2646
A. Lawer et al. / Bioorg. Med. Chem. Lett. 24 (2014) 2645–2647
hydrogen-bonding groups suggest that simplification of the struc-
ture should be a priority.
commenced. Since pohlianin C is a glycine-rich cyclic octapeptide,
it was expected to exhibit considerable conformational flexibility,
and so it was hypothesised that imposing a conformational restric-
tion may increase the antiplasmodial potency through entropic
means. Therefore, the immediate objective was to identify a loca-
tion within the molecule that was amenable to modification with-
out compromising the antiplasmodial activity. The glycine–glycine
segment of 3 seemed a logical candidate since this segment con-
tains no sidechains that might be important for target interactions,
Accordingly, the aims of the present work are two-fold: first, to
perform a total synthesis of the natural product in order to confirm
its reported structure and antiplasmodial activity, and to provide
material for mechanistic studies; and second, to identify groups
within the structure that are amenable to modification, as a first
step towards the production of more potent analogues for thera-
peutic applications.
The planned synthesis of 3 entailed macrocyclisation between
glycine and phenylalanine residues (Fig. 1). This point of cyclisa-
tion was chosen because the presence of glycine at one terminus
was expected to lead to reduced steric crowding during the peptide
coupling reaction. The perhaps more intuitive cyclisation point,
between the two adjacent glycine residues of 3, was not exploited
for reasons which are articulated later. The requisite linear peptide
7 (Table 1, entry 1) was readily assembled in near-quantitative
yield by Fmoc-strategy solid phase peptide synthesis (SPPS),
employing Wang resin as the solid support and HBTU/DIPEA as
the coupling reagents.9 After cleavage of the linear peptide 7 from
the resin, cyclisation was achieved using the coupling reagents
DMTMM.BF4/DIPEA10 at a peptide concentration of 5 mM. Purifica-
tion of the product by preparative reversed-phase HPLC furnished
the desired cyclic peptide 3 in moderate yield, thereby completing
the total synthesis of pohlianin C (Table 1, entry 1). Synthetic 3 was
found to be spectroscopically identical to the natural sample,11
confirming that the reported structure of 3 is accurate.
and also because a
variable-temperature NMR study6 had
previously indicated that both glycine NH protons were solvent-
exposed, that is, this segment was likely to be very flexible.
Accordingly, a small series of analogues (4–6) was targeted in
which the glycine–glycine segment was replaced with simple
d-amino acids (Table 1, entries 2–4).12
Analogue 4 (Table 1, entry 2) contained a simple d-aminopenta-
noic acid unit in place of the glycine–glycine segment of pohlianin C.
The requisite linear precursor 8 was readily obtained through SPPS,
and cyclisation of this peptide then delivered the target 4 in modest
yield. The next analogue, 5 (Table 1, entry 3) contained an aminom-
ethylbenzoic acid unit in place of the glycine–glycine segment of
pohlianin C; synthesis of this target was readily achieved by cyclisa-
tion of the corresponding linear precursor 9. Finally, analogue 6
(Table 1, entry 4) contained a p-aminobenzoic acid unit in place of
the glycine–glycine segment of pohlianin C. The requisite linear pre-
cursor 10 was obtained by SPPS under similar conditions employed
for synthesising the other linear peptides; however, the reduced
nucleophilicity of the aniline group within 10 necessitated a double
coupling of the next amino acid, and diisopropylcarbodiimide was
found to be a superior coupling reagent in this step. Finally, cyclisa-
tion of peptide 10 delivered the target 6 in low yield (Table 1, entry
4); this cyclisation reaction was rather challenging, delivering none
of the desired product at 5 mM and only a trace quantity (4%) at
1 mM. No side-products arising from epimerization or oligomeriza-
tion were identified in any of the cyclization reactions.
Attention was next turned to confirming the biological activity
of 3. The growth of the 3D7 (chloroquine-sensitive) P. falciparum
line was monitored in the presence of varying concentrations of
3 (n = 3), and the resulting inhibition curve (Fig. 2) allowed the
IC50 of 3 to be estimated at 25
lM. This is reasonably close to the
IC50 value of 16
l
M which was reported for the natural sample.6
Having confirmed the structure and antiplasmodial activity
of pohlianin C (3), a search for more potent analogues was
Table 1
a,bSynthesis of pohlianin C (3) and analogues 4–6
Entry
Linear peptide (yield from SPPS)
Cyclic peptide (cyclisation yield)
=
1
2
7 (99%)
3 (38%)
4 (20%)
8 (100%)
9 (86%)
3
4
5 (28%)
6 (4%)d
10 (74%)c
a
b
c
DMTMM.BF4 = 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methylmorpholinium tetrafluoroborate.
DIPEA = diisopropylethylamine.
Conditions for SPPS coupling of aniline group = Fmoc-Phe-OH, diisopropylcarbodiimide, DCM (2Â).
d
Cyclisation performed at 1 mM.
Lawer, Aggie
