 67 - 0 17 - 0 0 15 4 -x
a
COOCH3
1H-NMR of akuammicine
H20
H15
H13
H14
H12
H6b
H8b
H9b
H6a
H8a
H17
H4
H9a
H18a
H19
H18b
1H-NMR of enzyme reaction product
6
CHCl3
CH3CN
8
N
19
12
17
18
4
9
13
14
20
H
H
N
H
2
15
CO2CH3
7.0
6.5
6.0
5.5
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5 p.p.m.
b
c
Synthetic akuammicine
Enzymatic akuammicine
1.00 2.00
3.00
4.00
1.00 2.00 3.00 4.00 5.00 6.00 7.00 8.00 9.00
Time
Time
1
Fig. 3
C
h
a
S
r
u
a
p
c
t
p
e
r
i
m
z
a
e
t
n
i
o
t
n
o
y
f
F
t
i
h
g
e
s
G
S
1
2
/
3
G
)
O
. b
p
.
r
L
o
C
d
-
u
M
c
t
S
,
a
c
k
h
u
a
m
m
a
m
i
c
g
i
n
e
m
. a
s
.
H
N
M
s
R
t
o
f
d
e
a
n
z
y
r
m
a
e
t
r
i
c
s
a
e
l
l
p
y
p
r
s
o
e
d
u
c
e
d
m
a
n
k
u
c
a
o
m
m
m
i
c
i
n
n
g
e
c
o
n
m
p
e
a
t
r
i
e
c
d
a
t
k
o
u
a
s
m
y
n
t
h
e
e
t
i
c
w
s
t
h
a
n
t
d
h
a
e
r
d
(
s
e
e
a
l
s
o
l
e
a
r
1
9–
r
o
to
r
a
w
i
t
h
a
a
n
r
d
e
v
h
a
c
o
l
u
p
a
r
i
s
y
t
h
a
m
i
c
i
n
i
t
e
n
z
y
m
a
t
i
c
p
r
o
d
u
c
t
.
c
C
h
i
r
a
l
L
C
-
M
S
c
h
r
o
m
a
t
o
g
r
a
m
s
c
o
m
p
a
r
i
n
g
t
h
e
r
a
c
e
m
i
c
s
y
n
t
h
e
t
i
c
s
t
a
n
d
a
r
d
w
i
t
h
t
h
e
e
n
z
y
m
a
t
i
c
p
r
o
d
u
c
t
geissoschizine synthase), led to only minor changes in alkaloid led us to dismiss GS1/GS2 as candidates for geissoschizine
profiles (Supplementary Fig. 3). Since VIGS phenotypes can be synthase, we noted that in CYP71D1V1 silencing experiments, a
attenuated by factors such as enzyme redundancy or molecule compound with the same m/z and retention time as (16 R)-E-
transport, we nevertheless assayed the protein in vitro. When the isositsirikine accumulated (Supplementary Fig. 2b, Supplemen-
protein encoded by this gene (initially reported but not tary Fig. 17). This provided an important clue that the enzymatic
functionally characterized during a search for strictosidine reactions of GS1 and CYP71D1V1 are coupled. We hypothesized
biosynthetic genes as KF302079.116) was heterologously that GS1/GS2 initially produces geissoschizine, which is captured
expressed in E. coli (Supplementary Fig. 4) and assayed with and then oxidized to preakuammicine when the correct down-
strictosidine in the presence of SGD, the enzymatic reaction stream oxidase is present. However, in the absence of a
yielded a mixture of products, primarily consisting of two downstream enzyme, geissoschizine could be over-reduced by
compounds (m/z 355) (Fig. 2b). Isolation and NMR character- GS1/GS2 to the more stable (16 R)-Z, E-isositsirikine products.
ization showed that the two major products of this enzymatic Notably, the E-isositsirikine isomer, which is the major enzymatic
reaction are (16 R)-E-isositsirikine (high-resolution MS [M + H]+ product and also the isomer that accumulates in CYP71D1V1-
355.2020; isolated in ca. 38% yield) and (16 R)-Z-isositsirikine silenced tissues, has the same C19–20 bond geometry as
(high-resolution MS [M + H]+ 355.2014; isolated in ca. 7% yield) geissoschizine9.
(Fig. 1a, Supplementary Figs 5 and 16)24, 25. In addition, a
To test this hypothesis, CYP71D1V1, a microsomal enzyme,
homolog of GS1 (GS2, KF302078.116) was also identified and was heterologously expressed in a yeast strain harboring a
tested. This enzyme was purified in lower yields from E. coli cytochrome P450 reductase, and the microsomes were harvested.
(Supplementary Fig. 4) but produced the same products as GS1 in When SGD, CYP71D1V1 and GS1 were incubated together with
decreased quantity. While the formation of isositsirikine initially strictosidine, using selected ion monitoring at m/z 355 we noted
4
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
8
:
3
1
6
D
O
I
:
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
7
-
0
0
1
5
4
-
x
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
|
|
|

N
A
T
UR
E
C
O
M
M
UN
I
C
A
T
I
O
NS
|
D
O
I
:
1
0.
1
0
3
8
/
s
4
1
46
7
-
01
7
-
001
5
4
-
x
A
R
T
I
C
L
E
a
Formaldehyde release
Akuammicine production
35
30
25
20
15
10
5
35
30
25
20
15
10
5
0
0
0
20
40
60
80
100
120
0
20
40
60
80
100
120
Microsomal protein amount (μl)
Microsomal protein amount (μl)
b
H3CO2C
H3CO2C
H
N
N
H
H
H
[O]
X
HO
H
H
N
H
N
H
CO2CH3
H3CO2C
X
N
N
CO2CH3
H
H
X
NH
N
16-R-dehydro-
Akuammicine
preakuammicine X = O
Rhazimal X = O
Geissoschizine
N
[H-]
[H-]
N
CH2O
Rhazimol X = OH, H
16R-preakuammicine X = OH, H
Fig. 4
D
e
d
f
o
b
r
y
m
m
y
l
a
a
s
t
i
s
k
o
n
i
e
n
c
o
t
t
n
h
e
m
s
f
o
e
s
r
t
u
m
a
(
e
t
s
i
o
e
t
n
e
e
o
a
d
f
a
o
y
k
S
B
u
u
e
a
m
m
i
m
d et al 1 0
c
i
n
e
e
n
.
a
D
e
i
E
f
g
o
s
r
m
y
l
a
2
a
t
9
r
i
o
)
n
. b
a
s
P
r
m
e
s
n
a
s
l
u
e
s
r
e
m
d
e
d
w
i
t
a
d
h
n
N
a
s
h
s
o
r
c
f
e
e
t
a
n
h
g
a
r
e
n
o
e
t
c
o
r
p
r
r
l
r
e
e
a
l
a
t
e
s
w
i
t
h
a
k
u
a
m
m
i
c
i
n
e
p
r
o
d
u
c
t
i
o
n
a
s
m
e
a
s
u
r
e
s
p
r
o
r
y
l
s
p
p
l
e
t
a
r
y
F
2
7
–
o
s
i
b
c
h
i
s
t
i
c
r
i
f
o
a
r
r
a
n
g
e
m
e
n
t
o
f
g
e
i
s
s
o
s
c
h
i
z
i
n
e
t
o
t
h
e
p
r
e
a
k
u
a
m
m
i
c
i
n
e
s
e
l
e
t
a
g
g
s
b
n
a
y
a
.
.
r
r
o
r
b
s
r
e
p
e
s
e
t
t
a
n
a
r
e
r
r
o
r
e
r
e
i
c
t
e
s
the same product profile observed with GS1 lacking CYP71D1V1, level, YFP-labeling of GO and GS1 suggests that GS1 has a
but the amounts of the products appeared to be decreased nucleocytosolic location while GO is anchored to the ER,
(Fig. 2b). However, no new product corresponding to pre- potentially facilitating GS1-reaction product uptake (Supplemen-
akuammicine (m/z 353), dehydropreakuammicine (m/z 351) or tary Fig. 26b). Collectively, these data suggest that GS1 and GO
any potential isomers of these compounds, was observed. are located in close proximity in planta, consistent with a coupled
Notably, preakuammicine can non-enzymatically deformylate to role in vivo.
yield akuammicine (m/z 323)8, and is generally regarded as the
direct precursor to akuammicine10. Careful analysis of the mass
spectra of the GS1 (or GS2) and CYP71D1V1 reaction revealed a
compound with m/z 323 (high-resolution MS [M + H]+ 323.1760)
Mechanism of akuammicine formation. The mechanistic basis
(Fig. 2c). This enzymatically produced compound was isolated
(ca. 12.5% yield), characterized by NMR (Supplementary Figs 18–23)
and shown to be identical to a synthetic standard of akuammicine
(Fig. 3a, b)26. Moreover, the enzymatic akuammicine product was
observed to be enantiopure on a chiral column (Fig. 3c).
Formation of akuammicine was absolutely dependent on the
presence of CYP71D1V1-expressing microsomes (Fig. 2c). On
this basis, CYP71D1V1 was named geissoschizine oxidase (GO).
In addition, replacement of GS1 or GS2 with another medium
chain alcohol dehydrogenase, THAS1, which is involved in an
unrelated branch (heteroyohimbine) of monoterpene indole
alkaloid biosynthesis18 failed to yield akuammicine (Fig. 2c).
GS1 and GO must be added to strictosidine aglycone
simultaneously for akuammicine to be produced (Supplementary
Fig. 24), suggesting that the GS1 reaction product that acts as the
substrate for GO (i.e., geissoschizine) is rapidly metabolized.
A small amount of compound with a mass consistent with
geissoschizine (m/z 353) was observed in the GS1 enzymatic
reaction, but this compound could not be isolated in sufficient
quantities for characterization (Supplementary Fig. 25). In vivo,
both GS1 and GO transcripts are enriched in the epidermis
relative to whole leaf, consistent with the epidermal location of
other monoterpene indole alkaloid enzymes such as strictosidine
synthase and SGD (Supplementary Fig. 26a)27. At the subcellular
of the rearrangement of the tetrahydro-β-carboline scaffold of
strictosidine to the Strychnos scaffold remains to established, but
the literature provides some clues. Akuammicine can be gener-
ated from preakuammicine by non-enzymatic deformylation
(Fig. 4a)8, 10. Production of formaldehyde during the enzyme
assay in approximately equimolar concentration as akuammicine
provides indirect support for the formation of preakuammicine
(Fig. 4a, Supplementary Figs 27–29). Geissoschizine, which
has been shown by feeding studies to be the akuammicine
precursor, is proposed to undergo oxidative reaction to form
rhazimal. Recent elegant model chemistry suggests that rhazimal,
or its reduced counterpart, rhazimol, can rearrange to dehy-
dropreakuammicine or preakuammicine, respectively (Fig. 4b)10.
Therefore, we hypothesize that a reductase generates geissoschi-
zine from strictosidine aglycone (4,21-dehydrogeissoschizine
isomer). Then, an oxidase transforms geissoschizine to rhazimal,
which then rearranges to dehydropreakuammicine, which is
then reduced to preakuammicine. Alternatively, rhazimal
could be reduced to rhazimol, which rearranges directly to
preakuammicine. However, monoterpene indole alkaloid bio-
synthesis in strychnine biosynthesis likely proceeds via dehy-
dropreakuammicine28. GS1/GS2 or an endogenous reductase
present in the yeast microsomes could catalyze the reduction of
rhazimal or dehydropreakuammicine.
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
8
:
3
1
6
D
O
I
:
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
7
-
0
0
1
5
4
-
x
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
5
|
|
|

A R T I C L E
N A T U R E C O M MU N ICA T IO N S | D O I: 1 0 . 1 0 3 8 / s 4 1 4 67 - 0 17 - 0 0 15 4 -x
Discussion
ATGGAAATGGTCAGAAACAAGTGGGGTTTCACTAGATACCCTTATGTTT
TTGGACATGAGACCGCCGGTGAGGTGGTAGAAGTTGGGAAGAAAGTAG
AGAAATTCAAGGTTGGAGATAAGGTGGGCGTGGGATGTATGGTCGGAT
CTTGTGGCAAATGTTTCCATTGTCAAAACGAAATGGAGAATTACTGCC
CGGAGCCTAATTTGGCTGATGGATCTACTTACCGTGAAGAAGGAGAAC
GTTCCTATGGAGGTTGTTCAAATGTCATGGTTGTTGATGAAAAATTCG
TCCTTAGATGGCCCGAAAATTTGCCTCAAGATAAAGGAGTTCCTCTCCT
CTGTGCTGGGGTTGTTGTTTATAGCCCCATGAAATATATGGGATTTGAT
All monoterpene indole alkaloids emerge from rearrangement of
the same precursor, strictosidine1. A number of these alkaloids
(e.g., vinblastine, strychnine, quinine, camptothecin) are used
clinically1, 2. The different mechanisms by which strictosidine
rearranges to form these structurally diverse compounds have
been studied extensively, though only enzymes that catalyze the
simplest rearrangement have been reported to date17, 18. One of
the most important—and cryptic—rearrangements of strictosi-
dine is its conversion into the biosynthetic intermediate
preakuammicine, the precursor for the Strychnos alkaloids, which
include neurotoxic agents such as strychnine, as well as the
precursor for the aspidosperma and iboga alkaloids that include
the anti-cancer agent vinblastine (Supplementary Fig. 1)1.
While recent synthetic efforts have demonstrated how the
monoterpene indole alkaloids can be chemically synthesized from
a common scaffold29, the enzymes that control rearrangement
AAGCCAGGAAAGCATATTGGGGTTTTTGGGTTGGGTGGTCTTGGTTCCA
TTGCTGTTAAGTTTATTAAAGCTTTTGGTGGTAAGGCTACTGTTATTAGT
ACATCAAGGCGTAAAGAGAAGGAAGCCATTGAAGAGCATGGAGCTGA
TGCTTTTGTTGTCAACACTGACTCTGAACAATTGAAGGCTCTGGAAGGT
ACTATGGATGGTGTTGTGGACACCACCCCAGGTGGCCGCACTCCTAT
GCCACTTATGCTCAATTTGGTTAAGTTTGACGGCGCCGTTATACTCGTC
GGTGCACCGGAGACGCTATTTGAGCTCCCGTTGGAGCCTATGGGAAGGA
AAAAGATAATCGGAAGCTCCACTGGAGGTCTCAAAGAGTATCAAGAAGT
GCTTGATATTGCAGCCAAACACAACATTGTATGTGATACTGAGGTTAT
TGGGATTGATTATCTCAGCACTGCTATGGAACGCATCAAGAATTTGGAT
GTCAAGTACCGATTCGCGATCGACATTGGGAATACATTGAAATTTGAGG
AATAAAGCTTTCTAGACCAT.-3′
The CYP71D1V1 coding sequence was amplified using primers 5′-CTGAGAA
and chemical diversification of the central intermediate found in CTAGTATGGAGTTTTCTTTCTCCTCACCAG-3′ and 5′-CTGAGAACTAGTCT
AATCGTTAACAAGATGAGGAACCAATT-3′ and cloned into pESC-HIS
expression vector (Agilent Technologies) prior to protein expression in yeast.
the biological system have largely remained cryptic. In this
manuscript, we report the discovery of two enzymes, an alcohol
dehydrogenase (GS) and a cytochrome P450 (GO), that convert
strictosidine aglycone into the Strychnos alkaloid akuammicine, a
κ-opioid receptor agonist11, 12. Together, these three enzymes
catalyze a series of tandem reactions that lead to the remarkable
rearrangement of the tetrahydro-β-carboline strictosidine sub-
strate into the Strychnos-type scaffold. VIGS data are supportive
of a role in planta, though the weak phenotype of GS suggests that
there may be redundancy with this enzyme.
Though extensive model chemistry reported in the literature
has provided some mechanistic insights into this enzymatic
transformation, aspects of the mechanism of this transformation
remain cryptic. Regardless, the discovery of GO and GS1 provides
considerable insight into the biocatalytic processes involved in the
diversification of monoterpene indole alkaloid structures, and
provides an important step forward for the complete elucidation
of metabolic pathways leading to biologically active molecules in
this family, such as vinblastine. Importantly, this work has
highlighted that to unlock the chemical diversity of certain plant
pathways, the instability of the reaction intermediates requires
that enzymes must be tested in combination. Therefore, methods
to rapidly screen enzyme combinations are likely to be required to
identify the genes of many plant natural product pathways. In
summary, the discovery of these enzymes demonstrate the
extraordinary chemistry that takes place in plants, and this dis-
covery will enable metabolic engineering efforts to overproduce
these high-value molecules.
Protein expression. GS1 and GS2 enzymes were expressed in SoluBL21 (DE3)
E. coli cells (Genlantis) grown in LB medium. A starter culture was grown over-
night at 37˚C in 100 mL of 2xYT media supplemented with carbenicillin (100 μg/ml).
A 1:100 dilution in fresh 2xYT (2 × 1 l) media supplemented with antibiotics was
prepared and allowed to grow at 37˚C to an OD600 of 0.8. Protein production was
induced by addition of 0.5 mM IPTG and the cultures were shaken at 18˚C for 16 h.
Cells were collected by centrifugation, lysed by sonication in Buffer A (50 mM Tris-
HCl pH 8, 50 mM glycine, 500 mM NaCl, 5% v/v glycerol, 20 mM imidazole)
supplemented with EDTA-free protease inhibitor (Roche Diagnostics Ltd.) and
0.2 mg/ml lysozyme.
Cells were lysed using sonication for 3 min on ice using 2 s pulses. All
purification steps were performed at 4˚C on an ÄKTAxpress purifier
(GE Healthcare). His-tagged enzyme was purified using a HisTrap FF 5 ml column
(GE Healthcare) equilibrated with Buffer A. The sample was loaded at a flow rate of
4 ml/min and step-eluted with Buffer B (50 mM Tris-HCl pH 8, 50 mM glycine,
500 mM NaCl, 5% glycerol, 500 mM imidazole). Eluted protein was subjected to
further purification on a Superdex Hiload 26/60 S75 gel filtration column
(GE Healthcare) at a flow rate of 3.2 ml/min using Buffer C (20 mM Hepes pH 7.5,
150 mM NaCl) and collected into 8 ml fractions. After analysis by SDS-PAGE,
those fractions containing no traces of other contaminating proteins were pooled
and dialyzed in Buffer C (50 mM phosphate pH 7.6, 100 mM NaCl) and
concentrated. Protein concentration was measured at 280 nm with Nanodrop
according to the manufacturer’s software. Purified proteins were divided in 20 μl
aliquots, fast-frozen in liquid nitrogen and stored at −20 °C.
CYP71D1V1 enzyme was expressed in the Saccharomyces cerevisiae
WAT11 cells32 grown in synthetic complete without histidine medium (SC–His)
plus 2% Glucose. A starter culture was grown for 48 h at 30 °C in 100 mL of SC–His
media supplemented with 2% Glucose. A 1:100 dilution in fresh SC-His (4 × 1 l)
media supplemented with 2% Glucose was prepared and allowed to grow at 30˚C
24 h. Yeast cells were collected by centrifugation and resuspended in SC-His
(4 × 1 l) media and protein production was induced by addition of 2% galactose
and the cultures were shaken at 30˚C overnight. The following day (20 h later), cells
were collected by centrifugation, lysed by French Disruption Cell Press (one shot at
25 kPSI) in Buffer D (50 mM Tris, pH 7.4, 1 mM EDTA, 0.6 M sorbitol). After
centrifugation at 35,000×g for 10 min and precipitation of cell debris, the
supernatant was centrifuged at 100,000×g for 1 h. The pellet was resuspended in
Buffer E (50 mM Tris, pH 7.4, 1 mM EDTA, 20 % glycerol). In total, from 4 l of
yeast culture, a yield of 28 ml of microsomal protein resulted. Purified microsomal
protein was divided in 500 μl aliquots, fast-frozen in liquid nitrogen and stored at
−80 °C.
Methods
Correlation analysis for selection of P450 candidates. An expression matrix
was generated using the CDF97 consensus transcriptome30 after pseudo-mapping
reads from paired-end sequencing runs available from EBI ENA (ERR1512369,
ERR1512370, ERR1512371, ERR1512372, ERR1512373, ERR1512374,
ERR1512375, ERR1512376, ERR1512377, SRR1144633, SRR1144634, SRR1271857,
SRR1271858, SRR1271859, SRR342017, SRR342019, SRR342022, SRR342023,
SRR646572, SRR646596, SRR646604, SRR648705, SRR648707, SRR924147, and
SRR924148). Quantification was achieved with Salmon v0.7.2 in the variational
bayesian optimized (--vbo) quasi-mapping mode with bias correction (--biasCor-
rect)31. Correlations were calculated with R (R Development Core Team, 2015).
Enzyme assays. Purified GS1 or GS2 and purified SGD17 were used in all assays.
CYP71D1V1 was used as a microsome preparation as described above. Strictosi-
dine was purified by preparative reverse phase HPLC and quantified using
1H NMR as previously described17. The strictosidine aglycone substrate was gen-
erated by deglycosylating strictosidine (100 or 300 μM) by the addition of purified
SGD in the presence of 50 mM phosphate buffer (pH 7.0) at 37 °C for 15 min. The
reactions were started by the addition of GS1 or GS2 enzymes (5 μM), NADPH
(5 mM) and microsomal protein containing GO (CYP71D1V1) (25 μl) at 37 ˚C in a
total volume of 100 μl with light shaking. Caffeine (50 μM) was used as internal
standard. All reactions were performed in quadruplicate. Aliquots of the reaction
mixtures (10 μl) were sampled 2, 5, 15, 30, 45 and 60 min after addition of the GS
and GO (CYP71D1V1) enzyme. The reactions were stopped by the addition of 10
μl of 100% MeOH. Samples were diluted further 1:5 in mobile phase (H2O + 0.1%
Cloning. The entire coding sequence of GS1 was amplified from Catharanthus
roseus cDNA using primers 5′-AAGTTCTGTTTCAGGGCCCGGCTGGAGAAA
CAACAAAACTAG-3′ and 5′-ATGGTCTAGAAAGCTTTATTCCTCAAATTTC
AATGTATTT-3′ and cloned into pOPIN-F expression vector (N terminus 6x
HisTag) using In-Fusion cloning while GS2 was cloned into the same expression
vector using the following synthesized DNA fragment with GeneArt String method
(ThermoFischer, USA) 5′-AAGTTCTGTTTCAGGGCCCGGCTGGAGAAACAA
CAAAACTAGACCTTTCAGTGAAGGCCATCGGATGGGGTGCTGCAGATGC
ATCTGGCGTTCTTCAGCCCATTAGGTTCTATAGAAGAGCCCCTGGTGA
ACGGGATGTGAAGATTAGAGTTTTGTACTGTGGTGTGTGCAATTTCGAT
6
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
8
:
3
1
6
D
O
I
:
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
7
-
0
0
1
5
4
-
x
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
|
|
|

N
A
T
U
R
E
C
O
M
M
UN
I
CA
T
I
O
NS
|
D
OI
:
1
0
.
1
0
3
8
/
s
4
1
46
7
-
01
7
-
001
5
4
-
x
A
R
T
I
C
L
E
formic acid) and centrifuged for 10 min at 4000×g before UPLC-MS injection
(1 µl). The activity of enzymes was measured by UPLC-MS.
Large-scale enzyme assay. For full characterization of GS1 and GO
(CYP71D1V1) products, two large-scale enzyme assays were performed. For the
characterization of the compounds (16R)-E-isositsirikine (1) and (16R)-Z-iso-
sitsirikine (2) produced by GS1, 9.2 mg of strictosidine prepared as described
previously17 were dissolved in 50 mL of 50 mM phosphate buffer (pH = 7.0). To
generate strictosidine aglycone as substrate, 5 nM of SGD enzyme was added and
incubated at 37 ˚C for 20 min. The consumption of strictosidine was monitored by
flow injection MS performed on an Advion express-ion Compact Mass Spectro-
meter. NADPH (25 mg) and aliquots of GS1 to final concentration 5 µM were
added to the reaction mixture. A NADPH-generation/regeneration system con-
sisting of glucose, glucose dehydrogenase (Sigma-Aldrich) was used additionally.
The reaction was monitored by flow injection MS at m/z 355.3 and after 3 h was
stopped by addition of 10 mL MeOH. The reaction mixture was centrifuged for 10
min at 3,130 g. The supernatant was loaded onto an SPE cartridge (RP18; 10 g; 60
ml), the SPE cartridge washed with 2 volumes of water and the enzyme reaction
product was eluted with 1 volume of MeOH. The methanol SPE fraction was
concentrated to 4 ml, filtered and products of GS1 were purified by preparative
HPLC (details of the method are below).
For the characterization of compound 3 (akuammicine) produced by GS1 and
GO (CYP71D1V1), 8.4 mg of strictosidine were dissolved in 50 ml of 50 mM
phosphate buffer (pH = 7.0). To generate strictosidine aglycone as substrate, 5 nM
of SGD enzyme were added and incubated at 37 ˚C for 20 min. The consumption
of strictosidine was monitored by flow injection MS performed on an Advion
express-ion Compact Mass Spectrometer. NADPH (25 mg), aliquots of GS1 to final
concentration 5 µM and 12.5 ml of microsomal protein containing GO
(CYP71D1V1) were added to reaction mixture. A NADPH-generation/
regeneration system consisting of glucose, glucose dehydrogenase (Sigma-Aldrich)
was also used. The reaction was monitored by flow injection MS at m/z 323.2 and
after 3 h was stopped by addition of 10 ml MeOH. The reaction mixture was
centrifuged for 10 min at 3130×g. The supernatant was loaded onto an SPE
cartridge (RP18; 10×g; 60 ml), the SPE cartridge was washed with 2 volumes of
water and enzyme reaction product was eluted with 1 volume of MeOH. The
methanol SPE fraction was concentrated to 4 ml, filtered and products of enzymatic
reaction were purified by preparative HPLC.
Preparative HPLC was performed on a Thermo Dionex Ultimate 3000
chromatography apparatus (Ultimate 3000 Pump, Ultimate 3000 Various
Wavelength Detector) using a Waters XBridge BEH C18 5 μ 10 × 250 OBD.). The
solvents used ammonium hydroxide (NH4OH) as Solvent A3 and 100%
acetonitrile as Solvent B3, with a flow rate of 5.0 ml/min. Injection volume was 500
µl. For the isolation the gradient profile was 0 min, 30% B3; from 0 to 15 min, linear
gradient to 95% B3; wash at 95% B3 for 2.5 min; from 17.5 min to 18 min, back to
30% B3 for 6 min to re-equilibrate the column. Elution of (16 R)-E-isositsirikine
(1) and (16 R)-Z-isositsirikine (2) was monitored at 225 nm, while elution of
akuammicine (3) was monitored at 328 nm. (16 R)-E-isositsirikine (1) was
collected at 7.3 min, (16 R)-Z-isositsirikine (2) at 8.6 min and akuammicine (3) at
9.7 min. The collected fractions were analyzed on a Shimadzu LCMS-IT-TOF Mass
Spectrometer. The chromatographic separation was carried out on a Phenomenex
Kinetex column 2.6 u XB-C18 100 Å (100 × 2.10 mm, 2.6 μm), and the binary
solvent system consisted of Solvent A4, H2O + 0.1% formic acid, and Solvent B4,
acetonitrile, at a constant flow rate 600 μl/min. The LC gradient began with 10%
Solvent B4 and linearly increased to 30% Solvent B4 in 5 min, then increased to
90% B3 in 1 min, held for 1.5 min and brought back to 10% Solvent B4. The
fractions containing the compound of interest were combined, dried and used
further for compound characterization. In total, 2.3 mg of (16 R)-E-isositsirikine
(1)24, ≈ 400 µg of (16 R)-Z-isositsirikine (2)24 and ≈ 600 µg of akuammicine (3)26
were isolated.
UPLC-MS chromatography was performed on a BEH Shield RP18 (50 × 2.1
mm; 1.7 μm) column (Waters). The solvents used were H2O + 0.1% formic acid as
Solvent A1 and 100% acetonitrile as Solvent B1, with a flow rate of 0.6 ml/min.
Injection volume was 1 µl. The gradient profile was 0 min, 5% B1; from 0 to 3.5
min, linear gradient to 35% B1; from 3.5 min to 3.75 min, linear gradient to 100%
B1; wash at 100% B1 for 1 min; from 4.75 min to 6 min, back to 5% B1 for 1 min to
re-equilibrate the column.
Mass spectrometry detection was performed on a Waters Xevo TQ-S mass
spectrometer (Milford, MA, USA) equipped with an electrospray (ESI) source.
Capillary voltage was 2.5 kV in positive mode; the source was kept at 150 °C;
desolvation temperature was 500 °C; cone gas flow, 50 l/h; and desolvation gas flow,
900 l/h. Unit resolution was applied to each quadrupole.
Targeted methods for each compound were developed using either commercial
standards (caffeine, ajmalicine, serpentine, catharanthine, vindoline, vindorosine,
loganic acid, tetrahydroalstonine, and yohimbine were purchased from Sigma-
Aldrich and tabersonine from Avachem Scientific) or enzymatically produced
compounds isolated as described here or as previously reported (strictosidine17 and
isositsirikine, this work). Akuammicine was synthesized and provided as a gift26
Flow injections of each individual compound were used to optimize the MRM
conditions. This was done automatically using the Waters Intellistart software.
A minimum dwell time of 25 ms was applied to each MRM transition. Four
.
transitions were used to monitor the elution of tetrahydroalstonine and ajmalicine:
m/z 353.2 > 117.0 (Cone 50, Collision 46), m/z 353.2 > 144.0 (Cone 50, Collision
26), m/z 353.2 > 210.1 (Cone 50, Collision 20) and m/z 353.2 > 222.0 (Cone 50,
Collision 20). Transition m/z 353.2 > 144.0 was used for quantification of these two
compounds. Transitions m/z 195.2 > 110.1 (cone 36, Collision 22) and m/z 195.2 >
138.2 (cone 36, Collision 18) were used for caffeine; transitions m/z 351.3 > 144.1
(cone 28, Collision 24) and m/z 351.3 > 170.2 (cone 28, Collision 22) were used for
deglycosylated strictosidine; transitions m/z 531.3 > 144.1 (cone 32, Collision 36)
and m/z 351.3 > 352.2 (cone 32, Collision 24) were used for detection of
strictosidine. Transitions m/z 355.3 > 117.1 (Cone 48, Collision 42), m/z 355.3 >
144.1 (Cone 48, Collision 26), m/z 355.3 > 212.2 (Cone 48, Collision 18) and m/z
355.3 > 224.2 (Cone 48, Collision 18) were used for detection of isositsirikine,
yohimbine. Transitions m/z 323.2 > 182.0 (Cone 50, Collision 36), m/z 323.2 >
234.1 (Cone 50, Collision 34), m/z 323.2 > 291.3 (Cone 50, Collision 22) were used
for detection of akuammicine. Peak areas were calculated using the Waters
MassLynx software and normalized.
The apparent rate under saturating substrate concentrations (observed kcat) for
GS1 was measured using NADPH absorbance using strictosidine (100 µM),
NADPH (100 µM) and GS1 (0.83 µM) in an assay previously reported17. The
apparent kcat was 0.04 0.007 s−1. Initial rates for GO can be calculated from the
data presented in Fig. 3a: 9.73 nmol l−1 s−1 (100 µM concentration of strictosidine,
5 μM GS1, 5 mM NADPH) and 7.99 nmol l−1 s−1 (300 µM strictosidine, 5 μM GS1,
5 mM NADPH). The specific activity for GO was not calculated since it was used as
a crude microsomal preparation (as is standard for microsomal cytochrome
P450s).
Chiral column liquid chromatography. Synthetic akuammicine and enzymatically
produced akuammicine were subjected to chiral resolution analysis by LCMS.
Chromatography was performed on a Lux i-Cellulose 5 (150 × 3.0 mm, 3 μm)
column (Phenomenex) under isocratic conditions using 60% 10 mM ammonium
acetate and 40% acetonitrile for 10 min with a flow rate of 0.6 ml/min. Synthetic
akuammicine (1 µl of 125 ng/ml solution) and 1 µl of enzyme assay were injected.
Mass spectrometry detection was performed on a Waters Xevo TQ-S mass spec-
trometer using the transitions m/z 323.2 > 182.0, m/z 323.2 > 234.1, m/z 323.2 >
291.3 for detection of akuammicine.
Compound characterization. High-resolution electrospray ionization mass spec-
trometry spectra and UV-Vis spectra were measured with a Shimadzu IT-TOF
mass spectrometer as described above. NMR spectra (1H NMR, 13C NMR) were
acquired using a Bruker Avance III 400 NMR spectrometer equipped with a BBFO
plus 5 mm probe, operating at 400 MHz for 1H and 100 MHz for 13C. The residual
1H- and 13C NMR signals of CD3Cl (δ 7.26 for 1H and δ 77.36 for 13C) were used
as internal chemical shift references. The number of scans depended on the con-
centration of the sample. The 1H NMR spectra were compared with those of
standards and literature data24, 26. For NMR, mass spectrometry and spectroscopic
data, see Supplementary Figs 5–23.
Formaldehyde release detection assay. Release of formaldehyde as result of
rearrangement catalyzed by GO (CYP71D1V1) was monitored using a
fluorescence-based Nash assay as previously reported33. Nash reagent was prepared
by adding 0.3 ml of acetic acid and 0.2 ml of acetylacetone to 100 ml of 2 M
ammonium acetate. Enzyme assays were performed in a total volume of 200 µl,
using 300 μM strictosidine aglycone substrate, 5 μM GS1, 5 mM NADPH and
various quantities of GO (CYP71D1V1) microsomes ranging from 5 to 100 µl in
50 mM phosphate buffer (pH 7.0). After 20 min, enzyme assays were centrifuged
for 5 min at 3130×g for 5 min, 100 µl of enzyme assay were quenched with two
volumes of Nash reagent and incubated at 60 ˚C for 10 min to convert for-
maldehyde to diacetyldihydrolutidine, which was detected by fluorescence using a
BMG Labtech CLARIOstar microplate reader at λex = 412 nm and λem = 505 nm.
A calibration curve using formaldehyde (Sigma-Aldrich) from 0–50 µM was made
for quantification. Peak areas were calculated using the BMG Labtech CLARIOstar
software and normalized. Aliquots of the reaction mixtures (10 μl) after cen-
trifugation were quenched by the addition of 10 μl of 100% MeOH. Samples were
diluted further 1:5 in mobile phase (H2O + 0.1% formic acid) and centrifuged for
10 min at 4000×g before UPLC-MS injection (1 µl) and the activity of enzymes was
measured by UPLC-MS as described above. Peak areas were calculated using the
Waters MassLynx software and normalized.
(16R)-E-isositsirikine (1). 1H-NMR (400 MHz, CDCl3): δ 8.66 p.p.m. (s, br, 1
H),), 7.48 (d, br, J = 7.6 Hz, 1 H), 7.38 (d, br, J = 7.9 Hz, 1 H), 7.16 (ddd, J = 7.9 Hz,
J = 7.2 Hz, J = 1.1 Hz, 1 H), 7.10 (ddd, J = 7.6 Hz, J = 7.2 Hz, J = 1.1 Hz, 1 H), 5.63
(q, J = 6.8 Hz, 1 H), 4.32 (m, br, 1 H), 3.82 (s, 3 H), 3.57 (dd J = 11.3 Hz, J = 5.8 Hz,
1 H), 3.54 (d J = 12.3 Hz, 1 H), 3.51 (dd J = 11.3 Hz, J = 4.7 Hz, 1 H), 3.28 (ddd, J =
13.0 Hz, J = 5.7 Hz, J = 0.9 Hz, 1 H), 3.19–3.09 (m, 2 H), 3.05–2.96 (m, 1 H), 2.94
(d, J = 12.2 Hz, 1 H), 2.65 (ddd, J = 14.5 Hz, J = 4.5 Hz, J = 1.0 Hz, 1 H), 2.51
(ddd, J = 11.4 Hz, J = 7.8 Hz, J = 5.0 Hz, 1 H), 2.29–2.15 (m, 2 H), 1.68 (dd, J = 6.9
Hz, J = 1.7 Hz 3 H); 13C-NMR (125 MHz, CDCl3): δ 175.6, 136.3, 134.1, 133.7,
127.8, 123.6, 121.7, 119.7, 118.1, 111.4, 108.0, 62.3, 52.9, 52.5, 52.4, 51.6, 49.7, 32.7,
30.5, 17.8, 13.5; UV/vis: λmax 225, 280 nm; HRMS (m/z): [M + H]+ calcd for
C
21H27N2O3, 355.2016; found, 355.2020; analysis.
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
8
:
3
1
6
D
O
I
:
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
7
-
0
0
1
5
4
-
x
w
w
w
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
r
e
c
o
m
m
u
n
i
c
a
t
i
o
n
s
7
|
|
|

A R T I C L E
N A T U R E C O M MU N ICA T IO N S | D O I: 1 0 . 1 0 3 8 / s 4 1 4 67 - 0 17 - 0 0 15 4 -x
(16R)-Z-isositsirikine (2). 1H-NMR (400 MHz, CDCl3): δ 7.88 p.p.m. (s, br, 1 H),
7.46 (d, br, J = 7.6 Hz, 1 H), 7.32 (d, br, J = 7.9 Hz, 1 H), 7.14 (ddd, J = 7.9 Hz,
J = 7.2 Hz, J = 1.1 Hz, 1 H), 7.08 (ddd, J = 7.6 Hz, J = 7.2 Hz, J = 1.1 Hz, 1 H), 5.46
(q, J = 6.8 Hz, 1 H), 3.91 (dd J = 11.3 Hz, J = 7.8 Hz, 1 H), 3.81 (dd J = 11.3 Hz,
J = 4.5 Hz, 1 H), 3.78 (d J = 12.7 Hz, 1 H), 3.74 (s, 3 H), 3.63 (d, br, J = 10.5 Hz,
1 H), 3.17 (ddd, J = 10.8 Hz, J = 5.5 Hz, J = 1.0 Hz, 1 H), 3.05–2.93 (m, 2 H), 2.90
(d, J = 12.7 Hz, 1 H), 2.78–2.67 (m, 2 H), 2.67–2.58 (m, 1 H), 2.15 (ddd, J = 12.5 Hz,
J = 4.5 Hz, J = 4.0 Hz, 1 H), 1.72 (d, J = 6.8 Hz, 3 H), 1.70–1.65 (m, 1 H); 13C-NMR
(125 MHz, CDCl3) (extracted from 1H – 13C HSQC): δ 121.7, 119.7, 119.1, 118.3,
111.2, 62.7, 58.7, 54.7, 52.6, 52.0, 49.2, 41.1, 34.5, 21.2, 13.5 UV/vis: λmax 225, 279
nm; HRMS (m/z): [M + H]+ calcd for C21H27N2O3, 355.2016; found, 355.2014;
analysis.
compounds followed by an increase to 100% B6 at 18 min, a 2 min wash step and a
re-equilibration at 0% B6 for 3 min before the next injection. The column was kept
at 60 °C throughout the analysis and the flow rate was 0.6 ml/min. MS detection
was performed in positive mode ESI. Capillary voltage was 3.0 kV; the source was
kept at 150 °C; desolvation temperature was 500 °C; cone gas flow, 50 l/h and
desolvation gas flow, 800 l/h. Unit resolution was applied to each quadrupole.
Multiple Reactions Monitoring (MRM) signals were used for detection and
quantification of caffeine, heteroyohimbine alkaloids, strictosidine, akuammicine as
reported above.
For targeted metabolic analyses of the alkaloid content of the GO
(CYP71D1V1)-silenced plants, alkaloids were extracted from lyophilized leaves by
grinding tissues with a mixer mill (Restch, MM 400) during 3 min at the maximal
frequency. The resulting powders were incubated in 1 ml methanol (containing
0.1% formic acid) under vigorous shaking during 1 h at 24 °C. After centrifugation
(15,000×g, 15 min), supernatants were collected and used for quantification.
Alkaloid quantifications were performed using an UPLC-MS chromatography
system coupled to a SQD mass spectrometer equipped with an electrospray
ionization (ESI) source controlled by Masslynx 4.1 software (Waters, Milford, MA).
Analyte separation was performed on a Waters Acquity HSS T3 C18 column
(150 × 2.1 mm, i.d. 1.8 µm) with a flow rate of 0.4 ml/min at 55 °C and the volume
of injection was 5 µL. The following linear elution gradient was used: acetonitrile-
water-formic acid from 10:90:0.1 to 60:40:0.1 over 18 min. The capillary and
sample cone voltages were 3,000 V and 30 V, respectively. The cone and
desolvation gas flow rates were 60 and 800 l/h. MS experiments were carried out in
positive mode in the selected ion-monitoring mode using m/z 337 for
catharanthine ([M + H]+, RT = 12.33 min), m/z 457 for vindoline ([M + H]+,
RT = 14.69 min), m/z 427 for vindorosine ([M + H]+, RT = 15.03 min), m/z 353 for
ajmalicine ([M + H]+, RT = 11.7 min), m/z 349 for serpentine ([M + H]+,
RT = 13.01 min) and m/z 375 for loganic acid ([M-H]-, RT = 4.94 min). The
acquired data was processed by the QuanLynx™ software (Waters, UK). Relative
quantification was performed by correcting peak areas by sample masses.
Gene silencing was confirmed by qRT-PCR. For GS1-silenced plants, RNA
extraction was performed using the RNeasy Plant Mini Kit (Qiagen). RNA (1 µg)
was used to synthesize cDNA in 20 µl reactions using the iScript cDNA Synthesis
Kit (Bio-Rad). The cDNA served as template for quantitative PCR performed using
the CFX96 Real-Time PCR Detection System (Bio-Rad) using the SSO Advanced
SYBR Green Supermix (Bio-Rad). Each reaction was performed in a total reaction
volume of 20 µl containing an equal amount of cDNA, 0.25 mM forward and
reverse primers, and 1x SsoAdvanced SYBRGreen Supermix (Bio-Rad). The
reaction was initiated by a denaturation step at 95 °C for 10 min followed by
41 cycles at 95 °C for 15 s and 60 °C for 1 min. For GO (CYP71DV1) silenced
plants, RNA was extracted with the NucleoSpin RNA Plant kit (Macherey- Nagel)
and 1 µg from each extraction was retro-transcribed using the RevertAid first
strand cDNA synthesis kit (ThermoFischer Scientific) with random hexamers
(5 µM) according to the manufacturer’s instructions. Gene expression levels were
monitored by quantitative PCR performed using the Dynamo ColorFlash probe
qPCR kit (ThermoFischer Scientific) in a 15 µl final volume containing 6 µl diluted
template cDNA and the forward and reverse primers (0.5 µM). Amplifications were
performed on a CFX96 real-time SYBR system (Bio-Rad) using detection of SYBR
green with the following conditions: 95 °C for 7 min, 40 cycles at 95 °C for 10 s and
60 °C for 40 s. Melting curves were used to determine the specificity of the
amplifications. Relative quantification of gene expression was calculated according
to the delta-delta cycle threshold method using the 40 S ribosomal protein S9
(RPS9). The primers 5′-TTGAGCCGTATCAGAAATGC-3′ and 5′-CCCTCATCA
AGCAGACCATA-3′ were used for RPS9, and 5′-TACTGAAGTTATTGGGAT
TGA-3′ and 5′-TTCAATGTATTTCCAATGTCA-3′ were used for GS1 and
5′-GCTGAGTTTATGTTGGCTGCTATGTT-3′ and 5′-ATAGTTGGCAAAGA
CAGACTAATCGT-3′ for GO (CYP71D1V1). All primer pair efficiencies were
between 98% and 108%, and the individual efficiency values were considered in the
calculation of normalized relative expression, which was performed using the Gene
Study feature of CFX Manager Software. All biological samples were measured in
technical duplicates (GS1-silenced plants) or triplicates (GO (CYP71D1V1)-
silenced plants).
Akuammicine (3). 1H-NMR (600 MHz, CDCl3): δ 9.00 p.p.m. (s, br, 1 H), 7.24
(d, br, J = 7.6 Hz, 1 H), 7.15 (ddd, J = 7.6 Hz, J = 7.6 Hz, J = 1.0 Hz, 1 H), 6.89
(ddd, J = 7.6 Hz, J = 7.6 Hz, J = 1.0 Hz, 1 H), 6.82 (d, br, J = 7.6 Hz, 1 H), 5.35
(q, br, J = 6.2 Hz, 1 H), 4.03 (m, br, 1 H), 3.94 (m, br, 1 H), 3.89 (d, br, J = 14.9 Hz,
1 H), 3.28 (m, 1 H), 3.02 (dd, J = 12.4 Hz, J = 6.7 Hz, 1 H), 2.95 (d, J = 15.0 Hz,
1 H), 2.51 (ddd, J = 12.6 Hz, J = 12.6 Hz, J = 6.8 Hz, 1 H), 2.42 (ddd, J = 13.8 Hz,
J = 4.0 Hz, J = 2.3 Hz, 1 H), 1.82 (dd, J = 12.3 Hz, J = 5.5 Hz, 1 H), 1.60
(d, J = 6.9 Hz, 3 H), 1.30 (ddd, J = 13.8 Hz, J = 2.8 Hz, J = 2.8 Hz, 1 H); 13C-NMR
(150 MHz, CDCl3) (extracted from 1H – 13C HSQC & 1H – 13C HMBC): δ 167.8,
143.1, 139.3, 136.2, 127.8, 121.1, 120.9, 120.3, 109.5, 61.7, 57.4, 56.9, 56.2, 51.0, 46.1,
30.8, 29.7, 12.9. UV/vis: λmax 204, 225, 293, 328 nm; HRMS (m/z): [M + H]+ calcd
for C20H23N2O2, 323.1754; found, 323.1760; analysis.
Virus–induced gene silencing. The GS1 silencing fragment was amplified
with primers 5′-GGCGCGAUGTGTTTGCAATTTCGATATGG-3′ and
5′-GGTTGCGAUATAGGATCGTTCCCCTTG-3′, to give a gene fragment of 266
bp. However, the very high nucleotide sequence identity (95%) of GS2 to GS1
means that co-silencing of these two genes is unavoidable. The resulting fragment,
when subjected to a tnBLAST search against the C. roseus transcriptome at http://
medicinalplantgenomics.msu.edu, did not show consecutive 22 bp duplication
(minimal bp sequence needed for successful silencing) to any other gene, sug-
gesting that cross-silencing with any other gene is highly unlikely. This fragment
was cloned into the pTRV2u vector as described17 and was used to silence the GS
in C. roseus seedlings. Leaves from the first two pairs to emerge following inocu-
lation were harvested from eight plants transformed with the empty pTRV2u and
pTRV2u-GS1. Similarly, the GO (CYP71D1V1) silencing fragment was amplified
using primers 5′-CTGAGAGGATCCTACAGTATGGCCCGA-3′ and 3′-CTGA-
GAGGATCCATCGTTAACAAGATGAGGAACCAAT-5′, to generate a 298 bp
cDNA displaying low identity with other C. roseus transcripts to avoid potential
gene cross-silencing. After cloning this gene fragment into pTRV2u to generate
pTRV2u-71D1V1, gene silencing was achieved on C. roseus plantlets through the
biolistic-mediated inoculation of viral vectors34, 35. Leaves from the first pair to
develop post-transformation were harvested from eight plants transformed with
the empty pTRV2u or pTRV2u-71D1V1. For both types of transformation, the
collected leaves were frozen in liquid nitrogen, powdered using a pre-chilled mortar
and pestle (leaves from GS1- silenced plants) or using a mixer mill (Retsch MM400
for leaves from GO (CYP71D1V1) silenced plants), and subjected to LCMS and
qRT-PCR analysis. Each pair of plant leaves was analyzed separately, for a total of
8 biological replicates.
To assess the results of VIGS, the alkaloid content of silenced leaves was
determined by LCMS. To comprehensively assess the global effect of silencing on
C. roseus metabolism using the GS1 fragment, an untargeted metabolomics analysis
by LCMS was performed. Leaves were weighed (8–30 mg) and collected into a fixed
volume of methanol (200 μl) and incubated at 56 °C for 60 min. After a 30 min
centrifugation step at 5,000 g, an aliquot of the supernatant (20 μl) was diluted to
400 μl with water and analyzed on a Shimadzu LCMS-IT-TOF Mass Spectrometer.
The chromatographic separation was carried out on a Phenomenex Kinetex
column 2.6 µ XB-C18 100 Å (100 × 2.10 mm, 2.6 μm), and the binary solvent
system consisted of Solvent A5, H2O + 0.1% formic acid, and Solvent B5,
acetonitrile, at a constant flow rate 600 μl/min. The LC gradient began with 10%
Solvent B5 and linearly increased to 30% Solvent B5 in 5 min, then increased to
90% B5 in 1 min, held for 1.5 min and brought back to 10% Solvent B5. Peak areas
were calculated using the Shimadzu Profiling Solution software and normalized by
leaf mass (fresh weight). Analysis revealed that the only significant metabolic
change with the GS1 silencing fragment was the accumulation of
Transcript distribution analysis. Epidermis-enriched fractions were generated
according to a previously described procedure36. Briefly, a cotton swab was dipped
into a carborundum powder (particle size < 300 grit, Fischer) and used to abrade
both lower and upper epidermis layers of young C. roseus leaves. Abraded leaves
were dipped in 4 ml of Trizol (Life Technologies) for 5 s in a 15 mL centrifuge tube.
A total of 3 × 10 leaves were abraded and RNAs were extracted according to the
manufacturer’s protocol. The RNA pellet resulting from the isopropanol pre-
cipitation was washed with 70% ethanol and re-suspended in 100 µl of RNAse-free
water. Excess sugars were removed by precipitation with 10% ethanol for 5 min on
ice and centrifugation for 5 min at 15,000×g and 4 °C. The supernatant was
recovered and precipitated with 0.5 volume of 3 M sodium acetate (pH 5.2) and
2.5 volumes of 100% ethanol. The pellet was washed with 70% ethanol and
re-suspended with 20 µl of RNase-free water. Total RNAs from whole young leaves
were also extracted with Trizol (Life Technologies) according to the manufacturer’s
protocol. RNA from both fractions was quantified using a NanoDrop® ND-1000
and 1 µg was retro-transcribed with the RevertAid First Strand cDNA Synthesis Kit
according to provider’s instructions (ThermosFisher Scientific). Gene expression
heteroyohimbines. The diastereomers tetrahydroalstonine and ajmalicine, which
are both naturally present in C. roseus seedlings, do not separate under these
chromatographic conditions.
For targeted metabolomic analysis, the secondary metabolite content of GS
silenced leaves was determined by a different chromatographic method. Leaf
powder was weighed (1–6 mg), extracted with methanol (2 ml) and vortexed for
1 min. After a 10-min centrifugation step at 17,000 g, an aliquot of the supernatant
(200 µl) was filtered through 0.2 µm PTFE filters and analyzed on a Waters Xevo
TQ-MS. The chromatographic separation was carried out with an Acquity BEH
C18 1.7 μm 2.1 × 50 mm column and the binary solvent system consisted of solvent
A6, which was 0.1% NH4OH and solvent B6, which was 0.1% NH4OH. A linear
gradient from 0% to 65% B6 in 17.5 min was applied for separation of the
8
N
A
TU
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
8
:
3
1
6
D O I : 1 0 . 1 0 3 8 / s 41 4 6 7 - 0 1 7 - 0 0 1 5 4 - x w ww .n a t u r e . c o m / n a t u re c o m m u n i c a t i o n s
|
|
|

N
A
T
UR
E
C
O
M
M
U
N
I
CA
T
I
O
NS
|
D
OI
:
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
1
7
-
0
0
1
5
4
-
x
A
R
T
I
C
L
E
levels were measured by quantitative PCR run on a CFX96 Touch Real-Time PCR
System (Bio-Rad). Each reaction was performed in a total reaction volume of 25 μl
containing an equal amount of cDNAs (1/3 dilution), 1 × DyNAmo ColorFlash
Probe qPCR Kit (Thermo Fisher Scientific), and 0.05 μM forward and reverse
specific primers, qSTRfor 5′-CATAGCTCTGTGGGTATATTAGTGT-3′, qSTRrev
5′-CATAGCTCTGTGGGTATATTAGTGT-3′; qSGDfor 5′-CTTCGACAACTTC
GAATGGAA-3′; qSGDrev 5′-CTTCTTGACTAACTCAACTAGT-3′; qHDSfor
5′- GTCCCTTACTGAACCTCCAGAG-3′; qHDSrev 5′-AATCACCTGTCCTGC
GTTGG-3′; qRPS9for 5′- TTACAAGTCCCTTCGGTGGT-3′, qRPS9rev 5′-TGC
TTATTCTTCATCCTCTTCATC-3′: qGS1for 5′-TTGACTATCTCAGCACTGCT
ATGGA-3′, qGS1rev 5′-GAACCAAGATGAAGACGGCACCTAA-3′ and
qCYP71D1V1for 5′-ATGGATTTCTATGGAAGTAACTTTGAA-3′ and
8. Scott, A. I. & Qureshi, A. A. Biogenesis of Strychnos, Aspidosperma, and Iboga
alkaloids. Structure and reactions of preakuammicine. J. Am. Chem. Soc. 91,
5874–5876 (1969).
9. Eckermann, R. & Gaich, T. The double-bond configuration of corynanthean
alkaloids and its impact on monoterpenoid indole alkaloid biosynthesis.
Chemistry 22, 5749–5755 (2016).
10. Benayad, S., Ahamada, K., Lewin, G., Evanno, L. & Poupon, E.
Preakuammicine: A Long-Awaited Missing Link in the Biosynthesis of
Monoterpene Indole Alkaloids. Eur. J. Org. Chem. 2016, 1494–1499 (2016).
11. Menzies, J. R. W., Paterson, S. J., Duwiejua, M. & Corbett, A. D. Opioid activity
of alkaloids extracted from Picralima nitida (fam. Apocynaceae). Eur. J.
Pharmacol. 350, 101–108 (1998).
qCYP71D1V1rev 5′-AGTCCTCAGTCAAATCAATTTCTTCT-3′. The amplifica-
tion program was 95 °C for 7 min (polymerase heat activation), followed by 40
cycles containing 2 steps, 95 °C for 10 sec and 60 °C for 40 sec. Quantification of
transcript copy number was performed with calibration curves and normalization
with the RPS9 reference gene and expressed relative to the amount of transcript
measured in the whole leaf fraction. All amplifications were performed in triplicate
and repeated at least on two independent biological repeats. Expression of
hydroxymethylbutenyl 4-diphosphate synthase (HDS) known to accumulate within
12. Corbett, A. D., Menzies, J. R. W., MacDonald, A., Paterson, S. J. & Duwiejua, M.
The opioid activity of akuammine, akuammicine and akuammidine: alkaloids
extracted from Picralima nitida (fam. Apocynaceae). Br. J. Pharmacol 119, 334
(1996).
13. Gongora-Castillo, E. et al. Development of transcriptomic resources for
interrogating the biosynthesis of monoterpene indole alkaloids in medicinal
plant species. PLoS ONE 7, e52506 (2012).
14. Kellner, F. et al. Genome-guided investigation of plant natural product
biosynthesis. Plant J. 82, 680–92 (2015).
internal phloem associated parenchyma was used as a control37
.
15. Dugé de Bernonville, T. et al. Folivory elicits a strong defense reaction in
Catharanthus roseus: metabolomic and transcriptomic analyses reveal distinct
local and systemic responses. Sci. Rep 7, 40453 (2017).
16. Miettinen, K. et al. The seco-iridoid pathway from Catharanthus roseus. Nat.
Commun. 5, 3606 (2014).
17. Stavrinides, A. et al. Structural investigation of heteroyohimbine alkaloid
synthesis reveals active site elements that control stereoselectivity. Nat.
Commun. 7, 12116 (2016).
18. Stavrinides, A. et al. Unlocking the diversity of alkaloids in Catharanthus
roseus: nuclear localization suggests metabolic channeling in secondary
metabolism. Chem. Biol. 22, 336–341 (2015).
19. Werck-Reichhart, D. & Feyereisen, R. Cytochromes P450: a success story.
Genome Biol. 1, Reviews3003 (2000).
20. Irmler, S. et al. Indole alkaloid biosynthesis in Catharanthus roseus: new
enzyme activities and identification of cytochrome P450 CYP72A1 as
secologanin synthase. Plant J. 24, 797–804 (2000).
21. Yamamoto, H., Katano, N., Ooi, A. & Inoue, K. Secologanin synthase which
catalyzes the oxidative cleavage of loganin into secologanin is a cytochrome
P450. Phytochemistry 53, 7–12 (2000).
22. Huang, L. et al. Molecular characterization of the pentacyclic triterpenoid
biosynthetic pathway in Catharanthus roseus. Planta 236, 1571–1581 (2012).
23. Scott, A. I. Biosynthesis of the indole alkaloids. Acc. Chem. Res. 3, 151–157
(1970).
Subcellular localization. Determination of the subcellular localization of GS1 and
GO (CYP71D1V1) was performed by subcloning the full length GS1 and
CYP71D1V1 coding sequences into the pSCA-cassette vector as YFP fusions. To
avoid masking of the potential N-terminal transmembrane helix, CYP71D1V1 was
only expressed as N-terminal fusion with YFP (GO (CYP71D1V1)-YFP) while GS1
was expressed as both N-terminal or C-terminal fusions (GS1-YFP and YFP-GS1).
Amplifications were achieved with primers 5′-CTGAGAACTAGTATGGCCG
GAGAAACAACCAAAC-3′ and 5′-CTGAGAACTAGTTTCCTCAAATTTCAAT
GTATTTCCAATG-3′ for GS1; and 5′-CTGAGAACTAGTATGGAGTTTTCTT
TCTCCTCACCAG-3′ and 5′-CTGAGAACTAGTATCGTTAACAAGATGAG
GAACCAATT-3′ for CYP71D1V1. The nuclear (nucleus-CFP), nucleocytosolic
(CFP) and endoplasmic reticulum (ER-CFP) markers were described previously17
Transient transformation of C. roseus cells by particle bombardment and fluores-
cence imaging were performed following the procedures previously described35, 38
.
.
C. roseus cells were bombarded with DNA-coated gold particles (1 μm) and 1,100
psi rupture disc at a stopping-screen-to-target distance of 6 cm, using the Bio-Rad
PDS1000/He system and 400 ng of each plasmid per transformation. Cells were
cultivated for 16 h to 38 h before being harvested and observed. The subcellular
localization was determined using an Olympus BX-51 epifluorescence microscope
equipped with an Olympus DP-71 digital camera and a combination of YFP and
CFP filters.
Data availability. Sequence data for GS1 (KF302079.1), GS2 (KF302078.1). and
GO (JN613015) have been deposited in GenBank as described in the manuscript.
Transcriptome data used in this study has already been published as referenced in
the manuscript (see also, http://medicinalplantgenomics.msu.edu/). Data sup-
porting the findings of this study are available within the article and its
supplementary information files and from the corresponding author upon
reasonable request.
24. Lousnasmaa, M., Jokela, R., Hanhinen, P., Miettinen, J. & Salo, J. Preparation
and conformational study of Z- and E-Isositsirikine epimers and model
compounds. Determination of their C-16 configurations. Tetrahedron 50,
9207–9222 (1994).
25. Hirata, T., Lee, S.-L. & Scott, A.I. Biosynthesis of (16R)-isositsirikine from
geissoschizine with a cell-free extract of Catharanthus roseus callus. J. Chem.
Soc. Chem. Commun. 1979, 1081–1083 (1979).
26. Hong, A. Y. & Vanderwal, C. D. A synthesis of alsmaphorazine B demonstrates
the chemical feasibility of a new biogenetic hypothesis. J. Am. Chem. Soc. 137,
7306–7309 (2015).
Received: 3 February 2017 Accepted: 6 June 2017
27. Murata, J., Roepke, J., Gordon, H. & De Luca, V. The leaf epidermome of
Catharanthus roseus reveals Its biochemical specialization. Plant Cell 20,
524–542 (2008).
28. Heimberger, S. I. & Scott, I. A. Biosynthesis of strychnine. J. Chem. Soc. Chem.
Commun. 1973, 217–218 (1973).
29. Jones, S. B., Simmons, B., Mastracchio, A. & MacMillan, D. W. Collective
synthesis of natural products by means of organocascade catalysis. Nature 475,
183–188 (2011).
30. Duge de Bernonville, T. et al. Characterization of a second secologanin synthase
isoform producing both secologanin and secoxyloganin allows enhanced de
novo assembly of a Catharanthus roseus transcriptome. BMC Genomics 16, 619
(2015).
References
1. O’Connor, S. E. & Maresh, J. J. Chemistry and biology of monoterpene indole
alkaloid biosynthesis. Nat. Prod. Rep. 23, 532–547 (2006).
2. Courdavault, V. et al. A look inside an alkaloid multisite plant: the
Catharanthus logistics. Curr. Opin. Plant Biol. 19, 43–50 (2014).
3. Geerlings, A., Ibanez, M. M., Memelink, J., van Der Heijden, R. & Verpoorte, R.
Molecular cloning and analysis of strictosidine beta-D-glucosidase, an enzyme
in terpenoid indole alkaloid biosynthesis in Catharanthus roseus. J. Biol. Chem
275, 3051–3056 (2000).
31. Patro, R., Duggal, G., Love, M.I., Irizarry, R.A. & Kingsford, C. Salmon provides
fast and bias-aware quantification of transcript expression. Nat. Methods 14,
417–419 (2017).
4. Battersby, A. R. & Hall, E. S. Intermediacy of Geissoschizine in indole alkaloid
biosynthesis. Rearrangement to the Strychnos skeleton. J. Chem. Soc. D,
793-794 (1969).
32. Pompon, D., Louerat, B., Bronine, A. & Urban, P. Yeast expression of animal
and plant P450s in optimized redox environments. Methods Enzymol 272,
51–64 (1996).
5. Scott, A. I., Cherry, P. C. & Qureshi, A. A. Mechanism of indole alkaloid
biosynthesis. The Corynanthe-Strychnos relationship. J. Am. Chem. Soc. 91,
4932–4933 (1969).
33. Farrow, S. C. & Facchini, P. J. Dioxygenases catalyze O-demethylation and O,
O-demethylenation with widespread roles in benzylisoquinoline alkaloid
metabolism in opium poppy. J. Biol. Chem. 288, 8997–29012 (2013).
34. Carqueijeiro, I. et al. Virus-induced gene silencing in Catharanthus roseus by
biolistic inoculation of tobacco rattle virus vectors. Plant Biol. 17, 1242–1246
(2015).
6. Battersby, A. R. & Gibson, K. H. Further studies on rearrangements during
biosynthesis of indole alkaloids. J. Chem. Soc. D Chem. Commun. 902-904
(1971).
7. Qureshi, A. A. & Scott, A. I. Biosynthesis of indole alkaloids: sequential
precursor formation and biological conversion in Vinca rosea. Chem. Commun.
948-950 (1968).
N
A
T
U
R
E
C
OM
M
U
N
I
C
A
T
I
ON
S
8
:
3
1
6
D
O
I
:
1
0
.
1
0
3
8
/
s
4
1
46
7
-
0
1
7
-
0
0
1
5
4
-
x
w
ww
.
n
a
t
u
r
e
.
c
o
m
/
n
a
t
u
re
c
o
m
m
u
n
i
c
a
t
i
o
n
s
9
|
|
|

A R T I C L E
N A T U R E C O M MU N ICA T IO N S | D O I: 1 0 . 1 0 3 8 / s 4 1 4 67 - 0 17 - 0 0 15 4 -x
characterization of GS1 and T.D.d.B., A.O., A.L., F.L., and M.C. contributed to identi-
fication and characterization of GO.
35. Foureau, E. et al. Prequels to synthetic biology: From candidate gene
identification and validation to enzyme subcellular localization in plant and
yeast cells. Methods Enzymol 576, 167–206 (2016).
36. Payne, R., et al. An NPF transporter exports a central monoterpene indole
alkaloid intermediate from the vacuole. Nat. Plants 3, 16208 (2017).
37. Burlat, V., Oudin, A., Courtois, M., Rideau, M. & St-Pierre, B. Co-expression
of three MEP pathway genes and geraniol 10-hydroxylase in internal phloem
parenchyma of Catharanthus roseus implicates multicellular translocation of
intermediates during the biosynthesis of monoterpene indole alkaloids and
isoprenoid-derived primary metabolites. Plant J. 38, 131–141 (2004).
38. Guirimand, G. et al. Optimization of the transient transformation of
Catharanthus roseus cells by particle bombardment and its application to the
subcellular localization of hydroxymethylbutenyl 4-diphosphate synthase and
geraniol 10-hydroxylase. Plant Cell Rep. 28, 1215–1234 (2009).
Additional information
Supplementary Information accompanies this paper at doi:10.1038/s41467-017-00154-x.
Competing interests: The authors declare no competing financial interests.
Reprints and permission information is available online at http://npg.nature.com/
reprintsandpermissions/
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Acknowledgements
This work was supported by grants from the European Research Council (311363),
BBSRC (BB/J004561/1) (S.E.O.) and from the Région Centre-Val de Loire, France
(ABISAL grant and BioPROPHARM project –ARD2020) (V.C.). J.F. gratefully
acknowledges DFG postdoctoral funding (FR 3720/1-1) and T.T.-T.D. an EMBO Long
trem Fellowship (ALTF 239-2015). We gratefully acknowledge the generous gift of
akuammicine from Chris Vanderwal (UC Irvine).
Open Access This article is licensed under
a Creative Commons
Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons license, and indicate if changes were made. The images or other third party
material in this article are included in the article’s Creative Commons license, unless
indicated otherwise in a credit line to the material. If material is not included in the
article’s Creative Commons license and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this license, visit http://creativecommons.org/
licenses/by/4.0/.
Author contributions
E.C.T., I.C., S.E.O., and V.C. designed the experiments and wrote the manuscript. E.C.T.
characterized GS/GO in vitro, performed GS silencing, and performed all product
characterization. J.F. and T.-T.T.D. contributed to in vitro enzyme assays. I.C.,
T.D.d.B., and V.C. identified GO, performed GO silencing, and GO silencing results and
contributed to GS/GO in vitro assays and performed localization. I.C., E.C.T., and V.C.
identified the link between GS in vitro reactions. A.K.S. contributed to identification and
© The Author(s) 2017
10
N
A
TU
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
8
:
3
1
6
D O I : 1 0 . 1 0 3 8 / s 41 4 6 7 - 0 1 7 - 0 0 1 5 4 - x w ww .n a t u r e . c o m / n a t u re c o m m u n i c a t i o n s
|
|
|