Alkyl indole-based cannabinoid type 2 receptor tools: Exploration of linker and fluorophore attachment

Article abstract of DOI:10.1016/j.ejmech.2017.11.076

Cannabinoid type 2 (CB₂) receptor continues to emerge as a promising drug target for many diseases and conditions. New tools for studying CB₂ receptor are required to further inform how this receptor functions in healthy and diseased states. The alkyl indole scaffold is a well-recognised ligand for cannabinoid receptors, and in this study the indole C5-7 positions were explored for linker and fluorophore attachment. A new high affinity, CB₂ receptor selective inverse agonist was identified (16b) along with a general trend of C5-substituted indoles acting as agonists versus C7-substituted indoles acting as inverse agonists. The indole C7 position was found to be the most tolerant to linker extension and resulted in a high affinity inverse agonist with a medium length linker (19). Although a high affinity fluorescent ligand for CB₂ receptor was not identified in this study, the indole C7 position shows great promise for fluorophore or probe attachment.

Full text of DOI:10.1016/j.ejmech.2017.11.076

Accepted Manuscript
Alkyl indole-based cannabinoid type 2 receptor tools: Exploration of linker and
fluorophore attachment
Anna G. Cooper, Christa Macdonald, Michelle Glass, Sarah Hook, Joel D.A. Tyndall,
Andrea J. Vernall
PII:
S0223-5234(17)30978-9
10.1016/j.ejmech.2017.11.076
EJMECH 9951
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 23 September 2017
Revised Date: 23 November 2017
Accepted Date: 26 November 2017
Please cite this article as: A.G. Cooper, C. Macdonald, M. Glass, S. Hook, J.D.A. Tyndall, A.J. Vernall,
Alkyl indole-based cannabinoid type 2 receptor tools: Exploration of linker and fluorophore attachment,
European Journal of Medicinal Chemistry (2017), doi: 10.1016/j.ejmech.2017.11.076.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
hCB agonists
O
O
2
O
R2
4
R O
N
N
R1
R3
O
O
hCB inverse agonists
2
1
R = H, hCB K = 5.7 ± 1.4 nM, hCB IC = 30.6 ± 13.9 nM
2
i
2
50
O
1
R =
hCB K = 200.7 ± 4.9 nM
2 i
MeO

ACCEPTED MANUSCRIPT
Alkyl indole-based cannabinoid type 2 receptor tools: Exploration of linker and
fluorophore attachment
†
‡
‡
†
†
Anna G. Cooper, Christa Macdonald, Michelle Glass, Sarah Hook, Joel D. A. Tyndall,
†
Andrea J. Vernall *
†
School of Pharmacy, University of Otago, Dunedin, New Zealand
‡
Department of Pharmacology and Clinical Pharmacology, University of Auckland,
Auckland, New Zealand
*
School of Pharmacy, 18 Frederick Street, Dunedin 9054, New Zealand.
Ph: 64 3 479 4518
Email: andrea.vernall@otago.ac.nz
Abbreviations: BODIPY 630/650-X-OSu, 6-(((4,4-difluoro-5-(2-thienyl)-4-bora-3a,4a-diaza-s-
indacene-3-yl)styryloxy)acetyl) aminohexanoic acid, succinimidyl ester; PEG, polyethylene glycol.

ACCEPTED MANUSCRIPT
ABSTRACT
Cannabinoid type 2 (CB ) receptor continues to emerge as a promising drug target for many
2
diseases and conditions. New tools for studying CB receptor are required to further inform
2
how this receptor functions in healthy and diseased states. The alkyl indole scaffold is a well-
recognised ligand for cannabinoid receptors, and in this study the indole C5-7 positions were
explored for linker and fluorophore attachment. A new high affinity, CB receptor selective
2
inverse agonist was identified (16b) along with a general trend of C5-substituted indoles
acting as agonists versus C7-substituted indoles acting as inverse agonists. The indole C7
position was found to be the most tolerant to linker extension and resulted in a high affinity
inverse agonist with a medium length linker (19). Although a high affinity fluorescent ligand
for CB receptor was not identified in this study, the indole C7 position shows great promise
2
for fluorophore or probe attachment.
KEYWORDS
Cannabinoid type 2 receptor
Cannabinoid
Fluorescent probe
Alkyl indole
G protein-coupled receptor
2

ACCEPTED MANUSCRIPT
1
. INTRODUCTION
Cannabinoid type 2 (CB ) receptor is a member of the G protein-coupled receptor (GPCR)
2
superfamily. GPCRs are the largest family of transmembrane signalling proteins in the
human genome and a major therapeutic target [1]. CB receptor, cannabinoid type 1 (CB )
2
1
receptor, endocannabinoids and associated regulatory enzymes comprise the
endocannabinoid system, which plays a key role in the regulation of various neurotransmitter
systems [2]. CB receptor is one of the most abundant receptors in the brain and plays a
1
significant role in motivation and cognition [3]. In contrast, CB receptor is mostly expressed
2
on cells of the immune system and in lymphatic tissues, spleen and tonsils [4] and only at low
levels in the central nervous system [5]. CB receptor is crucial to immune system function,
2
including modulation of cytokine release and immune cell migration [6, 7].
The cannabinoid (CB) receptors are potential drug targets in multiple pathologies, including
CNS diseases (e.g. Alzheimer’s disease), cancer, cardiovascular disease, drug addiction, pain,
obesity and inflammatory disorders [8]. In particular, CB receptor has been implicated in
2
cancer [9] and inflammatory diseases such as neurodegenerative disorders [10],
atherosclerosis [11] and inflammatory bowel disease [12]. In order to realise the potential of
CB receptor as a drug target to treat these conditions there is a need for more detailed
2
knowledge of how this receptor functions in healthy and diseased states. Current tools
available for studying CB receptor include radioligands, covalent ligands and antibodies and
2
whilst these have been used to study receptor localisation, structure and function, they all
possess significant limitations in their utility [13]. Fluorescent ligands are powerful tools for
studying receptor localisation, structure, dynamics and function in whole live cells [14].
Fluorescent ligands are adaptable to different assay and imaging systems, including high
throughput screening [15], fluorescent resonance energy transfer [16], confocal microscopy,
scanning confocal microscopy, flow cytometry and fluorescence correlation spectroscopy
[
17].
Whilst fluorescent ligands have been successfully designed for some other GPCRs [18],
attempts at developing high affinity, selective CB receptor fluorescent ligands for use in high
2
resolution confocal imaging studies have been thwarted by high non-specific binding,
limiting their utility as imaging tools [19-24]. A number of these fluorescent ligands for CB2
receptor have been developed by conjugating varying fluorophores to a pharmacophore-
3

ACCEPTED MANUSCRIPT
linker (mbc94) derived from SR144528 and of these, NIR760-mbc94 has been shown to
identify CB receptor mediated inflammation in a mouse model [25]. The challenge of
2
developing a high affinity, selective CB receptor fluorescent ligand with little non-specific
2
membrane binding is increased further by the lipophilic nature of cannabinoids and many
fluorescent dyes. Conjugation of a fluorescent dye to a pharmacophore introduces a large
amount of steric bulk to the ligand, which can significantly alter the pharmacology and
physicochemical properties compared to the unconjugated pharmacophore. Variation of site
of linker attachment, linker length, linker composition and fluorescent dye choice can all
influence affinity and efficacy of the probe and the overall physicochemical properties.
It is necessary to select a region of the pharmacophore for linker attachment that is amenable
to change and the introduction of steric bulk/length. The aim of this study was to locate a
suitable linker position on the alkyl indole scaffold and use this for the development of a high
affinity, selective fluorescent CB receptor antagonist or inverse agonist.
2
Figure 1: CB receptor agonists WIN-48,098 (pravadoline) (1) and WIN-55,212-2 (2). CB receptor agonists
2
JWH-015 (3) and JWH-046 (4). CB receptor inverse agonist AM630 (5).
2
Development of the indole class of cannabinoids was spearheaded by the development of CB
receptor agonists WIN-48,098 (pravadoline) 1 [26] and WIN-55,212-2 2 [27] (Figure 1).
These were followed by variations to the scaffold to give CB receptor selective agonists such
2
as JWH-015 3 [28] and JWH-046 4 [29], and inverse agonists such as AM630 5
(
iodopravadoline; hCB K = 31.2 nM, hCB K = 5152 nM) [30]. Many CB receptor
2 i 1 i 2
4

ACCEPTED MANUSCRIPT
selective indoles have subsequently been reported [31, 32]. In this project, linker attachment
via indole C5-C7 was explored for a variety of reasons. Firstly, the presence of the bulky C6
iodine group in AM630 5 demonstrates that CB receptor binding is still possible with steric
2
bulk at this side of the indole ring. Secondly, a number of other studies have shown that a
variety of bulky groups at the C5, C6 and C7 positions, for example O-benzyl [33] and furan
[
34], maintain high affinity and selectivity for CB receptor and give varying function as
2
either agonists or inverse agonists. Thirdly, other studies that attached linkers and fluorescent
dyes at the C3 acyl position of the alkyl indole scaffold were unsuccessful and resulted in a
>
250 fold loss in affinity for CB receptor compared to the parent pharmacophore and
2
implied that there is a strict size limitation on the C3 acyl substituent in order to retain
affinity [35]. In addition, established structure activity relationships (SAR) dictate that the
indole scaffold is highly sensitive to variation of N1 substituents and that alkyl chains longer
than hexyl at this position are not tolerated [29]. In this project, the N1-alkyl and C3-acyl
groups were selected based on optimal affinity and selectivity reported previously by others
[
36, 37] and a C2-H was used instead of the C2-methyl as present in AM630 5 for synthetic
ease since the latter is not essential for CB receptor binding affinity [38].
2
2
2
. RESULTS AND DISCUSSION
.1 Chemistry
All compounds were synthesised from a commercially available 5-, 6- or 7-O-benzyl
1
substituted indole (6a-c) (Scheme 1). Indoles 6a-c were N-alkylated with R -mesylate using
sodium hydride to give 7a-e in good yield. N-Alkyl indoles 7a-e were acylated at C3 with
2
different R -nitriles in the presence of catalytic [(Phen)Pd(OAc) ] using a recently reported
2
adapted procedure [39, 40] to give 8a-h. The palladium-ligand catalytic complex was pre-
formed rather than generated in situ as this gave superior product yields. Previously, Friedel-
Crafts acylation has commonly been used for C3 acylation of this indole class [37], however,
this palladium-catalysed reaction provides a cleaner and safer procedure with robust yield.
The benzyl ether of 8a-h was cleaved using Pd/C and hydrogen to reveal hydroxyindoles 9a-
h. Although these compounds were all evaluated for CB receptor binding (Table 1), not all
2
1
2
combinations of R and R were progressed through further derivatisations.
5

ACCEPTED MANUSCRIPT
1
2
Scheme 1. Reagents and Conditions: (i) R -OMs, NaH, DMF, 45°C, 2 h, 54 - 87%; (ii) R -nitrile,
[
(Phen)Pd(OAc) ], H O, CH COOH, 1,4-dioxane, 140°C, 24 h, 18 - 81%; (iii) Pd/C, H , EtOH, EA, rt, 20 h, 8 -
2 2 3 2
6
7%; (iv) tert-butylbromoacetate, NaH, DMF, 60°C, 3 h, 45 - 61%; (v) TFA, DCM, rt, 1 h; (vi) 1. Fmoc-Ala-
Ala-trityl resin (prepared by SPPS*), piperidine, DMF, 2. HATU, DIPEA, DMF, 3. Resin cleavage (TFA,
DCM, rt, 1 h), 19 - 26%; (vii) H N-(CH ) -NHBoc or H N-(C H O) C H -NHBoc, HATU, DIPEA, DMF, rt, 12
2
2 8
2
2
4
2
2
4
h, 50 - 94%; (viii) TFA, DCM, rt, 1 h, 27 - 77%; (ix.) Ac O, DCM, r.t, 1 h, 97%; (x) BODIPY 630/650-X-OSu,
2
DIPEA, DMF, rt, 12 h, 36 - 88%. * 1. 1,2-diaminoethane trityl resin, Fmoc-Ala-OH, HBTU, DIPEA, DMF, 2.
Ac O, DIPEA, DMF, 3. piperidine, DMF, 4. Fmoc-Ala-OH, HBTU, DIPEA, DMF.
2
Hydroxyindoles 9a, 9c and 9e-f were alkylated using tert-butylbromoacetate to install a short
spacer (10a-d), followed by cleavage of the tert-butyl group to give carboxylic acids 11a-d.
6

ACCEPTED MANUSCRIPT
These carboxylic acids were coupled to three different linkers, either N-Boc-1,8-
octanediamine, N-Boc-2,2’-(ethylenedioxy)diethylamine (polyethylene glycol (PEG) linker)
or Ala-Ala-ethylene-resin-bound (synthesised by solid phase peptide synthesis, then the
product cleaved from the resin) to give Boc-protected 12a and 12c, 12b and 12d or primary
amines 13e-g respectively. Linkers were chosen to give a range of polarity and functionality,
from the hydrophobic ‘benign’ alkyl chain to a more hydrophilic short peptide, with the goal
in mind to create a fluorescent probe with suitable physicochemical properties and little non-
specific-membrane binding and also to generate tools for studying CB receptor with varied
2
polarity. Boc-deprotection of 12a-d yielded primary amines 13a-d, which along with 13e-g
were then coupled to BODIPY 630/650-X-OSu (6-(((4,4-difluoro-5-(2-thienyl)-4-bora-3a,4a-
diaza-s-indacene-3-yl)styryloxy)acetyl) aminohexanoic acid, succinimidyl ester) to give 15a-
g. BODIPY 630/650-X was chosen as a fluorophore due to success with other Class A GPCR
fluorescent probes [18, 41, 42] and due to its high photo and chemical stability, intense
absorption and good fluorescence quantum yield, and red-emitting wavelength meaning
minimal detection interference from cellular autofluorescence. The amine 13b was also
acylated to give 14.
Scheme 2. Reagents and Conditions: (i) 1-bromopropane or BocNH(C H O) C H Br, NaH, DMF, rt, 20 h, 31 -
2
4
3
2
4
8
3%; (ii) 1. TFA, DCM, rt, 1 h, 2. Ac O, DCM, r.t, 1.5 h, 71%.
2
Initial biological characterisation of compounds described thus far indicated the polar
carbonyl group/amide bond installed via the short 5- or 7-linker in close proximity to the
pharmacophore core may be detrimental to CB receptor binding. Therefore, hydroxyindoles
2
9
e-f were reacted by Williamson ether procedure with either a short alkyl bromide or a longer
7

ACCEPTED MANUSCRIPT
PEG-alkyl bromide to give 16a-b or 17 respectively (Scheme 2). Once in hand, 17 was Boc-
deprotected and acetylated to give 18. Attachment of a PEG-amine alcohol to 9e was initially
attempted using Mitsunobu conditions however, this did not give a satisfactory yield of 17.
Biological characterization (described in detail in sections 2.2 and 2.3 below) revealed 7-O-
propyl substituted 16b as a lead compound, therefore 7-position linker derivatives of 16b
1
2
maintaining the same R and R groups were explored (Scheme 3). Alkylation of 9f with
methyl 5-bromovalerate or N-Boc-bromohexylamine afforded 19 or 24 respectively. Methyl
ester hydrolysis of 19 gave carboxylic acid 20, which underwent an amide coupling with N-
Boc-2,2’-(ethylenedioxy)diethylamine to yield 21. Boc-deprotection of 21 followed by
coupling to BODIPY 630/650-X-OSu afforded 23. In a separate reaction, this amine
intermediate was also acetylated to give 22. In an analogous manner, 24 was converted to the
corresponding BODIPY 630/650-X (26) or acetylated (25) conjugate.
Scheme 3. Reagents and Conditions: (i) methyl 5-bromovalerate or 6-(Boc-amino)hexyl bromide, NaH, DMF,
rt, 20 h, 33 - 79%; (ii) 0.2 M aq. LiOH.H O, THF, 0°C, 1 h, 95%; (iii) N-Boc-1,8-diamino-3,6-dioxaoctane,
2
HATU, DIPEA, DMF, rt, 12 h, 82%; (iv) TFA, DCM, rt, 1 h, 92 - 95%; (v) Ac O, DIPEA, DCM, rt, 1 h, 95-
2
9
6%; (vi) BODIPY 630/650-X-OSu, DIPEA, DMF, rt, 12 h, 77 - 93%.
8

ACCEPTED MANUSCRIPT
2
.2. Radioligand binding assays
Key smaller alkyl indoles and indole-linker conjugates along with all fluorescent compounds
3
were subjected to a competition radioligand binding assay screen, using [ H]CP-55,940 as the
competing radioligand and human embryonic kidney 293 (HEK) cells over-expressing human
(h) CB receptor chimerized with three N-terminal haemagglutinin (HA) tags (3HA-hCB )
2 2
3
[
43]. For compounds that displaced [ H]CP55,940 by more than 50% at hCB receptor,
2
concentration response curves were obtained at hCB receptor to determine K values (Table
2
i
1
). Compounds that showed high affinity for hCB receptor were then subjected to a
2
competition binding assay screen in HEK cells over-expressing 3HA-hCB receptor [44]. CB
1
receptor inverse agonist SR144528 was analysed in the binding assays as a literature
comparison control. In this study, SR144528 displayed a K = 51.0 ± 3.0 nM at hCB and K =
i
2
i
5
682 ± 2890 nM at hCB receptor with a CB :CB selectivity of 110. This is comparable to
1 2 1
the recently reported SR144528 pK = 7.88 ± 0.06 at hCB (K = 13.2 nM) and pK = 5.77 ±
i
2
i
i
3
0
.09 at hCB receptor (K = 1698.2 nM; using [ H]CP55,940 K = 0.33 at hCB and 0.1 at
1 i d 2
hCB ), with a CB :CB selectivity of 129 [45].
1
2
1
9

ACCEPTED MANUSCRIPT
Table 1. Affinity of pharmacophore and pharmacophore-linker conjugates for hCB and hCB receptors.
2
1
1
2
3
Compound
R
R
R
K hCB2
i
K hCB1
i
Selectivity fold
a
a
position
(nM ± SEM)
(nM ± SEM)
>10000
no binding
hCB / hCB1
2
b
9
9
9
9
9
9
9
9
a
b
c
d
e
f
Et-morpholino
Et-morpholino
Et-morpholino
Et-morpholino
Me-THP
MeOPh
5
746.6 ± 139.8
>10000
> 13
-
c
MeOPh
6
5
5
5
7
5
5
b
naphthalene
cyclohexane
MeOPh
169.0 ± 16.7
895.3 ± 332.2
120.1 ± 23.4
2838 ±705.3
>10000
17
> 11
12
66
3
b
d
1406 ±463.6
3583 ± 427
823.8 ±214.6
no binding
d
Me-THP
MeOPh
53.9 ± 7.6
d
d
g
h
Me-THP
naphthalene
MeOPh
275.4 ± 82.7
Me-THF
>10000
-
1
1
1
1
1
1
1
2
1a
1b
1c
1d
6a
6b
9
Et-morpholino
Et-morpholino
Me-THP
MeOPh
naphthalene
MeOPh
MeOPh
MeOPh
MeOPh
MeOPh
MeOPh
5
5
5
7
5
7
7
7
>10000
>10000
>10000
>10000
>10000
4527 ± 676
-
>10000
-
1660 ± 114.2
581.8 ± 70.8
> 6
> 17
3
Me-THP
d
Me-THP
1350 ± 399.3
b
b
Me-THP
5.7 ± 1.4
504.6 ± 148.2
1583 ± 242.7
>10000
89
8
Me-THP
200.7 ± 4.9
b
0
Me-THP
1307 ± 302.5
> 7
e
1
1
1
1
1
1
2
1
1
1
2
3a
3b
4
Et-morpholino
Et-morpholino
Et-morpholino
Et-morpholino
Et-morpholino
Me-THP
MeOPh
5
5
5
5
5
5
7
5
5
5
7
>10000
>10000
>10000
>10000
>10000
n.d.
-
MeOPh
n.d.
n.d.
n.d.
n.d.
-
MeOPh
-
3c
3d
8
naphthalene
naphthalene
MeOPh
-
-
d
b
>10000
no binding
-
2
Me-THP
MeOPh
>10000
>10000
n.d.
-
3e
3f
3g
5
Et-morpholino
Et-morpholino
Me-THP
MeOPh
no binding
no binding
no binding
1969 ± 351.9
-
naphthalene
MeOPh
n.d.
-
n.d.
-
Me-THP
MeOPh
3918 ± 800.4
5628 ± 2890
2
SR144528
51.0 ± 3.0
110
a
3
K obtained by radioligand binding assay, using 2.5 nM [ H]CP55,940 (calculated with a K = 3 nM for hCB
i
d
2
receptor or K = 2 nM for hCB receptor) and transfected HEK 293 3HA cells. SEM = standard error of the
d
1
b
mean. Data is the mean of three individual experiments performed in triplicate, except which is four individual
d
experiments performed in triplicate and which is five individual experiments performed in triplicate. Low
affinity compounds that passed D’Agostino & Pearson normality test and showed significant competition with
1
0

ACCEPTED MANUSCRIPT
3
c
[
H]CP55,940 (as determined by a one sample t-test) were determined to have a K value >10000 nM. no
i
3
significant binding was detected, where compounds showed no significant competition with [ H]CP55,940 as
e
determined by a one sample t-test. n.d. = value was not determined.
Comparison between the hydroxyindole compound series (9a-h, Scheme 1) revealed 7-
hydroxy 9f to have the highest affinity for hCB receptor and the greatest selectivity (approx.
2
1
2
6
6-fold selective) over hCB receptor. With the same R and R groups present in each pair,
1
the 7-hydroxy position gave better hCB receptor affinity than the 5-hydroxy (compare 9f to
2
9
e) and the 5-hydroxy position better than the 6-hydroxy (compare 9a to 9b). The three
2
derivatives with the 5-hydroxy and 4-methoxyphenol R groups held constant, but with
1
varying R groups revealed 4-methyl tetrahydropyran (THP) (9e) to have higher affinity for
hCB receptor compared to 4-ethylmorpholino (9a) and 3-methyl tetrahydrofuran (THF)
2
(
9h), the latter of which showed no appreciable binding to either hCB or hCB receptor. The
1 2
2
1
optimal R group was determined by comparing 5-hydroxy and the same R group and it was
found that either 1-napthalene (compare 9c to 9a and 9d) or 4-methoxyphenyl (compare 9e to
1
9
g) afforded higher hCB receptor affinity. Therefore, the 6-hydroxy, 3-methylTHF R and
2
2
cyclohexane R groups were not included in further derivatisations. Frost et al also found that
1
for 5-hydroxyindole compounds 4-methylTHP at R provides a higher K than
i
ethylmorpholino (when in combination with a tetramethylcyclopropyl group) [33]. However,
Frost et al also found 5-hydroxy substitution conferred higher affinity than 7-hydroxy
1
2
substitution (R = 4-methylTHP, R = tetramethylcyclopropyl) indicating that the preferred
1
2
position of 7-hydroxy found in our study is specific to the R and R groups utilised.
Evaluation of the six derivatives (11a-d and 16a-b, Schemes 1 and 2) with a 4-atom
linker/linear group revealed that the 7-indole-linked position conferred higher hCB receptor
2
affinity than the 5-position. The 7-substituted 11d had a 3-fold higher affinity than 5-
substituted 11c, and 7-substituted 16b around 230-fold higher affinity for CB receptor than
2
1
6a (Table 1, Figure 2a). The weak binding of 11a and 11b compared to 11c indicated a
1
2
preference for R group 4-methylTHP and R group 4-methoxyphenyl. Of all the compounds
analysed, 7-O-propyl 16b (hCB K = 5.7 ± 1.4 nM, hCB K = 504.6 ± 148.2) had the highest
2
i
1
i
hCB receptor affinity and had the most pronounced subtype selectivity (89 fold hCB over
2
2
hCB receptor) of the new compounds reported in this study.
1
1
1

ACCEPTED MANUSCRIPT
Derivatives 19 and 20 were evaluated to see if linker extension from the propyl-chain of lead
compound 16b was tolerated. Despite having a 35-fold reduction in hCB receptor affinity
2
compared to 16b, 19 still retained reasonable hCB receptor affinity (hCB K = 200.7 ± 4.9
2
2
i
nM) (Figure 2a) and was considered a promising lead for a fluorescent conjugate. The
carboxylic acid 20 showed approximately a 6-fold loss in affinity for hCB receptor
2
compared to the methyl ester analogue 19, which indicated it was unlikely the methyl ester
was hydrolysed during the assay.
Conjugation of longer linkers onto pharmacophore-short linker scaffolds was generally not
well tolerated. The longest linked derivative to show appreciable hCB receptor binding was
2
alkyl-ether-acetylated-25. Alkyl, or PEG linkers introduced to derivatives 11a-b, 9e and 20 to
give 13a-d, 18 and 22 showed minimal affinity (hCB receptor K >10000 nM). The lack of
2
i
binding of acetylated analogue 14 compared with the corresponding primary amine 13b
demonstrated the primary amine of 13b is not solely responsible for the lack of affinity.
There was no detectable hCB receptor binding observed for compounds containing a
2
dipeptide linker (13e-g). Unfortunately, conjugation of BODIPY630/650-X did not furnish
high affinity hCB receptor ligands, even when the fluorescent dye was conjugated to the
2
most promising pharmacophore-linker conjugate 19. The alkyl linker containing fluorescent
ligands showed no discernible binding, whilst the PEG and dipeptide-linked fluorescent
ligands showed very minimal affinity (hCB receptor K = >10000 nM; Supplementary Table
2
i
1
).
1
2

ACCEPTED MANUSCRIPT
Figure 2: (A) Binding affinity of 16a, 16b and 19 at hCB receptor, determined using competition binding
2
3
3
curves against [ H]CP55,940. [ H]CP55,940 bound (ccpm) values are normalised to the specific binding
window (%). Data shown is representative of a single experiment performed in triplicate, with error bars
indicating ± SEM. (B) Concentration response curve of 9e, 9f and CP55,940, measuring forskolin-stimulated
cAMP (5µM). Area under the curve values are normalised so that forskolin only response = 100% and basal
response = 0%. Data shown is representative of a single experiment, with data points conducted in duplicate. (C)
Concentration response curve of 16b and SR144528, measuring forskolin-stimulated cAMP. Area under the
curve values are normalised so that forskolin only response = 100% and basal response = 0%. Data shown is
representative of a single experiment, with data points conducted in duplicate.
1
3

ACCEPTED MANUSCRIPT
2
.3. cAMP assays
Compounds that showed micromolar or nanomolar affinity for hCB receptor were also
2
analysed for functional activity using a bioluminescence resonance energy transfer (BRET)
sensor to measure modulation of forskolin-stimulated cyclic adenosine monophosphate
(
cAMP) at hCB and hCB receptors. Initially compounds were tested at 10 µM in the
2 1
presence and absence of CP 55,940. Since both CB and CB receptor are G coupled,
1
2
i
agonists cause a decrease in cellular cAMP, antagonists can prevent the inhibition produced
by an EC concentration of a full agonist (i.e. CP 55,940) and inverse agonists lead to an
9
0
increase in cellular cAMP. Compounds that demonstrated a functional response were then
subjected to a concentration response assay at CB receptor and the three highest CB
2
2
receptor affinity compounds, 9e, 9f and 16b were also analysed for potency at CB receptor
1
(Table 2).
1
2
The five indoles with a 5-hydroxyl but varying R and R groups (9a, 9c-e, 9g) all behaved as
hCB receptor agonists in the cAMP assay. The compound with the highest affinity for CB
2
2
receptor (9e) was also the most potent agonist. 5-Hydroxy-9e had the highest hCB receptor
2
agonist potency of all compounds tested (hCB EC = 4.4 ± 0.35 nM) but also showed
2
50
agonist activity at hCB receptor at a lower potency (hCB EC = 62.9 ± 29.7 nM).
1
1
50
Extension of the 5-hydroxy position with a short chain either retained agonist activity (11c)
or resulted in no response (16a).
In contrast to the 5-substituted agonists, derivatisation of the alkyl indole scaffold in the 7-
position led to inverse agonist activity at hCB receptor in five (9f, 16b, 19, 20, 25) out of six
2
compounds analysed. Of note is the dramatic difference in function between the regioisomers
5
-hyroxy-9e and 7-hydroxy-9f, which showed agonism or inverse agonism at hCB receptor
2
respectively (Figure 2b). These two compounds were also analysed for function at hCB1
receptor and both behaved as agonists, with 9e having the highest potency (hCB EC = 62.9
1
50
±
29.7 nM).
1
4

ACCEPTED MANUSCRIPT
Table 2: Functional data at hCB and hCB receptor.
2
1
hCB receptor
hCB receptor
2
1
a
b
a
b
Compound
EC50 or IC50
Emax
Function
EC50 or IC50
Emax
Function
e
e
(nM ± SEM)
(% ± SEM)
(nM ± SEM)
(% ± SEM)
a
9
9
9
9
9
1
1
a
5.4 ± 2.7
8.7 ± 1.6
67.5 ± 2.9
77.4 ± 6.1
74.1 ± 1.2
65.7 ± 3.7
64.4 ± 1.5
57.4 ± 3.2
-
Agonist
Agonist
Agonist
Agonist
Agonist
Agonist
No response
n.d.
n.d.
n.d.
70.4 ± 4.6
54.6 ± 5.4
74.3 ± 4.2
62.2 ±3.4
52.4 ± 1.9
60.4 ± 1.2
-
Agonist
Agonist
Agonist
Agonist
Agonist
Agonist
No response
a
c
a
a
a
a
d
18.6 ± 5.1
4.4 ± 0.35
14.7 ± 5.3
70.2 ± 6.2
-
a
e
62.9 ± 29.7
g
n.d.
n.d.
-
1c
6a
b
a
9
f
271.7 ± 130.7
189.8 ± 13.8
Inverse agonist
632.5 ± 315.1
64.3 ± 8.1
Agonist
b
1
1
1
2
2
6b
1d
9
30.6 ± 13.9
263.3 ± 14.4
84.9 ± 3.3
Inverse agonist
Agonist
-
-
No response
Agonist
Agonist
Agonist
Agonist
Agonist
a,
>10000
n.d.
n.d.
n.d.
n.d.
n.d.
-
80.5 ± 5.5
77.8 ± 3.6
82.5 ± 1.6
74.3 ± 4.9
56.5 ± 1.4
-
b, c
>10000
165.8 ± 7.1
123.6 ± 7.7
150.9 ± 10.7
41.0 ± 1.4
Inverse agonist
Inverse agonist
Inverse agonist
Agonist
b, c
0
>10000
b, c
5
>10000
a
CP-55,940
SR144528
5.6 ± 2.3
b,d
760 ± 61.7
212.2 ± 2.5
Inverse agonist
No response
a
Potency was determined by concentration response BRET assays measuring forskolin-stimulated cAMP. EC
5
0
b
c
for compounds showing agonist activity. IC for compounds showing inverse agonist activity. Compounds
50
did not reach a plateau for maximum efficacy. Data is the mean of at least three individual experiments
d
performed in duplicate, except which is two individual experiments performed in duplicate. SEM = standard
e
error of the mean. Area under the curve (AUC) analysis was used to determine E as a percentage of
max
normalised forskolin only response (100%) and vehicle (0%). A t-tests was used to determine if the Emax
observed was significantly different from the forskolin only response and if no significant difference was found,
the compounds were deemed to show no response. n.d. = value was not determined.
It remains to be determined if this regioisomer functional switch between 5- and 7-substituted
indoles is a more robust trend as there is no additional SAR available in the literature directly
comparing the function of 5- and 7-substituted analogues with regards to cAMP signalling.
Pasquini et al identified 5-aryl-substituted indoles that displayed inverse agonism with cAMP
signalling, however with very different N1 and C3 indole substituents than those utilised in
our study [34]. In an interesting study by Frost et al [33] using 4-, 5-, 6- and 7-hydroxyl, O-
benzyl and methoxy substituted indoles, the 7-substituted analogues failed to show agonism
in a fluorescence imaging plate reader (FLIPR) calcium flux assay (but were not tested for
1
5

ACCEPTED MANUSCRIPT
antagonism or inverse agonism), whilst two out of three 5-substituted analogues did display
agonism.
It is important to note that we only analysed the G pathway via cAMP production in this
i
study and no other CB receptor signalling pathways such as β-arrestin recruitment, Gq
coupling (intracellular calcium release), or G coupling (adenyl cyclase stimulation). GPCR
s
ligands are well known to exhibit signalling bias [46, 47], and therefore the new ligands
reported herein could possess varying functions via signalling pathways other than cAMP.
The compound with the highest affinity for CB receptor, 16b, also showed the highest hCB2
2
receptor potency (hCB receptor IC = 30.6 ± 13.9 nM) amongst the inverse agonists in this
2
50
series, and a higher potency and Emax than the control inverse agonist SR144528 (Figure 2c).
6b was also analysed for hCB receptor potency but failed to produce a significant response
1
1
at concentrations up to 10 µM. In comparison to previously reported alkyl indole scaffolds,
6b showed improved hCB receptor affinity over AM630 5 (hCB Ki = 32.1 nM,
1
2
2
hCB /hCB 165-fold) [30] but with less subtype selectivity. There are a very few higher
2
1
affinity alkyl indoles that have been reported as CB receptor inverse agonists, for example
2
3
an inverse agonist (hCB K = 0.37 nM, hCB K = 344.9 nM, calculated using [ H]CP55,940
2
i
1
i
hCB K = 0.31 nM, hCB K = 0.18 nM) reported by Pasquini et al [34]. In contrast, there
2
d
1
d
are many higher affinity CB receptor indole-based agonists reported, which could be
2
attributed to the increased interest in CB receptor agonists driven by drug discovery efforts
[
48]. 16b also showed a higher affinity than diarylpyrazole inverse agonist SR144528 in this
study.
Compound 19 was the highest affinity pharmacophore-linker conjugate, and 25 was the
longest linked derivative to show appreciable hCB receptor binding, however despite both
2
behaving as an inverse agonist at CB receptor, an IC could not be estimated for either as a
2
50
maximal response plateau was not reached at 10 µM.
All the compounds analysed for potency in the cAMP BRET assay were also tested at 10 µM
and 1 µM in wild type (WT) HEK 293 cells to verify that the observed effects at hCB and
2
hCB receptors were receptor mediated (Supplementary Table 2). 9c, 9f, 16a-b, and 20 all
1
invoked a small but significant response at 10 µM but not at 1 µM and as these compounds
all demonstrated a response in HEK293-hCB cells at 1 µM, it can be concluded that the WT
2
HEK cell response is not large enough to have significantly affected the calculated hCB and
1
1
6

ACCEPTED MANUSCRIPT
hCB receptor potencies. 19 and 25 showed a significant response in WT HEK cells at both
1
1
0 µM and 1 µM, but as both compounds demonstrated very weak potency at hCB receptor
2
(
IC >10000nM) the potential effect of non-receptor mediated activity on the measured CB
50
receptor response is of limited concern.
2
.4. Molecular Modelling
GPCRs have proved difficult to crystallise due to the inherent insolubility of membrane
bound proteins, low expression levels, conformational flexibility and lack of stability in
detergents [49]. However, recent technological advances such as protein engineering and
lipid-based crystallisation have led to a rapid increase in the numbers of GPCR crystal
structures obtained [50]. A crystal structure of CB receptor has not yet been reported,
2
however three structures were recently determined for CB receptor [51-53]. A homology
1
model of CB receptor based on the crystal structure of inverse-agonist-bound CB receptor
2
1
(PDB ID: 5TGZ) [51] was constructed. The secondary structure of a homology model is
heavily influenced by the template structure. CB receptor homology models reported prior to
2
the determination of CB crystal structures have utilised other, less closely related crystal
1
structures, such as β2-adrenoceptor, A adenosine receptor or rhodopsin and therefore will
2
A
differ slightly to the model reported here [54-56].
Ligand docking studies were carried out in an attempt to rationalise the hCB receptor affinity
2
of the inverse agonist pharmacophore-linker conjugate 19. Docking poses were clustered into
two groups, with the most consistent pose showing the methyl valerate linker oriented out of
the binding pocket between TM1 and TM7, into the lipid bilayer (Figure 3). It has previously
been proposed that ligands may enter CB and CB receptors via the lipid membrane [57] and
1
2
molecular dynamics simulations have shown a likely entry route between TM1 and TM7 for
tetrahydrocannabinol and anandamide binding to CB receptor [58]. A similar exit pathway
1
from the orthosteric binding site of CB receptor between TM1 and TM7 to that postulated
2
here for 19 has been proposed for a biotin-containing CB receptor probe [59]. In this pose,
2
the 7-ether demonstrates hydrogen bonding with S285, whilst the THP oxygen and the ester
carbonyl hydrogen bonds with the extracellular loop. The phenyl is positioned in a
hydrophobic pocket between TM3-6 and the indole core and phenyl are able to engage in van
der Waals interactions with multiple nearby aromatic residues (F183, W194, W258 and
F281). This linker pose reinforces that the indole C7 position holds promise for linker and
1
7

ACCEPTED MANUSCRIPT
fluorescent dye conjugation. The other, smaller cluster of poses (three out of ten) showed the
linker exiting through the extracellular loops, however the flexibility of extracellular loops
makes docking in this region difficult.
Figure 3: Inverse agonist 19 (cyan carbon atoms) docked into hCB receptor homology model (grey ribbon).
2
The methyl valerate linker is orientated out of the binding pocket between helices 1 (TM1) and 7 (TM7).
Hydrogen bonding (yellow dotted lines) is shown between the 7-ether and S285 (shown as sticks) as well as the
THP ether and the ester carbonyl with the extracellular loop. Side chains of residues within 4 Å of the ligand 19
are shown as sticks (green).
2
.5. Off target activity
Compounds 8e-h, 9e-g and 16a were accepted into the Eli Lilly and Company Open
Innovation Drug Discovery Program and the Tres Cantos Open Lab Foundation and did not
show significant proprotein convertase subtilisin kexin type 9 (PCSK9) inhibition, GPR120
agonism, disruption to IL-17 protein-protein interaction, Leishmania donovani growth
inhibition, voltage-gated potassium channel KCNQ2/3 agonism or nicotinamide N-
methyltransferase (NNMT) inhibition.
1
8

ACCEPTED MANUSCRIPT
3
. CONCLUSIONS
In this study, the C5-C7 positions of CB receptor alkyl indole ligands were explored for
2
tolerability of linker attachment, with the aim to develop a high affinity, selective CB2
receptor fluorescent ligand. Substitution of short linkers at the indole C7 resulted in higher
affinity and selectivity for hCB receptor compared to the analogous C5 compounds. These
2
C7 substituted compounds also showed inverse agonism in contrast to the C5-substituted
agonists in nearly all examples. C7-Short alkyl-linked indole 16b (hCB K = 5.7 ± 1.4 nM,
2
i
hCB K = 504.6 ± 148.2, hCB /hCB 88-fold) was identified as a new high affinity hCB
2
1
i
2
1
receptor inverse agonist.
The pharmacophore-linker conjugate 19 that retained CB receptor affinity demonstrated that
2
a linker is tolerated in this indole C7 position and molecular modelling suggests this would be
amenable to further extension and fluorophore conjugation. Unfortunately, conjugation of a
fluorescent dye, including extension of 19, failed to produce a high affinity CB receptor
2
fluorescent ligand in this study. However, 19 remains a promising pharmacophore-linker lead
and future studies could explore extension of 19 with other linker lengths and types in
addition to different fluorescent dyes. In the context of this study that investigated C5-C7
indole positions and in light of previous efforts of linker conjugation via the C3 acyl group
[
35] or extension of the N1 substituent [29] the search for an indole-based fluorescent ligand
for CB receptor remains elusive.
2
EXPERIMENTAL PRODECURES
Material and Methods
Chemistry
Chemicals and solvents were purchased from Sigma Aldrich, Merck or AK Scientific and
were used without further purification. BODIPY 630/650-X-OSu was obtained from Life
Technologies. Anhydrous grade solvents were used when a dry atmosphere was required.
Unless stated, all reactions were carried out at room temperature (rt) and under atmospheric
pressure. 1,2-diaminoethane trityl resin (200-400 mesh) was purchased from Merck.
1
9

ACCEPTED MANUSCRIPT
Thin layer chromatography was carried out on 0.2 mm aluminium-backed silica gel plates 60
F254 and visualised under UV light at λ = 254 nM and 365 nM and then with potassium
permanganate dip. Flash column chromatography was carried out using 40-63 µm silica.
Reverse phase high performance liquid chromatography (RP-HPLC) was carried out on an
Agilent 1260 Infinity system, using an YMC C8 5 µm (150 × 10 mm) column for semi-
preparative RP-HPLC and an YMC C8 5 µm (150 × 4.6 mm) column for analytical RP-
HPLC. The mobile phases used were A: H O (0.05% TFA) and B: 9:1 ACN:H O (0.05%
2
2
TFA). Analytical RP-HPLC retention times quoted below were determined with a standard
method - 5% A for 1 min, then a linear gradient of 5-95% B from 1-27 min (followed by 1
min at 95% B, 2 min linear gradient 95-5% B and then 4 min re-equilibration at 5% A). All
compounds analysed for biological activity were > 95% purity by UV detection at 254 and
3
80 nm (and 550 nm for fluorescent compounds) by analytical RP-HPLC. All compounds
HPLC purified as the trifluoroacetic acid (TFA) salt were neutralised using an Amberlyst
A21 ion exchange resin before biological testing.
High resolution electrospray ionization mass spectra (HRMS-ESI) were obtained on a
microTOF mass spectrometer. Proton and carbon nuclear magnetic resonance (NMR)
Q
spectra were obtained on either a 400 MHz or a 500 MHz Varian MR spectrometer. Two-
dimensional NMR experiments, including COSY, HSQC, HMBC and NOESY were used to
assign spectra. Chemical shifts are listed on the δ scale in ppm, referenced to CDCl , MeOD-
3
d or DMSO-d with residual solvent as the internal standard and coupling constants (J)
4
6
recorded in hertz (Hz). Signal multiplicities are assigned as: s, singlet; d, doublet; t, triplet; q,
quartet; quint, quintet; dd, doublet of doublets; dt, doublet of triplets; td, triplet of doublets;
br, broad; or m, multiplet.
5
� (Benzyloxy)� 1� [2� (morpholin� 4� yl)ethyl]� 1H� indole (7a)
Preparation of 2-(morpholin-4-yl)ethyl methanesulfonate. A stirred solution of 4-(2-
hydroxyethyl)-morpholine (2.5 mL, 20.7 mmol) and Et N (8.6 mL, 62.0 mmol) in anhydrous
3
DCM (48 mL) under N was cooled to 0°C and then methanesulfonyl chloride (2.4 mL, 31.0
2
mmol) was added dropwise. The mixture was stirred at rt for 2 h, filtered, the solid washed
with minimal DCM and this filtrate combined with the original filtrate was evaporated under
reduced pressure to give a yellow oil (10.24 g) as a mixture of the desired product 2-
+
+
(
morpholin-4-yl)ethyl methanesulfonate and a salt (Et NH Cl ). This material was used
3
2
0

ACCEPTED MANUSCRIPT
without further purification. A stirred solution of 5-benzyloxyindole (600 mg, 2.7 mmol) in
anhydrous DMF (11 mL) under N was cooled to 0°C, and then NaH (60% by mass
2
dispersion in mineral oil; 358 mg, 9.0 mmol) was added. The mixture was stirred at 0°C for
1
0 min and then at rt for 30 min. The reaction mixture was cooled to 0°C and a solution of 2-
(morpholin-4-yl)ethyl methanesulfonate (1.13 g, 5.4 mmol) in anhydrous DMF (5 mL) was
added. The mixture was stirred at 45°C for 2 h, cooled to rt, diluted with EA (20 mL) and
quenched with sat. aq. NH Cl (20 mL) and H O (10 mL). The layers were separated and the
4
2
aqueous layer extracted with EA (3 × 20 mL). The combined organics were washed with H O
2
(
4 × 70 mL), dried over MgSO , filtered and evaporated under reduced pressure. The crude
4
product was purified using flash silica column chromatography (2:1 PE:EA) to yield the
1
desired 7a (690 mg, 2.1 mmol, 76%) as a light brown wax (R 0.13, 2:1 PE:EA). H NMR
f
(
400 MHz, CDCl ): δ 7.52 – 7.46 (m, 2H, ArH Bn), 7.43 – 7.37 (m, 2H, ArH Bn), 7.37 – 7.30
3
(m, 1H, ArH Bn), 7.26 (d, J = 8.8 Hz, 1H, ArH indole), 7.18 (d, J = 2.3 Hz, 1H, ArH indole),
7
.13 (d, J = 3.1 Hz, 1H, ArH indole), 6.98 (dd, J = 8.9, 2.2 Hz, 1H ArH indole), 6.41 (dd, J =
.2, 1.1 Hz, 1H, ArH indole), 5.12 (s, 2H, CH Bn), 4.22 (t, J = 6.9 Hz, 2H, N1-CH ), 3.72 (t,
3
2
2
J = 4.6 Hz, 4H, O-CH morpholino), 2.75 (t, J = 6.9 Hz, 2H, N1-CH CH ), 2.49 (t, J = 4.6
2
2
2
1
3
Hz, 4H, N-CH morpholino). C NMR (101 MHz, CDCl ) δ 153.33, 137.86, 131.56, 128.98,
2
3
1
4
6
6
28.69, 128.61, 127.85, 127.64, 112.69, 110.03, 104.38, 101.03, 71.01, 67.03, 58.36, 53.99,
+
4.29. HRMS-ESI calculated for C H N O [M+H] 337.1911, found m/z 337.1884.
2
1
25
2
2
-(Benzyloxy)-1-[2-(morpholin-4-yl) ethyl]-1H indole (7b)
-Benzyloxyindole (1.34 g, 6.0 mmol), NaH (732 mg, 18.3 mmol), and 2-(morpholin-4-
yl)ethyl methanesulfonate (2.51 g, 12.0 mmol) in DMF (48 mL) were reacted as described in
the procedure for 7a. The crude product was purified using flash silica column
chromatography (2:1 hexane:EA) to yield the desired 7b (1.10 g, 3.3 mmol, 54%) as a brown
1
solid (R 0.13, 2:1 PE (petroleum ether):EA). H NMR (400 MHz, CDCl ) δ 7.55 – 7.44 (m,
f
3
3
H, ArH Bn & ArH indole), 7.44 – 7.36 (m, 2H, ArH Bn), 7.36 – 7.29 (m, 1H, ArH Bn), 7.04
(d, J = 3.2 Hz, 1H, ArH indole), 6.95 – 6.83 (m, 2H, ArH indole), 6.42 (dd, J = 3.2, 0.7 Hz,
1
H, ArH indole), 5.14 (s, 2H, CH Bn), 4.32 – 4.04 (br m, 2H, N1-CH ), 3.82 – 3.57 (br m,
2 2
4
H, O-CH morpholino), 2.83 – 2.62 (br m, 2H, N1-CH CH ), 2.60 – 2.29 (br m, 4H, N-CH
2
2
2
2
1
3
morpholino). C NMR (101 MHz, CDCl ) δ 155.44, 137.60, 136.67, 128.67, 127.97, 127.63,
3
1
27.26, 123.27, 121.68, 109.95, 101.38, 94.83, 70.94, 67.02, 58.08, 53.99, 44.10. HRMS-ESI
+
calculated for C H N O [M+H] 337.1911, found m/z 337.1880.
2
1
25
2
2
2
1

ACCEPTED MANUSCRIPT
5
-(Benzyloxy)-1-[(oxan-4-yl) methyl]-1H-indole (7c)
Preparation of (oxan� 4� yl)methyl methanesulfonate. 4-(Hydroxymethyl)
tetrahydropyran (2.9 mL, 25.0 mmol), Et N (10.4 mL, 74.9 mmol) and methanesulfonyl
3
chloride (2.9 mL, 37.5 mmol) in DCM (50 mL) were reacted as described in the procedure
for 2-(morpholin-4-yl)ethyl methanesulfonate to give an orange oil (7.77 g) as a mixture of
the desired (oxan‐ 4‐ yl)methyl methanesulfonate and a salt. This material was used without
further purification. 5-Benzyloxyindole (600 mg, 2.7 mmol), NaH (481 mg, 12.0 mmol) and
(oxan‐ 4‐ yl)methyl methanesulfonate (1.04 g, 5.4 mmol) in DMF (17 mL) were reacted as
described in the procedure for 7a. The crude product was purified using flash silica column
chromatography (2:1 PE:EA) to yield the desired 7c (570 mg, 1.8 mmol, 66%) as a light
1
brown oil (R 0.44, 2:1 PE:EA). H NMR (400 MHz, CDCl ) δ 7.52 – 7.44 (m, 2H, ArH Bn),
f
3
7
.43 – 7.36 (m, 2H, ArH Bn), 7.35 – 7.29 (m, 1H, ArH Bn), 7.23 (d, J = 8.9 Hz, 1H, ArH
indole), 7.17 (d, J = 2.4 Hz, 1H, ArH indole), 7.03 (d, J = 2.4 Hz, 1H, ArH indole), 6.96 (dd,
J = 8.9, 2.4 Hz, 1H, ArH indole), 6.40 (d, J = 3.0 Hz, 1H, ArH indole), 5.11 (s, 2H, CH Bn),
2
3
.99 – 3.91 (m, 4H, N-CH & O-CH tetrahydropyran (THP)), 3.31 (td, J = 11.7, 2.3 Hz, 2H,
2 2
13
O-CH THP), 2.16 – 2.02 (m, 1H, CH THP), 1.53 – 1.32 (m, 4H, O-CH CH THP). C
2
2
2
NMR (101 MHz, CDCl ) δ 153.29, 137.85, 131.77, 129.02, 128.96, 128.61, 127.85, 127.64,
3
1
12.68, 110.25, 104.24, 100.72, 70.97, 67.58, 52.65, 36.40, 30.98. HRMS-ESI calculated for
+
C H NO [M+H] 322.1802, found m/z 322.1778.
21
24
2
7
-(Benzyloxy)-1-[(oxan-4-yl) methyl]-1H-indole (7d)
7
-Benzyloxyindole (335 mg, 1.5 mmol), NaH (268 mg, 6.7 mmol) and (oxan‐ 4‐ yl)methyl
methanesulfonate (582 mg, 3.0 mmol) in DMF (14 mL) were reacted as described in the
procedure for 7a. The crude product was purified using flash silica column chromatography
(4:1 hexane:EA) to yield the desired 7d (418 mg, 1.3 mmol, 87%), as a pale violet wax (Rf
1
0
.63, 2:1 PE:EA). H NMR (400 MHz, CDCl ) δ 7.51 – 7.44 (m, 2H, ArH Bn), 7.44 – 7.33
3
(m, 3H, ArH Bn), 7.24 (dt, J = 8.0, 0.7 Hz, 1H, ArH indole), 7.00 (t, J = 7.8 Hz, 1H, ArH
indole), 6.91 (d, J = 2.9 Hz, 1H, ArH indole), 6.73 (dd, J = 7.8, 0.8 Hz, 1H, ArH indole), 6.40
(
d, J = 3.0 Hz, 1H, ArH indole), 5.14 (s, 2H, CH Bn), 4.08 (d, J = 7.2 Hz, 2H, N-CH ), 3.87
2
2
–
1
1
3.76 (m, 2H, O-CH THP), 3.13 (td, J = 11.6, 2.5 Hz, 2H, O-CH THP), 2.11 – 1.92 (m,
2 2
1
3
H, CH THP), 1.26 – 1.03 (m, 4H, O-CH CH THP). C NMR (101 MHz, CDCl ) δ 146.85,
2
2
3
36.98, 131.35, 130.02, 128.71, 128.56, 128.41, 125.57, 119.87, 114.12, 103.02, 100.83,
2
2

ACCEPTED MANUSCRIPT
+
7
3
5
0.60, 67.70, 55.41, 37.68, 30.43. HRMS-ESI calculated for C H NNaO [M+Na]
2
1
23
2
44.1621, found m/z 344.1591.
-(Benzyloxy)-1-[(oxolan-3-yl) methyl]-1H-indole (7e)
Preparation of (oxolan-3-yl)methyl methanesulfonate. Tetrahydro-3-furanmethanol (0.86
mL, 9.0 mmol), Et N (3.7 mL, 26.9 mmol) and methanesulfonyl chloride (1.0 mL, 13.4
3
mmol) in DCM (21 mL) were reacted as described in the procedure for 2-(morpholin-4-
yl)ethyl methanesulfonate to give a yellow oil (3.29 g) as a mixture of the desired (oxolan‐
3
‐ yl)methyl methanesulfonate and a salt. This material was used without further
purification. 5-Benzyloxyindole (600 mg, 2.7 mmol), NaH (322 mg, 8.1 mmol) and (oxolan‐
‐ yl)methyl methanesulfonate (969 mg, 5.4 mmol) in DMF (16 mL) were reacted as
3
described in the procedure for 7a. The crude product was purified using flash silica column
chromatography (2:1 PE:EA) to yield the desired 7e (582 mg, 1.9 mmol, 70%), as a red-
1
orange oil (R 0.63, 2:1 EA:PE). H NMR (400 MHz, CDCl ) δ 7.55 – 7.46 (m, 2H, ArH Bn),
f
3
7
.44 – 7.37 (m, 2H, ArH Bn), 7.37 – 7.30 (m, 1H, ArH Bn), 7.27 (d, J = 8.9 Hz, 1H, ArH
indole), 7.19 (d, J = 2.5 Hz, 1H, ArH indole), 7.08 (d, J = 3.1 Hz, 1H, ArH indole), 6.98 (dd,
J = 8.9, 2.4 Hz, 1H, ArH indole), 6.43 (dd, J = 3.1, 0.8 Hz, 1H, ArH indole), 5.12 (s, 2H, CH2
Bn), 4.06 (d, J = 7.8 Hz, 2H, N-CH ), 3.97 (td, J = 8.3, 5.5 Hz, 1H, O-CH tetrahydrofuran
2
2
(
THF)), 3.82 – 3.67 (m, 2H, O-CH THF), 3.61 (dd, J = 8.9, 4.7 Hz, 1H, O-CH THF), 2.90 –
2
2
1
3
2
.75 (m, 1H, CH THF), 2.10 – 1.94 (m, 1H, CH THF), 1.73 – 1.57 (m, 1H, CH THF). C
2 2
NMR (101 MHz, CDCl ) δ 153.39, 137.84, 131.69, 129.02, 128.63, 128.43, 127.87, 127.65,
3
1
12.84, 110.11, 104.36, 101.14, 71.13, 71.01, 67.68, 49.22, 40.16, 29.97. HRMS-ESI
+
calculated for C H NNaO [M+Na] 330.1464, found m/z 330.1441.
2
0
21
2
5
-(Benzyloxy)-3-(4-methoxybenzoyl)-1-[2-(morpholin-4-yl)ethyl]-1H-indole (8a)
Formation of [(Phen)Pd(OAc) ] complex. Palladium (II) acetate (324 mg, 1.5 mmol) was
2
dissolved in acetone (28 mL) and left unstirred for 30 min at rt, after which, undissolved solid
Pd(OAc) was removed by filtration with fine filter paper. 1,10-Phenanthroline (313 mg, 1.7
2
mmol) was added to the reddish-brown filtrate, the solution swirled for 1 min in which time a
precipitate started to form. The solution was left to sit unstirred for 30 min, the precipitate
was filtered, washed with cold acetone and vacuum dried, yielding [(Phen)Pd(OAc) ] (336
2
mg, 0.83 mmol, 57%) as a canary yellow solid, as described in the literature [60]. A stirred
solution of 7a (104 mg, 0.31 mmol), 4-methoxybenzonitrile (119 mg, 0.89 mmol) and
[
(Phen)Pd(OAc) ] (13 mg, 31.0 µmol) dissolved in H O (0.12 mL), glacial AcOH (0.18 mL)
2 2
2
3

ACCEPTED MANUSCRIPT
and 1,4-dioxane (0.6 mL) was heated to 140°C in a sealed pressure tube for 42 h. The
reaction mixture was cooled to rt, diluted with DCM (18 mL), filtered through a celite pad
and the celite washed with DCM (6 mL). The filtrate was evaporated under reduced pressure.
The crude residue was purified by flash silica column chromatography (99:1 DCM:MeOH
with 0.3% Et N to load silica), and recrystallized using MeOH to yield 8a (74 mg, 0.16
3
1
mmol, 51%) as dark brown crystals (R 0.53, 95:5 DCM:MeOH). H NMR (400 MHz,
f
CDCl ) δ 8.06 (d, J = 2.4 Hz, 1H, ArH indole), 7.90 – 7.81 (m, 2H, ArH MeOPh), 7.65 (s,
3
1
1
H, ArH indole), 7.53 – 7.47 (m, 2H, ArH Bn), 7.43 – 7.36 (m, 2H, ArH Bn), 7.35 – 7.31 (m,
H, ArH Bn), 7.29 (dd, J = 8.9, 0.5 Hz, 1H, ArH, indole), 7.05 (dd, J = 8.9, 2.5 Hz, 1H, ArH
indole), 7.02 – 6.96 (m, 2H, ArH MeOPh), 5.17 (s, 2H, CH Bn), 4.22 (t, J = 6.3 Hz, 2H, N1-
2
CH ), 3.90 (s, 3H, O-CH ), 3.70 (t, J = 4.6 Hz, 4H, O-CH morpholino), 2.77 (t, J = 6.4 Hz,
2
3
2
1
3
2
H, N1-CH CH ), 2.48 (t, J = 4.7 Hz, 4H, N-CH morpholino). C NMR (101 MHz, CDCl )
2 2 2 3
δ 189.85, 162.30, 155.70, 137.49, 137.18, 133.59, 131.90, 130.98, 128.64, 128.31, 127.95,
1
27.81, 115.46, 114.71, 113.63, 110.48, 105.36, 70.69, 67.02, 57.84, 55.58, 53.82, 44.45.
+
HRMS-ESI calculated for C H N O [M+H] 471.2278, found m/z 471.2239.
2
9
31
2
4
6
-(Benzyloxy)-3-(4-methoxybenzoyl)-1-[2-(morpholin-4-yl)ethyl]-1H-indole (8b)
A solution of 7b (123 mg, 0.37 mmol), 4-methoxybenzonitrile (144 mg, 1.1 mmol) and
(Phen)Pd(OAc) ] (14 mg, 37.0 µmol) in H O (0.14 mL), glacial AcOH (0.21 mL) and 1,4-
[
2
2
dioxane (0.7 mL) was reacted as described in the procedure for 8a. The crude residue was
purified by flash silica column chromatography (99:1 DCM:MeOH, with 0.3% ammonia to
load silica) to yield 8b (129 mg, 0.27 mmol, 75%) as a brown solid (R 0.47, 95:5
f
1
DCM:MeOH). H NMR (500 MHz, CDCl ) δ 8.29 (d, J = 8.7 Hz, 1H, ArH indole), 7.87 –
3
7
7
.80 (m, 2H, ArH MeOPh), 7.56 (s, 1H, ArH indole), 7.50 – 7.44 (m, 2H, ArH Bn), 7.42 –
.36 (m, 2H, ArH Bn), 7.36 – 7.30 (m, 1H, ArH Bn), 7.05 (dd, J = 8.7, 2.2 Hz, 1H, ArH
indole), 7.00 – 6.93 (m, 2H, ArH MeOPh), 6.90 (d, J = 2.2 Hz, 1H, ArH indole), 5.14 (s, 2H,
CH Bn), 4.15 (t, J = 6.4 Hz, 2H, N1-CH ), 3.87 (s, 3H, O-CH ), 3.68 (t, J = 4.6 Hz, 4H, O-
2
2
3
CH morpholino), 2.72 (t, J = 6.4 Hz, 2H, N1-CH CH ), 2.46 (t, J = 4.7 Hz, 4H, N-CH
2
2
2
2
1
3
morpholino). C NMR (126 MHz, CDCl ) δ 189.64, 162.25, 156.39, 137.55, 137.15, 136.38,
3
1
7
4
5
33.43, 130.93, 128.65, 128.02, 127.57, 123.54, 121.79, 115.74, 113.53, 112.23, 95.19,
+
0.75, 66.94, 57.51, 55.50, 53.74, 44.12. HRMS-ESI calculated for C H N O [M+H]
2
9
31
2
4
71.2276, found m/z 471.2278.
-(Benzyloxy)-1-[2-(morpholin-4-yl)ethyl]-3-(naphthalene-1-carbonyl)-1H-indole (8c)
2
4

ACCEPTED MANUSCRIPT
A solution of 7a (525 mg, 1.6 mmol), 1-naphthonitrile (717 mg, 4.7 mmol) and
[
(Phen)Pd(OAc) ] (76 mg, 0.19 mmol) in H O (0.63 mL), glacial AcOH (0.93 mL) and 1,4-
2 2
dioxane (3.1 mL) was reacted as described in the procedure for 8a. The crude residue was
st
purified by flash silica column chromatography twice (1 99:1 DCM:MeOH, with 0.3% Et N
3
nd
to load silica, 2 10:1 EA:hexane with 0.5% ammonia to load silica) to yield 8c (138 mg,
1
0
.28 mmol, 18%) as a brown solid (R 0.34, 4:1 EA:hexane). H NMR (400 MHz, CDCl ) δ
f
3
8
.22 – 8.15 (m, 2H, ArH indole & naphthalene), 8.00 – 7.89 (m, 2H, ArH naphthalene), 7.65
(dd, J = 7.0, 1.2 Hz, 1H, ArH naphthalene), 7.57 – 7.45 (m, 5H, ArH Bn & naphthalene),
7
8
3
4
1
1
4
5
.45 – 7.37 (m, 3H, ArH Bn & indole), 7.37 – 7.27 (m, 2H, ArH Bn & indole), 7.08 (dd, J =
.9, 2.5 Hz, 1H, ArH indole), 5.20 (s, 2H, CH Bn), 4.29 – 3.97 (br m, 2H, N1-CH ), 3.76 –
2
2
.36 (br m, 4H, O-CH morpholino), 2.85 – 2.58 (br m, 2H, N1-CH CH ), 2.57 – 2.19 (br m,
2
2
2
1
3
H, N-CH morpholino). C NMR (101 MHz, CDCl ) δ 192.18, 156.04, 139.21, 138.98,
2
3
37.42, 133.83, 132.12, 130.87, 130.04, 128.66, 128.32, 128.00, 127.86, 127.76, 126.89,
26.43, 126.05, 125.80, 124.60, 117.54, 114.86, 110.71, 105.55, 70.72, 66.87, 57.57, 53.65,
+
4.40. HRMS-ESI calculated for C H N O [M+H] 491.2329, found m/z 491.2303.
3
2
31
2
3
-(Benzyloxy)-3-cyclohexanecarbonyl-1-[2-(morpholin-4-yl)ethyl]-1H-indol (8d)
A solution of 7a (100 mg, 0.3 mmol), cyclohexanecarbonitrile (106 µL, 0.89 mmol) and
(Phen)Pd(OAc) ] (12 mg, 30.0 µmol) in H O (0.12 mL), glacial AcOH (0.18 mL) and 1,4-
[
2
2
dioxane (0.6 mL) was reacted as described in the procedure for 8a. The crude residue was
purified by flash silica column chromatography (99:1 DCM:MeOH, with 0.3% Et N to load
3
silica) to yield a red oil as a 4:1 mixture of 8d and 7a (100 mg) (R 8d 0.17, 99:1
f
DCM:MeOH). This mixture could not be separated further and was therefore used without
further purification in subsequent reactions. HRMS-ESI calculated for 8d C H N O
3
28
35
2
+
[
M+H] 447.2642, found m/z 447.2605 (from HRMS-ESI of the 4:1 mixture).
-(Benzyloxy)-3-(4-methoxybenzoyl)-1-[(oxan-4-yl)methyl]-1H-indole (8e)
A solution of 7c (103 mg, 0.32 mmol), 4-methoxybenzonitrile (127 mg, 0.95 mmol) and
(Phen)Pd(OAc) ] (14 mg, 32.0 µmol) in H O (0.12 mL), glacial AcOH (0.18 mL) and 1,4-
5
[
2
2
dioxane (0.6 mL) was reacted as described in the procedure for 8a. The crude residue was
purified by flash silica column chromatography (2:1 PE:EA) to yield 8e (118 mg, 0.26 mmol,
1
8
1
1%) as a brown oil (R 0.32, 1:1 PE:EA). H NMR (400 MHz, CDCl ) δ 8.04 (d, J = 2.5 Hz,
f
3
H, ArH indole), 7.87 – 7.80 (m, 2H, ArH MeOPh), 7.53 – 7.47 (m, 3H, ArH Bn & ArH
indole), 7.43 – 7.36 (m, 2H, ArH Bn), 7.36 – 7.30 (m, 1H, ArH Bn), 7.28 (dd, J = 8.9, 0.5 Hz,
2
5

ACCEPTED MANUSCRIPT
1
H, ArH indole), 7.06 (dd, J = 8.9, 2.5 Hz, 1H, ArH indole), 7.03 – 6.97 (m, 2H, ArH
MeOPh), 5.17 (s, 2H, CH Bn), 4.01 (d, J = 7.3 Hz, 2H, N-CH ), 3.99 – 3.93 (m, 2H, O-CH
2
2
2
THP), 3.90 (s, 3H, O-CH ), 3.31 (td, J = 11.8, 2.2 Hz, 2H, O-CH THP), 2.21 – 2.05 (m, 1H,
3
2
13
CH THP), 1.56 – 1.28 (m, 4H, O-CH CH THP). C NMR (101 MHz, CDCl ) δ 189.89,
2
2
3
1
62.34, 155.71, 137.47, 136.76, 133.54, 132.11, 130.96, 128.65, 128.39, 127.97, 127.82,
1
15.43, 114.88, 113.71, 110.83, 105.31, 70.69, 67.42, 55.58, 53.33, 36.06, 30.86. HRMS-ESI
+
calculated for C H NO [M+H] 456.2169, found m/z 456.2128.
2
9
30
4
7
-(Benzyloxy)-3-(4-methoxybenzoyl)-1-[(oxan-4-yl)methyl]-1H-indole (8f)
A solution of 7d (104 mg, 0.32 mmol), 4-methoxybenzonitrile (129 mg, 0.97 mmol) and
(Phen)Pd(OAc) ] (13 mg, 32.0 µmol) in H O (0.12 mL), glacial AcOH (0.18 mL) and 1,4-
[
2
2
dioxane (0.6 mL) was reacted as described in the procedure for 8a. The crude residue was
purified by flash silica column chromatography (2:1 hexane:EA) to yield 8f (85 mg, 0.19
1
mmol, 58%) as a pale pinkish-white solid (R 0.44, 1:1 hexane:EA). H NMR (400 MHz,
f
CDCl ) δ 8.01 (dd, J = 8.1, 0.9 Hz, 1H, ArH indole), 7.85 – 7.78 (m, 2H, ArH MeOPh), 7.51
3
–
8
7.46 (m, 2H, ArH Bn), 7.45 – 7.37 (m, 3H, ArH Bn), 7.37 (s, 1H, ArH indole), 7.23 (t, J =
.0 Hz, 1H, ArH indole), 7.02 – 6.95 (m, 2H, ArH MeOPh), 6.87 (dd, J = 7.9, 0.9 Hz, 1H,
ArH indole), 5.15 (s, 2H, CH Bn), 4.10 (d, J = 7.2 Hz, 2H, N-CH ), 3.89 (s, 3H, O-CH ),
2
2
3
3
.86 – 3.77 (m, 2H, O-CH THP), 3.12 (td, J = 11.6, 2.4 Hz, 2H, O-CH THP), 2.14 – 1.93
2 2
1
3
(
m, 1H, CH THP), 1.22 – 0.99 (m, 4H, O-CH CH THP). C NMR (101 MHz, CDCl ) δ
2 2 3
1
89.81, 162.33, 146.60, 137.85, 136.49, 133.54, 131.04, 130.25, 128.82, 128.74, 128.66,
1
26.27, 123.20, 115.56, 115.32, 113.64, 105.14, 70.84, 67.54, 56.29, 55.55, 37.10, 30.22.
+
HRMS-ESI calculated for C H NO [M+H] 456.2169, found m/z 456.2129.
2
9
30
4
5
-(Benzyloxy)-3-(naphthalene-1-carbonyl)-1-[(oxan-4-yl)methyl]-1H-indole (8g)
A solution of 7c (103 mg, 0.32 mmol), 1-naphthonitrile (147 mg, 0.96 mmol) and
(Phen)Pd(OAc) ] (14 mg, 32.0 µmol) in H O (0.12 mL), glacial AcOH (0.18 mL) and 1,4-
[
2
2
dioxane (0.6 mL) was reacted as described in the procedure for 8a. The crude residue was
purified by flash silica column chromatography (2:1 PE:EA) to yield 8g (87 mg, 0.18 mmol,
1
5
9%) as a brown oil (R 0.5, 1:1 PE:EA). H NMR (400 MHz, CDCl ) δ 8.19 (dd, J = 8.5,
f
3
1
.2 Hz, 1H, ArH naphthalene), 8.13 (d, J = 2.5 Hz, 1H, ArH indole), 7.98 (dd, J = 8.3, 1.1
Hz, 1H, ArH naphthalene), 7.95 – 7.89 (m, 1H, ArH naphthalene), 7.66 (dd, J = 7.0, 1.3 Hz,
H, ArH naphthalene), 7.60 – 7.44 (m, 5H, ArH Bn & naphthalene), 7.44 – 7.37 (m, 2H, ArH
Bn), 7.37 – 7.31 (m, 1H, ArH indole), 7.28 (d, J = 8.9 Hz, 1H, ArH Bn), 7.27 (s, 1H, ArH
1
2
6

ACCEPTED MANUSCRIPT
indole), 7.08 (dd, J = 8.9, 2.5 Hz, 1H, ArH indole), 5.18 (s, 2H, CH Bn), 4.00 – 3.85 (m, 4H,
2
N-CH & O-CH THP), 3.28 (td, J = 11.8, 2.1 Hz, 2H, O-CH THP), 2.13 – 1.96 (m, 1H, CH
2
2
2
13
THP), 1.48 – 1.24 (m, 4H, O-CH CH THP). C NMR (101 MHz, CDCl ) δ 192.14, 156.02,
2
2
3
1
1
3
4
5
39.12, 138.36, 137.40, 133.88, 132.31, 130.90, 130.16, 128.67, 128.33, 128.01, 127.88,
26.86, 126.45, 126.11, 125.97, 124.71,117.40, 114.99, 111.03, 105.44, 70.70, 67.36, 53.28,
+
5.93, 30.75. HRMS-ESI calculated for C H NNaO [M+Na] 498.2040, found m/z
3
2
29
3
98.1996.
-(Benzyloxy)-3-(4-methoxybenzoyl)-1-[(oxolan-3-yl)methyl]-1H-indole (8h)
A solution of 7e (87 mg, 0.28 mmol), 4-methoxybenzonitrile (130 mg, 0.98 mmol) and
(Phen)Pd(OAc) ] (13 mg, 28.0 µmol) in H O (0.12 mL), glacial AcOH (0.18 mL) and 1,4-
[
2
2
dioxane (0.6 mL) was reacted as described in the procedure for 8a. The crude residue was
purified by flash silica column chromatography (2:1 PE:EA) to yield 8h (84 mg, 0.19 mmol,
1
6
2
7%) as a light brown oil (R 0.12, 2:1 PE:EA). H NMR (400 MHz, CDCl ) δ 8.05 (d, J =
f
3
.5 Hz, 1H, ArH indole), 7.83 (dt, J = 8.7, 2.8, 2.6 Hz, 2H, ArH MeOPh), 7.54 (s, 1H, ArH
indole), 7.50 (d, J = 6.9 Hz, 2H, ArH Bn), 7.40 (t, J = 7.2 Hz, 2H, ArH Bn), 7.32 (dd, J = 9.4,
8
8
3
.1 Hz, 2H, ArH Bn, ArH indole), 7.07 (dd, J = 8.9, 2.5 Hz, 1H, ArH indole), 7.00 (dt, J =
.7, 2.8, 2.6 Hz, 2H, ArH MeOPh), 5.17 (s, 2H, CH Bn), 4.11 (d, J = 7.8 Hz, 2H, N-CH ),
2
2
.97 (td, J = 8.3, 5.4 Hz, 1H, O-CH THF), 3.90 (s, 3H, O-CH ), 3.77 (t, J = 8.3, 6.9 Hz, 1H,
2
3
O-CH THF), 3.71 (dd, J = 9.1, 6.4 Hz, 1H, O-CH THF), 3.61 (dd, J = 9.1, 4.4 Hz, 1H, O-
2
2
CH THF), 2.92 – 2.79 (m, 1H, CH THF), 2.12 – 1.99 (m, 1H, CH THF), 1.73 – 1.60 (m,
2
2
1
3
1
1
1
H, CH THF). C NMR (101 MHz, CDCl ) δ 189.91, 162.36, 155.77, 137.45, 136.23,
2 3
33.48, 132.00, 130.96, 128.65, 128.41, 127.97, 127.82, 115.71, 114.99, 113.73, 110.68,
05.34, 70.85, 70.69, 67.62, 55.58, 49.90, 39.74, 29.86. HRMS-ESI calculated for
+
C H NO [M+H] 442.2013, found m/z 442.1985.
28
28
4
3
-(4-Methoxybenzoyl)-1-[2-(morpholin-4-yl)ethyl]-1H-indol-5-ol (9a)
A flask containing a stirred solution of 8a (497 mg, 1.1 mmol), EtOH (12 mL), MeOH (12
mL) and AcOH (1 mL) was evacuated and purged with N , followed by addition of Pd/C
2
(10% by weight loading (dry basis), matrix carbon powder, wet support) (50 mg). The
solution was placed under an atmosphere of hydrogen (balloon) (evacuated and purged with
H three times) and stirred for 20 h at rt. The solution was then filtered through celite, the
2
celite washed with 1:1 EtOH:MeOH and then MeOH, and the filtrate evaporated under
reduced pressure. The crude residue was purified by precipitation in EA to yield the desired
2
7

ACCEPTED MANUSCRIPT
1
product 9a (54 mg, 0.14 mmol, 13%) as a pale fawn solid (R 0.29, 95:5 DCM:MeOH). H
f
NMR (400 MHz, DMSO-d ) δ 9.08 (s, 1H, OH), 7.93 (s, 1H, ArH indole), 7.77 (d, J = 8.3
6
Hz, 2H, ArH MeOPh), 7.66 (s, 1H, ArH indole), 7.41 (d, J = 8.8 Hz, 1H, ArH indole), 7.06
(d, J = 8.3 Hz, 2H, ArH MeOPh), 6.76 (d, J = 8.7 Hz, 1H, ArH indole), 4.30 (t, J = 6.2 Hz,
2
H, N1-CH ), 3.85 (s, 3H, O-CH ), 3.54 (t, J = 4.6 Hz, 4H, O-CH morpholino), 2.67 (t, J =
2 3 2
13
6
.3 Hz, 2H, N1-CH CH ), 2.47 – 2.35 (br m, 4H, N-CH morpholino). C NMR (101 MHz,
2
2
2
DMSO-d ) δ 188.12, 161.52, 153.35, 138.50, 133.22, 130.81, 130.44, 127.90, 113.54,
6
1
13.22, 112.71, 111.12, 106.06, 66.26, 57.30, 55.40, 53.17, 43.09. HRMS-ESI calculated for
+
C H N O [M+H] 381.1809, found m/z 381.1779. Analytical RP-HPLC R = 12.89 min.
22
25
2
4
t
3
-(4-Methoxybenzoyl)-1-[2-(mopholin-4-yl)ethyl]-1H-indol-6-ol (9b)
A mixture of 8b (75 mg, 0.16 mmol), EtOH (2.6 mL), EA (1.3 mL) and Pd/C (7.5 mg) was
reacted as described in the procedure for 9a. The reaction mixture was filtered through celite,
the celite washed with 2:1 EtOH:EA and then MeOH, and the filtrate evaporated under
reduced pressure. The crude residue was purified by precipitation in EA and semi-preparative
RP-HPLC to yield the desired product 9b (4.8 mg, 12.6 µmol, 8%) as a brown solid (R 0.21,
f
1
9
5:5 DCM:MeOH). H NMR (400 MHz, DMSO-d ) δ 9.38 (s, 1H, ArOH), 8.00 (d, J = 8.6
6
Hz, 1H, ArH indole), 7.84 (s, 1H, ArH indole), 7.81 – 7.75 (m, 2H, ArH MeOPh), 7.09 – 7.03
m, 2H, ArH MeOPh), 6.88 (d, J = 2.1 Hz, 1H, ArH indole), 6.76 (dd, J = 8.5, 2.1 Hz, 1H,
ArH indole), 4.25 (t, J = 6.3 Hz, 2H, N1-CH ), 3.85 (s, 3H, O-CH ), 3.55 (t, J = 4.6 Hz, 4H,
(
2
3
O-CH morpholino), 2.66 (t, J = 6.3 Hz, 2H, N1-CH CH ), 2.44 (t, J = 4.6 Hz, 4H, N-CH
2
2
2
2
1
3
morpholino). C NMR (126 MHz, DMSO-d ) δ 188.10, 161.59, 154.39, 137.76, 137.37,
6
1
4
33.07, 130.45, 122.27, 119.71, 114.01, 113.54, 112.04, 95.81, 66.25, 57.07, 55.41, 53.18,
+
2.91. HRMS-ESI calculated for C H N O [M+H] 381.1809, found m/z 381.1804.
2
2
25
2
4
Analytical RP-HPLC R = 13.78 min.
t
1
-[2-(Morpholin-4-yl)ethyl]-3-(naphthalene-1-carbonyl)-1H-indol-5-ol (9c)
A mixture of 8c (299 mg, 0.61 mmol), EtOH (11 mL), EA (5.5 mL) and Pd/C (30 mg) was
reacted as described in the procedure for 9a. The reaction mixture was filtered through celite,
the celite washed with 2:1 EtOH:EA and then MeOH, and the filtrate evaporated under
reduced pressure. The crude residue was purified by precipitation in EA to yield the desired
1
product 9c (95 mg, 0.24 mmol, 39%) as a white solid (R 0.35, 95:5 DCM:MeOH). H NMR
f
(
400 MHz, MeOD-d ) δ 8.03 (dd, J = 8.2, 4.6 Hz, 2H, ArH naphthalene), 7.99 – 7.94 (m, 1H,
4
ArH naphthalene), 7.82 (d, J = 2.4 Hz, 1H, ArH indole), 7.65 (dd, J = 7.0, 1.4 Hz, 1H, ArH
2
8