Functional Assemblies Emerging in Complex Mixtures of Peptides and Nucleic Acid–Peptide Chimeras

Article abstract of DOI:10.1002/chem.201800500

Striking synergy between nucleic acids and proteins is exhibited in living cells. Whether such mutual activity can be performed using simple supramolecular nucleic acid–peptide (NA-pep) architectures remains a mystery. To shed light on this question, we studied the emergence of a primitive synergy in assemblies of short DNA-peptide chimeras. Specifically, we characterized multiple structures forming along gradual mixing trajectory, in which a peptide solution was seeded with increasing amounts of NA-pep chimeras. We report on the systematic change from β-sheet-peptide-based fibrillar architectures into the spherical structures formed by the conjugates. Remarkably, we find that through forming onion-like structures, the conjugates exhibit increased DNA hybridization stability and bind small molecules more efficiently than the peptides or DNA alone. A brief discussion highlights the implications of our findings for the production of new materials and for research on the origin of life.

Full text of DOI:10.1002/chem.201800500

A Journal of
Accepted Article
Title: Functional Assemblies Emerging in Complex Mixtures of
Peptides and Nucleic Acid-Peptide Chimeras
Authors: Gonen Ashkenasy, Agata Chotera, Hava Sadihov, Rivka
Cohen-Luria, and Pierre-Alain Monnard
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201800500
Link to VoR: http://dx.doi.org/10.1002/chem.201800500
Supported by

10.1002/chem.201800500
Chemistry - A European Journal
FULL PAPER
Functional Assemblies Emerging in Complex Mixtures of
Peptides and Nucleic Acid–Peptide Chimeras
Agata Chotera,[a] Hava Sadihov,[a] Rivka Cohen-Luria,[a] Pierre-Alain Monnard,[b] and Gonen
Ashkenasy*[a]
Abstract: Striking synergy between nucleic acids and proteins is
exhibited in living cells. Whether such mutual activity can be
performed using simple supramolecular nucleic-acid-peptide (NA-
pep) architectures remains a mystery. To shed light on this question,
we studied the emergence of a primitive synergy in assemblies of
short DNA–peptide chimeras. Specifically, we characterized multiple
structures forming along gradual mixing trajectory, in which a peptide
solution was seeded with increasing amounts of NA-pep chimeras.
We report on the systematic change from β-sheet-peptide-based
fibrillar architectures into the spherical structures formed by the
conjugates. Remarkably, we find that through forming onion-like
structures, the conjugates exhibit increased DNA hybridization
stability and bind small molecules more efficiently than the peptides
or DNA alone. A brief discussion highlights the implications of our
findings for the production of new materials and for research on the
origin of life.
To address the above hypotheses, we prepared two novel DNA-
peptide conjugates (DNA1-pep and DNA2-pep, Fig. 1) and
studied their assembly along a putative mixing trajectory, in which
a solution containing peptide molecules was seeded with
increasing amounts of one of the conjugates (ssCon) or with pre-
hybridized two conjugates (dsCon). In this paper, we report a
remarkable step-by-step morphological transition, leading from β-
sheet peptide-dominated fibrillar architectures into dsCon-based
spherical, densely packed structures formed by concentric
bilayers. A set of experiments allows us to propose an interaction
model for the production of such spheres. Finally, we reveal that
through the formation of spherical particles, dsCon exhibits
increased stability towards elevated temperatures and pH values
and can efficiently and reversibly bind to small molecules, such
as the dye thioflavin T (ThT) and the drug doxorubicin (Dox).
Results and Discussion
Introduction
DNA1-pep and DNA2-pep conjugates contain segments of ten
nucleotides attached through a short triazole-based linker to an
amphiphilic peptide of nine amino acids (Fig. 1). The DNA
domains of the two conjugates are complementary to each other.
Each molecule contains a non-palindromic NA sequence and a
low ratio of guanine (G) bases in order to limit the formation of
undesired structures. In addition, these two sequences are
presumably too short to form a very stable duplex without auxiliary
interactions, such as those facilitated by conjugation to the
peptide. The peptide domain sequence derived from related
molecules previously studied by our group, containing repetitive
Glu-Phe dyads. As observed for other short amphiphilic
peptides,[4] the Glu-Phe peptides can readily assemble into -
sheet fibrils[5] and function as catalysts,[6] replicators,[7] and even
as charge transport entities.[8] The oligonucleotide and peptide
segments were prepared separately by solid phase synthesis,
and each segment was functionalized to be linked together by a
copper(II)-catalyzed ‘click’ reaction (Fig. 1).[9] After conjugation,
DNA1-pep and DNA2-pep were purified, and their purity and
molecular weights were characterized by HPLC and MALDI-TOF
MS (Figs. 1b and s1).
The enormously diverse biochemistry in living cells is made
possible by molecules from a small number of families, primarily
proteins, nucleic acids, fatty acids and sugars. While simple
operations, such as catalysis of
a
specific chemical
transformation, involve a single (enzyme) molecule or a small
complex of identical molecules, more elaborate tasks are often
driven by synergistic activity in assemblies of interacting
molecules from different families. Two major examples of such
synergy between nucleic acids and proteins are the specific
protein synthesis by the ribosome that translates RNA
sequences,[1] and RNA (or DNA) assembly and packaging by
proteins in virus particles, which control the entire viral replication
cycle.[2] The emergence of a primitive synergy in assemblies of
much simpler nucleic acid–peptide (NA-pep) chimeras is the topic
of the current paper. We posited that covalent interactions
between short nucleic acid and peptide segments and
spontaneous assembly of the chimeras will yield new structures
based on both the peptide association and sequence-specific
hybridization of the oligonucleotides.[3] We furthermore
hypothesized that these assemblies would be superior, in terms
of structure stability and function, to their peptide-only or nucleic-
acid-only counterparts.
Assemblies formed by the peptide molecules alone (N3-pep)
were characterized first with the aim to detect structure(s) likely to
form later on in mixtures with the DNA-pep conjugates. TEM and
AFM images revealed straight fibril-like structures that were
closely associated and bundled one next to the other (Fig. 2). The
width of these fibrils, ca. 4 nm, correlates well with a bilayer
packing where Phe side chains are hidden within the hydrophobic
fibril core, as we have previously observed for related peptides.[5a,
[a]
[b]
A. Chotera, Dr. R. Cohen-Luria and Prof. G. Ashkenasy*
Department of Chemistry, Ben-Gurion University of the Negev,
Beer-Sheva 84105, Israel. E-mail: gonenash@bgu.ac.il
Prof. P-A. Monnard
Institute for Physics, Chemistry and Pharmacy
University of Southern Denmark, 5230 Odense M, Denmark
8b]
Further characterization by CD (Fig. s2) and FTIR (Fig. s3)
revealed the typical characteristics of -sheet structures,
Additional table and figures, supporting information for this article,
are given via a link at the end of the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800500
Chemistry - A European Journal
FULL PAPER
including the amide bond IR absorbance peaks at 1623 cm-1 and
1694 cm-1, indicating the formation of antiparallel sheets.
(curved) fibrils with larger distances from one another (Fig. 3b),
probably since solvent-exposed DNA segments limited their
association. While under these conditions individual fibrils were
found (marked as 1 in Fig. 3b), the fibril length varied significantly
(150 nm up to a few microns), and inter-fibril association also led
to bundled morphologies of different width (Fig. 3b; 2 and 3).
Consistently, DNA1-pep or DNA2-pep molecules alone also
formed bundled architectures (AFM; Fig. s4a and s4c), or rather
shorter fibrils at higher concentrations (cryo-TEM; Fig. s4b).
Structure formation in mixtures containing N3-pep and
different amounts of DNA1-pep and DNA2-pep conjugates (1:1
ratio) was directed by first dissolving all three components in
buffer (pH 7; 10 mM Mg2+), incubating the mixture at ≥ 70 ºC (10
min), and then cooling it to RT for a further incubation of at least
3 h. Spatially isolated, apparently linear and stiff, fibril assemblies
were formed in mixtures containing N3-pep and low molar
fractions of dsCon (Fig. 3c,d). Remarkably different structures
were found in mixtures of N3-pep and higher molar fractions of the
DNA1-pep and DNA2-pep conjugates (1:1:1), namely, co-
assembly architectures composed of
a fibril-like skeleton
decorated with spherical objects (Figs. 3e and s5). A similar
‘beads-on-a-string’ configuration was previously obtained for the
co-assembly of two peptides having different hydrophobicity.[10]
Close inspection of the newly formed architectures showed that
all the fiber elements were of similar width (~4 nm) and that they
were sometimes bundled to wider aggregates. Current data
resolution allows us to propose that fiber and sphere elements are
bound through inter-assembly DNA hybridization, and/or
overlapping hydrophobic interactions between -sheet peptide
segments. We further note that isolated spheres can also form
under these conditions, supposedly made of only (or largely) the
dsCon duplexes, as discussed in detail in the following.
Figure 1. DNA-pep chimeras. a) DNA-pep sequences, highlighting the
complementary nature of the DNA segments and the -strand formed by the
peptide segments. Single-letter nomenclature is used to represent the
sequences. A = -alanine. b) HPLC and MALDI-TOF MS characterization of
DNA1-pep during 2h of a ‘click’ reaction between the N3-pep and 5’-hexynyl
precursors. Fig. s1 shows a similar characterization of the DNA2-pep conjugate.
To examine the final stage of this gradual mixing course, we
characterized the assembly products obtained after annealing
and equilibration of the two complementary conjugates DNA1-
pep and DNA2-pep (25 M each, [N3-pep] = 0; Fig. 4). Cryo-TEM
measurements (Figs. 4a and 4c), and dry TEM measurements
(Figs. 4d, 4e and s8) revealed the formation of rounded crowded
spheres with diameters ranging 20–180 nm. Likewise, DLS
characterization of the 50 M dsCon revealed the formation of
particles 20–80 nm in radius (mean radius = 37 nm, Fig. s6), and
DLS of a much more concentrated sample (500 M) showed only
slightly larger aggregates (mean radius = 51 nm, Fig. s6).
Analysis of the AFM data (Figs. 4b and s7) indicated rather flat
aggregates, shorter in the z direction (19 nm in average) than in
their lateral diameter (130 nm), suggesting that the spheres had
shrunk upon drying on the mica surfaces and/or that the images
had been subject to the AFM tip convolution effect. Statistical
analysis of an ensemble of the cryo-TEM images revealed that
dsCon formed spherical architectures with distinct diameter
values (Figs. 4f, s9 and s10). We therefore grouped together
structures of similar size, representing spheres with distinct
numbers of DNA-pep layers, when taking into account that the
assembly of each layer expands its diameter by ~6 nm.
Figure 2. N3-pep fibril assembly structures formed in aqueous solutions at pH 7
and characterized by TEM (a, 500 M) and AFM (b, 50 M).
Structure formation in mixtures (50 M total concentration each)
containing either N3-pep and a single type of conjugate ssCon or
N3-pep and the hybridized complementary conjugate dsCon was
then followed extensively by AFM and cryo-TEM (Fig. 3). We note
that N3-pep and both the DNA and peptide segments of DNA1-
pep and DNA2-pep are all negatively charged at neutral pH,
implying that their mutual assembly is driven primarily by peptide-
peptide interactions and inter-molecular, H-bonding directed,
hybridization of nucleic acid segments. AFM images obtained for
the assemblies formed in mixtures of N3-pep and low
concentrations of DNA1-pep showed narrow and bundled fibril
assemblies (Fig. 3a), similar to the fibrils formed by N3-pep alone.
Images obtained for mixtures with a higher ssCon:pep ratio (1:1)
revealed the formation of ca. 50% wider and more flexible
This article is protected by copyright. All rights reserved.

10.1002/chem.201800500
Chemistry - A European Journal
FULL PAPER
and dsCon (c-e). Traces below the AFM images show the respective height
cross-sections. Numbers 1-3 in panel b highlight the formation of individual
fibrils and fibril bundles of variable width. The structures were formed in mixtures
of 50 M total concentration; the ratio between peptide and conjugate(s) is given
in each panel (e.g., in panel d, ‘40:10’ represents 40 M N3-pep mixed together
with 5 M DNA1-pep and 5 M DNA2-pep). Additional images are given in Figs.
s4 and s5.
Remarkably, we indeed found narrow diameter distributions,
centering close to the predicted value for each of the sphere
groups (Fig. 4f).
Further insight into the dsCon molecular arrangement within the
spheres was obtained from the higher resolution TEM data in
Figures 4e and s8, allowing to detect ‘rings’ for the different
assembly layers. CD measurements (Fig. s11), in addition
showed that the DNA segments in dsCon primarily took the more
compact B-form,[11] unlike dsDNA of the same sequences, which
in control studies formed mostly A-type helix structures, as
typically observed for short G-C-rich sequences[11b]. Taken
together, our results indicate the formation of multi-lamellar
spherical architectures (Fig. 4f - inset), as was previously
observed for related assemblies formed by Janus soft-matter
building blocks,[12] such as lipid-DNA molecules.[13]
Figure 4. Characterization of the spherical assemblies formed by pure dsCon.
a) Cryo-TEM images. b) AFM images. c) Enlarged image of an individual sphere
observed in the Cryo-TEM measurements. d,e) Enlarged images of individual
spheres observed in dry TEM measurements; data collected after staining with
uranyl acetate. Additional images of dsCon spheres are given in Figures s8 and
s12. f) Statistical analysis using differential count of sphere diameters (bin = 0.5
nm; count numbers ≥ 2). Values in blue refer to the number of layers in the
corresponding group of spherical assemblies, calculated for the assumed round
figure of 6 nm per layer (see Scheme at the top right of this panel).
Figure 3. AFM and cryo-TEM images of self-assembled nanostructures formed
in mixtures containing either N3-pep and ssCon [DNA1-pep; (a, b)] or N3-pep
This article is protected by copyright. All rights reserved.

10.1002/chem.201800500
Chemistry - A European Journal
FULL PAPER
Several recently published works have described the self-
assembly of NA-pep conjugates into vesicle or fibril-like
architectures. In such studies, scientist have characterized
assemblies formed by NA moieties attached to either short
peptides[14] or to positively charged peptides that strongly interact
intramolecularly with the NA segments.[15] Co-assembly via non-
covalent association of positively charged peptides (or proteins)
and NA polymers have also been disclosed recently.[16] The
chimeras introduced here produce a different type of assembly, in
which minimal interaction takes place between DNA and the
peptide domains of the same molecules, implying that new
architectures are formed on the basis of synergy, due to a
collective set of interactions. Most remarkably, to the best of our
knowledge, this is the first systematic study showing the gradual
change from peptide fibrils to conjugate multi-lamellar spheres,
while also isolating the intermediate architectures formed along
the way (Scheme 1). In order to directly probe how such gradual
changes in structures may come about during simple mixing
processes, we have further followed structure formation in N3-pep
solutions that were gradually seeded with increasing amounts of
dsCon. AFM images (Fig. s13) indeed show the same trend as
observed for the assembly of the corresponding individual
mixtures (Fig. 3). Control experiments, carried out by simply
mixing solutions of dsDNA and N3-pep, did not result the same
structural changes, but rather the formation of simple fibril or tape-
like architectures (Fig. s14).
for mixtures containing dsCon that is about 6 °C higher than that
of the control unmodified dsDNA. Subsequently, stability studies
were also performed to probe the chemical (alkaline-pH-
dependent) denaturation of dsCon. At an alkaline pH, the activity
of OH- groups causes denaturation of DNA duplexes by removing
the protons involved in H-bond formation from the guanine and
thymine bases.[18] Duplex denaturation as a function of pH was
followed, and the denaturation fractions calculated as previously
reported (see SI).[19]
A markedly lower degree of duplex
denaturation was observed for dsCon than for dsDNA above pH
8 (Fig. 5b). AFM investigation confirmed that dsCon spheres can
be readily formed at pH 8 and that they are disrupted at pH ≥10
(Fig. s16).
NA-pep conjugates[20] and peptide nucleic acid (PNA)[21]
oligomers have mostly been utilized as soluble entities with
applications in the medical sciences, e.g., for drug delivery or
gene silencing. Little is known about the binding properties of NA-
pep conjugate assemblies, which could serve as (drug) carriers
or as catalysts for reactions of encapsulated substrates.
Therefore, we tested the dsCon spheres as binders for known
peptide fibril and dsDNA intercalators. Binding of ThT to -sheet-
rich structures or to DNA[22] induces a characteristic red-shift in its
fluorescence spectrum, enhancing the emission at ca. 480 nm. A
binding assay, performed by mixing 100 M ThT with 10 M
dsCon, revealed significant staining of the dsCon spheres (Fig.
s17). Remarkably, the increase in ThT fluorescence upon binding
to dsCon was higher than that observed in control experiments,
in which we followed ThT binding to dsDNA, N3-pep, or even to
the mixture of both dsDNA and N3-pep (Fig. s17). These findings
indicate that the dsCon sphere interior might provide additional
binding sites to encapsulate the ThT.
Scheme 1. Gradual structural changes driven by mixing amphiphilic peptides
with increasing amounts of DNA-pep chimeras.
We propose that each of the new architectures could find utility in
the future; for example, the stiff linear fibrils formed in the
dsCon:pep mixtures might be useful for molecular electronics,[17]
while the dsCon spheres could serve as efficient hosts for binding
and releasing small molecules (vide infra). Aggregation of dsCon
into spherical structures with concentric bilayers induces the
encapsulation of the hybridized DNA segments and can thus
confer additional duplex stability, namely, minimize denaturation
of dsDNA unless harsh conditions are applied. This hypothesis
was tested in thermal denaturation experiments performed by
following the changes in UV absorption at 260 nm upon heating
and transition of the DNA duplex into single strands (Fig. 5a and
Table s1). The results show a duplex melting temperature (Tm)
Figure 5. Stability and binding properties of the dsCon spherical structures.
a, b) Temperature-dependent and pH-dependent denaturation. In each panel,
the dsCon denaturing stability (solid line) is compared to that of dsDNA (dashed
line). c) Left – Fluorescence measurements (590 nm) reflecting the quenching
and restoration of emission, upon binding of Dox to dsCon and its release,
respectively. Right – Dox fluorescence upon binding to dsCon, dsDNA and N3-
pep assemblies (Fig. s18). 100% fluorescence was calibrated for the free Dox
emission. Inset – AFM images of dsCon spheres after DOX intercalation.
The binding of dsCon spheres to the Doxorubicin (Dox) molecule,
a DNA intercalator often used for cancer chemotherapy, was then
This article is protected by copyright. All rights reserved.

10.1002/chem.201800500
Chemistry - A European Journal
FULL PAPER
modified oligonucleotides were purchased from Bio-Synthesis, Inc.
Doxorubicin was generously provided by Prof. Ayelet David, School of
Pharmacy, Ben-Gurion University of the Negev.
analyzed by following the quenching of Dox auto-fluorescence
upon mixing with dsCon. Studies at a 1:1 host:drug molar ratio
(10 M each) showed almost complete fluorescence quenching,
attributed to Dox binding to and within dsCon structures (Fig. 5c).
Evidence for the reversible nature of this binding was obtained by
adding the denaturing agent guanidinium chloride to a mixture of
Dox and dsCon (1:1); the denaturing agent caused immediate
Dox release, as observed by the restoration of maximal
fluorescence (Fig. 5c - left). We therefore suggest that this system
might be useful for future controlled drug-delivery applications.
Structural characterization by AFM of Dox-binding dsCon
revealed that the spheres expanded substantially to
accommodate the bound aromatic entities (Fig. 5c – inset). In
addition, our results showed that mixing 10 M dsCon with a large
(10-fold) excess of Dox did not induce a significant change in Dox
auto-fluorescence, probably because the dsCon structures were
already saturated with Dox. This notion is further supported by the
mixture AFM images, showing fibril like structures, as also
observed for Dox molecules alone (Fig. s19). Control binding
experiments showed a small change due to Dox binding to dsDNA
and no quenching upon mixing of Dox with the peptide fibrils (Fig.
5c – right). Here, we again attribute the stronger Dox binding to
dsCon to additional binding sites provided by the DsCon self-
assembly, as does the slower Dox release back into the solution.
Main Text Paragraph.
Azide-modified peptide synthesis. Triflyl azide, used for modification of
amine at the N terminus or side chain was synthesized as follows: a
suspension of sodium azide (3.3 mmol) in 4 ml of pyridine was cooled in
an ice bath. Then triflic anhydride (2.78 mmol) was added to the mixture
by a syringe during 5 min of stirring. After the reaction had been maintained
for 2 h in the ice bath, the TfN3-containing solution (after filtration of the
salts) was added to the amine solution for subsequent diazo transfer
reaction. In parallel, the free amine peptide sequence was synthesized at
0.1-0.2 mmol scale on a Rink amide MBHA resin using standard Fmoc
protocols. Thereafter, 100 mg of peptide-attached resin was allowed to
swell in DMF for 20 min, followed by the addition of 0.02 mmol of CuSO4
in MeOH (1 ml) and triethylamine (0.02 mol). The mixture was cooled in
the ice bath for a few minutes, and then the solution of triflyl azide in
pyridine was added dropwise. The reaction mixture was allowed to warm
to room temperature and reacted for 5 h. The same procedure was
repeated 3 times with fresh reagents. Upon completion, the reaction
mixture was washed with DMF, DIPEA/DMF (99.5:0.5, 3×2 min), and
finally with DMF, DCM and ether. The azido-peptide was obtained after
cleavage and global deprotection with a mixture of TFA:TIS:water
(95:2.5:2.5). The resin was removed by filtration under reduced pressure
and the TFA was evaporated out of the peptide mixture. Cold ether, 8-10
times in volume, was added to precipitate the crude peptide. The peptide
was purified by preparative HPLC using a Dionex Ultimate 3000 instrument
with C18 reverse phase preparative column at a flow rate of 20 ml/min.
The identity and purity of the peptide were analyzed on a Dionex 1100
using a reverse phase C18 column at a flow rate of 1.5 ml/min. ESI-MS
mass spectrometry (LCMS Thermo Surveyor 355) analysis was performed
to verify the molecular weight (measured Mw = 1143).
Synthesis of DNA-peptide conjugates via ‘click’ chemistry. The azido-
peptide (1 eq.) and 5’ hexynyl DNA (1 eq.) were dissolved in water/t-BuOH
(1:1). CuSO4 (1% mole eq.) and TBTA (5% mole eq.) were added to the
solution, followed by sodium ascorbate (0.5 eq.). The mixture was shaken
at 37ºC. The reaction progress was monitored by analytical HPLC on a
Dionex 1100 using a reverse phase C18 column; upon completion, the
product was purified by HPLC, and analyzed by HPLC and MALDI-TOF
mass spectrometry (Figs. 1 and s1).
Conclusions
We have revealed
a variety of supramolecular structures
emerging in mixtures of amphiphilic peptides and DNA-pep
chimeras. The assemblies formed by the double-stranded
chimera confer higher stability on the DNA duplex and bind to
small molecule substrates better than their NA or peptide moieties
alone. We accordingly suggest that these molecules, or similar
conjugates, could serve as new materials for various applications,
such as catalysis, electron transfer and drug delivery. Further-
Sample preparation for biophysical studies. Aqueous solutions of the
chosen sample were prepared fresh, by dissolving lyophilized powder in
phosphate buffer (10 mM), pH ~ 7. Peptide containing samples were
dissolved by sonication for 10 min. After sonication, the samples were
vortexed to eliminate any undesired structures originating from the
sonication, and then equilibrated for at least 2 hours. Samples containing
dsDNA and dsCon were incubated at elevated temperatures (>70 ºC) for
10 min, cooled down to RT and then equilibrated for at least 3 more hours.
The concentrations of peptide stock solutions were determined by
measuring the UV absorption of a diluted peptide stock solution at 257 nm
(the maximum absorption of phenylalanine). Concentrations of DNA and
DNA-pep samples were determined by measuring the UV absorption of a
diluted stock solution at 260 nm (the maximum absorption of DNA), using
ε = 99000 L/(mole*cm) for DNA1 and ε = 85400 L/(mole*cm) for DNA2.
more, we propose that our research might offer
a new
understanding of early chemical evolution, towards the origin of
life. Since it has recently been shown that both peptide and
nucleic acid precursors can be formed in a prebiotic environment
from simple building blocks of unified origin,[23] we suggest that
synergistic activity such as that demonstrated here may have led
to the selection of functional chimeras. Together with other recent
results demonstrating structural and functional synergy in NA–
pep systems[3c, 24] and also synergy between simple prebiotically
relevant molecules from different families, such as NA–lipid,[13, 25]
pep–lipid,[26] and sugar–pep,[27] the current study shows the rapid
progression in complexity and systems chemistry[28] that would
have facilitated a large repertoire of functions.
Circular dichroism measurements. CD spectra were performed on a
Jasco J-815 CD-spectrometer. All samples were prepared by dissolving
powder of peptide, DNA or conjugates in the phosphate buffer, and re-
checking the concentration by UV-absorption. CD spectrum between 200
nm to 320 nm was recorded with data steps of 0.5 nm, a scan speed of 10
nm min-1, response time 4 s and a bandwidth of 2 nm. The baseline buffer
spectrum was measured and subtracted from the sample spectra. All
samples were measured in triplicate, and the results were averaged
automatically.
Atomic force microscopy (AFM) imaging. Samples were prepared with
an addition of 10 mM MgCl2, for better binding of negatively charged
compounds to the mica surfaces. A 3 μl aliquot of solution after
equilibration as described above was deposited on a freshly cleaved mica
surface. The sample was then slowly dried in air overnight. Topography
images were acquired by AFM (SolverPro, NT-MDT, Ru) in tapping mode
using noncontact tips BudgetSensors Multi75Al-G, 3 N/m, 75 kHz. Image
processing, which included second order polynomial line fitting and cross-
section analysis, was done using the NOVA AFM software. AFM
Experimental Section
General. Chemical and reagents were purchased from Aldrich or Merck
and unless otherwise specified used without further purification. Amino
acids, resins and coupling reagents were purchased from Novabiochem
and Alfa Aesar, and DMF was purchased from Biotech Grade. 5’ hexynyl
This article is protected by copyright. All rights reserved.

10.1002/chem.201800500
Chemistry - A European Journal
FULL PAPER
histograms of height distribution (Fig. s7) were prepared by counting 25
different AFM features.
containing dsDNA and dsCon were incubated at elevated temperatures (≥
70ºC) for 10 min, then 10 μM Doxorubicin was added, and spectra were
recorded on a 3001 Thermo VarioSkan fluorescence spectrometer,
equipped with a 96-microwellplate reader. Spectra of doxorubicin were
collected from 505 to 700 nm, using an excitation wavelength of 480 nm
(excitation slit width: 12 nm), and scan speed of 100 nm min-1. The intensity
of fluorescence signals at 590 nm was used to compare Dox binding to the
different assemblies; the values are shown as normalized emission
percentage, using the signal of free Dox in buffer as 100% emission.
Cryogenic transmission electron microscopy (Cryo-TEM) imaging.
Samples for direct imaging of the aqueous dispersions were prepared in
the controlled environment box of a vitrification robot (Vitrobot, FEI), as
follows: a 5 μl drop of the sample solution was deposited on a glow-
discharged TEM grid, the excess liquid was automatically blotted with a
filter paper, and the specimen was rapidly plunged into liquid ethane and
transferred to liquid nitrogen, where it was kept until use. The samples
were examined below −175 °C using an FEI Tecnai 12 G2 TWIN TEM that
operated at 120 kV in low-dose mode and with a few micrometers under
focus to increase the phase contrast. The images were recorded with a
Gatan charge-coupled device camera (model 794). The images of the
assemblies were analyzed by Digital Micrograph software (Version 3.1).
Dictating the double stranded DNA species melting temperatures.
The samples used for the thermal melting analysis contained a total
concentration of 5 μM, with DNA1-pep and DNA2-pep in 1:1 ratio in a total
volume of 200 µl. The samples were prepared in 10 mM phosphate buffer
pH = 7 containing 50 mM Na2SO4. dsDNA solutions of the same
concentrations were used as a control. Tm was found using the change in
absorption at 260 nm upon transition from duplex to single strands. The
scans were performed on an Agilent Cary 60 UV-Vis spectrometer with
Temperature Controller TC125 using a cuvette of 1 cm pass length. For
the melting and annealing measurements, first the sample was heated fast
to 80 ºC and incubated for 10 min, before cooling down to 20 ºC at a rate
of 0.5 ºC min-1. The sample was kept at 20ºC for 10 min, before heating
back up to 80 ºC at the same rate. The absorbance values were
normalized using the given equation:
Statistical analysis based on the Cryo-TEM data. Hundred individual
spherical objects were measured, and the diameter of each one was
determined as the average of 3 individual lines that pass through the
sphere center (Fig. s9). Differential count analysis of all the measured
diameters was obtained using the Excel statistical COUNTIF function, by
utilizing the following bins: 0.5, 1 and 2. The value range screened by the
COUNTIF function included all the 100 sphere diameters, and the criteria
applied as higher or equal given bin, but smaller than the next bin. For
example, for bin = 0.5 and diameter = 24, the function assigned the number
of measured objects with diameters equal to or greater than 24 nm, but
lower than 24.5 nm. For all three analyzed bin values, we have clearly
observed higher counts for diameters divisible by 6, correlating well with
the theoretical cross section of a single DNA-pep layer (Fig. 4 in the main
manuscript), which may suggest that the assembly of each layer adds ~6
nm to the sphere diameter.
Abs − 퐴푏푠
0
Normalized value =
퐴푏푠
푚푎푥
− 퐴푏푠
0
The Tm was found by determining the maximum of the first derivatives of
absorbance signals resulting from 3 separate melting and annealing
measurements.
Analyzing the pH-dependent denaturation of double stranded DNA
species. 1 M NaOH solution was used to denature the DNA in dsCon and
dsDNA assemblies. NaOH solution was slowly added to 200 μL dsCon or
dsDNA solutions, in order to obtain incremental one-unit jumps in the
solution pH. The mixture was homogenized with continuous pipetting and
vortex, after which the final absorbance (A260) was measured using the
UV/VIS spectrophotometer. The changes in 260 nm absorbance as a
function of the pH were applied to calculate the denaturation percentage,[1]
using a simplified equation. Variables measured and used in the equation
include: Initial A260, Final A260 and Blank A260.
Transmission electron microscopy (TEM) imaging. A TEM copper grid
with a 200-mesh carbon support (Electron Microscopy Sciences) was
covered with 10 μl of a peptide solution (1 mM) in 10 mM phosphate buffer
pH = 7, for 1 min before wicking the excess solution with filter paper. 10
μL of the staining solution, 2ꢀ% uranyl acetate, (Sigma-Aldrich), was added
and incubated for 2 min; excess solution was wicked away, and the grids
were placed in desiccators to dry under vacuum. A Hitachi H-7500
transmission electron microscope was used to image the samples at 75
kV.
Denaturation (%) = (퐹푖푛푎푙 퐴260−퐵푙푎푛푘 퐴260 − 1) 푥200
Initial A260
Attenuated total reflectance fourier transform infrared spectroscopy
(ATR-FTIR). Aliquot (10 μL) of peptide solution was dried as thin films on
an ATR diamond crystal. FT-IR spectra were acquired using a FTIR
Spectrometer 6700 at room temperature, and averaging 512 scans with 2
cm−1 resolution, using a sensitive liquid N2 cooled MCT-A detector with a
The results presented in the manuscript (Fig. 5) are the mean values
obtained for 3 repetitions of the experiment and the standard deviations.
5-mm aperture and a scanning speed of 4 mmꢀsec−1
.
Acknowledgements
Dynamic light scattering (DLS) measurements. Samples were
prepared at 50 µM and 500 µM concentrations in phosphate buffer (pH 7)
and transferred to a cylindrical glass ampoule (length ca. 8 cm; diameter
ca. 0.8 cm) of high quality optical glass. It was then placed in a vat filled
with toluene as the index matching fluid. During the course of the
measurements, the vat was kept at room temperature. The light source
was an argon ion laser and photons scattered by the sample were
collected by a photomultiplier tube mounted on the goniometer arm at 90º
to the direction of the incident radiation. Measurements were acquired
using a CGS-3 LSE-5004 correlator at room temperature.
This research is sponsored by an NSF-BSF grant (2015671). We
also acknowledge support from the COST actions CM1304 and
TD1308. We thank Prof. N. Ashkenasy (BGU) and Prof. D.G.
Lynn (Emory U.) for fruitful discussions.
Keywords: Nucleic acid – peptide conjugates • Systems
Chemistry • Nanostructures
Fluorescence measurement of thioflavin T (ThT) staining. Samples
used for the measurements contained a total concentration of 10 μM of
peptide, dsDNA or dsCon, in a volume of 70 µl of 10 mM phosphate buffer
pH = 7. Samples containing dsDNA and dsCon were first incubated at
elevated temperatures (≥ 70ºC) for 10 min, then 100 μM ThT was added,
and the spectra were measured. Fluorescence spectra were recorded on
a 3001 Thermo VarioSkan fluorescence spectrometer, equipped with a 96-
microwellplate reader. ThT Spectra were collected from 465 to 700 nm,
using an excitation wavelength of 440 nm (excitation slit width: 12 nm), at
scan speed of 100 nm min-1. The baseline of 100 μM thioflavin T in the
buffer was subtracted from the sample spectra. Fluorescence emission
signals at 480 nm were used to compare ThT binding to the different
assemblies.
[1]
[2]
A. Yonath, Angew. Chem., Int. Ed. 2010, 49, 4340-4354.
J. D. Perlmutter, M. F. Hagan, Annu. Rev. Phys. Chem. 2015, 66, 217-
239.
[3]
a) J. T. Goodwin, A. K. Mehta, D. G. Lynn, Acc. Chem. Res. 2012, 45,
2189-2199; b) A. Brack, Chem. Biodivers. 2007, 4, 665-679; c) A. N.
Lazar, A. W. Coleman, S. Terenzi, P. Strazewski, Chem. Commun. 2006,
63-65.
[4]
a) L. Adler-Abramovich, E. Gazit, Chem. Soc. Rev. 2014, 43, 6881-6893;
b) M. Reches, E. Gazit, Curr. Nanosci. 2006, 2, 105-111; c) S. Fleming,
R. V. Ulijn, Chem. Soc. Rev. 2014, 43, 8150-8177; d) M. J. Webber, E.
J. Berns, S. I. Stupp, Isr. J. Chem. 2013, 53, 530-554; e) S. Cavalli, F.
Albericio, A. Kros, Chem. Soc. Rev. 2010, 39, 241-263; f) Y. Gao, F.
Zhao, Q.-G. Wang, Y. Zhang, B. Xu, Chem. Soc. Rev. 2010, 39, 3425-
Fluorescence measurement of doxorubicin (Dox). Samples containing
a total concentration of 0-25 μM of peptide, dsDNA or dsCon in a volume
of 70 µl, were prepared in 10 mM phosphate buffer pH = 7. Samples
This article is protected by copyright. All rights reserved.

10.1002/chem.201800500
Chemistry - A European Journal
FULL PAPER
3433; g) C. A. E. Hauser, S. Zhang, Chem. Soc. Rev. 2010, 39, 2780-
2790; h) J. M. A. Carnall, C. A. Waudby, A. M. Belenguer, M. C. A. Stuart,
J. J. P. Peyralans, S. Otto, Science 2010, 327, 1502-1506; i) N. Singh,
M. Kumar, J. F. Miravet, R. V. Ulijn, B. Escuder, Chem. Eur. J. 2017,
981–993.
Meyer, C. Modlin, X. Zuo, V. P. Conticello, Y. Ke, J. Am. Chem. Soc.
2017, 139, 14025-14028.
[17] M. Mizrahi, A. Zakrassov, J. Lerner-Yardeni, N. Ashkenasy, Nanoscale
2012, 4, 518-524.
[18] S. Shin, L. A. Day, Anal. Biochem. 1995, 226, 202-206.
[19] a) X. F. Wang, A. Son, Environmental Science-Processes & Impacts
2013, 15, 2204-2212; b) X. Wang, H. J. Lim, A. Son, Environ Health
Toxicol 2014, 29, e2014007.
[5]
a) Y. Raz, B. Rubinov, M. Matmor, H. Rapaport, G. Ashkenasy, Y. Miller,
Chem. Commun. 2013, 49, 6561-6563; b) H. Rapaport, K. Kjaer, T. R.
Jensen, L. Leiserowitz, D. A. Tirrell, J. Am. Chem. Soc. 2000, 122,
12523-12529; c) H. Rapaport, Supramol. Chem. 2006, 18, 445-454.
M. Tena-Solsona, J. Nanda, S. Diaz-Oltra, A. Chotera, G. Ashkenasy, B.
Escuder, Chem. Eur. J. 2016, 22, 6687 - 6694.
[20] a) D. M. Copolovici, K. Langel, E. Eriste, U. Langel, ACS Nano 2014, 8,
1972-1994; b) J. J. Turner, G. D. Ivanova, B. Verbeure, D. Williams, A.
A. Arzumanov, S. Abes, B. Lebleu, M. J. Gait, Nucl. Acids Res. 2005, 33,
6837-6849; c) T. P. Wang, N. C. Ko, Y. C. Su, E. C. Wang, S. Severance,
C. C. Hwang, Y. T. Shih, M. H. Wu, Y. H. Chen, Bioconjug. Chem. 2012,
23, 2417-2433.
[6]
[7]
a) B. Rubinov, N. Wagner, H. Rapaport, G. Ashkenasy, Angew. Chem.
Int. Ed. 2009, 48, 6683-6686; b) B. Rubinov, N. Wagner, M. Matmor, O.
Regev, N. Ashkenasy, G. Ashkenasy, ACS nano 2012, 6, 7893-7901.
a) D. Ivnitski, M. Amit, B. Rubinov, R. Cohen-Luria, N. Ashkenasy, G.
Ashkenasy, Chem. Commun. 2014, 50, 6733-6736; b) D. Ivnitski, M.
Amit, O. Silberbush, Y. Atsmon-Raz, J. Nanda, R. Cohen-Luria, Y. Miller,
G. Ashkenasy, N. Ashkenasy, Angew. Chem. Int. Ed. 2016, 55, 9988-
9992.
[8]
[21] T. Shiraishi, P. E. Nielsen, Methods Mol. Biol. 2014, 1050, 193-205.
[22] A. Biancardi, T. Biver, A. Burgalassi, M. Mattonai, F. Secco, M. Venturini,
PCCP 2014, 16, 20061-20072.
[23] B. H. Patel, C. Percivalle, D. J. Ritson, C. D. Duffy, J. D. Sutherland, Nat.
Chem. 2015, 7, 301–307.
[9]
a) N. Venkatesan, B. H. Kim, Chem. Rev. 2006, 106, 3712-3761; b) K.
Lu, Q. P. Duan, L. Ma, D. X. Zhao, Bioconjug. Chem. 2010, 21, 187-202.
[24] a) R. Wieczorek, M. Doerr, A. Chotera, P. L. Luisi, P.-A. Monnard,
ChemBioChem 2013, 14, 217-223; b) J. E. Hein, E. Tse, D. G.
Blackmond, Nat. Chem. 2011, 3, 704-706; c) H. Griesser, M. Bechthold,
P. Tremmel, E. Kervio, C. Richert, Angew. Chem., Int. Ed. 2017, 56,
1224-1228; d) H. Griesser, P. Tremmel, E. Kervio, C. Pfeffer, U. E.
Steiner, C. Richert, Angew. Chem., Int. Ed. 2017, 56, 1219-1223; e) Y.
Ura, J. M. Beierle, L. J. Leman, L. E. Orgel, M. R. Ghadiri, Science 2009,
325, 73-77.
[10] S. Yuran, Y. Razvag, M. Reches, ACS Nano 2012, 6, 9559-9566.
[11] a) C. Lou, M. C. Martos-Maldonado, C. S. Madsen, R. P. Thomsen, S.
R. Midtgaard, N. J. Christensen, J. Kjems, P. W. Thulstrup, J. Wengel,
K. J. Jensen, Nat. Commun. 2016, 7, 12294-12302; b) J. Kypr, I.
Kejnovska´, D. Rencˇiuk, M. V. c. k. *, Nucl. Acids Res. 2009, 37, 1713–
1725.
[12] Y. Mai, A. Eisenberg, Chem. Soc. Rev. 2012, 41, 5969-5985.
[13] R. S. Shirazi, K. K. Ewert, B. F. B. Silva, C. Leal, Y. Li, C. R. Safinya,
Langmuir 2012, 28, 10495-10503.
[25] M. C. Wamberg, R. Wieczorek, S. B. Brier, J. W. d. Vries, M. Kwak, A.
Herrmann, P.-A. Monnard, Bioconjugate Chem. 2014, 25, 1678−1688.
[26] a) S. Murillo-Sanchez, D. Beaufils, J. M. Gonzalez Manas, R. Pascal, K.
Ruiz-Mirazo, Chem. Sci. 2016, 7, 3406-3413; b) E. C. Izgu, B. A., N. P.
Kamat, V. S. Lelyveld, W. Zhang, T. Z. Jia, J. W. Szostak, J. Am. Chem.
Soc. 2016, 138, 16669-16676; c) H. R. Marsden, I. Tomatsu, A. Kros,
Chem. Soc. Rev. 2011, 40, 1572-1585.
[14] a) X. Li, Y. Kuang, H. C. Lin, Y. Gao, J. Shi, B. Xu, Angew. Chem. Int.
Ed. 2011, 50, 9365-9369; b) N. Gour, J. N. Abraham, M. Chami, A.
Castillo, S. Verma, C. Vebert-Nardin, Chem. Commun. 2014, 50, 6863-
6865; c) P. Liu, R. Ni, A. K. Mehta, W. S. Childers, A. Lakdawala, S. V.
Pingali, P. Thiyagarajan, D. G. Lynn, J. Am. Chem. Soc. 2008, 130,
16867-16869; d) R. M. Turk-MacLeod, D. Puthenvedu, I. Majerfeld, M.
Yarus, J. Mol. Evol. 2012, 74, 217-225.
[27] A. J. Wagner, D. Y. Zubarev, A. Aspuru-Guzik, D. G. Blackmond, ACS
Cent. Sci. 2017, 3, 322-328.
[28] G. Ashkenasy, T. M. Hermans, S. Otto, A. F. Taylor, Chem. Soc. Rev.
2017, 46, 2543-2554.
[15] a) J. E. Noble, E. De Santis, J. Ravi, B. Lamarre, V. Castelletto, J. Mantell,
S. Ray, M. G. Ryadnov, J. Am. Chem. Soc. 2016, 138, 12202-12210; b)
M. Kye, Y.-b. Lim, Angew. Chem. Int. Ed. 2016, 55, 12003-12007; c) W.
M. Aumiller, Jr., C. D. Keating, Nat. Chem. 2016, 8, 129-137.
[16] a) R. Ni, Y. Chau, J. Am. Chem. Soc. 2014, 136, 17902-17905; b) R. Ni,
Y. Chau, Angew. Chem., Int. Ed. 2017, 56, 9356-9360; c) T. Jiang, T. A.
This article is protected by copyright. All rights reserved.

10.1002/chem.201800500
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents
FULL PAPER
Chemical evolution: systematic
changes from fibrillar architectures
into spherical nano-structures
Agata Chotera, Rivka Cohen-Luria,
Pierre-Alain Monnard, and Gonen
Ashkenasy*
observed upon mixing amphiphilic
peptides and DNA-peptide
Page No. – Page No.
conjugates. The spherical onion-like
structures exhibit increased stability
and efficiently bind small molecules.
Functional Assemblies Emerging
along the Chemical Evolution of
Nucleic Acid–Peptide Chimeras
Larger version of the TOC figure:
This article is protected by copyright. All rights reserved.