T. Ling et al. / European Journal of Medicinal Chemistry 110 (2016) 126e132
127
the isopropyl alkane through a base mediated bromination of
propargyl alcohol followed by alkylation under Gilman's reaction
[16a] conditions to afford the corresponding alkyne almost quan-
titatively. Hydroxyl directed Red-Al reduction [16b] of the prop-
argyl unit, and sequential quenching with iodine provided the
desired syn configuration as a sole regio-product. Chemical trans-
formation of the hydroxyl group to iodide 16a in 50 g batches was
mediated by zirconium tetrachloride with a 60% overall yield
(Scheme 2) [15a].
Both aromatic and alkynyl groups can be readily generated using
this effective two step (one pot) reaction. To evaluate the role of
larger groups and to introduce functional groups, the electrophiles
16b-e were synthesized in good yields as depicted in Scheme 2
(60e85% yields, for detail See SI).
Fig. 1. Representative natural product 11b-HSD1 inhibitors.
Our established synthesis of the core of colletoic acid has been
improved by adapting the readily available Evans chiral auxiliary
[17], which is easily removed under mild reductive conditions
(Scheme 3). The synthesis of compound 14 had previously been
reported by our group and can be generated in multigram quanti-
ties [15a]. Treatment of (R)-4-Benzyl-2-oxazolidinone with pivaloyl
chloride to generate the mix-anhydride [17], followed by addition
of 14, afforded 13 in multigram scale. The formation of the enolate
of compound 13 was mediated by NaHMDS and quenched with
electrophile 16a-e to afford the desired precursor 12 (ee ꢀ 98%) in
gram quantities with excellent diastereoselectivity (dr > 20: 1) and
chemical yield (Scheme 3). Removal of benzyl-2-oxazolidinone
with NaBH4 in methanol proceeded in good yield, and the corre-
sponding hydroxyl group was protected with TESCl and imidazole
to provide the Heck reaction precursor. The intramolecular Heck
reaction was mediated by palladium (0) in acetonitrile/THF at 60 ꢁC
fungus Colletotrichum gloeosporioides [13]. The acorane family is a
subgroup of the large sesquiterpenoid family with less than 100
natural products reported to date. The relative stereochemistry of
CA was confirmed by x-ray crystallography and its absolute ste-
reochemistry was elucidated by a modified Mosher's method [13].
CA was reported to be a selective 11b-HSD1 inhibitor with an IC50 of
13 nM and no detectable 11 -HSD2 inhibition [13]. CA possesses a
b
spiro- [4,5] decane core, decorated with four stereogenic centers,
three of which are successive. CA features an all-carbon quaternary
spirocenter, providing a unique platform to develop new 11b-HSD1
modulators. A diverse synthetic strategy to CA offers the opportu-
nity to access other acorane family members, which have not been
exploited for their biological properties. These remarkable small
natural products have shown potent activity such as the closely
related axisonitrile, a potent antimalarial agent [14]. These unusual
natural products offer new chemical scaffolds with rich stereo-
chemistry for drug discovery. Herein, we disclose our efforts to
improve the process chemistry involved in the synthesis of the CA
core, to access a diverse number of derivatives for biochemical
to generate the
a, b-unsaturated spirocycle 9 as a single diaste-
reoisomer after protecting group removal with aqueous HCl [15a].
Relative stereochemistry was confirmed via 2D NMR and X-ray
analysis of derivatives. The Heck reaction was observed with high
regio and stereo control in the tested substrates. It is presumed that
the exo- transition state is favored to avoid clashing interactions
between the palladium complex and the R group in the transition
state. With intermediate 9 available, we have generated various
colletoic acid-like compounds as well as colletoic acid in multi-
milligram scale for further mechanistic studies. Compound 9 is a
rich source of potential derivatives featuring congruent function-
alities. The conjugated system is prone to a multitude of nucleo-
philic additions while the electron rich olefin preferentially
undergoes electrophilic reactions.
evaluation against 11b-HSD1.
Previously we [15a] and others [15b] have reported the enan-
tioselective total synthesis of CA. We also validated its activity as a
potent 11b-HSD1 inhibitor in a whole cell assay [15a]. We have
further streamlined the synthesis to access CA and simplified de-
rivatives bearing the all-carbon spirocenter. In addition, our initial
exploration of the structure-activity relationship of these de-
rivatives against 11b-HSD1 inhibition is presented.
In the retrosynthetic analysis of CA and the corresponding an-
alogs the key intermediate 9 is envisioned for late-stage function-
With the readily available key intermediate 9 using the opti-
mized reaction sequence, sufficient quantities were generated for
synthesis of the corresponding colletoic acid core derivatives. The
objective was to synthesize derivatives to collect experimental in-
formation regarding the required functional groups for activity of
alization. Compound
9
would arise presumably from
a
stereoselective spiro center formation by a palladium mediated
Heck reaction. The preceding synthon 12 would be generated via
alkylation at C1 of the readily available Evan's amide 13 with
electrophile 16 (Scheme 1). In the retrosynthetic disassembly of the
system, 13 would be readily available from monoprotected 1, 4-
cyclohexadione 15, via a HornereWadswortheEmmons reaction
and olefin isomerization. The approach highlights the modular
aspects of the synthesis and feasibility of generating scale quanti-
ties of the natural product CA (more than 500 mg has been pre-
pared to date). The advantageous synthesis allows the introduction
of a broad range of electrophiles during the alkylation of 13 to
generate colletoic acid core derivatives 17e60 (Scheme 1). The
compound 9 is versatile because functional groups can be intro-
duced at C1, C4, and C7eC10.
CA against 11b-HSD1. The synthesis of a selection of derivatives
highlights the versatility of our synthetic approach. Hydroxylated
9a core was readily produced via chemoselective epoxidation and
oxirane opening (Scheme 4). Chemoselective epoxidation of 9a
promoted by aqueous H2O2 and NaOH had previously been satis-
factory [15a], but not optimal. It was reasoned that the nature of the
protecting group on the hydroxyl group would influence the
outcome. Protection of 9a with DHP, and PPTS in DCM provided the
THP protected hydroxyl group in quantitative yield; the resultant
product was then treated under basic epoxidation conditions to
afford the epoxide product in 80% yield as a single diastereoisomer.
Then, the compound was re-protected with TBSCl and imidazole in
DCM to afford the corresponding silyl ether, which was treated
A general synthesis strategy to the di-iodo substrates (16ae16e)
based on a two step (one pot) reductive iodonization (Scheme 2) to
access the distal syn product was applied, delivering excellent
chemical yields. The synthon 16a was generated by introduction of
with methyl magnesium bromide (b-face attack) to provide com-
pound 7 as a single diastereoisomer at C10 in 90% yield. The C3eC4
hydrogenation was mediated by Adam's catalyst (PtO2) under 1 atm