Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Soft Matter. Author manuscript; available in PMC 2011 January 1.
Published in final edited form as:
PMCID: PMC2915463

Tuning hydrogel properties and function using substituent effects


The physical properties and function of hydrogels are shown to depend on the substituents present in three novel 1,3,5-tri-azaadamantane (TAA) cross-linkers. Gel stiffness and degradation rate at varied pHs could be predictably tuned with the cross-linker substituents used to form the gel. Subsequently, protein release from the hydrogel were controlled with chemical structure of the cross-linker.

Hydrogels prepared by cross-linking hydrophilic polymers can resemble a natural extracellular matrix and, thus, have been used for numerous biomedical applications including drug delivery and cell encapsulation.1 The successful use of hydrogels in these applications very much relies on the ability to tune the hydrogel degradation rate at target tissues which can present different pH environments. Tumor and inflammatory tissues present an acidic environment relative to normal tissue, requiring a gel that degrades under acidic conditions.2 Other applications involving the implantation of hydrogels in normal tissue require a gel that degrades under neutral conditions.3 Degradable hydrogels are commonly prepared by incorporating hydrolytically labile ester,4a anhydride,4b or acetal4c,d groups into gel-forming polymers or cross-linkers, however there is still need for new degradable motifs displaying controlled degradation.

Herein we report a method to tune the hydrolysis rates of hydro-gels at a given pH by using cross-linking agents containing the 1,3,5-triazaadamantane (TAA) unit. As shown schematically in Fig. 1, these cross-linkers possess a hydrolytically labile TAA unit with three aromatic groups to control hydrolysis rates through substituent effects,5,6 three acrylamide groups for cross-linking, and a poly-(ethylene glycol) chain for water solubility. Reports by several groups have shown that the rate of degradation of acetal-containing hydrogels can be tuned under acidic conditions.7 In contrast, the TAA group allows the hydrolysis to occur in a highly tunable fashion across a much broader pH range, extending up into the basic regime.

Fig. 1
A schematic showing (a) generalized TAA trivalent cross-linking agent and (b) degradation of TAA unit within gel.

Three TAA cross-linkers were synthesized to present a similar size, architecture, and reactivity but each possesing different aromatic substituent groups (Scheme 1).8 Thus, TAAs 3a and 3b with electron withdrawing groups (amide and ester linkages, respectively) in the 4-position were expected to be comparatively stable in comparison to TAA 3c with 3,4-alkoxy-substituted (electron rich) aromatic groups. Whereas TAAs 3a and 3c were designed to be stable to base, ester-containing TAA 3b was designed to degrade under both acidic and basic conditions (Fig. S1).

Scheme 1
Synthesis of TAA cross-linkers 3a–c; mPEG ¼ methoxy poly(ethylene glycol).

Polyacrylamide, poly(N-isopropylacrylamide), and poly(2-hydroxy-ethyl acrylate) hydrogels were prepared by in situ photochemical or chemical polymerization of the appropriate monomer solution containing TAAs 3a–c.8 No gels formed in the absence of TAAs, supporting the role of the TAA functionalized acrylamides as a cross-linker (Fig. 2a vs. 2b). The concentration of the TAA determined the mechanical stiffness of the resultant poly-acrylamide hydrogels. Thus, the elastic modulus of the hydrogel increased with the ratio of TAA to acrylamide used in the polymerization (Fig. S2a). The molecular structure of the cross-linker also had an effect on the modulus (Fig. S2b). Hydrogels cross-linked with TAA 3a were approximately twice as stiff as those cross-linked with TAA 3b or 3c at the same monomer to cross-linker ratio. This result can be explained by the restricted free rotation around the amide bond.

Fig. 2
Acrylamide solution polymerized (a) with and (b) without TAA 3a. A 1 mol % concentration of TAA was used.

Changes in the aromatic substitution pattern of the TAA units also produced a noticeable change in the hydrogel degradation rates, as quantified by the increase of hydrogel swelling ratio over time (Fig. 3a). At pH 5.0, gels cross-linked by TAA 3c degraded >100-fold faster than those cross-linked by TAAs 3a or 3b (Fig. 3b). Gels prepared using 1 mol % of TAA 3c completely degraded within seven hours, whereas analogous gels prepared with TAA 3a or 3b slowly swelled over two months. These trends were observed at higher pH values, although the overall degradation rates were lower. At pH 7.4, discs prepared from TAAs 3a and 3b swelled very little over the course of 100 days, whereas hydrogels cross-linked by TAA 3c were fully degraded (Fig. 3c). A further increase of pH to 10 led to degradation of 3b-derived hydrogels as a result of ester hydrolysis (Fig. 3d).

Fig. 3
(a) Images of polyacrylamide hydrogel discs cross-linked with 1 mol % TAA 3c (left) before and (right) after significant TAA degradation. Changes in the surface area over time of polyacrylamide hydro-gels cross-linked by TAAs 3a–c at (b) pH 5.0, ...

The hydrogel degradation rates in acidic and neutral conditions were inversely correlated with the Hammett substituent constants (σ), which quantify the electronic contribution of aromatic substituent groups (Fig. 4).9,10 This result further confirms the role of TAA hydrolysis in the hydrogel degradation under acidic and neutral conditions. In contrast, the degradation rates of hydrogels incubated under basic conditions were independent of σ, suggesting that, in this event, the hydrogel degradation was controlled by base catalyzed cleavage of the ester linkage of gels prepared from 3b (Fig. S1).

Fig. 4
The correlation of Hammett substituent constants (σ) with the log of hydrogel degradation rates (k)11 at pH 5.0 of hydrogel discs cross-linked with TAAs 3a, 3b, or 3c.

We further examined the ability of TAA 3a and 3c-derived hydrogels to control the release of macromolecular drugs in a sustained manner. Thus, polyacrylamide gels were prepared using a 0.5% (w/v) solution of bovine serum albumin (BSA) and a 1 : 19 ratio of 3a or 3c to acrylamide. Incubation of the gel cross-linked by 3c in an acidic media of pH 5.0, resulted in significant protein release; whereas the BSA release from the gel cross-linked by 3a was highly limited (Fig. 5a). At pH 7.4, the amount of protein released from the gels cross-linked with 3c and 3a was greatly diminished as compared to the acidic conditions due to the decrease in cross-linker hydrolysis (Fig. 5b). The percentage of BSA released after 120 h from gels cross-linked with 3c under acidic conditions corresponds to the majority of entrapped protein (> 65%), while a significantly lower percentage (10–35%) was released using the other conditions (Fig. S4). Circular dichroism confirmed that the released BSA had not been denatured from the encapsulation and release processes (Fig. S5).

Fig. 5
The release profile of bovine serum albumin (BSA) from poly-acrylamide hydrogels cross-linked by TAAs 3a or 3c. Hydrogels cross-linked with 5 mol% TAA were incubated in PBS at 37.0 °C at (a) pH 5.0 and (b) pH 7.4.

Lastly, the cytotoxicity of the hydrogels was evaluated. In this study, hydrogels cross-linked with TAAs 3a–3c were fully hydrolyzed in strongly acidic media and varying amounts of the degradation products were added to media used for the culture of NIH3T3 fibroblasts. Cells were mostly viable up to 0.2 mg mL−1 of degraded hydrogel, but a significant decrease in viability was observed at concentrations above 2 mg mL−1. Compared to polyethyleneimine (PEI), a common polymer known for being toxic, the hydrogel degradation products were relatively biocompatable (Fig. 6). Additionally, polyacrylamide hydrogels prepared from all three cross-linking agents (3a–c) resulted in minimal inflammatory response over 4 days in vivo when they were implanted into a chick embryo chorioallantoic membrane (Fig. S6).

Fig. 6
Cytotoxicity of hydrogel evaluated with viability of NIH3T3 fibroblasts cultured with varying amounts of hydrogel degradation products and branched 25 kDa polyethyleneimine (PEI). Hydrogel degradation products of gels cross-linked with TAAs 3a, 3b, and ...

In conclusion, this study describes a novel method to tune hydrogel degradation kinetics using pH-responsive biodegradable cross-linkers. This approach relies on the electronic effect that aromatic substituents have on the hydrolysis rate of the TAA unit. The effects of the cross-linker on the hydrogel degradation rate and subsequent protein release profile in physiological conditions were fully predictable using Hammett constants. We expect that the hydrogel properties and function can be further tuned by controlling the cross-linking density and the polymer concentration in the gel. Such refined control over hydrogel properties and function are desirable for a broad array of applications such as drug delivery that should selectively occur in response to specific pH environments and three dimensional cell culture that should stably last regardless of pH change.12

Overall, the TAA cross-linking agents synthesized in this study may prove useful for controlling the properties and function of hydrogels formed from a broader array of monomers13a and pre-polymers.13b Furthermore, the use of TAAs with multiple reactive acrylamide groups may facilitate the processing of hydrogels in various sophisticated forms including microgel particles.7a,b,14,15 Ultimately, TAA cross-linkers may help expedite the use of hydrogels as targeted drug delivery devices and tissue engineering scaffolds in clinical settings.

Supplementary Material

Supplemental Information


The present work was supported by the National Institute of Health (1 R21 HL097314 A).


Electronic supplementary information (ESI) available: Experimental details. See DOI: 10.1039/c001548b

Notes and references

1. Park K, Shalaby WS, Park H. Biodegradable Hydrogels for Drug Delivery. Technomic Publishing Co; Lancaster, PA: 1993.
2. Tannock IF, Rotin D. Canc Res. 1989;49:4373. [PubMed]
3. Jeong B, Bae YH, Lee DS, Kim SW. Nature. 1997;388:860. [PubMed]
4. (a) Albertsson AC, Varma IK. Biomacromolecules. 2003;4:1466. [PubMed] (b) Kumar N, Langer RS, Domb AJ. Adv Drug Delivery Rev. 2002;54:889. [PubMed] (c) Themistou E, Patrickios CS. Macromolecules. 2006;39:73. (d) Falco EE, Patel M, Fisher JP. Pharm Res. 2008;25:2348. [PubMed]
5. Balija AM, Kohman RE, Zimmerman SC. Angew Chem, Int Ed. 2008;47:8072. [PMC free article] [PubMed]
6. Kohman RE, Zimmerman SC. Chem Commun. 2009:794. [PubMed]
7. (a) Murthy N, Thng YX, Schuck S, Xu MC, Fréchet JMJ. J Am Chem Soc. 2002;124:12398. [PubMed] (b) Griset AP, Walpole J, Liu R, Gaffey A, Colson YL, Grinstaff MW. J Am Chem Soc. 2009;131:2469. [PubMed] (c) Chen W, Meng F, Li F, Ji SJ, Zhong Z. Biomacromolecules. 2009;10:1727. [PubMed]
8. See ESI for further details.
9. (a) Anslyn EV, Dougherty DA. Modern Physical Organic Chemistry. University Science Books; Sausalito: 2006. (b) Hansch C, Leo A, Taft RW. Chem Rev. 1991;91:165.
10. σ value for 3a (0.36) was calculated from the sum of the σmeta value and σ+ value of the methyl ether substitution (ref. 9a) and the σ value for 3b (0.45) and 3c (−0.68) were taken as the σpara value of the methyl ester and methyl amide respectively (ref. 9b).
11. k values were taken as the slopes of the linear data in Fig. S3a and S3b.
12. Lim F, Sun AM. Science. 1980;210:908. [PubMed]
13. (a) Nguyen KT, West JL. Biomaterials. 2002;23:4307. [PubMed] (b) Nuttelman CR, Rice MA, Rydholm AE, Salinas CN, Shah DN, Anseth KS. Prog Polym Sci. 2008;33:167. [PubMed]
14. Debord JD, Lyon LA. J Phys Chem B. 2000;104:6327.
15. Chu C, Schaefer BW, DeVolder RJ, Kong H. Polymer. 2009;50:5288.