6086
T.S. Kale, J.D. Tovar / Tetrahedron 72 (2016) 6084e6090
amino acids, are known to induce flexibility within peptide
chains, while alanine, the next larger natural amino acid, is
residues are deprotonated. The repulsive forces among like charges
on the peptide backbones inhibit self-assembly and the peptide
molecules remain in what we consider to be a ‘molecularly dis-
solved’ state. Under acidic conditions, on the other hand, the
aspartic acid residues are protonated leading to self-assembly of
the peptide facilitated by the formal screening of the negative
charges. The self-assembly process leads to the formation of in-
4
0
a weak
solution.
b
-sheet former with a higher preference for
a
-helices in
Amino acids with progressively larger residues, such
-sheet formation,
with valine having the strongest preference in forming such tape-
4
1e43
as valine, isoleucine, and phenylalanine favor
b
42,43
like
b
-sheet rich motifs.
Variation in the primary peptide se-
quence can therefore lead to variation in the hierarchical com-
plexity of the 1-D supramolecular morphologies (stacked ribbons,
tapes, or fibrils) much like other amyloid-like oligopeptides.40,44
Starting from the short peptide sequence, DFAG, subtle single
amino acid substitutions at the internal locations, closest to the
chromophore unit, along the peptide sequences were made. These
substitutions were selected based on their relative size and hy-
drophobic nature, which could impact the intermolecular in-
teractions and influence hierarchical fibrillization processes.
Specifically, all four possible combinations of positioning the amino
acids alanine and glycine, namely DFAG (3), DFGG (4), DFAA (5) and
DFGA (6) were synthesized. The library of peptides thus generated
was then evaluated to understand the self-assembly properties as
well as the photophysical properties of the OPE units when em-
termolecular hydrogen-bonding among the amides (e.g.,
and forces the -electron systems into closer proximity, which
helps intermolecular electronic delocalization (Scheme 1). The
morphology of the nanostructures that form from the peptide-
peptide self-assembly process was studied using transmission
electron microscopy (TEM) and the photophysics of the embedded
chromophore units were evaluated using UVevis absorption,
photoluminescence (PL) and circular dichroism (CD) spectros-
copies. These studies were carried out using self-assembled nano-
structures generated in aqueous media where the pH was adjusted
to 10 using 1 M potassium hydroxide (KOH) solution or to 1 using
1 M hydrochloric acid (HCl).
b-sheets)
p
p
-
3.1. Effect of varying chromophore
bedded in the peptide backbone of these peptide-
molecules.
p-peptide
3.1.1. Nanostructure morphology. Low level energy minimization
calculations on these peptide-p-peptide molecules illustrate the
formation of 1-D nanostructures with the terminal phenyl rings on
the chromophore unit lying above and below the plane of the 1-D
peptide stack. The formation of such stacks due to self-assembly of
five molecules of peptides 1 and 3 is shown in Figure S1. Such as-
sembly may lead to formation of nanostructures that are con-
strained from ‘bundling’ due to steric hindrance from the terminal
phenyl rings preventing close inter-stack contacts. While the in-
ternal hydrogen-bonding networks within these stacks appeared to
deviate from ideal
enthalpic stabilization of the assembly. Previous molecular dy-
namics simulations of peptide- -peptide molecules found that the
deviations from ideal -sheet conformations can be attributed to
b-sheet architectures, they still allowed for
p
b
entropic mixing within each assembled stack as well as to internal
deformations caused by a combination of various stabilizing
2
0
hydrogen-bonding interactions. Also, the quadrupole associated
with the central -conjugated unit presents a distinctly non-
natural intermolecular interaction deviating these peptide assem-
blies from ideal protein secondary structures.
p
Chart 2. Structure of peptides 4e6.
The nanostructure morphologies of the peptide assemblies were
observed using TEM imaging. A solution of the peptide (1 mg/mL)
was acidified using 1 M HCl solution to generate a suspension of 1-
D nanostructures in aqueous media. A formvar coated copper grid
was then floated on a drop of this solution to adsorb the assemblies
onto the grid, followed by staining using 2% uranyl acetate and air
drying.
The self-assembling peptides were obtained by first synthesiz-
ing the peptide sequence using standard solid phase peptide syn-
thesis protocols starting with Wang resin loaded with Asp(OtBu)-
NH-Fmoc which was alternately deprotected (using 20% (v/v) so-
lution of piperidine in DMF) and coupled with the appropriate
amino acid (using HBTU as the coupling reagent and DIPEA as the
base). After completion of the required oligopeptide sequence, the
terminal free amine was subjected to a double N-acylation protocol
using 2,5-dibromoterephthalic acid and PyBOP as the coupling
agent leading to ‘on-resin’ dimerization of the peptide. The bromo-
functionalized peptide thus obtained was then reacted, while still
on the resin support, with the appropriate terminal alkyne using
As shown in Fig. 1, peptides 1e3 form 1-D nanostructures with
controlled, uniform diameters and variable nanostructure lengths.
The average nanostructure diameter calculated from at least 20
measurements was found to be 4.2ꢀ0.3 nm in the case of peptide 1,
4
.7ꢀ0.2 nm in the case of peptide 2 and 4.0ꢀ0.3 nm in the case of
peptide 3. The molecular lengths of these peptides were calculated
to be 3.06 nm, 3.08 nm and 3.19 nm for peptides 1e3, respectively,
using low level energy minimization calculations. This suggests
that the nanostructures obtained from these peptides are pre-
2
2
Sonogashira coupling conditions to yield the desired peptide-
peptide trimer which was then cleaved off the solid support and
purified using reverse phase HPLC. The peptide- -peptide mole-
cules were identified using ESI-MS and H NMR spectroscopy.
p-
p
1
dominantly comprised of isolated 1-D stacks of the peptide-p-
peptide molecules, in agreement with the model (Scheme 1b).
These stacks may resemble twisted tape-like stacks of individual
molecules which have limited interactions among adjacent stacks
due to the steric bulk of the oblique OPE units which project the
terminal phenyl rings above and below the average peptide plane.
3
. Results and discussion
The cruciform peptide-
p
-peptide molecules were designed to
undergo a pH triggered self-assembly process, similar to those in
our prior studies. Under alkaline conditions, the aspartic acid
This is in contrast to our previous work on ‘linear’ peptide-
tide molecules that shows significant bundling in some cases
p-pep-