Site selective C–H functionalization of Mitragyna alkaloids reveals a molecular switch for tuning opioid receptor signaling efficacy

Article abstract of DOI:10.1038/s41467-021-23736-2

Mitragynine (MG) is the most abundant alkaloid component of the psychoactive plant material “kratom”, which according to numerous anecdotal reports shows efficacy in self-medication for pain syndromes, depression, anxiety, and substance use disorders. We have developed a synthetic method for selective functionalization of the unexplored C11 position of the MG scaffold (C6 position in indole numbering) via the use of an indole-ethylene glycol adduct and subsequent iridium-catalyzed borylation. Through this work we discover that C11 represents a key locant for fine-tuning opioid receptor signaling efficacy. 7-Hydroxymitragynine (7OH), the parent compound with low efficacy on par with buprenorphine, is transformed to an?even lower efficacy agonist by introducing a fluorine substituent in this position (11-F-7OH), as demonstrated in vitro at both mouse and human mu opioid receptors (mMOR/hMOR) and in vivo in mouse analgesia tests. Low efficacy opioid agonists are of high interest as candidates for generating safer opioid medications with mitigated adverse effects.

Full text of DOI:10.1038/s41467-021-23736-2

A RT ICLE
https://doi.org/10.1038/s41467-021-23736-2
OPEN
S it e s elec tive C –H f u n c t i o n a l i z a t i o n o f Mitragyna
a lk al o i d s r eve a ls a m o lecu la r s wit ch fo r t un ing
o p io id rec ep to r s ign a l ing ef fica cy
1
,1 2
1, 2, 12
1
3,4
5,6
Sr i jita Bho wmi k
, J u raj G al eta
, Vá cl av H av el , M el i ss a N el s on
, A bde l fat tah F aou zi
,
Be n ja mi n Bec ha nd , Mi k e A ns on o ff 7, T om as F i al a 1,8, A m anda H un kel e 5, 9 , A ndr ew C. Kr ueg el1 ,
1
J o hn . E . P i nt ar 7, S us ru ta Maj u m dar 5, J o nat han A . J av i t ch
3,4
& D al i bor Sam es
1,1 0,11 ✉
Mitragyn in e (MG) is the most abundant alkaloid compon ent of th e psycho act ive plant material
kratom”, wh ich a ccording to numerous anecdot al rep orts show s effic a cy i n s e l f -m e di c a ti o n
“
for pain sy ndro mes, d epression, anxiety, and subst ance u se d isord ers. We h ave develo ped a
synthetic meth od fo r selective funct ionalization of t he un explored C11 po sitio n of the MG
scaffold (C 6 p ositio n in indole numbe ring) via the use o f an ind ole-eth ylen e glyco l add uct an d
su bsequ ent iridiu m-catalyzed borylation. Through t his work w e discover th at C 11 repre sen ts a
key loca nt fo r fin e-t un ing o pi oid r e ce p tor sig na lin g e f fic acy . 7-H yd rox ymit ra gyn in e ( 7O H) , th e
parent comp oun d w ith low efficacy on par with bupreno rph in e, is t ran sfo rmed t o an eve n
l owe r e ffica cy a go nist by introducing a fluorine substitu ent in th is po sition (11-F-7OH), as
demonstrated in vitro at both mouse and human mu op ioid recept ors (mMOR/h MOR) an d
in vivo in mou se ana lgesia t ests. Low e fficacy opioid agonists are o f high in terest as candid ate s
for gene ra ting safer opioid medicat ions with mitigat ed ad verse effects.
1
Department of Chemistry, Columbia University, New York, NY, USA. 2 Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences
3
(
IOCB Prague), 160 00 Prague 6, Czech Republic. Department of Psychiatry, and Department of Molecular Pharmacology and Therapeutics, Columbia
4
5
University, New York, NY, USA. Division of Molecular Therapeutics, New York State Psychiatric Institute, New York, NY, USA. Center for Clinical
Pharmacology, St Louis College of Pharmacy and Washington University School of Medicine, St Louis, MO 63110, USA. University of California San Diego,
La Jolla, CA 92161, USA. Department of Neuroscience and Cell Biology, Rutgers University, New Jersey, NJ 08854, USA. Laboratory of Organic Chemistry,
ETH Zürich, 8093 Zürich, Switzerland. Department of Neurology and Molecular Pharmacology, Memorial Sloan Kettering Cancer Center, New York, NY
6
7
8
9
0021, USA. 1 NeuroTechnology Center at Columbia University, New York, NY, USA. The Zuckerman Mind Brain Behavior Institute at Columbia University,
0
11
1
12
✉
New York, NY, USA. These authors contributed equally: Srijita Bhowmik, Juraj Galeta. email: ds584@columbia.edu
N A TU RE C OMM U N IC AT I ON S |
(2 0 2 1 ) 1 2 : 3 85 8 | h t t p s :/ /d o i . o r g /1 0 . 1 0 3 8 /s 4 1 4 67 - 0 2 1 - 2 3 7 3 6 - 2 | w ww . na tu r e . c o m /n atu r e c o m mu n i ca ti o n s
1

A R T I C L E
N AT URE CO M MU N ICAT I ON S | ht t ps: // doi. org /1 0 .10 38/ s41 46 7 -02 1 -2 3 7 36- 2
n search of molecules with robust clinical effects in the area of diseases)1,23. Therefore, safer opioid modulators represent pro-
central nervous system (CNS) disorders and ability to repair mising treatments for disorders spanning a wide spectrum of
synaptic function in the brain, we have been led to atypical physical and emotional pain. Accordingly, further understanding
I
1–4
modulators of endogenous opioid signaling . In this context, we of kratom and its alkaloids has major implications for public
became interested in the psychoactive plant Mitragyna speciosa that health.
has been used for centuries in Southeast Asia for treatment of pain,
Our previous report provided a basic structure–activity rela-
fatigue, opium dependence, and a number of other ailments. In the tionship (SAR) map of MG, in terms of in vitro opioid receptor
US, the use of the dry leaf material, known as “kratom”, has been on pharmacology, with respect to the substituents on the saturated
the rise in the last decade, along with the number of anecdotal rings of MG. This work was enabled by the total enantioselective
1
5
reports that point to the efficacy of kratom in a range of disorders synthesis of MG developed in our laboratories . Several other
with limited therapeutic options, including opioid dependence, research teams have also reported total syntheses of MG and
treatment-resistant depression, anxiety, and pain syndromes5–12. A related compounds24–28. These de novo synthetic approaches
number of alkaloids in this plant, including mitragynine (MG) and offer nearly unlimited exploration of complex MG-related
its oxidation product 7-hydroxymitragynine (7OH, Fig. 1), have molecules, but they are labor intensive due to the high number
been found to bind to opioid receptors and represent molecular of synthetic steps. Thus, the need to access many different deri-
1
3,14
scaffolds for the development of opioid receptor modulators
.
vatives at multiple positions demands more efficient synthetic
MG, as an indole alkaloid, shows no structural resemblance to strategies. MG can be extracted from kratom leaf matter in
1
4,15
traditional morphine-type compounds and represents an atypical multigram quantities (~1% of dry kratom mass
), and there-
opioid ligand with distinct signaling properties and physiological fore, there is a strong incentive to develop synthetic methods for
effects compared to clinically used opioid analgesics. For example, direct functionalization of MG, for example, via late-stage C–H
we have shown that MG is a partial MOR agonist (mu-opioid functionalization, rather than laborious total synthesis.
receptor agonist) with a potential bias for G protein signaling
With respect to derivatization of the indole nucleus, the
methoxy group at the C9 position is synthetically accessible by
1
5,16
(showing no β-arrestin-2 recruitment) in cell-based assays
.
Further, we have found that 7OH is a more potent and efficacious selective demethylation and subsequent functionalization of the
1
6
partial MOR agonist compared to MG, and that it acts as a potent free phenol . For the adjacent C10 position, a small series of
analgesic in mice1 . The relative activities of MG and 7OH are compounds has been prepared
5,16
14,29
, while the C11 position
highly relevant, as we recently reported that 7OH is formed in vivo remained unexplored due to the lack of functionalization meth-
1
7
from MG and mediates MG’s analgesic effects in mice . In addi- ods (see below). Our preliminary docking studies suggested that
tional preclinical studies, others have found that MG exhibits relatively small substituents (e.g., F, Cl, and Me) would be tol-
1
5
antinociceptive effects in dogs comparable to those of codeine but erated at C11 (Fig. 1) . However, it is presently difficult to
1
8
with less respiratory depression , and that the compound is not predict the effect of such substituents on opioid receptor activa-
self-administered by rats, but instead, inhibits self-administration of tion potency, efficacy, or signaling bias, demonstrating the need
morphine and heroin1
9,20
.
for C–H bond functionalization approaches compatible with the
The study and development of safer opioids is a long-standing structural complexity presented by Mitragyna alkaloids.
scientific and societal goal, and a part of the scientific strategy More than a decade ago, we formulated the general concept of
proposed by the National Institute of Health (NIH) initiative to “C–H bonds as ubiquitous functionality” and demonstrated the
2
1,22
address the opioid crisis
. Further, there is a growing interest strategic impact of C–H functionalization in both the construc-
in the use of opioid receptor modulators as medicaments tion of molecular frameworks and modification of existing
3
0–33
for depression and anxiety disorders (and other psychiatric complex cores
. For the latter, termed “complex core
N127
a/
Y75
OCH3
b/
Q124
7
N
W133
I322
extraction
OCH3
N
H
H
H
1
2
C217
H3CO
V143
O
kratom
C11 position
(
C6 indole)
mitragynine (MG)
I144
unexplored SAR
7
OH docked in
Oxidation
in vivo metabol.
mu opioid receptor
G protein signaling
H3CO
HO
N
OCH3
N
H
H3CO
O
7-hydroxymitragynine (7OH)
Fig. 1 The rationale for selective C–H bond functionalization of mitragynine (MG) and 7-hydroxymitragynine (7OH). a M G is re a dily obt ai ne d i n
mu l ti gr a m qu a nt i tie s f r om kr a t om pow de r by e x t ra ct i on, he nc e t he re is an inc e nt i ve t o deve l op se le c t iv e f unc t ional iz at ion of M G and re l at e d sca f f ol d s vi a
l a te -s t a ge C –H bo nd f un ct i ona l i zat i on. S pe cifica l ly , t he C1 1 pos i t ion has not be e n e xpl ore d in t e rms of mu-o pioid re c e pt or (M O R) si gna lin g a nd o t he r
b i ol o gi ca l e f f ec t s , du e t o t he l ac k of ch em is t ri e s fo r fu nct i ona l i zat ion of t his posi t ion. b T he t op-r ig ht pane l show s a doc k ing pose of 7 O H in hu man M OR
h i gh li gh ti ng t he a r oma t ic r i ng an d t he pr op ose d bi ndi ng poc ke t in t he re c e pt or .
2
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
(
2
0
2
1
)
1
2
:
3
8
5
8
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
2
1
-
2
3
7
3
6
-
2
|
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

NA TUR E COMM UNICAT IO NS | ht t ps ://doi .or g / 10 .1 0 38/ s 41 46 7 -02 1- 23 7 36-2
A R T I C L E
diversification” or “late-stage functionalization”, the synthetic approach was not applicable to MG as the TIPS-protected MG
power of this concept is readily apparent as the positional (as well decomposed under the catalytic conditions. On the other hand,
as stereoisomeric) analogs of complex starting materials are p-methoxybenzyl (PMB)-MG was either unreactive or formed
accessed with high efficiency when compared to lengthy de novo undesired products under the reaction conditions. Tert-
approaches3
0–36
. Late-stage functionalization approaches have butoxycarbonyl (Boc)-protected MG provided the C11-boronate
since been widely adopted and become a common part of che- ester, but the subsequent functionalization reaction was ineffi-
3
7–41
mists’ armamentarium
. These concepts are being extended cient (Supplementary Figs. 3 and 4). The latter approach was not
beyond C–H bonds to include the possibilities of skeleton mod- further optimized as an entirely different protection method was
ification via C–C and other bond activation 2–45.
4
found to be successful (see below).
In this paper, we describe application of these concepts to
Mitragyna alkaloids in the context of mapping their neuro- rearrangement of 3-bromoindolenines to 6-bromoindoles , which
pharmacology. We introduce a strategic temporary modification had been successfully applied in complex substrates en route to
of the MG alkaloid skeleton (“complex core restructuring”), to stephacidine A alkaloids . However, this approach also failed in the
create a distinct chemotype disposed toward the desired C–H MG scaffold under similar reaction conditions (Fig. 2c); namely,
functionalization chemistry, namely C11 functionalization of MG was unreactive in the presence of N-bromosuccinimide (NBS)
MG’s indole nucleus. In this manner, we offer a solution that under neutral conditions, while under acidic conditions, 12-Br-MG
provides for rapid and selective functionalization of the aromatic (3) was the major product (Fig. 2c). Apparently, the C9 methoxy
ring of MG at the C11 and C12 positions (C6 and C7 positions in group exerts an activating and directing effect in the benzene ring of
indole numbering). We subsequently found that these analogs the indole nucleus once the bromination reagent is sufficiently
Another avenue of enquiry was inspired by an intriguing thermal
53
4
2
5
4
enable fine-tuning of opioid receptor signaling efficacy, which activated .
represents one of the currently most promising strategies for
creating safer opioid therapeutics.
Finally, the methods relying on remote directing effect of
groups attached to the indole nitrogen, such as the copper-
catalyzed C6 functionalization of indoles using the phosphini-
5
5
mide directing group , were not examined, as (1) the deprotec-
tion step involves harsh reduction conditions (e.g., LiAlH4)
incompatible with MG’s functionalities or (2) the C6 functiona-
lization is too restrictive in terms of the functionalization
Results
C–H borylation of MG yields C12-substituted analogs. MG is a
complex natural product of a corynanthidine alkaloid type
decorated with a number of functional groups, including enol
ether, ester, tertiary amine, indole nitrogen, and aromatic methyl
ether, arranged in a specific constitutional and geometrical con-
figuration that underlies its opioid activity. It therefore poses an
exciting challenge for late-stage functionalization. In this study,
we focused on functionalization of the indole arene ring
5
6,57
chemistry or substrate requirements
.
Complex core restructuring: C–H borylation of mitragynine-
ethylene glycol adduct (MG-EG). Direct functionalization of MG
gave C12-halo or -boronate ester analogs, while re-routing this
regioselectivity was unsuccessful via either catalyst optimization
or indole nitrogen protection. The known methods for C6 indole
functionalization—that possess the required generality and scope
to pursue systematic SAR studies—failed to provide an efficient
synthetic route to C11-substituted MG analogs.
We therefore resorted to changing the reactivity of the MG
core by “removing” the indole double bond (2,3-π bond in indole
numbering) by either reduction or oxidation, or oxidation
followed by rearrangement (Fig. 3). 7OH was examined as an
indolenine substrate in the Ir-catalyzed borylation reaction, but
there was no conversion to a borylated product. 7OH can be
readily rearranged to mitragynine pseudoindoxyl with zinc triflate
(the rationale is discussed in “Introduction”). Iridium-catalyzed
arene C–H borylation was selected due to its wide substrate scope,
including basic heterocyclic compounds
4
6,47
.
After exploration and optimization of reaction conditions
Supplementary Table 1), we found that borylation is compatible
(
with the MG chemotype. Using bis(pinacolato)diboron (B Pin )
under catalytic [Ir(COD)OMe] and 4,4′-di-tert-butyl-2,2′-dipyr-
2
2
2
idyl (dtbpy) as the ligand in absolute heptane at 80 °C for ≥17 h,
C12-boronate ester 1 was obtained as the single isomer (Fig. 2a).
This compound proved to be unstable and decomposed during
silica gel column chromatography, and we thus confirmed its
formation in the crude material via nuclear magnetic resonance
spectroscopy (NMR), thin-layer chromatography (TLC), and
mass spectroscopy (MS). The crude ester 1 was transformed to
several derivatives without further purification in good yields.
Namely, 12-Cl- (2) and 12-Br-MG (3) were prepared in two steps
1
6
as reported by us previously , however this pseudoindoxyl
system afforded C12 borylation (Fig. 3), while the corresponding
2-boronate ester (not shown) was unreactive in the subsequent
1
halogenation reaction. Next, we investigated the possibility of
4
8
in 70% and 65% yield, respectively (Fig. 2a) . The observed
regioselectivity is consistent with the known directing effect of
58
using a reduced indole substrate, namely dihydromitragynine .
With this compound, we observed a partial conversion of starting
material to the corresponding 11-boronate ester (50% conversion
the indole nitrogen and C7-borylation of 2,3-disubstituted
indoles4
9,50
.
1
as determined by H NMR). However, the following halogenation
reaction gave complex mixtures of MG and 2,3-dihydromitragy-
Known C6 indole derivatization methods do not produce C11 nine as the major products, and 11-Br-MG and 11-Br-2,3-
analogs of MG. After achieving C12 functionalization we turned dihydromitragynine as the minor products (Fig. 3).
our attention to the C11 position in MG (C6 position in indole
Lastly, we focused on oxidative attachment of ethylene glycol to
numbering), which remained unexplored due to the limited the indole nucleus, rendering a distinct chemotype: mitragynine-
5
1
reaction repertoire available for C6 indole functionalization . We ethylene glycol (MG-EG, Fig. 4a). This unusual compound was
considered several known approaches for C6 indole functionali- first introduced by Takayama and co-authors to provide access to
zation to gain access to these derivatives. Specifically, we exam- the C10-substituted analogs via electrophilic aromatic substitu-
1
4,59
ined the possibility of blocking the indole nitrogen’s directing tion, and this scaffold showed potent activity at MOR
. Aside
effect with a protecting group, along the path demonstrated in the from these studies, this indole-ethylene glycol adduct has not
context of tryptophan substrates, where protection with a bulky been explored in the context of indole chemistry. We became
triisopropyl silane (TIPS) group led to C6-borylation under interested in this compound as the ethylene glycol group not only
5
2
optimized C–H borylation conditions (Fig. 2b) . However, this masks the indole, which is particularly relevant in this study, but
N A TU RE C OMM U N IC AT I ON S |
(2 0 2 1 ) 1 2 : 3 85 8 | h t t p s :/ /d o i . o r g /1 0 . 1 0 3 8 /s 4 1 4 67 - 0 2 1 - 2 3 7 3 6 - 2 | w ww . na tu r e . c o m /n atu r e c o m mu n i ca ti o n s
3

A R T I C L E
N AT URE CO M MU N ICAT I ON S | ht t ps: // doi. org /1 0 .10 38/ s41 46 7 -02 1 -2 3 7 36- 2
a/
H3CO
B2Pin2
H3CO
H3CO
N
[
Ir(OMe)(COD)]2
N
CuCl2
N
dtbpy
O
O
MeOH/H O
(4:1)
N
H
O
N
H
2
H
H
N
H
H
12
Cl
OCH3
B
OCH3
H CO
OCH3
dry heptane
H CO
3
H3CO
O
O
3
70% for 2 steps
1
2
>
80% by NMR
Mitragynine
H3CO
the only isomer detected
N
H
CuBr2
b/
O
MeOH/H2O
(4:1)
65% for 2 steps
N
H
H3CO
Br
OCH3
N
C-H borylation/
silylation
H3CO
OCH3
no Product
3
N
H
PG: TIPS, PMB
PG
O
OCH3
N-protected mitragynine
c/
H3CO
H3CO
Br
N
NBS
N
no Product
O
O
N
H
H
N
H
OCH3
H3CO
OCH3
H3CO
NBS, 0 oC, TFA, CH2Cl2
0%
3
6
Fig. 2 Examination of direct functionalization of mitragynine indole nucleus. a C –H bor ylation of M G pr ovid es e xc lu sively C12 - de riv ativ e s c onsi st en t wi t h
i n do l e di r ec t i ng e f f ec t s ( C7 i n i ndol e numb e ri ng). T he yi e l ds ar e a g e ne ral ra ng e bas ed on >10 re pe t i t ions by t wo e xpe r ime nt e rs. b T he posi t ional se le ct i vi t y
c a n no t be r e -r out e d t o C11 fu nct i ona l i zat i on v i a pr ev i ousl y r e por te d T IP S pr ot e ct i on of t he ind ole nit r og e n. c T he C3-i ndole hal og en ation–re a rra ng e men t
p ro c es s5 3 is no t a pp li c a bl e t o M G. Br omi na ti on und er ac i di c co ndit i ons g ave 12 - Br -M G as t he maj or pr oduc t .
H3CO
O
H3CO
H3CO HO
N
NaBH4
N
N
oxone
Zn(OTf)2
O
N
H
H
mitragynine
N H
H
O
CF3COOH
RT, 50% yield
NaHCO3
acetone
0% yield
N
H
toluene, 110 o
40% yield
C
OCH3
H3CO
OCH3
O
5
H3CO
7OH
H3CO
dihydromitragynine
mitragynine pseudoindoxyl OCH3
1
. Ir-cat. borylation
a, b
c
Ir-cat. borylation
1. Ir-cat. borylation
d, e
2
. CuX2
2. CuX2
partial C11 borylation
complex mixtures
no borylation
C12-borylation
Fig. 3 Catalytic borylation of MG analogs with reduced, oxidized, or rearranged indole nucleus. a B
2
P in
2
, dt bpy/M e
4
-phe n , [I r(CO D)(O M e )]
2
, 65 °C , d ry
1
h ep ta n e; 5 0% c on ve r s ion t o 1 1- bor on at e e st e r by H N MR. b CuB r
2
,
M
e
O
H
/
H
2
O
(
4
:
1
)
,
8
0
°
C
;
u
s
i
n
g
t
h
e
c
r
u
d
e
b
o
r
y
l
a
t
i
o
n
m
i
x
t
u
r
e
,
p
r
o
v
i
d
e
d
m
i
x
t
u
r
e
s
o
f
1
1
m
i
t
r
a
g
y
n
i
n
e
,
2
,
3
-
d
i
h
y
d
r
o
m
i
t
r
a
g
y
n
i
n
e
(
m
a
j
o
r
p
r
o
d
u
c
t
s
,
~
6
0
%
b
y
H
N
M
R
)
a
n
d
1
1
-
B
r
-
d
i
h
y
d
r
o
m
i
t
r
a
g
y
n
i
n
e
a
n
d
1
1
-
B
r
-
M
G
(
m
i
n
o
r
p
r
o
d
u
c
t
s
,
~
3
0
%
b
y
H
N
M
R
)
.
c B
2
P
i
n
2
,
d
t
b
p
y
/
M
e
4
-
p
h
e
n
,
[
I
r
(
C
O
D
)
(
O
M
e
)
]
2
,
6
5
°
C
,
d
r
y
h
e
p
t
a
n
e
;
n
o
b
o
r
y
l
a
t
i
o
n
p
r
o
d
u
c
t
w
a
s
d
e
t
e
c
t
e
d
.
d
B
2
P
i
n
2
,
d
t
b
p
y
/
M
e
4
2
1
C
,
d
r
y
h
e
p
t
a
n
e
;
c
o
m
p
l
e
t
e
c
o
n
v
e
r
s
i
o
n
t
o
C
1
2
-
b
o
r
o
n
a
t
e
b
y
H
N
M
R
)
.
e
C
u
B
r
2
,
M
e
O
H
/
H
2
O
(
4
:
1
)
,
8
0
°
C
;
n
o
r
e
a
c
t
i
o
n
.
also dramatically alters the shape of the entire alkaloid scaffold.
We reasoned that by converting the indole nucleus to an unusual
We crystallized MG-EG from methanol and confirmed its 3D aniline derivative, the directing effects of the nitrogen would be
structure (Fig. 4b, CCDC 1905559), which revealed the chair-like diminished while its steric effects would gain importance, resulting
conformation of the dioxane ring and the propeller-like in functionalization of the unhindered C11 position (after
arrangement of the three rings converging on the C–C bond of catalyst–ligand optimization). Boronates 4a and 4b were converted
4
8
the former indole ring.
to the bromo derivatives 5 and 6, respectively , using copper(II)
When we applied the catalytic borylation protocol (with the bromide. Thus, this sequence provided two 11-substituted MG-EG
dtbpy ligand) to MG-EG, a 1:1 ratio of C11 (4a) and C12 (4b) intermediates—boronate 4a and bromide 5—with versatile syn-
products was observed (Fig. 4c), clearly confirming the different thetic potential.
reactivity property of this chemotype, and unlocking the
possibility of optimizing the reaction conditions to favor C11
Preparation of C11-substituted analogs of mitragynine and
related scaffolds. The 11-boronate ester (4a) was converted to the
compounds 11-Cl-MG-EG (13) and 11-OH-MG-EG (14) using
functionalization.5
Table 3) showed that 3,4,7,8-tetramethyl-1,10-phenanthroline
Me -phen) as the ligand, together with B Pin and [Ir(COD)
2,60,61
Indeed, ligand screening (Supplementary
(
48,62
4
2
2
the appropriate substitution methods (Fig. 5a)
. For the pre-
OMe] in heptane at 65 °C, gave the 11-borylated product (4a) as
2
paration of additional derivatives, the 11-Br-MG-EG (5) served as
the key intermediate enabling preparation of compounds 8–12 in
the major product with ratio >16:1 of C11 (4a): C12 (4b)
products (Fig. 4c, confirmed by the subsequent bromination).
63
64
one synthetic step (Fig. 5a), where X = I (8) , Me (9) , Ph
4
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
(
2
0
2
1
)
1
2
:
3
8
5
8
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
2
1
-
2
3
7
3
6
-
2
|
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

NA TUR E COMM UNICAT IO NS | ht t ps ://doi .or g / 10 .1 0 38/ s 41 46 7 -02 1- 23 7 36-2
A R T I C L E
a/
b/
H3CO
O
H3CO
N
OH
O N
HO
O
N
H
H
O
N
H
H
PIFA, CH3CN
70% yield
OCH3
H3CO
OCH3
H3CO
MG-EG (4)
mitragynine
MG-EG
LIGAND (Me4-phen)
c/
H3CO
O
H3CO
O
N
N
MG-EG (4)
O N
CuBr2
O
N
B2Pin2
Ir(OMe)(COD)]2
11
O
MeOH/H2O
(4:1)
60-70%
O
[
O
N
H
H
N
H
H
B
O
Br
dry heptane
H CO
OCH3
OCH3
3
H3CO
for 2 steps
5
4a
C11:C12 > 16:1
[
Ir(OMe)(COD)]2
H3CO
O
O N
H3CO
O
4
a
O
O N
N
H
H
N
N
CuBr2
MeOH/H2O
4:1)
5%
for 2 steps
C11:C12 = 1:1
1
2
B
5
OCH3
O
O
O
H3CO
N
H
H
LIGAND (dtbpy)
(
4b
Br
OCH3
6
H3CO
6
Fig. 4 C11-selective C–H borylation via conversion of MG to mitragynine-ethylene glycol adduct (MG-EG). a O ne -st e p syn th e sis of M G-E G, a st a bl e
d er i va t iv e of MG. b Th e 3D st r uc t ur e of M G-E G w as co nfir me d by X -ra y c rys t all og rap hy. O RT E P re pr e se nt at io n of M G-E G st r uct ur e is shown wi t ho ut
h y d ro ge n a t oms f or c la r i ty . c C–H bor y la ti on of M G-E G g iv e s a m ixt ur e of C11 - and C12 - boronat e s and t he subs e que nt f unc t ional iz at ion pr oduc ts. T he h ig h
1
C1 1 s el e ct i vi t y wa s a c hi ev e d by l ig an d opt i mi z ati on. T he r at i o w as de t e rmi ned by H N M R of c rude re a ct i on mix t ure s. T he yi e lds, g iv en as a g en er al ra ng e ,
a r e b as e d on > 20 r e pe ti t ions by 3 e x pe r ime n te rs, on a sc al e up t o 10 0 mg of subs t rat e .
6
5
66
67
(
10) , CONH (11) , and CN (12) . Considering the com-
2
Table 1 Binding affinities of 7OH analogs at the mouse
opioid receptors.
plexity of these compounds, the yields were satisfactory and more
than sufficient to produce practical amounts of the compounds
for preliminary pharmacological evaluation. However, the
synthesis of 11-F-MG-EG (7) proved more difficult compared to
its other halogen relatives. Relevant reported procedures were not
effective in this context; for example, direct fluorination of 11-
boronate ester 4a using copper-mediated fluorination with
Compound
K
i
± SEM (nM)a
mMOR
mKOR
mDOR
7
O H
2 1. 5 ± 0 .8
13. 7 ± 1. 0
2 7 .1 ± 1. 1
32 . 4 ± 1. 4
11 9.0 ± 2 .1
2 1. 0 ± 3. 2
30 .7 ± 11 .4
81. 1 ± 7 .1
88. 5 ± 9.9
35 . 8 ± 2 .0
4 7 .2 ± 2 .4
5 7 .7 ± 6.7
11 -F- 7 OH (22)
11 -Cl- 7 OH ( 23)
11 -B r-7O H ( 24)
t
68,69
(
BuCN) CuOTf or Cu(OTf) py failed
, so did an indirect
2
2
4
7
0
sequence of stannylation–fluorination . Deoxyfluorination of the
corresponding phenol 14 using Phenofluor™ Mix was also
7
1
a
A l l dat a poi nt s r e pr e se n t me a n ± SE M ( nM ) of n = 3. [ 1 25 I]B Nt xA w as us e d as the sta nda r d
unsuccessful . Eventually, 11-F-MG-EG (7) was synthesized
from bromide 5 via the sequence of stannylation and fluorination
r adi ol a be l e d l i ga nd98
.
72
(
Fig. 5b) . The triflate 14a represents an alternative intermediate
(
to the corresponding bromide 5) for further functionalization;
for example, (1) 11-F-MG-EG (7) was also synthesized from
triflate through a series of stannylation and fluorination (Fig. 5c)
and (2) carboxamide 11 was prepared in 57% yield (versus 35%
via the bromide, Supplementary information).
In summary, we developed access to 11-substituted analogs in
three MG-related molecular series, compounds previously
inaccessible via electrophilic substitution or other approaches,
providing the means for a systematic SAR exploration of this
position in multiple MG-type scaffolds.
The ethylene glycol moiety can be readily removed under mild
reductive condition in one step to yield 11-substituted MG
analogs (15–21, Fig. 5d) in moderate-to-good yields. Our strategy
enables functionalization of the C11 position with a wide range of Neuropharmacology: in vitro modulation of opioid receptors
substituents starting directly from the MG natural product, thus by the 11-substituted 7OH analogs. With efficient access to the
permitting systematic exploration of SAR at this position.
C11 analogs, we examined the effect of C11 substitution on the
To convert the MG analogs to the corresponding 7OH series, opioid receptor pharmacology of MG analogs. We focused the
®
we optimized a set of conditions using OXONE in acetone initial set of biological assays on the 7OH series, as 7OH is an
Fig. 5d)7 . This procedure was successfully applied to the active metabolite of MG and exhibits an order of magnitude
3,74
(
1
1
7
1-substituted MG analogs, as demonstrated by preparation of greater potency compared to MG as an MOR agonist . First, we
the 11-halo derivatives’ 22–24 analogs in yields around 50%. This examined the 11-halo analogs of 7OH in radioligand binding
oxidation protocol is superior to the previously reported studies to assess the affinity for the opioid receptors (Table 1). We
approaches, namely PIFA gives complex mixtures of products found that the C11 halogen modulated affinity across all three
and a lower yield of 7OH, while lead tetraacetate involves a toxic opioid receptors: the 11-F compound 22 exhibited greater affinity
7
5
heavy metal and requires a second hydrolysis step .
compared to the parent 7OH, but this gain in binding was
N A TU RE C OMM U N IC AT I ON S |
(
2
0
2
1
)
1
2
:
3
8
5
8
|
h
t
t
p
s
:
/
/
d
o
i
.
o
r
g
/
1
0
.
1
0
3
8
/
s
4
1
4
6
7
-
0
2
1
-
2
3
7
3
6
-
2
|
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 AT URE CO M MU N ICAT I ON S | ht t ps: // doi. org /1 0 .10 38/ s41 46 7 -02 1 -2 3 7 36- 2
a/
1. B2Pin2
H3CO
O
H3CO
O
[
Ir(COD)OMe]2
Me4-phen
O N
O N
a - e
MG-EG (4)
O
O
2
. CuBr2
MeOH/H2O
4:1)
0-70%
for 2 steps
Br
N
H
H
N
H
H
X
OCH3
H3CO
OCH3
(
H CO
5
3
6
8, X = I, 59%a
, X = Me, 60%
0, X = Ph, 81%
1, X = CONH2, 35%
2, X = CN, 60%
b
9
1
1
c
H3CO
O
1
. B2Pin2
d
O N
[
Ir(COD)OMe]2
Me4-phen
1
e
O
N
H
H
X
2
. (f-g)
OCH3
H3CO
f
1
1
3, X = Cl, 60-70%
4, X = OH, 50%
g
b/
H3CO
O
H CO
O
I) Stannylationi
3
O N
II) Fluorinationj
O N
O
O
N
H
H
36% over two steps
N
H
Br
F
H
OCH3
OCH3
H3CO
7
H CO
5
3
c/
H3CO
O
H CO
O
3
H3CO
O
O N
h
O N
H
O N
H
(i, j), d
O
O
N
H
H
O
N
H
HO
N
H
X
TfO
OCH3
OCH3
i, j
H3CO
OCH3
3
H CO
H3CO
4a (70%)
14
1
7, X = F, 40% (over two steps)
1, X = CONH2, 57%
d
1
d/
oxone
acetone, NaHCO3
H3CO
O
H3CO
H3CO HO
1
. NaBH3CN
AcOH, RT
O N
N
N
O
O
O
N
H
H
N
H
H
N
H
X
X
X
OCH3
OCH3
OCH3
H3CO
H3CO
H3CO
MG-EG series
Mitragynine series
7OH series
MG-EG, X = H
MG, X = H
7OH, X = H, 50%
22, X = F, 52%
23, X = Cl, 50%
7
1
5
9
1
, X = F
15, X = F, 70%
16, X = Cl, 70%
17, X = Br, 80%
18, X = Me, 60%
19, X = Ph, 56%
20, X = CONH , 35%
21, X = CN, 34%
3, X = Cl
, X = Br
, X = Me
0, X = Ph
24, X = Br, 45%
11, X = CONH2
2
12, X = CN
Fig. 5 The MG-EG scaffold is compatible with a wide array of functionalization chemistries. a, b, c Sy nt he sis of C11 -subs tit ut e d ana log s of M G- EG . ( a)
Na I, C uI, N, N ’-d ime t hy l et hyl e ne dia mi ne , 1 10 °C, 46 h. (b ) P d (d ba) , X P hos, D ABA L-M e , T H F, 60 ° C, 2 h. (c ) P d(dpp f)Cl · CH Cl , P hB (O H ) , C sOAc,
T HF, 70 °C, 7 h. ( d) P d( OAc ) , dpp f, Co (CO) , D IP EA, N H Cl , i mida zo le, dio xane , 90 ° C, 15 h. (e ) P d (dba ) , Zn dust , Zn(C N) , D M F, RT , 3
O (4 :1) , 80 °C, 1 2 h. (g ) H , T HF, 0 ° C t o RT , 30 min. (h) P hN T f , P d (P P h
2
3
3
2
2
2
2
2
2
8
4
2
3
2
, [H P t Bu
3
]B F
Sn )
4
n
h
.
(
f
)
C
u
C
l
2
·
2
H
2
O
,
M
e
O
H
:
H
2
2
O
2
2
,
D
I
P
E
A
,
D
M
F
,
5
0
°
C
,
o
v
e
r
n
i
g
h
t
.
(
i
)
(
B
u
3
2
3
4
°
L
i
C
l
,
d
i
o
x
a
n
e
,
1
0
0
C
,
o
v
e
r
n
i
g
h
t
.
(
j
)
A
g
O
T
f
,
F
-
T
E
D
A
-
P
F
6
,
R
T
,
2
0
m
i
n
.
T
h
e
y
i
e
l
d
s
a
r
e
b
a
s
e
d
o
n
>
2
r
e
p
e
t
i
t
i
o
n
s
.
d
R
e
d
u
c
t
i
v
e
r
e
m
o
v
a
l
o
f
e
t
h
y
l
e
n
e
g
l
y
c
o
l
.
T
h
i
s
s te p r end e rs 11- su bs t i tu t ed M G an al og s, w hic h i n t ur n ca n be ox idi z e d t o t he c orr e spondi ng 7 O H de ri va ti ve s. T he yi e lds ar e av e rag e bas ed on a t l ea st 3
r ep eti t ion s.
progressively lost as the halogen became larger (compounds 23 decreasing amounts of cAMP as a result of conformational
and 24, Table 1). This effect was most pronounced at kappa- changes in the sensor. cAMP levels were raised prior to the assay
opioid receptor (KOR), where the affinity of 22 was six times by addition of forskolin as described previously (Fig. 6a, illus-
1
,76
greater than that of the parent compound 7OH. The compound tration created by the authors)
2 also had more than 2-fold greater affinity for delta-opioid In the cAMP assay, 7OH was a potent agonist with relatively
receptor (DOR) compared to 7OH. The effect of the C11 halogen high efficacy (E = 70%, compared to DAMGO ([D-Ala , N-
.
2
2
max
4
5
on MOR binding was subtle across the halogen series. Since the Me-Phe , Gly -ol]-enkephalin), EC = 48 nM, Fig. 6b, d).
50
parent compound 7OH acts as a potent agonist, we examined the Introduction of the 11-fluoro substitution in compound 22
functional modulation of mouse MOR (mMOR) activation in dramatically reduced the efficacy, rendering a low-efficacy agonist
living cells. Specifically, we examined the signaling consequences (Emax ~ 21%) (Fig. 6b, d). For the 23 and 24 derivatives, the
of G protein activation via detection of the downstream signaling agonist activity at mMOR was nearly abolished (Emax = 13% for
molecule cyclic adenosine monophosphate (cAMP). cAMP is 23 and Emax < 10% for 24). Accordingly, 11-halo substituents in
formed from adenosine triphosphate (ATP) by adenylyl cyclase the 7OH scaffold do not interfere with binding of these
(
AC), which is inhibited by activation of the MOR and its asso- compounds to mMOR, but profoundly modulate the signaling
ciated G proteins. Changes in the level of cAMP were measured efficacy at this receptor.
using a bioluminescence resonance energy transfer (BRET)
To determine if the efficacy modulation trend holds at
functional assay. The BRET signal increases in response to the human receptor, we examined the 11-X-7OH series in
6
N AT U RE CO M MU NI C A TI O N S |
( 2 0 21 ) 1 2 : 3 85 8 | h tt p s : / /d o i . o r g /1 0 . 1 0 3 8/ s 41 4 67 - 0 2 1 - 2 3 7 3 6- 2 | w ww . na tu r e . c o m /n atu r e c o m m u n ic a tio n s

NA TUR E COMM UNICAT IO NS | ht t ps ://doi .or g / 10 .1 0 38/ s 41 46 7 -02 1- 23 7 36-2
A R T I C L E
Fig. 6 Activity of 7-hydroxymitragynine (7OH) analogs at the mu-opioid receptor (MOR). a Conc e pt ual re pr e se nt at i on of t he M O R CAMY EL B RE T
a s sa y . To me a s ur e G pr o t e in ac t i va t i on, M OR (l i gh t g re e n) w as coe x pre sse d wi t h G pr ot e in subu nit s G αoB , β , γ , and t he B RE T CAM Y EL se nso r. Fo rsko l in
1
2
a c ti va t es a de ny l y l c y c la s e (A C), w hic h co nv er t s AT P t o cy cl i c ad e nosi ne monoph osphat e (c AM P ). O n ac t iv at ion of mM O R by li ga nd, α subu nit i nh ib i t s AC ,
r es u l ti ng in a de c r ea se of cA MP ac cu mul at i on. T he CAM YE L se nso r is c ompri se d of c it ri ne (mus ta rd ye l low) and Renilla luc i fe r ase (l ig ht blu e) wi t h h uma n
Ep a c 1 ( blu e li nk e r ) in t er s p ac ed be t we e n t he m. b Ag oni st ac t i vi t y of 7 O H at mM O R, pr es e nt ed as a pe rc e nt ag e of c AM P inh ibit i on pr oduc ed by t h e p osi t i ve
2
4
5
c o nt ro l DAMG O ( [D-Al a , N -M e -P he , Gl y - ol ]-e nk ep hal in ); c Ag onis t ac t iv it y of 7 O H ana log s at hM O R; c urv e s re pr e se nt t he av e rag e of n = 3 ,
i n de p end e nt e xp er i me nt s w it h e r r or ba r s r e pr e se nt i ng ± S EM . d Fun ct i onal ac t iv it y of 7 O H ana log s at mM O R as de t e rmi ned in CAM Y EL B RE T a ssay s.
a
e Fu n c ti on al a c t iv it y of 7 OH an al og s at hM OR as de t e r mine d i n CAM Y EL B RE T ass ays [ Ag onis t ac t iv it y ind ic at e d by E C 5 0 va lue s, max imal effica cy
i s ex pr e s se d a s ma x ima l i nhi bi t or y e ff e ct on cA MP l ev e ls r e la t i v e t o D AM GO ( Ema x = 10 0 % – c AM P (min) %). bAll dat a poi nt s re pr e se nt −log E C5 0 ± S E
(EC5 0 nM ) f or n = 3 r e pe t it i ons].
77
hMOR-CAMYEL assay (Fig. 6c) using the same experimental overexpression and downstream signaling amplification . To
setup as illustrated above (Fig. 6a). The parent compound 7OH examine this effect for 7OH and its analogs, we adopted an assay
was potent and highly efficacious (Emax = 96%, compared to the that detects an active conformation of MOR, using the Nb33
reference ligand DAMGO, EC50 = 16.4 nM, Fig. 6c, e), whereas nanobody sensor, and thus limits the signaling-related amplification
1
1-F-7OH (22) was a partial agonist with markedly reduced (Nb33 BRET assay using a luciferase-tagged MOR and Venus-
7
8
efficacy (Emax = 57%, EC50 = 29.4 nM, Fig. 6c, e). The 23 and tagged Nb33, see Supplementary information) . This assay was
7
9
2
4 derivatives showed very low efficacy (Emax = 23% for 23 and initially described by Stoeber et al. and has recently been
7
8
Emax = 14% for 24). Hence, we can conclude that the trend in rigorously validated by Gillis et al. via comparison to a number of
efficacy signaling modulation in the halogen 7OH series is independent signaling assays, and calibrated by direct comparison
comparable between human and mouse MOR.
of several opioid drugs including well-established partial agonists
The measured efficacy of receptor agonists depends greatly on (buprenorphine and morphine) and experimental opioids. Using
the receptor reserve pools and the extent of downstream signaling mMOR and hMOR Nb33-BRET assays, we found that morphine
amplification of each functional readout. Under the conditions and buprenorphine act as partial agonists, in comparison to
typically used in in vitro cell-based assays, such as the cAMP BRET DAMGO, with markedly different signaling efficacies (Fig. 7).
assay discussed above, the efficacy of partial agonists appear While morphine appears as a partial agonist with relatively
exaggerated owing to large receptor pools produced by receptor high efficacy (Emax ~ 67% (mMOR) and 72% (hMOR), Fig. 7),
N A TU RE C OMM U N IC AT I ON S |
(2 0 2 1 ) 1 2 : 3 85 8 | h t t p s :/ /d o i . o r g /1 0 . 1 0 3 8 /s 4 1 4 67 - 0 2 1 - 2 3 7 3 6 - 2 | w ww . na tu r e . c o m /n atu r e c o m mu n i ca ti o n s
7

A R T I C L E
N AT URE CO M MU N ICAT I ON S | ht t ps: // doi. org /1 0 .10 38/ s41 46 7 -02 1 -2 3 7 36- 2
Fig. 7 Receptor activation by 7OH series in comparison to known opioids in BRET-based Nb33 recruitment assay at mMOR and hMOR. I nt r in si c
r ec ep tor a c t iv a t io n e ffic a cy w as de t e r mine d by t he e x t en t of r e cr uit me nt of t he c onf orma ti onal ly se le c t iv e nano body N b33 t o hM O R or mM OR f us ed a s
i ts C te r min us t o Na no L u c, as me as ur e d by BR ET . P os it i v e co nt rol s use d ar e : D AM GO , bupr e norphine , mor phine , and par e nt 7 O H . a Ag onist a ct i vi t y o f
11 -F- 7OH (22) a s c omp a re d t o pos i ti v e co nt r ol s at mM OR. b Ag onis t ac t iv it y of 11 -F- 7 OH (22) as c ompar e d t o posi t ive c ont rol s at hM O R. C urve s
r ep re se nt t he a v er a ge of n = 3, i nde pe nde nt e x pe r iment s w it h e r ror bar s re pr e se nt in g ± SE M . c Fun ct i onal ac t iv it y of 11 -F- 7 OH ( 22) and posi t ive co nt r ol s a t
a
b
mMO R a nd hM OR in Nb33 BR ET As say [ Ag oni st ac t i vi t y i ndi cat e d by E C 5 0 va lue s, max imal e f fic ac y is e xpr e sse d re l at iv e t o D AM GO ( Ema x) . Al l d at a
p o in t s r e pr es e nt −lo gEC5 0 ± S E (E C50 µM ) fo r n = 3 r e pe t it i ons].
Fig. 8 Antagonist activity of 7OH analogs at mMOR and hMOR. T o ass ay G pr ot e in ac t iv at ion, mM O R and hM O R wa s c oe xpr e sse d wi t h G p rot e in
s u bu n i t s GαoB , β , γ , a nd t he BR ET CAM YEL se ns or . Ant a go ni sm wa s me asur e d by t he inh ibit i on of D AM GO’s e f fe c t on c AM P . a Compe t i ti ve a nt a g oni st
1 2
a c ti vi ty of 7 OH a na l ogs at mM OR; pos i ti v e co nt r ol = na lo xo ne. b Compe t i ti ve ant a goni st ac t iv it y of 7 O H ana log s at hM O R; posi t ive c ont rol = n al ox on e.
Cu r ves r e pr es e nt t he a v e r ag e of n = 3, i nde pe nde nt e x pe r ime n ts wi t h e rr or bar s re pr e se nt in g ± SE M . c Fun ct i onal ant a goni st ac t iv it y of 7 O H a na lo g s a t
mMO R a s de t er mi ne d in CAM YEL BR ET as sa ys . d F unc ti on al an t ag onis t ac t iv it y of 7 O H ana log s at hM O R as de t e rmine d in CAM Y EL B RE T a ssay s
[cA n ta go ni st a c t iv it y in di c at e d by IC 50 v al ue s fo r i nhi bi t io n of a re f e re n ce ag oni st (D AM GO), al l dat a poi nt s re pr e se nt −log I C5 0 ± SE (I C5 0 µM ) f or n = 3
r ep eti t ion s].
8
N AT U RE CO M MU NI C A TI O N S |
( 2 0 21 ) 1 2 : 3 85 8 | h tt p s : / /d o i . o r g /1 0 . 1 0 3 8/ s 41 4 67 - 0 2 1 - 2 3 7 3 6- 2 | w ww . na tu r e . c o m /n atu r e c o m m u n ic a tio n s

NA TUR E COMM UNICAT IO NS | ht t ps ://doi .or g / 10 .1 0 38/ s 41 46 7 -02 1- 23 7 36-2
A R T I C L E
b/
a/
1
00
1
00
1
1-F-7OH (22) (5mg/kg)
8
0
8
6
4
2
0
0
0
0
0
Vehicle
60
40
2
0
0
3
0
60
90
120
Vehicle
Time (min)
mg.kg, sc.
7OH, 2
1
1-F-7OH, 5 mg/kg, sc
1
1-F-7OH, 11 10 - Fm - 7g O/ k Hg , 2s 5c mg/kg, Ms co rphine 10 mg.kg, sc.
c/
d/
1
00
1
1-Cl-7OH (23)
(
5 mg/kg)
100
8
6
4
2
0
0
0
0
0
1
1-Cl-7OH (23)
(25mg/kg)
50
0
0
30
60
90
120 150
0.1
1
10
Log [drug] (M)
Time (min)
e/
100
Wild Type *
MOR KO
ns
50
0
0.1
1
10
Log [drug] (M)
Fig. 9 Antinociceptive effects of 7OH analogs in tail-flick assay in mice. a T ime c ours e of t he ant i noc ic e ptiv e e f fe c t of 22 (5 mg /k g . s. c. , n = 11 C D1 mi ce )
a n d ve hi c le (n = 6 CD1 mi ce ) i n t he t ai l -fli ck as sa y. E ach poi nt r e pre s e nt s me an ± SE M . % M P E, a pe rc e nt ag e of t he max imum poss ibl e e f fe c t as se t i n t h is
a s sa y ( s ee S up ple ment a r y i nfo r mat i on) . b T ai l -fli ck dos e–r e spo ns e f or t he hig h- dose ra ng e (5 , 10 , and 2 5 mg /k g , s. c. ; n = 8 CD 1 mic e e ac h g roup) o f 22, i n
d i re c t c omp a r is on t o c on tr ol s 7 OH an d mor ph in e. CD 1 mi ce ( n = 10 ), at pe ak ana lg e sic t ime poi nt . (V eh ic le ( −5 .4 , 7 .8 , 4 .0 ), 11 -F- 7 OH 5 mg /k g ( 17 .3 , 2 2 .4 ,
3
8
.
3
)
,
1
1
-
F
-
7
O
H
1
0
m
g
/
k
g
(
1
3
.
7
,
3
9
.
0
,
2
7
.
1
)
,
1
1
-
F
-
7
O
H
2
5
m
g
/
k
g
(
2
3
.
1
,
4
4
.
2
,
3
6
.
8
)
,
7
O
H
2
m
g
/
k
g
(
5
2
.
1
,
1
0
0
.
0
,
1
0
0
.
0
)
,
M
o
r
p
h
i
n
e
1
0
m
g
/
k
g
(
3
2
.
8
,
1
0
0
,
1
0
0
)
)
c
T
i
m
e
c
o
u
r
s
e
o
f
2
3
a
t
2
d
o
s
e
s
(
5
a
n
d
2
5
m
g
/
k
g
,
s
.
c
.
,
n
=
8
C
D
1
m
i
c
e
e
a
c
h
)
.
E
a
c
h
p
o
i
n
t
r
e
p
r
e
s
e
n
t
s
m
e
a
n
±
S
E
M
.
d
D
o
s
e
–
r
e
s
p
o
n
s
e
c
u
r
v
e
s
i
n
m
o
u
s
e
t ai l -fli ck a s s a y f or t he pa r en t 7 OH (s .c.) an d 22 (s .c.) , 1 5 mi n t i me poi nt , C5 7 B L/ 6 mic e (n = 10 ). E ac h poi nt re pr e se nt s me an ± SE M . e Ant in oc ic ept i ve
ef f ec t of 11- F-7 OH (22) is M OR de pe nde n t. D ose–r e spo nse cu r ve s of 22 (s. c .), in t he t ai l- flic k ass ay wa s e valua t e d 15 min post dr ug admi nist r at i on i n WT ,
C5 7 m ic e (n = 8 ) a nd MO R KO mi ce ( n = 5 ). E ach poi nt r e pr e se nt s me an ± SE M . O ne -wa y RM AN O VA show s st at i st ic all y si gnific ant re s ult f or WT * mi ce
n
s
wi t h *p = 0. 00 6 a nd s t a ti st i ca ll y non -s ig nifica nt r e sul t (M OR KO ) wi t h p = 0 .8 32 f or M O R KO mic e .
buprenorphine shows much lower efficacy (Emax ~16% (mMOR)
and 21% (hMOR)), results consistent with the previous report . into MOR signaling efficacy measurements. The Nb33 BRET
Remarkably, we found that 7OH exhibits low “intrinsic” efficacy assay, with limited signaling amplification, positioned 7OH close
The two assays employed provide complementary windows
7
8
(Emax ~14% (mMOR) and 22% (hMOR), Fig. 7) comparable to that to buprenorphine in terms of the broad range of signaling
of buprenorphine. In direct comparison, the efficacy of 22 is further efficacies of different opioids. Despite the narrow dynamic range
reduced to near the detection limit of the assay using both mMOR of this assay for low-efficacy compounds, it showed that
and hMOR (Fig. 7).
fluorination of the 11-position further impaired the signaling
N A TU RE C OMM U N IC AT I ON S |
(2 0 2 1 ) 1 2 : 3 85 8 | h t t p s :/ /d o i . o r g /1 0 . 1 0 3 8 /s 4 1 4 67 - 0 2 1 - 2 3 7 3 6 - 2 | w ww . na tu r e . c o m /n atu r e c o m mu n i ca ti o n s
9

A R T I C L E
N AT URE CO M MU N ICAT I ON S | ht t ps: // doi. org /1 0 .10 38/ s41 46 7 -02 1 -2 3 7 36- 2
efficacy of 7OH to near the limit of detection. In the amplified
Thus, the in vitro MOR pharmacological profile of the 7OH
cAMP CAMYEL BRET assay, with a greater dynamic range in the analogs translates well to the in vivo tail-flick test, an established
low end of efficacy readout, 22 exhibits greatly diminished but physiological readout for MOR agonists that measures anti-
readily detectable agonist signaling as compared to 7OH, whereas nociceptive effects of drugs to painful thermal stimuli. We next
the efficacy was essentially lost for the much less efficacious set out to rigorously confirm these results by: (1) using a different
compounds 23 and 24.
and widely used mouse strain, C57BL/6J mice; (2) performing the
To correlate the very low agonist efficacy with antagonism, we experiments in a different location and research group; and (3)
next determined the antagonist activity of the 11-X-7OH series in determining a full dose-curve for both the control (7OH) and the
a CAMYEL antagonism assay at mMOR (Fig. 8a, c) and hMOR compound 22 (Fig. 9d). 7OH showed a potent and high-efficacy
(
Fig. 8b, d) compared to the positive control naloxone. As analgesic effect (ED50 = 0.25 (0.20–0.31) mg/kg, 95% CI,
expected, the compounds were able to inhibit the response analgesic efficacy normalized to 100%, comparable to morphine),
1
7
elicited by the standard full agonist DAMGO at hMOR and consistent with previous results , while 22 also elicited a potent
mMOR down to the levels predicted based on their agonist analgesic effect (ED50 = 0.18 (0.04–0.78) mg/kg, 95% CI), but
efficacy as determined above.
with dramatically attenuated analgesic efficacy (Emax = 21%,
To complete the examination of this series at the opioid Fig. 9d). Thus, 22 is equipotent to 7OH but exhibits low efficacy
receptors, we performed the analogous cAMP BRET assays with in antinociceptive effects, whereas 7OH is comparable to
mDOR and hDOR, and rat KOR (rKOR) and hKOR. The parent morphine in efficacy.
7
7
OH was found to act as a partial agonist at both mDOR (Emax
=
To confirm that the low-efficacy analgesia is driven by MOR,
2%, EC50 = 71.4 nM) and hDOR (E = 45%, EC50 = 99.9 nM, and no other targets or mechanisms (e.g., other opioid receptors
max
8
0
Supplementary Fig. 7) . Again, 22 showed considerable reduc- or non-opioid off-targets), we performed the mouse tail-flick
tion in signaling efficacy at the receptors of both species, when assay in wild-type (WT), and MOR knockout (MOR KO) mice
compared to 7OH, without a major shift in potency (Emax = 41%, with the same genetic background (Fig. 9e). Despite the low range
EC50 = 102.1 nM at mDOR; Emax = 26%, EC50 = 150.8 nM at of the efficacy readout afforded by 22, we were able to show that
hDOR,). The 23 and 24 compounds had very low efficacy at the dose-dependent effect of 22 was statistically significant in WT
mDOR and hDOR (Supplementary Fig. 7). Thus, the C11 mice, but not in MOR KO animals.
halogen trend described for MORs was also observed at DORs. In
These results support an explanatory model where the
contrast, this compound series showed no agonist activity at the analgesic effect of 11-F-7OH is largely driven by MOR, and the
KOR receptors (Supplementary Fig. 8). Finally, we selected one reduced analgesic efficacy is a consequence of low-efficacy MOR
representative compound of this series, 11-Br-7OH (24), and activation/signaling induced by this compound. Thus, 7OH and
tested it for antagonism at the KOR and DOR (Supplementary 11-F-7OH provide an excellent probe pair for examining the
Figs. 9 and 10) revealing full inhibition of reference agonists effects of relative signaling efficacy on therapeutic (desired) and
DPDPE ([D-Pen2 ]-enkephalin) for DOR and U50,488 for KOR. adverse effects, and the resulting preclinical measure of a
In summary, these pharmacological studies indicate an therapeutic window.
,5
important role for the C11 substitution in selective and profound
modulation of the signaling efficacy at MORs and DORs of the
MG-related scaffolds and provide a strong rationale for examin- Discussion
ing this effect in vivo.
Most recent estimates of kratom use, based on the tonnage
of imported kratom and survey reports of dosing ranges,
8
3
suggest 10–16 million users in the US alone . Numerous
1-Fluoro-7-hydroxymitragynine (22) is a low-efficacy partial anecdotal reports, including those collected by the United
1
agonist in vivo in mouse analgesia tests. We set out to examine States Drug Enforcement Administration (>23,000 publicly
8
how the gradual signaling efficacy modulation found in vitro, available reports) , point to remarkable medicinal effects of
within the 7OH series, would translate to in vivo effects in living kratom in self-medication for chronic pain, depression, anxiety,
rodents. There is a strong rationale for pursuing such studies, as substance use disorders, and other ailments. Inspired by these
partial MOR agonists may provide a path to safer opioid ther- clinical indicators, our laboratories have been studying the
apeutics (see “Discussion” below for more details). We recently chemistry and biology of kratom alkaloids with the long-term
reported that 7OH was a potent and efficacious analgesic in mice goal of developing drug leads based on these compounds. In
1
6,17
using a tail-flick assay (thermal nociception assay)
with a previous report by others
, consistent this context, the aromatic ring SAR of MG has not been
8
1,82
. On the basis of the pre- mapped systematically, as only limited series of analogs have
viously determined ED50 value for 7OH in mouse tail-flick assay been synthesized and studied (e.g., C10 analogs) , and the
ED50 = 0.57 (0.19–1.7) mg/kg, 95% CI) , and similar in vitro C11 position has not been examined at all. To address this task,
binding and signaling potencies of 7OH and 22 at MOR, we and to complement our total synthesis efforts , we set out to
selected a relatively high dose of 22 (5 mg/kg, subcutaneous (s.c.) examine the late stage functionalization of MG, in this study
administration) for the initial time course profiling of the focusing on the C11 position of the indole nucleus.
1
4,16
1
7
(
1
5
analgesic effect in CD1 mice (Fig. 9a). The analgesic effect peaked
Years ago, we formulated the ideas of C–H bond functio-
at 15 min post injection, indicating rapid bioavailability in vivo nalization and late-stage functionalization as general concepts
comparable to the parent 7OH, followed by a decay to a residual with transformational potential for the thought and practice of
3
0–36
analgesic effect that lasted for at least 120 min. Noteworthy is the chemical synthesis of carbon-based substances
. These
partial analgesic efficacy of <40% MPE (maximum possible effect) concepts have since been widely adopted in both academia and
in the tail-flick assay. To confirm that the maximum effect was industry and the C–H functionalization mindset, as well as
reached, we examined even higher doses at a 15 min time point, corresponding methods, are indispensable components of
3
7–40
namely 10 and 25 mg/kg (s.c.), which showed that indeed the today’s synthetic repertoire
maximal analgesic effect was reached already at the 5 mg/kg dose methods for systematic and programmable functionalization of
Fig. 9b). The control compounds, 7OH and morphine, elicited a heteroarenes, which are essential building blocks in medicinal
greater analgesic effect in the same mouse strain. In contrast, 23 chemistry, including indoles, pyrroles, pyrazoles, imidazoles,
. Our team has been developing
(
3
0–36
showed no analgesic effect at 5 or 25 mg/kg dose (Fig. 9c).
pyridines, and others
. In the present study, we focused on
1
0
N AT U RE CO M MU NI C A TI O N S |
( 2 0 21 ) 1 2 : 3 85 8 | h tt p s : / /d o i . o r g /1 0 . 1 0 3 8/ s 41 4 67 - 0 2 1 - 2 3 7 3 6- 2 | w ww . na tu r e . c o m /n atu r e c o m m u n ic a tio n s

NA TUR E COMM UNICAT IO NS | ht t ps ://doi .or g / 10 .1 0 38/ s 41 46 7 -02 1- 23 7 36-2
A R T I C L E
the benzene ring of highly complex indoles, in the context of system. Considering the recent surveys that document much
specific neuropharmacological goals.
Our approach provided a practical route to C11-analogs of opioids
improved safety of kratom compared to the prescription or illicit
8
3,93,94
, a further decrease of MOR’s G protein signaling
MG, 7OH, and MG-EG, from the natural product, and systematic efficacy of the key mediator of kratom’s opioid-like effects,
1
7
SAR exploration of the C11 position. The results indicate that 7OH , may lead to even safer compounds, and illustrates the
substitution of this position plays an important role with respect potential importance of the present work. Further, a recent report
to MOR signaling, specifically in modulating the efficacy of showed that 7OH attenuated alcohol intake in drinking mice and
receptor-triggered signaling events. Designing the receptor sig- that this effect was mediated by DOR activation, while 7OH
8
0
naling parameters of GPCR ligands a priori remains an unfulfilled exhibited rewarding effects on its own . Thus, the attenuated
potential of computational methods, and thus, a systematic MOR signaling efficacy of 22, while maintaining sufficient DOR
examination of functional characteristics of receptor ligands and signaling, may provide a therapeutic lead for OUD and other
probes still relies largely on chemical synthesis. The chemical substance use disorders with no or much diminished abuse
approach described here provides a means to advance our potential. That we see only partial analgesia for 22 and effective
examination and understanding of how molecular interactions of analgesia with 7OH suggests that the ideal efficacy may be
MG-type scaffolds with the opioid receptors underlie functional somewhere between that of these two analogs, which can guide
outcomes such as G protein signaling and maximal signaling subsequent synthetic efforts in search of the ideal balance between
efficacy.
There has been considerable excitement and hope for gen-
therapeutic effects and side effects.
In summary, we describe a systematic examination of late-stage
erating MOR agonists that bias its signaling toward G protein- functionalization of kratom alkaloids, which provided efficient
initiated pathways as a means for rational design of improved access to MG analogs and identified 11-F-7OH (22) as an
safety and therapeutic index of opioid analgesics (“G protein bias important lead compound for further investigations.
8
4
hypothesis” of opioids) . It was suggested that some of the
typical adverse effects of opioids such as respiratory depression or
Methods
reinforcing effects are linked to arrestin signaling. According to General procedure for synthesis of 11-boronate ester MG-EG (4a). Starting
this approach, biasing the MOR-triggered signaling toward the G material MG-EG (4) (50 mg, 0.11 mmol), [Ir(COD)OMe] (3.6 mg, 5.5 µmol, 5 mol
), 3,4,7,8-tetramethyl-1,10-phenanthroline (3.9 mg, 16 µmol, 15 mol%) and
Pin (111 mg, 0.44 mmol, 4 equiv.) were balanced into an oven dried vial. The
2
%
protein and away from arrestin would result in attenuation of the
B
2
2
8
4
adverse effects, relative to the desired analgesic activity . How-
ever, this hypothesis has been seriously challenged using both
vial was purged with argon, dry heptane (2.5 mL) was added under argon, and the
vial was sealed with a Teflon-lined screw cap and heated to 65 °C. The RM became
7
8,85–90
genetic and pharmacological tools
. An emerging alter- a dark red-brown solution after 5–15 min of heating. After 17–24 h, when LR-MS
indicated complete consumption of SM, the RM was concentrated to give the crude
native hypothesis is that partial agonism of MOR can lead to
boronate ester. This intermediate was immediately used to prepare the 5, 13, and
4 derivatives without further purification. Compound 7 was synthesized either
from 14 through a sequence of triflation, stannylation and fluorination or from 5
reduced side effects while preserving therapeutic effects (“sig-
1
7
8,91
naling efficacy hypothesis”)
. An example is the superior
safety profile of buprenorphine (a partial agonist in vitro) com- through a sequence of stannylation and fluorination.
9
2
pared to the standard prescription opioids . Buprenorphine is as
efficacious as full agonists in treating pain but has substantially General procedure for synthesis of 11-X-MG derivatives (15–17). Starting
reduced ability to produce respiratory depression and thus material (5, 7, or 13, 0.11 mmol) was dissolved in AcOH (2.0 mL) under argon and
9
2
NaBH
at RT for 15 min, another portion of NaBH
added and stirring was continued for 1 h. After this time, MeOH (81 µL) was added
3
CN (13.7 mg, 0.22 mmol, 2 equiv.) was added to the solution. After stirring
reduced risk of opioid overdose death . Such low-efficacy com-
pounds may have been incorrectly identified as G protein biased
3
CN (13.7 mg, 0.22 mmol, 2 equiv.) was
due both to receptor reserve and to the much greater amplifica- and the RM was heated to 90 °C for 1 h (for 15) and 14 h (for 16 and 17). The
tion of the G protein signaling arm as compared to the much less reaction mixture was added into a cold concentrated NH OH solution and
extracted with DCM. After drying over Na SO , the DCM extract was evaporated.
4
7
8
sensitive arrestin pathway assays . Lower efficacy opioids are
2
4
Product was purified by preparative TLC using an appropriate solvent mixture.
also of great interest for the treatment of opioid use disorder
(
OUD) where such compounds may provide efficacious opioid
General procedure for synthesis of 11-X-7OH derivatives (22–24). Starting
material (15, 16, or 17; 73 µmol) was dissolved in acetone (2.2 mL), sat. aq.
NaHCO (1.5 mL) was added, and the stirred suspension was cooled in an ice bath
3
maintenance while having a reduced abuse potential. Therefore,
close structural derivatives with a varying degree of G protein
signaling efficacy are needed for direct and rigorous probing of (0 °C). OXONE® (1.4–1.5 equiv.) in H O (0.7 mL) was added dropwise over 20 min
2
this hypothesis.
with vigorous stirring (care should be taken that the RM does not form lumps and
should be stirred thoroughly). The reaction was monitored during the addition of
In this context the 11-halo series of 7OH provide valuable
pharmacological tools. We demonstrated that 7OH is a potent
low-efficacy agonist at mMOR and hMOR in the efficacy range extracts were washed with brine (10 mL), dried over Na SO , and concentrated.
®
OXONE by TLC. After 25 min from the first addition, the reaction mixture was
2
diluted with H O (10 mL) and extracted with EtOAc (3 × 10 mL). The combined
2
4
comparable to buprenorphine (Emax ~ 20% versus DAMGO), in a Product was purified by preparative TLC using an appropriate solvent mixture to
synthesize compounds 22, 23, and 24.
cell-based assay without signaling amplification (Nb33 BRET
assay). Comparatively, 11-F-7OH showed only minimal receptor
activation in this assay. In the amplified system (cAMP BRET
assay) with a greater dynamic range on the low end of efficacy
spectrum, the enhancement in apparent efficacy of 7OH was
much smaller for 22. Thus, using two assays with complementary housed 5 mice per cage in a vivarium following an IACUC-approved protocol. For
dynamic ranges of efficacy readouts, we showed that fluorination male C57BL/6 mice temperature was kept constant at 22 ± 2 °C, and relative
in the 11-position markedly reduces MOR signaling efficacy of an
already low-efficacy agonist, the parent 7OH.
Tail-flick mice assay
Mice. For analgesic dose–response experiments, male CD1 mice (20–32 g),
6
–8 weeks were obtained from Charles River Laboratories and male C57BL/6 mice
(
22–30 g), 8–15 weeks were obtained from Jackson Lab (Bar Harbor, ME) and
humidity was maintained at 50 ± 5%. For male CD1 mice (20–32 g) the tem-
perature was in the range of 20–26 °C and relative humidity maintained within the
range of 30–70%. Mice were given access to food and tap water ad libitum. All mice
used throughout the manuscript were opioid naïve. All mice were maintained on a
Interestingly, the correlation between the cAMP signaling
efficacy obtained in vitro in cultured cells and the in vivo 12 h light/dark cycle with Purina rodent chow and water available ad libitum and
housed in groups of five until testing.
For analgesic testing in KO animals, wild-type, male C57BL/6 mice (22–33 g),
analgesic efficacy determined in living animals, for the 11-X-7OH
series, suggests that the in vitro amplified system may mimic the
receptor reserve and/or receptor-effector situation in the neurons
1
0–12 weeks were purchased from the Jackson Lab (Bar Harbor, ME). These mice
were kept at a constant temperature of 22 ± 2 °C, and relative humidity was
of relevant pain circuitry, and thus provides a useful model maintained at 40–50%. Exon-1/Exon-11 MOR-1 KO mice on a C57 background
N A TU RE C OMM U N IC AT I ON S |
(2 0 2 1 ) 1 2 : 3 85 8 | h t t p s :/ /d o i . o r g /1 0 . 1 0 3 8 /s 4 1 4 67 - 0 2 1 - 2 3 7 3 6 - 2 | w ww . na tu r e . c o m /n atu r e c o m mu n i ca ti o n s
11

A R T I C L E
N AT URE CO M MU N ICAT I ON S | ht t ps: // doi. org /1 0 .10 38/ s41 46 7 -02 1 -2 3 7 36- 2
were bred in the Pintar laboratory at Rutgers University. All mice were maintained
on a 12-hour light/dark cycle with food and water available ad libitum, and housed
in groups of five until testing. All testing was done in the light cycle.
All animal studies were preapproved by the Institutional Animal Care and Use
Committees of Washington University School of Medicine and Columbia
University, in accordance with the 2002 National Institutes of Health Guide for the
Care and Use of Laboratory Animals.
6. Kruegel, A. C. & Grundmann, O. The medicinal chemistry and
neuropharmacology of kratom: a preliminary discussion of a promising
medicinal plant and analysis of its potential for abuse. Neuropharmacology
134, 108–120 (2018).
7. Swogger, M. T. et al. Experiences of kratom users: a qualitative analysis. J.
Psychoactive Drugs 47, 360–367 (2015).
8. DEA 3 Factor Analysis for mitragynine and 7-hydroxymitragynine https://
www.regulations.gov/document?D=DEA-2016-0015-0004 (August 2016).
9. Swogger, M. T. & Walsh, Z. Kratom use and mental health: a systematic
review. Drug Alcohol Depend. 183, 134–140 (2018).
10. Fluyau, D. & Revadigar, N. Biochemical benefits, diagnosis, and clinical risks
evaluation of kratom. Front. Psychiatry 8, 62 (2017).
Tail flick (dose–response). Tail-flick antinociception was determined using the
radiant heat tail-flick technique using an Ugo Basile model 37360 instrument as
9
5,96
previously described
. The intensity was set to achieve a baseline between 2 and
3
s. Baseline latencies were determined before experimental treatments for all mice.
Tail-flick antinociception was assessed as an increase in baseline latency, with a
maximal 15 s latency to minimize damage to the tail. Data were analyzed as percent
maximal effect, % MPE, and was calculated according to the formula: % MPE =
11. Chakraborty, S. & Majumdar, S. Natural products for the treatment of pain:
chemistry and pharmacology of salvinorin A, mitragynine, and collybolide.
Biochemistry 60, 1381–1400 (2021).
[
(observed latency – baseline latency)/(maximal latency – baseline latency)] × 100.
1
1
1
1
1
1
1
1
2. Wilson, L. L. et al. Kratom alkaloids, natural and semi-synthetic, show less
Compounds were injected s.c. and antinociception was assessed at the peak effect.
Mice were tested for analgesia with cumulative subcutaneous doses of the drug
until the mouse can withstand the maximal latency. Once the mouse reached the
maximal latency, the mouse was no longer given higher doses. The analgesia
experiments were performed by blinding the experimenter to the identity of 7OH
versus 11-F-7OH. In vivo experiments were evaluated using GraphPad Prism 8,
San Diego, CA as described above.
physical dependence and ameliorate opioid withdrawal. Cell Mol. Neurobiol.
4
1, 1131–1143 (2021).
3. Takayama, H. Chemistry and pharmacology of analgesic indole alkaloids from
the rubiaceous plant, Mitragyna speciosa. Chem. Pharm. Bull. 52, 916–928
(
2004).
4. Takayama, H. et al. New procedure to mask the 2,3-π bond of the indole
nucleus and its application to the preparation of potent opioid receptor
agonists with a corynanthe skeleton. Org. Lett. 8, 5705–5708 (2006).
5. Kruegel, A. C. et al. Synthetic and receptor signaling explorations of the
Mitragyna alkaloids: mitragynine as an atypical molecular framework for
opioid receptor modulators. J. Am. Chem. Soc. 138, 6754–6764 (2016).
6. Váradi, A. et al. Mitragynine/corynantheidine pseudoindoxyls as opioid
analgesics with Mu agonism and delta antagonism, which do not recruit
β-arrestin-2. J. Med. Chem. 59, 8381–8397 (2016).
Tail flick (MOR KO animals). Analgesia was tested in wild-type and MOR KO
animals by the radiant heat tail-flick technique using an IITC Model 33 Tail Flick
9
7
Analgesia Meter as previously described . The intensity was set to achieve a
baseline between 2 and 3 s. Tail-flick antinociception was assessed as an increase in
baseline latency, with a maximal 10 s latency to minimize damage to the tail. Data
were analyzed as percent maximal effect, % MPE, which was calculated according
to the formula: % MPE [(observed latency − baseline latency)/(maximal latency −
baseline latency)] × 100. Compounds were administered s.c. as indicated in the
figures, and analgesia was assessed at the peak effect (15 min). Mice were tested for
analgesia with cumulative subcutaneous doses of the drug until the mouse can
withstand the maximal latency. Once the mouse reached the maximal latency, the
mouse was no longer given higher doses. In vivo experiments were evaluated using
GraphPad Prism 8, San Diego, CA as described above.
7. Kruegel, A. C. et al. 7-Hydroxymitragynine is an active metabolite of
Mitragynine and a key mediator of its analgesic effects. ACS Cent. Sci. 5,
992–1001 (2019).
8. Macko, E., Weisbach, J. A. & Douglas, B. Some observations on the
pharmacology of mitragynine. Arch. Int. Pharmacodyn. Ther. 198, 145–161
(
1972).
9. Hemby, S. E., McIntosh, S., Leon, F., Cutler, S. J. & McCurdy, C. R. Abuse
liability and therapeutic potential of the Mitragyna speciosa (kratom) alkaloids
mitragynine and 7-hydroxymitragynine: kratom abuse liability. Addict. Biol.
Reporting summary. Further information on research design is available in the Nature
Research Reporting Summary linked to this article.
2
4, 874–885 (2019).
20. Yue, K., Kopajtic, T. A. & Katz, J. L. Abuse liability of mitragynine assessed
with a self-administration procedure in rats. Psychopharmacology 235,
Data availability
2823–2829 (2018).
The authors declare that all the data supporting the findings of this study are available
within the article and Supplementary Information files which contains synthetic
procedures and NMR spectra for the featured compounds, additional functional data at
rodent and human receptors, biological protocols for receptor binding, activity and
antinociception assays. The X-ray crystallographic coordinates for structure 4 reported in
this study have been deposited at the Cambridge Crystallographic Data Centre (CCDC),
under deposition numbers 1905559. This data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_request/cif. A
preprint version of this work has been deposited in the public depository ChemRxiv
platform https://doi.org/10.26434/chemrxiv.12799787. Source data are provided with
this paper.
2
2
2
1. Volkow, N. D. & Collins, F. S. The role of science in addressing the opioid
crisis. N. Engl. J. Med. 377, 391–394 (2017).
2. Majumdar, S. & Devi, L. A. Strategy for making safer opioids bolstered. Nature
5
53, 286–288 (2018).
3. Samuels, B. A. et al. The behavioral effects of the antidepressant tianeptine
require the Mu-opioid receptor. Neuropsychopharmacology 42, 2052–2063
(
2017).
2
2
4. Takayama, H. et al. The first total synthesis of (−)-mitragynine, an
analgesic indole alkaloid in Mitragyna speciosa. Tetrahedron Lett. 36,
9337–9340 (1995).
5. Ma, J., Yin, W., Zhou, H., Liao, X. & Cook, J. M. General approach to the total
synthesis of 9-methoxy-substituted indole alkaloids: synthesis of mitragynine,
Received: 2 August 2020; Accepted: 13 May 2021;
as well as 9-methoxygeissoschizol and 9-methoxy-N -methylgeissoschizol. J.
Org. Chem. 74, 264–273 (2009).
b
2
2
2
2
6. Kerschgens, I. P. et al. Total syntheses of mitragynine, paynantheine and
speciogynine via an enantioselective thiourea-catalysed Pictet–Spengler
reaction. Chem. Commun. 48, 12243–12245 (2012).
7. Kim, J., Schneekloth, J. S. & Sorensen, E. J. A chemical synthesis of 11-
methoxy mitragynine pseudoindoxyl featuring the interrupted Ugi reaction.
Chem. Sci. 3, 2849–2852 (2012).
8. Sun, X. & Ma, D. Organocatalytic approach for the syntheses of
corynantheidol, dihydrocorynantheol, protoemetinol, protoemetine, and
mitragynine. Chem. Asian J. 6, 2158–2165 (2011).
References
1.
2.
3.
4.
5.
Gassaway, M. M., Rives, M.-L., Kruegel, A. C., Javitch, J. A. & Sames, D. The
atypical antidepressant and neurorestorative agent tianeptine is a μ-opioid
receptor agonist. Transl. Psychiatry 4, e411–e411 (2014).
Kruegel, A. C., Rakshit, S., Li, X. & Sames, D. Constructing Iboga alkaloids via
C–H bond functionalization: examination of the direct and catalytic union of
heteroarenes and isoquinuclidine alkenes. J. Org. Chem. 80, 2062–2071 (2015).
Gassaway, M. M. et al. Deconstructing the Iboga alkaloid skeleton:
potentiation of FGF2-induced glial cell line-derived neurotrophic factor
release by a novel compound. ACS Chem. Biol. 11, 77–87 (2016).
Marton, S. et al. Ibogaine administration modifies GDNF and BDNF
expression in brain regions involved in mesocorticolimbic and nigral
dopaminergic circuits. Front. Pharmacol. 10, 193 (2019).
9. Matsumoto, K. et al. Orally active opioid / dual agonist MGM-16, a derivative
of the indole alkaloid Mitragynine, exhibits potent antiallodynic effect on
neuropathic pain in mice. J. Pharmacol. Exp. Ther. 348, 383–392 (2014).
0. Godula, K. & Sames, D. C–H bond functionalization in complex organic
synthesis. Science 312, 67–72 (2006).
3
3
3
1. Pastine, S. J., Gribkov, D. V. & Sames, D. sp C−H bond arylation directed by
amidine protecting group: α-arylation of pyrrolidines and piperidines. J. Am.
Chem. Soc. 128, 14220–14221 (2006).
3
2. Guo, P., Joo, J. M., Rakshit, S. & Sames, D. C–H arylation of pyridines: high
regioselectivity as a consequence of the electronic character of C–H bonds and
heteroarene ring. J. Am. Chem. Soc. 133, 16338–16341 (2011).
Adkins, J. E., Boyer, E. W. & McCurdy, C. R. Mitragyna speciosa, a
psychoactive tree from Southeast Asia with opioid activity. Curr. Top. Med.
Chem. 11, 1165–1175 (2011).
1
2
N AT U RE CO M MU NI C A TI O N S |
( 2 0 21 ) 1 2 : 3 85 8 | h tt p s : / /d o i . o r g /1 0 . 1 0 3 8/ s 41 4 67 - 0 2 1 - 2 3 7 3 6- 2 | w ww . na tu r e . c o m /n atu r e c o m m u n ic a tio n s

NA TUR E COMM UNICAT IO NS | ht t ps ://doi .or g / 10 .1 0 38/ s 41 46 7 -02 1- 23 7 36-2
A R T I C L E
3
3
3
3. Genovino, J., Sames, D., Hamann, L. G. & Touré, B. B. Accessing drug
metabolites via transition-metal catalyzed C−H oxidation: the liver as
synthetic inspiration. Angew. Chem. Int. Ed. 55, 14218–14238 (2016).
4. Lane, B. S., Brown, M. A. & Sames, D. Direct palladium-catalyzed C-2 and C-3
arylation of indoles: a mechanistic rationale for regioselectivity. J. Am. Chem.
Soc. 127, 8050–8057 (2005).
5. Goikhman, R., Jacques, T. L. & Sames, D. C−H bonds as ubiquitous
functionality: a general approach to complex arylated pyrazoles via sequential
regioselective C-arylation and N-alkylation enabled by SEM-group
transposition. J. Am. Chem. Soc. 131, 3042–3048 (2009).
61. Litvinas, N. D., Fier, P. S. & Hartwig, J. F. A general strategy for the
perfluoroalkylation of arenes and arylbromides by using arylboronate esters
F
and [(phen)CuR ]. Angew. Chem. Int. Ed. 51, 536–539 (2012).
62. Eastabrook, A. S., Wang, C., Davison, E. K. & Sperry, J. A procedure for
transforming indoles into indolequinones. J. Org. Chem. 80, 1006–1017
(2015).
63. Meyer-Eppler, G. et al. Cheap and easy synthesis of highly functionalized
(Het)aryl iodides via the aromatic Finkelstein reaction. Synthesis 46,
1085–1090 (2014).
64. Cooper, T., Novak, A., Humphreys, L. D., Walker, M. D. & Woodward, S.
User-friendly methylation of aryl and vinyl halides and pseudohalides with
3
6. Wang, X., Lane, B. S. & Sames, D. Direct C-arylation of free (NH)-indoles and
pyrroles catalyzed by Ar−Rh(III) complexes assembled in situ. J. Am. Chem.
Soc. 127, 4996–4997 (2005).
3
DABAL-Me . Adv. Synth. Catal. 348, 686–690 (2006).
65. Wang, B., Sun, H.-X. & Sun, Z.-H. A general and efficient Suzuki-Miyaura
cross-coupling protocol using weak base and no water: the essential role of
acetate. Eur. J. Org. Chem. 2009, 3688–3692 (2009).
3
7. He, J., Wasa, M., Chan, K. S. L., Shao, Q. & Yu, J.-Q. Palladium-catalyzed
transformations of alkyl C–H bonds. Chem. Rev. 117, 8754–8786 (2017).
8. McMurray, L., O’Hara, F. & Gaunt, M. J. Recent developments in natural
product synthesis using metal-catalysed C–H bond functionalisation. Chem.
Soc. Rev. 40, 1885–1898 (2011).
3
66. Suresh, A. S., Baburajan, P. & Ahmed, M. Synthesis of primary amides by
3
aminocarbonylation of aryl/hetero halides using non-gaseous NH and CO
sources. Tetrahedron Lett. 56, 4864–4867 (2015).
3
4
4
9. Cernak, T., Dykstra, K. D., Tyagarajan, S., Vachal, P. & Krska, S. W. The
medicinal chemist’s toolbox for late stage functionalization of drug-like
molecules. Chem. Soc. Rev. 45, 546–576 (2016).
0. Yang, L. & Huang, H. Transition-metal-catalyzed direct addition of
unactivated C–H bonds to polar unsaturated bonds. Chem. Rev. 115,
67. Ramnauth, J., Bhardwaj, N., Renton, P., Rakhit, S. & Maddaford, S. P.
The room-temperature palladium-catalyzed cyanation of aryl bromides
and iodides with tri-t-butylphosphine as ligand. Synlett 2003, 2237–2239
(2003).
68. Fier, P. S., Luo, J. & Hartwig, J. F. Copper-mediated fluorination of
arylboronate esters. identification of a copper(III) fluoride complex. J. Am.
Chem. Soc. 135, 2552–2559 (2013).
69. Taylor, N. J. et al. Derisking the Cu-mediated 18F-fluorination of heterocyclic
positron emission tomography radioligands. J. Am. Chem. Soc. 139,
8267–8276 (2017).
3468–3517 (2015).
1. Gensch, T., Hopkinson, M. N., Glorius, F. & Wencel-Delord, J. Mild metal-
catalyzed C–H activation: examples and concepts. Chem. Soc. Rev. 45,
2900–2936 (2016).
4
4
2. Mukai, K. et al. Bioinspired chemical synthesis of monomeric and dimeric
stephacidin A congeners. Nat. Chem. 10, 38–44 (2017).
3. Kerschgens, I., Rovira, A. R. & Sarpong, R. Total synthesis of (−)-xishacorene
B from (R)-carvone using a C–C activation strategy. J. Am. Chem. Soc. 140,
70. Furuya, T. & Ritter, T. Fluorination of boronic acids mediated by silver(I)
triflate. Org. Lett. 11, 2860–2863 (2009).
71. Tang, P., Wang, W. & Ritter, T. Deoxyfluorination of phenols. J. Am. Chem.
Soc. 133, 11482–11484 (2011).
9810–9813 (2018).
4
4. Roque, J. B., Kuroda, Y., Göttemann, L. T. & Sarpong, R. Deconstructive
72. Furuya, T., Strom, A. E. & Ritter, T. Silver-mediated fluorination of
functionalized aryl stannanes. J. Am. Chem. Soc. 131, 1662–1663 (2009).
73. Liu, S., Scotti, J. S. & Kozmin, S. A. Emulating the logic of monoterpenoid
alkaloid biogenesis to access a skeletally diverse chemical library. J. Org. Chem.
78, 8645–8654 (2013).
fluorination of cyclic amines by carbon–carbon cleavage. Science 361, 171–174
(
2018).
4
5. Liao, K. et al. Site-selective and stereoselective functionalization of non-
activated tertiary C–H bonds. Nature 551, 609–613 (2017).
4
6. Larsen, M. A. & Hartwig, J. F. Iridium-catalyzed C–H borylation of
heteroarenes: scope, regioselectivity, application to late-stage
functionalization, and mechanism. J. Am. Chem. Soc. 136, 4287–4299 (2014).
7. Preshlock, S. M. et al. A traceless directing group for C–H borylation. Angew.
Chem. Int. Ed. 52, 12915–12919 (2013).
74. Movassaghi, M., Schmidt, M. A. & Ashenhurst, J. A. Stereoselective oxidative
rearrangement of 2-aryl tryptamine derivatives. Org. Lett. 10, 4009–4012
(2008).
75. Ishikawa, H., Takayama, H. & Aimi, N. Dimerization of indole derivatives
with hypervalent iodines(III): a new entry for the concise total synthesis of
rac- and meso-chimonanthines. Tetrahedron Lett. 43, 5637–5639 (2002).
76. Jiang, L. I. et al. Use of a cAMP BRET sensor to characterize a novel regulation
of cAMP by the sphingosine 1-phosphate/G13 pathway. J. Biol. Chem. 282,
10576–10584 (2007).
77. Kenakin, T. A scale of agonism and allosteric modulation for assessment of
selectivity, bias, and receptor mutation. Mol. Pharm. 92, 414–424 (2017).
78. Gillis, A. et al. Low intrinsic efficacy for G protein activation can explain the
improved side effect profiles of new opioid agonists. Sci. Signal. 13, eaaz3140
(2020).
79. Stoeber, M. et al. Agonist-selective recruitment of engineered protein probes
and of GRK2 by opioid receptors in living cells. eLife 9, e54208 (2020).
80. Gutridge, A. M. et al. G protein-biased kratom-alkaloids and synthetic
carfentanil-amide opioids as potential treatments for alcohol use disorder. Br.
J. Pharm. 177, 1497–1513 (2020).
4
4
8. Murphy, J. M., Liao, X. & Hartwig, J. F. Meta halogenation of 1,3-disubstituted
arenes via iridium-catalyzed arene borylation. J. Am. Chem. Soc. 129,
15434–15435 (2007).
4
5
5
5
9. Paul, S. et al. Ir-catalyzed functionalization of 2-substituted indoles at the 7-
position: nitrogen-directed aromatic borylation. J. Am. Chem. Soc. 128,
1
5552–15553 (2006).
0. Homer, J. A. & Sperry, J. A short synthesis of the endogenous plant metabolite
-hydroxyoxindole-3-acetic acid (7-OH-OxIAA) using simultaneous C–H
7
borylations. Tetrahedron Lett. 55, 5798–5800 (2014).
1. Leitch, J. A., Bhonoah, Y. & Frost, C. G. Beyond C2 and C3: transition-
metal-catalyzed C–H functionalization of indole. ACS Catal. 7, 5618–5627
(
2017).
2. Feng, Y. et al. Total synthesis of verruculogen and fumitremorgin A enabled
by ligand-controlled C–H borylation. J. Am. Chem. Soc. 137, 10160–10163
(
2015).
81. Matsumoto, K. et al. Antinociceptive effect of 7-hydroxymitragynine in mice:
Discovery of an orally active opioid analgesic from the Thai medicinal herb
Mitragyna speciosa. Life Sci. 74, 2143–2155 (2004).
5
5
3. Ikeda, M. & Tamura, Y. 3-Haloindolenines—versatile intermediates in the
indole chemistry. Heterocycles 14, 867–888 (1980).
4. Vadola, P. A. & Sames, D. Catalytic coupling of arene C–H bonds and
alkynes for the synthesis of coumarins: substrate scope and application to
the development of neuroimaging agents. J. Org. Chem. 77, 7804–7814 (2012).
5. Yang, Y., Li, R., Zhao, Y., Zhao, D. & Shi, Z. Cu-catalyzed direct C6-arylation
of indoles. J. Am. Chem. Soc. 138, 8734–8737 (2016).
82. Matsumoto, K. et al. Involvement of μ-opioid receptors in antinociception and
inhibition of gastrointestinal transit induced by 7-hydroxymitragynine,
isolated from Thai herbal medicine Mitragyna speciosa. Eur. J. Pharmacol.
549, 63–70 (2006).
83. Henningfield, J. E. et al. Risk of death associated with kratom use compared to
opioids. Preventive Med. 128, 105851 (2019).
5
5
6. Yang, G. et al. Pd(II)-catalyzed meta-C–H olefination, arylation, and
acetoxylation of indolines using a U-shaped template. J. Am. Chem. Soc. 136,
84. Schmid, C. L. et al. Bias factor and therapeutic window correlate to predict
safer opioid analgesics. Cell 171, 1165–1175.e13 (2017).
10807–10813 (2014).
5
5
5
6
7. Leitch, J. A., McMullin, C. L., Mahon, M. F., Bhonoah, Y. & Frost, C. G.
Remote C6-selective ruthenium-catalyzed C–H alkylation of indole derivatives
via σ-activation. ACS Catal. 7, 2616–2623 (2017).
85. Johnson, T. A. et al. Identification of the first marine-derived opioid receptor
“balanced” agonist with a signaling profile that resembles the endorphins. ACS
Chem. Neurosci. 8, 473–485 (2017).
86. Hill, R. et al. The novel μ-opioid receptor agonist PZM21 depresses respiration
and induces tolerance to antinociception: PZM21 depresses respiration. Br. J.
Pharmacol. 175, 2653–2661 (2018).
87. Kliewer, A. et al. Phosphorylation-deficient G-protein-biased μ-opioid
receptors improve analgesia and diminish tolerance but worsen opioid side
effects. Nat. Commun. 10, 367 (2019).
88. Kliewer, A. et al. Morphine-induced respiratory depression is independent of
β‐arrestin2 signalling. Br. J. Pharmacol. 177, 2923–2931 (2020).
8. Gribble, G. W., Johnson, J. L. & Saulnier, M. G. Stereoselective reduction of
1,2,3,4,6,7,12,12b-octahydroindolo[2,3-a]quinolizine with sodium
borohydride in trifluoeoacetic acid. Heterocycles 16, 2109–2114 (1981).
9. Okada, N., Misawa, K., Kitajima, M. & Takayama, H. Preparation of ethylene
glycol adducts at 2,3-positions of indoles with hypervalent iodine.
Heterocycles 74, 461–472 (2007).
0. Saito, Y., Segawa, Y. & Itami, K. para-C–H borylation of benzene derivatives
by a bulky iridium catalyst. J. Am. Chem. Soc. 137, 5193–5198 (2015).
N
A
T
U
R
E
C
O
M
M
U
N
I
C
A
T
I
O
N
S
|
(2 0 2 1 ) 1 2 : 3 85 8 | h t t p s :/ /d o i . o r g /1 0 . 1 0 3 8 /s 4 1 4 67 - 0 2 1 - 2 3 7 3 6 - 2 | w ww . na tu r e . c o m /n atu r e c o m mu n i ca ti o n s
13

A
R
T
I
C
L
E
N AT URE CO M MU N ICAT I ON S | ht t ps: // doi. org /1 0 .10 38/ s41 46 7 -02 1 -2 3 7 36- 2
8
9. Conibear, A. E. & Kelly, E. A biased view of μ-opioid receptors? Mol. Pharm.
6, 542–549 (2019).
0. Faouzi, A. et al. Synthesis and pharmacology of a novel μ–δ opioid receptor
heteromer-selective agonist based on the carfentanyl template. J. Med. Chem.
characterization. V.H. contributed to the synthesis and characterization of compounds,
T.F. contributed to reaction optimization and compound characterization. M.N. per-
formed the BRET functional assays, A.C.K. contributed to the design of compounds and
supervised the interpretation of the pharmacological data, S.M. supervised the binding
assays, tail-flick mice assay and contributed to global planning of the project, A.H.
performed the binding assays, A.F. and B.B. performed the tail-flick mice assay, M.A.
performed the tail-flick mice assay comparing WT and MOR/KO mice, J.E.P. supervised
the tail-flick mice assay with MOR/KO mice, J.A.J. supervised the BRET functional assays
and interpretation of the corresponding results. S.B. and D.S. wrote the manuscript with
significant help from J.G., V.H., J.A.J., and A.C.K. All authors contributed to editing of
the manuscript.
9
9
6
3, 13618–13637 (2020).
9
9
9
1. Uprety, R. et al. Controlling opioid receptor functional selectivity by targeting
distinct subpockets of the orthosteric site. eLife 10, e56519 (2021).
2. Pergolizzi, J. et al. Current knowledge of buprenorphine and its unique
pharmacological profile. Pain Pract. 10, 428–450 (2010).
3. Saref, A. et al. Self-reported prevalence and severity of opioid and kratom
(Mitragyna speciosa korth.) side effects. J. Ethnopharmacol. 238, 111876
(
2019).
9
9
4. Garcia-Romeu, A., Cox, D. J., Smith, K. E., Dunn, K. E. & Griffiths, R. R.
Kratom (Mitragyna speciosa): user demographics, use patterns, and
implications for the opioid epidemic. Drug Alcohol Depend. 208, 107849
Competing interests
S.B., J.G., V.H., A.C.K., J.A.J., S.M., and D.S., are named as inventors on patent appli-
cations related to mitragynine analogs. A.C.K., J.A.J., and D.S. are co-founders of Kures,
Inc. S.M. is a co-founder of Sparian biosciences. The authors M.N., A.F., B.B., M.A., T.F.,
A.H., and J.E.P. declare no competing interests.
(
2020).
5. Kozai, T. D. Y., Jaquins-Gerstl, A. S., Vazquez, A. L., Michael, A. C. & Cui, X.
T. Brain tissue responses to neural implants impact signal sensitivity and
intervention strategies. ACS Chem. Neurosci. 6, 48–67 (2015).
9
6. Váradi, A. et al. Synthesis of carfentanil amide opioids using the Ugi
multicomponent reaction. ACS Chem. Neurosci. 6, 1570–1577 (2015).
7. Schuller, A. G. P. et al. Retention of heroin and morphine–6β–glucuronide
analgesia in a new line of mice lacking exon 1 of MOR–1. Nat. Neurosci. 2,
Additional information
Supplementary information The online version contains supplementary material
available at https://doi.org/10.1038/s41467-021-23736-2.
9
151–156 (1999).
Correspondence and requests for materials should be addressed to D.S.
9
8. Majumdar, S. et al. Generation of novel radiolabeled opiates through site-
selective iodination. Bioorg. Med. Chem. Lett. 21, 4001–4004 (2011).
Peer review information Nature Communications thanks Zhuangzhi Shi and the other,
anonymous, reviewer(s) for their contribution to the peer review of this work.
Acknowledgements
Reprints and permission information is available at http://www.nature.com/reprints
This work was supported by the National Institute on Drug Addiction (NIDA) of the
National Institute of Health (NIH), grants R01DA046487 (D.S., S.M., and J.A.J.),
R21DA045884 (S.M. and D.S.) and R21MH1164620 (J.E.P.), and the Hope for Depres-
sion Research Foundation (J.A.J.). S.M. also acknowledges Faculty Research Incentive
Funding (FRIF) from STLCOP and AA026949 from NIAAA. A.F. acknowledges support
from the Philippe Foundation. T.F. is the recipient of the Alfred Bader Fellowship in
Organic Chemistry. We thank Prof. Gerard Parkin and Matthew James Hammond for
solving the X-ray crystal structure of MG-EG and helpful discussions. We also thank Dr.
Mu Yang, Co-Director of the Mouse NeuroBehavior Core (MNBC) facility for oversight
of the in vivo experiments at Columbia University. We thank Dr. Nevin Lambert at the
Medical College of Georgia and Dr. Meritxell Canals at the University of Nottingham for
providing the plasmid coding for human MOR-nanoluc (hMOR-nluc) and the plasmid
coding for the nanobody-33-Venus (Nb-33) construct, respectively. This research was
funded in part through the NIH/NCI Cancer Center Support Grant P30 CA008748 to
MSKCC.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
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
D.S. conceptualized and supervised the work. S.B. and J.G. carried out majority of the
synthetic work including reaction development and optimization, and compound
© The Author(s) 2021
1
4
N AT U RE CO M MU NI C A TI O N S |
( 2 0 21 ) 1 2 : 3 85 8 | h tt p s : / /d o i . o r g /1 0 . 1 0 3 8/ s 41 4 67 - 0 2 1 - 2 3 7 3 6- 2 | w ww . na tu r e . c o m /n atu r e c o m m u n ic a tio n s