Synthesis, structural characterization and effect on human granulocyte intracellular cAMP levels of abscisic acid analogs

Article abstract of DOI:10.1016/j.bmc.2014.11.035

The phytohormone abscisic acid (ABA), in addition to regulating physiological functions in plants, is also produced and released by several mammalian cell types, including human granulocytes, where it stimulates innate immune functions via an increase of the intracellular cAMP concentration ([cAMP]i). We synthesized several ABA analogs and evaluated the structure-activity relationship, by the systematical modification of selected regions of these analogs. The resulting molecules were tested for their ability to inhibit the ABA-induced increase of [cAMP]i in human granulocytes. The analogs with modified configurations at C-2′ and C-3′ abrogated the ABA-induced increase of the [cAMP]i and also inhibited several pro-inflammatory effects induced by exogenous ABA on granulocytes and monocytes. Accordingly, these analogs could be suitable as novel putative anti-inflammatory compounds.

Full text of DOI:10.1016/j.bmc.2014.11.035

Bioorganic & Medicinal Chemistry 23 (2015) 22–32
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Synthesis, structural characterization and effect on human
granulocyte intracellular cAMP levels of abscisic acid analogs
Marta Bellotti a,b, , Annalisa Salis b,c, , Alessia Grozio a, Gianluca Damonte a,b, Tiziana Vigliarolo a,
Andrea Galatini d, Elena Zocchi a,b, Umberto Benatti a,b, Enrico Millo a,b,
⇑
a Department of Experimental Medicine, Section of Biochemistry, University of Genoa, Viale Benedetto XV 1, 16132 Genoa, Italy
b Center of Excellence for Biomedical Research (CEBR), University of Genoa, Viale Benedetto XV 5, 16132 Genoa, Italy
c Department of Hearth Environmental and Life Science (DISTAV), University of Genoa, Corso Europa 26, 16132 Genoa, Italy
d Department of Chemistry and Industrial Chemistry, University of Genoa, Via Dodecaneso 31, 16146 Genoa, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 July 2014
Revised 11 November 2014
Accepted 21 November 2014
Available online 27 November 2014
The phytohormone abscisic acid (ABA), in addition to regulating physiological functions in plants, is also
produced and released by several mammalian cell types, including human granulocytes, where it stim-
ulates innate immune functions via an increase of the intracellular cAMP concentration ([cAMP]i).
We synthesized several ABA analogs and evaluated the structure–activity relationship, by the system-
atical modification of selected regions of these analogs. The resulting molecules were tested for their abil-
ity to inhibit the ABA-induced increase of [cAMP]i in human granulocytes. The analogs with modified
configurations at C-20 and C-30 abrogated the ABA-induced increase of the [cAMP]i and also inhibited sev-
eral pro-inflammatory effects induced by exogenous ABA on granulocytes and monocytes. Accordingly,
these analogs could be suitable as novel putative anti-inflammatory compounds.
Keywords:
Abscisic acid
Abscisic acid analogs
cAMP
Granulocytes
Inflammatory disease
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
The discovery that ABA is involved in the response of lower
Metazoa to environmental stress suggested that it may function
The phytohormone abscisic acid (ABA) coordinates important
physiological functions involved in the response to abiotic stress
(temperature, light, drought), such as regulation of seed dormancy
and germination, control of stomatal closure and of gene transcrip-
tion. ABA synthesis is also stimulated by thermal stress in marine
sponges by light in hydroids. In these lower Metazoa, ABA activates
an ADP-ribosyl cyclase (ADPRC) through a protein kinase A
(PKA)-stimulated phosphorylation; ADPRC activation leads to an
increase of the intracellular cyclic ADP-ribose (cADPR) concentra-
tion ([cADPR]i), which in turn induces intracellular calcium release.
The increase of the intracellular calcium concentration [Ca2+]i
stimulates oxygen consumption, water filtration and protein
synthesis in sponges and tissue regeneration in hydroids.1,2
as a ‘stress hormone’ also in higher animals. Granulocytes are the
first line of defense of mammals against pathogens and an increase
of the [Ca2+]i via cADPR is known to stimulate several functional
activities of human granulocytes.3
Indeed, human granulocytes exposed to physical or chemical
stimuli release ABA and the hormone stimulates migration, phago-
cytosis, reactive oxygen species and nitric oxide production in an
autocrine manner. The signaling pathway activated by ABA in
granulocytes sequentially involves binding to its G-protein-
coupled receptor lanthionine synthetase C-like protein 2 (LANCL-2),
activation of adenylate cyclase (AC), cAMP-dependent activation
of protein kinase A (PKA), phosphorylation of the cADPR CD38
and consequent cADPR overproduction, leading to an increase of
the intracellular Ca2+ concentration.3–6
Monocyte/macrophages also release ABA upon stimulation with
activated platelets or quartz particles and autocrine ABA activates
a signaling pathway involving AC and PKA, eventually stimulating
cell migration, release of prostaglandin E2 (PGE2), of the cytokines
monocyte chemoattractant protein-1 (MCP-1) and tumour
Abbreviations: ABA, abscisic acid; [cAMP]i, intracellular cAMP concentration;
[cADPR]i, intracellular cADPR concentration; HATU, O-(7-azabenzotriazol-1-yl)-
1,1,3,3 tetramethyluroniumhexafluorophosphate; NaBH4, sodiumborohydride;
DIBAL, diisobutylaluminium hydride.
necrosis factor-
a (TNF-a
) and of metalloprotease-9 (MMP-9).7
⇑
Corresponding author. Tel.: +39 (0)103533032 33; fax: +39 (0)103533024.
E-mail address: enrico.millo@unige.it (E. Millo).
These inflammatory mediators are known to participate in the
atherosclerotic process.6
These authors contributed equally to this work.
http://dx.doi.org/10.1016/j.bmc.2014.11.035
0968-0896/Ó 2014 Elsevier Ltd. All rights reserved.

M. Bellotti et al. / Bioorg. Med. Chem. 23 (2015) 22–32
23
Interestingly, ABA also stimulates insulin release from human
and murine b-pancreatic cells, via a PKA-dependent activation of
CD38, similarly to the signaling pathway of granulocytes. The dual
activity of ABA in mammals, as a pro-inflammatory cytokine and as
an insulin secretagogue, suggests that ABA may play a role in the
metabolic syndrome (MS), a condition characterized by chronic
adipose tissue inflammation, inducing insulin resistance and
hyperinsulinemia.8 The stimulatory effect of endogenous ABA on
the activation of innate immune cells identifies this hormone as
a new target of anti-inflammatory drugs. The possible role of
ABA in the pathogenesis of the MS is a further incentive for the
identification of ABA antagonists capable of preventing the pro-
inflammatory actions of the hormone.
The scientific literature regarding the synthesis and functional
characterization of ABA analogs is focused on the identification of
the structural requirements for ABA action in plants,9,10 with the
aim of ameliorating plant response to abiotic11–13 and other
stresses.14,15 Structure–activity studies performed in a variety of
plant tissues have identified a number of determinants of the
ABA molecule that need to be conserved in analogs for biological
activity.
In plants, assessing the molecular requirements of ABA action
from analog-bioassay studies is controversial, as metabolic path-
ways may activate or deactivate synthesized compounds.16,17
Moreover, while previous studies have investigated synthesis of
ABA analogs, the data concerning determination of their interac-
tion and affinity with the ABA receptor are conflicting.18–23
A structure–activity study of ABA analogs on mammalian cells
has not yet been attempted. In principle, it is possible that the
structural determinants of the ABA molecule required for its bio-
logical activity are different in plant and animal cells. In plants,
several proteins have been demonstrated to bind ABA and trans-
duce its signal.24 Among these proteins, GPCR2 shares a significant
sequence homology with the only mammalian ABA receptor
described so far, LANCL-2.25 This homology suggests the evolution-
ary conservation of a protein fold capable of binding to conserved
structural determinants of the ABA molecule.
Figure 1. Structural formula of ABA, with the conventional carbon numbering. The
regions of the molecule that were modified in the various analogs are boxed.
Alterations were made to the functional group at C-1, to the configuration at the
C-20–C-30 bond, to the carbonyl group at C-40 and to the cyclohexenic ring
saturation. Hydrogen atoms are omitted for clarity.
replaced with an ester, an amide or with carbon chains of different
length, the double bond between C-20 and C-30 was replaced by an
epoxidic bridge, the carbonyl group at C-40 was reduced or substi-
tuted with an amino-group, and finally the cyclohexenic group was
substituted with an aromatic ring (Fig. 2). To define the relative
contribution of each structural modification to the ABA-agonist
or -antagonist action of the compound, some ABA derivatives with
simultaneous modifications on different parts of the molecule
were also studied (Fig. 2). Another structural modification consid-
ered outside of this scheme was a two-carbon truncation of the
normal 2-cis–4-trans pentadienoic side-chain, to yield the
b-ionone derivatives.
a- and
In the majority of the compounds, the following features were
left unmodified: the C10 hydroxyl group, the 70 methyl group, the
2-Z-enoic acid on the side chain and the methyl group at
position 3 (C6). Preliminary experiments suggested that modifica-
tions at these positions impaired the ability of the analogs to
displace ABA from its binding to granulocyte membranes (not
shown).
The aim of this study was to design and synthesize ABA analogs
and to test their biological activity as ABA agonists or antagonists
on the [cAMP]i in human granulocytes. This screening assay was
chosen because activation by ABA of AC, followed by cAMP-depen-
dent activation of PKA and phosphorylation of CD38 appear to be
the common feature of the ABA-signaling pathway in both innate
immune cells and b-pancreatic cells.3–6 Different regions of the
ABA backbone were systematically modified. In particular, altera-
tions were made to the pentadienoic side-chain (functionality at
C-1), to the configuration at the C-20–C-30 bond, to the carbonyl
group at C-40 and to the cyclohexenic ring saturation. As a func-
tional screening assay, all compounds were tested for their ability
to affect the ABA-induced increase of the intracellular cAMP
concentration ([cAMP]i).
Some analogs were synthesized using methods reported by
others or adapted from published protocols,26–35 while others
(analogs 1, 2, 3, 4, 6, 7b, 9, 14b, 17) were synthesized for the first
time in this study.
2.2. Effect of ABA analogs on the [cAMP]i in human
granulocytes
In principle, the fact that chemical synthesis of the various ABA
analogs yielded two enantiomers might complicate the evaluation
of their biological activity, without prior separation.
However, in plants the (10S) and the (1R0) enantiomers show
similar biological activity in different assay systems,36–38 indicat-
ing that in plants the synthetic isomer (1R0) can mimic at least
some of the effects of the natural one (10S). In addition, in mamma-
lian cells, the functional effects elicited by ABA on different cell
types were observed with either one of the enantiomers at the
same concentration.3 In our work, the ABA-triggered increase of
the [cAMP]i in human granulocytes was similar with both ABA
enantiomers. In line with previous results, the [cAMP]i increased
by 35 5% (n = 24) in human granulocytes incubated for 1 min at
2. Results and discussion
2.1. Synthesis
To identify possible ABA antagonists and to assess a possible
structure–activity relationship, ABA analogs were synthesized by
systematic modification of each of the three regions of the ABA
molecule indicated in Figure 1. The ABA molecule contains a chiral
center (C10) and the racemic mixture was used as a starting mate-
rial for analog synthesis, yielding the enantiomers (10S)(+)ABA and
(10R)(ꢀ)ABA. First, the modifications were made in one region at a
time, to determine the contribution of a particular structural attri-
bute to bioactivity. Thus, the acidic functionality at C-1 was
25 °C in the presence of 1 lM ( )-ABA, as compared to untreated
cells. The 20 synthetic ABA analogs shown in Figure 2 were tested
on human granulocytes for their ability to induce an increase of the
[cAMP]i, or to inhibit the ABA-triggered increase of the [cAMP]i. In
order to favor the possible competition with ABA of analogs behav-
ing as ABA-antagonists, each analog was tested at a concentration
of 50 lM, 10 lM and 1 lM. Results are shown in Figure 3. Most

24
M. Bellotti et al. / Bioorg. Med. Chem. 23 (2015) 22–32
Figure 2. Structures of the ABA analogs tested in this study. Analogs were prepared by total synthesis, starting from the racemic mixture of ( )-ABA. Thus, all compounds
were racemic mixtures.
analogs, when tested alone, behaved as ABA-agonists, inducing an
increase of the [cAMP]i which was similar or even higher
(e.g., analogs 1, 2, 3 and 9) than that elicited by ABA, at all three
concentrations tested. However, analogs 10 and 11 reduced or
prevented the ABA-triggered increase of the [cAMP]i, behaving as
ABA-antagonists.
the absence of ABA. Interestingly, when compounds 1, 2 and 3
were tested at the concentration of 1 M in the presence of an
l
equal concentration of ABA, the effect of compounds 2 and 3 on
the [cAMP]i was not additive with that of ABA alone, while com-
pound 1 exhibited a synergistic activity (Fig. 3C). This suggests that
compounds 2 and 3 may act on the same receptor as ABA, while
compound 1 may interact with a different target.
2.3. SAR of ABA derivatives
The methyl ester of ABA (analog 7, Fig. 2) was obtained in near
quantitative yield by using the direct reaction of ABA with
dimethyl sulfate.28 Analog 7 per se induced an increase of the
[cAMP]i of granulocytes and appeared to have a slightly additive
effect if administered together with ABA (Fig. 3A).
The tert-butyl ester (analog 7b) obtained by using tert-butylac-
etate and HClO4 per se induced an increase of the [cAMP]i similarly
to methyl ester but without an additive effect if administered
together with ABA (Fig. 3A). Also analogs 7 and 7b were tested at
2.3.1. Modifications of functionality at C-1
To evaluate the relevance of the carboxylic group of ABA to its
biological function, we synthesized a panel of different analogs
where the carboxylic group was modified. Most of the molecules
synthesized and tested were modified only in the pentadienoic
side-chain, without varying the cyclohexenoic ring.
In a first series of analogs, the carboxyl group was changed into
an amide (Fig. 2, analogs 1, 2, 3, 4), introducing aliphatic or aro-
matic amino groups. The amide modification of the C1 carboxylic
group of ABA has apparently never been tested on plant cells,
unlike C1 derivatives with ester, alcohol or aldehydic groups.18,19
These amide derivatives were synthesized with a simple reaction
between the carboxylic group, activated with an uronium salt,
and the amino group of the various amines. Collectively, the
ABA-amide analogs behaved as ABA agonists (Fig. 3A), except for
analog 4, which did not affect the [cAMP]i per se, nor did it antag-
onize the increase of the [cAMP]i induced by ABA. In fact, the addi-
the concentration of 10 lM and 1 lM and, at this concentration
they lost the slight agonistic activity shown at 50
lM (data not
shown).
Esters in general and methyl esters in particular may undergo
hydrolysis by cellular esterases to produce the free acid. However,
the very brief incubation time of granulocytes with the analog
(1 min), and the apparently additive effect of analog 7 and ABA,
make this possibility unlikely.
To obtain an alcoholic group at the C-1 position, the unstable
acid chloride of ABA was reduced with NaBH4 at low tempera-
ture.28 This alcohol (analog 8) induced a [cAMP]i increase similar
to that observed with ABA (Fig. 3A) and for this reason no further
experiments were performed at lower concentrations.
tion of ABA to the analog 4 at the concentration of 50
and 1 M in the reaction mixture is able to restore the original
effect of ABA (data not shown).
lM, 10 lM
l
Moreover, our attention was focused on compounds 1, 2 and 3,
which showed the strongest agonist activity at the concentration
In plants, abscisyl alcohol can be converted to ABA;39 it is pos-
sible that the same oxidation takes place also in granulocytes,
although the brief incubation time of the cAMP assay (1 min)
argues against a significant analog transformation.
of 50
lM. For these molecules, the same effect was confirmed at
10 M (data not shown) and at 1
l
l
M (Fig. 3C), when tested in

M. Bellotti et al. / Bioorg. Med. Chem. 23 (2015) 22–32
25
Figure 3. Screening of synthetic ABA analogs for their ability to inhibit the ABA-induced increase of the [cAMP]i in human granulocytes at 50
l
M and 1
lM. Granulocytes
were incubated for 1 min without (control) or with 1 lM ABA. 50 lM (panel A) or 1 lM (panel C) of the indicated analog was tested, in the presence or in the absence of 1 lM
ABA. In panels A and C results are expressed as [cAMP]i increase relative to control, untreated cells are mean from triplicate incubations (SD <5% of the mean for all values).
The black bar indicates the effect of ABA alone. Panel B. Results are expressed as percentage stimulation or inhibition of the [cAMP]i increase induced by ABA in the presence
of the indicated analogs at 50 lM, relative to that induced by ABA alone. Analogs behaving as ABA-antagonists are highlighted in grey.
We also tested a derivative in which the carboxylic group was
substituted with an aldehyde; its effect on the [cAMP]i was similar
to that of ABA (data not shown).
reaction with tert-butylbromoacetate in an anhydrous organic sol-
vent allowed to obtain the desired compound with good yield.29
Interestingly, analog 6 at 50
granulocytes and it abrogated the [cAMP]i rise induced by 1
ABA (Fig. 3A); conversely, analog 9 per se induced a [cAMP]i
increase higher than that elicited by ABA alone also at 10 M (data
not shown) and 1 M (Fig. 3C).
Besides, when analog 9 was tested at the concentration of 1
lM did not increase the [cAMP]i in
To investigate whether a modification of the pentadienoic chain
could confer ABA-antagonistic activity, two derivatives were syn-
thesized (analogs 6 and 9) in which this side chain was modified
by lengthening of the backbone. In particular, a linker was intro-
duced, containing carbon and oxygen atoms and ending with a car-
boxylic group. Analog 6, which has a free carboxylic group, was
obtained by reaction of ABA with bromoacetic acid. In the case of
derivative 9, which has a protected carboxylic group, a simple
lM
l
l
lM
in the presence of an equal concentration of ABA, the increase of the
[cAMP]i was comparable to that observed with ABA alone (Fig. 3C),
a behavior similar to that described above for analogs 2 and 3.

26
M. Bellotti et al. / Bioorg. Med. Chem. 23 (2015) 22–32
The antagonistic activity of analog 6 on the [cAMP]i rise induced
by 1 M ABA was negligible at 10 M analog and absent at 1
(data not shown).
2.3.3. Effect of the modification of the 40-carbonyl oxygen
l
l
lM
We previously reported that ABA biotinylated at the C40 keto
group binds to granulocyte membranes and that this effect is
antagonized by excess of unmodified ABA.3
Elongation of the ABA side-chain diminished the ability of the
analog to increase the [cAMP]i, but only in the presence of a free
acid in the same chain. Thus, it seems likely that one of the require-
ments for activity is the presence of a double bond conjugated to
the carboxylic group. In analog 9, the side chain is longer but the
presence of tert-butyl group increases the hydrophobicity of the
molecule. This feature seems to promote a significant [cAMP]i
increase (Fig. 3A). This result also highlights the importance of
the orientation of the ABA carboxyl group.
In analog 5, the pentadienoic chain was substituted with a five
ring cycle (Fig. 2). This analog does not have the –OH group in C10
that is reported to be critical for biological activity in plants,18,19
but contains a carbonyl group in C40. This compound, derived from
the reaction of ABA with pyrophosphoric acid,34 had a similar
effect on granulocyte [cAMP]i as ABA at all concentrations tested,
even though the side chain is devoid of the carboxyl group and
has a fairly hydrophobic lactone cycle.
The 40-carbonyl oxygen on the ABA ring was selectively reduced
to a hydroxyl group with NaBH4, generating the 10,40 diol (analog
13). The final product was a racemic mixture of the trans- and of
the cis-form derivatives.32 The racemic mixture and the purified
cis and trans isomers of 10,40-diol ABA all showed a similar effect
on the granulocyte [cAMP]i as ABA (Fig. 3A).
The same results were obtained at 10 lM and 1 lM (not
shown). This result was surprising, given that the keto-to-hydroxyl
reduction and the consequent absence of the double bond signifi-
cantly modify the conformation of the ABA cyclohexene ring. The
similar effect of ABA and of its 10,40-diol-derivative on granulocyte
[cAMP]i levels suggested that the C40 carbonyl group is not
essential for ABA activity. To further investigate this hypothesis,
the cis and trans 10,40-diols were selectively acetylated with acetic
anhydride in pyridine, obtaining the corresponding 40-O-acetates
(analog 14).26 The racemic mixture and the purified cis- and
trans-isomers of the 40-acetate analog all showed, as expected, a
similar ABA-like activity on granulocyte [cAMP]i (Fig. 3A).
2.3.2. Effect of the modification of the configuration at C-20 and
C-30
Other derivatives containing different groups at the C40 position
(e.g., ketimine or secondary amine groups) were also synthesized
and tested, but none behaved as an ABA-antagonist or agonist on
granulocyte [cAMP]i (data not shown). As an example, the com-
pound obtained by the reaction of ABA and aniline followed by
reductive amination with sodium cyanoborohydride27 (analog
14b) showed an ABA-like activity on granulocyte [cAMP]i (Fig. 3A).
Together, these results suggest that the 40-carbonyl group of
ABA is not fundamental for its effect on the [cAMP]i.
With this group of analogs, we tried to determine how the
degree of ABA ring unsaturation affects the biological activity of
the hormone.
Analog 10 contains a 20–30epoxy bridge, which eliminates the
unsaturation in the ring. Epoxidation of ( )-ABA with H2O2 in an
alkaline solution produced a racemic mixture of one diastereoiso-
mer, because addition of the hydrogen peroxide occurs only at the
less hindered a
-face, opposite to the side chain.30 Analog 10 com-
pletely inhibited the ABA-induced [cAMP]i rise (Fig. 3A), in line with
previously published results.40 Analog 10 has been already shown to
compete with ABA for binding to recombinant LANCL2, with an IC50
value of 6.56 nM, and to inhibit ABA-induced chemotaxis, phagocy-
tosis, ROS and PGE2 production in human granulocytes.40
Given the unusual behavior of analog 10, which appears to
antagonize the pro-inflammatory effects of ABA on granulocytes,
several other derivatives containing an epoxy bridge were synthe-
sized and tested on granulocyte cAMP. Analogs 11 and 12 (Fig. 1)
were obtained by the direct oxidation of alpha- and beta-ionone,
respectively, with m-chloroperoxybenzoic acid,35 and showed
two different actions. Analog 11, where the epoxy group is in the
same position as in analog 10 (C20–C30), behaved as an ABA-antag-
2.3.4. Effect of the modification of the ring saturation
Several ABA analogs containing a cyclohexane ring have been
tested on plants, but none containing an aromatic ring.18–22 In ana-
log 15 (Fig. 2), the cyclohexene is substituted by a benzene moiety.
Treatment of ABA methyl ester with acetic anhydride and toluene-
p-sulfonic acid led to the simultaneous aromatization of the
original cyclohexene ring and to the introduction of an acetate
group at position 40 with a previously described mechanism.31 This
ABA analog retains the 70- and 80-methyl groups, but lacks the
10-hydroxyl and the 90-methyl group of ABA that are believed to
play a significant role in ABA activity in plants.18–22 This derivative
behaved as a very weak agonist on granulocyte [cAMP]i and did
not alter the effect of ABA (Fig. 3A). For this reason no further
experiments were performed with analog 15 at lower concentrations.
Thus, presence of the aromatic ring apparently prevents analog
15 from competing with ABA in human granulocytes: in fact, the
geometry of the ring is profoundly modified, possibly preventing
binding of this analog to the receptor.
onist at 1 lM too, inhibiting the ABA-induced [cAMP]i increase in
granulocytes similarly to analog 10 (data not shown); conversely,
analog 12, in which the epoxy group is in a different position of
the ring (C10–C20), did not antagonize the effect of ABA, and rather
induced a [cAMP]i increase of its own at 50
ity was completely lost at 10 M and 1 M (not shown). The antag-
onistic activity of analog 11 on the [cAMP]i rise induced by 1
lM (Fig. 3A). This activ-
l
l
lM
ABA was negligible at 10nM analog (not shown), unlike that of ana-
log 10, which had an EC50 of 0.7 0.2 nM.40 However, analog 11
lacks the keto group in C40 and the pentadienoic side chain with
the carboxylic group, which are present in analog 10. In fact, the
1-carboxyl group seems to have a specific electronic interaction
with the receptor and shortening of this side-chain of ABA seri-
ously diminishes activity.
These results suggest that the ABA-antagonistic activity of ana-
log 10 is due not only to the presence of the epoxy group in the
ring, but also to presence of the C40 keto group and of the pentadi-
enoic side chain with a free carboxyl group. The antagonistic effect
of analogs 10 and 11 may be due to steric reasons as the epoxy ring
in both cases retains the methyl group at C20, coplanar with the
ring, but at the same time apparently creates a new point of
interaction with the receptor.
2.3.5. Effect of the concomitant modification of different
positions in the ABA structure
So far, we examined the activity of many compounds, with the
exclusion of analog 15, that harbored a single variation of the ABA
molecule.
Next, we synthesized several analogs containing modifications
in different positions of the ABA molecule. Firstly, functionalities
at C1 and C40 were concomitantly modified: in analogs 16 and 17
(Fig. 2) the ketone group at C40 was substituted with a hydroxyl
group and the carboxylic function at C1 was substituted with an
ester. Neither compound had any effect on the [cAMP]i nor did it
show any inhibitory activity on the ABA-induced [cAMP]i increase
in granulocytes (Fig. 3A). Analog 18 (Fig. 2), obtained by the selec-
tive reduction of the ABA methyl ester in C1 and of the carbonyl
group in C40 with diisobutylaluminium hydride,28 showed an

M. Bellotti et al. / Bioorg. Med. Chem. 23 (2015) 22–32
27
ABA-like activity on granulocyte [cAMP]i, similar to that of analog
7 (ABA methyl ester), that is perhaps explainable with the oxida-
tion to acid or with the presence of the electrostatic characteristic
of the alcohol. These slight activities were not confirmed using
and length of the side chain and of the presence of a keto group
in 40. For this reason, analog 11, where the epoxy group is in the
same position as in analog 10 (20–30), also behaved similarly as
an ABA-antagonist, although it lacked the keto group and had a
shortened side-chain. On the contrary, analog 12, in which the
epoxy group is in a different position of the ring, between C10and
C20, but which shares a ionone-like backbone with analog 11, does
not antagonize the effect of ABA, instead inducing itself a [cAMP]i
analogs 16, 17, 18 at 10 lM or 1 lM (data not shown).
As the presence of the epoxy group confers an antagonistic
activity to analogs 10 and 11, other derivatives containing this
modification were synthesized. Due to the reactivity of the epoxy
group, the only part of the ABA molecule that could be concomi-
tantly modified, without altering the epoxy functionality, was the
carboxyl group in C1.41 We tested several derivatives containing
the epoxy group in C20 and C30 as well as ester substitutions of
the carboxyl group. Almost all derivatives tested indeed behaved
as ABA antagonists on granulocyte [cAMP]i, although when tested
increase. At 1 lM, the antagonistic activity of analog 11 on the
[cAMP]i rise induced by ABA was lower than that of analog 10
(data not shown). Thus, it appears that the ABA-antagonistic effect
is more pronounced if the pentadienoic chain and the C-40ketone
group are left unchanged.
These results are in agreement with a recent paper where sub-
stitution in the 30 position of abscisic acid’s ring produced a potent
ABA antagonist that was sufficiently active to block stress-induced
ABA responses in vivo on plants.42
at concentrations lower than 50 lM, they were all less active than
analog 10 (data not shown). These results, however, confirm that
the epoxy modification in C20–C30 is critical for ABA-antagonism.
Once the tridimensional structure of ABA receptor(s) in mam-
mals is defined, it will be possible to verify the results and conse-
quent predictions reported above and to probe the binding pockets
with structurally defined compounds and hence design new deriv-
atives to fit different binding sites. Given the present state of
knowledge, we propose that these analogs, in particular 10 and
11 could be useful for biochemical studies aimed at identifying
novel putative compounds for the treatment of inflammatory
diseases.
3. Conclusion
We described the synthesis, structural characterization and bio-
logical activity ‘in vitro’ of a series of derivatives of ABA, in an
attempt to identify the molecular determinants associated with
ABA-agonistic or antagonistic effects on human granulocytes. Thus,
the compound library was tested on human granulocytes, to find
ABA analogs capable of inhibiting the [cAMP]i rise induced by
ABA. The systematic modification of the various parts of the mole-
cule aimed to clarify which functional groups of ABA are important
in determining its biological effect.
4. Experimental
4.1. General
Specifically, results obtained by modifying the ABA structure
only at the C40 suggest that presence of the carbonyl group is not
essential for ABA-like activity on granulocyte [cAMP]i.
The analysis of the intermediates and the raw products was per-
formed by liquid chromatography-electrospray mass spectrometry
(HPLC–ESI–MS) using an Agilent 1100 series LC/MSD ion trap
instrument. The products were then purified by preparative
reverse phase high performance liquid chromatography (RP-HPLC)
on a Shimadzu LC-9A preparative HPLC equipped with a Phenom-
enex C18 Luna column (21.20 ꢁ 250 mm). The separation was per-
formed in gradient starting with 15% solvent B for 5 min, linearly
increasing to 70% solvent B in 30 min and up to 100% B in
10 min. The solvent used was 0.1% trifluoroacetic acid (TFA) in
water (A) and 0.1% TFA in acetonitrile (B).
All the analogs in which the carbonyl group is present and con-
jugated with a 20 30 double bond (analog 1, 2, 3, 5, 7, 7b, 8, 9) show
ABA-like activity at 50
log 1, 2, 3, 9 also at 10
derivatives do not contain this keto group (13, 14, 14b, 18) but
have, as a common feature, a polar chain containing carboxylic acid
or alcohol group that allow an increase of the [cAMP]i even if this
l
M. This action was confirmed only for ana-
l
M (not shown) and 1 M (Fig. 3C). Four
l
activity is evident only at 50 lM.
Analog 14, in which this carbonyl is substituted with an acetyl
group, retains the ability to stimulate the production of [cAMP]i, as
the acetyl group allows conjugation with the cyclohexenic double
bond.
The molecular weight of the products were finally confirmed by
electrospray mass spectrometry using an ion trap analyzer that
allowed to perform MS and MS2 analysis.
Particularly interesting is the presence of the alcohol group in
position 40. In fact, alcohol group in position 40 differently affects
the [cAMP]i, depending on the chemical–physical characteristics
of the group in C1. Presence of a hydrophobic group in C1 (such
as the methyl and tertbutyl esters in analogs 16 and 17, respec-
tively) abrogates stimulation of [cAMP]i production, while the
increase of the [cAMP]i is maintained if C1 is polar, carboxylic or
alcoholic (analogs 13 and 18) only at high analog concentration.
If the carbonyl group in 40 is preserved, the side chain deter-
mines the activity of the analogs tested. Elongation of the side
chain with hydrophobic esters or amides ensures an ABA-like
activity, as demonstrated in particular for analogs 1, 2, 3 and 9.
On the contrary, the presence of a free carboxylic group (analog
6) or the introduction of the polar amino group as substituent in
the amide (analog 4) abrogate the effect on the [cAMP]i albeit only
1H NMR and 13C NMR spectra were recorded on a Varian Mer-
cury Plus at 300 MHz and 75 MHz, respectively. Chemicals shifts
are reported as d values (ppm).
4.2. Chemicals
Synthetic ( )-ABA and the natural (S)-(+)-ABA are available
commercially and both the resolution and HPLC separation of the
racemic mixture have been reported.36
However, because of the high cost, the use of (+)ABA as a start-
ing material for analogue synthesis has been impractical and, until
now, most ABA analogues have been prepared by total synthesis
starting from of the racemic form as ( )-ABA.
at 50
lM.
4.3. Synthesis of abscisic acid analogs
Among the analogs tested, only those where the configuration
at C-20 and C-30 was modified by introduction of an epoxidic ring
abrogated the [cAMP]i increase induced by ABA on granulocytes.
The introduction of an epoxidic group in the position C-20 and
C-30 conferred the ABA-antagonistic effect regardless of the type
4.3.1. (2Z,4E)-N,N-Diethyl-5-(10-hydroxy-20,60,60-trimethyl-40-
oxo-20-cyclohexen-10-yl)-3 methylpenta-2,4-dienamide (1)
( )-Abscisic acid (10 mg, 0.038 mmol, 1 equiv) was dissolved in
1 ml of N,N dimethylformamide and O-(7-azabenzotriazol-1-yl)-

28
M. Bellotti et al. / Bioorg. Med. Chem. 23 (2015) 22–32
1,1,3,3 tetramethyluroniumhexafluorophosphate (HATU) (14.5 mg,
0.038 mmol, 1 equiv) and N,N-Diisopropylethylamine (DIPEA)
vacuum and the residue was dissolved in deionized water. The
pH was adjusted to 9 with addition of NaOH 0.1 N and the solution
was extracted with diethyl ether (3 ꢁ 5 ml). The organic solution
was finally evaporated under vacuum. The residue was purified
by RP-HPLC to obtain the title compound (12.7 mg, yield of 46%)
as brown oil.
(13
ture at room temperature in the dark. After few minutes a solution
of diethylamine (8 l, 0.076 mmol, 2 equiv) was added and the
ll, 0.076 mmol, 2 equiv) were slowly added to the stirred mix-
l
reaction mixture was mixed at room temperature, protected from
light. When reaction was complete, about 24 h, the solvent was
evaporated under vacuum and the residue was dissolved in
deionized water. The pH was adjusted to 9 with addition of NaOH
0.1 N and the solution was extracted with diethyl ether (3 ꢁ 5 ml).
The organic solution was finally evaporated under vacuum. The
residue was purified by reverse phase high performance liquid
chromatography (RP-HPLC) to give the title compound (8 mg, yield
of 66%) as oil.
ESI–MS m/z: 369.2.
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.93 [s, 3H, 90]; 0.96
[s, 3H, 80]; 1.84 [d, 3H, 70, J = 1.4 Hz]; 1.98 [d, 3H, 6, J = 1.0 Hz]; 2.09
[d, 1H, 50, J = 16.7 Hz], 2.53 [d, 1H, 50, J = 16.7 Hz]; 3.71 [s, 3H,
OCH3] 5.19 [s, 1H, OH]; 5.84–5.79 [m, 1H, 30]; 5.88 [m, 1H, 2];
6.16 [d, 1H, 5, J = 16.0 Hz], 6.48 [d, 1H, 4, J = 16.0 Hz]; 6.91–6.83
[m, 2H, CH]; 7.58–7.50 [m, 2H, CH]; 9.87 [s, 1H, –NH].
13C NMR: DMSO-d6, T = 25 °C; d 18.96 [70]; 21.08 [6]; 23.24 [80];
24.22 [90]; 41.35 [60]; 49.41 [50]; 55.17 [O–CH3]; 78.44 [10];
113.84[CH phenyl]; 120.57[CH phenyl]; 121.46 [2]; 125.84 [30];
127.78 [4]; 132.59[C–NH]; 136.09 [5]; 145.39 [3]; 155.12[C–O–
CH3]; 163.55 [20]; 163.67 [1]; 197.35 [40].
ESI–MS m/z: 319.3.
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.97 [s, 3H, 80]; 0.96
[s, 3H, 90]; 1.35 [m, 6H, CH3]; 1.93 [s, 3H, 70]; 1.98 [d, 3H, 6,
J = 1.2 Hz]; 2.28 (d, 1H, 50, J = 15.4 Hz] 2.74 [d, 1H, 50, J = 15.5 Hz];
3.75 [m, 4H, N–CH2]; 3.64 [s, 1H, OH]; 5.91 [s, 1H, 5]; 6.08 [d,
1H, 30, J = 16.2 Hz]; 6.48 [d, 1H, 4, J = 16.2 Hz]; 6.93 [d, 1H, 2,
J = 16 Hz].
4.3.4. (2Z,4E)-N-(4-aminophenyl)-5-(10-hydroxy-20,60,60-
trimethyl-40-oxo-20-cyclohexen-10-yl)-3-methylpenta-2,4-
dienamide (4)
( )-Abscisic acid (10 mg, 0.038 mmol, 1 equiv) was dissolved in
0.5 ml of N,N dimethylformamide and HATU (14 mg, 0.038 mmol,
13C NMR: (DMSO-d6), T = 25 °C; d 12.95[–CH3]; 18.85 [70]; 20.86
[6]; 23.19 [80]; 24.16 [90]; 41.31 [60]; 42.45 [–N–CH2]; 49.31[50];
50.87 [O–CH3]; 78.41 [10]; 116.63 [2]; 126.04 [30]; 126.99 [4];
138.45 [5]; 150.77 [3]; 163.05 [20]; 165.69 [1]; 197.21 [40].
1 equiv) and DIPEA (13 ll, 0.076 mmol, 2 equiv) were slowly added
to the stirred mixture in the dark at room temperature. After a few
minutes, a solution of 1,4 phenylendiamine (16 mg, 0.076 mmol,
2 equiv) in 300 ll of N,N dimethylformamide was added and the
4.3.2. (2Z,4E)-5-(10-Hydroxy-20,60,60-trimethyl-40-oxo-20-
cyclohexen-10-yl)-3-methyl-N-phenylpenta-2,4-dienamide (2)
( )-Abscisic acid (10 mg, 0.038 mmol, 1 equiv) was dissolved in
1 ml of N,N dimethylformamide and HATU (14.5 mg, 0.038 mmol,
reaction mixture was mixed at room temperature, protected from
light, overnight monitoring the reaction with thin layer chroma-
tography (TLC) (ethanol/ethyl acetate/ammonia solution 25%
10:90:0.25).
When reaction was complete the solvent was evaporated under
vacuum and the residue was dissolved in deionized water. The
solution was extracted with diethyl ether (3 ꢁ 2 ml). The aqueous
solution was finally extracted with ethyl acetate (EtOAc)
(3 ꢁ 2 ml). The organic solution was then evaporated under
vacuum. The residue was purified by RP-HPLC to obtain the title
compound (7 mg, yield of 52%) as oil.
1 equiv) and DIPEA (13
to the stirred mixture in the dark at room temperature. After few
minutes a solution of aniline (6.8 l, 0.076 mmol, 2 equiv) was
ll, 0.076 mmol, 2 equiv) were slowly added
l
added and the reaction mixture was mixed at room temperature,
protected from light, for about 48 h.
When reaction was complete, the solvent was evaporated under
vacuum and the residue was dissolved in deionized water. The pH
was adjusted to 9 with addition of NaOH 0.1 N and the solution
was extracted with diethyl ether (3 ꢁ 5 ml). The organic solution
was finally evaporated under vacuum. The residue was purified
by RP-HPLC to give the title compound (7 mg, yield of 54%) as oil.
ESI–MS m/z: 339.2.
ESI–MS m/z: 354.2.
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.93 [s, 3H, 90]; 0.96
[s, 3H, 80]; 1.84 [d, 3H, 70, J = 1.4 Hz]; 1.98 [d, 3H, 6, J = 1.0 Hz]; 2.09
[d, 1H, 50, J = 16.7 Hz], 2.53 [d, 1H, 50, J = 16.7 Hz]; 5.19 [s, 1H, OH];
5.84–5.79 [m, 1H, 30]; 5.88 [m, 1H, 2]; 6.16 [d, 1H, 5, J = 16.0 Hz];
6.48 [d, 1H, 4, J = 16.0 Hz]; 6.50–6.58 [m, 2H, CH]; 6.29 [s, 2H,
NH2]; 7.41–7.43 [m, 2H, CH]; 9.87 [s, 1H, –NH].
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.93 [s, 3H, 90]; 0.96
[s, 3H, 80]; 1.84 [d, 3H, 70, J = 1.4 Hz]; 1.98 [d, 3H, 6, J = 1.0 Hz]; 2.09
[d, 1H, 50, J = 16.7 Hz], 2.53 [d, 1H, 50, J = 16.7 Hz]; 5.19 [s, 1H, OH];
5.84–5.79 [m, 1H, 30]; 5.88 [m, 1H, 2]; 6.16 [d, 1H, 5, J = 16.0 Hz],
6.48 [d, 1H, 4, J = 16.0 Hz;] 7.19 [m, 1H, CH]; 7.41–7.43 [m, 2H,
CH]; 7.50–7.58 [m, 2H, CH]; 9.87 [s, 1H, –NH].
13C NMR: DMSO-d6, T = 25 °C; d 18.96 [70]; 21.08 [6]; 23.24 [80];
24.22 [90]; 41.35 [60]; 49.41 [50]; 78.44 [10]; 113.84 [CH phenyl];
120.57 [CH phenyl] 121.46 [2]; 125.84 [30]; 127.78 [4];
132.48[C–NH]; 136.09 [5]; 145.39 [3]; 155.12 [CH–NH2]; 163.55
[20]; 163.67 [1]; 197.35 [40].
13C NMR: DMSO-d6, T = 25 °C; d 18.96 [70]; 21.08 [6]; 23.24 [80];
24.22 [90]; 41.35 [60]; 49.41 [50]; 78.44 [10]; 113.84[CH phenyl];
120.57[CH phenyl]; 121.46 [2]; 125.84 [30]; 127.78 [4]; 132.59
[C–NH]; 136.09 [5]; 145.39 [3]; 153.12 [CH phenyl]; 163.55 [20];
163.67 [1]; 197.35 [40].
4.3.5. (Z)-4-Methyl-5-((20,60,60-trimethyl-40-oxo-20-cyclohexen-10-yl)
methylene)furan-2(5H)-one (5)
4.3.3. (2Z,4E)-5-(10-Hydroxy-20,60,60-trimethyl-40-oxo-20-
cyclohexen-10-yl)-N-(4 methoxyphenyl)-3-methylpenta-2,4-
dienamide (3)
( )-Abscisic acid (20 mg, 0.075 mmol, 1 equiv), was dissolved in
2 ml of N,N dimethylformamide and HATU (30 mg, 0.075 mmol,
Pyrophosphoric acid (200 mg, 1.2 mmol, 10 equiv) was dis-
solved in 5 ml of methanol and the mixture was heated at 90° for
5 min. ( )-Abscisic acid was added to this solution (30 mg,
0.12 mmol, 1 equiv) and stirred at the same temperature until com-
plete dissolution. When the reaction was complete, it was allowed
to reach the ambient temperature and the solvent was evaporated
under vacuum. The residue was then purified by RP-HPLC to obtain
the title compound (23 mg, yield of 74%) as brown oil.
ESI–MS m/z: 246.1.
1 equiv) and DIPEA (26 ll, 0.15 mmol, 2 equiv) were slowly added
to the stirred mixture in the dark. After few minutes a solution of
p-anisidine (19 mg, 0.151 mmol, 2 equiv) was added and the reac-
tion mixture was mixed at room temperature, protected from light,
for about 48 h monitoring the reaction with HPLC analysis. When
the reaction was complete, the solvent was evaporated under
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; 0.97 [s, 3H, 90]; d 1.00
[s, 3H, 80]; 1.81 [d, 3H, 70, J = 1.4 Hz]; 2.21 [d, 3H, 6, J = 1.5 Hz]; 2.42

M. Bellotti et al. / Bioorg. Med. Chem. 23 (2015) 22–32
29
[d, 1H, 50, J = 17 Hz]; 2.6 (d, 1H, 50, J = 16.9 Hz]; 3.2 [d, 1H, 10,
J = 11 Hz]; 5.59 [d, 1H, 5, J = 11 Hz]; 5.88 [s, 1H, 30]; 6.20 [m, 1H, 2].
13C NMR: (DMSO-d6): T = 25 °C; d 11.4 [6]; 22.9 [70]; 27. 21 [80];
26.34 [90]; 36.2 [60]; 47.28 [50]; 125.21 [30]; 160.31 [20]; 48.7 [10];
116.12 [2]; 152.9 [4]; 109.8 [5]; 155.71 [3]; 167.95 [1]; 197.10 [40].
hydrochloric acid in water. The resultant aqueous solution was
adjusted to pH 9 by addition of 10% K2CO3, and then extracted with
diethyl ether (3 ꢁ 1 mL). The combined organic phases were dried
with anhydrous sodium sulfate, filtered and concentrated to give
an oil. The residue was purified by RP-HPLC to obtain the title com-
pound (5 mg, yield of 21%) as oil.
4.3.6. 2-(2-(((2Z,4E)-5-(10-Hydroxy-20,60,60-trimethyl-40-oxo-20-
cyclohexen-10-yl)-3-methylpenta-2,4-dienoyl)oxy)acetoxy)
acetic acid (6)
ESI–MS m/z: 320.4.
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.95 [s, 3H, 80]; 0.96
[s, 3H, 90]; 1.38 [m, 9H, (CH3)3]; 1.83 [s, 3H, 70]; 1.98 [d, 3H, 6,
J = 1.2 Hz]; 2.28 (d, 1H, 50, J = 15.4 Hz] 2.74 [d, 1H, 50, J = 15.5 Hz];
3.64 [s, 1H, OH]; 5.63 [d, 1H, 2, J = 16 Hz]; 5.91 [s, 1H, 5]; 6.08 [d,
1H, 30, J = 16.2 Hz]; 6.48 [d, 1H, 4, J = 16.2 Hz].
2-bromoacetic acid (14 ll, 0.1 mmol, 2 equiv) and K2CO3
(24 mg, 0.35 mmol, 7 equiv) were added to a solution of ( )-ABA
(13 mg, 0.05 mmol, 1 equiv) in 2 ml of dry N,N dimethylformam-
ide. The reaction was mixed at room temperature, protected from
light for 24 h, monitoring the reaction with TLC (ethanol/ethylace-
tate/acetic acid 10:90:0.25). When the reaction was complete, the
solvent was evaporated under vacuum and the residue was dis-
solved in deionized water. The pH was adjusted to 7 and the solu-
tion was extracted with diethyl ether (3 ꢁ 3 mL). The organic
extracts were dried with anhydrous sodium sulfate and filtered.
The organic solution was evaporated under vacuum. The residue
was purified by RP-HPLC to obtain the title compound (6 mg, yield
of 32%) as white powder.
13C NMR: (DMSO-d6): T = 25 °C; d 18.85 [70]; 20.86 [6]; 23.19
[80]; 24.16 [90]; 28.7 [(CH3)3] 41.31 [60]; 49.31[50]; 78.41 [10];
80.98 [O–C]; 116.63 [2]; 126.04 [30]; 126.99 [4]; 138.45 [5];
150.77 [3]; 163.05 [20]; 165.69 [1]; 197.21 [40].
4.3.9. (2Z,4E)-5-(10-Hydroxy-40-oxo-20,60,60-trimethyl-20-
cyclohexenyl)-3-methyl-pentadien-1-ol (8)
Oxalyl chloride (0.025 ml, 0.28 mmol, 2 equiv) was added drop-
wise to a solution of ( )-ABA (50 mg, 0.19 mmol, 1 equiv) in 2 ml of
dry N,N dimethylformamide at T = 10 °C. After 30 min the reaction
was stopped and the solvent was evaporated under vacuum. The
residue was dissolved in 1 ml of dry tetrahydrofuran (THF) and
NaBH4 (55 mg, 1.4 mmol, 1 equiv) was added at T = ꢀ55 °C to the
ESI–MS m/z: 380.1.
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.92 [s, 3H, 80]; 0.95
[s, 3H, 90]; 1.81 [d, 3H, 70, J = 1.3 Hz]; 2.04 [d, 3H, 6, J = 1.2 Hz]; 2.10
[d, 1H, 50, J = 17.0 Hz]; 2.54 [d, 1H, 50, J = 17.0 Hz]; 4.94 [s, 2H,
O–CH2–CO]; 5.11[s, 2H, O–CH2–CO]; 5.26 [s, 1H, OH]; 5.85–5.79
[m, 2H, 30 e 2]; 6.35 [d, 1H, 5, J = 15.9 Hz]; 7.71 [d, 1H, 4, J = 15.9,
0.6 Hz]. 11.00 [s, 1H, OH].
stirred solution. When the reaction was complete, 100 ll of
0.1 M acetic acid was added to the mixture. The reaction was
poured into cold water (1 mL), and finally extracted with EtOAc
(3 ꢁ 3 mL). The EtOAc extract was dried with anhydrous sodium
sulfate, filtered and concentrated under vacuum to afford the final
product.
13C NMR: (DMSO-d6): T = 25 °C; d 18.82 [70]; 20.94 [6]; 24.19
[80]; 27.68 [90]; 41.32 [60]; 49.31 [50]; 62.60 [O–CH2–]; 63.70
[O–CH2–]; 78.40 [10]; 115.82 [2]; 126.08 [30]; 126.91 [4]; 139.07
[5]; 152.25 [3]; 162.89 [20]; 164.52 [1]; 166.99 [C@O]; 171.99
[C@O]; 197.18 [40].
The residue was purified by RP-HPLC to obtain the title com-
pound (18 mg, yield of 37%) as yellow oil.
ESI–MS m/z: 250.2.
1H NMR (CDCl3, 300 MHz): T = 25 °C; d 0.97 [s, 3H, 80]; 0.99 [s,
3H, 90]; 1.85 [3H, s, 70]; 1. 98 [s 3H, 6]; 2.26 [d, 1H, 50, J = 17 Hz],
2.45 [d, 1H, 50, J = 17 Hz]; 4.29 [d, 2H, –CH2–OH, J = 6.5 Hz]; 5.25
[s, 1H, OH]; 5.62 [t, 1H, 2, J = 6.5 Hz] 5.80 [d, 1H, 5, J = 15.5 Hz];
5.85 [m, 1H, 30]; 5.90 [s, –OH, 1H,]; 6.76 [d, 1H, 4, J = 15.5 Hz].
4.3.7. (2Z,4E)-methyl5-(10-hydroxy-20,60,60-trimethyl-40-oxo-20-
cyclohexen-10-yl)-3-methylpenta-2,4-dienoate (7)
A stirred solution of ( )-ABA (50 mg, 0.189 mmol, 1 equiv) dim-
ethylsulfate (60 mg, 0.473 mmol, 2.5 equiv) and K2CO3 (104 mg,
0.756 mmol, 4 equiv) in acetone (4 ml) was heated under reflux
for 1 h. The reaction was allowed to warm to ambient temperature,
extracted with ethyl acetate and the combined organic layer
washed with a solution of ammoniumhydrogencarbonate 0.1 M.
The organic phase was dried with anhydrous sodium sulfate, fil-
tered, concentrated under vacuum and liophylized to afford a
white solid.
4.3.10. (2Z,4E)-2-(tert-butoxy)-2-oxoethyl-5-(10-hydroxy-20,60,60-
trimethyl-40-oxo-20cyclohexen-10-yl)-3-methylpenta-2,4-
dienoate (9)
Tertbutyl-2-bromoacetate (100 ll, 0.605 mmol, 4 equiv) and
K2CO3 (48 mg, 0.35 mmol, 2.3 equiv) were added to a solution of
( )-ABA (40 mg, 0.151 mmol, 1 equiv) in 2 ml of dry N,N dimethyl-
formamide. The reaction was mixed at room temperature, pro-
tected from light, up to completeness. The solvent was
evaporated under vacuum and the residue was dissolved in about
3 ml of deionized water. The pH was adjusted to 9 with addition of
NaOH 0.1 N and the solution was extracted with EtOAc (3 ꢁ 3 mL).
The EtOAc extract was dried with anhydrous sodium sulfate and
filtered. The organic solution was evaporated under vacuum. The
residue was purified by RP-HPLC to obtain the title compound
(35 mg, yield of 61%) as powder. ESI–MS m/z: 378.20.
The residue was purified by RP-HPLC to obtain the title com-
pound (44 mg, yield of 85%) as powder.
ESI–MS m/z: 278.1.
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.92 [s, 3H, 90]; 0.96
[s, 3H, 80]; 1.82 [d, 3H, 70, J = 1.4 Hz]; 2.01[d, 3H, 6 J = 1.3 Hz]; 2.10
[d, 1H, 50, J = 16.6 Hz]; 2.53 [d, 1H, 50, J = 16.6 Hz]; 3.61 [s, 3H,
O–CH3]; 5.25 [s, 1H, OH]; 5.77–5.73 [m, 1H, 2]; 5.81–5.85 [m,
1H, 30]; 6.30 [d, 1H, 5, J = 15.9 Hz]; 7.70 [dd, 1H, 4, J = 15.9, 0.8 Hz].
13C NMR: (DMSO-d6): T = 25 °C; d 18.85 [70]; 20.86 [6]; 23.19
[80]; 24.16 [90]; 41.31 [60]; 49.31 [50]; 50.87 [O–CH3]; 78.41 [10];
116.63 [2]; 126.04 [30]; 126.99 [4]; 138.45 [5]; 150.77 [3]; 163.05
[20]; 165.69 [1]; 197.21 [40].
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.92 [s, 3H, 80]; 0.95
[s, 3H, 90]; 1.41 [s, 9H, 3(3HC)CO–]; 1.82 [d, 3H, 70, J = 1.4 Hz]; 2.04
[d, 3H, 6, J = 1.2 Hz]; 2.10 [d, 1H, 50, J = 17.0 Hz]; 2.54 [d, 1H, 50,
J = 17.0 Hz]; 4.54 [s, 2H, O–CH2–CO]; 5.26 [s, 1H, OH]; 5.85–5.79
[m, 2H, 30 e 2]; 6.35 [d, 1H, 5, J = 15.9 Hz]; 7.71 [dd, 1H, 4,
J = 15.9, 0.6 Hz].
4.3.8. (2Z,4E)-tert-butyl 5-(10-hydroxy-20,60,60-trimethyl-40-oxo-
20-cyclohexen-10-yl)-3-methylpenta-2,4-dienoate (7b)
HClO4 (7
tion of ( )-ABA (20 mg, 0.076 mmol, 1 equiv) in tert-butyl acetate
(600 l, 3.8 mmol, 50 equiv) at T = 0 °C. The reaction mixture was
stirred at room temperature for 12 h, then washed with 1 N
ll, 0.12 mmol, 1.5 equiv) was slowly added to a solu-
13C NMR: (DMSO-d6): T = 25 °C; d 18.82 [70]; 20.94 [6]; 24.19
[80]; 27.68 [90]; 41.32 [60]; 49.31[50]; 60.60 [O–CH2–]; 78.40 [10];
81.48 [(H3C)C]; 115.82 [2]; 126.08 [30]; 126.91 [4]; 139.07 [5];
152.25 [3]; 162.89 [20]; 164.52 [1]; 166.99 [C@O]; 197.18 [40].
l

30
M. Bellotti et al. / Bioorg. Med. Chem. 23 (2015) 22–32
4.3.11. (2Z,4E)-5-(2-hydroxy-1,3,3-trimethyl-5-oxo-7-
oxabicyclo[4.1.0]heptan-2-yl)-3-methylpenta-2,4-dienoic acid
(10)
The reaction was poured into water (5 mL), acidified to pH 2 with
1 N HCI and finally extracted with EtOAc (3 ꢁ 1 mL). The EtOAc
extract was dried with anhydrous sodium sulfate, filtered, and con-
centrated under vacuum to give a colorless solid.
30% H2O2 (250 ll, 8.09 mmol, 70 equiv) and 6 N NaOH (50 ll)
were added at 0 °C to a stirred solution of ( )-ABA (30 mg,
0.113 mmol, 1 equiv) in methanol (MeOH) (1 ml). The mixture
was stirred, protected from light, for 96 h at 0 °C and then diluted
with H2O.
After lowering the pH to 2 with 3 N HCI, the mixture was
extracted with EtOAc (3 ꢁ 3 ml). The organic layer was concen-
trated under reduced pressure. The residue was purified by RP-
HPLC to obtain the title compound (25 mg, yield of 76%) as yellow
oil.
The residue was purified by RP-HPLC to obtain two different
compounds (trans-diol ABA 4 mg, yield of 40%; cis-diol ABA
4.8 mg, yield of 47%) as oils.
ESI–MS m/z: 266.2.
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.92 [s, 3H, 90]; 0.98
[s, 3H, 80]; 1.65 [d, 3H, 70, J = 1.4 Hz]; 1.73 [d, 1H, 50, J = 16.6 Hz]
1.81 [d, 1H, 50, J = 16.6 Hz] 2.01 [d, 3H, 6 J = 1.3 Hz]; 2.80 [s, 1H,
OH]; 4.25 [m, 1H, 40]; 5.25 [s, 1H, OH]; 5.63 [m, 1H, 30]; 5.69 [m,
1H, 2]; 6.05 [d, 1H, 5, J = 15.9 Hz]; 7.70 [dd, 1H, 4, J = 15.9 Hz],
11.00 [s, 1H, OH].
ESI–MS m/z 280.1.
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.79 [s, 3H, 80]; 0.96
[s, 3H, 90]; 1.27 [s, 3H, 70]; 1.93 (d, 1H, 50, J = 15.4 Hz]; 1.98 [d, 3H, 6,
J = 1.2 Hz]; 2.74 [d, 1H, 50, J = 15.5 Hz]; 3.22 [s, 1H, 30]; 5.25 [s, 1H,
OH]; 5.66–5.68 [m, 1H, 2]; 6.22 [dd, 1H, 5, J = 15.8, 0.5 Hz]; 7.81
[dd, 1H, 4, J = 15.8, 0.7 Hz] 11.00 [s, 1H, OH].
1H NMR: (DMSO-d6, 300 MHz): T = 25 °C; d 0.90 [s, 3H, 90]; 1.00
[s, 3H, 80]; 1.63 [d, 3H, 70, J = 1.4 Hz]; 1.73 [d, 1H, 50, J = 16.6 Hz]
1.78 [d, 1H, 50, J = 16.6 Hz] 2.02[d, 3H, 6 J = 1.3 Hz]; 2.80 [s, 1H,
OH]; 4.22 [m, 1H, 40]; 5.25 [s, 1H, OH]; 5.67 [m, 1H, 30]; 5.69 [m,
1H, 2]; 6.11 [d, 1H, 5, J = 15.9 Hz]; 7.25 [dd, 1H, 4, J = 15.9 Hz].
11.00 [s, 1H, OH].
13C NMR: (DMSO-d6), T = 25 °C; d 19.65 [70]; 20.93 [6]; 24.92
[80]; 26.02 [90]; 41.55 [60]; 47.28 [50]; 62.11 [30]; 66.71 [20]; 76.15
[10]; 118.12 [2]; 126.9 [4]; 138.18 [5]; 149.31 [3]; 166.95 [1];
206.00 [40].
4.3.15. (2Z,4E)-5-(40-acetoxy-10-hydroxy-20,60,60-trimethyl 20-
cyclohexen-10-yl)-3-methylpenta-2,4-dienoic acid (14)
Trans ABA diol (27 mg, 0.096 mmol, 1 equiv) was dissolved in a
1:2 (v:v) mixture of acetic anhydride and pyridine (1 mL), and the
solution was stirred at T = 40 °C for about 3 h, monitoring the reac-
tion with TLC (ethanol/ethylacetate/acetic acid 10:90:0.25). The
reaction was poured into water, finally extracted with EtOAc
(3 ꢁ 1 mL). The EtOAc extract was dried with anhydrous sodium
sulfate, filtered, and concentrated under vacuum to give a colorless
solid. The residue was purified by RP-HPLC to obtain the title com-
pound (23 mg, yield of 82%) as colorless solid.
4.3.12. 4-(20,30-Epoxy-20,60,60-trimethylcyclohexyl)-3-buten-2-
one (11)
m-Chloroperoxybenzoic acid (13.4 mg, 0.078 mmol, 1.5 equiv)
was added to a solution of (+)-
a-Ionone (11
ll, 0.052 mmol,
1 equiv) in 600
bath.
ll of methylene chloride, being cooled in an ice
The reaction mixture was stirred at room temperature for 8 h
and then filtered and washed with aqueous ammonium hydrogen
carbonate 0.1 M. The organic phase was dried with anhydrous
sodium sulfate, and evaporated under reduced pressure. The resi-
due was purified by RP-HPLC to obtain a the title compound
(8.7 mg, yield of 80%) as an yellow oil.
ESI–MS m/z: 308.2.
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.90 [s, 3H, 90]; 0.94
[s, 3H, 80]; 1.56 [d, 3H, 7 J = 1.4 Hz]; 1.71 [d, 1H, 50, J = 16 Hz] 1.80
[d, 1H, 50, J = 16 Hz]; 1.91[d, 3H, 6 J = 1.1 Hz]; 1.97 [s, 3H, O–
COCH3]; 5.20 [m, 1H, 40]; 5.25 [s, 1H, OH]; 5.33 [m, 1H, 30]; 5.55
[m, 1H, 2]; 6.02 [d, 1H, 5, J = 15.9 Hz]; 7.61 [dd, 1H, 4, J = 15.9,
0.8 Hz]. 11.00 [s, 1H, OH].
ESI–MS m/z: 208.2.
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.76 [s, 3H, 90]; 0.94
[s, 3H, 80]; 1.01 [m, 1H, 50]; 1.26 [s, 3H, 70]; 1.40 [m, 1H, 50]; 1.80–
2.00 [m, 2H, 40]; 2.09 [d, 1H, 10, J = 10 Hz]; 2.30 [s, 3H, CH3CO]; 3.10
[m, 1H, 30]; 6.10 [d, 1H, 3, J = 16 Hz]; 6.73 [dd, 1H, 4, J = 16 Hz].
The cis diol (20 mg, 0.075 mmol, 1 equiv) was treated as above
and the mixture was stirred at T = 40° for about 2 h monitoring
the reaction with TLC (ethanol/ethylacetate/acetic acid
10:90:0.25). The reaction was poured into water, finally extracted
with EtOAc (3 ꢁ 1 mL). The EtOAc extract was dried with anhy-
drous sodium sulfate, filtered and concentrated under vacuum to
give a colorless solid. The residue was purified by RP-HPLC to
obtain the title compound (18 mg, yield of 77%) as solid.
ESI–MS m/z: 308.2.
4.3.13. 4-(10,20-Epoxy-20,60,60-trimethylcyclohexyl)-3-buten-2-
one (12)
m-Chloroperoxybenzoic acid (13.4 mg, 0.078 mmol, 1.5 equiv)
was added to a solution of (+)-b-ionone (11
ll, 0.052 mmol,
1 equiv) in 600
bath.
ll of methylene chloride, being cooled in an ice
The reaction mixture was stirred at room temperature for about
12 h and then filtered and washed with aqueous ammonium
hydrogen carbonate 0.1 M. The organic phase was dried with anhy-
drous sodium sulfate, and evaporated under reduced pressure. The
residue was purified by RP-HPLC to obtain the title compound
(7.4 mg, yield of 68%) as an yellow oil.
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.94 [s, 3H, 90]; 0.97
[s, 3H, 80]; 1.62 [d, 3H, 7 J = 1.4 Hz]; 1.72 [d, 1H, 50, J = 16 Hz] 1.81
[d, 1H, 50, J = 16 Hz]; 1.93[d, 3H, 6 J = 1.1 Hz]; 2.01 [s, 3H, O–
COCH3]; 5.25 [m, 1H, 40]; 5.50 [m, 1H, 2]; 5.68 [m, 1H, 30]; 6.21
[d, 1H, 5, J = 15.9 Hz]; 7.61 [dd,1H, 4, J = 15.9, 0.8 Hz].
ESI–MS m/z: 208.2.
4.3.16. (2Z,4E)-5-(10-hydroxy-20,60,60-trimethyl-40-
(phenylamino) 20-cyclohexen-10-yl)-3-methylpenta-2,4-dienoic
acid (14b)
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; 0.94 [s, 3H, 80]; 1.14[s,
6H, 70e 90]; 1.40 [m, 1H, 50]; 1.30–1.9 [m, 6H, 40, 30, 50]; 2.30 [s, 3H,
CH3CO]; 6.10 [d, 1H, 3, J = 16 Hz]; 7.01 [dd, 1H, 4, J = 16 Hz].
( )-ABA (30 mg, 0.113 mmol, 1 equiv) was dissolved in 5 ml of
methanol at room temperature for few minutes. Aniline (16
ll,
4.3.14. (2Z,4E)-5-(10,40-Dihydroxy-20,60,60-trimethyl-40-oxo-20-
0.170 mmol, 1.5 equiv) and acetic acid (100 l, 1.748 mmol,
l
cyclohexen-10-yl)-3-methylpenta-2,4-dienoic acid (13)
10 equiv) were added. The mixture was refluxed for about 4 h.
The reaction was allowed to warm to ambient temperature, then,
sodium cyanoborohydride (20 mg, 1.917 mmol, 11 equiv) was
added in several aliquots over a period of 30 min. The reaction mix-
ture was stirred at room temperature, protected from light, for
24 h, monitoring the reaction with TLC (hexane/ethylacetate/acetic
( )-ABA (10 mg, 0.038 mmol, 1 equiv) in 800
ll of methanol and
100 l of deionized water was cooled to 0 °C. Sodium borohydride
l
(7 mg, 0.18 mmol, 4.5 equiv) was added to the resulting solution
and the mixture was stirred at T = 0 °C for about 6 h, monitoring
the reaction by TLC (ethanol/ethylacetate/acetic acid 10/90/0.25).

M. Bellotti et al. / Bioorg. Med. Chem. 23 (2015) 22–32
31
acid 50:35:1). The reaction was poured into cold water (5 mL), the
pH was adjusted to 8 with addition of aqueous ammonium hydro-
gen carbonate 0.1 M and finally extracted with EtOAc (3 ꢁ 10 mL).
The EtOAc extract was dried with anhydrous sodium sulfate, fil-
tered and concentrated under vacuum to afford a mixture of the
two different products.
[m, 1H, 2]; 5.75 [m, 1H, 30]; 6.10 [d, 1H, 5, J = 15.9 Hz]; 7.69 [dd,
1H, 4, J = 15.9, 0.8 Hz].
13C NMR: (DMSO-d6): T = 25 °C; d 18.85 [70]; 20.86 [6]; 23.19
[80]; 24.16 [90]; 41.31 [60]; 43.31 [50]; 50.87 [O–CH3]; 67.25 [40];
78.41 [10]; 116.63 [2]; 126.04 [20]; 126.99 [4]; 138.45 [5]; 141.05
[30]; 150.77 [3]; 165.69 [1].
The residue was purified by RP-HPLC to obtain the title com-
pound (30 mg, 73%) as yellow oil.
4.3.19. (2Z,4E)-2-(tert-Butoxy)-2-oxoethyl 5-(10,40-dihydroxy-
20,60,60 trimethyl 20-cyclohexen-10-yl)-3-methylpenta-2,4-
dienoate (17)
ESI–MS m/z: 341.4.
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.90 [s, 3H, 90]; 0.94
[s, 3H, 80]; 1.56 [d, 3H, 7 J = 1.4 Hz]; 1.71 [d, 1H, 50, J = 16 Hz] 1.80
[d, 1H, 50, J = 16 Hz]; 1.91[d, 3H, 6 J = 1.1 Hz]; 3.20 [m, 1H, 40];
4.05[s, 1H, NH] 5.25 [s, 1H, OH]; 5.33 [m, 1H, 30]; 5.55 [m, 1H,
2]; 6.02 [d, 1H, 5, J = 15.9 Hz]; 7.19 [m, 1H, CH phenyl]; 7.41–
7.43 [m, 2H, CH phenyl]; 7.50–7.58 [m, 2H, CH phenyl]; 7.61
[dd, 1H, 4, J = 15.9, 0.8 Hz]. 11.00 [s, 1H, OH].
A solution of analog 9 (5 mg, 0.013 mmol, 1 equiv) in 800
ll of
methanol and 100 l of deionized water was cooled to T = 0 °C.
l
Sodium borohydride (5 mg, 0.13 mmol, 10 equiv) was added to
the resulting solution and the mixture was stirred at T = 0 °C for
about 20 h, monitoring the reaction with TLC (ethanol/ethylace-
tate/acetic acid 10:90:0.25). The reaction was poured into water
acidified to pH 2 with 1 N HCI and finally extracted with EtOAc
(3 ꢁ 1 mL). The EtOAc extract was dried with anhydrous sodium
sulfate, filtered, and concentrated under vacuum to give a colorless
solid. The residue was purified by RP-HPLC to obtain the title com-
pound, containing a mixture of two different isomers (4.5 mg, yield
of 90%) as oils.
13C NMR: (DMSO-d6): T = 25 °C; d 18.85 [70]; 20.86 [6]; 23.19
[80]; 24.16 [90]; 41.31 [60]; 43.34 [50]; 50.87 [O–CH3]; 54.7 [40];
78.41 [10]; 116.63 [2]; 118.04 [30]; 120.11 [C phenyl]; 122.32 [C
phenyl]; 126.99 [4]; 129.50 [C phenyl]; 137.45 [5]; 139.05 [20];
146.85 [C phenyl]; 150.77 [3]; 169.69 [1].
ESI–MS m/z: 380.2.
4.3.17. (2Z,4E)-methyl-5-(40-acetoxy-20,50,60-trimethylphenyl)-3-
methylpenta-2,4-dienoate (15)
Analog 7 (10 mg, 0.036 mmol, 1 equiv) was stirred with acetic
anhydride (500 ll, 5.23 mmol) and p-toluene-p-sulfonic acid
(25 mg, 0.145 mmol, 3.8 equiv) for 24 h at room temperature,
monitoring the reaction with TLC (hexane/ethylacetate/acetic acid
50:35:1). About 1 ml of water was added and the final solution was
extracted with EtOAc (3 ꢁ 1 mL). The organic phase was dried with
anhydrous sodium sulfate, filtered, and concentrated under vac-
uum. The residue was purified by RP-HPLC to obtain the title com-
pound (8 mg, yield of 73%) as a gum.
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.92 [s, 3H, 80]; 0.95
[s, 3H, 90]; 1.41 [s, 9H, 3(3HC)CO–]; 1.67 [d, 1H, 50, J = 17.0 Hz]; 1.81
[d, 3H, 70, J = 1.4 Hz]; 1.85 [d, 1H, 50, J = 17.0 Hz]; 2.04 [d, 3H, 6,
J = 1.2 Hz]; 2.85 [s, 1H, OH]; 4.22 [m, 1H, 40], 4.54 [s, 2H, O–CH2–
CO]; 5.26 [s, 1H, OH]; 5.55–5.74 [m, 2H, 30 e 2]; 6.35 [d, 1H, 5,
J = 15.9 Hz]; 7.71 [dd, 1H, 4, J = 15.9, 0.6 Hz].
13C NMR: (DMSO-d6): T = 25 °C; d 18.82 [70]; 20.94 [6]; 24.19
[80]; 27.68 [90]; 41.32 [60]; 49.31[50]; 60.60 [O–CH2–]; 78.40 [10];
81.48 [(H3C)C]; 115.82 [2]; 126.08 [30]; 126.91 [4]; 139.07 [5];
152.25 [3]; 162.89 [20]; 164.52 [1]; 166.99 [C@O]; 197.18 [40].
ESI–MS m/z: 302.1.
4.3.20. (2Z,4E)-5-(10,40-dihydroxy-20,60,60-trimethyl-20-
cyclohexen-1-yl)-3-methyl-2,4-pentadien-1-ol (18)
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 2.02 [s, 3H, 6]; 2.15
[s, 3H, 50]; 2.24 [s, 3H, 60]; 2.25 [d, 3H, 70, J = 1.4 Hz]; 2.31 [s, 1H,
40], 5.79 [s, 1H, 2]; 6.78 [s, 1H, 30]; 6.93 [d, 1H, 4, J = 16.0 Hz],
7.73 [d, 1H, 5, J = 16.0 Hz] 11.00 [s, 1H, OH].
Diisobutylaluminium hydride (1.5 M in toluene, 253 ll,
0.38 mmol, 4 equiv) was carefully added with a syringe to a stirred
solution of analog 7 (25 mg, 0.089 mmol, 1 equiv) in 1.5 ml dieth-
ylether at T = ꢀ20° under N2 atmosphere. The solution was stirred
at the same temperature for about 2 h, monitoring the reaction
4.3.18. (2Z,4E)-methyl-5-(10,40-dihydroxy-20,60,60-trimethyl 20-
cyclohexen-10-yl)-3-methylpenta-2,4-dienoate (16)
with HPLC analysis. Then, 100 ll of methanol was added to the
Analog 7 (10 mg, 0.036 mmol, 1 equiv) in 800
ll of methanol
mixture to stop the reaction. The mixture was stirred at room tem-
perature for 30 min and finally the solvent was evaporated under
vacuum and the residue was dissolved in about 1 ml of deionized
water. The solution was extracted with EtOAc (3 ꢁ 1 mL). The
EtOAc extract was dried with anhydrous sodium sulfate, filtered,
and concentrated under vacuum to give a colorless solid.
The residue was purified by RP-HPLC to obtain the title com-
pound containing two different isomers (a cis analog 4.5 mg, yield
20% and a trans analog, 6.5 mg, yield 30%) as oils.
and 100 l of deionized water was cooled to 0 °C. Sodium borohy-
l
dride (7 mg, 0.18 mmol, 5 equiv) was added to the resulting solu-
tion and the mixture was stirred at T = 0° for about 24 h,
monitoring the reaction with TLC (ethanol/ethylacetate/acetic acid
10:90:0.25). The reaction was poured into water (5 mL), acidified
to pH
2 with 1 N HCI and finally extracted with EtOAc
(3 ꢁ 1 mL). The EtOAc extract was dried with anhydrous sodium
sulfate, filtered, and concentrated under vacuum to give a colorless
solid.
The residue was purified by RP-HPLC to obtain two different
products (a trans-diol methylester derivative, 3 mg, yield of 40%
and a cis-diol methylester derivative, 4 mg, yield of 47%) as oils.
ESI–MS m/z 280.2.
ESI–MS m/z: 252.2.
1H NMR (CDCl3, 300 MHz): T = 25 °C; d 0.90 [s, 3H, 80]; 0.94 [s,
3H, 90]; 1.64 [3H, s, 70]; 1.73 [2H, d, 50]; 1.85 [3H, s, 6]; 2.59 [3H,
br s, 3OH]; 4.28 [m, 3H, –CH2OH e 40]; 5.57 [m, 2H, 30e 2] 5.67
[d, 1H, 5, J = 15.5 Hz]; 6.73 [d, 1H, 4, J = 15.5 Hz].
1H NMR (DMSO-d6, 300 MHz): T = 25 °C; d 0.94 [s, 3H, 90]; 0.99
[s, 3H, 80]; 1.68 [d, 3H, 70 J = 1.4 Hz]; 1.69 [d, 1H, 50, J = 16.6 Hz] 1.81
[d, 1H, 50, J = 16.6 Hz]; 2.00[d, 3H, 6 J = 1.1 Hz]; 2.81 [s, 1H, OH];
3.71 [s, 3H, O–CH3]; 4.25 [m, 1H, 40]; 5.25 [s, 1H, OH]; 5.70 [m,
1H, 2]; 5.81 [m, 1H, 30]; 6.05 [d, 1H, 5, J = 15.9 Hz]; 7.72 [dd, 1H,
4, J = 15.9, 0.8 Hz].
1H NMR (CDCl3, 300 MHz): T = 25 °C; d 0.87 [s, 3H, 80]; 0.98 [s,
3H, 90]; 1.51 [dd, 1H, J = 10.13 Hz, 50]; 1.62 [3H, s, 70]; 1.70 [dd,
1H, 50, J = 6.13 Hz]; 1.85 [s, 3H, 6]; 2.75 [3H, br s, 3OH]; 4.20 [m,
3H, –CH2OH e 40]; 5.56[m, 2H, 30e 2] 5.74 [d, 1H, 5, J = 15.5 Hz];
6.56 [d, 1H, 4, J = 15.5 Hz].
1H NMR (DMSO-d6, 300 MHz), T = 25 °C; d 0.92 [s, 3H, 90]; 1.01
[s, 3H, 80]; 1.67 [d, 3H, 70 J = 1.4 Hz]; 1.69 [d, 1H, 50, J = 16.6 Hz]
1.83 [d, 1H, 50, J = 16.6 Hz]; 2.02[d, 3H, 6 J = 1.1 Hz]; 2.81 [s, 1H,
OH]; 3.69 [s, 3H, O–CH3]; 4.22 [m, 1H, 40]; 5.25 [s, 1H, OH]; 5.70
4.3.21. Determination of intracellular cAMP in human
granulocytes
Human granulocytes were isolated by density gradient centrifu-
gation of buffy coats,3 prepared from freshly drawn blood of

32
M. Bellotti et al. / Bioorg. Med. Chem. 23 (2015) 22–32
8. Bruzzone, S.; Ameri, P.; Briatore, L.; Mannino, E.; Basile, G.; Andraghetti, G.;
Grozio, A.; Magnone, M.; Guida, L.; Scarfì, S.; Salis, A.; Damonte, G.; Sturla, L.;
Nencioni, A.; Fenoglio, D.; Fiory, F.; Miele, C.; Beguinot, F.; Ruvolo, V.; Bormioli,
M.; Colombo, G.; Maggi, D.; Murialdo, G.; Cordera, R.; De Flora, A.; Zocchi, E.
FASEB J. 2012, 26, 1251.
9. Gusta, L. V.; Reaney, M. J. T.; Abrams, S. R.; Robertson, A. J.; Abrams, G. D.
Comments Agric. Food Chem. 1990, 2, 143.
10. Walker-Simmons, M. K.; Anderberg, R. J.; Rose, P. A.; Abrams, S. R. Plant Physiol.
1992, 99, 501.
11. Blake, T. J.; Tan, W.; Abrams, S. R. Physiol. Plant. 1990, 80, 365.
12. Flores, A.; Dorffling, K. J. Plant Growth Regul. 1990, 9, 133.
13. Flores, A.; Grau, A.; Laurich, F.; Dorffling, K. J. Plant Physiol. 1988, 132, 362.
14. Reaney, M. J. T.; Gusta, L. V.; Abrams, S. R.; Shaw, A. C.; Ewan, B. Plant Physiol.
1990, 93, S-7.
15. Schubert, J.; Roser, K.; Grossmann, K.; Sauter, H.; Jung, J. J. Plant Growth Regul.
1991, 10, 27.
16. Walton, D. C. In Abscisic Acid; Addicott, F. T., Ed.; Praeger: New York, 1983; p
113.
17. Milborrow, B. V. In Abscisic acid; Letham, D. S., Goodwin, P. B., Higgins, T. J. V.,
Eds.; Elsevier/North Holland Biomedical Press: New York, 1978; p 295. Vol. 1.
18. Churchill, G. C.; Ewan, B.; Reaney, M. J. T.; Abrams, S. R.; Gusta, L. W. Plant
Physiol. 1992, 100, 2024.
19. Churchill, G. C.; Ewan, B.; Reaney, M. J. T.; Abrams, S. R.; Gusta, L. W. Plant
Growth Regul. 1998, 25, 35.
20. Walker-Simmons, M. K.; Reaney, M. J. T.; Quarrie, S. A.; Perata, P.; Vernieri, P.;
Abrams, S. R. Plant Physiol. 1991, 95, 46.
21. Walker-Simmons, M. K.; Rose, P. A.; Shaw, A. C.; Abrams, S. R. Plant Physiol.
1994, 106, 1279.
22. Perras, M.; Rose, P. A.; Pass, E. W.; Chatson, K. B.; Balsevich, J. J.; Abrams, S. R.
Phytochemistry 1997, 46, 215.
23. Ueno, K.; Araki, Y.; Hirai, N.; Saito, S.; Mizutani, M.; Sakata, K.; Todoroki, Y.
Bioorg. Med. Chem. 2005, 13, 3359.
24. McCourt, P. a.; Creelman, R. b. Curr. Opin. Plant Biol. 2008, 11, 474.
25. Sturla, L.; Fresia, C.; Guida, L.; Grozio, A.; Vigliarolo, T.; Mannino, E.; Millo, E.;
Bagnasco, L.; Bruzzone, S.; De Flora, A.; Zocchi, E. Biochem. Biophys. Res.
Commun. 2011, 415, 390.
healthy human volunteers. Granulocytes were then incubated
overnight in DMEM/F12 supplemented with 10% of fetal bovine
serum (FBS), penicillin (50 units/ml) and streptomycin (50 lg/ml)
(DMEM/F12 complete). The physical stimulation of the cells during
their purification results in an artificial elevation of basal [cAMP]i
values, compromising the ability of the cells to respond to ABA.
After overnight rest, cells were centrifuged at 200ꢁg for 5 min
and resuspended at 1.5 ꢁ 107/ml in Hanks’ Balanced Salt Solution
(HBSS) with calcium and magnesium. Granulocytes were preincu-
bated in the presence of 10
4-[(3-butoxy-4-methoxyphenyl)-methyl]-2-imidazolidinone (Ro-
20-1724) for 5 min at 25 °C, duplicate 200 l-aliquots of cells were
incubated without (control) or with 1 M ( )-ABA, in the absence
or in the presence of different analogs at 50 M final concentration,
for 1 min at 25 °C. The reaction was stopped by the addition of
13 l of 0.6 M perchloric acid (PCA) at 4 °C. The cell lysates were
lM cAMP phosphodiesterase inhibitor
l
l
l
l
centrifuged for 5 min at 17,000ꢁg at 4 °C and the supernatant
was adjusted to pH 7.5 with 3 M potassium carbonate (K2CO3).
The reaction led to the development of CO2 gas and the formation
of a white precipitate of potassium perchlorate (KClO4), which was
removed by a 5-min centrifugation at 17,000ꢁg at 4 °C.
The intracellular cAMP concentration was determined with a
competitive radioimmunological test [3H]-cAMP assay system
(Amersham, GE Healthcare), following the manufacturer’s
instructions.
Author disclosure statement
26. Belsevich, J.; Bishop, G.; Jacques, S. L.; Hogge, L. R.; Olson, D. J. H.; Laganiere, N.
Can. J. Chem. 1996, 74, 112.
27. Liu, X.; Ma, L.; Lin, Y.; Tang, Lu. Y. J. Chromatogr. A 2003, 1021, 209.
28. Ward, D. E.; Yuanzhu, G. Synth. Commun. 1997, 27, 2133.
29. Kim, K. H.; Fan, X. J.; Nielsen, P. E. Bioconjug. Chem. 2007, 18, 567.
30. Todoroki, Y.; Hirai, N. Tetrahedron 1995, 25, 6911.
The authors declare no competing financial interest.
Acknowledgements
31. Beale, M. H.; Chen, A.; Harrison, P. A.; Willis, C. L. J. Chem. Soc. Perkin. Trans. 1
1993, 3061.
32. Hirai, N.; Yoshida, R.; Todoroki, Y.; Ohigashi, N. Biosci. Biotechnol. Biochem.
2000, 64, 1448.
33. Kim, B. T.; Min, Y. K.; Asami, T.; Park, N. K.; Kwon, O. Y.; Cho, K. Y.; Yoshida, S. J.
Agric. Food Chem. 1999, 47, 313.
34. Mallaby, R.; Ryback, G. J. Chem. Soc. Perkin. Trans. 2 1972, 2, 919.
35. Aleu, J.; Brenna, E.; Fuganti, C.; Serra, S. J. Chem. Soc. Perkin Trans. 1 1999, 271.
36. Zaharia, L. I.; Walker-Simmon, M. K.; Rodriguez, C. N.; Abrams, S. R. J. Plant
Growth Regul. 2005, 24, 274.
The authors would like to thank Dr. Paola Fossa for helping in
revision manuscript. This work was supported in part by the Italian
Ministry of Education University and Scientific Research, by the
University of Genova, by the Fondazione CARIGE, by the Compag-
nia di S. Paolo, and by Regione Liguria.
References and notes
37. Lin, B. L.; Wang, H. J.; Wang, J. S.; Zaharia, L. I.; Abrams, S. R. J. Exp. Bot. 2005, 56,
2935.
38. Ueno, K.; Yoneyama, H.; Mizutani, M.; Hirai, N.; Todoroki, Y. Bioorg. Med. Chem.
2007, 15, 6311.
39. Rock, C. D.; Heath, T. G.; Gage, D. A.; Zeevaart, J. A. Plant Physiol. 1991, 97, 670.
40. Grozio, A.; Millo, E.; Guida, L.; Vigliarolo, T.; Bellotti, M.; Salis, A.; Fresia, C.;
Sturla, L.; Magnone, M.; Galatini, A.; Damonte, G.; De Flora, A.; Bruzzone, S.;
Bagnasco, L.; Zocchi, E. Biochem. Biophys. Res. Commun. 2011, 415, 696.
41. Zocchi, E.; De Flora, A.; Millo, E.; Bruzzone, S.; Guida, L.; Grozio, A.; Salis, A.;
Magnone, M. European Patent 2511272A1, 2012.
42. Takeuchi, J.; Okamoto, M.; Akiyama, T.; Muto, T.; Yajima, S.; Sue, M.; Seo, M.;
Kanno, Y.; Kamo, T.; Endo, A.; Nambara, E.; Hirai, N.; Ohnishi, T.; Cutler, S. R.;
Todoroki, Y. Nat. Chem. Biol. 2014, 10, 477.
1. Nambara, E.; Marion, P. Annu. Rev. Plant Biol. 2005, 56, 165.
2. Xue-Xuan, X.; Hong-Bo, S.; Yuan-Yuan, M.; Jun-Na, S.; Dong-Gang, G.; Cheng-
Jiang, R. Crit. Rev. Biotechnol. 2010, 30, 222.
3. Bruzzone, S.; Moreschi, I.; Usai, C.; Guida, L.; Damonte, G.; Salis, A.; Scarfì, S.;
Millo, E.; De Flora, A.; Zocchi, E. Proc. Natl. Acad. Sci. U.S.A. 2007, 104, 5759.
4. Bruzzone, S.; Bodrato, N.; Usai, C.; Guida, L.; Moreschi, I.; Nano, R.; Antonioli, B.;
Fruscione, F.; Magnone, M.; Scarfì, S.; De Flora, A.; Zocchi, E. J. Biol. Chem. 2008,
283, 32188.
5. Bodrato, N.; Franco, L.; Fresia, C.; Guida, L.; Usai, C.; Salis, A.; Moreschi, I.;
Ferraris, C.; Verderio, C.; Basile, G.; Bruzzone, S.; Scarfì, S.; De Flora, A.; Zocchi,
E. J. Biol. Chem. 2009, 284, 14777.
6. Magnone, M.; Bruzzone, S.; Guida, L.; Damonte, G.; Millo, E.; Scarfì, S.; Usai, C.;
Sturla, L.; Palombo, D.; De Flora, A.; Zocchi, E. J. Biol. Chem. 2009, 284, 17808.
7. Magnone, M.; Sturla, L.; Jacchetti, E.; Scarfì, S.; Bruzzone, S.; Usai, C.; Guida, L.;
Salis, A.; Damonte, G.; De Flora, A.; Zocchi, E. FASEB J. 2012, 26, 1261.