1964
J.Vizzavona et al./ Bioorg.Med.Chem.Lett.12 (2002) 1963–1965
Alternatively, we have also used 2 to form the Gly
derivative on the resin after reaction with the resin-
bound bromoacetyl-peptide. However, we felt more
confident to incorporate the building block after syn-
thesis according to the Fmoc strategy, since standard
coupling procedures can be applied.
scavengers in TFA was not successful. Surprisingly, an
earlier eluting peak as compared to the purified ligation
product was observed when injecting directly the TFA
cleavage solution onto an RP-HPLC column. Attempts
to isolate this new, more polar compound failed. Both
ligation products, 9 and 11, displayed a similar beha-
viour upon treatment with TFA. LC–MS analysis
revealed the identical mass for the early eluting peak
and the ligation product, thus we assumed the existence
of an equilibrium between the amide bond-containing
ligation product and the intermediate thioester which is
formed in the first stepof the ligation (Fig. 3A).
In this study, two different peptide-thioesters, H-Tyr-
Ser-Leu-SBzl 6 and Z-Lys-SBzl 7, have been used to
produce the target peptides listed in Table 1.
In the case of the Dmmb group, the reaction is thought
to pass through a six-membered ring in the transition
state, followed by a rearrangement to produce the
desired amide bond. In the first step, we have examined
the kinetics in comparison to the NCL method, in which
ligation proceeds through a five-membered ring transi-
tion state.
The following results support this suggestion and are
rationalized in Figure 3:
1. the earlier eluting peak completely disappears if
the pH is raised to 8;
2. no earlier eluting peak is produced after selective
methylation of the sulfur of the Dmmb groupto
obtain 132 followed by TFA treatment;
3. no earlier eluting peak could be detected after
oxidation to the disulfide containing dimer 14
followed by TFA treatment.
The time course of the ligation reactions under standard
conditions, that is 0.1 M phosphate buffer, 6 M guani-
dine, 40 mM TCEP, 3% thiophenol, pH 7.4, was fol-
lowed by analytical HPLC. In the native situation 8, the
reaction was finished after 6 h (Fig. 2). In the case of the
Dmmb group 9, 25 h were required to achieve quantitative
conversion to the ligated product.
Interestingly, the modified Dmmb groupattached to
compound 13 could be cleaved upon prolonged treat-
ment (20 h) in a TFA solution (93% TFA, 5% H2O, 1%
TIPS, 1% EDT). Under these conditions at ambient
temperature, compound 11 was almost completely
converted to the putative internal thioester.
We have anticipated that the Dmmb group would be
cleaved under acidic conditions, for example >90% of
TFA, as commonly applied in peptide synthesis. How-
ever, the removal of the Dmmb groupusing a variety of
Since postsynthetic methylation may be cumbersome,
we were looking into further methods for Dmmb
removal. For this reason, cocktails containing 1 M
TMSBr, 1 M thioanisole, 1% EDT in TFA and 1 M
TFMSA, 1 M thioanisole in TFA9 have been tested. The
cleavage reactions to obtain compounds 10 and 12 were
carried out at ambient temperature for 1 and 20 h,
respectively. The Dmmb group could be successfully
cleaved and in the case of the TMSBr-assisted cleavage,
Table 1. List of synthetic peptides
Compd
Peptide
8
9
Z-Lys-Cys-Pro-Trp-Trp-OH
Z-Lys-(Dmmb)Gly-Pro-Trp-Trp-OH
Z-Lys-Gly-Pro-Trp-Trp-OH
H-Tyr-Ser-Leu-(Dmmb)Gly-Ala-Tyr-OH
H-Tyr-Ser-Leu-Gly-Ala-Tyr-OH
10
11
12
13
14
H-Tyr-Ser-Leu-(S-methyl-Dmmb)Gly-Ala-Tyr-OH
(H-Tyr-Ser-Leu-(Dmmb)Gly-Ala-Tyr-OH)2
Figure 2. Plot of the ligation kinetics. Formation of 8 (solid line) and 9
(dotted line). Peptide concentrations were 6 mM and the reaction was
carried out at ambient temperature. Relative peak areas in the HPLC
chromatograms were used to calculate the percentage of conversion.
Figure 3. Rearrangement of the ligation product under acidic condi-
tions: (A) internal thioester; (B) ligation product containing the amide
bond.
Vizzavona, Jean
