Probing the carboxyester side chain in controlled deactivation (-)-δ8-tetrahydrocannabinols

Article abstract of DOI:10.1021/jm501165d

We recently reported on a controlled deactivation/detoxification approach for obtaining cannabinoids with improved druggability. Our design incorporates a metabolically labile ester group at strategic positions within the THC structure. We have now synthesized a series of (-)-δ⁸-THC analogues encompassing a carboxyester group within the 3-alkyl chain in an effort to explore this novel cannabinergic chemotype for CB receptor binding affinity, in vitro and in vivo potency and efficacy, as well as controlled deactivation by plasma esterases. We have also probed the chain's polar characteristics with regard to fast onset and short duration of action. Our lead molecule, namely 2-[(6aR,10aR)-6a,7,10,10a-tetrahydro-1-hydroxy-6,6,9-trimethyl-6H-dibenzo[b,d]pyran-3-yl]-2-methyl-propanoic acid 3-cyano-propyl ester (AM7438), showed picomolar affinity for CB receptors and is deactivated by plasma esterases while the respective acid metabolite is inactive. In further in vitro and in vivo experiments, the compound was found to be a remarkably potent and efficacious CB1 receptor agonist with relatively fast onset/offset of action.

Full text of DOI:10.1021/jm501165d

Subscriber access provided by MONASH UNIVERSITY
Article
Probing the Carboxyester Side Chain in Controlled
8
Deactivation (-)-#-Tetrahydrocannabinols
Spyros P. Nikas, Rishi Sharma, Carol A. Paronis, Shashank Kulkarni, Ganesh A Thakur, Dow P.
Hurst, JodiAnne T Wood, Roger S Gifford, Girija Rajarshi, Yingpeng Liu, Jimit Girish Raghav, Jason
Jianxin Guo, Torbjörn U.C. Järbe, Patricia H. Reggio, Jack Bergman, and Alexandros Makriyannis
J. Med. Chem., Just Accepted Manuscript • Publication Date (Web): 03 Dec 2014
Downloaded from http://pubs.acs.org on December 4, 2014
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 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Probing the Carboxyester Side Chain in Controlled Deactivation (ꢀ)ꢀ
8
ꢁ ꢀTetrahydrocannabinols
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
†
,§
†§
†
†
†
Spyros P. Nikas, Rishi Sharma, Carol A. Paronis, Shashank Kulkarni, Ganesh A. Thakur,
£
†
†
†
†
Dow Hurst, JodiAnne T. Wood, Roger S. Gifford, Girija Rajarshi, , Yingpeng Liu, Jimit Girish
†
†
†
£
‡
Raghav, Jason Jianxin Guo, Torbjörn U.C. Järbe, Patricia H. Reggio, Jack Bergman, and
,
†,#
Alexandros Makriyannis.*
†
Center for Drug Discovery and Departments of Chemistry and Chemical Biology and
Pharmaceutical Sciences, Northeastern University, Boston, Massachusetts 02115, United States
£
Center for Drug Discovery, University of North Carolina at Greensboro, Greensboro, North
Carolina 27402, United States
‡
McLean Hospital, Harvard Medical School, Belmont, Massachusetts 02478, United States
#
King Abdulaziz University, Jeddah, 22254, Saudi Arabia
*
To whom correspondence should be addressed.
§
These two authors contributed equally to this work.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 67
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
We recently reported on a controlled deactivation/detoxification approach for obtaining
cannabinoids with improved druggability. Our design incorporates a metabolically labile ester
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
group at strategic positions within the THC structure. We have now synthesized a series of (ꢀ)ꢀꢁ ꢀ
THC analogues encompassing a carboxyester group within the 3ꢀalkyl chain in an effort to explore
this novel cannabinergic chemotype for CB receptor binding affinity, in vitro and in vivo potency
and efficacy as well as controlled deactivation by plasma esterases. We have also probed the
chain’s polar characteristics with regard to fast onset and short duration of action. Our lead
molecule,
namely
2ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀtetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀ
dibenzo[b,d]pyranꢀ3ꢀyl]ꢀ2ꢀmethylꢀpropanoic acid 3ꢀcyanoꢀpropyl ester (AM7438) showed
picomolar affinity for CB receptors and is deactivated by plasma esterases while the respective acid
metabolite is inactive. In further in vitro and in vivo experiments the compound was found to be a
remarkably potent and efficacious CB1 receptor agonist with relatively fast onset/offset of action.
ACS Paragon Plus Environment

Page 3 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Introduction
9
1
9
2ꢀ4
(
ꢀ)ꢀꢁ ꢀTetrahydrocannabinol [(ꢀ)ꢀꢁ ꢀTHC, 1, Figure 1] and its congeners act at CB1 and CB2,
two G ꢀproteinꢀcoupled cannabinoid receptors that are currently being targeted for various
i/o
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
conditions including pain, inflammation, neurodegeneration, glaucoma, eating and mental disorders
5
ꢀ14
as well as cancer.
Owing to the undesirable side effects associated with CB1 receptor
activation/deactivation as well as poor pharmacokinetic/pharmacodynamic (PK/PD) properties only
1
5
a limited number of cannabinergic drugs have been approved to date. Thus, the development of
safer THCꢀbased medications with favorable oral bioavailability, consistent efficacy, and
predictable time course of action and detoxification remains to be addressed.
Towards this end we recently reported on a controlled deactivation/detoxification approach where
the “soft” analogue/drug concept of enzymatic deactivation was combined with a “depot effect” that
9
is commonly observed with ꢁ ꢀTHC and other lipophilic cannabinoids. In our design, the
compound’s systemic half life is determined by two factors. The first reflects the ability of the
compound to sequester in some tissue reservoir such as fatty tissue (depot effect). The tissue
sequestration affects the availability of the compound for receptor activation as well as for
hydrolytic deactivation through systemic circulation. This process is dependent on the compound’s
physicochemical properties and can be modulated by adjusting log P and PSA. Thus, more
lipophilic compounds are slowly released in the bloodstream from the depot while more polar
compounds are expected to have less of a depot effect. We have also confirmed that, in the
compounds discussed here, the hydrolytic metabolic pathway is the most dominant and is
considerably faster than other metabolic options such as reactions involving microsomal enzymes.
The second is the rate of enzymatic hydrolysis of a metabolically labile ester group [ꢀC(O)ꢀOꢀ] by
blood esterases. This can be calibrated by incorporating suitable stereochemical features in the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 67
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
16
vicinity of the hydrolyzable moiety (enzymatic effect). Currently, we have developed two
controlledꢀdeactivation cannabinergic templates based on the tricyclic classical cannabinoid
1
7
prototype. In the first, the Cꢀring in THC was replaced by a hydrolyzable sevenꢀmembered lactone
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
while in the second, the metabolically labile ester group was placed at the 2'ꢀposition of the side
1
6
chain pharmacophore (Figure 1).
8
In the present SAR study, we sought to probe the ester side chain pharmacophore in the (ꢀ)ꢀꢁ ꢀTHC
prototype within a novel series of analogs with regard to affinity for the CB1 and CB2 cannabinoid
receptors, in vitro and in vivo potency and efficacy as well as ability to control the halfꢀlives of
deactivation through enzymatic action. An additional goal of this work was to explore the polar
characteristics of the side chain and gain information related to the modulation of the depot effect
and its role in the in vivo pharmacokinetic profile of the analogs. The above considerations led us to
design ligands that retain the metabolically labile group at the 2'ꢀposition of the side chain and
replace the ester group [ꢀC(O)ꢀOꢀ] with the reverse ester [ꢀOꢀC(O)ꢀ] as well as with the
corresponding thioester [ꢀC(O)ꢀSꢀ] and the hydrolytically more stable amide group [ꢀC(O)ꢀNHꢀ]
(
Figure 2). To explore the chain length we synthesized analogs with fourꢀ to nineꢀatomꢀlong side
chains. Additionally, to probe the ligand’s polarity we incorporated bromoꢀ, cyanoꢀ and imidazolylꢀ
1
6
groups at the distal side chain carbon atom. As with our previous work, to regulate the rates of
enzymatic inactivation while enhancing the compound’s affinities for CB receptors we introduced
benzylic substituents contiguous to the metabolically vulnerable ester group.
All novel compounds were characterized by determining their in vitro CB1 and CB2 receptor
affinities. Of these, the most interesting were assessed for their functional activities and for their in
vitro metabolic stabilities toward mouse plasma esterases while a limited representative set of key
compounds was evaluated for their hypothermic and analgesic effects in vivo. Our data show that (ꢀ
ACS Paragon Plus Environment

Page 5 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
8
)
ꢀꢁ ꢀTHC analogs carrying fiveꢀ to nineꢀatomꢀlong side chains that are substituted with geminal
dimethyl and cyclobutyl groups at the C1'ꢀposition exhibit remarkably high affinities for the CB1
and CB2 receptors. As predicted, all novel analogues were found to be susceptible to enzymatic
deactivation by plasma esterases in a controllable manner, while their metabolites showed no or
very low cannabinergic activity. Additionally, all key compounds were shown to be agonists for the
CB1 receptor when tested for their abilities to reduce cAMP levels and also produce the
characteristic CB1ꢀmediated hypothermia and analgesia in rats and mice. Equally important, in the
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
8
hypothermia assay the 6'ꢀcyanoꢀ2'ꢀcarboxyꢀꢁ ꢀTHC analog 10a (AM7438) exhibited 10ꢀfold higher
potency as well as faster onset and shorter duration of action than the less polar carboxy counterpart
2
b. Also, the offꢀset of the analgesic effect was significantly faster for 10a compared to 2b.
Congruent with our rational design, these in vivo results suggest that the depot effect of our
controlled deactivation analogs can suitably be modulated by enhancing the polar characteristics of
the side chain pharmacophore without any significant loss of potency.
Chemistry
Syntheses of the 4'ꢀbromoꢀ and 4'ꢀcyanoꢀbutyl esters 6aꢀ6c and 7aꢀ7c are summarized in Scheme 1.
8
The required (ꢀ)ꢀꢁ ꢀTHC carboxylic acids 5aꢀ5c were synthesized from commercially available
(
3,5ꢀdimethoxyphenyl)acetonitrile and (+)ꢀcis/transꢀpꢀmenthaꢀ2,8ꢀdienꢀ1ꢀol in three to four steps
1
6
following our recently reported procedures. Alkylation of the respective carboxylate anions with
1
,4ꢀdibromobutane under microwave heating led to the corresponding esters 6aꢀ6c in 45ꢀ63%
1
6
18
yields. Treatment of these bromides with sodium cyanide in dimethyl sulfoxide produced the
respective side chain cyanoꢀsubstituted analogs 7aꢀ7c in 54ꢀ71% yields.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 67
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 a similar fashion, the tricyclic carboxylic acid 5b was transformed to the side chain homologues
a, 8b and 9aꢀ9d in 47ꢀ87% yields (Scheme 2). Reaction of the 3ꢀbromoꢀpropyl ester 9b with
8
sodium cyanide or imidazole in the presence of potassium carbonate, in dimethyl sulfoxide, led to
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
the end carbon substituted derivatives 10a and 10b (41ꢀ98% yields).
1
9
Synthesis of the reverse ester analogue 16 involves a Mitsunobu esterification reaction as the key
2
0
o
step (Scheme 3). Thus, reduction of nitrile 11 with diisobutylaluminum hydride at ꢀ78 C led to
2
1
aldehyde 12 (92% yield) which upon exposure to sodium borohydride in methanol afforded the
respective alcohol 13 in excellent yield (94%). Cleavage of the methyl ether groups in 13 using
18
boron tribromide produced resorcinol 14 in 47% yield. Acid catalyzed condensation of this
8
intermediate with chiral terpenoid alcohol 17 in refluxing chloroform for 4ꢀ6 hours, gave (ꢀ)ꢀꢁ ꢀ
THC alcohol 15 in 21% isolated yield along with unidentified byproducts. We were able to improve
16
this yield (28%) using microwave conditions over a much shorter time period (10 min).
Subsequently, Mitsunobu esterification of 15 with nꢀvaleric acid using triphenylphosphine and
diethyl azodicarboxylate led to the final ester 16 (41% yield).
Treatment of the α,αꢀdimethylꢀcarboxylic acid 5b with [bis(2ꢀmethoxyethyl)amino]sulfur trifluoride
2
2
8
and coupling of the in situ generated acyl fluoride with nꢀpentylamine afforded the (ꢀ)ꢀꢁ ꢀTHC
amide 18 in 70% yield (Scheme 4). Exposure of the same starting material (5b) to the
2
3
benzotriazole/thionyl chloride reagent followed by treatment of the intermediate acyl chloride 19
with nꢀpropanethiol led to thioester 20 in 23% yield for the two steps.
Cannabinoid receptor affinities
The abilities of compounds 6aꢀ6c, 7aꢀ7c, 8a, 8b, 9aꢀ9d, 10a, 10b, 15, 16, 18 and 20 to displace the
radiolabeled CB1/CB2 agonist CPꢀ55,940 from membranes prepared from rat brain (source of
ACS Paragon Plus Environment

Page 7 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
CB1) and HEK293 cells expressing either mouse CB2 or human CB2 were determined as described
1
8,20
earlier
and inhibition constant values (Ki) from the respective competition binding curves are
8
8
listed in Table 1 in which our prototype (ꢀ)ꢀꢁ ꢀTHC, as well as the first generation carboxyꢀꢁ ꢀ
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
THCs 2aꢀ2c are included for comparison. The rat, mouse, and human CB1 receptors have 97ꢀ99%
1
6
sequence identity across species, and as shown earlier are not expected to exhibit variations in
2
4
their Ki values. However, mouse CB2 (mCB2) exhibits only 82% sequence identity with the
3
human clone (hCB2). This divergent nature of mCB2 and hCB2 receptors was shown in earlier
2
5,26
work
to be associated with speciesꢀbased differences in affinity. For this reason, the analogs
were also tested on hCB2.
8
The compounds included in this study are (ꢀ)ꢀꢁ ꢀTHC analogues in which a fourꢀ to nineꢀatomꢀlong
side chain with or without 1'ꢀsubstituents incorporates an enzymatically vulnerable group at the 2'ꢀ
or 3'ꢀpositions. Optimization of the novel carboxy or thiocarboxyꢀester side chains was probed
through the synthesis of analogues carrying bromoꢀ, cyanoꢀ and imidazolylꢀsubstituents at the distal
carbon atom. As predicted based on our earlier results, the hydrolytic metabolites 5aꢀ5c and 15
have no significant affinities for CB1 and CB2 receptors, thus, minimizing the possibility of
8
undesirable cannabinoid receptor related side effects. Comparison of the binding data of (ꢀ)ꢀꢁ ꢀ
THC and its carboxyester congeners 2a, 6a and 7a suggests that extension of the linear chain from
five to sevenꢀnine atoms, along with incorporation of an ester group at the 2',3'ꢀpositions enhances
the binding affinities of these analogues for both the CB1 and CB2 receptors. We also observed that
incorporation of the bromo as well as the more polar cyano substituents at the terminal carbon of
8
the side chain is well tolerated. Comparison of the binding data of our prototype (ꢀ)ꢀꢁ ꢀTHC and
compound 8b demonstrates the remarkable effects of 1',1'ꢀdimethylꢀ2'ꢀcarboxyester substitution [ꢀ
C(CH ) ꢀC(O)Oꢀ] on the side chain pharmacophore. Thus, 8b with the five atom long side chain
3
2
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 67
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
exhibits 10ꢀ, 78ꢀ and 24ꢀfold higher binding affinities for rCB1, mCB2 and hCB2 receptors,
8
respectively, when compared to (ꢀ)ꢀꢁ ꢀTHC. The oneꢀcarbon shorter homologue 8a has
significantly reduced affinities for both CB receptors, indicating that a minimum requirement for
substantial affinity for both receptors in this series is a five atom long side chain. This high affinity
can be maintained all through side chain lengths of nine atoms with or without terminal bromine
substituents (analogues 2b, 6b and 9aꢀc). Interestingly, this holds true when the terminal two or
three atoms of the chain are replaced by the polar cyano group or the bulkier and amphiprotic
imidazole ring (compounds 10a, 10b and 7b). Notably, the cyano analogs 10a and 7b exhibit
remarkably high affinities for the CB1 and CB2 receptors. However, extension of the ωꢀsubstituted
imidazolyl chain of 10b by one methylene group (compound 9d) results in a reduction of the
ligand’s affinity for both cannabinoid receptors, an effect more accentuated in CB1 (~ 8ꢀfold
reduction). Taken together, these data suggest that the pharmacophoric limits for an ωꢀsubstituted
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
',1'ꢀdimethylꢀ2'ꢀcarboxyester chain could not be extended beyond the nine atoms.
A comparison of the binding affinities of the nonꢀsubstituted analogs 6a and 7a with their
respective gemꢀdimethyl counterparts 6b and 7b shows that introduction of two methyl substituents
at the 1'ꢀposition of an ωꢀsubstituted carboxylated chain leads to an enhancement (20ꢀfold
maximum) in CB1 and CB2 receptor affinities. This increase in the ligand’s affinities for both CB1
and CB2 receptors holds true when the gemꢀdimethyl substitution is modified to the bulkier
cyclobutyl ring (analogues 6c and 7c). An examination of the binding data of the 2',3'ꢀcarboxyester
analog 2b and its sulfur and nitrogen congeners 18 and 20 shows that the ester moiety [ꢀC(O)ꢀOꢀ]
can be replaced by the respective thioester [ꢀC(O)ꢀSꢀ] and amide [ꢀC(O)ꢀNHꢀ] groups. Likewise,
analogue 16 incorporating the sterically less hindered retro ester group [ꢀOꢀC(O)ꢀ] at the 3', 4'ꢀ
position maintains high affinity for both the CB1 and CB2 receptors.
ACS Paragon Plus Environment

Page 9 of 67
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
In summary, the detailed SAR reported here shows that a fiveꢀ to nineꢀatomꢀlong side chain with
gemꢀdimethyl or cyclobutyl substituents at the 1'ꢀposition and an enzymatically vulnerable group
within the 2' or 3' chain segment results in analogues with remarkably high affinities for both CB1
and CB2 receptors. Importantly, addition of the bromoꢀ or cyanoꢀgroups as well as the bulky and
amphiprotic imidazole ring at the terminal carbon maintains or enhances the affinity of the ligand
for the CB receptors.
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
In vitro plasma stability studies
Representative analogues within this series were assessed for their in vitro plasma stability toward
1
6,17
mouse plasma esterases as detailed in Experimantal section.
It should be noted that blood
contains various esterases which play the major role in the hydrolysis of compounds carrying ester,
carbamate, or phosphate bonds. These esterases include acetylcholinesterases (ACHE),
butyrylcholinesterases (BCHE), paraoxonases and carboxyesterases (in mice and rat but not in
human plasma). Esterase activity can be found mainly in plasma, with less activity in red blood
cells. Plasma albumin itself may also act as an esterase under certain conditions. For example,
albumin contributes about 20% of the total hydrolysis of aspirin to salicylic acid in human plasma.
The esterase activity in blood seems to be more extensive in small animals such as rats than in large
27
animals and humans. A comparison of the halfꢀlives (t , Table 1) of the alkylbromide and nitrile
1
/2
having no substitution at the 1'ꢀposition (6a, 7a) with their 1'ꢀgemꢀdimethyl (6b, 7b) and 1'ꢀ
cyclobutyl (6c, 7c) counterparts shows that the plasma esterase stabilities of the analogues correlate
well with the presence and the size of 1'ꢀsubstituents. Thus, the order of metabolic stabilities for the
bromoꢀ and cyanoꢀsubstituted analogues is 6a < 7a < 6b < 7b < 6c < 7c with the compounds
carrying the bulkier cyclobutyl group being the most hydrolytically stable. This trend with the ωꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 67
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
substituted eightꢀ to nineꢀatomꢀlong side chain analogues parallels our earlier observations with the
1
6
shorter (sevenꢀatoms) and unsubstituted side chain congeners 2a, 2b and 2c. Another general
observation through this structureꢀstability relationship study is that regardless of the 1'ꢀsubstituent,
addition of a bromoꢀ or cyanoꢀgroup at the terminal carbon of a sevenꢀatomꢀlong chain increases its
stability towards esterases with the cyanoꢀsubstituted chain exhibiting the highest stability
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
(comparison of 2a with 6a and 7a, 2b with 6b and 7b, and 2c with 6c and 7c).
The effect of ωꢀsubstitution on the chain’s enzymatic stability profile was studied in more detail in
the 1'ꢀgemꢀdimethyl series through the assessment of the halfꢀlives of the cyanoꢀ and imidazolylꢀ
substituted analogues 10a, 10b, 7b and 9b. We observe that presence of a terminal cyano group or
the larger imidazole ring reduce the enzymatic lability of the 1'ꢀgemꢀdimethyl chain (analogue 2b)
by 3ꢀ to 31ꢀfold. As expected, the sterically less hindered retro ester analogue 16 was more prone to
enzymatic hydrolysis when compared to the α,αꢀdimethylꢀcarboxylated counterpart 2b, while the
amide group (18) was shown to have remarkable plasma stability. In agreement with earlier work
2
8
on the rates of hydrolysis of ethyl butyrate and ethyl thiobutyrate by pig liver esterases, the
thioester 20 exhibits higher stability in mouse plasma esterases than the respective oxoester 2b.
8
Overall, our data show that the rate of enzymatic inactivation of our side chain carboxyester ꢁ ꢀ
THCs can be controlled by 1) structural features in the vicinity of the hydrolyzable group and 2)
substituents at the terminal carbon.
Functional characterization
Functional characterization of key compounds for the rCB1 receptor was carried out by measuring
1
6,18
the decrease in forskolinꢀstimulated cAMP, as detailed earlier.
Compounds were initially
screened in three different concentrations (three point data) and the approximate EC values were
50
ACS Paragon Plus Environment

Page 11 of 67
Journal of Medicinal Chemistry
1
2
3
calculated (Table 2). Subsequently the accurate EC value of the most potent analogue 10a was
50
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 from two independent experiments (eight point data) and listed in Table 2 along with
data for the related compounds 2b and 2c for comparison. We observe that all tested analogues are
full agonists at the CB1 receptor while their EC values correlate well with their respective binding
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
8
affinities. Of the compounds tested the carboxyester ꢁ ꢀTHC analogues 2b and 2c as well as their
ωꢀsubstituted cyanide congener 10a were found to be the most potent and efficacious compounds
within the series with EC s of 0.5, 0.4 and 0.9 nM, respectively.
5
0
Molecular modeling
As an aid in the interpretation of the data, we carried out conformational analyses and docking
studies (detailed procedures are given under Experimental) involving our lead analogue 10a, the
8
8
nonhydrolyzable counterpart (ꢀ)ꢀꢁ ꢀTHCꢀDMH as well as the parent compound (ꢀ)ꢀꢁ ꢀTHC.
8
8
Conformational analysis. Initial conformational analyses of (ꢀ)ꢀꢁ ꢀTHC, (ꢀ)ꢀꢁ ꢀTHCꢀDMH and 10a
revealed a difference in the conformation of the side chains in their global minimum energy
8
conformers. Both (ꢀ)ꢀꢁ ꢀTHCꢀDMH and 10a have their side chains oriented orthogonal to the
8
phenol ring, while for (ꢀ)ꢀꢁ ꢀTHC, the five carbon side chain extends in the same plane as its
phenol ring. Initial in vacuo conformational analyses of 10a identified an intramolecular hydrogen
bond between the cyano nitrogen and the phenolic hydroxyl. The 10a phenolic hydroxyl to cyano
nitrogen hydrogen bond heteroatom distance (N−O) and hydrogen bond (O−H− −N) angle were
3
.22 Å and 144°, respectively. The intramolecular hydrogen bond in 10a was not found in the
global minimum energy conformation from a second conformational search performed using the
OPSL2005 forcefield with an implicit GB/SA solvent model for water. Preliminary results from a
7
0 ns NAMD molecular dynamics simulation of 10a in a fully hydrated POPC lipid bilayer found a
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 67
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
very low incidence of this internal hydrogen bond as well (unpublished data). For this reason, the
nearest low energy conformer without an intramolecular hydrogen bond (which was 1.77 kcal/mol
above the initial global min) was used for the calculation of conformational energy costs for 10a.
The first and second side chain dihedrals of this conformer compare well with the global minimum
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
energy conformer of (ꢀ)ꢀꢁ ꢀTHCꢀDMH, shown in Table 3. Finally, the phenolic hydroxyls of all
three ligands were found to prefer the proton directed toward the C2 phenyl ring position nearer the
side chain and with the value of the C2ꢀC1ꢀO1ꢀH1 dihedral near zero. A comparison of the lowest
8
energy conformation of 10a without the internal hydrogen bond to those of (ꢀ)ꢀꢁ ꢀTHCꢀDMH and
8
(
ꢀ)ꢀꢁ ꢀTHC is shown in Figure 3.
Docking studies. To provide a representation for the interaction of the above three compounds with
the hCB1 receptor we carried out docking studies with an activated form of a CB1 receptor model
and calculated relative ligand/receptor interaction energies. Glide docking studies of 10a in the
activated CB1 receptor revealed a significant role for the ester group at the 2' position. The carbonyl
oxygen within the ester group has a hydrogen bonding interaction with T3.33(197) as shown in
Figure 4A. The hydrogen bond heteroatom distance (O−O) and hydrogen bond (O−H− −O) angle
are 2.56 Å and 165°, respectively. The T3.33(197) hydrogen bonding interaction is not available to
8
8
the (ꢀ)ꢀꢁ ꢀTHCꢀDMH counterpart or the parent compound (ꢀ)ꢀꢁ ꢀTHC as shown in Figures 4B, 4C.
However, all three compounds modelled in the Glide docking studies have a phenolic hydroxyl that
interacts directly with K3.28(192). The K3.28(192) and 10a phenolic hydroxyl hydrogen bond
heteroatom distance (N−O) and hydrogen bond (N−H− −O) angle are 2.73 Å and 170°,
8
respectively. The K3.28(192) and (ꢀ)ꢀꢁ ꢀTHCꢀDMH phenolic hydroxyl hydrogen bond heteroatom
distance (N−O) and hydrogen bond (N−H− −O) angle are 2.69 Å and 171°, respectively. Finally,
ACS Paragon Plus Environment

Page 13 of 67
Journal of Medicinal Chemistry
1
2
3
8
the K3.28(192) and (ꢀ)ꢀꢁ ꢀTHC phenolic hydroxyl hydrogen bond heteroatom distance (N−O) and
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
hydrogen bond (N−H− −O) angle are 2.64 Å and 168°, respectively.
Glide XP scores adjusted for ligand strain reflect the rank order of Ki’s (Table 3) in rCB1 with (10a
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
8
=
ꢀ5.7 kcal/mol) ≈ ((ꢀ)ꢀꢁ ꢀTHCꢀDMH = ꢀ5.4 kcal/mol) < ((ꢀ)ꢀꢁ ꢀTHC = ꢀ4.1 kcal/mol). For a list of
Glide XP scores in rank order and adjusted by ligand conformational cost, see Supplemental Tables
S2ꢀ4. A breakdown of ligand/receptor interaction energies revealed the importance of van der
8
Waals interactions for each of ligands and how the two 1',1'ꢀdimethyl analogs (10a and (ꢀ)ꢀꢁ ꢀTHCꢀ
8
8
DMH) have far more of these interactions than (ꢀ)ꢀꢁ ꢀTHC. The (ꢀ)ꢀꢁ ꢀTHCꢀDMH and 10a have
8
nine and eight van der Waals interactions better than ꢀ2.0 kcal/mol, respectively, while (ꢀ)ꢀꢁ ꢀTHC
has five such interactions. The most important van der Waals interactions in common for the 1',1'ꢀ
dimethyl analogs are F3.25(189), L3.29(193), F(268), M6.55(363), M(371), S7.39(383) and
8
C7.42(386). The (ꢀ)ꢀꢁ ꢀTHC parent interacts in a likewise manner with only F3.25(189),
L3.29(193), F(268), M(371) and S7.39(383). Of these residues, L3.29(193), F(268), M6.55(363)
and M(371) form part of a tunnel shaped hydrophobic enclosure between TMHs 3 and 6 for the
side chain of each analog to occupy. Unique to 10a is the van der Waals interaction with
8
V3.32(196), and unique to (ꢀ)ꢀꢁ ꢀTHCꢀDMH is F7.35(329). A full listing of all ligand/receptor
interaction energies is given under Supporting information (Tables 5ꢀ7).
Total ligand/receptor interaction energies also follow the trend of the Glide XP scores and Ki’s in
8
8
rCB1 with (10a = ꢀ42.9 kcal/mol) ≈ ((ꢀ)ꢀꢁ ꢀTHCꢀDMH = ꢀ44.4 kcal/mol) < ((ꢀ)ꢀꢁ ꢀTHC = ꢀ35.6
kcal/mol). The docked conformational cost for each ligand after the post Glide minimization (10a,
8
8
0
.18 kcal/mol; (ꢀ)ꢀꢁ ꢀTHCꢀDMH, 0.28 kcal/mol; and (ꢀ)ꢀꢁ ꢀTHC, 0.13 kcal/mol) was small.
8
Compound 10a did not have a statistically significant lower Ki compared to (ꢀ)ꢀꢁ ꢀTHCꢀDMH,
even though Glide docking studies identified two hydrogen bonds for 10a compared to one
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 67
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
hydrogen bond for (ꢀ)ꢀꢁ ꢀTHCꢀDMH. This may be due to the energetic cost for 10a to place its
hydrophilic ester and cyano moieties in a generally hydrophobic pocket in the CB1 receptor. This
cost likely causes the two Kis to be essentially equivalent.
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
In vivo behavioral characterization
Representative analogues within this series were initially screened using the hypothermia test in rats
(
data not shown) and the order of the potency was found to be 10a > 7b ≥ 6b > 2b ≈ 20 ≥ 16 ≈ 8b ≥
18. Subsequently, the most promising compound 10a was studied in more detail and its in vivo
hypothermic and antinociceptive profiles were compared to those of the earlier analog 2b as well as
8
with the nonꢀhydrolyzable parent compound (ꢀ)ꢀꢁ ꢀTHCꢀDMH.
Hypothermia testing. Body temperature was measured in isolated rats over a 6 h period following
drug injection (detailed procedures are given in the Experimental section). Compound 10a
decreased core body temperature in a doseꢀdependent manner, with a dose of 0.1 mg/kg reducing
o
o
body temperature up to 4.5 ± 0.5 C from an average baseline of 37.8 ± 0.1 C (Figure 5). Its ED
5
0
o
value (i.e., the dose required to reduce temperature by 3 C) was 0.034 mg/Kg (95% CI: 0.026,
0.041 mg/Kg). For comparison, the effects of our first generation analogue 2b and those of the (ꢀ)ꢀ
8
ꢁ
ꢀTHCꢀDMH are also shown. The ED values were 0.37 (0.14, 0.73 with 95% CI) for 2b and
50
8
0
.29 (0.18, 0.45 with 95% CI) for (ꢀ)ꢀꢁ ꢀTHCꢀDMH. Thus, 10a is approximately 8ꢀ to 10ꢀfold more
8
potent than (ꢀ)ꢀꢁ ꢀTHCꢀDMH and 2b. Limited extended studies comparing hypothermia induced by
8
equipotent doses of the 6'ꢀcyanoꢀ2'ꢀcarboxyꢀꢁ ꢀTHC analogue 10a and its structurally related
8
analogs 2b and (ꢀ)ꢀꢁ ꢀTHCꢀDMH for 12 hours provided an estimate of their in vivo times of action.
Thus, at a dose of 0.03 mg/kg, near its ED value, compound 10a reduced body temperature by 2
5
0
o
C within 3 h of injection and these effects were not observed 6 h after injection (see supporting
ACS Paragon Plus Environment

Page 15 of 67
Journal of Medicinal Chemistry
1
2
3
information Figure S1). This was compared with the effects of equivalent doses (near their ED50
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
values) of 2b and (ꢀ)ꢀꢁ ꢀTHCꢀDMH. Thus, for 2b, a dose of 0.3 mg/kg produced effects that
8
persisted for at least 6 h after injection while a dose of 0.3 mg/kg of (ꢀ)ꢀꢁ ꢀTHCꢀDMH produced an
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
equivalent reduction in temperature beyond 12 h after injection. It should be noted that in order to
compare equiactive doses of the most potent compound 10a with those of the in vivo less potent (ꢀ)ꢀ
8
ꢁ
ꢀTHCꢀDMH and 2b we have used different doses of the three drugs that are consistent with their
ED values. Our data indicate that the depot effect as reflected in the log P and PSA values (Table
5
0
4) plays the most dominant role in determining the in vivo hypothermic half lives of compounds
8
1
0a, 2b and, the nonꢀhydrolyzable (ꢀ)ꢀꢁ ꢀTHCꢀDMH. Thus, notwithstanding the shorter in vitro
hydrolytic half life of compound 2b, when compared to 10a, this compound has a longer in vivo
half life. These results are represented in Table 4.
Analgesia testing. The pharmacokinetic profiles of analogs bearing the carboxyester (2b) and the
more polar cyanoꢀcarboxyester (10a) side chains were further studied in the CB1 receptorꢀ
characteristic analgesia test in mice and the results are depicted in Figure 6 (the dose response graph
for four doses of 10a are given under supporting information, Figure S2). A mixed model repeated
®
measures ANOVA (IBM software package SPSS, v.21) applied to the tailꢀflick latency data
produced by compounds 10a and 2b during the descending phase (180 and 360 min) showed
significant main effects for drug (D) [F1,44 = 8.98; p ≤ 0.004], doseꢀlevel (L) [F1,44 = 40.95; p ≤
0.001], and time (T) [F1,44 = 28.46; p ≤ 0.001]. Of more interest, however, is that pairꢀwise
comparisons within these parameters (D, L, and T) using Sidak multiple comparison tꢀtest
procedure suggested significant differences for all three parameters (p = 0.05). Thus, the tailꢀflick
latencies were doseꢀ and time dependent for both compounds in a similar manner and since the pairꢀ
wise comparison for drug was also significant, the offꢀset of the analgesia effect was significantly
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 67
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
faster for 10a compared to 2b. The average (± sem) baseꢀline tailꢀflick withdrawal latency for all
mice (N = 24) examined with compound 10a was 0.99 ± 0.05 seconds.
Overall, our in vivo experiments show that in rats, compound 10a is approximately a 10ꢀ to 30ꢀfold
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
more potent as a cannabinoid agonist when compared to 2b and (ꢀ)ꢀꢁ ꢀTHCꢀDMH. Also, this more
potent analogue has a faster onset and shorter duration of action when compared to the less polar
8
counterpart 2b and the lipophilic and nonꢀhydrolizable parent compound (ꢀ)ꢀꢁ ꢀTHCꢀDMH. Our
antinociception data in mice clearly show that the cyanoꢀcarboxylꢀanalog 10a has a shorter duration
of action than its congener with no cyano substitution (2b).
Conclusions
In summary, as a continuation of our earlier work on the controlled deactivation/detoxification
ligand development project, we sought to probe the novel carboxyester side chain pharmacophore
8
in (ꢀ)ꢀꢁ ꢀTHCs for CB receptor binding affinity, in vitro and in vivo potency and efficacy as well as
its effects on the halfꢀlives of deactivation through enzymatic activity. We have also explored the
chain’s polar characteristics that are associated with the depot effect, in an effort to produce
cannabinoids with faster onset/offset and shorter duration of action than the currently existing (ꢀ)ꢀ
8
ꢁ
ꢀTHC analogs. In connection with our earlier work where we focused on the chain’s benzylic
1
8,20,29ꢀ33
position and the related subsite within the receptor’s binding domain,
the current SAR study
extends the mapping of the chain's pharmacophoric space beyond the 1'ꢀcarbon and argues that both
CB receptors can tolerate polar groups and atoms within the second and the fourth position of the
chain. We are hopeful that these observations will provide us with new opportunities for the design
of high affinity/efficacy novel analogs with improved selectivities for the two cannabinoid
receptors. The most successful compound identified through this careful study, namely 2ꢀ
ACS Paragon Plus Environment

Page 17 of 67
Journal of Medicinal Chemistry
1
2
3
4
[
(6aR,10aR)ꢀ6a,7,10,10aꢀtetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]ꢀ2ꢀ
5
methylꢀpropanoic acid 3ꢀcyanoꢀpropyl ester (10a), is a remarkably potent and efficacious CB1
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
receptor agonist with relatively shorter duration of action than its less polar 2'ꢀcarboxyꢀ(ꢀ)ꢀꢁ ꢀTHCꢀ
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
DMH counterpart and the highly lipophilic and long lasting (ꢀ)ꢀꢁ ꢀTHCꢀDMH. Our data support the
hypothesis that the pharmacological halfꢀlives of our selectively detoxified analogs can be
controlled by the joint modulation of their relative stabilities for plasma esterases as well as through
variation of their polar characteristics and thus to the depot effects.
Experimental section
Materials. All reagents and solvents were purchased from Aldrich Chemical Company, unless
otherwise specified, and used without further purification. All anhydrous reactions were performed
under a static argon atmosphere in flameꢀdried glassware using scrupulously dry solvents. Flash
column chromatography employed silica gel 60 (230ꢀ400 mesh). All compounds were
demonstrated to be homogeneous by analytical TLC on preꢀcoated silica gel TLC plates (Merck, 60
F245 on glass, layer thickness 250 ꢀm), and chromatograms were visualized by phosphomolybdic
acid staining. Melting points were determined on a microꢀmelting point apparatus and are
uncorrected. IR spectra were recorded on a Perkin Elmer Spectrum One FTꢀIR spectrometer. NMR
1
spectra were recorded in CDCl , unless otherwise stated, on a Bruker Ultra Shield 400 WB plus ( H
3
1
3
1
13
at 400 MHz, C at 100 MHz) or on a Varian INOVAꢀ500 ( H at 500 MHz, C at 125 MHz)
spectrometers and chemical shifts are reported in units of relative to internal TMS. Multiplicities
δ
are indicated as br (broadened), s (singlet), d (doublet), t (triplet), q (quartet), m (multiplet) and
coupling constants (J) are reported in hertz (Hz). Low and highꢀresolution mass spectra were
performed in School of Chemical Sciences, University of Illinois at UrbanaꢀChampaign. Mass
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 67
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
spectral data are reported in the form of m/z (intensity relative to base = 100). Elemental analyses
were obtained in Baron Consulting Co, Milford, CT, and were within ± 0.4% of the theoretical
values (see supporting information). Purities of the tested compounds were determined by
elemental analysis or by HPLC (using Waters Alliance HPLC system, 4.6 x 250 mm, Supelco
discovery column, acetonitrile/water with 8.5% oꢀphosphoric acid) or by LC/MS analysis using a
Waters MicroMass ZQ system [electrosprayꢀionization (ESI) with Watersꢀ2525 binary gradient
module coupled to a Photodiode Array Detector (Watersꢀ2996) and ELS detector (Watersꢀ2424)
using a XTerra MS C18, 5 µm, 4.6 mm x 50 mm column and acetonitrile/water] and were > 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
2
ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]
acetic acid 4ꢀbromoꢀbutyl ester (6a). A stirred mixture of 5a (175 mg, 0.58 mmol),
dibromobutane (313 mg, 1.45 mmol) and sodium bicarbonate (73 mg, 0.87 mmol) in DMF (2 mL)
was heated at 165°C for 12 min using microwave irradiation. The reaction mixture was cooled to
room temperature and diluted with water and ethyl acetate. The organic layer was separated and the
aqueous phase was extraced with ethyl acetate. The combined organic layer was washed with brine,
dried (MgSO ) and concentrated under reduced pressure. Purification by flash column
4
chromatography on silica gel gave 6a (159 mg, 63% yield) as a light yellow gum. IR (CHCl3):
ꢀ1 1
3
407, 2967, 2928, 1712 (s, >C=O), 1621, 1583, 1431, 1254 cm ; H NMR (500 MHz, CDCl ) δ
3
6.33 (d, J = 1.0 Hz, 1H, 4ꢀH), 6.24 (d, J = 1.0 Hz, 1H, 2ꢀH), 5.42 (m as d, J = 4.0 Hz, 1H, 8ꢀH),
5
2
1
.23 (s, 1H, OH), 4.12 (t, J = 6.5 Hz, 2H, ꢀOCH ꢀ), 3.45 (s, 2H, ꢀCH ꢀC(O)Oꢀ), 3.39 (t, J = 6.5 Hz,
2 2
H, ꢀCH Br), 3.19 (dd, J = 15.0 Hz, J = 4.5 Hz, 1H, 10αꢀH), 2.69 (td, J = 11.0 Hz, J = 4.5 Hz, 1H,
2
0aꢀH), 2.19ꢀ2.10 (m, 1H, 7αꢀH), 1.94ꢀ1.76 (m, 7H, 10βꢀH, 7βꢀH, 6aꢀH, ꢀCH ꢀCH ꢀCH ꢀBr, ꢀCH ꢀ
2
2
2
2
CH ꢀCH ꢀBr), 1.69 (s, 3H, 9ꢀCH ), 1.37 (s, 3H, 6βꢀCH ), 1.09 (s, 3H, 6αꢀCH ); mass spectrum (EI)
2
2
3
3
3
ACS Paragon Plus Environment

Page 19 of 67
Journal of Medicinal Chemistry
1
2
3
+
+
m/z (relative intensity) 438 (M + 2, 97), 436 (M , 97), 395 (38), 393 (38), 355 (95), 353 (95), 330
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
(
60), 316 (38), 287 (46), 273 (48), 257 (45), 247 (98), 233 (60), 213 (100); Exact mass (EI)
+
calculated for C H O Br (M ), 436.1249; found 436.1252. HPLC (4.6 x 250 mm, Supelco
2
2
29
4
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
discovery column, acetonitrile/water) showed purity 97.5% and retention time 12.6 min for the title
compound. LC/MS analysis (Waters MicroMass ZQ system) showed purity 97% and retention time
7
.3 min for the title compound.
2ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]ꢀ
ꢀmethylꢀpropanoic acid 4ꢀbromoꢀbutyl ester (6b). The synthesis was carried out as described
2
for 6a using 5b (200 mg, 0.61 mmol), dibromobutane (330 mg, 1.53 mmol), and sodium
bicarbonate (77 mg, 0.92 mmol) in DMF and gave 6b (148 mg, 53% yield) as a light yellow gum.
ꢀ1
1
IR (neat): 3403, 2970, 2917, 1727 and 1703 (s, >C=O), 1620, 1577, 1416, 1256 cm ; H NMR (500
MHz, CDCl ) δ 6.41 (d, J = 2.0 Hz, 1H, 4ꢀH), 6.25 (d, J = 2.0 Hz, 1H, 2ꢀH), 5.42 (m as d, J = 5.0
3
Hz, 1H, 8ꢀH), 5.18 (s, 1H, OH), 4.09 (t, J = 6.5 Hz, 2H, ꢀOCH ꢀ), 3.32 (t, J = 6.5 Hz, 2H, ꢀCH Br),
2
2
3
.19 (dd, J = 15.0 Hz, J = 4.5 Hz, 1H, 10αꢀH), 2.69 (td, J = 11.0 Hz, J = 4.5 Hz, 1H, 10aꢀH), 2.18ꢀ
2.10 (m, 1H, 7αꢀH), 1.85ꢀ1.68 (m and s overlapping, 10H, 10βꢀH, 7βꢀH, 6aꢀH, ꢀCH ꢀCH ꢀ of the
2
2
side chain and 9ꢀCH , especially 1.70 s, 9ꢀCH ), 1.51 (s, 6H, ꢀC(CH ) ꢀ), 1.38 (s, 3H, 6βꢀCH ), 1.10
3
3
3 2
3
1
3
(s, 3H, 6αꢀCH ); C NMR (125 MHz, CDCl ) δ 177.8 (>C=O), 155.7 (Cꢀ1 or Cꢀ5), 155.1 (Cꢀ5 or
3
3
Cꢀ1), 144.0, 134.9, 119.4, 112.0, 107.1, 105.3, 77.0 (Cꢀ6), 64.3 (ꢀOCH ꢀ), 46.3, 45.0, 35.9, 33.3,
2
3
1.7, 29.3, 28.0, 27.7, 27.2, 26.2, 26.2, 23.8, 18.7; mass spectrum (ESI) m/z (relative intensity) 467
+
+
(M + 2 + H, 100), 465 (M + H, 100), 285 (30); Exact mass (ESI) calculated for C H O Br
24 34 4
+
(M +H), 465.1640; found 465.1647. LC/MS analysis (Waters MicroMass ZQ system) showed
purity 98% and retention time 7.4 min for the title compound. Anal. (C H BrO ) C, H.
2
4
33
4
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 67
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
1
ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]ꢀ
cyclobutanecarboxylic acid 4ꢀbromoꢀbutyl ester (6c). The synthesis was carried out as described
for 6a using 5c (145.0 mg, 0.42 mmol), dibromobutane (136 mg, 0.63 mmol), and sodium
bicarbonate (39 mg, 0.46 mmol) in DMF and gave 6c (90 mg, 45% yield) as a light yellow gum. IR
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 1
(neat): 3407, 2968, 2915, 1728 and 1705 (s, >C=O), 1622, 1416, 1258 cm ; H NMR (500 MHz,
CDCl ) δ 6.39 (d, J = 2.0 Hz, 1H, 4ꢀH), 6.24 (d, J = 2.0 Hz, 1H, 2ꢀH), 5.85 (s, 1H, ꢀOH), 5.42 (m
3
as d, J = 4.5 Hz, 1H, 8ꢀH), 4.10 (m, 2H, ꢀOCH ꢀ), 3.31 (t, J = 6.5 Hz, 2H, ꢀCH Br), 3.24 (dd, J =
2
2
1
6.0 Hz, J = 4.0 Hz, 1H, 10αꢀH), 2.79ꢀ2.66 (m, 3H, 10aꢀH, 2H of the cyclobutane ring,
overlapping), 2.46 (m as qt, J = 9.0 Hz, 2H of the cyclobutane ring), 2.19ꢀ2.10 ( m, 1H, 7αꢀH),
1.99ꢀ1.90 (m, 1H of the cyclobutane ring), 1.89ꢀ1.64 (m and s overlapping, 11H, 10βꢀH, 7βꢀH, 6aꢀ
H, 1H of the cyclobutane ring, ꢀCH CH CH Br, 9ꢀCH , especially 1.69 s, 9ꢀCH ), 1.38 (s, 3H, 6βꢀ
2
2
2
3
3
+
CH ), 1.09 (s, 3H, 6αꢀCH ); mass spectrum (ESI) m/z (relative intensity) 501 (M + 2 + Na, 33), 499
3
3
+
+
+
(M + Na, 33), 479 (M + 2 + H, 100), 477 (M + H, 100), 297 (30); Exact mass (ESI) calculated for
+
C H O Br (M +H), 477.1640; found 477.1636. LC/MS analysis (Waters MicroMass ZQ system)
25
34
4
showed purity 97% and retention time 5.6 min for the title compound.
2
ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]
acetic acid 4ꢀcyanoꢀbutyl ester (7a). A solution of bromide 6a (80 mg, 0.18 mmol) and sodium
o
cyanide (88 mg, 1.8 mmol) in anhydrous DMSO (5mL) was stirred at 50 C for 3 hours under argon.
The reaction mixture was cooled to room temperature and diluted with water and ethyl acetate. The
organic layer was separated and the aqueous layer was extracted with ethyl acetate. The combined
organic layer was washed with brine, dried (MgSO ), and evaporated under reduced pressure.
4
ACS Paragon Plus Environment

Page 21 of 67
Journal of Medicinal Chemistry
1
2
3
Purification by flash column chromatography on silica gel (20% ethyl acetate in hexane) afforded
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
1
6
a (50 mg, 71% yield) as a light yellow gum. IR (neat): 3405, 2926, 2249 (CN), 1732 (s, >C=O),
ꢀ1 1
621, 1583, 1431, 1262, 1182 cm ; H NMR (500 MHz, CDCl ) δ 6.31 (d, J = 2.0 Hz, 1H, 4ꢀH),
3
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
.27 (d, J = 2.0 Hz, 1H, 2ꢀH), 5.72 (s, 1H, OH), 5.42 (m as d, J = 4.0 Hz, 1H, 8ꢀH), 4.14 (t, J = 6.0
Hz, 2H, ꢀOCH ꢀ), 3.46 (s, 2H, ꢀCH ꢀC(O)Oꢀ), 3.22 (dd, J = 15.0 Hz, J = 4.5 Hz, 1H, 10αꢀH), 2.69
2
2
(td, J = 11.0 Hz, J = 4.5 Hz, 1H, 10aꢀH), 2.35 (t, J = 7.0 Hz, 2H, ꢀCH CN), 2.17ꢀ2.09 (m, 1H, 7αꢀ
2
H), 1.85ꢀ1.70 (m, 7H, 10βꢀH, 7βꢀH, 6aꢀH, ꢀCH ꢀCH ꢀ of the side chain), 1.68 (s, 3H, 9ꢀCH ), 1.37
2
2
3
+
(s, 3H, 6βꢀCH ), 1.09 (s, 3H, 6αꢀCH ); mass spectrum (EI) m/z (relative intensity) 383 (M , 32),
3
3
3
68 (10), 340 (12), 315 (8), 300 (40), 279 (13), 213 (38), 205 (29), 97 (38), 83 (41), 69 (71), 57
+
(100); Exact mass (EI) calculated for C H O N (M ), 383.2097; found 383.2092. LC/MS analysis
23 29
4
(Waters MicroMass ZQ system) showed purity 96.5% and retention time 6.8 min for the title
compound.
2
ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]ꢀ
ꢀmethylꢀpropanoic acid 4ꢀcyanoꢀbutyl ester (7b). The synthesis was carried out as described for
2
7a using 6b (20 mg, 0.04 mmol) and sodium cyanide (21 mg, 0.43 mmol) in anhydrous DMSO and
gave 7b (9.5 mg, 54% yield) as light yellow gum. IR (CHCl ): 3405, 2973, 2249 (CN), 1727 and
3
ꢀ1 1
1
705 (s, >C=O), 1620, 1578, 1417, 1258, 1184 cm ; H NMR (500 MHz, CDCl ) δ 6.41 (d, J = 2.0
3
Hz, 1H, 4ꢀH), 6.26 (d, J = 2.0 Hz, 1H, 2ꢀH), 5.42 (m as d, J = 5.0 Hz, 1H, 8ꢀH), 5.26 (s, 1H, OH),
.11 (t, J = 6.5 Hz, 2H, ꢀOCH ꢀ), 3.20 (dd, J = 15.0 Hz, J = 4.5 Hz, 1H, 10αꢀH), 2.69 (td, J = 11.0
4
2
Hz, J = 4.5 Hz, 1H, 10aꢀH), 2.28 (t, J = 6.5 Hz, 2H, ꢀCH CN), 2.18ꢀ2.10 (m, 1H, 7αꢀH), 1.87ꢀ1.72
2
(m, 5H, 10βꢀH, 7βꢀH, 6aꢀH, ꢀCH ꢀ of the side chain), 1.70 (s, 3H, 9ꢀCH ), 1.68ꢀ1.58 (m, 2H, ꢀCH ꢀ
2 3 2
of the side chain), 1.51 (s, 3H, ꢀC(CH ) ꢀ), 1.50 (s, 3H, ꢀC(CH ) ꢀ),1.38 (s, 3H, 6βꢀCH ), 1.10 (s,
3
2
3 2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 67
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
+
+
3
H, 6αꢀCH ); mass spectrum (ESI) m/z (relative intensity) 434 (M + Na, 10), 412 (M + H, 100),
3
+
2
85 (5); Exact mass (ESI) calculated for C H NO (M +H), 412.2488; found 412.2480. LC/MS
25 34
4
analysis (Waters MicroMass ZQ system) showed purity 98% and retention time 6.8 min for the title
compound.
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
ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]ꢀ
cyclobutanecarboxylic acid 4ꢀcyanoꢀbutyl ester (7c). The synthesis was carried out as described
for 7a using 6c (180 mg, 0.38 mmol) and sodium cyanide (186 mg, 3.8 mmol) in anhydrous DMSO
and gave 7c (100 mg, 63% yield) as light yellow gum. IR (CHCl ): 3400, 2970, 2250 (CN), 1723
3
ꢀ1 1
and 1703 (s, >C=O), 1619, 1577, 1417, 1268, 1183 cm ; H NMR (500 MHz, CDCl ) δ 6.37 (d, J
3
= 1.5 Hz, 1H, 4ꢀH), 6.26 (d, J = 1.5 Hz, 1H, 2ꢀH), 5.95 (s, 1H, ꢀOH), 5.42 (m as d, J = 4.0 Hz, 1H,
8
3
ꢀH), 4.11 (t, J = 5.5 Hz, 2H, ꢀOCH ꢀ), 3.26 (dd, J = 16.5 Hz, J = 5.0 Hz, 1H, 10αꢀH), 2.79ꢀ2.67 (m,
2
H, 10aꢀH, 2H of the cyclobutane ring, overlapping), 2.46 (m, 2H of the cyclobutane ring), 2.26 (t,
J = 7.0 Hz, 2H, ꢀCH CN), 2.18ꢀ2.10 ( m, 1H, 7αꢀH), 2.00ꢀ1.90 (m, 1H of the cyclobutane ring),
2
1
.89ꢀ1.71 (m, 6H, 10βꢀH, 7βꢀH, 6aꢀH, 1H of the cyclobutane ring, ꢀCH ꢀ of the side chain), 1.69 (s,
2
3H, 9ꢀCH ), 1.63ꢀ1.54 (m, 2H, ꢀCH ꢀ of the side chain), 1.38 (s, 3H, 6βꢀCH ), 1.09 (s, 3H, 6αꢀCH );
3
2
3
3
+
+
mass spectrum (ESI) m/z (relative intensity) 446 (M + Na, 50), 424 (M + H, 100), 297 (5); Exact
+
mass (ESI) calculated for C H NO (M +H), 424.2488; found 424.2491. LC/MS analysis (Waters
26
34
4
MicroMass ZQ system) showed purity 98% and retention time 4.9 min for the title compound.
Anal. (C H NO ) C, H.
2
6
33
4
2
ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]ꢀ
ꢀmethylꢀpropanoic acid methyl ester (8a). The synthesis was carried out as described for 6a
2
ACS Paragon Plus Environment

Page 23 of 67
Journal of Medicinal Chemistry
1
2
3
using 5b (240 mg, 0.73 mmol), iodomethane (260 mg, 1.83 mmol), and sodium bicarbonate (92 mg,
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
1
2
6
.09 mmol) in DMF and gave 8a (160 mg, 64% yield) as a light yellow gum. IR (neat): 3412, 2962,
ꢀ1 1
932, 1728 and 1702 (s, >C=O), cm ; H NMR (500 MHz, CDCl ) δ 6.41 (d, J = 2.0 Hz, 1H, 4ꢀH),
3
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
.26 (d, J = 2.0 Hz, 1H, 2ꢀH), 5.75 (s, 1H, OH), 5.42 (m as d, J = 5.0 Hz, 1H, 8ꢀH), 3.66 (s, 3H, ꢀ
OCH ), 3.22 (dd, J = 15.0 Hz, J = 4.5 Hz, 1H, 10αꢀH), 2.69 (td, J = 11.0 Hz, J = 4.5 Hz, 1H, 10aꢀ
3
H), 2.19ꢀ2.10 (m, 1H, 7αꢀH), 1.86ꢀ1.74 (m, 3H, 10βꢀH, 7βꢀH, 6aꢀH), 1.69 (s, 3H, 9ꢀCH ), 1.51 (s,
3
1
3
6
H, ꢀC(CH ) ꢀ), 1.38 (s, 3H, 6βꢀCH ), 1.11 (s, 3H, 6αꢀCH ); C NMR (125 MHz, CDCl ) δ 177.8
3 2 3 3 3
(>C=O), 155.4 (Cꢀ1 or Cꢀ5), 155.1 (Cꢀ5 or Cꢀ1), 144.4, 135.0, 119.4, 111.9, 107.5, 105.4, 77.1 (Cꢀ
6
), 52.6 (ꢀOCH ), 46.3, 44.9, 35.9, 31.7, 28.1, 27.8, 26.5, 26.4, 23.8, 18.8; mass spectrum (EI) m/z
3
+
(relative intensity) 344 (M , 91), 329 (15), 301 (38), 285 (22), 276 (18), 261 (100), 241 (22), 223
+
(19); Exact mass (EI) calculated for C H O (M ), 344.1988; found 344.1992. HPLC (4.6 x 250
2
1
28
4
mm, Supelco discovery column, acetonitrile/water) showed purity 96.5% and retention time 12 min
for the title compound. LC/MS analysis (Waters MicroMass ZQ system) showed purity 96.5% and
retention time 6.6 min for the title compound.
2ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]ꢀ
2
ꢀmethylꢀpropanoic acid ethyl ester (8b). The synthesis was carried out as described for 6a using
b (300 mg, 0.91 mmol), iodoethane (356 mg, 2.28 mmol), and sodium bicarbonate (115 mg, 1.37
5
mmol) in DMF and gave 8b (283 mg, 87% yield) as a light yellow gum. IR (neat): 3410, 2959,
ꢀ1
1
2
6
7
931, 1728 and 1702 (s, >C=O), cm ; H NMR (500 MHz, CDCl ) δ 6.41 (d, J = 1.0 Hz, 1H, 4ꢀH),
3
.28 (d, J = 1.0 Hz, 1H, 2ꢀH), 5.74 (s, 1H, OH), 5.42 (m as d, J = 5.0 Hz, 1H, 8ꢀH), 4.12 (q, , J =
.0 Hz, 2H, ꢀOCH ꢀ), 3.22 (dd, J = 15.0 Hz, J = 4.5 Hz, 1H, 10αꢀH), 2.69 (td, J = 11.0 Hz, J = 4.5
2
Hz, 1H, 10aꢀH), 2.19ꢀ2.10 (m, 1H, 7αꢀH), 1.88ꢀ1.75 (m, 3H, 10βꢀH, 7βꢀH, 6aꢀH), 1.69 (s, 3H, 9ꢀ
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 67
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
CH ), 1.50 (s, 6H, ꢀC(CH ) ꢀ), 1.38 (s, 3H, 6βꢀCH ), 1.19 (t, J = 7.0 Hz, 3H, ꢀOCH CH ), 1.10 (s,
3
3 2
3
2
3
+
3
H, 6αꢀCH ); mass spectrum (EI) m/z (relative intensity) 358 (M , 92), 343 (17), 315 (39), 285 (48),
3
+
2
75 (100), 241 (35); Exact mass (EI) calculated for C H O (M ), 358.2144; found 358.2145.
2
2
30
4
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
LC/MS analysis (Waters MicroMass ZQ system) showed purity 97% and retention time 6.9 min for
the title compound.
2
ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]ꢀ
2ꢀmethylꢀpropanoic acid 2ꢀbromoꢀethyl ester (9a). The synthesis was carried out as described for
a using 5b (200 mg, 0.61 mmol), 1,2ꢀdibromoethane (287 mg, 1.53 mmol), and sodium
bicarbonate (77 mg, 0.92 mmol) in DMF and gave 9a (127 mg, 48% yield) as a light yellow gum.
6
ꢀ1 1
IR (neat): 3405, 2972, 2918, 1725 and 1701 (s, >C=O), 1621, 1577, 1415, 1262 cm ; H NMR (500
MHz, CDCl ) δ 6.42 (d, J = 1.5 Hz, 1H, 4ꢀH), 6.27 (d, J = 1.5 Hz, 1H, 2ꢀH), 5.42 (m as d, J = 4.0
3
Hz, 1H, 8ꢀH), 5.04 (br s, 1H, OH), 4.36 (m as td, J = 5.5 Hz, J = 2.0 Hz, 2H, ꢀOCH ꢀ), 3.46 (t, J =
2
6
.0 Hz, 2H, ꢀCH Br), 3.19 (dd, J = 15.0 Hz, J = 4.5 Hz, 1H, 10αꢀH), 2.69 (td, J = 11.0 Hz, J = 4.5
2
Hz, 1H, 10aꢀH), 2.18ꢀ2.10 (m, 1H, 7αꢀH), 1.89ꢀ1.72 (m, 3H, 10βꢀH, 7βꢀH, 6aꢀH), 1.70 (s, 3H, 9ꢀ
CH ), 1.53 (s, 6H, ꢀC(CH ) ꢀ), 1.38 (s, 3H, 6βꢀCH ), 1.10 (s, 3H, 6αꢀCH ); mass spectrum (EI) m/z
3
3 2
3
3
+
+
(
relative intensity) 438 (M + 2, 68), 436 (M , 68), 395 (22), 393 (22), 355 (67), 353 (67), 285 (44),
+
2
41 (45), 59 (100); Exact mass (EI) calculated for C H O Br (M ), 436.1249; found 436.1250.
2
2
29
4
LC/MS analysis (Waters MicroMass ZQ system) showed purity 98% and retention time 8.5 min for
the title compound.
2
ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]ꢀ
ꢀmethylꢀpropanoic acid 3ꢀbromoꢀpropyl ester (9b). The synthesis was carried out as described
2
ACS Paragon Plus Environment

Page 25 of 67
Journal of Medicinal Chemistry
1
2
3
for 6a using 5b (405 mg, 1.23 mmol), 1,3ꢀdibromopropane (622 mg, 3.08 mmol), and sodium
bicarbonate (155 mg, 1.85 mmol) in DMF and gave 9b (282 mg, 51% yield) as a light yellow gum.
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
ꢀ1 1
IR (neat): 3401, 2973, 2917, 1727 and 1703 (s, >C=O), 1621, 1576, 1416, 1260 cm ; H NMR (500
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
MHz, CDCl ) δ 6.41 (d, J = 1.5 Hz, 1H, 4ꢀH), 6.27 (d, J = 1.5 Hz, 1H, 2ꢀH), 5.89 (s, 1H, OH), 5.42
3
(m as d, J = 5.0 Hz, 1H, 8ꢀH), 4.19 (m as td, J = 6.5 Hz, J = 1.5 Hz, 2H, ꢀOCH ꢀ), 3.30ꢀ3.18 (t and
2
dd overlapping, 3H, ꢀCH Br, 10αꢀH, especially 3.24, t, J = 6.5 Hz, 2H, ꢀCH Br), 2.69 (td, J = 11.0
2
2
Hz, J = 4.5 Hz, 1H, 10aꢀH), 2.18ꢀ2.10 (m, 1H, 7αꢀH), 2.09 (qt, J = 6.5 Hz, 2H, ꢀCH CH Br), 1.87ꢀ
2
2
1.74 (m, 3H, 10βꢀH, 7βꢀH, 6aꢀH), 1.69 (s, 3H, 9ꢀCH ), 1.51 (s, 6H, ꢀC(CH ) ꢀ), 1.39 (s, 3H, 6βꢀ
3
3 2
+
CH ), 1.11 (s, 3H, 6αꢀCH ); mass spectrum (ESI) m/z (relative intensity) 453 (M + 2 + H, 55), 450
3
3
+
(
M + H, 55), 285 (100); LC/MS analysis (Waters MicroMass ZQ system) showed purity 96% and
retention time 7.6 min for the title compound. Anal. (C H BrO ) C, H.
2
3
31
4
2
ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]ꢀ
ꢀmethylꢀpropanoic acid 5ꢀbromoꢀpentyl ester (9c). The synthesis was carried out as described
2
for 6a using 5b (350 mg, 1.06 mmol), 1,5ꢀdibromopentane (609 mg, 2.65 mmol), and sodium
bicarbonate (134 mg, 1.59 mmol) in DMF and gave 9c (239 mg, 47% yield) as a light yellow gum.
ꢀ1 1
IR (CHCl ): 3408, 2973, 2931, 1728 and 1702 (s, >C=O), 1621, 1578, 1417, 1261 cm ; H NMR
3
(
500 MHz, CDCl ) δ 6.42 (d, J = 2.0 Hz, 1H, 4ꢀH), 6.25 (d, J = 2.0 Hz, 1H, 2ꢀH), 5.42 (m as d, J =
3
5.0 Hz, 1H, 8ꢀH), 5.07 (s, 1H, OH), 4.03 (t, J = 6.5 Hz, 2H, ꢀOCH ꢀ), 3.29 (t, J = 6.5 Hz, 2H, ꢀ
2
CH Br), 3.19 (dd, J = 15.0 Hz, J = 4.5 Hz, 1H, 10αꢀH), 2.69 (td, J = 11.0 Hz, J = 4.5 Hz, 1H, 10aꢀ
2
H), 2.18ꢀ2.10 (m, 1H, 7αꢀH), 1.84ꢀ1.67 (m and s overlapping, 10H, 10βꢀH, 7βꢀH, 6aꢀH, ꢀCH ꢀCH ꢀ
2
2
of the side chain and 9ꢀCH , especially 1.70 s, 9ꢀCH ), 1.51 (s, 6H, ꢀC(CH ) ꢀ), 1.38 (s, 3H, 6βꢀ
3
3
3 2
CH ), 1.28ꢀ1.23 (m, 2H, ꢀCH ꢀ of the side chain), 1.10 (s, 3H, 6αꢀCH ); mass spectrum (EI) m/z
3
2
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 67
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
+
+
+
(
relative intensity) 480 (M +2, 2), 478 (M , 2), 398 (M ꢀ Br, 82), 355 (19), 330 (21), 315 (100), 285
+
(
60), 241 (39); Exact mass (EI) calculated for C H O Br (M ), 478.1719; found 478.1715. LC/MS
25 35
4
analysis (Waters MicroMass ZQ system) showed purity 96% and retention time 7.5 min for the title
compound.
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
ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]ꢀ
ꢀmethylꢀpropanoic acid 4ꢀ(1Hꢀimidazolꢀ1ꢀyl)butyl ester (9d). The synthesis was carried out as
2
described for 6a using 5b (900 mg, 2.72 mmol), 4ꢀ(1Hꢀimidazolꢀ1ꢀyl)butyl bromide (1.12 g, 5.44
mmol), and sodium bicarbonate (343 mg, 4.08 mmol) in DMF and gave 9d (579 mg, 47% yield) as
o
a light yellow solid, m p 73ꢀ75 C. IR (CHCl ): 2971, 2932, 1727 (s, >C=O), 1618, 1578, 1512,
3
ꢀ1 1
1417, 1251 cm ; H NMR (500 MHz, CDCl ) δ 7.50 (s, 1H, imidazole), 7.10 (s, 1H, imidazole),
3
6
.86 (s, 1H, imidazole), 6.35 (d, J = 2.0 Hz, 1H, 4ꢀH), 6.23 (d, J = 2.0 Hz, 1H, 2ꢀH), 5.41 (m as d, J
=
=
5.0 Hz, 1H, 8ꢀH), 3.95 (m as td, J = 6.5 Hz, J = 2.0 Hz, 2H, ꢀOCH ꢀ), 3.88 (m as td, J = 7.5 Hz, J
2
3.0 Hz, 2H, ꢀCH ꢀN<), 3.37 (dd, J = 15.0 Hz, J = 4.5 Hz, 1H, 10αꢀH), 2.72 (td, J = 11.0 Hz, J =
2
4
.5 Hz, 1H, 10aꢀH), 2.18ꢀ2.09 (m, 1H, 7αꢀH), 1.84ꢀ1.68 (m and s overlapping, 10H, 10βꢀH, 7βꢀH,
6aꢀH, ꢀCH ꢀCH ꢀ of the side chain and 9ꢀCH , especially 1.70 s, 9ꢀCH ), 1.47 (s, 3H, ꢀC(CH ) ꢀ),
2
2
3
3
3 2
1
.46 (s, 3H, ꢀC(CH ) ꢀ), 1.38 (s, 3H, 6βꢀCH ), 1.09 (s, 3H, 6αꢀCH ); mass spectrum (EI) m/z
3 2 3 3
+
(
relative intensity) 452 (M , 100), 437 (20), 402 (48), 367 (56), 318 (23), 303 (38), 265 (31), 91
+
(18); Exact mass (EI) calculated for C H N O (M ), 452.2675; found 452.2666. Anal.
2
7
36
2
4
(C H N O ) C, H, N.
27 36 2 4
2
ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]ꢀ
ꢀmethylꢀpropanoic acid 3ꢀcyanoꢀpropyl ester (10a). The synthesis was carried out as described
2
ACS Paragon Plus Environment

Page 27 of 67
Journal of Medicinal Chemistry
1
2
3
for 7a using 9b (110 mg, 0.24 mmol) and sodium cyanide (118 mg, 2.4 mmol) in anhydrous DMSO
and gave 10a (95 mg, 98% yield) as light yellow gum. IR (CHCl ): 3403, 2971, 2248 (CN), 1727
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
3
ꢀ1 1
and 1704 (s, >C=O), 1621, 1576, 1421, 1261, 1185 cm ; H NMR (500 MHz, CDCl ) δ 6.40 (d, J =
3
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
.0 Hz, 1H, 4ꢀH), 6.29 (d, J = 2.0 Hz, 1H, 2ꢀH), 5.82 (s, 1H, OH), 5.42 (m as d, J = 4.0 Hz, 1H, 8ꢀ
H), 4.17 (m, AB system, 2H, ꢀOCH ꢀ), 3.22 (dd, J = 15.0 Hz, J = 4.5 Hz, 1H, 10αꢀH), 2.70 (td, J =
2
1
1.0 Hz, J = 4.5 Hz, 1H, 10aꢀH), 2.21 (t, J = 7.5 Hz, 2H, ꢀCH CN), 2.18ꢀ2.10 (m, 1H, 7αꢀH), 1.93
2
(
m as qt, J = 6.0 Hz, 2H, ꢀCH ꢀ of the side chain), 1.87ꢀ1.75 (m, 3H, 10βꢀH, 7βꢀH, 6aꢀH), 1.70 (s,
2
3H, 9ꢀCH ), 1.52 (s, 6H, ꢀC(CH ) ꢀ), 1.38 (s, 3H, 6βꢀCH ), 1.10 (s, 3H, 6αꢀCH ); mass spectrum
3
3 2
3
3
+
(
EI) m/z (relative intensity) 397 (M , 79), 382 (12), 354 (9), 329 (13), 314 (100), 285 (27), 276 (21),
+
2
41 (78), 149 (21), 70 (41); Exact mass (EI) calculated for C H NO (M ), 397.2253; found
24 31
4
397.2254. LC/MS analysis (Waters MicroMass ZQ system) showed purity 98% and retention time
1
1.4 min for the title compound. Anal. (C H NO ) C, H, N.
24 31 4
2
ꢀ[(6aR,10aR)ꢀ6a,7,10,10aꢀTetrahydroꢀ1ꢀhydroxyꢀ6,6,9ꢀtrimethylꢀ6Hꢀdibenzo[b,d]pyranꢀ3ꢀyl]ꢀ
ꢀmethylꢀpropanoic acid 3ꢀ(1Hꢀimidazolꢀ1ꢀyl)propyl ester (10b). To a stirred suspension of 9b
2
(196.0 mg, 0.43 mmol) and potassium carbonate (594 mg, 4.30 mmol) in DMSO (5 mL) was added
imidazole (146 mg, 2.15 mmol) at room temperature under an argon atmosphere. Stirring was
continued for 14 hours and then the mixture was diluted with water and ethyl acetate. The organic
layer was separated and the aqueous layer was extracted with ethyl acetate. The combined organic
layer was washed with water and brine, dried (MgSO ), and concentrated in vacuo. Purification by
4
flash column chromatography on silica gel (50% acetone in hexane) gave 10b (78 mg, 41% yield)
ꢀ1
as light yellow gum. IR (CHCl ): 2971, 2932, 1727 (s, >C=O), 1618, 1578, 1512, 1417, 1251 cm ;
3
1
H NMR (500 MHz, CDCl ) δ 7.33 (s, 1H, imidazole), 7.04 (s, 1H, imidazole), 6.44 (s, 1H,
3
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 67
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
imidazole), 6.40 (d, J = 2.0 Hz, 1H, 4ꢀH), 6.32 (d, J = 2.0 Hz, 1H, 2ꢀH), 5.40 (m as d, J = 5.0 Hz,
H, 8ꢀH), 3.93 (td, J = 17.0 Hz, J = 5.5 Hz, 1H, ꢀOCH ꢀ), 3.85 (td, J = 17.0 Hz, J = 5.5 Hz, 1H, ꢀ
1
2
OCH ꢀ), 3.72 (t, J = 7.0 Hz, 2H, ꢀCH ꢀN<), 3.41 (dd, J = 16.0 Hz, J = 4.5 Hz, 1H, 10αꢀH), 2.73 (td,
2
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
J = 11.0 Hz, J = 4.5 Hz, 1H, 10aꢀH), 2.17ꢀ2.09 (m, 1H, 7αꢀH), 2.01ꢀ1.93 (m, 2H, ꢀOCH ꢀCH ꢀ),
2
2
1.85ꢀ1.71 (m, 3H, 10βꢀH, 7βꢀH, 6aꢀH), 1.60 (s, 3H, 9ꢀCH ), 1.50 (s, 3H, ꢀC(CH ) ꢀ), 1.49 (s, 3H, ꢀ
3
3 2
1
3
C(CH ) ꢀ), 1.38 (s, 3H, 6βꢀCH ), 1.08 (s, 3H, 6αꢀCH ); C NMR (100 MHz, CDCl ) δ 176.8
3
2
3
3
3
(>C=O), 157.4 (Cꢀ1 or Cꢀ5), 155.1 (Cꢀ5 or Cꢀ1), 143.8, 137.5 (imidazole), 135.4, 128.7
(imidazole), 119.4, 119.2, 112.6, 106.0 (Cꢀ2 or Cꢀ4), 105.7 (Cꢀ4 or Cꢀ2), 77.0 (Cꢀ6), 60.9 (ꢀOCH ꢀ),
2
4
6.2, 45.2, 43.8, 36.0, 31.9, 29.8, 28.1, 27.8, 26.1, 25.8, 23.7, 18.7; mass spectrum (EI) m/z (relative
+
intensity) 438 (M , 62), 423 (12), 395 (80, 355 (7), 317 (6), 304 (10), 285 (7), 201 (6), 149 (10, 88
+
(18), 56 (100); Exact mass (EI) calculated for C H N O (M ), 438.2519; found 438.2518. HPLC
2
6
34
2
4
(
4.6 x 250 mm, Supelco discovery column, acetonitrile/water) showed purity 98% and retention
time 7 min for the title compound. LC/MS analysis (Waters MicroMass ZQ system) showed purity
8% and retention time 5.5 min for the title compound.
9
3
4
2ꢀ(3,5ꢀDimethoxyphenyl)ꢀ2ꢀmethylpropanal (12). To a solution of 11 (3.0 g, 14.6 mmol) in dry
CH Cl (50 mL) at ꢀ78°C under an argon atmosphere was added dropwise DIBALꢀH (37 mL, 37
2
2
mmol, 1 M solution in CH Cl ). The reaction mixture was stirred at the same temperature for 1 hour
2
2
and then quenched by dropwise addition of potassium sodium tartrate (10% solution in water). The
resulting mixture was warmed to room temperature, stirred vigorously for 40 min, and then diluted
with ethyl acetate. The organic phase was separated and the aqueous phase was extracted with ethyl
acetate. The combined organic layer was washed with brine, dried (MgSO ), and concentrated
4
under reduced pressure. Purification by flash column chromatography on silica gel (20% ethyl
ACS Paragon Plus Environment

Page 29 of 67
Journal of Medicinal Chemistry
1
2
3
acetate in hexane) gave 12 (2.8 g, 94%) as a colorless oil. IR (neat): 2970, 2937, 1725 (s, >C=O),
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
ꢀ
1 1
1
592, 1456, 1423, 1204, 1159, 1048 cm ; H NMR (500 MHz, CDCl ) δ 9.46 (s, ꢀCH=O), 6.40 (d,
3
J = 2.0 Hz, 2H, ArH), 6.39 (t, J = 2.0 Hz, 1H, ArH), 3.79 (s, 6H, ꢀOCH ), 1.43 (s, 6H, ꢀC(CH ) ꢀ);
3
3 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
+
+
mass spectrum (EI) m/z (relative intensity) 208 (M , 25), 196 (16), 179 (M ꢀCHO), 165 (25), 151
+
(14), 139 (39), 91 (20), 77 (20); Exact mass (EI) calculated for C H O (M ), 208.1099; found,
1
2
16
3
2
08.1108.
2ꢀ(3,5ꢀDimethoxyphenyl)ꢀ2ꢀmethylpropanꢀ1ꢀol (13). To a solution of 12 (1.70 g, 8.16 mmol) in
anhydrous methanol (40 mL) at room temperature under an argon atmosphere was added sodium
borohydride (1.39 g, 36.7 mmol) portionwise. The reaction mixture was stirred for 1 hour and then
quenched by the addition of aqueous 1N HCl and extracted with ethyl acetate. The organic layer
was washed with brine, dried (MgSO ), and evaporated. Purification by flash column
4
chromatograph on silica gel (20% ethyl acetate in hexane) gave 13 (1.6 g, 94%) as a colorless oil.
ꢀ1 1
IR (neat): 3433 (br, OH), 2961, 2836, 1592, 1456, 1422, 1202, 1159, 1048 cm ; H NMR (500
MHz, CDCl ) δ 6.53 (d, J = 2.0 Hz, 2H, ArH), 6.34 (t, J = 2.0 Hz, 1H, ArH), 3.80 (s, 6H, ꢀOCH ),
3
3
3.58 (s, 2H, ꢀCH OH), 1.30 (s, 6H, ꢀC(CH ) ꢀ); mass spectrum (ESI) m/z (relative intensity) 211
2
3 2
+
(M + H, 100).
5ꢀ(1ꢀHydroxyꢀ2ꢀmethylpropanꢀ2ꢀyl)resorcinol (14). To a stirred solution of 13 (400 mg, 1.9
mmol) in dry CH Cl (20 mL) at ꢀ78°C, under an argon atmosphere, was added boron tribromide
2
2
(
6.6 mL, 6.6 mmol, 1M solution in CH Cl ). Following this addition, the reaction temperature was
2 2
o
gradually raised over a period of 3 hours to 25 C, and the stirring was continued at that temperature
until the reaction was completed (4 hours). Unreacted boron tribromide was destroyed by the
ACS Paragon Plus Environment