Skip to content
Publicly Available Published by De Gruyter September 8, 2015

Amino acid sequence controls the self-assembled superstructure morphology of N-acetylated tri-β3-peptides

  • Rania S. Seoudi , Annette Dowd , Mark Del Borgo , Ketav Kulkarni , Patrick Perlmutter , Marie-Isabel Aguilar and Adam Mechler EMAIL logo

Abstract

Peptides based on unnatural β3-amino acids offer a versatile platform for the design of self-assembling nanostructures due to the folding stability of the 14-helix and the high symmetry of the side chains inherent in this geometry. We have previously described that N-terminal acetylation (Ac-) forms a supramolecular self-assembly motif that allows β3-peptides to assemble head-to-tail into a helical nanorod which then further bundles into hierarchical superstructures. Here we investigate the effect of the topography of the 14-helical nanorod on lateral self-assembly. Specifically, we report on the variations in the superstructure of three isomeric peptides comprising the same three β3-amino acid residues: β3-leucine (L), β3-isoleucine (I) β3-alanine (A) to give peptides Ac-β3[LIA], Ac-β3[IAL] and Ac-β3[ALI]. AFM imaging shows markedly different superstructures for the three peptides. Well defined synchrotron far-infrared spectra reveal uniform geometries with a high degree of similarity between the isomeric peptides in the amide modes of the 400–650 wavenumber range. Far-IR also confirms that the C-terminal carboxyl group is free in the assemblies, thus it is solvated in the dispersant. Hence, the differences in the superstructures formed by the fibers are defined primarily by van der Waals energy minimization between the varied cross sectional morphologies of the core nanorods.

Introduction

Self-assembly offers an elegant way to design and build nanostructures from specifically-designed small, simple molecular building blocks into a variety of superstructures, from fibers and films to three dimensional nano-architectures [1–8]. Bioinspired materials, such as peptide-based α-helices or β-sheets are often employed due to their well-defined structure and biocompatibility [9–11]. β3-peptide foldamers offer a particularly attractive vehicle for molecular design given that their 14-helical structure contains almost exactly three amino acids per turn, offering a triangular alignment of the side chains along the central helix [3, 4, 12, 13]. We have previously reported the design of a supramolecular self-assembly motif that allows head-to-tail assembly of helical N-acetyl-β3-peptides thereby propagating the 14-helical structure of β3-peptides intermolecularly and yielding a core nanorod held together by three H-bonds per turn [3, 4]. Thus the side chains are in geometrically well-defined positions, without influencing the assembly of the nanorod. We have also demonstrated that the lateral association of these nanorods leads to superstructures of diverse geometries which are strongly influenced by the physicochemical properties of the solvent. While this geometrically defined structure offers important design flexibility in terms of the three functional sites to implement specific binding, the design rules are yet to be established for controlled self-assembly into novel 14-helix-based bio- and nano-materials, particularly in terms of the role of the amino acid side chains in the hierarchical self-assembly process [4].

Characterization of the atomic structural features of supramolecular self-assembling systems is an on-going challenge [2, 14]. Hence the characterization of fibrous supramolecular assemblies often relies on X-ray fiber diffraction [15] or high resolution imaging methods, such as transmission electron microscopy or atomic force microscopy (AFM) [3, 4, 14]. However, it is also possible to indirectly probe the second order interactions if the secondary structure of the fiber-forming peptides and the binding patterns of the self-assembly motif are well defined. Hydrogen bonding of the backbone amide groups in small soluble polypeptides, including inter-molecular H-bonding in small oligomers, may be identified with 1H NMR by the N–H proton chemical shifts [16, 17], while dihedral angles, specific to folding geometries can be estimated from the J coupling of adjacent protons [18]. Molecular vibrations of the terahertz range (far-infrared, far-IR) are highly sensitive to, and thus may be used to reveal structural strains and the chemical environment of specific functional groups [19–21].

The purpose of the present work is to explore the role of cross section morphology, as a control of van der Waals energy minimization, on the superstructure formed by the lateral association of helical nanorods [3] formed through the self-assembly of small helical unnatural peptides. Hence we report on the effect of sequence variation of the same three β3-amino acids on the superstructures formed by N-terminal acetylated tri-β3-peptides Ac-β3[LIA], Ac-β3[IAL] and Ac-β3[ALI]. We used AFM imaging to show the morphological effect of these structural differences. We performed synchrotron far-infrared spectroscopy to ascertain the similarity of the folding of the peptides.

Experimental

Materials

Spectrophotometric grade methanol as well as HPLC grade ethanol and isopropanol were purchased from Sigma-Aldrich Pty. LTD. (Castle Hill, NSW, 2154 Australia). For all aqueous experiments Ultrapure water with a resistivity of 18.2 MΩ-cm was used (Sartorius AG, Germany). The N-acetyl β3-peptides Ac-β3[LIA], Ac-β3[ALI] and Ac-β3[IAL] were synthesized in house using solid phase synthesis as described previously [3].

Atomic force microscopy (AFM) imaging

Stock solutions containing 1 mg peptide were prepared in 1 mL of either methanol, ethanol, isopropanol or water. For AFM imaging the stock solution was deposited onto freshly cleaved mica surface. All samples were dried overnight and then imaged using an Ntegra AFM platform (NT-MDT, Russia) under ambient conditions using golden silicon probes with nominal 10 nm apex radius and of a typical spring constant of 72 Nm−1. For imaging, 512 × 512 pixel resolution and 0.5−1 Hz scan rate was used.

Synchrotron far-IR experiments

Far infrared spectra of lyophilized peptides were collected at the Far-IR Beamline at the Australian Synchrotron. A total of ~0.1–0.2 mg of each peptide was placed in diamond or polyethylene cells at the focal point of the beam and room temperature transmission spectra were recorded with a Bruker IFS125/HR Fourier transform spectrometer with a He-cooled Si bolometer. The scan range was limited to 700 cm−1 and the resolution was 2.0 cm−1. 200 scans were collected in each experiment. Spectra represent the average of ten experiments per sample.

Results and discussion

Peptide design

The primary self-assembly motif in the N-acetylated tri-β3-peptides is the three point hydrogen bonding motif that creates an extended pseudo-14-helical structure [3, 4]. The lateral association of these core nanorods is largely determined by the amino acid sequence of the monomers via a combination of hydrophobic attraction, van der Waals (vdW) interaction between the amino acid side chains, and inter-fibril H-bonding from the exposed C-terminal carboxyl groups [4]. These interactions are strongly solvent dependent and their relative strengths and the topographic complementarity between the neighbouring fibrils control the lateral assembly. In this study we analyzed the effect of nanorod surface topography on the superstructure morphology resulting from the lateral association of the self-assembled β3-peptide nanorods. This was achieved by comparing the superstructure morphologies formed by tri-β3-peptides containing the same β3-amino acid composition with different sequences, namely Ac-β3[LIA], Ac-β3[IAL] and Ac-β3[ALI] (Fig. 1). Upon self-assembly via the three-point H-bonded motif, the surface topography of the respective core nanorods is defined by four faces characterized by the aligned alanine, leucine and isoleucine side chains and the C-terminal carboxyl group (Fig. 1 surface models). In suspended nanorods, the face defined by the hydrophilic carboxyl is solvated by protic solvents while the hydrophobic areas aggregate to reduce the exposed surface; the aggregation allows vdW interactions to define the assembly geometry. The relative contributions of these two properties are expected to have a strong influence on the self-assembled superstructure and the effect of amino acid sequence on this superstructure was analyzed by AFM.

Fig. 1: 
            Different representations of the three peptides Ac-β3[LIA], Ac-β3[ALI] and Ac-β3[IAL]. From top to bottom: schematic structure; stick frame in 14-helix geometry; surface model of a nanorod comprising three self-assembled peptides in a 14-helix geometry; the 14-helix nanorod assembly with the amide and carboxyl groups highlighted in space-filling mode. Heteroatoms are depicted with conventional colours: oxygen, red; nitrogen, blue; hydrogen, white.
Fig. 1:

Different representations of the three peptides Ac-β3[LIA], Ac-β3[ALI] and Ac-β3[IAL]. From top to bottom: schematic structure; stick frame in 14-helix geometry; surface model of a nanorod comprising three self-assembled peptides in a 14-helix geometry; the 14-helix nanorod assembly with the amide and carboxyl groups highlighted in space-filling mode. Heteroatoms are depicted with conventional colours: oxygen, red; nitrogen, blue; hydrogen, white.

Morphological analysis

The effect of a range of protic solvents on the morphology of each tri-β3-peptide was analyzed by AFM. Specifically, peptides were deposited from solutions of water, methanol, ethanol and isopropanol which provide a series of solvents of decreasing dielectric constants and H-bonding strength. Significantly, all three peptides exhibited distinct morphologies following self-assembly clearly demonstrating that the nanorod cross-sectional topography plays an important role in the hierarchical self-assembly of β3-peptides. When deposited from water solution onto mica surface, Ac-β3[LIA] formed large bundles of considerable thickness up to several microns (Fig. 2a). In striking contrast, Ac-β3[ALI] formed a layer of interwoven, aligned but much thinner fibers, suggesting much weaker hydrophobic effect (Fig. 2b). Finally, a mesh of non-aligned fibers was seen for Ac-β3[IAL] (Fig. 2c).

Fig. 2: 
            AFM of peptide dissolved in water and deposited directly on the surface of fresh mica with a scanning range of 20 × 20 μm. (a) Ac-β3[LIA], height scale 900 nm; (b) Ac-β3[ALI], height scale 71 nm; (c) Ac-β3[IAL], height scale 61 nm.
Fig. 2:

AFM of peptide dissolved in water and deposited directly on the surface of fresh mica with a scanning range of 20 × 20 μm. (a) Ac-β3[LIA], height scale 900 nm; (b) Ac-β3[ALI], height scale 71 nm; (c) Ac-β3[IAL], height scale 61 nm.

The main characteristics of the tri-β3-peptide superstructures are defined by the alignment imposed on the nanorod assemblies by the van der Waals/hydrophobic packing of the side chains, and the solvation of the C-terminus, which in non-buffered solvents is likely to be involved in direct hydrogen bonding between the carboxyl group and solvent molecules [3, 4]. In alcohols of decreasing dielectric constants, the diminishing solvophobic forces that drive the aggregation of individual nanorods will be compensated by the increasing van der Waals attraction between the nanorods. Accordingly, for Ac-β3[LIA] the increasing dominance of van der Waals interaction facilitates a switch from the linear bundle geometry to a quasi three-dimensional dendritic structure over the range of dielectric constants from 80 (water) to 18.3 (isopropanol) (Figs. 2a and 3a,d,g). For this peptide, the decreasing alignment effect of the solvation of the C-terminus also contributes to the increased three-dimensionality of the structures in isopropanol, as previously reported [4].

Fig. 3: 
            AFM images of the superstructures formed by peptides Ac-β3[LIA] (left), Ac-β3[ALI] (middle) and Ac-β3[IAL] (right). Scanning range 20 × 20 μm. (a), (b), (c) Peptides deposited from methanol solution; height scale (a) 86 nm, (b) 74 nm, (c) 57 nm. (d), (e), (f) Peptides deposited from ethanol solution; height scale (d) 70 nm, (e) 52 nm, (f) 30 nm. (g), (h), (i) Peptides deposited from isopropanol solution; height scale (g) 159 nm, (h) 90 nm, (i) 57 nm.
Fig. 3:

AFM images of the superstructures formed by peptides Ac-β3[LIA] (left), Ac-β3[ALI] (middle) and Ac-β3[IAL] (right). Scanning range 20 × 20 μm. (a), (b), (c) Peptides deposited from methanol solution; height scale (a) 86 nm, (b) 74 nm, (c) 57 nm. (d), (e), (f) Peptides deposited from ethanol solution; height scale (d) 70 nm, (e) 52 nm, (f) 30 nm. (g), (h), (i) Peptides deposited from isopropanol solution; height scale (g) 159 nm, (h) 90 nm, (i) 57 nm.

Ac-β3[ALI] assembly revealed a weaker role of the solvophobic effect in water than Ac-β3[LIA] and it formed branched dendritic structures already in methanol, consistent with van der Waals packing (Fig. 3b). Thicker structures formed in ethanol (Fig. 3e) with clearly visible twisted rope-like branches while in isopropanol a three-dimensional tightly woven dendritic hierarchical structure was observed (Fig. 3h).

In contrast, the fiber mesh formed by Ac-β3[IAL] in water (Fig. 2c) suggests that the strength of lateral interaction is insufficient to form well-defined bundles. Ac-β3[IAL] formed similar structures in methanol (Fig. 3c), somewhat better aligned but largely two dimensional structures in ethanol (Fig. 3f) and the superstructure exhibited a branching assembly only in isopropanol (Fig. 3i), which is nevertheless less well-defined than for the other two peptides.

The space filling models of each peptide (Fig. 1) show clear differences in surface contours and overall shape and the different morphologies evident in Figs. 2 and 3 demonstrate that the differences in geometric complementarity as a result of sequence differences control the nanorod self-assembly process. Hence the oval geometry of Ac-β3[IAL] promotes pair-wise assembly, in particular in high dielectric constant media where the van der Waals interactions are weakened and thus only the contact of large surface areas result in permanent assembly. A high resolution image of Ac-β3[IAL] in ethanol supports this interpretation, with fibrils aligning in loose two dimensional bundles (Fig. 4).

Fig. 4: 
            Ac-β3[IAL] deposited from ethanol solution.
Fig. 4:

Ac-β3[IAL] deposited from ethanol solution.

Vibrational spectroscopy

The interpretation of the superstructures as the outcome of morphologic compatibility of nanorod cross sectional geometries presumes that the core helices are identical. It is possible to probe similarities and differences of molecular geometries using mid- to far-infrared spectroscopic methods although a full structural analysis of such a dynamic system is extremely challenging. The accuracy of peak assignment can be improved if it is informed by structural and morphological data, in the present work by the crystal structures of similar peptides [3] and by the superstructure morphologies obtained from AFM imaging. Synchrotron far-infrared spectroscopy was used to characterize the fibrils in terms of the effect of geometric factors and second order interactions on molecular vibrations. Absorption peaks in the far-IR spectra of Ac-β3[LIA], Ac-β3[ALI] and Ac-β3[IAL] at room temperature (Fig. 5) were well-defined compared to spectra obtained from Ala-rich polypeptides and proteins [20], suggesting a high degree of conformational homogeneity within each of the three peptides. The 400–650 wavenumber range offers the most information for structural analysis, as this range contains amide and carboxyl vibration modes that are sensitive to conformation [19, 20].

Fig. 5: 
            Far-infrared spectra (absorbance) of Ac-β3[LIA], Ac-β3[ALI] and Ac-β3[IAL].
Fig. 5:

Far-infrared spectra (absorbance) of Ac-β3[LIA], Ac-β3[ALI] and Ac-β3[IAL].

The band at ~600 cm−1 is generally associated with out-of-plane C=O bending (Amide VI) [22–24] and is prominent for all three peptides, although somewhat broadened in the case of Ac-β3[IAL] and Ac-β3[ALI]. These modes are known to be sensitive to the chemical environment and the folding geometry, varying between 550 cm−1 and 650 cm−1 in hydrogen bonded tri-α-peptides, poly-α-peptides and polyamides [22–24]. However, the peak position in Fig. 5 is the same for all three peptides, suggesting that all three peptides adopt the same core structure without appreciable variation, which from our previous results [3, 4] is likely to be the 14-helical structure.

The strong band at ~470 cm−1 may be assigned to O=C–N deformation [25]. These vibrations are very sensitive to the secondary structure, in particular the chain order and geometry of H-bonds, and are therefore sensitive to intermolecular interactions [25]. This band is shifted to different degrees in the three isomeric peptides. The peak is the narrowest in Ac-β3[IAL]. In Ac-β3[ALI], the peak is found at ~445 cm−1. Such large shifts can result from a difference in hydrogen bond pattern and strength, and from differences in coupling to skeletal and side chain motion. It has been previously shown, for example, that far infrared bands of backbone amides correlate with the folding of α-peptides in either alpha helix or beta sheet conformations, e.g., L-Ala in a polypeptide with β-sheet conformation exhibited a characteristic peak at ~445 cm−1 but in α-helix conformation the characteristic peak occurred at 375 cm−1 [21]. Thus the results suggest that in the backbone of Ac-β3[ALI] adopts a different structure. Indeed, considering the two small groups at the N terminal region (alanine and the terminal methyl) it is feasible to assume the N terminus has a higher conformational flexibility, stretching the 14-helical symmetry.

The C-terminal carboxyl of these peptides is protonated and is free for inter-fibril H-bonding, or to be solvated. Accordingly, carboxyl bands carry information about the packing of the fibers. The COO mode at ~617 cm−1 [26] is seen in all three peptides at the same position, suggesting that the carboxyl groups do not participate in H-bonds in the dry deposit. Given that the measurements were performed on fibers purified from water:acetonitrile (~65:35 mol:mol), this is consistent with the full hydration of the C-terminus in the solution phase.

The region between 400 and 500 cm−1 is predicted to also contain contributions from internal vibrational modes of side chains (i.e., modes which are unaffected by backbone geometry) [27, 28]; remarkably the three isomeric peptides have no common peaks in this region, which indicates that the conformation of the side chains is dependent on their position in the peptide sequence. Spectra in the region below 200 cm−1 (not shown) display clear but different modes (in contrast to near featureless spectra of α-polypeptides [20]) which are related to intramolecular movements/H-bond network [29], but are not possible to clearly assign without extensive molecular mechanic modeling. Ac-β3[IAL] has the fewest and broadest modes compared with the other tripeptides which indicates it has the least ordered network.

Thus the far-infrared spectroscopic results suggest that the supramolecular structures formed by the three isomeric peptides are very similar but not identical, with subtle differences in the backbone of Ac-β3[IAL] and in the side chain orientations of Ac-β3[ALI]. Due to H-bonding to water throughout the deposition process, the C-terminus does not contribute as a H-bonding partner to the inter-fibril interactions and thus the superstructure is largely defined by the van der Waals interactions between the side chains with the COOH faces of the nanorods oriented outwards.

Conclusion

We have investigated the effect of the surface topography of β3-tripeptide nanorods on their lateral self-assembly into hierarchical superstructures. We achieved different superstructure morphologies by a variation of the sequence of the same three residues. AFM imaging shows markedly different superstructures for the three peptides. Far-infrared spectroscopy confirms the similarity of the core structures, and that the carboxyl group is free in the assemblies, thus it is mostly solvated in the dispersant. Overall, the results demonstrate that the interfibril self-assembly and the superstructure morphology of the three peptides is determined primarily by the degree of geometric compatibility of the cross section of the core fibrils, with an orienting role of the solvation of the C terminus. Overall, we have shown that the spatial distribution of different non-polar side chains provides a subtle but powerful approach to impose fundamentally different structural characteristics on the resulting self-assembled nanomaterial.


Article note

A collection of invited papers based on presentations at the 2nd International Conference on Bioinspired and Biobased Chemistry and Materials: Nature Inspires Chemical Engineers (NICE-2014), Nice, France, 15–17 October 2014.



Corresponding author: Adam Mechler, La Trobe Institute for Molecular Science, Department of Chemistry and Physics, La Trobe University, Melbourne, Victoria, Australia, e-mail:
aThese authors contributed equally.

Acknowledgments

The authors acknowledge Dr. Joonsup Lee (University of Sydney) for helping with far-infrared spectroscopic data collection and Dr. Dominique Appadoo at the Australian Synchrotron for his invaluable insight and technical assistance. Part of this research was undertaken on the THz/Far-IR beamline at the Australian Synchrotron, Victoria, Australia.

References

[1] Y. Cui, S. N. Kim, R. R. Naik, M. C. McAlpine. Acc. Chem. Res. 45, 696 (2012).Search in Google Scholar

[2] D. N. Woolfson, Z. N. Mahmoud. Chem. Soc. Rev. 39, 3464 (2010).Search in Google Scholar

[3] M. P. Del Borgo, A. I. Mechler, D. Traore, C. Forsyth, J. A. Wilce, M. C. J. Wilce, M. I. Aguilar, P. Perlmutter. Angew. Chem. Int. Ed. 52, 8266 (2013).Search in Google Scholar

[4] R. S. Seoudi, M. P. Del Borgo, K. Kulkarni, P. Perlmutter, M.-I. Aguilar, A. Mechler. New J. Chem. 39, 3280 (2015).Search in Google Scholar

[5] S. Mann. Nat. Mater. 8, 781 (2009).10.1038/nmat2496Search in Google Scholar PubMed

[6] G. M. Whitesides, J. P. Mathias, C. T. Seto. Science 254, 1312 (1991).10.1126/science.1962191Search in Google Scholar PubMed

[7] G. M. Whitesides, B. Grzybowski. Science 295, 2418 (2002).10.1126/science.1070821Search in Google Scholar PubMed

[8] G. M. Whitesides, M. Boncheva. Proc. Natl. Acad. Sci. USA 99, 4769 (2002).10.1073/pnas.082065899Search in Google Scholar PubMed PubMed Central

[9] S. T. R. Walsh, H. Cheng, J. W. Bryson, H. Roder, W. F. DeGrado. Proc. Natl. Acad. Sci. USA 96, 5486 (1999).10.1073/pnas.96.10.5486Search in Google Scholar PubMed PubMed Central

[10] R. V. Ulijn, A. M. Smith. Chem. Soc. Rev. 37, 664 (2008).Search in Google Scholar

[11] S. G. Zhang. Biotechnol. Adv. 20, 321 (2002).10.1007/s12033-002-0004-5Search in Google Scholar

[12] R. P. Cheng, S. H. Gellman, W. F. DeGrado. Chem. Rev. 101, 3219 (2001).Search in Google Scholar

[13] W. S. Horne, S. H. Gellman. Acc. Chem. Res. 41, 1399 (2008).Search in Google Scholar

[14] S. Zhang, M. Andreasen, J. T. Nielsen, L. Liu, E. H. Nielsen, J. Song, G. Ji, F. Sun, T. Skrydstrup, F. Besenbacher, N. C. Nielsen, D. E. Otzen, M. D. Dong. Proc. Natl. Acad. Sci. USA 110, 2798 (2013).10.1073/pnas.1209955110Search in Google Scholar PubMed PubMed Central

[15] K. Morris, L. Serpell. Chem. Soc. Rev. 39, 3445 (2010).Search in Google Scholar

[16] E. S. Stevens, N. Sugawara, G. M. Bonora, C. Toniolo. J. Am. Chem. Soc. 102, 7048 (1980).Search in Google Scholar

[17] R. Kaur, M. Ramesh, P. V. Bharatam, R. Kishore. J. Phys. Chem. B 118, 9199 (2014).10.1021/jp5042074Search in Google Scholar PubMed

[18] T. A. Martinek, F. Fulop. Eur. J. Biochem. 270, 3657 (2003).Search in Google Scholar

[19] D. F. Plusquellic, K. Siegrist, E. J. Heilweil, O. Esenturk. ChemPhysChem 8, 2412 (2007).10.1002/cphc.200700332Search in Google Scholar PubMed

[20] T. Ding, A. P. J. Middelberg, T. Huber, R. J. Falconer. Vib. Spectrosc 61, 144 (2012).10.1016/j.vibspec.2012.02.020Search in Google Scholar

[21] K. Itoh, H. Katabuchi. Biopolymers 12, 921 (1973).10.1002/bip.1973.360120419Search in Google Scholar

[22] J. Bandekar, S. Krimm. Biopolymers 27, 885 (1988).10.1002/bip.360270602Search in Google Scholar

[23] K. Itoh, T. Shimanouchi. Biopolymers 9, 383 (1970).10.1002/bip.1970.360090402Search in Google Scholar

[24] H. Tadokoro, M. Kobayash, H. Yoshidom, K. Tai, D. Makino. J. Chem. Phys. 49, 3359 (1968).Search in Google Scholar

[25] T. C. Cheam, S. Krimm. J. Mol. Struct. 146, 175 (1986).Search in Google Scholar

[26] M. T. S. Rosado, M. L. R. S. Duarte, R. Fausto. J. Mol. Struct. 410, 343 (1997).Search in Google Scholar

[27] S. J. La’Verne, S. Srivastava, S. Srivastava, V. D. Gupta. Int. J. Biol. Macromol. 46, 131 (2010).Search in Google Scholar

[28] S. Srivastava, P. Tandon, V. D. Gupta, S. Rastogi. Polymer 37, 5401 (1996).10.1016/S0032-3861(96)00389-8Search in Google Scholar

[29] Y. F. Ding, L. Yang, S. Y. Zhang, Y. Wang, Y. L. Du, J. Pu, G. Peng, Y. Chen, H. N. Zhang, J. H. Yu, H. Y. Hang, P. Wu, F. Q. Yang, H. Y. Yang, A. Steinbuchel, P. S. Liu. J. Lipid Res. 53, 399 (2012).Search in Google Scholar

Published Online: 2015-09-08
Published in Print: 2015-10-01

©2015 IUPAC & De Gruyter

Downloaded on 1.5.2024 from https://www.degruyter.com/document/doi/10.1515/pac-2015-0108/html
Scroll to top button