Tracking morphologies at the nanoscale: self-assembly of ... · Self-assembling amphiphilic designer peptides have been explored as building blocks for the development of controllable,

Nano Res

1

Tracking morphologies at the nanoscale: self-assembly

of an amphiphilic designer peptide into a double helix

superstructure

Karin Kornmueller1, Ilse Letofsky-Papst2, Kerstin Gradauer1, Christian Mikl1, Fernando Cacho-Nerin3,†,

Mario Leypold4, Walter Keller5, Gerd Leitinger6,7, Heinz Amenitsch3, and Ruth Prassl1 ()

Nano Res., Just Accepted Manuscript • DOI 10.1007/s12274-014-0683-9

http://www.thenanoresearch.com on December 8, 2014

© Tsinghua University Press 2014

Just Accepted

This is a “Just Accepted” manuscript, which has been examined by the peer-review process and has been

accepted for publication. A “Just Accepted” manuscript is published online shortly after its acceptance,

which is prior to technical editing and formatting and author proofing. Tsinghua University Press (TUP)

provides “Just Accepted” as an optional and free service which allows authors to make their results available

to the research community as soon as possible after acceptance. After a manuscript has been technically

edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP

article. Please note that technical editing may introduce minor changes to the manuscript text and/or

graphics which may affect the content, and all legal disclaimers that apply to the journal pertain. In no event

shall TUP be held responsible for errors or consequences arising from the use of any information contained

in these “Just Accepted” manuscripts. To cite this manuscript please use its Digital Object Identifier (DOI®),

which is identical for all formats of publication.

Nano Research

DOI 10.1007/s12274-014-0683-9

TABLE OF CONTENTS (TOC)



superstructure

Karin Kornmueller1, Ilse Letofsky-Papst2, Kerstin

Gradauer1, Christian Mikl1, Fernando Cacho-Nerin3,†,

Mario Leypold4, Walter Keller5, Gerd Leitinger6,7, Heinz

Amenitsch3, and Ruth Prassl1*

1Medical University of Graz, BioTechMed-Graz, Austria

2Graz University of Technology and Graz Centre for

Electron Microscopy, Austria

3,4Graz University of Technology, Austria

5Karl-Franzens-University Graz, Austria

6,7Medical University of Graz, Austria

†Present address: Diamond Light Source, United Kingdom

We tracked the concentration-dependent morphologies of an

amphiphilic designer peptide by combining small angle X-ray

scattering (SAXS) with Cryo-TEM and provide a detailed

global characterization of the supramolecular assemblies, as

well as insights into the internal organization at the

low-nanometer range. For the first time, a stable double helical

architecture within the class of amphipathic designer peptides is

presented, which holds promise as template for innovative

bioinspired nanomaterials.



superstructure



Received: day month year

Revised: day month year

Accepted: day month year

(automatically inserted by

the publisher)

© Tsinghua University Press

and Springer-Verlag Berlin

Heidelberg 2014

KEYWORDS

Double helix, amphiphilic

designer peptide,

self-assembly, SAXS,

TEM, spectroscopy

ABSTRACT

Hierarchical self-assembly is a fundamental principle in nature, which gives

rise to astonishing supramolecular architectures that offer an inspiration for the

development of innovative materials in nanotechnology. Here we present the

unique structure of a cone-shaped amphiphilic designer peptide. When

tracking its concentration-dependent morphologies, we observed elongated

bilayered single tapes at the beginning of the assembly process, which further

developed into novel double-helix-like superstructures at increased

concentrations. This architecture is characterized by a tight intertwisting of two

individual helices, resulting in a periodic pitch size over their total lengths of

several hundred nanometers. Solution X-ray scattering data revealed a marked

2-layered internal organization. All these characteristics remained unaltered for

the investigated period of almost three months. In their collective morphology

the assemblies are integrated into a network with hydrogel characteristics. Such

a peptide based structure holds promise for a building block of next-generation

nanostructured biomaterials.

Address correspondence to Ruth Prassl, [email protected]

Nano Research

DOI (automatically inserted by the publisher)

Research Article

1. INTRODUCTION. Nature offers elegant and

efficient ways to create highly organized systems

over several orders of magnitude by hierarchical

self-assembly. This basic principle has been

mimicked with the design of synthetic amphiphilic

molecules in order to create novel materials [1-3].

Our study is focused on the class of short

amphiphilic designer peptides. These molecules are

exclusively composed of amino acids and exhibit

lengths of 4-10 residues. As their basic characteristic,

they contain antagonistic hydrophilic and

hydrophobic residues within the same molecule [4].

Self-assembling amphiphilic designer peptides have

been explored as building blocks for the

development of controllable, tailor-made,

biocompatible nanomaterials only recently. Due to

their simplicity yet variability, there lies a huge

scientific as well as economic potential in this class

of molecules. Applications span the range from

material sciences to chemistry and nanotechnology,

and are of particular interest in nanomedicine due

to the peptides’ biocompatibility and

biodegradability [5-7]. A constantly increasing

global demand in innovative materials makes the

elucidation of their assembly mechanism highly

important.

A variety of supramolecular peptide-based

architectures has been reported so far, including

vesicles [8-10], spherical micelles [11-13], bilayered

tapes [14,15], donuts [16], helical and twisted

ribbons [11,12,17,18], fibers [12-14,18,19], rods

[14,19,20] and tubes [8-11,17,21,22]. Here we report

on a new structural motif that can be added to the

list of assemblies - for the first time a

double-helix-like morphology could be observed.

Only a few studies approached the elucidation of

supramolecular peptide-structures with scattering

techniques so far [11,17-19,21] and none of the

reported peptides showed the propensity to form

double helices. We used small angle X-ray

scattering (SAXS), complemented with transmission

electron microscopy (TEM), to investigate the

different morphologies of nanostructures derived

from varying concentrations of the amphiphilic

designer peptide GAAVILRR. It is important to

emphasize, that both techniques are operating in

different length scales and therefore give insights

into different parts of the assembled nanostructures.

SAXS allows to monitor the self-assembly of

biomolecules in solution without the need of

immobilization, labeling, or any other modification.

It provides a detailed insight into the internal

organization of structures in the low-nanometer

range (~ 3-60 nm) and therefore can be used to

describe the cross-section of extended structures in

high detail. Longitudinal extensions, which often

exceed the nanometer range, are well resolved with

TEM. In combination, both techniques give a

detailed picture of the individual structure as well

as the collective morphologies in solution.

Our data indicates that peptide concentration

directly relates to the morphology, order, and

complexity of the assembled structure. With

increasing peptide concentration we monitored

transitions from monomeric systems to elongated

single and double tapes and finally to

double-helix-like structures.

2. RESULTS AND DISCUSSION. The newly

designed amphiphilic peptide GAAVILRR is

composed of eight amino acid residues providing a

hydrophilic head and a hydrophobic tail. The head

consists of two bulky positively charged arginine

residues, whereas the tail is composed of aliphatic

amino acids with consecutively increasing

hydrophobicity. The N-terminus is acetylated,

keeping it uncharged. The C-terminus is amidated,

resulting in an overall charge distribution of two

positive net charges. The counter ion is

trifluoroacetate. Figure 1 illustrates the structure of

a monomer. Lyophilized samples of varying

concentrations were dissolved in water, resulting in

optically clear, homogeneous, colorless solutions.

The highest concentration (75 mM) exhibited a

www.theNanoResearch.com∣www.Springer.com/journal/12274 | Nano Research

3 Nano Res.

gel-like appearance. In order to obtain fully

developed structures, the samples were aged for 1

week. The molecule’s amphiphilic nature endows a

strong self-assembling ability in aqueous solution

above the critical aggregation concentration (CAC).

For GAAVILRR the CAC was estimated by

fluorescence depolarization [23] to be about 30 mM

in water (see Electronic Supplementary Material

(ESM), Fig. S1).

Figure 1 Molecular model of the amphiphilic designer-peptide

Ac-GAAVILRR-NH2. The hydrophobic tail is composed of

amino acids with consecutively increasing hydrophobicity and

size. Two arginine residues provide the positively charged

hydrophilic head. Peptide modifications include acetylation of

the N-terminus and amidation of the C-terminus. The

monomer’s length is approximately 3 nm and its widest lateral

extension is around 1.5 nm. The overall structure of the

molecule looks like a cone. Color code: green: carbon, blue:

nitrogen, gray: oxygen, and white: hydrogen.

Synchrotron SAXS was applied to investigate the

self-assembly of GAAVILRR in solution as a

function of peptide concentration. The individual

scattering curves showed concentration-dependent

characteristics and can broadly be divided into

three groups (see Fig. 2).

2.1 SAXS analysis of the low-concentration regime.

First, we examined the peptide’s morphology at

concentrations below or close to the CAC. This data

could be modeled with the theoretical form factor of

the peptide monomer adding a hard sphere

interaction term [24]. The form factor was

calculated with the program FoXS [25,26] from a

theoretical atomic model of the monomer (Fig. S2 in

the ESM). Figure 2a shows the scattering curves of

the peptide at concentrations ranging from 10-45

mM with their respective fits (green curves) as a

function of the scattering vector q. All curves show

the same characteristics with an almost linear

region in the low q-range, followed by a broad

shoulder at q = 1.3 nm-1. The parameters to describe

molecular interactions are obtained from the fitting

function. As seen in Fig. 2b, the hard-sphere radius

(Rs) and the volume fraction () remain constant for

the measured concentrations. The averaged value of

1.6 nm for Rs is in good agreement with a

monomer’s theoretical total length of 3 nm (≈ 2Rs).

The linear increase in the extrapolated intensity at

zero angle (I0) is directly related to the increasing

peptide concentration, also visible in the constant

value when I0 is normalized to the concentration.

Based on these characteristics we conclude that the

transition from a dilute, non-interacting colloidal

system (10 mM) to a more concentrated system (45

mM) occurs without supramolecular structure

formation.

2.2 Structural heterogeneity dominates the

intermediate concentration regime. A pronounced

change in structure was observed when the

peptide’s concentration was increased well above

the CAC. Here, we distinguished between a

transition region (55-65 mM, Fig. 2c/d) and the final

structure (75 mM, Fig. 2e/f). The SAXS pattern of

the 55 mM sample shows a large intensity increase

in the lower q-range, indicating the first steps of

peptide assembly. At a slightly elevated

concentration (60 mM) we observed a SAXS pattern

typical of elongated structures. Increasing the

concentration to 65 mM leads to a SAXS pattern

that already shows distinct features of the endpoint

structure, but does not show the high viscosity of

the 75 mM sample.

Figure 2 The concentration-dependent self-assembly of GAAVILRR. Scattering curves of GAAVILRR at different concentrations

after 1 week of aging are shown with their respective fits, schematic illustrations of the assembled structures and the corresponding

Cryo-TEM images. (a) Low concentration regime (10-45 mM). The used fitting function combines a hard sphere model with the

theoretical form factor of the peptide monomer, the fits are represented by the green curves. (b) illustrates the concentration

dependence of the fitted hard-sphere parameters, showing that the structures do not change over the investigated concentration regime.

The hard-sphere radius Rs and the volume fraction remain constant, whereas the extrapolated intensity at zero angle (I0) rises

linearly with concentration. I0 normalized to the concentration results in a constant value. (c) Intermediate concentration regime


5 Nano Res.

(55-65 mM). Several models obtained equally good fits (see ESM, Table S5 and Fig. S6). This is in agreement with the observation of

different intermediate structures with Cryo-TEM. (d) shows a selection of possible intermediates found at 60 mM. Single tapes (d1) as

well as parallel aligned double tapes (d2) are found. Double-helix-like structures are the dominant species (d3) and are characterized

by a periodicity between intersection points. The double helices are several hundreds of nm in length and 15-20 nm in width. All

intermediates have a related (single) tape width. (e) Hydrogel forming sample (75 mM). The scattering curve was fitted with a double

helix model combined with the theoretical form factor of the peptide monomer. (f) Cryo-TEM image of the double-helix-like

structures at a concentration of 75 mM. They are defined by the identical parameters as the assemblies in (d3) except for a slightly

shorter distance between the intersection points of ~ 80 nm.

SAXS measurements were complemented with

TEM, which allows for a direct visualization of the

assemblies. TEM pictures revealed a quite

heterogeneous distribution of structures in the

intermediate concentration regime:

double-helix-like structures are the dominant

species, in coexistence with extended tapes and

double tapes (see Fig. 2, d1-d3). Uniformity in

tape-width was observed throughout all

intermediate morphologies in TEM.

2.3 Model development to fit the scattering curve

of the highest peptide concentration. Increasing

the peptide’s concentration to 75 mM results in a

highly viscous solution, typical of hydrogels that

are frequently reported in the context of

amphiphilic peptides [12,27]. Its scattering pattern

is shown in Fig. 2e, exhibiting four distinct features:

a broad shoulder at q ≈ 0.1-0.35 nm-1 and three side

maxima with consecutively decreasing intensity at

q = 0.5, 1.18 and 2.4 nm-1, respectively. In a first

approach and for completeness, we tested various

torus-like models to fit this curve, although TEM

images suggested elongated structures. The

rationale behind was that Khoe et al. [16]

investigated a peptide with a similar sequence,

lacking just one alanine residue, forming donut-like

structures in solution. However, neither a torus, nor

a torus with an elliptical cross-section, shell-models

thereof [28], nor elongated stacked tori could

properly describe our scattering curve (for a

description of the fitting functions and the

respective fits see ESM, Fig. S3 and Table S1). Also a

twisted ribbon model [29,30] can be excluded due to

its non-perfect fit (see ESM Fig. S4 and Table S2).

When we applied a helical fitting function, the fit

improved vastly (see ESM, Fig. S5 and Table S3).

But it was only after extending the helical model [30]

to a layered double helix, based on the equations of

Pringle et al. [31] that fitting results could be

obtained with a minimal chi-square value [32]. In

our study, a double helix is defined as two single

helices that fit into each other and are displaced by

the angle . This formalism introduces a coupling of

the different layer lines typical for the helical

diffraction pattern [33]. The amplitudes of the single

layer lines given by the values of tape width and

discriminate between single or double helices.

Equally good fits could be achieved by using single

and double helix models for the intermediate

structures from 55 to 65 mM (for details see ESM,

Table S5 and Fig. S6). As stated above, this

concentration regime represents an intermediate

state of a dynamic structure evolution process,

where coexistence of a variety of morphologies is

given. Single helical tapes as well as

double-helix-like structures are present, and even

the existence of twisted ribbons often described as

precursors in helix formation cannot be excluded.

This heterogeneity is represented by a non-perfect

fit of any single model to the SAXS data of the

intermediate concentrations. Ensemble analysis

shows a distribution of approximately 65 % single

helices and 35 % double helices (ESM Fig. S7 and

Table S6). The situation is different for the highest

peptide concentration: here the population of

| www.editorialmanager.com/nare/default.asp

6 Nano Res.

structures is more homogenous. Visually, the

assemblies can be best described as two individual

ribbons that are intertwined – by no means loosely

connected, but tightly associated, showing highly

regular repeat distances at their intersection points

(see Fig. 2f). Due to this regularity and homogeneity

in their total widths, an interaction between the two

strands seems likely. For this reason we describe

them as double-helix-like structures, and indeed,

fitting with the developed 2-layered double helix

model resulted in an excellent fit to the

experimentally obtained scattering data.

2.4 The SAXS-derived double helical model. The

X-ray derived parameters describing the novel

2-layered double helical model are illustrated in Fig.

3 (see also ESM Table S4). The double helix exhibits

a pitch (P) of ~ 60 nm and a mean radius (R) of 6 nm,

resulting in a helical twist angle of ~ 60°. The tape

height along the longitudinal axis (wh) is ~ 23 nm

(Fig. 3a). The diameter of the double helix is ~ 16

nm (Fig. 3b), indicating that no change in the

overall width of the tape has occurred upon

hydrogel formation. Most importantly, the electron

density profile () derived from the SAXS data

allows for a sectioning of the tape into several layers

(Fig. 3c). Accordingly, the peptide-containing region

in total accounts for ~ 3.5 nm. The inner

hydrophobic core has the highest electron density

and spans around 1 nm (D1). In this innermost

region we expect antiparallel stacking of the

peptide monomers due to intermolecular hydrogen

bonding. Given a monomer length of ~ 3 nm we

assume a very tight packing and interdigitation of

two monomers and we suggest that this

organization determines the high stability of the

supramolecular structures. Figure 3d shows one

sterically permitted arrangement for the stacking of

a peptide pair and the respective hydrogen bond

distribution. The hydrophobic core region is

flanked by a region providing the space for the

hydrophilic head groups. A decrease in the electron

density from 1 = 1.4 to 2 = 0.8 electrons/Å 3 points

to a less densely packed zone which indicates a

higher flexibility of the head groups. In addition,

the electron density distribution map shows an area

which is characterized by a diffuse scattering

contribution (3). We attribute this region to a low

fraction of twisted ribbons (below 5 %), which are

coexistent in the sample. They are characterized by

the same tape width (12 nm) as the tapes forming

the double-helix-like structures. The structural

morphology of the highest peptide concentration

was also approached by three different

TEM-techniques (freeze-fracture, negative-staining

and Cryo-TEM). All three techniques support our

model of two individual tapes that intertwine to

form the double-helix-like structures. The images,

as well as a discussion on the strengths and

limitations of either technique, can be found in the

ESM (Fig. S10).

2.5 Spectroscopic techniques to monitor the

structure’s internal organization. To obtain

additional information on the internal organization

of the structures and to confirm our model of

antiparallel organized monomers (as proposed in

Fig. 3d), we applied circular dichroism (CD) and

attenuated total reflection Fourier transform

infrared (ATR FT-IR) spectroscopy. In principle, the

obtained CD spectra demonstrate a transition from

mainly random-coil (1-45 mM, Fig. 4a) to

predominantly -type structures (60 mM, Fig. 4b).

The lowest peptide concentrations show a large

negative p→p* transition at 198-202 nm followed by

a negative band with a maximum at 220 nm, which

can be assigned to the n→p* transition. These

characteristics are associated with a mainly

random-coil structure. The large negative peak

becomes less pronounced with increasing peptide

concentration and is shifted to higher wavelengths.

This change goes in line with a rapid increase in the

high tension (HT) voltage and therefore these parts

of the spectra (dotted lines) should be considered

Figure 3 Structural parameters of the double helical model. Values are based on the 75 mM sample fit presented in Fig. 2e. (a)

Side-view of the double helix, composed of GAAVILRR molecules. Two individual strands are intertwisted to form a

double-helix-like structure. The illustrated parameters are the pitch (P), the helical twist angle (), the tape thickness (w), the tape

height (wh), the free space between one tape (h), and the corresponding free space in the bilayer plane (d). (b) The cross-section is

showing the mean radius (R) and the angle by which the two tapes are displaced with respect to each other. (c) Electron density

distribution of the cross-section, which can be divided into individual shells with D1-D2 being the shell-distances with the electron

densities 1-2. 3 (gray mesh) reflects the contribution of a low fraction of twisted ribbons in the sample. The dashed line represents

the symmetry line. In the innermost region (highlighted in green) we expect hydrogen bonding between antiparallel aligned peptide

molecules. (d) represents the hypothesized arrangement of two peptide molecules together with a proposed hydrogen bond

distribution pattern depicted in the chemical formulas. Color code of the molecular model of GAAVILRR: green: carbon, blue:

nitrogen, gray: oxygen, and white: hydrogen.

with caution. Nevertheless, all spectra of the low

concentration regime point to a random-coil

structure. Within the 60 mM sample the -strand

content significantly increased and the

spectroscopic characteristics developed towards a

negative maximum at 230 nm (Fig. 4b). Although

very few reports exist on how to interpret this large

negative peak at 230 nm there is a consensus that it

is related to -type structures [18,34,35]. A red shift

in signal has been associated with a higher degree

of twisting in -sheets compared to planar -sheets

[27]. CD spectra with striking similarity to ours

were reported in literature [36,37], interestingly all

of which were related to double helices. We now

hypothesize that a CD spectrum like ours -

characterized by the absence of the typical positive

peak at 190-200 nm and a single large negative band

around 230 nm - could be directly associated with a

double helix morphology, held together by

-strands.

ATR-FTIR measurements confirmed the presence of

-sheet structures (Fig. 4c). The major amide I peak

at 1620 nm-1 is characteristic of -sheet interactions.

The weak band at 1690 nm-1 can be assigned to an

antiparallel component, whereas the contribution at

1673 nm-1 arises from residual trifluoroacetic acid

(TFA) in the sample. In comparison, the 60 mM and

75 mM samples showed in both cases the same

spectroscopic characteristics, but also the typical

scaling of the signal with concentration.

The results of CD and ATR-FTIR spectroscopy go in

line with our proposed model of tightly packed,

interdigitated peptide molecules that are held

together by hydrogen bonding due to antiparallel

stacking.

Figure 4 The transition from random-coil (1-45 mM) to -type

structures (60-75 mM) measured by circular dichroism and

infrared spectroscopy. (a) CD spectra of the low-concentration

samples have been recorded after 10 days of incubation. (b) CD

spectra of the 60 mM sample were recorded at different time

points over an incubation time of 4 days. Dotted lines represent

parts of the spectra where a threshold of 600 Volts in the high

tension (HT) signal was exceeded. (c) ATR-FTIR spectroscopy

confirms the presence of antiparallel -sheets (peaks at 1620

and 1690 cm-1), the peak at 1673 nm-1 is indicative of residual

TFA.


9 Nano Res.

2.6 Evolution of structure over time. To test,

whether the double-helix-like structures are

intermediates and develop towards often described

tubular structures or whether they are a final state

of the assembly, we performed a series of SAXS and

Cryo-TEM measurements at different time points

over 11 weeks. The SAXS pattern of the 75 mM

sample remained unchanged over the observed

period of time (1, 4 and 11 weeks), revealing that

the overall structure and the internal organization

do not change (see ESM, Fig. S8). These results were

further supported by Cryo-TEM micrographs,

where stable double-helix-like structures were

observed at every time point (see Fig. 5). Already 30

minutes after dissolving, a double-helix-like

morphology developed and showed the final

widths of ~ 15-20 nm and a mean distance between

periodic intersection points of ~ 80 nm (see white

arrows in Fig. 5). These characteristics remained

unaltered for at least 11 weeks.

Figure 5 Cryo-TEM of 75 mM GAAVILRR at different time points shows no alteration in the global morphology. (a) freshly

dissolved (b) after 1 week (c) 4 weeks (d) 11 weeks. The peptide forms double helices at every time point, with similar structural

parameters: width ~ 15-20 nm and intersection points every ~ 80 nm (see white arrows). Scale bar = 200 nm.

An interesting point that needs to be addressed is

the strikingly different morphology of

GAAVILRR compared to the structure of the

peptide GAVILRR reported by Khoe et al. [16],

which differs from our sequence in just one

lacking alanine residue. While GAVILRR forms

donut-like structures in solution, we have found

double-helix-like assemblies for GAAVILRR.

These results highlight the fact that minor

differences in the amino acid composition can

have a huge influence in the resulting structures.

In fact, this feature is not unique and systematic

variations of the tail length or the substitution of

single amino acids in a given sequence are a

matter of intense research [12,13,18,27,38,39]. A

deeper understanding of the role of every single

amino acid opens up the opportunity for tunable

structures. They are highly desirable, when

thinking of the multifaceted application areas for

supramolecular self-assembled nanostructures.

Whether the double helical arrangement of

amphiphilic designer peptides has been

overlooked before due to experimental

limitations or is a unique, novel feature related to

peptide-sequence remains an open issue.

3. CONCLUSIONS. We have shown that the

structure of supramolecular peptide-assemblies

in solution strongly depends on the concentration

used. Below the CAC the peptides are dispersed

as monomers and do not display superstructural

features. According to Cenker et al. [40], as soon

as lyophilized peptides are dispersed in water

above their CAC, one part of the molecules is

present in the form of aggregates, whereas the

other part (equal to the concentration of the CAC)

is in its molecularly dissolved state. In a

time-dependent crystal growth process the

aggregates now grow by uniaxial extension

through the assembly of monomers. In a

dissolution-reassembly-mechanism structures

with a high level of perfection are grown within

minutes to a few hours. At this stage, the

structures already exhibit a defined cross-section

with their equilibrium diameters [40]. We

observed bilayered tapes, where we expect the

hydrophilic arginine residues being exposed to

the solvent and the hydrophobic tail residues

being interdigitated and tightly packed in the

interior, stabilized by intermolecular hydrogen

bonding. Due to the peptides’ intrinsic chirality,

monomers do not assemble strictly parallel but

slightly tilted with each other, leading to twisted

structures. In addition, a non-planar alignment of

the monomers leads to bending. This accounts for

a twisted-to-helical transition on the one hand

and on the other hand limits lateral growth, but

favors uniaxial elongation [29,41,42]. Hence,

double-helix-like structures with several hundred

nanometers to micrometers in length but only

around 16 nm in width can develop. Due to the

high regularity of pitch distances, the orientation

of the two individual strands seems to be the

consequence of a sensitively related interplay

between geometrical constraints, intertape

hydrogen bonding interactions, the hydrophobic

affinity of the tail residues and the electrostatic

repulsion of positively charged arginine residues.

Finally, the double-helix-like superstructures

assemble into a network displaying gel-like

features. As these intertwined assemblies are

stable for several months, this architecture seems

energetically and sterically highly favorable.

Having in mind the manifold applications for

supramolecular self-assembled structures

[4,5,43-47], a double-helical-like tape might be the

building block for the next generation of

bioinspired nanomaterials.

4. METHODS

4.1 Peptide Preparation. Ac-GAAVILRR-NH2

was custom synthesized and purified by

piCHEM GmbH (Graz, Austria) following the


11 Nano Res.

general procedure for Fmoc solid-phase peptide

synthesis. Samples were prepared by dissolving

the lyophilized peptide in double distilled water

to obtain the desired concentrations. The samples

were incubated at 4 °C under argon atmosphere

for 1 week prior to SAXS and TEM

measurements.

4.2 Small Angle X-Ray Scattering (SAXS).

Synchrotron X-ray scattering data was collected

at the Austrian SAXS beamline at ELETTRA

(Trieste, Italy) [48]. Measurements were

conducted at a wavelength of 0.154 nm and a

sample-detector distance of 1.1 m. The photon

energy was 8 keV. Data was recorded with a

Pilatus detector (PILATUS 100K, DECTRIS Ltd.,

Villigen PSI, Switzerland), calibrated with silver

behenate. The scattering intensity was measured

as a function of the scattering vector q where

q = 4 (sin)/

with 2 being the scattering angle and being

the wavelength. Samples were measured either in

a 1.5 mm glass capillary (10-60 mM) or in a gel

sample holder (75 mM). A discussion on the

influence of shearing forces when filling the

capillary can be found in the ESM (Table S7 and

Fig. S8). Data analysis was done with Fit2D [49]

and Igor Pro (Version 6.22A, WaveMetrics Inc.,

USA). Experimental intensities were normalized

to the sample transmission and corrected for

background. Absolute calibration was done by

using water as secondary standard after the

method described by Orthaber [50].

4.3 Formal description of a helix and a double

helix – fitting parameters. The mean radius (R) is

the distance between the helix center to the center

of the innermost shell. The cross section of our

model comprises two to three layers with their

respective distances (D1-D3) and electron

densities (1-3). solvent is the electron density of

the solvent, which was kept constant at 0.335 e/Å 3.

The mean radius and the pitch (P) provide the

helical twist angle , where = arctan[P/(2R)]. It

describes the angle of the helix with respect to the

equatorial plane. The rotation angle reflects the

degree of opening of the helix. The thickness of

the tape w is defined as w = D3sin, the tape

height wh = w/cos, and the free space between

the tapes h = P-wh. The free distance (d) between

the tapes in the bilayer plane is defined as d =

(wh)/wh [30,51].

A double helix results from two single helical

tapes (with the same properties as described

above) that are displaced by the angle with

each other. The pitch of a single helix needs to be

greater than the tape height wh in order to leave

room for the second tape.

4.4 Cryogenic Transmission Electron

Microscopy (Cryo-TEM). Samples were adsorbed

onto a glow-discharged carbon-coated copper

grid, blotted to create a thin film, plunged into

liquid ethane and transferred to liquid nitrogen.

Vitrified specimens were transferred onto a

Gatan 626-DH cryo transfer specimen holder and

imaged at cryogenic temperatures in a Tecnai T12

(FEI, The Netherlands) microscope, operated at

120 kV.

4.5 Circular Dichroism (CD) Spectroscopy.

Measurements were conducted on a JASCO J-715

Circular Dichroism Spectrometer (Jasco Inc.,

Easton, MD) using a 0.1 mm pathlength cell.

Spectra were recorded between 185 and 260 nm

in an interval scan mode with 0.2 nm resolution

and an average time of 1 sec. Scan speed was 50

nm/min. Spectra of the low concentration

samples (10-45 mM) were recorded after 10 days

of incubation. The 60 mM sample was filled into

the capillary directly after dissolution and its

structure evolution was monitored over 4 days.

Due to the high viscosity of the 75 mM sample it

was not possible to fill the sample into a


12 Nano Res.

measurement cell. The final spectra were

corrected for background by subtracting the

corresponding buffer spectra obtained under

identical conditions. Data processing and

analysis were performed with CDtool [52] and

Igor Pro (Version 6.22A, WaveMetrics Inc., USA).

The threshold of the high tension (HT) voltage

signal was set to 600 Volts. Parts of the CD

spectra exceeding this threshold are marked by

dotted lines.

4.6 Attenuated Total Reflection Fourier

Transform Infrared (ATR FT-IR) Spectroscopy.

Infrared spectra were recorded from 4000 to 600

cm-1 on a Bruker TENSOR 37 FT-IR spectrometer,

equipped with an ATR accessory (Bruker Optics

Inc., Billerica, USA). Spectra were averages of 128

scans with a resolution of 4 cm-1. Data analysis

was performed with the OPUS software (Version

6.5, Bruker Optics Inc., Billerica, USA).

4.7 Artwork. Illustrations of the assembled

structures were created with Blender (Version

2.66, Blender Foundation) [53] and Adobe

Photoshop CS5 Extended (Version 12.0.3 x32,

Adobe Systems Inc., USA). Chemical formulas

were drawn with ChemDraw (Version 6.0.1,

Perkin Elmer Informatics, USA).

Acknowledgements

This work has been supported by the Austrian

Science Fund (FWF Project No. I 1109-N28 to R. P.).

We thank Elisabeth Bock and Gertrud Havlicek for

technical assistance with electron microscopy. We

gratefully acknowledge Rolf Breinbauer for

providing access to infrared spectroscopy

instrumentation.

Electronic Supplementary Material:

Supplementary material (Experimental details;

determination of the CAC; SAXS: fitting functions:

hard-sphere model, theoretical form factor of a

peptide monomer, torus, torus with elliptical

cross-section, core-shell model of a torus and a

torus with elliptical cross-section, stacked models

thereof, helix, double helix; fits and derived

parameters of the torus-, twisted ribbon and

helix-models, ensemble analysis for coexisting

structures, time evolution of the 75 mM sample.

The influence of shearing forces on the resulting

SAXS pattern when filling the capillary.

Comparison of freeze-fracture, negative-staining

and Cryo-TEM techniques for the direct

visualization of superstructures) is available in the

online version of this article at

http://dx.doi.org/10.1007/s12274-***-****-*

(automatically inserted by the publisher). References

[1] Luo, Z.; Zhang, S. Designer nanomaterials using chiral

self-assembling peptide systems and their emerging

benefit for society. Chem. Soc. Rev. 2012, 41, 4736-4754.

[2] Zelzer, M.; Ulijn, R. V. Next-generation peptide

nanomaterials: molecular networks, interfaces and

supramolecular functionality. Chem. Soc. Rev. 2010, 39,

3351-3357.

[3] Aida, T.; Meijer, E. W.; Stupp, S. I. Functional

supramolecular polymers. Science 2012, 335, 813-817.

[4] Zhao, X.; Pan, F.; Xu, H.; Yaseen, M.; Shan, H.; Hauser,

C. A.; Zhang, S.; Lu, J. R. Molecular self-assembly and

applications of designer peptide amphiphiles. Chem. Soc.

Rev. 2010, 39, 3480-3498.

[5] Ellis-Behnke, R. G.; Liang, Y. X.; You, S. W.; Tay, D. K.;

Zhang, S.; So, K. F.; Schneider, G. E. Nano neuro

knitting: peptide nanofiber scaffold for brain repair and

axon regeneration with functional return of vision. Proc.

Natl. Acad. Sci. U.S.A. 2006, 103, 5054-5059.

[6] Ellis-Behnke, R. G.; Schneider, G. E. Peptide

amphiphiles and porous biodegradable scaffolds for

tissue regeneration in the brain and spinal cord. Methods

Mol. Biol. 2011, 726, 259-281.

[7] Hartgerink, J. D.; Beniash, E.; Stupp, S. I. Self-assembly

and mineralization of peptide-amphiphile nanofibers.

Science 2001, 294, 1684-1688.

[8] Vauthey, S.; Santoso, S.; Gong, H.; Watson, N.; Zhang, S.

Molecular self-assembly of surfactant-like peptides to

form nanotubes and nanovesicles. Proc. Natl. Acad. Sci.

U.S.A. 2002, 99, 5355-5360.

[9] Santoso, S.; Hwang, W.; Hartman, H.; Zhang, S. G.

Self-assembly of surfactant-like peptides with variable

glycine tails to form nanotubes and nanovesicles. Nano

Lett. 2002, 2, 687-691.

[10] von Maltzahn, G.; Vauthey, S.; Santoso, S.; Zhang, S. U.

Positively charged surfactant-like peptides self-assemble

into nanostructures. Langmuir 2003, 19, 4332-4337.

http://dx.doi.org/10.1007/s12274-***-****-*


13 Nano Res.

[11] Xu, H.; Wang, Y.; Ge, X.; Han, S.; Wang, S.; Zhou, P.;

Shan, H.; Zhao, X.; Lu, J. R. Twisted Nanotubes Formed

from Ultrashort Amphiphilic Peptide I3K and Their

Templating for the Fabrication of Silica Nanotubes.

Chem. Mater. 2010, 22, 5165-5173.

[12] Lakshmanan, A.; Hauser, C. A. Ultrasmall peptides

self-assemble into diverse nanostructures: morphological

evaluation and potential implications. Int. J. Mol. Sci.

2011, 12, 5736-5746.

[13] Han, S.; Cao, S.; Wang, Y.; Wang, J.; Xia, D.; Xu, H.;

Zhao, X.; Lu, J. R. Self-assembly of short peptide

amphiphiles: the cooperative effect of hydrophobic

interaction and hydrogen bonding. Chem. - Eur. J. 2011,

17, 13095-13102.

[14] Xu, H.; Wang, J.; Han, S.; Wang, J.; Yu, D.; Zhang, H.;

Xia, D.; Zhao, X.; Waigh, T. A.; Lu, J. R.

Hydrophobic-region-induced transitions in

self-assembled peptide nanostructures. Langmuir 2009,

25, 4115-4123.

[15] Pan, F.; Zhao, X.; Perumal, S.; Waigh, T. A.; Lu, J. R.;

Webster, J. R. P. Interfacial Dynamic Adsorption and

Structure of Molecular Layers of Peptide Surfactants.

Langmuir 2009, 26, 5690-5696.

[16] Khoe, U.; Yang, Y. L.; Zhang, S. G. Self-Assembly of

Nanodonut Structure from a Cone-Shaped Designer

Lipid-like Peptide Surfactant. Langmuir 2009, 25,

4111-4114.

[17] Castelletto, V.; Nutt, D. R.; Hamley, I. W.; Bucak, S.;

Cenker, C.; Olsson, U. Structure of single-wall peptide

nanotubes: in situ flow aligning X-ray diffraction. Chem.

Commun. 2010, 46, 6270-6272.

[18] Hauser, C. A.; Deng, R.; Mishra, A.; Loo, Y.; Khoe, U.;

Zhuang, F.; Cheong, D. W.; Accardo, A.; Sullivan, M. B.;

Riekel, C. et al. Natural tri- to hexapeptides

self-assemble in water to amyloid beta-type fiber

aggregates by unexpected alpha-helical intermediate

structures. Proc. Natl. Acad. Sci. U.S.A. 2011, 108,

1361-1366.

[19] Wang, J.; Han, S.; Meng, G.; Xu, H.; Xia, D.; Zhao, X.;

Schweins, R.; Lu, J. R. Dynamic self-assembly of

surfactant-like peptides A6K and A9K. Soft Matter 2009,

5, 3870-3878.

[20] Baumann, M. K.; Textor, M.; Reimhult, E.

Understanding Self-Assembled Amphiphilic Peptide

Supramolecular Structures from Primary Structure Helix

Propensity. Langmuir 2008, 24, 7645-7647.

[21] Bucak, S.; Cenker, C.; Nasir, I.; Olsson, U.; Zackrisson,

M. Peptide Nanotube Nematic Phase. Langmuir 2009, 25,

4262-4265.

[22] Wang, S.; Ge, X.; Xue, J.; Fan, H.; Mu, L.; Li, Y.; Xu, H.;

Lu, J. R. Mechanistic Processes Underlying Biomimetic

Synthesis of Silica Nanotubes from Self-Assembled

Ultrashort Peptide Templates. Chem. Mater. 2011, 23,

2466-2474.

[23] Yang, S.; Zhang, S. Self-assembling behavior of designer

lipid-like peptides. Supramol. Chem. 2006, 18, 289-396.

[24] Pontoni, D.; Finet, S.; Narayanan, T.; Rennie, A. R.

Interactions and kinetic arrest in an adhesive hard-sphere

colloidal system. J. Chem. Phys. 2003, 119, 6157.

[25] Schneidman-Duhovny, D.; Hammel, M.; Tainer, J. A.;

Sali, A. Accurate SAXS profile computation and its

assessment by contrast variation experiments. Biophys. J.

2013, 105, 962-974.

[26] Schneidman-Duhovny, D.; Hammel, M.; Sali, A. FoXS:

a web server for rapid computation and fitting of SAXS

profiles. Nucleic Acids Res. 2010, 38, W540-W544.

[27] Pashuck, E. T.; Cui, H.; Stupp, S. I. Tuning

supramolecular rigidity of peptide fibers through

molecular structure. J. Am. Chem. Soc. 2010, 132,

6041-6046.

[28] Kawaguchi, T. Radii of gyration and scattering functions

of a torus and its derivatives. J. Appl. Crystallogr. 2001,

34, 580-584.

[29] Selinger, J. V.; Spector, M. S.; Schnur, J. M. Theory of

Self-Assembled Tubules and Helical Ribbons. J. Phys.

Chem. B 2001, 105, 7157-7169.

[30] Teixeira, C. V.; Amenitsch, H.; Fukushima, T.; Hill, J. P.;

Jin, W.; Aida, T.; Hotokka, M.; Lindén, M. Form factor

of an N-layered helical tape and its application to

nanotube formation of

hexa-peri-hexabenzocoronene-based molecules. J. Appl.

Crystallogr. 2010, 43, 850-857.

[31] Pringle, O. A.; Schmidt, P. W. Small-angle X-ray

scattering from helical macromolecules. J. Appl.

Crystallogr. 1971, 4, 290-293.

[32] Pedersen, J. S. Analysis of small-angle scattering data

from colloid and polymer solutions: modeling and

least-square fitting. Adv. Colloid Interface Sci. 1997, 70 ,

171-210.

[33] Hamley, I. W. Form Factor of Helical Ribbons.

Macromolecules 2008, 41, 8948-8950.

[34] Rudra, J. S.; Tian, Y. F.; Jung, J. P.; Collier, J. H. A

self-assembling peptide acting as an immune adjuvant.

Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 622-627.

[35] Collier, J. H.; Messersmith, P. B. Enzymatic

Modification of Self-Assembled Peptide Structures with

Tissue Transglutaminase. Bioconjugate Chem. 2003, 14,

748-755.

[36] Hicks, M. R.; Damianoglou, A.; Rodger, A.; Dafforn, T.

R. Folding and Membrane Insertion of the Pore-Forming

Peptide Gramicidin Occur as a Concerted Process. J. Mol.

Biol. 2008, 383, 358-366.

[37] Chen, C. L.; Zhang, P.; Rosi, N. L. A New Peptide-Based

Method for the Design and Synthesis of Nanoparticle

Superstructures: Construction of Highly Ordered Gold

Nanoparticle Double Helices. J. Am. Chem. Soc. 2008,

130, 13555-13557.

[38] Paramonov, S. E.; Jun, H. W.; Hartgerink, J. D.

Self-assembly of peptide-amphiphile nanofibers: the

roles of hydrogen bonding and amphiphilic packing. J.

Am. Chem. Soc. 2006, 128, 7291-7298.

[39] Dong, H.; Paramonov, S. E.; Aulisa, L.; Bakota, E. L.;

Hartgerink, J. D. Self-assembly of multidomain peptides:

balancing molecular frustration controls conformation

and nanostructure. J. Am. Chem. Soc. 2007, 129,

12468-12472.

[40] Cenker, C. C.; Bomans, P. H. H.; Friedrich, H.; Dedeoglu,

B.; Aviyente, V.; Olsson, U.; Sommerdijk, N. A. J. M.;


14 Nano Res.

Bucak, S. Peptide nanotube formation: a crystal growth

process. Soft Matter 2012, 8, 7463-7470.

[41] Fishwick, C. W. G.; Beevers, A. J.; Carrick, L. M.;

Whitehouse, C. D.; Aggeli, A.; Boden, N. Structures of

Helical -Tapes and Twisted Ribbons: The Role of

Side-Chain Interactions on Twist and Bend Behavior.

Nano Lett. 2003, 3, 1475-1479.

[42] Armon, S.; Aharoni, H.; Moshe, M.; Sharon, E. Shape

selection in chiral ribbons: from seed pods to

supramolecular assemblies. Soft Matter 2014, 10,

2733-2740.

[43] Cui, H.; Webber, M. J.; Stupp, S. I. Self-assembly of

peptide amphiphiles: from molecules to nanostructures to

biomaterials. Biopolymers 2010, 94, 1-18.

[44] Hamley, I. W. Self-assembly of amphiphilic peptides.

Soft Matter 2011, 7, 4122–4138.

[45] Yao, Y.; Xue, M.; Chen, J.; Zhang, M.; Huang, F. An

amphiphilic pillar[5]arene: synthesis, controllable

self-assembly in water, and application in calcein release

and TNT adsorption. J. Am. Chem. Soc. 2012, 134,

15712-15715.

[46] Yao, Y.; Xue, M.; Zhang, Z.; Zhang, M.; Wang, Y.;

Huang, F. Gold nanoparticles stabilized by an

amphiphilic pillar[5]arene: preparation, self-assembly

into composite microtubes in water and application in

green catalysis. Chem. Sci. 2013, 4, 3667-3672.

[47] Yu, G.; Ma, Y.; Han, C.; Yao, Y.; Tang, G.; Mao, Z.; Gao,

C.; Huang, F. A sugar-functionalized amphiphilic

pillar[5]arene: synthesis, self-assembly in water, and

application in bacterial cell agglutination. J. Am. Chem.

Soc. 2013, 135, 10310-10313.

[48] Amenitsch, H.; Rappolt, M.; Kriechbaum, M.; Mio, H.;

Laggner, P.; Bernstorff, S. First performance assessment

of the small-angle X-ray scattering beamline at

ELETTRA. J. Synchrotron Rad. 1998, 5, 506-508.

[49] European Synchrotron Radiation Facility, Fit2D.

http://www.esrf.eu/computing/scientific/FIT2D/

(accessed August, 2014).

[50] Orthaber, D.; Bergmann, A.; Glatter, O. SAXS

experiments on absolute scale with Kratky systems using

water as a secondary standard. J. Appl. Crystallogr. 2000,

33, 218-225.

[51] Chung, D. S.; Benedek, G. B.; Konikoff, F. M.; Donovan,

J. M. Elastic free energy of anisotropic helical ribbons as

metastable intermediates in the crystallization of

cholesterol. Proc. Natl. Acad. Sci. U.S.A. 1993, 90,

11341-11345.

[52] Lees, J. G.; Smith, B. R.; Wien, F.; Miles, A. J.; Wallace,

B. A. CDtool-an integrated software package for circular

dichroism spectroscopic data processing, analysis, and

archiving. Anal. Biochem. 2004, 332, 285-289.

[53] Blender Foundation. http://www.blender.org (accessed

August, 2014).


Nano Res.

Electronic Supplementary Material



superstructure



Supporting information to DOI 10.1007/s12274-****-****-* (automatically inserted by the publisher)


Nano Res.

Determination of the critical aggregation concentration (CAC)

The CAC is defined as the concentration of a surfactant above which highly organized supramolecular

structures spontaneously form. The CAC was determined with a fluorescence assay described by Yang [1].

1,6-Diphenyl-1,3,5-hexatriene (DPH) was purchased from Sigma-Aldrich (Vienna, Austria). 5 µ l of 30 mM

DPH in chloroform were added to 150 µl of peptide solution at varying concentrations. Samples were

incubated overnight at 4°C, gently shaking. Fluorescence spectra were recorded on a SPEX FluoroMax-3

spectrofluorimeter (Jobin-Yvon Horiba, USA). The excitation wavelength was 350 nm, fluorescence emission

was recorded at 425 nm.

Below the CAC, surfactant molecules as well as the hydrophobic probe DPH are randomly oriented and thus

the fluorescence signal is quenched. As soon as surfactant molecules assemble into supramolecular structures

DPH is aligned within their hydrophobic milieus and the fluorescence intensity increases linearly. When the

measured fluorescence intensity is plotted against the peptide concentration, two straight lines can be fitted

into the plot – the low intensity region and the linearly increasing intensity region. The concentration at the

intersection point of these two lines is defined as the CAC of the peptide. The CAC of GAAVILRR was

determined to be around 30 mM in water (Figure S1).

Figure S1 Determination of the CAC.


Nano Res.

Small angle X-ray scattering (SAXS) - development of the structure factors

1.1 The low concentration regime (10-45 mM)

a) Atomic model

A theoretical atomic model of the peptide monomer Ac-GAAVILRR-NH2 was generated using the

builder-function in PyMOL, a molecular visualization software for rendering and animating 3D molecular

structures (The PyMOL Molecular Graphics System, Version 1.2r1, Schrödinger, LLC.). The generated model

was saved in the Protein Data Bank (PDB) format (http://www.wwpdb.org/docs.html).

b) Theoretical form factor Ftheo(q)2

In order to obtain the theoretical form factor of the peptide monomer, a theoretical scattering curve was

calculated using the program FoXS [2]. The atomic coordinates were derived from the atomic model

generated with the program PyMOL. The configuration was done by using the default input parameters.

Figure S2 illustrates the theoretical atomic model of GAAVILRR and the derived theoretical scattering curve.

Figure S2 Theoretical form factor of the peptide monomer. The theoretical atomic model of GAAVILRR was generated with PyMOL

and the derived theoretical scattering curve was created with FoXS. Color code: green: carbon, blue: nitrogen, gray: oxygen, and white:

hydrogen.

c) Structure factor

A hard sphere model [3] Shardsphere was combined with the theoretical form factor Ftheo(q)2 of the peptide

monomer.

2,,)( qFqRSCqI theoshardsphere

(1)

where is the volume fraction, Rs the hard sphere radius and C is a scaling constant. Within the low

concentration regime I0 - the extrapolated intensity at zero angle - was derived by averaging the intensities in


Nano Res.

the linear region between q = 0.22 and 0.50 nm-1 of the individual scattering curves.

1.2 Form factor of a helix and a double helix

The form factor of an N-layered helical tape [4] Fhelical(q)2 was combined with the theoretical form factor of the

peptide monomer. As the structural dimensions between the helical superstructure and the monomers are

very different, the helical form factor has only contributions in the regime, in which the scattering from the

peptide monomer is constant. As a consequence the helical superstructure can be treated as an additive term.

220 ,,)()( qFqRSCqFIqI theoshardspherehelical

(2)

with the fitting parameters I0 and C, besides the structural parameters of the helical ribbon. The parameters Rs

and have been fixed with the values derived from the monomers at low concentration.

The helical form factor was based on the one for a single tape [4] and has been extended to a double helix

system according to Pringle and Schmidt [5]. The total scattering intensity can be generalized for a double

helical tape consisting of N shells as:

22

2

2

2

2 ),,,(

2

2sin

2cos)( qPRGn

nn

qLCqF

nshelln

n

shelln

helical

(3)

where

mR

nm

N

mmmmnshell

Pq

nqrrdrJqPRG

0

2

11

212)(),,,(

(4)

and

N

mmmmshelln RC

1

21

(5)

L denotes the length of the tape, Cn-shell is a normalization term for the calculation of the form factor, P is the

pitch, and Rm and m are the radius and the electron density of the mth shell, respectively. In the above

equations o is the electron density of the solvent solvent. 0 = 1 and n = 2 for n≥1, n indicates the number of

layer line. is the angle by which one helix is rotated with respect to the second one. The rotation angle

reflects the degree of opening of a tape. Jn(x) is the Bessel function of the first kind and order n. Although

written as an infinite series, only layer lines excited at a certain q-value satisfying the condition: n ≤ Pq/2

contribute to the sum (all other terms are zero).

This can be simplified for our system (Nl-shells mirrored to the central one, as drawn in Fig. 3c for three

layers)

2m

m

DRR

(6)

for mNl


Nano Res.

2

12

mN

ml

DRR

(7)

for m>Nl

R is the mean radius, measuring the distance from the helix center to the center of the innermost shell, Dm is

the thickness of the mth layer, running from m=1 to m=Nl. In this way Cn-shell can be replaced by Cn-layer and Gshell

n by Glayer n, given by:

2

2

2

11

212,,,

m

m

l

n

DR

DR

n

N

mmmmmlayer

Pq

nqrrdrJqPRG

(8)

and

lN

m

mmmmlayern

DR

DRC

1

221 )

2()

2()(

(9)

For the single helical tape the angle by which one tape is rotated with respect to the second one is set to zero.

1.3 Form factor of a torus and its derivatives

A very similar peptide (GAVILRR) was reported to form donut-like structures in solution [6]. Therefore in a

first approach we tried to fit the scattering curve of the 75 mM GAAVILRR solution with various torus-like

models. The fitting functions were developed based on the work done by Takeshi Kawaguchi [7]. According

to Kawaguchi, the scattering from donut-structures arises from randomly oriented objects where the

rotational average of the form factor has to be performed. In the special case of donut-like structures this

average simplifies to:

2/

0

22 ),()sin()(

qFdqF

(10)

with being the angle between the scattering vector q and the vertical axis.


Nano Res.

The form factors F(q,) of the various models are:

a) Torus [7]

)cos(

))cos(sin()sin(4),( 0

q

qqrJrdrqF

ab

ab

(11)

with =[a2-(r-b)2]1/2, b stands for the radius of the torus, a is its cross-section radius as shown in Fig. S3. J0(x) is

the Bessel function of zero order.

b) Hollow torus [7]

),(),(),( 12 qFqFqF (12)

with the corresponding use of an inner torus cross-section radius a1 and an outer cross section radius a2.

c) Torus with elliptical cross section [7]

The form factor of a torus with an elliptical cross-section is derived from equation (11) with

=(c/a)[a2-(r-b)2]1/2, where c and a are the half-axes of the elliptical torus cross-section.

d) Hollow torus with elliptical cross section

),(),(),( 12 qFqFqF (13)

with the half-axes of the inner and outer torus cross-section being c1,a1 and c2,a2.

e) Stack of tori with elliptical cross section

),(),()sin()(2/

0

22

qSqFdqF

(14)

)cos(

))cos(sin()sin(4),( 0

q

qqrJrdrqF

ab

ab

(15)


Nano Res.

lN

kll dqkkNNqS

1

))cos(cos()(),(

(16)

with Nl being the number of stacked tori, d is the distance between the tori (d=2a+), the other parameters are

as in (c)

f) Stack of hollow tori with elliptical cross section

),(),(),( 12 qFqFqF (17)

derived in the same way as in (e), but with a modified form factor amplitude as in equation (13)

Figure S3 Torus model fitting. Scattering curve (black) of the 75 mM GAAVILRR sample with fitting approaches of various torus-like

models (red curves).


Nano Res.

Table S1 Derived fitting parameters of the torus-like model fits.

torus core-shell elliptical

cross-section

elliptical

cross-section,

core-shell

elliptical

cross-section,

stacked

elliptical cross-section,

core-shell, stacked

back 0.021 0.019 0.020 0.020 0.020 0.020

I0 (a.u.) 4.9e-6 3.6e-7 5.4e-7 1.2e-7 8.5e-7 6.5e-8

b (nm) 6.7 6.2 8.9 8.9 7.4 8.8

a (nm) 2.7 - 2.3 - 5.3 -

a1 (nm) - 2.1 - 2.3 - 2.1

a2 (nm) - 5.6 - 6.6 - 5.7

ratio - 6.3 - 8.4 - 6.9

- - 0.70 0.78 0.44 0.94

Nl - - - - 1 1

(nm) - - - - 0.14 0.59

Reduced,

weighted chi2 327.1 13.9 201.3 26.1 68.2 27.8


Nano Res.

1.4 Fitting of the 75 mM GAAVILRR sample with a twisted ribbon form factor

sample: 75 mM GAAVILRR

fitting function: twisted ribbon [4]

Figure S4 Fitting of a twisted ribbon model. Scattering curve (black) of the 75 mM GAAVILRR sample and the twisted ribbon model

fit (red curve). The number of points representing the experimentally measured scattering curve has been reduced from 490 to 125

points to the benefit of calculating time.

Table S2 Derived twisted ribbon fitting parameters.

back R P ribbon I0 Rs chi2

(nm) (nm) (e/Å3) (deg) (a.u.) (nm)

0.0011 7.7 45 1.0 227 4.3e-6 0.089 1.2 129


Nano Res.

1.5 Fitting of the 75 mM GAAVILRR sample with the helix form factor

In this approach we used the same helix model as used for the intermediate region (60 mM sample).

Compared to the ribbon and torus-models, the fit could be considerably improved. Nevertheless, in the low

q-range there is a strong deviation from the experimental scattering curve, which prompted us to further

improve the model.


fitting function: 3-layered single helical tape

Figure S5 Helix model fitting. Scattering curve (black) of the 75 mM GAAVILRR sample and the 3-layered helical model fit (red

curve).

Table S3 Derived helix fitting parameters.

back R D3 D2 D1 P sol 3 2 1 I0 Rs chi2

(nm) (nm) (nm) (nm) (deg) (nm) (e/Å3) (e/Å3) (e/Å3) (e/Å3) (a.u.) (nm)

0.0018 5.6 10.2 3.3 1.0 225 21 0.335 0.5 1.8 2.7 4.1e-6 0.14 1.2 3.7


Nano Res.

1.6 Fitting of the 75 mM GAAVILRR sample with the double helix form factor


fitting function: 3-layered double helical tape

Table S4 Derived double helix fitting parameters.

back R D3 D2 D1 P φ sol 3 2 1 I0 Rs chi2

(nm) (nm) (nm) (nm) (deg) (nm) (deg) (e/Å3) (e/Å3) (e/Å3) (e/Å3) (a.u.) (nm)

0.0011 6.0 12.3 3.3 1.0 69 58 119 0.335 0.5 0.8 1.4 4.0e-6 0.10 1.1 2.7


Nano Res.

1.7 Fitting of the intermediate concentration regime


fitting functions: 3-layered single helical tape & 3-layered double helical tape

Figure S6 Intermediate concentration regime fitting. The 60 mM concentration was fitted with 2 different models, either a 3-layered

helical model (a) or a 3-layered double helical model (b), both in combination with the theoretical form factor of the monomer. Both fits

are equally good, regarding their reduced, weighted chi2 value (see ESM Table S5).

Table S5 Derived helix & double helix fitting parameters.

back R D3 D2 = D1 P φ sol 3 1 = 2 I0 Rs chi2

(nm) (nm) (nm) (deg) (nm) (deg) (e/Å3) (e/Å3) (e/Å3) (a.u.) (nm)

helix 0.0011 7.8 9.0 1.0 189 44 - 0.335 0.5 2.9 1.5e-6 0.07 1.5 5.9

double

helix 0.0011 5.6 8.8 1.0 143 49 76 0.335 0.5 2.2 1.5e-6 0.07 1.5 6.5

When fitting the scattering curve of the intermediate region, both, the single and the double helical model,

provide good and comparable results, but none of the two gave a perfect fit. We attributed this to the

following facts: First, SAXS is able to determine the cross-section of extended structures in high detail. The

underlying parameters which define the cross-section of the helix and the double helix are identical (for a

description of the differences between the two models, please refer to the main article, Section 2.3 “Model

development to fit the scattering curve of the highest peptide concentration”, and Methods Section 4.3

“Formal description of a helix and a double helix – fitting parameters”). Second, Cryo-TEM showed a

heterogeneous population of structures regarding their long range order in the investigated sample.

Interestingly, their widths are highly homogeneous: irrespective whether they are extended tapes, double

tapes or double helices – one tape’s width (and as a conclusion its cross sectional characteristics) remained the

same throughout all structural intermediates. Therefore the presence of the various intermediates in the

sample explains the non-perfect fit of the extended structure to the SAXS data.


Nano Res.

Ensemble Analysis

Ensemble fitting was applied to determine the ratio of single and double helices in the 60 mM and 75 mM

samples.

Figure S7 Ensemble analysis. The scattering curves of the 60 mM and 75 mM concentrations were fitted with 2 different models, either

a 3-layered helical model (blue) or a 3-layered double helical model (green). (a) shows the single and double helical fits for the 60 mM

sample, (b) represents the experimental scattering curve (black) with the ensemble analysis of the two models (red), which allows to

determine the ratio of the individual components. The 60 mM sample contains approximately 65 % single helices and 35 % double

helices. The two fitting curves for the 75 mM sample are shown in (c). Ensemble analysis (d) reveals a distribution of 24 % single

helices and 76 % double helices.

Table S6 Ensemble analysis.

60 mM 75 mM

single helix (%) 65 24

double helix (%) 35 76


Nano Res.

Influence of shearing forces when filling the capillaries (60 mM sample)

We suspected that the shearing force which is going along with the filling of the SAXS capillary might change

the inherent self-assembled structure of the peptide. To test this effect we left the before measured 60 mM

sample (Fig. S8, lower curve, green fit) undisturbed in the capillary and repeated the measurement after 2

hours. In fact, the SAXS pattern has changed considerably (Fig. S8, upper curve, blue fit). Both curves were

fitted by combining the theoretical form factor of the peptide monomer with a 3-layered single helical tape.

The results only differ in a clearly decreased pitch size for the second sample, the cross-sectional

characteristics and the tape width remained the same. These results hint at a condensation process. It is highly

interesting that fully developed structures undergo that change within a few hours. We suppose that the shear

stress temporarily forced the structures into the elongated form and that under restored equilibrium

conditions they recovered elastically their coiled form. However, this temporal change highlights the dynamic

nature of the molecules and might even mirror the initial steps of a coiling process.

Table S7 Helix fitting parameters at 2 different time points

back R D3 D2 = D1 P sol 3 2 = 1 I0 Rs chi2

sample (nm) (nm) (nm) (deg) (nm) (e/Å3) (e/Å3) (e/Å3) (a.u.) (nm)

immediately

(green) 0.0011 7.2 10.0 1.0 183 62* 0.335 0.5 3.9 1.7e-6 0.07 1.5 6.4

after 2 h

(blue) 0.0011 7.8 9.0 1.0 189 44* 0.335 0.5 2.9 1.5e-6 0.07 1.5 5.9

*experimental constraint: Pmax = 2/qmin = 62 nm

Figure S8 Influence of shearing forces. The 60 mM concentration was measured twice: directly after filling of the capillary (lower

curve, green fit) and after leaving the sample undisturbed in the capillary for 2 hours (upper curve, blue fit). The optimal fits were

achieved with a 3-layered helical model in combination with the theoretical form factor of the monomer (see also ESM Table S7).


Nano Res.

Time-dependent evolution of the SAXS curve (75 mM)

To test, if the double helix model represents an intermediate or the thermodynamically endpoint structure, we

performed SAXS measurements after 1, 4 and 11 weeks of aging. As seen in Fig. S9, all curves show the same

scattering pattern, regardless their aging time. We therefore conclude that the double helices are the final

structure. This result perfectly agrees with Cryo-TEM measurements, where we did not observe structural

changes over the investigated period of 11 weeks either (see main article, Fig. 5).

Figure S9 Time-dependent evolution of the SAXS signal. Scattering curves of 75 mM GAAVILRR show identical characteristics after

1 week (red), 4 weeks (blue) and 11 weeks (green) of incubation.


Nano Res.

Comparison of various TEM methods for double helix visualization

We used 3 different transmission electron microscopy techniques (freeze-fracture, negative-staining and

cryogenic TEM) to visualize the morphology of the assembled structures at 75 mM after 1 week of aging (Fig.

S10). All 3 methods showed the double helical arrangement and provided consistent size information.

Freeze-fracture is the only technique that gives an impression on the depth of the assemblies (Fig. S10a).

Unfortunately we do not see the entire extension of the double helix, we rather see small fragments thereof. In

return, these fractions are very well resolved and the individual tapes are clearly visible. Interestingly, the

tapes rise quite steeply. This is in agreement with the calculated helical twist angle of ~ 60° from SAXS

experiments. The total diameters of the double helices range between 20-25 nm. Negative staining TEM

images highlight the double taped arrangement and illustrate the coiling of the tapes (Fig. S10b). The lengths

of the double helices are well over some hundreds of nanometers, whereas their widths are generally smaller

(10-15 nm) compared to SAXS. This could be due to shrinking during TEM sample preparation. Associated

with the double helices are large vesicles, with diameters of around 25 nm. They are believed to develop

during the sample preparation process when the solution is dried on the grid. These structures are not present

in the native solution and were neither observed with SAXS, nor with other TEM techniques. We conclude

that negative-staining TEM nicely visualizes the coiling of tapes, but that this technique is more susceptible to

produce artifacts than the others. Cryo-TEM is the least invasive method and preserves the structures at

cryogenic conditions. The double-helix-like assemblies are clearly visible and exhibit lengths of several

hundreds of nanometers, widths of 15-20 nm and periodic intersection points every ~ 80 nm (Fig. S10c).

Figure S10 Comparison of 3 different TEM techniques. In order to directly visualize the double-helix-like morphology, the same

sample (75 mM GAAVILRR, 1 week of aging) was investigated by (a) freeze-fracture TEM, (b) negative-staining TEM and (c)

cryo-TEM.

Experimental

Freeze-fracture Transmission Electron Microscopy (TEM)

The peptide-hydrogel samples were cut to cubes with an edge length of about 1.5 mm and were placed on

3-mm gold specimen carriers, pretreated with a thin layer of 30 % glycerol. The samples were frozen rapidly

in liquid propane (-140 °C to -150 °C) by using a cryo-fixation device (KF80, Reichert-Jung, Vienna, Austria).


Nano Res.

The frozen samples were stored in liquid nitrogen, transferred onto the cold stage of the freeze-fracture

system BAF 400 D (Balzers, Liechtenstein) and fractured at -100 °C in vacuum. The sample surface was etched

for 30 min. by placing the cooled knife (-196 °C) directly above the samples. Replicas were obtained by

shadowing the samples with platinum (20 nm) at a 30° angle and carbon (200 nm) applied from overhead. The

film-thickness was controlled by a quartz crystal thin-film monitor. In order to degrade the peptides, the

replicas were transferred to 10 % sodium hypochloride for 3 h, followed by storage in 50 % sodium hydroxide

overnight. The replicas were washed several times in double-distilled water and placed onto a copper grid.

Visualization of the samples was accomplished using a Fei Tecnai G² 20 transmission electron microscope

(Eindhoven, the Netherlands) operating at an acceleration voltage of 120 kV. Digital images were made using

a Gatan US1000 CCD camera at 2K x 2K resolution.

Negative-staining Transmission Electron Microscopy (TEM)

Samples were prepared by dissolving the lyophilized peptide in double distilled water to obtain the desired

concentrations and aged for 1 week at 4 °C under argon atmosphere. 5 µ l of each sample were adsorbed onto a

glow-discharged carbon-coated copper grid and allowed to settle for 1 minute. Excess fluid was carefully

removed with filter paper and immediately replaced by 5 µ l of a 2 % (w/v) uranyl acetate staining solution,

which was centrifuged for 6 minutes at 10000 rpm prior to usage in order to precipitate potentially present

aggregates. The uranyl acetate solution was allowed to settle for 30 seconds, blotted and replaced by another 5

µ l of fresh solution. After 30 seconds of incubation the staining solution was removed and the sample was

air-dried.

Visualization of the samples was achieved by using a Fei Tecnai G² 20 transmission electron microscope

(Eindhoven, The Netherlands) operating at an acceleration voltage of 120 kV. Digital images were made using

a Gatan US1000 CCD camera at 2K x 2K resolution.

Cryogenic Transmission Electron Microscopy (Cryo-TEM)

See main article (Methods Section)

References

[1] Yang, S.; Zhang, S. Self-assembling behavior of designer lipid-like peptides. Supramol. Chem. 2006, 18,

289-396.

[2] Schneidman-Duhovny, D.; Hammel, M.; Tainer, J. A.; Sali, A. Accurate SAXS profile computation and

its assessment by contrast variation experiments. Biophys. J. 2013, 105, 962-974.

Schneidman-Duhovny, D.; Hammel, M.; Sali, A. FoXS: a web server for rapid computation and fitting

of SAXS profiles. Nucleic Acids Res. 2010, 38, W540-W544.

[3] Pontoni, D.; Finet, S.; Narayanan, T.; Rennie, A. R. Interactions and kinetic arrest in an adhesive

hard-sphere colloidal system. J. Chem. Phys. 2003, 119, 6157.


Nano Res.

[4] Teixeira, C. V.; Amenitsch, H.; Fukushima, T.; Hill, J. P.; Jin, W.; Aida, T.; Hotokka, M.; Lindén, M.

Form factor of an N-layered helical tape and its application to nanotube formation of

hexa-peri-hexabenzocoronene-based molecules. J. Appl. Crystallogr. 2010, 43, 850-857.

[5] Pringle, O. A.; Schmidt, P. W. Small-angle X-ray scattering from helical macromolecules. J. Appl.

Crystallogr. 1971, 4, 290-293.

[6] Khoe, U.; Yang, Y. L.; Zhang, S. G. Self-Assembly of Nanodonut Structure from a Cone-Shaped

Designer Lipid-like Peptide Surfactant. Langmuir 2009, 25, 4111-4114.

[7] Kawaguchi, T. Radii of gyration and scattering functions of a torus and its derivatives. J. Appl. Crystallogr.

2001, 34, 580-584.

Documents

Tracking morphologies at the nanoscale: self-assembly of ... · Self-assembling amphiphilic designer peptides have been explored as building blocks for the development of controllable,