Rubber-like DNA hydrogel enabled by fast-shrinking-induced entanglement

Abstract

Biomass DNA holds immense potential as a sustainable material, but the scalable production of robust DNA materials with sufficient mechanical strength remains an essential challenge. Herein, we report a Fast-Shrinking-Induced Entanglement (FaSIE) process to achieve rubber-like hydrogels solely composed of DNA (stiffness >800 kPa, toughness >5 MJ/m³, and stretchability > 1000%). Fast shrinking kinetic restricts the chain relaxation, while the ultra-long chain feature of biomass DNA further suppress chain reptation. This synergistic effect enables substantial enhancement of entanglement density, exceeding the threshold required for high mechanical strength. We highlight the broad applicability of this strategy for high-resolution 3D printing of mechanically robust DNA hydrogels and the fabrication of DNA based soft magnetic robots. This approach paves the way for the large-scale production of resilient hydrogel materials derived from biomass DNA for versatile applications.

Biomass DNA has potential in sustainable materials, but scalable production of robust DNA materials is challenging. Here, the authors report the development of a fast shrinking induced entanglement strategy for the preparation of rubber-like hydrogels composed entirely of DNA.

Introduction

The unprecedented accumulation of synthetic materials poses a serious threat to the biosphere, which has led to a critical need in developing sustainable materials from biomass based resources^1–3. Over the years, the researches on sustainable materials are mainly focused on polysaccharides and proteins^4–7. On the other hand, DNA is an inexhaustible biopolymer, which can be extracted from any organisms including plants, animals, and microorganisms. There are ~50 billion metric tons of biomass DNA on the Earth, and utilizing ~0.7% of it would be sufficient to replace the total annual production of synthetic polymers⁸. However, developing biomass DNA as sustainable materials is still in its infancy and significantly underestimated. A major obstacle is the lack of bulk production methods for preparing DNA materials with sufficient mechanical strength⁹. For instance, conventional approaches via strand hybridization and chemical crosslinking result in soft jelly-like DNA hydrogels with elastic moduli and tensile strengths below 50 kPa (Fig. 1A)^10,11. Therefore, although DNA hydrogels have demonstrated attractive biocompatibility and bioactivity for applications such as cell culture matrices, injectable fillings, and localized drug delivery^12–14, their broader utilization as robust biomaterials are significantly constrained. Enhancing the mechanical properties of DNA hydrogels could significantly expand the application scope of biomass DNA.

Fig. 1. Rubber-like DNA hydrogel formed by fast-shrinking induced entanglement (FaSIE).

A The conventional strategies to fabricate DNA hydrogel via base pairing and chemical crosslinking. Both strategies lead to soft and fragile hydrogels. B Current strategy to achieve entangled polymer hydrogel by in situ polymerization at low water-to-monomer ratio. C The illustration of the FaSIE strategy in this work. Long chain biomass DNA (demonstrated as salmon sperm DNA) was dissolved into concentrated solution surpassing the entanglement concentration. The 1:1 (v/v) mixture of glacial acetic acid and Triton X-100 (AA/TX) was introduced to facilitate fast shrinking of this solution and obtain highly entangled rubber-like DNA hydrogel. The dashed box illustrates the shrinking process, where the extraordinary chain length and the fast shrinking kinetic minimized chain disentanglement and maximized the compression caused entanglement. D The photographs showed the appearance of the salmon sperm DNA, the concentrated DNA solution and the resulted DNA hydrogel stained by GelRed^TM DNA staining, respectively.

Recent studies have introduced a fascinating approach to enhance the strength and resilience of hydrogels by inducing dense polymer entanglements^15–18. Unlike covalent crosslinking, these entanglements function as slip crosslinks within the hydrogel¹⁹, increasing stiffness while maintaining toughness. When the hydrogel is stretched, tension is transmitted along the chain and to multiple other chains via entanglements, storing elastic energy within deformed conformations of the polymer strands before chain breakage. This mechanism could reinforce hydrogels and enable remarkable elasticity. Highly entangled synthetic polymers were produced by in situ polymerization at exceptionally high monomer concentrations, leading to extraordinary stiffness and toughness¹⁵ (Fig. 1B). Meanwhile, the denaturation and refolding of proteins at high concentration could fabricate densely entangled protein hydrogels¹⁶. However, these strategies cannot be adapted to introduce dense entanglements for biomass DNA. Therefore, although the exceptional chain length of biomass DNA (normally in the range of 10⁷-10¹²Da)²⁰ is favourable for entanglement engineering, and studies have revealed that entanglements generally exist in DNA hydrogel^21,22, the fabrication of entanglement dominant DNA hydrogels remains exceptionally challenging. methods to engage dense entanglements into biomass DNA might promote revolutionary strategies for producing high-strength DNA hydrogels.

In this study, we demonstrate a fast-shrinking-induced entanglement (FaSIE) strategy that facilitates the fabrication of densely entangled, rubber-like hydrogels from biomass DNA. The underlying mechanism stems from distinct polymer chain dynamics under varying dehydration rates: slow solvent evaporation allows sufficient chain relaxation and entropy-driven disentanglement, producing soft polymeric materials, while rapid dehydration kinetically traps non-equilibrium chain conformations through constrained relaxation, leading to increased entanglement density and mechanical rigidity. Building upon this phenomenon, we designed a dehydrating phase separation process for DNA solutions surpassing the entanglement threshold concentration, which triggers rapid volumetric shrinkage within seconds and generates ultrahigh-density topological constraints. The exceptionally long chains of biomass DNA further stabilize these non-equilibrium entanglements through pronounced reptation hindrance (Fig. 1C). This synergistic integration of kinetic control and macromolecular architecture enables instantaneous formation of entanglement-dominated hydrogels, demonstrating tens- to hundreds-fold increase in stiffness (E > 800 kPa) and toughness (W > 5 MJ/m³) relative to chemically crosslinked counterparts at equivalent water content (60 wt%). Notably, slow evaporation of DNA solutions to identical water content produces only viscous liquids, underscoring the pivotal role of dehydration kinetics. Comprehensive analysis identifies three fundamental requirements for hydrogel formation: high initial DNA concentration, rapid shrinking kinetic, and ultra-long chains. While challenging to satisfy with most synthetic polymers, these criteria are readily applicable to other natural biopolymers with extended chain lengths, such as giant polysaccharides. The FaSIE mechanism further empowers advanced material processing, including high-resolution 3D printing of intricate hydrogel architectures and seamless integration with functional components (e.g., magnetic nanoparticles). We validate this multifunctional capacity through prototype magnetoactive soft robots. Overall, the FaSIE strategy facilitated robust fabrication of DNA hydrogels with rubber-like mechanical strength that outperform all existing DNA-only hydrogels, thereby broadening the application horizons of DNA based materials.

Results

The entangled DNA hydrogel (E-Gel) prepared by fast-shrinking-induced entanglement (FaSIE)

To demonstrate the FaSIE process, the commercially available salmon sperm DNA was utilized as the model system. It consists of double-stranded DNA with an average molecular weight of 13200 kDa (N ~ 20000 base pairs (bp))²³. The contour length could reach about 6.8 µm, which is comparable to 600 kDa fully extended straight-chain polyethyleneglycol that is rare achieved synthetically. The 5% (w/v) salmon sperm DNA solution was prepared and a range of solvents was screened to induce dehydrating phase separation to this solution (Supplementary Fig. 1 and Fig. 3). Among these trials, the combination of glacial acetic acid (AA) and Triton X-100 (TX) was identified as the most effective (Supplementary Fig. 2). When the 1:1 mixture of AA and TX (v/v) was added, the volume of the DNA solution could shrink to ~50% within seconds, leading to the formation of a hydrogel instantly (Fig. 1D, Supplementary Video. 1). Further equilibration in the AA/TX solution would lead to additional shrinking of the hydrogel for hours. The resultant hydrogel, named the Entangled DNA Hydrogel (E-Gel), retained ~60% water content after equilibration. The infrared (IR) spectrum showed that the hydrogel primarily consisted of DNA without detectable residues of AA and TX after washing (Supplementary Fig. 4), indicating that AA and TX do not function as crosslinkers. Moreover, GoldView^TM staining confirmed the predominance of double-stranded DNA, and no obvious dissociation of double strands was observed even after stretching (Supplementary Figs. 5, 6). These data indicate the DNA structures were well maintained in the E-Gel, and no chemical changes likely occurred.

Fig. 3. Preparation and mechanical properties of E-Mg-Gel and E-Mg-PEG-Gel.

A Preparation of the E-Mg-Gel and E-Mg-PEG-Gel by adding divalent cations and moderate chemical crosslinking. B the E-Mg-PEG-Gel could be twisted and knotted while resisting sharp scalpel cutting. C Stress-strain curves of different samples under uniaxial tension stretch. D Stretching-relaxation stress-strain curves of E-Mg-PEG-Gel show hysteresis due to the additional sacrificial bonds. Inset: the deformation recovery kinetics of the hydrogel. About 85% of the energy dissipation capability can be recovered right after unloading, and the remaining can be recovered following a single-exponential kinetics, with k₁ of 0.051 s⁻¹ (E) Stress-strain curves of different samples under compression tests. F–H Ashby diagrams of stiffness versus ultimate tensile strength (F), ultimate tensile strength versus toughness (G) and compressive modulus versus compressive strength (H) of E-Gel, E-Mg-PEG-Gel, regular biopolymer hydrogels and reported entangled hydrogels. The data used are summarized in Supplementary Table. 3-4.

The E-Gel demonstrated an average total elongation (TE) at fracture of 1107.0 ± 104.2%, Young’s modulus (E) of 60.48 ± 5.65 kPa and an ultimate tensile strength (UTS) exceeding 0.4 MPa (Fig. 2A, C, and Supplementary Video. 3). When subjected to cyclic stretching at various amplitudes and a constant rate, the hysteresis of E-Gel was minimal (Supplementary Fig. 7). Such excellent elasticity is a feature of entanglement dominant hydrogel that is consistent with previous reports^24,25. In contrast, a conventional chemically crosslinked DNA hydrogel (C-Gel) was also prepared using the same salmon sperm DNA at the same concentration but crosslinked with PEG-DE (Poly(ethylene glycol) diglycidyl ether) as per prior methods (detailed preparation in SI)²⁶. It displayed significantly weaker mechanical properties (TE < 250%, UTS < 20 kPa, and E < 3 kPa, Fig. 2A, Supplementary Video. 2), even with the same water content (Supplementary Fig. 8). The compressive modulus (Y) of the E-Gel could reach 18.65 ± 2.46 kPa, which is also significantly higher than that of the C-Gel (Y = 1.52 ± 0.76 kPa) (Fig. 2B). Upon compression to 90%, the E-Gel could swiftly regain its original shape within seconds and sustain a stress as 1.304 ± 0.188 MPa without fracture, while the C-Gel collapsed entirely at 70% with a much lower stress of 20.19 ± 7,34 kPa (Fig. 2D). Scanning electron microscopy (SEM) of the E-Gel revealed a uniform structure without obvious pores (Fig. 2E), and E-Gel exhibited a homogeneous morphology throughout its volume (Supplementary Fig. 9). In contrast, immersing the E-Gel in water would result in enormously swelling, reaching up to 350 times of its original weight within hours, and a clearly porous structure could be observed in the SEM after swelling (Supplementary Fig. 10, and Fig. 2E). If left in ample water for over 30 days, the E-Gel could eventually dissolve which is a phenomenon consistent with dynamic interactions in the gel (Supplementary Fig. 10). The E-Gel could fully degrade upon treatment with DNase (Fig. 2F, and Supplementary Fig. 11A), consistent with the features of the other known DNA hydrogels. However, it exhibited resistance to heat up to 100 °C and is stable in a wide range of pH from 2 to 10, which is distinctive compared to strand-hybridized DNA hydrogels (Supplementary Fig. 11B–E).

Fig. 2. Mechanical properties and concentration/chain length dependency of E-Gel.

A The stress-strain curves of E-Gel and C-Gel under uniaxial tension. Insert: zoomed-in view of the stress-strain curves to show the C-Gel. B Compressive stress-strain curves of the E-gel and C-gel. Insert: zoomed-in view of the stress strain curves to show the C-Gel. C Photographs of the E-Gel and C-Gel during tensile tests. D Photographs of the E-Gel and C-Gel during compression tests. E Scanning electron microscopy (SEM) images and photographs of the E-Gel before and after swelling in PBS for 12 h. Representative images are shown from three independent experiments with similar results. F E-Gel digestion after treatment with DNase I (0.2 U/μL, 500 μL) for 24 h. G The chain length and concentration dependency of FaSIE process. The phase diagram was plotted using the data in Supplementary Fig. 16. H The stress-strain curves of E-Gel with different C_i under uniaxial tension. (I) The stress-strain curves of E-Gel with different C_i under compression tests.

The elastic modulus and ultimate rupture stretch of E-Gel fit well to the physical model for entanglement dominant hydrogel. In a highly entangled network, the Young’s modulus is intricately linked to the entanglement density through the equation E = 3ρRT/Me (Eq. 1)²⁷, where ρ represents polymer density, R is the gas constant, T is the absolute temperature, and Me corresponds to the molecular weight of polymer strands between entanglements²⁸. Consequently, the calculated Me is 61817 ± 5889 g/mol, indicating an average of 214 entanglements per DNA chain. Based on these data, we estimate the rupture strain to be ~1000% using a freely-jointed model (see more details in SI)^24,29, which aligns well with the experimental data (Fig. 2A).

Furthermore, when the DNA solution was initially crosslinked with PEG-DE and subsequently treated with AA/TX for several hours—until the pre-crosslinked DNA hydrogel reached a water content comparable to that of E-Gel—a weak hydrogel was yielded, exhibiting properties similar to the C-Gel (Supplementary Fig. 12), suggesting that pre-chemical crosslinking hindered DNA chain mobility and impeded chain entanglement.

Shrinking kinetic, initial concentration and chain length are the key factors for FaSIE based hydrogel

The FaSIE phenomenon is in general existing for all polymers. However, for most synthetic polymers, such entanglement is not stable to support sufficient mechanical strength. Entropy drive disentanglement could easily occur to reduce the entanglement density³⁰. In case of long chain nature biopolymers, like biomass DNA, significant chain interpenetrations are existing in solution with a practically high concentration. The AA/TX induce phase separation and fast dehydration of the DNA solution, thus immediately compress the volume of the interpenetrated DNA network. The extraordinary chain length and pre-existing entanglements prohibit the disentanglement tendency³¹, and the fast shrinking kinetic further reinforced the entanglements and constrained the DNA chains into the highly entangled network³². Based on this hypothesis, we further verified the influence of the dehydration time, the initial DNA concentration and the chain length toward the physical properties of the resulting hydrogels.

To compare with the solvent induced fast shrinking, 5% salmon sperm DNA was exposed to gradual dehydration with mild heating until reaching 60% water content as E-Gel. However, no gelation occurred and a viscous paste was obtained. By assessing the plateau elastic modulus (E_plateau) (Supplementary Fig. 13) in rheological test, the Me of this solution was determined to be 861764 g/mol according to E_plateau = ρRT/Me (Eq. 2), indicating an average of only 15 entanglement points on each DNA molecule. This supported that dense entanglements in the hydrogel arise from a kinetically driven process. Slow water evaporation allowed sufficient time for DNA chain disentanglement, resulting in insufficient entanglement density for gelation³³. Consistently, if the E-Gel is immersed in salted water for 2 days to prevent swelling while still permitting adequate water exchange, chain relaxation will also occur, thus leading to a notable weakening of the hydrogel (Supplementary Fig. 14). However, storing the hydrogel in a 4 °C fridge at 40% humidity for over one week does not visibly weaken it (Supplementary Fig. 15), indicating that E-Gel can maintain stability in air.

The initial DNA concentration (C_i) and the chain length of the DNA also significantly influent the FaSIE process. At low C_i, the AA/TX is miscible with the DNA solution, resulting in a clear liquid. As the C_i increases, DNA aggregation and precipitation become noticeable after AA/TX treatment (the concentration at which precipitation occurs is defined as C_p). Gelation requires an even higher concentration, termed the gelation concentration (C_g) (Supplementary Fig. 16, 17). To investigate these phenomena, we estimated the critical entanglement concentration (C_e) of DNA with varying molecular weights by monitoring the onset of the plateau elastic modulus and viscosity (see Supplementary Figs. 18, 19)^34,35. We found a strong correlation between C_e and C_p, emphasizing the importance of pre-existing chain interpenetration in the DNA solution, which is further reinforced by AA/TX-induced phase separation (Supplementary Table. 1). When C_i significantly exceeds C_e, a complete entangled network emerges throughout the DNA solution, leading to gelation. For high molecular weight DNA (e.g., 20,000 bp), C_e is relatively low (0.5%-1%), permitting the formation of stable hydrogels with robust mechanical properties within a practical range of C_i (5% to 15%). Conversely, as molecular weight decreases, C_e increases. When the DNA length is below 1000 bp, it becomes impractical to obtain a hydrogel within the above concentration ranges (Fig. 2G).

The mechanical properties of the hydrogels correlate closely with DNA molecular weight and C_i: higher molecular weights yield increased G’ and G” at a fixed C_i (Supplementary Fig. 20), while an increase in C_i significantly enhances stiffness, strength, and toughness (Fig. 2H, and Supplementary Fig. 21). For instance, in uniaxial tensile tests, increasing C_i from 5% to 12.5% boosted stiffness from 60 kPa to 800 kPa (Supplementary Fig. 22B), and tensile strength increased from 0.4 MPa to over 1.2 MPa (Supplementary Fig. 22C). Stretchability decreased with higher C_i (Supplementary Fig. 22D), consistent with increased entanglement density. By fitting these data with the Mooney-Rivlin model and non-affine tube model (Supplementary Fig. 23–24)²⁸, we found almost all of the modulus were contributed by the entanglements rather than by crosslinking. These results again supported that the E-Gel is an entanglement-dominated network. Compression tests also reflected this trend (Fig. 2I): at C_i = 10%, the compressive modulus (Y) reached 138.50 ± 28.89 kPa, and compressive strength rose to 6.83 ± 0.24 MPa (Supplementary Fig. 25), comparable to typical stiff and tough double network hydrogels made from synthetic polymers (Supplementary Table. 4). We observed a decline of compressive modulus and strength at an initial concentration of 12.5%, likely due to this concentration having already exceeded the saturation limit (Supplementary Fig. 26).

Stabilization by electrostatic interaction and chemical crosslinking

E-Gel is prepared in the non-swollen state and exhibits limited mechanical stability in aqueous solutions due to disentanglement, making it suitable for use in non-aqueous environments. To enhance the applicability of E-Gel in aqueous environments, Mg²⁺ and PEG-DE could be added post-gelation to introduce additional electrostatic interactions and chemical crosslinks (exemplified by C_i = 5% E-Gel with relative low DNA content and representative mechanical properties) (Fig. 3A, and Supplementary Fig. 27). Both the Mg²⁺-stabilized hydrogel (E-Mg-Gel) and the Mg²⁺/PEG-DE-stabilized hydrogel (E-Mg-PEG-Gel) could prevent the incessant swelling and obtain an equilibrate hydrogel with a water content of 70–80%. An increase in crosslinking density was also observed after Mg²⁺ and PEG-DE treatment (Supplementary Fig. 28). The UTS of the fully swelled hydrogels exceeded 1 MPa, with TE of ~1200% and a stiffness of 150 kPa (Fig. 3C). These hydrogels displayed more hysteresis comparing to E-Gel during cyclic stretching and compression, indicating increased energy dissipation due to sacrificial bonds (Supplementary Fig. 29–31). Nevertheless, the hydrogel could rapid recover from stretching and compression (Fig. 3D, Supplementary Fig. 31). At large strains (100%), the original energy dissipation capability could recover within 1 min. Crack propagation tests on single-edge notched samples showed a fracture energy increase from 20.41 ± 5.80 J/m² (C-Gel) to 1546.0 ± 420.3 J/m² (E-Gel) and 1013.0 ± 465.8 J/m² (E-Mg-PEG-Gel) (Supplementary Fig. 32, Supplementary Table. 2)³⁶. The E-Mg-PEG-Gel demonstrated excellent elasticity, allowing for easy knotting and twisting (Fig. 3B). It also exhibited notable compressive properties with strengths up to 5 MPa at 90% compression without fracture (Y = 221.9 ± 57.0 kPa) (Fig. 3E), that is resistance to sharp scalpel cutting (Fig. 3B). Compared with previously reported hydrogels composed solely of DNA, both E-Gel and E-Mg-PEG-Gel exhibit orders-of-magnitude enhancements in stiffness, toughness, ultimate tensile strength (UTS), compressive modulus, and compressive strength (Fig. 3F–H, and Supplementary Fig. 32, Supplementary Table. 2). Moreover, these enhanced mechanical properties also surpass those of DNA hybrid hydrogels^37–45 (Supplementary Table. 3) and biopolymer-based hydrogels^46,47, and rival the performance of reported entangled polymer hydrogels and double network hydrogels^{15,25,48–52} (Supplementary Table. 4).

High resolution 3D printing of biomass DNA hydrogel

The concentrated DNA solution exhibited high viscosity and shear-thinning behaviour, which is ideal for extrusion-based direct ink writing (DIW) 3D printing (Supplementary Fig. 19). The FaSIE process could be used as a post-printing treatment to obtain DNA hydrogel with micrometer scale resolution. The fast gelation process maintained the original shape of the DNA ink and the uniformed shrinkage significantly increased the structure resolution. For instance, the as-designed DNA frame structure was printed in four layers, with an original size of 10 mm × 10 mm and an original filament diameter of 647.0 ± 71.3 μm (Fig. 4D). After treatment with AA/TX, the gel quickly formed along with drastic volume shrinkage, resulting in a final DNA frame structure with a filament diameter of 66.1 ± 10.8 μm (Fig. 4A–D). Such resolution is among the best resolutions for extrusion-based hydrogel 3D printing^53–62, and it has never been achieved by DNA 3D printing (Supplementary Fig. 33). The different layers were also clearly distinguishable on SEM, supporting the excellent structure fidelity (Fig. 4C). The zoom-in images revealed the microstrutctures of the filaments were as same as E-Gel (Supplementary Fig. 34). By this way, the DNA hydrogels with various structures were printed, such as a cube, star, snowflake, and HUST (Fig. 4E). The FaSIE based 3D printing showed great shape fidelity, while the chemical crosslinking method (see details in supplementary discussion) could barely maintain its shape and quickly collapsed due to the time-consuming gelation process (Fig. 4E). The statistical shrinkage factor was found to be 4.39 ± 0.98 in the lateral dimension and 4.01 ± 0.69 in the axial dimension (Fig. 4F), indicating a favourable uniformed shrinkage.

Fig. 4. High-resolution 3D printing and functionalization of biomass DNA hydrogel.

A Photograph of high-resolution 3D printed DNA frame. B Super-depth-of-field microscopic image and (C) SEM image of 3D printed DNA frame. D Comparison of filament diameters before and after FaSIE. Data are presented as mean ± standard deviations (s.d.) of n = 6 replicates (independent samples). E Various structures printed via FaSIE based high resolution 3D printing (E-Gel 3D printing) and conventional chemical crosslinking based 3D printing (C-Gel 3D printing). F Statistical shrinking factor in lateral and axial dimension. Data are presented as mean ± standard deviations (s.d.) of n = 5 replicates (independent samples). G Fabrication of magnetic DNA soft robot. (i) The non-functionalized DNA solution and magnetic nanoparticles blended DNA solution was filled to the mold separately, and subsequent FaSIE process result in a whole piece hydrogel as a magnetic fork. (ii) The magnetic fork could be controlled by a magnet to lift up object.

Functionalization of biomass DNA hydrogel and extension of FaSIE to other biopolymers

The FaSIE process offers versatility for integrating with various functional components. The high viscosity of the initial DNA solution, combined with the rapid gelation process, also facilitates the fabrication of functional hydrogels with varying components in specific areas. For instance, FaSIE treatment of a DNA solution mixed with Fe₃O₄ nanoparticles could result in a magnetic DNA hydrogel. Thus, a magnetic DNA soft robot could be prepared by fusion the magnetic forks at both ends with a non-magnetic handle in the middle, which is capable of directional movement and can lift objects in response to a magnetic field (Fig. 4G, and Supplementary Video. 4). Additionally, the hydrogel retains the feature of sequence-specific hybridization inherent in DNA-based materials. When a fluorescein-labeled ssDNA probe with a complementary sequence to salmon sperm DNA is incubated with the E-Mg-PEG-Gel, stable integration of green fluorescence is achieved even after thorough washing with PBS. In contrast, free fluorescein dye lacking a DNA probe or with an uncomplimentary DNA probe would be readily washed away by PBS (Supplementary Fig. 35). Moreover, both E-Mg-Gel and E-Mg-PEG-Gel exhibited excellent biocompatibility with mammalian cells (Supplementary Fig. 36). Together, these results highlight the broad application potential of DNA hydrogels.

In addition to Salmon Sperm DNA, we have applied this strategy to another type of biomass DNA, namely Calf Thymus DNA, which exhibited similar mechanical properties to Salmon Sperm DNA (Supplementary Fig. 37, Supplementary Table. 5), demonstrating the generality of this strategy for various long-strand biomass DNAs. With the success of the FaSIE process in DNA hydrogels, the potential for reinforce other biopolymer hydrogels was also explored. Similar trend was observed with sodium alginate, sodium hyaluronate and carboxymethyl cellulose (Supplementary Fig. 38, Supplementary Table. 6). The FaSIE process is able to achieve hydrogels with stiffness and toughness 10 to 270 times enhanced comparing to conventional chemical crosslinking. Notably, these results were obtained with the same protocol used for DNA hydrogels. Additional optimization of the dehydration solution is expected to further enhance the FaSIE performance for individual biopolymers. Nevertheless, the preliminary data have already demonstrated the promising potential of FaSIE to enhance the mechanical properties of various biopolymers.

Discussion

Overall, we have established a fast-shrinking-induced entanglement (FaSIE) process that produces stiff and tough hydrogels solely composed of biomass DNA. By applying a phase-separation dehydrating solvent, rapid shrinkage of the concentrated DNA solution led to a dramatic entanglement reinforcement in the confined volume. The super long chain feature of biomass DNA hindered entropy driven disentanglement, therefore ensured the hydrogel formation. The resulting DNA hydrogel exhibited excellent elasticity, stiffness, and toughness, which outperformed reported DNA-only hydrogels by orders of magnitude. Moreover, additional electrostatic interactions and chemical crosslinking after the FaSIE process further reduce hydrogel swelling and enhance mechanical strength. Furthermore, the FaSIE method also ensures uniform shrinkage in both the x and y axes, making it suitable for high-resolution DNA 3D printing with micrometer-scale resolution. This process is also easily adaptable to the incorporation of functional materials, paving a way to the development of soft robots and electronic sensors using DNA hydrogels. Such efficient manufacturing techniques would potentially promote DNA as a new class of sustainable biomass material resources alongside proteins and polysaccharides for diverse applications. Additionally, the FaSIE phenomenon is not exclusive to biomass DNA; it also occurs in other long-chain biopolymers. Therefore, this strategy also holds great promise for improving the mechanical strength of a variety of long-chain biopolymer-based materials, thus adding a valuable toolkit for biomacromolecular engineering.

Methods

Materials

Salmon sperm DNA is purchased from Solarbio. Herring sperm DNA and DNA (Calf thymus) are purchased from Sigma-Aldrich. Acetic acid is purchased from Sinopharm. Triton-X 100 is purchased from Sangon. Magnesium chloride is purchased from Sangon. PEGDE (average MW 500) is purchased from TCI.

Synthesis of highly entangled DNA hydrogel (E-Gel) induced by fast shrinking

5% (w/v) salmon sperm DNA was dissolved in 1 × PBS (Sangon) buffer, for instance, 10 mg DNA with 200 μL PBS, mixed well until completely dissolved. The 5% salmon sperm DNA is highly viscous and exhibits semi-solid-like properties, causing it to flow slowly. It was then carefully poured into pre-designed molds, and the surface of the solution was flattened. 1 volume fold 1:1 (v/v) of AA/TX mixture was added into the mold right away to cover the DNA solution. The hydrogel would form quickly in seconds, and then the mold was kept in the oven at 37 °C for 2 h for further shrinking and equilibration with AA/TX.

Synthesis of E-Mg-Gel and E-Mg-PEG-Gel

To obtain the E-Mg-Gel, E-Gel was immersed in a high concentration of MgCl₂ (2.5 M) for 5 minutes, repeated three times. The E-Mg-Gel remained stable in PBS buffer and swelled to a final water content of 70-80%. To further strengthen and stiffen the hydrogel, E-Mg-Gel was chemically crosslinked with PEGDE. E-Mg-Gel, without swelling, was taken out of the tube and placed onto a plate for crosslinking; the hydrogel should not be in a solution environment. Then, 5% (w/v) PEGDE (original volume of the DNA solution) was added to the surface of the hydrogel, allowing the penetration of the crosslinker for 15-20 minutes. After that, 0.44% (w/v) TEMED was added to catalyze the reaction for 2 h, resulting in the E-Mg-PEG-Gel, then the gel was immersed in the 1 × PBS solution for at least 12 h to swell, which could swell to a final water content of 70-80%.

Synthesis of C-Gel and C-Mg-Gel

The same concentration, 5% DNA, was used to fabricate chemically crosslinked DNA hydrogel. As described, 5% (w/v) PEGDE was added to the DNA solution, along with 0.44% (w/v) TEMED, and the mixture was stirred well to obtain a uniform DNA solution. Then, the DNA solution was carefully poured into the mold, and the mold was left in the oven at 42 °C for 3 hours and at room temperature overnight for the complete reaction. To obtain the C-Mg-Gel, the same process as with E-Gel was followed: the C-Gel was placed in a high concentration MgCl₂ solution for 5 minutes, repeated three times.

Functionalization of biomass DNA hydrogel

Various functional components, such as Fe₃O₄, can be easily introduced into the hydrogel. To fabricate a locally magnetic-responsive DNA hydrogel, a 5% DNA solution was prepared and then poured into the mold. For the magnetic-responsive part, 2-5% (w/w) Fe₃O₄ was added to the DNA solution and stirred well. The mixed DNA solution could then be added at any desired location in the mold. The DNA solution and magnetic-responsive DNA solution were left to fusion for 5 minutes; the added component did not diffuse due to the high viscosity of the DNA solution. Then, AA/TX was added to fabricate the hydrogel.

Synthesis of crosslinked biopolymer hydrogels

6% sodium alginate was dissolved in the pure water and then immersed into 20 mM CaCl₂ solution for 30 min to obtained crosslinked alginate hydrogel. 5% sodium hyaluronate was dissolved in the 0.25 M NaOH solution and then 5% (w/v) PEGDE was added to the solution, the mixture was placed at 50 °C for 4 h to complete the gelation. 5% carboxymethyl cellulose was dissolved in the 0.4 M NaOH solution and then 5% (w/v) PEGDE was added to the solution, the mixture was placed at 50 °C for 4 h to complete the gelation, the hydrogel was collected and stored at 4 °C for use.

Synthesis of highly entangled biopolymer hydrogels

Highly entangled hydrogels were fabricated using solutions containing 6% sodium alginate, 5% sodium hyaluronate, and 5% carboxymethyl cellulose. An equal volume of the AA/TX mixture (1:1, v/v) was added to the mold containing the biopolymer solution. Rapid gelation occurred within seconds, after which the mold was incubated at 37 °C for 2 h to allow for further shrinking and equilibration.

Tensile tests

Various hydrogels were made into rectangular shaped species, with E-Gel, E-Mg-Gel and E-Mg-PEG-Gel of 15 mm × 5 mm × 1 mm. uniaxial tensile tests were performed on a commercial tensile machine (CTM6050) at room temperature, both ends of the samples were glued with a layer of sandpaper to reduce the slippage. The tensile tests were performed at a loading rate of 30 mm min^-1. The nominal stress and the elongation in the tensile direction were measured and calculated by machine. The Young’s modulus was calculated from the initial slope of the stress-strain curve. Toughness was determined by integrating stress–strain curves where specimens were loaded directly to failure.

Cyclic Tensile tests

Cyclic tensile tests were also performed using the same experimental setup. The samples were firstly stretched to a certain strain and then immediately unload, and they were stretched to an increased strain and unloaded again with the constant loading and unloading rates of 50 mm min^-1. The mechanical hysteresis was calculated by integrating loop area between loading and unloading stress–strain curves. In deformation recovery experiments, a pulling rate of 50 mm min⁻¹ was used. The same hydrogel was loaded and unloaded with various time intervals. The nominal stress and strain were recorded.

Determination of the fracture energy G_c

Fracture tests were performed using the classical single edge notch test. A notch of 1 mm length in the middle of the hydrogel was made which the total width of the hydrogel was about 5 mm. the hydrogel was clamped and stretched with recording the nominal stress and strain with a constant rates of 30 mm min^-1. The fracture energy G_c was calculated using the methodology as described previously reported³⁶. Which the fracture energy is given by

$$G_c=\frac{6\cdot W\cdot c}{\sqrt{\lambda c}}$$ 1

With c was the length of the crack, λc the strain at the break in single edge notch experiment and W the strain energy density calculated by integration of the stress versus strain of un-notched samples, until λc.

Compression and cyclic tests

Uniaxial compression tests were performed on cylinder-shaped hydrogels with a diameter of 6 mm and a height 6 mm using the same mechanical machine. The compression tests were performed at a constant rate of 5 mm min^-1. And the stress-strain curve was measured and recorded. The compressive modulus was measured at a strain of 10–20%. The ultimate stress was determined at 90% strain. Energy dissipation was calculated by integrating the loop area between compressing and relaxing stress–strain curves. In hysteresis recovery experiments, a compression rate of 50 mm min⁻¹ was used.

Water contents of hydrogels

The water contents of hydrogels were calculated from the weights of the gel before and after drying, the water content, denoted as Ww, using the following equation:

$$W_w=1-\frac{W_{dry}}{W_{wet}}$$ 2

Swelling ratio

The initial weight of E-Gel was recorded as W₀, and the E-Gel was immersed in different buffer solutions for different time, at selected time intervals, the hydrogels were taken out and wiped with filter paper to remove excess water from the gel surface and then weighted Ws, the swelling ratio was calculated using the following equation:

$$Swellingratio=\frac{W_s}{W_0}$$ 3

DNA hydrogel staining

The hydrogel was stained by GelRed (Mei5Bio) to check whether the hydrogel was formed by DNA. To study the DNA conformation in the hydrogel, the DNA hydrogel was stained by GoldView (Biosharp), a dye could emit red color fluorescence when bind with single strand DNA and green for double strand DNA⁶³.

DNA hydrogel digestion by DNase I

The hydrogel was put into the DNase I solution (0.2U/μL, Sangon) supplied with Mg²⁺ and Ca²⁺ for 24 h, the hydrogel was pictured at different time intervals for evaluating the digestion condition of DNA hydrogel.

Attenuated total reflection-Fourier transform infrared spectrometer (ATR-FTIR)

ATR-FTIR spectra were measured using Thermo Scientific Nicolet iS5 from 600 cm⁻¹ to 4000 cm⁻¹. For AA, Triton-X 100, and AA/TX, the solution was applied to the surface of the instrument for detection. The pure salmon DNA was a fiber-like solid; the probe of the spectrometer was pressed onto the surface of the DNA fiber, and the spectrum was measured. The E-Gel was immersed in PBS solution and washed several times to remove excess AA/TX residues. It was then freeze-dried to obtain a lyophilized hydrogel for ATR-FTIR detection.

Gel electrophoresis characterization

The different molecular weight of DNA was measured by 2% agarose gel. the gel was run in iced 1 × TBE buffer at 90 V for 2 h, stained by GelRed, and pictured on the Bio-Rad GelDoc XR.

Scanning electron microscopy (SEM) imaging

To observe the micro-structure of the E-Gel before and after swelling, the hydrogels were shock-frozen in liquid nitrogen and then quickly transferred into the freeze drier for at least 24 h to obtain the lyophilized hydrogel. and the hydrogel was coated by sputtering gold onto the surface before scanning. The measurement was performed with HITACHI SU8010. 3D printed DNA structure was directly scanned.

Confocal laser scanning microscopy (CLSM)

CLSM images were obtained to visualize the homogeneity of the hydrogels. 10 μL of 10 × GoldView was added into 5% DNA solution and then used to form the hydrogel. The CLSM images were obtained with a Olympus FV3000 using an air lens. The three-dimensional images were rendered at 5 µm intervals with a range 500 μm.

Small angle X-ray scattering (SAXS) measurements

The SAXS measurement was performed at BL19U2 beamline of Shanghai Synchrotron Radiation Facility (SSRF, China). The X-ray energy used was 12 keV, and the distance between sample and detector was 5353 mm. A Mar 165 CCD detector (2048 × 2048 pixels with pixel size of 172 μm) was used. to collect two-dimensional (2D) scattering patterns with a data acquisition time of 10 s per frame.

Rheology measurements

The elastic modulus and loss modulus of DNA solution, DNA hydrogel was measured by using a rheometer (AntonPaar MCR 102) equipped with 8 mm diameter steel plate requiring as low as 50 μL of samples. To measure the entanglement concentration of different chain length DNA, the frequency dependence experiments were performed. The frequency sweeping was carried out in the 0.01-10 rad s^-1 range at a constant shear strain of 1%, DNA was dissolved in pure water at different concentration. To compare the elastic modulus of different DNA hydrogels, the frequency sweeping was kept in the range of 0.01-10 Hz at a constant shear strain of 1%.

Viscosity measurements

Viscosity could also help to identify the entanglement concentration. The dynamic viscosity behavior of salmon sperm DNA was studied through flow measurement by rheometer equipped with a 20 mm diameter steel cone and an angle of 2°, with the concentration from the dilute regime through semi-dilute regime to entanglement regime. Simple shear steady state measurements were performed in a shear rate range from 10^-3 to 1000 s^-1, using five points per decade.

High-resolution 3D printing DNA hydrogel

The printing DNA ink was prepared by dissolving 4-5% (w/v) salmon sperm DNA in PBS solution. The DNA ink was mounted on the self-made 3D printer and extruded by pressure. To print the high-resolution DNA frame structure, a 250 µm diameter nozzle was used with a print speed of 20 mm/s, yielding a 600 µm diameter DNA line. When the DNA frame structure was finished, AA/TX (combination with 1 M MgCl₂) was added to trigger the gelation of the DNA solution. After treatment for a few minutes, the DNA frame was transferred into AA/TX solution and immersed for 24 hours to obtain a high-resolution 3D-printed DNA structure, the high-resolution frame structure was imaged by Camera (Sony), Super-Depth-of-Field Microscopy (VHX-1000E) and SEM (HITACHI). Other structures were also printed using a 500 µm diameter nozzle and subjected to the same treatment as the DNA frame. For comparison with chemical crosslinking post-treatment 3D printing, 4-5% DNA was dissolved in the PBS solution with 5% (w/v) PEGDE and 0.44% (w/v) TEMED and then mounted on the 3D printer, the printing process was the same with E-Gel, after the printing, the chemical crosslinking ink was put aside at 40 °C 2 h for crosslinking to obtain the printed hydrogel. Each printing model was built with 3D modeling software (Blender) and then sent to slicer software (Simplify3D) to obtain the print path (G-code).

Cytotoxicity of biomass DNA hydrogel

Cell Counting Kit-8 (CCK-8) (Sangon, China) was used to assess cell viability following the manufacturer’s instructions. Briefly, 1 × 10⁴ L929 cells were seeded in each well of a 96-well plate and incubated under standard conditions (37 °C, 5% CO₂) for 24 h. In one group, 5% E-Mg-Gel or E-Mg-PEG-Gel of different masses was added to the cells, and the mixtures were incubated under 5% CO₂ at 37 °C for 4 h. The hydrogels were then carefully removed from the wells using sterile tweezers. Subsequently, 100 µL of DMEM mixed with 10 µL of CCK-8 solution was added to each well and incubated at 37 °C for 4 h. The absorbance of the medium in each well was measured at 450 nm using a microplate reader. In another group, 400 mg of 5% E-Mg-Gel or E-Mg-PEG-Gel was incubated with DMEM for 24 h. The resulting extract solutions were then added to the cells at volumes adjusted to maintain the same DNA-to-medium mass ratio as in group one, and the mixtures were incubated under 5% CO₂ at 37 °C for 24 h.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Author contributions

Z.L. and Y.W. conceived of the project. Z.L. Z.Z., G.Z. L.Z. G.W., and Y.W. designed the study and interpreted the results. Z.L., S.F., Q.H., X.X. and J.J. performed the experiments. Y.W. and Z.L. wrote the manuscript. Y.W. supervised the research. All authors discussed the result and commented on the manuscript.

Peer review

Peer review information

Nature Communications thanks Arghya Paul and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

The data are available within the article and Supplementary Files. Source data are provided with this paper. All data are available from the corresponding author upon request. Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at 10.1038/s41467-026-68363-x.