Total Synthesis of the Endocannabinoid Uptake Inhibitor Guineensine and SAR Studies

Article abstract of DOI:10.1002/cmdc.201900390

Guineensine ((2E,4E,12E)-13-(benzo[d][1,3]dioxol-5-yl)-N-isobutyltrideca-2,4,12-trienamide) is a plant-derived natural product that inhibits reuptake of the endocannabinoid anandamide with sub-micromolar potency. We have established a highly efficient tot

Full text of DOI:10.1002/cmdc.201900390

Accepted Article
Title: Total Synthesis of the Endocannabinoid Uptake Inhibitor
Guineensine and SAR Studies
Authors: Karl-Heinz Altmann, Ruben Bartholomäus, Simon Nicolussi,
Mark Rau, Ana Catarina Simao, Alice Baumann, and Jürg
Gertsch
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemMedChem 10.1002/cmdc.201900390
Link to VoR: http://dx.doi.org/10.1002/cmdc.201900390
A Journal of
www.chemmedchem.org

10.1002/cmdc.201900390
ChemMedChem
FULL PAPER
Total Synthesis of the Endocannabinoid Uptake Inhibitor
Guineensine and SAR Studies
Ruben Bartholomäus,[a] Simon Nicolussi,[b] Alice Baumann,[a] Mark Rau,[b] Ana Catarina Simão,[a] Jürg
Gertsch,[b] and Karl-Heinz Altmann*[a]
[a]
Dr. R. Bartholomäus (ORCID ID 0000-0003-1696-5045), Msc. ETH A. Baumann, Dr. A. C. Simão, Prof. Dr. K.-H. Altmann (ORCID ID 0000-0002-0747-
9734)
Department of Chemistry and Applied Biosciences
ETH Zürich
HCI H405
Vladimir-Prelog-Weg 4
CH-8093 Zürich, Switzerland
E-mail: karl-heinz.altmann@pharma.ethz.ch
[b]
Dr. S. Nicolussi (ORCID ID 0000-0003-1260-3282), MSc. M. Rau, Prof. Dr. Jürg Gertsch (ORCID ID 0000-0003-0978-1555)
Institute of Biochemistry and Molecular Medicine
University of Bern, Switzerland
Bühlstrasse 28
CH-3012 Bern, Switzerland
Supporting information for this article is given via a link at the end of the document.
Abstract: Guineensine ((2E,4E,12E)-13-(benzo[d][1,3]dioxol-5-yl)-N-
isobutyltrideca-2,4,12-trienamide) is a plant-derived natural product
that inhibits the reuptake of the endocannabinoid anandamide with
sub-M potency. We have established a highly efficient total synthesis
of guineensine, which provided the natural product in only 5 steps
from commercially available 3-nonyn-1-ol in 17% overall yield, relying
on the attachment of the benzodioxolyl moiety to the unsaturated fatty
acid chain by means of a Suzuki coupling as the key step. Subsequent
SAR studies revealed that the replacement of the N-iso-butyl group in
the natural product by different alkyl, arylalkyl, or aryl groups is
generally well tolerated and derivatives could be identified that are
slightly more potent anandamide reuptake inhibitors than guineensine
itself. In contrast, modifications of the benzodioxolyl moiety led to
reduced activity. Intriguingly, a change in the configuration of the C4-
C5 double bond from E to Z was found to be very well tolerated, in
spite of the associated change in the overall geometry of the
molecule.
amide hydrolase (FAAH) or monoacylglycerol lipase (MAGL) nor
interacting with the cannabinoid receptors CB1 or CB2 to any
significant extent. In vivo, guineensine (1) dose-dependently
induced cannabimimetic effects in mice, such as strong
catalepsy, hypothermia, reduced locomotion and analgesia;[7a] in
addition, it was also shown to exhibit pronounced anti-
inflammatory effects.[7b] Guineensine (1) is a more potent inhibitor
of endocannabinoid uptake than either retrofractamides A/B or
brachystamide B, thus highlighting the importance of the length of
the connecting chain between the terminal amide group
(designated here as the head group) and the benzodioxolyl
moiety (tail group) for the inhibition of endocannabinoid uptake.[7a]
Introduction
Guineensine (1) (Fig. 1) is an unsaturated fatty acid-derived
natural product that belongs to the family of piperamides and was
first isolated in 1974 by Okogun et al. from the fruits of Piper
guineense collected in the western part of Nigeria.[1] Additional
members of this natural product family include, e. g.,
While both the physiological role of FAAH in EC metabolism and
the pharmacology of FAAH inhibitors have been well studied,[8]
the exact mechanism of EC membrane transport and the
pharmacological consequences of the direct, specific inhibition of
this process are much less well understood. In particular, it is still
unclear if EC membrane transport is mediated by a dedicated
membrane transport mechanism[9]. Given its ability to inhibit EC
cellular uptake in a potent and selective fashion, the guineensine
scaffold represents an attractive structural platform for the
development of tool compounds for mechanistic studies, including
compounds that would allow to selectively target EC membrane
transport. At the same time, guineensine analogs with further
improved potency and selectivity could be potential drug
retrofractamides A[2] and
B
(also known as pipercide),[3]
brachystamide B,[4] or laetispicine.[5]
Piperamides have been reported to exhibit a broad range of
biological activities, including antibacterial, antitumor, anti-
inflammatory, antidiabetic, and antidepressive activities as well as
pesticidal effects.[6] Most recently, guineensine (1) has attracted
significant attention because of its potent and selective inhibition
of endocannabinoid (EC) cellular uptake.[7a] Thus, guineensine (1)
strongly reduces the cellular reuptake of the main
endocannabinoid anandamide (AEA) in U937 cells (EC50 = 290
nM), while neither inhibiting the EC-degrading enzymes fatty acid
1
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900390
ChemMedChem
FULL PAPER
candidates for the treatment of anxiety, pain, chronic stress, or a
number of other disorders.[10] The only other potent and selective
inhibitor of EC reuptake reported to date is (2E,4E)-N-[2-(3,4-
dimethoxyphenyl)ethyl]dodeca-2,4-dienamide (WOBE437) that
we recently characterized in detail in vitro and in vivo.[11]
the mixture of E/Z isomers), the E/Z selectivity of the reaction was
poor. In fact, the major product of the reaction was the undesired
Z olefin, which proved to be inseparable from the desired E
isomer. Subsequent experiments with NaHDMS or KHMDS
produced 4 in an even more unfavorable E/Z ratio (Scheme 1).
These findings are in contrast to the results reported by Shingala
et al.[14] for the reaction of 2 with the bis-homo (C10) analog of
aldehyde 3. It appears, however, that the reaction in Shingala’s
case was performed under Barbier conditions, which might affect
its stereochemical outcome. (Ref.[14] does not include any
experimental protocols). In agreement with our observations, the
predominant formation of the Z isomer has also been observed in
the KHMDS-mediated reaction between 5-(benzylsulfonyl)-1-
phenyl-1H-tetrazole and decanal in DME.[15] In light of these
findings, the Julia-Kocienski approach towards construction of the
C12-C13 double bond was abandoned.
To date, three total syntheses of guineensine (1) have been
reported in the literature. Thus, Okwute et al.[12] described a 9-
step synthesis of 1 from catechol already in 1979, but the
compound was obtained only in 0.34% overall yield (1.47% from
1,2-(methylenedioxy)benzene). A more efficient approach was
subsequently developed by Vig et al.,[13] who were able to access
the acid precursor of 1 in 7 steps and ca. 7% overall yield from
1,8-octandiol (no yield was reported for the amide bond-forming
step). In the most recent synthesis of 1, Shingala at al. employed
a Julia-Kocienski olefination to establish the C12-C13 double
bond, while the dienoate moiety was constructed via a Corey-
Fuchs alkynylation/Rychnovsky-Trost isomerization sequence.
Shingala’s synthesis comprised a total of 12 steps from
commercial starting materials and gave 1 in 27% overall yield.[14]
No structure-activity relationship (SAR) studies on the effects of
guineensine analogs on EC membrane transport have been
reported so far, with the exception of a small series of naturally
occurring piperamides, including retrofractamides A/B and
brachystamide B.[7a] All of these compounds were found to be less
potent inhibitors of anandamide cellular uptake than 1.[7a]
Instead, we turned our attention to the use of a Suzuki cross-
coupling reaction for the construction of the terminal styryl moiety
in 1 (Fig. 1). Given the commercial availability of a broad range of
boronic acids, this approach would also enable late stage
diversification of the aromatic tail end moiety of the guineensine
scaffold as an attractive feature.
We were interested in the total synthesis of guineensine (1) in the
broader context of a project aiming (i) at the development of tools
that would enable us to target and probe EC cellular transport and
(ii) at the identification of highly potent and selective inhibitors of
EC transport as leads for drug discovery. To provide a chemical
basis for these efforts, we required an effective, concise, and also
modular synthesis of 1, in order to allow diversification of the
scaffold both at the arene moiety as well as the amide part. In this
report we now disclose a new, efficient total synthesis of
guineensine (1). The chemistry developed as part of the total
synthesis work was subsequently applied to the synthesis of
analogs for SAR studies, which have provided first insights into
the effect of modifications at either end of the guineensine scaffold
on the inhibitory activity for EC membrane transport.
Figure 1. Key retrosynthetic disconnections for the 2nd generation approach
towards 1.
As outlined in Scheme 2, our second generation approach
towards 1 departed from (commercially available) 3-nonyn-1-ol
(5), which was first subjected to an alkyne zipper reaction[16] to
shift the triple bond to the terminal position. This was followed by
Swern oxidation and subsequent Horner-Wadsworth-Emmons
olefination of the ensuing aldehyde with phosphonate 6, to
produce dienoate 7 in 45% overall yield (for the three-step
sequence from 5) and with high stereoselectivity (4,5-E/Z = 10:1,
(E/Z-2,3 > 20:1). Ester saponification and subsequent amide
bond formation via the acid chloride proceeded smoothly. The
remaining undesired Z-4,5 isomer could be removed at this stage
by flash chromatography, thus providing dienamide 8 as a single
isomer in 84% overall yield from ester 7. Pd-catalyzed
hydrostannylation of the terminal alkyne moiety followed by in situ
treatment of the resulting stannane with iodine then gave vinyl
iodide 9 in 64% yield. However, this material was only ca. 90%
pure; isolation of the intermediate stannane prior to tin-iodide
exchange did not yield material of higher purity.[17]
Results and Discussion
Our initial approach towards the synthesis of 1 was to be based
on the construction of the C12-C13 E-double bond by Julia-
Kocienski olefination between sulfone 2 and aldehyde 3 (Scheme
1), in analogy to work reported previously by Shingala et al.[14]
The crucial Suzuki cross-coupling of 9 with boronic acid 10 was
initially performed with Pd(PPh3)4 and Na2CO3 in DME/H2O 3:1,
but these conditions gave guineensine (1) in only 20% yield with
an unsatisfactory purity of 70% (after flash chromatography).
Both the yield and purity of the coupling product could be
significantly improved by the use of thallium ethoxide[18,19] as an
additive, which also reduced the reaction time to 5-10 min. The
TlOEt-mediated coupling reaction gave guineensine (1) in 34%
yield as a ca. 9:1 mixture with what we assume to be the geminally
substituted olefin as a side product; the latter could not be
removed by conventional flash chromatography. Purification of
this mixture was possible by preparative RP-HPLC, to provide
Scheme 1. a) 1. Sulfone 2, base, THF, -78 °C, 15 min; 2. 3, THF, -78 °C, 30
min.
Unfortunately, while treatment of 2 with LiHMDS followed by
reaction with aldehyde 3 gave the olefin in excellent yield (95% of
2
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900390
ChemMedChem
FULL PAPER
pure guineensine (1), whose analytical data were in full
agreement with those reported in the literature.[1,12b,20]
Scheme 3. a) Bu3SnH, Pd2(dba)3, Cy3PHBF4, DIEA, CH2Cl2, 0 °C, 98%; b)
catecholborane, THF, 70 °C, then H2O, rt, 60%; c) Pd(PPh3)4, DMF, 75 °C, 5 h,
46%; d) Pd(PPh3)4, Na2CO3, DME/H2O 3:1, 80 °C, 74%; e) (COCl)2, DMSO,
Et3N, CH2Cl2, -78 °C to rt, 88%; f) LDA, THF, -78 °C to -20 °C, 68%, 4,5-E/Z =
10:1, 2,3-E/Z >20:1; g) 2M NaOH, MeOH, 60 °C, 99%; h) (COCl)2, then iso-
butylamine, Et3N, CH2Cl2, 86%; i) LDA, THF, -78 °C to rt, 55%, 4,5-E/Z = 8:1,
Scheme 2. a) NaH, 1,3-diaminopropane, 70 °C, 80%; b) (COCl)2, DMSO, Et3N,
CH2Cl2, -78 °C to rt, 88%; c) 6, LDA, THF, -78 °C to -20 °C, 64%, 4,5-E/Z =
10:1, 2,3-E/Z > 20:1; d) 2M NaOH, MeOH, 65 °C, quant; e) 1. (COCl)2, CH2Cl2,
then iso-butylamine, Et3N, 84%; f) 1. Bu3SnH, Pd2(dba)3, Cy3PHBF4, DIEA,
CH2Cl2, 0 °C; 2. I2, CH2Cl2, rt, 64%; g) 10, TlOEt, Pd(PPh3)4, THF/H2O 3:1, rt,
34%, 4,5-E/Z > 20:1, 2,3-E/Z > 20:1.

2,3-E/Z > 20:1.
While HPLC purification of compounds for biological testing after
the final synthetic step is fully acceptable in the context of SAR
studies, from a total synthesis perspective, the incomplete
selectivity in the transformation of alkyne 8 into vinyl iodide 9 and
the resulting formation of an inseparable side product in the
subsequent Suzuki coupling step were still unsatisfactory. We
thus evaluated an alternative cross-coupling-based strategy
towards 1 that involved the early installation of the benzodioxole
moiety through Suzuki or Stille coupling between stannane 12 or
boronic acid 13, respectively, and bromide 14 (Scheme 3).
sequence of five steps (from 5). This represents the shortest and
most efficient synthesis of this natural product reported to date.
With an efficient total synthesis of the natural product established,
we turned our attention to the synthesis of selected analogs, in
order to gain first insights into the SAR of 1 with regard to inhibition
of EC transport. As illustrated in Scheme 4, we have utilized acid
18 to prepare head-group modified variants of guineensine with
variations in the size, lipophilicity and aliphatic/aromatic character
of the amide moiety. In all cases, the coupling reaction via the in
situ generated acid chloride proceeded smoothly and provided
the desired products 19 in yields between 59% and 86% after
purification by flash column chromatography. For most of the
analogs 19, the 4,5-Z isomer originating from the imperfect
stereoselectivity of the HWE olefination between 16 and 6 could
be removed after the coupling reaction. In all other cases the
fraction of 4,5-Z isomer in the final product was lower than 10%.
Stannane 12 and boronic acid 13 were both obtained from 8-
nonyn-1-ol (11) (prepared in one step from 5; Scheme 2) by
palladium-catalyzed hydrostannylation and hydroboration,
respectively. Both compounds were obtained as single regio- and
stereoisomers after flash column chromatography. While
stannane 12 was obtained from 11 in higher yield than boronic
acid 13, the Suzuki coupling of the latter with bromide 14 was
more efficient than Stille coupling of the former.[21] Thus, the
overall yield for the transformation of 11 into 15 was comparable
for both routes (44% vs. 43%). Alcohol 15 was then elaborated
into guineensine (1) by Swern oxidation, to provide aldehyde 16,
Scheme 4. a) 1. (COCl)2, CH2Cl2, rt, 30 min. 2. R-NH2, Et3N, CH2Cl2, rt, 15 min,
59-86%. For the structure of substituents R see Table 1.
followed by HWE-olefination with phosphonate
proceeded with excellent stereoselectivity (4,5-E/Z = 10:1, 2,3
E/Z > 20:1)), ester saponification and finally amide bond formation
via the acid chloride. The undesired, minor
4,5-Z isomer
originating from the HWE reaction could be removed after the
coupling step by flash column chromatography, to give
guineensine (1) as a single isomer in a total yield of 18% for the
seven-step sequence from 5 (as the ultimate commercial starting
material). The number of linear steps could be further reduced by
directly reacting aldehyde 16 with phosphonate 17 (for the
preparation of 17 see the SI). This highly convergent approach
provided guineensine (1) in a total yield of 17% for a longest linear
6 (which
-
The preparation of guineensine analogs with modifications of the
aryl moiety was based on vinyl iodide 9 (10:1 mixture of isomers,
vide supra), which was converted into structures 20 by means of
Suzuki coupling with a series of boronic acids under the
conditions elaborated for the synthesis of the natural product.
(Scheme 5).

3
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900390
ChemMedChem
FULL PAPER
All compounds were assessed for their ability to inhibit AEA
transport into U937 cells, using radiolabeled AEA as the transport
substrate.[7a] As AEA uptake is also reduced upon inhibition of
FAAH, the inhibitory activity of guineensine analogs against
FAAH was also determined independently.[7a] The corresponding
data are summarized in Tables 1 and 2 for head group (cpds. 19)
and tail group (cpds. 20) modifications, respectively.
Scheme 5. a) Ar-B(OH)2, TlOEt, Pd(PPh3)4, THF/H2O 3:1, rt, 37-52% (ca. 90%
purity). For the structures of Ar see Table 2.
As for the synthesis of 1, analogs 20 were generally contaminated
with ca. 10% of a side product that could not be removed by flash
column chromatography; however, all products 20 could be
obtained in analytically pure form by final RP-HPLC purification.
In addition to the assessment of head and tail group modifications,
we have also investigated analogs 21 and 22, which are
characterized by a higher and lower level of saturation of the C12-
C13 bond, respectively, and we have determined the importance
of the configuration of the C4-C5 double (analog 23).
Table 1. Inhibition of AEA uptake into U937 cells by head group-modified
guineensine analogs 19.
% FAAH
inhibition at
1 M[d]
Maximum
inhibition (%)[c]
Cpd.[a]
R[a]
EC50 [M] (95% CI)[b]
1
>60
>60
>60
>60
44.1[e]
0.288 (0.190-0.437)
0.409 (0.267-0.626)
0.374 (0.268-0.523)
0.596 (0.394-0.900)
19a
19b
19c
0
0
37
19d
0.661 (0.345-1.266)
1.50 (0.911-5.028)
0.972 (0.286-1.613)
0.438 (0.252-0.764)
>60
>60
>60
>60
12
14[f]
3[f]
19e
19f
19g
19h
23[f]
As illustrated in Scheme 6, analog 21 was elaborated from THP-
ether 4 (mixture of double bond isomers; Scheme 1), which was
converted into aldehyde 24 by THP-ether cleavage, followed by
catalytic hydrogenation and subsequent DMP oxidation in 80%
overall yield. Aldehyde 24 was then elaborated into 21 in analogy
to the preparation of guineensine (1) from 16 (cf. Scheme 3).
0.344 (0.201-0.590)
0.123 (0.095-0.158)
>60
>60
22
30
19i
19j
0.097 (0.062-0.218)
>60
7
>10
45
0[f]
42
35
19k
19l
0.096 (0.071-0.130)
0.112 (0.074-0.169)
>60
>60
19m
Scheme 6. a) TsOH, MeOH, 95%; b) H2, Pd/C, EtOH, 95%; c) (COCl)2, DMSO,
Et3N, CH2Cl2, -78 °C to rt, 84%; d) 6, LDA, THF, -78 °C to -20 °C, 70%, 4,5-E/Z
= 10:1; e) 2M NaOH, MeOH, 70 °C, 76%; f) EDC, HOBt, Et3N, iso-butylamine,
CH2Cl2, 59%.
0.463 (0.267-0.801)
19n
>60
17
0.81 (0.058-0.112)
0.116 (0.062-0.218)
19o
19p
>60
>60
2
7
Analog 22 was obtained from alkyne 8 (Scheme 2) and bromide
14 (Scheme 3) by Sonogashira coupling (Scheme 7), while 23
was isolated as a minor side product from the reaction of acid 18
(as a ca. 10:1 mixture of 4,5-E/Z isomers) with iso-butyl amine
(vide supra).
19q
19r
0.232 (0.145-0.370)
1.70 (1.04-2.80)
>60
ND[g]
83
8
0[f]
0
19s
0.579 (0.337-0.994)
[a] For the general structure of analogs 19, cf. Scheme 4. [b] Concentration
required for half-maximal inhibition of uptake of AEA into U937 cells. EC50
values represent the mean of at least three independent experiments. 95% CI,
95% confidence interval. [c] Maximal inhibition of uptake of AEA into U937 cells.
Maximum inhibition data in our uptake assay data varied significantly; thus, at
this point, we consider numbers >60% as essentially indistinguishable. [d]
Inhibition of FAAH in U937 cell homogenates. In general, 1 M was the highest
concentration tested. [e] IC50 value. [f] IC50 >10 M. [g] ND, not determined.
Scheme 7. a) 14, Pd(PPh3)2Cl2, CuI, n-BuNH2, 60 °C, 30 min, 11%.
4
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900390
ChemMedChem
FULL PAPER
With one single exception, none of the head group modifications
investigated here, which include the replacement of the natural N-
iso-butyl group by other alkyl groups as well as differently
substituted arylalkyl and aryl moieties, led to a substantial
decrease in AEA uptake inhibition (Table 1). In fact, 16 out of 19
different analogs exhibited sub-M EC50 values, with even
compounds 19e and 19r still being active in the low single digit
M range. Interestingly, while N-methyl amide 19e is less active
than all other N-alkyl amides, a re-increase in potency is observed
for the less hydrophobic primary amide 19g; the latter is not
significantly less active than guineensine (1). As has been
reported for guineensine (1), none of the compounds is able to
block AEA uptake completely, which is in line with previous
findings on other types of AEA uptake inhibitors. The possible
reasons for this observations have been discussed[7a] and the
arguments shall not be repeated here. Importantly, none of the
analogs showed effects on FAAH activity that would suggest a
major contribution of FAAH inhibition to the reduction in AEA
uptake, certainly not for those analogs with activity similar to or
better than that of guineensine (1). Five compounds (19i, j, l, m,
p) appear to be between 2- and 3-fold more potent than 1,
although the 95% confidence interval in all cases overlaps with
that of 1 and the differences, thus, should not be overinterpreted.
It should be noted, however, that the most potent compound
identified here (19j) incorporates a (3,4-dimethoxyphenyl)ethyl
substituent on the amide nitrogen. The same amide head group
is also part of the recently described potent and selective AEA
uptake inhibitor (2E,4E)-N-(3,4-dimethoxyphenethyl)dodeca-2,4-
dienamide (WOBE437)[11] and thus seems to be a privileged
structural feature for fatty acid amide-based AEA cellular uptake
inhibitors. Noteworthy, the ortho-dimethoxyphenyl motif is also
found in benzylamide 19p; the inhibitory activities of N-(3,4-
dimethoxyphenyl)ethyl amide 19j and of N-3,4-dimethoxybenzyl
amide 19p are 3-fold and 6-fold higher, respectively, than for the
corresponding unsubstituted parent compounds 19i and 19o.
As the sole example within the series of head group-modified
guineensine analogs investigated, compound 19k is dramatically
less potent than the natural product. In light of the sub-M activity
of analogs 19i and 19j, the loss in potency incurred by 19k is
unexpected and may in fact raise questions about the validity of
the result. However, the poor activity of 19k was reproduced with
two different batches of material (i. e. originating from two
separate synthesis campaigns) and we are confident about the
reliability of the data also for this analog.
and perhaps not too surprisingly, the complete removal of the
benzo[d][1,3]dioxole moiety from guineensine (1) leads to a sharp
drop in activity, although not to the same extent as for analogs
20b and 20f.
Table 2. Inhibition of AEA uptake into U937 cells by tail group-modified
guineensine analogs 20.
% FAAH
inhibition at
10 M[d]
Maximum
inhibition (%)[c]
Cpd.[a]
R[a]
EC50 [M] (95% CI)[b]
3.59 (1.00-12.86)
>100
>60
41
24
20a
20b
29[e]
1.3 (0.9-1.9)
11.0 (12.0-6.1)
6.0 (3.6-10.1)
>100
>60
>60
60
35
33
20c
20d
20e
20f
21[f]
12[g]
32
0.849
50
31[g]
20g
1.66 (1.16-2.36)
5.75 (4.33-7.63)
>60
>60
2.4[h]
11
20h
8
-
[a] For the general structure of analogs 19, cf. Scheme 4. [b] Concentration
required for half-maximal inhibition of uptake of AEA into U937 cells. EC50
values represent the mean of at least three independent experiments. 95% CI,
95% confidence interval. [c] Maximal inhibition of uptake of AEA into U937 cells.
Maximum inhibition data in our uptake assay data varied significantly; thus, at
this point, we consider numbers >60% as essentially indistinguishable. [d]
Inhibition of FAAH in U937 cell homogenates at 10 M compound
concentration. [e] 100 M. [f] 5 M. [g] 1 M. [h] IC50 value [M].
A very moderate loss in activity is also observed upon formal
reduction of the C12-C13 double bond in guineensine (1) (ca. 2-
fold, Table 3).
In contrast to head group-modified guineensine analogs 19, all of
the investigated tail group-modified variants 20 exhibit reduced
activity compared to 1 (Table 2). Strikingly, a dramatic drop in
AEA uptake inhibition is observed for analogs 20b and 20f, both
of which incorporate a methoxy group in the para position of the
terminal styryl moiety. This finding suggests that substitution at
this position is not tolerated outside of the context of the bicyclic
benzo[d][1,3]dioxole (or perhaps a closely related bicyclic) moiety
that is present in the natural product, although this may also
depend on the exact nature of the substituent. We note, however,
that we cannot rule out the possibility that 20b and 20f exhibit
particularly low solubility, which would obscure the results.
Interestingly, the replacement of the benzo[d][1,3]dioxolyl group
by an indole moiety (analog 20h) produces a reasonably potent
FAAH inhibitor and the inhibition of AEA uptake by this compound
may largely be attributed to its FAAH inhibitory activity. Finally,
Table 3. Inhibition of AEA uptake into U937 cells by backbone-modified
guineensine analogs 21-23.
% FAAH
inhibition at
1 M[d]
Maximum
inhibition (%)[c]
Cpd.[a]
EC50 [M] (95% CI)[b]
0.533 (0.317-0.895)
1.16 (0.70-1.92)
>60
>60
47
7
0
0
21
22
23
0.134 (0.047-0.384)
[a] For the structure of analogs 21-23 see text. [b] Concentration required for
half-maximal inhibition of uptake of AEA into U937 cells. EC50 values represent
the mean of at least three independent experiments. 95% CI, 95% confidence
interval. [c] Maximal inhibition of uptake of AEA into U937 cells. Maximum
inhibition data in our uptake assay data varied significantly; thus, at this point,
we consider numbers >60% as essentially indistinguishable. [d] Inhibition of
FAAH in U937 cell homogenates at 1 M compound concentration.
5
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900390
ChemMedChem
FULL PAPER
A somewhat more pronounced effect (ca. 4-fold reduction in
potency compared to 1) is associated with the replacement of the
C12-C13 double bond by a triple bond, although this modification
leads to a distinct change in the orientation of the aromatic tail
group relative to the alkyl chain. Intriguingly, the change in the
configuration of the C4-C5 double bond from E to Z does not
affect potency, in spite of the significantly different overall shape
of the corresponding analog 23. However, compared to 1, the
maximum efficacy observed for analog 23 is lower; although the
significance of this observation is not clear at this point.
Research (NCCR) TransCure. We are indebted to Dr. Bernhard
Pfeiffer and Leo Betschart for NMR support, to Dr. Xiangyang
Zhang, Louis Bertschi, Rolf Häfliger, and Oswald Greter for
HRMS spectra, and to Kurt Hauenstein for general technical
support. We thank Patricia Schenker und Tatiana Hofer for
excellent support with compound testing.
Conflict of interest
This authors declare no conflict of interest.
Conclusions
Keywords: endocannabinoids • endocannabinoid mebrane
transport • guineensine • natural product • total synthesis
We have established a new and highly efficient total synthesis of
the natural endocannabinoid reuptake inhibitor guineensine (1),
which provides the natural product in 5 steps (longest linear
sequence) and 17% overall yield. The chemistry developed in the
course of the total synthesis work was then exploited for the
synthesis of a series of analogs, which have offered first insights
into the SAR of guineensine-derived AEA uptake inhibitors. Thus,
the natural N-iso-butyl substituent on the terminal amide group
can be replaced by a number of alkyl, arylalkyl, or aryl groups
without a significant decrease in potency; some analogs are even
slightly more potent than 1. At first glance, the inhibition of AEA
uptake appears to track with the hydrophobicity of the group
attached to the amide nitrogen. However, as illustrated by the
excellent activity of primary amide 19g and the poor activity of
19k, the situation must be more complex, even if our findings
cannot be rationalized at this point. Compared to the amide
moiety, the benzo[d][1,3]dioxole tail group appears to be more
sensitive to structural changes and all of the analogs 20 were
found to be less potent than 1. While we have only studied a very
limited number of tail group-modified guineensine variants at this
point, the data again suggest that AEA cellular uptake inhibition,
as for the amide moiety, is not simply related to the presence of a
(any) hydrophobic tail group, but clearly depends on the exact
structure of this group. In line with our previous studies[11][22], AEA
membrane transport can be potently inhibited independent of
FAAH with a distinctive SAR indicative of a protein target. Overall,
our work presents the basis for the design and synthesis of more
advanced guineensine analogs or, more generally, fatty acid
amide-derived AEA cellular uptake inhibitors, with the ultimate
goal of developing more drug-like structures. Work along these
lines in combination with efforts aiming at the identification of the
target of EC membrane transport inhibitors are currently ongoing
in our laboratories.
References:
[1]
[2]
J. I. Okogun, D. E. U. Ekong, J. Chem. Soc. Perkin Trans. 1 1974, 1-4.
A. Banerji, D. Bandyopadhyay, M. Sarkar, A. K. Siddhanta, S. C. Pal, S.
Ghosh, K. Abraham, J. N. Shoolery, Phytochemistry 1985, 24, 279-284.
M. Miyakado, I. Nakayama, H. Yoshioka, N. Nakatani, Agricult. Biol.
Chem. 1979, 43, 1609-1611.
[3]`
[4]
[5]
A. Banerji, C. Das, Phytochemistry 1989, 28, 3039-3042.
C. Y. Yao, J. Wang, D. Dong, F. G. Qian, J. Xie, S. L., Pan,
Phytomedicine 2009, 16, 823-829.
[6]
[7]
For recent reviews, cf.: a) S. K. Okwute, H. O. Egharevba, Int. J. Chem.
2013, 5, 99-122. b) J. C. do Nascimento, V. F. de Paula, J. M. David, J.
P. David, Quim. Nova 2012, 35, 2288-2311.
a) S. Nicolussi, J. M. Viveros-Paredes, M. S. Gachet, M. Rau, M. E.
Flores-Soto, M. Blunder, J. Gertsch, Pharmacol. Res., 2014, 80, 52–65.
b) I. Reynoso-Moreno, I. Najar-Guerrero, N. Escareꢀo, M. Eduardo
Flores-Soto, J. Gertsch, J. M. Viveros-Paredes, J. Agric. Food Chem.
2017, 65, 9435-9442.
[8]
[9]
For recent reviews, see: a) W. Tuo, N. Leleu-Chavain, J. Spencer, S.
Sansook, R. Millet, P. Chavatte, Philippe, J. Med. Chem. 2017, 60, 4-46.
b) S. Ogawa, H. Kunugi, Curr. Neuropharmacol. 2015, 13, 760-775.
a) S. Nicolussi, J. Gertsch, Vitam. Horm. 2015, 98, 441-485. b) C. J.
Fowler, Trends Pharmacol. Sci. 2012, 33, 181-185. c) R. Mechoulam, D.
G. Deutsch, Proc. Natl. Acad. Sci. USA 2005, 102, 17541–17542.
[10] P. Pacher, Pharmacol. Rev. 2006, 58, 389-462.
[11] A. Chicca, S. Nicolussi, R. Bartholomäus, M. Blunder,d, A. Aparisi Rey,
V. Petruccia, I. del Carmen Reynoso-Moreno, J. M. Viveros-Paredes, M.
D. Gensa, B. Lutz, H. B. Schiöth, M. Soeberdt, C. Abels, R.-P. Charles,
K.-H. Altmann, J. Gertsch, Proc. Natl. Acad. Sci. USA 2017, 114, E5006-
E5015
[12] a) S. K. Okwute, D. A. Okorie, J. I. Okogun, Nig. J. Nat. Sci. 1979, 1, 9-
11. The synthesis was later republished with a 1.08% overall yield from
1,2-(methylenedioxy)benzene: b) S. K. Okwute, J. I. Okogun, D. A.
Okorie, Tetrahedron 1984, 40, 2541-2545.
[13] O. P. Vig, S. D. Sharma, O. P. Sood and S. S. Bari, Indian J. Chem.
1980, 19B, 350-352.
[14] S. D. Shingala, G. V. Reddy, R. S. C. Kumar, P. A. Yadav and K. S.
Babu, J. Asian Nat. Prod. Res. 2011, 13, 128-135.
Experimental Section
[15] P. J. Kocienski, A. Bell, P. R. Blakemore, Synlett 2000, 365-366.
[16] C. A. Brown, A. Yarnashita, J. Am. Chem. Soc. 1975, 97, 891-892.
[17] The generation of a vinyl iodide functionality already at the stage of ester
7 would have provided a valuable substrate for two-site diversification,
but was not successful. A complex mixture of products was obtained
upon attempted hydrostannylation of 7, probably caused by reduction of
the dienoate moiety by Bu3SnH.
Detailed protocols for the synthesis of all final products and
intermediates, analytical data for all new compounds, and copies
of the relevant 1H- and 13C-NMR spectra can be found in the SI.
[18] S. Frank, H. Chen, R. K. Kunz, M. J. Schnaderbeck, W. R. Roush, Org.
Lett. 2000, 2, 2691-2694.
Acknowledgements
[19] F. Glaus, K.-H. Altmann, Angew. Chem. Int. Ed. 2015, 54, 1937-1940.
[20] F. Kiuchi, N. Nakamura, Y. Tsuda, K. Kondo, H. Yoshimura, Chem.
Pharm. Bull. 1988, 36, 2452-2465.
This research was supported by the Swiss National Science
Foundation through the National Centre of Competence in
6
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900390
ChemMedChem
FULL PAPER
[21] Stille coupling between stannane 12 and triflate 24 gave 15 in essentially
the same yield as the reaction with bromide 14.
[22] S. Nicolussi, A. Chicca, M. Rau, S. Rihs, M. Soeberdt, C. Abels, J.
Gertsch, Biochem. Pharmacol. 2014, 92, 669-689.
[23] M. Rau, S. Nicolussi, A. Chicca, J. Gertsch, Meth. Mol. Biol. 2016, 1412,
191-203.
7
This article is protected by copyright. All rights reserved.

10.1002/cmdc.201900390
ChemMedChem
FULL PAPER
Entry for the Table of Contents
The piperamide guineensine is a potent inhibitor of endocannabinoid uptake. We have developed an efficient and scalable total
synthesis of guineensine, which has also served as a platform for the synthesis of analogs with modified end groups. While
modifications of the N-substituent on the amide nitrogen were generally well tolerated, the benzodioxolyl group was more sensitive to
structural changes. This group may represent a specific structural requirement for inhibition of endocannabinoid uptake by
guineensine and other piperamides.
8
This article is protected by copyright. All rights reserved.