Skip to content
Publicly Available Published by De Gruyter September 26, 2017

Co-assembly of helical β3-peptides: a self-assembled analogue of a statistical copolymer

  • Claire Buchanan , Christopher J. Garvey , Patrick Perlmutter and Adam Mechler EMAIL logo

Abstract

Unnatural peptide self-assembly offers the means to design hierarchical nanostructures of controlled geometries, chemical function and physical properties. N-acyl β3 peptides, where all residues are unnatural amino acids, are able to form helical fibrous structures by a head-to-tail assembly of helical monomers, extending the helix via a three point supramolecular hydrogen bonding motif. These helical nanorods were shown to be stable under a wide range of physical conditions, offering a self-assembled analogue of polymeric fibres. Hitherto the self-assembly has only been demonstrated between identical monomers; however the self-assembly motif is sequence-independent, offering the possibility of hetero-assembly of different peptide monomers. Here we present a proof of principle study of head-to-tail co-assembly of two different helical unnatural peptides Ac-β3[WELWEL] and Ac-β3[LIA], where the letters denote the β3 analogues of natural amino acids. By atomic force microscopy imaging it was demonstrated that the homo-assembly and co-assembly of these peptides yield characteristically different structures. Synchrotron small angle X-ray scattering experiments have confirmed the presence of the fibres in the solution and the averaged diameters from modelled data correlate well to the results of AFM imaging. Hence, there is evidence of co-assembly of the fibrous superstructures; given that different monomers may be used to introduce variations into chemical and physical properties, the results demonstrate a self-assembled analogue of a statistical co-polymer that can be used in designing complex functional nanomaterials.

Introduction

Self-assembly is a powerful way to build nanostructures of defined geometries [1], [2], [3], [4]. Strategies to encode the target structure in the building blocks range from simple amphiphiles that yield micellar assemblies [5], [6], [7], [8] to supramolecular binding motifs that implement specific and geometrically defined docking sites [9], [10]. The latter requires building blocks with inherent, stable geometries that can be chemically modified without altering the core structure [2]. Taking inspiration from the structural complexity of proteins, molecular designs based on polypeptides are frequently considered for this purpose [2], [4].

Oligopeptide-based self-assembling materials have been proposed for applications mostly in medicine, such as tissue scaffolding and drug delivery [11], [12]. Peptides consisting of natural α-amino acids, while structurally and functionally suitable for the design of hierarchical structures and smart nano-materials, are considered chemically unstable with weak secondary structures. Peptides made entirely from “unnatural” β3 amino acids offer stability and versatility far surpassing α-peptides [13], [14]. As opposed to the “binary” α-helix and β-sheet conformations of natural oligopeptides, β-peptides exhibit a range of – mostly helical – secondary structures [15]. Of particular interest are β3-peptides that fold into helices consisting 14-membered loops through hydrogen bonding between C=O of residues i and the NH of the i-3 residue [16], [17]. Accordingly, these so-called 14helices exhibit a pitch of 3.0–3.1 amino acids per turn that is a substantial advantage over the α-helix that exhibits 3.6 residues per turn [16], [18], [19]. In the 14-helix the side chains align along the helix, providing a geometrically determined structure that is an ideal scaffold for a self-assembling building block [18], [20].

It was demonstrated previously that N-terminally acetylated β3-amino acid sequences self-assemble into helical nanorods via a supramolecular 3-point hydrogen-bonding motif [21]. These nanorods then further assemble into hierarchical superstructures defined by the surface morphology of the constituent peptides and second order (non-covalent) interactions between the helices [21], [22], [23]. Considering that the core nanorods are truly one dimensional materials with sub-nm diameters and lengths in the μm range, potentially reaching millimetres, the self-assembly of N-acetylated β3-peptides yields fibrous materials that are analogous to covalently linked polymers. Thus, by the analogy with co-polymers, here we explore the possibility of co-assembling N-acyl β3-peptides of different chemical composition to alter the properties of the resulting superstructures.

Materials and methods

Materials

N-acetylated β3 peptides Ac-β3[WELWEL] and Ac-β3[LIA] (Fig. 1) where the letters W, E, L, I, A denote the β3 analogues of the respective α amino acids tryptophan, glutamic acid, leucine, isoleucine and alanine were synthesised using routine solid phase synthesis protocol as previously described [21]. Solutions for both AFM and SAXS analysis were prepared by dissolving peptides in LC-MS grade Isopropanol (Sigma Aldrich Pty. Ltd. Castle Hill NSW 2154 Australia) at 1 mg/mL concentration. For the mixed solution, 0.5 mg of each peptide was used, resulting in a 1:2.5 Ac-β3[WELWEL]:Ac-β3[LIA] molar ratio. Due to the slow exchange rate identified in previous work, solutions were aged for 9 months before analysis.

Fig. 1: 
            Structural representation of Ac-β3[LIA] and Ac-β3[WELWEL]. The top row shows the primary structure, while the bottom row shows 3D stick model with mesh surface structure representations of nanorods, looking down the barrel of the helix (left) and across the helix (right).
Fig. 1:

Structural representation of Ac-β3[LIA] and Ac-β3[WELWEL]. The top row shows the primary structure, while the bottom row shows 3D stick model with mesh surface structure representations of nanorods, looking down the barrel of the helix (left) and across the helix (right).

Atomic force microscopy (AFM)

AFM sample preparation was performed by pipetting a dilute suspension of the self-assembled material onto atomically smooth substrate surfaces and evaporating the solvent, i.e. a process analogous to polymer drop casting, albeit using much more diluted samples. Following this process 1 μL of the aged Ac-β3[WELWEL] and the mixed peptide solutions were deposited on freshly cleaved graphite and dried overnight at room temperature; Acβ3[LIA] sample was deposited on freshly cleaved mica as the different surface chemistry helped with AFM imaging. Semi-contact imaging was performed using an Ntegra AFM platform (NT-MDT, Russia) with silicon probes with 200–400 kHz resonance frequency (~5–30 N/m spring constant) and a nominal 10 nm apex radius. Pixel resolution was 512×512 and the scan rate was set between 0.35 and 0.6 Hz. Gwyddion software was used for all post-imaging processing.

Small angle X-ray scattering (SAXS)

All SAXS experiments were performed at the Australian Synchrotron. Solutions were placed in the SAXS beamline in 1 mm quartz capillaries to record the resulting scattering patterns. Beam energy was set to 11 keV and the detector distance was 2.7 m. The isotropic data was reduced on-site using scatterBrain software, using an isopropanol-filled capillary as a blank. Model fitting was performed using SasView software to determine diameters of structures in the solution phase. Two distributions were visible on each profile, and so samples were fitted to a bimodal cylindrical model according to eq. 1. f(q) is given by equation 2, where V is the volume of the cylinder, R is the radius of the cylinder, L is the length of the cylinder, J1 is the Bessel function, and α is the angle between the q-vector and the axis of the cylinder.

As two curves were visible on each profile, a summation of eq. 2 was used to make a bimodal cylindrical model.

(1) P ( q , α ) = Scale V 1 f 1 2 ( α ) + Scale V 2 f 2 2 ( α ) + B a c k g r o u n d

(2) f ( q , α ) = ( 2 ( Δ ρ ) V s i n ( q L  cos ( α 2 ) ) q L  cos ( α 2 ) ) × ( J 1  ( q R  cos ( α )) q R  cos ( α ) )

Results and discussion

N-acylated β3 peptides self-assemble head-to-tail into extended “nanorods” following a three point hydrogen bonding motif previously described [21]. The two peptides in this study, Acβ3[LIA] and Ac-β3[WELWEL] are comprised of distinctly different residues, resulting in different chemical characteristics of the core nanorod (Fig. 1). Ac-β3[LIA] self-assembly yields a pseudo-helical tube with three aliphatic faces, while the terminal carboxyl forms a fourth polar face; Ac-β3[WELWEL] consist of three chemically distinct faces (polar, aromatic and aliphatic), while the terminal carboxyl group is sterically less free and is only present once per two rises of the helix. Hence, the nanorods formed by the neat peptides are expected to associate/bundle via different physical interactions, resulting in distinct superstructures, while co-assembly is expected to yield a hybrid structure.

The self-assembly motif is identical in the two peptides and hence the energy difference between homo- and hetero-assembly is expected to be minimal but not identical due to van der Waals contribution of the side chains. In equilibrium, monomers can detach from the end of a chain and re-attach to another, thus the final structure is defined by the lowest energy minimum. The result is analogous to a statistical co-polymer, albeit where both energetics and concentrations have an effect on the final sequence.

Isopropanol was the solvent used for all samples presented here; its low polarity and dielectric constant suppresses hydrophobic effects and promotes van der Waals interactions. Samples of neat Ac-β3[LIA] and Ac-β3[WELWEL] assemblies, as well as the mixture of the two, were deposited on freshly cleaved mica and graphite surfaces, respectively, and imaged with AFM (Fig. 2). All three samples showed characteristic differences in both size and organisation of the supramolecular structures. Ac-β3[LIA] geometries have been described before and are used as a reference here [23]. The sample shows large dendritic bundled structures comprised of fibres ranging from 10 to 40 nm (Fig. 2A–D). In contrast, while occasional fibres of ~20 nm thickness tapering to 3 nm were present in the Ac-β3[WELWEL] samples, the majority of the peptide was found as a thin layer of branching structures (2–4 nm tall) on the graphite surface (Fig. 2E–H), with little evidence of lateral interactions.

Fig. 2: 
          AFM imagaes of neat peptides deposited from isopropanol onto a smooth graphite surface (A–C) Ac-β3[LIA] sample, (D) height profile of line on B; (F–H) Ac-β3[WELWEL]; (E) height profiles of lines indicated on F and G.
Fig. 2:

AFM imagaes of neat peptides deposited from isopropanol onto a smooth graphite surface (A–C) Ac-β3[LIA] sample, (D) height profile of line on B; (F–H) Ac-β3[WELWEL]; (E) height profiles of lines indicated on F and G.

The diverse nature of the residues present in Ac-β3[WELWEL] makes orientation highly important for any lateral interactions to occur between the helical nanorods, limiting both van der Waals interactions between leucine residues and pi-stacking interactions between tryptophan residues. Carboxy dimerization between the glutamic acid residues and/or the C-terminal carboxyl groups is also possible, however the solvent competes for the hydrogen bonding sites and thus it is expected that such an effect would only become dominant after drying the sample. The mostly unorganised nature of the structure indicates inherently weak inter-helix interactions.

The combined Ac-β3[LIA] and Ac-β3[WELWEL] sample showed distinct differences from both the neat peptide samples, with areas of aligned fibres of 3 nm diameter covered by large aggregation structures (Fig. 3A–D). When depositing the mixture on graphite, air bubbles formed in the solution on the surface of the graphite, leading to areas of no sample coverage surrounded by a ring of high peptide concentration. Scanning the higher concentration edges showed large, mostly disordered clusters, from which individual fibres (5–2 nm diameter) protrude (Fig. 3E–H). Close examination of the fibres (Fig. 3E, H) shows a regular pattern of 5 nm height at every 6 nm in length. The periodicity is indicative of two or more fibrils twisted together, and reveals the hierarchical nature of peptide self-assembly.

Fig. 3: 
          AFM images of combined Ac-β3[WELWEL] and Ac-β3[LIA] deposited from isopropanol onto a smooth graphite surface. (A–C) Images from even solution coverage, (D) height profile of lines on B, (F–H) images from highly concentrated area surround due to air bubble formation, (E) height profile of lines on G. C and H are three-dimensional rendering of the cutout in B and G, respectively.
Fig. 3:

AFM images of combined Ac-β3[WELWEL] and Ac-β3[LIA] deposited from isopropanol onto a smooth graphite surface. (A–C) Images from even solution coverage, (D) height profile of lines on B, (F–H) images from highly concentrated area surround due to air bubble formation, (E) height profile of lines on G. C and H are three-dimensional rendering of the cutout in B and G, respectively.

The multiple structures formed from the mixed peptide suggest that the co-assembly allows for some variation in the order of the monomer attachment to the core fibres, analogously to a statistical co-polymer. The thin aligned sample shows similarities to the neat Ac-β3[WELWEL] sample, but with a higher degree of orientation, rigidity and inter-helix interactions to yield straight, aligned fibres. The large fibres seen in Fig. 3G weakly resemble the Ac-β3[LIA] sample in that it contains long fibrous bundles, however instead of the dendritic structures, straight “braided” threads are observed that are unique to the co-assembled sample. Hence, the two dominant structures seen in the co-assembled sample appear to have been derived from the structures formed by the neat peptides but with modified properties, presenting a striking analogy to covalently linked co-polymers.

In order to probe the characteristic differences in self-assembly, the solution phase samples were characterised by synchrotron SAXS. Each sample yielded a unique scattering profile (Fig. 4A). The scattering profile of the combined sample is markedly different from the neat peptides. Each profile showed two curves across the q-range used, and so a bimodal cylindrical model was used to fit each curve (Fig. 4B–D) to get a cross sectional radius of the fibres present (Fig. 4, table). The lengths of the structures in each profile were estimated using AFM data, and were restricted to ±30% of that estimate. The values in the table of Fig. 4 show the numerical results. Due to the hierarchical nature of peptide self-assembly, the solutions contain structures of different, distinct diameters; however there is variation in the lengths of these structures, as seen in the AFM images. This complicates analysis, skewing detection towards the larger structures within the SAXS range of detection. As such, all numbers presented here are for the dominant fibres seen within each measured q range region.

Fig. 4: 
          SAXS profiles of solutions of neat and mixed Ac-β3[LIA] Ac-β3[WELWEL] peptides with bimodal cylindrical fitting of Ac-β3[LIA], Ac-β3[WELWEL] and the combined fibres, respectively; the table radii of cylinders fitted, along with the relative contribution from each population to the overall curve.
Fig. 4:

SAXS profiles of solutions of neat and mixed Ac-β3[LIA] Ac-β3[WELWEL] peptides with bimodal cylindrical fitting of Ac-β3[LIA], Ac-β3[WELWEL] and the combined fibres, respectively; the table radii of cylinders fitted, along with the relative contribution from each population to the overall curve.

The fitting of the SAXS data of the Ac-β3[LIA] sample identified radii of 7.21 and 18.4 nm (Fig. 4, table). At the lowest q, the fitting of the profile is poor, indicating the beginning of another length-scale outside the q-range. This may correspond to the larger bundles seen in the AFM images. These values provide an excellent match to the sizes found by AFM, indicating that the hierarchical self-assembly takes place in a deterministic manner, without substantial variations. The Ac-β3[WELWEL] sample showed structures with radii of 8.81 nm and 20.29 nm, while the combined peptide sample has structures with radii of 4.01 nm and 17.97 nm (Fig. 4, table). In the latter two cases, the small peptides found by AFM (Figs. 2C and 3C) are not detected in the SAXS profile, suggesting that the remaining population is of low volume fraction and monomeric and thus contributes weakly to the SAXS curve and is barely resolvable above the background signal. Hence, the small oligomeric structures seen in AFM are likely formed during the deposition/drying process (Fig. 3).

The differences found in the SAXS profiles of all three samples, specifically the small structures seen in the high q-range, also confirm the occurrence of co-assembly. The similarities in the average radii at high-q of both the neat peptides (approx. 8 nm), make the difference found in the combined sample especially notable; in a system of no co-assembly, the mixed sample would be expected to have discrete structures of Ac-β3[LIA] and Ac-β3[WELWEL] assembly, and as such would be expected to have a similar average radius value. The smaller value of 4.01 nm indicates that the two peptides are interacting, which ultimately affects the supramolecular structures formed.

Conclusions

We have demonstrated that Ac-β3-peptides of different sequences are capable of interacting with each other to co-assemble in a way that is analogous to co-polymers. Both AFM and SAXS analysis of the supramolecular structures formed from Ac-β3-peptides show characteristic differences between the supramolecular structures formed from neat and combined peptide samples. AFM imaging of neat and combined peptides show very distinct and unique supramolecular morphologies of the deposited samples, indicating at least two new supramolecular structures: large, flexible rope-like structures and a 2-dimensional array of small aligned fibres. Fitting the SAXS profiles of the samples with bi-modal cylindrical models revealed similar averaged radii for the neat peptides and a smaller radius for the mixed sample, confirming that co-assembly takes place in the solution phase.


Article note

A collection of invited papers based on presentations at the 3rd International Conference on Bioinspired and Biobased Chemistry and Materials: Nature Inspires Creativity Engineers (NICE-2016), Nice, France, 16–19 October 2016.


Acknowledgements

The authors acknowledge Khadeeja Hussein, Ketav Kulkarni and Mark Del Borgo for preparing the peptides, as well as Marie-Isabel Aguilar for her inspirational work on beta peptide design. Rania S. Seoudi has kindly contributed AFM images of Ac-β3[LIA]. Small angle X-ray scattering experiments were undertaken on the SAXS/WAXS beamline of the Australian Synchrotron. The authors acknowledge the beamline staff, Tim Ryan, Nigel Kirby, Stephen Mudie and Adrain Hawley.

References

[1] S. Mann. Self-assembly and transformation of hybrid nano-objects and nanostructures under equilibrium and non-equilibrium conditions. Nat. Mater.8, 781 (2009).10.1038/nmat2496Search in Google Scholar PubMed

[2] A. L. Boyle, E. H. C. Bromley, G. J. Bartlett, R. B. Sessions, T. H. Sharp, C. L. Williams, P. M. G. Curmi, N. R. Forde, H. Linke, D. N. Woolfson. Squaring the circle in peptide assembly: from fibers to discrete nanostructures by de novo design. J. Am. Chem. Soc.134, 15457 (2012).10.1021/ja3053943Search in Google Scholar PubMed

[3] D. Papapostolou, A. M. Smith, E. D. T. Atkins, S. J. Oliver, M. G. Ryadnov, L. C. Serpell, D. N. Woolfson. Engineering nanoscale order into a designed protein fiber. Proc. Natl. Acad. Sci. USA104, 10853 (2007).10.1073/pnas.0700801104Search in Google Scholar PubMed PubMed Central

[4] M. G. Ryadnov, D. N. Woolfson. MaP peptides: programming the self-assembly of peptide-based mesoscopic matrices. J. Am. Chem. Soc.127, 12407 (2005).10.1021/ja052972iSearch in Google Scholar PubMed

[5] H. G. Cui, M. J. Webber, S. I. Stupp. Self-assembly of peptide amphiphiles: from molecules to nanostructures to biomaterials. Biopolymers94, 1 (2010).10.1002/bip.21328Search in Google Scholar PubMed PubMed Central

[6] X. D. Xu, Y. Jin, Y. Liu, X. Z. Zhang, R. X. Zhuo. Self-assembly behavior of peptide amphiphiles (PAs) with different length of hydrophobic alkyl tails. Coll. Surf. B81, 329 (2010).10.1016/j.colsurfb.2010.07.027Search in Google Scholar PubMed

[7] X. Z. Yan, S. J. Li, T. R. Cook, X. F. Ji, Y. Yao, J. B. Pollock, Y. H. Shi, G. C. Yu, J. Y. Li, F. H. Huang, P. J. Stang. Hierarchical self-assembly: well-defined supramolecular nanostructures and metallohydrogels via amphiphilic discrete organoplatinum(II) metallacycles. J. Am. Chem. Soc.135, 14036 (2013).10.1021/ja406877bSearch in Google Scholar PubMed

[8] Z. Wu, Y. Yan, J. B. Huang. Advanced molecular self-assemblies facilitated by simple molecules. Langmuir30, 14375 (2014).10.1021/la501361fSearch in Google Scholar PubMed

[9] S. I. Stupp, R. H. Zha, L. C. Palmer, H. G. Cui, R. Bitton. Self-assembly of biomolecular soft matter. Farad. Discuss.166, 9 (2013).10.1039/c3fd00120bSearch in Google Scholar PubMed PubMed Central

[10] T. Aida, E. W. Meijer, S. I. Stupp. Functional supramolecular polymers. Science335, 813 (2012).10.1126/science.1205962Search in Google Scholar PubMed PubMed Central

[11] S. G. Zhang. Emerging biological materials through molecular self-assembly. Biotechnol. Adv.20, 321 (2002).10.1016/S0734-9750(02)00026-5Search in Google Scholar

[12] S. G. Zhang, D. M. Marini, W. Hwang, S. Santoso. Design of nanostructured biological materials through self-assembly of peptides and proteins. Curr. Opin. Chem. Biol.6, 865 (2002).10.1016/S1367-5931(02)00391-5Search in Google Scholar PubMed

[13] T. Beke, C. Somlai, A. Perczel. Toward a rational design of beta-peptide structures. J. Comput. Chem.27, 20 (2006).10.1002/jcc.20299Search in Google Scholar PubMed

[14] C. M. Goodman, S. Choi, S. Shandler, W. F. DeGrado. Foldamers as versatile frameworks for the design and evolution of function. Nat. Chem. Biol.3, 252 (2007).10.1038/nchembio876Search in Google Scholar PubMed PubMed Central

[15] D. Seebach, D. F. Hook, A. Glattli. Helices and other secondary structures of beta- and gamma-peptides. Biopolymers84, 23 (2006).10.1002/bip.20391Search in Google Scholar PubMed

[16] A. M. Bruckner, P. Chakraborty, S. H. Gellman, U. Diederichsen. Molecular architecture with functionalized beta-peptide helices. Angew. Chem. Int. Ed.42, 4395 (2003).10.1002/anie.200351871Search in Google Scholar PubMed

[17] C. L. Pizzey, W. C. Pomerantz, B. J. Sung, V. M. Yuwono, S. H. Gellman, J. D. Hartgerink, A. Yethiraj, N. L. Abbott. Characterization of nanofibers formed by self-assembly of beta-peptide oligomers using small angle X-ray scattering. J. Chem. Phys.129, 8 (2008).10.1063/1.2955745Search in Google Scholar PubMed

[18] R. P. Cheng, S. H. Gellman, W. F. DeGrado. Beta-peptides: from structure to function. Chem. Rev.101, 3219 (2001).10.1021/cr000045iSearch in Google Scholar PubMed

[19] S. A. Hart, A. B. F. Bahadoor, E. E. Matthews, X. Y. J. Qiu, A. Schepartz. Helix macrodipole control of beta(3)-peptide 14-helix stability in water. J. Am. Chem. Soc.125, 4022 (2003).10.1021/ja029868aSearch in Google Scholar PubMed

[20] R. D. Gopalan, M. P. Del Borgo, A. Mechler, P. Perlmutter, M. I. Aguilar. Geometrically precise building blocks: the self-assembly of beta-peptides. J. Chem. Biol.22, 1417 (2015).10.1016/j.chembiol.2015.10.005Search in Google Scholar PubMed

[21] M. P. Del Borgo, A. I. Mechler, D. Traore, C. Forsyth, J. A. Wilce, M. C. J. Wilce, M. I. Aguilar, P. Perlmutter. Supramolecular self-assembly of n-acetyl-capped beta-peptides leads to nano- to macroscale fiber formation. Angew. Chem. Int. Ed.52, 8266 (2013).10.1002/anie.201303175Search in Google Scholar PubMed

[22] R. S. Seoudi, A. Dowda, M. Del Borgo, K. Kulkarni, P. Perlmutter, M.-I. Aguilar, A. Mechler. Amino acid sequence controls the self-assembled superstructure morphology of N-acetylated tri-beta3-peptides. Pure Appl. Chem.87, 1021 (2015).10.1515/pac-2015-0108Search in Google Scholar

[23] R. S. Seoudi, M. P. Del Borgo, K. Kulkarni, P. Perlmutter, M. I. Aguilar, A. Mechler. Supramolecular self-assembly of 14-helical nanorods with tunable linear and dendritic hierarchical morphologies. New J. Chem.39, 3280 (2015).10.1039/C4NJ01926ASearch in Google Scholar

Published Online: 2017-09-26
Published in Print: 2017-11-27

©2017 IUPAC & De Gruyter. This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License. For more information, please visit: http://creativecommons.org/licenses/by-nc-nd/4.0/

Downloaded on 25.4.2024 from https://www.degruyter.com/document/doi/10.1515/pac-2017-0709/html
Scroll to top button