Adaptive hydrogels with spatiotemporal stiffening using pH-modulating enzymes

Abstract

Adaptive material systems that autonomously respond to external stimuli are crucial for advancing next-generation smart devices. Biological systems achieve autonomous behavior by utilizing chemical energy from out-of-equilibrium reactions to power life-like functions without requiring external energy inputs. Although responsive hydrogels with embedded enzymatic reactions offer a promising platform for implementing adaptive behavior in synthetic systems, previous studies have focused on controlling the supramolecular self-assembly of responsive building blocks rather than modulating network crosslinking. Here, we demonstrate direct enzymatic modulation of crosslinking density in a double-network hydrogel to achieve autonomous self-stiffening in response to a chemical stimulus. Our adaptive system embeds glucose oxidase within a polyacrylamide-alginate double-network hydrogel containing Ca(EDTA)2- complexes that render the crosslinked alginate network pH-responsive through a competitive calcium binding mechanism. Chemical waves emerging from enzymatic reaction activation propagate at speeds ranging from 15 to 44 um/min, driving spatiotemporal mechanical transitions that increase material stiffness by up to 2.1-fold. By integrating signal sensing and chemomechanical transduction within this responsive hydrogel, we realized adaptive behavior that autonomously converts localized chemical inputs into system-wide mechanical outputs. This positions our adaptive hydrogels as promising model systems to guide the design of intelligent materials for soft robotics and biomedical devices.

Figures (17)

• Figure 1: Design and mechanism of trigger-activated autonomous mechanical transition in adaptive DN hydrogels. a) The GOx enzymatic system and its pH-dependent activity profile. Top: GOx catalyzes the oxidation of glucose to produce gluconic acid, decreasing the system's pH. Bottom: GOx's pH-dependent activity profile generates fast, autocatalytic positive feedback, followed by a negative feedback regulation, enabling alkaline-to-acidic pH transition. At high pH, GOx exhibits minimal catalytic activity, maintaining the system in a quiescent (semi-dormant), excitable state that can be triggered by pH reduction prior to the spontaneous activation of GOx autocatalysis. b) Spatiotemporal signal amplification and propagation from localized acidic triggering. Local acidic signal activates GOx, initiating autocatalytic acid production that diffuses to sequentially activate adjacent GOx molecules and generates self-sustained chemical waves. c) Conversion of chemical signals into mechanical transitions (chemomechanical transduction). The PAM-alginate DN hydrogel undergoes pH-dependent stiffening as EDTA releases calcium ions (Ca^2+) upon enzymatically-driven alkaline-to-acidic pH transition, enabling progressive increase in crosslinking density of alginate chains. The hydrogel transitions from a soft state (alkaline pH, Ca^2+ sequestered by EDTA) to a stiff state (acidic pH, Ca-alginate crosslinking).
• Figure 2: Spontaneous autonomous mechanical transition in pH-responsive alginate induced by GOx autocatalysis. a) Time-sweep rheology with in situ pH monitoring measurements of autonomous temporal mechanical transitions in CaEDTA-alginate system. Stock solutions containing Fi and CaEDTA-alginate are mixed immediately before measurement, enabling monitoring of storage modulus evolution during the GOx-driven pH transition. b) Averaged pH transition and time-sweep curves for Fi concentrations ranging from 7.5112.5 at constant CaEDTA and glucose (50 and 100, respectively). c) Apparent storage modulus as a function of the final pH. Symbols and colors represent different Fi and CaEDTA concentrations, respectively; color scale on the right side indicates alginate crosslinking state from (a). Shaded areas in (b) represent standard deviation. Error bars in (c) show standard deviation of the values acquired in the final 20 of measurement.
• Figure 3: Spontaneous autonomous mechanical transition in pH-responsive DN alginate-based hydrogels induced by GOx autocatalysis. a) Experimental protocol for quasi-static tensile testing of PAM-CaEDTA-alginate DN hydrogels. The PAM network is crosslinked with bisacrylamide (MBA) in the presence of uncrosslinked alginate and CaEDTA, followed by substrate and mediator incorporation via by bath immersion. Tensile tests are performed after GOx-driven pH transition (adaptive) or without substrate/mediator (non-adaptive). b) Averaged stress-strain curves for adaptive and non-adaptive samples (50 CaEDTA, 22.5Fi, 100 glucose, $N = 6$), calculated over the strain interval from 0 to mean strain at rupture. c) Relative increase in apparent modulus and work of fracture of adaptive and non-adaptive samples. Apparent modulus is determined from a linear fit of the 50100 strain interval; work of fracture is calculated from the area under the stress-strain curves. Shaded areas in (b) represent the standard deviation. Error bars in (c) the standard deviation calculated via statistical error propagation.
• Figure 4: Propagation of self-sustained chemical waves in DN hydrogels. a) Experimental protocol for initiating chemical wave propagation in adaptive DN hydrogels. A local chemical signal is delivered by contacting the DN hydrogel with a PAM hydrogel pre-equilibrated at pH 1 for at least 24. b) Representative blue channel images showing duplicate samples at three time points. The trigger hydrogel is false-colored orange ($t = \qty{0}{\minute}$) and defines the triggered region (dashed line), with the propagating chemical wave front marked by a solid line. Sample compositions: GOx (GOx, 250), Fi (Fi, 22.5), CaEDTA (CaEDTA, 50), PAM (12.5) and alginate (2). Scale bar: 5. c) Representative kymograph showing the spatiotemporal progression of the self-sustained chemical wave in the adaptive DN hydrogel. d) Temporal evolution of wave front positions in four adaptive DN hydrogel samples from the same batch, presented in shades of blue. For comparison, the front position of a purely diffusive system (yellow) is shown, determined by Alizarin Red S color change upon 1-minute contact with a hydrogel at pH 1. Sample compositions: adaptive system contained GOx (350), Fi (7.5), glucose (100), and CaEDTA (50); diffusive system contained CaEDTA (50), both contained PAM (12.5) and alginate (2). The trigger hydrogel was removed at $t = \qty{0}{\minute}$ for all experiments.
• Figure 5: Chemical wave propagation kinetics in adaptive DN hydrogels. a) Temporal evolution of wavefront positions in adaptive hydrogels containing varying GOx concentrations (250, 300, 350, and 400) at fixed Fi (Fi, 7.5) and CaEDTA (CaEDTA, 50). Data points represent mean values whereas shaded regions indicate the standard deviation ($N = 4$). b) Wave propagation speed versus GOx concentration for two Fi concentrations (7.5 and 22.5), at constant CaEDTA (50). Solid and open circles indicate self-sustained and damped wave propagations, respectively. c) Correlation between propagation speed and time to spontaneous autocatalysis (time-to-autocatalysis) for the dataset in (b). d) Propagation speed versus CaEDTA concentration at fixed Fi (22.5) and GOx (100). Inset shows propagation speed versus time-to-autocatalysis. Dashed lines in (b–d) represent power law fits ($v \propto x^p$) with exponents indicated in the legends. All adaptive DN hydrogels contained a fixed glucose (100), PAM (12.5) and alginate (2) concentrations.