Formation of molecular hydrogels from a bile acid derivative and selected carboxylic acids

Abstract

Bile acids are natural compounds that can be made into dimers by covalently linking two of them through diethylenetriamine. A cholic acid dimer of this kind is synthesized and is found to form thermally reversible hydrogels with selected carboxylic acids through combined hydrogen bonding and ionic interactions. The gelation and viscoelastic properties of the hydrogels may be varied by judicious choice of the carboxylic mono- and diacids. The total organic content (the dimer and carboxylic acid) represents about 2% or less by weight in the ternary mixture. The molecular arrangement between the dimer and carboxylic acid is proposed to illustrate the formation mechanism of the hydrogels. The marginal solubility of the dimer–acid mixtures seems to be the deciding factor in obtaining the hydrogels.

---

Molecular gels are an important class of soft materials that may be potentially used for drug delivery, tissue engineering, and sensing.^1–3 In molecular gels, solvents are immobilized by a small amount (typically <2 wt%) of low molecular weight gelators (LMWGs).⁴ The small molecules self-assemble to form fibrils and 3-dimensional networks facilitated by supramolecular interactions such as hydrogen bonding, hydrophobic interaction, π–π stacking, electrostatic attraction, and charge transfer interactions between the gelators and solvents.⁵ Such gels often show advantages of biodegradability and diversity of functionality due to noncovalent interactions and precisely controlled structures.⁶ No general rule based on molecular structure seems to exist to predict the gelation behaviour, and many LMWGs have been discovered by serendipity.^7–9

Bile acids are natural compounds found in humans and most animals. In their salt form or after suitable chemical modification, they may form hydrogels by self-assembly.^10–17 For example, sodium cholate may interact with metal cations (Cu²⁺, Zn²⁺, Co²⁺, Ag⁺, trivalent lanthanides, etc.) to form hydrogels.^18–21 A cholic acid-based trimer was reported to form a hydrogel in the presence of up to 20% of acetic acid in aqueous solutions,^22–24 while strong inorganic acid such as HCl may not favour the gelation of the trimer. The necessity of the acids and the interactions between the trimer and acetic acid during the gelation remain unclear. In our work, we discovered that an otherwise insoluble cholic acid-based dimer linked through amide groups and bearing a secondary amine forms a hydrogel in the presence of certain carboxylic acids. The properties of the hydrogels appear to depend on the structure of the acids. Therefore, investigation of their gelation behaviours may lead to a better understanding of these molecular hydrogels.

The dimer based on cholic acid (Fig. 1) was synthesized according to a literature procedure with a simple two-step synthesis and a high yield (Scheme S1†), and it was previously reported to manifest interesting selective antifungal activities.²⁵ We have found that the dimer has limited solubility in water, even in the presence of HCl, but may be solubilized in the presence of certain amounts of selected carboxylic acids to form hydrogels (Chart 1A). Carboxylic acids that are either too strong or too hydrophobic do not favour the formation of hydrogels (Chart 1B, Fig. S1A and B†), which we discuss in more detail further on. In these experiments, hydrogels formed only when the carboxylic acid groups are in excess of the secondary amine groups of the cholic acid dimers. The dimer cannot be dissolved in water at a [dimer]/[–COOH] ratio of 1 : 1 and a cloudy liquid formed, but a transparent hydrogel was obtained at a [dimer]/[–COOH] ratio of 1 : 2 (see Fig. S1C and D†). Therefore, we fixed the concentration of the monoprotic acids at 40 mM and that of diprotic acids at 20 mM, keeping the ratio of [dimer]/[–COOH] at 1 : 2.

Fig. 1 Molecular structure of the dimer based on cholic acid.

Chart 1 The structure of various carboxylic acids tested for hydrogelation with the dimer: (A) acids that can form a hydrogel; (B) acids that cannot form a hydrogel. The values in brackets indicate the pK_a of the carboxylic acid groups.

Rheological studies may provide useful information related to the structure in gels. The mixture of 20 mM dimer and 40 mM acetic acid in water was studied as an example. Both G′ and G′′ increased during the gelation process (Fig. S2A†), reaching a plateau after 10 minutes. Neither G′ nor G′′ showed obvious change under a small oscillation stress (Fig. S2B†), demonstrating the stability of the 3-D networks in the hydrogel. A further increase in stress (>20 Pa) caused a yielding process and flow of the hydrogel due to disintegration of the 3-D fibrillar networks. The moduli (G′, G′′) of the hydrogel were found to depend on the frequency of stress oscillation (Fig. S2C†). G′ remained larger than G′′, although both increased with increasing frequency. G′′ increased at a faster rate than G′ and a crossover point would be expected at a higher frequency, e.g., between 100 and 1000 Hz, which is beyond the detection limit. Both G′ and G′′ decreased with increasing temperature, and their crossover point was observed at around 58 °C (Fig. S2D†), indicating a gel–sol transition. Further rheological tests illustrated that a deviation of the ratio of [dimer]/[–COOH] from 1 : 2 resulted in weaker hydrogels (lower G′ values and lower gel–sol transition temperatures, as shown in Fig. S3†).

The minimal gelation concentration (MGC) of the dimer was found to be dependent on the chemical structure of the carboxylic acids used in the hydrogel. Among the monoprotic acids tested at a concentration of 40 mM, the lowest MGC of the dimer was 5.5 mM obtained with formic acid and with butyric acid. In comparison, lower MGC values were obtained with 20 mM diprotic acids where the presence of more hydroxyl groups leads to even lower MGC values (Table S1†). The dimer may form hydrogels at a lower concentration when the content of the carboxylic acid decreases, which is similar to the gelation behaviour of a cholic acid trimer with acetic acid.²³

The mechanical properties of the hydrogels were also found to be related to the structure of the carboxylic acid. Whereas the dimer can form gels in the presence of all four monoacids having 1–4 carbon atoms (Fig. 2A), a longer alkyl chain on the acid results in better mechanical properties of the gel, likely due to enhanced hydrophobic interaction between the alkyl chains of the carboxylic acid.²⁶ Formic acid can form multiple hydrogen bonds,²⁷ thus improving the interactions between acid molecules that are complexed with the dimer. Therefore, the hydrogel made with formic acid showed the highest modulus among all the hydrogels in the monoacid series (Fig. 2A). In contrast, the hydrogels formed by the diacids have higher G′ than those formed by the monoacids (Fig. 2B).

Fig. 2 Oscillatory stress sweep experiments of the hydrogel formed from the dimer (20 mM) and (A) monoprotic acids (40 mM) and (B) diprotic acids (20 mM). [–COOH] = 40 mM.

The diacids in Fig. 2B all have 4 carbon atoms but different numbers of hydroxyl groups. The presence of additional hydroxyl groups led to better mechanical properties of the gels. For example, G′ increased by an order of magnitude when tartaric acid replaced succinic acid, implying that extra hydroxyl groups improve the interactions between gelator molecules in acid–dimer complexes to yield stronger self-assembled fibrillar networks (SAFINs) in the hydrogels.

The transmission electron microscopy (TEM) images (Fig. 3) show the existence of well-developed intertwined SAFINs in the acid–dimer hydrogels. Different morphologies were observed for hydrogels obtained with different carboxylic acids: interactions with acetic acid yielded mostly straight fibers of 20–40 nm in diameter (Fig. 3A) whereas more flexible and uniform-sized fibers of around 20 nm in diameter were observed when tartaric acid was used (Fig. 3C). Similar morphologies were also observed by AFM (Fig. S4†). Such morphology differences may explain the different gelation capabilities of the dimer with the different acids: thin, flexible, and highly entangled nanofibers may accompany systems with better gelation capabilities.²⁸ The self-assembled fibers had lengths of micrometers, and the 3-D network immobilizes water molecules such that the solution thickens to form a hydrogel. The concentration used for Fig. 3C corresponds to the lowest MGC of the dimer. The complex consisting of one dimer molecule and one tartaric acid molecule may immobilize more than 9 × 10⁴ water molecules, which indicates the good gelation capability of the dimer.

Fig. 3 TEM images of the fibrillar network in hydrogels formed by (A) 6 mM dimer and 12 mM acetic acid; (B) 1.2 mM dimer and 1.2 mM succinic acid; and (C) 0.6 mM dimer and 0.6 mM tartaric acid (black scale bar = 200 nm).

The ¹H NMR spectrum of acetic acid–dimer hydrogel in D₂O varied with temperature (Fig. S6†). The proton peaks from the dimer were broad and unresolved at room temperature but distinguishable when temperature was higher than 45 °C, indicating the increased mobility of the dimer due to the gel–sol transition process. However, the acetic acid peak in the ¹H NMR spectrum of the hydrogel was visible throughout the temperature range, and its integration values relative to the internal standard did not change with temperature. Therefore, acetic acid and the dimer in the hydrogel may exist in regions with different mobility.

FT-IR spectroscopy was used to study the physical interactions in the molecular hydrogels. The FT-IR spectrum of the aerogel formed from the dimer and succinic acid is similar to that of the dimer alone, except for the new peaks from the carboxylic acid groups on succinic acid (Fig. S7†). In Fig. 4A, the second derivative analysis of the dimer FT-IR spectrum shows the amide I band (C=O stretching) split into two peaks, 1651 and 1627 cm⁻¹, and the amide II band (complex of C–N stretching and N–H bending) at 1545 cm⁻¹. After gelation with succinic acid, only a single peak for the amide I band appears at 1650 cm⁻¹, while the amide II band shows a slight shift to 1549 cm⁻¹ (Fig. 4B). These results indicate that the environment of the amide groups changes upon gelation.

Fig. 4 Second derivative of the FT-IR spectra of (A) the dimer; (B) the aerogel of dimer–succinic acid (20 mM–20 mM).

The split peaks of the dimer's amide I band (1651 and 1627 cm⁻¹, Fig. 4A) are similar to those reported for the amide group of poly(N-isopropylacrylamide) in a hydrogel,²⁹ indicating the formation of amide–amide and amide–hydroxyl hydrogen bonds, respectively, in the dimer alone with a proposed molecular arrangement schematically shown in Fig. 5A. The disappearance of the peak at 1627 cm⁻¹ and the shift of the amide II band (Fig. 4B) in the acid–dimer aerogel indicates the disruption of the amide–hydroxyl hydrogen bonds of the dimer and the formation of new hydrogen bonds during gelation, where cholic acid hydroxyl groups are replaced by the carboxylic acid groups (Fig. 5B) due to the better H-acceptor ability of the carbonyl oxygen, and the better H-donor ability of the carboxylic acid hydrogen.³⁰ Formation of the amide–carboxylic acid complex may also help to solubilize the dimer in aqueous solutions upon heating, whereas the cooling process leads to the self-assembly of the dimer to form SAFINs and thus a hydrogel.

Fig. 5 The schematic representation of the hydrogen bonds (A) between the dimers; and (B) between the dimer and carboxylic acid after gelation in water.

For the formation of hydrogels with the dimer, the general requirements of the carboxylic acids are: (1) they should be soluble enough in water to interact with the dimer and form complexes that stabilize the dimer and induce formation of a fibrillar network under suitable conditions; (2) they should be able to protonate the dimer's secondary amine group to increase its polarity, which is feasible since the acids have pK_a lower than 5.0 (Chart 1) and aliphatic secondary amines have a pK_b value of around 3.0,³¹ which should be the case for the dimer; (3) they should be able to form hydrogen bonds with the dimer's amide groups to dissociate interactions between the dimer molecules. Strong carboxylic acids with low pK_a, such as trifluoroacetic acid, are mostly deprotonated in water, hindering the formation of hydrogen bonds with the amide group on the dimer as proposed in Fig. 5B.³²

Conclusions

In summary, the dimer based on cholic acid interacts with weak and hydrophilic carboxylic acids in water to form transparent hydrogels. It is interesting to note that that the total organic contents of cholic acid dimer and the carboxylic acid represent about 2% or less by weight in the mixture. The added carboxylic acid protonates the secondary amine group on the dimer making an otherwise insoluble dimer now marginally soluble in water. Note that no hydrogels are formed with carboxylic acids that are either strong electrolytes or too hydrophobic. The marginal solubility of the gelator seems to be the key in the formation of hydrogels. The formation and rearrangement of hydrogen bonds in the system lead to a stable 3-D fibrillar network in water to yield a hydrogel. All the hydrogels made from the dimer show thermo-reversible gel–sol transitions. Their mechanical properties and morphology vary with the chemical structure of the carboxylic acids used. Monoacids with longer alkyl chains form stronger gels due to hydrophobic interactions; diacids with hydroxyl groups also improve the mechanical properties, owing to the formation of multiple hydrogen bonds. This proves also the case for formic acid due to its greater hydrogen bond formation capabilities. The good gelation ability of the dimer in addition to its biocompatibility may make such hydrogels useful for a variety of biomedical applications. Moreover, a general rule for the gelation ability of the dimer with various carboxylic acids is now established. This may be useful in designing new functional molecular hydrogels and in predicting the gelation behaviour of the dimer with acids and other acidic biomolecules of natural origin. In addition, the reported antifungal activities of the dimer may be retained in the hydrogels formed, making them useful in bio-related applications.

Acknowledgements

Financial support from NSERC of Canada, FQRNT of Quebec, and the Canada Research Chair program is gratefully acknowledged. Meng Zhang thanks the Chinese Scholarship Council (CSC) for a scholarship.

Notes and references

1. R. G. Weiss and P. Terech, Molecular Gels: Materials with Self-Assembled Fibrillar Networks, Springer, Dordrecht, 2006.
2. B. Escuder and J. F. Miravet, Functional Molecular Gels, Royal Society of Chemistry, Cambridge, 2014.
3. A. Dasgupta, J. H. Mondal and D. Das, RSC Adv., 2013, 3, 9117–9149.
4. P. Terech and R. G. Weiss, Chem. Rev., 1997, 97, 3133–3160.
5. M. O. Piepenbrock, G. O. Lloyd, N. Clarke and J. W. Steed, Chem. Rev., 2010, 110, 1960–2004.
6. L. A. Estroff and A. D. Hamilton, Chem. Rev., 2004, 104, 1201–1218.
7. R. G. Weiss, J. Am. Chem. Soc., 2014, 136, 7519–7530.
8. J. Raeburn and D. J. Adams, Chem. Commun., 2015, 51, 5170–5180.
9. X. Du, J. Zhou, J. Shi and B. Xu, Chem. Rev., 2015, 115, 13165–13307.
10. S. Strandman, F. Le Devedec and X. X. Zhu, J. Phys. Chem. B, 2013, 117, 252–258.
11. H. Svobodova, V. Noponen, E. Kolehmainen and E. Sievanen, RSC Adv., 2012, 2, 4985–5007.
12. Y. L. Zhao and J. F. Stoddart, Langmuir, 2009, 25, 8442–8446.
13. V. H. Soto Tellini, A. Jover, F. Meijide, J. V. Tato, L. Galantini and N. V. Pavel, Adv. Mater., 2007, 19, 1752–1756.
14. L. Galantini, C. Leggio, A. Jover, F. Meijide, N. V. Pavel, V. H. Soto Tellini, J. V. Tato, R. Di Leonardo and G. Ruocco, Soft Matter, 2009, 5, 3018.
15. M. Gubitosi, L. Travaglini, A. D'Annibale, N. V. Pavel, J. Vazquez Tato, M. Obiols-Rabasa, S. Sennato, U. Olsson, K. Schillen and L. Galantini, Langmuir, 2014, 30, 6358–6366.
16. Y. Zhang, X. Xin, J. Shen, W. Tang, Y. Ren and L. Wang, RSC Adv., 2014, 4, 62262–62271.
17. M. Maity, V. S. Sajisha and U. Maitra, RSC Adv., 2015, 5, 90712–90719.
18. A. Chakrabarty, U. Maitra and A. D. Das, J. Mater. Chem., 2012, 22, 18268–18274.
19. R. Kandanelli, A. Sarkar and U. Maitra, Dalton Trans., 2013, 42, 15381–15386.
20. Y. Qiao, Y. Lin, Z. Yang, H. Chen, S. Zhang, Y. Yan and J. Huang, J. Phys. Chem. B, 2010, 114, 11725–11730.
21. S. Bhowmik, S. Banerjee and U. Maitra, Chem. Commun., 2010, 46, 8642–8644.
22. U. Maitra, S. Mukhopadhyay, A. Sarkar, P. Rao and S. S. Indi, Angew. Chem., Int. Ed., 2001, 40, 2281–2283.
23. S. Mukhopadhyay, U. Maitra, I. Ira, G. Krishnamoorthy, J. Schmidt and Y. Talmon, J. Am. Chem. Soc., 2004, 126, 15905–15914.
24. P. Terech and U. Maitra, J. Phys. Chem. B, 2008, 112, 13483–13492.
25. D. B. Salunke, B. G. Hazra, V. S. Pore, M. K. Bhat, P. B. Nahar and M. V. Deshpande, J. Med. Chem., 2004, 47, 1591–1594.
26. V. J. Nebot, J. Armengol, J. Smets, S. F. Prieto, B. Escuder and J. F. Miravet, Chem.–Eur. J., 2012, 18, 4063–4072.
27. L. Senthilkumar, T. K. Ghanty, S. K. Ghosh and P. Kolandaivel, J. Phys. Chem. A, 2006, 110, 12623–12638.
28. G. Zhu and J. S. Dordick, Chem. Mater., 2006, 18, 5988–5995.
29. Y. Hirashima, H. Sato and A. Suzuki, Macromolecules, 2005, 38, 9280–9286.
30. L. F. Pacios, Hydrogen Bonding: New Insights, Springer, Dordrecht, 2006.
31. S. Zhang, J. Comput. Chem., 2012, 33, 2469–2482.
32. C. D. Blundell, P. L. Deangelis and A. Almond, Biochem. J., 2006, 396, 487–498.

---

Footnote

† Electronic supplementary information (ESI) available: Materials and method, synthesis of the dimer, gelation behaviour and MGC of the dimer with various carboxylic acids, atomic-force microscopy (AFM) images, circular dichroism (CD) spectroscopy, ¹H NMR spectroscopy, FT-IR spectroscopy. See DOI: 10.1039/c6ra04536g

---