Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
A.M. Eliasen et al. / Tetrahedron xxx (2018) 1e8
3
xanthofulvin, namely the two partners needed to have singular
reactivity i.e. Michael donor or Michael acceptor, not both types of
reactivity as 5,6-dehydropolivione demonstrates.
Using a surrogate for dehydropolivione, an ynone such as 25
that can only function as a Michael acceptor provided a promising
subtarget. A proposed conjugate addition into the ynone 25 by a
ketoester would proceed through a related series of reactions to
form the xanthone core of xanthofulvin (Fig. 4). Previously this type
of reactivity had been documented with 3-(1-alkynyl)chromones
and 1,3-dicarbonyl compounds under basic conditions providing
support for our approach.15
The subtarget ynone 33 was prepared in nine steps from tetronic
acid using a similar synthetic pathway as developed for 5,6-
dehydropolivione (3) (Scheme 5). The addition of the sodium
anion of methyl acetoacetate, using sodium hydride, formed the
xanthone core in a single operation yielding xanthone 34. With
access to 34 the preparation of xanthofulvin required an additional
five synthetic steps. This provided the first synthesis of xantho-
fulvin. In addition to providing synthetic material for the first time
the synthesis resolved an inconsistency in the literature regarding
an alternative structural assignment of xanthofulvin as a proposed
structure dubbed 411J.16
With scalable access to both natural products, enabled by the
methods development, the evaluation of the mechanism of action
beyond Sema3a inhibition was made possible. Previously, the
pharmacological profile of vinaxanthone and xanthofulvin against
panels of enzymes and kinases was examined showing a high level
of selectivity for Sema3a.7,8 Given the dual anionic nature of the
compounds at physiological pH, it was expected that the molecules
would not readily enter cells. Accordingly, we investigated different
cell surface proteins as potential targets using the previously uti-
Fig. 2. Model system 14 of 5,6-dehydropolivione (3) for dimerization studies.
triketone, again in low yield. Exposure of enaminone 15 to
commercially available dioxinone (Entry 3) provided triketone 14
in 11% yield, which was an improvement but still not useful for the
process we sought to develop. Acyl Meldrum's acid (Entry 4 and 5)
underwent thermolysis in the presence of enaminone 15 and the
yield was increased to 16%. Optimization of this result led to an
enhanced protocol utilizing a pre-heated oil bath at 145 ꢀC ensuring
near immediate acyl ketene generation, and purification with
phosphoric acid-treated (pH ¼ 2) silica gel, which provided trike-
tone 14 in a reproducible 43% yield.
Hoye and co-workers had studied the chemistry of thio-
acetoacetates as acyl ketene donors, which can react under mild
ambient conditions in the presence of a thiophilic silver salt.14
Under this protocol (Entry 6, Table 1) a yield of 67% was achieved
improving from the previous reactions. Attempts to optimize this
further were not met with success, but the reaction was capable of
good material throughput.
With access to triketone 14 the proposed dimerization reaction
(Scheme 1) was made possible. Testing solvents and combinations,
without added acid or base, quickly arrived at the ideal solvent
system; a heated mixture of water/dioxane which cleanly furnished
the cyclized, dimerized adduct 21, assembling the carbogenic
scaffold of vinaxanthone in a single reaction from 14 (Table 2). The
structure of 21 was confirmed using 2D-NMR analysis. Noteworthy,
the reaction proceeds under neutral conditions and the addition of
acids or bases were deleterious, as an example treatment under
basic conditions with triethylamine furnished the hydroxylated
benzophenone 22 rather than the desired xanthone.
Using the methodological finding of the facile dimerization
applying it to the synthesis of vinxanthone required the synthesis
of the hypothetical natural product precursor 5,6-dehydropolivione
(3). Starting from tetronic acid (23) 5,6-dehydropolivione was
synthesized in eight steps (Scheme 3). Minor modification to the
reaction conditions used in the model system, namely replacing the
binary solvent system with water alone and lowering the temper-
ature (and extending the reaction time) the natural product
vinaxanthone was accessed for the second time by chemical syn-
thesis, the first was achieved in 200711 using Diels-Alder chemistry.
Our nine-step synthesis generated ample material for testing.
However, the application of this reaction to the synthesis of
lized concentrations of the natural products, 0.2 mM. Through these
studies commercial panel of 165 G-protein coupled receptors
(GPCRs) were assayed for activity against vinaxanthone and xan-
thofulvin. It is noteworthy that both compounds failed to function
as agonists nor antagonists for any of the GPCRs examined. How-
ever, it was found that both compounds markedly enhance the
affinity and efficacy of succinate receptor 1 (SUCNR1) towards its
ligand succinate although both natural products demonstrate no
agonistic activity. (Fig. 4).
Studies into SUCNR1 evolved rapidly following the discovery in
2004 that the then orphaned GPCR, GPR91, was activated by suc-
cinate at concentrations of 56
28
M (FLIPR assay).17 Activation by succinate occurs due to
increased concentration of succinate, relative to a baseline of
2e20
M in different tissues, 18,19 This in turn leads to transcription
8 mM (aequorin assay) and
5
m
m
of factors enabling angiogenesis and growth. This response occurs
at lower concentration than that required for succinate stabiliza-
tion of hypoxia inducible factor (HIF), providing evidence that many
of the effects of succinate that have been linked to HIF in fact may
be due to activation of SUCNR1.20 Since both the neuronal and
vascular system share similar signaling pathways, production of
growth factors not only leads to enhanced blood vessel develop-
ment but also to heightened axonal growth.21 The activation of
SUCNR1 leads to the pronounced growth promoting effects seen
xanthofulvin following from
a postulated reaction of 5,6-
dehydropolivione and polivione (24) failed, only generating
vinaxanthone regardless of addition sequences, stoichiometery and
a variety of other perturbations of the previously successful method
(Scheme 4). While this reaction did not translate it did provide
insight into the appropriate synthetic approach toward
Scheme 2. Proposed cascade reaction of enamine 15 with acylketene (16) to generate triketone 14.
Please cite this article in press as: Eliasen AM, et al., Cascade reactions leading to the mechanism of action of vinaxanthone and xanthofulvin,
Eliasen, Anders M.
