Modeling, Synthesis, and Biological Evaluation of Potential Retinoid X Receptor (RXR)-Selective Agonists: Analogues of 4-[1-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydro-2-naphthyl)ethynyl]benzoic Acid (Bexarotene) and 6-(Ethyl(5,5,8,8-tetrahydronaphthalen-2-yl)amino)nicotinic Acid (NEt-TMN)

Article abstract of DOI:10.1021/acs.jmedchem.6b00812

Sulfonic acid analogues of 4-[1-(3,5,5,8,8-pentamethyl-5,6,7,8-tetrahydro-2-naphthyl)ethynyl]benzoic acid (bexarotene, 1) as well as seven novel and two reported analogues of 6-(ethyl(5,5,8,8-tetramethyl-5,6,7,8-tetrahydronaphthalen-2-yl)amino)nicotinic acid (NEt-TMN) were synthesized and assessed for selective retinoid X receptor (RXR) agonism. Compound 1 is FDA-approved for treatment of cutaneous T-cell lymphoma (CTCL); however, 1 can provoke side effects by impacting RXR-dependent receptor pathways. All of the analogues in this study were evaluated for their potential to bind RXR through modeling and then assayed in an RXR-RXR mammalian-2-hybrid (M2H) system and in RXR-responsive element (RXRE)-mediated transcriptional experiments. The EC₅₀ profiles for these unique analogues and their analogous effectiveness to inhibit proliferation in CTCL cells relative to 1 suggest that these compounds possess similar or even enhanced therapeutic potential. Several compounds also displayed more selective RXR activation with minimal cross-signaling of the retinoic acid receptor. These results suggest that modifications of potent RXR agonists such as NEt-TMN can lead to improved biological selectivity and potency compared with the known therapeutic.

Full text of DOI:10.1021/acs.jmedchem.6b00812

Subscriber access provided by CORNELL UNIVERSITY LIBRARY
Article
Modeling, Synthesis and Biological Evaluation of Potential Retinoid-
X-Receptor (RXR) Selective Agonists: Analogs of 4-[1-(3,5,5,8,8-
Pentamethyl-5,6,7,8-tetrahydro-2-naphthyl)ethynyl]benzoic Acid (Bexarotene)
and 6-(ethyl(5,5,8,8-tetrahydronaphthalen-2-yl)amino)nicotinic Acid (NEt-TMN)
Michael C Heck, Carl Edward Wagner, Pritika H. Shahani, Mairi MacNeill, Aleksandra Grozic,
Tamana Darwaiz, Micah Shimabuku, David G Deans, Nathan M Robinson, Samer H Salama,
Joseph W. Ziller, Ning Ma, Arjan van der Vaart, Pamela A. Marshall, and Peter W. Jurutka
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.6b00812 • Publication Date (Web): 03 Sep 2016
Downloaded from http://pubs.acs.org on September 6, 2016
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 66
Journal of Medicinal Chemistry
1
2
3
Modeling, Synthesis and Biological Evaluation of Potential Retinoid-X-Receptor (RXR)
Selective Agonists: Analogs of 4-[1-(3,5,5,8,8-Pentamethyl-5,6,7,8-tetrahydro-2-
naphthyl)ethynyl]benzoic Acid (Bexarotene) and 6-(ethyl(5,5,8,8-
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
tetrahydronaphthalen-2-yl)amino)nicotinic Acid (NEt-TMN)
1
1
Michael C. Heck , Carl E. Wagner , Pritika H. Shahani, Mairi MacNeill, Aleksandra Grozic, Tamana
Darwaiz, Micah Shimabuku, David G. Deans, Nathan M. Robinson, Samer H. Salama, Joseph W.
§
‡
‡
Ziller , Ning Ma , Arjan van der Vaart , Pamela A. Marshall, Peter W. Jurutka*
School of Mathematical and Natural Sciences, New College of Interdisciplinary Arts and
Sciences, Arizona State University, 4701 W. Thunderbird Road, Glendale, AZ 85306.
§
Department of Chemistry, University of California, Irvine, 576 Rowland Hall, Irvine, CA 92697
‡
Department of Chemistry, University of South Florida, 4202 E Fowler Ave, CHE 205, Tampa,
FL, 33620.
1
These authors contributed equally to this work.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 66
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
ABSTRACT Sulphonic acid analogs of 4ꢀ[1ꢀ(3,5,5,8,8ꢀpentamethylꢀ5ꢀ6ꢀ7ꢀ8ꢀtetrahydroꢀ2ꢀ
naphthyl)ethynyl]benzoic acid (1)—or bexarotene—as well as seven novel and two reported
analogs of 6ꢀ(ethyl(5,5,8,8ꢀtetramethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)nicotinic acid
(NEt-TMN) were synthesized and assessed for selective retinoidꢀXꢀreceptor (RXR) agonism.
Compound 1 is FDAꢀapproved for treatment of cutaneous Tꢀcell lymphoma (CTCL); however, 1
can provoke sideꢀeffects by impacting RXRꢀdependent receptor pathways. All analogs in this
study were evaluated for their potential to bind RXR through modeling and then assayed in an
RXRꢀRXR mammalianꢀ2ꢀhybrid (M2H) system and in RXREꢀmediated transcriptional
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
experiments. The EC profiles for these unique analogs, and their analogous effectiveness to
50
inhibit proliferation in CTCL cells compared to 1, suggest these compounds possess similar or
even enhanced therapeutic potential. Several compounds also displayed more selective RXR
activation with minimal crossꢀsignaling of the retinoicꢀacidꢀreceptor. These results suggest that
modifications of potent RXR agonists such as NEt-TMN can lead to improved biological
selectivity and potency compared to the known therapeutic.
ACS Paragon Plus Environment

Page 3 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
1
There are three known isoforms (α, β, γ) of the human retinoid X receptors (hRXRs),
with one or more being expressed in every human tissue type, that primarily function as versatile
transcription factors. The RXRs often act as heterodimeric partners with other receptors of the
nuclear receptor (NR) superfamily comprising transcriptional regulators that include the retinoic
acid receptor (RAR), the thyroid hormone receptor (TR), the peroxisome proliferatorꢀactivated
receptor (PPAR), the vitamin D receptor (VDR), and the liver X receptor (LXR), to name only a
few. All nuclear receptors modulate gene expression, generally promoting transcription, when
signaled by a specific receptor ligand. This ligand, often a natural endogenous molecule,
associates with the receptor by binding to a receptor ligandꢀbinding domain (LBD) precipitating
a conformational change that permits high affinity receptor binding to the receptor’s
corresponding hormone responsive element (HRE) on the DNA. While HREs have increasingly
been located either upꢀ or downstream a considerable distance from the genes they regulate,
many HREs have been found close to or within the regulated gene promoter region. The HREs
exhibit sequence specificity, typically consisting of two hexad repeat sequences made up of halfꢀ
sites punctuated by a variable number of nucleotide spacers between these inverted, direct or
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
2
everted repeats. The corresponding HREs for VDR, TR and RAR, contain halfꢀsites separated
3
with 3, 4, and 5 nucleotide spacers, respectively.
While VDR, TR and RAR were first believed to bind as homodimers to their
4
corresponding HREs , they actually partner with RXR to form heterodimers in order to bind to
5
their HREs. When one of several natural RXR agonist ligands such as 9ꢀcisꢀretinoic acid (9ꢀcisꢀ
RA) binds to RXR, an RXR homodimer selfꢀassembles that then binds to RXR responsive
elements (RXREs). When RXR partners with other NRs in heterodimers, it does so either with
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 66
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
bound or unbound ligand within the RXR LBD. With respect to the RXRꢀVDR heterodimer,
6
RXR has been reported to be unoccupied. Alternatively, evidence suggests that RXR possesses
7
a ligand to function as a RXR:LXR heterodimer. Considering this versatility, it is not surprising
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
that RXR has been classified as the primary NR regulator, since it is often a prerequisite for
heterodimer formation and proper transcriptional activation of other NRs corresponding response
8
elements.
A multitude of studies employing rexinoids, RXR and a variety of different nuclear
receptors, have resulted in two primary classifications for RXRꢀheterodimer complexes—namely
permissive and nonpermissive heterodimers. Only ligands selective for the RXR partner can
activate purely nonpermissive RXR heterodimers, whereas either RXR specific ligands or the
9
partnerꢀselective ligands can activate permissive RXR heterodimers.
The generally
nonpermissive RXR heterodimers include RXRꢀVDR, RXRꢀRAR and RXRꢀTR. In most (but
not in all) circumstances, RXR is "silent" in VDR and TR heterodimers. The RXRꢀRAR
heterodimer, however, shows increased activity with select RXR agonists in combination with an
RAR ligand. In fact, a few rexinoids have been reported to stimulate RXRꢀRAR, despite the
1
0
absence of a RARꢀspecific ligand. In this respect, RXRꢀRAR has been termed conditionally
nonpermissive. Finally, the RXRꢀPPARs, RXRꢀLXRs, and RXRꢀFXR heterodimers have been
observed to be fully permissive.
Developing RXRꢀspecific agonists for clinical therapeutic treatments is complicated not
only by potential pleiotropic effects arising from the activation of permissive RXR heterodimers,
but also from a limited RXR pool that may be titrated away from RXRꢀcontaining heterodimers
to instead form RXR homodimers in certain cell conditions and tissue types—thereby limiting
the ability of the nonpermissive RXR heterodimers to form and function. For instance, RXR
ACS Paragon Plus Environment

Page 5 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
11
12
agonists such as 9ꢀcis RA can arrest nonpermissive heterodimer activation in VDR or TR.
Moreover, NR agonists like T3 or 1,25ꢀdihydroxyvitamin D (1,25D) that promote RXR
3
heterodimer formation with TR and VDR, respectively, may also drive crosstalk inhibition of
other RXRꢀdependent receptors utilizing the limited RXR pool. In fact, this crosstalk inhibition
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
3
has been observed for T ꢀTRꢀRXRꢀdirected inhibition of VDR transactivation
as well as
3
1
4 , 12
1,25DꢀVDRꢀRXRꢀmediated inhibition of TR function
; however, the crossꢀreceptor
squelching mechanism likely extends further than simple RXR sequestration. Nevertheless, two
key issues in developing therapeutic rexinoids with decreased pleiotropic effects concern RXR
potency and heterodimer selectivity. Hence, an approach to modify the structure of the central
RXR heterodimer agonist may lead to specific NR modulators (SNuRMs) that exhibit unique
1
5
characteristics arising from varying NR control.
Selective RXR ligand (rexinoid) SNuRMs are attractive medicinal drug targets because
activating RXR selectively versus RAR bestows chemotherapeutic effects in an array of human
1
6
cancer types without incurring toxicities arising from RAR crossover.
Following lengthy
17
development and testing of 9ꢀcis RA modeled compounds, 4ꢀ[1ꢀ(3,5,5,8,8ꢀpentamethyltetralinꢀ
1
8
2ꢀyl)ethynyl]benzoic acid (1) was demonstrated to possesses high RXR agonist specificity, and
Ligand Pharmaceuticals Inc. eventually gained FDA approval for the drug, more commonly
called bexarotene, for cutaneous Tꢀcell lymphoma (CTCL) treatment. Many structurally similar
1
9
bexarotene analogs, like disilabexarotene (2) , have been reported over the last twenty years and
many of these rexinoids possess RXR specific agonism similar to bexarotene.
Figure 1
While bexarotene (1) was approved as a treatment for CTCL, it has been explored for
20
21
therapeutic effects in colon cancer , mammary carcinoma and a proofꢀofꢀconcept (POC)
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 66
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
22
clinical trial has shown it effective in nonꢀsmall cell lung cancer , since the upꢀregulation of
certain RXRꢀcontrolled genes suppresses cell proliferation while synergizing apoptosis in cancer
2
0
cells subjected to chemotherapuetics. Several rexinoids, following compound 1, were
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
investigated in nonꢀinsulinꢀdependent diabetes mellitus (NIDDM) mouse models, because the
2
3
RXR:PPAR heterodimer impacts metabolism. Despite bexarotene’s RXRꢀspecificity versus
2
4
RAR, its known sideꢀeffects include hypothyroidism , hyperlipidemia, and cutaneous toxicity.
Like 9ꢀcisꢀRA, sideꢀeffects of bexarotene arise from disrupting nonpermissive receptor
2
5
activation, as in TR resulting in hypothyroidism, or stimulating permissive receptor activation,
26
27
as in LXR leading to hyperlipidemia or in RAR effecting cutaneous toxicity with therapeutic
dose concentrations. Hence, the approach to design, model and synthesize more potent and
selective rexinoids whose pleiotropic effects are attenuated is active area of research. Adding to
the motivation to create novel rexinoids, RXR selective agonists have recently been investigated
2
8
in neurodegenerative diseases such as Parkinson’s disease , and a recent POC trial of bexarotene
in patients with mild to moderate Alzheimer’s disease (AD)—driven by controversial murine
29
30
studies in AD —revealed provocative results in critical human biomarkers in the clinic.
Many RXR specific ligands modeled after 9ꢀcis RA and 1 have been reported. For
3
1
instance, the cyclopropylꢀdienoic acid (3) has been shown to be more potent as an RXR agonist
3
2
33
than its enantiomer. The 9cUAB30 (4) is in advanced clinical trials for breast cancer , and
3
4
many 9cUAB30 analogs have also been described. Arylꢀtrienoic acid RXR agonists either
35
36
37
unbranched , locked by one or with multipleꢀfused rings were reported by Boehm et al., and
5
exemplifies the latter. The substitution of a fluorine atom for a hydrogen atom adjacent to the
3
8
acid group on 1 gives compound 6 , and substituting fluorine for the last remaining hydrogen
39
atom adjacent to the acid group on 6 gives compound 7 . Both 6 and 7 show increased RXR
ACS Paragon Plus Environment

Page 7 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
activation in HCTꢀ116 and Cacoꢀ2 cell lines, respectively. Substituting a heteroaromatic
4
0
pyridine ring for the benzene ring of 1 gives compound 8 and modifying the bridging alkene
4
0
group of 8 to a bridging cyclopropyl group gives compound 9 (LGD100268) , which shows a
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
40
41
greater potency in CVꢀ1 cells than 1. Compound 10 possesses an unsaturation in the aliphatic
4
2
43
44
ring of 1, and has been produce in kilogram quantities . Acrylic acids 11 (CD3254) and 12
4
5
(
CD2915) are also potent RXR agonists. Our group recently reported rexinoids 13 – 19 , as
novel analogs of 8 – 11, and several of novel 13 – 19 exhibited differential sideꢀeffect profiles
4
6
and gene expression in vivo. Recently, Kakuta and coꢀworkers described a potent diꢀaryl amine
47
47
carboxylic acid (20) , more commonly referred to as NEt-TMN , and several assays
characterized this compound as a potent RXR agonist. Analogous compounds to 20 that have
4
8
48(a), 48(b)
been reported include the diarylꢀamine (21) and its methylated analog (22)
.
Figure 2
Kakuta and coworkers used 20 as a starting point to develop partial agonists to serve as
diabetes therapeutics without the side effects profile that comes from treatment with a potent
RXR agonist. However, in diseases such as cancer, there is still great need to develop additional
potent RXR agonists. While weighing the side effect profiles and therapeutic potential for
compounds, we must be cognizant that more effective therapeutics are often approved despite the
potential for sideꢀeffects. In the case of potent RXR agonists, the more severe sideꢀeffects of
hypothyroidism and triglyceridemia are often managed with hormone replacement therapy and
statins. Due to the selective and potent RXR agonism observed for compounds 1 and 20ꢀ22, our
laboratories have recently focused on preparing the seven novel analogous compounds, 23ꢀ29,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 66
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
modeled after compounds 20ꢀ22, as well as the sulfonic acid analogs of bexarotene, 30 and 31.
These novel rexinoids were subsequently assessed for RXR activation and EC profile, RAR
50
crossꢀover, potential for causing hypertriglyceridemia, ability to inhibit cancer cell proliferation,
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
and mutagenicity in sideꢀbyꢀside assays with 1 and compounds 20ꢀ22.
Figure 3
Results and Discussion
Molecular Modeling
Docking scores for compounds 20ꢀ32 relative to bexarotene are shown in Fig. 4; each circle
represents the best pose to a particular crystal structure, and ∆G < 0 indicates more favorable
predicted binding than bexarotene. The averaged calculated docking scores were within 1.2
kcal/mol of bexarotene, with the most favorable binding for compounds 30 and 31. Poses of the
docked compounds were largely similar to each other (this is illustrated for compounds 20 and
23 in Fig. 5), with the fused ringꢀsystems in near identical positions and small shifts in the
positions of the single rings and bridgeheads to optimize binding. The acidic groups on the single
ring of the compounds were observed to hydrogen bond to Arg 315.
Figure 4 & Figure 5
The Chemistry
4
7(a)
The route of Kakuta and coꢀworkers was followed to reproduce NEt-TMN (20)
from
commercially available 1,1,4,4ꢀtetramethylꢀ6ꢀnitroꢀ1,2,3,4ꢀtetrahydronaphthalene (32), and
details about synthetic intermediates (33ꢀ36) en route to 20 are provided (Supporting
ACS Paragon Plus Environment

Page 9 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Information). The known compound 4ꢀ(ethyl(5,5,8,8ꢀtetramethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ
2ꢀyl)amino)benzoic acid (21) was prepared using a palladium coupling method reported by
4
7(c)
Kakuta and coꢀworkers
, and experimental procedures for the synthetic intermediates (37ꢀ39)
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
en route to 21 are also provided (Supporting Information). Fortuitously, compounds 20 and 21
both formed transparent, single crystals from DMSO. Hence, Xꢀray diffraction studies were
performed and revealed the structures of 20—incorporating DMSO within the crystal in a 1:1
ratio—and 21 (Supporting Information).
The known compound, 4ꢀ(ethyl(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀ
yl)amino)benzoic acid (22), was synthesized in a similar method as 21. The method of Kakuta
4
7(e)
and coꢀworkers was used to nitrate 40 with nitric acid in acetic anhydride to give 41
in 50.4%
4
9
yield, and 41 was subsequently reduced by palladium catalyzed reduction in a flow reactor to
47(e)
give amine 42
system to
in 97% yield. Next, amine (42) was coupled to 37 in a palladium (0) catalytic
give methyl 4ꢀ((3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀ
yl)amino)benzoate (43) in 81% purified yield. Methyl ester 43 was treated with sodium hydride
in DMF followed by ethyl iodide to give methyl 4ꢀ(ethyl(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀ
tetrahydronaphthalenꢀ2ꢀyl)amino)benzoate (44) in 97% purified yield. The methyl ester of 44
was saponified under standard methanolic alkyline conditions at reflux followed by acidification
with 20% hydrochloric acid to give 4ꢀ(ethyl(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ
2ꢀyl)amino)benzoic acid (22) in 71% yield (Scheme 1). Again, a transparent crystal of 44 was
given from chloroform, and an Xꢀray diffraction study revealed disorder in the aliphatic ring
system (Supporting Information).
Scheme 1
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 66
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
The
synthesis
of
novel
compound
6ꢀ(ethyl(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀ
tetrahydronaphthalenꢀ2ꢀyl)amino)nicotinic acid (23) could not be accomplished starting with an
approach of coupling amine 42 to 34 through a nucleophilic aromatic substitution reaction with
conditions similar to those used in the synthesis of 20—running this type of reaction, we simply
recovered 42 and 34. Hence, we used the palladium coupling approach whereby 6ꢀbromoꢀ
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
4
5
1
,1,4,4,7ꢀpentamethylꢀ1,2,3,4ꢀtetrahydronaphthalene (45)
was coupled to methyl 6ꢀ
aminonicotinate (46) to give methyl 6ꢀ((3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀ
yl)amino)nicotinate (47) in 37% purified yield. Methyl ester 47 was ethylated under standard
substitution
conditions
to
give
methyl
6ꢀ(ethyl(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀ
tetrahydronaphthalenꢀ2ꢀyl)amino)nicotinate (48) in 78% purified yield. The methyl ester of 48
was saponified in methanolic potassium hydroxide followed by acidification to give 6ꢀ
(
ethyl(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)nicotinic acid (23) in 91%
yield (Scheme 2).
Scheme 2
4
7(a)
A similar route to that of Kakuta and coꢀworkers to make NEt-TMN
novel compound 2ꢀ(ethyl(5,5,8,8ꢀtetramethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀ
yl)amino)pyrimidineꢀ5ꢀcarboxylic acid (24). Amine (33) was mixed with methyl 2ꢀ
chloropyrimidine (49) and pꢀTsOH and refluxed in 1,4ꢀdioxane for 18 h to give methyl 2ꢀ
(5,5,8,8ꢀtetramethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)pyrimidineꢀ5ꢀcarboxylate (50) in
was followed to
make
(
7
5% yield. Amine (50) in DMF was ethylated to give methyl 2ꢀ(ethyl(5,5,8,8ꢀtetramethylꢀ
ACS Paragon Plus Environment

Page 11 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)pyrimidineꢀ5ꢀcarboxylate (51) in 75% yield. Finally,
methyl ester 51 was saponified to 24 in 97% yield (Scheme 3).
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
Scheme 3
The
novel
pyrimidine
compound
2ꢀ(ethyl(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀ
tetrahydronaphthalenꢀ2ꢀyl)amino)pyrimidineꢀ5ꢀcarboxylic acid (25), on the other hand, was
prepared using the palladium coupling methodology. Aryl bromide 45 was coupled to methyl 2ꢀ
aminopyrimidineꢀ5ꢀcarboxylate (52) in the usual palladium (0) catalytic system to give methyl 2ꢀ
(
(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)pyrimidineꢀ5ꢀcarboxylate (53)
in 45% purified yield. Methyl ester 53 ethylated to give methyl 2ꢀ(ethyl(3,5,5,8,8ꢀpentamethylꢀ
,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)pyrimidineꢀ5ꢀcarboxylate (54) in 65% purified yield.
5
Methyl ester 54 was saponified to give 2ꢀ(ethyl(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀ
tetrahydronaphthalenꢀ2ꢀyl)amino)pyrimidineꢀ5ꢀcarboxylic acid (25) in 84% yield after
acidification and purification (Scheme 2).
The pyrazine analog 5ꢀ(ethyl(5,5,8,8ꢀtetramethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀ
yl)amino)pyrazineꢀ2ꢀcarboxylic acid (26) of NEt-TMN was synthesized in an analogous manner
4
7(a)
using the same methodology as Kakuta.
Amine (33) was reacted with methyl 5ꢀ
chloropyrazineꢀ2ꢀcarboxylate (55) and pꢀTsOH by refluxing in 1,4ꢀdioxane for 18 h to give
methyl 5ꢀ((5,5,8,8ꢀtetramethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)pyrazineꢀ2ꢀcarboxylate
(
56) in 56 % yield. Amine (56) in DMF was deprotonated with sodium hydride, and the
resulting anion was reacted with ethyl iodide to produce methyl 5ꢀ(ethyl(5,5,8,8ꢀtetramethylꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 66
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
5
,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)pyrazineꢀ2ꢀcarboxylate (57) in 51% yield. Finally,
methyl ester 57 was saponified to 26 in 71% yield under standard conditions (Scheme 3).
Again, we alternately employed the palladium coupling to access novel compound 5ꢀ
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
(ethyl(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)pyrazineꢀ2ꢀcarboxylic
acid (27). First, arylꢀbromide 45 was coupled to methyl 5ꢀaminopyrazineꢀ2ꢀcarboxylate (58) to
give methyl 5ꢀ((3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)pyrazineꢀ2ꢀ
carboxylate (59) in 63% purified yield. Methyl ester 59 was treated with sodium hydride in
DMF followed by ethyl iodide to give methyl 5ꢀ(ethyl(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀ
tetrahydronaphthalenꢀ2ꢀyl)amino)pyrazineꢀ2ꢀcarboxylate (60) in 25% purified yield. The methyl
ester of 60 was saponified under the usual conditions to provide 27 in 57% yield. (Scheme 2).
Again, we were fortunate to isolate transparent crystals of 26 and 60 and Xꢀray diffraction
studies were performed (Supporting Information).
To probe how the fluorination of 21 effects its activity, we prepared 4ꢀ(ethyl(5,5,8,8ꢀ
tetramethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)ꢀ2ꢀfluorobenzoic acid (28) in a palladium
coupling reaction starting from the commercially available 2ꢀfluoroꢀ4ꢀiodobenzoic acid (61).
The benzoic acid 61 was converted to methylꢀester 62 by slowly adding thionyl chloride to a
solution of 61 in methanol at 0 °C followed by reflux for 1h in a modified method of Kakuta and
4
7(c)
coꢀworkers.
Subsequently, amine (33) was coupled to 62 in the standard palladium (0)
catalytic system to give methyl 2ꢀfluoroꢀ4ꢀ((5,5,8,8ꢀtetramethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀ
yl)amino)benzoate (63) in 52% purified yield. Methyl ester 63 was ethylated to provide methyl
4
ꢀ(ethyl(5,5,8,8ꢀtetramethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)ꢀ2ꢀfluorobenzoate (64) in
8% purified yield, and 64 was saponified to make 4ꢀ(ethyl(5,5,8,8ꢀtetramethylꢀ5,6,7,8ꢀ
6
ACS Paragon Plus Environment

Page 13 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
tetrahydronaphthalenꢀ2ꢀyl)amino)ꢀ2ꢀfluorobenzoic acid (28) in 85% yield after isolation
(
Scheme 1).
To complete our preparation of NEt-TMN analogs, novel compound 4ꢀ(ethyl(3,5,5,8,8ꢀ
pentamethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)ꢀ2ꢀfluorobenzoic acid (29) was
synthesized according to the route shown in Scheme 1. Amine (42) was coupled by palladium to
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
2
ꢀfluoroꢀ4ꢀiodoꢀmethylbenzoate (62) to access methyl 2ꢀfluoroꢀ4ꢀ((3,5,5,8,8ꢀpentamethylꢀ
,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)benzoate (65) in 65% purified yield. Methyl ester 65
5
was ethylated with ethyl iodide to methyl 4ꢀ(ethyl(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀ
tetrahydronaphthalenꢀ2ꢀyl)amino)ꢀ2ꢀfluorobenzoate (66) in 91% purified yield. Methylꢀester 66
was saponified to 29 in 84% yield (Scheme 1). We were pleased to recover a single crystal of 28
from DMSO, and the Xꢀray diffraction study structure is provided (Supporting Information).
Because of the excellent predicted binding for the sulphonic acid analog (30) of
bexarotene (1), we also set out to synthesize 4ꢀ(1ꢀ(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀ
tetrahydronaphthalenꢀ2ꢀyl)vinyl)benzenesulfonic acid (30) and its ketone analog 4ꢀ(3,5,5,8,8ꢀ
pentamethylꢀ5,6,7,8ꢀtetrahydronaphthaleneꢀ2ꢀcarbonyl)benzenesulfonic
acid
(31).
Commercially available potassium 4ꢀcarboxybenzenesulfonate (67) was refluxed in thionyl
chloride and thereby converted to 4ꢀ(chlorosulfonyl)benzoyl chloride (68). Benzoyl chloride
(
68) was mixed with 40, and when aluminum chloride was added to the mixture, the Friedelꢀ
Crafts acylation provided 4ꢀ(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀtetrahydronaphthaleneꢀ2ꢀ
carbonyl)benzeneꢀ1ꢀsulfonyl chloride (69) in 86% yield. Sulfonyl chloride 69 was converted to
1 in 56% yield by treatment of 69 with sodium hydroxide in acetone followed by acidification
with 20% HCl. Sulfonyl chloride 69 was also observed to react with 2ꢀmethylpropanol in
acetone and triethylamine to give isobutyl 4ꢀ(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀ
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 66
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
tetrahydronaphthaleneꢀ2ꢀcarbonyl)benzenesulfonate (70) in 87% yield. When the isobutyl
sulfonate ester 70 was treated with a solution of triphenylphosphine methylide, isobutyl 4ꢀ(1ꢀ
(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)vinyl)benzenesulfonate
(71)
was
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
produced in 48.9% yield. The vinyl sulfonate (71) was refluxed with KOH in methanol and then
treated with 20% HCl to give 4ꢀ(1ꢀ(3,5,5,8,8ꢀpentamethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀ
yl)vinyl)benzenesulfonic acid (72) in 80% yield (Scheme 4).
Scheme 4
Biological Assays and Rationale
Mammalian twoꢀhybrid assay demonstrates that several novel analogs induce RXR
homodimerization: Biologic evaluation of the synthetic analogs described above (compounds
20-31) was first conducted using the mammalian twoꢀhybrid system in the human colorectal
carcinoma cell line, HCT116. Employing this assay, ligand binding to a recombinant RXR
containing an activation domain will induce dimerization with an RXRꢀGal4 fusion protein. The
subsequent dimer will then bind to multiple Gal4 DNA response elements which are used to
drive expression of firefly luciferase.
We tested the indicated compounds at three different concentrations (25 nM, 100 nM and
1000 nM) in this system. The 25 nM assay revealed that five compounds (20, 23, 24, 25, 29, 22)
possessed transcriptional activity considerably higher than compound 1 (bexarotene).
Compounds 20, 23, 24 and 25 revealed transcriptional activity that was 2ꢀfold higher than
compound 1, while compound 29 and 22 was increased 40ꢀ50% (Figure 6A). Conversely,
ACS Paragon Plus Environment

Page 15 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compounds 31, 30, 27 and 26 were dramatically reduced in their ability to induce RXR
homodimerization and displayed activity of 0.6, 0.8, 3.4 and 11.3% when compared to
compound 1, respectively. At 100 nM treatment (Figure 6B), four compounds (20, 23, 24 and
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
2
5) exhibited increased transcriptional activity in the range of 135ꢀ146%, when compared to
compound 1. Additionally, compounds 29, 22, 28, 21 had activity similar to compound 1.
Analogous to the 25 nM treatment (Figure 6A), compounds 31 and 30 were not able to induce
RXR homodimerization, but analogs 26 and 27 exhibited a significant increase in activity (5 to
16ꢀfold) at the higher 100 nM analog concentration (compare Figure 6A and 6B). Furthermore,
at 1000 nM treatment (Figure 6C), compounds 20, 23, 24, 25 and 26 exhibited activities greater
than compound 1, with activity increased by 35ꢀ58%. In contrast, even at this elevated
concentration, compound 31 was not able to induce RXR homodimerization, while compound 30
had activity of 17% compared to compound 1. Interestingly, compounds 29 and 22 exhibited the
greatest transcriptional activity at 25 nM (140% and 150% of compound 1) and lower activity at
1
000 nM (51ꢀ56% of compound 1). These results indicate that novel analogs modeled after
compound 1 can be generated that increase RXR homodimerization, presumably via increased
binding to RXR (see Table 1). Furthermore, acidification of the molecule does not increase
RXR homodimerization, as illustrated with compounds 31 and 30. Also, fluorination does not
alter homodimerization, as shown by testing of compounds 29 and 28 compared to 22 and 21.
Figure 6
Determination of EC values: We utilized the mammalian twoꢀhybrid system in HCT116 colon
5
0
cancer cells to determine the EC value for compounds 20-31. Full dose response curves were
50
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 66
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
generated using a wide range of analog concentrations of 0.1 nM to 2000 nM for compounds 20-
2
2
3
9 and 1 µM to 80 µM for compounds 31 and 30. The results reveal that compounds 20, 23, 24,
5, 29, and 22 have an EC lower than compound 1, with EC values of 13.8, 13.8, 40.9, 18.2,
50
50
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
3.8 and 7.9 nM, respectively (Table 1). These results are not surprising because these
compounds exhibited greater RXR homodimerization at 25 nM compared to compound 1 (Figure
A). In contrast, compounds 26, 27, 28 and 21 have EC values that were 2.4 to 7ꢀfold higher
7
5
0
than compound 1. As observed in the mammalian twoꢀhybrid assay (Figure 6), acidification of
the parent compound 1 to generate 30 and 31 resulted in decreased homodimerization, and the
EC for compound 30 was determined to be approximately 18,200 nM which is 350ꢀfold higher
50
than compound 1, while an EC50 value for 31 could not be calculated even when dosing at
80,000 nM.
The EC50s for 30 and 31 were surprising, since docking predicted highly favorable
binding (Fig. 1), with hydrogen bonding between the sulfonate of the compounds and Arg 315 of
5
0
the RXR (residue numbering as in 1MVC ). This disagreement could stem from the implicit
51
solvent, which might mistreat dehydration. Alternatively, compounds 30 and 31 may bind to
RXR but not activate the receptor, a profile that would be more indicative of an antagonist.
Therefore, we conducted competitive binding experiments in which we treated HCTꢀ116 cells
with increasing concentrations of 30 and 31 and a standard concentration of bexarotene, and
these experiments revealed decreasing bexaroteneꢀdirected activation of RXR in the presence of
increasing 30 and 31 (Figure S12 in the Supporting Information). These results suggest that 30
and 31 likely reach the binding pocket of RXR, but they appear to exert antagonism compared to
1
. Finally, it is interesting to note that compounds 20 and 23 exhibit almost identical EC values
50
ACS Paragon Plus Environment

Page 17 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
(
Table 1), and the best docked poses (Fig. 4) suggest very similar modes of binding despite the
structural difference of a methyl substitution.
Table 1
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
Novel analogs are able to direct RXR/RXRE homodimerꢀmediated transcription:
The
mammalian twoꢀhybrid assay is a useful, initial screen for RXR agonist induced
homodimerization because of its accessibility, speed and sensitivity. However, it is necessary to
determine if RXR agonists are able to induce transcription in a natural protein and DNA
configuration. Therefore, we performed a second screening assay that utilizes transfection of
human RXRα and an authentic RXRE from the rat cellular retinol binding protein II gene. As
with the mammalian twoꢀhybrid assay, three concentrations of compounds were used, 25 nM,
100 nM and 1000 nM. At 25 nM (Figure 7A) compounds 20, 23, 24, 25, 29, 22 and 28 exhibited
increased RXRꢀmediated transcriptional activity when compared to compound 1, with the values
being amplified by 27ꢀ66%. The greatest activity was observed with compounds 20, 23, 24, and
25. These results are similar to those obtained in the mammalian twoꢀhybrid system (Figure
6A). Taken together, it is suggested that at 25nM compounds 20, 23, 24, 25, 29, 22 and 28 are
more efficient than compound 1 at binding to RXR and inducing RXR homodimerization, DNA
binding and initiation of transcription via an RXRE. Conversely, compounds 31, 30, 26 and 27
are not as effective at initiating transcription from an RXRE as compound 1. As expected from
the mammalian twoꢀhybrid assay, compound 31 was unable to initiate transcription above the
solvent control. At 100 nM, compounds 20, 23, 24 and 25 exhibited a 68ꢀ89% increase in
transcriptional activity when compared to compound 1 (Figure 7B). Compounds 29, 22, 28, and
21 showed slight increases in RXRE transcription of approximately 16ꢀ32%. At 1000 nM, only
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 66
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
compounds 20, 23, 24, 25 and 26 exhibited increased activity, ranging from 20ꢀ43% (Figure 7C).
Interestingly, compounds 27, 29, 22, 28 and 21 displayed similar RXRE activity as compound 1,
despite having reduced activity in the mammalian twoꢀhybrid system (compare Figure 6C 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
7C), suggesting that the specific DNA platform recognized by the RXR homodimer may play a
role in agonistꢀreceptor affinity/activity.
Figure 7
Quantitation of RAR agonist activity: Since compound 1 is known to possess “residual” retinoic
acid receptor (RAR) agonist activity, we evaluated the ability of compounds 20-31 to induce
transcription via the retinoic acid response element (RARE) and RAR. Human embryonic kidney
cells (HEK293) cells were transfected with human RARα and dosed with 100 nM of either allꢀ
trans retinoic acid (RA), the natural ligand for RARα, compound 1 or analogs. Within this
assay, compound 1 possessed 27% of the activity of RA (Figure 8A). Compound 20 displayed
the greatest RARE activation at 26% of RA, while compound 29 showed the lowest RARE
activity at 11%, which is indistinguishable from the ethanol control. Since one goal of this work
was to develop specific RXR agonists, an index of crossꢀreactivity (CR) was generated by
calculating the RARE:RXRE ratio, with compound 1 set to 1.0. Figure 8B reveals that RA, as
expected, has a very high index of crossꢀreactivity with a ratio of 9.5 indicating a preference to
activate RAR over RXR. Conversely, compounds 25, 29, 22 and 28 have the lowest crossꢀ
reactivity index at 0.35ꢀ0.37, indicating that they preferentially bind and activate RXR 3ꢀfold
more efficiently than RAR (compared to the parent compound 1). Compounds 20, 26 and 27
ACS Paragon Plus Environment

Page 19 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
have a CR>0.5. Taken together, this assay reveals that compounds 20-29 have a lower CR than
compound 1 suggesting that they are more selective RXR agonists than the parent compound.
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
Figure 8
Evaluation of SREBP promoter activation: One of the primary side effects of treatment with
bexarotene (compound 1) is hypertriglyceridemia. Lipid metabolism is regulated, in part, by the
sterol regulatory elementꢀbinding proteins (SREBPs), which are a group of transcription factors
that regulate genes involved in lipid biosynthesis. Expression of SREBP is regulated in part by
the liver X receptor (LXR)/RXR heterodimer via LXRE control elements in the SREBP
promoter. We tested the ability of compounds 20-31 to activate the SREBP promoter alone or in
the presence of the LXR agonist TO901317 (TO). Figure 9A shows that compounds 26, 27, 29,
2
2, 28 and 21 have decreased transcriptional activity, ranging from 54ꢀ79% of the activity of
compound 1, while compounds 20, 23, 24 and 25 have a similar activity profile relative to
compound 1. Not surprisingly, compounds 31 and 30, which do not activate RXR (Figures 6 and
7), do not stimulate the SREBP promoter and the activity of these analogs is similar to the
ethanol control. Furthermore, in the presence of analog and TO (activated LXR), only
compounds 26, 27, 28 and 21 show reduced SREBP promoter activation, compared to compound
1 and TO. Compound 26 + TO shows a reduction of 33% and compound 27 displays a 29%
reduction. Although compounds 20, 23, 24 and 25 have similar transcriptional activity as
compound 1, alone, there is a 13ꢀ17% decrease when combined with TO. The lipid risk
assessment (LR) was generated by calculating the SREBP:RXRE activity ratio and is a measure
of the potential risk of SREBPꢀdriven hypertriglyceridemia associated with patient treatment of
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 66
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
compound 1. Compounds 26, 29, 22 and 28 show a LR ranging from 0.48ꢀ0.65 (Figure 9B),
with compounds 29, 22, 28 and 21 possessing both an increase in RXRE transcriptional activity
(Figure 7B) and a decrease in SREBP transcription (Figure 9A), when compared to compound 1.
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
These results suggest that it is possible to generate analogs of compound 1 with increased RXR
activity, while potentially decreasing adverse side effects of hypertriglyceridemia. Conversely,
compound 27 possesses both a decrease in RXR/RXRE transcriptional assays (Figure 7B) and a
decrease in SREBP promoter activity (Figure 9A). Furthermore, the LR for compounds 20, 23,
24 and 25 range from 0.49ꢀ0.51 as a result of increased transcriptional activity in the RXRE
assay of 68ꢀ89% (Figure 7B), while maintaining similar SREBP transcriptional activity as
compound 1 (Figure 9A). In summary, the results for compounds 20, 23, 24, 25 and 28 indicate
that these compounds display at least a 2ꢀfold preference for activating RXR/RXRE transcription
versus SREBP (i.e., an LR of ≤ 0.50).
Figure 9
Evaluation of RXRꢀmediated induction of CYP3A4: CYP3A4 is a member of the cytochrome
P450 family of proteins and is responsible for the oxidation and clearance of approximately 50%
of xenobiotics by the liver. It has been shown that compound 1 is sufficient to induce CYP3A4
5
2
expression. We tested compounds 20-31 for their ability to induce CYP3A4 mRNA expression
using qPCR in the liver carcinoma cell line HepG2. There was variable induction of CYP3A4
ranging from a 3ꢀfold repression (compound 28), to a 2ꢀfold induction for compounds 31 and 24,
while compound 1 resulted in a 3.2ꢀfold induction (Figure 10). In addition to compound 28,
compound 21 also repressed CYP3A4 expression by 2.3ꢀfold. Compounds 23 and 29 did not
ACS Paragon Plus Environment

Page 21 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
alter CYP3A4 expression, while compounds 20, 25, 26 and 27 resulted in a modest increase of
approximately 1.5ꢀfold in CYP3A4 expression. These results demonstrate that for one metabolic
breakdown pathway, compounds 20-31 exert repressive effects that in vivo (depending on the
totality of metabolism that occurs) could potentially translate into differential halfꢀlife profiles
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
4
6
for these compounds compared to 1, as previously observed by our group.
Figure 10
Inhibition of proliferation in CTCL cells: Since compound 1 is used clinically to treat cutaneous
Tꢀcell lymphoma (CTCL), compounds 20-31 were evaluated for their ability to inhibit growth of
the CTCL cell line, Hut78. Figure 11 indicates that none of the compounds tested were as
efficient at inhibiting Hut78 growth as compound 1. However, compounds 20, 23 and 24 were
able to inhibit growth >85% as efficiently as compound 1. Furthermore, compounds 25, 29 and
2
1 also maintained 55ꢀ80% of the antiꢀproliferative capacity of the parent compound 1. There
was a substantial decrease in the antiꢀproliferative effects of compounds 26, 27, 22 and 28, which
were only 35ꢀ51% as effective as compound 1.
Figure 11
Inhibition of HDAC1 expression in CTCL cells: Since none of the novel analogs tested were as
effective at inhibiting cell growth of Hut78 cells, we examined the ability of a subset of analogs,
2
0 and 25 to inhibit histone deacetylaseꢀ1 (HDAC1) expression. HDAC inhibitors have been in
development for several years with four receiving FDA approval, including Vorinostat for
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 66
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
53
treatment of CTCL.
Figure 12 shows that compound 1 was able to suppress HDAC1
expression by 10%, while compound 25 showed a 28% reduction. Surprisingly, compound 20
did not alter HDAC1 expression. These data suggest, that although compound 25 did not inhibit
growth as effectively as compound 1 in the proliferation assay (Figure 11), it may be more
effective at inducing toxicity or apoptosis by altering gene expression and protein function,
compared to compound 1.
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 12
Mutagenesis: Further determining the suitability of the compounds, we undertook a screen of the
compounds to test for their ability to mutate DNA, a very undesirable trait in any potential
3
9, 45
pharmaceutical. We assayed the compounds as in previous work
of the compounds are mutagenic in this assay.
and determined that none
Conclusions
Bexarotene (1) possesses FDA approval for cutaneous Tꢀcell lymphoma, and has been examined
in a plethora of other cancer types. Recent studies also suggest that 1 may be a promising
2
9, 54
candidate to treat Alzheimer's disease (AD).
The current study was interested in factors that
may mitigate the sideꢀeffect profile of 1, since bexarotene appears to effectively treat many
CTCL patients, but also has the potential to activate RXR heterodimer pathways and thereby
induce hyperlipidemia and hypothyroidism. There is compelling motivation to explore novel
RXR ligands for therapeutic treatment, since new rexinoids may possess unique activities and
39, 45
reduced sideꢀeffect profiles, and we have previously demonstrated
a number of novel
ACS Paragon Plus Environment

Page 23 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
rexinoids with different PK characteristics and differential ability to modulate RXRꢀcontrolled
4
6
pathways. The current study has produced 9 rexinoids, 5 of which were determined to possess
lower EC s than 1. More potent rexinoids, possessing lower EC values, may be able to
5
0
50
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
translate to clinical treatments permitting lower dose regimens to achieve the same effect in
patients, and this may have the potential to reduce doseꢀdependent sideꢀeffects. We address the
primary concern about rexinoids’ tendency to raise lipids by assessing this panel of analogs in an
SREBP assay. Interestingly, compounds 21, 22, 28 and 29 all exhibit greater RXRE activation
and reduced SREBP promotion versus 1. Furthermore, given the recent development of
vorinostat and romidepsin histone deacetylase inhibitors (HDACi), and FDA approved of these
HDACi for the treatment of cutaneous Tꢀcell lymphoma (CTCL), we found that compound 25
exhibited a greater suppression of HDAC1 than 1. Moreover, several of the rexinoids 20ꢀ31 also
demonstrated similar antiꢀproliferative effects in Hut78 cells as well as a lower induction of
CYP3A4 as 1—suggesting that some of the novel rexinoids may perhaps possess modified halfꢀ
life profiles in comparison to 1. Taken together, a number of our novel RXR agonists in this
study exhibit greater onꢀtarget potency, similar efficacy at inhibition of CTCL proliferation, and
a lower likelihood to induce side effects than 1.
Experimental Section
Ligand Docking and Scoring. Docking studies were carried out for bexarotene and ligands 20ꢀ
55
3
1. The ligands were prepared by Avogadro , following an energy minimization with the
5
6
57
Universal Force Field . The charge distribution on ligands were assigned by Autodock and
5
8
OpenBabel 2.3.0 , respectively. Prepared ligands were docked to a group of target proteins,
50
50
59
8
which include the human RXRα (protein data bank code 1MVC , 1MZN , 1RDT , 2P1T ,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 66
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
60
60
61
61
61
43
62
63
62
2
P1U , 2ZXZ , 2ZY0 , 3DZY , 3DZU , 3E00 , 3FUG , 3H0A , 3PCU , 4K6I ,
6
4
respectively) and RXRβ (protein data bank code 1H9U ). Docking was performed using the
5
7
39, 45
Autodock 4.2 software . As in our previous studies
, protein residues Arg 316 and Ile 268
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
(
residue numbering as in 1MVC) were treated flexible. Docking used the Lamarckian genetic
algorithm with 25,000,000 energy evaluations per dock and a total of 250 docks per compound.
Calculated binding free energies were used to score the compounds.
Mammalian Two-Hybrid Assay. HCTꢀ116 colorectal carcinoma cells were plated overnight at
80,000 cells per well in a 24ꢀwell plate and maintained in DMEM + glutamine + sodium
pyruvate (Mediatech, Manassas, VA) containing 10% FBS (Atlanta Biologicals, Atlanta, GA),
supplemented with 100 µg/mL streptomycin and 100 U/mL penicillin. (Gibco, Gaithersberg,
MD). Cells were coꢀtransfected using a human RXR binding domain (BD) vector (bait), a
human RXR activation domain (AD) vector (prey), pFRꢀLuc and renilla control plasmid.
Transfection was conducted using 1.25 µL of Polyjet (Signagen, Gaithersberg, MD) per well and
allowed to incubate for 24 hours. The cells were then treated with ethanol or DMSO vehicle
(
0.1%), compound 1 or the indicated analogs at a final concentrations of 25 nM, 100 nM or 1000
nM and incubated for 24 hours. The rexinoid activity was measured by luciferase output
utilizing a dualꢀluciferase report assay system according to the manufacturer’s instructions
(
Promega, Madison, WI) in a Sirus luminometer (Berthold Detection System, Pforzheim,
Germany). Three independent assays were conducted with triplicate samples for each treatment
group. The value of compound 1 is set to 100%.
EC50 Determination. The EC values were determined from full doseꢀresponse curves for
50
compound 1 and compounds 20ꢀ29 ranging from 0.1 to 2000 nM. EC curves for 30 and 31
50
were determined from doseꢀresponse curves ranging from 1 to 80 µM. HCTꢀ116 cells were
ACS Paragon Plus Environment

Page 25 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
plated, transfected and treated as above using the mammalian twoꢀhybrid assay protocols. Cells
were treated with ethanol or DMSO vehicle (0.1%) or analog (final concentration of 0.1, 0.5, 1,
2.5, 5, 10, 25, 50, 75, 100, 250, 500, 750, 1000, 2000 nM) for 24 hours. Alternatively,
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
compound 30 and 31 dose curves were generated using a final concentration of 1, 2.5, 5, 10, 20,
40, 80 µM. The amount of rexinoid activity at each concentration was measured using the same
luciferase assay described above. Dose response curves were generated using 4ꢀparameter
analysis utilizing Prism Software (GraphPad Inc., La Jolla, CA). Three independent assays were
conducted with triplicate samples for each treatment group.
RXRE-mediated Transcription Assay. HCTꢀ116 cells were plated at 80,000 cells per well in a
2
4ꢀwell plate and maintained as described above. The cells were coꢀtransfected using 250 ng of
RXREꢀluciferase reporter gene (RXRE from the naturally occurring responsive element in rat
cellular retinol binding protein II gene: 5’ꢀ
AAAATGAACTGTGACCTGTGACCTGTGACCTGTGACꢀ3’, DR1 responsive elements
underlined), 25 ng pSG5ꢀhuman RXRα and 20 ng of renilla control plasmid using 1.25 µL
Polyjet for 24 hours. The cells were treated with ethanol or DMSO vehicle (0.1%) or analog
(final concentration of 25 nM, 100 nM or 1000 nM) for 24 hours. The rexinoid activity was
measured by luciferase output as described above. Three independent assays were conducted
with triplicate samples for each treatment group. The value of compound 1 is set to 100%.
RARE-mediated Transcription Assay. HEK293, human embryonic kidney, cells were plated
at 60,000 cells per well in a 24ꢀwell plate and maintained as described above. After 24 hours, the
cells were transfected with 250 ng pTKꢀDR5(X2)ꢀLuc, 25 ng pCMXꢀhuman RARα and 20 ng
renilla utilizing 1.25 µL Polyjet per well for 24 hours. The sequence of the double DR5 RARE
is:
5’ꢀAAAGGTCACCGAAAGGTCACCATCCCGGGAGGTCACCGAAAGGTCACCꢀ3’
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 66
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
(
DR5 responsive elements underlined). The cells were treated with ethanol or DMSO vehicle
(0.1%) or allꢀtrans-retinoic acid (RA, the ligand for RAR) or the indicated rexinoid at a final
concentration of 100 nM. After 24 hours of treatment, the retinoid activity was measured as
described above. The activity of compound 1 or analog divided by the activity of allꢀtransꢀRA
(expressed as a percentage) represents the RAR agonist activity. Three independent assays were
conducted with triplicate samples for each treatment group. The value for RA is set to 100%.
The "index of crossꢀreactivity" was calculated by dividing RARE activity by 100 nM RXRE
activity with the RARE:RXRE ratio for compound 1 set to 1.0.
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
Sterol Regulatory Element-Binding Protein (SREBP)-mediated Transcription Assay.
HCT116 colorectal carcinoma cells were plated at 80,000 cells per well and maintained as
described above. After 24 hours, the cells were coꢀtransfected with 100 ng pBP1c(6500)ꢀLuc
which contains approximately 6500 base pairs of the flanking DNA from the murine sterol
regulatory element binding protein (SREBP)ꢀ1c natural promoter, along with 10 ng pCMXꢀ
hLXRα and 20 ng of renilla control plasmid using 1.25 µL Polyjet for 24 hours. The cells were
treated with 100 nM of the LXR agonist TO901317 alone or with 100 nM compound 1 or
analog. After 24 hours, the amount of SREBP promoter activity was measured by luciferase
output as described above. Three independent assays were conducted with triplicate samples for
each treatment group. The value of compound 1 is set to 100%. The "lipid risk assessment" was
calculated by dividing SREBP activity by 100 nM RXRE with the SREBP:RXRE ratio for
compound 1 set to 1.0.
Proliferation Assay. Human Tꢀcell lymphoma cells (Hut78) were maintained in RPMI media
(Mediatech, Manassas, VA) containing 10% FBS (Atlanta Biologicals, Atlanta, GA),
supplemented with 100 µg/mL streptomycin and 100 U/mL penicillin (Gibco, Gaithersberg,
ACS Paragon Plus Environment

Page 27 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
o
MD) at 37 C, 5% CO . Cells were plated at 7,500 cells per well in a 96ꢀwell plate and
2
immediately dosed with either media alone, ethanol, DMSO or 200 nM of compound 1 or analog
in a total volume of 100 µL. After 72 hours, 20 µL of CellTiter 96 AQueous One Solution Cell
Proliferation Assay (MTS) (Promega, Madison, WI) was added to each well and the plate was
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
o
incubated at 37 C according to manufacturer’s recommendations. MTS assays were quantified
using a BioTek ELx808 96ꢀwell plate reader (BioTek, Winooski, VT) at A490 nm. Percent
inhibition was calculated using media control set to 0.0.
Quantitative real-time PCR. Human hepatocellular carcinoma cells (HepG2) cells were
maintained in DMEM + glutamine + sodium pyruvate (Mediatech, Manassas, VA) containing
1
0% FBS (Atlanta Biologicals, Atlanta, GA), supplemented with 100 µg/mL streptomycin and
100 U/mL penicillin (Gibco, Gaithersberg, MD). Cells were plated at 400,000 cells per well in a
ꢀwell plate. After 16 hours, the media was replaced with DMEM containing 1% FBS for 24
6
hours. The cells were dosed in DMEMꢀ1% FBS containing ethanol, DMSO or compound 1 or
analog for 24 hours. Total RNA was isolated from each plate or well using an Aurum Total
RNA Mini kit (BioꢀRad, Hercules, CA) according to the manufacturer’s instructions. The RNA
obtained was quantified using A260/280 spectrophotometry.
Human Tꢀcell lymphoma cells (Hut78) were maintained in RPMI media as described above.
Cells were harvested by centrifugation, 10 minutes at 400 rcf at room temperature, and
resuspended in RPMI supplemented with 1% FBS, 100 µg/mL streptomycin and 100 U/mL
penicillin. Cells were plated at 750,000 cells per well in a 6ꢀwell plate and immediately dosed
with ethanol, DMSO, compound or analog at a final concentration of 100 nM for 48 hours. Total
RNA was extracted as described above.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 66
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
DNaseꢀtreated total RNA (0.5 µg) was reverseꢀtranscribed via the use of the iScript cDNA
Synthesis kit (BioꢀRad) to prepare 20 µL of first strand cDNA. Realꢀtime PCR reactions were
prepared by adding 1.5 µL of cDNA to 5 µL FastStart Universal SYBR Green Master Mix + Rox
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
(Roche Applied Science, Indianapolis IN) and primers for a total volume of 10 µL. Reactions
were performed in 96 well plates in a BioRad CFX96 thermal cycler using a 40 cycle
profile. Data analysis was performed using the comparative ꢁꢁCt method as the means of
relative quantitation, normalized to an endogenous reference (GAPDH) and relative to a
ꢀꢁꢁCt
calibrator (normalized Ct value from vehicleꢀtreated cells) and expressed as 2
according to
Applied Biosystems’ User Bulletin 2: Rev B, “Relative Quantitation of Gene Expression.” The
primers utilized during PCR experimentation are as follows: GAPDH forward: 5’ꢀ
ACAACTTTGGTATCGTGAAGGACꢀ3’,
CAGGGATGATGTTCTGGAGAGCꢀ3’;
GAPDH
HDAC1
reverse:
forward:
5’ꢀ
5’ꢀ
GAGATGACCAAGTACCACAGCꢀ3’ HDAC1 reverse: 5’ꢀTGACAGAACTCAAACAGGCCꢀ
’; CYP3A4 forward: 5’ꢀTCAGCCTGGTGCTCCTCTATCTAT ꢀ3’ CYP3A4 reverse: 5’ꢀ
3
AAGCCCTTATGGTAGGACAAAATATTT ꢀ3’.
Mutagenesis Assay. All compounds were tested for their ability to mutate DNA in an EPAꢀ
3
9, 45
approved Saccharomyces cerevisiae cell based system as in our previous work.
D7 yeast
6
5
cells were incubated with increasing concentrations of each compound solubilized in DMSO
(highest concentration being 0.1mg/ml) and then plated on yeast extract, peptone, dextrose
media; and synthetic media lacking isoleucine or tryptophan, to check for forward and reverse
6
6
mutations.
1
13
Instrumentation. A 400 MHz Bruker spectrometer was used to acquire H NMR and C NMR
spectra. Chemical shifts (δ) are listed in ppm against residual nonꢀdeuterated solvent peaks in a
ACS Paragon Plus Environment

Page 29 of 66
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
given deuterated solvent (e.g. CHCl in CDCl ) as an internal reference. Coupling constants (J)
3
3
are reported in Hz, and the abbreviations for splitting include: s, single; d, doublet; t, triplet; q,
1
3
quartet; p, pentet; m, multiplet; br, broad. All C NMR spectra were acquired on a Bruker
instrument at 100.6 MHz. Chemical shifts (δ) are listed in ppm against deuterated solvent
carbon peaks as an internal reference. High resolution mass spectra were recorded using either a
JEOL GCmate(2004), a JEOL LCmate(2002) high resolution mass spectrometer or an ABI
Mariner (1999) ESIꢀTOF mass spectrometer.
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
General Procedures. Removal of volatile solvents transpired under reduced pressure using a
Büchi rotary evaporator and is referred to as removing solvents in vacuo. Thin layer
chromatography was conducted on precoated (0.25 mm thickness) silica gel plates with 60Fꢀ254
indicator (Merck). Column chromatography was conducted using 230ꢀ400 mesh silica gel (E.
Merck reagent silica gel 60). All tested compounds were analyzed for purity by combustion
analysis through Columbia Analytical Services (formerly Desert Analytics in Tucson, AZ) and
were found to be > 95% pure.
47(a)
NEt-TMN (20).
To a 100 mL roundꢀbottomed flask equipped with a stir bar and charged
with 6ꢀ(ethyl(5,5,8,8ꢀtetramethylꢀ5,6,7,8ꢀtetrahydronaphthalenꢀ2ꢀyl)amino)nicotinate (0.3674 g,
.002 mmol) suspended in methanol (3.5 mL) was added a solution of potassium hydroxide
0.1767 g, 3.15 mmol) in water (0.22 mL). This reaction was stirred at reflux in an oil bath at 87
1
(
°C for 1 h. The reaction was then cooled to room temperature and acidified with 20% HCl (52
mL). The crude white precipitate was filtered and washed with cold water to provide crude 20
(0.3238 g, 91%) and this crude material was purified by column chromatography (25 mL SiO2,
ACS Paragon Plus Environment