Interactions of rationally designed small peptide dendrons functionalized with valine or sinapic acid with α-helix and β-sheet structures of poly-l-lysine and poly-l-glutamic acid

Article abstract of DOI:10.1002/pep2.24155

Uncontrolled protein oligomerization leading to deposition of fibrils is related to several diseases including neurodegeneration and diabetes. Involvement of natural compounds in regulation of this process has been documented. Therefore, the design and detailed study of bioinspired new molecular entities is one of the possible avenues to achieve better therapeutics. Here, we provide experimental data derived from the application of chiraloptic methods that rationally designed, bioinspired small branched peptides influenced the primary conformation of both poly-l-lysine (PLL) and poly-l-glutamic acid (PLGA) polypeptides in a structure- and concentration-dependent manner. In several cases, the circular dichroism (CD) spectra of polypeptide/dendron mixtures were considerably different from those corresponding to the individual polypeptides, in terms of significant reduction of intensity, discrete structure, and dislocation of characteristic bands. Data deconvolution suggested that compared to the individual homo-polypeptides, the resulting polypeptide/dendron complexes had a relative gain of distorted α-helix, and right- and left-hand twisted β-sheet forms, which may indicate a more diffuse structure. The electrostatic attraction and multiple hydrogen bonding between oppositely charged molecules, that is, between cationic-branched peptides and the β-sheet surface formed by anionic PLGA, might be the main cause of coaggregation that increased the variety and contribution of less ordered forms, and reduced the propensity for self-aggregation.

Full text of DOI:10.1002/pep2.24155

Received: 13 May 2019
Revised: 5 February 2020
Accepted: 14 February 2020
DOI: 10.1002/pep2.24155
F U L L P A P E R
Interactions of rationally designed small peptide dendrons
functionalized with valine or sinapic acid with α-helix and
β-sheet structures of poly-L-lysine and poly-L-glutamic acid
1
Maja Morawiak1 | Magdalena Stolarska2 | Maciej Cieslak
|
ꢀ
Zofia Urbanczyk-Lipkowska1
1Institute of Organic Chemistry Polish
Abstract
Academy of Sciences, Warsaw, Poland
2Department of Chemistry, Gdansk
Uncontrolled protein oligomerization leading to deposition of fibrils is related to sev-
eral diseases including neurodegeneration and diabetes. Involvement of natural com-
pounds in regulation of this process has been documented. Therefore, the design and
detailed study of bioinspired new molecular entities is one of the possible avenues to
achieve better therapeutics. Here, we provide experimental data derived from the
application of chiraloptic methods that rationally designed, bioinspired small
branched peptides influenced the primary conformation of both poly-L-lysine (PLL)
and poly-L-glutamic acid (PLGA) polypeptides in a structure- and concentration-
dependent manner. In several cases, the circular dichroism (CD) spectra of polypep-
tide/dendron mixtures were considerably different from those corresponding to the
individual polypeptides, in terms of significant reduction of intensity, discrete struc-
ture, and dislocation of characteristic bands. Data deconvolution suggested that com-
pared to the individual homo-polypeptides, the resulting polypeptide/dendron
complexes had a relative gain of distorted α-helix, and right- and left-hand twisted
β-sheet forms, which may indicate a more diffuse structure. The electrostatic attrac-
tion and multiple hydrogen bonding between oppositely charged molecules, that is,
between cationic-branched peptides and the β-sheet surface formed by anionic
PLGA, might be the main cause of coaggregation that increased the variety and con-
tribution of less ordered forms, and reduced the propensity for self-aggregation.
ꢀ
Polytechnic School, Gdansk, Poland
Correspondence
Zofia Urbanczyk-Lipkowska, Institute of
Organic Chemistry Polish Academy of
Sciences, Warsaw, Poland.
Email: zofia.lipkowska@icho.edu.pl
Funding information
National Science Centre of Poland, Grant/
Award Number: UMO-2012/07/B/ST5/01941
K E Y W O R D S
branched peptides, circular dichroism, fibrillation, sinapic acid, supramolecular interactions,
valine
1
|
INTRODUCTION
Moreover, as documented by numerous X-ray studies of proteins or
enzymes, formation of an active site is often assigned not to a mono-
Protein folding and aggregation are natural cellular processes coded
by the amino acid sequence. Folding enables construction of well-
defined 3D structures that are specific for a particular protein. The
resulting structure is characterized by a multifunctional surface that is
critical for various protein functions, such as molecular recognition,
transport, specificity of ligand binding, and further transformations.
mer unit but to protein aggregates. In homeostasis, proteins in mono-
meric and aggregated form are recognized, disassembled, cleaved, and
recirculated by the cell's enzymatic machinery (e.g., chaperones,
enzymes, etc.).[1,2] In pathological conditions misfolded proteins may
be involved in a process of self-assembly that cannot be controlled,
leading to formation of fibrillary deposits that are problematic for cell
Pept Sci. 2020;e24155.
wileyonlinelibrary.com/peptidesci
© 2020 Wiley Periodicals, Inc.
1 of 14
https://doi.org/10.1002/pep2.24155

2 of 14
MORAWIAK ET AL.
function and survival. Oligomerization of misfolded proteins under
impaired functioning of the chaperone protection system is thought
aggregation of amyloid peptides, for example, polyamidoamine
(PAMAM) and polypropyleneimine (PPI),[12] and poly-lysine[13] dendri-
mers, or show potential in antithrombotic therapy (PAMAM[14] or
PAMAM G4-Arginine-Tos[15]). However, densely charged cationic
dendrimers may interact with negatively charged blood and cell pro-
teins, and therefore amphiphilic branched peptides hold more thera-
peutic promise.
to be associated with
a number of human diseases, including
Alzheimer's disease, Parkinson's disease, and diabetes.[3]
Therefore, molecules that suppress self-aggregation or eliminate
abnormal and/or toxic protein aggregates are in high demand.
Recently two types of compounds have demonstrated their useful-
ness in controlling the aggregation process: small amphiphilic peptides
and natural polyphenols that are part of the molecular recognition
machinery and are regarded as defense against unwanted pro-
oxidation processes. The peptide-based design strategy was based on
the knowledge of strategic fragments in the amyloid beta peptide (Aβ)
sequence that, due to supramolecular interactions (electrostatic,
hydrophobic-hydrophobic or hydrogen bonding), are responsible for
Aβ self-recognition and further oligomerisation.[4] A great example is
the (Aβ17-21) sequence 5-peptide (Leu-Val-Phe-Phe-Ala) that has
been modified on the N- or C-terminus by coupling with basic amino
acid(s) or PEG fragments to obtain well-soluble 7 to 15 amino acid
long peptides that disturb fibril formation and promote disassembly of
the preformed fibrils.[5] As revealed in a recent study in an animal
model of Alzheimer disease, even more spectacular is the discovery of
the anti-aggregative potential of sinapic acid, a natural polyphenol
with a cinnamic acid structure.[6] It was found in two independent
in vivo studies that cognitive impairment associated with Alzheimer
disease, induced by injection of amyloid β (Aβ) 1-42 protein in mice[7]
or scopolamine in rats,[8] was abolished after oral administration of
sinapic acid.
Poly-L-lysine (PLL) and poly-L-glutamic acid (PLGA), known for
their ability to adopt pH and temperature-dependent, well-defined
proteinaceous conformations made them an important model in
studying amyloid-related disorders, such as Huntington's, Alzheimer's
or Parkinsons diseases.[16,17] Several studies have shown that the
aggregation propensity of homo-polypeptides (e.g., polyglutamine, Aib
homo-polypeptides, and PLGA) depends on the polypeptide. More-
over, polydispersity of the sample in terms of chain length enhances
transformation between unstructured monomeric and structured
aggregated forms.[18–21] Here we used polydispersed PLL or PLGA
homo-polypeptides for the preparation of proteinaceous secondary
structures. Circular dichroism spectroscopy (CD) was used to follow
interactions between these polypeptides and the designed dendrons
by studying the evolution of the secondary structure-related charac-
teristic bands as a function of dendron concentration. The relative
contributions of α-helix, β-sheet, “turn,” and “other” secondary struc-
tures were estimated using a deconvolution algorithm from the
BeStSel program.[22]
It was found that dendrons influenced the resulting CD spectra in
a
structure- and concentration-dependent fashion. Data
A fundamental principle of the normal or pathogenic fibrillation
process is the primary intermolecular recognition between protein
fragments of complementary chemical character. This idea was
recently used to explore chromatographic beads enriched with single
amino acids to form bulky hexapeptide combinatorial libraries for cap-
turing low abundant proteins from a mixture of natural origin.[9] It
appeared that within the single amino acids, a group of eight amino
acids (Arg, Lys, His, Phe, Tyr, Trp, Val, and Leu) called thereafter “great
catchers” could be selected that captured numerous proteins
suggesting that they might affect molecular recognition-related pro-
cesses in vivo as well.
deconvolution suggested that, compared to the individual homo-poly-
peptides, the resulting polypeptide/dendron aggregates exhibited a
varying rate of secondary structures, including distorted forms of
α-helix, as well as right- and left-hand twisted β-sheet forms in solu-
tion, which may indicate a more diffuse structure. Transmission elec-
tron microscopy (TEM) images showed that the fibrillation process of
polydispersed PLGA in a water solution occurred via self-assembly of
microspheres with a particle size distribution ~1 to 5 μm.
2
|
RESULTS
Therefore, the rationale for designing the present molecules was
to follow a peptide-based strategy to generate relatively small, but
bulky, first generation peptide dendrons constructed from three
amino acids from the above group: lysine (Lys) with a polar, basic side
chain (used here also for the development of a branched structure),
and two amino acids with hydrophobic side chains: valine (Val) with a
β-branched aliphatic and phenylalanine (Phe) with an aromatic side
chain. These dendrons were amphiphilic and exhibited high and vari-
able supramolecular potential in terms of the predominating form of
intermolecular interactions: hydrophobic for D1 and D2 and electro-
static and hydrogen bonding for the sinapic acid derivative D3. More-
over, as we found previously for compounds of similar architecture,
their molecular recognition potential might be enhanced, due to self-
assembly into larger aggregates.[10,11] Several other types of much
larger branched molecules have been discovered that abolish
2.1
|
Synthesis and conformational assignment of
dendrons D1 to D3
The general strategy for the synthesis of the D1 to D2 dendrons
involved application of the active esters method involving coupling in
solution of the N-orthogonally protected branching unit of the (Lys)
Lys(Lys)Phe(C12H23) structure, (Lys denotes lysine and Phe phenylala-
nine), with two (D1) or four (D2) (N-Cbz)-protected valines. After Boc-
deprotection, dendrons D1, and D2 were obtained. The (Lys) Lys (Lys)
Phe fragment of dendron D3 was synthesized on a solid support using
Phe-2chlorotrityl-resin preloaded with phenylalanine. In the final step,
Nε-amino groups were acylated with a dimethylformamide (DMF)
solution of sinapic acid. The crude products were purified by prepara-
tive high-performance liquid chromatography (HPLC) and the

MORAWIAK ET AL.
3 of 14
structure was confirmed on the basis of 1H and 13C nuclear magnetic
resonance (NMR) (at least 95% purity).
spectroscopy, apart from their intrinsic chirality, the studied branched
peptides self-aggregated, generating a structure-dependent induced
CD signal. The propensity for self-assembly was observed before for
amphiphilic dendrimeric peptides of similar architecture and is related
to their membrane disturbance mechanism.[10,11]
The details of the synthesis and analytical data for dendrons D1
to D3 are shown in Schemes 1 and 2 of the Experimental section.
As shown in Figure 1, lipodendrons D1 and D2 are amphiphilic
due to the presence of cationic amino groups supplemented with two
hydrophobic Boc and two Val residues (D1) or four Val residues (D2)
and a dodecyl residue located at the C-terminus. Dendron D3 is more
hydrophilic and prone to extensive intermolecular hydrogen bonding
due to the presence of two protonated amino groups and two pheno-
lic OH groups, as well as a carboxyl group located at the C-terminal
position. Each dendron was characterized by a structure-specific CD
spectrum whose shape and intensity for D1 and D2 was
concentration-dependent, and for D3 was concentration insensitive
(Figure 2). The CD spectrum for D2 exhibited two negative bands of
low intensity around 205 to 207 and 221 to 223 nm (< 4.8 mdeg for
500 μM). As seen in Figure 3, similar bands are typical for the α-helix
conformation. Both D1 and D3 exhibited an intense positive band at
~215 to 219 nm (< 15 mdeg for 500 μM). As revealed by CD
2.2
|
Preparation of secondary structures from
poly-lysine and poly-glutamic acid commercial samples
and their TEM images
The model polypeptides used were commercial samples of PLL hydro-
bromide and PLGA. These peptides were not monodispersed and con-
tained batch-dependent mixtures of polypeptides varying in chain
length and molecular weight (M.W.). (For PLL in the range
10 000-40 000 a.u.; for PLGA >50 000 a.u.) Literature data regarding
the generation of random, α-helix, and most of all β-sheet structures
from such samples report different conditions[18,20,21,23–26] and there-
fore, the necessary first step was to establish pH, temperature and
SCHEME 1 Synthesis of dendrons D1 and D2 in solution. Reagents and conditions: A, Boc-ROH, DCC/HOSu, DMF, 4 hours at 0 ꢀC to 5 ꢀC
and overnight at RT, yield: 95%; B, HCl/AcOEt, overnight at RT, yield 97%; C, Boc-Lys-BocOH, DCC/HOSu, DMF, 4 hours at 0 ꢀC to 5 ꢀC and
overnight at RT, yield: 91%; D, HCl/AcOEt, overnight at RT, yield 96%; E, Boc-Lys(Z)OH, DCC/HOSu, DMF, 4 hours at 0 ꢀC to 5 ꢀC and overnight at
RT, yield: 89%; F, H2, Pd/C, MeOH, 6 hours at RT, yield: 85%; G, Cbz-Val, DCC/HOSu, DMF, 4 hours at 0 ꢀC to 5 ꢀC and overnight at RT, yield:
86%; H, H2, Pd/C, MeOH, 6 hours at RT, yield 75%; I, HCl/AcOEt, overnight at RT, yield 92%; J, Cbz-Val, DCC/HOSu, DMF, 4 hours at 0 ꢀC to 5 ꢀC
and overnight at RT, yield: 80%; K, H2, Pd/C, MeOH, 6 hours at RT, yield 70%. DIC, diisopropylcarbodiimide; DMF, dimethylformamide

4 of 14
MORAWIAK ET AL.
SCHEME 2 Synthesis of dendron D3 on solid support. Reagents and conditions: A, 20% piperidine, 2 × 5 minutes at RT, B, Fmoc-Lys(Fmoc)-
OH, HATU, DIPEA, 4 hours at RT, C, 20% piperidine, 2 × 10 minutes at RT, D, Boc-Lys(Fmoc)-OH, HATU, DIPEA, 4 hours at RT, E, 20%
piperidine, 2 × 10 minutes at RT, F, sinapic acid, DIC, Oxyme, DIPEA, 4 hours at RT, G, 50% TFA/DCM, 2 hours at RT. Yield: 18%. DIC,
diisopropylcarbodiimide
FIGURE 1 Molecular structure of dendrons D1-D3
time ranges for the formation of a particular secondary structure in
solution. According to our experiments, at pH 12, the water solution
of PLL revealed a well resolved α-helix conformation, whereas further
heating to 52 ꢀC for 15 minutes induced an α-helix to β-sheet trans-
formation (Figure 3). The solution containing the polydispersed PLL
sample followed α-helix/β-sheet transformation kinetics observed for
the solutions of PLL containing samples of 84 000, 25 500, and 9200
a. u. molecular weight, respectively.[27] On the contrary, the α-helical
structure of PLGA in water was prompted by the pH decrease to 4.2,
which, when followed by heating the solution to 65 ꢀC for
30 minutes, showed the α-helix to β-sheet transformation. As shown
in Figure 4 and S1 the decreased α-helix content and increased
β-sheet structure analyzed by the deconvolution program involved an
increase of not only a percentage of the β-sheet structure, but also a
family of conformations defined as “others” and the appearance of
~10% of turn conformations. Prolonged heating (24 hours) of both
polypeptide solutions yielded essentially the same CD profile typical
for the β-structure but with lower intensity. Therefore, for studying
the interactions of dendrons with the β-structure of polypeptides, the
experimental conditions involved the stabilization of primary solutions
of dendrons and polypeptides at T > 50 ꢀC or T > 60 ꢀC, respectively
for PLL (Figure 3, left) and PLGA (Figure 3, right) and controlling the
temperature during CD spectra acquisition by a spectrometer Peltier
system.
Rapid CD experiments performed by Mendonça and Hache using
a nanosecond 82 MHz titanium-sapphire laser followed the helical
fraction evolution during heating a monodispersed PLGA sample
(molecular weight 84 000). The authors concluded that only CD at
220 nm is relevant to observe the unfolding of an α-helix and its
transformation into the β-sheet form, whereas no change is observed
for CD at 204 nm. This observation is in agreement with the present
CD data and spectra deconvolution.[28]
An interesting discovery was made when the morphology of
3-month-old solid deposits of both polypeptides in water was
observed under a transmission electron microscope. As shown in
Figure 5, because of partial water evaporation from the microscope
slide, the initially transparent solution containing primary fibrils
(pH 4, 25 ꢀC), reaches a critical micelle concentration (CMC) and
PLGA (probably in α-helical form) self-assembles into globular
aggregates with a particle size distribution ~1 to 5 μm, with the
majority of microspheres around 1 μm. A magnified picture of the
marked area shows that small microspheres had a tendency for
directional aggregation into a fibril-like structure. This unique type
of assembly was also revealed in the mature fibers where the
microspheres were slightly elongated in a direction perpendicular
to the fiber axis. These phenomena were not observed in the PLL
solutions, where much smaller nonfibrillar deposits precipitated
from the water solution.

MORAWIAK ET AL.
5 of 14
FIGURE 2 CD spectra of dendrons D1-D3 recorded in methanol (spectra not smoothed)
FIGURE 3 CD spectra of PLL and PLGA commercial samples (not smoothed), recorded in pH-adjusted water solutions showing the formation
of secondary structures at various temperatures and time of heating. PLL, poly-L-lysine; PLGA, poly-L-glutamic acid

6 of 14
MORAWIAK ET AL.
Structural similarity in terms of the advantage of parallel over
the existing secondary structures (Fourier-transform infrared spec-
troscopy, CD, fluorescence spectroscopy, X-Ray, atomic force micros-
copy and TEM), CD spectroscopy is often used for detecting the
in vitro secondary structures of peptides and proteins involved in the
development and progression of the disease and for studying factors
that interfere with this process.[32,33]
antiparallel β-sheet subpopulations in the primary PLGA β-sheet
model and in amyloid peptides[29] and Aβ1-40 oligomers was also
found.[30]
2.3
|
Detecting interactions of dendrons with
Our studies involved the formation of proteinaceous secondary
structures in water solution from polydispersed PLL hydrobromide or
PLGA and application of CD spectroscopy for studying their interac-
tions with three rationally designed branched peptides of different
structure. Samples used for CD spectra acquisition contained the
same amount (μg/mL) of the respective homo-polypeptides and the
same volume of dendron solution in an increasing concentration. To
diminish the contribution of the original conformation of dendrons to
the conformation-characterizing bands of the polypeptides, CD spec-
tra obtained for polypeptide/dendron mixtures were corrected by
subtracting the spectrum of the pure dendron at the appropriate
concentration.
secondary structures of PLL and PLGA by CD
spectroscopy
2.3.1
|
Interactions of dendrons with an α-helix
structure generated from PLL and PLGA
The formation of insoluble fibrillar aggregates rich in β-sheet structure
by a normally soluble protein is regarded as a hallmark for a range of
diseases including Alzheimer's disease, Parkinson's disease, prion's dis-
ease and type II diabetes.[31] In studies related to neurodegenerative
diseases, PLL and PLGA often serve as a model of protein aggregation
and fibrillogenesis. Among available techniques for the assignment of
Conformational transitions were observed as fluctuation of the
CD signal in the 200 to 260 nm wavelength range as a function of
dendron concentration. The relative contribution of a particular sec-
ondary structure can be derived using several algorithms for
deconvolution of the experimental CD spectra.[34] For the present
studies, deconvolution was carried out with the BeStSel system.[22]
This program was specifically designed for handling β-structures of
amyloid peptides and other proteins with a high content of the
β-sheet structure. Moreover, the program recognized parallel and anti-
parallel variations of this structure, as well as a twist of the originally
planar β-sheet that might be important when the pathway adopted
for transformation of the α-helix to the β-sheet structure was of
interest.
CD spectra of cationic PLL or anionic PLGA at increasing dendron
concentrations are shown in Figures 6–8. Figure 6 presents the
impact of dendrons on the α-helical conformation formed by PLL
(Figure 6A–C) or PLGA (Figure 6D). Spectra of pure homo-
polypeptides had two characteristic negative absorption bands, more
intense at ~208 nm and less intense at ~ 222 nm (the respective
FIGURE 4 Transformation of PLGA α-helix to β-sheet structure
upon heating at pH = 4. PLGA, poly-L-glutamic acid
FIGURE 5 Transmission electron
scanning microscope image of
3-month-old fibrils of PLGA above
the CMC at pH = 4 (left); magnified
view of inlet shows the formation of
micelles of various sizes and their
self-assembly into fibrillar forms
(right); scale bar = 20 μm. CMC,
critical micelle concentration; PLGA,
poly-L-glutamic acid

MORAWIAK ET AL.
7 of 14
FIGURE 6 CD studies on the influence of dendrons D1-D3 on preformed α-helix secondary structures generated from polycationic PLL, A-C,
and polyanionic PLGA, D, upon addition of an increasing amount of dendron (inlets show effective dendron concentrations after dilution). CD,
circular dichroism; PLL, poly-L-lysine; PLGA, poly-L-glutamic acid
π ! π* and n ! π* transition), corresponding to the α-helix of the
peptide bond (red line). The PLL intensity, but not the position of the
characteristic bands, slightly fluctuated when titrated with D1 and D2
at higher concentrations, suggesting that the resulting intermolecular
interactions were weak. More significant was the impact of sinapic
acid functionalized D3. After the addition of D3 above a concentra-
tion of 133 μM, a decrease in intensity in λ = 208 nm was observed
that was not associated with the corresponding changes in
λ = 222 nm. Such changes were often related to fluctuation in “turn”
structure content, since it had a maximum at this wavelength.
Table 1). This observation is in agreement with the results of rapid CD
experiments performed by Mendonça and Hache for monodispersed
PLGA sample (molecular weight 84 000).[28]
Populations of secondary structures detected during titration of
PLL or PLGA in the α-helix conformation with dendrons were calcu-
lated by the BeStSel program and are shown in Table 1. The results
revealed that the prevailing α-helix structure and a small amount of
β-sheet, as well as a significant amount of β-turn (10%-20%), and con-
formations defined as “others” were present in the primary solution.
Upon the addition of a small quantity of D1 to D3 dendrons to the
PLL solution, the amount of regular α-helix was slightly decreased,
whereas the amount of the distorted form was conserved. Addition of
75 μM of D1 or D2 resulted in an increase of the total amount of the
β-sheet structure, up to ca 21%, with a concomitant decrease of the
range of structures defined as “others.” Interestingly, only “relaxed”
and “right-hand-twisted” but not “left-hand twisted” β-sheet forms of
PLL were observed. Titration with D3 at 200 and 266 μM had a more
significant impact; the initial content of regular and distorted α-helix
The surface of PLGA molecules in either form was, at least in part,
negatively charged in a pH-dependent manner or contained neutral
carboxyl groups. Therefore, it was expected that the influence of rela-
tively small positively charged dendrons on the α-helix and β-sheet
structures of PLGA (M. W. >50 000 a.u) might be insignificant. Indeed,
dendrons D1 to D3 had almost no impact on the α-helical conforma-
tion formed by PLGA, with only a small intensity fluctuation in the CD
curve, as shown for the representative dendron D2 (Figure 6D and

8 of 14
MORAWIAK ET AL.
FIGURE 7 CD studies on the influence of D1-D3 dendrons on β-sheet structures generated from polycationic PLL. CD, circular dichroism;
PLL, poly-L-lysine
was reduced from ~70% to 75% to ~55%. The interrelation between
the β-turn and β-sheet structures was interesting. Increasing the D3
concentration to 133 μM resulted in the sudden increase of β-turn
content to ~21%, whereas at 200 μM, almost all the β-turn forms
were converted into the β-sheet form. Doubling the D3 concentration
to 266 μM resulted in the decreased content of the β-sheet form to
11%, with an increase in the β-turn content to ~ 32%. This may sug-
gest that upon interaction with amphiphilic D3, either longer PLL
chains became bent or D3 is connected to the C-terminal fragments
of two α-helices via formation of multiple hydrogen bonds and elec-
trostatic interactions. Both structures in the next step may have facili-
tated or strengthened the existing β-sheet pattern.
dendrons is shown in Figures 7 and 8, respectively. Micsonai et al.
showed by analysis of the CD spectra and the respective molecular
conformations for proteins deposited in the Protein Data Base, these
containing ca 50% α-helix show similar CD curve profiles, whereas
those having ca 50% β-sheet, even with negligible α-helical content,
show CD spectra with diverse amplitudes.[35]
Indeed, titration of PLL in the β-sheet conformation by D1 and
D2 resulted in a lowered intensity, more discrete structure and a
shifted position of the negative band at 216 to 219 nm to a higher
wavelength (~227 nm) (Figure 7AB). Moreover, even shifts in the CD
trace for most wavelengths were observed only upon titration with
D1, but not D2. The impact of dendron D3 on the β-sheet structure
of PLL was much smaller in terms of reducing the band intensity and
shifting the band position (Figure 7C). CD spectra deconvolution in
the case of D3 (Table 2) showed a decrease in the β-sheet content in
favor of an α-helix and a significant increase of structures defined as
“others” (up to 52%-59%). Spectral deconvolution can give reliable
results when the measured CD spectra are at least in the 200 to
250 nm λ range.[22] For this reason, calculations for several CD
2.3.2
|
Interactions of dendrons with the β-sheet
structure generated from PLL and PLGA
The influence of dendrons on the β-sheet conformation formed by
PLL and PLGA homo-polypeptides in the absence and presence of

MORAWIAK ET AL.
9 of 14
FIGURE 8 CD studies on the influence of D1-D3 dendrons on the β-sheet structure generated from polyanionic PLGA. CD, circular
dichroism; PLGA, poly-L-glutamic acid
spectra measured above 205 nm could not be performed: the entire
entry from Figure 7A,B: (PLL-β-sheet + D1) and (PLL-β-sheet + D2)
and those of higher concentrations of PLGA β-sheet + D2 (Figure 8B).
CD spectra deconvolutions calculated by the BeStSel program
(Table 2) showed that the total amount of the generated β-structure
in primary solution depended on the type of polypeptide. It was
around 46% for PLL, and only 24% to 33% in experiments involving
PLGA titrated with D1, D2, and D3, respectively. Primary solutions of
both homo-polypeptides expressed an advantage of parallel over anti-
parallel β-sheet conformations, and the presence of a significant
amount of structures defined as “others” (45%-54%) and ~ 10%) of
β-turn structures.
structures. According to the deconvolution procedure, this
accounted for the development of a range of conformations defined
as “others” (~44%). At 125 μM, the CD curve showed a new mini-
mum at 207 nm (Figure 8A). Structure deconvolution for this con-
centration revealed an increase in “turn” (~21%) and β-sheet (~32%)
content and a decreased percentage of conformations defined as
“others” (~32%). Dendron D2 [4 Val and (+) 4 charge] showed only
a small fluctuation in intensity but not in the positioning of the neg-
ative band at 217 nm (Figure 7C), thus suggesting structural cooper-
ativity with the β-sheet structure. In contrast, dendron D3 at a
higher concentration changed the shape, intensity, and structure of
the CD trace in a concentration-dependent manner. At 75 μM, the
CD trace had the shape of a wide plateau and a discrete structure
with the intensity approaching zero. However, reduction of the
twisted and a higher percentage of the relaxed β-sheet form might
suggest a stabilizing effect of D3. Interestingly, the observed differ-
ences in the shape and bands intensity of the CD spectra were not
reflected by spectra deconvolution, since only several percentage
As shown in Figure 8A-C, dendrons show structure-dependent
behaviour. From visual assignment, dendrons D1 and D2 interacted
in a different way with PLGA formatted to the β-sheet form.
Dendron D1 (2 Val and [+2] charge) in the 12.5 to 25 μM concen-
tration range showed the appearance of an α-helix. Above a con-
centration of 50 μM, spectra exhibited lower intensity and discrete

10 of 14
MORAWIAK ET AL.
TABLE 1 Populations (%) of secondary structures observed in solution of PLL or PLGA in α-helix form upon addition of increasing
concentrations of dendrimers
α-Helix (%)
β-Sheet (%)
Antiparallel
Turn
(%)
Other
(%)
α-helix Denrimer/conc. (μM)
PLL
Regular
25.4
27.0
25.7
28.8
23.5
25.4
19.9
19.0
29.8
18.9
55.2
51.5
54.8
52.4
46.2
33.9
48.5
51.3
51.1
50.4
50.4
49.8
Distorted
14.0
13.2
14.8
14.8
12,0
14.0
12.2
12.6
11.4
11.3
20.3
20.0
18.1
20.0
15.6
20.7
19.7
20.1
20.1
20.5
20.2
20.2
left-H
0
relaxed
right-H
0
parallel
1.5
0
6.2
11.5
0
15.2
14.0
15.2
15.5
14.4
15.2
12.4
14.4
12.8
12.7
6.6
37.6
34.1
37.2
29.3
28.3
37.6
35.7
38.2
38.7
37.8
18.0
18.5
11.0
0
PLL + D1/12
PLL + D1/25
PLL + D1/50
PLL + D1/75
PLL
0
0.2
7.1
12.4
0
0
0
0
0
0
0
21.3
6.2
7.2
3.3
6.3
5.0
0
0
0
0
1.5
4.3
4.5
3.8
2.7
0
PLL + D2/12
PLL + D2/25
PLL + D2/50
PLL + D2/75
PLL α-helix
0
8.2
8.0
8.5
11.7
0
0
0
0
0
PLL + D3/33
PLL + D3/66
PLL + D3/133
PLL + D3/200
PLL + D3/266
PLGA α-helix
PLGA + D2/30
PLGA + D2/60
PLGA + D2/120
PLGA + D2/180
PLGA + D2/266
0
0
0
0
10.0
10.0
22.7
1.4
0
0
13.9
0
0
4.1
0
0
0.8
0
0
19.8
11.0
3.4
3.2
1.8
4.1
1.9
1.5
17.0
2.5
0
0
0
31.9
10.6
9.9
0
0
0
17.9
15.4
17.2
18.2
18.0
18.2
0
0
0
0
0
0
9.8
0
0
0
6.7
0
0
0
9.6
0
0
0
10.3
Abbreviations: PLL, poly-L-lysine; PLGA, poly-L-glutamic acid.
fluctuations from the initial content of the secondary structures
were found (Table 2). One of the possible reasons is that it is hard
to interpret the “not classic” shape of the CD trace and the fact that
the development of deconvolution procedures are still mostly based
on databases of larger well-ordered proteins.[36]
interacting polypeptides and dendrons were formed with variable
conformation rates. In particular, the observed fluctuation between
relaxed, parallel and antiparallel right- and left-hand shifted β-sheet
may introduce “chiral noise” to the intensity and fine structure of the
CD signal. The highest impact on the α-helix structure formed by PLL
(pH 12) had dendron D3, whereas all cationic dendrons had little
impact on the α-helix structures formed by polyanionic PLGA (pH 4).
However, addition of the structurally different dendrons D1 and D3
influenced the shape, intensity, and band position in the CD spectra
of PLGA in the β-sheet conformation. Dendron D2 bearing four cat-
ionic centres flanked by bulky valine, dodecyl and tert-butoxycarbonyl
(Boc) groups affected the secondary structure composition to a lesser
degree, probably due to inhibition of the generation of multiple hydro-
gen bonds.
The amphiphilic dendrons D1 to D3, in addition to their intrinsic
chirality, had the ability to self-assemble and therefore, to form sec-
ondary structures which generated an induced CD signal. Due to
intermolecular interactions dominated by electrostatic and hydrogen
bond forces, aggregated particles influenced the populations of the
secondary structures formed by the homo-polypeptides PLL and
PLGA in a structure- and concentration-dependent manner. Confor-
mational transitions were followed by the evolution of the CD signals
at structure-characteristic wavelengths (α-helix, β-turn, β-sheet, etc.)
as a function of dendron concentration. CD spectra of several poly-
peptide/dendron mixtures were considerably different from those
corresponding to the individual polypeptides. The highest impact of
dendrons was visualized by the reduction of the intensity, discrete
structure, and dislocation of characteristic bands in the CD spectra.
CD spectra deconvolution showed that, in fact, complex mixtures of
Supramolecular mechanisms for dendron-polypeptide interac-
tions considering single dendron molecules could be different from
variations of capping-related mechanisms proposed by Klainert
et al., for explaining the decomposition of fibrillar forms of Aβ pep-
tides by third-generation PPI or third- to fifth-generation PAMAM
dendrimers.[12,37] First, the PPI and PAMAM dendrimers used in

MORAWIAK ET AL.
11 of 14
TABLE 2 Populations (%) of secondary structures observed in solution of PLL or PLGA in β-sheet form upon addition of increasing
concentrations of dendrimers
α-Helix (%)
β-Sheet (%)
antiparallel
Turn
(%)
Other
(%)
Polypeptide+ dendrimer
PLL β-Sheet Denrimer/conc. (μM)
PLL + D3/12
regular
0
distorted
2.3
left-H
0
relaxed
7.3
right-H
8.9
parallel
29.6
0
5.5
14.0
11.4
11.1
9.0
46.3
41.8
45.0
48.1
59.7
52.5
54.2
33.9
41.4
44.7
42.9
35.1
40.5
43.3
40.5
41.6
41.7
45.7
48.5
43.4
8.5
21.7
20.9
12.1
2.3
4.0
19.6
2.9
0
2.2
0
11.5
9.1
0.2
PLL + D3/24
12.8
11.3
2.3
0
0
0
PLL + D3/47
0
0
8.6
0
PLL + D3/74
0
0
10.4
11.2
0.6
15.9
16.9
14.7
7.7
0
PLL + D3/94
4.8
0
1.9
10.5
12.1
13.4
13.7
16.0
12.0
21.2
2.9
PLGA β-Sheet
5.6
4.7
0
4.1
PLGA + D1/12
PLGA + D1/25
PLGA + D1/50
PLGA + D1/75
PLGA + D1/125
PLGA β-Sheet
12.1
7.2
6.4
6.9
0
9.7
3.3
5.4
0
13.2
18.9
17.0
0
1.9
11.9
0
4.7
3.8
27.4
1.8
0
3.9
9.8
2.6
0
7.3
4.8
15.4
0
0
4.9
29.1
14.9
34.0
0
PLGA + D2/12
PLGA + D2/24
PLGA β-Sheet
5.1
7.8
8.2
3.4
4.6
2.3
2.7
5.5
15.3
9.5
3.1
8.7
0.5
0
7.4
5.9
6.0
1.7
0
6.2
18.6
17.1
19.8
19.1
18.2
11.2
12.1
13.3
12.9
14.4
13.0
13.4
11.4
13.6
13.4
PLGA + D3/12
PLGA + D3/25
PLGA + D3/50
PLGA + D3/75
5.1
0
3.9
1.7
0
3.3
2.3
2.8
0
Abbreviations: PLL, poly-L-lysine; PLGA, poly-L-glutamic acid.
those studies were relatively large, polycationic compounds bearing
on the surface 16 to 128 amino groups enabling the formation of
an extensive system of multiple hydrogen bonds and electrostatic
interactions with the existing of Aβ fibrils. In contrast, the presently
studied dendrons were much smaller molecules with only (+) 2 or
(+) 4 charges. However, their amphiphilic structure enabled by the
presence of cationic centers and several lipophilic groups might be
advantageous in the creation of multiple interactions with the exis-
ting secondary structures. Moreover, as revealed by CD spectros-
copy, as a part of their intrinsic chirality, the studied branched
peptides self-aggregated generating structure-dependent induced
CD signals. Such behavior is characteristic of natural antimicrobial
peptides that make large deposits on negatively charged microbial
membranes.[38]
electrospray ionization (ESI) time-of-flight mass spectrometer
(PerSeptive Biosystems) for the samples prepared in methanol
(MeOH). The 1H-NMR spectra were recorded using a Bruker Avance
spectrometer at 500/125 using deuterated solvents and TMS as an
internal standard. Chemical shifts are reported as δ values in parts
per million, and coupling constants are given in hertz. Thin layer
chromatography (TLC) was performed on aluminium sheets with the
silica gel 60 F254 from Merck. Column chromatography (CC) was
carried out using silica gel (230-400 mesh) from Merck or Sephadex
LH20. The TLC spots were visualized by treatment with 1% alcoholic
solution of ninhydrin and heating. PLL hydrobromide (M.W.
4000-20 000) and PLGA (M.W. > 50 000) were obtained from MP
Biomedicals. MeOH, sodium hydroxide, and hydrochloric acid were
purchased from Sigma-Aldrich. Water solutions of both polypeptides
were pH-adjusted with HCl or NaOH solutions and controlled by a
Mettler-Toledo pH-meter.
3
|
EXPERIMENTAL
Materials
3.1
|
3.2
|
Sample preparation
All solvents and reagents were of analytical grade and were used
without further purification. All reagents and solvents were obtained
from Sigma-Aldrich. Mass spectra were recorded with a Mariner
Stock solutions of D1 to D3 (1 mM) were prepared by dissolving the
respective sample in MeOH. The solutions were stored at room tem-
perature (RT).

12 of 14
MORAWIAK ET AL.
Polypeptides (2 mg/mL) were obtained by dissolution in distilled
and washed with DMF, the Fmoc group was removed, as previously
described. After draining the resin, the acylation procedure was
repeated for 4 hours with a solution containing Boc-Lys (Fmoc) OH
(1.2 g; 2.56 mmol), HATU (1.946 g; 5.12 mmol), and DIPEA (1.65 g;
12.77 mmol) in anhydrous DMF. After being drained and washed with
DMF, the Fmoc group was removed and the resin was washed with
DMF. The last acylation procedure was performed for 4 hours with a
solution of sinapic acid (0.574 g; 2.56 mmol), diisopropylcarbodiimide
(DIC) (0.646 g; 5.12 mmol), Oxyme (0,727 g; 5.12 mmol), and DIPEA
(1.65 g; 12.77 mmol) in anhydrous DMF. After being drained and
washed with DMF, the peptide-resin was deprotected and released
by treatment with a TFA/DCM (5:5) solution at RT for 2 hours. The
resin was filtered off and washed with ethyl acetate. The volatiles
were then removed in vacuo, and the crude product D3 was precipi-
tated twice with diethyl ether. The crude product D3 was purified by
preparative HPLC using a C18 column, 250 × 21.20 mm, particle size
15 μm and a pore diameter of 300 Å. The mobile phase consisted of a
gradient from 5% to 95% MeOH/H2O, 0.05% HCl, at a flow rate of
3.0 mL/min. Final yield: 18%. The synthetic pathway for dendron D3
is shown in Scheme 2.
water and then diluted to the appropriate concentration. The proper
dilution was determined on the basis of the high tension (HT) value,
which was determined during the CD measurement. The HT should not
be higher than 600 V (lower concentration = lower HT). The correct pH
for a given peptide was obtained by adding NaOH to PLL (pH 12) and
HCl to PLGA (pH 4), respectively. The pH value was measured using a
LE438 electrode with a built-in temperature sensor connected to a
Mettler-Toledo FiveGo pH meter. The peptides thus prepared were
mixed with methanolic dendron solutions at a suitable rate.
3.3
|
Synthesis and spectroscopic data of dendrons
D1 to D3
Peptide dendrons were obtained using the active esters method as
shown in Scheme 1. The substrate with free amino groups was dis-
solved in MeOH, and then 50% excess of corresponding active ester
in MeOH was added successively. The reaction was carried out at RT
for 5 days, then the solvent was evaporated to dryness and the post-
reaction mixture was dissolved in MeOH. The compounds were puri-
fied on a Sephadex LH-20. Then, the solvent was removed in vacuo
resulting in the title compound as lightly yellow oil.
Dendron D3: C49H67N7O13, M.W. 962.11, Exact Mass: 961.48,
m/z: 962.6, Elemental Analysis: C, 61.17; H, 7.02; N, 10.19; O, 21.62;
MS (ESI, MeOH): m/z: 984.6 [M + Na] +, 962.6 [M + H] +.
The synthetic pathway for dendrons D1 and D2 is shown in
Scheme 1.
1H NMR (600 MHz, MeOD), δ: 1.23 to 1.98 (m, 26H, 3xβCH2 Lys,
3xγCH2 Lys, 3xδCH2 Lys, 3xεCH2 Lys, βCH2 Phe), 2.97to 3.25 (m, 2H,
2xαCH Lys), 3.35 to 3.41 (m, 2H, αCH Phe, αCH Lys), 3.86 to 3.96 (m,
12H, 4xOCH3 sinapic acid), 6.37 to 6.54 (m, 8H, 4xH-Ar sinapic acid, 2x
─CH=CH─ sinapic acid), 6.85 to 6.95 (m, 9H, 5xεNH Lys, 2xαNH2 Lys),
7.12 to 7.36 (m, 5H, H-Ar Phe), 7.40 to 7.51 (m, 2H, 2xOH sinapic acid).
Dendron D1:
C59H106N10O10, M.W. 1115.53, Exact Mass:
1114.81, m/z: 1114.81, Elemental Analysis: C, 63.52; H, 9.58; N,
12.56; O, 14.34; MS (ESI, MeOH): m/z: 1138[M + Na]+.
1H NMR (500 MHz, MeOD), δ: 0.75 to 0.85 (m, 9H, CH3 DDA,
2xCH3 Val), 1.2 to 1.9 (m, 62H, 3xβCH2 Lys, 3xγCH2 Lys, 3xδCH2 Lys,
2xβCH2 Val, 2xC[CH3]3, CH2 C2-11 DDA), 2.5 to 2.8 (m, 10H, 3xεCH2
Lys, βCH2-Phe, CH2 C1 DDA), 2.9 to 3.1 (m, 4H, 2xCH2 Val), 3.9 to 4.3
(m, 6H, 3xαCH Lys, αCH Phe, 2xαCH Val), 7.3 to 7.65 (m, 5H, H-
Ar-Phe).
3.4
|
Circular dichroism measurements
During conformational studies on the influence of the D1 to D3 den-
drons on the α-helix structure generated from PLL or PLGA, CD spectra
were measured at RT with a Jasco J-715 spectropolarimeter using a
0.2 cm path length quartz cuvette and (1:1, 1:2) vol/vol ratio between
peptide and dendron. The HT of the anode did not exceed 600 V. For
studies concerning their impact on the β-sheet structure, CD spectra
were collected with a Jasco J-815 spectropolarimeter fitted with a
Peltier temperature controller, using a rectangular quartz cuvette with a
fitted cap and an optical path length of 1 to 0.2 cm and (3:1) vol/vol
ratio between peptide and dendron. CD measurements were performed
after the samples had been heated to the appropriate temperature at a
given time. PLL creates a β-sheet structure after 15 minutes at 52 ꢀC
and PLGA after 30 minutes at 65 ꢀC. Samples of both polypeptides
were dissolved in water with the addition of HCl or NaOH for pH
adjustment, whereas dendrons were dissolved in MeOH.
Dendron D2:
C59H108N12O8, M.W. 1113.56, Exact Mass:
1112.84, m/z: 1112.84, Elemental Analysis: C, 63.64; H, 9.78; N,
15.09; O, 11.49; MS (ESI, MeOH): m/z: 1114[M + H] +.
1H NMR (500 MHz, D2O), δ: 0.7 to 0.9 (m, 15H, CH3 DDA,
4xCH3 Val), 1.2 to 2.1 (m, 48H, 3xβCH2 Lys, 3xγCH2 Lys, 3xδCH2 Lys,
4xβCH2 Val, CH2 C2-11-DDA), 2.7 to 3.0 (m, 12H, 3xεCH2 Lys, 2xβCH2
Phe, CH2 C1-DDA), 2.8 to 3.1 (m, 8H, 4xCH2 Val), 3.9 to 4.5 (m, 8H,
3xαCH Lys, αCH Phe, 4xαCH Val), 7.15 to 7.6 (m, 5H, H-Ar Phe).
Synthesis of dendron D3 on solid support
Fmoc-protected Phe-2chlorotrityl-resin (resin preloaded with phenyl-
alanine) (1 g; equiv. 0.82 mmol/g) was swollen in DMF for 2 hours.
The Fmoc group was removed using two 5 minutes treatments with
2:8 piperidine/DMF, and washed thoroughly with DMF. Once dra-
ined, the resin was acylated with a solution containing Fmoc-Lys
(Fmoc) OH (0.775 g; 1.3 mmol)), [2-(7-Aza-1H-benzotriazole-1-yl)-1,
1, 3, 3-tetramethyluronium hexafluorophosphate] (HATU; 0.973 g;
2.56 mmol), and N, N-Diisopropylethylamine (DIPEA; 0.827 g;
6.4 mmol) in anhydrous DMF for 4 hours at RT. After being drained
The β-sheet structure cell conditioning at the appropriate tempera-
ture by the spectrometer Peltier system was continued for 15 minutes.
Then, the same volume of MeOH solution of dendrons at the appropriate
temperature and concentration was added to the cell and the CD spec-
trum was measured after ca 5 minutes. All spectra were averaged over at

MORAWIAK ET AL.
13 of 14
least five scans and baseline-corrected with a mixture of water with an
appropriate pH and MeOH as the reference. To diminish the contribution
of the original conformation of dendrons to the conformation-
characterizing bands of the polypeptides, the CD spectra obtained for the
polypeptide/dendron mixture were corrected by subtracting the spec-
trum of the pure dendron at the appropriate concentration.
REFERENCES
[1] P. Alam, K. Siddiqi, S. K. Chturvedi, R. H. Khan, Int. J. Biol. Macromol.
2017, 103, 208.
[2] R. A. M. Meijering, R. H. Henning, B. J. J. M. Brundel, Trends Cardi-
ovasc. Med. 2015, 25, 243.
[3] B. Mannini, F. Chiti, Front. Mol. Neurosci. 2017, 1, 98.
[4] Y. L. Xiao, B. Y. Ma, D. McElheny, S. Parthasarathy, F. Long, M. Hoshi,
R. Nussinov, Y. Ishii, Nat. Struct. Mol. Biol. 2015, 6, 499.
[5] C. K. Bett, W. K. Serem, K. R. Fontenot, R. P. Hammer, J. C. Garno,
ACS Chem. Neurosci. 2010, 1, 661.
3.5
|
TEM
[6] C. Chunye, Oxidative Med. Cell. Longev. 2016, 1, 3571614.
[7] H. E. Lee, D. H. Kim, S. J. Park, J. M. Kim, Y. W. Lee, J. M. Jung,
C. H. Lee, J. G. Hong, X. Liu, M. Cai, K. J. Park, D. S. Jang, J. H. Ryua,
Pharmacol. Biochem. Behav. 2012, 103, 260.
PLGA fibrillation was observed under a LSM780/Elyra PS.1 (Zeiss) elec-
tron microscope in transmission mode. Objects were registered using
transmitted light and ZEN 2012 (Zeiss) and the ×40 objective with a ×10
ocular set-up. All images were supplemented with a scale bar = 20 μm.
[8] F. Karakida, Y. Ikeya, M. Tsunakawa, T. Yamaguchi, Y. Ikarashi,
S. Takeda, M. Aburada, Biol. Pharm. Bull. 2007, 30, 514.
[9] E. Boschetti, P. G. Righetti, J. Proteome 2008, 71, 255.
[10] T. K. Lind, P. Zielinska, H. P. Wacklin, Z. Urbanczyk-Lipkowska,
M. Cardenas, ACS Nano 2014, 8, 396.
[11] T. K. Lind, P. Polcyn, P. Zielinska, M. Cardenas, Z. Urbanczyk-
Lipkowska, Molecules 2015, 20, 738.
4
|
CONCLUSIONS
[12] B. Klajnert, M. Cortijo-Arellano, J. Cladera, M. Bryszewska, Biochem.
Biophys. Res. Commun. 2006, 345, 21.
Aggregation is a multistep process where monomers arrange into olig-
omers, protofibrils, and mature fibrils. Preventing the formation or dis-
assembly of either oligomers or fibrils is, at present, one of the main
efforts in drug discovery. Involvement of natural compounds in the
regulation of this process has been documented. Therefore, the
design and detailed study of the aggregation process in the presence
of new molecular entities is one of the possible avenues to achieve
better therapeutics.
[13] I. M. Neelov, A. Janaszewska, B. Klajnert, M. Bryszewska,
N. Z. Makova, D. Hicks, H. A. Pearson, G. P. Vlasov, M. Y. Ilyash,
D. S. Vasilev, N. M. Dubrovskaya, N. L. Tumanova, I. A. Zhuravin, A. J.
N. N. Nalivaeva, Curr. Med. Chem. 2013, 20, 134.
[14] M. A. Dobrovolskaia, A. K. Patri, J. Simak, J. B. Hall, J. Semberova,
S. H. D. P. Lacerda, S. E. McNeil, Mol. Pharm. 2012, 9, 382.
[15] E. L. Guzman, J. Amalraj, E. Fuentes, M. Alarcon, I. Palomo, I. Palomo,
L. S. Santos, Eur. J. Med. Chem. 2013, 69, 601.
[16] H. Lee, J. H. Jeong, T. G. Park, J. Control. Release 2002, 79, 283.
[17] Y. Zhao, M. J. Haney, N. L. Klyachko, S. Li, S. L. Booth,
S. M. Higginbotham, J. Jones, M. C. Zimmerman, R. L. Mosley,
A. V. Kabanov, H. E. Gendelman, E. V. Batrakova, Nanomedicine 2010,
6, 25.
The CD spectra reveal that the small branched peptides carrying
several valine or sinapic acid residues significantly altered the second-
ary structures of PLL and PLGA model polypeptides in a structure-
and concentration-dependent fashion. The amphiphilic character, high
supramolecular capacity, and probable self-aggregation into larger
nanoparticle structures are structural factors that enabled multiple
contacts with polypeptides. In several cases, the CD spectra of the
polypeptide/dendron mixtures were considerably different from those
corresponding to the individual polypeptides in terms of the signifi-
cant reduction of intensity, discrete structure, and dislocation of char-
acteristic bands. CD spectra deconvolution indicated that such
spectral features resulted from the presence of a complex mixture,
involving not only those typical for well-ordered proteins but also
varying in the amount of distorted α-helix, and right- and left-hand
twisted β-sheet, and sometimes the alternating presence of β-turns
and β-sheets. The electrostatic attraction and multiple hydrogen
bonding between oppositely charged molecules, that is, cationic
branched peptides and the β-sheet surface formed by aggregating
PLGA, were probably the main cause of co-aggregation that increased
the variety and contribution of less ordered forms.
[18] R. H. Walters, R. M. Murphy, J. Mol. Biol. 2009, 393, 978.
[19] G. Bocchinfuso, P. Conflitti, S. Raniolo, M. Caruso, C. Mazzuca,
E. Gatto, E. Placidi, F. Formaggio, C. Toniolo, M. Venanzi, A. Palleschi,
J. Pept. Sci. 2014, 20, 494.
[20] A. Hernik, W. Puławski, B. Fedorczyk, D. Tymecka, A. Misicka,
S. Filipek, W. Dzwolak, Langmuir 2015, 31, 10500.
ꢀ
[21] A. Hernik-Magon, W. Puławski, B. Fedorczyk, D. Tymecka,
A. Misicka, P. Szymczak, W. Dzwolak, Biomacromolecules 2016, 17(4),
1376.
[22] A. Micsonai, F. Wien, É. Bulyáki, Y.-H. Lee, Y. Goto, M. Réfrégiers,
J. Kardos, Nucleic Acids Res. 2018, 46, W315.
[23] J. J. Grigsby, H. W. Blanch, J. M. Prausnitz, Biophys. Chem. 2002, 99, 107.
[24] M. Chittchang, H. H. Alur, A. K. Mitra, T. P. Johnston, J. Pharm.
Pharmacol. 2002, 54, 315.
[25] N. Yamamoto, O. Kambara, K. Yamamoto, A. Tamura, S. Saito,
K. Tominaga, Soft Matter 2012, 8, 1997.
[26] S. Song, S. A. Asher, J. Am. Chem. Soc. 1989, 1989(111), 4295.
[27] R. A. Hule, D. J. Pochan, J. Polym. Sci.
45, 252.
B Polym. Phys. 2007,
[28] L. Mendonça, F. Hache, Int. J. Mol. Sci. 2012, 13, 2239.
[29] J. J. Balbach, A. T. Petkova, N. A. Oyler, O. N. Antzutkin,
D. J. Gordon, S. C. Meredith, R. Tycko, Biophys. J. 2002, 83,
1205.
CONFLICT OF INTEREST
[30] S. Chimon, M. A. Shaibat, C. R. Jones, D. C. Calero, Y. Ishii, PNAS
1998, 95, 13407.
The authors declare no conflict of interest.
[31] M. C. Owen, D. Gnutt, M. M. Gao, S. K. T. S. Warmlander, J. Jarvet,
A. Graslund, R. Winter, S. Ebbinghaus, S. B. Strodel, Chem. Soc. Rev.
2019, 48, 3946.
ORCID
Zofia Urbanczyk-Lipkowska
https://orcid.org/0000-0003-2043-
[32] Nat. Protoc. 2006, 1, 2876.
487X

14 of 14
MORAWIAK ET AL.
[33] H. Yao, E. Wynendaele, X. Xu, A. Kosgei, B. De Spiegeleer, J. Pharm.
Biomed. Anal. 2018, 147, 50.
SUPPORTING INFORMATION
Additional supporting information may be found online in the
Supporting Information section at the end of this article.
[34] A. Perczel, M. Hollosi, G. Tusnady, G. D. Fasman, Protein Eng. 1991,
4, 669.
[35] A. Micsonai, F. Wien, L. Kernya, Y.-H. Lee, Y. Goto, M. Réfrégiers,
J. Kardos, PNAS 2015, 112, E3095.
ꢀ
How to cite this article: Morawiak M, Stolarska M, Cieslak M,
[36] M. Michaelis, N. Hildebrand, R. H. Meißner, N. Wurzler, Z. Li,
J. D. Hirst, A. Micsonai, J. Kardos, M. D. Piane, L. C. Ciacchi, J. Phys.
Chem. B 2019, 123, 6694.
Urbanczyk-Lipkowska Z. Interactions of rationally designed
small peptide dendrons functionalized with valine or sinapic
acid with α-helix and β-sheet structures of poly-L-lysine and
poly-L-glutamic acid. Pept Sci. 2020;e24155. https://doi.org/
10.1002/pep2.24155
[37] B. Klajnert, J. Cladera, M. Bryszewska, Biomacromolecules 2006, 7,
2186.
[38] D. I. Andersson, D. Hughes, J. Z. Kubicek-Sutherland, Drug Resist.
Updat. 2016, 26, 43.