Angewandte
Communications
Chemie
dehydration protocol, thus setting dissimilar cleavable sites
(b-hairpins) when a l-Pro is added in the turn following the
[2]
before assembling the two parallel strands. Infrared analysis
unequivocally proved the formation of the isocyanide group,
key d-Pro residue. It is noteworthy that there are no
significant differences between the N-alkylated b-sheet
peptides 5a, 6a, and 7a, bearing either neutral or charged
moieties as part of the turn N-modification. This indicates
that the N-alkylation, and not the nature of the functionality,
plays a positive effect on the folding. On the other hand,
comparison of the CD spectra (Scheme 2B, lower panel) of
the N-alkylated peptides 5b, 6b, and 7b without the addi-
tional geminal methyl groups to reference peptide 8 reveals
loss of b-sheet content. Indeed, the conformational preorga-
nization imposed by the geminal methyl groups of the d-Pro-
DADME turn is crucial for b-sheet folding in aqueous
solution, regardless of the presence of an N-alkylation.
To our knowledge, there are no methods available for
labeling or ligating a b-sheet peptide without affecting the
strand termini or side chains, which are known to be crucial
for the b-sheet stability. In this regard, the capability of the
MCR approach to easily functionalize a parallel b-sheet
without modifying the strands shows great prospect in the
pursuit of biological applications. To exemplify this possibil-
ity, the commercially available cysteamine-trityl resin was
converted into the isocyano-resin and subjected to the Ugi
reaction to assemble the resin-linked d-Pro-DADME frag-
ment 9. As shown in Scheme 3, subsequent assembly of the
À1
according to the band around 2100 cm . For the Ugi reaction
to be carried out on resin, preformation of the imine is
achieved in solution by stirring the Alloc mono-protected
diamine and the aldehyde component in DCM/MeOH 1:1 for
1
h. The mixture is added together with Fmoc-d-Pro-OH
(
both in four-fold excess) to the isocyano-resin and the
reaction is shaken for maximum 1 day at room temperature.
The success of the Ugi reactions was assessed by RP-HPLC/
ESI-MS analysis after acidic mini-cleavage and by quantita-
tive cleavage of the Fmoc group and determination of the
absorption at 301 nm (see the Supporting Information). This
latter analysis also enabled the calculation of the loading of
the different peptide-functionalized resins 1–4, which is
crucial for the subsequent solid-phase peptide synthesis
(
SPPS). All four types of amino-alkyl-modified resin proved
suitable to be converted into isonitriles and to perform the
subsequent Ugi reactions, thus providing optimum flexibility
with respect to solid support and cleavage sensitivity.
Three degrees of orthogonality were employed during the
assembly of the two peptide strands using the Fmoc/tBu
methodology. Initial cleavage of the Fmoc group enabled the
growth of the upper strand, keeping the Alloc-protected
residue unaffected. Removal of this latter protecting group
and assembly of the lower strand completed the synthesis of
peptides 5–8 in good overall yield and purity of the crude
products (Scheme 1). The amino acid sequence chosen for
peptides 5–8 follows some previously established crite-
[7,8,16]
ria.
sheet formation, b) choosing positions to enable favorable
hydrophobic or electrostatic) interactions between paired
These are, a) having an intrinsic propensity for b-
(
residues in each strand, and c) adding some basic amino acids
to generate positive charges and therefore reduce aggrega-
tion.
As shown in Scheme 2, parallel b-sheets bearing a carbox-
amide (5a,b), a carboxylic acid (6a,b), and an amine (7a,b)
were produced, either with or withou the geminal dimethyl
groups in the diamine residue. Thus, we aimed at finding out
whether the turn-inducing capacity of the N-alkylated pepti-
domimetic fragment is sufficient to fold the peptides into
parallel b-sheets without the positive influence of the Thorpe–
Ingold effect provided by the geminal methyl substituents.
The versatility of the MCR method also enabled the facile
synthesis of the non-N-alkylated b-sheet, peptide 8, using the
acid-labile 2,4-dimethoxyphenyl substituent (arising from
the aldehyde component), which allows the cleavage at that
specific position.
Scheme 2B (upper panel) shows a comparison of the
circular dichroism (CD) spectra in acetate buffer (AcB) of the
N-alkylated peptides 5a, 6a, and 7a with the non-N-alkylated
one (8), all based on the DADME residue. The increment in
the intensity of the maximum at 198 nm and the minimum at
Scheme 3. Synthesis of a thiol-functionalized parallel b-sheet peptide,
its conjugation to a fluorescent label, and circular dichroism analysis.
PB: phosphate buffer; TCEP: tris(2-carboxy-ethyl)phosphine.
two peptide strands followed by mild acidic cleavage from the
resin led to the thiol-functionalized parallel b-sheet 10 in
good yield and moderate purity. Purified peptide 10 was
readily conjugated to the maleimide-functionalized dansyl
tag 11 in presence of tris(2-carboxyethyl)phosphine (TCEP)
to furnish the fluorescently labeled parallel b-sheet 12 in
excellent yield. The CD spectra of both peptides 10 and 12
show the characteristic b-sheet pattern described above,
confirming that the type of functionality attached to the
turn N-alkylation does not significantly affect the b-sheet
content in water.
[
17]
2
15 nm, in peptides 5a, 6a, and 7a suggests that the N-
alkylation of the turn-inducing moiety leads to a higher b-
sheet population. It is plausible to accept a positive effect of
the additional amide N-alkylation on the parallel b-sheet-
folding propensity; similar to its role in antiparallel b-sheets
Further to the CD evidence for a higher b-sheet content in
Ugi-peptides, we sought confirmation by NMR spectroscopy.
Initially, the NMR analysis was employed to assess the
Angew. Chem. Int. Ed. 2019, 58, 1 – 6
ꢀ 2019 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
These are not the final page numbers!