Supramolecular hydrogels based on the epitope of potassium ion channels†

Yi Kuang , Yuan Gao , Junfeng Shi , Hsin-Chieh Lin and Bing Xu *
Department of Chemistry, Brandeis University, 415 South St., Waltham, MA 02454, USA. E-mail: bxu@brandeis.edu; Fax: +1 781-736-2516; Tel: +1 781-736-5201

First published on 24th June 2011

---

---

This communication reports a novel supramolecular hydrogelator, which consists of fluorene and the pentapeptide epitope (TIGYG) of potassium ion (K⁺) channels,¹ to self-assemble in water to form the tunable, hierarchical nanostructures dictated by the concentration of K⁺ ([K⁺]). Supramolecular hydrogels² rely on non-covalent interactions to drive the self-assembly of small molecules in water for forming molecular nanofibers and encapsulating water. As a new type of soft materials,³ supramolecular hydrogels have attracted considerable research attentions because of their promising applications in biomedicine. Although hydrophobic interactions originating from alkyl chains or aromatic rings are essential non-covalent interactions for the formation of ordered nanostructures in water, which afford supramolecular hydrogels, other additional non-covalent interactions also are useful for triggering the self-assembly or enhancing the nanostructures formed by self-assembly. For example, the electrostatic interaction between calcium cations and carboxylate anions renders the simple mixture of cell suspensions and peptide amphiphile molecules into a hydrogel to guide the differentiation of neuron progenitor cells,⁴ provides a convenient method to tailor the mechanical properties of hydrogels,⁵ or offers ionic bridging for significantly affecting the self-assembly of the peptide building blocks.⁶Zinc ion-triggered folding of β-hairpin results in the formation of molecular nanofibers and hydrogels.⁷ More interestingly, the binding of G-quartets (G₄) with a cation (e.g.Na⁺ or K⁺) has resulted in constitutionally dynamic hydrogels that permit multilevel self-assembly processes for phase-organization and phase-transition events.⁸ These seminal works have encouraged us to evaluate the use of selective interactions of K⁺ and small peptides for constructing supramolecular hydrogels.

Although there are scarce reports on the use of K⁺ and small peptides for producing organogels⁹ or hydrogels,¹⁰ the interaction between K⁺ and the peptides is still electrostatic (i.e., attraction between cations and anions), which inherently lacks selectivity. The selectivity exhibited by G₄ towards K⁺ suggests that it is easier to achieve selectivity toward alkaline metal ions by using a neutral ligand for binding the cations.¹¹ Based on this principle, we choose to learn from nature and use the K⁺-binding epitope of the K⁺ channels¹ as the building block for the hydrogelator and evaluate the specificity of the hydrogelator towards K⁺. We find that the attachment of a fluorene group to the pentapeptide epitope of potassium ion channels affords a supramolecular hydrogelator that self-assembles into nanofibers whose widths and crosslinking depend on [K⁺] (Scheme 1), but not on the concentration of Na⁺ ([Na⁺]), which suggests the selective binding of the nanofibers towards K⁺. As the first example of the generation of a hydrogel by mimicking the functions of sophisticated membrane proteins, this work opens a new venue to develop functional materials that exhibit specific responses to ions of biological importance.

Fig. 1A and B show the key structural feature of the K⁺ channel,¹ which serves as the base of the molecular design of the hydrogelator. In a typical potassium channel, four parallel K⁺ binding epitopes (TIGYGs) form a K⁺ filter that allows the selective in-flow of K⁺ ions.¹ The electro-negative and cation-attractive carbonyl oxygen atoms of the epitopes align towards the axis of the filter that favours the entry of K⁺ and disfavours the entry of Na⁺. This feature suggests that the arrangement of multiple TIGYGs in bundles may offer the selective binding to K⁺. Therefore, we decided to use aromatic–aromatic interaction from fluorene for promoting the self-assembly of TIGYGs. There are several additional advantages for this choice. (i) Aromatic–aromatic interactions can effectively enable the self-assembly of pentapeptides to form nanofibers¹² so that epitopes can arrange in a close proximity. (ii) Aromatic–aromatic interaction is orthogonal to the attraction of carbonyl oxygen to K⁺. (iii) As part of the protecting group for amino acids, it is easy to add fluorene to TIGYG. Thus, we connected the fluorenylmethyloxycarbonyl (Fmoc) group¹³ at the N-terminal of TIGYG to generate molecule 1 (Fig. 1C and D) via standard solid phase peptide synthesis and obtained 1 in a yield of about 90%.

To test the minimum gelation concentration (mgc) of 1, we evaluate the hydrogelation of 1 at different concentrations of 1, but at the constant ratio of [K⁺]/[1] = 2.33 because the adjustment of the pH (by adding 1.0 N KOH) inevitably introduces K⁺. To produce the hydrogel, we first adjust pH of the solutions of 1 to 8.0, then increase the temperature of the solutions to 50 °C before adding 1.0 N HCl to lower the pH to 4.0. We observe the formation of the hydrogels after placing the solutions of 1 at room temperature overnight and determine that the mgc of 1 is 0.10 wt% (1.37 mM). Below this value, 1 in water only results in a viscous solution (Fig. S1, ESI†).

Since the event of hydrogelation itself can serve as a simple visual assay for ligand–receptor interactions,¹⁴ we use the formation of the hydrogels to evaluate the interaction between K⁺ or Na⁺ and hydrogelator 1 by mixing the solution of KCl or NaCl at different molar ratios to 1 when the concentration of 1 is 0.05 wt% (below the mgc). If there is a sufficient interaction between K⁺ or Na⁺ and TIGYG, it will increase the crosslinking of the nanofibers and result in hydrogelation. As shown in Fig. 2, 0.05 wt% of 1 remains as a solution regardless of the increase of [Na⁺] from [Na⁺]/[1] = 2.33 up to [Na⁺]/[1] = 82.33, indicating that Na⁺ unlikely binds strongly with 1. In contrast to the case of Na⁺, the increase of [K⁺]/[1] from 2.33 to 12.33 triggers hydrogelation of the solution of 1 at 0.05 wt%, suggesting that K⁺ likely binds to 1 to promote the crosslinking of the nanofibers of 1 to form a sufficient network for the hydrogelation. When [K⁺]/[1] reaches 22.33, the hydrogel becomes softer. Upon the further increase of [K⁺] to [K⁺]/[1] = 42.33 and 82.33, the hydrogels appear too soft to maintain a perfect self-support. Plotting the density of the gels or the solution versus [M⁺]/[1] (Fig. S2, ESI†), we find that the difference in weight of the solutions or hydrogels of 1 with the increase of the amount of K⁺ or Na⁺ is negligible, suggesting that factors other than weight must affect the mechanical strength of the K⁺ containing hydrogels.

To explain the influence of K⁺ on 1, we used TEM to reveal the nanoscale structures of the hydrogels or viscous solutions of 1 at 0.05 wt% of 1 and at different [K⁺] or [Na⁺]. As shown in Fig. 3A–D, although long extended nanofibers are the major feature of both the solution and hydrogels of 1 at 0.05 wt% and pH 4.0, the increase of [K⁺] clearly changes both the width and the aggregation of the nanofibers. At [K⁺]/[1] = 2.33, the nanofibers have a width of 11.3 ± 1.2 nm and exhibit limited crosslinking. The increase of [K⁺]/[1] to 12.33 changes little the width of nanofibers (11.4 ± 0.6 nm) but dramatically increases the crosslinking between nanofibers. The interaction between TIGYG epitopes and K⁺ likely causes the increase of crosslinking. As the nanofiber comprises many monomers of 1, significant numbers of TIGYG epitopes likely present on the surfaces of the nanofibers. Among their possible conformations, some of the epitopes may adopt the conformation of TIGYG in the potassium filter, thus bind to K⁺. Although the binding to K⁺ can also happen within the nanofiber, the binding of K⁺ with multiple epitopes (mimicking the nature structure) on the surface is more accessible and naturally favoured. Thus, K⁺ should be able to bind with TIGYGs from different nanofibers, serving as the potassium bridges which, when concentration is high enough (e.g., [K⁺]/[1] = 12.33), are capable to “glue” the nanofibers together and dynamically crosslink the nanofibers. Multiple crosslinking leads to a stronger network and consequently enables formation of hydrogels. At [K⁺]/[1] = 22.33, bundles of parallel aligned nanofibers that have the width of 9.9 ± 1.2 nm become the dominate feature. The reduction in width of the nanofibers produces more surface area for the protruding TIGYG to bind with K⁺. The alignment of nanofibers leads to less capacity to withhold water and the decrease in the width of the nanofibers weakens the strength of the network, the two phenomena add up to give a hydrogel that is softer than that at [K⁺]/[1] = 12.33. Upon further increase of [K⁺]/[1] to 42.33 and 82.33, the width of nanofibers continues to decrease to 6.3 ± 0.7 and 7.2 ± 0.5 nm, respectively (Fig. S3, ESI†) thus generating more surface area. As thinner nanofibers have weaker mechanical strength, the hydrogels become softer at high [K⁺]. On the other hand, the width of the nanofibers and the aggregation of the nanofibers exhibit little difference at low or high [Na⁺] (widths are 8.3 ± 0.6 and 8.0 ± 1.0 nm at [Na⁺]/[1] = 2.33 and 82.33, respectively), suggesting that there is little interaction between 1 and Na⁺.

Fig. 4 shows the storage moduli of the hydrogels or viscous solutions formed by 1 (0.05 wt%, 0.68 mM) with K⁺ or Na⁺, indicating that the change of G′ with [K⁺] agrees with the morphological change. The values of G′ of the hydrogels with different [K⁺]/[1] peak at [K⁺]/[1] = 12.33. When the [K⁺]/[1] is above 12.33, G′ gradually decreases with increase of [K⁺], agreeing with the appearance of softer hydrogels. In the control test, in which 1 exhibits little interaction with Na⁺, the values of G′ only show a trend of slight decline, likely resulted from the salting-in effect¹⁶ that weakens the mechanical strength of the system. These rheological behaviours agree with that 1 selectively binds K⁺.

In conclusion, we designed and synthesized the first example of a hydrogelator with the functional epitope (TIGYG) from a natural ion channel protein. Gelation of the hydrogelator 1 occurs exclusively at a specific ratio of [K⁺]/[1], proving the conservation of inherent ability of the epitopes to bind with K⁺. In a control experiment to demonstrate the importance of the structure of the epitope to K⁺ binding, we change the structure of 1 by alternation of the pentapeptide sequence into TGGIY (2) which loses the hydrogelation ability upon the addition of either K⁺ or Na⁺ at 0.05 wt% of 2 (Fig. S6, ESI†). These results imply that self-assembly effectively generates repeats of epitopes for interacting with specific targets in the form of soft nanomaterials when multiple epitopes are necessary for the specific interaction with the target(s).

We are grateful for financial support from the NSF, a start-up grant from Brandeis University, and an HFSP grant (RGP0056/2008). We thank the EM facility at Brandeis University for assistance.

Notes and references

1. O. B. Clarke, A. T. Caputo, A. P. Hill, J. I. Vandenberg, B. J. Smith and J. M. Gulbis, Cell (Cambridge, Mass.), 2010, 141, 1018–1029 CrossRef CAS .
2. Z. Yang, G. Liang and B. Xu, Acc. Chem. Res., 2008, 41, 315–326 CrossRef CAS ; L. Haines-Butterick, K. Rajagopal, M. Branco, D. Salick, R. Rughani, M. Pilarz, M. S. Lamm, D. J. Pochan and J. P. Schneider, Proc. Natl. Acad. Sci. U. S. A., 2007, 104, 7791–7796 CrossRef ; L. A. Estroff and A. D. Hamilton, Chem. Rev., 2004, 104, 1201–1217 CrossRef ; M. Zelzer and R. V. Ulijn, Chem. Soc. Rev., 2010, 39, 3351–3357 RSC ; N. M. Sangeetha and U. Maitra, Chem. Soc. Rev., 2005, 34, 821–836 RSC ; P. Terech and R. G. Weiss, Chem. Rev., 1997, 97, 3133–3160 CrossRef ; S. Kiyonaka, K. Sugiyasu, S. Shinkai and I. Hamachi, J. Am. Chem. Soc., 2002, 124, 10954–10955 CrossRef .
3. G. M. Whitesides and D. J. Lipomi, Faraday Discuss., 2009, 143, 373–384 RSC ; A. Dawn, T. Shiraki, S. Haraguchi, S. Tamaru and S. Shinkai, Chem.–Asian J., 2011, 6, 266–282 CrossRef CAS ; D. C. Lin and F. Horkay, Soft Matter, 2008, 4, 669–682 RSC .
4. G. A. Silva, C. Czeisler, K. L. Niece, E. Beniash, D. A. Harrington, J. A. Kessler and S. I. Stupp, Science, 2004, 303, 1352–1355 CrossRef CAS .
5. M. A. Greenfield, J. R. Hoffman, M. O. de la Cruz and S. I. Stupp, Langmuir, 2010, 26, 3641–3647 CrossRef CAS .
6. J. C. Stendahl, M. S. Rao, M. O. Guler and S. I. Stupp, Adv. Funct. Mater., 2006, 16, 499–508 CrossRef CAS .
7. C. M. Micklitsch, P. J. Knerr, M. C. Branco, R. Nagarkar, D. J. Pochan and J. P. Schneider, Angew. Chem., Int. Ed., 2011, 50, 1577–1579 CrossRef CAS .
8. A. Ghoussoub and J. M. Lehn, Chem. Commun., 2005, 5763–5765 RSC ; V. Setnicka, M. Urbanova, K. Volka, S. Nampally and J. M. Lehn, Chem.–Eur. J., 2006, 12, 8735–8743 CrossRef CAS ; N. Sreenivasachary and J. M. Lehn, Proc. Natl. Acad. Sci. U. S. A., 2005, 102, 5938–5943 CrossRef ; L. E. Buerkle, Z. Li, A. M. Jamieson and S. J. Rowan, Langmuir, 2009, 25, 8833–8840 CrossRef .
9. M. Suzuki, M. Yumoto, M. Kimura, H. Shirai and K. Hanabusa, Helv. Chim. Acta, 2004, 87, 1–10 CrossRef CAS .
10. S. Ghosh, S. K. Singh and S. Verma, Chem. Commun., 2007, 2296–2298 RSC .
11. K. Suzuki, K. Watanabe, Y. Matsumoto, M. Kobayashi, S. Sato, D. Siswanta and H. Hisamoto, Anal. Chem., 1995, 67, 324–334 CrossRef CAS .
12. M. Ma, Y. Kuang, Y. Gao, Y. Zhang, P. Gao and B. Xu, J. Am. Chem. Soc., 2010, 132, 2719–2728 CrossRef CAS ; V. Jayawarna, M. Ali, T. A. Jowitt, A. E. Miller, A. Saiani, J. E. Gough and R. V. Ulijn, Adv. Mater., 2006, 18, 611–614 CrossRef .
13. J. W. Sadownik, J. Leckie and R. V. Ulijn, Chem. Commun., 2011, 47, 728–730 RSC .
14. Z. Yang and B. Xu, Chem. Commun., 2004, 2424–2425 RSC .
15. L. L. Frado and R. Craig, J. Mol. Biol., 1992, 223, 391–397 CrossRef CAS .
16. T. Arakawa and S. N. Timasheff, Biochemistry, 1982, 21, 6545–6552 CrossRef CAS .

---

Footnote

† Electronic supplementary information (ESI) available: Fig. S1 to S5. See DOI: 10.1039/c1cc13115j

This journal is © The Royal Society of Chemistry 2011