Diindolylmethane Derivatives: Potent Agonists of the Immunostimulatory Orphan G Protein-Coupled Receptor GPR84

Article abstract of DOI:10.1021/acs.jmedchem.6b01593

The G_i protein-coupled receptor GPR84, which is activated by (hydroxy)fatty acids, is highly expressed on immune cells. Recently, 3,3′-diindolylmethane was identified as a heterocyclic, nonlipid-like GPR84 agonist. We synthesized a broad range of diindolylmethane derivatives by condensation of indoles with formaldehyde in water under microwave irradiation. The products were evaluated at the human GPR84 in cAMP and β-arrestin assays. Structure-activity relationships (SARs) were steep. 3,3′-Diindolylmethanes bearing small lipophilic residues at the 5- and/or 7-position of the indole rings displayed the highest activity in cAMP assays, the most potent agonists being di(5-fluoro-1H-indole-3-yl)methane (38, PSB-15160, EC₅₀ 80.0 nM) and di(5,7-difluoro-1H-indole-3-yl)methane (57, PSB-16671, EC₅₀ 41.3 nM). In β-arrestin assays, SARs were different, indicating biased agonism. The new compounds were selective versus related fatty acid receptors and the arylhydrocarbon receptor. Selected compounds were further investigated and found to display an ago-allosteric mechanism of action and increased stability in comparison to the lead structure.

Full text of DOI:10.1021/acs.jmedchem.6b01593

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)
Article
Diindolylmethane Derivatives – Potent Agonists of the
Immunostimulatory Orphan G Protein-Coupled Receptor GPR84
Thanigaimalai Pillaiyar, Meryem Köse, Katharina Sylvester, Heike Weighardt, Dominik
Thimm, Gleice Borges, Irmgard Förster, Ivar von Kügelgen, and Christa E Müller
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b01593 • Publication Date (Web): 13 Apr 2017
Downloaded from http://pubs.acs.org on April 13, 2017
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of Medicinal Chemistry is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Diindolylmethane Derivatives – Potent Agonists of the
Immunostimulatory Orphan G ProteinꢀCoupled
Receptor GPR84
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1≠
1≠
1
2
1
Thanigaimalai Pillaiyar , Meryem Köse , Katharina Sylvester , Heike Weighardt , Dominik Thimm ,
1
2
1
3
Gleice Borges , Irmgard Förster , Ivar von Kügelgen and Christa E. Müller *
1
PharmaCenter Bonn, Pharmaceutical Institute, Pharmaceutical Chemistry I, University of Bonn, An der
Immenburg 4, Dꢀ53121 Bonn, Germany
2
Life and Medical Sciences (LIMES) Institute, Immunology and Environment, University of Bonn,
CarlꢀTrollꢀStraße 31, 53115 Bonn, Germany
3Department of Pharmacology and Toxicology, University of Bonn, 53105 Bonn, Germany
≠
Authors contributed equally
*
Corresponding author
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 70
KEYWORDS: Agonist, βꢀArrestin, Biased Signalling, Diindolylmethane, Free fatty acid receptor, G
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
proteinꢀcoupled receptors, GPR84, GPR40, GPR120, Orphan receptor, Synthesis, Structureꢀactivity
relationships
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ABSTRACT: The G proteinꢀcoupled receptor GPR84, which is activated by (hydroxy)fatty acids, is
i
highly expressed on immune cells. Recently 3,3’ꢀdiindolylmethane was identified as a heterocyclic,
nonꢀlipidꢀlike GPR84 agonist. We synthesized a broad range of diindolylmethane derivatives by
condensation of indoles with formaldehyde in water under microwave irradiation. The products were
evaluated at the human GPR84 in cAMP and βꢀarrestin assays. Structureꢀactivity relationships (SARs)
were steep. 3,3’ꢀDiindolylmethanes bearing small lipophilic residues at the 5ꢀ and/or 7ꢀposition of the
indole rings displayed the highest activity in cAMP assays, the most potent agonists being diꢀ(5ꢀfluoroꢀ
1Hꢀindoleꢀ3ꢀyl)methane (38, PSBꢀ15160, EC 80.0 nM) and diꢀ(5,7ꢀdifluoroꢀ1Hꢀindoleꢀ3ꢀyl)methane
50
(
57, PSBꢀ16671, EC 41.3 nM). In βꢀarrestin assays SARs were different indicating biased agonism.
50
The new compounds were selective versus related fatty acid receptors and the arylhydrocarbon receptor.
Selected compounds were further investigated and found to display an agoꢀallosteric mechanism of
action and increased stability in comparison to the lead structure.
ACS Paragon Plus Environment

Page 3 of 70
Journal of Medicinal Chemistry
INTRODUCTION
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
G proteinꢀcoupled receptors (GPCRs) constitute the largest known eukaryotic family of integral
membrane proteins with more than 800 human genes that encode for GPCRs. It has been estimated that
1
about one third of all modern drugs are targeting members of this receptor family. Approximately 50 %
are nonꢀolfactory receptors and about 100 of these are poorly characterized orphan receptors whose
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
natural ligand and physiological functions are still unknown or unconfirmed. These orphan receptors
are, nevertheless, expected to play important (patho)physiological roles. In the last decade an increasing
number of GPCRs has been deꢀorphanized and it was found that free fatty acids (FFAs) act as ligands
3
ꢀ8
for several GPCRs, i.e., GPR40, GPR41, GPR43, GPR120 and GPR84. FFAR1 (GPR40), a receptor
3
,6
for long chain free fatty acids, enhances insulin secretion in pancreatic βꢀcells, among other functions.
FFAR2 (GPR43) and FFAR3 (GPR41) are stimulated by short chain fatty acids and are involved in
4,7
energy regulation in adipocytes. FFAR4 (GPR120) is a receptor for long chain free fatty acids
5
mediating glucagon like peptideꢀ1 secretion. GPR84 is an exceptional fatty acidꢀactivated receptor
9
,10
which is predominantly present on immune cells.
In the periphery, GPR84 is mainly expressed in
bone marrow, spleen, lung, and peripheral blood leukocytes. In the CNS, GPR84 expression is restricted
1
0,11
to microglia. In addition, GPR84 is expressed in adipose tissues.
Table 1 provides a summary of the
3
,4,6,12ꢀ14
fatty acid receptors, their coupling and important functions.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 70
Table 1. G proteinꢀcoupled fatty acid receptors, their ligands, coupling and physiological functions
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Receptor
G protein
coupling
Important physiological function
Potent
agonists
(fatty acids)
FFAR1
Gq/11
Glucoseꢀdependent insulin release, inhibition of Long chain, C12ꢀ
1
5,16
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(GPR40)
osteoclastogenesis
taste preference for fatty acids
Immune function, hematopoiesis, mast cell
and regulation of
C18
1
7
FFAR2
Gq/11, Gi/o
Short chain,
C3=C4=C2
Short chain,
C3=C4>>C2
Long chain, C12ꢀ
C16
1
8
(
GPR43)
FFAR3
GPR41)
FFAR4
GPR120)
GPR84
activity, and adipogenesis.
Gi/o
Pancreatic peptide YY (PYY) secretion and
1
9
(
glucaconꢀlike peptide 1 (GLPꢀ1) secretion
5
Gq/11
GLPꢀ1 secretion
(
9
24
Gi/o
Immunostimulation, proinflammatory,
Medium chain,
C9ꢀC14
1
0
defense against pathogens
GPR84 was discovered by sequence tag data mining and cloned from a cDNA library prepared from
20,21
human peripheral blood neutrophils.
The human and mouse GPR84 genes encode for a protein of
2
0
3
96 amino acids in length with 85 % identity. GPR84 couples primarily to a pertussis toxin (PTX)ꢀ
sensitive G pathway in response to FFAs, including hydroxylated FFAs, of medium carbon chain
i/o
9
length (C9ꢀC14, MCFA). MCFAs, such as decanoic acid (capric acid), undecanoic acid, and
dodecanoic acid (lauric acid) were shown to act through GPR84 in RAW264.7 cells, a murine
macrophageꢀlike cell line, amplifying the stimulation of lipopolysaccharide (LPS)ꢀinduced interleukinꢀ
9
,22
1
2B (IL12B) production. The proꢀinflammatory cytokine ILꢀ12 plays a pivotal role in promoting cellꢀ
mediated immunity to eradicate pathogens by inducing and maintaining T helper 1 (Th1) responses and
inhibiting T helper 2 (Th2) response, which plays an important role in antibodyꢀmediated immunity.
The expression of GPR84 in the peripheral immune system and in microglia suggests a role in the
1
0
regulation of (neuro)inflammatory processes. Recent evidence showed that GPR84 deficiency
decreased the proliferation of leukemia cells and has therefore been proposed for the treatment of acute
ACS Paragon Plus Environment

Page 5 of 70
Journal of Medicinal Chemistry
2
3
myeloid leukemia. The expression of GPR84 in adipose tissues may indicate that the receptor plays a
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
2
4
role in obesity. GPR84 may be a promising drug target for a variety of inflammatory diseases
2
5
26
27
including Alzheimer’s disease, neuropathic pain, reflux esophagitis, and inflammatory bowel
28
disease.
MCFAs are only weakly potent activators of GPR84. Moreover, these putative endogenous ligands of
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
GPR84, including saturated fatty acids ranging in chain length from C9 to C14, also activate the free
6,29,30
fatty acid (FFA) receptors FFAR1 (GPR40) and FFAR4 (GPR120).
Suzuki et al. reported that 2ꢀ
hydroxyꢀ and 3ꢀhydroxyꢀfatty acids were more potent GPR84 agonists than nonꢀhydroxylated fatty
acids. In the same study, 6ꢀnꢀoctylaminouracil (6ꢀOAU, Figure 1), which displays a lipidꢀmimetic
structure, was reported as a synthetic agonist of human GPR84 with an EC50 value of 512 nM
3
5
13,29,31
determined in [ S]GTPɣS binding assays.
More recently, the structurally related 2ꢀ
32,33
(hexylthio)pyrimidineꢀ4,6ꢀdiol (Figure 1) and its derivatives
were identified as GPR84 agonists. In
addition, the natural product embelin (embelic acid, 3ꢀundecylꢀ2,5ꢀdihydroxyꢀ1,4ꢀbenzoquinone, Figure
3
4
1) has also been reported to be a GPR84 agonist with moderate potency. Embelin is a secondary
metabolite originally isolated from the plant Embelia ribes (Primulaceae), which was reported to have
anthelmintic, antifertility, antitumor, antiꢀapoptotic, antimicrobial, analgesic, and antiꢀinflammatory
3
5ꢀ37
activity.
Besides GPR84 activation, embelin has several other activities including inhibition of Xꢀ
38
38
linked inhibitor of apoptosis protein (XIAP, IC 4.1 µM), activation of caspase 9, and antiꢀoxidant
5
0
3
9
properties.
Another natural productꢀderived GPR84 agonist, 3,3’ꢀdiindolylmethane (1, Figure 1) was discovered,
which possibly acts via an allosteric binding site since its structure does not resemble fatty acids and is
4
0
41
not lipidꢀlike.
1 is the major metabolite of indoleꢀ3ꢀcarbinol which is a breakꢀdown product of
glucobrassicin, a glucosinolate that is present in high concentrations in some vegetables such as
4
2
broccoli. The compound was previously reported to activate the arylhydrocarbon receptor (AhR), a
4
3
ligandꢀactivated transcription factor associated with gastric carcinogenesis, as well as estrogen
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 70
4
4
45
receptors at concentrations of around 10 µM and higher. 1 Showed antiꢀobesity effects in mice. and
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
4
6
displayed antiꢀcancer activity in vitro and in vivo.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Selected GPR84 agonists
While most GPR84 agonists contain long alkyl chains and are thus highly lipophilic, 1 is a small
heterocyclic molecule consisting of indole rings, which are found in many perorally applied drug
4
7ꢀ49
molecules.
In the present study, we therefore selected 1 as a lead molecule and studied its structureꢀ
activity relationships (SARs) with the goal to improve its potency. The synthesized compounds were
evaluated at the human GPR84 in cAMP accumulation as well as βꢀarrestin recruitment assays.
ACS Paragon Plus Environment

Page 7 of 70
Journal of Medicinal Chemistry
RESULTS AND DISCUSSION
Compound design
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In order to explore the SARs of diindolylmethane derivatives at GPR84, the following modifications
were targeted (Figure 2): (A) substitution of the indole rings in various positions including symmetrical
and unsymmetrical substitution patterns, (B) monoꢀsubstitution of the 3,3’ꢀmethylene linker (C10) by
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
alkyl and aryl residues, (C) replacement of the 3,3’ꢀmethylene linker (CH ) by a carbonyl group, (D)
2
oxidation of the diindolylmethane ring system and (E) subsequent partial reduction yielding compounds
E.
Figure 2. Modifications of 3,3’ꢀdiindolylmethane (1)
Chemistry
It is well known that indoles react with aliphatic or aromatic aldehydes and ketones to produce
azafulvenium salts; these can undergo further addition with a second indole molecule to afford
diindolylmethanes. A number of synthetic methods for the preparation of 3,3’ꢀdiindolylmethanes have
5
0ꢀ52
53,54
been reported in the literature using protic acid (e.g. HCl),
Lewis acids (e.g. AlCl , BF , I ),
or
3
3
2
5
5ꢀ58
59
different lanthanide triflates or chlorides.
Recently, benzoic acid in water, sodium dodecylsulfate
60
(
SDS) as a surfactant in water, oxalic acid and NꢀacetylꢀN,N,Nꢀtrimethylammonium bromide (CTAB)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 70
6
1
57
62,63
in water, metal triflate in ionic liquid, in ionic liquids with or without Fe(III) salts
were reported
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
to provide efficient conditions for the transformation of indoles to diindolylmethanes. However, there
are still a number of drawbacks in the available catalytic systems including the requirement of
55,64
65
55,59,64
66
large
or stoichiometric amounts of catalysts, long reaction times
and low yields. Organic
reactions in water as a reaction medium offer advantages, including low cost, safe handling and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
environmentꢀcompatibility. There are few reports on the synthesis of diindolylmethanes in water via the
59,60
condensation of indole with aliphatic or aromatic aldehydes.
This led us to attempt the reaction of
indole with various aldehydes in aqueous media under microwave irradiation. In our initial attempt, the
o
reaction of indole (2, 10 mmol) with formaldehyde (37 %, 5 mmol) in water at 100 C for 5 minutes was
selected as a model reaction (entry 1, Table 2). The desired product 1 was isolated in reasonable
quantity (43 %) without a catalyst. In order to improve the yield, reaction time, temperature and catalyst
were optimized (Table 2). As shown in entries 2–4, the yield was gradually improved by increasing the
o
reaction time, and the best result was obtained for reactions at 100 C for 20 min (entry 4, 98 % yield).
The reaction temperature is one of the most important criteria as the reduction of the temperature from
o
o
1
00 C to 75 or 50 C in a 30 min reaction resulted in low yield (entry 5 and 6). The reaction was also
o
studied in the presence of acids as catalysts, concd. H SO or CH COOH, at 100 C for 30 min;
2
4
3
however, these conditions resulted in poor yields (20 % for entry 7 and 30 % for entry 8, Table 2).
ACS Paragon Plus Environment

Page 9 of 70
Journal of Medicinal Chemistry
Table 2. Microwaveꢀassisted synthesis of 3,3’ꢀdiindolylmethane in water under different conditions
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Temp
Time
(min)
Power
(W)
Yield
(%)
Entry
Catalyst
o
(
C)
1
2
3
4
5
6
7
8
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
100
100
100
100
75
5
105
105
105
105
75
43
84
91
98
42
28
20
30
10
15
20
30
50
30
75
concd. H SO 100
20ꢀ30
20ꢀ30
105
105
2
4
CH CO H
100
3
2
The optimized conditions were subsequently used for the synthesis of a series of diindolylmethane
derivatives (Scheme 1). The desired products 32ꢀ60 were obtained from the starting indoles 2ꢀ31 in very
good to excellent yields (84ꢀ98 % isolated). Detailed reaction conditions and purities of the products
(32ꢀ60) are summarized in Table 3. It should be noted that the microwaveꢀassisted reactions were
completed within only 20ꢀ30 min for most of the derivatives.
a
Scheme 1. Synthesis of 3,3’ꢀdiindolylmethane derivatives 32ꢀ60
a
o
b
1
2
3
Reagents and conditions: (a) H O, 100 C, microwave, 20–180 min, yield 45–98 %. For R , R and R
2
see Table 3.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 70
Table 3. Reaction times, yields, melting points and purities of 3,3’ꢀdiindolylmethane derivatives
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
Time
min)
Yield
(%)
Compd.
R
R
R
(
6
0
1
3
3
3
3
3
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
20
25
25
25
25
25
25
30
20
25
29
35
25
98
93
87
82
86
91
89
90
92
87
92
90
84
6
7
2
3
4
5
6
4ꢀCH3
4ꢀOCH3
4ꢀF
68
4ꢀCl
6
6
6
6
6
7
7
7
7
7
5ꢀCH3
5ꢀOCH3
5ꢀF
37
38
3
4
4
4
4
9
0
1
2
3
5ꢀCl
5ꢀBr
69
5ꢀCN
5ꢀNO2
5ꢀCOOMe
6
9
ACS Paragon Plus Environment

Page 11 of 70
Journal of Medicinal Chemistry
1
4
4
4
4
4
5
6
7
5ꢀCHO
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
CH3
25
25
35
25
25
30
25
25
25
30
30
20
20
30
25
25
25
85
79
75
88
82
87
90
93
85
83
76
87
83
80
89
82
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5ꢀCOOH
5ꢀOCH
6ꢀCH3
Ph
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
6
6
6
6
8
8
8
8
48
49
6ꢀOCH3
6ꢀF
50
51
52
53
54
6ꢀCl
7ꢀOCH
7ꢀF
3
4ꢀCl, 6ꢀCl
5ꢀF, 6ꢀCl
4ꢀF, 5ꢀF
5ꢀF, 6ꢀF
5ꢀF, 7ꢀF
H
55
56
5
5
5
6
7
8
9
0
70
68
CH3
CH3
see structure above
H
6
8
82
5
9,60,71
Surfactants can play an important role as emulsifiers in waterꢀmediated organic reactions.
In order
to avoid solubility problems, we used sodium dodecyl sulfate (SDS) in our next optimization step
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 70
reacting indoles with various aliphatic and aromatic aldehydes under microwave irradiation in water.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Indole 2 was treated with the appropriate aldehyde in the presence of 10 % SDS in water (5 mL), and
o
the resulting white turbid mixture was irradiated in a microwave oven at 100 C for 30ꢀ40 min (Scheme
2
A). As shown in Table 4, both, aliphatic (61ꢀ66) and aromatic aldehydes (67ꢀ74), afforded good yields
6
0,71ꢀ76
of the corresponding products (75ꢀ88).
Diindolylmethanone 92 was synthesized using a published
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
7
7
procedure as indicated in Scheme 2B. Indoleꢀ3ꢀcarboxylic acid (89) was converted to the
o
corresponding acid chloride 90 by reaction with thionyl chloride in dichloromethane at 45 C for 1 h.
The resulting acid chloride was subsequently treated with indole (2) in the presence of zirconium(IV)
chloride in dichloroethane at room temperature for 4 h to produce the desired product 91.
ACS Paragon Plus Environment

Page 13 of 70
Journal of Medicinal Chemistry
a
b
Scheme 2. Synthesis of diindolylmethane derivatives 75ꢀ88 and 91
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
A)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
B)
a
o
Reagents and conditions: (a) sodium dodecyl sulfate (SDS), H O, 95–100 C, microwave (MW), 30ꢀ40
2
b
o
min, yield 72–95 %. For R see Table 5; (a) SOCl , dichloroethane (DCE), 40 C, 1 h; b) ZrCl , DCE, 0
2
4
o
C to rt, 4 h, 69 %.
Table 4. Reaction times, yields, melting points and purities of products 75–88 and 91
1
Compd.
R
R
Reaction time
Yield (%)
(lit. yield)
(
min)
7
2
7
5
H
30
89
83
76
4ꢀF
30
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
7
7
H
30
30
30
35
40
40
40
40
40
40
40
40
80
7
3
4
5
2
2
84
73)
7
8
H
H
H
H
H
H
H
H
H
H
H
(
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
89
7
7
5
5
79
80
(62)
72
21)
(
(
(
85
65)
8
1
2
95
80)
8
8
3
93
52
87
8
4
(55)
80
8
5
8
6
91
94
7
6
2
87
88
(
89)
5
90
60)
(
ACS Paragon Plus Environment

Page 15 of 70
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
7
91
See structure above
40
69
Unsymmetrical 3,3’ꢀdiindolylmethanes were synthesized through an intermolecular Pummerer reaction
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
7
8
using indole as a nucleophile. As shown in Scheme 3, indole was treated with dimethyl sulfoxide
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
o
o
(
DMSO) and trifluoroacetic acid anhydride in 1,4ꢀdioxane at 0 C, and the mixture was heated at 80 C
for 30 min to obtain intermediate 92, which was subsequently converted to the unsymmetrical 3,3’ꢀ
diindolylmethanes 93–97 by heating it with substituted indole (8, 9, 14 or 29) and copper acetate for 2 h.
Product 97 was obtained in an excellent yield of 99 % after hydrolysis of the dimethyl ester 96, which
was achieved by treatment with 2ꢀN sodium hydroxide in ethanol under reflux for 2 h. Yields, melting
points and purities of synthesized derivatives are collected in Table 5.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 70
a
Scheme 3. Synthesis of unsymmetrical 3,3’ꢀdiindolylmethanes 93ꢀ98
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
o
o
Reagents and conditions: (a) DMSO, trifluoroacetic anhydride (TFAA), dioxane, 0 C then 80 C, 30
o
o
min; (b) Cu(OAc) , dioxane, 80 C, 1 h, yield 35–47%; c) 2 N NaOH, ethanol, 100 C, 1 h, yield 99%.
2
1
2
For R , R see Table 5.
Table 5. Yields, melting points and purities of synthesized unsymmetrically substituted 3,3’ꢀ
diindolylmethanes (93–97)
Yield
1
2
Compd.
R
R
(%)
93
94
95
96
97
H
CH
3
40
47
41
35
99
OCH
F
H
3
H
CO
CO
CH
H
H
2
3
H
2
ACS Paragon Plus Environment

Page 17 of 70
Journal of Medicinal Chemistry
2
ꢀOxoindole derivatives were synthesized as illustrated in Scheme 4. 2ꢀOxoindole derivative 98 or 99
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
was reacted with the appropriate aldehyde (100ꢀ103) in absolute ethanol in the presence of piperidine at
o
79ꢀ83
65 C for 2 h to produce α,βꢀunsaturated indoleꢀ2ꢀone derivatives 104ꢀ109.
These products were
o
reduced in the presence of sodium borohydride at 65 C for 2 h to yield the corresponding saturated
7
9
indoleꢀ2ꢀone derivatives 110ꢀ113.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
13
13
The structures of the synthesized compounds were confirmed by H, C, and/or C attached protein
1
3
13
test ( C ) NMR spectroscopy, a common experiment to assign CꢀH multiplicities in C NMR spectra.
apt
In addition, HPLC analysis coupled to electrospray ionization mass spectrometry (LC/ESIꢀMS) was
performed, which was also used to determine the purity of the compounds.
a
Scheme 4: Synthesis of unsaturated and saturated 2ꢀoxoindole derivatives 104ꢀ109 and 110–113
a
o
o
Reagents and condition: (a) piperidine, ethanol, 65 C, 2 h, yield 69–80%; (b) NaBH , ethanol, 65 C, 2
4
1
2
h, yield 73–95%. For R , R see Table 6.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 70
Table 6. Yields, melting points and purities of synthesized 2ꢀoxoindole derivatives 104ꢀ113
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
Compd.
R
R
Yield (%)
76
8
0
1
1
04
05
H
H
H
H
F
F
69
73
85
80
7
3
106
OCH3
F
81
1
07
108
F
OCH
3
83
1
09
See structure above
73
8
2
110
H
F
F
OCH3
F
89
96
91
111
112
113
OCH
3
See structure above
95
ACS Paragon Plus Environment

Page 19 of 70
Journal of Medicinal Chemistry
Pharmacological evaluation
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
All compounds were initially investigated in cAMP accumulation assays at a concentration of 10 µM
for their potency to inhibit forskolin (10 µM) induced cAMP accumulation in Chinese hamster ovary
(CHO) cells stably expressing the G proteinꢀcoupled human GPR84. For compounds leading to more
i
than 50 % inhibition of cAMP accumulation, full concentrationꢀresponse curves were determined and
EC values were calculated (see Table 7 and 8). In order to determine the compounds’ efficacy, their
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
50
maximal effects were compared to the maximal signal induced by decanoic acid (100 µM; EC50
7.42 µM). Compounds that did not show agonistic activity were tested for their potency to antagonize
receptor activation by decanoic acid (20 µM) at a concentration of 10 µM. To exclude GPR84ꢀ
independent effects, the most potent compounds were additionally investigated in the parent nonꢀ
GPR84ꢀtransfected CHO cell line. Selected compounds were further investigated in GPR84ꢀdependent
βꢀarrestin recruitment assays using the βꢀgalactosidase fragment complementation technology
8
4,85
(Pathhunter, DiscoverX)
(see Table 7 and 8). We additionally confirmed the allosteric interaction of
diindolylmethane and its optimized derivative 57 with GPR84 in cAMP accumulation assays applying
different combinations of the orthosteric agonist decanoic acid and diindolylmethanes. To investigate
GPR84 selectivity, potent agonists were evaluated at those human free fatty acid receptor subtypes that
show overlapping agonist specificity with GPR84 and are also activated by dodecanoic acid, namely
FFAR1 and FFAR4 (see Table S2 and S3). The G proteinꢀcoupled human FFAR1 was retrovirally
q
expressed in 1321N1 astrocytoma cells and calcium mobilization assays were performed. The human
FFAR4 was expressed in a CHO cell line suitable to perform βꢀarrestin recruitment assays using the βꢀ
galactosidase complementation assay technology. All compounds were tested for agonistic as well as
antagonistic activity at FFAR1 and FFAR4 in a final concentration of 10 µM. Selected
diindolylmethane derivatives were further evaluated for activation of the arylhydrocarbon receptor, the
most prominent target of lead compound 1.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 70
Table 7. Potency of symmetrical 3,3’ꢀdiindolylmethane derivatives 32–60 as agonists at the human
GPR84 determined in cAMP accumulation and βꢀarrestin recruitment assays
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
human GPR84
1
2
3
Compd.
R
R
R
a
cAMP assay
βꢀarrestin assay
EC ± SEM (µM)
EC ± SEM (µM)
5
0
50
(
or percent receptor
(or percent receptor
activation at 10 µM)
[efficacy]
activation at 10 µM)
b
c
[efficacy]
*
#
Decanoic acid
Embelin
7.42 ± 0.40 [100 %]
6.08 ± 0.36 [92 %]
§
$
0.795 ± 0.168 [86 %]
0.400 ± 0.142 [100 %]
1.64 ± 0.81 [60 %]
> 10 (24 ± 8 %)
n.d.*
1
3
3
H
H
H
H
H
H
H
H
H
0.252 ± 0.088 [125 %]
0.987 ± 0.235 [112 %]
≥ 10 (47 ± 17 %)
2
3
4ꢀCH
3
4ꢀOCH
3
3
1
3
4 (PSBꢀ 4ꢀF
0.328 ± 0.005 [110 %]
1.23 ± 0.03 [108 %]
6357)
5
6
4ꢀCl
H
H
H
H
H
H
1.14 ± 0.06 [111 %]
≥ 10 (45 ± 9 %)
> 10 (25 ± 14 %)
≥ 10 (49 ± 6 %)
3
5ꢀCH
3
3
7 (PSBꢀ 5ꢀOCH3
0.369 ± 0.025 [119 %]
1.20 ± 0.80 [103 %]
1
6105)
38 (PSBꢀ 5ꢀF
H
H
0.0800 ± 0.0212 [117 %] 4.33 ± 2.12 [125 %]
1
5160)
3
9
5ꢀCl
H
H
H
H
H
H
H
H
H
H
H
H
4.96 ± 3.08 [86 %]
3.44 ± 0.29 [106 %]
≥ 10 (49 %)
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
4
4
4
4
4
0
1
2
3
4
5ꢀBr
5ꢀCN
5ꢀNO
> 10 (31 %)
2
5ꢀCOOMe
5ꢀCHO
> 10 (11 %)
> 10 (ꢀ4 %)
ACS Paragon Plus Environment

Page 21 of 70
Journal of Medicinal Chemistry
4
4
4
4
5
6
7
8
5ꢀCOOH
H
H
H
H
H
H
H
H
H
H
≥ 10 (28 %)
>> 10 (ꢀ5 ± 6 %)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5ꢀOCH
Ph
> 10 (36 %)
n.d.
2
6ꢀCH
6ꢀOCH
> 10 (ꢀ22 %)
> 10 (31 %)
n.d.
3
n.d.
3
4
9 (PSBꢀ 6ꢀF
0.625 ± 0.070 [89 %]
7.53 ± 2.30 [48 %]
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
6358)
50
51
6ꢀCl
H
H
H
H
H
H
> 10 (4 %)
5.34 ± 1.84 [118 %]
n.d.
7ꢀOCH
> 10 (18 %)
3
52 (PSBꢀ 7ꢀF
0.113 ± 0.047 [129 %]
6.07 ± 3.37 [99 %]
1
6381)
5
3
4ꢀCl,6ꢀCl
5ꢀF,6ꢀCl
4ꢀF,5ꢀF
H
H
H
H
H
H
H
H
≥ 10 (51 %)
n.d.
5
4
5
10.8 ± 0.50 [83 %]
0.377 ± 0.030 [123 %]
0.187 ± 0.053 [121 %]
3.98 ± 0.86 [181 %]
>> 10 (ꢀ2 ± 9 %)
10.3 ± 5.6 [92 %]
5
5
1
5
6 (PSBꢀ 5ꢀF,6ꢀF
6586)
7 (PSBꢀ 5ꢀF,7ꢀF
H
H
0.0413 ± 0.0098 [131 %] 5.47 ± 0.13 [385 %]
16671)
5
8
9
H
H
H
CH
3
> 10 (ꢀ37 %)
≥ 10 (42 %)
> 10 (ꢀ12 %)
>> 10 (ꢀ3 %)
5
CH
3
H
> 10 (23 %)
60
see structure above
ꢀ
2.27 ± 0.42 [115 %]
*n.d., not determined
Inhibition of forskolin (10 µM) induced decrease in cAMP accumulation
Efficacy (Emax) relative to the max. effect of decanoic acid (100 µM) (= 100 %)
Efficacy (Emax) relative to the max. effect of embelin (10 µM) (= 100 %)
a
b
c
*
#
§
$
9
Literature value: EC = 4.5 ± 0.3 µM
5
0
84
Literature value: EC = 10 µM
50
85
Literature value: EC = 421 nM
5
0
84
Literature value: EC = 12.3 µM
5
0
Structureꢀactivity relationships
cAMP accumulation assays
The lead compound 1 induced an inhibition of forskolinꢀinduced cAMP accumulation in CHO cells
transfected with the human G ꢀcoupled GPR84 displaying an EC 252 nM. (see Fig. 4). It was 29ꢀfold
i
50
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 70
more potent than decanoic acid (p=0.0021,**) and showed a higher efficacy (125 %) compared to
decanoic acid (set at 100 %) that – in contrast to 1 ꢀ did not completely inhibit cAMP formation at the
highest concentration used (Fig. 4). We confirmed that 1 had no effect on forskolinꢀinduced cAMP
levels in nonꢀtransfected CHO cells (see Fig. 4). This clearly proves that the observed effect on cAMP
levels was due to GPR84 activation.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
When exploring the SARs of diindolylmethane derivatives we initially focused on substitution of the
indole rings (Table 7). SAR analysis revealed that many symmetrically substituted indole derivatives
were potent GPR84 agonists. However, substitution at several positions of the indole rings resulted in
loss of activity, and substitution was only permitted in few positions. These results indicated very steep
SARs, clearly not driven by an increase in lipophilicity.
The following substituents were introduced at the 4,4’ꢀpositions of the indole rings: 4,4’ꢀdimethyl (32),
4
,4’ꢀdimethoxy (33), 4,4’ꢀdifluoro (34) and 4,4’ꢀdichloro (35). A deleterious effect on activity was
observed with these substituents, except for compound 34 (EC50 328 nM) featuring 4,4’ꢀdifluoro
substitution, which exhibited comparable activity to the lead compound (1). Next, substituents were
introduced at the 5,5’ꢀpositions yielding compounds 36–46. Among them, the 5,5’ꢀdifluoro derivative
(
38: EC 80.0 nM) showed high activity being 3ꢀfold more potent than the lead compound 1. 5,5’ꢀ
50
Dimethoxy substitution (37: EC 369 nM) conferred moderate potency, whereas 5,5’ꢀdichloro (39:
50
EC50 4960 nM) and 5,5’ꢀdibromo derivatives (40: EC50 3440 nM) displayed only weak potency.
Substitution at the 6,6’ꢀpositions (compounds 47ꢀ50) did not improve agonistic potency, and the small
6
,6’ꢀdifluoro substituents were best tolerated (compound 49, EC 625 nM). Based on these results we
50
only prepared two 7,7’ꢀsubstituted diindolylmethanes with 7,7’ꢀdimethoxy (51) and 7,7’ꢀdifluoro (52)
derivatives. Interestingly, compound 52 (EC 113 nM) was found to be 6ꢀfold more potent than the
50
corresponding 6,6’ꢀdifluoroꢀsubstituted isomer 49 (EC 625 nM) (p=0.0057, **), and appeared to be
50
slightly more potent than lead compound 1.
ACS Paragon Plus Environment

Page 23 of 70
Journal of Medicinal Chemistry
Encouraged by these results our next effort was to introduce multiple halogen substituents with the goal
to identify the best combination (compounds 53ꢀ57). Chloro substitution at the 4ꢀ and 6ꢀ positions (53)
was not tolerated probably because of the size of the halogen atoms. When we introduced the smaller
fluorine atoms, the potency of the resulting compounds (54ꢀ57) was much higher. A combination of
fluorine atoms in the 5ꢀ and 7ꢀpositions (57) was found to be the best combination, and in fact
compound 57 (EC 41.3 nM) was the most potent GPR84 agonist of the present series.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
5
0
The replacement of the indole NH hydrogen atoms of lead structure 1 by methylation yielding the 1,1’ꢀ
dimethyl derivative 58 (EC50 >10 µM) or the 1ꢀmethyl derivative 59 (EC50 ≥10 µM, see Table 8)
abolished potency of the compound. This confirms that the indole NH functions are very important for
interaction with GPR84, presumably by the formation of hydrogen bonds. Substitution of the 2ꢀ
positions as in the 2,2’ꢀdimethyl derivative 59, or introduction of a nitrogen atom as in diꢀ(7ꢀ
azaindolyl)methane (60) was not tolerated. These results obtained from substitution and modification of
the indole rings clearly indicated that substituents other than fluoro were not well tolerated and in all
cases led to a decrease in agonistic potency at GPR84. The rank order of potency for substituents at
various positions of the indole rings is as follows: 5,5’ꢀdifluoro (38: EC 80.0 nM), 7,7’ꢀdifluoro (52:
50
EC 113 nM) > 4,4’ꢀdifluoro (34: EC 328 nM, p=0.0104, *) ≥ 5,5’ꢀdimethoxy (37: EC 369 nM) >
50
50
50
6
,6’ꢀdifluoro (49: EC 625 nM, p=0.0262 *). Only fluoroꢀsubstitution in the 7ꢀ and particularly in the
50
5ꢀposition improved the potency of the unsubstituted lead structure 1.
In the next set of experiment, we investigated the effects of monoꢀsubstitution at the 5ꢀposition of only
one of the indole rings (Table 8). The following rank order of potency was observed: 5ꢀfluoro (95: EC50
*
2
5
64 nM) > 5ꢀmethoxy (94: EC 1890 nM, p=0.0013, ) > 5ꢀcarboxymethyl ester (96: EC ≥10 µM) >
50
50
ꢀcarboxylic acid (97: EC >>10 µM). This means that also in the monoꢀsubstituted series only small
50
residues were tolerated in the 5ꢀposition, and a fluorine atom (compound 95) was best showing similar
potency as the unsubstituted lead structure 1. The potencies of the monoꢀsubstituted derivatives 94 (5ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 70
fluoro) and 95 (5ꢀmethoxy) were reduced when compared to the corresponding diꢀsubstituted
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
**
derivatives 37 (p=0.0014, ) and 38 (p=0.0776, ns). These results revealed that symmetrical substitution
of 1 is superior to monoꢀsubstitution yielding unsymmetrical diindolylmethane derivatives (compare
3
7/94 and 38/95).
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
We subsequently examined the effects of substitution at the 3,3'ꢀmethylene bridge (Cꢀ10) of 1 (Table 8).
A large variety of substituents including methyl (75ꢀ77), nꢀpropyl (78), nꢀbutyl (79), isopropyl (80), 4ꢀ
tolyl (81), 4ꢀanisyl (82), 4ꢀhydroxyphenyl (83), 4ꢀchlorophenyl (84), 4ꢀfluorophenyl (85), 4ꢀ(1,3ꢀ
benzodioxolyl) (86), 3ꢀindolyl (87) and 4ꢀnitrophenyl (88) were introduced. However, none of the
substituents was tolerated at that position, and all compounds were inactive. On the other hand, 3,3’ꢀ
diindolylmethanone (91: EC 553 nM), an oxidized derivative of 1, was found to be almost as potent as
50
the lead structure 1 (EC 252 nM).
5
0
ACS Paragon Plus Environment

Page 25 of 70
Journal of Medicinal Chemistry
Table 8. Potency of 3,3’ꢀdiindolylmethane derivatives 75ꢀ113 as agonists at the human GPR84
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
determined in cAMP accumulation and βꢀarrestin recruitment assays
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Human GPR84
a
cAMP assay
βꢀarrestin assay
1
2
Compound
R
R
EC ± SEM (µM)
EC ± SEM (µM)
5
0
50
(
or percent receptor
(or percent receptor
activation at 10 µM)
[efficacy]
activation at 10 µM)
b
c
[efficacy]
Structure A: Substitution of the methylene linker (C10)
75
–
> 10 (ꢀ5 %)
> 10 (5 %)
> 10 (3 %)
76
77
78
4ꢀF
–
> 10 (34 %)
> 10 (ꢀ24 %)
> 10 (ꢀ24 %)
> 10 (ꢀ4 %)
2.97 ± 1.03 [67 %]
6.07 ± 0.77 [199 %]
> 10 (ꢀ3 %)
–
79
–
8
0
–
–
> 10 (16 %)
> 10 (2 %)
7.29 ± 0.19 [167 %]
> 10 (10 %)
81
8
2
3
–
> 10 (ꢀ3 %)
n.d.
8
–
–
–
>
10 (7 %)
> 10 (ꢀ5 %)
n.d.
84
85
> 10 (ꢀ4 %)
> 10 (8 %)
n.d.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 70
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
8
6
–
> 10 (0 %)
> 10 (7 %)
8
7
8
–
> 10 (ꢀ10 %)
> 10 (6 %)
> 10 (10 %)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
8
–
n.d.
Structure B:
Diindolylmethanone and unsymmetrically substituted diindolylmethane derivatives
91
1
(PSBꢀ
–
0.553 ± 0.141 [104 %]
19.0 ± 8.1 [161 %]
> 10 (8 %)
6359)
–
9
3
≥ 10 (43 %)
–
–
–
–
94
1.89 ± 0.19 [97 %]
0.264 ± 0.075 [115 %]
ca. 10 (57 %)
6.47 ± 3.20 [110 %]
9
5
(PSBꢀ
1.08 ± 0.58 [72 %]
16244)
9
6
7
n.d.
n.d.
9
> 10 (5 %)
Structure C: Oxidized products of diindolylmethane derivatives (methene bridged)
104
H
> 10 (ꢀ17 %)
> 10 (ꢀ22 %)
ACS Paragon Plus Environment

Page 27 of 70
Journal of Medicinal Chemistry
F
1
2
1
1
1
05
H
> 10 (ꢀ1 %)
n.d.
n.d.
n.d.
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
N
H
06
07
H
> 10 (ꢀ10 %)
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
F
5ꢀF
> 10 (5 %)
N
H
108
5ꢀF
H
> 10 (7 %)
> 10 (ꢀ31 %)
> 10 (ꢀ7 %)
1
09
> 10 (ꢀ16 %)
Structure D: 2ꢀOxodihydroindolylꢀindolylmethane derivatives
110
111
112
H
> 10 (ꢀ21 %)
> 10 (16 %)
n.d.
n.d.
F
5ꢀF
N
H
5ꢀF
H
> 10 (ꢀ10 %)
> 10 (7 %)
n.d.
n.d.
113
a
Inhibition of forskolin (10 µM) induced decrease in cAMP accumulation
Efficacy (Emax) related to the max. effect of decanoic acid (100 µM) (= 100 %)
Efficacy (Emax) related to the max. effect of embelin (10 µM) (= 100 %)
b
c
To obtain deeper insight into the SARs of diindolylmethane derivatives, we altered the hydrogen
bonding characteristics of 1 by preparing 2ꢀoxyindoles. To this end we prepared a series of unsaturated
(
104–109) and saturated 2ꢀoxyindole derivatives (110–113). However none of these compounds was
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 70
found to interact with GPR84 (see Table 8) and this might be due to disruption of the planarity of the
molecules.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Figure 3 summarizes the structureꢀactivity relationships of the investigated compounds as agonists of
the human GPR84: (i) symmetrical diindolylmethane derivatives show the highest potency, (ii) small
substituents like methoxy and fluoro are tolerated at C4, C5, and C7 of the benzene rings, e.g., 4,4’ꢀ
difluoroꢀ (34), 5,5’ꢀdimethoxyꢀ (37) and 5ꢀfluoroꢀdiindolylmethane (95) show similar potency as the
unsubstituted lead structure, while 5,5’ꢀdifluoroꢀ (38), 7,7’ꢀdifluoroꢀ (52), and 5,5’,7,7’ꢀtetrafluoroꢀ
diindolylmethane (57) display enhanced agonistic potency; (iii) substitutions or modifications of the
methylene linker are not tolerated.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3: Structureꢀactivity relationships of diindolylmethane derivatives as agonists at human GPR84
Concentrationꢀresponse curves for the best two GPR84 agonists, 38 and 57, as well as for the known
agonists decanoic acid and diindolylmethane (1) are displayed in Figure 4. The diindolylmethanes were
additionally tested at the parent CHO cell line that does not express GPR84. While 1, 38 and 57 show a
concentrationꢀdependent inhibition of forskolinꢀinduced cAMP accumulation in CHO cells transfected
ACS Paragon Plus Environment

Page 29 of 70
Journal of Medicinal Chemistry
with the human GPR84, no inhibition is observed in nonꢀtransfected cells. This proves that the effects
observed for the diindolylmethane derivatives are due to GPR84 activation.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 4. Concentrationꢀresponse curves of selected GPR84 agonists determined in cAMP
accumulation assays. Nonꢀtransfected CHO cells or CHO cells stably expressing the human GPR84
(CHOꢀhGPR84) were preincubated with the respective test compounds at the indicated concentrations
for 5 min. Then 10 µM forskolin was added and the cells were incubated for additional 15 min. The
maximal forskolinꢀinduced cAMP accumulation in the absence of agonist stimulation was defined as
1
00 %. EC values for the investigated agonists were as follows: 38: 80.0 nM, 57: 41.3 nM; decanoic
50
acid: 7420 nM, diindolylmethane: 252 nM. Mean values ± SEM from 3 independent experiments
performed in duplicates are shown.
In cAMP assays diindolylmethane (1) showed somewhat higher efficacy (125 %) than the standard
***
agonist decanoic acid (set as 100 %, p=0.0006, ). Most potent diindolylmethane derivatives and
analogs displayed efficacies in the same range as 1, e.g. for the most active compounds 38, 52, and 57 it
ranged from 117ꢀ131 % of the maximal effect observed for decanoic acid. Only the 5,5’ꢀdichloroꢀ
ACS Paragon Plus Environment