Synthesis of a 6,6-Spiroketal Amino Acid
COMMUNICATION
À
linear peptide chain was constructed on 2-chlorotrityl resin
preloaded with leucine using automated Fmoc-SPPS condi-
tions. However, the valuable Fmoc-spiroketal amino acid 7
was coupled using a manual procedure employing HATU,
HOAt and DIPEA. The second Phe residue was also incor-
porated using a manual procedure employing the phospho-
nium coupling reagent, PyBoP, as initial attempts to couple
the second Phe residue using HBTU resulted in formation
of an N-terminal tetramethyl guanidine by-product.
full list). Notably, however, an important strong Ha Ha rOe
was observed between Trp3 and Phe8.
Based on rOe-derived distance restraints, average solution
structures were calculated for peptide 8 by restrained molec-
ular dynamics in torsion angle space using the simulated an-
nealing protocol in the program DYANA,[26] with the spiro-
ketal in the bis-anomerically stabilised conformation. The
resulting calculated average structures superimpose with a
backbone rmsd of approximately 1.2 ꢂ (Figure 4 and
Table 1), and clearly reflect a significant backbone flexibility,
Upon completion of the peptide sequence, the resin was
carefully treated with 20% AcOH/CH2Cl2, followed by
80% AcOH/CH2Cl2 to release the linear protected peptide
22 (Scheme 2). A dilute solution (1 mgmLÀ1) of this materi-
al was then treated with HATU to induce the macrocyclisa-
tion, the progress of which was monitored by observing the
consumption of the linear peptide 22 by analytical RP-
HPLC.[25] Upon completion of the reaction, the solvent was
removed in vacuo and the crude protected cyclic peptide 23
was immediately treated with 95% TFA in H2O to cleave
the side chain protecting groups (Scheme 2). Subsequent pu-
rification by preparative RP-HPLC afforded cyclic spiro-
ACHTUNGTRENNUNG
ketal peptide 8 with no epimerisation of Leu4 Ca observed
by NMR spectroscopy or HPLC analysis (>96% purity,
[M+H]+ =1364.4, Figure 3).
Figure 4. Left: backbone representation and superimposition of the 20
DYANA NMR structures for 8. All backbone heavy atoms were used for
the superimposition. Right: one typical structure is shown. MOLMOL[27]
and PyMOL[28] were used for structure analysis and visualisation.
with large torsion angle fluctuations (Table 2). Nevertheless,
the backbone conformations show similarities and adopt dis-
torted hairpin-like structures. Interestingly, the spiroketal
motif itself does not mimic the central i+1/i+2 region of a
b-turn. Rather, the calculated structures all contain the
spiroACTHUNRTGNEUNGketal in a b-turn-like structure, with the spiroketal oc-
cupying positions i and i+1, and Phe1 and Glu2 at positions
i+2 and i+3, respectively. Thus, the spiroketal does not
induce a full 1808 turn in the peptide chain, but rather indu-
ces an approximate 908 turn in the backbone conformation.
This is illustrated in Figure 5, showing the turn-like region
Figure 3. HPLC and MS spectra of purified cyclic spiroketal peptide 8
using a Gemini C18 column (5 mm, 4.6ꢃ150 mm), 1–81% CH3CN/H2O
over 20 min, 1.5 mLminÀ1
.
Table 1. Summary of experimental distance restraints used and resulting
statistics for the final 20 NMR structures calculated for peptide 8, as su-
perimposed and shown in Figure 4.
The solution structure of spiroketal peptide 8 was investi-
gated by 1H NMR spectroscopy in [D6]DMSO, due to its
limited solubility in water. At room temperature, the spec-
trum showed broadening of the NH resonances although
sharp signals were observed in the aromatic region (dH =
6.7–6.8 ppm). Reducing the peptide concentration had little
effect on the chemical shifts or shape of the signals, howev-
er, raising the temperature to 355 K in [D6]DMSO caused
sharpening of the NH resonances. These findings are consis-
tent with relatively slow exchange process(es) on the NMR
time scale at room temperature. In 2D ROESY plots, with a
mixing time of 250 ms, most of the observed rOes were
Experimental distance restraints
Calcd
number of NOE upper-distance limits
intraresidue
65
21
sequential
35
medium- and long-range
9
residual target function value [ꢂ2]
0.32Æ0.08
mean rmsd values [ꢂ]
all backbone atoms
all heavy atoms
1.17Æ0.40
2.61Æ0.80
residual NOE violations
number >0.2 ꢂ
5
intra
ACHTUNGTRENNUNGresidue or sequential, with only a few weak long range
maximum [ꢂ]
0.25
rOes being observed (see the Supporting Information for a
Chem. Eur. J. 2013, 19, 3807 – 3811
ꢁ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3809