Easy directed assembly of only nonionic azoamphiphile builds up functional azovesicles

Abstract

We obtained functional azovesicles based only on a simple nonionic azoamphiphile, C₁₂OazoE₃OH. The process combines evaporation-induced and solvent-induced directed assembly.

---

Directed assembly has emerged as a powerful strategy to obtain functional systems under non-equilibrium conditions controlled by the processing conditions.¹ In this context, aromatic rod-coil small molecules with amphiphilic features can be very useful because their self-assembly in water is accomplished by thermodynamically driven separation between the polar medium and the rigid π-surface. On the other hand, as π–π interactions are directional such self-assembly can produce structurally precise nano-structures from small molecules.² In general, the easy way to obtain advanced materials is evaporation – induced self-assembly (EISA).³ Typically, it consists in the controlled evaporation of the solvent under rotatory evaporation, allowing solute concentration and the formation of a thin organized mesostructure.^3a,b Once the thin film is hydrated at the transition temperature the formation of nano-objects in water might occur. By this methodology, vesicular systems such as liposomes and niosomes are built up.⁴ A further step is to obtain stimuli responsive vesicles; in this context azobenzene E–Z photoisomerization emerges as a simple molecular spatio-temporal switch which undergoes a large geometric change and dipole moment depending on light conditions.⁵ Previously, an ammonium azobenzene derivative was mixed with a commercial amphiphile obtaining different multi-state nanostructures depending on light conditions.⁶ Recently, other ammonium azobenzene amphiphile was mixed with cholesterol sulphate obtaining non-phospholipid fluid liposomes.⁷ Nonionic azoamphiphiles are seldom investigated; mainly because nonionic amphiphiles are known to form ill-defined aggregates in water.⁸ However, two pure azoamphiphiles with nonionic hydroxyl head form stable nano-objects in water by spontaneous self-assembly.⁹ As far as we are concerned, there are no reports of directed assembly of pure small azobenzene amphiphiles in water. Even less with nonionic features, although they are excellent candidates to form robust nano-objects in water. The major advantage is that the uncharged head allows that the tails conformation change due to photoisomerization is transferred efficiently to the whole aggregate.⁹ Recently, we have obtained a synthetically simple and versatile photoswitchable nonionic azoamphiphile, 4-dodecyloxi-4′-(1-hydroxytriethylenglycol) azobenzene, C₁₂OazoE₃OH (Scheme 1A).^10,§ The interfacial behaviour of C₁₂OazoE₃OH alone and among a model biomembrane was evaluated by Langmuir and Gibbs monolayers. We have proved that C₁₂OazoE₃OH might be an efficient photo-responsible probe to remote control of biological membrane properties.¹⁰

Scheme 1 (A) Photo-responsible nonionic azobenzene employed in this study as functional molecule, C₁₂OazoE₃OH. (B) Illustration of the hypothetical photoswitchable vesicles obtained by directed assembly of the functional molecule.

Here, with respect to previous paper, we point out that C₁₂OazoE₃OH develops functional pure azovesicles without the assistance of surfactants or lipids by EISA (Scheme 1B).

Our hypothesis is based on some relevant features of C₁₂OazoE₃OH. First, the length of head would result in an increase in the minimum aggregation number at which pure vesicles become geometrically allowed.¹¹ Second, the C12 hydrophobic tail would favor bilayer formation.¹¹ Third, C₁₂OazoE₃OH develops a Smetic C mesophase, which was confirmed recently by RDX.¹² Finally, at the interface pure C₁₂OazoE₃OH compressive modulus (bending elasticity) decrease from 100 mN m⁻¹ (E) to circa 60 mN m⁻¹ upon E–Z transformation.¹⁰ In other words, the change in the effective cross-sectional area of C₁₂OazoE₃OH due to photoisomerization led to a substantial change of the interfacial properties and probably this change occurs in bulk, too.^11,9b

We envisaged that directing the assembly of C₁₂OazoE₃OH into liquid crystalline state,¹³ followed by hydration of the film at the transition temperature will build up pure azovesicles in water. To check our working hypothesis, the lyotropic behaviour of C₁₂OazoE₃OH in chloroform was tested by Lawrence's penetration experiment.¹⁴ This experiment consists in melting the solid between two coverslips obtaining a homogenous film. Then, chloroform was allowed to penetrate the solid by capillarity and polarized optical microscopy (POM) observation showed the sequentially obtained mesophases. The result is a fast phase diagram at room temperature, showing the formation of a periodic hexagonal, cubic and lamellar mesophase on increasing concentration (Fig. 1A).^3a

Fig. 1 (A) Lawrence's experiment observed by POM (see arrow from right to left): hexagonal, cubic and lamellar mesophase, respectively. (B) SEM photomicrograph of giant vesicles of C₁₂OazoE₃OH obtained by EISA. Arrows shown the empty interior of the vesicles. (C) TEM photomicrograph of C₁₂OazoE₃OH nano-azovesicles obtained after mild sonication of giant C₁₂OazoE₃OH vesicles. (D) Diameter distribution of the particles observed in image (C).

The lamellar mesophase of C₁₂OazoE₃OH was confirmed after film evaporation at room temperature from a 19 mM solution followed by POM observation (see ESI†). As many aromatic amphiphiles, C₁₂OazoE₃OH is not soluble in water at room temperature, however the critical aggregation concentration (c.a.c) was determinated below 0.35 μM at 70 °C (Krafft point) (further details see ESI†).

Next, we performed EISA³ from a chloroform solution of C₁₂OazoE₃OH,§ using a round bottom flask and a rotary evaporator (30 °C). By this methodology, as the liquid flows away from the contact line it would result in the uniform solute deposition on the substrate. Subsequent hydration at Krafft temperature (70 °C) followed by shaking and ageing for two days led to stable giant vesicles of pure C₁₂OazoE₃OH of 1 to 10 μm diameter, as detected by microscopic techniques (Fig. 1B; further details see ESI†).

Mild sonication led to nanovesicles of 73 ± 1 nm, detected by transmission electron microscopy (Fig. 1C and D see ESI†).

The molecular organization of the azobenzene in the vesicles and the efficiency of photoisomerization in the aggregate state were evaluated by simple UV/Vis absorption spectroscopy. In general, azobenzenes are prone to form J-aggregates and H-aggregates, which possess characteristic UV-Vis spectra.¹⁵ In water, the two bands corresponding to the π–π* transition of (E) C₁₂OazoE₃OH were observable; one at 260 nm, and the second at 311 nm (Fig. 2A). In chloroform, the second band was observed at 358 nm.¹⁰ The blue-shifted of the band from 358 to 311 nm is indicative of strong interaction of the chromophores and is characteristic of H-aggregates in bilayers and cylindrical micelles.¹⁵ Spherical micelles are not able to form H aggregates because the packing is too loose and the chromophores cannot adopt the required configuration in the micelle.¹⁵ This supports the formation of nano-(E) azovesicles and it is well correlated with the morphology observed by TEM. After UV-light illumination a hypochromic shift of both bands at 260 nm and 311 nm occurs, showing that after photoisomerization the chromophores were still as H-aggregates, the photostationary state (pss) was E : Z (46 : 54), for simplicity we will refer to this state as (Z) azovesicles. The new (Z) azovesicles are less rigid than the (E) azovesicles; because of that, TEM observation was not suitable to determinate their size (ESI†). Photo-reversion was achieved after irradiation with a white light bulb (140 min 60 watts).

Fig. 2 (A) UV-Vis spectra of C₁₂OazoE₃OH under UV-light illumination in water (0.5 mM, 150 watts). (B) DLS measurements of hydrodynamic diameter of C₁₂OazoE₃OH before () and after () UV-light illumination in water (140 min, 150 watts). (C) Calcein release during the E–Z isomerization at 310 nm of C₁₂OazoE₃OH vesicles (excitation 415 nm and emission at 515 nm, 150 watts). Passive release was less than 10% after two days. (D) Nile red spectra of C₁₂OazoE₃OH system before () and after () UV-light illumination (12 min, 150 watts). Nile Red contribution in the same experimental conditions was subtracted. For experimental details, please refer to ESI.†

The photomodulation of the (E) azovesicles was further supported by particle size measurements from dynamic light scattering (Fig. 2B). Initially, the hydrodynamic diameter of (E) azovesicles were 166 ± 79 nm and after UV-light irradiation a contraction of the system occurs with a final diameter of 136 ± 41 nm. It seems that the change of the effective cross-sectional area of the C₁₂OazoE₃OH, due to E–Z photoisomerization promoted a change in the size/shape of the azovesicles.^9b,11,16

To prove the photo-delivery properties of C₁₂OazoE₃OH nanovesicles, an encapsulation experiment with the hydrophilic fluorescent dye, calcein is presented. To this aim, the mentioned EISA protocol was performed but, during the hydration step, a solution of 100 mM of calcein in HEPES buffer was employed.¹⁷ After size exclusion chromatography, the free calcein was separated from the calcein-loaded (E) azovesicles.¶ Photoisomerization experiments were performed using the spectrofluorometer (310 nm, 6 cycles of 3 minutes irradiation each). An increase of calcein fluorescence at 515 nm was observed depending on irradiation time (Fig. 2C). The maximum release was 97% considering the maximal release of calcein-loaded (E) azovesicles upon Triton X-100 treatment (ESI†).¹⁷ To further investigate the nature of the azovesicles, the hydrophobic Nile red dye (NR) was added to a dispersion of (E) azovesicles. Nile red senses microenvironment changes by a large blueshift of the emission λ_max depending on probe microenvironment (108 nm from water to hexane) and it is well used to stain neutral lipids in cells. Additionally, NR fluorescence intensity decreased in polar media and/or in the presence of hydrophilic environment.¹⁸

Upon binding to the preformed (E) azovesicles, NR emission was blueshifted from λ_max 658 nm to 643 nm (Fig. 2D and ESI†). After E–Z photoisomerization, NR emission was blueshifted again to λ_max 615 nm indicating that NR is in more hydrophobic environment. Moreover, the emission spectrum after E–Z isomerization was broad indicating different binding modes. The decrease of the fluorescence intensity after E–Z photoisomerization has been previously explained considering the increase of dipolar moment of the Z isomer.¹⁹ Furthermore; it shows that, after irradiation the system is more fluid allowing the NR dye to penetrate better into the superstructure with the consequence blueshift of the emission wavelength. The observed behavior is in good agreement with the aforementioned decrease of the bending elasticity at the interface¹¹ and justify the calcein release upon UV-light illumination.

In conclusion, we develop functional azovesicles or azoniosomes obtained from a small molecule by simple EISA procedures. Taking advantage of the simplicity of the protocols, this strategy might open new opportunities to build up advanced materials from pure pi-conjugated amphiphilic small molecules.

Acknowledgements

This work was supported by UNS-PGI (Universidad Nacional del Sur), CONICET (National Scientific and Technical Research Council) and ANCyPT PICT-PRH 2010–2013 (National Agency for Promotion of Science and Technology) Argentinian Research grants. M. A. S., M. G. H., Z. B. Q. are also grateful for their CONICET fellowship. The authors thank Dr L. Benedini for technical support in c.a.c evaluation. V. I. D. acknowledges her Alexander von Humboldt Foundation fellowship (HERMES).

Notes and references

1. (a) J. L. Jiménez Blanco, et.al., Chem. Commun., 2016, 52, 10117; (b) A. Nikoubashman, V. E. Lee, C. Sosa, R. K. Prud’homme, R. D. Priestley and A. Z. Panagiotopoulos, ACS Nano, 2016, 10, 1425; (c) M. D. Kim, S. A. Derguno and E. Pinkhassik, Langmuir, 2015, 31, 2561.
2. (a) J.-H. Ryu, D.-J. Hong and M. Lee, Chem. Commun., 2008, 1043; (b) M. Molla and S. Ghosh, Phys. Chem. Chem. Phys., 2014, 16, 26672; (c) M. J. Hollamby, et al., Nat. Chem., 2014, 6, 690.
3. (a) C. J. Brinker, Y. Lu, A. Sellinger and H. Fan, Adv. Mater., 1999, 11, 579; (b) D. Grosso, F. Cagnol, G. J. d. A. A. Soler-Illia, E. L. Crepaldi, H. Amenitsch, A. Brunet-Bruneau, A. Bourgeois and C. Sanchez, Adv. Funct. Mater., 2004, 14, 309; (c) W. Han and Z. Lin, Angew. Chem., Int. Ed., 2012, 51, 1534.
4. (a) F. M. Menger and K. Gabrielson, Angew. Chem., Int. Ed., 1995, 34, 2091; (b) P. Walde, H. Umakoshi, P. Stano and F. Mavelli, Chem. Commun., 2014, 50, 10177; (c) S. Gupta, R. Tyagi, V. Parmar, S. Sharma and R. Haag, Polymer, 2012, 53, 3053; (d) E. Soussan, S. Cassel, M. Blanzat and I. Rico-Lattes, Angew. Chem., Int. Ed., 2009, 48, 274.
5. B. Feringa, Molecular Switches, ed. B. Feringa, Wiley-VCH, Verlag GmbH, 2001, p. 399.
6. Y. Wang, N. Ma, Z. Wang and X. Zhang, Soft Matter, 2010, 6, 902.
7. Z.-K. Cui, T. Phoeung, P.-A. Rousseau, G. Rydzek, Q. Zhang, C. G. Bazuin and M. Lafleur, Langmuir, 2014, 30, 10818.
8. (a) C. Tanford in The Hydrophobic Effect, Wiley, New York, 1973; (b) R. Nagarajan and E. Ruckenstein, J. Colloid Interface Sci., 1979, 71, 580.
9. (a) C. Kördel, C. Popeney and R. Haag, Chem. Commun., 2011, 47, 6584; (b) T. Shang, K. Smith and T. Hatton, Langmuir, 2006, 22, 1436.
10. L. A. Benedini, M. A. Sequeira, M. L. Fanani, B. Maggio and V. I. Dodero, J. Phys. Chem. B, 2016, 120, 4053.
11. (a) J. Israelachvili, J. Mitchell and B. Ninham, Biochim. Biophys. Acta, Biomembr., 1977, 470, 185; (b) T. Kunitake and Y. Okahata, J. Am. Chem. Soc., 1977, 99, 3860.
12. X. Tan, R. Zhang, C. Guo, X. Cheng, H. Gao, F. Liu, J. R. Bruckner, F. Giesselmann, M. Prehm and C. Tschierske, J. Mater. Chem. C, 2015, 3, 11202.
13. M. Lee and Y.-S. Yoo, J. Mater. Chem., 2002, 12, 2161.
14. E. S. Blackmore and G. J. T. Tiddy, J. Chem. Soc., Faraday Trans. 2, 1988, 84, 1115.
15. (a) X. Song, J. Perlstein and D. Whitten, J. Am. Chem. Soc., 1997, 119, 9144; (b) J. Kuiper and J. Engberts, Langmuir, 2004, 20, 1152; (c) R. A. Moss and W. Jiang, Langmuir, 1995, 11, 4217; (d) J. Kuiper, M. Stuart and J. Engberts, Langmuir, 2008, 24, 426.
16. (a) Y. Orihara, A. Matsumura, Y. Saito, N. Ogawa, T. Saji, A. Yamaguchi, H. Sakai and M. Abe, Langmuir, 2001, 17, 6072; (b) A. Pulido-Companys, R. Albalat, J. Garcia-Amoros, D. Velasco and J. Ignes-Mullol, Langmuir, 2013, 29, 9635.
17. (a) C. L. Ávila, et al., J. Biol. Chem., 2014, 289, 13838; (b) S. L. Huang and R. C. MacDonald, Biochim. Biophys. Acta, 2004, 1665, 134.
18. (a) P. Greenspan, E. P. Mayer and S. D. Fowler, J. Cell Biol., 1985, 100, 965; (b) F. Deye, T. A. Berger and A. G. Anderson, Anal. Chem., 1990, 62, 615; (c) M. G. Herrera, T. V. Veuthey and V. I. Dodero, Colloids Surf., B, 2016, 141, 565.
19. (a) M. Kumari, M. Billamboz, E. Leonard, C. Len, C. Böttcher, A. K. Prasad, R. Haag and S. K. Sharma, RSC Adv., 2015, 5, 48301; (b) T. Yuan, J. Dong, G. Hana and G. Wang, RSC Adv., 2016, 6, 10904.

---

Footnotes

† Electronic supplementary information (ESI) available: Experimental details and additional figures. See DOI: 10.1039/c6ra20933e

‡ Present Address: Chemistry Department, Bielefeld University, Germany. E-mail; E-mail: veronica.dodero@uni-bielefeld.de

§ In CHCl₃ solution, pure C₁₂OazoE₃OH exists as photostationary state (pss) of E : Z (95 : 5) which, upon UV-light illumination, photoisomerizes in one minute to a second pss E : Z (10 : 90). Complete physical and spectroscopic characterization is in ref. 10.

¶ Passive release was less than 10% after two days in darkness.

---